diff options
Diffstat (limited to 'drivers/base')
103 files changed, 15377 insertions, 10165 deletions
diff --git a/drivers/base/Kconfig b/drivers/base/Kconfig index ffcbe2bc460e..1786d87b29e2 100644 --- a/drivers/base/Kconfig +++ b/drivers/base/Kconfig @@ -62,6 +62,17 @@ config DEVTMPFS_MOUNT rescue mode with init=/bin/sh, even when the /dev directory on the rootfs is completely empty. +config DEVTMPFS_SAFE + bool "Use nosuid,noexec mount options on devtmpfs" + depends on DEVTMPFS + help + This instructs the kernel to include the MS_NOEXEC and MS_NOSUID mount + flags when mounting devtmpfs. + + Notice: If enabled, things like /dev/mem cannot be mmapped + with the PROT_EXEC flag. This can break, for example, non-KMS + video drivers. + config STANDALONE bool "Select only drivers that don't need compile-time external firmware" default y @@ -156,6 +167,12 @@ config PM_QOS_KUNIT_TEST depends on KUNIT=y default KUNIT_ALL_TESTS +config PM_RUNTIME_KUNIT_TEST + tristate "KUnit Tests for runtime PM" if !KUNIT_ALL_TESTS + depends on KUNIT + depends on PM + default KUNIT_ALL_TESTS + config HMEM_REPORTING bool default n @@ -215,8 +232,21 @@ config GENERIC_ARCH_TOPOLOGY config GENERIC_ARCH_NUMA bool + select NUMA_MEMBLKS help Enable support for generic NUMA implementation. Currently, RISC-V and ARM64 use it. +config FW_DEVLINK_SYNC_STATE_TIMEOUT + bool "sync_state() behavior defaults to timeout instead of strict" + help + This is build time equivalent of adding kernel command line parameter + "fw_devlink.sync_state=timeout". Give up waiting on consumers and + call sync_state() on any devices that haven't yet received their + sync_state() calls after deferred_probe_timeout has expired or by + late_initcall() if !CONFIG_MODULES. You should almost always want to + select N here unless you have already successfully tested with the + command line option on every system/board your kernel is expected to + work on. + endmenu diff --git a/drivers/base/Makefile b/drivers/base/Makefile index ef8e44a7d288..8074a10183dc 100644 --- a/drivers/base/Makefile +++ b/drivers/base/Makefile @@ -6,25 +6,27 @@ obj-y := component.o core.o bus.o dd.o syscore.o \ cpu.o firmware.o init.o map.o devres.o \ attribute_container.o transport_class.o \ topology.o container.o property.o cacheinfo.o \ - swnode.o + swnode.o faux.o obj-$(CONFIG_AUXILIARY_BUS) += auxiliary.o obj-$(CONFIG_DEVTMPFS) += devtmpfs.o obj-y += power/ obj-$(CONFIG_ISA_BUS_API) += isa.o obj-y += firmware_loader/ obj-$(CONFIG_NUMA) += node.o -obj-$(CONFIG_MEMORY_HOTPLUG_SPARSE) += memory.o +obj-$(CONFIG_MEMORY_HOTPLUG) += memory.o ifeq ($(CONFIG_SYSFS),y) obj-$(CONFIG_MODULES) += module.o +obj-$(CONFIG_AUXILIARY_BUS) += auxiliary_sysfs.o endif obj-$(CONFIG_SYS_HYPERVISOR) += hypervisor.o obj-$(CONFIG_REGMAP) += regmap/ obj-$(CONFIG_SOC_BUS) += soc.o obj-$(CONFIG_PINCTRL) += pinctrl.o obj-$(CONFIG_DEV_COREDUMP) += devcoredump.o -obj-$(CONFIG_GENERIC_MSI_IRQ_DOMAIN) += platform-msi.o +obj-$(CONFIG_GENERIC_MSI_IRQ) += platform-msi.o obj-$(CONFIG_GENERIC_ARCH_TOPOLOGY) += arch_topology.o obj-$(CONFIG_GENERIC_ARCH_NUMA) += arch_numa.o +obj-$(CONFIG_ACPI) += physical_location.o obj-y += test/ diff --git a/drivers/base/arch_numa.c b/drivers/base/arch_numa.c index 4cc4e117727d..c99f2ab105e5 100644 --- a/drivers/base/arch_numa.c +++ b/drivers/base/arch_numa.c @@ -12,16 +12,12 @@ #include <linux/memblock.h> #include <linux/module.h> #include <linux/of.h> +#include <linux/numa_memblks.h> #include <asm/sections.h> -struct pglist_data *node_data[MAX_NUMNODES] __read_mostly; -EXPORT_SYMBOL(node_data); -nodemask_t numa_nodes_parsed __initdata; static int cpu_to_node_map[NR_CPUS] = { [0 ... NR_CPUS-1] = NUMA_NO_NODE }; -static int numa_distance_cnt; -static u8 *numa_distance; bool numa_off; static __init int numa_parse_early_param(char *opt) @@ -30,6 +26,8 @@ static __init int numa_parse_early_param(char *opt) return -EINVAL; if (str_has_prefix(opt, "off")) numa_off = true; + if (!strncmp(opt, "fake=", 5)) + return numa_emu_cmdline(opt + 5); return 0; } @@ -61,6 +59,7 @@ EXPORT_SYMBOL(cpumask_of_node); #endif +#ifndef CONFIG_NUMA_EMU static void numa_update_cpu(unsigned int cpu, bool remove) { int nid = cpu_to_node(cpu); @@ -83,6 +82,7 @@ void numa_remove_cpu(unsigned int cpu) { numa_update_cpu(cpu, true); } +#endif void numa_clear_node(unsigned int cpu) { @@ -144,7 +144,7 @@ void __init early_map_cpu_to_node(unsigned int cpu, int nid) unsigned long __per_cpu_offset[NR_CPUS] __read_mostly; EXPORT_SYMBOL(__per_cpu_offset); -static int __init early_cpu_to_node(int cpu) +int early_cpu_to_node(int cpu) { return cpu_to_node_map[cpu]; } @@ -154,36 +154,34 @@ static int __init pcpu_cpu_distance(unsigned int from, unsigned int to) return node_distance(early_cpu_to_node(from), early_cpu_to_node(to)); } -static void * __init pcpu_fc_alloc(unsigned int cpu, size_t size, - size_t align) -{ - int nid = early_cpu_to_node(cpu); - - return memblock_alloc_try_nid(size, align, - __pa(MAX_DMA_ADDRESS), MEMBLOCK_ALLOC_ACCESSIBLE, nid); -} - -static void __init pcpu_fc_free(void *ptr, size_t size) -{ - memblock_free_early(__pa(ptr), size); -} - void __init setup_per_cpu_areas(void) { unsigned long delta; unsigned int cpu; - int rc; + int rc = -EINVAL; + + if (pcpu_chosen_fc != PCPU_FC_PAGE) { + /* + * Always reserve area for module percpu variables. That's + * what the legacy allocator did. + */ + rc = pcpu_embed_first_chunk(PERCPU_MODULE_RESERVE, + PERCPU_DYNAMIC_RESERVE, PAGE_SIZE, + pcpu_cpu_distance, + early_cpu_to_node); +#ifdef CONFIG_NEED_PER_CPU_PAGE_FIRST_CHUNK + if (rc < 0) + pr_warn("PERCPU: %s allocator failed (%d), falling back to page size\n", + pcpu_fc_names[pcpu_chosen_fc], rc); +#endif + } - /* - * Always reserve area for module percpu variables. That's - * what the legacy allocator did. - */ - rc = pcpu_embed_first_chunk(PERCPU_MODULE_RESERVE, - PERCPU_DYNAMIC_RESERVE, PAGE_SIZE, - pcpu_cpu_distance, - pcpu_fc_alloc, pcpu_fc_free); +#ifdef CONFIG_NEED_PER_CPU_PAGE_FIRST_CHUNK + if (rc < 0) + rc = pcpu_page_first_chunk(PERCPU_MODULE_RESERVE, early_cpu_to_node); +#endif if (rc < 0) - panic("Failed to initialize percpu areas."); + panic("Failed to initialize percpu areas (err=%d).", rc); delta = (unsigned long)pcpu_base_addr - (unsigned long)__per_cpu_start; for_each_possible_cpu(cpu) @@ -191,179 +189,28 @@ void __init setup_per_cpu_areas(void) } #endif -/** - * numa_add_memblk() - Set node id to memblk - * @nid: NUMA node ID of the new memblk - * @start: Start address of the new memblk - * @end: End address of the new memblk - * - * RETURNS: - * 0 on success, -errno on failure. - */ -int __init numa_add_memblk(int nid, u64 start, u64 end) -{ - int ret; - - ret = memblock_set_node(start, (end - start), &memblock.memory, nid); - if (ret < 0) { - pr_err("memblock [0x%llx - 0x%llx] failed to add on node %d\n", - start, (end - 1), nid); - return ret; - } - - node_set(nid, numa_nodes_parsed); - return ret; -} - /* * Initialize NODE_DATA for a node on the local memory */ static void __init setup_node_data(int nid, u64 start_pfn, u64 end_pfn) { - const size_t nd_size = roundup(sizeof(pg_data_t), SMP_CACHE_BYTES); - u64 nd_pa; - void *nd; - int tnid; - if (start_pfn >= end_pfn) pr_info("Initmem setup node %d [<memory-less node>]\n", nid); - nd_pa = memblock_phys_alloc_try_nid(nd_size, SMP_CACHE_BYTES, nid); - if (!nd_pa) - panic("Cannot allocate %zu bytes for node %d data\n", - nd_size, nid); - - nd = __va(nd_pa); + alloc_node_data(nid); - /* report and initialize */ - pr_info("NODE_DATA [mem %#010Lx-%#010Lx]\n", - nd_pa, nd_pa + nd_size - 1); - tnid = early_pfn_to_nid(nd_pa >> PAGE_SHIFT); - if (tnid != nid) - pr_info("NODE_DATA(%d) on node %d\n", nid, tnid); - - node_data[nid] = nd; - memset(NODE_DATA(nid), 0, sizeof(pg_data_t)); NODE_DATA(nid)->node_id = nid; NODE_DATA(nid)->node_start_pfn = start_pfn; NODE_DATA(nid)->node_spanned_pages = end_pfn - start_pfn; } -/* - * numa_free_distance - * - * The current table is freed. - */ -void __init numa_free_distance(void) -{ - size_t size; - - if (!numa_distance) - return; - - size = numa_distance_cnt * numa_distance_cnt * - sizeof(numa_distance[0]); - - memblock_free(__pa(numa_distance), size); - numa_distance_cnt = 0; - numa_distance = NULL; -} - -/* - * Create a new NUMA distance table. - */ -static int __init numa_alloc_distance(void) -{ - size_t size; - u64 phys; - int i, j; - - size = nr_node_ids * nr_node_ids * sizeof(numa_distance[0]); - phys = memblock_find_in_range(0, PFN_PHYS(max_pfn), - size, PAGE_SIZE); - if (WARN_ON(!phys)) - return -ENOMEM; - - memblock_reserve(phys, size); - - numa_distance = __va(phys); - numa_distance_cnt = nr_node_ids; - - /* fill with the default distances */ - for (i = 0; i < numa_distance_cnt; i++) - for (j = 0; j < numa_distance_cnt; j++) - numa_distance[i * numa_distance_cnt + j] = i == j ? - LOCAL_DISTANCE : REMOTE_DISTANCE; - - pr_debug("Initialized distance table, cnt=%d\n", numa_distance_cnt); - - return 0; -} - -/** - * numa_set_distance() - Set inter node NUMA distance from node to node. - * @from: the 'from' node to set distance - * @to: the 'to' node to set distance - * @distance: NUMA distance - * - * Set the distance from node @from to @to to @distance. - * If distance table doesn't exist, a warning is printed. - * - * If @from or @to is higher than the highest known node or lower than zero - * or @distance doesn't make sense, the call is ignored. - */ -void __init numa_set_distance(int from, int to, int distance) -{ - if (!numa_distance) { - pr_warn_once("Warning: distance table not allocated yet\n"); - return; - } - - if (from >= numa_distance_cnt || to >= numa_distance_cnt || - from < 0 || to < 0) { - pr_warn_once("Warning: node ids are out of bound, from=%d to=%d distance=%d\n", - from, to, distance); - return; - } - - if ((u8)distance != distance || - (from == to && distance != LOCAL_DISTANCE)) { - pr_warn_once("Warning: invalid distance parameter, from=%d to=%d distance=%d\n", - from, to, distance); - return; - } - - numa_distance[from * numa_distance_cnt + to] = distance; -} - -/* - * Return NUMA distance @from to @to - */ -int __node_distance(int from, int to) -{ - if (from >= numa_distance_cnt || to >= numa_distance_cnt) - return from == to ? LOCAL_DISTANCE : REMOTE_DISTANCE; - return numa_distance[from * numa_distance_cnt + to]; -} -EXPORT_SYMBOL(__node_distance); - static int __init numa_register_nodes(void) { int nid; - struct memblock_region *mblk; - - /* Check that valid nid is set to memblks */ - for_each_mem_region(mblk) { - int mblk_nid = memblock_get_region_node(mblk); - phys_addr_t start = mblk->base; - phys_addr_t end = mblk->base + mblk->size - 1; - - if (mblk_nid == NUMA_NO_NODE || mblk_nid >= MAX_NUMNODES) { - pr_warn("Warning: invalid memblk node %d [mem %pap-%pap]\n", - mblk_nid, &start, &end); - return -EINVAL; - } - } + + /* Check the validity of the memblock/node mapping */ + if (!memblock_validate_numa_coverage(0)) + return -EINVAL; /* Finally register nodes. */ for_each_node_mask(nid, numa_nodes_parsed) { @@ -388,11 +235,7 @@ static int __init numa_init(int (*init_func)(void)) nodes_clear(node_possible_map); nodes_clear(node_online_map); - ret = numa_alloc_distance(); - if (ret < 0) - return ret; - - ret = init_func(); + ret = numa_memblks_init(init_func, /* memblock_force_top_down */ false); if (ret < 0) goto out_free_distance; @@ -410,7 +253,7 @@ static int __init numa_init(int (*init_func)(void)) return 0; out_free_distance: - numa_free_distance(); + numa_reset_distance(); return ret; } @@ -440,6 +283,7 @@ static int __init dummy_numa_init(void) pr_err("NUMA init failed\n"); return ret; } + node_set(0, numa_nodes_parsed); numa_off = true; return 0; @@ -452,7 +296,7 @@ static int __init arch_acpi_numa_init(void) ret = acpi_numa_init(); if (ret) { - pr_info("Failed to initialise from firmware\n"); + pr_debug("Failed to initialise from firmware\n"); return ret; } @@ -482,3 +326,54 @@ void __init arch_numa_init(void) numa_init(dummy_numa_init); } + +#ifdef CONFIG_NUMA_EMU +void __init numa_emu_update_cpu_to_node(int *emu_nid_to_phys, + unsigned int nr_emu_nids) +{ + int i, j; + + /* + * Transform cpu_to_node_map table to use emulated nids by + * reverse-mapping phys_nid. The maps should always exist but fall + * back to zero just in case. + */ + for (i = 0; i < ARRAY_SIZE(cpu_to_node_map); i++) { + if (cpu_to_node_map[i] == NUMA_NO_NODE) + continue; + for (j = 0; j < nr_emu_nids; j++) + if (cpu_to_node_map[i] == emu_nid_to_phys[j]) + break; + cpu_to_node_map[i] = j < nr_emu_nids ? j : 0; + } +} + +u64 __init numa_emu_dma_end(void) +{ + return memblock_start_of_DRAM() + SZ_4G; +} + +void debug_cpumask_set_cpu(unsigned int cpu, int node, bool enable) +{ + struct cpumask *mask; + + if (node == NUMA_NO_NODE) + return; + + mask = node_to_cpumask_map[node]; + if (!cpumask_available(mask)) { + pr_err("node_to_cpumask_map[%i] NULL\n", node); + dump_stack(); + return; + } + + if (enable) + cpumask_set_cpu(cpu, mask); + else + cpumask_clear_cpu(cpu, mask); + + pr_debug("%s cpu %d node %d: mask now %*pbl\n", + enable ? "numa_add_cpu" : "numa_remove_cpu", + cpu, node, cpumask_pr_args(mask)); +} +#endif /* CONFIG_NUMA_EMU */ diff --git a/drivers/base/arch_topology.c b/drivers/base/arch_topology.c index 921312a8d957..84ec92bff642 100644 --- a/drivers/base/arch_topology.c +++ b/drivers/base/arch_topology.c @@ -7,24 +7,30 @@ */ #include <linux/acpi.h> +#include <linux/cacheinfo.h> +#include <linux/cleanup.h> #include <linux/cpu.h> #include <linux/cpufreq.h> +#include <linux/cpu_smt.h> #include <linux/device.h> #include <linux/of.h> #include <linux/slab.h> -#include <linux/string.h> #include <linux/sched/topology.h> #include <linux/cpuset.h> #include <linux/cpumask.h> #include <linux/init.h> -#include <linux/percpu.h> #include <linux/rcupdate.h> #include <linux/sched.h> -#include <linux/smp.h> +#include <linux/units.h> + +#define CREATE_TRACE_POINTS +#include <trace/events/hw_pressure.h> static DEFINE_PER_CPU(struct scale_freq_data __rcu *, sft_data); static struct cpumask scale_freq_counters_mask; static bool scale_freq_invariant; +DEFINE_PER_CPU(unsigned long, capacity_freq_ref) = 0; +EXPORT_PER_CPU_SYMBOL_GPL(capacity_freq_ref); static bool supports_scale_freq_counters(const struct cpumask *cpus) { @@ -148,57 +154,54 @@ void topology_set_freq_scale(const struct cpumask *cpus, unsigned long cur_freq, per_cpu(arch_freq_scale, i) = scale; } -DEFINE_PER_CPU(unsigned long, cpu_scale) = SCHED_CAPACITY_SCALE; +DEFINE_PER_CPU(unsigned long, hw_pressure); -void topology_set_cpu_scale(unsigned int cpu, unsigned long capacity) +/** + * topology_update_hw_pressure() - Update HW pressure for CPUs + * @cpus : The related CPUs for which capacity has been reduced + * @capped_freq : The maximum allowed frequency that CPUs can run at + * + * Update the value of HW pressure for all @cpus in the mask. The + * cpumask should include all (online+offline) affected CPUs, to avoid + * operating on stale data when hot-plug is used for some CPUs. The + * @capped_freq reflects the currently allowed max CPUs frequency due to + * HW capping. It might be also a boost frequency value, which is bigger + * than the internal 'capacity_freq_ref' max frequency. In such case the + * pressure value should simply be removed, since this is an indication that + * there is no HW throttling. The @capped_freq must be provided in kHz. + */ +void topology_update_hw_pressure(const struct cpumask *cpus, + unsigned long capped_freq) { - per_cpu(cpu_scale, cpu) = capacity; -} + unsigned long max_capacity, capacity, pressure; + u32 max_freq; + int cpu; -DEFINE_PER_CPU(unsigned long, thermal_pressure); + cpu = cpumask_first(cpus); + max_capacity = arch_scale_cpu_capacity(cpu); + max_freq = arch_scale_freq_ref(cpu); -void topology_set_thermal_pressure(const struct cpumask *cpus, - unsigned long th_pressure) -{ - int cpu; + /* + * Handle properly the boost frequencies, which should simply clean + * the HW pressure value. + */ + if (max_freq <= capped_freq) + capacity = max_capacity; + else + capacity = mult_frac(max_capacity, capped_freq, max_freq); - for_each_cpu(cpu, cpus) - WRITE_ONCE(per_cpu(thermal_pressure, cpu), th_pressure); -} + pressure = max_capacity - capacity; -static ssize_t cpu_capacity_show(struct device *dev, - struct device_attribute *attr, - char *buf) -{ - struct cpu *cpu = container_of(dev, struct cpu, dev); + trace_hw_pressure_update(cpu, pressure); - return sysfs_emit(buf, "%lu\n", topology_get_cpu_scale(cpu->dev.id)); + for_each_cpu(cpu, cpus) + WRITE_ONCE(per_cpu(hw_pressure, cpu), pressure); } +EXPORT_SYMBOL_GPL(topology_update_hw_pressure); static void update_topology_flags_workfn(struct work_struct *work); static DECLARE_WORK(update_topology_flags_work, update_topology_flags_workfn); -static DEVICE_ATTR_RO(cpu_capacity); - -static int register_cpu_capacity_sysctl(void) -{ - int i; - struct device *cpu; - - for_each_possible_cpu(i) { - cpu = get_cpu_device(i); - if (!cpu) { - pr_err("%s: too early to get CPU%d device!\n", - __func__, i); - continue; - } - device_create_file(cpu, &dev_attr_cpu_capacity); - } - - return 0; -} -subsys_initcall(register_cpu_capacity_sysctl); - static int update_topology; int topology_update_cpu_topology(void) @@ -218,7 +221,6 @@ static void update_topology_flags_workfn(struct work_struct *work) update_topology = 0; } -static DEFINE_PER_CPU(u32, freq_factor) = 1; static u32 *raw_capacity; static int free_raw_capacity(void) @@ -240,13 +242,15 @@ void topology_normalize_cpu_scale(void) capacity_scale = 1; for_each_possible_cpu(cpu) { - capacity = raw_capacity[cpu] * per_cpu(freq_factor, cpu); + capacity = raw_capacity[cpu] * + (per_cpu(capacity_freq_ref, cpu) ?: 1); capacity_scale = max(capacity, capacity_scale); } pr_debug("cpu_capacity: capacity_scale=%llu\n", capacity_scale); for_each_possible_cpu(cpu) { - capacity = raw_capacity[cpu] * per_cpu(freq_factor, cpu); + capacity = raw_capacity[cpu] * + (per_cpu(capacity_freq_ref, cpu) ?: 1); capacity = div64_u64(capacity << SCHED_CAPACITY_SHIFT, capacity_scale); topology_set_cpu_scale(cpu, capacity); @@ -282,15 +286,15 @@ bool __init topology_parse_cpu_capacity(struct device_node *cpu_node, int cpu) cpu_node, raw_capacity[cpu]); /* - * Update freq_factor for calculating early boot cpu capacities. + * Update capacity_freq_ref for calculating early boot CPU capacities. * For non-clk CPU DVFS mechanism, there's no way to get the * frequency value now, assuming they are running at the same - * frequency (by keeping the initial freq_factor value). + * frequency (by keeping the initial capacity_freq_ref value). */ cpu_clk = of_clk_get(cpu_node, 0); - if (!PTR_ERR_OR_ZERO(cpu_clk)) { - per_cpu(freq_factor, cpu) = - clk_get_rate(cpu_clk) / 1000; + if (!IS_ERR_OR_NULL(cpu_clk)) { + per_cpu(capacity_freq_ref, cpu) = + clk_get_rate(cpu_clk) / HZ_PER_KHZ; clk_put(cpu_clk); } } else { @@ -306,6 +310,70 @@ bool __init topology_parse_cpu_capacity(struct device_node *cpu_node, int cpu) return !ret; } +void __weak freq_inv_set_max_ratio(int cpu, u64 max_rate) +{ +} + +#ifdef CONFIG_ACPI_CPPC_LIB +#include <acpi/cppc_acpi.h> + +static inline void topology_init_cpu_capacity_cppc(void) +{ + u64 capacity, capacity_scale = 0; + struct cppc_perf_caps perf_caps; + int cpu; + + if (likely(!acpi_cpc_valid())) + return; + + raw_capacity = kcalloc(num_possible_cpus(), sizeof(*raw_capacity), + GFP_KERNEL); + if (!raw_capacity) + return; + + for_each_possible_cpu(cpu) { + if (!cppc_get_perf_caps(cpu, &perf_caps) && + (perf_caps.highest_perf >= perf_caps.nominal_perf) && + (perf_caps.highest_perf >= perf_caps.lowest_perf)) { + raw_capacity[cpu] = perf_caps.highest_perf; + capacity_scale = max_t(u64, capacity_scale, raw_capacity[cpu]); + + per_cpu(capacity_freq_ref, cpu) = cppc_perf_to_khz(&perf_caps, raw_capacity[cpu]); + + pr_debug("cpu_capacity: CPU%d cpu_capacity=%u (raw).\n", + cpu, raw_capacity[cpu]); + continue; + } + + pr_err("cpu_capacity: CPU%d missing/invalid highest performance.\n", cpu); + pr_err("cpu_capacity: partial information: fallback to 1024 for all CPUs\n"); + goto exit; + } + + for_each_possible_cpu(cpu) { + freq_inv_set_max_ratio(cpu, + per_cpu(capacity_freq_ref, cpu) * HZ_PER_KHZ); + + capacity = raw_capacity[cpu]; + capacity = div64_u64(capacity << SCHED_CAPACITY_SHIFT, + capacity_scale); + topology_set_cpu_scale(cpu, capacity); + pr_debug("cpu_capacity: CPU%d cpu_capacity=%lu\n", + cpu, topology_get_cpu_scale(cpu)); + } + + schedule_work(&update_topology_flags_work); + pr_debug("cpu_capacity: cpu_capacity initialization done\n"); + +exit: + free_raw_capacity(); +} +void acpi_processor_init_invariance_cppc(void) +{ + topology_init_cpu_capacity_cppc(); +} +#endif + #ifdef CONFIG_CPU_FREQ static cpumask_var_t cpus_to_visit; static void parsing_done_workfn(struct work_struct *work); @@ -319,9 +387,6 @@ init_cpu_capacity_callback(struct notifier_block *nb, struct cpufreq_policy *policy = data; int cpu; - if (!raw_capacity) - return 0; - if (val != CPUFREQ_CREATE_POLICY) return 0; @@ -331,13 +396,18 @@ init_cpu_capacity_callback(struct notifier_block *nb, cpumask_andnot(cpus_to_visit, cpus_to_visit, policy->related_cpus); - for_each_cpu(cpu, policy->related_cpus) - per_cpu(freq_factor, cpu) = policy->cpuinfo.max_freq / 1000; + for_each_cpu(cpu, policy->related_cpus) { + per_cpu(capacity_freq_ref, cpu) = policy->cpuinfo.max_freq; + freq_inv_set_max_ratio(cpu, + per_cpu(capacity_freq_ref, cpu) * HZ_PER_KHZ); + } if (cpumask_empty(cpus_to_visit)) { - topology_normalize_cpu_scale(); - schedule_work(&update_topology_flags_work); - free_raw_capacity(); + if (raw_capacity) { + topology_normalize_cpu_scale(); + schedule_work(&update_topology_flags_work); + free_raw_capacity(); + } pr_debug("cpu_capacity: parsing done\n"); schedule_work(&parsing_done_work); } @@ -354,11 +424,10 @@ static int __init register_cpufreq_notifier(void) int ret; /* - * on ACPI-based systems we need to use the default cpu capacity - * until we have the necessary code to parse the cpu capacity, so - * skip registering cpufreq notifier. + * On ACPI-based systems skip registering cpufreq notifier as cpufreq + * information is not needed for cpu capacity initialization. */ - if (!acpi_disabled || !raw_capacity) + if (!acpi_disabled) return -EINVAL; if (!alloc_cpumask_var(&cpus_to_visit, GFP_KERNEL)) @@ -388,6 +457,10 @@ core_initcall(free_raw_capacity); #endif #if defined(CONFIG_ARM64) || defined(CONFIG_RISCV) + +/* Used to enable the SMT control */ +static unsigned int max_smt_thread_num = 1; + /* * This function returns the logic cpu number of the node. * There are basically three kinds of return values: @@ -400,10 +473,10 @@ core_initcall(free_raw_capacity); */ static int __init get_cpu_for_node(struct device_node *node) { - struct device_node *cpu_node; int cpu; + struct device_node *cpu_node __free(device_node) = + of_parse_phandle(node, "cpu", 0); - cpu_node = of_parse_phandle(node, "cpu", 0); if (!cpu_node) return -1; @@ -414,38 +487,40 @@ static int __init get_cpu_for_node(struct device_node *node) pr_info("CPU node for %pOF exist but the possible cpu range is :%*pbl\n", cpu_node, cpumask_pr_args(cpu_possible_mask)); - of_node_put(cpu_node); return cpu; } static int __init parse_core(struct device_node *core, int package_id, - int core_id) + int cluster_id, int core_id) { char name[20]; bool leaf = true; int i = 0; int cpu; - struct device_node *t; do { snprintf(name, sizeof(name), "thread%d", i); - t = of_get_child_by_name(core, name); - if (t) { - leaf = false; - cpu = get_cpu_for_node(t); - if (cpu >= 0) { - cpu_topology[cpu].package_id = package_id; - cpu_topology[cpu].core_id = core_id; - cpu_topology[cpu].thread_id = i; - } else if (cpu != -ENODEV) { - pr_err("%pOF: Can't get CPU for thread\n", t); - of_node_put(t); - return -EINVAL; - } - of_node_put(t); + struct device_node *t __free(device_node) = + of_get_child_by_name(core, name); + + if (!t) + break; + + leaf = false; + cpu = get_cpu_for_node(t); + if (cpu >= 0) { + cpu_topology[cpu].package_id = package_id; + cpu_topology[cpu].cluster_id = cluster_id; + cpu_topology[cpu].core_id = core_id; + cpu_topology[cpu].thread_id = i; + } else if (cpu != -ENODEV) { + pr_err("%pOF: Can't get CPU for thread\n", t); + return -EINVAL; } i++; - } while (t); + } while (1); + + max_smt_thread_num = max_t(unsigned int, max_smt_thread_num, i); cpu = get_cpu_for_node(core); if (cpu >= 0) { @@ -456,6 +531,7 @@ static int __init parse_core(struct device_node *core, int package_id, } cpu_topology[cpu].package_id = package_id; + cpu_topology[cpu].cluster_id = cluster_id; cpu_topology[cpu].core_id = core_id; } else if (leaf && cpu != -ENODEV) { pr_err("%pOF: Can't get CPU for leaf core\n", core); @@ -465,13 +541,12 @@ static int __init parse_core(struct device_node *core, int package_id, return 0; } -static int __init parse_cluster(struct device_node *cluster, int depth) +static int __init parse_cluster(struct device_node *cluster, int package_id, + int cluster_id, int depth) { char name[20]; bool leaf = true; bool has_cores = false; - struct device_node *c; - static int package_id __initdata; int core_id = 0; int i, ret; @@ -483,63 +558,103 @@ static int __init parse_cluster(struct device_node *cluster, int depth) i = 0; do { snprintf(name, sizeof(name), "cluster%d", i); - c = of_get_child_by_name(cluster, name); - if (c) { - leaf = false; - ret = parse_cluster(c, depth + 1); - of_node_put(c); - if (ret != 0) - return ret; - } + struct device_node *c __free(device_node) = + of_get_child_by_name(cluster, name); + + if (!c) + break; + + leaf = false; + ret = parse_cluster(c, package_id, i, depth + 1); + if (depth > 0) + pr_warn("Topology for clusters of clusters not yet supported\n"); + if (ret != 0) + return ret; i++; - } while (c); + } while (1); /* Now check for cores */ i = 0; do { snprintf(name, sizeof(name), "core%d", i); - c = of_get_child_by_name(cluster, name); - if (c) { - has_cores = true; - - if (depth == 0) { - pr_err("%pOF: cpu-map children should be clusters\n", - c); - of_node_put(c); - return -EINVAL; - } + struct device_node *c __free(device_node) = + of_get_child_by_name(cluster, name); - if (leaf) { - ret = parse_core(c, package_id, core_id++); - } else { - pr_err("%pOF: Non-leaf cluster with core %s\n", - cluster, name); - ret = -EINVAL; - } + if (!c) + break; + + has_cores = true; + + if (depth == 0) { + pr_err("%pOF: cpu-map children should be clusters\n", c); + return -EINVAL; + } - of_node_put(c); + if (leaf) { + ret = parse_core(c, package_id, cluster_id, core_id++); if (ret != 0) return ret; + } else { + pr_err("%pOF: Non-leaf cluster with core %s\n", + cluster, name); + return -EINVAL; } + i++; - } while (c); + } while (1); if (leaf && !has_cores) pr_warn("%pOF: empty cluster\n", cluster); - if (leaf) + return 0; +} + +static int __init parse_socket(struct device_node *socket) +{ + char name[20]; + bool has_socket = false; + int package_id = 0, ret; + + do { + snprintf(name, sizeof(name), "socket%d", package_id); + struct device_node *c __free(device_node) = + of_get_child_by_name(socket, name); + + if (!c) + break; + + has_socket = true; + ret = parse_cluster(c, package_id, -1, 0); + if (ret != 0) + return ret; + package_id++; + } while (1); - return 0; + if (!has_socket) + ret = parse_cluster(socket, 0, -1, 0); + + /* + * Reset the max_smt_thread_num to 1 on failure. Since on failure + * we need to notify the framework the SMT is not supported, but + * max_smt_thread_num can be initialized to the SMT thread number + * of the cores which are successfully parsed. + */ + if (ret) + max_smt_thread_num = 1; + + cpu_smt_set_num_threads(max_smt_thread_num, max_smt_thread_num); + + return ret; } static int __init parse_dt_topology(void) { - struct device_node *cn, *map; int ret = 0; int cpu; + struct device_node *cn __free(device_node) = + of_find_node_by_path("/cpus"); - cn = of_find_node_by_path("/cpus"); if (!cn) { pr_err("No CPU information found in DT\n"); return 0; @@ -549,13 +664,15 @@ static int __init parse_dt_topology(void) * When topology is provided cpu-map is essentially a root * cluster with restricted subnodes. */ - map = of_get_child_by_name(cn, "cpu-map"); + struct device_node *map __free(device_node) = + of_get_child_by_name(cn, "cpu-map"); + if (!map) - goto out; + return ret; - ret = parse_cluster(map, 0); + ret = parse_socket(map); if (ret != 0) - goto out_map; + return ret; topology_normalize_cpu_scale(); @@ -564,13 +681,10 @@ static int __init parse_dt_topology(void) * only mark cores described in the DT as possible. */ for_each_possible_cpu(cpu) - if (cpu_topology[cpu].package_id == -1) - ret = -EINVAL; + if (cpu_topology[cpu].package_id < 0) { + return -EINVAL; + } -out_map: - of_node_put(map); -out: - of_node_put(cn); return ret; } #endif @@ -590,24 +704,51 @@ const struct cpumask *cpu_coregroup_mask(int cpu) /* not numa in package, lets use the package siblings */ core_mask = &cpu_topology[cpu].core_sibling; } - if (cpu_topology[cpu].llc_id != -1) { + + if (last_level_cache_is_valid(cpu)) { if (cpumask_subset(&cpu_topology[cpu].llc_sibling, core_mask)) core_mask = &cpu_topology[cpu].llc_sibling; } + /* + * For systems with no shared cpu-side LLC but with clusters defined, + * extend core_mask to cluster_siblings. The sched domain builder will + * then remove MC as redundant with CLS if SCHED_CLUSTER is enabled. + */ + if (IS_ENABLED(CONFIG_SCHED_CLUSTER) && + cpumask_subset(core_mask, &cpu_topology[cpu].cluster_sibling)) + core_mask = &cpu_topology[cpu].cluster_sibling; + return core_mask; } +const struct cpumask *cpu_clustergroup_mask(int cpu) +{ + /* + * Forbid cpu_clustergroup_mask() to span more or the same CPUs as + * cpu_coregroup_mask(). + */ + if (cpumask_subset(cpu_coregroup_mask(cpu), + &cpu_topology[cpu].cluster_sibling)) + return topology_sibling_cpumask(cpu); + + return &cpu_topology[cpu].cluster_sibling; +} + void update_siblings_masks(unsigned int cpuid) { struct cpu_topology *cpu_topo, *cpuid_topo = &cpu_topology[cpuid]; - int cpu; + int cpu, ret; + + ret = detect_cache_attributes(cpuid); + if (ret && ret != -ENOENT) + pr_info("Early cacheinfo allocation failed, ret = %d\n", ret); /* update core and thread sibling masks */ for_each_online_cpu(cpu) { cpu_topo = &cpu_topology[cpu]; - if (cpuid_topo->llc_id == cpu_topo->llc_id) { + if (last_level_cache_is_shared(cpu, cpuid)) { cpumask_set_cpu(cpu, &cpuid_topo->llc_sibling); cpumask_set_cpu(cpuid, &cpu_topo->llc_sibling); } @@ -618,6 +759,14 @@ void update_siblings_masks(unsigned int cpuid) cpumask_set_cpu(cpuid, &cpu_topo->core_sibling); cpumask_set_cpu(cpu, &cpuid_topo->core_sibling); + if (cpuid_topo->cluster_id != cpu_topo->cluster_id) + continue; + + if (cpuid_topo->cluster_id >= 0) { + cpumask_set_cpu(cpu, &cpuid_topo->cluster_sibling); + cpumask_set_cpu(cpuid, &cpu_topo->cluster_sibling); + } + if (cpuid_topo->core_id != cpu_topo->core_id) continue; @@ -633,6 +782,9 @@ static void clear_cpu_topology(int cpu) cpumask_clear(&cpu_topo->llc_sibling); cpumask_set_cpu(cpu, &cpu_topo->llc_sibling); + cpumask_clear(&cpu_topo->cluster_sibling); + cpumask_set_cpu(cpu, &cpu_topo->cluster_sibling); + cpumask_clear(&cpu_topo->core_sibling); cpumask_set_cpu(cpu, &cpu_topo->core_sibling); cpumask_clear(&cpu_topo->thread_sibling); @@ -648,8 +800,8 @@ void __init reset_cpu_topology(void) cpu_topo->thread_id = -1; cpu_topo->core_id = -1; + cpu_topo->cluster_id = -1; cpu_topo->package_id = -1; - cpu_topo->llc_id = -1; clear_cpu_topology(cpu); } @@ -663,29 +815,158 @@ void remove_cpu_topology(unsigned int cpu) cpumask_clear_cpu(cpu, topology_core_cpumask(sibling)); for_each_cpu(sibling, topology_sibling_cpumask(cpu)) cpumask_clear_cpu(cpu, topology_sibling_cpumask(sibling)); + for_each_cpu(sibling, topology_cluster_cpumask(cpu)) + cpumask_clear_cpu(cpu, topology_cluster_cpumask(sibling)); for_each_cpu(sibling, topology_llc_cpumask(cpu)) cpumask_clear_cpu(cpu, topology_llc_cpumask(sibling)); clear_cpu_topology(cpu); } +#if defined(CONFIG_ARM64) || defined(CONFIG_RISCV) +struct cpu_smt_info { + unsigned int thread_num; + int core_id; +}; + +static bool __init acpi_cpu_is_threaded(int cpu) +{ + int is_threaded = acpi_pptt_cpu_is_thread(cpu); + + /* + * if the PPTT doesn't have thread information, check for architecture + * specific fallback if available + */ + if (is_threaded < 0) + is_threaded = arch_cpu_is_threaded(); + + return !!is_threaded; +} + +/* + * Propagate the topology information of the processor_topology_node tree to the + * cpu_topology array. + */ __weak int __init parse_acpi_topology(void) { + unsigned int max_smt_thread_num = 1; + struct cpu_smt_info *entry; + struct xarray hetero_cpu; + unsigned long hetero_id; + int cpu, topology_id; + + if (acpi_disabled) + return 0; + + xa_init(&hetero_cpu); + + for_each_possible_cpu(cpu) { + topology_id = find_acpi_cpu_topology(cpu, 0); + if (topology_id < 0) + return topology_id; + + if (acpi_cpu_is_threaded(cpu)) { + cpu_topology[cpu].thread_id = topology_id; + topology_id = find_acpi_cpu_topology(cpu, 1); + cpu_topology[cpu].core_id = topology_id; + + /* + * In the PPTT, CPUs below a node with the 'identical + * implementation' flag have the same number of threads. + * Count the number of threads for only one CPU (i.e. + * one core_id) among those with the same hetero_id. + * See the comment of find_acpi_cpu_topology_hetero_id() + * for more details. + * + * One entry is created for each node having: + * - the 'identical implementation' flag + * - its parent not having the flag + */ + hetero_id = find_acpi_cpu_topology_hetero_id(cpu); + entry = xa_load(&hetero_cpu, hetero_id); + if (!entry) { + entry = kzalloc(sizeof(*entry), GFP_KERNEL); + WARN_ON_ONCE(!entry); + + if (entry) { + entry->core_id = topology_id; + entry->thread_num = 1; + xa_store(&hetero_cpu, hetero_id, + entry, GFP_KERNEL); + } + } else if (entry->core_id == topology_id) { + entry->thread_num++; + } + } else { + cpu_topology[cpu].thread_id = -1; + cpu_topology[cpu].core_id = topology_id; + } + topology_id = find_acpi_cpu_topology_cluster(cpu); + cpu_topology[cpu].cluster_id = topology_id; + topology_id = find_acpi_cpu_topology_package(cpu); + cpu_topology[cpu].package_id = topology_id; + } + + /* + * This is a short loop since the number of XArray elements is the + * number of heterogeneous CPU clusters. On a homogeneous system + * there's only one entry in the XArray. + */ + xa_for_each(&hetero_cpu, hetero_id, entry) { + max_smt_thread_num = max(max_smt_thread_num, entry->thread_num); + xa_erase(&hetero_cpu, hetero_id); + kfree(entry); + } + + cpu_smt_set_num_threads(max_smt_thread_num, max_smt_thread_num); + xa_destroy(&hetero_cpu); return 0; } -#if defined(CONFIG_ARM64) || defined(CONFIG_RISCV) void __init init_cpu_topology(void) { + int cpu, ret; + reset_cpu_topology(); + ret = parse_acpi_topology(); + if (!ret) + ret = of_have_populated_dt() && parse_dt_topology(); - /* - * Discard anything that was parsed if we hit an error so we - * don't use partial information. - */ - if (parse_acpi_topology()) - reset_cpu_topology(); - else if (of_have_populated_dt() && parse_dt_topology()) + if (ret) { + /* + * Discard anything that was parsed if we hit an error so we + * don't use partial information. But do not return yet to give + * arch-specific early cache level detection a chance to run. + */ reset_cpu_topology(); + } + + for_each_possible_cpu(cpu) { + ret = fetch_cache_info(cpu); + if (!ret) + continue; + else if (ret != -ENOENT) + pr_err("Early cacheinfo failed, ret = %d\n", ret); + return; + } +} + +void store_cpu_topology(unsigned int cpuid) +{ + struct cpu_topology *cpuid_topo = &cpu_topology[cpuid]; + + if (cpuid_topo->package_id != -1) + goto topology_populated; + + cpuid_topo->thread_id = -1; + cpuid_topo->core_id = cpuid; + cpuid_topo->package_id = cpu_to_node(cpuid); + + pr_debug("CPU%u: package %d core %d thread %d\n", + cpuid, cpuid_topo->package_id, cpuid_topo->core_id, + cpuid_topo->thread_id); + +topology_populated: + update_siblings_masks(cpuid); } #endif diff --git a/drivers/base/attribute_container.c b/drivers/base/attribute_container.c index 01ef796c2055..b6f941a6ab69 100644 --- a/drivers/base/attribute_container.c +++ b/drivers/base/attribute_container.c @@ -346,8 +346,7 @@ attribute_container_device_trigger_safe(struct device *dev, * @fn: the function to execute for each classdev. * * This function is for executing a trigger when you need to know both - * the container and the classdev. If you only care about the - * container, then use attribute_container_trigger() instead. + * the container and the classdev. */ void attribute_container_device_trigger(struct device *dev, @@ -379,33 +378,6 @@ attribute_container_device_trigger(struct device *dev, } /** - * attribute_container_trigger - trigger a function for each matching container - * - * @dev: The generic device to activate the trigger for - * @fn: the function to trigger - * - * This routine triggers a function that only needs to know the - * matching containers (not the classdev) associated with a device. - * It is more lightweight than attribute_container_device_trigger, so - * should be used in preference unless the triggering function - * actually needs to know the classdev. - */ -void -attribute_container_trigger(struct device *dev, - int (*fn)(struct attribute_container *, - struct device *)) -{ - struct attribute_container *cont; - - mutex_lock(&attribute_container_mutex); - list_for_each_entry(cont, &attribute_container_list, node) { - if (cont->match(cont, dev)) - fn(cont, dev); - } - mutex_unlock(&attribute_container_mutex); -} - -/** * attribute_container_add_attrs - add attributes * * @classdev: The class device @@ -459,24 +431,6 @@ attribute_container_add_class_device(struct device *classdev) } /** - * attribute_container_add_class_device_adapter - simple adapter for triggers - * - * @cont: the container to register. - * @dev: the generic device to activate the trigger for - * @classdev: the class device to add - * - * This function is identical to attribute_container_add_class_device except - * that it is designed to be called from the triggers - */ -int -attribute_container_add_class_device_adapter(struct attribute_container *cont, - struct device *dev, - struct device *classdev) -{ - return attribute_container_add_class_device(classdev); -} - -/** * attribute_container_remove_attrs - remove any attribute files * * @classdev: The class device to remove the files from diff --git a/drivers/base/auxiliary.c b/drivers/base/auxiliary.c index adc199dfba3c..04bdbff4dbe5 100644 --- a/drivers/base/auxiliary.c +++ b/drivers/base/auxiliary.c @@ -17,34 +17,186 @@ #include <linux/auxiliary_bus.h> #include "base.h" +/** + * DOC: PURPOSE + * + * In some subsystems, the functionality of the core device (PCI/ACPI/other) is + * too complex for a single device to be managed by a monolithic driver (e.g. + * Sound Open Firmware), multiple devices might implement a common intersection + * of functionality (e.g. NICs + RDMA), or a driver may want to export an + * interface for another subsystem to drive (e.g. SIOV Physical Function export + * Virtual Function management). A split of the functionality into child- + * devices representing sub-domains of functionality makes it possible to + * compartmentalize, layer, and distribute domain-specific concerns via a Linux + * device-driver model. + * + * An example for this kind of requirement is the audio subsystem where a + * single IP is handling multiple entities such as HDMI, Soundwire, local + * devices such as mics/speakers etc. The split for the core's functionality + * can be arbitrary or be defined by the DSP firmware topology and include + * hooks for test/debug. This allows for the audio core device to be minimal + * and focused on hardware-specific control and communication. + * + * Each auxiliary_device represents a part of its parent functionality. The + * generic behavior can be extended and specialized as needed by encapsulating + * an auxiliary_device within other domain-specific structures and the use of + * .ops callbacks. Devices on the auxiliary bus do not share any structures and + * the use of a communication channel with the parent is domain-specific. + * + * Note that ops are intended as a way to augment instance behavior within a + * class of auxiliary devices, it is not the mechanism for exporting common + * infrastructure from the parent. Consider EXPORT_SYMBOL_NS() to convey + * infrastructure from the parent module to the auxiliary module(s). + */ + +/** + * DOC: USAGE + * + * The auxiliary bus is to be used when a driver and one or more kernel + * modules, who share a common header file with the driver, need a mechanism to + * connect and provide access to a shared object allocated by the + * auxiliary_device's registering driver. The registering driver for the + * auxiliary_device(s) and the kernel module(s) registering auxiliary_drivers + * can be from the same subsystem, or from multiple subsystems. + * + * The emphasis here is on a common generic interface that keeps subsystem + * customization out of the bus infrastructure. + * + * One example is a PCI network device that is RDMA-capable and exports a child + * device to be driven by an auxiliary_driver in the RDMA subsystem. The PCI + * driver allocates and registers an auxiliary_device for each physical + * function on the NIC. The RDMA driver registers an auxiliary_driver that + * claims each of these auxiliary_devices. This conveys data/ops published by + * the parent PCI device/driver to the RDMA auxiliary_driver. + * + * Another use case is for the PCI device to be split out into multiple sub + * functions. For each sub function an auxiliary_device is created. A PCI sub + * function driver binds to such devices that creates its own one or more class + * devices. A PCI sub function auxiliary device is likely to be contained in a + * struct with additional attributes such as user defined sub function number + * and optional attributes such as resources and a link to the parent device. + * These attributes could be used by systemd/udev; and hence should be + * initialized before a driver binds to an auxiliary_device. + * + * A key requirement for utilizing the auxiliary bus is that there is no + * dependency on a physical bus, device, register accesses or regmap support. + * These individual devices split from the core cannot live on the platform bus + * as they are not physical devices that are controlled by DT/ACPI. The same + * argument applies for not using MFD in this scenario as MFD relies on + * individual function devices being physical devices. + */ + +/** + * DOC: EXAMPLE + * + * Auxiliary devices are created and registered by a subsystem-level core + * device that needs to break up its functionality into smaller fragments. One + * way to extend the scope of an auxiliary_device is to encapsulate it within a + * domain-specific structure defined by the parent device. This structure + * contains the auxiliary_device and any associated shared data/callbacks + * needed to establish the connection with the parent. + * + * An example is: + * + * .. code-block:: c + * + * struct foo { + * struct auxiliary_device auxdev; + * void (*connect)(struct auxiliary_device *auxdev); + * void (*disconnect)(struct auxiliary_device *auxdev); + * void *data; + * }; + * + * The parent device then registers the auxiliary_device by calling + * auxiliary_device_init(), and then auxiliary_device_add(), with the pointer + * to the auxdev member of the above structure. The parent provides a name for + * the auxiliary_device that, combined with the parent's KBUILD_MODNAME, + * creates a match_name that is be used for matching and binding with a driver. + * + * Whenever an auxiliary_driver is registered, based on the match_name, the + * auxiliary_driver's probe() is invoked for the matching devices. The + * auxiliary_driver can also be encapsulated inside custom drivers that make + * the core device's functionality extensible by adding additional + * domain-specific ops as follows: + * + * .. code-block:: c + * + * struct my_ops { + * void (*send)(struct auxiliary_device *auxdev); + * void (*receive)(struct auxiliary_device *auxdev); + * }; + * + * + * struct my_driver { + * struct auxiliary_driver auxiliary_drv; + * const struct my_ops ops; + * }; + * + * An example of this type of usage is: + * + * .. code-block:: c + * + * const struct auxiliary_device_id my_auxiliary_id_table[] = { + * { .name = "foo_mod.foo_dev" }, + * { }, + * }; + * + * const struct my_ops my_custom_ops = { + * .send = my_tx, + * .receive = my_rx, + * }; + * + * const struct my_driver my_drv = { + * .auxiliary_drv = { + * .name = "myauxiliarydrv", + * .id_table = my_auxiliary_id_table, + * .probe = my_probe, + * .remove = my_remove, + * .shutdown = my_shutdown, + * }, + * .ops = my_custom_ops, + * }; + * + * Please note that such custom ops approach is valid, but it is hard to implement + * it right without global locks per-device to protect from auxiliary_drv removal + * during call to that ops. In addition, this implementation lacks proper module + * dependency, which causes to load/unload races between auxiliary parent and devices + * modules. + * + * The most easiest way to provide these ops reliably without needing to + * have a lock is to EXPORT_SYMBOL*() them and rely on already existing + * modules infrastructure for validity and correct dependencies chains. + */ + static const struct auxiliary_device_id *auxiliary_match_id(const struct auxiliary_device_id *id, const struct auxiliary_device *auxdev) { - for (; id->name[0]; id++) { - const char *p = strrchr(dev_name(&auxdev->dev), '.'); - int match_size; + const char *auxdev_name = dev_name(&auxdev->dev); + const char *p = strrchr(auxdev_name, '.'); + int match_size; - if (!p) - continue; - match_size = p - dev_name(&auxdev->dev); + if (!p) + return NULL; + match_size = p - auxdev_name; + for (; id->name[0]; id++) { /* use dev_name(&auxdev->dev) prefix before last '.' char to match to */ if (strlen(id->name) == match_size && - !strncmp(dev_name(&auxdev->dev), id->name, match_size)) + !strncmp(auxdev_name, id->name, match_size)) return id; } return NULL; } -static int auxiliary_match(struct device *dev, struct device_driver *drv) +static int auxiliary_match(struct device *dev, const struct device_driver *drv) { struct auxiliary_device *auxdev = to_auxiliary_dev(dev); - struct auxiliary_driver *auxdrv = to_auxiliary_drv(drv); + const struct auxiliary_driver *auxdrv = to_auxiliary_drv(drv); return !!auxiliary_match_id(auxdrv->id_table, auxdev); } -static int auxiliary_uevent(struct device *dev, struct kobj_uevent_env *env) +static int auxiliary_uevent(const struct device *dev, struct kobj_uevent_env *env) { const char *name, *p; @@ -62,38 +214,32 @@ static const struct dev_pm_ops auxiliary_dev_pm_ops = { static int auxiliary_bus_probe(struct device *dev) { - struct auxiliary_driver *auxdrv = to_auxiliary_drv(dev->driver); + const struct auxiliary_driver *auxdrv = to_auxiliary_drv(dev->driver); struct auxiliary_device *auxdev = to_auxiliary_dev(dev); int ret; - ret = dev_pm_domain_attach(dev, true); + ret = dev_pm_domain_attach(dev, PD_FLAG_ATTACH_POWER_ON | + PD_FLAG_DETACH_POWER_OFF); if (ret) { dev_warn(dev, "Failed to attach to PM Domain : %d\n", ret); return ret; } - ret = auxdrv->probe(auxdev, auxiliary_match_id(auxdrv->id_table, auxdev)); - if (ret) - dev_pm_domain_detach(dev, true); - - return ret; + return auxdrv->probe(auxdev, auxiliary_match_id(auxdrv->id_table, auxdev)); } -static int auxiliary_bus_remove(struct device *dev) +static void auxiliary_bus_remove(struct device *dev) { - struct auxiliary_driver *auxdrv = to_auxiliary_drv(dev->driver); + const struct auxiliary_driver *auxdrv = to_auxiliary_drv(dev->driver); struct auxiliary_device *auxdev = to_auxiliary_dev(dev); if (auxdrv->remove) auxdrv->remove(auxdev); - dev_pm_domain_detach(dev, true); - - return 0; } static void auxiliary_bus_shutdown(struct device *dev) { - struct auxiliary_driver *auxdrv = NULL; + const struct auxiliary_driver *auxdrv = NULL; struct auxiliary_device *auxdev; if (dev->driver) { @@ -105,7 +251,7 @@ static void auxiliary_bus_shutdown(struct device *dev) auxdrv->shutdown(auxdev); } -static struct bus_type auxiliary_bus_type = { +static const struct bus_type auxiliary_bus_type = { .name = "auxiliary", .probe = auxiliary_bus_probe, .remove = auxiliary_bus_remove, @@ -119,7 +265,7 @@ static struct bus_type auxiliary_bus_type = { * auxiliary_device_init - check auxiliary_device and initialize * @auxdev: auxiliary device struct * - * This is the first step in the two-step process to register an + * This is the second step in the three-step process to register an * auxiliary_device. * * When this function returns an error code, then the device_initialize will @@ -148,6 +294,7 @@ int auxiliary_device_init(struct auxiliary_device *auxdev) dev->bus = &auxiliary_bus_type; device_initialize(&auxdev->dev); + mutex_init(&auxdev->sysfs.lock); return 0; } EXPORT_SYMBOL_GPL(auxiliary_device_init); @@ -157,7 +304,7 @@ EXPORT_SYMBOL_GPL(auxiliary_device_init); * @auxdev: auxiliary bus device to add to the bus * @modname: name of the parent device's driver module * - * This is the second step in the two-step process to register an + * This is the third step in the three-step process to register an * auxiliary_device. * * This function must be called after a successful call to @@ -196,41 +343,21 @@ int __auxiliary_device_add(struct auxiliary_device *auxdev, const char *modname) EXPORT_SYMBOL_GPL(__auxiliary_device_add); /** - * auxiliary_find_device - auxiliary device iterator for locating a particular device. - * @start: Device to begin with - * @data: Data to pass to match function - * @match: Callback function to check device - * - * This function returns a reference to a device that is 'found' - * for later use, as determined by the @match callback. - * - * The callback should return 0 if the device doesn't match and non-zero - * if it does. If the callback returns non-zero, this function will - * return to the caller and not iterate over any more devices. - */ -struct auxiliary_device *auxiliary_find_device(struct device *start, - const void *data, - int (*match)(struct device *dev, const void *data)) -{ - struct device *dev; - - dev = bus_find_device(&auxiliary_bus_type, start, data, match); - if (!dev) - return NULL; - - return to_auxiliary_dev(dev); -} -EXPORT_SYMBOL_GPL(auxiliary_find_device); - -/** * __auxiliary_driver_register - register a driver for auxiliary bus devices * @auxdrv: auxiliary_driver structure * @owner: owning module/driver * @modname: KBUILD_MODNAME for parent driver + * + * The expectation is that users will call the "auxiliary_driver_register" + * macro so that the caller's KBUILD_MODNAME is automatically inserted for the + * modname parameter. Only if a user requires a custom name would this version + * be called directly. */ int __auxiliary_driver_register(struct auxiliary_driver *auxdrv, struct module *owner, const char *modname) { + int ret; + if (WARN_ON(!auxdrv->probe) || WARN_ON(!auxdrv->id_table)) return -EINVAL; @@ -246,7 +373,11 @@ int __auxiliary_driver_register(struct auxiliary_driver *auxdrv, auxdrv->driver.bus = &auxiliary_bus_type; auxdrv->driver.mod_name = modname; - return driver_register(&auxdrv->driver); + ret = driver_register(&auxdrv->driver); + if (ret) + kfree(auxdrv->driver.name); + + return ret; } EXPORT_SYMBOL_GPL(__auxiliary_driver_register); @@ -261,6 +392,116 @@ void auxiliary_driver_unregister(struct auxiliary_driver *auxdrv) } EXPORT_SYMBOL_GPL(auxiliary_driver_unregister); +static void auxiliary_device_release(struct device *dev) +{ + struct auxiliary_device *auxdev = to_auxiliary_dev(dev); + + of_node_put(dev->of_node); + kfree(auxdev); +} + +/** + * auxiliary_device_create - create a device on the auxiliary bus + * @dev: parent device + * @modname: module name used to create the auxiliary driver name. + * @devname: auxiliary bus device name + * @platform_data: auxiliary bus device platform data + * @id: auxiliary bus device id + * + * Helper to create an auxiliary bus device. + * The device created matches driver 'modname.devname' on the auxiliary bus. + */ +struct auxiliary_device *auxiliary_device_create(struct device *dev, + const char *modname, + const char *devname, + void *platform_data, + int id) +{ + struct auxiliary_device *auxdev; + int ret; + + auxdev = kzalloc(sizeof(*auxdev), GFP_KERNEL); + if (!auxdev) + return NULL; + + auxdev->id = id; + auxdev->name = devname; + auxdev->dev.parent = dev; + auxdev->dev.platform_data = platform_data; + auxdev->dev.release = auxiliary_device_release; + device_set_of_node_from_dev(&auxdev->dev, dev); + + ret = auxiliary_device_init(auxdev); + if (ret) { + of_node_put(auxdev->dev.of_node); + kfree(auxdev); + return NULL; + } + + ret = __auxiliary_device_add(auxdev, modname); + if (ret) { + /* + * It may look odd but auxdev should not be freed here. + * auxiliary_device_uninit() calls device_put() which call + * the device release function, freeing auxdev. + */ + auxiliary_device_uninit(auxdev); + return NULL; + } + + return auxdev; +} +EXPORT_SYMBOL_GPL(auxiliary_device_create); + +/** + * auxiliary_device_destroy - remove an auxiliary device + * @auxdev: pointer to the auxdev to be removed + * + * Helper to remove an auxiliary device created with + * auxiliary_device_create() + */ +void auxiliary_device_destroy(void *auxdev) +{ + struct auxiliary_device *_auxdev = auxdev; + + auxiliary_device_delete(_auxdev); + auxiliary_device_uninit(_auxdev); +} +EXPORT_SYMBOL_GPL(auxiliary_device_destroy); + +/** + * __devm_auxiliary_device_create - create a managed device on the auxiliary bus + * @dev: parent device + * @modname: module name used to create the auxiliary driver name. + * @devname: auxiliary bus device name + * @platform_data: auxiliary bus device platform data + * @id: auxiliary bus device id + * + * Device managed helper to create an auxiliary bus device. + * The device created matches driver 'modname.devname' on the auxiliary bus. + */ +struct auxiliary_device *__devm_auxiliary_device_create(struct device *dev, + const char *modname, + const char *devname, + void *platform_data, + int id) +{ + struct auxiliary_device *auxdev; + int ret; + + auxdev = auxiliary_device_create(dev, modname, devname, platform_data, id); + if (!auxdev) + return NULL; + + ret = devm_add_action_or_reset(dev, auxiliary_device_destroy, + auxdev); + if (ret) + return NULL; + + return auxdev; +} +EXPORT_SYMBOL_GPL(__devm_auxiliary_device_create); + void __init auxiliary_bus_init(void) { WARN_ON(bus_register(&auxiliary_bus_type)); diff --git a/drivers/base/auxiliary_sysfs.c b/drivers/base/auxiliary_sysfs.c new file mode 100644 index 000000000000..754f21730afd --- /dev/null +++ b/drivers/base/auxiliary_sysfs.c @@ -0,0 +1,113 @@ +// SPDX-License-Identifier: GPL-2.0 +/* + * Copyright (c) 2024, NVIDIA CORPORATION & AFFILIATES + */ + +#include <linux/auxiliary_bus.h> +#include <linux/slab.h> + +#define AUXILIARY_MAX_IRQ_NAME 11 + +struct auxiliary_irq_info { + struct device_attribute sysfs_attr; + char name[AUXILIARY_MAX_IRQ_NAME]; +}; + +static struct attribute *auxiliary_irq_attrs[] = { + NULL +}; + +static const struct attribute_group auxiliary_irqs_group = { + .name = "irqs", + .attrs = auxiliary_irq_attrs, +}; + +static int auxiliary_irq_dir_prepare(struct auxiliary_device *auxdev) +{ + int ret = 0; + + guard(mutex)(&auxdev->sysfs.lock); + if (auxdev->sysfs.irq_dir_exists) + return 0; + + ret = devm_device_add_group(&auxdev->dev, &auxiliary_irqs_group); + if (ret) + return ret; + + auxdev->sysfs.irq_dir_exists = true; + xa_init(&auxdev->sysfs.irqs); + return 0; +} + +/** + * auxiliary_device_sysfs_irq_add - add a sysfs entry for the given IRQ + * @auxdev: auxiliary bus device to add the sysfs entry. + * @irq: The associated interrupt number. + * + * This function should be called after auxiliary device have successfully + * received the irq. + * The driver is responsible to add a unique irq for the auxiliary device. The + * driver can invoke this function from multiple thread context safely for + * unique irqs of the auxiliary devices. The driver must not invoke this API + * multiple times if the irq is already added previously. + * + * Return: zero on success or an error code on failure. + */ +int auxiliary_device_sysfs_irq_add(struct auxiliary_device *auxdev, int irq) +{ + struct auxiliary_irq_info *info __free(kfree) = NULL; + struct device *dev = &auxdev->dev; + int ret; + + ret = auxiliary_irq_dir_prepare(auxdev); + if (ret) + return ret; + + info = kzalloc(sizeof(*info), GFP_KERNEL); + if (!info) + return -ENOMEM; + + sysfs_attr_init(&info->sysfs_attr.attr); + snprintf(info->name, AUXILIARY_MAX_IRQ_NAME, "%d", irq); + + ret = xa_insert(&auxdev->sysfs.irqs, irq, info, GFP_KERNEL); + if (ret) + return ret; + + info->sysfs_attr.attr.name = info->name; + ret = sysfs_add_file_to_group(&dev->kobj, &info->sysfs_attr.attr, + auxiliary_irqs_group.name); + if (ret) + goto sysfs_add_err; + + xa_store(&auxdev->sysfs.irqs, irq, no_free_ptr(info), GFP_KERNEL); + return 0; + +sysfs_add_err: + xa_erase(&auxdev->sysfs.irqs, irq); + return ret; +} +EXPORT_SYMBOL_GPL(auxiliary_device_sysfs_irq_add); + +/** + * auxiliary_device_sysfs_irq_remove - remove a sysfs entry for the given IRQ + * @auxdev: auxiliary bus device to add the sysfs entry. + * @irq: the IRQ to remove. + * + * This function should be called to remove an IRQ sysfs entry. + * The driver must invoke this API when IRQ is released by the device. + */ +void auxiliary_device_sysfs_irq_remove(struct auxiliary_device *auxdev, int irq) +{ + struct auxiliary_irq_info *info __free(kfree) = xa_load(&auxdev->sysfs.irqs, irq); + struct device *dev = &auxdev->dev; + + if (!info) { + dev_err(&auxdev->dev, "IRQ %d doesn't exist\n", irq); + return; + } + sysfs_remove_file_from_group(&dev->kobj, &info->sysfs_attr.attr, + auxiliary_irqs_group.name); + xa_erase(&auxdev->sysfs.irqs, irq); +} +EXPORT_SYMBOL_GPL(auxiliary_device_sysfs_irq_remove); diff --git a/drivers/base/base.h b/drivers/base/base.h index 404db83ee5ec..430cbefbc97f 100644 --- a/drivers/base/base.h +++ b/drivers/base/base.h @@ -27,11 +27,13 @@ * on this bus. * @bus - pointer back to the struct bus_type that this structure is associated * with. + * @dev_root: Default device to use as the parent. * * @glue_dirs - "glue" directory to put in-between the parent device to * avoid namespace conflicts * @class - pointer back to the struct class that this structure is associated * with. + * @lock_key: Lock class key for use by the lock validator * * This structure is the one that is the actual kobject allowing struct * bus_type/class to be statically allocated safely. Nothing outside of the @@ -48,12 +50,31 @@ struct subsys_private { struct klist klist_drivers; struct blocking_notifier_head bus_notifier; unsigned int drivers_autoprobe:1; - struct bus_type *bus; + const struct bus_type *bus; + struct device *dev_root; struct kset glue_dirs; - struct class *class; + const struct class *class; + + struct lock_class_key lock_key; }; -#define to_subsys_private(obj) container_of(obj, struct subsys_private, subsys.kobj) +#define to_subsys_private(obj) container_of_const(obj, struct subsys_private, subsys.kobj) + +static inline struct subsys_private *subsys_get(struct subsys_private *sp) +{ + if (sp) + kset_get(&sp->subsys); + return sp; +} + +static inline void subsys_put(struct subsys_private *sp) +{ + if (sp) + kset_put(&sp->subsys); +} + +struct subsys_private *bus_to_subsys(const struct bus_type *bus); +struct subsys_private *class_to_subsys(const struct class *class); struct driver_private { struct kobject kobj; @@ -64,6 +85,18 @@ struct driver_private { }; #define to_driver(obj) container_of(obj, struct driver_private, kobj) +#ifdef CONFIG_RUST +/** + * struct driver_type - Representation of a Rust driver type. + */ +struct driver_type { + /** + * @id: Representation of core::any::TypeId. + */ + u8 id[16]; +} __packed; +#endif + /** * struct device_private - structure to hold the private to the driver core portions of the device structure. * @@ -79,6 +112,7 @@ struct driver_private { * @async_driver - pointer to device driver awaiting probe via async_probe * @device - pointer back to the struct device that this structure is * associated with. + * @driver_type - The type of the bound Rust driver. * @dead - This device is currently either in the process of or has been * removed from the system. Any asynchronous events scheduled for this * device should exit without taking any action. @@ -92,9 +126,12 @@ struct device_private { struct klist_node knode_bus; struct klist_node knode_class; struct list_head deferred_probe; - struct device_driver *async_driver; + const struct device_driver *async_driver; char *deferred_probe_reason; struct device *device; +#ifdef CONFIG_RUST + struct driver_type driver_type; +#endif u8 dead:1; }; #define to_device_private_parent(obj) \ @@ -107,90 +144,137 @@ struct device_private { container_of(obj, struct device_private, knode_class) /* initialisation functions */ -extern int devices_init(void); -extern int buses_init(void); -extern int classes_init(void); -extern int firmware_init(void); +int devices_init(void); +int buses_init(void); +int classes_init(void); +int firmware_init(void); #ifdef CONFIG_SYS_HYPERVISOR -extern int hypervisor_init(void); +int hypervisor_init(void); #else static inline int hypervisor_init(void) { return 0; } #endif -extern int platform_bus_init(void); -extern void cpu_dev_init(void); -extern void container_dev_init(void); +int platform_bus_init(void); +int faux_bus_init(void); +void cpu_dev_init(void); +void container_dev_init(void); #ifdef CONFIG_AUXILIARY_BUS -extern void auxiliary_bus_init(void); +void auxiliary_bus_init(void); #else static inline void auxiliary_bus_init(void) { } #endif -struct kobject *virtual_device_parent(struct device *dev); +struct kobject *virtual_device_parent(void); -extern int bus_add_device(struct device *dev); -extern void bus_probe_device(struct device *dev); -extern void bus_remove_device(struct device *dev); +int bus_add_device(struct device *dev); +void bus_probe_device(struct device *dev); +void bus_remove_device(struct device *dev); +void bus_notify(struct device *dev, enum bus_notifier_event value); +bool bus_is_registered(const struct bus_type *bus); -extern int bus_add_driver(struct device_driver *drv); -extern void bus_remove_driver(struct device_driver *drv); -extern void device_release_driver_internal(struct device *dev, - struct device_driver *drv, - struct device *parent); +int bus_add_driver(struct device_driver *drv); +void bus_remove_driver(struct device_driver *drv); +void device_release_driver_internal(struct device *dev, const struct device_driver *drv, + struct device *parent); -extern void driver_detach(struct device_driver *drv); -extern void driver_deferred_probe_del(struct device *dev); -extern void device_set_deferred_probe_reason(const struct device *dev, - struct va_format *vaf); -static inline int driver_match_device(struct device_driver *drv, +void driver_detach(const struct device_driver *drv); +void driver_deferred_probe_del(struct device *dev); +void device_set_deferred_probe_reason(const struct device *dev, struct va_format *vaf); +static inline int driver_match_device(const struct device_driver *drv, struct device *dev) { return drv->bus->match ? drv->bus->match(dev, drv) : 1; } -extern bool driver_allows_async_probing(struct device_driver *drv); -extern int driver_add_groups(struct device_driver *drv, - const struct attribute_group **groups); -extern void driver_remove_groups(struct device_driver *drv, - const struct attribute_group **groups); +static inline void dev_sync_state(struct device *dev) +{ + if (dev->bus->sync_state) + dev->bus->sync_state(dev); + else if (dev->driver && dev->driver->sync_state) + dev->driver->sync_state(dev); +} + +int driver_add_groups(const struct device_driver *drv, const struct attribute_group **groups); +void driver_remove_groups(const struct device_driver *drv, const struct attribute_group **groups); void device_driver_detach(struct device *dev); -extern char *make_class_name(const char *name, struct kobject *kobj); +static inline void device_set_driver(struct device *dev, const struct device_driver *drv) +{ + /* + * Majority (all?) read accesses to dev->driver happens either + * while holding device lock or in bus/driver code that is only + * invoked when the device is bound to a driver and there is no + * concern of the pointer being changed while it is being read. + * However when reading device's uevent file we read driver pointer + * without taking device lock (so we do not block there for + * arbitrary amount of time). We use WRITE_ONCE() here to prevent + * tearing so that READ_ONCE() can safely be used in uevent code. + */ + // FIXME - this cast should not be needed "soon" + WRITE_ONCE(dev->driver, (struct device_driver *)drv); +} -extern int devres_release_all(struct device *dev); -extern void device_block_probing(void); -extern void device_unblock_probing(void); +int devres_release_all(struct device *dev); +void device_block_probing(void); +void device_unblock_probing(void); +void deferred_probe_extend_timeout(void); +void driver_deferred_probe_trigger(void); +const char *device_get_devnode(const struct device *dev, umode_t *mode, + kuid_t *uid, kgid_t *gid, const char **tmp); /* /sys/devices directory */ extern struct kset *devices_kset; -extern void devices_kset_move_last(struct device *dev); +void devices_kset_move_last(struct device *dev); #if defined(CONFIG_MODULES) && defined(CONFIG_SYSFS) -extern void module_add_driver(struct module *mod, struct device_driver *drv); -extern void module_remove_driver(struct device_driver *drv); +int module_add_driver(struct module *mod, const struct device_driver *drv); +void module_remove_driver(const struct device_driver *drv); #else -static inline void module_add_driver(struct module *mod, - struct device_driver *drv) { } +static inline int module_add_driver(struct module *mod, + struct device_driver *drv) +{ + return 0; +} static inline void module_remove_driver(struct device_driver *drv) { } #endif #ifdef CONFIG_DEVTMPFS -extern int devtmpfs_init(void); +int devtmpfs_init(void); #else static inline int devtmpfs_init(void) { return 0; } #endif +#ifdef CONFIG_BLOCK +extern const struct class block_class; +static inline bool is_blockdev(struct device *dev) +{ + return dev->class == &block_class; +} +#else +static inline bool is_blockdev(struct device *dev) { return false; } +#endif + /* Device links support */ -extern int device_links_read_lock(void); -extern void device_links_read_unlock(int idx); -extern int device_links_read_lock_held(void); -extern int device_links_check_suppliers(struct device *dev); -extern void device_links_force_bind(struct device *dev); -extern void device_links_driver_bound(struct device *dev); -extern void device_links_driver_cleanup(struct device *dev); -extern void device_links_no_driver(struct device *dev); -extern bool device_links_busy(struct device *dev); -extern void device_links_unbind_consumers(struct device *dev); -extern void fw_devlink_drivers_done(void); +int device_links_read_lock(void); +void device_links_read_unlock(int idx); +int device_links_read_lock_held(void); +int device_links_check_suppliers(struct device *dev); +void device_links_force_bind(struct device *dev); +void device_links_driver_bound(struct device *dev); +void device_links_driver_cleanup(struct device *dev); +void device_links_no_driver(struct device *dev); +bool device_links_busy(struct device *dev); +void device_links_unbind_consumers(struct device *dev); +bool device_link_flag_is_sync_state_only(u32 flags); +void fw_devlink_drivers_done(void); +void fw_devlink_probing_done(void); + +#define dev_for_each_link_to_supplier(__link, __dev) \ + list_for_each_entry_srcu(__link, &(__dev)->links.suppliers, c_node, \ + device_links_read_lock_held()) + +#define dev_for_each_link_to_consumer(__link, __dev) \ + list_for_each_entry_srcu(__link, &(__dev)->links.consumers, s_node, \ + device_links_read_lock_held()) /* device pm support */ void device_pm_move_to_tail(struct device *dev); @@ -202,3 +286,6 @@ int devtmpfs_delete_node(struct device *dev); static inline int devtmpfs_create_node(struct device *dev) { return 0; } static inline int devtmpfs_delete_node(struct device *dev) { return 0; } #endif + +void software_node_notify(struct device *dev); +void software_node_notify_remove(struct device *dev); diff --git a/drivers/base/bus.c b/drivers/base/bus.c index 1f6b4bd61056..9eb7771706f0 100644 --- a/drivers/base/bus.c +++ b/drivers/base/bus.c @@ -6,6 +6,7 @@ * Copyright (c) 2002-3 Open Source Development Labs * Copyright (c) 2007 Greg Kroah-Hartman <gregkh@suse.de> * Copyright (c) 2007 Novell Inc. + * Copyright (c) 2023 Greg Kroah-Hartman <gregkh@linuxfoundation.org> */ #include <linux/async.h> @@ -24,6 +25,9 @@ /* /sys/devices/system */ static struct kset *system_kset; +/* /sys/bus */ +static struct kset *bus_kset; + #define to_bus_attr(_attr) container_of(_attr, struct bus_attribute, attr) /* @@ -39,19 +43,63 @@ static struct kset *system_kset; static int __must_check bus_rescan_devices_helper(struct device *dev, void *data); -static struct bus_type *bus_get(struct bus_type *bus) +/** + * bus_to_subsys - Turn a struct bus_type into a struct subsys_private + * + * @bus: pointer to the struct bus_type to look up + * + * The driver core internals needs to work on the subsys_private structure, not + * the external struct bus_type pointer. This function walks the list of + * registered busses in the system and finds the matching one and returns the + * internal struct subsys_private that relates to that bus. + * + * Note, the reference count of the return value is INCREMENTED if it is not + * NULL. A call to subsys_put() must be done when finished with the pointer in + * order for it to be properly freed. + */ +struct subsys_private *bus_to_subsys(const struct bus_type *bus) { - if (bus) { - kset_get(&bus->p->subsys); - return bus; + struct subsys_private *sp = NULL; + struct kobject *kobj; + + if (!bus || !bus_kset) + return NULL; + + spin_lock(&bus_kset->list_lock); + + if (list_empty(&bus_kset->list)) + goto done; + + list_for_each_entry(kobj, &bus_kset->list, entry) { + struct kset *kset = container_of(kobj, struct kset, kobj); + + sp = container_of_const(kset, struct subsys_private, subsys); + if (sp->bus == bus) + goto done; } + sp = NULL; +done: + sp = subsys_get(sp); + spin_unlock(&bus_kset->list_lock); + return sp; +} + +static const struct bus_type *bus_get(const struct bus_type *bus) +{ + struct subsys_private *sp = bus_to_subsys(bus); + + if (sp) + return bus; return NULL; } -static void bus_put(struct bus_type *bus) +static void bus_put(const struct bus_type *bus) { - if (bus) - kset_put(&bus->p->subsys); + struct subsys_private *sp = bus_to_subsys(bus); + + /* two puts are required as the call to bus_to_subsys incremented it again */ + subsys_put(sp); + subsys_put(sp); } static ssize_t drv_attr_show(struct kobject *kobj, struct attribute *attr, @@ -91,7 +139,7 @@ static void driver_release(struct kobject *kobj) kfree(drv_priv); } -static struct kobj_type driver_ktype = { +static const struct kobj_type driver_ktype = { .sysfs_ops = &driver_sysfs_ops, .release = driver_release, }; @@ -104,7 +152,8 @@ static ssize_t bus_attr_show(struct kobject *kobj, struct attribute *attr, { struct bus_attribute *bus_attr = to_bus_attr(attr); struct subsys_private *subsys_priv = to_subsys_private(kobj); - ssize_t ret = 0; + /* return -EIO for reading a bus attribute without show() */ + ssize_t ret = -EIO; if (bus_attr->show) ret = bus_attr->show(subsys_priv->bus, buf); @@ -116,7 +165,8 @@ static ssize_t bus_attr_store(struct kobject *kobj, struct attribute *attr, { struct bus_attribute *bus_attr = to_bus_attr(attr); struct subsys_private *subsys_priv = to_subsys_private(kobj); - ssize_t ret = 0; + /* return -EIO for writing a bus attribute without store() */ + ssize_t ret = -EIO; if (bus_attr->store) ret = bus_attr->store(subsys_priv->bus, buf, count); @@ -128,44 +178,49 @@ static const struct sysfs_ops bus_sysfs_ops = { .store = bus_attr_store, }; -int bus_create_file(struct bus_type *bus, struct bus_attribute *attr) +int bus_create_file(const struct bus_type *bus, struct bus_attribute *attr) { + struct subsys_private *sp = bus_to_subsys(bus); int error; - if (bus_get(bus)) { - error = sysfs_create_file(&bus->p->subsys.kobj, &attr->attr); - bus_put(bus); - } else - error = -EINVAL; + + if (!sp) + return -EINVAL; + + error = sysfs_create_file(&sp->subsys.kobj, &attr->attr); + + subsys_put(sp); return error; } EXPORT_SYMBOL_GPL(bus_create_file); -void bus_remove_file(struct bus_type *bus, struct bus_attribute *attr) +void bus_remove_file(const struct bus_type *bus, struct bus_attribute *attr) { - if (bus_get(bus)) { - sysfs_remove_file(&bus->p->subsys.kobj, &attr->attr); - bus_put(bus); - } + struct subsys_private *sp = bus_to_subsys(bus); + + if (!sp) + return; + + sysfs_remove_file(&sp->subsys.kobj, &attr->attr); + subsys_put(sp); } EXPORT_SYMBOL_GPL(bus_remove_file); static void bus_release(struct kobject *kobj) { struct subsys_private *priv = to_subsys_private(kobj); - struct bus_type *bus = priv->bus; + lockdep_unregister_key(&priv->lock_key); kfree(priv); - bus->p = NULL; } -static struct kobj_type bus_ktype = { +static const struct kobj_type bus_ktype = { .sysfs_ops = &bus_sysfs_ops, .release = bus_release, }; -static int bus_uevent_filter(struct kset *kset, struct kobject *kobj) +static int bus_uevent_filter(const struct kobject *kobj) { - struct kobj_type *ktype = get_ktype(kobj); + const struct kobj_type *ktype = get_ktype(kobj); if (ktype == &bus_ktype) return 1; @@ -176,13 +231,11 @@ static const struct kset_uevent_ops bus_uevent_ops = { .filter = bus_uevent_filter, }; -static struct kset *bus_kset; - /* Manually detach a device from its associated driver. */ static ssize_t unbind_store(struct device_driver *drv, const char *buf, size_t count) { - struct bus_type *bus = bus_get(drv->bus); + const struct bus_type *bus = bus_get(drv->bus); struct device *dev; int err = -ENODEV; @@ -195,7 +248,7 @@ static ssize_t unbind_store(struct device_driver *drv, const char *buf, bus_put(bus); return err; } -static DRIVER_ATTR_IGNORE_LOCKDEP(unbind, S_IWUSR, NULL, unbind_store); +static DRIVER_ATTR_IGNORE_LOCKDEP(unbind, 0200, NULL, unbind_store); /* * Manually attach a device to a driver. @@ -205,7 +258,7 @@ static DRIVER_ATTR_IGNORE_LOCKDEP(unbind, S_IWUSR, NULL, unbind_store); static ssize_t bind_store(struct device_driver *drv, const char *buf, size_t count) { - struct bus_type *bus = bus_get(drv->bus); + const struct bus_type *bus = bus_get(drv->bus); struct device *dev; int err = -ENODEV; @@ -221,24 +274,39 @@ static ssize_t bind_store(struct device_driver *drv, const char *buf, bus_put(bus); return err; } -static DRIVER_ATTR_IGNORE_LOCKDEP(bind, S_IWUSR, NULL, bind_store); +static DRIVER_ATTR_IGNORE_LOCKDEP(bind, 0200, NULL, bind_store); -static ssize_t drivers_autoprobe_show(struct bus_type *bus, char *buf) +static ssize_t drivers_autoprobe_show(const struct bus_type *bus, char *buf) { - return sysfs_emit(buf, "%d\n", bus->p->drivers_autoprobe); + struct subsys_private *sp = bus_to_subsys(bus); + int ret; + + if (!sp) + return -EINVAL; + + ret = sysfs_emit(buf, "%d\n", sp->drivers_autoprobe); + subsys_put(sp); + return ret; } -static ssize_t drivers_autoprobe_store(struct bus_type *bus, +static ssize_t drivers_autoprobe_store(const struct bus_type *bus, const char *buf, size_t count) { + struct subsys_private *sp = bus_to_subsys(bus); + + if (!sp) + return -EINVAL; + if (buf[0] == '0') - bus->p->drivers_autoprobe = 0; + sp->drivers_autoprobe = 0; else - bus->p->drivers_autoprobe = 1; + sp->drivers_autoprobe = 1; + + subsys_put(sp); return count; } -static ssize_t drivers_probe_store(struct bus_type *bus, +static ssize_t drivers_probe_store(const struct bus_type *bus, const char *buf, size_t count) { struct device *dev; @@ -266,6 +334,19 @@ static struct device *next_device(struct klist_iter *i) return dev; } +static struct device *prev_device(struct klist_iter *i) +{ + struct klist_node *n = klist_prev(i); + struct device *dev = NULL; + struct device_private *dev_prv; + + if (n) { + dev_prv = to_device_private_bus(n); + dev = dev_prv->device; + } + return dev; +} + /** * bus_for_each_dev - device iterator. * @bus: bus type. @@ -285,21 +366,23 @@ static struct device *next_device(struct klist_iter *i) * to retain this data, it should do so, and increment the reference * count in the supplied callback. */ -int bus_for_each_dev(struct bus_type *bus, struct device *start, - void *data, int (*fn)(struct device *, void *)) +int bus_for_each_dev(const struct bus_type *bus, struct device *start, + void *data, device_iter_t fn) { + struct subsys_private *sp = bus_to_subsys(bus); struct klist_iter i; struct device *dev; int error = 0; - if (!bus || !bus->p) + if (!sp) return -EINVAL; - klist_iter_init_node(&bus->p->klist_devices, &i, + klist_iter_init_node(&sp->klist_devices, &i, (start ? &start->p->knode_bus : NULL)); while (!error && (dev = next_device(&i))) error = fn(dev, data); klist_iter_exit(&i); + subsys_put(sp); return error; } EXPORT_SYMBOL_GPL(bus_for_each_dev); @@ -319,66 +402,55 @@ EXPORT_SYMBOL_GPL(bus_for_each_dev); * if it does. If the callback returns non-zero, this function will * return to the caller and not iterate over any more devices. */ -struct device *bus_find_device(struct bus_type *bus, +struct device *bus_find_device(const struct bus_type *bus, struct device *start, const void *data, - int (*match)(struct device *dev, const void *data)) + device_match_t match) { + struct subsys_private *sp = bus_to_subsys(bus); struct klist_iter i; struct device *dev; - if (!bus || !bus->p) + if (!sp) return NULL; - klist_iter_init_node(&bus->p->klist_devices, &i, + klist_iter_init_node(&sp->klist_devices, &i, (start ? &start->p->knode_bus : NULL)); - while ((dev = next_device(&i))) - if (match(dev, data) && get_device(dev)) + while ((dev = next_device(&i))) { + if (match(dev, data)) { + get_device(dev); break; + } + } klist_iter_exit(&i); + subsys_put(sp); return dev; } EXPORT_SYMBOL_GPL(bus_find_device); -/** - * subsys_find_device_by_id - find a device with a specific enumeration number - * @subsys: subsystem - * @id: index 'id' in struct device - * @hint: device to check first - * - * Check the hint's next object and if it is a match return it directly, - * otherwise, fall back to a full list search. Either way a reference for - * the returned object is taken. - */ -struct device *subsys_find_device_by_id(struct bus_type *subsys, unsigned int id, - struct device *hint) +struct device *bus_find_device_reverse(const struct bus_type *bus, + struct device *start, const void *data, + device_match_t match) { + struct subsys_private *sp = bus_to_subsys(bus); struct klist_iter i; struct device *dev; - if (!subsys) + if (!sp) return NULL; - if (hint) { - klist_iter_init_node(&subsys->p->klist_devices, &i, &hint->p->knode_bus); - dev = next_device(&i); - if (dev && dev->id == id && get_device(dev)) { - klist_iter_exit(&i); - return dev; - } - klist_iter_exit(&i); - } - - klist_iter_init_node(&subsys->p->klist_devices, &i, NULL); - while ((dev = next_device(&i))) { - if (dev->id == id && get_device(dev)) { - klist_iter_exit(&i); - return dev; + klist_iter_init_node(&sp->klist_devices, &i, + (start ? &start->p->knode_bus : NULL)); + while ((dev = prev_device(&i))) { + if (match(dev, data)) { + get_device(dev); + break; } } klist_iter_exit(&i); - return NULL; + subsys_put(sp); + return dev; } -EXPORT_SYMBOL_GPL(subsys_find_device_by_id); +EXPORT_SYMBOL_GPL(bus_find_device_reverse); static struct device_driver *next_driver(struct klist_iter *i) { @@ -411,21 +483,23 @@ static struct device_driver *next_driver(struct klist_iter *i) * in the callback. It must also be sure to increment the refcount * so it doesn't disappear before returning to the caller. */ -int bus_for_each_drv(struct bus_type *bus, struct device_driver *start, +int bus_for_each_drv(const struct bus_type *bus, struct device_driver *start, void *data, int (*fn)(struct device_driver *, void *)) { + struct subsys_private *sp = bus_to_subsys(bus); struct klist_iter i; struct device_driver *drv; int error = 0; - if (!bus) + if (!sp) return -EINVAL; - klist_iter_init_node(&bus->p->klist_drivers, &i, + klist_iter_init_node(&sp->klist_drivers, &i, start ? &start->p->knode_bus : NULL); while ((drv = next_driver(&i)) && !error) error = fn(drv, data); klist_iter_exit(&i); + subsys_put(sp); return error; } EXPORT_SYMBOL_GPL(bus_for_each_drv); @@ -440,32 +514,46 @@ EXPORT_SYMBOL_GPL(bus_for_each_drv); */ int bus_add_device(struct device *dev) { - struct bus_type *bus = bus_get(dev->bus); - int error = 0; + struct subsys_private *sp = bus_to_subsys(dev->bus); + int error; - if (bus) { - pr_debug("bus: '%s': add device %s\n", bus->name, dev_name(dev)); - error = device_add_groups(dev, bus->dev_groups); - if (error) - goto out_put; - error = sysfs_create_link(&bus->p->devices_kset->kobj, - &dev->kobj, dev_name(dev)); - if (error) - goto out_groups; - error = sysfs_create_link(&dev->kobj, - &dev->bus->p->subsys.kobj, "subsystem"); - if (error) - goto out_subsys; - klist_add_tail(&dev->p->knode_bus, &bus->p->klist_devices); + if (!sp) { + /* + * This is a normal operation for many devices that do not + * have a bus assigned to them, just say that all went + * well. + */ + return 0; } + + /* + * Reference in sp is now incremented and will be dropped when + * the device is removed from the bus + */ + + pr_debug("bus: '%s': add device %s\n", sp->bus->name, dev_name(dev)); + + error = device_add_groups(dev, sp->bus->dev_groups); + if (error) + goto out_put; + + error = sysfs_create_link(&sp->devices_kset->kobj, &dev->kobj, dev_name(dev)); + if (error) + goto out_groups; + + error = sysfs_create_link(&dev->kobj, &sp->subsys.kobj, "subsystem"); + if (error) + goto out_subsys; + + klist_add_tail(&dev->p->knode_bus, &sp->klist_devices); return 0; out_subsys: - sysfs_remove_link(&bus->p->devices_kset->kobj, dev_name(dev)); + sysfs_remove_link(&sp->devices_kset->kobj, dev_name(dev)); out_groups: - device_remove_groups(dev, bus->dev_groups); + device_remove_groups(dev, sp->bus->dev_groups); out_put: - bus_put(dev->bus); + subsys_put(sp); return error; } @@ -477,20 +565,20 @@ out_put: */ void bus_probe_device(struct device *dev) { - struct bus_type *bus = dev->bus; + struct subsys_private *sp = bus_to_subsys(dev->bus); struct subsys_interface *sif; - if (!bus) + if (!sp) return; - if (bus->p->drivers_autoprobe) - device_initial_probe(dev); + device_initial_probe(dev); - mutex_lock(&bus->p->mutex); - list_for_each_entry(sif, &bus->p->interfaces, node) + mutex_lock(&sp->mutex); + list_for_each_entry(sif, &sp->interfaces, node) if (sif->add_dev) sif->add_dev(dev, sif); - mutex_unlock(&bus->p->mutex); + mutex_unlock(&sp->mutex); + subsys_put(sp); } /** @@ -505,21 +593,20 @@ void bus_probe_device(struct device *dev) */ void bus_remove_device(struct device *dev) { - struct bus_type *bus = dev->bus; + struct subsys_private *sp = bus_to_subsys(dev->bus); struct subsys_interface *sif; - if (!bus) + if (!sp) return; - mutex_lock(&bus->p->mutex); - list_for_each_entry(sif, &bus->p->interfaces, node) + mutex_lock(&sp->mutex); + list_for_each_entry(sif, &sp->interfaces, node) if (sif->remove_dev) sif->remove_dev(dev, sif); - mutex_unlock(&bus->p->mutex); + mutex_unlock(&sp->mutex); sysfs_remove_link(&dev->kobj, "subsystem"); - sysfs_remove_link(&dev->bus->p->devices_kset->kobj, - dev_name(dev)); + sysfs_remove_link(&sp->devices_kset->kobj, dev_name(dev)); device_remove_groups(dev, dev->bus->dev_groups); if (klist_node_attached(&dev->p->knode_bus)) klist_del(&dev->p->knode_bus); @@ -527,7 +614,14 @@ void bus_remove_device(struct device *dev) pr_debug("bus: '%s': remove device %s\n", dev->bus->name, dev_name(dev)); device_release_driver(dev); - bus_put(dev->bus); + + /* + * Decrement the reference count twice, once for the bus_to_subsys() + * call in the start of this function, and the second one from the + * reference increment in bus_add_device() + */ + subsys_put(sp); + subsys_put(sp); } static int __must_check add_bind_files(struct device_driver *drv) @@ -552,7 +646,7 @@ static void remove_bind_files(struct device_driver *drv) static BUS_ATTR_WO(drivers_probe); static BUS_ATTR_RW(drivers_autoprobe); -static int add_probe_files(struct bus_type *bus) +static int add_probe_files(const struct bus_type *bus) { int retval; @@ -567,7 +661,7 @@ out: return retval; } -static void remove_probe_files(struct bus_type *bus) +static void remove_probe_files(const struct bus_type *bus) { bus_remove_file(bus, &bus_attr_drivers_autoprobe); bus_remove_file(bus, &bus_attr_drivers_probe); @@ -589,15 +683,18 @@ static DRIVER_ATTR_WO(uevent); */ int bus_add_driver(struct device_driver *drv) { - struct bus_type *bus; + struct subsys_private *sp = bus_to_subsys(drv->bus); struct driver_private *priv; int error = 0; - bus = bus_get(drv->bus); - if (!bus) + if (!sp) return -EINVAL; - pr_debug("bus: '%s': add driver %s\n", bus->name, drv->name); + /* + * Reference in sp is now incremented and will be dropped when + * the driver is removed from the bus + */ + pr_debug("bus: '%s': add driver %s\n", sp->bus->name, drv->name); priv = kzalloc(sizeof(*priv), GFP_KERNEL); if (!priv) { @@ -607,26 +704,31 @@ int bus_add_driver(struct device_driver *drv) klist_init(&priv->klist_devices, NULL, NULL); priv->driver = drv; drv->p = priv; - priv->kobj.kset = bus->p->drivers_kset; + priv->kobj.kset = sp->drivers_kset; error = kobject_init_and_add(&priv->kobj, &driver_ktype, NULL, "%s", drv->name); if (error) goto out_unregister; - klist_add_tail(&priv->knode_bus, &bus->p->klist_drivers); - if (drv->bus->p->drivers_autoprobe) { + klist_add_tail(&priv->knode_bus, &sp->klist_drivers); + if (sp->drivers_autoprobe) { error = driver_attach(drv); if (error) - goto out_unregister; + goto out_del_list; + } + error = module_add_driver(drv->owner, drv); + if (error) { + printk(KERN_ERR "%s: failed to create module links for %s\n", + __func__, drv->name); + goto out_detach; } - module_add_driver(drv->owner, drv); error = driver_create_file(drv, &driver_attr_uevent); if (error) { printk(KERN_ERR "%s: uevent attr (%s) failed\n", __func__, drv->name); } - error = driver_add_groups(drv, bus->drv_groups); + error = driver_add_groups(drv, sp->bus->drv_groups); if (error) { /* How the hell do we get out of this pickle? Give up */ printk(KERN_ERR "%s: driver_add_groups(%s) failed\n", @@ -644,12 +746,16 @@ int bus_add_driver(struct device_driver *drv) return 0; +out_detach: + driver_detach(drv); +out_del_list: + klist_del(&priv->knode_bus); out_unregister: kobject_put(&priv->kobj); /* drv->p is freed in driver_release() */ drv->p = NULL; out_put_bus: - bus_put(bus); + subsys_put(sp); return error; } @@ -663,19 +769,29 @@ out_put_bus: */ void bus_remove_driver(struct device_driver *drv) { - if (!drv->bus) + struct subsys_private *sp = bus_to_subsys(drv->bus); + + if (!sp) return; + pr_debug("bus: '%s': remove driver %s\n", sp->bus->name, drv->name); + if (!drv->suppress_bind_attrs) remove_bind_files(drv); - driver_remove_groups(drv, drv->bus->drv_groups); + driver_remove_groups(drv, sp->bus->drv_groups); driver_remove_file(drv, &driver_attr_uevent); klist_remove(&drv->p->knode_bus); - pr_debug("bus: '%s': remove driver %s\n", drv->bus->name, drv->name); driver_detach(drv); module_remove_driver(drv); kobject_put(&drv->p->kobj); - bus_put(drv->bus); + + /* + * Decrement the reference count twice, once for the bus_to_subsys() + * call in the start of this function, and the second one from the + * reference increment in bus_add_driver() + */ + subsys_put(sp); + subsys_put(sp); } /* Helper for bus_rescan_devices's iter */ @@ -702,7 +818,7 @@ static int __must_check bus_rescan_devices_helper(struct device *dev, * attached and rescan it against existing drivers to see if it matches * any by calling device_attach() for the unbound devices. */ -int bus_rescan_devices(struct bus_type *bus) +int bus_rescan_devices(const struct bus_type *bus) { return bus_for_each_dev(bus, NULL, NULL, bus_rescan_devices_helper); } @@ -725,18 +841,6 @@ int device_reprobe(struct device *dev) } EXPORT_SYMBOL_GPL(device_reprobe); -static int bus_add_groups(struct bus_type *bus, - const struct attribute_group **groups) -{ - return sysfs_create_groups(&bus->p->subsys.kobj, groups); -} - -static void bus_remove_groups(struct bus_type *bus, - const struct attribute_group **groups) -{ - sysfs_remove_groups(&bus->p->subsys.kobj, groups); -} - static void klist_devices_get(struct klist_node *n) { struct device_private *dev_prv = to_device_private_bus(n); @@ -753,13 +857,21 @@ static void klist_devices_put(struct klist_node *n) put_device(dev); } -static ssize_t bus_uevent_store(struct bus_type *bus, +static ssize_t bus_uevent_store(const struct bus_type *bus, const char *buf, size_t count) { - int rc; + struct subsys_private *sp = bus_to_subsys(bus); + int ret; - rc = kobject_synth_uevent(&bus->p->subsys.kobj, buf, count); - return rc ? rc : count; + if (!sp) + return -EINVAL; + + ret = kobject_synth_uevent(&sp->subsys.kobj, buf, count); + subsys_put(sp); + + if (ret) + return ret; + return count; } /* * "open code" the old BUS_ATTR() macro here. We want to use BUS_ATTR_WO() @@ -767,7 +879,7 @@ static ssize_t bus_uevent_store(struct bus_type *bus, * DEVICE_ATTR_WO(uevent), which would cause a clash with the with the store * function name. */ -static struct bus_attribute bus_attr_uevent = __ATTR(uevent, S_IWUSR, NULL, +static struct bus_attribute bus_attr_uevent = __ATTR(uevent, 0200, NULL, bus_uevent_store); /** @@ -778,27 +890,28 @@ static struct bus_attribute bus_attr_uevent = __ATTR(uevent, S_IWUSR, NULL, * infrastructure, then register the children subsystems it has: * the devices and drivers that belong to the subsystem. */ -int bus_register(struct bus_type *bus) +int bus_register(const struct bus_type *bus) { int retval; struct subsys_private *priv; - struct lock_class_key *key = &bus->lock_key; + struct kobject *bus_kobj; + struct lock_class_key *key; priv = kzalloc(sizeof(struct subsys_private), GFP_KERNEL); if (!priv) return -ENOMEM; priv->bus = bus; - bus->p = priv; BLOCKING_INIT_NOTIFIER_HEAD(&priv->bus_notifier); - retval = kobject_set_name(&priv->subsys.kobj, "%s", bus->name); + bus_kobj = &priv->subsys.kobj; + retval = kobject_set_name(bus_kobj, "%s", bus->name); if (retval) goto out; - priv->subsys.kobj.kset = bus_kset; - priv->subsys.kobj.ktype = &bus_ktype; + bus_kobj->kset = bus_kset; + bus_kobj->ktype = &bus_ktype; priv->drivers_autoprobe = 1; retval = kset_register(&priv->subsys); @@ -809,21 +922,21 @@ int bus_register(struct bus_type *bus) if (retval) goto bus_uevent_fail; - priv->devices_kset = kset_create_and_add("devices", NULL, - &priv->subsys.kobj); + priv->devices_kset = kset_create_and_add("devices", NULL, bus_kobj); if (!priv->devices_kset) { retval = -ENOMEM; goto bus_devices_fail; } - priv->drivers_kset = kset_create_and_add("drivers", NULL, - &priv->subsys.kobj); + priv->drivers_kset = kset_create_and_add("drivers", NULL, bus_kobj); if (!priv->drivers_kset) { retval = -ENOMEM; goto bus_drivers_fail; } INIT_LIST_HEAD(&priv->interfaces); + key = &priv->lock_key; + lockdep_register_key(key); __mutex_init(&priv->mutex, "subsys mutex", key); klist_init(&priv->klist_devices, klist_devices_get, klist_devices_put); klist_init(&priv->klist_drivers, NULL, NULL); @@ -832,7 +945,7 @@ int bus_register(struct bus_type *bus) if (retval) goto bus_probe_files_fail; - retval = bus_add_groups(bus, bus->bus_groups); + retval = sysfs_create_groups(bus_kobj, bus->bus_groups); if (retval) goto bus_groups_fail; @@ -842,16 +955,17 @@ int bus_register(struct bus_type *bus) bus_groups_fail: remove_probe_files(bus); bus_probe_files_fail: - kset_unregister(bus->p->drivers_kset); + kset_unregister(priv->drivers_kset); bus_drivers_fail: - kset_unregister(bus->p->devices_kset); + kset_unregister(priv->devices_kset); bus_devices_fail: bus_remove_file(bus, &bus_attr_uevent); bus_uevent_fail: - kset_unregister(&bus->p->subsys); + kset_unregister(&priv->subsys); + /* Above kset_unregister() will kfree @priv */ + priv = NULL; out: - kfree(bus->p); - bus->p = NULL; + kfree(priv); return retval; } EXPORT_SYMBOL_GPL(bus_register); @@ -863,43 +977,82 @@ EXPORT_SYMBOL_GPL(bus_register); * Unregister the child subsystems and the bus itself. * Finally, we call bus_put() to release the refcount */ -void bus_unregister(struct bus_type *bus) +void bus_unregister(const struct bus_type *bus) { + struct subsys_private *sp = bus_to_subsys(bus); + struct kobject *bus_kobj; + + if (!sp) + return; + pr_debug("bus: '%s': unregistering\n", bus->name); - if (bus->dev_root) - device_unregister(bus->dev_root); - bus_remove_groups(bus, bus->bus_groups); + if (sp->dev_root) + device_unregister(sp->dev_root); + + bus_kobj = &sp->subsys.kobj; + sysfs_remove_groups(bus_kobj, bus->bus_groups); remove_probe_files(bus); - kset_unregister(bus->p->drivers_kset); - kset_unregister(bus->p->devices_kset); bus_remove_file(bus, &bus_attr_uevent); - kset_unregister(&bus->p->subsys); + + kset_unregister(sp->drivers_kset); + kset_unregister(sp->devices_kset); + kset_unregister(&sp->subsys); + subsys_put(sp); } EXPORT_SYMBOL_GPL(bus_unregister); -int bus_register_notifier(struct bus_type *bus, struct notifier_block *nb) +int bus_register_notifier(const struct bus_type *bus, struct notifier_block *nb) { - return blocking_notifier_chain_register(&bus->p->bus_notifier, nb); + struct subsys_private *sp = bus_to_subsys(bus); + int retval; + + if (!sp) + return -EINVAL; + + retval = blocking_notifier_chain_register(&sp->bus_notifier, nb); + subsys_put(sp); + return retval; } EXPORT_SYMBOL_GPL(bus_register_notifier); -int bus_unregister_notifier(struct bus_type *bus, struct notifier_block *nb) +int bus_unregister_notifier(const struct bus_type *bus, struct notifier_block *nb) { - return blocking_notifier_chain_unregister(&bus->p->bus_notifier, nb); + struct subsys_private *sp = bus_to_subsys(bus); + int retval; + + if (!sp) + return -EINVAL; + retval = blocking_notifier_chain_unregister(&sp->bus_notifier, nb); + subsys_put(sp); + return retval; } EXPORT_SYMBOL_GPL(bus_unregister_notifier); -struct kset *bus_get_kset(struct bus_type *bus) +void bus_notify(struct device *dev, enum bus_notifier_event value) { - return &bus->p->subsys; + struct subsys_private *sp = bus_to_subsys(dev->bus); + + if (!sp) + return; + + blocking_notifier_call_chain(&sp->bus_notifier, value, dev); + subsys_put(sp); } -EXPORT_SYMBOL_GPL(bus_get_kset); -struct klist *bus_get_device_klist(struct bus_type *bus) +struct kset *bus_get_kset(const struct bus_type *bus) { - return &bus->p->klist_devices; + struct subsys_private *sp = bus_to_subsys(bus); + struct kset *kset; + + if (!sp) + return NULL; + + kset = &sp->subsys; + subsys_put(sp); + + return kset; } -EXPORT_SYMBOL_GPL(bus_get_device_klist); +EXPORT_SYMBOL_GPL(bus_get_kset); /* * Yes, this forcibly breaks the klist abstraction temporarily. It @@ -928,17 +1081,20 @@ static void device_insertion_sort_klist(struct device *a, struct list_head *list list_move_tail(&a->p->knode_bus.n_node, list); } -void bus_sort_breadthfirst(struct bus_type *bus, +void bus_sort_breadthfirst(const struct bus_type *bus, int (*compare)(const struct device *a, const struct device *b)) { + struct subsys_private *sp = bus_to_subsys(bus); LIST_HEAD(sorted_devices); struct klist_node *n, *tmp; struct device_private *dev_prv; struct device *dev; struct klist *device_klist; - device_klist = bus_get_device_klist(bus); + if (!sp) + return; + device_klist = &sp->klist_devices; spin_lock(&device_klist->k_lock); list_for_each_entry_safe(n, tmp, &device_klist->k_list, n_node) { @@ -948,13 +1104,19 @@ void bus_sort_breadthfirst(struct bus_type *bus, } list_splice(&sorted_devices, &device_klist->k_list); spin_unlock(&device_klist->k_lock); + subsys_put(sp); } EXPORT_SYMBOL_GPL(bus_sort_breadthfirst); +struct subsys_dev_iter { + struct klist_iter ki; + const struct device_type *type; +}; + /** * subsys_dev_iter_init - initialize subsys device iterator * @iter: subsys iterator to initialize - * @subsys: the subsys we wanna iterate over + * @sp: the subsys private (i.e. bus) we wanna iterate over * @start: the device to start iterating from, if any * @type: device_type of the devices to iterate over, NULL for all * @@ -963,17 +1125,16 @@ EXPORT_SYMBOL_GPL(bus_sort_breadthfirst); * otherwise if it is NULL, the iteration starts at the beginning of * the list. */ -void subsys_dev_iter_init(struct subsys_dev_iter *iter, struct bus_type *subsys, - struct device *start, const struct device_type *type) +static void subsys_dev_iter_init(struct subsys_dev_iter *iter, struct subsys_private *sp, + struct device *start, const struct device_type *type) { struct klist_node *start_knode = NULL; if (start) start_knode = &start->p->knode_bus; - klist_iter_init_node(&subsys->p->klist_devices, &iter->ki, start_knode); + klist_iter_init_node(&sp->klist_devices, &iter->ki, start_knode); iter->type = type; } -EXPORT_SYMBOL_GPL(subsys_dev_iter_init); /** * subsys_dev_iter_next - iterate to the next device @@ -987,7 +1148,7 @@ EXPORT_SYMBOL_GPL(subsys_dev_iter_init); * free to do whatever it wants to do with the device including * calling back into subsys code. */ -struct device *subsys_dev_iter_next(struct subsys_dev_iter *iter) +static struct device *subsys_dev_iter_next(struct subsys_dev_iter *iter) { struct klist_node *knode; struct device *dev; @@ -1001,7 +1162,6 @@ struct device *subsys_dev_iter_next(struct subsys_dev_iter *iter) return dev; } } -EXPORT_SYMBOL_GPL(subsys_dev_iter_next); /** * subsys_dev_iter_exit - finish iteration @@ -1010,34 +1170,38 @@ EXPORT_SYMBOL_GPL(subsys_dev_iter_next); * Finish an iteration. Always call this function after iteration is * complete whether the iteration ran till the end or not. */ -void subsys_dev_iter_exit(struct subsys_dev_iter *iter) +static void subsys_dev_iter_exit(struct subsys_dev_iter *iter) { klist_iter_exit(&iter->ki); } -EXPORT_SYMBOL_GPL(subsys_dev_iter_exit); int subsys_interface_register(struct subsys_interface *sif) { - struct bus_type *subsys; + struct subsys_private *sp; struct subsys_dev_iter iter; struct device *dev; if (!sif || !sif->subsys) return -ENODEV; - subsys = bus_get(sif->subsys); - if (!subsys) + sp = bus_to_subsys(sif->subsys); + if (!sp) return -EINVAL; - mutex_lock(&subsys->p->mutex); - list_add_tail(&sif->node, &subsys->p->interfaces); + /* + * Reference in sp is now incremented and will be dropped when + * the interface is removed from the bus + */ + + mutex_lock(&sp->mutex); + list_add_tail(&sif->node, &sp->interfaces); if (sif->add_dev) { - subsys_dev_iter_init(&iter, subsys, NULL, NULL); + subsys_dev_iter_init(&iter, sp, NULL, NULL); while ((dev = subsys_dev_iter_next(&iter))) sif->add_dev(dev, sif); subsys_dev_iter_exit(&iter); } - mutex_unlock(&subsys->p->mutex); + mutex_unlock(&sp->mutex); return 0; } @@ -1045,26 +1209,34 @@ EXPORT_SYMBOL_GPL(subsys_interface_register); void subsys_interface_unregister(struct subsys_interface *sif) { - struct bus_type *subsys; + struct subsys_private *sp; struct subsys_dev_iter iter; struct device *dev; if (!sif || !sif->subsys) return; - subsys = sif->subsys; + sp = bus_to_subsys(sif->subsys); + if (!sp) + return; - mutex_lock(&subsys->p->mutex); + mutex_lock(&sp->mutex); list_del_init(&sif->node); if (sif->remove_dev) { - subsys_dev_iter_init(&iter, subsys, NULL, NULL); + subsys_dev_iter_init(&iter, sp, NULL, NULL); while ((dev = subsys_dev_iter_next(&iter))) sif->remove_dev(dev, sif); subsys_dev_iter_exit(&iter); } - mutex_unlock(&subsys->p->mutex); - - bus_put(subsys); + mutex_unlock(&sp->mutex); + + /* + * Decrement the reference count twice, once for the bus_to_subsys() + * call in the start of this function, and the second one from the + * reference increment in subsys_interface_register() + */ + subsys_put(sp); + subsys_put(sp); } EXPORT_SYMBOL_GPL(subsys_interface_unregister); @@ -1073,10 +1245,11 @@ static void system_root_device_release(struct device *dev) kfree(dev); } -static int subsys_register(struct bus_type *subsys, +static int subsys_register(const struct bus_type *subsys, const struct attribute_group **groups, struct kobject *parent_of_root) { + struct subsys_private *sp; struct device *dev; int err; @@ -1084,6 +1257,12 @@ static int subsys_register(struct bus_type *subsys, if (err < 0) return err; + sp = bus_to_subsys(subsys); + if (!sp) { + err = -EINVAL; + goto err_sp; + } + dev = kzalloc(sizeof(struct device), GFP_KERNEL); if (!dev) { err = -ENOMEM; @@ -1102,7 +1281,8 @@ static int subsys_register(struct bus_type *subsys, if (err < 0) goto err_dev_reg; - subsys->dev_root = dev; + sp->dev_root = dev; + subsys_put(sp); return 0; err_dev_reg: @@ -1111,6 +1291,8 @@ err_dev_reg: err_name: kfree(dev); err_dev: + subsys_put(sp); +err_sp: bus_unregister(subsys); return err; } @@ -1133,7 +1315,7 @@ err_dev: * directory itself and not some create fake root-device placed in * /sys/devices/system/<name>. */ -int subsys_system_register(struct bus_type *subsys, +int subsys_system_register(const struct bus_type *subsys, const struct attribute_group **groups) { return subsys_register(subsys, groups, &system_kset->kobj); @@ -1146,17 +1328,17 @@ EXPORT_SYMBOL_GPL(subsys_system_register); * @groups: default attributes for the root device * * All 'virtual' subsystems have a /sys/devices/system/<name> root device - * with the name of the subystem. The root device can carry subsystem-wide + * with the name of the subsystem. The root device can carry subsystem-wide * attributes. All registered devices are below this single root device. * There's no restriction on device naming. This is for kernel software * constructs which need sysfs interface. */ -int subsys_virtual_register(struct bus_type *subsys, +int subsys_virtual_register(const struct bus_type *subsys, const struct attribute_group **groups) { struct kobject *virtual_dir; - virtual_dir = virtual_device_parent(NULL); + virtual_dir = virtual_device_parent(); if (!virtual_dir) return -ENOMEM; @@ -1164,6 +1346,82 @@ int subsys_virtual_register(struct bus_type *subsys, } EXPORT_SYMBOL_GPL(subsys_virtual_register); +/** + * driver_find - locate driver on a bus by its name. + * @name: name of the driver. + * @bus: bus to scan for the driver. + * + * Call kset_find_obj() to iterate over list of drivers on + * a bus to find driver by name. Return driver if found. + * + * This routine provides no locking to prevent the driver it returns + * from being unregistered or unloaded while the caller is using it. + * The caller is responsible for preventing this. + */ +struct device_driver *driver_find(const char *name, const struct bus_type *bus) +{ + struct subsys_private *sp = bus_to_subsys(bus); + struct kobject *k; + struct driver_private *priv; + + if (!sp) + return NULL; + + k = kset_find_obj(sp->drivers_kset, name); + subsys_put(sp); + if (!k) + return NULL; + + priv = to_driver(k); + + /* Drop reference added by kset_find_obj() */ + kobject_put(k); + return priv->driver; +} +EXPORT_SYMBOL_GPL(driver_find); + +/* + * Warning, the value could go to "removed" instantly after calling this function, so be very + * careful when calling it... + */ +bool bus_is_registered(const struct bus_type *bus) +{ + struct subsys_private *sp = bus_to_subsys(bus); + bool is_initialized = false; + + if (sp) { + is_initialized = true; + subsys_put(sp); + } + return is_initialized; +} + +/** + * bus_get_dev_root - return a pointer to the "device root" of a bus + * @bus: bus to return the device root of. + * + * If a bus has a "device root" structure, return it, WITH THE REFERENCE + * COUNT INCREMENTED. + * + * Note, when finished with the device, a call to put_device() is required. + * + * If the device root is not present (or bus is not a valid pointer), NULL + * will be returned. + */ +struct device *bus_get_dev_root(const struct bus_type *bus) +{ + struct subsys_private *sp = bus_to_subsys(bus); + struct device *dev_root; + + if (!sp) + return NULL; + + dev_root = get_device(sp->dev_root); + subsys_put(sp); + return dev_root; +} +EXPORT_SYMBOL_GPL(bus_get_dev_root); + int __init buses_init(void) { bus_kset = kset_create_and_add("bus", &bus_uevent_ops, NULL); @@ -1171,8 +1429,13 @@ int __init buses_init(void) return -ENOMEM; system_kset = kset_create_and_add("system", NULL, &devices_kset->kobj); - if (!system_kset) + if (!system_kset) { + /* Do error handling here as devices_init() do */ + kset_unregister(bus_kset); + bus_kset = NULL; + pr_err("%s: failed to create and add kset 'bus'\n", __func__); return -ENOMEM; + } return 0; } diff --git a/drivers/base/cacheinfo.c b/drivers/base/cacheinfo.c index bfc095956dd1..613410705a47 100644 --- a/drivers/base/cacheinfo.c +++ b/drivers/base/cacheinfo.c @@ -8,6 +8,7 @@ #define pr_fmt(fmt) KBUILD_MODNAME ": " fmt #include <linux/acpi.h> +#include <linux/bitfield.h> #include <linux/bitops.h> #include <linux/cacheinfo.h> #include <linux/compiler.h> @@ -25,19 +26,66 @@ static DEFINE_PER_CPU(struct cpu_cacheinfo, ci_cpu_cacheinfo); #define ci_cacheinfo(cpu) (&per_cpu(ci_cpu_cacheinfo, cpu)) #define cache_leaves(cpu) (ci_cacheinfo(cpu)->num_leaves) #define per_cpu_cacheinfo(cpu) (ci_cacheinfo(cpu)->info_list) +#define per_cpu_cacheinfo_idx(cpu, idx) \ + (per_cpu_cacheinfo(cpu) + (idx)) + +/* Set if no cache information is found in DT/ACPI. */ +static bool use_arch_info; struct cpu_cacheinfo *get_cpu_cacheinfo(unsigned int cpu) { return ci_cacheinfo(cpu); } -#ifdef CONFIG_OF static inline bool cache_leaves_are_shared(struct cacheinfo *this_leaf, struct cacheinfo *sib_leaf) { + /* + * For non DT/ACPI systems, assume unique level 1 caches, + * system-wide shared caches for all other levels. + */ + if (!(IS_ENABLED(CONFIG_OF) || IS_ENABLED(CONFIG_ACPI)) || + use_arch_info) + return (this_leaf->level != 1) && (sib_leaf->level != 1); + + if ((sib_leaf->attributes & CACHE_ID) && + (this_leaf->attributes & CACHE_ID)) + return sib_leaf->id == this_leaf->id; + return sib_leaf->fw_token == this_leaf->fw_token; } +bool last_level_cache_is_valid(unsigned int cpu) +{ + struct cacheinfo *llc; + + if (!cache_leaves(cpu) || !per_cpu_cacheinfo(cpu)) + return false; + + llc = per_cpu_cacheinfo_idx(cpu, cache_leaves(cpu) - 1); + + return (llc->attributes & CACHE_ID) || !!llc->fw_token; + +} + +bool last_level_cache_is_shared(unsigned int cpu_x, unsigned int cpu_y) +{ + struct cacheinfo *llc_x, *llc_y; + + if (!last_level_cache_is_valid(cpu_x) || + !last_level_cache_is_valid(cpu_y)) + return false; + + llc_x = per_cpu_cacheinfo_idx(cpu_x, cache_leaves(cpu_x) - 1); + llc_y = per_cpu_cacheinfo_idx(cpu_y, cache_leaves(cpu_y) - 1); + + return cache_leaves_are_shared(llc_x, llc_y); +} + +#ifdef CONFIG_OF + +static bool of_check_cache_nodes(struct device_node *np); + /* OF properties to query for a given cache type */ struct cache_type_info { const char *size_prop; @@ -136,6 +184,54 @@ static bool cache_node_is_unified(struct cacheinfo *this_leaf, return of_property_read_bool(np, "cache-unified"); } +static bool match_cache_node(struct device_node *cpu, + const struct device_node *cache_node) +{ + struct device_node *prev, *cache = of_find_next_cache_node(cpu); + + while (cache) { + if (cache == cache_node) { + of_node_put(cache); + return true; + } + + prev = cache; + cache = of_find_next_cache_node(cache); + of_node_put(prev); + } + + return false; +} + +#ifndef arch_compact_of_hwid +#define arch_compact_of_hwid(_x) (_x) +#endif + +static void cache_of_set_id(struct cacheinfo *this_leaf, + struct device_node *cache_node) +{ + struct device_node *cpu; + u32 min_id = ~0; + + for_each_of_cpu_node(cpu) { + u64 id = of_get_cpu_hwid(cpu, 0); + + id = arch_compact_of_hwid(id); + if (FIELD_GET(GENMASK_ULL(63, 32), id)) { + of_node_put(cpu); + return; + } + + if (match_cache_node(cpu, cache_node)) + min_id = min(min_id, id); + } + + if (min_id != ~0) { + this_leaf->id = min_id; + this_leaf->attributes |= CACHE_ID; + } +} + static void cache_of_set_props(struct cacheinfo *this_leaf, struct device_node *np) { @@ -151,39 +247,32 @@ static void cache_of_set_props(struct cacheinfo *this_leaf, cache_get_line_size(this_leaf, np); cache_nr_sets(this_leaf, np); cache_associativity(this_leaf); + cache_of_set_id(this_leaf, np); } static int cache_setup_of_node(unsigned int cpu) { - struct device_node *np; struct cacheinfo *this_leaf; - struct device *cpu_dev = get_cpu_device(cpu); - struct cpu_cacheinfo *this_cpu_ci = get_cpu_cacheinfo(cpu); unsigned int index = 0; - /* skip if fw_token is already populated */ - if (this_cpu_ci->info_list->fw_token) { - return 0; - } - - if (!cpu_dev) { - pr_err("No cpu device for CPU %d\n", cpu); - return -ENODEV; - } - np = cpu_dev->of_node; + struct device_node *np __free(device_node) = of_cpu_device_node_get(cpu); if (!np) { pr_err("Failed to find cpu%d device node\n", cpu); return -ENOENT; } + if (!of_check_cache_nodes(np)) { + return -ENOENT; + } + while (index < cache_leaves(cpu)) { - this_leaf = this_cpu_ci->info_list + index; - if (this_leaf->level != 1) + this_leaf = per_cpu_cacheinfo_idx(cpu, index); + if (this_leaf->level != 1) { + struct device_node *prev __free(device_node) = np; np = of_find_next_cache_node(np); - else - np = of_node_get(np);/* cpu node itself */ - if (!np) - break; + if (!np) + break; + } cache_of_set_props(this_leaf, np); this_leaf->fw_token = np; index++; @@ -194,18 +283,87 @@ static int cache_setup_of_node(unsigned int cpu) return 0; } -#else -static inline int cache_setup_of_node(unsigned int cpu) { return 0; } -static inline bool cache_leaves_are_shared(struct cacheinfo *this_leaf, - struct cacheinfo *sib_leaf) + +static bool of_check_cache_nodes(struct device_node *np) { - /* - * For non-DT/ACPI systems, assume unique level 1 caches, system-wide - * shared caches for all other levels. This will be used only if - * arch specific code has not populated shared_cpu_map - */ - return !(this_leaf->level == 1); + if (of_property_present(np, "cache-size") || + of_property_present(np, "i-cache-size") || + of_property_present(np, "d-cache-size") || + of_property_present(np, "cache-unified")) + return true; + + struct device_node *next __free(device_node) = of_find_next_cache_node(np); + if (next) { + return true; + } + + return false; +} + +static int of_count_cache_leaves(struct device_node *np) +{ + unsigned int leaves = 0; + + if (of_property_present(np, "cache-size")) + ++leaves; + if (of_property_present(np, "i-cache-size")) + ++leaves; + if (of_property_present(np, "d-cache-size")) + ++leaves; + + if (!leaves) { + /* The '[i-|d-|]cache-size' property is required, but + * if absent, fallback on the 'cache-unified' property. + */ + if (of_property_read_bool(np, "cache-unified")) + return 1; + else + return 2; + } + + return leaves; } + +int init_of_cache_level(unsigned int cpu) +{ + struct cpu_cacheinfo *this_cpu_ci = get_cpu_cacheinfo(cpu); + struct device_node *np __free(device_node) = of_cpu_device_node_get(cpu); + unsigned int levels = 0, leaves, level; + + if (!of_check_cache_nodes(np)) { + return -ENOENT; + } + + leaves = of_count_cache_leaves(np); + if (leaves > 0) + levels = 1; + + while (1) { + struct device_node *prev __free(device_node) = np; + np = of_find_next_cache_node(np); + if (!np) + break; + + if (!of_device_is_compatible(np, "cache")) + return -EINVAL; + if (of_property_read_u32(np, "cache-level", &level)) + return -EINVAL; + if (level <= levels) + return -EINVAL; + + leaves += of_count_cache_leaves(np); + levels = level; + } + + this_cpu_ci->num_levels = levels; + this_cpu_ci->num_leaves = leaves; + + return 0; +} + +#else +static inline int cache_setup_of_node(unsigned int cpu) { return 0; } +int init_of_cache_level(unsigned int cpu) { return 0; } #endif int __weak cache_setup_acpi(unsigned int cpu) @@ -215,42 +373,69 @@ int __weak cache_setup_acpi(unsigned int cpu) unsigned int coherency_max_size; +static int cache_setup_properties(unsigned int cpu) +{ + int ret = 0; + + if (of_have_populated_dt()) + ret = cache_setup_of_node(cpu); + else if (!acpi_disabled) + ret = cache_setup_acpi(cpu); + + // Assume there is no cache information available in DT/ACPI from now. + if (ret && use_arch_cache_info()) + use_arch_info = true; + + return ret; +} + static int cache_shared_cpu_map_setup(unsigned int cpu) { struct cpu_cacheinfo *this_cpu_ci = get_cpu_cacheinfo(cpu); struct cacheinfo *this_leaf, *sib_leaf; - unsigned int index; + unsigned int index, sib_index; int ret = 0; if (this_cpu_ci->cpu_map_populated) return 0; - if (of_have_populated_dt()) - ret = cache_setup_of_node(cpu); - else if (!acpi_disabled) - ret = cache_setup_acpi(cpu); - - if (ret) - return ret; + /* + * skip setting up cache properties if LLC is valid, just need + * to update the shared cpu_map if the cache attributes were + * populated early before all the cpus are brought online + */ + if (!last_level_cache_is_valid(cpu) && !use_arch_info) { + ret = cache_setup_properties(cpu); + if (ret) + return ret; + } for (index = 0; index < cache_leaves(cpu); index++) { unsigned int i; - this_leaf = this_cpu_ci->info_list + index; - /* skip if shared_cpu_map is already populated */ - if (!cpumask_empty(&this_leaf->shared_cpu_map)) - continue; + this_leaf = per_cpu_cacheinfo_idx(cpu, index); cpumask_set_cpu(cpu, &this_leaf->shared_cpu_map); for_each_online_cpu(i) { - struct cpu_cacheinfo *sib_cpu_ci = get_cpu_cacheinfo(i); - - if (i == cpu || !sib_cpu_ci->info_list) + if (i == cpu || !per_cpu_cacheinfo(i)) continue;/* skip if itself or no cacheinfo */ - sib_leaf = sib_cpu_ci->info_list + index; - if (cache_leaves_are_shared(this_leaf, sib_leaf)) { - cpumask_set_cpu(cpu, &sib_leaf->shared_cpu_map); - cpumask_set_cpu(i, &this_leaf->shared_cpu_map); + for (sib_index = 0; sib_index < cache_leaves(i); sib_index++) { + sib_leaf = per_cpu_cacheinfo_idx(i, sib_index); + + /* + * Comparing cache IDs only makes sense if the leaves + * belong to the same cache level of same type. Skip + * the check if level and type do not match. + */ + if (sib_leaf->level != this_leaf->level || + sib_leaf->type != this_leaf->type) + continue; + + if (cache_leaves_are_shared(this_leaf, sib_leaf)) { + cpumask_set_cpu(cpu, &sib_leaf->shared_cpu_map); + cpumask_set_cpu(i, &this_leaf->shared_cpu_map); + break; + } } } /* record the maximum cache line size */ @@ -258,6 +443,8 @@ static int cache_shared_cpu_map_setup(unsigned int cpu) coherency_max_size = this_leaf->coherency_line_size; } + /* shared_cpu_map is now populated for the cpu */ + this_cpu_ci->cpu_map_populated = true; return 0; } @@ -265,27 +452,37 @@ static void cache_shared_cpu_map_remove(unsigned int cpu) { struct cpu_cacheinfo *this_cpu_ci = get_cpu_cacheinfo(cpu); struct cacheinfo *this_leaf, *sib_leaf; - unsigned int sibling, index; + unsigned int sibling, index, sib_index; for (index = 0; index < cache_leaves(cpu); index++) { - this_leaf = this_cpu_ci->info_list + index; + this_leaf = per_cpu_cacheinfo_idx(cpu, index); for_each_cpu(sibling, &this_leaf->shared_cpu_map) { - struct cpu_cacheinfo *sib_cpu_ci; - - if (sibling == cpu) /* skip itself */ - continue; - - sib_cpu_ci = get_cpu_cacheinfo(sibling); - if (!sib_cpu_ci->info_list) - continue; + if (sibling == cpu || !per_cpu_cacheinfo(sibling)) + continue;/* skip if itself or no cacheinfo */ - sib_leaf = sib_cpu_ci->info_list + index; - cpumask_clear_cpu(cpu, &sib_leaf->shared_cpu_map); - cpumask_clear_cpu(sibling, &this_leaf->shared_cpu_map); + for (sib_index = 0; sib_index < cache_leaves(sibling); sib_index++) { + sib_leaf = per_cpu_cacheinfo_idx(sibling, sib_index); + + /* + * Comparing cache IDs only makes sense if the leaves + * belong to the same cache level of same type. Skip + * the check if level and type do not match. + */ + if (sib_leaf->level != this_leaf->level || + sib_leaf->type != this_leaf->type) + continue; + + if (cache_leaves_are_shared(this_leaf, sib_leaf)) { + cpumask_clear_cpu(cpu, &sib_leaf->shared_cpu_map); + cpumask_clear_cpu(sibling, &this_leaf->shared_cpu_map); + break; + } + } } - if (of_have_populated_dt()) - of_node_put(this_leaf->fw_token); } + + /* cpu is no longer populated in the shared map */ + this_cpu_ci->cpu_map_populated = false; } static void free_cache_attributes(unsigned int cpu) @@ -294,9 +491,11 @@ static void free_cache_attributes(unsigned int cpu) return; cache_shared_cpu_map_remove(cpu); +} - kfree(per_cpu_cacheinfo(cpu)); - per_cpu_cacheinfo(cpu) = NULL; +int __weak early_cache_level(unsigned int cpu) +{ + return -ENOENT; } int __weak init_cache_level(unsigned int cpu) @@ -309,25 +508,113 @@ int __weak populate_cache_leaves(unsigned int cpu) return -ENOENT; } -static int detect_cache_attributes(unsigned int cpu) +static inline int allocate_cache_info(int cpu) { + per_cpu_cacheinfo(cpu) = kcalloc(cache_leaves(cpu), sizeof(struct cacheinfo), GFP_ATOMIC); + if (!per_cpu_cacheinfo(cpu)) { + cache_leaves(cpu) = 0; + return -ENOMEM; + } + + return 0; +} + +int fetch_cache_info(unsigned int cpu) +{ + struct cpu_cacheinfo *this_cpu_ci = get_cpu_cacheinfo(cpu); + unsigned int levels = 0, split_levels = 0; int ret; + if (acpi_disabled) { + ret = init_of_cache_level(cpu); + } else { + ret = acpi_get_cache_info(cpu, &levels, &split_levels); + if (!ret) { + this_cpu_ci->num_levels = levels; + /* + * This assumes that: + * - there cannot be any split caches (data/instruction) + * above a unified cache + * - data/instruction caches come by pair + */ + this_cpu_ci->num_leaves = levels + split_levels; + } + } + + if (ret || !cache_leaves(cpu)) { + ret = early_cache_level(cpu); + if (ret) + return ret; + + if (!cache_leaves(cpu)) + return -ENOENT; + + this_cpu_ci->early_ci_levels = true; + } + + return allocate_cache_info(cpu); +} + +static inline int init_level_allocate_ci(unsigned int cpu) +{ + unsigned int early_leaves = cache_leaves(cpu); + + /* Since early initialization/allocation of the cacheinfo is allowed + * via fetch_cache_info() and this also gets called as CPU hotplug + * callbacks via cacheinfo_cpu_online, the init/alloc can be skipped + * as it will happen only once (the cacheinfo memory is never freed). + * Just populate the cacheinfo. However, if the cacheinfo has been + * allocated early through the arch-specific early_cache_level() call, + * there is a chance the info is wrong (this can happen on arm64). In + * that case, call init_cache_level() anyway to give the arch-specific + * code a chance to make things right. + */ + if (per_cpu_cacheinfo(cpu) && !ci_cacheinfo(cpu)->early_ci_levels) + return 0; + if (init_cache_level(cpu) || !cache_leaves(cpu)) return -ENOENT; - per_cpu_cacheinfo(cpu) = kcalloc(cache_leaves(cpu), - sizeof(struct cacheinfo), GFP_KERNEL); - if (per_cpu_cacheinfo(cpu) == NULL) - return -ENOMEM; + /* + * Now that we have properly initialized the cache level info, make + * sure we don't try to do that again the next time we are called + * (e.g. as CPU hotplug callbacks). + */ + ci_cacheinfo(cpu)->early_ci_levels = false; /* - * populate_cache_leaves() may completely setup the cache leaves and - * shared_cpu_map or it may leave it partially setup. + * Some architectures (e.g., x86) do not use early initialization. + * Allocate memory now in such case. */ - ret = populate_cache_leaves(cpu); + if (cache_leaves(cpu) <= early_leaves && per_cpu_cacheinfo(cpu)) + return 0; + + kfree(per_cpu_cacheinfo(cpu)); + return allocate_cache_info(cpu); +} + +int detect_cache_attributes(unsigned int cpu) +{ + int ret; + + ret = init_level_allocate_ci(cpu); if (ret) - goto free_ci; + return ret; + + /* + * If LLC is valid the cache leaves were already populated so just go to + * update the cpu map. + */ + if (!last_level_cache_is_valid(cpu)) { + /* + * populate_cache_leaves() may completely setup the cache leaves and + * shared_cpu_map or it may leave it partially setup. + */ + ret = populate_cache_leaves(cpu); + if (ret) + goto free_ci; + } + /* * For systems using DT for cache hierarchy, fw_token * and shared_cpu_map will be set up here only if they are @@ -613,7 +900,6 @@ static int cache_add_dev(unsigned int cpu) int rc; struct device *ci_dev, *parent; struct cacheinfo *this_leaf; - struct cpu_cacheinfo *this_cpu_ci = get_cpu_cacheinfo(cpu); const struct attribute_group **cache_groups; rc = cpu_cache_sysfs_init(cpu); @@ -622,7 +908,7 @@ static int cache_add_dev(unsigned int cpu) parent = per_cpu_cache_dev(cpu); for (i = 0; i < cache_leaves(cpu); i++) { - this_leaf = this_cpu_ci->info_list + i; + this_leaf = per_cpu_cacheinfo_idx(cpu, i); if (this_leaf->disable_sysfs) continue; if (this_leaf->type == CACHE_TYPE_NOCACHE) @@ -644,24 +930,111 @@ err: return rc; } +static unsigned int cpu_map_shared_cache(bool online, unsigned int cpu, + cpumask_t **map) +{ + struct cacheinfo *llc, *sib_llc; + unsigned int sibling; + + if (!last_level_cache_is_valid(cpu)) + return 0; + + llc = per_cpu_cacheinfo_idx(cpu, cache_leaves(cpu) - 1); + + if (llc->type != CACHE_TYPE_DATA && llc->type != CACHE_TYPE_UNIFIED) + return 0; + + if (online) { + *map = &llc->shared_cpu_map; + return cpumask_weight(*map); + } + + /* shared_cpu_map of offlined CPU will be cleared, so use sibling map */ + for_each_cpu(sibling, &llc->shared_cpu_map) { + if (sibling == cpu || !last_level_cache_is_valid(sibling)) + continue; + sib_llc = per_cpu_cacheinfo_idx(sibling, cache_leaves(sibling) - 1); + *map = &sib_llc->shared_cpu_map; + return cpumask_weight(*map); + } + + return 0; +} + +/* + * Calculate the size of the per-CPU data cache slice. This can be + * used to estimate the size of the data cache slice that can be used + * by one CPU under ideal circumstances. UNIFIED caches are counted + * in addition to DATA caches. So, please consider code cache usage + * when use the result. + * + * Because the cache inclusive/non-inclusive information isn't + * available, we just use the size of the per-CPU slice of LLC to make + * the result more predictable across architectures. + */ +static void update_per_cpu_data_slice_size_cpu(unsigned int cpu) +{ + struct cpu_cacheinfo *ci; + struct cacheinfo *llc; + unsigned int nr_shared; + + if (!last_level_cache_is_valid(cpu)) + return; + + ci = ci_cacheinfo(cpu); + llc = per_cpu_cacheinfo_idx(cpu, cache_leaves(cpu) - 1); + + if (llc->type != CACHE_TYPE_DATA && llc->type != CACHE_TYPE_UNIFIED) + return; + + nr_shared = cpumask_weight(&llc->shared_cpu_map); + if (nr_shared) + ci->per_cpu_data_slice_size = llc->size / nr_shared; +} + +static void update_per_cpu_data_slice_size(bool cpu_online, unsigned int cpu, + cpumask_t *cpu_map) +{ + unsigned int icpu; + + for_each_cpu(icpu, cpu_map) { + if (!cpu_online && icpu == cpu) + continue; + update_per_cpu_data_slice_size_cpu(icpu); + setup_pcp_cacheinfo(icpu); + } +} + static int cacheinfo_cpu_online(unsigned int cpu) { int rc = detect_cache_attributes(cpu); + cpumask_t *cpu_map; if (rc) return rc; rc = cache_add_dev(cpu); if (rc) - free_cache_attributes(cpu); + goto err; + if (cpu_map_shared_cache(true, cpu, &cpu_map)) + update_per_cpu_data_slice_size(true, cpu, cpu_map); + return 0; +err: + free_cache_attributes(cpu); return rc; } static int cacheinfo_cpu_pre_down(unsigned int cpu) { + cpumask_t *cpu_map; + unsigned int nr_shared; + + nr_shared = cpu_map_shared_cache(false, cpu, &cpu_map); if (cpumask_test_and_clear_cpu(cpu, &cache_dev_map)) cpu_cache_sysfs_exit(cpu); free_cache_attributes(cpu); + if (nr_shared > 1) + update_per_cpu_data_slice_size(false, cpu, cpu_map); return 0; } diff --git a/drivers/base/class.c b/drivers/base/class.c index 7476f393df97..2526c57d924e 100644 --- a/drivers/base/class.c +++ b/drivers/base/class.c @@ -16,12 +16,56 @@ #include <linux/kdev_t.h> #include <linux/err.h> #include <linux/slab.h> -#include <linux/genhd.h> +#include <linux/blkdev.h> #include <linux/mutex.h> #include "base.h" +/* /sys/class */ +static struct kset *class_kset; + #define to_class_attr(_attr) container_of(_attr, struct class_attribute, attr) +/** + * class_to_subsys - Turn a struct class into a struct subsys_private + * + * @class: pointer to the struct bus_type to look up + * + * The driver core internals need to work on the subsys_private structure, not + * the external struct class pointer. This function walks the list of + * registered classes in the system and finds the matching one and returns the + * internal struct subsys_private that relates to that class. + * + * Note, the reference count of the return value is INCREMENTED if it is not + * NULL. A call to subsys_put() must be done when finished with the pointer in + * order for it to be properly freed. + */ +struct subsys_private *class_to_subsys(const struct class *class) +{ + struct subsys_private *sp = NULL; + struct kobject *kobj; + + if (!class || !class_kset) + return NULL; + + spin_lock(&class_kset->list_lock); + + if (list_empty(&class_kset->list)) + goto done; + + list_for_each_entry(kobj, &class_kset->list, entry) { + struct kset *kset = container_of(kobj, struct kset, kobj); + + sp = container_of_const(kset, struct subsys_private, subsys); + if (sp->class == class) + goto done; + } + sp = NULL; +done: + sp = subsys_get(sp); + spin_unlock(&class_kset->list_lock); + return sp; +} + static ssize_t class_attr_show(struct kobject *kobj, struct attribute *attr, char *buf) { @@ -49,7 +93,7 @@ static ssize_t class_attr_store(struct kobject *kobj, struct attribute *attr, static void class_release(struct kobject *kobj) { struct subsys_private *cp = to_subsys_private(kobj); - struct class *class = cp->class; + const struct class *class = cp->class; pr_debug("class '%s': release.\n", class->name); @@ -59,13 +103,14 @@ static void class_release(struct kobject *kobj) pr_debug("class '%s' does not have a release() function, " "be careful\n", class->name); + lockdep_unregister_key(&cp->lock_key); kfree(cp); } -static const struct kobj_ns_type_operations *class_child_ns_type(struct kobject *kobj) +static const struct kobj_ns_type_operations *class_child_ns_type(const struct kobject *kobj) { - struct subsys_private *cp = to_subsys_private(kobj); - struct class *class = cp->class; + const struct subsys_private *cp = to_subsys_private(kobj); + const struct class *class = cp->class; return class->ns_type; } @@ -75,48 +120,40 @@ static const struct sysfs_ops class_sysfs_ops = { .store = class_attr_store, }; -static struct kobj_type class_ktype = { +static const struct kobj_type class_ktype = { .sysfs_ops = &class_sysfs_ops, .release = class_release, .child_ns_type = class_child_ns_type, }; -/* Hotplug events for classes go to the class subsys */ -static struct kset *class_kset; - - -int class_create_file_ns(struct class *cls, const struct class_attribute *attr, +int class_create_file_ns(const struct class *cls, const struct class_attribute *attr, const void *ns) { + struct subsys_private *sp = class_to_subsys(cls); int error; - if (cls) - error = sysfs_create_file_ns(&cls->p->subsys.kobj, - &attr->attr, ns); - else - error = -EINVAL; + if (!sp) + return -EINVAL; + + error = sysfs_create_file_ns(&sp->subsys.kobj, &attr->attr, ns); + subsys_put(sp); + return error; } +EXPORT_SYMBOL_GPL(class_create_file_ns); -void class_remove_file_ns(struct class *cls, const struct class_attribute *attr, +void class_remove_file_ns(const struct class *cls, const struct class_attribute *attr, const void *ns) { - if (cls) - sysfs_remove_file_ns(&cls->p->subsys.kobj, &attr->attr, ns); -} + struct subsys_private *sp = class_to_subsys(cls); -static struct class *class_get(struct class *cls) -{ - if (cls) - kset_get(&cls->p->subsys); - return cls; -} + if (!sp) + return; -static void class_put(struct class *cls) -{ - if (cls) - kset_put(&cls->p->subsys); + sysfs_remove_file_ns(&sp->subsys.kobj, &attr->attr, ns); + subsys_put(sp); } +EXPORT_SYMBOL_GPL(class_remove_file_ns); static struct device *klist_class_to_dev(struct klist_node *n) { @@ -138,82 +175,85 @@ static void klist_class_dev_put(struct klist_node *n) put_device(dev); } -static int class_add_groups(struct class *cls, - const struct attribute_group **groups) -{ - return sysfs_create_groups(&cls->p->subsys.kobj, groups); -} - -static void class_remove_groups(struct class *cls, - const struct attribute_group **groups) -{ - return sysfs_remove_groups(&cls->p->subsys.kobj, groups); -} - -int __class_register(struct class *cls, struct lock_class_key *key) +int class_register(const struct class *cls) { struct subsys_private *cp; + struct lock_class_key *key; int error; pr_debug("device class '%s': registering\n", cls->name); + if (cls->ns_type && !cls->namespace) { + pr_err("%s: class '%s' does not have namespace\n", + __func__, cls->name); + return -EINVAL; + } + if (!cls->ns_type && cls->namespace) { + pr_err("%s: class '%s' does not have ns_type\n", + __func__, cls->name); + return -EINVAL; + } + cp = kzalloc(sizeof(*cp), GFP_KERNEL); if (!cp) return -ENOMEM; klist_init(&cp->klist_devices, klist_class_dev_get, klist_class_dev_put); INIT_LIST_HEAD(&cp->interfaces); kset_init(&cp->glue_dirs); + key = &cp->lock_key; + lockdep_register_key(key); __mutex_init(&cp->mutex, "subsys mutex", key); error = kobject_set_name(&cp->subsys.kobj, "%s", cls->name); - if (error) { - kfree(cp); - return error; - } - - /* set the default /sys/dev directory for devices of this class */ - if (!cls->dev_kobj) - cls->dev_kobj = sysfs_dev_char_kobj; + if (error) + goto err_out; -#if defined(CONFIG_BLOCK) - /* let the block class directory show up in the root of sysfs */ - if (!sysfs_deprecated || cls != &block_class) - cp->subsys.kobj.kset = class_kset; -#else cp->subsys.kobj.kset = class_kset; -#endif cp->subsys.kobj.ktype = &class_ktype; cp->class = cls; - cls->p = cp; error = kset_register(&cp->subsys); + if (error) + goto err_out; + + error = sysfs_create_groups(&cp->subsys.kobj, cls->class_groups); if (error) { - kfree(cp); - return error; + kobject_del(&cp->subsys.kobj); + kfree_const(cp->subsys.kobj.name); + goto err_out; } - error = class_add_groups(class_get(cls), cls->class_groups); - class_put(cls); + return 0; + +err_out: + lockdep_unregister_key(key); + kfree(cp); return error; } -EXPORT_SYMBOL_GPL(__class_register); +EXPORT_SYMBOL_GPL(class_register); -void class_unregister(struct class *cls) +void class_unregister(const struct class *cls) { + struct subsys_private *sp = class_to_subsys(cls); + + if (!sp) + return; + pr_debug("device class '%s': unregistering\n", cls->name); - class_remove_groups(cls, cls->class_groups); - kset_unregister(&cls->p->subsys); + + sysfs_remove_groups(&sp->subsys.kobj, cls->class_groups); + kset_unregister(&sp->subsys); + subsys_put(sp); } +EXPORT_SYMBOL_GPL(class_unregister); -static void class_create_release(struct class *cls) +static void class_create_release(const struct class *cls) { pr_debug("%s called for %s\n", __func__, cls->name); kfree(cls); } /** - * __class_create - create a struct class structure - * @owner: pointer to the module that is to "own" this struct class + * class_create - create a struct class structure * @name: pointer to a string for the name of this class. - * @key: the lock_class_key for this class; used by mutex lock debugging * * This is used to create a struct class pointer that can then be used * in calls to device_create(). @@ -223,8 +263,7 @@ static void class_create_release(struct class *cls) * Note, the pointer created here is to be destroyed when finished by * making a call to class_destroy(). */ -struct class *__class_create(struct module *owner, const char *name, - struct lock_class_key *key) +struct class *class_create(const char *name) { struct class *cls; int retval; @@ -236,10 +275,9 @@ struct class *__class_create(struct module *owner, const char *name, } cls->name = name; - cls->owner = owner; cls->class_release = class_create_release; - retval = __class_register(cls, key); + retval = class_register(cls); if (retval) goto error; @@ -249,7 +287,7 @@ error: kfree(cls); return ERR_PTR(retval); } -EXPORT_SYMBOL_GPL(__class_create); +EXPORT_SYMBOL_GPL(class_create); /** * class_destroy - destroys a struct class structure @@ -258,13 +296,14 @@ EXPORT_SYMBOL_GPL(__class_create); * Note, the pointer to be destroyed must have been created with a call * to class_create(). */ -void class_destroy(struct class *cls) +void class_destroy(const struct class *cls) { - if ((cls == NULL) || (IS_ERR(cls))) + if (IS_ERR_OR_NULL(cls)) return; class_unregister(cls); } +EXPORT_SYMBOL_GPL(class_destroy); /** * class_dev_iter_init - initialize class device iterator @@ -278,15 +317,24 @@ void class_destroy(struct class *cls) * otherwise if it is NULL, the iteration starts at the beginning of * the list. */ -void class_dev_iter_init(struct class_dev_iter *iter, struct class *class, - struct device *start, const struct device_type *type) +void class_dev_iter_init(struct class_dev_iter *iter, const struct class *class, + const struct device *start, const struct device_type *type) { + struct subsys_private *sp = class_to_subsys(class); struct klist_node *start_knode = NULL; + memset(iter, 0, sizeof(*iter)); + if (!sp) { + pr_crit("%s: class %p was not registered yet\n", + __func__, class); + return; + } + if (start) start_knode = &start->p->knode_class; - klist_iter_init_node(&class->p->klist_devices, &iter->ki, start_knode); + klist_iter_init_node(&sp->klist_devices, &iter->ki, start_knode); iter->type = type; + iter->sp = sp; } EXPORT_SYMBOL_GPL(class_dev_iter_init); @@ -307,6 +355,9 @@ struct device *class_dev_iter_next(struct class_dev_iter *iter) struct klist_node *knode; struct device *dev; + if (!iter->sp) + return NULL; + while (1) { knode = klist_next(&iter->ki); if (!knode) @@ -328,6 +379,7 @@ EXPORT_SYMBOL_GPL(class_dev_iter_next); void class_dev_iter_exit(struct class_dev_iter *iter) { klist_iter_exit(&iter->ki); + subsys_put(iter->sp); } EXPORT_SYMBOL_GPL(class_dev_iter_exit); @@ -349,17 +401,18 @@ EXPORT_SYMBOL_GPL(class_dev_iter_exit); * @fn is allowed to do anything including calling back into class * code. There's no locking restriction. */ -int class_for_each_device(struct class *class, struct device *start, - void *data, int (*fn)(struct device *, void *)) +int class_for_each_device(const struct class *class, const struct device *start, + void *data, device_iter_t fn) { + struct subsys_private *sp = class_to_subsys(class); struct class_dev_iter iter; struct device *dev; int error = 0; if (!class) return -EINVAL; - if (!class->p) { - WARN(1, "%s called for class '%s' before it was initialized", + if (!sp) { + WARN(1, "%s called for class '%s' before it was registered", __func__, class->name); return -EINVAL; } @@ -371,6 +424,7 @@ int class_for_each_device(struct class *class, struct device *start, break; } class_dev_iter_exit(&iter); + subsys_put(sp); return error; } @@ -396,17 +450,17 @@ EXPORT_SYMBOL_GPL(class_for_each_device); * @match is allowed to do anything including calling back into class * code. There's no locking restriction. */ -struct device *class_find_device(struct class *class, struct device *start, - const void *data, - int (*match)(struct device *, const void *)) +struct device *class_find_device(const struct class *class, const struct device *start, + const void *data, device_match_t match) { + struct subsys_private *sp = class_to_subsys(class); struct class_dev_iter iter; struct device *dev; if (!class) return NULL; - if (!class->p) { - WARN(1, "%s called for class '%s' before it was initialized", + if (!sp) { + WARN(1, "%s called for class '%s' before it was registered", __func__, class->name); return NULL; } @@ -419,6 +473,7 @@ struct device *class_find_device(struct class *class, struct device *start, } } class_dev_iter_exit(&iter); + subsys_put(sp); return dev; } @@ -426,54 +481,74 @@ EXPORT_SYMBOL_GPL(class_find_device); int class_interface_register(struct class_interface *class_intf) { - struct class *parent; + struct subsys_private *sp; + const struct class *parent; struct class_dev_iter iter; struct device *dev; if (!class_intf || !class_intf->class) return -ENODEV; - parent = class_get(class_intf->class); - if (!parent) + parent = class_intf->class; + sp = class_to_subsys(parent); + if (!sp) return -EINVAL; - mutex_lock(&parent->p->mutex); - list_add_tail(&class_intf->node, &parent->p->interfaces); + /* + * Reference in sp is now incremented and will be dropped when + * the interface is removed in the call to class_interface_unregister() + */ + + mutex_lock(&sp->mutex); + list_add_tail(&class_intf->node, &sp->interfaces); if (class_intf->add_dev) { class_dev_iter_init(&iter, parent, NULL, NULL); while ((dev = class_dev_iter_next(&iter))) - class_intf->add_dev(dev, class_intf); + class_intf->add_dev(dev); class_dev_iter_exit(&iter); } - mutex_unlock(&parent->p->mutex); + mutex_unlock(&sp->mutex); return 0; } +EXPORT_SYMBOL_GPL(class_interface_register); void class_interface_unregister(struct class_interface *class_intf) { - struct class *parent = class_intf->class; + struct subsys_private *sp; + const struct class *parent = class_intf->class; struct class_dev_iter iter; struct device *dev; if (!parent) return; - mutex_lock(&parent->p->mutex); + sp = class_to_subsys(parent); + if (!sp) + return; + + mutex_lock(&sp->mutex); list_del_init(&class_intf->node); if (class_intf->remove_dev) { class_dev_iter_init(&iter, parent, NULL, NULL); while ((dev = class_dev_iter_next(&iter))) - class_intf->remove_dev(dev, class_intf); + class_intf->remove_dev(dev); class_dev_iter_exit(&iter); } - mutex_unlock(&parent->p->mutex); + mutex_unlock(&sp->mutex); - class_put(parent); + /* + * Decrement the reference count twice, once for the class_to_subsys() + * call in the start of this function, and the second one from the + * reference increment in class_interface_register() + */ + subsys_put(sp); + subsys_put(sp); } +EXPORT_SYMBOL_GPL(class_interface_unregister); -ssize_t show_class_attr_string(struct class *class, - struct class_attribute *attr, char *buf) +ssize_t show_class_attr_string(const struct class *class, + const struct class_attribute *attr, char *buf) { struct class_attribute_string *cs; @@ -526,30 +601,10 @@ EXPORT_SYMBOL_GPL(class_compat_unregister); * a bus device * @cls: the compatibility class * @dev: the target bus device - * @device_link: an optional device to which a "device" link should be created */ -int class_compat_create_link(struct class_compat *cls, struct device *dev, - struct device *device_link) +int class_compat_create_link(struct class_compat *cls, struct device *dev) { - int error; - - error = sysfs_create_link(cls->kobj, &dev->kobj, dev_name(dev)); - if (error) - return error; - - /* - * Optionally add a "device" link (typically to the parent), as a - * class device would have one and we want to provide as much - * backwards compatibility as possible. - */ - if (device_link) { - error = sysfs_create_link(&dev->kobj, &device_link->kobj, - "device"); - if (error) - sysfs_remove_link(cls->kobj, dev_name(dev)); - } - - return error; + return sysfs_create_link(cls->kobj, &dev->kobj, dev_name(dev)); } EXPORT_SYMBOL_GPL(class_compat_create_link); @@ -558,18 +613,38 @@ EXPORT_SYMBOL_GPL(class_compat_create_link); * a bus device * @cls: the compatibility class * @dev: the target bus device - * @device_link: an optional device to which a "device" link was previously - * created */ -void class_compat_remove_link(struct class_compat *cls, struct device *dev, - struct device *device_link) +void class_compat_remove_link(struct class_compat *cls, struct device *dev) { - if (device_link) - sysfs_remove_link(&dev->kobj, "device"); sysfs_remove_link(cls->kobj, dev_name(dev)); } EXPORT_SYMBOL_GPL(class_compat_remove_link); +/** + * class_is_registered - determine if at this moment in time, a class is + * registered in the driver core or not. + * @class: the class to check + * + * Returns a boolean to state if the class is registered in the driver core + * or not. Note that the value could switch right after this call is made, + * so only use this in places where you "know" it is safe to do so (usually + * to determine if the specific class has been registered yet or not). + * + * Be careful in using this. + */ +bool class_is_registered(const struct class *class) +{ + struct subsys_private *sp = class_to_subsys(class); + bool is_initialized = false; + + if (sp) { + is_initialized = true; + subsys_put(sp); + } + return is_initialized; +} +EXPORT_SYMBOL_GPL(class_is_registered); + int __init classes_init(void) { class_kset = kset_create_and_add("class", NULL, NULL); @@ -577,11 +652,3 @@ int __init classes_init(void) return -ENOMEM; return 0; } - -EXPORT_SYMBOL_GPL(class_create_file_ns); -EXPORT_SYMBOL_GPL(class_remove_file_ns); -EXPORT_SYMBOL_GPL(class_unregister); -EXPORT_SYMBOL_GPL(class_destroy); - -EXPORT_SYMBOL_GPL(class_interface_register); -EXPORT_SYMBOL_GPL(class_interface_unregister); diff --git a/drivers/base/component.c b/drivers/base/component.c index 5e79299f6c3f..024ad9471b8a 100644 --- a/drivers/base/component.c +++ b/drivers/base/component.c @@ -1,17 +1,12 @@ // SPDX-License-Identifier: GPL-2.0 /* * Componentized device handling. - * - * This is work in progress. We gather up the component devices into a list, - * and bind them when instructed. At the moment, we're specific to the DRM - * subsystem, and only handles one master device, but this doesn't have to be - * the case. */ #include <linux/component.h> #include <linux/device.h> -#include <linux/kref.h> #include <linux/list.h> #include <linux/mutex.h> +#include <linux/of.h> #include <linux/slab.h> #include <linux/debugfs.h> @@ -58,7 +53,7 @@ struct component_match { struct component_match_array *compare; }; -struct master { +struct aggregate_device { struct list_head node; bool bound; @@ -69,7 +64,7 @@ struct master { struct component { struct list_head node; - struct master *master; + struct aggregate_device *adev; bool bound; const struct component_ops *ops; @@ -79,7 +74,7 @@ struct component { static DEFINE_MUTEX(component_mutex); static LIST_HEAD(component_list); -static LIST_HEAD(masters); +static LIST_HEAD(aggregate_devices); #ifdef CONFIG_DEBUG_FS @@ -87,22 +82,22 @@ static struct dentry *component_debugfs_dir; static int component_devices_show(struct seq_file *s, void *data) { - struct master *m = s->private; + struct aggregate_device *m = s->private; struct component_match *match = m->match; size_t i; mutex_lock(&component_mutex); - seq_printf(s, "%-40s %20s\n", "master name", "status"); - seq_puts(s, "-------------------------------------------------------------\n"); - seq_printf(s, "%-40s %20s\n\n", + seq_printf(s, "%-50s %20s\n", "aggregate_device name", "status"); + seq_puts(s, "-----------------------------------------------------------------------\n"); + seq_printf(s, "%-50s %20s\n\n", dev_name(m->parent), m->bound ? "bound" : "not bound"); - seq_printf(s, "%-40s %20s\n", "device name", "status"); - seq_puts(s, "-------------------------------------------------------------\n"); + seq_printf(s, "%-50s %20s\n", "device name", "status"); + seq_puts(s, "-----------------------------------------------------------------------\n"); for (i = 0; i < match->num; i++) { struct component *component = match->compare[i].component; - seq_printf(s, "%-40s %20s\n", + seq_printf(s, "%-50s %20s\n", component ? dev_name(component->dev) : "(unknown)", component ? (component->bound ? "bound" : "not bound") : "not registered"); } @@ -122,46 +117,46 @@ static int __init component_debug_init(void) core_initcall(component_debug_init); -static void component_master_debugfs_add(struct master *m) +static void component_debugfs_add(struct aggregate_device *m) { debugfs_create_file(dev_name(m->parent), 0444, component_debugfs_dir, m, &component_devices_fops); } -static void component_master_debugfs_del(struct master *m) +static void component_debugfs_del(struct aggregate_device *m) { - debugfs_remove(debugfs_lookup(dev_name(m->parent), component_debugfs_dir)); + debugfs_lookup_and_remove(dev_name(m->parent), component_debugfs_dir); } #else -static void component_master_debugfs_add(struct master *m) +static void component_debugfs_add(struct aggregate_device *m) { } -static void component_master_debugfs_del(struct master *m) +static void component_debugfs_del(struct aggregate_device *m) { } #endif -static struct master *__master_find(struct device *parent, +static struct aggregate_device *__aggregate_find(struct device *parent, const struct component_master_ops *ops) { - struct master *m; + struct aggregate_device *m; - list_for_each_entry(m, &masters, node) + list_for_each_entry(m, &aggregate_devices, node) if (m->parent == parent && (!ops || m->ops == ops)) return m; return NULL; } -static struct component *find_component(struct master *master, +static struct component *find_component(struct aggregate_device *adev, struct component_match_array *mc) { struct component *c; list_for_each_entry(c, &component_list, node) { - if (c->master && c->master != master) + if (c->adev && c->adev != adev) continue; if (mc->compare && mc->compare(c->dev, mc->data)) @@ -175,101 +170,103 @@ static struct component *find_component(struct master *master, return NULL; } -static int find_components(struct master *master) +static int find_components(struct aggregate_device *adev) { - struct component_match *match = master->match; + struct component_match *match = adev->match; size_t i; int ret = 0; /* * Scan the array of match functions and attach - * any components which are found to this master. + * any components which are found to this adev. */ for (i = 0; i < match->num; i++) { struct component_match_array *mc = &match->compare[i]; struct component *c; - dev_dbg(master->parent, "Looking for component %zu\n", i); + dev_dbg(adev->parent, "Looking for component %zu\n", i); if (match->compare[i].component) continue; - c = find_component(master, mc); + c = find_component(adev, mc); if (!c) { ret = -ENXIO; break; } - dev_dbg(master->parent, "found component %s, duplicate %u\n", dev_name(c->dev), !!c->master); + dev_dbg(adev->parent, "found component %s, duplicate %u\n", + dev_name(c->dev), !!c->adev); - /* Attach this component to the master */ - match->compare[i].duplicate = !!c->master; + /* Attach this component to the adev */ + match->compare[i].duplicate = !!c->adev; match->compare[i].component = c; - c->master = master; + c->adev = adev; } return ret; } -/* Detach component from associated master */ -static void remove_component(struct master *master, struct component *c) +/* Detach component from associated aggregate_device */ +static void remove_component(struct aggregate_device *adev, struct component *c) { size_t i; - /* Detach the component from this master. */ - for (i = 0; i < master->match->num; i++) - if (master->match->compare[i].component == c) - master->match->compare[i].component = NULL; + /* Detach the component from this adev. */ + for (i = 0; i < adev->match->num; i++) + if (adev->match->compare[i].component == c) + adev->match->compare[i].component = NULL; } /* - * Try to bring up a master. If component is NULL, we're interested in - * this master, otherwise it's a component which must be present to try - * and bring up the master. + * Try to bring up an aggregate device. If component is NULL, we're interested + * in this aggregate device, otherwise it's a component which must be present + * to try and bring up the aggregate device. * * Returns 1 for successful bringup, 0 if not ready, or -ve errno. */ -static int try_to_bring_up_master(struct master *master, +static int try_to_bring_up_aggregate_device(struct aggregate_device *adev, struct component *component) { int ret; - dev_dbg(master->parent, "trying to bring up master\n"); + dev_dbg(adev->parent, "trying to bring up adev\n"); - if (find_components(master)) { - dev_dbg(master->parent, "master has incomplete components\n"); + if (find_components(adev)) { + dev_dbg(adev->parent, "master has incomplete components\n"); return 0; } - if (component && component->master != master) { - dev_dbg(master->parent, "master is not for this component (%s)\n", + if (component && component->adev != adev) { + dev_dbg(adev->parent, "master is not for this component (%s)\n", dev_name(component->dev)); return 0; } - if (!devres_open_group(master->parent, NULL, GFP_KERNEL)) + if (!devres_open_group(adev->parent, adev, GFP_KERNEL)) return -ENOMEM; /* Found all components */ - ret = master->ops->bind(master->parent); + ret = adev->ops->bind(adev->parent); if (ret < 0) { - devres_release_group(master->parent, NULL); + devres_release_group(adev->parent, NULL); if (ret != -EPROBE_DEFER) - dev_info(master->parent, "master bind failed: %d\n", ret); + dev_info(adev->parent, "adev bind failed: %d\n", ret); return ret; } - master->bound = true; + devres_close_group(adev->parent, NULL); + adev->bound = true; return 1; } static int try_to_bring_up_masters(struct component *component) { - struct master *m; + struct aggregate_device *adev; int ret = 0; - list_for_each_entry(m, &masters, node) { - if (!m->bound) { - ret = try_to_bring_up_master(m, component); + list_for_each_entry(adev, &aggregate_devices, node) { + if (!adev->bound) { + ret = try_to_bring_up_aggregate_device(adev, component); if (ret != 0) break; } @@ -278,15 +275,72 @@ static int try_to_bring_up_masters(struct component *component) return ret; } -static void take_down_master(struct master *master) +static void take_down_aggregate_device(struct aggregate_device *adev) { - if (master->bound) { - master->ops->unbind(master->parent); - devres_release_group(master->parent, NULL); - master->bound = false; + if (adev->bound) { + adev->ops->unbind(adev->parent); + devres_release_group(adev->parent, adev); + adev->bound = false; } } +/** + * component_compare_of - A common component compare function for of_node + * @dev: component device + * @data: @compare_data from component_match_add_release() + * + * A common compare function when compare_data is device of_node. e.g. + * component_match_add_release(masterdev, &match, component_release_of, + * component_compare_of, component_dev_of_node) + */ +int component_compare_of(struct device *dev, void *data) +{ + return device_match_of_node(dev, data); +} +EXPORT_SYMBOL_GPL(component_compare_of); + +/** + * component_release_of - A common component release function for of_node + * @dev: component device + * @data: @compare_data from component_match_add_release() + * + * About the example, Please see component_compare_of(). + */ +void component_release_of(struct device *dev, void *data) +{ + of_node_put(data); +} +EXPORT_SYMBOL_GPL(component_release_of); + +/** + * component_compare_dev - A common component compare function for dev + * @dev: component device + * @data: @compare_data from component_match_add_release() + * + * A common compare function when compare_data is struce device. e.g. + * component_match_add(masterdev, &match, component_compare_dev, component_dev) + */ +int component_compare_dev(struct device *dev, void *data) +{ + return dev == data; +} +EXPORT_SYMBOL_GPL(component_compare_dev); + +/** + * component_compare_dev_name - A common component compare function for device name + * @dev: component device + * @data: @compare_data from component_match_add_release() + * + * A common compare function when compare_data is device name string. e.g. + * component_match_add(masterdev, &match, component_compare_dev_name, + * "component_dev_name") + */ +int component_compare_dev_name(struct device *dev, void *data) +{ + return device_match_name(dev, data); +} +EXPORT_SYMBOL_GPL(component_compare_dev_name); + static void devm_component_match_release(struct device *parent, void *res) { struct component_match *match = res; @@ -324,7 +378,7 @@ static int component_match_realloc(struct component_match *match, size_t num) return 0; } -static void __component_match_add(struct device *master, +static void __component_match_add(struct device *parent, struct component_match **matchptr, void (*release)(struct device *, void *), int (*compare)(struct device *, void *), @@ -344,7 +398,7 @@ static void __component_match_add(struct device *master, return; } - devres_add(master, match); + devres_add(parent, match); *matchptr = match; } @@ -370,13 +424,13 @@ static void __component_match_add(struct device *master, /** * component_match_add_release - add a component match entry with release callback - * @master: device with the aggregate driver + * @parent: parent device of the aggregate driver * @matchptr: pointer to the list of component matches * @release: release function for @compare_data * @compare: compare function to match against all components * @compare_data: opaque pointer passed to the @compare function * - * Adds a new component match to the list stored in @matchptr, which the @master + * Adds a new component match to the list stored in @matchptr, which the * aggregate driver needs to function. The list of component matches pointed to * by @matchptr must be initialized to NULL before adding the first match. This * only matches against components added with component_add(). @@ -388,24 +442,24 @@ static void __component_match_add(struct device *master, * * See also component_match_add() and component_match_add_typed(). */ -void component_match_add_release(struct device *master, +void component_match_add_release(struct device *parent, struct component_match **matchptr, void (*release)(struct device *, void *), int (*compare)(struct device *, void *), void *compare_data) { - __component_match_add(master, matchptr, release, compare, NULL, + __component_match_add(parent, matchptr, release, compare, NULL, compare_data); } EXPORT_SYMBOL(component_match_add_release); /** * component_match_add_typed - add a component match entry for a typed component - * @master: device with the aggregate driver + * @parent: parent device of the aggregate driver * @matchptr: pointer to the list of component matches * @compare_typed: compare function to match against all typed components * @compare_data: opaque pointer passed to the @compare function * - * Adds a new component match to the list stored in @matchptr, which the @master + * Adds a new component match to the list stored in @matchptr, which the * aggregate driver needs to function. The list of component matches pointed to * by @matchptr must be initialized to NULL before adding the first match. This * only matches against components added with component_add_typed(). @@ -415,32 +469,32 @@ EXPORT_SYMBOL(component_match_add_release); * * See also component_match_add_release() and component_match_add_typed(). */ -void component_match_add_typed(struct device *master, +void component_match_add_typed(struct device *parent, struct component_match **matchptr, int (*compare_typed)(struct device *, int, void *), void *compare_data) { - __component_match_add(master, matchptr, NULL, NULL, compare_typed, + __component_match_add(parent, matchptr, NULL, NULL, compare_typed, compare_data); } EXPORT_SYMBOL(component_match_add_typed); -static void free_master(struct master *master) +static void free_aggregate_device(struct aggregate_device *adev) { - struct component_match *match = master->match; + struct component_match *match = adev->match; int i; - component_master_debugfs_del(master); - list_del(&master->node); + component_debugfs_del(adev); + list_del(&adev->node); if (match) { for (i = 0; i < match->num; i++) { struct component *c = match->compare[i].component; if (c) - c->master = NULL; + c->adev = NULL; } } - kfree(master); + kfree(adev); } /** @@ -459,7 +513,7 @@ int component_master_add_with_match(struct device *parent, const struct component_master_ops *ops, struct component_match *match) { - struct master *master; + struct aggregate_device *adev; int ret; /* Reallocate the match array for its true size */ @@ -467,23 +521,23 @@ int component_master_add_with_match(struct device *parent, if (ret) return ret; - master = kzalloc(sizeof(*master), GFP_KERNEL); - if (!master) + adev = kzalloc(sizeof(*adev), GFP_KERNEL); + if (!adev) return -ENOMEM; - master->parent = parent; - master->ops = ops; - master->match = match; + adev->parent = parent; + adev->ops = ops; + adev->match = match; - component_master_debugfs_add(master); - /* Add to the list of available masters. */ + component_debugfs_add(adev); + /* Add to the list of available aggregate devices. */ mutex_lock(&component_mutex); - list_add(&master->node, &masters); + list_add(&adev->node, &aggregate_devices); - ret = try_to_bring_up_master(master, NULL); + ret = try_to_bring_up_aggregate_device(adev, NULL); if (ret < 0) - free_master(master); + free_aggregate_device(adev); mutex_unlock(&component_mutex); @@ -503,25 +557,43 @@ EXPORT_SYMBOL_GPL(component_master_add_with_match); void component_master_del(struct device *parent, const struct component_master_ops *ops) { - struct master *master; + struct aggregate_device *adev; mutex_lock(&component_mutex); - master = __master_find(parent, ops); - if (master) { - take_down_master(master); - free_master(master); + adev = __aggregate_find(parent, ops); + if (adev) { + take_down_aggregate_device(adev); + free_aggregate_device(adev); } mutex_unlock(&component_mutex); } EXPORT_SYMBOL_GPL(component_master_del); +bool component_master_is_bound(struct device *parent, + const struct component_master_ops *ops) +{ + struct aggregate_device *adev; + + guard(mutex)(&component_mutex); + adev = __aggregate_find(parent, ops); + if (!adev) + return 0; + + return adev->bound; +} +EXPORT_SYMBOL_GPL(component_master_is_bound); + static void component_unbind(struct component *component, - struct master *master, void *data) + struct aggregate_device *adev, void *data) { - WARN_ON(!component->bound); + if (WARN_ON(!component->bound)) + return; + + dev_dbg(adev->parent, "unbinding %s component %p (ops %ps)\n", + dev_name(component->dev), component, component->ops); if (component->ops && component->ops->unbind) - component->ops->unbind(component->dev, master->parent, data); + component->ops->unbind(component->dev, adev->parent, data); component->bound = false; /* Release all resources claimed in the binding of this component */ @@ -539,26 +611,26 @@ static void component_unbind(struct component *component, */ void component_unbind_all(struct device *parent, void *data) { - struct master *master; + struct aggregate_device *adev; struct component *c; size_t i; WARN_ON(!mutex_is_locked(&component_mutex)); - master = __master_find(parent, NULL); - if (!master) + adev = __aggregate_find(parent, NULL); + if (!adev) return; /* Unbind components in reverse order */ - for (i = master->match->num; i--; ) - if (!master->match->compare[i].duplicate) { - c = master->match->compare[i].component; - component_unbind(c, master, data); + for (i = adev->match->num; i--; ) + if (!adev->match->compare[i].duplicate) { + c = adev->match->compare[i].component; + component_unbind(c, adev, data); } } EXPORT_SYMBOL_GPL(component_unbind_all); -static int component_bind(struct component *component, struct master *master, +static int component_bind(struct component *component, struct aggregate_device *adev, void *data) { int ret; @@ -568,7 +640,7 @@ static int component_bind(struct component *component, struct master *master, * This allows us to roll-back a failed component without * affecting anything else. */ - if (!devres_open_group(master->parent, NULL, GFP_KERNEL)) + if (!devres_open_group(adev->parent, NULL, GFP_KERNEL)) return -ENOMEM; /* @@ -577,14 +649,14 @@ static int component_bind(struct component *component, struct master *master, * at the appropriate moment. */ if (!devres_open_group(component->dev, component, GFP_KERNEL)) { - devres_release_group(master->parent, NULL); + devres_release_group(adev->parent, NULL); return -ENOMEM; } - dev_dbg(master->parent, "binding %s (ops %ps)\n", + dev_dbg(adev->parent, "binding %s (ops %ps)\n", dev_name(component->dev), component->ops); - ret = component->ops->bind(component->dev, master->parent, data); + ret = component->ops->bind(component->dev, adev->parent, data); if (!ret) { component->bound = true; @@ -595,16 +667,16 @@ static int component_bind(struct component *component, struct master *master, * can clean those resources up independently. */ devres_close_group(component->dev, NULL); - devres_remove_group(master->parent, NULL); + devres_remove_group(adev->parent, NULL); - dev_info(master->parent, "bound %s (ops %ps)\n", + dev_info(adev->parent, "bound %s (ops %ps)\n", dev_name(component->dev), component->ops); } else { devres_release_group(component->dev, NULL); - devres_release_group(master->parent, NULL); + devres_release_group(adev->parent, NULL); if (ret != -EPROBE_DEFER) - dev_err(master->parent, "failed to bind %s (ops %ps): %d\n", + dev_err(adev->parent, "failed to bind %s (ops %ps): %d\n", dev_name(component->dev), component->ops, ret); } @@ -622,31 +694,31 @@ static int component_bind(struct component *component, struct master *master, */ int component_bind_all(struct device *parent, void *data) { - struct master *master; + struct aggregate_device *adev; struct component *c; size_t i; int ret = 0; WARN_ON(!mutex_is_locked(&component_mutex)); - master = __master_find(parent, NULL); - if (!master) + adev = __aggregate_find(parent, NULL); + if (!adev) return -EINVAL; /* Bind components in match order */ - for (i = 0; i < master->match->num; i++) - if (!master->match->compare[i].duplicate) { - c = master->match->compare[i].component; - ret = component_bind(c, master, data); + for (i = 0; i < adev->match->num; i++) + if (!adev->match->compare[i].duplicate) { + c = adev->match->compare[i].component; + ret = component_bind(c, adev, data); if (ret) break; } if (ret != 0) { for (; i > 0; i--) - if (!master->match->compare[i - 1].duplicate) { - c = master->match->compare[i - 1].component; - component_unbind(c, master, data); + if (!adev->match->compare[i - 1].duplicate) { + c = adev->match->compare[i - 1].component; + component_unbind(c, adev, data); } } @@ -675,8 +747,8 @@ static int __component_add(struct device *dev, const struct component_ops *ops, ret = try_to_bring_up_masters(component); if (ret < 0) { - if (component->master) - remove_component(component->master, component); + if (component->adev) + remove_component(component->adev, component); list_del(&component->node); kfree(component); @@ -697,7 +769,7 @@ static int __component_add(struct device *dev, const struct component_ops *ops, * component_bind_all(). See also &struct component_ops. * * @subcomponent must be nonzero and is used to differentiate between multiple - * components registerd on the same device @dev. These components are match + * components registered on the same device @dev. These components are match * using component_match_add_typed(). * * The component needs to be unregistered at driver unload/disconnect by @@ -727,7 +799,7 @@ EXPORT_SYMBOL_GPL(component_add_typed); * The component needs to be unregistered at driver unload/disconnect by * calling component_del(). * - * See also component_add_typed() for a variant that allows multipled different + * See also component_add_typed() for a variant that allows multiple different * components on the same device. */ int component_add(struct device *dev, const struct component_ops *ops) @@ -757,9 +829,9 @@ void component_del(struct device *dev, const struct component_ops *ops) break; } - if (component && component->master) { - take_down_master(component->master); - remove_component(component->master, component); + if (component && component->adev) { + take_down_aggregate_device(component->adev); + remove_component(component->adev, component); } mutex_unlock(&component_mutex); diff --git a/drivers/base/container.c b/drivers/base/container.c index 1ba42d2d3532..f40588ebc3f5 100644 --- a/drivers/base/container.c +++ b/drivers/base/container.c @@ -24,7 +24,7 @@ static int container_offline(struct device *dev) return cdev->offline ? cdev->offline(cdev) : 0; } -struct bus_type container_subsys = { +const struct bus_type container_subsys = { .name = CONTAINER_BUS_NAME, .dev_name = CONTAINER_BUS_NAME, .online = trivial_online, diff --git a/drivers/base/core.c b/drivers/base/core.c index cadcade65825..40de2f51a1b1 100644 --- a/drivers/base/core.c +++ b/drivers/base/core.c @@ -9,54 +9,49 @@ */ #include <linux/acpi.h> +#include <linux/blkdev.h> +#include <linux/cleanup.h> #include <linux/cpufreq.h> #include <linux/device.h> +#include <linux/dma-map-ops.h> /* for dma_default_coherent */ #include <linux/err.h> #include <linux/fwnode.h> #include <linux/init.h> -#include <linux/module.h> -#include <linux/slab.h> -#include <linux/string.h> #include <linux/kdev_t.h> +#include <linux/kstrtox.h> +#include <linux/module.h> +#include <linux/mutex.h> +#include <linux/netdevice.h> #include <linux/notifier.h> #include <linux/of.h> #include <linux/of_device.h> -#include <linux/genhd.h> -#include <linux/mutex.h> #include <linux/pm_runtime.h> -#include <linux/netdevice.h> -#include <linux/sched/signal.h> #include <linux/sched/mm.h> +#include <linux/sched/signal.h> +#include <linux/slab.h> +#include <linux/string_helpers.h> +#include <linux/swiotlb.h> #include <linux/sysfs.h> -#include <linux/dma-map-ops.h> /* for dma_default_coherent */ #include "base.h" +#include "physical_location.h" #include "power/power.h" -#ifdef CONFIG_SYSFS_DEPRECATED -#ifdef CONFIG_SYSFS_DEPRECATED_V2 -long sysfs_deprecated = 1; -#else -long sysfs_deprecated = 0; -#endif -static int __init sysfs_deprecated_setup(char *arg) -{ - return kstrtol(arg, 10, &sysfs_deprecated); -} -early_param("sysfs.deprecated", sysfs_deprecated_setup); -#endif - /* Device links support. */ static LIST_HEAD(deferred_sync); static unsigned int defer_sync_state_count = 1; static DEFINE_MUTEX(fwnode_link_lock); static bool fw_devlink_is_permissive(void); +static void __fw_devlink_link_to_consumers(struct device *dev); static bool fw_devlink_drv_reg_done; +static bool fw_devlink_best_effort; +static struct workqueue_struct *device_link_wq; /** - * fwnode_link_add - Create a link between two fwnode_handles. + * __fwnode_link_add - Create a link between two fwnode_handles. * @con: Consumer end of the link. * @sup: Supplier end of the link. + * @flags: Link flags. * * Create a fwnode link between fwnode handles @con and @sup. The fwnode link * represents the detail that the firmware lists @sup fwnode as supplying a @@ -70,34 +65,69 @@ static bool fw_devlink_drv_reg_done; * Attempts to create duplicate links between the same pair of fwnode handles * are ignored and there is no reference counting. */ -int fwnode_link_add(struct fwnode_handle *con, struct fwnode_handle *sup) +static int __fwnode_link_add(struct fwnode_handle *con, + struct fwnode_handle *sup, u8 flags) { struct fwnode_link *link; - int ret = 0; - - mutex_lock(&fwnode_link_lock); list_for_each_entry(link, &sup->consumers, s_hook) - if (link->consumer == con) - goto out; + if (link->consumer == con) { + link->flags |= flags; + return 0; + } link = kzalloc(sizeof(*link), GFP_KERNEL); - if (!link) { - ret = -ENOMEM; - goto out; - } + if (!link) + return -ENOMEM; link->supplier = sup; INIT_LIST_HEAD(&link->s_hook); link->consumer = con; INIT_LIST_HEAD(&link->c_hook); + link->flags = flags; list_add(&link->s_hook, &sup->consumers); list_add(&link->c_hook, &con->suppliers); -out: - mutex_unlock(&fwnode_link_lock); + pr_debug("%pfwf Linked as a fwnode consumer to %pfwf\n", + con, sup); - return ret; + return 0; +} + +int fwnode_link_add(struct fwnode_handle *con, struct fwnode_handle *sup, + u8 flags) +{ + guard(mutex)(&fwnode_link_lock); + + return __fwnode_link_add(con, sup, flags); +} + +/** + * __fwnode_link_del - Delete a link between two fwnode_handles. + * @link: the fwnode_link to be deleted + * + * The fwnode_link_lock needs to be held when this function is called. + */ +static void __fwnode_link_del(struct fwnode_link *link) +{ + pr_debug("%pfwf Dropping the fwnode link to %pfwf\n", + link->consumer, link->supplier); + list_del(&link->s_hook); + list_del(&link->c_hook); + kfree(link); +} + +/** + * __fwnode_link_cycle - Mark a fwnode link as being part of a cycle. + * @link: the fwnode_link to be marked + * + * The fwnode_link_lock needs to be held when this function is called. + */ +static void __fwnode_link_cycle(struct fwnode_link *link) +{ + pr_debug("%pfwf: cycle: depends on %pfwf\n", + link->consumer, link->supplier); + link->flags |= FWLINK_FLAG_CYCLE; } /** @@ -110,13 +140,10 @@ static void fwnode_links_purge_suppliers(struct fwnode_handle *fwnode) { struct fwnode_link *link, *tmp; - mutex_lock(&fwnode_link_lock); - list_for_each_entry_safe(link, tmp, &fwnode->suppliers, c_hook) { - list_del(&link->s_hook); - list_del(&link->c_hook); - kfree(link); - } - mutex_unlock(&fwnode_link_lock); + guard(mutex)(&fwnode_link_lock); + + list_for_each_entry_safe(link, tmp, &fwnode->suppliers, c_hook) + __fwnode_link_del(link); } /** @@ -129,13 +156,10 @@ static void fwnode_links_purge_consumers(struct fwnode_handle *fwnode) { struct fwnode_link *link, *tmp; - mutex_lock(&fwnode_link_lock); - list_for_each_entry_safe(link, tmp, &fwnode->consumers, s_hook) { - list_del(&link->s_hook); - list_del(&link->c_hook); - kfree(link); - } - mutex_unlock(&fwnode_link_lock); + guard(mutex)(&fwnode_link_lock); + + list_for_each_entry_safe(link, tmp, &fwnode->consumers, s_hook) + __fwnode_link_del(link); } /** @@ -166,7 +190,51 @@ void fw_devlink_purge_absent_suppliers(struct fwnode_handle *fwnode) } EXPORT_SYMBOL_GPL(fw_devlink_purge_absent_suppliers); -#ifdef CONFIG_SRCU +/** + * __fwnode_links_move_consumers - Move consumer from @from to @to fwnode_handle + * @from: move consumers away from this fwnode + * @to: move consumers to this fwnode + * + * Move all consumer links from @from fwnode to @to fwnode. + */ +static void __fwnode_links_move_consumers(struct fwnode_handle *from, + struct fwnode_handle *to) +{ + struct fwnode_link *link, *tmp; + + list_for_each_entry_safe(link, tmp, &from->consumers, s_hook) { + __fwnode_link_add(link->consumer, to, link->flags); + __fwnode_link_del(link); + } +} + +/** + * __fw_devlink_pickup_dangling_consumers - Pick up dangling consumers + * @fwnode: fwnode from which to pick up dangling consumers + * @new_sup: fwnode of new supplier + * + * If the @fwnode has a corresponding struct device and the device supports + * probing (that is, added to a bus), then we want to let fw_devlink create + * MANAGED device links to this device, so leave @fwnode and its descendant's + * fwnode links alone. + * + * Otherwise, move its consumers to the new supplier @new_sup. + */ +static void __fw_devlink_pickup_dangling_consumers(struct fwnode_handle *fwnode, + struct fwnode_handle *new_sup) +{ + struct fwnode_handle *child; + + if (fwnode->dev && fwnode->dev->bus) + return; + + fwnode->flags |= FWNODE_FLAG_NOT_DEVICE; + __fwnode_links_move_consumers(fwnode, new_sup); + + fwnode_for_each_available_child_node(fwnode, child) + __fw_devlink_pickup_dangling_consumers(child, new_sup); +} + static DEFINE_MUTEX(device_links_lock); DEFINE_STATIC_SRCU(device_links_srcu); @@ -205,47 +273,6 @@ static void device_link_remove_from_lists(struct device_link *link) list_del_rcu(&link->s_node); list_del_rcu(&link->c_node); } -#else /* !CONFIG_SRCU */ -static DECLARE_RWSEM(device_links_lock); - -static inline void device_links_write_lock(void) -{ - down_write(&device_links_lock); -} - -static inline void device_links_write_unlock(void) -{ - up_write(&device_links_lock); -} - -int device_links_read_lock(void) -{ - down_read(&device_links_lock); - return 0; -} - -void device_links_read_unlock(int not_used) -{ - up_read(&device_links_lock); -} - -#ifdef CONFIG_DEBUG_LOCK_ALLOC -int device_links_read_lock_held(void) -{ - return lockdep_is_held(&device_links_lock); -} -#endif - -static inline void device_link_synchronize_removal(void) -{ -} - -static void device_link_remove_from_lists(struct device_link *link) -{ - list_del(&link->s_node); - list_del(&link->c_node); -} -#endif /* !CONFIG_SRCU */ static bool device_is_ancestor(struct device *dev, struct device *target) { @@ -257,6 +284,14 @@ static bool device_is_ancestor(struct device *dev, struct device *target) return false; } +#define DL_MARKER_FLAGS (DL_FLAG_INFERRED | \ + DL_FLAG_CYCLE | \ + DL_FLAG_MANAGED) +bool device_link_flag_is_sync_state_only(u32 flags) +{ + return (flags & ~DL_MARKER_FLAGS) == DL_FLAG_SYNC_STATE_ONLY; +} + /** * device_is_dependent - Check if one device depends on another one * @dev: Device to check dependencies for. @@ -265,7 +300,7 @@ static bool device_is_ancestor(struct device *dev, struct device *target) * Check if @target depends on @dev or any device dependent on it (its child or * its consumer etc). Return 1 if that is the case or 0 otherwise. */ -int device_is_dependent(struct device *dev, void *target) +static int device_is_dependent(struct device *dev, void *target) { struct device_link *link; int ret; @@ -283,8 +318,7 @@ int device_is_dependent(struct device *dev, void *target) return ret; list_for_each_entry(link, &dev->links.consumers, s_node) { - if ((link->flags & ~DL_FLAG_INFERRED) == - (DL_FLAG_SYNC_STATE_ONLY | DL_FLAG_MANAGED)) + if (device_link_flag_is_sync_state_only(link->flags)) continue; if (link->consumer == target) @@ -357,8 +391,7 @@ static int device_reorder_to_tail(struct device *dev, void *not_used) device_for_each_child(dev, NULL, device_reorder_to_tail); list_for_each_entry(link, &dev->links.consumers, s_node) { - if ((link->flags & ~DL_FLAG_INFERRED) == - (DL_FLAG_SYNC_STATE_ONLY | DL_FLAG_MANAGED)) + if (device_link_flag_is_sync_state_only(link->flags)) continue; device_reorder_to_tail(link->consumer, NULL); } @@ -427,9 +460,9 @@ static ssize_t auto_remove_on_show(struct device *dev, struct device_link *link = to_devlink(dev); const char *output; - if (link->flags & DL_FLAG_AUTOREMOVE_SUPPLIER) + if (device_link_test(link, DL_FLAG_AUTOREMOVE_SUPPLIER)) output = "supplier unbind"; - else if (link->flags & DL_FLAG_AUTOREMOVE_CONSUMER) + else if (device_link_test(link, DL_FLAG_AUTOREMOVE_CONSUMER)) output = "consumer unbind"; else output = "never"; @@ -443,7 +476,7 @@ static ssize_t runtime_pm_show(struct device *dev, { struct device_link *link = to_devlink(dev); - return sysfs_emit(buf, "%d\n", !!(link->flags & DL_FLAG_PM_RUNTIME)); + return sysfs_emit(buf, "%d\n", device_link_test(link, DL_FLAG_PM_RUNTIME)); } static DEVICE_ATTR_RO(runtime_pm); @@ -452,8 +485,7 @@ static ssize_t sync_state_only_show(struct device *dev, { struct device_link *link = to_devlink(dev); - return sysfs_emit(buf, "%d\n", - !!(link->flags & DL_FLAG_SYNC_STATE_ONLY)); + return sysfs_emit(buf, "%d\n", device_link_test(link, DL_FLAG_SYNC_STATE_ONLY)); } static DEVICE_ATTR_RO(sync_state_only); @@ -473,8 +505,18 @@ static void device_link_release_fn(struct work_struct *work) /* Ensure that all references to the link object have been dropped. */ device_link_synchronize_removal(); - while (refcount_dec_not_one(&link->rpm_active)) - pm_runtime_put(link->supplier); + pm_runtime_release_supplier(link); + /* + * If supplier_preactivated is set, the link has been dropped between + * the pm_runtime_get_suppliers() and pm_runtime_put_suppliers() calls + * in __driver_probe_device(). In that case, drop the supplier's + * PM-runtime usage counter to remove the reference taken by + * pm_runtime_get_suppliers(). + */ + if (link->supplier_preactivated) + pm_runtime_put_noidle(link->supplier); + + pm_request_idle(link->supplier); put_device(link->consumer); put_device(link->supplier); @@ -489,36 +531,39 @@ static void devlink_dev_release(struct device *dev) /* * It may take a while to complete this work because of the SRCU * synchronization in device_link_release_fn() and if the consumer or - * supplier devices get deleted when it runs, so put it into the "long" - * workqueue. + * supplier devices get deleted when it runs, so put it into the + * dedicated workqueue. */ - queue_work(system_long_wq, &link->rm_work); + queue_work(device_link_wq, &link->rm_work); } -static struct class devlink_class = { +/** + * device_link_wait_removal - Wait for ongoing devlink removal jobs to terminate + */ +void device_link_wait_removal(void) +{ + /* + * devlink removal jobs are queued in the dedicated work queue. + * To be sure that all removal jobs are terminated, ensure that any + * scheduled work has run to completion. + */ + flush_workqueue(device_link_wq); +} +EXPORT_SYMBOL_GPL(device_link_wait_removal); + +static const struct class devlink_class = { .name = "devlink", - .owner = THIS_MODULE, .dev_groups = devlink_groups, .dev_release = devlink_dev_release, }; -static int devlink_add_symlinks(struct device *dev, - struct class_interface *class_intf) +static int devlink_add_symlinks(struct device *dev) { + char *buf_con __free(kfree) = NULL, *buf_sup __free(kfree) = NULL; int ret; - size_t len; struct device_link *link = to_devlink(dev); struct device *sup = link->supplier; struct device *con = link->consumer; - char *buf; - - len = max(strlen(dev_bus_name(sup)) + strlen(dev_name(sup)), - strlen(dev_bus_name(con)) + strlen(dev_name(con))); - len += strlen(":"); - len += strlen("supplier:") + 1; - buf = kzalloc(len, GFP_KERNEL); - if (!buf) - return -ENOMEM; ret = sysfs_create_link(&link->link_dev.kobj, &sup->kobj, "supplier"); if (ret) @@ -528,57 +573,64 @@ static int devlink_add_symlinks(struct device *dev, if (ret) goto err_con; - snprintf(buf, len, "consumer:%s:%s", dev_bus_name(con), dev_name(con)); - ret = sysfs_create_link(&sup->kobj, &link->link_dev.kobj, buf); + buf_con = kasprintf(GFP_KERNEL, "consumer:%s:%s", dev_bus_name(con), dev_name(con)); + if (!buf_con) { + ret = -ENOMEM; + goto err_con_dev; + } + + ret = sysfs_create_link(&sup->kobj, &link->link_dev.kobj, buf_con); if (ret) goto err_con_dev; - snprintf(buf, len, "supplier:%s:%s", dev_bus_name(sup), dev_name(sup)); - ret = sysfs_create_link(&con->kobj, &link->link_dev.kobj, buf); + buf_sup = kasprintf(GFP_KERNEL, "supplier:%s:%s", dev_bus_name(sup), dev_name(sup)); + if (!buf_sup) { + ret = -ENOMEM; + goto err_sup_dev; + } + + ret = sysfs_create_link(&con->kobj, &link->link_dev.kobj, buf_sup); if (ret) goto err_sup_dev; goto out; err_sup_dev: - snprintf(buf, len, "consumer:%s:%s", dev_bus_name(con), dev_name(con)); - sysfs_remove_link(&sup->kobj, buf); + sysfs_remove_link(&sup->kobj, buf_con); err_con_dev: sysfs_remove_link(&link->link_dev.kobj, "consumer"); err_con: sysfs_remove_link(&link->link_dev.kobj, "supplier"); out: - kfree(buf); return ret; } -static void devlink_remove_symlinks(struct device *dev, - struct class_interface *class_intf) +static void devlink_remove_symlinks(struct device *dev) { + char *buf_con __free(kfree) = NULL, *buf_sup __free(kfree) = NULL; struct device_link *link = to_devlink(dev); - size_t len; struct device *sup = link->supplier; struct device *con = link->consumer; - char *buf; sysfs_remove_link(&link->link_dev.kobj, "consumer"); sysfs_remove_link(&link->link_dev.kobj, "supplier"); - len = max(strlen(dev_bus_name(sup)) + strlen(dev_name(sup)), - strlen(dev_bus_name(con)) + strlen(dev_name(con))); - len += strlen(":"); - len += strlen("supplier:") + 1; - buf = kzalloc(len, GFP_KERNEL); - if (!buf) { - WARN(1, "Unable to properly free device link symlinks!\n"); - return; + if (device_is_registered(con)) { + buf_sup = kasprintf(GFP_KERNEL, "supplier:%s:%s", dev_bus_name(sup), dev_name(sup)); + if (!buf_sup) + goto out; + sysfs_remove_link(&con->kobj, buf_sup); } - snprintf(buf, len, "supplier:%s:%s", dev_bus_name(sup), dev_name(sup)); - sysfs_remove_link(&con->kobj, buf); - snprintf(buf, len, "consumer:%s:%s", dev_bus_name(con), dev_name(con)); - sysfs_remove_link(&sup->kobj, buf); - kfree(buf); + buf_con = kasprintf(GFP_KERNEL, "consumer:%s:%s", dev_bus_name(con), dev_name(con)); + if (!buf_con) + goto out; + sysfs_remove_link(&sup->kobj, buf_con); + + return; + +out: + WARN(1, "Unable to properly free device link symlinks!\n"); } static struct class_interface devlink_class_intf = { @@ -607,7 +659,8 @@ postcore_initcall(devlink_class_init); DL_FLAG_AUTOREMOVE_SUPPLIER | \ DL_FLAG_AUTOPROBE_CONSUMER | \ DL_FLAG_SYNC_STATE_ONLY | \ - DL_FLAG_INFERRED) + DL_FLAG_INFERRED | \ + DL_FLAG_CYCLE) #define DL_ADD_VALID_FLAGS (DL_MANAGED_LINK_FLAGS | DL_FLAG_STATELESS | \ DL_FLAG_PM_RUNTIME | DL_FLAG_RPM_ACTIVE) @@ -618,6 +671,9 @@ postcore_initcall(devlink_class_init); * @supplier: Supplier end of the link. * @flags: Link flags. * + * Return: On success, a device_link struct will be returned. + * On error or invalid flag settings, NULL will be returned. + * * The caller is responsible for the proper synchronization of the link creation * with runtime PM. First, setting the DL_FLAG_PM_RUNTIME flag will cause the * runtime PM framework to take the link into account. Second, if the @@ -673,10 +729,9 @@ struct device_link *device_link_add(struct device *consumer, { struct device_link *link; - if (!consumer || !supplier || flags & ~DL_ADD_VALID_FLAGS || + if (!consumer || !supplier || consumer == supplier || + flags & ~DL_ADD_VALID_FLAGS || (flags & DL_FLAG_STATELESS && flags & DL_MANAGED_LINK_FLAGS) || - (flags & DL_FLAG_SYNC_STATE_ONLY && - (flags & ~DL_FLAG_INFERRED) != DL_FLAG_SYNC_STATE_ONLY) || (flags & DL_FLAG_AUTOPROBE_CONSUMER && flags & (DL_FLAG_AUTOREMOVE_CONSUMER | DL_FLAG_AUTOREMOVE_SUPPLIER))) @@ -692,6 +747,10 @@ struct device_link *device_link_add(struct device *consumer, if (!(flags & DL_FLAG_STATELESS)) flags |= DL_FLAG_MANAGED; + if (flags & DL_FLAG_SYNC_STATE_ONLY && + !device_link_flag_is_sync_state_only(flags)) + return NULL; + device_links_write_lock(); device_pm_lock(); @@ -732,12 +791,12 @@ struct device_link *device_link_add(struct device *consumer, if (link->consumer != consumer) continue; - if (link->flags & DL_FLAG_INFERRED && + if (device_link_test(link, DL_FLAG_INFERRED) && !(flags & DL_FLAG_INFERRED)) link->flags &= ~DL_FLAG_INFERRED; if (flags & DL_FLAG_PM_RUNTIME) { - if (!(link->flags & DL_FLAG_PM_RUNTIME)) { + if (!device_link_test(link, DL_FLAG_PM_RUNTIME)) { pm_runtime_new_link(consumer); link->flags |= DL_FLAG_PM_RUNTIME; } @@ -747,8 +806,8 @@ struct device_link *device_link_add(struct device *consumer, if (flags & DL_FLAG_STATELESS) { kref_get(&link->kref); - if (link->flags & DL_FLAG_SYNC_STATE_ONLY && - !(link->flags & DL_FLAG_STATELESS)) { + if (device_link_test(link, DL_FLAG_SYNC_STATE_ONLY) && + !device_link_test(link, DL_FLAG_STATELESS)) { link->flags |= DL_FLAG_STATELESS; goto reorder; } else { @@ -763,7 +822,7 @@ struct device_link *device_link_add(struct device *consumer, * update the existing link to stay around longer. */ if (flags & DL_FLAG_AUTOREMOVE_SUPPLIER) { - if (link->flags & DL_FLAG_AUTOREMOVE_CONSUMER) { + if (device_link_test(link, DL_FLAG_AUTOREMOVE_CONSUMER)) { link->flags &= ~DL_FLAG_AUTOREMOVE_CONSUMER; link->flags |= DL_FLAG_AUTOREMOVE_SUPPLIER; } @@ -771,12 +830,12 @@ struct device_link *device_link_add(struct device *consumer, link->flags &= ~(DL_FLAG_AUTOREMOVE_CONSUMER | DL_FLAG_AUTOREMOVE_SUPPLIER); } - if (!(link->flags & DL_FLAG_MANAGED)) { + if (!device_link_test(link, DL_FLAG_MANAGED)) { kref_get(&link->kref); link->flags |= DL_FLAG_MANAGED; device_link_init_status(link, consumer, supplier); } - if (link->flags & DL_FLAG_SYNC_STATE_ONLY && + if (device_link_test(link, DL_FLAG_SYNC_STATE_ONLY) && !(flags & DL_FLAG_SYNC_STATE_ONLY)) { link->flags &= ~DL_FLAG_SYNC_STATE_ONLY; goto reorder; @@ -806,9 +865,7 @@ struct device_link *device_link_add(struct device *consumer, dev_bus_name(supplier), dev_name(supplier), dev_bus_name(consumer), dev_name(consumer)); if (device_register(&link->link_dev)) { - put_device(consumer); - put_device(supplier); - kfree(link); + put_device(&link->link_dev); link = NULL; goto out; } @@ -882,8 +939,10 @@ static void __device_link_del(struct kref *kref) static void device_link_put_kref(struct device_link *link) { - if (link->flags & DL_FLAG_STATELESS) + if (device_link_test(link, DL_FLAG_STATELESS)) kref_put(&link->kref, __device_link_del); + else if (!device_is_registered(link->consumer)) + __device_link_del(&link->kref); else WARN(1, "Unable to drop a managed device link reference\n"); } @@ -944,12 +1003,34 @@ static void device_links_missing_supplier(struct device *dev) if (link->supplier->links.status == DL_DEV_DRIVER_BOUND) { WRITE_ONCE(link->status, DL_STATE_AVAILABLE); } else { - WARN_ON(!(link->flags & DL_FLAG_SYNC_STATE_ONLY)); + WARN_ON(!device_link_test(link, DL_FLAG_SYNC_STATE_ONLY)); WRITE_ONCE(link->status, DL_STATE_DORMANT); } } } +static bool dev_is_best_effort(struct device *dev) +{ + return (fw_devlink_best_effort && dev->can_match) || + (dev->fwnode && (dev->fwnode->flags & FWNODE_FLAG_BEST_EFFORT)); +} + +static struct fwnode_handle *fwnode_links_check_suppliers( + struct fwnode_handle *fwnode) +{ + struct fwnode_link *link; + + if (!fwnode || fw_devlink_is_permissive()) + return NULL; + + list_for_each_entry(link, &fwnode->suppliers, c_hook) + if (!(link->flags & + (FWLINK_FLAG_CYCLE | FWLINK_FLAG_IGNORE))) + return link->supplier; + + return NULL; +} + /** * device_links_check_suppliers - Check presence of supplier drivers. * @dev: Consumer device. @@ -969,36 +1050,43 @@ static void device_links_missing_supplier(struct device *dev) int device_links_check_suppliers(struct device *dev) { struct device_link *link; - int ret = 0; + int ret = 0, fwnode_ret = 0; + struct fwnode_handle *sup_fw; /* * Device waiting for supplier to become available is not allowed to * probe. */ - mutex_lock(&fwnode_link_lock); - if (dev->fwnode && !list_empty(&dev->fwnode->suppliers) && - !fw_devlink_is_permissive()) { - dev_dbg(dev, "probe deferral - wait for supplier %pfwP\n", - list_first_entry(&dev->fwnode->suppliers, - struct fwnode_link, - c_hook)->supplier); - mutex_unlock(&fwnode_link_lock); - return -EPROBE_DEFER; + scoped_guard(mutex, &fwnode_link_lock) { + sup_fw = fwnode_links_check_suppliers(dev->fwnode); + if (sup_fw) { + if (dev_is_best_effort(dev)) + fwnode_ret = -EAGAIN; + else + return dev_err_probe(dev, -EPROBE_DEFER, + "wait for supplier %pfwf\n", sup_fw); + } } - mutex_unlock(&fwnode_link_lock); device_links_write_lock(); list_for_each_entry(link, &dev->links.suppliers, c_node) { - if (!(link->flags & DL_FLAG_MANAGED)) + if (!device_link_test(link, DL_FLAG_MANAGED)) continue; if (link->status != DL_STATE_AVAILABLE && - !(link->flags & DL_FLAG_SYNC_STATE_ONLY)) { + !device_link_test(link, DL_FLAG_SYNC_STATE_ONLY)) { + + if (dev_is_best_effort(dev) && + device_link_test(link, DL_FLAG_INFERRED) && + !link->supplier->can_match) { + ret = -EAGAIN; + continue; + } + device_links_missing_supplier(dev); - dev_dbg(dev, "probe deferral - supplier %s not ready\n", - dev_name(link->supplier)); - ret = -EPROBE_DEFER; + ret = dev_err_probe(dev, -EPROBE_DEFER, + "supplier %s not ready\n", dev_name(link->supplier)); break; } WRITE_ONCE(link->status, DL_STATE_CONSUMER_PROBE); @@ -1006,7 +1094,8 @@ int device_links_check_suppliers(struct device *dev) dev->links.status = DL_DEV_PROBING; device_links_write_unlock(); - return ret; + + return ret ? ret : fwnode_ret; } /** @@ -1038,7 +1127,7 @@ static void __device_links_queue_sync_state(struct device *dev, return; list_for_each_entry(link, &dev->links.consumers, s_node) { - if (!(link->flags & DL_FLAG_MANAGED)) + if (!device_link_test(link, DL_FLAG_MANAGED)) continue; if (link->status != DL_STATE_ACTIVE) return; @@ -1079,10 +1168,7 @@ static void device_links_flush_sync_list(struct list_head *list, if (dev != dont_lock_dev) device_lock(dev); - if (dev->bus->sync_state) - dev->bus->sync_state(dev); - else if (dev->driver && dev->driver->sync_state) - dev->driver->sync_state(dev); + dev_sync_state(dev); if (dev != dont_lock_dev) device_unlock(dev); @@ -1153,7 +1239,8 @@ static ssize_t waiting_for_supplier_show(struct device *dev, bool val; device_lock(dev); - val = !list_empty(&dev->fwnode->suppliers); + scoped_guard(mutex, &fwnode_link_lock) + val = !!fwnode_links_check_suppliers(dev->fwnode); device_unlock(dev); return sysfs_emit(buf, "%u\n", val); } @@ -1180,7 +1267,7 @@ void device_links_force_bind(struct device *dev) device_links_write_lock(); list_for_each_entry_safe(link, ln, &dev->links.suppliers, c_node) { - if (!(link->flags & DL_FLAG_MANAGED)) + if (!device_link_test(link, DL_FLAG_MANAGED)) continue; if (link->status != DL_STATE_AVAILABLE) { @@ -1216,23 +1303,32 @@ void device_links_driver_bound(struct device *dev) * them. So, fw_devlink no longer needs to create device links to any * of the device's suppliers. * - * Also, if a child firmware node of this bound device is not added as - * a device by now, assume it is never going to be added and make sure - * other devices don't defer probe indefinitely by waiting for such a - * child device. + * Also, if a child firmware node of this bound device is not added as a + * device by now, assume it is never going to be added. Make this bound + * device the fallback supplier to the dangling consumers of the child + * firmware node because this bound device is probably implementing the + * child firmware node functionality and we don't want the dangling + * consumers to defer probe indefinitely waiting for a device for the + * child firmware node. */ if (dev->fwnode && dev->fwnode->dev == dev) { struct fwnode_handle *child; + fwnode_links_purge_suppliers(dev->fwnode); + + guard(mutex)(&fwnode_link_lock); + fwnode_for_each_available_child_node(dev->fwnode, child) - fw_devlink_purge_absent_suppliers(child); + __fw_devlink_pickup_dangling_consumers(child, + dev->fwnode); + __fw_devlink_link_to_consumers(dev); } device_remove_file(dev, &dev_attr_waiting_for_supplier); device_links_write_lock(); list_for_each_entry(link, &dev->links.consumers, s_node) { - if (!(link->flags & DL_FLAG_MANAGED)) + if (!device_link_test(link, DL_FLAG_MANAGED)) continue; /* @@ -1248,7 +1344,7 @@ void device_links_driver_bound(struct device *dev) WARN_ON(link->status != DL_STATE_DORMANT); WRITE_ONCE(link->status, DL_STATE_AVAILABLE); - if (link->flags & DL_FLAG_AUTOPROBE_CONSUMER) + if (device_link_test(link, DL_FLAG_AUTOPROBE_CONSUMER)) driver_deferred_probe_add(link->consumer); } @@ -1260,17 +1356,29 @@ void device_links_driver_bound(struct device *dev) list_for_each_entry_safe(link, ln, &dev->links.suppliers, c_node) { struct device *supplier; - if (!(link->flags & DL_FLAG_MANAGED)) + if (!device_link_test(link, DL_FLAG_MANAGED)) continue; supplier = link->supplier; - if (link->flags & DL_FLAG_SYNC_STATE_ONLY) { + if (device_link_test(link, DL_FLAG_SYNC_STATE_ONLY)) { /* * When DL_FLAG_SYNC_STATE_ONLY is set, it means no * other DL_MANAGED_LINK_FLAGS have been set. So, it's * save to drop the managed link completely. */ device_link_drop_managed(link); + } else if (dev_is_best_effort(dev) && + device_link_test(link, DL_FLAG_INFERRED) && + link->status != DL_STATE_CONSUMER_PROBE && + !link->supplier->can_match) { + /* + * When dev_is_best_effort() is true, we ignore device + * links to suppliers that don't have a driver. If the + * consumer device still managed to probe, there's no + * point in maintaining a device link in a weird state + * (consumer probed before supplier). So delete it. + */ + device_link_drop_managed(link); } else { WARN_ON(link->status != DL_STATE_CONSUMER_PROBE); WRITE_ONCE(link->status, DL_STATE_ACTIVE); @@ -1312,10 +1420,10 @@ static void __device_links_no_driver(struct device *dev) struct device_link *link, *ln; list_for_each_entry_safe_reverse(link, ln, &dev->links.suppliers, c_node) { - if (!(link->flags & DL_FLAG_MANAGED)) + if (!device_link_test(link, DL_FLAG_MANAGED)) continue; - if (link->flags & DL_FLAG_AUTOREMOVE_CONSUMER) { + if (device_link_test(link, DL_FLAG_AUTOREMOVE_CONSUMER)) { device_link_drop_managed(link); continue; } @@ -1327,7 +1435,7 @@ static void __device_links_no_driver(struct device *dev) if (link->supplier->links.status == DL_DEV_DRIVER_BOUND) { WRITE_ONCE(link->status, DL_STATE_AVAILABLE); } else { - WARN_ON(!(link->flags & DL_FLAG_SYNC_STATE_ONLY)); + WARN_ON(!device_link_test(link, DL_FLAG_SYNC_STATE_ONLY)); WRITE_ONCE(link->status, DL_STATE_DORMANT); } } @@ -1352,7 +1460,7 @@ void device_links_no_driver(struct device *dev) device_links_write_lock(); list_for_each_entry(link, &dev->links.consumers, s_node) { - if (!(link->flags & DL_FLAG_MANAGED)) + if (!device_link_test(link, DL_FLAG_MANAGED)) continue; /* @@ -1389,10 +1497,10 @@ void device_links_driver_cleanup(struct device *dev) device_links_write_lock(); list_for_each_entry_safe(link, ln, &dev->links.consumers, s_node) { - if (!(link->flags & DL_FLAG_MANAGED)) + if (!device_link_test(link, DL_FLAG_MANAGED)) continue; - WARN_ON(link->flags & DL_FLAG_AUTOREMOVE_CONSUMER); + WARN_ON(device_link_test(link, DL_FLAG_AUTOREMOVE_CONSUMER)); WARN_ON(link->status != DL_STATE_SUPPLIER_UNBIND); /* @@ -1401,7 +1509,7 @@ void device_links_driver_cleanup(struct device *dev) * has moved to DL_STATE_SUPPLIER_UNBIND. */ if (link->status == DL_STATE_SUPPLIER_UNBIND && - link->flags & DL_FLAG_AUTOREMOVE_SUPPLIER) + device_link_test(link, DL_FLAG_AUTOREMOVE_SUPPLIER)) device_link_drop_managed(link); WRITE_ONCE(link->status, DL_STATE_DORMANT); @@ -1435,7 +1543,7 @@ bool device_links_busy(struct device *dev) device_links_write_lock(); list_for_each_entry(link, &dev->links.consumers, s_node) { - if (!(link->flags & DL_FLAG_MANAGED)) + if (!device_link_test(link, DL_FLAG_MANAGED)) continue; if (link->status == DL_STATE_CONSUMER_PROBE @@ -1477,8 +1585,8 @@ void device_links_unbind_consumers(struct device *dev) list_for_each_entry(link, &dev->links.consumers, s_node) { enum device_link_state status; - if (!(link->flags & DL_FLAG_MANAGED) || - link->flags & DL_FLAG_SYNC_STATE_ONLY) + if (!device_link_test(link, DL_FLAG_MANAGED) || + device_link_test(link, DL_FLAG_SYNC_STATE_ONLY)) continue; status = link->status; @@ -1544,7 +1652,7 @@ static void device_links_purge(struct device *dev) #define FW_DEVLINK_FLAGS_RPM (FW_DEVLINK_FLAGS_ON | \ DL_FLAG_PM_RUNTIME) -static u32 fw_devlink_flags = FW_DEVLINK_FLAGS_ON; +static u32 fw_devlink_flags = FW_DEVLINK_FLAGS_RPM; static int __init fw_devlink_setup(char *arg) { if (!arg) @@ -1566,12 +1674,40 @@ early_param("fw_devlink", fw_devlink_setup); static bool fw_devlink_strict; static int __init fw_devlink_strict_setup(char *arg) { - return strtobool(arg, &fw_devlink_strict); + return kstrtobool(arg, &fw_devlink_strict); } early_param("fw_devlink.strict", fw_devlink_strict_setup); -u32 fw_devlink_get_flags(void) +#define FW_DEVLINK_SYNC_STATE_STRICT 0 +#define FW_DEVLINK_SYNC_STATE_TIMEOUT 1 + +#ifndef CONFIG_FW_DEVLINK_SYNC_STATE_TIMEOUT +static int fw_devlink_sync_state; +#else +static int fw_devlink_sync_state = FW_DEVLINK_SYNC_STATE_TIMEOUT; +#endif + +static int __init fw_devlink_sync_state_setup(char *arg) +{ + if (!arg) + return -EINVAL; + + if (strcmp(arg, "strict") == 0) { + fw_devlink_sync_state = FW_DEVLINK_SYNC_STATE_STRICT; + return 0; + } else if (strcmp(arg, "timeout") == 0) { + fw_devlink_sync_state = FW_DEVLINK_SYNC_STATE_TIMEOUT; + return 0; + } + return -EINVAL; +} +early_param("fw_devlink.sync_state", fw_devlink_sync_state_setup); + +static inline u32 fw_devlink_get_flags(u8 fwlink_flags) { + if (fwlink_flags & FWLINK_FLAG_CYCLE) + return FW_DEVLINK_FLAGS_PERMISSIVE | DL_FLAG_CYCLE; + return fw_devlink_flags; } @@ -1606,10 +1742,10 @@ static void fw_devlink_parse_fwtree(struct fwnode_handle *fwnode) static void fw_devlink_relax_link(struct device_link *link) { - if (!(link->flags & DL_FLAG_INFERRED)) + if (!device_link_test(link, DL_FLAG_INFERRED)) return; - if (link->flags == (DL_FLAG_MANAGED | FW_DEVLINK_FLAGS_PERMISSIVE)) + if (device_link_flag_is_sync_state_only(link->flags)) return; pm_runtime_drop_link(link); @@ -1637,6 +1773,100 @@ void fw_devlink_drivers_done(void) device_links_write_unlock(); } +static int fw_devlink_dev_sync_state(struct device *dev, void *data) +{ + struct device_link *link = to_devlink(dev); + struct device *sup = link->supplier; + + if (!device_link_test(link, DL_FLAG_MANAGED) || + link->status == DL_STATE_ACTIVE || sup->state_synced || + !dev_has_sync_state(sup)) + return 0; + + if (fw_devlink_sync_state == FW_DEVLINK_SYNC_STATE_STRICT) { + dev_info(sup, "sync_state() pending due to %s\n", + dev_name(link->consumer)); + return 0; + } + + if (!list_empty(&sup->links.defer_sync)) + return 0; + + dev_warn(sup, "Timed out. Forcing sync_state()\n"); + sup->state_synced = true; + get_device(sup); + list_add_tail(&sup->links.defer_sync, data); + + return 0; +} + +void fw_devlink_probing_done(void) +{ + LIST_HEAD(sync_list); + + device_links_write_lock(); + class_for_each_device(&devlink_class, NULL, &sync_list, + fw_devlink_dev_sync_state); + device_links_write_unlock(); + device_links_flush_sync_list(&sync_list, NULL); +} + +/** + * wait_for_init_devices_probe - Try to probe any device needed for init + * + * Some devices might need to be probed and bound successfully before the kernel + * boot sequence can finish and move on to init/userspace. For example, a + * network interface might need to be bound to be able to mount a NFS rootfs. + * + * With fw_devlink=on by default, some of these devices might be blocked from + * probing because they are waiting on a optional supplier that doesn't have a + * driver. While fw_devlink will eventually identify such devices and unblock + * the probing automatically, it might be too late by the time it unblocks the + * probing of devices. For example, the IP4 autoconfig might timeout before + * fw_devlink unblocks probing of the network interface. + * + * This function is available to temporarily try and probe all devices that have + * a driver even if some of their suppliers haven't been added or don't have + * drivers. + * + * The drivers can then decide which of the suppliers are optional vs mandatory + * and probe the device if possible. By the time this function returns, all such + * "best effort" probes are guaranteed to be completed. If a device successfully + * probes in this mode, we delete all fw_devlink discovered dependencies of that + * device where the supplier hasn't yet probed successfully because they have to + * be optional dependencies. + * + * Any devices that didn't successfully probe go back to being treated as if + * this function was never called. + * + * This also means that some devices that aren't needed for init and could have + * waited for their optional supplier to probe (when the supplier's module is + * loaded later on) would end up probing prematurely with limited functionality. + * So call this function only when boot would fail without it. + */ +void __init wait_for_init_devices_probe(void) +{ + if (!fw_devlink_flags || fw_devlink_is_permissive()) + return; + + /* + * Wait for all ongoing probes to finish so that the "best effort" is + * only applied to devices that can't probe otherwise. + */ + wait_for_device_probe(); + + pr_info("Trying to probe devices needed for running init ...\n"); + fw_devlink_best_effort = true; + driver_deferred_probe_trigger(); + + /* + * Wait for all "best effort" probes to finish before going back to + * normal enforcement. + */ + wait_for_device_probe(); + fw_devlink_best_effort = false; +} + static void fw_devlink_unblock_consumers(struct device *dev) { struct device_link *link; @@ -1650,44 +1880,204 @@ static void fw_devlink_unblock_consumers(struct device *dev) device_links_write_unlock(); } +static bool fwnode_init_without_drv(struct fwnode_handle *fwnode) +{ + struct device *dev; + bool ret; + + if (!(fwnode->flags & FWNODE_FLAG_INITIALIZED)) + return false; + + dev = get_dev_from_fwnode(fwnode); + ret = !dev || dev->links.status == DL_DEV_NO_DRIVER; + put_device(dev); + + return ret; +} + +static bool fwnode_ancestor_init_without_drv(struct fwnode_handle *fwnode) +{ + struct fwnode_handle *parent; + + fwnode_for_each_parent_node(fwnode, parent) { + if (fwnode_init_without_drv(parent)) { + fwnode_handle_put(parent); + return true; + } + } + + return false; +} + /** - * fw_devlink_relax_cycle - Convert cyclic links to SYNC_STATE_ONLY links - * @con: Device to check dependencies for. - * @sup: Device to check against. - * - * Check if @sup depends on @con or any device dependent on it (its child or - * its consumer etc). When such a cyclic dependency is found, convert all - * device links created solely by fw_devlink into SYNC_STATE_ONLY device links. - * This is the equivalent of doing fw_devlink=permissive just between the - * devices in the cycle. We need to do this because, at this point, fw_devlink - * can't tell which of these dependencies is not a real dependency. - * - * Return 1 if a cycle is found. Otherwise, return 0. + * fwnode_is_ancestor_of - Test if @ancestor is ancestor of @child + * @ancestor: Firmware which is tested for being an ancestor + * @child: Firmware which is tested for being the child + * + * A node is considered an ancestor of itself too. + * + * Return: true if @ancestor is an ancestor of @child. Otherwise, returns false. */ -static int fw_devlink_relax_cycle(struct device *con, void *sup) +static bool fwnode_is_ancestor_of(const struct fwnode_handle *ancestor, + const struct fwnode_handle *child) { - struct device_link *link; - int ret; + struct fwnode_handle *parent; - if (con == sup) - return 1; + if (IS_ERR_OR_NULL(ancestor)) + return false; - ret = device_for_each_child(con, sup, fw_devlink_relax_cycle); - if (ret) - return ret; + if (child == ancestor) + return true; - list_for_each_entry(link, &con->links.consumers, s_node) { - if ((link->flags & ~DL_FLAG_INFERRED) == - (DL_FLAG_SYNC_STATE_ONLY | DL_FLAG_MANAGED)) - continue; + fwnode_for_each_parent_node(child, parent) { + if (parent == ancestor) { + fwnode_handle_put(parent); + return true; + } + } + return false; +} - if (!fw_devlink_relax_cycle(link->consumer, sup)) +/** + * fwnode_get_next_parent_dev - Find device of closest ancestor fwnode + * @fwnode: firmware node + * + * Given a firmware node (@fwnode), this function finds its closest ancestor + * firmware node that has a corresponding struct device and returns that struct + * device. + * + * The caller is responsible for calling put_device() on the returned device + * pointer. + * + * Return: a pointer to the device of the @fwnode's closest ancestor. + */ +static struct device *fwnode_get_next_parent_dev(const struct fwnode_handle *fwnode) +{ + struct fwnode_handle *parent; + struct device *dev; + + fwnode_for_each_parent_node(fwnode, parent) { + dev = get_dev_from_fwnode(parent); + if (dev) { + fwnode_handle_put(parent); + return dev; + } + } + return NULL; +} + +/** + * __fw_devlink_relax_cycles - Relax and mark dependency cycles. + * @con_handle: Potential consumer device fwnode. + * @sup_handle: Potential supplier's fwnode. + * + * Needs to be called with fwnode_lock and device link lock held. + * + * Check if @sup_handle or any of its ancestors or suppliers direct/indirectly + * depend on @con. This function can detect multiple cyles between @sup_handle + * and @con. When such dependency cycles are found, convert all device links + * created solely by fw_devlink into SYNC_STATE_ONLY device links. Also, mark + * all fwnode links in the cycle with FWLINK_FLAG_CYCLE so that when they are + * converted into a device link in the future, they are created as + * SYNC_STATE_ONLY device links. This is the equivalent of doing + * fw_devlink=permissive just between the devices in the cycle. We need to do + * this because, at this point, fw_devlink can't tell which of these + * dependencies is not a real dependency. + * + * Return true if one or more cycles were found. Otherwise, return false. + */ +static bool __fw_devlink_relax_cycles(struct fwnode_handle *con_handle, + struct fwnode_handle *sup_handle) +{ + struct device *sup_dev = NULL, *par_dev = NULL, *con_dev = NULL; + struct fwnode_link *link; + struct device_link *dev_link; + bool ret = false; + + if (!sup_handle) + return false; + + /* + * We aren't trying to find all cycles. Just a cycle between con and + * sup_handle. + */ + if (sup_handle->flags & FWNODE_FLAG_VISITED) + return false; + + sup_handle->flags |= FWNODE_FLAG_VISITED; + + /* Termination condition. */ + if (sup_handle == con_handle) { + pr_debug("----- cycle: start -----\n"); + ret = true; + goto out; + } + + sup_dev = get_dev_from_fwnode(sup_handle); + con_dev = get_dev_from_fwnode(con_handle); + /* + * If sup_dev is bound to a driver and @con hasn't started binding to a + * driver, sup_dev can't be a consumer of @con. So, no need to check + * further. + */ + if (sup_dev && sup_dev->links.status == DL_DEV_DRIVER_BOUND && + con_dev && con_dev->links.status == DL_DEV_NO_DRIVER) { + ret = false; + goto out; + } + + list_for_each_entry(link, &sup_handle->suppliers, c_hook) { + if (link->flags & FWLINK_FLAG_IGNORE) continue; - ret = 1; + if (__fw_devlink_relax_cycles(con_handle, link->supplier)) { + __fwnode_link_cycle(link); + ret = true; + } + } - fw_devlink_relax_link(link); + /* + * Give priority to device parent over fwnode parent to account for any + * quirks in how fwnodes are converted to devices. + */ + if (sup_dev) + par_dev = get_device(sup_dev->parent); + else + par_dev = fwnode_get_next_parent_dev(sup_handle); + + if (par_dev && __fw_devlink_relax_cycles(con_handle, par_dev->fwnode)) { + pr_debug("%pfwf: cycle: child of %pfwf\n", sup_handle, + par_dev->fwnode); + ret = true; } + + if (!sup_dev) + goto out; + + list_for_each_entry(dev_link, &sup_dev->links.suppliers, c_node) { + /* + * Ignore a SYNC_STATE_ONLY flag only if it wasn't marked as + * such due to a cycle. + */ + if (device_link_flag_is_sync_state_only(dev_link->flags) && + !device_link_test(dev_link, DL_FLAG_CYCLE)) + continue; + + if (__fw_devlink_relax_cycles(con_handle, + dev_link->supplier->fwnode)) { + pr_debug("%pfwf: cycle: depends on %pfwf\n", sup_handle, + dev_link->supplier->fwnode); + fw_devlink_relax_link(dev_link); + dev_link->flags |= DL_FLAG_CYCLE; + ret = true; + } + } + +out: + sup_handle->flags &= ~FWNODE_FLAG_VISITED; + put_device(sup_dev); + put_device(con_dev); + put_device(par_dev); return ret; } @@ -1695,7 +2085,7 @@ static int fw_devlink_relax_cycle(struct device *con, void *sup) * fw_devlink_create_devlink - Create a device link from a consumer to fwnode * @con: consumer device for the device link * @sup_handle: fwnode handle of supplier - * @flags: devlink flags + * @link: fwnode link that's being converted to a device link * * This function will try to create a device link between the consumer device * @con and the supplier device represented by @sup_handle. @@ -1712,12 +2102,59 @@ static int fw_devlink_relax_cycle(struct device *con, void *sup) * possible to do that in the future */ static int fw_devlink_create_devlink(struct device *con, - struct fwnode_handle *sup_handle, u32 flags) + struct fwnode_handle *sup_handle, + struct fwnode_link *link) { struct device *sup_dev; int ret = 0; + u32 flags; + + if (link->flags & FWLINK_FLAG_IGNORE) + return 0; + + /* + * In some cases, a device P might also be a supplier to its child node + * C. However, this would defer the probe of C until the probe of P + * completes successfully. This is perfectly fine in the device driver + * model. device_add() doesn't guarantee probe completion of the device + * by the time it returns. + * + * However, there are a few drivers that assume C will finish probing + * as soon as it's added and before P finishes probing. So, we provide + * a flag to let fw_devlink know not to delay the probe of C until the + * probe of P completes successfully. + * + * When such a flag is set, we can't create device links where P is the + * supplier of C as that would delay the probe of C. + */ + if (sup_handle->flags & FWNODE_FLAG_NEEDS_CHILD_BOUND_ON_ADD && + fwnode_is_ancestor_of(sup_handle, con->fwnode)) + return -EINVAL; + + /* + * Don't try to optimize by not calling the cycle detection logic under + * certain conditions. There's always some corner case that won't get + * detected. + */ + device_links_write_lock(); + if (__fw_devlink_relax_cycles(link->consumer, sup_handle)) { + __fwnode_link_cycle(link); + pr_debug("----- cycle: end -----\n"); + pr_info("%pfwf: Fixed dependency cycle(s) with %pfwf\n", + link->consumer, sup_handle); + } + device_links_write_unlock(); + + if (con->fwnode == link->consumer) + flags = fw_devlink_get_flags(link->flags); + else + flags = FW_DEVLINK_FLAGS_PERMISSIVE; + + if (sup_handle->flags & FWNODE_FLAG_NOT_DEVICE) + sup_dev = fwnode_get_next_parent_dev(sup_handle); + else + sup_dev = get_dev_from_fwnode(sup_handle); - sup_dev = get_dev_from_fwnode(sup_handle); if (sup_dev) { /* * If it's one of those drivers that don't actually bind to @@ -1726,64 +2163,34 @@ static int fw_devlink_create_devlink(struct device *con, */ if (sup_dev->links.status == DL_DEV_NO_DRIVER && sup_handle->flags & FWNODE_FLAG_INITIALIZED) { + dev_dbg(con, + "Not linking %pfwf - dev might never probe\n", + sup_handle); ret = -EINVAL; goto out; } - /* - * If this fails, it is due to cycles in device links. Just - * give up on this link and treat it as invalid. - */ - if (!device_link_add(con, sup_dev, flags) && - !(flags & DL_FLAG_SYNC_STATE_ONLY)) { - dev_info(con, "Fixing up cyclic dependency with %s\n", - dev_name(sup_dev)); - device_links_write_lock(); - fw_devlink_relax_cycle(con, sup_dev); - device_links_write_unlock(); - device_link_add(con, sup_dev, - FW_DEVLINK_FLAGS_PERMISSIVE); + if (con != sup_dev && !device_link_add(con, sup_dev, flags)) { + dev_err(con, "Failed to create device link (0x%x) with supplier %s for %pfwf\n", + flags, dev_name(sup_dev), link->consumer); ret = -EINVAL; } goto out; } - /* Supplier that's already initialized without a struct device. */ - if (sup_handle->flags & FWNODE_FLAG_INITIALIZED) - return -EINVAL; - /* - * DL_FLAG_SYNC_STATE_ONLY doesn't block probing and supports - * cycles. So cycle detection isn't necessary and shouldn't be - * done. + * Supplier or supplier's ancestor already initialized without a struct + * device or being probed by a driver. */ - if (flags & DL_FLAG_SYNC_STATE_ONLY) - return -EAGAIN; - - /* - * If we can't find the supplier device from its fwnode, it might be - * due to a cyclic dependency between fwnodes. Some of these cycles can - * be broken by applying logic. Check for these types of cycles and - * break them so that devices in the cycle probe properly. - * - * If the supplier's parent is dependent on the consumer, then - * the consumer-supplier dependency is a false dependency. So, - * treat it as an invalid link. - */ - sup_dev = fwnode_get_next_parent_dev(sup_handle); - if (sup_dev && device_is_dependent(con, sup_dev)) { - dev_dbg(con, "Not linking to %pfwP - False link\n", + if (fwnode_init_without_drv(sup_handle) || + fwnode_ancestor_init_without_drv(sup_handle)) { + dev_dbg(con, "Not linking %pfwf - might never become dev\n", sup_handle); - ret = -EINVAL; - } else { - /* - * Can't check for cycles or no cycles. So let's try - * again later. - */ - ret = -EAGAIN; + return -EINVAL; } + ret = -EAGAIN; out: put_device(sup_dev); return ret; @@ -1811,7 +2218,6 @@ static void __fw_devlink_link_to_consumers(struct device *dev) struct fwnode_link *link, *tmp; list_for_each_entry_safe(link, tmp, &fwnode->consumers, s_hook) { - u32 dl_flags = fw_devlink_get_flags(); struct device *con_dev; bool own_link = true; int ret; @@ -1841,21 +2247,18 @@ static void __fw_devlink_link_to_consumers(struct device *dev) con_dev = NULL; } else { own_link = false; - dl_flags = FW_DEVLINK_FLAGS_PERMISSIVE; } } if (!con_dev) continue; - ret = fw_devlink_create_devlink(con_dev, fwnode, dl_flags); + ret = fw_devlink_create_devlink(con_dev, fwnode, link); put_device(con_dev); if (!own_link || ret == -EAGAIN) continue; - list_del(&link->s_hook); - list_del(&link->c_hook); - kfree(link); + __fwnode_link_del(link); } } @@ -1870,10 +2273,7 @@ static void __fw_devlink_link_to_consumers(struct device *dev) * * The function creates normal (non-SYNC_STATE_ONLY) device links between @dev * and the real suppliers of @dev. Once these device links are created, the - * fwnode links are deleted. When such device links are successfully created, - * this function is called recursively on those supplier devices. This is - * needed to detect and break some invalid cycles in fwnode links. See - * fw_devlink_create_devlink() for more details. + * fwnode links are deleted. * * In addition, it also looks at all the suppliers of the entire fwnode tree * because some of the child devices of @dev that have not been added yet @@ -1891,46 +2291,16 @@ static void __fw_devlink_link_to_suppliers(struct device *dev, bool own_link = (dev->fwnode == fwnode); struct fwnode_link *link, *tmp; struct fwnode_handle *child = NULL; - u32 dl_flags; - - if (own_link) - dl_flags = fw_devlink_get_flags(); - else - dl_flags = FW_DEVLINK_FLAGS_PERMISSIVE; list_for_each_entry_safe(link, tmp, &fwnode->suppliers, c_hook) { int ret; - struct device *sup_dev; struct fwnode_handle *sup = link->supplier; - ret = fw_devlink_create_devlink(dev, sup, dl_flags); + ret = fw_devlink_create_devlink(dev, sup, link); if (!own_link || ret == -EAGAIN) continue; - list_del(&link->s_hook); - list_del(&link->c_hook); - kfree(link); - - /* If no device link was created, nothing more to do. */ - if (ret) - continue; - - /* - * If a device link was successfully created to a supplier, we - * now need to try and link the supplier to all its suppliers. - * - * This is needed to detect and delete false dependencies in - * fwnode links that haven't been converted to a device link - * yet. See comments in fw_devlink_create_devlink() for more - * details on the false dependency. - * - * Without deleting these false dependencies, some devices will - * never probe because they'll keep waiting for their false - * dependency fwnode links to be converted to device links. - */ - sup_dev = get_dev_from_fwnode(sup); - __fw_devlink_link_to_suppliers(sup_dev, sup_dev->fwnode); - put_device(sup_dev); + __fwnode_link_del(link); } /* @@ -1952,19 +2322,21 @@ static void fw_devlink_link_device(struct device *dev) fw_devlink_parse_fwtree(fwnode); - mutex_lock(&fwnode_link_lock); + guard(mutex)(&fwnode_link_lock); + __fw_devlink_link_to_consumers(dev); __fw_devlink_link_to_suppliers(dev, fwnode); - mutex_unlock(&fwnode_link_lock); } /* Device links support end. */ -int (*platform_notify)(struct device *dev) = NULL; -int (*platform_notify_remove)(struct device *dev) = NULL; static struct kobject *dev_kobj; -struct kobject *sysfs_dev_char_kobj; -struct kobject *sysfs_dev_block_kobj; + +/* /sys/dev/char */ +static struct kobject *sysfs_dev_char_kobj; + +/* /sys/dev/block */ +static struct kobject *sysfs_dev_block_kobj; static DEFINE_MUTEX(device_hotplug_lock); @@ -2000,24 +2372,18 @@ static inline int device_is_not_partition(struct device *dev) } #endif -static int -device_platform_notify(struct device *dev, enum kobject_action action) +static void device_platform_notify(struct device *dev) { - int ret; + acpi_device_notify(dev); - ret = acpi_platform_notify(dev, action); - if (ret) - return ret; + software_node_notify(dev); +} - ret = software_node_notify(dev, action); - if (ret) - return ret; +static void device_platform_notify_remove(struct device *dev) +{ + software_node_notify_remove(dev); - if (platform_notify && action == KOBJ_ADD) - platform_notify(dev); - else if (platform_notify_remove && action == KOBJ_REMOVE) - platform_notify_remove(dev); - return 0; + acpi_device_notify_remove(dev); } /** @@ -2140,7 +2506,7 @@ ssize_t device_store_bool(struct device *dev, struct device_attribute *attr, { struct dev_ext_attribute *ea = to_ext_attr(attr); - if (strtobool(buf, ea->var) < 0) + if (kstrtobool(buf, ea->var) < 0) return -EINVAL; return size; @@ -2156,6 +2522,15 @@ ssize_t device_show_bool(struct device *dev, struct device_attribute *attr, } EXPORT_SYMBOL_GPL(device_show_bool); +ssize_t device_show_string(struct device *dev, + struct device_attribute *attr, char *buf) +{ + struct dev_ext_attribute *ea = to_ext_attr(attr); + + return sysfs_emit(buf, "%s\n", (char *)ea->var); +} +EXPORT_SYMBOL_GPL(device_show_string); + /** * device_release - free device structure. * @kobj: device's kobject. @@ -2194,26 +2569,26 @@ static void device_release(struct kobject *kobj) kfree(p); } -static const void *device_namespace(struct kobject *kobj) +static const void *device_namespace(const struct kobject *kobj) { - struct device *dev = kobj_to_dev(kobj); + const struct device *dev = kobj_to_dev(kobj); const void *ns = NULL; - if (dev->class && dev->class->ns_type) + if (dev->class && dev->class->namespace) ns = dev->class->namespace(dev); return ns; } -static void device_get_ownership(struct kobject *kobj, kuid_t *uid, kgid_t *gid) +static void device_get_ownership(const struct kobject *kobj, kuid_t *uid, kgid_t *gid) { - struct device *dev = kobj_to_dev(kobj); + const struct device *dev = kobj_to_dev(kobj); if (dev->class && dev->class->get_ownership) dev->class->get_ownership(dev, uid, gid); } -static struct kobj_type device_ktype = { +static const struct kobj_type device_ktype = { .release = device_release, .sysfs_ops = &dev_sysfs_ops, .namespace = device_namespace, @@ -2221,12 +2596,12 @@ static struct kobj_type device_ktype = { }; -static int dev_uevent_filter(struct kset *kset, struct kobject *kobj) +static int dev_uevent_filter(const struct kobject *kobj) { - struct kobj_type *ktype = get_ktype(kobj); + const struct kobj_type *ktype = get_ktype(kobj); if (ktype == &device_ktype) { - struct device *dev = kobj_to_dev(kobj); + const struct device *dev = kobj_to_dev(kobj); if (dev->bus) return 1; if (dev->class) @@ -2235,9 +2610,9 @@ static int dev_uevent_filter(struct kset *kset, struct kobject *kobj) return 0; } -static const char *dev_uevent_name(struct kset *kset, struct kobject *kobj) +static const char *dev_uevent_name(const struct kobject *kobj) { - struct device *dev = kobj_to_dev(kobj); + const struct device *dev = kobj_to_dev(kobj); if (dev->bus) return dev->bus->name; @@ -2246,10 +2621,38 @@ static const char *dev_uevent_name(struct kset *kset, struct kobject *kobj) return NULL; } -static int dev_uevent(struct kset *kset, struct kobject *kobj, - struct kobj_uevent_env *env) +/* + * Try filling "DRIVER=<name>" uevent variable for a device. Because this + * function may race with binding and unbinding the device from a driver, + * we need to be careful. Binding is generally safe, at worst we miss the + * fact that the device is already bound to a driver (but the driver + * information that is delivered through uevents is best-effort, it may + * become obsolete as soon as it is generated anyways). Unbinding is more + * risky as driver pointer is transitioning to NULL, so READ_ONCE() should + * be used to make sure we are dealing with the same pointer, and to + * ensure that driver structure is not going to disappear from under us + * we take bus' drivers klist lock. The assumption that only registered + * driver can be bound to a device, and to unregister a driver bus code + * will take the same lock. + */ +static void dev_driver_uevent(const struct device *dev, struct kobj_uevent_env *env) { - struct device *dev = kobj_to_dev(kobj); + struct subsys_private *sp = bus_to_subsys(dev->bus); + + if (sp) { + scoped_guard(spinlock, &sp->klist_drivers.k_lock) { + struct device_driver *drv = READ_ONCE(dev->driver); + if (drv) + add_uevent_var(env, "DRIVER=%s", drv->name); + } + + subsys_put(sp); + } +} + +static int dev_uevent(const struct kobject *kobj, struct kobj_uevent_env *env) +{ + const struct device *dev = kobj_to_dev(kobj); int retval = 0; /* add device node properties if present */ @@ -2278,8 +2681,8 @@ static int dev_uevent(struct kset *kset, struct kobject *kobj, if (dev->type && dev->type->name) add_uevent_var(env, "DEVTYPE=%s", dev->type->name); - if (dev->driver) - add_uevent_var(env, "DRIVER=%s", dev->driver->name); + /* Add "DRIVER=%s" variable if the device is bound to a driver */ + dev_driver_uevent(dev, env); /* Add common DT information about the device */ of_device_uevent(dev, env); @@ -2342,7 +2745,7 @@ static ssize_t uevent_show(struct device *dev, struct device_attribute *attr, /* respect filter */ if (kset->uevent_ops && kset->uevent_ops->filter) - if (!kset->uevent_ops->filter(kset, &dev->kobj)) + if (!kset->uevent_ops->filter(&dev->kobj)) goto out; env = kzalloc(sizeof(struct kobj_uevent_env), GFP_KERNEL); @@ -2350,7 +2753,7 @@ static ssize_t uevent_show(struct device *dev, struct device_attribute *attr, return -ENOMEM; /* let the kset specific function add its keys */ - retval = kset->uevent_ops->uevent(kset, &dev->kobj, env); + retval = kset->uevent_ops->uevent(&dev->kobj, env); if (retval) goto out; @@ -2370,7 +2773,7 @@ static ssize_t uevent_store(struct device *dev, struct device_attribute *attr, rc = kobject_synth_uevent(&dev->kobj, buf, count); if (rc) { - dev_err(dev, "uevent: failed to send synthetic uevent\n"); + dev_err(dev, "uevent: failed to send synthetic uevent: %d\n", rc); return rc; } @@ -2395,7 +2798,7 @@ static ssize_t online_store(struct device *dev, struct device_attribute *attr, bool val; int ret; - ret = strtobool(buf, &val); + ret = kstrtobool(buf, &val); if (ret < 0) return ret; @@ -2446,11 +2849,6 @@ union device_attr_group_devres { const struct attribute_group **groups; }; -static int devm_attr_group_match(struct device *dev, void *res, void *data) -{ - return ((union device_attr_group_devres *)res)->group == data; -} - static void devm_attr_group_remove(struct device *dev, void *res) { union device_attr_group_devres *devres = res; @@ -2460,15 +2858,6 @@ static void devm_attr_group_remove(struct device *dev, void *res) sysfs_remove_group(&dev->kobj, group); } -static void devm_attr_groups_remove(struct device *dev, void *res) -{ - union device_attr_group_devres *devres = res; - const struct attribute_group **groups = devres->groups; - - dev_dbg(dev, "%s: removing groups %p\n", __func__, groups); - sysfs_remove_groups(&dev->kobj, groups); -} - /** * devm_device_add_group - given a device, create a managed attribute group * @dev: The device to create the group for @@ -2501,79 +2890,9 @@ int devm_device_add_group(struct device *dev, const struct attribute_group *grp) } EXPORT_SYMBOL_GPL(devm_device_add_group); -/** - * devm_device_remove_group: remove a managed group from a device - * @dev: device to remove the group from - * @grp: group to remove - * - * This function removes a group of attributes from a device. The attributes - * previously have to have been created for this group, otherwise it will fail. - */ -void devm_device_remove_group(struct device *dev, - const struct attribute_group *grp) -{ - WARN_ON(devres_release(dev, devm_attr_group_remove, - devm_attr_group_match, - /* cast away const */ (void *)grp)); -} -EXPORT_SYMBOL_GPL(devm_device_remove_group); - -/** - * devm_device_add_groups - create a bunch of managed attribute groups - * @dev: The device to create the group for - * @groups: The attribute groups to create, NULL terminated - * - * This function creates a bunch of managed attribute groups. If an error - * occurs when creating a group, all previously created groups will be - * removed, unwinding everything back to the original state when this - * function was called. It will explicitly warn and error if any of the - * attribute files being created already exist. - * - * Returns 0 on success or error code from sysfs_create_group on failure. - */ -int devm_device_add_groups(struct device *dev, - const struct attribute_group **groups) -{ - union device_attr_group_devres *devres; - int error; - - devres = devres_alloc(devm_attr_groups_remove, - sizeof(*devres), GFP_KERNEL); - if (!devres) - return -ENOMEM; - - error = sysfs_create_groups(&dev->kobj, groups); - if (error) { - devres_free(devres); - return error; - } - - devres->groups = groups; - devres_add(dev, devres); - return 0; -} -EXPORT_SYMBOL_GPL(devm_device_add_groups); - -/** - * devm_device_remove_groups - remove a list of managed groups - * - * @dev: The device for the groups to be removed from - * @groups: NULL terminated list of groups to be removed - * - * If groups is not NULL, remove the specified groups from the device. - */ -void devm_device_remove_groups(struct device *dev, - const struct attribute_group **groups) -{ - WARN_ON(devres_release(dev, devm_attr_groups_remove, - devm_attr_group_match, - /* cast away const */ (void *)groups)); -} -EXPORT_SYMBOL_GPL(devm_device_remove_groups); - static int device_add_attrs(struct device *dev) { - struct class *class = dev->class; + const struct class *class = dev->class; const struct device_type *type = dev->type; int error; @@ -2611,8 +2930,17 @@ static int device_add_attrs(struct device *dev) goto err_remove_dev_waiting_for_supplier; } + if (dev_add_physical_location(dev)) { + error = device_add_group(dev, + &dev_attr_physical_location_group); + if (error) + goto err_remove_dev_removable; + } + return 0; + err_remove_dev_removable: + device_remove_file(dev, &dev_attr_removable); err_remove_dev_waiting_for_supplier: device_remove_file(dev, &dev_attr_waiting_for_supplier); err_remove_dev_online: @@ -2631,9 +2959,14 @@ static int device_add_attrs(struct device *dev) static void device_remove_attrs(struct device *dev) { - struct class *class = dev->class; + const struct class *class = dev->class; const struct device_type *type = dev->type; + if (dev->physical_location) { + device_remove_group(dev, &dev_attr_physical_location_group); + kfree(dev->physical_location); + } + device_remove_file(dev, &dev_attr_removable); device_remove_file(dev, &dev_attr_waiting_for_supplier); device_remove_file(dev, &dev_attr_online); @@ -2826,17 +3159,11 @@ void device_initialize(struct device *dev) kobject_init(&dev->kobj, &device_ktype); INIT_LIST_HEAD(&dev->dma_pools); mutex_init(&dev->mutex); -#ifdef CONFIG_PROVE_LOCKING - mutex_init(&dev->lockdep_mutex); -#endif lockdep_set_novalidate_class(&dev->mutex); spin_lock_init(&dev->devres_lock); INIT_LIST_HEAD(&dev->devres_head); device_pm_init(dev); - set_dev_node(dev, -1); -#ifdef CONFIG_GENERIC_MSI_IRQ - INIT_LIST_HEAD(&dev->msi_list); -#endif + set_dev_node(dev, NUMA_NO_NODE); INIT_LIST_HEAD(&dev->links.consumers); INIT_LIST_HEAD(&dev->links.suppliers); INIT_LIST_HEAD(&dev->links.defer_sync); @@ -2846,10 +3173,11 @@ void device_initialize(struct device *dev) defined(CONFIG_ARCH_HAS_SYNC_DMA_FOR_CPU_ALL) dev->dma_coherent = dma_default_coherent; #endif + swiotlb_dev_init(dev); } EXPORT_SYMBOL_GPL(device_initialize); -struct kobject *virtual_device_parent(struct device *dev) +struct kobject *virtual_device_parent(void) { static struct kobject *virtual_dir = NULL; @@ -2862,7 +3190,7 @@ struct kobject *virtual_device_parent(struct device *dev) struct class_dir { struct kobject kobj; - struct class *class; + const struct class *class; }; #define to_class_dir(obj) container_of(obj, struct class_dir, kobj) @@ -2874,20 +3202,20 @@ static void class_dir_release(struct kobject *kobj) } static const -struct kobj_ns_type_operations *class_dir_child_ns_type(struct kobject *kobj) +struct kobj_ns_type_operations *class_dir_child_ns_type(const struct kobject *kobj) { - struct class_dir *dir = to_class_dir(kobj); + const struct class_dir *dir = to_class_dir(kobj); return dir->class->ns_type; } -static struct kobj_type class_dir_ktype = { +static const struct kobj_type class_dir_ktype = { .release = class_dir_release, .sysfs_ops = &kobj_sysfs_ops, .child_ns_type = class_dir_child_ns_type }; -static struct kobject * -class_dir_create_and_add(struct class *class, struct kobject *parent_kobj) +static struct kobject *class_dir_create_and_add(struct subsys_private *sp, + struct kobject *parent_kobj) { struct class_dir *dir; int retval; @@ -2896,12 +3224,12 @@ class_dir_create_and_add(struct class *class, struct kobject *parent_kobj) if (!dir) return ERR_PTR(-ENOMEM); - dir->class = class; + dir->class = sp->class; kobject_init(&dir->kobj, &class_dir_ktype); - dir->kobj.kset = &class->p->glue_dirs; + dir->kobj.kset = &sp->glue_dirs; - retval = kobject_add(&dir->kobj, parent_kobj, "%s", class->name); + retval = kobject_add(&dir->kobj, parent_kobj, "%s", sp->class->name); if (retval < 0) { kobject_put(&dir->kobj); return ERR_PTR(retval); @@ -2914,57 +3242,61 @@ static DEFINE_MUTEX(gdp_mutex); static struct kobject *get_device_parent(struct device *dev, struct device *parent) { - if (dev->class) { - struct kobject *kobj = NULL; + struct subsys_private *sp = class_to_subsys(dev->class); + struct kobject *kobj = NULL; + + if (sp) { struct kobject *parent_kobj; struct kobject *k; -#ifdef CONFIG_BLOCK - /* block disks show up in /sys/block */ - if (sysfs_deprecated && dev->class == &block_class) { - if (parent && parent->class == &block_class) - return &parent->kobj; - return &block_class.p->subsys.kobj; - } -#endif - /* * If we have no parent, we live in "virtual". * Class-devices with a non class-device as parent, live * in a "glue" directory to prevent namespace collisions. */ if (parent == NULL) - parent_kobj = virtual_device_parent(dev); - else if (parent->class && !dev->class->ns_type) + parent_kobj = virtual_device_parent(); + else if (parent->class && !dev->class->ns_type) { + subsys_put(sp); return &parent->kobj; - else + } else { parent_kobj = &parent->kobj; + } mutex_lock(&gdp_mutex); /* find our class-directory at the parent and reference it */ - spin_lock(&dev->class->p->glue_dirs.list_lock); - list_for_each_entry(k, &dev->class->p->glue_dirs.list, entry) + spin_lock(&sp->glue_dirs.list_lock); + list_for_each_entry(k, &sp->glue_dirs.list, entry) if (k->parent == parent_kobj) { kobj = kobject_get(k); break; } - spin_unlock(&dev->class->p->glue_dirs.list_lock); + spin_unlock(&sp->glue_dirs.list_lock); if (kobj) { mutex_unlock(&gdp_mutex); + subsys_put(sp); return kobj; } /* or create a new class-directory at the parent device */ - k = class_dir_create_and_add(dev->class, parent_kobj); + k = class_dir_create_and_add(sp, parent_kobj); /* do not emit an uevent for this simple "glue" directory */ mutex_unlock(&gdp_mutex); + subsys_put(sp); return k; } /* subsystems can specify a default root directory for their devices */ - if (!parent && dev->bus && dev->bus->dev_root) - return &dev->bus->dev_root->kobj; + if (!parent && dev->bus) { + struct device *dev_root = bus_get_dev_root(dev->bus); + + if (dev_root) { + kobj = &dev_root->kobj; + put_device(dev_root); + return kobj; + } + } if (parent) return &parent->kobj; @@ -2974,10 +3306,23 @@ static struct kobject *get_device_parent(struct device *dev, static inline bool live_in_glue_dir(struct kobject *kobj, struct device *dev) { - if (!kobj || !dev->class || - kobj->kset != &dev->class->p->glue_dirs) + struct subsys_private *sp; + bool retval; + + if (!kobj || !dev->class) return false; - return true; + + sp = class_to_subsys(dev->class); + if (!sp) + return false; + + if (kobj->kset == &sp->glue_dirs) + retval = true; + else + retval = false; + + subsys_put(sp); + return retval; } static inline struct kobject *get_glue_dir(struct device *dev) @@ -2985,6 +3330,23 @@ static inline struct kobject *get_glue_dir(struct device *dev) return dev->kobj.parent; } +/** + * kobject_has_children - Returns whether a kobject has children. + * @kobj: the object to test + * + * This will return whether a kobject has other kobjects as children. + * + * It does NOT account for the presence of attribute files, only sub + * directories. It also assumes there is no concurrent addition or + * removal of such children, and thus relies on external locking. + */ +static inline bool kobject_has_children(struct kobject *kobj) +{ + WARN_ON_ONCE(kref_read(&kobj->kref) == 0); + + return kobj->sd && kobj->sd->dir.subdirs; +} + /* * make sure cleaning up dir as the last step, we need to make * sure .release handler of kobject is run with holding the @@ -3057,6 +3419,7 @@ static void cleanup_glue_dir(struct device *dev, struct kobject *glue_dir) static int device_add_class_symlinks(struct device *dev) { struct device_node *of_node = dev_of_node(dev); + struct subsys_private *sp; int error; if (of_node) { @@ -3066,12 +3429,11 @@ static int device_add_class_symlinks(struct device *dev) /* An error here doesn't warrant bringing down the device */ } - if (!dev->class) + sp = class_to_subsys(dev->class); + if (!sp) return 0; - error = sysfs_create_link(&dev->kobj, - &dev->class->p->subsys.kobj, - "subsystem"); + error = sysfs_create_link(&dev->kobj, &sp->subsys.kobj, "subsystem"); if (error) goto out_devnode; @@ -3082,46 +3444,38 @@ static int device_add_class_symlinks(struct device *dev) goto out_subsys; } -#ifdef CONFIG_BLOCK - /* /sys/block has directories and does not need symlinks */ - if (sysfs_deprecated && dev->class == &block_class) - return 0; -#endif - /* link in the class directory pointing to the device */ - error = sysfs_create_link(&dev->class->p->subsys.kobj, - &dev->kobj, dev_name(dev)); + error = sysfs_create_link(&sp->subsys.kobj, &dev->kobj, dev_name(dev)); if (error) goto out_device; - - return 0; + goto exit; out_device: sysfs_remove_link(&dev->kobj, "device"); - out_subsys: sysfs_remove_link(&dev->kobj, "subsystem"); out_devnode: sysfs_remove_link(&dev->kobj, "of_node"); +exit: + subsys_put(sp); return error; } static void device_remove_class_symlinks(struct device *dev) { + struct subsys_private *sp = class_to_subsys(dev->class); + if (dev_of_node(dev)) sysfs_remove_link(&dev->kobj, "of_node"); - if (!dev->class) + if (!sp) return; if (dev->parent && device_is_not_partition(dev)) sysfs_remove_link(&dev->kobj, "device"); sysfs_remove_link(&dev->kobj, "subsystem"); -#ifdef CONFIG_BLOCK - if (sysfs_deprecated && dev->class == &block_class) - return; -#endif - sysfs_delete_link(&dev->class->p->subsys.kobj, &dev->kobj, dev_name(dev)); + sysfs_delete_link(&sp->subsys.kobj, &dev->kobj, dev_name(dev)); + subsys_put(sp); } /** @@ -3141,27 +3495,13 @@ int dev_set_name(struct device *dev, const char *fmt, ...) } EXPORT_SYMBOL_GPL(dev_set_name); -/** - * device_to_dev_kobj - select a /sys/dev/ directory for the device - * @dev: device - * - * By default we select char/ for new entries. Setting class->dev_obj - * to NULL prevents an entry from being created. class->dev_kobj must - * be set (or cleared) before any devices are registered to the class - * otherwise device_create_sys_dev_entry() and - * device_remove_sys_dev_entry() will disagree about the presence of - * the link. - */ +/* select a /sys/dev/ directory for the device */ static struct kobject *device_to_dev_kobj(struct device *dev) { - struct kobject *kobj; - - if (dev->class) - kobj = dev->class->dev_kobj; + if (is_blockdev(dev)) + return sysfs_dev_block_kobj; else - kobj = sysfs_dev_char_kobj; - - return kobj; + return sysfs_dev_char_kobj; } static int device_create_sys_dev_entry(struct device *dev) @@ -3230,6 +3570,7 @@ static int device_private_init(struct device *dev) */ int device_add(struct device *dev) { + struct subsys_private *sp; struct device *parent; struct kobject *kobj; struct class_interface *class_intf; @@ -3252,18 +3593,19 @@ int device_add(struct device *dev) * the name, and force the use of dev_name() */ if (dev->init_name) { - dev_set_name(dev, "%s", dev->init_name); + error = dev_set_name(dev, "%s", dev->init_name); dev->init_name = NULL; } + if (dev_name(dev)) + error = 0; /* subsystems can specify simple device enumeration */ - if (!dev_name(dev) && dev->bus && dev->bus->dev_name) - dev_set_name(dev, "%s%u", dev->bus->dev_name, dev->id); - - if (!dev_name(dev)) { + else if (dev->bus && dev->bus->dev_name) + error = dev_set_name(dev, "%s%u", dev->bus->dev_name, dev->id); + else error = -EINVAL; + if (error) goto name_error; - } pr_debug("device: '%s': %s\n", dev_name(dev), __func__); @@ -3284,14 +3626,12 @@ int device_add(struct device *dev) /* we require the name to be set before, and pass NULL */ error = kobject_add(&dev->kobj, dev->kobj.parent, NULL); if (error) { - glue_dir = get_glue_dir(dev); + glue_dir = kobj; goto Error; } /* notify platform of device entry */ - error = device_platform_notify(dev, KOBJ_ADD); - if (error) - goto platform_error; + device_platform_notify(dev); error = device_create_file(dev, &dev_attr_uevent); if (error) @@ -3326,10 +3666,7 @@ int device_add(struct device *dev) /* Notify clients of device addition. This call must come * after dpm_sysfs_add() and before kobject_uevent(). */ - if (dev->bus) - blocking_notifier_call_chain(&dev->bus->p->bus_notifier, - BUS_NOTIFY_ADD_DEVICE, dev); - + bus_notify(dev, BUS_NOTIFY_ADD_DEVICE); kobject_uevent(&dev->kobj, KOBJ_ADD); /* @@ -3363,18 +3700,18 @@ int device_add(struct device *dev) klist_add_tail(&dev->p->knode_parent, &parent->p->klist_children); - if (dev->class) { - mutex_lock(&dev->class->p->mutex); + sp = class_to_subsys(dev->class); + if (sp) { + mutex_lock(&sp->mutex); /* tie the class to the device */ - klist_add_tail(&dev->p->knode_class, - &dev->class->p->klist_devices); + klist_add_tail(&dev->p->knode_class, &sp->klist_devices); /* notify any interfaces that the device is here */ - list_for_each_entry(class_intf, - &dev->class->p->interfaces, node) + list_for_each_entry(class_intf, &sp->interfaces, node) if (class_intf->add_dev) - class_intf->add_dev(dev, class_intf); - mutex_unlock(&dev->class->p->mutex); + class_intf->add_dev(dev); + mutex_unlock(&sp->mutex); + subsys_put(sp); } done: put_device(dev); @@ -3386,6 +3723,7 @@ done: device_pm_remove(dev); dpm_sysfs_remove(dev); DPMError: + device_set_driver(dev, NULL); bus_remove_device(dev); BusError: device_remove_attrs(dev); @@ -3394,8 +3732,7 @@ done: SymlinkError: device_remove_file(dev, &dev_attr_uevent); attrError: - device_platform_notify(dev, KOBJ_REMOVE); -platform_error: + device_platform_notify_remove(dev); kobject_uevent(&dev->kobj, KOBJ_REMOVE); glue_dir = get_glue_dir(dev); kobject_del(&dev->kobj); @@ -3494,6 +3831,7 @@ EXPORT_SYMBOL_GPL(kill_device); */ void device_del(struct device *dev) { + struct subsys_private *sp; struct device *parent = dev->parent; struct kobject *glue_dir = NULL; struct class_interface *class_intf; @@ -3510,9 +3848,7 @@ void device_del(struct device *dev) * before dpm_sysfs_remove(). */ noio_flag = memalloc_noio_save(); - if (dev->bus) - blocking_notifier_call_chain(&dev->bus->p->bus_notifier, - BUS_NOTIFY_DEL_DEVICE, dev); + bus_notify(dev, BUS_NOTIFY_DEL_DEVICE); dpm_sysfs_remove(dev); if (parent) @@ -3522,31 +3858,41 @@ void device_del(struct device *dev) device_remove_sys_dev_entry(dev); device_remove_file(dev, &dev_attr_dev); } - if (dev->class) { + + sp = class_to_subsys(dev->class); + if (sp) { device_remove_class_symlinks(dev); - mutex_lock(&dev->class->p->mutex); + mutex_lock(&sp->mutex); /* notify any interfaces that the device is now gone */ - list_for_each_entry(class_intf, - &dev->class->p->interfaces, node) + list_for_each_entry(class_intf, &sp->interfaces, node) if (class_intf->remove_dev) - class_intf->remove_dev(dev, class_intf); + class_intf->remove_dev(dev); /* remove the device from the class list */ klist_del(&dev->p->knode_class); - mutex_unlock(&dev->class->p->mutex); + mutex_unlock(&sp->mutex); + subsys_put(sp); } device_remove_file(dev, &dev_attr_uevent); device_remove_attrs(dev); bus_remove_device(dev); device_pm_remove(dev); driver_deferred_probe_del(dev); - device_platform_notify(dev, KOBJ_REMOVE); - device_remove_properties(dev); + device_platform_notify_remove(dev); device_links_purge(dev); - if (dev->bus) - blocking_notifier_call_chain(&dev->bus->p->bus_notifier, - BUS_NOTIFY_REMOVED_DEVICE, dev); + /* + * If a device does not have a driver attached, we need to clean + * up any managed resources. We do this in device_release(), but + * it's never called (and we leak the device) if a managed + * resource holds a reference to the device. So release all + * managed resources here, like we do in driver_detach(). We + * still need to do so again in device_release() in case someone + * adds a new resource after this point, though. + */ + devres_release_all(dev); + + bus_notify(dev, BUS_NOTIFY_REMOVED_DEVICE); kobject_uevent(&dev->kobj, KOBJ_REMOVE); glue_dir = get_glue_dir(dev); kobject_del(&dev->kobj); @@ -3614,7 +3960,7 @@ static struct device *next_device(struct klist_iter *i) * a name. This memory is returned in tmp and needs to be * freed by the caller. */ -const char *device_get_devnode(struct device *dev, +const char *device_get_devnode(const struct device *dev, umode_t *mode, kuid_t *uid, kgid_t *gid, const char **tmp) { @@ -3639,18 +3985,17 @@ const char *device_get_devnode(struct device *dev, return dev_name(dev); /* replace '!' in the name with '/' */ - s = kstrdup(dev_name(dev), GFP_KERNEL); + s = kstrdup_and_replace(dev_name(dev), '!', '/', GFP_KERNEL); if (!s) return NULL; - strreplace(s, '!', '/'); return *tmp = s; } /** * device_for_each_child - device child iterator. * @parent: parent struct device. - * @fn: function to be called for each device. * @data: data for the callback. + * @fn: function to be called for each device. * * Iterate over @parent's child devices, and call @fn for each, * passing it @data. @@ -3659,13 +4004,13 @@ const char *device_get_devnode(struct device *dev, * other than 0, we break out and return that value. */ int device_for_each_child(struct device *parent, void *data, - int (*fn)(struct device *dev, void *data)) + device_iter_t fn) { struct klist_iter i; struct device *child; int error = 0; - if (!parent->p) + if (!parent || !parent->p) return 0; klist_iter_init(&parent->p->klist_children, &i); @@ -3679,8 +4024,8 @@ EXPORT_SYMBOL_GPL(device_for_each_child); /** * device_for_each_child_reverse - device child iterator in reversed order. * @parent: parent struct device. - * @fn: function to be called for each device. * @data: data for the callback. + * @fn: function to be called for each device. * * Iterate over @parent's child devices, and call @fn for each, * passing it @data. @@ -3689,13 +4034,13 @@ EXPORT_SYMBOL_GPL(device_for_each_child); * other than 0, we break out and return that value. */ int device_for_each_child_reverse(struct device *parent, void *data, - int (*fn)(struct device *dev, void *data)) + device_iter_t fn) { struct klist_iter i; struct device *child; int error = 0; - if (!parent->p) + if (!parent || !parent->p) return 0; klist_iter_init(&parent->p->klist_children, &i); @@ -3707,67 +4052,77 @@ int device_for_each_child_reverse(struct device *parent, void *data, EXPORT_SYMBOL_GPL(device_for_each_child_reverse); /** - * device_find_child - device iterator for locating a particular device. - * @parent: parent struct device - * @match: Callback function to check device - * @data: Data to pass to match function - * - * This is similar to the device_for_each_child() function above, but it - * returns a reference to a device that is 'found' for later use, as - * determined by the @match callback. + * device_for_each_child_reverse_from - device child iterator in reversed order. + * @parent: parent struct device. + * @from: optional starting point in child list + * @data: data for the callback. + * @fn: function to be called for each device. * - * The callback should return 0 if the device doesn't match and non-zero - * if it does. If the callback returns non-zero and a reference to the - * current device can be obtained, this function will return to the caller - * and not iterate over any more devices. + * Iterate over @parent's child devices, starting at @from, and call @fn + * for each, passing it @data. This helper is identical to + * device_for_each_child_reverse() when @from is NULL. * - * NOTE: you will need to drop the reference with put_device() after use. + * @fn is checked each iteration. If it returns anything other than 0, + * iteration stop and that value is returned to the caller of + * device_for_each_child_reverse_from(); */ -struct device *device_find_child(struct device *parent, void *data, - int (*match)(struct device *dev, void *data)) +int device_for_each_child_reverse_from(struct device *parent, + struct device *from, void *data, + device_iter_t fn) { struct klist_iter i; struct device *child; + int error = 0; - if (!parent) - return NULL; + if (!parent || !parent->p) + return 0; - klist_iter_init(&parent->p->klist_children, &i); - while ((child = next_device(&i))) - if (match(child, data) && get_device(child)) - break; + klist_iter_init_node(&parent->p->klist_children, &i, + (from ? &from->p->knode_parent : NULL)); + while ((child = prev_device(&i)) && !error) + error = fn(child, data); klist_iter_exit(&i); - return child; + return error; } -EXPORT_SYMBOL_GPL(device_find_child); +EXPORT_SYMBOL_GPL(device_for_each_child_reverse_from); /** - * device_find_child_by_name - device iterator for locating a child device. + * device_find_child - device iterator for locating a particular device. * @parent: parent struct device - * @name: name of the child device + * @data: Data to pass to match function + * @match: Callback function to check device * - * This is similar to the device_find_child() function above, but it - * returns a reference to a device that has the name @name. + * This is similar to the device_for_each_child() function above, but it + * returns a reference to a device that is 'found' for later use, as + * determined by the @match callback. + * + * The callback should return 0 if the device doesn't match and non-zero + * if it does. If the callback returns non-zero and a reference to the + * current device can be obtained, this function will return to the caller + * and not iterate over any more devices. * * NOTE: you will need to drop the reference with put_device() after use. */ -struct device *device_find_child_by_name(struct device *parent, - const char *name) +struct device *device_find_child(struct device *parent, const void *data, + device_match_t match) { struct klist_iter i; struct device *child; - if (!parent) + if (!parent || !parent->p) return NULL; klist_iter_init(&parent->p->klist_children, &i); - while ((child = next_device(&i))) - if (sysfs_streq(dev_name(child), name) && get_device(child)) + while ((child = next_device(&i))) { + if (match(child, data)) { + get_device(child); break; + } + } klist_iter_exit(&i); return child; } -EXPORT_SYMBOL_GPL(device_find_child_by_name); +EXPORT_SYMBOL_GPL(device_find_child); int __init devices_init(void) { @@ -3783,9 +4138,14 @@ int __init devices_init(void) sysfs_dev_char_kobj = kobject_create_and_add("char", dev_kobj); if (!sysfs_dev_char_kobj) goto char_kobj_err; + device_link_wq = alloc_workqueue("device_link_wq", WQ_PERCPU, 0); + if (!device_link_wq) + goto wq_err; return 0; + wq_err: + kobject_put(sysfs_dev_char_kobj); char_kobj_err: kobject_put(sysfs_dev_block_kobj); block_kobj_err: @@ -3979,7 +4339,7 @@ static void device_create_release(struct device *dev) } static __printf(6, 0) struct device * -device_create_groups_vargs(struct class *class, struct device *parent, +device_create_groups_vargs(const struct class *class, struct device *parent, dev_t devt, void *drvdata, const struct attribute_group **groups, const char *fmt, va_list args) @@ -3987,7 +4347,7 @@ device_create_groups_vargs(struct class *class, struct device *parent, struct device *dev = NULL; int retval = -ENODEV; - if (class == NULL || IS_ERR(class)) + if (IS_ERR_OR_NULL(class)) goto error; dev = kzalloc(sizeof(*dev), GFP_KERNEL); @@ -4039,11 +4399,8 @@ error: * pointer. * * Returns &struct device pointer on success, or ERR_PTR() on error. - * - * Note: the struct class passed to this function must have previously - * been created with a call to class_create(). */ -struct device *device_create(struct class *class, struct device *parent, +struct device *device_create(const struct class *class, struct device *parent, dev_t devt, void *drvdata, const char *fmt, ...) { va_list vargs; @@ -4080,11 +4437,8 @@ EXPORT_SYMBOL_GPL(device_create); * pointer. * * Returns &struct device pointer on success, or ERR_PTR() on error. - * - * Note: the struct class passed to this function must have previously - * been created with a call to class_create(). */ -struct device *device_create_with_groups(struct class *class, +struct device *device_create_with_groups(const struct class *class, struct device *parent, dev_t devt, void *drvdata, const struct attribute_group **groups, @@ -4109,7 +4463,7 @@ EXPORT_SYMBOL_GPL(device_create_with_groups); * This call unregisters and cleans up a device that was created with a * call to device_create(). */ -void device_destroy(struct class *class, dev_t devt) +void device_destroy(const struct class *class, dev_t devt) { struct device *dev; @@ -4131,9 +4485,12 @@ EXPORT_SYMBOL_GPL(device_destroy); * on the same device to ensure that new_name is valid and * won't conflict with other devices. * - * Note: Don't call this function. Currently, the networking layer calls this - * function, but that will change. The following text from Kay Sievers offers - * some insight: + * Note: given that some subsystems (networking and infiniband) use this + * function, with no immediate plans for this to change, we cannot assume or + * require that this function not be called at all. + * + * However, if you're writing new code, do not call this function. The following + * text from Kay Sievers offers some insight: * * Renaming devices is racy at many levels, symlinks and other stuff are not * replaced atomically, and you get a "move" uevent, but it's not easy to @@ -4147,13 +4504,6 @@ EXPORT_SYMBOL_GPL(device_destroy); * kernel device renaming. Besides that, it's not even implemented now for * other things than (driver-core wise very simple) network devices. * - * We are currently about to change network renaming in udev to completely - * disallow renaming of devices in the same namespace as the kernel uses, - * because we can't solve the problems properly, that arise with swapping names - * of multiple interfaces without races. Means, renaming of eth[0-9]* will only - * be allowed to some other name than eth[0-9]*, for the aforementioned - * reasons. - * * Make up a "real" name in the driver before you register anything, or add * some other attributes for userspace to find the device, or use udev to add * symlinks -- but never rename kernel devices later, it's a complete mess. We @@ -4162,9 +4512,11 @@ EXPORT_SYMBOL_GPL(device_destroy); */ int device_rename(struct device *dev, const char *new_name) { + struct subsys_private *sp = NULL; struct kobject *kobj = &dev->kobj; char *old_device_name = NULL; int error; + bool is_link_renamed = false; dev = get_device(dev); if (!dev) @@ -4179,18 +4531,28 @@ int device_rename(struct device *dev, const char *new_name) } if (dev->class) { - error = sysfs_rename_link_ns(&dev->class->p->subsys.kobj, - kobj, old_device_name, + sp = class_to_subsys(dev->class); + + if (!sp) { + error = -EINVAL; + goto out; + } + + error = sysfs_rename_link_ns(&sp->subsys.kobj, kobj, old_device_name, new_name, kobject_namespace(kobj)); if (error) goto out; + + is_link_renamed = true; } error = kobject_rename(kobj, new_name); - if (error) - goto out; - out: + if (error && is_link_renamed) + sysfs_rename_link_ns(&sp->subsys.kobj, kobj, new_name, + old_device_name, kobject_namespace(kobj)); + subsys_put(sp); + put_device(dev); kfree(old_device_name); @@ -4306,7 +4668,7 @@ static int device_attrs_change_owner(struct device *dev, kuid_t kuid, kgid_t kgid) { struct kobject *kobj = &dev->kobj; - struct class *class = dev->class; + const struct class *class = dev->class; const struct device_type *type = dev->type; int error; @@ -4364,6 +4726,7 @@ int device_change_owner(struct device *dev, kuid_t kuid, kgid_t kgid) { int error; struct kobject *kobj = &dev->kobj; + struct subsys_private *sp; dev = get_device(dev); if (!dev) @@ -4400,21 +4763,19 @@ int device_change_owner(struct device *dev, kuid_t kuid, kgid_t kgid) if (error) goto out; -#ifdef CONFIG_BLOCK - if (sysfs_deprecated && dev->class == &block_class) - goto out; -#endif - /* * Change the owner of the symlink located in the class directory of * the device class associated with @dev which points to the actual * directory entry for @dev to @kuid/@kgid. This ensures that the * symlink shows the same permissions as its target. */ - error = sysfs_link_change_owner(&dev->class->p->subsys.kobj, &dev->kobj, - dev_name(dev), kuid, kgid); - if (error) + sp = class_to_subsys(dev->class); + if (!sp) { + error = -EINVAL; goto out; + } + error = sysfs_link_change_owner(&sp->subsys.kobj, &dev->kobj, dev_name(dev), kuid, kgid); + subsys_put(sp); out: put_device(dev); @@ -4513,7 +4874,7 @@ set_dev_info(const struct device *dev, struct dev_printk_info *dev_info) else return; - strscpy(dev_info->subsystem, subsys, sizeof(dev_info->subsystem)); + strscpy(dev_info->subsystem, subsys); /* * Add device identifier DEVICE=: @@ -4579,8 +4940,8 @@ static void __dev_printk(const char *level, const struct device *dev, printk("%s(NULL device *): %pV", level, vaf); } -void dev_printk(const char *level, const struct device *dev, - const char *fmt, ...) +void _dev_printk(const char *level, const struct device *dev, + const char *fmt, ...) { struct va_format vaf; va_list args; @@ -4594,7 +4955,7 @@ void dev_printk(const char *level, const struct device *dev, va_end(args); } -EXPORT_SYMBOL(dev_printk); +EXPORT_SYMBOL(_dev_printk); #define define_dev_printk_level(func, kern_level) \ void func(const struct device *dev, const char *fmt, ...) \ @@ -4623,6 +4984,49 @@ define_dev_printk_level(_dev_info, KERN_INFO); #endif +static void __dev_probe_failed(const struct device *dev, int err, bool fatal, + const char *fmt, va_list vargsp) +{ + struct va_format vaf; + va_list vargs; + + /* + * On x86_64 and possibly on other architectures, va_list is actually a + * size-1 array containing a structure. As a result, function parameter + * vargsp decays from T[1] to T*, and &vargsp has type T** rather than + * T(*)[1], which is expected by its assignment to vaf.va below. + * + * One standard way to solve this mess is by creating a copy in a local + * variable of type va_list and then using a pointer to that local copy + * instead, which is the approach employed here. + */ + va_copy(vargs, vargsp); + + vaf.fmt = fmt; + vaf.va = &vargs; + + switch (err) { + case -EPROBE_DEFER: + device_set_deferred_probe_reason(dev, &vaf); + dev_dbg(dev, "error %pe: %pV", ERR_PTR(err), &vaf); + break; + + case -ENOMEM: + /* Don't print anything on -ENOMEM, there's already enough output */ + break; + + default: + /* Log fatal final failures as errors, otherwise produce warnings */ + if (fatal) + dev_err(dev, "error %pe: %pV", ERR_PTR(err), &vaf); + else + dev_warn(dev, "error %pe: %pV", ERR_PTR(err), &vaf); + break; + } + + va_end(vargs); +} + /** * dev_err_probe - probe error check and log helper * @dev: the pointer to the struct device @@ -4635,7 +5039,7 @@ define_dev_printk_level(_dev_info, KERN_INFO); * -EPROBE_DEFER and propagate error upwards. * In case of -EPROBE_DEFER it sets also defer probe reason, which can be * checked later by reading devices_deferred debugfs attribute. - * It replaces code sequence:: + * It replaces the following code sequence:: * * if (err != -EPROBE_DEFER) * dev_err(dev, ...); @@ -4647,31 +5051,78 @@ define_dev_printk_level(_dev_info, KERN_INFO); * * return dev_err_probe(dev, err, ...); * - * Returns @err. + * Using this helper in your probe function is totally fine even if @err + * is known to never be -EPROBE_DEFER. + * The benefit compared to a normal dev_err() is the standardized format + * of the error code, which is emitted symbolically (i.e. you get "EAGAIN" + * instead of "-35"), and having the error code returned allows more + * compact error paths. * + * Returns @err. */ int dev_err_probe(const struct device *dev, int err, const char *fmt, ...) { - struct va_format vaf; - va_list args; + va_list vargs; - va_start(args, fmt); - vaf.fmt = fmt; - vaf.va = &args; + va_start(vargs, fmt); - if (err != -EPROBE_DEFER) { - dev_err(dev, "error %pe: %pV", ERR_PTR(err), &vaf); - } else { - device_set_deferred_probe_reason(dev, &vaf); - dev_dbg(dev, "error %pe: %pV", ERR_PTR(err), &vaf); - } + /* Use dev_err() for logging when err doesn't equal -EPROBE_DEFER */ + __dev_probe_failed(dev, err, true, fmt, vargs); - va_end(args); + va_end(vargs); return err; } EXPORT_SYMBOL_GPL(dev_err_probe); +/** + * dev_warn_probe - probe error check and log helper + * @dev: the pointer to the struct device + * @err: error value to test + * @fmt: printf-style format string + * @...: arguments as specified in the format string + * + * This helper implements common pattern present in probe functions for error + * checking: print debug or warning message depending if the error value is + * -EPROBE_DEFER and propagate error upwards. + * In case of -EPROBE_DEFER it sets also defer probe reason, which can be + * checked later by reading devices_deferred debugfs attribute. + * It replaces the following code sequence:: + * + * if (err != -EPROBE_DEFER) + * dev_warn(dev, ...); + * else + * dev_dbg(dev, ...); + * return err; + * + * with:: + * + * return dev_warn_probe(dev, err, ...); + * + * Using this helper in your probe function is totally fine even if @err + * is known to never be -EPROBE_DEFER. + * The benefit compared to a normal dev_warn() is the standardized format + * of the error code, which is emitted symbolically (i.e. you get "EAGAIN" + * instead of "-35"), and having the error code returned allows more + * compact error paths. + * + * Returns @err. + */ +int dev_warn_probe(const struct device *dev, int err, const char *fmt, ...) +{ + va_list vargs; + + va_start(vargs, fmt); + + /* Use dev_warn() for logging when err doesn't equal -EPROBE_DEFER */ + __dev_probe_failed(dev, err, false, fmt, vargs); + + va_end(vargs); + + return err; +} +EXPORT_SYMBOL_GPL(dev_warn_probe); + static inline bool fwnode_is_primary(struct fwnode_handle *fwnode) { return fwnode && !IS_ERR(fwnode->secondary); @@ -4708,9 +5159,13 @@ void set_primary_fwnode(struct device *dev, struct fwnode_handle *fwnode) } else { if (fwnode_is_primary(fn)) { dev->fwnode = fn->secondary; + + /* Skip nullifying fn->secondary if the primary is shared */ + if (parent && fn == parent->fwnode) + return; + /* Set fn->secondary = NULL, so fn remains the primary fwnode */ - if (!(parent && fn == parent->fwnode)) - fn->secondary = NULL; + fn->secondary = NULL; } else { dev->fwnode = NULL; } @@ -4740,6 +5195,67 @@ void set_secondary_fwnode(struct device *dev, struct fwnode_handle *fwnode) EXPORT_SYMBOL_GPL(set_secondary_fwnode); /** + * device_remove_of_node - Remove an of_node from a device + * @dev: device whose device tree node is being removed + */ +void device_remove_of_node(struct device *dev) +{ + dev = get_device(dev); + if (!dev) + return; + + if (!dev->of_node) + goto end; + + if (dev->fwnode == of_fwnode_handle(dev->of_node)) + dev->fwnode = NULL; + + of_node_put(dev->of_node); + dev->of_node = NULL; + +end: + put_device(dev); +} +EXPORT_SYMBOL_GPL(device_remove_of_node); + +/** + * device_add_of_node - Add an of_node to an existing device + * @dev: device whose device tree node is being added + * @of_node: of_node to add + * + * Return: 0 on success or error code on failure. + */ +int device_add_of_node(struct device *dev, struct device_node *of_node) +{ + int ret; + + if (!of_node) + return -EINVAL; + + dev = get_device(dev); + if (!dev) + return -EINVAL; + + if (dev->of_node) { + dev_err(dev, "Cannot replace node %pOF with %pOF\n", + dev->of_node, of_node); + ret = -EBUSY; + goto end; + } + + dev->of_node = of_node_get(of_node); + + if (!dev->fwnode) + dev->fwnode = of_fwnode_handle(of_node); + + ret = 0; +end: + put_device(dev); + return ret; +} +EXPORT_SYMBOL_GPL(device_add_of_node); + +/** * device_set_of_node_from_dev - reuse device-tree node of another device * @dev: device whose device-tree node is being set * @dev2: device whose device-tree node is being reused @@ -4762,21 +5278,52 @@ void device_set_node(struct device *dev, struct fwnode_handle *fwnode) } EXPORT_SYMBOL_GPL(device_set_node); +/** + * get_dev_from_fwnode - Obtain a reference count of the struct device the + * struct fwnode_handle is associated with. + * @fwnode: The pointer to the struct fwnode_handle to obtain the struct device + * reference count of. + * + * This function obtains a reference count of the device the device pointer + * embedded in the struct fwnode_handle points to. + * + * Note that the struct device pointer embedded in struct fwnode_handle does + * *not* have a reference count of the struct device itself. + * + * Hence, it is a UAF (and thus a bug) to call this function if the caller can't + * guarantee that the last reference count of the corresponding struct device is + * not dropped concurrently. + * + * This is possible since struct fwnode_handle has its own reference count and + * hence can out-live the struct device it is associated with. + */ +struct device *get_dev_from_fwnode(struct fwnode_handle *fwnode) +{ + return get_device((fwnode)->dev); +} +EXPORT_SYMBOL_GPL(get_dev_from_fwnode); + int device_match_name(struct device *dev, const void *name) { return sysfs_streq(dev_name(dev), name); } EXPORT_SYMBOL_GPL(device_match_name); +int device_match_type(struct device *dev, const void *type) +{ + return dev->type == type; +} +EXPORT_SYMBOL_GPL(device_match_type); + int device_match_of_node(struct device *dev, const void *np) { - return dev->of_node == np; + return np && dev->of_node == np; } EXPORT_SYMBOL_GPL(device_match_of_node); int device_match_fwnode(struct device *dev, const void *fwnode) { - return dev_fwnode(dev) == fwnode; + return fwnode && dev_fwnode(dev) == fwnode; } EXPORT_SYMBOL_GPL(device_match_fwnode); @@ -4788,10 +5335,16 @@ EXPORT_SYMBOL_GPL(device_match_devt); int device_match_acpi_dev(struct device *dev, const void *adev) { - return ACPI_COMPANION(dev) == adev; + return adev && ACPI_COMPANION(dev) == adev; } EXPORT_SYMBOL(device_match_acpi_dev); +int device_match_acpi_handle(struct device *dev, const void *handle) +{ + return handle && ACPI_HANDLE(dev) == handle; +} +EXPORT_SYMBOL(device_match_acpi_handle); + int device_match_any(struct device *dev, const void *unused) { return 1; diff --git a/drivers/base/cpu.c b/drivers/base/cpu.c index 5ef14db97904..c6c57b6f61c6 100644 --- a/drivers/base/cpu.c +++ b/drivers/base/cpu.c @@ -19,13 +19,14 @@ #include <linux/cpufeature.h> #include <linux/tick.h> #include <linux/pm_qos.h> +#include <linux/delay.h> #include <linux/sched/isolation.h> #include "base.h" static DEFINE_PER_CPU(struct device *, cpu_sys_devices); -static int cpu_subsys_match(struct device *dev, struct device_driver *drv) +static int cpu_subsys_match(struct device *dev, const struct device_driver *drv) { /* ACPI style match is the only one that may succeed. */ if (acpi_driver_match_device(dev, drv)) @@ -50,12 +51,30 @@ static int cpu_subsys_online(struct device *dev) int cpuid = dev->id; int from_nid, to_nid; int ret; + int retries = 0; from_nid = cpu_to_node(cpuid); if (from_nid == NUMA_NO_NODE) return -ENODEV; +retry: ret = cpu_device_up(dev); + + /* + * If -EBUSY is returned, it is likely that hotplug is temporarily + * disabled when cpu_hotplug_disable() was called. This condition is + * transient. So we retry after waiting for an exponentially + * increasing delay up to a total of at least 620ms as some PCI + * device initialization can take quite a while. + */ + if (ret == -EBUSY) { + retries++; + if (retries > 5) + return ret; + msleep(10 * (1 << retries)); + goto retry; + } + /* * When hot adding memory to memoryless node and enabling a cpu * on the node, node number of the cpu may internally change. @@ -76,6 +95,7 @@ void unregister_cpu(struct cpu *cpu) { int logical_cpu = cpu->dev.id; + set_cpu_enabled(logical_cpu, false); unregister_cpu_under_node(logical_cpu, cpu_to_node(logical_cpu)); device_unregister(&cpu->dev); @@ -125,18 +145,7 @@ static DEVICE_ATTR(release, S_IWUSR, NULL, cpu_release_store); #endif /* CONFIG_ARCH_CPU_PROBE_RELEASE */ #endif /* CONFIG_HOTPLUG_CPU */ -struct bus_type cpu_subsys = { - .name = "cpu", - .dev_name = "cpu", - .match = cpu_subsys_match, -#ifdef CONFIG_HOTPLUG_CPU - .online = cpu_subsys_online, - .offline = cpu_subsys_offline, -#endif -}; -EXPORT_SYMBOL_GPL(cpu_subsys); - -#ifdef CONFIG_KEXEC +#ifdef CONFIG_CRASH_DUMP #include <linux/kexec.h> static ssize_t crash_notes_show(struct device *dev, @@ -181,14 +190,14 @@ static const struct attribute_group crash_note_cpu_attr_group = { #endif static const struct attribute_group *common_cpu_attr_groups[] = { -#ifdef CONFIG_KEXEC +#ifdef CONFIG_CRASH_DUMP &crash_note_cpu_attr_group, #endif NULL }; static const struct attribute_group *hotplugable_cpu_attr_groups[] = { -#ifdef CONFIG_KEXEC +#ifdef CONFIG_CRASH_DUMP &crash_note_cpu_attr_group, #endif NULL @@ -265,6 +274,13 @@ static ssize_t print_cpus_offline(struct device *dev, } static DEVICE_ATTR(offline, 0444, print_cpus_offline, NULL); +static ssize_t print_cpus_enabled(struct device *dev, + struct device_attribute *attr, char *buf) +{ + return sysfs_emit(buf, "%*pbl\n", cpumask_pr_args(cpu_enabled_mask)); +} +static DEVICE_ATTR(enabled, 0444, print_cpus_enabled, NULL); + static ssize_t print_cpus_isolated(struct device *dev, struct device_attribute *attr, char *buf) { @@ -275,7 +291,7 @@ static ssize_t print_cpus_isolated(struct device *dev, return -ENOMEM; cpumask_andnot(isolated, cpu_possible_mask, - housekeeping_cpumask(HK_FLAG_DOMAIN)); + housekeeping_cpumask(HK_TYPE_DOMAIN)); len = sysfs_emit(buf, "%*pbl\n", cpumask_pr_args(isolated)); free_cpumask_var(isolated); @@ -284,13 +300,40 @@ static ssize_t print_cpus_isolated(struct device *dev, } static DEVICE_ATTR(isolated, 0444, print_cpus_isolated, NULL); +static ssize_t housekeeping_show(struct device *dev, + struct device_attribute *attr, char *buf) +{ + const struct cpumask *hk_mask; + + hk_mask = housekeeping_cpumask(HK_TYPE_KERNEL_NOISE); + + if (housekeeping_enabled(HK_TYPE_KERNEL_NOISE)) + return sysfs_emit(buf, "%*pbl\n", cpumask_pr_args(hk_mask)); + return sysfs_emit(buf, "\n"); +} +static DEVICE_ATTR_RO(housekeeping); + #ifdef CONFIG_NO_HZ_FULL -static ssize_t print_cpus_nohz_full(struct device *dev, - struct device_attribute *attr, char *buf) +static ssize_t nohz_full_show(struct device *dev, + struct device_attribute *attr, + char *buf) +{ + if (cpumask_available(tick_nohz_full_mask)) + return sysfs_emit(buf, "%*pbl\n", + cpumask_pr_args(tick_nohz_full_mask)); + return sysfs_emit(buf, "\n"); +} +static DEVICE_ATTR_RO(nohz_full); +#endif + +#ifdef CONFIG_CRASH_HOTPLUG +static ssize_t crash_hotplug_show(struct device *dev, + struct device_attribute *attr, + char *buf) { - return sysfs_emit(buf, "%*pbl\n", cpumask_pr_args(tick_nohz_full_mask)); + return sysfs_emit(buf, "%d\n", crash_check_hotplug_support()); } -static DEVICE_ATTR(nohz_full, 0444, print_cpus_nohz_full, NULL); +static DEVICE_ATTR_RO(crash_hotplug); #endif static void cpu_device_release(struct device *dev) @@ -299,7 +342,7 @@ static void cpu_device_release(struct device *dev) * This is an empty function to prevent the driver core from spitting a * warning at us. Yes, I know this is directly opposite of what the * documentation for the driver core and kobjects say, and the author - * of this code has already been publically ridiculed for doing + * of this code has already been publicly ridiculed for doing * something as foolish as this. However, at this point in time, it is * the only way to handle the issue of statically allocated cpu * devices. The different architectures will have their cpu device @@ -336,7 +379,7 @@ static ssize_t print_cpu_modalias(struct device *dev, return len; } -static int cpu_uevent(struct device *dev, struct kobj_uevent_env *env) +static int cpu_uevent(const struct device *dev, struct kobj_uevent_env *env) { char *buf = kzalloc(PAGE_SIZE, GFP_KERNEL); if (buf) { @@ -348,6 +391,20 @@ static int cpu_uevent(struct device *dev, struct kobj_uevent_env *env) } #endif +const struct bus_type cpu_subsys = { + .name = "cpu", + .dev_name = "cpu", + .match = cpu_subsys_match, +#ifdef CONFIG_HOTPLUG_CPU + .online = cpu_subsys_online, + .offline = cpu_subsys_offline, +#endif +#ifdef CONFIG_GENERIC_CPU_AUTOPROBE + .uevent = cpu_uevent, +#endif +}; +EXPORT_SYMBOL_GPL(cpu_subsys); + /* * register_cpu - Setup a sysfs device for a CPU. * @cpu - cpu->hotpluggable field set to 1 will generate a control file in @@ -368,9 +425,6 @@ int register_cpu(struct cpu *cpu, int num) cpu->dev.offline_disabled = !cpu->hotpluggable; cpu->dev.offline = !cpu_online(num); cpu->dev.of_node = of_get_cpu_node(num, NULL); -#ifdef CONFIG_GENERIC_CPU_AUTOPROBE - cpu->dev.bus->uevent = cpu_uevent; -#endif cpu->dev.groups = common_cpu_attr_groups; if (cpu->hotpluggable) cpu->dev.groups = hotplugable_cpu_attr_groups; @@ -384,11 +438,12 @@ int register_cpu(struct cpu *cpu, int num) register_cpu_under_node(num, cpu_to_node(num)); dev_pm_qos_expose_latency_limit(&cpu->dev, PM_QOS_RESUME_LATENCY_NO_CONSTRAINT); + set_cpu_enabled(num, true); return 0; } -struct device *get_cpu_device(unsigned cpu) +struct device *get_cpu_device(unsigned int cpu) { if (cpu < nr_cpu_ids && cpu_possible(cpu)) return per_cpu(cpu_sys_devices, cpu); @@ -465,10 +520,15 @@ static struct attribute *cpu_root_attrs[] = { &cpu_attrs[2].attr.attr, &dev_attr_kernel_max.attr, &dev_attr_offline.attr, + &dev_attr_enabled.attr, &dev_attr_isolated.attr, + &dev_attr_housekeeping.attr, #ifdef CONFIG_NO_HZ_FULL &dev_attr_nohz_full.attr, #endif +#ifdef CONFIG_CRASH_HOTPLUG + &dev_attr_crash_hotplug.attr, +#endif #ifdef CONFIG_GENERIC_CPU_AUTOPROBE &dev_attr_modalias.attr, #endif @@ -484,86 +544,85 @@ static const struct attribute_group *cpu_root_attr_groups[] = { NULL, }; -bool cpu_is_hotpluggable(unsigned cpu) +bool cpu_is_hotpluggable(unsigned int cpu) { struct device *dev = get_cpu_device(cpu); - return dev && container_of(dev, struct cpu, dev)->hotpluggable; + return dev && container_of(dev, struct cpu, dev)->hotpluggable + && tick_nohz_cpu_hotpluggable(cpu); } EXPORT_SYMBOL_GPL(cpu_is_hotpluggable); #ifdef CONFIG_GENERIC_CPU_DEVICES -static DEFINE_PER_CPU(struct cpu, cpu_devices); -#endif +DEFINE_PER_CPU(struct cpu, cpu_devices); -static void __init cpu_dev_register_generic(void) +bool __weak arch_cpu_is_hotpluggable(int cpu) { -#ifdef CONFIG_GENERIC_CPU_DEVICES - int i; - - for_each_possible_cpu(i) { - if (register_cpu(&per_cpu(cpu_devices, i), i)) - panic("Failed to register CPU device"); - } -#endif + return false; } -#ifdef CONFIG_GENERIC_CPU_VULNERABILITIES - -ssize_t __weak cpu_show_meltdown(struct device *dev, - struct device_attribute *attr, char *buf) +int __weak arch_register_cpu(int cpu) { - return sysfs_emit(buf, "Not affected\n"); -} + struct cpu *c = &per_cpu(cpu_devices, cpu); -ssize_t __weak cpu_show_spectre_v1(struct device *dev, - struct device_attribute *attr, char *buf) -{ - return sysfs_emit(buf, "Not affected\n"); -} + c->hotpluggable = arch_cpu_is_hotpluggable(cpu); -ssize_t __weak cpu_show_spectre_v2(struct device *dev, - struct device_attribute *attr, char *buf) -{ - return sysfs_emit(buf, "Not affected\n"); + return register_cpu(c, cpu); } -ssize_t __weak cpu_show_spec_store_bypass(struct device *dev, - struct device_attribute *attr, char *buf) -{ - return sysfs_emit(buf, "Not affected\n"); -} - -ssize_t __weak cpu_show_l1tf(struct device *dev, - struct device_attribute *attr, char *buf) +#ifdef CONFIG_HOTPLUG_CPU +void __weak arch_unregister_cpu(int num) { - return sysfs_emit(buf, "Not affected\n"); + unregister_cpu(&per_cpu(cpu_devices, num)); } +#endif /* CONFIG_HOTPLUG_CPU */ +#endif /* CONFIG_GENERIC_CPU_DEVICES */ -ssize_t __weak cpu_show_mds(struct device *dev, - struct device_attribute *attr, char *buf) +static void __init cpu_dev_register_generic(void) { - return sysfs_emit(buf, "Not affected\n"); -} + int i, ret; -ssize_t __weak cpu_show_tsx_async_abort(struct device *dev, - struct device_attribute *attr, - char *buf) -{ - return sysfs_emit(buf, "Not affected\n"); -} + if (!IS_ENABLED(CONFIG_GENERIC_CPU_DEVICES)) + return; -ssize_t __weak cpu_show_itlb_multihit(struct device *dev, - struct device_attribute *attr, char *buf) -{ - return sysfs_emit(buf, "Not affected\n"); + for_each_present_cpu(i) { + ret = arch_register_cpu(i); + if (ret && ret != -EPROBE_DEFER) + pr_warn("register_cpu %d failed (%d)\n", i, ret); + } } -ssize_t __weak cpu_show_srbds(struct device *dev, +#ifdef CONFIG_GENERIC_CPU_VULNERABILITIES +static ssize_t cpu_show_not_affected(struct device *dev, struct device_attribute *attr, char *buf) { return sysfs_emit(buf, "Not affected\n"); } +#define CPU_SHOW_VULN_FALLBACK(func) \ + ssize_t cpu_show_##func(struct device *, \ + struct device_attribute *, char *) \ + __attribute__((weak, alias("cpu_show_not_affected"))) + +CPU_SHOW_VULN_FALLBACK(meltdown); +CPU_SHOW_VULN_FALLBACK(spectre_v1); +CPU_SHOW_VULN_FALLBACK(spectre_v2); +CPU_SHOW_VULN_FALLBACK(spec_store_bypass); +CPU_SHOW_VULN_FALLBACK(l1tf); +CPU_SHOW_VULN_FALLBACK(mds); +CPU_SHOW_VULN_FALLBACK(tsx_async_abort); +CPU_SHOW_VULN_FALLBACK(itlb_multihit); +CPU_SHOW_VULN_FALLBACK(srbds); +CPU_SHOW_VULN_FALLBACK(mmio_stale_data); +CPU_SHOW_VULN_FALLBACK(retbleed); +CPU_SHOW_VULN_FALLBACK(spec_rstack_overflow); +CPU_SHOW_VULN_FALLBACK(gds); +CPU_SHOW_VULN_FALLBACK(reg_file_data_sampling); +CPU_SHOW_VULN_FALLBACK(ghostwrite); +CPU_SHOW_VULN_FALLBACK(old_microcode); +CPU_SHOW_VULN_FALLBACK(indirect_target_selection); +CPU_SHOW_VULN_FALLBACK(tsa); +CPU_SHOW_VULN_FALLBACK(vmscape); + static DEVICE_ATTR(meltdown, 0444, cpu_show_meltdown, NULL); static DEVICE_ATTR(spectre_v1, 0444, cpu_show_spectre_v1, NULL); static DEVICE_ATTR(spectre_v2, 0444, cpu_show_spectre_v2, NULL); @@ -573,6 +632,16 @@ static DEVICE_ATTR(mds, 0444, cpu_show_mds, NULL); static DEVICE_ATTR(tsx_async_abort, 0444, cpu_show_tsx_async_abort, NULL); static DEVICE_ATTR(itlb_multihit, 0444, cpu_show_itlb_multihit, NULL); static DEVICE_ATTR(srbds, 0444, cpu_show_srbds, NULL); +static DEVICE_ATTR(mmio_stale_data, 0444, cpu_show_mmio_stale_data, NULL); +static DEVICE_ATTR(retbleed, 0444, cpu_show_retbleed, NULL); +static DEVICE_ATTR(spec_rstack_overflow, 0444, cpu_show_spec_rstack_overflow, NULL); +static DEVICE_ATTR(gather_data_sampling, 0444, cpu_show_gds, NULL); +static DEVICE_ATTR(reg_file_data_sampling, 0444, cpu_show_reg_file_data_sampling, NULL); +static DEVICE_ATTR(ghostwrite, 0444, cpu_show_ghostwrite, NULL); +static DEVICE_ATTR(old_microcode, 0444, cpu_show_old_microcode, NULL); +static DEVICE_ATTR(indirect_target_selection, 0444, cpu_show_indirect_target_selection, NULL); +static DEVICE_ATTR(tsa, 0444, cpu_show_tsa, NULL); +static DEVICE_ATTR(vmscape, 0444, cpu_show_vmscape, NULL); static struct attribute *cpu_root_vulnerabilities_attrs[] = { &dev_attr_meltdown.attr, @@ -584,6 +653,16 @@ static struct attribute *cpu_root_vulnerabilities_attrs[] = { &dev_attr_tsx_async_abort.attr, &dev_attr_itlb_multihit.attr, &dev_attr_srbds.attr, + &dev_attr_mmio_stale_data.attr, + &dev_attr_retbleed.attr, + &dev_attr_spec_rstack_overflow.attr, + &dev_attr_gather_data_sampling.attr, + &dev_attr_reg_file_data_sampling.attr, + &dev_attr_ghostwrite.attr, + &dev_attr_old_microcode.attr, + &dev_attr_indirect_target_selection.attr, + &dev_attr_tsa.attr, + &dev_attr_vmscape.attr, NULL }; @@ -594,9 +673,13 @@ static const struct attribute_group cpu_root_vulnerabilities_group = { static void __init cpu_register_vulnerabilities(void) { - if (sysfs_create_group(&cpu_subsys.dev_root->kobj, - &cpu_root_vulnerabilities_group)) - pr_err("Unable to register CPU vulnerabilities\n"); + struct device *dev = bus_get_dev_root(&cpu_subsys); + + if (dev) { + if (sysfs_create_group(&dev->kobj, &cpu_root_vulnerabilities_group)) + pr_err("Unable to register CPU vulnerabilities\n"); + put_device(dev); + } } #else diff --git a/drivers/base/dd.c b/drivers/base/dd.c index daeb9b5763ae..349f31bedfa1 100644 --- a/drivers/base/dd.c +++ b/drivers/base/dd.c @@ -25,6 +25,7 @@ #include <linux/kthread.h> #include <linux/wait.h> #include <linux/async.h> +#include <linux/pm_domain.h> #include <linux/pm_runtime.h> #include <linux/pinctrl/devinfo.h> #include <linux/slab.h> @@ -60,6 +61,7 @@ static bool initcalls_done; /* Save the async probe drivers' name from kernel cmdline */ #define ASYNC_DRV_NAMES_MAX_LEN 256 static char async_probe_drv_names[ASYNC_DRV_NAMES_MAX_LEN]; +static bool async_probe_default; /* * In some cases, like suspend to RAM or hibernation, It might be reasonable @@ -152,7 +154,7 @@ void driver_deferred_probe_del(struct device *dev) mutex_unlock(&deferred_probe_mutex); } -static bool driver_deferred_probe_enable = false; +static bool driver_deferred_probe_enable; /** * driver_deferred_probe_trigger() - Kick off re-probing deferred devices * @@ -171,7 +173,7 @@ static bool driver_deferred_probe_enable = false; * changes in the midst of a probe, then deferred processing should be triggered * again. */ -static void driver_deferred_probe_trigger(void) +void driver_deferred_probe_trigger(void) { if (!driver_deferred_probe_enable) return; @@ -191,7 +193,7 @@ static void driver_deferred_probe_trigger(void) * Kick the re-probe thread. It may already be scheduled, but it is * safe to kick it again. */ - queue_work(system_unbound_wq, &deferred_probe_work); + queue_work(system_dfl_wq, &deferred_probe_work); } /** @@ -247,7 +249,7 @@ static int deferred_devs_show(struct seq_file *s, void *data) list_for_each_entry(curr, &deferred_probe_pending_list, deferred_probe) seq_printf(s, "%s\t%s", dev_name(curr->device), - curr->device->p->deferred_probe_reason ?: "\n"); + curr->deferred_probe_reason ?: "\n"); mutex_unlock(&deferred_probe_mutex); @@ -255,9 +257,11 @@ static int deferred_devs_show(struct seq_file *s, void *data) } DEFINE_SHOW_ATTRIBUTE(deferred_devs); -int driver_deferred_probe_timeout; -EXPORT_SYMBOL_GPL(driver_deferred_probe_timeout); -static DECLARE_WAIT_QUEUE_HEAD(probe_timeout_waitqueue); +#ifdef CONFIG_MODULES +static int driver_deferred_probe_timeout = 10; +#else +static int driver_deferred_probe_timeout; +#endif static int __init deferred_probe_timeout_setup(char *str) { @@ -274,10 +278,10 @@ __setup("deferred_probe_timeout=", deferred_probe_timeout_setup); * @dev: device to check * * Return: - * -ENODEV if initcalls have completed and modules are disabled. - * -ETIMEDOUT if the deferred probe timeout was set and has expired - * and modules are enabled. - * -EPROBE_DEFER in other cases. + * * -ENODEV if initcalls have completed and modules are disabled. + * * -ETIMEDOUT if the deferred probe timeout was set and has expired + * and modules are enabled. + * * -EPROBE_DEFER in other cases. * * Drivers or subsystems can opt-in to calling this function instead of directly * returning -EPROBE_DEFER. @@ -296,6 +300,7 @@ int driver_deferred_probe_check_state(struct device *dev) return -EPROBE_DEFER; } +EXPORT_SYMBOL_GPL(driver_deferred_probe_check_state); static void deferred_probe_timeout_work_func(struct work_struct *work) { @@ -309,12 +314,27 @@ static void deferred_probe_timeout_work_func(struct work_struct *work) mutex_lock(&deferred_probe_mutex); list_for_each_entry(p, &deferred_probe_pending_list, deferred_probe) - dev_info(p->device, "deferred probe pending\n"); + dev_warn(p->device, "deferred probe pending: %s", p->deferred_probe_reason ?: "(reason unknown)\n"); mutex_unlock(&deferred_probe_mutex); - wake_up_all(&probe_timeout_waitqueue); + + fw_devlink_probing_done(); } static DECLARE_DELAYED_WORK(deferred_probe_timeout_work, deferred_probe_timeout_work_func); +void deferred_probe_extend_timeout(void) +{ + /* + * If the work hasn't been queued yet or if the work expired, don't + * start a new one. + */ + if (cancel_delayed_work(&deferred_probe_timeout_work)) { + schedule_delayed_work(&deferred_probe_timeout_work, + driver_deferred_probe_timeout * HZ); + pr_debug("Extended deferred probe timeout by %d secs\n", + driver_deferred_probe_timeout); + } +} + /** * deferred_probe_initcall() - Enable probing of deferred devices * @@ -347,13 +367,17 @@ static int deferred_probe_initcall(void) schedule_delayed_work(&deferred_probe_timeout_work, driver_deferred_probe_timeout * HZ); } + + if (!IS_ENABLED(CONFIG_MODULES)) + fw_devlink_probing_done(); + return 0; } late_initcall(deferred_probe_initcall); static void __exit deferred_probe_exit(void) { - debugfs_remove_recursive(debugfs_lookup("devices_deferred", NULL)); + debugfs_lookup_and_remove("devices_deferred", NULL); } __exitcall(deferred_probe_exit); @@ -370,17 +394,17 @@ bool device_is_bound(struct device *dev) { return dev->p && klist_node_attached(&dev->p->knode_driver); } +EXPORT_SYMBOL_GPL(device_is_bound); static void driver_bound(struct device *dev) { if (device_is_bound(dev)) { - pr_warn("%s: device %s already bound\n", - __func__, kobject_name(&dev->kobj)); + dev_warn(dev, "%s: device already bound\n", __func__); return; } - pr_debug("driver: '%s': %s: bound to device '%s'\n", dev->driver->name, - __func__, dev_name(dev)); + dev_dbg(dev, "driver: '%s': %s: bound to device\n", dev->driver->name, + __func__); klist_add_tail(&dev->p->knode_driver, &dev->driver->p->klist_devices); device_links_driver_bound(dev); @@ -394,10 +418,7 @@ static void driver_bound(struct device *dev) driver_deferred_probe_del(dev); driver_deferred_probe_trigger(); - if (dev->bus) - blocking_notifier_call_chain(&dev->bus->p->bus_notifier, - BUS_NOTIFY_BOUND_DRIVER, dev); - + bus_notify(dev, BUS_NOTIFY_BOUND_DRIVER); kobject_uevent(&dev->kobj, KOBJ_BIND); } @@ -416,9 +437,7 @@ static int driver_sysfs_add(struct device *dev) { int ret; - if (dev->bus) - blocking_notifier_call_chain(&dev->bus->p->bus_notifier, - BUS_NOTIFY_BIND_DRIVER, dev); + bus_notify(dev, BUS_NOTIFY_BIND_DRIVER); ret = sysfs_create_link(&dev->driver->p->kobj, &dev->kobj, kobject_name(&dev->kobj)); @@ -483,9 +502,8 @@ int device_bind_driver(struct device *dev) device_links_force_bind(dev); driver_bound(dev); } - else if (dev->bus) - blocking_notifier_call_chain(&dev->bus->p->bus_notifier, - BUS_NOTIFY_DRIVER_NOT_BOUND, dev); + else + bus_notify(dev, BUS_NOTIFY_DRIVER_NOT_BOUND); return ret; } EXPORT_SYMBOL_GPL(device_bind_driver); @@ -493,6 +511,27 @@ EXPORT_SYMBOL_GPL(device_bind_driver); static atomic_t probe_count = ATOMIC_INIT(0); static DECLARE_WAIT_QUEUE_HEAD(probe_waitqueue); +static ssize_t state_synced_store(struct device *dev, + struct device_attribute *attr, + const char *buf, size_t count) +{ + int ret = 0; + + if (strcmp("1", buf)) + return -EINVAL; + + device_lock(dev); + if (!dev->state_synced) { + dev->state_synced = true; + dev_sync_state(dev); + } else { + ret = -EINVAL; + } + device_unlock(dev); + + return ret ? ret : count; +} + static ssize_t state_synced_show(struct device *dev, struct device_attribute *attr, char *buf) { @@ -504,10 +543,35 @@ static ssize_t state_synced_show(struct device *dev, return sysfs_emit(buf, "%u\n", val); } -static DEVICE_ATTR_RO(state_synced); +static DEVICE_ATTR_RW(state_synced); +static void device_unbind_cleanup(struct device *dev) +{ + devres_release_all(dev); + arch_teardown_dma_ops(dev); + kfree(dev->dma_range_map); + dev->dma_range_map = NULL; + device_set_driver(dev, NULL); + dev_set_drvdata(dev, NULL); + dev_pm_domain_detach(dev, dev->power.detach_power_off); + if (dev->pm_domain && dev->pm_domain->dismiss) + dev->pm_domain->dismiss(dev); + pm_runtime_reinit(dev); + dev_pm_set_driver_flags(dev, 0); +} -static int call_driver_probe(struct device *dev, struct device_driver *drv) +static void device_remove(struct device *dev) +{ + device_remove_file(dev, &dev_attr_state_synced); + device_remove_groups(dev, dev->driver->dev_groups); + + if (dev->bus && dev->bus->remove) + dev->bus->remove(dev); + else if (dev->driver->remove) + dev->driver->remove(dev); +} + +static int call_driver_probe(struct device *dev, const struct device_driver *drv) { int ret = 0; @@ -525,24 +589,24 @@ static int call_driver_probe(struct device *dev, struct device_driver *drv) break; case -ENODEV: case -ENXIO: - pr_debug("%s: probe of %s rejects match %d\n", - drv->name, dev_name(dev), ret); + dev_dbg(dev, "probe with driver %s rejects match %d\n", + drv->name, ret); break; default: /* driver matched but the probe failed */ - pr_warn("%s: probe of %s failed with error %d\n", - drv->name, dev_name(dev), ret); + dev_err(dev, "probe with driver %s failed with error %d\n", + drv->name, ret); break; } return ret; } -static int really_probe(struct device *dev, struct device_driver *drv) +static int really_probe(struct device *dev, const struct device_driver *drv) { bool test_remove = IS_ENABLED(CONFIG_DEBUG_TEST_DRIVER_REMOVE) && !drv->suppress_bind_attrs; - int ret; + int ret, link_ret; if (defer_all_probes) { /* @@ -554,12 +618,12 @@ static int really_probe(struct device *dev, struct device_driver *drv) return -EPROBE_DEFER; } - ret = device_links_check_suppliers(dev); - if (ret) - return ret; + link_ret = device_links_check_suppliers(dev); + if (link_ret == -EPROBE_DEFER) + return link_ret; - pr_debug("bus: '%s': %s: probing driver %s with device %s\n", - drv->bus->name, __func__, drv->name, dev_name(dev)); + dev_dbg(dev, "bus: '%s': %s: probing driver %s with device\n", + drv->bus->name, __func__, drv->name); if (!list_empty(&dev->devres_head)) { dev_crit(dev, "Resources present before probing\n"); ret = -EBUSY; @@ -567,7 +631,7 @@ static int really_probe(struct device *dev, struct device_driver *drv) } re_probe: - dev->driver = drv; + device_set_driver(dev, drv); /* If using pinctrl, bind pins now before probing */ ret = pinctrl_bind_pins(dev); @@ -577,13 +641,13 @@ re_probe: if (dev->bus->dma_configure) { ret = dev->bus->dma_configure(dev); if (ret) - goto probe_failed; + goto pinctrl_bind_failed; } - if (driver_sysfs_add(dev)) { - pr_err("%s: driver_sysfs_add(%s) failed\n", - __func__, dev_name(dev)); - goto probe_failed; + ret = driver_sysfs_add(dev); + if (ret) { + dev_err(dev, "%s: driver_sysfs_add failed\n", __func__); + goto sysfs_failed; } if (dev->pm_domain && dev->pm_domain->activate) { @@ -595,6 +659,15 @@ re_probe: ret = call_driver_probe(dev, drv); if (ret) { /* + * If fw_devlink_best_effort is active (denoted by -EAGAIN), the + * device might actually probe properly once some of its missing + * suppliers have probed. So, treat this as if the driver + * returned -EPROBE_DEFER. + */ + if (link_ret == -EAGAIN) + ret = -EPROBE_DEFER; + + /* * Return probe errors as positive values so that the callers * can distinguish them from other errors. */ @@ -602,35 +675,28 @@ re_probe: goto probe_failed; } - if (device_add_groups(dev, drv->dev_groups)) { + ret = device_add_groups(dev, drv->dev_groups); + if (ret) { dev_err(dev, "device_add_groups() failed\n"); goto dev_groups_failed; } - if (dev_has_sync_state(dev) && - device_create_file(dev, &dev_attr_state_synced)) { - dev_err(dev, "state_synced sysfs add failed\n"); - goto dev_sysfs_state_synced_failed; + if (dev_has_sync_state(dev)) { + ret = device_create_file(dev, &dev_attr_state_synced); + if (ret) { + dev_err(dev, "state_synced sysfs add failed\n"); + goto dev_sysfs_state_synced_failed; + } } if (test_remove) { test_remove = false; - device_remove_file(dev, &dev_attr_state_synced); - device_remove_groups(dev, drv->dev_groups); - - if (dev->bus->remove) - dev->bus->remove(dev); - else if (drv->remove) - drv->remove(dev); - - devres_release_all(dev); + device_remove(dev); driver_sysfs_remove(dev); - dev->driver = NULL; - dev_set_drvdata(dev, NULL); - if (dev->pm_domain && dev->pm_domain->dismiss) - dev->pm_domain->dismiss(dev); - pm_runtime_reinit(dev); + if (dev->bus && dev->bus->dma_cleanup) + dev->bus->dma_cleanup(dev); + device_unbind_cleanup(dev); goto re_probe; } @@ -641,34 +707,22 @@ re_probe: dev->pm_domain->sync(dev); driver_bound(dev); - pr_debug("bus: '%s': %s: bound device %s to driver %s\n", - drv->bus->name, __func__, dev_name(dev), drv->name); + dev_dbg(dev, "bus: '%s': %s: bound device to driver %s\n", + drv->bus->name, __func__, drv->name); goto done; dev_sysfs_state_synced_failed: - device_remove_groups(dev, drv->dev_groups); dev_groups_failed: - if (dev->bus->remove) - dev->bus->remove(dev); - else if (drv->remove) - drv->remove(dev); + device_remove(dev); probe_failed: - kfree(dev->dma_range_map); - dev->dma_range_map = NULL; - if (dev->bus) - blocking_notifier_call_chain(&dev->bus->p->bus_notifier, - BUS_NOTIFY_DRIVER_NOT_BOUND, dev); + driver_sysfs_remove(dev); +sysfs_failed: + bus_notify(dev, BUS_NOTIFY_DRIVER_NOT_BOUND); + if (dev->bus && dev->bus->dma_cleanup) + dev->bus->dma_cleanup(dev); pinctrl_bind_failed: device_links_no_driver(dev); - devres_release_all(dev); - arch_teardown_dma_ops(dev); - driver_sysfs_remove(dev); - dev->driver = NULL; - dev_set_drvdata(dev, NULL); - if (dev->pm_domain && dev->pm_domain->dismiss) - dev->pm_domain->dismiss(dev); - pm_runtime_reinit(dev); - dev_pm_set_driver_flags(dev, 0); + device_unbind_cleanup(dev); done: return ret; } @@ -676,7 +730,7 @@ done: /* * For initcall_debug, show the driver probe time. */ -static int really_probe_debug(struct device *dev, struct device_driver *drv) +static int really_probe_debug(struct device *dev, const struct device_driver *drv) { ktime_t calltime, rettime; int ret; @@ -684,7 +738,12 @@ static int really_probe_debug(struct device *dev, struct device_driver *drv) calltime = ktime_get(); ret = really_probe(dev, drv); rettime = ktime_get(); - pr_debug("probe of %s returned %d after %lld usecs\n", + /* + * Don't change this to pr_debug() because that requires + * CONFIG_DYNAMIC_DEBUG and we want a simple 'initcall_debug' on the + * kernel commandline to print this all the time at the debug level. + */ + printk(KERN_DEBUG "probe of %s returned %d after %lld usecs\n", dev_name(dev), ret, ktime_us_delta(rettime, calltime)); return ret; } @@ -695,14 +754,12 @@ static int really_probe_debug(struct device *dev, struct device_driver *drv) * * Should somehow figure out how to use a semaphore, not an atomic variable... */ -int driver_probe_done(void) +bool __init driver_probe_done(void) { int local_probe_count = atomic_read(&probe_count); pr_debug("%s: probe_count = %d\n", __func__, local_probe_count); - if (local_probe_count) - return -EBUSY; - return 0; + return !local_probe_count; } /** @@ -711,9 +768,6 @@ int driver_probe_done(void) */ void wait_for_device_probe(void) { - /* wait for probe timeout */ - wait_event(probe_timeout_waitqueue, !driver_deferred_probe_timeout); - /* wait for the deferred probe workqueue to finish */ flush_work(&deferred_probe_work); @@ -723,7 +777,7 @@ void wait_for_device_probe(void) } EXPORT_SYMBOL_GPL(wait_for_device_probe); -static int __driver_probe_device(struct device_driver *drv, struct device *dev) +static int __driver_probe_device(const struct device_driver *drv, struct device *dev) { int ret = 0; @@ -733,8 +787,8 @@ static int __driver_probe_device(struct device_driver *drv, struct device *dev) return -EBUSY; dev->can_match = true; - pr_debug("bus: '%s': %s: matched device %s with driver %s\n", - drv->bus->name, __func__, dev_name(dev), drv->name); + dev_dbg(dev, "bus: '%s': %s: matched device with driver %s\n", + drv->bus->name, __func__, drv->name); pm_runtime_get_suppliers(dev); if (dev->parent) @@ -768,7 +822,7 @@ static int __driver_probe_device(struct device_driver *drv, struct device *dev) * * If the device has a parent, runtime-resume the parent before driver probing. */ -static int driver_probe_device(struct device_driver *drv, struct device *dev) +static int driver_probe_device(const struct device_driver *drv, struct device *dev) { int trigger_count = atomic_read(&deferred_trigger_count); int ret; @@ -792,7 +846,11 @@ static int driver_probe_device(struct device_driver *drv, struct device *dev) static inline bool cmdline_requested_async_probing(const char *drv_name) { - return parse_option_str(async_probe_drv_names, drv_name); + bool async_drv; + + async_drv = parse_option_str(async_probe_drv_names, drv_name); + + return (async_probe_default != async_drv); } /* The option format is "driver_async_probe=drv_name1,drv_name2,..." */ @@ -801,12 +859,14 @@ static int __init save_async_options(char *buf) if (strlen(buf) >= ASYNC_DRV_NAMES_MAX_LEN) pr_warn("Too long list of driver names for 'driver_async_probe'!\n"); - strlcpy(async_probe_drv_names, buf, ASYNC_DRV_NAMES_MAX_LEN); - return 0; + strscpy(async_probe_drv_names, buf, ASYNC_DRV_NAMES_MAX_LEN); + async_probe_default = parse_option_str(async_probe_drv_names, "*"); + + return 1; } __setup("driver_async_probe=", save_async_options); -bool driver_allows_async_probing(struct device_driver *drv) +static bool driver_allows_async_probing(const struct device_driver *drv) { switch (drv->probe_type) { case PROBE_PREFER_ASYNCHRONOUS: @@ -830,7 +890,7 @@ struct device_attach_data { struct device *dev; /* - * Indicates whether we are are considering asynchronous probing or + * Indicates whether we are considering asynchronous probing or * not. Only initial binding after device or driver registration * (including deferral processing) may be done asynchronously, the * rest is always synchronous, as we expect it is being done by @@ -874,6 +934,11 @@ static int __device_attach_driver(struct device_driver *drv, void *_data) dev_dbg(dev, "Device match requests probe deferral\n"); dev->can_match = true; driver_deferred_probe_add(dev); + /* + * Device can't match with a driver right now, so don't attempt + * to match or bind with other drivers on the bus. + */ + return ret; } else if (ret < 0) { dev_dbg(dev, "Bus failed to match device: %d\n", ret); return ret; @@ -936,6 +1001,7 @@ out_unlock: static int __device_attach(struct device *dev, bool allow_async) { int ret = 0; + bool async = false; device_lock(dev); if (dev->p->dead) { @@ -949,7 +1015,7 @@ static int __device_attach(struct device *dev, bool allow_async) if (ret == 0) ret = 1; else { - dev->driver = NULL; + device_set_driver(dev, NULL); ret = 0; } } else { @@ -974,7 +1040,7 @@ static int __device_attach(struct device *dev, bool allow_async) */ dev_dbg(dev, "scheduling asynchronous probe\n"); get_device(dev); - async_schedule_dev(__device_attach_async_helper, dev); + async = true; } else { pm_request_idle(dev); } @@ -984,6 +1050,8 @@ static int __device_attach(struct device *dev, bool allow_async) } out_unlock: device_unlock(dev); + if (async) + async_schedule_dev(__device_attach_async_helper, dev); return ret; } @@ -1009,7 +1077,15 @@ EXPORT_SYMBOL_GPL(device_attach); void device_initial_probe(struct device *dev) { - __device_attach(dev, true); + struct subsys_private *sp = bus_to_subsys(dev->bus); + + if (!sp) + return; + + if (sp->drivers_autoprobe) + __device_attach(dev, true); + + subsys_put(sp); } /* @@ -1034,7 +1110,7 @@ static void __device_driver_lock(struct device *dev, struct device *parent) * @parent: Parent device. Needed if the bus requires parent lock * * This function will release the required locks for manipulating dev->drv. - * Normally this will just be the the @dev lock, but when called for a + * Normally this will just be the @dev lock, but when called for a * USB interface, @parent lock will be released as well. */ static void __device_driver_unlock(struct device *dev, struct device *parent) @@ -1052,7 +1128,7 @@ static void __device_driver_unlock(struct device *dev, struct device *parent) * Manually attach driver to a device. Will acquire both @dev lock and * @dev->parent lock if needed. Returns 0 on success, -ERR on failure. */ -int device_driver_attach(struct device_driver *drv, struct device *dev) +int device_driver_attach(const struct device_driver *drv, struct device *dev) { int ret; @@ -1072,11 +1148,12 @@ EXPORT_SYMBOL_GPL(device_driver_attach); static void __driver_attach_async_helper(void *_dev, async_cookie_t cookie) { struct device *dev = _dev; - struct device_driver *drv; + const struct device_driver *drv; int ret; __device_driver_lock(dev, dev->parent); drv = dev->p->async_driver; + dev->p->async_driver = NULL; ret = driver_probe_device(drv, dev); __device_driver_unlock(dev, dev->parent); @@ -1087,7 +1164,8 @@ static void __driver_attach_async_helper(void *_dev, async_cookie_t cookie) static int __driver_attach(struct device *dev, void *data) { - struct device_driver *drv = data; + const struct device_driver *drv = data; + bool async = false; int ret; /* @@ -1108,9 +1186,18 @@ static int __driver_attach(struct device *dev, void *data) dev_dbg(dev, "Device match requests probe deferral\n"); dev->can_match = true; driver_deferred_probe_add(dev); + /* + * Driver could not match with device, but may match with + * another device on the bus. + */ + return 0; } else if (ret < 0) { dev_dbg(dev, "Bus failed to match device: %d\n", ret); - return ret; + /* + * Driver could not match with device, but may match with + * another device on the bus. + */ + return 0; } /* ret > 0 means positive match */ if (driver_allows_async_probing(drv)) { @@ -1123,12 +1210,14 @@ static int __driver_attach(struct device *dev, void *data) */ dev_dbg(dev, "probing driver %s asynchronously\n", drv->name); device_lock(dev); - if (!dev->driver) { + if (!dev->driver && !dev->p->async_driver) { get_device(dev); dev->p->async_driver = drv; - async_schedule_dev(__driver_attach_async_helper, dev); + async = true; } device_unlock(dev); + if (async) + async_schedule_dev(__driver_attach_async_helper, dev); return 0; } @@ -1148,9 +1237,10 @@ static int __driver_attach(struct device *dev, void *data) * returns 0 and the @dev->driver is set, we've found a * compatible pair. */ -int driver_attach(struct device_driver *drv) +int driver_attach(const struct device_driver *drv) { - return bus_for_each_dev(drv->bus, NULL, drv, __driver_attach); + /* The (void *) will be put back to const * in __driver_attach() */ + return bus_for_each_dev(drv->bus, NULL, (void *)drv, __driver_attach); } EXPORT_SYMBOL_GPL(driver_attach); @@ -1185,45 +1275,28 @@ static void __device_release_driver(struct device *dev, struct device *parent) driver_sysfs_remove(dev); - if (dev->bus) - blocking_notifier_call_chain(&dev->bus->p->bus_notifier, - BUS_NOTIFY_UNBIND_DRIVER, - dev); + bus_notify(dev, BUS_NOTIFY_UNBIND_DRIVER); pm_runtime_put_sync(dev); - device_remove_file(dev, &dev_attr_state_synced); - device_remove_groups(dev, drv->dev_groups); + device_remove(dev); - if (dev->bus && dev->bus->remove) - dev->bus->remove(dev); - else if (drv->remove) - drv->remove(dev); + if (dev->bus && dev->bus->dma_cleanup) + dev->bus->dma_cleanup(dev); + device_unbind_cleanup(dev); device_links_driver_cleanup(dev); - devres_release_all(dev); - arch_teardown_dma_ops(dev); - dev->driver = NULL; - dev_set_drvdata(dev, NULL); - if (dev->pm_domain && dev->pm_domain->dismiss) - dev->pm_domain->dismiss(dev); - pm_runtime_reinit(dev); - dev_pm_set_driver_flags(dev, 0); - klist_remove(&dev->p->knode_driver); device_pm_check_callbacks(dev); - if (dev->bus) - blocking_notifier_call_chain(&dev->bus->p->bus_notifier, - BUS_NOTIFY_UNBOUND_DRIVER, - dev); + bus_notify(dev, BUS_NOTIFY_UNBOUND_DRIVER); kobject_uevent(&dev->kobj, KOBJ_UNBIND); } } void device_release_driver_internal(struct device *dev, - struct device_driver *drv, + const struct device_driver *drv, struct device *parent) { __device_driver_lock(dev, parent); @@ -1272,7 +1345,7 @@ void device_driver_detach(struct device *dev) * driver_detach - detach driver from all devices it controls. * @drv: driver. */ -void driver_detach(struct device_driver *drv) +void driver_detach(const struct device_driver *drv) { struct device_private *dev_prv; struct device *dev; diff --git a/drivers/base/devcoredump.c b/drivers/base/devcoredump.c index f4d794d6bb85..55bdc7f5e59d 100644 --- a/drivers/base/devcoredump.c +++ b/drivers/base/devcoredump.c @@ -18,17 +18,51 @@ static struct class devcd_class; /* global disable flag, for security purposes */ static bool devcd_disabled; -/* if data isn't read by userspace after 5 minutes then delete it */ -#define DEVCD_TIMEOUT (HZ * 60 * 5) - struct devcd_entry { struct device devcd_dev; void *data; size_t datalen; + /* + * There are 2 races for which mutex is required. + * + * The first race is between device creation and userspace writing to + * schedule immediately destruction. + * + * This race is handled by arming the timer before device creation, but + * when device creation fails the timer still exists. + * + * To solve this, hold the mutex during device_add(), and set + * init_completed on success before releasing the mutex. + * + * That way the timer will never fire until device_add() is called, + * it will do nothing if init_completed is not set. The timer is also + * cancelled in that case. + * + * The second race involves multiple parallel invocations of devcd_free(), + * add a deleted flag so only 1 can call the destructor. + */ + struct mutex mutex; + bool init_completed, deleted; struct module *owner; ssize_t (*read)(char *buffer, loff_t offset, size_t count, void *data, size_t datalen); void (*free)(void *data); + /* + * If nothing interferes and device_add() was returns success, + * del_wk will destroy the device after the timer fires. + * + * Multiple userspace processes can interfere in the working of the timer: + * - Writing to the coredump will reschedule the timer to run immediately, + * if still armed. + * + * This is handled by using "if (cancel_delayed_work()) { + * schedule_delayed_work() }", to prevent re-arming after having + * been previously fired. + * - Writing to /sys/class/devcoredump/disabled will destroy the + * coredump synchronously. + * This is handled by using disable_delayed_work_sync(), and then + * checking if deleted flag is set with &devcd->mutex held. + */ struct delayed_work del_wk; struct device *failing_dev; }; @@ -57,18 +91,31 @@ static void devcd_dev_release(struct device *dev) kfree(devcd); } +static void __devcd_del(struct devcd_entry *devcd) +{ + devcd->deleted = true; + device_del(&devcd->devcd_dev); + put_device(&devcd->devcd_dev); +} + static void devcd_del(struct work_struct *wk) { struct devcd_entry *devcd; + bool init_completed; devcd = container_of(wk, struct devcd_entry, del_wk.work); - device_del(&devcd->devcd_dev); - put_device(&devcd->devcd_dev); + /* devcd->mutex serializes against dev_coredumpm_timeout */ + mutex_lock(&devcd->mutex); + init_completed = devcd->init_completed; + mutex_unlock(&devcd->mutex); + + if (init_completed) + __devcd_del(devcd); } static ssize_t devcd_data_read(struct file *filp, struct kobject *kobj, - struct bin_attribute *bin_attr, + const struct bin_attribute *bin_attr, char *buffer, loff_t offset, size_t count) { struct device *dev = kobj_to_dev(kobj); @@ -78,25 +125,26 @@ static ssize_t devcd_data_read(struct file *filp, struct kobject *kobj, } static ssize_t devcd_data_write(struct file *filp, struct kobject *kobj, - struct bin_attribute *bin_attr, + const struct bin_attribute *bin_attr, char *buffer, loff_t offset, size_t count) { struct device *dev = kobj_to_dev(kobj); struct devcd_entry *devcd = dev_to_devcd(dev); - mod_delayed_work(system_wq, &devcd->del_wk, 0); + /* + * Although it's tempting to use mod_delayed work here, + * that will cause a reschedule if the timer already fired. + */ + if (cancel_delayed_work(&devcd->del_wk)) + schedule_delayed_work(&devcd->del_wk, 0); return count; } -static struct bin_attribute devcd_attr_data = { - .attr = { .name = "data", .mode = S_IRUSR | S_IWUSR, }, - .size = 0, - .read = devcd_data_read, - .write = devcd_data_write, -}; +static const struct bin_attribute devcd_attr_data = + __BIN_ATTR(data, 0600, devcd_data_read, devcd_data_write, 0); -static struct bin_attribute *devcd_dev_bin_attrs[] = { +static const struct bin_attribute *const devcd_dev_bin_attrs[] = { &devcd_attr_data, NULL, }; @@ -112,17 +160,54 @@ static int devcd_free(struct device *dev, void *data) { struct devcd_entry *devcd = dev_to_devcd(dev); - flush_delayed_work(&devcd->del_wk); + /* + * To prevent a race with devcd_data_write(), disable work and + * complete manually instead. + * + * We cannot rely on the return value of + * disable_delayed_work_sync() here, because it might be in the + * middle of a cancel_delayed_work + schedule_delayed_work pair. + * + * devcd->mutex here guards against multiple parallel invocations + * of devcd_free(). + */ + disable_delayed_work_sync(&devcd->del_wk); + mutex_lock(&devcd->mutex); + if (!devcd->deleted) + __devcd_del(devcd); + mutex_unlock(&devcd->mutex); return 0; } -static ssize_t disabled_show(struct class *class, struct class_attribute *attr, +static ssize_t disabled_show(const struct class *class, const struct class_attribute *attr, char *buf) { return sysfs_emit(buf, "%d\n", devcd_disabled); } -static ssize_t disabled_store(struct class *class, struct class_attribute *attr, +/* + * + * disabled_store() worker() + * class_for_each_device(&devcd_class, + * NULL, NULL, devcd_free) + * ... + * ... + * while ((dev = class_dev_iter_next(&iter)) + * devcd_del() + * device_del() + * put_device() <- last reference + * error = fn(dev, data) devcd_dev_release() + * devcd_free(dev, data) kfree(devcd) + * + * + * In the above diagram, it looks like disabled_store() would be racing with parallelly + * running devcd_del() and result in memory abort after dropping its last reference with + * put_device(). However, this will not happens as fn(dev, data) runs + * with its own reference to device via klist_node so it is not its last reference. + * so, above situation would not occur. + */ + +static ssize_t disabled_store(const struct class *class, const struct class_attribute *attr, const char *buf, size_t count) { long tmp = simple_strtol(buf, NULL, 10); @@ -151,7 +236,6 @@ ATTRIBUTE_GROUPS(devcd_class); static struct class devcd_class = { .name = "devcoredump", - .owner = THIS_MODULE, .dev_release = devcd_dev_release, .dev_groups = devcd_dev_groups, .class_groups = devcd_class_groups, @@ -214,6 +298,8 @@ static void devcd_free_sgtable(void *data) * @offset: start copy from @offset@ bytes from the head of the data * in the given scatterlist * @data_len: the length of the data in the sg_table + * + * Returns: the number of bytes copied */ static ssize_t devcd_read_from_sgtable(char *buffer, loff_t offset, size_t buf_len, void *data, @@ -231,7 +317,31 @@ static ssize_t devcd_read_from_sgtable(char *buffer, loff_t offset, } /** - * dev_coredumpm - create device coredump with read/free methods + * dev_coredump_put - remove device coredump + * @dev: the struct device for the crashed device + * + * dev_coredump_put() removes coredump, if exists, for a given device from + * the file system and free its associated data otherwise, does nothing. + * + * It is useful for modules that do not want to keep coredump + * available after its unload. + */ +void dev_coredump_put(struct device *dev) +{ + struct device *existing; + + existing = class_find_device(&devcd_class, NULL, dev, + devcd_match_failing); + if (existing) { + devcd_free(existing, NULL); + put_device(existing); + } +} +EXPORT_SYMBOL_GPL(dev_coredump_put); + +/** + * dev_coredumpm_timeout - create device coredump with read/free methods with a + * custom timeout. * @dev: the struct device for the crashed device * @owner: the module that contains the read/free functions, use %THIS_MODULE * @data: data cookie for the @read/@free functions @@ -239,17 +349,20 @@ static ssize_t devcd_read_from_sgtable(char *buffer, loff_t offset, * @gfp: allocation flags * @read: function to read from the given buffer * @free: function to free the given buffer + * @timeout: time in jiffies to remove coredump * * Creates a new device coredump for the given device. If a previous one hasn't * been read yet, the new coredump is discarded. The data lifetime is determined * by the device coredump framework and when it is no longer needed the @free * function will be called to free the data. */ -void dev_coredumpm(struct device *dev, struct module *owner, - void *data, size_t datalen, gfp_t gfp, - ssize_t (*read)(char *buffer, loff_t offset, size_t count, - void *data, size_t datalen), - void (*free)(void *data)) +void dev_coredumpm_timeout(struct device *dev, struct module *owner, + void *data, size_t datalen, gfp_t gfp, + ssize_t (*read)(char *buffer, loff_t offset, + size_t count, void *data, + size_t datalen), + void (*free)(void *data), + unsigned long timeout) { static atomic_t devcd_count = ATOMIC_INIT(0); struct devcd_entry *devcd; @@ -278,13 +391,23 @@ void dev_coredumpm(struct device *dev, struct module *owner, devcd->read = read; devcd->free = free; devcd->failing_dev = get_device(dev); + devcd->deleted = false; + mutex_init(&devcd->mutex); device_initialize(&devcd->devcd_dev); dev_set_name(&devcd->devcd_dev, "devcd%d", atomic_inc_return(&devcd_count)); devcd->devcd_dev.class = &devcd_class; + dev_set_uevent_suppress(&devcd->devcd_dev, true); + + /* devcd->mutex prevents devcd_del() completing until init finishes */ + mutex_lock(&devcd->mutex); + devcd->init_completed = false; + INIT_DELAYED_WORK(&devcd->del_wk, devcd_del); + schedule_delayed_work(&devcd->del_wk, timeout); + if (device_add(&devcd->devcd_dev)) goto put_device; @@ -299,18 +422,28 @@ void dev_coredumpm(struct device *dev, struct module *owner, "devcoredump")) dev_warn(dev, "devcoredump create_link failed\n"); - INIT_DELAYED_WORK(&devcd->del_wk, devcd_del); - schedule_delayed_work(&devcd->del_wk, DEVCD_TIMEOUT); + dev_set_uevent_suppress(&devcd->devcd_dev, false); + kobject_uevent(&devcd->devcd_dev.kobj, KOBJ_ADD); + /* + * Safe to run devcd_del() now that we are done with devcd_dev. + * Alternatively we could have taken a ref on devcd_dev before + * dropping the lock. + */ + devcd->init_completed = true; + mutex_unlock(&devcd->mutex); return; put_device: + mutex_unlock(&devcd->mutex); + cancel_delayed_work_sync(&devcd->del_wk); put_device(&devcd->devcd_dev); + put_module: module_put(owner); free: free(data); } -EXPORT_SYMBOL_GPL(dev_coredumpm); +EXPORT_SYMBOL_GPL(dev_coredumpm_timeout); /** * dev_coredumpsg - create device coredump that uses scatterlist as data diff --git a/drivers/base/devres.c b/drivers/base/devres.c index eaa9a5cd1db9..f54db6d138ab 100644 --- a/drivers/base/devres.c +++ b/drivers/base/devres.c @@ -29,10 +29,10 @@ struct devres { * Some archs want to perform DMA into kmalloc caches * and need a guaranteed alignment larger than * the alignment of a 64-bit integer. - * Thus we use ARCH_KMALLOC_MINALIGN here and get exactly the same - * buffer alignment as if it was allocated by plain kmalloc(). + * Thus we use ARCH_DMA_MINALIGN for data[] which will force the same + * alignment for struct devres when allocated by kmalloc(). */ - u8 __aligned(ARCH_KMALLOC_MINALIGN) data[]; + u8 __aligned(ARCH_DMA_MINALIGN) data[]; }; struct devres_group { @@ -85,7 +85,7 @@ static void group_close_release(struct device *dev, void *res) /* noop */ } -static struct devres_group * node_to_group(struct devres_node *node) +static struct devres_group *node_to_group(struct devres_node *node) { if (node->release == &group_open_release) return container_of(node, struct devres_group, node[0]); @@ -101,11 +101,14 @@ static bool check_dr_size(size_t size, size_t *tot_size) size, tot_size))) return false; + /* Actually allocate the full kmalloc bucket size. */ + *tot_size = kmalloc_size_roundup(*tot_size); + return true; } -static __always_inline struct devres * alloc_dr(dr_release_t release, - size_t size, gfp_t gfp, int nid) +static __always_inline struct devres *alloc_dr(dr_release_t release, + size_t size, gfp_t gfp, int nid) { size_t tot_size; struct devres *dr; @@ -117,7 +120,9 @@ static __always_inline struct devres * alloc_dr(dr_release_t release, if (unlikely(!dr)) return NULL; - memset(dr, 0, offsetof(struct devres, data)); + /* No need to clear memory twice */ + if (!(gfp & __GFP_ZERO)) + memset(dr, 0, offsetof(struct devres, data)); INIT_LIST_HEAD(&dr->node.entry); dr->node.release = release; @@ -278,8 +283,8 @@ static struct devres *find_dr(struct device *dev, dr_release_t release, * RETURNS: * Pointer to found devres, NULL if not found. */ -void * devres_find(struct device *dev, dr_release_t release, - dr_match_t match, void *match_data) +void *devres_find(struct device *dev, dr_release_t release, + dr_match_t match, void *match_data) { struct devres *dr; unsigned long flags; @@ -308,8 +313,8 @@ EXPORT_SYMBOL_GPL(devres_find); * RETURNS: * Pointer to found or added devres. */ -void * devres_get(struct device *dev, void *new_res, - dr_match_t match, void *match_data) +void *devres_get(struct device *dev, void *new_res, + dr_match_t match, void *match_data) { struct devres *new_dr = container_of(new_res, struct devres, data); struct devres *dr; @@ -344,8 +349,8 @@ EXPORT_SYMBOL_GPL(devres_get); * RETURNS: * Pointer to removed devres on success, NULL if not found. */ -void * devres_remove(struct device *dev, dr_release_t release, - dr_match_t match, void *match_data) +void *devres_remove(struct device *dev, dr_release_t release, + dr_match_t match, void *match_data) { struct devres *dr; unsigned long flags; @@ -544,7 +549,7 @@ int devres_release_all(struct device *dev) * RETURNS: * ID of the new group, NULL on failure. */ -void * devres_open_group(struct device *dev, void *id, gfp_t gfp) +void *devres_open_group(struct device *dev, void *id, gfp_t gfp) { struct devres_group *grp; unsigned long flags; @@ -562,6 +567,7 @@ void * devres_open_group(struct device *dev, void *id, gfp_t gfp) grp->id = grp; if (id) grp->id = id; + grp->color = 0; spin_lock_irqsave(&dev->devres_lock, flags); add_dr(dev, &grp->node[0]); @@ -570,8 +576,11 @@ void * devres_open_group(struct device *dev, void *id, gfp_t gfp) } EXPORT_SYMBOL_GPL(devres_open_group); -/* Find devres group with ID @id. If @id is NULL, look for the latest. */ -static struct devres_group * find_group(struct device *dev, void *id) +/* + * Find devres group with ID @id. If @id is NULL, look for the latest open + * group. + */ +static struct devres_group *find_group(struct device *dev, void *id) { struct devres_node *node; @@ -681,6 +690,13 @@ int devres_release_group(struct device *dev, void *id) spin_unlock_irqrestore(&dev->devres_lock, flags); release_nodes(dev, &todo); + } else if (list_empty(&dev->devres_head)) { + /* + * dev is probably dying via devres_release_all(): groups + * have already been removed and are on the process of + * being released - don't touch and don't warn. + */ + spin_unlock_irqrestore(&dev->devres_lock, flags); } else { WARN_ON(1); spin_unlock_irqrestore(&dev->devres_lock, flags); @@ -692,7 +708,7 @@ EXPORT_SYMBOL_GPL(devres_release_group); /* * Custom devres actions allow inserting a simple function call - * into the teadown sequence. + * into the teardown sequence. */ struct action_devres { @@ -717,20 +733,21 @@ static void devm_action_release(struct device *dev, void *res) } /** - * devm_add_action() - add a custom action to list of managed resources + * __devm_add_action() - add a custom action to list of managed resources * @dev: Device that owns the action * @action: Function that should be called * @data: Pointer to data passed to @action implementation + * @name: Name of the resource (for debugging purposes) * * This adds a custom action to the list of managed resources so that * it gets executed as part of standard resource unwinding. */ -int devm_add_action(struct device *dev, void (*action)(void *), void *data) +int __devm_add_action(struct device *dev, void (*action)(void *), void *data, const char *name) { struct action_devres *devres; - devres = devres_alloc(devm_action_release, - sizeof(struct action_devres), GFP_KERNEL); + devres = __devres_alloc_node(devm_action_release, sizeof(struct action_devres), + GFP_KERNEL, NUMA_NO_NODE, name); if (!devres) return -ENOMEM; @@ -740,28 +757,52 @@ int devm_add_action(struct device *dev, void (*action)(void *), void *data) devres_add(dev, devres); return 0; } -EXPORT_SYMBOL_GPL(devm_add_action); +EXPORT_SYMBOL_GPL(__devm_add_action); + +bool devm_is_action_added(struct device *dev, void (*action)(void *), void *data) +{ + struct action_devres devres = { + .data = data, + .action = action, + }; + + return devres_find(dev, devm_action_release, devm_action_match, &devres); +} +EXPORT_SYMBOL_GPL(devm_is_action_added); /** - * devm_remove_action() - removes previously added custom action + * devm_remove_action_nowarn() - removes previously added custom action * @dev: Device that owns the action * @action: Function implementing the action * @data: Pointer to data passed to @action implementation * * Removes instance of @action previously added by devm_add_action(). * Both action and data should match one of the existing entries. + * + * In contrast to devm_remove_action(), this function does not WARN() if no + * entry could have been found. + * + * This should only be used if the action is contained in an object with + * independent lifetime management, e.g. the Devres rust abstraction. + * + * Causing the warning from regular driver code most likely indicates an abuse + * of the devres API. + * + * Returns: 0 on success, -ENOENT if no entry could have been found. */ -void devm_remove_action(struct device *dev, void (*action)(void *), void *data) +int devm_remove_action_nowarn(struct device *dev, + void (*action)(void *), + void *data) { struct action_devres devres = { .data = data, .action = action, }; - WARN_ON(devres_destroy(dev, devm_action_release, devm_action_match, - &devres)); + return devres_destroy(dev, devm_action_release, devm_action_match, + &devres); } -EXPORT_SYMBOL_GPL(devm_remove_action); +EXPORT_SYMBOL_GPL(devm_remove_action_nowarn); /** * devm_release_action() - release previously added custom action @@ -890,9 +931,12 @@ void *devm_krealloc(struct device *dev, void *ptr, size_t new_size, gfp_t gfp) /* * Otherwise: allocate new, larger chunk. We need to allocate before * taking the lock as most probably the caller uses GFP_KERNEL. + * alloc_dr() will call check_dr_size() to reserve extra memory + * for struct devres automatically, so size @new_size user request + * is delivered to it directly as devm_kmalloc() does. */ new_dr = alloc_dr(devm_kmalloc_release, - total_new_size, gfp, dev_to_node(dev)); + new_size, gfp, dev_to_node(dev)); if (!new_dr) return NULL; @@ -916,7 +960,7 @@ void *devm_krealloc(struct device *dev, void *ptr, size_t new_size, gfp_t gfp) /* * We can copy the memory contents after releasing the lock as we're - * no longer modyfing the list links. + * no longer modifying the list links. */ memcpy(new_dr->data, old_dr->data, total_old_size - offsetof(struct devres, data)); @@ -943,17 +987,10 @@ EXPORT_SYMBOL_GPL(devm_krealloc); */ char *devm_kstrdup(struct device *dev, const char *s, gfp_t gfp) { - size_t size; - char *buf; - if (!s) return NULL; - size = strlen(s) + 1; - buf = devm_kmalloc(dev, size, gfp); - if (buf) - memcpy(buf, s, size); - return buf; + return devm_kmemdup(dev, s, strlen(s) + 1, gfp); } EXPORT_SYMBOL_GPL(devm_kstrdup); @@ -1080,6 +1117,27 @@ void *devm_kmemdup(struct device *dev, const void *src, size_t len, gfp_t gfp) } EXPORT_SYMBOL_GPL(devm_kmemdup); +/** + * devm_kmemdup_const - conditionally duplicate and manage a region of memory + * + * @dev: Device this memory belongs to + * @src: memory region to duplicate + * @len: memory region length, + * @gfp: GFP mask to use + * + * Return: source address if it is in .rodata or the return value of kmemdup() + * to which the function falls back otherwise. + */ +const void * +devm_kmemdup_const(struct device *dev, const void *src, size_t len, gfp_t gfp) +{ + if (is_kernel_rodata((unsigned long)src)) + return src; + + return devm_kmemdup(dev, src, len, gfp); +} +EXPORT_SYMBOL_GPL(devm_kmemdup_const); + struct pages_devres { unsigned long addr; unsigned int order; @@ -1164,13 +1222,6 @@ static void devm_percpu_release(struct device *dev, void *pdata) free_percpu(p); } -static int devm_percpu_match(struct device *dev, void *data, void *p) -{ - struct devres *devr = container_of(data, struct devres, data); - - return *(void **)devr->data == p; -} - /** * __devm_alloc_percpu - Resource-managed alloc_percpu * @dev: Device to allocate per-cpu memory for @@ -1206,17 +1257,3 @@ void __percpu *__devm_alloc_percpu(struct device *dev, size_t size, return pcpu; } EXPORT_SYMBOL_GPL(__devm_alloc_percpu); - -/** - * devm_free_percpu - Resource-managed free_percpu - * @dev: Device this memory belongs to - * @pdata: Per-cpu memory to free - * - * Free memory allocated with devm_alloc_percpu(). - */ -void devm_free_percpu(struct device *dev, void __percpu *pdata) -{ - WARN_ON(devres_destroy(dev, devm_percpu_release, devm_percpu_match, - (__force void *)pdata)); -} -EXPORT_SYMBOL_GPL(devm_free_percpu); diff --git a/drivers/base/devtmpfs.c b/drivers/base/devtmpfs.c index 8be352ab4ddb..194b44075ac7 100644 --- a/drivers/base/devtmpfs.c +++ b/drivers/base/devtmpfs.c @@ -13,11 +13,13 @@ * overwrite the default setting if needed. */ +#define pr_fmt(fmt) "devtmpfs: " fmt + #include <linux/kernel.h> #include <linux/syscalls.h> #include <linux/mount.h> #include <linux/device.h> -#include <linux/genhd.h> +#include <linux/blkdev.h> #include <linux/namei.h> #include <linux/fs.h> #include <linux/shmem_fs.h> @@ -29,6 +31,12 @@ #include <uapi/linux/mount.h> #include "base.h" +#ifdef CONFIG_DEVTMPFS_SAFE +#define DEVTMPFS_MFLAGS (MS_SILENT | MS_NOEXEC | MS_NOSUID) +#else +#define DEVTMPFS_MFLAGS (MS_SILENT) +#endif + static struct task_struct *thread; static int __initdata mount_dev = IS_ENABLED(CONFIG_DEVTMPFS_MOUNT); @@ -55,40 +63,51 @@ __setup("devtmpfs.mount=", mount_param); static struct vfsmount *mnt; -static struct dentry *public_dev_mount(struct file_system_type *fs_type, int flags, - const char *dev_name, void *data) -{ - struct super_block *s = mnt->mnt_sb; - atomic_inc(&s->s_active); - down_write(&s->s_umount); - return dget(s->s_root); -} - static struct file_system_type internal_fs_type = { .name = "devtmpfs", #ifdef CONFIG_TMPFS .init_fs_context = shmem_init_fs_context, - .parameters = shmem_fs_parameters, #else .init_fs_context = ramfs_init_fs_context, - .parameters = ramfs_fs_parameters, #endif - .kill_sb = kill_litter_super, + .kill_sb = kill_anon_super, }; -static struct file_system_type dev_fs_type = { - .name = "devtmpfs", - .mount = public_dev_mount, -}; - -#ifdef CONFIG_BLOCK -static inline int is_blockdev(struct device *dev) +/* Simply take a ref on the existing mount */ +static int devtmpfs_get_tree(struct fs_context *fc) { - return dev->class == &block_class; + struct super_block *sb = mnt->mnt_sb; + + atomic_inc(&sb->s_active); + down_write(&sb->s_umount); + fc->root = dget(sb->s_root); + return 0; } + +/* Ops are filled in during init depending on underlying shmem or ramfs type */ +struct fs_context_operations devtmpfs_context_ops = {}; + +/* Call the underlying initialization and set to our ops */ +static int devtmpfs_init_fs_context(struct fs_context *fc) +{ + int ret; +#ifdef CONFIG_TMPFS + ret = shmem_init_fs_context(fc); #else -static inline int is_blockdev(struct device *dev) { return 0; } + ret = ramfs_init_fs_context(fc); #endif + if (ret < 0) + return ret; + + fc->ops = &devtmpfs_context_ops; + + return 0; +} + +static struct file_system_type dev_fs_type = { + .name = "devtmpfs", + .init_fs_context = devtmpfs_init_fs_context, +}; static int devtmpfs_submit_req(struct req *req, const char *tmp) { @@ -156,18 +175,17 @@ static int dev_mkdir(const char *name, umode_t mode) { struct dentry *dentry; struct path path; - int err; - dentry = kern_path_create(AT_FDCWD, name, &path, LOOKUP_DIRECTORY); + dentry = start_creating_path(AT_FDCWD, name, &path, LOOKUP_DIRECTORY); if (IS_ERR(dentry)) return PTR_ERR(dentry); - err = vfs_mkdir(&init_user_ns, d_inode(path.dentry), dentry, mode); - if (!err) + dentry = vfs_mkdir(&nop_mnt_idmap, d_inode(path.dentry), dentry, mode, NULL); + if (!IS_ERR(dentry)) /* mark as kernel-created inode */ d_inode(dentry)->i_private = &thread; - done_path_create(&path, dentry); - return err; + end_creating_path(&path, dentry); + return PTR_ERR_OR_ZERO(dentry); } static int create_path(const char *nodepath) @@ -204,16 +222,16 @@ static int handle_create(const char *nodename, umode_t mode, kuid_t uid, struct path path; int err; - dentry = kern_path_create(AT_FDCWD, nodename, &path, 0); + dentry = start_creating_path(AT_FDCWD, nodename, &path, 0); if (dentry == ERR_PTR(-ENOENT)) { create_path(nodename); - dentry = kern_path_create(AT_FDCWD, nodename, &path, 0); + dentry = start_creating_path(AT_FDCWD, nodename, &path, 0); } if (IS_ERR(dentry)) return PTR_ERR(dentry); - err = vfs_mknod(&init_user_ns, d_inode(path.dentry), dentry, mode, - dev->devt); + err = vfs_mknod(&nop_mnt_idmap, d_inode(path.dentry), dentry, mode, + dev->devt, NULL); if (!err) { struct iattr newattrs; @@ -222,13 +240,13 @@ static int handle_create(const char *nodename, umode_t mode, kuid_t uid, newattrs.ia_gid = gid; newattrs.ia_valid = ATTR_MODE|ATTR_UID|ATTR_GID; inode_lock(d_inode(dentry)); - notify_change(&init_user_ns, dentry, &newattrs, NULL); + notify_change(&nop_mnt_idmap, dentry, &newattrs, NULL); inode_unlock(d_inode(dentry)); /* mark as kernel-created inode */ d_inode(dentry)->i_private = &thread; } - done_path_create(&path, dentry); + end_creating_path(&path, dentry); return err; } @@ -238,21 +256,16 @@ static int dev_rmdir(const char *name) struct dentry *dentry; int err; - dentry = kern_path_locked(name, &parent); + dentry = start_removing_path(name, &parent); if (IS_ERR(dentry)) return PTR_ERR(dentry); - if (d_really_is_positive(dentry)) { - if (d_inode(dentry)->i_private == &thread) - err = vfs_rmdir(&init_user_ns, d_inode(parent.dentry), - dentry); - else - err = -EPERM; - } else { - err = -ENOENT; - } - dput(dentry); - inode_unlock(d_inode(parent.dentry)); - path_put(&parent); + if (d_inode(dentry)->i_private == &thread) + err = vfs_rmdir(&nop_mnt_idmap, d_inode(parent.dentry), + dentry, NULL); + else + err = -EPERM; + + end_removing_path(&parent, dentry); return err; } @@ -281,7 +294,7 @@ static int delete_path(const char *nodepath) return err; } -static int dev_mynode(struct device *dev, struct inode *inode, struct kstat *stat) +static int dev_mynode(struct device *dev, struct inode *inode) { /* did we create it */ if (inode->i_private != &thread) @@ -289,13 +302,13 @@ static int dev_mynode(struct device *dev, struct inode *inode, struct kstat *sta /* does the dev_t match */ if (is_blockdev(dev)) { - if (!S_ISBLK(stat->mode)) + if (!S_ISBLK(inode->i_mode)) return 0; } else { - if (!S_ISCHR(stat->mode)) + if (!S_ISCHR(inode->i_mode)) return 0; } - if (stat->rdev != dev->devt) + if (inode->i_rdev != dev->devt) return 0; /* ours */ @@ -306,44 +319,36 @@ static int handle_remove(const char *nodename, struct device *dev) { struct path parent; struct dentry *dentry; + struct inode *inode; int deleted = 0; - int err; + int err = 0; - dentry = kern_path_locked(nodename, &parent); + dentry = start_removing_path(nodename, &parent); if (IS_ERR(dentry)) return PTR_ERR(dentry); - if (d_really_is_positive(dentry)) { - struct kstat stat; - struct path p = {.mnt = parent.mnt, .dentry = dentry}; - err = vfs_getattr(&p, &stat, STATX_TYPE | STATX_MODE, - AT_STATX_SYNC_AS_STAT); - if (!err && dev_mynode(dev, d_inode(dentry), &stat)) { - struct iattr newattrs; - /* - * before unlinking this node, reset permissions - * of possible references like hardlinks - */ - newattrs.ia_uid = GLOBAL_ROOT_UID; - newattrs.ia_gid = GLOBAL_ROOT_GID; - newattrs.ia_mode = stat.mode & ~0777; - newattrs.ia_valid = - ATTR_UID|ATTR_GID|ATTR_MODE; - inode_lock(d_inode(dentry)); - notify_change(&init_user_ns, dentry, &newattrs, NULL); - inode_unlock(d_inode(dentry)); - err = vfs_unlink(&init_user_ns, d_inode(parent.dentry), - dentry, NULL); - if (!err || err == -ENOENT) - deleted = 1; - } - } else { - err = -ENOENT; + inode = d_inode(dentry); + if (dev_mynode(dev, inode)) { + struct iattr newattrs; + /* + * before unlinking this node, reset permissions + * of possible references like hardlinks + */ + newattrs.ia_uid = GLOBAL_ROOT_UID; + newattrs.ia_gid = GLOBAL_ROOT_GID; + newattrs.ia_mode = inode->i_mode & ~0777; + newattrs.ia_valid = + ATTR_UID|ATTR_GID|ATTR_MODE; + inode_lock(d_inode(dentry)); + notify_change(&nop_mnt_idmap, dentry, &newattrs, NULL); + inode_unlock(d_inode(dentry)); + err = vfs_unlink(&nop_mnt_idmap, d_inode(parent.dentry), + dentry, NULL); + if (!err || err == -ENOENT) + deleted = 1; } - dput(dentry); - inode_unlock(d_inode(parent.dentry)); + end_removing_path(&parent, dentry); - path_put(&parent); if (deleted && strchr(nodename, '/')) delete_path(nodename); return err; @@ -363,11 +368,11 @@ int __init devtmpfs_mount(void) if (!thread) return 0; - err = init_mount("devtmpfs", "dev", "devtmpfs", MS_SILENT, NULL); + err = init_mount("devtmpfs", "dev", "devtmpfs", DEVTMPFS_MFLAGS, NULL); if (err) - printk(KERN_INFO "devtmpfs: error mounting %i\n", err); + pr_info("error mounting %d\n", err); else - printk(KERN_INFO "devtmpfs: mounted\n"); + pr_info("mounted\n"); return err; } @@ -412,7 +417,7 @@ static noinline int __init devtmpfs_setup(void *p) err = ksys_unshare(CLONE_NEWNS); if (err) goto out; - err = init_mount("devtmpfs", "/", "devtmpfs", MS_SILENT, NULL); + err = init_mount("devtmpfs", "/", "devtmpfs", DEVTMPFS_MFLAGS, NULL); if (err) goto out; init_chdir("/.."); /* will traverse into overmounted root */ @@ -439,6 +444,31 @@ static int __ref devtmpfsd(void *p) } /* + * Get the underlying (shmem/ramfs) context ops to build ours + */ +static int devtmpfs_configure_context(void) +{ + struct fs_context *fc; + + fc = fs_context_for_reconfigure(mnt->mnt_root, mnt->mnt_sb->s_flags, + MS_RMT_MASK); + if (IS_ERR(fc)) + return PTR_ERR(fc); + + /* Set up devtmpfs_context_ops based on underlying type */ + devtmpfs_context_ops.free = fc->ops->free; + devtmpfs_context_ops.dup = fc->ops->dup; + devtmpfs_context_ops.parse_param = fc->ops->parse_param; + devtmpfs_context_ops.parse_monolithic = fc->ops->parse_monolithic; + devtmpfs_context_ops.get_tree = &devtmpfs_get_tree; + devtmpfs_context_ops.reconfigure = fc->ops->reconfigure; + + put_fs_context(fc); + + return 0; +} + +/* * Create devtmpfs instance, driver-core devices will add their device * nodes here. */ @@ -449,14 +479,19 @@ int __init devtmpfs_init(void) mnt = vfs_kern_mount(&internal_fs_type, 0, "devtmpfs", opts); if (IS_ERR(mnt)) { - printk(KERN_ERR "devtmpfs: unable to create devtmpfs %ld\n", - PTR_ERR(mnt)); + pr_err("unable to create devtmpfs %ld\n", PTR_ERR(mnt)); return PTR_ERR(mnt); } + + err = devtmpfs_configure_context(); + if (err) { + pr_err("unable to configure devtmpfs type %d\n", err); + return err; + } + err = register_filesystem(&dev_fs_type); if (err) { - printk(KERN_ERR "devtmpfs: unable to register devtmpfs " - "type %i\n", err); + pr_err("unable to register devtmpfs type %d\n", err); return err; } @@ -469,11 +504,12 @@ int __init devtmpfs_init(void) } if (err) { - printk(KERN_ERR "devtmpfs: unable to create devtmpfs %i\n", err); + pr_err("unable to create devtmpfs %d\n", err); unregister_filesystem(&dev_fs_type); + thread = NULL; return err; } - printk(KERN_INFO "devtmpfs: initialized\n"); + pr_info("initialized\n"); return 0; } diff --git a/drivers/base/driver.c b/drivers/base/driver.c index 8c0d33e182fd..8ab010ddf709 100644 --- a/drivers/base/driver.c +++ b/drivers/base/driver.c @@ -31,6 +31,81 @@ static struct device *next_device(struct klist_iter *i) } /** + * driver_set_override() - Helper to set or clear driver override. + * @dev: Device to change + * @override: Address of string to change (e.g. &device->driver_override); + * The contents will be freed and hold newly allocated override. + * @s: NUL-terminated string, new driver name to force a match, pass empty + * string to clear it ("" or "\n", where the latter is only for sysfs + * interface). + * @len: length of @s + * + * Helper to set or clear driver override in a device, intended for the cases + * when the driver_override field is allocated by driver/bus code. + * + * Returns: 0 on success or a negative error code on failure. + */ +int driver_set_override(struct device *dev, const char **override, + const char *s, size_t len) +{ + const char *new, *old; + char *cp; + + if (!override || !s) + return -EINVAL; + + /* + * The stored value will be used in sysfs show callback (sysfs_emit()), + * which has a length limit of PAGE_SIZE and adds a trailing newline. + * Thus we can store one character less to avoid truncation during sysfs + * show. + */ + if (len >= (PAGE_SIZE - 1)) + return -EINVAL; + + /* + * Compute the real length of the string in case userspace sends us a + * bunch of \0 characters like python likes to do. + */ + len = strlen(s); + + if (!len) { + /* Empty string passed - clear override */ + device_lock(dev); + old = *override; + *override = NULL; + device_unlock(dev); + kfree(old); + + return 0; + } + + cp = strnchr(s, len, '\n'); + if (cp) + len = cp - s; + + new = kstrndup(s, len, GFP_KERNEL); + if (!new) + return -ENOMEM; + + device_lock(dev); + old = *override; + if (cp != s) { + *override = new; + } else { + /* "\n" passed - clear override */ + kfree(new); + *override = NULL; + } + device_unlock(dev); + + kfree(old); + + return 0; +} +EXPORT_SYMBOL_GPL(driver_set_override); + +/** * driver_for_each_device - Iterator for devices bound to a driver. * @drv: Driver we're iterating. * @start: Device to begin with @@ -40,7 +115,7 @@ static struct device *next_device(struct klist_iter *i) * Iterate over the @drv's list of devices calling @fn for each one. */ int driver_for_each_device(struct device_driver *drv, struct device *start, - void *data, int (*fn)(struct device *, void *)) + void *data, device_iter_t fn) { struct klist_iter i; struct device *dev; @@ -73,9 +148,9 @@ EXPORT_SYMBOL_GPL(driver_for_each_device); * if it does. If the callback returns non-zero, this function will * return to the caller and not iterate over any more devices. */ -struct device *driver_find_device(struct device_driver *drv, +struct device *driver_find_device(const struct device_driver *drv, struct device *start, const void *data, - int (*match)(struct device *dev, const void *data)) + device_match_t match) { struct klist_iter i; struct device *dev; @@ -85,9 +160,12 @@ struct device *driver_find_device(struct device_driver *drv, klist_iter_init_node(&drv->p->klist_devices, &i, (start ? &start->p->knode_driver : NULL)); - while ((dev = next_device(&i))) - if (match(dev, data) && get_device(dev)) + while ((dev = next_device(&i))) { + if (match(dev, data)) { + get_device(dev); break; + } + } klist_iter_exit(&i); return dev; } @@ -98,7 +176,7 @@ EXPORT_SYMBOL_GPL(driver_find_device); * @drv: driver. * @attr: driver attribute descriptor. */ -int driver_create_file(struct device_driver *drv, +int driver_create_file(const struct device_driver *drv, const struct driver_attribute *attr) { int error; @@ -116,7 +194,7 @@ EXPORT_SYMBOL_GPL(driver_create_file); * @drv: driver. * @attr: driver attribute descriptor. */ -void driver_remove_file(struct device_driver *drv, +void driver_remove_file(const struct device_driver *drv, const struct driver_attribute *attr) { if (drv) @@ -124,13 +202,13 @@ void driver_remove_file(struct device_driver *drv, } EXPORT_SYMBOL_GPL(driver_remove_file); -int driver_add_groups(struct device_driver *drv, +int driver_add_groups(const struct device_driver *drv, const struct attribute_group **groups) { return sysfs_create_groups(&drv->p->kobj, groups); } -void driver_remove_groups(struct device_driver *drv, +void driver_remove_groups(const struct device_driver *drv, const struct attribute_group **groups) { sysfs_remove_groups(&drv->p->kobj, groups); @@ -149,7 +227,7 @@ int driver_register(struct device_driver *drv) int ret; struct device_driver *other; - if (!drv->bus->p) { + if (!bus_is_registered(drv->bus)) { pr_err("Driver '%s' was unable to register with bus_type '%s' because the bus was not initialized.\n", drv->name, drv->bus->name); return -EINVAL; @@ -177,6 +255,7 @@ int driver_register(struct device_driver *drv) return ret; } kobject_uevent(&drv->p->kobj, KOBJ_ADD); + deferred_probe_extend_timeout(); return ret; } @@ -198,30 +277,3 @@ void driver_unregister(struct device_driver *drv) bus_remove_driver(drv); } EXPORT_SYMBOL_GPL(driver_unregister); - -/** - * driver_find - locate driver on a bus by its name. - * @name: name of the driver. - * @bus: bus to scan for the driver. - * - * Call kset_find_obj() to iterate over list of drivers on - * a bus to find driver by name. Return driver if found. - * - * This routine provides no locking to prevent the driver it returns - * from being unregistered or unloaded while the caller is using it. - * The caller is responsible for preventing this. - */ -struct device_driver *driver_find(const char *name, struct bus_type *bus) -{ - struct kobject *k = kset_find_obj(bus->p->drivers_kset, name); - struct driver_private *priv; - - if (k) { - /* Drop reference added by kset_find_obj() */ - kobject_put(k); - priv = to_driver(k); - return priv->driver; - } - return NULL; -} -EXPORT_SYMBOL_GPL(driver_find); diff --git a/drivers/base/faux.c b/drivers/base/faux.c new file mode 100644 index 000000000000..21dd02124231 --- /dev/null +++ b/drivers/base/faux.c @@ -0,0 +1,261 @@ +// SPDX-License-Identifier: GPL-2.0-only +/* + * Copyright (c) 2025 Greg Kroah-Hartman <gregkh@linuxfoundation.org> + * Copyright (c) 2025 The Linux Foundation + * + * A "simple" faux bus that allows devices to be created and added + * automatically to it. This is to be used whenever you need to create a + * device that is not associated with any "real" system resources, and do + * not want to have to deal with a bus/driver binding logic. It is + * intended to be very simple, with only a create and a destroy function + * available. + */ +#include <linux/err.h> +#include <linux/init.h> +#include <linux/slab.h> +#include <linux/string.h> +#include <linux/container_of.h> +#include <linux/device/faux.h> +#include "base.h" + +/* + * Internal wrapper structure so we can hold a pointer to the + * faux_device_ops for this device. + */ +struct faux_object { + struct faux_device faux_dev; + const struct faux_device_ops *faux_ops; + const struct attribute_group **groups; +}; +#define to_faux_object(dev) container_of_const(dev, struct faux_object, faux_dev.dev) + +static struct device faux_bus_root = { + .init_name = "faux", +}; + +static int faux_match(struct device *dev, const struct device_driver *drv) +{ + /* Match always succeeds, we only have one driver */ + return 1; +} + +static int faux_probe(struct device *dev) +{ + struct faux_object *faux_obj = to_faux_object(dev); + struct faux_device *faux_dev = &faux_obj->faux_dev; + const struct faux_device_ops *faux_ops = faux_obj->faux_ops; + int ret; + + if (faux_ops && faux_ops->probe) { + ret = faux_ops->probe(faux_dev); + if (ret) + return ret; + } + + /* + * Add groups after the probe succeeds to ensure resources are + * initialized correctly + */ + ret = device_add_groups(dev, faux_obj->groups); + if (ret && faux_ops && faux_ops->remove) + faux_ops->remove(faux_dev); + + return ret; +} + +static void faux_remove(struct device *dev) +{ + struct faux_object *faux_obj = to_faux_object(dev); + struct faux_device *faux_dev = &faux_obj->faux_dev; + const struct faux_device_ops *faux_ops = faux_obj->faux_ops; + + device_remove_groups(dev, faux_obj->groups); + + if (faux_ops && faux_ops->remove) + faux_ops->remove(faux_dev); +} + +static const struct bus_type faux_bus_type = { + .name = "faux", + .match = faux_match, + .probe = faux_probe, + .remove = faux_remove, +}; + +static struct device_driver faux_driver = { + .name = "faux_driver", + .bus = &faux_bus_type, + .probe_type = PROBE_FORCE_SYNCHRONOUS, + .suppress_bind_attrs = true, +}; + +static void faux_device_release(struct device *dev) +{ + struct faux_object *faux_obj = to_faux_object(dev); + + kfree(faux_obj); +} + +/** + * faux_device_create_with_groups - Create and register with the driver + * core a faux device and populate the device with an initial + * set of sysfs attributes. + * @name: The name of the device we are adding, must be unique for + * all faux devices. + * @parent: Pointer to a potential parent struct device. If set to + * NULL, the device will be created in the "root" of the faux + * device tree in sysfs. + * @faux_ops: struct faux_device_ops that the new device will call back + * into, can be NULL. + * @groups: The set of sysfs attributes that will be created for this + * device when it is registered with the driver core. + * + * Create a new faux device and register it in the driver core properly. + * If present, callbacks in @faux_ops will be called with the device that + * for the caller to do something with at the proper time given the + * device's lifecycle. + * + * Note, when this function is called, the functions specified in struct + * faux_ops can be called before the function returns, so be prepared for + * everything to be properly initialized before that point in time. If the + * probe callback (if one is present) does NOT succeed, the creation of the + * device will fail and NULL will be returned. + * + * Return: + * * NULL if an error happened with creating the device + * * pointer to a valid struct faux_device that is registered with sysfs + */ +struct faux_device *faux_device_create_with_groups(const char *name, + struct device *parent, + const struct faux_device_ops *faux_ops, + const struct attribute_group **groups) +{ + struct faux_object *faux_obj; + struct faux_device *faux_dev; + struct device *dev; + int ret; + + faux_obj = kzalloc(sizeof(*faux_obj), GFP_KERNEL); + if (!faux_obj) + return NULL; + + /* Save off the callbacks and groups so we can use them in the future */ + faux_obj->faux_ops = faux_ops; + faux_obj->groups = groups; + + /* Initialize the device portion and register it with the driver core */ + faux_dev = &faux_obj->faux_dev; + dev = &faux_dev->dev; + + device_initialize(dev); + dev->release = faux_device_release; + if (parent) + dev->parent = parent; + else + dev->parent = &faux_bus_root; + dev->bus = &faux_bus_type; + dev_set_name(dev, "%s", name); + device_set_pm_not_required(dev); + + ret = device_add(dev); + if (ret) { + pr_err("%s: device_add for faux device '%s' failed with %d\n", + __func__, name, ret); + put_device(dev); + return NULL; + } + + /* + * Verify that we did bind the driver to the device (i.e. probe worked), + * if not, let's fail the creation as trying to guess if probe was + * successful is almost impossible to determine by the caller. + */ + if (!dev->driver) { + dev_dbg(dev, "probe did not succeed, tearing down the device\n"); + faux_device_destroy(faux_dev); + faux_dev = NULL; + } + + return faux_dev; +} +EXPORT_SYMBOL_GPL(faux_device_create_with_groups); + +/** + * faux_device_create - create and register with the driver core a faux device + * @name: The name of the device we are adding, must be unique for all + * faux devices. + * @parent: Pointer to a potential parent struct device. If set to + * NULL, the device will be created in the "root" of the faux + * device tree in sysfs. + * @faux_ops: struct faux_device_ops that the new device will call back + * into, can be NULL. + * + * Create a new faux device and register it in the driver core properly. + * If present, callbacks in @faux_ops will be called with the device that + * for the caller to do something with at the proper time given the + * device's lifecycle. + * + * Note, when this function is called, the functions specified in struct + * faux_ops can be called before the function returns, so be prepared for + * everything to be properly initialized before that point in time. + * + * Return: + * * NULL if an error happened with creating the device + * * pointer to a valid struct faux_device that is registered with sysfs + */ +struct faux_device *faux_device_create(const char *name, + struct device *parent, + const struct faux_device_ops *faux_ops) +{ + return faux_device_create_with_groups(name, parent, faux_ops, NULL); +} +EXPORT_SYMBOL_GPL(faux_device_create); + +/** + * faux_device_destroy - destroy a faux device + * @faux_dev: faux device to destroy + * + * Unregisters and cleans up a device that was created with a call to + * faux_device_create() + */ +void faux_device_destroy(struct faux_device *faux_dev) +{ + struct device *dev = &faux_dev->dev; + + if (!faux_dev) + return; + + device_del(dev); + + /* The final put_device() will clean up the memory we allocated for this device. */ + put_device(dev); +} +EXPORT_SYMBOL_GPL(faux_device_destroy); + +int __init faux_bus_init(void) +{ + int ret; + + ret = device_register(&faux_bus_root); + if (ret) { + put_device(&faux_bus_root); + return ret; + } + + ret = bus_register(&faux_bus_type); + if (ret) + goto error_bus; + + ret = driver_register(&faux_driver); + if (ret) + goto error_driver; + + return ret; + +error_driver: + bus_unregister(&faux_bus_type); + +error_bus: + device_unregister(&faux_bus_root); + return ret; +} diff --git a/drivers/base/firmware_loader/Kconfig b/drivers/base/firmware_loader/Kconfig index 5b24f3959255..15eff8a4b505 100644 --- a/drivers/base/firmware_loader/Kconfig +++ b/drivers/base/firmware_loader/Kconfig @@ -3,6 +3,7 @@ menu "Firmware loader" config FW_LOADER tristate "Firmware loading facility" if EXPERT + select CRYPTO_LIB_SHA256 if FW_LOADER_DEBUG default y help This enables the firmware loading facility in the kernel. The kernel @@ -24,11 +25,31 @@ config FW_LOADER You also want to be sure to enable this built-in if you are going to enable built-in firmware (CONFIG_EXTRA_FIRMWARE). +config FW_LOADER_DEBUG + bool "Log filenames and checksums for loaded firmware" + depends on DYNAMIC_DEBUG + depends on FW_LOADER + default FW_LOADER + help + Select this option to use dynamic debug to log firmware filenames and + SHA256 checksums to the kernel log for each firmware file that is + loaded. + +config RUST_FW_LOADER_ABSTRACTIONS + bool "Rust Firmware Loader abstractions" + depends on RUST + select FW_LOADER + help + This enables the Rust abstractions for the firmware loader API. + if FW_LOADER config FW_LOADER_PAGED_BUF bool +config FW_LOADER_SYSFS + bool + config EXTRA_FIRMWARE string "Build named firmware blobs into the kernel binary" help @@ -60,6 +81,8 @@ config EXTRA_FIRMWARE image since it combines both GPL and non-GPL work. You should consult a lawyer of your own before distributing such an image. + NOTE: Compressed files are not supported in EXTRA_FIRMWARE. + config EXTRA_FIRMWARE_DIR string "Firmware blobs root directory" depends on EXTRA_FIRMWARE != "" @@ -70,6 +93,7 @@ config EXTRA_FIRMWARE_DIR config FW_LOADER_USER_HELPER bool "Enable the firmware sysfs fallback mechanism" + select FW_LOADER_SYSFS select FW_LOADER_PAGED_BUF help This option enables a sysfs loading facility to enable firmware @@ -157,17 +181,33 @@ config FW_LOADER_USER_HELPER_FALLBACK config FW_LOADER_COMPRESS bool "Enable compressed firmware support" - select FW_LOADER_PAGED_BUF - select XZ_DEC help This option enables the support for loading compressed firmware files. The caller of firmware API receives the decompressed file content. The compressed file is loaded as a fallback, only after loading the raw file failed at first. - Currently only XZ-compressed files are supported, and they have to - be compressed with either none or crc32 integrity check type (pass - "-C crc32" option to xz command). + Compressed firmware support does not apply to firmware images + that are built into the kernel image (CONFIG_EXTRA_FIRMWARE). + +if FW_LOADER_COMPRESS +config FW_LOADER_COMPRESS_XZ + bool "Enable XZ-compressed firmware support" + select FW_LOADER_PAGED_BUF + select XZ_DEC + default y + help + This option adds the support for XZ-compressed files. + The files have to be compressed with either none or crc32 + integrity check type (pass "-C crc32" option to xz command). + +config FW_LOADER_COMPRESS_ZSTD + bool "Enable ZSTD-compressed firmware support" + select ZSTD_DECOMPRESS + help + This option adds the support for ZSTD-compressed files. + +endif # FW_LOADER_COMPRESS config FW_CACHE bool "Enable firmware caching during suspend" @@ -181,5 +221,19 @@ config FW_CACHE If unsure, say Y. +config FW_UPLOAD + bool "Enable users to initiate firmware updates using sysfs" + select FW_LOADER_SYSFS + select FW_LOADER_PAGED_BUF + help + Enabling this option will allow device drivers to expose a persistent + sysfs interface that allows firmware updates to be initiated from + userspace. For example, FPGA based PCIe cards load firmware and FPGA + images from local FLASH when the card boots. The images in FLASH may + be updated with new images provided by the user. Enable this device + to support cards that rely on user-initiated updates for firmware files. + + If unsure, say N. + endif # FW_LOADER endmenu diff --git a/drivers/base/firmware_loader/Makefile b/drivers/base/firmware_loader/Makefile index e87843408fe6..60d19f9e0ddc 100644 --- a/drivers/base/firmware_loader/Makefile +++ b/drivers/base/firmware_loader/Makefile @@ -6,5 +6,7 @@ obj-$(CONFIG_FW_LOADER) += firmware_class.o firmware_class-objs := main.o firmware_class-$(CONFIG_FW_LOADER_USER_HELPER) += fallback.o firmware_class-$(CONFIG_EFI_EMBEDDED_FIRMWARE) += fallback_platform.o +firmware_class-$(CONFIG_FW_LOADER_SYSFS) += sysfs.o +firmware_class-$(CONFIG_FW_UPLOAD) += sysfs_upload.o obj-y += builtin/ diff --git a/drivers/base/firmware_loader/builtin/Makefile b/drivers/base/firmware_loader/builtin/Makefile index 101754ad48d9..6c067dedc01e 100644 --- a/drivers/base/firmware_loader/builtin/Makefile +++ b/drivers/base/firmware_loader/builtin/Makefile @@ -1,11 +1,13 @@ # SPDX-License-Identifier: GPL-2.0 +obj-y += main.o # Create $(fwdir) from $(CONFIG_EXTRA_FIRMWARE_DIR) -- if it doesn't have a # leading /, it's relative to $(srctree). -fwdir := $(subst $(quote),,$(CONFIG_EXTRA_FIRMWARE_DIR)) +fwdir := $(CONFIG_EXTRA_FIRMWARE_DIR) fwdir := $(addprefix $(srctree)/,$(filter-out /%,$(fwdir)))$(filter /%,$(fwdir)) -obj-y := $(addsuffix .gen.o, $(subst $(quote),,$(CONFIG_EXTRA_FIRMWARE))) +firmware := $(addsuffix .gen.o, $(CONFIG_EXTRA_FIRMWARE)) +obj-y += $(firmware) FWNAME = $(patsubst $(obj)/%.gen.S,%,$@) FWSTR = $(subst $(comma),_,$(subst /,_,$(subst .,_,$(subst -,_,$(FWNAME))))) @@ -34,7 +36,7 @@ $(obj)/%.gen.S: FORCE $(call filechk,fwbin) # The .o files depend on the binaries directly; the .S files don't. -$(addprefix $(obj)/, $(obj-y)): $(obj)/%.gen.o: $(fwdir)/% +$(addprefix $(obj)/, $(firmware)): $(obj)/%.gen.o: $(fwdir)/% targets := $(patsubst $(obj)/%,%, \ $(shell find $(obj) -name \*.gen.S 2>/dev/null)) diff --git a/drivers/base/firmware_loader/builtin/main.c b/drivers/base/firmware_loader/builtin/main.c new file mode 100644 index 000000000000..d36befebb1b9 --- /dev/null +++ b/drivers/base/firmware_loader/builtin/main.c @@ -0,0 +1,106 @@ +// SPDX-License-Identifier: GPL-2.0 +/* Builtin firmware support */ + +#include <linux/firmware.h> +#include "../firmware.h" + +/* Only if FW_LOADER=y */ +#ifdef CONFIG_FW_LOADER + +struct builtin_fw { + char *name; + void *data; + unsigned long size; +}; + +extern struct builtin_fw __start_builtin_fw[]; +extern struct builtin_fw __end_builtin_fw[]; + +static bool fw_copy_to_prealloc_buf(struct firmware *fw, + void *buf, size_t size) +{ + if (!buf) + return true; + if (size < fw->size) + return false; + memcpy(buf, fw->data, fw->size); + return true; +} + +/** + * firmware_request_builtin() - load builtin firmware + * @fw: pointer to firmware struct + * @name: name of firmware file + * + * Some use cases in the kernel have a requirement so that no memory allocator + * is involved as these calls take place early in boot process. An example is + * the x86 CPU microcode loader. In these cases all the caller wants is to see + * if the firmware was built-in and if so use it right away. This can be used + * for such cases. + * + * This looks for the firmware in the built-in kernel. Only if the kernel was + * built-in with the firmware you are looking for will this return successfully. + * + * Callers of this API do not need to use release_firmware() as the pointer to + * the firmware is expected to be provided locally on the stack of the caller. + **/ +bool firmware_request_builtin(struct firmware *fw, const char *name) +{ + struct builtin_fw *b_fw; + + if (!fw) + return false; + + for (b_fw = __start_builtin_fw; b_fw != __end_builtin_fw; b_fw++) { + if (strcmp(name, b_fw->name) == 0) { + fw->size = b_fw->size; + fw->data = b_fw->data; + return true; + } + } + + return false; +} +EXPORT_SYMBOL_NS_GPL(firmware_request_builtin, "TEST_FIRMWARE"); + +/** + * firmware_request_builtin_buf() - load builtin firmware into optional buffer + * @fw: pointer to firmware struct + * @name: name of firmware file + * @buf: If set this lets you use a pre-allocated buffer so that the built-in + * firmware into is copied into. This field can be NULL. It is used by + * callers such as request_firmware_into_buf() and + * request_partial_firmware_into_buf() + * @size: if buf was provided, the max size of the allocated buffer available. + * If the built-in firmware does not fit into the pre-allocated @buf this + * call will fail. + * + * This looks for the firmware in the built-in kernel. Only if the kernel was + * built-in with the firmware you are looking for will this call possibly + * succeed. If you passed a @buf the firmware will be copied into it *iff* the + * built-in firmware fits into the pre-allocated buffer size specified in + * @size. + * + * This caller is to be used internally by the firmware_loader only. + **/ +bool firmware_request_builtin_buf(struct firmware *fw, const char *name, + void *buf, size_t size) +{ + if (!firmware_request_builtin(fw, name)) + return false; + + return fw_copy_to_prealloc_buf(fw, buf, size); +} + +bool firmware_is_builtin(const struct firmware *fw) +{ + struct builtin_fw *b_fw; + + for (b_fw = __start_builtin_fw; b_fw != __end_builtin_fw; b_fw++) + if (fw->data == b_fw->data) + return true; + + return false; +} + +#endif diff --git a/drivers/base/firmware_loader/fallback.c b/drivers/base/firmware_loader/fallback.c index 91899d185e31..3ef0b312ae71 100644 --- a/drivers/base/firmware_loader/fallback.c +++ b/drivers/base/firmware_loader/fallback.c @@ -3,12 +3,9 @@ #include <linux/types.h> #include <linux/kconfig.h> #include <linux/list.h> -#include <linux/slab.h> #include <linux/security.h> -#include <linux/highmem.h> #include <linux/umh.h> #include <linux/sysctl.h> -#include <linux/vmalloc.h> #include <linux/module.h> #include "fallback.h" @@ -18,22 +15,6 @@ * firmware fallback mechanism */ -MODULE_IMPORT_NS(FIRMWARE_LOADER_PRIVATE); - -extern struct firmware_fallback_config fw_fallback_config; - -/* These getters are vetted to use int properly */ -static inline int __firmware_loading_timeout(void) -{ - return fw_fallback_config.loading_timeout; -} - -/* These setters are vetted to use int properly */ -static void __fw_fallback_set_timeout(int timeout) -{ - fw_fallback_config.loading_timeout = timeout; -} - /* * use small loading timeout for caching devices' firmware because all these * firmware images have been loaded successfully at lease once, also system is @@ -58,56 +39,14 @@ static long firmware_loading_timeout(void) __firmware_loading_timeout() * HZ : MAX_JIFFY_OFFSET; } -static inline bool fw_sysfs_done(struct fw_priv *fw_priv) -{ - return __fw_state_check(fw_priv, FW_STATUS_DONE); -} - -static inline bool fw_sysfs_loading(struct fw_priv *fw_priv) -{ - return __fw_state_check(fw_priv, FW_STATUS_LOADING); -} - static inline int fw_sysfs_wait_timeout(struct fw_priv *fw_priv, long timeout) { return __fw_state_wait_common(fw_priv, timeout); } -struct fw_sysfs { - bool nowait; - struct device dev; - struct fw_priv *fw_priv; - struct firmware *fw; -}; - -static struct fw_sysfs *to_fw_sysfs(struct device *dev) -{ - return container_of(dev, struct fw_sysfs, dev); -} - -static void __fw_load_abort(struct fw_priv *fw_priv) -{ - /* - * There is a small window in which user can write to 'loading' - * between loading done and disappearance of 'loading' - */ - if (fw_sysfs_done(fw_priv)) - return; - - list_del_init(&fw_priv->pending_list); - fw_state_aborted(fw_priv); -} - -static void fw_load_abort(struct fw_sysfs *fw_sysfs) -{ - struct fw_priv *fw_priv = fw_sysfs->fw_priv; - - __fw_load_abort(fw_priv); -} - static LIST_HEAD(pending_fw_head); -void kill_pending_fw_fallback_reqs(bool only_kill_custom) +void kill_pending_fw_fallback_reqs(bool kill_all) { struct fw_priv *fw_priv; struct fw_priv *next; @@ -115,376 +54,14 @@ void kill_pending_fw_fallback_reqs(bool only_kill_custom) mutex_lock(&fw_lock); list_for_each_entry_safe(fw_priv, next, &pending_fw_head, pending_list) { - if (!fw_priv->need_uevent || !only_kill_custom) + if (kill_all || !fw_priv->need_uevent) __fw_load_abort(fw_priv); } - mutex_unlock(&fw_lock); -} - -static ssize_t timeout_show(struct class *class, struct class_attribute *attr, - char *buf) -{ - return sysfs_emit(buf, "%d\n", __firmware_loading_timeout()); -} - -/** - * timeout_store() - set number of seconds to wait for firmware - * @class: device class pointer - * @attr: device attribute pointer - * @buf: buffer to scan for timeout value - * @count: number of bytes in @buf - * - * Sets the number of seconds to wait for the firmware. Once - * this expires an error will be returned to the driver and no - * firmware will be provided. - * - * Note: zero means 'wait forever'. - **/ -static ssize_t timeout_store(struct class *class, struct class_attribute *attr, - const char *buf, size_t count) -{ - int tmp_loading_timeout = simple_strtol(buf, NULL, 10); - - if (tmp_loading_timeout < 0) - tmp_loading_timeout = 0; - - __fw_fallback_set_timeout(tmp_loading_timeout); - - return count; -} -static CLASS_ATTR_RW(timeout); - -static struct attribute *firmware_class_attrs[] = { - &class_attr_timeout.attr, - NULL, -}; -ATTRIBUTE_GROUPS(firmware_class); - -static void fw_dev_release(struct device *dev) -{ - struct fw_sysfs *fw_sysfs = to_fw_sysfs(dev); - - kfree(fw_sysfs); -} - -static int do_firmware_uevent(struct fw_sysfs *fw_sysfs, struct kobj_uevent_env *env) -{ - if (add_uevent_var(env, "FIRMWARE=%s", fw_sysfs->fw_priv->fw_name)) - return -ENOMEM; - if (add_uevent_var(env, "TIMEOUT=%i", __firmware_loading_timeout())) - return -ENOMEM; - if (add_uevent_var(env, "ASYNC=%d", fw_sysfs->nowait)) - return -ENOMEM; - - return 0; -} - -static int firmware_uevent(struct device *dev, struct kobj_uevent_env *env) -{ - struct fw_sysfs *fw_sysfs = to_fw_sysfs(dev); - int err = 0; - - mutex_lock(&fw_lock); - if (fw_sysfs->fw_priv) - err = do_firmware_uevent(fw_sysfs, env); - mutex_unlock(&fw_lock); - return err; -} - -static struct class firmware_class = { - .name = "firmware", - .class_groups = firmware_class_groups, - .dev_uevent = firmware_uevent, - .dev_release = fw_dev_release, -}; - -int register_sysfs_loader(void) -{ - return class_register(&firmware_class); -} - -void unregister_sysfs_loader(void) -{ - class_unregister(&firmware_class); -} - -static ssize_t firmware_loading_show(struct device *dev, - struct device_attribute *attr, char *buf) -{ - struct fw_sysfs *fw_sysfs = to_fw_sysfs(dev); - int loading = 0; - - mutex_lock(&fw_lock); - if (fw_sysfs->fw_priv) - loading = fw_sysfs_loading(fw_sysfs->fw_priv); - mutex_unlock(&fw_lock); - - return sysfs_emit(buf, "%d\n", loading); -} - -/** - * firmware_loading_store() - set value in the 'loading' control file - * @dev: device pointer - * @attr: device attribute pointer - * @buf: buffer to scan for loading control value - * @count: number of bytes in @buf - * - * The relevant values are: - * - * 1: Start a load, discarding any previous partial load. - * 0: Conclude the load and hand the data to the driver code. - * -1: Conclude the load with an error and discard any written data. - **/ -static ssize_t firmware_loading_store(struct device *dev, - struct device_attribute *attr, - const char *buf, size_t count) -{ - struct fw_sysfs *fw_sysfs = to_fw_sysfs(dev); - struct fw_priv *fw_priv; - ssize_t written = count; - int loading = simple_strtol(buf, NULL, 10); - - mutex_lock(&fw_lock); - fw_priv = fw_sysfs->fw_priv; - if (fw_state_is_aborted(fw_priv)) - goto out; - - switch (loading) { - case 1: - /* discarding any previous partial load */ - if (!fw_sysfs_done(fw_priv)) { - fw_free_paged_buf(fw_priv); - fw_state_start(fw_priv); - } - break; - case 0: - if (fw_sysfs_loading(fw_priv)) { - int rc; - - /* - * Several loading requests may be pending on - * one same firmware buf, so let all requests - * see the mapped 'buf->data' once the loading - * is completed. - * */ - rc = fw_map_paged_buf(fw_priv); - if (rc) - dev_err(dev, "%s: map pages failed\n", - __func__); - else - rc = security_kernel_post_load_data(fw_priv->data, - fw_priv->size, - LOADING_FIRMWARE, "blob"); - - /* - * Same logic as fw_load_abort, only the DONE bit - * is ignored and we set ABORT only on failure. - */ - list_del_init(&fw_priv->pending_list); - if (rc) { - fw_state_aborted(fw_priv); - written = rc; - } else { - fw_state_done(fw_priv); - } - break; - } - fallthrough; - default: - dev_err(dev, "%s: unexpected value (%d)\n", __func__, loading); - fallthrough; - case -1: - fw_load_abort(fw_sysfs); - break; - } -out: - mutex_unlock(&fw_lock); - return written; -} -static DEVICE_ATTR(loading, 0644, firmware_loading_show, firmware_loading_store); - -static void firmware_rw_data(struct fw_priv *fw_priv, char *buffer, - loff_t offset, size_t count, bool read) -{ - if (read) - memcpy(buffer, fw_priv->data + offset, count); - else - memcpy(fw_priv->data + offset, buffer, count); -} + if (kill_all) + fw_load_abort_all = true; -static void firmware_rw(struct fw_priv *fw_priv, char *buffer, - loff_t offset, size_t count, bool read) -{ - while (count) { - void *page_data; - int page_nr = offset >> PAGE_SHIFT; - int page_ofs = offset & (PAGE_SIZE-1); - int page_cnt = min_t(size_t, PAGE_SIZE - page_ofs, count); - - page_data = kmap(fw_priv->pages[page_nr]); - - if (read) - memcpy(buffer, page_data + page_ofs, page_cnt); - else - memcpy(page_data + page_ofs, buffer, page_cnt); - - kunmap(fw_priv->pages[page_nr]); - buffer += page_cnt; - offset += page_cnt; - count -= page_cnt; - } -} - -static ssize_t firmware_data_read(struct file *filp, struct kobject *kobj, - struct bin_attribute *bin_attr, - char *buffer, loff_t offset, size_t count) -{ - struct device *dev = kobj_to_dev(kobj); - struct fw_sysfs *fw_sysfs = to_fw_sysfs(dev); - struct fw_priv *fw_priv; - ssize_t ret_count; - - mutex_lock(&fw_lock); - fw_priv = fw_sysfs->fw_priv; - if (!fw_priv || fw_sysfs_done(fw_priv)) { - ret_count = -ENODEV; - goto out; - } - if (offset > fw_priv->size) { - ret_count = 0; - goto out; - } - if (count > fw_priv->size - offset) - count = fw_priv->size - offset; - - ret_count = count; - - if (fw_priv->data) - firmware_rw_data(fw_priv, buffer, offset, count, true); - else - firmware_rw(fw_priv, buffer, offset, count, true); - -out: mutex_unlock(&fw_lock); - return ret_count; -} - -static int fw_realloc_pages(struct fw_sysfs *fw_sysfs, int min_size) -{ - int err; - - err = fw_grow_paged_buf(fw_sysfs->fw_priv, - PAGE_ALIGN(min_size) >> PAGE_SHIFT); - if (err) - fw_load_abort(fw_sysfs); - return err; -} - -/** - * firmware_data_write() - write method for firmware - * @filp: open sysfs file - * @kobj: kobject for the device - * @bin_attr: bin_attr structure - * @buffer: buffer being written - * @offset: buffer offset for write in total data store area - * @count: buffer size - * - * Data written to the 'data' attribute will be later handed to - * the driver as a firmware image. - **/ -static ssize_t firmware_data_write(struct file *filp, struct kobject *kobj, - struct bin_attribute *bin_attr, - char *buffer, loff_t offset, size_t count) -{ - struct device *dev = kobj_to_dev(kobj); - struct fw_sysfs *fw_sysfs = to_fw_sysfs(dev); - struct fw_priv *fw_priv; - ssize_t retval; - - if (!capable(CAP_SYS_RAWIO)) - return -EPERM; - - mutex_lock(&fw_lock); - fw_priv = fw_sysfs->fw_priv; - if (!fw_priv || fw_sysfs_done(fw_priv)) { - retval = -ENODEV; - goto out; - } - - if (fw_priv->data) { - if (offset + count > fw_priv->allocated_size) { - retval = -ENOMEM; - goto out; - } - firmware_rw_data(fw_priv, buffer, offset, count, false); - retval = count; - } else { - retval = fw_realloc_pages(fw_sysfs, offset + count); - if (retval) - goto out; - - retval = count; - firmware_rw(fw_priv, buffer, offset, count, false); - } - - fw_priv->size = max_t(size_t, offset + count, fw_priv->size); -out: - mutex_unlock(&fw_lock); - return retval; -} - -static struct bin_attribute firmware_attr_data = { - .attr = { .name = "data", .mode = 0644 }, - .size = 0, - .read = firmware_data_read, - .write = firmware_data_write, -}; - -static struct attribute *fw_dev_attrs[] = { - &dev_attr_loading.attr, - NULL -}; - -static struct bin_attribute *fw_dev_bin_attrs[] = { - &firmware_attr_data, - NULL -}; - -static const struct attribute_group fw_dev_attr_group = { - .attrs = fw_dev_attrs, - .bin_attrs = fw_dev_bin_attrs, -}; - -static const struct attribute_group *fw_dev_attr_groups[] = { - &fw_dev_attr_group, - NULL -}; - -static struct fw_sysfs * -fw_create_instance(struct firmware *firmware, const char *fw_name, - struct device *device, u32 opt_flags) -{ - struct fw_sysfs *fw_sysfs; - struct device *f_dev; - - fw_sysfs = kzalloc(sizeof(*fw_sysfs), GFP_KERNEL); - if (!fw_sysfs) { - fw_sysfs = ERR_PTR(-ENOMEM); - goto exit; - } - - fw_sysfs->nowait = !!(opt_flags & FW_OPT_NOWAIT); - fw_sysfs->fw = firmware; - f_dev = &fw_sysfs->dev; - - device_initialize(f_dev); - dev_set_name(f_dev, "%s", fw_name); - f_dev->parent = device; - f_dev->class = &firmware_class; - f_dev->groups = fw_dev_attr_groups; -exit: - return fw_sysfs; } /** @@ -513,6 +90,11 @@ static int fw_load_sysfs_fallback(struct fw_sysfs *fw_sysfs, long timeout) } mutex_lock(&fw_lock); + if (fw_load_abort_all || fw_state_is_aborted(fw_priv)) { + mutex_unlock(&fw_lock); + retval = -EINTR; + goto out; + } list_add(&fw_priv->pending_list, &pending_fw_head); mutex_unlock(&fw_lock); @@ -535,11 +117,10 @@ static int fw_load_sysfs_fallback(struct fw_sysfs *fw_sysfs, long timeout) if (fw_state_is_aborted(fw_priv)) { if (retval == -ERESTARTSYS) retval = -EINTR; - else - retval = -EAGAIN; } else if (fw_priv->is_paged_buf && !fw_priv->data) retval = -ENOMEM; +out: device_del(f_dev); err_put_dev: put_device(f_dev); diff --git a/drivers/base/firmware_loader/fallback.h b/drivers/base/firmware_loader/fallback.h index 3af7205b302f..ccf912bef6ca 100644 --- a/drivers/base/firmware_loader/fallback.h +++ b/drivers/base/firmware_loader/fallback.h @@ -6,42 +6,18 @@ #include <linux/device.h> #include "firmware.h" - -/** - * struct firmware_fallback_config - firmware fallback configuration settings - * - * Helps describe and fine tune the fallback mechanism. - * - * @force_sysfs_fallback: force the sysfs fallback mechanism to be used - * as if one had enabled CONFIG_FW_LOADER_USER_HELPER_FALLBACK=y. - * Useful to help debug a CONFIG_FW_LOADER_USER_HELPER_FALLBACK=y - * functionality on a kernel where that config entry has been disabled. - * @ignore_sysfs_fallback: force to disable the sysfs fallback mechanism. - * This emulates the behaviour as if we had set the kernel - * config CONFIG_FW_LOADER_USER_HELPER=n. - * @old_timeout: for internal use - * @loading_timeout: the timeout to wait for the fallback mechanism before - * giving up, in seconds. - */ -struct firmware_fallback_config { - unsigned int force_sysfs_fallback; - unsigned int ignore_sysfs_fallback; - int old_timeout; - int loading_timeout; -}; +#include "sysfs.h" #ifdef CONFIG_FW_LOADER_USER_HELPER int firmware_fallback_sysfs(struct firmware *fw, const char *name, struct device *device, u32 opt_flags, int ret); -void kill_pending_fw_fallback_reqs(bool only_kill_custom); +void kill_pending_fw_fallback_reqs(bool kill_all); void fw_fallback_set_cache_timeout(void); void fw_fallback_set_default_timeout(void); -int register_sysfs_loader(void); -void unregister_sysfs_loader(void); #else /* CONFIG_FW_LOADER_USER_HELPER */ static inline int firmware_fallback_sysfs(struct firmware *fw, const char *name, struct device *device, @@ -52,18 +28,9 @@ static inline int firmware_fallback_sysfs(struct firmware *fw, const char *name, return ret; } -static inline void kill_pending_fw_fallback_reqs(bool only_kill_custom) { } +static inline void kill_pending_fw_fallback_reqs(bool kill_all) { } static inline void fw_fallback_set_cache_timeout(void) { } static inline void fw_fallback_set_default_timeout(void) { } - -static inline int register_sysfs_loader(void) -{ - return 0; -} - -static inline void unregister_sysfs_loader(void) -{ -} #endif /* CONFIG_FW_LOADER_USER_HELPER */ #ifdef CONFIG_EFI_EMBEDDED_FIRMWARE diff --git a/drivers/base/firmware_loader/fallback_table.c b/drivers/base/firmware_loader/fallback_table.c index 46a731dede6f..c8afc501a8a4 100644 --- a/drivers/base/firmware_loader/fallback_table.c +++ b/drivers/base/firmware_loader/fallback_table.c @@ -4,6 +4,7 @@ #include <linux/kconfig.h> #include <linux/list.h> #include <linux/slab.h> +#include <linux/export.h> #include <linux/security.h> #include <linux/highmem.h> #include <linux/umh.h> @@ -21,10 +22,10 @@ struct firmware_fallback_config fw_fallback_config = { .loading_timeout = 60, .old_timeout = 60, }; -EXPORT_SYMBOL_NS_GPL(fw_fallback_config, FIRMWARE_LOADER_PRIVATE); +EXPORT_SYMBOL_NS_GPL(fw_fallback_config, "FIRMWARE_LOADER_PRIVATE"); #ifdef CONFIG_SYSCTL -struct ctl_table firmware_config_table[] = { +static const struct ctl_table firmware_config_table[] = { { .procname = "force_sysfs_fallback", .data = &fw_fallback_config.force_sysfs_fallback, @@ -43,6 +44,25 @@ struct ctl_table firmware_config_table[] = { .extra1 = SYSCTL_ZERO, .extra2 = SYSCTL_ONE, }, - { } }; -#endif + +static struct ctl_table_header *firmware_config_sysct_table_header; +int register_firmware_config_sysctl(void) +{ + firmware_config_sysct_table_header = + register_sysctl("kernel/firmware_config", + firmware_config_table); + if (!firmware_config_sysct_table_header) + return -ENOMEM; + return 0; +} +EXPORT_SYMBOL_NS_GPL(register_firmware_config_sysctl, "FIRMWARE_LOADER_PRIVATE"); + +void unregister_firmware_config_sysctl(void) +{ + unregister_sysctl_table(firmware_config_sysct_table_header); + firmware_config_sysct_table_header = NULL; +} +EXPORT_SYMBOL_NS_GPL(unregister_firmware_config_sysctl, "FIRMWARE_LOADER_PRIVATE"); + +#endif /* CONFIG_SYSCTL */ diff --git a/drivers/base/firmware_loader/firmware.h b/drivers/base/firmware_loader/firmware.h index 63bd29fdcb9c..e891742ba264 100644 --- a/drivers/base/firmware_loader/firmware.h +++ b/drivers/base/firmware_loader/firmware.h @@ -9,8 +9,6 @@ #include <linux/list.h> #include <linux/completion.h> -#include <generated/utsrelease.h> - /** * enum fw_opt - options to control firmware loading behaviour * @@ -87,6 +85,8 @@ struct fw_priv { }; extern struct mutex fw_lock; +extern struct firmware_cache fw_cache; +extern bool fw_load_abort_all; static inline bool __fw_state_check(struct fw_priv *fw_priv, enum fw_status status) @@ -117,8 +117,16 @@ static inline void __fw_state_set(struct fw_priv *fw_priv, WRITE_ONCE(fw_st->status, status); - if (status == FW_STATUS_DONE || status == FW_STATUS_ABORTED) + if (status == FW_STATUS_DONE || status == FW_STATUS_ABORTED) { +#ifdef CONFIG_FW_LOADER_USER_HELPER + /* + * Doing this here ensures that the fw_priv is deleted from + * the pending list in all abort/done paths. + */ + list_del_init(&fw_priv->pending_list); +#endif complete_all(&fw_st->completion); + } } static inline void fw_state_aborted(struct fw_priv *fw_priv) @@ -141,7 +149,39 @@ static inline void fw_state_done(struct fw_priv *fw_priv) __fw_state_set(fw_priv, FW_STATUS_DONE); } +static inline bool fw_state_is_done(struct fw_priv *fw_priv) +{ + return __fw_state_check(fw_priv, FW_STATUS_DONE); +} + +static inline bool fw_state_is_loading(struct fw_priv *fw_priv) +{ + return __fw_state_check(fw_priv, FW_STATUS_LOADING); +} + +int alloc_lookup_fw_priv(const char *fw_name, struct firmware_cache *fwc, + struct fw_priv **fw_priv, void *dbuf, size_t size, + size_t offset, u32 opt_flags); int assign_fw(struct firmware *fw, struct device *device); +void free_fw_priv(struct fw_priv *fw_priv); +void fw_state_init(struct fw_priv *fw_priv); + +#ifdef CONFIG_FW_LOADER +bool firmware_is_builtin(const struct firmware *fw); +bool firmware_request_builtin_buf(struct firmware *fw, const char *name, + void *buf, size_t size); +#else /* module case */ +static inline bool firmware_is_builtin(const struct firmware *fw) +{ + return false; +} +static inline bool firmware_request_builtin_buf(struct firmware *fw, + const char *name, + void *buf, size_t size) +{ + return false; +} +#endif #ifdef CONFIG_FW_LOADER_PAGED_BUF void fw_free_paged_buf(struct fw_priv *fw_priv); diff --git a/drivers/base/firmware_loader/main.c b/drivers/base/firmware_loader/main.c index 4fdb8219cd08..4ebdca9e4da4 100644 --- a/drivers/base/firmware_loader/main.c +++ b/drivers/base/firmware_loader/main.c @@ -35,6 +35,7 @@ #include <linux/syscore_ops.h> #include <linux/reboot.h> #include <linux/security.h> +#include <linux/zstd.h> #include <linux/xz.h> #include <generated/utsrelease.h> @@ -91,68 +92,10 @@ static inline struct fw_priv *to_fw_priv(struct kref *ref) * guarding for corner cases a global lock should be OK */ DEFINE_MUTEX(fw_lock); -static struct firmware_cache fw_cache; +struct firmware_cache fw_cache; +bool fw_load_abort_all; -/* Builtin firmware support */ - -#ifdef CONFIG_FW_LOADER - -extern struct builtin_fw __start_builtin_fw[]; -extern struct builtin_fw __end_builtin_fw[]; - -static void fw_copy_to_prealloc_buf(struct firmware *fw, - void *buf, size_t size) -{ - if (!buf || size < fw->size) - return; - memcpy(buf, fw->data, fw->size); -} - -static bool fw_get_builtin_firmware(struct firmware *fw, const char *name, - void *buf, size_t size) -{ - struct builtin_fw *b_fw; - - for (b_fw = __start_builtin_fw; b_fw != __end_builtin_fw; b_fw++) { - if (strcmp(name, b_fw->name) == 0) { - fw->size = b_fw->size; - fw->data = b_fw->data; - fw_copy_to_prealloc_buf(fw, buf, size); - - return true; - } - } - - return false; -} - -static bool fw_is_builtin_firmware(const struct firmware *fw) -{ - struct builtin_fw *b_fw; - - for (b_fw = __start_builtin_fw; b_fw != __end_builtin_fw; b_fw++) - if (fw->data == b_fw->data) - return true; - - return false; -} - -#else /* Module case - no builtin firmware support */ - -static inline bool fw_get_builtin_firmware(struct firmware *fw, - const char *name, void *buf, - size_t size) -{ - return false; -} - -static inline bool fw_is_builtin_firmware(const struct firmware *fw) -{ - return false; -} -#endif - -static void fw_state_init(struct fw_priv *fw_priv) +void fw_state_init(struct fw_priv *fw_priv) { struct fw_state *fw_st = &fw_priv->fw_st; @@ -165,7 +108,7 @@ static inline int fw_state_wait(struct fw_priv *fw_priv) return __fw_state_wait_common(fw_priv, MAX_SCHEDULE_TIMEOUT); } -static int fw_cache_piggyback_on_request(const char *name); +static void fw_cache_piggyback_on_request(struct fw_priv *fw_priv); static struct fw_priv *__allocate_fw_priv(const char *fw_name, struct firmware_cache *fwc, @@ -222,13 +165,9 @@ static struct fw_priv *__lookup_fw_priv(const char *fw_name) } /* Returns 1 for batching firmware requests with the same name */ -static int alloc_lookup_fw_priv(const char *fw_name, - struct firmware_cache *fwc, - struct fw_priv **fw_priv, - void *dbuf, - size_t size, - size_t offset, - u32 opt_flags) +int alloc_lookup_fw_priv(const char *fw_name, struct firmware_cache *fwc, + struct fw_priv **fw_priv, void *dbuf, size_t size, + size_t offset, u32 opt_flags) { struct fw_priv *tmp; @@ -283,7 +222,7 @@ static void __free_fw_priv(struct kref *ref) kfree(fw_priv); } -static void free_fw_priv(struct fw_priv *fw_priv) +void free_fw_priv(struct fw_priv *fw_priv) { struct firmware_cache *fwc = fw_priv->fwc; spin_lock(&fwc->lock); @@ -312,6 +251,8 @@ void fw_free_paged_buf(struct fw_priv *fw_priv) fw_priv->pages = NULL; fw_priv->page_array_size = 0; fw_priv->nr_pages = 0; + fw_priv->data = NULL; + fw_priv->size = 0; } int fw_grow_paged_buf(struct fw_priv *fw_priv, int pages_needed) @@ -364,9 +305,73 @@ int fw_map_paged_buf(struct fw_priv *fw_priv) #endif /* + * ZSTD-compressed firmware support + */ +#ifdef CONFIG_FW_LOADER_COMPRESS_ZSTD +static int fw_decompress_zstd(struct device *dev, struct fw_priv *fw_priv, + size_t in_size, const void *in_buffer) +{ + size_t len, out_size, workspace_size; + void *workspace, *out_buf; + zstd_dctx *ctx; + int err; + + if (fw_priv->allocated_size) { + out_size = fw_priv->allocated_size; + out_buf = fw_priv->data; + } else { + zstd_frame_header params; + + if (zstd_get_frame_header(¶ms, in_buffer, in_size) || + params.frameContentSize == ZSTD_CONTENTSIZE_UNKNOWN) { + dev_dbg(dev, "%s: invalid zstd header\n", __func__); + return -EINVAL; + } + out_size = params.frameContentSize; + out_buf = vzalloc(out_size); + if (!out_buf) + return -ENOMEM; + } + + workspace_size = zstd_dctx_workspace_bound(); + workspace = kvzalloc(workspace_size, GFP_KERNEL); + if (!workspace) { + err = -ENOMEM; + goto error; + } + + ctx = zstd_init_dctx(workspace, workspace_size); + if (!ctx) { + dev_dbg(dev, "%s: failed to initialize context\n", __func__); + err = -EINVAL; + goto error; + } + + len = zstd_decompress_dctx(ctx, out_buf, out_size, in_buffer, in_size); + if (zstd_is_error(len)) { + dev_dbg(dev, "%s: failed to decompress: %d\n", __func__, + zstd_get_error_code(len)); + err = -EINVAL; + goto error; + } + + if (!fw_priv->allocated_size) + fw_priv->data = out_buf; + fw_priv->size = len; + err = 0; + + error: + kvfree(workspace); + if (err && !fw_priv->allocated_size) + vfree(out_buf); + return err; +} +#endif /* CONFIG_FW_LOADER_COMPRESS_ZSTD */ + +/* * XZ-compressed firmware support */ -#ifdef CONFIG_FW_LOADER_COMPRESS +#ifdef CONFIG_FW_LOADER_COMPRESS_XZ /* show an error and return the standard error code */ static int fw_decompress_xz_error(struct device *dev, enum xz_ret xz_ret) { @@ -431,11 +436,11 @@ static int fw_decompress_xz_pages(struct device *dev, struct fw_priv *fw_priv, /* decompress onto the new allocated page */ page = fw_priv->pages[fw_priv->nr_pages - 1]; - xz_buf.out = kmap(page); + xz_buf.out = kmap_local_page(page); xz_buf.out_pos = 0; xz_buf.out_size = PAGE_SIZE; xz_ret = xz_dec_run(xz_dec, &xz_buf); - kunmap(page); + kunmap_local(xz_buf.out); fw_priv->size += xz_buf.out_pos; /* partial decompression means either end or error */ if (xz_buf.out_pos != PAGE_SIZE) @@ -460,7 +465,7 @@ static int fw_decompress_xz(struct device *dev, struct fw_priv *fw_priv, else return fw_decompress_xz_pages(dev, fw_priv, in_size, in_buffer); } -#endif /* CONFIG_FW_LOADER_COMPRESS */ +#endif /* CONFIG_FW_LOADER_COMPRESS_XZ */ /* direct firmware loading support */ static char fw_path_para[256]; @@ -489,9 +494,9 @@ fw_get_filesystem_firmware(struct device *device, struct fw_priv *fw_priv, const void *in_buffer)) { size_t size; - int i, len; + int i, len, maxlen = 0; int rc = -ENOENT; - char *path; + char *path, *nt = NULL; size_t msize = INT_MAX; void *buffer = NULL; @@ -514,8 +519,17 @@ fw_get_filesystem_firmware(struct device *device, struct fw_priv *fw_priv, if (!fw_path[i][0]) continue; - len = snprintf(path, PATH_MAX, "%s/%s%s", - fw_path[i], fw_priv->fw_name, suffix); + /* strip off \n from customized path */ + maxlen = strlen(fw_path[i]); + if (i == 0) { + nt = strchr(fw_path[i], '\n'); + if (nt) + maxlen = nt - fw_path[i]; + } + + len = snprintf(path, PATH_MAX, "%.*s/%s%s", + maxlen, fw_path[i], + fw_priv->fw_name, suffix); if (len >= PATH_MAX) { rc = -ENAMETOOLONG; break; @@ -537,12 +551,16 @@ fw_get_filesystem_firmware(struct device *device, struct fw_priv *fw_priv, file_size_ptr, READING_FIRMWARE); if (rc < 0) { - if (rc != -ENOENT) - dev_warn(device, "loading %s failed with error %d\n", - path, rc); - else - dev_dbg(device, "loading %s failed for no such file or directory.\n", - path); + if (!(fw_priv->opt_flags & FW_OPT_NO_WARN)) { + if (rc != -ENOENT) + dev_warn(device, + "loading %s failed with error %d\n", + path, rc); + else + dev_dbg(device, + "loading %s failed for no such file or directory.\n", + path); + } continue; } size = rc; @@ -707,10 +725,8 @@ int assign_fw(struct firmware *fw, struct device *device) * on request firmware. */ if (!(fw_priv->opt_flags & FW_OPT_NOCACHE) && - fw_priv->fwc->state == FW_LOADER_START_CACHE) { - if (fw_cache_piggyback_on_request(fw_priv->fw_name)) - kref_get(&fw_priv->ref); - } + fw_priv->fwc->state == FW_LOADER_START_CACHE) + fw_cache_piggyback_on_request(fw_priv); /* pass the pages buffer to driver at the last minute */ fw_set_page_data(fw_priv, fw); @@ -738,7 +754,7 @@ _request_firmware_prepare(struct firmware **firmware_p, const char *name, return -ENOMEM; } - if (fw_get_builtin_firmware(firmware, name, dbuf, size)) { + if (firmware_request_builtin_buf(firmware, name, dbuf, size)) { dev_dbg(device, "using built-in %s\n", name); return 0; /* assigned */ } @@ -783,9 +799,28 @@ static void fw_abort_batch_reqs(struct firmware *fw) return; fw_priv = fw->priv; + mutex_lock(&fw_lock); if (!fw_state_is_aborted(fw_priv)) fw_state_aborted(fw_priv); + mutex_unlock(&fw_lock); +} + +#if defined(CONFIG_FW_LOADER_DEBUG) +#include <crypto/sha2.h> + +static void fw_log_firmware_info(const struct firmware *fw, const char *name, struct device *device) +{ + u8 digest[SHA256_DIGEST_SIZE]; + + sha256(fw->data, fw->size, digest); + dev_dbg(device, "Loaded FW: %s, sha256: %*phN\n", + name, SHA256_DIGEST_SIZE, digest); } +#else +static void fw_log_firmware_info(const struct firmware *fw, const char *name, + struct device *device) +{} +#endif /* called from request_firmware() and request_firmware_work_func() */ static int @@ -805,41 +840,75 @@ _request_firmware(const struct firmware **firmware_p, const char *name, goto out; } + + /* + * Reject firmware file names with ".." path components. + * There are drivers that construct firmware file names from + * device-supplied strings, and we don't want some device to be + * able to tell us "I would like to be sent my firmware from + * ../../../etc/shadow, please". + * + * This intentionally only looks at the firmware name, not at + * the firmware base directory or at symlink contents. + */ + if (name_contains_dotdot(name)) { + dev_warn(device, + "Firmware load for '%s' refused, path contains '..' component\n", + name); + ret = -EINVAL; + goto out; + } + ret = _request_firmware_prepare(&fw, name, device, buf, size, offset, opt_flags); if (ret <= 0) /* error or already assigned */ goto out; - ret = fw_get_filesystem_firmware(device, fw->priv, "", NULL); + /* + * We are about to try to access the firmware file. Because we may have been + * called by a driver when serving an unrelated request from userland, we use + * the kernel credentials to read the file. + */ + scoped_with_kernel_creds() { + ret = fw_get_filesystem_firmware(device, fw->priv, "", NULL); - /* Only full reads can support decompression, platform, and sysfs. */ - if (!(opt_flags & FW_OPT_PARTIAL)) - nondirect = true; + /* Only full reads can support decompression, platform, and sysfs. */ + if (!(opt_flags & FW_OPT_PARTIAL)) + nondirect = true; -#ifdef CONFIG_FW_LOADER_COMPRESS - if (ret == -ENOENT && nondirect) - ret = fw_get_filesystem_firmware(device, fw->priv, ".xz", - fw_decompress_xz); +#ifdef CONFIG_FW_LOADER_COMPRESS_ZSTD + if (ret == -ENOENT && nondirect) + ret = fw_get_filesystem_firmware(device, fw->priv, ".zst", + fw_decompress_zstd); +#endif +#ifdef CONFIG_FW_LOADER_COMPRESS_XZ + if (ret == -ENOENT && nondirect) + ret = fw_get_filesystem_firmware(device, fw->priv, ".xz", + fw_decompress_xz); #endif - if (ret == -ENOENT && nondirect) - ret = firmware_fallback_platform(fw->priv); + if (ret == -ENOENT && nondirect) + ret = firmware_fallback_platform(fw->priv); - if (ret) { - if (!(opt_flags & FW_OPT_NO_WARN)) - dev_warn(device, - "Direct firmware load for %s failed with error %d\n", - name, ret); - if (nondirect) - ret = firmware_fallback_sysfs(fw, name, device, - opt_flags, ret); - } else - ret = assign_fw(fw, device); + if (ret) { + if (!(opt_flags & FW_OPT_NO_WARN)) + dev_warn(device, + "Direct firmware load for %s failed with error %d\n", + name, ret); + if (nondirect) + ret = firmware_fallback_sysfs(fw, name, device, + opt_flags, ret); + } else { + ret = assign_fw(fw, device); + } + } - out: +out: if (ret < 0) { fw_abort_batch_reqs(fw); release_firmware(fw); fw = NULL; + } else { + fw_log_firmware_info(fw, name, device); } *firmware_p = fw; @@ -860,6 +929,8 @@ _request_firmware(const struct firmware **firmware_p, const char *name, * @name will be used as $FIRMWARE in the uevent environment and * should be distinctive enough not to be confused with any other * firmware image for this or any other device. + * It must not contain any ".." path components - "foo/bar..bin" is + * allowed, but "foo/../bar.bin" is not. * * Caller must hold the reference count of @device. * @@ -959,8 +1030,8 @@ EXPORT_SYMBOL_GPL(firmware_request_platform); /** * firmware_request_cache() - cache firmware for suspend so resume can use it - * @name: name of firmware file * @device: device for which firmware should be cached for + * @name: name of firmware file * * There are some devices with an optimization that enables the device to not * require loading firmware on system reboot. This optimization may still @@ -1051,7 +1122,7 @@ EXPORT_SYMBOL(request_partial_firmware_into_buf); void release_firmware(const struct firmware *fw) { if (fw) { - if (!fw_is_builtin_firmware(fw)) + if (!firmware_is_builtin(fw)) firmware_free_data(fw); kfree(fw); } @@ -1086,34 +1157,11 @@ static void request_firmware_work_func(struct work_struct *work) kfree(fw_work); } -/** - * request_firmware_nowait() - asynchronous version of request_firmware - * @module: module requesting the firmware - * @uevent: sends uevent to copy the firmware image if this flag - * is non-zero else the firmware copy must be done manually. - * @name: name of firmware file - * @device: device for which firmware is being loaded - * @gfp: allocation flags - * @context: will be passed over to @cont, and - * @fw may be %NULL if firmware request fails. - * @cont: function will be called asynchronously when the firmware - * request is over. - * - * Caller must hold the reference count of @device. - * - * Asynchronous variant of request_firmware() for user contexts: - * - sleep for as small periods as possible since it may - * increase kernel boot time of built-in device drivers - * requesting firmware in their ->probe() methods, if - * @gfp is GFP_KERNEL. - * - * - can't sleep at all if @gfp is GFP_ATOMIC. - **/ -int -request_firmware_nowait( + +static int _request_firmware_nowait( struct module *module, bool uevent, const char *name, struct device *device, gfp_t gfp, void *context, - void (*cont)(const struct firmware *fw, void *context)) + void (*cont)(const struct firmware *fw, void *context), bool nowarn) { struct firmware_work *fw_work; @@ -1131,7 +1179,8 @@ request_firmware_nowait( fw_work->context = context; fw_work->cont = cont; fw_work->opt_flags = FW_OPT_NOWAIT | - (uevent ? FW_OPT_UEVENT : FW_OPT_USERHELPER); + (uevent ? FW_OPT_UEVENT : FW_OPT_USERHELPER) | + (nowarn ? FW_OPT_NO_WARN : 0); if (!uevent && fw_cache_is_setup(device, name)) { kfree_const(fw_work->name); @@ -1150,8 +1199,66 @@ request_firmware_nowait( schedule_work(&fw_work->work); return 0; } + +/** + * request_firmware_nowait() - asynchronous version of request_firmware + * @module: module requesting the firmware + * @uevent: sends uevent to copy the firmware image if this flag + * is non-zero else the firmware copy must be done manually. + * @name: name of firmware file + * @device: device for which firmware is being loaded + * @gfp: allocation flags + * @context: will be passed over to @cont, and + * @fw may be %NULL if firmware request fails. + * @cont: function will be called asynchronously when the firmware + * request is over. + * + * Caller must hold the reference count of @device. + * + * Asynchronous variant of request_firmware() for user contexts: + * - sleep for as small periods as possible since it may + * increase kernel boot time of built-in device drivers + * requesting firmware in their ->probe() methods, if + * @gfp is GFP_KERNEL. + * + * - can't sleep at all if @gfp is GFP_ATOMIC. + **/ +int request_firmware_nowait( + struct module *module, bool uevent, + const char *name, struct device *device, gfp_t gfp, void *context, + void (*cont)(const struct firmware *fw, void *context)) +{ + return _request_firmware_nowait(module, uevent, name, device, gfp, + context, cont, false); + +} EXPORT_SYMBOL(request_firmware_nowait); +/** + * firmware_request_nowait_nowarn() - async version of request_firmware_nowarn + * @module: module requesting the firmware + * @name: name of firmware file + * @device: device for which firmware is being loaded + * @gfp: allocation flags + * @context: will be passed over to @cont, and + * @fw may be %NULL if firmware request fails. + * @cont: function will be called asynchronously when the firmware + * request is over. + * + * Similar in function to request_firmware_nowait(), but doesn't print a warning + * when the firmware file could not be found and always sends a uevent to copy + * the firmware image. + */ +int firmware_request_nowait_nowarn( + struct module *module, const char *name, + struct device *device, gfp_t gfp, void *context, + void (*cont)(const struct firmware *fw, void *context)) +{ + return _request_firmware_nowait(module, FW_ACTION_UEVENT, name, device, + gfp, context, cont, true); +} +EXPORT_SYMBOL_GPL(firmware_request_nowait_nowarn); + #ifdef CONFIG_FW_CACHE static ASYNC_DOMAIN_EXCLUSIVE(fw_cache_domain); @@ -1215,7 +1322,7 @@ static int uncache_firmware(const char *fw_name) pr_debug("%s: %s\n", __func__, fw_name); - if (fw_get_builtin_firmware(&fw, fw_name, NULL, 0)) + if (firmware_request_builtin(&fw, fw_name)) return 0; fw_priv = lookup_fw_priv(fw_name); @@ -1257,11 +1364,11 @@ static int __fw_entry_found(const char *name) return 0; } -static int fw_cache_piggyback_on_request(const char *name) +static void fw_cache_piggyback_on_request(struct fw_priv *fw_priv) { - struct firmware_cache *fwc = &fw_cache; + const char *name = fw_priv->fw_name; + struct firmware_cache *fwc = fw_priv->fwc; struct fw_cache_entry *fce; - int ret = 0; spin_lock(&fwc->name_lock); if (__fw_entry_found(name)) @@ -1269,13 +1376,12 @@ static int fw_cache_piggyback_on_request(const char *name) fce = alloc_fw_cache_entry(name); if (fce) { - ret = 1; list_add(&fce->list, &fwc->fw_names); + kref_get(&fw_priv->ref); pr_debug("%s: fw: %s\n", __func__, name); } found: spin_unlock(&fwc->name_lock); - return ret; } static void free_fw_cache_entry(struct fw_cache_entry *fce) @@ -1444,10 +1550,10 @@ static int fw_pm_notify(struct notifier_block *notify_block, case PM_SUSPEND_PREPARE: case PM_RESTORE_PREPARE: /* - * kill pending fallback requests with a custom fallback - * to avoid stalling suspend. + * Here, kill pending fallback requests will only kill + * non-uevent firmware request to avoid stalling suspend. */ - kill_pending_fw_fallback_reqs(true); + kill_pending_fw_fallback_reqs(false); device_cache_fw_images(); break; @@ -1470,16 +1576,20 @@ static int fw_pm_notify(struct notifier_block *notify_block, } /* stop caching firmware once syscore_suspend is reached */ -static int fw_suspend(void) +static int fw_suspend(void *data) { fw_cache.state = FW_LOADER_NO_CACHE; return 0; } -static struct syscore_ops fw_syscore_ops = { +static const struct syscore_ops fw_syscore_ops = { .suspend = fw_suspend, }; +static struct syscore fw_syscore = { + .ops = &fw_syscore_ops, +}; + static int __init register_fw_pm_ops(void) { int ret; @@ -1495,20 +1605,19 @@ static int __init register_fw_pm_ops(void) if (ret) return ret; - register_syscore_ops(&fw_syscore_ops); + register_syscore(&fw_syscore); return ret; } static inline void unregister_fw_pm_ops(void) { - unregister_syscore_ops(&fw_syscore_ops); + unregister_syscore(&fw_syscore); unregister_pm_notifier(&fw_cache.pm_notify); } #else -static int fw_cache_piggyback_on_request(const char *name) +static void fw_cache_piggyback_on_request(struct fw_priv *fw_priv) { - return 0; } static inline int register_fw_pm_ops(void) { @@ -1533,7 +1642,7 @@ static int fw_shutdown_notify(struct notifier_block *unused1, * Kill all pending fallback requests to avoid both stalling shutdown, * and avoid a deadlock with the usermode_lock. */ - kill_pending_fw_fallback_reqs(false); + kill_pending_fw_fallback_reqs(true); return NOTIFY_DONE; } diff --git a/drivers/base/firmware_loader/sysfs.c b/drivers/base/firmware_loader/sysfs.c new file mode 100644 index 000000000000..92e91050f96a --- /dev/null +++ b/drivers/base/firmware_loader/sysfs.c @@ -0,0 +1,425 @@ +// SPDX-License-Identifier: GPL-2.0 + +#include <linux/highmem.h> +#include <linux/module.h> +#include <linux/security.h> +#include <linux/slab.h> +#include <linux/types.h> + +#include "sysfs.h" + +/* + * sysfs support for firmware loader + */ + +void __fw_load_abort(struct fw_priv *fw_priv) +{ + /* + * There is a small window in which user can write to 'loading' + * between loading done/aborted and disappearance of 'loading' + */ + if (fw_state_is_aborted(fw_priv) || fw_state_is_done(fw_priv)) + return; + + fw_state_aborted(fw_priv); +} + +#ifdef CONFIG_FW_LOADER_USER_HELPER +static ssize_t timeout_show(const struct class *class, const struct class_attribute *attr, + char *buf) +{ + return sysfs_emit(buf, "%d\n", __firmware_loading_timeout()); +} + +/** + * timeout_store() - set number of seconds to wait for firmware + * @class: device class pointer + * @attr: device attribute pointer + * @buf: buffer to scan for timeout value + * @count: number of bytes in @buf + * + * Sets the number of seconds to wait for the firmware. Once + * this expires an error will be returned to the driver and no + * firmware will be provided. + * + * Note: zero means 'wait forever'. + **/ +static ssize_t timeout_store(const struct class *class, const struct class_attribute *attr, + const char *buf, size_t count) +{ + int tmp_loading_timeout; + + if (kstrtoint(buf, 10, &tmp_loading_timeout)) + return -EINVAL; + + if (tmp_loading_timeout < 0) + tmp_loading_timeout = 0; + + __fw_fallback_set_timeout(tmp_loading_timeout); + + return count; +} +static CLASS_ATTR_RW(timeout); + +static struct attribute *firmware_class_attrs[] = { + &class_attr_timeout.attr, + NULL, +}; +ATTRIBUTE_GROUPS(firmware_class); + +static int do_firmware_uevent(const struct fw_sysfs *fw_sysfs, struct kobj_uevent_env *env) +{ + if (add_uevent_var(env, "FIRMWARE=%s", fw_sysfs->fw_priv->fw_name)) + return -ENOMEM; + if (add_uevent_var(env, "TIMEOUT=%i", __firmware_loading_timeout())) + return -ENOMEM; + if (add_uevent_var(env, "ASYNC=%d", fw_sysfs->nowait)) + return -ENOMEM; + + return 0; +} + +static int firmware_uevent(const struct device *dev, struct kobj_uevent_env *env) +{ + const struct fw_sysfs *fw_sysfs = to_fw_sysfs(dev); + int err = 0; + + mutex_lock(&fw_lock); + if (fw_sysfs->fw_priv) + err = do_firmware_uevent(fw_sysfs, env); + mutex_unlock(&fw_lock); + return err; +} +#endif /* CONFIG_FW_LOADER_USER_HELPER */ + +static void fw_dev_release(struct device *dev) +{ + struct fw_sysfs *fw_sysfs = to_fw_sysfs(dev); + + if (fw_sysfs->fw_upload_priv) + fw_upload_free(fw_sysfs); + + kfree(fw_sysfs); +} + +static struct class firmware_class = { + .name = "firmware", +#ifdef CONFIG_FW_LOADER_USER_HELPER + .class_groups = firmware_class_groups, + .dev_uevent = firmware_uevent, +#endif + .dev_release = fw_dev_release, +}; + +int register_sysfs_loader(void) +{ + int ret = class_register(&firmware_class); + + if (ret != 0) + return ret; + return register_firmware_config_sysctl(); +} + +void unregister_sysfs_loader(void) +{ + unregister_firmware_config_sysctl(); + class_unregister(&firmware_class); +} + +static ssize_t firmware_loading_show(struct device *dev, + struct device_attribute *attr, char *buf) +{ + struct fw_sysfs *fw_sysfs = to_fw_sysfs(dev); + int loading = 0; + + mutex_lock(&fw_lock); + if (fw_sysfs->fw_priv) + loading = fw_state_is_loading(fw_sysfs->fw_priv); + mutex_unlock(&fw_lock); + + return sysfs_emit(buf, "%d\n", loading); +} + +/** + * firmware_loading_store() - set value in the 'loading' control file + * @dev: device pointer + * @attr: device attribute pointer + * @buf: buffer to scan for loading control value + * @count: number of bytes in @buf + * + * The relevant values are: + * + * 1: Start a load, discarding any previous partial load. + * 0: Conclude the load and hand the data to the driver code. + * -1: Conclude the load with an error and discard any written data. + **/ +static ssize_t firmware_loading_store(struct device *dev, + struct device_attribute *attr, + const char *buf, size_t count) +{ + struct fw_sysfs *fw_sysfs = to_fw_sysfs(dev); + struct fw_priv *fw_priv; + ssize_t written = count; + int loading; + + if (kstrtoint(buf, 10, &loading)) + return -EINVAL; + + mutex_lock(&fw_lock); + fw_priv = fw_sysfs->fw_priv; + if (fw_state_is_aborted(fw_priv) || fw_state_is_done(fw_priv)) + goto out; + + switch (loading) { + case 1: + /* discarding any previous partial load */ + fw_free_paged_buf(fw_priv); + fw_state_start(fw_priv); + break; + case 0: + if (fw_state_is_loading(fw_priv)) { + int rc; + + /* + * Several loading requests may be pending on + * one same firmware buf, so let all requests + * see the mapped 'buf->data' once the loading + * is completed. + */ + rc = fw_map_paged_buf(fw_priv); + if (rc) + dev_err(dev, "%s: map pages failed\n", + __func__); + else + rc = security_kernel_post_load_data(fw_priv->data, + fw_priv->size, + LOADING_FIRMWARE, + "blob"); + + /* + * Same logic as fw_load_abort, only the DONE bit + * is ignored and we set ABORT only on failure. + */ + if (rc) { + fw_state_aborted(fw_priv); + written = rc; + } else { + fw_state_done(fw_priv); + + /* + * If this is a user-initiated firmware upload + * then start the upload in a worker thread now. + */ + rc = fw_upload_start(fw_sysfs); + if (rc) + written = rc; + } + break; + } + fallthrough; + default: + dev_err(dev, "%s: unexpected value (%d)\n", __func__, loading); + fallthrough; + case -1: + fw_load_abort(fw_sysfs); + if (fw_sysfs->fw_upload_priv) + fw_state_init(fw_sysfs->fw_priv); + + break; + } +out: + mutex_unlock(&fw_lock); + return written; +} + +DEVICE_ATTR(loading, 0644, firmware_loading_show, firmware_loading_store); + +static void firmware_rw_data(struct fw_priv *fw_priv, char *buffer, + loff_t offset, size_t count, bool read) +{ + if (read) + memcpy(buffer, fw_priv->data + offset, count); + else + memcpy(fw_priv->data + offset, buffer, count); +} + +static void firmware_rw(struct fw_priv *fw_priv, char *buffer, + loff_t offset, size_t count, bool read) +{ + while (count) { + int page_nr = offset >> PAGE_SHIFT; + int page_ofs = offset & (PAGE_SIZE - 1); + int page_cnt = min_t(size_t, PAGE_SIZE - page_ofs, count); + + if (read) + memcpy_from_page(buffer, fw_priv->pages[page_nr], + page_ofs, page_cnt); + else + memcpy_to_page(fw_priv->pages[page_nr], page_ofs, + buffer, page_cnt); + + buffer += page_cnt; + offset += page_cnt; + count -= page_cnt; + } +} + +static ssize_t firmware_data_read(struct file *filp, struct kobject *kobj, + const struct bin_attribute *bin_attr, + char *buffer, loff_t offset, size_t count) +{ + struct device *dev = kobj_to_dev(kobj); + struct fw_sysfs *fw_sysfs = to_fw_sysfs(dev); + struct fw_priv *fw_priv; + ssize_t ret_count; + + mutex_lock(&fw_lock); + fw_priv = fw_sysfs->fw_priv; + if (!fw_priv || fw_state_is_done(fw_priv)) { + ret_count = -ENODEV; + goto out; + } + if (offset > fw_priv->size) { + ret_count = 0; + goto out; + } + if (count > fw_priv->size - offset) + count = fw_priv->size - offset; + + ret_count = count; + + if (fw_priv->data) + firmware_rw_data(fw_priv, buffer, offset, count, true); + else + firmware_rw(fw_priv, buffer, offset, count, true); + +out: + mutex_unlock(&fw_lock); + return ret_count; +} + +static int fw_realloc_pages(struct fw_sysfs *fw_sysfs, int min_size) +{ + int err; + + err = fw_grow_paged_buf(fw_sysfs->fw_priv, + PAGE_ALIGN(min_size) >> PAGE_SHIFT); + if (err) + fw_load_abort(fw_sysfs); + return err; +} + +/** + * firmware_data_write() - write method for firmware + * @filp: open sysfs file + * @kobj: kobject for the device + * @bin_attr: bin_attr structure + * @buffer: buffer being written + * @offset: buffer offset for write in total data store area + * @count: buffer size + * + * Data written to the 'data' attribute will be later handed to + * the driver as a firmware image. + **/ +static ssize_t firmware_data_write(struct file *filp, struct kobject *kobj, + const struct bin_attribute *bin_attr, + char *buffer, loff_t offset, size_t count) +{ + struct device *dev = kobj_to_dev(kobj); + struct fw_sysfs *fw_sysfs = to_fw_sysfs(dev); + struct fw_priv *fw_priv; + ssize_t retval; + + if (!capable(CAP_SYS_RAWIO)) + return -EPERM; + + mutex_lock(&fw_lock); + fw_priv = fw_sysfs->fw_priv; + if (!fw_priv || fw_state_is_done(fw_priv)) { + retval = -ENODEV; + goto out; + } + + if (fw_priv->data) { + if (offset + count > fw_priv->allocated_size) { + retval = -ENOMEM; + goto out; + } + firmware_rw_data(fw_priv, buffer, offset, count, false); + retval = count; + } else { + retval = fw_realloc_pages(fw_sysfs, offset + count); + if (retval) + goto out; + + retval = count; + firmware_rw(fw_priv, buffer, offset, count, false); + } + + fw_priv->size = max_t(size_t, offset + count, fw_priv->size); +out: + mutex_unlock(&fw_lock); + return retval; +} + +static const struct bin_attribute firmware_attr_data = { + .attr = { .name = "data", .mode = 0644 }, + .size = 0, + .read = firmware_data_read, + .write = firmware_data_write, +}; + +static struct attribute *fw_dev_attrs[] = { + &dev_attr_loading.attr, +#ifdef CONFIG_FW_UPLOAD + &dev_attr_cancel.attr, + &dev_attr_status.attr, + &dev_attr_error.attr, + &dev_attr_remaining_size.attr, +#endif + NULL +}; + +static const struct bin_attribute *const fw_dev_bin_attrs[] = { + &firmware_attr_data, + NULL +}; + +static const struct attribute_group fw_dev_attr_group = { + .attrs = fw_dev_attrs, + .bin_attrs = fw_dev_bin_attrs, +#ifdef CONFIG_FW_UPLOAD + .is_visible = fw_upload_is_visible, +#endif +}; + +static const struct attribute_group *fw_dev_attr_groups[] = { + &fw_dev_attr_group, + NULL +}; + +struct fw_sysfs * +fw_create_instance(struct firmware *firmware, const char *fw_name, + struct device *device, u32 opt_flags) +{ + struct fw_sysfs *fw_sysfs; + struct device *f_dev; + + fw_sysfs = kzalloc(sizeof(*fw_sysfs), GFP_KERNEL); + if (!fw_sysfs) { + fw_sysfs = ERR_PTR(-ENOMEM); + goto exit; + } + + fw_sysfs->nowait = !!(opt_flags & FW_OPT_NOWAIT); + fw_sysfs->fw = firmware; + f_dev = &fw_sysfs->dev; + + device_initialize(f_dev); + dev_set_name(f_dev, "%s", fw_name); + f_dev->parent = device; + f_dev->class = &firmware_class; + f_dev->groups = fw_dev_attr_groups; +exit: + return fw_sysfs; +} diff --git a/drivers/base/firmware_loader/sysfs.h b/drivers/base/firmware_loader/sysfs.h new file mode 100644 index 000000000000..1cabea544a40 --- /dev/null +++ b/drivers/base/firmware_loader/sysfs.h @@ -0,0 +1,118 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +#ifndef __FIRMWARE_SYSFS_H +#define __FIRMWARE_SYSFS_H + +#include <linux/device.h> + +#include "firmware.h" + +MODULE_IMPORT_NS("FIRMWARE_LOADER_PRIVATE"); + +extern struct firmware_fallback_config fw_fallback_config; +extern struct device_attribute dev_attr_loading; + +#ifdef CONFIG_FW_LOADER_USER_HELPER +/** + * struct firmware_fallback_config - firmware fallback configuration settings + * + * Helps describe and fine tune the fallback mechanism. + * + * @force_sysfs_fallback: force the sysfs fallback mechanism to be used + * as if one had enabled CONFIG_FW_LOADER_USER_HELPER_FALLBACK=y. + * Useful to help debug a CONFIG_FW_LOADER_USER_HELPER_FALLBACK=y + * functionality on a kernel where that config entry has been disabled. + * @ignore_sysfs_fallback: force to disable the sysfs fallback mechanism. + * This emulates the behaviour as if we had set the kernel + * config CONFIG_FW_LOADER_USER_HELPER=n. + * @old_timeout: for internal use + * @loading_timeout: the timeout to wait for the fallback mechanism before + * giving up, in seconds. + */ +struct firmware_fallback_config { + unsigned int force_sysfs_fallback; + unsigned int ignore_sysfs_fallback; + int old_timeout; + int loading_timeout; +}; + +/* These getters are vetted to use int properly */ +static inline int __firmware_loading_timeout(void) +{ + return fw_fallback_config.loading_timeout; +} + +/* These setters are vetted to use int properly */ +static inline void __fw_fallback_set_timeout(int timeout) +{ + fw_fallback_config.loading_timeout = timeout; +} +#endif + +#ifdef CONFIG_FW_LOADER_SYSFS +int register_sysfs_loader(void); +void unregister_sysfs_loader(void); +#if defined(CONFIG_FW_LOADER_USER_HELPER) && defined(CONFIG_SYSCTL) +int register_firmware_config_sysctl(void); +void unregister_firmware_config_sysctl(void); +#else +static inline int register_firmware_config_sysctl(void) +{ + return 0; +} + +static inline void unregister_firmware_config_sysctl(void) { } +#endif /* CONFIG_FW_LOADER_USER_HELPER && CONFIG_SYSCTL */ +#else /* CONFIG_FW_LOADER_SYSFS */ +static inline int register_sysfs_loader(void) +{ + return 0; +} + +static inline void unregister_sysfs_loader(void) +{ +} +#endif /* CONFIG_FW_LOADER_SYSFS */ + +struct fw_sysfs { + bool nowait; + struct device dev; + struct fw_priv *fw_priv; + struct firmware *fw; + void *fw_upload_priv; +}; +#define to_fw_sysfs(__dev) container_of_const(__dev, struct fw_sysfs, dev) + +void __fw_load_abort(struct fw_priv *fw_priv); + +static inline void fw_load_abort(struct fw_sysfs *fw_sysfs) +{ + struct fw_priv *fw_priv = fw_sysfs->fw_priv; + + __fw_load_abort(fw_priv); +} + +struct fw_sysfs * +fw_create_instance(struct firmware *firmware, const char *fw_name, + struct device *device, u32 opt_flags); + +#ifdef CONFIG_FW_UPLOAD +extern struct device_attribute dev_attr_status; +extern struct device_attribute dev_attr_error; +extern struct device_attribute dev_attr_cancel; +extern struct device_attribute dev_attr_remaining_size; + +int fw_upload_start(struct fw_sysfs *fw_sysfs); +void fw_upload_free(struct fw_sysfs *fw_sysfs); +umode_t fw_upload_is_visible(struct kobject *kobj, struct attribute *attr, int n); +#else +static inline int fw_upload_start(struct fw_sysfs *fw_sysfs) +{ + return 0; +} + +static inline void fw_upload_free(struct fw_sysfs *fw_sysfs) +{ +} +#endif + +#endif /* __FIRMWARE_SYSFS_H */ diff --git a/drivers/base/firmware_loader/sysfs_upload.c b/drivers/base/firmware_loader/sysfs_upload.c new file mode 100644 index 000000000000..c3797b93c5f5 --- /dev/null +++ b/drivers/base/firmware_loader/sysfs_upload.c @@ -0,0 +1,410 @@ +// SPDX-License-Identifier: GPL-2.0 + +#include <linux/firmware.h> +#include <linux/module.h> +#include <linux/slab.h> + +#include "sysfs_upload.h" + +/* + * Support for user-space to initiate a firmware upload to a device. + */ + +static const char * const fw_upload_prog_str[] = { + [FW_UPLOAD_PROG_IDLE] = "idle", + [FW_UPLOAD_PROG_RECEIVING] = "receiving", + [FW_UPLOAD_PROG_PREPARING] = "preparing", + [FW_UPLOAD_PROG_TRANSFERRING] = "transferring", + [FW_UPLOAD_PROG_PROGRAMMING] = "programming" +}; + +static const char * const fw_upload_err_str[] = { + [FW_UPLOAD_ERR_NONE] = "none", + [FW_UPLOAD_ERR_HW_ERROR] = "hw-error", + [FW_UPLOAD_ERR_TIMEOUT] = "timeout", + [FW_UPLOAD_ERR_CANCELED] = "user-abort", + [FW_UPLOAD_ERR_BUSY] = "device-busy", + [FW_UPLOAD_ERR_INVALID_SIZE] = "invalid-file-size", + [FW_UPLOAD_ERR_RW_ERROR] = "read-write-error", + [FW_UPLOAD_ERR_WEAROUT] = "flash-wearout", + [FW_UPLOAD_ERR_FW_INVALID] = "firmware-invalid", +}; + +static const char *fw_upload_progress(struct device *dev, + enum fw_upload_prog prog) +{ + const char *status = "unknown-status"; + + if (prog < FW_UPLOAD_PROG_MAX) + status = fw_upload_prog_str[prog]; + else + dev_err(dev, "Invalid status during secure update: %d\n", prog); + + return status; +} + +static const char *fw_upload_error(struct device *dev, + enum fw_upload_err err_code) +{ + const char *error = "unknown-error"; + + if (err_code < FW_UPLOAD_ERR_MAX) + error = fw_upload_err_str[err_code]; + else + dev_err(dev, "Invalid error code during secure update: %d\n", + err_code); + + return error; +} + +static ssize_t +status_show(struct device *dev, struct device_attribute *attr, char *buf) +{ + struct fw_upload_priv *fwlp = to_fw_sysfs(dev)->fw_upload_priv; + + return sysfs_emit(buf, "%s\n", fw_upload_progress(dev, fwlp->progress)); +} +DEVICE_ATTR_RO(status); + +static ssize_t +error_show(struct device *dev, struct device_attribute *attr, char *buf) +{ + struct fw_upload_priv *fwlp = to_fw_sysfs(dev)->fw_upload_priv; + int ret; + + mutex_lock(&fwlp->lock); + + if (fwlp->progress != FW_UPLOAD_PROG_IDLE) + ret = -EBUSY; + else if (!fwlp->err_code) + ret = 0; + else + ret = sysfs_emit(buf, "%s:%s\n", + fw_upload_progress(dev, fwlp->err_progress), + fw_upload_error(dev, fwlp->err_code)); + + mutex_unlock(&fwlp->lock); + + return ret; +} +DEVICE_ATTR_RO(error); + +static ssize_t cancel_store(struct device *dev, struct device_attribute *attr, + const char *buf, size_t count) +{ + struct fw_upload_priv *fwlp = to_fw_sysfs(dev)->fw_upload_priv; + int ret = count; + bool cancel; + + if (kstrtobool(buf, &cancel) || !cancel) + return -EINVAL; + + mutex_lock(&fwlp->lock); + if (fwlp->progress == FW_UPLOAD_PROG_IDLE) { + mutex_unlock(&fwlp->lock); + return -ENODEV; + } + + fwlp->ops->cancel(fwlp->fw_upload); + mutex_unlock(&fwlp->lock); + + return ret; +} +DEVICE_ATTR_WO(cancel); + +static ssize_t remaining_size_show(struct device *dev, + struct device_attribute *attr, char *buf) +{ + struct fw_upload_priv *fwlp = to_fw_sysfs(dev)->fw_upload_priv; + + return sysfs_emit(buf, "%u\n", fwlp->remaining_size); +} +DEVICE_ATTR_RO(remaining_size); + +umode_t +fw_upload_is_visible(struct kobject *kobj, struct attribute *attr, int n) +{ + static struct fw_sysfs *fw_sysfs; + + fw_sysfs = to_fw_sysfs(kobj_to_dev(kobj)); + + if (fw_sysfs->fw_upload_priv || attr == &dev_attr_loading.attr) + return attr->mode; + + return 0; +} + +static void fw_upload_update_progress(struct fw_upload_priv *fwlp, + enum fw_upload_prog new_progress) +{ + mutex_lock(&fwlp->lock); + fwlp->progress = new_progress; + mutex_unlock(&fwlp->lock); +} + +static void fw_upload_set_error(struct fw_upload_priv *fwlp, + enum fw_upload_err err_code) +{ + mutex_lock(&fwlp->lock); + fwlp->err_progress = fwlp->progress; + fwlp->err_code = err_code; + mutex_unlock(&fwlp->lock); +} + +static void fw_upload_prog_complete(struct fw_upload_priv *fwlp) +{ + mutex_lock(&fwlp->lock); + fwlp->progress = FW_UPLOAD_PROG_IDLE; + mutex_unlock(&fwlp->lock); +} + +static void fw_upload_main(struct work_struct *work) +{ + struct fw_upload_priv *fwlp; + struct fw_sysfs *fw_sysfs; + u32 written = 0, offset = 0; + enum fw_upload_err ret; + struct device *fw_dev; + struct fw_upload *fwl; + + fwlp = container_of(work, struct fw_upload_priv, work); + fwl = fwlp->fw_upload; + fw_sysfs = (struct fw_sysfs *)fwl->priv; + fw_dev = &fw_sysfs->dev; + + fw_upload_update_progress(fwlp, FW_UPLOAD_PROG_PREPARING); + ret = fwlp->ops->prepare(fwl, fwlp->data, fwlp->remaining_size); + if (ret != FW_UPLOAD_ERR_NONE) { + fw_upload_set_error(fwlp, ret); + goto putdev_exit; + } + + fw_upload_update_progress(fwlp, FW_UPLOAD_PROG_TRANSFERRING); + while (fwlp->remaining_size) { + ret = fwlp->ops->write(fwl, fwlp->data, offset, + fwlp->remaining_size, &written); + if (ret != FW_UPLOAD_ERR_NONE || !written) { + if (ret == FW_UPLOAD_ERR_NONE) { + dev_warn(fw_dev, "write-op wrote zero data\n"); + ret = FW_UPLOAD_ERR_RW_ERROR; + } + fw_upload_set_error(fwlp, ret); + goto done; + } + + fwlp->remaining_size -= written; + offset += written; + } + + fw_upload_update_progress(fwlp, FW_UPLOAD_PROG_PROGRAMMING); + ret = fwlp->ops->poll_complete(fwl); + if (ret != FW_UPLOAD_ERR_NONE) + fw_upload_set_error(fwlp, ret); + +done: + if (fwlp->ops->cleanup) + fwlp->ops->cleanup(fwl); + +putdev_exit: + put_device(fw_dev->parent); + + /* + * Note: fwlp->remaining_size is left unmodified here to provide + * additional information on errors. It will be reinitialized when + * the next firmeware upload begins. + */ + mutex_lock(&fw_lock); + fw_free_paged_buf(fw_sysfs->fw_priv); + fw_state_init(fw_sysfs->fw_priv); + mutex_unlock(&fw_lock); + fwlp->data = NULL; + fw_upload_prog_complete(fwlp); +} + +/* + * Start a worker thread to upload data to the parent driver. + * Must be called with fw_lock held. + */ +int fw_upload_start(struct fw_sysfs *fw_sysfs) +{ + struct fw_priv *fw_priv = fw_sysfs->fw_priv; + struct device *fw_dev = &fw_sysfs->dev; + struct fw_upload_priv *fwlp; + + if (!fw_sysfs->fw_upload_priv) + return 0; + + if (!fw_priv->size) { + fw_free_paged_buf(fw_priv); + fw_state_init(fw_sysfs->fw_priv); + return 0; + } + + fwlp = fw_sysfs->fw_upload_priv; + mutex_lock(&fwlp->lock); + + /* Do not interfere with an on-going fw_upload */ + if (fwlp->progress != FW_UPLOAD_PROG_IDLE) { + mutex_unlock(&fwlp->lock); + return -EBUSY; + } + + get_device(fw_dev->parent); /* released in fw_upload_main */ + + fwlp->progress = FW_UPLOAD_PROG_RECEIVING; + fwlp->err_code = 0; + fwlp->remaining_size = fw_priv->size; + fwlp->data = fw_priv->data; + + pr_debug("%s: fw-%s fw_priv=%p data=%p size=%u\n", + __func__, fw_priv->fw_name, + fw_priv, fw_priv->data, + (unsigned int)fw_priv->size); + + queue_work(system_long_wq, &fwlp->work); + mutex_unlock(&fwlp->lock); + + return 0; +} + +void fw_upload_free(struct fw_sysfs *fw_sysfs) +{ + struct fw_upload_priv *fw_upload_priv = fw_sysfs->fw_upload_priv; + + free_fw_priv(fw_sysfs->fw_priv); + kfree(fw_upload_priv->fw_upload); + kfree(fw_upload_priv); +} + +/** + * firmware_upload_register() - register for the firmware upload sysfs API + * @module: kernel module of this device + * @parent: parent device instantiating firmware upload + * @name: firmware name to be associated with this device + * @ops: pointer to structure of firmware upload ops + * @dd_handle: pointer to parent driver private data + * + * @name must be unique among all users of firmware upload. The firmware + * sysfs files for this device will be found at /sys/class/firmware/@name. + * + * Return: struct fw_upload pointer or ERR_PTR() + * + **/ +struct fw_upload * +firmware_upload_register(struct module *module, struct device *parent, + const char *name, const struct fw_upload_ops *ops, + void *dd_handle) +{ + u32 opt_flags = FW_OPT_NOCACHE; + struct fw_upload *fw_upload; + struct fw_upload_priv *fw_upload_priv; + struct fw_sysfs *fw_sysfs; + struct fw_priv *fw_priv; + struct device *fw_dev; + int ret; + + if (!name || name[0] == '\0') + return ERR_PTR(-EINVAL); + + if (!ops || !ops->cancel || !ops->prepare || + !ops->write || !ops->poll_complete) { + dev_err(parent, "Attempt to register without all required ops\n"); + return ERR_PTR(-EINVAL); + } + + if (!try_module_get(module)) + return ERR_PTR(-EFAULT); + + fw_upload = kzalloc(sizeof(*fw_upload), GFP_KERNEL); + if (!fw_upload) { + ret = -ENOMEM; + goto exit_module_put; + } + + fw_upload_priv = kzalloc(sizeof(*fw_upload_priv), GFP_KERNEL); + if (!fw_upload_priv) { + ret = -ENOMEM; + goto free_fw_upload; + } + + fw_upload_priv->fw_upload = fw_upload; + fw_upload_priv->ops = ops; + mutex_init(&fw_upload_priv->lock); + fw_upload_priv->module = module; + fw_upload_priv->name = name; + fw_upload_priv->err_code = 0; + fw_upload_priv->progress = FW_UPLOAD_PROG_IDLE; + INIT_WORK(&fw_upload_priv->work, fw_upload_main); + fw_upload->dd_handle = dd_handle; + + fw_sysfs = fw_create_instance(NULL, name, parent, opt_flags); + if (IS_ERR(fw_sysfs)) { + ret = PTR_ERR(fw_sysfs); + goto free_fw_upload_priv; + } + fw_upload->priv = fw_sysfs; + fw_sysfs->fw_upload_priv = fw_upload_priv; + fw_dev = &fw_sysfs->dev; + + ret = alloc_lookup_fw_priv(name, &fw_cache, &fw_priv, NULL, 0, 0, + FW_OPT_NOCACHE); + if (ret != 0) { + if (ret > 0) + ret = -EINVAL; + goto free_fw_sysfs; + } + fw_priv->is_paged_buf = true; + fw_sysfs->fw_priv = fw_priv; + + ret = device_add(fw_dev); + if (ret) { + dev_err(fw_dev, "%s: device_register failed\n", __func__); + put_device(fw_dev); + goto exit_module_put; + } + + return fw_upload; + +free_fw_sysfs: + kfree(fw_sysfs); + +free_fw_upload_priv: + kfree(fw_upload_priv); + +free_fw_upload: + kfree(fw_upload); + +exit_module_put: + module_put(module); + + return ERR_PTR(ret); +} +EXPORT_SYMBOL_GPL(firmware_upload_register); + +/** + * firmware_upload_unregister() - Unregister firmware upload interface + * @fw_upload: pointer to struct fw_upload + **/ +void firmware_upload_unregister(struct fw_upload *fw_upload) +{ + struct fw_sysfs *fw_sysfs = fw_upload->priv; + struct fw_upload_priv *fw_upload_priv = fw_sysfs->fw_upload_priv; + struct module *module = fw_upload_priv->module; + + mutex_lock(&fw_upload_priv->lock); + if (fw_upload_priv->progress == FW_UPLOAD_PROG_IDLE) { + mutex_unlock(&fw_upload_priv->lock); + goto unregister; + } + + fw_upload_priv->ops->cancel(fw_upload); + mutex_unlock(&fw_upload_priv->lock); + + /* Ensure lower-level device-driver is finished */ + flush_work(&fw_upload_priv->work); + +unregister: + device_unregister(&fw_sysfs->dev); + module_put(module); +} +EXPORT_SYMBOL_GPL(firmware_upload_unregister); diff --git a/drivers/base/firmware_loader/sysfs_upload.h b/drivers/base/firmware_loader/sysfs_upload.h new file mode 100644 index 000000000000..31931ff7808a --- /dev/null +++ b/drivers/base/firmware_loader/sysfs_upload.h @@ -0,0 +1,41 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +#ifndef __SYSFS_UPLOAD_H +#define __SYSFS_UPLOAD_H + +#include <linux/device.h> + +#include "sysfs.h" + +/** + * enum fw_upload_prog - firmware upload progress codes + * @FW_UPLOAD_PROG_IDLE: there is no firmware upload in progress + * @FW_UPLOAD_PROG_RECEIVING: worker thread is receiving firmware data + * @FW_UPLOAD_PROG_PREPARING: target device is preparing for firmware upload + * @FW_UPLOAD_PROG_TRANSFERRING: data is being copied to the device + * @FW_UPLOAD_PROG_PROGRAMMING: device is performing the firmware update + * @FW_UPLOAD_PROG_MAX: Maximum progress code marker + */ +enum fw_upload_prog { + FW_UPLOAD_PROG_IDLE, + FW_UPLOAD_PROG_RECEIVING, + FW_UPLOAD_PROG_PREPARING, + FW_UPLOAD_PROG_TRANSFERRING, + FW_UPLOAD_PROG_PROGRAMMING, + FW_UPLOAD_PROG_MAX +}; + +struct fw_upload_priv { + struct fw_upload *fw_upload; + struct module *module; + const char *name; + const struct fw_upload_ops *ops; + struct mutex lock; /* protect data structure contents */ + struct work_struct work; + const u8 *data; /* pointer to update data */ + u32 remaining_size; /* size remaining to transfer */ + enum fw_upload_prog progress; + enum fw_upload_prog err_progress; /* progress at time of failure */ + enum fw_upload_err err_code; /* security manager error code */ +}; + +#endif /* __SYSFS_UPLOAD_H */ diff --git a/drivers/base/init.c b/drivers/base/init.c index a9f57c22fb9e..9d2b06d65dfc 100644 --- a/drivers/base/init.c +++ b/drivers/base/init.c @@ -8,6 +8,7 @@ #include <linux/init.h> #include <linux/memory.h> #include <linux/of.h> +#include <linux/backing-dev.h> #include "base.h" @@ -20,6 +21,7 @@ void __init driver_init(void) { /* These are the core pieces */ + bdi_init(&noop_backing_dev_info); devtmpfs_init(); devices_init(); buses_init(); @@ -30,10 +32,12 @@ void __init driver_init(void) /* These are also core pieces, but must come after the * core core pieces. */ + faux_bus_init(); of_core_init(); platform_bus_init(); auxiliary_bus_init(); - cpu_dev_init(); memory_dev_init(); + node_dev_init(); + cpu_dev_init(); container_dev_init(); } diff --git a/drivers/base/isa.c b/drivers/base/isa.c index aa4737667026..bfd9215c9070 100644 --- a/drivers/base/isa.c +++ b/drivers/base/isa.c @@ -23,7 +23,7 @@ struct isa_dev { #define to_isa_dev(x) container_of((x), struct isa_dev, dev) -static int isa_bus_match(struct device *dev, struct device_driver *driver) +static int isa_bus_match(struct device *dev, const struct device_driver *driver) { struct isa_driver *isa_driver = to_isa_driver(driver); @@ -46,14 +46,12 @@ static int isa_bus_probe(struct device *dev) return 0; } -static int isa_bus_remove(struct device *dev) +static void isa_bus_remove(struct device *dev) { struct isa_driver *isa_driver = dev->platform_data; if (isa_driver && isa_driver->remove) isa_driver->remove(dev, to_isa_dev(dev)->id); - - return 0; } static void isa_bus_shutdown(struct device *dev) @@ -84,7 +82,7 @@ static int isa_bus_resume(struct device *dev) return 0; } -static struct bus_type isa_bus_type = { +static const struct bus_type isa_bus_type = { .name = "isa", .match = isa_bus_match, .probe = isa_bus_probe, @@ -151,11 +149,8 @@ int isa_register_driver(struct isa_driver *isa_driver, unsigned int ndev) break; } - if (isa_dev->dev.platform_data) { - isa_dev->next = isa_driver->devices; - isa_driver->devices = &isa_dev->dev; - } else - device_unregister(&isa_dev->dev); + isa_dev->next = isa_driver->devices; + isa_driver->devices = &isa_dev->dev; } if (!error && !isa_driver->devices) diff --git a/drivers/base/map.c b/drivers/base/map.c index 5650ab2b247a..83aeb09ca161 100644 --- a/drivers/base/map.c +++ b/drivers/base/map.c @@ -33,9 +33,9 @@ int kobj_map(struct kobj_map *domain, dev_t dev, unsigned long range, struct module *module, kobj_probe_t *probe, int (*lock)(dev_t, void *), void *data) { - unsigned n = MAJOR(dev + range - 1) - MAJOR(dev) + 1; - unsigned index = MAJOR(dev); - unsigned i; + unsigned int n = MAJOR(dev + range - 1) - MAJOR(dev) + 1; + unsigned int index = MAJOR(dev); + unsigned int i; struct probe *p; if (n > 255) @@ -67,9 +67,9 @@ int kobj_map(struct kobj_map *domain, dev_t dev, unsigned long range, void kobj_unmap(struct kobj_map *domain, dev_t dev, unsigned long range) { - unsigned n = MAJOR(dev + range - 1) - MAJOR(dev) + 1; - unsigned index = MAJOR(dev); - unsigned i; + unsigned int n = MAJOR(dev + range - 1) - MAJOR(dev) + 1; + unsigned int index = MAJOR(dev); + unsigned int i; struct probe *found = NULL; if (n > 255) diff --git a/drivers/base/memory.c b/drivers/base/memory.c index aa31a21f33d7..751f248ca4a8 100644 --- a/drivers/base/memory.c +++ b/drivers/base/memory.c @@ -22,6 +22,7 @@ #include <linux/stat.h> #include <linux/slab.h> #include <linux/xarray.h> +#include <linux/export.h> #include <linux/atomic.h> #include <linux/uaccess.h> @@ -48,27 +49,13 @@ int mhp_online_type_from_str(const char *str) #define to_memory_block(dev) container_of(dev, struct memory_block, dev) -static int sections_per_block; - -static inline unsigned long memory_block_id(unsigned long section_nr) -{ - return section_nr / sections_per_block; -} - -static inline unsigned long pfn_to_block_id(unsigned long pfn) -{ - return memory_block_id(pfn_to_section_nr(pfn)); -} - -static inline unsigned long phys_to_block_id(unsigned long phys) -{ - return pfn_to_block_id(PFN_DOWN(phys)); -} +int sections_per_block; +EXPORT_SYMBOL(sections_per_block); static int memory_subsys_online(struct device *dev); static int memory_subsys_offline(struct device *dev); -static struct bus_type memory_subsys = { +static const struct bus_type memory_subsys = { .name = MEMORY_CLASS_NAME, .dev_name = MEMORY_CLASS_NAME, .online = memory_subsys_online, @@ -82,6 +69,12 @@ static struct bus_type memory_subsys = { */ static DEFINE_XARRAY(memory_blocks); +/* + * Memory groups, indexed by memory group id (mgid). + */ +static DEFINE_XARRAY_FLAGS(memory_groups, XA_FLAGS_ALLOC); +#define MEMORY_GROUP_MARK_DYNAMIC XA_MARK_1 + static BLOCKING_NOTIFIER_HEAD(memory_chain); int register_memory_notifier(struct notifier_block *nb) @@ -99,28 +92,75 @@ EXPORT_SYMBOL(unregister_memory_notifier); static void memory_block_release(struct device *dev) { struct memory_block *mem = to_memory_block(dev); - + /* Verify that the altmap is freed */ + WARN_ON(mem->altmap); kfree(mem); } + +/* Max block size to be set by memory_block_advise_max_size */ +static unsigned long memory_block_advised_size; +static bool memory_block_advised_size_queried; + +/** + * memory_block_advise_max_size() - advise memory hotplug on the max suggested + * block size, usually for alignment. + * @size: suggestion for maximum block size. must be aligned on power of 2. + * + * Early boot software (pre-allocator init) may advise archs on the max block + * size. This value can only decrease after initialization, as the intent is + * to identify the largest supported alignment for all sources. + * + * Use of this value is arch-defined, as is min/max block size. + * + * Return: 0 on success + * -EINVAL if size is 0 or not pow2 aligned + * -EBUSY if value has already been probed + */ +int __init memory_block_advise_max_size(unsigned long size) +{ + if (!size || !is_power_of_2(size)) + return -EINVAL; + + if (memory_block_advised_size_queried) + return -EBUSY; + + if (memory_block_advised_size) + memory_block_advised_size = min(memory_block_advised_size, size); + else + memory_block_advised_size = size; + + return 0; +} + +/** + * memory_block_advised_max_size() - query advised max hotplug block size. + * + * After the first call, the value can never change. Callers looking for the + * actual block size should use memory_block_size_bytes. This interface is + * intended for use by arch-init when initializing the hotplug block size. + * + * Return: advised size in bytes, or 0 if never set. + */ +unsigned long memory_block_advised_max_size(void) +{ + memory_block_advised_size_queried = true; + return memory_block_advised_size; +} + unsigned long __weak memory_block_size_bytes(void) { return MIN_MEMORY_BLOCK_SIZE; } EXPORT_SYMBOL_GPL(memory_block_size_bytes); -/* - * Show the first physical section index (number) of this memory block. - */ +/* Show the memory block ID, relative to the memory block size */ static ssize_t phys_index_show(struct device *dev, struct device_attribute *attr, char *buf) { struct memory_block *mem = to_memory_block(dev); - unsigned long phys_index; - - phys_index = mem->start_section_nr / sections_per_block; - return sysfs_emit(buf, "%08lx\n", phys_index); + return sysfs_emit(buf, "%08lx\n", memory_block_id(mem->start_section_nr)); } /* @@ -158,26 +198,42 @@ static ssize_t state_show(struct device *dev, struct device_attribute *attr, break; default: WARN_ON(1); - return sysfs_emit(buf, "ERROR-UNKNOWN-%ld\n", mem->state); + return sysfs_emit(buf, "ERROR-UNKNOWN-%d\n", mem->state); } return sysfs_emit(buf, "%s\n", output); } -int memory_notify(unsigned long val, void *v) +int memory_notify(enum memory_block_state state, void *v) { - return blocking_notifier_call_chain(&memory_chain, val, v); + return blocking_notifier_call_chain(&memory_chain, state, v); } +#if defined(CONFIG_MEMORY_FAILURE) && defined(CONFIG_MEMORY_HOTPLUG) +static unsigned long memblk_nr_poison(struct memory_block *mem); +#else +static inline unsigned long memblk_nr_poison(struct memory_block *mem) +{ + return 0; +} +#endif + +/* + * Must acquire mem_hotplug_lock in write mode. + */ static int memory_block_online(struct memory_block *mem) { unsigned long start_pfn = section_nr_to_pfn(mem->start_section_nr); unsigned long nr_pages = PAGES_PER_SECTION * sections_per_block; - unsigned long nr_vmemmap_pages = mem->nr_vmemmap_pages; + unsigned long nr_vmemmap_pages = 0; struct zone *zone; int ret; - zone = zone_for_pfn_range(mem->online_type, mem->nid, start_pfn, nr_pages); + if (memblk_nr_poison(mem)) + return -EHWPOISON; + + zone = zone_for_pfn_range(mem->online_type, mem->nid, mem->group, + start_pfn, nr_pages); /* * Although vmemmap pages have a different lifecycle than the pages @@ -186,18 +242,22 @@ static int memory_block_online(struct memory_block *mem) * stage helps to keep accounting easier to follow - e.g vmemmaps * belong to the same zone as the memory they backed. */ + if (mem->altmap) + nr_vmemmap_pages = mem->altmap->free; + + mem_hotplug_begin(); if (nr_vmemmap_pages) { ret = mhp_init_memmap_on_memory(start_pfn, nr_vmemmap_pages, zone); if (ret) - return ret; + goto out; } ret = online_pages(start_pfn + nr_vmemmap_pages, - nr_pages - nr_vmemmap_pages, zone); + nr_pages - nr_vmemmap_pages, zone, mem->group); if (ret) { if (nr_vmemmap_pages) mhp_deinit_memmap_on_memory(start_pfn, nr_vmemmap_pages); - return ret; + goto out; } /* @@ -205,40 +265,56 @@ static int memory_block_online(struct memory_block *mem) * now already properly populated. */ if (nr_vmemmap_pages) - adjust_present_page_count(zone, nr_vmemmap_pages); + adjust_present_page_count(pfn_to_page(start_pfn), mem->group, + nr_vmemmap_pages); + mem->zone = zone; +out: + mem_hotplug_done(); return ret; } +/* + * Must acquire mem_hotplug_lock in write mode. + */ static int memory_block_offline(struct memory_block *mem) { unsigned long start_pfn = section_nr_to_pfn(mem->start_section_nr); unsigned long nr_pages = PAGES_PER_SECTION * sections_per_block; - unsigned long nr_vmemmap_pages = mem->nr_vmemmap_pages; - struct zone *zone; + unsigned long nr_vmemmap_pages = 0; int ret; + if (!mem->zone) + return -EINVAL; + /* * Unaccount before offlining, such that unpopulated zone and kthreads * can properly be torn down in offline_pages(). */ - if (nr_vmemmap_pages) { - zone = page_zone(pfn_to_page(start_pfn)); - adjust_present_page_count(zone, -nr_vmemmap_pages); - } + if (mem->altmap) + nr_vmemmap_pages = mem->altmap->free; + + mem_hotplug_begin(); + if (nr_vmemmap_pages) + adjust_present_page_count(pfn_to_page(start_pfn), mem->group, + -nr_vmemmap_pages); ret = offline_pages(start_pfn + nr_vmemmap_pages, - nr_pages - nr_vmemmap_pages); + nr_pages - nr_vmemmap_pages, mem->zone, mem->group); if (ret) { /* offline_pages() failed. Account back. */ if (nr_vmemmap_pages) - adjust_present_page_count(zone, nr_vmemmap_pages); - return ret; + adjust_present_page_count(pfn_to_page(start_pfn), + mem->group, nr_vmemmap_pages); + goto out; } if (nr_vmemmap_pages) mhp_deinit_memmap_on_memory(start_pfn, nr_vmemmap_pages); + mem->zone = NULL; +out: + mem_hotplug_done(); return ret; } @@ -374,12 +450,13 @@ static ssize_t phys_device_show(struct device *dev, #ifdef CONFIG_MEMORY_HOTREMOVE static int print_allowed_zone(char *buf, int len, int nid, + struct memory_group *group, unsigned long start_pfn, unsigned long nr_pages, int online_type, struct zone *default_zone) { struct zone *zone; - zone = zone_for_pfn_range(online_type, nid, start_pfn, nr_pages); + zone = zone_for_pfn_range(online_type, nid, group, start_pfn, nr_pages); if (zone == default_zone) return 0; @@ -392,9 +469,10 @@ static ssize_t valid_zones_show(struct device *dev, struct memory_block *mem = to_memory_block(dev); unsigned long start_pfn = section_nr_to_pfn(mem->start_section_nr); unsigned long nr_pages = PAGES_PER_SECTION * sections_per_block; + struct memory_group *group = mem->group; struct zone *default_zone; - int len = 0; - int nid; + int nid = mem->nid; + int len; /* * Check the existing zone. Make sure that we do that only on the @@ -402,27 +480,21 @@ static ssize_t valid_zones_show(struct device *dev, */ if (mem->state == MEM_ONLINE) { /* - * The block contains more than one zone can not be offlined. - * This can happen e.g. for ZONE_DMA and ZONE_DMA32 + * If !mem->zone, the memory block spans multiple zones and + * cannot get offlined. */ - default_zone = test_pages_in_a_zone(start_pfn, - start_pfn + nr_pages); - if (!default_zone) - return sysfs_emit(buf, "%s\n", "none"); - len += sysfs_emit_at(buf, len, "%s", default_zone->name); - goto out; + return sysfs_emit(buf, "%s\n", + mem->zone ? mem->zone->name : "none"); } - nid = mem->nid; - default_zone = zone_for_pfn_range(MMOP_ONLINE, nid, start_pfn, - nr_pages); + default_zone = zone_for_pfn_range(MMOP_ONLINE, nid, group, + start_pfn, nr_pages); - len += sysfs_emit_at(buf, len, "%s", default_zone->name); - len += print_allowed_zone(buf, len, nid, start_pfn, nr_pages, + len = sysfs_emit(buf, "%s", default_zone->name); + len += print_allowed_zone(buf, len, nid, group, start_pfn, nr_pages, MMOP_ONLINE_KERNEL, default_zone); - len += print_allowed_zone(buf, len, nid, start_pfn, nr_pages, + len += print_allowed_zone(buf, len, nid, group, start_pfn, nr_pages, MMOP_ONLINE_MOVABLE, default_zone); -out: len += sysfs_emit_at(buf, len, "\n"); return len; } @@ -453,7 +525,7 @@ static ssize_t auto_online_blocks_show(struct device *dev, struct device_attribute *attr, char *buf) { return sysfs_emit(buf, "%s\n", - online_type_to_str[mhp_default_online_type]); + online_type_to_str[mhp_get_default_online_type()]); } static ssize_t auto_online_blocks_store(struct device *dev, @@ -465,12 +537,22 @@ static ssize_t auto_online_blocks_store(struct device *dev, if (online_type < 0) return -EINVAL; - mhp_default_online_type = online_type; + mhp_set_default_online_type(online_type); return count; } static DEVICE_ATTR_RW(auto_online_blocks); +#ifdef CONFIG_CRASH_HOTPLUG +#include <linux/kexec.h> +static ssize_t crash_hotplug_show(struct device *dev, + struct device_attribute *attr, char *buf) +{ + return sysfs_emit(buf, "%d\n", crash_check_hotplug_support()); +} +static DEVICE_ATTR_RO(crash_hotplug); +#endif + /* * Some architectures will have custom drivers to do this, and * will not need to do it from userspace. The fake hot-add code @@ -546,7 +628,9 @@ static ssize_t hard_offline_page_store(struct device *dev, if (kstrtoull(buf, 0, &pfn) < 0) return -EINVAL; pfn >>= PAGE_SHIFT; - ret = memory_failure(pfn, 0); + ret = memory_failure(pfn, MF_SW_SIMULATED); + if (ret == -EOPNOTSUPP) + ret = 0; return ret ? ret : count; } @@ -565,7 +649,7 @@ int __weak arch_get_memory_phys_device(unsigned long start_pfn) * * Called under device_hotplug_lock. */ -static struct memory_block *find_memory_block_by_id(unsigned long block_id) +struct memory_block *find_memory_block_by_id(unsigned long block_id) { struct memory_block *mem; @@ -578,9 +662,9 @@ static struct memory_block *find_memory_block_by_id(unsigned long block_id) /* * Called under device_hotplug_lock. */ -struct memory_block *find_memory_block(struct mem_section *section) +struct memory_block *find_memory_block(unsigned long section_nr) { - unsigned long block_id = memory_block_id(__section_nr(section)); + unsigned long block_id = memory_block_id(section_nr); return find_memory_block_by_id(block_id); } @@ -605,11 +689,7 @@ static const struct attribute_group *memory_memblk_attr_groups[] = { NULL, }; -/* - * register_memory - Setup a sysfs device for a memory block - */ -static -int register_memory(struct memory_block *memory) +static int __add_memory_block(struct memory_block *memory) { int ret; @@ -626,15 +706,91 @@ int register_memory(struct memory_block *memory) } ret = xa_err(xa_store(&memory_blocks, memory->dev.id, memory, GFP_KERNEL)); - if (ret) { - put_device(&memory->dev); + if (ret) device_unregister(&memory->dev); - } + return ret; } -static int init_memory_block(unsigned long block_id, unsigned long state, - unsigned long nr_vmemmap_pages) +static struct zone *early_node_zone_for_memory_block(struct memory_block *mem, + int nid) +{ + const unsigned long start_pfn = section_nr_to_pfn(mem->start_section_nr); + const unsigned long nr_pages = PAGES_PER_SECTION * sections_per_block; + struct zone *zone, *matching_zone = NULL; + pg_data_t *pgdat = NODE_DATA(nid); + int i; + + /* + * This logic only works for early memory, when the applicable zones + * already span the memory block. We don't expect overlapping zones on + * a single node for early memory. So if we're told that some PFNs + * of a node fall into this memory block, we can assume that all node + * zones that intersect with the memory block are actually applicable. + * No need to look at the memmap. + */ + for (i = 0; i < MAX_NR_ZONES; i++) { + zone = pgdat->node_zones + i; + if (!populated_zone(zone)) + continue; + if (!zone_intersects(zone, start_pfn, nr_pages)) + continue; + if (!matching_zone) { + matching_zone = zone; + continue; + } + /* Spans multiple zones ... */ + matching_zone = NULL; + break; + } + return matching_zone; +} + +#ifdef CONFIG_NUMA +/** + * memory_block_add_nid_early() - Indicate that early system RAM falling into + * this memory block device (partially) belongs + * to the given node. + * @mem: The memory block device. + * @nid: The node id. + * + * Indicate that early system RAM falling into this memory block (partially) + * belongs to the given node. This will also properly set/adjust mem->zone based + * on the zone ranges of the given node. + * + * Memory hotplug handles this on memory block creation, where we can only have + * a single nid span a memory block. + */ +void memory_block_add_nid_early(struct memory_block *mem, int nid) +{ + if (mem->nid != nid) { + /* + * For early memory we have to determine the zone when setting + * the node id and handle multiple nodes spanning a single + * memory block by indicate via zone == NULL that we're not + * dealing with a single zone. So if we're setting the node id + * the first time, determine if there is a single zone. If we're + * setting the node id a second time to a different node, + * invalidate the single detected zone. + */ + if (mem->nid == NUMA_NO_NODE) + mem->zone = early_node_zone_for_memory_block(mem, nid); + else + mem->zone = NULL; + /* + * If this memory block spans multiple nodes, we only indicate + * the last processed node. If we span multiple nodes (not applicable + * to hotplugged memory), zone == NULL will prohibit memory offlining + * and consequently unplug. + */ + mem->nid = nid; + } +} +#endif + +static int add_memory_block(unsigned long block_id, int nid, unsigned long state, + struct vmem_altmap *altmap, + struct memory_group *group) { struct memory_block *mem; int ret = 0; @@ -650,37 +806,45 @@ static int init_memory_block(unsigned long block_id, unsigned long state, mem->start_section_nr = block_id * sections_per_block; mem->state = state; - mem->nid = NUMA_NO_NODE; - mem->nr_vmemmap_pages = nr_vmemmap_pages; - - ret = register_memory(mem); + mem->nid = nid; + mem->altmap = altmap; + INIT_LIST_HEAD(&mem->group_next); - return ret; -} +#ifndef CONFIG_NUMA + if (state == MEM_ONLINE) + /* + * MEM_ONLINE at this point implies early memory. With NUMA, + * we'll determine the zone when setting the node id via + * memory_block_add_nid(). Memory hotplug updated the zone + * manually when memory onlining/offlining succeeds. + */ + mem->zone = early_node_zone_for_memory_block(mem, NUMA_NO_NODE); +#endif /* CONFIG_NUMA */ -static int add_memory_block(unsigned long base_section_nr) -{ - int section_count = 0; - unsigned long nr; + ret = __add_memory_block(mem); + if (ret) + return ret; - for (nr = base_section_nr; nr < base_section_nr + sections_per_block; - nr++) - if (present_section_nr(nr)) - section_count++; + if (group) { + mem->group = group; + list_add(&mem->group_next, &group->memory_blocks); + } - if (section_count == 0) - return 0; - return init_memory_block(memory_block_id(base_section_nr), - MEM_ONLINE, 0); + return 0; } -static void unregister_memory(struct memory_block *memory) +static void remove_memory_block(struct memory_block *memory) { if (WARN_ON_ONCE(memory->dev.bus != &memory_subsys)) return; WARN_ON(xa_erase(&memory_blocks, memory->dev.id) == NULL); + if (memory->group) { + list_del(&memory->group_next); + memory->group = NULL; + } + /* drop the ref. we got via find_memory_block() */ put_device(&memory->dev); device_unregister(&memory->dev); @@ -694,7 +858,8 @@ static void unregister_memory(struct memory_block *memory) * Called under device_hotplug_lock. */ int create_memory_block_devices(unsigned long start, unsigned long size, - unsigned long vmemmap_pages) + int nid, struct vmem_altmap *altmap, + struct memory_group *group) { const unsigned long start_block_id = pfn_to_block_id(PFN_DOWN(start)); unsigned long end_block_id = pfn_to_block_id(PFN_DOWN(start + size)); @@ -707,7 +872,7 @@ int create_memory_block_devices(unsigned long start, unsigned long size, return -EINVAL; for (block_id = start_block_id; block_id != end_block_id; block_id++) { - ret = init_memory_block(block_id, MEM_OFFLINE, vmemmap_pages); + ret = add_memory_block(block_id, nid, MEM_OFFLINE, altmap, group); if (ret) break; } @@ -718,7 +883,7 @@ int create_memory_block_devices(unsigned long start, unsigned long size, mem = find_memory_block_by_id(block_id); if (WARN_ON_ONCE(!mem)) continue; - unregister_memory(mem); + remove_memory_block(mem); } } return ret; @@ -746,17 +911,12 @@ void remove_memory_block_devices(unsigned long start, unsigned long size) mem = find_memory_block_by_id(block_id); if (WARN_ON_ONCE(!mem)) continue; + num_poisoned_pages_sub(-1UL, memblk_nr_poison(mem)); unregister_memory_block_under_nodes(mem); - unregister_memory(mem); + remove_memory_block(mem); } } -/* return true if the memory block is offlined, otherwise, return false */ -bool is_memblock_offlined(struct memory_block *mem) -{ - return mem->state == MEM_OFFLINE; -} - static struct attribute *memory_root_attrs[] = { #ifdef CONFIG_ARCH_MEMORY_PROBE &dev_attr_probe.attr, @@ -769,6 +929,9 @@ static struct attribute *memory_root_attrs[] = { &dev_attr_block_size_bytes.attr, &dev_attr_auto_online_blocks.attr, +#ifdef CONFIG_CRASH_HOTPLUG + &dev_attr_crash_hotplug.attr, +#endif NULL }; @@ -789,7 +952,7 @@ static const struct attribute_group *memory_root_attr_groups[] = { void __init memory_dev_init(void) { int ret; - unsigned long block_sz, nr; + unsigned long block_sz, block_id, nr; /* Validate the configured memory block size */ block_sz = memory_block_size_bytes(); @@ -802,15 +965,23 @@ void __init memory_dev_init(void) panic("%s() failed to register subsystem: %d\n", __func__, ret); /* - * Create entries for memory sections that were found - * during boot and have been initialized + * Create entries for memory sections that were found during boot + * and have been initialized. Use @block_id to track the last + * handled block and initialize it to an invalid value (ULONG_MAX) + * to bypass the block ID matching check for the first present + * block so that it can be covered. */ - for (nr = 0; nr <= __highest_present_section_nr; - nr += sections_per_block) { - ret = add_memory_block(nr); - if (ret) - panic("%s() failed to add memory block: %d\n", __func__, - ret); + block_id = ULONG_MAX; + for_each_present_section_nr(0, nr) { + if (block_id != ULONG_MAX && memory_block_id(nr) == block_id) + continue; + + block_id = memory_block_id(nr); + ret = add_memory_block(block_id, NUMA_NO_NODE, MEM_ONLINE, NULL, NULL); + if (ret) { + panic("%s() failed to add memory block: %d\n", + __func__, ret); + } } } @@ -891,3 +1062,189 @@ int for_each_memory_block(void *arg, walk_memory_blocks_func_t func) return bus_for_each_dev(&memory_subsys, NULL, &cb_data, for_each_memory_block_cb); } + +/* + * This is an internal helper to unify allocation and initialization of + * memory groups. Note that the passed memory group will be copied to a + * dynamically allocated memory group. After this call, the passed + * memory group should no longer be used. + */ +static int memory_group_register(struct memory_group group) +{ + struct memory_group *new_group; + uint32_t mgid; + int ret; + + if (!node_possible(group.nid)) + return -EINVAL; + + new_group = kzalloc(sizeof(group), GFP_KERNEL); + if (!new_group) + return -ENOMEM; + *new_group = group; + INIT_LIST_HEAD(&new_group->memory_blocks); + + ret = xa_alloc(&memory_groups, &mgid, new_group, xa_limit_31b, + GFP_KERNEL); + if (ret) { + kfree(new_group); + return ret; + } else if (group.is_dynamic) { + xa_set_mark(&memory_groups, mgid, MEMORY_GROUP_MARK_DYNAMIC); + } + return mgid; +} + +/** + * memory_group_register_static() - Register a static memory group. + * @nid: The node id. + * @max_pages: The maximum number of pages we'll have in this static memory + * group. + * + * Register a new static memory group and return the memory group id. + * All memory in the group belongs to a single unit, such as a DIMM. All + * memory belonging to a static memory group is added in one go to be removed + * in one go -- it's static. + * + * Returns an error if out of memory, if the node id is invalid, if no new + * memory groups can be registered, or if max_pages is invalid (0). Otherwise, + * returns the new memory group id. + */ +int memory_group_register_static(int nid, unsigned long max_pages) +{ + struct memory_group group = { + .nid = nid, + .s = { + .max_pages = max_pages, + }, + }; + + if (!max_pages) + return -EINVAL; + return memory_group_register(group); +} +EXPORT_SYMBOL_GPL(memory_group_register_static); + +/** + * memory_group_register_dynamic() - Register a dynamic memory group. + * @nid: The node id. + * @unit_pages: Unit in pages in which is memory added/removed in this dynamic + * memory group. + * + * Register a new dynamic memory group and return the memory group id. + * Memory within a dynamic memory group is added/removed dynamically + * in unit_pages. + * + * Returns an error if out of memory, if the node id is invalid, if no new + * memory groups can be registered, or if unit_pages is invalid (0, not a + * power of two, smaller than a single memory block). Otherwise, returns the + * new memory group id. + */ +int memory_group_register_dynamic(int nid, unsigned long unit_pages) +{ + struct memory_group group = { + .nid = nid, + .is_dynamic = true, + .d = { + .unit_pages = unit_pages, + }, + }; + + if (!unit_pages || !is_power_of_2(unit_pages) || + unit_pages < PHYS_PFN(memory_block_size_bytes())) + return -EINVAL; + return memory_group_register(group); +} +EXPORT_SYMBOL_GPL(memory_group_register_dynamic); + +/** + * memory_group_unregister() - Unregister a memory group. + * @mgid: the memory group id + * + * Unregister a memory group. If any memory block still belongs to this + * memory group, unregistering will fail. + * + * Returns -EINVAL if the memory group id is invalid, returns -EBUSY if some + * memory blocks still belong to this memory group and returns 0 if + * unregistering succeeded. + */ +int memory_group_unregister(int mgid) +{ + struct memory_group *group; + + if (mgid < 0) + return -EINVAL; + + group = xa_load(&memory_groups, mgid); + if (!group) + return -EINVAL; + if (!list_empty(&group->memory_blocks)) + return -EBUSY; + xa_erase(&memory_groups, mgid); + kfree(group); + return 0; +} +EXPORT_SYMBOL_GPL(memory_group_unregister); + +/* + * This is an internal helper only to be used in core memory hotplug code to + * lookup a memory group. We don't care about locking, as we don't expect a + * memory group to get unregistered while adding memory to it -- because + * the group and the memory is managed by the same driver. + */ +struct memory_group *memory_group_find_by_id(int mgid) +{ + return xa_load(&memory_groups, mgid); +} + +/* + * This is an internal helper only to be used in core memory hotplug code to + * walk all dynamic memory groups excluding a given memory group, either + * belonging to a specific node, or belonging to any node. + */ +int walk_dynamic_memory_groups(int nid, walk_memory_groups_func_t func, + struct memory_group *excluded, void *arg) +{ + struct memory_group *group; + unsigned long index; + int ret = 0; + + xa_for_each_marked(&memory_groups, index, group, + MEMORY_GROUP_MARK_DYNAMIC) { + if (group == excluded) + continue; +#ifdef CONFIG_NUMA + if (nid != NUMA_NO_NODE && group->nid != nid) + continue; +#endif /* CONFIG_NUMA */ + ret = func(group, arg); + if (ret) + break; + } + return ret; +} + +#if defined(CONFIG_MEMORY_FAILURE) && defined(CONFIG_MEMORY_HOTPLUG) +void memblk_nr_poison_inc(unsigned long pfn) +{ + const unsigned long block_id = pfn_to_block_id(pfn); + struct memory_block *mem = find_memory_block_by_id(block_id); + + if (mem) + atomic_long_inc(&mem->nr_hwpoison); +} + +void memblk_nr_poison_sub(unsigned long pfn, long i) +{ + const unsigned long block_id = pfn_to_block_id(pfn); + struct memory_block *mem = find_memory_block_by_id(block_id); + + if (mem) + atomic_long_sub(i, &mem->nr_hwpoison); +} + +static unsigned long memblk_nr_poison(struct memory_block *mem) +{ + return atomic_long_read(&mem->nr_hwpoison); +} +#endif diff --git a/drivers/base/module.c b/drivers/base/module.c index 46ad4d636731..218aaa096455 100644 --- a/drivers/base/module.c +++ b/drivers/base/module.c @@ -9,7 +9,7 @@ #include <linux/string.h> #include "base.h" -static char *make_driver_name(struct device_driver *drv) +static char *make_driver_name(const struct device_driver *drv) { char *driver_name; @@ -30,46 +30,67 @@ static void module_create_drivers_dir(struct module_kobject *mk) mutex_unlock(&drivers_dir_mutex); } -void module_add_driver(struct module *mod, struct device_driver *drv) +int module_add_driver(struct module *mod, const struct device_driver *drv) { char *driver_name; - int no_warn; struct module_kobject *mk = NULL; + int ret; if (!drv) - return; + return 0; if (mod) mk = &mod->mkobj; else if (drv->mod_name) { - struct kobject *mkobj; - - /* Lookup built-in module entry in /sys/modules */ - mkobj = kset_find_obj(module_kset, drv->mod_name); - if (mkobj) { - mk = container_of(mkobj, struct module_kobject, kobj); + /* Lookup or create built-in module entry in /sys/modules */ + mk = lookup_or_create_module_kobject(drv->mod_name); + if (mk) { /* remember our module structure */ drv->p->mkobj = mk; - /* kset_find_obj took a reference */ - kobject_put(mkobj); + /* lookup_or_create_module_kobject took a reference */ + kobject_put(&mk->kobj); } } if (!mk) - return; + return 0; + + ret = sysfs_create_link(&drv->p->kobj, &mk->kobj, "module"); + if (ret) + return ret; - /* Don't check return codes; these calls are idempotent */ - no_warn = sysfs_create_link(&drv->p->kobj, &mk->kobj, "module"); driver_name = make_driver_name(drv); - if (driver_name) { - module_create_drivers_dir(mk); - no_warn = sysfs_create_link(mk->drivers_dir, &drv->p->kobj, - driver_name); - kfree(driver_name); + if (!driver_name) { + ret = -ENOMEM; + goto out_remove_kobj; } + + module_create_drivers_dir(mk); + if (!mk->drivers_dir) { + ret = -EINVAL; + goto out_free_driver_name; + } + + ret = sysfs_create_link(mk->drivers_dir, &drv->p->kobj, driver_name); + if (ret) + goto out_remove_drivers_dir; + + kfree(driver_name); + + return 0; + +out_remove_drivers_dir: + sysfs_remove_link(mk->drivers_dir, driver_name); + +out_free_driver_name: + kfree(driver_name); + +out_remove_kobj: + sysfs_remove_link(&drv->p->kobj, "module"); + return ret; } -void module_remove_driver(struct device_driver *drv) +void module_remove_driver(const struct device_driver *drv) { struct module_kobject *mk = NULL; char *driver_name; diff --git a/drivers/base/node.c b/drivers/base/node.c index 4a4ae868ad9f..00cf4532f121 100644 --- a/drivers/base/node.c +++ b/drivers/base/node.c @@ -7,6 +7,7 @@ #include <linux/init.h> #include <linux/mm.h> #include <linux/memory.h> +#include <linux/mempolicy.h> #include <linux/vmstat.h> #include <linux/notifier.h> #include <linux/node.h> @@ -20,49 +21,54 @@ #include <linux/pm_runtime.h> #include <linux/swap.h> #include <linux/slab.h> +#include <linux/memblock.h> -static struct bus_type node_subsys = { +static const struct bus_type node_subsys = { .name = "node", .dev_name = "node", }; - -static ssize_t node_read_cpumap(struct device *dev, bool list, char *buf) +static inline ssize_t cpumap_read(struct file *file, struct kobject *kobj, + const struct bin_attribute *attr, char *buf, + loff_t off, size_t count) { - ssize_t n; - cpumask_var_t mask; + struct device *dev = kobj_to_dev(kobj); struct node *node_dev = to_node(dev); - - /* 2008/04/07: buf currently PAGE_SIZE, need 9 chars per 32 bits. */ - BUILD_BUG_ON((NR_CPUS/32 * 9) > (PAGE_SIZE-1)); + cpumask_var_t mask; + ssize_t n; if (!alloc_cpumask_var(&mask, GFP_KERNEL)) return 0; cpumask_and(mask, cpumask_of_node(node_dev->dev.id), cpu_online_mask); - n = cpumap_print_to_pagebuf(list, buf, mask); + n = cpumap_print_bitmask_to_buf(buf, mask, off, count); free_cpumask_var(mask); return n; } -static inline ssize_t cpumap_show(struct device *dev, - struct device_attribute *attr, - char *buf) +static const BIN_ATTR_RO(cpumap, CPUMAP_FILE_MAX_BYTES); + +static inline ssize_t cpulist_read(struct file *file, struct kobject *kobj, + const struct bin_attribute *attr, char *buf, + loff_t off, size_t count) { - return node_read_cpumap(dev, false, buf); -} + struct device *dev = kobj_to_dev(kobj); + struct node *node_dev = to_node(dev); + cpumask_var_t mask; + ssize_t n; -static DEVICE_ATTR_RO(cpumap); + if (!alloc_cpumask_var(&mask, GFP_KERNEL)) + return 0; -static inline ssize_t cpulist_show(struct device *dev, - struct device_attribute *attr, - char *buf) -{ - return node_read_cpumap(dev, true, buf); + cpumask_and(mask, cpumask_of_node(node_dev->dev.id), cpu_online_mask); + n = cpumap_print_list_to_buf(buf, mask, off, count); + free_cpumask_var(mask); + + return n; } -static DEVICE_ATTR_RO(cpulist); +static const BIN_ATTR_RO(cpulist, CPULIST_FILE_MAX_BYTES); /** * struct node_access_nodes - Access class device to hold user visible @@ -70,14 +76,14 @@ static DEVICE_ATTR_RO(cpulist); * @dev: Device for this memory access class * @list_node: List element in the node's access list * @access: The access class rank - * @hmem_attrs: Heterogeneous memory performance attributes + * @coord: Heterogeneous memory performance coordinates */ struct node_access_nodes { struct device dev; struct list_head list_node; - unsigned access; + unsigned int access; #ifdef CONFIG_HMEM_REPORTING - struct node_hmem_attrs hmem_attrs; + struct access_coordinate coord; #endif }; #define to_access_nodes(dev) container_of(dev, struct node_access_nodes, dev) @@ -106,6 +112,27 @@ static const struct attribute_group *node_access_node_groups[] = { NULL, }; +#ifdef CONFIG_MEMORY_HOTPLUG +static BLOCKING_NOTIFIER_HEAD(node_chain); + +int register_node_notifier(struct notifier_block *nb) +{ + return blocking_notifier_chain_register(&node_chain, nb); +} +EXPORT_SYMBOL(register_node_notifier); + +void unregister_node_notifier(struct notifier_block *nb) +{ + blocking_notifier_chain_unregister(&node_chain, nb); +} +EXPORT_SYMBOL(unregister_node_notifier); + +int node_notify(unsigned long val, void *v) +{ + return blocking_notifier_call_chain(&node_chain, val, v); +} +#endif + static void node_remove_accesses(struct node *node) { struct node_access_nodes *c, *cnext; @@ -122,7 +149,7 @@ static void node_access_release(struct device *dev) } static struct node_access_nodes *node_init_node_access(struct node *node, - unsigned access) + enum access_coordinate_class access) { struct node_access_nodes *access_node; struct device *dev; @@ -157,15 +184,15 @@ free: } #ifdef CONFIG_HMEM_REPORTING -#define ACCESS_ATTR(name) \ -static ssize_t name##_show(struct device *dev, \ +#define ACCESS_ATTR(property) \ +static ssize_t property##_show(struct device *dev, \ struct device_attribute *attr, \ char *buf) \ { \ return sysfs_emit(buf, "%u\n", \ - to_access_nodes(dev)->hmem_attrs.name); \ + to_access_nodes(dev)->coord.property); \ } \ -static DEVICE_ATTR_RO(name) +static DEVICE_ATTR_RO(property) ACCESS_ATTR(read_bandwidth); ACCESS_ATTR(read_latency); @@ -183,11 +210,11 @@ static struct attribute *access_attrs[] = { /** * node_set_perf_attrs - Set the performance values for given access class * @nid: Node identifier to be set - * @hmem_attrs: Heterogeneous memory performance attributes + * @coord: Heterogeneous memory performance coordinates * @access: The access class the for the given attributes */ -void node_set_perf_attrs(unsigned int nid, struct node_hmem_attrs *hmem_attrs, - unsigned access) +void node_set_perf_attrs(unsigned int nid, struct access_coordinate *coord, + enum access_coordinate_class access) { struct node_access_nodes *c; struct node *node; @@ -201,7 +228,7 @@ void node_set_perf_attrs(unsigned int nid, struct node_hmem_attrs *hmem_attrs, if (!c) return; - c->hmem_attrs = *hmem_attrs; + c->coord = *coord; for (i = 0; access_attrs[i] != NULL; i++) { if (sysfs_add_file_to_group(&c->dev.kobj, access_attrs[i], "initiators")) { @@ -210,7 +237,54 @@ void node_set_perf_attrs(unsigned int nid, struct node_hmem_attrs *hmem_attrs, break; } } + + /* When setting CPU access coordinates, update mempolicy */ + if (access == ACCESS_COORDINATE_CPU) { + if (mempolicy_set_node_perf(nid, coord)) { + pr_info("failed to set mempolicy attrs for node %d\n", + nid); + } + } } +EXPORT_SYMBOL_GPL(node_set_perf_attrs); + +/** + * node_update_perf_attrs - Update the performance values for given access class + * @nid: Node identifier to be updated + * @coord: Heterogeneous memory performance coordinates + * @access: The access class for the given attributes + */ +void node_update_perf_attrs(unsigned int nid, struct access_coordinate *coord, + enum access_coordinate_class access) +{ + struct node_access_nodes *access_node; + struct node *node; + int i; + + if (WARN_ON_ONCE(!node_online(nid))) + return; + + node = node_devices[nid]; + list_for_each_entry(access_node, &node->access_list, list_node) { + if (access_node->access != access) + continue; + + access_node->coord = *coord; + for (i = 0; access_attrs[i]; i++) { + sysfs_notify(&access_node->dev.kobj, + NULL, access_attrs[i]->name); + } + break; + } + + /* When setting CPU access coordinates, update mempolicy */ + if (access != ACCESS_COORDINATE_CPU) + return; + + if (mempolicy_set_node_perf(nid, coord)) + pr_info("failed to set mempolicy attrs for node %d\n", nid); +} +EXPORT_SYMBOL_GPL(node_update_perf_attrs); /** * struct node_cache_info - Internal tracking for memory node caches @@ -239,12 +313,14 @@ CACHE_ATTR(size, "%llu") CACHE_ATTR(line_size, "%u") CACHE_ATTR(indexing, "%u") CACHE_ATTR(write_policy, "%u") +CACHE_ATTR(address_mode, "%#x") static struct attribute *cache_attrs[] = { &dev_attr_indexing.attr, &dev_attr_size.attr, &dev_attr_line_size.attr, &dev_attr_write_policy.attr, + &dev_attr_address_mode.attr, NULL, }; ATTRIBUTE_GROUPS(cache); @@ -429,6 +505,7 @@ static ssize_t node_read_meminfo(struct device *dev, "Node %d ShadowCallStack:%8lu kB\n" #endif "Node %d PageTables: %8lu kB\n" + "Node %d SecPageTables: %8lu kB\n" "Node %d NFS_Unstable: %8lu kB\n" "Node %d Bounce: %8lu kB\n" "Node %d WritebackTmp: %8lu kB\n" @@ -440,8 +517,11 @@ static ssize_t node_read_meminfo(struct device *dev, "Node %d AnonHugePages: %8lu kB\n" "Node %d ShmemHugePages: %8lu kB\n" "Node %d ShmemPmdMapped: %8lu kB\n" - "Node %d FileHugePages: %8lu kB\n" - "Node %d FilePmdMapped: %8lu kB\n" + "Node %d FileHugePages: %8lu kB\n" + "Node %d FilePmdMapped: %8lu kB\n" +#endif +#ifdef CONFIG_UNACCEPTED_MEMORY + "Node %d Unaccepted: %8lu kB\n" #endif , nid, K(node_page_state(pgdat, NR_FILE_DIRTY)), @@ -455,9 +535,10 @@ static ssize_t node_read_meminfo(struct device *dev, nid, node_page_state(pgdat, NR_KERNEL_SCS_KB), #endif nid, K(node_page_state(pgdat, NR_PAGETABLE)), + nid, K(node_page_state(pgdat, NR_SECONDARY_PAGETABLE)), + nid, 0UL, + nid, 0UL, nid, 0UL, - nid, K(sum_zone_node_page_state(nid, NR_BOUNCE)), - nid, K(node_page_state(pgdat, NR_WRITEBACK_TEMP)), nid, K(sreclaimable + node_page_state(pgdat, NR_KERNEL_MISC_RECLAIMABLE)), nid, K(sreclaimable + sunreclaimable), @@ -471,6 +552,10 @@ static ssize_t node_read_meminfo(struct device *dev, nid, K(node_page_state(pgdat, NR_FILE_THPS)), nid, K(node_page_state(pgdat, NR_FILE_PMDMAPPED)) #endif +#ifdef CONFIG_UNACCEPTED_MEMORY + , + nid, K(sum_zone_node_page_state(nid, NR_UNACCEPTED)) +#endif ); len += hugetlb_report_node_meminfo(buf, len, nid); return len; @@ -557,115 +642,38 @@ static ssize_t node_read_distance(struct device *dev, static DEVICE_ATTR(distance, 0444, node_read_distance, NULL); static struct attribute *node_dev_attrs[] = { - &dev_attr_cpumap.attr, - &dev_attr_cpulist.attr, &dev_attr_meminfo.attr, &dev_attr_numastat.attr, &dev_attr_distance.attr, &dev_attr_vmstat.attr, NULL }; -ATTRIBUTE_GROUPS(node_dev); - -#ifdef CONFIG_HUGETLBFS -/* - * hugetlbfs per node attributes registration interface: - * When/if hugetlb[fs] subsystem initializes [sometime after this module], - * it will register its per node attributes for all online nodes with - * memory. It will also call register_hugetlbfs_with_node(), below, to - * register its attribute registration functions with this node driver. - * Once these hooks have been initialized, the node driver will call into - * the hugetlb module to [un]register attributes for hot-plugged nodes. - */ -static node_registration_func_t __hugetlb_register_node; -static node_registration_func_t __hugetlb_unregister_node; -static inline bool hugetlb_register_node(struct node *node) -{ - if (__hugetlb_register_node && - node_state(node->dev.id, N_MEMORY)) { - __hugetlb_register_node(node); - return true; - } - return false; -} - -static inline void hugetlb_unregister_node(struct node *node) -{ - if (__hugetlb_unregister_node) - __hugetlb_unregister_node(node); -} +static const struct bin_attribute *node_dev_bin_attrs[] = { + &bin_attr_cpumap, + &bin_attr_cpulist, + NULL +}; -void register_hugetlbfs_with_node(node_registration_func_t doregister, - node_registration_func_t unregister) -{ - __hugetlb_register_node = doregister; - __hugetlb_unregister_node = unregister; -} -#else -static inline void hugetlb_register_node(struct node *node) {} +static const struct attribute_group node_dev_group = { + .attrs = node_dev_attrs, + .bin_attrs = node_dev_bin_attrs, +}; -static inline void hugetlb_unregister_node(struct node *node) {} +static const struct attribute_group *node_dev_groups[] = { + &node_dev_group, +#ifdef CONFIG_HAVE_ARCH_NODE_DEV_GROUP + &arch_node_dev_group, #endif - -static void node_device_release(struct device *dev) -{ - struct node *node = to_node(dev); - -#if defined(CONFIG_MEMORY_HOTPLUG_SPARSE) && defined(CONFIG_HUGETLBFS) - /* - * We schedule the work only when a memory section is - * onlined/offlined on this node. When we come here, - * all the memory on this node has been offlined, - * so we won't enqueue new work to this work. - * - * The work is using node->node_work, so we should - * flush work before freeing the memory. - */ - flush_work(&node->node_work); +#ifdef CONFIG_MEMORY_FAILURE + &memory_failure_attr_group, #endif - kfree(node); -} - -/* - * register_node - Setup a sysfs device for a node. - * @num - Node number to use when creating the device. - * - * Initialize and register the node device. - */ -static int register_node(struct node *node, int num) -{ - int error; - - node->dev.id = num; - node->dev.bus = &node_subsys; - node->dev.release = node_device_release; - node->dev.groups = node_dev_groups; - error = device_register(&node->dev); - - if (error) - put_device(&node->dev); - else { - hugetlb_register_node(node); - - compaction_register_node(node); - } - return error; -} + NULL +}; -/** - * unregister_node - unregister a node device - * @node: node going away - * - * Unregisters a node device @node. All the devices on the node must be - * unregistered before calling this function. - */ -void unregister_node(struct node *node) +static void node_device_release(struct device *dev) { - hugetlb_unregister_node(node); /* no-op, if memoryless node */ - node_remove_accesses(node); - node_remove_caches(node); - device_unregister(&node->dev); + kfree(to_node(dev)); } struct node *node_devices[MAX_NUMNODES]; @@ -711,7 +719,7 @@ int register_cpu_under_node(unsigned int cpu, unsigned int nid) */ int register_memory_node_under_compute_node(unsigned int mem_nid, unsigned int cpu_nid, - unsigned access) + enum access_coordinate_class access) { struct node *init_node, *targ_node; struct node_access_nodes *initiator, *target; @@ -765,29 +773,12 @@ int unregister_cpu_under_node(unsigned int cpu, unsigned int nid) return 0; } -#ifdef CONFIG_MEMORY_HOTPLUG_SPARSE -static int __ref get_nid_for_pfn(unsigned long pfn) -{ - if (!pfn_valid_within(pfn)) - return -1; -#ifdef CONFIG_DEFERRED_STRUCT_PAGE_INIT - if (system_state < SYSTEM_RUNNING) - return early_pfn_to_nid(pfn); -#endif - return pfn_to_nid(pfn); -} - +#ifdef CONFIG_MEMORY_HOTPLUG static void do_register_memory_block_under_node(int nid, struct memory_block *mem_blk) { int ret; - /* - * If this memory block spans multiple nodes, we only indicate - * the last processed node. - */ - mem_blk->nid = nid; - ret = sysfs_create_link_nowarn(&node_devices[nid]->dev.kobj, &mem_blk->dev.kobj, kobject_name(&mem_blk->dev.kobj)); @@ -806,46 +797,6 @@ static void do_register_memory_block_under_node(int nid, ret); } -/* register memory section under specified node if it spans that node */ -static int register_mem_block_under_node_early(struct memory_block *mem_blk, - void *arg) -{ - unsigned long memory_block_pfns = memory_block_size_bytes() / PAGE_SIZE; - unsigned long start_pfn = section_nr_to_pfn(mem_blk->start_section_nr); - unsigned long end_pfn = start_pfn + memory_block_pfns - 1; - int nid = *(int *)arg; - unsigned long pfn; - - for (pfn = start_pfn; pfn <= end_pfn; pfn++) { - int page_nid; - - /* - * memory block could have several absent sections from start. - * skip pfn range from absent section - */ - if (!pfn_in_present_section(pfn)) { - pfn = round_down(pfn + PAGES_PER_SECTION, - PAGES_PER_SECTION) - 1; - continue; - } - - /* - * We need to check if page belongs to nid only at the boot - * case because node's ranges can be interleaved. - */ - page_nid = get_nid_for_pfn(pfn); - if (page_nid < 0) - continue; - if (page_nid != nid) - continue; - - do_register_memory_block_under_node(nid, mem_blk); - return 0; - } - /* mem section does not span the specified node */ - return 0; -} - /* * During hotplug we know that all pages in the memory block belong to the same * node. @@ -874,99 +825,77 @@ void unregister_memory_block_under_nodes(struct memory_block *mem_blk) kobject_name(&node_devices[mem_blk->nid]->dev.kobj)); } -void link_mem_sections(int nid, unsigned long start_pfn, unsigned long end_pfn, - enum meminit_context context) +/* register all memory blocks under the corresponding nodes */ +static void register_memory_blocks_under_nodes(void) { - walk_memory_blocks_func_t func; + struct memblock_region *r; - if (context == MEMINIT_HOTPLUG) - func = register_mem_block_under_node_hotplug; - else - func = register_mem_block_under_node_early; + for_each_mem_region(r) { + const unsigned long start_block_id = phys_to_block_id(r->base); + const unsigned long end_block_id = phys_to_block_id(r->base + r->size - 1); + const int nid = memblock_get_region_node(r); + unsigned long block_id; - walk_memory_blocks(PFN_PHYS(start_pfn), PFN_PHYS(end_pfn - start_pfn), - (void *)&nid, func); - return; -} - -#ifdef CONFIG_HUGETLBFS -/* - * Handle per node hstate attribute [un]registration on transistions - * to/from memoryless state. - */ -static void node_hugetlb_work(struct work_struct *work) -{ - struct node *node = container_of(work, struct node, node_work); + if (!node_online(nid)) + continue; - /* - * We only get here when a node transitions to/from memoryless state. - * We can detect which transition occurred by examining whether the - * node has memory now. hugetlb_register_node() already check this - * so we try to register the attributes. If that fails, then the - * node has transitioned to memoryless, try to unregister the - * attributes. - */ - if (!hugetlb_register_node(node)) - hugetlb_unregister_node(node); -} + for (block_id = start_block_id; block_id <= end_block_id; block_id++) { + struct memory_block *mem; -static void init_node_hugetlb_work(int nid) -{ - INIT_WORK(&node_devices[nid]->node_work, node_hugetlb_work); -} + mem = find_memory_block_by_id(block_id); + if (!mem) + continue; -static int node_memory_callback(struct notifier_block *self, - unsigned long action, void *arg) -{ - struct memory_notify *mnb = arg; - int nid = mnb->status_change_nid; - - switch (action) { - case MEM_ONLINE: - case MEM_OFFLINE: - /* - * offload per node hstate [un]registration to a work thread - * when transitioning to/from memoryless state. - */ - if (nid != NUMA_NO_NODE) - schedule_work(&node_devices[nid]->node_work); - break; + memory_block_add_nid_early(mem, nid); + do_register_memory_block_under_node(nid, mem); + put_device(&mem->dev); + } - case MEM_GOING_ONLINE: - case MEM_GOING_OFFLINE: - case MEM_CANCEL_ONLINE: - case MEM_CANCEL_OFFLINE: - default: - break; } - - return NOTIFY_OK; } -#endif /* CONFIG_HUGETLBFS */ -#endif /* CONFIG_MEMORY_HOTPLUG_SPARSE */ -#if !defined(CONFIG_MEMORY_HOTPLUG_SPARSE) || \ - !defined(CONFIG_HUGETLBFS) -static inline int node_memory_callback(struct notifier_block *self, - unsigned long action, void *arg) +void register_memory_blocks_under_node_hotplug(int nid, unsigned long start_pfn, + unsigned long end_pfn) { - return NOTIFY_OK; + walk_memory_blocks(PFN_PHYS(start_pfn), PFN_PHYS(end_pfn - start_pfn), + (void *)&nid, register_mem_block_under_node_hotplug); + return; } +#endif /* CONFIG_MEMORY_HOTPLUG */ -static void init_node_hugetlb_work(int nid) { } - -#endif - -int __register_one_node(int nid) +/** + * register_node - Initialize and register the node device. + * @nid: Node number to use when creating the device. + * + * Return: 0 on success, -errno otherwise + */ +int register_node(int nid) { int error; int cpu; + struct node *node; - node_devices[nid] = kzalloc(sizeof(struct node), GFP_KERNEL); - if (!node_devices[nid]) + node = kzalloc(sizeof(struct node), GFP_KERNEL); + if (!node) return -ENOMEM; - error = register_node(node_devices[nid], nid); + INIT_LIST_HEAD(&node->access_list); + + node->dev.id = nid; + node->dev.bus = &node_subsys; + node->dev.release = node_device_release; + node->dev.groups = node_dev_groups; + + error = device_register(&node->dev); + if (error) { + put_device(&node->dev); + return error; + } + + node_devices[nid] = node; + hugetlb_register_node(node); + compaction_register_node(node); + reclaim_register_node(node); /* link cpu under this node */ for_each_present_cpu(cpu) { @@ -974,20 +903,30 @@ int __register_one_node(int nid) register_cpu_under_node(cpu, nid); } - INIT_LIST_HEAD(&node_devices[nid]->access_list); - /* initialize work queue for memory hot plug */ - init_node_hugetlb_work(nid); node_init_caches(nid); return error; } - -void unregister_one_node(int nid) +/** + * unregister_node - unregister a node device + * @nid: nid of the node going away + * + * Unregisters the node device at node id @nid. All the devices on the + * node must be unregistered before calling this function. + */ +void unregister_node(int nid) { - if (!node_devices[nid]) + struct node *node = node_devices[nid]; + + if (!node) return; - unregister_node(node_devices[nid]); + hugetlb_unregister_node(node); + compaction_unregister_node(node); + reclaim_unregister_node(node); + node_remove_accesses(node); + node_remove_caches(node); + device_unregister(&node->dev); node_devices[nid] = NULL; } @@ -1047,27 +986,26 @@ static const struct attribute_group *cpu_root_attr_groups[] = { NULL, }; -#define NODE_CALLBACK_PRI 2 /* lower than SLAB */ -static int __init register_node_type(void) +void __init node_dev_init(void) { - int ret; + int ret, i; BUILD_BUG_ON(ARRAY_SIZE(node_state_attr) != NR_NODE_STATES); BUILD_BUG_ON(ARRAY_SIZE(node_state_attrs)-1 != NR_NODE_STATES); ret = subsys_system_register(&node_subsys, cpu_root_attr_groups); - if (!ret) { - static struct notifier_block node_memory_callback_nb = { - .notifier_call = node_memory_callback, - .priority = NODE_CALLBACK_PRI, - }; - register_hotmemory_notifier(&node_memory_callback_nb); - } + if (ret) + panic("%s() failed to register subsystem: %d\n", __func__, ret); /* - * Note: we're not going to unregister the node class if we fail - * to register the node state class attribute files. + * Create all node devices, which will properly link the node + * to already created cpu devices. */ - return ret; + for_each_online_node(i) { + ret = register_node(i); + if (ret) + panic("%s() failed to add node: %d\n", __func__, ret); + } + + register_memory_blocks_under_nodes(); } -postcore_initcall(register_node_type); diff --git a/drivers/base/physical_location.c b/drivers/base/physical_location.c new file mode 100644 index 000000000000..a5539e294d4d --- /dev/null +++ b/drivers/base/physical_location.c @@ -0,0 +1,145 @@ +// SPDX-License-Identifier: GPL-2.0 +/* + * Device physical location support + * + * Author: Won Chung <wonchung@google.com> + */ + +#include <linux/acpi.h> +#include <linux/sysfs.h> +#include <linux/string_choices.h> + +#include "physical_location.h" + +bool dev_add_physical_location(struct device *dev) +{ + struct acpi_pld_info *pld; + + if (!has_acpi_companion(dev)) + return false; + + if (!acpi_get_physical_device_location(ACPI_HANDLE(dev), &pld)) + return false; + + dev->physical_location = + kzalloc(sizeof(*dev->physical_location), GFP_KERNEL); + if (!dev->physical_location) { + ACPI_FREE(pld); + return false; + } + + dev->physical_location->panel = pld->panel; + dev->physical_location->vertical_position = pld->vertical_position; + dev->physical_location->horizontal_position = pld->horizontal_position; + dev->physical_location->dock = pld->dock; + dev->physical_location->lid = pld->lid; + + ACPI_FREE(pld); + return true; +} + +static ssize_t panel_show(struct device *dev, struct device_attribute *attr, + char *buf) +{ + const char *panel; + + switch (dev->physical_location->panel) { + case DEVICE_PANEL_TOP: + panel = "top"; + break; + case DEVICE_PANEL_BOTTOM: + panel = "bottom"; + break; + case DEVICE_PANEL_LEFT: + panel = "left"; + break; + case DEVICE_PANEL_RIGHT: + panel = "right"; + break; + case DEVICE_PANEL_FRONT: + panel = "front"; + break; + case DEVICE_PANEL_BACK: + panel = "back"; + break; + default: + panel = "unknown"; + } + return sysfs_emit(buf, "%s\n", panel); +} +static DEVICE_ATTR_RO(panel); + +static ssize_t vertical_position_show(struct device *dev, + struct device_attribute *attr, char *buf) +{ + const char *vertical_position; + + switch (dev->physical_location->vertical_position) { + case DEVICE_VERT_POS_UPPER: + vertical_position = "upper"; + break; + case DEVICE_VERT_POS_CENTER: + vertical_position = "center"; + break; + case DEVICE_VERT_POS_LOWER: + vertical_position = "lower"; + break; + default: + vertical_position = "unknown"; + } + return sysfs_emit(buf, "%s\n", vertical_position); +} +static DEVICE_ATTR_RO(vertical_position); + +static ssize_t horizontal_position_show(struct device *dev, + struct device_attribute *attr, char *buf) +{ + const char *horizontal_position; + + switch (dev->physical_location->horizontal_position) { + case DEVICE_HORI_POS_LEFT: + horizontal_position = "left"; + break; + case DEVICE_HORI_POS_CENTER: + horizontal_position = "center"; + break; + case DEVICE_HORI_POS_RIGHT: + horizontal_position = "right"; + break; + default: + horizontal_position = "unknown"; + } + return sysfs_emit(buf, "%s\n", horizontal_position); +} +static DEVICE_ATTR_RO(horizontal_position); + +static ssize_t dock_show(struct device *dev, struct device_attribute *attr, + char *buf) +{ + return sysfs_emit(buf, "%s\n", + str_yes_no(dev->physical_location->dock)); +} +static DEVICE_ATTR_RO(dock); + +static ssize_t lid_show(struct device *dev, struct device_attribute *attr, + char *buf) +{ + return sysfs_emit(buf, "%s\n", + str_yes_no(dev->physical_location->lid)); +} +static DEVICE_ATTR_RO(lid); + +static struct attribute *dev_attr_physical_location[] = { + &dev_attr_panel.attr, + &dev_attr_vertical_position.attr, + &dev_attr_horizontal_position.attr, + &dev_attr_dock.attr, + &dev_attr_lid.attr, + NULL, +}; + +const struct attribute_group dev_attr_physical_location_group = { + .name = "physical_location", + .attrs = dev_attr_physical_location, +}; + diff --git a/drivers/base/physical_location.h b/drivers/base/physical_location.h new file mode 100644 index 000000000000..3f3f61307998 --- /dev/null +++ b/drivers/base/physical_location.h @@ -0,0 +1,16 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +/* + * Device physical location support + * + * Author: Won Chung <wonchung@google.com> + */ + +#include <linux/device.h> + +#ifdef CONFIG_ACPI +bool dev_add_physical_location(struct device *dev); +extern const struct attribute_group dev_attr_physical_location_group; +#else +static inline bool dev_add_physical_location(struct device *dev) { return false; }; +static const struct attribute_group dev_attr_physical_location_group = {}; +#endif diff --git a/drivers/base/platform-msi.c b/drivers/base/platform-msi.c index 0b72b134a304..70db08f3ac6f 100644 --- a/drivers/base/platform-msi.c +++ b/drivers/base/platform-msi.c @@ -4,417 +4,97 @@ * * Copyright (C) 2015 ARM Limited, All Rights Reserved. * Author: Marc Zyngier <marc.zyngier@arm.com> + * Copyright (C) 2022 Linutronix GmbH */ #include <linux/device.h> -#include <linux/idr.h> -#include <linux/irq.h> #include <linux/irqdomain.h> #include <linux/msi.h> -#include <linux/slab.h> - -#define DEV_ID_SHIFT 21 -#define MAX_DEV_MSIS (1 << (32 - DEV_ID_SHIFT)) /* - * Internal data structure containing a (made up, but unique) devid - * and the callback to write the MSI message. + * This indirection can go when platform_device_msi_init_and_alloc_irqs() + * is switched to a proper irq_chip::irq_write_msi_msg() callback. Keep it + * simple for now. */ -struct platform_msi_priv_data { - struct device *dev; - void *host_data; - msi_alloc_info_t arg; - irq_write_msi_msg_t write_msg; - int devid; -}; - -/* The devid allocator */ -static DEFINE_IDA(platform_msi_devid_ida); - -#ifdef GENERIC_MSI_DOMAIN_OPS -/* - * Convert an msi_desc to a globaly unique identifier (per-device - * devid + msi_desc position in the msi_list). - */ -static irq_hw_number_t platform_msi_calc_hwirq(struct msi_desc *desc) +static void platform_msi_write_msi_msg(struct irq_data *d, struct msi_msg *msg) { - u32 devid; - - devid = desc->platform.msi_priv_data->devid; + irq_write_msi_msg_t cb = d->chip_data; - return (devid << (32 - DEV_ID_SHIFT)) | desc->platform.msi_index; + cb(irq_data_get_msi_desc(d), msg); } static void platform_msi_set_desc(msi_alloc_info_t *arg, struct msi_desc *desc) { arg->desc = desc; - arg->hwirq = platform_msi_calc_hwirq(desc); -} - -static int platform_msi_init(struct irq_domain *domain, - struct msi_domain_info *info, - unsigned int virq, irq_hw_number_t hwirq, - msi_alloc_info_t *arg) -{ - return irq_domain_set_hwirq_and_chip(domain, virq, hwirq, - info->chip, info->chip_data); -} - -static void platform_msi_set_proxy_dev(msi_alloc_info_t *arg) -{ - arg->flags |= MSI_ALLOC_FLAGS_PROXY_DEVICE; -} -#else -#define platform_msi_set_desc NULL -#define platform_msi_init NULL -#define platform_msi_set_proxy_dev(x) do {} while(0) -#endif - -static void platform_msi_update_dom_ops(struct msi_domain_info *info) -{ - struct msi_domain_ops *ops = info->ops; - - BUG_ON(!ops); - - if (ops->msi_init == NULL) - ops->msi_init = platform_msi_init; - if (ops->set_desc == NULL) - ops->set_desc = platform_msi_set_desc; -} - -static void platform_msi_write_msg(struct irq_data *data, struct msi_msg *msg) -{ - struct msi_desc *desc = irq_data_get_msi_desc(data); - struct platform_msi_priv_data *priv_data; - - priv_data = desc->platform.msi_priv_data; - - priv_data->write_msg(desc, msg); -} - -static void platform_msi_update_chip_ops(struct msi_domain_info *info) -{ - struct irq_chip *chip = info->chip; - - BUG_ON(!chip); - if (!chip->irq_mask) - chip->irq_mask = irq_chip_mask_parent; - if (!chip->irq_unmask) - chip->irq_unmask = irq_chip_unmask_parent; - if (!chip->irq_eoi) - chip->irq_eoi = irq_chip_eoi_parent; - if (!chip->irq_set_affinity) - chip->irq_set_affinity = msi_domain_set_affinity; - if (!chip->irq_write_msi_msg) - chip->irq_write_msi_msg = platform_msi_write_msg; - if (WARN_ON((info->flags & MSI_FLAG_LEVEL_CAPABLE) && - !(chip->flags & IRQCHIP_SUPPORTS_LEVEL_MSI))) - info->flags &= ~MSI_FLAG_LEVEL_CAPABLE; -} - -static void platform_msi_free_descs(struct device *dev, int base, int nvec) -{ - struct msi_desc *desc, *tmp; - - list_for_each_entry_safe(desc, tmp, dev_to_msi_list(dev), list) { - if (desc->platform.msi_index >= base && - desc->platform.msi_index < (base + nvec)) { - list_del(&desc->list); - free_msi_entry(desc); - } - } -} - -static int platform_msi_alloc_descs_with_irq(struct device *dev, int virq, - int nvec, - struct platform_msi_priv_data *data) - -{ - struct msi_desc *desc; - int i, base = 0; - - if (!list_empty(dev_to_msi_list(dev))) { - desc = list_last_entry(dev_to_msi_list(dev), - struct msi_desc, list); - base = desc->platform.msi_index + 1; - } - - for (i = 0; i < nvec; i++) { - desc = alloc_msi_entry(dev, 1, NULL); - if (!desc) - break; - - desc->platform.msi_priv_data = data; - desc->platform.msi_index = base + i; - desc->irq = virq ? virq + i : 0; - - list_add_tail(&desc->list, dev_to_msi_list(dev)); - } - - if (i != nvec) { - /* Clean up the mess */ - platform_msi_free_descs(dev, base, nvec); - - return -ENOMEM; - } - - return 0; -} - -static int platform_msi_alloc_descs(struct device *dev, int nvec, - struct platform_msi_priv_data *data) - -{ - return platform_msi_alloc_descs_with_irq(dev, 0, nvec, data); -} - -/** - * platform_msi_create_irq_domain - Create a platform MSI interrupt domain - * @fwnode: Optional fwnode of the interrupt controller - * @info: MSI domain info - * @parent: Parent irq domain - * - * Updates the domain and chip ops and creates a platform MSI - * interrupt domain. - * - * Returns: - * A domain pointer or NULL in case of failure. - */ -struct irq_domain *platform_msi_create_irq_domain(struct fwnode_handle *fwnode, - struct msi_domain_info *info, - struct irq_domain *parent) -{ - struct irq_domain *domain; - - if (info->flags & MSI_FLAG_USE_DEF_DOM_OPS) - platform_msi_update_dom_ops(info); - if (info->flags & MSI_FLAG_USE_DEF_CHIP_OPS) - platform_msi_update_chip_ops(info); - - domain = msi_create_irq_domain(fwnode, info, parent); - if (domain) - irq_domain_update_bus_token(domain, DOMAIN_BUS_PLATFORM_MSI); - - return domain; + arg->hwirq = desc->msi_index; } -static struct platform_msi_priv_data * -platform_msi_alloc_priv_data(struct device *dev, unsigned int nvec, - irq_write_msi_msg_t write_msi_msg) -{ - struct platform_msi_priv_data *datap; - /* - * Limit the number of interrupts to 2048 per device. Should we - * need to bump this up, DEV_ID_SHIFT should be adjusted - * accordingly (which would impact the max number of MSI - * capable devices). - */ - if (!dev->msi_domain || !write_msi_msg || !nvec || nvec > MAX_DEV_MSIS) - return ERR_PTR(-EINVAL); - - if (dev->msi_domain->bus_token != DOMAIN_BUS_PLATFORM_MSI) { - dev_err(dev, "Incompatible msi_domain, giving up\n"); - return ERR_PTR(-EINVAL); - } - - /* Already had a helping of MSI? Greed... */ - if (!list_empty(dev_to_msi_list(dev))) - return ERR_PTR(-EBUSY); - - datap = kzalloc(sizeof(*datap), GFP_KERNEL); - if (!datap) - return ERR_PTR(-ENOMEM); +static const struct msi_domain_template platform_msi_template = { + .chip = { + .name = "pMSI", + .irq_mask = irq_chip_mask_parent, + .irq_unmask = irq_chip_unmask_parent, + .irq_write_msi_msg = platform_msi_write_msi_msg, + /* The rest is filled in by the platform MSI parent */ + }, - datap->devid = ida_simple_get(&platform_msi_devid_ida, - 0, 1 << DEV_ID_SHIFT, GFP_KERNEL); - if (datap->devid < 0) { - int err = datap->devid; - kfree(datap); - return ERR_PTR(err); - } + .ops = { + .set_desc = platform_msi_set_desc, + }, - datap->write_msg = write_msi_msg; - datap->dev = dev; - - return datap; -} - -static void platform_msi_free_priv_data(struct platform_msi_priv_data *data) -{ - ida_simple_remove(&platform_msi_devid_ida, data->devid); - kfree(data); -} + .info = { + .bus_token = DOMAIN_BUS_DEVICE_MSI, + }, +}; /** - * platform_msi_domain_alloc_irqs - Allocate MSI interrupts for @dev + * platform_device_msi_init_and_alloc_irqs - Initialize platform device MSI + * and allocate interrupts for @dev * @dev: The device for which to allocate interrupts * @nvec: The number of interrupts to allocate * @write_msi_msg: Callback to write an interrupt message for @dev * * Returns: * Zero for success, or an error code in case of failure - */ -int platform_msi_domain_alloc_irqs(struct device *dev, unsigned int nvec, - irq_write_msi_msg_t write_msi_msg) -{ - struct platform_msi_priv_data *priv_data; - int err; - - priv_data = platform_msi_alloc_priv_data(dev, nvec, write_msi_msg); - if (IS_ERR(priv_data)) - return PTR_ERR(priv_data); - - err = platform_msi_alloc_descs(dev, nvec, priv_data); - if (err) - goto out_free_priv_data; - - err = msi_domain_alloc_irqs(dev->msi_domain, dev, nvec); - if (err) - goto out_free_desc; - - return 0; - -out_free_desc: - platform_msi_free_descs(dev, 0, nvec); -out_free_priv_data: - platform_msi_free_priv_data(priv_data); - - return err; -} -EXPORT_SYMBOL_GPL(platform_msi_domain_alloc_irqs); - -/** - * platform_msi_domain_free_irqs - Free MSI interrupts for @dev - * @dev: The device for which to free interrupts - */ -void platform_msi_domain_free_irqs(struct device *dev) -{ - if (!list_empty(dev_to_msi_list(dev))) { - struct msi_desc *desc; - - desc = first_msi_entry(dev); - platform_msi_free_priv_data(desc->platform.msi_priv_data); - } - - msi_domain_free_irqs(dev->msi_domain, dev); - platform_msi_free_descs(dev, 0, MAX_DEV_MSIS); -} -EXPORT_SYMBOL_GPL(platform_msi_domain_free_irqs); - -/** - * platform_msi_get_host_data - Query the private data associated with - * a platform-msi domain - * @domain: The platform-msi domain - * - * Returns the private data provided when calling - * platform_msi_create_device_domain. - */ -void *platform_msi_get_host_data(struct irq_domain *domain) -{ - struct platform_msi_priv_data *data = domain->host_data; - return data->host_data; -} - -/** - * __platform_msi_create_device_domain - Create a platform-msi domain * - * @dev: The device generating the MSIs - * @nvec: The number of MSIs that need to be allocated - * @is_tree: flag to indicate tree hierarchy - * @write_msi_msg: Callback to write an interrupt message for @dev - * @ops: The hierarchy domain operations to use - * @host_data: Private data associated to this domain + * This creates a MSI domain on @dev which has @dev->msi.domain as + * parent. The parent domain sets up the new domain. The domain has + * a fixed size of @nvec. The domain is managed by devres and will + * be removed when the device is removed. * - * Returns an irqdomain for @nvec interrupts + * Note: For migration purposes this falls back to the original platform_msi code + * up to the point where all platforms have been converted to the MSI + * parent model. */ -struct irq_domain * -__platform_msi_create_device_domain(struct device *dev, - unsigned int nvec, - bool is_tree, - irq_write_msi_msg_t write_msi_msg, - const struct irq_domain_ops *ops, - void *host_data) +int platform_device_msi_init_and_alloc_irqs(struct device *dev, unsigned int nvec, + irq_write_msi_msg_t write_msi_msg) { - struct platform_msi_priv_data *data; - struct irq_domain *domain; - int err; - - data = platform_msi_alloc_priv_data(dev, nvec, write_msi_msg); - if (IS_ERR(data)) - return NULL; - - data->host_data = host_data; - domain = irq_domain_create_hierarchy(dev->msi_domain, 0, - is_tree ? 0 : nvec, - dev->fwnode, ops, data); - if (!domain) - goto free_priv; - - platform_msi_set_proxy_dev(&data->arg); - err = msi_domain_prepare_irqs(domain->parent, dev, nvec, &data->arg); - if (err) - goto free_domain; + struct irq_domain *domain = dev->msi.domain; - return domain; + if (!domain || !write_msi_msg) + return -EINVAL; -free_domain: - irq_domain_remove(domain); -free_priv: - platform_msi_free_priv_data(data); - return NULL; -} - -/** - * platform_msi_domain_free - Free interrupts associated with a platform-msi - * domain - * - * @domain: The platform-msi domain - * @virq: The base irq from which to perform the free operation - * @nvec: How many interrupts to free from @virq - */ -void platform_msi_domain_free(struct irq_domain *domain, unsigned int virq, - unsigned int nvec) -{ - struct platform_msi_priv_data *data = domain->host_data; - struct msi_desc *desc, *tmp; - for_each_msi_entry_safe(desc, tmp, data->dev) { - if (WARN_ON(!desc->irq || desc->nvec_used != 1)) - return; - if (!(desc->irq >= virq && desc->irq < (virq + nvec))) - continue; + /* + * @write_msi_msg is stored in the resulting msi_domain_info::data. + * The underlying domain creation mechanism will assign that + * callback to the resulting irq chip. + */ + if (!msi_create_device_irq_domain(dev, MSI_DEFAULT_DOMAIN, + &platform_msi_template, + nvec, NULL, write_msi_msg)) + return -ENODEV; - irq_domain_free_irqs_common(domain, desc->irq, 1); - list_del(&desc->list); - free_msi_entry(desc); - } + return msi_domain_alloc_irqs_range(dev, MSI_DEFAULT_DOMAIN, 0, nvec - 1); } +EXPORT_SYMBOL_GPL(platform_device_msi_init_and_alloc_irqs); /** - * platform_msi_domain_alloc - Allocate interrupts associated with - * a platform-msi domain - * - * @domain: The platform-msi domain - * @virq: The base irq from which to perform the allocate operation - * @nr_irqs: How many interrupts to free from @virq - * - * Return 0 on success, or an error code on failure. Must be called - * with irq_domain_mutex held (which can only be done as part of a - * top-level interrupt allocation). + * platform_device_msi_free_irqs_all - Free all interrupts for @dev + * @dev: The device for which to free interrupts */ -int platform_msi_domain_alloc(struct irq_domain *domain, unsigned int virq, - unsigned int nr_irqs) +void platform_device_msi_free_irqs_all(struct device *dev) { - struct platform_msi_priv_data *data = domain->host_data; - int err; - - err = platform_msi_alloc_descs_with_irq(data->dev, virq, nr_irqs, data); - if (err) - return err; - - err = msi_domain_populate_irqs(domain->parent, data->dev, - virq, nr_irqs, &data->arg); - if (err) - platform_msi_domain_free(domain, virq, nr_irqs); - - return err; + msi_domain_free_irqs_all(dev, MSI_DEFAULT_DOMAIN); + msi_remove_device_irq_domain(dev, MSI_DEFAULT_DOMAIN); } +EXPORT_SYMBOL_GPL(platform_device_msi_free_irqs_all); diff --git a/drivers/base/platform.c b/drivers/base/platform.c index 8640578f45e9..b45d41b018ca 100644 --- a/drivers/base/platform.c +++ b/drivers/base/platform.c @@ -30,6 +30,8 @@ #include <linux/property.h> #include <linux/kmemleak.h> #include <linux/types.h> +#include <linux/iommu.h> +#include <linux/dma-map-ops.h> #include "base.h" #include "power/power.h" @@ -148,25 +150,37 @@ devm_platform_ioremap_resource_byname(struct platform_device *pdev, EXPORT_SYMBOL_GPL(devm_platform_ioremap_resource_byname); #endif /* CONFIG_HAS_IOMEM */ +static const struct cpumask *get_irq_affinity(struct platform_device *dev, + unsigned int num) +{ + const struct cpumask *mask = NULL; +#ifndef CONFIG_SPARC + struct fwnode_handle *fwnode = dev_fwnode(&dev->dev); + + if (is_of_node(fwnode)) + mask = of_irq_get_affinity(to_of_node(fwnode), num); + else if (is_acpi_device_node(fwnode)) + mask = acpi_irq_get_affinity(ACPI_HANDLE_FWNODE(fwnode), num); +#endif + + return mask ?: cpu_possible_mask; +} + /** - * platform_get_irq_optional - get an optional IRQ for a device - * @dev: platform device - * @num: IRQ number index + * platform_get_irq_affinity - get an optional IRQ and its affinity for a device + * @dev: platform device + * @num: interrupt number index + * @affinity: optional cpumask pointer to get the affinity of a per-cpu interrupt * - * Gets an IRQ for a platform device. Device drivers should check the return - * value for errors so as to not pass a negative integer value to the - * request_irq() APIs. This is the same as platform_get_irq(), except that it - * does not print an error message if an IRQ can not be obtained. - * - * For example:: + * Gets an interupt for a platform device. Device drivers should check the + * return value for errors so as to not pass a negative integer value to + * the request_irq() APIs. Optional affinity information is provided in the + * affinity pointer if available, and NULL otherwise. * - * int irq = platform_get_irq_optional(pdev, 0); - * if (irq < 0) - * return irq; - * - * Return: non-zero IRQ number on success, negative error number on failure. + * Return: non-zero interrupt number on success, negative error number on failure. */ -int platform_get_irq_optional(struct platform_device *dev, unsigned int num) +int platform_get_irq_affinity(struct platform_device *dev, unsigned int num, + const struct cpumask **affinity) { int ret; #ifdef CONFIG_SPARC @@ -176,18 +190,19 @@ int platform_get_irq_optional(struct platform_device *dev, unsigned int num) ret = dev->archdata.irqs[num]; goto out; #else + struct fwnode_handle *fwnode = dev_fwnode(&dev->dev); struct resource *r; - if (IS_ENABLED(CONFIG_OF_IRQ) && dev->dev.of_node) { - ret = of_irq_get(dev->dev.of_node, num); + if (is_of_node(fwnode)) { + ret = of_irq_get(to_of_node(fwnode), num); if (ret > 0 || ret == -EPROBE_DEFER) goto out; } r = platform_get_resource(dev, IORESOURCE_IRQ, num); - if (has_acpi_companion(&dev->dev)) { + if (is_acpi_device_node(fwnode)) { if (r && r->flags & IORESOURCE_DISABLED) { - ret = acpi_irq_get(ACPI_HANDLE(&dev->dev), num, r); + ret = acpi_irq_get(ACPI_HANDLE_FWNODE(fwnode), num, r); if (ret) goto out; } @@ -220,8 +235,8 @@ int platform_get_irq_optional(struct platform_device *dev, unsigned int num) * the device will only expose one IRQ, and this fallback * allows a common code path across either kind of resource. */ - if (num == 0 && has_acpi_companion(&dev->dev)) { - ret = acpi_dev_gpio_irq_get(ACPI_COMPANION(&dev->dev), num); + if (num == 0 && is_acpi_device_node(fwnode)) { + ret = acpi_dev_gpio_irq_get(to_acpi_device_node(fwnode), num); /* Our callers expect -ENXIO for missing IRQs. */ if (ret >= 0 || ret == -EPROBE_DEFER) goto out; @@ -231,9 +246,39 @@ int platform_get_irq_optional(struct platform_device *dev, unsigned int num) out_not_found: ret = -ENXIO; out: - WARN(ret == 0, "0 is an invalid IRQ number\n"); + if (WARN(!ret, "0 is an invalid IRQ number\n")) + return -EINVAL; + + if (ret > 0 && affinity) + *affinity = get_irq_affinity(dev, num); + return ret; } +EXPORT_SYMBOL_GPL(platform_get_irq_affinity); + +/** + * platform_get_irq_optional - get an optional interrupt for a device + * @dev: platform device + * @num: interrupt number index + * + * Gets an interrupt for a platform device. Device drivers should check the + * return value for errors so as to not pass a negative integer value to + * the request_irq() APIs. This is the same as platform_get_irq(), except + * that it does not print an error message if an interrupt can not be + * obtained. + * + * For example:: + * + * int irq = platform_get_irq_optional(pdev, 0); + * if (irq < 0) + * return irq; + * + * Return: non-zero interrupt number on success, negative error number on failure. + */ +int platform_get_irq_optional(struct platform_device *dev, unsigned int num) +{ + return platform_get_irq_affinity(dev, num, NULL); +} EXPORT_SYMBOL_GPL(platform_get_irq_optional); /** @@ -258,8 +303,9 @@ int platform_get_irq(struct platform_device *dev, unsigned int num) int ret; ret = platform_get_irq_optional(dev, num); - if (ret < 0 && ret != -EPROBE_DEFER) - dev_err(&dev->dev, "IRQ index %u not found\n", num); + if (ret < 0) + return dev_err_probe(&dev->dev, ret, + "IRQ index %u not found\n", num); return ret; } @@ -287,7 +333,7 @@ EXPORT_SYMBOL_GPL(platform_irq_count); struct irq_affinity_devres { unsigned int count; - unsigned int irq[]; + unsigned int irq[] __counted_by(count); }; static void platform_disable_acpi_irq(struct platform_device *pdev, int index) @@ -308,7 +354,7 @@ static void devm_platform_get_irqs_affinity_release(struct device *dev, for (i = 0; i < ptr->count; i++) { irq_dispose_mapping(ptr->irq[i]); - if (has_acpi_companion(dev)) + if (is_acpi_device_node(dev_fwnode(dev))) platform_disable_acpi_irq(to_platform_device(dev), i); } } @@ -437,15 +483,14 @@ static int __platform_get_irq_byname(struct platform_device *dev, struct resource *r; int ret; - if (IS_ENABLED(CONFIG_OF_IRQ) && dev->dev.of_node) { - ret = of_irq_get_byname(dev->dev.of_node, name); - if (ret > 0 || ret == -EPROBE_DEFER) - return ret; - } + ret = fwnode_irq_get_byname(dev_fwnode(&dev->dev), name); + if (ret > 0 || ret == -EPROBE_DEFER) + return ret; r = platform_get_resource_byname(dev, IORESOURCE_IRQ, name); if (r) { - WARN(r->start == 0, "0 is an invalid IRQ number\n"); + if (WARN(!r->start, "0 is an invalid IRQ number\n")) + return -EINVAL; return r->start; } @@ -466,9 +511,9 @@ int platform_get_irq_byname(struct platform_device *dev, const char *name) int ret; ret = __platform_get_irq_byname(dev, name); - if (ret < 0 && ret != -EPROBE_DEFER) - dev_err(&dev->dev, "IRQ %s not found\n", name); - + if (ret < 0) + return dev_err_probe(&dev->dev, ret, "IRQ %s not found\n", + name); return ret; } EXPORT_SYMBOL_GPL(platform_get_irq_byname); @@ -494,6 +539,8 @@ EXPORT_SYMBOL_GPL(platform_get_irq_byname_optional); * platform_add_devices - add a numbers of platform devices * @devs: array of platform devices to add * @num: number of platform devices in array + * + * Return: 0 on success, negative error number on failure. */ int platform_add_devices(struct platform_device **devs, int num) { @@ -602,7 +649,7 @@ int platform_device_add_resources(struct platform_device *pdev, struct resource *r = NULL; if (res) { - r = kmemdup(res, sizeof(struct resource) * num, GFP_KERNEL); + r = kmemdup_array(res, num, sizeof(*r), GFP_KERNEL); if (!r) return -ENOMEM; } @@ -642,22 +689,6 @@ int platform_device_add_data(struct platform_device *pdev, const void *data, EXPORT_SYMBOL_GPL(platform_device_add_data); /** - * platform_device_add_properties - add built-in properties to a platform device - * @pdev: platform device to add properties to - * @properties: null terminated array of properties to add - * - * The function will take deep copy of @properties and attach the copy to the - * platform device. The memory associated with properties will be freed when the - * platform device is released. - */ -int platform_device_add_properties(struct platform_device *pdev, - const struct property_entry *properties) -{ - return device_add_properties(&pdev->dev, properties); -} -EXPORT_SYMBOL_GPL(platform_device_add_properties); - -/** * platform_device_add - add a platform device to device hierarchy * @pdev: platform device we're adding * @@ -666,23 +697,21 @@ EXPORT_SYMBOL_GPL(platform_device_add_properties); */ int platform_device_add(struct platform_device *pdev) { + struct device *dev = &pdev->dev; u32 i; int ret; - if (!pdev) - return -EINVAL; + if (!dev->parent) + dev->parent = &platform_bus; - if (!pdev->dev.parent) - pdev->dev.parent = &platform_bus; - - pdev->dev.bus = &platform_bus_type; + dev->bus = &platform_bus_type; switch (pdev->id) { default: - dev_set_name(&pdev->dev, "%s.%d", pdev->name, pdev->id); + dev_set_name(dev, "%s.%d", pdev->name, pdev->id); break; case PLATFORM_DEVID_NONE: - dev_set_name(&pdev->dev, "%s", pdev->name); + dev_set_name(dev, "%s", pdev->name); break; case PLATFORM_DEVID_AUTO: /* @@ -692,10 +721,10 @@ int platform_device_add(struct platform_device *pdev) */ ret = ida_alloc(&platform_devid_ida, GFP_KERNEL); if (ret < 0) - goto err_out; + return ret; pdev->id = ret; pdev->id_auto = true; - dev_set_name(&pdev->dev, "%s.%d.auto", pdev->name, pdev->id); + dev_set_name(dev, "%s.%d.auto", pdev->name, pdev->id); break; } @@ -703,7 +732,7 @@ int platform_device_add(struct platform_device *pdev) struct resource *p, *r = &pdev->resource[i]; if (r->name == NULL) - r->name = dev_name(&pdev->dev); + r->name = dev_name(dev); p = r->parent; if (!p) { @@ -716,18 +745,20 @@ int platform_device_add(struct platform_device *pdev) if (p) { ret = insert_resource(p, r); if (ret) { - dev_err(&pdev->dev, "failed to claim resource %d: %pR\n", i, r); + dev_err(dev, "failed to claim resource %d: %pR\n", i, r); goto failed; } } } - pr_debug("Registering platform device '%s'. Parent at %s\n", - dev_name(&pdev->dev), dev_name(pdev->dev.parent)); + pr_debug("Registering platform device '%s'. Parent at %s\n", dev_name(dev), + dev_name(dev->parent)); - ret = device_add(&pdev->dev); - if (ret == 0) - return ret; + ret = device_add(dev); + if (ret) + goto failed; + + return 0; failed: if (pdev->id_auto) { @@ -741,7 +772,6 @@ int platform_device_add(struct platform_device *pdev) release_resource(r); } - err_out: return ret; } EXPORT_SYMBOL_GPL(platform_device_add); @@ -778,6 +808,10 @@ EXPORT_SYMBOL_GPL(platform_device_del); /** * platform_device_register - add a platform-level device * @pdev: platform device we're adding + * + * NOTE: _Never_ directly free @pdev after calling this function, even if it + * returned an error! Always use platform_device_put() to give up the + * reference initialised in this function instead. */ int platform_device_register(struct platform_device *pdev) { @@ -842,8 +876,8 @@ struct platform_device *platform_device_register_full( goto err; if (pdevinfo->properties) { - ret = platform_device_add_properties(pdev, - pdevinfo->properties); + ret = device_create_managed_software_node(&pdev->dev, + pdevinfo->properties, NULL); if (ret) goto err; } @@ -890,6 +924,13 @@ static int platform_probe_fail(struct platform_device *pdev) return -ENXIO; } +static int is_bound_to_driver(struct device *dev, void *driver) +{ + if (dev->driver == driver) + return 1; + return 0; +} + /** * __platform_driver_probe - register driver for non-hotpluggable device * @drv: platform driver structure @@ -913,7 +954,7 @@ static int platform_probe_fail(struct platform_device *pdev) int __init_or_module __platform_driver_probe(struct platform_driver *drv, int (*probe)(struct platform_device *), struct module *module) { - int retval, code; + int retval; if (drv->driver.probe_type == PROBE_PREFER_ASYNCHRONOUS) { pr_err("%s: drivers registered with %s can not be probed asynchronously\n", @@ -939,24 +980,21 @@ int __init_or_module __platform_driver_probe(struct platform_driver *drv, /* temporary section violation during probe() */ drv->probe = probe; - retval = code = __platform_driver_register(drv, module); + retval = __platform_driver_register(drv, module); if (retval) return retval; - /* - * Fixup that section violation, being paranoid about code scanning - * the list of drivers in order to probe new devices. Check to see - * if the probe was successful, and make sure any forced probes of - * new devices fail. - */ - spin_lock(&drv->driver.bus->p->klist_drivers.k_lock); + /* Force all new probes of this driver to fail */ drv->probe = platform_probe_fail; - if (code == 0 && list_empty(&drv->driver.p->klist_devices.k_list)) - retval = -ENODEV; - spin_unlock(&drv->driver.bus->p->klist_drivers.k_lock); - if (code != retval) + /* Walk all platform devices and see if any actually bound to this driver. + * If not, return an error as the device should have done so by now. + */ + if (!bus_for_each_dev(&platform_bus_type, NULL, &drv->driver, is_bound_to_driver)) { + retval = -ENODEV; platform_driver_unregister(drv); + } + return retval; } EXPORT_SYMBOL_GPL(__platform_driver_probe); @@ -985,7 +1023,7 @@ struct platform_device * __init_or_module __platform_create_bundle( struct platform_device *pdev; int error; - pdev = platform_device_alloc(driver->driver.name, -1); + pdev = platform_device_alloc(driver->driver.name, PLATFORM_DEVID_NONE); if (!pdev) { error = -ENOMEM; goto err_out; @@ -1125,7 +1163,7 @@ static int platform_legacy_resume(struct device *dev) int platform_pm_suspend(struct device *dev) { - struct device_driver *drv = dev->driver; + const struct device_driver *drv = dev->driver; int ret = 0; if (!drv) @@ -1143,7 +1181,7 @@ int platform_pm_suspend(struct device *dev) int platform_pm_resume(struct device *dev) { - struct device_driver *drv = dev->driver; + const struct device_driver *drv = dev->driver; int ret = 0; if (!drv) @@ -1165,7 +1203,7 @@ int platform_pm_resume(struct device *dev) int platform_pm_freeze(struct device *dev) { - struct device_driver *drv = dev->driver; + const struct device_driver *drv = dev->driver; int ret = 0; if (!drv) @@ -1183,7 +1221,7 @@ int platform_pm_freeze(struct device *dev) int platform_pm_thaw(struct device *dev) { - struct device_driver *drv = dev->driver; + const struct device_driver *drv = dev->driver; int ret = 0; if (!drv) @@ -1201,7 +1239,7 @@ int platform_pm_thaw(struct device *dev) int platform_pm_poweroff(struct device *dev) { - struct device_driver *drv = dev->driver; + const struct device_driver *drv = dev->driver; int ret = 0; if (!drv) @@ -1219,7 +1257,7 @@ int platform_pm_poweroff(struct device *dev) int platform_pm_restore(struct device *dev) { - struct device_driver *drv = dev->driver; + const struct device_driver *drv = dev->driver; int ret = 0; if (!drv) @@ -1286,31 +1324,11 @@ static ssize_t driver_override_store(struct device *dev, const char *buf, size_t count) { struct platform_device *pdev = to_platform_device(dev); - char *driver_override, *old, *cp; - - /* We need to keep extra room for a newline */ - if (count >= (PAGE_SIZE - 1)) - return -EINVAL; - - driver_override = kstrndup(buf, count, GFP_KERNEL); - if (!driver_override) - return -ENOMEM; - - cp = strchr(driver_override, '\n'); - if (cp) - *cp = '\0'; - - device_lock(dev); - old = pdev->driver_override; - if (strlen(driver_override)) { - pdev->driver_override = driver_override; - } else { - kfree(driver_override); - pdev->driver_override = NULL; - } - device_unlock(dev); + int ret; - kfree(old); + ret = driver_set_override(dev, &pdev->driver_override, buf, count); + if (ret) + return ret; return count; } @@ -1355,7 +1373,7 @@ __ATTRIBUTE_GROUPS(platform_dev); * and compare it against the name of the driver. Return whether they match * or not. */ -static int platform_match(struct device *dev, struct device_driver *drv) +static int platform_match(struct device *dev, const struct device_driver *drv) { struct platform_device *pdev = to_platform_device(dev); struct platform_driver *pdrv = to_platform_driver(drv); @@ -1380,9 +1398,9 @@ static int platform_match(struct device *dev, struct device_driver *drv) return (strcmp(pdev->name, drv->name) == 0); } -static int platform_uevent(struct device *dev, struct kobj_uevent_env *env) +static int platform_uevent(const struct device *dev, struct kobj_uevent_env *env) { - struct platform_device *pdev = to_platform_device(dev); + const struct platform_device *pdev = to_platform_device(dev); int rc; /* Some devices have extra OF data and an OF-style MODALIAS */ @@ -1419,15 +1437,13 @@ static int platform_probe(struct device *_dev) if (ret < 0) return ret; - ret = dev_pm_domain_attach(_dev, true); + ret = dev_pm_domain_attach(_dev, PD_FLAG_ATTACH_POWER_ON | + PD_FLAG_DETACH_POWER_OFF); if (ret) goto out; - if (drv->probe) { + if (drv->probe) ret = drv->probe(dev); - if (ret) - dev_pm_domain_detach(_dev, true); - } out: if (drv->prevent_deferred_probe && ret == -EPROBE_DEFER) { @@ -1438,20 +1454,13 @@ out: return ret; } -static int platform_remove(struct device *_dev) +static void platform_remove(struct device *_dev) { struct platform_driver *drv = to_platform_driver(_dev->driver); struct platform_device *dev = to_platform_device(_dev); - if (drv->remove) { - int ret = drv->remove(dev); - - if (ret) - dev_warn(_dev, "remove callback returned a non-zero value. This will be ignored.\n"); - } - dev_pm_domain_detach(_dev, true); - - return 0; + if (drv->remove) + drv->remove(dev); } static void platform_shutdown(struct device *_dev) @@ -1467,29 +1476,44 @@ static void platform_shutdown(struct device *_dev) drv->shutdown(dev); } - -int platform_dma_configure(struct device *dev) +static int platform_dma_configure(struct device *dev) { + struct device_driver *drv = READ_ONCE(dev->driver); + struct fwnode_handle *fwnode = dev_fwnode(dev); enum dev_dma_attr attr; int ret = 0; - if (dev->of_node) { - ret = of_dma_configure(dev, dev->of_node, true); - } else if (has_acpi_companion(dev)) { - attr = acpi_get_dma_attr(to_acpi_device_node(dev->fwnode)); + if (is_of_node(fwnode)) { + ret = of_dma_configure(dev, to_of_node(fwnode), true); + } else if (is_acpi_device_node(fwnode)) { + attr = acpi_get_dma_attr(to_acpi_device_node(fwnode)); ret = acpi_dma_configure(dev, attr); } + /* @dev->driver may not be valid when we're called from the IOMMU layer */ + if (ret || !drv || to_platform_driver(drv)->driver_managed_dma) + return ret; + + ret = iommu_device_use_default_domain(dev); + if (ret) + arch_teardown_dma_ops(dev); return ret; } +static void platform_dma_cleanup(struct device *dev) +{ + struct platform_driver *drv = to_platform_driver(dev->driver); + + if (!drv->driver_managed_dma) + iommu_device_unuse_default_domain(dev); +} + static const struct dev_pm_ops platform_dev_pm_ops = { - .runtime_suspend = pm_generic_runtime_suspend, - .runtime_resume = pm_generic_runtime_resume, + SET_RUNTIME_PM_OPS(pm_generic_runtime_suspend, pm_generic_runtime_resume, NULL) USE_PLATFORM_PM_SLEEP_OPS }; -struct bus_type platform_bus_type = { +const struct bus_type platform_bus_type = { .name = "platform", .dev_groups = platform_dev_groups, .match = platform_match, @@ -1498,6 +1522,7 @@ struct bus_type platform_bus_type = { .remove = platform_remove, .shutdown = platform_shutdown, .dma_configure = platform_dma_configure, + .dma_cleanup = platform_dma_cleanup, .pm = &platform_dev_pm_ops, }; EXPORT_SYMBOL_GPL(platform_bus_type); @@ -1537,6 +1562,6 @@ int __init platform_bus_init(void) error = bus_register(&platform_bus_type); if (error) device_unregister(&platform_bus); - of_platform_register_reconfig_notifier(); + return error; } diff --git a/drivers/base/power/Makefile b/drivers/base/power/Makefile index 8fdd0073eeeb..2989e42d0161 100644 --- a/drivers/base/power/Makefile +++ b/drivers/base/power/Makefile @@ -2,8 +2,8 @@ obj-$(CONFIG_PM) += sysfs.o generic_ops.o common.o qos.o runtime.o wakeirq.o obj-$(CONFIG_PM_SLEEP) += main.o wakeup.o wakeup_stats.o obj-$(CONFIG_PM_TRACE_RTC) += trace.o -obj-$(CONFIG_PM_GENERIC_DOMAINS) += domain.o domain_governor.o obj-$(CONFIG_HAVE_CLK) += clock_ops.o obj-$(CONFIG_PM_QOS_KUNIT_TEST) += qos-test.o +obj-$(CONFIG_PM_RUNTIME_KUNIT_TEST) += runtime-test.o ccflags-$(CONFIG_DEBUG_DRIVER) := -DDEBUG diff --git a/drivers/base/power/clock_ops.c b/drivers/base/power/clock_ops.c index 0251f3e6e61d..b69bcb37c830 100644 --- a/drivers/base/power/clock_ops.c +++ b/drivers/base/power/clock_ops.c @@ -259,39 +259,6 @@ int pm_clk_add_clk(struct device *dev, struct clk *clk) } EXPORT_SYMBOL_GPL(pm_clk_add_clk); - -/** - * of_pm_clk_add_clk - Start using a device clock for power management. - * @dev: Device whose clock is going to be used for power management. - * @name: Name of clock that is going to be used for power management. - * - * Add the clock described in the 'clocks' device-tree node that matches - * with the 'name' provided, to the list of clocks used for the power - * management of @dev. On success, returns 0. Returns a negative error - * code if the clock is not found or cannot be added. - */ -int of_pm_clk_add_clk(struct device *dev, const char *name) -{ - struct clk *clk; - int ret; - - if (!dev || !dev->of_node || !name) - return -EINVAL; - - clk = of_clk_get_by_name(dev->of_node, name); - if (IS_ERR(clk)) - return PTR_ERR(clk); - - ret = pm_clk_add_clk(dev, clk); - if (ret) { - clk_put(clk); - return ret; - } - - return 0; -} -EXPORT_SYMBOL_GPL(of_pm_clk_add_clk); - /** * of_pm_clk_add_clks - Start using device clock(s) for power management. * @dev: Device whose clock(s) is going to be used for power management. @@ -377,46 +344,6 @@ static void __pm_clk_remove(struct pm_clock_entry *ce) } /** - * pm_clk_remove - Stop using a device clock for power management. - * @dev: Device whose clock should not be used for PM any more. - * @con_id: Connection ID of the clock. - * - * Remove the clock represented by @con_id from the list of clocks used for - * the power management of @dev. - */ -void pm_clk_remove(struct device *dev, const char *con_id) -{ - struct pm_subsys_data *psd = dev_to_psd(dev); - struct pm_clock_entry *ce; - - if (!psd) - return; - - pm_clk_list_lock(psd); - - list_for_each_entry(ce, &psd->clock_list, node) { - if (!con_id && !ce->con_id) - goto remove; - else if (!con_id || !ce->con_id) - continue; - else if (!strcmp(con_id, ce->con_id)) - goto remove; - } - - pm_clk_list_unlock(psd); - return; - - remove: - list_del(&ce->node); - if (ce->enabled_when_prepared) - psd->clock_op_might_sleep--; - pm_clk_list_unlock(psd); - - __pm_clk_remove(ce); -} -EXPORT_SYMBOL_GPL(pm_clk_remove); - -/** * pm_clk_remove_clk - Stop using a device clock for power management. * @dev: Device whose clock should not be used for PM any more. * @clk: Clock pointer @@ -519,6 +446,23 @@ void pm_clk_destroy(struct device *dev) } EXPORT_SYMBOL_GPL(pm_clk_destroy); +static void pm_clk_destroy_action(void *data) +{ + pm_clk_destroy(data); +} + +int devm_pm_clk_create(struct device *dev) +{ + int ret; + + ret = pm_clk_create(dev); + if (ret) + return ret; + + return devm_add_action_or_reset(dev, pm_clk_destroy_action, dev); +} +EXPORT_SYMBOL_GPL(devm_pm_clk_create); + /** * pm_clk_suspend - Disable clocks in a device's PM clock list. * @dev: Device to disable the clocks for. @@ -776,7 +720,7 @@ static int pm_clk_notify(struct notifier_block *nb, * the remaining members of @clknb should be populated prior to calling this * routine. */ -void pm_clk_add_notifier(struct bus_type *bus, +void pm_clk_add_notifier(const struct bus_type *bus, struct pm_clk_notifier_block *clknb) { if (!bus || !clknb) diff --git a/drivers/base/power/common.c b/drivers/base/power/common.c index bbddb267c2e6..6ecf9ce4a4e6 100644 --- a/drivers/base/power/common.c +++ b/drivers/base/power/common.c @@ -11,6 +11,7 @@ #include <linux/pm_clock.h> #include <linux/acpi.h> #include <linux/pm_domain.h> +#include <linux/pm_opp.h> #include "power.h" @@ -82,7 +83,7 @@ EXPORT_SYMBOL_GPL(dev_pm_put_subsys_data); /** * dev_pm_domain_attach - Attach a device to its PM domain. * @dev: Device to attach. - * @power_on: Used to indicate whether we should power on the device. + * @flags: indicate whether we should power on/off the device on attach/detach * * The @dev may only be attached to a single PM domain. By iterating through * the available alternatives we try to find a valid PM domain for the device. @@ -99,17 +100,20 @@ EXPORT_SYMBOL_GPL(dev_pm_put_subsys_data); * Returns 0 on successfully attached PM domain, or when it is found that the * device doesn't need a PM domain, else a negative error code. */ -int dev_pm_domain_attach(struct device *dev, bool power_on) +int dev_pm_domain_attach(struct device *dev, u32 flags) { int ret; if (dev->pm_domain) return 0; - ret = acpi_dev_pm_attach(dev, power_on); + ret = acpi_dev_pm_attach(dev, !!(flags & PD_FLAG_ATTACH_POWER_ON)); if (!ret) ret = genpd_dev_pm_attach(dev); + if (dev->pm_domain) + dev->power.detach_power_off = !!(flags & PD_FLAG_DETACH_POWER_OFF); + return ret < 0 ? ret : 0; } EXPORT_SYMBOL_GPL(dev_pm_domain_attach); @@ -168,14 +172,187 @@ struct device *dev_pm_domain_attach_by_name(struct device *dev, EXPORT_SYMBOL_GPL(dev_pm_domain_attach_by_name); /** + * dev_pm_domain_attach_list - Associate a device with its PM domains. + * @dev: The device used to lookup the PM domains for. + * @data: The data used for attaching to the PM domains. + * @list: An out-parameter with an allocated list of attached PM domains. + * + * This function helps to attach a device to its multiple PM domains. The + * caller, which is typically a driver's probe function, may provide a list of + * names for the PM domains that we should try to attach the device to, but it + * may also provide an empty list, in case the attach should be done for all of + * the available PM domains. + * + * Callers must ensure proper synchronization of this function with power + * management callbacks. + * + * Returns the number of attached PM domains or a negative error code in case of + * a failure. Note that, to detach the list of PM domains, the driver shall call + * dev_pm_domain_detach_list(), typically during the remove phase. + */ +int dev_pm_domain_attach_list(struct device *dev, + const struct dev_pm_domain_attach_data *data, + struct dev_pm_domain_list **list) +{ + struct device_node *np = dev->of_node; + struct dev_pm_domain_list *pds; + struct device *pd_dev = NULL; + int ret, i, num_pds = 0; + bool by_id = true; + size_t size; + u32 pd_flags = data ? data->pd_flags : 0; + u32 link_flags = pd_flags & PD_FLAG_NO_DEV_LINK ? 0 : + DL_FLAG_STATELESS | DL_FLAG_PM_RUNTIME; + + if (dev->pm_domain) + return -EEXIST; + + /* For now this is limited to OF based platforms. */ + if (!np) + return 0; + + if (data && data->pd_names) { + num_pds = data->num_pd_names; + by_id = false; + } else { + num_pds = of_count_phandle_with_args(np, "power-domains", + "#power-domain-cells"); + } + + if (num_pds <= 0) + return 0; + + pds = kzalloc(sizeof(*pds), GFP_KERNEL); + if (!pds) + return -ENOMEM; + + size = sizeof(*pds->pd_devs) + sizeof(*pds->pd_links) + + sizeof(*pds->opp_tokens); + pds->pd_devs = kcalloc(num_pds, size, GFP_KERNEL); + if (!pds->pd_devs) { + ret = -ENOMEM; + goto free_pds; + } + pds->pd_links = (void *)(pds->pd_devs + num_pds); + pds->opp_tokens = (void *)(pds->pd_links + num_pds); + + if (link_flags && pd_flags & PD_FLAG_DEV_LINK_ON) + link_flags |= DL_FLAG_RPM_ACTIVE; + + for (i = 0; i < num_pds; i++) { + if (by_id) + pd_dev = dev_pm_domain_attach_by_id(dev, i); + else + pd_dev = dev_pm_domain_attach_by_name(dev, + data->pd_names[i]); + if (IS_ERR_OR_NULL(pd_dev)) { + ret = pd_dev ? PTR_ERR(pd_dev) : -ENODEV; + goto err_attach; + } + + if (pd_flags & PD_FLAG_REQUIRED_OPP) { + struct dev_pm_opp_config config = { + .required_dev = pd_dev, + .required_dev_index = i, + }; + + ret = dev_pm_opp_set_config(dev, &config); + if (ret < 0) + goto err_link; + + pds->opp_tokens[i] = ret; + } + + if (link_flags) { + struct device_link *link; + + link = device_link_add(dev, pd_dev, link_flags); + if (!link) { + ret = -ENODEV; + goto err_link; + } + + pds->pd_links[i] = link; + } + + pds->pd_devs[i] = pd_dev; + } + + pds->num_pds = num_pds; + *list = pds; + return num_pds; + +err_link: + dev_pm_opp_clear_config(pds->opp_tokens[i]); + dev_pm_domain_detach(pd_dev, true); +err_attach: + while (--i >= 0) { + dev_pm_opp_clear_config(pds->opp_tokens[i]); + if (pds->pd_links[i]) + device_link_del(pds->pd_links[i]); + dev_pm_domain_detach(pds->pd_devs[i], true); + } + kfree(pds->pd_devs); +free_pds: + kfree(pds); + return ret; +} +EXPORT_SYMBOL_GPL(dev_pm_domain_attach_list); + +/** + * devm_pm_domain_detach_list - devres-enabled version of dev_pm_domain_detach_list. + * @_list: The list of PM domains to detach. + * + * This function reverse the actions from devm_pm_domain_attach_list(). + * it will be invoked during the remove phase from drivers implicitly if driver + * uses devm_pm_domain_attach_list() to attach the PM domains. + */ +static void devm_pm_domain_detach_list(void *_list) +{ + struct dev_pm_domain_list *list = _list; + + dev_pm_domain_detach_list(list); +} + +/** + * devm_pm_domain_attach_list - devres-enabled version of dev_pm_domain_attach_list + * @dev: The device used to lookup the PM domains for. + * @data: The data used for attaching to the PM domains. + * @list: An out-parameter with an allocated list of attached PM domains. + * + * NOTE: this will also handle calling devm_pm_domain_detach_list() for + * you during remove phase. + * + * Returns the number of attached PM domains or a negative error code in case of + * a failure. + */ +int devm_pm_domain_attach_list(struct device *dev, + const struct dev_pm_domain_attach_data *data, + struct dev_pm_domain_list **list) +{ + int ret, num_pds; + + num_pds = dev_pm_domain_attach_list(dev, data, list); + if (num_pds <= 0) + return num_pds; + + ret = devm_add_action_or_reset(dev, devm_pm_domain_detach_list, *list); + if (ret) + return ret; + + return num_pds; +} +EXPORT_SYMBOL_GPL(devm_pm_domain_attach_list); + +/** * dev_pm_domain_detach - Detach a device from its PM domain. * @dev: Device to detach. * @power_off: Used to indicate whether we should power off the device. * - * This functions will reverse the actions from dev_pm_domain_attach() and - * dev_pm_domain_attach_by_id(), thus it detaches @dev from its PM domain. - * Typically it should be invoked during the remove phase, either from - * subsystem level code or from drivers. + * This functions will reverse the actions from dev_pm_domain_attach(), + * dev_pm_domain_attach_by_id() and dev_pm_domain_attach_by_name(), thus it + * detaches @dev from its PM domain. Typically it should be invoked during the + * remove phase, either from subsystem level code or from drivers. * * Callers must ensure proper synchronization of this function with power * management callbacks. @@ -188,6 +365,35 @@ void dev_pm_domain_detach(struct device *dev, bool power_off) EXPORT_SYMBOL_GPL(dev_pm_domain_detach); /** + * dev_pm_domain_detach_list - Detach a list of PM domains. + * @list: The list of PM domains to detach. + * + * This function reverse the actions from dev_pm_domain_attach_list(). + * Typically it should be invoked during the remove phase from drivers. + * + * Callers must ensure proper synchronization of this function with power + * management callbacks. + */ +void dev_pm_domain_detach_list(struct dev_pm_domain_list *list) +{ + int i; + + if (!list) + return; + + for (i = 0; i < list->num_pds; i++) { + dev_pm_opp_clear_config(list->opp_tokens[i]); + if (list->pd_links[i]) + device_link_del(list->pd_links[i]); + dev_pm_domain_detach(list->pd_devs[i], true); + } + + kfree(list->pd_devs); + kfree(list); +} +EXPORT_SYMBOL_GPL(dev_pm_domain_detach_list); + +/** * dev_pm_domain_start - Start the device through its PM domain. * @dev: Device to start. * @@ -228,3 +434,24 @@ void dev_pm_domain_set(struct device *dev, struct dev_pm_domain *pd) device_pm_check_callbacks(dev); } EXPORT_SYMBOL_GPL(dev_pm_domain_set); + +/** + * dev_pm_domain_set_performance_state - Request a new performance state. + * @dev: The device to make the request for. + * @state: Target performance state for the device. + * + * This function should be called when a new performance state needs to be + * requested for a device that is attached to a PM domain. Note that, the + * support for performance scaling for PM domains is optional. + * + * Returns 0 on success and when performance scaling isn't supported, negative + * error code on failure. + */ +int dev_pm_domain_set_performance_state(struct device *dev, unsigned int state) +{ + if (dev->pm_domain && dev->pm_domain->set_performance_state) + return dev->pm_domain->set_performance_state(dev, state); + + return 0; +} +EXPORT_SYMBOL_GPL(dev_pm_domain_set_performance_state); diff --git a/drivers/base/power/domain.c b/drivers/base/power/domain.c deleted file mode 100644 index a934c679e6ce..000000000000 --- a/drivers/base/power/domain.c +++ /dev/null @@ -1,3294 +0,0 @@ -// SPDX-License-Identifier: GPL-2.0 -/* - * drivers/base/power/domain.c - Common code related to device power domains. - * - * Copyright (C) 2011 Rafael J. Wysocki <rjw@sisk.pl>, Renesas Electronics Corp. - */ -#define pr_fmt(fmt) "PM: " fmt - -#include <linux/delay.h> -#include <linux/kernel.h> -#include <linux/io.h> -#include <linux/platform_device.h> -#include <linux/pm_opp.h> -#include <linux/pm_runtime.h> -#include <linux/pm_domain.h> -#include <linux/pm_qos.h> -#include <linux/pm_clock.h> -#include <linux/slab.h> -#include <linux/err.h> -#include <linux/sched.h> -#include <linux/suspend.h> -#include <linux/export.h> -#include <linux/cpu.h> -#include <linux/debugfs.h> - -#include "power.h" - -#define GENPD_RETRY_MAX_MS 250 /* Approximate */ - -#define GENPD_DEV_CALLBACK(genpd, type, callback, dev) \ -({ \ - type (*__routine)(struct device *__d); \ - type __ret = (type)0; \ - \ - __routine = genpd->dev_ops.callback; \ - if (__routine) { \ - __ret = __routine(dev); \ - } \ - __ret; \ -}) - -static LIST_HEAD(gpd_list); -static DEFINE_MUTEX(gpd_list_lock); - -struct genpd_lock_ops { - void (*lock)(struct generic_pm_domain *genpd); - void (*lock_nested)(struct generic_pm_domain *genpd, int depth); - int (*lock_interruptible)(struct generic_pm_domain *genpd); - void (*unlock)(struct generic_pm_domain *genpd); -}; - -static void genpd_lock_mtx(struct generic_pm_domain *genpd) -{ - mutex_lock(&genpd->mlock); -} - -static void genpd_lock_nested_mtx(struct generic_pm_domain *genpd, - int depth) -{ - mutex_lock_nested(&genpd->mlock, depth); -} - -static int genpd_lock_interruptible_mtx(struct generic_pm_domain *genpd) -{ - return mutex_lock_interruptible(&genpd->mlock); -} - -static void genpd_unlock_mtx(struct generic_pm_domain *genpd) -{ - return mutex_unlock(&genpd->mlock); -} - -static const struct genpd_lock_ops genpd_mtx_ops = { - .lock = genpd_lock_mtx, - .lock_nested = genpd_lock_nested_mtx, - .lock_interruptible = genpd_lock_interruptible_mtx, - .unlock = genpd_unlock_mtx, -}; - -static void genpd_lock_spin(struct generic_pm_domain *genpd) - __acquires(&genpd->slock) -{ - unsigned long flags; - - spin_lock_irqsave(&genpd->slock, flags); - genpd->lock_flags = flags; -} - -static void genpd_lock_nested_spin(struct generic_pm_domain *genpd, - int depth) - __acquires(&genpd->slock) -{ - unsigned long flags; - - spin_lock_irqsave_nested(&genpd->slock, flags, depth); - genpd->lock_flags = flags; -} - -static int genpd_lock_interruptible_spin(struct generic_pm_domain *genpd) - __acquires(&genpd->slock) -{ - unsigned long flags; - - spin_lock_irqsave(&genpd->slock, flags); - genpd->lock_flags = flags; - return 0; -} - -static void genpd_unlock_spin(struct generic_pm_domain *genpd) - __releases(&genpd->slock) -{ - spin_unlock_irqrestore(&genpd->slock, genpd->lock_flags); -} - -static const struct genpd_lock_ops genpd_spin_ops = { - .lock = genpd_lock_spin, - .lock_nested = genpd_lock_nested_spin, - .lock_interruptible = genpd_lock_interruptible_spin, - .unlock = genpd_unlock_spin, -}; - -#define genpd_lock(p) p->lock_ops->lock(p) -#define genpd_lock_nested(p, d) p->lock_ops->lock_nested(p, d) -#define genpd_lock_interruptible(p) p->lock_ops->lock_interruptible(p) -#define genpd_unlock(p) p->lock_ops->unlock(p) - -#define genpd_status_on(genpd) (genpd->status == GENPD_STATE_ON) -#define genpd_is_irq_safe(genpd) (genpd->flags & GENPD_FLAG_IRQ_SAFE) -#define genpd_is_always_on(genpd) (genpd->flags & GENPD_FLAG_ALWAYS_ON) -#define genpd_is_active_wakeup(genpd) (genpd->flags & GENPD_FLAG_ACTIVE_WAKEUP) -#define genpd_is_cpu_domain(genpd) (genpd->flags & GENPD_FLAG_CPU_DOMAIN) -#define genpd_is_rpm_always_on(genpd) (genpd->flags & GENPD_FLAG_RPM_ALWAYS_ON) - -static inline bool irq_safe_dev_in_no_sleep_domain(struct device *dev, - const struct generic_pm_domain *genpd) -{ - bool ret; - - ret = pm_runtime_is_irq_safe(dev) && !genpd_is_irq_safe(genpd); - - /* - * Warn once if an IRQ safe device is attached to a no sleep domain, as - * to indicate a suboptimal configuration for PM. For an always on - * domain this isn't case, thus don't warn. - */ - if (ret && !genpd_is_always_on(genpd)) - dev_warn_once(dev, "PM domain %s will not be powered off\n", - genpd->name); - - return ret; -} - -static int genpd_runtime_suspend(struct device *dev); - -/* - * Get the generic PM domain for a particular struct device. - * This validates the struct device pointer, the PM domain pointer, - * and checks that the PM domain pointer is a real generic PM domain. - * Any failure results in NULL being returned. - */ -static struct generic_pm_domain *dev_to_genpd_safe(struct device *dev) -{ - if (IS_ERR_OR_NULL(dev) || IS_ERR_OR_NULL(dev->pm_domain)) - return NULL; - - /* A genpd's always have its ->runtime_suspend() callback assigned. */ - if (dev->pm_domain->ops.runtime_suspend == genpd_runtime_suspend) - return pd_to_genpd(dev->pm_domain); - - return NULL; -} - -/* - * This should only be used where we are certain that the pm_domain - * attached to the device is a genpd domain. - */ -static struct generic_pm_domain *dev_to_genpd(struct device *dev) -{ - if (IS_ERR_OR_NULL(dev->pm_domain)) - return ERR_PTR(-EINVAL); - - return pd_to_genpd(dev->pm_domain); -} - -static int genpd_stop_dev(const struct generic_pm_domain *genpd, - struct device *dev) -{ - return GENPD_DEV_CALLBACK(genpd, int, stop, dev); -} - -static int genpd_start_dev(const struct generic_pm_domain *genpd, - struct device *dev) -{ - return GENPD_DEV_CALLBACK(genpd, int, start, dev); -} - -static bool genpd_sd_counter_dec(struct generic_pm_domain *genpd) -{ - bool ret = false; - - if (!WARN_ON(atomic_read(&genpd->sd_count) == 0)) - ret = !!atomic_dec_and_test(&genpd->sd_count); - - return ret; -} - -static void genpd_sd_counter_inc(struct generic_pm_domain *genpd) -{ - atomic_inc(&genpd->sd_count); - smp_mb__after_atomic(); -} - -#ifdef CONFIG_DEBUG_FS -static struct dentry *genpd_debugfs_dir; - -static void genpd_debug_add(struct generic_pm_domain *genpd); - -static void genpd_debug_remove(struct generic_pm_domain *genpd) -{ - struct dentry *d; - - d = debugfs_lookup(genpd->name, genpd_debugfs_dir); - debugfs_remove(d); -} - -static void genpd_update_accounting(struct generic_pm_domain *genpd) -{ - ktime_t delta, now; - - now = ktime_get(); - delta = ktime_sub(now, genpd->accounting_time); - - /* - * If genpd->status is active, it means we are just - * out of off and so update the idle time and vice - * versa. - */ - if (genpd->status == GENPD_STATE_ON) { - int state_idx = genpd->state_idx; - - genpd->states[state_idx].idle_time = - ktime_add(genpd->states[state_idx].idle_time, delta); - } else { - genpd->on_time = ktime_add(genpd->on_time, delta); - } - - genpd->accounting_time = now; -} -#else -static inline void genpd_debug_add(struct generic_pm_domain *genpd) {} -static inline void genpd_debug_remove(struct generic_pm_domain *genpd) {} -static inline void genpd_update_accounting(struct generic_pm_domain *genpd) {} -#endif - -static int _genpd_reeval_performance_state(struct generic_pm_domain *genpd, - unsigned int state) -{ - struct generic_pm_domain_data *pd_data; - struct pm_domain_data *pdd; - struct gpd_link *link; - - /* New requested state is same as Max requested state */ - if (state == genpd->performance_state) - return state; - - /* New requested state is higher than Max requested state */ - if (state > genpd->performance_state) - return state; - - /* Traverse all devices within the domain */ - list_for_each_entry(pdd, &genpd->dev_list, list_node) { - pd_data = to_gpd_data(pdd); - - if (pd_data->performance_state > state) - state = pd_data->performance_state; - } - - /* - * Traverse all sub-domains within the domain. This can be - * done without any additional locking as the link->performance_state - * field is protected by the parent genpd->lock, which is already taken. - * - * Also note that link->performance_state (subdomain's performance state - * requirement to parent domain) is different from - * link->child->performance_state (current performance state requirement - * of the devices/sub-domains of the subdomain) and so can have a - * different value. - * - * Note that we also take vote from powered-off sub-domains into account - * as the same is done for devices right now. - */ - list_for_each_entry(link, &genpd->parent_links, parent_node) { - if (link->performance_state > state) - state = link->performance_state; - } - - return state; -} - -static int genpd_xlate_performance_state(struct generic_pm_domain *genpd, - struct generic_pm_domain *parent, - unsigned int pstate) -{ - if (!parent->set_performance_state) - return pstate; - - return dev_pm_opp_xlate_performance_state(genpd->opp_table, - parent->opp_table, - pstate); -} - -static int _genpd_set_performance_state(struct generic_pm_domain *genpd, - unsigned int state, int depth) -{ - struct generic_pm_domain *parent; - struct gpd_link *link; - int parent_state, ret; - - if (state == genpd->performance_state) - return 0; - - /* Propagate to parents of genpd */ - list_for_each_entry(link, &genpd->child_links, child_node) { - parent = link->parent; - - /* Find parent's performance state */ - ret = genpd_xlate_performance_state(genpd, parent, state); - if (unlikely(ret < 0)) - goto err; - - parent_state = ret; - - genpd_lock_nested(parent, depth + 1); - - link->prev_performance_state = link->performance_state; - link->performance_state = parent_state; - parent_state = _genpd_reeval_performance_state(parent, - parent_state); - ret = _genpd_set_performance_state(parent, parent_state, depth + 1); - if (ret) - link->performance_state = link->prev_performance_state; - - genpd_unlock(parent); - - if (ret) - goto err; - } - - if (genpd->set_performance_state) { - ret = genpd->set_performance_state(genpd, state); - if (ret) - goto err; - } - - genpd->performance_state = state; - return 0; - -err: - /* Encountered an error, lets rollback */ - list_for_each_entry_continue_reverse(link, &genpd->child_links, - child_node) { - parent = link->parent; - - genpd_lock_nested(parent, depth + 1); - - parent_state = link->prev_performance_state; - link->performance_state = parent_state; - - parent_state = _genpd_reeval_performance_state(parent, - parent_state); - if (_genpd_set_performance_state(parent, parent_state, depth + 1)) { - pr_err("%s: Failed to roll back to %d performance state\n", - parent->name, parent_state); - } - - genpd_unlock(parent); - } - - return ret; -} - -static int genpd_set_performance_state(struct device *dev, unsigned int state) -{ - struct generic_pm_domain *genpd = dev_to_genpd(dev); - struct generic_pm_domain_data *gpd_data = dev_gpd_data(dev); - unsigned int prev_state; - int ret; - - prev_state = gpd_data->performance_state; - if (prev_state == state) - return 0; - - gpd_data->performance_state = state; - state = _genpd_reeval_performance_state(genpd, state); - - ret = _genpd_set_performance_state(genpd, state, 0); - if (ret) - gpd_data->performance_state = prev_state; - - return ret; -} - -static int genpd_drop_performance_state(struct device *dev) -{ - unsigned int prev_state = dev_gpd_data(dev)->performance_state; - - if (!genpd_set_performance_state(dev, 0)) - return prev_state; - - return 0; -} - -static void genpd_restore_performance_state(struct device *dev, - unsigned int state) -{ - if (state) - genpd_set_performance_state(dev, state); -} - -/** - * dev_pm_genpd_set_performance_state- Set performance state of device's power - * domain. - * - * @dev: Device for which the performance-state needs to be set. - * @state: Target performance state of the device. This can be set as 0 when the - * device doesn't have any performance state constraints left (And so - * the device wouldn't participate anymore to find the target - * performance state of the genpd). - * - * It is assumed that the users guarantee that the genpd wouldn't be detached - * while this routine is getting called. - * - * Returns 0 on success and negative error values on failures. - */ -int dev_pm_genpd_set_performance_state(struct device *dev, unsigned int state) -{ - struct generic_pm_domain *genpd; - int ret; - - genpd = dev_to_genpd_safe(dev); - if (!genpd) - return -ENODEV; - - if (WARN_ON(!dev->power.subsys_data || - !dev->power.subsys_data->domain_data)) - return -EINVAL; - - genpd_lock(genpd); - ret = genpd_set_performance_state(dev, state); - genpd_unlock(genpd); - - return ret; -} -EXPORT_SYMBOL_GPL(dev_pm_genpd_set_performance_state); - -/** - * dev_pm_genpd_set_next_wakeup - Notify PM framework of an impending wakeup. - * - * @dev: Device to handle - * @next: impending interrupt/wakeup for the device - * - * - * Allow devices to inform of the next wakeup. It's assumed that the users - * guarantee that the genpd wouldn't be detached while this routine is getting - * called. Additionally, it's also assumed that @dev isn't runtime suspended - * (RPM_SUSPENDED)." - * Although devices are expected to update the next_wakeup after the end of - * their usecase as well, it is possible the devices themselves may not know - * about that, so stale @next will be ignored when powering off the domain. - */ -void dev_pm_genpd_set_next_wakeup(struct device *dev, ktime_t next) -{ - struct generic_pm_domain_data *gpd_data; - struct generic_pm_domain *genpd; - - genpd = dev_to_genpd_safe(dev); - if (!genpd) - return; - - gpd_data = to_gpd_data(dev->power.subsys_data->domain_data); - gpd_data->next_wakeup = next; -} -EXPORT_SYMBOL_GPL(dev_pm_genpd_set_next_wakeup); - -static int _genpd_power_on(struct generic_pm_domain *genpd, bool timed) -{ - unsigned int state_idx = genpd->state_idx; - ktime_t time_start; - s64 elapsed_ns; - int ret; - - /* Notify consumers that we are about to power on. */ - ret = raw_notifier_call_chain_robust(&genpd->power_notifiers, - GENPD_NOTIFY_PRE_ON, - GENPD_NOTIFY_OFF, NULL); - ret = notifier_to_errno(ret); - if (ret) - return ret; - - if (!genpd->power_on) - goto out; - - if (!timed) { - ret = genpd->power_on(genpd); - if (ret) - goto err; - - goto out; - } - - time_start = ktime_get(); - ret = genpd->power_on(genpd); - if (ret) - goto err; - - elapsed_ns = ktime_to_ns(ktime_sub(ktime_get(), time_start)); - if (elapsed_ns <= genpd->states[state_idx].power_on_latency_ns) - goto out; - - genpd->states[state_idx].power_on_latency_ns = elapsed_ns; - genpd->max_off_time_changed = true; - pr_debug("%s: Power-%s latency exceeded, new value %lld ns\n", - genpd->name, "on", elapsed_ns); - -out: - raw_notifier_call_chain(&genpd->power_notifiers, GENPD_NOTIFY_ON, NULL); - return 0; -err: - raw_notifier_call_chain(&genpd->power_notifiers, GENPD_NOTIFY_OFF, - NULL); - return ret; -} - -static int _genpd_power_off(struct generic_pm_domain *genpd, bool timed) -{ - unsigned int state_idx = genpd->state_idx; - ktime_t time_start; - s64 elapsed_ns; - int ret; - - /* Notify consumers that we are about to power off. */ - ret = raw_notifier_call_chain_robust(&genpd->power_notifiers, - GENPD_NOTIFY_PRE_OFF, - GENPD_NOTIFY_ON, NULL); - ret = notifier_to_errno(ret); - if (ret) - return ret; - - if (!genpd->power_off) - goto out; - - if (!timed) { - ret = genpd->power_off(genpd); - if (ret) - goto busy; - - goto out; - } - - time_start = ktime_get(); - ret = genpd->power_off(genpd); - if (ret) - goto busy; - - elapsed_ns = ktime_to_ns(ktime_sub(ktime_get(), time_start)); - if (elapsed_ns <= genpd->states[state_idx].power_off_latency_ns) - goto out; - - genpd->states[state_idx].power_off_latency_ns = elapsed_ns; - genpd->max_off_time_changed = true; - pr_debug("%s: Power-%s latency exceeded, new value %lld ns\n", - genpd->name, "off", elapsed_ns); - -out: - raw_notifier_call_chain(&genpd->power_notifiers, GENPD_NOTIFY_OFF, - NULL); - return 0; -busy: - raw_notifier_call_chain(&genpd->power_notifiers, GENPD_NOTIFY_ON, NULL); - return ret; -} - -/** - * genpd_queue_power_off_work - Queue up the execution of genpd_power_off(). - * @genpd: PM domain to power off. - * - * Queue up the execution of genpd_power_off() unless it's already been done - * before. - */ -static void genpd_queue_power_off_work(struct generic_pm_domain *genpd) -{ - queue_work(pm_wq, &genpd->power_off_work); -} - -/** - * genpd_power_off - Remove power from a given PM domain. - * @genpd: PM domain to power down. - * @one_dev_on: If invoked from genpd's ->runtime_suspend|resume() callback, the - * RPM status of the releated device is in an intermediate state, not yet turned - * into RPM_SUSPENDED. This means genpd_power_off() must allow one device to not - * be RPM_SUSPENDED, while it tries to power off the PM domain. - * @depth: nesting count for lockdep. - * - * If all of the @genpd's devices have been suspended and all of its subdomains - * have been powered down, remove power from @genpd. - */ -static int genpd_power_off(struct generic_pm_domain *genpd, bool one_dev_on, - unsigned int depth) -{ - struct pm_domain_data *pdd; - struct gpd_link *link; - unsigned int not_suspended = 0; - int ret; - - /* - * Do not try to power off the domain in the following situations: - * (1) The domain is already in the "power off" state. - * (2) System suspend is in progress. - */ - if (!genpd_status_on(genpd) || genpd->prepared_count > 0) - return 0; - - /* - * Abort power off for the PM domain in the following situations: - * (1) The domain is configured as always on. - * (2) When the domain has a subdomain being powered on. - */ - if (genpd_is_always_on(genpd) || - genpd_is_rpm_always_on(genpd) || - atomic_read(&genpd->sd_count) > 0) - return -EBUSY; - - list_for_each_entry(pdd, &genpd->dev_list, list_node) { - enum pm_qos_flags_status stat; - - stat = dev_pm_qos_flags(pdd->dev, PM_QOS_FLAG_NO_POWER_OFF); - if (stat > PM_QOS_FLAGS_NONE) - return -EBUSY; - - /* - * Do not allow PM domain to be powered off, when an IRQ safe - * device is part of a non-IRQ safe domain. - */ - if (!pm_runtime_suspended(pdd->dev) || - irq_safe_dev_in_no_sleep_domain(pdd->dev, genpd)) - not_suspended++; - } - - if (not_suspended > 1 || (not_suspended == 1 && !one_dev_on)) - return -EBUSY; - - if (genpd->gov && genpd->gov->power_down_ok) { - if (!genpd->gov->power_down_ok(&genpd->domain)) - return -EAGAIN; - } - - /* Default to shallowest state. */ - if (!genpd->gov) - genpd->state_idx = 0; - - /* Don't power off, if a child domain is waiting to power on. */ - if (atomic_read(&genpd->sd_count) > 0) - return -EBUSY; - - ret = _genpd_power_off(genpd, true); - if (ret) { - genpd->states[genpd->state_idx].rejected++; - return ret; - } - - genpd->status = GENPD_STATE_OFF; - genpd_update_accounting(genpd); - genpd->states[genpd->state_idx].usage++; - - list_for_each_entry(link, &genpd->child_links, child_node) { - genpd_sd_counter_dec(link->parent); - genpd_lock_nested(link->parent, depth + 1); - genpd_power_off(link->parent, false, depth + 1); - genpd_unlock(link->parent); - } - - return 0; -} - -/** - * genpd_power_on - Restore power to a given PM domain and its parents. - * @genpd: PM domain to power up. - * @depth: nesting count for lockdep. - * - * Restore power to @genpd and all of its parents so that it is possible to - * resume a device belonging to it. - */ -static int genpd_power_on(struct generic_pm_domain *genpd, unsigned int depth) -{ - struct gpd_link *link; - int ret = 0; - - if (genpd_status_on(genpd)) - return 0; - - /* - * The list is guaranteed not to change while the loop below is being - * executed, unless one of the parents' .power_on() callbacks fiddles - * with it. - */ - list_for_each_entry(link, &genpd->child_links, child_node) { - struct generic_pm_domain *parent = link->parent; - - genpd_sd_counter_inc(parent); - - genpd_lock_nested(parent, depth + 1); - ret = genpd_power_on(parent, depth + 1); - genpd_unlock(parent); - - if (ret) { - genpd_sd_counter_dec(parent); - goto err; - } - } - - ret = _genpd_power_on(genpd, true); - if (ret) - goto err; - - genpd->status = GENPD_STATE_ON; - genpd_update_accounting(genpd); - - return 0; - - err: - list_for_each_entry_continue_reverse(link, - &genpd->child_links, - child_node) { - genpd_sd_counter_dec(link->parent); - genpd_lock_nested(link->parent, depth + 1); - genpd_power_off(link->parent, false, depth + 1); - genpd_unlock(link->parent); - } - - return ret; -} - -static int genpd_dev_pm_start(struct device *dev) -{ - struct generic_pm_domain *genpd = dev_to_genpd(dev); - - return genpd_start_dev(genpd, dev); -} - -static int genpd_dev_pm_qos_notifier(struct notifier_block *nb, - unsigned long val, void *ptr) -{ - struct generic_pm_domain_data *gpd_data; - struct device *dev; - - gpd_data = container_of(nb, struct generic_pm_domain_data, nb); - dev = gpd_data->base.dev; - - for (;;) { - struct generic_pm_domain *genpd; - struct pm_domain_data *pdd; - - spin_lock_irq(&dev->power.lock); - - pdd = dev->power.subsys_data ? - dev->power.subsys_data->domain_data : NULL; - if (pdd) { - to_gpd_data(pdd)->td.constraint_changed = true; - genpd = dev_to_genpd(dev); - } else { - genpd = ERR_PTR(-ENODATA); - } - - spin_unlock_irq(&dev->power.lock); - - if (!IS_ERR(genpd)) { - genpd_lock(genpd); - genpd->max_off_time_changed = true; - genpd_unlock(genpd); - } - - dev = dev->parent; - if (!dev || dev->power.ignore_children) - break; - } - - return NOTIFY_DONE; -} - -/** - * genpd_power_off_work_fn - Power off PM domain whose subdomain count is 0. - * @work: Work structure used for scheduling the execution of this function. - */ -static void genpd_power_off_work_fn(struct work_struct *work) -{ - struct generic_pm_domain *genpd; - - genpd = container_of(work, struct generic_pm_domain, power_off_work); - - genpd_lock(genpd); - genpd_power_off(genpd, false, 0); - genpd_unlock(genpd); -} - -/** - * __genpd_runtime_suspend - walk the hierarchy of ->runtime_suspend() callbacks - * @dev: Device to handle. - */ -static int __genpd_runtime_suspend(struct device *dev) -{ - int (*cb)(struct device *__dev); - - if (dev->type && dev->type->pm) - cb = dev->type->pm->runtime_suspend; - else if (dev->class && dev->class->pm) - cb = dev->class->pm->runtime_suspend; - else if (dev->bus && dev->bus->pm) - cb = dev->bus->pm->runtime_suspend; - else - cb = NULL; - - if (!cb && dev->driver && dev->driver->pm) - cb = dev->driver->pm->runtime_suspend; - - return cb ? cb(dev) : 0; -} - -/** - * __genpd_runtime_resume - walk the hierarchy of ->runtime_resume() callbacks - * @dev: Device to handle. - */ -static int __genpd_runtime_resume(struct device *dev) -{ - int (*cb)(struct device *__dev); - - if (dev->type && dev->type->pm) - cb = dev->type->pm->runtime_resume; - else if (dev->class && dev->class->pm) - cb = dev->class->pm->runtime_resume; - else if (dev->bus && dev->bus->pm) - cb = dev->bus->pm->runtime_resume; - else - cb = NULL; - - if (!cb && dev->driver && dev->driver->pm) - cb = dev->driver->pm->runtime_resume; - - return cb ? cb(dev) : 0; -} - -/** - * genpd_runtime_suspend - Suspend a device belonging to I/O PM domain. - * @dev: Device to suspend. - * - * Carry out a runtime suspend of a device under the assumption that its - * pm_domain field points to the domain member of an object of type - * struct generic_pm_domain representing a PM domain consisting of I/O devices. - */ -static int genpd_runtime_suspend(struct device *dev) -{ - struct generic_pm_domain *genpd; - bool (*suspend_ok)(struct device *__dev); - struct generic_pm_domain_data *gpd_data = dev_gpd_data(dev); - struct gpd_timing_data *td = &gpd_data->td; - bool runtime_pm = pm_runtime_enabled(dev); - ktime_t time_start; - s64 elapsed_ns; - int ret; - - dev_dbg(dev, "%s()\n", __func__); - - genpd = dev_to_genpd(dev); - if (IS_ERR(genpd)) - return -EINVAL; - - /* - * A runtime PM centric subsystem/driver may re-use the runtime PM - * callbacks for other purposes than runtime PM. In those scenarios - * runtime PM is disabled. Under these circumstances, we shall skip - * validating/measuring the PM QoS latency. - */ - suspend_ok = genpd->gov ? genpd->gov->suspend_ok : NULL; - if (runtime_pm && suspend_ok && !suspend_ok(dev)) - return -EBUSY; - - /* Measure suspend latency. */ - time_start = 0; - if (runtime_pm) - time_start = ktime_get(); - - ret = __genpd_runtime_suspend(dev); - if (ret) - return ret; - - ret = genpd_stop_dev(genpd, dev); - if (ret) { - __genpd_runtime_resume(dev); - return ret; - } - - /* Update suspend latency value if the measured time exceeds it. */ - if (runtime_pm) { - elapsed_ns = ktime_to_ns(ktime_sub(ktime_get(), time_start)); - if (elapsed_ns > td->suspend_latency_ns) { - td->suspend_latency_ns = elapsed_ns; - dev_dbg(dev, "suspend latency exceeded, %lld ns\n", - elapsed_ns); - genpd->max_off_time_changed = true; - td->constraint_changed = true; - } - } - - /* - * If power.irq_safe is set, this routine may be run with - * IRQs disabled, so suspend only if the PM domain also is irq_safe. - */ - if (irq_safe_dev_in_no_sleep_domain(dev, genpd)) - return 0; - - genpd_lock(genpd); - gpd_data->rpm_pstate = genpd_drop_performance_state(dev); - genpd_power_off(genpd, true, 0); - genpd_unlock(genpd); - - return 0; -} - -/** - * genpd_runtime_resume - Resume a device belonging to I/O PM domain. - * @dev: Device to resume. - * - * Carry out a runtime resume of a device under the assumption that its - * pm_domain field points to the domain member of an object of type - * struct generic_pm_domain representing a PM domain consisting of I/O devices. - */ -static int genpd_runtime_resume(struct device *dev) -{ - struct generic_pm_domain *genpd; - struct generic_pm_domain_data *gpd_data = dev_gpd_data(dev); - struct gpd_timing_data *td = &gpd_data->td; - bool runtime_pm = pm_runtime_enabled(dev); - ktime_t time_start; - s64 elapsed_ns; - int ret; - bool timed = true; - - dev_dbg(dev, "%s()\n", __func__); - - genpd = dev_to_genpd(dev); - if (IS_ERR(genpd)) - return -EINVAL; - - /* - * As we don't power off a non IRQ safe domain, which holds - * an IRQ safe device, we don't need to restore power to it. - */ - if (irq_safe_dev_in_no_sleep_domain(dev, genpd)) { - timed = false; - goto out; - } - - genpd_lock(genpd); - ret = genpd_power_on(genpd, 0); - if (!ret) - genpd_restore_performance_state(dev, gpd_data->rpm_pstate); - genpd_unlock(genpd); - - if (ret) - return ret; - - out: - /* Measure resume latency. */ - time_start = 0; - if (timed && runtime_pm) - time_start = ktime_get(); - - ret = genpd_start_dev(genpd, dev); - if (ret) - goto err_poweroff; - - ret = __genpd_runtime_resume(dev); - if (ret) - goto err_stop; - - /* Update resume latency value if the measured time exceeds it. */ - if (timed && runtime_pm) { - elapsed_ns = ktime_to_ns(ktime_sub(ktime_get(), time_start)); - if (elapsed_ns > td->resume_latency_ns) { - td->resume_latency_ns = elapsed_ns; - dev_dbg(dev, "resume latency exceeded, %lld ns\n", - elapsed_ns); - genpd->max_off_time_changed = true; - td->constraint_changed = true; - } - } - - return 0; - -err_stop: - genpd_stop_dev(genpd, dev); -err_poweroff: - if (!pm_runtime_is_irq_safe(dev) || genpd_is_irq_safe(genpd)) { - genpd_lock(genpd); - gpd_data->rpm_pstate = genpd_drop_performance_state(dev); - genpd_power_off(genpd, true, 0); - genpd_unlock(genpd); - } - - return ret; -} - -static bool pd_ignore_unused; -static int __init pd_ignore_unused_setup(char *__unused) -{ - pd_ignore_unused = true; - return 1; -} -__setup("pd_ignore_unused", pd_ignore_unused_setup); - -/** - * genpd_power_off_unused - Power off all PM domains with no devices in use. - */ -static int __init genpd_power_off_unused(void) -{ - struct generic_pm_domain *genpd; - - if (pd_ignore_unused) { - pr_warn("genpd: Not disabling unused power domains\n"); - return 0; - } - - mutex_lock(&gpd_list_lock); - - list_for_each_entry(genpd, &gpd_list, gpd_list_node) - genpd_queue_power_off_work(genpd); - - mutex_unlock(&gpd_list_lock); - - return 0; -} -late_initcall(genpd_power_off_unused); - -#ifdef CONFIG_PM_SLEEP - -/** - * genpd_sync_power_off - Synchronously power off a PM domain and its parents. - * @genpd: PM domain to power off, if possible. - * @use_lock: use the lock. - * @depth: nesting count for lockdep. - * - * Check if the given PM domain can be powered off (during system suspend or - * hibernation) and do that if so. Also, in that case propagate to its parents. - * - * This function is only called in "noirq" and "syscore" stages of system power - * transitions. The "noirq" callbacks may be executed asynchronously, thus in - * these cases the lock must be held. - */ -static void genpd_sync_power_off(struct generic_pm_domain *genpd, bool use_lock, - unsigned int depth) -{ - struct gpd_link *link; - - if (!genpd_status_on(genpd) || genpd_is_always_on(genpd)) - return; - - if (genpd->suspended_count != genpd->device_count - || atomic_read(&genpd->sd_count) > 0) - return; - - /* Choose the deepest state when suspending */ - genpd->state_idx = genpd->state_count - 1; - if (_genpd_power_off(genpd, false)) - return; - - genpd->status = GENPD_STATE_OFF; - - list_for_each_entry(link, &genpd->child_links, child_node) { - genpd_sd_counter_dec(link->parent); - - if (use_lock) - genpd_lock_nested(link->parent, depth + 1); - - genpd_sync_power_off(link->parent, use_lock, depth + 1); - - if (use_lock) - genpd_unlock(link->parent); - } -} - -/** - * genpd_sync_power_on - Synchronously power on a PM domain and its parents. - * @genpd: PM domain to power on. - * @use_lock: use the lock. - * @depth: nesting count for lockdep. - * - * This function is only called in "noirq" and "syscore" stages of system power - * transitions. The "noirq" callbacks may be executed asynchronously, thus in - * these cases the lock must be held. - */ -static void genpd_sync_power_on(struct generic_pm_domain *genpd, bool use_lock, - unsigned int depth) -{ - struct gpd_link *link; - - if (genpd_status_on(genpd)) - return; - - list_for_each_entry(link, &genpd->child_links, child_node) { - genpd_sd_counter_inc(link->parent); - - if (use_lock) - genpd_lock_nested(link->parent, depth + 1); - - genpd_sync_power_on(link->parent, use_lock, depth + 1); - - if (use_lock) - genpd_unlock(link->parent); - } - - _genpd_power_on(genpd, false); - genpd->status = GENPD_STATE_ON; -} - -/** - * genpd_prepare - Start power transition of a device in a PM domain. - * @dev: Device to start the transition of. - * - * Start a power transition of a device (during a system-wide power transition) - * under the assumption that its pm_domain field points to the domain member of - * an object of type struct generic_pm_domain representing a PM domain - * consisting of I/O devices. - */ -static int genpd_prepare(struct device *dev) -{ - struct generic_pm_domain *genpd; - int ret; - - dev_dbg(dev, "%s()\n", __func__); - - genpd = dev_to_genpd(dev); - if (IS_ERR(genpd)) - return -EINVAL; - - genpd_lock(genpd); - - if (genpd->prepared_count++ == 0) - genpd->suspended_count = 0; - - genpd_unlock(genpd); - - ret = pm_generic_prepare(dev); - if (ret < 0) { - genpd_lock(genpd); - - genpd->prepared_count--; - - genpd_unlock(genpd); - } - - /* Never return 1, as genpd don't cope with the direct_complete path. */ - return ret >= 0 ? 0 : ret; -} - -/** - * genpd_finish_suspend - Completion of suspend or hibernation of device in an - * I/O pm domain. - * @dev: Device to suspend. - * @poweroff: Specifies if this is a poweroff_noirq or suspend_noirq callback. - * - * Stop the device and remove power from the domain if all devices in it have - * been stopped. - */ -static int genpd_finish_suspend(struct device *dev, bool poweroff) -{ - struct generic_pm_domain *genpd; - int ret = 0; - - genpd = dev_to_genpd(dev); - if (IS_ERR(genpd)) - return -EINVAL; - - if (poweroff) - ret = pm_generic_poweroff_noirq(dev); - else - ret = pm_generic_suspend_noirq(dev); - if (ret) - return ret; - - if (device_wakeup_path(dev) && genpd_is_active_wakeup(genpd)) - return 0; - - if (genpd->dev_ops.stop && genpd->dev_ops.start && - !pm_runtime_status_suspended(dev)) { - ret = genpd_stop_dev(genpd, dev); - if (ret) { - if (poweroff) - pm_generic_restore_noirq(dev); - else - pm_generic_resume_noirq(dev); - return ret; - } - } - - genpd_lock(genpd); - genpd->suspended_count++; - genpd_sync_power_off(genpd, true, 0); - genpd_unlock(genpd); - - return 0; -} - -/** - * genpd_suspend_noirq - Completion of suspend of device in an I/O PM domain. - * @dev: Device to suspend. - * - * Stop the device and remove power from the domain if all devices in it have - * been stopped. - */ -static int genpd_suspend_noirq(struct device *dev) -{ - dev_dbg(dev, "%s()\n", __func__); - - return genpd_finish_suspend(dev, false); -} - -/** - * genpd_resume_noirq - Start of resume of device in an I/O PM domain. - * @dev: Device to resume. - * - * Restore power to the device's PM domain, if necessary, and start the device. - */ -static int genpd_resume_noirq(struct device *dev) -{ - struct generic_pm_domain *genpd; - int ret; - - dev_dbg(dev, "%s()\n", __func__); - - genpd = dev_to_genpd(dev); - if (IS_ERR(genpd)) - return -EINVAL; - - if (device_wakeup_path(dev) && genpd_is_active_wakeup(genpd)) - return pm_generic_resume_noirq(dev); - - genpd_lock(genpd); - genpd_sync_power_on(genpd, true, 0); - genpd->suspended_count--; - genpd_unlock(genpd); - - if (genpd->dev_ops.stop && genpd->dev_ops.start && - !pm_runtime_status_suspended(dev)) { - ret = genpd_start_dev(genpd, dev); - if (ret) - return ret; - } - - return pm_generic_resume_noirq(dev); -} - -/** - * genpd_freeze_noirq - Completion of freezing a device in an I/O PM domain. - * @dev: Device to freeze. - * - * Carry out a late freeze of a device under the assumption that its - * pm_domain field points to the domain member of an object of type - * struct generic_pm_domain representing a power domain consisting of I/O - * devices. - */ -static int genpd_freeze_noirq(struct device *dev) -{ - const struct generic_pm_domain *genpd; - int ret = 0; - - dev_dbg(dev, "%s()\n", __func__); - - genpd = dev_to_genpd(dev); - if (IS_ERR(genpd)) - return -EINVAL; - - ret = pm_generic_freeze_noirq(dev); - if (ret) - return ret; - - if (genpd->dev_ops.stop && genpd->dev_ops.start && - !pm_runtime_status_suspended(dev)) - ret = genpd_stop_dev(genpd, dev); - - return ret; -} - -/** - * genpd_thaw_noirq - Early thaw of device in an I/O PM domain. - * @dev: Device to thaw. - * - * Start the device, unless power has been removed from the domain already - * before the system transition. - */ -static int genpd_thaw_noirq(struct device *dev) -{ - const struct generic_pm_domain *genpd; - int ret = 0; - - dev_dbg(dev, "%s()\n", __func__); - - genpd = dev_to_genpd(dev); - if (IS_ERR(genpd)) - return -EINVAL; - - if (genpd->dev_ops.stop && genpd->dev_ops.start && - !pm_runtime_status_suspended(dev)) { - ret = genpd_start_dev(genpd, dev); - if (ret) - return ret; - } - - return pm_generic_thaw_noirq(dev); -} - -/** - * genpd_poweroff_noirq - Completion of hibernation of device in an - * I/O PM domain. - * @dev: Device to poweroff. - * - * Stop the device and remove power from the domain if all devices in it have - * been stopped. - */ -static int genpd_poweroff_noirq(struct device *dev) -{ - dev_dbg(dev, "%s()\n", __func__); - - return genpd_finish_suspend(dev, true); -} - -/** - * genpd_restore_noirq - Start of restore of device in an I/O PM domain. - * @dev: Device to resume. - * - * Make sure the domain will be in the same power state as before the - * hibernation the system is resuming from and start the device if necessary. - */ -static int genpd_restore_noirq(struct device *dev) -{ - struct generic_pm_domain *genpd; - int ret = 0; - - dev_dbg(dev, "%s()\n", __func__); - - genpd = dev_to_genpd(dev); - if (IS_ERR(genpd)) - return -EINVAL; - - /* - * At this point suspended_count == 0 means we are being run for the - * first time for the given domain in the present cycle. - */ - genpd_lock(genpd); - if (genpd->suspended_count++ == 0) { - /* - * The boot kernel might put the domain into arbitrary state, - * so make it appear as powered off to genpd_sync_power_on(), - * so that it tries to power it on in case it was really off. - */ - genpd->status = GENPD_STATE_OFF; - } - - genpd_sync_power_on(genpd, true, 0); - genpd_unlock(genpd); - - if (genpd->dev_ops.stop && genpd->dev_ops.start && - !pm_runtime_status_suspended(dev)) { - ret = genpd_start_dev(genpd, dev); - if (ret) - return ret; - } - - return pm_generic_restore_noirq(dev); -} - -/** - * genpd_complete - Complete power transition of a device in a power domain. - * @dev: Device to complete the transition of. - * - * Complete a power transition of a device (during a system-wide power - * transition) under the assumption that its pm_domain field points to the - * domain member of an object of type struct generic_pm_domain representing - * a power domain consisting of I/O devices. - */ -static void genpd_complete(struct device *dev) -{ - struct generic_pm_domain *genpd; - - dev_dbg(dev, "%s()\n", __func__); - - genpd = dev_to_genpd(dev); - if (IS_ERR(genpd)) - return; - - pm_generic_complete(dev); - - genpd_lock(genpd); - - genpd->prepared_count--; - if (!genpd->prepared_count) - genpd_queue_power_off_work(genpd); - - genpd_unlock(genpd); -} - -static void genpd_switch_state(struct device *dev, bool suspend) -{ - struct generic_pm_domain *genpd; - bool use_lock; - - genpd = dev_to_genpd_safe(dev); - if (!genpd) - return; - - use_lock = genpd_is_irq_safe(genpd); - - if (use_lock) - genpd_lock(genpd); - - if (suspend) { - genpd->suspended_count++; - genpd_sync_power_off(genpd, use_lock, 0); - } else { - genpd_sync_power_on(genpd, use_lock, 0); - genpd->suspended_count--; - } - - if (use_lock) - genpd_unlock(genpd); -} - -/** - * dev_pm_genpd_suspend - Synchronously try to suspend the genpd for @dev - * @dev: The device that is attached to the genpd, that can be suspended. - * - * This routine should typically be called for a device that needs to be - * suspended during the syscore suspend phase. It may also be called during - * suspend-to-idle to suspend a corresponding CPU device that is attached to a - * genpd. - */ -void dev_pm_genpd_suspend(struct device *dev) -{ - genpd_switch_state(dev, true); -} -EXPORT_SYMBOL_GPL(dev_pm_genpd_suspend); - -/** - * dev_pm_genpd_resume - Synchronously try to resume the genpd for @dev - * @dev: The device that is attached to the genpd, which needs to be resumed. - * - * This routine should typically be called for a device that needs to be resumed - * during the syscore resume phase. It may also be called during suspend-to-idle - * to resume a corresponding CPU device that is attached to a genpd. - */ -void dev_pm_genpd_resume(struct device *dev) -{ - genpd_switch_state(dev, false); -} -EXPORT_SYMBOL_GPL(dev_pm_genpd_resume); - -#else /* !CONFIG_PM_SLEEP */ - -#define genpd_prepare NULL -#define genpd_suspend_noirq NULL -#define genpd_resume_noirq NULL -#define genpd_freeze_noirq NULL -#define genpd_thaw_noirq NULL -#define genpd_poweroff_noirq NULL -#define genpd_restore_noirq NULL -#define genpd_complete NULL - -#endif /* CONFIG_PM_SLEEP */ - -static struct generic_pm_domain_data *genpd_alloc_dev_data(struct device *dev) -{ - struct generic_pm_domain_data *gpd_data; - int ret; - - ret = dev_pm_get_subsys_data(dev); - if (ret) - return ERR_PTR(ret); - - gpd_data = kzalloc(sizeof(*gpd_data), GFP_KERNEL); - if (!gpd_data) { - ret = -ENOMEM; - goto err_put; - } - - gpd_data->base.dev = dev; - gpd_data->td.constraint_changed = true; - gpd_data->td.effective_constraint_ns = PM_QOS_RESUME_LATENCY_NO_CONSTRAINT_NS; - gpd_data->nb.notifier_call = genpd_dev_pm_qos_notifier; - gpd_data->next_wakeup = KTIME_MAX; - - spin_lock_irq(&dev->power.lock); - - if (dev->power.subsys_data->domain_data) { - ret = -EINVAL; - goto err_free; - } - - dev->power.subsys_data->domain_data = &gpd_data->base; - - spin_unlock_irq(&dev->power.lock); - - return gpd_data; - - err_free: - spin_unlock_irq(&dev->power.lock); - kfree(gpd_data); - err_put: - dev_pm_put_subsys_data(dev); - return ERR_PTR(ret); -} - -static void genpd_free_dev_data(struct device *dev, - struct generic_pm_domain_data *gpd_data) -{ - spin_lock_irq(&dev->power.lock); - - dev->power.subsys_data->domain_data = NULL; - - spin_unlock_irq(&dev->power.lock); - - kfree(gpd_data); - dev_pm_put_subsys_data(dev); -} - -static void genpd_update_cpumask(struct generic_pm_domain *genpd, - int cpu, bool set, unsigned int depth) -{ - struct gpd_link *link; - - if (!genpd_is_cpu_domain(genpd)) - return; - - list_for_each_entry(link, &genpd->child_links, child_node) { - struct generic_pm_domain *parent = link->parent; - - genpd_lock_nested(parent, depth + 1); - genpd_update_cpumask(parent, cpu, set, depth + 1); - genpd_unlock(parent); - } - - if (set) - cpumask_set_cpu(cpu, genpd->cpus); - else - cpumask_clear_cpu(cpu, genpd->cpus); -} - -static void genpd_set_cpumask(struct generic_pm_domain *genpd, int cpu) -{ - if (cpu >= 0) - genpd_update_cpumask(genpd, cpu, true, 0); -} - -static void genpd_clear_cpumask(struct generic_pm_domain *genpd, int cpu) -{ - if (cpu >= 0) - genpd_update_cpumask(genpd, cpu, false, 0); -} - -static int genpd_get_cpu(struct generic_pm_domain *genpd, struct device *dev) -{ - int cpu; - - if (!genpd_is_cpu_domain(genpd)) - return -1; - - for_each_possible_cpu(cpu) { - if (get_cpu_device(cpu) == dev) - return cpu; - } - - return -1; -} - -static int genpd_add_device(struct generic_pm_domain *genpd, struct device *dev, - struct device *base_dev) -{ - struct generic_pm_domain_data *gpd_data; - int ret; - - dev_dbg(dev, "%s()\n", __func__); - - if (IS_ERR_OR_NULL(genpd) || IS_ERR_OR_NULL(dev)) - return -EINVAL; - - gpd_data = genpd_alloc_dev_data(dev); - if (IS_ERR(gpd_data)) - return PTR_ERR(gpd_data); - - gpd_data->cpu = genpd_get_cpu(genpd, base_dev); - - ret = genpd->attach_dev ? genpd->attach_dev(genpd, dev) : 0; - if (ret) - goto out; - - genpd_lock(genpd); - - genpd_set_cpumask(genpd, gpd_data->cpu); - dev_pm_domain_set(dev, &genpd->domain); - - genpd->device_count++; - genpd->max_off_time_changed = true; - - list_add_tail(&gpd_data->base.list_node, &genpd->dev_list); - - genpd_unlock(genpd); - out: - if (ret) - genpd_free_dev_data(dev, gpd_data); - else - dev_pm_qos_add_notifier(dev, &gpd_data->nb, - DEV_PM_QOS_RESUME_LATENCY); - - return ret; -} - -/** - * pm_genpd_add_device - Add a device to an I/O PM domain. - * @genpd: PM domain to add the device to. - * @dev: Device to be added. - */ -int pm_genpd_add_device(struct generic_pm_domain *genpd, struct device *dev) -{ - int ret; - - mutex_lock(&gpd_list_lock); - ret = genpd_add_device(genpd, dev, dev); - mutex_unlock(&gpd_list_lock); - - return ret; -} -EXPORT_SYMBOL_GPL(pm_genpd_add_device); - -static int genpd_remove_device(struct generic_pm_domain *genpd, - struct device *dev) -{ - struct generic_pm_domain_data *gpd_data; - struct pm_domain_data *pdd; - int ret = 0; - - dev_dbg(dev, "%s()\n", __func__); - - pdd = dev->power.subsys_data->domain_data; - gpd_data = to_gpd_data(pdd); - dev_pm_qos_remove_notifier(dev, &gpd_data->nb, - DEV_PM_QOS_RESUME_LATENCY); - - genpd_lock(genpd); - - if (genpd->prepared_count > 0) { - ret = -EAGAIN; - goto out; - } - - genpd->device_count--; - genpd->max_off_time_changed = true; - - genpd_clear_cpumask(genpd, gpd_data->cpu); - dev_pm_domain_set(dev, NULL); - - list_del_init(&pdd->list_node); - - genpd_unlock(genpd); - - if (genpd->detach_dev) - genpd->detach_dev(genpd, dev); - - genpd_free_dev_data(dev, gpd_data); - - return 0; - - out: - genpd_unlock(genpd); - dev_pm_qos_add_notifier(dev, &gpd_data->nb, DEV_PM_QOS_RESUME_LATENCY); - - return ret; -} - -/** - * pm_genpd_remove_device - Remove a device from an I/O PM domain. - * @dev: Device to be removed. - */ -int pm_genpd_remove_device(struct device *dev) -{ - struct generic_pm_domain *genpd = dev_to_genpd_safe(dev); - - if (!genpd) - return -EINVAL; - - return genpd_remove_device(genpd, dev); -} -EXPORT_SYMBOL_GPL(pm_genpd_remove_device); - -/** - * dev_pm_genpd_add_notifier - Add a genpd power on/off notifier for @dev - * - * @dev: Device that should be associated with the notifier - * @nb: The notifier block to register - * - * Users may call this function to add a genpd power on/off notifier for an - * attached @dev. Only one notifier per device is allowed. The notifier is - * sent when genpd is powering on/off the PM domain. - * - * It is assumed that the user guarantee that the genpd wouldn't be detached - * while this routine is getting called. - * - * Returns 0 on success and negative error values on failures. - */ -int dev_pm_genpd_add_notifier(struct device *dev, struct notifier_block *nb) -{ - struct generic_pm_domain *genpd; - struct generic_pm_domain_data *gpd_data; - int ret; - - genpd = dev_to_genpd_safe(dev); - if (!genpd) - return -ENODEV; - - if (WARN_ON(!dev->power.subsys_data || - !dev->power.subsys_data->domain_data)) - return -EINVAL; - - gpd_data = to_gpd_data(dev->power.subsys_data->domain_data); - if (gpd_data->power_nb) - return -EEXIST; - - genpd_lock(genpd); - ret = raw_notifier_chain_register(&genpd->power_notifiers, nb); - genpd_unlock(genpd); - - if (ret) { - dev_warn(dev, "failed to add notifier for PM domain %s\n", - genpd->name); - return ret; - } - - gpd_data->power_nb = nb; - return 0; -} -EXPORT_SYMBOL_GPL(dev_pm_genpd_add_notifier); - -/** - * dev_pm_genpd_remove_notifier - Remove a genpd power on/off notifier for @dev - * - * @dev: Device that is associated with the notifier - * - * Users may call this function to remove a genpd power on/off notifier for an - * attached @dev. - * - * It is assumed that the user guarantee that the genpd wouldn't be detached - * while this routine is getting called. - * - * Returns 0 on success and negative error values on failures. - */ -int dev_pm_genpd_remove_notifier(struct device *dev) -{ - struct generic_pm_domain *genpd; - struct generic_pm_domain_data *gpd_data; - int ret; - - genpd = dev_to_genpd_safe(dev); - if (!genpd) - return -ENODEV; - - if (WARN_ON(!dev->power.subsys_data || - !dev->power.subsys_data->domain_data)) - return -EINVAL; - - gpd_data = to_gpd_data(dev->power.subsys_data->domain_data); - if (!gpd_data->power_nb) - return -ENODEV; - - genpd_lock(genpd); - ret = raw_notifier_chain_unregister(&genpd->power_notifiers, - gpd_data->power_nb); - genpd_unlock(genpd); - - if (ret) { - dev_warn(dev, "failed to remove notifier for PM domain %s\n", - genpd->name); - return ret; - } - - gpd_data->power_nb = NULL; - return 0; -} -EXPORT_SYMBOL_GPL(dev_pm_genpd_remove_notifier); - -static int genpd_add_subdomain(struct generic_pm_domain *genpd, - struct generic_pm_domain *subdomain) -{ - struct gpd_link *link, *itr; - int ret = 0; - - if (IS_ERR_OR_NULL(genpd) || IS_ERR_OR_NULL(subdomain) - || genpd == subdomain) - return -EINVAL; - - /* - * If the domain can be powered on/off in an IRQ safe - * context, ensure that the subdomain can also be - * powered on/off in that context. - */ - if (!genpd_is_irq_safe(genpd) && genpd_is_irq_safe(subdomain)) { - WARN(1, "Parent %s of subdomain %s must be IRQ safe\n", - genpd->name, subdomain->name); - return -EINVAL; - } - - link = kzalloc(sizeof(*link), GFP_KERNEL); - if (!link) - return -ENOMEM; - - genpd_lock(subdomain); - genpd_lock_nested(genpd, SINGLE_DEPTH_NESTING); - - if (!genpd_status_on(genpd) && genpd_status_on(subdomain)) { - ret = -EINVAL; - goto out; - } - - list_for_each_entry(itr, &genpd->parent_links, parent_node) { - if (itr->child == subdomain && itr->parent == genpd) { - ret = -EINVAL; - goto out; - } - } - - link->parent = genpd; - list_add_tail(&link->parent_node, &genpd->parent_links); - link->child = subdomain; - list_add_tail(&link->child_node, &subdomain->child_links); - if (genpd_status_on(subdomain)) - genpd_sd_counter_inc(genpd); - - out: - genpd_unlock(genpd); - genpd_unlock(subdomain); - if (ret) - kfree(link); - return ret; -} - -/** - * pm_genpd_add_subdomain - Add a subdomain to an I/O PM domain. - * @genpd: Leader PM domain to add the subdomain to. - * @subdomain: Subdomain to be added. - */ -int pm_genpd_add_subdomain(struct generic_pm_domain *genpd, - struct generic_pm_domain *subdomain) -{ - int ret; - - mutex_lock(&gpd_list_lock); - ret = genpd_add_subdomain(genpd, subdomain); - mutex_unlock(&gpd_list_lock); - - return ret; -} -EXPORT_SYMBOL_GPL(pm_genpd_add_subdomain); - -/** - * pm_genpd_remove_subdomain - Remove a subdomain from an I/O PM domain. - * @genpd: Leader PM domain to remove the subdomain from. - * @subdomain: Subdomain to be removed. - */ -int pm_genpd_remove_subdomain(struct generic_pm_domain *genpd, - struct generic_pm_domain *subdomain) -{ - struct gpd_link *l, *link; - int ret = -EINVAL; - - if (IS_ERR_OR_NULL(genpd) || IS_ERR_OR_NULL(subdomain)) - return -EINVAL; - - genpd_lock(subdomain); - genpd_lock_nested(genpd, SINGLE_DEPTH_NESTING); - - if (!list_empty(&subdomain->parent_links) || subdomain->device_count) { - pr_warn("%s: unable to remove subdomain %s\n", - genpd->name, subdomain->name); - ret = -EBUSY; - goto out; - } - - list_for_each_entry_safe(link, l, &genpd->parent_links, parent_node) { - if (link->child != subdomain) - continue; - - list_del(&link->parent_node); - list_del(&link->child_node); - kfree(link); - if (genpd_status_on(subdomain)) - genpd_sd_counter_dec(genpd); - - ret = 0; - break; - } - -out: - genpd_unlock(genpd); - genpd_unlock(subdomain); - - return ret; -} -EXPORT_SYMBOL_GPL(pm_genpd_remove_subdomain); - -static void genpd_free_default_power_state(struct genpd_power_state *states, - unsigned int state_count) -{ - kfree(states); -} - -static int genpd_set_default_power_state(struct generic_pm_domain *genpd) -{ - struct genpd_power_state *state; - - state = kzalloc(sizeof(*state), GFP_KERNEL); - if (!state) - return -ENOMEM; - - genpd->states = state; - genpd->state_count = 1; - genpd->free_states = genpd_free_default_power_state; - - return 0; -} - -static void genpd_lock_init(struct generic_pm_domain *genpd) -{ - if (genpd->flags & GENPD_FLAG_IRQ_SAFE) { - spin_lock_init(&genpd->slock); - genpd->lock_ops = &genpd_spin_ops; - } else { - mutex_init(&genpd->mlock); - genpd->lock_ops = &genpd_mtx_ops; - } -} - -/** - * pm_genpd_init - Initialize a generic I/O PM domain object. - * @genpd: PM domain object to initialize. - * @gov: PM domain governor to associate with the domain (may be NULL). - * @is_off: Initial value of the domain's power_is_off field. - * - * Returns 0 on successful initialization, else a negative error code. - */ -int pm_genpd_init(struct generic_pm_domain *genpd, - struct dev_power_governor *gov, bool is_off) -{ - int ret; - - if (IS_ERR_OR_NULL(genpd)) - return -EINVAL; - - INIT_LIST_HEAD(&genpd->parent_links); - INIT_LIST_HEAD(&genpd->child_links); - INIT_LIST_HEAD(&genpd->dev_list); - RAW_INIT_NOTIFIER_HEAD(&genpd->power_notifiers); - genpd_lock_init(genpd); - genpd->gov = gov; - INIT_WORK(&genpd->power_off_work, genpd_power_off_work_fn); - atomic_set(&genpd->sd_count, 0); - genpd->status = is_off ? GENPD_STATE_OFF : GENPD_STATE_ON; - genpd->device_count = 0; - genpd->max_off_time_ns = -1; - genpd->max_off_time_changed = true; - genpd->provider = NULL; - genpd->has_provider = false; - genpd->accounting_time = ktime_get(); - genpd->domain.ops.runtime_suspend = genpd_runtime_suspend; - genpd->domain.ops.runtime_resume = genpd_runtime_resume; - genpd->domain.ops.prepare = genpd_prepare; - genpd->domain.ops.suspend_noirq = genpd_suspend_noirq; - genpd->domain.ops.resume_noirq = genpd_resume_noirq; - genpd->domain.ops.freeze_noirq = genpd_freeze_noirq; - genpd->domain.ops.thaw_noirq = genpd_thaw_noirq; - genpd->domain.ops.poweroff_noirq = genpd_poweroff_noirq; - genpd->domain.ops.restore_noirq = genpd_restore_noirq; - genpd->domain.ops.complete = genpd_complete; - genpd->domain.start = genpd_dev_pm_start; - - if (genpd->flags & GENPD_FLAG_PM_CLK) { - genpd->dev_ops.stop = pm_clk_suspend; - genpd->dev_ops.start = pm_clk_resume; - } - - /* Always-on domains must be powered on at initialization. */ - if ((genpd_is_always_on(genpd) || genpd_is_rpm_always_on(genpd)) && - !genpd_status_on(genpd)) - return -EINVAL; - - if (genpd_is_cpu_domain(genpd) && - !zalloc_cpumask_var(&genpd->cpus, GFP_KERNEL)) - return -ENOMEM; - - /* Use only one "off" state if there were no states declared */ - if (genpd->state_count == 0) { - ret = genpd_set_default_power_state(genpd); - if (ret) { - if (genpd_is_cpu_domain(genpd)) - free_cpumask_var(genpd->cpus); - return ret; - } - } else if (!gov && genpd->state_count > 1) { - pr_warn("%s: no governor for states\n", genpd->name); - } - - device_initialize(&genpd->dev); - dev_set_name(&genpd->dev, "%s", genpd->name); - - mutex_lock(&gpd_list_lock); - list_add(&genpd->gpd_list_node, &gpd_list); - mutex_unlock(&gpd_list_lock); - genpd_debug_add(genpd); - - return 0; -} -EXPORT_SYMBOL_GPL(pm_genpd_init); - -static int genpd_remove(struct generic_pm_domain *genpd) -{ - struct gpd_link *l, *link; - - if (IS_ERR_OR_NULL(genpd)) - return -EINVAL; - - genpd_lock(genpd); - - if (genpd->has_provider) { - genpd_unlock(genpd); - pr_err("Provider present, unable to remove %s\n", genpd->name); - return -EBUSY; - } - - if (!list_empty(&genpd->parent_links) || genpd->device_count) { - genpd_unlock(genpd); - pr_err("%s: unable to remove %s\n", __func__, genpd->name); - return -EBUSY; - } - - list_for_each_entry_safe(link, l, &genpd->child_links, child_node) { - list_del(&link->parent_node); - list_del(&link->child_node); - kfree(link); - } - - genpd_debug_remove(genpd); - list_del(&genpd->gpd_list_node); - genpd_unlock(genpd); - cancel_work_sync(&genpd->power_off_work); - if (genpd_is_cpu_domain(genpd)) - free_cpumask_var(genpd->cpus); - if (genpd->free_states) - genpd->free_states(genpd->states, genpd->state_count); - - pr_debug("%s: removed %s\n", __func__, genpd->name); - - return 0; -} - -/** - * pm_genpd_remove - Remove a generic I/O PM domain - * @genpd: Pointer to PM domain that is to be removed. - * - * To remove the PM domain, this function: - * - Removes the PM domain as a subdomain to any parent domains, - * if it was added. - * - Removes the PM domain from the list of registered PM domains. - * - * The PM domain will only be removed, if the associated provider has - * been removed, it is not a parent to any other PM domain and has no - * devices associated with it. - */ -int pm_genpd_remove(struct generic_pm_domain *genpd) -{ - int ret; - - mutex_lock(&gpd_list_lock); - ret = genpd_remove(genpd); - mutex_unlock(&gpd_list_lock); - - return ret; -} -EXPORT_SYMBOL_GPL(pm_genpd_remove); - -#ifdef CONFIG_PM_GENERIC_DOMAINS_OF - -/* - * Device Tree based PM domain providers. - * - * The code below implements generic device tree based PM domain providers that - * bind device tree nodes with generic PM domains registered in the system. - * - * Any driver that registers generic PM domains and needs to support binding of - * devices to these domains is supposed to register a PM domain provider, which - * maps a PM domain specifier retrieved from the device tree to a PM domain. - * - * Two simple mapping functions have been provided for convenience: - * - genpd_xlate_simple() for 1:1 device tree node to PM domain mapping. - * - genpd_xlate_onecell() for mapping of multiple PM domains per node by - * index. - */ - -/** - * struct of_genpd_provider - PM domain provider registration structure - * @link: Entry in global list of PM domain providers - * @node: Pointer to device tree node of PM domain provider - * @xlate: Provider-specific xlate callback mapping a set of specifier cells - * into a PM domain. - * @data: context pointer to be passed into @xlate callback - */ -struct of_genpd_provider { - struct list_head link; - struct device_node *node; - genpd_xlate_t xlate; - void *data; -}; - -/* List of registered PM domain providers. */ -static LIST_HEAD(of_genpd_providers); -/* Mutex to protect the list above. */ -static DEFINE_MUTEX(of_genpd_mutex); - -/** - * genpd_xlate_simple() - Xlate function for direct node-domain mapping - * @genpdspec: OF phandle args to map into a PM domain - * @data: xlate function private data - pointer to struct generic_pm_domain - * - * This is a generic xlate function that can be used to model PM domains that - * have their own device tree nodes. The private data of xlate function needs - * to be a valid pointer to struct generic_pm_domain. - */ -static struct generic_pm_domain *genpd_xlate_simple( - struct of_phandle_args *genpdspec, - void *data) -{ - return data; -} - -/** - * genpd_xlate_onecell() - Xlate function using a single index. - * @genpdspec: OF phandle args to map into a PM domain - * @data: xlate function private data - pointer to struct genpd_onecell_data - * - * This is a generic xlate function that can be used to model simple PM domain - * controllers that have one device tree node and provide multiple PM domains. - * A single cell is used as an index into an array of PM domains specified in - * the genpd_onecell_data struct when registering the provider. - */ -static struct generic_pm_domain *genpd_xlate_onecell( - struct of_phandle_args *genpdspec, - void *data) -{ - struct genpd_onecell_data *genpd_data = data; - unsigned int idx = genpdspec->args[0]; - - if (genpdspec->args_count != 1) - return ERR_PTR(-EINVAL); - - if (idx >= genpd_data->num_domains) { - pr_err("%s: invalid domain index %u\n", __func__, idx); - return ERR_PTR(-EINVAL); - } - - if (!genpd_data->domains[idx]) - return ERR_PTR(-ENOENT); - - return genpd_data->domains[idx]; -} - -/** - * genpd_add_provider() - Register a PM domain provider for a node - * @np: Device node pointer associated with the PM domain provider. - * @xlate: Callback for decoding PM domain from phandle arguments. - * @data: Context pointer for @xlate callback. - */ -static int genpd_add_provider(struct device_node *np, genpd_xlate_t xlate, - void *data) -{ - struct of_genpd_provider *cp; - - cp = kzalloc(sizeof(*cp), GFP_KERNEL); - if (!cp) - return -ENOMEM; - - cp->node = of_node_get(np); - cp->data = data; - cp->xlate = xlate; - fwnode_dev_initialized(&np->fwnode, true); - - mutex_lock(&of_genpd_mutex); - list_add(&cp->link, &of_genpd_providers); - mutex_unlock(&of_genpd_mutex); - pr_debug("Added domain provider from %pOF\n", np); - - return 0; -} - -static bool genpd_present(const struct generic_pm_domain *genpd) -{ - bool ret = false; - const struct generic_pm_domain *gpd; - - mutex_lock(&gpd_list_lock); - list_for_each_entry(gpd, &gpd_list, gpd_list_node) { - if (gpd == genpd) { - ret = true; - break; - } - } - mutex_unlock(&gpd_list_lock); - - return ret; -} - -/** - * of_genpd_add_provider_simple() - Register a simple PM domain provider - * @np: Device node pointer associated with the PM domain provider. - * @genpd: Pointer to PM domain associated with the PM domain provider. - */ -int of_genpd_add_provider_simple(struct device_node *np, - struct generic_pm_domain *genpd) -{ - int ret; - - if (!np || !genpd) - return -EINVAL; - - if (!genpd_present(genpd)) - return -EINVAL; - - genpd->dev.of_node = np; - - /* Parse genpd OPP table */ - if (genpd->set_performance_state) { - ret = dev_pm_opp_of_add_table(&genpd->dev); - if (ret) { - if (ret != -EPROBE_DEFER) - dev_err(&genpd->dev, "Failed to add OPP table: %d\n", - ret); - return ret; - } - - /* - * Save table for faster processing while setting performance - * state. - */ - genpd->opp_table = dev_pm_opp_get_opp_table(&genpd->dev); - WARN_ON(IS_ERR(genpd->opp_table)); - } - - ret = genpd_add_provider(np, genpd_xlate_simple, genpd); - if (ret) { - if (genpd->set_performance_state) { - dev_pm_opp_put_opp_table(genpd->opp_table); - dev_pm_opp_of_remove_table(&genpd->dev); - } - - return ret; - } - - genpd->provider = &np->fwnode; - genpd->has_provider = true; - - return 0; -} -EXPORT_SYMBOL_GPL(of_genpd_add_provider_simple); - -/** - * of_genpd_add_provider_onecell() - Register a onecell PM domain provider - * @np: Device node pointer associated with the PM domain provider. - * @data: Pointer to the data associated with the PM domain provider. - */ -int of_genpd_add_provider_onecell(struct device_node *np, - struct genpd_onecell_data *data) -{ - struct generic_pm_domain *genpd; - unsigned int i; - int ret = -EINVAL; - - if (!np || !data) - return -EINVAL; - - if (!data->xlate) - data->xlate = genpd_xlate_onecell; - - for (i = 0; i < data->num_domains; i++) { - genpd = data->domains[i]; - - if (!genpd) - continue; - if (!genpd_present(genpd)) - goto error; - - genpd->dev.of_node = np; - - /* Parse genpd OPP table */ - if (genpd->set_performance_state) { - ret = dev_pm_opp_of_add_table_indexed(&genpd->dev, i); - if (ret) { - if (ret != -EPROBE_DEFER) - dev_err(&genpd->dev, "Failed to add OPP table for index %d: %d\n", - i, ret); - goto error; - } - - /* - * Save table for faster processing while setting - * performance state. - */ - genpd->opp_table = dev_pm_opp_get_opp_table(&genpd->dev); - WARN_ON(IS_ERR(genpd->opp_table)); - } - - genpd->provider = &np->fwnode; - genpd->has_provider = true; - } - - ret = genpd_add_provider(np, data->xlate, data); - if (ret < 0) - goto error; - - return 0; - -error: - while (i--) { - genpd = data->domains[i]; - - if (!genpd) - continue; - - genpd->provider = NULL; - genpd->has_provider = false; - - if (genpd->set_performance_state) { - dev_pm_opp_put_opp_table(genpd->opp_table); - dev_pm_opp_of_remove_table(&genpd->dev); - } - } - - return ret; -} -EXPORT_SYMBOL_GPL(of_genpd_add_provider_onecell); - -/** - * of_genpd_del_provider() - Remove a previously registered PM domain provider - * @np: Device node pointer associated with the PM domain provider - */ -void of_genpd_del_provider(struct device_node *np) -{ - struct of_genpd_provider *cp, *tmp; - struct generic_pm_domain *gpd; - - mutex_lock(&gpd_list_lock); - mutex_lock(&of_genpd_mutex); - list_for_each_entry_safe(cp, tmp, &of_genpd_providers, link) { - if (cp->node == np) { - /* - * For each PM domain associated with the - * provider, set the 'has_provider' to false - * so that the PM domain can be safely removed. - */ - list_for_each_entry(gpd, &gpd_list, gpd_list_node) { - if (gpd->provider == &np->fwnode) { - gpd->has_provider = false; - - if (!gpd->set_performance_state) - continue; - - dev_pm_opp_put_opp_table(gpd->opp_table); - dev_pm_opp_of_remove_table(&gpd->dev); - } - } - - fwnode_dev_initialized(&cp->node->fwnode, false); - list_del(&cp->link); - of_node_put(cp->node); - kfree(cp); - break; - } - } - mutex_unlock(&of_genpd_mutex); - mutex_unlock(&gpd_list_lock); -} -EXPORT_SYMBOL_GPL(of_genpd_del_provider); - -/** - * genpd_get_from_provider() - Look-up PM domain - * @genpdspec: OF phandle args to use for look-up - * - * Looks for a PM domain provider under the node specified by @genpdspec and if - * found, uses xlate function of the provider to map phandle args to a PM - * domain. - * - * Returns a valid pointer to struct generic_pm_domain on success or ERR_PTR() - * on failure. - */ -static struct generic_pm_domain *genpd_get_from_provider( - struct of_phandle_args *genpdspec) -{ - struct generic_pm_domain *genpd = ERR_PTR(-ENOENT); - struct of_genpd_provider *provider; - - if (!genpdspec) - return ERR_PTR(-EINVAL); - - mutex_lock(&of_genpd_mutex); - - /* Check if we have such a provider in our array */ - list_for_each_entry(provider, &of_genpd_providers, link) { - if (provider->node == genpdspec->np) - genpd = provider->xlate(genpdspec, provider->data); - if (!IS_ERR(genpd)) - break; - } - - mutex_unlock(&of_genpd_mutex); - - return genpd; -} - -/** - * of_genpd_add_device() - Add a device to an I/O PM domain - * @genpdspec: OF phandle args to use for look-up PM domain - * @dev: Device to be added. - * - * Looks-up an I/O PM domain based upon phandle args provided and adds - * the device to the PM domain. Returns a negative error code on failure. - */ -int of_genpd_add_device(struct of_phandle_args *genpdspec, struct device *dev) -{ - struct generic_pm_domain *genpd; - int ret; - - mutex_lock(&gpd_list_lock); - - genpd = genpd_get_from_provider(genpdspec); - if (IS_ERR(genpd)) { - ret = PTR_ERR(genpd); - goto out; - } - - ret = genpd_add_device(genpd, dev, dev); - -out: - mutex_unlock(&gpd_list_lock); - - return ret; -} -EXPORT_SYMBOL_GPL(of_genpd_add_device); - -/** - * of_genpd_add_subdomain - Add a subdomain to an I/O PM domain. - * @parent_spec: OF phandle args to use for parent PM domain look-up - * @subdomain_spec: OF phandle args to use for subdomain look-up - * - * Looks-up a parent PM domain and subdomain based upon phandle args - * provided and adds the subdomain to the parent PM domain. Returns a - * negative error code on failure. - */ -int of_genpd_add_subdomain(struct of_phandle_args *parent_spec, - struct of_phandle_args *subdomain_spec) -{ - struct generic_pm_domain *parent, *subdomain; - int ret; - - mutex_lock(&gpd_list_lock); - - parent = genpd_get_from_provider(parent_spec); - if (IS_ERR(parent)) { - ret = PTR_ERR(parent); - goto out; - } - - subdomain = genpd_get_from_provider(subdomain_spec); - if (IS_ERR(subdomain)) { - ret = PTR_ERR(subdomain); - goto out; - } - - ret = genpd_add_subdomain(parent, subdomain); - -out: - mutex_unlock(&gpd_list_lock); - - return ret == -ENOENT ? -EPROBE_DEFER : ret; -} -EXPORT_SYMBOL_GPL(of_genpd_add_subdomain); - -/** - * of_genpd_remove_subdomain - Remove a subdomain from an I/O PM domain. - * @parent_spec: OF phandle args to use for parent PM domain look-up - * @subdomain_spec: OF phandle args to use for subdomain look-up - * - * Looks-up a parent PM domain and subdomain based upon phandle args - * provided and removes the subdomain from the parent PM domain. Returns a - * negative error code on failure. - */ -int of_genpd_remove_subdomain(struct of_phandle_args *parent_spec, - struct of_phandle_args *subdomain_spec) -{ - struct generic_pm_domain *parent, *subdomain; - int ret; - - mutex_lock(&gpd_list_lock); - - parent = genpd_get_from_provider(parent_spec); - if (IS_ERR(parent)) { - ret = PTR_ERR(parent); - goto out; - } - - subdomain = genpd_get_from_provider(subdomain_spec); - if (IS_ERR(subdomain)) { - ret = PTR_ERR(subdomain); - goto out; - } - - ret = pm_genpd_remove_subdomain(parent, subdomain); - -out: - mutex_unlock(&gpd_list_lock); - - return ret; -} -EXPORT_SYMBOL_GPL(of_genpd_remove_subdomain); - -/** - * of_genpd_remove_last - Remove the last PM domain registered for a provider - * @np: Pointer to device node associated with provider - * - * Find the last PM domain that was added by a particular provider and - * remove this PM domain from the list of PM domains. The provider is - * identified by the 'provider' device structure that is passed. The PM - * domain will only be removed, if the provider associated with domain - * has been removed. - * - * Returns a valid pointer to struct generic_pm_domain on success or - * ERR_PTR() on failure. - */ -struct generic_pm_domain *of_genpd_remove_last(struct device_node *np) -{ - struct generic_pm_domain *gpd, *tmp, *genpd = ERR_PTR(-ENOENT); - int ret; - - if (IS_ERR_OR_NULL(np)) - return ERR_PTR(-EINVAL); - - mutex_lock(&gpd_list_lock); - list_for_each_entry_safe(gpd, tmp, &gpd_list, gpd_list_node) { - if (gpd->provider == &np->fwnode) { - ret = genpd_remove(gpd); - genpd = ret ? ERR_PTR(ret) : gpd; - break; - } - } - mutex_unlock(&gpd_list_lock); - - return genpd; -} -EXPORT_SYMBOL_GPL(of_genpd_remove_last); - -static void genpd_release_dev(struct device *dev) -{ - of_node_put(dev->of_node); - kfree(dev); -} - -static struct bus_type genpd_bus_type = { - .name = "genpd", -}; - -/** - * genpd_dev_pm_detach - Detach a device from its PM domain. - * @dev: Device to detach. - * @power_off: Currently not used - * - * Try to locate a corresponding generic PM domain, which the device was - * attached to previously. If such is found, the device is detached from it. - */ -static void genpd_dev_pm_detach(struct device *dev, bool power_off) -{ - struct generic_pm_domain *pd; - unsigned int i; - int ret = 0; - - pd = dev_to_genpd(dev); - if (IS_ERR(pd)) - return; - - dev_dbg(dev, "removing from PM domain %s\n", pd->name); - - for (i = 1; i < GENPD_RETRY_MAX_MS; i <<= 1) { - ret = genpd_remove_device(pd, dev); - if (ret != -EAGAIN) - break; - - mdelay(i); - cond_resched(); - } - - if (ret < 0) { - dev_err(dev, "failed to remove from PM domain %s: %d", - pd->name, ret); - return; - } - - /* Check if PM domain can be powered off after removing this device. */ - genpd_queue_power_off_work(pd); - - /* Unregister the device if it was created by genpd. */ - if (dev->bus == &genpd_bus_type) - device_unregister(dev); -} - -static void genpd_dev_pm_sync(struct device *dev) -{ - struct generic_pm_domain *pd; - - pd = dev_to_genpd(dev); - if (IS_ERR(pd)) - return; - - genpd_queue_power_off_work(pd); -} - -static int __genpd_dev_pm_attach(struct device *dev, struct device *base_dev, - unsigned int index, bool power_on) -{ - struct of_phandle_args pd_args; - struct generic_pm_domain *pd; - int ret; - - ret = of_parse_phandle_with_args(dev->of_node, "power-domains", - "#power-domain-cells", index, &pd_args); - if (ret < 0) - return ret; - - mutex_lock(&gpd_list_lock); - pd = genpd_get_from_provider(&pd_args); - of_node_put(pd_args.np); - if (IS_ERR(pd)) { - mutex_unlock(&gpd_list_lock); - dev_dbg(dev, "%s() failed to find PM domain: %ld\n", - __func__, PTR_ERR(pd)); - return driver_deferred_probe_check_state(base_dev); - } - - dev_dbg(dev, "adding to PM domain %s\n", pd->name); - - ret = genpd_add_device(pd, dev, base_dev); - mutex_unlock(&gpd_list_lock); - - if (ret < 0) { - if (ret != -EPROBE_DEFER) - dev_err(dev, "failed to add to PM domain %s: %d", - pd->name, ret); - return ret; - } - - dev->pm_domain->detach = genpd_dev_pm_detach; - dev->pm_domain->sync = genpd_dev_pm_sync; - - if (power_on) { - genpd_lock(pd); - ret = genpd_power_on(pd, 0); - genpd_unlock(pd); - } - - if (ret) - genpd_remove_device(pd, dev); - - return ret ? -EPROBE_DEFER : 1; -} - -/** - * genpd_dev_pm_attach - Attach a device to its PM domain using DT. - * @dev: Device to attach. - * - * Parse device's OF node to find a PM domain specifier. If such is found, - * attaches the device to retrieved pm_domain ops. - * - * Returns 1 on successfully attached PM domain, 0 when the device don't need a - * PM domain or when multiple power-domains exists for it, else a negative error - * code. Note that if a power-domain exists for the device, but it cannot be - * found or turned on, then return -EPROBE_DEFER to ensure that the device is - * not probed and to re-try again later. - */ -int genpd_dev_pm_attach(struct device *dev) -{ - if (!dev->of_node) - return 0; - - /* - * Devices with multiple PM domains must be attached separately, as we - * can only attach one PM domain per device. - */ - if (of_count_phandle_with_args(dev->of_node, "power-domains", - "#power-domain-cells") != 1) - return 0; - - return __genpd_dev_pm_attach(dev, dev, 0, true); -} -EXPORT_SYMBOL_GPL(genpd_dev_pm_attach); - -/** - * genpd_dev_pm_attach_by_id - Associate a device with one of its PM domains. - * @dev: The device used to lookup the PM domain. - * @index: The index of the PM domain. - * - * Parse device's OF node to find a PM domain specifier at the provided @index. - * If such is found, creates a virtual device and attaches it to the retrieved - * pm_domain ops. To deal with detaching of the virtual device, the ->detach() - * callback in the struct dev_pm_domain are assigned to genpd_dev_pm_detach(). - * - * Returns the created virtual device if successfully attached PM domain, NULL - * when the device don't need a PM domain, else an ERR_PTR() in case of - * failures. If a power-domain exists for the device, but cannot be found or - * turned on, then ERR_PTR(-EPROBE_DEFER) is returned to ensure that the device - * is not probed and to re-try again later. - */ -struct device *genpd_dev_pm_attach_by_id(struct device *dev, - unsigned int index) -{ - struct device *virt_dev; - int num_domains; - int ret; - - if (!dev->of_node) - return NULL; - - /* Verify that the index is within a valid range. */ - num_domains = of_count_phandle_with_args(dev->of_node, "power-domains", - "#power-domain-cells"); - if (index >= num_domains) - return NULL; - - /* Allocate and register device on the genpd bus. */ - virt_dev = kzalloc(sizeof(*virt_dev), GFP_KERNEL); - if (!virt_dev) - return ERR_PTR(-ENOMEM); - - dev_set_name(virt_dev, "genpd:%u:%s", index, dev_name(dev)); - virt_dev->bus = &genpd_bus_type; - virt_dev->release = genpd_release_dev; - virt_dev->of_node = of_node_get(dev->of_node); - - ret = device_register(virt_dev); - if (ret) { - put_device(virt_dev); - return ERR_PTR(ret); - } - - /* Try to attach the device to the PM domain at the specified index. */ - ret = __genpd_dev_pm_attach(virt_dev, dev, index, false); - if (ret < 1) { - device_unregister(virt_dev); - return ret ? ERR_PTR(ret) : NULL; - } - - pm_runtime_enable(virt_dev); - genpd_queue_power_off_work(dev_to_genpd(virt_dev)); - - return virt_dev; -} -EXPORT_SYMBOL_GPL(genpd_dev_pm_attach_by_id); - -/** - * genpd_dev_pm_attach_by_name - Associate a device with one of its PM domains. - * @dev: The device used to lookup the PM domain. - * @name: The name of the PM domain. - * - * Parse device's OF node to find a PM domain specifier using the - * power-domain-names DT property. For further description see - * genpd_dev_pm_attach_by_id(). - */ -struct device *genpd_dev_pm_attach_by_name(struct device *dev, const char *name) -{ - int index; - - if (!dev->of_node) - return NULL; - - index = of_property_match_string(dev->of_node, "power-domain-names", - name); - if (index < 0) - return NULL; - - return genpd_dev_pm_attach_by_id(dev, index); -} - -static const struct of_device_id idle_state_match[] = { - { .compatible = "domain-idle-state", }, - { } -}; - -static int genpd_parse_state(struct genpd_power_state *genpd_state, - struct device_node *state_node) -{ - int err; - u32 residency; - u32 entry_latency, exit_latency; - - err = of_property_read_u32(state_node, "entry-latency-us", - &entry_latency); - if (err) { - pr_debug(" * %pOF missing entry-latency-us property\n", - state_node); - return -EINVAL; - } - - err = of_property_read_u32(state_node, "exit-latency-us", - &exit_latency); - if (err) { - pr_debug(" * %pOF missing exit-latency-us property\n", - state_node); - return -EINVAL; - } - - err = of_property_read_u32(state_node, "min-residency-us", &residency); - if (!err) - genpd_state->residency_ns = 1000 * residency; - - genpd_state->power_on_latency_ns = 1000 * exit_latency; - genpd_state->power_off_latency_ns = 1000 * entry_latency; - genpd_state->fwnode = &state_node->fwnode; - - return 0; -} - -static int genpd_iterate_idle_states(struct device_node *dn, - struct genpd_power_state *states) -{ - int ret; - struct of_phandle_iterator it; - struct device_node *np; - int i = 0; - - ret = of_count_phandle_with_args(dn, "domain-idle-states", NULL); - if (ret <= 0) - return ret == -ENOENT ? 0 : ret; - - /* Loop over the phandles until all the requested entry is found */ - of_for_each_phandle(&it, ret, dn, "domain-idle-states", NULL, 0) { - np = it.node; - if (!of_match_node(idle_state_match, np)) - continue; - if (states) { - ret = genpd_parse_state(&states[i], np); - if (ret) { - pr_err("Parsing idle state node %pOF failed with err %d\n", - np, ret); - of_node_put(np); - return ret; - } - } - i++; - } - - return i; -} - -/** - * of_genpd_parse_idle_states: Return array of idle states for the genpd. - * - * @dn: The genpd device node - * @states: The pointer to which the state array will be saved. - * @n: The count of elements in the array returned from this function. - * - * Returns the device states parsed from the OF node. The memory for the states - * is allocated by this function and is the responsibility of the caller to - * free the memory after use. If any or zero compatible domain idle states is - * found it returns 0 and in case of errors, a negative error code is returned. - */ -int of_genpd_parse_idle_states(struct device_node *dn, - struct genpd_power_state **states, int *n) -{ - struct genpd_power_state *st; - int ret; - - ret = genpd_iterate_idle_states(dn, NULL); - if (ret < 0) - return ret; - - if (!ret) { - *states = NULL; - *n = 0; - return 0; - } - - st = kcalloc(ret, sizeof(*st), GFP_KERNEL); - if (!st) - return -ENOMEM; - - ret = genpd_iterate_idle_states(dn, st); - if (ret <= 0) { - kfree(st); - return ret < 0 ? ret : -EINVAL; - } - - *states = st; - *n = ret; - - return 0; -} -EXPORT_SYMBOL_GPL(of_genpd_parse_idle_states); - -/** - * pm_genpd_opp_to_performance_state - Gets performance state of the genpd from its OPP node. - * - * @genpd_dev: Genpd's device for which the performance-state needs to be found. - * @opp: struct dev_pm_opp of the OPP for which we need to find performance - * state. - * - * Returns performance state encoded in the OPP of the genpd. This calls - * platform specific genpd->opp_to_performance_state() callback to translate - * power domain OPP to performance state. - * - * Returns performance state on success and 0 on failure. - */ -unsigned int pm_genpd_opp_to_performance_state(struct device *genpd_dev, - struct dev_pm_opp *opp) -{ - struct generic_pm_domain *genpd = NULL; - int state; - - genpd = container_of(genpd_dev, struct generic_pm_domain, dev); - - if (unlikely(!genpd->opp_to_performance_state)) - return 0; - - genpd_lock(genpd); - state = genpd->opp_to_performance_state(genpd, opp); - genpd_unlock(genpd); - - return state; -} -EXPORT_SYMBOL_GPL(pm_genpd_opp_to_performance_state); - -static int __init genpd_bus_init(void) -{ - return bus_register(&genpd_bus_type); -} -core_initcall(genpd_bus_init); - -#endif /* CONFIG_PM_GENERIC_DOMAINS_OF */ - - -/*** debugfs support ***/ - -#ifdef CONFIG_DEBUG_FS -/* - * TODO: This function is a slightly modified version of rtpm_status_show - * from sysfs.c, so generalize it. - */ -static void rtpm_status_str(struct seq_file *s, struct device *dev) -{ - static const char * const status_lookup[] = { - [RPM_ACTIVE] = "active", - [RPM_RESUMING] = "resuming", - [RPM_SUSPENDED] = "suspended", - [RPM_SUSPENDING] = "suspending" - }; - const char *p = ""; - - if (dev->power.runtime_error) - p = "error"; - else if (dev->power.disable_depth) - p = "unsupported"; - else if (dev->power.runtime_status < ARRAY_SIZE(status_lookup)) - p = status_lookup[dev->power.runtime_status]; - else - WARN_ON(1); - - seq_printf(s, "%-25s ", p); -} - -static void perf_status_str(struct seq_file *s, struct device *dev) -{ - struct generic_pm_domain_data *gpd_data; - - gpd_data = to_gpd_data(dev->power.subsys_data->domain_data); - seq_put_decimal_ull(s, "", gpd_data->performance_state); -} - -static int genpd_summary_one(struct seq_file *s, - struct generic_pm_domain *genpd) -{ - static const char * const status_lookup[] = { - [GENPD_STATE_ON] = "on", - [GENPD_STATE_OFF] = "off" - }; - struct pm_domain_data *pm_data; - const char *kobj_path; - struct gpd_link *link; - char state[16]; - int ret; - - ret = genpd_lock_interruptible(genpd); - if (ret) - return -ERESTARTSYS; - - if (WARN_ON(genpd->status >= ARRAY_SIZE(status_lookup))) - goto exit; - if (!genpd_status_on(genpd)) - snprintf(state, sizeof(state), "%s-%u", - status_lookup[genpd->status], genpd->state_idx); - else - snprintf(state, sizeof(state), "%s", - status_lookup[genpd->status]); - seq_printf(s, "%-30s %-50s %u", genpd->name, state, genpd->performance_state); - - /* - * Modifications on the list require holding locks on both - * parent and child, so we are safe. - * Also genpd->name is immutable. - */ - list_for_each_entry(link, &genpd->parent_links, parent_node) { - if (list_is_first(&link->parent_node, &genpd->parent_links)) - seq_printf(s, "\n%48s", " "); - seq_printf(s, "%s", link->child->name); - if (!list_is_last(&link->parent_node, &genpd->parent_links)) - seq_puts(s, ", "); - } - - list_for_each_entry(pm_data, &genpd->dev_list, list_node) { - kobj_path = kobject_get_path(&pm_data->dev->kobj, - genpd_is_irq_safe(genpd) ? - GFP_ATOMIC : GFP_KERNEL); - if (kobj_path == NULL) - continue; - - seq_printf(s, "\n %-50s ", kobj_path); - rtpm_status_str(s, pm_data->dev); - perf_status_str(s, pm_data->dev); - kfree(kobj_path); - } - - seq_puts(s, "\n"); -exit: - genpd_unlock(genpd); - - return 0; -} - -static int summary_show(struct seq_file *s, void *data) -{ - struct generic_pm_domain *genpd; - int ret = 0; - - seq_puts(s, "domain status children performance\n"); - seq_puts(s, " /device runtime status\n"); - seq_puts(s, "----------------------------------------------------------------------------------------------\n"); - - ret = mutex_lock_interruptible(&gpd_list_lock); - if (ret) - return -ERESTARTSYS; - - list_for_each_entry(genpd, &gpd_list, gpd_list_node) { - ret = genpd_summary_one(s, genpd); - if (ret) - break; - } - mutex_unlock(&gpd_list_lock); - - return ret; -} - -static int status_show(struct seq_file *s, void *data) -{ - static const char * const status_lookup[] = { - [GENPD_STATE_ON] = "on", - [GENPD_STATE_OFF] = "off" - }; - - struct generic_pm_domain *genpd = s->private; - int ret = 0; - - ret = genpd_lock_interruptible(genpd); - if (ret) - return -ERESTARTSYS; - - if (WARN_ON_ONCE(genpd->status >= ARRAY_SIZE(status_lookup))) - goto exit; - - if (genpd->status == GENPD_STATE_OFF) - seq_printf(s, "%s-%u\n", status_lookup[genpd->status], - genpd->state_idx); - else - seq_printf(s, "%s\n", status_lookup[genpd->status]); -exit: - genpd_unlock(genpd); - return ret; -} - -static int sub_domains_show(struct seq_file *s, void *data) -{ - struct generic_pm_domain *genpd = s->private; - struct gpd_link *link; - int ret = 0; - - ret = genpd_lock_interruptible(genpd); - if (ret) - return -ERESTARTSYS; - - list_for_each_entry(link, &genpd->parent_links, parent_node) - seq_printf(s, "%s\n", link->child->name); - - genpd_unlock(genpd); - return ret; -} - -static int idle_states_show(struct seq_file *s, void *data) -{ - struct generic_pm_domain *genpd = s->private; - unsigned int i; - int ret = 0; - - ret = genpd_lock_interruptible(genpd); - if (ret) - return -ERESTARTSYS; - - seq_puts(s, "State Time Spent(ms) Usage Rejected\n"); - - for (i = 0; i < genpd->state_count; i++) { - ktime_t delta = 0; - s64 msecs; - - if ((genpd->status == GENPD_STATE_OFF) && - (genpd->state_idx == i)) - delta = ktime_sub(ktime_get(), genpd->accounting_time); - - msecs = ktime_to_ms( - ktime_add(genpd->states[i].idle_time, delta)); - seq_printf(s, "S%-13i %-14lld %-14llu %llu\n", i, msecs, - genpd->states[i].usage, genpd->states[i].rejected); - } - - genpd_unlock(genpd); - return ret; -} - -static int active_time_show(struct seq_file *s, void *data) -{ - struct generic_pm_domain *genpd = s->private; - ktime_t delta = 0; - int ret = 0; - - ret = genpd_lock_interruptible(genpd); - if (ret) - return -ERESTARTSYS; - - if (genpd->status == GENPD_STATE_ON) - delta = ktime_sub(ktime_get(), genpd->accounting_time); - - seq_printf(s, "%lld ms\n", ktime_to_ms( - ktime_add(genpd->on_time, delta))); - - genpd_unlock(genpd); - return ret; -} - -static int total_idle_time_show(struct seq_file *s, void *data) -{ - struct generic_pm_domain *genpd = s->private; - ktime_t delta = 0, total = 0; - unsigned int i; - int ret = 0; - - ret = genpd_lock_interruptible(genpd); - if (ret) - return -ERESTARTSYS; - - for (i = 0; i < genpd->state_count; i++) { - - if ((genpd->status == GENPD_STATE_OFF) && - (genpd->state_idx == i)) - delta = ktime_sub(ktime_get(), genpd->accounting_time); - - total = ktime_add(total, genpd->states[i].idle_time); - } - total = ktime_add(total, delta); - - seq_printf(s, "%lld ms\n", ktime_to_ms(total)); - - genpd_unlock(genpd); - return ret; -} - - -static int devices_show(struct seq_file *s, void *data) -{ - struct generic_pm_domain *genpd = s->private; - struct pm_domain_data *pm_data; - const char *kobj_path; - int ret = 0; - - ret = genpd_lock_interruptible(genpd); - if (ret) - return -ERESTARTSYS; - - list_for_each_entry(pm_data, &genpd->dev_list, list_node) { - kobj_path = kobject_get_path(&pm_data->dev->kobj, - genpd_is_irq_safe(genpd) ? - GFP_ATOMIC : GFP_KERNEL); - if (kobj_path == NULL) - continue; - - seq_printf(s, "%s\n", kobj_path); - kfree(kobj_path); - } - - genpd_unlock(genpd); - return ret; -} - -static int perf_state_show(struct seq_file *s, void *data) -{ - struct generic_pm_domain *genpd = s->private; - - if (genpd_lock_interruptible(genpd)) - return -ERESTARTSYS; - - seq_printf(s, "%u\n", genpd->performance_state); - - genpd_unlock(genpd); - return 0; -} - -DEFINE_SHOW_ATTRIBUTE(summary); -DEFINE_SHOW_ATTRIBUTE(status); -DEFINE_SHOW_ATTRIBUTE(sub_domains); -DEFINE_SHOW_ATTRIBUTE(idle_states); -DEFINE_SHOW_ATTRIBUTE(active_time); -DEFINE_SHOW_ATTRIBUTE(total_idle_time); -DEFINE_SHOW_ATTRIBUTE(devices); -DEFINE_SHOW_ATTRIBUTE(perf_state); - -static void genpd_debug_add(struct generic_pm_domain *genpd) -{ - struct dentry *d; - - if (!genpd_debugfs_dir) - return; - - d = debugfs_create_dir(genpd->name, genpd_debugfs_dir); - - debugfs_create_file("current_state", 0444, - d, genpd, &status_fops); - debugfs_create_file("sub_domains", 0444, - d, genpd, &sub_domains_fops); - debugfs_create_file("idle_states", 0444, - d, genpd, &idle_states_fops); - debugfs_create_file("active_time", 0444, - d, genpd, &active_time_fops); - debugfs_create_file("total_idle_time", 0444, - d, genpd, &total_idle_time_fops); - debugfs_create_file("devices", 0444, - d, genpd, &devices_fops); - if (genpd->set_performance_state) - debugfs_create_file("perf_state", 0444, - d, genpd, &perf_state_fops); -} - -static int __init genpd_debug_init(void) -{ - struct generic_pm_domain *genpd; - - genpd_debugfs_dir = debugfs_create_dir("pm_genpd", NULL); - - debugfs_create_file("pm_genpd_summary", S_IRUGO, genpd_debugfs_dir, - NULL, &summary_fops); - - list_for_each_entry(genpd, &gpd_list, gpd_list_node) - genpd_debug_add(genpd); - - return 0; -} -late_initcall(genpd_debug_init); - -static void __exit genpd_debug_exit(void) -{ - debugfs_remove_recursive(genpd_debugfs_dir); -} -__exitcall(genpd_debug_exit); -#endif /* CONFIG_DEBUG_FS */ diff --git a/drivers/base/power/domain_governor.c b/drivers/base/power/domain_governor.c deleted file mode 100644 index cd08c5885190..000000000000 --- a/drivers/base/power/domain_governor.c +++ /dev/null @@ -1,406 +0,0 @@ -// SPDX-License-Identifier: GPL-2.0 -/* - * drivers/base/power/domain_governor.c - Governors for device PM domains. - * - * Copyright (C) 2011 Rafael J. Wysocki <rjw@sisk.pl>, Renesas Electronics Corp. - */ -#include <linux/kernel.h> -#include <linux/pm_domain.h> -#include <linux/pm_qos.h> -#include <linux/hrtimer.h> -#include <linux/cpuidle.h> -#include <linux/cpumask.h> -#include <linux/ktime.h> - -static int dev_update_qos_constraint(struct device *dev, void *data) -{ - s64 *constraint_ns_p = data; - s64 constraint_ns; - - if (dev->power.subsys_data && dev->power.subsys_data->domain_data) { - /* - * Only take suspend-time QoS constraints of devices into - * account, because constraints updated after the device has - * been suspended are not guaranteed to be taken into account - * anyway. In order for them to take effect, the device has to - * be resumed and suspended again. - */ - constraint_ns = dev_gpd_data(dev)->td.effective_constraint_ns; - } else { - /* - * The child is not in a domain and there's no info on its - * suspend/resume latencies, so assume them to be negligible and - * take its current PM QoS constraint (that's the only thing - * known at this point anyway). - */ - constraint_ns = dev_pm_qos_read_value(dev, DEV_PM_QOS_RESUME_LATENCY); - constraint_ns *= NSEC_PER_USEC; - } - - if (constraint_ns < *constraint_ns_p) - *constraint_ns_p = constraint_ns; - - return 0; -} - -/** - * default_suspend_ok - Default PM domain governor routine to suspend devices. - * @dev: Device to check. - */ -static bool default_suspend_ok(struct device *dev) -{ - struct gpd_timing_data *td = &dev_gpd_data(dev)->td; - unsigned long flags; - s64 constraint_ns; - - dev_dbg(dev, "%s()\n", __func__); - - spin_lock_irqsave(&dev->power.lock, flags); - - if (!td->constraint_changed) { - bool ret = td->cached_suspend_ok; - - spin_unlock_irqrestore(&dev->power.lock, flags); - return ret; - } - td->constraint_changed = false; - td->cached_suspend_ok = false; - td->effective_constraint_ns = 0; - constraint_ns = __dev_pm_qos_resume_latency(dev); - - spin_unlock_irqrestore(&dev->power.lock, flags); - - if (constraint_ns == 0) - return false; - - constraint_ns *= NSEC_PER_USEC; - /* - * We can walk the children without any additional locking, because - * they all have been suspended at this point and their - * effective_constraint_ns fields won't be modified in parallel with us. - */ - if (!dev->power.ignore_children) - device_for_each_child(dev, &constraint_ns, - dev_update_qos_constraint); - - if (constraint_ns == PM_QOS_RESUME_LATENCY_NO_CONSTRAINT_NS) { - /* "No restriction", so the device is allowed to suspend. */ - td->effective_constraint_ns = PM_QOS_RESUME_LATENCY_NO_CONSTRAINT_NS; - td->cached_suspend_ok = true; - } else if (constraint_ns == 0) { - /* - * This triggers if one of the children that don't belong to a - * domain has a zero PM QoS constraint and it's better not to - * suspend then. effective_constraint_ns is zero already and - * cached_suspend_ok is false, so bail out. - */ - return false; - } else { - constraint_ns -= td->suspend_latency_ns + - td->resume_latency_ns; - /* - * effective_constraint_ns is zero already and cached_suspend_ok - * is false, so if the computed value is not positive, return - * right away. - */ - if (constraint_ns <= 0) - return false; - - td->effective_constraint_ns = constraint_ns; - td->cached_suspend_ok = true; - } - - /* - * The children have been suspended already, so we don't need to take - * their suspend latencies into account here. - */ - return td->cached_suspend_ok; -} - -static void update_domain_next_wakeup(struct generic_pm_domain *genpd, ktime_t now) -{ - ktime_t domain_wakeup = KTIME_MAX; - ktime_t next_wakeup; - struct pm_domain_data *pdd; - struct gpd_link *link; - - if (!(genpd->flags & GENPD_FLAG_MIN_RESIDENCY)) - return; - - /* - * Devices that have a predictable wakeup pattern, may specify - * their next wakeup. Let's find the next wakeup from all the - * devices attached to this domain and from all the sub-domains. - * It is possible that component's a next wakeup may have become - * stale when we read that here. We will ignore to ensure the domain - * is able to enter its optimal idle state. - */ - list_for_each_entry(pdd, &genpd->dev_list, list_node) { - next_wakeup = to_gpd_data(pdd)->next_wakeup; - if (next_wakeup != KTIME_MAX && !ktime_before(next_wakeup, now)) - if (ktime_before(next_wakeup, domain_wakeup)) - domain_wakeup = next_wakeup; - } - - list_for_each_entry(link, &genpd->parent_links, parent_node) { - next_wakeup = link->child->next_wakeup; - if (next_wakeup != KTIME_MAX && !ktime_before(next_wakeup, now)) - if (ktime_before(next_wakeup, domain_wakeup)) - domain_wakeup = next_wakeup; - } - - genpd->next_wakeup = domain_wakeup; -} - -static bool next_wakeup_allows_state(struct generic_pm_domain *genpd, - unsigned int state, ktime_t now) -{ - ktime_t domain_wakeup = genpd->next_wakeup; - s64 idle_time_ns, min_sleep_ns; - - min_sleep_ns = genpd->states[state].power_off_latency_ns + - genpd->states[state].residency_ns; - - idle_time_ns = ktime_to_ns(ktime_sub(domain_wakeup, now)); - - return idle_time_ns >= min_sleep_ns; -} - -static bool __default_power_down_ok(struct dev_pm_domain *pd, - unsigned int state) -{ - struct generic_pm_domain *genpd = pd_to_genpd(pd); - struct gpd_link *link; - struct pm_domain_data *pdd; - s64 min_off_time_ns; - s64 off_on_time_ns; - - off_on_time_ns = genpd->states[state].power_off_latency_ns + - genpd->states[state].power_on_latency_ns; - - min_off_time_ns = -1; - /* - * Check if subdomains can be off for enough time. - * - * All subdomains have been powered off already at this point. - */ - list_for_each_entry(link, &genpd->parent_links, parent_node) { - struct generic_pm_domain *sd = link->child; - s64 sd_max_off_ns = sd->max_off_time_ns; - - if (sd_max_off_ns < 0) - continue; - - /* - * Check if the subdomain is allowed to be off long enough for - * the current domain to turn off and on (that's how much time - * it will have to wait worst case). - */ - if (sd_max_off_ns <= off_on_time_ns) - return false; - - if (min_off_time_ns > sd_max_off_ns || min_off_time_ns < 0) - min_off_time_ns = sd_max_off_ns; - } - - /* - * Check if the devices in the domain can be off enough time. - */ - list_for_each_entry(pdd, &genpd->dev_list, list_node) { - struct gpd_timing_data *td; - s64 constraint_ns; - - /* - * Check if the device is allowed to be off long enough for the - * domain to turn off and on (that's how much time it will - * have to wait worst case). - */ - td = &to_gpd_data(pdd)->td; - constraint_ns = td->effective_constraint_ns; - /* - * Zero means "no suspend at all" and this runs only when all - * devices in the domain are suspended, so it must be positive. - */ - if (constraint_ns == PM_QOS_RESUME_LATENCY_NO_CONSTRAINT_NS) - continue; - - if (constraint_ns <= off_on_time_ns) - return false; - - if (min_off_time_ns > constraint_ns || min_off_time_ns < 0) - min_off_time_ns = constraint_ns; - } - - /* - * If the computed minimum device off time is negative, there are no - * latency constraints, so the domain can spend arbitrary time in the - * "off" state. - */ - if (min_off_time_ns < 0) - return true; - - /* - * The difference between the computed minimum subdomain or device off - * time and the time needed to turn the domain on is the maximum - * theoretical time this domain can spend in the "off" state. - */ - genpd->max_off_time_ns = min_off_time_ns - - genpd->states[state].power_on_latency_ns; - return true; -} - -/** - * _default_power_down_ok - Default generic PM domain power off governor routine. - * @pd: PM domain to check. - * @now: current ktime. - * - * This routine must be executed under the PM domain's lock. - */ -static bool _default_power_down_ok(struct dev_pm_domain *pd, ktime_t now) -{ - struct generic_pm_domain *genpd = pd_to_genpd(pd); - int state_idx = genpd->state_count - 1; - struct gpd_link *link; - - /* - * Find the next wakeup from devices that can determine their own wakeup - * to find when the domain would wakeup and do it for every device down - * the hierarchy. It is not worth while to sleep if the state's residency - * cannot be met. - */ - update_domain_next_wakeup(genpd, now); - if ((genpd->flags & GENPD_FLAG_MIN_RESIDENCY) && (genpd->next_wakeup != KTIME_MAX)) { - /* Let's find out the deepest domain idle state, the devices prefer */ - while (state_idx >= 0) { - if (next_wakeup_allows_state(genpd, state_idx, now)) { - genpd->max_off_time_changed = true; - break; - } - state_idx--; - } - - if (state_idx < 0) { - state_idx = 0; - genpd->cached_power_down_ok = false; - goto done; - } - } - - if (!genpd->max_off_time_changed) { - genpd->state_idx = genpd->cached_power_down_state_idx; - return genpd->cached_power_down_ok; - } - - /* - * We have to invalidate the cached results for the parents, so - * use the observation that default_power_down_ok() is not - * going to be called for any parent until this instance - * returns. - */ - list_for_each_entry(link, &genpd->child_links, child_node) - link->parent->max_off_time_changed = true; - - genpd->max_off_time_ns = -1; - genpd->max_off_time_changed = false; - genpd->cached_power_down_ok = true; - - /* - * Find a state to power down to, starting from the state - * determined by the next wakeup. - */ - while (!__default_power_down_ok(pd, state_idx)) { - if (state_idx == 0) { - genpd->cached_power_down_ok = false; - break; - } - state_idx--; - } - -done: - genpd->state_idx = state_idx; - genpd->cached_power_down_state_idx = genpd->state_idx; - return genpd->cached_power_down_ok; -} - -static bool default_power_down_ok(struct dev_pm_domain *pd) -{ - return _default_power_down_ok(pd, ktime_get()); -} - -static bool always_on_power_down_ok(struct dev_pm_domain *domain) -{ - return false; -} - -#ifdef CONFIG_CPU_IDLE -static bool cpu_power_down_ok(struct dev_pm_domain *pd) -{ - struct generic_pm_domain *genpd = pd_to_genpd(pd); - struct cpuidle_device *dev; - ktime_t domain_wakeup, next_hrtimer; - ktime_t now = ktime_get(); - s64 idle_duration_ns; - int cpu, i; - - /* Validate dev PM QoS constraints. */ - if (!_default_power_down_ok(pd, now)) - return false; - - if (!(genpd->flags & GENPD_FLAG_CPU_DOMAIN)) - return true; - - /* - * Find the next wakeup for any of the online CPUs within the PM domain - * and its subdomains. Note, we only need the genpd->cpus, as it already - * contains a mask of all CPUs from subdomains. - */ - domain_wakeup = ktime_set(KTIME_SEC_MAX, 0); - for_each_cpu_and(cpu, genpd->cpus, cpu_online_mask) { - dev = per_cpu(cpuidle_devices, cpu); - if (dev) { - next_hrtimer = READ_ONCE(dev->next_hrtimer); - if (ktime_before(next_hrtimer, domain_wakeup)) - domain_wakeup = next_hrtimer; - } - } - - /* The minimum idle duration is from now - until the next wakeup. */ - idle_duration_ns = ktime_to_ns(ktime_sub(domain_wakeup, now)); - if (idle_duration_ns <= 0) - return false; - - /* - * Find the deepest idle state that has its residency value satisfied - * and by also taking into account the power off latency for the state. - * Start at the state picked by the dev PM QoS constraint validation. - */ - i = genpd->state_idx; - do { - if (idle_duration_ns >= (genpd->states[i].residency_ns + - genpd->states[i].power_off_latency_ns)) { - genpd->state_idx = i; - return true; - } - } while (--i >= 0); - - return false; -} - -struct dev_power_governor pm_domain_cpu_gov = { - .suspend_ok = default_suspend_ok, - .power_down_ok = cpu_power_down_ok, -}; -#endif - -struct dev_power_governor simple_qos_governor = { - .suspend_ok = default_suspend_ok, - .power_down_ok = default_power_down_ok, -}; - -/** - * pm_genpd_gov_always_on - A governor implementing an always-on policy - */ -struct dev_power_governor pm_domain_always_on_gov = { - .power_down_ok = always_on_power_down_ok, - .suspend_ok = default_suspend_ok, -}; diff --git a/drivers/base/power/generic_ops.c b/drivers/base/power/generic_ops.c index 4fa525668cb7..af99bbcf281c 100644 --- a/drivers/base/power/generic_ops.c +++ b/drivers/base/power/generic_ops.c @@ -8,6 +8,13 @@ #include <linux/pm_runtime.h> #include <linux/export.h> +#define CALL_PM_OP(dev, op) \ +({ \ + struct device *_dev = (dev); \ + const struct dev_pm_ops *pm = _dev->driver ? _dev->driver->pm : NULL; \ + pm && pm->op ? pm->op(_dev) : 0; \ +}) + #ifdef CONFIG_PM /** * pm_generic_runtime_suspend - Generic runtime suspend callback for subsystems. @@ -19,12 +26,7 @@ */ int pm_generic_runtime_suspend(struct device *dev) { - const struct dev_pm_ops *pm = dev->driver ? dev->driver->pm : NULL; - int ret; - - ret = pm && pm->runtime_suspend ? pm->runtime_suspend(dev) : 0; - - return ret; + return CALL_PM_OP(dev, runtime_suspend); } EXPORT_SYMBOL_GPL(pm_generic_runtime_suspend); @@ -38,12 +40,7 @@ EXPORT_SYMBOL_GPL(pm_generic_runtime_suspend); */ int pm_generic_runtime_resume(struct device *dev) { - const struct dev_pm_ops *pm = dev->driver ? dev->driver->pm : NULL; - int ret; - - ret = pm && pm->runtime_resume ? pm->runtime_resume(dev) : 0; - - return ret; + return CALL_PM_OP(dev, runtime_resume); } EXPORT_SYMBOL_GPL(pm_generic_runtime_resume); #endif /* CONFIG_PM */ @@ -72,9 +69,7 @@ int pm_generic_prepare(struct device *dev) */ int pm_generic_suspend_noirq(struct device *dev) { - const struct dev_pm_ops *pm = dev->driver ? dev->driver->pm : NULL; - - return pm && pm->suspend_noirq ? pm->suspend_noirq(dev) : 0; + return CALL_PM_OP(dev, suspend_noirq); } EXPORT_SYMBOL_GPL(pm_generic_suspend_noirq); @@ -84,9 +79,7 @@ EXPORT_SYMBOL_GPL(pm_generic_suspend_noirq); */ int pm_generic_suspend_late(struct device *dev) { - const struct dev_pm_ops *pm = dev->driver ? dev->driver->pm : NULL; - - return pm && pm->suspend_late ? pm->suspend_late(dev) : 0; + return CALL_PM_OP(dev, suspend_late); } EXPORT_SYMBOL_GPL(pm_generic_suspend_late); @@ -96,9 +89,7 @@ EXPORT_SYMBOL_GPL(pm_generic_suspend_late); */ int pm_generic_suspend(struct device *dev) { - const struct dev_pm_ops *pm = dev->driver ? dev->driver->pm : NULL; - - return pm && pm->suspend ? pm->suspend(dev) : 0; + return CALL_PM_OP(dev, suspend); } EXPORT_SYMBOL_GPL(pm_generic_suspend); @@ -108,33 +99,17 @@ EXPORT_SYMBOL_GPL(pm_generic_suspend); */ int pm_generic_freeze_noirq(struct device *dev) { - const struct dev_pm_ops *pm = dev->driver ? dev->driver->pm : NULL; - - return pm && pm->freeze_noirq ? pm->freeze_noirq(dev) : 0; + return CALL_PM_OP(dev, freeze_noirq); } EXPORT_SYMBOL_GPL(pm_generic_freeze_noirq); /** - * pm_generic_freeze_late - Generic freeze_late callback for subsystems. - * @dev: Device to freeze. - */ -int pm_generic_freeze_late(struct device *dev) -{ - const struct dev_pm_ops *pm = dev->driver ? dev->driver->pm : NULL; - - return pm && pm->freeze_late ? pm->freeze_late(dev) : 0; -} -EXPORT_SYMBOL_GPL(pm_generic_freeze_late); - -/** * pm_generic_freeze - Generic freeze callback for subsystems. * @dev: Device to freeze. */ int pm_generic_freeze(struct device *dev) { - const struct dev_pm_ops *pm = dev->driver ? dev->driver->pm : NULL; - - return pm && pm->freeze ? pm->freeze(dev) : 0; + return CALL_PM_OP(dev, freeze); } EXPORT_SYMBOL_GPL(pm_generic_freeze); @@ -144,9 +119,7 @@ EXPORT_SYMBOL_GPL(pm_generic_freeze); */ int pm_generic_poweroff_noirq(struct device *dev) { - const struct dev_pm_ops *pm = dev->driver ? dev->driver->pm : NULL; - - return pm && pm->poweroff_noirq ? pm->poweroff_noirq(dev) : 0; + return CALL_PM_OP(dev, poweroff_noirq); } EXPORT_SYMBOL_GPL(pm_generic_poweroff_noirq); @@ -156,9 +129,7 @@ EXPORT_SYMBOL_GPL(pm_generic_poweroff_noirq); */ int pm_generic_poweroff_late(struct device *dev) { - const struct dev_pm_ops *pm = dev->driver ? dev->driver->pm : NULL; - - return pm && pm->poweroff_late ? pm->poweroff_late(dev) : 0; + return CALL_PM_OP(dev, poweroff_late); } EXPORT_SYMBOL_GPL(pm_generic_poweroff_late); @@ -168,9 +139,7 @@ EXPORT_SYMBOL_GPL(pm_generic_poweroff_late); */ int pm_generic_poweroff(struct device *dev) { - const struct dev_pm_ops *pm = dev->driver ? dev->driver->pm : NULL; - - return pm && pm->poweroff ? pm->poweroff(dev) : 0; + return CALL_PM_OP(dev, poweroff); } EXPORT_SYMBOL_GPL(pm_generic_poweroff); @@ -180,33 +149,17 @@ EXPORT_SYMBOL_GPL(pm_generic_poweroff); */ int pm_generic_thaw_noirq(struct device *dev) { - const struct dev_pm_ops *pm = dev->driver ? dev->driver->pm : NULL; - - return pm && pm->thaw_noirq ? pm->thaw_noirq(dev) : 0; + return CALL_PM_OP(dev, thaw_noirq); } EXPORT_SYMBOL_GPL(pm_generic_thaw_noirq); /** - * pm_generic_thaw_early - Generic thaw_early callback for subsystems. - * @dev: Device to thaw. - */ -int pm_generic_thaw_early(struct device *dev) -{ - const struct dev_pm_ops *pm = dev->driver ? dev->driver->pm : NULL; - - return pm && pm->thaw_early ? pm->thaw_early(dev) : 0; -} -EXPORT_SYMBOL_GPL(pm_generic_thaw_early); - -/** * pm_generic_thaw - Generic thaw callback for subsystems. * @dev: Device to thaw. */ int pm_generic_thaw(struct device *dev) { - const struct dev_pm_ops *pm = dev->driver ? dev->driver->pm : NULL; - - return pm && pm->thaw ? pm->thaw(dev) : 0; + return CALL_PM_OP(dev, thaw); } EXPORT_SYMBOL_GPL(pm_generic_thaw); @@ -216,9 +169,7 @@ EXPORT_SYMBOL_GPL(pm_generic_thaw); */ int pm_generic_resume_noirq(struct device *dev) { - const struct dev_pm_ops *pm = dev->driver ? dev->driver->pm : NULL; - - return pm && pm->resume_noirq ? pm->resume_noirq(dev) : 0; + return CALL_PM_OP(dev, resume_noirq); } EXPORT_SYMBOL_GPL(pm_generic_resume_noirq); @@ -228,9 +179,7 @@ EXPORT_SYMBOL_GPL(pm_generic_resume_noirq); */ int pm_generic_resume_early(struct device *dev) { - const struct dev_pm_ops *pm = dev->driver ? dev->driver->pm : NULL; - - return pm && pm->resume_early ? pm->resume_early(dev) : 0; + return CALL_PM_OP(dev, resume_early); } EXPORT_SYMBOL_GPL(pm_generic_resume_early); @@ -240,9 +189,7 @@ EXPORT_SYMBOL_GPL(pm_generic_resume_early); */ int pm_generic_resume(struct device *dev) { - const struct dev_pm_ops *pm = dev->driver ? dev->driver->pm : NULL; - - return pm && pm->resume ? pm->resume(dev) : 0; + return CALL_PM_OP(dev, resume); } EXPORT_SYMBOL_GPL(pm_generic_resume); @@ -252,9 +199,7 @@ EXPORT_SYMBOL_GPL(pm_generic_resume); */ int pm_generic_restore_noirq(struct device *dev) { - const struct dev_pm_ops *pm = dev->driver ? dev->driver->pm : NULL; - - return pm && pm->restore_noirq ? pm->restore_noirq(dev) : 0; + return CALL_PM_OP(dev, restore_noirq); } EXPORT_SYMBOL_GPL(pm_generic_restore_noirq); @@ -264,9 +209,7 @@ EXPORT_SYMBOL_GPL(pm_generic_restore_noirq); */ int pm_generic_restore_early(struct device *dev) { - const struct dev_pm_ops *pm = dev->driver ? dev->driver->pm : NULL; - - return pm && pm->restore_early ? pm->restore_early(dev) : 0; + return CALL_PM_OP(dev, restore_early); } EXPORT_SYMBOL_GPL(pm_generic_restore_early); @@ -276,9 +219,7 @@ EXPORT_SYMBOL_GPL(pm_generic_restore_early); */ int pm_generic_restore(struct device *dev) { - const struct dev_pm_ops *pm = dev->driver ? dev->driver->pm : NULL; - - return pm && pm->restore ? pm->restore(dev) : 0; + return CALL_PM_OP(dev, restore); } EXPORT_SYMBOL_GPL(pm_generic_restore); diff --git a/drivers/base/power/main.c b/drivers/base/power/main.c index d568772152c2..97a8b4fcf471 100644 --- a/drivers/base/power/main.c +++ b/drivers/base/power/main.c @@ -32,19 +32,15 @@ #include <linux/suspend.h> #include <trace/events/power.h> #include <linux/cpufreq.h> -#include <linux/cpuidle.h> #include <linux/devfreq.h> #include <linux/timer.h> +#include <linux/nmi.h> #include "../base.h" #include "power.h" typedef int (*pm_callback_t)(struct device *); -#define list_for_each_entry_rcu_locked(pos, head, member) \ - list_for_each_entry_rcu(pos, head, member, \ - device_links_read_lock_held()) - /* * The entries in the dpm_list list are in a depth first order, simply * because children are guaranteed to be discovered after parents, and @@ -61,12 +57,26 @@ static LIST_HEAD(dpm_suspended_list); static LIST_HEAD(dpm_late_early_list); static LIST_HEAD(dpm_noirq_list); -struct suspend_stats suspend_stats; static DEFINE_MUTEX(dpm_list_mtx); static pm_message_t pm_transition; +static DEFINE_MUTEX(async_wip_mtx); static int async_error; +/** + * pm_hibernate_is_recovering - if recovering from hibernate due to error. + * + * Used to query if dev_pm_ops.thaw() is called for normal hibernation case or + * recovering from some error. + * + * Return: true for error case, false for normal case. + */ +bool pm_hibernate_is_recovering(void) +{ + return pm_transition.event == PM_EVENT_RECOVER; +} +EXPORT_SYMBOL_GPL(pm_hibernate_is_recovering); + static const char *pm_verb(int event) { switch (event) { @@ -86,6 +96,8 @@ static const char *pm_verb(int event) return "restore"; case PM_EVENT_RECOVER: return "recover"; + case PM_EVENT_POWEROFF: + return "poweroff"; default: return "(unknown PM event)"; } @@ -210,7 +222,7 @@ static ktime_t initcall_debug_start(struct device *dev, void *cb) if (!pm_print_times_enabled) return 0; - dev_info(dev, "calling %pS @ %i, parent: %s\n", cb, + dev_info(dev, "calling %ps @ %i, parent: %s\n", cb, task_pid_nr(current), dev->parent ? dev_name(dev->parent) : "none"); return ktime_get(); @@ -225,7 +237,7 @@ static void initcall_debug_report(struct device *dev, ktime_t calltime, return; rettime = ktime_get(); - dev_info(dev, "%pS returned %d after %Ld usecs\n", cb, error, + dev_info(dev, "%ps returned %d after %Ld usecs\n", cb, error, (unsigned long long)ktime_us_delta(rettime, calltime)); } @@ -251,7 +263,7 @@ static int dpm_wait_fn(struct device *dev, void *async_ptr) static void dpm_wait_for_children(struct device *dev, bool async) { - device_for_each_child(dev, &async, dpm_wait_fn); + device_for_each_child(dev, &async, dpm_wait_fn); } static void dpm_wait_for_suppliers(struct device *dev, bool async) @@ -268,8 +280,9 @@ static void dpm_wait_for_suppliers(struct device *dev, bool async) * callbacks freeing the link objects for the links in the list we're * walking. */ - list_for_each_entry_rcu_locked(link, &dev->links.suppliers, c_node) - if (READ_ONCE(link->status) != DL_STATE_DORMANT) + dev_for_each_link_to_supplier(link, dev) + if (READ_ONCE(link->status) != DL_STATE_DORMANT && + !device_link_flag_is_sync_state_only(link->flags)) dpm_wait(link->supplier, async); device_links_read_unlock(idx); @@ -325,8 +338,9 @@ static void dpm_wait_for_consumers(struct device *dev, bool async) * continue instead of trying to continue in parallel with its * unregistration). */ - list_for_each_entry_rcu_locked(link, &dev->links.consumers, s_node) - if (READ_ONCE(link->status) != DL_STATE_DORMANT) + dev_for_each_link_to_consumer(link, dev) + if (READ_ONCE(link->status) != DL_STATE_DORMANT && + !device_link_flag_is_sync_state_only(link->flags)) dpm_wait(link->consumer, async); device_links_read_unlock(idx); @@ -356,6 +370,7 @@ static pm_callback_t pm_op(const struct dev_pm_ops *ops, pm_message_t state) case PM_EVENT_FREEZE: case PM_EVENT_QUIESCE: return ops->freeze; + case PM_EVENT_POWEROFF: case PM_EVENT_HIBERNATE: return ops->poweroff; case PM_EVENT_THAW: @@ -390,6 +405,7 @@ static pm_callback_t pm_late_early_op(const struct dev_pm_ops *ops, case PM_EVENT_FREEZE: case PM_EVENT_QUIESCE: return ops->freeze_late; + case PM_EVENT_POWEROFF: case PM_EVENT_HIBERNATE: return ops->poweroff_late; case PM_EVENT_THAW: @@ -424,6 +440,7 @@ static pm_callback_t pm_noirq_op(const struct dev_pm_ops *ops, pm_message_t stat case PM_EVENT_FREEZE: case PM_EVENT_QUIESCE: return ops->freeze_noirq; + case PM_EVENT_POWEROFF: case PM_EVENT_HIBERNATE: return ops->poweroff_noirq; case PM_EVENT_THAW: @@ -486,7 +503,7 @@ static int dpm_run_callback(pm_callback_t cb, struct device *dev, trace_device_pm_callback_start(dev, info, state.event); error = cb(dev); trace_device_pm_callback_end(dev, error); - suspend_report_result(cb, error); + suspend_report_result(dev, cb, error); initcall_debug_report(dev, calltime, cb, error); @@ -498,11 +515,17 @@ struct dpm_watchdog { struct device *dev; struct task_struct *tsk; struct timer_list timer; + bool fatal; }; #define DECLARE_DPM_WATCHDOG_ON_STACK(wd) \ struct dpm_watchdog wd +static bool __read_mostly dpm_watchdog_all_cpu_backtrace; +module_param(dpm_watchdog_all_cpu_backtrace, bool, 0644); +MODULE_PARM_DESC(dpm_watchdog_all_cpu_backtrace, + "Backtrace all CPUs on DPM watchdog timeout"); + /** * dpm_watchdog_handler - Driver suspend / resume watchdog handler. * @t: The timer that PM watchdog depends on. @@ -513,12 +536,28 @@ struct dpm_watchdog { */ static void dpm_watchdog_handler(struct timer_list *t) { - struct dpm_watchdog *wd = from_timer(wd, t, timer); + struct dpm_watchdog *wd = timer_container_of(wd, t, timer); + struct timer_list *timer = &wd->timer; + unsigned int time_left; + + if (wd->fatal) { + unsigned int this_cpu = smp_processor_id(); - dev_emerg(wd->dev, "**** DPM device timeout ****\n"); - show_stack(wd->tsk, NULL, KERN_EMERG); - panic("%s %s: unrecoverable failure\n", - dev_driver_string(wd->dev), dev_name(wd->dev)); + dev_emerg(wd->dev, "**** DPM device timeout ****\n"); + show_stack(wd->tsk, NULL, KERN_EMERG); + if (dpm_watchdog_all_cpu_backtrace) + trigger_allbutcpu_cpu_backtrace(this_cpu); + panic("%s %s: unrecoverable failure\n", + dev_driver_string(wd->dev), dev_name(wd->dev)); + } + + time_left = CONFIG_DPM_WATCHDOG_TIMEOUT - CONFIG_DPM_WATCHDOG_WARNING_TIMEOUT; + dev_warn(wd->dev, "**** DPM device timeout after %u seconds; %u seconds until panic ****\n", + CONFIG_DPM_WATCHDOG_WARNING_TIMEOUT, time_left); + show_stack(wd->tsk, NULL, KERN_WARNING); + + wd->fatal = true; + mod_timer(timer, jiffies + HZ * time_left); } /** @@ -532,10 +571,11 @@ static void dpm_watchdog_set(struct dpm_watchdog *wd, struct device *dev) wd->dev = dev; wd->tsk = current; + wd->fatal = CONFIG_DPM_WATCHDOG_TIMEOUT == CONFIG_DPM_WATCHDOG_WARNING_TIMEOUT; timer_setup_on_stack(timer, dpm_watchdog_handler, 0); /* use same timeout value for both suspend and resume */ - timer->expires = jiffies + HZ * CONFIG_DPM_WATCHDOG_TIMEOUT; + timer->expires = jiffies + HZ * CONFIG_DPM_WATCHDOG_WARNING_TIMEOUT; add_timer(timer); } @@ -547,8 +587,8 @@ static void dpm_watchdog_clear(struct dpm_watchdog *wd) { struct timer_list *timer = &wd->timer; - del_timer_sync(timer); - destroy_timer_on_stack(timer); + timer_delete_sync(timer); + timer_destroy_on_stack(timer); } #else #define DECLARE_DPM_WATCHDOG_ON_STACK(wd) @@ -579,6 +619,102 @@ bool dev_pm_skip_resume(struct device *dev) return !dev->power.must_resume; } +static bool is_async(struct device *dev) +{ + return dev->power.async_suspend && pm_async_enabled + && !pm_trace_is_enabled(); +} + +static bool __dpm_async(struct device *dev, async_func_t func) +{ + if (dev->power.work_in_progress) + return true; + + if (!is_async(dev)) + return false; + + dev->power.work_in_progress = true; + + get_device(dev); + + if (async_schedule_dev_nocall(func, dev)) + return true; + + put_device(dev); + + return false; +} + +static bool dpm_async_fn(struct device *dev, async_func_t func) +{ + guard(mutex)(&async_wip_mtx); + + return __dpm_async(dev, func); +} + +static int dpm_async_with_cleanup(struct device *dev, void *fn) +{ + guard(mutex)(&async_wip_mtx); + + if (!__dpm_async(dev, fn)) + dev->power.work_in_progress = false; + + return 0; +} + +static void dpm_async_resume_children(struct device *dev, async_func_t func) +{ + /* + * Prevent racing with dpm_clear_async_state() during initial list + * walks in dpm_noirq_resume_devices(), dpm_resume_early(), and + * dpm_resume(). + */ + guard(mutex)(&dpm_list_mtx); + + /* + * Start processing "async" children of the device unless it's been + * started already for them. + */ + device_for_each_child(dev, func, dpm_async_with_cleanup); +} + +static void dpm_async_resume_subordinate(struct device *dev, async_func_t func) +{ + struct device_link *link; + int idx; + + dpm_async_resume_children(dev, func); + + idx = device_links_read_lock(); + + /* Start processing the device's "async" consumers. */ + dev_for_each_link_to_consumer(link, dev) + if (READ_ONCE(link->status) != DL_STATE_DORMANT) + dpm_async_with_cleanup(link->consumer, func); + + device_links_read_unlock(idx); +} + +static void dpm_clear_async_state(struct device *dev) +{ + reinit_completion(&dev->power.completion); + dev->power.work_in_progress = false; +} + +static bool dpm_root_device(struct device *dev) +{ + lockdep_assert_held(&dpm_list_mtx); + + /* + * Since this function is required to run under dpm_list_mtx, the + * list_empty() below will only return true if the device's list of + * consumers is actually empty before calling it. + */ + return !dev->parent && list_empty(&dev->links.suppliers); +} + +static void async_resume_noirq(void *data, async_cookie_t cookie); + /** * device_resume_noirq - Execute a "noirq resume" callback for given device. * @dev: Device to handle. @@ -588,7 +724,7 @@ bool dev_pm_skip_resume(struct device *dev) * The driver of @dev will not receive interrupts while this function is being * executed. */ -static int device_resume_noirq(struct device *dev, pm_message_t state, bool async) +static void device_resume_noirq(struct device *dev, pm_message_t state, bool async) { pm_callback_t callback = NULL; const char *info = NULL; @@ -601,8 +737,20 @@ static int device_resume_noirq(struct device *dev, pm_message_t state, bool asyn if (dev->power.syscore || dev->power.direct_complete) goto Out; - if (!dev->power.is_noirq_suspended) + if (!dev->power.is_noirq_suspended) { + /* + * This means that system suspend has been aborted in the noirq + * phase before invoking the noirq suspend callback for the + * device, so if device_suspend_late() has left it in suspend, + * device_resume_early() should leave it in suspend either in + * case the early resume of it depends on the noirq resume that + * has not run. + */ + if (dev_pm_skip_suspend(dev)) + dev->power.must_resume = false; + goto Out; + } if (!dpm_wait_for_superior(dev, async)) goto Out; @@ -615,12 +763,12 @@ static int device_resume_noirq(struct device *dev, pm_message_t state, bool asyn * so change its status accordingly. * * Otherwise, the device is going to be resumed, so set its PM-runtime - * status to "active", but do that only if DPM_FLAG_SMART_SUSPEND is set - * to avoid confusing drivers that don't use it. + * status to "active" unless its power.smart_suspend flag is clear, in + * which case it is not necessary to update its PM-runtime status. */ if (skip_resume) pm_runtime_set_suspended(dev); - else if (dev_pm_skip_suspend(dev)) + else if (dev_pm_smart_suspend(dev)) pm_runtime_set_active(dev); if (dev->pm_domain) { @@ -656,37 +804,21 @@ Skip: Out: complete_all(&dev->power.completion); TRACE_RESUME(error); - return error; -} - -static bool is_async(struct device *dev) -{ - return dev->power.async_suspend && pm_async_enabled - && !pm_trace_is_enabled(); -} -static bool dpm_async_fn(struct device *dev, async_func_t func) -{ - reinit_completion(&dev->power.completion); - - if (is_async(dev)) { - get_device(dev); - async_schedule_dev(func, dev); - return true; + if (error) { + WRITE_ONCE(async_error, error); + dpm_save_failed_dev(dev_name(dev)); + pm_dev_err(dev, state, async ? " async noirq" : " noirq", error); } - return false; + dpm_async_resume_subordinate(dev, async_resume_noirq); } static void async_resume_noirq(void *data, async_cookie_t cookie) { - struct device *dev = (struct device *)data; - int error; - - error = device_resume_noirq(dev, pm_transition, true); - if (error) - pm_dev_err(dev, pm_transition, " async", error); + struct device *dev = data; + device_resume_noirq(dev, pm_transition, true); put_device(dev); } @@ -696,41 +828,44 @@ static void dpm_noirq_resume_devices(pm_message_t state) ktime_t starttime = ktime_get(); trace_suspend_resume(TPS("dpm_resume_noirq"), state.event, true); - mutex_lock(&dpm_list_mtx); + + async_error = 0; pm_transition = state; + mutex_lock(&dpm_list_mtx); + /* - * Advanced the async threads upfront, - * in case the starting of async threads is - * delayed by non-async resuming devices. + * Start processing "async" root devices upfront so they don't wait for + * the "sync" devices they don't depend on. */ - list_for_each_entry(dev, &dpm_noirq_list, power.entry) - dpm_async_fn(dev, async_resume_noirq); + list_for_each_entry(dev, &dpm_noirq_list, power.entry) { + dpm_clear_async_state(dev); + if (dpm_root_device(dev)) + dpm_async_with_cleanup(dev, async_resume_noirq); + } while (!list_empty(&dpm_noirq_list)) { dev = to_device(dpm_noirq_list.next); - get_device(dev); list_move_tail(&dev->power.entry, &dpm_late_early_list); - mutex_unlock(&dpm_list_mtx); - if (!is_async(dev)) { - int error; + if (!dpm_async_fn(dev, async_resume_noirq)) { + get_device(dev); - error = device_resume_noirq(dev, state, false); - if (error) { - suspend_stats.failed_resume_noirq++; - dpm_save_failed_step(SUSPEND_RESUME_NOIRQ); - dpm_save_failed_dev(dev_name(dev)); - pm_dev_err(dev, state, " noirq", error); - } - } + mutex_unlock(&dpm_list_mtx); - mutex_lock(&dpm_list_mtx); - put_device(dev); + device_resume_noirq(dev, state, false); + + put_device(dev); + + mutex_lock(&dpm_list_mtx); + } } mutex_unlock(&dpm_list_mtx); async_synchronize_full(); dpm_show_time(starttime, state, 0, "noirq"); + if (READ_ONCE(async_error)) + dpm_save_failed_step(SUSPEND_RESUME_NOIRQ); + trace_suspend_resume(TPS("dpm_resume_noirq"), state.event, false); } @@ -747,10 +882,10 @@ void dpm_resume_noirq(pm_message_t state) resume_device_irqs(); device_wakeup_disarm_wake_irqs(); - - cpuidle_resume(); } +static void async_resume_early(void *data, async_cookie_t cookie); + /** * device_resume_early - Execute an "early resume" callback for given device. * @dev: Device to handle. @@ -759,7 +894,7 @@ void dpm_resume_noirq(pm_message_t state) * * Runtime PM is disabled for @dev while this function is being executed. */ -static int device_resume_early(struct device *dev, pm_message_t state, bool async) +static void device_resume_early(struct device *dev, pm_message_t state, bool async) { pm_callback_t callback = NULL; const char *info = NULL; @@ -768,12 +903,15 @@ static int device_resume_early(struct device *dev, pm_message_t state, bool asyn TRACE_DEVICE(dev); TRACE_RESUME(0); - if (dev->power.syscore || dev->power.direct_complete) + if (dev->power.direct_complete) goto Out; if (!dev->power.is_late_suspended) goto Out; + if (dev->power.syscore) + goto Skip; + if (!dpm_wait_for_superior(dev, async)) goto Out; @@ -806,24 +944,27 @@ Run: Skip: dev->power.is_late_suspended = false; + pm_runtime_enable(dev); Out: TRACE_RESUME(error); - pm_runtime_enable(dev); complete_all(&dev->power.completion); - return error; + + if (error) { + WRITE_ONCE(async_error, error); + dpm_save_failed_dev(dev_name(dev)); + pm_dev_err(dev, state, async ? " async early" : " early", error); + } + + dpm_async_resume_subordinate(dev, async_resume_early); } static void async_resume_early(void *data, async_cookie_t cookie) { - struct device *dev = (struct device *)data; - int error; - - error = device_resume_early(dev, pm_transition, true); - if (error) - pm_dev_err(dev, pm_transition, " async", error); + struct device *dev = data; + device_resume_early(dev, pm_transition, true); put_device(dev); } @@ -837,40 +978,44 @@ void dpm_resume_early(pm_message_t state) ktime_t starttime = ktime_get(); trace_suspend_resume(TPS("dpm_resume_early"), state.event, true); - mutex_lock(&dpm_list_mtx); + + async_error = 0; pm_transition = state; + mutex_lock(&dpm_list_mtx); + /* - * Advanced the async threads upfront, - * in case the starting of async threads is - * delayed by non-async resuming devices. + * Start processing "async" root devices upfront so they don't wait for + * the "sync" devices they don't depend on. */ - list_for_each_entry(dev, &dpm_late_early_list, power.entry) - dpm_async_fn(dev, async_resume_early); + list_for_each_entry(dev, &dpm_late_early_list, power.entry) { + dpm_clear_async_state(dev); + if (dpm_root_device(dev)) + dpm_async_with_cleanup(dev, async_resume_early); + } while (!list_empty(&dpm_late_early_list)) { dev = to_device(dpm_late_early_list.next); - get_device(dev); list_move_tail(&dev->power.entry, &dpm_suspended_list); - mutex_unlock(&dpm_list_mtx); - if (!is_async(dev)) { - int error; + if (!dpm_async_fn(dev, async_resume_early)) { + get_device(dev); - error = device_resume_early(dev, state, false); - if (error) { - suspend_stats.failed_resume_early++; - dpm_save_failed_step(SUSPEND_RESUME_EARLY); - dpm_save_failed_dev(dev_name(dev)); - pm_dev_err(dev, state, " early", error); - } + mutex_unlock(&dpm_list_mtx); + + device_resume_early(dev, state, false); + + put_device(dev); + + mutex_lock(&dpm_list_mtx); } - mutex_lock(&dpm_list_mtx); - put_device(dev); } mutex_unlock(&dpm_list_mtx); async_synchronize_full(); dpm_show_time(starttime, state, 0, "early"); + if (READ_ONCE(async_error)) + dpm_save_failed_step(SUSPEND_RESUME_EARLY); + trace_suspend_resume(TPS("dpm_resume_early"), state.event, false); } @@ -885,13 +1030,15 @@ void dpm_resume_start(pm_message_t state) } EXPORT_SYMBOL_GPL(dpm_resume_start); +static void async_resume(void *data, async_cookie_t cookie); + /** * device_resume - Execute "resume" callbacks for given device. * @dev: Device to handle. * @state: PM transition of the system being carried out. * @async: If true, the device is being resumed asynchronously. */ -static int device_resume(struct device *dev, pm_message_t state, bool async) +static void device_resume(struct device *dev, pm_message_t state, bool async) { pm_callback_t callback = NULL; const char *info = NULL; @@ -904,8 +1051,20 @@ static int device_resume(struct device *dev, pm_message_t state, bool async) if (dev->power.syscore) goto Complete; + if (!dev->power.is_suspended) + goto Complete; + + dev->power.is_suspended = false; + if (dev->power.direct_complete) { - /* Match the pm_runtime_disable() in __device_suspend(). */ + /* + * Allow new children to be added under the device after this + * point if it has no PM callbacks. + */ + if (dev->power.no_pm_callbacks) + dev->power.is_prepared = false; + + /* Match the pm_runtime_disable() in device_suspend(). */ pm_runtime_enable(dev); goto Complete; } @@ -922,9 +1081,6 @@ static int device_resume(struct device *dev, pm_message_t state, bool async) */ dev->power.is_prepared = false; - if (!dev->power.is_suspended) - goto Unlock; - if (dev->pm_domain) { info = "power domain "; callback = pm_op(&dev->pm_domain->ops, state); @@ -962,9 +1118,7 @@ static int device_resume(struct device *dev, pm_message_t state, bool async) End: error = dpm_run_callback(callback, dev, state, info); - dev->power.is_suspended = false; - Unlock: device_unlock(dev); dpm_watchdog_clear(&wd); @@ -973,17 +1127,20 @@ static int device_resume(struct device *dev, pm_message_t state, bool async) TRACE_RESUME(error); - return error; + if (error) { + WRITE_ONCE(async_error, error); + dpm_save_failed_dev(dev_name(dev)); + pm_dev_err(dev, state, async ? " async" : "", error); + } + + dpm_async_resume_subordinate(dev, async_resume); } static void async_resume(void *data, async_cookie_t cookie) { - struct device *dev = (struct device *)data; - int error; + struct device *dev = data; - error = device_resume(dev, pm_transition, true); - if (error) - pm_dev_err(dev, pm_transition, " async", error); + device_resume(dev, pm_transition, true); put_device(dev); } @@ -1000,40 +1157,43 @@ void dpm_resume(pm_message_t state) ktime_t starttime = ktime_get(); trace_suspend_resume(TPS("dpm_resume"), state.event, true); - might_sleep(); - mutex_lock(&dpm_list_mtx); pm_transition = state; async_error = 0; - list_for_each_entry(dev, &dpm_suspended_list, power.entry) - dpm_async_fn(dev, async_resume); + mutex_lock(&dpm_list_mtx); + + /* + * Start processing "async" root devices upfront so they don't wait for + * the "sync" devices they don't depend on. + */ + list_for_each_entry(dev, &dpm_suspended_list, power.entry) { + dpm_clear_async_state(dev); + if (dpm_root_device(dev)) + dpm_async_with_cleanup(dev, async_resume); + } while (!list_empty(&dpm_suspended_list)) { dev = to_device(dpm_suspended_list.next); - get_device(dev); - if (!is_async(dev)) { - int error; + list_move_tail(&dev->power.entry, &dpm_prepared_list); + + if (!dpm_async_fn(dev, async_resume)) { + get_device(dev); mutex_unlock(&dpm_list_mtx); - error = device_resume(dev, state, false); - if (error) { - suspend_stats.failed_resume++; - dpm_save_failed_step(SUSPEND_RESUME); - dpm_save_failed_dev(dev_name(dev)); - pm_dev_err(dev, state, "", error); - } + device_resume(dev, state, false); + + put_device(dev); mutex_lock(&dpm_list_mtx); } - if (!list_empty(&dev->power.entry)) - list_move_tail(&dev->power.entry, &dpm_prepared_list); - put_device(dev); } mutex_unlock(&dpm_list_mtx); async_synchronize_full(); dpm_show_time(starttime, state, 0, NULL); + if (READ_ONCE(async_error)) + dpm_save_failed_step(SUSPEND_RESUME); cpufreq_resume(); devfreq_resume(); @@ -1051,7 +1211,7 @@ static void device_complete(struct device *dev, pm_message_t state) const char *info = NULL; if (dev->power.syscore) - return; + goto out; device_lock(dev); @@ -1081,6 +1241,9 @@ static void device_complete(struct device *dev, pm_message_t state) device_unlock(dev); +out: + /* If enabling runtime PM for the device is blocked, unblock it. */ + pm_runtime_unblock(dev); pm_runtime_put(dev); } @@ -1096,7 +1259,6 @@ void dpm_complete(pm_message_t state) struct list_head list; trace_suspend_resume(TPS("dpm_complete"), state.event, true); - might_sleep(); INIT_LIST_HEAD(&list); mutex_lock(&dpm_list_mtx); @@ -1106,14 +1268,16 @@ void dpm_complete(pm_message_t state) get_device(dev); dev->power.is_prepared = false; list_move(&dev->power.entry, &list); + mutex_unlock(&dpm_list_mtx); trace_device_pm_callback_start(dev, "", state.event); device_complete(dev, state); trace_device_pm_callback_end(dev, 0); - mutex_lock(&dpm_list_mtx); put_device(dev); + + mutex_lock(&dpm_list_mtx); } list_splice(&list, &dpm_list); mutex_unlock(&dpm_list_mtx); @@ -1133,6 +1297,7 @@ void dpm_complete(pm_message_t state) void dpm_resume_end(pm_message_t state) { dpm_resume(state); + pm_restore_gfp_mask(); dpm_complete(state); } EXPORT_SYMBOL_GPL(dpm_resume_end); @@ -1140,6 +1305,82 @@ EXPORT_SYMBOL_GPL(dpm_resume_end); /*------------------------- Suspend routines -------------------------*/ +static bool dpm_leaf_device(struct device *dev) +{ + struct device *child; + + lockdep_assert_held(&dpm_list_mtx); + + child = device_find_any_child(dev); + if (child) { + put_device(child); + + return false; + } + + /* + * Since this function is required to run under dpm_list_mtx, the + * list_empty() below will only return true if the device's list of + * consumers is actually empty before calling it. + */ + return list_empty(&dev->links.consumers); +} + +static bool dpm_async_suspend_parent(struct device *dev, async_func_t func) +{ + guard(mutex)(&dpm_list_mtx); + + /* + * If the device is suspended asynchronously and the parent's callback + * deletes both the device and the parent itself, the parent object may + * be freed while this function is running, so avoid that by checking + * if the device has been deleted already as the parent cannot be + * deleted before it. + */ + if (!device_pm_initialized(dev)) + return false; + + /* Start processing the device's parent if it is "async". */ + if (dev->parent) + dpm_async_with_cleanup(dev->parent, func); + + return true; +} + +static void dpm_async_suspend_superior(struct device *dev, async_func_t func) +{ + struct device_link *link; + int idx; + + if (!dpm_async_suspend_parent(dev, func)) + return; + + idx = device_links_read_lock(); + + /* Start processing the device's "async" suppliers. */ + dev_for_each_link_to_supplier(link, dev) + if (READ_ONCE(link->status) != DL_STATE_DORMANT) + dpm_async_with_cleanup(link->supplier, func); + + device_links_read_unlock(idx); +} + +static void dpm_async_suspend_complete_all(struct list_head *device_list) +{ + struct device *dev; + + guard(mutex)(&async_wip_mtx); + + list_for_each_entry_reverse(dev, device_list, power.entry) { + /* + * In case the device is being waited for and async processing + * has not started for it yet, let the waiters make progress. + */ + if (!dev->power.work_in_progress) + complete_all(&dev->power.completion); + } +} + /** * resume_event - Return a "resume" message for given "suspend" sleep state. * @sleep_state: PM message representing a sleep state. @@ -1171,14 +1412,16 @@ static void dpm_superior_set_must_resume(struct device *dev) idx = device_links_read_lock(); - list_for_each_entry_rcu_locked(link, &dev->links.suppliers, c_node) + dev_for_each_link_to_supplier(link, dev) link->supplier->power.must_resume = true; device_links_read_unlock(idx); } +static void async_suspend_noirq(void *data, async_cookie_t cookie); + /** - * __device_suspend_noirq - Execute a "noirq suspend" callback for given device. + * device_suspend_noirq - Execute a "noirq suspend" callback for given device. * @dev: Device to handle. * @state: PM transition of the system being carried out. * @async: If true, the device is being suspended asynchronously. @@ -1186,7 +1429,7 @@ static void dpm_superior_set_must_resume(struct device *dev) * The driver of @dev will not receive interrupts while this function is being * executed. */ -static int __device_suspend_noirq(struct device *dev, pm_message_t state, bool async) +static void device_suspend_noirq(struct device *dev, pm_message_t state, bool async) { pm_callback_t callback = NULL; const char *info = NULL; @@ -1197,7 +1440,7 @@ static int __device_suspend_noirq(struct device *dev, pm_message_t state, bool a dpm_wait_for_subordinate(dev, async); - if (async_error) + if (READ_ONCE(async_error)) goto Complete; if (dev->power.syscore || dev->power.direct_complete) @@ -1230,7 +1473,9 @@ static int __device_suspend_noirq(struct device *dev, pm_message_t state, bool a Run: error = dpm_run_callback(callback, dev, state, info); if (error) { - async_error = error; + WRITE_ONCE(async_error, error); + dpm_save_failed_dev(dev_name(dev)); + pm_dev_err(dev, state, async ? " async noirq" : " noirq", error); goto Complete; } @@ -1238,14 +1483,13 @@ Skip: dev->power.is_noirq_suspended = true; /* - * Skipping the resume of devices that were in use right before the - * system suspend (as indicated by their PM-runtime usage counters) - * would be suboptimal. Also resume them if doing that is not allowed - * to be skipped. + * Devices must be resumed unless they are explicitly allowed to be left + * in suspend, but even in that case skipping the resume of devices that + * were in use right before the system suspend (as indicated by their + * runtime PM usage counters and child counters) would be suboptimal. */ - if (atomic_read(&dev->power.usage_count) > 1 || - !(dev_pm_test_driver_flags(dev, DPM_FLAG_MAY_SKIP_RESUME) && - dev->power.may_skip_resume)) + if (!(dev_pm_test_driver_flags(dev, DPM_FLAG_MAY_SKIP_RESUME) && + dev->power.may_skip_resume) || !pm_runtime_need_not_resume(dev)) dev->power.must_resume = true; if (dev->power.must_resume) @@ -1254,72 +1498,81 @@ Skip: Complete: complete_all(&dev->power.completion); TRACE_SUSPEND(error); - return error; -} - -static void async_suspend_noirq(void *data, async_cookie_t cookie) -{ - struct device *dev = (struct device *)data; - int error; - error = __device_suspend_noirq(dev, pm_transition, true); - if (error) { - dpm_save_failed_dev(dev_name(dev)); - pm_dev_err(dev, pm_transition, " async", error); - } + if (error || READ_ONCE(async_error)) + return; - put_device(dev); + dpm_async_suspend_superior(dev, async_suspend_noirq); } -static int device_suspend_noirq(struct device *dev) +static void async_suspend_noirq(void *data, async_cookie_t cookie) { - if (dpm_async_fn(dev, async_suspend_noirq)) - return 0; + struct device *dev = data; - return __device_suspend_noirq(dev, pm_transition, false); + device_suspend_noirq(dev, pm_transition, true); + put_device(dev); } static int dpm_noirq_suspend_devices(pm_message_t state) { ktime_t starttime = ktime_get(); - int error = 0; + struct device *dev; + int error; trace_suspend_resume(TPS("dpm_suspend_noirq"), state.event, true); - mutex_lock(&dpm_list_mtx); + pm_transition = state; async_error = 0; + mutex_lock(&dpm_list_mtx); + + /* + * Start processing "async" leaf devices upfront so they don't need to + * wait for the "sync" devices they don't depend on. + */ + list_for_each_entry_reverse(dev, &dpm_late_early_list, power.entry) { + dpm_clear_async_state(dev); + if (dpm_leaf_device(dev)) + dpm_async_with_cleanup(dev, async_suspend_noirq); + } + while (!list_empty(&dpm_late_early_list)) { - struct device *dev = to_device(dpm_late_early_list.prev); + dev = to_device(dpm_late_early_list.prev); + + list_move(&dev->power.entry, &dpm_noirq_list); + + if (dpm_async_fn(dev, async_suspend_noirq)) + continue; get_device(dev); + mutex_unlock(&dpm_list_mtx); - error = device_suspend_noirq(dev); + device_suspend_noirq(dev, state, false); - mutex_lock(&dpm_list_mtx); - if (error) { - pm_dev_err(dev, state, " noirq", error); - dpm_save_failed_dev(dev_name(dev)); - put_device(dev); - break; - } - if (!list_empty(&dev->power.entry)) - list_move(&dev->power.entry, &dpm_noirq_list); put_device(dev); - if (async_error) + mutex_lock(&dpm_list_mtx); + + if (READ_ONCE(async_error)) { + dpm_async_suspend_complete_all(&dpm_late_early_list); + /* + * Move all devices to the target list to resume them + * properly. + */ + list_splice_init(&dpm_late_early_list, &dpm_noirq_list); break; + } } + mutex_unlock(&dpm_list_mtx); + async_synchronize_full(); - if (!error) - error = async_error; - if (error) { - suspend_stats.failed_suspend_noirq++; + error = READ_ONCE(async_error); + if (error) dpm_save_failed_step(SUSPEND_SUSPEND_NOIRQ); - } + dpm_show_time(starttime, state, error, "noirq"); trace_suspend_resume(TPS("dpm_suspend_noirq"), state.event, false); return error; @@ -1336,8 +1589,6 @@ int dpm_suspend_noirq(pm_message_t state) { int ret; - cpuidle_pause(); - device_wakeup_arm_wake_irqs(); suspend_device_irqs(); @@ -1363,15 +1614,17 @@ static void dpm_propagate_wakeup_to_parent(struct device *dev) spin_unlock_irq(&parent->power.lock); } +static void async_suspend_late(void *data, async_cookie_t cookie); + /** - * __device_suspend_late - Execute a "late suspend" callback for given device. + * device_suspend_late - Execute a "late suspend" callback for given device. * @dev: Device to handle. * @state: PM transition of the system being carried out. * @async: If true, the device is being suspended asynchronously. * * Runtime PM is disabled for @dev while this function is being executed. */ -static int __device_suspend_late(struct device *dev, pm_message_t state, bool async) +static void device_suspend_late(struct device *dev, pm_message_t state, bool async) { pm_callback_t callback = NULL; const char *info = NULL; @@ -1380,21 +1633,28 @@ static int __device_suspend_late(struct device *dev, pm_message_t state, bool as TRACE_DEVICE(dev); TRACE_SUSPEND(0); - __pm_runtime_disable(dev, false); - dpm_wait_for_subordinate(dev, async); - if (async_error) + if (READ_ONCE(async_error)) goto Complete; if (pm_wakeup_pending()) { - async_error = -EBUSY; + WRITE_ONCE(async_error, -EBUSY); goto Complete; } - if (dev->power.syscore || dev->power.direct_complete) + if (dev->power.direct_complete) goto Complete; + /* + * Disable runtime PM for the device without checking if there is a + * pending resume request for it. + */ + __pm_runtime_disable(dev, false); + + if (dev->power.syscore) + goto Skip; + if (dev->pm_domain) { info = "late power domain "; callback = pm_late_early_op(&dev->pm_domain->ops, state); @@ -1422,7 +1682,10 @@ static int __device_suspend_late(struct device *dev, pm_message_t state, bool as Run: error = dpm_run_callback(callback, dev, state, info); if (error) { - async_error = error; + WRITE_ONCE(async_error, error); + dpm_save_failed_dev(dev_name(dev)); + pm_dev_err(dev, state, async ? " async late" : " late", error); + pm_runtime_enable(dev); goto Complete; } dpm_propagate_wakeup_to_parent(dev); @@ -1433,28 +1696,19 @@ Skip: Complete: TRACE_SUSPEND(error); complete_all(&dev->power.completion); - return error; -} -static void async_suspend_late(void *data, async_cookie_t cookie) -{ - struct device *dev = (struct device *)data; - int error; + if (error || READ_ONCE(async_error)) + return; - error = __device_suspend_late(dev, pm_transition, true); - if (error) { - dpm_save_failed_dev(dev_name(dev)); - pm_dev_err(dev, pm_transition, " async", error); - } - put_device(dev); + dpm_async_suspend_superior(dev, async_suspend_late); } -static int device_suspend_late(struct device *dev) +static void async_suspend_late(void *data, async_cookie_t cookie) { - if (dpm_async_fn(dev, async_suspend_late)) - return 0; + struct device *dev = data; - return __device_suspend_late(dev, pm_transition, false); + device_suspend_late(dev, pm_transition, true); + put_device(dev); } /** @@ -1464,42 +1718,63 @@ static int device_suspend_late(struct device *dev) int dpm_suspend_late(pm_message_t state) { ktime_t starttime = ktime_get(); - int error = 0; + struct device *dev; + int error; trace_suspend_resume(TPS("dpm_suspend_late"), state.event, true); - mutex_lock(&dpm_list_mtx); + pm_transition = state; async_error = 0; + wake_up_all_idle_cpus(); + + mutex_lock(&dpm_list_mtx); + + /* + * Start processing "async" leaf devices upfront so they don't need to + * wait for the "sync" devices they don't depend on. + */ + list_for_each_entry_reverse(dev, &dpm_suspended_list, power.entry) { + dpm_clear_async_state(dev); + if (dpm_leaf_device(dev)) + dpm_async_with_cleanup(dev, async_suspend_late); + } + while (!list_empty(&dpm_suspended_list)) { - struct device *dev = to_device(dpm_suspended_list.prev); + dev = to_device(dpm_suspended_list.prev); + + list_move(&dev->power.entry, &dpm_late_early_list); + + if (dpm_async_fn(dev, async_suspend_late)) + continue; get_device(dev); + mutex_unlock(&dpm_list_mtx); - error = device_suspend_late(dev); + device_suspend_late(dev, state, false); + + put_device(dev); mutex_lock(&dpm_list_mtx); - if (!list_empty(&dev->power.entry)) - list_move(&dev->power.entry, &dpm_late_early_list); - if (error) { - pm_dev_err(dev, state, " late", error); - dpm_save_failed_dev(dev_name(dev)); - put_device(dev); + if (READ_ONCE(async_error)) { + dpm_async_suspend_complete_all(&dpm_suspended_list); + /* + * Move all devices to the target list to resume them + * properly. + */ + list_splice_init(&dpm_suspended_list, &dpm_late_early_list); break; } - put_device(dev); - - if (async_error) - break; } + mutex_unlock(&dpm_list_mtx); + async_synchronize_full(); - if (!error) - error = async_error; + + error = READ_ONCE(async_error); if (error) { - suspend_stats.failed_suspend_late++; dpm_save_failed_step(SUSPEND_SUSPEND_LATE); dpm_resume_early(resume_event(state)); } @@ -1550,7 +1825,7 @@ static int legacy_suspend(struct device *dev, pm_message_t state, trace_device_pm_callback_start(dev, info, state.event); error = cb(dev, state); trace_device_pm_callback_end(dev, error); - suspend_report_result(cb, error); + suspend_report_result(dev, cb, error); initcall_debug_report(dev, calltime, cb, error); @@ -1570,7 +1845,7 @@ static void dpm_clear_superiors_direct_complete(struct device *dev) idx = device_links_read_lock(); - list_for_each_entry_rcu_locked(link, &dev->links.suppliers, c_node) { + dev_for_each_link_to_supplier(link, dev) { spin_lock_irq(&link->supplier->power.lock); link->supplier->power.direct_complete = false; spin_unlock_irq(&link->supplier->power.lock); @@ -1579,13 +1854,15 @@ static void dpm_clear_superiors_direct_complete(struct device *dev) device_links_read_unlock(idx); } +static void async_suspend(void *data, async_cookie_t cookie); + /** - * __device_suspend - Execute "suspend" callbacks for given device. + * device_suspend - Execute "suspend" callbacks for given device. * @dev: Device to handle. * @state: PM transition of the system being carried out. * @async: If true, the device is being suspended asynchronously. */ -static int __device_suspend(struct device *dev, pm_message_t state, bool async) +static void device_suspend(struct device *dev, pm_message_t state, bool async) { pm_callback_t callback = NULL; const char *info = NULL; @@ -1597,7 +1874,7 @@ static int __device_suspend(struct device *dev, pm_message_t state, bool async) dpm_wait_for_subordinate(dev, async); - if (async_error) { + if (READ_ONCE(async_error)) { dev->power.direct_complete = false; goto Complete; } @@ -1617,7 +1894,7 @@ static int __device_suspend(struct device *dev, pm_message_t state, bool async) if (pm_wakeup_pending()) { dev->power.direct_complete = false; - async_error = -EBUSY; + WRITE_ONCE(async_error, -EBUSY); goto Complete; } @@ -1633,6 +1910,7 @@ static int __device_suspend(struct device *dev, pm_message_t state, bool async) pm_runtime_disable(dev); if (pm_runtime_status_suspended(dev)) { pm_dev_dbg(dev, state, "direct-complete "); + dev->power.is_suspended = true; goto Complete; } @@ -1642,7 +1920,7 @@ static int __device_suspend(struct device *dev, pm_message_t state, bool async) } dev->power.may_skip_resume = true; - dev->power.must_resume = false; + dev->power.must_resume = !dev_pm_test_driver_flags(dev, DPM_FLAG_MAY_SKIP_RESUME); dpm_watchdog_set(&wd, dev); device_lock(dev); @@ -1699,34 +1977,27 @@ static int __device_suspend(struct device *dev, pm_message_t state, bool async) dpm_watchdog_clear(&wd); Complete: - if (error) - async_error = error; + if (error) { + WRITE_ONCE(async_error, error); + dpm_save_failed_dev(dev_name(dev)); + pm_dev_err(dev, state, async ? " async" : "", error); + } complete_all(&dev->power.completion); TRACE_SUSPEND(error); - return error; -} -static void async_suspend(void *data, async_cookie_t cookie) -{ - struct device *dev = (struct device *)data; - int error; - - error = __device_suspend(dev, pm_transition, true); - if (error) { - dpm_save_failed_dev(dev_name(dev)); - pm_dev_err(dev, pm_transition, " async", error); - } + if (error || READ_ONCE(async_error)) + return; - put_device(dev); + dpm_async_suspend_superior(dev, async_suspend); } -static int device_suspend(struct device *dev) +static void async_suspend(void *data, async_cookie_t cookie) { - if (dpm_async_fn(dev, async_suspend)) - return 0; + struct device *dev = data; - return __device_suspend(dev, pm_transition, false); + device_suspend(dev, pm_transition, true); + put_device(dev); } /** @@ -1736,7 +2007,8 @@ static int device_suspend(struct device *dev) int dpm_suspend(pm_message_t state) { ktime_t starttime = ktime_get(); - int error = 0; + struct device *dev; + int error; trace_suspend_resume(TPS("dpm_suspend"), state.event, true); might_sleep(); @@ -1744,43 +2016,103 @@ int dpm_suspend(pm_message_t state) devfreq_suspend(); cpufreq_suspend(); - mutex_lock(&dpm_list_mtx); pm_transition = state; async_error = 0; + + mutex_lock(&dpm_list_mtx); + + /* + * Start processing "async" leaf devices upfront so they don't need to + * wait for the "sync" devices they don't depend on. + */ + list_for_each_entry_reverse(dev, &dpm_prepared_list, power.entry) { + dpm_clear_async_state(dev); + if (dpm_leaf_device(dev)) + dpm_async_with_cleanup(dev, async_suspend); + } + while (!list_empty(&dpm_prepared_list)) { - struct device *dev = to_device(dpm_prepared_list.prev); + dev = to_device(dpm_prepared_list.prev); + + list_move(&dev->power.entry, &dpm_suspended_list); + + if (dpm_async_fn(dev, async_suspend)) + continue; get_device(dev); + mutex_unlock(&dpm_list_mtx); - error = device_suspend(dev); + device_suspend(dev, state, false); + + put_device(dev); mutex_lock(&dpm_list_mtx); - if (error) { - pm_dev_err(dev, state, "", error); - dpm_save_failed_dev(dev_name(dev)); - put_device(dev); + + if (READ_ONCE(async_error)) { + dpm_async_suspend_complete_all(&dpm_prepared_list); + /* + * Move all devices to the target list to resume them + * properly. + */ + list_splice_init(&dpm_prepared_list, &dpm_suspended_list); break; } - if (!list_empty(&dev->power.entry)) - list_move(&dev->power.entry, &dpm_suspended_list); - put_device(dev); - if (async_error) - break; } + mutex_unlock(&dpm_list_mtx); + async_synchronize_full(); - if (!error) - error = async_error; - if (error) { - suspend_stats.failed_suspend++; + + error = READ_ONCE(async_error); + if (error) dpm_save_failed_step(SUSPEND_SUSPEND); - } + dpm_show_time(starttime, state, error, NULL); trace_suspend_resume(TPS("dpm_suspend"), state.event, false); return error; } +static bool device_prepare_smart_suspend(struct device *dev) +{ + struct device_link *link; + bool ret = true; + int idx; + + /* + * The "smart suspend" feature is enabled for devices whose drivers ask + * for it and for devices without PM callbacks. + * + * However, if "smart suspend" is not enabled for the device's parent + * or any of its suppliers that take runtime PM into account, it cannot + * be enabled for the device either. + */ + if (!dev->power.no_pm_callbacks && + !dev_pm_test_driver_flags(dev, DPM_FLAG_SMART_SUSPEND)) + return false; + + if (dev->parent && !dev_pm_smart_suspend(dev->parent) && + !dev->parent->power.ignore_children && !pm_runtime_blocked(dev->parent)) + return false; + + idx = device_links_read_lock(); + + dev_for_each_link_to_supplier(link, dev) { + if (!device_link_test(link, DL_FLAG_PM_RUNTIME)) + continue; + + if (!dev_pm_smart_suspend(link->supplier) && + !pm_runtime_blocked(link->supplier)) { + ret = false; + break; + } + } + + device_links_read_unlock(idx); + + return ret; +} + /** * device_prepare - Prepare a device for system power transition. * @dev: Device to handle. @@ -1792,11 +2124,9 @@ int dpm_suspend(pm_message_t state) static int device_prepare(struct device *dev, pm_message_t state) { int (*callback)(struct device *) = NULL; + bool smart_suspend; int ret = 0; - if (dev->power.syscore) - return 0; - /* * If a device's parent goes into runtime suspend at the wrong time, * it won't be possible to resume the device. To prevent this we @@ -1804,10 +2134,21 @@ static int device_prepare(struct device *dev, pm_message_t state) * it again during the complete phase. */ pm_runtime_get_noresume(dev); + /* + * If runtime PM is disabled for the device at this point and it has + * never been enabled so far, it should not be enabled until this system + * suspend-resume cycle is complete, so prepare to trigger a warning on + * subsequent attempts to enable it. + */ + smart_suspend = !pm_runtime_block_if_disabled(dev); + + if (dev->power.syscore) + return 0; device_lock(dev); dev->power.wakeup_path = false; + dev->power.out_band_wakeup = false; if (dev->power.no_pm_callbacks) goto unlock; @@ -1831,10 +2172,17 @@ unlock: device_unlock(dev); if (ret < 0) { - suspend_report_result(callback, ret); + suspend_report_result(dev, callback, ret); pm_runtime_put(dev); return ret; } + /* Do not enable "smart suspend" for devices with disabled runtime PM. */ + if (smart_suspend) + smart_suspend = device_prepare_smart_suspend(dev); + + spin_lock_irq(&dev->power.lock); + + dev->power.smart_suspend = smart_suspend; /* * A positive return value from ->prepare() means "this device appears * to be runtime-suspended and its state is fine, so if it really is @@ -1842,11 +2190,12 @@ unlock: * will do the same thing with all of its descendants". This only * applies to suspend transitions, however. */ - spin_lock_irq(&dev->power.lock); dev->power.direct_complete = state.event == PM_EVENT_SUSPEND && (ret > 0 || dev->power.no_pm_callbacks) && !dev_pm_test_driver_flags(dev, DPM_FLAG_NO_DIRECT_COMPLETE); + spin_unlock_irq(&dev->power.lock); + return 0; } @@ -1861,7 +2210,6 @@ int dpm_prepare(pm_message_t state) int error = 0; trace_suspend_resume(TPS("dpm_prepare"), state.event, true); - might_sleep(); /* * Give a chance for the known devices to complete their probes, before @@ -1878,10 +2226,11 @@ int dpm_prepare(pm_message_t state) device_block_probing(); mutex_lock(&dpm_list_mtx); - while (!list_empty(&dpm_list)) { + while (!list_empty(&dpm_list) && !error) { struct device *dev = to_device(dpm_list.next); get_device(dev); + mutex_unlock(&dpm_list_mtx); trace_device_pm_callback_start(dev, "", state.event); @@ -1889,21 +2238,23 @@ int dpm_prepare(pm_message_t state) trace_device_pm_callback_end(dev, error); mutex_lock(&dpm_list_mtx); - if (error) { - if (error == -EAGAIN) { - put_device(dev); - error = 0; - continue; - } + + if (!error) { + dev->power.is_prepared = true; + if (!list_empty(&dev->power.entry)) + list_move_tail(&dev->power.entry, &dpm_prepared_list); + } else if (error == -EAGAIN) { + error = 0; + } else { dev_info(dev, "not prepared for power transition: code %d\n", error); - put_device(dev); - break; } - dev->power.is_prepared = true; - if (!list_empty(&dev->power.entry)) - list_move_tail(&dev->power.entry, &dpm_prepared_list); + + mutex_unlock(&dpm_list_mtx); + put_device(dev); + + mutex_lock(&dpm_list_mtx); } mutex_unlock(&dpm_list_mtx); trace_suspend_resume(TPS("dpm_prepare"), state.event, false); @@ -1923,20 +2274,22 @@ int dpm_suspend_start(pm_message_t state) int error; error = dpm_prepare(state); - if (error) { - suspend_stats.failed_prepare++; + if (error) dpm_save_failed_step(SUSPEND_PREPARE); - } else + else { + pm_restrict_gfp_mask(); error = dpm_suspend(state); + } + dpm_show_time(starttime, state, error, "start"); return error; } EXPORT_SYMBOL_GPL(dpm_suspend_start); -void __suspend_report_result(const char *function, void *fn, int ret) +void __suspend_report_result(const char *function, struct device *dev, void *fn, int ret) { if (ret) - pr_err("%s(): %pS returns %d\n", function, fn, ret); + dev_err(dev, "%s(): %ps returns %d\n", function, fn, ret); } EXPORT_SYMBOL_GPL(__suspend_report_result); @@ -1991,7 +2344,9 @@ static bool pm_ops_is_empty(const struct dev_pm_ops *ops) void device_pm_check_callbacks(struct device *dev) { - spin_lock_irq(&dev->power.lock); + unsigned long flags; + + spin_lock_irqsave(&dev->power.lock, flags); dev->power.no_pm_callbacks = (!dev->bus || (pm_ops_is_empty(dev->bus->pm) && !dev->bus->suspend && !dev->bus->resume)) && @@ -2000,11 +2355,10 @@ void device_pm_check_callbacks(struct device *dev) (!dev->pm_domain || pm_ops_is_empty(&dev->pm_domain->ops)) && (!dev->driver || (pm_ops_is_empty(dev->driver->pm) && !dev->driver->suspend && !dev->driver->resume)); - spin_unlock_irq(&dev->power.lock); + spin_unlock_irqrestore(&dev->power.lock, flags); } bool dev_pm_skip_suspend(struct device *dev) { - return dev_pm_test_driver_flags(dev, DPM_FLAG_SMART_SUSPEND) && - pm_runtime_status_suspended(dev); + return dev_pm_smart_suspend(dev) && pm_runtime_status_suspended(dev); } diff --git a/drivers/base/power/power.h b/drivers/base/power/power.h index 54292cdd7808..922ed457db19 100644 --- a/drivers/base/power/power.h +++ b/drivers/base/power/power.h @@ -25,8 +25,11 @@ extern u64 pm_runtime_active_time(struct device *dev); #define WAKE_IRQ_DEDICATED_ALLOCATED BIT(0) #define WAKE_IRQ_DEDICATED_MANAGED BIT(1) +#define WAKE_IRQ_DEDICATED_REVERSE BIT(2) #define WAKE_IRQ_DEDICATED_MASK (WAKE_IRQ_DEDICATED_ALLOCATED | \ - WAKE_IRQ_DEDICATED_MANAGED) + WAKE_IRQ_DEDICATED_MANAGED | \ + WAKE_IRQ_DEDICATED_REVERSE) +#define WAKE_IRQ_DEDICATED_ENABLED BIT(3) struct wake_irq { struct device *dev; @@ -39,7 +42,8 @@ extern void dev_pm_arm_wake_irq(struct wake_irq *wirq); extern void dev_pm_disarm_wake_irq(struct wake_irq *wirq); extern void dev_pm_enable_wake_irq_check(struct device *dev, bool can_change_status); -extern void dev_pm_disable_wake_irq_check(struct device *dev); +extern void dev_pm_disable_wake_irq_check(struct device *dev, bool cond_disable); +extern void dev_pm_enable_wake_irq_complete(struct device *dev); #ifdef CONFIG_PM_SLEEP diff --git a/drivers/base/power/qos.c b/drivers/base/power/qos.c index 8e93167f1783..ff393cba7649 100644 --- a/drivers/base/power/qos.c +++ b/drivers/base/power/qos.c @@ -137,6 +137,7 @@ s32 dev_pm_qos_read_value(struct device *dev, enum dev_pm_qos_req_type type) return ret; } +EXPORT_SYMBOL_GPL(dev_pm_qos_read_value); /** * apply_constraint - Add/modify/remove device PM QoS request. @@ -201,7 +202,7 @@ static int dev_pm_qos_constraints_allocate(struct device *dev) if (!qos) return -ENOMEM; - n = kzalloc(3 * sizeof(*n), GFP_KERNEL); + n = kcalloc(3, sizeof(*n), GFP_KERNEL); if (!n) { kfree(qos); return -ENOMEM; diff --git a/drivers/base/power/runtime-test.c b/drivers/base/power/runtime-test.c new file mode 100644 index 000000000000..1535ad2b0264 --- /dev/null +++ b/drivers/base/power/runtime-test.c @@ -0,0 +1,249 @@ +// SPDX-License-Identifier: GPL-2.0 +/* + * Copyright 2025 Google, Inc. + */ + +#include <linux/cleanup.h> +#include <linux/pm_runtime.h> +#include <kunit/device.h> +#include <kunit/test.h> + +#define DEVICE_NAME "pm_runtime_test_device" + +static void pm_runtime_depth_test(struct kunit *test) +{ + struct device *dev = kunit_device_register(test, DEVICE_NAME); + + KUNIT_ASSERT_NOT_ERR_OR_NULL(test, dev); + + pm_runtime_enable(dev); + + KUNIT_EXPECT_TRUE(test, pm_runtime_suspended(dev)); + KUNIT_EXPECT_EQ(test, 0, pm_runtime_get_sync(dev)); + KUNIT_EXPECT_TRUE(test, pm_runtime_active(dev)); + KUNIT_EXPECT_EQ(test, 1, pm_runtime_get_sync(dev)); /* "already active" */ + KUNIT_EXPECT_EQ(test, 0, pm_runtime_put_sync(dev)); + KUNIT_EXPECT_EQ(test, 0, pm_runtime_put_sync(dev)); + KUNIT_EXPECT_TRUE(test, pm_runtime_suspended(dev)); +} + +/* Test pm_runtime_put() and friends when already suspended. */ +static void pm_runtime_already_suspended_test(struct kunit *test) +{ + struct device *dev = kunit_device_register(test, DEVICE_NAME); + + KUNIT_ASSERT_NOT_ERR_OR_NULL(test, dev); + + pm_runtime_enable(dev); + KUNIT_EXPECT_TRUE(test, pm_runtime_suspended(dev)); + + pm_runtime_get_noresume(dev); + KUNIT_EXPECT_EQ(test, 1, pm_runtime_put_sync(dev)); + + KUNIT_EXPECT_EQ(test, 1, pm_runtime_suspend(dev)); + KUNIT_EXPECT_EQ(test, 1, pm_runtime_autosuspend(dev)); + KUNIT_EXPECT_EQ(test, 1, pm_request_autosuspend(dev)); + + pm_runtime_get_noresume(dev); + KUNIT_EXPECT_EQ(test, 1, pm_runtime_put_sync_autosuspend(dev)); + + pm_runtime_get_noresume(dev); + pm_runtime_put_autosuspend(dev); + + /* Grab 2 refcounts */ + pm_runtime_get_noresume(dev); + pm_runtime_get_noresume(dev); + /* The first put() sees usage_count 1 */ + KUNIT_EXPECT_EQ(test, 0, pm_runtime_put_sync_autosuspend(dev)); + /* The second put() sees usage_count 0 but tells us "already suspended". */ + KUNIT_EXPECT_EQ(test, 1, pm_runtime_put_sync_autosuspend(dev)); + + /* Should have remained suspended the whole time. */ + KUNIT_EXPECT_TRUE(test, pm_runtime_suspended(dev)); +} + +static void pm_runtime_idle_test(struct kunit *test) +{ + struct device *dev = kunit_device_register(test, DEVICE_NAME); + + KUNIT_ASSERT_NOT_ERR_OR_NULL(test, dev); + + pm_runtime_enable(dev); + + KUNIT_EXPECT_TRUE(test, pm_runtime_suspended(dev)); + KUNIT_EXPECT_EQ(test, 0, pm_runtime_get_sync(dev)); + KUNIT_EXPECT_TRUE(test, pm_runtime_active(dev)); + KUNIT_EXPECT_EQ(test, -EAGAIN, pm_runtime_idle(dev)); + KUNIT_EXPECT_TRUE(test, pm_runtime_active(dev)); + pm_runtime_put_noidle(dev); + KUNIT_EXPECT_TRUE(test, pm_runtime_active(dev)); + KUNIT_EXPECT_EQ(test, 0, pm_runtime_idle(dev)); + KUNIT_EXPECT_TRUE(test, pm_runtime_suspended(dev)); + KUNIT_EXPECT_EQ(test, -EAGAIN, pm_runtime_idle(dev)); + KUNIT_EXPECT_EQ(test, -EAGAIN, pm_request_idle(dev)); +} + +static void pm_runtime_disabled_test(struct kunit *test) +{ + struct device *dev = kunit_device_register(test, DEVICE_NAME); + + KUNIT_ASSERT_NOT_ERR_OR_NULL(test, dev); + + /* Never called pm_runtime_enable() */ + KUNIT_EXPECT_FALSE(test, pm_runtime_enabled(dev)); + + /* "disabled" is treated as "active" */ + KUNIT_EXPECT_TRUE(test, pm_runtime_active(dev)); + KUNIT_EXPECT_FALSE(test, pm_runtime_suspended(dev)); + + /* + * Note: these "fail", but they still acquire/release refcounts, so + * keep them balanced. + */ + KUNIT_EXPECT_EQ(test, -EACCES, pm_runtime_get(dev)); + pm_runtime_put(dev); + + KUNIT_EXPECT_EQ(test, -EACCES, pm_runtime_get_sync(dev)); + KUNIT_EXPECT_EQ(test, -EACCES, pm_runtime_put_sync(dev)); + + KUNIT_EXPECT_EQ(test, -EACCES, pm_runtime_get(dev)); + pm_runtime_put_autosuspend(dev); + + KUNIT_EXPECT_EQ(test, -EACCES, pm_runtime_resume_and_get(dev)); + KUNIT_EXPECT_EQ(test, -EACCES, pm_runtime_idle(dev)); + KUNIT_EXPECT_EQ(test, -EACCES, pm_request_idle(dev)); + KUNIT_EXPECT_EQ(test, -EACCES, pm_request_resume(dev)); + KUNIT_EXPECT_EQ(test, -EACCES, pm_request_autosuspend(dev)); + KUNIT_EXPECT_EQ(test, -EACCES, pm_runtime_suspend(dev)); + KUNIT_EXPECT_EQ(test, -EACCES, pm_runtime_resume(dev)); + KUNIT_EXPECT_EQ(test, -EACCES, pm_runtime_autosuspend(dev)); + + /* Still disabled */ + KUNIT_EXPECT_TRUE(test, pm_runtime_active(dev)); + KUNIT_EXPECT_FALSE(test, pm_runtime_enabled(dev)); +} + +static void pm_runtime_error_test(struct kunit *test) +{ + struct device *dev = kunit_device_register(test, DEVICE_NAME); + + KUNIT_ASSERT_NOT_ERR_OR_NULL(test, dev); + + pm_runtime_enable(dev); + KUNIT_EXPECT_TRUE(test, pm_runtime_suspended(dev)); + + /* Fake a .runtime_resume() error */ + dev->power.runtime_error = -EIO; + + /* + * Note: these "fail", but they still acquire/release refcounts, so + * keep them balanced. + */ + KUNIT_EXPECT_EQ(test, -EINVAL, pm_runtime_get(dev)); + pm_runtime_put(dev); + + KUNIT_EXPECT_EQ(test, -EINVAL, pm_runtime_get_sync(dev)); + KUNIT_EXPECT_EQ(test, -EINVAL, pm_runtime_put_sync(dev)); + + KUNIT_EXPECT_EQ(test, -EINVAL, pm_runtime_get(dev)); + pm_runtime_put_autosuspend(dev); + + KUNIT_EXPECT_EQ(test, -EINVAL, pm_runtime_get(dev)); + KUNIT_EXPECT_EQ(test, -EINVAL, pm_runtime_put_sync_autosuspend(dev)); + + KUNIT_EXPECT_EQ(test, -EINVAL, pm_runtime_resume_and_get(dev)); + KUNIT_EXPECT_EQ(test, -EINVAL, pm_runtime_idle(dev)); + KUNIT_EXPECT_EQ(test, -EINVAL, pm_request_idle(dev)); + KUNIT_EXPECT_EQ(test, -EINVAL, pm_request_resume(dev)); + KUNIT_EXPECT_EQ(test, -EINVAL, pm_request_autosuspend(dev)); + KUNIT_EXPECT_EQ(test, -EINVAL, pm_runtime_suspend(dev)); + KUNIT_EXPECT_EQ(test, -EINVAL, pm_runtime_resume(dev)); + KUNIT_EXPECT_EQ(test, -EINVAL, pm_runtime_autosuspend(dev)); + + /* Error is still pending */ + KUNIT_EXPECT_TRUE(test, pm_runtime_suspended(dev)); + KUNIT_EXPECT_EQ(test, -EIO, dev->power.runtime_error); + /* Clear error */ + KUNIT_EXPECT_EQ(test, 0, pm_runtime_set_suspended(dev)); + KUNIT_EXPECT_EQ(test, 0, dev->power.runtime_error); + /* Still suspended */ + KUNIT_EXPECT_TRUE(test, pm_runtime_suspended(dev)); + + KUNIT_EXPECT_EQ(test, 0, pm_runtime_get(dev)); + pm_runtime_barrier(dev); + pm_runtime_put(dev); + pm_runtime_suspend(dev); /* flush the put(), to suspend */ + KUNIT_EXPECT_TRUE(test, pm_runtime_suspended(dev)); + + KUNIT_EXPECT_EQ(test, 0, pm_runtime_get_sync(dev)); + KUNIT_EXPECT_EQ(test, 0, pm_runtime_put_sync(dev)); + + KUNIT_EXPECT_EQ(test, 0, pm_runtime_get_sync(dev)); + pm_runtime_put_autosuspend(dev); + + KUNIT_EXPECT_EQ(test, 0, pm_runtime_resume_and_get(dev)); + + /* + * The following should all return -EAGAIN (usage is non-zero) or 1 + * (already resumed). + */ + KUNIT_EXPECT_EQ(test, -EAGAIN, pm_runtime_idle(dev)); + KUNIT_EXPECT_EQ(test, -EAGAIN, pm_request_idle(dev)); + KUNIT_EXPECT_EQ(test, 1, pm_request_resume(dev)); + KUNIT_EXPECT_EQ(test, -EAGAIN, pm_request_autosuspend(dev)); + KUNIT_EXPECT_EQ(test, -EAGAIN, pm_runtime_suspend(dev)); + KUNIT_EXPECT_EQ(test, 1, pm_runtime_resume(dev)); + KUNIT_EXPECT_EQ(test, -EAGAIN, pm_runtime_autosuspend(dev)); + + KUNIT_EXPECT_EQ(test, 0, pm_runtime_put_sync(dev)); + + /* Suspended again */ + KUNIT_EXPECT_TRUE(test, pm_runtime_suspended(dev)); +} + +/* + * Explore a typical probe() sequence in which a device marks itself powered, + * but doesn't hold any runtime PM reference, so it suspends as soon as it goes + * idle. + */ +static void pm_runtime_probe_active_test(struct kunit *test) +{ + struct device *dev = kunit_device_register(test, DEVICE_NAME); + + KUNIT_ASSERT_NOT_ERR_OR_NULL(test, dev); + + KUNIT_EXPECT_TRUE(test, pm_runtime_status_suspended(dev)); + + KUNIT_EXPECT_EQ(test, 0, pm_runtime_set_active(dev)); + KUNIT_EXPECT_TRUE(test, pm_runtime_active(dev)); + + pm_runtime_enable(dev); + KUNIT_EXPECT_TRUE(test, pm_runtime_active(dev)); + + /* Nothing to flush. We stay active. */ + pm_runtime_barrier(dev); + KUNIT_EXPECT_TRUE(test, pm_runtime_active(dev)); + + /* Ask for idle? Now we suspend. */ + KUNIT_EXPECT_EQ(test, 0, pm_runtime_idle(dev)); + KUNIT_EXPECT_TRUE(test, pm_runtime_suspended(dev)); +} + +static struct kunit_case pm_runtime_test_cases[] = { + KUNIT_CASE(pm_runtime_depth_test), + KUNIT_CASE(pm_runtime_already_suspended_test), + KUNIT_CASE(pm_runtime_idle_test), + KUNIT_CASE(pm_runtime_disabled_test), + KUNIT_CASE(pm_runtime_error_test), + KUNIT_CASE(pm_runtime_probe_active_test), + {} +}; + +static struct kunit_suite pm_runtime_test_suite = { + .name = "pm_runtime_test_cases", + .test_cases = pm_runtime_test_cases, +}; + +kunit_test_suite(pm_runtime_test_suite); +MODULE_DESCRIPTION("Runtime power management unit test suite"); +MODULE_LICENSE("GPL"); diff --git a/drivers/base/power/runtime.c b/drivers/base/power/runtime.c index 8a66eaf731e4..84676cc24221 100644 --- a/drivers/base/power/runtime.c +++ b/drivers/base/power/runtime.c @@ -11,6 +11,7 @@ #include <linux/export.h> #include <linux/pm_runtime.h> #include <linux/pm_wakeirq.h> +#include <linux/rculist.h> #include <trace/events/rpm.h> #include "../base.h" @@ -18,10 +19,24 @@ typedef int (*pm_callback_t)(struct device *); +static inline pm_callback_t get_callback_ptr(const void *start, size_t offset) +{ + return *(pm_callback_t *)(start + offset); +} + +static pm_callback_t __rpm_get_driver_callback(struct device *dev, + size_t cb_offset) +{ + if (dev->driver && dev->driver->pm) + return get_callback_ptr(dev->driver->pm, cb_offset); + + return NULL; +} + static pm_callback_t __rpm_get_callback(struct device *dev, size_t cb_offset) { - pm_callback_t cb; const struct dev_pm_ops *ops; + pm_callback_t cb = NULL; if (dev->pm_domain) ops = &dev->pm_domain->ops; @@ -35,12 +50,10 @@ static pm_callback_t __rpm_get_callback(struct device *dev, size_t cb_offset) ops = NULL; if (ops) - cb = *(pm_callback_t *)((void *)ops + cb_offset); - else - cb = NULL; + cb = get_callback_ptr(ops, cb_offset); - if (!cb && dev->driver && dev->driver->pm) - cb = *(pm_callback_t *)((void *)dev->driver->pm + cb_offset); + if (!cb) + cb = __rpm_get_driver_callback(dev, cb_offset); return cb; } @@ -77,7 +90,7 @@ static void update_pm_runtime_accounting(struct device *dev) /* * Because ktime_get_mono_fast_ns() is not monotonic during * timekeeping updates, ensure that 'now' is after the last saved - * timesptamp. + * timestamp. */ if (now < last) return; @@ -93,6 +106,7 @@ static void update_pm_runtime_accounting(struct device *dev) static void __update_runtime_status(struct device *dev, enum rpm_status status) { update_pm_runtime_accounting(dev); + trace_rpm_status(dev, status); dev->power.runtime_status = status; } @@ -203,7 +217,7 @@ static int dev_memalloc_noio(struct device *dev, void *data) * resume/suspend callback of any one of its ancestors(or the * block device itself), the deadlock may be triggered inside the * memory allocation since it might not complete until the block - * device becomes active and the involed page I/O finishes. The + * device becomes active and the involved page I/O finishes. The * situation is pointed out first by Alan Stern. Network device * are involved in iSCSI kind of situation. * @@ -243,8 +257,7 @@ void pm_runtime_set_memalloc_noio(struct device *dev, bool enable) * flag was set by any one of the descendants. */ if (!dev || (!enable && - device_for_each_child(dev, NULL, - dev_memalloc_noio))) + device_for_each_child(dev, NULL, dev_memalloc_noio))) break; } mutex_unlock(&dev_hotplug_mutex); @@ -263,17 +276,15 @@ static int rpm_check_suspend_allowed(struct device *dev) retval = -EINVAL; else if (dev->power.disable_depth > 0) retval = -EACCES; - else if (atomic_read(&dev->power.usage_count) > 0) + else if (atomic_read(&dev->power.usage_count)) retval = -EAGAIN; - else if (!dev->power.ignore_children && - atomic_read(&dev->power.child_count)) + else if (!dev->power.ignore_children && atomic_read(&dev->power.child_count)) retval = -EBUSY; /* Pending resume requests take precedence over suspends. */ - else if ((dev->power.deferred_resume - && dev->power.runtime_status == RPM_SUSPENDING) - || (dev->power.request_pending - && dev->power.request == RPM_REQ_RESUME)) + else if ((dev->power.deferred_resume && + dev->power.runtime_status == RPM_SUSPENDING) || + (dev->power.request_pending && dev->power.request == RPM_REQ_RESUME)) retval = -EAGAIN; else if (__dev_pm_qos_resume_latency(dev) == 0) retval = -EPERM; @@ -291,7 +302,7 @@ static int rpm_get_suppliers(struct device *dev) device_links_read_lock_held()) { int retval; - if (!(link->flags & DL_FLAG_PM_RUNTIME)) + if (!device_link_test(link, DL_FLAG_PM_RUNTIME)) continue; retval = pm_runtime_get_sync(link->supplier); @@ -305,16 +316,34 @@ static int rpm_get_suppliers(struct device *dev) return 0; } +/** + * pm_runtime_release_supplier - Drop references to device link's supplier. + * @link: Target device link. + * + * Drop all runtime PM references associated with @link to its supplier device. + */ +void pm_runtime_release_supplier(struct device_link *link) +{ + struct device *supplier = link->supplier; + + /* + * The additional power.usage_count check is a safety net in case + * the rpm_active refcount becomes saturated, in which case + * refcount_dec_not_one() would return true forever, but it is not + * strictly necessary. + */ + while (refcount_dec_not_one(&link->rpm_active) && + atomic_read(&supplier->power.usage_count) > 0) + pm_runtime_put_noidle(supplier); +} + static void __rpm_put_suppliers(struct device *dev, bool try_to_suspend) { struct device_link *link; list_for_each_entry_rcu(link, &dev->links.suppliers, c_node, device_links_read_lock_held()) { - - while (refcount_dec_not_one(&link->rpm_active)) - pm_runtime_put_noidle(link->supplier); - + pm_runtime_release_supplier(link); if (try_to_suspend) pm_request_idle(link->supplier); } @@ -386,9 +415,9 @@ static int __rpm_callback(int (*cb)(struct device *), struct device *dev) * * Do that if resume fails too. */ - if (use_links - && ((dev->power.runtime_status == RPM_SUSPENDING && !retval) - || (dev->power.runtime_status == RPM_RESUMING && retval))) { + if (use_links && + ((dev->power.runtime_status == RPM_SUSPENDING && !retval) || + (dev->power.runtime_status == RPM_RESUMING && retval))) { idx = device_links_read_lock(); __rpm_put_suppliers(dev, false); @@ -404,6 +433,49 @@ fail: } /** + * rpm_callback - Run a given runtime PM callback for a given device. + * @cb: Runtime PM callback to run. + * @dev: Device to run the callback for. + */ +static int rpm_callback(int (*cb)(struct device *), struct device *dev) +{ + int retval; + + if (dev->power.memalloc_noio) { + unsigned int noio_flag; + + /* + * Deadlock might be caused if memory allocation with + * GFP_KERNEL happens inside runtime_suspend and + * runtime_resume callbacks of one block device's + * ancestor or the block device itself. Network + * device might be thought as part of iSCSI block + * device, so network device and its ancestor should + * be marked as memalloc_noio too. + */ + noio_flag = memalloc_noio_save(); + retval = __rpm_callback(cb, dev); + memalloc_noio_restore(noio_flag); + } else { + retval = __rpm_callback(cb, dev); + } + + /* + * Since -EACCES means that runtime PM is disabled for the given device, + * it should not be returned by runtime PM callbacks. If it is returned + * nevertheless, assume it to be a transient error and convert it to + * -EAGAIN. + */ + if (retval == -EACCES) + retval = -EAGAIN; + + if (retval != -EAGAIN && retval != -EBUSY) + dev->power.runtime_error = retval; + + return retval; +} + +/** * rpm_idle - Notify device bus type if the device can be suspended. * @dev: Device to notify the bus type about. * @rpmflags: Flag bits. @@ -421,11 +493,14 @@ static int rpm_idle(struct device *dev, int rpmflags) int (*callback)(struct device *); int retval; - trace_rpm_idle_rcuidle(dev, rpmflags); + trace_rpm_idle(dev, rpmflags); retval = rpm_check_suspend_allowed(dev); if (retval < 0) ; /* Conditions are wrong. */ + else if ((rpmflags & RPM_GET_PUT) && retval == 1) + ; /* put() is allowed in RPM_SUSPENDED */ + /* Idle notifications are allowed only in the RPM_ACTIVE state. */ else if (dev->power.runtime_status != RPM_ACTIVE) retval = -EAGAIN; @@ -441,6 +516,7 @@ static int rpm_idle(struct device *dev, int rpmflags) /* Act as though RPM_NOWAIT is always set. */ else if (dev->power.idle_notification) retval = -EINPROGRESS; + if (retval) goto out; @@ -460,55 +536,33 @@ static int rpm_idle(struct device *dev, int rpmflags) dev->power.request_pending = true; queue_work(pm_wq, &dev->power.work); } - trace_rpm_return_int_rcuidle(dev, _THIS_IP_, 0); + trace_rpm_return_int(dev, _THIS_IP_, 0); return 0; } dev->power.idle_notification = true; - retval = __rpm_callback(callback, dev); + if (dev->power.irq_safe) + spin_unlock(&dev->power.lock); + else + spin_unlock_irq(&dev->power.lock); + + retval = callback(dev); + + if (dev->power.irq_safe) + spin_lock(&dev->power.lock); + else + spin_lock_irq(&dev->power.lock); dev->power.idle_notification = false; wake_up_all(&dev->power.wait_queue); out: - trace_rpm_return_int_rcuidle(dev, _THIS_IP_, retval); + trace_rpm_return_int(dev, _THIS_IP_, retval); return retval ? retval : rpm_suspend(dev, rpmflags | RPM_AUTO); } /** - * rpm_callback - Run a given runtime PM callback for a given device. - * @cb: Runtime PM callback to run. - * @dev: Device to run the callback for. - */ -static int rpm_callback(int (*cb)(struct device *), struct device *dev) -{ - int retval; - - if (dev->power.memalloc_noio) { - unsigned int noio_flag; - - /* - * Deadlock might be caused if memory allocation with - * GFP_KERNEL happens inside runtime_suspend and - * runtime_resume callbacks of one block device's - * ancestor or the block device itself. Network - * device might be thought as part of iSCSI block - * device, so network device and its ancestor should - * be marked as memalloc_noio too. - */ - noio_flag = memalloc_noio_save(); - retval = __rpm_callback(cb, dev); - memalloc_noio_restore(noio_flag); - } else { - retval = __rpm_callback(cb, dev); - } - - dev->power.runtime_error = retval; - return retval != -EACCES ? retval : -EIO; -} - -/** * rpm_suspend - Carry out runtime suspend of given device. * @dev: Device to suspend. * @rpmflags: Flag bits. @@ -536,7 +590,7 @@ static int rpm_suspend(struct device *dev, int rpmflags) struct device *parent = NULL; int retval; - trace_rpm_suspend_rcuidle(dev, rpmflags); + trace_rpm_suspend(dev, rpmflags); repeat: retval = rpm_check_suspend_allowed(dev); @@ -546,12 +600,12 @@ static int rpm_suspend(struct device *dev, int rpmflags) /* Synchronous suspends are not allowed in the RPM_RESUMING state. */ if (dev->power.runtime_status == RPM_RESUMING && !(rpmflags & RPM_ASYNC)) retval = -EAGAIN; + if (retval) goto out; /* If the autosuspend_delay time hasn't expired yet, reschedule. */ - if ((rpmflags & RPM_AUTO) - && dev->power.runtime_status != RPM_SUSPENDING) { + if ((rpmflags & RPM_AUTO) && dev->power.runtime_status != RPM_SUSPENDING) { u64 expires = pm_runtime_autosuspend_expiration(dev); if (expires != 0) { @@ -566,7 +620,7 @@ static int rpm_suspend(struct device *dev, int rpmflags) * rest. */ if (!(dev->power.timer_expires && - dev->power.timer_expires <= expires)) { + dev->power.timer_expires <= expires)) { /* * We add a slack of 25% to gather wakeups * without sacrificing the granularity. @@ -576,9 +630,9 @@ static int rpm_suspend(struct device *dev, int rpmflags) dev->power.timer_expires = expires; hrtimer_start_range_ns(&dev->power.suspend_timer, - ns_to_ktime(expires), - slack, - HRTIMER_MODE_ABS); + ns_to_ktime(expires), + slack, + HRTIMER_MODE_ABS); } dev->power.timer_autosuspends = 1; goto out; @@ -645,6 +699,8 @@ static int rpm_suspend(struct device *dev, int rpmflags) if (retval) goto fail; + dev_pm_enable_wake_irq_complete(dev); + no_callback: __update_runtime_status(dev, RPM_SUSPENDED); pm_runtime_deactivate_timer(dev); @@ -685,31 +741,28 @@ static int rpm_suspend(struct device *dev, int rpmflags) } out: - trace_rpm_return_int_rcuidle(dev, _THIS_IP_, retval); + trace_rpm_return_int(dev, _THIS_IP_, retval); return retval; fail: - dev_pm_disable_wake_irq_check(dev); + dev_pm_disable_wake_irq_check(dev, true); __update_runtime_status(dev, RPM_ACTIVE); dev->power.deferred_resume = false; wake_up_all(&dev->power.wait_queue); - if (retval == -EAGAIN || retval == -EBUSY) { - dev->power.runtime_error = 0; + /* + * On transient errors, if the callback routine failed an autosuspend, + * and if the last_busy time has been updated so that there is a new + * autosuspend expiration time, automatically reschedule another + * autosuspend. + */ + if (!dev->power.runtime_error && (rpmflags & RPM_AUTO) && + pm_runtime_autosuspend_expiration(dev) != 0) + goto repeat; + + pm_runtime_cancel_pending(dev); - /* - * If the callback routine failed an autosuspend, and - * if the last_busy time has been updated so that there - * is a new autosuspend expiration time, automatically - * reschedule another autosuspend. - */ - if ((rpmflags & RPM_AUTO) && - pm_runtime_autosuspend_expiration(dev) != 0) - goto repeat; - } else { - pm_runtime_cancel_pending(dev); - } goto out; } @@ -737,16 +790,20 @@ static int rpm_resume(struct device *dev, int rpmflags) struct device *parent = NULL; int retval = 0; - trace_rpm_resume_rcuidle(dev, rpmflags); + trace_rpm_resume(dev, rpmflags); repeat: - if (dev->power.runtime_error) + if (dev->power.runtime_error) { retval = -EINVAL; - else if (dev->power.disable_depth == 1 && dev->power.is_suspended - && dev->power.runtime_status == RPM_ACTIVE) - retval = 1; - else if (dev->power.disable_depth > 0) - retval = -EACCES; + } else if (dev->power.disable_depth > 0) { + if (dev->power.runtime_status == RPM_ACTIVE && + dev->power.last_status == RPM_ACTIVE) + retval = 1; + else if (rpmflags & RPM_TRANSPARENT) + goto out; + else + retval = -EACCES; + } if (retval) goto out; @@ -765,15 +822,18 @@ static int rpm_resume(struct device *dev, int rpmflags) goto out; } - if (dev->power.runtime_status == RPM_RESUMING - || dev->power.runtime_status == RPM_SUSPENDING) { + if (dev->power.runtime_status == RPM_RESUMING || + dev->power.runtime_status == RPM_SUSPENDING) { DEFINE_WAIT(wait); if (rpmflags & (RPM_ASYNC | RPM_NOWAIT)) { - if (dev->power.runtime_status == RPM_SUSPENDING) + if (dev->power.runtime_status == RPM_SUSPENDING) { dev->power.deferred_resume = true; - else + if (rpmflags & RPM_NOWAIT) + retval = -EINPROGRESS; + } else { retval = -EINPROGRESS; + } goto out; } @@ -790,8 +850,8 @@ static int rpm_resume(struct device *dev, int rpmflags) for (;;) { prepare_to_wait(&dev->power.wait_queue, &wait, TASK_UNINTERRUPTIBLE); - if (dev->power.runtime_status != RPM_RESUMING - && dev->power.runtime_status != RPM_SUSPENDING) + if (dev->power.runtime_status != RPM_RESUMING && + dev->power.runtime_status != RPM_SUSPENDING) break; spin_unlock_irq(&dev->power.lock); @@ -811,9 +871,9 @@ static int rpm_resume(struct device *dev, int rpmflags) */ if (dev->power.no_callbacks && !parent && dev->parent) { spin_lock_nested(&dev->parent->power.lock, SINGLE_DEPTH_NESTING); - if (dev->parent->power.disable_depth > 0 - || dev->parent->power.ignore_children - || dev->parent->power.runtime_status == RPM_ACTIVE) { + if (dev->parent->power.disable_depth > 0 || + dev->parent->power.ignore_children || + dev->parent->power.runtime_status == RPM_ACTIVE) { atomic_inc(&dev->parent->power.child_count); spin_unlock(&dev->parent->power.lock); retval = 1; @@ -842,6 +902,7 @@ static int rpm_resume(struct device *dev, int rpmflags) parent = dev->parent; if (dev->power.irq_safe) goto skip_parent; + spin_unlock(&dev->power.lock); pm_runtime_get_noresume(parent); @@ -851,8 +912,8 @@ static int rpm_resume(struct device *dev, int rpmflags) * Resume the parent if it has runtime PM enabled and not been * set to ignore its children. */ - if (!parent->power.disable_depth - && !parent->power.ignore_children) { + if (!parent->power.disable_depth && + !parent->power.ignore_children) { rpm_resume(parent, 0); if (parent->power.runtime_status != RPM_ACTIVE) retval = -EBUSY; @@ -862,6 +923,7 @@ static int rpm_resume(struct device *dev, int rpmflags) spin_lock(&dev->power.lock); if (retval) goto out; + goto repeat; } skip_parent: @@ -873,7 +935,7 @@ static int rpm_resume(struct device *dev, int rpmflags) callback = RPM_GET_CALLBACK(dev, runtime_resume); - dev_pm_disable_wake_irq_check(dev); + dev_pm_disable_wake_irq_check(dev, false); retval = rpm_callback(callback, dev); if (retval) { __update_runtime_status(dev, RPM_SUSPENDED); @@ -900,7 +962,7 @@ static int rpm_resume(struct device *dev, int rpmflags) spin_lock_irq(&dev->power.lock); } - trace_rpm_return_int_rcuidle(dev, _THIS_IP_, retval); + trace_rpm_return_int(dev, _THIS_IP_, retval); return retval; } @@ -966,7 +1028,7 @@ static enum hrtimer_restart pm_suspend_timer_fn(struct hrtimer *timer) * If 'expires' is after the current time, we've been called * too early. */ - if (expires > 0 && expires < ktime_get_mono_fast_ns()) { + if (expires > 0 && expires <= ktime_get_mono_fast_ns()) { dev->power.timer_expires = 0; rpm_suspend(dev, dev->power.timer_autosuspends ? (RPM_ASYNC | RPM_AUTO) : RPM_ASYNC); @@ -1014,13 +1076,33 @@ int pm_schedule_suspend(struct device *dev, unsigned int delay) } EXPORT_SYMBOL_GPL(pm_schedule_suspend); +static int rpm_drop_usage_count(struct device *dev) +{ + int ret; + + ret = atomic_sub_return(1, &dev->power.usage_count); + if (ret >= 0) + return ret; + + /* + * Because rpm_resume() does not check the usage counter, it will resume + * the device even if the usage counter is 0 or negative, so it is + * sufficient to increment the usage counter here to reverse the change + * made above. + */ + atomic_inc(&dev->power.usage_count); + dev_warn(dev, "Runtime PM usage count underflow!\n"); + return -EINVAL; +} + /** * __pm_runtime_idle - Entry point for runtime idle operations. * @dev: Device to send idle notification for. * @rpmflags: Flag bits. * * If the RPM_GET_PUT flag is set, decrement the device's usage count and - * return immediately if it is larger than zero. Then carry out an idle + * return immediately if it is larger than zero (if it becomes negative, log a + * warning, increment it, and return an error). Then carry out an idle * notification, either synchronous or asynchronous. * * This routine may be called in atomic context if the RPM_ASYNC flag is set, @@ -1032,8 +1114,11 @@ int __pm_runtime_idle(struct device *dev, int rpmflags) int retval; if (rpmflags & RPM_GET_PUT) { - if (!atomic_dec_and_test(&dev->power.usage_count)) { - trace_rpm_usage_rcuidle(dev, rpmflags); + retval = rpm_drop_usage_count(dev); + if (retval < 0) { + return retval; + } else if (retval > 0) { + trace_rpm_usage(dev, rpmflags); return 0; } } @@ -1054,7 +1139,8 @@ EXPORT_SYMBOL_GPL(__pm_runtime_idle); * @rpmflags: Flag bits. * * If the RPM_GET_PUT flag is set, decrement the device's usage count and - * return immediately if it is larger than zero. Then carry out a suspend, + * return immediately if it is larger than zero (if it becomes negative, log a + * warning, increment it, and return an error). Then carry out a suspend, * either synchronous or asynchronous. * * This routine may be called in atomic context if the RPM_ASYNC flag is set, @@ -1066,8 +1152,11 @@ int __pm_runtime_suspend(struct device *dev, int rpmflags) int retval; if (rpmflags & RPM_GET_PUT) { - if (!atomic_dec_and_test(&dev->power.usage_count)) { - trace_rpm_usage_rcuidle(dev, rpmflags); + retval = rpm_drop_usage_count(dev); + if (retval < 0) { + return retval; + } else if (retval > 0) { + trace_rpm_usage(dev, rpmflags); return 0; } } @@ -1113,16 +1202,18 @@ int __pm_runtime_resume(struct device *dev, int rpmflags) EXPORT_SYMBOL_GPL(__pm_runtime_resume); /** - * pm_runtime_get_if_active - Conditionally bump up device usage counter. + * pm_runtime_get_conditional - Conditionally bump up device usage counter. * @dev: Device to handle. * @ign_usage_count: Whether or not to look at the current usage counter value. * * Return -EINVAL if runtime PM is disabled for @dev. * - * Otherwise, if the runtime PM status of @dev is %RPM_ACTIVE and either - * @ign_usage_count is %true or the runtime PM usage counter of @dev is not - * zero, increment the usage counter of @dev and return 1. Otherwise, return 0 - * without changing the usage counter. + * Otherwise, if its runtime PM status is %RPM_ACTIVE and (1) @ign_usage_count + * is set, or (2) @dev is not ignoring children and its active child count is + * nonzero, or (3) the runtime PM usage counter of @dev is not zero, increment + * the usage counter of @dev and return 1. + * + * Otherwise, return 0 without changing the usage counter. * * If @ign_usage_count is %true, this function can be used to prevent suspending * the device when its runtime PM status is %RPM_ACTIVE. @@ -1134,7 +1225,7 @@ EXPORT_SYMBOL_GPL(__pm_runtime_resume); * The caller is responsible for decrementing the runtime PM usage counter of * @dev after this function has returned a positive value for it. */ -int pm_runtime_get_if_active(struct device *dev, bool ign_usage_count) +static int pm_runtime_get_conditional(struct device *dev, bool ign_usage_count) { unsigned long flags; int retval; @@ -1144,20 +1235,58 @@ int pm_runtime_get_if_active(struct device *dev, bool ign_usage_count) retval = -EINVAL; } else if (dev->power.runtime_status != RPM_ACTIVE) { retval = 0; - } else if (ign_usage_count) { + } else if (ign_usage_count || (!dev->power.ignore_children && + atomic_read(&dev->power.child_count) > 0)) { retval = 1; atomic_inc(&dev->power.usage_count); } else { retval = atomic_inc_not_zero(&dev->power.usage_count); } - trace_rpm_usage_rcuidle(dev, 0); + trace_rpm_usage(dev, 0); spin_unlock_irqrestore(&dev->power.lock, flags); return retval; } + +/** + * pm_runtime_get_if_active - Bump up runtime PM usage counter if the device is + * in active state + * @dev: Target device. + * + * Increment the runtime PM usage counter of @dev if its runtime PM status is + * %RPM_ACTIVE, in which case it returns 1. If the device is in a different + * state, 0 is returned. -EINVAL is returned if runtime PM is disabled for the + * device, in which case also the usage_count will remain unmodified. + */ +int pm_runtime_get_if_active(struct device *dev) +{ + return pm_runtime_get_conditional(dev, true); +} EXPORT_SYMBOL_GPL(pm_runtime_get_if_active); /** + * pm_runtime_get_if_in_use - Conditionally bump up runtime PM usage counter. + * @dev: Target device. + * + * Increment the runtime PM usage counter of @dev if its runtime PM status is + * %RPM_ACTIVE and its runtime PM usage counter is greater than 0 or it is not + * ignoring children and its active child count is nonzero. 1 is returned in + * this case. + * + * If @dev is in a different state or it is not in use (that is, its usage + * counter is 0, or it is ignoring children, or its active child count is 0), + * 0 is returned. + * + * -EINVAL is returned if runtime PM is disabled for the device, in which case + * also the usage counter of @dev is not updated. + */ +int pm_runtime_get_if_in_use(struct device *dev) +{ + return pm_runtime_get_conditional(dev, false); +} +EXPORT_SYMBOL_GPL(pm_runtime_get_if_in_use); + +/** * __pm_runtime_set_status - Set runtime PM status of a device. * @dev: Device to handle. * @status: New runtime PM status of the device. @@ -1185,12 +1314,13 @@ int __pm_runtime_set_status(struct device *dev, unsigned int status) { struct device *parent = dev->parent; bool notify_parent = false; + unsigned long flags; int error = 0; if (status != RPM_ACTIVE && status != RPM_SUSPENDED) return -EINVAL; - spin_lock_irq(&dev->power.lock); + spin_lock_irqsave(&dev->power.lock, flags); /* * Prevent PM-runtime from being enabled for the device or return an @@ -1201,7 +1331,7 @@ int __pm_runtime_set_status(struct device *dev, unsigned int status) else error = -EAGAIN; - spin_unlock_irq(&dev->power.lock); + spin_unlock_irqrestore(&dev->power.lock, flags); if (error) return error; @@ -1222,7 +1352,7 @@ int __pm_runtime_set_status(struct device *dev, unsigned int status) device_links_read_unlock(idx); } - spin_lock_irq(&dev->power.lock); + spin_lock_irqsave(&dev->power.lock, flags); if (dev->power.runtime_status == status || !parent) goto out_set; @@ -1238,9 +1368,9 @@ int __pm_runtime_set_status(struct device *dev, unsigned int status) * not active, has runtime PM enabled and the * 'power.ignore_children' flag unset. */ - if (!parent->power.disable_depth - && !parent->power.ignore_children - && parent->power.runtime_status != RPM_ACTIVE) { + if (!parent->power.disable_depth && + !parent->power.ignore_children && + parent->power.runtime_status != RPM_ACTIVE) { dev_err(dev, "runtime PM trying to activate child device %s but parent (%s) is not active\n", dev_name(dev), dev_name(parent)); @@ -1263,7 +1393,7 @@ int __pm_runtime_set_status(struct device *dev, unsigned int status) dev->power.runtime_error = 0; out: - spin_unlock_irq(&dev->power.lock); + spin_unlock_irqrestore(&dev->power.lock, flags); if (notify_parent) pm_request_idle(parent); @@ -1305,9 +1435,9 @@ static void __pm_runtime_barrier(struct device *dev) dev->power.request_pending = false; } - if (dev->power.runtime_status == RPM_SUSPENDING - || dev->power.runtime_status == RPM_RESUMING - || dev->power.idle_notification) { + if (dev->power.runtime_status == RPM_SUSPENDING || + dev->power.runtime_status == RPM_RESUMING || + dev->power.idle_notification) { DEFINE_WAIT(wait); /* Suspend, wake-up or idle notification in progress. */ @@ -1337,47 +1467,48 @@ static void __pm_runtime_barrier(struct device *dev) * Next, make sure that all pending requests for the device have been flushed * from pm_wq and wait for all runtime PM operations involving the device in * progress to complete. - * - * Return value: - * 1, if there was a resume request pending and the device had to be woken up, - * 0, otherwise */ -int pm_runtime_barrier(struct device *dev) +void pm_runtime_barrier(struct device *dev) { - int retval = 0; - pm_runtime_get_noresume(dev); spin_lock_irq(&dev->power.lock); if (dev->power.request_pending - && dev->power.request == RPM_REQ_RESUME) { + && dev->power.request == RPM_REQ_RESUME) rpm_resume(dev, 0); - retval = 1; - } __pm_runtime_barrier(dev); spin_unlock_irq(&dev->power.lock); pm_runtime_put_noidle(dev); - - return retval; } EXPORT_SYMBOL_GPL(pm_runtime_barrier); -/** - * __pm_runtime_disable - Disable runtime PM of a device. - * @dev: Device to handle. - * @check_resume: If set, check if there's a resume request for the device. - * - * Increment power.disable_depth for the device and if it was zero previously, - * cancel all pending runtime PM requests for the device and wait for all - * operations in progress to complete. The device can be either active or - * suspended after its runtime PM has been disabled. - * - * If @check_resume is set and there's a resume request pending when - * __pm_runtime_disable() is called and power.disable_depth is zero, the - * function will wake up the device before disabling its runtime PM. - */ +bool pm_runtime_block_if_disabled(struct device *dev) +{ + bool ret; + + spin_lock_irq(&dev->power.lock); + + ret = !pm_runtime_enabled(dev); + if (ret && dev->power.last_status == RPM_INVALID) + dev->power.last_status = RPM_BLOCKED; + + spin_unlock_irq(&dev->power.lock); + + return ret; +} + +void pm_runtime_unblock(struct device *dev) +{ + spin_lock_irq(&dev->power.lock); + + if (dev->power.last_status == RPM_BLOCKED) + dev->power.last_status = RPM_INVALID; + + spin_unlock_irq(&dev->power.lock); +} + void __pm_runtime_disable(struct device *dev, bool check_resume) { spin_lock_irq(&dev->power.lock); @@ -1392,8 +1523,8 @@ void __pm_runtime_disable(struct device *dev, bool check_resume) * means there probably is some I/O to process and disabling runtime PM * shouldn't prevent the device from processing the I/O. */ - if (check_resume && dev->power.request_pending - && dev->power.request == RPM_REQ_RESUME) { + if (check_resume && dev->power.request_pending && + dev->power.request == RPM_REQ_RESUME) { /* * Prevent suspends and idle notifications from being carried * out after we have woken up the device. @@ -1408,8 +1539,10 @@ void __pm_runtime_disable(struct device *dev, bool check_resume) /* Update time accounting before disabling PM-runtime. */ update_pm_runtime_accounting(dev); - if (!dev->power.disable_depth++) + if (!dev->power.disable_depth++) { __pm_runtime_barrier(dev); + dev->power.last_status = dev->power.runtime_status; + } out: spin_unlock_irq(&dev->power.lock); @@ -1426,34 +1559,107 @@ void pm_runtime_enable(struct device *dev) spin_lock_irqsave(&dev->power.lock, flags); - if (dev->power.disable_depth > 0) { - dev->power.disable_depth--; - - /* About to enable runtime pm, set accounting_timestamp to now */ - if (!dev->power.disable_depth) - dev->power.accounting_timestamp = ktime_get_mono_fast_ns(); - } else { + if (!dev->power.disable_depth) { dev_warn(dev, "Unbalanced %s!\n", __func__); + goto out; + } + + if (--dev->power.disable_depth > 0) + goto out; + + if (dev->power.last_status == RPM_BLOCKED) { + dev_warn(dev, "Attempt to enable runtime PM when it is blocked\n"); + dump_stack(); } + dev->power.last_status = RPM_INVALID; + dev->power.accounting_timestamp = ktime_get_mono_fast_ns(); - WARN(!dev->power.disable_depth && - dev->power.runtime_status == RPM_SUSPENDED && - !dev->power.ignore_children && - atomic_read(&dev->power.child_count) > 0, - "Enabling runtime PM for inactive device (%s) with active children\n", - dev_name(dev)); + if (dev->power.runtime_status == RPM_SUSPENDED && + !dev->power.ignore_children && + atomic_read(&dev->power.child_count) > 0) + dev_warn(dev, "Enabling runtime PM for inactive device with active children\n"); +out: spin_unlock_irqrestore(&dev->power.lock, flags); } EXPORT_SYMBOL_GPL(pm_runtime_enable); +static void pm_runtime_set_suspended_action(void *data) +{ + pm_runtime_set_suspended(data); +} + +/** + * devm_pm_runtime_set_active_enabled - set_active version of devm_pm_runtime_enable. + * + * @dev: Device to handle. + */ +int devm_pm_runtime_set_active_enabled(struct device *dev) +{ + int err; + + err = pm_runtime_set_active(dev); + if (err) + return err; + + err = devm_add_action_or_reset(dev, pm_runtime_set_suspended_action, dev); + if (err) + return err; + + return devm_pm_runtime_enable(dev); +} +EXPORT_SYMBOL_GPL(devm_pm_runtime_set_active_enabled); + +static void pm_runtime_disable_action(void *data) +{ + pm_runtime_dont_use_autosuspend(data); + pm_runtime_disable(data); +} + +/** + * devm_pm_runtime_enable - devres-enabled version of pm_runtime_enable. + * + * NOTE: this will also handle calling pm_runtime_dont_use_autosuspend() for + * you at driver exit time if needed. + * + * @dev: Device to handle. + */ +int devm_pm_runtime_enable(struct device *dev) +{ + pm_runtime_enable(dev); + + return devm_add_action_or_reset(dev, pm_runtime_disable_action, dev); +} +EXPORT_SYMBOL_GPL(devm_pm_runtime_enable); + +static void pm_runtime_put_noidle_action(void *data) +{ + pm_runtime_put_noidle(data); +} + +/** + * devm_pm_runtime_get_noresume - devres-enabled version of pm_runtime_get_noresume. + * + * @dev: Device to handle. + */ +int devm_pm_runtime_get_noresume(struct device *dev) +{ + pm_runtime_get_noresume(dev); + + return devm_add_action_or_reset(dev, pm_runtime_put_noidle_action, dev); +} +EXPORT_SYMBOL_GPL(devm_pm_runtime_get_noresume); + /** * pm_runtime_forbid - Block runtime PM of a device. * @dev: Device to handle. * - * Increase the device's usage count and clear its power.runtime_auto flag, - * so that it cannot be suspended at run time until pm_runtime_allow() is called - * for it. + * Resume @dev if already suspended and block runtime suspend of @dev in such + * a way that it can be unblocked via the /sys/devices/.../power/control + * interface, or otherwise by calling pm_runtime_allow(). + * + * Calling this function many times in a row has the same effect as calling it + * once. */ void pm_runtime_forbid(struct device *dev) { @@ -1474,19 +1680,28 @@ EXPORT_SYMBOL_GPL(pm_runtime_forbid); * pm_runtime_allow - Unblock runtime PM of a device. * @dev: Device to handle. * - * Decrease the device's usage count and set its power.runtime_auto flag. + * Unblock runtime suspend of @dev after it has been blocked by + * pm_runtime_forbid() (for instance, if it has been blocked via the + * /sys/devices/.../power/control interface), check if @dev can be + * suspended and suspend it in that case. + * + * Calling this function many times in a row has the same effect as calling it + * once. */ void pm_runtime_allow(struct device *dev) { + int ret; + spin_lock_irq(&dev->power.lock); if (dev->power.runtime_auto) goto out; dev->power.runtime_auto = true; - if (atomic_dec_and_test(&dev->power.usage_count)) + ret = rpm_drop_usage_count(dev); + if (ret == 0) rpm_idle(dev, RPM_AUTO | RPM_ASYNC); - else - trace_rpm_usage_rcuidle(dev, RPM_AUTO | RPM_ASYNC); + else if (ret > 0) + trace_rpm_usage(dev, RPM_AUTO | RPM_ASYNC); out: spin_unlock_irq(&dev->power.lock); @@ -1526,6 +1741,7 @@ void pm_runtime_irq_safe(struct device *dev) { if (dev->parent) pm_runtime_get_sync(dev->parent); + spin_lock_irq(&dev->power.lock); dev->power.irq_safe = 1; spin_unlock_irq(&dev->power.lock); @@ -1555,7 +1771,7 @@ static void update_autosuspend(struct device *dev, int old_delay, int old_use) atomic_inc(&dev->power.usage_count); rpm_resume(dev, 0); } else { - trace_rpm_usage_rcuidle(dev, 0); + trace_rpm_usage(dev, 0); } } @@ -1621,6 +1837,7 @@ EXPORT_SYMBOL_GPL(__pm_runtime_use_autosuspend); void pm_runtime_init(struct device *dev) { dev->power.runtime_status = RPM_SUSPENDED; + dev->power.last_status = RPM_INVALID; dev->power.idle_notification = false; dev->power.disable_depth = 1; @@ -1635,12 +1852,12 @@ void pm_runtime_init(struct device *dev) dev->power.request_pending = false; dev->power.request = RPM_REQ_NONE; dev->power.deferred_resume = false; - dev->power.needs_force_resume = 0; + dev->power.needs_force_resume = false; INIT_WORK(&dev->power.work, pm_runtime_work); dev->power.timer_expires = 0; - hrtimer_init(&dev->power.suspend_timer, CLOCK_MONOTONIC, HRTIMER_MODE_ABS); - dev->power.suspend_timer.function = pm_suspend_timer_fn; + hrtimer_setup(&dev->power.suspend_timer, pm_suspend_timer_fn, CLOCK_MONOTONIC, + HRTIMER_MODE_ABS); init_waitqueue_head(&dev->power.wait_queue); } @@ -1662,6 +1879,11 @@ void pm_runtime_reinit(struct device *dev) pm_runtime_put(dev->parent); } } + /* + * Clear power.needs_force_resume in case it has been set by + * pm_runtime_force_suspend() invoked from a driver remove callback. + */ + dev->power.needs_force_resume = false; } /** @@ -1685,12 +1907,10 @@ void pm_runtime_get_suppliers(struct device *dev) idx = device_links_read_lock(); - list_for_each_entry_rcu(link, &dev->links.suppliers, c_node, - device_links_read_lock_held()) - if (link->flags & DL_FLAG_PM_RUNTIME) { + dev_for_each_link_to_supplier(link, dev) + if (device_link_test(link, DL_FLAG_PM_RUNTIME)) { link->supplier_preactivated = true; pm_runtime_get_sync(link->supplier); - refcount_inc(&link->rpm_active); } device_links_read_unlock(idx); @@ -1703,8 +1923,6 @@ void pm_runtime_get_suppliers(struct device *dev) void pm_runtime_put_suppliers(struct device *dev) { struct device_link *link; - unsigned long flags; - bool put; int idx; idx = device_links_read_lock(); @@ -1713,12 +1931,7 @@ void pm_runtime_put_suppliers(struct device *dev) device_links_read_lock_held()) if (link->supplier_preactivated) { link->supplier_preactivated = false; - spin_lock_irqsave(&dev->power.lock, flags); - put = pm_runtime_status_suspended(dev) && - refcount_dec_not_one(&link->rpm_active); - spin_unlock_irqrestore(&dev->power.lock, flags); - if (put) - pm_runtime_put(link->supplier); + pm_runtime_put(link->supplier); } device_links_read_unlock(idx); @@ -1749,22 +1962,31 @@ static void pm_runtime_drop_link_count(struct device *dev) */ void pm_runtime_drop_link(struct device_link *link) { - if (!(link->flags & DL_FLAG_PM_RUNTIME)) + if (!device_link_test(link, DL_FLAG_PM_RUNTIME)) return; pm_runtime_drop_link_count(link->consumer); - - while (refcount_dec_not_one(&link->rpm_active)) - pm_runtime_put(link->supplier); + pm_runtime_release_supplier(link); + pm_request_idle(link->supplier); } -static bool pm_runtime_need_not_resume(struct device *dev) +static pm_callback_t get_callback(struct device *dev, size_t cb_offset) { - return atomic_read(&dev->power.usage_count) <= 1 && - (atomic_read(&dev->power.child_count) == 0 || - dev->power.ignore_children); + /* + * Setting power.strict_midlayer means that the middle layer + * code does not want its runtime PM callbacks to be invoked via + * pm_runtime_force_suspend() and pm_runtime_force_resume(), so + * return a direct pointer to the driver callback in that case. + */ + if (dev_pm_strict_midlayer_is_set(dev)) + return __rpm_get_driver_callback(dev, cb_offset); + + return __rpm_get_callback(dev, cb_offset); } +#define GET_CALLBACK(dev, callback) \ + get_callback(dev, offsetof(struct dev_pm_ops, callback)) + /** * pm_runtime_force_suspend - Force a device into suspend state if needed. * @dev: Device to suspend. @@ -1788,74 +2010,106 @@ int pm_runtime_force_suspend(struct device *dev) int ret; pm_runtime_disable(dev); - if (pm_runtime_status_suspended(dev)) + if (pm_runtime_status_suspended(dev) || dev->power.needs_force_resume) return 0; - callback = RPM_GET_CALLBACK(dev, runtime_suspend); + callback = GET_CALLBACK(dev, runtime_suspend); + dev_pm_enable_wake_irq_check(dev, true); ret = callback ? callback(dev) : 0; if (ret) goto err; + dev_pm_enable_wake_irq_complete(dev); + /* * If the device can stay in suspend after the system-wide transition * to the working state that will follow, drop the children counter of - * its parent, but set its status to RPM_SUSPENDED anyway in case this - * function will be called again for it in the meantime. + * its parent and the usage counters of its suppliers. Otherwise, set + * power.needs_force_resume to let pm_runtime_force_resume() know that + * the device needs to be taken care of and to prevent this function + * from handling the device again in case the device is passed to it + * once more subsequently. */ - if (pm_runtime_need_not_resume(dev)) { + if (pm_runtime_need_not_resume(dev)) pm_runtime_set_suspended(dev); - } else { - __update_runtime_status(dev, RPM_SUSPENDED); - dev->power.needs_force_resume = 1; - } + else + dev->power.needs_force_resume = true; return 0; err: + dev_pm_disable_wake_irq_check(dev, true); pm_runtime_enable(dev); return ret; } EXPORT_SYMBOL_GPL(pm_runtime_force_suspend); +#ifdef CONFIG_PM_SLEEP + /** * pm_runtime_force_resume - Force a device into resume state if needed. * @dev: Device to resume. * - * Prior invoking this function we expect the user to have brought the device - * into low power state by a call to pm_runtime_force_suspend(). Here we reverse - * those actions and bring the device into full power, if it is expected to be - * used on system resume. In the other case, we defer the resume to be managed - * via runtime PM. + * This function expects that either pm_runtime_force_suspend() has put the + * device into a low-power state prior to calling it, or the device had been + * runtime-suspended before the preceding system-wide suspend transition and it + * was left in suspend during that transition. * - * Typically this function may be invoked from a system resume callback. + * The actions carried out by pm_runtime_force_suspend(), or by a runtime + * suspend in general, are reversed and the device is brought back into full + * power if it is expected to be used on system resume, which is the case when + * its needs_force_resume flag is set or when its smart_suspend flag is set and + * its runtime PM status is "active". + * + * In other cases, the resume is deferred to be managed via runtime PM. + * + * Typically, this function may be invoked from a system resume callback. */ int pm_runtime_force_resume(struct device *dev) { int (*callback)(struct device *); int ret = 0; - if (!pm_runtime_status_suspended(dev) || !dev->power.needs_force_resume) + if (!dev->power.needs_force_resume && (!dev_pm_smart_suspend(dev) || + pm_runtime_status_suspended(dev))) goto out; - /* - * The value of the parent's children counter is correct already, so - * just update the status of the device. - */ - __update_runtime_status(dev, RPM_ACTIVE); - - callback = RPM_GET_CALLBACK(dev, runtime_resume); + callback = GET_CALLBACK(dev, runtime_resume); + dev_pm_disable_wake_irq_check(dev, false); ret = callback ? callback(dev) : 0; if (ret) { pm_runtime_set_suspended(dev); + dev_pm_enable_wake_irq_check(dev, false); goto out; } pm_runtime_mark_last_busy(dev); + out: - dev->power.needs_force_resume = 0; + /* + * The smart_suspend flag can be cleared here because it is not going + * to be necessary until the next system-wide suspend transition that + * will update it again. + */ + dev->power.smart_suspend = false; + /* + * Also clear needs_force_resume to make this function skip devices that + * have been seen by it once. + */ + dev->power.needs_force_resume = false; + pm_runtime_enable(dev); return ret; } EXPORT_SYMBOL_GPL(pm_runtime_force_resume); + +bool pm_runtime_need_not_resume(struct device *dev) +{ + return atomic_read(&dev->power.usage_count) <= 1 && + (atomic_read(&dev->power.child_count) == 0 || + dev->power.ignore_children); +} + +#endif /* CONFIG_PM_SLEEP */ diff --git a/drivers/base/power/sysfs.c b/drivers/base/power/sysfs.c index a1474fb67db9..13b31a3adc77 100644 --- a/drivers/base/power/sysfs.c +++ b/drivers/base/power/sysfs.c @@ -6,7 +6,6 @@ #include <linux/export.h> #include <linux/pm_qos.h> #include <linux/pm_runtime.h> -#include <linux/pm_wakeup.h> #include <linux/atomic.h> #include <linux/jiffies.h> #include "power.h" @@ -509,14 +508,6 @@ static ssize_t wakeup_last_time_ms_show(struct device *dev, return sysfs_emit(buf, "%lld\n", msec); } -static inline int dpm_sysfs_wakeup_change_owner(struct device *dev, kuid_t kuid, - kgid_t kgid) -{ - if (dev->power.wakeup && dev->power.wakeup->dev) - return device_change_owner(dev->power.wakeup->dev, kuid, kgid); - return 0; -} - static DEVICE_ATTR_RO(wakeup_last_time_ms); #ifdef CONFIG_PM_AUTOSLEEP @@ -541,6 +532,15 @@ static ssize_t wakeup_prevent_sleep_time_ms_show(struct device *dev, static DEVICE_ATTR_RO(wakeup_prevent_sleep_time_ms); #endif /* CONFIG_PM_AUTOSLEEP */ + +static inline int dpm_sysfs_wakeup_change_owner(struct device *dev, kuid_t kuid, + kgid_t kgid) +{ + if (dev->power.wakeup && dev->power.wakeup->dev) + return device_change_owner(dev->power.wakeup->dev, kuid, kgid); + return 0; +} + #else /* CONFIG_PM_SLEEP */ static inline int dpm_sysfs_wakeup_change_owner(struct device *dev, kuid_t kuid, kgid_t kgid) @@ -611,15 +611,9 @@ static DEVICE_ATTR_RW(async); #endif /* CONFIG_PM_ADVANCED_DEBUG */ static struct attribute *power_attrs[] = { -#ifdef CONFIG_PM_ADVANCED_DEBUG -#ifdef CONFIG_PM_SLEEP +#if defined(CONFIG_PM_ADVANCED_DEBUG) && defined(CONFIG_PM_SLEEP) &dev_attr_async.attr, #endif - &dev_attr_runtime_status.attr, - &dev_attr_runtime_usage.attr, - &dev_attr_runtime_active_kids.attr, - &dev_attr_runtime_enabled.attr, -#endif /* CONFIG_PM_ADVANCED_DEBUG */ NULL, }; static const struct attribute_group pm_attr_group = { @@ -650,13 +644,16 @@ static const struct attribute_group pm_wakeup_attr_group = { }; static struct attribute *runtime_attrs[] = { -#ifndef CONFIG_PM_ADVANCED_DEBUG &dev_attr_runtime_status.attr, -#endif &dev_attr_control.attr, &dev_attr_runtime_suspended_time.attr, &dev_attr_runtime_active_time.attr, &dev_attr_autosuspend_delay_ms.attr, +#ifdef CONFIG_PM_ADVANCED_DEBUG + &dev_attr_runtime_usage.attr, + &dev_attr_runtime_active_kids.attr, + &dev_attr_runtime_enabled.attr, +#endif NULL, }; static const struct attribute_group pm_runtime_attr_group = { diff --git a/drivers/base/power/trace.c b/drivers/base/power/trace.c index a97f33d0c59f..d8da7195bb00 100644 --- a/drivers/base/power/trace.c +++ b/drivers/base/power/trace.c @@ -13,6 +13,7 @@ #include <linux/export.h> #include <linux/rtc.h> #include <linux/suspend.h> +#include <linux/init.h> #include <linux/mc146818rtc.h> @@ -119,7 +120,11 @@ static unsigned int read_magic_time(void) struct rtc_time time; unsigned int val; - mc146818_get_time(&time); + if (mc146818_get_time(&time, 1000) < 0) { + pr_err("Unable to read current time from RTC\n"); + return 0; + } + pr_info("RTC time: %ptRt, date: %ptRd\n", &time, &time); val = time.tm_year; /* 100 years */ if (val > 100) @@ -165,6 +170,9 @@ void generate_pm_trace(const void *tracedata, unsigned int user) const char *file = *(const char **)(tracedata + 2); unsigned int user_hash_value, file_hash_value; + if (!x86_platform.legacy.rtc) + return; + user_hash_value = user % USERHASH; file_hash_value = hash_string(lineno, file, FILEHASH); set_magic_time(user_hash_value, file_hash_value, dev_hash_value); @@ -230,10 +238,8 @@ int show_trace_dev_match(char *buf, size_t size) unsigned int hash = hash_string(DEVSEED, dev_name(dev), DEVHASH); if (hash == value) { - int len = snprintf(buf, size, "%s\n", + int len = scnprintf(buf, size, "%s\n", dev_driver_string(dev)); - if (len > size) - len = size; buf += len; ret += len; size -= len; @@ -267,6 +273,9 @@ static struct notifier_block pm_trace_nb = { static int __init early_resume_init(void) { + if (!x86_platform.legacy.rtc) + return 0; + hash_value_early_read = read_magic_time(); register_pm_notifier(&pm_trace_nb); return 0; @@ -277,6 +286,9 @@ static int __init late_resume_init(void) unsigned int val = hash_value_early_read; unsigned int user, file, dev; + if (!x86_platform.legacy.rtc) + return 0; + user = val % USERHASH; val = val / USERHASH; file = val % FILEHASH; diff --git a/drivers/base/power/wakeirq.c b/drivers/base/power/wakeirq.c index 3bad3266a2ad..8aa28c08b289 100644 --- a/drivers/base/power/wakeirq.c +++ b/drivers/base/power/wakeirq.c @@ -12,14 +12,11 @@ /** * dev_pm_attach_wake_irq - Attach device interrupt as a wake IRQ * @dev: Device entry - * @irq: Device wake-up capable interrupt * @wirq: Wake irq specific data * - * Internal function to attach either a device IO interrupt or a - * dedicated wake-up interrupt as a wake IRQ. + * Internal function to attach a dedicated wake-up interrupt as a wake IRQ. */ -static int dev_pm_attach_wake_irq(struct device *dev, int irq, - struct wake_irq *wirq) +static int dev_pm_attach_wake_irq(struct device *dev, struct wake_irq *wirq) { unsigned long flags; @@ -65,7 +62,7 @@ int dev_pm_set_wake_irq(struct device *dev, int irq) wirq->dev = dev; wirq->irq = irq; - err = dev_pm_attach_wake_irq(dev, irq, wirq); + err = dev_pm_attach_wake_irq(dev, wirq); if (err) kfree(wirq); @@ -106,6 +103,32 @@ void dev_pm_clear_wake_irq(struct device *dev) } EXPORT_SYMBOL_GPL(dev_pm_clear_wake_irq); +static void devm_pm_clear_wake_irq(void *dev) +{ + dev_pm_clear_wake_irq(dev); +} + +/** + * devm_pm_set_wake_irq - device-managed variant of dev_pm_set_wake_irq + * @dev: Device entry + * @irq: Device IO interrupt + * + * + * Attach a device IO interrupt as a wake IRQ, same with dev_pm_set_wake_irq, + * but the device will be auto clear wake capability on driver detach. + */ +int devm_pm_set_wake_irq(struct device *dev, int irq) +{ + int ret; + + ret = dev_pm_set_wake_irq(dev, irq); + if (ret) + return ret; + + return devm_add_action_or_reset(dev, devm_pm_clear_wake_irq, dev); +} +EXPORT_SYMBOL_GPL(devm_pm_set_wake_irq); + /** * handle_threaded_wake_irq - Handler for dedicated wake-up interrupts * @irq: Device specific dedicated wake-up interrupt @@ -145,24 +168,7 @@ static irqreturn_t handle_threaded_wake_irq(int irq, void *_wirq) return IRQ_HANDLED; } -/** - * dev_pm_set_dedicated_wake_irq - Request a dedicated wake-up interrupt - * @dev: Device entry - * @irq: Device wake-up interrupt - * - * Unless your hardware has separate wake-up interrupts in addition - * to the device IO interrupts, you don't need this. - * - * Sets up a threaded interrupt handler for a device that has - * a dedicated wake-up interrupt in addition to the device IO - * interrupt. - * - * The interrupt starts disabled, and needs to be managed for - * the device by the bus code or the device driver using - * dev_pm_enable_wake_irq() and dev_pm_disable_wake_irq() - * functions. - */ -int dev_pm_set_dedicated_wake_irq(struct device *dev, int irq) +static int __dev_pm_set_dedicated_wake_irq(struct device *dev, int irq, unsigned int flag) { struct wake_irq *wirq; int err; @@ -196,11 +202,11 @@ int dev_pm_set_dedicated_wake_irq(struct device *dev, int irq) if (err) goto err_free_name; - err = dev_pm_attach_wake_irq(dev, irq, wirq); + err = dev_pm_attach_wake_irq(dev, wirq); if (err) goto err_free_irq; - wirq->status = WAKE_IRQ_DEDICATED_ALLOCATED; + wirq->status = WAKE_IRQ_DEDICATED_ALLOCATED | flag; return err; @@ -213,45 +219,45 @@ err_free: return err; } -EXPORT_SYMBOL_GPL(dev_pm_set_dedicated_wake_irq); /** - * dev_pm_enable_wake_irq - Enable device wake-up interrupt - * @dev: Device + * dev_pm_set_dedicated_wake_irq - Request a dedicated wake-up interrupt + * @dev: Device entry + * @irq: Device wake-up interrupt * - * Optionally called from the bus code or the device driver for - * runtime_resume() to override the PM runtime core managed wake-up - * interrupt handling to enable the wake-up interrupt. + * Unless your hardware has separate wake-up interrupts in addition + * to the device IO interrupts, you don't need this. * - * Note that for runtime_suspend()) the wake-up interrupts - * should be unconditionally enabled unlike for suspend() - * that is conditional. + * Sets up a threaded interrupt handler for a device that has + * a dedicated wake-up interrupt in addition to the device IO + * interrupt. */ -void dev_pm_enable_wake_irq(struct device *dev) +int dev_pm_set_dedicated_wake_irq(struct device *dev, int irq) { - struct wake_irq *wirq = dev->power.wakeirq; - - if (wirq && (wirq->status & WAKE_IRQ_DEDICATED_ALLOCATED)) - enable_irq(wirq->irq); + return __dev_pm_set_dedicated_wake_irq(dev, irq, 0); } -EXPORT_SYMBOL_GPL(dev_pm_enable_wake_irq); +EXPORT_SYMBOL_GPL(dev_pm_set_dedicated_wake_irq); /** - * dev_pm_disable_wake_irq - Disable device wake-up interrupt - * @dev: Device + * dev_pm_set_dedicated_wake_irq_reverse - Request a dedicated wake-up interrupt + * with reverse enable ordering + * @dev: Device entry + * @irq: Device wake-up interrupt * - * Optionally called from the bus code or the device driver for - * runtime_suspend() to override the PM runtime core managed wake-up - * interrupt handling to disable the wake-up interrupt. + * Unless your hardware has separate wake-up interrupts in addition + * to the device IO interrupts, you don't need this. + * + * Sets up a threaded interrupt handler for a device that has a dedicated + * wake-up interrupt in addition to the device IO interrupt. It sets + * the status of WAKE_IRQ_DEDICATED_REVERSE to tell rpm_suspend() + * to enable dedicated wake-up interrupt after running the runtime suspend + * callback for @dev. */ -void dev_pm_disable_wake_irq(struct device *dev) +int dev_pm_set_dedicated_wake_irq_reverse(struct device *dev, int irq) { - struct wake_irq *wirq = dev->power.wakeirq; - - if (wirq && (wirq->status & WAKE_IRQ_DEDICATED_ALLOCATED)) - disable_irq_nosync(wirq->irq); + return __dev_pm_set_dedicated_wake_irq(dev, irq, WAKE_IRQ_DEDICATED_REVERSE); } -EXPORT_SYMBOL_GPL(dev_pm_disable_wake_irq); +EXPORT_SYMBOL_GPL(dev_pm_set_dedicated_wake_irq_reverse); /** * dev_pm_enable_wake_irq_check - Checks and enables wake-up interrupt @@ -260,7 +266,7 @@ EXPORT_SYMBOL_GPL(dev_pm_disable_wake_irq); * * Enables wakeirq conditionally. We need to enable wake-up interrupt * lazily on the first rpm_suspend(). This is needed as the consumer device - * starts in RPM_SUSPENDED state, and the the first pm_runtime_get() would + * starts in RPM_SUSPENDED state, and the first pm_runtime_get() would * otherwise try to disable already disabled wakeirq. The wake-up interrupt * starts disabled with IRQ_NOAUTOEN set. * @@ -285,25 +291,58 @@ void dev_pm_enable_wake_irq_check(struct device *dev, return; enable: - enable_irq(wirq->irq); + if (!can_change_status || !(wirq->status & WAKE_IRQ_DEDICATED_REVERSE)) { + enable_irq(wirq->irq); + wirq->status |= WAKE_IRQ_DEDICATED_ENABLED; + } } /** * dev_pm_disable_wake_irq_check - Checks and disables wake-up interrupt * @dev: Device + * @cond_disable: if set, also check WAKE_IRQ_DEDICATED_REVERSE * * Disables wake-up interrupt conditionally based on status. * Should be only called from rpm_suspend() and rpm_resume() path. */ -void dev_pm_disable_wake_irq_check(struct device *dev) +void dev_pm_disable_wake_irq_check(struct device *dev, bool cond_disable) { struct wake_irq *wirq = dev->power.wakeirq; if (!wirq || !(wirq->status & WAKE_IRQ_DEDICATED_MASK)) return; - if (wirq->status & WAKE_IRQ_DEDICATED_MANAGED) + if (cond_disable && (wirq->status & WAKE_IRQ_DEDICATED_REVERSE)) + return; + + if (wirq->status & WAKE_IRQ_DEDICATED_MANAGED) { + wirq->status &= ~WAKE_IRQ_DEDICATED_ENABLED; disable_irq_nosync(wirq->irq); + } +} + +/** + * dev_pm_enable_wake_irq_complete - enable wake IRQ not enabled before + * @dev: Device using the wake IRQ + * + * Enable wake IRQ conditionally based on status, mainly used if want to + * enable wake IRQ after running ->runtime_suspend() which depends on + * WAKE_IRQ_DEDICATED_REVERSE. + * + * Should be only called from rpm_suspend() path. + */ +void dev_pm_enable_wake_irq_complete(struct device *dev) +{ + struct wake_irq *wirq = dev->power.wakeirq; + + if (!wirq || !(wirq->status & WAKE_IRQ_DEDICATED_MASK)) + return; + + if (wirq->status & WAKE_IRQ_DEDICATED_MANAGED && + wirq->status & WAKE_IRQ_DEDICATED_REVERSE) { + enable_irq(wirq->irq); + wirq->status |= WAKE_IRQ_DEDICATED_ENABLED; + } } /** @@ -320,7 +359,7 @@ void dev_pm_arm_wake_irq(struct wake_irq *wirq) if (device_may_wakeup(wirq->dev)) { if (wirq->status & WAKE_IRQ_DEDICATED_ALLOCATED && - !pm_runtime_status_suspended(wirq->dev)) + !(wirq->status & WAKE_IRQ_DEDICATED_ENABLED)) enable_irq(wirq->irq); enable_irq_wake(wirq->irq); @@ -343,7 +382,7 @@ void dev_pm_disarm_wake_irq(struct wake_irq *wirq) disable_irq_wake(wirq->irq); if (wirq->status & WAKE_IRQ_DEDICATED_ALLOCATED && - !pm_runtime_status_suspended(wirq->dev)) + !(wirq->status & WAKE_IRQ_DEDICATED_ENABLED)) disable_irq_nosync(wirq->irq); } } diff --git a/drivers/base/power/wakeup.c b/drivers/base/power/wakeup.c index f0b37c188514..1e1a0e7eeac5 100644 --- a/drivers/base/power/wakeup.c +++ b/drivers/base/power/wakeup.c @@ -19,11 +19,6 @@ #include "power.h" -#ifndef CONFIG_SUSPEND -suspend_state_t pm_suspend_target_state; -#define pm_suspend_target_state (PM_SUSPEND_ON) -#endif - #define list_for_each_entry_rcu_locked(pos, head, member) \ list_for_each_entry_rcu(pos, head, member, \ srcu_read_lock_held(&wakeup_srcu)) @@ -34,7 +29,8 @@ suspend_state_t pm_suspend_target_state; bool events_check_enabled __read_mostly; /* First wakeup IRQ seen by the kernel in the last cycle. */ -unsigned int pm_wakeup_irq __read_mostly; +static unsigned int wakeup_irq[2] __read_mostly; +static DEFINE_RAW_SPINLOCK(wakeup_irq_lock); /* If greater than 0 and the system is suspending, terminate the suspend. */ static atomic_t pm_abort_suspend __read_mostly; @@ -81,7 +77,7 @@ static DEFINE_IDA(wakeup_ida); * wakeup_source_create - Create a struct wakeup_source object. * @name: Name of the new wakeup source. */ -struct wakeup_source *wakeup_source_create(const char *name) +static struct wakeup_source *wakeup_source_create(const char *name) { struct wakeup_source *ws; const char *ws_name; @@ -110,7 +106,6 @@ err_name: err_ws: return NULL; } -EXPORT_SYMBOL_GPL(wakeup_source_create); /* * Record wakeup_source statistics being deleted into a dummy wakeup_source. @@ -153,7 +148,7 @@ static void wakeup_source_free(struct wakeup_source *ws) * * Use only for wakeup source objects created with wakeup_source_create(). */ -void wakeup_source_destroy(struct wakeup_source *ws) +static void wakeup_source_destroy(struct wakeup_source *ws) { if (!ws) return; @@ -162,13 +157,12 @@ void wakeup_source_destroy(struct wakeup_source *ws) wakeup_source_record(ws); wakeup_source_free(ws); } -EXPORT_SYMBOL_GPL(wakeup_source_destroy); /** * wakeup_source_add - Add given object to the list of wakeup sources. * @ws: Wakeup source object to add to the list. */ -void wakeup_source_add(struct wakeup_source *ws) +static void wakeup_source_add(struct wakeup_source *ws) { unsigned long flags; @@ -183,32 +177,29 @@ void wakeup_source_add(struct wakeup_source *ws) list_add_rcu(&ws->entry, &wakeup_sources); raw_spin_unlock_irqrestore(&events_lock, flags); } -EXPORT_SYMBOL_GPL(wakeup_source_add); /** * wakeup_source_remove - Remove given object from the wakeup sources list. * @ws: Wakeup source object to remove from the list. */ -void wakeup_source_remove(struct wakeup_source *ws) +static void wakeup_source_remove(struct wakeup_source *ws) { unsigned long flags; if (WARN_ON(!ws)) return; + /* + * After shutting down the timer, wakeup_source_activate() will warn if + * the given wakeup source is passed to it. + */ + timer_shutdown_sync(&ws->timer); + raw_spin_lock_irqsave(&events_lock, flags); list_del_rcu(&ws->entry); raw_spin_unlock_irqrestore(&events_lock, flags); synchronize_srcu(&wakeup_srcu); - - del_timer_sync(&ws->timer); - /* - * Clear timer.function to make wakeup_source_not_registered() treat - * this wakeup source as not registered. - */ - ws->timer.function = NULL; } -EXPORT_SYMBOL_GPL(wakeup_source_remove); /** * wakeup_source_register - Create wakeup source and add it to the list. @@ -341,7 +332,7 @@ int device_wakeup_enable(struct device *dev) if (!dev || !dev->power.can_wakeup) return -EINVAL; - if (pm_suspend_target_state != PM_SUSPEND_ON) + if (pm_sleep_transition_in_progress()) dev_dbg(dev, "Suspicious %s() during system transition!\n", __func__); ws = wakeup_source_register(dev, dev_name(dev)); @@ -455,16 +446,15 @@ static struct wakeup_source *device_wakeup_detach(struct device *dev) * Detach the @dev's wakeup source object from it, unregister this wakeup source * object and destroy it. */ -int device_wakeup_disable(struct device *dev) +void device_wakeup_disable(struct device *dev) { struct wakeup_source *ws; if (!dev || !dev->power.can_wakeup) - return -EINVAL; + return; ws = device_wakeup_detach(dev); wakeup_source_unregister(ws); - return 0; } EXPORT_SYMBOL_GPL(device_wakeup_disable); @@ -500,55 +490,29 @@ void device_set_wakeup_capable(struct device *dev, bool capable) EXPORT_SYMBOL_GPL(device_set_wakeup_capable); /** - * device_init_wakeup - Device wakeup initialization. - * @dev: Device to handle. - * @enable: Whether or not to enable @dev as a wakeup device. - * - * By default, most devices should leave wakeup disabled. The exceptions are - * devices that everyone expects to be wakeup sources: keyboards, power buttons, - * possibly network interfaces, etc. Also, devices that don't generate their - * own wakeup requests but merely forward requests from one bus to another - * (like PCI bridges) should have wakeup enabled by default. - */ -int device_init_wakeup(struct device *dev, bool enable) -{ - int ret = 0; - - if (!dev) - return -EINVAL; - - if (enable) { - device_set_wakeup_capable(dev, true); - ret = device_wakeup_enable(dev); - } else { - device_wakeup_disable(dev); - device_set_wakeup_capable(dev, false); - } - - return ret; -} -EXPORT_SYMBOL_GPL(device_init_wakeup); - -/** * device_set_wakeup_enable - Enable or disable a device to wake up the system. * @dev: Device to handle. * @enable: enable/disable flag */ int device_set_wakeup_enable(struct device *dev, bool enable) { - return enable ? device_wakeup_enable(dev) : device_wakeup_disable(dev); + if (enable) + return device_wakeup_enable(dev); + + device_wakeup_disable(dev); + return 0; } EXPORT_SYMBOL_GPL(device_set_wakeup_enable); /** - * wakeup_source_not_registered - validate the given wakeup source. + * wakeup_source_not_usable - validate the given wakeup source. * @ws: Wakeup source to be validated. */ -static bool wakeup_source_not_registered(struct wakeup_source *ws) +static bool wakeup_source_not_usable(struct wakeup_source *ws) { /* - * Use timer struct to check if the given source is initialized - * by wakeup_source_add. + * Use the timer struct to check if the given wakeup source has been + * initialized by wakeup_source_add() and it is not going away. */ return ws->timer.function != pm_wakeup_timer_fn; } @@ -586,15 +550,14 @@ static bool wakeup_source_not_registered(struct wakeup_source *ws) * @ws: Wakeup source to handle. * * Update the @ws' statistics and, if @ws has just been activated, notify the PM - * core of the event by incrementing the counter of of wakeup events being + * core of the event by incrementing the counter of the wakeup events being * processed. */ static void wakeup_source_activate(struct wakeup_source *ws) { unsigned int cec; - if (WARN_ONCE(wakeup_source_not_registered(ws), - "unregistered wakeup source\n")) + if (WARN_ONCE(wakeup_source_not_usable(ws), "unusable wakeup source\n")) return; ws->active = true; @@ -644,7 +607,7 @@ void __pm_stay_awake(struct wakeup_source *ws) spin_lock_irqsave(&ws->lock, flags); wakeup_source_report_event(ws, false); - del_timer(&ws->timer); + timer_delete(&ws->timer); ws->timer_expires = 0; spin_unlock_irqrestore(&ws->lock, flags); @@ -724,7 +687,7 @@ static void wakeup_source_deactivate(struct wakeup_source *ws) ws->max_time = duration; ws->last_time = now; - del_timer(&ws->timer); + timer_delete(&ws->timer); ws->timer_expires = 0; if (ws->autosleep_enabled) @@ -732,7 +695,7 @@ static void wakeup_source_deactivate(struct wakeup_source *ws) /* * Increment the counter of registered wakeup events and decrement the - * couter of wakeup events in progress simultaneously. + * counter of wakeup events in progress simultaneously. */ cec = atomic_add_return(MAX_IN_PROGRESS, &combined_event_count); trace_wakeup_source_deactivate(ws->name, cec); @@ -794,7 +757,7 @@ EXPORT_SYMBOL_GPL(pm_relax); */ static void pm_wakeup_timer_fn(struct timer_list *t) { - struct wakeup_source *ws = from_timer(ws, t, timer); + struct wakeup_source *ws = timer_container_of(ws, t, timer); unsigned long flags; spin_lock_irqsave(&ws->lock, flags); @@ -929,6 +892,7 @@ bool pm_wakeup_pending(void) return ret || atomic_read(&pm_abort_suspend) > 0; } +EXPORT_SYMBOL_GPL(pm_wakeup_pending); void pm_system_wakeup(void) { @@ -942,19 +906,47 @@ void pm_system_cancel_wakeup(void) atomic_dec_if_positive(&pm_abort_suspend); } -void pm_wakeup_clear(bool reset) +void pm_wakeup_clear(unsigned int irq_number) { - pm_wakeup_irq = 0; - if (reset) + raw_spin_lock_irq(&wakeup_irq_lock); + + if (irq_number && wakeup_irq[0] == irq_number) + wakeup_irq[0] = wakeup_irq[1]; + else + wakeup_irq[0] = 0; + + wakeup_irq[1] = 0; + + raw_spin_unlock_irq(&wakeup_irq_lock); + + if (!irq_number) atomic_set(&pm_abort_suspend, 0); } void pm_system_irq_wakeup(unsigned int irq_number) { - if (pm_wakeup_irq == 0) { - pm_wakeup_irq = irq_number; + unsigned long flags; + + raw_spin_lock_irqsave(&wakeup_irq_lock, flags); + + if (wakeup_irq[0] == 0) + wakeup_irq[0] = irq_number; + else if (wakeup_irq[1] == 0) + wakeup_irq[1] = irq_number; + else + irq_number = 0; + + pm_pr_dbg("Triggering wakeup from IRQ %d\n", irq_number); + + raw_spin_unlock_irqrestore(&wakeup_irq_lock, flags); + + if (irq_number) pm_system_wakeup(); - } +} + +unsigned int pm_wakeup_irq(void) +{ + return wakeup_irq[0]; } /** @@ -1180,7 +1172,7 @@ static const struct file_operations wakeup_sources_stats_fops = { static int __init wakeup_sources_debugfs_init(void) { - debugfs_create_file("wakeup_sources", S_IRUGO, NULL, NULL, + debugfs_create_file("wakeup_sources", 0444, NULL, NULL, &wakeup_sources_stats_fops); return 0; } diff --git a/drivers/base/power/wakeup_stats.c b/drivers/base/power/wakeup_stats.c index 924fac493c4f..3ffd427248e8 100644 --- a/drivers/base/power/wakeup_stats.c +++ b/drivers/base/power/wakeup_stats.c @@ -34,6 +34,7 @@ wakeup_attr(active_count); wakeup_attr(event_count); wakeup_attr(wakeup_count); wakeup_attr(expire_count); +wakeup_attr(relax_count); static ssize_t active_time_ms_show(struct device *dev, struct device_attribute *attr, char *buf) @@ -119,6 +120,7 @@ static struct attribute *wakeup_source_attrs[] = { &dev_attr_event_count.attr, &dev_attr_wakeup_count.attr, &dev_attr_expire_count.attr, + &dev_attr_relax_count.attr, &dev_attr_active_time_ms.attr, &dev_attr_total_time_ms.attr, &dev_attr_max_time_ms.attr, @@ -210,7 +212,7 @@ void wakeup_source_sysfs_remove(struct wakeup_source *ws) static int __init wakeup_sources_sysfs_init(void) { - wakeup_class = class_create(THIS_MODULE, "wakeup"); + wakeup_class = class_create("wakeup"); return PTR_ERR_OR_ZERO(wakeup_class); } diff --git a/drivers/base/property.c b/drivers/base/property.c index d0874f6c29bb..6a63860579dd 100644 --- a/drivers/base/property.c +++ b/drivers/base/property.c @@ -7,23 +7,30 @@ * Mika Westerberg <mika.westerberg@linux.intel.com> */ -#include <linux/acpi.h> +#include <linux/device.h> +#include <linux/err.h> #include <linux/export.h> -#include <linux/kernel.h> +#include <linux/kconfig.h> #include <linux/of.h> -#include <linux/of_address.h> -#include <linux/of_graph.h> -#include <linux/of_irq.h> #include <linux/property.h> -#include <linux/etherdevice.h> #include <linux/phy.h> +#include <linux/slab.h> +#include <linux/string.h> +#include <linux/types.h> -struct fwnode_handle *dev_fwnode(struct device *dev) +struct fwnode_handle *__dev_fwnode(struct device *dev) { return IS_ENABLED(CONFIG_OF) && dev->of_node ? of_fwnode_handle(dev->of_node) : dev->fwnode; } -EXPORT_SYMBOL_GPL(dev_fwnode); +EXPORT_SYMBOL_GPL(__dev_fwnode); + +const struct fwnode_handle *__dev_fwnode_const(const struct device *dev) +{ + return IS_ENABLED(CONFIG_OF) && dev->of_node ? + of_fwnode_handle(dev->of_node) : dev->fwnode; +} +EXPORT_SYMBOL_GPL(__dev_fwnode_const); /** * device_property_present - check if a property of a device is present @@ -31,8 +38,10 @@ EXPORT_SYMBOL_GPL(dev_fwnode); * @propname: Name of the property * * Check if property @propname is present in the device firmware description. + * + * Return: true if property @propname is present. Otherwise, returns false. */ -bool device_property_present(struct device *dev, const char *propname) +bool device_property_present(const struct device *dev, const char *propname) { return fwnode_property_present(dev_fwnode(dev), propname); } @@ -42,22 +51,64 @@ EXPORT_SYMBOL_GPL(device_property_present); * fwnode_property_present - check if a property of a firmware node is present * @fwnode: Firmware node whose property to check * @propname: Name of the property + * + * Return: true if property @propname is present. Otherwise, returns false. */ bool fwnode_property_present(const struct fwnode_handle *fwnode, const char *propname) { bool ret; + if (IS_ERR_OR_NULL(fwnode)) + return false; + ret = fwnode_call_bool_op(fwnode, property_present, propname); - if (ret == false && !IS_ERR_OR_NULL(fwnode) && - !IS_ERR_OR_NULL(fwnode->secondary)) - ret = fwnode_call_bool_op(fwnode->secondary, property_present, - propname); - return ret; + if (ret) + return ret; + + return fwnode_call_bool_op(fwnode->secondary, property_present, propname); } EXPORT_SYMBOL_GPL(fwnode_property_present); /** + * device_property_read_bool - Return the value for a boolean property of a device + * @dev: Device whose property is being checked + * @propname: Name of the property + * + * Return if property @propname is true or false in the device firmware description. + * + * Return: true if property @propname is present. Otherwise, returns false. + */ +bool device_property_read_bool(const struct device *dev, const char *propname) +{ + return fwnode_property_read_bool(dev_fwnode(dev), propname); +} +EXPORT_SYMBOL_GPL(device_property_read_bool); + +/** + * fwnode_property_read_bool - Return the value for a boolean property of a firmware node + * @fwnode: Firmware node whose property to check + * @propname: Name of the property + * + * Return if property @propname is true or false in the firmware description. + */ +bool fwnode_property_read_bool(const struct fwnode_handle *fwnode, + const char *propname) +{ + bool ret; + + if (IS_ERR_OR_NULL(fwnode)) + return false; + + ret = fwnode_call_bool_op(fwnode, property_read_bool, propname); + if (ret) + return ret; + + return fwnode_call_bool_op(fwnode->secondary, property_read_bool, propname); +} +EXPORT_SYMBOL_GPL(fwnode_property_read_bool); + +/** * device_property_read_u8_array - return a u8 array property of a device * @dev: Device to get the property of * @propname: Name of the property @@ -67,6 +118,9 @@ EXPORT_SYMBOL_GPL(fwnode_property_present); * Function reads an array of u8 properties with @propname from the device * firmware description and stores them to @val if found. * + * It's recommended to call device_property_count_u8() instead of calling + * this function with @val equals %NULL and @nval equals 0. + * * Return: number of values if @val was %NULL, * %0 if the property was found (success), * %-EINVAL if given arguments are not valid, @@ -75,7 +129,7 @@ EXPORT_SYMBOL_GPL(fwnode_property_present); * %-EOVERFLOW if the size of the property is not as expected. * %-ENXIO if no suitable firmware interface is present. */ -int device_property_read_u8_array(struct device *dev, const char *propname, +int device_property_read_u8_array(const struct device *dev, const char *propname, u8 *val, size_t nval) { return fwnode_property_read_u8_array(dev_fwnode(dev), propname, val, nval); @@ -92,6 +146,9 @@ EXPORT_SYMBOL_GPL(device_property_read_u8_array); * Function reads an array of u16 properties with @propname from the device * firmware description and stores them to @val if found. * + * It's recommended to call device_property_count_u16() instead of calling + * this function with @val equals %NULL and @nval equals 0. + * * Return: number of values if @val was %NULL, * %0 if the property was found (success), * %-EINVAL if given arguments are not valid, @@ -100,7 +157,7 @@ EXPORT_SYMBOL_GPL(device_property_read_u8_array); * %-EOVERFLOW if the size of the property is not as expected. * %-ENXIO if no suitable firmware interface is present. */ -int device_property_read_u16_array(struct device *dev, const char *propname, +int device_property_read_u16_array(const struct device *dev, const char *propname, u16 *val, size_t nval) { return fwnode_property_read_u16_array(dev_fwnode(dev), propname, val, nval); @@ -117,6 +174,9 @@ EXPORT_SYMBOL_GPL(device_property_read_u16_array); * Function reads an array of u32 properties with @propname from the device * firmware description and stores them to @val if found. * + * It's recommended to call device_property_count_u32() instead of calling + * this function with @val equals %NULL and @nval equals 0. + * * Return: number of values if @val was %NULL, * %0 if the property was found (success), * %-EINVAL if given arguments are not valid, @@ -125,7 +185,7 @@ EXPORT_SYMBOL_GPL(device_property_read_u16_array); * %-EOVERFLOW if the size of the property is not as expected. * %-ENXIO if no suitable firmware interface is present. */ -int device_property_read_u32_array(struct device *dev, const char *propname, +int device_property_read_u32_array(const struct device *dev, const char *propname, u32 *val, size_t nval) { return fwnode_property_read_u32_array(dev_fwnode(dev), propname, val, nval); @@ -142,6 +202,9 @@ EXPORT_SYMBOL_GPL(device_property_read_u32_array); * Function reads an array of u64 properties with @propname from the device * firmware description and stores them to @val if found. * + * It's recommended to call device_property_count_u64() instead of calling + * this function with @val equals %NULL and @nval equals 0. + * * Return: number of values if @val was %NULL, * %0 if the property was found (success), * %-EINVAL if given arguments are not valid, @@ -150,7 +213,7 @@ EXPORT_SYMBOL_GPL(device_property_read_u32_array); * %-EOVERFLOW if the size of the property is not as expected. * %-ENXIO if no suitable firmware interface is present. */ -int device_property_read_u64_array(struct device *dev, const char *propname, +int device_property_read_u64_array(const struct device *dev, const char *propname, u64 *val, size_t nval) { return fwnode_property_read_u64_array(dev_fwnode(dev), propname, val, nval); @@ -167,6 +230,9 @@ EXPORT_SYMBOL_GPL(device_property_read_u64_array); * Function reads an array of string properties with @propname from the device * firmware description and stores them to @val if found. * + * It's recommended to call device_property_string_array_count() instead of calling + * this function with @val equals %NULL and @nval equals 0. + * * Return: number of values read on success if @val is non-NULL, * number of values available on success if @val is NULL, * %-EINVAL if given arguments are not valid, @@ -175,7 +241,7 @@ EXPORT_SYMBOL_GPL(device_property_read_u64_array); * %-EOVERFLOW if the size of the property is not as expected. * %-ENXIO if no suitable firmware interface is present. */ -int device_property_read_string_array(struct device *dev, const char *propname, +int device_property_read_string_array(const struct device *dev, const char *propname, const char **val, size_t nval) { return fwnode_property_read_string_array(dev_fwnode(dev), propname, val, nval); @@ -197,7 +263,7 @@ EXPORT_SYMBOL_GPL(device_property_read_string_array); * %-EPROTO or %-EILSEQ if the property type is not a string. * %-ENXIO if no suitable firmware interface is present. */ -int device_property_read_string(struct device *dev, const char *propname, +int device_property_read_string(const struct device *dev, const char *propname, const char **val) { return fwnode_property_read_string(dev_fwnode(dev), propname, val); @@ -213,13 +279,13 @@ EXPORT_SYMBOL_GPL(device_property_read_string); * Find a given string in a string array and if it is found return the * index back. * - * Return: %0 if the property was found (success), + * Return: index, starting from %0, if the property was found (success), * %-EINVAL if given arguments are not valid, * %-ENODATA if the property does not have a value, * %-EPROTO if the property is not an array of strings, * %-ENXIO if no suitable firmware interface is present. */ -int device_property_match_string(struct device *dev, const char *propname, +int device_property_match_string(const struct device *dev, const char *propname, const char *string) { return fwnode_property_match_string(dev_fwnode(dev), propname, string); @@ -233,15 +299,16 @@ static int fwnode_property_read_int_array(const struct fwnode_handle *fwnode, { int ret; + if (IS_ERR_OR_NULL(fwnode)) + return -EINVAL; + ret = fwnode_call_int_op(fwnode, property_read_int_array, propname, elem_size, val, nval); - if (ret == -EINVAL && !IS_ERR_OR_NULL(fwnode) && - !IS_ERR_OR_NULL(fwnode->secondary)) - ret = fwnode_call_int_op( - fwnode->secondary, property_read_int_array, propname, - elem_size, val, nval); + if (ret != -EINVAL) + return ret; - return ret; + return fwnode_call_int_op(fwnode->secondary, property_read_int_array, propname, + elem_size, val, nval); } /** @@ -254,6 +321,9 @@ static int fwnode_property_read_int_array(const struct fwnode_handle *fwnode, * Read an array of u8 properties with @propname from @fwnode and stores them to * @val if found. * + * It's recommended to call fwnode_property_count_u8() instead of calling + * this function with @val equals %NULL and @nval equals 0. + * * Return: number of values if @val was %NULL, * %0 if the property was found (success), * %-EINVAL if given arguments are not valid, @@ -280,6 +350,9 @@ EXPORT_SYMBOL_GPL(fwnode_property_read_u8_array); * Read an array of u16 properties with @propname from @fwnode and store them to * @val if found. * + * It's recommended to call fwnode_property_count_u16() instead of calling + * this function with @val equals %NULL and @nval equals 0. + * * Return: number of values if @val was %NULL, * %0 if the property was found (success), * %-EINVAL if given arguments are not valid, @@ -306,6 +379,9 @@ EXPORT_SYMBOL_GPL(fwnode_property_read_u16_array); * Read an array of u32 properties with @propname from @fwnode store them to * @val if found. * + * It's recommended to call fwnode_property_count_u32() instead of calling + * this function with @val equals %NULL and @nval equals 0. + * * Return: number of values if @val was %NULL, * %0 if the property was found (success), * %-EINVAL if given arguments are not valid, @@ -332,6 +408,9 @@ EXPORT_SYMBOL_GPL(fwnode_property_read_u32_array); * Read an array of u64 properties with @propname from @fwnode and store them to * @val if found. * + * It's recommended to call fwnode_property_count_u64() instead of calling + * this function with @val equals %NULL and @nval equals 0. + * * Return: number of values if @val was %NULL, * %0 if the property was found (success), * %-EINVAL if given arguments are not valid, @@ -358,6 +437,9 @@ EXPORT_SYMBOL_GPL(fwnode_property_read_u64_array); * Read an string list property @propname from the given firmware node and store * them to @val if found. * + * It's recommended to call fwnode_property_string_array_count() instead of calling + * this function with @val equals %NULL and @nval equals 0. + * * Return: number of values read on success if @val is non-NULL, * number of values available on success if @val is NULL, * %-EINVAL if given arguments are not valid, @@ -372,14 +454,16 @@ int fwnode_property_read_string_array(const struct fwnode_handle *fwnode, { int ret; + if (IS_ERR_OR_NULL(fwnode)) + return -EINVAL; + ret = fwnode_call_int_op(fwnode, property_read_string_array, propname, val, nval); - if (ret == -EINVAL && !IS_ERR_OR_NULL(fwnode) && - !IS_ERR_OR_NULL(fwnode->secondary)) - ret = fwnode_call_int_op(fwnode->secondary, - property_read_string_array, propname, - val, nval); - return ret; + if (ret != -EINVAL) + return ret; + + return fwnode_call_int_op(fwnode->secondary, property_read_string_array, propname, + val, nval); } EXPORT_SYMBOL_GPL(fwnode_property_read_string_array); @@ -416,7 +500,7 @@ EXPORT_SYMBOL_GPL(fwnode_property_read_string); * Find a given string in a string array and if it is found return the * index back. * - * Return: %0 if the property was found (success), + * Return: index, starting from %0, if the property was found (success), * %-EINVAL if given arguments are not valid, * %-ENODATA if the property does not have a value, * %-EPROTO if the property is not an array of strings, @@ -428,7 +512,7 @@ int fwnode_property_match_string(const struct fwnode_handle *fwnode, const char **values; int nval, ret; - nval = fwnode_property_read_string_array(fwnode, propname, NULL, 0); + nval = fwnode_property_string_array_count(fwnode, propname); if (nval < 0) return nval; @@ -441,35 +525,72 @@ int fwnode_property_match_string(const struct fwnode_handle *fwnode, ret = fwnode_property_read_string_array(fwnode, propname, values, nval); if (ret < 0) - goto out; + goto out_free; ret = match_string(values, nval, string); if (ret < 0) ret = -ENODATA; -out: + +out_free: kfree(values); return ret; } EXPORT_SYMBOL_GPL(fwnode_property_match_string); /** + * fwnode_property_match_property_string - find a property string value in an array and return index + * @fwnode: Firmware node to get the property of + * @propname: Name of the property holding the string value + * @array: String array to search in + * @n: Size of the @array + * + * Find a property string value in a given @array and if it is found return + * the index back. + * + * Return: index, starting from %0, if the string value was found in the @array (success), + * %-ENOENT when the string value was not found in the @array, + * %-EINVAL if given arguments are not valid, + * %-ENODATA if the property does not have a value, + * %-EPROTO or %-EILSEQ if the property is not a string, + * %-ENXIO if no suitable firmware interface is present. + */ +int fwnode_property_match_property_string(const struct fwnode_handle *fwnode, + const char *propname, const char * const *array, size_t n) +{ + const char *string; + int ret; + + ret = fwnode_property_read_string(fwnode, propname, &string); + if (ret) + return ret; + + ret = match_string(array, n, string); + if (ret < 0) + ret = -ENOENT; + + return ret; +} +EXPORT_SYMBOL_GPL(fwnode_property_match_property_string); + +/** * fwnode_property_get_reference_args() - Find a reference with arguments * @fwnode: Firmware node where to look for the reference * @prop: The name of the property * @nargs_prop: The name of the property telling the number of * arguments in the referred node. NULL if @nargs is known, - * otherwise @nargs is ignored. Only relevant on OF. + * otherwise @nargs is ignored. * @nargs: Number of arguments. Ignored if @nargs_prop is non-NULL. * @index: Index of the reference, from zero onwards. * @args: Result structure with reference and integer arguments. + * May be NULL. * * Obtain a reference based on a named property in an fwnode, with * integer arguments. * - * Caller is responsible to call fwnode_handle_put() on the returned - * args->fwnode pointer. + * The caller is responsible for calling fwnode_handle_put() on the returned + * @args->fwnode pointer. * - * Returns: %0 on success + * Return: %0 on success * %-ENOENT when the index is out of bounds, the index has an empty * reference or the property was not found * %-EINVAL on parse error @@ -479,7 +600,20 @@ int fwnode_property_get_reference_args(const struct fwnode_handle *fwnode, unsigned int nargs, unsigned int index, struct fwnode_reference_args *args) { - return fwnode_call_int_op(fwnode, get_reference_args, prop, nargs_prop, + int ret; + + if (IS_ERR_OR_NULL(fwnode)) + return -ENOENT; + + ret = fwnode_call_int_op(fwnode, get_reference_args, prop, nargs_prop, + nargs, index, args); + if (ret == 0) + return ret; + + if (IS_ERR_OR_NULL(fwnode->secondary)) + return ret; + + return fwnode_call_int_op(fwnode->secondary, get_reference_args, prop, nargs_prop, nargs, index, args); } EXPORT_SYMBOL_GPL(fwnode_property_get_reference_args); @@ -492,8 +626,11 @@ EXPORT_SYMBOL_GPL(fwnode_property_get_reference_args); * * @index can be used when the named reference holds a table of references. * - * Returns pointer to the reference fwnode, or ERR_PTR. Caller is responsible to - * call fwnode_handle_put() on the returned fwnode pointer. + * The caller is responsible for calling fwnode_handle_put() on the returned + * fwnode pointer. + * + * Return: a pointer to the reference fwnode, when found. Otherwise, + * returns an error pointer. */ struct fwnode_handle *fwnode_find_reference(const struct fwnode_handle *fwnode, const char *name, @@ -509,58 +646,10 @@ struct fwnode_handle *fwnode_find_reference(const struct fwnode_handle *fwnode, EXPORT_SYMBOL_GPL(fwnode_find_reference); /** - * device_remove_properties - Remove properties from a device object. - * @dev: Device whose properties to remove. - * - * The function removes properties previously associated to the device - * firmware node with device_add_properties(). Memory allocated to the - * properties will also be released. - */ -void device_remove_properties(struct device *dev) -{ - struct fwnode_handle *fwnode = dev_fwnode(dev); - - if (!fwnode) - return; - - if (is_software_node(fwnode->secondary)) { - fwnode_remove_software_node(fwnode->secondary); - set_secondary_fwnode(dev, NULL); - } -} -EXPORT_SYMBOL_GPL(device_remove_properties); - -/** - * device_add_properties - Add a collection of properties to a device object. - * @dev: Device to add properties to. - * @properties: Collection of properties to add. - * - * Associate a collection of device properties represented by @properties with - * @dev. The function takes a copy of @properties. - * - * WARNING: The callers should not use this function if it is known that there - * is no real firmware node associated with @dev! In that case the callers - * should create a software node and assign it to @dev directly. - */ -int device_add_properties(struct device *dev, - const struct property_entry *properties) -{ - struct fwnode_handle *fwnode; - - fwnode = fwnode_create_software_node(properties, NULL); - if (IS_ERR(fwnode)) - return PTR_ERR(fwnode); - - set_secondary_fwnode(dev, fwnode); - return 0; -} -EXPORT_SYMBOL_GPL(device_add_properties); - -/** * fwnode_get_name - Return the name of a node * @fwnode: The firmware node * - * Returns a pointer to the node name. + * Return: a pointer to the node name, or %NULL. */ const char *fwnode_get_name(const struct fwnode_handle *fwnode) { @@ -572,7 +661,7 @@ EXPORT_SYMBOL_GPL(fwnode_get_name); * fwnode_get_name_prefix - Return the prefix of node for printing purposes * @fwnode: The firmware node * - * Returns the prefix of a node, intended to be printed right before the node. + * Return: the prefix of a node, intended to be printed right before the node. * The prefix works also as a separator between the nodes. */ const char *fwnode_get_name_prefix(const struct fwnode_handle *fwnode) @@ -581,10 +670,41 @@ const char *fwnode_get_name_prefix(const struct fwnode_handle *fwnode) } /** + * fwnode_name_eq - Return true if node name is equal + * @fwnode: The firmware node + * @name: The name to which to compare the node name + * + * Compare the name provided as an argument to the name of the node, stopping + * the comparison at either NUL or '@' character, whichever comes first. This + * function is generally used for comparing node names while ignoring the + * possible unit address of the node. + * + * Return: true if the node name matches with the name provided in the @name + * argument, false otherwise. + */ +bool fwnode_name_eq(const struct fwnode_handle *fwnode, const char *name) +{ + const char *node_name; + ptrdiff_t len; + + node_name = fwnode_get_name(fwnode); + if (!node_name) + return false; + + len = strchrnul(node_name, '@') - node_name; + + return str_has_prefix(node_name, name) == len; +} +EXPORT_SYMBOL_GPL(fwnode_name_eq); + +/** * fwnode_get_parent - Return parent firwmare node * @fwnode: Firmware whose parent is retrieved * - * Return parent firmware node of the given node if possible or %NULL if no + * The caller is responsible for calling fwnode_handle_put() on the returned + * fwnode pointer. + * + * Return: parent firmware node of the given node if possible or %NULL if no * parent was available. */ struct fwnode_handle *fwnode_get_parent(const struct fwnode_handle *fwnode) @@ -601,8 +721,12 @@ EXPORT_SYMBOL_GPL(fwnode_get_parent); * on the passed node, making it suitable for iterating through a * node's parents. * - * Returns a node pointer with refcount incremented, use - * fwnode_handle_node() on it when done. + * The caller is responsible for calling fwnode_handle_put() on the returned + * fwnode pointer. Note that this function also puts a reference to @fwnode + * unconditionally. + * + * Return: parent firmware node of the given node if possible or %NULL if no + * parent was available. */ struct fwnode_handle *fwnode_get_next_parent(struct fwnode_handle *fwnode) { @@ -615,46 +739,18 @@ struct fwnode_handle *fwnode_get_next_parent(struct fwnode_handle *fwnode) EXPORT_SYMBOL_GPL(fwnode_get_next_parent); /** - * fwnode_get_next_parent_dev - Find device of closest ancestor fwnode - * @fwnode: firmware node - * - * Given a firmware node (@fwnode), this function finds its closest ancestor - * firmware node that has a corresponding struct device and returns that struct - * device. - * - * The caller of this function is expected to call put_device() on the returned - * device when they are done. - */ -struct device *fwnode_get_next_parent_dev(struct fwnode_handle *fwnode) -{ - struct device *dev; - - fwnode_handle_get(fwnode); - do { - fwnode = fwnode_get_next_parent(fwnode); - if (!fwnode) - return NULL; - dev = get_dev_from_fwnode(fwnode); - } while (!dev); - fwnode_handle_put(fwnode); - return dev; -} - -/** * fwnode_count_parents - Return the number of parents a node has * @fwnode: The node the parents of which are to be counted * - * Returns the number of parents a node has. + * Return: the number of parents a node has. */ unsigned int fwnode_count_parents(const struct fwnode_handle *fwnode) { - struct fwnode_handle *__fwnode; - unsigned int count; - - __fwnode = fwnode_get_parent(fwnode); + struct fwnode_handle *parent; + unsigned int count = 0; - for (count = 0; __fwnode; count++) - __fwnode = fwnode_get_next_parent(__fwnode); + fwnode_for_each_parent_node(fwnode, parent) + count++; return count; } @@ -665,58 +761,37 @@ EXPORT_SYMBOL_GPL(fwnode_count_parents); * @fwnode: The node the parent of which is requested * @depth: Distance of the parent from the node * - * Returns the nth parent of a node. If there is no parent at the requested + * The caller is responsible for calling fwnode_handle_put() on the returned + * fwnode pointer. + * + * Return: the nth parent of a node. If there is no parent at the requested * @depth, %NULL is returned. If @depth is 0, the functionality is equivalent to * fwnode_handle_get(). For @depth == 1, it is fwnode_get_parent() and so on. - * - * The caller is responsible for calling fwnode_handle_put() for the returned - * node. */ struct fwnode_handle *fwnode_get_nth_parent(struct fwnode_handle *fwnode, unsigned int depth) { - unsigned int i; - - fwnode_handle_get(fwnode); - - for (i = 0; i < depth && fwnode; i++) - fwnode = fwnode_get_next_parent(fwnode); - - return fwnode; -} -EXPORT_SYMBOL_GPL(fwnode_get_nth_parent); + struct fwnode_handle *parent; -/** - * fwnode_is_ancestor_of - Test if @test_ancestor is ancestor of @test_child - * @test_ancestor: Firmware which is tested for being an ancestor - * @test_child: Firmware which is tested for being the child - * - * A node is considered an ancestor of itself too. - * - * Returns true if @test_ancestor is an ancestor of @test_child. - * Otherwise, returns false. - */ -bool fwnode_is_ancestor_of(struct fwnode_handle *test_ancestor, - struct fwnode_handle *test_child) -{ - if (!test_ancestor) - return false; + if (depth == 0) + return fwnode_handle_get(fwnode); - fwnode_handle_get(test_child); - while (test_child) { - if (test_child == test_ancestor) { - fwnode_handle_put(test_child); - return true; - } - test_child = fwnode_get_next_parent(test_child); + fwnode_for_each_parent_node(fwnode, parent) { + if (--depth == 0) + return parent; } - return false; + return NULL; } +EXPORT_SYMBOL_GPL(fwnode_get_nth_parent); /** * fwnode_get_next_child_node - Return the next child node handle for a node * @fwnode: Firmware node to find the next child node for. * @child: Handle to one of the node's child nodes or a %NULL handle. + * + * The caller is responsible for calling fwnode_handle_put() on the returned + * fwnode pointer. Note that this function also puts a reference to @child + * unconditionally. */ struct fwnode_handle * fwnode_get_next_child_node(const struct fwnode_handle *fwnode, @@ -727,10 +802,13 @@ fwnode_get_next_child_node(const struct fwnode_handle *fwnode, EXPORT_SYMBOL_GPL(fwnode_get_next_child_node); /** - * fwnode_get_next_available_child_node - Return the next - * available child node handle for a node + * fwnode_get_next_available_child_node - Return the next available child node handle for a node * @fwnode: Firmware node to find the next child node for. * @child: Handle to one of the node's child nodes or a %NULL handle. + * + * The caller is responsible for calling fwnode_handle_put() on the returned + * fwnode pointer. Note that this function also puts a reference to @child + * unconditionally. */ struct fwnode_handle * fwnode_get_next_available_child_node(const struct fwnode_handle *fwnode, @@ -738,7 +816,7 @@ fwnode_get_next_available_child_node(const struct fwnode_handle *fwnode, { struct fwnode_handle *next_child = child; - if (!fwnode) + if (IS_ERR_OR_NULL(fwnode)) return NULL; do { @@ -754,24 +832,28 @@ EXPORT_SYMBOL_GPL(fwnode_get_next_available_child_node); /** * device_get_next_child_node - Return the next child node handle for a device * @dev: Device to find the next child node for. - * @child: Handle to one of the device's child nodes or a null handle. + * @child: Handle to one of the device's child nodes or a %NULL handle. + * + * The caller is responsible for calling fwnode_handle_put() on the returned + * fwnode pointer. Note that this function also puts a reference to @child + * unconditionally. */ -struct fwnode_handle *device_get_next_child_node(struct device *dev, +struct fwnode_handle *device_get_next_child_node(const struct device *dev, struct fwnode_handle *child) { const struct fwnode_handle *fwnode = dev_fwnode(dev); struct fwnode_handle *next; + if (IS_ERR_OR_NULL(fwnode)) + return NULL; + /* Try to find a child in primary fwnode */ next = fwnode_get_next_child_node(fwnode, child); if (next) return next; /* When no more children in primary, continue with secondary */ - if (fwnode && !IS_ERR_OR_NULL(fwnode->secondary)) - next = fwnode_get_next_child_node(fwnode->secondary, child); - - return next; + return fwnode_get_next_child_node(fwnode->secondary, child); } EXPORT_SYMBOL_GPL(device_get_next_child_node); @@ -779,6 +861,9 @@ EXPORT_SYMBOL_GPL(device_get_next_child_node); * fwnode_get_named_child_node - Return first matching named child node handle * @fwnode: Firmware node to find the named child node for. * @childname: String to match child node name against. + * + * The caller is responsible for calling fwnode_handle_put() on the returned + * fwnode pointer. */ struct fwnode_handle * fwnode_get_named_child_node(const struct fwnode_handle *fwnode, @@ -792,8 +877,11 @@ EXPORT_SYMBOL_GPL(fwnode_get_named_child_node); * device_get_named_child_node - Return first matching named child node handle * @dev: Device to find the named child node for. * @childname: String to match child node name against. + * + * The caller is responsible for calling fwnode_handle_put() on the returned + * fwnode pointer. */ -struct fwnode_handle *device_get_named_child_node(struct device *dev, +struct fwnode_handle *device_get_named_child_node(const struct device *dev, const char *childname) { return fwnode_get_named_child_node(dev_fwnode(dev), childname); @@ -804,7 +892,10 @@ EXPORT_SYMBOL_GPL(device_get_named_child_node); * fwnode_handle_get - Obtain a reference to a device node * @fwnode: Pointer to the device node to obtain the reference to. * - * Returns the fwnode handle. + * The caller is responsible for calling fwnode_handle_put() on the returned + * fwnode pointer. + * + * Return: the fwnode handle. */ struct fwnode_handle *fwnode_handle_get(struct fwnode_handle *fwnode) { @@ -816,28 +907,19 @@ struct fwnode_handle *fwnode_handle_get(struct fwnode_handle *fwnode) EXPORT_SYMBOL_GPL(fwnode_handle_get); /** - * fwnode_handle_put - Drop reference to a device node - * @fwnode: Pointer to the device node to drop the reference to. - * - * This has to be used when terminating device_for_each_child_node() iteration - * with break or return to prevent stale device node references from being left - * behind. - */ -void fwnode_handle_put(struct fwnode_handle *fwnode) -{ - fwnode_call_void_op(fwnode, put); -} -EXPORT_SYMBOL_GPL(fwnode_handle_put); - -/** * fwnode_device_is_available - check if a device is available for use * @fwnode: Pointer to the fwnode of the device. * + * Return: true if device is available for use. Otherwise, returns false. + * * For fwnode node types that don't implement the .device_is_available() * operation, this function returns true. */ bool fwnode_device_is_available(const struct fwnode_handle *fwnode) { + if (IS_ERR_OR_NULL(fwnode)) + return false; + if (!fwnode_has_op(fwnode, device_is_available)) return true; @@ -846,50 +928,62 @@ bool fwnode_device_is_available(const struct fwnode_handle *fwnode) EXPORT_SYMBOL_GPL(fwnode_device_is_available); /** - * device_get_child_node_count - return the number of child nodes for device - * @dev: Device to cound the child nodes for + * fwnode_get_child_node_count - return the number of child nodes for a given firmware node + * @fwnode: Pointer to the parent firmware node + * + * Return: the number of child nodes for a given firmware node. */ -unsigned int device_get_child_node_count(struct device *dev) +unsigned int fwnode_get_child_node_count(const struct fwnode_handle *fwnode) { struct fwnode_handle *child; unsigned int count = 0; - device_for_each_child_node(dev, child) + fwnode_for_each_child_node(fwnode, child) count++; return count; } -EXPORT_SYMBOL_GPL(device_get_child_node_count); +EXPORT_SYMBOL_GPL(fwnode_get_child_node_count); -bool device_dma_supported(struct device *dev) +/** + * fwnode_get_named_child_node_count - number of child nodes with given name + * @fwnode: Node which child nodes are counted. + * @name: String to match child node name against. + * + * Scan child nodes and count all the nodes with a specific name. Potential + * 'number' -ending after the 'at sign' for scanned names is ignored. + * E.g.:: + * fwnode_get_named_child_node_count(fwnode, "channel"); + * would match all the nodes:: + * channel { }, channel@0 {}, channel@0xabba {}... + * + * Return: the number of child nodes with a matching name for a given device. + */ +unsigned int fwnode_get_named_child_node_count(const struct fwnode_handle *fwnode, + const char *name) { - const struct fwnode_handle *fwnode = dev_fwnode(dev); + struct fwnode_handle *child; + unsigned int count = 0; - /* For DT, this is always supported. - * For ACPI, this depends on CCA, which - * is determined by the acpi_dma_supported(). - */ - if (is_of_node(fwnode)) - return true; + fwnode_for_each_named_child_node(fwnode, child, name) + count++; - return acpi_dma_supported(to_acpi_device_node(fwnode)); + return count; } -EXPORT_SYMBOL_GPL(device_dma_supported); +EXPORT_SYMBOL_GPL(fwnode_get_named_child_node_count); -enum dev_dma_attr device_get_dma_attr(struct device *dev) +bool device_dma_supported(const struct device *dev) { - const struct fwnode_handle *fwnode = dev_fwnode(dev); - enum dev_dma_attr attr = DEV_DMA_NOT_SUPPORTED; + return fwnode_call_bool_op(dev_fwnode(dev), device_dma_supported); +} +EXPORT_SYMBOL_GPL(device_dma_supported); - if (is_of_node(fwnode)) { - if (of_dma_is_coherent(to_of_node(fwnode))) - attr = DEV_DMA_COHERENT; - else - attr = DEV_DMA_NON_COHERENT; - } else - attr = acpi_get_dma_attr(to_acpi_device_node(fwnode)); +enum dev_dma_attr device_get_dma_attr(const struct device *dev) +{ + if (!fwnode_has_op(dev_fwnode(dev), device_get_dma_attr)) + return DEV_DMA_NOT_SUPPORTED; - return attr; + return fwnode_call_int_op(dev_fwnode(dev), device_get_dma_attr); } EXPORT_SYMBOL_GPL(device_get_dma_attr); @@ -901,7 +995,7 @@ EXPORT_SYMBOL_GPL(device_get_dma_attr); * 'phy-connection-type', and return its index in phy_modes table, or errno in * error case. */ -int fwnode_get_phy_mode(struct fwnode_handle *fwnode) +int fwnode_get_phy_mode(const struct fwnode_handle *fwnode) { const char *pm; int err, i; @@ -935,105 +1029,109 @@ int device_get_phy_mode(struct device *dev) } EXPORT_SYMBOL_GPL(device_get_phy_mode); -static void *fwnode_get_mac_addr(struct fwnode_handle *fwnode, - const char *name, char *addr, - int alen) -{ - int ret = fwnode_property_read_u8_array(fwnode, name, addr, alen); - - if (ret == 0 && alen == ETH_ALEN && is_valid_ether_addr(addr)) - return addr; - return NULL; -} - /** - * fwnode_get_mac_address - Get the MAC from the firmware node + * fwnode_iomap - Maps the memory mapped IO for a given fwnode * @fwnode: Pointer to the firmware node - * @addr: Address of buffer to store the MAC in - * @alen: Length of the buffer pointed to by addr, should be ETH_ALEN - * - * Search the firmware node for the best MAC address to use. 'mac-address' is - * checked first, because that is supposed to contain to "most recent" MAC - * address. If that isn't set, then 'local-mac-address' is checked next, - * because that is the default address. If that isn't set, then the obsolete - * 'address' is checked, just in case we're using an old device tree. - * - * Note that the 'address' property is supposed to contain a virtual address of - * the register set, but some DTS files have redefined that property to be the - * MAC address. - * - * All-zero MAC addresses are rejected, because those could be properties that - * exist in the firmware tables, but were not updated by the firmware. For - * example, the DTS could define 'mac-address' and 'local-mac-address', with - * zero MAC addresses. Some older U-Boots only initialized 'local-mac-address'. - * In this case, the real MAC is in 'local-mac-address', and 'mac-address' - * exists but is all zeros. -*/ -void *fwnode_get_mac_address(struct fwnode_handle *fwnode, char *addr, int alen) -{ - char *res; - - res = fwnode_get_mac_addr(fwnode, "mac-address", addr, alen); - if (res) - return res; - - res = fwnode_get_mac_addr(fwnode, "local-mac-address", addr, alen); - if (res) - return res; - - return fwnode_get_mac_addr(fwnode, "address", addr, alen); -} -EXPORT_SYMBOL(fwnode_get_mac_address); - -/** - * device_get_mac_address - Get the MAC for a given device - * @dev: Pointer to the device - * @addr: Address of buffer to store the MAC in - * @alen: Length of the buffer pointed to by addr, should be ETH_ALEN + * @index: Index of the IO range + * + * Return: a pointer to the mapped memory. */ -void *device_get_mac_address(struct device *dev, char *addr, int alen) +void __iomem *fwnode_iomap(struct fwnode_handle *fwnode, int index) { - return fwnode_get_mac_address(dev_fwnode(dev), addr, alen); + return fwnode_call_ptr_op(fwnode, iomap, index); } -EXPORT_SYMBOL(device_get_mac_address); +EXPORT_SYMBOL(fwnode_iomap); /** * fwnode_irq_get - Get IRQ directly from a fwnode * @fwnode: Pointer to the firmware node * @index: Zero-based index of the IRQ * - * Returns Linux IRQ number on success. Other values are determined - * accordingly to acpi_/of_ irq_get() operation. + * Return: Linux IRQ number on success. Negative errno on failure. */ int fwnode_irq_get(const struct fwnode_handle *fwnode, unsigned int index) { - struct resource res; int ret; - if (is_of_node(fwnode)) - return of_irq_get(to_of_node(fwnode), index); + ret = fwnode_call_int_op(fwnode, irq_get, index); + /* We treat mapping errors as invalid case */ + if (ret == 0) + return -EINVAL; - ret = acpi_irq_get(ACPI_HANDLE_FWNODE(fwnode), index, &res); - if (ret) - return ret; - - return res.start; + return ret; } EXPORT_SYMBOL(fwnode_irq_get); /** + * fwnode_irq_get_byname - Get IRQ from a fwnode using its name + * @fwnode: Pointer to the firmware node + * @name: IRQ name + * + * Description: + * Find a match to the string @name in the 'interrupt-names' string array + * in _DSD for ACPI, or of_node for Device Tree. Then get the Linux IRQ + * number of the IRQ resource corresponding to the index of the matched + * string. + * + * Return: Linux IRQ number on success, or negative errno otherwise. + */ +int fwnode_irq_get_byname(const struct fwnode_handle *fwnode, const char *name) +{ + int index; + + if (!name) + return -EINVAL; + + index = fwnode_property_match_string(fwnode, "interrupt-names", name); + if (index < 0) + return index; + + return fwnode_irq_get(fwnode, index); +} +EXPORT_SYMBOL(fwnode_irq_get_byname); + +/** * fwnode_graph_get_next_endpoint - Get next endpoint firmware node * @fwnode: Pointer to the parent firmware node * @prev: Previous endpoint node or %NULL to get the first * - * Returns an endpoint firmware node pointer or %NULL if no more endpoints + * The caller is responsible for calling fwnode_handle_put() on the returned + * fwnode pointer. Note that this function also puts a reference to @prev + * unconditionally. + * + * Return: an endpoint firmware node pointer or %NULL if no more endpoints * are available. */ struct fwnode_handle * fwnode_graph_get_next_endpoint(const struct fwnode_handle *fwnode, struct fwnode_handle *prev) { - return fwnode_call_ptr_op(fwnode, graph_get_next_endpoint, prev); + struct fwnode_handle *ep, *port_parent = NULL; + const struct fwnode_handle *parent; + + /* + * If this function is in a loop and the previous iteration returned + * an endpoint from fwnode->secondary, then we need to use the secondary + * as parent rather than @fwnode. + */ + if (prev) { + port_parent = fwnode_graph_get_port_parent(prev); + parent = port_parent; + } else { + parent = fwnode; + } + if (IS_ERR_OR_NULL(parent)) + return NULL; + + ep = fwnode_call_ptr_op(parent, graph_get_next_endpoint, prev); + if (ep) + goto out_put_port_parent; + + ep = fwnode_graph_get_next_endpoint(parent->secondary, NULL); + +out_put_port_parent: + fwnode_handle_put(port_parent); + return ep; } EXPORT_SYMBOL_GPL(fwnode_graph_get_next_endpoint); @@ -1041,6 +1139,9 @@ EXPORT_SYMBOL_GPL(fwnode_graph_get_next_endpoint); * fwnode_graph_get_port_parent - Return the device fwnode of a port endpoint * @endpoint: Endpoint firmware node of the port * + * The caller is responsible for calling fwnode_handle_put() on the returned + * fwnode pointer. + * * Return: the firmware node of the device the @endpoint belongs to. */ struct fwnode_handle * @@ -1062,6 +1163,9 @@ EXPORT_SYMBOL_GPL(fwnode_graph_get_port_parent); * @fwnode: Endpoint firmware node pointing to the remote endpoint * * Extracts firmware node of a remote device the @fwnode points to. + * + * The caller is responsible for calling fwnode_handle_put() on the returned + * fwnode pointer. */ struct fwnode_handle * fwnode_graph_get_remote_port_parent(const struct fwnode_handle *fwnode) @@ -1082,6 +1186,9 @@ EXPORT_SYMBOL_GPL(fwnode_graph_get_remote_port_parent); * @fwnode: Endpoint firmware node pointing to the remote endpoint * * Extracts firmware node of a remote port the @fwnode points to. + * + * The caller is responsible for calling fwnode_handle_put() on the returned + * fwnode pointer. */ struct fwnode_handle * fwnode_graph_get_remote_port(const struct fwnode_handle *fwnode) @@ -1095,6 +1202,9 @@ EXPORT_SYMBOL_GPL(fwnode_graph_get_remote_port); * @fwnode: Endpoint firmware node pointing to the remote endpoint * * Extracts firmware node of a remote endpoint the @fwnode points to. + * + * The caller is responsible for calling fwnode_handle_put() on the returned + * fwnode pointer. */ struct fwnode_handle * fwnode_graph_get_remote_endpoint(const struct fwnode_handle *fwnode) @@ -1103,43 +1213,17 @@ fwnode_graph_get_remote_endpoint(const struct fwnode_handle *fwnode) } EXPORT_SYMBOL_GPL(fwnode_graph_get_remote_endpoint); -/** - * fwnode_graph_get_remote_node - get remote parent node for given port/endpoint - * @fwnode: pointer to parent fwnode_handle containing graph port/endpoint - * @port_id: identifier of the parent port node - * @endpoint_id: identifier of the endpoint node - * - * Return: Remote fwnode handle associated with remote endpoint node linked - * to @node. Use fwnode_node_put() on it when done. - */ -struct fwnode_handle * -fwnode_graph_get_remote_node(const struct fwnode_handle *fwnode, u32 port_id, - u32 endpoint_id) +static bool fwnode_graph_remote_available(struct fwnode_handle *ep) { - struct fwnode_handle *endpoint = NULL; - - while ((endpoint = fwnode_graph_get_next_endpoint(fwnode, endpoint))) { - struct fwnode_endpoint fwnode_ep; - struct fwnode_handle *remote; - int ret; + struct fwnode_handle *dev_node; + bool available; - ret = fwnode_graph_parse_endpoint(endpoint, &fwnode_ep); - if (ret < 0) - continue; - - if (fwnode_ep.port != port_id || fwnode_ep.id != endpoint_id) - continue; + dev_node = fwnode_graph_get_remote_port_parent(ep); + available = fwnode_device_is_available(dev_node); + fwnode_handle_put(dev_node); - remote = fwnode_graph_get_remote_port_parent(endpoint); - if (!remote) - return NULL; - - return fwnode_device_is_available(remote) ? remote : NULL; - } - - return NULL; + return available; } -EXPORT_SYMBOL_GPL(fwnode_graph_get_remote_node); /** * fwnode_graph_get_endpoint_by_id - get endpoint by port and endpoint numbers @@ -1148,42 +1232,34 @@ EXPORT_SYMBOL_GPL(fwnode_graph_get_remote_node); * @endpoint: identifier of the endpoint node under the port node * @flags: fwnode lookup flags * - * Return the fwnode handle of the local endpoint corresponding the port and - * endpoint IDs or NULL if not found. + * The caller is responsible for calling fwnode_handle_put() on the returned + * fwnode pointer. + * + * Return: the fwnode handle of the local endpoint corresponding the port and + * endpoint IDs or %NULL if not found. * * If FWNODE_GRAPH_ENDPOINT_NEXT is passed in @flags and the specified endpoint * has not been found, look for the closest endpoint ID greater than the * specified one and return the endpoint that corresponds to it, if present. * - * Do not return endpoints that belong to disabled devices, unless - * FWNODE_GRAPH_DEVICE_DISABLED is passed in @flags. - * - * The returned endpoint needs to be released by calling fwnode_handle_put() on - * it when it is not needed any more. + * Does not return endpoints that belong to disabled devices or endpoints that + * are unconnected, unless FWNODE_GRAPH_DEVICE_DISABLED is passed in @flags. */ struct fwnode_handle * fwnode_graph_get_endpoint_by_id(const struct fwnode_handle *fwnode, u32 port, u32 endpoint, unsigned long flags) { - struct fwnode_handle *ep = NULL, *best_ep = NULL; + struct fwnode_handle *ep, *best_ep = NULL; unsigned int best_ep_id = 0; bool endpoint_next = flags & FWNODE_GRAPH_ENDPOINT_NEXT; bool enabled_only = !(flags & FWNODE_GRAPH_DEVICE_DISABLED); - while ((ep = fwnode_graph_get_next_endpoint(fwnode, ep))) { + fwnode_graph_for_each_endpoint(fwnode, ep) { struct fwnode_endpoint fwnode_ep = { 0 }; int ret; - if (enabled_only) { - struct fwnode_handle *dev_node; - bool available; - - dev_node = fwnode_graph_get_remote_port_parent(ep); - available = fwnode_device_is_available(dev_node); - fwnode_handle_put(dev_node); - if (!available) - continue; - } + if (enabled_only && !fwnode_graph_remote_available(ep)) + continue; ret = fwnode_graph_parse_endpoint(ep, &fwnode_ep); if (ret < 0) @@ -1212,16 +1288,34 @@ fwnode_graph_get_endpoint_by_id(const struct fwnode_handle *fwnode, best_ep_id = fwnode_ep.id; } - if (best_ep) - return best_ep; + return best_ep; +} +EXPORT_SYMBOL_GPL(fwnode_graph_get_endpoint_by_id); - if (fwnode && !IS_ERR_OR_NULL(fwnode->secondary)) - return fwnode_graph_get_endpoint_by_id(fwnode->secondary, port, - endpoint, flags); +/** + * fwnode_graph_get_endpoint_count - Count endpoints on a device node + * @fwnode: The node related to a device + * @flags: fwnode lookup flags + * Count endpoints in a device node. + * + * If FWNODE_GRAPH_DEVICE_DISABLED flag is specified, also unconnected endpoints + * and endpoints connected to disabled devices are counted. + */ +unsigned int fwnode_graph_get_endpoint_count(const struct fwnode_handle *fwnode, + unsigned long flags) +{ + struct fwnode_handle *ep; + unsigned int count = 0; - return NULL; + fwnode_graph_for_each_endpoint(fwnode, ep) { + if (flags & FWNODE_GRAPH_DEVICE_DISABLED || + fwnode_graph_remote_available(ep)) + count++; + } + + return count; } -EXPORT_SYMBOL_GPL(fwnode_graph_get_endpoint_by_id); +EXPORT_SYMBOL_GPL(fwnode_graph_get_endpoint_count); /** * fwnode_graph_parse_endpoint - parse common endpoint node properties @@ -1241,55 +1335,75 @@ int fwnode_graph_parse_endpoint(const struct fwnode_handle *fwnode, } EXPORT_SYMBOL(fwnode_graph_parse_endpoint); -const void *device_get_match_data(struct device *dev) +const void *device_get_match_data(const struct device *dev) { return fwnode_call_ptr_op(dev_fwnode(dev), device_get_match_data, dev); } EXPORT_SYMBOL_GPL(device_get_match_data); -static void * -fwnode_graph_devcon_match(struct fwnode_handle *fwnode, const char *con_id, - void *data, devcon_match_fn_t match) +static unsigned int fwnode_graph_devcon_matches(const struct fwnode_handle *fwnode, + const char *con_id, void *data, + devcon_match_fn_t match, + void **matches, + unsigned int matches_len) { struct fwnode_handle *node; struct fwnode_handle *ep; + unsigned int count = 0; void *ret; fwnode_graph_for_each_endpoint(fwnode, ep) { + if (matches && count >= matches_len) { + fwnode_handle_put(ep); + break; + } + node = fwnode_graph_get_remote_port_parent(ep); - if (!fwnode_device_is_available(node)) + if (!fwnode_device_is_available(node)) { + fwnode_handle_put(node); continue; + } ret = match(node, con_id, data); fwnode_handle_put(node); if (ret) { - fwnode_handle_put(ep); - return ret; + if (matches) + matches[count] = ret; + count++; } } - return NULL; + return count; } -static void * -fwnode_devcon_match(struct fwnode_handle *fwnode, const char *con_id, - void *data, devcon_match_fn_t match) +static unsigned int fwnode_devcon_matches(const struct fwnode_handle *fwnode, + const char *con_id, void *data, + devcon_match_fn_t match, + void **matches, + unsigned int matches_len) { struct fwnode_handle *node; + unsigned int count = 0; + unsigned int i; void *ret; - int i; for (i = 0; ; i++) { + if (matches && count >= matches_len) + break; + node = fwnode_find_reference(fwnode, con_id, i); if (IS_ERR(node)) break; ret = match(node, NULL, data); fwnode_handle_put(node); - if (ret) - return ret; + if (ret) { + if (matches) + matches[count] = ret; + count++; + } } - return NULL; + return count; } /** @@ -1303,19 +1417,66 @@ fwnode_devcon_match(struct fwnode_handle *fwnode, const char *con_id, * device node. @match will be used to convert the connection description to * data the caller is expecting to be returned. */ -void *fwnode_connection_find_match(struct fwnode_handle *fwnode, +void *fwnode_connection_find_match(const struct fwnode_handle *fwnode, const char *con_id, void *data, devcon_match_fn_t match) { + unsigned int count; void *ret; if (!fwnode || !match) return NULL; - ret = fwnode_graph_devcon_match(fwnode, con_id, data, match); - if (ret) + count = fwnode_graph_devcon_matches(fwnode, con_id, data, match, &ret, 1); + if (count) return ret; - return fwnode_devcon_match(fwnode, con_id, data, match); + count = fwnode_devcon_matches(fwnode, con_id, data, match, &ret, 1); + return count ? ret : NULL; } EXPORT_SYMBOL_GPL(fwnode_connection_find_match); + +/** + * fwnode_connection_find_matches - Find connections from a device node + * @fwnode: Device node with the connection + * @con_id: Identifier for the connection + * @data: Data for the match function + * @match: Function to check and convert the connection description + * @matches: (Optional) array of pointers to fill with matches + * @matches_len: Length of @matches + * + * Find up to @matches_len connections with unique identifier @con_id between + * @fwnode and other device nodes. @match will be used to convert the + * connection description to data the caller is expecting to be returned + * through the @matches array. + * + * If @matches is %NULL @matches_len is ignored and the total number of resolved + * matches is returned. + * + * Return: Number of matches resolved, or negative errno. + */ +int fwnode_connection_find_matches(const struct fwnode_handle *fwnode, + const char *con_id, void *data, + devcon_match_fn_t match, + void **matches, unsigned int matches_len) +{ + unsigned int count_graph; + unsigned int count_ref; + + if (!fwnode || !match) + return -EINVAL; + + count_graph = fwnode_graph_devcon_matches(fwnode, con_id, data, match, + matches, matches_len); + + if (matches) { + matches += count_graph; + matches_len -= count_graph; + } + + count_ref = fwnode_devcon_matches(fwnode, con_id, data, match, + matches, matches_len); + + return count_graph + count_ref; +} +EXPORT_SYMBOL_GPL(fwnode_connection_find_matches); diff --git a/drivers/base/regmap/Kconfig b/drivers/base/regmap/Kconfig index 159bac6c5046..ffb2ef488298 100644 --- a/drivers/base/regmap/Kconfig +++ b/drivers/base/regmap/Kconfig @@ -4,15 +4,32 @@ # subsystems should select the appropriate symbols. config REGMAP - default y if (REGMAP_I2C || REGMAP_SPI || REGMAP_SPMI || REGMAP_W1 || REGMAP_AC97 || REGMAP_MMIO || REGMAP_IRQ || REGMAP_SOUNDWIRE || REGMAP_SOUNDWIRE_MBQ || REGMAP_SCCB || REGMAP_I3C || REGMAP_SPI_AVMM || REGMAP_MDIO) - select IRQ_DOMAIN if REGMAP_IRQ - select MDIO_BUS if REGMAP_MDIO - bool - -config REGCACHE_COMPRESSED - select LZO_COMPRESS - select LZO_DECOMPRESS bool + default y if (REGMAP_I2C || REGMAP_SPI || REGMAP_SPMI || REGMAP_W1 || REGMAP_AC97 || REGMAP_MMIO || REGMAP_IRQ || REGMAP_SOUNDWIRE || REGMAP_SOUNDWIRE_MBQ || REGMAP_SCCB || REGMAP_I3C || REGMAP_SPI_AVMM || REGMAP_MDIO || REGMAP_FSI) + help + Enable support for the Register Map (regmap) access API. + + Usually, this option is automatically selected when needed. + However, you may want to enable it manually for running the regmap + KUnit tests. + + If unsure, say N. + +config REGMAP_KUNIT + tristate "KUnit tests for regmap" + depends on KUNIT && REGMAP + default KUNIT_ALL_TESTS + select REGMAP_RAM + +config REGMAP_BUILD + bool "Enable regmap build" + depends on KUNIT + select REGMAP + help + This option exists purely to allow the regmap KUnit tests to + be enabled without having to enable some driver that uses + regmap due to unfortunate issues with how KUnit tests are + normally enabled. config REGMAP_AC97 tristate @@ -39,12 +56,17 @@ config REGMAP_W1 config REGMAP_MDIO tristate + select MDIO_BUS config REGMAP_MMIO tristate config REGMAP_IRQ bool + select IRQ_DOMAIN + +config REGMAP_RAM + tristate config REGMAP_SOUNDWIRE tristate @@ -65,3 +87,7 @@ config REGMAP_I3C config REGMAP_SPI_AVMM tristate depends on SPI + +config REGMAP_FSI + tristate + depends on FSI diff --git a/drivers/base/regmap/Makefile b/drivers/base/regmap/Makefile index 11facb32a027..5fdd0845b45e 100644 --- a/drivers/base/regmap/Makefile +++ b/drivers/base/regmap/Makefile @@ -3,11 +3,12 @@ CFLAGS_regmap.o := -I$(src) obj-$(CONFIG_REGMAP) += regmap.o regcache.o -obj-$(CONFIG_REGMAP) += regcache-rbtree.o regcache-flat.o -obj-$(CONFIG_REGCACHE_COMPRESSED) += regcache-lzo.o +obj-$(CONFIG_REGMAP) += regcache-rbtree.o regcache-flat.o regcache-maple.o obj-$(CONFIG_DEBUG_FS) += regmap-debugfs.o +obj-$(CONFIG_REGMAP_KUNIT) += regmap-kunit.o obj-$(CONFIG_REGMAP_AC97) += regmap-ac97.o obj-$(CONFIG_REGMAP_I2C) += regmap-i2c.o +obj-$(CONFIG_REGMAP_RAM) += regmap-ram.o regmap-raw-ram.o obj-$(CONFIG_REGMAP_SLIMBUS) += regmap-slimbus.o obj-$(CONFIG_REGMAP_SPI) += regmap-spi.o obj-$(CONFIG_REGMAP_SPMI) += regmap-spmi.o @@ -20,3 +21,4 @@ obj-$(CONFIG_REGMAP_SCCB) += regmap-sccb.o obj-$(CONFIG_REGMAP_I3C) += regmap-i3c.o obj-$(CONFIG_REGMAP_SPI_AVMM) += regmap-spi-avmm.o obj-$(CONFIG_REGMAP_MDIO) += regmap-mdio.o +obj-$(CONFIG_REGMAP_FSI) += regmap-fsi.o diff --git a/drivers/base/regmap/internal.h b/drivers/base/regmap/internal.h index 0097696c31de..1477329410ec 100644 --- a/drivers/base/regmap/internal.h +++ b/drivers/base/regmap/internal.h @@ -32,6 +32,7 @@ struct regmap_format { size_t reg_bytes; size_t pad_bytes; size_t val_bytes; + s8 reg_shift; void (*format_write)(struct regmap *map, unsigned int reg, unsigned int val); void (*format_reg)(void *buf, unsigned int reg, unsigned int shift); @@ -53,11 +54,17 @@ struct regmap { spinlock_t spinlock; unsigned long spinlock_flags; }; + struct { + raw_spinlock_t raw_spinlock; + unsigned long raw_spinlock_flags; + }; }; + struct lock_class_key *lock_key; regmap_lock lock; regmap_unlock unlock; void *lock_arg; /* This is passed to lock/unlock functions */ gfp_t alloc_flags; + unsigned int reg_base; struct device *dev; /* Device we do I/O on */ void *work_buf; /* Scratch buffer used to format I/O */ @@ -66,12 +73,12 @@ struct regmap { void *bus_context; const char *name; - bool async; spinlock_t async_lock; wait_queue_head_t async_waitq; struct list_head async_list; struct list_head async_free; int async_ret; + bool async; #ifdef CONFIG_DEBUG_FS bool debugfs_disable; @@ -87,6 +94,7 @@ struct regmap { #endif unsigned int max_register; + bool max_register_is_set; bool (*writeable_reg)(struct device *dev, unsigned int reg); bool (*readable_reg)(struct device *dev, unsigned int reg); bool (*volatile_reg)(struct device *dev, unsigned int reg); @@ -104,8 +112,10 @@ struct regmap { int (*reg_write)(void *context, unsigned int reg, unsigned int val); int (*reg_update_bits)(void *context, unsigned int reg, unsigned int mask, unsigned int val); - - bool defer_caching; + /* Bulk read/write */ + int (*read)(void *context, const void *reg_buf, size_t reg_size, + void *val_buf, size_t val_size); + int (*write)(void *context, const void *data, size_t count); unsigned long read_flag_mask; unsigned long write_flag_mask; @@ -115,6 +125,11 @@ struct regmap { int reg_stride; int reg_stride_order; + bool defer_caching; + + /* If set, will always write field to HW. */ + bool force_write_field; + /* regcache specific members */ const struct regcache_ops *cache_ops; enum regcache_type cache_type; @@ -146,6 +161,9 @@ struct regmap { struct reg_sequence *patch; int patch_regs; + /* if set, the regmap core can sleep */ + bool can_sleep; + /* if set, converts bulk read to single read */ bool use_single_read; /* if set, converts bulk write to single write */ @@ -161,9 +179,6 @@ struct regmap { void *selector_work_buf; /* Scratch buffer used for selector */ struct hwspinlock *hwlock; - - /* if set, the regmap core can sleep */ - bool can_sleep; }; struct regcache_ops { @@ -171,6 +186,7 @@ struct regcache_ops { enum regcache_type type; int (*init)(struct regmap *map); int (*exit)(struct regmap *map); + int (*populate)(struct regmap *map); #ifdef CONFIG_DEBUG_FS void (*debugfs_init)(struct regmap *map); #endif @@ -247,6 +263,8 @@ int regcache_sync_block(struct regmap *map, void *block, unsigned long *cache_present, unsigned int block_base, unsigned int start, unsigned int end); +bool regcache_reg_needs_sync(struct regmap *map, unsigned int reg, + unsigned int val); static inline const void *regcache_get_val_addr(struct regmap *map, const void *base, @@ -257,9 +275,10 @@ static inline const void *regcache_get_val_addr(struct regmap *map, unsigned int regcache_get_val(struct regmap *map, const void *base, unsigned int idx); -bool regcache_set_val(struct regmap *map, void *base, unsigned int idx, +void regcache_set_val(struct regmap *map, void *base, unsigned int idx, unsigned int val); int regcache_lookup_reg(struct regmap *map, unsigned int reg); +int regcache_sync_val(struct regmap *map, unsigned int reg, unsigned int val); int _regmap_raw_write(struct regmap *map, unsigned int reg, const void *val, size_t val_len, bool noinc); @@ -270,8 +289,9 @@ enum regmap_endian regmap_get_val_endian(struct device *dev, const struct regmap_bus *bus, const struct regmap_config *config); +extern struct regcache_ops regcache_flat_sparse_ops; extern struct regcache_ops regcache_rbtree_ops; -extern struct regcache_ops regcache_lzo_ops; +extern struct regcache_ops regcache_maple_ops; extern struct regcache_ops regcache_flat_ops; static inline const char *regmap_name(const struct regmap *map) @@ -297,4 +317,34 @@ static inline unsigned int regcache_get_index_by_order(const struct regmap *map, return reg >> map->reg_stride_order; } +struct regmap_ram_data { + unsigned int *vals; /* Allocatd by caller */ + bool *read; + bool *written; + enum regmap_endian reg_endian; + bool (*noinc_reg)(struct regmap_ram_data *data, unsigned int reg); +}; + +/* + * Create a test register map with data stored in RAM, not intended + * for practical use. + */ +struct regmap *__regmap_init_ram(struct device *dev, + const struct regmap_config *config, + struct regmap_ram_data *data, + struct lock_class_key *lock_key, + const char *lock_name); + +#define regmap_init_ram(dev, config, data) \ + __regmap_lockdep_wrapper(__regmap_init_ram, #dev, dev, config, data) + +struct regmap *__regmap_init_raw_ram(struct device *dev, + const struct regmap_config *config, + struct regmap_ram_data *data, + struct lock_class_key *lock_key, + const char *lock_name); + +#define regmap_init_raw_ram(dev, config, data) \ + __regmap_lockdep_wrapper(__regmap_init_raw_ram, #dev, dev, config, data) + #endif diff --git a/drivers/base/regmap/regcache-flat.c b/drivers/base/regmap/regcache-flat.c index b7e4b2464102..53cc59c84e2f 100644 --- a/drivers/base/regmap/regcache-flat.c +++ b/drivers/base/regmap/regcache-flat.c @@ -6,7 +6,11 @@ // // Author: Mark Brown <broonie@opensource.wolfsonmicro.com> +#include <linux/bitmap.h> +#include <linux/bitops.h> #include <linux/device.h> +#include <linux/limits.h> +#include <linux/overflow.h> #include <linux/seq_file.h> #include <linux/slab.h> @@ -18,46 +22,92 @@ static inline unsigned int regcache_flat_get_index(const struct regmap *map, return regcache_get_index_by_order(map, reg); } +struct regcache_flat_data { + unsigned long *valid; + unsigned int data[]; +}; + static int regcache_flat_init(struct regmap *map) { - int i; - unsigned int *cache; + unsigned int cache_size; + struct regcache_flat_data *cache; - if (!map || map->reg_stride_order < 0 || !map->max_register) + if (!map || map->reg_stride_order < 0 || !map->max_register_is_set) return -EINVAL; - map->cache = kcalloc(regcache_flat_get_index(map, map->max_register) - + 1, sizeof(unsigned int), GFP_KERNEL); - if (!map->cache) + cache_size = regcache_flat_get_index(map, map->max_register) + 1; + cache = kzalloc(struct_size(cache, data, cache_size), map->alloc_flags); + if (!cache) return -ENOMEM; - cache = map->cache; + cache->valid = bitmap_zalloc(cache_size, map->alloc_flags); + if (!cache->valid) + goto err_free; + + map->cache = cache; + + return 0; + +err_free: + kfree(cache); + return -ENOMEM; +} + +static int regcache_flat_exit(struct regmap *map) +{ + struct regcache_flat_data *cache = map->cache; + + if (cache) + bitmap_free(cache->valid); + + kfree(cache); + map->cache = NULL; + + return 0; +} + +static int regcache_flat_populate(struct regmap *map) +{ + struct regcache_flat_data *cache = map->cache; + unsigned int i; for (i = 0; i < map->num_reg_defaults; i++) { unsigned int reg = map->reg_defaults[i].reg; unsigned int index = regcache_flat_get_index(map, reg); - cache[index] = map->reg_defaults[i].def; + cache->data[index] = map->reg_defaults[i].def; + __set_bit(index, cache->valid); } return 0; } -static int regcache_flat_exit(struct regmap *map) +static int regcache_flat_read(struct regmap *map, + unsigned int reg, unsigned int *value) { - kfree(map->cache); - map->cache = NULL; + struct regcache_flat_data *cache = map->cache; + unsigned int index = regcache_flat_get_index(map, reg); + + /* legacy behavior: ignore validity, but warn the user */ + if (unlikely(!test_bit(index, cache->valid))) + dev_warn_once(map->dev, + "using zero-initialized flat cache, this may cause unexpected behavior"); + + *value = cache->data[index]; return 0; } -static int regcache_flat_read(struct regmap *map, - unsigned int reg, unsigned int *value) +static int regcache_flat_sparse_read(struct regmap *map, + unsigned int reg, unsigned int *value) { - unsigned int *cache = map->cache; + struct regcache_flat_data *cache = map->cache; unsigned int index = regcache_flat_get_index(map, reg); - *value = cache[index]; + if (unlikely(!test_bit(index, cache->valid))) + return -ENOENT; + + *value = cache->data[index]; return 0; } @@ -65,10 +115,23 @@ static int regcache_flat_read(struct regmap *map, static int regcache_flat_write(struct regmap *map, unsigned int reg, unsigned int value) { - unsigned int *cache = map->cache; + struct regcache_flat_data *cache = map->cache; unsigned int index = regcache_flat_get_index(map, reg); - cache[index] = value; + cache->data[index] = value; + __set_bit(index, cache->valid); + + return 0; +} + +static int regcache_flat_drop(struct regmap *map, unsigned int min, + unsigned int max) +{ + struct regcache_flat_data *cache = map->cache; + unsigned int bitmap_min = regcache_flat_get_index(map, min); + unsigned int bitmap_max = regcache_flat_get_index(map, max); + + bitmap_clear(cache->valid, bitmap_min, bitmap_max + 1 - bitmap_min); return 0; } @@ -78,6 +141,18 @@ struct regcache_ops regcache_flat_ops = { .name = "flat", .init = regcache_flat_init, .exit = regcache_flat_exit, + .populate = regcache_flat_populate, .read = regcache_flat_read, .write = regcache_flat_write, }; + +struct regcache_ops regcache_flat_sparse_ops = { + .type = REGCACHE_FLAT_S, + .name = "flat-sparse", + .init = regcache_flat_init, + .exit = regcache_flat_exit, + .populate = regcache_flat_populate, + .read = regcache_flat_sparse_read, + .write = regcache_flat_write, + .drop = regcache_flat_drop, +}; diff --git a/drivers/base/regmap/regcache-lzo.c b/drivers/base/regmap/regcache-lzo.c deleted file mode 100644 index 7886303eb026..000000000000 --- a/drivers/base/regmap/regcache-lzo.c +++ /dev/null @@ -1,368 +0,0 @@ -// SPDX-License-Identifier: GPL-2.0 -// -// Register cache access API - LZO caching support -// -// Copyright 2011 Wolfson Microelectronics plc -// -// Author: Dimitris Papastamos <dp@opensource.wolfsonmicro.com> - -#include <linux/device.h> -#include <linux/lzo.h> -#include <linux/slab.h> - -#include "internal.h" - -static int regcache_lzo_exit(struct regmap *map); - -struct regcache_lzo_ctx { - void *wmem; - void *dst; - const void *src; - size_t src_len; - size_t dst_len; - size_t decompressed_size; - unsigned long *sync_bmp; - int sync_bmp_nbits; -}; - -#define LZO_BLOCK_NUM 8 -static int regcache_lzo_block_count(struct regmap *map) -{ - return LZO_BLOCK_NUM; -} - -static int regcache_lzo_prepare(struct regcache_lzo_ctx *lzo_ctx) -{ - lzo_ctx->wmem = kmalloc(LZO1X_MEM_COMPRESS, GFP_KERNEL); - if (!lzo_ctx->wmem) - return -ENOMEM; - return 0; -} - -static int regcache_lzo_compress(struct regcache_lzo_ctx *lzo_ctx) -{ - size_t compress_size; - int ret; - - ret = lzo1x_1_compress(lzo_ctx->src, lzo_ctx->src_len, - lzo_ctx->dst, &compress_size, lzo_ctx->wmem); - if (ret != LZO_E_OK || compress_size > lzo_ctx->dst_len) - return -EINVAL; - lzo_ctx->dst_len = compress_size; - return 0; -} - -static int regcache_lzo_decompress(struct regcache_lzo_ctx *lzo_ctx) -{ - size_t dst_len; - int ret; - - dst_len = lzo_ctx->dst_len; - ret = lzo1x_decompress_safe(lzo_ctx->src, lzo_ctx->src_len, - lzo_ctx->dst, &dst_len); - if (ret != LZO_E_OK || dst_len != lzo_ctx->dst_len) - return -EINVAL; - return 0; -} - -static int regcache_lzo_compress_cache_block(struct regmap *map, - struct regcache_lzo_ctx *lzo_ctx) -{ - int ret; - - lzo_ctx->dst_len = lzo1x_worst_compress(PAGE_SIZE); - lzo_ctx->dst = kmalloc(lzo_ctx->dst_len, GFP_KERNEL); - if (!lzo_ctx->dst) { - lzo_ctx->dst_len = 0; - return -ENOMEM; - } - - ret = regcache_lzo_compress(lzo_ctx); - if (ret < 0) - return ret; - return 0; -} - -static int regcache_lzo_decompress_cache_block(struct regmap *map, - struct regcache_lzo_ctx *lzo_ctx) -{ - int ret; - - lzo_ctx->dst_len = lzo_ctx->decompressed_size; - lzo_ctx->dst = kmalloc(lzo_ctx->dst_len, GFP_KERNEL); - if (!lzo_ctx->dst) { - lzo_ctx->dst_len = 0; - return -ENOMEM; - } - - ret = regcache_lzo_decompress(lzo_ctx); - if (ret < 0) - return ret; - return 0; -} - -static inline int regcache_lzo_get_blkindex(struct regmap *map, - unsigned int reg) -{ - return ((reg / map->reg_stride) * map->cache_word_size) / - DIV_ROUND_UP(map->cache_size_raw, - regcache_lzo_block_count(map)); -} - -static inline int regcache_lzo_get_blkpos(struct regmap *map, - unsigned int reg) -{ - return (reg / map->reg_stride) % - (DIV_ROUND_UP(map->cache_size_raw, - regcache_lzo_block_count(map)) / - map->cache_word_size); -} - -static inline int regcache_lzo_get_blksize(struct regmap *map) -{ - return DIV_ROUND_UP(map->cache_size_raw, - regcache_lzo_block_count(map)); -} - -static int regcache_lzo_init(struct regmap *map) -{ - struct regcache_lzo_ctx **lzo_blocks; - size_t bmp_size; - int ret, i, blksize, blkcount; - const char *p, *end; - unsigned long *sync_bmp; - - ret = 0; - - blkcount = regcache_lzo_block_count(map); - map->cache = kcalloc(blkcount, sizeof(*lzo_blocks), - GFP_KERNEL); - if (!map->cache) - return -ENOMEM; - lzo_blocks = map->cache; - - /* - * allocate a bitmap to be used when syncing the cache with - * the hardware. Each time a register is modified, the corresponding - * bit is set in the bitmap, so we know that we have to sync - * that register. - */ - bmp_size = map->num_reg_defaults_raw; - sync_bmp = bitmap_zalloc(bmp_size, GFP_KERNEL); - if (!sync_bmp) { - ret = -ENOMEM; - goto err; - } - - /* allocate the lzo blocks and initialize them */ - for (i = 0; i < blkcount; i++) { - lzo_blocks[i] = kzalloc(sizeof **lzo_blocks, - GFP_KERNEL); - if (!lzo_blocks[i]) { - bitmap_free(sync_bmp); - ret = -ENOMEM; - goto err; - } - lzo_blocks[i]->sync_bmp = sync_bmp; - lzo_blocks[i]->sync_bmp_nbits = bmp_size; - /* alloc the working space for the compressed block */ - ret = regcache_lzo_prepare(lzo_blocks[i]); - if (ret < 0) - goto err; - } - - blksize = regcache_lzo_get_blksize(map); - p = map->reg_defaults_raw; - end = map->reg_defaults_raw + map->cache_size_raw; - /* compress the register map and fill the lzo blocks */ - for (i = 0; i < blkcount; i++, p += blksize) { - lzo_blocks[i]->src = p; - if (p + blksize > end) - lzo_blocks[i]->src_len = end - p; - else - lzo_blocks[i]->src_len = blksize; - ret = regcache_lzo_compress_cache_block(map, - lzo_blocks[i]); - if (ret < 0) - goto err; - lzo_blocks[i]->decompressed_size = - lzo_blocks[i]->src_len; - } - - return 0; -err: - regcache_lzo_exit(map); - return ret; -} - -static int regcache_lzo_exit(struct regmap *map) -{ - struct regcache_lzo_ctx **lzo_blocks; - int i, blkcount; - - lzo_blocks = map->cache; - if (!lzo_blocks) - return 0; - - blkcount = regcache_lzo_block_count(map); - /* - * the pointer to the bitmap used for syncing the cache - * is shared amongst all lzo_blocks. Ensure it is freed - * only once. - */ - if (lzo_blocks[0]) - bitmap_free(lzo_blocks[0]->sync_bmp); - for (i = 0; i < blkcount; i++) { - if (lzo_blocks[i]) { - kfree(lzo_blocks[i]->wmem); - kfree(lzo_blocks[i]->dst); - } - /* each lzo_block is a pointer returned by kmalloc or NULL */ - kfree(lzo_blocks[i]); - } - kfree(lzo_blocks); - map->cache = NULL; - return 0; -} - -static int regcache_lzo_read(struct regmap *map, - unsigned int reg, unsigned int *value) -{ - struct regcache_lzo_ctx *lzo_block, **lzo_blocks; - int ret, blkindex, blkpos; - size_t tmp_dst_len; - void *tmp_dst; - - /* index of the compressed lzo block */ - blkindex = regcache_lzo_get_blkindex(map, reg); - /* register index within the decompressed block */ - blkpos = regcache_lzo_get_blkpos(map, reg); - lzo_blocks = map->cache; - lzo_block = lzo_blocks[blkindex]; - - /* save the pointer and length of the compressed block */ - tmp_dst = lzo_block->dst; - tmp_dst_len = lzo_block->dst_len; - - /* prepare the source to be the compressed block */ - lzo_block->src = lzo_block->dst; - lzo_block->src_len = lzo_block->dst_len; - - /* decompress the block */ - ret = regcache_lzo_decompress_cache_block(map, lzo_block); - if (ret >= 0) - /* fetch the value from the cache */ - *value = regcache_get_val(map, lzo_block->dst, blkpos); - - kfree(lzo_block->dst); - /* restore the pointer and length of the compressed block */ - lzo_block->dst = tmp_dst; - lzo_block->dst_len = tmp_dst_len; - - return ret; -} - -static int regcache_lzo_write(struct regmap *map, - unsigned int reg, unsigned int value) -{ - struct regcache_lzo_ctx *lzo_block, **lzo_blocks; - int ret, blkindex, blkpos; - size_t tmp_dst_len; - void *tmp_dst; - - /* index of the compressed lzo block */ - blkindex = regcache_lzo_get_blkindex(map, reg); - /* register index within the decompressed block */ - blkpos = regcache_lzo_get_blkpos(map, reg); - lzo_blocks = map->cache; - lzo_block = lzo_blocks[blkindex]; - - /* save the pointer and length of the compressed block */ - tmp_dst = lzo_block->dst; - tmp_dst_len = lzo_block->dst_len; - - /* prepare the source to be the compressed block */ - lzo_block->src = lzo_block->dst; - lzo_block->src_len = lzo_block->dst_len; - - /* decompress the block */ - ret = regcache_lzo_decompress_cache_block(map, lzo_block); - if (ret < 0) { - kfree(lzo_block->dst); - goto out; - } - - /* write the new value to the cache */ - if (regcache_set_val(map, lzo_block->dst, blkpos, value)) { - kfree(lzo_block->dst); - goto out; - } - - /* prepare the source to be the decompressed block */ - lzo_block->src = lzo_block->dst; - lzo_block->src_len = lzo_block->dst_len; - - /* compress the block */ - ret = regcache_lzo_compress_cache_block(map, lzo_block); - if (ret < 0) { - kfree(lzo_block->dst); - kfree(lzo_block->src); - goto out; - } - - /* set the bit so we know we have to sync this register */ - set_bit(reg / map->reg_stride, lzo_block->sync_bmp); - kfree(tmp_dst); - kfree(lzo_block->src); - return 0; -out: - lzo_block->dst = tmp_dst; - lzo_block->dst_len = tmp_dst_len; - return ret; -} - -static int regcache_lzo_sync(struct regmap *map, unsigned int min, - unsigned int max) -{ - struct regcache_lzo_ctx **lzo_blocks; - unsigned int val; - int i; - int ret; - - lzo_blocks = map->cache; - i = min; - for_each_set_bit_from(i, lzo_blocks[0]->sync_bmp, - lzo_blocks[0]->sync_bmp_nbits) { - if (i > max) - continue; - - ret = regcache_read(map, i, &val); - if (ret) - return ret; - - /* Is this the hardware default? If so skip. */ - ret = regcache_lookup_reg(map, i); - if (ret > 0 && val == map->reg_defaults[ret].def) - continue; - - map->cache_bypass = true; - ret = _regmap_write(map, i, val); - map->cache_bypass = false; - if (ret) - return ret; - dev_dbg(map->dev, "Synced register %#x, value %#x\n", - i, val); - } - - return 0; -} - -struct regcache_ops regcache_lzo_ops = { - .type = REGCACHE_COMPRESSED, - .name = "lzo", - .init = regcache_lzo_init, - .exit = regcache_lzo_exit, - .read = regcache_lzo_read, - .write = regcache_lzo_write, - .sync = regcache_lzo_sync -}; diff --git a/drivers/base/regmap/regcache-maple.c b/drivers/base/regmap/regcache-maple.c new file mode 100644 index 000000000000..ca1c72b68f31 --- /dev/null +++ b/drivers/base/regmap/regcache-maple.c @@ -0,0 +1,395 @@ +// SPDX-License-Identifier: GPL-2.0 +// +// Register cache access API - maple tree based cache +// +// Copyright 2023 Arm, Ltd +// +// Author: Mark Brown <broonie@kernel.org> + +#include <linux/debugfs.h> +#include <linux/device.h> +#include <linux/maple_tree.h> +#include <linux/slab.h> + +#include "internal.h" + +static int regcache_maple_read(struct regmap *map, + unsigned int reg, unsigned int *value) +{ + struct maple_tree *mt = map->cache; + MA_STATE(mas, mt, reg, reg); + unsigned long *entry; + + rcu_read_lock(); + + entry = mas_walk(&mas); + if (!entry) { + rcu_read_unlock(); + return -ENOENT; + } + + *value = entry[reg - mas.index]; + + rcu_read_unlock(); + + return 0; +} + +static int regcache_maple_write(struct regmap *map, unsigned int reg, + unsigned int val) +{ + struct maple_tree *mt = map->cache; + MA_STATE(mas, mt, reg, reg); + unsigned long *entry, *upper, *lower; + unsigned long index, last; + size_t lower_sz, upper_sz; + int ret; + + rcu_read_lock(); + + entry = mas_walk(&mas); + if (entry) { + entry[reg - mas.index] = val; + rcu_read_unlock(); + return 0; + } + + /* Any adjacent entries to extend/merge? */ + mas_set_range(&mas, reg - 1, reg + 1); + index = reg; + last = reg; + + lower = mas_find(&mas, reg - 1); + if (lower) { + index = mas.index; + lower_sz = (mas.last - mas.index + 1) * sizeof(unsigned long); + } + + upper = mas_find(&mas, reg + 1); + if (upper) { + last = mas.last; + upper_sz = (mas.last - mas.index + 1) * sizeof(unsigned long); + } + + rcu_read_unlock(); + + entry = kmalloc_array(last - index + 1, sizeof(*entry), map->alloc_flags); + if (!entry) + return -ENOMEM; + + if (lower) + memcpy(entry, lower, lower_sz); + entry[reg - index] = val; + if (upper) + memcpy(&entry[reg - index + 1], upper, upper_sz); + + /* + * This is safe because the regmap lock means the Maple lock + * is redundant, but we need to take it due to lockdep asserts + * in the maple tree code. + */ + mas_lock(&mas); + + mas_set_range(&mas, index, last); + ret = mas_store_gfp(&mas, entry, map->alloc_flags); + + mas_unlock(&mas); + + if (ret == 0) { + kfree(lower); + kfree(upper); + } + + return ret; +} + +static int regcache_maple_drop(struct regmap *map, unsigned int min, + unsigned int max) +{ + struct maple_tree *mt = map->cache; + MA_STATE(mas, mt, min, max); + unsigned long *entry, *lower, *upper; + /* initialized to work around false-positive -Wuninitialized warning */ + unsigned long lower_index = 0, lower_last = 0; + unsigned long upper_index, upper_last; + int ret = 0; + + lower = NULL; + upper = NULL; + + mas_lock(&mas); + + mas_for_each(&mas, entry, max) { + /* + * This is safe because the regmap lock means the + * Maple lock is redundant, but we need to take it due + * to lockdep asserts in the maple tree code. + */ + mas_unlock(&mas); + + /* Do we need to save any of this entry? */ + if (mas.index < min) { + lower_index = mas.index; + lower_last = min -1; + + lower = kmemdup_array(entry, + min - mas.index, sizeof(*lower), + map->alloc_flags); + if (!lower) { + ret = -ENOMEM; + goto out_unlocked; + } + } + + if (mas.last > max) { + upper_index = max + 1; + upper_last = mas.last; + + upper = kmemdup_array(&entry[max - mas.index + 1], + mas.last - max, sizeof(*upper), + map->alloc_flags); + if (!upper) { + ret = -ENOMEM; + goto out_unlocked; + } + } + + kfree(entry); + mas_lock(&mas); + mas_erase(&mas); + + /* Insert new nodes with the saved data */ + if (lower) { + mas_set_range(&mas, lower_index, lower_last); + ret = mas_store_gfp(&mas, lower, map->alloc_flags); + if (ret != 0) + goto out; + lower = NULL; + } + + if (upper) { + mas_set_range(&mas, upper_index, upper_last); + ret = mas_store_gfp(&mas, upper, map->alloc_flags); + if (ret != 0) + goto out; + upper = NULL; + } + } + +out: + mas_unlock(&mas); +out_unlocked: + kfree(lower); + kfree(upper); + + return ret; +} + +static int regcache_maple_sync_block(struct regmap *map, unsigned long *entry, + struct ma_state *mas, + unsigned int min, unsigned int max) +{ + void *buf; + unsigned long r; + size_t val_bytes = map->format.val_bytes; + int ret = 0; + + mas_pause(mas); + rcu_read_unlock(); + + /* + * Use a raw write if writing more than one register to a + * device that supports raw writes to reduce transaction + * overheads. + */ + if (max - min > 1 && regmap_can_raw_write(map)) { + buf = kmalloc_array(max - min, val_bytes, map->alloc_flags); + if (!buf) { + ret = -ENOMEM; + goto out; + } + + /* Render the data for a raw write */ + for (r = min; r < max; r++) { + regcache_set_val(map, buf, r - min, + entry[r - mas->index]); + } + + ret = _regmap_raw_write(map, min, buf, (max - min) * val_bytes, + false); + + kfree(buf); + } else { + for (r = min; r < max; r++) { + ret = _regmap_write(map, r, + entry[r - mas->index]); + if (ret != 0) + goto out; + } + } + +out: + rcu_read_lock(); + + return ret; +} + +static int regcache_maple_sync(struct regmap *map, unsigned int min, + unsigned int max) +{ + struct maple_tree *mt = map->cache; + unsigned long *entry; + MA_STATE(mas, mt, min, max); + unsigned long lmin = min; + unsigned long lmax = max; + unsigned int r, v, sync_start; + int ret = 0; + bool sync_needed = false; + + map->cache_bypass = true; + + rcu_read_lock(); + + mas_for_each(&mas, entry, max) { + for (r = max(mas.index, lmin); r <= min(mas.last, lmax); r++) { + v = entry[r - mas.index]; + + if (regcache_reg_needs_sync(map, r, v)) { + if (!sync_needed) { + sync_start = r; + sync_needed = true; + } + continue; + } + + if (!sync_needed) + continue; + + ret = regcache_maple_sync_block(map, entry, &mas, + sync_start, r); + if (ret != 0) + goto out; + sync_needed = false; + } + + if (sync_needed) { + ret = regcache_maple_sync_block(map, entry, &mas, + sync_start, r); + if (ret != 0) + goto out; + sync_needed = false; + } + } + +out: + rcu_read_unlock(); + + map->cache_bypass = false; + + return ret; +} + +static int regcache_maple_init(struct regmap *map) +{ + struct maple_tree *mt; + + mt = kmalloc(sizeof(*mt), map->alloc_flags); + if (!mt) + return -ENOMEM; + map->cache = mt; + + mt_init(mt); + + if (!mt_external_lock(mt) && map->lock_key) + lockdep_set_class_and_subclass(&mt->ma_lock, map->lock_key, 1); + + return 0; +} + +static int regcache_maple_exit(struct regmap *map) +{ + struct maple_tree *mt = map->cache; + MA_STATE(mas, mt, 0, UINT_MAX); + unsigned int *entry; + + /* if we've already been called then just return */ + if (!mt) + return 0; + + mas_lock(&mas); + mas_for_each(&mas, entry, UINT_MAX) + kfree(entry); + __mt_destroy(mt); + mas_unlock(&mas); + + kfree(mt); + map->cache = NULL; + + return 0; +} + +static int regcache_maple_insert_block(struct regmap *map, int first, + int last) +{ + struct maple_tree *mt = map->cache; + MA_STATE(mas, mt, first, last); + unsigned long *entry; + int i, ret; + + entry = kmalloc_array(last - first + 1, sizeof(*entry), map->alloc_flags); + if (!entry) + return -ENOMEM; + + for (i = 0; i < last - first + 1; i++) + entry[i] = map->reg_defaults[first + i].def; + + mas_lock(&mas); + + mas_set_range(&mas, map->reg_defaults[first].reg, + map->reg_defaults[last].reg); + ret = mas_store_gfp(&mas, entry, map->alloc_flags); + + mas_unlock(&mas); + + if (ret) + kfree(entry); + + return ret; +} + +static int regcache_maple_populate(struct regmap *map) +{ + int i; + int ret; + int range_start; + + range_start = 0; + + /* Scan for ranges of contiguous registers */ + for (i = 1; i < map->num_reg_defaults; i++) { + if (map->reg_defaults[i].reg != + map->reg_defaults[i - 1].reg + 1) { + ret = regcache_maple_insert_block(map, range_start, + i - 1); + if (ret != 0) + return ret; + + range_start = i; + } + } + + /* Add the last block */ + return regcache_maple_insert_block(map, range_start, map->num_reg_defaults - 1); +} + +struct regcache_ops regcache_maple_ops = { + .type = REGCACHE_MAPLE, + .name = "maple", + .init = regcache_maple_init, + .exit = regcache_maple_exit, + .populate = regcache_maple_populate, + .read = regcache_maple_read, + .write = regcache_maple_write, + .drop = regcache_maple_drop, + .sync = regcache_maple_sync, +}; diff --git a/drivers/base/regmap/regcache-rbtree.c b/drivers/base/regmap/regcache-rbtree.c index cfa29dc89bbf..3344b82c3799 100644 --- a/drivers/base/regmap/regcache-rbtree.c +++ b/drivers/base/regmap/regcache-rbtree.c @@ -22,7 +22,7 @@ struct regcache_rbtree_node { /* block of adjacent registers */ void *block; /* Which registers are present */ - long *cache_present; + unsigned long *cache_present; /* base register handled by this block */ unsigned int base_reg; /* number of registers available in the block */ @@ -184,10 +184,8 @@ static void rbtree_debugfs_init(struct regmap *map) static int regcache_rbtree_init(struct regmap *map) { struct regcache_rbtree_ctx *rbtree_ctx; - int i; - int ret; - map->cache = kmalloc(sizeof *rbtree_ctx, GFP_KERNEL); + map->cache = kmalloc(sizeof *rbtree_ctx, map->alloc_flags); if (!map->cache) return -ENOMEM; @@ -195,19 +193,7 @@ static int regcache_rbtree_init(struct regmap *map) rbtree_ctx->root = RB_ROOT; rbtree_ctx->cached_rbnode = NULL; - for (i = 0; i < map->num_reg_defaults; i++) { - ret = regcache_rbtree_write(map, - map->reg_defaults[i].reg, - map->reg_defaults[i].def); - if (ret) - goto err; - } - return 0; - -err: - regcache_rbtree_exit(map); - return ret; } static int regcache_rbtree_exit(struct regmap *map) @@ -239,6 +225,22 @@ static int regcache_rbtree_exit(struct regmap *map) return 0; } +static int regcache_rbtree_populate(struct regmap *map) +{ + unsigned int i; + int ret; + + for (i = 0; i < map->num_reg_defaults; i++) { + ret = regcache_rbtree_write(map, + map->reg_defaults[i].reg, + map->reg_defaults[i].def); + if (ret) + return ret; + } + + return 0; +} + static int regcache_rbtree_read(struct regmap *map, unsigned int reg, unsigned int *value) { @@ -275,20 +277,18 @@ static int regcache_rbtree_insert_to_block(struct regmap *map, pos = (reg - base_reg) / map->reg_stride; offset = (rbnode->base_reg - base_reg) / map->reg_stride; - blk = krealloc(rbnode->block, - blklen * map->cache_word_size, - GFP_KERNEL); + blk = krealloc_array(rbnode->block, blklen, map->cache_word_size, map->alloc_flags); if (!blk) return -ENOMEM; + rbnode->block = blk; + if (BITS_TO_LONGS(blklen) > BITS_TO_LONGS(rbnode->blklen)) { - present = krealloc(rbnode->cache_present, - BITS_TO_LONGS(blklen) * sizeof(*present), - GFP_KERNEL); - if (!present) { - kfree(blk); + present = krealloc_array(rbnode->cache_present, + BITS_TO_LONGS(blklen), sizeof(*present), + map->alloc_flags); + if (!present) return -ENOMEM; - } memset(present + BITS_TO_LONGS(rbnode->blklen), 0, (BITS_TO_LONGS(blklen) - BITS_TO_LONGS(rbnode->blklen)) @@ -305,7 +305,6 @@ static int regcache_rbtree_insert_to_block(struct regmap *map, } /* update the rbnode block, its size and the base register */ - rbnode->block = blk; rbnode->blklen = blklen; rbnode->base_reg = base_reg; rbnode->cache_present = present; @@ -321,7 +320,7 @@ regcache_rbtree_node_alloc(struct regmap *map, unsigned int reg) const struct regmap_range *range; int i; - rbnode = kzalloc(sizeof(*rbnode), GFP_KERNEL); + rbnode = kzalloc(sizeof(*rbnode), map->alloc_flags); if (!rbnode) return NULL; @@ -347,13 +346,13 @@ regcache_rbtree_node_alloc(struct regmap *map, unsigned int reg) } rbnode->block = kmalloc_array(rbnode->blklen, map->cache_word_size, - GFP_KERNEL); + map->alloc_flags); if (!rbnode->block) goto err_free; rbnode->cache_present = kcalloc(BITS_TO_LONGS(rbnode->blklen), sizeof(*rbnode->cache_present), - GFP_KERNEL); + map->alloc_flags); if (!rbnode->cache_present) goto err_free_block; @@ -454,7 +453,8 @@ static int regcache_rbtree_write(struct regmap *map, unsigned int reg, if (!rbnode) return -ENOMEM; regcache_rbtree_set_register(map, rbnode, - reg - rbnode->base_reg, value); + (reg - rbnode->base_reg) / map->reg_stride, + value); regcache_rbtree_insert(map, &rbtree_ctx->root, rbnode); rbtree_ctx->cached_rbnode = rbnode; } @@ -472,6 +472,8 @@ static int regcache_rbtree_sync(struct regmap *map, unsigned int min, unsigned int start, end; int ret; + map->async = true; + rbtree_ctx = map->cache; for (node = rb_first(&rbtree_ctx->root); node; node = rb_next(node)) { rbnode = rb_entry(node, struct regcache_rbtree_node, node); @@ -500,6 +502,8 @@ static int regcache_rbtree_sync(struct regmap *map, unsigned int min, return ret; } + map->async = false; + return regmap_async_complete(map); } @@ -544,6 +548,7 @@ struct regcache_ops regcache_rbtree_ops = { .name = "rbtree", .init = regcache_rbtree_init, .exit = regcache_rbtree_exit, + .populate = regcache_rbtree_populate, #ifdef CONFIG_DEBUG_FS .debugfs_init = rbtree_debugfs_init, #endif diff --git a/drivers/base/regmap/regcache.c b/drivers/base/regmap/regcache.c index f2469d3435ca..319c342bf5a0 100644 --- a/drivers/base/regmap/regcache.c +++ b/drivers/base/regmap/regcache.c @@ -16,13 +16,32 @@ #include "internal.h" static const struct regcache_ops *cache_types[] = { + ®cache_flat_sparse_ops, ®cache_rbtree_ops, -#if IS_ENABLED(CONFIG_REGCACHE_COMPRESSED) - ®cache_lzo_ops, -#endif + ®cache_maple_ops, ®cache_flat_ops, }; +static int regcache_defaults_cmp(const void *a, const void *b) +{ + const struct reg_default *x = a; + const struct reg_default *y = b; + + if (x->reg > y->reg) + return 1; + else if (x->reg < y->reg) + return -1; + else + return 0; +} + +void regcache_sort_defaults(struct reg_default *defaults, unsigned int ndefaults) +{ + sort(defaults, ndefaults, sizeof(*defaults), + regcache_defaults_cmp, NULL); +} +EXPORT_SYMBOL_GPL(regcache_sort_defaults); + static int regcache_hw_init(struct regmap *map) { int i, j; @@ -133,6 +152,12 @@ int regcache_init(struct regmap *map, const struct regmap_config *config) return -EINVAL; } + if (config->num_reg_defaults && !config->reg_defaults) { + dev_err(map->dev, + "Register defaults number are set without the reg!\n"); + return -EINVAL; + } + for (i = 0; i < config->num_reg_defaults; i++) if (config->reg_defaults[i].reg % map->reg_stride) return -EINVAL; @@ -142,7 +167,7 @@ int regcache_init(struct regmap *map, const struct regmap_config *config) break; if (i == ARRAY_SIZE(cache_types)) { - dev_err(map->dev, "Could not match compress type: %d\n", + dev_err(map->dev, "Could not match cache type: %d\n", map->cache_type); return -EINVAL; } @@ -150,7 +175,7 @@ int regcache_init(struct regmap *map, const struct regmap_config *config) map->num_reg_defaults = config->num_reg_defaults; map->num_reg_defaults_raw = config->num_reg_defaults_raw; map->reg_defaults_raw = config->reg_defaults_raw; - map->cache_word_size = DIV_ROUND_UP(config->val_bits, 8); + map->cache_word_size = BITS_TO_BYTES(config->val_bits); map->cache_size_raw = map->cache_word_size * config->num_reg_defaults_raw; map->cache = NULL; @@ -166,8 +191,8 @@ int regcache_init(struct regmap *map, const struct regmap_config *config) * a copy of it. */ if (config->reg_defaults) { - tmp_buf = kmemdup(config->reg_defaults, map->num_reg_defaults * - sizeof(struct reg_default), GFP_KERNEL); + tmp_buf = kmemdup_array(config->reg_defaults, map->num_reg_defaults, + sizeof(*map->reg_defaults), GFP_KERNEL); if (!tmp_buf) return -ENOMEM; map->reg_defaults = tmp_buf; @@ -183,18 +208,38 @@ int regcache_init(struct regmap *map, const struct regmap_config *config) return 0; } - if (!map->max_register) - map->max_register = map->num_reg_defaults_raw; + if (!map->max_register_is_set && map->num_reg_defaults_raw) { + map->max_register = (map->num_reg_defaults_raw - 1) * map->reg_stride; + map->max_register_is_set = true; + } if (map->cache_ops->init) { dev_dbg(map->dev, "Initializing %s cache\n", map->cache_ops->name); + map->lock(map->lock_arg); ret = map->cache_ops->init(map); + map->unlock(map->lock_arg); if (ret) goto err_free; } + + if (map->num_reg_defaults && map->cache_ops->populate) { + dev_dbg(map->dev, "Populating %s cache\n", map->cache_ops->name); + map->lock(map->lock_arg); + ret = map->cache_ops->populate(map); + map->unlock(map->lock_arg); + if (ret) + goto err_exit; + } return 0; +err_exit: + if (map->cache_ops->exit) { + dev_dbg(map->dev, "Destroying %s cache\n", map->cache_ops->name); + map->lock(map->lock_arg); + ret = map->cache_ops->exit(map); + map->unlock(map->lock_arg); + } err_free: kfree(map->reg_defaults); if (map->cache_free) @@ -217,7 +262,9 @@ void regcache_exit(struct regmap *map) if (map->cache_ops->exit) { dev_dbg(map->dev, "Destroying %s cache\n", map->cache_ops->name); + map->lock(map->lock_arg); map->cache_ops->exit(map); + map->unlock(map->lock_arg); } } @@ -236,7 +283,7 @@ int regcache_read(struct regmap *map, int ret; if (map->cache_type == REGCACHE_NONE) - return -ENOSYS; + return -EINVAL; BUG_ON(!map->cache_ops); @@ -275,11 +322,14 @@ int regcache_write(struct regmap *map, return 0; } -static bool regcache_reg_needs_sync(struct regmap *map, unsigned int reg, - unsigned int val) +bool regcache_reg_needs_sync(struct regmap *map, unsigned int reg, + unsigned int val) { int ret; + if (!regmap_writeable(map, reg)) + return false; + /* If we don't know the chip just got reset, then sync everything. */ if (!map->no_sync_defaults) return true; @@ -305,6 +355,8 @@ static int regcache_default_sync(struct regmap *map, unsigned int min, continue; ret = regcache_read(map, reg, &val); + if (ret == -ENOENT) + continue; if (ret) return ret; @@ -325,6 +377,11 @@ static int regcache_default_sync(struct regmap *map, unsigned int min, return 0; } +static int rbtree_all(const void *key, const struct rb_node *node) +{ + return 0; +} + /** * regcache_sync - Sync the register cache with the hardware. * @@ -342,6 +399,10 @@ int regcache_sync(struct regmap *map) unsigned int i; const char *name; bool bypass; + struct rb_node *node; + + if (WARN_ON(map->cache_type == REGCACHE_NONE)) + return -EINVAL; BUG_ON(!map->cache_ops); @@ -356,8 +417,6 @@ int regcache_sync(struct regmap *map) if (!map->cache_dirty) goto out; - map->async = true; - /* Apply any patch first */ map->cache_bypass = true; for (i = 0; i < map->patch_regs; i++) { @@ -380,9 +439,31 @@ int regcache_sync(struct regmap *map) out: /* Restore the bypass state */ - map->async = false; map->cache_bypass = bypass; map->no_sync_defaults = false; + + /* + * If we did any paging with cache bypassed and a cached + * paging register then the register and cache state might + * have gone out of sync, force writes of all the paging + * registers. + */ + rb_for_each(node, NULL, &map->range_tree, rbtree_all) { + struct regmap_range_node *this = + rb_entry(node, struct regmap_range_node, node); + + /* If there's nothing in the cache there's nothing to sync */ + if (regcache_read(map, this->selector_reg, &i) != 0) + continue; + + ret = _regmap_write(map, this->selector_reg, i); + if (ret != 0) { + dev_err(map->dev, "Failed to write %x = %x: %d\n", + this->selector_reg, i, ret); + break; + } + } + map->unlock(map->lock_arg); regmap_async_complete(map); @@ -412,6 +493,9 @@ int regcache_sync_region(struct regmap *map, unsigned int min, const char *name; bool bypass; + if (WARN_ON(map->cache_type == REGCACHE_NONE)) + return -EINVAL; + BUG_ON(!map->cache_ops); map->lock(map->lock_arg); @@ -495,7 +579,8 @@ EXPORT_SYMBOL_GPL(regcache_drop_region); void regcache_cache_only(struct regmap *map, bool enable) { map->lock(map->lock_arg); - WARN_ON(map->cache_bypass && enable); + WARN_ON(map->cache_type != REGCACHE_NONE && + map->cache_bypass && enable); map->cache_only = enable; trace_regmap_cache_only(map, enable); map->unlock(map->lock_arg); @@ -531,7 +616,7 @@ EXPORT_SYMBOL_GPL(regcache_mark_dirty); * @enable: flag if changes should not be written to the cache * * When a register map is marked with the cache bypass option, writes - * to the register map API will only update the hardware and not the + * to the register map API will only update the hardware and not * the cache directly. This is useful when syncing the cache back to * the hardware. */ @@ -545,17 +630,37 @@ void regcache_cache_bypass(struct regmap *map, bool enable) } EXPORT_SYMBOL_GPL(regcache_cache_bypass); -bool regcache_set_val(struct regmap *map, void *base, unsigned int idx, - unsigned int val) +/** + * regcache_reg_cached - Check if a register is cached + * + * @map: map to check + * @reg: register to check + * + * Reports if a register is cached. + */ +bool regcache_reg_cached(struct regmap *map, unsigned int reg) { - if (regcache_get_val(map, base, idx) == val) - return true; + unsigned int val; + int ret; + + map->lock(map->lock_arg); + ret = regcache_read(map, reg, &val); + + map->unlock(map->lock_arg); + + return ret == 0; +} +EXPORT_SYMBOL_GPL(regcache_reg_cached); + +void regcache_set_val(struct regmap *map, void *base, unsigned int idx, + unsigned int val) +{ /* Use device native format if possible */ if (map->format.format_val) { map->format.format_val(base + (map->cache_word_size * idx), val, 0); - return false; + return; } switch (map->cache_word_size) { @@ -577,18 +682,9 @@ bool regcache_set_val(struct regmap *map, void *base, unsigned int idx, cache[idx] = val; break; } -#ifdef CONFIG_64BIT - case 8: { - u64 *cache = base; - - cache[idx] = val; - break; - } -#endif default: BUG(); } - return false; } unsigned int regcache_get_val(struct regmap *map, const void *base, @@ -618,13 +714,6 @@ unsigned int regcache_get_val(struct regmap *map, const void *base, return cache[idx]; } -#ifdef CONFIG_64BIT - case 8: { - const u64 *cache = base; - - return cache[idx]; - } -#endif default: BUG(); } @@ -665,6 +754,30 @@ static bool regcache_reg_present(unsigned long *cache_present, unsigned int idx) return test_bit(idx, cache_present); } +int regcache_sync_val(struct regmap *map, unsigned int reg, unsigned int val) +{ + int ret; + + if (!regcache_reg_needs_sync(map, reg, val)) + return 0; + + map->cache_bypass = true; + + ret = _regmap_write(map, reg, val); + + map->cache_bypass = false; + + if (ret != 0) { + dev_err(map->dev, "Unable to sync register %#x. %d\n", + reg, ret); + return ret; + } + dev_dbg(map->dev, "Synced register %#x, value %#x\n", + reg, val); + + return 0; +} + static int regcache_sync_block_single(struct regmap *map, void *block, unsigned long *cache_present, unsigned int block_base, @@ -681,21 +794,9 @@ static int regcache_sync_block_single(struct regmap *map, void *block, continue; val = regcache_get_val(map, block, i); - if (!regcache_reg_needs_sync(map, regtmp, val)) - continue; - - map->cache_bypass = true; - - ret = _regmap_write(map, regtmp, val); - - map->cache_bypass = false; - if (ret != 0) { - dev_err(map->dev, "Unable to sync register %#x. %d\n", - regtmp, ret); + ret = regcache_sync_val(map, regtmp, val); + if (ret != 0) return ret; - } - dev_dbg(map->dev, "Synced register %#x, value %#x\n", - regtmp, val); } return 0; diff --git a/drivers/base/regmap/regmap-ac97.c b/drivers/base/regmap/regmap-ac97.c index b9f76bdf74a9..a561971c459c 100644 --- a/drivers/base/regmap/regmap-ac97.c +++ b/drivers/base/regmap/regmap-ac97.c @@ -86,4 +86,5 @@ struct regmap *__devm_regmap_init_ac97(struct snd_ac97 *ac97, } EXPORT_SYMBOL_GPL(__devm_regmap_init_ac97); +MODULE_DESCRIPTION("Register map access API - AC'97 support"); MODULE_LICENSE("GPL v2"); diff --git a/drivers/base/regmap/regmap-debugfs.c b/drivers/base/regmap/regmap-debugfs.c index 211a335a608d..c9b4c04b1cf6 100644 --- a/drivers/base/regmap/regmap-debugfs.c +++ b/drivers/base/regmap/regmap-debugfs.c @@ -48,7 +48,7 @@ static ssize_t regmap_name_read_file(struct file *file, name = map->dev->driver->name; ret = snprintf(buf, PAGE_SIZE, "%s\n", name); - if (ret < 0) { + if (ret >= PAGE_SIZE) { kfree(buf); return ret; } @@ -226,8 +226,8 @@ static ssize_t regmap_read_debugfs(struct regmap *map, unsigned int from, if (*ppos < 0 || !count) return -EINVAL; - if (count > (PAGE_SIZE << (MAX_ORDER - 1))) - count = PAGE_SIZE << (MAX_ORDER - 1); + if (count > (PAGE_SIZE << MAX_PAGE_ORDER)) + count = PAGE_SIZE << MAX_PAGE_ORDER; buf = kmalloc(count, GFP_KERNEL); if (!buf) @@ -368,13 +368,13 @@ static ssize_t regmap_reg_ranges_read_file(struct file *file, char *buf; char *entry; int ret; - unsigned entry_len; + unsigned int entry_len; if (*ppos < 0 || !count) return -EINVAL; - if (count > (PAGE_SIZE << (MAX_ORDER - 1))) - count = PAGE_SIZE << (MAX_ORDER - 1); + if (count > (PAGE_SIZE << MAX_PAGE_ORDER)) + count = PAGE_SIZE << MAX_PAGE_ORDER; buf = kmalloc(count, GFP_KERNEL); if (!buf) @@ -470,10 +470,6 @@ static ssize_t regmap_cache_only_write_file(struct file *file, if (err) return count; - err = debugfs_file_get(file->f_path.dentry); - if (err) - return err; - map->lock(map->lock_arg); if (new_val && !map->cache_only) { @@ -486,7 +482,6 @@ static ssize_t regmap_cache_only_write_file(struct file *file, map->cache_only = new_val; map->unlock(map->lock_arg); - debugfs_file_put(file->f_path.dentry); if (require_sync) { err = regcache_sync(map); @@ -517,10 +512,6 @@ static ssize_t regmap_cache_bypass_write_file(struct file *file, if (err) return count; - err = debugfs_file_get(file->f_path.dentry); - if (err) - return err; - map->lock(map->lock_arg); if (new_val && !map->cache_bypass) { @@ -532,7 +523,6 @@ static ssize_t regmap_cache_bypass_write_file(struct file *file, map->cache_bypass = new_val; map->unlock(map->lock_arg); - debugfs_file_put(file->f_path.dentry); return count; } @@ -598,7 +588,7 @@ void regmap_debugfs_init(struct regmap *map) map->debugfs_name = kasprintf(GFP_KERNEL, "dummy%d", dummy_index); if (!map->debugfs_name) - return; + return; name = map->debugfs_name; dummy_index++; } @@ -636,6 +626,17 @@ void regmap_debugfs_init(struct regmap *map) ®map_cache_bypass_fops); } + /* + * This could interfere with driver operation. Therefore, don't provide + * any real compile time configuration option for this feature. One will + * have to modify the source code directly in order to use it. + */ +#undef REGMAP_ALLOW_FORCE_WRITE_FIELD_DEBUGFS +#ifdef REGMAP_ALLOW_FORCE_WRITE_FIELD_DEBUGFS + debugfs_create_bool("force_write_field", 0600, map->debugfs, + &map->force_write_field); +#endif + next = rb_first(&map->range_tree); while (next) { range_node = rb_entry(next, struct regmap_range_node, node); diff --git a/drivers/base/regmap/regmap-fsi.c b/drivers/base/regmap/regmap-fsi.c new file mode 100644 index 000000000000..3d2f3cb31d5e --- /dev/null +++ b/drivers/base/regmap/regmap-fsi.c @@ -0,0 +1,231 @@ +// SPDX-License-Identifier: GPL-2.0 +// +// Register map access API - FSI support +// +// Copyright 2022 IBM Corp +// +// Author: Eddie James <eajames@linux.ibm.com> + +#include <linux/fsi.h> +#include <linux/module.h> +#include <linux/regmap.h> + +#include "internal.h" + +static int regmap_fsi32_reg_read(void *context, unsigned int reg, unsigned int *val) +{ + u32 v; + int ret; + + ret = fsi_slave_read(context, reg, &v, sizeof(v)); + if (ret) + return ret; + + *val = v; + return 0; +} + +static int regmap_fsi32_reg_write(void *context, unsigned int reg, unsigned int val) +{ + u32 v = val; + + return fsi_slave_write(context, reg, &v, sizeof(v)); +} + +static const struct regmap_bus regmap_fsi32 = { + .reg_write = regmap_fsi32_reg_write, + .reg_read = regmap_fsi32_reg_read, +}; + +static int regmap_fsi32le_reg_read(void *context, unsigned int reg, unsigned int *val) +{ + __be32 v; + int ret; + + ret = fsi_slave_read(context, reg, &v, sizeof(v)); + if (ret) + return ret; + + *val = be32_to_cpu(v); + return 0; +} + +static int regmap_fsi32le_reg_write(void *context, unsigned int reg, unsigned int val) +{ + __be32 v = cpu_to_be32(val); + + return fsi_slave_write(context, reg, &v, sizeof(v)); +} + +static const struct regmap_bus regmap_fsi32le = { + .reg_write = regmap_fsi32le_reg_write, + .reg_read = regmap_fsi32le_reg_read, +}; + +static int regmap_fsi16_reg_read(void *context, unsigned int reg, unsigned int *val) +{ + u16 v; + int ret; + + ret = fsi_slave_read(context, reg, &v, sizeof(v)); + if (ret) + return ret; + + *val = v; + return 0; +} + +static int regmap_fsi16_reg_write(void *context, unsigned int reg, unsigned int val) +{ + u16 v; + + if (val > 0xffff) + return -EINVAL; + + v = val; + return fsi_slave_write(context, reg, &v, sizeof(v)); +} + +static const struct regmap_bus regmap_fsi16 = { + .reg_write = regmap_fsi16_reg_write, + .reg_read = regmap_fsi16_reg_read, +}; + +static int regmap_fsi16le_reg_read(void *context, unsigned int reg, unsigned int *val) +{ + __be16 v; + int ret; + + ret = fsi_slave_read(context, reg, &v, sizeof(v)); + if (ret) + return ret; + + *val = be16_to_cpu(v); + return 0; +} + +static int regmap_fsi16le_reg_write(void *context, unsigned int reg, unsigned int val) +{ + __be16 v; + + if (val > 0xffff) + return -EINVAL; + + v = cpu_to_be16(val); + return fsi_slave_write(context, reg, &v, sizeof(v)); +} + +static const struct regmap_bus regmap_fsi16le = { + .reg_write = regmap_fsi16le_reg_write, + .reg_read = regmap_fsi16le_reg_read, +}; + +static int regmap_fsi8_reg_read(void *context, unsigned int reg, unsigned int *val) +{ + u8 v; + int ret; + + ret = fsi_slave_read(context, reg, &v, sizeof(v)); + if (ret) + return ret; + + *val = v; + return 0; +} + +static int regmap_fsi8_reg_write(void *context, unsigned int reg, unsigned int val) +{ + u8 v; + + if (val > 0xff) + return -EINVAL; + + v = val; + return fsi_slave_write(context, reg, &v, sizeof(v)); +} + +static const struct regmap_bus regmap_fsi8 = { + .reg_write = regmap_fsi8_reg_write, + .reg_read = regmap_fsi8_reg_read, +}; + +static const struct regmap_bus *regmap_get_fsi_bus(struct fsi_device *fsi_dev, + const struct regmap_config *config) +{ + const struct regmap_bus *bus = NULL; + + if (config->reg_bits == 8 || config->reg_bits == 16 || config->reg_bits == 32) { + switch (config->val_bits) { + case 8: + bus = ®map_fsi8; + break; + case 16: + switch (regmap_get_val_endian(&fsi_dev->dev, NULL, config)) { + case REGMAP_ENDIAN_LITTLE: +#ifdef __LITTLE_ENDIAN + case REGMAP_ENDIAN_NATIVE: +#endif + bus = ®map_fsi16le; + break; + case REGMAP_ENDIAN_DEFAULT: + case REGMAP_ENDIAN_BIG: +#ifdef __BIG_ENDIAN + case REGMAP_ENDIAN_NATIVE: +#endif + bus = ®map_fsi16; + break; + default: + break; + } + break; + case 32: + switch (regmap_get_val_endian(&fsi_dev->dev, NULL, config)) { + case REGMAP_ENDIAN_LITTLE: +#ifdef __LITTLE_ENDIAN + case REGMAP_ENDIAN_NATIVE: +#endif + bus = ®map_fsi32le; + break; + case REGMAP_ENDIAN_DEFAULT: + case REGMAP_ENDIAN_BIG: +#ifdef __BIG_ENDIAN + case REGMAP_ENDIAN_NATIVE: +#endif + bus = ®map_fsi32; + break; + default: + break; + } + break; + } + } + + return bus ?: ERR_PTR(-EOPNOTSUPP); +} + +struct regmap *__regmap_init_fsi(struct fsi_device *fsi_dev, const struct regmap_config *config, + struct lock_class_key *lock_key, const char *lock_name) +{ + const struct regmap_bus *bus = regmap_get_fsi_bus(fsi_dev, config); + + if (IS_ERR(bus)) + return ERR_CAST(bus); + + return __regmap_init(&fsi_dev->dev, bus, fsi_dev->slave, config, lock_key, lock_name); +} +EXPORT_SYMBOL_GPL(__regmap_init_fsi); + +struct regmap *__devm_regmap_init_fsi(struct fsi_device *fsi_dev, + const struct regmap_config *config, + struct lock_class_key *lock_key, const char *lock_name) +{ + const struct regmap_bus *bus = regmap_get_fsi_bus(fsi_dev, config); + + if (IS_ERR(bus)) + return ERR_CAST(bus); + + return __devm_regmap_init(&fsi_dev->dev, bus, fsi_dev->slave, config, lock_key, lock_name); +} +EXPORT_SYMBOL_GPL(__devm_regmap_init_fsi); + +MODULE_LICENSE("GPL"); diff --git a/drivers/base/regmap/regmap-i2c.c b/drivers/base/regmap/regmap-i2c.c index 980e5ce6a3a3..c9b39a02278e 100644 --- a/drivers/base/regmap/regmap-i2c.c +++ b/drivers/base/regmap/regmap-i2c.c @@ -242,8 +242,8 @@ static int regmap_i2c_smbus_i2c_read(void *context, const void *reg, static const struct regmap_bus regmap_i2c_smbus_i2c_block = { .write = regmap_i2c_smbus_i2c_write, .read = regmap_i2c_smbus_i2c_read, - .max_raw_read = I2C_SMBUS_BLOCK_MAX, - .max_raw_write = I2C_SMBUS_BLOCK_MAX, + .max_raw_read = I2C_SMBUS_BLOCK_MAX - 1, + .max_raw_write = I2C_SMBUS_BLOCK_MAX - 1, }; static int regmap_i2c_smbus_i2c_write_reg16(void *context, const void *data, @@ -299,8 +299,8 @@ static int regmap_i2c_smbus_i2c_read_reg16(void *context, const void *reg, static const struct regmap_bus regmap_i2c_smbus_i2c_block_reg16 = { .write = regmap_i2c_smbus_i2c_write_reg16, .read = regmap_i2c_smbus_i2c_read_reg16, - .max_raw_read = I2C_SMBUS_BLOCK_MAX, - .max_raw_write = I2C_SMBUS_BLOCK_MAX, + .max_raw_read = I2C_SMBUS_BLOCK_MAX - 2, + .max_raw_write = I2C_SMBUS_BLOCK_MAX - 2, }; static const struct regmap_bus *regmap_get_i2c_bus(struct i2c_client *i2c, @@ -350,7 +350,8 @@ static const struct regmap_bus *regmap_get_i2c_bus(struct i2c_client *i2c, if (quirks->max_write_len && (bus->max_raw_write == 0 || bus->max_raw_write > quirks->max_write_len)) - max_write = quirks->max_write_len; + max_write = quirks->max_write_len - + (config->reg_bits + config->pad_bits) / BITS_PER_BYTE; if (max_read || max_write) { ret_bus = kmemdup(bus, sizeof(*bus), GFP_KERNEL); @@ -396,4 +397,5 @@ struct regmap *__devm_regmap_init_i2c(struct i2c_client *i2c, } EXPORT_SYMBOL_GPL(__devm_regmap_init_i2c); +MODULE_DESCRIPTION("Register map access API - I2C support"); MODULE_LICENSE("GPL"); diff --git a/drivers/base/regmap/regmap-i3c.c b/drivers/base/regmap/regmap-i3c.c index 1578fb506683..863b348704dc 100644 --- a/drivers/base/regmap/regmap-i3c.c +++ b/drivers/base/regmap/regmap-i3c.c @@ -1,6 +1,7 @@ // SPDX-License-Identifier: GPL-2.0 // Copyright (c) 2018 Synopsys, Inc. and/or its affiliates. +#include <linux/array_size.h> #include <linux/regmap.h> #include <linux/i3c/device.h> #include <linux/i3c/master.h> @@ -10,7 +11,7 @@ static int regmap_i3c_write(void *context, const void *data, size_t count) { struct device *dev = context; struct i3c_device *i3c = dev_to_i3cdev(dev); - struct i3c_priv_xfer xfers[] = { + struct i3c_xfer xfers[] = { { .rnw = false, .len = count, @@ -18,7 +19,7 @@ static int regmap_i3c_write(void *context, const void *data, size_t count) }, }; - return i3c_device_do_priv_xfers(i3c, xfers, 1); + return i3c_device_do_xfers(i3c, xfers, ARRAY_SIZE(xfers), I3C_SDR); } static int regmap_i3c_read(void *context, @@ -27,7 +28,7 @@ static int regmap_i3c_read(void *context, { struct device *dev = context; struct i3c_device *i3c = dev_to_i3cdev(dev); - struct i3c_priv_xfer xfers[2]; + struct i3c_xfer xfers[2]; xfers[0].rnw = false; xfers[0].len = reg_size; @@ -37,10 +38,10 @@ static int regmap_i3c_read(void *context, xfers[1].len = val_size; xfers[1].data.in = val; - return i3c_device_do_priv_xfers(i3c, xfers, 2); + return i3c_device_do_xfers(i3c, xfers, ARRAY_SIZE(xfers), I3C_SDR); } -static struct regmap_bus regmap_i3c = { +static const struct regmap_bus regmap_i3c = { .write = regmap_i3c_write, .read = regmap_i3c_read, }; @@ -56,5 +57,5 @@ struct regmap *__devm_regmap_init_i3c(struct i3c_device *i3c, EXPORT_SYMBOL_GPL(__devm_regmap_init_i3c); MODULE_AUTHOR("Vitor Soares <vitor.soares@synopsys.com>"); -MODULE_DESCRIPTION("Regmap I3C Module"); +MODULE_DESCRIPTION("regmap I3C Module"); MODULE_LICENSE("GPL v2"); diff --git a/drivers/base/regmap/regmap-irq.c b/drivers/base/regmap/regmap-irq.c index d2656581a608..6112d942499b 100644 --- a/drivers/base/regmap/regmap-irq.c +++ b/drivers/base/regmap/regmap-irq.c @@ -6,11 +6,13 @@ // // Author: Mark Brown <broonie@opensource.wolfsonmicro.com> +#include <linux/array_size.h> #include <linux/device.h> #include <linux/export.h> #include <linux/interrupt.h> #include <linux/irq.h> #include <linux/irqdomain.h> +#include <linux/overflow.h> #include <linux/pm_runtime.h> #include <linux/regmap.h> #include <linux/slab.h> @@ -19,6 +21,7 @@ struct regmap_irq_chip_data { struct mutex lock; + struct lock_class_key lock_key; struct irq_chip irq_chip; struct regmap *map; @@ -33,39 +36,21 @@ struct regmap_irq_chip_data { void *status_reg_buf; unsigned int *main_status_buf; unsigned int *status_buf; + unsigned int *prev_status_buf; unsigned int *mask_buf; unsigned int *mask_buf_def; unsigned int *wake_buf; unsigned int *type_buf; unsigned int *type_buf_def; - unsigned int **virt_buf; + unsigned int **config_buf; unsigned int irq_reg_stride; - unsigned int type_reg_stride; - bool clear_status:1; -}; - -static int sub_irq_reg(struct regmap_irq_chip_data *data, - unsigned int base_reg, int i) -{ - const struct regmap_irq_chip *chip = data->chip; - struct regmap *map = data->map; - struct regmap_irq_sub_irq_map *subreg; - unsigned int offset; - int reg = 0; - - if (!chip->sub_reg_offsets || !chip->not_fixed_stride) { - /* Assume linear mapping */ - reg = base_reg + (i * map->reg_stride * data->irq_reg_stride); - } else { - subreg = &chip->sub_reg_offsets[i]; - offset = subreg->offset[0]; - reg = base_reg + offset; - } + unsigned int (*get_irq_reg)(struct regmap_irq_chip_data *data, + unsigned int base, int index); - return reg; -} + unsigned int clear_status:1; +}; static inline const struct regmap_irq *irq_to_regmap_irq(struct regmap_irq_chip_data *data, @@ -74,21 +59,25 @@ struct regmap_irq *irq_to_regmap_irq(struct regmap_irq_chip_data *data, return &data->chip->irqs[irq]; } -static void regmap_irq_lock(struct irq_data *data) +static bool regmap_irq_can_bulk_read_status(struct regmap_irq_chip_data *data) { - struct regmap_irq_chip_data *d = irq_data_get_irq_chip_data(data); + struct regmap *map = data->map; - mutex_lock(&d->lock); + /* + * While possible that a user-defined ->get_irq_reg() callback might + * be linear enough to support bulk reads, most of the time it won't. + * Therefore only allow them if the default callback is being used. + */ + return data->irq_reg_stride == 1 && map->reg_stride == 1 && + data->get_irq_reg == regmap_irq_get_irq_reg_linear && + !map->use_single_read; } -static int regmap_irq_update_bits(struct regmap_irq_chip_data *d, - unsigned int reg, unsigned int mask, - unsigned int val) +static void regmap_irq_lock(struct irq_data *data) { - if (d->chip->mask_writeonly) - return regmap_write_bits(d->map, reg, mask, val); - else - return regmap_update_bits(d->map, reg, mask, val); + struct regmap_irq_chip_data *d = irq_data_get_irq_chip_data(data); + + mutex_lock(&d->lock); } static void regmap_irq_sync_unlock(struct irq_data *data) @@ -97,7 +86,6 @@ static void regmap_irq_sync_unlock(struct irq_data *data) struct regmap *map = d->map; int i, j, ret; u32 reg; - u32 unmask_offset; u32 val; if (d->chip->runtime_pm) { @@ -109,7 +97,7 @@ static void regmap_irq_sync_unlock(struct irq_data *data) if (d->clear_status) { for (i = 0; i < d->chip->num_regs; i++) { - reg = sub_irq_reg(d, d->chip->status_base, i); + reg = d->get_irq_reg(d, d->chip->status_base, i); ret = regmap_read(map, reg, &val); if (ret) @@ -126,44 +114,37 @@ static void regmap_irq_sync_unlock(struct irq_data *data) * suppress pointless writes. */ for (i = 0; i < d->chip->num_regs; i++) { - if (!d->chip->mask_base) - continue; + if (d->chip->handle_mask_sync) + d->chip->handle_mask_sync(i, d->mask_buf_def[i], + d->mask_buf[i], + d->chip->irq_drv_data); + + if (d->chip->mask_base && !d->chip->handle_mask_sync) { + reg = d->get_irq_reg(d, d->chip->mask_base, i); + ret = regmap_update_bits(d->map, reg, + d->mask_buf_def[i], + d->mask_buf[i]); + if (ret) + dev_err(d->map->dev, "Failed to sync masks in %x\n", reg); + } - reg = sub_irq_reg(d, d->chip->mask_base, i); - if (d->chip->mask_invert) { - ret = regmap_irq_update_bits(d, reg, - d->mask_buf_def[i], ~d->mask_buf[i]); - } else if (d->chip->unmask_base) { - /* set mask with mask_base register */ - ret = regmap_irq_update_bits(d, reg, + if (d->chip->unmask_base && !d->chip->handle_mask_sync) { + reg = d->get_irq_reg(d, d->chip->unmask_base, i); + ret = regmap_update_bits(d->map, reg, d->mask_buf_def[i], ~d->mask_buf[i]); - if (ret < 0) - dev_err(d->map->dev, - "Failed to sync unmasks in %x\n", + if (ret) + dev_err(d->map->dev, "Failed to sync masks in %x\n", reg); - unmask_offset = d->chip->unmask_base - - d->chip->mask_base; - /* clear mask with unmask_base register */ - ret = regmap_irq_update_bits(d, - reg + unmask_offset, - d->mask_buf_def[i], - d->mask_buf[i]); - } else { - ret = regmap_irq_update_bits(d, reg, - d->mask_buf_def[i], d->mask_buf[i]); } - if (ret != 0) - dev_err(d->map->dev, "Failed to sync masks in %x\n", - reg); - reg = sub_irq_reg(d, d->chip->wake_base, i); + reg = d->get_irq_reg(d, d->chip->wake_base, i); if (d->wake_buf) { if (d->chip->wake_invert) - ret = regmap_irq_update_bits(d, reg, + ret = regmap_update_bits(d->map, reg, d->mask_buf_def[i], ~d->wake_buf[i]); else - ret = regmap_irq_update_bits(d, reg, + ret = regmap_update_bits(d->map, reg, d->mask_buf_def[i], d->wake_buf[i]); if (ret != 0) @@ -180,7 +161,7 @@ static void regmap_irq_sync_unlock(struct irq_data *data) * it'll be ignored in irq handler, then may introduce irq storm */ if (d->mask_buf[i] && (d->chip->ack_base || d->chip->use_ack)) { - reg = sub_irq_reg(d, d->chip->ack_base, i); + reg = d->get_irq_reg(d, d->chip->ack_base, i); /* some chips ack by write 0 */ if (d->chip->ack_invert) @@ -189,11 +170,9 @@ static void regmap_irq_sync_unlock(struct irq_data *data) ret = regmap_write(map, reg, d->mask_buf[i]); if (d->chip->clear_ack) { if (d->chip->ack_invert && !ret) - ret = regmap_write(map, reg, - d->mask_buf[i]); + ret = regmap_write(map, reg, UINT_MAX); else if (!ret) - ret = regmap_write(map, reg, - ~d->mask_buf[i]); + ret = regmap_write(map, reg, 0); } if (ret != 0) dev_err(d->map->dev, "Failed to ack 0x%x: %d\n", @@ -201,35 +180,14 @@ static void regmap_irq_sync_unlock(struct irq_data *data) } } - /* Don't update the type bits if we're using mask bits for irq type. */ - if (!d->chip->type_in_mask) { - for (i = 0; i < d->chip->num_type_reg; i++) { - if (!d->type_buf_def[i]) - continue; - reg = sub_irq_reg(d, d->chip->type_base, i); - if (d->chip->type_invert) - ret = regmap_irq_update_bits(d, reg, - d->type_buf_def[i], ~d->type_buf[i]); - else - ret = regmap_irq_update_bits(d, reg, - d->type_buf_def[i], d->type_buf[i]); - if (ret != 0) - dev_err(d->map->dev, "Failed to sync type in %x\n", - reg); - } - } - - if (d->chip->num_virt_regs) { - for (i = 0; i < d->chip->num_virt_regs; i++) { - for (j = 0; j < d->chip->num_regs; j++) { - reg = sub_irq_reg(d, d->chip->virt_reg_base[i], - j); - ret = regmap_write(map, reg, d->virt_buf[i][j]); - if (ret != 0) - dev_err(d->map->dev, - "Failed to write virt 0x%x: %d\n", - reg, ret); - } + for (i = 0; i < d->chip->num_config_bases; i++) { + for (j = 0; j < d->chip->num_config_regs; j++) { + reg = d->get_irq_reg(d, d->chip->config_base[i], j); + ret = regmap_write(map, reg, d->config_buf[i][j]); + if (ret) + dev_err(d->map->dev, + "Failed to write config %x: %d\n", + reg, ret); } } @@ -239,10 +197,10 @@ static void regmap_irq_sync_unlock(struct irq_data *data) /* If we've changed our wakeup count propagate it to the parent */ if (d->wake_count < 0) for (i = d->wake_count; i < 0; i++) - irq_set_irq_wake(d->irq, 0); + disable_irq_wake(d->irq); else if (d->wake_count > 0) for (i = 0; i < d->wake_count; i++) - irq_set_irq_wake(d->irq, 1); + enable_irq_wake(d->irq); d->wake_count = 0; @@ -254,30 +212,28 @@ static void regmap_irq_enable(struct irq_data *data) struct regmap_irq_chip_data *d = irq_data_get_irq_chip_data(data); struct regmap *map = d->map; const struct regmap_irq *irq_data = irq_to_regmap_irq(d, data->hwirq); - unsigned int mask, type; - - type = irq_data->type.type_falling_val | irq_data->type.type_rising_val; + unsigned int reg = irq_data->reg_offset / map->reg_stride; + unsigned int mask; /* * The type_in_mask flag means that the underlying hardware uses - * separate mask bits for rising and falling edge interrupts, but - * we want to make them into a single virtual interrupt with - * configurable edge. + * separate mask bits for each interrupt trigger type, but we want + * to have a single logical interrupt with a configurable type. * - * If the interrupt we're enabling defines the falling or rising - * masks then instead of using the regular mask bits for this - * interrupt, use the value previously written to the type buffer - * at the corresponding offset in regmap_irq_set_type(). + * If the interrupt we're enabling defines any supported types + * then instead of using the regular mask bits for this interrupt, + * use the value previously written to the type buffer at the + * corresponding offset in regmap_irq_set_type(). */ - if (d->chip->type_in_mask && type) - mask = d->type_buf[irq_data->reg_offset / map->reg_stride]; + if (d->chip->type_in_mask && irq_data->type.types_supported) + mask = d->type_buf[reg] & irq_data->mask; else mask = irq_data->mask; if (d->chip->clear_on_unmask) d->clear_status = true; - d->mask_buf[irq_data->reg_offset / map->reg_stride] &= ~mask; + d->mask_buf[reg] &= ~mask; } static void regmap_irq_disable(struct irq_data *data) @@ -294,7 +250,7 @@ static int regmap_irq_set_type(struct irq_data *data, unsigned int type) struct regmap_irq_chip_data *d = irq_data_get_irq_chip_data(data); struct regmap *map = d->map; const struct regmap_irq *irq_data = irq_to_regmap_irq(d, data->hwirq); - int reg; + int reg, ret; const struct regmap_irq_type *t = &irq_data->type; if ((t->types_supported & type) != type) @@ -302,41 +258,19 @@ static int regmap_irq_set_type(struct irq_data *data, unsigned int type) reg = t->type_reg_offset / map->reg_stride; - if (t->type_reg_mask) - d->type_buf[reg] &= ~t->type_reg_mask; - else - d->type_buf[reg] &= ~(t->type_falling_val | - t->type_rising_val | - t->type_level_low_val | - t->type_level_high_val); - switch (type) { - case IRQ_TYPE_EDGE_FALLING: - d->type_buf[reg] |= t->type_falling_val; - break; - - case IRQ_TYPE_EDGE_RISING: - d->type_buf[reg] |= t->type_rising_val; - break; - - case IRQ_TYPE_EDGE_BOTH: - d->type_buf[reg] |= (t->type_falling_val | - t->type_rising_val); - break; - - case IRQ_TYPE_LEVEL_HIGH: - d->type_buf[reg] |= t->type_level_high_val; - break; - - case IRQ_TYPE_LEVEL_LOW: - d->type_buf[reg] |= t->type_level_low_val; - break; - default: - return -EINVAL; + if (d->chip->type_in_mask) { + ret = regmap_irq_set_type_config_simple(&d->type_buf, type, + irq_data, reg, d->chip->irq_drv_data); + if (ret) + return ret; } - if (d->chip->set_type_virt) - return d->chip->set_type_virt(d->virt_buf, type, data->hwirq, - reg); + if (d->chip->set_type_config) { + ret = d->chip->set_type_config(d->config_buf, type, irq_data, + reg, d->chip->irq_drv_data); + if (ret) + return ret; + } return 0; } @@ -375,29 +309,26 @@ static inline int read_sub_irq_data(struct regmap_irq_chip_data *data, unsigned int b) { const struct regmap_irq_chip *chip = data->chip; + const struct regmap_irq_sub_irq_map *subreg; struct regmap *map = data->map; - struct regmap_irq_sub_irq_map *subreg; + unsigned int reg; int i, ret = 0; if (!chip->sub_reg_offsets) { - /* Assume linear mapping */ - ret = regmap_read(map, chip->status_base + - (b * map->reg_stride * data->irq_reg_stride), - &data->status_buf[b]); + reg = data->get_irq_reg(data, chip->status_base, b); + ret = regmap_read(map, reg, &data->status_buf[b]); } else { + /* + * Note we can't use ->get_irq_reg() here because the offsets + * in 'subreg' are *not* interchangeable with indices. + */ subreg = &chip->sub_reg_offsets[b]; for (i = 0; i < subreg->num_regs; i++) { unsigned int offset = subreg->offset[i]; + unsigned int index = offset / map->reg_stride; - if (chip->not_fixed_stride) - ret = regmap_read(map, - chip->status_base + offset, - &data->status_buf[b]); - else - ret = regmap_read(map, - chip->status_base + offset, - &data->status_buf[offset]); - + ret = regmap_read(map, chip->status_base + offset, + &data->status_buf[index]); if (ret) break; } @@ -405,43 +336,29 @@ static inline int read_sub_irq_data(struct regmap_irq_chip_data *data, return ret; } -static irqreturn_t regmap_irq_thread(int irq, void *d) +static int read_irq_data(struct regmap_irq_chip_data *data) { - struct regmap_irq_chip_data *data = d; const struct regmap_irq_chip *chip = data->chip; struct regmap *map = data->map; int ret, i; - bool handled = false; u32 reg; - if (chip->handle_pre_irq) - chip->handle_pre_irq(chip->irq_drv_data); - - if (chip->runtime_pm) { - ret = pm_runtime_get_sync(map->dev); - if (ret < 0) { - dev_err(map->dev, "IRQ thread failed to resume: %d\n", - ret); - goto exit; - } - } - /* * Read only registers with active IRQs if the chip has 'main status * register'. Else read in the statuses, using a single bulk read if * possible in order to reduce the I/O overheads. */ - if (chip->num_main_regs) { + if (chip->no_status) { + /* no status register so default to all active */ + memset32(data->status_buf, GENMASK(31, 0), chip->num_regs); + } else if (chip->num_main_regs) { unsigned int max_main_bits; - unsigned long size; - - size = chip->num_regs * sizeof(unsigned int); max_main_bits = (chip->num_main_status_bits) ? chip->num_main_status_bits : chip->num_regs; /* Clear the status buf as we don't read all status regs */ - memset(data->status_buf, 0, size); + memset32(data->status_buf, 0, chip->num_regs); /* We could support bulk read for main status registers * but I don't expect to see devices with really many main @@ -449,15 +366,11 @@ static irqreturn_t regmap_irq_thread(int irq, void *d) * sake of simplicity. and add bulk reads only if needed */ for (i = 0; i < chip->num_main_regs; i++) { - ret = regmap_read(map, chip->main_status + - (i * map->reg_stride - * data->irq_reg_stride), - &data->main_status_buf[i]); + reg = data->get_irq_reg(data, chip->main_status, i); + ret = regmap_read(map, reg, &data->main_status_buf[i]); if (ret) { - dev_err(map->dev, - "Failed to read IRQ status %d\n", - ret); - goto exit; + dev_err(map->dev, "Failed to read IRQ status %d\n", ret); + return ret; } } @@ -473,16 +386,13 @@ static irqreturn_t regmap_irq_thread(int irq, void *d) ret = read_sub_irq_data(data, b); if (ret != 0) { - dev_err(map->dev, - "Failed to read IRQ status %d\n", - ret); - goto exit; + dev_err(map->dev, "Failed to read IRQ status %d\n", ret); + return ret; } } } - } else if (!map->use_single_read && map->reg_stride == 1 && - data->irq_reg_stride == 1) { + } else if (regmap_irq_can_bulk_read_status(data)) { u8 *buf8 = data->status_reg_buf; u16 *buf16 = data->status_reg_buf; @@ -494,9 +404,8 @@ static irqreturn_t regmap_irq_thread(int irq, void *d) data->status_reg_buf, chip->num_regs); if (ret != 0) { - dev_err(map->dev, "Failed to read IRQ status: %d\n", - ret); - goto exit; + dev_err(map->dev, "Failed to read IRQ status: %d\n", ret); + return ret; } for (i = 0; i < data->chip->num_regs; i++) { @@ -512,21 +421,19 @@ static irqreturn_t regmap_irq_thread(int irq, void *d) break; default: BUG(); - goto exit; + return -EIO; } } } else { for (i = 0; i < data->chip->num_regs; i++) { - unsigned int reg = sub_irq_reg(data, + unsigned int reg = data->get_irq_reg(data, data->chip->status_base, i); ret = regmap_read(map, reg, &data->status_buf[i]); if (ret != 0) { - dev_err(map->dev, - "Failed to read IRQ status: %d\n", - ret); - goto exit; + dev_err(map->dev, "Failed to read IRQ status: %d\n", ret); + return ret; } } } @@ -535,9 +442,45 @@ static irqreturn_t regmap_irq_thread(int irq, void *d) for (i = 0; i < data->chip->num_regs; i++) data->status_buf[i] = ~data->status_buf[i]; + return 0; +} + +static irqreturn_t regmap_irq_thread(int irq, void *d) +{ + struct regmap_irq_chip_data *data = d; + const struct regmap_irq_chip *chip = data->chip; + struct regmap *map = data->map; + int ret, i; + bool handled = false; + u32 reg; + + if (chip->handle_pre_irq) + chip->handle_pre_irq(chip->irq_drv_data); + + if (chip->runtime_pm) { + ret = pm_runtime_get_sync(map->dev); + if (ret < 0) { + dev_err(map->dev, "IRQ thread failed to resume: %d\n", ret); + goto exit; + } + } + + ret = read_irq_data(data); + if (ret < 0) + goto exit; + + if (chip->status_is_level) { + for (i = 0; i < data->chip->num_regs; i++) { + unsigned int val = data->status_buf[i]; + + data->status_buf[i] ^= data->prev_status_buf[i]; + data->prev_status_buf[i] = val; + } + } + /* * Ignore masked IRQs and ack if we need to; we ack early so - * there is no race between handling and acknowleding the + * there is no race between handling and acknowledging the * interrupt. We assume that typically few of the interrupts * will fire simultaneously so don't worry about overhead from * doing a write per register. @@ -546,7 +489,7 @@ static irqreturn_t regmap_irq_thread(int irq, void *d) data->status_buf[i] &= ~data->mask_buf[i]; if (data->status_buf[i] && (chip->ack_base || chip->use_ack)) { - reg = sub_irq_reg(data, data->chip->ack_base, i); + reg = data->get_irq_reg(data, data->chip->ack_base, i); if (chip->ack_invert) ret = regmap_write(map, reg, @@ -556,11 +499,9 @@ static irqreturn_t regmap_irq_thread(int irq, void *d) data->status_buf[i]); if (chip->clear_ack) { if (chip->ack_invert && !ret) - ret = regmap_write(map, reg, - data->status_buf[i]); + ret = regmap_write(map, reg, UINT_MAX); else if (!ret) - ret = regmap_write(map, reg, - ~data->status_buf[i]); + ret = regmap_write(map, reg, 0); } if (ret != 0) dev_err(map->dev, "Failed to ack 0x%x: %d\n", @@ -577,24 +518,28 @@ static irqreturn_t regmap_irq_thread(int irq, void *d) } exit: - if (chip->runtime_pm) - pm_runtime_put(map->dev); - if (chip->handle_post_irq) chip->handle_post_irq(chip->irq_drv_data); + if (chip->runtime_pm) + pm_runtime_put(map->dev); + if (handled) return IRQ_HANDLED; else return IRQ_NONE; } +static struct lock_class_key regmap_irq_lock_class; +static struct lock_class_key regmap_irq_request_class; + static int regmap_irq_map(struct irq_domain *h, unsigned int virq, irq_hw_number_t hw) { struct regmap_irq_chip_data *data = h->host_data; irq_set_chip_data(virq, data); + irq_set_lockdep_class(virq, ®map_irq_lock_class, ®map_irq_request_class); irq_set_chip(virq, &data->irq_chip); irq_set_nested_thread(virq, 1); irq_set_parent(virq, data->irq); @@ -609,6 +554,105 @@ static const struct irq_domain_ops regmap_domain_ops = { }; /** + * regmap_irq_get_irq_reg_linear() - Linear IRQ register mapping callback. + * @data: Data for the &struct regmap_irq_chip + * @base: Base register + * @index: Register index + * + * Returns the register address corresponding to the given @base and @index + * by the formula ``base + index * regmap_stride * irq_reg_stride``. + */ +unsigned int regmap_irq_get_irq_reg_linear(struct regmap_irq_chip_data *data, + unsigned int base, int index) +{ + struct regmap *map = data->map; + + return base + index * map->reg_stride * data->irq_reg_stride; +} +EXPORT_SYMBOL_GPL(regmap_irq_get_irq_reg_linear); + +/** + * regmap_irq_set_type_config_simple() - Simple IRQ type configuration callback. + * @buf: Buffer containing configuration register values, this is a 2D array of + * `num_config_bases` rows, each of `num_config_regs` elements. + * @type: The requested IRQ type. + * @irq_data: The IRQ being configured. + * @idx: Index of the irq's config registers within each array `buf[i]` + * @irq_drv_data: Driver specific IRQ data + * + * This is a &struct regmap_irq_chip->set_type_config callback suitable for + * chips with one config register. Register values are updated according to + * the &struct regmap_irq_type data associated with an IRQ. + */ +int regmap_irq_set_type_config_simple(unsigned int **buf, unsigned int type, + const struct regmap_irq *irq_data, + int idx, void *irq_drv_data) +{ + const struct regmap_irq_type *t = &irq_data->type; + + if (t->type_reg_mask) + buf[0][idx] &= ~t->type_reg_mask; + else + buf[0][idx] &= ~(t->type_falling_val | + t->type_rising_val | + t->type_level_low_val | + t->type_level_high_val); + + switch (type) { + case IRQ_TYPE_EDGE_FALLING: + buf[0][idx] |= t->type_falling_val; + break; + + case IRQ_TYPE_EDGE_RISING: + buf[0][idx] |= t->type_rising_val; + break; + + case IRQ_TYPE_EDGE_BOTH: + buf[0][idx] |= (t->type_falling_val | + t->type_rising_val); + break; + + case IRQ_TYPE_LEVEL_HIGH: + buf[0][idx] |= t->type_level_high_val; + break; + + case IRQ_TYPE_LEVEL_LOW: + buf[0][idx] |= t->type_level_low_val; + break; + + default: + return -EINVAL; + } + + return 0; +} +EXPORT_SYMBOL_GPL(regmap_irq_set_type_config_simple); + +static int regmap_irq_create_domain(struct fwnode_handle *fwnode, int irq_base, + const struct regmap_irq_chip *chip, + struct regmap_irq_chip_data *d) +{ + struct irq_domain_info info = { + .fwnode = fwnode, + .size = chip->num_irqs, + .hwirq_max = chip->num_irqs, + .virq_base = irq_base, + .ops = ®map_domain_ops, + .host_data = d, + .name_suffix = chip->domain_suffix, + }; + + d->domain = irq_domain_instantiate(&info); + if (IS_ERR(d->domain)) { + dev_err(d->map->dev, "Failed to create IRQ domain\n"); + return PTR_ERR(d->domain); + } + + return 0; +} + + +/** * regmap_add_irq_chip_fwnode() - Use standard regmap IRQ controller handling * * @fwnode: The firmware node where the IRQ domain should be added to. @@ -634,9 +678,7 @@ int regmap_add_irq_chip_fwnode(struct fwnode_handle *fwnode, struct regmap_irq_chip_data *d; int i; int ret = -ENOMEM; - int num_type_reg; u32 reg; - u32 unmask_offset; if (chip->num_regs <= 0) return -EINVAL; @@ -644,6 +686,9 @@ int regmap_add_irq_chip_fwnode(struct fwnode_handle *fwnode, if (chip->clear_on_unmask && (chip->ack_base || chip->use_ack)) return -EINVAL; + if (chip->mask_base && chip->unmask_base && !chip->mask_unmask_non_inverted) + return -EINVAL; + for (i = 0; i < chip->num_irqs; i++) { if (chip->irqs[i].reg_offset % map->reg_stride) return -EINVAL; @@ -652,12 +697,6 @@ int regmap_add_irq_chip_fwnode(struct fwnode_handle *fwnode, return -EINVAL; } - if (chip->not_fixed_stride) { - for (i = 0; i < chip->num_regs; i++) - if (chip->sub_reg_offsets[i].num_regs != 1) - return -EINVAL; - } - if (irq_base) { irq_base = irq_alloc_descs(irq_base, 0, chip->num_irqs, 0); if (irq_base < 0) { @@ -673,62 +712,67 @@ int regmap_add_irq_chip_fwnode(struct fwnode_handle *fwnode, if (chip->num_main_regs) { d->main_status_buf = kcalloc(chip->num_main_regs, - sizeof(unsigned int), + sizeof(*d->main_status_buf), GFP_KERNEL); if (!d->main_status_buf) goto err_alloc; } - d->status_buf = kcalloc(chip->num_regs, sizeof(unsigned int), + d->status_buf = kcalloc(chip->num_regs, sizeof(*d->status_buf), GFP_KERNEL); if (!d->status_buf) goto err_alloc; - d->mask_buf = kcalloc(chip->num_regs, sizeof(unsigned int), + if (chip->status_is_level) { + d->prev_status_buf = kcalloc(chip->num_regs, sizeof(*d->prev_status_buf), + GFP_KERNEL); + if (!d->prev_status_buf) + goto err_alloc; + } + + d->mask_buf = kcalloc(chip->num_regs, sizeof(*d->mask_buf), GFP_KERNEL); if (!d->mask_buf) goto err_alloc; - d->mask_buf_def = kcalloc(chip->num_regs, sizeof(unsigned int), + d->mask_buf_def = kcalloc(chip->num_regs, sizeof(*d->mask_buf_def), GFP_KERNEL); if (!d->mask_buf_def) goto err_alloc; if (chip->wake_base) { - d->wake_buf = kcalloc(chip->num_regs, sizeof(unsigned int), + d->wake_buf = kcalloc(chip->num_regs, sizeof(*d->wake_buf), GFP_KERNEL); if (!d->wake_buf) goto err_alloc; } - num_type_reg = chip->type_in_mask ? chip->num_regs : chip->num_type_reg; - if (num_type_reg) { - d->type_buf_def = kcalloc(num_type_reg, - sizeof(unsigned int), GFP_KERNEL); + if (chip->type_in_mask) { + d->type_buf_def = kcalloc(chip->num_regs, + sizeof(*d->type_buf_def), GFP_KERNEL); if (!d->type_buf_def) goto err_alloc; - d->type_buf = kcalloc(num_type_reg, sizeof(unsigned int), - GFP_KERNEL); + d->type_buf = kcalloc(chip->num_regs, sizeof(*d->type_buf), GFP_KERNEL); if (!d->type_buf) goto err_alloc; } - if (chip->num_virt_regs) { + if (chip->num_config_bases && chip->num_config_regs) { /* - * Create virt_buf[chip->num_extra_config_regs][chip->num_regs] + * Create config_buf[num_config_bases][num_config_regs] */ - d->virt_buf = kcalloc(chip->num_virt_regs, sizeof(*d->virt_buf), - GFP_KERNEL); - if (!d->virt_buf) + d->config_buf = kcalloc(chip->num_config_bases, + sizeof(*d->config_buf), GFP_KERNEL); + if (!d->config_buf) goto err_alloc; - for (i = 0; i < chip->num_virt_regs; i++) { - d->virt_buf[i] = kcalloc(chip->num_regs, - sizeof(unsigned int), - GFP_KERNEL); - if (!d->virt_buf[i]) + for (i = 0; i < chip->num_config_bases; i++) { + d->config_buf[i] = kcalloc(chip->num_config_regs, + sizeof(**d->config_buf), + GFP_KERNEL); + if (!d->config_buf[i]) goto err_alloc; } } @@ -745,13 +789,12 @@ int regmap_add_irq_chip_fwnode(struct fwnode_handle *fwnode, else d->irq_reg_stride = 1; - if (chip->type_reg_stride) - d->type_reg_stride = chip->type_reg_stride; + if (chip->get_irq_reg) + d->get_irq_reg = chip->get_irq_reg; else - d->type_reg_stride = 1; + d->get_irq_reg = regmap_irq_get_irq_reg_linear; - if (!map->use_single_read && map->reg_stride == 1 && - d->irq_reg_stride == 1) { + if (regmap_irq_can_bulk_read_status(d)) { d->status_reg_buf = kmalloc_array(chip->num_regs, map->format.val_bytes, GFP_KERNEL); @@ -759,7 +802,13 @@ int regmap_add_irq_chip_fwnode(struct fwnode_handle *fwnode, goto err_alloc; } - mutex_init(&d->lock); + /* + * If one regmap-irq is the parent of another then we'll try + * to lock the child with the parent locked, use an explicit + * lock_key so lockdep can figure out what's going on. + */ + lockdep_register_key(&d->lock_key); + mutex_init_with_key(&d->lock, &d->lock_key); for (i = 0; i < chip->num_irqs; i++) d->mask_buf_def[chip->irqs[i].reg_offset / map->reg_stride] @@ -768,47 +817,60 @@ int regmap_add_irq_chip_fwnode(struct fwnode_handle *fwnode, /* Mask all the interrupts by default */ for (i = 0; i < chip->num_regs; i++) { d->mask_buf[i] = d->mask_buf_def[i]; - if (!chip->mask_base) - continue; - reg = sub_irq_reg(d, d->chip->mask_base, i); - - if (chip->mask_invert) - ret = regmap_irq_update_bits(d, reg, - d->mask_buf[i], ~d->mask_buf[i]); - else if (d->chip->unmask_base) { - unmask_offset = d->chip->unmask_base - - d->chip->mask_base; - ret = regmap_irq_update_bits(d, - reg + unmask_offset, - d->mask_buf[i], - d->mask_buf[i]); - } else - ret = regmap_irq_update_bits(d, reg, - d->mask_buf[i], d->mask_buf[i]); - if (ret != 0) { - dev_err(map->dev, "Failed to set masks in 0x%x: %d\n", - reg, ret); - goto err_alloc; + if (chip->handle_mask_sync) { + ret = chip->handle_mask_sync(i, d->mask_buf_def[i], + d->mask_buf[i], + chip->irq_drv_data); + if (ret) + goto err_mutex; + } + + if (chip->mask_base && !chip->handle_mask_sync) { + reg = d->get_irq_reg(d, chip->mask_base, i); + ret = regmap_update_bits(d->map, reg, + d->mask_buf_def[i], + d->mask_buf[i]); + if (ret) { + dev_err(map->dev, "Failed to set masks in 0x%x: %d\n", + reg, ret); + goto err_mutex; + } + } + + if (chip->unmask_base && !chip->handle_mask_sync) { + reg = d->get_irq_reg(d, chip->unmask_base, i); + ret = regmap_update_bits(d->map, reg, + d->mask_buf_def[i], ~d->mask_buf[i]); + if (ret) { + dev_err(map->dev, "Failed to set masks in 0x%x: %d\n", + reg, ret); + goto err_mutex; + } } if (!chip->init_ack_masked) continue; /* Ack masked but set interrupts */ - reg = sub_irq_reg(d, d->chip->status_base, i); - ret = regmap_read(map, reg, &d->status_buf[i]); - if (ret != 0) { - dev_err(map->dev, "Failed to read IRQ status: %d\n", - ret); - goto err_alloc; + if (d->chip->no_status) { + /* no status register so default to all active */ + d->status_buf[i] = UINT_MAX; + } else { + reg = d->get_irq_reg(d, d->chip->status_base, i); + ret = regmap_read(map, reg, &d->status_buf[i]); + if (ret != 0) { + dev_err(map->dev, "Failed to read IRQ status: %d\n", + ret); + goto err_mutex; + } } if (chip->status_invert) d->status_buf[i] = ~d->status_buf[i]; if (d->status_buf[i] && (chip->ack_base || chip->use_ack)) { - reg = sub_irq_reg(d, d->chip->ack_base, i); + reg = d->get_irq_reg(d, d->chip->ack_base, i); if (chip->ack_invert) ret = regmap_write(map, reg, ~(d->status_buf[i] & d->mask_buf[i])); @@ -817,18 +879,14 @@ int regmap_add_irq_chip_fwnode(struct fwnode_handle *fwnode, d->status_buf[i] & d->mask_buf[i]); if (chip->clear_ack) { if (chip->ack_invert && !ret) - ret = regmap_write(map, reg, - (d->status_buf[i] & - d->mask_buf[i])); + ret = regmap_write(map, reg, UINT_MAX); else if (!ret) - ret = regmap_write(map, reg, - ~(d->status_buf[i] & - d->mask_buf[i])); + ret = regmap_write(map, reg, 0); } if (ret != 0) { dev_err(map->dev, "Failed to ack 0x%x: %d\n", reg, ret); - goto err_alloc; + goto err_mutex; } } } @@ -837,53 +895,37 @@ int regmap_add_irq_chip_fwnode(struct fwnode_handle *fwnode, if (d->wake_buf) { for (i = 0; i < chip->num_regs; i++) { d->wake_buf[i] = d->mask_buf_def[i]; - reg = sub_irq_reg(d, d->chip->wake_base, i); + reg = d->get_irq_reg(d, d->chip->wake_base, i); if (chip->wake_invert) - ret = regmap_irq_update_bits(d, reg, + ret = regmap_update_bits(d->map, reg, d->mask_buf_def[i], 0); else - ret = regmap_irq_update_bits(d, reg, + ret = regmap_update_bits(d->map, reg, d->mask_buf_def[i], d->wake_buf[i]); if (ret != 0) { dev_err(map->dev, "Failed to set masks in 0x%x: %d\n", reg, ret); - goto err_alloc; + goto err_mutex; } } } - if (chip->num_type_reg && !chip->type_in_mask) { - for (i = 0; i < chip->num_type_reg; ++i) { - reg = sub_irq_reg(d, d->chip->type_base, i); - - ret = regmap_read(map, reg, &d->type_buf_def[i]); - - if (d->chip->type_invert) - d->type_buf_def[i] = ~d->type_buf_def[i]; + /* Store current levels */ + if (chip->status_is_level) { + ret = read_irq_data(d); + if (ret < 0) + goto err_mutex; - if (ret) { - dev_err(map->dev, "Failed to get type defaults at 0x%x: %d\n", - reg, ret); - goto err_alloc; - } - } + memcpy(d->prev_status_buf, d->status_buf, + array_size(d->chip->num_regs, sizeof(d->prev_status_buf[0]))); } - if (irq_base) - d->domain = irq_domain_create_legacy(fwnode, chip->num_irqs, - irq_base, 0, - ®map_domain_ops, d); - else - d->domain = irq_domain_create_linear(fwnode, chip->num_irqs, - ®map_domain_ops, d); - if (!d->domain) { - dev_err(map->dev, "Failed to create IRQ domain\n"); - ret = -ENOMEM; - goto err_alloc; - } + ret = regmap_irq_create_domain(fwnode, irq_base, chip, d); + if (ret) + goto err_mutex; ret = request_threaded_irq(irq, NULL, regmap_irq_thread, irq_flags | IRQF_ONESHOT, @@ -900,18 +942,23 @@ int regmap_add_irq_chip_fwnode(struct fwnode_handle *fwnode, err_domain: /* Should really dispose of the domain but... */ +err_mutex: + mutex_destroy(&d->lock); + lockdep_unregister_key(&d->lock_key); err_alloc: kfree(d->type_buf); kfree(d->type_buf_def); kfree(d->wake_buf); kfree(d->mask_buf_def); kfree(d->mask_buf); + kfree(d->main_status_buf); kfree(d->status_buf); + kfree(d->prev_status_buf); kfree(d->status_reg_buf); - if (d->virt_buf) { - for (i = 0; i < chip->num_virt_regs; i++) - kfree(d->virt_buf[i]); - kfree(d->virt_buf); + if (d->config_buf) { + for (i = 0; i < chip->num_config_bases; i++) + kfree(d->config_buf[i]); + kfree(d->config_buf); } kfree(d); return ret; @@ -953,7 +1000,7 @@ EXPORT_SYMBOL_GPL(regmap_add_irq_chip); void regmap_del_irq_chip(int irq, struct regmap_irq_chip_data *d) { unsigned int virq; - int hwirq; + int i, hwirq; if (!d) return; @@ -981,8 +1028,17 @@ void regmap_del_irq_chip(int irq, struct regmap_irq_chip_data *d) kfree(d->wake_buf); kfree(d->mask_buf_def); kfree(d->mask_buf); + kfree(d->main_status_buf); kfree(d->status_reg_buf); kfree(d->status_buf); + kfree(d->prev_status_buf); + if (d->config_buf) { + for (i = 0; i < d->chip->num_config_bases; i++) + kfree(d->config_buf[i]); + kfree(d->config_buf); + } + mutex_destroy(&d->lock); + lockdep_unregister_key(&d->lock_key); kfree(d); } EXPORT_SYMBOL_GPL(regmap_del_irq_chip); @@ -1053,7 +1109,7 @@ int devm_regmap_add_irq_chip_fwnode(struct device *dev, EXPORT_SYMBOL_GPL(devm_regmap_add_irq_chip_fwnode); /** - * devm_regmap_add_irq_chip() - Resource manager regmap_add_irq_chip() + * devm_regmap_add_irq_chip() - Resource managed regmap_add_irq_chip() * * @dev: The device pointer on which irq_chip belongs to. * @map: The regmap for the device. @@ -1082,7 +1138,7 @@ EXPORT_SYMBOL_GPL(devm_regmap_add_irq_chip); /** * devm_regmap_del_irq_chip() - Resource managed regmap_del_irq_chip() * - * @dev: Device for which which resource was allocated. + * @dev: Device for which the resource was allocated. * @irq: Primary IRQ for the device. * @data: ®map_irq_chip_data allocated by regmap_add_irq_chip(). * diff --git a/drivers/base/regmap/regmap-kunit.c b/drivers/base/regmap/regmap-kunit.c new file mode 100644 index 000000000000..f6fc5ed016da --- /dev/null +++ b/drivers/base/regmap/regmap-kunit.c @@ -0,0 +1,2131 @@ +// SPDX-License-Identifier: GPL-2.0 +// +// regmap KUnit tests +// +// Copyright 2023 Arm Ltd + +#include <kunit/device.h> +#include <kunit/resource.h> +#include <kunit/test.h> +#include "internal.h" + +#define BLOCK_TEST_SIZE 12 + +KUNIT_DEFINE_ACTION_WRAPPER(regmap_exit_action, regmap_exit, struct regmap *); + +struct regmap_test_priv { + struct device *dev; +}; + +struct regmap_test_param { + enum regcache_type cache; + enum regmap_endian val_endian; + + unsigned int from_reg; + bool fast_io; +}; + +static void get_changed_bytes(void *orig, void *new, size_t size) +{ + char *o = orig; + char *n = new; + int i; + + get_random_bytes(new, size); + + /* + * This could be nicer and more efficient but we shouldn't + * super care. + */ + for (i = 0; i < size; i++) + while (n[i] == o[i]) + get_random_bytes(&n[i], 1); +} + +static const struct regmap_config test_regmap_config = { + .reg_stride = 1, + .val_bits = sizeof(unsigned int) * 8, +}; + +static const char *regcache_type_name(enum regcache_type type) +{ + switch (type) { + case REGCACHE_NONE: + return "none"; + case REGCACHE_FLAT: + return "flat"; + case REGCACHE_FLAT_S: + return "flat-sparse"; + case REGCACHE_RBTREE: + return "rbtree"; + case REGCACHE_MAPLE: + return "maple"; + default: + return NULL; + } +} + +static const char *regmap_endian_name(enum regmap_endian endian) +{ + switch (endian) { + case REGMAP_ENDIAN_BIG: + return "big"; + case REGMAP_ENDIAN_LITTLE: + return "little"; + case REGMAP_ENDIAN_DEFAULT: + return "default"; + case REGMAP_ENDIAN_NATIVE: + return "native"; + default: + return NULL; + } +} + +static void param_to_desc(const struct regmap_test_param *param, char *desc) +{ + snprintf(desc, KUNIT_PARAM_DESC_SIZE, "%s-%s%s @%#x", + regcache_type_name(param->cache), + regmap_endian_name(param->val_endian), + param->fast_io ? " fast I/O" : "", + param->from_reg); +} + +static const struct regmap_test_param regcache_types_list[] = { + { .cache = REGCACHE_NONE }, + { .cache = REGCACHE_NONE, .fast_io = true }, + { .cache = REGCACHE_FLAT }, + { .cache = REGCACHE_FLAT, .fast_io = true }, + { .cache = REGCACHE_FLAT_S }, + { .cache = REGCACHE_FLAT_S, .fast_io = true }, + { .cache = REGCACHE_RBTREE }, + { .cache = REGCACHE_RBTREE, .fast_io = true }, + { .cache = REGCACHE_MAPLE }, + { .cache = REGCACHE_MAPLE, .fast_io = true }, +}; + +KUNIT_ARRAY_PARAM(regcache_types, regcache_types_list, param_to_desc); + +static const struct regmap_test_param real_cache_types_only_list[] = { + { .cache = REGCACHE_FLAT }, + { .cache = REGCACHE_FLAT, .fast_io = true }, + { .cache = REGCACHE_FLAT_S }, + { .cache = REGCACHE_FLAT_S, .fast_io = true }, + { .cache = REGCACHE_RBTREE }, + { .cache = REGCACHE_RBTREE, .fast_io = true }, + { .cache = REGCACHE_MAPLE }, + { .cache = REGCACHE_MAPLE, .fast_io = true }, +}; + +KUNIT_ARRAY_PARAM(real_cache_types_only, real_cache_types_only_list, param_to_desc); + +static const struct regmap_test_param real_cache_types_list[] = { + { .cache = REGCACHE_FLAT, .from_reg = 0 }, + { .cache = REGCACHE_FLAT, .from_reg = 0, .fast_io = true }, + { .cache = REGCACHE_FLAT, .from_reg = 0x2001 }, + { .cache = REGCACHE_FLAT, .from_reg = 0x2002 }, + { .cache = REGCACHE_FLAT, .from_reg = 0x2003 }, + { .cache = REGCACHE_FLAT, .from_reg = 0x2004 }, + { .cache = REGCACHE_FLAT_S, .from_reg = 0 }, + { .cache = REGCACHE_FLAT_S, .from_reg = 0, .fast_io = true }, + { .cache = REGCACHE_FLAT_S, .from_reg = 0x2001 }, + { .cache = REGCACHE_FLAT_S, .from_reg = 0x2002 }, + { .cache = REGCACHE_FLAT_S, .from_reg = 0x2003 }, + { .cache = REGCACHE_FLAT_S, .from_reg = 0x2004 }, + { .cache = REGCACHE_RBTREE, .from_reg = 0 }, + { .cache = REGCACHE_RBTREE, .from_reg = 0, .fast_io = true }, + { .cache = REGCACHE_RBTREE, .from_reg = 0x2001 }, + { .cache = REGCACHE_RBTREE, .from_reg = 0x2002 }, + { .cache = REGCACHE_RBTREE, .from_reg = 0x2003 }, + { .cache = REGCACHE_RBTREE, .from_reg = 0x2004 }, + { .cache = REGCACHE_MAPLE, .from_reg = 0 }, + { .cache = REGCACHE_MAPLE, .from_reg = 0, .fast_io = true }, + { .cache = REGCACHE_MAPLE, .from_reg = 0x2001 }, + { .cache = REGCACHE_MAPLE, .from_reg = 0x2002 }, + { .cache = REGCACHE_MAPLE, .from_reg = 0x2003 }, + { .cache = REGCACHE_MAPLE, .from_reg = 0x2004 }, +}; + +KUNIT_ARRAY_PARAM(real_cache_types, real_cache_types_list, param_to_desc); + +static const struct regmap_test_param sparse_cache_types_list[] = { + { .cache = REGCACHE_FLAT_S, .from_reg = 0 }, + { .cache = REGCACHE_FLAT_S, .from_reg = 0, .fast_io = true }, + { .cache = REGCACHE_FLAT_S, .from_reg = 0x2001 }, + { .cache = REGCACHE_FLAT_S, .from_reg = 0x2002 }, + { .cache = REGCACHE_FLAT_S, .from_reg = 0x2003 }, + { .cache = REGCACHE_FLAT_S, .from_reg = 0x2004 }, + { .cache = REGCACHE_RBTREE, .from_reg = 0 }, + { .cache = REGCACHE_RBTREE, .from_reg = 0, .fast_io = true }, + { .cache = REGCACHE_RBTREE, .from_reg = 0x2001 }, + { .cache = REGCACHE_RBTREE, .from_reg = 0x2002 }, + { .cache = REGCACHE_RBTREE, .from_reg = 0x2003 }, + { .cache = REGCACHE_RBTREE, .from_reg = 0x2004 }, + { .cache = REGCACHE_MAPLE, .from_reg = 0 }, + { .cache = REGCACHE_MAPLE, .from_reg = 0, .fast_io = true }, + { .cache = REGCACHE_MAPLE, .from_reg = 0x2001 }, + { .cache = REGCACHE_MAPLE, .from_reg = 0x2002 }, + { .cache = REGCACHE_MAPLE, .from_reg = 0x2003 }, + { .cache = REGCACHE_MAPLE, .from_reg = 0x2004 }, +}; + +KUNIT_ARRAY_PARAM(sparse_cache_types, sparse_cache_types_list, param_to_desc); + +static struct regmap *gen_regmap(struct kunit *test, + struct regmap_config *config, + struct regmap_ram_data **data) +{ + const struct regmap_test_param *param = test->param_value; + struct regmap_test_priv *priv = test->priv; + unsigned int *buf; + struct regmap *ret = ERR_PTR(-ENOMEM); + size_t size; + int i, error; + struct reg_default *defaults; + + config->cache_type = param->cache; + config->fast_io = param->fast_io; + + if (config->max_register == 0) { + config->max_register = param->from_reg; + if (config->num_reg_defaults) + config->max_register += (config->num_reg_defaults - 1) * + config->reg_stride; + else + config->max_register += (BLOCK_TEST_SIZE * config->reg_stride); + } + + size = array_size(config->max_register + 1, sizeof(*buf)); + buf = kmalloc(size, GFP_KERNEL); + if (!buf) + return ERR_PTR(-ENOMEM); + + get_random_bytes(buf, size); + + *data = kzalloc(sizeof(**data), GFP_KERNEL); + if (!(*data)) + goto out_free; + (*data)->vals = buf; + + if (config->num_reg_defaults) { + defaults = kunit_kcalloc(test, + config->num_reg_defaults, + sizeof(struct reg_default), + GFP_KERNEL); + if (!defaults) + goto out_free; + + config->reg_defaults = defaults; + + for (i = 0; i < config->num_reg_defaults; i++) { + defaults[i].reg = param->from_reg + (i * config->reg_stride); + defaults[i].def = buf[param->from_reg + (i * config->reg_stride)]; + } + } + + ret = regmap_init_ram(priv->dev, config, *data); + if (IS_ERR(ret)) + goto out_free; + + /* This calls regmap_exit() on failure, which frees buf and *data */ + error = kunit_add_action_or_reset(test, regmap_exit_action, ret); + if (error) + ret = ERR_PTR(error); + + return ret; + +out_free: + kfree(buf); + kfree(*data); + + return ret; +} + +static bool reg_5_false(struct device *dev, unsigned int reg) +{ + struct kunit *test = dev_get_drvdata(dev); + const struct regmap_test_param *param = test->param_value; + + return reg != (param->from_reg + 5); +} + +static void basic_read_write(struct kunit *test) +{ + struct regmap *map; + struct regmap_config config; + struct regmap_ram_data *data; + unsigned int val, rval; + + config = test_regmap_config; + + map = gen_regmap(test, &config, &data); + KUNIT_ASSERT_FALSE(test, IS_ERR(map)); + if (IS_ERR(map)) + return; + + get_random_bytes(&val, sizeof(val)); + + /* If we write a value to a register we can read it back */ + KUNIT_EXPECT_EQ(test, 0, regmap_write(map, 0, val)); + KUNIT_EXPECT_EQ(test, 0, regmap_read(map, 0, &rval)); + KUNIT_EXPECT_EQ(test, val, rval); + + /* If using a cache the cache satisfied the read */ + KUNIT_EXPECT_EQ(test, config.cache_type == REGCACHE_NONE, data->read[0]); +} + +static void bulk_write(struct kunit *test) +{ + struct regmap *map; + struct regmap_config config; + struct regmap_ram_data *data; + unsigned int val[BLOCK_TEST_SIZE], rval[BLOCK_TEST_SIZE]; + int i; + + config = test_regmap_config; + + map = gen_regmap(test, &config, &data); + KUNIT_ASSERT_FALSE(test, IS_ERR(map)); + if (IS_ERR(map)) + return; + + get_random_bytes(&val, sizeof(val)); + + /* + * Data written via the bulk API can be read back with single + * reads. + */ + KUNIT_EXPECT_EQ(test, 0, regmap_bulk_write(map, 0, val, + BLOCK_TEST_SIZE)); + for (i = 0; i < BLOCK_TEST_SIZE; i++) + KUNIT_EXPECT_EQ(test, 0, regmap_read(map, i, &rval[i])); + + KUNIT_EXPECT_MEMEQ(test, val, rval, sizeof(val)); + + /* If using a cache the cache satisfied the read */ + for (i = 0; i < BLOCK_TEST_SIZE; i++) + KUNIT_EXPECT_EQ(test, config.cache_type == REGCACHE_NONE, data->read[i]); +} + +static void bulk_read(struct kunit *test) +{ + struct regmap *map; + struct regmap_config config; + struct regmap_ram_data *data; + unsigned int val[BLOCK_TEST_SIZE], rval[BLOCK_TEST_SIZE]; + int i; + + config = test_regmap_config; + + map = gen_regmap(test, &config, &data); + KUNIT_ASSERT_FALSE(test, IS_ERR(map)); + if (IS_ERR(map)) + return; + + get_random_bytes(&val, sizeof(val)); + + /* Data written as single writes can be read via the bulk API */ + for (i = 0; i < BLOCK_TEST_SIZE; i++) + KUNIT_EXPECT_EQ(test, 0, regmap_write(map, i, val[i])); + KUNIT_EXPECT_EQ(test, 0, regmap_bulk_read(map, 0, rval, + BLOCK_TEST_SIZE)); + KUNIT_EXPECT_MEMEQ(test, val, rval, sizeof(val)); + + /* If using a cache the cache satisfied the read */ + for (i = 0; i < BLOCK_TEST_SIZE; i++) + KUNIT_EXPECT_EQ(test, config.cache_type == REGCACHE_NONE, data->read[i]); +} + +static void multi_write(struct kunit *test) +{ + struct regmap *map; + struct regmap_config config; + struct regmap_ram_data *data; + struct reg_sequence sequence[BLOCK_TEST_SIZE]; + unsigned int val[BLOCK_TEST_SIZE], rval[BLOCK_TEST_SIZE]; + int i; + + config = test_regmap_config; + + map = gen_regmap(test, &config, &data); + KUNIT_ASSERT_FALSE(test, IS_ERR(map)); + if (IS_ERR(map)) + return; + + get_random_bytes(&val, sizeof(val)); + + /* + * Data written via the multi API can be read back with single + * reads. + */ + for (i = 0; i < BLOCK_TEST_SIZE; i++) { + sequence[i].reg = i; + sequence[i].def = val[i]; + sequence[i].delay_us = 0; + } + KUNIT_EXPECT_EQ(test, 0, + regmap_multi_reg_write(map, sequence, BLOCK_TEST_SIZE)); + for (i = 0; i < BLOCK_TEST_SIZE; i++) + KUNIT_EXPECT_EQ(test, 0, regmap_read(map, i, &rval[i])); + + KUNIT_EXPECT_MEMEQ(test, val, rval, sizeof(val)); + + /* If using a cache the cache satisfied the read */ + for (i = 0; i < BLOCK_TEST_SIZE; i++) + KUNIT_EXPECT_EQ(test, config.cache_type == REGCACHE_NONE, data->read[i]); +} + +static void multi_read(struct kunit *test) +{ + struct regmap *map; + struct regmap_config config; + struct regmap_ram_data *data; + unsigned int regs[BLOCK_TEST_SIZE]; + unsigned int val[BLOCK_TEST_SIZE], rval[BLOCK_TEST_SIZE]; + int i; + + config = test_regmap_config; + + map = gen_regmap(test, &config, &data); + KUNIT_ASSERT_FALSE(test, IS_ERR(map)); + if (IS_ERR(map)) + return; + + get_random_bytes(&val, sizeof(val)); + + /* Data written as single writes can be read via the multi API */ + for (i = 0; i < BLOCK_TEST_SIZE; i++) { + regs[i] = i; + KUNIT_EXPECT_EQ(test, 0, regmap_write(map, i, val[i])); + } + KUNIT_EXPECT_EQ(test, 0, + regmap_multi_reg_read(map, regs, rval, BLOCK_TEST_SIZE)); + KUNIT_EXPECT_MEMEQ(test, val, rval, sizeof(val)); + + /* If using a cache the cache satisfied the read */ + for (i = 0; i < BLOCK_TEST_SIZE; i++) + KUNIT_EXPECT_EQ(test, config.cache_type == REGCACHE_NONE, data->read[i]); +} + +static void read_bypassed(struct kunit *test) +{ + const struct regmap_test_param *param = test->param_value; + struct regmap *map; + struct regmap_config config; + struct regmap_ram_data *data; + unsigned int val[BLOCK_TEST_SIZE], rval; + int i; + + config = test_regmap_config; + + map = gen_regmap(test, &config, &data); + KUNIT_ASSERT_FALSE(test, IS_ERR(map)); + if (IS_ERR(map)) + return; + + KUNIT_EXPECT_FALSE(test, map->cache_bypass); + + get_random_bytes(&val, sizeof(val)); + + /* Write some test values */ + KUNIT_EXPECT_EQ(test, 0, regmap_bulk_write(map, param->from_reg, val, ARRAY_SIZE(val))); + + regcache_cache_only(map, true); + + /* + * While in cache-only regmap_read_bypassed() should return the register + * value and leave the map in cache-only. + */ + for (i = 0; i < ARRAY_SIZE(val); i++) { + /* Put inverted bits in rval to prove we really read the value */ + rval = ~val[i]; + KUNIT_EXPECT_EQ(test, 0, regmap_read(map, param->from_reg + i, &rval)); + KUNIT_EXPECT_EQ(test, val[i], rval); + + rval = ~val[i]; + KUNIT_EXPECT_EQ(test, 0, regmap_read_bypassed(map, param->from_reg + i, &rval)); + KUNIT_EXPECT_EQ(test, val[i], rval); + KUNIT_EXPECT_TRUE(test, map->cache_only); + KUNIT_EXPECT_FALSE(test, map->cache_bypass); + } + + /* + * Change the underlying register values to prove it is returning + * real values not cached values. + */ + for (i = 0; i < ARRAY_SIZE(val); i++) { + val[i] = ~val[i]; + data->vals[param->from_reg + i] = val[i]; + } + + for (i = 0; i < ARRAY_SIZE(val); i++) { + rval = ~val[i]; + KUNIT_EXPECT_EQ(test, 0, regmap_read(map, param->from_reg + i, &rval)); + KUNIT_EXPECT_NE(test, val[i], rval); + + rval = ~val[i]; + KUNIT_EXPECT_EQ(test, 0, regmap_read_bypassed(map, param->from_reg + i, &rval)); + KUNIT_EXPECT_EQ(test, val[i], rval); + KUNIT_EXPECT_TRUE(test, map->cache_only); + KUNIT_EXPECT_FALSE(test, map->cache_bypass); + } +} + +static void read_bypassed_volatile(struct kunit *test) +{ + const struct regmap_test_param *param = test->param_value; + struct regmap *map; + struct regmap_config config; + struct regmap_ram_data *data; + unsigned int val[BLOCK_TEST_SIZE], rval; + int i; + + config = test_regmap_config; + /* All registers except #5 volatile */ + config.volatile_reg = reg_5_false; + + map = gen_regmap(test, &config, &data); + KUNIT_ASSERT_FALSE(test, IS_ERR(map)); + if (IS_ERR(map)) + return; + + KUNIT_EXPECT_FALSE(test, map->cache_bypass); + + get_random_bytes(&val, sizeof(val)); + + /* Write some test values */ + KUNIT_EXPECT_EQ(test, 0, regmap_bulk_write(map, param->from_reg, val, ARRAY_SIZE(val))); + + regcache_cache_only(map, true); + + /* + * While in cache-only regmap_read_bypassed() should return the register + * value and leave the map in cache-only. + */ + for (i = 0; i < ARRAY_SIZE(val); i++) { + /* Register #5 is non-volatile so should read from cache */ + KUNIT_EXPECT_EQ(test, (i == 5) ? 0 : -EBUSY, + regmap_read(map, param->from_reg + i, &rval)); + + /* Put inverted bits in rval to prove we really read the value */ + rval = ~val[i]; + KUNIT_EXPECT_EQ(test, 0, regmap_read_bypassed(map, param->from_reg + i, &rval)); + KUNIT_EXPECT_EQ(test, val[i], rval); + KUNIT_EXPECT_TRUE(test, map->cache_only); + KUNIT_EXPECT_FALSE(test, map->cache_bypass); + } + + /* + * Change the underlying register values to prove it is returning + * real values not cached values. + */ + for (i = 0; i < ARRAY_SIZE(val); i++) { + val[i] = ~val[i]; + data->vals[param->from_reg + i] = val[i]; + } + + for (i = 0; i < ARRAY_SIZE(val); i++) { + if (i == 5) + continue; + + rval = ~val[i]; + KUNIT_EXPECT_EQ(test, 0, regmap_read_bypassed(map, param->from_reg + i, &rval)); + KUNIT_EXPECT_EQ(test, val[i], rval); + KUNIT_EXPECT_TRUE(test, map->cache_only); + KUNIT_EXPECT_FALSE(test, map->cache_bypass); + } +} + +static void write_readonly(struct kunit *test) +{ + struct regmap *map; + struct regmap_config config; + struct regmap_ram_data *data; + unsigned int val; + int i; + + config = test_regmap_config; + config.num_reg_defaults = BLOCK_TEST_SIZE; + config.writeable_reg = reg_5_false; + + map = gen_regmap(test, &config, &data); + KUNIT_ASSERT_FALSE(test, IS_ERR(map)); + if (IS_ERR(map)) + return; + + get_random_bytes(&val, sizeof(val)); + + for (i = 0; i < BLOCK_TEST_SIZE; i++) + data->written[i] = false; + + /* Change the value of all registers, readonly should fail */ + for (i = 0; i < BLOCK_TEST_SIZE; i++) + KUNIT_EXPECT_EQ(test, i != 5, regmap_write(map, i, val) == 0); + + /* Did that match what we see on the device? */ + for (i = 0; i < BLOCK_TEST_SIZE; i++) + KUNIT_EXPECT_EQ(test, i != 5, data->written[i]); +} + +static void read_writeonly(struct kunit *test) +{ + struct regmap *map; + struct regmap_config config; + struct regmap_ram_data *data; + unsigned int val; + int i; + + config = test_regmap_config; + config.readable_reg = reg_5_false; + + map = gen_regmap(test, &config, &data); + KUNIT_ASSERT_FALSE(test, IS_ERR(map)); + if (IS_ERR(map)) + return; + + for (i = 0; i < BLOCK_TEST_SIZE; i++) + data->read[i] = false; + + /* + * Try to read all the registers, the writeonly one should + * fail if we aren't using the flat cache. + */ + for (i = 0; i < BLOCK_TEST_SIZE; i++) { + if (config.cache_type != REGCACHE_FLAT) { + KUNIT_EXPECT_EQ(test, i != 5, + regmap_read(map, i, &val) == 0); + } else { + KUNIT_EXPECT_EQ(test, 0, regmap_read(map, i, &val)); + } + } + + /* Did we trigger a hardware access? */ + KUNIT_EXPECT_FALSE(test, data->read[5]); +} + +static void reg_defaults(struct kunit *test) +{ + struct regmap *map; + struct regmap_config config; + struct regmap_ram_data *data; + unsigned int rval[BLOCK_TEST_SIZE]; + int i; + + config = test_regmap_config; + config.num_reg_defaults = BLOCK_TEST_SIZE; + + map = gen_regmap(test, &config, &data); + KUNIT_ASSERT_FALSE(test, IS_ERR(map)); + if (IS_ERR(map)) + return; + + /* Read back the expected default data */ + KUNIT_EXPECT_EQ(test, 0, regmap_bulk_read(map, 0, rval, + BLOCK_TEST_SIZE)); + KUNIT_EXPECT_MEMEQ(test, data->vals, rval, sizeof(rval)); + + /* The data should have been read from cache if there was one */ + for (i = 0; i < BLOCK_TEST_SIZE; i++) + KUNIT_EXPECT_EQ(test, config.cache_type == REGCACHE_NONE, data->read[i]); +} + +static void reg_defaults_read_dev(struct kunit *test) +{ + struct regmap *map; + struct regmap_config config; + struct regmap_ram_data *data; + unsigned int rval[BLOCK_TEST_SIZE]; + int i; + + config = test_regmap_config; + config.num_reg_defaults_raw = BLOCK_TEST_SIZE; + + map = gen_regmap(test, &config, &data); + KUNIT_ASSERT_FALSE(test, IS_ERR(map)); + if (IS_ERR(map)) + return; + + /* We should have read the cache defaults back from the map */ + for (i = 0; i < BLOCK_TEST_SIZE; i++) { + KUNIT_EXPECT_EQ(test, config.cache_type != REGCACHE_NONE, data->read[i]); + data->read[i] = false; + } + + /* Read back the expected default data */ + KUNIT_EXPECT_EQ(test, 0, regmap_bulk_read(map, 0, rval, + BLOCK_TEST_SIZE)); + KUNIT_EXPECT_MEMEQ(test, data->vals, rval, sizeof(rval)); + + /* The data should have been read from cache if there was one */ + for (i = 0; i < BLOCK_TEST_SIZE; i++) + KUNIT_EXPECT_EQ(test, config.cache_type == REGCACHE_NONE, data->read[i]); +} + +static void register_patch(struct kunit *test) +{ + struct regmap *map; + struct regmap_config config; + struct regmap_ram_data *data; + struct reg_sequence patch[2]; + unsigned int rval[BLOCK_TEST_SIZE]; + int i; + + /* We need defaults so readback works */ + config = test_regmap_config; + config.num_reg_defaults = BLOCK_TEST_SIZE; + + map = gen_regmap(test, &config, &data); + KUNIT_ASSERT_FALSE(test, IS_ERR(map)); + if (IS_ERR(map)) + return; + + /* Stash the original values */ + KUNIT_EXPECT_EQ(test, 0, regmap_bulk_read(map, 0, rval, + BLOCK_TEST_SIZE)); + + /* Patch a couple of values */ + patch[0].reg = 2; + patch[0].def = rval[2] + 1; + patch[0].delay_us = 0; + patch[1].reg = 5; + patch[1].def = rval[5] + 1; + patch[1].delay_us = 0; + KUNIT_EXPECT_EQ(test, 0, regmap_register_patch(map, patch, + ARRAY_SIZE(patch))); + + /* Only the patched registers are written */ + for (i = 0; i < BLOCK_TEST_SIZE; i++) { + switch (i) { + case 2: + case 5: + KUNIT_EXPECT_TRUE(test, data->written[i]); + KUNIT_EXPECT_EQ(test, data->vals[i], rval[i] + 1); + break; + default: + KUNIT_EXPECT_FALSE(test, data->written[i]); + KUNIT_EXPECT_EQ(test, data->vals[i], rval[i]); + break; + } + } +} + +static void stride(struct kunit *test) +{ + struct regmap *map; + struct regmap_config config; + struct regmap_ram_data *data; + unsigned int rval; + int i; + + config = test_regmap_config; + config.reg_stride = 2; + config.num_reg_defaults = BLOCK_TEST_SIZE / 2; + + /* + * Allow one extra register so that the read/written arrays + * are sized big enough to include an entry for the odd + * address past the final reg_default register. + */ + config.max_register = BLOCK_TEST_SIZE; + + map = gen_regmap(test, &config, &data); + KUNIT_ASSERT_FALSE(test, IS_ERR(map)); + if (IS_ERR(map)) + return; + + /* Only even addresses can be accessed, try both read and write */ + for (i = 0; i < BLOCK_TEST_SIZE; i++) { + data->read[i] = false; + data->written[i] = false; + + if (i % 2) { + KUNIT_EXPECT_NE(test, 0, regmap_read(map, i, &rval)); + KUNIT_EXPECT_NE(test, 0, regmap_write(map, i, rval)); + KUNIT_EXPECT_FALSE(test, data->read[i]); + KUNIT_EXPECT_FALSE(test, data->written[i]); + } else { + KUNIT_EXPECT_EQ(test, 0, regmap_read(map, i, &rval)); + KUNIT_EXPECT_EQ(test, data->vals[i], rval); + KUNIT_EXPECT_EQ(test, config.cache_type == REGCACHE_NONE, + data->read[i]); + + KUNIT_EXPECT_EQ(test, 0, regmap_write(map, i, rval)); + KUNIT_EXPECT_TRUE(test, data->written[i]); + } + } +} + +static const struct regmap_range_cfg test_range = { + .selector_reg = 1, + .selector_mask = 0xff, + + .window_start = 4, + .window_len = 10, + + .range_min = 20, + .range_max = 40, +}; + +static bool test_range_window_volatile(struct device *dev, unsigned int reg) +{ + if (reg >= test_range.window_start && + reg <= test_range.window_start + test_range.window_len) + return true; + + return false; +} + +static bool test_range_all_volatile(struct device *dev, unsigned int reg) +{ + if (test_range_window_volatile(dev, reg)) + return true; + + if (reg >= test_range.range_min && reg <= test_range.range_max) + return true; + + return false; +} + +static void basic_ranges(struct kunit *test) +{ + struct regmap *map; + struct regmap_config config; + struct regmap_ram_data *data; + unsigned int val; + int i; + + config = test_regmap_config; + config.volatile_reg = test_range_all_volatile; + config.ranges = &test_range; + config.num_ranges = 1; + config.max_register = test_range.range_max; + + map = gen_regmap(test, &config, &data); + KUNIT_ASSERT_FALSE(test, IS_ERR(map)); + if (IS_ERR(map)) + return; + + for (i = test_range.range_min; i < test_range.range_max; i++) { + data->read[i] = false; + data->written[i] = false; + } + + /* Reset the page to a non-zero value to trigger a change */ + KUNIT_EXPECT_EQ(test, 0, regmap_write(map, test_range.selector_reg, + test_range.range_max)); + + /* Check we set the page and use the window for writes */ + data->written[test_range.selector_reg] = false; + data->written[test_range.window_start] = false; + KUNIT_EXPECT_EQ(test, 0, regmap_write(map, test_range.range_min, 0)); + KUNIT_EXPECT_TRUE(test, data->written[test_range.selector_reg]); + KUNIT_EXPECT_TRUE(test, data->written[test_range.window_start]); + + data->written[test_range.selector_reg] = false; + data->written[test_range.window_start] = false; + KUNIT_EXPECT_EQ(test, 0, regmap_write(map, + test_range.range_min + + test_range.window_len, + 0)); + KUNIT_EXPECT_TRUE(test, data->written[test_range.selector_reg]); + KUNIT_EXPECT_TRUE(test, data->written[test_range.window_start]); + + /* Same for reads */ + data->written[test_range.selector_reg] = false; + data->read[test_range.window_start] = false; + KUNIT_EXPECT_EQ(test, 0, regmap_read(map, test_range.range_min, &val)); + KUNIT_EXPECT_TRUE(test, data->written[test_range.selector_reg]); + KUNIT_EXPECT_TRUE(test, data->read[test_range.window_start]); + + data->written[test_range.selector_reg] = false; + data->read[test_range.window_start] = false; + KUNIT_EXPECT_EQ(test, 0, regmap_read(map, + test_range.range_min + + test_range.window_len, + &val)); + KUNIT_EXPECT_TRUE(test, data->written[test_range.selector_reg]); + KUNIT_EXPECT_TRUE(test, data->read[test_range.window_start]); + + /* No physical access triggered in the virtual range */ + for (i = test_range.range_min; i < test_range.range_max; i++) { + KUNIT_EXPECT_FALSE(test, data->read[i]); + KUNIT_EXPECT_FALSE(test, data->written[i]); + } +} + +/* Try to stress dynamic creation of cache data structures */ +static void stress_insert(struct kunit *test) +{ + struct regmap *map; + struct regmap_config config; + struct regmap_ram_data *data; + unsigned int rval, *vals; + size_t buf_sz; + int i; + + config = test_regmap_config; + config.max_register = 300; + + map = gen_regmap(test, &config, &data); + KUNIT_ASSERT_FALSE(test, IS_ERR(map)); + if (IS_ERR(map)) + return; + + buf_sz = array_size(sizeof(*vals), config.max_register); + vals = kunit_kmalloc(test, buf_sz, GFP_KERNEL); + KUNIT_ASSERT_FALSE(test, vals == NULL); + + get_random_bytes(vals, buf_sz); + + /* Write data into the map/cache in ever decreasing strides */ + for (i = 0; i < config.max_register; i += 100) + KUNIT_EXPECT_EQ(test, 0, regmap_write(map, i, vals[i])); + for (i = 0; i < config.max_register; i += 50) + KUNIT_EXPECT_EQ(test, 0, regmap_write(map, i, vals[i])); + for (i = 0; i < config.max_register; i += 25) + KUNIT_EXPECT_EQ(test, 0, regmap_write(map, i, vals[i])); + for (i = 0; i < config.max_register; i += 10) + KUNIT_EXPECT_EQ(test, 0, regmap_write(map, i, vals[i])); + for (i = 0; i < config.max_register; i += 5) + KUNIT_EXPECT_EQ(test, 0, regmap_write(map, i, vals[i])); + for (i = 0; i < config.max_register; i += 3) + KUNIT_EXPECT_EQ(test, 0, regmap_write(map, i, vals[i])); + for (i = 0; i < config.max_register; i += 2) + KUNIT_EXPECT_EQ(test, 0, regmap_write(map, i, vals[i])); + for (i = 0; i < config.max_register; i++) + KUNIT_EXPECT_EQ(test, 0, regmap_write(map, i, vals[i])); + + /* Do reads from the cache (if there is one) match? */ + for (i = 0; i < config.max_register; i ++) { + KUNIT_EXPECT_EQ(test, 0, regmap_read(map, i, &rval)); + KUNIT_EXPECT_EQ(test, rval, vals[i]); + KUNIT_EXPECT_EQ(test, config.cache_type == REGCACHE_NONE, data->read[i]); + } +} + +static void cache_bypass(struct kunit *test) +{ + const struct regmap_test_param *param = test->param_value; + struct regmap *map; + struct regmap_config config; + struct regmap_ram_data *data; + unsigned int val, rval; + + config = test_regmap_config; + + map = gen_regmap(test, &config, &data); + KUNIT_ASSERT_FALSE(test, IS_ERR(map)); + if (IS_ERR(map)) + return; + + get_random_bytes(&val, sizeof(val)); + + /* Ensure the cache has a value in it */ + KUNIT_EXPECT_EQ(test, 0, regmap_write(map, param->from_reg, val)); + + /* Bypass then write a different value */ + regcache_cache_bypass(map, true); + KUNIT_EXPECT_EQ(test, 0, regmap_write(map, param->from_reg, val + 1)); + + /* Read the bypassed value */ + KUNIT_EXPECT_EQ(test, 0, regmap_read(map, param->from_reg, &rval)); + KUNIT_EXPECT_EQ(test, val + 1, rval); + KUNIT_EXPECT_EQ(test, data->vals[param->from_reg], rval); + + /* Disable bypass, the cache should still return the original value */ + regcache_cache_bypass(map, false); + KUNIT_EXPECT_EQ(test, 0, regmap_read(map, param->from_reg, &rval)); + KUNIT_EXPECT_EQ(test, val, rval); +} + +static void cache_sync_marked_dirty(struct kunit *test) +{ + const struct regmap_test_param *param = test->param_value; + struct regmap *map; + struct regmap_config config; + struct regmap_ram_data *data; + unsigned int val[BLOCK_TEST_SIZE]; + int i; + + config = test_regmap_config; + + map = gen_regmap(test, &config, &data); + KUNIT_ASSERT_FALSE(test, IS_ERR(map)); + if (IS_ERR(map)) + return; + + get_random_bytes(&val, sizeof(val)); + + /* Put some data into the cache */ + KUNIT_EXPECT_EQ(test, 0, regmap_bulk_write(map, param->from_reg, val, + BLOCK_TEST_SIZE)); + for (i = 0; i < BLOCK_TEST_SIZE; i++) + data->written[param->from_reg + i] = false; + + /* Trash the data on the device itself then resync */ + regcache_mark_dirty(map); + memset(data->vals, 0, sizeof(val)); + KUNIT_EXPECT_EQ(test, 0, regcache_sync(map)); + + /* Did we just write the correct data out? */ + KUNIT_EXPECT_MEMEQ(test, &data->vals[param->from_reg], val, sizeof(val)); + for (i = 0; i < BLOCK_TEST_SIZE; i++) + KUNIT_EXPECT_EQ(test, true, data->written[param->from_reg + i]); +} + +static void cache_sync_after_cache_only(struct kunit *test) +{ + const struct regmap_test_param *param = test->param_value; + struct regmap *map; + struct regmap_config config; + struct regmap_ram_data *data; + unsigned int val[BLOCK_TEST_SIZE]; + unsigned int val_mask; + int i; + + config = test_regmap_config; + + map = gen_regmap(test, &config, &data); + KUNIT_ASSERT_FALSE(test, IS_ERR(map)); + if (IS_ERR(map)) + return; + + val_mask = GENMASK(config.val_bits - 1, 0); + get_random_bytes(&val, sizeof(val)); + + /* Put some data into the cache */ + KUNIT_EXPECT_EQ(test, 0, regmap_bulk_write(map, param->from_reg, val, + BLOCK_TEST_SIZE)); + for (i = 0; i < BLOCK_TEST_SIZE; i++) + data->written[param->from_reg + i] = false; + + /* Set cache-only and change the values */ + regcache_cache_only(map, true); + for (i = 0; i < ARRAY_SIZE(val); ++i) + val[i] = ~val[i] & val_mask; + + KUNIT_EXPECT_EQ(test, 0, regmap_bulk_write(map, param->from_reg, val, + BLOCK_TEST_SIZE)); + for (i = 0; i < BLOCK_TEST_SIZE; i++) + KUNIT_EXPECT_FALSE(test, data->written[param->from_reg + i]); + + KUNIT_EXPECT_MEMNEQ(test, &data->vals[param->from_reg], val, sizeof(val)); + + /* Exit cache-only and sync the cache without marking hardware registers dirty */ + regcache_cache_only(map, false); + + KUNIT_EXPECT_EQ(test, 0, regcache_sync(map)); + + /* Did we just write the correct data out? */ + KUNIT_EXPECT_MEMEQ(test, &data->vals[param->from_reg], val, sizeof(val)); + for (i = 0; i < BLOCK_TEST_SIZE; i++) + KUNIT_EXPECT_TRUE(test, data->written[param->from_reg + i]); +} + +static void cache_sync_defaults_marked_dirty(struct kunit *test) +{ + const struct regmap_test_param *param = test->param_value; + struct regmap *map; + struct regmap_config config; + struct regmap_ram_data *data; + unsigned int val; + int i; + + config = test_regmap_config; + config.num_reg_defaults = BLOCK_TEST_SIZE; + + map = gen_regmap(test, &config, &data); + KUNIT_ASSERT_FALSE(test, IS_ERR(map)); + if (IS_ERR(map)) + return; + + get_random_bytes(&val, sizeof(val)); + + /* Change the value of one register */ + KUNIT_EXPECT_EQ(test, 0, regmap_write(map, param->from_reg + 2, val)); + + /* Resync */ + regcache_mark_dirty(map); + for (i = 0; i < BLOCK_TEST_SIZE; i++) + data->written[param->from_reg + i] = false; + KUNIT_EXPECT_EQ(test, 0, regcache_sync(map)); + + /* Did we just sync the one register we touched? */ + for (i = 0; i < BLOCK_TEST_SIZE; i++) + KUNIT_EXPECT_EQ(test, i == 2, data->written[param->from_reg + i]); + + /* Rewrite registers back to their defaults */ + for (i = 0; i < config.num_reg_defaults; ++i) + KUNIT_EXPECT_EQ(test, 0, regmap_write(map, config.reg_defaults[i].reg, + config.reg_defaults[i].def)); + + /* + * Resync after regcache_mark_dirty() should not write out registers + * that are at default value + */ + for (i = 0; i < BLOCK_TEST_SIZE; i++) + data->written[param->from_reg + i] = false; + regcache_mark_dirty(map); + KUNIT_EXPECT_EQ(test, 0, regcache_sync(map)); + for (i = 0; i < BLOCK_TEST_SIZE; i++) + KUNIT_EXPECT_FALSE(test, data->written[param->from_reg + i]); +} + +static void cache_sync_default_after_cache_only(struct kunit *test) +{ + const struct regmap_test_param *param = test->param_value; + struct regmap *map; + struct regmap_config config; + struct regmap_ram_data *data; + unsigned int orig_val; + int i; + + config = test_regmap_config; + config.num_reg_defaults = BLOCK_TEST_SIZE; + + map = gen_regmap(test, &config, &data); + KUNIT_ASSERT_FALSE(test, IS_ERR(map)); + if (IS_ERR(map)) + return; + + KUNIT_EXPECT_EQ(test, 0, regmap_read(map, param->from_reg + 2, &orig_val)); + + /* Enter cache-only and change the value of one register */ + regcache_cache_only(map, true); + KUNIT_EXPECT_EQ(test, 0, regmap_write(map, param->from_reg + 2, orig_val + 1)); + + /* Exit cache-only and resync, should write out the changed register */ + regcache_cache_only(map, false); + for (i = 0; i < BLOCK_TEST_SIZE; i++) + data->written[param->from_reg + i] = false; + KUNIT_EXPECT_EQ(test, 0, regcache_sync(map)); + + /* Was the register written out? */ + KUNIT_EXPECT_TRUE(test, data->written[param->from_reg + 2]); + KUNIT_EXPECT_EQ(test, data->vals[param->from_reg + 2], orig_val + 1); + + /* Enter cache-only and write register back to its default value */ + regcache_cache_only(map, true); + KUNIT_EXPECT_EQ(test, 0, regmap_write(map, param->from_reg + 2, orig_val)); + + /* Resync should write out the new value */ + regcache_cache_only(map, false); + for (i = 0; i < BLOCK_TEST_SIZE; i++) + data->written[param->from_reg + i] = false; + + KUNIT_EXPECT_EQ(test, 0, regcache_sync(map)); + KUNIT_EXPECT_TRUE(test, data->written[param->from_reg + 2]); + KUNIT_EXPECT_EQ(test, data->vals[param->from_reg + 2], orig_val); +} + +static void cache_sync_readonly(struct kunit *test) +{ + const struct regmap_test_param *param = test->param_value; + struct regmap *map; + struct regmap_config config; + struct regmap_ram_data *data; + unsigned int val; + int i; + + config = test_regmap_config; + config.writeable_reg = reg_5_false; + + map = gen_regmap(test, &config, &data); + KUNIT_ASSERT_FALSE(test, IS_ERR(map)); + if (IS_ERR(map)) + return; + + /* Read all registers to fill the cache */ + for (i = 0; i < BLOCK_TEST_SIZE; i++) + KUNIT_EXPECT_EQ(test, 0, regmap_read(map, param->from_reg + i, &val)); + + /* Change the value of all registers, readonly should fail */ + get_random_bytes(&val, sizeof(val)); + regcache_cache_only(map, true); + for (i = 0; i < BLOCK_TEST_SIZE; i++) + KUNIT_EXPECT_EQ(test, i != 5, regmap_write(map, param->from_reg + i, val) == 0); + regcache_cache_only(map, false); + + /* Resync */ + for (i = 0; i < BLOCK_TEST_SIZE; i++) + data->written[param->from_reg + i] = false; + KUNIT_EXPECT_EQ(test, 0, regcache_sync(map)); + + /* Did that match what we see on the device? */ + for (i = 0; i < BLOCK_TEST_SIZE; i++) + KUNIT_EXPECT_EQ(test, i != 5, data->written[param->from_reg + i]); +} + +static void cache_sync_patch(struct kunit *test) +{ + const struct regmap_test_param *param = test->param_value; + struct regmap *map; + struct regmap_config config; + struct regmap_ram_data *data; + struct reg_sequence patch[2]; + unsigned int rval[BLOCK_TEST_SIZE], val; + int i; + + /* We need defaults so readback works */ + config = test_regmap_config; + config.num_reg_defaults = BLOCK_TEST_SIZE; + + map = gen_regmap(test, &config, &data); + KUNIT_ASSERT_FALSE(test, IS_ERR(map)); + if (IS_ERR(map)) + return; + + /* Stash the original values */ + KUNIT_EXPECT_EQ(test, 0, regmap_bulk_read(map, param->from_reg, rval, + BLOCK_TEST_SIZE)); + + /* Patch a couple of values */ + patch[0].reg = param->from_reg + 2; + patch[0].def = rval[2] + 1; + patch[0].delay_us = 0; + patch[1].reg = param->from_reg + 5; + patch[1].def = rval[5] + 1; + patch[1].delay_us = 0; + KUNIT_EXPECT_EQ(test, 0, regmap_register_patch(map, patch, + ARRAY_SIZE(patch))); + + /* Sync the cache */ + regcache_mark_dirty(map); + for (i = 0; i < BLOCK_TEST_SIZE; i++) + data->written[param->from_reg + i] = false; + KUNIT_EXPECT_EQ(test, 0, regcache_sync(map)); + + /* The patch should be on the device but not in the cache */ + for (i = 0; i < BLOCK_TEST_SIZE; i++) { + KUNIT_EXPECT_EQ(test, 0, regmap_read(map, param->from_reg + i, &val)); + KUNIT_EXPECT_EQ(test, val, rval[i]); + + switch (i) { + case 2: + case 5: + KUNIT_EXPECT_EQ(test, true, data->written[param->from_reg + i]); + KUNIT_EXPECT_EQ(test, data->vals[param->from_reg + i], rval[i] + 1); + break; + default: + KUNIT_EXPECT_EQ(test, false, data->written[param->from_reg + i]); + KUNIT_EXPECT_EQ(test, data->vals[param->from_reg + i], rval[i]); + break; + } + } +} + +static void cache_drop(struct kunit *test) +{ + const struct regmap_test_param *param = test->param_value; + struct regmap *map; + struct regmap_config config; + struct regmap_ram_data *data; + unsigned int rval[BLOCK_TEST_SIZE]; + int i; + + config = test_regmap_config; + config.num_reg_defaults = BLOCK_TEST_SIZE; + + map = gen_regmap(test, &config, &data); + KUNIT_ASSERT_FALSE(test, IS_ERR(map)); + if (IS_ERR(map)) + return; + + /* Ensure the data is read from the cache */ + for (i = 0; i < BLOCK_TEST_SIZE; i++) + data->read[param->from_reg + i] = false; + KUNIT_EXPECT_EQ(test, 0, regmap_bulk_read(map, param->from_reg, rval, + BLOCK_TEST_SIZE)); + for (i = 0; i < BLOCK_TEST_SIZE; i++) { + KUNIT_EXPECT_FALSE(test, data->read[param->from_reg + i]); + data->read[param->from_reg + i] = false; + } + KUNIT_EXPECT_MEMEQ(test, &data->vals[param->from_reg], rval, sizeof(rval)); + + /* Drop some registers */ + KUNIT_EXPECT_EQ(test, 0, regcache_drop_region(map, param->from_reg + 3, + param->from_reg + 5)); + + /* Reread and check only the dropped registers hit the device. */ + KUNIT_EXPECT_EQ(test, 0, regmap_bulk_read(map, param->from_reg, rval, + BLOCK_TEST_SIZE)); + for (i = 0; i < BLOCK_TEST_SIZE; i++) + KUNIT_EXPECT_EQ(test, data->read[param->from_reg + i], i >= 3 && i <= 5); + KUNIT_EXPECT_MEMEQ(test, &data->vals[param->from_reg], rval, sizeof(rval)); +} + +static void cache_drop_with_non_contiguous_ranges(struct kunit *test) +{ + const struct regmap_test_param *param = test->param_value; + struct regmap *map; + struct regmap_config config; + struct regmap_ram_data *data; + unsigned int val[4][BLOCK_TEST_SIZE]; + unsigned int reg; + const int num_ranges = ARRAY_SIZE(val) * 2; + int rangeidx, i; + + static_assert(ARRAY_SIZE(val) == 4); + + config = test_regmap_config; + config.max_register = param->from_reg + (num_ranges * BLOCK_TEST_SIZE); + + map = gen_regmap(test, &config, &data); + KUNIT_ASSERT_FALSE(test, IS_ERR(map)); + if (IS_ERR(map)) + return; + + for (i = 0; i < config.max_register + 1; i++) + data->written[i] = false; + + /* Create non-contiguous cache blocks by writing every other range */ + get_random_bytes(&val, sizeof(val)); + for (rangeidx = 0; rangeidx < num_ranges; rangeidx += 2) { + reg = param->from_reg + (rangeidx * BLOCK_TEST_SIZE); + KUNIT_EXPECT_EQ(test, 0, regmap_bulk_write(map, reg, + &val[rangeidx / 2], + BLOCK_TEST_SIZE)); + KUNIT_EXPECT_MEMEQ(test, &data->vals[reg], + &val[rangeidx / 2], sizeof(val[rangeidx / 2])); + } + + /* Check that odd ranges weren't written */ + for (rangeidx = 1; rangeidx < num_ranges; rangeidx += 2) { + reg = param->from_reg + (rangeidx * BLOCK_TEST_SIZE); + for (i = 0; i < BLOCK_TEST_SIZE; i++) + KUNIT_EXPECT_FALSE(test, data->written[reg + i]); + } + + /* Drop range 2 */ + reg = param->from_reg + (2 * BLOCK_TEST_SIZE); + KUNIT_EXPECT_EQ(test, 0, regcache_drop_region(map, reg, reg + BLOCK_TEST_SIZE - 1)); + + /* Drop part of range 4 */ + reg = param->from_reg + (4 * BLOCK_TEST_SIZE); + KUNIT_EXPECT_EQ(test, 0, regcache_drop_region(map, reg + 3, reg + 5)); + + /* Mark dirty and reset mock registers to 0 */ + regcache_mark_dirty(map); + for (i = 0; i < config.max_register + 1; i++) { + data->vals[i] = 0; + data->written[i] = false; + } + + /* The registers that were dropped from range 4 should now remain at 0 */ + val[4 / 2][3] = 0; + val[4 / 2][4] = 0; + val[4 / 2][5] = 0; + + /* Sync and check that the expected register ranges were written */ + KUNIT_EXPECT_EQ(test, 0, regcache_sync(map)); + + /* Check that odd ranges weren't written */ + for (rangeidx = 1; rangeidx < num_ranges; rangeidx += 2) { + reg = param->from_reg + (rangeidx * BLOCK_TEST_SIZE); + for (i = 0; i < BLOCK_TEST_SIZE; i++) + KUNIT_EXPECT_FALSE(test, data->written[reg + i]); + } + + /* Check that even ranges (except 2 and 4) were written */ + for (rangeidx = 0; rangeidx < num_ranges; rangeidx += 2) { + if ((rangeidx == 2) || (rangeidx == 4)) + continue; + + reg = param->from_reg + (rangeidx * BLOCK_TEST_SIZE); + for (i = 0; i < BLOCK_TEST_SIZE; i++) + KUNIT_EXPECT_TRUE(test, data->written[reg + i]); + + KUNIT_EXPECT_MEMEQ(test, &data->vals[reg], + &val[rangeidx / 2], sizeof(val[rangeidx / 2])); + } + + /* Check that range 2 wasn't written */ + reg = param->from_reg + (2 * BLOCK_TEST_SIZE); + for (i = 0; i < BLOCK_TEST_SIZE; i++) + KUNIT_EXPECT_FALSE(test, data->written[reg + i]); + + /* Check that range 4 was partially written */ + reg = param->from_reg + (4 * BLOCK_TEST_SIZE); + for (i = 0; i < BLOCK_TEST_SIZE; i++) + KUNIT_EXPECT_EQ(test, data->written[reg + i], i < 3 || i > 5); + + KUNIT_EXPECT_MEMEQ(test, &data->vals[reg], &val[4 / 2], sizeof(val[4 / 2])); + + /* Nothing before param->from_reg should have been written */ + for (i = 0; i < param->from_reg; i++) + KUNIT_EXPECT_FALSE(test, data->written[i]); +} + +static void cache_drop_all_and_sync_marked_dirty(struct kunit *test) +{ + const struct regmap_test_param *param = test->param_value; + struct regmap *map; + struct regmap_config config; + struct regmap_ram_data *data; + unsigned int rval[BLOCK_TEST_SIZE]; + int i; + + config = test_regmap_config; + config.num_reg_defaults = BLOCK_TEST_SIZE; + + map = gen_regmap(test, &config, &data); + KUNIT_ASSERT_FALSE(test, IS_ERR(map)); + if (IS_ERR(map)) + return; + + /* Ensure the data is read from the cache */ + for (i = 0; i < BLOCK_TEST_SIZE; i++) + data->read[param->from_reg + i] = false; + KUNIT_EXPECT_EQ(test, 0, regmap_bulk_read(map, param->from_reg, rval, + BLOCK_TEST_SIZE)); + KUNIT_EXPECT_MEMEQ(test, &data->vals[param->from_reg], rval, sizeof(rval)); + + /* Change all values in cache from defaults */ + for (i = 0; i < BLOCK_TEST_SIZE; i++) + KUNIT_EXPECT_EQ(test, 0, regmap_write(map, param->from_reg + i, rval[i] + 1)); + + /* Drop all registers */ + KUNIT_EXPECT_EQ(test, 0, regcache_drop_region(map, 0, config.max_register)); + + /* Mark dirty and cache sync should not write anything. */ + regcache_mark_dirty(map); + for (i = 0; i < BLOCK_TEST_SIZE; i++) + data->written[param->from_reg + i] = false; + + KUNIT_EXPECT_EQ(test, 0, regcache_sync(map)); + for (i = 0; i <= config.max_register; i++) + KUNIT_EXPECT_FALSE(test, data->written[i]); +} + +static void cache_drop_all_and_sync_no_defaults(struct kunit *test) +{ + const struct regmap_test_param *param = test->param_value; + struct regmap *map; + struct regmap_config config; + struct regmap_ram_data *data; + unsigned int rval[BLOCK_TEST_SIZE]; + int i; + + config = test_regmap_config; + + map = gen_regmap(test, &config, &data); + KUNIT_ASSERT_FALSE(test, IS_ERR(map)); + if (IS_ERR(map)) + return; + + /* Ensure the data is read from the cache */ + for (i = 0; i < BLOCK_TEST_SIZE; i++) + data->read[param->from_reg + i] = false; + KUNIT_EXPECT_EQ(test, 0, regmap_bulk_read(map, param->from_reg, rval, + BLOCK_TEST_SIZE)); + KUNIT_EXPECT_MEMEQ(test, &data->vals[param->from_reg], rval, sizeof(rval)); + + /* Change all values in cache */ + for (i = 0; i < BLOCK_TEST_SIZE; i++) + KUNIT_EXPECT_EQ(test, 0, regmap_write(map, param->from_reg + i, rval[i] + 1)); + + /* Drop all registers */ + KUNIT_EXPECT_EQ(test, 0, regcache_drop_region(map, 0, config.max_register)); + + /* + * Sync cache without marking it dirty. All registers were dropped + * so the cache should not have any entries to write out. + */ + for (i = 0; i < BLOCK_TEST_SIZE; i++) + data->written[param->from_reg + i] = false; + + KUNIT_EXPECT_EQ(test, 0, regcache_sync(map)); + for (i = 0; i <= config.max_register; i++) + KUNIT_EXPECT_FALSE(test, data->written[i]); +} + +static void cache_drop_all_and_sync_has_defaults(struct kunit *test) +{ + const struct regmap_test_param *param = test->param_value; + struct regmap *map; + struct regmap_config config; + struct regmap_ram_data *data; + unsigned int rval[BLOCK_TEST_SIZE]; + int i; + + config = test_regmap_config; + config.num_reg_defaults = BLOCK_TEST_SIZE; + + map = gen_regmap(test, &config, &data); + KUNIT_ASSERT_FALSE(test, IS_ERR(map)); + if (IS_ERR(map)) + return; + + /* Ensure the data is read from the cache */ + for (i = 0; i < BLOCK_TEST_SIZE; i++) + data->read[param->from_reg + i] = false; + KUNIT_EXPECT_EQ(test, 0, regmap_bulk_read(map, param->from_reg, rval, + BLOCK_TEST_SIZE)); + KUNIT_EXPECT_MEMEQ(test, &data->vals[param->from_reg], rval, sizeof(rval)); + + /* Change all values in cache from defaults */ + for (i = 0; i < BLOCK_TEST_SIZE; i++) + KUNIT_EXPECT_EQ(test, 0, regmap_write(map, param->from_reg + i, rval[i] + 1)); + + /* Drop all registers */ + KUNIT_EXPECT_EQ(test, 0, regcache_drop_region(map, 0, config.max_register)); + + /* + * Sync cache without marking it dirty. All registers were dropped + * so the cache should not have any entries to write out. + */ + for (i = 0; i < BLOCK_TEST_SIZE; i++) + data->written[param->from_reg + i] = false; + + KUNIT_EXPECT_EQ(test, 0, regcache_sync(map)); + for (i = 0; i <= config.max_register; i++) + KUNIT_EXPECT_FALSE(test, data->written[i]); +} + +static void cache_present(struct kunit *test) +{ + const struct regmap_test_param *param = test->param_value; + struct regmap *map; + struct regmap_config config; + struct regmap_ram_data *data; + unsigned int val; + int i; + + config = test_regmap_config; + + map = gen_regmap(test, &config, &data); + KUNIT_ASSERT_FALSE(test, IS_ERR(map)); + if (IS_ERR(map)) + return; + + for (i = 0; i < BLOCK_TEST_SIZE; i++) + data->read[param->from_reg + i] = false; + + /* No defaults so no registers cached. */ + for (i = 0; i < BLOCK_TEST_SIZE; i++) + KUNIT_ASSERT_FALSE(test, regcache_reg_cached(map, param->from_reg + i)); + + /* We didn't trigger any reads */ + for (i = 0; i < BLOCK_TEST_SIZE; i++) + KUNIT_ASSERT_FALSE(test, data->read[param->from_reg + i]); + + /* Fill the cache */ + for (i = 0; i < BLOCK_TEST_SIZE; i++) + KUNIT_EXPECT_EQ(test, 0, regmap_read(map, param->from_reg + i, &val)); + + /* Now everything should be cached */ + for (i = 0; i < BLOCK_TEST_SIZE; i++) + KUNIT_ASSERT_TRUE(test, regcache_reg_cached(map, param->from_reg + i)); +} + +static void cache_write_zero(struct kunit *test) +{ + const struct regmap_test_param *param = test->param_value; + struct regmap *map; + struct regmap_config config; + struct regmap_ram_data *data; + unsigned int val; + int i; + + config = test_regmap_config; + + map = gen_regmap(test, &config, &data); + KUNIT_ASSERT_FALSE(test, IS_ERR(map)); + if (IS_ERR(map)) + return; + + for (i = 0; i < BLOCK_TEST_SIZE; i++) + data->read[param->from_reg + i] = false; + + /* No defaults so no registers cached. */ + for (i = 0; i < BLOCK_TEST_SIZE; i++) + KUNIT_ASSERT_FALSE(test, regcache_reg_cached(map, param->from_reg + i)); + + /* We didn't trigger any reads */ + for (i = 0; i < BLOCK_TEST_SIZE; i++) + KUNIT_ASSERT_FALSE(test, data->read[param->from_reg + i]); + + /* Write a zero value */ + KUNIT_EXPECT_EQ(test, 0, regmap_write(map, 1, 0)); + + /* Read that zero value back */ + KUNIT_EXPECT_EQ(test, 0, regmap_read(map, 1, &val)); + KUNIT_EXPECT_EQ(test, 0, val); + + /* From the cache? */ + KUNIT_ASSERT_TRUE(test, regcache_reg_cached(map, 1)); + + /* Try to throw it away */ + KUNIT_EXPECT_EQ(test, 0, regcache_drop_region(map, 1, 1)); + KUNIT_ASSERT_FALSE(test, regcache_reg_cached(map, 1)); +} + +/* Check that caching the window register works with sync */ +static void cache_range_window_reg(struct kunit *test) +{ + struct regmap *map; + struct regmap_config config; + struct regmap_ram_data *data; + unsigned int val; + int i; + + config = test_regmap_config; + config.volatile_reg = test_range_window_volatile; + config.ranges = &test_range; + config.num_ranges = 1; + config.max_register = test_range.range_max; + + map = gen_regmap(test, &config, &data); + KUNIT_ASSERT_FALSE(test, IS_ERR(map)); + if (IS_ERR(map)) + return; + + /* Write new values to the entire range */ + for (i = test_range.range_min; i <= test_range.range_max; i++) + KUNIT_ASSERT_EQ(test, 0, regmap_write(map, i, 0)); + + val = data->vals[test_range.selector_reg] & test_range.selector_mask; + KUNIT_ASSERT_EQ(test, val, 2); + + /* Write to the first register in the range to reset the page */ + KUNIT_ASSERT_EQ(test, 0, regmap_write(map, test_range.range_min, 0)); + val = data->vals[test_range.selector_reg] & test_range.selector_mask; + KUNIT_ASSERT_EQ(test, val, 0); + + /* Trigger a cache sync */ + regcache_mark_dirty(map); + KUNIT_ASSERT_EQ(test, 0, regcache_sync(map)); + + /* Write to the first register again, the page should be reset */ + KUNIT_ASSERT_EQ(test, 0, regmap_write(map, test_range.range_min, 0)); + val = data->vals[test_range.selector_reg] & test_range.selector_mask; + KUNIT_ASSERT_EQ(test, val, 0); + + /* Trigger another cache sync */ + regcache_mark_dirty(map); + KUNIT_ASSERT_EQ(test, 0, regcache_sync(map)); + + /* Write to the last register again, the page should be reset */ + KUNIT_ASSERT_EQ(test, 0, regmap_write(map, test_range.range_max, 0)); + val = data->vals[test_range.selector_reg] & test_range.selector_mask; + KUNIT_ASSERT_EQ(test, val, 2); +} + +static const struct regmap_test_param raw_types_list[] = { + { .cache = REGCACHE_NONE, .val_endian = REGMAP_ENDIAN_LITTLE }, + { .cache = REGCACHE_NONE, .val_endian = REGMAP_ENDIAN_BIG }, + { .cache = REGCACHE_FLAT, .val_endian = REGMAP_ENDIAN_LITTLE }, + { .cache = REGCACHE_FLAT, .val_endian = REGMAP_ENDIAN_BIG }, + { .cache = REGCACHE_FLAT_S, .val_endian = REGMAP_ENDIAN_LITTLE }, + { .cache = REGCACHE_FLAT_S, .val_endian = REGMAP_ENDIAN_BIG }, + { .cache = REGCACHE_RBTREE, .val_endian = REGMAP_ENDIAN_LITTLE }, + { .cache = REGCACHE_RBTREE, .val_endian = REGMAP_ENDIAN_BIG }, + { .cache = REGCACHE_MAPLE, .val_endian = REGMAP_ENDIAN_LITTLE }, + { .cache = REGCACHE_MAPLE, .val_endian = REGMAP_ENDIAN_BIG }, +}; + +KUNIT_ARRAY_PARAM(raw_test_types, raw_types_list, param_to_desc); + +static const struct regmap_test_param raw_cache_types_list[] = { + { .cache = REGCACHE_FLAT, .val_endian = REGMAP_ENDIAN_LITTLE }, + { .cache = REGCACHE_FLAT, .val_endian = REGMAP_ENDIAN_BIG }, + { .cache = REGCACHE_FLAT_S, .val_endian = REGMAP_ENDIAN_LITTLE }, + { .cache = REGCACHE_FLAT_S, .val_endian = REGMAP_ENDIAN_BIG }, + { .cache = REGCACHE_RBTREE, .val_endian = REGMAP_ENDIAN_LITTLE }, + { .cache = REGCACHE_RBTREE, .val_endian = REGMAP_ENDIAN_BIG }, + { .cache = REGCACHE_MAPLE, .val_endian = REGMAP_ENDIAN_LITTLE }, + { .cache = REGCACHE_MAPLE, .val_endian = REGMAP_ENDIAN_BIG }, +}; + +KUNIT_ARRAY_PARAM(raw_test_cache_types, raw_cache_types_list, param_to_desc); + +static const struct regmap_config raw_regmap_config = { + .max_register = BLOCK_TEST_SIZE, + + .reg_format_endian = REGMAP_ENDIAN_LITTLE, + .reg_bits = 16, + .val_bits = 16, +}; + +static struct regmap *gen_raw_regmap(struct kunit *test, + struct regmap_config *config, + struct regmap_ram_data **data) +{ + struct regmap_test_priv *priv = test->priv; + const struct regmap_test_param *param = test->param_value; + u16 *buf; + struct regmap *ret = ERR_PTR(-ENOMEM); + int i, error; + struct reg_default *defaults; + size_t size; + + config->cache_type = param->cache; + config->val_format_endian = param->val_endian; + config->disable_locking = config->cache_type == REGCACHE_RBTREE || + config->cache_type == REGCACHE_MAPLE; + + size = array_size(config->max_register + 1, BITS_TO_BYTES(config->reg_bits)); + buf = kmalloc(size, GFP_KERNEL); + if (!buf) + return ERR_PTR(-ENOMEM); + + get_random_bytes(buf, size); + + *data = kzalloc(sizeof(**data), GFP_KERNEL); + if (!(*data)) + goto out_free; + (*data)->vals = (void *)buf; + + config->num_reg_defaults = config->max_register + 1; + defaults = kunit_kcalloc(test, + config->num_reg_defaults, + sizeof(struct reg_default), + GFP_KERNEL); + if (!defaults) + goto out_free; + config->reg_defaults = defaults; + + for (i = 0; i < config->num_reg_defaults; i++) { + defaults[i].reg = i; + switch (param->val_endian) { + case REGMAP_ENDIAN_LITTLE: + defaults[i].def = le16_to_cpu(buf[i]); + break; + case REGMAP_ENDIAN_BIG: + defaults[i].def = be16_to_cpu(buf[i]); + break; + default: + ret = ERR_PTR(-EINVAL); + goto out_free; + } + } + + /* + * We use the defaults in the tests but they don't make sense + * to the core if there's no cache. + */ + if (config->cache_type == REGCACHE_NONE) + config->num_reg_defaults = 0; + + ret = regmap_init_raw_ram(priv->dev, config, *data); + if (IS_ERR(ret)) + goto out_free; + + /* This calls regmap_exit() on failure, which frees buf and *data */ + error = kunit_add_action_or_reset(test, regmap_exit_action, ret); + if (error) + ret = ERR_PTR(error); + + return ret; + +out_free: + kfree(buf); + kfree(*data); + + return ret; +} + +static void raw_read_defaults_single(struct kunit *test) +{ + struct regmap *map; + struct regmap_config config; + struct regmap_ram_data *data; + unsigned int rval; + int i; + + config = raw_regmap_config; + + map = gen_raw_regmap(test, &config, &data); + KUNIT_ASSERT_FALSE(test, IS_ERR(map)); + if (IS_ERR(map)) + return; + + /* Check that we can read the defaults via the API */ + for (i = 0; i < config.max_register + 1; i++) { + KUNIT_EXPECT_EQ(test, 0, regmap_read(map, i, &rval)); + KUNIT_EXPECT_EQ(test, config.reg_defaults[i].def, rval); + } +} + +static void raw_read_defaults(struct kunit *test) +{ + struct regmap *map; + struct regmap_config config; + struct regmap_ram_data *data; + u16 *rval; + u16 def; + size_t val_len; + int i; + + config = raw_regmap_config; + + map = gen_raw_regmap(test, &config, &data); + KUNIT_ASSERT_FALSE(test, IS_ERR(map)); + if (IS_ERR(map)) + return; + + val_len = array_size(sizeof(*rval), config.max_register + 1); + rval = kunit_kmalloc(test, val_len, GFP_KERNEL); + KUNIT_ASSERT_TRUE(test, rval != NULL); + if (!rval) + return; + + /* Check that we can read the defaults via the API */ + KUNIT_EXPECT_EQ(test, 0, regmap_raw_read(map, 0, rval, val_len)); + for (i = 0; i < config.max_register + 1; i++) { + def = config.reg_defaults[i].def; + if (config.val_format_endian == REGMAP_ENDIAN_BIG) { + KUNIT_EXPECT_EQ(test, def, be16_to_cpu((__force __be16)rval[i])); + } else { + KUNIT_EXPECT_EQ(test, def, le16_to_cpu((__force __le16)rval[i])); + } + } +} + +static void raw_write_read_single(struct kunit *test) +{ + struct regmap *map; + struct regmap_config config; + struct regmap_ram_data *data; + u16 val; + unsigned int rval; + + config = raw_regmap_config; + + map = gen_raw_regmap(test, &config, &data); + KUNIT_ASSERT_FALSE(test, IS_ERR(map)); + if (IS_ERR(map)) + return; + + get_random_bytes(&val, sizeof(val)); + + /* If we write a value to a register we can read it back */ + KUNIT_EXPECT_EQ(test, 0, regmap_write(map, 0, val)); + KUNIT_EXPECT_EQ(test, 0, regmap_read(map, 0, &rval)); + KUNIT_EXPECT_EQ(test, val, rval); +} + +static void raw_write(struct kunit *test) +{ + struct regmap *map; + struct regmap_config config; + struct regmap_ram_data *data; + u16 *hw_buf; + u16 val[2]; + unsigned int rval; + int i; + + config = raw_regmap_config; + + map = gen_raw_regmap(test, &config, &data); + KUNIT_ASSERT_FALSE(test, IS_ERR(map)); + if (IS_ERR(map)) + return; + + hw_buf = (u16 *)data->vals; + + get_random_bytes(&val, sizeof(val)); + + /* Do a raw write */ + KUNIT_EXPECT_EQ(test, 0, regmap_raw_write(map, 2, val, sizeof(val))); + + /* We should read back the new values, and defaults for the rest */ + for (i = 0; i < config.max_register + 1; i++) { + KUNIT_EXPECT_EQ(test, 0, regmap_read(map, i, &rval)); + + switch (i) { + case 2: + case 3: + if (config.val_format_endian == REGMAP_ENDIAN_BIG) { + KUNIT_EXPECT_EQ(test, rval, + be16_to_cpu((__force __be16)val[i % 2])); + } else { + KUNIT_EXPECT_EQ(test, rval, + le16_to_cpu((__force __le16)val[i % 2])); + } + break; + default: + KUNIT_EXPECT_EQ(test, config.reg_defaults[i].def, rval); + break; + } + } + + /* The values should appear in the "hardware" */ + KUNIT_EXPECT_MEMEQ(test, &hw_buf[2], val, sizeof(val)); +} + +static bool reg_zero(struct device *dev, unsigned int reg) +{ + return reg == 0; +} + +static bool ram_reg_zero(struct regmap_ram_data *data, unsigned int reg) +{ + return reg == 0; +} + +static void raw_noinc_write(struct kunit *test) +{ + struct regmap *map; + struct regmap_config config; + struct regmap_ram_data *data; + unsigned int val; + u16 val_test, val_last; + u16 val_array[BLOCK_TEST_SIZE]; + + config = raw_regmap_config; + config.volatile_reg = reg_zero; + config.writeable_noinc_reg = reg_zero; + config.readable_noinc_reg = reg_zero; + + map = gen_raw_regmap(test, &config, &data); + KUNIT_ASSERT_FALSE(test, IS_ERR(map)); + if (IS_ERR(map)) + return; + + data->noinc_reg = ram_reg_zero; + + get_random_bytes(&val_array, sizeof(val_array)); + + if (config.val_format_endian == REGMAP_ENDIAN_BIG) { + val_test = be16_to_cpu(val_array[1]) + 100; + val_last = be16_to_cpu(val_array[BLOCK_TEST_SIZE - 1]); + } else { + val_test = le16_to_cpu(val_array[1]) + 100; + val_last = le16_to_cpu(val_array[BLOCK_TEST_SIZE - 1]); + } + + /* Put some data into the register following the noinc register */ + KUNIT_EXPECT_EQ(test, 0, regmap_write(map, 1, val_test)); + + /* Write some data to the noinc register */ + KUNIT_EXPECT_EQ(test, 0, regmap_noinc_write(map, 0, val_array, + sizeof(val_array))); + + /* We should read back the last value written */ + KUNIT_EXPECT_EQ(test, 0, regmap_read(map, 0, &val)); + KUNIT_ASSERT_EQ(test, val_last, val); + + /* Make sure we didn't touch the register after the noinc register */ + KUNIT_EXPECT_EQ(test, 0, regmap_read(map, 1, &val)); + KUNIT_ASSERT_EQ(test, val_test, val); +} + +static void raw_sync(struct kunit *test) +{ + struct regmap *map; + struct regmap_config config; + struct regmap_ram_data *data; + u16 val[3]; + u16 *hw_buf; + unsigned int rval; + int i; + + config = raw_regmap_config; + + map = gen_raw_regmap(test, &config, &data); + KUNIT_ASSERT_FALSE(test, IS_ERR(map)); + if (IS_ERR(map)) + return; + + hw_buf = (u16 *)data->vals; + + get_changed_bytes(&hw_buf[2], &val[0], sizeof(val)); + + /* Do a regular write and a raw write in cache only mode */ + regcache_cache_only(map, true); + KUNIT_EXPECT_EQ(test, 0, regmap_raw_write(map, 2, val, + sizeof(u16) * 2)); + KUNIT_EXPECT_EQ(test, 0, regmap_write(map, 4, val[2])); + + /* We should read back the new values, and defaults for the rest */ + for (i = 0; i < config.max_register + 1; i++) { + KUNIT_EXPECT_EQ(test, 0, regmap_read(map, i, &rval)); + + switch (i) { + case 2: + case 3: + if (config.val_format_endian == REGMAP_ENDIAN_BIG) { + KUNIT_EXPECT_EQ(test, rval, + be16_to_cpu((__force __be16)val[i - 2])); + } else { + KUNIT_EXPECT_EQ(test, rval, + le16_to_cpu((__force __le16)val[i - 2])); + } + break; + case 4: + KUNIT_EXPECT_EQ(test, rval, val[i - 2]); + break; + default: + KUNIT_EXPECT_EQ(test, config.reg_defaults[i].def, rval); + break; + } + } + + /* + * The value written via _write() was translated by the core, + * translate the original copy for comparison purposes. + */ + if (config.val_format_endian == REGMAP_ENDIAN_BIG) + val[2] = cpu_to_be16(val[2]); + else + val[2] = cpu_to_le16(val[2]); + + /* The values should not appear in the "hardware" */ + KUNIT_EXPECT_MEMNEQ(test, &hw_buf[2], &val[0], sizeof(val)); + + for (i = 0; i < config.max_register + 1; i++) + data->written[i] = false; + + /* Do the sync */ + regcache_cache_only(map, false); + regcache_mark_dirty(map); + KUNIT_EXPECT_EQ(test, 0, regcache_sync(map)); + + /* The values should now appear in the "hardware" */ + KUNIT_EXPECT_MEMEQ(test, &hw_buf[2], &val[0], sizeof(val)); +} + +static void raw_ranges(struct kunit *test) +{ + struct regmap *map; + struct regmap_config config; + struct regmap_ram_data *data; + unsigned int val; + int i; + + config = raw_regmap_config; + config.volatile_reg = test_range_all_volatile; + config.ranges = &test_range; + config.num_ranges = 1; + config.max_register = test_range.range_max; + + map = gen_raw_regmap(test, &config, &data); + KUNIT_ASSERT_FALSE(test, IS_ERR(map)); + if (IS_ERR(map)) + return; + + /* Reset the page to a non-zero value to trigger a change */ + KUNIT_EXPECT_EQ(test, 0, regmap_write(map, test_range.selector_reg, + test_range.range_max)); + + /* Check we set the page and use the window for writes */ + data->written[test_range.selector_reg] = false; + data->written[test_range.window_start] = false; + KUNIT_EXPECT_EQ(test, 0, regmap_write(map, test_range.range_min, 0)); + KUNIT_EXPECT_TRUE(test, data->written[test_range.selector_reg]); + KUNIT_EXPECT_TRUE(test, data->written[test_range.window_start]); + + data->written[test_range.selector_reg] = false; + data->written[test_range.window_start] = false; + KUNIT_EXPECT_EQ(test, 0, regmap_write(map, + test_range.range_min + + test_range.window_len, + 0)); + KUNIT_EXPECT_TRUE(test, data->written[test_range.selector_reg]); + KUNIT_EXPECT_TRUE(test, data->written[test_range.window_start]); + + /* Same for reads */ + data->written[test_range.selector_reg] = false; + data->read[test_range.window_start] = false; + KUNIT_EXPECT_EQ(test, 0, regmap_read(map, test_range.range_min, &val)); + KUNIT_EXPECT_TRUE(test, data->written[test_range.selector_reg]); + KUNIT_EXPECT_TRUE(test, data->read[test_range.window_start]); + + data->written[test_range.selector_reg] = false; + data->read[test_range.window_start] = false; + KUNIT_EXPECT_EQ(test, 0, regmap_read(map, + test_range.range_min + + test_range.window_len, + &val)); + KUNIT_EXPECT_TRUE(test, data->written[test_range.selector_reg]); + KUNIT_EXPECT_TRUE(test, data->read[test_range.window_start]); + + /* No physical access triggered in the virtual range */ + for (i = test_range.range_min; i < test_range.range_max; i++) { + KUNIT_EXPECT_FALSE(test, data->read[i]); + KUNIT_EXPECT_FALSE(test, data->written[i]); + } +} + +static struct kunit_case regmap_test_cases[] = { + KUNIT_CASE_PARAM(basic_read_write, regcache_types_gen_params), + KUNIT_CASE_PARAM(read_bypassed, real_cache_types_gen_params), + KUNIT_CASE_PARAM(read_bypassed_volatile, real_cache_types_gen_params), + KUNIT_CASE_PARAM(bulk_write, regcache_types_gen_params), + KUNIT_CASE_PARAM(bulk_read, regcache_types_gen_params), + KUNIT_CASE_PARAM(multi_write, regcache_types_gen_params), + KUNIT_CASE_PARAM(multi_read, regcache_types_gen_params), + KUNIT_CASE_PARAM(write_readonly, regcache_types_gen_params), + KUNIT_CASE_PARAM(read_writeonly, regcache_types_gen_params), + KUNIT_CASE_PARAM(reg_defaults, regcache_types_gen_params), + KUNIT_CASE_PARAM(reg_defaults_read_dev, regcache_types_gen_params), + KUNIT_CASE_PARAM(register_patch, regcache_types_gen_params), + KUNIT_CASE_PARAM(stride, regcache_types_gen_params), + KUNIT_CASE_PARAM(basic_ranges, regcache_types_gen_params), + KUNIT_CASE_PARAM(stress_insert, regcache_types_gen_params), + KUNIT_CASE_PARAM(cache_bypass, real_cache_types_gen_params), + KUNIT_CASE_PARAM(cache_sync_marked_dirty, real_cache_types_gen_params), + KUNIT_CASE_PARAM(cache_sync_after_cache_only, real_cache_types_gen_params), + KUNIT_CASE_PARAM(cache_sync_defaults_marked_dirty, real_cache_types_gen_params), + KUNIT_CASE_PARAM(cache_sync_default_after_cache_only, real_cache_types_gen_params), + KUNIT_CASE_PARAM(cache_sync_readonly, real_cache_types_gen_params), + KUNIT_CASE_PARAM(cache_sync_patch, real_cache_types_gen_params), + KUNIT_CASE_PARAM(cache_drop, sparse_cache_types_gen_params), + KUNIT_CASE_PARAM(cache_drop_with_non_contiguous_ranges, sparse_cache_types_gen_params), + KUNIT_CASE_PARAM(cache_drop_all_and_sync_marked_dirty, sparse_cache_types_gen_params), + KUNIT_CASE_PARAM(cache_drop_all_and_sync_no_defaults, sparse_cache_types_gen_params), + KUNIT_CASE_PARAM(cache_drop_all_and_sync_has_defaults, sparse_cache_types_gen_params), + KUNIT_CASE_PARAM(cache_present, sparse_cache_types_gen_params), + KUNIT_CASE_PARAM(cache_write_zero, sparse_cache_types_gen_params), + KUNIT_CASE_PARAM(cache_range_window_reg, real_cache_types_only_gen_params), + + KUNIT_CASE_PARAM(raw_read_defaults_single, raw_test_types_gen_params), + KUNIT_CASE_PARAM(raw_read_defaults, raw_test_types_gen_params), + KUNIT_CASE_PARAM(raw_write_read_single, raw_test_types_gen_params), + KUNIT_CASE_PARAM(raw_write, raw_test_types_gen_params), + KUNIT_CASE_PARAM(raw_noinc_write, raw_test_types_gen_params), + KUNIT_CASE_PARAM(raw_sync, raw_test_cache_types_gen_params), + KUNIT_CASE_PARAM(raw_ranges, raw_test_cache_types_gen_params), + {} +}; + +static int regmap_test_init(struct kunit *test) +{ + struct regmap_test_priv *priv; + struct device *dev; + + priv = kunit_kzalloc(test, sizeof(*priv), GFP_KERNEL); + if (!priv) + return -ENOMEM; + + test->priv = priv; + + dev = kunit_device_register(test, "regmap_test"); + if (IS_ERR(dev)) + return PTR_ERR(dev); + + priv->dev = get_device(dev); + dev_set_drvdata(dev, test); + + return 0; +} + +static void regmap_test_exit(struct kunit *test) +{ + struct regmap_test_priv *priv = test->priv; + + /* Destroy the dummy struct device */ + if (priv && priv->dev) + put_device(priv->dev); +} + +static struct kunit_suite regmap_test_suite = { + .name = "regmap", + .init = regmap_test_init, + .exit = regmap_test_exit, + .test_cases = regmap_test_cases, +}; +kunit_test_suite(regmap_test_suite); + +MODULE_DESCRIPTION("Regmap KUnit tests"); +MODULE_LICENSE("GPL v2"); diff --git a/drivers/base/regmap/regmap-mdio.c b/drivers/base/regmap/regmap-mdio.c index 6a20201299f5..9573bf3b52f4 100644 --- a/drivers/base/regmap/regmap-mdio.c +++ b/drivers/base/regmap/regmap-mdio.c @@ -10,31 +10,21 @@ /* Clause-45 mask includes the device type (5 bit) and actual register number (16 bit) */ #define REGNUM_C45_MASK GENMASK(20, 0) -static int regmap_mdio_read(struct mdio_device *mdio_dev, u32 reg, unsigned int *val) +static int regmap_mdio_c22_read(void *context, unsigned int reg, unsigned int *val) { + struct mdio_device *mdio_dev = context; int ret; - ret = mdiobus_read(mdio_dev->bus, mdio_dev->addr, reg); + if (unlikely(reg & ~REGNUM_C22_MASK)) + return -ENXIO; + + ret = mdiodev_read(mdio_dev, reg); if (ret < 0) return ret; *val = ret & REGVAL_MASK; - return 0; -} - -static int regmap_mdio_write(struct mdio_device *mdio_dev, u32 reg, unsigned int val) -{ - return mdiobus_write(mdio_dev->bus, mdio_dev->addr, reg, val); -} - -static int regmap_mdio_c22_read(void *context, unsigned int reg, unsigned int *val) -{ - struct mdio_device *mdio_dev = context; - if (unlikely(reg & ~REGNUM_C22_MASK)) - return -ENXIO; - - return regmap_mdio_read(mdio_dev, reg, val); + return 0; } static int regmap_mdio_c22_write(void *context, unsigned int reg, unsigned int val) @@ -44,7 +34,7 @@ static int regmap_mdio_c22_write(void *context, unsigned int reg, unsigned int v if (unlikely(reg & ~REGNUM_C22_MASK)) return -ENXIO; - return mdiobus_write(mdio_dev->bus, mdio_dev->addr, reg, val); + return mdiodev_write(mdio_dev, reg, val); } static const struct regmap_bus regmap_mdio_c22_bus = { @@ -55,21 +45,36 @@ static const struct regmap_bus regmap_mdio_c22_bus = { static int regmap_mdio_c45_read(void *context, unsigned int reg, unsigned int *val) { struct mdio_device *mdio_dev = context; + unsigned int devad; + int ret; if (unlikely(reg & ~REGNUM_C45_MASK)) return -ENXIO; - return regmap_mdio_read(mdio_dev, MII_ADDR_C45 | reg, val); + devad = reg >> REGMAP_MDIO_C45_DEVAD_SHIFT; + reg = reg & REGMAP_MDIO_C45_REGNUM_MASK; + + ret = mdiodev_c45_read(mdio_dev, devad, reg); + if (ret < 0) + return ret; + + *val = ret & REGVAL_MASK; + + return 0; } static int regmap_mdio_c45_write(void *context, unsigned int reg, unsigned int val) { struct mdio_device *mdio_dev = context; + unsigned int devad; if (unlikely(reg & ~REGNUM_C45_MASK)) return -ENXIO; - return regmap_mdio_write(mdio_dev, MII_ADDR_C45 | reg, val); + devad = reg >> REGMAP_MDIO_C45_DEVAD_SHIFT; + reg = reg & REGMAP_MDIO_C45_REGNUM_MASK; + + return mdiodev_c45_write(mdio_dev, devad, reg, val); } static const struct regmap_bus regmap_mdio_c45_bus = { @@ -112,5 +117,5 @@ struct regmap *__devm_regmap_init_mdio(struct mdio_device *mdio_dev, EXPORT_SYMBOL_GPL(__devm_regmap_init_mdio); MODULE_AUTHOR("Sander Vanheule <sander@svanheule.net>"); -MODULE_DESCRIPTION("Regmap MDIO Module"); +MODULE_DESCRIPTION("regmap MDIO Module"); MODULE_LICENSE("GPL v2"); diff --git a/drivers/base/regmap/regmap-mmio.c b/drivers/base/regmap/regmap-mmio.c index f9cd51afb9d2..29e5f3175301 100644 --- a/drivers/base/regmap/regmap-mmio.c +++ b/drivers/base/regmap/regmap-mmio.c @@ -10,13 +10,14 @@ #include <linux/module.h> #include <linux/regmap.h> #include <linux/slab.h> +#include <linux/swab.h> #include "internal.h" struct regmap_mmio_context { void __iomem *regs; - unsigned val_bytes; - bool relaxed_mmio; + unsigned int val_bytes; + bool big_endian; bool attached_clk; struct clk *clk; @@ -33,9 +34,6 @@ static int regmap_mmio_regbits_check(size_t reg_bits) case 8: case 16: case 32: -#ifdef CONFIG_64BIT - case 64: -#endif return 0; default: return -EINVAL; @@ -50,18 +48,13 @@ static int regmap_mmio_get_min_stride(size_t val_bits) case 8: /* The core treats 0 as 1 */ min_stride = 0; - return 0; + break; case 16: min_stride = 2; break; case 32: min_stride = 4; break; -#ifdef CONFIG_64BIT - case 64: - min_stride = 8; - break; -#endif default: return -EINVAL; } @@ -83,6 +76,12 @@ static void regmap_mmio_write8_relaxed(struct regmap_mmio_context *ctx, writeb_relaxed(val, ctx->regs + reg); } +static void regmap_mmio_iowrite8(struct regmap_mmio_context *ctx, + unsigned int reg, unsigned int val) +{ + iowrite8(val, ctx->regs + reg); +} + static void regmap_mmio_write16le(struct regmap_mmio_context *ctx, unsigned int reg, unsigned int val) @@ -97,10 +96,22 @@ static void regmap_mmio_write16le_relaxed(struct regmap_mmio_context *ctx, writew_relaxed(val, ctx->regs + reg); } +static void regmap_mmio_iowrite16le(struct regmap_mmio_context *ctx, + unsigned int reg, unsigned int val) +{ + iowrite16(val, ctx->regs + reg); +} + static void regmap_mmio_write16be(struct regmap_mmio_context *ctx, unsigned int reg, unsigned int val) { + writew(swab16(val), ctx->regs + reg); +} + +static void regmap_mmio_iowrite16be(struct regmap_mmio_context *ctx, + unsigned int reg, unsigned int val) +{ iowrite16be(val, ctx->regs + reg); } @@ -118,28 +129,24 @@ static void regmap_mmio_write32le_relaxed(struct regmap_mmio_context *ctx, writel_relaxed(val, ctx->regs + reg); } -static void regmap_mmio_write32be(struct regmap_mmio_context *ctx, - unsigned int reg, - unsigned int val) +static void regmap_mmio_iowrite32le(struct regmap_mmio_context *ctx, + unsigned int reg, unsigned int val) { - iowrite32be(val, ctx->regs + reg); + iowrite32(val, ctx->regs + reg); } -#ifdef CONFIG_64BIT -static void regmap_mmio_write64le(struct regmap_mmio_context *ctx, +static void regmap_mmio_write32be(struct regmap_mmio_context *ctx, unsigned int reg, unsigned int val) { - writeq(val, ctx->regs + reg); + writel(swab32(val), ctx->regs + reg); } -static void regmap_mmio_write64le_relaxed(struct regmap_mmio_context *ctx, - unsigned int reg, - unsigned int val) +static void regmap_mmio_iowrite32be(struct regmap_mmio_context *ctx, + unsigned int reg, unsigned int val) { - writeq_relaxed(val, ctx->regs + reg); + iowrite32be(val, ctx->regs + reg); } -#endif static int regmap_mmio_write(void *context, unsigned int reg, unsigned int val) { @@ -160,6 +167,69 @@ static int regmap_mmio_write(void *context, unsigned int reg, unsigned int val) return 0; } +static int regmap_mmio_noinc_write(void *context, unsigned int reg, + const void *val, size_t val_count) +{ + struct regmap_mmio_context *ctx = context; + int ret = 0; + int i; + + if (!IS_ERR(ctx->clk)) { + ret = clk_enable(ctx->clk); + if (ret < 0) + return ret; + } + + /* + * There are no native, assembly-optimized write single register + * operations for big endian, so fall back to emulation if this + * is needed. (Single bytes are fine, they are not affected by + * endianness.) + */ + if (ctx->big_endian && (ctx->val_bytes > 1)) { + switch (ctx->val_bytes) { + case 2: + { + const u16 *valp = (const u16 *)val; + for (i = 0; i < val_count; i++) + writew(swab16(valp[i]), ctx->regs + reg); + goto out_clk; + } + case 4: + { + const u32 *valp = (const u32 *)val; + for (i = 0; i < val_count; i++) + writel(swab32(valp[i]), ctx->regs + reg); + goto out_clk; + } + default: + ret = -EINVAL; + goto out_clk; + } + } + + switch (ctx->val_bytes) { + case 1: + writesb(ctx->regs + reg, (const u8 *)val, val_count); + break; + case 2: + writesw(ctx->regs + reg, (const u16 *)val, val_count); + break; + case 4: + writesl(ctx->regs + reg, (const u32 *)val, val_count); + break; + default: + ret = -EINVAL; + break; + } + +out_clk: + if (!IS_ERR(ctx->clk)) + clk_disable(ctx->clk); + + return ret; +} + static unsigned int regmap_mmio_read8(struct regmap_mmio_context *ctx, unsigned int reg) { @@ -172,6 +242,12 @@ static unsigned int regmap_mmio_read8_relaxed(struct regmap_mmio_context *ctx, return readb_relaxed(ctx->regs + reg); } +static unsigned int regmap_mmio_ioread8(struct regmap_mmio_context *ctx, + unsigned int reg) +{ + return ioread8(ctx->regs + reg); +} + static unsigned int regmap_mmio_read16le(struct regmap_mmio_context *ctx, unsigned int reg) { @@ -184,9 +260,21 @@ static unsigned int regmap_mmio_read16le_relaxed(struct regmap_mmio_context *ctx return readw_relaxed(ctx->regs + reg); } +static unsigned int regmap_mmio_ioread16le(struct regmap_mmio_context *ctx, + unsigned int reg) +{ + return ioread16(ctx->regs + reg); +} + static unsigned int regmap_mmio_read16be(struct regmap_mmio_context *ctx, unsigned int reg) { + return swab16(readw(ctx->regs + reg)); +} + +static unsigned int regmap_mmio_ioread16be(struct regmap_mmio_context *ctx, + unsigned int reg) +{ return ioread16be(ctx->regs + reg); } @@ -202,25 +290,23 @@ static unsigned int regmap_mmio_read32le_relaxed(struct regmap_mmio_context *ctx return readl_relaxed(ctx->regs + reg); } -static unsigned int regmap_mmio_read32be(struct regmap_mmio_context *ctx, - unsigned int reg) +static unsigned int regmap_mmio_ioread32le(struct regmap_mmio_context *ctx, + unsigned int reg) { - return ioread32be(ctx->regs + reg); + return ioread32(ctx->regs + reg); } -#ifdef CONFIG_64BIT -static unsigned int regmap_mmio_read64le(struct regmap_mmio_context *ctx, +static unsigned int regmap_mmio_read32be(struct regmap_mmio_context *ctx, unsigned int reg) { - return readq(ctx->regs + reg); + return swab32(readl(ctx->regs + reg)); } -static unsigned int regmap_mmio_read64le_relaxed(struct regmap_mmio_context *ctx, - unsigned int reg) +static unsigned int regmap_mmio_ioread32be(struct regmap_mmio_context *ctx, + unsigned int reg) { - return readq_relaxed(ctx->regs + reg); + return ioread32be(ctx->regs + reg); } -#endif static int regmap_mmio_read(void *context, unsigned int reg, unsigned int *val) { @@ -241,6 +327,61 @@ static int regmap_mmio_read(void *context, unsigned int reg, unsigned int *val) return 0; } +static int regmap_mmio_noinc_read(void *context, unsigned int reg, + void *val, size_t val_count) +{ + struct regmap_mmio_context *ctx = context; + int ret = 0; + + if (!IS_ERR(ctx->clk)) { + ret = clk_enable(ctx->clk); + if (ret < 0) + return ret; + } + + switch (ctx->val_bytes) { + case 1: + readsb(ctx->regs + reg, (u8 *)val, val_count); + break; + case 2: + readsw(ctx->regs + reg, (u16 *)val, val_count); + break; + case 4: + readsl(ctx->regs + reg, (u32 *)val, val_count); + break; + default: + ret = -EINVAL; + goto out_clk; + } + + /* + * There are no native, assembly-optimized write single register + * operations for big endian, so fall back to emulation if this + * is needed. (Single bytes are fine, they are not affected by + * endianness.) + */ + if (ctx->big_endian && (ctx->val_bytes > 1)) { + switch (ctx->val_bytes) { + case 2: + swab16_array(val, val_count); + break; + case 4: + swab32_array(val, val_count); + break; + default: + ret = -EINVAL; + break; + } + } + +out_clk: + if (!IS_ERR(ctx->clk)) + clk_disable(ctx->clk); + + return ret; +} + + static void regmap_mmio_free_context(void *context) { struct regmap_mmio_context *ctx = context; @@ -257,6 +398,8 @@ static const struct regmap_bus regmap_mmio = { .fast_io = true, .reg_write = regmap_mmio_write, .reg_read = regmap_mmio_read, + .reg_noinc_write = regmap_mmio_noinc_write, + .reg_noinc_read = regmap_mmio_noinc_read, .free_context = regmap_mmio_free_context, .val_format_endian_default = REGMAP_ENDIAN_LITTLE, }; @@ -281,7 +424,10 @@ static struct regmap_mmio_context *regmap_mmio_gen_context(struct device *dev, if (min_stride < 0) return ERR_PTR(min_stride); - if (config->reg_stride < min_stride) + if (config->reg_stride && config->reg_stride < min_stride) + return ERR_PTR(-EINVAL); + + if (config->use_relaxed_mmio && config->io_port) return ERR_PTR(-EINVAL); ctx = kzalloc(sizeof(*ctx), GFP_KERNEL); @@ -290,7 +436,6 @@ static struct regmap_mmio_context *regmap_mmio_gen_context(struct device *dev, ctx->regs = regs; ctx->val_bytes = config->val_bits / 8; - ctx->relaxed_mmio = config->use_relaxed_mmio; ctx->clk = ERR_PTR(-ENODEV); switch (regmap_get_val_endian(dev, ®map_mmio, config)) { @@ -301,7 +446,10 @@ static struct regmap_mmio_context *regmap_mmio_gen_context(struct device *dev, #endif switch (config->val_bits) { case 8: - if (ctx->relaxed_mmio) { + if (config->io_port) { + ctx->reg_read = regmap_mmio_ioread8; + ctx->reg_write = regmap_mmio_iowrite8; + } else if (config->use_relaxed_mmio) { ctx->reg_read = regmap_mmio_read8_relaxed; ctx->reg_write = regmap_mmio_write8_relaxed; } else { @@ -310,7 +458,10 @@ static struct regmap_mmio_context *regmap_mmio_gen_context(struct device *dev, } break; case 16: - if (ctx->relaxed_mmio) { + if (config->io_port) { + ctx->reg_read = regmap_mmio_ioread16le; + ctx->reg_write = regmap_mmio_iowrite16le; + } else if (config->use_relaxed_mmio) { ctx->reg_read = regmap_mmio_read16le_relaxed; ctx->reg_write = regmap_mmio_write16le_relaxed; } else { @@ -319,7 +470,10 @@ static struct regmap_mmio_context *regmap_mmio_gen_context(struct device *dev, } break; case 32: - if (ctx->relaxed_mmio) { + if (config->io_port) { + ctx->reg_read = regmap_mmio_ioread32le; + ctx->reg_write = regmap_mmio_iowrite32le; + } else if (config->use_relaxed_mmio) { ctx->reg_read = regmap_mmio_read32le_relaxed; ctx->reg_write = regmap_mmio_write32le_relaxed; } else { @@ -327,17 +481,6 @@ static struct regmap_mmio_context *regmap_mmio_gen_context(struct device *dev, ctx->reg_write = regmap_mmio_write32le; } break; -#ifdef CONFIG_64BIT - case 64: - if (ctx->relaxed_mmio) { - ctx->reg_read = regmap_mmio_read64le_relaxed; - ctx->reg_write = regmap_mmio_write64le_relaxed; - } else { - ctx->reg_read = regmap_mmio_read64le; - ctx->reg_write = regmap_mmio_write64le; - } - break; -#endif default: ret = -EINVAL; goto err_free; @@ -347,18 +490,34 @@ static struct regmap_mmio_context *regmap_mmio_gen_context(struct device *dev, #ifdef __BIG_ENDIAN case REGMAP_ENDIAN_NATIVE: #endif + ctx->big_endian = true; switch (config->val_bits) { case 8: - ctx->reg_read = regmap_mmio_read8; - ctx->reg_write = regmap_mmio_write8; + if (config->io_port) { + ctx->reg_read = regmap_mmio_ioread8; + ctx->reg_write = regmap_mmio_iowrite8; + } else { + ctx->reg_read = regmap_mmio_read8; + ctx->reg_write = regmap_mmio_write8; + } break; case 16: - ctx->reg_read = regmap_mmio_read16be; - ctx->reg_write = regmap_mmio_write16be; + if (config->io_port) { + ctx->reg_read = regmap_mmio_ioread16be; + ctx->reg_write = regmap_mmio_iowrite16be; + } else { + ctx->reg_read = regmap_mmio_read16be; + ctx->reg_write = regmap_mmio_write16be; + } break; case 32: - ctx->reg_read = regmap_mmio_read32be; - ctx->reg_write = regmap_mmio_write32be; + if (config->io_port) { + ctx->reg_read = regmap_mmio_ioread32be; + ctx->reg_write = regmap_mmio_iowrite32be; + } else { + ctx->reg_read = regmap_mmio_read32be; + ctx->reg_write = regmap_mmio_write32be; + } break; default: ret = -EINVAL; @@ -450,4 +609,5 @@ void regmap_mmio_detach_clk(struct regmap *map) } EXPORT_SYMBOL_GPL(regmap_mmio_detach_clk); +MODULE_DESCRIPTION("regmap MMIO Module"); MODULE_LICENSE("GPL v2"); diff --git a/drivers/base/regmap/regmap-ram.c b/drivers/base/regmap/regmap-ram.c new file mode 100644 index 000000000000..4e5b4518ce4d --- /dev/null +++ b/drivers/base/regmap/regmap-ram.c @@ -0,0 +1,87 @@ +// SPDX-License-Identifier: GPL-2.0 +// +// Register map access API - Memory region +// +// This is intended for testing only +// +// Copyright (c) 2023, Arm Ltd + +#include <linux/clk.h> +#include <linux/err.h> +#include <linux/io.h> +#include <linux/module.h> +#include <linux/regmap.h> +#include <linux/slab.h> +#include <linux/swab.h> + +#include "internal.h" + +static int regmap_ram_write(void *context, unsigned int reg, unsigned int val) +{ + struct regmap_ram_data *data = context; + + data->vals[reg] = val; + data->written[reg] = true; + + return 0; +} + +static int regmap_ram_read(void *context, unsigned int reg, unsigned int *val) +{ + struct regmap_ram_data *data = context; + + *val = data->vals[reg]; + data->read[reg] = true; + + return 0; +} + +static void regmap_ram_free_context(void *context) +{ + struct regmap_ram_data *data = context; + + kfree(data->vals); + kfree(data->read); + kfree(data->written); + kfree(data); +} + +static const struct regmap_bus regmap_ram = { + .fast_io = true, + .reg_write = regmap_ram_write, + .reg_read = regmap_ram_read, + .free_context = regmap_ram_free_context, +}; + +struct regmap *__regmap_init_ram(struct device *dev, + const struct regmap_config *config, + struct regmap_ram_data *data, + struct lock_class_key *lock_key, + const char *lock_name) +{ + struct regmap *map; + + if (!config->max_register) { + pr_crit("No max_register specified for RAM regmap\n"); + return ERR_PTR(-EINVAL); + } + + data->read = kcalloc(config->max_register + 1, sizeof(bool), + GFP_KERNEL); + if (!data->read) + return ERR_PTR(-ENOMEM); + + data->written = kcalloc(config->max_register + 1, sizeof(bool), + GFP_KERNEL); + if (!data->written) + return ERR_PTR(-ENOMEM); + + map = __regmap_init(dev, ®map_ram, data, config, + lock_key, lock_name); + + return map; +} +EXPORT_SYMBOL_GPL(__regmap_init_ram); + +MODULE_DESCRIPTION("Register map access API - Memory region"); +MODULE_LICENSE("GPL v2"); diff --git a/drivers/base/regmap/regmap-raw-ram.c b/drivers/base/regmap/regmap-raw-ram.c new file mode 100644 index 000000000000..76c98814fb8a --- /dev/null +++ b/drivers/base/regmap/regmap-raw-ram.c @@ -0,0 +1,146 @@ +// SPDX-License-Identifier: GPL-2.0 +// +// Register map access API - Memory region with raw access +// +// This is intended for testing only +// +// Copyright (c) 2023, Arm Ltd + +#include <linux/clk.h> +#include <linux/err.h> +#include <linux/io.h> +#include <linux/module.h> +#include <linux/regmap.h> +#include <linux/slab.h> +#include <linux/swab.h> + +#include "internal.h" + +static unsigned int decode_reg(enum regmap_endian endian, const void *reg) +{ + const u16 *r = reg; + + if (endian == REGMAP_ENDIAN_BIG) + return be16_to_cpu(*r); + else + return le16_to_cpu(*r); +} + +static int regmap_raw_ram_gather_write(void *context, + const void *reg, size_t reg_len, + const void *val, size_t val_len) +{ + struct regmap_ram_data *data = context; + unsigned int r; + u16 *our_buf = (u16 *)data->vals; + int i; + + if (reg_len != 2) + return -EINVAL; + if (val_len % 2) + return -EINVAL; + + r = decode_reg(data->reg_endian, reg); + if (data->noinc_reg && data->noinc_reg(data, r)) { + memcpy(&our_buf[r], val + val_len - 2, 2); + data->written[r] = true; + } else { + memcpy(&our_buf[r], val, val_len); + + for (i = 0; i < val_len / 2; i++) + data->written[r + i] = true; + } + + return 0; +} + +static int regmap_raw_ram_write(void *context, const void *data, size_t count) +{ + return regmap_raw_ram_gather_write(context, data, 2, + data + 2, count - 2); +} + +static int regmap_raw_ram_read(void *context, + const void *reg, size_t reg_len, + void *val, size_t val_len) +{ + struct regmap_ram_data *data = context; + unsigned int r; + u16 *our_buf = (u16 *)data->vals; + int i; + + if (reg_len != 2) + return -EINVAL; + if (val_len % 2) + return -EINVAL; + + r = decode_reg(data->reg_endian, reg); + if (data->noinc_reg && data->noinc_reg(data, r)) { + for (i = 0; i < val_len; i += 2) + memcpy(val + i, &our_buf[r], 2); + data->read[r] = true; + } else { + memcpy(val, &our_buf[r], val_len); + + for (i = 0; i < val_len / 2; i++) + data->read[r + i] = true; + } + + return 0; +} + +static void regmap_raw_ram_free_context(void *context) +{ + struct regmap_ram_data *data = context; + + kfree(data->vals); + kfree(data->read); + kfree(data->written); + kfree(data); +} + +static const struct regmap_bus regmap_raw_ram = { + .fast_io = true, + .write = regmap_raw_ram_write, + .gather_write = regmap_raw_ram_gather_write, + .read = regmap_raw_ram_read, + .free_context = regmap_raw_ram_free_context, +}; + +struct regmap *__regmap_init_raw_ram(struct device *dev, + const struct regmap_config *config, + struct regmap_ram_data *data, + struct lock_class_key *lock_key, + const char *lock_name) +{ + struct regmap *map; + + if (config->reg_bits != 16) + return ERR_PTR(-EINVAL); + + if (!config->max_register) { + pr_crit("No max_register specified for RAM regmap\n"); + return ERR_PTR(-EINVAL); + } + + data->read = kcalloc(config->max_register + 1, sizeof(bool), + GFP_KERNEL); + if (!data->read) + return ERR_PTR(-ENOMEM); + + data->written = kcalloc(config->max_register + 1, sizeof(bool), + GFP_KERNEL); + if (!data->written) + return ERR_PTR(-ENOMEM); + + data->reg_endian = config->reg_format_endian; + + map = __regmap_init(dev, ®map_raw_ram, data, config, + lock_key, lock_name); + + return map; +} +EXPORT_SYMBOL_GPL(__regmap_init_raw_ram); + +MODULE_DESCRIPTION("Register map access API - Memory region with raw access"); +MODULE_LICENSE("GPL v2"); diff --git a/drivers/base/regmap/regmap-sccb.c b/drivers/base/regmap/regmap-sccb.c index 597042e2d009..12bbbb03e5f2 100644 --- a/drivers/base/regmap/regmap-sccb.c +++ b/drivers/base/regmap/regmap-sccb.c @@ -80,7 +80,7 @@ static int regmap_sccb_write(void *context, unsigned int reg, unsigned int val) return i2c_smbus_write_byte_data(i2c, reg, val); } -static struct regmap_bus regmap_sccb_bus = { +static const struct regmap_bus regmap_sccb_bus = { .reg_write = regmap_sccb_write, .reg_read = regmap_sccb_read, }; @@ -125,4 +125,5 @@ struct regmap *__devm_regmap_init_sccb(struct i2c_client *i2c, } EXPORT_SYMBOL_GPL(__devm_regmap_init_sccb); +MODULE_DESCRIPTION("Register map access API - SCCB support"); MODULE_LICENSE("GPL v2"); diff --git a/drivers/base/regmap/regmap-sdw-mbq.c b/drivers/base/regmap/regmap-sdw-mbq.c index fe3ac26b66ad..6a61629f5f89 100644 --- a/drivers/base/regmap/regmap-sdw-mbq.c +++ b/drivers/base/regmap/regmap-sdw-mbq.c @@ -1,48 +1,190 @@ // SPDX-License-Identifier: GPL-2.0 // Copyright(c) 2020 Intel Corporation. +#include <linux/bits.h> +#include <linux/delay.h> #include <linux/device.h> #include <linux/errno.h> +#include <linux/iopoll.h> #include <linux/module.h> #include <linux/regmap.h> #include <linux/soundwire/sdw.h> #include <linux/soundwire/sdw_registers.h> +#include <sound/sdca_function.h> #include "internal.h" +struct regmap_mbq_context { + struct device *dev; + struct sdw_slave *sdw; + + bool (*readable_reg)(struct device *dev, unsigned int reg); + + struct regmap_sdw_mbq_cfg cfg; + + int val_size; +}; + +static int regmap_sdw_mbq_size(struct regmap_mbq_context *ctx, unsigned int reg) +{ + int size = ctx->val_size; + + if (ctx->cfg.mbq_size) { + size = ctx->cfg.mbq_size(ctx->dev, reg); + if (!size || size > ctx->val_size) + return -EINVAL; + } + + return size; +} + +static bool regmap_sdw_mbq_deferrable(struct regmap_mbq_context *ctx, unsigned int reg) +{ + if (ctx->cfg.deferrable) + return ctx->cfg.deferrable(ctx->dev, reg); + + return false; +} + +static int regmap_sdw_mbq_poll_busy(struct sdw_slave *slave, unsigned int reg, + struct regmap_mbq_context *ctx) +{ + struct device *dev = ctx->dev; + int val, ret = 0; + + dev_dbg(dev, "Deferring transaction for 0x%x\n", reg); + + reg = SDW_SDCA_CTL(SDW_SDCA_CTL_FUNC(reg), 0, + SDCA_CTL_ENTITY_0_FUNCTION_STATUS, 0); + + if (ctx->readable_reg(dev, reg)) { + ret = read_poll_timeout(sdw_read_no_pm, val, + val < 0 || !(val & SDCA_CTL_ENTITY_0_FUNCTION_BUSY), + ctx->cfg.timeout_us, ctx->cfg.retry_us, + false, slave, reg); + if (val < 0) + return val; + if (ret) + dev_err(dev, "Function busy timed out 0x%x: %d\n", reg, val); + } else { + fsleep(ctx->cfg.timeout_us); + } + + return ret; +} + +static int regmap_sdw_mbq_write_impl(struct sdw_slave *slave, + unsigned int reg, unsigned int val, + int mbq_size, bool deferrable) +{ + int shift = mbq_size * BITS_PER_BYTE; + int ret; + + while (--mbq_size > 0) { + shift -= BITS_PER_BYTE; + + ret = sdw_write_no_pm(slave, SDW_SDCA_MBQ_CTL(reg), + (val >> shift) & 0xff); + if (ret < 0) + return ret; + } + + ret = sdw_write_no_pm(slave, reg, val & 0xff); + if (deferrable && ret == -ENODATA) + return -EAGAIN; + + return ret; +} + static int regmap_sdw_mbq_write(void *context, unsigned int reg, unsigned int val) { - struct device *dev = context; - struct sdw_slave *slave = dev_to_sdw_dev(dev); + struct regmap_mbq_context *ctx = context; + struct sdw_slave *slave = ctx->sdw; + bool deferrable = regmap_sdw_mbq_deferrable(ctx, reg); + int mbq_size = regmap_sdw_mbq_size(ctx, reg); int ret; - ret = sdw_write_no_pm(slave, SDW_SDCA_MBQ_CTL(reg), (val >> 8) & 0xff); - if (ret < 0) - return ret; + if (mbq_size < 0) + return mbq_size; + + /* + * Technically the spec does allow a device to set itself to busy for + * internal reasons, but since it doesn't provide any information on + * how to handle timeouts in that case, for now the code will only + * process a single wait/timeout on function busy and a single retry + * of the transaction. + */ + ret = regmap_sdw_mbq_write_impl(slave, reg, val, mbq_size, deferrable); + if (ret == -EAGAIN) { + ret = regmap_sdw_mbq_poll_busy(slave, reg, ctx); + if (ret) + return ret; + + ret = regmap_sdw_mbq_write_impl(slave, reg, val, mbq_size, false); + } - return sdw_write_no_pm(slave, reg, val & 0xff); + return ret; } -static int regmap_sdw_mbq_read(void *context, unsigned int reg, unsigned int *val) +static int regmap_sdw_mbq_read_impl(struct sdw_slave *slave, + unsigned int reg, unsigned int *val, + int mbq_size, bool deferrable) { - struct device *dev = context; - struct sdw_slave *slave = dev_to_sdw_dev(dev); - int read0; - int read1; + int shift = BITS_PER_BYTE; + int read; + + read = sdw_read_no_pm(slave, reg); + if (read < 0) { + if (deferrable && read == -ENODATA) + return -EAGAIN; + + return read; + } - read0 = sdw_read_no_pm(slave, reg); - if (read0 < 0) - return read0; + *val = read; - read1 = sdw_read_no_pm(slave, SDW_SDCA_MBQ_CTL(reg)); - if (read1 < 0) - return read1; + while (--mbq_size > 0) { + read = sdw_read_no_pm(slave, SDW_SDCA_MBQ_CTL(reg)); + if (read < 0) + return read; - *val = (read1 << 8) | read0; + *val |= read << shift; + shift += BITS_PER_BYTE; + } return 0; } -static struct regmap_bus regmap_sdw_mbq = { +static int regmap_sdw_mbq_read(void *context, unsigned int reg, unsigned int *val) +{ + struct regmap_mbq_context *ctx = context; + struct sdw_slave *slave = ctx->sdw; + bool deferrable = regmap_sdw_mbq_deferrable(ctx, reg); + int mbq_size = regmap_sdw_mbq_size(ctx, reg); + int ret; + + if (mbq_size < 0) + return mbq_size; + + /* + * Technically the spec does allow a device to set itself to busy for + * internal reasons, but since it doesn't provide any information on + * how to handle timeouts in that case, for now the code will only + * process a single wait/timeout on function busy and a single retry + * of the transaction. + */ + ret = regmap_sdw_mbq_read_impl(slave, reg, val, mbq_size, deferrable); + if (ret == -EAGAIN) { + ret = regmap_sdw_mbq_poll_busy(slave, reg, ctx); + if (ret) + return ret; + + ret = regmap_sdw_mbq_read_impl(slave, reg, val, mbq_size, false); + } + + return ret; +} + +static const struct regmap_bus regmap_sdw_mbq = { .reg_read = regmap_sdw_mbq_read, .reg_write = regmap_sdw_mbq_write, .reg_format_endian_default = REGMAP_ENDIAN_LITTLE, @@ -51,8 +193,7 @@ static struct regmap_bus regmap_sdw_mbq = { static int regmap_sdw_mbq_config_check(const struct regmap_config *config) { - /* MBQ-based controls are only 16-bits for now */ - if (config->val_bits != 16) + if (config->val_bits > (sizeof(unsigned int) * BITS_PER_BYTE)) return -ENOTSUPP; /* Registers are 32 bits wide */ @@ -65,37 +206,73 @@ static int regmap_sdw_mbq_config_check(const struct regmap_config *config) return 0; } -struct regmap *__regmap_init_sdw_mbq(struct sdw_slave *sdw, +static struct regmap_mbq_context * +regmap_sdw_mbq_gen_context(struct device *dev, + struct sdw_slave *sdw, + const struct regmap_config *config, + const struct regmap_sdw_mbq_cfg *mbq_config) +{ + struct regmap_mbq_context *ctx; + + ctx = devm_kzalloc(dev, sizeof(*ctx), GFP_KERNEL); + if (!ctx) + return ERR_PTR(-ENOMEM); + + ctx->dev = dev; + ctx->sdw = sdw; + + if (mbq_config) + ctx->cfg = *mbq_config; + + ctx->val_size = config->val_bits / BITS_PER_BYTE; + ctx->readable_reg = config->readable_reg; + + return ctx; +} + +struct regmap *__regmap_init_sdw_mbq(struct device *dev, struct sdw_slave *sdw, const struct regmap_config *config, + const struct regmap_sdw_mbq_cfg *mbq_config, struct lock_class_key *lock_key, const char *lock_name) { + struct regmap_mbq_context *ctx; int ret; ret = regmap_sdw_mbq_config_check(config); if (ret) return ERR_PTR(ret); - return __regmap_init(&sdw->dev, ®map_sdw_mbq, - &sdw->dev, config, lock_key, lock_name); + ctx = regmap_sdw_mbq_gen_context(dev, sdw, config, mbq_config); + if (IS_ERR(ctx)) + return ERR_CAST(ctx); + + return __regmap_init(dev, ®map_sdw_mbq, ctx, + config, lock_key, lock_name); } EXPORT_SYMBOL_GPL(__regmap_init_sdw_mbq); -struct regmap *__devm_regmap_init_sdw_mbq(struct sdw_slave *sdw, +struct regmap *__devm_regmap_init_sdw_mbq(struct device *dev, struct sdw_slave *sdw, const struct regmap_config *config, + const struct regmap_sdw_mbq_cfg *mbq_config, struct lock_class_key *lock_key, const char *lock_name) { + struct regmap_mbq_context *ctx; int ret; ret = regmap_sdw_mbq_config_check(config); if (ret) return ERR_PTR(ret); - return __devm_regmap_init(&sdw->dev, ®map_sdw_mbq, - &sdw->dev, config, lock_key, lock_name); + ctx = regmap_sdw_mbq_gen_context(dev, sdw, config, mbq_config); + if (IS_ERR(ctx)) + return ERR_CAST(ctx); + + return __devm_regmap_init(dev, ®map_sdw_mbq, ctx, + config, lock_key, lock_name); } EXPORT_SYMBOL_GPL(__devm_regmap_init_sdw_mbq); -MODULE_DESCRIPTION("Regmap SoundWire MBQ Module"); +MODULE_DESCRIPTION("regmap SoundWire MBQ Module"); MODULE_LICENSE("GPL"); diff --git a/drivers/base/regmap/regmap-sdw.c b/drivers/base/regmap/regmap-sdw.c index 966de8a136d9..ea631ac7c7ec 100644 --- a/drivers/base/regmap/regmap-sdw.c +++ b/drivers/base/regmap/regmap-sdw.c @@ -6,50 +6,63 @@ #include <linux/module.h> #include <linux/regmap.h> #include <linux/soundwire/sdw.h> +#include <linux/types.h> #include "internal.h" -static int regmap_sdw_write(void *context, unsigned int reg, unsigned int val) +static int regmap_sdw_write(void *context, const void *val_buf, size_t val_size) { struct device *dev = context; struct sdw_slave *slave = dev_to_sdw_dev(dev); + /* First word of buffer contains the destination address */ + u32 addr = le32_to_cpu(*(const __le32 *)val_buf); + const u8 *val = val_buf; - return sdw_write_no_pm(slave, reg, val); + return sdw_nwrite_no_pm(slave, addr, val_size - sizeof(addr), val + sizeof(addr)); } -static int regmap_sdw_read(void *context, unsigned int reg, unsigned int *val) +static int regmap_sdw_gather_write(void *context, + const void *reg_buf, size_t reg_size, + const void *val_buf, size_t val_size) { struct device *dev = context; struct sdw_slave *slave = dev_to_sdw_dev(dev); - int read; + u32 addr = le32_to_cpu(*(const __le32 *)reg_buf); - read = sdw_read_no_pm(slave, reg); - if (read < 0) - return read; + return sdw_nwrite_no_pm(slave, addr, val_size, val_buf); +} - *val = read; - return 0; +static int regmap_sdw_read(void *context, + const void *reg_buf, size_t reg_size, + void *val_buf, size_t val_size) +{ + struct device *dev = context; + struct sdw_slave *slave = dev_to_sdw_dev(dev); + u32 addr = le32_to_cpu(*(const __le32 *)reg_buf); + + return sdw_nread_no_pm(slave, addr, val_size, val_buf); } -static struct regmap_bus regmap_sdw = { - .reg_read = regmap_sdw_read, - .reg_write = regmap_sdw_write, +static const struct regmap_bus regmap_sdw = { + .write = regmap_sdw_write, + .gather_write = regmap_sdw_gather_write, + .read = regmap_sdw_read, .reg_format_endian_default = REGMAP_ENDIAN_LITTLE, .val_format_endian_default = REGMAP_ENDIAN_LITTLE, }; static int regmap_sdw_config_check(const struct regmap_config *config) { - /* All register are 8-bits wide as per MIPI Soundwire 1.0 Spec */ - if (config->val_bits != 8) - return -ENOTSUPP; - - /* Registers are 32 bits wide */ + /* Register addresses are 32 bits wide */ if (config->reg_bits != 32) return -ENOTSUPP; if (config->pad_bits != 0) return -ENOTSUPP; + /* Only bulk writes are supported not multi-register writes */ + if (config->can_multi_write) + return -ENOTSUPP; + return 0; } @@ -85,5 +98,5 @@ struct regmap *__devm_regmap_init_sdw(struct sdw_slave *sdw, } EXPORT_SYMBOL_GPL(__devm_regmap_init_sdw); -MODULE_DESCRIPTION("Regmap SoundWire Module"); +MODULE_DESCRIPTION("regmap SoundWire Module"); MODULE_LICENSE("GPL v2"); diff --git a/drivers/base/regmap/regmap-slimbus.c b/drivers/base/regmap/regmap-slimbus.c index 0968059f1ef5..e523fae73004 100644 --- a/drivers/base/regmap/regmap-slimbus.c +++ b/drivers/base/regmap/regmap-slimbus.c @@ -22,7 +22,7 @@ static int regmap_slimbus_read(void *context, const void *reg, size_t reg_size, return slim_read(sdev, *(u16 *)reg, val_size, val); } -static struct regmap_bus regmap_slimbus_bus = { +static const struct regmap_bus regmap_slimbus_bus = { .write = regmap_slimbus_write, .read = regmap_slimbus_read, .reg_format_endian_default = REGMAP_ENDIAN_LITTLE, @@ -48,8 +48,7 @@ struct regmap *__regmap_init_slimbus(struct slim_device *slimbus, if (IS_ERR(bus)) return ERR_CAST(bus); - return __regmap_init(&slimbus->dev, bus, &slimbus->dev, config, - lock_key, lock_name); + return __regmap_init(&slimbus->dev, bus, slimbus, config, lock_key, lock_name); } EXPORT_SYMBOL_GPL(__regmap_init_slimbus); @@ -63,9 +62,9 @@ struct regmap *__devm_regmap_init_slimbus(struct slim_device *slimbus, if (IS_ERR(bus)) return ERR_CAST(bus); - return __devm_regmap_init(&slimbus->dev, bus, &slimbus, config, - lock_key, lock_name); + return __devm_regmap_init(&slimbus->dev, bus, slimbus, config, lock_key, lock_name); } EXPORT_SYMBOL_GPL(__devm_regmap_init_slimbus); +MODULE_DESCRIPTION("Register map access API - SLIMbus support"); MODULE_LICENSE("GPL v2"); diff --git a/drivers/base/regmap/regmap-spi-avmm.c b/drivers/base/regmap/regmap-spi-avmm.c index ad1da83e849f..d86a06cadcdb 100644 --- a/drivers/base/regmap/regmap-spi-avmm.c +++ b/drivers/base/regmap/regmap-spi-avmm.c @@ -7,6 +7,7 @@ #include <linux/module.h> #include <linux/regmap.h> #include <linux/spi/spi.h> +#include <linux/swab.h> /* * This driver implements the regmap operations for a generic SPI @@ -162,19 +163,12 @@ struct spi_avmm_bridge { /* bridge buffer used in translation between protocol layers */ char trans_buf[TRANS_BUF_SIZE]; char phy_buf[PHY_BUF_SIZE]; - void (*swap_words)(char *buf, unsigned int len); + void (*swap_words)(void *buf, unsigned int len); }; -static void br_swap_words_32(char *buf, unsigned int len) +static void br_swap_words_32(void *buf, unsigned int len) { - u32 *p = (u32 *)buf; - unsigned int count; - - count = len / 4; - while (count--) { - *p = swab32p(p); - p++; - } + swab32_array(buf, len / 4); } /* @@ -716,4 +710,5 @@ struct regmap *__devm_regmap_init_spi_avmm(struct spi_device *spi, } EXPORT_SYMBOL_GPL(__devm_regmap_init_spi_avmm); +MODULE_DESCRIPTION("Register map access API - SPI AVMM support"); MODULE_LICENSE("GPL v2"); diff --git a/drivers/base/regmap/regmap-spi.c b/drivers/base/regmap/regmap-spi.c index c1894e93c378..14b1d88997cb 100644 --- a/drivers/base/regmap/regmap-spi.c +++ b/drivers/base/regmap/regmap-spi.c @@ -109,13 +109,44 @@ static const struct regmap_bus regmap_spi = { .val_format_endian_default = REGMAP_ENDIAN_BIG, }; +static const struct regmap_bus *regmap_get_spi_bus(struct spi_device *spi, + const struct regmap_config *config) +{ + size_t max_size = spi_max_transfer_size(spi); + size_t max_msg_size, reg_reserve_size; + struct regmap_bus *bus; + + if (max_size != SIZE_MAX) { + bus = kmemdup(®map_spi, sizeof(*bus), GFP_KERNEL); + if (!bus) + return ERR_PTR(-ENOMEM); + + max_msg_size = spi_max_message_size(spi); + reg_reserve_size = (config->reg_bits + config->pad_bits) / BITS_PER_BYTE; + if (max_size + reg_reserve_size > max_msg_size) + max_size -= reg_reserve_size; + + bus->free_on_exit = true; + bus->max_raw_read = max_size; + bus->max_raw_write = max_size; + + return bus; + } + + return ®map_spi; +} + struct regmap *__regmap_init_spi(struct spi_device *spi, const struct regmap_config *config, struct lock_class_key *lock_key, const char *lock_name) { - return __regmap_init(&spi->dev, ®map_spi, &spi->dev, config, - lock_key, lock_name); + const struct regmap_bus *bus = regmap_get_spi_bus(spi, config); + + if (IS_ERR(bus)) + return ERR_CAST(bus); + + return __regmap_init(&spi->dev, bus, &spi->dev, config, lock_key, lock_name); } EXPORT_SYMBOL_GPL(__regmap_init_spi); @@ -124,9 +155,14 @@ struct regmap *__devm_regmap_init_spi(struct spi_device *spi, struct lock_class_key *lock_key, const char *lock_name) { - return __devm_regmap_init(&spi->dev, ®map_spi, &spi->dev, config, - lock_key, lock_name); + const struct regmap_bus *bus = regmap_get_spi_bus(spi, config); + + if (IS_ERR(bus)) + return ERR_CAST(bus); + + return __devm_regmap_init(&spi->dev, bus, &spi->dev, config, lock_key, lock_name); } EXPORT_SYMBOL_GPL(__devm_regmap_init_spi); +MODULE_DESCRIPTION("regmap SPI Module"); MODULE_LICENSE("GPL"); diff --git a/drivers/base/regmap/regmap-spmi.c b/drivers/base/regmap/regmap-spmi.c index cdf12d2aa3a1..347bfe9544ce 100644 --- a/drivers/base/regmap/regmap-spmi.c +++ b/drivers/base/regmap/regmap-spmi.c @@ -222,4 +222,5 @@ struct regmap *__devm_regmap_init_spmi_ext(struct spmi_device *sdev, } EXPORT_SYMBOL_GPL(__devm_regmap_init_spmi_ext); +MODULE_DESCRIPTION("Register map access API - SPMI support"); MODULE_LICENSE("GPL"); diff --git a/drivers/base/regmap/regmap-w1.c b/drivers/base/regmap/regmap-w1.c index 1fbaaad71ca5..29fd24f9c7ed 100644 --- a/drivers/base/regmap/regmap-w1.c +++ b/drivers/base/regmap/regmap-w1.c @@ -172,17 +172,17 @@ static int w1_reg_a16_v16_write(void *context, unsigned int reg, * Various types of supported bus addressing */ -static struct regmap_bus regmap_w1_bus_a8_v8 = { +static const struct regmap_bus regmap_w1_bus_a8_v8 = { .reg_read = w1_reg_a8_v8_read, .reg_write = w1_reg_a8_v8_write, }; -static struct regmap_bus regmap_w1_bus_a8_v16 = { +static const struct regmap_bus regmap_w1_bus_a8_v16 = { .reg_read = w1_reg_a8_v16_read, .reg_write = w1_reg_a8_v16_write, }; -static struct regmap_bus regmap_w1_bus_a16_v16 = { +static const struct regmap_bus regmap_w1_bus_a16_v16 = { .reg_read = w1_reg_a16_v16_read, .reg_write = w1_reg_a16_v16_write, }; @@ -234,4 +234,5 @@ struct regmap *__devm_regmap_init_w1(struct device *w1_dev, } EXPORT_SYMBOL_GPL(__devm_regmap_init_w1); +MODULE_DESCRIPTION("Register map access API - W1 (1-Wire) support"); MODULE_LICENSE("GPL"); diff --git a/drivers/base/regmap/regmap.c b/drivers/base/regmap/regmap.c index fe3e38dd5324..ce9be3989a21 100644 --- a/drivers/base/regmap/regmap.c +++ b/drivers/base/regmap/regmap.c @@ -17,7 +17,7 @@ #include <linux/delay.h> #include <linux/log2.h> #include <linux/hwspinlock.h> -#include <asm/unaligned.h> +#include <linux/unaligned.h> #define CREATE_TRACE_POINTS #include "trace.h" @@ -89,7 +89,7 @@ EXPORT_SYMBOL_GPL(regmap_check_range_table); bool regmap_writeable(struct regmap *map, unsigned int reg) { - if (map->max_register && reg > map->max_register) + if (map->max_register_is_set && reg > map->max_register) return false; if (map->writeable_reg) @@ -112,7 +112,7 @@ bool regmap_cached(struct regmap *map, unsigned int reg) if (!map->cache_ops) return false; - if (map->max_register && reg > map->max_register) + if (map->max_register_is_set && reg > map->max_register) return false; map->lock(map->lock_arg); @@ -129,7 +129,7 @@ bool regmap_readable(struct regmap *map, unsigned int reg) if (!map->reg_read) return false; - if (map->max_register && reg > map->max_register) + if (map->max_register_is_set && reg > map->max_register) return false; if (map->format.format_write) @@ -288,15 +288,9 @@ static void regmap_format_16_native(void *buf, unsigned int val, memcpy(buf, &v, sizeof(v)); } -static void regmap_format_24(void *buf, unsigned int val, unsigned int shift) +static void regmap_format_24_be(void *buf, unsigned int val, unsigned int shift) { - u8 *b = buf; - - val <<= shift; - - b[0] = val >> 16; - b[1] = val >> 8; - b[2] = val; + put_unaligned_be24(val << shift, buf); } static void regmap_format_32_be(void *buf, unsigned int val, unsigned int shift) @@ -317,26 +311,6 @@ static void regmap_format_32_native(void *buf, unsigned int val, memcpy(buf, &v, sizeof(v)); } -#ifdef CONFIG_64BIT -static void regmap_format_64_be(void *buf, unsigned int val, unsigned int shift) -{ - put_unaligned_be64((u64) val << shift, buf); -} - -static void regmap_format_64_le(void *buf, unsigned int val, unsigned int shift) -{ - put_unaligned_le64((u64) val << shift, buf); -} - -static void regmap_format_64_native(void *buf, unsigned int val, - unsigned int shift) -{ - u64 v = (u64) val << shift; - - memcpy(buf, &v, sizeof(v)); -} -#endif - static void regmap_parse_inplace_noop(void *buf) { } @@ -380,14 +354,9 @@ static unsigned int regmap_parse_16_native(const void *buf) return v; } -static unsigned int regmap_parse_24(const void *buf) +static unsigned int regmap_parse_24_be(const void *buf) { - const u8 *b = buf; - unsigned int ret = b[2]; - ret |= ((unsigned int)b[1]) << 8; - ret |= ((unsigned int)b[0]) << 16; - - return ret; + return get_unaligned_be24(buf); } static unsigned int regmap_parse_32_be(const void *buf) @@ -422,40 +391,6 @@ static unsigned int regmap_parse_32_native(const void *buf) return v; } -#ifdef CONFIG_64BIT -static unsigned int regmap_parse_64_be(const void *buf) -{ - return get_unaligned_be64(buf); -} - -static unsigned int regmap_parse_64_le(const void *buf) -{ - return get_unaligned_le64(buf); -} - -static void regmap_parse_64_be_inplace(void *buf) -{ - u64 v = get_unaligned_be64(buf); - - memcpy(buf, &v, sizeof(v)); -} - -static void regmap_parse_64_le_inplace(void *buf) -{ - u64 v = get_unaligned_le64(buf); - - memcpy(buf, &v, sizeof(v)); -} - -static unsigned int regmap_parse_64_native(const void *buf) -{ - u64 v; - - memcpy(&v, buf, sizeof(v)); - return v; -} -#endif - static void regmap_lock_hwlock(void *__map) { struct regmap *map = __map; @@ -533,6 +468,23 @@ __releases(&map->spinlock) spin_unlock_irqrestore(&map->spinlock, map->spinlock_flags); } +static void regmap_lock_raw_spinlock(void *__map) +__acquires(&map->raw_spinlock) +{ + struct regmap *map = __map; + unsigned long flags; + + raw_spin_lock_irqsave(&map->raw_spinlock, flags); + map->raw_spinlock_flags = flags; +} + +static void regmap_unlock_raw_spinlock(void *__map) +__releases(&map->raw_spinlock) +{ + struct regmap *map = __map; + raw_spin_unlock_irqrestore(&map->raw_spinlock, map->raw_spinlock_flags); +} + static void dev_get_regmap_release(struct device *dev, void *res) { /* @@ -630,6 +582,7 @@ int regmap_attach_dev(struct device *dev, struct regmap *map, if (ret) return ret; + regmap_debugfs_exit(map); regmap_debugfs_init(map); /* Add a devres resource for dev_get_regmap() */ @@ -645,6 +598,17 @@ int regmap_attach_dev(struct device *dev, struct regmap *map, } EXPORT_SYMBOL_GPL(regmap_attach_dev); +static int dev_get_regmap_match(struct device *dev, void *res, void *data); + +static int regmap_detach_dev(struct device *dev, struct regmap *map) +{ + if (!dev) + return 0; + + return devres_release(dev, dev_get_regmap_release, + dev_get_regmap_match, (void *)map->name); +} + static enum regmap_endian regmap_get_reg_endian(const struct regmap_bus *bus, const struct regmap_config *config) { @@ -770,11 +734,19 @@ struct regmap *__regmap_init(struct device *dev, } else { if ((bus && bus->fast_io) || config->fast_io) { - spin_lock_init(&map->spinlock); - map->lock = regmap_lock_spinlock; - map->unlock = regmap_unlock_spinlock; - lockdep_set_class_and_name(&map->spinlock, - lock_key, lock_name); + if (config->use_raw_spinlock) { + raw_spin_lock_init(&map->raw_spinlock); + map->lock = regmap_lock_raw_spinlock; + map->unlock = regmap_unlock_raw_spinlock; + lockdep_set_class_and_name(&map->raw_spinlock, + lock_key, lock_name); + } else { + spin_lock_init(&map->spinlock); + map->lock = regmap_lock_spinlock; + map->unlock = regmap_unlock_spinlock; + lockdep_set_class_and_name(&map->spinlock, + lock_key, lock_name); + } } else { mutex_init(&map->mutex); map->lock = regmap_lock_mutex; @@ -784,6 +756,7 @@ struct regmap *__regmap_init(struct device *dev, lock_key, lock_name); } map->lock_arg = map; + map->lock_key = lock_key; } /* @@ -795,12 +768,14 @@ struct regmap *__regmap_init(struct device *dev, else map->alloc_flags = GFP_KERNEL; - map->format.reg_bytes = DIV_ROUND_UP(config->reg_bits, 8); - map->format.pad_bytes = config->pad_bits / 8; - map->format.val_bytes = DIV_ROUND_UP(config->val_bits, 8); - map->format.buf_size = DIV_ROUND_UP(config->reg_bits + - config->val_bits + config->pad_bits, 8); + map->reg_base = config->reg_base; map->reg_shift = config->pad_bits % 8; + + map->format.pad_bytes = config->pad_bits / 8; + map->format.reg_shift = config->reg_shift; + map->format.reg_bytes = BITS_TO_BYTES(config->reg_bits); + map->format.val_bytes = BITS_TO_BYTES(config->val_bits); + map->format.buf_size = BITS_TO_BYTES(config->reg_bits + config->val_bits + config->pad_bits); if (config->reg_stride) map->reg_stride = config->reg_stride; else @@ -809,17 +784,21 @@ struct regmap *__regmap_init(struct device *dev, map->reg_stride_order = ilog2(map->reg_stride); else map->reg_stride_order = -1; - map->use_single_read = config->use_single_read || !bus || !bus->read; - map->use_single_write = config->use_single_write || !bus || !bus->write; - map->can_multi_write = config->can_multi_write && bus && bus->write; + map->use_single_read = config->use_single_read || !(config->read || (bus && bus->read)); + map->use_single_write = config->use_single_write || !(config->write || (bus && bus->write)); + map->can_multi_write = config->can_multi_write && (config->write || (bus && bus->write)); if (bus) { map->max_raw_read = bus->max_raw_read; map->max_raw_write = bus->max_raw_write; + } else if (config->max_raw_read && config->max_raw_write) { + map->max_raw_read = config->max_raw_read; + map->max_raw_write = config->max_raw_write; } map->dev = dev; map->bus = bus; map->bus_context = bus_context; map->max_register = config->max_register; + map->max_register_is_set = map->max_register ?: config->max_register_is_0; map->wr_table = config->wr_table; map->rd_table = config->rd_table; map->volatile_table = config->volatile_table; @@ -848,9 +827,21 @@ struct regmap *__regmap_init(struct device *dev, map->read_flag_mask = bus->read_flag_mask; } - if (!bus) { + if (config->read && config->write) { + map->reg_read = _regmap_bus_read; + if (config->reg_update_bits) + map->reg_update_bits = config->reg_update_bits; + + /* Bulk read/write */ + map->read = config->read; + map->write = config->write; + + reg_endian = REGMAP_ENDIAN_NATIVE; + val_endian = REGMAP_ENDIAN_NATIVE; + } else if (!bus) { map->reg_read = config->reg_read; map->reg_write = config->reg_write; + map->reg_update_bits = config->reg_update_bits; map->defer_caching = false; goto skip_format_initialization; @@ -864,10 +855,13 @@ struct regmap *__regmap_init(struct device *dev, } else { map->reg_read = _regmap_bus_read; map->reg_update_bits = bus->reg_update_bits; - } + /* Bulk read/write */ + map->read = bus->read; + map->write = bus->write; - reg_endian = regmap_get_reg_endian(bus, config); - val_endian = regmap_get_val_endian(dev, bus, config); + reg_endian = regmap_get_reg_endian(bus, config); + val_endian = regmap_get_val_endian(dev, bus, config); + } switch (config->reg_bits + map->reg_shift) { case 2: @@ -944,44 +938,30 @@ struct regmap *__regmap_init(struct device *dev, break; case 24: - if (reg_endian != REGMAP_ENDIAN_BIG) - goto err_hwlock; - map->format.format_reg = regmap_format_24; - break; - - case 32: switch (reg_endian) { case REGMAP_ENDIAN_BIG: - map->format.format_reg = regmap_format_32_be; - break; - case REGMAP_ENDIAN_LITTLE: - map->format.format_reg = regmap_format_32_le; - break; - case REGMAP_ENDIAN_NATIVE: - map->format.format_reg = regmap_format_32_native; + map->format.format_reg = regmap_format_24_be; break; default: goto err_hwlock; } break; -#ifdef CONFIG_64BIT - case 64: + case 32: switch (reg_endian) { case REGMAP_ENDIAN_BIG: - map->format.format_reg = regmap_format_64_be; + map->format.format_reg = regmap_format_32_be; break; case REGMAP_ENDIAN_LITTLE: - map->format.format_reg = regmap_format_64_le; + map->format.format_reg = regmap_format_32_le; break; case REGMAP_ENDIAN_NATIVE: - map->format.format_reg = regmap_format_64_native; + map->format.format_reg = regmap_format_32_native; break; default: goto err_hwlock; } break; -#endif default: goto err_hwlock; @@ -1017,10 +997,14 @@ struct regmap *__regmap_init(struct device *dev, } break; case 24: - if (val_endian != REGMAP_ENDIAN_BIG) + switch (val_endian) { + case REGMAP_ENDIAN_BIG: + map->format.format_val = regmap_format_24_be; + map->format.parse_val = regmap_parse_24_be; + break; + default: goto err_hwlock; - map->format.format_val = regmap_format_24; - map->format.parse_val = regmap_parse_24; + } break; case 32: switch (val_endian) { @@ -1042,28 +1026,6 @@ struct regmap *__regmap_init(struct device *dev, goto err_hwlock; } break; -#ifdef CONFIG_64BIT - case 64: - switch (val_endian) { - case REGMAP_ENDIAN_BIG: - map->format.format_val = regmap_format_64_be; - map->format.parse_val = regmap_parse_64_be; - map->format.parse_inplace = regmap_parse_64_be_inplace; - break; - case REGMAP_ENDIAN_LITTLE: - map->format.format_val = regmap_format_64_le; - map->format.parse_val = regmap_parse_64_le; - map->format.parse_inplace = regmap_parse_64_le_inplace; - break; - case REGMAP_ENDIAN_NATIVE: - map->format.format_val = regmap_format_64_native; - map->format.parse_val = regmap_parse_64_native; - break; - default: - goto err_hwlock; - } - break; -#endif } if (map->format.format_write) { @@ -1100,13 +1062,13 @@ skip_format_initialization: /* Sanity check */ if (range_cfg->range_max < range_cfg->range_min) { - dev_err(map->dev, "Invalid range %d: %d < %d\n", i, + dev_err(map->dev, "Invalid range %d: %u < %u\n", i, range_cfg->range_max, range_cfg->range_min); goto err_range; } if (range_cfg->range_max > map->max_register) { - dev_err(map->dev, "Invalid range %d: %d > %d\n", i, + dev_err(map->dev, "Invalid range %d: %u > %u\n", i, range_cfg->range_max, map->max_register); goto err_range; } @@ -1126,10 +1088,10 @@ skip_format_initialization: /* Make sure, that this register range has no selector or data window within its boundary */ for (j = 0; j < config->num_ranges; j++) { - unsigned sel_reg = config->ranges[j].selector_reg; - unsigned win_min = config->ranges[j].window_start; - unsigned win_max = win_min + - config->ranges[j].window_len - 1; + unsigned int sel_reg = config->ranges[j].selector_reg; + unsigned int win_min = config->ranges[j].window_start; + unsigned int win_max = win_min + + config->ranges[j].window_len - 1; /* Allow data window inside its own virtual range */ if (j == i) @@ -1211,6 +1173,8 @@ err_name: err_map: kfree(map); err: + if (bus && bus->free_on_exit) + kfree(bus); return ERR_PTR(ret); } EXPORT_SYMBOL_GPL(__regmap_init); @@ -1253,6 +1217,9 @@ static void regmap_field_init(struct regmap_field *rm_field, rm_field->reg = reg_field.reg; rm_field->shift = reg_field.lsb; rm_field->mask = GENMASK(reg_field.msb, reg_field.lsb); + + WARN_ONCE(rm_field->mask == 0, "invalid empty mask defined\n"); + rm_field->id_size = reg_field.id_size; rm_field->id_offset = reg_field.id_offset; } @@ -1298,7 +1265,7 @@ EXPORT_SYMBOL_GPL(devm_regmap_field_alloc); */ int regmap_field_bulk_alloc(struct regmap *regmap, struct regmap_field **rm_field, - struct reg_field *reg_field, + const struct reg_field *reg_field, int num_fields) { struct regmap_field *rf; @@ -1334,7 +1301,7 @@ EXPORT_SYMBOL_GPL(regmap_field_bulk_alloc); int devm_regmap_field_bulk_alloc(struct device *dev, struct regmap *regmap, struct regmap_field **rm_field, - struct reg_field *reg_field, + const struct reg_field *reg_field, int num_fields) { struct regmap_field *rf; @@ -1459,6 +1426,7 @@ int regmap_reinit_cache(struct regmap *map, const struct regmap_config *config) regmap_debugfs_exit(map); map->max_register = config->max_register; + map->max_register_is_set = map->max_register ?: config->max_register_is_0; map->writeable_reg = config->writeable_reg; map->readable_reg = config->readable_reg; map->volatile_reg = config->volatile_reg; @@ -1489,7 +1457,9 @@ void regmap_exit(struct regmap *map) { struct regmap_async *async; + regmap_detach_dev(map->dev, map); regcache_exit(map); + regmap_debugfs_exit(map); regmap_range_exit(map); if (map->bus && map->bus->free_context) @@ -1525,7 +1495,7 @@ static int dev_get_regmap_match(struct device *dev, void *res, void *data) /* If the user didn't specify a name match any */ if (data) - return !strcmp((*r)->name, data); + return (*r)->name && !strcmp((*r)->name, data); else return 1; } @@ -1629,6 +1599,18 @@ static void regmap_set_work_buf_flag_mask(struct regmap *map, int max_bytes, buf[i] |= (mask >> (8 * i)) & 0xff; } +static unsigned int regmap_reg_addr(struct regmap *map, unsigned int reg) +{ + reg += map->reg_base; + + if (map->format.reg_shift > 0) + reg >>= map->format.reg_shift; + else if (map->format.reg_shift < 0) + reg <<= -(map->format.reg_shift); + + return reg; +} + static int _regmap_raw_write_impl(struct regmap *map, unsigned int reg, const void *val, size_t val_len, bool noinc) { @@ -1641,8 +1623,6 @@ static int _regmap_raw_write_impl(struct regmap *map, unsigned int reg, size_t len; int i; - WARN_ON(!map->bus); - /* Check for unwritable or noinc registers in range * before we start */ @@ -1657,17 +1637,19 @@ static int _regmap_raw_write_impl(struct regmap *map, unsigned int reg, } if (!map->cache_bypass && map->format.parse_val) { - unsigned int ival; + unsigned int ival, offset; int val_bytes = map->format.val_bytes; - for (i = 0; i < val_len / val_bytes; i++) { - ival = map->format.parse_val(val + (i * val_bytes)); - ret = regcache_write(map, - reg + regmap_get_offset(map, i), - ival); + + /* Cache the last written value for noinc writes */ + i = noinc ? val_len - val_bytes : 0; + for (; i < val_len; i += val_bytes) { + ival = map->format.parse_val(val + i); + offset = noinc ? 0 : regmap_get_offset(map, i / val_bytes); + ret = regcache_write(map, reg + offset, ival); if (ret) { dev_err(map->dev, "Error in caching of register: %x ret: %d\n", - reg + i, ret); + reg + offset, ret); return ret; } } @@ -1708,6 +1690,7 @@ static int _regmap_raw_write_impl(struct regmap *map, unsigned int reg, return ret; } + reg = regmap_reg_addr(map, reg); map->format.format_reg(map->work_buf, reg, map->reg_shift); regmap_set_work_buf_flag_mask(map, map->format.reg_bytes, map->write_flag_mask); @@ -1722,7 +1705,7 @@ static int _regmap_raw_write_impl(struct regmap *map, unsigned int reg, val = work_val; } - if (map->async && map->bus->async_write) { + if (map->async && map->bus && map->bus->async_write) { struct regmap_async *async; trace_regmap_async_write_start(map, reg, val_len); @@ -1790,11 +1773,11 @@ static int _regmap_raw_write_impl(struct regmap *map, unsigned int reg, * write. */ if (val == work_val) - ret = map->bus->write(map->bus_context, map->work_buf, - map->format.reg_bytes + - map->format.pad_bytes + - val_len); - else if (map->bus->gather_write) + ret = map->write(map->bus_context, map->work_buf, + map->format.reg_bytes + + map->format.pad_bytes + + val_len); + else if (map->bus && map->bus->gather_write) ret = map->bus->gather_write(map->bus_context, map->work_buf, map->format.reg_bytes + map->format.pad_bytes, @@ -1812,7 +1795,7 @@ static int _regmap_raw_write_impl(struct regmap *map, unsigned int reg, memcpy(buf, map->work_buf, map->format.reg_bytes); memcpy(buf + map->format.reg_bytes + map->format.pad_bytes, val, val_len); - ret = map->bus->write(map->bus_context, buf, len); + ret = map->write(map->bus_context, buf, len); kfree(buf); } else if (ret != 0 && !map->cache_bypass && map->format.parse_val) { @@ -1835,8 +1818,7 @@ static int _regmap_raw_write_impl(struct regmap *map, unsigned int reg, */ bool regmap_can_raw_write(struct regmap *map) { - return map->bus && map->bus->write && map->format.format_val && - map->format.format_reg; + return map->write && map->format.format_val && map->format.format_reg; } EXPORT_SYMBOL_GPL(regmap_can_raw_write); @@ -1869,7 +1851,7 @@ static int _regmap_bus_formatted_write(void *context, unsigned int reg, struct regmap_range_node *range; struct regmap *map = context; - WARN_ON(!map->bus || !map->format.format_write); + WARN_ON(!map->format.format_write); range = _regmap_range_lookup(map, reg); if (range) { @@ -1878,12 +1860,12 @@ static int _regmap_bus_formatted_write(void *context, unsigned int reg, return ret; } + reg = regmap_reg_addr(map, reg); map->format.format_write(map, reg, val); trace_regmap_hw_write_start(map, reg, 1); - ret = map->bus->write(map->bus_context, map->work_buf, - map->format.buf_size); + ret = map->write(map->bus_context, map->work_buf, map->format.buf_size); trace_regmap_hw_write_done(map, reg, 1); @@ -1894,7 +1876,17 @@ static int _regmap_bus_reg_write(void *context, unsigned int reg, unsigned int val) { struct regmap *map = context; + struct regmap_range_node *range; + int ret; + range = _regmap_range_lookup(map, reg); + if (range) { + ret = _regmap_select_page(map, ®, range, 1); + if (ret != 0) + return ret; + } + + reg = regmap_reg_addr(map, reg); return map->bus->reg_write(map->bus_context, reg, val); } @@ -1903,7 +1895,7 @@ static int _regmap_bus_raw_write(void *context, unsigned int reg, { struct regmap *map = context; - WARN_ON(!map->bus || !map->format.format_val); + WARN_ON(!map->format.format_val); map->format.format_val(map->work_buf + map->format.reg_bytes + map->format.pad_bytes, val, 0); @@ -1917,7 +1909,7 @@ static int _regmap_bus_raw_write(void *context, unsigned int reg, static inline void *_regmap_map_get_context(struct regmap *map) { - return (map->bus) ? map : map->bus_context; + return (map->bus || (!map->bus && map->read)) ? map : map->bus_context; } int _regmap_write(struct regmap *map, unsigned int reg, @@ -2082,8 +2074,86 @@ int regmap_raw_write(struct regmap *map, unsigned int reg, } EXPORT_SYMBOL_GPL(regmap_raw_write); +static int regmap_noinc_readwrite(struct regmap *map, unsigned int reg, + void *val, unsigned int val_len, bool write) +{ + size_t val_bytes = map->format.val_bytes; + size_t val_count = val_len / val_bytes; + unsigned int lastval; + u8 *u8p; + u16 *u16p; + u32 *u32p; + int ret; + int i; + + switch (val_bytes) { + case 1: + u8p = val; + if (write) + lastval = (unsigned int)u8p[val_count - 1]; + break; + case 2: + u16p = val; + if (write) + lastval = (unsigned int)u16p[val_count - 1]; + break; + case 4: + u32p = val; + if (write) + lastval = (unsigned int)u32p[val_count - 1]; + break; + default: + return -EINVAL; + } + + /* + * Update the cache with the last value we write, the rest is just + * gone down in the hardware FIFO. We can't cache FIFOs. This makes + * sure a single read from the cache will work. + */ + if (write) { + if (!map->cache_bypass && !map->defer_caching) { + ret = regcache_write(map, reg, lastval); + if (ret != 0) + return ret; + if (map->cache_only) { + map->cache_dirty = true; + return 0; + } + } + ret = map->bus->reg_noinc_write(map->bus_context, reg, val, val_count); + } else { + ret = map->bus->reg_noinc_read(map->bus_context, reg, val, val_count); + } + + if (!ret && regmap_should_log(map)) { + dev_info(map->dev, "%x %s [", reg, write ? "<=" : "=>"); + for (i = 0; i < val_count; i++) { + switch (val_bytes) { + case 1: + pr_cont("%x", u8p[i]); + break; + case 2: + pr_cont("%x", u16p[i]); + break; + case 4: + pr_cont("%x", u32p[i]); + break; + default: + break; + } + if (i == (val_count - 1)) + pr_cont("]\n"); + else + pr_cont(","); + } + } + + return 0; +} + /** - * regmap_noinc_write(): Write data from a register without incrementing the + * regmap_noinc_write(): Write data to a register without incrementing the * register number * * @map: Register map to write to @@ -2109,10 +2179,8 @@ int regmap_noinc_write(struct regmap *map, unsigned int reg, size_t write_len; int ret; - if (!map->bus) + if (!map->write && !(map->bus && map->bus->reg_noinc_write)) return -EINVAL; - if (!map->bus->write) - return -ENOTSUPP; if (val_len % map->format.val_bytes) return -EINVAL; if (!IS_ALIGNED(reg, map->reg_stride)) @@ -2127,6 +2195,15 @@ int regmap_noinc_write(struct regmap *map, unsigned int reg, goto out_unlock; } + /* + * Use the accelerated operation if we can. The val drops the const + * typing in order to facilitate code reuse in regmap_noinc_readwrite(). + */ + if (map->bus->reg_noinc_write) { + ret = regmap_noinc_readwrite(map, reg, (void *)val, val_len, true); + goto out_unlock; + } + while (val_len) { if (map->max_raw_write && map->max_raw_write < val_len) write_len = map->max_raw_write; @@ -2175,6 +2252,30 @@ int regmap_field_update_bits_base(struct regmap_field *field, EXPORT_SYMBOL_GPL(regmap_field_update_bits_base); /** + * regmap_field_test_bits() - Check if all specified bits are set in a + * register field. + * + * @field: Register field to operate on + * @bits: Bits to test + * + * Returns negative errno if the underlying regmap_field_read() fails, + * 0 if at least one of the tested bits is not set and 1 if all tested + * bits are set. + */ +int regmap_field_test_bits(struct regmap_field *field, unsigned int bits) +{ + unsigned int val; + int ret; + + ret = regmap_field_read(field, &val); + if (ret) + return ret; + + return (val & bits) == bits; +} +EXPORT_SYMBOL_GPL(regmap_field_test_bits); + +/** * regmap_fields_update_bits_base() - Perform a read/modify/write cycle a * register field with port ID * @@ -2232,7 +2333,7 @@ int regmap_bulk_write(struct regmap *map, unsigned int reg, const void *val, * Some devices don't support bulk write, for them we have a series of * single write operations. */ - if (!map->bus || !map->format.parse_inplace) { + if (!map->write || !map->format.parse_inplace) { map->lock(map->lock_arg); for (i = 0; i < val_count; i++) { unsigned int ival; @@ -2247,11 +2348,6 @@ int regmap_bulk_write(struct regmap *map, unsigned int reg, const void *val, case 4: ival = *(u32 *)(val + (i * val_bytes)); break; -#ifdef CONFIG_64BIT - case 8: - ival = *(u64 *)(val + (i * val_bytes)); - break; -#endif default: ret = -EINVAL; goto out; @@ -2268,7 +2364,7 @@ out: } else { void *wval; - wval = kmemdup(val, val_count * val_bytes, map->alloc_flags); + wval = kmemdup_array(val, val_count, val_bytes, map->alloc_flags); if (!wval) return -ENOMEM; @@ -2279,6 +2375,10 @@ out: kfree(wval); } + + if (!ret) + trace_regmap_bulk_write(map, reg, val, val_bytes * val_count); + return ret; } EXPORT_SYMBOL_GPL(regmap_bulk_write); @@ -2319,6 +2419,7 @@ static int _regmap_raw_multi_reg_write(struct regmap *map, unsigned int reg = regs[i].reg; unsigned int val = regs[i].def; trace_regmap_hw_write_start(map, reg, 1); + reg = regmap_reg_addr(map, reg); map->format.format_reg(u8, reg, map->reg_shift); u8 += reg_bytes + pad_bytes; map->format.format_val(u8, val, 0); @@ -2327,7 +2428,7 @@ static int _regmap_raw_multi_reg_write(struct regmap *map, u8 = buf; *u8 |= map->write_flag_mask; - ret = map->bus->write(map->bus_context, buf, len); + ret = map->write(map->bus_context, buf, len); kfree(buf); @@ -2633,9 +2734,7 @@ static int _regmap_raw_read(struct regmap *map, unsigned int reg, void *val, struct regmap_range_node *range; int ret; - WARN_ON(!map->bus); - - if (!map->bus || !map->bus->read) + if (!map->read) return -EINVAL; range = _regmap_range_lookup(map, reg); @@ -2646,14 +2745,15 @@ static int _regmap_raw_read(struct regmap *map, unsigned int reg, void *val, return ret; } + reg = regmap_reg_addr(map, reg); map->format.format_reg(map->work_buf, reg, map->reg_shift); regmap_set_work_buf_flag_mask(map, map->format.reg_bytes, map->read_flag_mask); trace_regmap_hw_read_start(map, reg, val_len / map->format.val_bytes); - ret = map->bus->read(map->bus_context, map->work_buf, - map->format.reg_bytes + map->format.pad_bytes, - val, val_len); + ret = map->read(map->bus_context, map->work_buf, + map->format.reg_bytes + map->format.pad_bytes, + val, val_len); trace_regmap_hw_read_done(map, reg, val_len / map->format.val_bytes); @@ -2664,7 +2764,17 @@ static int _regmap_bus_reg_read(void *context, unsigned int reg, unsigned int *val) { struct regmap *map = context; + struct regmap_range_node *range; + int ret; + range = _regmap_range_lookup(map, reg); + if (range) { + ret = _regmap_select_page(map, ®, range, 1); + if (ret != 0) + return ret; + } + + reg = regmap_reg_addr(map, reg); return map->bus->reg_read(map->bus_context, reg, val); } @@ -2746,6 +2856,43 @@ int regmap_read(struct regmap *map, unsigned int reg, unsigned int *val) EXPORT_SYMBOL_GPL(regmap_read); /** + * regmap_read_bypassed() - Read a value from a single register direct + * from the device, bypassing the cache + * + * @map: Register map to read from + * @reg: Register to be read from + * @val: Pointer to store read value + * + * A value of zero will be returned on success, a negative errno will + * be returned in error cases. + */ +int regmap_read_bypassed(struct regmap *map, unsigned int reg, unsigned int *val) +{ + int ret; + bool bypass, cache_only; + + if (!IS_ALIGNED(reg, map->reg_stride)) + return -EINVAL; + + map->lock(map->lock_arg); + + bypass = map->cache_bypass; + cache_only = map->cache_only; + map->cache_bypass = true; + map->cache_only = false; + + ret = _regmap_read(map, reg, val); + + map->cache_bypass = bypass; + map->cache_only = cache_only; + + map->unlock(map->lock_arg); + + return ret; +} +EXPORT_SYMBOL_GPL(regmap_read_bypassed); + +/** * regmap_raw_read() - Read raw data from the device * * @map: Register map to read from @@ -2764,8 +2911,6 @@ int regmap_raw_read(struct regmap *map, unsigned int reg, void *val, unsigned int v; int ret, i; - if (!map->bus) - return -EINVAL; if (val_len % map->format.val_bytes) return -EINVAL; if (!IS_ALIGNED(reg, map->reg_stride)) @@ -2780,7 +2925,12 @@ int regmap_raw_read(struct regmap *map, unsigned int reg, void *val, size_t chunk_count, chunk_bytes; size_t chunk_regs = val_count; - if (!map->bus->read) { + if (!map->cache_bypass && map->cache_only) { + ret = -EBUSY; + goto out; + } + + if (!map->read) { ret = -ENOTSUPP; goto out; } @@ -2840,7 +2990,7 @@ EXPORT_SYMBOL_GPL(regmap_raw_read); * @val: Pointer to data buffer * @val_len: Length of output buffer in bytes. * - * The regmap API usually assumes that bulk bus read operations will read a + * The regmap API usually assumes that bulk read operations will read a * range of registers. Some devices have certain registers for which a read * operation read will read from an internal FIFO. * @@ -2858,10 +3008,9 @@ int regmap_noinc_read(struct regmap *map, unsigned int reg, size_t read_len; int ret; - if (!map->bus) - return -EINVAL; - if (!map->bus->read) + if (!map->read) return -ENOTSUPP; + if (val_len % map->format.val_bytes) return -EINVAL; if (!IS_ALIGNED(reg, map->reg_stride)) @@ -2876,6 +3025,23 @@ int regmap_noinc_read(struct regmap *map, unsigned int reg, goto out_unlock; } + /* + * We have not defined the FIFO semantics for cache, as the + * cache is just one value deep. Should we return the last + * written value? Just avoid this by always reading the FIFO + * even when using cache. Cache only will not work. + */ + if (!map->cache_bypass && map->cache_only) { + ret = -EBUSY; + goto out_unlock; + } + + /* Use the accelerated operation if we can */ + if (map->bus->reg_noinc_read) { + ret = regmap_noinc_readwrite(map, reg, val, val_len, false); + goto out_unlock; + } + while (val_len) { if (map->max_raw_read && map->max_raw_read < val_len) read_len = map->max_raw_read; @@ -2952,8 +3118,53 @@ int regmap_fields_read(struct regmap_field *field, unsigned int id, } EXPORT_SYMBOL_GPL(regmap_fields_read); +static int _regmap_bulk_read(struct regmap *map, unsigned int reg, + const unsigned int *regs, void *val, size_t val_count) +{ + u32 *u32 = val; + u16 *u16 = val; + u8 *u8 = val; + int ret, i; + + map->lock(map->lock_arg); + + for (i = 0; i < val_count; i++) { + unsigned int ival; + + if (regs) { + if (!IS_ALIGNED(regs[i], map->reg_stride)) { + ret = -EINVAL; + goto out; + } + ret = _regmap_read(map, regs[i], &ival); + } else { + ret = _regmap_read(map, reg + regmap_get_offset(map, i), &ival); + } + if (ret != 0) + goto out; + + switch (map->format.val_bytes) { + case 4: + u32[i] = ival; + break; + case 2: + u16[i] = ival; + break; + case 1: + u8[i] = ival; + break; + default: + ret = -EINVAL; + goto out; + } + } +out: + map->unlock(map->lock_arg); + return ret; +} + /** - * regmap_bulk_read() - Read multiple registers from the device + * regmap_bulk_read() - Read multiple sequential registers from the device * * @map: Register map to read from * @reg: First register to be read from @@ -2975,7 +3186,7 @@ int regmap_bulk_read(struct regmap *map, unsigned int reg, void *val, if (val_count == 0) return -EINVAL; - if (map->bus && map->format.parse_inplace && (vol || map->cache_type == REGCACHE_NONE)) { + if (map->read && map->format.parse_inplace && (vol || map->cache_type == REGCACHE_NONE)) { ret = regmap_raw_read(map, reg, val, val_bytes * val_count); if (ret != 0) return ret; @@ -2983,52 +3194,35 @@ int regmap_bulk_read(struct regmap *map, unsigned int reg, void *val, for (i = 0; i < val_count * val_bytes; i += val_bytes) map->format.parse_inplace(val + i); } else { -#ifdef CONFIG_64BIT - u64 *u64 = val; -#endif - u32 *u32 = val; - u16 *u16 = val; - u8 *u8 = val; - - map->lock(map->lock_arg); - - for (i = 0; i < val_count; i++) { - unsigned int ival; - - ret = _regmap_read(map, reg + regmap_get_offset(map, i), - &ival); - if (ret != 0) - goto out; - - switch (map->format.val_bytes) { -#ifdef CONFIG_64BIT - case 8: - u64[i] = ival; - break; -#endif - case 4: - u32[i] = ival; - break; - case 2: - u16[i] = ival; - break; - case 1: - u8[i] = ival; - break; - default: - ret = -EINVAL; - goto out; - } - } - -out: - map->unlock(map->lock_arg); + ret = _regmap_bulk_read(map, reg, NULL, val, val_count); } - + if (!ret) + trace_regmap_bulk_read(map, reg, val, val_bytes * val_count); return ret; } EXPORT_SYMBOL_GPL(regmap_bulk_read); +/** + * regmap_multi_reg_read() - Read multiple non-sequential registers from the device + * + * @map: Register map to read from + * @regs: Array of registers to read from + * @val: Pointer to store read value, in native register size for device + * @val_count: Number of registers to read + * + * A value of zero will be returned on success, a negative errno will + * be returned in error cases. + */ +int regmap_multi_reg_read(struct regmap *map, const unsigned int *regs, void *val, + size_t val_count) +{ + if (val_count == 0) + return -EINVAL; + + return _regmap_bulk_read(map, 0, regs, val, val_count); +} +EXPORT_SYMBOL_GPL(regmap_multi_reg_read); + static int _regmap_update_bits(struct regmap *map, unsigned int reg, unsigned int mask, unsigned int val, bool *change, bool force_write) @@ -3040,6 +3234,7 @@ static int _regmap_update_bits(struct regmap *map, unsigned int reg, *change = false; if (regmap_volatile(map, reg) && map->reg_update_bits) { + reg = regmap_reg_addr(map, reg); ret = map->reg_update_bits(map->bus_context, reg, mask, val); if (ret == 0 && change) *change = true; @@ -3051,7 +3246,7 @@ static int _regmap_update_bits(struct regmap *map, unsigned int reg, tmp = orig & ~mask; tmp |= val & mask; - if (force_write || (tmp != orig)) { + if (force_write || (tmp != orig) || map->force_write_field) { ret = _regmap_write(map, reg, tmp); if (ret == 0 && change) *change = true; @@ -3116,7 +3311,8 @@ EXPORT_SYMBOL_GPL(regmap_update_bits_base); */ int regmap_test_bits(struct regmap *map, unsigned int reg, unsigned int bits) { - unsigned int val, ret; + unsigned int val; + int ret; ret = regmap_read(map, reg, &val); if (ret) @@ -3277,7 +3473,7 @@ EXPORT_SYMBOL_GPL(regmap_get_val_bytes); */ int regmap_get_max_register(struct regmap *map) { - return map->max_register ? map->max_register : -EINVAL; + return map->max_register_is_set ? map->max_register : -EINVAL; } EXPORT_SYMBOL_GPL(regmap_get_max_register); @@ -3295,6 +3491,19 @@ int regmap_get_reg_stride(struct regmap *map) } EXPORT_SYMBOL_GPL(regmap_get_reg_stride); +/** + * regmap_might_sleep() - Returns whether a regmap access might sleep. + * + * @map: Register map to operate on. + * + * Returns true if an access to the register might sleep, else false. + */ +bool regmap_might_sleep(struct regmap *map) +{ + return map->can_sleep; +} +EXPORT_SYMBOL_GPL(regmap_might_sleep); + int regmap_parse_val(struct regmap *map, const void *buf, unsigned int *val) { diff --git a/drivers/base/regmap/trace.h b/drivers/base/regmap/trace.h index 9abee14df9ee..bcc5a8b226a6 100644 --- a/drivers/base/regmap/trace.h +++ b/drivers/base/regmap/trace.h @@ -27,14 +27,12 @@ DECLARE_EVENT_CLASS(regmap_reg, ), TP_fast_assign( - __assign_str(name, regmap_name(map)); + __assign_str(name); __entry->reg = reg; __entry->val = val; ), - TP_printk("%s reg=%x val=%x", __get_str(name), - (unsigned int)__entry->reg, - (unsigned int)__entry->val) + TP_printk("%s reg=%x val=%x", __get_str(name), __entry->reg, __entry->val) ); DEFINE_EVENT(regmap_reg, regmap_reg_write, @@ -43,7 +41,6 @@ DEFINE_EVENT(regmap_reg, regmap_reg_write, unsigned int val), TP_ARGS(map, reg, val) - ); DEFINE_EVENT(regmap_reg, regmap_reg_read, @@ -52,7 +49,6 @@ DEFINE_EVENT(regmap_reg, regmap_reg_read, unsigned int val), TP_ARGS(map, reg, val) - ); DEFINE_EVENT(regmap_reg, regmap_reg_read_cache, @@ -61,7 +57,47 @@ DEFINE_EVENT(regmap_reg, regmap_reg_read_cache, unsigned int val), TP_ARGS(map, reg, val) +); + +DECLARE_EVENT_CLASS(regmap_bulk, + TP_PROTO(struct regmap *map, unsigned int reg, + const void *val, int val_len), + + TP_ARGS(map, reg, val, val_len), + + TP_STRUCT__entry( + __string(name, regmap_name(map)) + __field(unsigned int, reg) + __dynamic_array(char, buf, val_len) + __field(int, val_len) + ), + + TP_fast_assign( + __assign_str(name); + __entry->reg = reg; + __entry->val_len = val_len; + memcpy(__get_dynamic_array(buf), val, val_len); + ), + + TP_printk("%s reg=%x val=%s", __get_str(name), __entry->reg, + __print_hex(__get_dynamic_array(buf), __entry->val_len)) +); + +DEFINE_EVENT(regmap_bulk, regmap_bulk_write, + + TP_PROTO(struct regmap *map, unsigned int reg, + const void *val, int val_len), + + TP_ARGS(map, reg, val, val_len) +); + +DEFINE_EVENT(regmap_bulk, regmap_bulk_read, + + TP_PROTO(struct regmap *map, unsigned int reg, + const void *val, int val_len), + + TP_ARGS(map, reg, val, val_len) ); DECLARE_EVENT_CLASS(regmap_block, @@ -77,14 +113,12 @@ DECLARE_EVENT_CLASS(regmap_block, ), TP_fast_assign( - __assign_str(name, regmap_name(map)); + __assign_str(name); __entry->reg = reg; __entry->count = count; ), - TP_printk("%s reg=%x count=%d", __get_str(name), - (unsigned int)__entry->reg, - (int)__entry->count) + TP_printk("%s reg=%x count=%d", __get_str(name), __entry->reg, __entry->count) ); DEFINE_EVENT(regmap_block, regmap_hw_read_start, @@ -129,9 +163,9 @@ TRACE_EVENT(regcache_sync, ), TP_fast_assign( - __assign_str(name, regmap_name(map)); - __assign_str(status, status); - __assign_str(type, type); + __assign_str(name); + __assign_str(status); + __assign_str(type); ), TP_printk("%s type=%s status=%s", __get_str(name), @@ -150,12 +184,11 @@ DECLARE_EVENT_CLASS(regmap_bool, ), TP_fast_assign( - __assign_str(name, regmap_name(map)); + __assign_str(name); __entry->flag = flag; ), - TP_printk("%s flag=%d", __get_str(name), - (int)__entry->flag) + TP_printk("%s flag=%d", __get_str(name), __entry->flag) ); DEFINE_EVENT(regmap_bool, regmap_cache_only, @@ -163,7 +196,6 @@ DEFINE_EVENT(regmap_bool, regmap_cache_only, TP_PROTO(struct regmap *map, bool flag), TP_ARGS(map, flag) - ); DEFINE_EVENT(regmap_bool, regmap_cache_bypass, @@ -171,7 +203,6 @@ DEFINE_EVENT(regmap_bool, regmap_cache_bypass, TP_PROTO(struct regmap *map, bool flag), TP_ARGS(map, flag) - ); DECLARE_EVENT_CLASS(regmap_async, @@ -185,7 +216,7 @@ DECLARE_EVENT_CLASS(regmap_async, ), TP_fast_assign( - __assign_str(name, regmap_name(map)); + __assign_str(name); ), TP_printk("%s", __get_str(name)) @@ -203,7 +234,6 @@ DEFINE_EVENT(regmap_async, regmap_async_io_complete, TP_PROTO(struct regmap *map), TP_ARGS(map) - ); DEFINE_EVENT(regmap_async, regmap_async_complete_start, @@ -211,7 +241,6 @@ DEFINE_EVENT(regmap_async, regmap_async_complete_start, TP_PROTO(struct regmap *map), TP_ARGS(map) - ); DEFINE_EVENT(regmap_async, regmap_async_complete_done, @@ -219,7 +248,6 @@ DEFINE_EVENT(regmap_async, regmap_async_complete_done, TP_PROTO(struct regmap *map), TP_ARGS(map) - ); TRACE_EVENT(regcache_drop_region, @@ -236,13 +264,12 @@ TRACE_EVENT(regcache_drop_region, ), TP_fast_assign( - __assign_str(name, regmap_name(map)); + __assign_str(name); __entry->from = from; __entry->to = to; ), - TP_printk("%s %u-%u", __get_str(name), (unsigned int)__entry->from, - (unsigned int)__entry->to) + TP_printk("%s %u-%u", __get_str(name), __entry->from, __entry->to) ); #endif /* _TRACE_REGMAP_H */ diff --git a/drivers/base/soc.c b/drivers/base/soc.c index 0af5363a582c..282c38aece0d 100644 --- a/drivers/base/soc.c +++ b/drivers/base/soc.c @@ -7,6 +7,7 @@ #include <linux/sysfs.h> #include <linux/init.h> +#include <linux/of.h> #include <linux/stat.h> #include <linux/slab.h> #include <linux/idr.h> @@ -27,9 +28,10 @@ struct soc_device { int soc_dev_num; }; -static struct bus_type soc_bus_type = { +static const struct bus_type soc_bus_type = { .name = "soc", }; +static bool soc_bus_registered; static DEVICE_ATTR(machine, 0444, soc_info_show, NULL); static DEVICE_ATTR(family, 0444, soc_info_show, NULL); @@ -104,11 +106,23 @@ static void soc_release(struct device *dev) { struct soc_device *soc_dev = container_of(dev, struct soc_device, dev); - ida_simple_remove(&soc_ida, soc_dev->soc_dev_num); + ida_free(&soc_ida, soc_dev->soc_dev_num); kfree(soc_dev->dev.groups); kfree(soc_dev); } +static void soc_device_get_machine(struct soc_device_attribute *soc_dev_attr) +{ + struct device_node *np; + + if (soc_dev_attr->machine) + return; + + np = of_find_node_by_path("/"); + of_property_read_string(np, "model", &soc_dev_attr->machine); + of_node_put(np); +} + static struct soc_device_attribute *early_soc_dev_attr; struct soc_device *soc_device_register(struct soc_device_attribute *soc_dev_attr) @@ -117,7 +131,9 @@ struct soc_device *soc_device_register(struct soc_device_attribute *soc_dev_attr const struct attribute_group **soc_attr_groups; int ret; - if (!soc_bus_type.p) { + soc_device_get_machine(soc_dev_attr); + + if (!soc_bus_registered) { if (early_soc_dev_attr) return ERR_PTR(-EBUSY); early_soc_dev_attr = soc_dev_attr; @@ -139,7 +155,7 @@ struct soc_device *soc_device_register(struct soc_device_attribute *soc_dev_attr soc_attr_groups[1] = soc_dev_attr->custom_attr_group; /* Fetch a unique (reclaimable) SOC ID. */ - ret = ida_simple_get(&soc_ida, 0, 0, GFP_KERNEL); + ret = ida_alloc(&soc_ida, GFP_KERNEL); if (ret < 0) goto out3; soc_dev->soc_dev_num = ret; @@ -183,6 +199,7 @@ static int __init soc_bus_register(void) ret = bus_register(&soc_bus_type); if (ret) return ret; + soc_bus_registered = true; if (early_soc_dev_attr) return PTR_ERR(soc_device_register(early_soc_dev_attr)); @@ -241,15 +258,13 @@ static int soc_device_match_one(struct device *dev, void *arg) const struct soc_device_attribute *soc_device_match( const struct soc_device_attribute *matches) { - int ret = 0; + int ret; if (!matches) return NULL; - while (!ret) { - if (!(matches->machine || matches->family || - matches->revision || matches->soc_id)) - break; + while (matches->machine || matches->family || matches->revision || + matches->soc_id) { ret = bus_for_each_dev(&soc_bus_type, NULL, (void *)matches, soc_device_match_one); if (ret < 0 && early_soc_dev_attr) @@ -257,10 +272,10 @@ const struct soc_device_attribute *soc_device_match( matches); if (ret < 0) return NULL; - if (!ret) - matches++; - else + if (ret) return matches; + + matches++; } return NULL; } diff --git a/drivers/base/swnode.c b/drivers/base/swnode.c index d1f1a8240120..16a8301c25d6 100644 --- a/drivers/base/swnode.c +++ b/drivers/base/swnode.c @@ -6,10 +6,23 @@ * Author: Heikki Krogerus <heikki.krogerus@linux.intel.com> */ +#include <linux/container_of.h> #include <linux/device.h> -#include <linux/kernel.h> +#include <linux/err.h> +#include <linux/export.h> +#include <linux/idr.h> +#include <linux/init.h> +#include <linux/kobject.h> +#include <linux/kstrtox.h> +#include <linux/list.h> #include <linux/property.h> #include <linux/slab.h> +#include <linux/spinlock.h> +#include <linux/string.h> +#include <linux/sysfs.h> +#include <linux/types.h> + +#include "base.h" struct swnode { struct kobject kobj; @@ -411,9 +424,6 @@ software_node_get_name(const struct fwnode_handle *fwnode) { const struct swnode *swnode = to_swnode(fwnode); - if (!swnode) - return "(null)"; - return kobject_name(&swnode->kobj); } @@ -505,9 +515,6 @@ software_node_get_reference_args(const struct fwnode_handle *fwnode, int error; int i; - if (!swnode) - return -ENOENT; - prop = property_entry_get(swnode->node->properties, propname); if (!prop) return -ENOENT; @@ -522,20 +529,35 @@ software_node_get_reference_args(const struct fwnode_handle *fwnode, if (prop->is_inline) return -EINVAL; - if (index * sizeof(*ref) >= prop->length) + if ((index + 1) * sizeof(*ref) > prop->length) return -ENOENT; ref_array = prop->pointer; ref = &ref_array[index]; - refnode = software_node_fwnode(ref->node); + /* + * A software node can reference other software nodes or firmware + * nodes (which are the abstraction layer sitting on top of them). + * This is done to ensure we can create references to static software + * nodes before they're registered with the firmware node framework. + * At the time the reference is being resolved, we expect the swnodes + * in question to already have been registered and to be backed by + * a firmware node. This is why we use the fwnode API below to read the + * relevant properties and bump the reference count. + */ + + if (ref->swnode) + refnode = software_node_fwnode(ref->swnode); + else if (ref->fwnode) + refnode = ref->fwnode; + else + return -EINVAL; + if (!refnode) return -ENOENT; if (nargs_prop) { - error = property_entry_read_int_array(swnode->node->properties, - nargs_prop, sizeof(u32), - &nargs_prop_val, 1); + error = fwnode_property_read_u32(refnode, nargs_prop, &nargs_prop_val); if (error) return error; @@ -545,7 +567,10 @@ software_node_get_reference_args(const struct fwnode_handle *fwnode, if (nargs > NR_FWNODE_REFERENCE_ARGS) return -EINVAL; - args->fwnode = software_node_get(refnode); + if (!args) + return 0; + + args->fwnode = fwnode_handle_get(refnode); args->nargs = nargs; for (i = 0; i < nargs; i++) @@ -625,7 +650,10 @@ software_node_graph_get_remote_endpoint(const struct fwnode_handle *fwnode) ref = prop->pointer; - return software_node_get(software_node_fwnode(ref[0].node)); + if (!ref->swnode) + return NULL; + + return software_node_get(software_node_fwnode(ref->swnode)); } static struct fwnode_handle * @@ -667,6 +695,7 @@ static const struct fwnode_operations software_node_ops = { .get = software_node_get, .put = software_node_put, .property_present = software_node_property_present, + .property_read_bool = software_node_property_present, .property_read_int_array = software_node_read_int_array, .property_read_string_array = software_node_read_string_array, .get_name = software_node_get_name, @@ -751,10 +780,10 @@ static void software_node_release(struct kobject *kobj) struct swnode *swnode = kobj_to_swnode(kobj); if (swnode->parent) { - ida_simple_remove(&swnode->parent->child_ids, swnode->id); + ida_free(&swnode->parent->child_ids, swnode->id); list_del(&swnode->entry); } else { - ida_simple_remove(&swnode_root_ids, swnode->id); + ida_free(&swnode_root_ids, swnode->id); } if (swnode->allocated) @@ -764,7 +793,7 @@ static void software_node_release(struct kobject *kobj) kfree(swnode); } -static struct kobj_type software_node_type = { +static const struct kobj_type software_node_type = { .release = software_node_release, .sysfs_ops = &kobj_sysfs_ops, }; @@ -780,8 +809,8 @@ swnode_register(const struct software_node *node, struct swnode *parent, if (!swnode) return ERR_PTR(-ENOMEM); - ret = ida_simple_get(parent ? &parent->child_ids : &swnode_root_ids, - 0, 0, GFP_KERNEL); + ret = ida_alloc(parent ? &parent->child_ids : &swnode_root_ids, + GFP_KERNEL); if (ret < 0) { kfree(swnode); return ERR_PTR(ret); @@ -824,67 +853,6 @@ swnode_register(const struct software_node *node, struct swnode *parent, } /** - * software_node_register_nodes - Register an array of software nodes - * @nodes: Zero terminated array of software nodes to be registered - * - * Register multiple software nodes at once. If any node in the array - * has its .parent pointer set (which can only be to another software_node), - * then its parent **must** have been registered before it is; either outside - * of this function or by ordering the array such that parent comes before - * child. - */ -int software_node_register_nodes(const struct software_node *nodes) -{ - int ret; - int i; - - for (i = 0; nodes[i].name; i++) { - const struct software_node *parent = nodes[i].parent; - - if (parent && !software_node_to_swnode(parent)) { - ret = -EINVAL; - goto err_unregister_nodes; - } - - ret = software_node_register(&nodes[i]); - if (ret) - goto err_unregister_nodes; - } - - return 0; - -err_unregister_nodes: - software_node_unregister_nodes(nodes); - return ret; -} -EXPORT_SYMBOL_GPL(software_node_register_nodes); - -/** - * software_node_unregister_nodes - Unregister an array of software nodes - * @nodes: Zero terminated array of software nodes to be unregistered - * - * Unregister multiple software nodes at once. If parent pointers are set up - * in any of the software nodes then the array **must** be ordered such that - * parents come before their children. - * - * NOTE: If you are uncertain whether the array is ordered such that - * parents will be unregistered before their children, it is wiser to - * remove the nodes individually, in the correct order (child before - * parent). - */ -void software_node_unregister_nodes(const struct software_node *nodes) -{ - unsigned int i = 0; - - while (nodes[i].name) - i++; - - while (i--) - software_node_unregister(&nodes[i]); -} -EXPORT_SYMBOL_GPL(software_node_unregister_nodes); - -/** * software_node_register_node_group - Register a group of software nodes * @node_group: NULL terminated array of software node pointers to be registered * @@ -894,7 +862,7 @@ EXPORT_SYMBOL_GPL(software_node_unregister_nodes); * of this function or by ordering the array such that parent comes before * child. */ -int software_node_register_node_group(const struct software_node **node_group) +int software_node_register_node_group(const struct software_node * const *node_group) { unsigned int i; int ret; @@ -927,8 +895,7 @@ EXPORT_SYMBOL_GPL(software_node_register_node_group); * remove the nodes individually, in the correct order (child before * parent). */ -void software_node_unregister_node_group( - const struct software_node **node_group) +void software_node_unregister_node_group(const struct software_node * const *node_group) { unsigned int i = 0; @@ -1053,7 +1020,7 @@ int device_add_software_node(struct device *dev, const struct software_node *nod * balance. */ if (device_is_registered(dev)) - software_node_notify(dev, KOBJ_ADD); + software_node_notify(dev); return 0; } @@ -1074,7 +1041,8 @@ void device_remove_software_node(struct device *dev) return; if (device_is_registered(dev)) - software_node_notify(dev, KOBJ_REMOVE); + software_node_notify_remove(dev); + set_secondary_fwnode(dev, NULL); kobject_put(&swnode->kobj); } @@ -1113,48 +1081,50 @@ int device_create_managed_software_node(struct device *dev, to_swnode(fwnode)->managed = true; set_secondary_fwnode(dev, fwnode); + if (device_is_registered(dev)) + software_node_notify(dev); + return 0; } EXPORT_SYMBOL_GPL(device_create_managed_software_node); -int software_node_notify(struct device *dev, unsigned long action) +void software_node_notify(struct device *dev) { struct swnode *swnode; int ret; swnode = dev_to_swnode(dev); if (!swnode) - return 0; + return; - switch (action) { - case KOBJ_ADD: - ret = sysfs_create_link(&dev->kobj, &swnode->kobj, "software_node"); - if (ret) - break; + kobject_get(&swnode->kobj); + ret = sysfs_create_link(&dev->kobj, &swnode->kobj, "software_node"); + if (ret) + return; - ret = sysfs_create_link(&swnode->kobj, &dev->kobj, - dev_name(dev)); - if (ret) { - sysfs_remove_link(&dev->kobj, "software_node"); - break; - } - kobject_get(&swnode->kobj); - break; - case KOBJ_REMOVE: - sysfs_remove_link(&swnode->kobj, dev_name(dev)); + ret = sysfs_create_link(&swnode->kobj, &dev->kobj, dev_name(dev)); + if (ret) { sysfs_remove_link(&dev->kobj, "software_node"); - kobject_put(&swnode->kobj); - - if (swnode->managed) { - set_secondary_fwnode(dev, NULL); - kobject_put(&swnode->kobj); - } - break; - default: - break; + return; } +} - return 0; +void software_node_notify_remove(struct device *dev) +{ + struct swnode *swnode; + + swnode = dev_to_swnode(dev); + if (!swnode) + return; + + sysfs_remove_link(&swnode->kobj, dev_name(dev)); + sysfs_remove_link(&dev->kobj, "software_node"); + kobject_put(&swnode->kobj); + + if (swnode->managed) { + set_secondary_fwnode(dev, NULL); + kobject_put(&swnode->kobj); + } } static int __init software_node_init(void) diff --git a/drivers/base/syscore.c b/drivers/base/syscore.c index 13db1f78d2ce..483adb796654 100644 --- a/drivers/base/syscore.c +++ b/drivers/base/syscore.c @@ -11,32 +11,32 @@ #include <linux/suspend.h> #include <trace/events/power.h> -static LIST_HEAD(syscore_ops_list); -static DEFINE_MUTEX(syscore_ops_lock); +static LIST_HEAD(syscore_list); +static DEFINE_MUTEX(syscore_lock); /** - * register_syscore_ops - Register a set of system core operations. - * @ops: System core operations to register. + * register_syscore - Register a set of system core operations. + * @syscore: System core operations to register. */ -void register_syscore_ops(struct syscore_ops *ops) +void register_syscore(struct syscore *syscore) { - mutex_lock(&syscore_ops_lock); - list_add_tail(&ops->node, &syscore_ops_list); - mutex_unlock(&syscore_ops_lock); + mutex_lock(&syscore_lock); + list_add_tail(&syscore->node, &syscore_list); + mutex_unlock(&syscore_lock); } -EXPORT_SYMBOL_GPL(register_syscore_ops); +EXPORT_SYMBOL_GPL(register_syscore); /** - * unregister_syscore_ops - Unregister a set of system core operations. - * @ops: System core operations to unregister. + * unregister_syscore - Unregister a set of system core operations. + * @syscore: System core operations to unregister. */ -void unregister_syscore_ops(struct syscore_ops *ops) +void unregister_syscore(struct syscore *syscore) { - mutex_lock(&syscore_ops_lock); - list_del(&ops->node); - mutex_unlock(&syscore_ops_lock); + mutex_lock(&syscore_lock); + list_del(&syscore->node); + mutex_unlock(&syscore_lock); } -EXPORT_SYMBOL_GPL(unregister_syscore_ops); +EXPORT_SYMBOL_GPL(unregister_syscore); #ifdef CONFIG_PM_SLEEP /** @@ -46,7 +46,7 @@ EXPORT_SYMBOL_GPL(unregister_syscore_ops); */ int syscore_suspend(void) { - struct syscore_ops *ops; + struct syscore *syscore; int ret = 0; trace_suspend_resume(TPS("syscore_suspend"), 0, true); @@ -59,25 +59,27 @@ int syscore_suspend(void) WARN_ONCE(!irqs_disabled(), "Interrupts enabled before system core suspend.\n"); - list_for_each_entry_reverse(ops, &syscore_ops_list, node) - if (ops->suspend) { - pm_pr_dbg("Calling %pS\n", ops->suspend); - ret = ops->suspend(); + list_for_each_entry_reverse(syscore, &syscore_list, node) + if (syscore->ops->suspend) { + pm_pr_dbg("Calling %pS\n", syscore->ops->suspend); + ret = syscore->ops->suspend(syscore->data); if (ret) goto err_out; WARN_ONCE(!irqs_disabled(), - "Interrupts enabled after %pS\n", ops->suspend); + "Interrupts enabled after %pS\n", + syscore->ops->suspend); } trace_suspend_resume(TPS("syscore_suspend"), 0, false); return 0; err_out: - pr_err("PM: System core suspend callback %pS failed.\n", ops->suspend); + pr_err("PM: System core suspend callback %pS failed.\n", + syscore->ops->suspend); - list_for_each_entry_continue(ops, &syscore_ops_list, node) - if (ops->resume) - ops->resume(); + list_for_each_entry_continue(syscore, &syscore_list, node) + if (syscore->ops->resume) + syscore->ops->resume(syscore->data); return ret; } @@ -90,18 +92,19 @@ EXPORT_SYMBOL_GPL(syscore_suspend); */ void syscore_resume(void) { - struct syscore_ops *ops; + struct syscore *syscore; trace_suspend_resume(TPS("syscore_resume"), 0, true); WARN_ONCE(!irqs_disabled(), "Interrupts enabled before system core resume.\n"); - list_for_each_entry(ops, &syscore_ops_list, node) - if (ops->resume) { - pm_pr_dbg("Calling %pS\n", ops->resume); - ops->resume(); + list_for_each_entry(syscore, &syscore_list, node) + if (syscore->ops->resume) { + pm_pr_dbg("Calling %pS\n", syscore->ops->resume); + syscore->ops->resume(syscore->data); WARN_ONCE(!irqs_disabled(), - "Interrupts enabled after %pS\n", ops->resume); + "Interrupts enabled after %pS\n", + syscore->ops->resume); } trace_suspend_resume(TPS("syscore_resume"), 0, false); } @@ -113,16 +116,17 @@ EXPORT_SYMBOL_GPL(syscore_resume); */ void syscore_shutdown(void) { - struct syscore_ops *ops; + struct syscore *syscore; - mutex_lock(&syscore_ops_lock); + mutex_lock(&syscore_lock); - list_for_each_entry_reverse(ops, &syscore_ops_list, node) - if (ops->shutdown) { + list_for_each_entry_reverse(syscore, &syscore_list, node) + if (syscore->ops->shutdown) { if (initcall_debug) - pr_info("PM: Calling %pS\n", ops->shutdown); - ops->shutdown(); + pr_info("PM: Calling %pS\n", + syscore->ops->shutdown); + syscore->ops->shutdown(syscore->data); } - mutex_unlock(&syscore_ops_lock); + mutex_unlock(&syscore_lock); } diff --git a/drivers/base/test/.kunitconfig b/drivers/base/test/.kunitconfig new file mode 100644 index 000000000000..473923f0998b --- /dev/null +++ b/drivers/base/test/.kunitconfig @@ -0,0 +1,2 @@ +CONFIG_KUNIT=y +CONFIG_DM_KUNIT_TEST=y diff --git a/drivers/base/test/Kconfig b/drivers/base/test/Kconfig index 2f3fa31a948e..2756870615cc 100644 --- a/drivers/base/test/Kconfig +++ b/drivers/base/test/Kconfig @@ -8,7 +8,13 @@ config TEST_ASYNC_DRIVER_PROBE The module name will be test_async_driver_probe.ko If unsure say N. + +config DM_KUNIT_TEST + tristate "KUnit Tests for the device model" if !KUNIT_ALL_TESTS + depends on KUNIT + default KUNIT_ALL_TESTS + config DRIVER_PE_KUNIT_TEST - bool "KUnit Tests for property entry API" if !KUNIT_ALL_TESTS - depends on KUNIT=y + tristate "KUnit Tests for property entry API" if !KUNIT_ALL_TESTS + depends on KUNIT default KUNIT_ALL_TESTS diff --git a/drivers/base/test/Makefile b/drivers/base/test/Makefile index 64b2f3d744d5..e321dfc7e922 100644 --- a/drivers/base/test/Makefile +++ b/drivers/base/test/Makefile @@ -1,5 +1,8 @@ # SPDX-License-Identifier: GPL-2.0 obj-$(CONFIG_TEST_ASYNC_DRIVER_PROBE) += test_async_driver_probe.o +obj-$(CONFIG_DM_KUNIT_TEST) += root-device-test.o +obj-$(CONFIG_DM_KUNIT_TEST) += platform-device-test.o + obj-$(CONFIG_DRIVER_PE_KUNIT_TEST) += property-entry-test.o -CFLAGS_REMOVE_property-entry-test.o += -fplugin-arg-structleak_plugin-byref -fplugin-arg-structleak_plugin-byref-all +CFLAGS_property-entry-test.o += $(DISABLE_STRUCTLEAK_PLUGIN) diff --git a/drivers/base/test/platform-device-test.c b/drivers/base/test/platform-device-test.c new file mode 100644 index 000000000000..6355a2231b74 --- /dev/null +++ b/drivers/base/test/platform-device-test.c @@ -0,0 +1,263 @@ +// SPDX-License-Identifier: GPL-2.0 + +#include <kunit/platform_device.h> +#include <kunit/resource.h> + +#include <linux/device.h> +#include <linux/device/bus.h> +#include <linux/of_platform.h> +#include <linux/platform_device.h> + +#define DEVICE_NAME "test" + +struct test_priv { + bool probe_done; + bool release_done; + wait_queue_head_t probe_wq; + wait_queue_head_t release_wq; + struct device *dev; +}; + +static int platform_device_devm_init(struct kunit *test) +{ + struct test_priv *priv; + + priv = kunit_kzalloc(test, sizeof(*priv), GFP_KERNEL); + KUNIT_ASSERT_NOT_ERR_OR_NULL(test, priv); + init_waitqueue_head(&priv->probe_wq); + init_waitqueue_head(&priv->release_wq); + + test->priv = priv; + + return 0; +} + +static void devm_device_action(void *ptr) +{ + struct test_priv *priv = ptr; + + priv->release_done = true; + wake_up_interruptible(&priv->release_wq); +} + +static void devm_put_device_action(void *ptr) +{ + struct test_priv *priv = ptr; + + put_device(priv->dev); + priv->release_done = true; + wake_up_interruptible(&priv->release_wq); +} + +#define RELEASE_TIMEOUT_MS 100 + +/* + * Tests that a platform bus, non-probed device will run its + * device-managed actions when unregistered. + */ +static void platform_device_devm_register_unregister_test(struct kunit *test) +{ + struct platform_device *pdev; + struct test_priv *priv = test->priv; + int ret; + + pdev = platform_device_alloc(DEVICE_NAME, PLATFORM_DEVID_NONE); + KUNIT_ASSERT_NOT_ERR_OR_NULL(test, pdev); + + ret = platform_device_add(pdev); + KUNIT_ASSERT_EQ(test, ret, 0); + + priv->dev = &pdev->dev; + + ret = devm_add_action_or_reset(priv->dev, devm_device_action, priv); + KUNIT_ASSERT_EQ(test, ret, 0); + + platform_device_unregister(pdev); + + ret = wait_event_interruptible_timeout(priv->release_wq, priv->release_done, + msecs_to_jiffies(RELEASE_TIMEOUT_MS)); + KUNIT_EXPECT_GT(test, ret, 0); +} + +/* + * Tests that a platform bus, non-probed device will run its + * device-managed actions when unregistered, even if someone still holds + * a reference to it. + */ +static void platform_device_devm_register_get_unregister_with_devm_test(struct kunit *test) +{ + struct platform_device *pdev; + struct test_priv *priv = test->priv; + int ret; + + pdev = platform_device_alloc(DEVICE_NAME, PLATFORM_DEVID_NONE); + KUNIT_ASSERT_NOT_ERR_OR_NULL(test, pdev); + + ret = platform_device_add(pdev); + KUNIT_ASSERT_EQ(test, ret, 0); + + priv->dev = &pdev->dev; + + get_device(priv->dev); + + ret = devm_add_action_or_reset(priv->dev, devm_put_device_action, priv); + KUNIT_ASSERT_EQ(test, ret, 0); + + platform_device_unregister(pdev); + + ret = wait_event_interruptible_timeout(priv->release_wq, priv->release_done, + msecs_to_jiffies(RELEASE_TIMEOUT_MS)); + KUNIT_EXPECT_GT(test, ret, 0); +} + +static int fake_probe(struct platform_device *pdev) +{ + struct test_priv *priv = platform_get_drvdata(pdev); + + priv->probe_done = true; + wake_up_interruptible(&priv->probe_wq); + + return 0; +} + +static struct platform_driver fake_driver = { + .probe = fake_probe, + .driver = { + .name = DEVICE_NAME, + }, +}; + +/* + * Tests that a platform bus, probed device will run its device-managed + * actions when unregistered. + */ +static void probed_platform_device_devm_register_unregister_test(struct kunit *test) +{ + struct platform_device *pdev; + struct test_priv *priv = test->priv; + int ret; + + ret = platform_driver_register(&fake_driver); + KUNIT_ASSERT_EQ(test, ret, 0); + + pdev = platform_device_alloc(DEVICE_NAME, PLATFORM_DEVID_NONE); + KUNIT_ASSERT_NOT_ERR_OR_NULL(test, pdev); + + priv->dev = &pdev->dev; + platform_set_drvdata(pdev, priv); + + ret = platform_device_add(pdev); + KUNIT_ASSERT_EQ(test, ret, 0); + + ret = wait_event_interruptible_timeout(priv->probe_wq, priv->probe_done, + msecs_to_jiffies(RELEASE_TIMEOUT_MS)); + KUNIT_ASSERT_GT(test, ret, 0); + + ret = devm_add_action_or_reset(priv->dev, devm_device_action, priv); + KUNIT_ASSERT_EQ(test, ret, 0); + + platform_device_unregister(pdev); + + ret = wait_event_interruptible_timeout(priv->release_wq, priv->release_done, + msecs_to_jiffies(RELEASE_TIMEOUT_MS)); + KUNIT_EXPECT_GT(test, ret, 0); + + platform_driver_unregister(&fake_driver); +} + +/* + * Tests that a platform bus, probed device will run its device-managed + * actions when unregistered, even if someone still holds a reference to + * it. + */ +static void probed_platform_device_devm_register_get_unregister_with_devm_test(struct kunit *test) +{ + struct platform_device *pdev; + struct test_priv *priv = test->priv; + int ret; + + ret = platform_driver_register(&fake_driver); + KUNIT_ASSERT_EQ(test, ret, 0); + + pdev = platform_device_alloc(DEVICE_NAME, PLATFORM_DEVID_NONE); + KUNIT_ASSERT_NOT_ERR_OR_NULL(test, pdev); + + priv->dev = &pdev->dev; + platform_set_drvdata(pdev, priv); + + ret = platform_device_add(pdev); + KUNIT_ASSERT_EQ(test, ret, 0); + + ret = wait_event_interruptible_timeout(priv->probe_wq, priv->probe_done, + msecs_to_jiffies(RELEASE_TIMEOUT_MS)); + KUNIT_ASSERT_GT(test, ret, 0); + + get_device(priv->dev); + + ret = devm_add_action_or_reset(priv->dev, devm_put_device_action, priv); + KUNIT_ASSERT_EQ(test, ret, 0); + + platform_device_unregister(pdev); + + ret = wait_event_interruptible_timeout(priv->release_wq, priv->release_done, + msecs_to_jiffies(RELEASE_TIMEOUT_MS)); + KUNIT_EXPECT_GT(test, ret, 0); + + platform_driver_unregister(&fake_driver); +} + +static struct kunit_case platform_device_devm_tests[] = { + KUNIT_CASE(platform_device_devm_register_unregister_test), + KUNIT_CASE(platform_device_devm_register_get_unregister_with_devm_test), + KUNIT_CASE(probed_platform_device_devm_register_unregister_test), + KUNIT_CASE(probed_platform_device_devm_register_get_unregister_with_devm_test), + {} +}; + +static struct kunit_suite platform_device_devm_test_suite = { + .name = "platform-device-devm", + .init = platform_device_devm_init, + .test_cases = platform_device_devm_tests, +}; + +static void platform_device_find_by_null_test(struct kunit *test) +{ + struct platform_device *pdev; + int ret; + + pdev = kunit_platform_device_alloc(test, DEVICE_NAME, PLATFORM_DEVID_NONE); + KUNIT_ASSERT_NOT_ERR_OR_NULL(test, pdev); + + ret = kunit_platform_device_add(test, pdev); + KUNIT_ASSERT_EQ(test, ret, 0); + + KUNIT_EXPECT_PTR_EQ(test, of_find_device_by_node(NULL), NULL); + + KUNIT_EXPECT_PTR_EQ(test, bus_find_device_by_of_node(&platform_bus_type, NULL), NULL); + KUNIT_EXPECT_PTR_EQ(test, bus_find_device_by_fwnode(&platform_bus_type, NULL), NULL); + KUNIT_EXPECT_PTR_EQ(test, bus_find_device_by_acpi_dev(&platform_bus_type, NULL), NULL); + + KUNIT_EXPECT_FALSE(test, device_match_of_node(&pdev->dev, NULL)); + KUNIT_EXPECT_FALSE(test, device_match_fwnode(&pdev->dev, NULL)); + KUNIT_EXPECT_FALSE(test, device_match_acpi_dev(&pdev->dev, NULL)); + KUNIT_EXPECT_FALSE(test, device_match_acpi_handle(&pdev->dev, NULL)); +} + +static struct kunit_case platform_device_match_tests[] = { + KUNIT_CASE(platform_device_find_by_null_test), + {} +}; + +static struct kunit_suite platform_device_match_test_suite = { + .name = "platform-device-match", + .test_cases = platform_device_match_tests, +}; + +kunit_test_suites( + &platform_device_devm_test_suite, + &platform_device_match_test_suite, +); + +MODULE_DESCRIPTION("Test module for platform devices"); +MODULE_AUTHOR("Maxime Ripard <mripard@kernel.org>"); +MODULE_LICENSE("GPL"); diff --git a/drivers/base/test/property-entry-test.c b/drivers/base/test/property-entry-test.c index 6071d5bc128c..a8657eb06f94 100644 --- a/drivers/base/test/property-entry-test.c +++ b/drivers/base/test/property-entry-test.c @@ -405,20 +405,18 @@ static void pe_test_move_inline_str(struct kunit *test) /* Handling of reference properties */ static void pe_test_reference(struct kunit *test) { - static const struct software_node nodes[] = { - { .name = "1", }, - { .name = "2", }, - { } - }; + static const struct software_node node1 = { .name = "1" }; + static const struct software_node node2 = { .name = "2" }; + static const struct software_node *group[] = { &node1, &node2, NULL }; static const struct software_node_ref_args refs[] = { - SOFTWARE_NODE_REFERENCE(&nodes[0]), - SOFTWARE_NODE_REFERENCE(&nodes[1], 3, 4), + SOFTWARE_NODE_REFERENCE(&node1), + SOFTWARE_NODE_REFERENCE(&node2, 3, 4), }; const struct property_entry entries[] = { - PROPERTY_ENTRY_REF("ref-1", &nodes[0]), - PROPERTY_ENTRY_REF("ref-2", &nodes[1], 1, 2), + PROPERTY_ENTRY_REF("ref-1", &node1), + PROPERTY_ENTRY_REF("ref-2", &node2, 1, 2), PROPERTY_ENTRY_REF_ARRAY("ref-3", refs), { } }; @@ -427,7 +425,7 @@ static void pe_test_reference(struct kunit *test) struct fwnode_reference_args ref; int error; - error = software_node_register_nodes(nodes); + error = software_node_register_node_group(group); KUNIT_ASSERT_EQ(test, error, 0); node = fwnode_create_software_node(entries, NULL); @@ -436,7 +434,7 @@ static void pe_test_reference(struct kunit *test) error = fwnode_property_get_reference_args(node, "ref-1", NULL, 0, 0, &ref); KUNIT_ASSERT_EQ(test, error, 0); - KUNIT_EXPECT_PTR_EQ(test, to_software_node(ref.fwnode), &nodes[0]); + KUNIT_EXPECT_PTR_EQ(test, to_software_node(ref.fwnode), &node1); KUNIT_EXPECT_EQ(test, ref.nargs, 0U); /* wrong index */ @@ -447,7 +445,7 @@ static void pe_test_reference(struct kunit *test) error = fwnode_property_get_reference_args(node, "ref-2", NULL, 1, 0, &ref); KUNIT_ASSERT_EQ(test, error, 0); - KUNIT_EXPECT_PTR_EQ(test, to_software_node(ref.fwnode), &nodes[1]); + KUNIT_EXPECT_PTR_EQ(test, to_software_node(ref.fwnode), &node2); KUNIT_EXPECT_EQ(test, ref.nargs, 1U); KUNIT_EXPECT_EQ(test, ref.args[0], 1LLU); @@ -455,7 +453,7 @@ static void pe_test_reference(struct kunit *test) error = fwnode_property_get_reference_args(node, "ref-2", NULL, 3, 0, &ref); KUNIT_ASSERT_EQ(test, error, 0); - KUNIT_EXPECT_PTR_EQ(test, to_software_node(ref.fwnode), &nodes[1]); + KUNIT_EXPECT_PTR_EQ(test, to_software_node(ref.fwnode), &node2); KUNIT_EXPECT_EQ(test, ref.nargs, 3U); KUNIT_EXPECT_EQ(test, ref.args[0], 1LLU); KUNIT_EXPECT_EQ(test, ref.args[1], 2LLU); @@ -470,14 +468,14 @@ static void pe_test_reference(struct kunit *test) error = fwnode_property_get_reference_args(node, "ref-3", NULL, 0, 0, &ref); KUNIT_ASSERT_EQ(test, error, 0); - KUNIT_EXPECT_PTR_EQ(test, to_software_node(ref.fwnode), &nodes[0]); + KUNIT_EXPECT_PTR_EQ(test, to_software_node(ref.fwnode), &node1); KUNIT_EXPECT_EQ(test, ref.nargs, 0U); /* second reference in the array */ error = fwnode_property_get_reference_args(node, "ref-3", NULL, 2, 1, &ref); KUNIT_ASSERT_EQ(test, error, 0); - KUNIT_EXPECT_PTR_EQ(test, to_software_node(ref.fwnode), &nodes[1]); + KUNIT_EXPECT_PTR_EQ(test, to_software_node(ref.fwnode), &node2); KUNIT_EXPECT_EQ(test, ref.nargs, 2U); KUNIT_EXPECT_EQ(test, ref.args[0], 3LLU); KUNIT_EXPECT_EQ(test, ref.args[1], 4LLU); @@ -488,7 +486,7 @@ static void pe_test_reference(struct kunit *test) KUNIT_EXPECT_NE(test, error, 0); fwnode_remove_software_node(node); - software_node_unregister_nodes(nodes); + software_node_unregister_node_group(group); } static struct kunit_case property_entry_test_cases[] = { @@ -508,3 +506,7 @@ static struct kunit_suite property_entry_test_suite = { }; kunit_test_suite(property_entry_test_suite); + +MODULE_DESCRIPTION("Test module for the property entry API"); +MODULE_AUTHOR("Dmitry Torokhov <dtor@chromium.org>"); +MODULE_LICENSE("GPL"); diff --git a/drivers/base/test/root-device-test.c b/drivers/base/test/root-device-test.c new file mode 100644 index 000000000000..9aea23c9123e --- /dev/null +++ b/drivers/base/test/root-device-test.c @@ -0,0 +1,112 @@ +// SPDX-License-Identifier: GPL-2.0 +// Copyright 2023 Maxime Ripard <mripard@kernel.org> + +#include <kunit/resource.h> + +#include <linux/device.h> + +#define DEVICE_NAME "test" + +struct test_priv { + bool probe_done; + bool release_done; + wait_queue_head_t release_wq; + struct device *dev; +}; + +static int root_device_devm_init(struct kunit *test) +{ + struct test_priv *priv; + + priv = kunit_kzalloc(test, sizeof(*priv), GFP_KERNEL); + KUNIT_ASSERT_NOT_ERR_OR_NULL(test, priv); + init_waitqueue_head(&priv->release_wq); + + test->priv = priv; + + return 0; +} + +static void devm_device_action(void *ptr) +{ + struct test_priv *priv = ptr; + + priv->release_done = true; + wake_up_interruptible(&priv->release_wq); +} + +#define RELEASE_TIMEOUT_MS 100 + +/* + * Tests that a bus-less, non-probed device will run its device-managed + * actions when unregistered. + */ +static void root_device_devm_register_unregister_test(struct kunit *test) +{ + struct test_priv *priv = test->priv; + int ret; + + priv->dev = root_device_register(DEVICE_NAME); + KUNIT_ASSERT_NOT_ERR_OR_NULL(test, priv->dev); + + ret = devm_add_action_or_reset(priv->dev, devm_device_action, priv); + KUNIT_ASSERT_EQ(test, ret, 0); + + root_device_unregister(priv->dev); + + ret = wait_event_interruptible_timeout(priv->release_wq, priv->release_done, + msecs_to_jiffies(RELEASE_TIMEOUT_MS)); + KUNIT_EXPECT_GT(test, ret, 0); +} + +static void devm_put_device_action(void *ptr) +{ + struct test_priv *priv = ptr; + + put_device(priv->dev); + priv->release_done = true; + wake_up_interruptible(&priv->release_wq); +} + +/* + * Tests that a bus-less, non-probed device will run its device-managed + * actions when unregistered, even if someone still holds a reference to + * it. + */ +static void root_device_devm_register_get_unregister_with_devm_test(struct kunit *test) +{ + struct test_priv *priv = test->priv; + int ret; + + priv->dev = root_device_register(DEVICE_NAME); + KUNIT_ASSERT_NOT_ERR_OR_NULL(test, priv->dev); + + get_device(priv->dev); + + ret = devm_add_action_or_reset(priv->dev, devm_put_device_action, priv); + KUNIT_ASSERT_EQ(test, ret, 0); + + root_device_unregister(priv->dev); + + ret = wait_event_interruptible_timeout(priv->release_wq, priv->release_done, + msecs_to_jiffies(RELEASE_TIMEOUT_MS)); + KUNIT_EXPECT_GT(test, ret, 0); +} + +static struct kunit_case root_device_devm_tests[] = { + KUNIT_CASE(root_device_devm_register_unregister_test), + KUNIT_CASE(root_device_devm_register_get_unregister_with_devm_test), + {} +}; + +static struct kunit_suite root_device_devm_test_suite = { + .name = "root-device-devm", + .init = root_device_devm_init, + .test_cases = root_device_devm_tests, +}; + +kunit_test_suite(root_device_devm_test_suite); + +MODULE_DESCRIPTION("Test module for root devices"); +MODULE_AUTHOR("Maxime Ripard <mripard@kernel.org>"); +MODULE_LICENSE("GPL"); diff --git a/drivers/base/test/test_async_driver_probe.c b/drivers/base/test/test_async_driver_probe.c index 3bb7beb127a9..3465800baa6c 100644 --- a/drivers/base/test/test_async_driver_probe.c +++ b/drivers/base/test/test_async_driver_probe.c @@ -84,7 +84,7 @@ test_platform_device_register_node(char *name, int id, int nid) pdev = platform_device_alloc(name, id); if (!pdev) - return NULL; + return ERR_PTR(-ENOMEM); if (nid != NUMA_NO_NODE) set_dev_node(&pdev->dev, nid); @@ -104,7 +104,7 @@ static int __init test_async_probe_init(void) struct platform_device **pdev = NULL; int async_id = 0, sync_id = 0; unsigned long long duration; - ktime_t calltime, delta; + ktime_t calltime; int err, nid, cpu; pr_info("registering first set of asynchronous devices...\n"); @@ -133,8 +133,7 @@ static int __init test_async_probe_init(void) goto err_unregister_async_devs; } - delta = ktime_sub(ktime_get(), calltime); - duration = (unsigned long long) ktime_to_ms(delta); + duration = (unsigned long long)ktime_ms_delta(ktime_get(), calltime); pr_info("registration took %lld msecs\n", duration); if (duration > TEST_PROBE_THRESHOLD) { pr_err("test failed: probe took too long\n"); @@ -146,7 +145,7 @@ static int __init test_async_probe_init(void) calltime = ktime_get(); for_each_online_cpu(cpu) { nid = cpu_to_node(cpu); - pdev = &sync_dev[sync_id]; + pdev = &async_dev[async_id]; *pdev = test_platform_device_register_node("test_async_driver", async_id, @@ -161,8 +160,7 @@ static int __init test_async_probe_init(void) async_id++; } - delta = ktime_sub(ktime_get(), calltime); - duration = (unsigned long long) ktime_to_ms(delta); + duration = (unsigned long long)ktime_ms_delta(ktime_get(), calltime); dev_info(&(*pdev)->dev, "registration took %lld msecs\n", duration); if (duration > TEST_PROBE_THRESHOLD) { @@ -197,8 +195,7 @@ static int __init test_async_probe_init(void) goto err_unregister_sync_devs; } - delta = ktime_sub(ktime_get(), calltime); - duration = (unsigned long long) ktime_to_ms(delta); + duration = (unsigned long long)ktime_ms_delta(ktime_get(), calltime); pr_info("registration took %lld msecs\n", duration); if (duration < TEST_PROBE_THRESHOLD) { dev_err(&(*pdev)->dev, @@ -223,8 +220,7 @@ static int __init test_async_probe_init(void) sync_id++; - delta = ktime_sub(ktime_get(), calltime); - duration = (unsigned long long) ktime_to_ms(delta); + duration = (unsigned long long)ktime_ms_delta(ktime_get(), calltime); dev_info(&(*pdev)->dev, "registration took %lld msecs\n", duration); if (duration < TEST_PROBE_THRESHOLD) { diff --git a/drivers/base/topology.c b/drivers/base/topology.c index 4d254fcc93d1..c890e2a5b428 100644 --- a/drivers/base/topology.c +++ b/drivers/base/topology.c @@ -14,106 +14,173 @@ #include <linux/hardirq.h> #include <linux/topology.h> -#define define_id_show_func(name) \ +#define define_id_show_func(name, fmt) \ static ssize_t name##_show(struct device *dev, \ struct device_attribute *attr, char *buf) \ { \ - return sysfs_emit(buf, "%d\n", topology_##name(dev->id)); \ + return sysfs_emit(buf, fmt "\n", topology_##name(dev->id)); \ } -#define define_siblings_show_map(name, mask) \ -static ssize_t name##_show(struct device *dev, \ - struct device_attribute *attr, char *buf) \ -{ \ - return cpumap_print_to_pagebuf(false, buf, topology_##mask(dev->id));\ +#define define_siblings_read_func(name, mask) \ +static ssize_t name##_read(struct file *file, struct kobject *kobj, \ + const struct bin_attribute *attr, char *buf, \ + loff_t off, size_t count) \ +{ \ + struct device *dev = kobj_to_dev(kobj); \ + cpumask_var_t mask; \ + ssize_t n; \ + \ + if (!alloc_cpumask_var(&mask, GFP_KERNEL)) \ + return -ENOMEM; \ + \ + cpumask_copy(mask, topology_##mask(dev->id)); \ + n = cpumap_print_bitmask_to_buf(buf, mask, off, count); \ + free_cpumask_var(mask); \ + \ + return n; \ +} \ + \ +static ssize_t name##_list_read(struct file *file, struct kobject *kobj, \ + const struct bin_attribute *attr, char *buf, \ + loff_t off, size_t count) \ +{ \ + struct device *dev = kobj_to_dev(kobj); \ + cpumask_var_t mask; \ + ssize_t n; \ + \ + if (!alloc_cpumask_var(&mask, GFP_KERNEL)) \ + return -ENOMEM; \ + \ + cpumask_copy(mask, topology_##mask(dev->id)); \ + n = cpumap_print_list_to_buf(buf, mask, off, count); \ + free_cpumask_var(mask); \ + \ + return n; \ } -#define define_siblings_show_list(name, mask) \ -static ssize_t name##_list_show(struct device *dev, \ - struct device_attribute *attr, \ - char *buf) \ -{ \ - return cpumap_print_to_pagebuf(true, buf, topology_##mask(dev->id));\ -} - -#define define_siblings_show_func(name, mask) \ - define_siblings_show_map(name, mask); \ - define_siblings_show_list(name, mask) - -define_id_show_func(physical_package_id); +define_id_show_func(physical_package_id, "%d"); static DEVICE_ATTR_RO(physical_package_id); -define_id_show_func(die_id); +#ifdef TOPOLOGY_DIE_SYSFS +define_id_show_func(die_id, "%d"); static DEVICE_ATTR_RO(die_id); +#endif + +#ifdef TOPOLOGY_CLUSTER_SYSFS +define_id_show_func(cluster_id, "%d"); +static DEVICE_ATTR_RO(cluster_id); +#endif -define_id_show_func(core_id); +define_id_show_func(core_id, "%d"); static DEVICE_ATTR_RO(core_id); -define_siblings_show_func(thread_siblings, sibling_cpumask); -static DEVICE_ATTR_RO(thread_siblings); -static DEVICE_ATTR_RO(thread_siblings_list); +define_id_show_func(ppin, "0x%llx"); +static DEVICE_ATTR_ADMIN_RO(ppin); + +define_siblings_read_func(thread_siblings, sibling_cpumask); +static const BIN_ATTR_RO(thread_siblings, CPUMAP_FILE_MAX_BYTES); +static const BIN_ATTR_RO(thread_siblings_list, CPULIST_FILE_MAX_BYTES); -define_siblings_show_func(core_cpus, sibling_cpumask); -static DEVICE_ATTR_RO(core_cpus); -static DEVICE_ATTR_RO(core_cpus_list); +define_siblings_read_func(core_cpus, sibling_cpumask); +static const BIN_ATTR_RO(core_cpus, CPUMAP_FILE_MAX_BYTES); +static const BIN_ATTR_RO(core_cpus_list, CPULIST_FILE_MAX_BYTES); -define_siblings_show_func(core_siblings, core_cpumask); -static DEVICE_ATTR_RO(core_siblings); -static DEVICE_ATTR_RO(core_siblings_list); +define_siblings_read_func(core_siblings, core_cpumask); +static const BIN_ATTR_RO(core_siblings, CPUMAP_FILE_MAX_BYTES); +static const BIN_ATTR_RO(core_siblings_list, CPULIST_FILE_MAX_BYTES); -define_siblings_show_func(die_cpus, die_cpumask); -static DEVICE_ATTR_RO(die_cpus); -static DEVICE_ATTR_RO(die_cpus_list); +#ifdef TOPOLOGY_CLUSTER_SYSFS +define_siblings_read_func(cluster_cpus, cluster_cpumask); +static const BIN_ATTR_RO(cluster_cpus, CPUMAP_FILE_MAX_BYTES); +static const BIN_ATTR_RO(cluster_cpus_list, CPULIST_FILE_MAX_BYTES); +#endif + +#ifdef TOPOLOGY_DIE_SYSFS +define_siblings_read_func(die_cpus, die_cpumask); +static const BIN_ATTR_RO(die_cpus, CPUMAP_FILE_MAX_BYTES); +static const BIN_ATTR_RO(die_cpus_list, CPULIST_FILE_MAX_BYTES); +#endif -define_siblings_show_func(package_cpus, core_cpumask); -static DEVICE_ATTR_RO(package_cpus); -static DEVICE_ATTR_RO(package_cpus_list); +define_siblings_read_func(package_cpus, core_cpumask); +static const BIN_ATTR_RO(package_cpus, CPUMAP_FILE_MAX_BYTES); +static const BIN_ATTR_RO(package_cpus_list, CPULIST_FILE_MAX_BYTES); -#ifdef CONFIG_SCHED_BOOK -define_id_show_func(book_id); +#ifdef TOPOLOGY_BOOK_SYSFS +define_id_show_func(book_id, "%d"); static DEVICE_ATTR_RO(book_id); -define_siblings_show_func(book_siblings, book_cpumask); -static DEVICE_ATTR_RO(book_siblings); -static DEVICE_ATTR_RO(book_siblings_list); +define_siblings_read_func(book_siblings, book_cpumask); +static const BIN_ATTR_RO(book_siblings, CPUMAP_FILE_MAX_BYTES); +static const BIN_ATTR_RO(book_siblings_list, CPULIST_FILE_MAX_BYTES); #endif -#ifdef CONFIG_SCHED_DRAWER -define_id_show_func(drawer_id); +#ifdef TOPOLOGY_DRAWER_SYSFS +define_id_show_func(drawer_id, "%d"); static DEVICE_ATTR_RO(drawer_id); -define_siblings_show_func(drawer_siblings, drawer_cpumask); -static DEVICE_ATTR_RO(drawer_siblings); -static DEVICE_ATTR_RO(drawer_siblings_list); +define_siblings_read_func(drawer_siblings, drawer_cpumask); +static const BIN_ATTR_RO(drawer_siblings, CPUMAP_FILE_MAX_BYTES); +static const BIN_ATTR_RO(drawer_siblings_list, CPULIST_FILE_MAX_BYTES); #endif +static const struct bin_attribute *const bin_attrs[] = { + &bin_attr_core_cpus, + &bin_attr_core_cpus_list, + &bin_attr_thread_siblings, + &bin_attr_thread_siblings_list, + &bin_attr_core_siblings, + &bin_attr_core_siblings_list, +#ifdef TOPOLOGY_CLUSTER_SYSFS + &bin_attr_cluster_cpus, + &bin_attr_cluster_cpus_list, +#endif +#ifdef TOPOLOGY_DIE_SYSFS + &bin_attr_die_cpus, + &bin_attr_die_cpus_list, +#endif + &bin_attr_package_cpus, + &bin_attr_package_cpus_list, +#ifdef TOPOLOGY_BOOK_SYSFS + &bin_attr_book_siblings, + &bin_attr_book_siblings_list, +#endif +#ifdef TOPOLOGY_DRAWER_SYSFS + &bin_attr_drawer_siblings, + &bin_attr_drawer_siblings_list, +#endif + NULL +}; + static struct attribute *default_attrs[] = { &dev_attr_physical_package_id.attr, +#ifdef TOPOLOGY_DIE_SYSFS &dev_attr_die_id.attr, +#endif +#ifdef TOPOLOGY_CLUSTER_SYSFS + &dev_attr_cluster_id.attr, +#endif &dev_attr_core_id.attr, - &dev_attr_thread_siblings.attr, - &dev_attr_thread_siblings_list.attr, - &dev_attr_core_cpus.attr, - &dev_attr_core_cpus_list.attr, - &dev_attr_core_siblings.attr, - &dev_attr_core_siblings_list.attr, - &dev_attr_die_cpus.attr, - &dev_attr_die_cpus_list.attr, - &dev_attr_package_cpus.attr, - &dev_attr_package_cpus_list.attr, -#ifdef CONFIG_SCHED_BOOK +#ifdef TOPOLOGY_BOOK_SYSFS &dev_attr_book_id.attr, - &dev_attr_book_siblings.attr, - &dev_attr_book_siblings_list.attr, #endif -#ifdef CONFIG_SCHED_DRAWER +#ifdef TOPOLOGY_DRAWER_SYSFS &dev_attr_drawer_id.attr, - &dev_attr_drawer_siblings.attr, - &dev_attr_drawer_siblings_list.attr, #endif + &dev_attr_ppin.attr, NULL }; +static umode_t topology_is_visible(struct kobject *kobj, + struct attribute *attr, int unused) +{ + if (attr == &dev_attr_ppin.attr && !topology_ppin(kobj_to_dev(kobj)->id)) + return 0; + + return attr->mode; +} + static const struct attribute_group topology_attr_group = { .attrs = default_attrs, + .bin_attrs = bin_attrs, + .is_visible = topology_is_visible, .name = "topology" }; @@ -141,3 +208,55 @@ static int __init topology_sysfs_init(void) } device_initcall(topology_sysfs_init); + +DEFINE_PER_CPU(unsigned long, cpu_scale) = SCHED_CAPACITY_SCALE; +EXPORT_PER_CPU_SYMBOL_GPL(cpu_scale); + +void topology_set_cpu_scale(unsigned int cpu, unsigned long capacity) +{ + per_cpu(cpu_scale, cpu) = capacity; +} + +static ssize_t cpu_capacity_show(struct device *dev, + struct device_attribute *attr, + char *buf) +{ + struct cpu *cpu = container_of(dev, struct cpu, dev); + + return sysfs_emit(buf, "%lu\n", topology_get_cpu_scale(cpu->dev.id)); +} + +static DEVICE_ATTR_RO(cpu_capacity); + +static int cpu_capacity_sysctl_add(unsigned int cpu) +{ + struct device *cpu_dev = get_cpu_device(cpu); + + if (!cpu_dev) + return -ENOENT; + + device_create_file(cpu_dev, &dev_attr_cpu_capacity); + + return 0; +} + +static int cpu_capacity_sysctl_remove(unsigned int cpu) +{ + struct device *cpu_dev = get_cpu_device(cpu); + + if (!cpu_dev) + return -ENOENT; + + device_remove_file(cpu_dev, &dev_attr_cpu_capacity); + + return 0; +} + +static int register_cpu_capacity_sysctl(void) +{ + cpuhp_setup_state(CPUHP_AP_ONLINE_DYN, "topology/cpu-capacity", + cpu_capacity_sysctl_add, cpu_capacity_sysctl_remove); + + return 0; +} +subsys_initcall(register_cpu_capacity_sysctl); diff --git a/drivers/base/trace.h b/drivers/base/trace.h index 3192e18f877e..3b83b13a57ff 100644 --- a/drivers/base/trace.h +++ b/drivers/base/trace.h @@ -24,18 +24,18 @@ DECLARE_EVENT_CLASS(devres, __field(struct device *, dev) __field(const char *, op) __field(void *, node) - __field(const char *, name) + __string(name, name) __field(size_t, size) ), TP_fast_assign( - __assign_str(devname, dev_name(dev)); + __assign_str(devname); __entry->op = op; __entry->node = node; - __entry->name = name; + __assign_str(name); __entry->size = size; ), TP_printk("%s %3s %p %s (%zu bytes)", __get_str(devname), - __entry->op, __entry->node, __entry->name, __entry->size) + __entry->op, __entry->node, __get_str(name), __entry->size) ); DEFINE_EVENT(devres, devres_log, diff --git a/drivers/base/transport_class.c b/drivers/base/transport_class.c index ccc86206e508..09ee2a1e35bb 100644 --- a/drivers/base/transport_class.c +++ b/drivers/base/transport_class.c @@ -155,12 +155,27 @@ static int transport_add_class_device(struct attribute_container *cont, struct device *dev, struct device *classdev) { + struct transport_class *tclass = class_to_transport_class(cont->class); int error = attribute_container_add_class_device(classdev); struct transport_container *tcont = attribute_container_to_transport_container(cont); - if (!error && tcont->statistics) + if (error) + goto err_remove; + + if (tcont->statistics) { error = sysfs_create_group(&classdev->kobj, tcont->statistics); + if (error) + goto err_del; + } + + return 0; + +err_del: + attribute_container_class_device_del(classdev); +err_remove: + if (tclass->remove) + tclass->remove(tcont, dev, classdev); return error; } |