diff options
Diffstat (limited to 'arch/s390/pci')
| -rw-r--r-- | arch/s390/pci/Makefile | 6 | ||||
| -rw-r--r-- | arch/s390/pci/pci.c | 697 | ||||
| -rw-r--r-- | arch/s390/pci/pci_bus.c | 374 | ||||
| -rw-r--r-- | arch/s390/pci/pci_bus.h | 42 | ||||
| -rw-r--r-- | arch/s390/pci/pci_clp.c | 304 | ||||
| -rw-r--r-- | arch/s390/pci/pci_debug.c | 24 | ||||
| -rw-r--r-- | arch/s390/pci/pci_dma.c | 684 | ||||
| -rw-r--r-- | arch/s390/pci/pci_event.c | 355 | ||||
| -rw-r--r-- | arch/s390/pci/pci_insn.c | 177 | ||||
| -rw-r--r-- | arch/s390/pci/pci_iov.c | 99 | ||||
| -rw-r--r-- | arch/s390/pci/pci_iov.h | 30 | ||||
| -rw-r--r-- | arch/s390/pci/pci_irq.c | 150 | ||||
| -rw-r--r-- | arch/s390/pci/pci_kvm_hook.c | 11 | ||||
| -rw-r--r-- | arch/s390/pci/pci_mmio.c | 292 | ||||
| -rw-r--r-- | arch/s390/pci/pci_sysfs.c | 137 |
15 files changed, 2065 insertions, 1317 deletions
diff --git a/arch/s390/pci/Makefile b/arch/s390/pci/Makefile index 748626a33028..0547a10406e7 100644 --- a/arch/s390/pci/Makefile +++ b/arch/s390/pci/Makefile @@ -3,5 +3,7 @@ # Makefile for the s390 PCI subsystem. # -obj-$(CONFIG_PCI) += pci.o pci_irq.o pci_dma.o pci_clp.o pci_sysfs.o \ - pci_event.o pci_debug.o pci_insn.o pci_mmio.o +obj-$(CONFIG_PCI) += pci.o pci_irq.o pci_clp.o pci_sysfs.o \ + pci_event.o pci_debug.o pci_insn.o pci_mmio.o \ + pci_bus.o pci_kvm_hook.o +obj-$(CONFIG_PCI_IOV) += pci_iov.o diff --git a/arch/s390/pci/pci.c b/arch/s390/pci/pci.c index 8e872951c07b..26afde0d1ed3 100644 --- a/arch/s390/pci/pci.c +++ b/arch/s390/pci/pci.c @@ -28,6 +28,7 @@ #include <linux/jump_label.h> #include <linux/pci.h> #include <linux/printk.h> +#include <linux/lockdep.h> #include <asm/isc.h> #include <asm/airq.h> @@ -36,17 +37,22 @@ #include <asm/pci_clp.h> #include <asm/pci_dma.h> +#include "pci_bus.h" +#include "pci_iov.h" + /* list of all detected zpci devices */ static LIST_HEAD(zpci_list); static DEFINE_SPINLOCK(zpci_list_lock); -static DECLARE_BITMAP(zpci_domain, ZPCI_NR_DEVICES); +static DECLARE_BITMAP(zpci_domain, ZPCI_DOMAIN_BITMAP_SIZE); static DEFINE_SPINLOCK(zpci_domain_lock); #define ZPCI_IOMAP_ENTRIES \ min(((unsigned long) ZPCI_NR_DEVICES * PCI_STD_NUM_BARS / 2), \ ZPCI_IOMAP_MAX_ENTRIES) +unsigned int s390_pci_no_rid; + static DEFINE_SPINLOCK(zpci_iomap_lock); static unsigned long *zpci_iomap_bitmap; struct zpci_iomap_entry *zpci_iomap_start; @@ -56,6 +62,12 @@ DEFINE_STATIC_KEY_FALSE(have_mio); static struct kmem_cache *zdev_fmb_cache; +/* AEN structures that must be preserved over KVM module re-insertion */ +union zpci_sic_iib *zpci_aipb; +EXPORT_SYMBOL_GPL(zpci_aipb); +struct airq_iv *zpci_aif_sbv; +EXPORT_SYMBOL_GPL(zpci_aif_sbv); + struct zpci_dev *get_zdev_by_fid(u32 fid) { struct zpci_dev *tmp, *zdev = NULL; @@ -64,6 +76,7 @@ struct zpci_dev *get_zdev_by_fid(u32 fid) list_for_each_entry(tmp, &zpci_list, entry) { if (tmp->fid == fid) { zdev = tmp; + zpci_zdev_get(zdev); break; } } @@ -87,17 +100,12 @@ void zpci_remove_reserved_devices(void) spin_unlock(&zpci_list_lock); list_for_each_entry_safe(zdev, tmp, &remove, entry) - zpci_remove_device(zdev); -} - -static struct zpci_dev *get_zdev_by_bus(struct pci_bus *bus) -{ - return (bus && bus->sysdata) ? (struct zpci_dev *) bus->sysdata : NULL; + zpci_device_reserved(zdev); } int pci_domain_nr(struct pci_bus *bus) { - return ((struct zpci_dev *) bus->sysdata)->domain; + return ((struct zpci_bus *) bus->sysdata)->domain_nr; } EXPORT_SYMBOL_GPL(pci_domain_nr); @@ -109,18 +117,27 @@ EXPORT_SYMBOL_GPL(pci_proc_domain); /* Modify PCI: Register I/O address translation parameters */ int zpci_register_ioat(struct zpci_dev *zdev, u8 dmaas, - u64 base, u64 limit, u64 iota) + u64 base, u64 limit, u64 iota, u8 *status) { u64 req = ZPCI_CREATE_REQ(zdev->fh, dmaas, ZPCI_MOD_FC_REG_IOAT); struct zpci_fib fib = {0}; - u8 status; + u8 cc; WARN_ON_ONCE(iota & 0x3fff); fib.pba = base; - fib.pal = limit; + /* Work around off by one in ISM virt device */ + if (zdev->pft == PCI_FUNC_TYPE_ISM && limit > base) + fib.pal = limit + (1 << 12); + else + fib.pal = limit; fib.iota = iota | ZPCI_IOTA_RTTO_FLAG; - return zpci_mod_fc(req, &fib, &status) ? -EIO : 0; + fib.gd = zdev->gisa; + cc = zpci_mod_fc(req, &fib, status); + if (cc) + zpci_dbg(3, "reg ioat fid:%x, cc:%d, status:%d\n", zdev->fid, cc, *status); + return cc; } +EXPORT_SYMBOL_GPL(zpci_register_ioat); /* Modify PCI: Unregister I/O address translation parameters */ int zpci_unregister_ioat(struct zpci_dev *zdev, u8 dmaas) @@ -129,16 +146,19 @@ int zpci_unregister_ioat(struct zpci_dev *zdev, u8 dmaas) struct zpci_fib fib = {0}; u8 cc, status; + fib.gd = zdev->gisa; + cc = zpci_mod_fc(req, &fib, &status); - if (cc == 3) /* Function already gone. */ - cc = 0; - return cc ? -EIO : 0; + if (cc) + zpci_dbg(3, "unreg ioat fid:%x, cc:%d, status:%d\n", zdev->fid, cc, status); + return cc; } /* Modify PCI: Set PCI function measurement parameters */ int zpci_fmb_enable_device(struct zpci_dev *zdev) { u64 req = ZPCI_CREATE_REQ(zdev->fh, 0, ZPCI_MOD_FC_SET_MEASURE); + struct zpci_iommu_ctrs *ctrs; struct zpci_fib fib = {0}; u8 cc, status; @@ -151,11 +171,18 @@ int zpci_fmb_enable_device(struct zpci_dev *zdev) WARN_ON((u64) zdev->fmb & 0xf); /* reset software counters */ - atomic64_set(&zdev->allocated_pages, 0); - atomic64_set(&zdev->mapped_pages, 0); - atomic64_set(&zdev->unmapped_pages, 0); + ctrs = zpci_get_iommu_ctrs(zdev); + if (ctrs) { + atomic64_set(&ctrs->mapped_pages, 0); + atomic64_set(&ctrs->unmapped_pages, 0); + atomic64_set(&ctrs->global_rpcits, 0); + atomic64_set(&ctrs->sync_map_rpcits, 0); + atomic64_set(&ctrs->sync_rpcits, 0); + } + fib.fmb_addr = virt_to_phys(zdev->fmb); + fib.gd = zdev->gisa; cc = zpci_mod_fc(req, &fib, &status); if (cc) { kmem_cache_free(zdev_fmb_cache, zdev->fmb); @@ -174,6 +201,8 @@ int zpci_fmb_disable_device(struct zpci_dev *zdev) if (!zdev->fmb) return -EINVAL; + fib.gd = zdev->gisa; + /* Function measurement is disabled if fmb address is zero */ cc = zpci_mod_fc(req, &fib, &status); if (cc == 3) /* Function already gone. */ @@ -224,41 +253,28 @@ resource_size_t pcibios_align_resource(void *data, const struct resource *res, /* combine single writes by using store-block insn */ void __iowrite64_copy(void __iomem *to, const void *from, size_t count) { - zpci_memcpy_toio(to, from, count); + zpci_memcpy_toio(to, from, count * 8); } -void __iomem *ioremap(unsigned long ioaddr, unsigned long size) +void __iomem *ioremap_prot(phys_addr_t phys_addr, size_t size, + unsigned long prot) { - struct vm_struct *area; - unsigned long offset; - - if (!size) - return NULL; - + /* + * When PCI MIO instructions are unavailable the "physical" address + * encodes a hint for accessing the PCI memory space it represents. + * Just pass it unchanged such that ioread/iowrite can decode it. + */ if (!static_branch_unlikely(&have_mio)) - return (void __iomem *) ioaddr; + return (void __iomem *)phys_addr; - offset = ioaddr & ~PAGE_MASK; - ioaddr &= PAGE_MASK; - size = PAGE_ALIGN(size + offset); - area = get_vm_area(size, VM_IOREMAP); - if (!area) - return NULL; - - if (ioremap_page_range((unsigned long) area->addr, - (unsigned long) area->addr + size, - ioaddr, PAGE_KERNEL)) { - vunmap(area->addr); - return NULL; - } - return (void __iomem *) ((unsigned long) area->addr + offset); + return generic_ioremap_prot(phys_addr, size, __pgprot(prot)); } -EXPORT_SYMBOL(ioremap); +EXPORT_SYMBOL(ioremap_prot); void iounmap(volatile void __iomem *addr) { if (static_branch_likely(&have_mio)) - vunmap((__force void *) ((unsigned long) addr & PAGE_MASK)); + generic_iounmap(addr); } EXPORT_SYMBOL(iounmap); @@ -372,29 +388,17 @@ EXPORT_SYMBOL(pci_iounmap); static int pci_read(struct pci_bus *bus, unsigned int devfn, int where, int size, u32 *val) { - struct zpci_dev *zdev = get_zdev_by_bus(bus); - int ret; - - if (!zdev || devfn != ZPCI_DEVFN) - ret = -ENODEV; - else - ret = zpci_cfg_load(zdev, where, val, size); + struct zpci_dev *zdev = zdev_from_bus(bus, devfn); - return ret; + return (zdev) ? zpci_cfg_load(zdev, where, val, size) : -ENODEV; } static int pci_write(struct pci_bus *bus, unsigned int devfn, int where, int size, u32 val) { - struct zpci_dev *zdev = get_zdev_by_bus(bus); - int ret; + struct zpci_dev *zdev = zdev_from_bus(bus, devfn); - if (!zdev || devfn != ZPCI_DEVFN) - ret = -ENODEV; - else - ret = zpci_cfg_store(zdev, where, val, size); - - return ret; + return (zdev) ? zpci_cfg_store(zdev, where, val, size) : -ENODEV; } static struct pci_ops pci_root_ops = { @@ -402,15 +406,6 @@ static struct pci_ops pci_root_ops = { .write = pci_write, }; -#ifdef CONFIG_PCI_IOV -static struct resource iov_res = { - .name = "PCI IOV res", - .start = 0, - .end = -1, - .flags = IORESOURCE_MEM, -}; -#endif - static void zpci_map_resources(struct pci_dev *pdev) { struct zpci_dev *zdev = to_zpci(pdev); @@ -424,23 +419,14 @@ static void zpci_map_resources(struct pci_dev *pdev) if (zpci_use_mio(zdev)) pdev->resource[i].start = - (resource_size_t __force) zdev->bars[i].mio_wb; + (resource_size_t __force) zdev->bars[i].mio_wt; else pdev->resource[i].start = (resource_size_t __force) pci_iomap_range_fh(pdev, i, 0, 0); pdev->resource[i].end = pdev->resource[i].start + len - 1; } -#ifdef CONFIG_PCI_IOV - for (i = 0; i < PCI_SRIOV_NUM_BARS; i++) { - int bar = i + PCI_IOV_RESOURCES; - - len = pci_resource_len(pdev, bar); - if (!len) - continue; - pdev->resource[bar].parent = &iov_res; - } -#endif + zpci_iov_map_resources(pdev); } static void zpci_unmap_resources(struct pci_dev *pdev) @@ -484,6 +470,34 @@ static void zpci_free_iomap(struct zpci_dev *zdev, int entry) spin_unlock(&zpci_iomap_lock); } +static void zpci_do_update_iomap_fh(struct zpci_dev *zdev, u32 fh) +{ + int bar, idx; + + spin_lock(&zpci_iomap_lock); + for (bar = 0; bar < PCI_STD_NUM_BARS; bar++) { + if (!zdev->bars[bar].size) + continue; + idx = zdev->bars[bar].map_idx; + if (!zpci_iomap_start[idx].count) + continue; + WRITE_ONCE(zpci_iomap_start[idx].fh, zdev->fh); + } + spin_unlock(&zpci_iomap_lock); +} + +void zpci_update_fh(struct zpci_dev *zdev, u32 fh) +{ + if (!fh || zdev->fh == fh) + return; + + zdev->fh = fh; + if (zpci_use_mio(zdev)) + return; + if (zdev->has_resources && zdev_enabled(zdev)) + zpci_do_update_iomap_fh(zdev, fh); +} + static struct resource *__alloc_res(struct zpci_dev *zdev, unsigned long start, unsigned long size, unsigned long flags) { @@ -505,15 +519,14 @@ static struct resource *__alloc_res(struct zpci_dev *zdev, unsigned long start, return r; } -static int zpci_setup_bus_resources(struct zpci_dev *zdev, - struct list_head *resources) +int zpci_setup_bus_resources(struct zpci_dev *zdev) { unsigned long addr, size, flags; struct resource *res; int i, entry; snprintf(zdev->res_name, sizeof(zdev->res_name), - "PCI Bus %04x:%02x", zdev->domain, ZPCI_BUS_NR); + "PCI Bus %04x:%02x", zdev->uid, ZPCI_BUS_NR); for (i = 0; i < PCI_STD_NUM_BARS; i++) { if (!zdev->bars[i].size) @@ -531,7 +544,7 @@ static int zpci_setup_bus_resources(struct zpci_dev *zdev, flags |= IORESOURCE_MEM_64; if (zpci_use_mio(zdev)) - addr = (unsigned long) zdev->bars[i].mio_wb; + addr = (unsigned long) zdev->bars[i].mio_wt; else addr = ZPCI_ADDR(entry); size = 1UL << zdev->bars[i].size; @@ -542,36 +555,45 @@ static int zpci_setup_bus_resources(struct zpci_dev *zdev, return -ENOMEM; } zdev->bars[i].res = res; - pci_add_resource(resources, res); } + zdev->has_resources = 1; return 0; } static void zpci_cleanup_bus_resources(struct zpci_dev *zdev) { + struct resource *res; int i; + pci_lock_rescan_remove(); for (i = 0; i < PCI_STD_NUM_BARS; i++) { - if (!zdev->bars[i].size || !zdev->bars[i].res) + res = zdev->bars[i].res; + if (!res) continue; + release_resource(res); + pci_bus_remove_resource(zdev->zbus->bus, res); zpci_free_iomap(zdev, zdev->bars[i].map_idx); - release_resource(zdev->bars[i].res); - kfree(zdev->bars[i].res); + zdev->bars[i].res = NULL; + kfree(res); } + zdev->has_resources = 0; + pci_unlock_rescan_remove(); } -int pcibios_add_device(struct pci_dev *pdev) +int pcibios_device_add(struct pci_dev *pdev) { + struct zpci_dev *zdev = to_zpci(pdev); struct resource *res; int i; + /* The pdev has a reference to the zdev via its bus */ + zpci_zdev_get(zdev); if (pdev->is_physfn) pdev->no_vf_scan = 1; pdev->dev.groups = zpci_attr_groups; - pdev->dev.dma_ops = &s390_pci_dma_ops; zpci_map_resources(pdev); for (i = 0; i < PCI_STD_NUM_BARS; i++) { @@ -586,7 +608,10 @@ int pcibios_add_device(struct pci_dev *pdev) void pcibios_release_device(struct pci_dev *pdev) { + struct zpci_dev *zdev = to_zpci(pdev); + zpci_unmap_resources(pdev); + zpci_zdev_put(zdev); } int pcibios_enable_device(struct pci_dev *pdev, int mask) @@ -607,210 +632,330 @@ void pcibios_disable_device(struct pci_dev *pdev) zpci_debug_exit_device(zdev); } -#ifdef CONFIG_HIBERNATE_CALLBACKS -static int zpci_restore(struct device *dev) +static int __zpci_register_domain(int domain) { - struct pci_dev *pdev = to_pci_dev(dev); - struct zpci_dev *zdev = to_zpci(pdev); - int ret = 0; - - if (zdev->state != ZPCI_FN_STATE_ONLINE) - goto out; - - ret = clp_enable_fh(zdev, ZPCI_NR_DMA_SPACES); - if (ret) - goto out; - - zpci_map_resources(pdev); - zpci_register_ioat(zdev, 0, zdev->start_dma, zdev->end_dma, - (u64) zdev->dma_table); - -out: - return ret; + spin_lock(&zpci_domain_lock); + if (test_bit(domain, zpci_domain)) { + spin_unlock(&zpci_domain_lock); + pr_err("Domain %04x is already assigned\n", domain); + return -EEXIST; + } + set_bit(domain, zpci_domain); + spin_unlock(&zpci_domain_lock); + return domain; } -static int zpci_freeze(struct device *dev) +static int __zpci_alloc_domain(void) { - struct pci_dev *pdev = to_pci_dev(dev); - struct zpci_dev *zdev = to_zpci(pdev); + int domain; - if (zdev->state != ZPCI_FN_STATE_ONLINE) - return 0; - - zpci_unregister_ioat(zdev, 0); - zpci_unmap_resources(pdev); - return clp_disable_fh(zdev); + spin_lock(&zpci_domain_lock); + /* + * We can always auto allocate domains below ZPCI_NR_DEVICES. + * There is either a free domain or we have reached the maximum in + * which case we would have bailed earlier. + */ + domain = find_first_zero_bit(zpci_domain, ZPCI_NR_DEVICES); + set_bit(domain, zpci_domain); + spin_unlock(&zpci_domain_lock); + return domain; } -struct dev_pm_ops pcibios_pm_ops = { - .thaw_noirq = zpci_restore, - .freeze_noirq = zpci_freeze, - .restore_noirq = zpci_restore, - .poweroff_noirq = zpci_freeze, -}; -#endif /* CONFIG_HIBERNATE_CALLBACKS */ - -static int zpci_alloc_domain(struct zpci_dev *zdev) +int zpci_alloc_domain(int domain) { if (zpci_unique_uid) { - zdev->domain = (u16) zdev->uid; - if (zdev->domain >= ZPCI_NR_DEVICES) - return 0; - - spin_lock(&zpci_domain_lock); - if (test_bit(zdev->domain, zpci_domain)) { - spin_unlock(&zpci_domain_lock); - pr_err("Adding PCI function %08x failed because domain %04x is already assigned\n", - zdev->fid, zdev->domain); - return -EEXIST; - } - set_bit(zdev->domain, zpci_domain); - spin_unlock(&zpci_domain_lock); - return 0; - } - - spin_lock(&zpci_domain_lock); - zdev->domain = find_first_zero_bit(zpci_domain, ZPCI_NR_DEVICES); - if (zdev->domain == ZPCI_NR_DEVICES) { - spin_unlock(&zpci_domain_lock); - pr_err("Adding PCI function %08x failed because the configured limit of %d is reached\n", - zdev->fid, ZPCI_NR_DEVICES); - return -ENOSPC; + if (domain) + return __zpci_register_domain(domain); + pr_warn("UID checking was active but no UID is provided: switching to automatic domain allocation\n"); + update_uid_checking(false); } - set_bit(zdev->domain, zpci_domain); - spin_unlock(&zpci_domain_lock); - return 0; + return __zpci_alloc_domain(); } -static void zpci_free_domain(struct zpci_dev *zdev) +void zpci_free_domain(int domain) { - if (zdev->domain >= ZPCI_NR_DEVICES) - return; - spin_lock(&zpci_domain_lock); - clear_bit(zdev->domain, zpci_domain); + clear_bit(domain, zpci_domain); spin_unlock(&zpci_domain_lock); } -void pcibios_remove_bus(struct pci_bus *bus) + +int zpci_enable_device(struct zpci_dev *zdev) { - struct zpci_dev *zdev = get_zdev_by_bus(bus); + u32 fh = zdev->fh; + int rc = 0; - zpci_exit_slot(zdev); - zpci_cleanup_bus_resources(zdev); - zpci_destroy_iommu(zdev); - zpci_free_domain(zdev); + if (clp_enable_fh(zdev, &fh, ZPCI_NR_DMA_SPACES)) + rc = -EIO; + else + zpci_update_fh(zdev, fh); + return rc; +} +EXPORT_SYMBOL_GPL(zpci_enable_device); - spin_lock(&zpci_list_lock); - list_del(&zdev->entry); - spin_unlock(&zpci_list_lock); +int zpci_disable_device(struct zpci_dev *zdev) +{ + u32 fh = zdev->fh; + int cc, rc = 0; - zpci_dbg(3, "rem fid:%x\n", zdev->fid); - kfree(zdev); + cc = clp_disable_fh(zdev, &fh); + if (!cc) { + zpci_update_fh(zdev, fh); + } else if (cc == CLP_RC_SETPCIFN_ALRDY) { + pr_info("Disabling PCI function %08x had no effect as it was already disabled\n", + zdev->fid); + /* Function is already disabled - update handle */ + rc = clp_refresh_fh(zdev->fid, &fh); + if (!rc) { + zpci_update_fh(zdev, fh); + rc = -EINVAL; + } + } else { + rc = -EIO; + } + return rc; } +EXPORT_SYMBOL_GPL(zpci_disable_device); -static int zpci_scan_bus(struct zpci_dev *zdev) +/** + * zpci_hot_reset_device - perform a reset of the given zPCI function + * @zdev: the slot which should be reset + * + * Performs a low level reset of the zPCI function. The reset is low level in + * the sense that the zPCI function can be reset without detaching it from the + * common PCI subsystem. The reset may be performed while under control of + * either DMA or IOMMU APIs in which case the existing DMA/IOMMU translation + * table is reinstated at the end of the reset. + * + * After the reset the functions internal state is reset to an initial state + * equivalent to its state during boot when first probing a driver. + * Consequently after reset the PCI function requires re-initialization via the + * common PCI code including re-enabling IRQs via pci_alloc_irq_vectors() + * and enabling the function via e.g. pci_enable_device_flags(). The caller + * must guard against concurrent reset attempts. + * + * In most cases this function should not be called directly but through + * pci_reset_function() or pci_reset_bus() which handle the save/restore and + * locking - asserted by lockdep. + * + * Return: 0 on success and an error value otherwise + */ +int zpci_hot_reset_device(struct zpci_dev *zdev) { - LIST_HEAD(resources); - int ret; + u8 status; + int rc; - ret = zpci_setup_bus_resources(zdev, &resources); - if (ret) - goto error; + lockdep_assert_held(&zdev->state_lock); + zpci_dbg(3, "rst fid:%x, fh:%x\n", zdev->fid, zdev->fh); + if (zdev_enabled(zdev)) { + /* Disables device access, DMAs and IRQs (reset state) */ + rc = zpci_disable_device(zdev); + /* + * Due to a z/VM vs LPAR inconsistency in the error state the + * FH may indicate an enabled device but disable says the + * device is already disabled don't treat it as an error here. + */ + if (rc == -EINVAL) + rc = 0; + if (rc) + return rc; + } - zdev->bus = pci_scan_root_bus(NULL, ZPCI_BUS_NR, &pci_root_ops, - zdev, &resources); - if (!zdev->bus) { - ret = -EIO; - goto error; + rc = zpci_enable_device(zdev); + if (rc) + return rc; + + if (zdev->dma_table) + rc = zpci_register_ioat(zdev, 0, zdev->start_dma, zdev->end_dma, + virt_to_phys(zdev->dma_table), &status); + if (rc) { + zpci_disable_device(zdev); + return rc; } - zdev->bus->max_bus_speed = zdev->max_bus_speed; - pci_bus_add_devices(zdev->bus); - return 0; -error: - zpci_cleanup_bus_resources(zdev); - pci_free_resource_list(&resources); - return ret; + return 0; } -int zpci_enable_device(struct zpci_dev *zdev) +/** + * zpci_create_device() - Create a new zpci_dev and add it to the zbus + * @fid: Function ID of the device to be created + * @fh: Current Function Handle of the device to be created + * @state: Initial state after creation either Standby or Configured + * + * Creates a new zpci device and adds it to its, possibly newly created, zbus + * as well as zpci_list. + * + * Returns: the zdev on success or an error pointer otherwise + */ +struct zpci_dev *zpci_create_device(u32 fid, u32 fh, enum zpci_state state) { + struct zpci_dev *zdev; int rc; - rc = clp_enable_fh(zdev, ZPCI_NR_DMA_SPACES); + zpci_dbg(1, "add fid:%x, fh:%x, c:%d\n", fid, fh, state); + zdev = kzalloc(sizeof(*zdev), GFP_KERNEL); + if (!zdev) + return ERR_PTR(-ENOMEM); + + /* FID and Function Handle are the static/dynamic identifiers */ + zdev->fid = fid; + zdev->fh = fh; + + /* Query function properties and update zdev */ + rc = clp_query_pci_fn(zdev); if (rc) - goto out; + goto error; + zdev->state = state; + + kref_init(&zdev->kref); + mutex_init(&zdev->state_lock); + mutex_init(&zdev->fmb_lock); + mutex_init(&zdev->kzdev_lock); - rc = zpci_dma_init_device(zdev); + rc = zpci_init_iommu(zdev); if (rc) - goto out_dma; + goto error; - zdev->state = ZPCI_FN_STATE_ONLINE; - return 0; + rc = zpci_bus_device_register(zdev, &pci_root_ops); + if (rc) + goto error_destroy_iommu; -out_dma: - clp_disable_fh(zdev); -out: - return rc; + spin_lock(&zpci_list_lock); + list_add_tail(&zdev->entry, &zpci_list); + spin_unlock(&zpci_list_lock); + + return zdev; + +error_destroy_iommu: + zpci_destroy_iommu(zdev); +error: + zpci_dbg(0, "add fid:%x, rc:%d\n", fid, rc); + kfree(zdev); + return ERR_PTR(rc); } -EXPORT_SYMBOL_GPL(zpci_enable_device); -int zpci_disable_device(struct zpci_dev *zdev) +bool zpci_is_device_configured(struct zpci_dev *zdev) { - zpci_dma_exit_device(zdev); - return clp_disable_fh(zdev); + enum zpci_state state = zdev->state; + + return state != ZPCI_FN_STATE_RESERVED && + state != ZPCI_FN_STATE_STANDBY; +} + +/** + * zpci_scan_configured_device() - Scan a freshly configured zpci_dev + * @zdev: The zpci_dev to be configured + * @fh: The general function handle supplied by the platform + * + * Given a device in the configuration state Configured, enables, scans and + * adds it to the common code PCI subsystem if possible. If any failure occurs, + * the zpci_dev is left disabled. + * + * Return: 0 on success, or an error code otherwise + */ +int zpci_scan_configured_device(struct zpci_dev *zdev, u32 fh) +{ + zpci_update_fh(zdev, fh); + return zpci_bus_scan_device(zdev); } -EXPORT_SYMBOL_GPL(zpci_disable_device); -int zpci_create_device(struct zpci_dev *zdev) +/** + * zpci_deconfigure_device() - Deconfigure a zpci_dev + * @zdev: The zpci_dev to configure + * + * Deconfigure a zPCI function that is currently configured and possibly known + * to the common code PCI subsystem. + * If any failure occurs the device is left as is. + * + * Return: 0 on success, or an error code otherwise + */ +int zpci_deconfigure_device(struct zpci_dev *zdev) { int rc; - rc = zpci_alloc_domain(zdev); - if (rc) - goto out; + lockdep_assert_held(&zdev->state_lock); + if (zdev->state != ZPCI_FN_STATE_CONFIGURED) + return 0; - rc = zpci_init_iommu(zdev); - if (rc) - goto out_free; + if (zdev->zbus->bus) + zpci_bus_remove_device(zdev, false); - mutex_init(&zdev->lock); - if (zdev->state == ZPCI_FN_STATE_CONFIGURED) { - rc = zpci_enable_device(zdev); + if (zdev_enabled(zdev)) { + rc = zpci_disable_device(zdev); if (rc) - goto out_destroy_iommu; + return rc; } - rc = zpci_scan_bus(zdev); + + rc = sclp_pci_deconfigure(zdev->fid); + zpci_dbg(3, "deconf fid:%x, rc:%d\n", zdev->fid, rc); if (rc) - goto out_disable; + return rc; + zdev->state = ZPCI_FN_STATE_STANDBY; + return 0; +} + +/** + * zpci_device_reserved() - Mark device as reserved + * @zdev: the zpci_dev that was reserved + * + * Handle the case that a given zPCI function was reserved by another system. + * After a call to this function the zpci_dev can not be found via + * get_zdev_by_fid() anymore but may still be accessible via existing + * references though it will not be functional anymore. + */ +void zpci_device_reserved(struct zpci_dev *zdev) +{ + /* + * Remove device from zpci_list as it is going away. This also + * makes sure we ignore subsequent zPCI events for this device. + */ spin_lock(&zpci_list_lock); - list_add_tail(&zdev->entry, &zpci_list); + list_del(&zdev->entry); spin_unlock(&zpci_list_lock); + zdev->state = ZPCI_FN_STATE_RESERVED; + zpci_dbg(3, "rsv fid:%x\n", zdev->fid); + zpci_zdev_put(zdev); +} - zpci_init_slot(zdev); +void zpci_release_device(struct kref *kref) +{ + struct zpci_dev *zdev = container_of(kref, struct zpci_dev, kref); + int ret; - return 0; + if (zdev->has_hp_slot) + zpci_exit_slot(zdev); -out_disable: - if (zdev->state == ZPCI_FN_STATE_ONLINE) - zpci_disable_device(zdev); -out_destroy_iommu: - zpci_destroy_iommu(zdev); -out_free: - zpci_free_domain(zdev); -out: - return rc; -} + if (zdev->zbus->bus) + zpci_bus_remove_device(zdev, false); -void zpci_remove_device(struct zpci_dev *zdev) -{ - if (!zdev->bus) - return; + if (zdev_enabled(zdev)) + zpci_disable_device(zdev); - pci_stop_root_bus(zdev->bus); - pci_remove_root_bus(zdev->bus); + switch (zdev->state) { + case ZPCI_FN_STATE_CONFIGURED: + ret = sclp_pci_deconfigure(zdev->fid); + zpci_dbg(3, "deconf fid:%x, rc:%d\n", zdev->fid, ret); + fallthrough; + case ZPCI_FN_STATE_STANDBY: + if (zdev->has_hp_slot) + zpci_exit_slot(zdev); + spin_lock(&zpci_list_lock); + list_del(&zdev->entry); + spin_unlock(&zpci_list_lock); + zpci_dbg(3, "rsv fid:%x\n", zdev->fid); + fallthrough; + case ZPCI_FN_STATE_RESERVED: + if (zdev->has_resources) + zpci_cleanup_bus_resources(zdev); + zpci_bus_device_unregister(zdev); + zpci_destroy_iommu(zdev); + fallthrough; + default: + break; + } + zpci_dbg(3, "rem fid:%x\n", zdev->fid); + kfree_rcu(zdev, rcu); } int zpci_report_error(struct pci_dev *pdev, @@ -822,6 +967,59 @@ int zpci_report_error(struct pci_dev *pdev, } EXPORT_SYMBOL(zpci_report_error); +/** + * zpci_clear_error_state() - Clears the zPCI error state of the device + * @zdev: The zdev for which the zPCI error state should be reset + * + * Clear the zPCI error state of the device. If clearing the zPCI error state + * fails the device is left in the error state. In this case it may make sense + * to call zpci_io_perm_failure() on the associated pdev if it exists. + * + * Returns: 0 on success, -EIO otherwise + */ +int zpci_clear_error_state(struct zpci_dev *zdev) +{ + u64 req = ZPCI_CREATE_REQ(zdev->fh, 0, ZPCI_MOD_FC_RESET_ERROR); + struct zpci_fib fib = {0}; + u8 status; + int cc; + + cc = zpci_mod_fc(req, &fib, &status); + if (cc) { + zpci_dbg(3, "ces fid:%x, cc:%d, status:%x\n", zdev->fid, cc, status); + return -EIO; + } + + return 0; +} + +/** + * zpci_reset_load_store_blocked() - Re-enables L/S from error state + * @zdev: The zdev for which to unblock load/store access + * + * Re-enables load/store access for a PCI function in the error state while + * keeping DMA blocked. In this state drivers can poke MMIO space to determine + * if error recovery is possible while catching any rogue DMA access from the + * device. + * + * Returns: 0 on success, -EIO otherwise + */ +int zpci_reset_load_store_blocked(struct zpci_dev *zdev) +{ + u64 req = ZPCI_CREATE_REQ(zdev->fh, 0, ZPCI_MOD_FC_RESET_BLOCK); + struct zpci_fib fib = {0}; + u8 status; + int cc; + + cc = zpci_mod_fc(req, &fib, &status); + if (cc) { + zpci_dbg(3, "rls fid:%x, cc:%d, status:%x\n", zdev->fid, cc, status); + return -EIO; + } + + return 0; +} + static int zpci_mem_init(void) { BUILD_BUG_ON(!is_power_of_2(__alignof__(struct zpci_fmb)) || @@ -842,6 +1040,9 @@ static int zpci_mem_init(void) if (!zpci_iomap_bitmap) goto error_iomap_bitmap; + if (static_branch_likely(&have_mio)) + clp_setup_writeback_mio(); + return 0; error_iomap_bitmap: kfree(zpci_iomap_start); @@ -859,7 +1060,6 @@ static void zpci_mem_exit(void) } static unsigned int s390_pci_probe __initdata = 1; -static unsigned int s390_pci_no_mio __initdata; unsigned int s390_pci_force_floating __initdata; static unsigned int s390_pci_initialized; @@ -870,13 +1070,17 @@ char * __init pcibios_setup(char *str) return NULL; } if (!strcmp(str, "nomio")) { - s390_pci_no_mio = 1; + S390_lowcore.machine_flags &= ~MACHINE_FLAG_PCI_MIO; return NULL; } if (!strcmp(str, "force_floating")) { s390_pci_force_floating = 1; return NULL; } + if (!strcmp(str, "norid")) { + s390_pci_no_rid = 1; + return NULL; + } return str; } @@ -892,12 +1096,14 @@ static int __init pci_base_init(void) if (!s390_pci_probe) return 0; - if (!test_facility(69) || !test_facility(71)) + if (!test_facility(69) || !test_facility(71)) { + pr_info("PCI is not supported because CPU facilities 69 or 71 are not available\n"); return 0; + } - if (test_facility(153) && !s390_pci_no_mio) { + if (MACHINE_HAS_PCI_MIO) { static_branch_enable(&have_mio); - ctl_set_bit(2, 5); + system_ctl_set_bit(2, CR2_MIO_ADDRESSING_BIT); } rc = zpci_debug_init(); @@ -912,20 +1118,15 @@ static int __init pci_base_init(void) if (rc) goto out_irq; - rc = zpci_dma_init(); - if (rc) - goto out_dma; - rc = clp_scan_pci_devices(); if (rc) goto out_find; + zpci_bus_scan_busses(); s390_pci_initialized = 1; return 0; out_find: - zpci_dma_exit(); -out_dma: zpci_irq_exit(); out_irq: zpci_mem_exit(); @@ -935,9 +1136,3 @@ out: return rc; } subsys_initcall_sync(pci_base_init); - -void zpci_rescan(void) -{ - if (zpci_is_enabled()) - clp_rescan_pci_devices_simple(); -} diff --git a/arch/s390/pci/pci_bus.c b/arch/s390/pci/pci_bus.c new file mode 100644 index 000000000000..daa5d7450c7d --- /dev/null +++ b/arch/s390/pci/pci_bus.c @@ -0,0 +1,374 @@ +// SPDX-License-Identifier: GPL-2.0 +/* + * Copyright IBM Corp. 2020 + * + * Author(s): + * Pierre Morel <pmorel@linux.ibm.com> + * + */ + +#define KMSG_COMPONENT "zpci" +#define pr_fmt(fmt) KMSG_COMPONENT ": " fmt + +#include <linux/kernel.h> +#include <linux/slab.h> +#include <linux/err.h> +#include <linux/export.h> +#include <linux/delay.h> +#include <linux/seq_file.h> +#include <linux/jump_label.h> +#include <linux/pci.h> +#include <linux/printk.h> + +#include <asm/pci_clp.h> +#include <asm/pci_dma.h> + +#include "pci_bus.h" +#include "pci_iov.h" + +static LIST_HEAD(zbus_list); +static DEFINE_MUTEX(zbus_list_lock); +static int zpci_nb_devices; + +/* zpci_bus_prepare_device - Prepare a zPCI function for scanning + * @zdev: the zPCI function to be prepared + * + * The PCI resources for the function are set up and added to its zbus and the + * function is enabled. The function must be added to a zbus which must have + * a PCI bus created. If an error occurs the zPCI function is not enabled. + * + * Return: 0 on success, an error code otherwise + */ +static int zpci_bus_prepare_device(struct zpci_dev *zdev) +{ + int rc, i; + + if (!zdev_enabled(zdev)) { + rc = zpci_enable_device(zdev); + if (rc) + return rc; + } + + if (!zdev->has_resources) { + zpci_setup_bus_resources(zdev); + for (i = 0; i < PCI_STD_NUM_BARS; i++) { + if (zdev->bars[i].res) + pci_bus_add_resource(zdev->zbus->bus, zdev->bars[i].res, 0); + } + } + + return 0; +} + +/* zpci_bus_scan_device - Scan a single device adding it to the PCI core + * @zdev: the zdev to be scanned + * + * Scans the PCI function making it available to the common PCI code. + * + * Return: 0 on success, an error value otherwise + */ +int zpci_bus_scan_device(struct zpci_dev *zdev) +{ + struct pci_dev *pdev; + int rc; + + rc = zpci_bus_prepare_device(zdev); + if (rc) + return rc; + + pdev = pci_scan_single_device(zdev->zbus->bus, zdev->devfn); + if (!pdev) + return -ENODEV; + + pci_lock_rescan_remove(); + pci_bus_add_device(pdev); + pci_unlock_rescan_remove(); + + return 0; +} + +/* zpci_bus_remove_device - Removes the given zdev from the PCI core + * @zdev: the zdev to be removed from the PCI core + * @set_error: if true the device's error state is set to permanent failure + * + * Sets a zPCI device to a configured but offline state; the zPCI + * device is still accessible through its hotplug slot and the zPCI + * API but is removed from the common code PCI bus, making it + * no longer available to drivers. + */ +void zpci_bus_remove_device(struct zpci_dev *zdev, bool set_error) +{ + struct zpci_bus *zbus = zdev->zbus; + struct pci_dev *pdev; + + if (!zdev->zbus->bus) + return; + + pdev = pci_get_slot(zbus->bus, zdev->devfn); + if (pdev) { + if (set_error) + pdev->error_state = pci_channel_io_perm_failure; + if (pdev->is_virtfn) { + zpci_iov_remove_virtfn(pdev, zdev->vfn); + /* balance pci_get_slot */ + pci_dev_put(pdev); + return; + } + pci_stop_and_remove_bus_device_locked(pdev); + /* balance pci_get_slot */ + pci_dev_put(pdev); + } +} + +/* zpci_bus_scan_bus - Scan all configured zPCI functions on the bus + * @zbus: the zbus to be scanned + * + * Enables and scans all PCI functions on the bus making them available to the + * common PCI code. If a PCI function fails to be initialized an error will be + * returned but attempts will still be made for all other functions on the bus. + * + * Return: 0 on success, an error value otherwise + */ +int zpci_bus_scan_bus(struct zpci_bus *zbus) +{ + struct zpci_dev *zdev; + int devfn, rc, ret = 0; + + for (devfn = 0; devfn < ZPCI_FUNCTIONS_PER_BUS; devfn++) { + zdev = zbus->function[devfn]; + if (zdev && zdev->state == ZPCI_FN_STATE_CONFIGURED) { + rc = zpci_bus_prepare_device(zdev); + if (rc) + ret = -EIO; + } + } + + pci_lock_rescan_remove(); + pci_scan_child_bus(zbus->bus); + pci_bus_add_devices(zbus->bus); + pci_unlock_rescan_remove(); + + return ret; +} + +/* zpci_bus_scan_busses - Scan all registered busses + * + * Scan all available zbusses + * + */ +void zpci_bus_scan_busses(void) +{ + struct zpci_bus *zbus = NULL; + + mutex_lock(&zbus_list_lock); + list_for_each_entry(zbus, &zbus_list, bus_next) { + zpci_bus_scan_bus(zbus); + cond_resched(); + } + mutex_unlock(&zbus_list_lock); +} + +/* zpci_bus_create_pci_bus - Create the PCI bus associated with this zbus + * @zbus: the zbus holding the zdevices + * @fr: PCI root function that will determine the bus's domain, and bus speeed + * @ops: the pci operations + * + * The PCI function @fr determines the domain (its UID), multifunction property + * and maximum bus speed of the entire bus. + * + * Return: 0 on success, an error code otherwise + */ +static int zpci_bus_create_pci_bus(struct zpci_bus *zbus, struct zpci_dev *fr, struct pci_ops *ops) +{ + struct pci_bus *bus; + int domain; + + domain = zpci_alloc_domain((u16)fr->uid); + if (domain < 0) + return domain; + + zbus->domain_nr = domain; + zbus->multifunction = fr->rid_available; + zbus->max_bus_speed = fr->max_bus_speed; + + /* + * Note that the zbus->resources are taken over and zbus->resources + * is empty after a successful call + */ + bus = pci_create_root_bus(NULL, ZPCI_BUS_NR, ops, zbus, &zbus->resources); + if (!bus) { + zpci_free_domain(zbus->domain_nr); + return -EFAULT; + } + + zbus->bus = bus; + + return 0; +} + +static void zpci_bus_release(struct kref *kref) +{ + struct zpci_bus *zbus = container_of(kref, struct zpci_bus, kref); + + if (zbus->bus) { + pci_lock_rescan_remove(); + pci_stop_root_bus(zbus->bus); + + zpci_free_domain(zbus->domain_nr); + pci_free_resource_list(&zbus->resources); + + pci_remove_root_bus(zbus->bus); + pci_unlock_rescan_remove(); + } + + mutex_lock(&zbus_list_lock); + list_del(&zbus->bus_next); + mutex_unlock(&zbus_list_lock); + kfree(zbus); +} + +static void zpci_bus_put(struct zpci_bus *zbus) +{ + kref_put(&zbus->kref, zpci_bus_release); +} + +static struct zpci_bus *zpci_bus_get(int pchid) +{ + struct zpci_bus *zbus; + + mutex_lock(&zbus_list_lock); + list_for_each_entry(zbus, &zbus_list, bus_next) { + if (pchid == zbus->pchid) { + kref_get(&zbus->kref); + goto out_unlock; + } + } + zbus = NULL; +out_unlock: + mutex_unlock(&zbus_list_lock); + return zbus; +} + +static struct zpci_bus *zpci_bus_alloc(int pchid) +{ + struct zpci_bus *zbus; + + zbus = kzalloc(sizeof(*zbus), GFP_KERNEL); + if (!zbus) + return NULL; + + zbus->pchid = pchid; + INIT_LIST_HEAD(&zbus->bus_next); + mutex_lock(&zbus_list_lock); + list_add_tail(&zbus->bus_next, &zbus_list); + mutex_unlock(&zbus_list_lock); + + kref_init(&zbus->kref); + INIT_LIST_HEAD(&zbus->resources); + + zbus->bus_resource.start = 0; + zbus->bus_resource.end = ZPCI_BUS_NR; + zbus->bus_resource.flags = IORESOURCE_BUS; + pci_add_resource(&zbus->resources, &zbus->bus_resource); + + return zbus; +} + +void pcibios_bus_add_device(struct pci_dev *pdev) +{ + struct zpci_dev *zdev = to_zpci(pdev); + + /* + * With pdev->no_vf_scan the common PCI probing code does not + * perform PF/VF linking. + */ + if (zdev->vfn) { + zpci_iov_setup_virtfn(zdev->zbus, pdev, zdev->vfn); + pdev->no_command_memory = 1; + } +} + +static int zpci_bus_add_device(struct zpci_bus *zbus, struct zpci_dev *zdev) +{ + int rc = -EINVAL; + + if (zbus->function[zdev->devfn]) { + pr_err("devfn %04x is already assigned\n", zdev->devfn); + return rc; + } + + zdev->zbus = zbus; + zbus->function[zdev->devfn] = zdev; + zpci_nb_devices++; + + if (zbus->multifunction && !zdev->rid_available) { + WARN_ONCE(1, "rid_available not set for multifunction\n"); + goto error; + } + rc = zpci_init_slot(zdev); + if (rc) + goto error; + zdev->has_hp_slot = 1; + + return 0; + +error: + zbus->function[zdev->devfn] = NULL; + zdev->zbus = NULL; + zpci_nb_devices--; + return rc; +} + +int zpci_bus_device_register(struct zpci_dev *zdev, struct pci_ops *ops) +{ + struct zpci_bus *zbus = NULL; + int rc = -EBADF; + + if (zpci_nb_devices == ZPCI_NR_DEVICES) { + pr_warn("Adding PCI function %08x failed because the configured limit of %d is reached\n", + zdev->fid, ZPCI_NR_DEVICES); + return -ENOSPC; + } + + if (zdev->devfn >= ZPCI_FUNCTIONS_PER_BUS) + return -EINVAL; + + if (!s390_pci_no_rid && zdev->rid_available) + zbus = zpci_bus_get(zdev->pchid); + + if (!zbus) { + zbus = zpci_bus_alloc(zdev->pchid); + if (!zbus) + return -ENOMEM; + } + + if (!zbus->bus) { + /* The UID of the first PCI function registered with a zpci_bus + * is used as the domain number for that bus. Currently there + * is exactly one zpci_bus per domain. + */ + rc = zpci_bus_create_pci_bus(zbus, zdev, ops); + if (rc) + goto error; + } + + rc = zpci_bus_add_device(zbus, zdev); + if (rc) + goto error; + + return 0; + +error: + pr_err("Adding PCI function %08x failed\n", zdev->fid); + zpci_bus_put(zbus); + return rc; +} + +void zpci_bus_device_unregister(struct zpci_dev *zdev) +{ + struct zpci_bus *zbus = zdev->zbus; + + zpci_nb_devices--; + zbus->function[zdev->devfn] = NULL; + zpci_bus_put(zbus); +} diff --git a/arch/s390/pci/pci_bus.h b/arch/s390/pci/pci_bus.h new file mode 100644 index 000000000000..af9f0ac79a1b --- /dev/null +++ b/arch/s390/pci/pci_bus.h @@ -0,0 +1,42 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +/* + * Copyright IBM Corp. 2020 + * + * Author(s): + * Pierre Morel <pmorel@linux.ibm.com> + * + */ + +int zpci_bus_device_register(struct zpci_dev *zdev, struct pci_ops *ops); +void zpci_bus_device_unregister(struct zpci_dev *zdev); + +int zpci_bus_scan_bus(struct zpci_bus *zbus); +void zpci_bus_scan_busses(void); + +int zpci_bus_scan_device(struct zpci_dev *zdev); +void zpci_bus_remove_device(struct zpci_dev *zdev, bool set_error); + +void zpci_release_device(struct kref *kref); +static inline void zpci_zdev_put(struct zpci_dev *zdev) +{ + if (zdev) + kref_put(&zdev->kref, zpci_release_device); +} + +static inline void zpci_zdev_get(struct zpci_dev *zdev) +{ + kref_get(&zdev->kref); +} + +int zpci_alloc_domain(int domain); +void zpci_free_domain(int domain); +int zpci_setup_bus_resources(struct zpci_dev *zdev); + +static inline struct zpci_dev *zdev_from_bus(struct pci_bus *bus, + unsigned int devfn) +{ + struct zpci_bus *zbus = bus->sysdata; + + return (devfn >= ZPCI_FUNCTIONS_PER_BUS) ? NULL : zbus->function[devfn]; +} + diff --git a/arch/s390/pci/pci_clp.c b/arch/s390/pci/pci_clp.c index 4c613e569fe0..ee90a91ed888 100644 --- a/arch/s390/pci/pci_clp.c +++ b/arch/s390/pci/pci_clp.c @@ -17,17 +17,20 @@ #include <linux/delay.h> #include <linux/pci.h> #include <linux/uaccess.h> +#include <asm/asm-extable.h> #include <asm/pci_debug.h> #include <asm/pci_clp.h> #include <asm/clp.h> #include <uapi/asm/clp.h> +#include "pci_bus.h" + bool zpci_unique_uid; -static void update_uid_checking(bool new) +void update_uid_checking(bool new) { if (zpci_unique_uid != new) - zpci_dbg(1, "uid checking:%d\n", new); + zpci_dbg(3, "uid checking:%d\n", new); zpci_unique_uid = new; } @@ -102,6 +105,9 @@ static void clp_store_query_pci_fngrp(struct zpci_dev *zdev, zdev->msi_addr = response->msia; zdev->max_msi = response->noi; zdev->fmb_update = response->mui; + zdev->version = response->version; + zdev->maxstbl = response->maxstbl; + zdev->dtsm = response->dtsm; switch (response->version) { case 1: @@ -155,13 +161,19 @@ static int clp_store_query_pci_fn(struct zpci_dev *zdev, zdev->pfgid = response->pfgid; zdev->pft = response->pft; zdev->vfn = response->vfn; + zdev->port = response->port; zdev->uid = response->uid; zdev->fmb_length = sizeof(u32) * response->fmb_len; + zdev->rid_available = response->rid_avail; + zdev->is_physfn = response->is_physfn; + if (!s390_pci_no_rid && zdev->rid_available) + zdev->devfn = response->rid & ZPCI_RID_MASK_DEVFN; memcpy(zdev->pfip, response->pfip, sizeof(zdev->pfip)); if (response->util_str_avail) { memcpy(zdev->util_str, response->util_str, sizeof(zdev->util_str)); + zdev->util_str_avail = 1; } zdev->mio_capable = response->mio_addr_avail; for (i = 0; i < PCI_STD_NUM_BARS; i++) { @@ -174,7 +186,7 @@ static int clp_store_query_pci_fn(struct zpci_dev *zdev, return 0; } -static int clp_query_pci_fn(struct zpci_dev *zdev, u32 fh) +int clp_query_pci_fn(struct zpci_dev *zdev) { struct clp_req_rsp_query_pci *rrb; int rc; @@ -187,7 +199,7 @@ static int clp_query_pci_fn(struct zpci_dev *zdev, u32 fh) rrb->request.hdr.len = sizeof(rrb->request); rrb->request.hdr.cmd = CLP_QUERY_PCI_FN; rrb->response.hdr.len = sizeof(rrb->response); - rrb->request.fh = fh; + rrb->request.fh = zdev->fh; rc = clp_req(rrb, CLP_LPS_PCI); if (!rc && rrb->response.hdr.rsp == CLP_RC_OK) { @@ -205,60 +217,39 @@ out: return rc; } -int clp_add_pci_device(u32 fid, u32 fh, int configured) -{ - struct zpci_dev *zdev; - int rc = -ENOMEM; - - zpci_dbg(3, "add fid:%x, fh:%x, c:%d\n", fid, fh, configured); - zdev = kzalloc(sizeof(*zdev), GFP_KERNEL); - if (!zdev) - goto error; - - zdev->fh = fh; - zdev->fid = fid; - - /* Query function properties and update zdev */ - rc = clp_query_pci_fn(zdev, fh); - if (rc) - goto error; - - if (configured) - zdev->state = ZPCI_FN_STATE_CONFIGURED; - else - zdev->state = ZPCI_FN_STATE_STANDBY; - - rc = zpci_create_device(zdev); - if (rc) - goto error; - return 0; - -error: - zpci_dbg(0, "add fid:%x, rc:%d\n", fid, rc); - kfree(zdev); - return rc; -} - -/* - * Enable/Disable a given PCI function defined by its function handle. +/** + * clp_set_pci_fn() - Execute a command on a PCI function + * @zdev: Function that will be affected + * @fh: Out parameter for updated function handle + * @nr_dma_as: DMA address space number + * @command: The command code to execute + * + * Returns: 0 on success, < 0 for Linux errors (e.g. -ENOMEM), and + * > 0 for non-success platform responses */ -static int clp_set_pci_fn(u32 *fh, u8 nr_dma_as, u8 command) +static int clp_set_pci_fn(struct zpci_dev *zdev, u32 *fh, u8 nr_dma_as, u8 command) { struct clp_req_rsp_set_pci *rrb; int rc, retries = 100; + u32 gisa = 0; + *fh = 0; rrb = clp_alloc_block(GFP_KERNEL); if (!rrb) return -ENOMEM; + if (command != CLP_SET_DISABLE_PCI_FN) + gisa = zdev->gisa; + do { memset(rrb, 0, sizeof(*rrb)); rrb->request.hdr.len = sizeof(rrb->request); rrb->request.hdr.cmd = CLP_SET_PCI_FN; rrb->response.hdr.len = sizeof(rrb->response); - rrb->request.fh = *fh; + rrb->request.fh = zdev->fh; rrb->request.oc = command; rrb->request.ndas = nr_dma_as; + rrb->request.gisa = gisa; rc = clp_req(rrb, CLP_LPS_PCI); if (rrb->response.hdr.rsp == CLP_RC_SETPCIFN_BUSY) { @@ -269,50 +260,107 @@ static int clp_set_pci_fn(u32 *fh, u8 nr_dma_as, u8 command) } } while (rrb->response.hdr.rsp == CLP_RC_SETPCIFN_BUSY); - if (!rc && rrb->response.hdr.rsp == CLP_RC_OK) + if (!rc && rrb->response.hdr.rsp == CLP_RC_OK) { *fh = rrb->response.fh; - else { + } else { zpci_err("Set PCI FN:\n"); zpci_err_clp(rrb->response.hdr.rsp, rc); - rc = -EIO; + if (!rc) + rc = rrb->response.hdr.rsp; } clp_free_block(rrb); return rc; } -int clp_enable_fh(struct zpci_dev *zdev, u8 nr_dma_as) +int clp_setup_writeback_mio(void) { - u32 fh = zdev->fh; + struct clp_req_rsp_slpc_pci *rrb; + u8 wb_bit_pos; int rc; - rc = clp_set_pci_fn(&fh, nr_dma_as, CLP_SET_ENABLE_PCI_FN); - zpci_dbg(3, "ena fid:%x, fh:%x, rc:%d\n", zdev->fid, fh, rc); - if (rc) - goto out; + rrb = clp_alloc_block(GFP_KERNEL); + if (!rrb) + return -ENOMEM; + + memset(rrb, 0, sizeof(*rrb)); + rrb->request.hdr.len = sizeof(rrb->request); + rrb->request.hdr.cmd = CLP_SLPC; + rrb->response.hdr.len = sizeof(rrb->response); + + rc = clp_req(rrb, CLP_LPS_PCI); + if (!rc && rrb->response.hdr.rsp == CLP_RC_OK) { + if (rrb->response.vwb) { + wb_bit_pos = rrb->response.mio_wb; + set_bit_inv(wb_bit_pos, &mio_wb_bit_mask); + zpci_dbg(3, "wb bit: %d\n", wb_bit_pos); + } else { + zpci_dbg(3, "wb bit: n.a.\n"); + } + + } else { + zpci_err("SLPC PCI:\n"); + zpci_err_clp(rrb->response.hdr.rsp, rc); + rc = -EIO; + } + clp_free_block(rrb); + return rc; +} + +int clp_enable_fh(struct zpci_dev *zdev, u32 *fh, u8 nr_dma_as) +{ + int rc; - zdev->fh = fh; - if (zpci_use_mio(zdev)) { - rc = clp_set_pci_fn(&fh, nr_dma_as, CLP_SET_ENABLE_MIO); - zpci_dbg(3, "ena mio fid:%x, fh:%x, rc:%d\n", zdev->fid, fh, rc); + rc = clp_set_pci_fn(zdev, fh, nr_dma_as, CLP_SET_ENABLE_PCI_FN); + zpci_dbg(3, "ena fid:%x, fh:%x, rc:%d\n", zdev->fid, *fh, rc); + if (!rc && zpci_use_mio(zdev)) { + rc = clp_set_pci_fn(zdev, fh, nr_dma_as, CLP_SET_ENABLE_MIO); + zpci_dbg(3, "ena mio fid:%x, fh:%x, rc:%d\n", + zdev->fid, *fh, rc); if (rc) - clp_disable_fh(zdev); + clp_disable_fh(zdev, fh); } -out: return rc; } -int clp_disable_fh(struct zpci_dev *zdev) +int clp_disable_fh(struct zpci_dev *zdev, u32 *fh) { - u32 fh = zdev->fh; int rc; if (!zdev_enabled(zdev)) return 0; - rc = clp_set_pci_fn(&fh, 0, CLP_SET_DISABLE_PCI_FN); - zpci_dbg(3, "dis fid:%x, fh:%x, rc:%d\n", zdev->fid, fh, rc); - if (!rc) - zdev->fh = fh; + rc = clp_set_pci_fn(zdev, fh, 0, CLP_SET_DISABLE_PCI_FN); + zpci_dbg(3, "dis fid:%x, fh:%x, rc:%d\n", zdev->fid, *fh, rc); + return rc; +} + +static int clp_list_pci_req(struct clp_req_rsp_list_pci *rrb, + u64 *resume_token, int *nentries) +{ + int rc; + + memset(rrb, 0, sizeof(*rrb)); + rrb->request.hdr.len = sizeof(rrb->request); + rrb->request.hdr.cmd = CLP_LIST_PCI; + /* store as many entries as possible */ + rrb->response.hdr.len = CLP_BLK_SIZE - LIST_PCI_HDR_LEN; + rrb->request.resume_token = *resume_token; + + /* Get PCI function handle list */ + rc = clp_req(rrb, CLP_LPS_PCI); + if (rc || rrb->response.hdr.rsp != CLP_RC_OK) { + zpci_err("List PCI FN:\n"); + zpci_err_clp(rrb->response.hdr.rsp, rc); + return -EIO; + } + + update_uid_checking(rrb->response.uid_checking); + WARN_ON_ONCE(rrb->response.entry_size != + sizeof(struct clp_fh_list_entry)); + + *nentries = (rrb->response.hdr.len - LIST_PCI_HDR_LEN) / + rrb->response.entry_size; + *resume_token = rrb->response.resume_token; return rc; } @@ -321,53 +369,43 @@ static int clp_list_pci(struct clp_req_rsp_list_pci *rrb, void *data, void (*cb)(struct clp_fh_list_entry *, void *)) { u64 resume_token = 0; - int entries, i, rc; + int nentries, i, rc; do { - memset(rrb, 0, sizeof(*rrb)); - rrb->request.hdr.len = sizeof(rrb->request); - rrb->request.hdr.cmd = CLP_LIST_PCI; - /* store as many entries as possible */ - rrb->response.hdr.len = CLP_BLK_SIZE - LIST_PCI_HDR_LEN; - rrb->request.resume_token = resume_token; - - /* Get PCI function handle list */ - rc = clp_req(rrb, CLP_LPS_PCI); - if (rc || rrb->response.hdr.rsp != CLP_RC_OK) { - zpci_err("List PCI FN:\n"); - zpci_err_clp(rrb->response.hdr.rsp, rc); - rc = -EIO; - goto out; - } - - update_uid_checking(rrb->response.uid_checking); - WARN_ON_ONCE(rrb->response.entry_size != - sizeof(struct clp_fh_list_entry)); - - entries = (rrb->response.hdr.len - LIST_PCI_HDR_LEN) / - rrb->response.entry_size; - - resume_token = rrb->response.resume_token; - for (i = 0; i < entries; i++) + rc = clp_list_pci_req(rrb, &resume_token, &nentries); + if (rc) + return rc; + for (i = 0; i < nentries; i++) cb(&rrb->response.fh_list[i], data); } while (resume_token); -out: + return rc; } -static void __clp_add(struct clp_fh_list_entry *entry, void *data) +static int clp_find_pci(struct clp_req_rsp_list_pci *rrb, u32 fid, + struct clp_fh_list_entry *entry) { - struct zpci_dev *zdev; + struct clp_fh_list_entry *fh_list; + u64 resume_token = 0; + int nentries, i, rc; - if (!entry->vendor_id) - return; + do { + rc = clp_list_pci_req(rrb, &resume_token, &nentries); + if (rc) + return rc; + fh_list = rrb->response.fh_list; + for (i = 0; i < nentries; i++) { + if (fh_list[i].fid == fid) { + *entry = fh_list[i]; + return 0; + } + } + } while (resume_token); - zdev = get_zdev_by_fid(entry->fid); - if (!zdev) - clp_add_pci_device(entry->fid, entry->fh, entry->config_state); + return -ENODEV; } -static void __clp_update(struct clp_fh_list_entry *entry, void *data) +static void __clp_add(struct clp_fh_list_entry *entry, void *data) { struct zpci_dev *zdev; @@ -375,10 +413,11 @@ static void __clp_update(struct clp_fh_list_entry *entry, void *data) return; zdev = get_zdev_by_fid(entry->fid); - if (!zdev) + if (zdev) { + zpci_zdev_put(zdev); return; - - zdev->fh = entry->fh; + } + zpci_create_device(entry->fid, entry->fh, entry->config_state); } int clp_scan_pci_devices(void) @@ -396,66 +435,44 @@ int clp_scan_pci_devices(void) return rc; } -int clp_rescan_pci_devices(void) -{ - struct clp_req_rsp_list_pci *rrb; - int rc; - - zpci_remove_reserved_devices(); - - rrb = clp_alloc_block(GFP_KERNEL); - if (!rrb) - return -ENOMEM; - - rc = clp_list_pci(rrb, NULL, __clp_add); - - clp_free_block(rrb); - return rc; -} - -int clp_rescan_pci_devices_simple(void) +/* + * Get the current function handle of the function matching @fid + */ +int clp_refresh_fh(u32 fid, u32 *fh) { struct clp_req_rsp_list_pci *rrb; + struct clp_fh_list_entry entry; int rc; rrb = clp_alloc_block(GFP_NOWAIT); if (!rrb) return -ENOMEM; - rc = clp_list_pci(rrb, NULL, __clp_update); + rc = clp_find_pci(rrb, fid, &entry); + if (!rc) + *fh = entry.fh; clp_free_block(rrb); return rc; } -struct clp_state_data { - u32 fid; - enum zpci_state state; -}; - -static void __clp_get_state(struct clp_fh_list_entry *entry, void *data) -{ - struct clp_state_data *sd = data; - - if (entry->fid != sd->fid) - return; - - sd->state = entry->config_state; -} - int clp_get_state(u32 fid, enum zpci_state *state) { struct clp_req_rsp_list_pci *rrb; - struct clp_state_data sd = {fid, ZPCI_FN_STATE_RESERVED}; + struct clp_fh_list_entry entry; int rc; rrb = clp_alloc_block(GFP_ATOMIC); if (!rrb) return -ENOMEM; - rc = clp_list_pci(rrb, &sd, __clp_get_state); - if (!rc) - *state = sd.state; + rc = clp_find_pci(rrb, fid, &entry); + if (!rc) { + *state = entry.config_state; + } else if (rc == -ENODEV) { + *state = ZPCI_FN_STATE_RESERVED; + rc = 0; + } clp_free_block(rrb); return rc; @@ -481,7 +498,7 @@ static int clp_base_command(struct clp_req *req, struct clp_req_hdr *lpcb) } } -static int clp_pci_slpc(struct clp_req *req, struct clp_req_rsp_slpc *lpcb) +static int clp_pci_slpc(struct clp_req *req, struct clp_req_rsp_slpc_pci *lpcb) { unsigned long limit = PAGE_SIZE - sizeof(lpcb->request); @@ -649,9 +666,4 @@ static struct miscdevice clp_misc_device = { .fops = &clp_misc_fops, }; -static int __init clp_misc_init(void) -{ - return misc_register(&clp_misc_device); -} - -device_initcall(clp_misc_init); +builtin_misc_device(clp_misc_device); diff --git a/arch/s390/pci/pci_debug.c b/arch/s390/pci/pci_debug.c index 3408c0df3ebf..2cb5043a997d 100644 --- a/arch/s390/pci/pci_debug.c +++ b/arch/s390/pci/pci_debug.c @@ -53,9 +53,11 @@ static char *pci_fmt3_names[] = { }; static char *pci_sw_names[] = { - "Allocated pages", "Mapped pages", "Unmapped pages", + "Global RPCITs", + "Sync Map RPCITs", + "Sync RPCITs", }; static void pci_fmb_show(struct seq_file *m, char *name[], int length, @@ -69,10 +71,14 @@ static void pci_fmb_show(struct seq_file *m, char *name[], int length, static void pci_sw_counter_show(struct seq_file *m) { - struct zpci_dev *zdev = m->private; - atomic64_t *counter = &zdev->allocated_pages; + struct zpci_iommu_ctrs *ctrs = zpci_get_iommu_ctrs(m->private); + atomic64_t *counter; int i; + if (!ctrs) + return; + + counter = &ctrs->mapped_pages; for (i = 0; i < ARRAY_SIZE(pci_sw_names); i++, counter++) seq_printf(m, "%26s:\t%llu\n", pci_sw_names[i], atomic64_read(counter)); @@ -85,9 +91,9 @@ static int pci_perf_show(struct seq_file *m, void *v) if (!zdev) return 0; - mutex_lock(&zdev->lock); + mutex_lock(&zdev->fmb_lock); if (!zdev->fmb) { - mutex_unlock(&zdev->lock); + mutex_unlock(&zdev->fmb_lock); seq_puts(m, "FMB statistics disabled\n"); return 0; } @@ -124,7 +130,7 @@ static int pci_perf_show(struct seq_file *m, void *v) } pci_sw_counter_show(m); - mutex_unlock(&zdev->lock); + mutex_unlock(&zdev->fmb_lock); return 0; } @@ -142,7 +148,7 @@ static ssize_t pci_perf_seq_write(struct file *file, const char __user *ubuf, if (rc) return rc; - mutex_lock(&zdev->lock); + mutex_lock(&zdev->fmb_lock); switch (val) { case 0: rc = zpci_fmb_disable_device(zdev); @@ -151,7 +157,7 @@ static ssize_t pci_perf_seq_write(struct file *file, const char __user *ubuf, rc = zpci_fmb_enable_device(zdev); break; } - mutex_unlock(&zdev->lock); + mutex_unlock(&zdev->fmb_lock); return rc ? rc : count; } @@ -196,7 +202,7 @@ int __init zpci_debug_init(void) if (!pci_debug_err_id) return -EINVAL; debug_register_view(pci_debug_err_id, &debug_hex_ascii_view); - debug_set_level(pci_debug_err_id, 6); + debug_set_level(pci_debug_err_id, 3); debugfs_root = debugfs_create_dir("pci", NULL); return 0; diff --git a/arch/s390/pci/pci_dma.c b/arch/s390/pci/pci_dma.c deleted file mode 100644 index 64b1399a73f0..000000000000 --- a/arch/s390/pci/pci_dma.c +++ /dev/null @@ -1,684 +0,0 @@ -// SPDX-License-Identifier: GPL-2.0 -/* - * Copyright IBM Corp. 2012 - * - * Author(s): - * Jan Glauber <jang@linux.vnet.ibm.com> - */ - -#include <linux/kernel.h> -#include <linux/slab.h> -#include <linux/export.h> -#include <linux/iommu-helper.h> -#include <linux/dma-mapping.h> -#include <linux/vmalloc.h> -#include <linux/pci.h> -#include <asm/pci_dma.h> - -static struct kmem_cache *dma_region_table_cache; -static struct kmem_cache *dma_page_table_cache; -static int s390_iommu_strict; - -static int zpci_refresh_global(struct zpci_dev *zdev) -{ - return zpci_refresh_trans((u64) zdev->fh << 32, zdev->start_dma, - zdev->iommu_pages * PAGE_SIZE); -} - -unsigned long *dma_alloc_cpu_table(void) -{ - unsigned long *table, *entry; - - table = kmem_cache_alloc(dma_region_table_cache, GFP_ATOMIC); - if (!table) - return NULL; - - for (entry = table; entry < table + ZPCI_TABLE_ENTRIES; entry++) - *entry = ZPCI_TABLE_INVALID; - return table; -} - -static void dma_free_cpu_table(void *table) -{ - kmem_cache_free(dma_region_table_cache, table); -} - -static unsigned long *dma_alloc_page_table(void) -{ - unsigned long *table, *entry; - - table = kmem_cache_alloc(dma_page_table_cache, GFP_ATOMIC); - if (!table) - return NULL; - - for (entry = table; entry < table + ZPCI_PT_ENTRIES; entry++) - *entry = ZPCI_PTE_INVALID; - return table; -} - -static void dma_free_page_table(void *table) -{ - kmem_cache_free(dma_page_table_cache, table); -} - -static unsigned long *dma_get_seg_table_origin(unsigned long *entry) -{ - unsigned long *sto; - - if (reg_entry_isvalid(*entry)) - sto = get_rt_sto(*entry); - else { - sto = dma_alloc_cpu_table(); - if (!sto) - return NULL; - - set_rt_sto(entry, sto); - validate_rt_entry(entry); - entry_clr_protected(entry); - } - return sto; -} - -static unsigned long *dma_get_page_table_origin(unsigned long *entry) -{ - unsigned long *pto; - - if (reg_entry_isvalid(*entry)) - pto = get_st_pto(*entry); - else { - pto = dma_alloc_page_table(); - if (!pto) - return NULL; - set_st_pto(entry, pto); - validate_st_entry(entry); - entry_clr_protected(entry); - } - return pto; -} - -unsigned long *dma_walk_cpu_trans(unsigned long *rto, dma_addr_t dma_addr) -{ - unsigned long *sto, *pto; - unsigned int rtx, sx, px; - - rtx = calc_rtx(dma_addr); - sto = dma_get_seg_table_origin(&rto[rtx]); - if (!sto) - return NULL; - - sx = calc_sx(dma_addr); - pto = dma_get_page_table_origin(&sto[sx]); - if (!pto) - return NULL; - - px = calc_px(dma_addr); - return &pto[px]; -} - -void dma_update_cpu_trans(unsigned long *entry, void *page_addr, int flags) -{ - if (flags & ZPCI_PTE_INVALID) { - invalidate_pt_entry(entry); - } else { - set_pt_pfaa(entry, page_addr); - validate_pt_entry(entry); - } - - if (flags & ZPCI_TABLE_PROTECTED) - entry_set_protected(entry); - else - entry_clr_protected(entry); -} - -static int __dma_update_trans(struct zpci_dev *zdev, unsigned long pa, - dma_addr_t dma_addr, size_t size, int flags) -{ - unsigned int nr_pages = PAGE_ALIGN(size) >> PAGE_SHIFT; - u8 *page_addr = (u8 *) (pa & PAGE_MASK); - unsigned long irq_flags; - unsigned long *entry; - int i, rc = 0; - - if (!nr_pages) - return -EINVAL; - - spin_lock_irqsave(&zdev->dma_table_lock, irq_flags); - if (!zdev->dma_table) { - rc = -EINVAL; - goto out_unlock; - } - - for (i = 0; i < nr_pages; i++) { - entry = dma_walk_cpu_trans(zdev->dma_table, dma_addr); - if (!entry) { - rc = -ENOMEM; - goto undo_cpu_trans; - } - dma_update_cpu_trans(entry, page_addr, flags); - page_addr += PAGE_SIZE; - dma_addr += PAGE_SIZE; - } - -undo_cpu_trans: - if (rc && ((flags & ZPCI_PTE_VALID_MASK) == ZPCI_PTE_VALID)) { - flags = ZPCI_PTE_INVALID; - while (i-- > 0) { - page_addr -= PAGE_SIZE; - dma_addr -= PAGE_SIZE; - entry = dma_walk_cpu_trans(zdev->dma_table, dma_addr); - if (!entry) - break; - dma_update_cpu_trans(entry, page_addr, flags); - } - } -out_unlock: - spin_unlock_irqrestore(&zdev->dma_table_lock, irq_flags); - return rc; -} - -static int __dma_purge_tlb(struct zpci_dev *zdev, dma_addr_t dma_addr, - size_t size, int flags) -{ - unsigned long irqflags; - int ret; - - /* - * With zdev->tlb_refresh == 0, rpcit is not required to establish new - * translations when previously invalid translation-table entries are - * validated. With lazy unmap, rpcit is skipped for previously valid - * entries, but a global rpcit is then required before any address can - * be re-used, i.e. after each iommu bitmap wrap-around. - */ - if ((flags & ZPCI_PTE_VALID_MASK) == ZPCI_PTE_VALID) { - if (!zdev->tlb_refresh) - return 0; - } else { - if (!s390_iommu_strict) - return 0; - } - - ret = zpci_refresh_trans((u64) zdev->fh << 32, dma_addr, - PAGE_ALIGN(size)); - if (ret == -ENOMEM && !s390_iommu_strict) { - /* enable the hypervisor to free some resources */ - if (zpci_refresh_global(zdev)) - goto out; - - spin_lock_irqsave(&zdev->iommu_bitmap_lock, irqflags); - bitmap_andnot(zdev->iommu_bitmap, zdev->iommu_bitmap, - zdev->lazy_bitmap, zdev->iommu_pages); - bitmap_zero(zdev->lazy_bitmap, zdev->iommu_pages); - spin_unlock_irqrestore(&zdev->iommu_bitmap_lock, irqflags); - ret = 0; - } -out: - return ret; -} - -static int dma_update_trans(struct zpci_dev *zdev, unsigned long pa, - dma_addr_t dma_addr, size_t size, int flags) -{ - int rc; - - rc = __dma_update_trans(zdev, pa, dma_addr, size, flags); - if (rc) - return rc; - - rc = __dma_purge_tlb(zdev, dma_addr, size, flags); - if (rc && ((flags & ZPCI_PTE_VALID_MASK) == ZPCI_PTE_VALID)) - __dma_update_trans(zdev, pa, dma_addr, size, ZPCI_PTE_INVALID); - - return rc; -} - -void dma_free_seg_table(unsigned long entry) -{ - unsigned long *sto = get_rt_sto(entry); - int sx; - - for (sx = 0; sx < ZPCI_TABLE_ENTRIES; sx++) - if (reg_entry_isvalid(sto[sx])) - dma_free_page_table(get_st_pto(sto[sx])); - - dma_free_cpu_table(sto); -} - -void dma_cleanup_tables(unsigned long *table) -{ - int rtx; - - if (!table) - return; - - for (rtx = 0; rtx < ZPCI_TABLE_ENTRIES; rtx++) - if (reg_entry_isvalid(table[rtx])) - dma_free_seg_table(table[rtx]); - - dma_free_cpu_table(table); -} - -static unsigned long __dma_alloc_iommu(struct device *dev, - unsigned long start, int size) -{ - struct zpci_dev *zdev = to_zpci(to_pci_dev(dev)); - unsigned long boundary_size; - - boundary_size = ALIGN(dma_get_seg_boundary(dev) + 1, - PAGE_SIZE) >> PAGE_SHIFT; - return iommu_area_alloc(zdev->iommu_bitmap, zdev->iommu_pages, - start, size, zdev->start_dma >> PAGE_SHIFT, - boundary_size, 0); -} - -static dma_addr_t dma_alloc_address(struct device *dev, int size) -{ - struct zpci_dev *zdev = to_zpci(to_pci_dev(dev)); - unsigned long offset, flags; - - spin_lock_irqsave(&zdev->iommu_bitmap_lock, flags); - offset = __dma_alloc_iommu(dev, zdev->next_bit, size); - if (offset == -1) { - if (!s390_iommu_strict) { - /* global flush before DMA addresses are reused */ - if (zpci_refresh_global(zdev)) - goto out_error; - - bitmap_andnot(zdev->iommu_bitmap, zdev->iommu_bitmap, - zdev->lazy_bitmap, zdev->iommu_pages); - bitmap_zero(zdev->lazy_bitmap, zdev->iommu_pages); - } - /* wrap-around */ - offset = __dma_alloc_iommu(dev, 0, size); - if (offset == -1) - goto out_error; - } - zdev->next_bit = offset + size; - spin_unlock_irqrestore(&zdev->iommu_bitmap_lock, flags); - - return zdev->start_dma + offset * PAGE_SIZE; - -out_error: - spin_unlock_irqrestore(&zdev->iommu_bitmap_lock, flags); - return DMA_MAPPING_ERROR; -} - -static void dma_free_address(struct device *dev, dma_addr_t dma_addr, int size) -{ - struct zpci_dev *zdev = to_zpci(to_pci_dev(dev)); - unsigned long flags, offset; - - offset = (dma_addr - zdev->start_dma) >> PAGE_SHIFT; - - spin_lock_irqsave(&zdev->iommu_bitmap_lock, flags); - if (!zdev->iommu_bitmap) - goto out; - - if (s390_iommu_strict) - bitmap_clear(zdev->iommu_bitmap, offset, size); - else - bitmap_set(zdev->lazy_bitmap, offset, size); - -out: - spin_unlock_irqrestore(&zdev->iommu_bitmap_lock, flags); -} - -static inline void zpci_err_dma(unsigned long rc, unsigned long addr) -{ - struct { - unsigned long rc; - unsigned long addr; - } __packed data = {rc, addr}; - - zpci_err_hex(&data, sizeof(data)); -} - -static dma_addr_t s390_dma_map_pages(struct device *dev, struct page *page, - unsigned long offset, size_t size, - enum dma_data_direction direction, - unsigned long attrs) -{ - struct zpci_dev *zdev = to_zpci(to_pci_dev(dev)); - unsigned long pa = page_to_phys(page) + offset; - int flags = ZPCI_PTE_VALID; - unsigned long nr_pages; - dma_addr_t dma_addr; - int ret; - - /* This rounds up number of pages based on size and offset */ - nr_pages = iommu_num_pages(pa, size, PAGE_SIZE); - dma_addr = dma_alloc_address(dev, nr_pages); - if (dma_addr == DMA_MAPPING_ERROR) { - ret = -ENOSPC; - goto out_err; - } - - /* Use rounded up size */ - size = nr_pages * PAGE_SIZE; - - if (direction == DMA_NONE || direction == DMA_TO_DEVICE) - flags |= ZPCI_TABLE_PROTECTED; - - ret = dma_update_trans(zdev, pa, dma_addr, size, flags); - if (ret) - goto out_free; - - atomic64_add(nr_pages, &zdev->mapped_pages); - return dma_addr + (offset & ~PAGE_MASK); - -out_free: - dma_free_address(dev, dma_addr, nr_pages); -out_err: - zpci_err("map error:\n"); - zpci_err_dma(ret, pa); - return DMA_MAPPING_ERROR; -} - -static void s390_dma_unmap_pages(struct device *dev, dma_addr_t dma_addr, - size_t size, enum dma_data_direction direction, - unsigned long attrs) -{ - struct zpci_dev *zdev = to_zpci(to_pci_dev(dev)); - int npages, ret; - - npages = iommu_num_pages(dma_addr, size, PAGE_SIZE); - dma_addr = dma_addr & PAGE_MASK; - ret = dma_update_trans(zdev, 0, dma_addr, npages * PAGE_SIZE, - ZPCI_PTE_INVALID); - if (ret) { - zpci_err("unmap error:\n"); - zpci_err_dma(ret, dma_addr); - return; - } - - atomic64_add(npages, &zdev->unmapped_pages); - dma_free_address(dev, dma_addr, npages); -} - -static void *s390_dma_alloc(struct device *dev, size_t size, - dma_addr_t *dma_handle, gfp_t flag, - unsigned long attrs) -{ - struct zpci_dev *zdev = to_zpci(to_pci_dev(dev)); - struct page *page; - unsigned long pa; - dma_addr_t map; - - size = PAGE_ALIGN(size); - page = alloc_pages(flag | __GFP_ZERO, get_order(size)); - if (!page) - return NULL; - - pa = page_to_phys(page); - map = s390_dma_map_pages(dev, page, 0, size, DMA_BIDIRECTIONAL, 0); - if (dma_mapping_error(dev, map)) { - free_pages(pa, get_order(size)); - return NULL; - } - - atomic64_add(size / PAGE_SIZE, &zdev->allocated_pages); - if (dma_handle) - *dma_handle = map; - return (void *) pa; -} - -static void s390_dma_free(struct device *dev, size_t size, - void *pa, dma_addr_t dma_handle, - unsigned long attrs) -{ - struct zpci_dev *zdev = to_zpci(to_pci_dev(dev)); - - size = PAGE_ALIGN(size); - atomic64_sub(size / PAGE_SIZE, &zdev->allocated_pages); - s390_dma_unmap_pages(dev, dma_handle, size, DMA_BIDIRECTIONAL, 0); - free_pages((unsigned long) pa, get_order(size)); -} - -/* Map a segment into a contiguous dma address area */ -static int __s390_dma_map_sg(struct device *dev, struct scatterlist *sg, - size_t size, dma_addr_t *handle, - enum dma_data_direction dir) -{ - unsigned long nr_pages = PAGE_ALIGN(size) >> PAGE_SHIFT; - struct zpci_dev *zdev = to_zpci(to_pci_dev(dev)); - dma_addr_t dma_addr_base, dma_addr; - int flags = ZPCI_PTE_VALID; - struct scatterlist *s; - unsigned long pa = 0; - int ret; - - dma_addr_base = dma_alloc_address(dev, nr_pages); - if (dma_addr_base == DMA_MAPPING_ERROR) - return -ENOMEM; - - dma_addr = dma_addr_base; - if (dir == DMA_NONE || dir == DMA_TO_DEVICE) - flags |= ZPCI_TABLE_PROTECTED; - - for (s = sg; dma_addr < dma_addr_base + size; s = sg_next(s)) { - pa = page_to_phys(sg_page(s)); - ret = __dma_update_trans(zdev, pa, dma_addr, - s->offset + s->length, flags); - if (ret) - goto unmap; - - dma_addr += s->offset + s->length; - } - ret = __dma_purge_tlb(zdev, dma_addr_base, size, flags); - if (ret) - goto unmap; - - *handle = dma_addr_base; - atomic64_add(nr_pages, &zdev->mapped_pages); - - return ret; - -unmap: - dma_update_trans(zdev, 0, dma_addr_base, dma_addr - dma_addr_base, - ZPCI_PTE_INVALID); - dma_free_address(dev, dma_addr_base, nr_pages); - zpci_err("map error:\n"); - zpci_err_dma(ret, pa); - return ret; -} - -static int s390_dma_map_sg(struct device *dev, struct scatterlist *sg, - int nr_elements, enum dma_data_direction dir, - unsigned long attrs) -{ - struct scatterlist *s = sg, *start = sg, *dma = sg; - unsigned int max = dma_get_max_seg_size(dev); - unsigned int size = s->offset + s->length; - unsigned int offset = s->offset; - int count = 0, i; - - for (i = 1; i < nr_elements; i++) { - s = sg_next(s); - - s->dma_address = DMA_MAPPING_ERROR; - s->dma_length = 0; - - if (s->offset || (size & ~PAGE_MASK) || - size + s->length > max) { - if (__s390_dma_map_sg(dev, start, size, - &dma->dma_address, dir)) - goto unmap; - - dma->dma_address += offset; - dma->dma_length = size - offset; - - size = offset = s->offset; - start = s; - dma = sg_next(dma); - count++; - } - size += s->length; - } - if (__s390_dma_map_sg(dev, start, size, &dma->dma_address, dir)) - goto unmap; - - dma->dma_address += offset; - dma->dma_length = size - offset; - - return count + 1; -unmap: - for_each_sg(sg, s, count, i) - s390_dma_unmap_pages(dev, sg_dma_address(s), sg_dma_len(s), - dir, attrs); - - return 0; -} - -static void s390_dma_unmap_sg(struct device *dev, struct scatterlist *sg, - int nr_elements, enum dma_data_direction dir, - unsigned long attrs) -{ - struct scatterlist *s; - int i; - - for_each_sg(sg, s, nr_elements, i) { - if (s->dma_length) - s390_dma_unmap_pages(dev, s->dma_address, s->dma_length, - dir, attrs); - s->dma_address = 0; - s->dma_length = 0; - } -} - -int zpci_dma_init_device(struct zpci_dev *zdev) -{ - int rc; - - /* - * At this point, if the device is part of an IOMMU domain, this would - * be a strong hint towards a bug in the IOMMU API (common) code and/or - * simultaneous access via IOMMU and DMA API. So let's issue a warning. - */ - WARN_ON(zdev->s390_domain); - - spin_lock_init(&zdev->iommu_bitmap_lock); - spin_lock_init(&zdev->dma_table_lock); - - zdev->dma_table = dma_alloc_cpu_table(); - if (!zdev->dma_table) { - rc = -ENOMEM; - goto out; - } - - /* - * Restrict the iommu bitmap size to the minimum of the following: - * - main memory size - * - 3-level pagetable address limit minus start_dma offset - * - DMA address range allowed by the hardware (clp query pci fn) - * - * Also set zdev->end_dma to the actual end address of the usable - * range, instead of the theoretical maximum as reported by hardware. - */ - zdev->start_dma = PAGE_ALIGN(zdev->start_dma); - zdev->iommu_size = min3((u64) high_memory, - ZPCI_TABLE_SIZE_RT - zdev->start_dma, - zdev->end_dma - zdev->start_dma + 1); - zdev->end_dma = zdev->start_dma + zdev->iommu_size - 1; - zdev->iommu_pages = zdev->iommu_size >> PAGE_SHIFT; - zdev->iommu_bitmap = vzalloc(zdev->iommu_pages / 8); - if (!zdev->iommu_bitmap) { - rc = -ENOMEM; - goto free_dma_table; - } - if (!s390_iommu_strict) { - zdev->lazy_bitmap = vzalloc(zdev->iommu_pages / 8); - if (!zdev->lazy_bitmap) { - rc = -ENOMEM; - goto free_bitmap; - } - - } - rc = zpci_register_ioat(zdev, 0, zdev->start_dma, zdev->end_dma, - (u64) zdev->dma_table); - if (rc) - goto free_bitmap; - - return 0; -free_bitmap: - vfree(zdev->iommu_bitmap); - zdev->iommu_bitmap = NULL; - vfree(zdev->lazy_bitmap); - zdev->lazy_bitmap = NULL; -free_dma_table: - dma_free_cpu_table(zdev->dma_table); - zdev->dma_table = NULL; -out: - return rc; -} - -void zpci_dma_exit_device(struct zpci_dev *zdev) -{ - /* - * At this point, if the device is part of an IOMMU domain, this would - * be a strong hint towards a bug in the IOMMU API (common) code and/or - * simultaneous access via IOMMU and DMA API. So let's issue a warning. - */ - WARN_ON(zdev->s390_domain); - - if (zpci_unregister_ioat(zdev, 0)) - return; - - dma_cleanup_tables(zdev->dma_table); - zdev->dma_table = NULL; - vfree(zdev->iommu_bitmap); - zdev->iommu_bitmap = NULL; - vfree(zdev->lazy_bitmap); - zdev->lazy_bitmap = NULL; - - zdev->next_bit = 0; -} - -static int __init dma_alloc_cpu_table_caches(void) -{ - dma_region_table_cache = kmem_cache_create("PCI_DMA_region_tables", - ZPCI_TABLE_SIZE, ZPCI_TABLE_ALIGN, - 0, NULL); - if (!dma_region_table_cache) - return -ENOMEM; - - dma_page_table_cache = kmem_cache_create("PCI_DMA_page_tables", - ZPCI_PT_SIZE, ZPCI_PT_ALIGN, - 0, NULL); - if (!dma_page_table_cache) { - kmem_cache_destroy(dma_region_table_cache); - return -ENOMEM; - } - return 0; -} - -int __init zpci_dma_init(void) -{ - return dma_alloc_cpu_table_caches(); -} - -void zpci_dma_exit(void) -{ - kmem_cache_destroy(dma_page_table_cache); - kmem_cache_destroy(dma_region_table_cache); -} - -const struct dma_map_ops s390_pci_dma_ops = { - .alloc = s390_dma_alloc, - .free = s390_dma_free, - .map_sg = s390_dma_map_sg, - .unmap_sg = s390_dma_unmap_sg, - .map_page = s390_dma_map_pages, - .unmap_page = s390_dma_unmap_pages, - .mmap = dma_common_mmap, - .get_sgtable = dma_common_get_sgtable, - /* dma_supported is unconditionally true without a callback */ -}; -EXPORT_SYMBOL_GPL(s390_pci_dma_ops); - -static int __init s390_iommu_setup(char *str) -{ - if (!strcmp(str, "strict")) - s390_iommu_strict = 1; - return 1; -} - -__setup("s390_iommu=", s390_iommu_setup); diff --git a/arch/s390/pci/pci_event.c b/arch/s390/pci/pci_event.c index 8d6ee4af4230..dbe95ec5917e 100644 --- a/arch/s390/pci/pci_event.c +++ b/arch/s390/pci/pci_event.c @@ -12,8 +12,11 @@ #include <linux/kernel.h> #include <linux/pci.h> #include <asm/pci_debug.h> +#include <asm/pci_dma.h> #include <asm/sclp.h> +#include "pci_bus.h" + /* Content Code Description for PCI Function Error */ struct zpci_ccdf_err { u32 reserved1; @@ -44,25 +47,257 @@ struct zpci_ccdf_avail { u16 pec; /* PCI event code */ } __packed; +static inline bool ers_result_indicates_abort(pci_ers_result_t ers_res) +{ + switch (ers_res) { + case PCI_ERS_RESULT_CAN_RECOVER: + case PCI_ERS_RESULT_RECOVERED: + case PCI_ERS_RESULT_NEED_RESET: + return false; + default: + return true; + } +} + +static bool is_passed_through(struct pci_dev *pdev) +{ + struct zpci_dev *zdev = to_zpci(pdev); + bool ret; + + mutex_lock(&zdev->kzdev_lock); + ret = !!zdev->kzdev; + mutex_unlock(&zdev->kzdev_lock); + + return ret; +} + +static bool is_driver_supported(struct pci_driver *driver) +{ + if (!driver || !driver->err_handler) + return false; + if (!driver->err_handler->error_detected) + return false; + if (!driver->err_handler->slot_reset) + return false; + if (!driver->err_handler->resume) + return false; + return true; +} + +static pci_ers_result_t zpci_event_notify_error_detected(struct pci_dev *pdev, + struct pci_driver *driver) +{ + pci_ers_result_t ers_res = PCI_ERS_RESULT_DISCONNECT; + + ers_res = driver->err_handler->error_detected(pdev, pdev->error_state); + if (ers_result_indicates_abort(ers_res)) + pr_info("%s: Automatic recovery failed after initial reporting\n", pci_name(pdev)); + else if (ers_res == PCI_ERS_RESULT_NEED_RESET) + pr_debug("%s: Driver needs reset to recover\n", pci_name(pdev)); + + return ers_res; +} + +static pci_ers_result_t zpci_event_do_error_state_clear(struct pci_dev *pdev, + struct pci_driver *driver) +{ + pci_ers_result_t ers_res = PCI_ERS_RESULT_DISCONNECT; + struct zpci_dev *zdev = to_zpci(pdev); + int rc; + + pr_info("%s: Unblocking device access for examination\n", pci_name(pdev)); + rc = zpci_reset_load_store_blocked(zdev); + if (rc) { + pr_err("%s: Unblocking device access failed\n", pci_name(pdev)); + /* Let's try a full reset instead */ + return PCI_ERS_RESULT_NEED_RESET; + } + + if (driver->err_handler->mmio_enabled) { + ers_res = driver->err_handler->mmio_enabled(pdev); + if (ers_result_indicates_abort(ers_res)) { + pr_info("%s: Automatic recovery failed after MMIO re-enable\n", + pci_name(pdev)); + return ers_res; + } else if (ers_res == PCI_ERS_RESULT_NEED_RESET) { + pr_debug("%s: Driver needs reset to recover\n", pci_name(pdev)); + return ers_res; + } + } + + pr_debug("%s: Unblocking DMA\n", pci_name(pdev)); + rc = zpci_clear_error_state(zdev); + if (!rc) { + pdev->error_state = pci_channel_io_normal; + } else { + pr_err("%s: Unblocking DMA failed\n", pci_name(pdev)); + /* Let's try a full reset instead */ + return PCI_ERS_RESULT_NEED_RESET; + } + + return ers_res; +} + +static pci_ers_result_t zpci_event_do_reset(struct pci_dev *pdev, + struct pci_driver *driver) +{ + pci_ers_result_t ers_res = PCI_ERS_RESULT_DISCONNECT; + + pr_info("%s: Initiating reset\n", pci_name(pdev)); + if (zpci_hot_reset_device(to_zpci(pdev))) { + pr_err("%s: The reset request failed\n", pci_name(pdev)); + return ers_res; + } + pdev->error_state = pci_channel_io_normal; + ers_res = driver->err_handler->slot_reset(pdev); + if (ers_result_indicates_abort(ers_res)) { + pr_info("%s: Automatic recovery failed after slot reset\n", pci_name(pdev)); + return ers_res; + } + + return ers_res; +} + +/* zpci_event_attempt_error_recovery - Try to recover the given PCI function + * @pdev: PCI function to recover currently in the error state + * + * We follow the scheme outlined in Documentation/PCI/pci-error-recovery.rst. + * With the simplification that recovery always happens per function + * and the platform determines which functions are affected for + * multi-function devices. + */ +static pci_ers_result_t zpci_event_attempt_error_recovery(struct pci_dev *pdev) +{ + pci_ers_result_t ers_res = PCI_ERS_RESULT_DISCONNECT; + struct pci_driver *driver; + + /* + * Ensure that the PCI function is not removed concurrently, no driver + * is unbound or probed and that userspace can't access its + * configuration space while we perform recovery. + */ + pci_dev_lock(pdev); + if (pdev->error_state == pci_channel_io_perm_failure) { + ers_res = PCI_ERS_RESULT_DISCONNECT; + goto out_unlock; + } + pdev->error_state = pci_channel_io_frozen; + + if (is_passed_through(pdev)) { + pr_info("%s: Cannot be recovered in the host because it is a pass-through device\n", + pci_name(pdev)); + goto out_unlock; + } + + driver = to_pci_driver(pdev->dev.driver); + if (!is_driver_supported(driver)) { + if (!driver) + pr_info("%s: Cannot be recovered because no driver is bound to the device\n", + pci_name(pdev)); + else + pr_info("%s: The %s driver bound to the device does not support error recovery\n", + pci_name(pdev), + driver->name); + goto out_unlock; + } + + ers_res = zpci_event_notify_error_detected(pdev, driver); + if (ers_result_indicates_abort(ers_res)) + goto out_unlock; + + if (ers_res == PCI_ERS_RESULT_CAN_RECOVER) { + ers_res = zpci_event_do_error_state_clear(pdev, driver); + if (ers_result_indicates_abort(ers_res)) + goto out_unlock; + } + + if (ers_res == PCI_ERS_RESULT_NEED_RESET) + ers_res = zpci_event_do_reset(pdev, driver); + + if (ers_res != PCI_ERS_RESULT_RECOVERED) { + pr_err("%s: Automatic recovery failed; operator intervention is required\n", + pci_name(pdev)); + goto out_unlock; + } + + pr_info("%s: The device is ready to resume operations\n", pci_name(pdev)); + if (driver->err_handler->resume) + driver->err_handler->resume(pdev); +out_unlock: + pci_dev_unlock(pdev); + + return ers_res; +} + +/* zpci_event_io_failure - Report PCI channel failure state to driver + * @pdev: PCI function for which to report + * @es: PCI channel failure state to report + */ +static void zpci_event_io_failure(struct pci_dev *pdev, pci_channel_state_t es) +{ + struct pci_driver *driver; + + pci_dev_lock(pdev); + pdev->error_state = es; + /** + * While vfio-pci's error_detected callback notifies user-space QEMU + * reacts to this by freezing the guest. In an s390 environment PCI + * errors are rarely fatal so this is overkill. Instead in the future + * we will inject the error event and let the guest recover the device + * itself. + */ + if (is_passed_through(pdev)) + goto out; + driver = to_pci_driver(pdev->dev.driver); + if (driver && driver->err_handler && driver->err_handler->error_detected) + driver->err_handler->error_detected(pdev, pdev->error_state); +out: + pci_dev_unlock(pdev); +} + static void __zpci_event_error(struct zpci_ccdf_err *ccdf) { struct zpci_dev *zdev = get_zdev_by_fid(ccdf->fid); struct pci_dev *pdev = NULL; + pci_ers_result_t ers_res; + zpci_dbg(3, "err fid:%x, fh:%x, pec:%x\n", + ccdf->fid, ccdf->fh, ccdf->pec); zpci_err("error CCDF:\n"); zpci_err_hex(ccdf, sizeof(*ccdf)); - if (zdev) - pdev = pci_get_slot(zdev->bus, ZPCI_DEVFN); + if (zdev) { + mutex_lock(&zdev->state_lock); + zpci_update_fh(zdev, ccdf->fh); + if (zdev->zbus->bus) + pdev = pci_get_slot(zdev->zbus->bus, zdev->devfn); + } pr_err("%s: Event 0x%x reports an error for PCI function 0x%x\n", pdev ? pci_name(pdev) : "n/a", ccdf->pec, ccdf->fid); if (!pdev) - return; + goto no_pdev; - pdev->error_state = pci_channel_io_perm_failure; + switch (ccdf->pec) { + case 0x003a: /* Service Action or Error Recovery Successful */ + ers_res = zpci_event_attempt_error_recovery(pdev); + if (ers_res != PCI_ERS_RESULT_RECOVERED) + zpci_event_io_failure(pdev, pci_channel_io_perm_failure); + break; + default: + /* + * Mark as frozen not permanently failed because the device + * could be subsequently recovered by the platform. + */ + zpci_event_io_failure(pdev, pci_channel_io_frozen); + break; + } pci_dev_put(pdev); +no_pdev: + if (zdev) + mutex_unlock(&zdev->state_lock); + zpci_zdev_put(zdev); } void zpci_event_error(void *data) @@ -71,90 +306,94 @@ void zpci_event_error(void *data) __zpci_event_error(data); } +static void zpci_event_hard_deconfigured(struct zpci_dev *zdev, u32 fh) +{ + zpci_update_fh(zdev, fh); + /* Give the driver a hint that the function is + * already unusable. + */ + zpci_bus_remove_device(zdev, true); + /* Even though the device is already gone we still + * need to free zPCI resources as part of the disable. + */ + if (zdev_enabled(zdev)) + zpci_disable_device(zdev); + zdev->state = ZPCI_FN_STATE_STANDBY; +} + static void __zpci_event_availability(struct zpci_ccdf_avail *ccdf) { struct zpci_dev *zdev = get_zdev_by_fid(ccdf->fid); - struct pci_dev *pdev = NULL; + bool existing_zdev = !!zdev; enum zpci_state state; - int ret; - if (zdev) - pdev = pci_get_slot(zdev->bus, ZPCI_DEVFN); + zpci_dbg(3, "avl fid:%x, fh:%x, pec:%x\n", + ccdf->fid, ccdf->fh, ccdf->pec); - pr_info("%s: Event 0x%x reconfigured PCI function 0x%x\n", - pdev ? pci_name(pdev) : "n/a", ccdf->pec, ccdf->fid); - zpci_err("avail CCDF:\n"); - zpci_err_hex(ccdf, sizeof(*ccdf)); + if (existing_zdev) + mutex_lock(&zdev->state_lock); switch (ccdf->pec) { case 0x0301: /* Reserved|Standby -> Configured */ if (!zdev) { - ret = clp_add_pci_device(ccdf->fid, ccdf->fh, 0); - if (ret) + zdev = zpci_create_device(ccdf->fid, ccdf->fh, ZPCI_FN_STATE_CONFIGURED); + if (IS_ERR(zdev)) + break; + } else { + /* the configuration request may be stale */ + if (zdev->state != ZPCI_FN_STATE_STANDBY) break; - zdev = get_zdev_by_fid(ccdf->fid); + zdev->state = ZPCI_FN_STATE_CONFIGURED; } - if (!zdev || zdev->state != ZPCI_FN_STATE_STANDBY) - break; - zdev->state = ZPCI_FN_STATE_CONFIGURED; - zdev->fh = ccdf->fh; - ret = zpci_enable_device(zdev); - if (ret) - break; - pci_lock_rescan_remove(); - pci_rescan_bus(zdev->bus); - pci_unlock_rescan_remove(); + zpci_scan_configured_device(zdev, ccdf->fh); break; case 0x0302: /* Reserved -> Standby */ if (!zdev) - clp_add_pci_device(ccdf->fid, ccdf->fh, 0); + zpci_create_device(ccdf->fid, ccdf->fh, ZPCI_FN_STATE_STANDBY); + else + zpci_update_fh(zdev, ccdf->fh); break; case 0x0303: /* Deconfiguration requested */ - if (!zdev) - break; - if (pdev) - pci_stop_and_remove_bus_device_locked(pdev); - - ret = zpci_disable_device(zdev); - if (ret) - break; - - ret = sclp_pci_deconfigure(zdev->fid); - zpci_dbg(3, "deconf fid:%x, rc:%d\n", zdev->fid, ret); - if (!ret) - zdev->state = ZPCI_FN_STATE_STANDBY; - + if (zdev) { + /* The event may have been queued before we configured + * the device. + */ + if (zdev->state != ZPCI_FN_STATE_CONFIGURED) + break; + zpci_update_fh(zdev, ccdf->fh); + zpci_deconfigure_device(zdev); + } break; case 0x0304: /* Configured -> Standby|Reserved */ - if (!zdev) - break; - if (pdev) { - /* Give the driver a hint that the function is - * already unusable. */ - pdev->error_state = pci_channel_io_perm_failure; - pci_stop_and_remove_bus_device_locked(pdev); - } - - zdev->fh = ccdf->fh; - zpci_disable_device(zdev); - zdev->state = ZPCI_FN_STATE_STANDBY; - if (!clp_get_state(ccdf->fid, &state) && - state == ZPCI_FN_STATE_RESERVED) { - zpci_remove_device(zdev); + if (zdev) { + /* The event may have been queued before we configured + * the device.: + */ + if (zdev->state == ZPCI_FN_STATE_CONFIGURED) + zpci_event_hard_deconfigured(zdev, ccdf->fh); + /* The 0x0304 event may immediately reserve the device */ + if (!clp_get_state(zdev->fid, &state) && + state == ZPCI_FN_STATE_RESERVED) { + zpci_device_reserved(zdev); + } } break; case 0x0306: /* 0x308 or 0x302 for multiple devices */ - clp_rescan_pci_devices(); + zpci_remove_reserved_devices(); + clp_scan_pci_devices(); break; case 0x0308: /* Standby -> Reserved */ if (!zdev) break; - zpci_remove_device(zdev); + zpci_device_reserved(zdev); break; default: break; } - pci_dev_put(pdev); + if (existing_zdev) { + mutex_unlock(&zdev->state_lock); + zpci_zdev_put(zdev); + } } void zpci_event_availability(void *data) diff --git a/arch/s390/pci/pci_insn.c b/arch/s390/pci/pci_insn.c index 02f9505c99a8..56480be48244 100644 --- a/arch/s390/pci/pci_insn.c +++ b/arch/s390/pci/pci_insn.c @@ -9,6 +9,7 @@ #include <linux/errno.h> #include <linux/delay.h> #include <linux/jump_label.h> +#include <asm/asm-extable.h> #include <asm/facility.h> #include <asm/pci_insn.h> #include <asm/pci_debug.h> @@ -17,16 +18,40 @@ #define ZPCI_INSN_BUSY_DELAY 1 /* 1 microsecond */ -static inline void zpci_err_insn(u8 cc, u8 status, u64 req, u64 offset) +struct zpci_err_insn_data { + u8 insn; + u8 cc; + u8 status; + union { + struct { + u64 req; + u64 offset; + }; + struct { + u64 addr; + u64 len; + }; + }; +} __packed; + +static inline void zpci_err_insn_req(int lvl, u8 insn, u8 cc, u8 status, + u64 req, u64 offset) { - struct { - u64 req; - u64 offset; - u8 cc; - u8 status; - } __packed data = {req, offset, cc, status}; - - zpci_err_hex(&data, sizeof(data)); + struct zpci_err_insn_data data = { + .insn = insn, .cc = cc, .status = status, + .req = req, .offset = offset}; + + zpci_err_hex_level(lvl, &data, sizeof(data)); +} + +static inline void zpci_err_insn_addr(int lvl, u8 insn, u8 cc, u8 status, + u64 addr, u64 len) +{ + struct zpci_err_insn_data data = { + .insn = insn, .cc = cc, .status = status, + .addr = addr, .len = len}; + + zpci_err_hex_level(lvl, &data, sizeof(data)); } /* Modify PCI Function Controls */ @@ -46,33 +71,41 @@ static inline u8 __mpcifc(u64 req, struct zpci_fib *fib, u8 *status) u8 zpci_mod_fc(u64 req, struct zpci_fib *fib, u8 *status) { + bool retried = false; u8 cc; do { cc = __mpcifc(req, fib, status); - if (cc == 2) + if (cc == 2) { msleep(ZPCI_INSN_BUSY_DELAY); + if (!retried) { + zpci_err_insn_req(1, 'M', cc, *status, req, 0); + retried = true; + } + } } while (cc == 2); if (cc) - zpci_err_insn(cc, *status, req, 0); + zpci_err_insn_req(0, 'M', cc, *status, req, 0); + else if (retried) + zpci_err_insn_req(1, 'M', cc, *status, req, 0); return cc; } +EXPORT_SYMBOL_GPL(zpci_mod_fc); /* Refresh PCI Translations */ static inline u8 __rpcit(u64 fn, u64 addr, u64 range, u8 *status) { - register u64 __addr asm("2") = addr; - register u64 __range asm("3") = range; + union register_pair addr_range = {.even = addr, .odd = range}; u8 cc; asm volatile ( - " .insn rre,0xb9d30000,%[fn],%[addr]\n" + " .insn rre,0xb9d30000,%[fn],%[addr_range]\n" " ipm %[cc]\n" " srl %[cc],28\n" : [cc] "=d" (cc), [fn] "+d" (fn) - : [addr] "d" (__addr), "d" (__range) + : [addr_range] "d" (addr_range.pair) : "cc"); *status = fn >> 24 & 0xff; return cc; @@ -80,16 +113,24 @@ static inline u8 __rpcit(u64 fn, u64 addr, u64 range, u8 *status) int zpci_refresh_trans(u64 fn, u64 addr, u64 range) { + bool retried = false; u8 cc, status; do { cc = __rpcit(fn, addr, range, &status); - if (cc == 2) + if (cc == 2) { udelay(ZPCI_INSN_BUSY_DELAY); + if (!retried) { + zpci_err_insn_addr(1, 'R', cc, status, addr, range); + retried = true; + } + } } while (cc == 2); if (cc) - zpci_err_insn(cc, status, addr, range); + zpci_err_insn_addr(0, 'R', cc, status, addr, range); + else if (retried) + zpci_err_insn_addr(1, 'R', cc, status, addr, range); if (cc == 1 && (status == 4 || status == 16)) return -ENOMEM; @@ -98,7 +139,7 @@ int zpci_refresh_trans(u64 fn, u64 addr, u64 range) } /* Set Interruption Controls */ -int __zpci_set_irq_ctrl(u16 ctl, u8 isc, union zpci_sic_iib *iib) +int zpci_set_irq_ctrl(u16 ctl, u8 isc, union zpci_sic_iib *iib) { if (!test_facility(72)) return -EIO; @@ -109,25 +150,24 @@ int __zpci_set_irq_ctrl(u16 ctl, u8 isc, union zpci_sic_iib *iib) return 0; } +EXPORT_SYMBOL_GPL(zpci_set_irq_ctrl); /* PCI Load */ static inline int ____pcilg(u64 *data, u64 req, u64 offset, u8 *status) { - register u64 __req asm("2") = req; - register u64 __offset asm("3") = offset; + union register_pair req_off = {.even = req, .odd = offset}; int cc = -ENXIO; u64 __data; asm volatile ( - " .insn rre,0xb9d20000,%[data],%[req]\n" + " .insn rre,0xb9d20000,%[data],%[req_off]\n" "0: ipm %[cc]\n" " srl %[cc],28\n" "1:\n" EX_TABLE(0b, 1b) - : [cc] "+d" (cc), [data] "=d" (__data), [req] "+d" (__req) - : "d" (__offset) - : "cc"); - *status = __req >> 24 & 0xff; + : [cc] "+d" (cc), [data] "=d" (__data), + [req_off] "+&d" (req_off.pair) :: "cc"); + *status = req_off.even >> 24 & 0xff; *data = __data; return cc; } @@ -146,17 +186,25 @@ static inline int __pcilg(u64 *data, u64 req, u64 offset, u8 *status) int __zpci_load(u64 *data, u64 req, u64 offset) { + bool retried = false; u8 status; int cc; do { cc = __pcilg(data, req, offset, &status); - if (cc == 2) + if (cc == 2) { udelay(ZPCI_INSN_BUSY_DELAY); + if (!retried) { + zpci_err_insn_req(1, 'l', cc, status, req, offset); + retried = true; + } + } } while (cc == 2); if (cc) - zpci_err_insn(cc, status, req, offset); + zpci_err_insn_req(0, 'l', cc, status, req, offset); + else if (retried) + zpci_err_insn_req(1, 'l', cc, status, req, offset); return (cc > 0) ? -EIO : cc; } @@ -166,28 +214,26 @@ static inline int zpci_load_fh(u64 *data, const volatile void __iomem *addr, unsigned long len) { struct zpci_iomap_entry *entry = &zpci_iomap_start[ZPCI_IDX(addr)]; - u64 req = ZPCI_CREATE_REQ(entry->fh, entry->bar, len); + u64 req = ZPCI_CREATE_REQ(READ_ONCE(entry->fh), entry->bar, len); return __zpci_load(data, req, ZPCI_OFFSET(addr)); } static inline int __pcilg_mio(u64 *data, u64 ioaddr, u64 len, u8 *status) { - register u64 addr asm("2") = ioaddr; - register u64 r3 asm("3") = len; + union register_pair ioaddr_len = {.even = ioaddr, .odd = len}; int cc = -ENXIO; u64 __data; asm volatile ( - " .insn rre,0xb9d60000,%[data],%[ioaddr]\n" + " .insn rre,0xb9d60000,%[data],%[ioaddr_len]\n" "0: ipm %[cc]\n" " srl %[cc],28\n" "1:\n" EX_TABLE(0b, 1b) - : [cc] "+d" (cc), [data] "=d" (__data), "+d" (r3) - : [ioaddr] "d" (addr) - : "cc"); - *status = r3 >> 24 & 0xff; + : [cc] "+d" (cc), [data] "=d" (__data), + [ioaddr_len] "+&d" (ioaddr_len.pair) :: "cc"); + *status = ioaddr_len.odd >> 24 & 0xff; *data = __data; return cc; } @@ -202,7 +248,7 @@ int zpci_load(u64 *data, const volatile void __iomem *addr, unsigned long len) cc = __pcilg_mio(data, (__force u64) addr, len, &status); if (cc) - zpci_err_insn(cc, status, 0, (__force u64) addr); + zpci_err_insn_addr(0, 'L', cc, status, (__force u64) addr, len); return (cc > 0) ? -EIO : cc; } @@ -211,36 +257,43 @@ EXPORT_SYMBOL_GPL(zpci_load); /* PCI Store */ static inline int __pcistg(u64 data, u64 req, u64 offset, u8 *status) { - register u64 __req asm("2") = req; - register u64 __offset asm("3") = offset; + union register_pair req_off = {.even = req, .odd = offset}; int cc = -ENXIO; asm volatile ( - " .insn rre,0xb9d00000,%[data],%[req]\n" + " .insn rre,0xb9d00000,%[data],%[req_off]\n" "0: ipm %[cc]\n" " srl %[cc],28\n" "1:\n" EX_TABLE(0b, 1b) - : [cc] "+d" (cc), [req] "+d" (__req) - : "d" (__offset), [data] "d" (data) + : [cc] "+d" (cc), [req_off] "+&d" (req_off.pair) + : [data] "d" (data) : "cc"); - *status = __req >> 24 & 0xff; + *status = req_off.even >> 24 & 0xff; return cc; } int __zpci_store(u64 data, u64 req, u64 offset) { + bool retried = false; u8 status; int cc; do { cc = __pcistg(data, req, offset, &status); - if (cc == 2) + if (cc == 2) { udelay(ZPCI_INSN_BUSY_DELAY); + if (!retried) { + zpci_err_insn_req(1, 's', cc, status, req, offset); + retried = true; + } + } } while (cc == 2); if (cc) - zpci_err_insn(cc, status, req, offset); + zpci_err_insn_req(0, 's', cc, status, req, offset); + else if (retried) + zpci_err_insn_req(1, 's', cc, status, req, offset); return (cc > 0) ? -EIO : cc; } @@ -250,27 +303,26 @@ static inline int zpci_store_fh(const volatile void __iomem *addr, u64 data, unsigned long len) { struct zpci_iomap_entry *entry = &zpci_iomap_start[ZPCI_IDX(addr)]; - u64 req = ZPCI_CREATE_REQ(entry->fh, entry->bar, len); + u64 req = ZPCI_CREATE_REQ(READ_ONCE(entry->fh), entry->bar, len); return __zpci_store(data, req, ZPCI_OFFSET(addr)); } static inline int __pcistg_mio(u64 data, u64 ioaddr, u64 len, u8 *status) { - register u64 addr asm("2") = ioaddr; - register u64 r3 asm("3") = len; + union register_pair ioaddr_len = {.even = ioaddr, .odd = len}; int cc = -ENXIO; asm volatile ( - " .insn rre,0xb9d40000,%[data],%[ioaddr]\n" + " .insn rre,0xb9d40000,%[data],%[ioaddr_len]\n" "0: ipm %[cc]\n" " srl %[cc],28\n" "1:\n" EX_TABLE(0b, 1b) - : [cc] "+d" (cc), "+d" (r3) - : [data] "d" (data), [ioaddr] "d" (addr) - : "cc"); - *status = r3 >> 24 & 0xff; + : [cc] "+d" (cc), [ioaddr_len] "+&d" (ioaddr_len.pair) + : [data] "d" (data) + : "cc", "memory"); + *status = ioaddr_len.odd >> 24 & 0xff; return cc; } @@ -284,7 +336,7 @@ int zpci_store(const volatile void __iomem *addr, u64 data, unsigned long len) cc = __pcistg_mio(data, (__force u64) addr, len, &status); if (cc) - zpci_err_insn(cc, status, 0, (__force u64) addr); + zpci_err_insn_addr(0, 'S', cc, status, (__force u64) addr, len); return (cc > 0) ? -EIO : cc; } @@ -310,17 +362,25 @@ static inline int __pcistb(const u64 *data, u64 req, u64 offset, u8 *status) int __zpci_store_block(const u64 *data, u64 req, u64 offset) { + bool retried = false; u8 status; int cc; do { cc = __pcistb(data, req, offset, &status); - if (cc == 2) + if (cc == 2) { udelay(ZPCI_INSN_BUSY_DELAY); + if (!retried) { + zpci_err_insn_req(0, 'b', cc, status, req, offset); + retried = true; + } + } } while (cc == 2); if (cc) - zpci_err_insn(cc, status, req, offset); + zpci_err_insn_req(0, 'b', cc, status, req, offset); + else if (retried) + zpci_err_insn_req(1, 'b', cc, status, req, offset); return (cc > 0) ? -EIO : cc; } @@ -364,7 +424,7 @@ int zpci_write_block(volatile void __iomem *dst, cc = __pcistb_mio(src, (__force u64) dst, len, &status); if (cc) - zpci_err_insn(cc, status, 0, (__force u64) dst); + zpci_err_insn_addr(0, 'B', cc, status, (__force u64) dst, len); return (cc > 0) ? -EIO : cc; } @@ -372,10 +432,7 @@ EXPORT_SYMBOL_GPL(zpci_write_block); static inline void __pciwb_mio(void) { - unsigned long unused = 0; - - asm volatile (".insn rre,0xb9d50000,%[op],%[op]\n" - : [op] "+d" (unused)); + asm volatile (".insn rre,0xb9d50000,0,0\n"); } void zpci_barrier(void) diff --git a/arch/s390/pci/pci_iov.c b/arch/s390/pci/pci_iov.c new file mode 100644 index 000000000000..ead062bf2b41 --- /dev/null +++ b/arch/s390/pci/pci_iov.c @@ -0,0 +1,99 @@ +// SPDX-License-Identifier: GPL-2.0 +/* + * Copyright IBM Corp. 2020 + * + * Author(s): + * Niklas Schnelle <schnelle@linux.ibm.com> + * + */ + +#define KMSG_COMPONENT "zpci" +#define pr_fmt(fmt) KMSG_COMPONENT ": " fmt + +#include <linux/kernel.h> +#include <linux/pci.h> + +#include "pci_iov.h" + +static struct resource iov_res = { + .name = "PCI IOV res", + .start = 0, + .end = -1, + .flags = IORESOURCE_MEM, +}; + +void zpci_iov_map_resources(struct pci_dev *pdev) +{ + resource_size_t len; + int i; + + for (i = 0; i < PCI_SRIOV_NUM_BARS; i++) { + int bar = i + PCI_IOV_RESOURCES; + + len = pci_resource_len(pdev, bar); + if (!len) + continue; + pdev->resource[bar].parent = &iov_res; + } +} + +void zpci_iov_remove_virtfn(struct pci_dev *pdev, int vfn) +{ + pci_lock_rescan_remove(); + /* Linux' vfid's start at 0 vfn at 1 */ + pci_iov_remove_virtfn(pdev->physfn, vfn - 1); + pci_unlock_rescan_remove(); +} + +static int zpci_iov_link_virtfn(struct pci_dev *pdev, struct pci_dev *virtfn, int vfid) +{ + int rc; + + rc = pci_iov_sysfs_link(pdev, virtfn, vfid); + if (rc) + return rc; + + virtfn->is_virtfn = 1; + virtfn->multifunction = 0; + virtfn->physfn = pci_dev_get(pdev); + + return 0; +} + +int zpci_iov_setup_virtfn(struct zpci_bus *zbus, struct pci_dev *virtfn, int vfn) +{ + int i, cand_devfn; + struct zpci_dev *zdev; + struct pci_dev *pdev; + int vfid = vfn - 1; /* Linux' vfid's start at 0 vfn at 1*/ + int rc = 0; + + if (!zbus->multifunction) + return 0; + + /* If the parent PF for the given VF is also configured in the + * instance, it must be on the same zbus. + * We can then identify the parent PF by checking what + * devfn the VF would have if it belonged to that PF using the PF's + * stride and offset. Only if this candidate devfn matches the + * actual devfn will we link both functions. + */ + for (i = 0; i < ZPCI_FUNCTIONS_PER_BUS; i++) { + zdev = zbus->function[i]; + if (zdev && zdev->is_physfn) { + pdev = pci_get_slot(zbus->bus, zdev->devfn); + if (!pdev) + continue; + cand_devfn = pci_iov_virtfn_devfn(pdev, vfid); + if (cand_devfn == virtfn->devfn) { + rc = zpci_iov_link_virtfn(pdev, virtfn, vfid); + /* balance pci_get_slot() */ + pci_dev_put(pdev); + break; + } + /* balance pci_get_slot() */ + pci_dev_put(pdev); + } + } + return rc; +} diff --git a/arch/s390/pci/pci_iov.h b/arch/s390/pci/pci_iov.h new file mode 100644 index 000000000000..b2c828003bad --- /dev/null +++ b/arch/s390/pci/pci_iov.h @@ -0,0 +1,30 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +/* + * Copyright IBM Corp. 2020 + * + * Author(s): + * Niklas Schnelle <schnelle@linux.ibm.com> + * + */ + +#ifndef __S390_PCI_IOV_H +#define __S390_PCI_IOV_H + +#ifdef CONFIG_PCI_IOV +void zpci_iov_remove_virtfn(struct pci_dev *pdev, int vfn); + +void zpci_iov_map_resources(struct pci_dev *pdev); + +int zpci_iov_setup_virtfn(struct zpci_bus *zbus, struct pci_dev *virtfn, int vfn); + +#else /* CONFIG_PCI_IOV */ +static inline void zpci_iov_remove_virtfn(struct pci_dev *pdev, int vfn) {} + +static inline void zpci_iov_map_resources(struct pci_dev *pdev) {} + +static inline int zpci_iov_setup_virtfn(struct zpci_bus *zbus, struct pci_dev *virtfn, int vfn) +{ + return 0; +} +#endif /* CONFIG_PCI_IOV */ +#endif /* __S390_PCI_IOV_h */ diff --git a/arch/s390/pci/pci_irq.c b/arch/s390/pci/pci_irq.c index fbe97ab2e228..ff8f24854c64 100644 --- a/arch/s390/pci/pci_irq.c +++ b/arch/s390/pci/pci_irq.c @@ -11,16 +11,10 @@ #include <asm/isc.h> #include <asm/airq.h> +#include <asm/tpi.h> static enum {FLOATING, DIRECTED} irq_delivery; -#define SIC_IRQ_MODE_ALL 0 -#define SIC_IRQ_MODE_SINGLE 1 -#define SIC_IRQ_MODE_DIRECT 4 -#define SIC_IRQ_MODE_D_ALL 16 -#define SIC_IRQ_MODE_D_SINGLE 17 -#define SIC_IRQ_MODE_SET_CPU 18 - /* * summary bit vector * FLOATING - summary bit per function @@ -35,7 +29,7 @@ static struct airq_iv *zpci_sbv; */ static struct airq_iv **zpci_ibv; -/* Modify PCI: Register adapter interruptions */ +/* Modify PCI: Register floating adapter interruptions */ static int zpci_set_airq(struct zpci_dev *zdev) { u64 req = ZPCI_CREATE_REQ(zdev->fh, 0, ZPCI_MOD_FC_REG_INT); @@ -45,21 +39,24 @@ static int zpci_set_airq(struct zpci_dev *zdev) fib.fmt0.isc = PCI_ISC; fib.fmt0.sum = 1; /* enable summary notifications */ fib.fmt0.noi = airq_iv_end(zdev->aibv); - fib.fmt0.aibv = (unsigned long) zdev->aibv->vector; + fib.fmt0.aibv = virt_to_phys(zdev->aibv->vector); fib.fmt0.aibvo = 0; /* each zdev has its own interrupt vector */ - fib.fmt0.aisb = (unsigned long) zpci_sbv->vector + (zdev->aisb/64)*8; + fib.fmt0.aisb = virt_to_phys(zpci_sbv->vector) + (zdev->aisb / 64) * 8; fib.fmt0.aisbo = zdev->aisb & 63; + fib.gd = zdev->gisa; return zpci_mod_fc(req, &fib, &status) ? -EIO : 0; } -/* Modify PCI: Unregister adapter interruptions */ +/* Modify PCI: Unregister floating adapter interruptions */ static int zpci_clear_airq(struct zpci_dev *zdev) { u64 req = ZPCI_CREATE_REQ(zdev->fh, 0, ZPCI_MOD_FC_DEREG_INT); struct zpci_fib fib = {0}; u8 cc, status; + fib.gd = zdev->gisa; + cc = zpci_mod_fc(req, &fib, &status); if (cc == 3 || (cc == 1 && status == 24)) /* Function already gone or IRQs already deregistered. */ @@ -78,6 +75,7 @@ static int zpci_set_directed_irq(struct zpci_dev *zdev) fib.fmt = 1; fib.fmt1.noi = zdev->msi_nr_irqs; fib.fmt1.dibvo = zdev->msi_first_bit; + fib.gd = zdev->gisa; return zpci_mod_fc(req, &fib, &status) ? -EIO : 0; } @@ -90,6 +88,7 @@ static int zpci_clear_directed_irq(struct zpci_dev *zdev) u8 cc, status; fib.fmt = 1; + fib.gd = zdev->gisa; cc = zpci_mod_fc(req, &fib, &status); if (cc == 3 || (cc == 1 && status == 24)) /* Function already gone or IRQs already deregistered. */ @@ -98,14 +97,47 @@ static int zpci_clear_directed_irq(struct zpci_dev *zdev) return cc ? -EIO : 0; } +/* Register adapter interruptions */ +static int zpci_set_irq(struct zpci_dev *zdev) +{ + int rc; + + if (irq_delivery == DIRECTED) + rc = zpci_set_directed_irq(zdev); + else + rc = zpci_set_airq(zdev); + + if (!rc) + zdev->irqs_registered = 1; + + return rc; +} + +/* Clear adapter interruptions */ +static int zpci_clear_irq(struct zpci_dev *zdev) +{ + int rc; + + if (irq_delivery == DIRECTED) + rc = zpci_clear_directed_irq(zdev); + else + rc = zpci_clear_airq(zdev); + + if (!rc) + zdev->irqs_registered = 0; + + return rc; +} + static int zpci_set_irq_affinity(struct irq_data *data, const struct cpumask *dest, bool force) { - struct msi_desc *entry = irq_get_msi_desc(data->irq); + struct msi_desc *entry = irq_data_get_msi_desc(data); struct msi_msg msg = entry->msg; + int cpu_addr = smp_cpu_get_cpu_address(cpumask_first(dest)); msg.address_lo &= 0xff0000ff; - msg.address_lo |= (cpumask_first(dest) << 8); + msg.address_lo |= (cpu_addr << 8); pci_write_msi_msg(data->irq, &msg); return IRQ_SET_MASK_OK; @@ -115,12 +147,12 @@ static struct irq_chip zpci_irq_chip = { .name = "PCI-MSI", .irq_unmask = pci_msi_unmask_irq, .irq_mask = pci_msi_mask_irq, - .irq_set_affinity = zpci_set_irq_affinity, }; static void zpci_handle_cpu_local_irq(bool rescan) { struct airq_iv *dibv = zpci_ibv[smp_processor_id()]; + union zpci_sic_iib iib = {{0}}; unsigned long bit; int irqs_on = 0; @@ -131,8 +163,8 @@ static void zpci_handle_cpu_local_irq(bool rescan) if (!rescan || irqs_on++) /* End of second scan with interrupts on. */ break; - /* First scan complete, reenable interrupts. */ - if (zpci_set_irq_ctrl(SIC_IRQ_MODE_D_SINGLE, PCI_ISC)) + /* First scan complete, re-enable interrupts. */ + if (zpci_set_irq_ctrl(SIC_IRQ_MODE_D_SINGLE, PCI_ISC, &iib)) break; bit = 0; continue; @@ -160,6 +192,7 @@ static void zpci_handle_remote_irq(void *data) static void zpci_handle_fallback_irq(void) { struct cpu_irq_data *cpu_data; + union zpci_sic_iib iib = {{0}}; unsigned long cpu; int irqs_on = 0; @@ -169,8 +202,8 @@ static void zpci_handle_fallback_irq(void) if (irqs_on++) /* End of second scan with interrupts on. */ break; - /* First scan complete, reenable interrupts. */ - if (zpci_set_irq_ctrl(SIC_IRQ_MODE_SINGLE, PCI_ISC)) + /* First scan complete, re-enable interrupts. */ + if (zpci_set_irq_ctrl(SIC_IRQ_MODE_SINGLE, PCI_ISC, &iib)) break; cpu = 0; continue; @@ -179,15 +212,16 @@ static void zpci_handle_fallback_irq(void) if (atomic_inc_return(&cpu_data->scheduled) > 1) continue; - cpu_data->csd.func = zpci_handle_remote_irq; - cpu_data->csd.info = &cpu_data->scheduled; - cpu_data->csd.flags = 0; + INIT_CSD(&cpu_data->csd, zpci_handle_remote_irq, &cpu_data->scheduled); smp_call_function_single_async(cpu, &cpu_data->csd); } } -static void zpci_directed_irq_handler(struct airq_struct *airq, bool floating) +static void zpci_directed_irq_handler(struct airq_struct *airq, + struct tpi_info *tpi_info) { + bool floating = !tpi_info->directed_irq; + if (floating) { inc_irq_stat(IRQIO_PCF); zpci_handle_fallback_irq(); @@ -197,8 +231,10 @@ static void zpci_directed_irq_handler(struct airq_struct *airq, bool floating) } } -static void zpci_floating_irq_handler(struct airq_struct *airq, bool floating) +static void zpci_floating_irq_handler(struct airq_struct *airq, + struct tpi_info *tpi_info) { + union zpci_sic_iib iib = {{0}}; unsigned long si, ai; struct airq_iv *aibv; int irqs_on = 0; @@ -211,8 +247,8 @@ static void zpci_floating_irq_handler(struct airq_struct *airq, bool floating) if (irqs_on++) /* End of second scan with interrupts on. */ break; - /* First scan complete, reenable interrupts. */ - if (zpci_set_irq_ctrl(SIC_IRQ_MODE_SINGLE, PCI_ISC)) + /* First scan complete, re-enable interrupts. */ + if (zpci_set_irq_ctrl(SIC_IRQ_MODE_SINGLE, PCI_ISC, &iib)) break; si = 0; continue; @@ -239,6 +275,7 @@ int arch_setup_msi_irqs(struct pci_dev *pdev, int nvec, int type) unsigned long bit; struct msi_desc *msi; struct msi_msg msg; + int cpu_addr; int rc, irq; zdev->aisb = -1UL; @@ -260,7 +297,7 @@ int arch_setup_msi_irqs(struct pci_dev *pdev, int nvec, int type) zdev->aisb = bit; /* Create adapter interrupt vector */ - zdev->aibv = airq_iv_create(msi_vecs, AIRQ_IV_DATA | AIRQ_IV_BITLOCK); + zdev->aibv = airq_iv_create(msi_vecs, AIRQ_IV_DATA | AIRQ_IV_BITLOCK, NULL); if (!zdev->aibv) return -ENOMEM; @@ -272,11 +309,13 @@ int arch_setup_msi_irqs(struct pci_dev *pdev, int nvec, int type) /* Request MSI interrupts */ hwirq = bit; - for_each_pci_msi_entry(msi, pdev) { + msi_for_each_desc(msi, &pdev->dev, MSI_DESC_NOTASSOCIATED) { rc = -EIO; if (hwirq - bit >= msi_vecs) break; - irq = __irq_alloc_descs(-1, 0, 1, 0, THIS_MODULE, msi->affinity); + irq = __irq_alloc_descs(-1, 0, 1, 0, THIS_MODULE, + (irq_delivery == DIRECTED) ? + msi->affinity : NULL); if (irq < 0) return -ENOMEM; rc = irq_set_msi_desc(irq, msi); @@ -286,9 +325,15 @@ int arch_setup_msi_irqs(struct pci_dev *pdev, int nvec, int type) handle_percpu_irq); msg.data = hwirq - bit; if (irq_delivery == DIRECTED) { + if (msi->affinity) + cpu = cpumask_first(&msi->affinity->mask); + else + cpu = 0; + cpu_addr = smp_cpu_get_cpu_address(cpu); + msg.address_lo = zdev->msi_addr & 0xff0000ff; - msg.address_lo |= msi->affinity ? - (cpumask_first(&msi->affinity->mask) << 8) : 0; + msg.address_lo |= (cpu_addr << 8); + for_each_possible_cpu(cpu) { airq_iv_set_data(zpci_ibv[cpu], hwirq, irq); } @@ -304,10 +349,7 @@ int arch_setup_msi_irqs(struct pci_dev *pdev, int nvec, int type) zdev->msi_first_bit = bit; zdev->msi_nr_irqs = msi_vecs; - if (irq_delivery == DIRECTED) - rc = zpci_set_directed_irq(zdev); - else - rc = zpci_set_airq(zdev); + rc = zpci_set_irq(zdev); if (rc) return rc; @@ -321,21 +363,12 @@ void arch_teardown_msi_irqs(struct pci_dev *pdev) int rc; /* Disable interrupts */ - if (irq_delivery == DIRECTED) - rc = zpci_clear_directed_irq(zdev); - else - rc = zpci_clear_airq(zdev); + rc = zpci_clear_irq(zdev); if (rc) return; /* Release MSI interrupts */ - for_each_pci_msi_entry(msi, pdev) { - if (!msi->irq) - continue; - if (msi->msi_attrib.is_msix) - __pci_msix_desc_mask_irq(msi, 1); - else - __pci_msi_desc_mask_irq(msi, 1, 1); + msi_for_each_desc(msi, &pdev->dev, MSI_DESC_ASSOCIATED) { irq_set_msi_desc(msi->irq, NULL); irq_free_desc(msi->irq); msi->msg.address_lo = 0; @@ -358,6 +391,15 @@ void arch_teardown_msi_irqs(struct pci_dev *pdev) airq_iv_free(zpci_ibv[0], zdev->msi_first_bit, zdev->msi_nr_irqs); } +bool arch_restore_msi_irqs(struct pci_dev *pdev) +{ + struct zpci_dev *zdev = to_zpci(pdev); + + if (!zdev->irqs_registered) + zpci_set_irq(zdev); + return true; +} + static struct airq_struct zpci_airq = { .handler = zpci_floating_irq_handler, .isc = PCI_ISC, @@ -366,11 +408,12 @@ static struct airq_struct zpci_airq = { static void __init cpu_enable_directed_irq(void *unused) { union zpci_sic_iib iib = {{0}}; + union zpci_sic_iib ziib = {{0}}; iib.cdiib.dibv_addr = (u64) zpci_ibv[smp_processor_id()]->vector; - __zpci_set_irq_ctrl(SIC_IRQ_MODE_SET_CPU, 0, &iib); - zpci_set_irq_ctrl(SIC_IRQ_MODE_D_SINGLE, PCI_ISC); + zpci_set_irq_ctrl(SIC_IRQ_MODE_SET_CPU, 0, &iib); + zpci_set_irq_ctrl(SIC_IRQ_MODE_D_SINGLE, PCI_ISC, &ziib); } static int __init zpci_directed_irq_init(void) @@ -378,14 +421,14 @@ static int __init zpci_directed_irq_init(void) union zpci_sic_iib iib = {{0}}; unsigned int cpu; - zpci_sbv = airq_iv_create(num_possible_cpus(), 0); + zpci_sbv = airq_iv_create(num_possible_cpus(), 0, NULL); if (!zpci_sbv) return -ENOMEM; iib.diib.isc = PCI_ISC; iib.diib.nr_cpus = num_possible_cpus(); - iib.diib.disb_addr = (u64) zpci_sbv->vector; - __zpci_set_irq_ctrl(SIC_IRQ_MODE_DIRECT, 0, &iib); + iib.diib.disb_addr = virt_to_phys(zpci_sbv->vector); + zpci_set_irq_ctrl(SIC_IRQ_MODE_DIRECT, 0, &iib); zpci_ibv = kcalloc(num_possible_cpus(), sizeof(*zpci_ibv), GFP_KERNEL); @@ -400,7 +443,7 @@ static int __init zpci_directed_irq_init(void) zpci_ibv[cpu] = airq_iv_create(cache_line_size() * BITS_PER_BYTE, AIRQ_IV_DATA | AIRQ_IV_CACHELINE | - (!cpu ? AIRQ_IV_ALLOC : 0)); + (!cpu ? AIRQ_IV_ALLOC : 0), NULL); if (!zpci_ibv[cpu]) return -ENOMEM; } @@ -417,7 +460,7 @@ static int __init zpci_floating_irq_init(void) if (!zpci_ibv) return -ENOMEM; - zpci_sbv = airq_iv_create(ZPCI_NR_DEVICES, AIRQ_IV_ALLOC); + zpci_sbv = airq_iv_create(ZPCI_NR_DEVICES, AIRQ_IV_ALLOC, NULL); if (!zpci_sbv) goto out_free; @@ -430,6 +473,7 @@ out_free: int __init zpci_irq_init(void) { + union zpci_sic_iib iib = {{0}}; int rc; irq_delivery = sclp.has_dirq ? DIRECTED : FLOATING; @@ -461,7 +505,7 @@ int __init zpci_irq_init(void) * Enable floating IRQs (with suppression after one IRQ). When using * directed IRQs this enables the fallback path. */ - zpci_set_irq_ctrl(SIC_IRQ_MODE_SINGLE, PCI_ISC); + zpci_set_irq_ctrl(SIC_IRQ_MODE_SINGLE, PCI_ISC, &iib); return 0; out_airq: diff --git a/arch/s390/pci/pci_kvm_hook.c b/arch/s390/pci/pci_kvm_hook.c new file mode 100644 index 000000000000..ff34baf50a3e --- /dev/null +++ b/arch/s390/pci/pci_kvm_hook.c @@ -0,0 +1,11 @@ +// SPDX-License-Identifier: GPL-2.0-only +/* + * VFIO ZPCI devices support + * + * Copyright (C) IBM Corp. 2022. All rights reserved. + * Author(s): Pierre Morel <pmorel@linux.ibm.com> + */ +#include <linux/kvm_host.h> + +struct zpci_kvm_hook zpci_kvm_hook; +EXPORT_SYMBOL_GPL(zpci_kvm_hook); diff --git a/arch/s390/pci/pci_mmio.c b/arch/s390/pci/pci_mmio.c index 7d42a8794f10..a90499c087f0 100644 --- a/arch/s390/pci/pci_mmio.c +++ b/arch/s390/pci/pci_mmio.c @@ -11,25 +11,108 @@ #include <linux/mm.h> #include <linux/errno.h> #include <linux/pci.h> +#include <asm/asm-extable.h> +#include <asm/pci_io.h> +#include <asm/pci_debug.h> -static long get_pfn(unsigned long user_addr, unsigned long access, - unsigned long *pfn) +static inline void zpci_err_mmio(u8 cc, u8 status, u64 offset) { - struct vm_area_struct *vma; - long ret; + struct { + u64 offset; + u8 cc; + u8 status; + } data = {offset, cc, status}; - down_read(¤t->mm->mmap_sem); - ret = -EINVAL; - vma = find_vma(current->mm, user_addr); - if (!vma) - goto out; - ret = -EACCES; - if (!(vma->vm_flags & access)) - goto out; - ret = follow_pfn(vma, user_addr, pfn); -out: - up_read(¤t->mm->mmap_sem); - return ret; + zpci_err_hex(&data, sizeof(data)); +} + +static inline int __pcistb_mio_inuser( + void __iomem *ioaddr, const void __user *src, + u64 len, u8 *status) +{ + int cc = -ENXIO; + + asm volatile ( + " sacf 256\n" + "0: .insn rsy,0xeb00000000d4,%[len],%[ioaddr],%[src]\n" + "1: ipm %[cc]\n" + " srl %[cc],28\n" + "2: sacf 768\n" + EX_TABLE(0b, 2b) EX_TABLE(1b, 2b) + : [cc] "+d" (cc), [len] "+d" (len) + : [ioaddr] "a" (ioaddr), [src] "Q" (*((u8 __force *)src)) + : "cc", "memory"); + *status = len >> 24 & 0xff; + return cc; +} + +static inline int __pcistg_mio_inuser( + void __iomem *ioaddr, const void __user *src, + u64 ulen, u8 *status) +{ + union register_pair ioaddr_len = {.even = (u64 __force)ioaddr, .odd = ulen}; + int cc = -ENXIO; + u64 val = 0; + u64 cnt = ulen; + u8 tmp; + + /* + * copy 0 < @len <= 8 bytes from @src into the right most bytes of + * a register, then store it to PCI at @ioaddr while in secondary + * address space. pcistg then uses the user mappings. + */ + asm volatile ( + " sacf 256\n" + "0: llgc %[tmp],0(%[src])\n" + "4: sllg %[val],%[val],8\n" + " aghi %[src],1\n" + " ogr %[val],%[tmp]\n" + " brctg %[cnt],0b\n" + "1: .insn rre,0xb9d40000,%[val],%[ioaddr_len]\n" + "2: ipm %[cc]\n" + " srl %[cc],28\n" + "3: sacf 768\n" + EX_TABLE(0b, 3b) EX_TABLE(4b, 3b) EX_TABLE(1b, 3b) EX_TABLE(2b, 3b) + : + [src] "+a" (src), [cnt] "+d" (cnt), + [val] "+d" (val), [tmp] "=d" (tmp), + [cc] "+d" (cc), [ioaddr_len] "+&d" (ioaddr_len.pair) + :: "cc", "memory"); + *status = ioaddr_len.odd >> 24 & 0xff; + + /* did we read everything from user memory? */ + if (!cc && cnt != 0) + cc = -EFAULT; + + return cc; +} + +static inline int __memcpy_toio_inuser(void __iomem *dst, + const void __user *src, size_t n) +{ + int size, rc = 0; + u8 status = 0; + + if (!src) + return -EINVAL; + + while (n > 0) { + size = zpci_get_max_io_size((u64 __force) dst, + (u64 __force) src, n, + ZPCI_MAX_WRITE_SIZE); + if (size > 8) /* main path */ + rc = __pcistb_mio_inuser(dst, src, size, &status); + else + rc = __pcistg_mio_inuser(dst, src, size, &status); + if (rc) + break; + src += size; + dst += size; + n -= size; + } + if (rc) + zpci_err_mmio(rc, status, (__force u64) dst); + return rc; } SYSCALL_DEFINE3(s390_pci_mmio_write, unsigned long, mmio_addr, @@ -38,7 +121,9 @@ SYSCALL_DEFINE3(s390_pci_mmio_write, unsigned long, mmio_addr, u8 local_buf[64]; void __iomem *io_addr; void *buf; - unsigned long pfn; + struct vm_area_struct *vma; + pte_t *ptep; + spinlock_t *ptl; long ret; if (!zpci_is_enabled()) @@ -46,6 +131,22 @@ SYSCALL_DEFINE3(s390_pci_mmio_write, unsigned long, mmio_addr, if (length <= 0 || PAGE_SIZE - (mmio_addr & ~PAGE_MASK) < length) return -EINVAL; + + /* + * We only support write access to MIO capable devices if we are on + * a MIO enabled system. Otherwise we would have to check for every + * address if it is a special ZPCI_ADDR and would have to do + * a pfn lookup which we don't need for MIO capable devices. Currently + * ISM devices are the only devices without MIO support and there is no + * known need for accessing these from userspace. + */ + if (static_branch_likely(&have_mio)) { + ret = __memcpy_toio_inuser((void __iomem *) mmio_addr, + user_buffer, + length); + return ret; + } + if (length > 64) { buf = kmalloc(length, GFP_KERNEL); if (!buf) @@ -53,32 +154,118 @@ SYSCALL_DEFINE3(s390_pci_mmio_write, unsigned long, mmio_addr, } else buf = local_buf; - ret = get_pfn(mmio_addr, VM_WRITE, &pfn); + ret = -EFAULT; + if (copy_from_user(buf, user_buffer, length)) + goto out_free; + + mmap_read_lock(current->mm); + ret = -EINVAL; + vma = vma_lookup(current->mm, mmio_addr); + if (!vma) + goto out_unlock_mmap; + if (!(vma->vm_flags & (VM_IO | VM_PFNMAP))) + goto out_unlock_mmap; + ret = -EACCES; + if (!(vma->vm_flags & VM_WRITE)) + goto out_unlock_mmap; + + ret = follow_pte(vma->vm_mm, mmio_addr, &ptep, &ptl); if (ret) - goto out; - io_addr = (void __iomem *)((pfn << PAGE_SHIFT) | (mmio_addr & ~PAGE_MASK)); + goto out_unlock_mmap; - ret = -EFAULT; - if ((unsigned long) io_addr < ZPCI_IOMAP_ADDR_BASE) - goto out; + io_addr = (void __iomem *)((pte_pfn(*ptep) << PAGE_SHIFT) | + (mmio_addr & ~PAGE_MASK)); - if (copy_from_user(buf, user_buffer, length)) - goto out; + if ((unsigned long) io_addr < ZPCI_IOMAP_ADDR_BASE) + goto out_unlock_pt; ret = zpci_memcpy_toio(io_addr, buf, length); -out: +out_unlock_pt: + pte_unmap_unlock(ptep, ptl); +out_unlock_mmap: + mmap_read_unlock(current->mm); +out_free: if (buf != local_buf) kfree(buf); return ret; } +static inline int __pcilg_mio_inuser( + void __user *dst, const void __iomem *ioaddr, + u64 ulen, u8 *status) +{ + union register_pair ioaddr_len = {.even = (u64 __force)ioaddr, .odd = ulen}; + u64 cnt = ulen; + int shift = ulen * 8; + int cc = -ENXIO; + u64 val, tmp; + + /* + * read 0 < @len <= 8 bytes from the PCI memory mapped at @ioaddr (in + * user space) into a register using pcilg then store these bytes at + * user address @dst + */ + asm volatile ( + " sacf 256\n" + "0: .insn rre,0xb9d60000,%[val],%[ioaddr_len]\n" + "1: ipm %[cc]\n" + " srl %[cc],28\n" + " ltr %[cc],%[cc]\n" + " jne 4f\n" + "2: ahi %[shift],-8\n" + " srlg %[tmp],%[val],0(%[shift])\n" + "3: stc %[tmp],0(%[dst])\n" + "5: aghi %[dst],1\n" + " brctg %[cnt],2b\n" + "4: sacf 768\n" + EX_TABLE(0b, 4b) EX_TABLE(1b, 4b) EX_TABLE(3b, 4b) EX_TABLE(5b, 4b) + : + [ioaddr_len] "+&d" (ioaddr_len.pair), + [cc] "+d" (cc), [val] "=d" (val), + [dst] "+a" (dst), [cnt] "+d" (cnt), [tmp] "=d" (tmp), + [shift] "+d" (shift) + :: "cc", "memory"); + + /* did we write everything to the user space buffer? */ + if (!cc && cnt != 0) + cc = -EFAULT; + + *status = ioaddr_len.odd >> 24 & 0xff; + return cc; +} + +static inline int __memcpy_fromio_inuser(void __user *dst, + const void __iomem *src, + unsigned long n) +{ + int size, rc = 0; + u8 status; + + while (n > 0) { + size = zpci_get_max_io_size((u64 __force) src, + (u64 __force) dst, n, + ZPCI_MAX_READ_SIZE); + rc = __pcilg_mio_inuser(dst, src, size, &status); + if (rc) + break; + src += size; + dst += size; + n -= size; + } + if (rc) + zpci_err_mmio(rc, status, (__force u64) dst); + return rc; +} + SYSCALL_DEFINE3(s390_pci_mmio_read, unsigned long, mmio_addr, void __user *, user_buffer, size_t, length) { u8 local_buf[64]; void __iomem *io_addr; void *buf; - unsigned long pfn; + struct vm_area_struct *vma; + pte_t *ptep; + spinlock_t *ptl; long ret; if (!zpci_is_enabled()) @@ -86,29 +273,62 @@ SYSCALL_DEFINE3(s390_pci_mmio_read, unsigned long, mmio_addr, if (length <= 0 || PAGE_SIZE - (mmio_addr & ~PAGE_MASK) < length) return -EINVAL; + + /* + * We only support read access to MIO capable devices if we are on + * a MIO enabled system. Otherwise we would have to check for every + * address if it is a special ZPCI_ADDR and would have to do + * a pfn lookup which we don't need for MIO capable devices. Currently + * ISM devices are the only devices without MIO support and there is no + * known need for accessing these from userspace. + */ + if (static_branch_likely(&have_mio)) { + ret = __memcpy_fromio_inuser( + user_buffer, (const void __iomem *)mmio_addr, + length); + return ret; + } + if (length > 64) { buf = kmalloc(length, GFP_KERNEL); if (!buf) return -ENOMEM; - } else + } else { buf = local_buf; + } - ret = get_pfn(mmio_addr, VM_READ, &pfn); + mmap_read_lock(current->mm); + ret = -EINVAL; + vma = vma_lookup(current->mm, mmio_addr); + if (!vma) + goto out_unlock_mmap; + if (!(vma->vm_flags & (VM_IO | VM_PFNMAP))) + goto out_unlock_mmap; + ret = -EACCES; + if (!(vma->vm_flags & VM_WRITE)) + goto out_unlock_mmap; + + ret = follow_pte(vma->vm_mm, mmio_addr, &ptep, &ptl); if (ret) - goto out; - io_addr = (void __iomem *)((pfn << PAGE_SHIFT) | (mmio_addr & ~PAGE_MASK)); + goto out_unlock_mmap; + + io_addr = (void __iomem *)((pte_pfn(*ptep) << PAGE_SHIFT) | + (mmio_addr & ~PAGE_MASK)); if ((unsigned long) io_addr < ZPCI_IOMAP_ADDR_BASE) { ret = -EFAULT; - goto out; + goto out_unlock_pt; } ret = zpci_memcpy_fromio(buf, io_addr, length); - if (ret) - goto out; - if (copy_to_user(user_buffer, buf, length)) + +out_unlock_pt: + pte_unmap_unlock(ptep, ptl); +out_unlock_mmap: + mmap_read_unlock(current->mm); + + if (!ret && copy_to_user(user_buffer, buf, length)) ret = -EFAULT; -out: if (buf != local_buf) kfree(buf); return ret; diff --git a/arch/s390/pci/pci_sysfs.c b/arch/s390/pci/pci_sysfs.c index a433ba01a317..a0b872b74fe3 100644 --- a/arch/s390/pci/pci_sysfs.c +++ b/arch/s390/pci/pci_sysfs.c @@ -13,6 +13,8 @@ #include <linux/stat.h> #include <linux/pci.h> +#include "../../../drivers/pci/pci.h" + #include <asm/sclp.h> #define zpci_attr(name, fmt, member) \ @@ -31,6 +33,7 @@ zpci_attr(pchid, "0x%04x\n", pchid); zpci_attr(pfgid, "0x%02x\n", pfgid); zpci_attr(vfn, "0x%04x\n", vfn); zpci_attr(pft, "0x%02x\n", pft); +zpci_attr(port, "%d\n", port); zpci_attr(uid, "0x%x\n", uid); zpci_attr(segment0, "0x%02x\n", pfip[0]); zpci_attr(segment1, "0x%02x\n", pfip[1]); @@ -46,34 +49,87 @@ static ssize_t mio_enabled_show(struct device *dev, } static DEVICE_ATTR_RO(mio_enabled); -static ssize_t recover_store(struct device *dev, struct device_attribute *attr, - const char *buf, size_t count) +static int _do_recover(struct pci_dev *pdev, struct zpci_dev *zdev) { - struct pci_dev *pdev = to_pci_dev(dev); - struct zpci_dev *zdev = to_zpci(pdev); + u8 status; int ret; - if (!device_remove_file_self(dev, attr)) - return count; - - pci_lock_rescan_remove(); pci_stop_and_remove_bus_device(pdev); - ret = zpci_disable_device(zdev); - if (ret) - goto error; + if (zdev_enabled(zdev)) { + ret = zpci_disable_device(zdev); + /* + * Due to a z/VM vs LPAR inconsistency in the error + * state the FH may indicate an enabled device but + * disable says the device is already disabled don't + * treat it as an error here. + */ + if (ret == -EINVAL) + ret = 0; + if (ret) + return ret; + } ret = zpci_enable_device(zdev); if (ret) - goto error; + return ret; + + if (zdev->dma_table) { + ret = zpci_register_ioat(zdev, 0, zdev->start_dma, zdev->end_dma, + virt_to_phys(zdev->dma_table), &status); + if (ret) + zpci_disable_device(zdev); + } + return ret; +} - pci_rescan_bus(zdev->bus); +static ssize_t recover_store(struct device *dev, struct device_attribute *attr, + const char *buf, size_t count) +{ + struct kernfs_node *kn; + struct pci_dev *pdev = to_pci_dev(dev); + struct zpci_dev *zdev = to_zpci(pdev); + int ret = 0; + + /* Can't use device_remove_self() here as that would lead us to lock + * the pci_rescan_remove_lock while holding the device' kernfs lock. + * This would create a possible deadlock with disable_slot() which is + * not directly protected by the device' kernfs lock but takes it + * during the device removal which happens under + * pci_rescan_remove_lock. + * + * This is analogous to sdev_store_delete() in + * drivers/scsi/scsi_sysfs.c + */ + kn = sysfs_break_active_protection(&dev->kobj, &attr->attr); + WARN_ON_ONCE(!kn); + + /* Device needs to be configured and state must not change */ + mutex_lock(&zdev->state_lock); + if (zdev->state != ZPCI_FN_STATE_CONFIGURED) + goto out; + + /* device_remove_file() serializes concurrent calls ignoring all but + * the first + */ + device_remove_file(dev, attr); + + /* A concurrent call to recover_store() may slip between + * sysfs_break_active_protection() and the sysfs file removal. + * Once it unblocks from pci_lock_rescan_remove() the original pdev + * will already be removed. + */ + pci_lock_rescan_remove(); + if (pci_dev_is_added(pdev)) { + ret = _do_recover(pdev, zdev); + } + pci_rescan_bus(zdev->zbus->bus); pci_unlock_rescan_remove(); - return count; - -error: - pci_unlock_rescan_remove(); - return ret; +out: + mutex_unlock(&zdev->state_lock); + if (kn) + sysfs_unbreak_active_protection(kn); + return ret ? ret : count; } static DEVICE_ATTR_WO(recover); @@ -109,6 +165,45 @@ static ssize_t report_error_write(struct file *filp, struct kobject *kobj, } static BIN_ATTR(report_error, S_IWUSR, NULL, report_error_write, PAGE_SIZE); +static ssize_t uid_is_unique_show(struct device *dev, + struct device_attribute *attr, char *buf) +{ + return sysfs_emit(buf, "%d\n", zpci_unique_uid ? 1 : 0); +} +static DEVICE_ATTR_RO(uid_is_unique); + +#ifndef CONFIG_DMI +/* analogous to smbios index */ +static ssize_t index_show(struct device *dev, + struct device_attribute *attr, char *buf) +{ + struct zpci_dev *zdev = to_zpci(to_pci_dev(dev)); + u32 index = ~0; + + if (zpci_unique_uid) + index = zdev->uid; + + return sysfs_emit(buf, "%u\n", index); +} +static DEVICE_ATTR_RO(index); + +static umode_t zpci_index_is_visible(struct kobject *kobj, + struct attribute *attr, int n) +{ + return zpci_unique_uid ? attr->mode : 0; +} + +static struct attribute *zpci_ident_attrs[] = { + &dev_attr_index.attr, + NULL, +}; + +static struct attribute_group zpci_ident_attr_group = { + .attrs = zpci_ident_attrs, + .is_visible = zpci_index_is_visible, +}; +#endif + static struct bin_attribute *zpci_bin_attrs[] = { &bin_attr_util_string, &bin_attr_report_error, @@ -121,12 +216,15 @@ static struct attribute *zpci_dev_attrs[] = { &dev_attr_pchid.attr, &dev_attr_pfgid.attr, &dev_attr_pft.attr, + &dev_attr_port.attr, &dev_attr_vfn.attr, &dev_attr_uid.attr, &dev_attr_recover.attr, &dev_attr_mio_enabled.attr, + &dev_attr_uid_is_unique.attr, NULL, }; + static struct attribute_group zpci_attr_group = { .attrs = zpci_dev_attrs, .bin_attrs = zpci_bin_attrs, @@ -147,5 +245,8 @@ static struct attribute_group pfip_attr_group = { const struct attribute_group *zpci_attr_groups[] = { &zpci_attr_group, &pfip_attr_group, +#ifndef CONFIG_DMI + &zpci_ident_attr_group, +#endif NULL, }; |