diff options
Diffstat (limited to 'arch/parisc/kernel')
69 files changed, 3935 insertions, 2855 deletions
diff --git a/arch/parisc/kernel/.gitignore b/arch/parisc/kernel/.gitignore index c5f676c3c224..bbb90f92d051 100644 --- a/arch/parisc/kernel/.gitignore +++ b/arch/parisc/kernel/.gitignore @@ -1 +1,2 @@ +# SPDX-License-Identifier: GPL-2.0-only vmlinux.lds diff --git a/arch/parisc/kernel/Makefile b/arch/parisc/kernel/Makefile index 2663c8f8be11..5ab0467be70a 100644 --- a/arch/parisc/kernel/Makefile +++ b/arch/parisc/kernel/Makefile @@ -3,14 +3,14 @@ # Makefile for arch/parisc/kernel # -extra-y := head.o vmlinux.lds +extra-y := vmlinux.lds -obj-y := cache.o pacache.o setup.o pdt.o traps.o time.o irq.o \ - pa7300lc.o syscall.o entry.o sys_parisc.o firmware.o \ +obj-y := head.o cache.o pacache.o setup.o pdt.o traps.o time.o irq.o \ + syscall.o entry.o sys_parisc.o firmware.o \ ptrace.o hardware.o inventory.o drivers.o alternative.o \ signal.o hpmc.o real2.o parisc_ksyms.o unaligned.o \ process.o processor.o pdc_cons.o pdc_chassis.o unwind.o \ - patch.o + patch.o toc.o toc_asm.o ifdef CONFIG_FUNCTION_TRACER # Do not profile debug and lowlevel utilities @@ -21,6 +21,9 @@ CFLAGS_REMOVE_unwind.o = $(CC_FLAGS_FTRACE) CFLAGS_REMOVE_patch.o = $(CC_FLAGS_FTRACE) endif +CFLAGS_REMOVE_sys_parisc.o = -Wmissing-prototypes -Wmissing-declarations +CFLAGS_REMOVE_sys_parisc32.o = -Wmissing-prototypes -Wmissing-declarations + obj-$(CONFIG_SMP) += smp.o obj-$(CONFIG_PA11) += pci-dma.o obj-$(CONFIG_PCI) += pci.o @@ -31,11 +34,16 @@ obj-$(CONFIG_AUDIT) += audit.o obj64-$(CONFIG_AUDIT) += compat_audit.o # only supported for PCX-W/U in 64-bit mode at the moment obj-$(CONFIG_64BIT) += perf.o perf_asm.o $(obj64-y) -obj-$(CONFIG_PARISC_CPU_TOPOLOGY) += topology.o +obj-$(CONFIG_GENERIC_ARCH_TOPOLOGY) += topology.o obj-$(CONFIG_FUNCTION_TRACER) += ftrace.o obj-$(CONFIG_FUNCTION_GRAPH_TRACER) += ftrace.o obj-$(CONFIG_JUMP_LABEL) += jump_label.o obj-$(CONFIG_KGDB) += kgdb.o obj-$(CONFIG_KPROBES) += kprobes.o -obj-$(CONFIG_KEXEC) += kexec.o relocate_kernel.o +obj-$(CONFIG_KEXEC_CORE) += kexec.o relocate_kernel.o obj-$(CONFIG_KEXEC_FILE) += kexec_file.o + +# vdso +obj-y += vdso.o +obj-$(CONFIG_64BIT) += vdso64/ +obj-y += vdso32/ diff --git a/arch/parisc/kernel/alternative.c b/arch/parisc/kernel/alternative.c index 3c66d5c4d90d..25c4d6c3375d 100644 --- a/arch/parisc/kernel/alternative.c +++ b/arch/parisc/kernel/alternative.c @@ -8,6 +8,7 @@ #include <asm/processor.h> #include <asm/sections.h> #include <asm/alternative.h> +#include <asm/cacheflush.h> #include <linux/module.h> @@ -24,12 +25,29 @@ void __init_or_module apply_alternatives(struct alt_instr *start, { struct alt_instr *entry; int index = 0, applied = 0; - int num_cpus = num_online_cpus(); + int num_cpus = num_present_cpus(); + u16 cond_check; + + cond_check = ALT_COND_ALWAYS | + ((num_cpus == 1) ? ALT_COND_NO_SMP : 0) | + ((cache_info.dc_size == 0) ? ALT_COND_NO_DCACHE : 0) | + ((cache_info.ic_size == 0) ? ALT_COND_NO_ICACHE : 0) | + (running_on_qemu ? ALT_COND_RUN_ON_QEMU : 0) | + ((split_tlb == 0) ? ALT_COND_NO_SPLIT_TLB : 0) | + /* + * If the PDC_MODEL capabilities has Non-coherent IO-PDIR bit + * set (bit #61, big endian), we have to flush and sync every + * time IO-PDIR is changed in Ike/Astro. + */ + (((boot_cpu_data.cpu_type > pcxw_) && + ((boot_cpu_data.pdc.capabilities & PDC_MODEL_IOPDIR_FDC) == 0)) + ? ALT_COND_NO_IOC_FDC : 0); for (entry = start; entry < end; entry++, index++) { - u32 *from, cond, replacement; - s32 len; + u32 *from, replacement; + u16 cond; + s16 len; from = (u32 *)((ulong)&entry->orig_offset + entry->orig_offset); len = entry->len; @@ -38,29 +56,14 @@ void __init_or_module apply_alternatives(struct alt_instr *start, WARN_ON(!cond); - if (cond != ALT_COND_ALWAYS && no_alternatives) + if ((cond & ALT_COND_ALWAYS) == 0 && no_alternatives) continue; pr_debug("Check %d: Cond 0x%x, Replace %02d instructions @ 0x%px with 0x%08x\n", index, cond, len, from, replacement); - if ((cond & ALT_COND_NO_SMP) && (num_cpus != 1)) - continue; - if ((cond & ALT_COND_NO_DCACHE) && (cache_info.dc_size != 0)) - continue; - if ((cond & ALT_COND_NO_ICACHE) && (cache_info.ic_size != 0)) - continue; - if ((cond & ALT_COND_RUN_ON_QEMU) && !running_on_qemu) - continue; - - /* - * If the PDC_MODEL capabilities has Non-coherent IO-PDIR bit - * set (bit #61, big endian), we have to flush and sync every - * time IO-PDIR is changed in Ike/Astro. - */ - if ((cond & ALT_COND_NO_IOC_FDC) && - ((boot_cpu_data.cpu_type <= pcxw_) || - (boot_cpu_data.pdc.capabilities & PDC_MODEL_IOPDIR_FDC))) + /* Bounce out if none of the conditions are true. */ + if ((cond & cond_check) == 0) continue; /* Want to replace pdtlb by a pdtlb,l instruction? */ @@ -106,5 +109,14 @@ void __init apply_alternatives_all(void) apply_alternatives((struct alt_instr *) &__alt_instructions, (struct alt_instr *) &__alt_instructions_end, NULL); + if (cache_info.dc_size == 0 && cache_info.ic_size == 0) { + pr_info("alternatives: optimizing cache-flushes.\n"); + static_branch_disable(&parisc_has_cache); + } + if (cache_info.dc_size == 0) + static_branch_disable(&parisc_has_dcache); + if (cache_info.ic_size == 0) + static_branch_disable(&parisc_has_icache); + set_kernel_text_rw(0); } diff --git a/arch/parisc/kernel/asm-offsets.c b/arch/parisc/kernel/asm-offsets.c index aa79d35dedfa..757816a7bd4b 100644 --- a/arch/parisc/kernel/asm-offsets.c +++ b/arch/parisc/kernel/asm-offsets.c @@ -20,19 +20,17 @@ #include <linux/ptrace.h> #include <linux/hardirq.h> #include <linux/kbuild.h> +#include <linux/pgtable.h> -#include <asm/pgtable.h> +#include <asm/assembly.h> #include <asm/ptrace.h> #include <asm/processor.h> #include <asm/pdc.h> +#include <uapi/asm/sigcontext.h> +#include <asm/ucontext.h> +#include <asm/rt_sigframe.h> #include <linux/uaccess.h> - -#ifdef CONFIG_64BIT -#define FRAME_SIZE 128 -#else -#define FRAME_SIZE 64 -#endif -#define FRAME_ALIGN 64 +#include "signal32.h" /* Add FRAME_SIZE to the size x and align it to y. All definitions * that use align_frame will include space for a frame. @@ -41,14 +39,12 @@ int main(void) { - DEFINE(TASK_THREAD_INFO, offsetof(struct task_struct, stack)); - DEFINE(TASK_STATE, offsetof(struct task_struct, state)); - DEFINE(TASK_FLAGS, offsetof(struct task_struct, flags)); - DEFINE(TASK_SIGPENDING, offsetof(struct task_struct, pending)); - DEFINE(TASK_PTRACE, offsetof(struct task_struct, ptrace)); - DEFINE(TASK_MM, offsetof(struct task_struct, mm)); - DEFINE(TASK_PERSONALITY, offsetof(struct task_struct, personality)); - DEFINE(TASK_PID, offsetof(struct task_struct, pid)); + DEFINE(TASK_TI_FLAGS, offsetof(struct task_struct, thread_info.flags)); +#ifdef CONFIG_SMP + DEFINE(TASK_TI_CPU, offsetof(struct task_struct, thread_info.cpu)); +#endif + DEFINE(TASK_STACK, offsetof(struct task_struct, stack)); + DEFINE(TASK_PAGEFAULT_DISABLED, offsetof(struct task_struct, pagefault_disabled)); BLANK(); DEFINE(TASK_REGS, offsetof(struct task_struct, thread.regs)); DEFINE(TASK_PT_PSW, offsetof(struct task_struct, thread.regs.gr[ 0])); @@ -136,10 +132,6 @@ int main(void) DEFINE(TASK_PT_ISR, offsetof(struct task_struct, thread.regs.isr)); DEFINE(TASK_PT_IOR, offsetof(struct task_struct, thread.regs.ior)); BLANK(); - DEFINE(TASK_SZ, sizeof(struct task_struct)); - /* TASK_SZ_ALGN includes space for a stack frame. */ - DEFINE(TASK_SZ_ALGN, align_frame(sizeof(struct task_struct), FRAME_ALIGN)); - BLANK(); DEFINE(PT_PSW, offsetof(struct pt_regs, gr[ 0])); DEFINE(PT_GR1, offsetof(struct pt_regs, gr[ 1])); DEFINE(PT_GR2, offsetof(struct pt_regs, gr[ 2])); @@ -224,18 +216,21 @@ int main(void) DEFINE(PT_IIR, offsetof(struct pt_regs, iir)); DEFINE(PT_ISR, offsetof(struct pt_regs, isr)); DEFINE(PT_IOR, offsetof(struct pt_regs, ior)); - DEFINE(PT_SIZE, sizeof(struct pt_regs)); /* PT_SZ_ALGN includes space for a stack frame. */ DEFINE(PT_SZ_ALGN, align_frame(sizeof(struct pt_regs), FRAME_ALIGN)); BLANK(); - DEFINE(TI_TASK, offsetof(struct thread_info, task)); DEFINE(TI_FLAGS, offsetof(struct thread_info, flags)); - DEFINE(TI_CPU, offsetof(struct thread_info, cpu)); - DEFINE(TI_SEGMENT, offsetof(struct thread_info, addr_limit)); - DEFINE(TI_PRE_COUNT, offsetof(struct thread_info, preempt_count)); - DEFINE(THREAD_SZ, sizeof(struct thread_info)); - /* THREAD_SZ_ALGN includes space for a stack frame. */ - DEFINE(THREAD_SZ_ALGN, align_frame(sizeof(struct thread_info), FRAME_ALIGN)); + DEFINE(TI_PRE_COUNT, offsetof(struct task_struct, thread_info.preempt_count)); + BLANK(); + DEFINE(ASM_SIGFRAME_SIZE, PARISC_RT_SIGFRAME_SIZE); + DEFINE(SIGFRAME_CONTEXT_REGS, offsetof(struct rt_sigframe, uc.uc_mcontext) - PARISC_RT_SIGFRAME_SIZE); +#ifdef CONFIG_64BIT + DEFINE(ASM_SIGFRAME_SIZE32, PARISC_RT_SIGFRAME_SIZE32); + DEFINE(SIGFRAME_CONTEXT_REGS32, offsetof(struct compat_rt_sigframe, uc.uc_mcontext) - PARISC_RT_SIGFRAME_SIZE32); +#else + DEFINE(ASM_SIGFRAME_SIZE32, PARISC_RT_SIGFRAME_SIZE); + DEFINE(SIGFRAME_CONTEXT_REGS32, offsetof(struct rt_sigframe, uc.uc_mcontext) - PARISC_RT_SIGFRAME_SIZE); +#endif BLANK(); DEFINE(ICACHE_BASE, offsetof(struct pdc_cache_info, ic_base)); DEFINE(ICACHE_STRIDE, offsetof(struct pdc_cache_info, ic_stride)); @@ -268,7 +263,6 @@ int main(void) DEFINE(ASM_BITS_PER_PGD, BITS_PER_PGD); DEFINE(ASM_BITS_PER_PMD, BITS_PER_PMD); DEFINE(ASM_BITS_PER_PTE, BITS_PER_PTE); - DEFINE(ASM_PGD_PMD_OFFSET, -(PAGE_SIZE << PGD_ORDER)); DEFINE(ASM_PMD_ENTRY, ((PAGE_OFFSET & PMD_MASK) >> PMD_SHIFT)); DEFINE(ASM_PGD_ENTRY, PAGE_OFFSET >> PGDIR_SHIFT); DEFINE(ASM_PGD_ENTRY_SIZE, PGD_ENTRY_SIZE); @@ -281,6 +275,8 @@ int main(void) * and kernel data on physical huge pages */ #ifdef CONFIG_HUGETLB_PAGE DEFINE(HUGEPAGE_SIZE, 1UL << REAL_HPAGE_SHIFT); +#elif !defined(CONFIG_64BIT) + DEFINE(HUGEPAGE_SIZE, 4*1024*1024); #else DEFINE(HUGEPAGE_SIZE, PAGE_SIZE); #endif diff --git a/arch/parisc/kernel/audit.c b/arch/parisc/kernel/audit.c index 9eb47b2225d2..375cd73b5281 100644 --- a/arch/parisc/kernel/audit.c +++ b/arch/parisc/kernel/audit.c @@ -40,20 +40,21 @@ int audit_classify_arch(int arch) int audit_classify_syscall(int abi, unsigned syscall) { -#ifdef CONFIG_COMPAT - extern int parisc32_classify_syscall(unsigned); - if (abi == AUDIT_ARCH_PARISC) - return parisc32_classify_syscall(syscall); -#endif switch (syscall) { case __NR_open: - return 2; + return AUDITSC_OPEN; case __NR_openat: - return 3; + return AUDITSC_OPENAT; case __NR_execve: - return 5; + return AUDITSC_EXECVE; + case __NR_openat2: + return AUDITSC_OPENAT2; default: - return 0; +#ifdef CONFIG_COMPAT + if (abi == AUDIT_ARCH_PARISC) + return AUDITSC_COMPAT; +#endif + return AUDITSC_NATIVE; } } diff --git a/arch/parisc/kernel/cache.c b/arch/parisc/kernel/cache.c index 1eedfecc5137..422f3e1e6d9c 100644 --- a/arch/parisc/kernel/cache.c +++ b/arch/parisc/kernel/cache.c @@ -19,15 +19,17 @@ #include <linux/pagemap.h> #include <linux/sched.h> #include <linux/sched/mm.h> +#include <linux/syscalls.h> #include <asm/pdc.h> #include <asm/cache.h> #include <asm/cacheflush.h> #include <asm/tlbflush.h> #include <asm/page.h> -#include <asm/pgalloc.h> #include <asm/processor.h> #include <asm/sections.h> #include <asm/shmparam.h> +#include <asm/mmu_context.h> +#include <asm/cachectl.h> int split_tlb __ro_after_init; int dcache_stride __ro_after_init; @@ -39,6 +41,9 @@ EXPORT_SYMBOL(flush_dcache_page_asm); void purge_dcache_page_asm(unsigned long phys_addr, unsigned long vaddr); void flush_icache_page_asm(unsigned long phys_addr, unsigned long vaddr); +/* Internal implementation in arch/parisc/kernel/pacache.S */ +void flush_data_cache_local(void *); /* flushes local data-cache only */ +void flush_instruction_cache_local(void); /* flushes local code-cache only */ /* On some machines (i.e., ones with the Merced bus), there can be * only a single PxTLB broadcast at a time; this must be guaranteed @@ -47,47 +52,53 @@ void flush_icache_page_asm(unsigned long phys_addr, unsigned long vaddr); */ DEFINE_SPINLOCK(pa_tlb_flush_lock); -/* Swapper page setup lock. */ -DEFINE_SPINLOCK(pa_swapper_pg_lock); - #if defined(CONFIG_64BIT) && defined(CONFIG_SMP) int pa_serialize_tlb_flushes __ro_after_init; #endif struct pdc_cache_info cache_info __ro_after_init; #ifndef CONFIG_PA20 -static struct pdc_btlb_info btlb_info __ro_after_init; +struct pdc_btlb_info btlb_info; #endif -#ifdef CONFIG_SMP -void -flush_data_cache(void) +DEFINE_STATIC_KEY_TRUE(parisc_has_cache); +DEFINE_STATIC_KEY_TRUE(parisc_has_dcache); +DEFINE_STATIC_KEY_TRUE(parisc_has_icache); + +static void cache_flush_local_cpu(void *dummy) { - on_each_cpu(flush_data_cache_local, NULL, 1); + if (static_branch_likely(&parisc_has_icache)) + flush_instruction_cache_local(); + if (static_branch_likely(&parisc_has_dcache)) + flush_data_cache_local(NULL); } -void -flush_instruction_cache(void) + +void flush_cache_all_local(void) { - on_each_cpu(flush_instruction_cache_local, NULL, 1); + cache_flush_local_cpu(NULL); } -#endif -void -flush_cache_all_local(void) +void flush_cache_all(void) +{ + if (static_branch_likely(&parisc_has_cache)) + on_each_cpu(cache_flush_local_cpu, NULL, 1); +} + +static inline void flush_data_cache(void) { - flush_instruction_cache_local(NULL); - flush_data_cache_local(NULL); + if (static_branch_likely(&parisc_has_dcache)) + on_each_cpu(flush_data_cache_local, NULL, 1); } -EXPORT_SYMBOL(flush_cache_all_local); -/* Virtual address of pfn. */ + +/* Kernel virtual address of pfn. */ #define pfn_va(pfn) __va(PFN_PHYS(pfn)) -void -update_mmu_cache(struct vm_area_struct *vma, unsigned long address, pte_t *ptep) +void __update_cache(pte_t pte) { - unsigned long pfn = pte_pfn(*ptep); - struct page *page; + unsigned long pfn = pte_pfn(pte); + struct folio *folio; + unsigned int nr; /* We don't have pte special. As a result, we can be called with an invalid pfn and we don't need to flush the kernel dcache page. @@ -95,13 +106,17 @@ update_mmu_cache(struct vm_area_struct *vma, unsigned long address, pte_t *ptep) if (!pfn_valid(pfn)) return; - page = pfn_to_page(pfn); - if (page_mapping_file(page) && - test_bit(PG_dcache_dirty, &page->flags)) { - flush_kernel_dcache_page_addr(pfn_va(pfn)); - clear_bit(PG_dcache_dirty, &page->flags); + folio = page_folio(pfn_to_page(pfn)); + pfn = folio_pfn(folio); + nr = folio_nr_pages(folio); + if (folio_flush_mapping(folio) && + test_bit(PG_dcache_dirty, &folio->flags)) { + while (nr--) + flush_kernel_dcache_page_addr(pfn_va(pfn + nr)); + clear_bit(PG_dcache_dirty, &folio->flags); } else if (parisc_requires_coherency()) - flush_kernel_dcache_page_addr(pfn_va(pfn)); + while (nr--) + flush_kernel_dcache_page_addr(pfn_va(pfn + nr)); } void @@ -113,11 +128,13 @@ show_cache_info(struct seq_file *m) cache_info.ic_size/1024 ); if (cache_info.dc_loop != 1) snprintf(buf, 32, "%lu-way associative", cache_info.dc_loop); - seq_printf(m, "D-cache\t\t: %ld KB (%s%s, %s)\n", + seq_printf(m, "D-cache\t\t: %ld KB (%s%s, %s, alias=%d)\n", cache_info.dc_size/1024, (cache_info.dc_conf.cc_wt ? "WT":"WB"), (cache_info.dc_conf.cc_sh ? ", shared I/D":""), - ((cache_info.dc_loop == 1) ? "direct mapped" : buf)); + ((cache_info.dc_loop == 1) ? "direct mapped" : buf), + cache_info.dc_conf.cc_alias + ); seq_printf(m, "ITLB entries\t: %ld\n" "DTLB entries\t: %ld%s\n", cache_info.it_size, cache_info.dt_size, @@ -247,11 +264,9 @@ parisc_cache_init(void) icache_stride = CAFL_STRIDE(cache_info.ic_conf); #undef CAFL_STRIDE -#ifndef CONFIG_PA20 - if (pdc_btlb_info(&btlb_info) < 0) { - memset(&btlb_info, 0, sizeof btlb_info); - } -#endif + /* stride needs to be non-zero, otherwise cache flushes will not work */ + WARN_ON(cache_info.dc_size && dcache_stride == 0); + WARN_ON(cache_info.ic_size && icache_stride == 0); if ((boot_cpu_data.pdc.capabilities & PDC_MODEL_NVA_MASK) == PDC_MODEL_NVA_UNSUPPORTED) { @@ -262,7 +277,7 @@ parisc_cache_init(void) } } -void __init disable_sr_hashing(void) +void disable_sr_hashing(void) { int srhash_type, retval; unsigned long space_bits; @@ -304,6 +319,8 @@ static inline void __flush_cache_page(struct vm_area_struct *vma, unsigned long vmaddr, unsigned long physaddr) { + if (!static_branch_likely(&parisc_has_cache)) + return; preempt_disable(); flush_dcache_page_asm(physaddr, vmaddr); if (vma->vm_flags & VM_EXEC) @@ -311,86 +328,202 @@ __flush_cache_page(struct vm_area_struct *vma, unsigned long vmaddr, preempt_enable(); } -static inline void -__purge_cache_page(struct vm_area_struct *vma, unsigned long vmaddr, - unsigned long physaddr) +static void flush_user_cache_page(struct vm_area_struct *vma, unsigned long vmaddr) { + unsigned long flags, space, pgd, prot; +#ifdef CONFIG_TLB_PTLOCK + unsigned long pgd_lock; +#endif + + vmaddr &= PAGE_MASK; + preempt_disable(); - purge_dcache_page_asm(physaddr, vmaddr); + + /* Set context for flush */ + local_irq_save(flags); + prot = mfctl(8); + space = mfsp(SR_USER); + pgd = mfctl(25); +#ifdef CONFIG_TLB_PTLOCK + pgd_lock = mfctl(28); +#endif + switch_mm_irqs_off(NULL, vma->vm_mm, NULL); + local_irq_restore(flags); + + flush_user_dcache_range_asm(vmaddr, vmaddr + PAGE_SIZE); if (vma->vm_flags & VM_EXEC) - flush_icache_page_asm(physaddr, vmaddr); + flush_user_icache_range_asm(vmaddr, vmaddr + PAGE_SIZE); + flush_tlb_page(vma, vmaddr); + + /* Restore previous context */ + local_irq_save(flags); +#ifdef CONFIG_TLB_PTLOCK + mtctl(pgd_lock, 28); +#endif + mtctl(pgd, 25); + mtsp(space, SR_USER); + mtctl(prot, 8); + local_irq_restore(flags); + preempt_enable(); } -void flush_dcache_page(struct page *page) +void flush_icache_pages(struct vm_area_struct *vma, struct page *page, + unsigned int nr) +{ + void *kaddr = page_address(page); + + for (;;) { + flush_kernel_dcache_page_addr(kaddr); + flush_kernel_icache_page(kaddr); + if (--nr == 0) + break; + kaddr += PAGE_SIZE; + } +} + +static inline pte_t *get_ptep(struct mm_struct *mm, unsigned long addr) +{ + pte_t *ptep = NULL; + pgd_t *pgd = mm->pgd; + p4d_t *p4d; + pud_t *pud; + pmd_t *pmd; + + if (!pgd_none(*pgd)) { + p4d = p4d_offset(pgd, addr); + if (!p4d_none(*p4d)) { + pud = pud_offset(p4d, addr); + if (!pud_none(*pud)) { + pmd = pmd_offset(pud, addr); + if (!pmd_none(*pmd)) + ptep = pte_offset_map(pmd, addr); + } + } + } + return ptep; +} + +static inline bool pte_needs_flush(pte_t pte) { - struct address_space *mapping = page_mapping_file(page); - struct vm_area_struct *mpnt; - unsigned long offset; + return (pte_val(pte) & (_PAGE_PRESENT | _PAGE_ACCESSED | _PAGE_NO_CACHE)) + == (_PAGE_PRESENT | _PAGE_ACCESSED); +} + +void flush_dcache_folio(struct folio *folio) +{ + struct address_space *mapping = folio_flush_mapping(folio); + struct vm_area_struct *vma; unsigned long addr, old_addr = 0; + void *kaddr; + unsigned long count = 0; + unsigned long i, nr, flags; pgoff_t pgoff; if (mapping && !mapping_mapped(mapping)) { - set_bit(PG_dcache_dirty, &page->flags); + set_bit(PG_dcache_dirty, &folio->flags); return; } - flush_kernel_dcache_page(page); + nr = folio_nr_pages(folio); + kaddr = folio_address(folio); + for (i = 0; i < nr; i++) + flush_kernel_dcache_page_addr(kaddr + i * PAGE_SIZE); if (!mapping) return; - pgoff = page->index; + pgoff = folio->index; - /* We have carefully arranged in arch_get_unmapped_area() that + /* + * We have carefully arranged in arch_get_unmapped_area() that * *any* mappings of a file are always congruently mapped (whether * declared as MAP_PRIVATE or MAP_SHARED), so we only need - * to flush one address here for them all to become coherent */ - - flush_dcache_mmap_lock(mapping); - vma_interval_tree_foreach(mpnt, &mapping->i_mmap, pgoff, pgoff) { - offset = (pgoff - mpnt->vm_pgoff) << PAGE_SHIFT; - addr = mpnt->vm_start + offset; - - /* The TLB is the engine of coherence on parisc: The - * CPU is entitled to speculate any page with a TLB - * mapping, so here we kill the mapping then flush the - * page along a special flush only alias mapping. - * This guarantees that the page is no-longer in the - * cache for any process and nor may it be - * speculatively read in (until the user or kernel - * specifically accesses it, of course) */ - - flush_tlb_page(mpnt, addr); - if (old_addr == 0 || (old_addr & (SHM_COLOUR - 1)) - != (addr & (SHM_COLOUR - 1))) { - __flush_cache_page(mpnt, addr, page_to_phys(page)); - if (parisc_requires_coherency() && old_addr) - printk(KERN_ERR "INEQUIVALENT ALIASES 0x%lx and 0x%lx in file %pD\n", old_addr, addr, mpnt->vm_file); - old_addr = addr; + * to flush one address here for them all to become coherent + * on machines that support equivalent aliasing + */ + flush_dcache_mmap_lock_irqsave(mapping, flags); + vma_interval_tree_foreach(vma, &mapping->i_mmap, pgoff, pgoff + nr - 1) { + unsigned long offset = pgoff - vma->vm_pgoff; + unsigned long pfn = folio_pfn(folio); + + addr = vma->vm_start; + nr = folio_nr_pages(folio); + if (offset > -nr) { + pfn -= offset; + nr += offset; + } else { + addr += offset * PAGE_SIZE; + } + if (addr + nr * PAGE_SIZE > vma->vm_end) + nr = (vma->vm_end - addr) / PAGE_SIZE; + + if (parisc_requires_coherency()) { + for (i = 0; i < nr; i++) { + pte_t *ptep = get_ptep(vma->vm_mm, + addr + i * PAGE_SIZE); + if (!ptep) + continue; + if (pte_needs_flush(*ptep)) + flush_user_cache_page(vma, + addr + i * PAGE_SIZE); + /* Optimise accesses to the same table? */ + pte_unmap(ptep); + } + } else { + /* + * The TLB is the engine of coherence on parisc: + * The CPU is entitled to speculate any page + * with a TLB mapping, so here we kill the + * mapping then flush the page along a special + * flush only alias mapping. This guarantees that + * the page is no-longer in the cache for any + * process and nor may it be speculatively read + * in (until the user or kernel specifically + * accesses it, of course) + */ + for (i = 0; i < nr; i++) + flush_tlb_page(vma, addr + i * PAGE_SIZE); + if (old_addr == 0 || (old_addr & (SHM_COLOUR - 1)) + != (addr & (SHM_COLOUR - 1))) { + for (i = 0; i < nr; i++) + __flush_cache_page(vma, + addr + i * PAGE_SIZE, + (pfn + i) * PAGE_SIZE); + /* + * Software is allowed to have any number + * of private mappings to a page. + */ + if (!(vma->vm_flags & VM_SHARED)) + continue; + if (old_addr) + pr_err("INEQUIVALENT ALIASES 0x%lx and 0x%lx in file %pD\n", + old_addr, addr, vma->vm_file); + if (nr == folio_nr_pages(folio)) + old_addr = addr; + } } + WARN_ON(++count == 4096); } - flush_dcache_mmap_unlock(mapping); + flush_dcache_mmap_unlock_irqrestore(mapping, flags); } -EXPORT_SYMBOL(flush_dcache_page); +EXPORT_SYMBOL(flush_dcache_folio); /* Defined in arch/parisc/kernel/pacache.S */ EXPORT_SYMBOL(flush_kernel_dcache_range_asm); -EXPORT_SYMBOL(flush_kernel_dcache_page_asm); -EXPORT_SYMBOL(flush_data_cache_local); EXPORT_SYMBOL(flush_kernel_icache_range_asm); #define FLUSH_THRESHOLD 0x80000 /* 0.5MB */ static unsigned long parisc_cache_flush_threshold __ro_after_init = FLUSH_THRESHOLD; #define FLUSH_TLB_THRESHOLD (16*1024) /* 16 KiB minimum TLB threshold */ -static unsigned long parisc_tlb_flush_threshold __ro_after_init = FLUSH_TLB_THRESHOLD; +static unsigned long parisc_tlb_flush_threshold __ro_after_init = ~0UL; void __init parisc_setup_cache_timing(void) { unsigned long rangetime, alltime; - unsigned long size, start; - unsigned long threshold; + unsigned long size; + unsigned long threshold, threshold2; alltime = mfctl(16); flush_data_cache(); @@ -404,11 +537,16 @@ void __init parisc_setup_cache_timing(void) printk(KERN_DEBUG "Whole cache flush %lu cycles, flushing %lu bytes %lu cycles\n", alltime, size, rangetime); - threshold = L1_CACHE_ALIGN(size * alltime / rangetime); - if (threshold > cache_info.dc_size) - threshold = cache_info.dc_size; - if (threshold) - parisc_cache_flush_threshold = threshold; + threshold = L1_CACHE_ALIGN((unsigned long)((uint64_t)size * alltime / rangetime)); + pr_info("Calculated flush threshold is %lu KiB\n", + threshold/1024); + + /* + * The threshold computed above isn't very reliable. The following + * heuristic works reasonably well on c8000/rp3440. + */ + threshold2 = cache_info.dc_size * num_online_cpus(); + parisc_cache_flush_threshold = threshold2; printk(KERN_INFO "Cache flush threshold set to %lu KiB\n", parisc_cache_flush_threshold/1024); @@ -423,14 +561,9 @@ void __init parisc_setup_cache_timing(void) goto set_tlb_threshold; } - size = 0; - start = (unsigned long) _text; + size = (unsigned long)_end - (unsigned long)_text; rangetime = mfctl(16); - while (start < (unsigned long) _end) { - flush_tlb_kernel_range(start, start + PAGE_SIZE); - start += PAGE_SIZE; - size += PAGE_SIZE; - } + flush_tlb_kernel_range((unsigned long)_text, (unsigned long)_end); rangetime = mfctl(16) - rangetime; alltime = mfctl(16); @@ -445,8 +578,11 @@ void __init parisc_setup_cache_timing(void) threshold/1024); set_tlb_threshold: - if (threshold > parisc_tlb_flush_threshold) + if (threshold > FLUSH_TLB_THRESHOLD) parisc_tlb_flush_threshold = threshold; + else + parisc_tlb_flush_threshold = FLUSH_TLB_THRESHOLD; + printk(KERN_INFO "TLB flush threshold set to %lu KiB\n", parisc_tlb_flush_threshold/1024); } @@ -455,30 +591,64 @@ extern void purge_kernel_dcache_page_asm(unsigned long); extern void clear_user_page_asm(void *, unsigned long); extern void copy_user_page_asm(void *, void *, unsigned long); -void flush_kernel_dcache_page_addr(void *addr) +void flush_kernel_dcache_page_addr(const void *addr) { unsigned long flags; flush_kernel_dcache_page_asm(addr); purge_tlb_start(flags); - pdtlb_kernel(addr); + pdtlb(SR_KERNEL, addr); purge_tlb_end(flags); } EXPORT_SYMBOL(flush_kernel_dcache_page_addr); -void copy_user_page(void *vto, void *vfrom, unsigned long vaddr, - struct page *pg) +static void flush_cache_page_if_present(struct vm_area_struct *vma, + unsigned long vmaddr, unsigned long pfn) { - /* Copy using kernel mapping. No coherency is needed (all in - kunmap) for the `to' page. However, the `from' page needs to - be flushed through a mapping equivalent to the user mapping - before it can be accessed through the kernel mapping. */ - preempt_disable(); - flush_dcache_page_asm(__pa(vfrom), vaddr); - copy_page_asm(vto, vfrom); - preempt_enable(); + bool needs_flush = false; + pte_t *ptep; + + /* + * The pte check is racy and sometimes the flush will trigger + * a non-access TLB miss. Hopefully, the page has already been + * flushed. + */ + ptep = get_ptep(vma->vm_mm, vmaddr); + if (ptep) { + needs_flush = pte_needs_flush(*ptep); + pte_unmap(ptep); + } + if (needs_flush) + flush_cache_page(vma, vmaddr, pfn); +} + +void copy_user_highpage(struct page *to, struct page *from, + unsigned long vaddr, struct vm_area_struct *vma) +{ + void *kto, *kfrom; + + kfrom = kmap_local_page(from); + kto = kmap_local_page(to); + flush_cache_page_if_present(vma, vaddr, page_to_pfn(from)); + copy_page_asm(kto, kfrom); + kunmap_local(kto); + kunmap_local(kfrom); +} + +void copy_to_user_page(struct vm_area_struct *vma, struct page *page, + unsigned long user_vaddr, void *dst, void *src, int len) +{ + flush_cache_page_if_present(vma, user_vaddr, page_to_pfn(page)); + memcpy(dst, src, len); + flush_kernel_dcache_range_asm((unsigned long)dst, (unsigned long)dst + len); +} + +void copy_from_user_page(struct vm_area_struct *vma, struct page *page, + unsigned long user_vaddr, void *dst, void *src, int len) +{ + flush_cache_page_if_present(vma, user_vaddr, page_to_pfn(page)); + memcpy(dst, src, len); } -EXPORT_SYMBOL(copy_user_page); /* __flush_tlb_range() * @@ -500,152 +670,127 @@ int __flush_tlb_range(unsigned long sid, unsigned long start, but cause a purge request to be broadcast to other TLBs. */ while (start < end) { purge_tlb_start(flags); - mtsp(sid, 1); - pdtlb(start); - pitlb(start); + mtsp(sid, SR_TEMP1); + pdtlb(SR_TEMP1, start); + pitlb(SR_TEMP1, start); purge_tlb_end(flags); start += PAGE_SIZE; } return 0; } -static void cacheflush_h_tmp_function(void *dummy) -{ - flush_cache_all_local(); -} - -void flush_cache_all(void) +static void flush_cache_pages(struct vm_area_struct *vma, unsigned long start, unsigned long end) { - on_each_cpu(cacheflush_h_tmp_function, NULL, 1); + unsigned long addr, pfn; + pte_t *ptep; + + for (addr = start; addr < end; addr += PAGE_SIZE) { + bool needs_flush = false; + /* + * The vma can contain pages that aren't present. Although + * the pte search is expensive, we need the pte to find the + * page pfn and to check whether the page should be flushed. + */ + ptep = get_ptep(vma->vm_mm, addr); + if (ptep) { + needs_flush = pte_needs_flush(*ptep); + pfn = pte_pfn(*ptep); + pte_unmap(ptep); + } + if (needs_flush) { + if (parisc_requires_coherency()) { + flush_user_cache_page(vma, addr); + } else { + if (WARN_ON(!pfn_valid(pfn))) + return; + __flush_cache_page(vma, addr, PFN_PHYS(pfn)); + } + } + } } static inline unsigned long mm_total_size(struct mm_struct *mm) { struct vm_area_struct *vma; unsigned long usize = 0; + VMA_ITERATOR(vmi, mm, 0); - for (vma = mm->mmap; vma; vma = vma->vm_next) + for_each_vma(vmi, vma) { + if (usize >= parisc_cache_flush_threshold) + break; usize += vma->vm_end - vma->vm_start; - return usize; -} - -static inline pte_t *get_ptep(pgd_t *pgd, unsigned long addr) -{ - pte_t *ptep = NULL; - - if (!pgd_none(*pgd)) { - p4d_t *p4d = p4d_offset(pgd, addr); - if (!p4d_none(*p4d)) { - pud_t *pud = pud_offset(p4d, addr); - if (!pud_none(*pud)) { - pmd_t *pmd = pmd_offset(pud, addr); - if (!pmd_none(*pmd)) - ptep = pte_offset_map(pmd, addr); - } - } } - return ptep; + return usize; } void flush_cache_mm(struct mm_struct *mm) { struct vm_area_struct *vma; - pgd_t *pgd; - - /* Flushing the whole cache on each cpu takes forever on - rp3440, etc. So, avoid it if the mm isn't too big. */ - if ((!IS_ENABLED(CONFIG_SMP) || !arch_irqs_disabled()) && - mm_total_size(mm) >= parisc_cache_flush_threshold) { - if (mm->context) - flush_tlb_all(); + VMA_ITERATOR(vmi, mm, 0); + + /* + * Flushing the whole cache on each cpu takes forever on + * rp3440, etc. So, avoid it if the mm isn't too big. + * + * Note that we must flush the entire cache on machines + * with aliasing caches to prevent random segmentation + * faults. + */ + if (!parisc_requires_coherency() + || mm_total_size(mm) >= parisc_cache_flush_threshold) { + if (WARN_ON(IS_ENABLED(CONFIG_SMP) && arch_irqs_disabled())) + return; + flush_tlb_all(); flush_cache_all(); return; } - if (mm->context == mfsp(3)) { - for (vma = mm->mmap; vma; vma = vma->vm_next) { - flush_user_dcache_range_asm(vma->vm_start, vma->vm_end); - if (vma->vm_flags & VM_EXEC) - flush_user_icache_range_asm(vma->vm_start, vma->vm_end); - flush_tlb_range(vma, vma->vm_start, vma->vm_end); - } - return; - } - - pgd = mm->pgd; - for (vma = mm->mmap; vma; vma = vma->vm_next) { - unsigned long addr; - - for (addr = vma->vm_start; addr < vma->vm_end; - addr += PAGE_SIZE) { - unsigned long pfn; - pte_t *ptep = get_ptep(pgd, addr); - if (!ptep) - continue; - pfn = pte_pfn(*ptep); - if (!pfn_valid(pfn)) - continue; - if (unlikely(mm->context)) { - flush_tlb_page(vma, addr); - __flush_cache_page(vma, addr, PFN_PHYS(pfn)); - } else { - __purge_cache_page(vma, addr, PFN_PHYS(pfn)); - } - } - } + /* Flush mm */ + for_each_vma(vmi, vma) + flush_cache_pages(vma, vma->vm_start, vma->vm_end); } -void flush_cache_range(struct vm_area_struct *vma, - unsigned long start, unsigned long end) +void flush_cache_range(struct vm_area_struct *vma, unsigned long start, unsigned long end) { - pgd_t *pgd; - unsigned long addr; - - if ((!IS_ENABLED(CONFIG_SMP) || !arch_irqs_disabled()) && - end - start >= parisc_cache_flush_threshold) { - if (vma->vm_mm->context) - flush_tlb_range(vma, start, end); + if (!parisc_requires_coherency() + || end - start >= parisc_cache_flush_threshold) { + if (WARN_ON(IS_ENABLED(CONFIG_SMP) && arch_irqs_disabled())) + return; + flush_tlb_range(vma, start, end); flush_cache_all(); return; } - if (vma->vm_mm->context == mfsp(3)) { - flush_user_dcache_range_asm(start, end); - if (vma->vm_flags & VM_EXEC) - flush_user_icache_range_asm(start, end); - flush_tlb_range(vma, start, end); - return; - } + flush_cache_pages(vma, start, end); +} - pgd = vma->vm_mm->pgd; - for (addr = vma->vm_start; addr < vma->vm_end; addr += PAGE_SIZE) { - unsigned long pfn; - pte_t *ptep = get_ptep(pgd, addr); - if (!ptep) - continue; - pfn = pte_pfn(*ptep); - if (pfn_valid(pfn)) { - if (unlikely(vma->vm_mm->context)) { - flush_tlb_page(vma, addr); - __flush_cache_page(vma, addr, PFN_PHYS(pfn)); - } else { - __purge_cache_page(vma, addr, PFN_PHYS(pfn)); - } - } - } +void flush_cache_page(struct vm_area_struct *vma, unsigned long vmaddr, unsigned long pfn) +{ + if (WARN_ON(!pfn_valid(pfn))) + return; + if (parisc_requires_coherency()) + flush_user_cache_page(vma, vmaddr); + else + __flush_cache_page(vma, vmaddr, PFN_PHYS(pfn)); } -void -flush_cache_page(struct vm_area_struct *vma, unsigned long vmaddr, unsigned long pfn) +void flush_anon_page(struct vm_area_struct *vma, struct page *page, unsigned long vmaddr) { - if (pfn_valid(pfn)) { - if (likely(vma->vm_mm->context)) { - flush_tlb_page(vma, vmaddr); - __flush_cache_page(vma, vmaddr, PFN_PHYS(pfn)); - } else { - __purge_cache_page(vma, vmaddr, PFN_PHYS(pfn)); - } + if (!PageAnon(page)) + return; + + if (parisc_requires_coherency()) { + if (vma->vm_flags & VM_SHARED) + flush_data_cache(); + else + flush_user_cache_page(vma, vmaddr); + return; } + + flush_tlb_page(vma, vmaddr); + preempt_disable(); + flush_dcache_page_asm(page_to_phys(page), vmaddr); + preempt_enable(); } void flush_kernel_vmap_range(void *vaddr, int size) @@ -670,6 +815,9 @@ void invalidate_kernel_vmap_range(void *vaddr, int size) unsigned long start = (unsigned long)vaddr; unsigned long end = start + size; + /* Ensure DMA is complete */ + asm_syncdma(); + if ((!IS_ENABLED(CONFIG_SMP) || !arch_irqs_disabled()) && (unsigned long)size >= parisc_cache_flush_threshold) { flush_tlb_kernel_range(start, end); @@ -681,3 +829,50 @@ void invalidate_kernel_vmap_range(void *vaddr, int size) flush_tlb_kernel_range(start, end); } EXPORT_SYMBOL(invalidate_kernel_vmap_range); + + +SYSCALL_DEFINE3(cacheflush, unsigned long, addr, unsigned long, bytes, + unsigned int, cache) +{ + unsigned long start, end; + ASM_EXCEPTIONTABLE_VAR(error); + + if (bytes == 0) + return 0; + if (!access_ok((void __user *) addr, bytes)) + return -EFAULT; + + end = addr + bytes; + + if (cache & DCACHE) { + start = addr; + __asm__ __volatile__ ( +#ifdef CONFIG_64BIT + "1: cmpb,*<<,n %0,%2,1b\n" +#else + "1: cmpb,<<,n %0,%2,1b\n" +#endif + " fic,m %3(%4,%0)\n" + "2: sync\n" + ASM_EXCEPTIONTABLE_ENTRY_EFAULT(1b, 2b, "%1") + : "+r" (start), "+r" (error) + : "r" (end), "r" (dcache_stride), "i" (SR_USER)); + } + + if (cache & ICACHE && error == 0) { + start = addr; + __asm__ __volatile__ ( +#ifdef CONFIG_64BIT + "1: cmpb,*<<,n %0,%2,1b\n" +#else + "1: cmpb,<<,n %0,%2,1b\n" +#endif + " fdc,m %3(%4,%0)\n" + "2: sync\n" + ASM_EXCEPTIONTABLE_ENTRY_EFAULT(1b, 2b, "%1") + : "+r" (start), "+r" (error) + : "r" (end), "r" (icache_stride), "i" (SR_USER)); + } + + return error; +} diff --git a/arch/parisc/kernel/compat_audit.c b/arch/parisc/kernel/compat_audit.c index 20c39c9d86a9..3ac53f1ab860 100644 --- a/arch/parisc/kernel/compat_audit.c +++ b/arch/parisc/kernel/compat_audit.c @@ -1,4 +1,5 @@ // SPDX-License-Identifier: GPL-2.0 +#include <linux/audit_arch.h> #include <asm/unistd.h> unsigned int parisc32_dir_class[] = { @@ -25,17 +26,3 @@ unsigned int parisc32_signal_class[] = { #include <asm-generic/audit_signal.h> ~0U }; - -int parisc32_classify_syscall(unsigned syscall) -{ - switch (syscall) { - case __NR_open: - return 2; - case __NR_openat: - return 3; - case __NR_execve: - return 5; - default: - return 1; - } -} diff --git a/arch/parisc/kernel/drivers.c b/arch/parisc/kernel/drivers.c index 3b330e58a4f0..ac19d685e4a5 100644 --- a/arch/parisc/kernel/drivers.c +++ b/arch/parisc/kernel/drivers.c @@ -4,7 +4,7 @@ * * Copyright (c) 1999 The Puffin Group * Copyright (c) 2001 Matthew Wilcox for Hewlett Packard - * Copyright (c) 2001 Helge Deller <deller@gmx.de> + * Copyright (c) 2001-2023 Helge Deller <deller@gmx.de> * Copyright (c) 2001,2002 Ryan Bradetich * Copyright (c) 2004-2005 Thibaut VARENE <varenet@parisc-linux.org> * @@ -30,6 +30,7 @@ #include <linux/spinlock.h> #include <linux/string.h> #include <linux/export.h> +#include <linux/dma-map-ops.h> #include <asm/hardware.h> #include <asm/io.h> #include <asm/pdc.h> @@ -73,13 +74,13 @@ static int descend_children(struct device * dev, void * data) } /** - * for_each_padev - Iterate over all devices in the tree - * @fn: Function to call for each device. - * @data: Data to pass to the called function. + * for_each_padev - Iterate over all devices in the tree + * @fn: Function to call for each device. + * @data: Data to pass to the called function. * - * This performs a depth-first traversal of the tree, calling the - * function passed for each node. It calls the function for parents - * before children. + * This performs a depth-first traversal of the tree, calling the + * function passed for each node. It calls the function for parents + * before children. */ static int for_each_padev(int (*fn)(struct device *, void *), void * data) @@ -132,14 +133,13 @@ static int parisc_driver_probe(struct device *dev) return rc; } -static int __exit parisc_driver_remove(struct device *dev) +static void __exit parisc_driver_remove(struct device *dev) { struct parisc_device *pa_dev = to_parisc_device(dev); struct parisc_driver *pa_drv = to_parisc_driver(dev->driver); + if (pa_drv->remove) pa_drv->remove(pa_dev); - - return 0; } @@ -280,7 +280,7 @@ int __init machine_has_merced_bus(void) /** * find_pa_parent_type - Find a parent of a specific type - * @dev: The device to start searching from + * @padev: The device to start searching from * @type: The device type to search for. * * Walks up the device tree looking for a device of the specified type. @@ -344,8 +344,8 @@ static char *print_hwpath(struct hardware_path *path, char *output) /** * print_pa_hwpath - Returns hardware path for PA devices - * dev: The device to return the path for - * output: Pointer to a previously-allocated array to place the path in. + * @dev: The device to return the path for + * @output: Pointer to a previously-allocated array to place the path in. * * This function fills in the output array with a human-readable path * to a PA device. This string is compatible with that used by PDC, and @@ -379,8 +379,8 @@ EXPORT_SYMBOL(get_pci_node_path); /** * print_pci_hwpath - Returns hardware path for PCI devices - * dev: The device to return the path for - * output: Pointer to a previously-allocated array to place the path in. + * @dev: The device to return the path for + * @output: Pointer to a previously-allocated array to place the path in. * * This function fills in the output array with a human-readable path * to a PCI device. This string is compatible with that used by PDC, and @@ -415,7 +415,8 @@ static void setup_bus_id(struct parisc_device *padev) dev_set_name(&padev->dev, name); } -struct parisc_device * __init create_tree_node(char id, struct device *parent) +static struct parisc_device * __init create_tree_node(char id, + struct device *parent) { struct parisc_device *dev = kzalloc(sizeof(*dev), GFP_KERNEL); if (!dev) @@ -520,7 +521,6 @@ alloc_pa_dev(unsigned long hpa, struct hardware_path *mod_path) dev->id.hversion_rev = iodc_data[1] & 0x0f; dev->id.sversion = ((iodc_data[4] & 0x0f) << 16) | (iodc_data[5] << 8) | iodc_data[6]; - dev->hpa.name = parisc_pathname(dev); dev->hpa.start = hpa; /* This is awkward. The STI spec says that gfx devices may occupy * 32MB or 64MB. Unfortunately, we don't know how to tell whether @@ -534,10 +534,10 @@ alloc_pa_dev(unsigned long hpa, struct hardware_path *mod_path) dev->hpa.end = hpa + 0xfff; } dev->hpa.flags = IORESOURCE_MEM; - name = parisc_hardware_description(&dev->id); - if (name) { - strlcpy(dev->name, name, sizeof(dev->name)); - } + dev->hpa.name = dev->name; + name = parisc_hardware_description(&dev->id) ? : "unknown"; + snprintf(dev->name, sizeof(dev->name), "%s [%s]", + name, parisc_pathname(dev)); /* Silently fail things like mouse ports which are subsumed within * the keyboard controller @@ -553,7 +553,7 @@ static int parisc_generic_match(struct device *dev, struct device_driver *drv) return match_device(to_parisc_driver(drv), to_parisc_device(dev)); } -static ssize_t make_modalias(struct device *dev, char *buf) +static ssize_t make_modalias(const struct device *dev, char *buf) { const struct parisc_device *padev = to_parisc_device(dev); const struct parisc_device_id *id = &padev->id; @@ -563,7 +563,7 @@ static ssize_t make_modalias(struct device *dev, char *buf) (u32)id->sversion); } -static int parisc_uevent(struct device *dev, struct kobj_uevent_env *env) +static int parisc_uevent(const struct device *dev, struct kobj_uevent_env *env) { const struct parisc_device *padev; char modalias[40]; @@ -618,7 +618,7 @@ static struct attribute *parisc_device_attrs[] = { }; ATTRIBUTE_GROUPS(parisc_device); -struct bus_type parisc_bus_type = { +const struct bus_type parisc_bus_type = { .name = "parisc", .match = parisc_generic_match, .uevent = parisc_uevent, @@ -742,7 +742,7 @@ parse_tree_node(struct device *parent, int index, struct hardware_path *modpath) }; if (device_for_each_child(parent, &recurse_data, descend_children)) - /* nothing */; + { /* nothing */ } return d.dev; } @@ -772,8 +772,8 @@ EXPORT_SYMBOL(hwpath_to_device); /** * device_to_hwpath - Populates the hwpath corresponding to the given device. - * @param dev the target device - * @param path pointer to a previously allocated hwpath struct to be filled in + * @dev: the target device + * @path: pointer to a previously allocated hwpath struct to be filled in */ void device_to_hwpath(struct device *dev, struct hardware_path *path) { @@ -810,7 +810,7 @@ EXPORT_SYMBOL(device_to_hwpath); static void walk_native_bus(unsigned long io_io_low, unsigned long io_io_high, struct device *parent); -static void walk_lower_bus(struct parisc_device *dev) +static void __init walk_lower_bus(struct parisc_device *dev) { unsigned long io_io_low, io_io_high; @@ -883,15 +883,13 @@ void __init walk_central_bus(void) &root); } -static void print_parisc_device(struct parisc_device *dev) +static __init void print_parisc_device(struct parisc_device *dev) { - char hw_path[64]; - static int count; + static int count __initdata; - print_pa_hwpath(dev, hw_path); - pr_info("%d. %s at 0x%px [%s] { %d, 0x%x, 0x%.3x, 0x%.5x }", - ++count, dev->name, (void*) dev->hpa.start, hw_path, dev->id.hw_type, - dev->id.hversion_rev, dev->id.hversion, dev->id.sversion); + pr_info("%d. %s at %pap { type:%d, hv:%#x, sv:%#x, rev:%#x }", + ++count, dev->name, &(dev->hpa.start), dev->id.hw_type, + dev->id.hversion, dev->id.sversion, dev->id.hversion_rev); if (dev->num_addrs) { int k; @@ -927,10 +925,10 @@ static __init void qemu_header(void) pr_info("#define PARISC_MODEL \"%s\"\n\n", boot_cpu_data.pdc.sys_model_name); - pr_info("#define PARISC_PDC_MODEL 0x%lx, 0x%lx, 0x%lx, " - "0x%lx, 0x%lx, 0x%lx, 0x%lx, 0x%lx, 0x%lx\n\n", #define p ((unsigned long *)&boot_cpu_data.pdc.model) - p[0], p[1], p[2], p[3], p[4], p[5], p[6], p[7], p[8]); + pr_info("#define PARISC_PDC_MODEL 0x%lx, 0x%lx, 0x%lx, " + "0x%lx, 0x%lx, 0x%lx, 0x%lx, 0x%lx, 0x%lx, 0x%lx\n\n", + p[0], p[1], p[2], p[3], p[4], p[5], p[6], p[7], p[8], p[9]); #undef p pr_info("#define PARISC_PDC_VERSION 0x%04lx\n\n", @@ -1006,6 +1004,9 @@ static __init int qemu_print_iodc_data(struct device *lin_dev, void *data) pr_info("\n"); + /* Prevent hung task messages when printing on serial console */ + cond_resched(); + pr_info("#define HPA_%08lx_DESCRIPTION \"%s\"\n", hpa, parisc_hardware_description(&dev->id)); @@ -1080,7 +1081,7 @@ static __init int qemu_print_iodc_data(struct device *lin_dev, void *data) -static int print_one_device(struct device * dev, void * data) +static __init int print_one_device(struct device * dev, void * data) { struct parisc_device * pdev = to_parisc_device(dev); diff --git a/arch/parisc/kernel/entry.S b/arch/parisc/kernel/entry.S index b96d74496977..ab23e61a6f01 100644 --- a/arch/parisc/kernel/entry.S +++ b/arch/parisc/kernel/entry.S @@ -19,15 +19,16 @@ #include <asm/psw.h> #include <asm/cache.h> /* for L1_CACHE_SHIFT */ #include <asm/assembly.h> /* for LDREG/STREG defines */ -#include <asm/pgtable.h> #include <asm/signal.h> #include <asm/unistd.h> #include <asm/ldcw.h> #include <asm/traps.h> #include <asm/thread_info.h> #include <asm/alternative.h> +#include <asm/spinlock_types.h> #include <linux/linkage.h> +#include <linux/pgtable.h> #ifdef CONFIG_64BIT .level 2.0w @@ -35,10 +36,27 @@ .level 2.0 #endif - .import pa_tlb_lock,data - .macro load_pa_tlb_lock reg - mfctl %cr25,\reg - addil L%(PAGE_SIZE << (PGD_ALLOC_ORDER - 1)),\reg +/* + * We need seven instructions after a TLB insert for it to take effect. + * The PA8800/PA8900 processors are an exception and need 12 instructions. + * The RFI changes both IAOQ_Back and IAOQ_Front, so it counts as one. + */ +#ifdef CONFIG_64BIT +#define NUM_PIPELINE_INSNS 12 +#else +#define NUM_PIPELINE_INSNS 7 +#endif + + /* Insert num nops */ + .macro insert_nops num + .rept \num + nop + .endr + .endm + + /* Get aligned page_table_lock address for this mm from cr28/tr4 */ + .macro get_ptl reg + mfctl %cr28,\reg .endm /* space_to_prot macro creates a prot id from a space id */ @@ -52,30 +70,6 @@ extrd,u \spc,(64 - (SPACEID_SHIFT)),32,\prot .endm #endif - - /* Switch to virtual mapping, trashing only %r1 */ - .macro virt_map - /* pcxt_ssm_bug */ - rsm PSW_SM_I, %r0 /* barrier for "Relied upon Translation */ - mtsp %r0, %sr4 - mtsp %r0, %sr5 - mtsp %r0, %sr6 - tovirt_r1 %r29 - load32 KERNEL_PSW, %r1 - - rsm PSW_SM_QUIET,%r0 /* second "heavy weight" ctl op */ - mtctl %r0, %cr17 /* Clear IIASQ tail */ - mtctl %r0, %cr17 /* Clear IIASQ head */ - mtctl %r1, %ipsw - load32 4f, %r1 - mtctl %r1, %cr18 /* Set IIAOQ tail */ - ldo 4(%r1), %r1 - mtctl %r1, %cr18 /* Set IIAOQ head */ - rfir - nop -4: - .endm - /* * The "get_stack" macros are responsible for determining the * kernel stack value. @@ -88,8 +82,8 @@ * Need to set up a kernel stack, so call the * get_stack_use_cr30 macro to set up a pointer * to the pt_regs structure contained within the - * task pointer pointed to by cr30. Set the stack - * pointer to point to the end of the task structure. + * task pointer pointed to by cr30. Load the stack + * pointer from the task structure. * * Note that we use shadowed registers for temps until * we can save %r26 and %r29. %r26 is used to preserve @@ -101,8 +95,6 @@ * or handle_interruption. %r29 is used to hold a pointer * the register save area, and once again, it needs to * be a non-shadowed register so that it survives the rfir. - * - * N.B. TASK_SZ_ALGN and PT_SZ_ALGN include space for a stack frame. */ .macro get_stack_use_cr30 @@ -111,12 +103,11 @@ copy %r30, %r17 mfctl %cr30, %r1 - ldo THREAD_SZ_ALGN(%r1), %r30 - mtsp %r0,%sr7 + tophys %r1,%r9 /* task_struct */ + LDREG TASK_STACK(%r9),%r30 + ldo PT_SZ_ALGN(%r30),%r30 + mtsp %r0,%sr7 /* clear sr7 after kernel stack was set! */ mtsp %r16,%sr3 - tophys %r1,%r9 - LDREG TI_TASK(%r9), %r1 /* thread_info -> task_struct */ - tophys %r1,%r9 ldo TASK_REGS(%r9),%r9 STREG %r17,PT_GR30(%r9) STREG %r29,PT_GR29(%r9) @@ -394,69 +385,47 @@ */ .macro L2_ptep pmd,pte,index,va,fault #if CONFIG_PGTABLE_LEVELS == 3 - extru \va,31-ASM_PMD_SHIFT,ASM_BITS_PER_PMD,\index + extru_safe \va,31-ASM_PMD_SHIFT,ASM_BITS_PER_PMD,\index #else -# if defined(CONFIG_64BIT) - extrd,u \va,63-ASM_PGDIR_SHIFT,ASM_BITS_PER_PGD,\index - #else - # if PAGE_SIZE > 4096 - extru \va,31-ASM_PGDIR_SHIFT,32-ASM_PGDIR_SHIFT,\index - # else - extru \va,31-ASM_PGDIR_SHIFT,ASM_BITS_PER_PGD,\index - # endif -# endif + extru_safe \va,31-ASM_PGDIR_SHIFT,ASM_BITS_PER_PGD,\index #endif dep %r0,31,PAGE_SHIFT,\pmd /* clear offset */ +#if CONFIG_PGTABLE_LEVELS < 3 copy %r0,\pte +#endif ldw,s \index(\pmd),\pmd bb,>=,n \pmd,_PxD_PRESENT_BIT,\fault dep %r0,31,PxD_FLAG_SHIFT,\pmd /* clear flags */ SHLREG \pmd,PxD_VALUE_SHIFT,\pmd - extru \va,31-PAGE_SHIFT,ASM_BITS_PER_PTE,\index + extru_safe \va,31-PAGE_SHIFT,ASM_BITS_PER_PTE,\index dep %r0,31,PAGE_SHIFT,\pmd /* clear offset */ shladd \index,BITS_PER_PTE_ENTRY,\pmd,\pmd /* pmd is now pte */ .endm - /* Look up PTE in a 3-Level scheme. - * - * Here we implement a Hybrid L2/L3 scheme: we allocate the - * first pmd adjacent to the pgd. This means that we can - * subtract a constant offset to get to it. The pmd and pgd - * sizes are arranged so that a single pmd covers 4GB (giving - * a full LP64 process access to 8TB) so our lookups are - * effectively L2 for the first 4GB of the kernel (i.e. for - * all ILP32 processes and all the kernel for machines with - * under 4GB of memory) */ + /* Look up PTE in a 3-Level scheme. */ .macro L3_ptep pgd,pte,index,va,fault -#if CONFIG_PGTABLE_LEVELS == 3 /* we might have a 2-Level scheme, e.g. with 16kb page size */ +#if CONFIG_PGTABLE_LEVELS == 3 + copy %r0,\pte extrd,u \va,63-ASM_PGDIR_SHIFT,ASM_BITS_PER_PGD,\index - extrd,u,*= \va,63-ASM_PGDIR_SHIFT,64-ASM_PGDIR_SHIFT,%r0 ldw,s \index(\pgd),\pgd - extrd,u,*= \va,63-ASM_PGDIR_SHIFT,64-ASM_PGDIR_SHIFT,%r0 bb,>=,n \pgd,_PxD_PRESENT_BIT,\fault - extrd,u,*= \va,63-ASM_PGDIR_SHIFT,64-ASM_PGDIR_SHIFT,%r0 - shld \pgd,PxD_VALUE_SHIFT,\index - extrd,u,*= \va,63-ASM_PGDIR_SHIFT,64-ASM_PGDIR_SHIFT,%r0 - copy \index,\pgd - extrd,u,*<> \va,63-ASM_PGDIR_SHIFT,64-ASM_PGDIR_SHIFT,%r0 - ldo ASM_PGD_PMD_OFFSET(\pgd),\pgd + shld \pgd,PxD_VALUE_SHIFT,\pgd #endif L2_ptep \pgd,\pte,\index,\va,\fault .endm - /* Acquire pa_tlb_lock lock and check page is present. */ - .macro tlb_lock spc,ptp,pte,tmp,tmp1,fault -#ifdef CONFIG_SMP + /* Acquire page_table_lock and check page is present. */ + .macro ptl_lock spc,ptp,pte,tmp,tmp1,fault +#ifdef CONFIG_TLB_PTLOCK 98: cmpib,COND(=),n 0,\spc,2f - load_pa_tlb_lock \tmp + get_ptl \tmp 1: LDCW 0(\tmp),\tmp1 cmpib,COND(=) 0,\tmp1,1b nop LDREG 0(\ptp),\pte bb,<,n \pte,_PAGE_PRESENT_BIT,3f - LDCW 0(\tmp),\tmp1 b \fault - stw \spc,0(\tmp) + stw \tmp1,0(\tmp) 99: ALTERNATIVE(98b, 99b, ALT_COND_NO_SMP, INSN_NOP) #endif 2: LDREG 0(\ptp),\pte @@ -464,23 +433,20 @@ 3: .endm - /* Release pa_tlb_lock lock without reloading lock address. */ - .macro tlb_unlock0 spc,tmp,tmp1 -#ifdef CONFIG_SMP -98: or,COND(=) %r0,\spc,%r0 - LDCW 0(\tmp),\tmp1 + /* Release page_table_lock if for user space. We use an ordered + store to ensure all prior accesses are performed prior to + releasing the lock. Note stw may not be executed, so we + provide one extra nop when CONFIG_TLB_PTLOCK is defined. */ + .macro ptl_unlock spc,tmp,tmp2 +#ifdef CONFIG_TLB_PTLOCK +98: get_ptl \tmp + ldi __ARCH_SPIN_LOCK_UNLOCKED_VAL, \tmp2 or,COND(=) %r0,\spc,%r0 - stw \spc,0(\tmp) + stw,ma \tmp2,0(\tmp) 99: ALTERNATIVE(98b, 99b, ALT_COND_NO_SMP, INSN_NOP) -#endif - .endm - - /* Release pa_tlb_lock lock. */ - .macro tlb_unlock1 spc,tmp,tmp1 -#ifdef CONFIG_SMP -98: load_pa_tlb_lock \tmp -99: ALTERNATIVE(98b, 99b, ALT_COND_NO_SMP, INSN_NOP) - tlb_unlock0 \spc,\tmp,\tmp1 + insert_nops NUM_PIPELINE_INSNS - 4 +#else + insert_nops NUM_PIPELINE_INSNS - 1 #endif .endm @@ -509,13 +475,13 @@ * to a CPU TLB 4k PFN (4k => 12 bits to shift) */ #define PAGE_ADD_SHIFT (PAGE_SHIFT-12) #define PAGE_ADD_HUGE_SHIFT (REAL_HPAGE_SHIFT-12) + #define PFN_START_BIT (63-ASM_PFN_PTE_SHIFT+(63-58)-PAGE_ADD_SHIFT) /* Drop prot bits and convert to page addr for iitlbt and idtlbt */ .macro convert_for_tlb_insert20 pte,tmp #ifdef CONFIG_HUGETLB_PAGE copy \pte,\tmp - extrd,u \tmp,(63-ASM_PFN_PTE_SHIFT)+(63-58)+PAGE_ADD_SHIFT,\ - 64-PAGE_SHIFT-PAGE_ADD_SHIFT,\pte + extrd,u \tmp,PFN_START_BIT,PFN_START_BIT+1,\pte depdi _PAGE_SIZE_ENCODING_DEFAULT,63,\ (63-58)+PAGE_ADD_SHIFT,\pte @@ -523,8 +489,7 @@ depdi _HUGE_PAGE_SIZE_ENCODING_DEFAULT,63,\ (63-58)+PAGE_ADD_HUGE_SHIFT,\pte #else /* Huge pages disabled */ - extrd,u \pte,(63-ASM_PFN_PTE_SHIFT)+(63-58)+PAGE_ADD_SHIFT,\ - 64-PAGE_SHIFT-PAGE_ADD_SHIFT,\pte + extrd,u \pte,PFN_START_BIT,PFN_START_BIT+1,\pte depdi _PAGE_SIZE_ENCODING_DEFAULT,63,\ (63-58)+PAGE_ADD_SHIFT,\pte #endif @@ -546,6 +511,10 @@ * Finally, _PAGE_READ goes in the top bit of PL1 (so we * trigger an access rights trap in user space if the user * tries to read an unreadable page */ +#if _PAGE_SPECIAL_BIT == _PAGE_DMB_BIT + /* need to drop DMB bit, as it's used as SPECIAL flag */ + depi 0,_PAGE_SPECIAL_BIT,1,\pte +#endif depd \pte,8,7,\prot /* PAGE_USER indicates the page can be read with user privileges, @@ -576,6 +545,10 @@ * makes the tlb entry for the differently formatted pa11 * insertion instructions */ .macro make_insert_tlb_11 spc,pte,prot +#if _PAGE_SPECIAL_BIT == _PAGE_DMB_BIT + /* need to drop DMB bit, as it's used as SPECIAL flag */ + depi 0,_PAGE_SPECIAL_BIT,1,\pte +#endif zdep \spc,30,15,\prot dep \pte,8,7,\prot extru,= \pte,_PAGE_NO_CACHE_BIT,1,%r0 @@ -601,8 +574,9 @@ extrd,s \pte,63,25,\pte .endm - /* The alias region is an 8MB aligned 16MB to do clear and - * copy user pages at addresses congruent with the user + /* The alias region is comprised of a pair of 4 MB regions + * aligned to 8 MB. It is used to clear/copy/flush user pages + * using kernel virtual addresses congruent with the user * virtual address. * * To use the alias page, you set %r26 up with the to TLB @@ -612,13 +586,8 @@ .macro do_alias spc,tmp,tmp1,va,pte,prot,fault,patype cmpib,COND(<>),n 0,\spc,\fault ldil L%(TMPALIAS_MAP_START),\tmp -#if defined(CONFIG_64BIT) && (TMPALIAS_MAP_START >= 0x80000000) - /* on LP64, ldi will sign extend into the upper 32 bits, - * which is behaviour we don't want */ - depdi 0,31,32,\tmp -#endif copy \va,\tmp1 - depi 0,31,23,\tmp1 + depi_safe 0,31,TMPALIAS_SIZE_BITS+1,\tmp1 cmpb,COND(<>),n \tmp,\tmp1,\fault mfctl %cr19,\tmp /* iir */ /* get the opcode (first six bits) into \tmp */ @@ -651,13 +620,13 @@ * OK, it is in the temp alias region, check whether "from" or "to". * Check "subtle" note in pacache.S re: r23/r26. */ -#ifdef CONFIG_64BIT - extrd,u,*= \va,41,1,%r0 -#else - extrw,u,= \va,9,1,%r0 -#endif + extrw,u,= \va,31-TMPALIAS_SIZE_BITS,1,%r0 or,COND(tr) %r23,%r0,\pte or %r26,%r0,\pte + + /* convert phys addr in \pte (from r23 or r26) to tlb insert format */ + SHRREG \pte,PAGE_SHIFT+PAGE_ADD_SHIFT-5, \pte + depi_safe _PAGE_SIZE_ENCODING_DEFAULT, 31,5, \pte .endm @@ -769,7 +738,7 @@ ENTRY(ret_from_kernel_thread) BL schedule_tail, %r2 nop - LDREG TI_TASK-THREAD_SZ_ALGN-FRAME_SIZE(%r30), %r1 + mfctl %cr30,%r1 /* task_struct */ LDREG TASK_PT_GR25(%r1), %r26 #ifdef CONFIG_64BIT LDREG TASK_PT_GR27(%r1), %r27 @@ -800,7 +769,6 @@ ENTRY_CFI(_switch_to) STREG %r30, TASK_PT_KSP(%r26) LDREG TASK_PT_KSP(%r25), %r30 - LDREG TASK_THREAD_INFO(%r25), %r25 bv %r0(%r2) mtctl %r25,%cr30 @@ -831,17 +799,16 @@ ENDPROC_CFI(_switch_to) .align PAGE_SIZE ENTRY_CFI(syscall_exit_rfi) - mfctl %cr30,%r16 - LDREG TI_TASK(%r16), %r16 /* thread_info -> task_struct */ + mfctl %cr30,%r16 /* task_struct */ ldo TASK_REGS(%r16),%r16 /* Force iaoq to userspace, as the user has had access to our current * context via sigcontext. Also Filter the PSW for the same reason. */ LDREG PT_IAOQ0(%r16),%r19 - depi 3,31,2,%r19 + depi PRIV_USER,31,2,%r19 STREG %r19,PT_IAOQ0(%r16) LDREG PT_IAOQ1(%r16),%r19 - depi 3,31,2,%r19 + depi PRIV_USER,31,2,%r19 STREG %r19,PT_IAOQ1(%r16) LDREG PT_PSW(%r16),%r19 load32 USER_PSW_MASK,%r1 @@ -877,15 +844,15 @@ ENTRY_CFI(syscall_exit_rfi) ENTRY(intr_return) /* check for reschedule */ mfctl %cr30,%r1 - LDREG TI_FLAGS(%r1),%r19 /* sched.h: TIF_NEED_RESCHED */ + LDREG TASK_TI_FLAGS(%r1),%r19 /* sched.h: TIF_NEED_RESCHED */ bb,<,n %r19,31-TIF_NEED_RESCHED,intr_do_resched /* forward */ .import do_notify_resume,code intr_check_sig: /* As above */ mfctl %cr30,%r1 - LDREG TI_FLAGS(%r1),%r19 - ldi (_TIF_SIGPENDING|_TIF_NOTIFY_RESUME), %r20 + LDREG TASK_TI_FLAGS(%r1),%r19 + ldi (_TIF_USER_WORK_MASK & ~_TIF_NEED_RESCHED), %r20 and,COND(<>) %r19, %r20, %r0 b,n intr_restore /* skip past if we've nothing to do */ @@ -897,20 +864,20 @@ intr_check_sig: * Only do signals if we are returning to user space */ LDREG PT_IASQ0(%r16), %r20 - cmpib,COND(=),n LINUX_GATEWAY_SPACE, %r20, intr_restore /* backward */ + cmpib,COND(=),n LINUX_GATEWAY_SPACE, %r20, intr_restore /* forward */ LDREG PT_IASQ1(%r16), %r20 - cmpib,COND(=),n LINUX_GATEWAY_SPACE, %r20, intr_restore /* backward */ - - /* NOTE: We need to enable interrupts if we have to deliver - * signals. We used to do this earlier but it caused kernel - * stack overflows. */ - ssm PSW_SM_I, %r0 + cmpib,COND(=),n LINUX_GATEWAY_SPACE, %r20, intr_restore /* forward */ copy %r0, %r25 /* long in_syscall = 0 */ #ifdef CONFIG_64BIT ldo -16(%r30),%r29 /* Reference param save area */ #endif + /* NOTE: We need to enable interrupts if we have to deliver + * signals. We used to do this earlier but it caused kernel + * stack overflows. */ + ssm PSW_SM_I, %r0 + BL do_notify_resume,%r2 copy %r16, %r26 /* struct pt_regs *regs */ @@ -940,14 +907,14 @@ intr_restore: rfi nop -#ifndef CONFIG_PREEMPT +#ifndef CONFIG_PREEMPTION # define intr_do_preempt intr_restore -#endif /* !CONFIG_PREEMPT */ +#endif /* !CONFIG_PREEMPTION */ .import schedule,code intr_do_resched: /* Only call schedule on return to userspace. If we're returning - * to kernel space, we may schedule if CONFIG_PREEMPT, otherwise + * to kernel space, we may schedule if CONFIG_PREEMPTION, otherwise * we jump back to intr_restore. */ LDREG PT_IASQ0(%r16), %r20 @@ -979,15 +946,15 @@ intr_do_resched: * and preempt_count is 0. otherwise, we continue on * our merry way back to the current running task. */ -#ifdef CONFIG_PREEMPT +#ifdef CONFIG_PREEMPTION .import preempt_schedule_irq,code intr_do_preempt: rsm PSW_SM_I, %r0 /* disable interrupts */ /* current_thread_info()->preempt_count */ mfctl %cr30, %r1 - LDREG TI_PRE_COUNT(%r1), %r19 - cmpib,COND(<>) 0, %r19, intr_restore /* if preempt_count > 0 */ + ldw TI_PRE_COUNT(%r1), %r19 + cmpib,<> 0, %r19, intr_restore /* if preempt_count > 0 */ nop /* prev insn branched backwards */ /* check if we interrupted a critical path */ @@ -995,11 +962,18 @@ intr_do_preempt: bb,<,n %r20, 31 - PSW_SM_I, intr_restore nop + /* ssm PSW_SM_I done later in intr_restore */ +#ifdef CONFIG_MLONGCALLS + ldil L%intr_restore, %r2 + load32 preempt_schedule_irq, %r1 + bv %r0(%r1) + ldo R%intr_restore(%r2), %r2 +#else + ldil L%intr_restore, %r1 BL preempt_schedule_irq, %r2 - nop - - b,n intr_restore /* ssm PSW_SM_I done by intr_restore */ -#endif /* CONFIG_PREEMPT */ + ldo R%intr_restore(%r1), %r2 +#endif +#endif /* CONFIG_PREEMPTION */ /* * External interrupts. @@ -1156,14 +1130,14 @@ dtlb_miss_20w: L3_ptep ptp,pte,t0,va,dtlb_check_alias_20w - tlb_lock spc,ptp,pte,t0,t1,dtlb_check_alias_20w + ptl_lock spc,ptp,pte,t0,t1,dtlb_check_alias_20w update_accessed ptp,pte,t0,t1 make_insert_tlb spc,pte,prot,t1 idtlbt pte,prot - tlb_unlock1 spc,t0,t1 + ptl_unlock spc,t0,t1 rfir nop @@ -1172,6 +1146,7 @@ dtlb_check_alias_20w: idtlbt pte,prot + insert_nops NUM_PIPELINE_INSNS - 1 rfir nop @@ -1182,14 +1157,14 @@ nadtlb_miss_20w: L3_ptep ptp,pte,t0,va,nadtlb_check_alias_20w - tlb_lock spc,ptp,pte,t0,t1,nadtlb_check_alias_20w + ptl_lock spc,ptp,pte,t0,t1,nadtlb_check_alias_20w update_accessed ptp,pte,t0,t1 make_insert_tlb spc,pte,prot,t1 idtlbt pte,prot - tlb_unlock1 spc,t0,t1 + ptl_unlock spc,t0,t1 rfir nop @@ -1198,6 +1173,7 @@ nadtlb_check_alias_20w: idtlbt pte,prot + insert_nops NUM_PIPELINE_INSNS - 1 rfir nop @@ -1210,7 +1186,7 @@ dtlb_miss_11: L2_ptep ptp,pte,t0,va,dtlb_check_alias_11 - tlb_lock spc,ptp,pte,t0,t1,dtlb_check_alias_11 + ptl_lock spc,ptp,pte,t0,t1,dtlb_check_alias_11 update_accessed ptp,pte,t0,t1 make_insert_tlb_11 spc,pte,prot @@ -1223,7 +1199,7 @@ dtlb_miss_11: mtsp t1, %sr1 /* Restore sr1 */ - tlb_unlock1 spc,t0,t1 + ptl_unlock spc,t0,t1 rfir nop @@ -1233,6 +1209,7 @@ dtlb_check_alias_11: idtlba pte,(va) idtlbp prot,(va) + insert_nops NUM_PIPELINE_INSNS - 1 rfir nop @@ -1243,7 +1220,7 @@ nadtlb_miss_11: L2_ptep ptp,pte,t0,va,nadtlb_check_alias_11 - tlb_lock spc,ptp,pte,t0,t1,nadtlb_check_alias_11 + ptl_lock spc,ptp,pte,t0,t1,nadtlb_check_alias_11 update_accessed ptp,pte,t0,t1 make_insert_tlb_11 spc,pte,prot @@ -1256,7 +1233,7 @@ nadtlb_miss_11: mtsp t1, %sr1 /* Restore sr1 */ - tlb_unlock1 spc,t0,t1 + ptl_unlock spc,t0,t1 rfir nop @@ -1266,6 +1243,7 @@ nadtlb_check_alias_11: idtlba pte,(va) idtlbp prot,(va) + insert_nops NUM_PIPELINE_INSNS - 1 rfir nop @@ -1276,7 +1254,7 @@ dtlb_miss_20: L2_ptep ptp,pte,t0,va,dtlb_check_alias_20 - tlb_lock spc,ptp,pte,t0,t1,dtlb_check_alias_20 + ptl_lock spc,ptp,pte,t0,t1,dtlb_check_alias_20 update_accessed ptp,pte,t0,t1 make_insert_tlb spc,pte,prot,t1 @@ -1285,7 +1263,7 @@ dtlb_miss_20: idtlbt pte,prot - tlb_unlock1 spc,t0,t1 + ptl_unlock spc,t0,t1 rfir nop @@ -1294,6 +1272,7 @@ dtlb_check_alias_20: idtlbt pte,prot + insert_nops NUM_PIPELINE_INSNS - 1 rfir nop @@ -1304,7 +1283,7 @@ nadtlb_miss_20: L2_ptep ptp,pte,t0,va,nadtlb_check_alias_20 - tlb_lock spc,ptp,pte,t0,t1,nadtlb_check_alias_20 + ptl_lock spc,ptp,pte,t0,t1,nadtlb_check_alias_20 update_accessed ptp,pte,t0,t1 make_insert_tlb spc,pte,prot,t1 @@ -1313,7 +1292,7 @@ nadtlb_miss_20: idtlbt pte,prot - tlb_unlock1 spc,t0,t1 + ptl_unlock spc,t0,t1 rfir nop @@ -1322,6 +1301,7 @@ nadtlb_check_alias_20: idtlbt pte,prot + insert_nops NUM_PIPELINE_INSNS - 1 rfir nop @@ -1330,74 +1310,12 @@ nadtlb_check_alias_20: nadtlb_emulate: /* - * Non access misses can be caused by fdc,fic,pdc,lpa,probe and - * probei instructions. We don't want to fault for these - * instructions (not only does it not make sense, it can cause - * deadlocks, since some flushes are done with the mmap - * semaphore held). If the translation doesn't exist, we can't - * insert a translation, so have to emulate the side effects - * of the instruction. Since we don't insert a translation - * we can get a lot of faults during a flush loop, so it makes - * sense to try to do it here with minimum overhead. We only - * emulate fdc,fic,pdc,probew,prober instructions whose base - * and index registers are not shadowed. We defer everything - * else to the "slow" path. + * Non-access misses can be caused by fdc,fic,pdc,lpa,probe and + * probei instructions. The kernel no longer faults doing flushes. + * Use of lpa and probe instructions is rare. Given the issue + * with shadow registers, we defer everything to the "slow" path. */ - - mfctl %cr19,%r9 /* Get iir */ - - /* PA 2.0 Arch Ref. Book pg 382 has a good description of the insn bits. - Checks for fdc,fdce,pdc,"fic,4f",prober,probeir,probew, probeiw */ - - /* Checks for fdc,fdce,pdc,"fic,4f" only */ - ldi 0x280,%r16 - and %r9,%r16,%r17 - cmpb,<>,n %r16,%r17,nadtlb_probe_check - bb,>=,n %r9,26,nadtlb_nullify /* m bit not set, just nullify */ - BL get_register,%r25 - extrw,u %r9,15,5,%r8 /* Get index register # */ - cmpib,COND(=),n -1,%r1,nadtlb_fault /* have to use slow path */ - copy %r1,%r24 - BL get_register,%r25 - extrw,u %r9,10,5,%r8 /* Get base register # */ - cmpib,COND(=),n -1,%r1,nadtlb_fault /* have to use slow path */ - BL set_register,%r25 - add,l %r1,%r24,%r1 /* doesn't affect c/b bits */ - -nadtlb_nullify: - mfctl %ipsw,%r8 - ldil L%PSW_N,%r9 - or %r8,%r9,%r8 /* Set PSW_N */ - mtctl %r8,%ipsw - - rfir - nop - - /* - When there is no translation for the probe address then we - must nullify the insn and return zero in the target register. - This will indicate to the calling code that it does not have - write/read privileges to this address. - - This should technically work for prober and probew in PA 1.1, - and also probe,r and probe,w in PA 2.0 - - WARNING: USE ONLY NON-SHADOW REGISTERS WITH PROBE INSN! - THE SLOW-PATH EMULATION HAS NOT BEEN WRITTEN YET. - - */ -nadtlb_probe_check: - ldi 0x80,%r16 - and %r9,%r16,%r17 - cmpb,<>,n %r16,%r17,nadtlb_fault /* Must be probe,[rw]*/ - BL get_register,%r25 /* Find the target register */ - extrw,u %r9,31,5,%r8 /* Get target register */ - cmpib,COND(=),n -1,%r1,nadtlb_fault /* have to use slow path */ - BL set_register,%r25 - copy %r0,%r1 /* Write zero to target register */ - b nadtlb_nullify /* Nullify return insn */ - nop - + b,n nadtlb_fault #ifdef CONFIG_64BIT itlb_miss_20w: @@ -1413,14 +1331,14 @@ itlb_miss_20w: L3_ptep ptp,pte,t0,va,itlb_fault - tlb_lock spc,ptp,pte,t0,t1,itlb_fault + ptl_lock spc,ptp,pte,t0,t1,itlb_fault update_accessed ptp,pte,t0,t1 make_insert_tlb spc,pte,prot,t1 iitlbt pte,prot - tlb_unlock1 spc,t0,t1 + ptl_unlock spc,t0,t1 rfir nop @@ -1437,14 +1355,14 @@ naitlb_miss_20w: L3_ptep ptp,pte,t0,va,naitlb_check_alias_20w - tlb_lock spc,ptp,pte,t0,t1,naitlb_check_alias_20w + ptl_lock spc,ptp,pte,t0,t1,naitlb_check_alias_20w update_accessed ptp,pte,t0,t1 make_insert_tlb spc,pte,prot,t1 iitlbt pte,prot - tlb_unlock1 spc,t0,t1 + ptl_unlock spc,t0,t1 rfir nop @@ -1453,6 +1371,7 @@ naitlb_check_alias_20w: iitlbt pte,prot + insert_nops NUM_PIPELINE_INSNS - 1 rfir nop @@ -1465,7 +1384,7 @@ itlb_miss_11: L2_ptep ptp,pte,t0,va,itlb_fault - tlb_lock spc,ptp,pte,t0,t1,itlb_fault + ptl_lock spc,ptp,pte,t0,t1,itlb_fault update_accessed ptp,pte,t0,t1 make_insert_tlb_11 spc,pte,prot @@ -1478,7 +1397,7 @@ itlb_miss_11: mtsp t1, %sr1 /* Restore sr1 */ - tlb_unlock1 spc,t0,t1 + ptl_unlock spc,t0,t1 rfir nop @@ -1489,7 +1408,7 @@ naitlb_miss_11: L2_ptep ptp,pte,t0,va,naitlb_check_alias_11 - tlb_lock spc,ptp,pte,t0,t1,naitlb_check_alias_11 + ptl_lock spc,ptp,pte,t0,t1,naitlb_check_alias_11 update_accessed ptp,pte,t0,t1 make_insert_tlb_11 spc,pte,prot @@ -1502,7 +1421,7 @@ naitlb_miss_11: mtsp t1, %sr1 /* Restore sr1 */ - tlb_unlock1 spc,t0,t1 + ptl_unlock spc,t0,t1 rfir nop @@ -1512,6 +1431,7 @@ naitlb_check_alias_11: iitlba pte,(%sr0, va) iitlbp prot,(%sr0, va) + insert_nops NUM_PIPELINE_INSNS - 1 rfir nop @@ -1523,7 +1443,7 @@ itlb_miss_20: L2_ptep ptp,pte,t0,va,itlb_fault - tlb_lock spc,ptp,pte,t0,t1,itlb_fault + ptl_lock spc,ptp,pte,t0,t1,itlb_fault update_accessed ptp,pte,t0,t1 make_insert_tlb spc,pte,prot,t1 @@ -1532,7 +1452,7 @@ itlb_miss_20: iitlbt pte,prot - tlb_unlock1 spc,t0,t1 + ptl_unlock spc,t0,t1 rfir nop @@ -1543,7 +1463,7 @@ naitlb_miss_20: L2_ptep ptp,pte,t0,va,naitlb_check_alias_20 - tlb_lock spc,ptp,pte,t0,t1,naitlb_check_alias_20 + ptl_lock spc,ptp,pte,t0,t1,naitlb_check_alias_20 update_accessed ptp,pte,t0,t1 make_insert_tlb spc,pte,prot,t1 @@ -1552,7 +1472,7 @@ naitlb_miss_20: iitlbt pte,prot - tlb_unlock1 spc,t0,t1 + ptl_unlock spc,t0,t1 rfir nop @@ -1561,6 +1481,7 @@ naitlb_check_alias_20: iitlbt pte,prot + insert_nops NUM_PIPELINE_INSNS - 1 rfir nop @@ -1575,14 +1496,14 @@ dbit_trap_20w: L3_ptep ptp,pte,t0,va,dbit_fault - tlb_lock spc,ptp,pte,t0,t1,dbit_fault + ptl_lock spc,ptp,pte,t0,t1,dbit_fault update_dirty ptp,pte,t1 make_insert_tlb spc,pte,prot,t1 idtlbt pte,prot - tlb_unlock0 spc,t0,t1 + ptl_unlock spc,t0,t1 rfir nop #else @@ -1595,7 +1516,7 @@ dbit_trap_11: L2_ptep ptp,pte,t0,va,dbit_fault - tlb_lock spc,ptp,pte,t0,t1,dbit_fault + ptl_lock spc,ptp,pte,t0,t1,dbit_fault update_dirty ptp,pte,t1 make_insert_tlb_11 spc,pte,prot @@ -1608,7 +1529,7 @@ dbit_trap_11: mtsp t1, %sr1 /* Restore sr1 */ - tlb_unlock0 spc,t0,t1 + ptl_unlock spc,t0,t1 rfir nop @@ -1619,7 +1540,7 @@ dbit_trap_20: L2_ptep ptp,pte,t0,va,dbit_fault - tlb_lock spc,ptp,pte,t0,t1,dbit_fault + ptl_lock spc,ptp,pte,t0,t1,dbit_fault update_dirty ptp,pte,t1 make_insert_tlb spc,pte,prot,t1 @@ -1628,7 +1549,7 @@ dbit_trap_20: idtlbt pte,prot - tlb_unlock0 spc,t0,t1 + ptl_unlock spc,t0,t1 rfir nop #endif @@ -1721,7 +1642,7 @@ dtlb_fault: .macro fork_like name ENTRY_CFI(sys_\name\()_wrapper) - LDREG TI_TASK-THREAD_SZ_ALGN-FRAME_SIZE(%r30), %r1 + mfctl %cr30,%r1 ldo TASK_REGS(%r1),%r1 reg_save %r1 mfctl %cr27, %r28 @@ -1741,7 +1662,7 @@ ENTRY(child_return) BL schedule_tail, %r2 nop finish_child_return: - LDREG TI_TASK-THREAD_SZ_ALGN-FRAME_SIZE(%r30), %r1 + mfctl %cr30,%r1 ldo TASK_REGS(%r1),%r1 /* get pt regs */ LDREG PT_CR27(%r1), %r3 @@ -1752,7 +1673,7 @@ finish_child_return: END(child_return) ENTRY_CFI(sys_rt_sigreturn_wrapper) - LDREG TI_TASK-THREAD_SZ_ALGN-FRAME_SIZE(%r30),%r26 + mfctl %cr30,%r26 ldo TASK_REGS(%r26),%r26 /* get pt regs */ /* Don't save regs, we are going to restore them from sigcontext. */ STREG %r2, -RP_OFFSET(%r30) @@ -1769,7 +1690,7 @@ ENTRY_CFI(sys_rt_sigreturn_wrapper) LDREG -RP_OFFSET(%r30), %r2 /* FIXME: I think we need to restore a few more things here. */ - LDREG TI_TASK-THREAD_SZ_ALGN-FRAME_SIZE(%r30),%r1 + mfctl %cr30,%r1 ldo TASK_REGS(%r1),%r1 /* get pt regs */ reg_restore %r1 @@ -1788,9 +1709,7 @@ ENTRY(syscall_exit) */ /* save return value now */ - mfctl %cr30, %r1 - LDREG TI_TASK(%r1),%r1 STREG %r28,TASK_PT_GR28(%r1) /* Seems to me that dp could be wrong here, if the syscall involved @@ -1801,14 +1720,15 @@ ENTRY(syscall_exit) syscall_check_resched: /* check for reschedule */ - - LDREG TI_FLAGS-THREAD_SZ_ALGN-FRAME_SIZE(%r30),%r19 /* long */ + mfctl %cr30,%r19 + LDREG TASK_TI_FLAGS(%r19),%r19 /* long */ bb,<,n %r19, 31-TIF_NEED_RESCHED, syscall_do_resched /* forward */ .import do_signal,code syscall_check_sig: - LDREG TI_FLAGS-THREAD_SZ_ALGN-FRAME_SIZE(%r30),%r19 - ldi (_TIF_SIGPENDING|_TIF_NOTIFY_RESUME), %r26 + mfctl %cr30,%r19 + LDREG TASK_TI_FLAGS(%r19),%r19 + ldi (_TIF_USER_WORK_MASK & ~_TIF_NEED_RESCHED), %r26 and,COND(<>) %r19, %r26, %r0 b,n syscall_restore /* skip past if we've nothing to do */ @@ -1818,7 +1738,7 @@ syscall_do_signal: * consistent with all the relevant state of the process * before the syscall. We need to verify this. */ - LDREG TI_TASK-THREAD_SZ_ALGN-FRAME_SIZE(%r30),%r1 + mfctl %cr30,%r1 ldo TASK_REGS(%r1), %r26 /* struct pt_regs *regs */ reg_save %r26 @@ -1829,18 +1749,18 @@ syscall_do_signal: BL do_notify_resume,%r2 ldi 1, %r25 /* long in_syscall = 1 */ - LDREG TI_TASK-THREAD_SZ_ALGN-FRAME_SIZE(%r30),%r1 + mfctl %cr30,%r1 ldo TASK_REGS(%r1), %r20 /* reload pt_regs */ reg_restore %r20 b,n syscall_check_sig syscall_restore: - LDREG TI_TASK-THREAD_SZ_ALGN-FRAME_SIZE(%r30),%r1 + mfctl %cr30,%r1 /* Are we being ptraced? */ - ldw TASK_FLAGS(%r1),%r19 - ldi _TIF_SYSCALL_TRACE_MASK,%r2 + LDREG TASK_TI_FLAGS(%r1),%r19 + ldi _TIF_SINGLESTEP|_TIF_BLOCKSTEP,%r2 and,COND(=) %r19,%r2,%r0 b,n syscall_restore_rfi @@ -1878,7 +1798,7 @@ syscall_restore: mtsp %r1,%sr5 /* Restore sr5 */ mtsp %r1,%sr6 /* Restore sr6 */ - depi 3,31,2,%r31 /* ensure return to user mode. */ + depi PRIV_USER,31,2,%r31 /* ensure return to user mode. */ #ifdef CONFIG_64BIT /* decide whether to reset the wide mode bit @@ -1954,7 +1874,7 @@ syscall_restore_rfi: STREG %r0,TASK_PT_SR2(%r1) LDREG TASK_PT_GR31(%r1),%r2 - depi 3,31,2,%r2 /* ensure return to user mode. */ + depi PRIV_USER,31,2,%r2 /* ensure return to user mode. */ STREG %r2,TASK_PT_IAOQ0(%r1) ldo 4(%r2),%r2 STREG %r2,TASK_PT_IAOQ1(%r1) @@ -1963,10 +1883,10 @@ syscall_restore_rfi: pt_regs_ok: LDREG TASK_PT_IAOQ0(%r1),%r2 - depi 3,31,2,%r2 /* ensure return to user mode. */ + depi PRIV_USER,31,2,%r2 /* ensure return to user mode. */ STREG %r2,TASK_PT_IAOQ0(%r1) LDREG TASK_PT_IAOQ1(%r1),%r2 - depi 3,31,2,%r2 + depi PRIV_USER,31,2,%r2 STREG %r2,TASK_PT_IAOQ1(%r1) b intr_restore copy %r25,%r16 diff --git a/arch/parisc/kernel/firmware.c b/arch/parisc/kernel/firmware.c index 1d976f2ebff0..c69f6d5946e9 100644 --- a/arch/parisc/kernel/firmware.c +++ b/arch/parisc/kernel/firmware.c @@ -4,7 +4,8 @@ * * PDC == Processor Dependent Code * - * See http://www.parisc-linux.org/documentation/index.html + * See PDC documentation at + * https://parisc.wiki.kernel.org/index.php/Technical_Documentation * for documentation describing the entry points and calling * conventions defined below. * @@ -50,7 +51,7 @@ * prumpf 991016 */ -#include <stdarg.h> +#include <linux/stdarg.h> #include <linux/delay.h> #include <linux/init.h> @@ -73,16 +74,16 @@ static DEFINE_SPINLOCK(pdc_lock); #endif -extern unsigned long pdc_result[NUM_PDC_RESULT]; -extern unsigned long pdc_result2[NUM_PDC_RESULT]; +static unsigned long pdc_result[NUM_PDC_RESULT] __aligned(8); +static unsigned long pdc_result2[NUM_PDC_RESULT] __aligned(8); #ifdef CONFIG_64BIT -#define WIDE_FIRMWARE 0x1 -#define NARROW_FIRMWARE 0x2 +#define WIDE_FIRMWARE PDC_MODEL_OS64 +#define NARROW_FIRMWARE PDC_MODEL_OS32 -/* Firmware needs to be initially set to narrow to determine the +/* Firmware needs to be initially set to narrow to determine the * actual firmware width. */ -int parisc_narrow_firmware __ro_after_init = 1; +int parisc_narrow_firmware __ro_after_init = NARROW_FIRMWARE; #endif /* On most currently-supported platforms, IODC I/O calls are 32-bit calls @@ -122,10 +123,10 @@ static unsigned long f_extend(unsigned long address) #ifdef CONFIG_64BIT if(unlikely(parisc_narrow_firmware)) { if((address & 0xff000000) == 0xf0000000) - return 0xf0f0f0f000000000UL | (u32)address; + return (0xfffffff0UL << 32) | (u32)address; if((address & 0xf0000000) == 0xf0000000) - return 0xffffffff00000000UL | (u32)address; + return (0xffffffffUL << 32) | (u32)address; } #endif return address; @@ -133,7 +134,7 @@ static unsigned long f_extend(unsigned long address) /** * convert_to_wide - Convert the return buffer addresses into kernel addresses. - * @address: The return buffer from PDC. + * @addr: The return buffer from PDC. * * This function is used to convert the return buffer addresses retrieved from PDC * into kernel addresses when the PDC address size and kernel address size are @@ -159,20 +160,27 @@ void set_firmware_width_unlocked(void) ret = mem_pdc_call(PDC_MODEL, PDC_MODEL_CAPABILITIES, __pa(pdc_result), 0); + if (ret < 0) + return; convert_to_wide(pdc_result); if (pdc_result[0] != NARROW_FIRMWARE) parisc_narrow_firmware = 0; } - + /** * set_firmware_width - Determine if the firmware is wide or narrow. - * + * * This function must be called before any pdc_* function that uses the * convert_to_wide function. */ void set_firmware_width(void) { unsigned long flags; + + /* already initialized? */ + if (parisc_narrow_firmware != NARROW_FIRMWARE) + return; + spin_lock_irqsave(&pdc_lock, flags); set_firmware_width_unlocked(); spin_unlock_irqrestore(&pdc_lock, flags); @@ -249,8 +257,8 @@ int __init pdc_instr(unsigned int *instr) /** * pdc_chassis_info - Return chassis information. - * @result: The return buffer. * @chassis_info: The memory buffer address. + * @led_info: The size of the memory buffer address. * @len: The size of the memory buffer address. * * An HVERSION dependent call for returning the chassis information. @@ -274,7 +282,8 @@ int __init pdc_chassis_info(struct pdc_chassis_info *chassis_info, void *led_inf /** * pdc_pat_chassis_send_log - Sends a PDC PAT CHASSIS log message. - * @retval: -1 on error, 0 on success. Other value are PDC errors + * @state: state of the machine + * @data: value for that state * * Must be correctly formatted or expect system crash */ @@ -297,7 +306,7 @@ int pdc_pat_chassis_send_log(unsigned long state, unsigned long data) /** * pdc_chassis_disp - Updates chassis code - * @retval: -1 on error, 0 on success + * @disp: value to show on display */ int pdc_chassis_disp(unsigned long disp) { @@ -312,8 +321,7 @@ int pdc_chassis_disp(unsigned long disp) } /** - * pdc_cpu_rendenzvous - Stop currently executing CPU - * @retval: -1 on error, 0 on success + * __pdc_cpu_rendezvous - Stop currently executing CPU and do not return. */ int __pdc_cpu_rendezvous(void) { @@ -323,10 +331,47 @@ int __pdc_cpu_rendezvous(void) return mem_pdc_call(PDC_PROC, 1, 0); } +/** + * pdc_cpu_rendezvous_lock - Lock PDC while transitioning to rendezvous state + */ +void pdc_cpu_rendezvous_lock(void) __acquires(&pdc_lock) +{ + spin_lock(&pdc_lock); +} + +/** + * pdc_cpu_rendezvous_unlock - Unlock PDC after reaching rendezvous state + */ +void pdc_cpu_rendezvous_unlock(void) __releases(&pdc_lock) +{ + spin_unlock(&pdc_lock); +} + +/** + * pdc_pat_get_PDC_entrypoint - Get PDC entry point for current CPU + * @pdc_entry: pointer to where the PDC entry point should be stored + */ +int pdc_pat_get_PDC_entrypoint(unsigned long *pdc_entry) +{ + int retval = 0; + unsigned long flags; + + if (!IS_ENABLED(CONFIG_SMP) || !is_pdc_pat()) { + *pdc_entry = MEM_PDC; + return 0; + } + spin_lock_irqsave(&pdc_lock, flags); + retval = mem_pdc_call(PDC_PAT_CPU, PDC_PAT_CPU_GET_PDC_ENTRYPOINT, + __pa(pdc_result)); + *pdc_entry = pdc_result[0]; + spin_unlock_irqrestore(&pdc_lock, flags); + + return retval; +} /** * pdc_chassis_warn - Fetches chassis warnings - * @retval: -1 on error, 0 on success + * @warn: The warning value to be shown */ int pdc_chassis_warn(unsigned long *warn) { @@ -478,20 +523,21 @@ int pdc_model_info(struct pdc_model *model) /** * pdc_model_sysmodel - Get the system model name. + * @os_id: The operating system ID asked for (an OS_ID_* value) * @name: A char array of at least 81 characters. * * Get system model name from PDC ROM (e.g. 9000/715 or 9000/778/B160L). * Using OS_ID_HPUX will return the equivalent of the 'modelname' command * on HP/UX. */ -int pdc_model_sysmodel(char *name) +int pdc_model_sysmodel(unsigned int os_id, char *name) { int retval; unsigned long flags; spin_lock_irqsave(&pdc_lock, flags); retval = mem_pdc_call(PDC_MODEL, PDC_MODEL_SYSMODEL, __pa(pdc_result), - OS_ID_HPUX, __pa(name)); + os_id, __pa(name)); convert_to_wide(pdc_result); if (retval == PDC_OK) { @@ -506,7 +552,7 @@ int pdc_model_sysmodel(char *name) /** * pdc_model_versions - Identify the version number of each processor. - * @cpu_id: The return buffer. + * @versions: The return buffer. * @id: The id of the processor to check. * * Returns the version number for each processor component. @@ -641,7 +687,6 @@ int pdc_spaceid_bits(unsigned long *space_bits) return retval; } -#ifndef CONFIG_PA20 /** * pdc_btlb_info - Return block TLB information. * @btlb: The return buffer. @@ -650,18 +695,51 @@ int pdc_spaceid_bits(unsigned long *space_bits) */ int pdc_btlb_info(struct pdc_btlb_info *btlb) { - int retval; + int retval; unsigned long flags; - spin_lock_irqsave(&pdc_lock, flags); - retval = mem_pdc_call(PDC_BLOCK_TLB, PDC_BTLB_INFO, __pa(pdc_result), 0); - memcpy(btlb, pdc_result, sizeof(*btlb)); - spin_unlock_irqrestore(&pdc_lock, flags); + if (IS_ENABLED(CONFIG_PA20)) + return PDC_BAD_PROC; - if(retval < 0) { - btlb->max_size = 0; - } - return retval; + spin_lock_irqsave(&pdc_lock, flags); + retval = mem_pdc_call(PDC_BLOCK_TLB, PDC_BTLB_INFO, __pa(pdc_result), 0); + memcpy(btlb, pdc_result, sizeof(*btlb)); + spin_unlock_irqrestore(&pdc_lock, flags); + + if(retval < 0) { + btlb->max_size = 0; + } + return retval; +} + +int pdc_btlb_insert(unsigned long long vpage, unsigned long physpage, unsigned long len, + unsigned long entry_info, unsigned long slot) +{ + int retval; + unsigned long flags; + + if (IS_ENABLED(CONFIG_PA20)) + return PDC_BAD_PROC; + + spin_lock_irqsave(&pdc_lock, flags); + retval = mem_pdc_call(PDC_BLOCK_TLB, PDC_BTLB_INSERT, (unsigned long) (vpage >> 32), + (unsigned long) vpage, physpage, len, entry_info, slot); + spin_unlock_irqrestore(&pdc_lock, flags); + return retval; +} + +int pdc_btlb_purge_all(void) +{ + int retval; + unsigned long flags; + + if (IS_ENABLED(CONFIG_PA20)) + return PDC_BAD_PROC; + + spin_lock_irqsave(&pdc_lock, flags); + retval = mem_pdc_call(PDC_BLOCK_TLB, PDC_BTLB_PURGE_ALL); + spin_unlock_irqrestore(&pdc_lock, flags); + return retval; } /** @@ -682,6 +760,9 @@ int pdc_mem_map_hpa(struct pdc_memory_map *address, int retval; unsigned long flags; + if (IS_ENABLED(CONFIG_PA20)) + return PDC_BAD_PROC; + spin_lock_irqsave(&pdc_lock, flags); memcpy(pdc_result2, mod_path, sizeof(*mod_path)); retval = mem_pdc_call(PDC_MEM_MAP, PDC_MEM_MAP_HPA, __pa(pdc_result), @@ -691,7 +772,6 @@ int pdc_mem_map_hpa(struct pdc_memory_map *address, return retval; } -#endif /* !CONFIG_PA20 */ /** * pdc_lan_station_id - Get the LAN address. @@ -953,8 +1033,8 @@ int pdc_pci_irt(unsigned long num_entries, unsigned long hpa, void *tbl) /** * pdc_pci_config_read - read PCI config space. - * @hpa token from PDC to indicate which PCI device - * @pci_addr configuration space address to read from + * @hpa: Token from PDC to indicate which PCI device + * @cfg_addr: Configuration space address to read from * * Read PCI Configuration space *before* linux PCI subsystem is running. */ @@ -976,9 +1056,9 @@ unsigned int pdc_pci_config_read(void *hpa, unsigned long cfg_addr) /** * pdc_pci_config_write - read PCI config space. - * @hpa token from PDC to indicate which PCI device - * @pci_addr configuration space address to write - * @val value we want in the 32-bit register + * @hpa: Token from PDC to indicate which PCI device + * @cfg_addr: Configuration space address to write + * @val: Value we want in the 32-bit register * * Write PCI Configuration space *before* linux PCI subsystem is running. */ @@ -1061,6 +1141,38 @@ int pdc_mem_pdt_read_entries(struct pdc_mem_read_pdt *pret, } /** + * pdc_pim_toc11 - Fetch TOC PIM 1.1 data from firmware. + * @ret: pointer to return buffer + */ +int pdc_pim_toc11(struct pdc_toc_pim_11 *ret) +{ + int retval; + unsigned long flags; + + spin_lock_irqsave(&pdc_lock, flags); + retval = mem_pdc_call(PDC_PIM, PDC_PIM_TOC, __pa(pdc_result), + __pa(ret), sizeof(*ret)); + spin_unlock_irqrestore(&pdc_lock, flags); + return retval; +} + +/** + * pdc_pim_toc20 - Fetch TOC PIM 2.0 data from firmware. + * @ret: pointer to return buffer + */ +int pdc_pim_toc20(struct pdc_toc_pim_20 *ret) +{ + int retval; + unsigned long flags; + + spin_lock_irqsave(&pdc_lock, flags); + retval = mem_pdc_call(PDC_PIM, PDC_PIM_TOC, __pa(pdc_result), + __pa(ret), sizeof(*ret)); + spin_unlock_irqrestore(&pdc_lock, flags); + return retval; +} + +/** * pdc_tod_set - Set the Time-Of-Day clock. * @sec: The number of seconds since epoch. * @usec: The number of micro seconds. @@ -1157,15 +1269,18 @@ int __init pdc_soft_power_info(unsigned long *power_reg) } /* - * pdc_soft_power_button - Control the soft power button behaviour - * @sw_control: 0 for hardware control, 1 for software control + * pdc_soft_power_button{_panic} - Control the soft power button behaviour + * @sw_control: 0 for hardware control, 1 for software control * * * This PDC function places the soft power button under software or * hardware control. - * Under software control the OS may control to when to allow to shut - * down the system. Under hardware control pressing the power button + * Under software control the OS may control to when to allow to shut + * down the system. Under hardware control pressing the power button * powers off the system immediately. + * + * The _panic version relies on spin_trylock to prevent deadlock + * on panic path. */ int pdc_soft_power_button(int sw_control) { @@ -1179,6 +1294,22 @@ int pdc_soft_power_button(int sw_control) return retval; } +int pdc_soft_power_button_panic(int sw_control) +{ + int retval; + unsigned long flags; + + if (!spin_trylock_irqsave(&pdc_lock, flags)) { + pr_emerg("Couldn't enable soft power button\n"); + return -EBUSY; /* ignored by the panic notifier */ + } + + retval = mem_pdc_call(PDC_SOFT_POWER, PDC_SOFT_POWER_ENABLE, __pa(pdc_result), sw_control); + spin_unlock_irqrestore(&pdc_lock, flags); + + return retval; +} + /* * pdc_io_reset - Hack to avoid overlapping range registers of Bridges devices. * Primarily a problem on T600 (which parisc-linux doesn't support) but @@ -1213,9 +1344,8 @@ void pdc_io_reset_devices(void) #endif /* defined(BOOTLOADER) */ -/* locked by pdc_console_lock */ -static int __attribute__((aligned(8))) iodc_retbuf[32]; -static char __attribute__((aligned(64))) iodc_dbuf[4096]; +/* locked by pdc_lock */ +static char iodc_dbuf[4096] __page_aligned_bss; /** * pdc_iodc_print - Console print using IODC. @@ -1229,15 +1359,19 @@ static char __attribute__((aligned(64))) iodc_dbuf[4096]; */ int pdc_iodc_print(const unsigned char *str, unsigned count) { - unsigned int i; + unsigned int i, found = 0; unsigned long flags; + count = min_t(unsigned int, count, sizeof(iodc_dbuf)); + + spin_lock_irqsave(&pdc_lock, flags); for (i = 0; i < count;) { switch(str[i]) { case '\n': iodc_dbuf[i+0] = '\r'; iodc_dbuf[i+1] = '\n'; i += 2; + found = 1; goto print; default: iodc_dbuf[i] = str[i]; @@ -1247,14 +1381,13 @@ int pdc_iodc_print(const unsigned char *str, unsigned count) } print: - spin_lock_irqsave(&pdc_lock, flags); - real32_call(PAGE0->mem_cons.iodc_io, - (unsigned long)PAGE0->mem_cons.hpa, ENTRY_IO_COUT, - PAGE0->mem_cons.spa, __pa(PAGE0->mem_cons.dp.layers), - __pa(iodc_retbuf), 0, __pa(iodc_dbuf), i, 0); - spin_unlock_irqrestore(&pdc_lock, flags); + real32_call(PAGE0->mem_cons.iodc_io, + (unsigned long)PAGE0->mem_cons.hpa, ENTRY_IO_COUT, + PAGE0->mem_cons.spa, __pa(PAGE0->mem_cons.dp.layers), + __pa(pdc_result), 0, __pa(iodc_dbuf), i, 0); + spin_unlock_irqrestore(&pdc_lock, flags); - return i; + return i - found; } #if !defined(BOOTLOADER) @@ -1279,10 +1412,11 @@ int pdc_iodc_getc(void) real32_call(PAGE0->mem_kbd.iodc_io, (unsigned long)PAGE0->mem_kbd.hpa, ENTRY_IO_CIN, PAGE0->mem_kbd.spa, __pa(PAGE0->mem_kbd.dp.layers), - __pa(iodc_retbuf), 0, __pa(iodc_dbuf), 1, 0); + __pa(pdc_result), 0, __pa(iodc_dbuf), 1, 0); ch = *iodc_dbuf; - status = *iodc_retbuf; + /* like convert_to_wide() but for first return value only: */ + status = *(int *)&pdc_result; spin_unlock_irqrestore(&pdc_lock, flags); if (status == 0) @@ -1292,17 +1426,25 @@ int pdc_iodc_getc(void) } int pdc_sti_call(unsigned long func, unsigned long flags, - unsigned long inptr, unsigned long outputr, - unsigned long glob_cfg) + unsigned long inptr, unsigned long outputr, + unsigned long glob_cfg, int do_call64) { - int retval; + int retval = 0; unsigned long irqflags; - spin_lock_irqsave(&pdc_lock, irqflags); - retval = real32_call(func, flags, inptr, outputr, glob_cfg); - spin_unlock_irqrestore(&pdc_lock, irqflags); + spin_lock_irqsave(&pdc_lock, irqflags); + if (IS_ENABLED(CONFIG_64BIT) && do_call64) { +#ifdef CONFIG_64BIT + retval = real64_call(func, flags, inptr, outputr, glob_cfg); +#else + WARN_ON(1); +#endif + } else { + retval = real32_call(func, flags, inptr, outputr, glob_cfg); + } + spin_unlock_irqrestore(&pdc_lock, irqflags); - return retval; + return retval; } EXPORT_SYMBOL(pdc_sti_call); @@ -1452,7 +1594,7 @@ int pdc_pat_get_irt(void *r_addr, unsigned long cell_num) /** * pdc_pat_pd_get_addr_map - Retrieve information about memory address ranges. - * @actlen: The return buffer. + * @actual_len: The return buffer. * @mem_addr: Pointer to the memory buffer. * @count: The number of bytes to read from the buffer. * @offset: The offset with respect to the beginning of the buffer. @@ -1475,7 +1617,7 @@ int pdc_pat_pd_get_addr_map(unsigned long *actual_len, void *mem_addr, } /** - * pdc_pat_pd_get_PDC_interface_revisions - Retrieve PDC interface revisions. + * pdc_pat_pd_get_pdc_revisions - Retrieve PDC interface revisions. * @legacy_rev: The legacy revision. * @pat_rev: The PAT revision. * @pdc_cap: The PDC capabilities. @@ -1530,7 +1672,7 @@ int pdc_pat_io_pci_cfg_read(unsigned long pci_addr, int pci_size, u32 *mem_addr) * pdc_pat_io_pci_cfg_write - Retrieve information about memory address ranges. * @pci_addr: PCI configuration space address for which the write request is being made. * @pci_size: Size of write in bytes. Valid values are 1, 2, and 4. - * @value: Pointer to 1, 2, or 4 byte value in low order end of argument to be + * @val: Pointer to 1, 2, or 4 byte value in low order end of argument to be * written to PCI Config space. * */ @@ -1548,7 +1690,7 @@ int pdc_pat_io_pci_cfg_write(unsigned long pci_addr, int pci_size, u32 val) } /** - * pdc_pat_mem_pdc_info - Retrieve information about page deallocation table + * pdc_pat_mem_pdt_info - Retrieve information about page deallocation table * @rinfo: memory pdt information * */ diff --git a/arch/parisc/kernel/ftrace.c b/arch/parisc/kernel/ftrace.c index b836fc61a24f..621a4b386ae4 100644 --- a/arch/parisc/kernel/ftrace.c +++ b/arch/parisc/kernel/ftrace.c @@ -15,15 +15,18 @@ #include <linux/uaccess.h> #include <linux/kprobes.h> #include <linux/ptrace.h> +#include <linux/jump_label.h> #include <asm/assembly.h> #include <asm/sections.h> #include <asm/ftrace.h> #include <asm/patch.h> -#define __hot __attribute__ ((__section__ (".text.hot"))) +#define __hot __section(".text.hot") #ifdef CONFIG_FUNCTION_GRAPH_TRACER +static DEFINE_STATIC_KEY_FALSE(ftrace_graph_enable); + /* * Hook the return address and push it in the stack of return addrs * in current thread info. @@ -48,24 +51,19 @@ static void __hot prepare_ftrace_return(unsigned long *parent, } #endif /* CONFIG_FUNCTION_GRAPH_TRACER */ -void notrace __hot ftrace_function_trampoline(unsigned long parent, +static ftrace_func_t ftrace_func; + +asmlinkage void notrace __hot ftrace_function_trampoline(unsigned long parent, unsigned long self_addr, unsigned long org_sp_gr3, - struct pt_regs *regs) + struct ftrace_regs *fregs) { -#ifndef CONFIG_DYNAMIC_FTRACE - extern ftrace_func_t ftrace_trace_function; -#endif extern struct ftrace_ops *function_trace_op; - if (function_trace_op->flags & FTRACE_OPS_FL_ENABLED && - ftrace_trace_function != ftrace_stub) - ftrace_trace_function(self_addr, parent, - function_trace_op, regs); + ftrace_func(self_addr, parent, function_trace_op, fregs); #ifdef CONFIG_FUNCTION_GRAPH_TRACER - if (ftrace_graph_return != (trace_func_graph_ret_t) ftrace_stub || - ftrace_graph_entry != ftrace_graph_entry_stub) { + if (static_branch_unlikely(&ftrace_graph_enable)) { unsigned long *parent_rp; /* calculate pointer to %rp in stack */ @@ -80,26 +78,24 @@ void notrace __hot ftrace_function_trampoline(unsigned long parent, #endif } -#ifdef CONFIG_FUNCTION_GRAPH_TRACER +#if defined(CONFIG_DYNAMIC_FTRACE) && defined(CONFIG_FUNCTION_GRAPH_TRACER) int ftrace_enable_ftrace_graph_caller(void) { + static_key_enable(&ftrace_graph_enable.key); return 0; } int ftrace_disable_ftrace_graph_caller(void) { + static_key_enable(&ftrace_graph_enable.key); return 0; } #endif #ifdef CONFIG_DYNAMIC_FTRACE - -int __init ftrace_dyn_arch_init(void) -{ - return 0; -} int ftrace_update_ftrace_func(ftrace_func_t func) { + ftrace_func = func; return 0; } @@ -172,7 +168,7 @@ int ftrace_make_call(struct dyn_ftrace *rec, unsigned long addr) ip = (void *)(rec->ip + 4 - size); - ret = probe_kernel_read(insn, ip, size); + ret = copy_from_kernel_nofault(insn, ip, size); if (ret) return ret; @@ -203,17 +199,25 @@ int ftrace_make_nop(struct module *mod, struct dyn_ftrace *rec, #ifdef CONFIG_KPROBES_ON_FTRACE void kprobe_ftrace_handler(unsigned long ip, unsigned long parent_ip, - struct ftrace_ops *ops, struct pt_regs *regs) + struct ftrace_ops *ops, struct ftrace_regs *fregs) { struct kprobe_ctlblk *kcb; - struct kprobe *p = get_kprobe((kprobe_opcode_t *)ip); + struct pt_regs *regs; + struct kprobe *p; + int bit; - if (unlikely(!p) || kprobe_disabled(p)) + bit = ftrace_test_recursion_trylock(ip, parent_ip); + if (bit < 0) return; + regs = ftrace_get_regs(fregs); + p = get_kprobe((kprobe_opcode_t *)ip); + if (unlikely(!p) || kprobe_disabled(p)) + goto out; + if (kprobe_running()) { kprobes_inc_nmissed_count(p); - return; + goto out; } __this_cpu_write(current_kprobe, p); @@ -234,6 +238,8 @@ void kprobe_ftrace_handler(unsigned long ip, unsigned long parent_ip, } } __this_cpu_write(current_kprobe, NULL); +out: + ftrace_test_recursion_unlock(bit); } NOKPROBE_SYMBOL(kprobe_ftrace_handler); diff --git a/arch/parisc/kernel/hardware.c b/arch/parisc/kernel/hardware.c index 98c5203c1ab0..357d9cdab7ce 100644 --- a/arch/parisc/kernel/hardware.c +++ b/arch/parisc/kernel/hardware.c @@ -6,7 +6,8 @@ * * Based on the document "PA-RISC 1.1 I/O Firmware Architecture * Reference Specification", March 7, 1999, version 0.96. This - * is available at http://parisc-linux.org/documentation/ + * is available at + * https://parisc.wiki.kernel.org/index.php/Technical_Documentation * * Copyright 1999 by Alex deVries <alex@onefishtwo.ca> * and copyright 1999 The Puffin Group Inc. @@ -22,9 +23,6 @@ * HP PARISC Hardware Database * Access to this database is only possible during bootup * so don't reference this table after starting the init process - * - * NOTE: Product names which are listed here and ends with a '?' - * are guessed. If you know the correct name, please let us know. */ static struct hp_hardware hp_hardware_list[] __initdata = { @@ -211,7 +209,7 @@ static struct hp_hardware hp_hardware_list[] __initdata = { {HPHW_NPROC,0x5DD,0x4,0x81,"Duet W2"}, {HPHW_NPROC,0x5DE,0x4,0x81,"Piccolo W+"}, {HPHW_NPROC,0x5DF,0x4,0x81,"Cantata W2"}, - {HPHW_NPROC,0x5DF,0x0,0x00,"Marcato W+ (rp5470)?"}, + {HPHW_NPROC,0x5DF,0x0,0x00,"Marcato W+ (rp5470)"}, {HPHW_NPROC,0x5E0,0x4,0x91,"Cantata DC- W2"}, {HPHW_NPROC,0x5E1,0x4,0x91,"Crescendo DC- W2"}, {HPHW_NPROC,0x5E2,0x4,0x91,"Crescendo 650 W2"}, @@ -265,11 +263,11 @@ static struct hp_hardware hp_hardware_list[] __initdata = { {HPHW_NPROC,0x888,0x4,0x91,"Storm Peak Fast DC-"}, {HPHW_NPROC,0x889,0x4,0x91,"Storm Peak Fast"}, {HPHW_NPROC,0x88A,0x4,0x91,"Crestone Peak Slow"}, - {HPHW_NPROC,0x88B,0x4,0x91,"Crestone Peak Fast?"}, + {HPHW_NPROC,0x88B,0x4,0x91,"Crestone Peak Fast"}, {HPHW_NPROC,0x88C,0x4,0x91,"Orca Mako+"}, {HPHW_NPROC,0x88D,0x4,0x91,"Rainier/Medel Mako+ Slow"}, {HPHW_NPROC,0x88E,0x4,0x91,"Rainier/Medel Mako+ Fast"}, - {HPHW_NPROC,0x892,0x4,0x91,"Mt. Hamilton Slow Mako+?"}, + {HPHW_NPROC,0x892,0x4,0x91,"Mt. Hamilton Slow Mako+"}, {HPHW_NPROC,0x894,0x4,0x91,"Mt. Hamilton Fast Mako+"}, {HPHW_NPROC,0x895,0x4,0x91,"Storm Peak Slow Mako+"}, {HPHW_NPROC,0x896,0x4,0x91,"Storm Peak Fast Mako+"}, @@ -1197,7 +1195,7 @@ static struct hp_hardware hp_hardware_list[] __initdata = { {HPHW_FIO, 0x004, 0x00340, 0x0, "BARCO CX4500 VME Grphx Cnsl"}, {HPHW_FIO, 0x004, 0x00360, 0x0, "Hughes TOG VME FDDI"}, {HPHW_FIO, 0x076, 0x000AD, 0x0, "Crestone Peak Core RS-232"}, - {HPHW_FIO, 0x077, 0x000AD, 0x0, "Crestone Peak Fast? Core RS-232"}, + {HPHW_FIO, 0x077, 0x000AD, 0x0, "Crestone Peak Fast Core RS-232"}, {HPHW_IOA, 0x185, 0x0000B, 0x00, "Java BC Summit Port"}, {HPHW_IOA, 0x1FF, 0x0000B, 0x00, "Hitachi Ghostview Summit Port"}, {HPHW_IOA, 0x580, 0x0000B, 0x10, "U2-IOA BC Runway Port"}, diff --git a/arch/parisc/kernel/head.S b/arch/parisc/kernel/head.S index 951a339369dd..96e0264ac961 100644 --- a/arch/parisc/kernel/head.S +++ b/arch/parisc/kernel/head.S @@ -17,12 +17,12 @@ #include <asm/pdc.h> #include <asm/assembly.h> -#include <asm/pgtable.h> #include <linux/linkage.h> #include <linux/init.h> +#include <linux/pgtable.h> - .level PA_ASM_LEVEL + .level 1.1 __INITDATA ENTRY(boot_args) @@ -35,7 +35,8 @@ END(boot_args) __HEAD .align 4 - .import init_thread_union,data + .import init_task,data + .import init_stack,data .import fault_vector_20,code /* IVA parisc 2.0 32 bit */ #ifndef CONFIG_64BIT .import fault_vector_11,code /* IVA parisc 1.1 32 bit */ @@ -69,6 +70,46 @@ $bss_loop: stw,ma %arg2,4(%r1) stw,ma %arg3,4(%r1) +#if defined(CONFIG_PA20) + /* check for 64-bit capable CPU as required by current kernel */ + ldi 32,%r10 + mtctl %r10,%cr11 + .level 2.0 + mfctl,w %cr11,%r10 + .level 1.1 + comib,<>,n 0,%r10,$cpu_ok + + load32 PA(msg1),%arg0 + ldi msg1_end-msg1,%arg1 +$iodc_panic: + copy %arg0, %r10 + copy %arg1, %r11 + load32 PA(init_stack),%sp +#define MEM_CONS 0x3A0 + ldw MEM_CONS+32(%r0),%arg0 // HPA + ldi ENTRY_IO_COUT,%arg1 + ldw MEM_CONS+36(%r0),%arg2 // SPA + ldw MEM_CONS+8(%r0),%arg3 // layers + load32 PA(__bss_start),%r1 + stw %r1,-52(%sp) // arg4 + stw %r0,-56(%sp) // arg5 + stw %r10,-60(%sp) // arg6 = ptr to text + stw %r11,-64(%sp) // arg7 = len + stw %r0,-68(%sp) // arg8 + load32 PA(.iodc_panic_ret), %rp + ldw MEM_CONS+40(%r0),%r1 // ENTRY_IODC + bv,n (%r1) +.iodc_panic_ret: + b . /* wait endless with ... */ + or %r10,%r10,%r10 /* qemu idle sleep */ +msg1: .ascii "Can't boot kernel which was built for PA8x00 CPUs on this machine.\r\n" +msg1_end: + +$cpu_ok: +#endif + + .level PA_ASM_LEVEL + /* Initialize startup VM. Just map first 16/32 MB of memory */ load32 PA(swapper_pg_dir),%r4 mtctl %r4,%cr24 /* Initialize kernel root pointer */ @@ -123,12 +164,12 @@ $pgt_fill_loop: load32 start_parisc,%r11 /* And the initial task pointer */ - load32 init_thread_union,%r6 + load32 init_task,%r6 mtctl %r6,%cr30 /* And the stack pointer too */ - ldo THREAD_SZ_ALGN(%r6),%sp - + load32 init_stack,%sp + tophys_r1 %sp #if defined(CONFIG_64BIT) && defined(CONFIG_FUNCTION_TRACER) .import _mcount,data /* initialize mcount FPTR */ @@ -138,10 +179,10 @@ $pgt_fill_loop: std %dp,0x18(%r10) #endif -#ifdef CONFIG_64BIT - /* Get PDCE_PROC for monarch CPU. */ #define MEM_PDC_LO 0x388 #define MEM_PDC_HI 0x35C +#ifdef CONFIG_64BIT + /* Get PDCE_PROC for monarch CPU. */ ldw MEM_PDC_LO(%r0),%r3 ldw MEM_PDC_HI(%r0),%r10 depd %r10, 31, 32, %r3 /* move to upper word */ @@ -161,6 +202,15 @@ $pgt_fill_loop: /* FALLTHROUGH */ .procend +#ifdef CONFIG_HOTPLUG_CPU + /* common_stext is far away in another section... jump there */ + load32 PA(common_stext), %rp + bv,n (%rp) + + /* common_stext and smp_slave_stext needs to be in text section */ + .text +#endif + /* ** Code Common to both Monarch and Slave processors. ** Entry: @@ -186,12 +236,11 @@ common_stext: #endif /*CONFIG_SMP*/ #ifdef CONFIG_64BIT - tophys_r1 %sp + mfctl %cr30,%r6 /* PCX-W2 firmware bug */ + tophys_r1 %r6 /* Save the rfi target address */ - ldd TI_TASK-THREAD_SZ_ALGN(%sp), %r10 - tophys_r1 %r10 - std %r11, TASK_PT_GR11(%r10) + STREG %r11, TASK_PT_GR11(%r6) /* Switch to wide mode Superdome doesn't support narrow PDC ** calls. */ @@ -206,7 +255,6 @@ common_stext: ** Someday, palo might not do this for the Monarch either. */ 2: - mfctl %cr30,%r6 /* PCX-W2 firmware bug */ ldo PDC_PSW(%r0),%arg0 /* 21 */ ldo PDC_PSW_SET_DEFAULTS(%r0),%arg1 /* 2 */ @@ -216,15 +264,21 @@ common_stext: copy %r0,%arg3 stext_pdc_ret: + LDREG TASK_PT_GR11(%r6), %r11 + tovirt_r1 %r6 mtctl %r6,%cr30 /* restore task thread info */ +#endif - /* restore rfi target address*/ - ldd TI_TASK-THREAD_SZ_ALGN(%sp), %r10 - tophys_r1 %r10 - ldd TASK_PT_GR11(%r10), %r11 - tovirt_r1 %sp +#ifndef CONFIG_64BIT + /* clear all BTLBs */ + ldi PDC_BLOCK_TLB,%arg0 + load32 PA(stext_pdc_btlb_ret), %rp + ldw MEM_PDC_LO(%r0),%r3 + bv (%r3) + ldi PDC_BTLB_PURGE_ALL,%arg1 +stext_pdc_btlb_ret: #endif - + /* PARANOID: clear user scratch/user space SR's */ mtsp %r0,%sr0 mtsp %r0,%sr1 @@ -287,7 +341,9 @@ aligned_rfi: load32 KERNEL_PSW,%r10 mtctl %r10,%ipsw - + + tovirt_r1 %sp + /* Jump through hyperspace to Virt Mode */ rfi nop @@ -343,12 +399,13 @@ smp_slave_stext: #endif /* Initialize the SP - monarch sets up smp_init_current_idle_task */ - load32 PA(smp_init_current_idle_task),%sp - LDREG 0(%sp),%sp /* load task address */ + load32 PA(smp_init_current_idle_task),%r6 + LDREG 0(%r6),%r6 + mtctl %r6,%cr30 + tophys_r1 %r6 + LDREG TASK_STACK(%r6),%sp tophys_r1 %sp - LDREG TASK_THREAD_INFO(%sp),%sp - mtctl %sp,%cr30 /* store in cr30 */ - ldo THREAD_SZ_ALGN(%sp),%sp + ldo FRAME_SIZE(%sp),%sp /* point CPU to kernel page tables */ load32 PA(swapper_pg_dir),%r4 @@ -373,8 +430,6 @@ smp_slave_stext: .procend #endif /* CONFIG_SMP */ -ENDPROC(parisc_kernel_start) - #ifndef CONFIG_64BIT .section .data..ro_after_init diff --git a/arch/parisc/kernel/hpmc.S b/arch/parisc/kernel/hpmc.S index 81de5e2b391c..eb2e4bd67035 100644 --- a/arch/parisc/kernel/hpmc.S +++ b/arch/parisc/kernel/hpmc.S @@ -43,10 +43,8 @@ * IODC requires 7K byte stack. That leaves 1K byte for os_hpmc. */ - __PAGE_ALIGNED_BSS - .align 4096 -hpmc_stack: - .block 16384 + .import toc_stack,data +#define hpmc_stack toc_stack /* re-use the TOC stack */ #define HPMC_IODC_BUF_SIZE 0x8000 @@ -289,13 +287,3 @@ os_hpmc_6: b . nop .align 16 /* make function length multiple of 16 bytes */ -.os_hpmc_end: - - - __INITRODATA -.globl os_hpmc_size - .align 4 - .type os_hpmc_size, @object - .size os_hpmc_size, 4 -os_hpmc_size: - .word .os_hpmc_end-.os_hpmc diff --git a/arch/parisc/kernel/inventory.c b/arch/parisc/kernel/inventory.c index 9298f2285510..7ab2f2a54400 100644 --- a/arch/parisc/kernel/inventory.c +++ b/arch/parisc/kernel/inventory.c @@ -19,6 +19,7 @@ #include <linux/init.h> #include <linux/slab.h> #include <linux/mm.h> +#include <linux/platform_device.h> #include <asm/hardware.h> #include <asm/io.h> #include <asm/mmzone.h> @@ -641,4 +642,33 @@ void __init do_device_inventory(void) if (pa_serialize_tlb_flushes) pr_info("Merced bus found: Enable PxTLB serialization.\n"); #endif + +#if defined(CONFIG_FW_CFG_SYSFS) + if (running_on_qemu) { + struct resource res[3] = {0,}; + unsigned int base; + + base = ((unsigned long long) PAGE0->pad0[2] << 32) + | PAGE0->pad0[3]; /* SeaBIOS stored it here */ + + res[0].name = "fw_cfg"; + res[0].start = base; + res[0].end = base + 8 - 1; + res[0].flags = IORESOURCE_MEM; + + res[1].name = "ctrl"; + res[1].start = 0; + res[1].flags = IORESOURCE_REG; + + res[2].name = "data"; + res[2].start = 4; + res[2].flags = IORESOURCE_REG; + + if (base) { + pr_info("Found qemu fw_cfg interface at %#08x\n", base); + platform_device_register_simple("fw_cfg", + PLATFORM_DEVID_NONE, res, 3); + } + } +#endif } diff --git a/arch/parisc/kernel/irq.c b/arch/parisc/kernel/irq.c index e5fcfb70cc7c..dff66be65d29 100644 --- a/arch/parisc/kernel/irq.c +++ b/arch/parisc/kernel/irq.c @@ -15,16 +15,15 @@ #include <linux/kernel_stat.h> #include <linux/seq_file.h> #include <linux/types.h> +#include <linux/sched/task_stack.h> #include <asm/io.h> +#include <asm/softirq_stack.h> #include <asm/smp.h> #include <asm/ldcw.h> #undef PARISC_IRQ_CR16_COUNTS -extern irqreturn_t timer_interrupt(int, void *); -extern irqreturn_t ipi_interrupt(int, void *); - #define EIEM_MASK(irq) (1UL<<(CPU_IRQ_MAX - irq)) /* Bits in EIEM correlate with cpu_irq_action[]. @@ -103,28 +102,12 @@ int cpu_check_affinity(struct irq_data *d, const struct cpumask *dest) if (irqd_is_per_cpu(d)) return -EINVAL; - /* whatever mask they set, we just allow one CPU */ - cpu_dest = cpumask_next_and(d->irq & (num_online_cpus()-1), - dest, cpu_online_mask); + cpu_dest = cpumask_first_and(dest, cpu_online_mask); if (cpu_dest >= nr_cpu_ids) - cpu_dest = cpumask_first_and(dest, cpu_online_mask); + cpu_dest = cpumask_first(cpu_online_mask); return cpu_dest; } - -static int cpu_set_affinity_irq(struct irq_data *d, const struct cpumask *dest, - bool force) -{ - int cpu_dest; - - cpu_dest = cpu_check_affinity(d, dest); - if (cpu_dest < 0) - return -1; - - cpumask_copy(irq_data_get_affinity_mask(d), dest); - - return 0; -} #endif static struct irq_chip cpu_interrupt_type = { @@ -133,9 +116,6 @@ static struct irq_chip cpu_interrupt_type = { .irq_unmask = cpu_unmask_irq, .irq_ack = cpu_ack_irq, .irq_eoi = cpu_eoi_irq, -#ifdef CONFIG_SMP - .irq_set_affinity = cpu_set_affinity_irq, -#endif /* XXX: Needs to be written. We managed without it so far, but * we really ought to write it. */ @@ -216,12 +196,9 @@ int show_interrupts(struct seq_file *p, void *v) if (!action) goto skip; seq_printf(p, "%3d: ", i); -#ifdef CONFIG_SMP + for_each_online_cpu(j) - seq_printf(p, "%10u ", kstat_irqs_cpu(i, j)); -#else - seq_printf(p, "%10u ", kstat_irqs(i)); -#endif + seq_printf(p, "%10u ", irq_desc_kstat_cpu(desc, j)); seq_printf(p, " %14s", irq_desc_get_chip(desc)->name); #ifndef PARISC_IRQ_CR16_COUNTS @@ -335,7 +312,7 @@ unsigned long txn_affinity_addr(unsigned int irq, int cpu) { #ifdef CONFIG_SMP struct irq_data *d = irq_get_irq_data(irq); - cpumask_copy(irq_data_get_affinity_mask(d), cpumask_of(cpu)); + irq_data_update_affinity(d, cpumask_of(cpu)); #endif return per_cpu(cpu_data, cpu).txn_addr; @@ -376,7 +353,11 @@ static inline int eirr_to_irq(unsigned long eirr) /* * IRQ STACK - used for irq handler */ +#ifdef CONFIG_64BIT +#define IRQ_STACK_SIZE (4096 << 4) /* 64k irq stack size */ +#else #define IRQ_STACK_SIZE (4096 << 3) /* 32k irq stack size */ +#endif union irq_stack_union { unsigned long stack[IRQ_STACK_SIZE/sizeof(unsigned long)]; @@ -384,7 +365,7 @@ union irq_stack_union { volatile unsigned int lock[1]; }; -DEFINE_PER_CPU(union irq_stack_union, irq_stack_union) = { +static DEFINE_PER_CPU(union irq_stack_union, irq_stack_union) = { .slock = { 1,1,1,1 }, }; #endif @@ -397,8 +378,7 @@ static inline void stack_overflow_check(struct pt_regs *regs) #ifdef CONFIG_DEBUG_STACKOVERFLOW #define STACK_MARGIN (256*6) - /* Our stack starts directly behind the thread_info struct. */ - unsigned long stack_start = (unsigned long) current_thread_info(); + unsigned long stack_start = (unsigned long) task_stack_page(current); unsigned long sp = regs->gr[30]; unsigned long stack_usage; unsigned int *last_usage; @@ -474,7 +454,7 @@ static void execute_on_irq_stack(void *func, unsigned long param1) union_ptr = &per_cpu(irq_stack_union, smp_processor_id()); irq_stack = (unsigned long) &union_ptr->stack; irq_stack = ALIGN(irq_stack + sizeof(irq_stack_union.slock), - 64); /* align for stack frame usage */ + FRAME_ALIGN); /* align for stack frame usage */ /* We may be called recursive. If we are already using the irq stack, * just continue to use it. Use spinlocks to serialize @@ -497,14 +477,16 @@ static void execute_on_irq_stack(void *func, unsigned long param1) *irq_stack_in_use = 1; } +#ifdef CONFIG_SOFTIRQ_ON_OWN_STACK void do_softirq_own_stack(void) { execute_on_irq_stack(__do_softirq, 0); } +#endif #endif /* CONFIG_IRQSTACKS */ /* ONLY called from entry.S:intr_extint() */ -void do_cpu_irq_mask(struct pt_regs *regs) +asmlinkage void do_cpu_irq_mask(struct pt_regs *regs) { struct pt_regs *old_regs; unsigned long eirr_val; @@ -516,7 +498,7 @@ void do_cpu_irq_mask(struct pt_regs *regs) old_regs = set_irq_regs(regs); local_irq_disable(); - irq_enter(); + irq_enter_rcu(); eirr_val = mfctl(23) & cpu_eiem & per_cpu(local_ack_eiem, cpu); if (!eirr_val) @@ -551,7 +533,7 @@ void do_cpu_irq_mask(struct pt_regs *regs) #endif /* CONFIG_IRQSTACKS */ out: - irq_exit(); + irq_exit_rcu(); set_irq_regs(old_regs); return; @@ -560,37 +542,27 @@ void do_cpu_irq_mask(struct pt_regs *regs) goto out; } -static struct irqaction timer_action = { - .handler = timer_interrupt, - .name = "timer", - .flags = IRQF_TIMER | IRQF_PERCPU | IRQF_IRQPOLL, -}; - -#ifdef CONFIG_SMP -static struct irqaction ipi_action = { - .handler = ipi_interrupt, - .name = "IPI", - .flags = IRQF_PERCPU, -}; -#endif - static void claim_cpu_irqs(void) { + unsigned long flags = IRQF_TIMER | IRQF_PERCPU | IRQF_IRQPOLL; int i; + for (i = CPU_IRQ_BASE; i <= CPU_IRQ_MAX; i++) { irq_set_chip_and_handler(i, &cpu_interrupt_type, handle_percpu_irq); } irq_set_handler(TIMER_IRQ, handle_percpu_irq); - setup_irq(TIMER_IRQ, &timer_action); + if (request_irq(TIMER_IRQ, timer_interrupt, flags, "timer", NULL)) + pr_err("Failed to register timer interrupt\n"); #ifdef CONFIG_SMP irq_set_handler(IPI_IRQ, handle_percpu_irq); - setup_irq(IPI_IRQ, &ipi_action); + if (request_irq(IPI_IRQ, ipi_interrupt, IRQF_PERCPU, "IPI", NULL)) + pr_err("Failed to register IPI interrupt\n"); #endif } -void __init init_IRQ(void) +void init_IRQ(void) { local_irq_disable(); /* PARANOID - should already be disabled */ mtctl(~0UL, 23); /* EIRR : clear all pending external intr */ diff --git a/arch/parisc/kernel/jump_label.c b/arch/parisc/kernel/jump_label.c index d2f3cb12e282..e253b134500d 100644 --- a/arch/parisc/kernel/jump_label.c +++ b/arch/parisc/kernel/jump_label.c @@ -42,14 +42,3 @@ void arch_jump_label_transform(struct jump_entry *entry, patch_text(addr, insn); } - -void arch_jump_label_transform_static(struct jump_entry *entry, - enum jump_label_type type) -{ - /* - * We use the architected NOP in arch_static_branch, so there's no - * need to patch an identical NOP over the top of it here. The core - * will call arch_jump_label_transform from a module notifier if the - * NOP needs to be replaced by a branch. - */ -} diff --git a/arch/parisc/kernel/kexec.c b/arch/parisc/kernel/kexec.c index 5eb7f30edc1f..db57345a9daf 100644 --- a/arch/parisc/kernel/kexec.c +++ b/arch/parisc/kernel/kexec.c @@ -4,6 +4,8 @@ #include <linux/console.h> #include <linux/kexec.h> #include <linux/delay.h> +#include <linux/reboot.h> + #include <asm/cacheflush.h> #include <asm/sections.h> diff --git a/arch/parisc/kernel/kexec_file.c b/arch/parisc/kernel/kexec_file.c index 8c534204f0fd..3fc82130b6c3 100644 --- a/arch/parisc/kernel/kexec_file.c +++ b/arch/parisc/kernel/kexec_file.c @@ -38,8 +38,8 @@ static void *elf_load(struct kimage *image, char *kernel_buf, for (i = 0; i < image->nr_segments; i++) image->segment[i].mem = __pa(image->segment[i].mem); - pr_debug("Loaded the kernel at 0x%lx, entry at 0x%lx\n", - kernel_load_addr, image->start); + kexec_dprintk("Loaded the kernel at 0x%lx, entry at 0x%lx\n", + kernel_load_addr, image->start); if (initrd != NULL) { kbuf.buffer = initrd; @@ -51,7 +51,7 @@ static void *elf_load(struct kimage *image, char *kernel_buf, if (ret) goto out; - pr_debug("Loaded initrd at 0x%lx\n", kbuf.mem); + kexec_dprintk("Loaded initrd at 0x%lx\n", kbuf.mem); image->arch.initrd_start = kbuf.mem; image->arch.initrd_end = kbuf.mem + initrd_len; } @@ -68,7 +68,7 @@ static void *elf_load(struct kimage *image, char *kernel_buf, if (ret) goto out; - pr_debug("Loaded cmdline at 0x%lx\n", kbuf.mem); + kexec_dprintk("Loaded cmdline at 0x%lx\n", kbuf.mem); image->arch.cmdline = kbuf.mem; } out: diff --git a/arch/parisc/kernel/kgdb.c b/arch/parisc/kernel/kgdb.c index 664278db9b97..b16fa9bac5f4 100644 --- a/arch/parisc/kernel/kgdb.c +++ b/arch/parisc/kernel/kgdb.c @@ -3,6 +3,7 @@ * PA-RISC KGDB support * * Copyright (c) 2019 Sven Schnelle <svens@stackframe.org> + * Copyright (c) 2022 Helge Deller <deller@gmx.de> * */ @@ -154,8 +155,8 @@ void kgdb_arch_set_pc(struct pt_regs *regs, unsigned long ip) int kgdb_arch_set_breakpoint(struct kgdb_bkpt *bpt) { - int ret = probe_kernel_read(bpt->saved_instr, (char *)bpt->bpt_addr, - BREAK_INSTR_SIZE); + int ret = copy_from_kernel_nofault(bpt->saved_instr, + (char *)bpt->bpt_addr, BREAK_INSTR_SIZE); if (ret) return ret; diff --git a/arch/parisc/kernel/kprobes.c b/arch/parisc/kernel/kprobes.c index 77ec51818916..6e0b86652f30 100644 --- a/arch/parisc/kernel/kprobes.c +++ b/arch/parisc/kernel/kprobes.c @@ -5,6 +5,7 @@ * PA-RISC kprobes implementation * * Copyright (c) 2019 Sven Schnelle <svens@stackframe.org> + * Copyright (c) 2022 Helge Deller <deller@gmx.de> */ #include <linux/types.h> @@ -25,9 +26,14 @@ int __kprobes arch_prepare_kprobe(struct kprobe *p) if (!p->ainsn.insn) return -ENOMEM; - memcpy(p->ainsn.insn, p->addr, - MAX_INSN_SIZE * sizeof(kprobe_opcode_t)); + /* + * Set up new instructions. Second break instruction will + * trigger call of parisc_kprobe_ss_handler(). + */ p->opcode = *p->addr; + p->ainsn.insn[0] = p->opcode; + p->ainsn.insn[1] = PARISC_KPROBES_BREAK_INSN2; + flush_insn_slot(p); return 0; } @@ -73,9 +79,7 @@ static void __kprobes setup_singlestep(struct kprobe *p, { kcb->iaoq[0] = regs->iaoq[0]; kcb->iaoq[1] = regs->iaoq[1]; - regs->iaoq[0] = (unsigned long)p->ainsn.insn; - mtctl(0, 0); - regs->gr[0] |= PSW_R; + instruction_pointer_set(regs, (unsigned long)p->ainsn.insn); } int __kprobes parisc_kprobe_break_handler(struct pt_regs *regs) @@ -148,7 +152,7 @@ int __kprobes parisc_kprobe_ss_handler(struct pt_regs *regs) /* for absolute branch instructions we can copy iaoq_b. for relative * branch instructions we need to calculate the new address based on the * difference between iaoq_f and iaoq_b. We cannot use iaoq_b without - * modificationt because it's based on our ainsn.insn address. + * modifications because it's based on our ainsn.insn address. */ if (p->post_handler) @@ -165,9 +169,8 @@ int __kprobes parisc_kprobe_ss_handler(struct pt_regs *regs) regs->iaoq[0] = kcb->iaoq[1]; break; default: - regs->iaoq[1] = kcb->iaoq[0]; - regs->iaoq[1] += (regs->iaoq[1] - regs->iaoq[0]) + 4; regs->iaoq[0] = kcb->iaoq[1]; + regs->iaoq[1] = regs->iaoq[0] + 4; break; } kcb->kprobe_status = KPROBE_HIT_SSDONE; @@ -175,7 +178,7 @@ int __kprobes parisc_kprobe_ss_handler(struct pt_regs *regs) return 1; } -static inline void kretprobe_trampoline(void) +void __kretprobe_trampoline(void) { asm volatile("nop"); asm volatile("nop"); @@ -191,87 +194,22 @@ static struct kprobe trampoline_p = { static int __kprobes trampoline_probe_handler(struct kprobe *p, struct pt_regs *regs) { - struct kretprobe_instance *ri = NULL; - struct hlist_head *head, empty_rp; - struct hlist_node *tmp; - unsigned long flags, orig_ret_address = 0; - unsigned long trampoline_address = (unsigned long)trampoline_p.addr; - kprobe_opcode_t *correct_ret_addr = NULL; - - INIT_HLIST_HEAD(&empty_rp); - kretprobe_hash_lock(current, &head, &flags); - - /* - * It is possible to have multiple instances associated with a given - * task either because multiple functions in the call path have - * a return probe installed on them, and/or more than one return - * probe was registered for a target function. - * - * We can handle this because: - * - instances are always inserted at the head of the list - * - when multiple return probes are registered for the same - * function, the first instance's ret_addr will point to the - * real return address, and all the rest will point to - * kretprobe_trampoline - */ - hlist_for_each_entry_safe(ri, tmp, head, hlist) { - if (ri->task != current) - /* another task is sharing our hash bucket */ - continue; - - orig_ret_address = (unsigned long)ri->ret_addr; - - if (orig_ret_address != trampoline_address) - /* - * This is the real return address. Any other - * instances associated with this task are for - * other calls deeper on the call stack - */ - break; - } + __kretprobe_trampoline_handler(regs, NULL); - kretprobe_assert(ri, orig_ret_address, trampoline_address); - - correct_ret_addr = ri->ret_addr; - hlist_for_each_entry_safe(ri, tmp, head, hlist) { - if (ri->task != current) - /* another task is sharing our hash bucket */ - continue; - - orig_ret_address = (unsigned long)ri->ret_addr; - if (ri->rp && ri->rp->handler) { - __this_cpu_write(current_kprobe, &ri->rp->kp); - get_kprobe_ctlblk()->kprobe_status = KPROBE_HIT_ACTIVE; - ri->ret_addr = correct_ret_addr; - ri->rp->handler(ri, regs); - __this_cpu_write(current_kprobe, NULL); - } - - recycle_rp_inst(ri, &empty_rp); - - if (orig_ret_address != trampoline_address) - /* - * This is the real return address. Any other - * instances associated with this task are for - * other calls deeper on the call stack - */ - break; - } - - kretprobe_hash_unlock(current, &flags); - - hlist_for_each_entry_safe(ri, tmp, &empty_rp, hlist) { - hlist_del(&ri->hlist); - kfree(ri); - } - instruction_pointer_set(regs, orig_ret_address); return 1; } +void arch_kretprobe_fixup_return(struct pt_regs *regs, + kprobe_opcode_t *correct_ret_addr) +{ + regs->gr[2] = (unsigned long)correct_ret_addr; +} + void __kprobes arch_prepare_kretprobe(struct kretprobe_instance *ri, struct pt_regs *regs) { ri->ret_addr = (kprobe_opcode_t *)regs->gr[2]; + ri->fp = NULL; /* Replace the return addr with trampoline addr. */ regs->gr[2] = (unsigned long)trampoline_p.addr; @@ -285,6 +223,6 @@ int __kprobes arch_trampoline_kprobe(struct kprobe *p) int __init arch_init_kprobes(void) { trampoline_p.addr = (kprobe_opcode_t *) - dereference_function_descriptor(kretprobe_trampoline); + dereference_function_descriptor(__kretprobe_trampoline); return register_kprobe(&trampoline_p); } diff --git a/arch/parisc/kernel/module.c b/arch/parisc/kernel/module.c index 1c50093e2ebe..d214bbe3c2af 100644 --- a/arch/parisc/kernel/module.c +++ b/arch/parisc/kernel/module.c @@ -3,9 +3,9 @@ * * The best reference for this stuff is probably the Processor- * Specific ELF Supplement for PA-RISC: - * http://ftp.parisc-linux.org/docs/arch/elf-pa-hp.pdf + * https://parisc.wiki.kernel.org/index.php/File:Elf-pa-hp.pdf * - * Linux/PA-RISC Project (http://www.parisc-linux.org/) + * Linux/PA-RISC Project * Copyright (C) 2003 Randolph Chung <tausq at debian . org> * Copyright (C) 2008 Helge Deller <deller@gmx.de> * @@ -27,9 +27,9 @@ * We are not doing SEGREL32 handling correctly. According to the ABI, we * should do a value offset, like this: * if (in_init(me, (void *)val)) - * val -= (uint32_t)me->init_layout.base; + * val -= (uint32_t)me->mem[MOD_INIT_TEXT].base; * else - * val -= (uint32_t)me->core_layout.base; + * val -= (uint32_t)me->mem[MOD_TEXT].base; * However, SEGREL32 is used only for PARISC unwind entries, and we want * those entries to have an absolute address, and not just an offset. * @@ -50,7 +50,6 @@ #include <linux/mm.h> #include <linux/slab.h> -#include <asm/pgtable.h> #include <asm/unwind.h> #include <asm/sections.h> @@ -77,25 +76,6 @@ * allows us to allocate up to 4095 GOT entries. */ #define MAX_GOTS 4095 -/* three functions to determine where in the module core - * or init pieces the location is */ -static inline int in_init(struct module *me, void *loc) -{ - return (loc >= me->init_layout.base && - loc <= (me->init_layout.base + me->init_layout.size)); -} - -static inline int in_core(struct module *me, void *loc) -{ - return (loc >= me->core_layout.base && - loc <= (me->core_layout.base + me->core_layout.size)); -} - -static inline int in_local(struct module *me, void *loc) -{ - return in_init(me, loc) || in_core(me, loc); -} - #ifndef CONFIG_64BIT struct got_entry { Elf32_Addr addr; @@ -303,6 +283,7 @@ int module_frob_arch_sections(CONST Elf_Ehdr *hdr, { unsigned long gots = 0, fdescs = 0, len; unsigned int i; + struct module_memory *mod_mem; len = hdr->e_shnum * sizeof(me->arch.section[0]); me->arch.section = kzalloc(len, GFP_KERNEL); @@ -347,14 +328,15 @@ int module_frob_arch_sections(CONST Elf_Ehdr *hdr, me->arch.section[s].stub_entries += count; } + mod_mem = &me->mem[MOD_TEXT]; /* align things a bit */ - me->core_layout.size = ALIGN(me->core_layout.size, 16); - me->arch.got_offset = me->core_layout.size; - me->core_layout.size += gots * sizeof(struct got_entry); + mod_mem->size = ALIGN(mod_mem->size, 16); + me->arch.got_offset = mod_mem->size; + mod_mem->size += gots * sizeof(struct got_entry); - me->core_layout.size = ALIGN(me->core_layout.size, 16); - me->arch.fdesc_offset = me->core_layout.size; - me->core_layout.size += fdescs * sizeof(Elf_Fdesc); + mod_mem->size = ALIGN(mod_mem->size, 16); + me->arch.fdesc_offset = mod_mem->size; + mod_mem->size += fdescs * sizeof(Elf_Fdesc); me->arch.got_max = gots; me->arch.fdesc_max = fdescs; @@ -372,7 +354,7 @@ static Elf64_Word get_got(struct module *me, unsigned long value, long addend) BUG_ON(value == 0); - got = me->core_layout.base + me->arch.got_offset; + got = me->mem[MOD_TEXT].base + me->arch.got_offset; for (i = 0; got[i].addr; i++) if (got[i].addr == value) goto out; @@ -390,7 +372,7 @@ static Elf64_Word get_got(struct module *me, unsigned long value, long addend) #ifdef CONFIG_64BIT static Elf_Addr get_fdesc(struct module *me, unsigned long value) { - Elf_Fdesc *fdesc = me->core_layout.base + me->arch.fdesc_offset; + Elf_Fdesc *fdesc = me->mem[MOD_TEXT].base + me->arch.fdesc_offset; if (!value) { printk(KERN_ERR "%s: zero OPD requested!\n", me->name); @@ -408,7 +390,7 @@ static Elf_Addr get_fdesc(struct module *me, unsigned long value) /* Create new one */ fdesc->addr = value; - fdesc->gp = (Elf_Addr)me->core_layout.base + me->arch.got_offset; + fdesc->gp = (Elf_Addr)me->mem[MOD_TEXT].base + me->arch.got_offset; return (Elf_Addr)fdesc; } #endif /* CONFIG_64BIT */ @@ -743,7 +725,7 @@ int apply_relocate_add(Elf_Shdr *sechdrs, loc, val); val += addend; /* can we reach it locally? */ - if (in_local(me, (void *)val)) { + if (within_module(val, me)) { /* this is the case where the symbol is local * to the module, but in a different section, * so stub the jump in case it's more than 22 @@ -802,7 +784,7 @@ int apply_relocate_add(Elf_Shdr *sechdrs, break; case R_PARISC_FPTR64: /* 64-bit function address */ - if(in_local(me, (void *)(val + addend))) { + if (within_module(val + addend, me)) { *loc64 = get_fdesc(me, val+addend); pr_debug("FDESC for %s at %llx points to %llx\n", strtab + sym->st_name, *loc64, @@ -840,7 +822,7 @@ register_unwind_table(struct module *me, table = (unsigned char *)sechdrs[me->arch.unwind_section].sh_addr; end = table + sechdrs[me->arch.unwind_section].sh_size; - gp = (Elf_Addr)me->core_layout.base + me->arch.got_offset; + gp = (Elf_Addr)me->mem[MOD_TEXT].base + me->arch.got_offset; pr_debug("register_unwind_table(), sect = %d at 0x%p - 0x%p (gp=0x%lx)\n", me->arch.unwind_section, table, end, gp); @@ -863,7 +845,7 @@ int module_finalize(const Elf_Ehdr *hdr, const char *strtab = NULL; const Elf_Shdr *s; char *secstrings; - int symindex = -1; + int symindex __maybe_unused = -1; Elf_Sym *newptr, *oldptr; Elf_Shdr *symhdr = NULL; #ifdef DEBUG @@ -978,7 +960,7 @@ void module_arch_cleanup(struct module *mod) #ifdef CONFIG_64BIT void *dereference_module_function_descriptor(struct module *mod, void *ptr) { - unsigned long start_opd = (Elf64_Addr)mod->core_layout.base + + unsigned long start_opd = (Elf64_Addr)mod->mem[MOD_TEXT].base + mod->arch.fdesc_offset; unsigned long end_opd = start_opd + mod->arch.fdesc_count * sizeof(Elf64_Fdesc); diff --git a/arch/parisc/kernel/pa7300lc.c b/arch/parisc/kernel/pa7300lc.c deleted file mode 100644 index 0d770ac83f70..000000000000 --- a/arch/parisc/kernel/pa7300lc.c +++ /dev/null @@ -1,51 +0,0 @@ -// SPDX-License-Identifier: GPL-2.0 -/* - * linux/arch/parisc/kernel/pa7300lc.c - * - PA7300LC-specific functions - * - * Copyright (C) 2000 Philipp Rumpf */ - -#include <linux/sched.h> -#include <linux/sched/debug.h> -#include <linux/smp.h> -#include <linux/kernel.h> -#include <asm/io.h> -#include <asm/ptrace.h> -#include <asm/machdep.h> - -/* CPU register indices */ - -#define MIOC_STATUS 0xf040 -#define MIOC_CONTROL 0xf080 -#define MDERRADD 0xf0e0 -#define DMAERR 0xf0e8 -#define DIOERR 0xf0ec -#define HIDMAMEM 0xf0f4 - -/* this returns the HPA of the CPU it was called on */ -static u32 cpu_hpa(void) -{ - return 0xfffb0000; -} - -static void pa7300lc_lpmc(int code, struct pt_regs *regs) -{ - u32 hpa; - printk(KERN_WARNING "LPMC on CPU %d\n", smp_processor_id()); - - show_regs(regs); - - hpa = cpu_hpa(); - printk(KERN_WARNING - "MIOC_CONTROL %08x\n" "MIOC_STATUS %08x\n" - "MDERRADD %08x\n" "DMAERR %08x\n" - "DIOERR %08x\n" "HIDMAMEM %08x\n", - gsc_readl(hpa+MIOC_CONTROL), gsc_readl(hpa+MIOC_STATUS), - gsc_readl(hpa+MDERRADD), gsc_readl(hpa+DMAERR), - gsc_readl(hpa+DIOERR), gsc_readl(hpa+HIDMAMEM)); -} - -void pa7300lc_init(void) -{ - cpu_lpmc = pa7300lc_lpmc; -} diff --git a/arch/parisc/kernel/pacache.S b/arch/parisc/kernel/pacache.S index fa092ed1e837..541370d14559 100644 --- a/arch/parisc/kernel/pacache.S +++ b/arch/parisc/kernel/pacache.S @@ -21,12 +21,12 @@ #include <asm/psw.h> #include <asm/assembly.h> -#include <asm/pgtable.h> #include <asm/cache.h> #include <asm/ldcw.h> #include <asm/alternative.h> #include <linux/linkage.h> #include <linux/init.h> +#include <linux/pgtable.h> .section .text.hot .align 16 @@ -300,7 +300,6 @@ fdoneloop2: fdce,m %arg1(%sr1, %arg0) /* Fdce for one loop */ fdsync: - syncdma sync mtsm %r22 /* restore I-bit */ 89: ALTERNATIVE(88b, 89b, ALT_COND_NO_DCACHE, INSN_NOP) @@ -488,6 +487,8 @@ ENDPROC_CFI(copy_page_asm) * parisc chip designers that there will not ever be a parisc * chip with a larger alias boundary (Never say never :-) ). * + * Yah, what about the PA8800 and PA8900 processors? + * * Subtle: the dtlb miss handlers support the temp alias region by * "knowing" that if a dtlb miss happens within the temp alias * region it must have occurred while in clear_user_page. Since @@ -499,19 +500,10 @@ ENDPROC_CFI(copy_page_asm) * miss on the translation, the dtlb miss handler inserts the * translation into the tlb using these values: * - * %r26 physical page (shifted for tlb insert) of "to" translation - * %r23 physical page (shifted for tlb insert) of "from" translation + * %r26 physical address of "to" translation + * %r23 physical address of "from" translation */ - /* Drop prot bits and convert to page addr for iitlbt and idtlbt */ - #define PAGE_ADD_SHIFT (PAGE_SHIFT-12) - .macro convert_phys_for_tlb_insert20 phys - extrd,u \phys, 56-PAGE_ADD_SHIFT, 32-PAGE_ADD_SHIFT, \phys -#if _PAGE_SIZE_ENCODING_DEFAULT - depdi _PAGE_SIZE_ENCODING_DEFAULT, 63, (63-58), \phys -#endif - .endm - /* * copy_user_page_asm() performs a page copy using mappings * equivalent to the user page mappings. It can be used to @@ -540,24 +532,10 @@ ENTRY_CFI(copy_user_page_asm) sub %r25, %r1, %r23 ldil L%(TMPALIAS_MAP_START), %r28 -#ifdef CONFIG_64BIT -#if (TMPALIAS_MAP_START >= 0x80000000) - depdi 0, 31,32, %r28 /* clear any sign extension */ -#endif - convert_phys_for_tlb_insert20 %r26 /* convert phys addr to tlb insert format */ - convert_phys_for_tlb_insert20 %r23 /* convert phys addr to tlb insert format */ - depd %r24,63,22, %r28 /* Form aliased virtual address 'to' */ - depdi 0, 63,PAGE_SHIFT, %r28 /* Clear any offset bits */ - copy %r28, %r29 - depdi 1, 41,1, %r29 /* Form aliased virtual address 'from' */ -#else - extrw,u %r26, 24,25, %r26 /* convert phys addr to tlb insert format */ - extrw,u %r23, 24,25, %r23 /* convert phys addr to tlb insert format */ - depw %r24, 31,22, %r28 /* Form aliased virtual address 'to' */ - depwi 0, 31,PAGE_SHIFT, %r28 /* Clear any offset bits */ + dep_safe %r24, 31,TMPALIAS_SIZE_BITS, %r28 /* Form aliased virtual address 'to' */ + depi_safe 0, 31,PAGE_SHIFT, %r28 /* Clear any offset bits */ copy %r28, %r29 - depwi 1, 9,1, %r29 /* Form aliased virtual address 'from' */ -#endif + depi_safe 1, 31-TMPALIAS_SIZE_BITS,1, %r29 /* Form aliased virtual address 'from' */ /* Purge any old translations */ @@ -687,18 +665,8 @@ ENTRY_CFI(clear_user_page_asm) tophys_r1 %r26 ldil L%(TMPALIAS_MAP_START), %r28 -#ifdef CONFIG_64BIT -#if (TMPALIAS_MAP_START >= 0x80000000) - depdi 0, 31,32, %r28 /* clear any sign extension */ -#endif - convert_phys_for_tlb_insert20 %r26 /* convert phys addr to tlb insert format */ - depd %r25, 63,22, %r28 /* Form aliased virtual address 'to' */ - depdi 0, 63,PAGE_SHIFT, %r28 /* Clear any offset bits */ -#else - extrw,u %r26, 24,25, %r26 /* convert phys addr to tlb insert format */ - depw %r25, 31,22, %r28 /* Form aliased virtual address 'to' */ - depwi 0, 31,PAGE_SHIFT, %r28 /* Clear any offset bits */ -#endif + dep_safe %r25, 31,TMPALIAS_SIZE_BITS, %r28 /* Form aliased virtual address 'to' */ + depi_safe 0, 31,PAGE_SHIFT, %r28 /* Clear any offset bits */ /* Purge any old translation */ @@ -763,18 +731,8 @@ ENDPROC_CFI(clear_user_page_asm) ENTRY_CFI(flush_dcache_page_asm) ldil L%(TMPALIAS_MAP_START), %r28 -#ifdef CONFIG_64BIT -#if (TMPALIAS_MAP_START >= 0x80000000) - depdi 0, 31,32, %r28 /* clear any sign extension */ -#endif - convert_phys_for_tlb_insert20 %r26 /* convert phys addr to tlb insert format */ - depd %r25, 63,22, %r28 /* Form aliased virtual address 'to' */ - depdi 0, 63,PAGE_SHIFT, %r28 /* Clear any offset bits */ -#else - extrw,u %r26, 24,25, %r26 /* convert phys addr to tlb insert format */ - depw %r25, 31,22, %r28 /* Form aliased virtual address 'to' */ - depwi 0, 31,PAGE_SHIFT, %r28 /* Clear any offset bits */ -#endif + dep_safe %r25, 31,TMPALIAS_SIZE_BITS, %r28 /* Form aliased virtual address 'to' */ + depi_safe 0, 31,PAGE_SHIFT, %r28 /* Clear any offset bits */ /* Purge any old translation */ @@ -822,18 +780,8 @@ ENDPROC_CFI(flush_dcache_page_asm) ENTRY_CFI(purge_dcache_page_asm) ldil L%(TMPALIAS_MAP_START), %r28 -#ifdef CONFIG_64BIT -#if (TMPALIAS_MAP_START >= 0x80000000) - depdi 0, 31,32, %r28 /* clear any sign extension */ -#endif - convert_phys_for_tlb_insert20 %r26 /* convert phys addr to tlb insert format */ - depd %r25, 63,22, %r28 /* Form aliased virtual address 'to' */ - depdi 0, 63,PAGE_SHIFT, %r28 /* Clear any offset bits */ -#else - extrw,u %r26, 24,25, %r26 /* convert phys addr to tlb insert format */ - depw %r25, 31,22, %r28 /* Form aliased virtual address 'to' */ - depwi 0, 31,PAGE_SHIFT, %r28 /* Clear any offset bits */ -#endif + dep_safe %r25, 31,TMPALIAS_SIZE_BITS, %r28 /* Form aliased virtual address 'to' */ + depi_safe 0, 31,PAGE_SHIFT, %r28 /* Clear any offset bits */ /* Purge any old translation */ @@ -881,18 +829,8 @@ ENDPROC_CFI(purge_dcache_page_asm) ENTRY_CFI(flush_icache_page_asm) ldil L%(TMPALIAS_MAP_START), %r28 -#ifdef CONFIG_64BIT -#if (TMPALIAS_MAP_START >= 0x80000000) - depdi 0, 31,32, %r28 /* clear any sign extension */ -#endif - convert_phys_for_tlb_insert20 %r26 /* convert phys addr to tlb insert format */ - depd %r25, 63,22, %r28 /* Form aliased virtual address 'to' */ - depdi 0, 63,PAGE_SHIFT, %r28 /* Clear any offset bits */ -#else - extrw,u %r26, 24,25, %r26 /* convert phys addr to tlb insert format */ - depw %r25, 31,22, %r28 /* Form aliased virtual address 'to' */ - depwi 0, 31,PAGE_SHIFT, %r28 /* Clear any offset bits */ -#endif + dep_safe %r25, 31,TMPALIAS_SIZE_BITS, %r28 /* Form aliased virtual address 'to' */ + depi_safe 0, 31,PAGE_SHIFT, %r28 /* Clear any offset bits */ /* Purge any old translation. Note that the FIC instruction * may use either the instruction or data TLB. Given that we @@ -951,6 +889,7 @@ ENDPROC_CFI(flush_icache_page_asm) ENTRY_CFI(flush_kernel_dcache_page_asm) 88: ldil L%dcache_stride, %r1 ldw R%dcache_stride(%r1), %r23 + depi_safe 0, 31,PAGE_SHIFT, %r26 /* Clear any offset bits */ #ifdef CONFIG_64BIT depdi,z 1, 63-PAGE_SHIFT,1, %r25 @@ -987,6 +926,7 @@ ENDPROC_CFI(flush_kernel_dcache_page_asm) ENTRY_CFI(purge_kernel_dcache_page_asm) 88: ldil L%dcache_stride, %r1 ldw R%dcache_stride(%r1), %r23 + depi_safe 0, 31,PAGE_SHIFT, %r26 /* Clear any offset bits */ #ifdef CONFIG_64BIT depdi,z 1, 63-PAGE_SHIFT,1, %r25 @@ -1098,7 +1038,6 @@ ENTRY_CFI(flush_kernel_dcache_range_asm) sync 89: ALTERNATIVE(88b, 89b, ALT_COND_NO_DCACHE, INSN_NOP) - syncdma bv %r0(%r2) nop ENDPROC_CFI(flush_kernel_dcache_range_asm) @@ -1140,7 +1079,6 @@ ENTRY_CFI(purge_kernel_dcache_range_asm) sync 89: ALTERNATIVE(88b, 89b, ALT_COND_NO_DCACHE, INSN_NOP) - syncdma bv %r0(%r2) nop ENDPROC_CFI(purge_kernel_dcache_range_asm) @@ -1264,7 +1202,7 @@ ENTRY_CFI(flush_kernel_icache_range_asm) nop ENDPROC_CFI(flush_kernel_icache_range_asm) - __INIT + .text /* align should cover use of rfi in disable_sr_hashing_asm and * srdis_done. diff --git a/arch/parisc/kernel/parisc_ksyms.c b/arch/parisc/kernel/parisc_ksyms.c index 8ed409ecec93..6f0c92e8149d 100644 --- a/arch/parisc/kernel/parisc_ksyms.c +++ b/arch/parisc/kernel/parisc_ksyms.c @@ -14,13 +14,10 @@ #include <linux/module.h> #include <linux/kernel.h> #include <linux/syscalls.h> +#include <linux/libgcc.h> #include <linux/string.h> EXPORT_SYMBOL(memset); -EXPORT_SYMBOL(strlen); -EXPORT_SYMBOL(strcpy); -EXPORT_SYMBOL(strncpy); -EXPORT_SYMBOL(strcat); #include <linux/atomic.h> EXPORT_SYMBOL(__xchg8); @@ -36,7 +33,6 @@ EXPORT_SYMBOL(__xchg64); #include <linux/uaccess.h> EXPORT_SYMBOL(lclear_user); -EXPORT_SYMBOL(lstrnlen_user); #ifndef CONFIG_64BIT /* Needed so insmod can set dp value */ @@ -97,12 +93,6 @@ EXPORT_SYMBOL($$divI_12); EXPORT_SYMBOL($$divI_14); EXPORT_SYMBOL($$divI_15); -extern void __ashrdi3(void); -extern void __ashldi3(void); -extern void __lshrdi3(void); -extern void __muldi3(void); -extern void __ucmpdi2(void); - EXPORT_SYMBOL(__ashrdi3); EXPORT_SYMBOL(__ashldi3); EXPORT_SYMBOL(__lshrdi3); diff --git a/arch/parisc/kernel/patch.c b/arch/parisc/kernel/patch.c index 80a0ab372802..e59574f65e64 100644 --- a/arch/parisc/kernel/patch.c +++ b/arch/parisc/kernel/patch.c @@ -40,10 +40,7 @@ static void __kprobes *patch_map(void *addr, int fixmap, unsigned long *flags, *need_unmap = 1; set_fixmap(fixmap, page_to_phys(page)); - if (flags) - raw_spin_lock_irqsave(&patch_lock, *flags); - else - __acquire(&patch_lock); + raw_spin_lock_irqsave(&patch_lock, *flags); return (void *) (__fix_to_virt(fixmap) + (uintaddr & ~PAGE_MASK)); } @@ -52,10 +49,7 @@ static void __kprobes patch_unmap(int fixmap, unsigned long *flags) { clear_fixmap(fixmap); - if (flags) - raw_spin_unlock_irqrestore(&patch_lock, *flags); - else - __release(&patch_lock); + raw_spin_unlock_irqrestore(&patch_lock, *flags); } void __kprobes __patch_text_multiple(void *addr, u32 *insn, unsigned int len) @@ -67,8 +61,9 @@ void __kprobes __patch_text_multiple(void *addr, u32 *insn, unsigned int len) int mapped; /* Make sure we don't have any aliases in cache */ - flush_kernel_vmap_range(addr, len); - flush_icache_range(start, end); + flush_kernel_dcache_range_asm(start, end); + flush_kernel_icache_range_asm(start, end); + flush_tlb_kernel_range(start, end); p = fixmap = patch_map(addr, FIX_TEXT_POKE0, &flags, &mapped); @@ -81,8 +76,10 @@ void __kprobes __patch_text_multiple(void *addr, u32 *insn, unsigned int len) * We're crossing a page boundary, so * need to remap */ - flush_kernel_vmap_range((void *)fixmap, - (p-fixmap) * sizeof(*p)); + flush_kernel_dcache_range_asm((unsigned long)fixmap, + (unsigned long)p); + flush_tlb_kernel_range((unsigned long)fixmap, + (unsigned long)p); if (mapped) patch_unmap(FIX_TEXT_POKE0, &flags); p = fixmap = patch_map(addr, FIX_TEXT_POKE0, &flags, @@ -90,10 +87,10 @@ void __kprobes __patch_text_multiple(void *addr, u32 *insn, unsigned int len) } } - flush_kernel_vmap_range((void *)fixmap, (p-fixmap) * sizeof(*p)); + flush_kernel_dcache_range_asm((unsigned long)fixmap, (unsigned long)p); + flush_tlb_kernel_range((unsigned long)fixmap, (unsigned long)p); if (mapped) patch_unmap(FIX_TEXT_POKE0, &flags); - flush_icache_range(start, end); } void __kprobes __patch_text(void *addr, u32 insn) diff --git a/arch/parisc/kernel/pci-dma.c b/arch/parisc/kernel/pci-dma.c index 0f1b460ee715..bf9f192c826e 100644 --- a/arch/parisc/kernel/pci-dma.c +++ b/arch/parisc/kernel/pci-dma.c @@ -3,7 +3,7 @@ ** PARISC 1.1 Dynamic DMA mapping support. ** This implementation is for PA-RISC platforms that do not support ** I/O TLBs (aka DMA address translation hardware). -** See Documentation/DMA-API-HOWTO.txt for interface definitions. +** See Documentation/core-api/dma-api-howto.rst for interface definitions. ** ** (c) Copyright 1999,2000 Hewlett-Packard Company ** (c) Copyright 2000 Grant Grundler @@ -26,21 +26,20 @@ #include <linux/string.h> #include <linux/types.h> #include <linux/dma-direct.h> -#include <linux/dma-noncoherent.h> +#include <linux/dma-map-ops.h> #include <asm/cacheflush.h> #include <asm/dma.h> /* for DMA_CHUNK_SIZE */ #include <asm/io.h> #include <asm/page.h> /* get_order */ -#include <asm/pgalloc.h> #include <linux/uaccess.h> #include <asm/tlbflush.h> /* for purge_tlb_*() macros */ static struct proc_dir_entry * proc_gsc_root __read_mostly = NULL; -static unsigned long pcxl_used_bytes __read_mostly = 0; -static unsigned long pcxl_used_pages __read_mostly = 0; +static unsigned long pcxl_used_bytes __read_mostly; +static unsigned long pcxl_used_pages __read_mostly; -extern unsigned long pcxl_dma_start; /* Start of pcxl dma mapping area */ +unsigned long pcxl_dma_start __ro_after_init; /* pcxl dma mapping area start */ static DEFINE_SPINLOCK(pcxl_res_lock); static char *pcxl_res_map; static int pcxl_res_hint; @@ -92,7 +91,7 @@ static inline int map_pte_uncached(pte_t * pte, printk(KERN_ERR "map_pte_uncached: page already exists\n"); purge_tlb_start(flags); set_pte(pte, __mk_pte(*paddr_ptr, PAGE_KERNEL_UNC)); - pdtlb_kernel(orig_vaddr); + pdtlb(SR_KERNEL, orig_vaddr); purge_tlb_end(flags); vaddr += PAGE_SIZE; orig_vaddr += PAGE_SIZE; @@ -165,7 +164,7 @@ static inline void unmap_uncached_pte(pmd_t * pmd, unsigned long vaddr, pmd_clear(pmd); return; } - pte = pte_offset_map(pmd, vaddr); + pte = pte_offset_kernel(pmd, vaddr); vaddr &= ~PMD_MASK; end = vaddr + size; if (end > PMD_SIZE) @@ -176,7 +175,7 @@ static inline void unmap_uncached_pte(pmd_t * pmd, unsigned long vaddr, pte_clear(&init_mm, vaddr, pte); purge_tlb_start(flags); - pdtlb_kernel(orig_vaddr); + pdtlb(SR_KERNEL, orig_vaddr); purge_tlb_end(flags); vaddr += PAGE_SIZE; orig_vaddr += PAGE_SIZE; @@ -201,7 +200,7 @@ static inline void unmap_uncached_pmd(pgd_t * dir, unsigned long vaddr, pgd_clear(dir); return; } - pmd = pmd_offset(dir, vaddr); + pmd = pmd_offset(pud_offset(p4d_offset(dir, vaddr), vaddr), vaddr); vaddr &= ~PGDIR_MASK; end = vaddr + size; if (end > PGDIR_SIZE) @@ -246,7 +245,7 @@ static void unmap_uncached_pages(unsigned long vaddr, unsigned long size) PCXL_SEARCH_LOOP(idx, mask, size); \ } -unsigned long +static unsigned long pcxl_alloc_range(size_t size) { int res_idx; @@ -336,7 +335,7 @@ pcxl_free_range(unsigned long vaddr, size_t size) dump_resmap(); } -static int proc_pcxl_dma_show(struct seq_file *m, void *v) +static int __maybe_unused proc_pcxl_dma_show(struct seq_file *m, void *v) { #if 0 u_long i = 0; @@ -382,7 +381,7 @@ pcxl_dma_init(void) pcxl_res_map = (char *)__get_free_pages(GFP_KERNEL, get_order(pcxl_res_size)); memset(pcxl_res_map, 0, pcxl_res_size); - proc_gsc_root = proc_mkdir("gsc", NULL); + proc_gsc_root = proc_mkdir("bus/gsc", NULL); if (!proc_gsc_root) printk(KERN_WARNING "pcxl_dma_init: Unable to create gsc /proc dir entry\n"); @@ -418,14 +417,6 @@ void *arch_dma_alloc(struct device *dev, size_t size, map_uncached_pages(vaddr, size, paddr); *dma_handle = (dma_addr_t) paddr; -#if 0 -/* This probably isn't needed to support EISA cards. -** ISA cards will certainly only support 24-bit DMA addressing. -** Not clear if we can, want, or need to support ISA. -*/ - if (!dev || *dev->coherent_dma_mask < 0xffffffff) - gfp |= GFP_DMA; -#endif return (void *)vaddr; } @@ -447,17 +438,27 @@ void arch_dma_free(struct device *dev, size_t size, void *vaddr, void arch_sync_dma_for_device(phys_addr_t paddr, size_t size, enum dma_data_direction dir) { + /* + * fdc: The data cache line is written back to memory, if and only if + * it is dirty, and then invalidated from the data cache. + */ flush_kernel_dcache_range((unsigned long)phys_to_virt(paddr), size); } void arch_sync_dma_for_cpu(phys_addr_t paddr, size_t size, enum dma_data_direction dir) { - flush_kernel_dcache_range((unsigned long)phys_to_virt(paddr), size); -} + unsigned long addr = (unsigned long) phys_to_virt(paddr); -void arch_dma_cache_sync(struct device *dev, void *vaddr, size_t size, - enum dma_data_direction direction) -{ - flush_kernel_dcache_range((unsigned long)vaddr, size); + switch (dir) { + case DMA_TO_DEVICE: + case DMA_BIDIRECTIONAL: + flush_kernel_dcache_range(addr, size); + return; + case DMA_FROM_DEVICE: + purge_kernel_dcache_range_asm(addr, addr + size); + return; + default: + BUG(); + } } diff --git a/arch/parisc/kernel/pdc_chassis.c b/arch/parisc/kernel/pdc_chassis.c index 75ae88d13909..d477d0177c2f 100644 --- a/arch/parisc/kernel/pdc_chassis.c +++ b/arch/parisc/kernel/pdc_chassis.c @@ -20,6 +20,7 @@ #include <linux/init.h> #include <linux/module.h> #include <linux/kernel.h> +#include <linux/panic_notifier.h> #include <linux/reboot.h> #include <linux/notifier.h> #include <linux/cache.h> @@ -30,6 +31,7 @@ #include <asm/processor.h> #include <asm/pdc.h> #include <asm/pdcpat.h> +#include <asm/led.h> #define PDC_CHASSIS_VER "0.05" @@ -39,7 +41,7 @@ static unsigned int pdc_chassis_enabled __read_mostly = 1; /** * pdc_chassis_setup() - Enable/disable pdc_chassis code at boot time. - * @str configuration param: 0 to disable chassis log + * @str: configuration param: 0 to disable chassis log * @return 1 */ @@ -54,7 +56,6 @@ __setup("pdcchassis=", pdc_chassis_setup); /** * pdc_chassis_checkold() - Checks for old PDC_CHASSIS compatibility - * @pdc_chassis_old: 1 if old pdc chassis style * * Currently, only E class and A180 are known to work with this. * Inspired by Christoph Plattner @@ -79,6 +80,9 @@ static void __init pdc_chassis_checkold(void) /** * pdc_chassis_panic_event() - Called by the panic handler. + * @this: unused + * @event: unused + * @ptr: unused * * As soon as a panic occurs, we should inform the PDC. */ @@ -87,7 +91,7 @@ static int pdc_chassis_panic_event(struct notifier_block *this, unsigned long event, void *ptr) { pdc_chassis_send_status(PDC_CHASSIS_DIRECT_PANIC); - return NOTIFY_DONE; + return NOTIFY_DONE; } @@ -98,7 +102,10 @@ static struct notifier_block pdc_chassis_panic_block = { /** - * parisc_reboot_event() - Called by the reboot handler. + * pdc_chassis_reboot_event() - Called by the reboot handler. + * @this: unused + * @event: unused + * @ptr: unused * * As soon as a reboot occurs, we should inform the PDC. */ @@ -107,7 +114,7 @@ static int pdc_chassis_reboot_event(struct notifier_block *this, unsigned long event, void *ptr) { pdc_chassis_send_status(PDC_CHASSIS_DIRECT_SHUTDOWN); - return NOTIFY_DONE; + return NOTIFY_DONE; } @@ -147,7 +154,7 @@ void __init parisc_pdc_chassis_init(void) /** * pdc_chassis_send_status() - Sends a predefined message to the chassis, * and changes the front panel LEDs according to the new system state - * @retval: PDC call return value. + * @message: Type of message, one of PDC_CHASSIS_DIRECT_* values. * * Only machines with 64 bits PDC PAT and those reported in * pdc_chassis_checkold() are supported atm. @@ -228,6 +235,11 @@ int pdc_chassis_send_status(int message) } else retval = -1; #endif /* CONFIG_64BIT */ } /* if (pdc_chassis_enabled) */ + + /* if system has LCD display, update current string */ + if (retval != -1 && IS_ENABLED(CONFIG_CHASSIS_LCD_LED)) + lcd_print(NULL); + #endif /* CONFIG_PDC_CHASSIS */ return retval; } diff --git a/arch/parisc/kernel/pdc_cons.c b/arch/parisc/kernel/pdc_cons.c index 7ed404c60a9e..cf3bf8232374 100644 --- a/arch/parisc/kernel/pdc_cons.c +++ b/arch/parisc/kernel/pdc_cons.c @@ -1,267 +1,65 @@ // SPDX-License-Identifier: GPL-2.0-or-later /* - * PDC Console support - ie use firmware to dump text via boot console + * PDC early console support - use PDC firmware to dump text via boot console * - * Copyright (C) 1999-2003 Matthew Wilcox <willy at parisc-linux.org> - * Copyright (C) 2000 Martin K Petersen <mkp at mkp.net> - * Copyright (C) 2000 John Marvin <jsm at parisc-linux.org> - * Copyright (C) 2000-2003 Paul Bame <bame at parisc-linux.org> - * Copyright (C) 2000 Philipp Rumpf <prumpf with tux.org> - * Copyright (C) 2000 Michael Ang <mang with subcarrier.org> - * Copyright (C) 2000 Grant Grundler <grundler with parisc-linux.org> - * Copyright (C) 2001-2002 Ryan Bradetich <rbrad at parisc-linux.org> - * Copyright (C) 2001 Helge Deller <deller at parisc-linux.org> - * Copyright (C) 2001 Thomas Bogendoerfer <tsbogend at parisc-linux.org> - * Copyright (C) 2002 Randolph Chung <tausq with parisc-linux.org> - * Copyright (C) 2010 Guy Martin <gmsoft at tuxicoman.be> + * Copyright (C) 2001-2022 Helge Deller <deller@gmx.de> */ -/* - * The PDC console is a simple console, which can be used for debugging - * boot related problems on HP PA-RISC machines. It is also useful when no - * other console works. - * - * This code uses the ROM (=PDC) based functions to read and write characters - * from and to PDC's boot path. - */ - -/* Define EARLY_BOOTUP_DEBUG to debug kernel related boot problems. - * On production kernels EARLY_BOOTUP_DEBUG should be undefined. */ -#define EARLY_BOOTUP_DEBUG - - -#include <linux/kernel.h> #include <linux/console.h> -#include <linux/string.h> #include <linux/init.h> -#include <linux/major.h> -#include <linux/tty.h> +#include <linux/serial_core.h> +#include <linux/kgdb.h> #include <asm/page.h> /* for PAGE0 */ #include <asm/pdc.h> /* for iodc_call() proto and friends */ -static DEFINE_SPINLOCK(pdc_console_lock); -static struct console pdc_cons; - static void pdc_console_write(struct console *co, const char *s, unsigned count) { int i = 0; - unsigned long flags; - spin_lock_irqsave(&pdc_console_lock, flags); do { i += pdc_iodc_print(s + i, count - i); } while (i < count); - spin_unlock_irqrestore(&pdc_console_lock, flags); -} - -int pdc_console_poll_key(struct console *co) -{ - int c; - unsigned long flags; - - spin_lock_irqsave(&pdc_console_lock, flags); - c = pdc_iodc_getc(); - spin_unlock_irqrestore(&pdc_console_lock, flags); - - return c; -} - -static int pdc_console_setup(struct console *co, char *options) -{ - return 0; -} - -#if defined(CONFIG_PDC_CONSOLE) -#include <linux/vt_kern.h> -#include <linux/tty_flip.h> - -#define PDC_CONS_POLL_DELAY (30 * HZ / 1000) - -static void pdc_console_poll(struct timer_list *unused); -static DEFINE_TIMER(pdc_console_timer, pdc_console_poll); -static struct tty_port tty_port; - -static int pdc_console_tty_open(struct tty_struct *tty, struct file *filp) -{ - tty_port_tty_set(&tty_port, tty); - mod_timer(&pdc_console_timer, jiffies + PDC_CONS_POLL_DELAY); - - return 0; } -static void pdc_console_tty_close(struct tty_struct *tty, struct file *filp) +#ifdef CONFIG_KGDB +static int kgdb_pdc_read_char(void) { - if (tty->count == 1) { - del_timer_sync(&pdc_console_timer); - tty_port_tty_set(&tty_port, NULL); - } -} - -static int pdc_console_tty_write(struct tty_struct *tty, const unsigned char *buf, int count) -{ - pdc_console_write(NULL, buf, count); - return count; -} + int c = pdc_iodc_getc(); -static int pdc_console_tty_write_room(struct tty_struct *tty) -{ - return 32768; /* no limit, no buffer used */ + return (c <= 0) ? NO_POLL_CHAR : c; } -static int pdc_console_tty_chars_in_buffer(struct tty_struct *tty) +static void kgdb_pdc_write_char(u8 chr) { - return 0; /* no buffer */ + /* no need to print char as it's shown on standard console */ + /* pdc_iodc_print(&chr, 1); */ } -static const struct tty_operations pdc_console_tty_ops = { - .open = pdc_console_tty_open, - .close = pdc_console_tty_close, - .write = pdc_console_tty_write, - .write_room = pdc_console_tty_write_room, - .chars_in_buffer = pdc_console_tty_chars_in_buffer, +static struct kgdb_io kgdb_pdc_io_ops = { + .name = "kgdb_pdc", + .read_char = kgdb_pdc_read_char, + .write_char = kgdb_pdc_write_char, }; - -static void pdc_console_poll(struct timer_list *unused) -{ - int data, count = 0; - - while (1) { - data = pdc_console_poll_key(NULL); - if (data == -1) - break; - tty_insert_flip_char(&tty_port, data & 0xFF, TTY_NORMAL); - count ++; - } - - if (count) - tty_flip_buffer_push(&tty_port); - - if (pdc_cons.flags & CON_ENABLED) - mod_timer(&pdc_console_timer, jiffies + PDC_CONS_POLL_DELAY); -} - -static struct tty_driver *pdc_console_tty_driver; - -static int __init pdc_console_tty_driver_init(void) -{ - int err; - - /* Check if the console driver is still registered. - * It is unregistered if the pdc console was not selected as the - * primary console. */ - - struct console *tmp; - - console_lock(); - for_each_console(tmp) - if (tmp == &pdc_cons) - break; - console_unlock(); - - if (!tmp) { - printk(KERN_INFO "PDC console driver not registered anymore, not creating %s\n", pdc_cons.name); - return -ENODEV; - } - - printk(KERN_INFO "The PDC console driver is still registered, removing CON_BOOT flag\n"); - pdc_cons.flags &= ~CON_BOOT; - - pdc_console_tty_driver = alloc_tty_driver(1); - - if (!pdc_console_tty_driver) - return -ENOMEM; - - tty_port_init(&tty_port); - - pdc_console_tty_driver->driver_name = "pdc_cons"; - pdc_console_tty_driver->name = "ttyB"; - pdc_console_tty_driver->major = MUX_MAJOR; - pdc_console_tty_driver->minor_start = 0; - pdc_console_tty_driver->type = TTY_DRIVER_TYPE_SYSTEM; - pdc_console_tty_driver->init_termios = tty_std_termios; - pdc_console_tty_driver->flags = TTY_DRIVER_REAL_RAW | - TTY_DRIVER_RESET_TERMIOS; - tty_set_operations(pdc_console_tty_driver, &pdc_console_tty_ops); - tty_port_link_device(&tty_port, pdc_console_tty_driver, 0); - - err = tty_register_driver(pdc_console_tty_driver); - if (err) { - printk(KERN_ERR "Unable to register the PDC console TTY driver\n"); - tty_port_destroy(&tty_port); - return err; - } - - return 0; -} -device_initcall(pdc_console_tty_driver_init); - -static struct tty_driver * pdc_console_device (struct console *c, int *index) -{ - *index = c->index; - return pdc_console_tty_driver; -} -#else -#define pdc_console_device NULL #endif -static struct console pdc_cons = { - .name = "ttyB", - .write = pdc_console_write, - .device = pdc_console_device, - .setup = pdc_console_setup, - .flags = CON_BOOT | CON_PRINTBUFFER, - .index = -1, -}; - -static int pdc_console_initialized; - -static void pdc_console_init_force(void) +static int __init pdc_earlycon_setup(struct earlycon_device *device, + const char *opt) { - if (pdc_console_initialized) - return; - ++pdc_console_initialized; - + struct console *earlycon_console; + /* If the console is duplex then copy the COUT parameters to CIN. */ if (PAGE0->mem_cons.cl_class == CL_DUPLEX) memcpy(&PAGE0->mem_kbd, &PAGE0->mem_cons, sizeof(PAGE0->mem_cons)); - /* register the pdc console */ - register_console(&pdc_cons); -} + earlycon_console = device->con; + earlycon_console->write = pdc_console_write; + device->port.iotype = UPIO_MEM32BE; -void __init pdc_console_init(void) -{ -#if defined(EARLY_BOOTUP_DEBUG) || defined(CONFIG_PDC_CONSOLE) - pdc_console_init_force(); +#ifdef CONFIG_KGDB + kgdb_register_io_module(&kgdb_pdc_io_ops); #endif -#ifdef EARLY_BOOTUP_DEBUG - printk(KERN_INFO "Initialized PDC Console for debugging.\n"); -#endif -} - -/* - * Used for emergencies. Currently only used if an HPMC occurs. If an - * HPMC occurs, it is possible that the current console may not be - * properly initialised after the PDC IO reset. This routine unregisters - * all of the current consoles, reinitializes the pdc console and - * registers it. - */ - -void pdc_console_restart(void) -{ - struct console *console; - - if (pdc_console_initialized) - return; - - /* If we've already seen the output, don't bother to print it again */ - if (console_drivers != NULL) - pdc_cons.flags &= ~CON_PRINTBUFFER; - - while ((console = console_drivers) != NULL) - unregister_console(console_drivers); - - /* force registering the pdc console */ - pdc_console_init_force(); + return 0; } + +EARLYCON_DECLARE(pdc, pdc_earlycon_setup); diff --git a/arch/parisc/kernel/pdt.c b/arch/parisc/kernel/pdt.c index 36434d4da381..0f9b3b5914cf 100644 --- a/arch/parisc/kernel/pdt.c +++ b/arch/parisc/kernel/pdt.c @@ -16,7 +16,10 @@ #include <linux/memblock.h> #include <linux/seq_file.h> #include <linux/kthread.h> +#include <linux/proc_fs.h> #include <linux/initrd.h> +#include <linux/pgtable.h> +#include <linux/mm.h> #include <asm/pdc.h> #include <asm/pdcpat.h> @@ -230,6 +233,7 @@ void __init pdc_pdt_init(void) /* mark memory page bad */ memblock_reserve(pdt_entry[i] & PAGE_MASK, PAGE_SIZE); + num_poisoned_pages_inc(addr >> PAGE_SHIFT); } } @@ -327,8 +331,7 @@ static int pdt_mainloop(void *unused) ((pde & PDT_ADDR_SINGLE_ERR) == 0)) memory_failure(pde >> PAGE_SHIFT, 0); else - soft_offline_page( - pfn_to_page(pde >> PAGE_SHIFT), 0); + soft_offline_page(pde >> PAGE_SHIFT, 0); #else pr_crit("PDT: memory error at 0x%lx ignored.\n" "Rebuild kernel with CONFIG_MEMORY_FAILURE=y " @@ -350,13 +353,9 @@ static int __init pdt_initcall(void) if (pdt_type == PDT_NONE) return -ENODEV; - kpdtd_task = kthread_create(pdt_mainloop, NULL, "kpdtd"); - if (IS_ERR(kpdtd_task)) - return PTR_ERR(kpdtd_task); + kpdtd_task = kthread_run(pdt_mainloop, NULL, "kpdtd"); - wake_up_process(kpdtd_task); - - return 0; + return PTR_ERR_OR_ZERO(kpdtd_task); } late_initcall(pdt_initcall); diff --git a/arch/parisc/kernel/perf.c b/arch/parisc/kernel/perf.c index 676683641d00..b0f0816879df 100644 --- a/arch/parisc/kernel/perf.c +++ b/arch/parisc/kernel/perf.c @@ -57,7 +57,7 @@ struct rdr_tbl_ent { static int perf_processor_interface __read_mostly = UNKNOWN_INTF; static int perf_enabled __read_mostly; static DEFINE_SPINLOCK(perf_lock); -struct parisc_device *cpu_device __read_mostly; +static struct parisc_device *cpu_device __read_mostly; /* RDRs to write for PCX-W */ static const int perf_rdrs_W[] = @@ -288,7 +288,7 @@ static ssize_t perf_read(struct file *file, char __user *buf, size_t cnt, loff_t static ssize_t perf_write(struct file *file, const char __user *buf, size_t count, loff_t *ppos) { - size_t image_size; + size_t image_size __maybe_unused; uint32_t image_type; uint32_t interface_type; uint32_t test; @@ -300,7 +300,7 @@ static ssize_t perf_write(struct file *file, const char __user *buf, else return -EFAULT; - if (!capable(CAP_SYS_ADMIN)) + if (!perfmon_capable()) return -EACCES; if (count != sizeof(uint32_t)) @@ -792,7 +792,7 @@ static int perf_write_image(uint64_t *memaddr) return -1; } - runway = ioremap_nocache(cpu_device->hpa.start, 4096); + runway = ioremap(cpu_device->hpa.start, 4096); if (!runway) { pr_err("perf_write_image: ioremap failed!\n"); return -ENOMEM; diff --git a/arch/parisc/kernel/process.c b/arch/parisc/kernel/process.c index ecc5c2771208..ed93bd8c1545 100644 --- a/arch/parisc/kernel/process.c +++ b/arch/parisc/kernel/process.c @@ -17,9 +17,6 @@ * Copyright (C) 2001-2014 Helge Deller <deller@gmx.de> * Copyright (C) 2002 Randolph Chung <tausq with parisc-linux.org> */ - -#include <stdarg.h> - #include <linux/elf.h> #include <linux/errno.h> #include <linux/kernel.h> @@ -29,6 +26,7 @@ #include <linux/module.h> #include <linux/personality.h> #include <linux/ptrace.h> +#include <linux/reboot.h> #include <linux/sched.h> #include <linux/sched/debug.h> #include <linux/sched/task.h> @@ -41,15 +39,16 @@ #include <linux/rcupdate.h> #include <linux/random.h> #include <linux/nmi.h> +#include <linux/sched/hotplug.h> #include <asm/io.h> #include <asm/asm-offsets.h> #include <asm/assembly.h> #include <asm/pdc.h> #include <asm/pdc_chassis.h> -#include <asm/pgalloc.h> #include <asm/unwind.h> #include <asm/sections.h> +#include <asm/cacheflush.h> #define COMMAND_GLOBAL F_EXTEND(0xfffe0030) #define CMD_RESET 5 /* reset any module */ @@ -98,18 +97,12 @@ void machine_restart(char *cmd) } -void (*chassis_power_off)(void); - /* * This routine is called from sys_reboot to actually turn off the * machine */ void machine_power_off(void) { - /* If there is a registered power off handler, call it. */ - if (chassis_power_off) - chassis_power_off(); - /* Put the soft power button back under hardware control. * If the user had already pressed the power button, the * following call will immediately power off. */ @@ -118,19 +111,23 @@ void machine_power_off(void) pdc_chassis_send_status(PDC_CHASSIS_DIRECT_SHUTDOWN); /* ipmi_poweroff may have been installed. */ - if (pm_power_off) - pm_power_off(); + do_kernel_power_off(); /* It seems we have no way to power the system off via * software. The user has to press the button himself. */ - printk(KERN_EMERG "System shut down completed.\n" - "Please power this system off now."); + printk("Power off or press RETURN to reboot.\n"); /* prevent soft lockup/stalled CPU messages for endless loop. */ rcu_sysrq_start(); lockup_detector_soft_poweroff(); - for (;;); + while (1) { + /* reboot if user presses RETURN key */ + if (pdc_iodc_getc() == 13) { + printk("Rebooting...\n"); + machine_restart(NULL); + } + } } void (*pm_power_off)(void); @@ -148,29 +145,6 @@ void flush_thread(void) */ } -void release_thread(struct task_struct *dead_task) -{ -} - -/* - * Fill in the FPU structure for a core dump. - */ - -int dump_fpu (struct pt_regs * regs, elf_fpregset_t *r) -{ - if (regs == NULL) - return 0; - - memcpy(r, regs->fr, sizeof *r); - return 1; -} - -int dump_task_fpu (struct task_struct *tsk, elf_fpregset_t *r) -{ - memcpy(r, tsk->thread.regs.fr, sizeof(*r)); - return 1; -} - /* * Idle thread support * @@ -181,16 +155,33 @@ int dump_task_fpu (struct task_struct *tsk, elf_fpregset_t *r) int running_on_qemu __ro_after_init; EXPORT_SYMBOL(running_on_qemu); -void __cpuidle arch_cpu_idle_dead(void) +/* + * Called from the idle thread for the CPU which has been shutdown. + */ +void __noreturn arch_cpu_idle_dead(void) { - /* nop on real hardware, qemu will offline CPU. */ - asm volatile("or %%r31,%%r31,%%r31\n":::); +#ifdef CONFIG_HOTPLUG_CPU + idle_task_exit(); + + local_irq_disable(); + + /* Tell the core that this CPU is now safe to dispose of. */ + cpuhp_ap_report_dead(); + + /* Ensure that the cache lines are written out. */ + flush_cache_all_local(); + flush_tlb_all_local(NULL); + + /* Let PDC firmware put CPU into firmware idle loop. */ + __pdc_cpu_rendezvous(); + + pr_warn("PDC does not provide rendezvous function.\n"); +#endif + while (1); } void __cpuidle arch_cpu_idle(void) { - local_irq_enable(); - /* nop on real hardware, qemu will idle sleep. */ asm volatile("or %%r10,%%r10,%%r10\n":::); } @@ -208,9 +199,11 @@ arch_initcall(parisc_idle_init); * Copy architecture-specific thread state */ int -copy_thread(unsigned long clone_flags, unsigned long usp, - unsigned long kthread_arg, struct task_struct *p) +copy_thread(struct task_struct *p, const struct kernel_clone_args *args) { + unsigned long clone_flags = args->flags; + unsigned long usp = args->stack; + unsigned long tls = args->tls; struct pt_regs *cregs = &(p->thread.regs); void *stack = task_stack_page(p); @@ -220,27 +213,27 @@ copy_thread(unsigned long clone_flags, unsigned long usp, extern void * const ret_from_kernel_thread; extern void * const child_return; - if (unlikely(p->flags & PF_KTHREAD)) { + if (unlikely(args->fn)) { /* kernel thread */ memset(cregs, 0, sizeof(struct pt_regs)); - if (!usp) /* idle thread */ + if (args->idle) /* idle thread */ return 0; /* Must exit via ret_from_kernel_thread in order * to call schedule_tail() */ - cregs->ksp = (unsigned long)stack + THREAD_SZ_ALGN + FRAME_SIZE; + cregs->ksp = (unsigned long) stack + FRAME_SIZE + PT_SZ_ALGN; cregs->kpc = (unsigned long) &ret_from_kernel_thread; /* * Copy function and argument to be called from * ret_from_kernel_thread. */ #ifdef CONFIG_64BIT - cregs->gr[27] = ((unsigned long *)usp)[3]; - cregs->gr[26] = ((unsigned long *)usp)[2]; + cregs->gr[27] = ((unsigned long *)args->fn)[3]; + cregs->gr[26] = ((unsigned long *)args->fn)[2]; #else - cregs->gr[26] = usp; + cregs->gr[26] = (unsigned long) args->fn; #endif - cregs->gr[25] = kthread_arg; + cregs->gr[25] = (unsigned long) args->fn_arg; } else { /* user thread */ /* usp must be word aligned. This also prevents users from @@ -251,27 +244,24 @@ copy_thread(unsigned long clone_flags, unsigned long usp, if (likely(usp)) cregs->gr[30] = usp; } - cregs->ksp = (unsigned long)stack + THREAD_SZ_ALGN + FRAME_SIZE; + cregs->ksp = (unsigned long) stack + FRAME_SIZE; cregs->kpc = (unsigned long) &child_return; - /* Setup thread TLS area from the 4th parameter in clone */ + /* Setup thread TLS area */ if (clone_flags & CLONE_SETTLS) - cregs->cr27 = cregs->gr[23]; + cregs->cr27 = tls; } return 0; } unsigned long -get_wchan(struct task_struct *p) +__get_wchan(struct task_struct *p) { struct unwind_frame_info info; unsigned long ip; int count = 0; - if (!p || p == current || p->state == TASK_RUNNING) - return 0; - /* * These bracket the sleeping functions.. */ @@ -280,44 +270,11 @@ get_wchan(struct task_struct *p) do { if (unwind_once(&info) < 0) return 0; + if (task_is_running(p)) + return 0; ip = info.ip; if (!in_sched_functions(ip)) return ip; } while (count++ < MAX_UNWIND_ENTRIES); return 0; } - -#ifdef CONFIG_64BIT -void *dereference_function_descriptor(void *ptr) -{ - Elf64_Fdesc *desc = ptr; - void *p; - - if (!probe_kernel_address(&desc->addr, p)) - ptr = p; - return ptr; -} - -void *dereference_kernel_function_descriptor(void *ptr) -{ - if (ptr < (void *)__start_opd || - ptr >= (void *)__end_opd) - return ptr; - - return dereference_function_descriptor(ptr); -} -#endif - -static inline unsigned long brk_rnd(void) -{ - return (get_random_int() & BRK_RND_MASK) << PAGE_SHIFT; -} - -unsigned long arch_randomize_brk(struct mm_struct *mm) -{ - unsigned long ret = PAGE_ALIGN(mm->brk + brk_rnd()); - - if (ret < mm->brk) - return mm->brk; - return ret; -} diff --git a/arch/parisc/kernel/processor.c b/arch/parisc/kernel/processor.c index 13f771f74ee3..bf73562706b2 100644 --- a/arch/parisc/kernel/processor.c +++ b/arch/parisc/kernel/processor.c @@ -19,12 +19,14 @@ #include <linux/random.h> #include <linux/slab.h> #include <linux/cpu.h> +#include <asm/topology.h> #include <asm/param.h> #include <asm/cache.h> #include <asm/hardware.h> /* for register_parisc_driver() stuff */ #include <asm/processor.h> #include <asm/page.h> #include <asm/pdc.h> +#include <asm/smp.h> #include <asm/pdcpat.h> #include <asm/irq.h> /* for struct irq_region */ #include <asm/parisc-device.h> @@ -57,7 +59,7 @@ DEFINE_PER_CPU(struct cpuinfo_parisc, cpu_data); */ /** - * init_cpu_profiler - enable/setup per cpu profiling hooks. + * init_percpu_prof - enable/setup per cpu profiling hooks. * @cpunum: The processor instance. * * FIXME: doesn't do much yet... @@ -163,7 +165,6 @@ static int __init processor_probe(struct parisc_device *dev) if (cpuid) memset(p, 0, sizeof(struct cpuinfo_parisc)); - p->loops_per_jiffy = loops_per_jiffy; p->dev = dev; /* Save IODC data in case we need it */ p->hpa = dev->hpa.start; /* save CPU hpa */ p->cpuid = cpuid; /* save CPU id */ @@ -212,7 +213,7 @@ static int __init processor_probe(struct parisc_device *dev) #ifdef CONFIG_SMP if (cpuid) { set_cpu_present(cpuid, true); - cpu_up(cpuid); + add_cpu(cpuid); } #endif @@ -240,9 +241,9 @@ void __init collect_boot_cpu_data(void) /* get CPU-Model Information... */ #define p ((unsigned long *)&boot_cpu_data.pdc.model) if (pdc_model_info(&boot_cpu_data.pdc.model) == PDC_OK) { - printk(KERN_INFO - "model %08lx %08lx %08lx %08lx %08lx %08lx %08lx %08lx %08lx\n", - p[0], p[1], p[2], p[3], p[4], p[5], p[6], p[7], p[8]); + printk(KERN_INFO + "model %08lx %08lx %08lx %08lx %08lx %08lx %08lx %08lx %08lx %08lx\n", + p[0], p[1], p[2], p[3], p[4], p[5], p[6], p[7], p[8], p[9]); add_device_randomness(&boot_cpu_data.pdc.model, sizeof(boot_cpu_data.pdc.model)); @@ -271,10 +272,15 @@ void __init collect_boot_cpu_data(void) printk(KERN_INFO "capabilities 0x%lx\n", boot_cpu_data.pdc.capabilities); - if (pdc_model_sysmodel(boot_cpu_data.pdc.sys_model_name) == PDC_OK) - printk(KERN_INFO "model %s\n", + if (pdc_model_sysmodel(OS_ID_HPUX, boot_cpu_data.pdc.sys_model_name) == PDC_OK) + pr_info("HP-UX model name: %s\n", boot_cpu_data.pdc.sys_model_name); + serial_no[0] = 0; + if (pdc_model_sysmodel(OS_ID_MPEXL, serial_no) == PDC_OK && + serial_no[0]) + pr_info("MPE/iX model name: %s\n", serial_no); + dump_stack_set_arch_desc("%s", boot_cpu_data.pdc.sys_model_name); boot_cpu_data.hversion = boot_cpu_data.pdc.model.hversion; @@ -318,7 +324,7 @@ void __init collect_boot_cpu_data(void) * * o Enable CPU profiling hooks. */ -int __init init_per_cpu(int cpunum) +int init_per_cpu(int cpunum) { int ret; struct pdc_coproc_cfg coproc_cfg; @@ -326,8 +332,6 @@ int __init init_per_cpu(int cpunum) set_firmware_width(); ret = pdc_coproc_cfg(&coproc_cfg); - store_cpu_topology(cpunum); - if(ret >= 0 && coproc_cfg.ccr_functional) { mtctl(coproc_cfg.ccr_functional, 10); /* 10 == Coprocessor Control Reg */ @@ -363,6 +367,8 @@ int __init init_per_cpu(int cpunum) /* FUTURE: Enable Performance Monitor : ccr bit 0x20 */ init_percpu_prof(cpunum); + btlb_init_per_cpu(); + return ret; } @@ -373,10 +379,18 @@ int show_cpuinfo (struct seq_file *m, void *v) { unsigned long cpu; + char cpu_name[60], *p; + + /* strip PA path from CPU name to not confuse lscpu */ + strscpy(cpu_name, per_cpu(cpu_data, 0).dev->name, sizeof(cpu_name)); + p = strrchr(cpu_name, '['); + if (p) + *(--p) = 0; for_each_online_cpu(cpu) { - const struct cpuinfo_parisc *cpuinfo = &per_cpu(cpu_data, cpu); #ifdef CONFIG_SMP + const struct cpuinfo_parisc *cpuinfo = &per_cpu(cpu_data, cpu); + if (0 == cpuinfo->hpa) continue; #endif @@ -391,7 +405,7 @@ show_cpuinfo (struct seq_file *m, void *v) boot_cpu_data.cpu_hz / 1000000, boot_cpu_data.cpu_hz % 1000000 ); -#ifdef CONFIG_PARISC_CPU_TOPOLOGY +#ifdef CONFIG_GENERIC_ARCH_TOPOLOGY seq_printf(m, "physical id\t: %d\n", topology_physical_package_id(cpu)); seq_printf(m, "siblings\t: %d\n", @@ -419,11 +433,9 @@ show_cpuinfo (struct seq_file *m, void *v) } seq_printf(m, " (0x%02lx)\n", boot_cpu_data.pdc.capabilities); - seq_printf(m, "model\t\t: %s\n" - "model name\t: %s\n", + seq_printf(m, "model\t\t: %s - %s\n", boot_cpu_data.pdc.sys_model_name, - cpuinfo->dev ? - cpuinfo->dev->name : "Unknown"); + cpu_name); seq_printf(m, "hversion\t: 0x%08x\n" "sversion\t: 0x%08x\n", @@ -434,8 +446,8 @@ show_cpuinfo (struct seq_file *m, void *v) show_cache_info(m); seq_printf(m, "bogomips\t: %lu.%02lu\n", - cpuinfo->loops_per_jiffy / (500000 / HZ), - (cpuinfo->loops_per_jiffy / (5000 / HZ)) % 100); + loops_per_jiffy / (500000 / HZ), + loops_per_jiffy / (5000 / HZ) % 100); seq_printf(m, "software id\t: %ld\n\n", boot_cpu_data.pdc.model.sw_id); @@ -461,5 +473,6 @@ static struct parisc_driver cpu_driver __refdata = { */ void __init processor_init(void) { + reset_cpu_topology(); register_parisc_driver(&cpu_driver); } diff --git a/arch/parisc/kernel/ptrace.c b/arch/parisc/kernel/ptrace.c index f8c07dcbfb49..ceb45f51d52e 100644 --- a/arch/parisc/kernel/ptrace.c +++ b/arch/parisc/kernel/ptrace.c @@ -15,7 +15,6 @@ #include <linux/elf.h> #include <linux/errno.h> #include <linux/ptrace.h> -#include <linux/tracehook.h> #include <linux/user.h> #include <linux/personality.h> #include <linux/regset.h> @@ -26,7 +25,6 @@ #include <linux/audit.h> #include <linux/uaccess.h> -#include <asm/pgtable.h> #include <asm/processor.h> #include <asm/asm-offsets.h> @@ -128,6 +126,12 @@ long arch_ptrace(struct task_struct *child, long request, unsigned long tmp; long ret = -EIO; + unsigned long user_regs_struct_size = sizeof(struct user_regs_struct); +#ifdef CONFIG_64BIT + if (is_compat_task()) + user_regs_struct_size /= 2; +#endif + switch (request) { /* Read the word at location addr in the USER area. For ptraced @@ -168,7 +172,7 @@ long arch_ptrace(struct task_struct *child, long request, addr >= sizeof(struct pt_regs)) break; if (addr == PT_IAOQ0 || addr == PT_IAOQ1) { - data |= 3; /* ensure userspace privilege */ + data |= PRIV_USER; /* ensure userspace privilege */ } if ((addr >= PT_GR1 && addr <= PT_GR31) || addr == PT_IAOQ0 || addr == PT_IAOQ1 || @@ -183,14 +187,14 @@ long arch_ptrace(struct task_struct *child, long request, return copy_regset_to_user(child, task_user_regset_view(current), REGSET_GENERAL, - 0, sizeof(struct user_regs_struct), + 0, user_regs_struct_size, datap); case PTRACE_SETREGS: /* Set all gp regs in the child. */ return copy_regset_from_user(child, task_user_regset_view(current), REGSET_GENERAL, - 0, sizeof(struct user_regs_struct), + 0, user_regs_struct_size, datap); case PTRACE_GETFPREGS: /* Get the child FPU state. */ @@ -287,7 +291,7 @@ long compat_arch_ptrace(struct task_struct *child, compat_long_t request, if (addr >= sizeof(struct pt_regs)) break; if (addr == PT_IAOQ0+4 || addr == PT_IAOQ1+4) { - data |= 3; /* ensure userspace privilege */ + data |= PRIV_USER; /* ensure userspace privilege */ } if (addr >= PT_FR0 && addr <= PT_FR31 + 4) { /* Special case, fp regs are 64 bits anyway */ @@ -304,6 +308,11 @@ long compat_arch_ptrace(struct task_struct *child, compat_long_t request, } } break; + case PTRACE_GETREGS: + case PTRACE_SETREGS: + case PTRACE_GETFPREGS: + case PTRACE_SETFPREGS: + return arch_ptrace(child, request, addr, data); default: ret = compat_ptrace_request(child, request, addr, data); @@ -317,7 +326,7 @@ long compat_arch_ptrace(struct task_struct *child, compat_long_t request, long do_syscall_trace_enter(struct pt_regs *regs) { if (test_thread_flag(TIF_SYSCALL_TRACE)) { - int rc = tracehook_report_syscall_entry(regs); + int rc = ptrace_report_syscall_entry(regs); /* * As tracesys_next does not set %r28 to -ENOSYS @@ -328,7 +337,7 @@ long do_syscall_trace_enter(struct pt_regs *regs) if (rc) { /* * A nonzero return code from - * tracehook_report_syscall_entry() tells us + * ptrace_report_syscall_entry() tells us * to prevent the syscall execution. Skip * the syscall call and the syscall restart handling. * @@ -382,7 +391,7 @@ void do_syscall_trace_exit(struct pt_regs *regs) #endif if (stepping || test_thread_flag(TIF_SYSCALL_TRACE)) - tracehook_report_syscall_exit(regs, stepping); + ptrace_report_syscall_exit(regs, stepping); } @@ -392,31 +401,11 @@ void do_syscall_trace_exit(struct pt_regs *regs) static int fpr_get(struct task_struct *target, const struct user_regset *regset, - unsigned int pos, unsigned int count, - void *kbuf, void __user *ubuf) + struct membuf to) { struct pt_regs *regs = task_regs(target); - __u64 *k = kbuf; - __u64 __user *u = ubuf; - __u64 reg; - pos /= sizeof(reg); - count /= sizeof(reg); - - if (kbuf) - for (; count > 0 && pos < ELF_NFPREG; --count) - *k++ = regs->fr[pos++]; - else - for (; count > 0 && pos < ELF_NFPREG; --count) - if (__put_user(regs->fr[pos++], u++)) - return -EFAULT; - - kbuf = k; - ubuf = u; - pos *= sizeof(reg); - count *= sizeof(reg); - return user_regset_copyout_zero(&pos, &count, &kbuf, &ubuf, - ELF_NFPREG * sizeof(reg), -1); + return membuf_write(&to, regs->fr, ELF_NFPREG * sizeof(__u64)); } static int fpr_set(struct task_struct *target, @@ -446,8 +435,9 @@ static int fpr_set(struct task_struct *target, ubuf = u; pos *= sizeof(reg); count *= sizeof(reg); - return user_regset_copyin_ignore(&pos, &count, &kbuf, &ubuf, - ELF_NFPREG * sizeof(reg), -1); + user_regset_copyin_ignore(&pos, &count, &kbuf, &ubuf, + ELF_NFPREG * sizeof(reg), -1); + return 0; } #define RI(reg) (offsetof(struct user_regs_struct,reg) / sizeof(long)) @@ -505,7 +495,7 @@ static void set_reg(struct pt_regs *regs, int num, unsigned long val) case RI(iaoq[0]): case RI(iaoq[1]): /* set 2 lowest bits to ensure userspace privilege: */ - regs->iaoq[num - RI(iaoq[0])] = val | 3; + regs->iaoq[num - RI(iaoq[0])] = val | PRIV_USER; return; case RI(sar): regs->sar = val; return; @@ -528,30 +518,14 @@ static void set_reg(struct pt_regs *regs, int num, unsigned long val) static int gpr_get(struct task_struct *target, const struct user_regset *regset, - unsigned int pos, unsigned int count, - void *kbuf, void __user *ubuf) + struct membuf to) { struct pt_regs *regs = task_regs(target); - unsigned long *k = kbuf; - unsigned long __user *u = ubuf; - unsigned long reg; - - pos /= sizeof(reg); - count /= sizeof(reg); + unsigned int pos; - if (kbuf) - for (; count > 0 && pos < ELF_NGREG; --count) - *k++ = get_reg(regs, pos++); - else - for (; count > 0 && pos < ELF_NGREG; --count) - if (__put_user(get_reg(regs, pos++), u++)) - return -EFAULT; - kbuf = k; - ubuf = u; - pos *= sizeof(reg); - count *= sizeof(reg); - return user_regset_copyout_zero(&pos, &count, &kbuf, &ubuf, - ELF_NGREG * sizeof(reg), -1); + for (pos = 0; pos < ELF_NGREG; pos++) + membuf_store(&to, get_reg(regs, pos)); + return 0; } static int gpr_set(struct task_struct *target, @@ -581,20 +555,21 @@ static int gpr_set(struct task_struct *target, ubuf = u; pos *= sizeof(reg); count *= sizeof(reg); - return user_regset_copyin_ignore(&pos, &count, &kbuf, &ubuf, - ELF_NGREG * sizeof(reg), -1); + user_regset_copyin_ignore(&pos, &count, &kbuf, &ubuf, + ELF_NGREG * sizeof(reg), -1); + return 0; } static const struct user_regset native_regsets[] = { [REGSET_GENERAL] = { .core_note_type = NT_PRSTATUS, .n = ELF_NGREG, .size = sizeof(long), .align = sizeof(long), - .get = gpr_get, .set = gpr_set + .regset_get = gpr_get, .set = gpr_set }, [REGSET_FP] = { .core_note_type = NT_PRFPREG, .n = ELF_NFPREG, .size = sizeof(__u64), .align = sizeof(__u64), - .get = fpr_get, .set = fpr_set + .regset_get = fpr_get, .set = fpr_set } }; @@ -604,35 +579,17 @@ static const struct user_regset_view user_parisc_native_view = { }; #ifdef CONFIG_64BIT -#include <linux/compat.h> - static int gpr32_get(struct task_struct *target, const struct user_regset *regset, - unsigned int pos, unsigned int count, - void *kbuf, void __user *ubuf) + struct membuf to) { struct pt_regs *regs = task_regs(target); - compat_ulong_t *k = kbuf; - compat_ulong_t __user *u = ubuf; - compat_ulong_t reg; + unsigned int pos; - pos /= sizeof(reg); - count /= sizeof(reg); - - if (kbuf) - for (; count > 0 && pos < ELF_NGREG; --count) - *k++ = get_reg(regs, pos++); - else - for (; count > 0 && pos < ELF_NGREG; --count) - if (__put_user((compat_ulong_t) get_reg(regs, pos++), u++)) - return -EFAULT; + for (pos = 0; pos < ELF_NGREG; pos++) + membuf_store(&to, (compat_ulong_t)get_reg(regs, pos)); - kbuf = k; - ubuf = u; - pos *= sizeof(reg); - count *= sizeof(reg); - return user_regset_copyout_zero(&pos, &count, &kbuf, &ubuf, - ELF_NGREG * sizeof(reg), -1); + return 0; } static int gpr32_set(struct task_struct *target, @@ -662,8 +619,9 @@ static int gpr32_set(struct task_struct *target, ubuf = u; pos *= sizeof(reg); count *= sizeof(reg); - return user_regset_copyin_ignore(&pos, &count, &kbuf, &ubuf, - ELF_NGREG * sizeof(reg), -1); + user_regset_copyin_ignore(&pos, &count, &kbuf, &ubuf, + ELF_NGREG * sizeof(reg), -1); + return 0; } /* @@ -673,12 +631,12 @@ static const struct user_regset compat_regsets[] = { [REGSET_GENERAL] = { .core_note_type = NT_PRSTATUS, .n = ELF_NGREG, .size = sizeof(compat_long_t), .align = sizeof(compat_long_t), - .get = gpr32_get, .set = gpr32_set + .regset_get = gpr32_get, .set = gpr32_set }, [REGSET_FP] = { .core_note_type = NT_PRFPREG, .n = ELF_NFPREG, .size = sizeof(__u64), .align = sizeof(__u64), - .get = fpr_get, .set = fpr_set + .regset_get = fpr_get, .set = fpr_set } }; diff --git a/arch/parisc/kernel/real2.S b/arch/parisc/kernel/real2.S index 2b16d8d6598f..509d18b8e0e6 100644 --- a/arch/parisc/kernel/real2.S +++ b/arch/parisc/kernel/real2.S @@ -15,28 +15,15 @@ #include <linux/linkage.h> - - .section .bss - - .export pdc_result - .export pdc_result2 - .align 8 -pdc_result: - .block ASM_PDC_RESULT_SIZE -pdc_result2: - .block ASM_PDC_RESULT_SIZE - .export real_stack - .export real32_stack .export real64_stack - .align 64 + __PAGE_ALIGNED_BSS real_stack: -real32_stack: real64_stack: .block 8192 #define N_SAVED_REGS 9 - + .section .bss save_cr_space: .block REG_SZ * N_SAVED_REGS save_cr_end: @@ -248,9 +235,6 @@ ENTRY_CFI(real64_call_asm) /* save fn */ copy %arg2, %r31 - /* set up the new ap */ - ldo 64(%arg1), %r29 - /* load up the arg registers from the saved arg area */ /* 32-bit calling convention passes first 4 args in registers */ ldd 0*REG_SZ(%arg1), %arg0 /* note overwriting arg0 */ @@ -262,7 +246,9 @@ ENTRY_CFI(real64_call_asm) ldd 7*REG_SZ(%arg1), %r19 ldd 1*REG_SZ(%arg1), %arg1 /* do this one last! */ + /* set up real-mode stack and real-mode ap */ tophys_r1 %sp + ldo -16(%sp), %r29 /* Reference param save area */ b,l rfi_virt2real,%r2 nop diff --git a/arch/parisc/kernel/setup.c b/arch/parisc/kernel/setup.c index 53a21ce927de..ace483b6f19a 100644 --- a/arch/parisc/kernel/setup.c +++ b/arch/parisc/kernel/setup.c @@ -31,7 +31,6 @@ #include <asm/sections.h> #include <asm/pdc.h> #include <asm/led.h> -#include <asm/machdep.h> /* for pa7300lc_init() proto */ #include <asm/pdc_chassis.h> #include <asm/io.h> #include <asm/setup.h> @@ -40,40 +39,48 @@ static char __initdata command_line[COMMAND_LINE_SIZE]; -/* Intended for ccio/sba/cpu statistics under /proc/bus/{runway|gsc} */ -struct proc_dir_entry * proc_runway_root __read_mostly = NULL; -struct proc_dir_entry * proc_gsc_root __read_mostly = NULL; -struct proc_dir_entry * proc_mckinley_root __read_mostly = NULL; - -void __init setup_cmdline(char **cmdline_p) +static void __init setup_cmdline(char **cmdline_p) { extern unsigned int boot_args[]; + char *p; - /* Collect stuff passed in from the boot loader */ + *cmdline_p = command_line; /* boot_args[0] is free-mem start, boot_args[1] is ptr to command line */ - if (boot_args[0] < 64) { - /* called from hpux boot loader */ - boot_command_line[0] = '\0'; - } else { - strlcpy(boot_command_line, (char *)__va(boot_args[1]), - COMMAND_LINE_SIZE); + if (boot_args[0] < 64) + return; /* return if called from hpux boot loader */ + + /* Collect stuff passed in from the boot loader */ + strscpy(boot_command_line, (char *)__va(boot_args[1]), + COMMAND_LINE_SIZE); + + /* autodetect console type (if not done by palo yet) */ + p = boot_command_line; + if (!str_has_prefix(p, "console=") && !strstr(p, " console=")) { + strlcat(p, " console=", COMMAND_LINE_SIZE); + if (PAGE0->mem_cons.cl_class == CL_DUPLEX) + strlcat(p, "ttyS0", COMMAND_LINE_SIZE); + else + strlcat(p, "tty0", COMMAND_LINE_SIZE); + } + + /* default to use early console */ + if (!strstr(p, "earlycon")) + strlcat(p, " earlycon=pdc", COMMAND_LINE_SIZE); #ifdef CONFIG_BLK_DEV_INITRD - if (boot_args[2] != 0) /* did palo pass us a ramdisk? */ - { - initrd_start = (unsigned long)__va(boot_args[2]); - initrd_end = (unsigned long)__va(boot_args[3]); - } -#endif + /* did palo pass us a ramdisk? */ + if (boot_args[2] != 0) { + initrd_start = (unsigned long)__va(boot_args[2]); + initrd_end = (unsigned long)__va(boot_args[3]); } +#endif - strcpy(command_line, boot_command_line); - *cmdline_p = command_line; + strscpy(command_line, boot_command_line, COMMAND_LINE_SIZE); } #ifdef CONFIG_PA11 -void __init dma_ops_init(void) +static void __init dma_ops_init(void) { switch (boot_cpu_data.cpu_type) { case pcx: @@ -85,21 +92,14 @@ void __init dma_ops_init(void) "the PA-RISC 1.1 or 2.0 architecture specification.\n"); case pcxl2: - pa7300lc_init(); - break; default: break; } } #endif -extern void collect_boot_cpu_data(void); - void __init setup_arch(char **cmdline_p) { -#ifdef CONFIG_64BIT - extern int parisc_narrow_firmware; -#endif unwind_init(); init_per_cpu(smp_processor_id()); /* Set Modes & Enable FP */ @@ -128,8 +128,6 @@ void __init setup_arch(char **cmdline_p) if (__pa((unsigned long) &_end) >= KERNEL_INITIAL_SIZE) panic("KERNEL_INITIAL_ORDER too small!"); - pdc_console_init(); - #ifdef CONFIG_64BIT if(parisc_narrow_firmware) { printk(KERN_INFO "Kernel is using PDC in 32-bit mode.\n"); @@ -142,28 +140,16 @@ void __init setup_arch(char **cmdline_p) parisc_cache_init(); paging_init(); -#ifdef CONFIG_CHASSIS_LCD_LED - /* initialize the LCD/LED after boot_cpu_data is available ! */ - led_init(); /* LCD/LED initialization */ -#endif - #ifdef CONFIG_PA11 dma_ops_init(); #endif -#if defined(CONFIG_VT) && defined(CONFIG_DUMMY_CONSOLE) - conswitchp = &dummy_con; /* we use do_take_over_console() later ! */ -#endif - clear_sched_clock_stable(); } /* * Display CPU info for all CPUs. - * for parisc this is in processor.c */ -extern int show_cpuinfo (struct seq_file *m, void *v); - static void * c_start (struct seq_file *m, loff_t *pos) { @@ -194,48 +180,6 @@ const struct seq_operations cpuinfo_op = { .show = show_cpuinfo }; -static void __init parisc_proc_mkdir(void) -{ - /* - ** Can't call proc_mkdir() until after proc_root_init() has been - ** called by start_kernel(). In other words, this code can't - ** live in arch/.../setup.c because start_parisc() calls - ** start_kernel(). - */ - switch (boot_cpu_data.cpu_type) { - case pcxl: - case pcxl2: - if (NULL == proc_gsc_root) - { - proc_gsc_root = proc_mkdir("bus/gsc", NULL); - } - break; - case pcxt_: - case pcxu: - case pcxu_: - case pcxw: - case pcxw_: - case pcxw2: - if (NULL == proc_runway_root) - { - proc_runway_root = proc_mkdir("bus/runway", NULL); - } - break; - case mako: - case mako2: - if (NULL == proc_mckinley_root) - { - proc_mckinley_root = proc_mkdir("bus/mckinley", NULL); - } - break; - default: - /* FIXME: this was added to prevent the compiler - * complaining about missing pcx, pcxs and pcxt - * I'm assuming they have neither gsc nor runway */ - break; - } -} - static struct resource central_bus = { .name = "Central Bus", .start = F_EXTEND(0xfff80000), @@ -272,7 +216,7 @@ static int __init parisc_init_resources(void) result = request_resource(&iomem_resource, &local_broadcast); if (result < 0) { printk(KERN_ERR - "%s: failed to claim %saddress space!\n", + "%s: failed to claim %s address space!\n", __FILE__, local_broadcast.name); return result; } @@ -288,21 +232,10 @@ static int __init parisc_init_resources(void) return 0; } -extern void gsc_init(void); -extern void processor_init(void); -extern void ccio_init(void); -extern void hppb_init(void); -extern void dino_init(void); -extern void iosapic_init(void); -extern void lba_init(void); -extern void sba_init(void); -extern void eisa_init(void); - static int __init parisc_init(void) { u32 osid = (OS_ID_LINUX << 16); - parisc_proc_mkdir(); parisc_init_resources(); do_device_inventory(); /* probe for hardware */ @@ -337,55 +270,12 @@ static int __init parisc_init(void) apply_alternatives_all(); parisc_setup_cache_timing(); - - /* These are in a non-obvious order, will fix when we have an iotree */ -#if defined(CONFIG_IOSAPIC) - iosapic_init(); -#endif -#if defined(CONFIG_IOMMU_SBA) - sba_init(); -#endif -#if defined(CONFIG_PCI_LBA) - lba_init(); -#endif - - /* CCIO before any potential subdevices */ -#if defined(CONFIG_IOMMU_CCIO) - ccio_init(); -#endif - - /* - * Need to register Asp & Wax before the EISA adapters for the IRQ - * regions. EISA must come before PCI to be sure it gets IRQ region - * 0. - */ -#if defined(CONFIG_GSC_LASI) || defined(CONFIG_GSC_WAX) - gsc_init(); -#endif -#ifdef CONFIG_EISA - eisa_init(); -#endif - -#if defined(CONFIG_HPPB) - hppb_init(); -#endif - -#if defined(CONFIG_GSC_DINO) - dino_init(); -#endif - -#ifdef CONFIG_CHASSIS_LCD_LED - register_led_regions(); /* register LED port info in procfs */ -#endif - return 0; } arch_initcall(parisc_init); void __init start_parisc(void) { - extern void early_trap_init(void); - int ret, cpunum; struct pdc_coproc_cfg coproc_cfg; diff --git a/arch/parisc/kernel/signal.c b/arch/parisc/kernel/signal.c index 02895a8f2c55..e8d27def6c52 100644 --- a/arch/parisc/kernel/signal.c +++ b/arch/parisc/kernel/signal.c @@ -1,16 +1,13 @@ // SPDX-License-Identifier: GPL-2.0 /* - * linux/arch/parisc/kernel/signal.c: Architecture-specific signal - * handling support. + * PA-RISC architecture-specific signal handling support. * * Copyright (C) 2000 David Huggins-Daines <dhd@debian.org> * Copyright (C) 2000 Linuxcare, Inc. + * Copyright (C) 2000-2022 Helge Deller <deller@gmx.de> + * Copyright (C) 2022 John David Anglin <dave.anglin@bell.net> * * Based on the ia64, i386, and alpha versions. - * - * Like the IA-64, we are a recent enough port (we are *starting* - * with glibc2.2) that we do not need to support the old non-realtime - * Linux signals. Therefore we don't. */ #include <linux/sched.h> @@ -22,7 +19,7 @@ #include <linux/errno.h> #include <linux/wait.h> #include <linux/ptrace.h> -#include <linux/tracehook.h> +#include <linux/resume_user_mode.h> #include <linux/unistd.h> #include <linux/stddef.h> #include <linux/compat.h> @@ -30,9 +27,9 @@ #include <asm/ucontext.h> #include <asm/rt_sigframe.h> #include <linux/uaccess.h> -#include <asm/pgalloc.h> #include <asm/cacheflush.h> #include <asm/asm-offsets.h> +#include <asm/vdso.h> #ifdef CONFIG_COMPAT #include "signal32.h" @@ -60,14 +57,6 @@ * Do a signal return - restore sigcontext. */ -/* Trampoline for calling rt_sigreturn() */ -#define INSN_LDI_R25_0 0x34190000 /* ldi 0,%r25 (in_syscall=0) */ -#define INSN_LDI_R25_1 0x34190002 /* ldi 1,%r25 (in_syscall=1) */ -#define INSN_LDI_R20 0x3414015a /* ldi __NR_rt_sigreturn,%r20 */ -#define INSN_BLE_SR2_R0 0xe4008200 /* be,l 0x100(%sr2,%r0),%sr0,%r31 */ -/* For debugging */ -#define INSN_DIE_HORRIBLY 0x68000ccc /* stw %r0,0x666(%sr0,%r0) */ - static long restore_sigcontext(struct sigcontext __user *sc, struct pt_regs *regs) { @@ -78,13 +67,13 @@ restore_sigcontext(struct sigcontext __user *sc, struct pt_regs *regs) err |= __copy_from_user(regs->iaoq, sc->sc_iaoq, sizeof(regs->iaoq)); err |= __copy_from_user(regs->iasq, sc->sc_iasq, sizeof(regs->iasq)); err |= __get_user(regs->sar, &sc->sc_sar); - DBG(2,"restore_sigcontext: iaoq is %#lx / %#lx\n", - regs->iaoq[0],regs->iaoq[1]); - DBG(2,"restore_sigcontext: r28 is %ld\n", regs->gr[28]); + DBG(2, "%s: iaoq is %#lx / %#lx\n", + __func__, regs->iaoq[0], regs->iaoq[1]); + DBG(2, "%s: r28 is %ld\n", __func__, regs->gr[28]); return err; } -void +asmlinkage void sys_rt_sigreturn(struct pt_regs *regs, int in_syscall) { struct rt_sigframe __user *frame; @@ -103,7 +92,7 @@ sys_rt_sigreturn(struct pt_regs *regs, int in_syscall) /* Unwind the user stack to get the rt_sigframe structure. */ frame = (struct rt_sigframe __user *) (usp - sigframe_size); - DBG(2,"sys_rt_sigreturn: frame is %p\n", frame); + DBG(2, "%s: frame is %p pid %d\n", __func__, frame, task_pid_nr(current)); regs->orig_r28 = 1; /* no restarts for sigreturn */ @@ -111,7 +100,6 @@ sys_rt_sigreturn(struct pt_regs *regs, int in_syscall) compat_frame = (struct compat_rt_sigframe __user *)frame; if (is_compat_task()) { - DBG(2,"sys_rt_sigreturn: ELF32 process.\n"); if (get_compat_sigset(&set, &compat_frame->uc.uc_sigmask)) goto give_sigsegv; } else @@ -126,25 +114,25 @@ sys_rt_sigreturn(struct pt_regs *regs, int in_syscall) /* Good thing we saved the old gr[30], eh? */ #ifdef CONFIG_64BIT if (is_compat_task()) { - DBG(1,"sys_rt_sigreturn: compat_frame->uc.uc_mcontext 0x%p\n", - &compat_frame->uc.uc_mcontext); + DBG(1, "%s: compat_frame->uc.uc_mcontext 0x%p\n", + __func__, &compat_frame->uc.uc_mcontext); // FIXME: Load upper half from register file if (restore_sigcontext32(&compat_frame->uc.uc_mcontext, &compat_frame->regs, regs)) goto give_sigsegv; - DBG(1,"sys_rt_sigreturn: usp %#08lx stack 0x%p\n", - usp, &compat_frame->uc.uc_stack); + DBG(1, "%s: usp %#08lx stack 0x%p\n", + __func__, usp, &compat_frame->uc.uc_stack); if (compat_restore_altstack(&compat_frame->uc.uc_stack)) goto give_sigsegv; } else #endif { - DBG(1,"sys_rt_sigreturn: frame->uc.uc_mcontext 0x%p\n", - &frame->uc.uc_mcontext); + DBG(1, "%s: frame->uc.uc_mcontext 0x%p\n", + __func__, &frame->uc.uc_mcontext); if (restore_sigcontext(&frame->uc.uc_mcontext, regs)) goto give_sigsegv; - DBG(1,"sys_rt_sigreturn: usp %#08lx stack 0x%p\n", - usp, &frame->uc.uc_stack); + DBG(1, "%s: usp %#08lx stack 0x%p\n", + __func__, usp, &frame->uc.uc_stack); if (restore_altstack(&frame->uc.uc_stack)) goto give_sigsegv; } @@ -156,14 +144,11 @@ sys_rt_sigreturn(struct pt_regs *regs, int in_syscall) */ if (in_syscall) regs->gr[31] = regs->iaoq[0]; -#if DEBUG_SIG - DBG(1,"sys_rt_sigreturn: returning to %#lx, DUMPING REGS:\n", regs->iaoq[0]); - show_regs(regs); -#endif + return; give_sigsegv: - DBG(1,"sys_rt_sigreturn: Sending SIGSEGV\n"); + DBG(1, "%s: Sending SIGSEGV\n", __func__); force_sig(SIGSEGV); return; } @@ -178,20 +163,20 @@ get_sigframe(struct k_sigaction *ka, unsigned long sp, size_t frame_size) /*FIXME: ELF32 vs. ELF64 has different frame_size, but since we don't use the parameter it doesn't matter */ - DBG(1,"get_sigframe: ka = %#lx, sp = %#lx, frame_size = %#lx\n", - (unsigned long)ka, sp, frame_size); + DBG(1, "%s: ka = %#lx, sp = %#lx, frame_size = %zu\n", + __func__, (unsigned long)ka, sp, frame_size); /* Align alternate stack and reserve 64 bytes for the signal handler's frame marker. */ if ((ka->sa.sa_flags & SA_ONSTACK) != 0 && ! sas_ss_flags(sp)) sp = (current->sas_ss_sp + 0x7f) & ~0x3f; /* Stacks grow up! */ - DBG(1,"get_sigframe: Returning sp = %#lx\n", (unsigned long)sp); + DBG(1, "%s: Returning sp = %#lx\n", __func__, (unsigned long)sp); return (void __user *) sp; /* Stacks grow up. Fun. */ } static long -setup_sigcontext(struct sigcontext __user *sc, struct pt_regs *regs, int in_syscall) +setup_sigcontext(struct sigcontext __user *sc, struct pt_regs *regs, long in_syscall) { unsigned long flags = 0; @@ -206,55 +191,65 @@ setup_sigcontext(struct sigcontext __user *sc, struct pt_regs *regs, int in_sysc err |= __put_user(regs->gr[31]+4, &sc->sc_iaoq[1]); err |= __put_user(regs->sr[3], &sc->sc_iasq[0]); err |= __put_user(regs->sr[3], &sc->sc_iasq[1]); - DBG(1,"setup_sigcontext: iaoq %#lx / %#lx (in syscall)\n", - regs->gr[31], regs->gr[31]+4); + DBG(1, "%s: iaoq %#lx / %#lx (in syscall)\n", + __func__, regs->gr[31], regs->gr[31]+4); } else { err |= __copy_to_user(sc->sc_iaoq, regs->iaoq, sizeof(regs->iaoq)); err |= __copy_to_user(sc->sc_iasq, regs->iasq, sizeof(regs->iasq)); - DBG(1,"setup_sigcontext: iaoq %#lx / %#lx (not in syscall)\n", - regs->iaoq[0], regs->iaoq[1]); + DBG(1, "%s: iaoq %#lx / %#lx (not in syscall)\n", + __func__, regs->iaoq[0], regs->iaoq[1]); } err |= __put_user(flags, &sc->sc_flags); err |= __copy_to_user(sc->sc_gr, regs->gr, sizeof(regs->gr)); err |= __copy_to_user(sc->sc_fr, regs->fr, sizeof(regs->fr)); err |= __put_user(regs->sar, &sc->sc_sar); - DBG(1,"setup_sigcontext: r28 is %ld\n", regs->gr[28]); + DBG(1, "%s: r28 is %ld\n", __func__, regs->gr[28]); return err; } static long setup_rt_frame(struct ksignal *ksig, sigset_t *set, struct pt_regs *regs, - int in_syscall) + long in_syscall) { struct rt_sigframe __user *frame; unsigned long rp, usp; unsigned long haddr, sigframe_size; - unsigned long start, end; + unsigned long start; int err = 0; #ifdef CONFIG_64BIT struct compat_rt_sigframe __user * compat_frame; #endif usp = (regs->gr[30] & ~(0x01UL)); - /*FIXME: frame_size parameter is unused, remove it. */ - frame = get_sigframe(&ksig->ka, usp, sizeof(*frame)); + sigframe_size = PARISC_RT_SIGFRAME_SIZE; +#ifdef CONFIG_64BIT + if (is_compat_task()) { + /* The gcc alloca implementation leaves garbage in the upper 32 bits of sp */ + usp = (compat_uint_t)usp; + sigframe_size = PARISC_RT_SIGFRAME_SIZE32; + } +#endif + frame = get_sigframe(&ksig->ka, usp, sigframe_size); - DBG(1,"SETUP_RT_FRAME: START\n"); - DBG(1,"setup_rt_frame: frame %p info %p\n", frame, ksig->info); + DBG(1, "%s: frame %p info %p\n", __func__, frame, &ksig->info); + start = (unsigned long) frame; + if (start >= TASK_SIZE_MAX - sigframe_size) + return -EFAULT; #ifdef CONFIG_64BIT compat_frame = (struct compat_rt_sigframe __user *)frame; if (is_compat_task()) { - DBG(1,"setup_rt_frame: frame->info = 0x%p\n", &compat_frame->info); + DBG(1, "%s: frame->info = 0x%p\n", __func__, &compat_frame->info); err |= copy_siginfo_to_user32(&compat_frame->info, &ksig->info); err |= __compat_save_altstack( &compat_frame->uc.uc_stack, regs->gr[30]); - DBG(1,"setup_rt_frame: frame->uc = 0x%p\n", &compat_frame->uc); - DBG(1,"setup_rt_frame: frame->uc.uc_mcontext = 0x%p\n", &compat_frame->uc.uc_mcontext); + DBG(1, "%s: frame->uc = 0x%p\n", __func__, &compat_frame->uc); + DBG(1, "%s: frame->uc.uc_mcontext = 0x%p\n", + __func__, &compat_frame->uc.uc_mcontext); err |= setup_sigcontext32(&compat_frame->uc.uc_mcontext, &compat_frame->regs, regs, in_syscall); err |= put_compat_sigset(&compat_frame->uc.uc_sigmask, set, @@ -262,11 +257,12 @@ setup_rt_frame(struct ksignal *ksig, sigset_t *set, struct pt_regs *regs, } else #endif { - DBG(1,"setup_rt_frame: frame->info = 0x%p\n", &frame->info); + DBG(1, "%s: frame->info = 0x%p\n", __func__, &frame->info); err |= copy_siginfo_to_user(&frame->info, &ksig->info); err |= __save_altstack(&frame->uc.uc_stack, regs->gr[30]); - DBG(1,"setup_rt_frame: frame->uc = 0x%p\n", &frame->uc); - DBG(1,"setup_rt_frame: frame->uc.uc_mcontext = 0x%p\n", &frame->uc.uc_mcontext); + DBG(1, "%s: frame->uc = 0x%p\n", __func__, &frame->uc); + DBG(1, "%s: frame->uc.uc_mcontext = 0x%p\n", + __func__, &frame->uc.uc_mcontext); err |= setup_sigcontext(&frame->uc.uc_mcontext, regs, in_syscall); /* FIXME: Should probably be converted as well for the compat case */ err |= __copy_to_user(&frame->uc.uc_sigmask, set, sizeof(*set)); @@ -275,42 +271,15 @@ setup_rt_frame(struct ksignal *ksig, sigset_t *set, struct pt_regs *regs, if (err) return -EFAULT; - /* Set up to return from userspace. If provided, use a stub - already in userspace. The first words of tramp are used to - save the previous sigrestartblock trampoline that might be - on the stack. We start the sigreturn trampoline at - SIGRESTARTBLOCK_TRAMP+X. */ - err |= __put_user(in_syscall ? INSN_LDI_R25_1 : INSN_LDI_R25_0, - &frame->tramp[SIGRESTARTBLOCK_TRAMP+0]); - err |= __put_user(INSN_LDI_R20, - &frame->tramp[SIGRESTARTBLOCK_TRAMP+1]); - err |= __put_user(INSN_BLE_SR2_R0, - &frame->tramp[SIGRESTARTBLOCK_TRAMP+2]); - err |= __put_user(INSN_NOP, &frame->tramp[SIGRESTARTBLOCK_TRAMP+3]); - -#if DEBUG_SIG - /* Assert that we're flushing in the correct space... */ - { - unsigned long sid; - asm ("mfsp %%sr3,%0" : "=r" (sid)); - DBG(1,"setup_rt_frame: Flushing 64 bytes at space %#x offset %p\n", - sid, frame->tramp); - } +#ifdef CONFIG_64BIT + if (!is_compat_task()) + rp = VDSO64_SYMBOL(current, sigtramp_rt); + else #endif + rp = VDSO32_SYMBOL(current, sigtramp_rt); - start = (unsigned long) &frame->tramp[0]; - end = (unsigned long) &frame->tramp[TRAMP_SIZE]; - flush_user_dcache_range_asm(start, end); - flush_user_icache_range_asm(start, end); - - /* TRAMP Words 0-4, Length 5 = SIGRESTARTBLOCK_TRAMP - * TRAMP Words 5-9, Length 4 = SIGRETURN_TRAMP - * So the SIGRETURN_TRAMP is at the end of SIGRESTARTBLOCK_TRAMP - */ - rp = (unsigned long) &frame->tramp[SIGRESTARTBLOCK_TRAMP]; - - if (err) - return -EFAULT; + if (in_syscall) + rp += 4*4; /* skip 4 instructions and start at ldi 1,%r25 */ haddr = A(ksig->ka.sa.sa_handler); /* The sa_handler may be a pointer to a function descriptor */ @@ -341,23 +310,18 @@ setup_rt_frame(struct ksignal *ksig, sigset_t *set, struct pt_regs *regs, haddr = fdesc.addr; regs->gr[19] = fdesc.gp; - DBG(1,"setup_rt_frame: 64 bit signal, exe=%#lx, r19=%#lx, in_syscall=%d\n", - haddr, regs->gr[19], in_syscall); + DBG(1, "%s: 64 bit signal, exe=%#lx, r19=%#lx, in_syscall=%d\n", + __func__, haddr, regs->gr[19], in_syscall); } #endif /* The syscall return path will create IAOQ values from r31. */ - sigframe_size = PARISC_RT_SIGFRAME_SIZE; -#ifdef CONFIG_64BIT - if (is_compat_task()) - sigframe_size = PARISC_RT_SIGFRAME_SIZE32; -#endif if (in_syscall) { regs->gr[31] = haddr; #ifdef CONFIG_64BIT if (!test_thread_flag(TIF_32BIT)) - sigframe_size |= 1; + sigframe_size |= 1; /* XXX ???? */ #endif } else { unsigned long psw = USER_PSW; @@ -379,11 +343,11 @@ setup_rt_frame(struct ksignal *ksig, sigset_t *set, struct pt_regs *regs, } regs->gr[0] = psw; - regs->iaoq[0] = haddr | 3; + regs->iaoq[0] = haddr | PRIV_USER; regs->iaoq[1] = regs->iaoq[0] + 4; } - regs->gr[2] = rp; /* userland return pointer */ + regs->gr[2] = rp; /* userland return pointer */ regs->gr[26] = ksig->sig; /* signal number */ #ifdef CONFIG_64BIT @@ -397,15 +361,15 @@ setup_rt_frame(struct ksignal *ksig, sigset_t *set, struct pt_regs *regs, regs->gr[24] = A(&frame->uc); /* ucontext pointer */ } - DBG(1,"setup_rt_frame: making sigreturn frame: %#lx + %#lx = %#lx\n", + DBG(1, "%s: making sigreturn frame: %#lx + %#lx = %#lx\n", __func__, regs->gr[30], sigframe_size, regs->gr[30] + sigframe_size); /* Raise the user stack pointer to make a proper call frame. */ regs->gr[30] = (A(frame) + sigframe_size); - DBG(1,"setup_rt_frame: sig deliver (%s,%d) frame=0x%p sp=%#lx iaoq=%#lx/%#lx rp=%#lx\n", - current->comm, current->pid, frame, regs->gr[30], + DBG(1, "%s: sig deliver (%s,%d) frame=0x%p sp=%#lx iaoq=%#lx/%#lx rp=%#lx\n", + __func__, current->comm, current->pid, frame, regs->gr[30], regs->iaoq[0], regs->iaoq[1], rp); return 0; @@ -416,13 +380,13 @@ setup_rt_frame(struct ksignal *ksig, sigset_t *set, struct pt_regs *regs, */ static void -handle_signal(struct ksignal *ksig, struct pt_regs *regs, int in_syscall) +handle_signal(struct ksignal *ksig, struct pt_regs *regs, long in_syscall) { int ret; sigset_t *oldset = sigmask_to_save(); - DBG(1,"handle_signal: sig=%ld, ka=%p, info=%p, oldset=%p, regs=%p\n", - ksig->sig, ksig->ka, ksig->info, oldset, regs); + DBG(1, "%s: sig=%d, ka=%p, info=%p, oldset=%p, regs=%p\n", + __func__, ksig->sig, &ksig->ka, &ksig->info, oldset, regs); /* Set up the stack frame */ ret = setup_rt_frame(ksig, oldset, regs, in_syscall); @@ -430,8 +394,8 @@ handle_signal(struct ksignal *ksig, struct pt_regs *regs, int in_syscall) signal_setup_done(ret, ksig, test_thread_flag(TIF_SINGLESTEP) || test_thread_flag(TIF_BLOCKSTEP)); - DBG(1,KERN_DEBUG "do_signal: Exit (success), regs->gr[28] = %ld\n", - regs->gr[28]); + DBG(1, "%s: Exit (success), regs->gr[28] = %ld\n", + __func__, regs->gr[28]); } /* @@ -459,7 +423,7 @@ static void check_syscallno_in_delay_branch(struct pt_regs *regs) regs->gr[31] -= 8; /* delayed branching */ /* Get assembler opcode of code in delay branch */ - uaddr = (unsigned int *) ((regs->gr[31] & ~3) + 4); + uaddr = (u32 __user *) ((regs->gr[31] & ~3) + 4); err = get_user(opcode, uaddr); if (err) return; @@ -489,22 +453,27 @@ syscall_restart(struct pt_regs *regs, struct k_sigaction *ka) if (regs->orig_r28) return; regs->orig_r28 = 1; /* no more restarts */ + + DBG(1, "%s: orig_r28 = %ld pid %d r20 %ld\n", + __func__, regs->orig_r28, task_pid_nr(current), regs->gr[20]); + /* Check the return code */ switch (regs->gr[28]) { case -ERESTART_RESTARTBLOCK: case -ERESTARTNOHAND: - DBG(1,"ERESTARTNOHAND: returning -EINTR\n"); + DBG(1, "%s: ERESTARTNOHAND: returning -EINTR\n", __func__); regs->gr[28] = -EINTR; break; - case -ERESTARTSYS: if (!(ka->sa.sa_flags & SA_RESTART)) { - DBG(1,"ERESTARTSYS: putting -EINTR\n"); + DBG(1, "%s: ERESTARTSYS: putting -EINTR pid %d\n", + __func__, task_pid_nr(current)); regs->gr[28] = -EINTR; break; } - /* fallthrough */ + fallthrough; case -ERESTARTNOINTR: + DBG(1, "%s: %ld\n", __func__, regs->gr[28]); check_syscallno_in_delay_branch(regs); break; } @@ -516,46 +485,52 @@ insert_restart_trampoline(struct pt_regs *regs) if (regs->orig_r28) return; regs->orig_r28 = 1; /* no more restarts */ - switch(regs->gr[28]) { + + DBG(2, "%s: gr28 = %ld pid %d\n", + __func__, regs->gr[28], task_pid_nr(current)); + + switch (regs->gr[28]) { case -ERESTART_RESTARTBLOCK: { /* Restart the system call - no handlers present */ unsigned int *usp = (unsigned int *)regs->gr[30]; - unsigned long start = (unsigned long) &usp[2]; - unsigned long end = (unsigned long) &usp[5]; + unsigned long rp; long err = 0; - /* Setup a trampoline to restart the syscall - * with __NR_restart_syscall + /* check that we don't exceed the stack */ + if (A(&usp[0]) >= TASK_SIZE_MAX - 5 * sizeof(int)) + return; + + /* Call trampoline in vdso to restart the syscall + * with __NR_restart_syscall. + * Original return addresses are on stack like this: * * 0: <return address (orig r31)> * 4: <2nd half for 64-bit> - * 8: ldw 0(%sp), %r31 - * 12: be 0x100(%sr2, %r0) - * 16: ldi __NR_restart_syscall, %r20 */ #ifdef CONFIG_64BIT - err |= put_user(regs->gr[31] >> 32, &usp[0]); - err |= put_user(regs->gr[31] & 0xffffffff, &usp[1]); - err |= put_user(0x0fc010df, &usp[2]); -#else - err |= put_user(regs->gr[31], &usp[0]); - err |= put_user(0x0fc0109f, &usp[2]); + if (!is_compat_task()) { + err |= put_user(regs->gr[31] >> 32, &usp[0]); + err |= put_user(regs->gr[31] & 0xffffffff, &usp[1]); + rp = VDSO64_SYMBOL(current, restart_syscall); + } else #endif - err |= put_user(0xe0008200, &usp[3]); - err |= put_user(0x34140000, &usp[4]); - + { + err |= put_user(regs->gr[31], &usp[0]); + rp = VDSO32_SYMBOL(current, restart_syscall); + } WARN_ON(err); - /* flush data/instruction cache for new insns */ - flush_user_dcache_range_asm(start, end); - flush_user_icache_range_asm(start, end); - - regs->gr[31] = regs->gr[30] + 8; + regs->gr[31] = rp; + DBG(1, "%s: ERESTART_RESTARTBLOCK\n", __func__); return; } + case -EINTR: + /* ok, was handled before and should be returned. */ + break; case -ERESTARTNOHAND: case -ERESTARTSYS: case -ERESTARTNOINTR: + DBG(1, "%s: Type %ld\n", __func__, regs->gr[28]); check_syscallno_in_delay_branch(regs); return; default: @@ -564,51 +539,51 @@ insert_restart_trampoline(struct pt_regs *regs) } /* - * Note that 'init' is a special process: it doesn't get signals it doesn't - * want to handle. Thus you cannot kill init even with a SIGKILL even by - * mistake. - * * We need to be able to restore the syscall arguments (r21-r26) to * restart syscalls. Thus, the syscall path should save them in the * pt_regs structure (it's okay to do so since they are caller-save * registers). As noted below, the syscall number gets restored for * us due to the magic of delayed branching. */ -asmlinkage void -do_signal(struct pt_regs *regs, long in_syscall) +static void do_signal(struct pt_regs *regs, long in_syscall) { struct ksignal ksig; + int restart_syscall; + bool has_handler; + + has_handler = get_signal(&ksig); - DBG(1,"\ndo_signal: regs=0x%p, sr7 %#lx, in_syscall=%d\n", - regs, regs->sr[7], in_syscall); + restart_syscall = 0; + if (in_syscall) + restart_syscall = 1; - if (get_signal(&ksig)) { - DBG(3,"do_signal: signr = %d, regs->gr[28] = %ld\n", signr, regs->gr[28]); + if (has_handler) { /* Restart a system call if necessary. */ - if (in_syscall) + if (restart_syscall) syscall_restart(regs, &ksig.ka); handle_signal(&ksig, regs, in_syscall); + DBG(1, "%s: Handled signal pid %d\n", + __func__, task_pid_nr(current)); return; } - /* Did we come from a system call? */ - if (in_syscall) + /* Do we need to restart the system call? */ + if (restart_syscall) insert_restart_trampoline(regs); - DBG(1,"do_signal: Exit (not delivered), regs->gr[28] = %ld\n", - regs->gr[28]); + DBG(1, "%s: Exit (not delivered), regs->gr[28] = %ld orig_r28 = %ld pid %d\n", + __func__, regs->gr[28], regs->orig_r28, task_pid_nr(current)); restore_saved_sigmask(); } -void do_notify_resume(struct pt_regs *regs, long in_syscall) +asmlinkage void do_notify_resume(struct pt_regs *regs, long in_syscall) { - if (test_thread_flag(TIF_SIGPENDING)) + if (test_thread_flag(TIF_SIGPENDING) || + test_thread_flag(TIF_NOTIFY_SIGNAL)) do_signal(regs, in_syscall); - if (test_thread_flag(TIF_NOTIFY_RESUME)) { - clear_thread_flag(TIF_NOTIFY_RESUME); - tracehook_notify_resume(regs); - } + if (test_thread_flag(TIF_NOTIFY_RESUME)) + resume_user_mode_work(regs); } diff --git a/arch/parisc/kernel/signal32.h b/arch/parisc/kernel/signal32.h index f166250f2d06..c03eb1ed4c53 100644 --- a/arch/parisc/kernel/signal32.h +++ b/arch/parisc/kernel/signal32.h @@ -36,21 +36,12 @@ struct compat_regfile { compat_int_t rf_sar; }; -#define COMPAT_SIGRETURN_TRAMP 4 -#define COMPAT_SIGRESTARTBLOCK_TRAMP 5 -#define COMPAT_TRAMP_SIZE (COMPAT_SIGRETURN_TRAMP + \ - COMPAT_SIGRESTARTBLOCK_TRAMP) - struct compat_rt_sigframe { - /* XXX: Must match trampoline size in arch/parisc/kernel/signal.c - Secondary to that it must protect the ERESTART_RESTARTBLOCK - trampoline we left on the stack (we were bad and didn't - change sp so we could run really fast.) */ - compat_uint_t tramp[COMPAT_TRAMP_SIZE]; - compat_siginfo_t info; - struct compat_ucontext uc; - /* Hidden location of truncated registers, *must* be last. */ - struct compat_regfile regs; + unsigned int tramp[2]; /* holds original return address */ + compat_siginfo_t info; + struct compat_ucontext uc; + /* Hidden location of truncated registers, *must* be last. */ + struct compat_regfile regs; }; /* diff --git a/arch/parisc/kernel/smp.c b/arch/parisc/kernel/smp.c index e202c37e56af..444154271f23 100644 --- a/arch/parisc/kernel/smp.c +++ b/arch/parisc/kernel/smp.c @@ -29,6 +29,8 @@ #include <linux/bitops.h> #include <linux/ftrace.h> #include <linux/cpu.h> +#include <linux/kgdb.h> +#include <linux/sched/hotplug.h> #include <linux/atomic.h> #include <asm/current.h> @@ -39,8 +41,6 @@ #include <asm/irq.h> /* for CPU_IRQ_REGION and friends */ #include <asm/mmu_context.h> #include <asm/page.h> -#include <asm/pgtable.h> -#include <asm/pgalloc.h> #include <asm/processor.h> #include <asm/ptrace.h> #include <asm/unistd.h> @@ -61,8 +61,6 @@ volatile struct task_struct *smp_init_current_idle_task; /* track which CPU is booting */ static volatile int cpu_now_booting; -static int parisc_max_cpus = 1; - static DEFINE_PER_CPU(spinlock_t, ipi_lock); enum ipi_message_type { @@ -71,7 +69,10 @@ enum ipi_message_type { IPI_CALL_FUNC, IPI_CPU_START, IPI_CPU_STOP, - IPI_CPU_TEST + IPI_CPU_TEST, +#ifdef CONFIG_KGDB + IPI_ENTER_KGDB, +#endif }; @@ -169,15 +170,23 @@ ipi_interrupt(int irq, void *dev_id) case IPI_CPU_TEST: smp_debug(100, KERN_DEBUG "CPU%d is alive!\n", this_cpu); break; - +#ifdef CONFIG_KGDB + case IPI_ENTER_KGDB: + smp_debug(100, KERN_DEBUG "CPU%d ENTER_KGDB\n", this_cpu); + kgdb_nmicallback(raw_smp_processor_id(), get_irq_regs()); + break; +#endif default: printk(KERN_CRIT "Unknown IPI num on CPU%d: %lu\n", this_cpu, which); return IRQ_NONE; } /* Switch */ - /* let in any pending interrupts */ - local_irq_enable(); - local_irq_disable(); + + /* before doing more, let in any pending interrupts */ + if (ops) { + local_irq_enable(); + local_irq_disable(); + } } /* while (ops) */ } return IRQ_HANDLED; @@ -218,19 +227,27 @@ static inline void send_IPI_allbutself(enum ipi_message_type op) { int i; - + + preempt_disable(); for_each_online_cpu(i) { if (i != smp_processor_id()) send_IPI_single(i, op); } + preempt_enable(); } +#ifdef CONFIG_KGDB +void kgdb_roundup_cpus(void) +{ + send_IPI_allbutself(IPI_ENTER_KGDB); +} +#endif inline void smp_send_stop(void) { send_IPI_allbutself(IPI_CPU_STOP); } -void -smp_send_reschedule(int cpu) { send_IPI_single(cpu, IPI_RESCHEDULE); } +void +arch_smp_send_reschedule(int cpu) { send_IPI_single(cpu, IPI_RESCHEDULE); } void smp_send_all_nop(void) @@ -251,12 +268,9 @@ void arch_send_call_function_single_ipi(int cpu) /* * Called by secondaries to update state and initialize CPU registers. */ -static void __init +static void smp_cpu_init(int cpunum) { - extern void init_IRQ(void); /* arch/parisc/kernel/irq.c */ - extern void start_cpu_itimer(void); /* arch/parisc/kernel/time.c */ - /* Set modes and Enable floating point coprocessor */ init_per_cpu(cpunum); @@ -291,7 +305,7 @@ smp_cpu_init(int cpunum) * Slaves start using C here. Indirectly called from smp_slave_stext. * Do what start_kernel() and main() do for boot strap processor (aka monarch) */ -void __init smp_callin(unsigned long pdce_proc) +void smp_callin(unsigned long pdce_proc) { int slave_id = cpu_now_booting; @@ -301,7 +315,6 @@ void __init smp_callin(unsigned long pdce_proc) #endif smp_cpu_init(slave_id); - preempt_disable(); flush_cache_all_local(); /* start with known state */ flush_tlb_all_local(NULL); @@ -317,12 +330,27 @@ void __init smp_callin(unsigned long pdce_proc) /* * Bring one cpu online. */ -int smp_boot_one_cpu(int cpuid, struct task_struct *idle) +static int smp_boot_one_cpu(int cpuid, struct task_struct *idle) { const struct cpuinfo_parisc *p = &per_cpu(cpu_data, cpuid); long timeout; - task_thread_info(idle)->cpu = cpuid; +#ifdef CONFIG_HOTPLUG_CPU + int i; + + /* reset irq statistics for this CPU */ + memset(&per_cpu(irq_stat, cpuid), 0, sizeof(irq_cpustat_t)); + for (i = 0; i < NR_IRQS; i++) { + struct irq_desc *desc = irq_to_desc(i); + + if (desc && desc->kstat_irqs) + *per_cpu_ptr(desc->kstat_irqs, cpuid) = 0; + } +#endif + + /* wait until last booting CPU has started. */ + while (cpu_now_booting) + ; /* Let _start know what logical CPU we're booting ** (offset into init_tasks[],cpu_data[]) @@ -359,7 +387,6 @@ int smp_boot_one_cpu(int cpuid, struct task_struct *idle) if(cpu_online(cpuid)) { /* Which implies Slave has started up */ cpu_now_booting = 0; - smp_init_current_idle_task = NULL; goto alive ; } udelay(100); @@ -377,13 +404,7 @@ alive: void __init smp_prepare_boot_cpu(void) { - int bootstrap_processor = per_cpu(cpu_data, 0).cpuid; - - /* Setup BSP mappings */ - printk(KERN_INFO "SMP: bootstrap CPU ID is %d\n", bootstrap_processor); - - set_cpu_online(bootstrap_processor, true); - set_cpu_present(bootstrap_processor, true); + pr_info("SMP: bootstrap CPU ID is 0\n"); } @@ -400,30 +421,87 @@ void __init smp_prepare_cpus(unsigned int max_cpus) spin_lock_init(&per_cpu(ipi_lock, cpu)); init_cpu_present(cpumask_of(0)); - - parisc_max_cpus = max_cpus; - if (!max_cpus) - printk(KERN_INFO "SMP mode deactivated.\n"); } -void smp_cpus_done(unsigned int cpu_max) +void __init smp_cpus_done(unsigned int cpu_max) { - return; } int __cpu_up(unsigned int cpu, struct task_struct *tidle) { - if (cpu != 0 && cpu < parisc_max_cpus && smp_boot_one_cpu(cpu, tidle)) - return -ENOSYS; + if (cpu_online(cpu)) + return 0; + + if (num_online_cpus() < nr_cpu_ids && + num_online_cpus() < setup_max_cpus && + smp_boot_one_cpu(cpu, tidle)) + return -EIO; + + return cpu_online(cpu) ? 0 : -EIO; +} + +/* + * __cpu_disable runs on the processor to be shutdown. + */ +int __cpu_disable(void) +{ +#ifdef CONFIG_HOTPLUG_CPU + unsigned int cpu = smp_processor_id(); + + remove_cpu_topology(cpu); + + /* + * Take this CPU offline. Once we clear this, we can't return, + * and we must not schedule until we're ready to give up the cpu. + */ + set_cpu_online(cpu, false); + + /* Find a new timesync master */ + if (cpu == time_keeper_id) { + time_keeper_id = cpumask_first(cpu_online_mask); + pr_info("CPU %d is now promoted to time-keeper master\n", time_keeper_id); + } + + disable_percpu_irq(IPI_IRQ); + + irq_migrate_all_off_this_cpu(); + + flush_cache_all_local(); + flush_tlb_all_local(NULL); + + /* disable all irqs, including timer irq */ + local_irq_disable(); + + /* wait for next timer irq ... */ + mdelay(1000/HZ+100); + + /* ... and then clear all pending external irqs */ + set_eiem(0); + mtctl(~0UL, CR_EIRR); + mfctl(CR_EIRR); + mtctl(0, CR_EIRR); +#endif + return 0; +} - return cpu_online(cpu) ? 0 : -ENOSYS; +/* + * called on the thread which is asking for a CPU to be shutdown - + * waits until shutdown has completed, or it is timed out. + */ +void __cpu_die(unsigned int cpu) +{ + pdc_cpu_rendezvous_lock(); } -#ifdef CONFIG_PROC_FS -int setup_profiling_timer(unsigned int multiplier) +void arch_cpuhp_cleanup_dead_cpu(unsigned int cpu) { - return -EINVAL; + pr_info("CPU%u: is shutting down\n", cpu); + + /* set task's state to interruptible sleep */ + set_current_state(TASK_INTERRUPTIBLE); + schedule_timeout((IS_ENABLED(CONFIG_64BIT) ? 8:2) * HZ); + + pdc_cpu_rendezvous_unlock(); } -#endif diff --git a/arch/parisc/kernel/stacktrace.c b/arch/parisc/kernel/stacktrace.c index 34bf6d6bf6e8..023834ef582e 100644 --- a/arch/parisc/kernel/stacktrace.c +++ b/arch/parisc/kernel/stacktrace.c @@ -2,45 +2,42 @@ /* * Stack trace management functions * - * Copyright (C) 2009 Helge Deller <deller@gmx.de> + * Copyright (C) 2009-2021 Helge Deller <deller@gmx.de> * based on arch/x86/kernel/stacktrace.c by Ingo Molnar <mingo@redhat.com> * and parisc unwind functions by Randolph Chung <tausq@debian.org> * * TODO: Userspace stacktrace (CONFIG_USER_STACKTRACE_SUPPORT) */ -#include <linux/module.h> +#include <linux/kernel.h> #include <linux/stacktrace.h> #include <asm/unwind.h> -static void dump_trace(struct task_struct *task, struct stack_trace *trace) +static void notrace walk_stackframe(struct task_struct *task, + struct pt_regs *regs, bool (*fn)(void *, unsigned long), void *cookie) { struct unwind_frame_info info; unwind_frame_init_task(&info, task, NULL); - - /* unwind stack and save entries in stack_trace struct */ - trace->nr_entries = 0; - while (trace->nr_entries < trace->max_entries) { + while (1) { if (unwind_once(&info) < 0 || info.ip == 0) break; if (__kernel_text_address(info.ip)) - trace->entries[trace->nr_entries++] = info.ip; + if (!fn(cookie, info.ip)) + break; } } -/* - * Save stack-backtrace addresses into a stack_trace buffer. - */ -void save_stack_trace(struct stack_trace *trace) +void arch_stack_walk(stack_trace_consume_fn consume_entry, void *cookie, + struct task_struct *task, struct pt_regs *regs) { - dump_trace(current, trace); + walk_stackframe(task, regs, consume_entry, cookie); } -EXPORT_SYMBOL_GPL(save_stack_trace); -void save_stack_trace_tsk(struct task_struct *tsk, struct stack_trace *trace) +int arch_stack_walk_reliable(stack_trace_consume_fn consume_entry, void *cookie, + struct task_struct *task) { - dump_trace(tsk, trace); + walk_stackframe(task, NULL, consume_entry, cookie); + return 1; } -EXPORT_SYMBOL_GPL(save_stack_trace_tsk); diff --git a/arch/parisc/kernel/sys_parisc.c b/arch/parisc/kernel/sys_parisc.c index 5d458a44b09c..98af719d5f85 100644 --- a/arch/parisc/kernel/sys_parisc.c +++ b/arch/parisc/kernel/sys_parisc.c @@ -6,7 +6,7 @@ * Copyright (C) 1999-2003 Matthew Wilcox <willy at parisc-linux.org> * Copyright (C) 2000-2003 Paul Bame <bame at parisc-linux.org> * Copyright (C) 2001 Thomas Bogendoerfer <tsbogend at parisc-linux.org> - * Copyright (C) 1999-2014 Helge Deller <deller@gmx.de> + * Copyright (C) 1999-2020 Helge Deller <deller@gmx.de> */ #include <linux/uaccess.h> @@ -23,35 +23,51 @@ #include <linux/utsname.h> #include <linux/personality.h> #include <linux/random.h> +#include <linux/compat.h> +#include <linux/elf-randomize.h> -/* we construct an artificial offset for the mapping based on the physical - * address of the kernel mapping variable */ -#define GET_LAST_MMAP(filp) \ - (filp ? ((unsigned long) filp->f_mapping) >> 8 : 0UL) -#define SET_LAST_MMAP(filp, val) \ - { /* nothing */ } - -static int get_offset(unsigned int last_mmap) -{ - return (last_mmap & (SHM_COLOUR-1)) >> PAGE_SHIFT; -} +/* + * Construct an artificial page offset for the mapping based on the physical + * address of the kernel file mapping variable. + */ +#define GET_FILP_PGOFF(filp) \ + (filp ? (((unsigned long) filp->f_mapping) >> 8) \ + & ((SHM_COLOUR-1) >> PAGE_SHIFT) : 0UL) -static unsigned long shared_align_offset(unsigned int last_mmap, +static unsigned long shared_align_offset(unsigned long filp_pgoff, unsigned long pgoff) { - return (get_offset(last_mmap) + pgoff) << PAGE_SHIFT; + return (filp_pgoff + pgoff) << PAGE_SHIFT; } static inline unsigned long COLOR_ALIGN(unsigned long addr, - unsigned int last_mmap, unsigned long pgoff) + unsigned long filp_pgoff, unsigned long pgoff) { unsigned long base = (addr+SHM_COLOUR-1) & ~(SHM_COLOUR-1); unsigned long off = (SHM_COLOUR-1) & - (shared_align_offset(last_mmap, pgoff) << PAGE_SHIFT); - + shared_align_offset(filp_pgoff, pgoff); return base + off; } + +#define STACK_SIZE_DEFAULT (USER_WIDE_MODE \ + ? (1 << 30) /* 1 GB */ \ + : (CONFIG_STACK_MAX_DEFAULT_SIZE_MB*1024*1024)) + +unsigned long calc_max_stack_size(unsigned long stack_max) +{ +#ifdef CONFIG_COMPAT + if (!USER_WIDE_MODE && (stack_max == COMPAT_RLIM_INFINITY)) + stack_max = STACK_SIZE_DEFAULT; + else +#endif + if (stack_max == RLIM_INFINITY) + stack_max = STACK_SIZE_DEFAULT; + + return stack_max; +} + + /* * Top of mmap area (just below the process stack). */ @@ -61,15 +77,15 @@ static inline unsigned long COLOR_ALIGN(unsigned long addr, * indicating that "current" should be used instead of a passed-in * value from the exec bprm as done with arch_pick_mmap_layout(). */ -static unsigned long mmap_upper_limit(struct rlimit *rlim_stack) +unsigned long mmap_upper_limit(struct rlimit *rlim_stack) { unsigned long stack_base; /* Limit stack size - see setup_arg_pages() in fs/exec.c */ stack_base = rlim_stack ? rlim_stack->rlim_max : rlimit_max(RLIMIT_STACK); - if (stack_base > STACK_SIZE_MAX) - stack_base = STACK_SIZE_MAX; + + stack_base = calc_max_stack_size(stack_base); /* Add space for stack randomization. */ if (current->flags & PF_RANDOMIZE) @@ -78,92 +94,41 @@ static unsigned long mmap_upper_limit(struct rlimit *rlim_stack) return PAGE_ALIGN(STACK_TOP - stack_base); } +enum mmap_allocation_direction {UP, DOWN}; -unsigned long arch_get_unmapped_area(struct file *filp, unsigned long addr, - unsigned long len, unsigned long pgoff, unsigned long flags) +static unsigned long arch_get_unmapped_area_common(struct file *filp, + unsigned long addr, unsigned long len, unsigned long pgoff, + unsigned long flags, enum mmap_allocation_direction dir) { struct mm_struct *mm = current->mm; struct vm_area_struct *vma, *prev; - unsigned long task_size = TASK_SIZE; - int do_color_align, last_mmap; + unsigned long filp_pgoff; + int do_color_align; struct vm_unmapped_area_info info; - if (len > task_size) + if (unlikely(len > TASK_SIZE)) return -ENOMEM; do_color_align = 0; if (filp || (flags & MAP_SHARED)) do_color_align = 1; - last_mmap = GET_LAST_MMAP(filp); + filp_pgoff = GET_FILP_PGOFF(filp); if (flags & MAP_FIXED) { - if ((flags & MAP_SHARED) && last_mmap && - (addr - shared_align_offset(last_mmap, pgoff)) - & (SHM_COLOUR - 1)) + /* Even MAP_FIXED mappings must reside within TASK_SIZE */ + if (TASK_SIZE - len < addr) return -EINVAL; - goto found_addr; - } - - if (addr) { - if (do_color_align && last_mmap) - addr = COLOR_ALIGN(addr, last_mmap, pgoff); - else - addr = PAGE_ALIGN(addr); - - vma = find_vma_prev(mm, addr, &prev); - if (task_size - len >= addr && - (!vma || addr + len <= vm_start_gap(vma)) && - (!prev || addr >= vm_end_gap(prev))) - goto found_addr; - } - - info.flags = 0; - info.length = len; - info.low_limit = mm->mmap_legacy_base; - info.high_limit = mmap_upper_limit(NULL); - info.align_mask = last_mmap ? (PAGE_MASK & (SHM_COLOUR - 1)) : 0; - info.align_offset = shared_align_offset(last_mmap, pgoff); - addr = vm_unmapped_area(&info); - -found_addr: - if (do_color_align && !last_mmap && !(addr & ~PAGE_MASK)) - SET_LAST_MMAP(filp, addr - (pgoff << PAGE_SHIFT)); - - return addr; -} -unsigned long -arch_get_unmapped_area_topdown(struct file *filp, const unsigned long addr0, - const unsigned long len, const unsigned long pgoff, - const unsigned long flags) -{ - struct vm_area_struct *vma, *prev; - struct mm_struct *mm = current->mm; - unsigned long addr = addr0; - int do_color_align, last_mmap; - struct vm_unmapped_area_info info; - - /* requested length too big for entire address space */ - if (len > TASK_SIZE) - return -ENOMEM; - - do_color_align = 0; - if (filp || (flags & MAP_SHARED)) - do_color_align = 1; - last_mmap = GET_LAST_MMAP(filp); - - if (flags & MAP_FIXED) { - if ((flags & MAP_SHARED) && last_mmap && - (addr - shared_align_offset(last_mmap, pgoff)) - & (SHM_COLOUR - 1)) + if ((flags & MAP_SHARED) && filp && + (addr - shared_align_offset(filp_pgoff, pgoff)) + & (SHM_COLOUR - 1)) return -EINVAL; - goto found_addr; + return addr; } - /* requesting a specific address */ if (addr) { - if (do_color_align && last_mmap) - addr = COLOR_ALIGN(addr, last_mmap, pgoff); + if (do_color_align) + addr = COLOR_ALIGN(addr, filp_pgoff, pgoff); else addr = PAGE_ALIGN(addr); @@ -171,87 +136,51 @@ arch_get_unmapped_area_topdown(struct file *filp, const unsigned long addr0, if (TASK_SIZE - len >= addr && (!vma || addr + len <= vm_start_gap(vma)) && (!prev || addr >= vm_end_gap(prev))) - goto found_addr; + return addr; } - info.flags = VM_UNMAPPED_AREA_TOPDOWN; info.length = len; - info.low_limit = PAGE_SIZE; - info.high_limit = mm->mmap_base; - info.align_mask = last_mmap ? (PAGE_MASK & (SHM_COLOUR - 1)) : 0; - info.align_offset = shared_align_offset(last_mmap, pgoff); - addr = vm_unmapped_area(&info); - if (!(addr & ~PAGE_MASK)) - goto found_addr; - VM_BUG_ON(addr != -ENOMEM); - - /* - * A failed mmap() very likely causes application failure, - * so fall back to the bottom-up function here. This scenario - * can happen with large stack limits and large mmap() - * allocations. - */ - return arch_get_unmapped_area(filp, addr0, len, pgoff, flags); - -found_addr: - if (do_color_align && !last_mmap && !(addr & ~PAGE_MASK)) - SET_LAST_MMAP(filp, addr - (pgoff << PAGE_SHIFT)); - - return addr; -} - -static int mmap_is_legacy(void) -{ - if (current->personality & ADDR_COMPAT_LAYOUT) - return 1; - - /* parisc stack always grows up - so a unlimited stack should - * not be an indicator to use the legacy memory layout. - * if (rlimit(RLIMIT_STACK) == RLIM_INFINITY) - * return 1; - */ - - return sysctl_legacy_va_layout; -} - -static unsigned long mmap_rnd(void) -{ - unsigned long rnd = 0; - - if (current->flags & PF_RANDOMIZE) - rnd = get_random_int() & MMAP_RND_MASK; - - return rnd << PAGE_SHIFT; -} + info.align_mask = do_color_align ? (PAGE_MASK & (SHM_COLOUR - 1)) : 0; + info.align_offset = shared_align_offset(filp_pgoff, pgoff); + + if (dir == DOWN) { + info.flags = VM_UNMAPPED_AREA_TOPDOWN; + info.low_limit = PAGE_SIZE; + info.high_limit = mm->mmap_base; + addr = vm_unmapped_area(&info); + if (!(addr & ~PAGE_MASK)) + return addr; + VM_BUG_ON(addr != -ENOMEM); + + /* + * A failed mmap() very likely causes application failure, + * so fall back to the bottom-up function here. This scenario + * can happen with large stack limits and large mmap() + * allocations. + */ + } -unsigned long arch_mmap_rnd(void) -{ - return (get_random_int() & MMAP_RND_MASK) << PAGE_SHIFT; + info.flags = 0; + info.low_limit = mm->mmap_base; + info.high_limit = mmap_upper_limit(NULL); + return vm_unmapped_area(&info); } -static unsigned long mmap_legacy_base(void) +unsigned long arch_get_unmapped_area(struct file *filp, unsigned long addr, + unsigned long len, unsigned long pgoff, unsigned long flags) { - return TASK_UNMAPPED_BASE + mmap_rnd(); + return arch_get_unmapped_area_common(filp, + addr, len, pgoff, flags, UP); } -/* - * This function, called very early during the creation of a new - * process VM image, sets up which VM layout function to use: - */ -void arch_pick_mmap_layout(struct mm_struct *mm, struct rlimit *rlim_stack) +unsigned long arch_get_unmapped_area_topdown(struct file *filp, + unsigned long addr, unsigned long len, unsigned long pgoff, + unsigned long flags) { - mm->mmap_legacy_base = mmap_legacy_base(); - mm->mmap_base = mmap_upper_limit(rlim_stack); - - if (mmap_is_legacy()) { - mm->mmap_base = mm->mmap_legacy_base; - mm->get_unmapped_area = arch_get_unmapped_area; - } else { - mm->get_unmapped_area = arch_get_unmapped_area_topdown; - } + return arch_get_unmapped_area_common(filp, + addr, len, pgoff, flags, DOWN); } - asmlinkage unsigned long sys_mmap2(unsigned long addr, unsigned long len, unsigned long prot, unsigned long flags, unsigned long fd, unsigned long pgoff) @@ -359,7 +288,7 @@ asmlinkage long parisc_fallocate(int fd, int mode, u32 offhi, u32 offlo, ((u64)lenhi << 32) | lenlo); } -long parisc_personality(unsigned long personality) +asmlinkage long parisc_personality(unsigned long personality) { long err; @@ -373,3 +302,103 @@ long parisc_personality(unsigned long personality) return err; } + +/* + * Up to kernel v5.9 we defined O_NONBLOCK as 000200004, + * since then O_NONBLOCK is defined as 000200000. + * + * The following wrapper functions mask out the old + * O_NDELAY bit from calls which use O_NONBLOCK. + * + * XXX: Remove those in year 2022 (or later)? + */ + +#define O_NONBLOCK_OLD 000200004 +#define O_NONBLOCK_MASK_OUT (O_NONBLOCK_OLD & ~O_NONBLOCK) + +static int FIX_O_NONBLOCK(int flags) +{ + if ((flags & O_NONBLOCK_MASK_OUT) && + !test_thread_flag(TIF_NONBLOCK_WARNING)) { + set_thread_flag(TIF_NONBLOCK_WARNING); + pr_warn("%s(%d) uses a deprecated O_NONBLOCK value." + " Please recompile with newer glibc.\n", + current->comm, current->pid); + } + return flags & ~O_NONBLOCK_MASK_OUT; +} + +asmlinkage long parisc_timerfd_create(int clockid, int flags) +{ + flags = FIX_O_NONBLOCK(flags); + return sys_timerfd_create(clockid, flags); +} + +asmlinkage long parisc_signalfd4(int ufd, sigset_t __user *user_mask, + size_t sizemask, int flags) +{ + flags = FIX_O_NONBLOCK(flags); + return sys_signalfd4(ufd, user_mask, sizemask, flags); +} + +#ifdef CONFIG_COMPAT +asmlinkage long parisc_compat_signalfd4(int ufd, + compat_sigset_t __user *user_mask, + compat_size_t sizemask, int flags) +{ + flags = FIX_O_NONBLOCK(flags); + return compat_sys_signalfd4(ufd, user_mask, sizemask, flags); +} +#endif + +asmlinkage long parisc_eventfd2(unsigned int count, int flags) +{ + flags = FIX_O_NONBLOCK(flags); + return sys_eventfd2(count, flags); +} + +asmlinkage long parisc_userfaultfd(int flags) +{ + flags = FIX_O_NONBLOCK(flags); + return sys_userfaultfd(flags); +} + +asmlinkage long parisc_pipe2(int __user *fildes, int flags) +{ + flags = FIX_O_NONBLOCK(flags); + return sys_pipe2(fildes, flags); +} + +asmlinkage long parisc_inotify_init1(int flags) +{ + flags = FIX_O_NONBLOCK(flags); + return sys_inotify_init1(flags); +} + +/* + * madvise() wrapper + * + * Up to kernel v6.1 parisc has different values than all other + * platforms for the MADV_xxx flags listed below. + * To keep binary compatibility with existing userspace programs + * translate the former values to the new values. + * + * XXX: Remove this wrapper in year 2025 (or later) + */ + +asmlinkage notrace long parisc_madvise(unsigned long start, size_t len_in, int behavior) +{ + switch (behavior) { + case 65: behavior = MADV_MERGEABLE; break; + case 66: behavior = MADV_UNMERGEABLE; break; + case 67: behavior = MADV_HUGEPAGE; break; + case 68: behavior = MADV_NOHUGEPAGE; break; + case 69: behavior = MADV_DONTDUMP; break; + case 70: behavior = MADV_DODUMP; break; + case 71: behavior = MADV_WIPEONFORK; break; + case 72: behavior = MADV_KEEPONFORK; break; + case 73: behavior = MADV_COLLAPSE; break; + } + + return sys_madvise(start, len_in, behavior); +} diff --git a/arch/parisc/kernel/syscall.S b/arch/parisc/kernel/syscall.S index 97ac707c6bff..1f51aa9c8230 100644 --- a/arch/parisc/kernel/syscall.S +++ b/arch/parisc/kernel/syscall.S @@ -39,6 +39,7 @@ registers). #include <asm/assembly.h> #include <asm/processor.h> #include <asm/cache.h> +#include <asm/spinlock_types.h> #include <linux/linkage.h> @@ -50,6 +51,32 @@ registers). .level PA_ASM_LEVEL + .macro lws_pagefault_disable reg1,reg2 + mfctl %cr30, \reg2 + ldo TASK_PAGEFAULT_DISABLED(\reg2), \reg2 + ldw 0(%sr2,\reg2), \reg1 + ldo 1(\reg1), \reg1 + stw \reg1, 0(%sr2,\reg2) + .endm + + .macro lws_pagefault_enable reg1,reg2 + mfctl %cr30, \reg2 + ldo TASK_PAGEFAULT_DISABLED(\reg2), \reg2 + ldw 0(%sr2,\reg2), \reg1 + ldo -1(\reg1), \reg1 + stw \reg1, 0(%sr2,\reg2) + .endm + + /* raise exception if spinlock content is not zero or + * __ARCH_SPIN_LOCK_UNLOCKED_VAL */ + .macro spinlock_check spin_val,tmpreg +#ifdef CONFIG_LIGHTWEIGHT_SPINLOCK_CHECK + ldi __ARCH_SPIN_LOCK_UNLOCKED_VAL, \tmpreg + andcm,= \spin_val, \tmpreg, %r0 + .word SPINLOCK_BREAK_INSN +#endif + .endm + .text .import syscall_exit,code @@ -74,11 +101,11 @@ ENTRY(linux_gateway_page) /* ADDRESS 0xb0 to 0xb8, lws uses two insns for entry */ /* Light-weight-syscall entry must always be located at 0xb0 */ /* WARNING: Keep this number updated with table size changes */ -#define __NR_lws_entries (3) +#define __NR_lws_entries (5) lws_entry: gate lws_start, %r0 /* increase privilege */ - depi 3, 31, 2, %r31 /* Ensure we return into user mode. */ + depi PRIV_USER, 31, 2, %r31 /* Ensure we return into user mode. */ /* Fill from 0xb8 to 0xe0 */ .rept 10 @@ -89,7 +116,7 @@ lws_entry: mechanism to work. DO NOT MOVE THIS CODE EVER! */ set_thread_pointer: gate .+8, %r0 /* increase privilege */ - depi 3, 31, 2, %r31 /* Ensure we return into user mode. */ + depi PRIV_USER, 31, 2, %r31 /* Ensure we return into user mode. */ be 0(%sr7,%r31) /* return to user space */ mtctl %r26, %cr27 /* move arg0 to the control register */ @@ -139,9 +166,9 @@ linux_gateway_entry: xor %r1,%r30,%r30 /* ye olde xor trick */ xor %r1,%r30,%r1 xor %r1,%r30,%r30 - - ldo THREAD_SZ_ALGN+FRAME_SIZE(%r30),%r30 /* set up kernel stack */ + LDREG TASK_STACK(%r30),%r30 /* set up kernel stack */ + ldo FRAME_SIZE(%r30),%r30 /* N.B.: It is critical that we don't set sr7 to 0 until r30 * contains a valid kernel stack pointer. It is also * critical that we don't start using the kernel stack @@ -152,7 +179,6 @@ linux_gateway_entry: ssm PSW_SM_I, %r0 /* enable interrupts */ STREGM %r1,FRAME_SIZE(%r30) /* save r1 (usp) here for now */ mfctl %cr30,%r1 /* get task ptr in %r1 */ - LDREG TI_TASK(%r1),%r1 /* Save some registers for sigcontext and potential task switch (see entry.S for the details of which ones are @@ -207,7 +233,7 @@ linux_gateway_entry: /* Are we being ptraced? */ mfctl %cr30, %r1 - LDREG TI_FLAGS(%r1),%r1 + LDREG TASK_TI_FLAGS(%r1),%r1 ldi _TIF_SYSCALL_TRACE_MASK, %r19 and,COND(=) %r1, %r19, %r0 b,n .Ltracesys @@ -272,8 +298,7 @@ tracesys: * C bit set, a non-straced syscall entry results in C and D clear * in the saved PSW. */ - ldo -THREAD_SZ_ALGN-FRAME_SIZE(%r30),%r1 /* get task ptr */ - LDREG TI_TASK(%r1), %r1 + mfctl %cr30,%r1 /* get task ptr */ ssm 0,%r2 STREG %r2,TASK_PT_PSW(%r1) /* Lower 8 bits only!! */ mfsp %sr0,%r2 @@ -327,8 +352,7 @@ tracesys_next: */ copy %ret0,%r20 - ldo -THREAD_SZ_ALGN-FRAME_SIZE(%r30),%r1 /* get task ptr */ - LDREG TI_TASK(%r1), %r1 + mfctl %cr30,%r1 /* get task ptr */ LDREG TASK_PT_GR28(%r1), %r28 /* Restore return value */ LDREG TASK_PT_GR26(%r1), %r26 /* Restore the users args */ LDREG TASK_PT_GR25(%r1), %r25 @@ -385,16 +409,14 @@ tracesys_next: makes a direct call to syscall_trace. */ tracesys_exit: - ldo -THREAD_SZ_ALGN-FRAME_SIZE(%r30),%r1 /* get task ptr */ - LDREG TI_TASK(%r1), %r1 + mfctl %cr30,%r1 /* get task ptr */ #ifdef CONFIG_64BIT ldo -16(%r30),%r29 /* Reference param save area */ #endif ldo TASK_REGS(%r1),%r26 BL do_syscall_trace_exit,%r2 STREG %r28,TASK_PT_GR28(%r1) /* save return value now */ - ldo -THREAD_SZ_ALGN-FRAME_SIZE(%r30),%r1 /* get task ptr */ - LDREG TI_TASK(%r1), %r1 + mfctl %cr30,%r1 /* get task ptr */ LDREG TASK_PT_GR28(%r1), %r28 /* Restore return val. */ ldil L%syscall_exit,%r1 @@ -407,8 +429,7 @@ tracesys_exit: ldo R%tracesys_sigexit(%r2),%r2 tracesys_sigexit: - ldo -THREAD_SZ_ALGN-FRAME_SIZE(%r30),%r1 /* get task ptr */ - LDREG TI_TASK(%r1), %r1 + mfctl %cr30,%r1 /* get task ptr */ #ifdef CONFIG_64BIT ldo -16(%r30),%r29 /* Reference param save area */ #endif @@ -478,7 +499,7 @@ lws_start: extrd,u %r1,PSW_W_BIT,1,%r1 /* sp must be aligned on 4, so deposit the W bit setting into * the bottom of sp temporarily */ - or,ev %r1,%r30,%r30 + or,od %r1,%r30,%r30 /* Clip LWS number to a 32-bit value for 32-bit processes */ depdi 0, 31, 32, %r20 @@ -496,8 +517,36 @@ lws_start: /* Jump to lws, lws table pointers already relocated */ be,n 0(%sr2,%r21) +lws_exit_noerror: + lws_pagefault_enable %r1,%r21 + ldi __ARCH_SPIN_LOCK_UNLOCKED_VAL, %r21 + stw,ma %r21, 0(%sr2,%r20) + ssm PSW_SM_I, %r0 + b lws_exit + copy %r0, %r21 + +lws_wouldblock: + ssm PSW_SM_I, %r0 + ldo 2(%r0), %r28 + b lws_exit + ldo -EAGAIN(%r0), %r21 + +lws_pagefault: + lws_pagefault_enable %r1,%r21 + ldi __ARCH_SPIN_LOCK_UNLOCKED_VAL, %r21 + stw,ma %r21, 0(%sr2,%r20) + ssm PSW_SM_I, %r0 + ldo 3(%r0),%r28 + b lws_exit + ldo -EAGAIN(%r0),%r21 + +lws_fault: + ldo 1(%r0),%r28 + b lws_exit + ldo -EFAULT(%r0),%r21 + lws_exit_nosys: - ldo -ENOSYS(%r0),%r21 /* set errno */ + ldo -ENOSYS(%r0),%r21 /* Fall through: Return to userspace */ lws_exit: @@ -524,27 +573,19 @@ lws_exit: %r28 - Return prev through this register. %r21 - Kernel error code - If debugging is DISabled: - - %r21 has the following meanings: - + %r21 returns the following error codes: EAGAIN - CAS is busy, ldcw failed, try again. EFAULT - Read or write failed. - If debugging is enabled: - - EDEADLOCK - CAS called recursively. - EAGAIN && r28 == 1 - CAS is busy. Lock contended. - EAGAIN && r28 == 2 - CAS is busy. ldcw failed. - EFAULT - Read or write failed. + If EAGAIN is returned, %r28 indicates the busy reason: + r28 == 1 - CAS is busy. lock contended. + r28 == 2 - CAS is busy. ldcw failed. + r28 == 3 - CAS is busy. page fault. Scratch: r20, r28, r1 ****************************************************/ - /* Do not enable LWS debugging */ -#define ENABLE_LWS_DEBUG 0 - /* ELF64 Process entry path */ lws_compare_and_swap64: #ifdef CONFIG_64BIT @@ -557,61 +598,46 @@ lws_compare_and_swap64: b,n lws_exit_nosys #endif - /* ELF32 Process entry path */ + /* ELF32/ELF64 Process entry path */ lws_compare_and_swap32: #ifdef CONFIG_64BIT - /* Clip all the input registers */ + /* Wide mode user process? */ + bb,<,n %sp, 31, lws_compare_and_swap + + /* Clip all the input registers for 32-bit processes */ depdi 0, 31, 32, %r26 depdi 0, 31, 32, %r25 depdi 0, 31, 32, %r24 #endif lws_compare_and_swap: - /* Load start of lock table */ - ldil L%lws_lock_start, %r20 - ldo R%lws_lock_start(%r20), %r28 + /* Trigger memory reference interruptions without writing to memory */ +1: ldw 0(%r26), %r28 +2: stbys,e %r0, 0(%r26) - /* Extract four bits from r26 and hash lock (Bits 4-7) */ - extru %r26, 27, 4, %r20 + /* Calculate 8-bit hash index from virtual address */ + extru_safe %r26, 27, 8, %r20 + + /* Load start of lock table */ + ldil L%lws_lock_start, %r28 + ldo R%lws_lock_start(%r28), %r28 - /* Find lock to use, the hash is either one of 0 to - 15, multiplied by 16 (keep it 16-byte aligned) + /* Find lock to use, the hash index is one of 0 to + 255, multiplied by 16 (keep it 16-byte aligned) and add to the lock table offset. */ shlw %r20, 4, %r20 add %r20, %r28, %r20 -# if ENABLE_LWS_DEBUG - /* - DEBUG, check for deadlock! - If the thread register values are the same - then we were the one that locked it last and - this is a recurisve call that will deadlock. - We *must* giveup this call and fail. - */ - ldw 4(%sr2,%r20), %r28 /* Load thread register */ - /* WARNING: If cr27 cycles to the same value we have problems */ - mfctl %cr27, %r21 /* Get current thread register */ - cmpb,<>,n %r21, %r28, cas_lock /* Called recursive? */ - b lws_exit /* Return error! */ - ldo -EDEADLOCK(%r0), %r21 -cas_lock: - cmpb,=,n %r0, %r28, cas_nocontend /* Is nobody using it? */ - ldo 1(%r0), %r28 /* 1st case */ - b lws_exit /* Contended... */ - ldo -EAGAIN(%r0), %r21 /* Spin in userspace */ -cas_nocontend: -# endif -/* ENABLE_LWS_DEBUG */ - rsm PSW_SM_I, %r0 /* Disable interrupts */ - /* COW breaks can cause contention on UP systems */ - LDCW 0(%sr2,%r20), %r28 /* Try to acquire the lock */ - cmpb,<>,n %r0, %r28, cas_action /* Did we get it? */ -cas_wouldblock: - ldo 2(%r0), %r28 /* 2nd case */ - ssm PSW_SM_I, %r0 - b lws_exit /* Contended... */ - ldo -EAGAIN(%r0), %r21 /* Spin in userspace */ + + /* Try to acquire the lock */ + LDCW 0(%sr2,%r20), %r28 + spinlock_check %r28, %r21 + comclr,<> %r0, %r28, %r0 + b,n lws_wouldblock + + /* Disable page faults to prevent sleeping in critical region */ + lws_pagefault_disable %r21,%r28 /* prev = *addr; @@ -621,70 +647,35 @@ cas_wouldblock: */ /* NOTES: - This all works becuse intr_do_signal + This all works because intr_do_signal and schedule both check the return iasq and see that we are on the kernel page so this process is never scheduled off or is ever sent any signal of any sort, - thus it is wholly atomic from usrspaces + thus it is wholly atomic from usrspace's perspective */ -cas_action: -#if defined CONFIG_SMP && ENABLE_LWS_DEBUG - /* DEBUG */ - mfctl %cr27, %r1 - stw %r1, 4(%sr2,%r20) -#endif /* The load and store could fail */ -1: ldw 0(%r26), %r28 +3: ldw 0(%r26), %r28 sub,<> %r28, %r25, %r0 -2: stw %r24, 0(%r26) - /* Free lock */ -#ifdef CONFIG_SMP -98: LDCW 0(%sr2,%r20), %r1 /* Barrier */ -99: ALTERNATIVE(98b, 99b, ALT_COND_NO_SMP, INSN_NOP) -#endif - stw %r20, 0(%sr2,%r20) -#if ENABLE_LWS_DEBUG - /* Clear thread register indicator */ - stw %r0, 4(%sr2,%r20) -#endif - /* Enable interrupts */ - ssm PSW_SM_I, %r0 - /* Return to userspace, set no error */ - b lws_exit - copy %r0, %r21 +4: stw %r24, 0(%r26) + b,n lws_exit_noerror -3: - /* Error occurred on load or store */ - /* Free lock */ -#ifdef CONFIG_SMP -98: LDCW 0(%sr2,%r20), %r1 /* Barrier */ -99: ALTERNATIVE(98b, 99b, ALT_COND_NO_SMP, INSN_NOP) -#endif - stw %r20, 0(%sr2,%r20) -#if ENABLE_LWS_DEBUG - stw %r0, 4(%sr2,%r20) -#endif - ssm PSW_SM_I, %r0 - b lws_exit - ldo -EFAULT(%r0),%r21 /* set errno */ - nop - nop - nop - nop + /* A fault occurred on load or stbys,e store */ +5: b,n lws_fault + ASM_EXCEPTIONTABLE_ENTRY(1b-linux_gateway_page, 5b-linux_gateway_page) + ASM_EXCEPTIONTABLE_ENTRY(2b-linux_gateway_page, 5b-linux_gateway_page) - /* Two exception table entries, one for the load, - the other for the store. Either return -EFAULT. - Each of the entries must be relocated. */ - ASM_EXCEPTIONTABLE_ENTRY(1b-linux_gateway_page, 3b-linux_gateway_page) - ASM_EXCEPTIONTABLE_ENTRY(2b-linux_gateway_page, 3b-linux_gateway_page) + /* A page fault occurred in critical region */ +6: b,n lws_pagefault + ASM_EXCEPTIONTABLE_ENTRY(3b-linux_gateway_page, 6b-linux_gateway_page) + ASM_EXCEPTIONTABLE_ENTRY(4b-linux_gateway_page, 6b-linux_gateway_page) /*************************************************** New CAS implementation which uses pointers and variable size information. The value pointed by old and new MUST NOT change - while performing CAS. The lock only protect the value at %r26. + while performing CAS. The lock only protects the value at %r26. %r26 - Address to examine %r25 - Pointer to the value to check (old) @@ -693,25 +684,32 @@ cas_action: %r28 - Return non-zero on failure %r21 - Kernel error code - %r21 has the following meanings: - + %r21 returns the following error codes: EAGAIN - CAS is busy, ldcw failed, try again. EFAULT - Read or write failed. + If EAGAIN is returned, %r28 indicates the busy reason: + r28 == 1 - CAS is busy. lock contended. + r28 == 2 - CAS is busy. ldcw failed. + r28 == 3 - CAS is busy. page fault. + Scratch: r20, r22, r28, r29, r1, fr4 (32bit for 64bit CAS only) ****************************************************/ - /* ELF32 Process entry path */ lws_compare_and_swap_2: #ifdef CONFIG_64BIT - /* Clip the input registers. We don't need to clip %r23 as we - only use it for word operations */ + /* Wide mode user process? */ + bb,<,n %sp, 31, cas2_begin + + /* Clip the input registers for 32-bit processes. We don't + need to clip %r23 as we only use it for word operations */ depdi 0, 31, 32, %r26 depdi 0, 31, 32, %r25 depdi 0, 31, 32, %r24 #endif +cas2_begin: /* Check the validity of the size pointer */ subi,>>= 3, %r23, %r0 b,n lws_exit_nosys @@ -722,71 +720,78 @@ lws_compare_and_swap_2: blr %r29, %r0 nop - /* 8bit load */ -4: ldb 0(%r25), %r25 + /* 8-bit load */ +1: ldb 0(%r25), %r25 b cas2_lock_start -5: ldb 0(%r24), %r24 +2: ldb 0(%r24), %r24 nop nop nop nop nop - /* 16bit load */ -6: ldh 0(%r25), %r25 + /* 16-bit load */ +3: ldh 0(%r25), %r25 b cas2_lock_start -7: ldh 0(%r24), %r24 +4: ldh 0(%r24), %r24 nop nop nop nop nop - /* 32bit load */ -8: ldw 0(%r25), %r25 + /* 32-bit load */ +5: ldw 0(%r25), %r25 b cas2_lock_start -9: ldw 0(%r24), %r24 +6: ldw 0(%r24), %r24 nop nop nop nop nop - /* 64bit load */ + /* 64-bit load */ #ifdef CONFIG_64BIT -10: ldd 0(%r25), %r25 -11: ldd 0(%r24), %r24 +7: ldd 0(%r25), %r25 +8: ldd 0(%r24), %r24 #else /* Load old value into r22/r23 - high/low */ -10: ldw 0(%r25), %r22 -11: ldw 4(%r25), %r23 +7: ldw 0(%r25), %r22 +8: ldw 4(%r25), %r23 /* Load new value into fr4 for atomic store later */ -12: flddx 0(%r24), %fr4 +9: flddx 0(%r24), %fr4 #endif cas2_lock_start: - /* Load start of lock table */ - ldil L%lws_lock_start, %r20 - ldo R%lws_lock_start(%r20), %r28 + /* Trigger memory reference interruptions without writing to memory */ + copy %r26, %r28 + depi_safe 0, 31, 2, %r28 +10: ldw 0(%r28), %r1 +11: stbys,e %r0, 0(%r28) - /* Extract four bits from r26 and hash lock (Bits 4-7) */ - extru %r26, 27, 4, %r20 + /* Calculate 8-bit hash index from virtual address */ + extru_safe %r26, 27, 8, %r20 + + /* Load start of lock table */ + ldil L%lws_lock_start, %r28 + ldo R%lws_lock_start(%r28), %r28 - /* Find lock to use, the hash is either one of 0 to - 15, multiplied by 16 (keep it 16-byte aligned) + /* Find lock to use, the hash index is one of 0 to + 255, multiplied by 16 (keep it 16-byte aligned) and add to the lock table offset. */ shlw %r20, 4, %r20 add %r20, %r28, %r20 rsm PSW_SM_I, %r0 /* Disable interrupts */ - /* COW breaks can cause contention on UP systems */ - LDCW 0(%sr2,%r20), %r28 /* Try to acquire the lock */ - cmpb,<>,n %r0, %r28, cas2_action /* Did we get it? */ -cas2_wouldblock: - ldo 2(%r0), %r28 /* 2nd case */ - ssm PSW_SM_I, %r0 - b lws_exit /* Contended... */ - ldo -EAGAIN(%r0), %r21 /* Spin in userspace */ + + /* Try to acquire the lock */ + LDCW 0(%sr2,%r20), %r28 + spinlock_check %r28, %r21 + comclr,<> %r0, %r28, %r0 + b,n lws_wouldblock + + /* Disable page faults to prevent sleeping in critical region */ + lws_pagefault_disable %r21,%r28 /* prev = *addr; @@ -796,123 +801,495 @@ cas2_wouldblock: */ /* NOTES: - This all works becuse intr_do_signal + This all works because intr_do_signal and schedule both check the return iasq and see that we are on the kernel page so this process is never scheduled off or is ever sent any signal of any sort, - thus it is wholly atomic from usrspaces + thus it is wholly atomic from usrspace's perspective */ -cas2_action: + /* Jump to the correct function */ blr %r29, %r0 /* Set %r28 as non-zero for now */ ldo 1(%r0),%r28 - /* 8bit CAS */ -13: ldb 0(%r26), %r29 + /* 8-bit CAS */ +12: ldb 0(%r26), %r29 sub,= %r29, %r25, %r0 - b,n cas2_end -14: stb %r24, 0(%r26) - b cas2_end + b,n lws_exit_noerror +13: stb %r24, 0(%r26) + b lws_exit_noerror copy %r0, %r28 nop nop - /* 16bit CAS */ -15: ldh 0(%r26), %r29 + /* 16-bit CAS */ +14: ldh 0(%r26), %r29 sub,= %r29, %r25, %r0 - b,n cas2_end -16: sth %r24, 0(%r26) - b cas2_end + b,n lws_exit_noerror +15: sth %r24, 0(%r26) + b lws_exit_noerror copy %r0, %r28 nop nop - /* 32bit CAS */ -17: ldw 0(%r26), %r29 + /* 32-bit CAS */ +16: ldw 0(%r26), %r29 sub,= %r29, %r25, %r0 - b,n cas2_end -18: stw %r24, 0(%r26) - b cas2_end + b,n lws_exit_noerror +17: stw %r24, 0(%r26) + b lws_exit_noerror copy %r0, %r28 nop nop - /* 64bit CAS */ + /* 64-bit CAS */ #ifdef CONFIG_64BIT -19: ldd 0(%r26), %r29 +18: ldd 0(%r26), %r29 sub,*= %r29, %r25, %r0 - b,n cas2_end -20: std %r24, 0(%r26) + b,n lws_exit_noerror +19: std %r24, 0(%r26) copy %r0, %r28 #else /* Compare first word */ -19: ldw 0(%r26), %r29 +18: ldw 0(%r26), %r29 sub,= %r29, %r22, %r0 - b,n cas2_end + b,n lws_exit_noerror /* Compare second word */ -20: ldw 4(%r26), %r29 +19: ldw 4(%r26), %r29 sub,= %r29, %r23, %r0 - b,n cas2_end + b,n lws_exit_noerror /* Perform the store */ -21: fstdx %fr4, 0(%r26) +20: fstdx %fr4, 0(%r26) copy %r0, %r28 #endif + b lws_exit_noerror + copy %r0, %r28 -cas2_end: - /* Free lock */ -#ifdef CONFIG_SMP -98: LDCW 0(%sr2,%r20), %r1 /* Barrier */ -99: ALTERNATIVE(98b, 99b, ALT_COND_NO_SMP, INSN_NOP) + /* A fault occurred on load or stbys,e store */ +30: b,n lws_fault + ASM_EXCEPTIONTABLE_ENTRY(1b-linux_gateway_page, 30b-linux_gateway_page) + ASM_EXCEPTIONTABLE_ENTRY(2b-linux_gateway_page, 30b-linux_gateway_page) + ASM_EXCEPTIONTABLE_ENTRY(3b-linux_gateway_page, 30b-linux_gateway_page) + ASM_EXCEPTIONTABLE_ENTRY(4b-linux_gateway_page, 30b-linux_gateway_page) + ASM_EXCEPTIONTABLE_ENTRY(5b-linux_gateway_page, 30b-linux_gateway_page) + ASM_EXCEPTIONTABLE_ENTRY(6b-linux_gateway_page, 30b-linux_gateway_page) + ASM_EXCEPTIONTABLE_ENTRY(7b-linux_gateway_page, 30b-linux_gateway_page) + ASM_EXCEPTIONTABLE_ENTRY(8b-linux_gateway_page, 30b-linux_gateway_page) +#ifndef CONFIG_64BIT + ASM_EXCEPTIONTABLE_ENTRY(9b-linux_gateway_page, 30b-linux_gateway_page) #endif - stw %r20, 0(%sr2,%r20) - /* Enable interrupts */ - ssm PSW_SM_I, %r0 - /* Return to userspace, set no error */ - b lws_exit - copy %r0, %r21 -22: - /* Error occurred on load or store */ - /* Free lock */ -#ifdef CONFIG_SMP -98: LDCW 0(%sr2,%r20), %r1 /* Barrier */ -99: ALTERNATIVE(98b, 99b, ALT_COND_NO_SMP, INSN_NOP) + ASM_EXCEPTIONTABLE_ENTRY(10b-linux_gateway_page, 30b-linux_gateway_page) + ASM_EXCEPTIONTABLE_ENTRY(11b-linux_gateway_page, 30b-linux_gateway_page) + + /* A page fault occurred in critical region */ +31: b,n lws_pagefault + ASM_EXCEPTIONTABLE_ENTRY(12b-linux_gateway_page, 31b-linux_gateway_page) + ASM_EXCEPTIONTABLE_ENTRY(13b-linux_gateway_page, 31b-linux_gateway_page) + ASM_EXCEPTIONTABLE_ENTRY(14b-linux_gateway_page, 31b-linux_gateway_page) + ASM_EXCEPTIONTABLE_ENTRY(15b-linux_gateway_page, 31b-linux_gateway_page) + ASM_EXCEPTIONTABLE_ENTRY(16b-linux_gateway_page, 31b-linux_gateway_page) + ASM_EXCEPTIONTABLE_ENTRY(17b-linux_gateway_page, 31b-linux_gateway_page) + ASM_EXCEPTIONTABLE_ENTRY(18b-linux_gateway_page, 31b-linux_gateway_page) + ASM_EXCEPTIONTABLE_ENTRY(19b-linux_gateway_page, 31b-linux_gateway_page) +#ifndef CONFIG_64BIT + ASM_EXCEPTIONTABLE_ENTRY(20b-linux_gateway_page, 31b-linux_gateway_page) #endif - stw %r20, 0(%sr2,%r20) - ssm PSW_SM_I, %r0 + + + /*************************************************** + LWS atomic exchange. + + %r26 - Exchange address + %r25 - Size of the variable (0/1/2/3 for 8/16/32/64 bit) + %r24 - Address of new value + %r23 - Address of old value + %r28 - Return non-zero on failure + %r21 - Kernel error code + + %r21 returns the following error codes: + EAGAIN - CAS is busy, ldcw failed, try again. + EFAULT - Read or write failed. + + If EAGAIN is returned, %r28 indicates the busy reason: + r28 == 1 - CAS is busy. lock contended. + r28 == 2 - CAS is busy. ldcw failed. + r28 == 3 - CAS is busy. page fault. + + Scratch: r20, r1 + + ****************************************************/ + +lws_atomic_xchg: +#ifdef CONFIG_64BIT + /* Wide mode user process? */ + bb,<,n %sp, 31, atomic_xchg_begin + + /* Clip the input registers for 32-bit processes. We don't + need to clip %r23 as we only use it for word operations */ + depdi 0, 31, 32, %r26 + depdi 0, 31, 32, %r25 + depdi 0, 31, 32, %r24 + depdi 0, 31, 32, %r23 +#endif + +atomic_xchg_begin: + /* Check the validity of the size pointer */ + subi,>>= 3, %r25, %r0 + b,n lws_exit_nosys + + /* Jump to the functions which will load the old and new values into + registers depending on the their size */ + shlw %r25, 2, %r1 + blr %r1, %r0 + nop + + /* Perform exception checks */ + + /* 8-bit exchange */ +1: ldb 0(%r24), %r20 + copy %r23, %r20 + depi_safe 0, 31, 2, %r20 + b atomic_xchg_start +2: stbys,e %r0, 0(%r20) + nop + nop + nop + + /* 16-bit exchange */ +3: ldh 0(%r24), %r20 + copy %r23, %r20 + depi_safe 0, 31, 2, %r20 + b atomic_xchg_start +4: stbys,e %r0, 0(%r20) + nop + nop + nop + + /* 32-bit exchange */ +5: ldw 0(%r24), %r20 + b atomic_xchg_start +6: stbys,e %r0, 0(%r23) + nop + nop + nop + nop + nop + + /* 64-bit exchange */ +#ifdef CONFIG_64BIT +7: ldd 0(%r24), %r20 +8: stdby,e %r0, 0(%r23) +#else +7: ldw 0(%r24), %r20 +8: ldw 4(%r24), %r20 + copy %r23, %r20 + depi_safe 0, 31, 2, %r20 +9: stbys,e %r0, 0(%r20) +10: stbys,e %r0, 4(%r20) +#endif + +atomic_xchg_start: + /* Trigger memory reference interruptions without writing to memory */ + copy %r26, %r28 + depi_safe 0, 31, 2, %r28 +11: ldw 0(%r28), %r1 +12: stbys,e %r0, 0(%r28) + + /* Calculate 8-bit hash index from virtual address */ + extru_safe %r26, 27, 8, %r20 + + /* Load start of lock table */ + ldil L%lws_lock_start, %r28 + ldo R%lws_lock_start(%r28), %r28 + + /* Find lock to use, the hash index is one of 0 to + 255, multiplied by 16 (keep it 16-byte aligned) + and add to the lock table offset. */ + shlw %r20, 4, %r20 + add %r20, %r28, %r20 + + rsm PSW_SM_I, %r0 /* Disable interrupts */ + + /* Try to acquire the lock */ + LDCW 0(%sr2,%r20), %r28 + spinlock_check %r28, %r21 + comclr,<> %r0, %r28, %r0 + b,n lws_wouldblock + + /* Disable page faults to prevent sleeping in critical region */ + lws_pagefault_disable %r21,%r28 + + /* NOTES: + This all works because intr_do_signal + and schedule both check the return iasq + and see that we are on the kernel page + so this process is never scheduled off + or is ever sent any signal of any sort, + thus it is wholly atomic from userspace's + perspective + */ + + /* Jump to the correct function */ + blr %r1, %r0 + /* Set %r28 as non-zero for now */ ldo 1(%r0),%r28 - b lws_exit - ldo -EFAULT(%r0),%r21 /* set errno */ - nop - nop - nop - - /* Exception table entries, for the load and store, return EFAULT. - Each of the entries must be relocated. */ - ASM_EXCEPTIONTABLE_ENTRY(4b-linux_gateway_page, 22b-linux_gateway_page) - ASM_EXCEPTIONTABLE_ENTRY(5b-linux_gateway_page, 22b-linux_gateway_page) - ASM_EXCEPTIONTABLE_ENTRY(6b-linux_gateway_page, 22b-linux_gateway_page) - ASM_EXCEPTIONTABLE_ENTRY(7b-linux_gateway_page, 22b-linux_gateway_page) - ASM_EXCEPTIONTABLE_ENTRY(8b-linux_gateway_page, 22b-linux_gateway_page) - ASM_EXCEPTIONTABLE_ENTRY(9b-linux_gateway_page, 22b-linux_gateway_page) - ASM_EXCEPTIONTABLE_ENTRY(10b-linux_gateway_page, 22b-linux_gateway_page) - ASM_EXCEPTIONTABLE_ENTRY(11b-linux_gateway_page, 22b-linux_gateway_page) - ASM_EXCEPTIONTABLE_ENTRY(13b-linux_gateway_page, 22b-linux_gateway_page) - ASM_EXCEPTIONTABLE_ENTRY(14b-linux_gateway_page, 22b-linux_gateway_page) - ASM_EXCEPTIONTABLE_ENTRY(15b-linux_gateway_page, 22b-linux_gateway_page) - ASM_EXCEPTIONTABLE_ENTRY(16b-linux_gateway_page, 22b-linux_gateway_page) - ASM_EXCEPTIONTABLE_ENTRY(17b-linux_gateway_page, 22b-linux_gateway_page) - ASM_EXCEPTIONTABLE_ENTRY(18b-linux_gateway_page, 22b-linux_gateway_page) - ASM_EXCEPTIONTABLE_ENTRY(19b-linux_gateway_page, 22b-linux_gateway_page) - ASM_EXCEPTIONTABLE_ENTRY(20b-linux_gateway_page, 22b-linux_gateway_page) + + /* 8-bit exchange */ +14: ldb 0(%r26), %r1 +15: stb %r1, 0(%r23) +15: ldb 0(%r24), %r1 +17: stb %r1, 0(%r26) + b lws_exit_noerror + copy %r0, %r28 + nop + nop + + /* 16-bit exchange */ +18: ldh 0(%r26), %r1 +19: sth %r1, 0(%r23) +20: ldh 0(%r24), %r1 +21: sth %r1, 0(%r26) + b lws_exit_noerror + copy %r0, %r28 + nop + nop + + /* 32-bit exchange */ +22: ldw 0(%r26), %r1 +23: stw %r1, 0(%r23) +24: ldw 0(%r24), %r1 +25: stw %r1, 0(%r26) + b lws_exit_noerror + copy %r0, %r28 + nop + nop + + /* 64-bit exchange */ +#ifdef CONFIG_64BIT +26: ldd 0(%r26), %r1 +27: std %r1, 0(%r23) +28: ldd 0(%r24), %r1 +29: std %r1, 0(%r26) +#else +26: flddx 0(%r26), %fr4 +27: fstdx %fr4, 0(%r23) +28: flddx 0(%r24), %fr4 +29: fstdx %fr4, 0(%r26) +#endif + b lws_exit_noerror + copy %r0, %r28 + + /* A fault occurred on load or stbys,e store */ +30: b,n lws_fault + ASM_EXCEPTIONTABLE_ENTRY(1b-linux_gateway_page, 30b-linux_gateway_page) + ASM_EXCEPTIONTABLE_ENTRY(2b-linux_gateway_page, 30b-linux_gateway_page) + ASM_EXCEPTIONTABLE_ENTRY(3b-linux_gateway_page, 30b-linux_gateway_page) + ASM_EXCEPTIONTABLE_ENTRY(4b-linux_gateway_page, 30b-linux_gateway_page) + ASM_EXCEPTIONTABLE_ENTRY(5b-linux_gateway_page, 30b-linux_gateway_page) + ASM_EXCEPTIONTABLE_ENTRY(6b-linux_gateway_page, 30b-linux_gateway_page) + ASM_EXCEPTIONTABLE_ENTRY(7b-linux_gateway_page, 30b-linux_gateway_page) + ASM_EXCEPTIONTABLE_ENTRY(8b-linux_gateway_page, 30b-linux_gateway_page) +#ifndef CONFIG_64BIT + ASM_EXCEPTIONTABLE_ENTRY(9b-linux_gateway_page, 30b-linux_gateway_page) + ASM_EXCEPTIONTABLE_ENTRY(10b-linux_gateway_page, 30b-linux_gateway_page) +#endif + + ASM_EXCEPTIONTABLE_ENTRY(11b-linux_gateway_page, 30b-linux_gateway_page) + ASM_EXCEPTIONTABLE_ENTRY(12b-linux_gateway_page, 30b-linux_gateway_page) + + /* A page fault occurred in critical region */ +31: b,n lws_pagefault + ASM_EXCEPTIONTABLE_ENTRY(14b-linux_gateway_page, 31b-linux_gateway_page) + ASM_EXCEPTIONTABLE_ENTRY(15b-linux_gateway_page, 31b-linux_gateway_page) + ASM_EXCEPTIONTABLE_ENTRY(16b-linux_gateway_page, 31b-linux_gateway_page) + ASM_EXCEPTIONTABLE_ENTRY(17b-linux_gateway_page, 31b-linux_gateway_page) + ASM_EXCEPTIONTABLE_ENTRY(18b-linux_gateway_page, 31b-linux_gateway_page) + ASM_EXCEPTIONTABLE_ENTRY(19b-linux_gateway_page, 31b-linux_gateway_page) + ASM_EXCEPTIONTABLE_ENTRY(20b-linux_gateway_page, 31b-linux_gateway_page) + ASM_EXCEPTIONTABLE_ENTRY(21b-linux_gateway_page, 31b-linux_gateway_page) + ASM_EXCEPTIONTABLE_ENTRY(22b-linux_gateway_page, 31b-linux_gateway_page) + ASM_EXCEPTIONTABLE_ENTRY(23b-linux_gateway_page, 31b-linux_gateway_page) + ASM_EXCEPTIONTABLE_ENTRY(24b-linux_gateway_page, 31b-linux_gateway_page) + ASM_EXCEPTIONTABLE_ENTRY(25b-linux_gateway_page, 31b-linux_gateway_page) + ASM_EXCEPTIONTABLE_ENTRY(26b-linux_gateway_page, 31b-linux_gateway_page) + ASM_EXCEPTIONTABLE_ENTRY(27b-linux_gateway_page, 31b-linux_gateway_page) + ASM_EXCEPTIONTABLE_ENTRY(28b-linux_gateway_page, 31b-linux_gateway_page) + ASM_EXCEPTIONTABLE_ENTRY(29b-linux_gateway_page, 31b-linux_gateway_page) + + /*************************************************** + LWS atomic store. + + %r26 - Address to store + %r25 - Size of the variable (0/1/2/3 for 8/16/32/64 bit) + %r24 - Address of value to store + %r28 - Return non-zero on failure + %r21 - Kernel error code + + %r21 returns the following error codes: + EAGAIN - CAS is busy, ldcw failed, try again. + EFAULT - Read or write failed. + + If EAGAIN is returned, %r28 indicates the busy reason: + r28 == 1 - CAS is busy. lock contended. + r28 == 2 - CAS is busy. ldcw failed. + r28 == 3 - CAS is busy. page fault. + + Scratch: r20, r1 + + ****************************************************/ + +lws_atomic_store: +#ifdef CONFIG_64BIT + /* Wide mode user process? */ + bb,<,n %sp, 31, atomic_store_begin + + /* Clip the input registers for 32-bit processes. We don't + need to clip %r23 as we only use it for word operations */ + depdi 0, 31, 32, %r26 + depdi 0, 31, 32, %r25 + depdi 0, 31, 32, %r24 +#endif + +atomic_store_begin: + /* Check the validity of the size pointer */ + subi,>>= 3, %r25, %r0 + b,n lws_exit_nosys + + shlw %r25, 1, %r1 + blr %r1, %r0 + nop + + /* Perform exception checks */ + + /* 8-bit store */ +1: ldb 0(%r24), %r20 + b,n atomic_store_start + nop + nop + + /* 16-bit store */ +2: ldh 0(%r24), %r20 + b,n atomic_store_start + nop + nop + + /* 32-bit store */ +3: ldw 0(%r24), %r20 + b,n atomic_store_start + nop + nop + + /* 64-bit store */ +#ifdef CONFIG_64BIT +4: ldd 0(%r24), %r20 +#else +4: ldw 0(%r24), %r20 +5: ldw 4(%r24), %r20 +#endif + +atomic_store_start: + /* Trigger memory reference interruptions without writing to memory */ + copy %r26, %r28 + depi_safe 0, 31, 2, %r28 +6: ldw 0(%r28), %r1 +7: stbys,e %r0, 0(%r28) + + /* Calculate 8-bit hash index from virtual address */ + extru_safe %r26, 27, 8, %r20 + + /* Load start of lock table */ + ldil L%lws_lock_start, %r28 + ldo R%lws_lock_start(%r28), %r28 + + /* Find lock to use, the hash index is one of 0 to + 255, multiplied by 16 (keep it 16-byte aligned) + and add to the lock table offset. */ + shlw %r20, 4, %r20 + add %r20, %r28, %r20 + + rsm PSW_SM_I, %r0 /* Disable interrupts */ + + /* Try to acquire the lock */ + LDCW 0(%sr2,%r20), %r28 + spinlock_check %r28, %r21 + comclr,<> %r0, %r28, %r0 + b,n lws_wouldblock + + /* Disable page faults to prevent sleeping in critical region */ + lws_pagefault_disable %r21,%r28 + + /* NOTES: + This all works because intr_do_signal + and schedule both check the return iasq + and see that we are on the kernel page + so this process is never scheduled off + or is ever sent any signal of any sort, + thus it is wholly atomic from userspace's + perspective + */ + + /* Jump to the correct function */ + blr %r1, %r0 + /* Set %r28 as non-zero for now */ + ldo 1(%r0),%r28 + + /* 8-bit store */ +9: ldb 0(%r24), %r1 +10: stb %r1, 0(%r26) + b lws_exit_noerror + copy %r0, %r28 + + /* 16-bit store */ +11: ldh 0(%r24), %r1 +12: sth %r1, 0(%r26) + b lws_exit_noerror + copy %r0, %r28 + + /* 32-bit store */ +13: ldw 0(%r24), %r1 +14: stw %r1, 0(%r26) + b lws_exit_noerror + copy %r0, %r28 + + /* 64-bit store */ +#ifdef CONFIG_64BIT +15: ldd 0(%r24), %r1 +16: std %r1, 0(%r26) +#else +15: flddx 0(%r24), %fr4 +16: fstdx %fr4, 0(%r26) +#endif + b lws_exit_noerror + copy %r0, %r28 + + /* A fault occurred on load or stbys,e store */ +30: b,n lws_fault + ASM_EXCEPTIONTABLE_ENTRY(1b-linux_gateway_page, 30b-linux_gateway_page) + ASM_EXCEPTIONTABLE_ENTRY(2b-linux_gateway_page, 30b-linux_gateway_page) + ASM_EXCEPTIONTABLE_ENTRY(3b-linux_gateway_page, 30b-linux_gateway_page) + ASM_EXCEPTIONTABLE_ENTRY(4b-linux_gateway_page, 30b-linux_gateway_page) #ifndef CONFIG_64BIT - ASM_EXCEPTIONTABLE_ENTRY(12b-linux_gateway_page, 22b-linux_gateway_page) - ASM_EXCEPTIONTABLE_ENTRY(21b-linux_gateway_page, 22b-linux_gateway_page) + ASM_EXCEPTIONTABLE_ENTRY(5b-linux_gateway_page, 30b-linux_gateway_page) #endif + ASM_EXCEPTIONTABLE_ENTRY(6b-linux_gateway_page, 30b-linux_gateway_page) + ASM_EXCEPTIONTABLE_ENTRY(7b-linux_gateway_page, 30b-linux_gateway_page) + + /* A page fault occurred in critical region */ +31: b,n lws_pagefault + ASM_EXCEPTIONTABLE_ENTRY(9b-linux_gateway_page, 31b-linux_gateway_page) + ASM_EXCEPTIONTABLE_ENTRY(10b-linux_gateway_page, 31b-linux_gateway_page) + ASM_EXCEPTIONTABLE_ENTRY(11b-linux_gateway_page, 31b-linux_gateway_page) + ASM_EXCEPTIONTABLE_ENTRY(12b-linux_gateway_page, 31b-linux_gateway_page) + ASM_EXCEPTIONTABLE_ENTRY(13b-linux_gateway_page, 31b-linux_gateway_page) + ASM_EXCEPTIONTABLE_ENTRY(14b-linux_gateway_page, 31b-linux_gateway_page) + ASM_EXCEPTIONTABLE_ENTRY(15b-linux_gateway_page, 31b-linux_gateway_page) + ASM_EXCEPTIONTABLE_ENTRY(16b-linux_gateway_page, 31b-linux_gateway_page) + /* Make sure nothing else is placed on this page */ .align PAGE_SIZE END(linux_gateway_page) @@ -931,28 +1308,30 @@ ENTRY(end_linux_gateway_page) ENTRY(lws_table) LWS_ENTRY(compare_and_swap32) /* 0 - ELF32 Atomic 32bit CAS */ LWS_ENTRY(compare_and_swap64) /* 1 - ELF64 Atomic 32bit CAS */ - LWS_ENTRY(compare_and_swap_2) /* 2 - ELF32 Atomic 64bit CAS */ + LWS_ENTRY(compare_and_swap_2) /* 2 - Atomic 64bit CAS */ + LWS_ENTRY(atomic_xchg) /* 3 - Atomic Exchange */ + LWS_ENTRY(atomic_store) /* 4 - Atomic Store */ END(lws_table) /* End of lws table */ -#define __SYSCALL(nr, entry, nargs) ASM_ULONG_INSN entry - .align 8 -ENTRY(sys_call_table) - .export sys_call_table,data #ifdef CONFIG_64BIT -#include <asm/syscall_table_c32.h> /* Compat syscalls */ +#define __SYSCALL_WITH_COMPAT(nr, native, compat) __SYSCALL(nr, compat) #else -#include <asm/syscall_table_32.h> /* 32-bit native syscalls */ +#define __SYSCALL_WITH_COMPAT(nr, native, compat) __SYSCALL(nr, native) #endif +#define __SYSCALL(nr, entry) ASM_ULONG_INSN entry + .align 8 +ENTRY(sys_call_table) + .export sys_call_table,data +#include <asm/syscall_table_32.h> /* 32-bit syscalls */ END(sys_call_table) #ifdef CONFIG_64BIT .align 8 ENTRY(sys_call_table64) -#include <asm/syscall_table_64.h> /* 64-bit native syscalls */ +#include <asm/syscall_table_64.h> /* 64-bit syscalls */ END(sys_call_table64) #endif -#undef __SYSCALL /* All light-weight-syscall atomic operations @@ -966,9 +1345,9 @@ END(sys_call_table64) .align L1_CACHE_BYTES ENTRY(lws_lock_start) /* lws locks */ - .rept 16 + .rept 256 /* Keep locks aligned at 16-bytes */ - .word 1 + .word __ARCH_SPIN_LOCK_UNLOCKED_VAL .word 0 .word 0 .word 0 @@ -977,5 +1356,3 @@ END(lws_lock_start) .previous .end - - diff --git a/arch/parisc/kernel/syscalls/Makefile b/arch/parisc/kernel/syscalls/Makefile index c22a21c39f30..8440c16dfb22 100644 --- a/arch/parisc/kernel/syscalls/Makefile +++ b/arch/parisc/kernel/syscalls/Makefile @@ -2,54 +2,32 @@ kapi := arch/$(SRCARCH)/include/generated/asm uapi := arch/$(SRCARCH)/include/generated/uapi/asm -_dummy := $(shell [ -d '$(uapi)' ] || mkdir -p '$(uapi)') \ - $(shell [ -d '$(kapi)' ] || mkdir -p '$(kapi)') +$(shell mkdir -p $(uapi) $(kapi)) -syscall := $(srctree)/$(src)/syscall.tbl -syshdr := $(srctree)/$(src)/syscallhdr.sh -systbl := $(srctree)/$(src)/syscalltbl.sh +syscall := $(src)/syscall.tbl +syshdr := $(srctree)/scripts/syscallhdr.sh +systbl := $(srctree)/scripts/syscalltbl.sh quiet_cmd_syshdr = SYSHDR $@ - cmd_syshdr = $(CONFIG_SHELL) '$(syshdr)' '$<' '$@' \ - '$(syshdr_abis_$(basetarget))' \ - '$(syshdr_pfx_$(basetarget))' \ - '$(syshdr_offset_$(basetarget))' + cmd_syshdr = $(CONFIG_SHELL) $(syshdr) --emit-nr --abis common,$* $< $@ quiet_cmd_systbl = SYSTBL $@ - cmd_systbl = $(CONFIG_SHELL) '$(systbl)' '$<' '$@' \ - '$(systbl_abis_$(basetarget))' \ - '$(systbl_abi_$(basetarget))' \ - '$(systbl_offset_$(basetarget))' + cmd_systbl = $(CONFIG_SHELL) $(systbl) --abis common,$* $< $@ -syshdr_abis_unistd_32 := common,32 -$(uapi)/unistd_32.h: $(syscall) $(syshdr) +$(uapi)/unistd_%.h: $(syscall) $(syshdr) FORCE $(call if_changed,syshdr) -syshdr_abis_unistd_64 := common,64 -$(uapi)/unistd_64.h: $(syscall) $(syshdr) - $(call if_changed,syshdr) - -systbl_abis_syscall_table_32 := common,32 -$(kapi)/syscall_table_32.h: $(syscall) $(systbl) - $(call if_changed,systbl) - -systbl_abis_syscall_table_64 := common,64 -$(kapi)/syscall_table_64.h: $(syscall) $(systbl) - $(call if_changed,systbl) - -systbl_abis_syscall_table_c32 := common,32 -systbl_abi_syscall_table_c32 := c32 -$(kapi)/syscall_table_c32.h: $(syscall) $(systbl) +$(kapi)/syscall_table_%.h: $(syscall) $(systbl) FORCE $(call if_changed,systbl) uapisyshdr-y += unistd_32.h unistd_64.h kapisyshdr-y += syscall_table_32.h \ - syscall_table_64.h \ - syscall_table_c32.h + syscall_table_64.h -targets += $(uapisyshdr-y) $(kapisyshdr-y) +uapisyshdr-y := $(addprefix $(uapi)/, $(uapisyshdr-y)) +kapisyshdr-y := $(addprefix $(kapi)/, $(kapisyshdr-y)) +targets += $(addprefix ../../../../, $(uapisyshdr-y) $(kapisyshdr-y)) PHONY += all -all: $(addprefix $(uapi)/,$(uapisyshdr-y)) -all: $(addprefix $(kapi)/,$(kapisyshdr-y)) +all: $(uapisyshdr-y) $(kapisyshdr-y) @: diff --git a/arch/parisc/kernel/syscalls/syscall.tbl b/arch/parisc/kernel/syscalls/syscall.tbl index 285ff516150c..b236a84c4e12 100644 --- a/arch/parisc/kernel/syscalls/syscall.tbl +++ b/arch/parisc/kernel/syscalls/syscall.tbl @@ -29,7 +29,7 @@ 18 common stat sys_newstat compat_sys_newstat 19 common lseek sys_lseek compat_sys_lseek 20 common getpid sys_getpid -21 common mount sys_mount compat_sys_mount +21 common mount sys_mount 22 common bind sys_bind 23 common setuid sys_setuid 24 common getuid sys_getuid @@ -131,7 +131,7 @@ 116 common sysinfo sys_sysinfo compat_sys_sysinfo 117 common shutdown sys_shutdown 118 common fsync sys_fsync -119 common madvise sys_madvise +119 common madvise parisc_madvise 120 common clone sys_clone_wrapper 121 common setdomainname sys_setdomainname 122 common sendfile sys_sendfile compat_sys_sendfile @@ -147,7 +147,7 @@ 131 common quotactl sys_quotactl 132 common getpgid sys_getpgid 133 common fchdir sys_fchdir -134 common bdflush sys_bdflush +134 common bdflush sys_ni_syscall 135 common sysfs sys_sysfs 136 32 personality parisc_personality 136 64 personality sys_personality @@ -159,11 +159,11 @@ 142 common _newselect sys_select compat_sys_select 143 common flock sys_flock 144 common msync sys_msync -145 common readv sys_readv compat_sys_readv -146 common writev sys_writev compat_sys_writev +145 common readv sys_readv +146 common writev sys_writev 147 common getsid sys_getsid 148 common fdatasync sys_fdatasync -149 common _sysctl sys_sysctl compat_sys_sysctl +149 common _sysctl sys_ni_syscall 150 common mlock sys_mlock 151 common munlock sys_munlock 152 common mlockall sys_mlockall @@ -198,8 +198,8 @@ 178 common rt_sigqueueinfo sys_rt_sigqueueinfo compat_sys_rt_sigqueueinfo 179 common rt_sigsuspend sys_rt_sigsuspend compat_sys_rt_sigsuspend 180 common chown sys_chown -181 common setsockopt sys_setsockopt compat_sys_setsockopt -182 common getsockopt sys_getsockopt compat_sys_getsockopt +181 common setsockopt sys_setsockopt sys_setsockopt +182 common getsockopt sys_getsockopt sys_getsockopt 183 common sendmsg sys_sendmsg compat_sys_sendmsg 184 common recvmsg sys_recvmsg compat_sys_recvmsg 185 common semop sys_semop @@ -245,7 +245,7 @@ # 220 was alloc_hugepages # 221 was free_hugepages 222 common exit_group sys_exit_group -223 common lookup_dcookie sys_lookup_dcookie compat_sys_lookup_dcookie +223 common lookup_dcookie sys_ni_syscall 224 common epoll_create sys_epoll_create 225 common epoll_ctl sys_epoll_ctl 226 common epoll_wait sys_epoll_wait @@ -292,9 +292,9 @@ 258 32 clock_nanosleep sys_clock_nanosleep_time32 258 64 clock_nanosleep sys_clock_nanosleep 259 common tgkill sys_tgkill -260 common mbind sys_mbind compat_sys_mbind -261 common get_mempolicy sys_get_mempolicy compat_sys_get_mempolicy -262 common set_mempolicy sys_set_mempolicy compat_sys_set_mempolicy +260 common mbind sys_mbind +261 common get_mempolicy sys_get_mempolicy +262 common set_mempolicy sys_set_mempolicy # 263 was vserver 264 common add_key sys_add_key 265 common request_key sys_request_key @@ -330,8 +330,8 @@ 292 32 sync_file_range parisc_sync_file_range 292 64 sync_file_range sys_sync_file_range 293 common tee sys_tee -294 common vmsplice sys_vmsplice compat_sys_vmsplice -295 common move_pages sys_move_pages compat_sys_move_pages +294 common vmsplice sys_vmsplice +295 common move_pages sys_move_pages 296 common getcpu sys_getcpu 297 common epoll_pwait sys_epoll_pwait compat_sys_epoll_pwait 298 common statfs64 sys_statfs64 compat_sys_statfs64 @@ -344,17 +344,17 @@ 304 common eventfd sys_eventfd 305 32 fallocate parisc_fallocate 305 64 fallocate sys_fallocate -306 common timerfd_create sys_timerfd_create +306 common timerfd_create parisc_timerfd_create 307 32 timerfd_settime sys_timerfd_settime32 307 64 timerfd_settime sys_timerfd_settime 308 32 timerfd_gettime sys_timerfd_gettime32 308 64 timerfd_gettime sys_timerfd_gettime -309 common signalfd4 sys_signalfd4 compat_sys_signalfd4 -310 common eventfd2 sys_eventfd2 +309 common signalfd4 parisc_signalfd4 parisc_compat_signalfd4 +310 common eventfd2 parisc_eventfd2 311 common epoll_create1 sys_epoll_create1 312 common dup3 sys_dup3 -313 common pipe2 sys_pipe2 -314 common inotify_init1 sys_inotify_init1 +313 common pipe2 parisc_pipe2 +314 common inotify_init1 parisc_inotify_init1 315 common preadv sys_preadv compat_sys_preadv 316 common pwritev sys_pwritev compat_sys_pwritev 317 common rt_tgsigqueueinfo sys_rt_tgsigqueueinfo compat_sys_rt_tgsigqueueinfo @@ -372,8 +372,8 @@ 327 common syncfs sys_syncfs 328 common setns sys_setns 329 common sendmmsg sys_sendmmsg compat_sys_sendmmsg -330 common process_vm_readv sys_process_vm_readv compat_sys_process_vm_readv -331 common process_vm_writev sys_process_vm_writev compat_sys_process_vm_writev +330 common process_vm_readv sys_process_vm_readv +331 common process_vm_writev sys_process_vm_writev 332 common kcmp sys_kcmp 333 common finit_module sys_finit_module 334 common sched_setattr sys_sched_setattr @@ -387,7 +387,7 @@ 341 common bpf sys_bpf 342 common execveat sys_execveat compat_sys_execveat 343 common membarrier sys_membarrier -344 common userfaultfd sys_userfaultfd +344 common userfaultfd parisc_userfaultfd 345 common mlock2 sys_mlock2 346 common copy_file_range sys_copy_file_range 347 common preadv2 sys_preadv2 compat_sys_preadv2 @@ -400,6 +400,7 @@ 353 common pkey_free sys_pkey_free 354 common rseq sys_rseq 355 common kexec_file_load sys_kexec_file_load sys_kexec_file_load +356 common cacheflush sys_cacheflush # up to 402 is unassigned and reserved for arch specific syscalls 403 32 clock_gettime64 sys_clock_gettime sys_clock_gettime 404 32 clock_settime64 sys_clock_settime sys_clock_settime @@ -413,7 +414,7 @@ 412 32 utimensat_time64 sys_utimensat sys_utimensat 413 32 pselect6_time64 sys_pselect6 compat_sys_pselect6_time64 414 32 ppoll_time64 sys_ppoll compat_sys_ppoll_time64 -416 32 io_pgetevents_time64 sys_io_pgetevents sys_io_pgetevents +416 32 io_pgetevents_time64 sys_io_pgetevents compat_sys_io_pgetevents_time64 417 32 recvmmsg_time64 sys_recvmmsg compat_sys_recvmmsg_time64 418 32 mq_timedsend_time64 sys_mq_timedsend sys_mq_timedsend 419 32 mq_timedreceive_time64 sys_mq_timedreceive sys_mq_timedreceive @@ -433,3 +434,29 @@ 433 common fspick sys_fspick 434 common pidfd_open sys_pidfd_open 435 common clone3 sys_clone3_wrapper +436 common close_range sys_close_range +437 common openat2 sys_openat2 +438 common pidfd_getfd sys_pidfd_getfd +439 common faccessat2 sys_faccessat2 +440 common process_madvise sys_process_madvise +441 common epoll_pwait2 sys_epoll_pwait2 compat_sys_epoll_pwait2 +442 common mount_setattr sys_mount_setattr +443 common quotactl_fd sys_quotactl_fd +444 common landlock_create_ruleset sys_landlock_create_ruleset +445 common landlock_add_rule sys_landlock_add_rule +446 common landlock_restrict_self sys_landlock_restrict_self +# 447 reserved for memfd_secret +448 common process_mrelease sys_process_mrelease +449 common futex_waitv sys_futex_waitv +450 common set_mempolicy_home_node sys_set_mempolicy_home_node +451 common cachestat sys_cachestat +452 common fchmodat2 sys_fchmodat2 +453 common map_shadow_stack sys_map_shadow_stack +454 common futex_wake sys_futex_wake +455 common futex_wait sys_futex_wait +456 common futex_requeue sys_futex_requeue +457 common statmount sys_statmount +458 common listmount sys_listmount +459 common lsm_get_self_attr sys_lsm_get_self_attr +460 common lsm_set_self_attr sys_lsm_set_self_attr +461 common lsm_list_modules sys_lsm_list_modules diff --git a/arch/parisc/kernel/syscalls/syscallhdr.sh b/arch/parisc/kernel/syscalls/syscallhdr.sh deleted file mode 100644 index 50242b747d7c..000000000000 --- a/arch/parisc/kernel/syscalls/syscallhdr.sh +++ /dev/null @@ -1,36 +0,0 @@ -#!/bin/sh -# SPDX-License-Identifier: GPL-2.0 - -in="$1" -out="$2" -my_abis=`echo "($3)" | tr ',' '|'` -prefix="$4" -offset="$5" - -fileguard=_UAPI_ASM_PARISC_`basename "$out" | sed \ - -e 'y/abcdefghijklmnopqrstuvwxyz/ABCDEFGHIJKLMNOPQRSTUVWXYZ/' \ - -e 's/[^A-Z0-9_]/_/g' -e 's/__/_/g'` -grep -E "^[0-9A-Fa-fXx]+[[:space:]]+${my_abis}" "$in" | sort -n | ( - printf "#ifndef %s\n" "${fileguard}" - printf "#define %s\n" "${fileguard}" - printf "\n" - - nxt=0 - while read nr abi name entry compat ; do - if [ -z "$offset" ]; then - printf "#define __NR_%s%s\t%s\n" \ - "${prefix}" "${name}" "${nr}" - else - printf "#define __NR_%s%s\t(%s + %s)\n" \ - "${prefix}" "${name}" "${offset}" "${nr}" - fi - nxt=$((nr+1)) - done - - printf "\n" - printf "#ifdef __KERNEL__\n" - printf "#define __NR_syscalls\t%s\n" "${nxt}" - printf "#endif\n" - printf "\n" - printf "#endif /* %s */" "${fileguard}" -) > "$out" diff --git a/arch/parisc/kernel/syscalls/syscalltbl.sh b/arch/parisc/kernel/syscalls/syscalltbl.sh deleted file mode 100644 index 45b5bae26240..000000000000 --- a/arch/parisc/kernel/syscalls/syscalltbl.sh +++ /dev/null @@ -1,36 +0,0 @@ -#!/bin/sh -# SPDX-License-Identifier: GPL-2.0 - -in="$1" -out="$2" -my_abis=`echo "($3)" | tr ',' '|'` -my_abi="$4" -offset="$5" - -emit() { - t_nxt="$1" - t_nr="$2" - t_entry="$3" - - while [ $t_nxt -lt $t_nr ]; do - printf "__SYSCALL(%s, sys_ni_syscall, )\n" "${t_nxt}" - t_nxt=$((t_nxt+1)) - done - printf "__SYSCALL(%s, %s, )\n" "${t_nxt}" "${t_entry}" -} - -grep -E "^[0-9A-Fa-fXx]+[[:space:]]+${my_abis}" "$in" | sort -n | ( - nxt=0 - if [ -z "$offset" ]; then - offset=0 - fi - - while read nr abi name entry compat ; do - if [ "$my_abi" = "c32" ] && [ ! -z "$compat" ]; then - emit $((nxt+offset)) $((nr+offset)) $compat - else - emit $((nxt+offset)) $((nr+offset)) $entry - fi - nxt=$((nr+1)) - done -) > "$out" diff --git a/arch/parisc/kernel/time.c b/arch/parisc/kernel/time.c index 04508158815c..9714fbd7c42d 100644 --- a/arch/parisc/kernel/time.c +++ b/arch/parisc/kernel/time.c @@ -40,6 +40,8 @@ #include <linux/timex.h> +int time_keeper_id __read_mostly; /* CPU used for timekeeping. */ + static unsigned long clocktick __ro_after_init; /* timer cycles per tick */ /* @@ -70,8 +72,6 @@ irqreturn_t __irq_entry timer_interrupt(int irq, void *dev_id) /* gcc can optimize for "read-only" case with a local clocktick */ unsigned long cpt = clocktick; - profile_tick(CPU_PROFILING); - /* Initialize next_tick to the old expected tick time. */ next_tick = cpuinfo->it_value; @@ -86,10 +86,9 @@ irqreturn_t __irq_entry timer_interrupt(int irq, void *dev_id) cpuinfo->it_value = next_tick; /* Go do system house keeping. */ - if (cpu == 0) - xtime_update(ticks_elapsed); - - update_process_times(user_mode(get_irq_regs())); + if (IS_ENABLED(CONFIG_SMP) && (cpu != time_keeper_id)) + ticks_elapsed = 0; + legacy_timer_tick(ticks_elapsed); /* Skip clockticks on purpose if we know we would miss those. * The new CR16 must be "later" than current CR16 otherwise @@ -153,7 +152,7 @@ static struct clocksource clocksource_cr16 = { .flags = CLOCK_SOURCE_IS_CONTINUOUS, }; -void __init start_cpu_itimer(void) +void start_cpu_itimer(void) { unsigned int cpu = smp_processor_id(); unsigned long next_tick = mfctl(16) + clocktick; @@ -180,9 +179,16 @@ static int rtc_generic_get_time(struct device *dev, struct rtc_time *tm) static int rtc_generic_set_time(struct device *dev, struct rtc_time *tm) { time64_t secs = rtc_tm_to_time64(tm); - - if (pdc_tod_set(secs, 0) < 0) + int ret; + + /* hppa has Y2K38 problem: pdc_tod_set() takes an u32 value! */ + ret = pdc_tod_set(secs, 0); + if (ret != 0) { + pr_warn("pdc_tod_set(%lld) returned error %d\n", secs, ret); + if (ret == PDC_INVALID_ARG) + return -EINVAL; return -EOPNOTSUPP; + } return 0; } @@ -245,33 +251,14 @@ void __init time_init(void) static int __init init_cr16_clocksource(void) { /* - * The cr16 interval timers are not syncronized across CPUs on - * different sockets, so mark them unstable and lower rating on - * multi-socket SMP systems. + * The cr16 interval timers are not synchronized across CPUs. */ if (num_online_cpus() > 1 && !running_on_qemu) { - int cpu; - unsigned long cpu0_loc; - cpu0_loc = per_cpu(cpu_data, 0).cpu_loc; - - for_each_online_cpu(cpu) { - if (cpu == 0) - continue; - if ((cpu0_loc != 0) && - (cpu0_loc == per_cpu(cpu_data, cpu).cpu_loc)) - continue; - - clocksource_cr16.name = "cr16_unstable"; - clocksource_cr16.flags = CLOCK_SOURCE_UNSTABLE; - clocksource_cr16.rating = 0; - break; - } + clocksource_cr16.name = "cr16_unstable"; + clocksource_cr16.flags = CLOCK_SOURCE_UNSTABLE; + clocksource_cr16.rating = 0; } - /* XXX: We may want to mark sched_clock stable here if cr16 clocks are - * in sync: - * (clocksource_cr16.flags == CLOCK_SOURCE_IS_CONTINUOUS) */ - /* register at clocksource framework */ clocksource_register_hz(&clocksource_cr16, 100 * PAGE0->mem_10msec); diff --git a/arch/parisc/kernel/toc.c b/arch/parisc/kernel/toc.c new file mode 100644 index 000000000000..e4b48d07afbd --- /dev/null +++ b/arch/parisc/kernel/toc.c @@ -0,0 +1,126 @@ +// SPDX-License-Identifier: GPL-2.0 + +#include <linux/kernel.h> +#include <linux/kgdb.h> +#include <linux/printk.h> +#include <linux/sched/debug.h> +#include <linux/delay.h> +#include <linux/reboot.h> + +#include <asm/pdc.h> +#include <asm/pdc_chassis.h> +#include <asm/ldcw.h> +#include <asm/processor.h> + +static unsigned int __aligned(16) toc_lock = 1; +DEFINE_PER_CPU_PAGE_ALIGNED(char [16384], toc_stack) __visible; + +static void toc20_to_pt_regs(struct pt_regs *regs, struct pdc_toc_pim_20 *toc) +{ + int i; + + regs->gr[0] = (unsigned long)toc->cr[22]; + + for (i = 1; i < 32; i++) + regs->gr[i] = (unsigned long)toc->gr[i]; + + for (i = 0; i < 8; i++) + regs->sr[i] = (unsigned long)toc->sr[i]; + + regs->iasq[0] = (unsigned long)toc->cr[17]; + regs->iasq[1] = (unsigned long)toc->iasq_back; + regs->iaoq[0] = (unsigned long)toc->cr[18]; + regs->iaoq[1] = (unsigned long)toc->iaoq_back; + + regs->sar = (unsigned long)toc->cr[11]; + regs->iir = (unsigned long)toc->cr[19]; + regs->isr = (unsigned long)toc->cr[20]; + regs->ior = (unsigned long)toc->cr[21]; +} + +static void toc11_to_pt_regs(struct pt_regs *regs, struct pdc_toc_pim_11 *toc) +{ + int i; + + regs->gr[0] = toc->cr[22]; + + for (i = 1; i < 32; i++) + regs->gr[i] = toc->gr[i]; + + for (i = 0; i < 8; i++) + regs->sr[i] = toc->sr[i]; + + regs->iasq[0] = toc->cr[17]; + regs->iasq[1] = toc->iasq_back; + regs->iaoq[0] = toc->cr[18]; + regs->iaoq[1] = toc->iaoq_back; + + regs->sar = toc->cr[11]; + regs->iir = toc->cr[19]; + regs->isr = toc->cr[20]; + regs->ior = toc->cr[21]; +} + +void notrace __noreturn __cold toc_intr(struct pt_regs *regs) +{ + struct pdc_toc_pim_20 pim_data20; + struct pdc_toc_pim_11 pim_data11; + + /* verify we wrote regs to the correct stack */ + BUG_ON(regs != (struct pt_regs *)&per_cpu(toc_stack, raw_smp_processor_id())); + + if (boot_cpu_data.cpu_type >= pcxu) { + if (pdc_pim_toc20(&pim_data20)) + panic("Failed to get PIM data"); + toc20_to_pt_regs(regs, &pim_data20); + } else { + if (pdc_pim_toc11(&pim_data11)) + panic("Failed to get PIM data"); + toc11_to_pt_regs(regs, &pim_data11); + } + +#ifdef CONFIG_KGDB + nmi_enter(); + + if (atomic_read(&kgdb_active) != -1) + kgdb_nmicallback(raw_smp_processor_id(), regs); + kgdb_handle_exception(9, SIGTRAP, 0, regs); +#endif + + /* serialize output, otherwise all CPUs write backtrace at once */ + while (__ldcw(&toc_lock) == 0) + ; /* wait */ + show_regs(regs); + toc_lock = 1; /* release lock for next CPU */ + + if (raw_smp_processor_id() != 0) + while (1) ; /* all but monarch CPU will wait endless. */ + + /* give other CPUs time to show their backtrace */ + mdelay(2000); + + machine_restart("TOC"); + + /* should never reach this */ + panic("TOC"); +} + +static __init int setup_toc(void) +{ + unsigned int csum = 0; + unsigned long toc_code = (unsigned long)dereference_function_descriptor(toc_handler); + int i; + + PAGE0->vec_toc = __pa(toc_code) & 0xffffffff; +#ifdef CONFIG_64BIT + PAGE0->vec_toc_hi = __pa(toc_code) >> 32; +#endif + PAGE0->vec_toclen = toc_handler_size; + + for (i = 0; i < toc_handler_size/4; i++) + csum += ((u32 *)toc_code)[i]; + toc_handler_csum = -csum; + pr_info("TOC handler registered\n"); + return 0; +} +early_initcall(setup_toc); diff --git a/arch/parisc/kernel/toc_asm.S b/arch/parisc/kernel/toc_asm.S new file mode 100644 index 000000000000..570f5cef526e --- /dev/null +++ b/arch/parisc/kernel/toc_asm.S @@ -0,0 +1,75 @@ +/* SPDX-License-Identifier: GPL-2.0-or-later */ + +/* TOC (Transfer of Control) handler. */ + + .level 1.1 + +#include <asm/assembly.h> +#include <linux/threads.h> +#include <linux/linkage.h> + + .text + .import toc_intr,code + .import toc_stack,data + .align 16 +ENTRY_CFI(toc_handler) + load32 PA(toc_stack),%sp + +#ifdef CONFIG_SMP + /* get per-cpu toc_stack address. */ + mfctl %cr30, %r1 + tophys %r1,%r2 /* task_struct */ + LDREG TASK_TI_CPU(%r2),%r4 /* cpu */ + load32 PA(__per_cpu_offset),%r1 + LDREGX %r4(%r1),%r4 + add %r4,%sp,%sp +#endif + + /* + * setup pt_regs on stack and save the + * floating point registers. PIM_TOC doesn't + * save fp registers, so we're doing it here. + */ + copy %sp,%arg0 + ldo PT_SZ_ALGN(%sp), %sp + + /* clear pt_regs */ + copy %arg0,%r1 +0: cmpb,<<,n %r1,%sp,0b + stw,ma %r0,4(%r1) + + ldo PT_FR0(%arg0),%r25 + save_fp %r25 + + /* go virtual */ + load32 PA(swapper_pg_dir),%r4 + mtctl %r4,%cr24 + mtctl %r4,%cr25 + + /* Clear sr4-sr7 */ + mtsp %r0, %sr4 + mtsp %r0, %sr5 + mtsp %r0, %sr6 + mtsp %r0, %sr7 + + tovirt_r1 %sp + tovirt_r1 %arg0 + virt_map + + loadgp + +#ifdef CONFIG_64BIT + ldo -16(%sp),%r29 +#endif + load32 toc_intr,%r1 + be 0(%sr7,%r1) + nop +ENDPROC_CFI(toc_handler) + + /* + * keep this checksum here, as it is part of the toc_handler + * spanned by toc_handler_size (all words in toc_handler are + * added in PDC and the sum must equal to zero. + */ +SYM_DATA(toc_handler_csum, .long 0) +SYM_DATA(toc_handler_size, .long . - toc_handler) diff --git a/arch/parisc/kernel/topology.c b/arch/parisc/kernel/topology.c index 0a10e4ddc528..b9d845e314f8 100644 --- a/arch/parisc/kernel/topology.c +++ b/arch/parisc/kernel/topology.c @@ -13,56 +13,21 @@ #include <linux/percpu.h> #include <linux/sched.h> #include <linux/sched/topology.h> +#include <linux/cpu.h> #include <asm/topology.h> +#include <asm/sections.h> - /* - * cpu topology table - */ -struct cputopo_parisc cpu_topology[NR_CPUS] __read_mostly; -EXPORT_SYMBOL_GPL(cpu_topology); - -const struct cpumask *cpu_coregroup_mask(int cpu) -{ - return &cpu_topology[cpu].core_sibling; -} - -static void update_siblings_masks(unsigned int cpuid) -{ - struct cputopo_parisc *cpu_topo, *cpuid_topo = &cpu_topology[cpuid]; - int cpu; - - /* update core and thread sibling masks */ - for_each_possible_cpu(cpu) { - cpu_topo = &cpu_topology[cpu]; - - if (cpuid_topo->socket_id != cpu_topo->socket_id) - continue; - - cpumask_set_cpu(cpuid, &cpu_topo->core_sibling); - if (cpu != cpuid) - cpumask_set_cpu(cpu, &cpuid_topo->core_sibling); - - if (cpuid_topo->core_id != cpu_topo->core_id) - continue; - - cpumask_set_cpu(cpuid, &cpu_topo->thread_sibling); - if (cpu != cpuid) - cpumask_set_cpu(cpu, &cpuid_topo->thread_sibling); - } - smp_wmb(); -} - -static int dualcores_found __initdata; +static DEFINE_PER_CPU(struct cpu, cpu_devices); /* * store_cpu_topology is called at boot when only one cpu is running * and with the mutex cpu_hotplug.lock locked, when several cpus have booted, * which prevents simultaneous write access to cpu_topology array */ -void __init store_cpu_topology(unsigned int cpuid) +void store_cpu_topology(unsigned int cpuid) { - struct cputopo_parisc *cpuid_topo = &cpu_topology[cpuid]; + struct cpu_topology *cpuid_topo = &cpu_topology[cpuid]; struct cpuinfo_parisc *p; int max_socket = -1; unsigned long cpu; @@ -71,6 +36,12 @@ void __init store_cpu_topology(unsigned int cpuid) if (cpuid_topo->core_id != -1) return; +#ifdef CONFIG_HOTPLUG_CPU + per_cpu(cpu_devices, cpuid).hotpluggable = 1; +#endif + if (register_cpu(&per_cpu(cpu_devices, cpuid), cpuid)) + pr_warn("Failed to register CPU%d device", cpuid); + /* create cpu topology mapping */ cpuid_topo->thread_id = -1; cpuid_topo->core_id = 0; @@ -86,57 +57,31 @@ void __init store_cpu_topology(unsigned int cpuid) cpuid_topo->core_id = cpu_topology[cpu].core_id; if (p->cpu_loc) { cpuid_topo->core_id++; - cpuid_topo->socket_id = cpu_topology[cpu].socket_id; - dualcores_found = 1; + cpuid_topo->package_id = cpu_topology[cpu].package_id; continue; } } - if (cpuid_topo->socket_id == -1) - max_socket = max(max_socket, cpu_topology[cpu].socket_id); + if (cpuid_topo->package_id == -1) + max_socket = max(max_socket, cpu_topology[cpu].package_id); } - if (cpuid_topo->socket_id == -1) - cpuid_topo->socket_id = max_socket + 1; + if (cpuid_topo->package_id == -1) + cpuid_topo->package_id = max_socket + 1; update_siblings_masks(cpuid); - pr_info("CPU%u: thread %d, cpu %d, socket %d\n", - cpuid, cpu_topology[cpuid].thread_id, + pr_info("CPU%u: cpu core %d of socket %d\n", + cpuid, cpu_topology[cpuid].core_id, - cpu_topology[cpuid].socket_id); + cpu_topology[cpuid].package_id); } -static struct sched_domain_topology_level parisc_mc_topology[] = { -#ifdef CONFIG_SCHED_MC - { cpu_coregroup_mask, cpu_core_flags, SD_INIT_NAME(MC) }, -#endif - - { cpu_cpu_mask, SD_INIT_NAME(DIE) }, - { NULL, }, -}; - /* * init_cpu_topology is called at boot when only one cpu is running * which prevent simultaneous write access to cpu_topology array */ void __init init_cpu_topology(void) { - unsigned int cpu; - - /* init core mask and capacity */ - for_each_possible_cpu(cpu) { - struct cputopo_parisc *cpu_topo = &(cpu_topology[cpu]); - - cpu_topo->thread_id = -1; - cpu_topo->core_id = -1; - cpu_topo->socket_id = -1; - cpumask_clear(&cpu_topo->core_sibling); - cpumask_clear(&cpu_topo->thread_sibling); - } - smp_wmb(); - - /* Set scheduler topology descriptor */ - if (dualcores_found) - set_sched_topology(parisc_mc_topology); + reset_cpu_topology(); } diff --git a/arch/parisc/kernel/traps.c b/arch/parisc/kernel/traps.c index 82fc01189488..1107ca819ac8 100644 --- a/arch/parisc/kernel/traps.c +++ b/arch/parisc/kernel/traps.c @@ -30,6 +30,7 @@ #include <linux/ratelimit.h> #include <linux/uaccess.h> #include <linux/kdebug.h> +#include <linux/kfence.h> #include <asm/assembly.h> #include <asm/io.h> @@ -46,10 +47,14 @@ #include <linux/kgdb.h> #include <linux/kprobes.h> +#if defined(CONFIG_LIGHTWEIGHT_SPINLOCK_CHECK) +#include <asm/spinlock.h> +#endif + #include "../math-emu/math-emu.h" /* for handle_fpe() */ static void parisc_show_stack(struct task_struct *task, - struct pt_regs *regs); + struct pt_regs *regs, const char *loglvl); static int printbinary(char *buf, unsigned long x, int nbits) { @@ -75,7 +80,7 @@ static int printbinary(char *buf, unsigned long x, int nbits) lvl, f, (x), (x+3), (r)[(x)+0], (r)[(x)+1], \ (r)[(x)+2], (r)[(x)+3]) -static void print_gr(char *level, struct pt_regs *regs) +static void print_gr(const char *level, struct pt_regs *regs) { int i; char buf[64]; @@ -89,7 +94,7 @@ static void print_gr(char *level, struct pt_regs *regs) PRINTREGS(level, regs->gr, "r", RFMT, i); } -static void print_fr(char *level, struct pt_regs *regs) +static void print_fr(const char *level, struct pt_regs *regs) { int i; char buf[64]; @@ -119,7 +124,7 @@ static void print_fr(char *level, struct pt_regs *regs) void show_regs(struct pt_regs *regs) { int i, user; - char *level; + const char *level; unsigned long cr30, cr31; user = user_mode(regs); @@ -143,7 +148,7 @@ void show_regs(struct pt_regs *regs) printk("%s IIR: %08lx ISR: " RFMT " IOR: " RFMT "\n", level, regs->iir, regs->isr, regs->ior); printk("%s CPU: %8d CR30: " RFMT " CR31: " RFMT "\n", - level, current_thread_info()->cpu, cr30, cr31); + level, task_cpu(current), cr30, cr31); printk("%s ORIG_R28: " RFMT "\n", level, regs->orig_r28); if (user) { @@ -155,7 +160,7 @@ void show_regs(struct pt_regs *regs) printk("%s IAOQ[1]: %pS\n", level, (void *) regs->iaoq[1]); printk("%s RP(r2): %pS\n", level, (void *) regs->gr[2]); - parisc_show_stack(current, regs); + parisc_show_stack(current, regs, KERN_DEFAULT); } } @@ -170,37 +175,37 @@ static DEFINE_RATELIMIT_STATE(_hppa_rs, } -static void do_show_stack(struct unwind_frame_info *info) +static void do_show_stack(struct unwind_frame_info *info, const char *loglvl) { int i = 1; - printk(KERN_CRIT "Backtrace:\n"); + printk("%sBacktrace:\n", loglvl); while (i <= MAX_UNWIND_ENTRIES) { if (unwind_once(info) < 0 || info->ip == 0) break; if (__kernel_text_address(info->ip)) { - printk(KERN_CRIT " [<" RFMT ">] %pS\n", - info->ip, (void *) info->ip); + printk("%s [<" RFMT ">] %pS\n", + loglvl, info->ip, (void *) info->ip); i++; } } - printk(KERN_CRIT "\n"); + printk("%s\n", loglvl); } static void parisc_show_stack(struct task_struct *task, - struct pt_regs *regs) + struct pt_regs *regs, const char *loglvl) { struct unwind_frame_info info; unwind_frame_init_task(&info, task, regs); - do_show_stack(&info); + do_show_stack(&info, loglvl); } -void show_stack(struct task_struct *t, unsigned long *sp) +void show_stack(struct task_struct *t, unsigned long *sp, const char *loglvl) { - parisc_show_stack(t, NULL); + parisc_show_stack(t, NULL, loglvl); } int is_valid_bugaddr(unsigned long iaoq) @@ -238,13 +243,6 @@ void die_if_kernel(char *str, struct pt_regs *regs, long err) /* unlock the pdc lock if necessary */ pdc_emergency_unlock(); - /* maybe the kernel hasn't booted very far yet and hasn't been able - * to initialize the serial or STI console. In that case we should - * re-enable the pdc console, so that the user will be able to - * identify the problem. */ - if (!console_drivers) - pdc_console_restart(); - if (err) printk(KERN_CRIT "%s (pid %d): %s (code %ld)\n", current->comm, task_pid_nr(current), str, err); @@ -268,7 +266,7 @@ void die_if_kernel(char *str, struct pt_regs *regs, long err) panic("Fatal exception"); oops_exit(); - do_exit(SIGSEGV); + make_task_dead(SIGSEGV); } /* gdb uses break 4,8 */ @@ -297,21 +295,30 @@ static void handle_break(struct pt_regs *regs) } #ifdef CONFIG_KPROBES - if (unlikely(iir == PARISC_KPROBES_BREAK_INSN)) { + if (unlikely(iir == PARISC_KPROBES_BREAK_INSN && !user_mode(regs))) { parisc_kprobe_break_handler(regs); return; } - + if (unlikely(iir == PARISC_KPROBES_BREAK_INSN2 && !user_mode(regs))) { + parisc_kprobe_ss_handler(regs); + return; + } #endif #ifdef CONFIG_KGDB - if (unlikely(iir == PARISC_KGDB_COMPILED_BREAK_INSN || - iir == PARISC_KGDB_BREAK_INSN)) { + if (unlikely((iir == PARISC_KGDB_COMPILED_BREAK_INSN || + iir == PARISC_KGDB_BREAK_INSN)) && !user_mode(regs)) { kgdb_handle_exception(9, SIGTRAP, 0, regs); return; } #endif +#ifdef CONFIG_LIGHTWEIGHT_SPINLOCK_CHECK + if ((iir == SPINLOCK_BREAK_INSN) && !user_mode(regs)) { + die_if_kernel("Spinlock was trashed", regs, 1); + } +#endif + if (unlikely(iir != GDB_BREAK_INSN)) parisc_printk_ratelimited(0, regs, KERN_DEBUG "break %d,%d: pid=%d command='%s'\n", @@ -328,10 +335,7 @@ static void default_trap(int code, struct pt_regs *regs) show_regs(regs); } -void (*cpu_lpmc) (int code, struct pt_regs *regs) __read_mostly = default_trap; - - -void transfer_pim_to_trap_frame(struct pt_regs *regs) +static void transfer_pim_to_trap_frame(struct pt_regs *regs) { register int i; extern unsigned int hpmc_pim_data[]; @@ -425,10 +429,6 @@ void parisc_terminate(char *msg, struct pt_regs *regs, int code, unsigned long o /* unlock the pdc lock if necessary */ pdc_emergency_unlock(); - /* restart pdc console if necessary */ - if (!console_drivers) - pdc_console_restart(); - /* Not all paths will gutter the processor... */ switch(code){ @@ -437,7 +437,6 @@ void parisc_terminate(char *msg, struct pt_regs *regs, int code, unsigned long o break; default: - /* Fall through */ break; } @@ -446,7 +445,7 @@ void parisc_terminate(char *msg, struct pt_regs *regs, int code, unsigned long o /* show_stack(NULL, (unsigned long *)regs->gr[30]); */ struct unwind_frame_info info; unwind_frame_init(&info, current, regs); - do_show_stack(&info); + do_show_stack(&info, KERN_CRIT); } printk("\n"); @@ -466,7 +465,7 @@ void parisc_terminate(char *msg, struct pt_regs *regs, int code, unsigned long o * panic notifiers, and we should call panic * directly from the location that we wish. * e.g. We should not call panic from - * parisc_terminate, but rather the oter way around. + * parisc_terminate, but rather the other way around. * This hack works, prints the panic message twice, * and it enables reboot timers! */ @@ -479,9 +478,7 @@ void notrace handle_interruption(int code, struct pt_regs *regs) unsigned long fault_space = 0; int si_code; - if (code == 1) - pdc_console_restart(); /* switch back to pdc if HPMC */ - else + if (!irqs_disabled_flags(regs->gr[0])) local_irq_enable(); /* Security check: @@ -539,11 +536,6 @@ void notrace handle_interruption(int code, struct pt_regs *regs) /* Recovery counter trap */ regs->gr[0] &= ~PSW_R; -#ifdef CONFIG_KPROBES - if (parisc_kprobe_ss_handler(regs)) - return; -#endif - #ifdef CONFIG_KGDB if (kgdb_single_step) { kgdb_handle_exception(0, SIGTRAP, 0, regs); @@ -562,7 +554,7 @@ void notrace handle_interruption(int code, struct pt_regs *regs) flush_cache_all(); flush_tlb_all(); - cpu_lpmc(5, regs); + default_trap(code, regs); return; case PARISC_ITLB_TRAP: @@ -644,12 +636,12 @@ void notrace handle_interruption(int code, struct pt_regs *regs) case 15: /* Data TLB miss fault/Data page fault */ - /* Fall through */ + fallthrough; case 16: /* Non-access instruction TLB miss fault */ /* The instruction TLB entry needed for the target address of the FIC is absent, and hardware can't find it, so we get to cleanup */ - /* Fall through */ + fallthrough; case 17: /* Non-access data TLB miss fault/Non-access data page fault */ /* FIXME: @@ -662,6 +654,8 @@ void notrace handle_interruption(int code, struct pt_regs *regs) by hand. Technically we need to emulate: fdc,fdce,pdc,"fic,4f",prober,probeir,probew, probeiw */ + if (code == 17 && handle_nadtlb_fault(regs)) + return; fault_address = regs->ior; fault_space = regs->isr; break; @@ -673,7 +667,7 @@ void notrace handle_interruption(int code, struct pt_regs *regs) handle_unaligned(regs); return; } - /* Fall Through */ + fallthrough; case 26: /* PCXL: Data memory access rights trap */ fault_address = regs->ior; @@ -683,7 +677,7 @@ void notrace handle_interruption(int code, struct pt_regs *regs) case 19: /* Data memory break trap */ regs->gr[0] |= PSW_X; /* So we can single-step over the trap */ - /* fall thru */ + fallthrough; case 21: /* Page reference trap */ handle_gdb_break(regs, TRAP_HWBKPT); @@ -717,7 +711,7 @@ void notrace handle_interruption(int code, struct pt_regs *regs) if (user_mode(regs)) { struct vm_area_struct *vma; - down_read(¤t->mm->mmap_sem); + mmap_read_lock(current->mm); vma = find_vma(current->mm,regs->iaoq[0]); if (vma && (regs->iaoq[0] >= vma->vm_start) && (vma->vm_flags & VM_EXEC)) { @@ -725,12 +719,14 @@ void notrace handle_interruption(int code, struct pt_regs *regs) fault_address = regs->iaoq[0]; fault_space = regs->iasq[0]; - up_read(¤t->mm->mmap_sem); + mmap_read_unlock(current->mm); break; /* call do_page_fault() */ } - up_read(¤t->mm->mmap_sem); + mmap_read_unlock(current->mm); } - /* Fall Through */ + /* CPU could not fetch instruction, so clear stale IIR value. */ + regs->iir = 0xbaadf00d; + fallthrough; case 27: /* Data memory protection ID trap */ if (code == 27 && !user_mode(regs) && @@ -783,11 +779,15 @@ void notrace handle_interruption(int code, struct pt_regs *regs) * unless pagefault_disable() was called before. */ - if (fault_space == 0 && !faulthandler_disabled()) + if (faulthandler_disabled() || fault_space == 0) { /* Clean up and return if in exception table. */ if (fixup_exception(regs)) return; + /* Clean up and return if handled by kfence. */ + if (kfence_handle_page_fault(fault_address, + parisc_acctyp(code, regs->iir) == VM_WRITE, regs)) + return; pdc_chassis_send_status(PDC_CHASSIS_DIRECT_PANIC); parisc_terminate("Kernel Fault", regs, code, fault_address); } @@ -797,16 +797,14 @@ void notrace handle_interruption(int code, struct pt_regs *regs) } -void __init initialize_ivt(const void *iva) +static void __init initialize_ivt(const void *iva) { - extern u32 os_hpmc_size; extern const u32 os_hpmc[]; int i; u32 check = 0; u32 *ivap; - u32 *hpmcp; - u32 length, instr; + u32 instr; if (strcmp((const char *)iva, "cows can fly")) panic("IVT invalid"); @@ -837,18 +835,12 @@ void __init initialize_ivt(const void *iva) /* Setup IVA and compute checksum for HPMC handler */ ivap[6] = (u32)__pa(os_hpmc); - length = os_hpmc_size; - ivap[7] = length; - - hpmcp = (u32 *)os_hpmc; - - for (i=0; i<length/4; i++) - check += *hpmcp++; for (i=0; i<8; i++) check += ivap[i]; ivap[5] = -check; + pr_debug("initialize_ivt: IVA[6] = 0x%08x\n", ivap[6]); } @@ -865,7 +857,3 @@ void __init early_trap_init(void) initialize_ivt(&fault_vector_20); } - -void __init trap_init(void) -{ -} diff --git a/arch/parisc/kernel/unaligned.c b/arch/parisc/kernel/unaligned.c index 237d20dd5622..71e596ca5a86 100644 --- a/arch/parisc/kernel/unaligned.c +++ b/arch/parisc/kernel/unaligned.c @@ -3,17 +3,16 @@ * Unaligned memory access handler * * Copyright (C) 2001 Randolph Chung <tausq@debian.org> + * Copyright (C) 2022 Helge Deller <deller@gmx.de> * Significantly tweaked by LaMont Jones <lamont@debian.org> */ -#include <linux/jiffies.h> -#include <linux/kernel.h> -#include <linux/module.h> #include <linux/sched/signal.h> -#include <linux/sched/debug.h> #include <linux/signal.h> #include <linux/ratelimit.h> #include <linux/uaccess.h> +#include <linux/sysctl.h> +#include <asm/unaligned.h> #include <asm/hardirq.h> #include <asm/traps.h> @@ -25,18 +24,7 @@ #define DPRINTF(fmt, args...) #endif -#ifdef CONFIG_64BIT -#define RFMT "%016lx" -#else -#define RFMT "%08lx" -#endif - -#define FIXUP_BRANCH(lbl) \ - "\tldil L%%" #lbl ", %%r1\n" \ - "\tldo R%%" #lbl "(%%r1), %%r1\n" \ - "\tbv,n %%r0(%%r1)\n" -/* If you use FIXUP_BRANCH, then you must list this clobber */ -#define FIXUP_BRANCH_CLOBBER "r1" +#define RFMT "%#08lx" /* 1111 1100 0000 0000 0001 0011 1100 0000 */ #define OPCODE1(a,b,c) ((a)<<26|(b)<<12|(c)<<6) @@ -107,44 +95,37 @@ #define R1(i) (((i)>>21)&0x1f) #define R2(i) (((i)>>16)&0x1f) #define R3(i) ((i)&0x1f) -#define FR3(i) ((((i)<<1)&0x1f)|(((i)>>6)&1)) +#define FR3(i) ((((i)&0x1f)<<1)|(((i)>>6)&1)) #define IM(i,n) (((i)>>1&((1<<(n-1))-1))|((i)&1?((0-1L)<<(n-1)):0)) #define IM5_2(i) IM((i)>>16,5) #define IM5_3(i) IM((i),5) #define IM14(i) IM((i),14) #define ERR_NOTHANDLED -1 -#define ERR_PAGEFAULT -2 int unaligned_enabled __read_mostly = 1; static int emulate_ldh(struct pt_regs *regs, int toreg) { unsigned long saddr = regs->ior; - unsigned long val = 0; - int ret; + unsigned long val = 0, temp1; + ASM_EXCEPTIONTABLE_VAR(ret); DPRINTF("load " RFMT ":" RFMT " to r%d for 2 bytes\n", regs->isr, regs->ior, toreg); __asm__ __volatile__ ( " mtsp %4, %%sr1\n" -"1: ldbs 0(%%sr1,%3), %%r20\n" +"1: ldbs 0(%%sr1,%3), %2\n" "2: ldbs 1(%%sr1,%3), %0\n" -" depw %%r20, 23, 24, %0\n" -" copy %%r0, %1\n" +" depw %2, 23, 24, %0\n" "3: \n" -" .section .fixup,\"ax\"\n" -"4: ldi -2, %1\n" - FIXUP_BRANCH(3b) -" .previous\n" - ASM_EXCEPTIONTABLE_ENTRY(1b, 4b) - ASM_EXCEPTIONTABLE_ENTRY(2b, 4b) - : "=r" (val), "=r" (ret) - : "0" (val), "r" (saddr), "r" (regs->isr) - : "r20", FIXUP_BRANCH_CLOBBER ); - - DPRINTF("val = 0x" RFMT "\n", val); + ASM_EXCEPTIONTABLE_ENTRY_EFAULT(1b, 3b, "%1") + ASM_EXCEPTIONTABLE_ENTRY_EFAULT(2b, 3b, "%1") + : "+r" (val), "+r" (ret), "=&r" (temp1) + : "r" (saddr), "r" (regs->isr) ); + + DPRINTF("val = " RFMT "\n", val); if (toreg) regs->gr[toreg] = val; @@ -155,34 +136,28 @@ static int emulate_ldh(struct pt_regs *regs, int toreg) static int emulate_ldw(struct pt_regs *regs, int toreg, int flop) { unsigned long saddr = regs->ior; - unsigned long val = 0; - int ret; + unsigned long val = 0, temp1, temp2; + ASM_EXCEPTIONTABLE_VAR(ret); DPRINTF("load " RFMT ":" RFMT " to r%d for 4 bytes\n", regs->isr, regs->ior, toreg); __asm__ __volatile__ ( -" zdep %3,28,2,%%r19\n" /* r19=(ofs&3)*8 */ -" mtsp %4, %%sr1\n" -" depw %%r0,31,2,%3\n" -"1: ldw 0(%%sr1,%3),%0\n" -"2: ldw 4(%%sr1,%3),%%r20\n" -" subi 32,%%r19,%%r19\n" -" mtctl %%r19,11\n" -" vshd %0,%%r20,%0\n" -" copy %%r0, %1\n" +" zdep %4,28,2,%2\n" /* r19=(ofs&3)*8 */ +" mtsp %5, %%sr1\n" +" depw %%r0,31,2,%4\n" +"1: ldw 0(%%sr1,%4),%0\n" +"2: ldw 4(%%sr1,%4),%3\n" +" subi 32,%2,%2\n" +" mtctl %2,11\n" +" vshd %0,%3,%0\n" "3: \n" -" .section .fixup,\"ax\"\n" -"4: ldi -2, %1\n" - FIXUP_BRANCH(3b) -" .previous\n" - ASM_EXCEPTIONTABLE_ENTRY(1b, 4b) - ASM_EXCEPTIONTABLE_ENTRY(2b, 4b) - : "=r" (val), "=r" (ret) - : "0" (val), "r" (saddr), "r" (regs->isr) - : "r19", "r20", FIXUP_BRANCH_CLOBBER ); - - DPRINTF("val = 0x" RFMT "\n", val); + ASM_EXCEPTIONTABLE_ENTRY_EFAULT(1b, 3b, "%1") + ASM_EXCEPTIONTABLE_ENTRY_EFAULT(2b, 3b, "%1") + : "+r" (val), "+r" (ret), "=&r" (temp1), "=&r" (temp2) + : "r" (saddr), "r" (regs->isr) ); + + DPRINTF("val = " RFMT "\n", val); if (flop) ((__u32*)(regs->fr))[toreg] = val; @@ -194,65 +169,49 @@ static int emulate_ldw(struct pt_regs *regs, int toreg, int flop) static int emulate_ldd(struct pt_regs *regs, int toreg, int flop) { unsigned long saddr = regs->ior; + unsigned long shift, temp1; __u64 val = 0; - int ret; + ASM_EXCEPTIONTABLE_VAR(ret); DPRINTF("load " RFMT ":" RFMT " to r%d for 8 bytes\n", regs->isr, regs->ior, toreg); -#ifdef CONFIG_PA20 -#ifndef CONFIG_64BIT - if (!flop) - return -1; -#endif + if (!IS_ENABLED(CONFIG_64BIT) && !flop) + return ERR_NOTHANDLED; + +#ifdef CONFIG_64BIT __asm__ __volatile__ ( -" depd,z %3,60,3,%%r19\n" /* r19=(ofs&7)*8 */ -" mtsp %4, %%sr1\n" -" depd %%r0,63,3,%3\n" -"1: ldd 0(%%sr1,%3),%0\n" -"2: ldd 8(%%sr1,%3),%%r20\n" -" subi 64,%%r19,%%r19\n" -" mtsar %%r19\n" -" shrpd %0,%%r20,%%sar,%0\n" -" copy %%r0, %1\n" +" depd,z %2,60,3,%3\n" /* shift=(ofs&7)*8 */ +" mtsp %5, %%sr1\n" +" depd %%r0,63,3,%2\n" +"1: ldd 0(%%sr1,%2),%0\n" +"2: ldd 8(%%sr1,%2),%4\n" +" subi 64,%3,%3\n" +" mtsar %3\n" +" shrpd %0,%4,%%sar,%0\n" "3: \n" -" .section .fixup,\"ax\"\n" -"4: ldi -2, %1\n" - FIXUP_BRANCH(3b) -" .previous\n" - ASM_EXCEPTIONTABLE_ENTRY(1b,4b) - ASM_EXCEPTIONTABLE_ENTRY(2b,4b) - : "=r" (val), "=r" (ret) - : "0" (val), "r" (saddr), "r" (regs->isr) - : "r19", "r20", FIXUP_BRANCH_CLOBBER ); + ASM_EXCEPTIONTABLE_ENTRY_EFAULT(1b, 3b, "%1") + ASM_EXCEPTIONTABLE_ENTRY_EFAULT(2b, 3b, "%1") + : "+r" (val), "+r" (ret), "+r" (saddr), "=&r" (shift), "=&r" (temp1) + : "r" (regs->isr) ); #else - { - unsigned long valh=0,vall=0; __asm__ __volatile__ ( -" zdep %5,29,2,%%r19\n" /* r19=(ofs&3)*8 */ -" mtsp %6, %%sr1\n" -" dep %%r0,31,2,%5\n" -"1: ldw 0(%%sr1,%5),%0\n" -"2: ldw 4(%%sr1,%5),%1\n" -"3: ldw 8(%%sr1,%5),%%r20\n" -" subi 32,%%r19,%%r19\n" -" mtsar %%r19\n" -" vshd %0,%1,%0\n" -" vshd %1,%%r20,%1\n" -" copy %%r0, %2\n" +" zdep %2,29,2,%3\n" /* shift=(ofs&3)*8 */ +" mtsp %5, %%sr1\n" +" dep %%r0,31,2,%2\n" +"1: ldw 0(%%sr1,%2),%0\n" +"2: ldw 4(%%sr1,%2),%R0\n" +"3: ldw 8(%%sr1,%2),%4\n" +" subi 32,%3,%3\n" +" mtsar %3\n" +" vshd %0,%R0,%0\n" +" vshd %R0,%4,%R0\n" "4: \n" -" .section .fixup,\"ax\"\n" -"5: ldi -2, %2\n" - FIXUP_BRANCH(4b) -" .previous\n" - ASM_EXCEPTIONTABLE_ENTRY(1b,5b) - ASM_EXCEPTIONTABLE_ENTRY(2b,5b) - ASM_EXCEPTIONTABLE_ENTRY(3b,5b) - : "=r" (valh), "=r" (vall), "=r" (ret) - : "0" (valh), "1" (vall), "r" (saddr), "r" (regs->isr) - : "r19", "r20", FIXUP_BRANCH_CLOBBER ); - val=((__u64)valh<<32)|(__u64)vall; - } + ASM_EXCEPTIONTABLE_ENTRY_EFAULT(1b, 4b, "%1") + ASM_EXCEPTIONTABLE_ENTRY_EFAULT(2b, 4b, "%1") + ASM_EXCEPTIONTABLE_ENTRY_EFAULT(3b, 4b, "%1") + : "+r" (val), "+r" (ret), "+r" (saddr), "=&r" (shift), "=&r" (temp1) + : "r" (regs->isr) ); #endif DPRINTF("val = 0x%llx\n", val); @@ -267,31 +226,25 @@ static int emulate_ldd(struct pt_regs *regs, int toreg, int flop) static int emulate_sth(struct pt_regs *regs, int frreg) { - unsigned long val = regs->gr[frreg]; - int ret; + unsigned long val = regs->gr[frreg], temp1; + ASM_EXCEPTIONTABLE_VAR(ret); if (!frreg) val = 0; - DPRINTF("store r%d (0x" RFMT ") to " RFMT ":" RFMT " for 2 bytes\n", frreg, + DPRINTF("store r%d (" RFMT ") to " RFMT ":" RFMT " for 2 bytes\n", frreg, val, regs->isr, regs->ior); __asm__ __volatile__ ( -" mtsp %3, %%sr1\n" -" extrw,u %1, 23, 8, %%r19\n" -"1: stb %1, 1(%%sr1, %2)\n" -"2: stb %%r19, 0(%%sr1, %2)\n" -" copy %%r0, %0\n" +" mtsp %4, %%sr1\n" +" extrw,u %2, 23, 8, %1\n" +"1: stb %1, 0(%%sr1, %3)\n" +"2: stb %2, 1(%%sr1, %3)\n" "3: \n" -" .section .fixup,\"ax\"\n" -"4: ldi -2, %0\n" - FIXUP_BRANCH(3b) -" .previous\n" - ASM_EXCEPTIONTABLE_ENTRY(1b,4b) - ASM_EXCEPTIONTABLE_ENTRY(2b,4b) - : "=r" (ret) - : "r" (val), "r" (regs->ior), "r" (regs->isr) - : "r19", FIXUP_BRANCH_CLOBBER ); + ASM_EXCEPTIONTABLE_ENTRY_EFAULT(1b, 3b, "%0") + ASM_EXCEPTIONTABLE_ENTRY_EFAULT(2b, 3b, "%0") + : "+r" (ret), "=&r" (temp1) + : "r" (val), "r" (regs->ior), "r" (regs->isr) ); return ret; } @@ -299,7 +252,7 @@ static int emulate_sth(struct pt_regs *regs, int frreg) static int emulate_stw(struct pt_regs *regs, int frreg, int flop) { unsigned long val; - int ret; + ASM_EXCEPTIONTABLE_VAR(ret); if (flop) val = ((__u32*)(regs->fr))[frreg]; @@ -308,7 +261,7 @@ static int emulate_stw(struct pt_regs *regs, int frreg, int flop) else val = 0; - DPRINTF("store r%d (0x" RFMT ") to " RFMT ":" RFMT " for 4 bytes\n", frreg, + DPRINTF("store r%d (" RFMT ") to " RFMT ":" RFMT " for 4 bytes\n", frreg, val, regs->isr, regs->ior); @@ -328,24 +281,19 @@ static int emulate_stw(struct pt_regs *regs, int frreg, int flop) " or %%r1, %%r21, %%r21\n" " stw %%r20,0(%%sr1,%2)\n" " stw %%r21,4(%%sr1,%2)\n" -" copy %%r0, %0\n" "3: \n" -" .section .fixup,\"ax\"\n" -"4: ldi -2, %0\n" - FIXUP_BRANCH(3b) -" .previous\n" - ASM_EXCEPTIONTABLE_ENTRY(1b,4b) - ASM_EXCEPTIONTABLE_ENTRY(2b,4b) - : "=r" (ret) + ASM_EXCEPTIONTABLE_ENTRY_EFAULT(1b, 3b, "%0") + ASM_EXCEPTIONTABLE_ENTRY_EFAULT(2b, 3b, "%0") + : "+r" (ret) : "r" (val), "r" (regs->ior), "r" (regs->isr) - : "r19", "r20", "r21", "r22", "r1", FIXUP_BRANCH_CLOBBER ); + : "r19", "r20", "r21", "r22", "r1" ); - return 0; + return ret; } static int emulate_std(struct pt_regs *regs, int frreg, int flop) { __u64 val; - int ret; + ASM_EXCEPTIONTABLE_VAR(ret); if (flop) val = regs->fr[frreg]; @@ -357,11 +305,10 @@ static int emulate_std(struct pt_regs *regs, int frreg, int flop) DPRINTF("store r%d (0x%016llx) to " RFMT ":" RFMT " for 8 bytes\n", frreg, val, regs->isr, regs->ior); -#ifdef CONFIG_PA20 -#ifndef CONFIG_64BIT - if (!flop) - return -1; -#endif + if (!IS_ENABLED(CONFIG_64BIT) && !flop) + return ERR_NOTHANDLED; + +#ifdef CONFIG_64BIT __asm__ __volatile__ ( " mtsp %3, %%sr1\n" " depd,z %2, 60, 3, %%r19\n" @@ -378,54 +325,43 @@ static int emulate_std(struct pt_regs *regs, int frreg, int flop) " or %%r1, %%r21, %%r21\n" "3: std %%r20,0(%%sr1,%2)\n" "4: std %%r21,8(%%sr1,%2)\n" -" copy %%r0, %0\n" "5: \n" -" .section .fixup,\"ax\"\n" -"6: ldi -2, %0\n" - FIXUP_BRANCH(5b) -" .previous\n" - ASM_EXCEPTIONTABLE_ENTRY(1b,6b) - ASM_EXCEPTIONTABLE_ENTRY(2b,6b) - ASM_EXCEPTIONTABLE_ENTRY(3b,6b) - ASM_EXCEPTIONTABLE_ENTRY(4b,6b) - : "=r" (ret) + ASM_EXCEPTIONTABLE_ENTRY_EFAULT(1b, 5b, "%0") + ASM_EXCEPTIONTABLE_ENTRY_EFAULT(2b, 5b, "%0") + ASM_EXCEPTIONTABLE_ENTRY_EFAULT(3b, 5b, "%0") + ASM_EXCEPTIONTABLE_ENTRY_EFAULT(4b, 5b, "%0") + : "+r" (ret) : "r" (val), "r" (regs->ior), "r" (regs->isr) - : "r19", "r20", "r21", "r22", "r1", FIXUP_BRANCH_CLOBBER ); + : "r19", "r20", "r21", "r22", "r1" ); #else { - unsigned long valh=(val>>32),vall=(val&0xffffffffl); __asm__ __volatile__ ( -" mtsp %4, %%sr1\n" -" zdep %2, 29, 2, %%r19\n" +" mtsp %3, %%sr1\n" +" zdep %R1, 29, 2, %%r19\n" " dep %%r0, 31, 2, %2\n" " mtsar %%r19\n" " zvdepi -2, 32, %%r19\n" -"1: ldw 0(%%sr1,%3),%%r20\n" -"2: ldw 8(%%sr1,%3),%%r21\n" -" vshd %1, %2, %%r1\n" +"1: ldw 0(%%sr1,%2),%%r20\n" +"2: ldw 8(%%sr1,%2),%%r21\n" +" vshd %1, %R1, %%r1\n" " vshd %%r0, %1, %1\n" -" vshd %2, %%r0, %2\n" +" vshd %R1, %%r0, %R1\n" " and %%r20, %%r19, %%r20\n" " andcm %%r21, %%r19, %%r21\n" " or %1, %%r20, %1\n" -" or %2, %%r21, %2\n" -"3: stw %1,0(%%sr1,%1)\n" -"4: stw %%r1,4(%%sr1,%3)\n" -"5: stw %2,8(%%sr1,%3)\n" -" copy %%r0, %0\n" +" or %R1, %%r21, %R1\n" +"3: stw %1,0(%%sr1,%2)\n" +"4: stw %%r1,4(%%sr1,%2)\n" +"5: stw %R1,8(%%sr1,%2)\n" "6: \n" -" .section .fixup,\"ax\"\n" -"7: ldi -2, %0\n" - FIXUP_BRANCH(6b) -" .previous\n" - ASM_EXCEPTIONTABLE_ENTRY(1b,7b) - ASM_EXCEPTIONTABLE_ENTRY(2b,7b) - ASM_EXCEPTIONTABLE_ENTRY(3b,7b) - ASM_EXCEPTIONTABLE_ENTRY(4b,7b) - ASM_EXCEPTIONTABLE_ENTRY(5b,7b) - : "=r" (ret) - : "r" (valh), "r" (vall), "r" (regs->ior), "r" (regs->isr) - : "r19", "r20", "r21", "r1", FIXUP_BRANCH_CLOBBER ); + ASM_EXCEPTIONTABLE_ENTRY_EFAULT(1b, 6b, "%0") + ASM_EXCEPTIONTABLE_ENTRY_EFAULT(2b, 6b, "%0") + ASM_EXCEPTIONTABLE_ENTRY_EFAULT(3b, 6b, "%0") + ASM_EXCEPTIONTABLE_ENTRY_EFAULT(4b, 6b, "%0") + ASM_EXCEPTIONTABLE_ENTRY_EFAULT(5b, 6b, "%0") + : "+r" (ret) + : "r" (val), "r" (regs->ior), "r" (regs->isr) + : "r19", "r20", "r21", "r1" ); } #endif @@ -438,7 +374,6 @@ void handle_unaligned(struct pt_regs *regs) unsigned long newbase = R1(regs->iir)?regs->gr[R1(regs->iir)]:0; int modify = 0; int ret = ERR_NOTHANDLED; - register int flop=0; /* true if this is a flop */ __inc_irq_stat(irq_unaligned_count); @@ -450,10 +385,10 @@ void handle_unaligned(struct pt_regs *regs) if (!(current->thread.flags & PARISC_UAC_NOPRINT) && __ratelimit(&ratelimit)) { - char buf[256]; - sprintf(buf, "%s(%d): unaligned access to 0x" RFMT " at ip=0x" RFMT "\n", - current->comm, task_pid_nr(current), regs->ior, regs->iaoq[0]); - printk(KERN_WARNING "%s", buf); + printk(KERN_WARNING "%s(%d): unaligned access to " RFMT + " at ip " RFMT " (iir " RFMT ")\n", + current->comm, task_pid_nr(current), regs->ior, + regs->iaoq[0], regs->iir); #ifdef DEBUG_UNALIGNED show_regs(regs); #endif @@ -461,6 +396,13 @@ void handle_unaligned(struct pt_regs *regs) if (!unaligned_enabled) goto force_sigbus; + } else { + static DEFINE_RATELIMIT_STATE(kernel_ratelimit, 5 * HZ, 5); + if (!(current->thread.flags & PARISC_UAC_NOPRINT) && + __ratelimit(&kernel_ratelimit)) + pr_warn("Kernel: unaligned access to " RFMT " in %pS " + "(iir " RFMT ")\n", + regs->ior, (void *)regs->iaoq[0], regs->iir); } /* handle modification - OK, it's ugly, see the instruction manual */ @@ -535,7 +477,7 @@ void handle_unaligned(struct pt_regs *regs) case OPCODE_LDWA_I: case OPCODE_LDW_S: case OPCODE_LDWA_S: - ret = emulate_ldw(regs, R3(regs->iir),0); + ret = emulate_ldw(regs, R3(regs->iir), 0); break; case OPCODE_STH: @@ -544,20 +486,20 @@ void handle_unaligned(struct pt_regs *regs) case OPCODE_STW: case OPCODE_STWA: - ret = emulate_stw(regs, R2(regs->iir),0); + ret = emulate_stw(regs, R2(regs->iir), 0); break; -#ifdef CONFIG_PA20 +#ifdef CONFIG_64BIT case OPCODE_LDD_I: case OPCODE_LDDA_I: case OPCODE_LDD_S: case OPCODE_LDDA_S: - ret = emulate_ldd(regs, R3(regs->iir),0); + ret = emulate_ldd(regs, R3(regs->iir), 0); break; case OPCODE_STD: case OPCODE_STDA: - ret = emulate_std(regs, R2(regs->iir),0); + ret = emulate_std(regs, R2(regs->iir), 0); break; #endif @@ -565,28 +507,24 @@ void handle_unaligned(struct pt_regs *regs) case OPCODE_FLDWS: case OPCODE_FLDWXR: case OPCODE_FLDWSR: - flop=1; - ret = emulate_ldw(regs,FR3(regs->iir),1); + ret = emulate_ldw(regs, FR3(regs->iir), 1); break; case OPCODE_FLDDX: case OPCODE_FLDDS: - flop=1; - ret = emulate_ldd(regs,R3(regs->iir),1); + ret = emulate_ldd(regs, R3(regs->iir), 1); break; case OPCODE_FSTWX: case OPCODE_FSTWS: case OPCODE_FSTWXR: case OPCODE_FSTWSR: - flop=1; - ret = emulate_stw(regs,FR3(regs->iir),1); + ret = emulate_stw(regs, FR3(regs->iir), 1); break; case OPCODE_FSTDX: case OPCODE_FSTDS: - flop=1; - ret = emulate_std(regs,R3(regs->iir),1); + ret = emulate_std(regs, R3(regs->iir), 1); break; case OPCODE_LDCD_I: @@ -596,37 +534,33 @@ void handle_unaligned(struct pt_regs *regs) ret = ERR_NOTHANDLED; /* "undefined", but lets kill them. */ break; } -#ifdef CONFIG_PA20 switch (regs->iir & OPCODE2_MASK) { case OPCODE_FLDD_L: - flop=1; ret = emulate_ldd(regs,R2(regs->iir),1); break; case OPCODE_FSTD_L: - flop=1; ret = emulate_std(regs, R2(regs->iir),1); break; +#ifdef CONFIG_64BIT case OPCODE_LDD_L: ret = emulate_ldd(regs, R2(regs->iir),0); break; case OPCODE_STD_L: ret = emulate_std(regs, R2(regs->iir),0); break; - } #endif + } switch (regs->iir & OPCODE3_MASK) { case OPCODE_FLDW_L: - flop=1; - ret = emulate_ldw(regs, R2(regs->iir),0); + ret = emulate_ldw(regs, R2(regs->iir), 1); break; case OPCODE_LDW_M: - ret = emulate_ldw(regs, R2(regs->iir),1); + ret = emulate_ldw(regs, R2(regs->iir), 0); break; case OPCODE_FSTW_L: - flop=1; ret = emulate_stw(regs, R2(regs->iir),1); break; case OPCODE_STW_M: @@ -673,7 +607,7 @@ void handle_unaligned(struct pt_regs *regs) printk(KERN_CRIT "Unaligned handler failed, ret = %d\n", ret); die_if_kernel("Unaligned data reference", regs, 28); - if (ret == ERR_PAGEFAULT) + if (ret == -EFAULT) { force_sig_fault(SIGSEGV, SEGV_MAPERR, (void __user *)regs->ior); diff --git a/arch/parisc/kernel/unwind.c b/arch/parisc/kernel/unwind.c index 87ae476d1c4f..f7e0fee5ee55 100644 --- a/arch/parisc/kernel/unwind.c +++ b/arch/parisc/kernel/unwind.c @@ -14,6 +14,7 @@ #include <linux/sched.h> #include <linux/slab.h> #include <linux/sort.h> +#include <linux/sched/task_stack.h> #include <linux/uaccess.h> #include <asm/assembly.h> @@ -21,12 +22,15 @@ #include <asm/ptrace.h> #include <asm/unwind.h> +#include <asm/switch_to.h> +#include <asm/sections.h> +#include <asm/ftrace.h> /* #define DEBUG 1 */ #ifdef DEBUG #define dbg(x...) pr_debug(x) #else -#define dbg(x...) +#define dbg(x...) do { } while (0) #endif #define KERNEL_START (KERNEL_BINARY_TEXT_START) @@ -176,7 +180,7 @@ void unwind_table_remove(struct unwind_table *table) /* Called from setup_arch to import the kernel unwind info */ int __init unwind_init(void) { - long start, stop; + long start __maybe_unused, stop __maybe_unused; register unsigned long gp __asm__ ("r27"); start = (long)&__start___unwind[0]; @@ -203,6 +207,11 @@ int __init unwind_init(void) return 0; } +static bool pc_is_kernel_fn(unsigned long pc, void *fn) +{ + return (unsigned long)dereference_kernel_function_descriptor(fn) == pc; +} + static int unwind_special(struct unwind_frame_info *info, unsigned long pc, int frame_size) { /* @@ -212,7 +221,6 @@ static int unwind_special(struct unwind_frame_info *info, unsigned long pc, int * Note: We could use dereference_kernel_function_descriptor() * instead but we want to keep it simple here. */ - extern void * const handle_interruption; extern void * const ret_from_kernel_thread; extern void * const syscall_exit; extern void * const intr_return; @@ -221,7 +229,7 @@ static int unwind_special(struct unwind_frame_info *info, unsigned long pc, int extern void * const _call_on_stack; #endif /* CONFIG_IRQSTACKS */ - if (pc == (unsigned long) &handle_interruption) { + if (pc_is_kernel_fn(pc, handle_interruption)) { struct pt_regs *regs = (struct pt_regs *)(info->sp - frame_size - PT_SZ_ALGN); dbg("Unwinding through handle_interruption()\n"); info->prev_sp = regs->gr[30]; @@ -229,13 +237,13 @@ static int unwind_special(struct unwind_frame_info *info, unsigned long pc, int return 1; } - if (pc == (unsigned long) &ret_from_kernel_thread || - pc == (unsigned long) &syscall_exit) { + if (pc == (unsigned long)&ret_from_kernel_thread || + pc == (unsigned long)&syscall_exit) { info->prev_sp = info->prev_ip = 0; return 1; } - if (pc == (unsigned long) &intr_return) { + if (pc == (unsigned long)&intr_return) { struct pt_regs *regs; dbg("Found intr_return()\n"); @@ -246,20 +254,20 @@ static int unwind_special(struct unwind_frame_info *info, unsigned long pc, int return 1; } - if (pc == (unsigned long) &_switch_to_ret) { + if (pc_is_kernel_fn(pc, _switch_to) || + pc == (unsigned long)&_switch_to_ret) { info->prev_sp = info->sp - CALLEE_SAVE_FRAME_SIZE; info->prev_ip = *(unsigned long *)(info->prev_sp - RP_OFFSET); return 1; } #ifdef CONFIG_IRQSTACKS - if (pc == (unsigned long) &_call_on_stack) { + if (pc == (unsigned long)&_call_on_stack) { info->prev_sp = *(unsigned long *)(info->sp - FRAME_SIZE - REG_SZ); info->prev_ip = *(unsigned long *)(info->sp - FRAME_SIZE - RP_OFFSET); return 1; } #endif - return 0; } @@ -292,17 +300,15 @@ static void unwind_frame_regs(struct unwind_frame_info *info) info->prev_sp = sp - 64; info->prev_ip = 0; - /* The stack is at the end inside the thread_union - * struct. If we reach data, we have reached the - * beginning of the stack and should stop unwinding. */ - if (info->prev_sp >= (unsigned long) task_thread_info(info->t) && - info->prev_sp < ((unsigned long) task_thread_info(info->t) - + THREAD_SZ_ALGN)) { + /* Check if stack is inside kernel stack area */ + if ((info->prev_sp - (unsigned long) task_stack_page(info->t)) + >= THREAD_SIZE) { info->prev_sp = 0; break; } - if (get_user(tmp, (unsigned long *)(info->prev_sp - RP_OFFSET))) + if (copy_from_kernel_nofault(&tmp, + (void *)info->prev_sp - RP_OFFSET, sizeof(tmp))) break; info->prev_ip = tmp; sp = info->prev_sp; diff --git a/arch/parisc/kernel/vdso.c b/arch/parisc/kernel/vdso.c new file mode 100644 index 000000000000..c5cbfce7a84c --- /dev/null +++ b/arch/parisc/kernel/vdso.c @@ -0,0 +1,122 @@ +// SPDX-License-Identifier: GPL-2.0 +/* + * Copyright (c) 2022 Helge Deller <deller@gmx.de> + * + * based on arch/s390/kernel/vdso.c which is + * Copyright IBM Corp. 2008 + * Author(s): Martin Schwidefsky (schwidefsky@de.ibm.com) + */ + +#include <linux/kernel.h> +#include <linux/mm.h> +#include <linux/slab.h> +#include <linux/elf.h> +#include <linux/timekeeper_internal.h> +#include <linux/compat.h> +#include <linux/nsproxy.h> +#include <linux/time_namespace.h> +#include <linux/random.h> + +#include <asm/pgtable.h> +#include <asm/page.h> +#include <asm/sections.h> +#include <asm/vdso.h> +#include <asm/cacheflush.h> + +extern char vdso32_start, vdso32_end; +extern char vdso64_start, vdso64_end; + +static int vdso_mremap(const struct vm_special_mapping *sm, + struct vm_area_struct *vma) +{ + current->mm->context.vdso_base = vma->vm_start; + return 0; +} + +#ifdef CONFIG_64BIT +static struct vm_special_mapping vdso64_mapping = { + .name = "[vdso]", + .mremap = vdso_mremap, +}; +#endif + +static struct vm_special_mapping vdso32_mapping = { + .name = "[vdso]", + .mremap = vdso_mremap, +}; + +/* + * This is called from binfmt_elf, we create the special vma for the + * vDSO and insert it into the mm struct tree + */ +int arch_setup_additional_pages(struct linux_binprm *bprm, + int executable_stack) +{ + + unsigned long vdso_text_start, vdso_text_len, map_base; + struct vm_special_mapping *vdso_mapping; + struct mm_struct *mm = current->mm; + struct vm_area_struct *vma; + int rc; + + if (mmap_write_lock_killable(mm)) + return -EINTR; + +#ifdef CONFIG_64BIT + if (!is_compat_task()) { + vdso_text_len = &vdso64_end - &vdso64_start; + vdso_mapping = &vdso64_mapping; + } else +#endif + { + vdso_text_len = &vdso32_end - &vdso32_start; + vdso_mapping = &vdso32_mapping; + } + + map_base = mm->mmap_base; + if (current->flags & PF_RANDOMIZE) + map_base -= get_random_u32_below(0x20) * PAGE_SIZE; + + vdso_text_start = get_unmapped_area(NULL, map_base, vdso_text_len, 0, 0); + + /* VM_MAYWRITE for COW so gdb can set breakpoints */ + vma = _install_special_mapping(mm, vdso_text_start, vdso_text_len, + VM_READ|VM_EXEC| + VM_MAYREAD|VM_MAYWRITE|VM_MAYEXEC, + vdso_mapping); + if (IS_ERR(vma)) { + do_munmap(mm, vdso_text_start, PAGE_SIZE, NULL); + rc = PTR_ERR(vma); + } else { + current->mm->context.vdso_base = vdso_text_start; + rc = 0; + } + + mmap_write_unlock(mm); + return rc; +} + +static struct page ** __init vdso_setup_pages(void *start, void *end) +{ + int pages = (end - start) >> PAGE_SHIFT; + struct page **pagelist; + int i; + + pagelist = kcalloc(pages + 1, sizeof(struct page *), GFP_KERNEL); + if (!pagelist) + panic("%s: Cannot allocate page list for VDSO", __func__); + for (i = 0; i < pages; i++) + pagelist[i] = virt_to_page(start + i * PAGE_SIZE); + return pagelist; +} + +static int __init vdso_init(void) +{ +#ifdef CONFIG_64BIT + vdso64_mapping.pages = vdso_setup_pages(&vdso64_start, &vdso64_end); +#endif + if (IS_ENABLED(CONFIG_COMPAT) || !IS_ENABLED(CONFIG_64BIT)) + vdso32_mapping.pages = vdso_setup_pages(&vdso32_start, &vdso32_end); + return 0; +} +arch_initcall(vdso_init); diff --git a/arch/parisc/kernel/vdso32/Makefile b/arch/parisc/kernel/vdso32/Makefile new file mode 100644 index 000000000000..4459a48d2303 --- /dev/null +++ b/arch/parisc/kernel/vdso32/Makefile @@ -0,0 +1,53 @@ +# List of files in the vdso, has to be asm only for now + +obj-vdso32 = note.o sigtramp.o restart_syscall.o + +# Build rules + +targets := $(obj-vdso32) vdso32.so +obj-vdso32 := $(addprefix $(obj)/, $(obj-vdso32)) + +ccflags-y := -shared -fno-common -fbuiltin -mno-fast-indirect-calls -O2 -mno-long-calls +# -march=1.1 -mschedule=7100LC +ccflags-y += -nostdlib -Wl,-soname=linux-vdso32.so.1 \ + $(call ld-option, -Wl$(comma)--hash-style=sysv) +asflags-y := -D__VDSO32__ -s + +KBUILD_AFLAGS += -DBUILD_VDSO +KBUILD_CFLAGS += -DBUILD_VDSO -DDISABLE_BRANCH_PROFILING + +VDSO_LIBGCC := $(shell $(CROSS32CC) -print-libgcc-file-name) + +obj-y += vdso32_wrapper.o +extra-y += vdso32.lds +CPPFLAGS_vdso32.lds += -P -C # -U$(ARCH) + +$(obj)/vdso32_wrapper.o : $(obj)/vdso32.so FORCE + +# Force dependency (incbin is bad) +# link rule for the .so file, .lds has to be first +$(obj)/vdso32.so: $(src)/vdso32.lds $(obj-vdso32) $(obj-cvdso32) $(VDSO_LIBGCC) FORCE + $(call if_changed,vdso32ld) + +# assembly rules for the .S files +$(obj-vdso32): %.o: %.S FORCE + $(call if_changed_dep,vdso32as) + +$(obj-cvdso32): %.o: %.c FORCE + $(call if_changed_dep,vdso32cc) + +# actual build commands +quiet_cmd_vdso32ld = VDSO32L $@ + cmd_vdso32ld = $(CROSS32CC) $(c_flags) -Wl,-T $(filter-out FORCE, $^) -o $@ +quiet_cmd_vdso32as = VDSO32A $@ + cmd_vdso32as = $(CROSS32CC) $(a_flags) -c -o $@ $< +quiet_cmd_vdso32cc = VDSO32C $@ + cmd_vdso32cc = $(CROSS32CC) $(c_flags) -c -fPIC -mno-fast-indirect-calls -o $@ $< + +# Generate VDSO offsets using helper script +gen-vdsosym := $(srctree)/$(src)/gen_vdso_offsets.sh +quiet_cmd_vdsosym = VDSOSYM $@ + cmd_vdsosym = $(NM) $< | $(gen-vdsosym) | LC_ALL=C sort > $@ + +include/generated/vdso32-offsets.h: $(obj)/vdso32.so FORCE + $(call if_changed,vdsosym) diff --git a/arch/parisc/kernel/vdso32/gen_vdso_offsets.sh b/arch/parisc/kernel/vdso32/gen_vdso_offsets.sh new file mode 100755 index 000000000000..da39d6cff7f1 --- /dev/null +++ b/arch/parisc/kernel/vdso32/gen_vdso_offsets.sh @@ -0,0 +1,15 @@ +#!/bin/sh +# SPDX-License-Identifier: GPL-2.0 + +# +# Match symbols in the DSO that look like VDSO_*; produce a header file +# of constant offsets into the shared object. +# +# Doing this inside the Makefile will break the $(filter-out) function, +# causing Kbuild to rebuild the vdso-offsets header file every time. +# +# Inspired by arm64 version. +# + +LC_ALL=C +sed -n 's/\([0-9a-f]*\) . __kernel_\(.*\)/\#define vdso32_offset_\2\t0x\1/p' diff --git a/arch/parisc/kernel/vdso32/note.S b/arch/parisc/kernel/vdso32/note.S new file mode 100644 index 000000000000..bb350918bebd --- /dev/null +++ b/arch/parisc/kernel/vdso32/note.S @@ -0,0 +1,26 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +/* + * This supplies .note.* sections to go into the PT_NOTE inside the vDSO text. + * Here we can supply some information useful to userland. + */ + +#include <linux/uts.h> +#include <linux/version.h> + +#define ASM_ELF_NOTE_BEGIN(name, flags, vendor, type) \ + .section name, flags; \ + .balign 4; \ + .long 1f - 0f; /* name length */ \ + .long 3f - 2f; /* data length */ \ + .long type; /* note type */ \ +0: .asciz vendor; /* vendor name */ \ +1: .balign 4; \ +2: + +#define ASM_ELF_NOTE_END \ +3: .balign 4; /* pad out section */ \ + .previous + + ASM_ELF_NOTE_BEGIN(".note.kernel-version", "a", UTS_SYSNAME, 0) + .long LINUX_VERSION_CODE + ASM_ELF_NOTE_END diff --git a/arch/parisc/kernel/vdso32/restart_syscall.S b/arch/parisc/kernel/vdso32/restart_syscall.S new file mode 100644 index 000000000000..7e82008d7e40 --- /dev/null +++ b/arch/parisc/kernel/vdso32/restart_syscall.S @@ -0,0 +1,32 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +/* + * Syscall restart trampoline for 32 and 64 bits processes. + * + * Copyright (C) 2018-2022 Helge Deller <deller@gmx.de> + * Copyright (C) 2022 John David Anglin <dave.anglin@bell.net> + */ + +#include <asm/unistd.h> +#include <asm/vdso.h> + +#include <linux/linkage.h> + + .text + +ENTRY_CFI(__kernel_restart_syscall) + /* + * Setup a trampoline to restart the syscall + * with __NR_restart_syscall + */ + + /* load return pointer */ +#if defined(__VDSO64__) + ldd 0(%sp), %r31 +#elif defined(__VDSO32__) + ldw 0(%sp), %r31 +#endif + + be 0x100(%sr2, %r0) + ldi __NR_restart_syscall, %r20 + +ENDPROC_CFI(__kernel_restart_syscall) diff --git a/arch/parisc/kernel/vdso32/sigtramp.S b/arch/parisc/kernel/vdso32/sigtramp.S new file mode 100644 index 000000000000..192da7077869 --- /dev/null +++ b/arch/parisc/kernel/vdso32/sigtramp.S @@ -0,0 +1,195 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +/* + * Signal trampolines for 32 bit processes. + * + * Copyright (C) 2006 Randolph Chung <tausq@debian.org> + * Copyright (C) 2018-2022 Helge Deller <deller@gmx.de> + * Copyright (C) 2022 John David Anglin <dave.anglin@bell.net> + */ +#include <asm/unistd.h> +#include <linux/linkage.h> +#include <generated/asm-offsets.h> + + .text + +/* Gdb expects the trampoline is on the stack and the pc is offset from + a 64-byte boundary by 0, 4 or 5 instructions. Since the vdso trampoline + is not on the stack, we need a new variant with different offsets and + data to tell gdb where to find the signal context on the stack. + + Here we put the offset to the context data at the start of the trampoline + region and offset the first trampoline by 2 instructions. Please do + not change the trampoline as the code in gdb depends on the following + instruction sequence exactly. + */ + .align 64 + .word SIGFRAME_CONTEXT_REGS32 + +/* The nop here is a hack. The dwarf2 unwind routines subtract 1 from + the return address to get an address in the middle of the presumed + call instruction. Since we don't have a call here, we artifically + extend the range covered by the unwind info by adding a nop before + the real start. + */ + nop + + .globl __kernel_sigtramp_rt + .type __kernel_sigtramp_rt, @function +__kernel_sigtramp_rt: + .proc + .callinfo FRAME=ASM_SIGFRAME_SIZE32,CALLS,SAVE_RP + .entry + +.Lsigrt_start = . - 4 +0: ldi 0, %r25 /* (in_syscall=0) */ + ldi __NR_rt_sigreturn, %r20 + ble 0x100(%sr2, %r0) + nop + +1: ldi 1, %r25 /* (in_syscall=1) */ + ldi __NR_rt_sigreturn, %r20 + ble 0x100(%sr2, %r0) + nop +.Lsigrt_end: + .exit + .procend + .size __kernel_sigtramp_rt,.-__kernel_sigtramp_rt + + + .section .eh_frame,"a",@progbits + +/* This is where the mcontext_t struct can be found on the stack. */ +#define PTREGS SIGFRAME_CONTEXT_REGS32 /* 32-bit process offset is -672 */ + +/* Register REGNO can be found at offset OFS of the mcontext_t structure. */ + .macro rsave regno,ofs + .byte 0x05 /* DW_CFA_offset_extended */ + .uleb128 \regno; /* regno */ + .uleb128 \ofs /* factored offset */ + .endm + +.Lcie: + .long .Lcie_end - .Lcie_start +.Lcie_start: + .long 0 /* CIE ID */ + .byte 1 /* Version number */ + .stringz "zRS" /* NUL-terminated augmentation string */ + .uleb128 4 /* Code alignment factor */ + .sleb128 4 /* Data alignment factor */ + .byte 89 /* Return address register column, iaoq[0] */ + .uleb128 1 /* Augmentation value length */ + .byte 0x1b /* DW_EH_PE_pcrel | DW_EH_PE_sdata4. */ + .byte 0x0f /* DW_CFA_def_cfa_expresion */ + .uleb128 9f - 1f /* length */ +1: + .byte 0x8e /* DW_OP_breg30 */ + .sleb128 PTREGS +9: + .balign 4 +.Lcie_end: + + .long .Lfde0_end - .Lfde0_start +.Lfde0_start: + .long .Lfde0_start - .Lcie /* CIE pointer. */ + .long .Lsigrt_start - . /* PC start, length */ + .long .Lsigrt_end - .Lsigrt_start + .uleb128 0 /* Augmentation */ + + /* General registers */ + rsave 1, 2 + rsave 2, 3 + rsave 3, 4 + rsave 4, 5 + rsave 5, 6 + rsave 6, 7 + rsave 7, 8 + rsave 8, 9 + rsave 9, 10 + rsave 10, 11 + rsave 11, 12 + rsave 12, 13 + rsave 13, 14 + rsave 14, 15 + rsave 15, 16 + rsave 16, 17 + rsave 17, 18 + rsave 18, 19 + rsave 19, 20 + rsave 20, 21 + rsave 21, 22 + rsave 22, 23 + rsave 23, 24 + rsave 24, 25 + rsave 25, 26 + rsave 26, 27 + rsave 27, 28 + rsave 28, 29 + rsave 29, 30 + rsave 30, 31 + rsave 31, 32 + + /* Floating-point registers */ + rsave 32, 42 + rsave 33, 43 + rsave 34, 44 + rsave 35, 45 + rsave 36, 46 + rsave 37, 47 + rsave 38, 48 + rsave 39, 49 + rsave 40, 50 + rsave 41, 51 + rsave 42, 52 + rsave 43, 53 + rsave 44, 54 + rsave 45, 55 + rsave 46, 56 + rsave 47, 57 + rsave 48, 58 + rsave 49, 59 + rsave 50, 60 + rsave 51, 61 + rsave 52, 62 + rsave 53, 63 + rsave 54, 64 + rsave 55, 65 + rsave 56, 66 + rsave 57, 67 + rsave 58, 68 + rsave 59, 69 + rsave 60, 70 + rsave 61, 71 + rsave 62, 72 + rsave 63, 73 + rsave 64, 74 + rsave 65, 75 + rsave 66, 76 + rsave 67, 77 + rsave 68, 78 + rsave 69, 79 + rsave 70, 80 + rsave 71, 81 + rsave 72, 82 + rsave 73, 83 + rsave 74, 84 + rsave 75, 85 + rsave 76, 86 + rsave 77, 87 + rsave 78, 88 + rsave 79, 89 + rsave 80, 90 + rsave 81, 91 + rsave 82, 92 + rsave 83, 93 + rsave 84, 94 + rsave 85, 95 + rsave 86, 96 + rsave 87, 97 + + /* SAR register */ + rsave 88, 102 + + /* iaoq[0] return address register */ + rsave 89, 100 + .balign 4 +.Lfde0_end: diff --git a/arch/parisc/kernel/vdso32/vdso32.lds.S b/arch/parisc/kernel/vdso32/vdso32.lds.S new file mode 100644 index 000000000000..d4aff3af5262 --- /dev/null +++ b/arch/parisc/kernel/vdso32/vdso32.lds.S @@ -0,0 +1,111 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +/* + * This is the infamous ld script for the 32 bits vdso library + */ +#include <asm/vdso.h> +#include <asm/page.h> + +/* Default link addresses for the vDSOs */ +OUTPUT_FORMAT("elf32-hppa-linux") +OUTPUT_ARCH(hppa) +ENTRY(_start) + +SECTIONS +{ + . = VDSO_LBASE + SIZEOF_HEADERS; + .hash : { *(.hash) } :text + .gnu.hash : { *(.gnu.hash) } + .dynsym : { *(.dynsym) } + .dynstr : { *(.dynstr) } + .gnu.version : { *(.gnu.version) } + .gnu.version_d : { *(.gnu.version_d) } + .gnu.version_r : { *(.gnu.version_r) } + + .note : { *(.note.*) } :text :note + + . = ALIGN (16); + .text : + { + *(.text .stub .text.* .gnu.linkonce.t.*) + } + PROVIDE (__etext = .); + PROVIDE (_etext = .); + PROVIDE (etext = .); + + /* Other stuff is appended to the text segment: */ + .rodata : { *(.rodata .rodata.* .gnu.linkonce.r.*) } + .rodata1 : { *(.rodata1) } + .rodata2 : { *(.data.rel.ro) } + + .eh_frame_hdr : { *(.eh_frame_hdr) } :text :eh_frame_hdr + .eh_frame : { KEEP (*(.eh_frame)) } :text + .gcc_except_table : { *(.gcc_except_table) } + .fixup : { *(.fixup) } + + .dynamic : { *(.dynamic) } :text :dynamic + .plt : { *(.plt) } + .got : { *(.got) } + + _end = .; + __end = .; + PROVIDE (end = .); + + + /* Stabs debugging sections are here too + */ + .stab 0 : { *(.stab) } + .stabstr 0 : { *(.stabstr) } + .stab.excl 0 : { *(.stab.excl) } + .stab.exclstr 0 : { *(.stab.exclstr) } + .stab.index 0 : { *(.stab.index) } + .stab.indexstr 0 : { *(.stab.indexstr) } + .comment 0 : { *(.comment) } + .debug 0 : { *(.debug) } + .line 0 : { *(.line) } + + .debug_srcinfo 0 : { *(.debug_srcinfo) } + .debug_sfnames 0 : { *(.debug_sfnames) } + + .debug_aranges 0 : { *(.debug_aranges) } + .debug_pubnames 0 : { *(.debug_pubnames) } + + .debug_info 0 : { *(.debug_info .gnu.linkonce.wi.*) } + .debug_abbrev 0 : { *(.debug_abbrev) } + .debug_line 0 : { *(.debug_line) } + .debug_frame 0 : { *(.debug_frame) } + .debug_str 0 : { *(.debug_str) } + .debug_loc 0 : { *(.debug_loc) } + .debug_macinfo 0 : { *(.debug_macinfo) } + + .debug_weaknames 0 : { *(.debug_weaknames) } + .debug_funcnames 0 : { *(.debug_funcnames) } + .debug_typenames 0 : { *(.debug_typenames) } + .debug_varnames 0 : { *(.debug_varnames) } + + /DISCARD/ : { *(.note.GNU-stack) } + /DISCARD/ : { *(.data .data.* .gnu.linkonce.d.* .sdata*) } + /DISCARD/ : { *(.bss .sbss .dynbss .dynsbss) } +} + + +PHDRS +{ + text PT_LOAD FILEHDR PHDRS FLAGS(5); /* PF_R|PF_X */ + note PT_NOTE FLAGS(4); /* PF_R */ + dynamic PT_DYNAMIC FLAGS(4); /* PF_R */ + eh_frame_hdr PT_GNU_EH_FRAME; +} + + +/* + * This controls what symbols we export from the DSO. + */ +VERSION +{ + VDSO_VERSION_STRING { + global: + __kernel_sigtramp_rt32; + __kernel_restart_syscall32; + local: *; + }; +} diff --git a/arch/parisc/kernel/vdso32/vdso32_wrapper.S b/arch/parisc/kernel/vdso32/vdso32_wrapper.S new file mode 100644 index 000000000000..5ac06093e8ec --- /dev/null +++ b/arch/parisc/kernel/vdso32/vdso32_wrapper.S @@ -0,0 +1,14 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +#include <linux/linkage.h> +#include <asm/page.h> + + __PAGE_ALIGNED_DATA + + .globl vdso32_start, vdso32_end + .balign PAGE_SIZE +vdso32_start: + .incbin "arch/parisc/kernel/vdso32/vdso32.so" + .balign PAGE_SIZE +vdso32_end: + + .previous diff --git a/arch/parisc/kernel/vdso64/Makefile b/arch/parisc/kernel/vdso64/Makefile new file mode 100644 index 000000000000..f3d6045793f4 --- /dev/null +++ b/arch/parisc/kernel/vdso64/Makefile @@ -0,0 +1,48 @@ +# List of files in the vdso, has to be asm only for now + +obj-vdso64 = note.o sigtramp.o restart_syscall.o + +# Build rules + +targets := $(obj-vdso64) vdso64.so +obj-vdso64 := $(addprefix $(obj)/, $(obj-vdso64)) + + +ccflags-y := -shared -fno-common -fno-builtin +ccflags-y += -nostdlib -Wl,-soname=linux-vdso64.so.1 \ + $(call ld-option, -Wl$(comma)--hash-style=sysv) +asflags-y := -D__VDSO64__ -s + +KBUILD_AFLAGS += -DBUILD_VDSO +KBUILD_CFLAGS += -DBUILD_VDSO -DDISABLE_BRANCH_PROFILING + +VDSO_LIBGCC := $(shell $(CC) -print-libgcc-file-name) + +obj-y += vdso64_wrapper.o +extra-y += vdso64.lds +CPPFLAGS_vdso64.lds += -P -C -U$(ARCH) + +$(obj)/vdso64_wrapper.o : $(obj)/vdso64.so FORCE + +# Force dependency (incbin is bad) +# link rule for the .so file, .lds has to be first +$(obj)/vdso64.so: $(src)/vdso64.lds $(obj-vdso64) $(VDSO_LIBGCC) FORCE + $(call if_changed,vdso64ld) + +# assembly rules for the .S files +$(obj-vdso64): %.o: %.S FORCE + $(call if_changed_dep,vdso64as) + +# actual build commands +quiet_cmd_vdso64ld = VDSO64L $@ + cmd_vdso64ld = $(CC) $(c_flags) -Wl,-T $(filter-out FORCE, $^) -o $@ +quiet_cmd_vdso64as = VDSO64A $@ + cmd_vdso64as = $(CC) $(a_flags) -c -o $@ $< + +# Generate VDSO offsets using helper script +gen-vdsosym := $(srctree)/$(src)/gen_vdso_offsets.sh +quiet_cmd_vdsosym = VDSOSYM $@ + cmd_vdsosym = $(NM) $< | $(gen-vdsosym) | LC_ALL=C sort > $@ + +include/generated/vdso64-offsets.h: $(obj)/vdso64.so FORCE + $(call if_changed,vdsosym) diff --git a/arch/parisc/kernel/vdso64/gen_vdso_offsets.sh b/arch/parisc/kernel/vdso64/gen_vdso_offsets.sh new file mode 100755 index 000000000000..37f05cb38dad --- /dev/null +++ b/arch/parisc/kernel/vdso64/gen_vdso_offsets.sh @@ -0,0 +1,15 @@ +#!/bin/sh +# SPDX-License-Identifier: GPL-2.0 + +# +# Match symbols in the DSO that look like VDSO_*; produce a header file +# of constant offsets into the shared object. +# +# Doing this inside the Makefile will break the $(filter-out) function, +# causing Kbuild to rebuild the vdso-offsets header file every time. +# +# Inspired by arm64 version. +# + +LC_ALL=C +sed -n 's/\([0-9a-f]*\) . __kernel_\(.*\)/\#define vdso64_offset_\2\t0x\1/p' diff --git a/arch/parisc/kernel/vdso64/note.S b/arch/parisc/kernel/vdso64/note.S new file mode 100644 index 000000000000..bd1fa23597d6 --- /dev/null +++ b/arch/parisc/kernel/vdso64/note.S @@ -0,0 +1,2 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +#include "../vdso32/note.S" diff --git a/arch/parisc/kernel/vdso64/restart_syscall.S b/arch/parisc/kernel/vdso64/restart_syscall.S new file mode 100644 index 000000000000..83004451f6b1 --- /dev/null +++ b/arch/parisc/kernel/vdso64/restart_syscall.S @@ -0,0 +1,3 @@ +/* SPDX-License-Identifier: GPL-2.0 */ + +#include "../vdso32/restart_syscall.S" diff --git a/arch/parisc/kernel/vdso64/sigtramp.S b/arch/parisc/kernel/vdso64/sigtramp.S new file mode 100644 index 000000000000..66a6d2b241e1 --- /dev/null +++ b/arch/parisc/kernel/vdso64/sigtramp.S @@ -0,0 +1,166 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +/* + * Signal trampolines for 64 bit processes. + * + * Copyright (C) 2006 Randolph Chung <tausq@debian.org> + * Copyright (C) 2018-2022 Helge Deller <deller@gmx.de> + * Copyright (C) 2022 John David Anglin <dave.anglin@bell.net> + */ +#include <asm/unistd.h> +#include <linux/linkage.h> +#include <generated/asm-offsets.h> + + .text + +/* Gdb expects the trampoline is on the stack and the pc is offset from + a 64-byte boundary by 0, 4 or 5 instructions. Since the vdso trampoline + is not on the stack, we need a new variant with different offsets and + data to tell gdb where to find the signal context on the stack. + + Here we put the offset to the context data at the start of the trampoline + region and offset the first trampoline by 2 instructions. Please do + not change the trampoline as the code in gdb depends on the following + instruction sequence exactly. + */ + .align 64 + .word SIGFRAME_CONTEXT_REGS + +/* The nop here is a hack. The dwarf2 unwind routines subtract 1 from + the return address to get an address in the middle of the presumed + call instruction. Since we don't have a call here, we artifically + extend the range covered by the unwind info by adding a nop before + the real start. + */ + nop + + .globl __kernel_sigtramp_rt + .type __kernel_sigtramp_rt, @function +__kernel_sigtramp_rt: + .proc + .callinfo FRAME=ASM_SIGFRAME_SIZE,CALLS,SAVE_RP + .entry + +.Lsigrt_start = . - 4 +0: ldi 0, %r25 /* (in_syscall=0) */ + ldi __NR_rt_sigreturn, %r20 + ble 0x100(%sr2, %r0) + nop + +1: ldi 1, %r25 /* (in_syscall=1) */ + ldi __NR_rt_sigreturn, %r20 + ble 0x100(%sr2, %r0) + nop +.Lsigrt_end: + .exit + .procend + .size __kernel_sigtramp_rt,.-__kernel_sigtramp_rt + + .section .eh_frame,"a",@progbits + +/* This is where the mcontext_t struct can be found on the stack. */ +#define PTREGS SIGFRAME_CONTEXT_REGS /* 64-bit process offset is -720 */ + +/* Register REGNO can be found at offset OFS of the mcontext_t structure. */ + .macro rsave regno,ofs + .byte 0x05 /* DW_CFA_offset_extended */ + .uleb128 \regno; /* regno */ + .uleb128 \ofs /* factored offset */ + .endm + +.Lcie: + .long .Lcie_end - .Lcie_start +.Lcie_start: + .long 0 /* CIE ID */ + .byte 1 /* Version number */ + .stringz "zRS" /* NUL-terminated augmentation string */ + .uleb128 4 /* Code alignment factor */ + .sleb128 8 /* Data alignment factor */ + .byte 61 /* Return address register column, iaoq[0] */ + .uleb128 1 /* Augmentation value length */ + .byte 0x1b /* DW_EH_PE_pcrel | DW_EH_PE_sdata4. */ + .byte 0x0f /* DW_CFA_def_cfa_expresion */ + .uleb128 9f - 1f /* length */ +1: + .byte 0x8e /* DW_OP_breg30 */ + .sleb128 PTREGS +9: + .balign 8 +.Lcie_end: + + .long .Lfde0_end - .Lfde0_start +.Lfde0_start: + .long .Lfde0_start - .Lcie /* CIE pointer. */ + .long .Lsigrt_start - . /* PC start, length */ + .long .Lsigrt_end - .Lsigrt_start + .uleb128 0 /* Augmentation */ + + /* General registers */ + rsave 1, 2 + rsave 2, 3 + rsave 3, 4 + rsave 4, 5 + rsave 5, 6 + rsave 6, 7 + rsave 7, 8 + rsave 8, 9 + rsave 9, 10 + rsave 10, 11 + rsave 11, 12 + rsave 12, 13 + rsave 13, 14 + rsave 14, 15 + rsave 15, 16 + rsave 16, 17 + rsave 17, 18 + rsave 18, 19 + rsave 19, 20 + rsave 20, 21 + rsave 21, 22 + rsave 22, 23 + rsave 23, 24 + rsave 24, 25 + rsave 25, 26 + rsave 26, 27 + rsave 27, 28 + rsave 28, 29 + rsave 29, 30 + rsave 30, 31 + rsave 31, 32 + + /* Floating-point registers */ + rsave 32, 36 + rsave 33, 37 + rsave 34, 38 + rsave 35, 39 + rsave 36, 40 + rsave 37, 41 + rsave 38, 42 + rsave 39, 43 + rsave 40, 44 + rsave 41, 45 + rsave 42, 46 + rsave 43, 47 + rsave 44, 48 + rsave 45, 49 + rsave 46, 50 + rsave 47, 51 + rsave 48, 52 + rsave 49, 53 + rsave 50, 54 + rsave 51, 55 + rsave 52, 56 + rsave 53, 57 + rsave 54, 58 + rsave 55, 59 + rsave 56, 60 + rsave 57, 61 + rsave 58, 62 + rsave 59, 63 + + /* SAR register */ + rsave 60, 67 + + /* iaoq[0] return address register */ + rsave 61, 65 + .balign 8 +.Lfde0_end: diff --git a/arch/parisc/kernel/vdso64/vdso64.lds.S b/arch/parisc/kernel/vdso64/vdso64.lds.S new file mode 100644 index 000000000000..de1fb4b19286 --- /dev/null +++ b/arch/parisc/kernel/vdso64/vdso64.lds.S @@ -0,0 +1,109 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +/* + * This is the infamous ld script for the 64 bits vdso library + */ +#include <asm/vdso.h> + +/* Default link addresses for the vDSOs */ +OUTPUT_FORMAT("elf64-hppa-linux") +OUTPUT_ARCH(hppa:hppa2.0w) +ENTRY(_start) + +SECTIONS +{ + . = VDSO_LBASE + SIZEOF_HEADERS; + .hash : { *(.hash) } :text + .gnu.hash : { *(.gnu.hash) } + .dynsym : { *(.dynsym) } + .dynstr : { *(.dynstr) } + .gnu.version : { *(.gnu.version) } + .gnu.version_d : { *(.gnu.version_d) } + .gnu.version_r : { *(.gnu.version_r) } + + .note : { *(.note.*) } :text :note + + . = ALIGN (16); + .text : + { + *(.text .stub .text.* .gnu.linkonce.t.*) + } + PROVIDE (__etext = .); + PROVIDE (_etext = .); + PROVIDE (etext = .); + + /* Other stuff is appended to the text segment: */ + .rodata : { *(.rodata .rodata.* .gnu.linkonce.r.*) } + .rodata1 : { *(.rodata1) } + + .eh_frame_hdr : { *(.eh_frame_hdr) } :text :eh_frame_hdr + .eh_frame : { KEEP (*(.eh_frame)) } :text + .gcc_except_table : { *(.gcc_except_table) } + .fixup : { *(.fixup) } + + .dynamic : { *(.dynamic) } :text :dynamic + .plt : { *(.plt) } + .got : { *(.got) } + + _end = .; + __end = .; + PROVIDE (end = .); + + + /* Stabs debugging sections are here too + */ + .stab 0 : { *(.stab) } + .stabstr 0 : { *(.stabstr) } + .stab.excl 0 : { *(.stab.excl) } + .stab.exclstr 0 : { *(.stab.exclstr) } + .stab.index 0 : { *(.stab.index) } + .stab.indexstr 0 : { *(.stab.indexstr) } + .comment 0 : { *(.comment) } + .debug 0 : { *(.debug) } + .line 0 : { *(.line) } + + .debug_srcinfo 0 : { *(.debug_srcinfo) } + .debug_sfnames 0 : { *(.debug_sfnames) } + + .debug_aranges 0 : { *(.debug_aranges) } + .debug_pubnames 0 : { *(.debug_pubnames) } + + .debug_info 0 : { *(.debug_info .gnu.linkonce.wi.*) } + .debug_abbrev 0 : { *(.debug_abbrev) } + .debug_line 0 : { *(.debug_line) } + .debug_frame 0 : { *(.debug_frame) } + .debug_str 0 : { *(.debug_str) } + .debug_loc 0 : { *(.debug_loc) } + .debug_macinfo 0 : { *(.debug_macinfo) } + + .debug_weaknames 0 : { *(.debug_weaknames) } + .debug_funcnames 0 : { *(.debug_funcnames) } + .debug_typenames 0 : { *(.debug_typenames) } + .debug_varnames 0 : { *(.debug_varnames) } + + /DISCARD/ : { *(.note.GNU-stack) } + /DISCARD/ : { *(.data .data.* .gnu.linkonce.d.* .sdata*) } + /DISCARD/ : { *(.bss .sbss .dynbss .dynsbss) } +} + + +PHDRS +{ + text PT_LOAD FILEHDR PHDRS FLAGS(5); /* PF_R|PF_X */ + note PT_NOTE FLAGS(4); /* PF_R */ + dynamic PT_DYNAMIC FLAGS(4); /* PF_R */ + eh_frame_hdr PT_GNU_EH_FRAME; +} + + +/* + * This controls what symbols we export from the DSO. + */ +VERSION +{ + VDSO_VERSION_STRING { + global: + __kernel_sigtramp_rt64; + __kernel_restart_syscall64; + local: *; + }; +} diff --git a/arch/parisc/kernel/vdso64/vdso64_wrapper.S b/arch/parisc/kernel/vdso64/vdso64_wrapper.S new file mode 100644 index 000000000000..66f929482d3d --- /dev/null +++ b/arch/parisc/kernel/vdso64/vdso64_wrapper.S @@ -0,0 +1,14 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +#include <linux/linkage.h> +#include <asm/page.h> + + __PAGE_ALIGNED_DATA + + .globl vdso64_start, vdso64_end + .balign PAGE_SIZE +vdso64_start: + .incbin "arch/parisc/kernel/vdso64/vdso64.so" + .balign PAGE_SIZE +vdso64_end: + + .previous diff --git a/arch/parisc/kernel/vmlinux.lds.S b/arch/parisc/kernel/vmlinux.lds.S index 53e29d88f99c..b445e47903cf 100644 --- a/arch/parisc/kernel/vmlinux.lds.S +++ b/arch/parisc/kernel/vmlinux.lds.S @@ -86,7 +86,6 @@ SECTIONS TEXT_TEXT LOCK_TEXT SCHED_TEXT - CPUIDLE_TEXT KPROBES_TEXT IRQENTRY_TEXT SOFTIRQENTRY_TEXT @@ -128,9 +127,10 @@ SECTIONS } #endif - RO_DATA(8) + RO_DATA(PAGE_SIZE) /* unwind info */ + . = ALIGN(4); .PARISC.unwind : { __start___unwind = .; *(.PARISC.unwind) @@ -155,6 +155,7 @@ SECTIONS } /* End of data section */ + . = ALIGN(PAGE_SIZE); _edata = .; /* BSS */ @@ -164,6 +165,7 @@ SECTIONS _end = . ; STABS_DEBUG + ELF_DETAILS .note 0 : { *(.note) } /* Sections to be discarded */ |