diff options
231 files changed, 3123 insertions, 2793 deletions
diff --git a/Documentation/admin-guide/xfs.rst b/Documentation/admin-guide/xfs.rst index 6178153d3320..5422407a96d7 100644 --- a/Documentation/admin-guide/xfs.rst +++ b/Documentation/admin-guide/xfs.rst @@ -284,6 +284,9 @@ The following sysctls are available for the XFS filesystem: removes unused preallocation from clean inodes and releases the unused space back to the free pool. + fs.xfs.speculative_cow_prealloc_lifetime + This is an alias for speculative_prealloc_lifetime. + fs.xfs.error_level (Min: 0 Default: 3 Max: 11) A volume knob for error reporting when internal errors occur. This will generate detailed messages & backtraces for filesystem @@ -356,12 +359,13 @@ The following sysctls are available for the XFS filesystem: Deprecated Sysctls ================== -=========================== ================ - Name Removal Schedule -=========================== ================ -fs.xfs.irix_sgid_inherit September 2025 -fs.xfs.irix_symlink_mode September 2025 -=========================== ================ +=========================================== ================ + Name Removal Schedule +=========================================== ================ +fs.xfs.irix_sgid_inherit September 2025 +fs.xfs.irix_symlink_mode September 2025 +fs.xfs.speculative_cow_prealloc_lifetime September 2025 +=========================================== ================ Removed Sysctls diff --git a/Documentation/filesystems/porting.rst b/Documentation/filesystems/porting.rst index 633610253f45..0302035781be 100644 --- a/Documentation/filesystems/porting.rst +++ b/Documentation/filesystems/porting.rst @@ -883,3 +883,10 @@ For bvec based itererators bio_iov_iter_get_pages() now doesn't copy bvecs but uses the one provided. Anyone issuing kiocb-I/O should ensure that the bvec and page references stay until I/O has completed, i.e. until ->ki_complete() has been called or returned with non -EIOCBQUEUED code. + +--- + +**mandatory** + +mnt_want_write_file() can now only be paired with mnt_drop_write_file(), +whereas previously it could be paired with mnt_drop_write() as well. diff --git a/arch/alpha/kernel/process.c b/arch/alpha/kernel/process.c index 6c71554206cc..5112ab996394 100644 --- a/arch/alpha/kernel/process.c +++ b/arch/alpha/kernel/process.c @@ -249,7 +249,7 @@ int copy_thread(unsigned long clone_flags, unsigned long usp, childti->pcb.ksp = (unsigned long) childstack; childti->pcb.flags = 1; /* set FEN, clear everything else */ - if (unlikely(p->flags & PF_KTHREAD)) { + if (unlikely(p->flags & (PF_KTHREAD | PF_IO_WORKER))) { /* kernel thread */ memset(childstack, 0, sizeof(struct switch_stack) + sizeof(struct pt_regs)); diff --git a/arch/arc/kernel/process.c b/arch/arc/kernel/process.c index 37f724ad5e39..d838d0d57696 100644 --- a/arch/arc/kernel/process.c +++ b/arch/arc/kernel/process.c @@ -191,7 +191,7 @@ int copy_thread(unsigned long clone_flags, unsigned long usp, childksp[0] = 0; /* fp */ childksp[1] = (unsigned long)ret_from_fork; /* blink */ - if (unlikely(p->flags & PF_KTHREAD)) { + if (unlikely(p->flags & (PF_KTHREAD | PF_IO_WORKER))) { memset(c_regs, 0, sizeof(struct pt_regs)); c_callee->r13 = kthread_arg; diff --git a/arch/arm/kernel/process.c b/arch/arm/kernel/process.c index ee3aee69e444..5199a2bb4111 100644 --- a/arch/arm/kernel/process.c +++ b/arch/arm/kernel/process.c @@ -243,7 +243,7 @@ int copy_thread(unsigned long clone_flags, unsigned long stack_start, thread->cpu_domain = get_domain(); #endif - if (likely(!(p->flags & PF_KTHREAD))) { + if (likely(!(p->flags & (PF_KTHREAD | PF_IO_WORKER)))) { *childregs = *current_pt_regs(); childregs->ARM_r0 = 0; if (stack_start) diff --git a/arch/arm64/kernel/process.c b/arch/arm64/kernel/process.c index 4cc1ccc8d6ab..325c83b1a24d 100644 --- a/arch/arm64/kernel/process.c +++ b/arch/arm64/kernel/process.c @@ -398,7 +398,7 @@ int copy_thread(unsigned long clone_flags, unsigned long stack_start, ptrauth_thread_init_kernel(p); - if (likely(!(p->flags & PF_KTHREAD))) { + if (likely(!(p->flags & (PF_KTHREAD | PF_IO_WORKER)))) { *childregs = *current_pt_regs(); childregs->regs[0] = 0; diff --git a/arch/csky/Kconfig b/arch/csky/Kconfig index 89dd2fcf38fa..34e91224adc3 100644 --- a/arch/csky/Kconfig +++ b/arch/csky/Kconfig @@ -7,7 +7,7 @@ config CSKY select ARCH_HAS_SYNC_DMA_FOR_CPU select ARCH_HAS_SYNC_DMA_FOR_DEVICE select ARCH_USE_BUILTIN_BSWAP - select ARCH_USE_QUEUED_RWLOCKS if NR_CPUS>2 + select ARCH_USE_QUEUED_RWLOCKS select ARCH_WANT_FRAME_POINTERS if !CPU_CK610 select ARCH_WANT_DEFAULT_TOPDOWN_MMAP_LAYOUT select COMMON_CLK @@ -35,6 +35,9 @@ config CSKY select GENERIC_IRQ_MULTI_HANDLER select GENERIC_SCHED_CLOCK select GENERIC_SMP_IDLE_THREAD + select GENERIC_TIME_VSYSCALL + select GENERIC_VDSO_32 + select GENERIC_GETTIMEOFDAY select GX6605S_TIMER if CPU_CK610 select HAVE_ARCH_TRACEHOOK select HAVE_ARCH_AUDITSYSCALL @@ -43,11 +46,14 @@ config CSKY select HAVE_CONTEXT_TRACKING select HAVE_VIRT_CPU_ACCOUNTING_GEN select HAVE_DEBUG_BUGVERBOSE + select HAVE_DEBUG_KMEMLEAK select HAVE_DYNAMIC_FTRACE select HAVE_DYNAMIC_FTRACE_WITH_REGS + select HAVE_GENERIC_VDSO select HAVE_FUNCTION_TRACER select HAVE_FUNCTION_GRAPH_TRACER select HAVE_FUNCTION_ERROR_INJECTION + select HAVE_FUTEX_CMPXCHG if FUTEX && SMP select HAVE_FTRACE_MCOUNT_RECORD select HAVE_KERNEL_GZIP select HAVE_KERNEL_LZO @@ -192,6 +198,22 @@ config CPU_CK860 endchoice choice + prompt "PAGE OFFSET" + default PAGE_OFFSET_80000000 + +config PAGE_OFFSET_80000000 + bool "PAGE OFFSET 2G (user:kernel = 2:2)" + +config PAGE_OFFSET_A0000000 + bool "PAGE OFFSET 2.5G (user:kernel = 2.5:1.5)" +endchoice + +config PAGE_OFFSET + hex + default 0x80000000 if PAGE_OFFSET_80000000 + default 0xa0000000 if PAGE_OFFSET_A0000000 +choice + prompt "C-SKY PMU type" depends on PERF_EVENTS depends on CPU_CK807 || CPU_CK810 || CPU_CK860 diff --git a/arch/csky/abiv1/inc/abi/cacheflush.h b/arch/csky/abiv1/inc/abi/cacheflush.h index d3e04208d53c..6cab7afae962 100644 --- a/arch/csky/abiv1/inc/abi/cacheflush.h +++ b/arch/csky/abiv1/inc/abi/cacheflush.h @@ -1,5 +1,4 @@ /* SPDX-License-Identifier: GPL-2.0 */ -// Copyright (C) 2018 Hangzhou C-SKY Microsystems co.,ltd. #ifndef __ABI_CSKY_CACHEFLUSH_H #define __ABI_CSKY_CACHEFLUSH_H diff --git a/arch/csky/abiv1/inc/abi/ckmmu.h b/arch/csky/abiv1/inc/abi/ckmmu.h index ba8eb5870835..416b30c57983 100644 --- a/arch/csky/abiv1/inc/abi/ckmmu.h +++ b/arch/csky/abiv1/inc/abi/ckmmu.h @@ -1,5 +1,4 @@ /* SPDX-License-Identifier: GPL-2.0 */ -// Copyright (C) 2018 Hangzhou C-SKY Microsystems co.,ltd. #ifndef __ASM_CSKY_CKMMUV1_H #define __ASM_CSKY_CKMMUV1_H @@ -89,13 +88,14 @@ static inline void tlb_invalid_indexed(void) cpwcr("cpcr8", 0x02000000); } -static inline void setup_pgd(unsigned long pgd, bool kernel) +static inline void setup_pgd(pgd_t *pgd, int asid) { - cpwcr("cpcr29", pgd | BIT(0)); + cpwcr("cpcr29", __pa(pgd) | BIT(0)); + write_mmu_entryhi(asid); } -static inline unsigned long get_pgd(void) +static inline pgd_t *get_pgd(void) { - return cprcr("cpcr29") & ~BIT(0); + return __va(cprcr("cpcr29") & ~BIT(0)); } #endif /* __ASM_CSKY_CKMMUV1_H */ diff --git a/arch/csky/abiv1/inc/abi/entry.h b/arch/csky/abiv1/inc/abi/entry.h index 13c23e2c707c..b6a2109b895e 100644 --- a/arch/csky/abiv1/inc/abi/entry.h +++ b/arch/csky/abiv1/inc/abi/entry.h @@ -1,5 +1,4 @@ /* SPDX-License-Identifier: GPL-2.0 */ -// Copyright (C) 2018 Hangzhou C-SKY Microsystems co.,ltd. #ifndef __ASM_CSKY_ENTRY_H #define __ASM_CSKY_ENTRY_H diff --git a/arch/csky/abiv1/inc/abi/page.h b/arch/csky/abiv1/inc/abi/page.h index c864519117c7..2d2159933b76 100644 --- a/arch/csky/abiv1/inc/abi/page.h +++ b/arch/csky/abiv1/inc/abi/page.h @@ -1,5 +1,4 @@ /* SPDX-License-Identifier: GPL-2.0 */ -// Copyright (C) 2018 Hangzhou C-SKY Microsystems co.,ltd. #include <asm/shmparam.h> diff --git a/arch/csky/abiv1/inc/abi/pgtable-bits.h b/arch/csky/abiv1/inc/abi/pgtable-bits.h index d605445aad9a..752c8b3f9194 100644 --- a/arch/csky/abiv1/inc/abi/pgtable-bits.h +++ b/arch/csky/abiv1/inc/abi/pgtable-bits.h @@ -1,37 +1,49 @@ /* SPDX-License-Identifier: GPL-2.0 */ -// Copyright (C) 2018 Hangzhou C-SKY Microsystems co.,ltd. #ifndef __ASM_CSKY_PGTABLE_BITS_H #define __ASM_CSKY_PGTABLE_BITS_H /* implemented in software */ -#define _PAGE_ACCESSED (1<<3) -#define PAGE_ACCESSED_BIT (3) - +#define _PAGE_PRESENT (1<<0) #define _PAGE_READ (1<<1) #define _PAGE_WRITE (1<<2) -#define _PAGE_PRESENT (1<<0) - +#define _PAGE_ACCESSED (1<<3) #define _PAGE_MODIFIED (1<<4) -#define PAGE_MODIFIED_BIT (4) /* implemented in hardware */ #define _PAGE_GLOBAL (1<<6) - #define _PAGE_VALID (1<<7) -#define PAGE_VALID_BIT (7) - #define _PAGE_DIRTY (1<<8) -#define PAGE_DIRTY_BIT (8) #define _PAGE_CACHE (3<<9) #define _PAGE_UNCACHE (2<<9) #define _PAGE_SO _PAGE_UNCACHE - #define _CACHE_MASK (7<<9) -#define _CACHE_CACHED (_PAGE_VALID | _PAGE_CACHE) -#define _CACHE_UNCACHED (_PAGE_VALID | _PAGE_UNCACHE) +#define _CACHE_CACHED _PAGE_CACHE +#define _CACHE_UNCACHED _PAGE_UNCACHE + +#define _PAGE_PROT_NONE _PAGE_READ + +/* + * Encode and decode a swap entry + * + * Format of swap PTE: + * bit 0: _PAGE_PRESENT (zero) + * bit 1: _PAGE_READ (zero) + * bit 2 - 5: swap type[0 - 3] + * bit 6: _PAGE_GLOBAL (zero) + * bit 7: _PAGE_VALID (zero) + * bit 8: swap type[4] + * bit 9 - 31: swap offset + */ +#define __swp_type(x) ((((x).val >> 2) & 0xf) | \ + (((x).val >> 4) & 0x10)) +#define __swp_offset(x) ((x).val >> 9) +#define __swp_entry(type, offset) ((swp_entry_t) { \ + ((type & 0xf) << 2) | \ + ((type & 0x10) << 4) | \ + ((offset) << 9)}) #define HAVE_ARCH_UNMAPPED_AREA diff --git a/arch/csky/abiv1/inc/abi/reg_ops.h b/arch/csky/abiv1/inc/abi/reg_ops.h index a153bd3918f7..abd01a243388 100644 --- a/arch/csky/abiv1/inc/abi/reg_ops.h +++ b/arch/csky/abiv1/inc/abi/reg_ops.h @@ -1,5 +1,4 @@ /* SPDX-License-Identifier: GPL-2.0 */ -// Copyright (C) 2018 Hangzhou C-SKY Microsystems co.,ltd. #ifndef __ABI_REG_OPS_H #define __ABI_REG_OPS_H diff --git a/arch/csky/abiv1/inc/abi/regdef.h b/arch/csky/abiv1/inc/abi/regdef.h index 104707fbdcc1..7b386fd67070 100644 --- a/arch/csky/abiv1/inc/abi/regdef.h +++ b/arch/csky/abiv1/inc/abi/regdef.h @@ -1,10 +1,14 @@ /* SPDX-License-Identifier: GPL-2.0 */ -// Copyright (C) 2018 Hangzhou C-SKY Microsystems co.,ltd. #ifndef __ASM_CSKY_REGDEF_H #define __ASM_CSKY_REGDEF_H +#ifdef __ASSEMBLY__ #define syscallid r1 +#else +#define syscallid "r1" +#endif + #define regs_syscallid(regs) regs->regs[9] #define regs_fp(regs) regs->regs[2] diff --git a/arch/csky/abiv1/inc/abi/string.h b/arch/csky/abiv1/inc/abi/string.h index 0cd43384f8d2..9d95594b0feb 100644 --- a/arch/csky/abiv1/inc/abi/string.h +++ b/arch/csky/abiv1/inc/abi/string.h @@ -1,5 +1,4 @@ /* SPDX-License-Identifier: GPL-2.0 */ -// Copyright (C) 2018 Hangzhou C-SKY Microsystems co.,ltd. #ifndef __ABI_CSKY_STRING_H #define __ABI_CSKY_STRING_H diff --git a/arch/csky/abiv1/inc/abi/switch_context.h b/arch/csky/abiv1/inc/abi/switch_context.h index 17c82686498e..ec73fd7c9f87 100644 --- a/arch/csky/abiv1/inc/abi/switch_context.h +++ b/arch/csky/abiv1/inc/abi/switch_context.h @@ -1,5 +1,4 @@ /* SPDX-License-Identifier: GPL-2.0 */ -// Copyright (C) 2018 Hangzhou C-SKY Microsystems co.,ltd. #ifndef __ABI_CSKY_PTRACE_H #define __ABI_CSKY_PTRACE_H diff --git a/arch/csky/abiv1/inc/abi/vdso.h b/arch/csky/abiv1/inc/abi/vdso.h index 14352f524f1d..9e6d0a2fdd2b 100644 --- a/arch/csky/abiv1/inc/abi/vdso.h +++ b/arch/csky/abiv1/inc/abi/vdso.h @@ -1,17 +1,9 @@ /* SPDX-License-Identifier: GPL-2.0 */ -#include <linux/uaccess.h> +#ifndef __ABI_CSKY_VDSO_H +#define __ABI_CSKY_VDSO_H -static inline int setup_vdso_page(unsigned short *ptr) -{ - int err = 0; +/* movi r1, 127; addi r1, (139 - 127) */ +#define SET_SYSCALL_ID .long 0x20b167f1 - /* movi r1, 127 */ - err |= __put_user(0x67f1, ptr + 0); - /* addi r1, (139 - 127) */ - err |= __put_user(0x20b1, ptr + 1); - /* trap 0 */ - err |= __put_user(0x0008, ptr + 2); - - return err; -} +#endif /* __ABI_CSKY_VDSO_H */ diff --git a/arch/csky/abiv2/cacheflush.c b/arch/csky/abiv2/cacheflush.c index 790f1ebfba44..39c51399dd81 100644 --- a/arch/csky/abiv2/cacheflush.c +++ b/arch/csky/abiv2/cacheflush.c @@ -12,6 +12,9 @@ void update_mmu_cache(struct vm_area_struct *vma, unsigned long address, unsigned long addr; struct page *page; + if (!pfn_valid(pte_pfn(*pte))) + return; + page = pfn_to_page(pte_pfn(*pte)); if (page == ZERO_PAGE(0)) return; diff --git a/arch/csky/abiv2/inc/abi/ckmmu.h b/arch/csky/abiv2/inc/abi/ckmmu.h index 73ded7c72482..64215f2380f1 100644 --- a/arch/csky/abiv2/inc/abi/ckmmu.h +++ b/arch/csky/abiv2/inc/abi/ckmmu.h @@ -1,5 +1,4 @@ /* SPDX-License-Identifier: GPL-2.0 */ -// Copyright (C) 2018 Hangzhou C-SKY Microsystems co.,ltd. #ifndef __ASM_CSKY_CKMMUV2_H #define __ASM_CSKY_CKMMUV2_H @@ -78,8 +77,13 @@ static inline void tlb_read(void) static inline void tlb_invalid_all(void) { #ifdef CONFIG_CPU_HAS_TLBI - asm volatile("tlbi.alls\n":::"memory"); sync_is(); + asm volatile( + "tlbi.alls \n" + "sync.i \n" + : + : + : "memory"); #else mtcr("cr<8, 15>", 0x04000000); #endif @@ -88,8 +92,13 @@ static inline void tlb_invalid_all(void) static inline void local_tlb_invalid_all(void) { #ifdef CONFIG_CPU_HAS_TLBI - asm volatile("tlbi.all\n":::"memory"); sync_is(); + asm volatile( + "tlbi.all \n" + "sync.i \n" + : + : + : "memory"); #else tlb_invalid_all(); #endif @@ -100,16 +109,31 @@ static inline void tlb_invalid_indexed(void) mtcr("cr<8, 15>", 0x02000000); } -static inline void setup_pgd(unsigned long pgd, bool kernel) +#define NOP32 ".long 0x4820c400\n" + +static inline void setup_pgd(pgd_t *pgd, int asid) { - if (kernel) - mtcr("cr<28, 15>", pgd | BIT(0)); - else - mtcr("cr<29, 15>", pgd | BIT(0)); +#ifdef CONFIG_CPU_HAS_TLBI + sync_is(); +#else + mb(); +#endif + asm volatile( +#ifdef CONFIG_CPU_HAS_TLBI + "mtcr %1, cr<28, 15> \n" +#endif + "mtcr %1, cr<29, 15> \n" + "mtcr %0, cr< 4, 15> \n" + ".rept 64 \n" + NOP32 + ".endr \n" + : + :"r"(asid), "r"(__pa(pgd) | BIT(0)) + :"memory"); } -static inline unsigned long get_pgd(void) +static inline pgd_t *get_pgd(void) { - return mfcr("cr<29, 15>") & ~BIT(0); + return __va(mfcr("cr<29, 15>") & ~BIT(0)); } #endif /* __ASM_CSKY_CKMMUV2_H */ diff --git a/arch/csky/abiv2/inc/abi/entry.h b/arch/csky/abiv2/inc/abi/entry.h index bedcc6f06bba..cca63e699b58 100644 --- a/arch/csky/abiv2/inc/abi/entry.h +++ b/arch/csky/abiv2/inc/abi/entry.h @@ -1,5 +1,4 @@ /* SPDX-License-Identifier: GPL-2.0 */ -// Copyright (C) 2018 Hangzhou C-SKY Microsystems co.,ltd. #ifndef __ASM_CSKY_ENTRY_H #define __ASM_CSKY_ENTRY_H @@ -26,6 +25,9 @@ stw tls, (sp, 0) stw lr, (sp, 4) + RD_MEH lr + WR_MEH lr + mfcr lr, epc movi tls, \epc_inc add lr, tls @@ -231,6 +233,16 @@ mtcr \rx, cr<8, 15> .endm +#ifdef CONFIG_PAGE_OFFSET_80000000 +#define MSA_SET cr<30, 15> +#define MSA_CLR cr<31, 15> +#endif + +#ifdef CONFIG_PAGE_OFFSET_A0000000 +#define MSA_SET cr<31, 15> +#define MSA_CLR cr<30, 15> +#endif + .macro SETUP_MMU /* Init psr and enable ee */ lrw r6, DEFAULT_PSR_VALUE @@ -281,15 +293,15 @@ * 31 - 29 | 28 - 9 | 8 | 7 | 6 | 5 | 4 | 3 | 2 | 1 | 0 * BA Reserved SH WA B SO SEC C D V */ - mfcr r6, cr<30, 15> /* Get MSA0 */ + mfcr r6, MSA_SET /* Get MSA */ 2: lsri r6, 29 lsli r6, 29 addi r6, 0x1ce - mtcr r6, cr<30, 15> /* Set MSA0 */ + mtcr r6, MSA_SET /* Set MSA */ movi r6, 0 - mtcr r6, cr<31, 15> /* Clr MSA1 */ + mtcr r6, MSA_CLR /* Clr MSA */ /* enable MMU */ mfcr r6, cr18 diff --git a/arch/csky/abiv2/inc/abi/fpu.h b/arch/csky/abiv2/inc/abi/fpu.h index 09e2700a3693..aabb79355013 100644 --- a/arch/csky/abiv2/inc/abi/fpu.h +++ b/arch/csky/abiv2/inc/abi/fpu.h @@ -1,5 +1,4 @@ /* SPDX-License-Identifier: GPL-2.0 */ -// Copyright (C) 2018 Hangzhou C-SKY Microsystems co.,ltd. #ifndef __ASM_CSKY_FPU_H #define __ASM_CSKY_FPU_H diff --git a/arch/csky/abiv2/inc/abi/page.h b/arch/csky/abiv2/inc/abi/page.h index 0a70cb553dca..cf005f13cd15 100644 --- a/arch/csky/abiv2/inc/abi/page.h +++ b/arch/csky/abiv2/inc/abi/page.h @@ -1,5 +1,4 @@ /* SPDX-License-Identifier: GPL-2.0 */ -// Copyright (C) 2018 Hangzhou C-SKY Microsystems co.,ltd. static inline void clear_user_page(void *addr, unsigned long vaddr, struct page *page) diff --git a/arch/csky/abiv2/inc/abi/pgtable-bits.h b/arch/csky/abiv2/inc/abi/pgtable-bits.h index 137f7932c83b..7e7f389f546f 100644 --- a/arch/csky/abiv2/inc/abi/pgtable-bits.h +++ b/arch/csky/abiv2/inc/abi/pgtable-bits.h @@ -1,37 +1,48 @@ /* SPDX-License-Identifier: GPL-2.0 */ -// Copyright (C) 2018 Hangzhou C-SKY Microsystems co.,ltd. #ifndef __ASM_CSKY_PGTABLE_BITS_H #define __ASM_CSKY_PGTABLE_BITS_H /* implemented in software */ #define _PAGE_ACCESSED (1<<7) -#define PAGE_ACCESSED_BIT (7) - #define _PAGE_READ (1<<8) #define _PAGE_WRITE (1<<9) #define _PAGE_PRESENT (1<<10) - #define _PAGE_MODIFIED (1<<11) -#define PAGE_MODIFIED_BIT (11) /* implemented in hardware */ #define _PAGE_GLOBAL (1<<0) - #define _PAGE_VALID (1<<1) -#define PAGE_VALID_BIT (1) - #define _PAGE_DIRTY (1<<2) -#define PAGE_DIRTY_BIT (2) #define _PAGE_SO (1<<5) #define _PAGE_BUF (1<<6) - #define _PAGE_CACHE (1<<3) - #define _CACHE_MASK _PAGE_CACHE -#define _CACHE_CACHED (_PAGE_VALID | _PAGE_CACHE | _PAGE_BUF) -#define _CACHE_UNCACHED (_PAGE_VALID) +#define _CACHE_CACHED (_PAGE_CACHE | _PAGE_BUF) +#define _CACHE_UNCACHED (0) + +#define _PAGE_PROT_NONE _PAGE_WRITE + +/* + * Encode and decode a swap entry + * + * Format of swap PTE: + * bit 0: _PAGE_GLOBAL (zero) + * bit 1: _PAGE_VALID (zero) + * bit 2 - 6: swap type + * bit 7 - 8: swap offset[0 - 1] + * bit 9: _PAGE_WRITE (zero) + * bit 10: _PAGE_PRESENT (zero) + * bit 11 - 31: swap offset[2 - 22] + */ +#define __swp_type(x) (((x).val >> 2) & 0x1f) +#define __swp_offset(x) ((((x).val >> 7) & 0x3) | \ + (((x).val >> 9) & 0x7ffffc)) +#define __swp_entry(type, offset) ((swp_entry_t) { \ + ((type & 0x1f) << 2) | \ + ((offset & 0x3) << 7) | \ + ((offset & 0x7ffffc) << 9)}) #endif /* __ASM_CSKY_PGTABLE_BITS_H */ diff --git a/arch/csky/abiv2/inc/abi/reg_ops.h b/arch/csky/abiv2/inc/abi/reg_ops.h index ae82c3f26a6b..49ba18a64751 100644 --- a/arch/csky/abiv2/inc/abi/reg_ops.h +++ b/arch/csky/abiv2/inc/abi/reg_ops.h @@ -1,5 +1,4 @@ /* SPDX-License-Identifier: GPL-2.0 */ -// Copyright (C) 2018 Hangzhou C-SKY Microsystems co.,ltd. #ifndef __ABI_REG_OPS_H #define __ABI_REG_OPS_H diff --git a/arch/csky/abiv2/inc/abi/regdef.h b/arch/csky/abiv2/inc/abi/regdef.h index d7328bbc1ce7..0933addbc27b 100644 --- a/arch/csky/abiv2/inc/abi/regdef.h +++ b/arch/csky/abiv2/inc/abi/regdef.h @@ -1,10 +1,14 @@ /* SPDX-License-Identifier: GPL-2.0 */ -// Copyright (C) 2018 Hangzhou C-SKY Microsystems co.,ltd. #ifndef __ASM_CSKY_REGDEF_H #define __ASM_CSKY_REGDEF_H +#ifdef __ASSEMBLY__ #define syscallid r7 +#else +#define syscallid "r7" +#endif + #define regs_syscallid(regs) regs->regs[3] #define regs_fp(regs) regs->regs[4] diff --git a/arch/csky/abiv2/inc/abi/switch_context.h b/arch/csky/abiv2/inc/abi/switch_context.h index 73a81245a3b3..5dd5c3f4ee7e 100644 --- a/arch/csky/abiv2/inc/abi/switch_context.h +++ b/arch/csky/abiv2/inc/abi/switch_context.h @@ -1,5 +1,4 @@ /* SPDX-License-Identifier: GPL-2.0 */ -// Copyright (C) 2018 Hangzhou C-SKY Microsystems co.,ltd. #ifndef __ABI_CSKY_PTRACE_H #define __ABI_CSKY_PTRACE_H diff --git a/arch/csky/abiv2/inc/abi/vdso.h b/arch/csky/abiv2/inc/abi/vdso.h index b60d4a070326..40fd10d893ff 100644 --- a/arch/csky/abiv2/inc/abi/vdso.h +++ b/arch/csky/abiv2/inc/abi/vdso.h @@ -3,21 +3,7 @@ #ifndef __ABI_CSKY_VDSO_H #define __ABI_CSKY_VDSO_H -#include <linux/uaccess.h> +/* movi r7, 173 */ +#define SET_SYSCALL_ID .long 0x008bea07 -static inline int setup_vdso_page(unsigned short *ptr) -{ - int err = 0; - - /* movi r7, 173 */ - err |= __put_user(0xea07, ptr); - err |= __put_user(0x008b, ptr+1); - - /* trap 0 */ - err |= __put_user(0xc000, ptr+2); - err |= __put_user(0x2020, ptr+3); - - return err; -} - -#endif /* __ABI_CSKY_STRING_H */ +#endif /* __ABI_CSKY_VDSO_H */ diff --git a/arch/csky/abiv2/sysdep.h b/arch/csky/abiv2/sysdep.h index bbbedfd34777..61abe9201c50 100644 --- a/arch/csky/abiv2/sysdep.h +++ b/arch/csky/abiv2/sysdep.h @@ -1,5 +1,4 @@ /* SPDX-License-Identifier: GPL-2.0 */ -// Copyright (C) 2018 Hangzhou C-SKY Microsystems co.,ltd. #ifndef __SYSDEP_H #define __SYSDEP_H diff --git a/arch/csky/include/asm/addrspace.h b/arch/csky/include/asm/addrspace.h index d1c2ede692ed..6fc05d44536c 100644 --- a/arch/csky/include/asm/addrspace.h +++ b/arch/csky/include/asm/addrspace.h @@ -1,5 +1,4 @@ /* SPDX-License-Identifier: GPL-2.0 */ -// Copyright (C) 2018 Hangzhou C-SKY Microsystems co.,ltd. #ifndef __ASM_CSKY_ADDRSPACE_H #define __ASM_CSKY_ADDRSPACE_H diff --git a/arch/csky/include/asm/atomic.h b/arch/csky/include/asm/atomic.h deleted file mode 100644 index e369d73b13e3..000000000000 --- a/arch/csky/include/asm/atomic.h +++ /dev/null @@ -1,212 +0,0 @@ -/* SPDX-License-Identifier: GPL-2.0 */ - -#ifndef __ASM_CSKY_ATOMIC_H -#define __ASM_CSKY_ATOMIC_H - -#include <linux/version.h> -#include <asm/cmpxchg.h> -#include <asm/barrier.h> - -#ifdef CONFIG_CPU_HAS_LDSTEX - -#define __atomic_add_unless __atomic_add_unless -static inline int __atomic_add_unless(atomic_t *v, int a, int u) -{ - unsigned long tmp, ret; - - smp_mb(); - - asm volatile ( - "1: ldex.w %0, (%3) \n" - " mov %1, %0 \n" - " cmpne %0, %4 \n" - " bf 2f \n" - " add %0, %2 \n" - " stex.w %0, (%3) \n" - " bez %0, 1b \n" - "2: \n" - : "=&r" (tmp), "=&r" (ret) - : "r" (a), "r"(&v->counter), "r"(u) - : "memory"); - - if (ret != u) - smp_mb(); - - return ret; -} - -#define ATOMIC_OP(op, c_op) \ -static inline void atomic_##op(int i, atomic_t *v) \ -{ \ - unsigned long tmp; \ - \ - asm volatile ( \ - "1: ldex.w %0, (%2) \n" \ - " " #op " %0, %1 \n" \ - " stex.w %0, (%2) \n" \ - " bez %0, 1b \n" \ - : "=&r" (tmp) \ - : "r" (i), "r"(&v->counter) \ - : "memory"); \ -} - -#define ATOMIC_OP_RETURN(op, c_op) \ -static inline int atomic_##op##_return(int i, atomic_t *v) \ -{ \ - unsigned long tmp, ret; \ - \ - smp_mb(); \ - asm volatile ( \ - "1: ldex.w %0, (%3) \n" \ - " " #op " %0, %2 \n" \ - " mov %1, %0 \n" \ - " stex.w %0, (%3) \n" \ - " bez %0, 1b \n" \ - : "=&r" (tmp), "=&r" (ret) \ - : "r" (i), "r"(&v->counter) \ - : "memory"); \ - smp_mb(); \ - \ - return ret; \ -} - -#define ATOMIC_FETCH_OP(op, c_op) \ -static inline int atomic_fetch_##op(int i, atomic_t *v) \ -{ \ - unsigned long tmp, ret; \ - \ - smp_mb(); \ - asm volatile ( \ - "1: ldex.w %0, (%3) \n" \ - " mov %1, %0 \n" \ - " " #op " %0, %2 \n" \ - " stex.w %0, (%3) \n" \ - " bez %0, 1b \n" \ - : "=&r" (tmp), "=&r" (ret) \ - : "r" (i), "r"(&v->counter) \ - : "memory"); \ - smp_mb(); \ - \ - return ret; \ -} - -#else /* CONFIG_CPU_HAS_LDSTEX */ - -#include <linux/irqflags.h> - -#define __atomic_add_unless __atomic_add_unless -static inline int __atomic_add_unless(atomic_t *v, int a, int u) -{ - unsigned long tmp, ret, flags; - - raw_local_irq_save(flags); - - asm volatile ( - " ldw %0, (%3) \n" - " mov %1, %0 \n" - " cmpne %0, %4 \n" - " bf 2f \n" - " add %0, %2 \n" - " stw %0, (%3) \n" - "2: \n" - : "=&r" (tmp), "=&r" (ret) - : "r" (a), "r"(&v->counter), "r"(u) - : "memory"); - - raw_local_irq_restore(flags); - - return ret; -} - -#define ATOMIC_OP(op, c_op) \ -static inline void atomic_##op(int i, atomic_t *v) \ -{ \ - unsigned long tmp, flags; \ - \ - raw_local_irq_save(flags); \ - \ - asm volatile ( \ - " ldw %0, (%2) \n" \ - " " #op " %0, %1 \n" \ - " stw %0, (%2) \n" \ - : "=&r" (tmp) \ - : "r" (i), "r"(&v->counter) \ - : "memory"); \ - \ - raw_local_irq_restore(flags); \ -} - -#define ATOMIC_OP_RETURN(op, c_op) \ -static inline int atomic_##op##_return(int i, atomic_t *v) \ -{ \ - unsigned long tmp, ret, flags; \ - \ - raw_local_irq_save(flags); \ - \ - asm volatile ( \ - " ldw %0, (%3) \n" \ - " " #op " %0, %2 \n" \ - " stw %0, (%3) \n" \ - " mov %1, %0 \n" \ - : "=&r" (tmp), "=&r" (ret) \ - : "r" (i), "r"(&v->counter) \ - : "memory"); \ - \ - raw_local_irq_restore(flags); \ - \ - return ret; \ -} - -#define ATOMIC_FETCH_OP(op, c_op) \ -static inline int atomic_fetch_##op(int i, atomic_t *v) \ -{ \ - unsigned long tmp, ret, flags; \ - \ - raw_local_irq_save(flags); \ - \ - asm volatile ( \ - " ldw %0, (%3) \n" \ - " mov %1, %0 \n" \ - " " #op " %0, %2 \n" \ - " stw %0, (%3) \n" \ - : "=&r" (tmp), "=&r" (ret) \ - : "r" (i), "r"(&v->counter) \ - : "memory"); \ - \ - raw_local_irq_restore(flags); \ - \ - return ret; \ -} - -#endif /* CONFIG_CPU_HAS_LDSTEX */ - -#define atomic_add_return atomic_add_return -ATOMIC_OP_RETURN(add, +) -#define atomic_sub_return atomic_sub_return -ATOMIC_OP_RETURN(sub, -) - -#define atomic_fetch_add atomic_fetch_add -ATOMIC_FETCH_OP(add, +) -#define atomic_fetch_sub atomic_fetch_sub -ATOMIC_FETCH_OP(sub, -) -#define atomic_fetch_and atomic_fetch_and -ATOMIC_FETCH_OP(and, &) -#define atomic_fetch_or atomic_fetch_or -ATOMIC_FETCH_OP(or, |) -#define atomic_fetch_xor atomic_fetch_xor -ATOMIC_FETCH_OP(xor, ^) - -#define atomic_and atomic_and -ATOMIC_OP(and, &) -#define atomic_or atomic_or -ATOMIC_OP(or, |) -#define atomic_xor atomic_xor -ATOMIC_OP(xor, ^) - -#undef ATOMIC_FETCH_OP -#undef ATOMIC_OP_RETURN -#undef ATOMIC_OP - -#include <asm-generic/atomic.h> - -#endif /* __ASM_CSKY_ATOMIC_H */ diff --git a/arch/csky/include/asm/barrier.h b/arch/csky/include/asm/barrier.h index a430e7fddf35..84fc600c8b45 100644 --- a/arch/csky/include/asm/barrier.h +++ b/arch/csky/include/asm/barrier.h @@ -1,5 +1,4 @@ /* SPDX-License-Identifier: GPL-2.0 */ -// Copyright (C) 2018 Hangzhou C-SKY Microsystems co.,ltd. #ifndef __ASM_CSKY_BARRIER_H #define __ASM_CSKY_BARRIER_H @@ -8,6 +7,61 @@ #define nop() asm volatile ("nop\n":::"memory") +#ifdef CONFIG_SMP + +/* + * bar.brwarws: ordering barrier for all load/store instructions + * before/after + * + * |31|30 26|25 21|20 16|15 10|9 5|4 0| + * 1 10000 00000 00000 100001 00001 0 bw br aw ar + * + * b: before + * a: after + * r: read + * w: write + * + * Here are all combinations: + * + * bar.brw + * bar.br + * bar.bw + * bar.arw + * bar.ar + * bar.aw + * bar.brwarw + * bar.brarw + * bar.bwarw + * bar.brwar + * bar.brwaw + * bar.brar + * bar.bwaw + */ +#define __bar_brw() asm volatile (".long 0x842cc000\n":::"memory") +#define __bar_br() asm volatile (".long 0x8424c000\n":::"memory") +#define __bar_bw() asm volatile (".long 0x8428c000\n":::"memory") +#define __bar_arw() asm volatile (".long 0x8423c000\n":::"memory") +#define __bar_ar() asm volatile (".long 0x8421c000\n":::"memory") +#define __bar_aw() asm volatile (".long 0x8422c000\n":::"memory") +#define __bar_brwarw() asm volatile (".long 0x842fc000\n":::"memory") +#define __bar_brarw() asm volatile (".long 0x8427c000\n":::"memory") +#define __bar_bwarw() asm volatile (".long 0x842bc000\n":::"memory") +#define __bar_brwar() asm volatile (".long 0x842dc000\n":::"memory") +#define __bar_brwaw() asm volatile (".long 0x842ec000\n":::"memory") +#define __bar_brar() asm volatile (".long 0x8425c000\n":::"memory") +#define __bar_brar() asm volatile (".long 0x8425c000\n":::"memory") +#define __bar_bwaw() asm volatile (".long 0x842ac000\n":::"memory") + +#define __smp_mb() __bar_brwarw() +#define __smp_rmb() __bar_brar() +#define __smp_wmb() __bar_bwaw() + +#define ACQUIRE_FENCE ".long 0x8427c000\n" +#define __smp_acquire_fence() __bar_brarw() +#define __smp_release_fence() __bar_brwaw() + +#endif /* CONFIG_SMP */ + /* * sync: completion barrier, all sync.xx instructions * guarantee the last response recieved by bus transaction @@ -15,31 +69,14 @@ * sync.s: inherit from sync, but also shareable to other cores * sync.i: inherit from sync, but also flush cpu pipeline * sync.is: the same with sync.i + sync.s - * - * bar.brwarw: ordering barrier for all load/store instructions before it - * bar.brwarws: ordering barrier for all load/store instructions before it - * and shareable to other cores - * bar.brar: ordering barrier for all load instructions before it - * bar.brars: ordering barrier for all load instructions before it - * and shareable to other cores - * bar.bwaw: ordering barrier for all store instructions before it - * bar.bwaws: ordering barrier for all store instructions before it - * and shareable to other cores */ +#define mb() asm volatile ("sync\n":::"memory") #ifdef CONFIG_CPU_HAS_CACHEV2 -#define mb() asm volatile ("sync.s\n":::"memory") - -#ifdef CONFIG_SMP -#define __smp_mb() asm volatile ("bar.brwarws\n":::"memory") -#define __smp_rmb() asm volatile ("bar.brars\n":::"memory") -#define __smp_wmb() asm volatile ("bar.bwaws\n":::"memory") -#endif /* CONFIG_SMP */ - -#define sync_is() asm volatile ("sync.is\n":::"memory") - -#else /* !CONFIG_CPU_HAS_CACHEV2 */ -#define mb() asm volatile ("sync\n":::"memory") +/* + * Using three sync.is to prevent speculative PTW + */ +#define sync_is() asm volatile ("sync.is\nsync.is\nsync.is\n":::"memory") #endif #include <asm-generic/barrier.h> diff --git a/arch/csky/include/asm/bitops.h b/arch/csky/include/asm/bitops.h index 43b9838bff63..91818787d860 100644 --- a/arch/csky/include/asm/bitops.h +++ b/arch/csky/include/asm/bitops.h @@ -1,5 +1,4 @@ /* SPDX-License-Identifier: GPL-2.0 */ -// Copyright (C) 2018 Hangzhou C-SKY Microsystems co.,ltd. #ifndef __ASM_CSKY_BITOPS_H #define __ASM_CSKY_BITOPS_H diff --git a/arch/csky/include/asm/bug.h b/arch/csky/include/asm/bug.h index 33ebd16b9c78..03f1a5f9184a 100644 --- a/arch/csky/include/asm/bug.h +++ b/arch/csky/include/asm/bug.h @@ -1,5 +1,4 @@ /* SPDX-License-Identifier: GPL-2.0 */ -// Copyright (C) 2018 Hangzhou C-SKY Microsystems co.,ltd. #ifndef __ASM_CSKY_BUG_H #define __ASM_CSKY_BUG_H @@ -21,6 +20,8 @@ do { \ struct pt_regs; void die(struct pt_regs *regs, const char *str); +void do_trap(struct pt_regs *regs, int signo, int code, unsigned long addr); + void show_regs(struct pt_regs *regs); void show_code(struct pt_regs *regs); diff --git a/arch/csky/include/asm/cacheflush.h b/arch/csky/include/asm/cacheflush.h index f0b8f25429a2..d0f9eafe8988 100644 --- a/arch/csky/include/asm/cacheflush.h +++ b/arch/csky/include/asm/cacheflush.h @@ -1,5 +1,4 @@ /* SPDX-License-Identifier: GPL-2.0 */ -// Copyright (C) 2018 Hangzhou C-SKY Microsystems co.,ltd. #ifndef __ASM_CSKY_CACHEFLUSH_H #define __ASM_CSKY_CACHEFLUSH_H diff --git a/arch/csky/include/asm/checksum.h b/arch/csky/include/asm/checksum.h index 7685824291b1..aa12ef4b9080 100644 --- a/arch/csky/include/asm/checksum.h +++ b/arch/csky/include/asm/checksum.h @@ -1,5 +1,4 @@ /* SPDX-License-Identifier: GPL-2.0 */ -// Copyright (C) 2018 Hangzhou C-SKY Microsystems co.,ltd. #ifndef __ASM_CSKY_CHECKSUM_H #define __ASM_CSKY_CHECKSUM_H diff --git a/arch/csky/include/asm/clocksource.h b/arch/csky/include/asm/clocksource.h new file mode 100644 index 000000000000..54da0e49efa1 --- /dev/null +++ b/arch/csky/include/asm/clocksource.h @@ -0,0 +1,8 @@ +/* SPDX-License-Identifier: GPL-2.0 */ + +#ifndef __ASM_VDSO_CSKY_CLOCKSOURCE_H +#define __ASM_VDSO_CSKY_CLOCKSOURCE_H + +#include <asm/vdso/clocksource.h> + +#endif diff --git a/arch/csky/include/asm/cmpxchg.h b/arch/csky/include/asm/cmpxchg.h index 89224530a0ee..dabc8e46ce7b 100644 --- a/arch/csky/include/asm/cmpxchg.h +++ b/arch/csky/include/asm/cmpxchg.h @@ -3,12 +3,12 @@ #ifndef __ASM_CSKY_CMPXCHG_H #define __ASM_CSKY_CMPXCHG_H -#ifdef CONFIG_CPU_HAS_LDSTEX +#ifdef CONFIG_SMP #include <asm/barrier.h> extern void __bad_xchg(void); -#define __xchg(new, ptr, size) \ +#define __xchg_relaxed(new, ptr, size) \ ({ \ __typeof__(ptr) __ptr = (ptr); \ __typeof__(new) __new = (new); \ @@ -16,7 +16,6 @@ extern void __bad_xchg(void); unsigned long tmp; \ switch (size) { \ case 4: \ - smp_mb(); \ asm volatile ( \ "1: ldex.w %0, (%3) \n" \ " mov %1, %2 \n" \ @@ -25,7 +24,6 @@ extern void __bad_xchg(void); : "=&r" (__ret), "=&r" (tmp) \ : "r" (__new), "r"(__ptr) \ :); \ - smp_mb(); \ break; \ default: \ __bad_xchg(); \ @@ -33,9 +31,10 @@ extern void __bad_xchg(void); __ret; \ }) -#define xchg(ptr, x) (__xchg((x), (ptr), sizeof(*(ptr)))) +#define xchg_relaxed(ptr, x) \ + (__xchg_relaxed((x), (ptr), sizeof(*(ptr)))) -#define __cmpxchg(ptr, old, new, size) \ +#define __cmpxchg_relaxed(ptr, old, new, size) \ ({ \ __typeof__(ptr) __ptr = (ptr); \ __typeof__(new) __new = (new); \ @@ -44,7 +43,6 @@ extern void __bad_xchg(void); __typeof__(*(ptr)) __ret; \ switch (size) { \ case 4: \ - smp_mb(); \ asm volatile ( \ "1: ldex.w %0, (%3) \n" \ " cmpne %0, %4 \n" \ @@ -56,7 +54,6 @@ extern void __bad_xchg(void); : "=&r" (__ret), "=&r" (__tmp) \ : "r" (__new), "r"(__ptr), "r"(__old) \ :); \ - smp_mb(); \ break; \ default: \ __bad_xchg(); \ @@ -64,8 +61,18 @@ extern void __bad_xchg(void); __ret; \ }) -#define cmpxchg(ptr, o, n) \ - (__cmpxchg((ptr), (o), (n), sizeof(*(ptr)))) +#define cmpxchg_relaxed(ptr, o, n) \ + (__cmpxchg_relaxed((ptr), (o), (n), sizeof(*(ptr)))) + +#define cmpxchg(ptr, o, n) \ +({ \ + __typeof__(*(ptr)) __ret; \ + __smp_release_fence(); \ + __ret = cmpxchg_relaxed(ptr, o, n); \ + __smp_acquire_fence(); \ + __ret; \ +}) + #else #include <asm-generic/cmpxchg.h> #endif diff --git a/arch/csky/include/asm/elf.h b/arch/csky/include/asm/elf.h index eb2cc5a673b5..48b83e283ed4 100644 --- a/arch/csky/include/asm/elf.h +++ b/arch/csky/include/asm/elf.h @@ -1,5 +1,4 @@ /* SPDX-License-Identifier: GPL-2.0 */ -// Copyright (C) 2018 Hangzhou C-SKY Microsystems co.,ltd. #ifndef __ASM_CSKY_ELF_H #define __ASM_CSKY_ELF_H diff --git a/arch/csky/include/asm/fixmap.h b/arch/csky/include/asm/fixmap.h index 4b589cc20900..49a77cbbe2a9 100644 --- a/arch/csky/include/asm/fixmap.h +++ b/arch/csky/include/asm/fixmap.h @@ -1,5 +1,4 @@ /* SPDX-License-Identifier: GPL-2.0 */ -// Copyright (C) 2018 Hangzhou C-SKY Microsystems co.,ltd. #ifndef __ASM_CSKY_FIXMAP_H #define __ASM_CSKY_FIXMAP_H diff --git a/arch/csky/include/asm/ftrace.h b/arch/csky/include/asm/ftrace.h index fae72b0b1374..9b86341731b6 100644 --- a/arch/csky/include/asm/ftrace.h +++ b/arch/csky/include/asm/ftrace.h @@ -1,5 +1,4 @@ /* SPDX-License-Identifier: GPL-2.0 */ -// Copyright (C) 2018 Hangzhou C-SKY Microsystems co.,ltd. #ifndef __ASM_CSKY_FTRACE_H #define __ASM_CSKY_FTRACE_H diff --git a/arch/csky/include/asm/futex.h b/arch/csky/include/asm/futex.h new file mode 100644 index 000000000000..6cfd312723fa --- /dev/null +++ b/arch/csky/include/asm/futex.h @@ -0,0 +1,121 @@ +/* SPDX-License-Identifier: GPL-2.0 */ + +#ifndef __ASM_CSKY_FUTEX_H +#define __ASM_CSKY_FUTEX_H + +#ifndef CONFIG_SMP +#include <asm-generic/futex.h> +#else +#include <linux/atomic.h> +#include <linux/futex.h> +#include <linux/uaccess.h> +#include <linux/errno.h> + +#define __futex_atomic_op(insn, ret, oldval, uaddr, oparg) \ +{ \ + u32 tmp; \ + \ + __atomic_pre_full_fence(); \ + \ + __asm__ __volatile__ ( \ + "1: ldex.w %[ov], %[u] \n" \ + " "insn" \n" \ + "2: stex.w %[t], %[u] \n" \ + " bez %[t], 1b \n" \ + " br 4f \n" \ + "3: mov %[r], %[e] \n" \ + "4: \n" \ + " .section __ex_table,\"a\" \n" \ + " .balign 4 \n" \ + " .long 1b, 3b \n" \ + " .long 2b, 3b \n" \ + " .previous \n" \ + : [r] "+r" (ret), [ov] "=&r" (oldval), \ + [u] "+m" (*uaddr), [t] "=&r" (tmp) \ + : [op] "Jr" (oparg), [e] "jr" (-EFAULT) \ + : "memory"); \ + \ + __atomic_post_full_fence(); \ +} + +static inline int +arch_futex_atomic_op_inuser(int op, int oparg, int *oval, u32 __user *uaddr) +{ + int oldval = 0, ret = 0; + + if (!access_ok(uaddr, sizeof(u32))) + return -EFAULT; + + switch (op) { + case FUTEX_OP_SET: + __futex_atomic_op("mov %[t], %[ov]", + ret, oldval, uaddr, oparg); + break; + case FUTEX_OP_ADD: + __futex_atomic_op("add %[t], %[ov], %[op]", + ret, oldval, uaddr, oparg); + break; + case FUTEX_OP_OR: + __futex_atomic_op("or %[t], %[ov], %[op]", + ret, oldval, uaddr, oparg); + break; + case FUTEX_OP_ANDN: + __futex_atomic_op("and %[t], %[ov], %[op]", + ret, oldval, uaddr, ~oparg); + break; + case FUTEX_OP_XOR: + __futex_atomic_op("xor %[t], %[ov], %[op]", + ret, oldval, uaddr, oparg); + break; + default: + ret = -ENOSYS; + } + + if (!ret) + *oval = oldval; + + return ret; +} + + + +static inline int +futex_atomic_cmpxchg_inatomic(u32 *uval, u32 __user *uaddr, + u32 oldval, u32 newval) +{ + int ret = 0; + u32 val, tmp; + + if (!access_ok(uaddr, sizeof(u32))) + return -EFAULT; + + __atomic_pre_full_fence(); + + __asm__ __volatile__ ( + "1: ldex.w %[v], %[u] \n" + " cmpne %[v], %[ov] \n" + " bt 4f \n" + " mov %[t], %[nv] \n" + "2: stex.w %[t], %[u] \n" + " bez %[t], 1b \n" + " br 4f \n" + "3: mov %[r], %[e] \n" + "4: \n" + " .section __ex_table,\"a\" \n" + " .balign 4 \n" + " .long 1b, 3b \n" + " .long 2b, 3b \n" + " .previous \n" + : [r] "+r" (ret), [v] "=&r" (val), [u] "+m" (*uaddr), + [t] "=&r" (tmp) + : [ov] "Jr" (oldval), [nv] "Jr" (newval), [e] "Jr" (-EFAULT) + : "memory"); + + __atomic_post_full_fence(); + + *uval = val; + return ret; +} + +#endif /* CONFIG_SMP */ +#endif /* __ASM_CSKY_FUTEX_H */ diff --git a/arch/csky/include/asm/highmem.h b/arch/csky/include/asm/highmem.h index 1f4ed3f4c0d9..1ed810effb3d 100644 --- a/arch/csky/include/asm/highmem.h +++ b/arch/csky/include/asm/highmem.h @@ -1,5 +1,4 @@ /* SPDX-License-Identifier: GPL-2.0 */ -// Copyright (C) 2018 Hangzhou C-SKY Microsystems co.,ltd. #ifndef __ASM_CSKY_HIGHMEM_H #define __ASM_CSKY_HIGHMEM_H diff --git a/arch/csky/include/asm/io.h b/arch/csky/include/asm/io.h index e909587f24c5..f82654053dc0 100644 --- a/arch/csky/include/asm/io.h +++ b/arch/csky/include/asm/io.h @@ -1,5 +1,4 @@ /* SPDX-License-Identifier: GPL-2.0 */ -// Copyright (C) 2018 Hangzhou C-SKY Microsystems co.,ltd. #ifndef __ASM_CSKY_IO_H #define __ASM_CSKY_IO_H diff --git a/arch/csky/include/asm/memory.h b/arch/csky/include/asm/memory.h index a65c6759f537..d12179801ae3 100644 --- a/arch/csky/include/asm/memory.h +++ b/arch/csky/include/asm/memory.h @@ -10,7 +10,7 @@ #define FIXADDR_TOP _AC(0xffffc000, UL) #define PKMAP_BASE _AC(0xff800000, UL) -#define VMALLOC_START _AC(0xc0008000, UL) +#define VMALLOC_START (PAGE_OFFSET + LOWMEM_LIMIT + (PAGE_SIZE * 8)) #define VMALLOC_END (PKMAP_BASE - (PAGE_SIZE * 2)) #ifdef CONFIG_HAVE_TCM diff --git a/arch/csky/include/asm/mmu.h b/arch/csky/include/asm/mmu.h index 26fbb1d15df0..d78321901d06 100644 --- a/arch/csky/include/asm/mmu.h +++ b/arch/csky/include/asm/mmu.h @@ -1,5 +1,4 @@ /* SPDX-License-Identifier: GPL-2.0 */ -// Copyright (C) 2018 Hangzhou C-SKY Microsystems co.,ltd. #ifndef __ASM_CSKY_MMU_H #define __ASM_CSKY_MMU_H diff --git a/arch/csky/include/asm/mmu_context.h b/arch/csky/include/asm/mmu_context.h index b227d29393a8..95d99b30792c 100644 --- a/arch/csky/include/asm/mmu_context.h +++ b/arch/csky/include/asm/mmu_context.h @@ -1,5 +1,4 @@ /* SPDX-License-Identifier: GPL-2.0 */ -// Copyright (C) 2018 Hangzhou C-SKY Microsystems co.,ltd. #ifndef __ASM_CSKY_MMU_CONTEXT_H #define __ASM_CSKY_MMU_CONTEXT_H @@ -14,12 +13,6 @@ #include <linux/sched.h> #include <abi/ckmmu.h> -#define TLBMISS_HANDLER_SETUP_PGD(pgd) \ - setup_pgd(__pa(pgd), false) - -#define TLBMISS_HANDLER_SETUP_PGD_KERNEL(pgd) \ - setup_pgd(__pa(pgd), true) - #define ASID_MASK ((1 << CONFIG_CPU_ASID_BITS) - 1) #define cpu_asid(mm) (atomic64_read(&mm->context.asid) & ASID_MASK) @@ -36,8 +29,7 @@ switch_mm(struct mm_struct *prev, struct mm_struct *next, if (prev != next) check_and_switch_context(next, cpu); - TLBMISS_HANDLER_SETUP_PGD(next->pgd); - write_mmu_entryhi(next->context.asid.counter); + setup_pgd(next->pgd, next->context.asid.counter); flush_icache_deferred(next); } diff --git a/arch/csky/include/asm/page.h b/arch/csky/include/asm/page.h index 9b98bf31d57c..3b91fc3cf36f 100644 --- a/arch/csky/include/asm/page.h +++ b/arch/csky/include/asm/page.h @@ -24,7 +24,7 @@ * address region. We use them mapping kernel 1GB direct-map address area and * for more than 1GB of memory we use highmem. */ -#define PAGE_OFFSET 0x80000000 +#define PAGE_OFFSET CONFIG_PAGE_OFFSET #define SSEG_SIZE 0x20000000 #define LOWMEM_LIMIT (SSEG_SIZE * 2) diff --git a/arch/csky/include/asm/perf_event.h b/arch/csky/include/asm/perf_event.h index 572093e11001..249905d8a4e8 100644 --- a/arch/csky/include/asm/perf_event.h +++ b/arch/csky/include/asm/perf_event.h @@ -1,5 +1,4 @@ /* SPDX-License-Identifier: GPL-2.0 */ -// Copyright (C) 2018 Hangzhou C-SKY Microsystems co.,ltd. #ifndef __ASM_CSKY_PERF_EVENT_H #define __ASM_CSKY_PERF_EVENT_H diff --git a/arch/csky/include/asm/pgalloc.h b/arch/csky/include/asm/pgalloc.h index d58d8146b729..cd211aabbefd 100644 --- a/arch/csky/include/asm/pgalloc.h +++ b/arch/csky/include/asm/pgalloc.h @@ -1,5 +1,4 @@ /* SPDX-License-Identifier: GPL-2.0 */ -// Copyright (C) 2018 Hangzhou C-SKY Microsystems co.,ltd. #ifndef __ASM_CSKY_PGALLOC_H #define __ASM_CSKY_PGALLOC_H @@ -71,7 +70,7 @@ do { \ } while (0) extern void pagetable_init(void); -extern void pre_mmu_init(void); +extern void mmu_init(unsigned long min_pfn, unsigned long max_pfn); extern void pre_trap_init(void); #endif /* __ASM_CSKY_PGALLOC_H */ diff --git a/arch/csky/include/asm/pgtable.h b/arch/csky/include/asm/pgtable.h index 2002cb7f1053..0d60367b6bfa 100644 --- a/arch/csky/include/asm/pgtable.h +++ b/arch/csky/include/asm/pgtable.h @@ -1,5 +1,4 @@ /* SPDX-License-Identifier: GPL-2.0 */ -// Copyright (C) 2018 Hangzhou C-SKY Microsystems co.,ltd. #ifndef __ASM_CSKY_PGTABLE_H #define __ASM_CSKY_PGTABLE_H @@ -14,7 +13,7 @@ #define PGDIR_SIZE (1UL << PGDIR_SHIFT) #define PGDIR_MASK (~(PGDIR_SIZE-1)) -#define USER_PTRS_PER_PGD (0x80000000UL/PGDIR_SIZE) +#define USER_PTRS_PER_PGD (PAGE_OFFSET/PGDIR_SIZE) #define FIRST_USER_ADDRESS 0UL /* @@ -34,23 +33,13 @@ #define pmd_page(pmd) (pfn_to_page(pmd_phys(pmd) >> PAGE_SHIFT)) #define pte_clear(mm, addr, ptep) set_pte((ptep), \ - (((unsigned int) addr & PAGE_OFFSET) ? __pte(_PAGE_GLOBAL) : __pte(0))) + (((unsigned int) addr >= PAGE_OFFSET) ? __pte(_PAGE_GLOBAL) : __pte(0))) #define pte_none(pte) (!(pte_val(pte) & ~_PAGE_GLOBAL)) #define pte_present(pte) (pte_val(pte) & _PAGE_PRESENT) #define pte_pfn(x) ((unsigned long)((x).pte_low >> PAGE_SHIFT)) #define pfn_pte(pfn, prot) __pte(((unsigned long long)(pfn) << PAGE_SHIFT) \ | pgprot_val(prot)) -#define __READABLE (_PAGE_READ | _PAGE_VALID | _PAGE_ACCESSED) -#define __WRITEABLE (_PAGE_WRITE | _PAGE_DIRTY | _PAGE_MODIFIED) - -#define _PAGE_CHG_MASK (PAGE_MASK | _PAGE_ACCESSED | _PAGE_MODIFIED | \ - _CACHE_MASK) - -#define __swp_type(x) (((x).val >> 4) & 0xff) -#define __swp_offset(x) ((x).val >> 12) -#define __swp_entry(type, offset) ((swp_entry_t) {((type) << 4) | \ - ((offset) << 12) }) #define __pte_to_swp_entry(pte) ((swp_entry_t) { pte_val(pte) }) #define __swp_entry_to_pte(x) ((pte_t) { (x).val }) @@ -59,41 +48,52 @@ pgprot_val(pgprot)) /* - * CSKY can't do page protection for execute, and considers that the same like - * read. Also, write permissions imply read permissions. This is the closest - * we can get by reasonable means.. + * C-SKY only has VALID and DIRTY bit in hardware. So we need to use the + * two bits emulate PRESENT, READ, WRITE, EXEC, MODIFIED, ACCESSED. */ -#define PAGE_NONE __pgprot(_PAGE_PRESENT | _CACHE_CACHED) -#define PAGE_SHARED __pgprot(_PAGE_PRESENT | _PAGE_READ | _PAGE_WRITE | \ +#define _PAGE_BASE (_PAGE_PRESENT | _PAGE_ACCESSED) + +#define PAGE_NONE __pgprot(_PAGE_PROT_NONE) +#define PAGE_READ __pgprot(_PAGE_BASE | _PAGE_READ | \ _CACHE_CACHED) -#define PAGE_COPY __pgprot(_PAGE_PRESENT | _PAGE_READ | _CACHE_CACHED) -#define PAGE_READONLY __pgprot(_PAGE_PRESENT | _PAGE_READ | _CACHE_CACHED) -#define PAGE_KERNEL __pgprot(_PAGE_PRESENT | __READABLE | __WRITEABLE | \ - _PAGE_GLOBAL | _CACHE_CACHED) -#define PAGE_USERIO __pgprot(_PAGE_PRESENT | _PAGE_READ | _PAGE_WRITE | \ +#define PAGE_WRITE __pgprot(_PAGE_BASE | _PAGE_READ | _PAGE_WRITE | \ _CACHE_CACHED) +#define PAGE_SHARED PAGE_WRITE + +#define PAGE_KERNEL __pgprot(_PAGE_BASE | _PAGE_READ | _PAGE_VALID | \ + _PAGE_WRITE | _PAGE_DIRTY | _PAGE_MODIFIED | \ + _PAGE_GLOBAL | \ + _CACHE_CACHED) + +#define _PAGE_IOREMAP (_PAGE_BASE | _PAGE_READ | _PAGE_VALID | \ + _PAGE_WRITE | _PAGE_DIRTY | _PAGE_MODIFIED | \ + _PAGE_GLOBAL | \ + _CACHE_UNCACHED | _PAGE_SO) + +#define _PAGE_CHG_MASK (~(unsigned long) \ + (_PAGE_PRESENT | _PAGE_READ | _PAGE_WRITE | \ + _CACHE_MASK | _PAGE_GLOBAL)) -#define _PAGE_IOREMAP \ - (_PAGE_PRESENT | __READABLE | __WRITEABLE | _PAGE_GLOBAL | \ - _CACHE_UNCACHED | _PAGE_SO) +#define MAX_SWAPFILES_CHECK() \ + BUILD_BUG_ON(MAX_SWAPFILES_SHIFT != 5) #define __P000 PAGE_NONE -#define __P001 PAGE_READONLY -#define __P010 PAGE_COPY -#define __P011 PAGE_COPY -#define __P100 PAGE_READONLY -#define __P101 PAGE_READONLY -#define __P110 PAGE_COPY -#define __P111 PAGE_COPY +#define __P001 PAGE_READ +#define __P010 PAGE_READ +#define __P011 PAGE_READ +#define __P100 PAGE_READ +#define __P101 PAGE_READ +#define __P110 PAGE_READ +#define __P111 PAGE_READ #define __S000 PAGE_NONE -#define __S001 PAGE_READONLY -#define __S010 PAGE_SHARED -#define __S011 PAGE_SHARED -#define __S100 PAGE_READONLY -#define __S101 PAGE_READONLY -#define __S110 PAGE_SHARED -#define __S111 PAGE_SHARED +#define __S001 PAGE_READ +#define __S010 PAGE_WRITE +#define __S011 PAGE_WRITE +#define __S100 PAGE_READ +#define __S101 PAGE_READ +#define __S110 PAGE_WRITE +#define __S111 PAGE_WRITE extern unsigned long empty_zero_page[PAGE_SIZE / sizeof(unsigned long)]; #define ZERO_PAGE(vaddr) (virt_to_page(empty_zero_page)) diff --git a/arch/csky/include/asm/processor.h b/arch/csky/include/asm/processor.h index 4800f6563abb..9e933021fe8e 100644 --- a/arch/csky/include/asm/processor.h +++ b/arch/csky/include/asm/processor.h @@ -1,5 +1,4 @@ /* SPDX-License-Identifier: GPL-2.0 */ -// Copyright (C) 2018 Hangzhou C-SKY Microsystems co.,ltd. #ifndef __ASM_CSKY_PROCESSOR_H #define __ASM_CSKY_PROCESSOR_H @@ -28,7 +27,7 @@ extern struct cpuinfo_csky cpu_data[]; * for a 64 bit kernel expandable to 8192EB, of which the current CSKY * implementations will "only" be able to use 1TB ... */ -#define TASK_SIZE 0x7fff8000UL +#define TASK_SIZE (PAGE_OFFSET - (PAGE_SIZE * 8)) #ifdef __KERNEL__ #define STACK_TOP TASK_SIZE diff --git a/arch/csky/include/asm/ptrace.h b/arch/csky/include/asm/ptrace.h index 91ceb1b454c9..4202aab6df42 100644 --- a/arch/csky/include/asm/ptrace.h +++ b/arch/csky/include/asm/ptrace.h @@ -1,5 +1,4 @@ /* SPDX-License-Identifier: GPL-2.0 */ -// Copyright (C) 2018 Hangzhou C-SKY Microsystems co.,ltd. #ifndef __ASM_CSKY_PTRACE_H #define __ASM_CSKY_PTRACE_H diff --git a/arch/csky/include/asm/segment.h b/arch/csky/include/asm/segment.h index 79ede9b1a646..589e8321dc14 100644 --- a/arch/csky/include/asm/segment.h +++ b/arch/csky/include/asm/segment.h @@ -1,5 +1,4 @@ /* SPDX-License-Identifier: GPL-2.0 */ -// Copyright (C) 2018 Hangzhou C-SKY Microsystems co.,ltd. #ifndef __ASM_CSKY_SEGMENT_H #define __ASM_CSKY_SEGMENT_H @@ -10,7 +9,7 @@ typedef struct { #define KERNEL_DS ((mm_segment_t) { 0xFFFFFFFF }) -#define USER_DS ((mm_segment_t) { 0x80000000UL }) +#define USER_DS ((mm_segment_t) { PAGE_OFFSET }) #define get_fs() (current_thread_info()->addr_limit) #define set_fs(x) (current_thread_info()->addr_limit = (x)) #define uaccess_kernel() (get_fs().seg == KERNEL_DS.seg) diff --git a/arch/csky/include/asm/shmparam.h b/arch/csky/include/asm/shmparam.h index efafe4c79fed..2fe6cea0dae9 100644 --- a/arch/csky/include/asm/shmparam.h +++ b/arch/csky/include/asm/shmparam.h @@ -1,5 +1,4 @@ /* SPDX-License-Identifier: GPL-2.0 */ -// Copyright (C) 2018 Hangzhou C-SKY Microsystems co.,ltd. #ifndef __ASM_CSKY_SHMPARAM_H #define __ASM_CSKY_SHMPARAM_H diff --git a/arch/csky/include/asm/spinlock.h b/arch/csky/include/asm/spinlock.h index 7cf3f2b34cea..69f5aa249c5f 100644 --- a/arch/csky/include/asm/spinlock.h +++ b/arch/csky/include/asm/spinlock.h @@ -6,8 +6,6 @@ #include <linux/spinlock_types.h> #include <asm/barrier.h> -#ifdef CONFIG_QUEUED_RWLOCKS - /* * Ticket-based spin-locking. */ @@ -88,169 +86,4 @@ static inline int arch_spin_is_contended(arch_spinlock_t *lock) #include <asm/qrwlock.h> -/* See include/linux/spinlock.h */ -#define smp_mb__after_spinlock() smp_mb() - -#else /* CONFIG_QUEUED_RWLOCKS */ - -/* - * Test-and-set spin-locking. - */ -static inline void arch_spin_lock(arch_spinlock_t *lock) -{ - u32 *p = &lock->lock; - u32 tmp; - - asm volatile ( - "1: ldex.w %0, (%1) \n" - " bnez %0, 1b \n" - " movi %0, 1 \n" - " stex.w %0, (%1) \n" - " bez %0, 1b \n" - : "=&r" (tmp) - : "r"(p) - : "cc"); - smp_mb(); -} - -static inline void arch_spin_unlock(arch_spinlock_t *lock) -{ - smp_mb(); - WRITE_ONCE(lock->lock, 0); -} - -static inline int arch_spin_trylock(arch_spinlock_t *lock) -{ - u32 *p = &lock->lock; - u32 tmp; - - asm volatile ( - "1: ldex.w %0, (%1) \n" - " bnez %0, 2f \n" - " movi %0, 1 \n" - " stex.w %0, (%1) \n" - " bez %0, 1b \n" - " movi %0, 0 \n" - "2: \n" - : "=&r" (tmp) - : "r"(p) - : "cc"); - - if (!tmp) - smp_mb(); - - return !tmp; -} - -#define arch_spin_is_locked(x) (READ_ONCE((x)->lock) != 0) - -/* - * read lock/unlock/trylock - */ -static inline void arch_read_lock(arch_rwlock_t *lock) -{ - u32 *p = &lock->lock; - u32 tmp; - - asm volatile ( - "1: ldex.w %0, (%1) \n" - " blz %0, 1b \n" - " addi %0, 1 \n" - " stex.w %0, (%1) \n" - " bez %0, 1b \n" - : "=&r" (tmp) - : "r"(p) - : "cc"); - smp_mb(); -} - -static inline void arch_read_unlock(arch_rwlock_t *lock) -{ - u32 *p = &lock->lock; - u32 tmp; - - smp_mb(); - asm volatile ( - "1: ldex.w %0, (%1) \n" - " subi %0, 1 \n" - " stex.w %0, (%1) \n" - " bez %0, 1b \n" - : "=&r" (tmp) - : "r"(p) - : "cc"); -} - -static inline int arch_read_trylock(arch_rwlock_t *lock) -{ - u32 *p = &lock->lock; - u32 tmp; - - asm volatile ( - "1: ldex.w %0, (%1) \n" - " blz %0, 2f \n" - " addi %0, 1 \n" - " stex.w %0, (%1) \n" - " bez %0, 1b \n" - " movi %0, 0 \n" - "2: \n" - : "=&r" (tmp) - : "r"(p) - : "cc"); - - if (!tmp) - smp_mb(); - - return !tmp; -} - -/* - * write lock/unlock/trylock - */ -static inline void arch_write_lock(arch_rwlock_t *lock) -{ - u32 *p = &lock->lock; - u32 tmp; - - asm volatile ( - "1: ldex.w %0, (%1) \n" - " bnez %0, 1b \n" - " subi %0, 1 \n" - " stex.w %0, (%1) \n" - " bez %0, 1b \n" - : "=&r" (tmp) - : "r"(p) - : "cc"); - smp_mb(); -} - -static inline void arch_write_unlock(arch_rwlock_t *lock) -{ - smp_mb(); - WRITE_ONCE(lock->lock, 0); -} - -static inline int arch_write_trylock(arch_rwlock_t *lock) -{ - u32 *p = &lock->lock; - u32 tmp; - - asm volatile ( - "1: ldex.w %0, (%1) \n" - " bnez %0, 2f \n" - " subi %0, 1 \n" - " stex.w %0, (%1) \n" - " bez %0, 1b \n" - " movi %0, 0 \n" - "2: \n" - : "=&r" (tmp) - : "r"(p) - : "cc"); - - if (!tmp) - smp_mb(); - - return !tmp; -} - -#endif /* CONFIG_QUEUED_RWLOCKS */ #endif /* __ASM_CSKY_SPINLOCK_H */ diff --git a/arch/csky/include/asm/spinlock_types.h b/arch/csky/include/asm/spinlock_types.h index 88b82438b182..8ff0f6ff3a00 100644 --- a/arch/csky/include/asm/spinlock_types.h +++ b/arch/csky/include/asm/spinlock_types.h @@ -22,16 +22,6 @@ typedef struct { #define __ARCH_SPIN_LOCK_UNLOCKED { { 0 } } -#ifdef CONFIG_QUEUED_RWLOCKS #include <asm-generic/qrwlock_types.h> -#else /* CONFIG_NR_CPUS > 2 */ - -typedef struct { - u32 lock; -} arch_rwlock_t; - -#define __ARCH_RW_LOCK_UNLOCKED { 0 } - -#endif /* CONFIG_QUEUED_RWLOCKS */ #endif /* __ASM_CSKY_SPINLOCK_TYPES_H */ diff --git a/arch/csky/include/asm/string.h b/arch/csky/include/asm/string.h index 73142de18355..a0d81e9d6b8f 100644 --- a/arch/csky/include/asm/string.h +++ b/arch/csky/include/asm/string.h @@ -1,5 +1,4 @@ /* SPDX-License-Identifier: GPL-2.0 */ -// Copyright (C) 2018 Hangzhou C-SKY Microsystems co.,ltd. #ifndef _CSKY_STRING_MM_H_ #define _CSKY_STRING_MM_H_ diff --git a/arch/csky/include/asm/switch_to.h b/arch/csky/include/asm/switch_to.h index 35a39e88933d..731e466415e2 100644 --- a/arch/csky/include/asm/switch_to.h +++ b/arch/csky/include/asm/switch_to.h @@ -1,5 +1,4 @@ /* SPDX-License-Identifier: GPL-2.0 */ -// Copyright (C) 2018 Hangzhou C-SKY Microsystems co.,ltd. #ifndef __ASM_CSKY_SWITCH_TO_H #define __ASM_CSKY_SWITCH_TO_H diff --git a/arch/csky/include/asm/syscalls.h b/arch/csky/include/asm/syscalls.h index 5d48e5e0082e..ea9ce6138b9b 100644 --- a/arch/csky/include/asm/syscalls.h +++ b/arch/csky/include/asm/syscalls.h @@ -1,5 +1,4 @@ /* SPDX-License-Identifier: GPL-2.0 */ -// Copyright (C) 2018 Hangzhou C-SKY Microsystems co.,ltd. #ifndef __ASM_CSKY_SYSCALLS_H #define __ASM_CSKY_SYSCALLS_H diff --git a/arch/csky/include/asm/thread_info.h b/arch/csky/include/asm/thread_info.h index 21456a3737c2..8c349a8f904d 100644 --- a/arch/csky/include/asm/thread_info.h +++ b/arch/csky/include/asm/thread_info.h @@ -1,12 +1,10 @@ /* SPDX-License-Identifier: GPL-2.0 */ -// Copyright (C) 2018 Hangzhou C-SKY Microsystems co.,ltd. #ifndef _ASM_CSKY_THREAD_INFO_H #define _ASM_CSKY_THREAD_INFO_H #ifndef __ASSEMBLY__ -#include <linux/version.h> #include <asm/types.h> #include <asm/page.h> #include <asm/processor.h> diff --git a/arch/csky/include/asm/tlb.h b/arch/csky/include/asm/tlb.h index fdff9b8d70c8..3498e65f59f8 100644 --- a/arch/csky/include/asm/tlb.h +++ b/arch/csky/include/asm/tlb.h @@ -1,5 +1,4 @@ /* SPDX-License-Identifier: GPL-2.0 */ -// Copyright (C) 2018 Hangzhou C-SKY Microsystems co.,ltd. #ifndef __ASM_CSKY_TLB_H #define __ASM_CSKY_TLB_H diff --git a/arch/csky/include/asm/tlbflush.h b/arch/csky/include/asm/tlbflush.h index 6845b0667703..407160b4fde7 100644 --- a/arch/csky/include/asm/tlbflush.h +++ b/arch/csky/include/asm/tlbflush.h @@ -1,5 +1,4 @@ /* SPDX-License-Identifier: GPL-2.0 */ -// Copyright (C) 2018 Hangzhou C-SKY Microsystems co.,ltd. #ifndef __ASM_TLBFLUSH_H #define __ASM_TLBFLUSH_H diff --git a/arch/csky/include/asm/traps.h b/arch/csky/include/asm/traps.h index 1c081805b962..421a4195e2fe 100644 --- a/arch/csky/include/asm/traps.h +++ b/arch/csky/include/asm/traps.h @@ -1,5 +1,4 @@ /* SPDX-License-Identifier: GPL-2.0 */ -// Copyright (C) 2018 Hangzhou C-SKY Microsystems co.,ltd. #ifndef __ASM_CSKY_TRAPS_H #define __ASM_CSKY_TRAPS_H diff --git a/arch/csky/include/asm/uaccess.h b/arch/csky/include/asm/uaccess.h index 1633ffe5ae15..3dec272e1fa3 100644 --- a/arch/csky/include/asm/uaccess.h +++ b/arch/csky/include/asm/uaccess.h @@ -1,5 +1,4 @@ /* SPDX-License-Identifier: GPL-2.0 */ -// Copyright (C) 2018 Hangzhou C-SKY Microsystems co.,ltd. #ifndef __ASM_CSKY_UACCESS_H #define __ASM_CSKY_UACCESS_H diff --git a/arch/csky/include/asm/unistd.h b/arch/csky/include/asm/unistd.h index da7a18295615..9cf97de9a26d 100644 --- a/arch/csky/include/asm/unistd.h +++ b/arch/csky/include/asm/unistd.h @@ -1,5 +1,4 @@ /* SPDX-License-Identifier: GPL-2.0 */ -// Copyright (C) 2018 Hangzhou C-SKY Microsystems co.,ltd. #include <uapi/asm/unistd.h> diff --git a/arch/csky/include/asm/vdso.h b/arch/csky/include/asm/vdso.h index d963d691f3a1..eb5142f9c564 100644 --- a/arch/csky/include/asm/vdso.h +++ b/arch/csky/include/asm/vdso.h @@ -3,10 +3,25 @@ #ifndef __ASM_CSKY_VDSO_H #define __ASM_CSKY_VDSO_H -#include <abi/vdso.h> +#include <linux/types.h> -struct csky_vdso { - unsigned short rt_signal_retcode[4]; +#ifndef GENERIC_TIME_VSYSCALL +struct vdso_data { }; +#endif + +/* + * The VDSO symbols are mapped into Linux so we can just use regular symbol + * addressing to get their offsets in userspace. The symbols are mapped at an + * offset of 0, but since the linker must support setting weak undefined + * symbols to the absolute address 0 it also happens to support other low + * addresses even when the code model suggests those low addresses would not + * otherwise be availiable. + */ +#define VDSO_SYMBOL(base, name) \ +({ \ + extern const char __vdso_##name[]; \ + (void __user *)((unsigned long)(base) + __vdso_##name); \ +}) #endif /* __ASM_CSKY_VDSO_H */ diff --git a/arch/csky/include/asm/vdso/clocksource.h b/arch/csky/include/asm/vdso/clocksource.h new file mode 100644 index 000000000000..dfca7b4724b7 --- /dev/null +++ b/arch/csky/include/asm/vdso/clocksource.h @@ -0,0 +1,9 @@ +/* SPDX-License-Identifier: GPL-2.0 */ + +#ifndef __ASM_VDSO_CSKY_CLOCKSOURCE_H +#define __ASM_VDSO_CSKY_CLOCKSOURCE_H + +#define VDSO_ARCH_CLOCKMODES \ + VDSO_CLOCKMODE_ARCHTIMER + +#endif /* __ASM_VDSO_CSKY_CLOCKSOURCE_H */ diff --git a/arch/csky/include/asm/vdso/gettimeofday.h b/arch/csky/include/asm/vdso/gettimeofday.h new file mode 100644 index 000000000000..6c4f1446944f --- /dev/null +++ b/arch/csky/include/asm/vdso/gettimeofday.h @@ -0,0 +1,114 @@ +/* SPDX-License-Identifier: GPL-2.0 */ + +#ifndef __ASM_VDSO_CSKY_GETTIMEOFDAY_H +#define __ASM_VDSO_CSKY_GETTIMEOFDAY_H + +#ifndef __ASSEMBLY__ + +#include <asm/barrier.h> +#include <asm/unistd.h> +#include <abi/regdef.h> +#include <uapi/linux/time.h> + +#define VDSO_HAS_CLOCK_GETRES 1 + +static __always_inline +int gettimeofday_fallback(struct __kernel_old_timeval *_tv, + struct timezone *_tz) +{ + register struct __kernel_old_timeval *tv asm("a0") = _tv; + register struct timezone *tz asm("a1") = _tz; + register long ret asm("a0"); + register long nr asm(syscallid) = __NR_gettimeofday; + + asm volatile ("trap 0\n" + : "=r" (ret) + : "r"(tv), "r"(tz), "r"(nr) + : "memory"); + + return ret; +} + +static __always_inline +long clock_gettime_fallback(clockid_t _clkid, struct __kernel_timespec *_ts) +{ + register clockid_t clkid asm("a0") = _clkid; + register struct __kernel_timespec *ts asm("a1") = _ts; + register long ret asm("a0"); + register long nr asm(syscallid) = __NR_clock_gettime64; + + asm volatile ("trap 0\n" + : "=r" (ret) + : "r"(clkid), "r"(ts), "r"(nr) + : "memory"); + + return ret; +} + +static __always_inline +long clock_gettime32_fallback(clockid_t _clkid, struct old_timespec32 *_ts) +{ + register clockid_t clkid asm("a0") = _clkid; + register struct old_timespec32 *ts asm("a1") = _ts; + register long ret asm("a0"); + register long nr asm(syscallid) = __NR_clock_gettime; + + asm volatile ("trap 0\n" + : "=r" (ret) + : "r"(clkid), "r"(ts), "r"(nr) + : "memory"); + + return ret; +} + +static __always_inline +int clock_getres_fallback(clockid_t _clkid, struct __kernel_timespec *_ts) +{ + register clockid_t clkid asm("a0") = _clkid; + register struct __kernel_timespec *ts asm("a1") = _ts; + register long ret asm("a0"); + register long nr asm(syscallid) = __NR_clock_getres_time64; + + asm volatile ("trap 0\n" + : "=r" (ret) + : "r"(clkid), "r"(ts), "r"(nr) + : "memory"); + + return ret; +} + +static __always_inline +int clock_getres32_fallback(clockid_t _clkid, struct old_timespec32 *_ts) +{ + register clockid_t clkid asm("a0") = _clkid; + register struct old_timespec32 *ts asm("a1") = _ts; + register long ret asm("a0"); + register long nr asm(syscallid) = __NR_clock_getres; + + asm volatile ("trap 0\n" + : "=r" (ret) + : "r"(clkid), "r"(ts), "r"(nr) + : "memory"); + + return ret; +} + +uint64_t csky_pmu_read_cc(void); +static __always_inline u64 __arch_get_hw_counter(s32 clock_mode, + const struct vdso_data *vd) +{ +#ifdef CONFIG_CSKY_PMU_V1 + return csky_pmu_read_cc(); +#else + return 0; +#endif +} + +static __always_inline const struct vdso_data *__arch_get_vdso_data(void) +{ + return _vdso_data; +} + +#endif /* !__ASSEMBLY__ */ + +#endif /* __ASM_VDSO_CSKY_GETTIMEOFDAY_H */ diff --git a/arch/csky/include/asm/vdso/processor.h b/arch/csky/include/asm/vdso/processor.h new file mode 100644 index 000000000000..39a6b561d0cc --- /dev/null +++ b/arch/csky/include/asm/vdso/processor.h @@ -0,0 +1,12 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ + +#ifndef __ASM_VDSO_CSKY_PROCESSOR_H +#define __ASM_VDSO_CSKY_PROCESSOR_H + +#ifndef __ASSEMBLY__ + +#define cpu_relax() barrier() + +#endif /* __ASSEMBLY__ */ + +#endif /* __ASM_VDSO_CSKY_PROCESSOR_H */ diff --git a/arch/csky/include/asm/vdso/vsyscall.h b/arch/csky/include/asm/vdso/vsyscall.h new file mode 100644 index 000000000000..c276211a7c4d --- /dev/null +++ b/arch/csky/include/asm/vdso/vsyscall.h @@ -0,0 +1,22 @@ +/* SPDX-License-Identifier: GPL-2.0 */ + +#ifndef __ASM_VDSO_CSKY_VSYSCALL_H +#define __ASM_VDSO_CSKY_VSYSCALL_H + +#ifndef __ASSEMBLY__ + +#include <vdso/datapage.h> + +extern struct vdso_data *vdso_data; + +static __always_inline struct vdso_data *__csky_get_k_vdso_data(void) +{ + return vdso_data; +} +#define __arch_get_k_vdso_data __csky_get_k_vdso_data + +#include <asm-generic/vdso/vsyscall.h> + +#endif /* !__ASSEMBLY__ */ + +#endif /* __ASM_VDSO_CSKY_VSYSCALL_H */ diff --git a/arch/csky/include/uapi/asm/byteorder.h b/arch/csky/include/uapi/asm/byteorder.h index d150cd664873..1aedd513b65a 100644 --- a/arch/csky/include/uapi/asm/byteorder.h +++ b/arch/csky/include/uapi/asm/byteorder.h @@ -1,5 +1,4 @@ /* SPDX-License-Identifier: GPL-2.0 WITH Linux-syscall-note */ -// Copyright (C) 2018 Hangzhou C-SKY Microsystems co.,ltd. #ifndef __ASM_CSKY_BYTEORDER_H #define __ASM_CSKY_BYTEORDER_H diff --git a/arch/csky/include/uapi/asm/perf_regs.h b/arch/csky/include/uapi/asm/perf_regs.h index 49d4e147a559..d0a8ac6a1b77 100644 --- a/arch/csky/include/uapi/asm/perf_regs.h +++ b/arch/csky/include/uapi/asm/perf_regs.h @@ -1,5 +1,4 @@ /* SPDX-License-Identifier: GPL-2.0 WITH Linux-syscall-note */ -// Copyright (C) 2019 Hangzhou C-SKY Microsystems co.,ltd. #ifndef _ASM_CSKY_PERF_REGS_H #define _ASM_CSKY_PERF_REGS_H diff --git a/arch/csky/include/uapi/asm/ptrace.h b/arch/csky/include/uapi/asm/ptrace.h index 66b2268e324e..3be9c14334a6 100644 --- a/arch/csky/include/uapi/asm/ptrace.h +++ b/arch/csky/include/uapi/asm/ptrace.h @@ -1,5 +1,4 @@ /* SPDX-License-Identifier: GPL-2.0 WITH Linux-syscall-note */ -// Copyright (C) 2018 Hangzhou C-SKY Microsystems co.,ltd. #ifndef _CSKY_PTRACE_H #define _CSKY_PTRACE_H diff --git a/arch/csky/include/uapi/asm/sigcontext.h b/arch/csky/include/uapi/asm/sigcontext.h index 670c020f2cb8..859afb602477 100644 --- a/arch/csky/include/uapi/asm/sigcontext.h +++ b/arch/csky/include/uapi/asm/sigcontext.h @@ -1,5 +1,4 @@ /* SPDX-License-Identifier: GPL-2.0 WITH Linux-syscall-note */ -// Copyright (C) 2018 Hangzhou C-SKY Microsystems co.,ltd. #ifndef __ASM_CSKY_SIGCONTEXT_H #define __ASM_CSKY_SIGCONTEXT_H diff --git a/arch/csky/include/uapi/asm/unistd.h b/arch/csky/include/uapi/asm/unistd.h index ba4018929733..7ff6a2466af1 100644 --- a/arch/csky/include/uapi/asm/unistd.h +++ b/arch/csky/include/uapi/asm/unistd.h @@ -1,5 +1,4 @@ /* SPDX-License-Identifier: GPL-2.0 WITH Linux-syscall-note */ -// Copyright (C) 2018 Hangzhou C-SKY Microsystems co.,ltd. #define __ARCH_WANT_STAT64 #define __ARCH_WANT_NEW_STAT diff --git a/arch/csky/kernel/Makefile b/arch/csky/kernel/Makefile index 37f37c0e934a..6c0f36010ed0 100644 --- a/arch/csky/kernel/Makefile +++ b/arch/csky/kernel/Makefile @@ -1,7 +1,7 @@ # SPDX-License-Identifier: GPL-2.0-only extra-y := head.o vmlinux.lds -obj-y += entry.o atomic.o signal.o traps.o irq.o time.o vdso.o +obj-y += entry.o atomic.o signal.o traps.o irq.o time.o vdso.o vdso/ obj-y += power.o syscall.o syscall_table.o setup.o obj-y += process.o cpu-probe.o ptrace.o stacktrace.o obj-y += probes/ diff --git a/arch/csky/kernel/atomic.S b/arch/csky/kernel/atomic.S index 3821ef9b7567..e73e548f7855 100644 --- a/arch/csky/kernel/atomic.S +++ b/arch/csky/kernel/atomic.S @@ -14,6 +14,10 @@ */ ENTRY(csky_cmpxchg) USPTOKSP + + RD_MEH a3 + WR_MEH a3 + mfcr a3, epc addi a3, TRAP0_SIZE @@ -36,11 +40,11 @@ ENTRY(csky_cmpxchg) 2: sync.is #else -1: +GLOBAL(csky_cmpxchg_ldw) ldw a3, (a2) cmpne a0, a3 bt16 3f -2: +GLOBAL(csky_cmpxchg_stw) stw a1, (a2) 3: #endif @@ -55,19 +59,3 @@ ENTRY(csky_cmpxchg) KSPTOUSP rte END(csky_cmpxchg) - -#ifndef CONFIG_CPU_HAS_LDSTEX -/* - * Called from tlbmodified exception - */ -ENTRY(csky_cmpxchg_fixup) - mfcr a0, epc - lrw a1, 2b - cmpne a1, a0 - bt 1f - subi a1, (2b - 1b) - stw a1, (sp, LSAVE_PC) -1: - rts -END(csky_cmpxchg_fixup) -#endif diff --git a/arch/csky/kernel/entry.S b/arch/csky/kernel/entry.S index 5a5cabd076e1..c1bd7a6b4ab6 100644 --- a/arch/csky/kernel/entry.S +++ b/arch/csky/kernel/entry.S @@ -13,10 +13,6 @@ #include <asm/page.h> #include <asm/thread_info.h> -#define PTE_INDX_MSK 0xffc -#define PTE_INDX_SHIFT 10 -#define _PGDIR_SHIFT 22 - .macro zero_fp #ifdef CONFIG_STACKTRACE movi r8, 0 @@ -41,108 +37,15 @@ #endif .endm -.macro tlbop_begin name, val0, val1, val2 -ENTRY(csky_\name) - mtcr a3, ss2 - mtcr r6, ss3 - mtcr a2, ss4 - - RD_PGDR r6 - RD_MEH a3 -#ifdef CONFIG_CPU_HAS_TLBI - tlbi.vaas a3 - sync.is - - btsti a3, 31 - bf 1f - RD_PGDR_K r6 -1: -#else - bgeni a2, 31 - WR_MCIR a2 - bgeni a2, 25 - WR_MCIR a2 -#endif - bclri r6, 0 - lrw a2, va_pa_offset - ld.w a2, (a2, 0) - subu r6, a2 - bseti r6, 31 - - mov a2, a3 - lsri a2, _PGDIR_SHIFT - lsli a2, 2 - addu r6, a2 - ldw r6, (r6) - - lrw a2, va_pa_offset - ld.w a2, (a2, 0) - subu r6, a2 - bseti r6, 31 - - lsri a3, PTE_INDX_SHIFT - lrw a2, PTE_INDX_MSK - and a3, a2 - addu r6, a3 - ldw a3, (r6) - - movi a2, (_PAGE_PRESENT | \val0) - and a3, a2 - cmpne a3, a2 - bt \name - - /* First read/write the page, just update the flags */ - ldw a3, (r6) - bgeni a2, PAGE_VALID_BIT - bseti a2, PAGE_ACCESSED_BIT - bseti a2, \val1 - bseti a2, \val2 - or a3, a2 - stw a3, (r6) - - /* Some cpu tlb-hardrefill bypass the cache */ -#ifdef CONFIG_CPU_NEED_TLBSYNC - movi a2, 0x22 - bseti a2, 6 - mtcr r6, cr22 - mtcr a2, cr17 - sync -#endif - - mfcr a3, ss2 - mfcr r6, ss3 - mfcr a2, ss4 - rte -\name: - mfcr a3, ss2 - mfcr r6, ss3 - mfcr a2, ss4 +.text +ENTRY(csky_pagefault) SAVE_ALL 0 -.endm -.macro tlbop_end is_write zero_fp context_tracking - RD_MEH a2 - psrset ee, ie + psrset ee mov a0, sp - movi a1, \is_write jbsr do_page_fault jmpi ret_from_exception -.endm - -.text - -tlbop_begin tlbinvalidl, _PAGE_READ, PAGE_VALID_BIT, PAGE_ACCESSED_BIT -tlbop_end 0 - -tlbop_begin tlbinvalids, _PAGE_WRITE, PAGE_DIRTY_BIT, PAGE_MODIFIED_BIT -tlbop_end 1 - -tlbop_begin tlbmodified, _PAGE_WRITE, PAGE_DIRTY_BIT, PAGE_MODIFIED_BIT -#ifndef CONFIG_CPU_HAS_LDSTEX -jbsr csky_cmpxchg_fixup -#endif -tlbop_end 1 ENTRY(csky_systemcall) SAVE_ALL TRAP0_SIZE @@ -314,6 +217,9 @@ ENTRY(csky_trap) ENTRY(csky_get_tls) USPTOKSP + RD_MEH a0 + WR_MEH a0 + /* increase epc for continue */ mfcr a0, epc addi a0, TRAP0_SIZE diff --git a/arch/csky/kernel/head.S b/arch/csky/kernel/head.S index 17ed9d250480..7e3e4f15b052 100644 --- a/arch/csky/kernel/head.S +++ b/arch/csky/kernel/head.S @@ -21,10 +21,16 @@ END(_start) ENTRY(_start_smp_secondary) SETUP_MMU - /* copy msa1 from CPU0 */ - lrw r6, secondary_msa1 +#ifdef CONFIG_PAGE_OFFSET_80000000 + lrw r6, secondary_msa1 ld.w r6, (r6, 0) mtcr r6, cr<31, 15> +#endif + + lrw r6, secondary_pgd + ld.w r6, (r6, 0) + mtcr r6, cr<28, 15> + mtcr r6, cr<29, 15> /* set stack point */ lrw r6, secondary_stack diff --git a/arch/csky/kernel/perf_event.c b/arch/csky/kernel/perf_event.c index 1a29f1157449..e5f18420ce64 100644 --- a/arch/csky/kernel/perf_event.c +++ b/arch/csky/kernel/perf_event.c @@ -87,7 +87,7 @@ static int csky_pmu_irq; }) /* cycle counter */ -static uint64_t csky_pmu_read_cc(void) +uint64_t csky_pmu_read_cc(void) { uint32_t lo, hi, tmp; uint64_t result; @@ -1319,7 +1319,7 @@ int csky_pmu_device_probe(struct platform_device *pdev, pr_notice("[perf] PMU request irq fail!\n"); } - ret = cpuhp_setup_state(CPUHP_AP_PERF_ONLINE, "AP_PERF_ONLINE", + ret = cpuhp_setup_state(CPUHP_AP_PERF_CSKY_ONLINE, "AP_PERF_ONLINE", csky_pmu_starting_cpu, csky_pmu_dying_cpu); if (ret) { diff --git a/arch/csky/kernel/probes/simulate-insn.c b/arch/csky/kernel/probes/simulate-insn.c index 4e464fed52ec..d6e8d092c9b7 100644 --- a/arch/csky/kernel/probes/simulate-insn.c +++ b/arch/csky/kernel/probes/simulate-insn.c @@ -274,9 +274,9 @@ void __kprobes simulate_bnezad32(u32 opcode, long addr, struct pt_regs *regs) { unsigned long tmp = opcode & 0x1f; - unsigned long val; + long val; - csky_insn_reg_get_val(regs, tmp, &val); + csky_insn_reg_get_val(regs, tmp, (unsigned long *)&val); val -= 1; @@ -286,7 +286,7 @@ simulate_bnezad32(u32 opcode, long addr, struct pt_regs *regs) } else instruction_pointer_set(regs, addr + 4); - csky_insn_reg_set_val(regs, tmp, val); + csky_insn_reg_set_val(regs, tmp, (unsigned long)val); } void __kprobes @@ -297,13 +297,11 @@ simulate_bhsz32(u32 opcode, long addr, struct pt_regs *regs) csky_insn_reg_get_val(regs, tmp, &val); - if (val >= 0) { + if ((long) val >= 0) { instruction_pointer_set(regs, addr + sign_extend32((opcode & 0xffff0000) >> 15, 15)); } else instruction_pointer_set(regs, addr + 4); - - csky_insn_reg_set_val(regs, tmp, val); } void __kprobes @@ -314,13 +312,11 @@ simulate_bhz32(u32 opcode, long addr, struct pt_regs *regs) csky_insn_reg_get_val(regs, tmp, &val); - if (val > 0) { + if ((long) val > 0) { instruction_pointer_set(regs, addr + sign_extend32((opcode & 0xffff0000) >> 15, 15)); } else instruction_pointer_set(regs, addr + 4); - - csky_insn_reg_set_val(regs, tmp, val); } void __kprobes @@ -331,13 +327,11 @@ simulate_blsz32(u32 opcode, long addr, struct pt_regs *regs) csky_insn_reg_get_val(regs, tmp, &val); - if (val <= 0) { + if ((long) val <= 0) { instruction_pointer_set(regs, addr + sign_extend32((opcode & 0xffff0000) >> 15, 15)); } else instruction_pointer_set(regs, addr + 4); - - csky_insn_reg_set_val(regs, tmp, val); } void __kprobes @@ -348,13 +342,11 @@ simulate_blz32(u32 opcode, long addr, struct pt_regs *regs) csky_insn_reg_get_val(regs, tmp, &val); - if (val < 0) { + if ((long) val < 0) { instruction_pointer_set(regs, addr + sign_extend32((opcode & 0xffff0000) >> 15, 15)); } else instruction_pointer_set(regs, addr + 4); - - csky_insn_reg_set_val(regs, tmp, val); } void __kprobes diff --git a/arch/csky/kernel/process.c b/arch/csky/kernel/process.c index 69af6bc87e64..3d0ca22cd0e2 100644 --- a/arch/csky/kernel/process.c +++ b/arch/csky/kernel/process.c @@ -49,7 +49,7 @@ int copy_thread(unsigned long clone_flags, /* setup thread.sp for switch_to !!! */ p->thread.sp = (unsigned long)childstack; - if (unlikely(p->flags & PF_KTHREAD)) { + if (unlikely(p->flags & (PF_KTHREAD | PF_IO_WORKER))) { memset(childregs, 0, sizeof(struct pt_regs)); childstack->r15 = (unsigned long) ret_from_kernel_thread; childstack->r10 = kthread_arg; diff --git a/arch/csky/kernel/ptrace.c b/arch/csky/kernel/ptrace.c index d822144906ac..0105ac81b432 100644 --- a/arch/csky/kernel/ptrace.c +++ b/arch/csky/kernel/ptrace.c @@ -22,6 +22,7 @@ #include <asm/asm-offsets.h> #include <abi/regdef.h> +#include <abi/ckmmu.h> #define CREATE_TRACE_POINTS #include <trace/events/syscalls.h> @@ -83,7 +84,7 @@ static int gpr_get(struct task_struct *target, /* Abiv1 regs->tls is fake and we need sync here. */ regs->tls = task_thread_info(target)->tp_value; - return membuf_write(&to, regs, sizeof(regs)); + return membuf_write(&to, regs, sizeof(*regs)); } static int gpr_set(struct task_struct *target, @@ -343,6 +344,124 @@ asmlinkage void syscall_trace_exit(struct pt_regs *regs) trace_sys_exit(regs, syscall_get_return_value(current, regs)); } +#ifdef CONFIG_CPU_CK860 +static void show_iutlb(void) +{ + int entry, i; + unsigned long flags; + unsigned long oldpid; + unsigned long entryhi[16], entrylo0[16], entrylo1[16]; + + oldpid = read_mmu_entryhi(); + + entry = 0x8000; + + local_irq_save(flags); + + for (i = 0; i < 16; i++) { + write_mmu_index(entry); + tlb_read(); + entryhi[i] = read_mmu_entryhi(); + entrylo0[i] = read_mmu_entrylo0(); + entrylo1[i] = read_mmu_entrylo1(); + + entry++; + } + + local_irq_restore(flags); + + write_mmu_entryhi(oldpid); + + printk("\n\n\n"); + for (i = 0; i < 16; i++) + printk("iutlb[%d]: entryhi - 0x%lx; entrylo0 - 0x%lx;" + " entrylo1 - 0x%lx\n", + i, entryhi[i], entrylo0[i], entrylo1[i]); + printk("\n\n\n"); +} + +static void show_dutlb(void) +{ + int entry, i; + unsigned long flags; + unsigned long oldpid; + unsigned long entryhi[16], entrylo0[16], entrylo1[16]; + + oldpid = read_mmu_entryhi(); + + entry = 0x4000; + + local_irq_save(flags); + + for (i = 0; i < 16; i++) { + write_mmu_index(entry); + tlb_read(); + entryhi[i] = read_mmu_entryhi(); + entrylo0[i] = read_mmu_entrylo0(); + entrylo1[i] = read_mmu_entrylo1(); + + entry++; + } + + local_irq_restore(flags); + + write_mmu_entryhi(oldpid); + + printk("\n\n\n"); + for (i = 0; i < 16; i++) + printk("dutlb[%d]: entryhi - 0x%lx; entrylo0 - 0x%lx;" + " entrylo1 - 0x%lx\n", + i, entryhi[i], entrylo0[i], entrylo1[i]); + printk("\n\n\n"); +} + +static unsigned long entryhi[1024], entrylo0[1024], entrylo1[1024]; +static void show_jtlb(void) +{ + int entry; + unsigned long flags; + unsigned long oldpid; + + oldpid = read_mmu_entryhi(); + + entry = 0; + + local_irq_save(flags); + while (entry < 1024) { + write_mmu_index(entry); + tlb_read(); + entryhi[entry] = read_mmu_entryhi(); + entrylo0[entry] = read_mmu_entrylo0(); + entrylo1[entry] = read_mmu_entrylo1(); + + entry++; + } + local_irq_restore(flags); + + write_mmu_entryhi(oldpid); + + printk("\n\n\n"); + + for (entry = 0; entry < 1024; entry++) + printk("jtlb[%x]: entryhi - 0x%lx; entrylo0 - 0x%lx;" + " entrylo1 - 0x%lx\n", + entry, entryhi[entry], entrylo0[entry], entrylo1[entry]); + printk("\n\n\n"); +} + +static void show_tlb(void) +{ + show_iutlb(); + show_dutlb(); + show_jtlb(); +} +#else +static void show_tlb(void) +{ + return; +} +#endif + void show_regs(struct pt_regs *fp) { pr_info("\nCURRENT PROCESS:\n\n"); @@ -363,9 +482,10 @@ void show_regs(struct pt_regs *fp) pr_info("PC: 0x%08lx (%pS)\n", (long)fp->pc, (void *)fp->pc); pr_info("LR: 0x%08lx (%pS)\n", (long)fp->lr, (void *)fp->lr); - pr_info("SP: 0x%08lx\n", (long)fp); - pr_info("orig_a0: 0x%08lx\n", fp->orig_a0); + pr_info("SP: 0x%08lx\n", (long)fp->usp); pr_info("PSR: 0x%08lx\n", (long)fp->sr); + pr_info("orig_a0: 0x%08lx\n", fp->orig_a0); + pr_info("PT_REGS: 0x%08lx\n", (long)fp); pr_info(" a0: 0x%08lx a1: 0x%08lx a2: 0x%08lx a3: 0x%08lx\n", fp->a0, fp->a1, fp->a2, fp->a3); @@ -395,5 +515,7 @@ void show_regs(struct pt_regs *fp) fp->regs[8], fp->regs[9]); #endif + show_tlb(); + return; } diff --git a/arch/csky/kernel/setup.c b/arch/csky/kernel/setup.c index e4cab16056d6..e93bc6f74432 100644 --- a/arch/csky/kernel/setup.c +++ b/arch/csky/kernel/setup.c @@ -45,13 +45,17 @@ static void __init csky_memblock_init(void) if (size >= lowmem_size) { max_low_pfn = min_low_pfn + lowmem_size; +#ifdef CONFIG_PAGE_OFFSET_80000000 write_mmu_msa1(read_mmu_msa0() + SSEG_SIZE); +#endif } else if (size > sseg_size) { max_low_pfn = min_low_pfn + sseg_size; } max_zone_pfn[ZONE_NORMAL] = max_low_pfn; + mmu_init(min_low_pfn, max_low_pfn); + #ifdef CONFIG_HIGHMEM max_zone_pfn[ZONE_HIGHMEM] = max_pfn; @@ -101,16 +105,26 @@ void __init setup_arch(char **cmdline_p) unsigned long va_pa_offset; EXPORT_SYMBOL(va_pa_offset); +static inline unsigned long read_mmu_msa(void) +{ +#ifdef CONFIG_PAGE_OFFSET_80000000 + return read_mmu_msa0(); +#endif + +#ifdef CONFIG_PAGE_OFFSET_A0000000 + return read_mmu_msa1(); +#endif +} + asmlinkage __visible void __init csky_start(unsigned int unused, void *dtb_start) { /* Clean up bss section */ memset(__bss_start, 0, __bss_stop - __bss_start); - va_pa_offset = read_mmu_msa0() & ~(SSEG_SIZE - 1); + va_pa_offset = read_mmu_msa() & ~(SSEG_SIZE - 1); pre_trap_init(); - pre_mmu_init(); if (dtb_start == NULL) early_init_dt_scan(__dtb_start); diff --git a/arch/csky/kernel/signal.c b/arch/csky/kernel/signal.c index 37ea64ed3c12..312f046d452d 100644 --- a/arch/csky/kernel/signal.c +++ b/arch/csky/kernel/signal.c @@ -134,7 +134,6 @@ setup_rt_frame(struct ksignal *ksig, sigset_t *set, struct pt_regs *regs) { struct rt_sigframe *frame; int err = 0; - struct csky_vdso *vdso = current->mm->context.vdso; frame = get_sigframe(ksig, regs, sizeof(*frame)); if (!access_ok(frame, sizeof(*frame))) @@ -152,7 +151,8 @@ setup_rt_frame(struct ksignal *ksig, sigset_t *set, struct pt_regs *regs) return -EFAULT; /* Set up to return from userspace. */ - regs->lr = (unsigned long)(vdso->rt_signal_retcode); + regs->lr = (unsigned long)VDSO_SYMBOL( + current->mm->context.vdso, rt_sigreturn); /* * Set up registers for signal handler. diff --git a/arch/csky/kernel/smp.c b/arch/csky/kernel/smp.c index 041d0de6a1b6..0f9f5eef9338 100644 --- a/arch/csky/kernel/smp.c +++ b/arch/csky/kernel/smp.c @@ -203,8 +203,8 @@ volatile unsigned int secondary_hint; volatile unsigned int secondary_hint2; volatile unsigned int secondary_ccr; volatile unsigned int secondary_stack; - -unsigned long secondary_msa1; +volatile unsigned int secondary_msa1; +volatile unsigned int secondary_pgd; int __cpu_up(unsigned int cpu, struct task_struct *tidle) { @@ -216,6 +216,7 @@ int __cpu_up(unsigned int cpu, struct task_struct *tidle) secondary_hint2 = mfcr("cr<21, 1>"); secondary_ccr = mfcr("cr18"); secondary_msa1 = read_mmu_msa1(); + secondary_pgd = mfcr("cr<29, 15>"); /* * Because other CPUs are in reset status, we must flush data @@ -262,8 +263,6 @@ void csky_start_secondary(void) flush_tlb_all(); write_mmu_pagemask(0); - TLBMISS_HANDLER_SETUP_PGD(swapper_pg_dir); - TLBMISS_HANDLER_SETUP_PGD_KERNEL(swapper_pg_dir); #ifdef CONFIG_CPU_HAS_FPU init_fpu(); diff --git a/arch/csky/kernel/traps.c b/arch/csky/kernel/traps.c index 959a917c989d..e5fbf8653a21 100644 --- a/arch/csky/kernel/traps.c +++ b/arch/csky/kernel/traps.c @@ -39,9 +39,7 @@ asmlinkage void csky_cmpxchg(void); asmlinkage void csky_get_tls(void); asmlinkage void csky_irq(void); -asmlinkage void csky_tlbinvalidl(void); -asmlinkage void csky_tlbinvalids(void); -asmlinkage void csky_tlbmodified(void); +asmlinkage void csky_pagefault(void); /* Defined in head.S */ asmlinkage void _start_smp_secondary(void); @@ -66,9 +64,9 @@ void __init trap_init(void) VEC_INIT(VEC_TRAP3, csky_get_tls); /* setup MMU TLB exception */ - VEC_INIT(VEC_TLBINVALIDL, csky_tlbinvalidl); - VEC_INIT(VEC_TLBINVALIDS, csky_tlbinvalids); - VEC_INIT(VEC_TLBMODIFIED, csky_tlbmodified); + VEC_INIT(VEC_TLBINVALIDL, csky_pagefault); + VEC_INIT(VEC_TLBINVALIDS, csky_pagefault); + VEC_INIT(VEC_TLBMODIFIED, csky_pagefault); #ifdef CONFIG_CPU_HAS_FPU init_fpu(); diff --git a/arch/csky/kernel/vdso.c b/arch/csky/kernel/vdso.c index abc3dbc658d4..16c20d64d165 100644 --- a/arch/csky/kernel/vdso.c +++ b/arch/csky/kernel/vdso.c @@ -1,86 +1,107 @@ // SPDX-License-Identifier: GPL-2.0 // Copyright (C) 2018 Hangzhou C-SKY Microsystems co.,ltd. -#include <linux/kernel.h> -#include <linux/err.h> -#include <linux/sched.h> -#include <linux/mm.h> -#include <linux/init.h> #include <linux/binfmts.h> #include <linux/elf.h> -#include <linux/vmalloc.h> -#include <linux/unistd.h> -#include <linux/uaccess.h> +#include <linux/err.h> +#include <linux/mm.h> +#include <linux/slab.h> +#include <asm/page.h> +#ifdef GENERIC_TIME_VSYSCALL +#include <vdso/datapage.h> +#else #include <asm/vdso.h> -#include <asm/cacheflush.h> +#endif -static struct page *vdso_page; +extern char vdso_start[], vdso_end[]; -static int __init init_vdso(void) -{ - struct csky_vdso *vdso; - int err = 0; - - vdso_page = alloc_page(GFP_KERNEL); - if (!vdso_page) - panic("Cannot allocate vdso"); +static unsigned int vdso_pages; +static struct page **vdso_pagelist; - vdso = vmap(&vdso_page, 1, 0, PAGE_KERNEL); - if (!vdso) - panic("Cannot map vdso"); +/* + * The vDSO data page. + */ +static union { + struct vdso_data data; + u8 page[PAGE_SIZE]; +} vdso_data_store __page_aligned_data; +struct vdso_data *vdso_data = &vdso_data_store.data; - clear_page(vdso); - - err = setup_vdso_page(vdso->rt_signal_retcode); - if (err) - panic("Cannot set signal return code, err: %x.", err); +static int __init vdso_init(void) +{ + unsigned int i; + + vdso_pages = (vdso_end - vdso_start) >> PAGE_SHIFT; + vdso_pagelist = + kcalloc(vdso_pages + 1, sizeof(struct page *), GFP_KERNEL); + if (unlikely(vdso_pagelist == NULL)) { + pr_err("vdso: pagelist allocation failed\n"); + return -ENOMEM; + } - dcache_wb_range((unsigned long)vdso, (unsigned long)vdso + 16); + for (i = 0; i < vdso_pages; i++) { + struct page *pg; - vunmap(vdso); + pg = virt_to_page(vdso_start + (i << PAGE_SHIFT)); + vdso_pagelist[i] = pg; + } + vdso_pagelist[i] = virt_to_page(vdso_data); return 0; } -subsys_initcall(init_vdso); +arch_initcall(vdso_init); -int arch_setup_additional_pages(struct linux_binprm *bprm, int uses_interp) +int arch_setup_additional_pages(struct linux_binprm *bprm, + int uses_interp) { - int ret; - unsigned long addr; struct mm_struct *mm = current->mm; + unsigned long vdso_base, vdso_len; + int ret; - mmap_write_lock(mm); + vdso_len = (vdso_pages + 1) << PAGE_SHIFT; - addr = get_unmapped_area(NULL, STACK_TOP, PAGE_SIZE, 0, 0); - if (IS_ERR_VALUE(addr)) { - ret = addr; - goto up_fail; + mmap_write_lock(mm); + vdso_base = get_unmapped_area(NULL, 0, vdso_len, 0, 0); + if (IS_ERR_VALUE(vdso_base)) { + ret = vdso_base; + goto end; } - ret = install_special_mapping( - mm, - addr, - PAGE_SIZE, - VM_READ|VM_EXEC|VM_MAYREAD|VM_MAYWRITE|VM_MAYEXEC, - &vdso_page); - if (ret) - goto up_fail; + /* + * Put vDSO base into mm struct. We need to do this before calling + * install_special_mapping or the perf counter mmap tracking code + * will fail to recognise it as a vDSO (since arch_vma_name fails). + */ + mm->context.vdso = (void *)vdso_base; + + ret = + install_special_mapping(mm, vdso_base, vdso_pages << PAGE_SHIFT, + (VM_READ | VM_EXEC | VM_MAYREAD | VM_MAYWRITE | VM_MAYEXEC), + vdso_pagelist); + + if (unlikely(ret)) { + mm->context.vdso = NULL; + goto end; + } - mm->context.vdso = (void *)addr; + vdso_base += (vdso_pages << PAGE_SHIFT); + ret = install_special_mapping(mm, vdso_base, PAGE_SIZE, + (VM_READ | VM_MAYREAD), &vdso_pagelist[vdso_pages]); -up_fail: + if (unlikely(ret)) + mm->context.vdso = NULL; +end: mmap_write_unlock(mm); return ret; } const char *arch_vma_name(struct vm_area_struct *vma) { - if (vma->vm_mm == NULL) - return NULL; - - if (vma->vm_start == (long)vma->vm_mm->context.vdso) + if (vma->vm_mm && (vma->vm_start == (long)vma->vm_mm->context.vdso)) return "[vdso]"; - else - return NULL; + if (vma->vm_mm && (vma->vm_start == + (long)vma->vm_mm->context.vdso + PAGE_SIZE)) + return "[vdso_data]"; + return NULL; } diff --git a/arch/csky/kernel/vdso/.gitignore b/arch/csky/kernel/vdso/.gitignore new file mode 100644 index 000000000000..3a19def868ec --- /dev/null +++ b/arch/csky/kernel/vdso/.gitignore @@ -0,0 +1,4 @@ +# SPDX-License-Identifier: GPL-2.0-only +vdso.lds +*.tmp +vdso-syms.S diff --git a/arch/csky/kernel/vdso/Makefile b/arch/csky/kernel/vdso/Makefile new file mode 100644 index 000000000000..0b6909f10667 --- /dev/null +++ b/arch/csky/kernel/vdso/Makefile @@ -0,0 +1,72 @@ +# SPDX-License-Identifier: GPL-2.0-only + +# Absolute relocation type $(ARCH_REL_TYPE_ABS) needs to be defined before +# the inclusion of generic Makefile. +ARCH_REL_TYPE_ABS := R_CKCORE_ADDR32|R_CKCORE_JUMP_SLOT +include $(srctree)/lib/vdso/Makefile + +# Symbols present in the vdso +vdso-syms += rt_sigreturn +vdso-syms += vgettimeofday + +# Files to link into the vdso +obj-vdso = $(patsubst %, %.o, $(vdso-syms)) note.o + +ifneq ($(c-gettimeofday-y),) + CFLAGS_vgettimeofday.o += -include $(c-gettimeofday-y) +endif + +ccflags-y := -fno-stack-protector -DBUILD_VDSO32 + +# Build rules +targets := $(obj-vdso) vdso.so vdso.so.dbg vdso.lds vdso-dummy.o +obj-vdso := $(addprefix $(obj)/, $(obj-vdso)) + +obj-y += vdso.o vdso-syms.o +CPPFLAGS_vdso.lds += -P -C -U$(ARCH) + +# Disable gcov profiling for VDSO code +GCOV_PROFILE := n +KCOV_INSTRUMENT := n + +# Force dependency +$(obj)/vdso.o: $(obj)/vdso.so + +SYSCFLAGS_vdso.so.dbg = $(c_flags) +$(obj)/vdso.so.dbg: $(src)/vdso.lds $(obj-vdso) FORCE + $(call if_changed,vdsold) +SYSCFLAGS_vdso.so.dbg = -shared -s -Wl,-soname=linux-vdso.so.1 \ + -Wl,--build-id=sha1 -Wl,--hash-style=both + +$(obj)/vdso-syms.S: $(obj)/vdso.so FORCE + $(call if_changed,so2s) + +# strip rule for the .so file +$(obj)/%.so: OBJCOPYFLAGS := -S +$(obj)/%.so: $(obj)/%.so.dbg FORCE + $(call if_changed,objcopy) + +# actual build commands +# The DSO images are built using a special linker script +# Make sure only to export the intended __vdso_xxx symbol offsets. +quiet_cmd_vdsold = VDSOLD $@ + cmd_vdsold = $(CC) $(KBUILD_CFLAGS) $(call cc-option, -no-pie) -nostdlib -nostartfiles $(SYSCFLAGS_$(@F)) \ + -Wl,-T,$(filter-out FORCE,$^) -o $@.tmp && \ + $(CROSS_COMPILE)objcopy \ + $(patsubst %, -G __vdso_%, $(vdso-syms)) $@.tmp $@ && \ + rm $@.tmp + +# Extracts symbol offsets from the VDSO, converting them into an assembly file +# that contains the same symbols at the same offsets. +quiet_cmd_so2s = SO2S $@ + cmd_so2s = $(NM) -D $< | $(srctree)/$(src)/so2s.sh > $@ + +# install commands for the unstripped file +quiet_cmd_vdso_install = INSTALL $@ + cmd_vdso_install = cp $(obj)/$@.dbg $(MODLIB)/vdso/$@ + +vdso.so: $(obj)/vdso.so.dbg + @mkdir -p $(MODLIB)/vdso + $(call cmd,vdso_install) + +vdso_install: vdso.so diff --git a/arch/csky/kernel/vdso/note.S b/arch/csky/kernel/vdso/note.S new file mode 100644 index 000000000000..2a956c942211 --- /dev/null +++ b/arch/csky/kernel/vdso/note.S @@ -0,0 +1,12 @@ +/* SPDX-License-Identifier: GPL-2.0-or-later */ +/* + * 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/elfnote.h> +#include <linux/version.h> + +ELFNOTE_START(Linux, 0, "a") + .long LINUX_VERSION_CODE +ELFNOTE_END diff --git a/arch/csky/kernel/vdso/rt_sigreturn.S b/arch/csky/kernel/vdso/rt_sigreturn.S new file mode 100644 index 000000000000..0a6bd1216118 --- /dev/null +++ b/arch/csky/kernel/vdso/rt_sigreturn.S @@ -0,0 +1,14 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ + +#include <linux/linkage.h> +#include <asm/unistd.h> +#include <abi/vdso.h> + + .text +ENTRY(__vdso_rt_sigreturn) + .cfi_startproc + .cfi_signal_frame + SET_SYSCALL_ID + trap 0 + .cfi_endproc +ENDPROC(__vdso_rt_sigreturn) diff --git a/arch/csky/kernel/vdso/so2s.sh b/arch/csky/kernel/vdso/so2s.sh new file mode 100755 index 000000000000..69da3d529c6d --- /dev/null +++ b/arch/csky/kernel/vdso/so2s.sh @@ -0,0 +1,5 @@ +#!/bin/sh +# SPDX-License-Identifier: GPL-2.0+ + +sed 's!\([0-9a-f]*\) T \([a-z0-9_]*\)\(@@LINUX_5.10\)*!.global \2\n.set \2,0x\1!' \ +| grep '^\.' diff --git a/arch/csky/kernel/vdso/vdso.S b/arch/csky/kernel/vdso/vdso.S new file mode 100644 index 000000000000..5162ca069494 --- /dev/null +++ b/arch/csky/kernel/vdso/vdso.S @@ -0,0 +1,16 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ + +#include <linux/init.h> +#include <linux/linkage.h> +#include <asm/page.h> + + __PAGE_ALIGNED_DATA + + .globl vdso_start, vdso_end + .balign PAGE_SIZE +vdso_start: + .incbin "arch/csky/kernel/vdso/vdso.so" + .balign PAGE_SIZE +vdso_end: + + .previous diff --git a/arch/csky/kernel/vdso/vdso.lds.S b/arch/csky/kernel/vdso/vdso.lds.S new file mode 100644 index 000000000000..590a6c79fff7 --- /dev/null +++ b/arch/csky/kernel/vdso/vdso.lds.S @@ -0,0 +1,58 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ + +#include <asm/page.h> + +OUTPUT_ARCH(csky) + +SECTIONS +{ + PROVIDE(_vdso_data = . + PAGE_SIZE); + . = 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 + .dynamic : { *(.dynamic) } :text :dynamic + + .eh_frame_hdr : { *(.eh_frame_hdr) } :text :eh_frame_hdr + .eh_frame : { KEEP (*(.eh_frame)) } :text + + .rodata : { *(.rodata .rodata.* .gnu.linkonce.r.*) } + + . = 0x800; + .text : { *(.text .text.*) } :text + + .data : { + *(.got.plt) *(.got) + *(.data .data.* .gnu.linkonce.d.*) + *(.dynbss) + *(.bss .bss.* .gnu.linkonce.b.*) + } +} + +PHDRS +{ + text PT_LOAD FLAGS(5) FILEHDR PHDRS; /* PF_R|PF_X */ + dynamic PT_DYNAMIC FLAGS(4); /* PF_R */ + note PT_NOTE FLAGS(4); /* PF_R */ + eh_frame_hdr PT_GNU_EH_FRAME; +} + +VERSION +{ + LINUX_5.10 { + global: + __vdso_rt_sigreturn; + __vdso_clock_gettime; + __vdso_clock_gettime64; + __vdso_gettimeofday; + __vdso_clock_getres; + local: *; + }; +} diff --git a/arch/csky/kernel/vdso/vgettimeofday.c b/arch/csky/kernel/vdso/vgettimeofday.c new file mode 100644 index 000000000000..da491832c098 --- /dev/null +++ b/arch/csky/kernel/vdso/vgettimeofday.c @@ -0,0 +1,28 @@ +// SPDX-License-Identifier: GPL-2.0-only + +#include <linux/time.h> +#include <linux/types.h> + +int __vdso_clock_gettime(clockid_t clock, + struct old_timespec32 *ts) +{ + return __cvdso_clock_gettime32(clock, ts); +} + +int __vdso_clock_gettime64(clockid_t clock, + struct __kernel_timespec *ts) +{ + return __cvdso_clock_gettime(clock, ts); +} + +int __vdso_gettimeofday(struct __kernel_old_timeval *tv, + struct timezone *tz) +{ + return __cvdso_gettimeofday(tv, tz); +} + +int __vdso_clock_getres(clockid_t clock_id, + struct old_timespec32 *res) +{ + return __cvdso_clock_getres_time32(clock_id, res); +} diff --git a/arch/csky/kernel/vmlinux.lds.S b/arch/csky/kernel/vmlinux.lds.S index f03033e17c29..e8b1a4a49798 100644 --- a/arch/csky/kernel/vmlinux.lds.S +++ b/arch/csky/kernel/vmlinux.lds.S @@ -33,6 +33,7 @@ SECTIONS .text : AT(ADDR(.text) - LOAD_OFFSET) { _text = .; + VBR_BASE IRQENTRY_TEXT SOFTIRQENTRY_TEXT TEXT_TEXT @@ -104,7 +105,6 @@ SECTIONS EXCEPTION_TABLE(L1_CACHE_BYTES) BSS_SECTION(L1_CACHE_BYTES, PAGE_SIZE, L1_CACHE_BYTES) - VBR_BASE _end = . ; STABS_DEBUG diff --git a/arch/csky/mm/fault.c b/arch/csky/mm/fault.c index 081b178b41b1..1482de56f4f7 100644 --- a/arch/csky/mm/fault.c +++ b/arch/csky/mm/fault.c @@ -1,29 +1,10 @@ // SPDX-License-Identifier: GPL-2.0 // Copyright (C) 2018 Hangzhou C-SKY Microsystems co.,ltd. -#include <linux/signal.h> -#include <linux/module.h> -#include <linux/sched.h> -#include <linux/interrupt.h> -#include <linux/kernel.h> -#include <linux/errno.h> -#include <linux/string.h> -#include <linux/types.h> -#include <linux/ptrace.h> -#include <linux/mman.h> -#include <linux/mm.h> -#include <linux/smp.h> -#include <linux/version.h> -#include <linux/vt_kern.h> #include <linux/extable.h> -#include <linux/uaccess.h> -#include <linux/perf_event.h> #include <linux/kprobes.h> - -#include <asm/hardirq.h> -#include <asm/mmu_context.h> -#include <asm/traps.h> -#include <asm/page.h> +#include <linux/mmu_context.h> +#include <linux/perf_event.h> int fixup_exception(struct pt_regs *regs) { @@ -39,180 +20,287 @@ int fixup_exception(struct pt_regs *regs) return 0; } -/* - * This routine handles page faults. It determines the address, - * and the problem, and then passes it off to one of the appropriate - * routines. - */ -asmlinkage void do_page_fault(struct pt_regs *regs, unsigned long write, - unsigned long mmu_meh) +static inline bool is_write(struct pt_regs *regs) { - struct vm_area_struct *vma = NULL; - struct task_struct *tsk = current; - struct mm_struct *mm = tsk->mm; - int si_code; - int fault; - unsigned long address = mmu_meh & PAGE_MASK; + switch (trap_no(regs)) { + case VEC_TLBINVALIDS: + return true; + case VEC_TLBMODIFIED: + return true; + } - if (kprobe_page_fault(regs, tsk->thread.trap_no)) + return false; +} + +#ifdef CONFIG_CPU_HAS_LDSTEX +static inline void csky_cmpxchg_fixup(struct pt_regs *regs) +{ + return; +} +#else +extern unsigned long csky_cmpxchg_ldw; +extern unsigned long csky_cmpxchg_stw; +static inline void csky_cmpxchg_fixup(struct pt_regs *regs) +{ + if (trap_no(regs) != VEC_TLBMODIFIED) return; - si_code = SEGV_MAPERR; + if (instruction_pointer(regs) == csky_cmpxchg_stw) + instruction_pointer_set(regs, csky_cmpxchg_ldw); + return; +} +#endif + +static inline void no_context(struct pt_regs *regs, unsigned long addr) +{ + current->thread.trap_no = trap_no(regs); + + /* Are we prepared to handle this kernel fault? */ + if (fixup_exception(regs)) + return; -#ifndef CONFIG_CPU_HAS_TLBI /* - * We fault-in kernel-space virtual memory on-demand. The - * 'reference' page table is init_mm.pgd. - * - * NOTE! We MUST NOT take any locks for this case. We may - * be in an interrupt or a critical region, and should - * only copy the information from the master page table, - * nothing more. + * Oops. The kernel tried to access some bad page. We'll have to + * terminate things with extreme prejudice. */ - if (unlikely(address >= VMALLOC_START) && - unlikely(address <= VMALLOC_END)) { - /* - * Synchronize this task's top level page-table - * with the 'reference' page table. - * - * Do _not_ use "tsk" here. We might be inside - * an interrupt in the middle of a task switch.. - */ - int offset = pgd_index(address); - pgd_t *pgd, *pgd_k; - pud_t *pud, *pud_k; - pmd_t *pmd, *pmd_k; - pte_t *pte_k; + bust_spinlocks(1); + pr_alert("Unable to handle kernel paging request at virtual " + "addr 0x%08lx, pc: 0x%08lx\n", addr, regs->pc); + die(regs, "Oops"); + do_exit(SIGKILL); +} - unsigned long pgd_base; +static inline void mm_fault_error(struct pt_regs *regs, unsigned long addr, vm_fault_t fault) +{ + current->thread.trap_no = trap_no(regs); - pgd_base = (unsigned long)__va(get_pgd()); - pgd = (pgd_t *)pgd_base + offset; - pgd_k = init_mm.pgd + offset; + if (fault & VM_FAULT_OOM) { + /* + * We ran out of memory, call the OOM killer, and return the userspace + * (which will retry the fault, or kill us if we got oom-killed). + */ + if (!user_mode(regs)) { + no_context(regs, addr); + return; + } + pagefault_out_of_memory(); + return; + } else if (fault & VM_FAULT_SIGBUS) { + /* Kernel mode? Handle exceptions or die */ + if (!user_mode(regs)) { + no_context(regs, addr); + return; + } + do_trap(regs, SIGBUS, BUS_ADRERR, addr); + return; + } + BUG(); +} - if (!pgd_present(*pgd_k)) - goto no_context; - set_pgd(pgd, *pgd_k); +static inline void bad_area(struct pt_regs *regs, struct mm_struct *mm, int code, unsigned long addr) +{ + /* + * Something tried to access memory that isn't in our memory map. + * Fix it, but check if it's kernel or user first. + */ + mmap_read_unlock(mm); + /* User mode accesses just cause a SIGSEGV */ + if (user_mode(regs)) { + do_trap(regs, SIGSEGV, code, addr); + return; + } - pud = (pud_t *)pgd; - pud_k = (pud_t *)pgd_k; - if (!pud_present(*pud_k)) - goto no_context; + no_context(regs, addr); +} - pmd = pmd_offset(pud, address); - pmd_k = pmd_offset(pud_k, address); - if (!pmd_present(*pmd_k)) - goto no_context; - set_pmd(pmd, *pmd_k); +static inline void vmalloc_fault(struct pt_regs *regs, int code, unsigned long addr) +{ + pgd_t *pgd, *pgd_k; + pud_t *pud, *pud_k; + pmd_t *pmd, *pmd_k; + pte_t *pte_k; + int offset; - pte_k = pte_offset_kernel(pmd_k, address); - if (!pte_present(*pte_k)) - goto no_context; + /* User mode accesses just cause a SIGSEGV */ + if (user_mode(regs)) { + do_trap(regs, SIGSEGV, code, addr); return; } -#endif - perf_sw_event(PERF_COUNT_SW_PAGE_FAULTS, 1, regs, address); /* - * If we're in an interrupt or have no user - * context, we must not take the fault.. + * Synchronize this task's top level page-table + * with the 'reference' page table. + * + * Do _not_ use "tsk" here. We might be inside + * an interrupt in the middle of a task switch.. */ - if (in_atomic() || !mm) - goto bad_area_nosemaphore; + offset = pgd_index(addr); - mmap_read_lock(mm); - vma = find_vma(mm, address); - if (!vma) - goto bad_area; - if (vma->vm_start <= address) - goto good_area; - if (!(vma->vm_flags & VM_GROWSDOWN)) - goto bad_area; - if (expand_stack(vma, address)) - goto bad_area; - /* - * Ok, we have a good vm_area for this memory access, so - * we can handle it.. - */ -good_area: - si_code = SEGV_ACCERR; + pgd = get_pgd() + offset; + pgd_k = init_mm.pgd + offset; - if (write) { + if (!pgd_present(*pgd_k)) { + no_context(regs, addr); + return; + } + set_pgd(pgd, *pgd_k); + + pud = (pud_t *)pgd; + pud_k = (pud_t *)pgd_k; + if (!pud_present(*pud_k)) { + no_context(regs, addr); + return; + } + + pmd = pmd_offset(pud, addr); + pmd_k = pmd_offset(pud_k, addr); + if (!pmd_present(*pmd_k)) { + no_context(regs, addr); + return; + } + set_pmd(pmd, *pmd_k); + + pte_k = pte_offset_kernel(pmd_k, addr); + if (!pte_present(*pte_k)) { + no_context(regs, addr); + return; + } + + flush_tlb_one(addr); +} + +static inline bool access_error(struct pt_regs *regs, struct vm_area_struct *vma) +{ + if (is_write(regs)) { if (!(vma->vm_flags & VM_WRITE)) - goto bad_area; + return true; } else { if (unlikely(!vma_is_accessible(vma))) - goto bad_area; + return true; } + return false; +} + +/* + * This routine handles page faults. It determines the address and the + * problem, and then passes it off to one of the appropriate routines. + */ +asmlinkage void do_page_fault(struct pt_regs *regs) +{ + struct task_struct *tsk; + struct vm_area_struct *vma; + struct mm_struct *mm; + unsigned long addr = read_mmu_entryhi() & PAGE_MASK; + unsigned int flags = FAULT_FLAG_DEFAULT; + int code = SEGV_MAPERR; + vm_fault_t fault; + + tsk = current; + mm = tsk->mm; + + csky_cmpxchg_fixup(regs); + + if (kprobe_page_fault(regs, tsk->thread.trap_no)) + return; /* - * If for any reason at all we couldn't handle the fault, - * make sure we exit gracefully rather than endlessly redo - * the fault. + * Fault-in kernel-space virtual memory on-demand. + * The 'reference' page table is init_mm.pgd. + * + * NOTE! We MUST NOT take any locks for this case. We may + * be in an interrupt or a critical region, and should + * only copy the information from the master page table, + * nothing more. */ - fault = handle_mm_fault(vma, address, write ? FAULT_FLAG_WRITE : 0, - regs); - if (unlikely(fault & VM_FAULT_ERROR)) { - if (fault & VM_FAULT_OOM) - goto out_of_memory; - else if (fault & VM_FAULT_SIGBUS) - goto do_sigbus; - else if (fault & VM_FAULT_SIGSEGV) - goto bad_area; - BUG(); + if (unlikely((addr >= VMALLOC_START) && (addr <= VMALLOC_END))) { + vmalloc_fault(regs, code, addr); + return; } - mmap_read_unlock(mm); - return; + + /* Enable interrupts if they were enabled in the parent context. */ + if (likely(regs->sr & BIT(6))) + local_irq_enable(); /* - * Something tried to access memory that isn't in our memory map.. - * Fix it, but check if it's kernel or user first.. + * If we're in an interrupt, have no user context, or are running + * in an atomic region, then we must not take the fault. */ -bad_area: - mmap_read_unlock(mm); - -bad_area_nosemaphore: - /* User mode accesses just cause a SIGSEGV */ - if (user_mode(regs)) { - tsk->thread.trap_no = trap_no(regs); - force_sig_fault(SIGSEGV, si_code, (void __user *)address); + if (unlikely(faulthandler_disabled() || !mm)) { + no_context(regs, addr); return; } -no_context: - tsk->thread.trap_no = trap_no(regs); + if (user_mode(regs)) + flags |= FAULT_FLAG_USER; - /* Are we prepared to handle this kernel fault? */ - if (fixup_exception(regs)) + perf_sw_event(PERF_COUNT_SW_PAGE_FAULTS, 1, regs, addr); + + if (is_write(regs)) + flags |= FAULT_FLAG_WRITE; +retry: + mmap_read_lock(mm); + vma = find_vma(mm, addr); + if (unlikely(!vma)) { + bad_area(regs, mm, code, addr); + return; + } + if (likely(vma->vm_start <= addr)) + goto good_area; + if (unlikely(!(vma->vm_flags & VM_GROWSDOWN))) { + bad_area(regs, mm, code, addr); return; + } + if (unlikely(expand_stack(vma, addr))) { + bad_area(regs, mm, code, addr); + return; + } /* - * Oops. The kernel tried to access some bad page. We'll have to - * terminate things with extreme prejudice. + * Ok, we have a good vm_area for this memory access, so + * we can handle it. */ - bust_spinlocks(1); - pr_alert("Unable to handle kernel paging request at virtual " - "address 0x%08lx, pc: 0x%08lx\n", address, regs->pc); - die(regs, "Oops"); +good_area: + code = SEGV_ACCERR; + + if (unlikely(access_error(regs, vma))) { + bad_area(regs, mm, code, addr); + return; + } -out_of_memory: - tsk->thread.trap_no = trap_no(regs); + /* + * If for any reason at all we could not handle the fault, + * make sure we exit gracefully rather than endlessly redo + * the fault. + */ + fault = handle_mm_fault(vma, addr, flags, regs); /* - * We ran out of memory, call the OOM killer, and return the userspace - * (which will retry the fault, or kill us if we got oom-killed). + * If we need to retry but a fatal signal is pending, handle the + * signal first. We do not need to release the mmap_lock because it + * would already be released in __lock_page_or_retry in mm/filemap.c. */ - pagefault_out_of_memory(); - return; + if (fault_signal_pending(fault, regs)) { + if (!user_mode(regs)) + no_context(regs, addr); + return; + } -do_sigbus: - tsk->thread.trap_no = trap_no(regs); + if (unlikely((fault & VM_FAULT_RETRY) && (flags & FAULT_FLAG_ALLOW_RETRY))) { + flags |= FAULT_FLAG_TRIED; - mmap_read_unlock(mm); + /* + * No need to mmap_read_unlock(mm) as we would + * have already released it in __lock_page_or_retry + * in mm/filemap.c. + */ + goto retry; + } - /* Kernel mode? Handle exceptions or die */ - if (!user_mode(regs)) - goto no_context; + mmap_read_unlock(mm); - force_sig_fault(SIGBUS, BUS_ADRERR, (void __user *)address); + if (unlikely(fault & VM_FAULT_ERROR)) { + mm_fault_error(regs, addr, fault); + return; + } + return; } diff --git a/arch/csky/mm/init.c b/arch/csky/mm/init.c index af627128314f..894050a8ce09 100644 --- a/arch/csky/mm/init.c +++ b/arch/csky/mm/init.c @@ -28,9 +28,15 @@ #include <asm/mmu_context.h> #include <asm/sections.h> #include <asm/tlb.h> +#include <asm/cacheflush.h> + +#define PTRS_KERN_TABLE \ + ((PTRS_PER_PGD - USER_PTRS_PER_PGD) * PTRS_PER_PTE) pgd_t swapper_pg_dir[PTRS_PER_PGD] __page_aligned_bss; pte_t invalid_pte_table[PTRS_PER_PTE] __page_aligned_bss; +pte_t kernel_pte_tables[PTRS_KERN_TABLE] __page_aligned_bss; + EXPORT_SYMBOL(invalid_pte_table); unsigned long empty_zero_page[PAGE_SIZE / sizeof(unsigned long)] __page_aligned_bss; @@ -80,9 +86,9 @@ void __init mem_init(void) #ifdef CONFIG_HIGHMEM unsigned long tmp; - max_mapnr = highend_pfn; + set_max_mapnr(highend_pfn - ARCH_PFN_OFFSET); #else - max_mapnr = max_low_pfn; + set_max_mapnr(max_low_pfn - ARCH_PFN_OFFSET); #endif high_memory = (void *) __va(max_low_pfn << PAGE_SHIFT); @@ -104,24 +110,9 @@ void __init mem_init(void) mem_init_print_info(NULL); } -extern char __init_begin[], __init_end[]; - void free_initmem(void) { - unsigned long addr; - - addr = (unsigned long) &__init_begin; - - while (addr < (unsigned long) &__init_end) { - ClearPageReserved(virt_to_page(addr)); - init_page_count(virt_to_page(addr)); - free_page(addr); - totalram_pages_inc(); - addr += PAGE_SIZE; - } - - pr_info("Freeing unused kernel memory: %dk freed\n", - ((unsigned int)&__init_end - (unsigned int)&__init_begin) >> 10); + free_initmem_default(-1); } void pgd_init(unsigned long *p) @@ -130,20 +121,35 @@ void pgd_init(unsigned long *p) for (i = 0; i < PTRS_PER_PGD; i++) p[i] = __pa(invalid_pte_table); + + flush_tlb_all(); + local_icache_inv_all(NULL); } -void __init pre_mmu_init(void) +void __init mmu_init(unsigned long min_pfn, unsigned long max_pfn) { - /* - * Setup page-table and enable TLB-hardrefill - */ + int i; + + for (i = 0; i < USER_PTRS_PER_PGD; i++) + swapper_pg_dir[i].pgd = __pa(invalid_pte_table); + + for (i = USER_PTRS_PER_PGD; i < PTRS_PER_PGD; i++) + swapper_pg_dir[i].pgd = + __pa(kernel_pte_tables + (PTRS_PER_PTE * (i - USER_PTRS_PER_PGD))); + + for (i = 0; i < PTRS_KERN_TABLE; i++) + set_pte(&kernel_pte_tables[i], __pte(_PAGE_GLOBAL)); + + for (i = min_pfn; i < max_pfn; i++) + set_pte(&kernel_pte_tables[i - PFN_DOWN(va_pa_offset)], pfn_pte(i, PAGE_KERNEL)); + flush_tlb_all(); - pgd_init((unsigned long *)swapper_pg_dir); - TLBMISS_HANDLER_SETUP_PGD(swapper_pg_dir); - TLBMISS_HANDLER_SETUP_PGD_KERNEL(swapper_pg_dir); + local_icache_inv_all(NULL); /* Setup page mask to 4k */ write_mmu_pagemask(0); + + setup_pgd(swapper_pg_dir, 0); } void __init fixrange_init(unsigned long start, unsigned long end, diff --git a/arch/csky/mm/tlb.c b/arch/csky/mm/tlb.c index ed1512381112..9234c5e5ceaf 100644 --- a/arch/csky/mm/tlb.c +++ b/arch/csky/mm/tlb.c @@ -24,7 +24,13 @@ void flush_tlb_all(void) void flush_tlb_mm(struct mm_struct *mm) { #ifdef CONFIG_CPU_HAS_TLBI - asm volatile("tlbi.asids %0"::"r"(cpu_asid(mm))); + sync_is(); + asm volatile( + "tlbi.asids %0 \n" + "sync.i \n" + : + : "r" (cpu_asid(mm)) + : "memory"); #else tlb_invalid_all(); #endif @@ -53,11 +59,17 @@ void flush_tlb_range(struct vm_area_struct *vma, unsigned long start, end &= TLB_ENTRY_SIZE_MASK; #ifdef CONFIG_CPU_HAS_TLBI + sync_is(); while (start < end) { - asm volatile("tlbi.vas %0"::"r"(start | newpid)); + asm volatile( + "tlbi.vas %0 \n" + : + : "r" (start | newpid) + : "memory"); + start += 2*PAGE_SIZE; } - sync_is(); + asm volatile("sync.i\n"); #else { unsigned long flags, oldpid; @@ -87,11 +99,17 @@ void flush_tlb_kernel_range(unsigned long start, unsigned long end) end &= TLB_ENTRY_SIZE_MASK; #ifdef CONFIG_CPU_HAS_TLBI + sync_is(); while (start < end) { - asm volatile("tlbi.vaas %0"::"r"(start)); + asm volatile( + "tlbi.vaas %0 \n" + : + : "r" (start) + : "memory"); + start += 2*PAGE_SIZE; } - sync_is(); + asm volatile("sync.i\n"); #else { unsigned long flags, oldpid; @@ -121,8 +139,13 @@ void flush_tlb_page(struct vm_area_struct *vma, unsigned long addr) addr &= TLB_ENTRY_SIZE_MASK; #ifdef CONFIG_CPU_HAS_TLBI - asm volatile("tlbi.vas %0"::"r"(addr | newpid)); sync_is(); + asm volatile( + "tlbi.vas %0 \n" + "sync.i \n" + : + : "r" (addr | newpid) + : "memory"); #else { int oldpid, idx; @@ -147,8 +170,13 @@ void flush_tlb_one(unsigned long addr) addr &= TLB_ENTRY_SIZE_MASK; #ifdef CONFIG_CPU_HAS_TLBI - asm volatile("tlbi.vaas %0"::"r"(addr)); sync_is(); + asm volatile( + "tlbi.vaas %0 \n" + "sync.i \n" + : + : "r" (addr) + : "memory"); #else { int oldpid, idx; diff --git a/arch/h8300/kernel/process.c b/arch/h8300/kernel/process.c index bc1364db58fe..46b1342ce515 100644 --- a/arch/h8300/kernel/process.c +++ b/arch/h8300/kernel/process.c @@ -112,7 +112,7 @@ int copy_thread(unsigned long clone_flags, unsigned long usp, childregs = (struct pt_regs *) (THREAD_SIZE + task_stack_page(p)) - 1; - if (unlikely(p->flags & PF_KTHREAD)) { + if (unlikely(p->flags & (PF_KTHREAD | PF_IO_WORKER))) { memset(childregs, 0, sizeof(struct pt_regs)); childregs->retpc = (unsigned long) ret_from_kernel_thread; childregs->er4 = topstk; /* arg */ diff --git a/arch/hexagon/kernel/process.c b/arch/hexagon/kernel/process.c index 6a980cba7b29..c61165c99ae0 100644 --- a/arch/hexagon/kernel/process.c +++ b/arch/hexagon/kernel/process.c @@ -73,7 +73,7 @@ int copy_thread(unsigned long clone_flags, unsigned long usp, unsigned long arg, sizeof(*ss)); ss->lr = (unsigned long)ret_from_fork; p->thread.switch_sp = ss; - if (unlikely(p->flags & PF_KTHREAD)) { + if (unlikely(p->flags & (PF_KTHREAD | PF_IO_WORKER))) { memset(childregs, 0, sizeof(struct pt_regs)); /* r24 <- fn, r25 <- arg */ ss->r24 = usp; diff --git a/arch/ia64/kernel/process.c b/arch/ia64/kernel/process.c index 4ebbfa076a26..7e1a1525e202 100644 --- a/arch/ia64/kernel/process.c +++ b/arch/ia64/kernel/process.c @@ -338,7 +338,7 @@ copy_thread(unsigned long clone_flags, unsigned long user_stack_base, ia64_drop_fpu(p); /* don't pick up stale state from a CPU's fph */ - if (unlikely(p->flags & PF_KTHREAD)) { + if (unlikely(p->flags & (PF_KTHREAD | PF_IO_WORKER))) { if (unlikely(!user_stack_base)) { /* fork_idle() called us */ return 0; diff --git a/arch/m68k/kernel/process.c b/arch/m68k/kernel/process.c index 08359a6e058f..da83cc83e791 100644 --- a/arch/m68k/kernel/process.c +++ b/arch/m68k/kernel/process.c @@ -157,7 +157,7 @@ int copy_thread(unsigned long clone_flags, unsigned long usp, unsigned long arg, */ p->thread.fs = get_fs().seg; - if (unlikely(p->flags & PF_KTHREAD)) { + if (unlikely(p->flags & (PF_KTHREAD | PF_IO_WORKER))) { /* kernel thread */ memset(frame, 0, sizeof(struct fork_frame)); frame->regs.sr = PS_S; diff --git a/arch/microblaze/kernel/process.c b/arch/microblaze/kernel/process.c index 657c2beb665e..62aa237180b6 100644 --- a/arch/microblaze/kernel/process.c +++ b/arch/microblaze/kernel/process.c @@ -59,7 +59,7 @@ int copy_thread(unsigned long clone_flags, unsigned long usp, unsigned long arg, struct pt_regs *childregs = task_pt_regs(p); struct thread_info *ti = task_thread_info(p); - if (unlikely(p->flags & PF_KTHREAD)) { + if (unlikely(p->flags & (PF_KTHREAD | PF_IO_WORKER))) { /* if we're creating a new kernel thread then just zeroing all * the registers. That's OK for a brand new thread.*/ memset(childregs, 0, sizeof(struct pt_regs)); diff --git a/arch/mips/kernel/process.c b/arch/mips/kernel/process.c index af4c862ec5ff..7efa0d1a4c2b 100644 --- a/arch/mips/kernel/process.c +++ b/arch/mips/kernel/process.c @@ -120,7 +120,7 @@ int copy_thread(unsigned long clone_flags, unsigned long usp, /* Put the stack after the struct pt_regs. */ childksp = (unsigned long) childregs; p->thread.cp0_status = (read_c0_status() & ~(ST0_CU2|ST0_CU1)) | ST0_KERNEL_CUMASK; - if (unlikely(p->flags & PF_KTHREAD)) { + if (unlikely(p->flags & (PF_KTHREAD | PF_IO_WORKER))) { /* kernel thread */ unsigned long status = p->thread.cp0_status; memset(childregs, 0, sizeof(struct pt_regs)); diff --git a/arch/nds32/kernel/process.c b/arch/nds32/kernel/process.c index e01ad5d17224..c1327e552ec6 100644 --- a/arch/nds32/kernel/process.c +++ b/arch/nds32/kernel/process.c @@ -156,7 +156,7 @@ int copy_thread(unsigned long clone_flags, unsigned long stack_start, memset(&p->thread.cpu_context, 0, sizeof(struct cpu_context)); - if (unlikely(p->flags & PF_KTHREAD)) { + if (unlikely(p->flags & (PF_KTHREAD | PF_IO_WORKER))) { memset(childregs, 0, sizeof(struct pt_regs)); /* kernel thread fn */ p->thread.cpu_context.r6 = stack_start; diff --git a/arch/nios2/kernel/process.c b/arch/nios2/kernel/process.c index 50b4eb19a6cc..c5f916ca6845 100644 --- a/arch/nios2/kernel/process.c +++ b/arch/nios2/kernel/process.c @@ -109,7 +109,7 @@ int copy_thread(unsigned long clone_flags, unsigned long usp, unsigned long arg, struct switch_stack *childstack = ((struct switch_stack *)childregs) - 1; - if (unlikely(p->flags & PF_KTHREAD)) { + if (unlikely(p->flags & (PF_KTHREAD | PF_IO_WORKER))) { memset(childstack, 0, sizeof(struct switch_stack) + sizeof(struct pt_regs)); diff --git a/arch/openrisc/kernel/process.c b/arch/openrisc/kernel/process.c index 181448f74316..eb62429681fc 100644 --- a/arch/openrisc/kernel/process.c +++ b/arch/openrisc/kernel/process.c @@ -174,7 +174,7 @@ copy_thread(unsigned long clone_flags, unsigned long usp, unsigned long arg, sp -= sizeof(struct pt_regs); kregs = (struct pt_regs *)sp; - if (unlikely(p->flags & PF_KTHREAD)) { + if (unlikely(p->flags & (PF_KTHREAD | PF_IO_WORKER))) { memset(kregs, 0, sizeof(struct pt_regs)); kregs->gpr[20] = usp; /* fn, kernel thread */ kregs->gpr[22] = arg; diff --git a/arch/parisc/kernel/process.c b/arch/parisc/kernel/process.c index fda1c1a6a444..b144fbe29bc1 100644 --- a/arch/parisc/kernel/process.c +++ b/arch/parisc/kernel/process.c @@ -200,7 +200,7 @@ 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(p->flags & (PF_KTHREAD | PF_IO_WORKER))) { /* kernel thread */ memset(cregs, 0, sizeof(struct pt_regs)); if (!usp) /* idle thread */ diff --git a/arch/powerpc/kernel/process.c b/arch/powerpc/kernel/process.c index 924d023dad0a..3231c2df9e26 100644 --- a/arch/powerpc/kernel/process.c +++ b/arch/powerpc/kernel/process.c @@ -1670,7 +1670,7 @@ int copy_thread(unsigned long clone_flags, unsigned long usp, /* Copy registers */ sp -= sizeof(struct pt_regs); childregs = (struct pt_regs *) sp; - if (unlikely(p->flags & PF_KTHREAD)) { + if (unlikely(p->flags & (PF_KTHREAD | PF_IO_WORKER))) { /* kernel thread */ memset(childregs, 0, sizeof(struct pt_regs)); childregs->gpr[1] = sp + sizeof(struct pt_regs); diff --git a/arch/riscv/Kconfig b/arch/riscv/Kconfig index a998babc1237..85d626b8ce5e 100644 --- a/arch/riscv/Kconfig +++ b/arch/riscv/Kconfig @@ -314,6 +314,7 @@ endchoice # Common NUMA Features config NUMA bool "NUMA Memory Allocation and Scheduler Support" + depends on SMP select GENERIC_ARCH_NUMA select OF_NUMA select ARCH_SUPPORTS_NUMA_BALANCING diff --git a/arch/riscv/configs/defconfig b/arch/riscv/configs/defconfig index 8c3d1e451703..6c0625aa96c7 100644 --- a/arch/riscv/configs/defconfig +++ b/arch/riscv/configs/defconfig @@ -17,6 +17,7 @@ CONFIG_BPF_SYSCALL=y CONFIG_SOC_SIFIVE=y CONFIG_SOC_VIRT=y CONFIG_SMP=y +CONFIG_HOTPLUG_CPU=y CONFIG_JUMP_LABEL=y CONFIG_MODULES=y CONFIG_MODULE_UNLOAD=y diff --git a/arch/riscv/configs/rv32_defconfig b/arch/riscv/configs/rv32_defconfig index 2c2cda6cc1c5..8dd02b842fef 100644 --- a/arch/riscv/configs/rv32_defconfig +++ b/arch/riscv/configs/rv32_defconfig @@ -18,6 +18,7 @@ CONFIG_SOC_SIFIVE=y CONFIG_SOC_VIRT=y CONFIG_ARCH_RV32I=y CONFIG_SMP=y +CONFIG_HOTPLUG_CPU=y CONFIG_JUMP_LABEL=y CONFIG_MODULES=y CONFIG_MODULE_UNLOAD=y diff --git a/arch/riscv/kernel/process.c b/arch/riscv/kernel/process.c index 19f4688f2f36..6f728e731bed 100644 --- a/arch/riscv/kernel/process.c +++ b/arch/riscv/kernel/process.c @@ -124,7 +124,7 @@ int copy_thread(unsigned long clone_flags, unsigned long usp, unsigned long arg, struct pt_regs *childregs = task_pt_regs(p); /* p->thread holds context to be restored by __switch_to() */ - if (unlikely(p->flags & PF_KTHREAD)) { + if (unlikely(p->flags & (PF_KTHREAD | PF_IO_WORKER))) { /* Kernel thread */ memset(childregs, 0, sizeof(struct pt_regs)); childregs->gp = gp_in_global; diff --git a/arch/riscv/mm/init.c b/arch/riscv/mm/init.c index 9dfb8b2dffd6..067583ab1bd7 100644 --- a/arch/riscv/mm/init.c +++ b/arch/riscv/mm/init.c @@ -108,34 +108,17 @@ void __init mem_init(void) void __init setup_bootmem(void) { - phys_addr_t mem_start = 0; - phys_addr_t start, dram_end, end = 0; phys_addr_t vmlinux_end = __pa_symbol(&_end); phys_addr_t vmlinux_start = __pa_symbol(&_start); + phys_addr_t dram_end = memblock_end_of_DRAM(); phys_addr_t max_mapped_addr = __pa(~(ulong)0); - u64 i; - - /* Find the memory region containing the kernel */ - for_each_mem_range(i, &start, &end) { - phys_addr_t size = end - start; - if (!mem_start) - mem_start = start; - if (start <= vmlinux_start && vmlinux_end <= end) - BUG_ON(size == 0); - } - /* - * The maximal physical memory size is -PAGE_OFFSET. - * Make sure that any memory beyond mem_start + (-PAGE_OFFSET) is removed - * as it is unusable by kernel. - */ + /* The maximal physical memory size is -PAGE_OFFSET. */ memblock_enforce_memory_limit(-PAGE_OFFSET); /* Reserve from the start of the kernel to the end of the kernel */ memblock_reserve(vmlinux_start, vmlinux_end - vmlinux_start); - dram_end = memblock_end_of_DRAM(); - /* * memblock allocator is not aware of the fact that last 4K bytes of * the addressable memory can not be mapped because of IS_ERR_VALUE diff --git a/arch/s390/kernel/process.c b/arch/s390/kernel/process.c index 367bd000f6d1..e20bed1ed34a 100644 --- a/arch/s390/kernel/process.c +++ b/arch/s390/kernel/process.c @@ -130,7 +130,7 @@ int copy_thread(unsigned long clone_flags, unsigned long new_stackp, frame->sf.gprs[9] = (unsigned long)frame; /* Store access registers to kernel stack of new process. */ - if (unlikely(p->flags & PF_KTHREAD)) { + if (unlikely(p->flags & (PF_KTHREAD | PF_IO_WORKER))) { /* kernel thread */ memset(&frame->childregs, 0, sizeof(struct pt_regs)); frame->childregs.psw.mask = PSW_KERNEL_BITS | PSW_MASK_DAT | diff --git a/arch/sh/boards/mach-landisk/gio.c b/arch/sh/boards/mach-landisk/gio.c index 1c0da99dfc60..ff2200fec29a 100644 --- a/arch/sh/boards/mach-landisk/gio.c +++ b/arch/sh/boards/mach-landisk/gio.c @@ -27,11 +27,10 @@ static int openCnt; static int gio_open(struct inode *inode, struct file *filp) { - int minor; + int minor = iminor(inode); int ret = -ENOENT; preempt_disable(); - minor = MINOR(inode->i_rdev); if (minor < DEVCOUNT) { if (openCnt > 0) { ret = -EALREADY; @@ -46,9 +45,8 @@ static int gio_open(struct inode *inode, struct file *filp) static int gio_close(struct inode *inode, struct file *filp) { - int minor; + int minor = iminor(inode); - minor = MINOR(inode->i_rdev); if (minor < DEVCOUNT) { openCnt--; } diff --git a/arch/sh/kernel/process_32.c b/arch/sh/kernel/process_32.c index 80a5d1c66a51..1aa508eb0823 100644 --- a/arch/sh/kernel/process_32.c +++ b/arch/sh/kernel/process_32.c @@ -114,7 +114,7 @@ int copy_thread(unsigned long clone_flags, unsigned long usp, unsigned long arg, childregs = task_pt_regs(p); p->thread.sp = (unsigned long) childregs; - if (unlikely(p->flags & PF_KTHREAD)) { + if (unlikely(p->flags & (PF_KTHREAD | PF_IO_WORKER))) { memset(childregs, 0, sizeof(struct pt_regs)); p->thread.pc = (unsigned long) ret_from_kernel_thread; childregs->regs[4] = arg; diff --git a/arch/sparc/kernel/process_32.c b/arch/sparc/kernel/process_32.c index 7649b14d69f8..b91e88058e0c 100644 --- a/arch/sparc/kernel/process_32.c +++ b/arch/sparc/kernel/process_32.c @@ -309,7 +309,7 @@ int copy_thread(unsigned long clone_flags, unsigned long sp, unsigned long arg, ti->ksp = (unsigned long) new_stack; p->thread.kregs = childregs; - if (unlikely(p->flags & PF_KTHREAD)) { + if (unlikely(p->flags & (PF_KTHREAD | PF_IO_WORKER))) { extern int nwindows; unsigned long psr; memset(new_stack, 0, STACKFRAME_SZ + TRACEREG_SZ); diff --git a/arch/sparc/kernel/process_64.c b/arch/sparc/kernel/process_64.c index 6f8c7822fc06..7afd0a859a78 100644 --- a/arch/sparc/kernel/process_64.c +++ b/arch/sparc/kernel/process_64.c @@ -597,7 +597,7 @@ int copy_thread(unsigned long clone_flags, unsigned long sp, unsigned long arg, sizeof(struct sparc_stackf)); t->fpsaved[0] = 0; - if (unlikely(p->flags & PF_KTHREAD)) { + if (unlikely(p->flags & (PF_KTHREAD | PF_IO_WORKER))) { memset(child_trap_frame, 0, child_stack_sz); __thread_flag_byte_ptr(t)[TI_FLAG_BYTE_CWP] = (current_pt_regs()->tstate + 1) & TSTATE_CWP; diff --git a/arch/um/kernel/process.c b/arch/um/kernel/process.c index 81d508daf67c..c5011064b5dd 100644 --- a/arch/um/kernel/process.c +++ b/arch/um/kernel/process.c @@ -157,7 +157,7 @@ int copy_thread(unsigned long clone_flags, unsigned long sp, unsigned long arg, struct task_struct * p, unsigned long tls) { void (*handler)(void); - int kthread = current->flags & PF_KTHREAD; + int kthread = current->flags & (PF_KTHREAD | PF_IO_WORKER); int ret = 0; p->thread = (struct thread_struct) INIT_THREAD; diff --git a/arch/x86/kernel/process.c b/arch/x86/kernel/process.c index 145a7ac0c19a..9c214d7085a4 100644 --- a/arch/x86/kernel/process.c +++ b/arch/x86/kernel/process.c @@ -161,7 +161,7 @@ int copy_thread(unsigned long clone_flags, unsigned long sp, unsigned long arg, #endif /* Kernel thread ? */ - if (unlikely(p->flags & PF_KTHREAD)) { + if (unlikely(p->flags & (PF_KTHREAD | PF_IO_WORKER))) { memset(childregs, 0, sizeof(struct pt_regs)); kthread_frame_init(frame, sp, arg); return 0; diff --git a/arch/xtensa/kernel/process.c b/arch/xtensa/kernel/process.c index 397a7de56377..9534ef515d74 100644 --- a/arch/xtensa/kernel/process.c +++ b/arch/xtensa/kernel/process.c @@ -217,7 +217,7 @@ int copy_thread(unsigned long clone_flags, unsigned long usp_thread_fn, p->thread.sp = (unsigned long)childregs; - if (!(p->flags & PF_KTHREAD)) { + if (!(p->flags & (PF_KTHREAD | PF_IO_WORKER))) { struct pt_regs *regs = current_pt_regs(); unsigned long usp = usp_thread_fn ? usp_thread_fn : regs->areg[1]; diff --git a/block/bfq-iosched.c b/block/bfq-iosched.c index b398dde53af9..ec482e6641ff 100644 --- a/block/bfq-iosched.c +++ b/block/bfq-iosched.c @@ -125,6 +125,8 @@ #include <linux/delay.h> #include <linux/backing-dev.h> +#include <trace/events/block.h> + #include "blk.h" #include "blk-mq.h" #include "blk-mq-tag.h" @@ -5621,7 +5623,7 @@ static void bfq_insert_request(struct blk_mq_hw_ctx *hctx, struct request *rq, spin_unlock_irq(&bfqd->lock); - blk_mq_sched_request_inserted(rq); + trace_block_rq_insert(rq); spin_lock_irq(&bfqd->lock); bfqq = bfq_init_rq(rq); diff --git a/block/blk-core.c b/block/blk-core.c index 5e752840b41a..fc60ff208497 100644 --- a/block/blk-core.c +++ b/block/blk-core.c @@ -59,6 +59,7 @@ EXPORT_TRACEPOINT_SYMBOL_GPL(block_rq_remap); EXPORT_TRACEPOINT_SYMBOL_GPL(block_bio_complete); EXPORT_TRACEPOINT_SYMBOL_GPL(block_split); EXPORT_TRACEPOINT_SYMBOL_GPL(block_unplug); +EXPORT_TRACEPOINT_SYMBOL_GPL(block_rq_insert); DEFINE_IDA(blk_queue_ida); diff --git a/block/blk-crypto-fallback.c b/block/blk-crypto-fallback.c index e8327c50d7c9..c176b7af56a7 100644 --- a/block/blk-crypto-fallback.c +++ b/block/blk-crypto-fallback.c @@ -80,6 +80,7 @@ static struct blk_crypto_keyslot { static struct blk_keyslot_manager blk_crypto_ksm; static struct workqueue_struct *blk_crypto_wq; static mempool_t *blk_crypto_bounce_page_pool; +static struct bio_set crypto_bio_split; /* * This is the key we set when evicting a keyslot. This *should* be the all 0's @@ -224,7 +225,8 @@ static bool blk_crypto_split_bio_if_needed(struct bio **bio_ptr) if (num_sectors < bio_sectors(bio)) { struct bio *split_bio; - split_bio = bio_split(bio, num_sectors, GFP_NOIO, NULL); + split_bio = bio_split(bio, num_sectors, GFP_NOIO, + &crypto_bio_split); if (!split_bio) { bio->bi_status = BLK_STS_RESOURCE; return false; @@ -538,9 +540,13 @@ static int blk_crypto_fallback_init(void) prandom_bytes(blank_key, BLK_CRYPTO_MAX_KEY_SIZE); - err = blk_ksm_init(&blk_crypto_ksm, blk_crypto_num_keyslots); + err = bioset_init(&crypto_bio_split, 64, 0, 0); if (err) goto out; + + err = blk_ksm_init(&blk_crypto_ksm, blk_crypto_num_keyslots); + if (err) + goto fail_free_bioset; err = -ENOMEM; blk_crypto_ksm.ksm_ll_ops = blk_crypto_ksm_ll_ops; @@ -591,6 +597,8 @@ fail_free_wq: destroy_workqueue(blk_crypto_wq); fail_free_ksm: blk_ksm_destroy(&blk_crypto_ksm); +fail_free_bioset: + bioset_exit(&crypto_bio_split); out: return err; } diff --git a/block/blk-map.c b/block/blk-map.c index 21630dccac62..369e204d14d0 100644 --- a/block/blk-map.c +++ b/block/blk-map.c @@ -150,9 +150,7 @@ static int bio_copy_user_iov(struct request *rq, struct rq_map_data *map_data, bmd->is_our_pages = !map_data; bmd->is_null_mapped = (map_data && map_data->null_mapped); - nr_pages = DIV_ROUND_UP(offset + len, PAGE_SIZE); - if (nr_pages > BIO_MAX_PAGES) - nr_pages = BIO_MAX_PAGES; + nr_pages = bio_max_segs(DIV_ROUND_UP(offset + len, PAGE_SIZE)); ret = -ENOMEM; bio = bio_kmalloc(gfp_mask, nr_pages); diff --git a/block/blk-mq-sched.c b/block/blk-mq-sched.c index deff4e826e23..ddb65e9e6fd9 100644 --- a/block/blk-mq-sched.c +++ b/block/blk-mq-sched.c @@ -384,12 +384,6 @@ bool blk_mq_sched_try_insert_merge(struct request_queue *q, struct request *rq) } EXPORT_SYMBOL_GPL(blk_mq_sched_try_insert_merge); -void blk_mq_sched_request_inserted(struct request *rq) -{ - trace_block_rq_insert(rq); -} -EXPORT_SYMBOL_GPL(blk_mq_sched_request_inserted); - static bool blk_mq_sched_bypass_insert(struct blk_mq_hw_ctx *hctx, bool has_sched, struct request *rq) diff --git a/block/blk-mq-sched.h b/block/blk-mq-sched.h index 0476360f05f1..5b18ab915c65 100644 --- a/block/blk-mq-sched.h +++ b/block/blk-mq-sched.h @@ -7,7 +7,6 @@ void blk_mq_sched_assign_ioc(struct request *rq); -void blk_mq_sched_request_inserted(struct request *rq); bool blk_mq_sched_try_merge(struct request_queue *q, struct bio *bio, unsigned int nr_segs, struct request **merged_request); bool __blk_mq_sched_bio_merge(struct request_queue *q, struct bio *bio, diff --git a/block/blk-pm.h b/block/blk-pm.h index a2283cc9f716..8a5a0d4b357f 100644 --- a/block/blk-pm.h +++ b/block/blk-pm.h @@ -21,31 +21,6 @@ static inline void blk_pm_mark_last_busy(struct request *rq) if (rq->q->dev && !(rq->rq_flags & RQF_PM)) pm_runtime_mark_last_busy(rq->q->dev); } - -static inline void blk_pm_requeue_request(struct request *rq) -{ - lockdep_assert_held(&rq->q->queue_lock); - - if (rq->q->dev && !(rq->rq_flags & RQF_PM)) - rq->q->nr_pending--; -} - -static inline void blk_pm_add_request(struct request_queue *q, - struct request *rq) -{ - lockdep_assert_held(&q->queue_lock); - - if (q->dev && !(rq->rq_flags & RQF_PM)) - q->nr_pending++; -} - -static inline void blk_pm_put_request(struct request *rq) -{ - lockdep_assert_held(&rq->q->queue_lock); - - if (rq->q->dev && !(rq->rq_flags & RQF_PM)) - --rq->q->nr_pending; -} #else static inline int blk_pm_resume_queue(const bool pm, struct request_queue *q) { @@ -55,19 +30,6 @@ static inline int blk_pm_resume_queue(const bool pm, struct request_queue *q) static inline void blk_pm_mark_last_busy(struct request *rq) { } - -static inline void blk_pm_requeue_request(struct request *rq) -{ -} - -static inline void blk_pm_add_request(struct request_queue *q, - struct request *rq) -{ -} - -static inline void blk_pm_put_request(struct request *rq) -{ -} #endif #endif /* _BLOCK_BLK_PM_H_ */ diff --git a/block/blk-settings.c b/block/blk-settings.c index 7dd8be314ac6..b4aa2f37fab6 100644 --- a/block/blk-settings.c +++ b/block/blk-settings.c @@ -504,6 +504,14 @@ void blk_queue_io_opt(struct request_queue *q, unsigned int opt) } EXPORT_SYMBOL(blk_queue_io_opt); +static unsigned int blk_round_down_sectors(unsigned int sectors, unsigned int lbs) +{ + sectors = round_down(sectors, lbs >> SECTOR_SHIFT); + if (sectors < PAGE_SIZE >> SECTOR_SHIFT) + sectors = PAGE_SIZE >> SECTOR_SHIFT; + return sectors; +} + /** * blk_stack_limits - adjust queue_limits for stacked devices * @t: the stacking driver limits (top device) @@ -630,6 +638,10 @@ int blk_stack_limits(struct queue_limits *t, struct queue_limits *b, ret = -1; } + t->max_sectors = blk_round_down_sectors(t->max_sectors, t->logical_block_size); + t->max_hw_sectors = blk_round_down_sectors(t->max_hw_sectors, t->logical_block_size); + t->max_dev_sectors = blk_round_down_sectors(t->max_dev_sectors, t->logical_block_size); + /* Discard alignment and granularity */ if (b->discard_granularity) { alignment = queue_limit_discard_alignment(b, start); diff --git a/block/blk-sysfs.c b/block/blk-sysfs.c index ae39c7f3d83d..0f4f0c8a7825 100644 --- a/block/blk-sysfs.c +++ b/block/blk-sysfs.c @@ -434,10 +434,13 @@ static ssize_t queue_poll_store(struct request_queue *q, const char *page, if (ret < 0) return ret; - if (poll_on) + if (poll_on) { blk_queue_flag_set(QUEUE_FLAG_POLL, q); - else + } else { + blk_mq_freeze_queue(q); blk_queue_flag_clear(QUEUE_FLAG_POLL, q); + blk_mq_unfreeze_queue(q); + } return ret; } diff --git a/block/bounce.c b/block/bounce.c index fc55314aa426..87983a35079c 100644 --- a/block/bounce.c +++ b/block/bounce.c @@ -214,8 +214,7 @@ static void bounce_end_io_read_isa(struct bio *bio) __bounce_end_io_read(bio, &isa_page_pool); } -static struct bio *bounce_clone_bio(struct bio *bio_src, gfp_t gfp_mask, - struct bio_set *bs) +static struct bio *bounce_clone_bio(struct bio *bio_src) { struct bvec_iter iter; struct bio_vec bv; @@ -242,10 +241,12 @@ static struct bio *bounce_clone_bio(struct bio *bio_src, gfp_t gfp_mask, * asking for trouble and would force extra work on * __bio_clone_fast() anyways. */ - - bio = bio_alloc_bioset(gfp_mask, bio_segments(bio_src), bs); - if (!bio) - return NULL; + if (bio_is_passthrough(bio_src)) + bio = bio_kmalloc(GFP_NOIO | __GFP_NOFAIL, + bio_segments(bio_src)); + else + bio = bio_alloc_bioset(GFP_NOIO, bio_segments(bio_src), + &bounce_bio_set); bio->bi_bdev = bio_src->bi_bdev; if (bio_flagged(bio_src, BIO_REMAPPED)) bio_set_flag(bio, BIO_REMAPPED); @@ -269,11 +270,11 @@ static struct bio *bounce_clone_bio(struct bio *bio_src, gfp_t gfp_mask, break; } - if (bio_crypt_clone(bio, bio_src, gfp_mask) < 0) + if (bio_crypt_clone(bio, bio_src, GFP_NOIO) < 0) goto err_put; if (bio_integrity(bio_src) && - bio_integrity_clone(bio, bio_src, gfp_mask) < 0) + bio_integrity_clone(bio, bio_src, GFP_NOIO) < 0) goto err_put; bio_clone_blkg_association(bio, bio_src); @@ -296,7 +297,6 @@ static void __blk_queue_bounce(struct request_queue *q, struct bio **bio_orig, unsigned i = 0; bool bounce = false; int sectors = 0; - bool passthrough = bio_is_passthrough(*bio_orig); bio_for_each_segment(from, *bio_orig, iter) { if (i++ < BIO_MAX_PAGES) @@ -307,14 +307,14 @@ static void __blk_queue_bounce(struct request_queue *q, struct bio **bio_orig, if (!bounce) return; - if (!passthrough && sectors < bio_sectors(*bio_orig)) { + if (!bio_is_passthrough(*bio_orig) && + sectors < bio_sectors(*bio_orig)) { bio = bio_split(*bio_orig, sectors, GFP_NOIO, &bounce_bio_split); bio_chain(bio, *bio_orig); submit_bio_noacct(*bio_orig); *bio_orig = bio; } - bio = bounce_clone_bio(*bio_orig, GFP_NOIO, passthrough ? NULL : - &bounce_bio_set); + bio = bounce_clone_bio(*bio_orig); /* * Bvec table can't be updated by bio_for_each_segment_all(), diff --git a/block/genhd.c b/block/genhd.c index 36ff45bbaaaf..fcc530164b5a 100644 --- a/block/genhd.c +++ b/block/genhd.c @@ -74,7 +74,7 @@ bool set_capacity_and_notify(struct gendisk *disk, sector_t size) return false; pr_info("%s: detected capacity change from %lld to %lld\n", - disk->disk_name, size, capacity); + disk->disk_name, capacity, size); /* * Historically we did not send a uevent for changes to/from an empty @@ -476,7 +476,7 @@ void disk_uevent(struct gendisk *disk, enum kobject_action action) struct disk_part_iter piter; struct block_device *part; - disk_part_iter_init(&piter, disk, DISK_PITER_INCL_PART0); + disk_part_iter_init(&piter, disk, DISK_PITER_INCL_EMPTY_PART0); while ((part = disk_part_iter_next(&piter))) kobject_uevent(bdev_kobj(part), action); disk_part_iter_exit(&piter); diff --git a/block/ioctl.c b/block/ioctl.c index d61d652078f4..ff241e663c01 100644 --- a/block/ioctl.c +++ b/block/ioctl.c @@ -81,20 +81,27 @@ static int compat_blkpg_ioctl(struct block_device *bdev, } #endif -static int blkdev_reread_part(struct block_device *bdev) +static int blkdev_reread_part(struct block_device *bdev, fmode_t mode) { - int ret; + struct block_device *tmp; if (!disk_part_scan_enabled(bdev->bd_disk) || bdev_is_partition(bdev)) return -EINVAL; if (!capable(CAP_SYS_ADMIN)) return -EACCES; - mutex_lock(&bdev->bd_mutex); - ret = bdev_disk_changed(bdev, false); - mutex_unlock(&bdev->bd_mutex); + /* + * Reopen the device to revalidate the driver state and force a + * partition rescan. + */ + mode &= ~FMODE_EXCL; + set_bit(GD_NEED_PART_SCAN, &bdev->bd_disk->state); - return ret; + tmp = blkdev_get_by_dev(bdev->bd_dev, mode, NULL); + if (IS_ERR(tmp)) + return PTR_ERR(tmp); + blkdev_put(tmp, mode); + return 0; } static int blk_ioctl_discard(struct block_device *bdev, fmode_t mode, @@ -498,7 +505,7 @@ static int blkdev_common_ioctl(struct block_device *bdev, fmode_t mode, bdev->bd_bdi->ra_pages = (arg * 512) / PAGE_SIZE; return 0; case BLKRRPART: - return blkdev_reread_part(bdev); + return blkdev_reread_part(bdev, mode); case BLKTRACESTART: case BLKTRACESTOP: case BLKTRACETEARDOWN: diff --git a/block/kyber-iosched.c b/block/kyber-iosched.c index c25c41d0d061..33d34d69cade 100644 --- a/block/kyber-iosched.c +++ b/block/kyber-iosched.c @@ -13,6 +13,8 @@ #include <linux/module.h> #include <linux/sbitmap.h> +#include <trace/events/block.h> + #include "blk.h" #include "blk-mq.h" #include "blk-mq-debugfs.h" @@ -353,19 +355,9 @@ static void kyber_timer_fn(struct timer_list *t) } } -static unsigned int kyber_sched_tags_shift(struct request_queue *q) -{ - /* - * All of the hardware queues have the same depth, so we can just grab - * the shift of the first one. - */ - return q->queue_hw_ctx[0]->sched_tags->bitmap_tags->sb.shift; -} - static struct kyber_queue_data *kyber_queue_data_alloc(struct request_queue *q) { struct kyber_queue_data *kqd; - unsigned int shift; int ret = -ENOMEM; int i; @@ -400,9 +392,6 @@ static struct kyber_queue_data *kyber_queue_data_alloc(struct request_queue *q) kqd->latency_targets[i] = kyber_latency_targets[i]; } - shift = kyber_sched_tags_shift(q); - kqd->async_depth = (1U << shift) * KYBER_ASYNC_PERCENT / 100U; - return kqd; err_buckets: @@ -458,9 +447,19 @@ static void kyber_ctx_queue_init(struct kyber_ctx_queue *kcq) INIT_LIST_HEAD(&kcq->rq_list[i]); } -static int kyber_init_hctx(struct blk_mq_hw_ctx *hctx, unsigned int hctx_idx) +static void kyber_depth_updated(struct blk_mq_hw_ctx *hctx) { struct kyber_queue_data *kqd = hctx->queue->elevator->elevator_data; + struct blk_mq_tags *tags = hctx->sched_tags; + unsigned int shift = tags->bitmap_tags->sb.shift; + + kqd->async_depth = (1U << shift) * KYBER_ASYNC_PERCENT / 100U; + + sbitmap_queue_min_shallow_depth(tags->bitmap_tags, kqd->async_depth); +} + +static int kyber_init_hctx(struct blk_mq_hw_ctx *hctx, unsigned int hctx_idx) +{ struct kyber_hctx_data *khd; int i; @@ -502,8 +501,7 @@ static int kyber_init_hctx(struct blk_mq_hw_ctx *hctx, unsigned int hctx_idx) khd->batching = 0; hctx->sched_data = khd; - sbitmap_queue_min_shallow_depth(hctx->sched_tags->bitmap_tags, - kqd->async_depth); + kyber_depth_updated(hctx); return 0; @@ -602,7 +600,7 @@ static void kyber_insert_requests(struct blk_mq_hw_ctx *hctx, list_move_tail(&rq->queuelist, head); sbitmap_set_bit(&khd->kcq_map[sched_domain], rq->mq_ctx->index_hw[hctx->type]); - blk_mq_sched_request_inserted(rq); + trace_block_rq_insert(rq); spin_unlock(&kcq->lock); } } @@ -1022,6 +1020,7 @@ static struct elevator_type kyber_sched = { .completed_request = kyber_completed_request, .dispatch_request = kyber_dispatch_request, .has_work = kyber_has_work, + .depth_updated = kyber_depth_updated, }, #ifdef CONFIG_BLK_DEBUG_FS .queue_debugfs_attrs = kyber_queue_debugfs_attrs, diff --git a/block/mq-deadline.c b/block/mq-deadline.c index b57470e154c8..f3631a287466 100644 --- a/block/mq-deadline.c +++ b/block/mq-deadline.c @@ -18,6 +18,8 @@ #include <linux/rbtree.h> #include <linux/sbitmap.h> +#include <trace/events/block.h> + #include "blk.h" #include "blk-mq.h" #include "blk-mq-debugfs.h" @@ -496,7 +498,7 @@ static void dd_insert_request(struct blk_mq_hw_ctx *hctx, struct request *rq, if (blk_mq_sched_try_insert_merge(q, rq)) return; - blk_mq_sched_request_inserted(rq); + trace_block_rq_insert(rq); if (at_head || blk_rq_is_passthrough(rq)) { if (at_head) diff --git a/drivers/block/loop.c b/drivers/block/loop.c index 578fc034db3f..a370cde3ddd4 100644 --- a/drivers/block/loop.c +++ b/drivers/block/loop.c @@ -663,7 +663,7 @@ static inline int is_loop_device(struct file *file) { struct inode *i = file->f_mapping->host; - return i && S_ISBLK(i->i_mode) && MAJOR(i->i_rdev) == LOOP_MAJOR; + return i && S_ISBLK(i->i_mode) && imajor(i) == LOOP_MAJOR; } static int loop_validate_file(struct file *file, struct block_device *bdev) @@ -1212,6 +1212,9 @@ static int __loop_clr_fd(struct loop_device *lo, bool release) goto out_unlock; } + if (test_bit(QUEUE_FLAG_WC, &lo->lo_queue->queue_flags)) + blk_queue_write_cache(lo->lo_queue, false, false); + /* freeze request queue during the transition */ blk_mq_freeze_queue(lo->lo_queue); diff --git a/drivers/block/nbd.c b/drivers/block/nbd.c index 8b9622eb0a21..4ff71b579cfc 100644 --- a/drivers/block/nbd.c +++ b/drivers/block/nbd.c @@ -78,8 +78,7 @@ struct link_dead_args { #define NBD_RT_HAS_PID_FILE 3 #define NBD_RT_HAS_CONFIG_REF 4 #define NBD_RT_BOUND 5 -#define NBD_RT_DESTROY_ON_DISCONNECT 6 -#define NBD_RT_DISCONNECT_ON_CLOSE 7 +#define NBD_RT_DISCONNECT_ON_CLOSE 6 #define NBD_DESTROY_ON_DISCONNECT 0 #define NBD_DISCONNECT_REQUESTED 1 @@ -1904,12 +1903,21 @@ again: if (info->attrs[NBD_ATTR_CLIENT_FLAGS]) { u64 flags = nla_get_u64(info->attrs[NBD_ATTR_CLIENT_FLAGS]); if (flags & NBD_CFLAG_DESTROY_ON_DISCONNECT) { - set_bit(NBD_RT_DESTROY_ON_DISCONNECT, - &config->runtime_flags); - set_bit(NBD_DESTROY_ON_DISCONNECT, &nbd->flags); - put_dev = true; + /* + * We have 1 ref to keep the device around, and then 1 + * ref for our current operation here, which will be + * inherited by the config. If we already have + * DESTROY_ON_DISCONNECT set then we know we don't have + * that extra ref already held so we don't need the + * put_dev. + */ + if (!test_and_set_bit(NBD_DESTROY_ON_DISCONNECT, + &nbd->flags)) + put_dev = true; } else { - clear_bit(NBD_DESTROY_ON_DISCONNECT, &nbd->flags); + if (test_and_clear_bit(NBD_DESTROY_ON_DISCONNECT, + &nbd->flags)) + refcount_inc(&nbd->refs); } if (flags & NBD_CFLAG_DISCONNECT_ON_CLOSE) { set_bit(NBD_RT_DISCONNECT_ON_CLOSE, @@ -2080,15 +2088,13 @@ static int nbd_genl_reconfigure(struct sk_buff *skb, struct genl_info *info) if (info->attrs[NBD_ATTR_CLIENT_FLAGS]) { u64 flags = nla_get_u64(info->attrs[NBD_ATTR_CLIENT_FLAGS]); if (flags & NBD_CFLAG_DESTROY_ON_DISCONNECT) { - if (!test_and_set_bit(NBD_RT_DESTROY_ON_DISCONNECT, - &config->runtime_flags)) + if (!test_and_set_bit(NBD_DESTROY_ON_DISCONNECT, + &nbd->flags)) put_dev = true; - set_bit(NBD_DESTROY_ON_DISCONNECT, &nbd->flags); } else { - if (test_and_clear_bit(NBD_RT_DESTROY_ON_DISCONNECT, - &config->runtime_flags)) + if (test_and_clear_bit(NBD_DESTROY_ON_DISCONNECT, + &nbd->flags)) refcount_inc(&nbd->refs); - clear_bit(NBD_DESTROY_ON_DISCONNECT, &nbd->flags); } if (flags & NBD_CFLAG_DISCONNECT_ON_CLOSE) { diff --git a/drivers/block/xen-blkback/blkback.c b/drivers/block/xen-blkback/blkback.c index da16121140ca..1cdf09ff67b6 100644 --- a/drivers/block/xen-blkback/blkback.c +++ b/drivers/block/xen-blkback/blkback.c @@ -1326,9 +1326,7 @@ static int dispatch_rw_block_io(struct xen_blkif_ring *ring, pages[i]->page, seg[i].nsec << 9, seg[i].offset) == 0)) { - - int nr_iovecs = min_t(int, (nseg-i), BIO_MAX_PAGES); - bio = bio_alloc(GFP_KERNEL, nr_iovecs); + bio = bio_alloc(GFP_KERNEL, bio_max_segs(nseg - i)); if (unlikely(bio == NULL)) goto fail_put_bio; diff --git a/drivers/dax/super.c b/drivers/dax/super.c index cadbd0a1a1ef..5fa6ae9dbc8b 100644 --- a/drivers/dax/super.c +++ b/drivers/dax/super.c @@ -480,7 +480,7 @@ static void dax_free_inode(struct inode *inode) kfree(dax_dev->host); dax_dev->host = NULL; if (inode->i_rdev) - ida_simple_remove(&dax_minor_ida, MINOR(inode->i_rdev)); + ida_simple_remove(&dax_minor_ida, iminor(inode)); kmem_cache_free(dax_cache, dax_dev); } diff --git a/drivers/i2c/busses/i2c-brcmstb.c b/drivers/i2c/busses/i2c-brcmstb.c index d4e0a0f6732a..ba766d24219e 100644 --- a/drivers/i2c/busses/i2c-brcmstb.c +++ b/drivers/i2c/busses/i2c-brcmstb.c @@ -316,7 +316,7 @@ static int brcmstb_send_i2c_cmd(struct brcmstb_i2c_dev *dev, goto cmd_out; } - if ((CMD_RD || CMD_WR) && + if ((cmd == CMD_RD || cmd == CMD_WR) && bsc_readl(dev, iic_enable) & BSC_IIC_EN_NOACK_MASK) { rc = -EREMOTEIO; dev_dbg(dev->device, "controller received NOACK intr for %s\n", diff --git a/drivers/i2c/busses/i2c-designware-core.h b/drivers/i2c/busses/i2c-designware-core.h index 85307cfa7109..5392b82f68a4 100644 --- a/drivers/i2c/busses/i2c-designware-core.h +++ b/drivers/i2c/busses/i2c-designware-core.h @@ -38,6 +38,8 @@ #define DW_IC_CON_TX_EMPTY_CTRL BIT(8) #define DW_IC_CON_RX_FIFO_FULL_HLD_CTRL BIT(9) +#define DW_IC_DATA_CMD_DAT GENMASK(7, 0) + /* * Registers offset */ diff --git a/drivers/i2c/busses/i2c-designware-master.c b/drivers/i2c/busses/i2c-designware-master.c index d6425ad6e6a3..dd27b9dbe931 100644 --- a/drivers/i2c/busses/i2c-designware-master.c +++ b/drivers/i2c/busses/i2c-designware-master.c @@ -432,7 +432,7 @@ i2c_dw_read(struct dw_i2c_dev *dev) regmap_read(dev->map, DW_IC_DATA_CMD, &tmp); /* Ensure length byte is a valid value */ if (flags & I2C_M_RECV_LEN && - tmp <= I2C_SMBUS_BLOCK_MAX && tmp > 0) { + (tmp & DW_IC_DATA_CMD_DAT) <= I2C_SMBUS_BLOCK_MAX && tmp > 0) { len = i2c_dw_recv_len(dev, tmp); } *buf++ = tmp; diff --git a/drivers/i2c/busses/i2c-exynos5.c b/drivers/i2c/busses/i2c-exynos5.c index 20a9881a0d6c..5ac30d95650c 100644 --- a/drivers/i2c/busses/i2c-exynos5.c +++ b/drivers/i2c/busses/i2c-exynos5.c @@ -606,6 +606,7 @@ static void exynos5_i2c_message_start(struct exynos5_i2c *i2c, int stop) u32 i2c_ctl; u32 int_en = 0; u32 i2c_auto_conf = 0; + u32 i2c_addr = 0; u32 fifo_ctl; unsigned long flags; unsigned short trig_lvl; @@ -640,7 +641,12 @@ static void exynos5_i2c_message_start(struct exynos5_i2c *i2c, int stop) int_en |= HSI2C_INT_TX_ALMOSTEMPTY_EN; } - writel(HSI2C_SLV_ADDR_MAS(i2c->msg->addr), i2c->regs + HSI2C_ADDR); + i2c_addr = HSI2C_SLV_ADDR_MAS(i2c->msg->addr); + + if (i2c->op_clock >= I2C_MAX_FAST_MODE_PLUS_FREQ) + i2c_addr |= HSI2C_MASTER_ID(MASTER_ID(i2c->adap.nr)); + + writel(i2c_addr, i2c->regs + HSI2C_ADDR); writel(fifo_ctl, i2c->regs + HSI2C_FIFO_CTL); writel(i2c_ctl, i2c->regs + HSI2C_CTL); diff --git a/drivers/i2c/busses/i2c-qcom-geni.c b/drivers/i2c/busses/i2c-qcom-geni.c index c3f584795911..214b4c913a13 100644 --- a/drivers/i2c/busses/i2c-qcom-geni.c +++ b/drivers/i2c/busses/i2c-qcom-geni.c @@ -375,32 +375,6 @@ static void geni_i2c_tx_msg_cleanup(struct geni_i2c_dev *gi2c, } } -static void geni_i2c_stop_xfer(struct geni_i2c_dev *gi2c) -{ - int ret; - u32 geni_status; - struct i2c_msg *cur; - - /* Resume device, as runtime suspend can happen anytime during transfer */ - ret = pm_runtime_get_sync(gi2c->se.dev); - if (ret < 0) { - dev_err(gi2c->se.dev, "Failed to resume device: %d\n", ret); - return; - } - - geni_status = readl_relaxed(gi2c->se.base + SE_GENI_STATUS); - if (geni_status & M_GENI_CMD_ACTIVE) { - cur = gi2c->cur; - geni_i2c_abort_xfer(gi2c); - if (cur->flags & I2C_M_RD) - geni_i2c_rx_msg_cleanup(gi2c, cur); - else - geni_i2c_tx_msg_cleanup(gi2c, cur); - } - - pm_runtime_put_sync_suspend(gi2c->se.dev); -} - static int geni_i2c_rx_one_msg(struct geni_i2c_dev *gi2c, struct i2c_msg *msg, u32 m_param) { @@ -676,13 +650,6 @@ static int geni_i2c_remove(struct platform_device *pdev) return 0; } -static void geni_i2c_shutdown(struct platform_device *pdev) -{ - struct geni_i2c_dev *gi2c = platform_get_drvdata(pdev); - - geni_i2c_stop_xfer(gi2c); -} - static int __maybe_unused geni_i2c_runtime_suspend(struct device *dev) { int ret; @@ -747,7 +714,6 @@ MODULE_DEVICE_TABLE(of, geni_i2c_dt_match); static struct platform_driver geni_i2c_driver = { .probe = geni_i2c_probe, .remove = geni_i2c_remove, - .shutdown = geni_i2c_shutdown, .driver = { .name = "geni_i2c", .pm = &geni_i2c_pm_ops, diff --git a/drivers/md/dm-io.c b/drivers/md/dm-io.c index 4312007d2d34..2d3cda0acacb 100644 --- a/drivers/md/dm-io.c +++ b/drivers/md/dm-io.c @@ -341,8 +341,8 @@ static void do_region(int op, int op_flags, unsigned region, num_bvecs = 1; break; default: - num_bvecs = min_t(int, BIO_MAX_PAGES, - dm_sector_div_up(remaining, (PAGE_SIZE >> SECTOR_SHIFT))); + num_bvecs = bio_max_segs(dm_sector_div_up(remaining, + (PAGE_SIZE >> SECTOR_SHIFT))); } bio = bio_alloc_bioset(GFP_NOIO, num_bvecs, &io->client->bios); diff --git a/drivers/md/dm-log-writes.c b/drivers/md/dm-log-writes.c index e3d35c6c9f71..57882654ffee 100644 --- a/drivers/md/dm-log-writes.c +++ b/drivers/md/dm-log-writes.c @@ -264,15 +264,14 @@ static int write_inline_data(struct log_writes_c *lc, void *entry, size_t entrylen, void *data, size_t datalen, sector_t sector) { - int num_pages, bio_pages, pg_datalen, pg_sectorlen, i; + int bio_pages, pg_datalen, pg_sectorlen, i; struct page *page; struct bio *bio; size_t ret; void *ptr; while (datalen) { - num_pages = ALIGN(datalen, PAGE_SIZE) >> PAGE_SHIFT; - bio_pages = min(num_pages, BIO_MAX_PAGES); + bio_pages = bio_max_segs(DIV_ROUND_UP(datalen, PAGE_SIZE)); atomic_inc(&lc->io_blocks); @@ -364,7 +363,7 @@ static int log_one_block(struct log_writes_c *lc, goto out; atomic_inc(&lc->io_blocks); - bio = bio_alloc(GFP_KERNEL, min(block->vec_cnt, BIO_MAX_PAGES)); + bio = bio_alloc(GFP_KERNEL, bio_max_segs(block->vec_cnt)); if (!bio) { DMERR("Couldn't alloc log bio"); goto error; @@ -386,7 +385,8 @@ static int log_one_block(struct log_writes_c *lc, if (ret != block->vecs[i].bv_len) { atomic_inc(&lc->io_blocks); submit_bio(bio); - bio = bio_alloc(GFP_KERNEL, min(block->vec_cnt - i, BIO_MAX_PAGES)); + bio = bio_alloc(GFP_KERNEL, + bio_max_segs(block->vec_cnt - i)); if (!bio) { DMERR("Couldn't alloc log bio"); goto error; diff --git a/drivers/nvme/target/io-cmd-bdev.c b/drivers/nvme/target/io-cmd-bdev.c index 3d9a5d3ed9cd..9a8b3726a37c 100644 --- a/drivers/nvme/target/io-cmd-bdev.c +++ b/drivers/nvme/target/io-cmd-bdev.c @@ -185,7 +185,7 @@ static int nvmet_bdev_alloc_bip(struct nvmet_req *req, struct bio *bio, } bip = bio_integrity_alloc(bio, GFP_NOIO, - min_t(unsigned int, req->metadata_sg_cnt, BIO_MAX_PAGES)); + bio_max_segs(req->metadata_sg_cnt)); if (IS_ERR(bip)) { pr_err("Unable to allocate bio_integrity_payload\n"); return PTR_ERR(bip); @@ -225,7 +225,7 @@ static int nvmet_bdev_alloc_bip(struct nvmet_req *req, struct bio *bio, static void nvmet_bdev_execute_rw(struct nvmet_req *req) { - int sg_cnt = req->sg_cnt; + unsigned int sg_cnt = req->sg_cnt; struct bio *bio; struct scatterlist *sg; struct blk_plug plug; @@ -262,7 +262,7 @@ static void nvmet_bdev_execute_rw(struct nvmet_req *req) bio = &req->b.inline_bio; bio_init(bio, req->inline_bvec, ARRAY_SIZE(req->inline_bvec)); } else { - bio = bio_alloc(GFP_KERNEL, min(sg_cnt, BIO_MAX_PAGES)); + bio = bio_alloc(GFP_KERNEL, bio_max_segs(sg_cnt)); } bio_set_dev(bio, req->ns->bdev); bio->bi_iter.bi_sector = sector; @@ -289,7 +289,7 @@ static void nvmet_bdev_execute_rw(struct nvmet_req *req) } } - bio = bio_alloc(GFP_KERNEL, min(sg_cnt, BIO_MAX_PAGES)); + bio = bio_alloc(GFP_KERNEL, bio_max_segs(sg_cnt)); bio_set_dev(bio, req->ns->bdev); bio->bi_iter.bi_sector = sector; bio->bi_opf = op; diff --git a/drivers/nvme/target/passthru.c b/drivers/nvme/target/passthru.c index f50c7b2bf21c..26c587ccd152 100644 --- a/drivers/nvme/target/passthru.c +++ b/drivers/nvme/target/passthru.c @@ -26,7 +26,7 @@ static u16 nvmet_passthru_override_id_ctrl(struct nvmet_req *req) struct nvme_ctrl *pctrl = ctrl->subsys->passthru_ctrl; u16 status = NVME_SC_SUCCESS; struct nvme_id_ctrl *id; - int max_hw_sectors; + unsigned int max_hw_sectors; int page_shift; id = kzalloc(sizeof(*id), GFP_KERNEL); @@ -198,7 +198,7 @@ static int nvmet_passthru_map_sg(struct nvmet_req *req, struct request *rq) bio = &req->p.inline_bio; bio_init(bio, req->inline_bvec, ARRAY_SIZE(req->inline_bvec)); } else { - bio = bio_alloc(GFP_KERNEL, min(req->sg_cnt, BIO_MAX_PAGES)); + bio = bio_alloc(GFP_KERNEL, bio_max_segs(req->sg_cnt)); bio->bi_end_io = bio_put; } bio->bi_opf = req_op(rq); diff --git a/drivers/rtc/rtc-m41t80.c b/drivers/rtc/rtc-m41t80.c index cce5b5284658..89128fc29ccc 100644 --- a/drivers/rtc/rtc-m41t80.c +++ b/drivers/rtc/rtc-m41t80.c @@ -785,7 +785,7 @@ static long wdt_unlocked_ioctl(struct file *file, unsigned int cmd, */ static int wdt_open(struct inode *inode, struct file *file) { - if (MINOR(inode->i_rdev) == WATCHDOG_MINOR) { + if (iminor(inode) == WATCHDOG_MINOR) { mutex_lock(&m41t80_rtc_mutex); if (test_and_set_bit(0, &wdt_is_open)) { mutex_unlock(&m41t80_rtc_mutex); @@ -809,7 +809,7 @@ static int wdt_open(struct inode *inode, struct file *file) */ static int wdt_release(struct inode *inode, struct file *file) { - if (MINOR(inode->i_rdev) == WATCHDOG_MINOR) + if (iminor(inode) == WATCHDOG_MINOR) clear_bit(0, &wdt_is_open); return 0; } diff --git a/drivers/s390/char/vmur.c b/drivers/s390/char/vmur.c index 1bbf27b98cf6..68f49e2e964c 100644 --- a/drivers/s390/char/vmur.c +++ b/drivers/s390/char/vmur.c @@ -681,7 +681,7 @@ static int ur_open(struct inode *inode, struct file *file) * We treat the minor number as the devno of the ur device * to find in the driver tree. */ - devno = MINOR(file_inode(file)->i_rdev); + devno = iminor(file_inode(file)); urd = urdev_get_from_devno(devno); if (!urd) { diff --git a/drivers/scsi/aic7xxx/aic79xx.h b/drivers/scsi/aic7xxx/aic79xx.h index 77e1d6bb59a3..59e9321815c8 100644 --- a/drivers/scsi/aic7xxx/aic79xx.h +++ b/drivers/scsi/aic7xxx/aic79xx.h @@ -1175,7 +1175,7 @@ struct ahd_softc { uint8_t tqinfifonext; /* - * Cached verson of the hs_mailbox so we can avoid + * Cached version of the hs_mailbox so we can avoid * pausing the sequencer during mailbox updates. */ uint8_t hs_mailbox; diff --git a/drivers/scsi/aic7xxx/aic7xxx.h b/drivers/scsi/aic7xxx/aic7xxx.h index 11a09798e6b5..9bc755a0a2d3 100644 --- a/drivers/scsi/aic7xxx/aic7xxx.h +++ b/drivers/scsi/aic7xxx/aic7xxx.h @@ -896,8 +896,6 @@ union ahc_bus_softc { typedef void (*ahc_bus_intr_t)(struct ahc_softc *); typedef int (*ahc_bus_chip_init_t)(struct ahc_softc *); -typedef int (*ahc_bus_suspend_t)(struct ahc_softc *); -typedef int (*ahc_bus_resume_t)(struct ahc_softc *); typedef void ahc_callback_t (void *); struct ahc_softc { diff --git a/drivers/scsi/bnx2fc/Kconfig b/drivers/scsi/bnx2fc/Kconfig index 3cf7e08df809..ecdc0f0f4f4e 100644 --- a/drivers/scsi/bnx2fc/Kconfig +++ b/drivers/scsi/bnx2fc/Kconfig @@ -5,6 +5,7 @@ config SCSI_BNX2X_FCOE depends on (IPV6 || IPV6=n) depends on LIBFC depends on LIBFCOE + depends on MMU select NETDEVICES select ETHERNET select NET_VENDOR_BROADCOM diff --git a/drivers/scsi/bnx2i/bnx2i_iscsi.c b/drivers/scsi/bnx2i/bnx2i_iscsi.c index fdd446765311..1e6d8f62ea3c 100644 --- a/drivers/scsi/bnx2i/bnx2i_iscsi.c +++ b/drivers/scsi/bnx2i/bnx2i_iscsi.c @@ -1171,10 +1171,8 @@ static void bnx2i_cleanup_task(struct iscsi_task *task) bnx2i_send_cmd_cleanup_req(hba, task->dd_data); spin_unlock_bh(&conn->session->back_lock); - spin_unlock_bh(&conn->session->frwd_lock); wait_for_completion_timeout(&bnx2i_conn->cmd_cleanup_cmpl, msecs_to_jiffies(ISCSI_CMD_CLEANUP_TIMEOUT)); - spin_lock_bh(&conn->session->frwd_lock); spin_lock_bh(&conn->session->back_lock); } bnx2i_iscsi_unmap_sg_list(task->dd_data); diff --git a/drivers/scsi/hpsa.c b/drivers/scsi/hpsa.c index 337d3aa91945..38369766511c 100644 --- a/drivers/scsi/hpsa.c +++ b/drivers/scsi/hpsa.c @@ -1151,7 +1151,10 @@ static void __enqueue_cmd_and_start_io(struct ctlr_info *h, { dial_down_lockup_detection_during_fw_flash(h, c); atomic_inc(&h->commands_outstanding); - if (c->device) + /* + * Check to see if the command is being retried. + */ + if (c->device && !c->retry_pending) atomic_inc(&c->device->commands_outstanding); reply_queue = h->reply_map[raw_smp_processor_id()]; @@ -5567,7 +5570,8 @@ static inline void hpsa_cmd_partial_init(struct ctlr_info *h, int index, } static int hpsa_ioaccel_submit(struct ctlr_info *h, - struct CommandList *c, struct scsi_cmnd *cmd) + struct CommandList *c, struct scsi_cmnd *cmd, + bool retry) { struct hpsa_scsi_dev_t *dev = cmd->device->hostdata; int rc = IO_ACCEL_INELIGIBLE; @@ -5584,18 +5588,22 @@ static int hpsa_ioaccel_submit(struct ctlr_info *h, cmd->host_scribble = (unsigned char *) c; if (dev->offload_enabled) { - hpsa_cmd_init(h, c->cmdindex, c); + hpsa_cmd_init(h, c->cmdindex, c); /* Zeroes out all fields */ c->cmd_type = CMD_SCSI; c->scsi_cmd = cmd; c->device = dev; + if (retry) /* Resubmit but do not increment device->commands_outstanding. */ + c->retry_pending = true; rc = hpsa_scsi_ioaccel_raid_map(h, c); if (rc < 0) /* scsi_dma_map failed. */ rc = SCSI_MLQUEUE_HOST_BUSY; } else if (dev->hba_ioaccel_enabled) { - hpsa_cmd_init(h, c->cmdindex, c); + hpsa_cmd_init(h, c->cmdindex, c); /* Zeroes out all fields */ c->cmd_type = CMD_SCSI; c->scsi_cmd = cmd; c->device = dev; + if (retry) /* Resubmit but do not increment device->commands_outstanding. */ + c->retry_pending = true; rc = hpsa_scsi_ioaccel_direct_map(h, c); if (rc < 0) /* scsi_dma_map failed. */ rc = SCSI_MLQUEUE_HOST_BUSY; @@ -5628,7 +5636,8 @@ static void hpsa_command_resubmit_worker(struct work_struct *work) if (c2->error_data.serv_response == IOACCEL2_STATUS_SR_TASK_COMP_SET_FULL) { - rc = hpsa_ioaccel_submit(h, c, cmd); + /* Resubmit with the retry_pending flag set. */ + rc = hpsa_ioaccel_submit(h, c, cmd, true); if (rc == 0) return; if (rc == SCSI_MLQUEUE_HOST_BUSY) { @@ -5644,6 +5653,15 @@ static void hpsa_command_resubmit_worker(struct work_struct *work) } } hpsa_cmd_partial_init(c->h, c->cmdindex, c); + /* + * Here we have not come in though queue_command, so we + * can set the retry_pending flag to true for a driver initiated + * retry attempt (I.E. not a SML retry). + * I.E. We are submitting a driver initiated retry. + * Note: hpsa_ciss_submit does not zero out the command fields like + * ioaccel submit does. + */ + c->retry_pending = true; if (hpsa_ciss_submit(c->h, c, cmd, dev)) { /* * If we get here, it means dma mapping failed. Try @@ -5706,11 +5724,16 @@ static int hpsa_scsi_queue_command(struct Scsi_Host *sh, struct scsi_cmnd *cmd) /* * Call alternate submit routine for I/O accelerated commands. * Retries always go down the normal I/O path. + * Note: If cmd->retries is non-zero, then this is a SML + * initiated retry and not a driver initiated retry. + * This command has been obtained from cmd_tagged_alloc + * and is therefore a brand-new command. */ if (likely(cmd->retries == 0 && !blk_rq_is_passthrough(cmd->request) && h->acciopath_status)) { - rc = hpsa_ioaccel_submit(h, c, cmd); + /* Submit with the retry_pending flag unset. */ + rc = hpsa_ioaccel_submit(h, c, cmd, false); if (rc == 0) return 0; if (rc == SCSI_MLQUEUE_HOST_BUSY) { @@ -6105,6 +6128,7 @@ return_reset_status: * at init, and managed by cmd_tagged_alloc() and cmd_tagged_free() using the * block request tag as an index into a table of entries. cmd_tagged_free() is * the complement, although cmd_free() may be called instead. + * This function is only called for new requests from queue_command. */ static struct CommandList *cmd_tagged_alloc(struct ctlr_info *h, struct scsi_cmnd *scmd) @@ -6139,8 +6163,14 @@ static struct CommandList *cmd_tagged_alloc(struct ctlr_info *h, } atomic_inc(&c->refcount); - hpsa_cmd_partial_init(h, idx, c); + + /* + * This is a new command obtained from queue_command so + * there have not been any driver initiated retry attempts. + */ + c->retry_pending = false; + return c; } @@ -6208,6 +6238,13 @@ static struct CommandList *cmd_alloc(struct ctlr_info *h) } hpsa_cmd_partial_init(h, i, c); c->device = NULL; + + /* + * cmd_alloc is for "internal" commands and they are never + * retried. + */ + c->retry_pending = false; + return c; } diff --git a/drivers/scsi/hpsa_cmd.h b/drivers/scsi/hpsa_cmd.h index 46df2e3ff89b..d126bb877250 100644 --- a/drivers/scsi/hpsa_cmd.h +++ b/drivers/scsi/hpsa_cmd.h @@ -448,7 +448,7 @@ struct CommandList { */ struct hpsa_scsi_dev_t *phys_disk; - int abort_pending; + bool retry_pending; struct hpsa_scsi_dev_t *device; atomic_t refcount; /* Must be last to avoid memset in hpsa_cmd_init() */ } __aligned(COMMANDLIST_ALIGNMENT); diff --git a/drivers/scsi/isci/request.c b/drivers/scsi/isci/request.c index bee16850b236..58e62162882f 100644 --- a/drivers/scsi/isci/request.c +++ b/drivers/scsi/isci/request.c @@ -1480,8 +1480,6 @@ static enum sci_status stp_request_pio_await_h2d_completion_tc_event(struct isci_request *ireq, u32 completion_code) { - enum sci_status status = SCI_SUCCESS; - switch (SCU_GET_COMPLETION_TL_STATUS(completion_code)) { case SCU_MAKE_COMPLETION_STATUS(SCU_TASK_DONE_GOOD): ireq->scu_status = SCU_TASK_DONE_GOOD; @@ -1500,7 +1498,7 @@ stp_request_pio_await_h2d_completion_tc_event(struct isci_request *ireq, break; } - return status; + return SCI_SUCCESS; } static enum sci_status @@ -2152,8 +2150,6 @@ static enum sci_status stp_request_udma_await_tc_event(struct isci_request *ireq static enum sci_status atapi_raw_completion(struct isci_request *ireq, u32 completion_code, enum sci_base_request_states next) { - enum sci_status status = SCI_SUCCESS; - switch (SCU_GET_COMPLETION_TL_STATUS(completion_code)) { case SCU_MAKE_COMPLETION_STATUS(SCU_TASK_DONE_GOOD): ireq->scu_status = SCU_TASK_DONE_GOOD; @@ -2172,7 +2168,7 @@ static enum sci_status atapi_raw_completion(struct isci_request *ireq, u32 compl break; } - return status; + return SCI_SUCCESS; } static enum sci_status atapi_data_tc_completion_handler(struct isci_request *ireq, diff --git a/drivers/scsi/iscsi_tcp.c b/drivers/scsi/iscsi_tcp.c index a9ce6298b935..dd33ce0e3737 100644 --- a/drivers/scsi/iscsi_tcp.c +++ b/drivers/scsi/iscsi_tcp.c @@ -847,6 +847,7 @@ iscsi_sw_tcp_session_create(struct iscsi_endpoint *ep, uint16_t cmds_max, struct iscsi_session *session; struct iscsi_sw_tcp_host *tcp_sw_host; struct Scsi_Host *shost; + int rc; if (ep) { printk(KERN_ERR "iscsi_tcp: invalid ep %p.\n", ep); @@ -864,6 +865,11 @@ iscsi_sw_tcp_session_create(struct iscsi_endpoint *ep, uint16_t cmds_max, shost->max_channel = 0; shost->max_cmd_len = SCSI_MAX_VARLEN_CDB_SIZE; + rc = iscsi_host_get_max_scsi_cmds(shost, cmds_max); + if (rc < 0) + goto free_host; + shost->can_queue = rc; + if (iscsi_host_add(shost, NULL)) goto free_host; @@ -878,7 +884,6 @@ iscsi_sw_tcp_session_create(struct iscsi_endpoint *ep, uint16_t cmds_max, tcp_sw_host = iscsi_host_priv(shost); tcp_sw_host->session = session; - shost->can_queue = session->scsi_cmds_max; if (iscsi_tcp_r2tpool_alloc(session)) goto remove_session; return cls_session; @@ -981,7 +986,7 @@ static struct scsi_host_template iscsi_sw_tcp_sht = { .name = "iSCSI Initiator over TCP/IP", .queuecommand = iscsi_queuecommand, .change_queue_depth = scsi_change_queue_depth, - .can_queue = ISCSI_DEF_XMIT_CMDS_MAX - 1, + .can_queue = ISCSI_TOTAL_CMDS_MAX, .sg_tablesize = 4096, .max_sectors = 0xFFFF, .cmd_per_lun = ISCSI_DEF_CMD_PER_LUN, diff --git a/drivers/scsi/libiscsi.c b/drivers/scsi/libiscsi.c index 4e668aafbcca..7ad11e42306d 100644 --- a/drivers/scsi/libiscsi.c +++ b/drivers/scsi/libiscsi.c @@ -523,16 +523,6 @@ static void iscsi_complete_task(struct iscsi_task *task, int state) WARN_ON_ONCE(task->state == ISCSI_TASK_FREE); task->state = state; - spin_lock_bh(&conn->taskqueuelock); - if (!list_empty(&task->running)) { - pr_debug_once("%s while task on list", __func__); - list_del_init(&task->running); - } - spin_unlock_bh(&conn->taskqueuelock); - - if (conn->task == task) - conn->task = NULL; - if (READ_ONCE(conn->ping_task) == task) WRITE_ONCE(conn->ping_task, NULL); @@ -564,11 +554,41 @@ void iscsi_complete_scsi_task(struct iscsi_task *task, } EXPORT_SYMBOL_GPL(iscsi_complete_scsi_task); +/* + * Must be called with back and frwd lock + */ +static bool cleanup_queued_task(struct iscsi_task *task) +{ + struct iscsi_conn *conn = task->conn; + bool early_complete = false; + + /* Bad target might have completed task while it was still running */ + if (task->state == ISCSI_TASK_COMPLETED) + early_complete = true; + + if (!list_empty(&task->running)) { + list_del_init(&task->running); + /* + * If it's on a list but still running, this could be from + * a bad target sending a rsp early, cleanup from a TMF, or + * session recovery. + */ + if (task->state == ISCSI_TASK_RUNNING || + task->state == ISCSI_TASK_COMPLETED) + __iscsi_put_task(task); + } + + if (conn->task == task) { + conn->task = NULL; + __iscsi_put_task(task); + } + + return early_complete; +} /* - * session back_lock must be held and if not called for a task that is - * still pending or from the xmit thread, then xmit thread must - * be suspended. + * session frwd lock must be held and if not called for a task that is still + * pending or from the xmit thread, then xmit thread must be suspended */ static void fail_scsi_task(struct iscsi_task *task, int err) { @@ -576,14 +596,11 @@ static void fail_scsi_task(struct iscsi_task *task, int err) struct scsi_cmnd *sc; int state; - /* - * if a command completes and we get a successful tmf response - * we will hit this because the scsi eh abort code does not take - * a ref to the task. - */ - sc = task->sc; - if (!sc) + spin_lock_bh(&conn->session->back_lock); + if (cleanup_queued_task(task)) { + spin_unlock_bh(&conn->session->back_lock); return; + } if (task->state == ISCSI_TASK_PENDING) { /* @@ -598,11 +615,9 @@ static void fail_scsi_task(struct iscsi_task *task, int err) else state = ISCSI_TASK_ABRT_TMF; + sc = task->sc; sc->result = err << 16; scsi_set_resid(sc, scsi_bufflen(sc)); - - /* regular RX path uses back_lock */ - spin_lock_bh(&conn->session->back_lock); iscsi_complete_task(task, state); spin_unlock_bh(&conn->session->back_lock); } @@ -748,9 +763,7 @@ __iscsi_conn_send_pdu(struct iscsi_conn *conn, struct iscsi_hdr *hdr, if (session->tt->xmit_task(task)) goto free_task; } else { - spin_lock_bh(&conn->taskqueuelock); list_add_tail(&task->running, &conn->mgmtqueue); - spin_unlock_bh(&conn->taskqueuelock); iscsi_conn_queue_work(conn); } @@ -1411,31 +1424,61 @@ static int iscsi_check_cmdsn_window_closed(struct iscsi_conn *conn) return 0; } -static int iscsi_xmit_task(struct iscsi_conn *conn) +static int iscsi_xmit_task(struct iscsi_conn *conn, struct iscsi_task *task, + bool was_requeue) { - struct iscsi_task *task = conn->task; int rc; - if (test_bit(ISCSI_SUSPEND_BIT, &conn->suspend_tx)) - return -ENODATA; - spin_lock_bh(&conn->session->back_lock); - if (conn->task == NULL) { + + if (!conn->task) { + /* Take a ref so we can access it after xmit_task() */ + __iscsi_get_task(task); + } else { + /* Already have a ref from when we failed to send it last call */ + conn->task = NULL; + } + + /* + * If this was a requeue for a R2T we have an extra ref on the task in + * case a bad target sends a cmd rsp before we have handled the task. + */ + if (was_requeue) + __iscsi_put_task(task); + + /* + * Do this after dropping the extra ref because if this was a requeue + * it's removed from that list and cleanup_queued_task would miss it. + */ + if (test_bit(ISCSI_SUSPEND_BIT, &conn->suspend_tx)) { + /* + * Save the task and ref in case we weren't cleaning up this + * task and get woken up again. + */ + conn->task = task; spin_unlock_bh(&conn->session->back_lock); return -ENODATA; } - __iscsi_get_task(task); spin_unlock_bh(&conn->session->back_lock); + spin_unlock_bh(&conn->session->frwd_lock); rc = conn->session->tt->xmit_task(task); spin_lock_bh(&conn->session->frwd_lock); if (!rc) { /* done with this task */ task->last_xfer = jiffies; - conn->task = NULL; } /* regular RX path uses back_lock */ spin_lock(&conn->session->back_lock); + if (rc && task->state == ISCSI_TASK_RUNNING) { + /* + * get an extra ref that is released next time we access it + * as conn->task above. + */ + __iscsi_get_task(task); + conn->task = task; + } + __iscsi_put_task(task); spin_unlock(&conn->session->back_lock); return rc; @@ -1445,9 +1488,7 @@ static int iscsi_xmit_task(struct iscsi_conn *conn) * iscsi_requeue_task - requeue task to run from session workqueue * @task: task to requeue * - * LLDs that need to run a task from the session workqueue should call - * this. The session frwd_lock must be held. This should only be called - * by software drivers. + * Callers must have taken a ref to the task that is going to be requeued. */ void iscsi_requeue_task(struct iscsi_task *task) { @@ -1457,11 +1498,18 @@ void iscsi_requeue_task(struct iscsi_task *task) * this may be on the requeue list already if the xmit_task callout * is handling the r2ts while we are adding new ones */ - spin_lock_bh(&conn->taskqueuelock); - if (list_empty(&task->running)) + spin_lock_bh(&conn->session->frwd_lock); + if (list_empty(&task->running)) { list_add_tail(&task->running, &conn->requeue); - spin_unlock_bh(&conn->taskqueuelock); + } else { + /* + * Don't need the extra ref since it's already requeued and + * has a ref. + */ + iscsi_put_task(task); + } iscsi_conn_queue_work(conn); + spin_unlock_bh(&conn->session->frwd_lock); } EXPORT_SYMBOL_GPL(iscsi_requeue_task); @@ -1487,7 +1535,7 @@ static int iscsi_data_xmit(struct iscsi_conn *conn) } if (conn->task) { - rc = iscsi_xmit_task(conn); + rc = iscsi_xmit_task(conn, conn->task, false); if (rc) goto done; } @@ -1497,54 +1545,41 @@ static int iscsi_data_xmit(struct iscsi_conn *conn) * only have one nop-out as a ping from us and targets should not * overflow us with nop-ins */ - spin_lock_bh(&conn->taskqueuelock); check_mgmt: while (!list_empty(&conn->mgmtqueue)) { - conn->task = list_entry(conn->mgmtqueue.next, - struct iscsi_task, running); - list_del_init(&conn->task->running); - spin_unlock_bh(&conn->taskqueuelock); - if (iscsi_prep_mgmt_task(conn, conn->task)) { + task = list_entry(conn->mgmtqueue.next, struct iscsi_task, + running); + list_del_init(&task->running); + if (iscsi_prep_mgmt_task(conn, task)) { /* regular RX path uses back_lock */ spin_lock_bh(&conn->session->back_lock); - __iscsi_put_task(conn->task); + __iscsi_put_task(task); spin_unlock_bh(&conn->session->back_lock); - conn->task = NULL; - spin_lock_bh(&conn->taskqueuelock); continue; } - rc = iscsi_xmit_task(conn); + rc = iscsi_xmit_task(conn, task, false); if (rc) goto done; - spin_lock_bh(&conn->taskqueuelock); } /* process pending command queue */ while (!list_empty(&conn->cmdqueue)) { - conn->task = list_entry(conn->cmdqueue.next, struct iscsi_task, - running); - list_del_init(&conn->task->running); - spin_unlock_bh(&conn->taskqueuelock); + task = list_entry(conn->cmdqueue.next, struct iscsi_task, + running); + list_del_init(&task->running); if (conn->session->state == ISCSI_STATE_LOGGING_OUT) { - fail_scsi_task(conn->task, DID_IMM_RETRY); - spin_lock_bh(&conn->taskqueuelock); + fail_scsi_task(task, DID_IMM_RETRY); continue; } - rc = iscsi_prep_scsi_cmd_pdu(conn->task); + rc = iscsi_prep_scsi_cmd_pdu(task); if (rc) { - if (rc == -ENOMEM || rc == -EACCES) { - spin_lock_bh(&conn->taskqueuelock); - list_add_tail(&conn->task->running, - &conn->cmdqueue); - conn->task = NULL; - spin_unlock_bh(&conn->taskqueuelock); - goto done; - } else - fail_scsi_task(conn->task, DID_ABORT); - spin_lock_bh(&conn->taskqueuelock); + if (rc == -ENOMEM || rc == -EACCES) + fail_scsi_task(task, DID_IMM_RETRY); + else + fail_scsi_task(task, DID_ABORT); continue; } - rc = iscsi_xmit_task(conn); + rc = iscsi_xmit_task(conn, task, false); if (rc) goto done; /* @@ -1552,7 +1587,6 @@ check_mgmt: * we need to check the mgmt queue for nops that need to * be sent to aviod starvation */ - spin_lock_bh(&conn->taskqueuelock); if (!list_empty(&conn->mgmtqueue)) goto check_mgmt; } @@ -1566,21 +1600,17 @@ check_mgmt: task = list_entry(conn->requeue.next, struct iscsi_task, running); + if (iscsi_check_tmf_restrictions(task, ISCSI_OP_SCSI_DATA_OUT)) break; - conn->task = task; - list_del_init(&conn->task->running); - conn->task->state = ISCSI_TASK_RUNNING; - spin_unlock_bh(&conn->taskqueuelock); - rc = iscsi_xmit_task(conn); + list_del_init(&task->running); + rc = iscsi_xmit_task(conn, task, true); if (rc) goto done; - spin_lock_bh(&conn->taskqueuelock); if (!list_empty(&conn->mgmtqueue)) goto check_mgmt; } - spin_unlock_bh(&conn->taskqueuelock); spin_unlock_bh(&conn->session->frwd_lock); return -ENODATA; @@ -1746,9 +1776,7 @@ int iscsi_queuecommand(struct Scsi_Host *host, struct scsi_cmnd *sc) goto prepd_reject; } } else { - spin_lock_bh(&conn->taskqueuelock); list_add_tail(&task->running, &conn->cmdqueue); - spin_unlock_bh(&conn->taskqueuelock); iscsi_conn_queue_work(conn); } @@ -1855,27 +1883,39 @@ static int iscsi_exec_task_mgmt_fn(struct iscsi_conn *conn, } /* - * Fail commands. session lock held and recv side suspended and xmit - * thread flushed + * Fail commands. session frwd lock held and xmit thread flushed. */ static void fail_scsi_tasks(struct iscsi_conn *conn, u64 lun, int error) { + struct iscsi_session *session = conn->session; struct iscsi_task *task; int i; - for (i = 0; i < conn->session->cmds_max; i++) { - task = conn->session->cmds[i]; + spin_lock_bh(&session->back_lock); + for (i = 0; i < session->cmds_max; i++) { + task = session->cmds[i]; if (!task->sc || task->state == ISCSI_TASK_FREE) continue; if (lun != -1 && lun != task->sc->device->lun) continue; - ISCSI_DBG_SESSION(conn->session, + __iscsi_get_task(task); + spin_unlock_bh(&session->back_lock); + + ISCSI_DBG_SESSION(session, "failing sc %p itt 0x%x state %d\n", task->sc, task->itt, task->state); fail_scsi_task(task, error); + + spin_unlock_bh(&session->frwd_lock); + iscsi_put_task(task); + spin_lock_bh(&session->frwd_lock); + + spin_lock_bh(&session->back_lock); } + + spin_unlock_bh(&session->back_lock); } /** @@ -1953,6 +1993,7 @@ enum blk_eh_timer_return iscsi_eh_cmd_timed_out(struct scsi_cmnd *sc) ISCSI_DBG_EH(session, "scsi cmd %p timedout\n", sc); spin_lock_bh(&session->frwd_lock); + spin_lock(&session->back_lock); task = (struct iscsi_task *)sc->SCp.ptr; if (!task) { /* @@ -1960,8 +2001,11 @@ enum blk_eh_timer_return iscsi_eh_cmd_timed_out(struct scsi_cmnd *sc) * so let timeout code complete it now. */ rc = BLK_EH_DONE; + spin_unlock(&session->back_lock); goto done; } + __iscsi_get_task(task); + spin_unlock(&session->back_lock); if (session->state != ISCSI_STATE_LOGGED_IN) { /* @@ -2020,6 +2064,7 @@ enum blk_eh_timer_return iscsi_eh_cmd_timed_out(struct scsi_cmnd *sc) goto done; } + spin_lock(&session->back_lock); for (i = 0; i < conn->session->cmds_max; i++) { running_task = conn->session->cmds[i]; if (!running_task->sc || running_task == task || @@ -2052,10 +2097,12 @@ enum blk_eh_timer_return iscsi_eh_cmd_timed_out(struct scsi_cmnd *sc) "last xfer %lu/%lu. Last check %lu.\n", task->last_xfer, running_task->last_xfer, task->last_timeout); + spin_unlock(&session->back_lock); rc = BLK_EH_RESET_TIMER; goto done; } } + spin_unlock(&session->back_lock); /* Assumes nop timeout is shorter than scsi cmd timeout */ if (task->have_checked_conn) @@ -2077,9 +2124,12 @@ enum blk_eh_timer_return iscsi_eh_cmd_timed_out(struct scsi_cmnd *sc) rc = BLK_EH_RESET_TIMER; done: - if (task) - task->last_timeout = jiffies; spin_unlock_bh(&session->frwd_lock); + + if (task) { + task->last_timeout = jiffies; + iscsi_put_task(task); + } ISCSI_DBG_EH(session, "return %s\n", rc == BLK_EH_RESET_TIMER ? "timer reset" : "shutdown or nh"); return rc; @@ -2187,15 +2237,20 @@ int iscsi_eh_abort(struct scsi_cmnd *sc) conn->eh_abort_cnt++; age = session->age; + spin_lock(&session->back_lock); task = (struct iscsi_task *)sc->SCp.ptr; - ISCSI_DBG_EH(session, "aborting [sc %p itt 0x%x]\n", - sc, task->itt); - - /* task completed before time out */ - if (!task->sc) { + if (!task || !task->sc) { + /* task completed before time out */ ISCSI_DBG_EH(session, "sc completed while abort in progress\n"); - goto success; + + spin_unlock(&session->back_lock); + spin_unlock_bh(&session->frwd_lock); + mutex_unlock(&session->eh_mutex); + return SUCCESS; } + ISCSI_DBG_EH(session, "aborting [sc %p itt 0x%x]\n", sc, task->itt); + __iscsi_get_task(task); + spin_unlock(&session->back_lock); if (task->state == ISCSI_TASK_PENDING) { fail_scsi_task(task, DID_ABORT); @@ -2257,6 +2312,7 @@ success: success_unlocked: ISCSI_DBG_EH(session, "abort success [sc %p itt 0x%x]\n", sc, task->itt); + iscsi_put_task(task); mutex_unlock(&session->eh_mutex); return SUCCESS; @@ -2265,6 +2321,7 @@ failed: failed_unlocked: ISCSI_DBG_EH(session, "abort failed [sc %p itt 0x%x]\n", sc, task ? task->itt : 0); + iscsi_put_task(task); mutex_unlock(&session->eh_mutex); return FAILED; } @@ -2591,6 +2648,56 @@ void iscsi_pool_free(struct iscsi_pool *q) } EXPORT_SYMBOL_GPL(iscsi_pool_free); +int iscsi_host_get_max_scsi_cmds(struct Scsi_Host *shost, + uint16_t requested_cmds_max) +{ + int scsi_cmds, total_cmds = requested_cmds_max; + +check: + if (!total_cmds) + total_cmds = ISCSI_DEF_XMIT_CMDS_MAX; + /* + * The iscsi layer needs some tasks for nop handling and tmfs, + * so the cmds_max must at least be greater than ISCSI_MGMT_CMDS_MAX + * + 1 command for scsi IO. + */ + if (total_cmds < ISCSI_TOTAL_CMDS_MIN) { + printk(KERN_ERR "iscsi: invalid max cmds of %d. Must be a power of two that is at least %d.\n", + total_cmds, ISCSI_TOTAL_CMDS_MIN); + return -EINVAL; + } + + if (total_cmds > ISCSI_TOTAL_CMDS_MAX) { + printk(KERN_INFO "iscsi: invalid max cmds of %d. Must be a power of 2 less than or equal to %d. Using %d.\n", + requested_cmds_max, ISCSI_TOTAL_CMDS_MAX, + ISCSI_TOTAL_CMDS_MAX); + total_cmds = ISCSI_TOTAL_CMDS_MAX; + } + + if (!is_power_of_2(total_cmds)) { + total_cmds = rounddown_pow_of_two(total_cmds); + if (total_cmds < ISCSI_TOTAL_CMDS_MIN) { + printk(KERN_ERR "iscsi: invalid max cmds of %d. Must be a power of 2 greater than %d.\n", requested_cmds_max, ISCSI_TOTAL_CMDS_MIN); + return -EINVAL; + } + + printk(KERN_INFO "iscsi: invalid max cmds %d. Must be a power of 2. Rounding max cmds down to %d.\n", + requested_cmds_max, total_cmds); + } + + scsi_cmds = total_cmds - ISCSI_MGMT_CMDS_MAX; + if (shost->can_queue && scsi_cmds > shost->can_queue) { + total_cmds = shost->can_queue; + + printk(KERN_INFO "iscsi: requested max cmds %u is higher than driver limit. Using driver limit %u\n", + requested_cmds_max, shost->can_queue); + goto check; + } + + return scsi_cmds; +} +EXPORT_SYMBOL_GPL(iscsi_host_get_max_scsi_cmds); + /** * iscsi_host_add - add host to system * @shost: scsi host @@ -2681,8 +2788,6 @@ void iscsi_host_remove(struct Scsi_Host *shost) flush_signals(current); scsi_remove_host(shost); - if (ihost->workq) - destroy_workqueue(ihost->workq); } EXPORT_SYMBOL_GPL(iscsi_host_remove); @@ -2690,6 +2795,9 @@ void iscsi_host_free(struct Scsi_Host *shost) { struct iscsi_host *ihost = shost_priv(shost); + if (ihost->workq) + destroy_workqueue(ihost->workq); + kfree(ihost->netdev); kfree(ihost->hwaddress); kfree(ihost->initiatorname); @@ -2743,7 +2851,7 @@ iscsi_session_setup(struct iscsi_transport *iscsit, struct Scsi_Host *shost, struct iscsi_host *ihost = shost_priv(shost); struct iscsi_session *session; struct iscsi_cls_session *cls_session; - int cmd_i, scsi_cmds, total_cmds = cmds_max; + int cmd_i, scsi_cmds; unsigned long flags; spin_lock_irqsave(&ihost->lock, flags); @@ -2754,37 +2862,9 @@ iscsi_session_setup(struct iscsi_transport *iscsit, struct Scsi_Host *shost, ihost->num_sessions++; spin_unlock_irqrestore(&ihost->lock, flags); - if (!total_cmds) - total_cmds = ISCSI_DEF_XMIT_CMDS_MAX; - /* - * The iscsi layer needs some tasks for nop handling and tmfs, - * so the cmds_max must at least be greater than ISCSI_MGMT_CMDS_MAX - * + 1 command for scsi IO. - */ - if (total_cmds < ISCSI_TOTAL_CMDS_MIN) { - printk(KERN_ERR "iscsi: invalid can_queue of %d. can_queue " - "must be a power of two that is at least %d.\n", - total_cmds, ISCSI_TOTAL_CMDS_MIN); + scsi_cmds = iscsi_host_get_max_scsi_cmds(shost, cmds_max); + if (scsi_cmds < 0) goto dec_session_count; - } - - if (total_cmds > ISCSI_TOTAL_CMDS_MAX) { - printk(KERN_ERR "iscsi: invalid can_queue of %d. can_queue " - "must be a power of 2 less than or equal to %d.\n", - cmds_max, ISCSI_TOTAL_CMDS_MAX); - total_cmds = ISCSI_TOTAL_CMDS_MAX; - } - - if (!is_power_of_2(total_cmds)) { - printk(KERN_ERR "iscsi: invalid can_queue of %d. can_queue " - "must be a power of 2.\n", total_cmds); - total_cmds = rounddown_pow_of_two(total_cmds); - if (total_cmds < ISCSI_TOTAL_CMDS_MIN) - goto dec_session_count; - printk(KERN_INFO "iscsi: Rounding can_queue to %d.\n", - total_cmds); - } - scsi_cmds = total_cmds - ISCSI_MGMT_CMDS_MAX; cls_session = iscsi_alloc_session(shost, iscsit, sizeof(struct iscsi_session) + @@ -2800,7 +2880,7 @@ iscsi_session_setup(struct iscsi_transport *iscsit, struct Scsi_Host *shost, session->lu_reset_timeout = 15; session->abort_timeout = 10; session->scsi_cmds_max = scsi_cmds; - session->cmds_max = total_cmds; + session->cmds_max = scsi_cmds + ISCSI_MGMT_CMDS_MAX; session->queued_cmdsn = session->cmdsn = initial_cmdsn; session->exp_cmdsn = initial_cmdsn + 1; session->max_cmdsn = initial_cmdsn + 1; @@ -2919,7 +2999,6 @@ iscsi_conn_setup(struct iscsi_cls_session *cls_session, int dd_size, INIT_LIST_HEAD(&conn->mgmtqueue); INIT_LIST_HEAD(&conn->cmdqueue); INIT_LIST_HEAD(&conn->requeue); - spin_lock_init(&conn->taskqueuelock); INIT_WORK(&conn->xmitwork, iscsi_xmitworker); /* allocate login_task used for the login/text sequences */ @@ -3085,10 +3164,16 @@ fail_mgmt_tasks(struct iscsi_session *session, struct iscsi_conn *conn) ISCSI_DBG_SESSION(conn->session, "failing mgmt itt 0x%x state %d\n", task->itt, task->state); + + spin_lock_bh(&session->back_lock); + if (cleanup_queued_task(task)) { + spin_unlock_bh(&session->back_lock); + continue; + } + state = ISCSI_TASK_ABRT_SESS_RECOV; if (task->state == ISCSI_TASK_PENDING) state = ISCSI_TASK_COMPLETED; - spin_lock_bh(&session->back_lock); iscsi_complete_task(task, state); spin_unlock_bh(&session->back_lock); } @@ -3189,6 +3274,13 @@ int iscsi_conn_bind(struct iscsi_cls_session *cls_session, spin_unlock_bh(&session->frwd_lock); /* + * The target could have reduced it's window size between logins, so + * we have to reset max/exp cmdsn so we can see the new values. + */ + spin_lock_bh(&session->back_lock); + session->max_cmdsn = session->exp_cmdsn = session->cmdsn + 1; + spin_unlock_bh(&session->back_lock); + /* * Unblock xmitworker(), Login Phase will pass through. */ clear_bit(ISCSI_SUSPEND_BIT, &conn->suspend_rx); diff --git a/drivers/scsi/libiscsi_tcp.c b/drivers/scsi/libiscsi_tcp.c index 83f14b2c8804..2e9ffe3d1a55 100644 --- a/drivers/scsi/libiscsi_tcp.c +++ b/drivers/scsi/libiscsi_tcp.c @@ -524,48 +524,79 @@ static int iscsi_tcp_data_in(struct iscsi_conn *conn, struct iscsi_task *task) /** * iscsi_tcp_r2t_rsp - iSCSI R2T Response processing * @conn: iscsi connection - * @task: scsi command task + * @hdr: PDU header */ -static int iscsi_tcp_r2t_rsp(struct iscsi_conn *conn, struct iscsi_task *task) +static int iscsi_tcp_r2t_rsp(struct iscsi_conn *conn, struct iscsi_hdr *hdr) { struct iscsi_session *session = conn->session; - struct iscsi_tcp_task *tcp_task = task->dd_data; - struct iscsi_tcp_conn *tcp_conn = conn->dd_data; - struct iscsi_r2t_rsp *rhdr = (struct iscsi_r2t_rsp *)tcp_conn->in.hdr; + struct iscsi_tcp_task *tcp_task; + struct iscsi_tcp_conn *tcp_conn; + struct iscsi_r2t_rsp *rhdr; struct iscsi_r2t_info *r2t; - int r2tsn = be32_to_cpu(rhdr->r2tsn); + struct iscsi_task *task; u32 data_length; u32 data_offset; + int r2tsn; int rc; + spin_lock(&session->back_lock); + task = iscsi_itt_to_ctask(conn, hdr->itt); + if (!task) { + spin_unlock(&session->back_lock); + return ISCSI_ERR_BAD_ITT; + } else if (task->sc->sc_data_direction != DMA_TO_DEVICE) { + spin_unlock(&session->back_lock); + return ISCSI_ERR_PROTO; + } + /* + * A bad target might complete the cmd before we have handled R2Ts + * so get a ref to the task that will be dropped in the xmit path. + */ + if (task->state != ISCSI_TASK_RUNNING) { + spin_unlock(&session->back_lock); + /* Let the path that got the early rsp complete it */ + return 0; + } + task->last_xfer = jiffies; + __iscsi_get_task(task); + + tcp_conn = conn->dd_data; + rhdr = (struct iscsi_r2t_rsp *)tcp_conn->in.hdr; + /* fill-in new R2T associated with the task */ + iscsi_update_cmdsn(session, (struct iscsi_nopin *)rhdr); + spin_unlock(&session->back_lock); + if (tcp_conn->in.datalen) { iscsi_conn_printk(KERN_ERR, conn, "invalid R2t with datalen %d\n", tcp_conn->in.datalen); - return ISCSI_ERR_DATALEN; + rc = ISCSI_ERR_DATALEN; + goto put_task; } + tcp_task = task->dd_data; + r2tsn = be32_to_cpu(rhdr->r2tsn); if (tcp_task->exp_datasn != r2tsn){ ISCSI_DBG_TCP(conn, "task->exp_datasn(%d) != rhdr->r2tsn(%d)\n", tcp_task->exp_datasn, r2tsn); - return ISCSI_ERR_R2TSN; + rc = ISCSI_ERR_R2TSN; + goto put_task; } - /* fill-in new R2T associated with the task */ - iscsi_update_cmdsn(session, (struct iscsi_nopin*)rhdr); - - if (!task->sc || session->state != ISCSI_STATE_LOGGED_IN) { + if (session->state != ISCSI_STATE_LOGGED_IN) { iscsi_conn_printk(KERN_INFO, conn, "dropping R2T itt %d in recovery.\n", task->itt); - return 0; + rc = 0; + goto put_task; } data_length = be32_to_cpu(rhdr->data_length); if (data_length == 0) { iscsi_conn_printk(KERN_ERR, conn, "invalid R2T with zero data len\n"); - return ISCSI_ERR_DATALEN; + rc = ISCSI_ERR_DATALEN; + goto put_task; } if (data_length > session->max_burst) @@ -579,7 +610,8 @@ static int iscsi_tcp_r2t_rsp(struct iscsi_conn *conn, struct iscsi_task *task) "invalid R2T with data len %u at offset %u " "and total length %d\n", data_length, data_offset, task->sc->sdb.length); - return ISCSI_ERR_DATALEN; + rc = ISCSI_ERR_DATALEN; + goto put_task; } spin_lock(&tcp_task->pool2queue); @@ -589,7 +621,8 @@ static int iscsi_tcp_r2t_rsp(struct iscsi_conn *conn, struct iscsi_task *task) "Target has sent more R2Ts than it " "negotiated for or driver has leaked.\n"); spin_unlock(&tcp_task->pool2queue); - return ISCSI_ERR_PROTO; + rc = ISCSI_ERR_PROTO; + goto put_task; } r2t->exp_statsn = rhdr->statsn; @@ -607,6 +640,10 @@ static int iscsi_tcp_r2t_rsp(struct iscsi_conn *conn, struct iscsi_task *task) iscsi_requeue_task(task); return 0; + +put_task: + iscsi_put_task(task); + return rc; } /* @@ -730,20 +767,11 @@ iscsi_tcp_hdr_dissect(struct iscsi_conn *conn, struct iscsi_hdr *hdr) rc = iscsi_complete_pdu(conn, hdr, NULL, 0); break; case ISCSI_OP_R2T: - spin_lock(&conn->session->back_lock); - task = iscsi_itt_to_ctask(conn, hdr->itt); - spin_unlock(&conn->session->back_lock); - if (!task) - rc = ISCSI_ERR_BAD_ITT; - else if (ahslen) + if (ahslen) { rc = ISCSI_ERR_AHSLEN; - else if (task->sc->sc_data_direction == DMA_TO_DEVICE) { - task->last_xfer = jiffies; - spin_lock(&conn->session->frwd_lock); - rc = iscsi_tcp_r2t_rsp(conn, task); - spin_unlock(&conn->session->frwd_lock); - } else - rc = ISCSI_ERR_PROTO; + break; + } + rc = iscsi_tcp_r2t_rsp(conn, hdr); break; case ISCSI_OP_LOGIN_RSP: case ISCSI_OP_TEXT_RSP: diff --git a/drivers/scsi/mpt3sas/mpt3sas_base.c b/drivers/scsi/mpt3sas/mpt3sas_base.c index f5582c8e77c9..ac066f86bb14 100644 --- a/drivers/scsi/mpt3sas/mpt3sas_base.c +++ b/drivers/scsi/mpt3sas/mpt3sas_base.c @@ -3648,25 +3648,16 @@ _base_get_msix_index(struct MPT3SAS_ADAPTER *ioc, base_mod64(atomic64_add_return(1, &ioc->total_io_cnt), ioc->reply_queue_count) : 0; - return ioc->cpu_msix_table[raw_smp_processor_id()]; -} + if (scmd && ioc->shost->nr_hw_queues > 1) { + u32 tag = blk_mq_unique_tag(scmd->request); -/** - * _base_sdev_nr_inflight_request -get number of inflight requests - * of a request queue. - * @q: request_queue object - * - * returns number of inflight request of a request queue. - */ -inline unsigned long -_base_sdev_nr_inflight_request(struct request_queue *q) -{ - struct blk_mq_hw_ctx *hctx = q->queue_hw_ctx[0]; + return blk_mq_unique_tag_to_hwq(tag) + + ioc->high_iops_queues; + } - return atomic_read(&hctx->nr_active); + return ioc->cpu_msix_table[raw_smp_processor_id()]; } - /** * _base_get_high_iops_msix_index - get the msix index of * high iops queues @@ -3686,7 +3677,8 @@ _base_get_high_iops_msix_index(struct MPT3SAS_ADAPTER *ioc, * reply queues in terms of batch count 16 when outstanding * IOs on the target device is >=8. */ - if (_base_sdev_nr_inflight_request(scmd->device->request_queue) > + + if (atomic_read(&scmd->device->device_busy) > MPT3SAS_DEVICE_HIGH_IOPS_DEPTH) return base_mod64(( atomic64_add_return(1, &ioc->high_iops_outstanding) / @@ -3739,8 +3731,23 @@ mpt3sas_base_get_smid_scsiio(struct MPT3SAS_ADAPTER *ioc, u8 cb_idx, struct scsi_cmnd *scmd) { struct scsiio_tracker *request = scsi_cmd_priv(scmd); - unsigned int tag = scmd->request->tag; u16 smid; + u32 tag, unique_tag; + + unique_tag = blk_mq_unique_tag(scmd->request); + tag = blk_mq_unique_tag_to_tag(unique_tag); + + /* + * Store hw queue number corresponding to the tag. + * This hw queue number is used later to determine + * the unique_tag using the logic below. This unique_tag + * is used to retrieve the scmd pointer corresponding + * to tag using scsi_host_find_tag() API. + * + * tag = smid - 1; + * unique_tag = ioc->io_queue_num[tag] << BLK_MQ_UNIQUE_TAG_BITS | tag; + */ + ioc->io_queue_num[tag] = blk_mq_unique_tag_to_hwq(unique_tag); smid = tag + 1; request->cb_idx = cb_idx; @@ -3831,6 +3838,7 @@ mpt3sas_base_free_smid(struct MPT3SAS_ADAPTER *ioc, u16 smid) mpt3sas_base_clear_st(ioc, st); _base_recovery_check(ioc); + ioc->io_queue_num[smid - 1] = 0; return; } @@ -5362,6 +5370,9 @@ _base_release_memory_pools(struct MPT3SAS_ADAPTER *ioc) kfree(ioc->chain_lookup); ioc->chain_lookup = NULL; } + + kfree(ioc->io_queue_num); + ioc->io_queue_num = NULL; } /** @@ -5641,7 +5652,8 @@ _base_allocate_memory_pools(struct MPT3SAS_ADAPTER *ioc) reply_post_free_sz = ioc->reply_post_queue_depth * sizeof(Mpi2DefaultReplyDescriptor_t); rdpq_sz = reply_post_free_sz * RDPQ_MAX_INDEX_IN_ONE_CHUNK; - if (_base_is_controller_msix_enabled(ioc) && !ioc->rdpq_array_enable) + if ((_base_is_controller_msix_enabled(ioc) && !ioc->rdpq_array_enable) + || (ioc->reply_queue_count < RDPQ_MAX_INDEX_IN_ONE_CHUNK)) rdpq_sz = reply_post_free_sz * ioc->reply_queue_count; ret = base_alloc_rdpq_dma_pool(ioc, rdpq_sz); if (ret == -EAGAIN) { @@ -5772,6 +5784,11 @@ _base_allocate_memory_pools(struct MPT3SAS_ADAPTER *ioc) ioc_info(ioc, "internal(0x%p): depth(%d), start smid(%d)\n", ioc->internal, ioc->internal_depth, ioc->internal_smid)); + + ioc->io_queue_num = kcalloc(ioc->scsiio_depth, + sizeof(u16), GFP_KERNEL); + if (!ioc->io_queue_num) + goto out; /* * The number of NVMe page sized blocks needed is: * (((sg_tablesize * 8) - 1) / (page_size - 8)) + 1 @@ -8174,8 +8191,11 @@ mpt3sas_base_hard_reset_handler(struct MPT3SAS_ADAPTER *ioc, ioc_state = mpt3sas_base_get_iocstate(ioc, 0); if ((ioc_state & MPI2_IOC_STATE_MASK) == MPI2_IOC_STATE_FAULT || (ioc_state & MPI2_IOC_STATE_MASK) == - MPI2_IOC_STATE_COREDUMP) + MPI2_IOC_STATE_COREDUMP) { is_fault = 1; + ioc->htb_rel.trigger_info_dwords[1] = + (ioc_state & MPI2_DOORBELL_DATA_MASK); + } } _base_pre_reset_handler(ioc); mpt3sas_wait_for_commands_to_complete(ioc); diff --git a/drivers/scsi/mpt3sas/mpt3sas_base.h b/drivers/scsi/mpt3sas/mpt3sas_base.h index 2def7a340616..315aee6ef86f 100644 --- a/drivers/scsi/mpt3sas/mpt3sas_base.h +++ b/drivers/scsi/mpt3sas/mpt3sas_base.h @@ -77,8 +77,8 @@ #define MPT3SAS_DRIVER_NAME "mpt3sas" #define MPT3SAS_AUTHOR "Avago Technologies <MPT-FusionLinux.pdl@avagotech.com>" #define MPT3SAS_DESCRIPTION "LSI MPT Fusion SAS 3.0 Device Driver" -#define MPT3SAS_DRIVER_VERSION "36.100.00.00" -#define MPT3SAS_MAJOR_VERSION 36 +#define MPT3SAS_DRIVER_VERSION "37.100.00.00" +#define MPT3SAS_MAJOR_VERSION 37 #define MPT3SAS_MINOR_VERSION 100 #define MPT3SAS_BUILD_VERSION 0 #define MPT3SAS_RELEASE_VERSION 00 @@ -1073,6 +1073,50 @@ struct hba_port { #define MULTIPATH_DISABLED_PORT_ID 0xFF +/** + * struct htb_rel_query - diagnostic buffer release reason + * @unique_id - unique id associated with this buffer. + * @buffer_rel_condition - Release condition ioctl/sysfs/reset + * @reserved + * @trigger_type - Master/Event/scsi/MPI + * @trigger_info_dwords - Data Correspondig to trigger type + */ +struct htb_rel_query { + u16 buffer_rel_condition; + u16 reserved; + u32 trigger_type; + u32 trigger_info_dwords[2]; +}; + +/* Buffer_rel_condition bit fields */ + +/* Bit 0 - Diag Buffer not Released */ +#define MPT3_DIAG_BUFFER_NOT_RELEASED (0x00) +/* Bit 0 - Diag Buffer Released */ +#define MPT3_DIAG_BUFFER_RELEASED (0x01) + +/* + * Bit 1 - Diag Buffer Released by IOCTL, + * This bit is valid only if Bit 0 is one + */ +#define MPT3_DIAG_BUFFER_REL_IOCTL (0x02 | MPT3_DIAG_BUFFER_RELEASED) + +/* + * Bit 2 - Diag Buffer Released by Trigger, + * This bit is valid only if Bit 0 is one + */ +#define MPT3_DIAG_BUFFER_REL_TRIGGER (0x04 | MPT3_DIAG_BUFFER_RELEASED) + +/* + * Bit 3 - Diag Buffer Released by SysFs, + * This bit is valid only if Bit 0 is one + */ +#define MPT3_DIAG_BUFFER_REL_SYSFS (0x08 | MPT3_DIAG_BUFFER_RELEASED) + +/* DIAG RESET Master trigger flags */ +#define MPT_DIAG_RESET_ISSUED_BY_DRIVER 0x00000000 +#define MPT_DIAG_RESET_ISSUED_BY_USER 0x00000001 + typedef void (*MPT3SAS_FLUSH_RUNNING_CMDS)(struct MPT3SAS_ADAPTER *ioc); /** * struct MPT3SAS_ADAPTER - per adapter struct @@ -1439,6 +1483,7 @@ struct MPT3SAS_ADAPTER { spinlock_t scsi_lookup_lock; int pending_io_count; wait_queue_head_t reset_wq; + u16 *io_queue_num; /* PCIe SGL */ struct dma_pool *pcie_sgl_dma_pool; @@ -1529,6 +1574,8 @@ struct MPT3SAS_ADAPTER { u32 diagnostic_flags[MPI2_DIAG_BUF_TYPE_COUNT]; u32 ring_buffer_offset; u32 ring_buffer_sz; + struct htb_rel_query htb_rel; + u8 reset_from_user; u8 is_warpdrive; u8 is_mcpu_endpoint; u8 hide_ir_msg; @@ -1565,6 +1612,7 @@ struct mpt3sas_debugfs_buffer { }; #define MPT_DRV_SUPPORT_BITMAP_MEMMOVE 0x00000001 +#define MPT_DRV_SUPPORT_BITMAP_ADDNLQUERY 0x00000002 typedef u8 (*MPT_CALLBACK)(struct MPT3SAS_ADAPTER *ioc, u16 smid, u8 msix_index, u32 reply); diff --git a/drivers/scsi/mpt3sas/mpt3sas_ctl.c b/drivers/scsi/mpt3sas/mpt3sas_ctl.c index c8a0ce18f2c5..44f9a05db94e 100644 --- a/drivers/scsi/mpt3sas/mpt3sas_ctl.c +++ b/drivers/scsi/mpt3sas/mpt3sas_ctl.c @@ -479,6 +479,8 @@ void mpt3sas_ctl_pre_reset_handler(struct MPT3SAS_ADAPTER *ioc) ioc_info(ioc, "%s: Releasing the trace buffer due to adapter reset.", __func__); + ioc->htb_rel.buffer_rel_condition = + MPT3_DIAG_BUFFER_REL_TRIGGER; mpt3sas_send_diag_release(ioc, i, &issue_reset); } } @@ -1334,6 +1336,7 @@ _ctl_do_reset(struct MPT3SAS_ADAPTER *ioc, void __user *arg) dctlprintk(ioc, ioc_info(ioc, "%s: enter\n", __func__)); + ioc->reset_from_user = 1; retval = mpt3sas_base_hard_reset_handler(ioc, FORCE_BIG_HAMMER); ioc_info(ioc, "Ioctl: host reset: %s\n", ((!retval) ? "SUCCESS" : "FAILED")); @@ -1687,6 +1690,9 @@ _ctl_diag_register_2(struct MPT3SAS_ADAPTER *ioc, request_data = ioc->diag_buffer[buffer_type]; request_data_sz = diag_register->requested_buffer_size; ioc->unique_id[buffer_type] = diag_register->unique_id; + /* Reset ioc variables used for additional query commands */ + ioc->reset_from_user = 0; + memset(&ioc->htb_rel, 0, sizeof(struct htb_rel_query)); ioc->diag_buffer_status[buffer_type] &= MPT3_DIAG_BUFFER_IS_DRIVER_ALLOCATED; memcpy(ioc->product_specific[buffer_type], @@ -2469,7 +2475,61 @@ _ctl_diag_read_buffer(struct MPT3SAS_ADAPTER *ioc, void __user *arg) return rc; } +/** + * _ctl_addnl_diag_query - query relevant info associated with diag buffers + * @ioc: per adapter object + * @arg: user space buffer containing ioctl content + * + * The application will send only unique_id. Driver will + * inspect unique_id first, if valid, fill the details related to cause + * for diag buffer release. + */ +static long +_ctl_addnl_diag_query(struct MPT3SAS_ADAPTER *ioc, void __user *arg) +{ + struct mpt3_addnl_diag_query karg; + u32 buffer_type = 0; + if (copy_from_user(&karg, arg, sizeof(karg))) { + pr_err("%s: failure at %s:%d/%s()!\n", + ioc->name, __FILE__, __LINE__, __func__); + return -EFAULT; + } + dctlprintk(ioc, ioc_info(ioc, "%s\n", __func__)); + if (karg.unique_id == 0) { + ioc_err(ioc, "%s: unique_id is(0x%08x)\n", + __func__, karg.unique_id); + return -EPERM; + } + buffer_type = _ctl_diag_get_bufftype(ioc, karg.unique_id); + if (buffer_type == MPT3_DIAG_UID_NOT_FOUND) { + ioc_err(ioc, "%s: buffer with unique_id(0x%08x) not found\n", + __func__, karg.unique_id); + return -EPERM; + } + memset(&karg.buffer_rel_condition, 0, sizeof(struct htb_rel_query)); + if ((ioc->diag_buffer_status[buffer_type] & + MPT3_DIAG_BUFFER_IS_REGISTERED) == 0) { + ioc_info(ioc, "%s: buffer_type(0x%02x) is not registered\n", + __func__, buffer_type); + goto out; + } + if ((ioc->diag_buffer_status[buffer_type] & + MPT3_DIAG_BUFFER_IS_RELEASED) == 0) { + ioc_err(ioc, "%s: buffer_type(0x%02x) is not released\n", + __func__, buffer_type); + return -EPERM; + } + memcpy(&karg.buffer_rel_condition, &ioc->htb_rel, + sizeof(struct htb_rel_query)); +out: + if (copy_to_user(arg, &karg, sizeof(struct mpt3_addnl_diag_query))) { + ioc_err(ioc, "%s: unable to write mpt3_addnl_diag_query data @ %p\n", + __func__, arg); + return -EFAULT; + } + return 0; +} #ifdef CONFIG_COMPAT /** @@ -2533,7 +2593,7 @@ _ctl_ioctl_main(struct file *file, unsigned int cmd, void __user *arg, struct MPT3SAS_ADAPTER *ioc; struct mpt3_ioctl_header ioctl_header; enum block_state state; - long ret = -EINVAL; + long ret = -ENOIOCTLCMD; /* get IOCTL header */ if (copy_from_user(&ioctl_header, (char __user *)arg, @@ -2643,6 +2703,10 @@ _ctl_ioctl_main(struct file *file, unsigned int cmd, void __user *arg, if (_IOC_SIZE(cmd) == sizeof(struct mpt3_diag_read_buffer)) ret = _ctl_diag_read_buffer(ioc, arg); break; + case MPT3ADDNLDIAGQUERY: + if (_IOC_SIZE(cmd) == sizeof(struct mpt3_addnl_diag_query)) + ret = _ctl_addnl_diag_query(ioc, arg); + break; default: dctlprintk(ioc, ioc_info(ioc, "unsupported ioctl opcode(0x%08x)\n", @@ -3425,6 +3489,7 @@ host_trace_buffer_enable_store(struct device *cdev, MPT3_DIAG_BUFFER_IS_RELEASED)) goto out; ioc_info(ioc, "releasing host trace buffer\n"); + ioc->htb_rel.buffer_rel_condition = MPT3_DIAG_BUFFER_REL_SYSFS; mpt3sas_send_diag_release(ioc, MPI2_DIAG_BUF_TYPE_TRACE, &issue_reset); } diff --git a/drivers/scsi/mpt3sas/mpt3sas_ctl.h b/drivers/scsi/mpt3sas/mpt3sas_ctl.h index 0f7aa4ddade0..d2ccdafb8df2 100644 --- a/drivers/scsi/mpt3sas/mpt3sas_ctl.h +++ b/drivers/scsi/mpt3sas/mpt3sas_ctl.h @@ -94,6 +94,8 @@ struct mpt3_diag_query) #define MPT3DIAGREADBUFFER _IOWR(MPT3_MAGIC_NUMBER, 30, \ struct mpt3_diag_read_buffer) +#define MPT3ADDNLDIAGQUERY _IOWR(MPT3_MAGIC_NUMBER, 32, \ + struct mpt3_addnl_diag_query) /* Trace Buffer default UniqueId */ #define MPT2DIAGBUFFUNIQUEID (0x07075900) @@ -430,4 +432,24 @@ struct mpt3_diag_read_buffer { uint32_t diagnostic_data[1]; }; +/** + * struct mpt3_addnl_diag_query - diagnostic buffer release reason + * @hdr - generic header + * @unique_id - unique id associated with this buffer. + * @buffer_rel_condition - Release condition ioctl/sysfs/reset + * @reserved1 + * @trigger_type - Master/Event/scsi/MPI + * @trigger_info_dwords - Data Correspondig to trigger type + * @reserved2 + */ +struct mpt3_addnl_diag_query { + struct mpt3_ioctl_header hdr; + uint32_t unique_id; + uint16_t buffer_rel_condition; + uint16_t reserved1; + uint32_t trigger_type; + uint32_t trigger_info_dwords[2]; + uint32_t reserved2[2]; +}; + #endif /* MPT3SAS_CTL_H_INCLUDED */ diff --git a/drivers/scsi/mpt3sas/mpt3sas_scsih.c b/drivers/scsi/mpt3sas/mpt3sas_scsih.c index c8b09a81834d..ffca03064797 100644 --- a/drivers/scsi/mpt3sas/mpt3sas_scsih.c +++ b/drivers/scsi/mpt3sas/mpt3sas_scsih.c @@ -54,6 +54,7 @@ #include <linux/interrupt.h> #include <linux/aer.h> #include <linux/raid_class.h> +#include <linux/blk-mq-pci.h> #include <asm/unaligned.h> #include "mpt3sas_base.h" @@ -168,6 +169,11 @@ MODULE_PARM_DESC(multipath_on_hba, "\t SAS 2.0 & SAS 3.0 HBA - This will be disabled,\n\t\t" "\t SAS 3.5 HBA - This will be enabled)"); +static int host_tagset_enable = 1; +module_param(host_tagset_enable, int, 0444); +MODULE_PARM_DESC(host_tagset_enable, + "Shared host tagset enable/disable Default: enable(1)"); + /* raid transport support */ static struct raid_template *mpt3sas_raid_template; static struct raid_template *mpt2sas_raid_template; @@ -1743,10 +1749,12 @@ mpt3sas_scsih_scsi_lookup_get(struct MPT3SAS_ADAPTER *ioc, u16 smid) struct scsi_cmnd *scmd = NULL; struct scsiio_tracker *st; Mpi25SCSIIORequest_t *mpi_request; + u16 tag = smid - 1; if (smid > 0 && smid <= ioc->scsiio_depth - INTERNAL_SCSIIO_CMDS_COUNT) { - u32 unique_tag = smid - 1; + u32 unique_tag = + ioc->io_queue_num[tag] << BLK_MQ_UNIQUE_TAG_BITS | tag; mpi_request = mpt3sas_base_get_msg_frame(ioc, smid); @@ -11599,6 +11607,22 @@ scsih_scan_finished(struct Scsi_Host *shost, unsigned long time) return 1; } +/** + * scsih_map_queues - map reply queues with request queues + * @shost: SCSI host pointer + */ +static int scsih_map_queues(struct Scsi_Host *shost) +{ + struct MPT3SAS_ADAPTER *ioc = + (struct MPT3SAS_ADAPTER *)shost->hostdata; + + if (ioc->shost->nr_hw_queues == 1) + return 0; + + return blk_mq_pci_map_queues(&shost->tag_set.map[HCTX_TYPE_DEFAULT], + ioc->pdev, ioc->high_iops_queues); +} + /* shost template for SAS 2.0 HBA devices */ static struct scsi_host_template mpt2sas_driver_template = { .module = THIS_MODULE, @@ -11666,6 +11690,7 @@ static struct scsi_host_template mpt3sas_driver_template = { .sdev_attrs = mpt3sas_dev_attrs, .track_queue_depth = 1, .cmd_size = sizeof(struct scsiio_tracker), + .map_queues = scsih_map_queues, }; /* raid transport support for SAS 3.0 HBA devices */ @@ -11922,6 +11947,8 @@ _scsih_probe(struct pci_dev *pdev, const struct pci_device_id *id) * Enable MEMORY MOVE support flag. */ ioc->drv_support_bitmap |= MPT_DRV_SUPPORT_BITMAP_MEMMOVE; + /* Enable ADDITIONAL QUERY support flag. */ + ioc->drv_support_bitmap |= MPT_DRV_SUPPORT_BITMAP_ADDNLQUERY; ioc->enable_sdev_max_qd = enable_sdev_max_qd; @@ -12028,6 +12055,21 @@ _scsih_probe(struct pci_dev *pdev, const struct pci_device_id *id) } else ioc->hide_drives = 0; + shost->host_tagset = 0; + shost->nr_hw_queues = 1; + + if (ioc->is_gen35_ioc && ioc->reply_queue_count > 1 && + host_tagset_enable && ioc->smp_affinity_enable) { + + shost->host_tagset = 1; + shost->nr_hw_queues = + ioc->reply_queue_count - ioc->high_iops_queues; + + dev_info(&ioc->pdev->dev, + "Max SCSIIO MPT commands: %d shared with nr_hw_queues = %d\n", + shost->can_queue, shost->nr_hw_queues); + } + rv = scsi_add_host(shost, &pdev->dev); if (rv) { ioc_err(ioc, "failure at %s:%d/%s()!\n", diff --git a/drivers/scsi/mpt3sas/mpt3sas_trigger_diag.c b/drivers/scsi/mpt3sas/mpt3sas_trigger_diag.c index 8ec9bab20ec4..d9b7d0ee25b0 100644 --- a/drivers/scsi/mpt3sas/mpt3sas_trigger_diag.c +++ b/drivers/scsi/mpt3sas/mpt3sas_trigger_diag.c @@ -132,6 +132,35 @@ mpt3sas_process_trigger_data(struct MPT3SAS_ADAPTER *ioc, &issue_reset); } + ioc->htb_rel.buffer_rel_condition = MPT3_DIAG_BUFFER_REL_TRIGGER; + if (event_data) { + ioc->htb_rel.trigger_type = event_data->trigger_type; + switch (event_data->trigger_type) { + case MPT3SAS_TRIGGER_SCSI: + memcpy(&ioc->htb_rel.trigger_info_dwords, + &event_data->u.scsi, + sizeof(struct SL_WH_SCSI_TRIGGER_T)); + break; + case MPT3SAS_TRIGGER_MPI: + memcpy(&ioc->htb_rel.trigger_info_dwords, + &event_data->u.mpi, + sizeof(struct SL_WH_MPI_TRIGGER_T)); + break; + case MPT3SAS_TRIGGER_MASTER: + ioc->htb_rel.trigger_info_dwords[0] = + event_data->u.master.MasterData; + break; + case MPT3SAS_TRIGGER_EVENT: + memcpy(&ioc->htb_rel.trigger_info_dwords, + &event_data->u.event, + sizeof(struct SL_WH_EVENT_TRIGGER_T)); + break; + default: + ioc_err(ioc, "%d - Is not a valid Trigger type\n", + event_data->trigger_type); + break; + } + } _mpt3sas_raise_sigio(ioc, event_data); dTriggerDiagPrintk(ioc, ioc_info(ioc, "%s: exit\n", @@ -201,9 +230,14 @@ mpt3sas_trigger_master(struct MPT3SAS_ADAPTER *ioc, u32 trigger_bitmask) event_data.u.master.MasterData = trigger_bitmask; if (trigger_bitmask & MASTER_TRIGGER_FW_FAULT || - trigger_bitmask & MASTER_TRIGGER_ADAPTER_RESET) + trigger_bitmask & MASTER_TRIGGER_ADAPTER_RESET) { + ioc->htb_rel.trigger_type = MPT3SAS_TRIGGER_MASTER; + ioc->htb_rel.trigger_info_dwords[0] = trigger_bitmask; + if (ioc->reset_from_user) + ioc->htb_rel.trigger_info_dwords[1] = + MPT_DIAG_RESET_ISSUED_BY_USER; _mpt3sas_raise_sigio(ioc, &event_data); - else + } else mpt3sas_send_trigger_data_event(ioc, &event_data); out: diff --git a/drivers/scsi/pmcraid.h b/drivers/scsi/pmcraid.h index 15c962108075..6d36debde18e 100644 --- a/drivers/scsi/pmcraid.h +++ b/drivers/scsi/pmcraid.h @@ -244,7 +244,7 @@ struct pmcraid_ioarcb { __u8 hrrq_id; __u8 cdb[PMCRAID_MAX_CDB_LEN]; struct pmcraid_ioarcb_add_data add_data; -} __attribute__((packed, aligned(PMCRAID_IOARCB_ALIGNMENT))); +}; /* well known resource handle values */ #define PMCRAID_IOA_RES_HANDLE 0xffffffff @@ -1040,8 +1040,8 @@ struct pmcraid_passthrough_ioctl_buffer { struct pmcraid_ioctl_header ioctl_header; struct pmcraid_ioarcb ioarcb; struct pmcraid_ioasa ioasa; - u8 request_buffer[1]; -} __attribute__ ((packed)); + u8 request_buffer[]; +} __attribute__ ((packed, aligned(PMCRAID_IOARCB_ALIGNMENT))); /* * keys to differentiate between driver handled IOCTLs and passthrough diff --git a/drivers/scsi/qla2xxx/qla_target.c b/drivers/scsi/qla2xxx/qla_target.c index 0d09480b66cd..c48daf52725d 100644 --- a/drivers/scsi/qla2xxx/qla_target.c +++ b/drivers/scsi/qla2xxx/qla_target.c @@ -981,8 +981,7 @@ void qlt_free_session_done(struct work_struct *work) int rc; if (!own || - (own && - (own->iocb.u.isp24.status_subcode == ELS_PLOGI))) { + (own->iocb.u.isp24.status_subcode == ELS_PLOGI)) { rc = qla2x00_post_async_logout_work(vha, sess, NULL); if (rc != QLA_SUCCESS) diff --git a/drivers/scsi/qla4xxx/ql4_os.c b/drivers/scsi/qla4xxx/ql4_os.c index a4b014e1cd8c..7bd9a4a04ad5 100644 --- a/drivers/scsi/qla4xxx/ql4_os.c +++ b/drivers/scsi/qla4xxx/ql4_os.c @@ -841,7 +841,7 @@ static int __qla4xxx_is_chap_active(struct device *dev, void *data) sess = cls_session->dd_data; ddb_entry = sess->dd_data; - if (iscsi_session_chkready(cls_session)) + if (iscsi_is_session_online(cls_session)) goto exit_is_chap_active; if (ddb_entry->chap_tbl_idx == *chap_tbl_idx) diff --git a/drivers/scsi/scsi_transport_iscsi.c b/drivers/scsi/scsi_transport_iscsi.c index 2e68c0a87698..969d24d580e2 100644 --- a/drivers/scsi/scsi_transport_iscsi.c +++ b/drivers/scsi/scsi_transport_iscsi.c @@ -1701,10 +1701,8 @@ static const char *iscsi_session_state_name(int state) int iscsi_session_chkready(struct iscsi_cls_session *session) { - unsigned long flags; int err; - spin_lock_irqsave(&session->lock, flags); switch (session->state) { case ISCSI_SESSION_LOGGED_IN: err = 0; @@ -1719,7 +1717,6 @@ int iscsi_session_chkready(struct iscsi_cls_session *session) err = DID_NO_CONNECT << 16; break; } - spin_unlock_irqrestore(&session->lock, flags); return err; } EXPORT_SYMBOL_GPL(iscsi_session_chkready); diff --git a/drivers/scsi/sd.c b/drivers/scsi/sd.c index a3d2d4bc4a3d..ed0b1bb99f08 100644 --- a/drivers/scsi/sd.c +++ b/drivers/scsi/sd.c @@ -707,9 +707,9 @@ static int sd_sec_submit(void *data, u16 spsp, u8 secp, void *buffer, put_unaligned_be16(spsp, &cdb[2]); put_unaligned_be32(len, &cdb[6]); - ret = scsi_execute_req(sdev, cdb, - send ? DMA_TO_DEVICE : DMA_FROM_DEVICE, - buffer, len, NULL, SD_TIMEOUT, sdkp->max_retries, NULL); + ret = scsi_execute(sdev, cdb, send ? DMA_TO_DEVICE : DMA_FROM_DEVICE, + buffer, len, NULL, NULL, SD_TIMEOUT, sdkp->max_retries, 0, + RQF_PM, NULL); return ret <= 0 ? ret : -EIO; } #endif /* CONFIG_BLK_SED_OPAL */ @@ -3379,10 +3379,12 @@ static int sd_probe(struct device *dev) sdp->type != TYPE_RBC) goto out; -#ifndef CONFIG_BLK_DEV_ZONED - if (sdp->type == TYPE_ZBC) + if (!IS_ENABLED(CONFIG_BLK_DEV_ZONED) && sdp->type == TYPE_ZBC) { + sdev_printk(KERN_WARNING, sdp, + "Unsupported ZBC host-managed device.\n"); goto out; -#endif + } + SCSI_LOG_HLQUEUE(3, sdev_printk(KERN_INFO, sdp, "sd_probe\n")); diff --git a/drivers/scsi/sd_zbc.c b/drivers/scsi/sd_zbc.c index 03adb39293c2..ee558675eab4 100644 --- a/drivers/scsi/sd_zbc.c +++ b/drivers/scsi/sd_zbc.c @@ -704,6 +704,7 @@ int sd_zbc_revalidate_zones(struct scsi_disk *sdkp) unsigned int nr_zones = sdkp->rev_nr_zones; u32 max_append; int ret = 0; + unsigned int flags; /* * For all zoned disks, initialize zone append emulation data if not @@ -736,16 +737,19 @@ int sd_zbc_revalidate_zones(struct scsi_disk *sdkp) disk->queue->nr_zones == nr_zones) goto unlock; + flags = memalloc_noio_save(); sdkp->zone_blocks = zone_blocks; sdkp->nr_zones = nr_zones; - sdkp->rev_wp_offset = kvcalloc(nr_zones, sizeof(u32), GFP_NOIO); + sdkp->rev_wp_offset = kvcalloc(nr_zones, sizeof(u32), GFP_KERNEL); if (!sdkp->rev_wp_offset) { ret = -ENOMEM; + memalloc_noio_restore(flags); goto unlock; } ret = blk_revalidate_disk_zones(disk, sd_zbc_revalidate_zones_cb); + memalloc_noio_restore(flags); kvfree(sdkp->rev_wp_offset); sdkp->rev_wp_offset = NULL; diff --git a/drivers/scsi/ufs/ufshcd.c b/drivers/scsi/ufs/ufshcd.c index 721f55db181f..77161750c9fb 100644 --- a/drivers/scsi/ufs/ufshcd.c +++ b/drivers/scsi/ufs/ufshcd.c @@ -451,6 +451,8 @@ static void ufshcd_print_evt(struct ufs_hba *hba, u32 id, if (!found) dev_err(hba->dev, "No record of %s\n", err_name); + else + dev_err(hba->dev, "%s: total cnt=%llu\n", err_name, e->cnt); } static void ufshcd_print_evt_hist(struct ufs_hba *hba) @@ -1866,7 +1868,7 @@ static ssize_t ufshcd_clkgate_delay_show(struct device *dev, { struct ufs_hba *hba = dev_get_drvdata(dev); - return snprintf(buf, PAGE_SIZE, "%lu\n", hba->clk_gating.delay_ms); + return sysfs_emit(buf, "%lu\n", hba->clk_gating.delay_ms); } static ssize_t ufshcd_clkgate_delay_store(struct device *dev, @@ -1889,7 +1891,7 @@ static ssize_t ufshcd_clkgate_enable_show(struct device *dev, { struct ufs_hba *hba = dev_get_drvdata(dev); - return snprintf(buf, PAGE_SIZE, "%d\n", hba->clk_gating.is_enabled); + return sysfs_emit(buf, "%d\n", hba->clk_gating.is_enabled); } static ssize_t ufshcd_clkgate_enable_store(struct device *dev, diff --git a/drivers/scsi/ufs/ufshcd.h b/drivers/scsi/ufs/ufshcd.h index ee61f821f75d..18e56c1c1b30 100644 --- a/drivers/scsi/ufs/ufshcd.h +++ b/drivers/scsi/ufs/ufshcd.h @@ -570,7 +570,7 @@ enum ufshcd_quirks { /* * This quirk allows only sg entries aligned with page size. */ - UFSHCD_QUIRK_ALIGN_SG_WITH_PAGE_SIZE = 1 << 13, + UFSHCD_QUIRK_ALIGN_SG_WITH_PAGE_SIZE = 1 << 14, }; enum ufshcd_caps { diff --git a/drivers/staging/vme/devices/vme_user.c b/drivers/staging/vme/devices/vme_user.c index 35d7260e2271..e3fa38bd7f12 100644 --- a/drivers/staging/vme/devices/vme_user.c +++ b/drivers/staging/vme/devices/vme_user.c @@ -175,7 +175,7 @@ static ssize_t buffer_from_user(unsigned int minor, const char __user *buf, static ssize_t vme_user_read(struct file *file, char __user *buf, size_t count, loff_t *ppos) { - unsigned int minor = MINOR(file_inode(file)->i_rdev); + unsigned int minor = iminor(file_inode(file)); ssize_t retval; size_t image_size; @@ -218,7 +218,7 @@ static ssize_t vme_user_read(struct file *file, char __user *buf, size_t count, static ssize_t vme_user_write(struct file *file, const char __user *buf, size_t count, loff_t *ppos) { - unsigned int minor = MINOR(file_inode(file)->i_rdev); + unsigned int minor = iminor(file_inode(file)); ssize_t retval; size_t image_size; @@ -260,7 +260,7 @@ static ssize_t vme_user_write(struct file *file, const char __user *buf, static loff_t vme_user_llseek(struct file *file, loff_t off, int whence) { - unsigned int minor = MINOR(file_inode(file)->i_rdev); + unsigned int minor = iminor(file_inode(file)); size_t image_size; loff_t res; @@ -294,7 +294,7 @@ static int vme_user_ioctl(struct inode *inode, struct file *file, struct vme_slave slave; struct vme_irq_id irq_req; unsigned long copied; - unsigned int minor = MINOR(inode->i_rdev); + unsigned int minor = iminor(inode); int retval; dma_addr_t pci_addr; void __user *argp = (void __user *)arg; @@ -412,7 +412,7 @@ vme_user_unlocked_ioctl(struct file *file, unsigned int cmd, unsigned long arg) { int ret; struct inode *inode = file_inode(file); - unsigned int minor = MINOR(inode->i_rdev); + unsigned int minor = iminor(inode); mutex_lock(&image[minor].mutex); ret = vme_user_ioctl(inode, file, cmd, arg); @@ -481,7 +481,7 @@ static int vme_user_master_mmap(unsigned int minor, struct vm_area_struct *vma) static int vme_user_mmap(struct file *file, struct vm_area_struct *vma) { - unsigned int minor = MINOR(file_inode(file)->i_rdev); + unsigned int minor = iminor(file_inode(file)); if (type[minor] == MASTER_MINOR) return vme_user_master_mmap(minor, vma); diff --git a/drivers/target/sbp/sbp_target.c b/drivers/target/sbp/sbp_target.c index e4a9b9fe3dfb..2a6165febd3b 100644 --- a/drivers/target/sbp/sbp_target.c +++ b/drivers/target/sbp/sbp_target.c @@ -1006,7 +1006,7 @@ static void tgt_agent_fetch_work(struct work_struct *work) agent->state = AGENT_STATE_SUSPENDED; spin_unlock_bh(&agent->lock); - }; + } } static struct sbp_target_agent *sbp_target_agent_register( diff --git a/drivers/target/target_core_iblock.c b/drivers/target/target_core_iblock.c index 8ed93fd205c7..ee3d52061281 100644 --- a/drivers/target/target_core_iblock.c +++ b/drivers/target/target_core_iblock.c @@ -315,10 +315,8 @@ iblock_get_bio(struct se_cmd *cmd, sector_t lba, u32 sg_num, int op, * Only allocate as many vector entries as the bio code allows us to, * we'll loop later on until we have handled the whole request. */ - if (sg_num > BIO_MAX_PAGES) - sg_num = BIO_MAX_PAGES; - - bio = bio_alloc_bioset(GFP_NOIO, sg_num, &ib_dev->ibd_bio_set); + bio = bio_alloc_bioset(GFP_NOIO, bio_max_segs(sg_num), + &ib_dev->ibd_bio_set); if (!bio) { pr_err("Unable to allocate memory for bio\n"); return NULL; @@ -638,8 +636,7 @@ iblock_alloc_bip(struct se_cmd *cmd, struct bio *bio, return -ENODEV; } - bip = bio_integrity_alloc(bio, GFP_NOIO, - min_t(unsigned int, cmd->t_prot_nents, BIO_MAX_PAGES)); + bip = bio_integrity_alloc(bio, GFP_NOIO, bio_max_segs(cmd->t_prot_nents)); if (IS_ERR(bip)) { pr_err("Unable to allocate bio_integrity_payload\n"); return PTR_ERR(bip); diff --git a/drivers/target/target_core_pr.c b/drivers/target/target_core_pr.c index 14db5e568f22..d4cc43afe05b 100644 --- a/drivers/target/target_core_pr.c +++ b/drivers/target/target_core_pr.c @@ -3739,6 +3739,7 @@ core_scsi3_pri_read_keys(struct se_cmd *cmd) spin_unlock(&dev->t10_pr.registration_lock); put_unaligned_be32(add_len, &buf[4]); + target_set_cmd_data_length(cmd, 8 + add_len); transport_kunmap_data_sg(cmd); @@ -3757,7 +3758,7 @@ core_scsi3_pri_read_reservation(struct se_cmd *cmd) struct t10_pr_registration *pr_reg; unsigned char *buf; u64 pr_res_key; - u32 add_len = 16; /* Hardcoded to 16 when a reservation is held. */ + u32 add_len = 0; if (cmd->data_length < 8) { pr_err("PRIN SA READ_RESERVATIONS SCSI Data Length: %u" @@ -3775,8 +3776,9 @@ core_scsi3_pri_read_reservation(struct se_cmd *cmd) pr_reg = dev->dev_pr_res_holder; if (pr_reg) { /* - * Set the hardcoded Additional Length + * Set the Additional Length to 16 when a reservation is held */ + add_len = 16; put_unaligned_be32(add_len, &buf[4]); if (cmd->data_length < 22) @@ -3812,6 +3814,8 @@ core_scsi3_pri_read_reservation(struct se_cmd *cmd) (pr_reg->pr_res_type & 0x0f); } + target_set_cmd_data_length(cmd, 8 + add_len); + err: spin_unlock(&dev->dev_reservation_lock); transport_kunmap_data_sg(cmd); @@ -3830,7 +3834,7 @@ core_scsi3_pri_report_capabilities(struct se_cmd *cmd) struct se_device *dev = cmd->se_dev; struct t10_reservation *pr_tmpl = &dev->t10_pr; unsigned char *buf; - u16 add_len = 8; /* Hardcoded to 8. */ + u16 len = 8; /* Hardcoded to 8. */ if (cmd->data_length < 6) { pr_err("PRIN SA REPORT_CAPABILITIES SCSI Data Length:" @@ -3842,7 +3846,7 @@ core_scsi3_pri_report_capabilities(struct se_cmd *cmd) if (!buf) return TCM_LOGICAL_UNIT_COMMUNICATION_FAILURE; - put_unaligned_be16(add_len, &buf[0]); + put_unaligned_be16(len, &buf[0]); buf[2] |= 0x10; /* CRH: Compatible Reservation Hanlding bit. */ buf[2] |= 0x08; /* SIP_C: Specify Initiator Ports Capable bit */ buf[2] |= 0x04; /* ATP_C: All Target Ports Capable bit */ @@ -3871,6 +3875,8 @@ core_scsi3_pri_report_capabilities(struct se_cmd *cmd) buf[4] |= 0x02; /* PR_TYPE_WRITE_EXCLUSIVE */ buf[5] |= 0x01; /* PR_TYPE_EXCLUSIVE_ACCESS_ALLREG */ + target_set_cmd_data_length(cmd, len); + transport_kunmap_data_sg(cmd); return 0; @@ -4031,6 +4037,7 @@ core_scsi3_pri_read_full_status(struct se_cmd *cmd) * Set ADDITIONAL_LENGTH */ put_unaligned_be32(add_len, &buf[4]); + target_set_cmd_data_length(cmd, 8 + add_len); transport_kunmap_data_sg(cmd); diff --git a/drivers/target/target_core_pscsi.c b/drivers/target/target_core_pscsi.c index 33770e5808ce..3cbc074992bc 100644 --- a/drivers/target/target_core_pscsi.c +++ b/drivers/target/target_core_pscsi.c @@ -881,7 +881,7 @@ pscsi_map_sg(struct se_cmd *cmd, struct scatterlist *sgl, u32 sgl_nents, if (!bio) { new_bio: - nr_vecs = min_t(int, BIO_MAX_PAGES, nr_pages); + nr_vecs = bio_max_segs(nr_pages); nr_pages -= nr_vecs; /* * Calls bio_kmalloc() and sets bio->bi_end_io() diff --git a/drivers/target/target_core_transport.c b/drivers/target/target_core_transport.c index 93ea17cbad79..5ecb9f18a53d 100644 --- a/drivers/target/target_core_transport.c +++ b/drivers/target/target_core_transport.c @@ -879,11 +879,9 @@ void target_complete_cmd(struct se_cmd *cmd, u8 scsi_status) } EXPORT_SYMBOL(target_complete_cmd); -void target_complete_cmd_with_length(struct se_cmd *cmd, u8 scsi_status, int length) +void target_set_cmd_data_length(struct se_cmd *cmd, int length) { - if ((scsi_status == SAM_STAT_GOOD || - cmd->se_cmd_flags & SCF_TREAT_READ_AS_NORMAL) && - length < cmd->data_length) { + if (length < cmd->data_length) { if (cmd->se_cmd_flags & SCF_UNDERFLOW_BIT) { cmd->residual_count += cmd->data_length - length; } else { @@ -893,6 +891,15 @@ void target_complete_cmd_with_length(struct se_cmd *cmd, u8 scsi_status, int len cmd->data_length = length; } +} +EXPORT_SYMBOL(target_set_cmd_data_length); + +void target_complete_cmd_with_length(struct se_cmd *cmd, u8 scsi_status, int length) +{ + if (scsi_status == SAM_STAT_GOOD || + cmd->se_cmd_flags & SCF_TREAT_READ_AS_NORMAL) { + target_set_cmd_data_length(cmd, length); + } target_complete_cmd(cmd, scsi_status); } diff --git a/drivers/target/target_core_user.c b/drivers/target/target_core_user.c index a5991df23581..bf73cd5f4b04 100644 --- a/drivers/target/target_core_user.c +++ b/drivers/target/target_core_user.c @@ -1566,6 +1566,88 @@ static struct se_device *tcmu_alloc_device(struct se_hba *hba, const char *name) return &udev->se_dev; } +static void tcmu_dev_call_rcu(struct rcu_head *p) +{ + struct se_device *dev = container_of(p, struct se_device, rcu_head); + struct tcmu_dev *udev = TCMU_DEV(dev); + + kfree(udev->uio_info.name); + kfree(udev->name); + kfree(udev); +} + +static int tcmu_check_and_free_pending_cmd(struct tcmu_cmd *cmd) +{ + if (test_bit(TCMU_CMD_BIT_EXPIRED, &cmd->flags)) { + kmem_cache_free(tcmu_cmd_cache, cmd); + return 0; + } + return -EINVAL; +} + +static void tcmu_blocks_release(struct radix_tree_root *blocks, + int start, int end) +{ + int i; + struct page *page; + + for (i = start; i < end; i++) { + page = radix_tree_delete(blocks, i); + if (page) { + __free_page(page); + atomic_dec(&global_db_count); + } + } +} + +static void tcmu_remove_all_queued_tmr(struct tcmu_dev *udev) +{ + struct tcmu_tmr *tmr, *tmp; + + list_for_each_entry_safe(tmr, tmp, &udev->tmr_queue, queue_entry) { + list_del_init(&tmr->queue_entry); + kfree(tmr); + } +} + +static void tcmu_dev_kref_release(struct kref *kref) +{ + struct tcmu_dev *udev = container_of(kref, struct tcmu_dev, kref); + struct se_device *dev = &udev->se_dev; + struct tcmu_cmd *cmd; + bool all_expired = true; + int i; + + vfree(udev->mb_addr); + udev->mb_addr = NULL; + + spin_lock_bh(&timed_out_udevs_lock); + if (!list_empty(&udev->timedout_entry)) + list_del(&udev->timedout_entry); + spin_unlock_bh(&timed_out_udevs_lock); + + /* Upper layer should drain all requests before calling this */ + mutex_lock(&udev->cmdr_lock); + idr_for_each_entry(&udev->commands, cmd, i) { + if (tcmu_check_and_free_pending_cmd(cmd) != 0) + all_expired = false; + } + /* There can be left over TMR cmds. Remove them. */ + tcmu_remove_all_queued_tmr(udev); + if (!list_empty(&udev->qfull_queue)) + all_expired = false; + idr_destroy(&udev->commands); + WARN_ON(!all_expired); + + tcmu_blocks_release(&udev->data_blocks, 0, udev->dbi_max + 1); + bitmap_free(udev->data_bitmap); + mutex_unlock(&udev->cmdr_lock); + + pr_debug("dev_kref_release\n"); + + call_rcu(&dev->rcu_head, tcmu_dev_call_rcu); +} + static void run_qfull_queue(struct tcmu_dev *udev, bool fail) { struct tcmu_cmd *tcmu_cmd, *tmp_cmd; @@ -1678,6 +1760,25 @@ static struct page *tcmu_try_get_block_page(struct tcmu_dev *udev, uint32_t dbi) return page; } +static void tcmu_vma_open(struct vm_area_struct *vma) +{ + struct tcmu_dev *udev = vma->vm_private_data; + + pr_debug("vma_open\n"); + + kref_get(&udev->kref); +} + +static void tcmu_vma_close(struct vm_area_struct *vma) +{ + struct tcmu_dev *udev = vma->vm_private_data; + + pr_debug("vma_close\n"); + + /* release ref from tcmu_vma_open */ + kref_put(&udev->kref, tcmu_dev_kref_release); +} + static vm_fault_t tcmu_vma_fault(struct vm_fault *vmf) { struct tcmu_dev *udev = vmf->vma->vm_private_data; @@ -1716,6 +1817,8 @@ static vm_fault_t tcmu_vma_fault(struct vm_fault *vmf) } static const struct vm_operations_struct tcmu_vm_ops = { + .open = tcmu_vma_open, + .close = tcmu_vma_close, .fault = tcmu_vma_fault, }; @@ -1732,6 +1835,8 @@ static int tcmu_mmap(struct uio_info *info, struct vm_area_struct *vma) if (vma_pages(vma) != (udev->ring_size >> PAGE_SHIFT)) return -EINVAL; + tcmu_vma_open(vma); + return 0; } @@ -1744,93 +1849,12 @@ static int tcmu_open(struct uio_info *info, struct inode *inode) return -EBUSY; udev->inode = inode; - kref_get(&udev->kref); pr_debug("open\n"); return 0; } -static void tcmu_dev_call_rcu(struct rcu_head *p) -{ - struct se_device *dev = container_of(p, struct se_device, rcu_head); - struct tcmu_dev *udev = TCMU_DEV(dev); - - kfree(udev->uio_info.name); - kfree(udev->name); - kfree(udev); -} - -static int tcmu_check_and_free_pending_cmd(struct tcmu_cmd *cmd) -{ - if (test_bit(TCMU_CMD_BIT_EXPIRED, &cmd->flags)) { - kmem_cache_free(tcmu_cmd_cache, cmd); - return 0; - } - return -EINVAL; -} - -static void tcmu_blocks_release(struct radix_tree_root *blocks, - int start, int end) -{ - int i; - struct page *page; - - for (i = start; i < end; i++) { - page = radix_tree_delete(blocks, i); - if (page) { - __free_page(page); - atomic_dec(&global_db_count); - } - } -} - -static void tcmu_remove_all_queued_tmr(struct tcmu_dev *udev) -{ - struct tcmu_tmr *tmr, *tmp; - - list_for_each_entry_safe(tmr, tmp, &udev->tmr_queue, queue_entry) { - list_del_init(&tmr->queue_entry); - kfree(tmr); - } -} - -static void tcmu_dev_kref_release(struct kref *kref) -{ - struct tcmu_dev *udev = container_of(kref, struct tcmu_dev, kref); - struct se_device *dev = &udev->se_dev; - struct tcmu_cmd *cmd; - bool all_expired = true; - int i; - - vfree(udev->mb_addr); - udev->mb_addr = NULL; - - spin_lock_bh(&timed_out_udevs_lock); - if (!list_empty(&udev->timedout_entry)) - list_del(&udev->timedout_entry); - spin_unlock_bh(&timed_out_udevs_lock); - - /* Upper layer should drain all requests before calling this */ - mutex_lock(&udev->cmdr_lock); - idr_for_each_entry(&udev->commands, cmd, i) { - if (tcmu_check_and_free_pending_cmd(cmd) != 0) - all_expired = false; - } - /* There can be left over TMR cmds. Remove them. */ - tcmu_remove_all_queued_tmr(udev); - if (!list_empty(&udev->qfull_queue)) - all_expired = false; - idr_destroy(&udev->commands); - WARN_ON(!all_expired); - - tcmu_blocks_release(&udev->data_blocks, 0, udev->dbi_max + 1); - bitmap_free(udev->data_bitmap); - mutex_unlock(&udev->cmdr_lock); - - call_rcu(&dev->rcu_head, tcmu_dev_call_rcu); -} - static int tcmu_release(struct uio_info *info, struct inode *inode) { struct tcmu_dev *udev = container_of(info, struct tcmu_dev, uio_info); @@ -1838,8 +1862,7 @@ static int tcmu_release(struct uio_info *info, struct inode *inode) clear_bit(TCMU_DEV_BIT_OPEN, &udev->flags); pr_debug("close\n"); - /* release ref from open */ - kref_put(&udev->kref, tcmu_dev_kref_release); + return 0; } diff --git a/fs/9p/vfs_inode.c b/fs/9p/vfs_inode.c index 648eb4c4cf7f..8d97f0b45e9c 100644 --- a/fs/9p/vfs_inode.c +++ b/fs/9p/vfs_inode.c @@ -1139,9 +1139,6 @@ v9fs_stat2inode(struct p9_wstat *stat, struct inode *inode, struct super_block *sb, unsigned int flags) { umode_t mode; - char ext[32]; - char tag_name[14]; - unsigned int i_nlink; struct v9fs_session_info *v9ses = sb->s_fs_info; struct v9fs_inode *v9inode = V9FS_I(inode); @@ -1159,18 +1156,18 @@ v9fs_stat2inode(struct p9_wstat *stat, struct inode *inode, inode->i_gid = stat->n_gid; } if ((S_ISREG(inode->i_mode)) || (S_ISDIR(inode->i_mode))) { - if (v9fs_proto_dotu(v9ses) && (stat->extension[0] != '\0')) { + if (v9fs_proto_dotu(v9ses)) { + unsigned int i_nlink; /* - * Hadlink support got added later to - * to the .u extension. So there can be - * server out there that doesn't support - * this even with .u extension. So check - * for non NULL stat->extension + * Hadlink support got added later to the .u extension. + * So there can be a server out there that doesn't + * support this even with .u extension. That would + * just leave us with stat->extension being an empty + * string, though. */ - strlcpy(ext, stat->extension, sizeof(ext)); /* HARDLINKCOUNT %u */ - sscanf(ext, "%13s %u", tag_name, &i_nlink); - if (!strncmp(tag_name, "HARDLINKCOUNT", 13)) + if (sscanf(stat->extension, + " HARDLINKCOUNT %u", &i_nlink) == 1) set_nlink(inode, i_nlink); } } diff --git a/fs/block_dev.c b/fs/block_dev.c index 91ff9bfd7c1a..4aa1f88d5bf8 100644 --- a/fs/block_dev.c +++ b/fs/block_dev.c @@ -221,7 +221,7 @@ static void blkdev_bio_end_io_simple(struct bio *bio) static ssize_t __blkdev_direct_IO_simple(struct kiocb *iocb, struct iov_iter *iter, - int nr_pages) + unsigned int nr_pages) { struct file *file = iocb->ki_filp; struct block_device *bdev = I_BDEV(bdev_file_inode(file)); @@ -355,8 +355,8 @@ static void blkdev_bio_end_io(struct bio *bio) } } -static ssize_t -__blkdev_direct_IO(struct kiocb *iocb, struct iov_iter *iter, int nr_pages) +static ssize_t __blkdev_direct_IO(struct kiocb *iocb, struct iov_iter *iter, + unsigned int nr_pages) { struct file *file = iocb->ki_filp; struct inode *inode = bdev_file_inode(file); @@ -486,7 +486,7 @@ __blkdev_direct_IO(struct kiocb *iocb, struct iov_iter *iter, int nr_pages) static ssize_t blkdev_direct_IO(struct kiocb *iocb, struct iov_iter *iter) { - int nr_pages; + unsigned int nr_pages; if (!iov_iter_count(iter)) return 0; @@ -495,7 +495,7 @@ blkdev_direct_IO(struct kiocb *iocb, struct iov_iter *iter) if (is_sync_kiocb(iocb) && nr_pages <= BIO_MAX_PAGES) return __blkdev_direct_IO_simple(iocb, iter, nr_pages); - return __blkdev_direct_IO(iocb, iter, min(nr_pages, BIO_MAX_PAGES)); + return __blkdev_direct_IO(iocb, iter, bio_max_segs(nr_pages)); } static __init int blkdev_init(void) diff --git a/fs/direct-io.c b/fs/direct-io.c index 0957e1bb8eb2..b61491bf3166 100644 --- a/fs/direct-io.c +++ b/fs/direct-io.c @@ -695,7 +695,7 @@ static inline int dio_new_bio(struct dio *dio, struct dio_submit *sdio, if (ret) goto out; sector = start_sector << (sdio->blkbits - 9); - nr_pages = min(sdio->pages_in_io, BIO_MAX_PAGES); + nr_pages = bio_max_segs(sdio->pages_in_io); BUG_ON(nr_pages <= 0); dio_bio_alloc(dio, sdio, map_bh->b_bdev, sector, nr_pages); sdio->boundary = 0; diff --git a/fs/erofs/data.c b/fs/erofs/data.c index ea4f693bee22..f88851c5c250 100644 --- a/fs/erofs/data.c +++ b/fs/erofs/data.c @@ -215,10 +215,8 @@ submit_bio_retry: /* max # of continuous pages */ if (nblocks > DIV_ROUND_UP(map.m_plen, PAGE_SIZE)) nblocks = DIV_ROUND_UP(map.m_plen, PAGE_SIZE); - if (nblocks > BIO_MAX_PAGES) - nblocks = BIO_MAX_PAGES; - bio = bio_alloc(GFP_NOIO, nblocks); + bio = bio_alloc(GFP_NOIO, bio_max_segs(nblocks)); bio->bi_end_io = erofs_readendio; bio_set_dev(bio, sb->s_bdev); diff --git a/fs/ext4/readpage.c b/fs/ext4/readpage.c index f014c5e473a9..3db923403505 100644 --- a/fs/ext4/readpage.c +++ b/fs/ext4/readpage.c @@ -371,8 +371,7 @@ int ext4_mpage_readpages(struct inode *inode, * bio_alloc will _always_ be able to allocate a bio if * __GFP_DIRECT_RECLAIM is set, see bio_alloc_bioset(). */ - bio = bio_alloc(GFP_KERNEL, - min_t(int, nr_pages, BIO_MAX_PAGES)); + bio = bio_alloc(GFP_KERNEL, bio_max_segs(nr_pages)); fscrypt_set_bio_crypt_ctx(bio, inode, next_block, GFP_KERNEL); ext4_set_bio_post_read_ctx(bio, inode, page->index); diff --git a/fs/f2fs/data.c b/fs/f2fs/data.c index b9721c8f116c..7c95818639a6 100644 --- a/fs/f2fs/data.c +++ b/fs/f2fs/data.c @@ -969,8 +969,7 @@ static struct bio *f2fs_grab_read_bio(struct inode *inode, block_t blkaddr, unsigned int post_read_steps = 0; bio = bio_alloc_bioset(for_write ? GFP_NOIO : GFP_KERNEL, - min_t(int, nr_pages, BIO_MAX_PAGES), - &f2fs_bioset); + bio_max_segs(nr_pages), &f2fs_bioset); if (!bio) return ERR_PTR(-ENOMEM); diff --git a/fs/f2fs/node.c b/fs/f2fs/node.c index a8a0fb890e8d..4b0e2e3c2c88 100644 --- a/fs/f2fs/node.c +++ b/fs/f2fs/node.c @@ -2747,7 +2747,7 @@ int f2fs_restore_node_summary(struct f2fs_sb_info *sbi, sum_entry = &sum->entries[0]; for (i = 0; i < last_offset; i += nrpages, addr += nrpages) { - nrpages = min(last_offset - i, BIO_MAX_PAGES); + nrpages = bio_max_segs(last_offset - i); /* readahead node pages */ f2fs_ra_meta_pages(sbi, addr, nrpages, META_POR, true); diff --git a/fs/gfs2/inode.c b/fs/gfs2/inode.c index 68376fa03880..c9775d5c6594 100644 --- a/fs/gfs2/inode.c +++ b/fs/gfs2/inode.c @@ -491,8 +491,8 @@ static void init_dinode(struct gfs2_inode *dip, struct gfs2_inode *ip, di = (struct gfs2_dinode *)dibh->b_data; gfs2_dinode_out(ip, di); - di->di_major = cpu_to_be32(MAJOR(ip->i_inode.i_rdev)); - di->di_minor = cpu_to_be32(MINOR(ip->i_inode.i_rdev)); + di->di_major = cpu_to_be32(imajor(&ip->i_inode)); + di->di_minor = cpu_to_be32(iminor(&ip->i_inode)); di->__pad1 = 0; di->__pad2 = 0; di->__pad3 = 0; diff --git a/fs/inode.c b/fs/inode.c index 6dba963d3f6d..a047ab306f9a 100644 --- a/fs/inode.c +++ b/fs/inode.c @@ -142,6 +142,7 @@ int inode_init_always(struct super_block *sb, struct inode *inode) atomic_set(&inode->i_count, 1); inode->i_op = &empty_iops; inode->i_fop = &no_open_fops; + inode->i_ino = 0; inode->__i_nlink = 1; inode->i_opflags = 0; if (sb->s_xattr) diff --git a/fs/io-wq.c b/fs/io-wq.c index c36bbcd823ce..44e20248805a 100644 --- a/fs/io-wq.c +++ b/fs/io-wq.c @@ -13,13 +13,9 @@ #include <linux/sched/mm.h> #include <linux/percpu.h> #include <linux/slab.h> -#include <linux/kthread.h> #include <linux/rculist_nulls.h> -#include <linux/fs_struct.h> -#include <linux/task_work.h> -#include <linux/blk-cgroup.h> -#include <linux/audit.h> #include <linux/cpu.h> +#include <linux/tracehook.h> #include "../kernel/sched/sched.h" #include "io-wq.h" @@ -36,7 +32,6 @@ enum { enum { IO_WQ_BIT_EXIT = 0, /* wq exiting */ - IO_WQ_BIT_ERROR = 1, /* error on setup */ }; enum { @@ -57,14 +52,12 @@ struct io_worker { struct io_wq_work *cur_work; spinlock_t lock; - struct rcu_head rcu; - struct mm_struct *mm; -#ifdef CONFIG_BLK_CGROUP - struct cgroup_subsys_state *blkcg_css; -#endif const struct cred *cur_creds; const struct cred *saved_creds; - struct nsproxy *restore_nsproxy; + + struct completion ref_done; + + struct rcu_head rcu; }; #if BITS_PER_LONG == 64 @@ -93,7 +86,6 @@ struct io_wqe { struct { raw_spinlock_t lock; struct io_wq_work_list work_list; - unsigned long hash_map; unsigned flags; } ____cacheline_aligned_in_smp; @@ -103,6 +95,8 @@ struct io_wqe { struct hlist_nulls_head free_list; struct list_head all_list; + struct wait_queue_entry wait; + struct io_wq *wq; struct io_wq_work *hash_tail[IO_WQ_NR_HASH_BUCKETS]; }; @@ -119,12 +113,15 @@ struct io_wq { struct task_struct *manager; struct user_struct *user; + + struct io_wq_hash *hash; + refcount_t refs; struct completion done; struct hlist_node cpuhp_node; - refcount_t use_refs; + pid_t task_pid; }; static enum cpuhp_state io_wq_online; @@ -137,62 +134,7 @@ static bool io_worker_get(struct io_worker *worker) static void io_worker_release(struct io_worker *worker) { if (refcount_dec_and_test(&worker->ref)) - wake_up_process(worker->task); -} - -/* - * Note: drops the wqe->lock if returning true! The caller must re-acquire - * the lock in that case. Some callers need to restart handling if this - * happens, so we can't just re-acquire the lock on behalf of the caller. - */ -static bool __io_worker_unuse(struct io_wqe *wqe, struct io_worker *worker) -{ - bool dropped_lock = false; - - if (worker->saved_creds) { - revert_creds(worker->saved_creds); - worker->cur_creds = worker->saved_creds = NULL; - } - - if (current->files) { - __acquire(&wqe->lock); - raw_spin_unlock_irq(&wqe->lock); - dropped_lock = true; - - task_lock(current); - current->files = NULL; - current->nsproxy = worker->restore_nsproxy; - task_unlock(current); - } - - if (current->fs) - current->fs = NULL; - - /* - * If we have an active mm, we need to drop the wq lock before unusing - * it. If we do, return true and let the caller retry the idle loop. - */ - if (worker->mm) { - if (!dropped_lock) { - __acquire(&wqe->lock); - raw_spin_unlock_irq(&wqe->lock); - dropped_lock = true; - } - __set_current_state(TASK_RUNNING); - kthread_unuse_mm(worker->mm); - mmput(worker->mm); - worker->mm = NULL; - } - -#ifdef CONFIG_BLK_CGROUP - if (worker->blkcg_css) { - kthread_associate_blkcg(NULL); - worker->blkcg_css = NULL; - } -#endif - if (current->signal->rlim[RLIMIT_FSIZE].rlim_cur != RLIM_INFINITY) - current->signal->rlim[RLIMIT_FSIZE].rlim_cur = RLIM_INFINITY; - return dropped_lock; + complete(&worker->ref_done); } static inline struct io_wqe_acct *io_work_get_acct(struct io_wqe *wqe, @@ -204,9 +146,10 @@ static inline struct io_wqe_acct *io_work_get_acct(struct io_wqe *wqe, return &wqe->acct[IO_WQ_ACCT_BOUND]; } -static inline struct io_wqe_acct *io_wqe_get_acct(struct io_wqe *wqe, - struct io_worker *worker) +static inline struct io_wqe_acct *io_wqe_get_acct(struct io_worker *worker) { + struct io_wqe *wqe = worker->wqe; + if (worker->flags & IO_WORKER_F_BOUND) return &wqe->acct[IO_WQ_ACCT_BOUND]; @@ -216,39 +159,36 @@ static inline struct io_wqe_acct *io_wqe_get_acct(struct io_wqe *wqe, static void io_worker_exit(struct io_worker *worker) { struct io_wqe *wqe = worker->wqe; - struct io_wqe_acct *acct = io_wqe_get_acct(wqe, worker); + struct io_wqe_acct *acct = io_wqe_get_acct(worker); + unsigned flags; - /* - * If we're not at zero, someone else is holding a brief reference - * to the worker. Wait for that to go away. - */ - set_current_state(TASK_INTERRUPTIBLE); - if (!refcount_dec_and_test(&worker->ref)) - schedule(); - __set_current_state(TASK_RUNNING); + if (refcount_dec_and_test(&worker->ref)) + complete(&worker->ref_done); + wait_for_completion(&worker->ref_done); preempt_disable(); current->flags &= ~PF_IO_WORKER; - if (worker->flags & IO_WORKER_F_RUNNING) + flags = worker->flags; + worker->flags = 0; + if (flags & IO_WORKER_F_RUNNING) atomic_dec(&acct->nr_running); - if (!(worker->flags & IO_WORKER_F_BOUND)) - atomic_dec(&wqe->wq->user->processes); worker->flags = 0; preempt_enable(); + if (worker->saved_creds) { + revert_creds(worker->saved_creds); + worker->cur_creds = worker->saved_creds = NULL; + } + raw_spin_lock_irq(&wqe->lock); - hlist_nulls_del_rcu(&worker->nulls_node); + if (flags & IO_WORKER_F_FREE) + hlist_nulls_del_rcu(&worker->nulls_node); list_del_rcu(&worker->all_list); - if (__io_worker_unuse(wqe, worker)) { - __release(&wqe->lock); - raw_spin_lock_irq(&wqe->lock); - } acct->nr_workers--; raw_spin_unlock_irq(&wqe->lock); kfree_rcu(worker, rcu); - if (refcount_dec_and_test(&wqe->wq->refs)) - complete(&wqe->wq->done); + io_wq_put(wqe->wq); } static inline bool io_wqe_run_queue(struct io_wqe *wqe) @@ -306,33 +246,27 @@ static void io_wqe_wake_worker(struct io_wqe *wqe, struct io_wqe_acct *acct) wake_up_process(wqe->wq->manager); } -static void io_wqe_inc_running(struct io_wqe *wqe, struct io_worker *worker) +static void io_wqe_inc_running(struct io_worker *worker) { - struct io_wqe_acct *acct = io_wqe_get_acct(wqe, worker); + struct io_wqe_acct *acct = io_wqe_get_acct(worker); atomic_inc(&acct->nr_running); } -static void io_wqe_dec_running(struct io_wqe *wqe, struct io_worker *worker) +static void io_wqe_dec_running(struct io_worker *worker) __must_hold(wqe->lock) { - struct io_wqe_acct *acct = io_wqe_get_acct(wqe, worker); + struct io_wqe_acct *acct = io_wqe_get_acct(worker); + struct io_wqe *wqe = worker->wqe; if (atomic_dec_and_test(&acct->nr_running) && io_wqe_run_queue(wqe)) io_wqe_wake_worker(wqe, acct); } -static void io_worker_start(struct io_wqe *wqe, struct io_worker *worker) +static void io_worker_start(struct io_worker *worker) { - allow_kernel_signal(SIGINT); - - current->flags |= PF_IO_WORKER; - current->fs = NULL; - current->files = NULL; - worker->flags |= (IO_WORKER_F_UP | IO_WORKER_F_RUNNING); - worker->restore_nsproxy = current->nsproxy; - io_wqe_inc_running(wqe, worker); + io_wqe_inc_running(worker); } /* @@ -357,19 +291,17 @@ static void __io_worker_busy(struct io_wqe *wqe, struct io_worker *worker, worker_bound = (worker->flags & IO_WORKER_F_BOUND) != 0; work_bound = (work->flags & IO_WQ_WORK_UNBOUND) == 0; if (worker_bound != work_bound) { - io_wqe_dec_running(wqe, worker); + io_wqe_dec_running(worker); if (work_bound) { worker->flags |= IO_WORKER_F_BOUND; wqe->acct[IO_WQ_ACCT_UNBOUND].nr_workers--; wqe->acct[IO_WQ_ACCT_BOUND].nr_workers++; - atomic_dec(&wqe->wq->user->processes); } else { worker->flags &= ~IO_WORKER_F_BOUND; wqe->acct[IO_WQ_ACCT_UNBOUND].nr_workers++; wqe->acct[IO_WQ_ACCT_BOUND].nr_workers--; - atomic_inc(&wqe->wq->user->processes); } - io_wqe_inc_running(wqe, worker); + io_wqe_inc_running(worker); } } @@ -380,15 +312,17 @@ static void __io_worker_busy(struct io_wqe *wqe, struct io_worker *worker, * retry the loop in that case (we changed task state), we don't regrab * the lock if we return success. */ -static bool __io_worker_idle(struct io_wqe *wqe, struct io_worker *worker) +static void __io_worker_idle(struct io_wqe *wqe, struct io_worker *worker) __must_hold(wqe->lock) { if (!(worker->flags & IO_WORKER_F_FREE)) { worker->flags |= IO_WORKER_F_FREE; hlist_nulls_add_head_rcu(&worker->nulls_node, &wqe->free_list); } - - return __io_worker_unuse(wqe, worker); + if (worker->saved_creds) { + revert_creds(worker->saved_creds); + worker->cur_creds = worker->saved_creds = NULL; + } } static inline unsigned int io_get_work_hash(struct io_wq_work *work) @@ -396,14 +330,31 @@ static inline unsigned int io_get_work_hash(struct io_wq_work *work) return work->flags >> IO_WQ_HASH_SHIFT; } +static void io_wait_on_hash(struct io_wqe *wqe, unsigned int hash) +{ + struct io_wq *wq = wqe->wq; + + spin_lock(&wq->hash->wait.lock); + if (list_empty(&wqe->wait.entry)) { + __add_wait_queue(&wq->hash->wait, &wqe->wait); + if (!test_bit(hash, &wq->hash->map)) { + __set_current_state(TASK_RUNNING); + list_del_init(&wqe->wait.entry); + } + } + spin_unlock(&wq->hash->wait.lock); +} + static struct io_wq_work *io_get_next_work(struct io_wqe *wqe) __must_hold(wqe->lock) { struct io_wq_work_node *node, *prev; struct io_wq_work *work, *tail; - unsigned int hash; + unsigned int stall_hash = -1U; wq_list_for_each(node, prev, &wqe->work_list) { + unsigned int hash; + work = container_of(node, struct io_wq_work, list); /* not hashed, can run anytime */ @@ -412,111 +363,60 @@ static struct io_wq_work *io_get_next_work(struct io_wqe *wqe) return work; } - /* hashed, can run if not already running */ hash = io_get_work_hash(work); - if (!(wqe->hash_map & BIT(hash))) { - wqe->hash_map |= BIT(hash); - /* all items with this hash lie in [work, tail] */ - tail = wqe->hash_tail[hash]; + /* all items with this hash lie in [work, tail] */ + tail = wqe->hash_tail[hash]; + + /* hashed, can run if not already running */ + if (!test_and_set_bit(hash, &wqe->wq->hash->map)) { wqe->hash_tail[hash] = NULL; wq_list_cut(&wqe->work_list, &tail->list, prev); return work; } + if (stall_hash == -1U) + stall_hash = hash; + /* fast forward to a next hash, for-each will fix up @prev */ + node = &tail->list; } - return NULL; -} - -static void io_wq_switch_mm(struct io_worker *worker, struct io_wq_work *work) -{ - if (worker->mm) { - kthread_unuse_mm(worker->mm); - mmput(worker->mm); - worker->mm = NULL; + if (stall_hash != -1U) { + raw_spin_unlock(&wqe->lock); + io_wait_on_hash(wqe, stall_hash); + raw_spin_lock(&wqe->lock); } - if (mmget_not_zero(work->identity->mm)) { - kthread_use_mm(work->identity->mm); - worker->mm = work->identity->mm; - return; - } - - /* failed grabbing mm, ensure work gets cancelled */ - work->flags |= IO_WQ_WORK_CANCEL; + return NULL; } -static inline void io_wq_switch_blkcg(struct io_worker *worker, - struct io_wq_work *work) +static void io_flush_signals(void) { -#ifdef CONFIG_BLK_CGROUP - if (!(work->flags & IO_WQ_WORK_BLKCG)) - return; - if (work->identity->blkcg_css != worker->blkcg_css) { - kthread_associate_blkcg(work->identity->blkcg_css); - worker->blkcg_css = work->identity->blkcg_css; + if (unlikely(test_tsk_thread_flag(current, TIF_NOTIFY_SIGNAL))) { + if (current->task_works) + task_work_run(); + clear_tsk_thread_flag(current, TIF_NOTIFY_SIGNAL); } -#endif } static void io_wq_switch_creds(struct io_worker *worker, struct io_wq_work *work) { - const struct cred *old_creds = override_creds(work->identity->creds); + const struct cred *old_creds = override_creds(work->creds); - worker->cur_creds = work->identity->creds; + worker->cur_creds = work->creds; if (worker->saved_creds) put_cred(old_creds); /* creds set by previous switch */ else worker->saved_creds = old_creds; } -static void io_impersonate_work(struct io_worker *worker, - struct io_wq_work *work) -{ - if ((work->flags & IO_WQ_WORK_FILES) && - current->files != work->identity->files) { - task_lock(current); - current->files = work->identity->files; - current->nsproxy = work->identity->nsproxy; - task_unlock(current); - if (!work->identity->files) { - /* failed grabbing files, ensure work gets cancelled */ - work->flags |= IO_WQ_WORK_CANCEL; - } - } - if ((work->flags & IO_WQ_WORK_FS) && current->fs != work->identity->fs) - current->fs = work->identity->fs; - if ((work->flags & IO_WQ_WORK_MM) && work->identity->mm != worker->mm) - io_wq_switch_mm(worker, work); - if ((work->flags & IO_WQ_WORK_CREDS) && - worker->cur_creds != work->identity->creds) - io_wq_switch_creds(worker, work); - if (work->flags & IO_WQ_WORK_FSIZE) - current->signal->rlim[RLIMIT_FSIZE].rlim_cur = work->identity->fsize; - else if (current->signal->rlim[RLIMIT_FSIZE].rlim_cur != RLIM_INFINITY) - current->signal->rlim[RLIMIT_FSIZE].rlim_cur = RLIM_INFINITY; - io_wq_switch_blkcg(worker, work); -#ifdef CONFIG_AUDIT - current->loginuid = work->identity->loginuid; - current->sessionid = work->identity->sessionid; -#endif -} - static void io_assign_current_work(struct io_worker *worker, struct io_wq_work *work) { if (work) { - /* flush pending signals before assigning new work */ - if (signal_pending(current)) - flush_signals(current); + io_flush_signals(); cond_resched(); } -#ifdef CONFIG_AUDIT - current->loginuid = KUIDT_INIT(AUDIT_UID_UNSET); - current->sessionid = AUDIT_SID_UNSET; -#endif - spin_lock_irq(&worker->lock); worker->cur_work = work; spin_unlock_irq(&worker->lock); @@ -550,6 +450,7 @@ get_next: if (!work) break; io_assign_current_work(worker, work); + __set_current_state(TASK_RUNNING); /* handle a whole dependent link */ do { @@ -557,7 +458,8 @@ get_next: unsigned int hash = io_get_work_hash(work); next_hashed = wq_next_work(work); - io_impersonate_work(worker, work); + if (work->creds && worker->cur_creds != work->creds) + io_wq_switch_creds(worker, work); wq->do_work(work); io_assign_current_work(worker, NULL); @@ -572,8 +474,10 @@ get_next: io_wqe_enqueue(wqe, linked); if (hash != -1U && !next_hashed) { + clear_bit(hash, &wq->hash->map); + if (wq_has_sleeper(&wq->hash->wait)) + wake_up(&wq->hash->wait); raw_spin_lock_irq(&wqe->lock); - wqe->hash_map &= ~BIT_ULL(hash); wqe->flags &= ~IO_WQE_FLAG_STALLED; /* skip unnecessary unlock-lock wqe->lock */ if (!work) @@ -592,27 +496,23 @@ static int io_wqe_worker(void *data) struct io_wqe *wqe = worker->wqe; struct io_wq *wq = wqe->wq; - io_worker_start(wqe, worker); + io_worker_start(worker); while (!test_bit(IO_WQ_BIT_EXIT, &wq->state)) { set_current_state(TASK_INTERRUPTIBLE); loop: raw_spin_lock_irq(&wqe->lock); if (io_wqe_run_queue(wqe)) { - __set_current_state(TASK_RUNNING); io_worker_handle_work(worker); goto loop; } - /* drops the lock on success, retry */ - if (__io_worker_idle(wqe, worker)) { - __release(&wqe->lock); - goto loop; - } + __io_worker_idle(wqe, worker); raw_spin_unlock_irq(&wqe->lock); - if (signal_pending(current)) - flush_signals(current); + io_flush_signals(); if (schedule_timeout(WORKER_IDLE_TIMEOUT)) continue; + if (fatal_signal_pending(current)) + break; /* timed out, exit unless we're the fixed worker */ if (test_bit(IO_WQ_BIT_EXIT, &wq->state) || !(worker->flags & IO_WORKER_F_FIXED)) @@ -636,15 +536,16 @@ loop: */ void io_wq_worker_running(struct task_struct *tsk) { - struct io_worker *worker = kthread_data(tsk); - struct io_wqe *wqe = worker->wqe; + struct io_worker *worker = tsk->pf_io_worker; + if (!worker) + return; if (!(worker->flags & IO_WORKER_F_UP)) return; if (worker->flags & IO_WORKER_F_RUNNING) return; worker->flags |= IO_WORKER_F_RUNNING; - io_wqe_inc_running(wqe, worker); + io_wqe_inc_running(worker); } /* @@ -654,9 +555,10 @@ void io_wq_worker_running(struct task_struct *tsk) */ void io_wq_worker_sleeping(struct task_struct *tsk) { - struct io_worker *worker = kthread_data(tsk); - struct io_wqe *wqe = worker->wqe; + struct io_worker *worker = tsk->pf_io_worker; + if (!worker) + return; if (!(worker->flags & IO_WORKER_F_UP)) return; if (!(worker->flags & IO_WORKER_F_RUNNING)) @@ -664,32 +566,27 @@ void io_wq_worker_sleeping(struct task_struct *tsk) worker->flags &= ~IO_WORKER_F_RUNNING; - raw_spin_lock_irq(&wqe->lock); - io_wqe_dec_running(wqe, worker); - raw_spin_unlock_irq(&wqe->lock); + raw_spin_lock_irq(&worker->wqe->lock); + io_wqe_dec_running(worker); + raw_spin_unlock_irq(&worker->wqe->lock); } -static bool create_io_worker(struct io_wq *wq, struct io_wqe *wqe, int index) +static int task_thread(void *data, int index) { + struct io_worker *worker = data; + struct io_wqe *wqe = worker->wqe; struct io_wqe_acct *acct = &wqe->acct[index]; - struct io_worker *worker; + struct io_wq *wq = wqe->wq; + char buf[TASK_COMM_LEN]; - worker = kzalloc_node(sizeof(*worker), GFP_KERNEL, wqe->node); - if (!worker) - return false; + sprintf(buf, "iou-wrk-%d", wq->task_pid); + set_task_comm(current, buf); - refcount_set(&worker->ref, 1); - worker->nulls_node.pprev = NULL; - worker->wqe = wqe; - spin_lock_init(&worker->lock); + current->pf_io_worker = worker; + worker->task = current; - worker->task = kthread_create_on_node(io_wqe_worker, worker, wqe->node, - "io_wqe_worker-%d/%d", index, wqe->node); - if (IS_ERR(worker->task)) { - kfree(worker); - return false; - } - kthread_bind_mask(worker->task, cpumask_of_node(wqe->node)); + set_cpus_allowed_ptr(current, cpumask_of_node(wqe->node)); + current->flags |= PF_NO_SETAFFINITY; raw_spin_lock_irq(&wqe->lock); hlist_nulls_add_head_rcu(&worker->nulls_node, &wqe->free_list); @@ -702,11 +599,63 @@ static bool create_io_worker(struct io_wq *wq, struct io_wqe *wqe, int index) acct->nr_workers++; raw_spin_unlock_irq(&wqe->lock); - if (index == IO_WQ_ACCT_UNBOUND) - atomic_inc(&wq->user->processes); + io_wqe_worker(data); + do_exit(0); +} + +static int task_thread_bound(void *data) +{ + return task_thread(data, IO_WQ_ACCT_BOUND); +} + +static int task_thread_unbound(void *data) +{ + return task_thread(data, IO_WQ_ACCT_UNBOUND); +} + +pid_t io_wq_fork_thread(int (*fn)(void *), void *arg) +{ + unsigned long flags = CLONE_FS|CLONE_FILES|CLONE_SIGHAND|CLONE_THREAD| + CLONE_IO|SIGCHLD; + struct kernel_clone_args args = { + .flags = ((lower_32_bits(flags) | CLONE_VM | + CLONE_UNTRACED) & ~CSIGNAL), + .exit_signal = (lower_32_bits(flags) & CSIGNAL), + .stack = (unsigned long)fn, + .stack_size = (unsigned long)arg, + }; + + return kernel_clone(&args); +} + +static bool create_io_worker(struct io_wq *wq, struct io_wqe *wqe, int index) +{ + struct io_worker *worker; + pid_t pid; + + __set_current_state(TASK_RUNNING); + + worker = kzalloc_node(sizeof(*worker), GFP_KERNEL, wqe->node); + if (!worker) + return false; + + refcount_set(&worker->ref, 1); + worker->nulls_node.pprev = NULL; + worker->wqe = wqe; + spin_lock_init(&worker->lock); + init_completion(&worker->ref_done); refcount_inc(&wq->refs); - wake_up_process(worker->task); + + if (index == IO_WQ_ACCT_BOUND) + pid = io_wq_fork_thread(task_thread_bound, worker); + else + pid = io_wq_fork_thread(task_thread_unbound, worker); + if (pid < 0) { + io_wq_put(wq); + kfree(worker); + return false; + } return true; } @@ -752,93 +701,57 @@ static bool io_wq_worker_wake(struct io_worker *worker, void *data) return false; } -/* - * Manager thread. Tasked with creating new workers, if we need them. - */ -static int io_wq_manager(void *data) +static void io_wq_check_workers(struct io_wq *wq) { - struct io_wq *wq = data; int node; - /* create fixed workers */ - refcount_set(&wq->refs, 1); for_each_node(node) { + struct io_wqe *wqe = wq->wqes[node]; + bool fork_worker[2] = { false, false }; + if (!node_online(node)) continue; - if (create_io_worker(wq, wq->wqes[node], IO_WQ_ACCT_BOUND)) - continue; - set_bit(IO_WQ_BIT_ERROR, &wq->state); - set_bit(IO_WQ_BIT_EXIT, &wq->state); - goto out; - } - - complete(&wq->done); - - while (!kthread_should_stop()) { - if (current->task_works) - task_work_run(); - - for_each_node(node) { - struct io_wqe *wqe = wq->wqes[node]; - bool fork_worker[2] = { false, false }; - - if (!node_online(node)) - continue; - - raw_spin_lock_irq(&wqe->lock); - if (io_wqe_need_worker(wqe, IO_WQ_ACCT_BOUND)) - fork_worker[IO_WQ_ACCT_BOUND] = true; - if (io_wqe_need_worker(wqe, IO_WQ_ACCT_UNBOUND)) - fork_worker[IO_WQ_ACCT_UNBOUND] = true; - raw_spin_unlock_irq(&wqe->lock); - if (fork_worker[IO_WQ_ACCT_BOUND]) - create_io_worker(wq, wqe, IO_WQ_ACCT_BOUND); - if (fork_worker[IO_WQ_ACCT_UNBOUND]) - create_io_worker(wq, wqe, IO_WQ_ACCT_UNBOUND); - } - set_current_state(TASK_INTERRUPTIBLE); - schedule_timeout(HZ); - } - - if (current->task_works) - task_work_run(); -out: - if (refcount_dec_and_test(&wq->refs)) { - complete(&wq->done); - return 0; - } - /* if ERROR is set and we get here, we have workers to wake */ - if (test_bit(IO_WQ_BIT_ERROR, &wq->state)) { - rcu_read_lock(); - for_each_node(node) - io_wq_for_each_worker(wq->wqes[node], io_wq_worker_wake, NULL); - rcu_read_unlock(); + raw_spin_lock_irq(&wqe->lock); + if (io_wqe_need_worker(wqe, IO_WQ_ACCT_BOUND)) + fork_worker[IO_WQ_ACCT_BOUND] = true; + if (io_wqe_need_worker(wqe, IO_WQ_ACCT_UNBOUND)) + fork_worker[IO_WQ_ACCT_UNBOUND] = true; + raw_spin_unlock_irq(&wqe->lock); + if (fork_worker[IO_WQ_ACCT_BOUND]) + create_io_worker(wq, wqe, IO_WQ_ACCT_BOUND); + if (fork_worker[IO_WQ_ACCT_UNBOUND]) + create_io_worker(wq, wqe, IO_WQ_ACCT_UNBOUND); } - return 0; } -static bool io_wq_can_queue(struct io_wqe *wqe, struct io_wqe_acct *acct, - struct io_wq_work *work) +/* + * Manager thread. Tasked with creating new workers, if we need them. + */ +static int io_wq_manager(void *data) { - bool free_worker; - - if (!(work->flags & IO_WQ_WORK_UNBOUND)) - return true; - if (atomic_read(&acct->nr_running)) - return true; + struct io_wq *wq = data; + char buf[TASK_COMM_LEN]; - rcu_read_lock(); - free_worker = !hlist_nulls_empty(&wqe->free_list); - rcu_read_unlock(); - if (free_worker) - return true; + sprintf(buf, "iou-mgr-%d", wq->task_pid); + set_task_comm(current, buf); + current->flags |= PF_IO_WORKER; + wq->manager = current; - if (atomic_read(&wqe->wq->user->processes) >= acct->max_workers && - !(capable(CAP_SYS_RESOURCE) || capable(CAP_SYS_ADMIN))) - return false; + complete(&wq->done); - return true; + do { + set_current_state(TASK_INTERRUPTIBLE); + io_wq_check_workers(wq); + schedule_timeout(HZ); + if (fatal_signal_pending(current)) + set_bit(IO_WQ_BIT_EXIT, &wq->state); + } while (!test_bit(IO_WQ_BIT_EXIT, &wq->state)); + + io_wq_check_workers(wq); + wq->manager = NULL; + io_wq_put(wq); + do_exit(0); } static void io_run_cancel(struct io_wq_work *work, struct io_wqe *wqe) @@ -872,20 +785,37 @@ append: wq_list_add_after(&work->list, &tail->list, &wqe->work_list); } +static int io_wq_fork_manager(struct io_wq *wq) +{ + int ret; + + if (wq->manager) + return 0; + + clear_bit(IO_WQ_BIT_EXIT, &wq->state); + refcount_inc(&wq->refs); + current->flags |= PF_IO_WORKER; + ret = io_wq_fork_thread(io_wq_manager, wq); + current->flags &= ~PF_IO_WORKER; + if (ret >= 0) { + wait_for_completion(&wq->done); + return 0; + } + + io_wq_put(wq); + return ret; +} + static void io_wqe_enqueue(struct io_wqe *wqe, struct io_wq_work *work) { struct io_wqe_acct *acct = io_work_get_acct(wqe, work); int work_flags; unsigned long flags; - /* - * Do early check to see if we need a new unbound worker, and if we do, - * if we're allowed to do so. This isn't 100% accurate as there's a - * gap between this check and incrementing the value, but that's OK. - * It's close enough to not be an issue, fork() has the same delay. - */ - if (unlikely(!io_wq_can_queue(wqe, acct, work))) { - io_run_cancel(work, wqe); + /* Can only happen if manager creation fails after exec */ + if (unlikely(io_wq_fork_manager(wqe->wq))) { + work->flags |= IO_WQ_WORK_CANCEL; + wqe->wq->do_work(work); return; } @@ -939,7 +869,7 @@ static bool io_wq_worker_cancel(struct io_worker *worker, void *data) spin_lock_irqsave(&worker->lock, flags); if (worker->cur_work && match->fn(worker->cur_work, match->data)) { - send_sig(SIGINT, worker->task, 1); + set_notify_signal(worker->task); match->nr_running++; } spin_unlock_irqrestore(&worker->lock, flags); @@ -1043,6 +973,24 @@ enum io_wq_cancel io_wq_cancel_cb(struct io_wq *wq, work_cancel_fn *cancel, return IO_WQ_CANCEL_NOTFOUND; } +static int io_wqe_hash_wake(struct wait_queue_entry *wait, unsigned mode, + int sync, void *key) +{ + struct io_wqe *wqe = container_of(wait, struct io_wqe, wait); + int ret; + + list_del_init(&wait->entry); + + rcu_read_lock(); + ret = io_wqe_activate_free_worker(wqe); + rcu_read_unlock(); + + if (!ret) + wake_up_process(wqe->wq->manager); + + return 1; +} + struct io_wq *io_wq_create(unsigned bounded, struct io_wq_data *data) { int ret = -ENOMEM, node; @@ -1063,12 +1011,11 @@ struct io_wq *io_wq_create(unsigned bounded, struct io_wq_data *data) if (ret) goto err_wqes; + refcount_inc(&data->hash->refs); + wq->hash = data->hash; wq->free_work = data->free_work; wq->do_work = data->do_work; - /* caller must already hold a reference to this */ - wq->user = data->user; - ret = -ENOMEM; for_each_node(node) { struct io_wqe *wqe; @@ -1083,11 +1030,11 @@ struct io_wq *io_wq_create(unsigned bounded, struct io_wq_data *data) wqe->node = alloc_node; wqe->acct[IO_WQ_ACCT_BOUND].max_workers = bounded; atomic_set(&wqe->acct[IO_WQ_ACCT_BOUND].nr_running, 0); - if (wq->user) { - wqe->acct[IO_WQ_ACCT_UNBOUND].max_workers = + wqe->acct[IO_WQ_ACCT_UNBOUND].max_workers = task_rlimit(current, RLIMIT_NPROC); - } atomic_set(&wqe->acct[IO_WQ_ACCT_UNBOUND].nr_running, 0); + wqe->wait.func = io_wqe_hash_wake; + INIT_LIST_HEAD(&wqe->wait.entry); wqe->wq = wq; raw_spin_lock_init(&wqe->lock); INIT_WQ_LIST(&wqe->work_list); @@ -1095,23 +1042,16 @@ struct io_wq *io_wq_create(unsigned bounded, struct io_wq_data *data) INIT_LIST_HEAD(&wqe->all_list); } + wq->task_pid = current->pid; init_completion(&wq->done); + refcount_set(&wq->refs, 1); - wq->manager = kthread_create(io_wq_manager, wq, "io_wq_manager"); - if (!IS_ERR(wq->manager)) { - wake_up_process(wq->manager); - wait_for_completion(&wq->done); - if (test_bit(IO_WQ_BIT_ERROR, &wq->state)) { - ret = -ENOMEM; - goto err; - } - refcount_set(&wq->use_refs, 1); - reinit_completion(&wq->done); + ret = io_wq_fork_manager(wq); + if (!ret) return wq; - } - ret = PTR_ERR(wq->manager); - complete(&wq->done); + io_wq_put(wq); + io_wq_put_hash(data->hash); err: cpuhp_state_remove_instance_nocalls(io_wq_online, &wq->cpuhp_node); for_each_node(node) @@ -1123,15 +1063,7 @@ err_wq: return ERR_PTR(ret); } -bool io_wq_get(struct io_wq *wq, struct io_wq_data *data) -{ - if (data->free_work != wq->free_work || data->do_work != wq->do_work) - return false; - - return refcount_inc_not_zero(&wq->use_refs); -} - -static void __io_wq_destroy(struct io_wq *wq) +static void io_wq_destroy(struct io_wq *wq) { int node; @@ -1139,30 +1071,31 @@ static void __io_wq_destroy(struct io_wq *wq) set_bit(IO_WQ_BIT_EXIT, &wq->state); if (wq->manager) - kthread_stop(wq->manager); + wake_up_process(wq->manager); rcu_read_lock(); for_each_node(node) io_wq_for_each_worker(wq->wqes[node], io_wq_worker_wake, NULL); rcu_read_unlock(); - wait_for_completion(&wq->done); + spin_lock_irq(&wq->hash->wait.lock); + for_each_node(node) { + struct io_wqe *wqe = wq->wqes[node]; - for_each_node(node) - kfree(wq->wqes[node]); + list_del_init(&wqe->wait.entry); + kfree(wqe); + } + spin_unlock_irq(&wq->hash->wait.lock); + io_wq_put_hash(wq->hash); kfree(wq->wqes); kfree(wq); -} -void io_wq_destroy(struct io_wq *wq) -{ - if (refcount_dec_and_test(&wq->use_refs)) - __io_wq_destroy(wq); } -struct task_struct *io_wq_get_task(struct io_wq *wq) +void io_wq_put(struct io_wq *wq) { - return wq->manager; + if (refcount_dec_and_test(&wq->refs)) + io_wq_destroy(wq); } static bool io_wq_worker_affinity(struct io_worker *worker, void *data) diff --git a/fs/io-wq.h b/fs/io-wq.h index 096f1021018e..b6ca12b60c35 100644 --- a/fs/io-wq.h +++ b/fs/io-wq.h @@ -1,6 +1,7 @@ #ifndef INTERNAL_IO_WQ_H #define INTERNAL_IO_WQ_H +#include <linux/refcount.h> #include <linux/io_uring.h> struct io_wq; @@ -11,13 +12,6 @@ enum { IO_WQ_WORK_UNBOUND = 4, IO_WQ_WORK_CONCURRENT = 16, - IO_WQ_WORK_FILES = 32, - IO_WQ_WORK_FS = 64, - IO_WQ_WORK_MM = 128, - IO_WQ_WORK_CREDS = 256, - IO_WQ_WORK_BLKCG = 512, - IO_WQ_WORK_FSIZE = 1024, - IO_WQ_HASH_SHIFT = 24, /* upper 8 bits are used for hash key */ }; @@ -85,7 +79,7 @@ static inline void wq_list_del(struct io_wq_work_list *list, struct io_wq_work { struct io_wq_work_node list; - struct io_identity *identity; + const struct cred *creds; unsigned flags; }; @@ -100,20 +94,32 @@ static inline struct io_wq_work *wq_next_work(struct io_wq_work *work) typedef struct io_wq_work *(free_work_fn)(struct io_wq_work *); typedef void (io_wq_work_fn)(struct io_wq_work *); -struct io_wq_data { - struct user_struct *user; +struct io_wq_hash { + refcount_t refs; + unsigned long map; + struct wait_queue_head wait; +}; +static inline void io_wq_put_hash(struct io_wq_hash *hash) +{ + if (refcount_dec_and_test(&hash->refs)) + kfree(hash); +} + +struct io_wq_data { + struct io_wq_hash *hash; io_wq_work_fn *do_work; free_work_fn *free_work; }; struct io_wq *io_wq_create(unsigned bounded, struct io_wq_data *data); -bool io_wq_get(struct io_wq *wq, struct io_wq_data *data); -void io_wq_destroy(struct io_wq *wq); +void io_wq_put(struct io_wq *wq); void io_wq_enqueue(struct io_wq *wq, struct io_wq_work *work); void io_wq_hash_work(struct io_wq_work *work, void *val); +pid_t io_wq_fork_thread(int (*fn)(void *), void *arg); + static inline bool io_wq_is_hashed(struct io_wq_work *work) { return work->flags & IO_WQ_WORK_HASHED; @@ -124,8 +130,6 @@ typedef bool (work_cancel_fn)(struct io_wq_work *, void *); enum io_wq_cancel io_wq_cancel_cb(struct io_wq *wq, work_cancel_fn *cancel, void *data, bool cancel_all); -struct task_struct *io_wq_get_task(struct io_wq *wq); - #if defined(CONFIG_IO_WQ) extern void io_wq_worker_sleeping(struct task_struct *); extern void io_wq_worker_running(struct task_struct *); @@ -140,6 +144,7 @@ static inline void io_wq_worker_running(struct task_struct *tsk) static inline bool io_wq_current_is_worker(void) { - return in_task() && (current->flags & PF_IO_WORKER); + return in_task() && (current->flags & PF_IO_WORKER) && + current->pf_io_worker; } #endif diff --git a/fs/io_uring.c b/fs/io_uring.c index c9f5f295c2ac..4a088581b0f2 100644 --- a/fs/io_uring.c +++ b/fs/io_uring.c @@ -57,7 +57,6 @@ #include <linux/mman.h> #include <linux/percpu.h> #include <linux/slab.h> -#include <linux/kthread.h> #include <linux/blkdev.h> #include <linux/bvec.h> #include <linux/net.h> @@ -254,6 +253,11 @@ struct io_restriction { bool registered; }; +enum { + IO_SQ_THREAD_SHOULD_STOP = 0, + IO_SQ_THREAD_SHOULD_PARK, +}; + struct io_sq_data { refcount_t refs; struct mutex lock; @@ -267,6 +271,13 @@ struct io_sq_data { struct wait_queue_head wait; unsigned sq_thread_idle; + int sq_cpu; + pid_t task_pid; + + unsigned long state; + struct completion startup; + struct completion completion; + struct completion exited; }; #define IO_IOPOLL_BATCH 8 @@ -323,12 +334,12 @@ struct io_ring_ctx { struct { unsigned int flags; unsigned int compat: 1; - unsigned int limit_mem: 1; unsigned int cq_overflow_flushed: 1; unsigned int drain_next: 1; unsigned int eventfd_async: 1; unsigned int restricted: 1; unsigned int sqo_dead: 1; + unsigned int sqo_exec: 1; /* * Ring buffer of indices into array of io_uring_sqe, which is @@ -350,6 +361,9 @@ struct io_ring_ctx { unsigned cached_cq_overflow; unsigned long sq_check_overflow; + /* hashed buffered write serialization */ + struct io_wq_hash *hash_map; + struct list_head defer_list; struct list_head timeout_list; struct list_head cq_overflow_list; @@ -366,22 +380,14 @@ struct io_ring_ctx { struct io_rings *rings; - /* IO offload */ - struct io_wq *io_wq; - /* - * For SQPOLL usage - we hold a reference to the parent task, so we - * have access to the ->files + * For SQPOLL usage */ struct task_struct *sqo_task; /* Only used for accounting purposes */ struct mm_struct *mm_account; -#ifdef CONFIG_BLK_CGROUP - struct cgroup_subsys_state *sqo_blkcg_css; -#endif - struct io_sq_data *sq_data; /* if using sq thread polling */ struct wait_queue_head sqo_sq_wait; @@ -401,13 +407,6 @@ struct io_ring_ctx { struct user_struct *user; - const struct cred *creds; - -#ifdef CONFIG_AUDIT - kuid_t loginuid; - unsigned int sessionid; -#endif - struct completion ref_comp; struct completion sq_thread_comp; @@ -456,6 +455,11 @@ struct io_ring_ctx { struct io_restriction restrictions; + /* exit task_work */ + struct callback_head *exit_task_work; + + struct wait_queue_head hash_wait; + /* Keep this last, we don't need it for the fast path */ struct work_struct exit_work; }; @@ -838,7 +842,6 @@ struct io_op_def { unsigned plug : 1; /* size of async data needed, if any */ unsigned short async_size; - unsigned work_flags; }; static const struct io_op_def io_op_defs[] = { @@ -851,7 +854,6 @@ static const struct io_op_def io_op_defs[] = { .needs_async_data = 1, .plug = 1, .async_size = sizeof(struct io_async_rw), - .work_flags = IO_WQ_WORK_MM | IO_WQ_WORK_BLKCG, }, [IORING_OP_WRITEV] = { .needs_file = 1, @@ -861,12 +863,9 @@ static const struct io_op_def io_op_defs[] = { .needs_async_data = 1, .plug = 1, .async_size = sizeof(struct io_async_rw), - .work_flags = IO_WQ_WORK_MM | IO_WQ_WORK_BLKCG | - IO_WQ_WORK_FSIZE, }, [IORING_OP_FSYNC] = { .needs_file = 1, - .work_flags = IO_WQ_WORK_BLKCG, }, [IORING_OP_READ_FIXED] = { .needs_file = 1, @@ -874,7 +873,6 @@ static const struct io_op_def io_op_defs[] = { .pollin = 1, .plug = 1, .async_size = sizeof(struct io_async_rw), - .work_flags = IO_WQ_WORK_BLKCG | IO_WQ_WORK_MM, }, [IORING_OP_WRITE_FIXED] = { .needs_file = 1, @@ -883,8 +881,6 @@ static const struct io_op_def io_op_defs[] = { .pollout = 1, .plug = 1, .async_size = sizeof(struct io_async_rw), - .work_flags = IO_WQ_WORK_BLKCG | IO_WQ_WORK_FSIZE | - IO_WQ_WORK_MM, }, [IORING_OP_POLL_ADD] = { .needs_file = 1, @@ -893,7 +889,6 @@ static const struct io_op_def io_op_defs[] = { [IORING_OP_POLL_REMOVE] = {}, [IORING_OP_SYNC_FILE_RANGE] = { .needs_file = 1, - .work_flags = IO_WQ_WORK_BLKCG, }, [IORING_OP_SENDMSG] = { .needs_file = 1, @@ -901,8 +896,6 @@ static const struct io_op_def io_op_defs[] = { .pollout = 1, .needs_async_data = 1, .async_size = sizeof(struct io_async_msghdr), - .work_flags = IO_WQ_WORK_MM | IO_WQ_WORK_BLKCG | - IO_WQ_WORK_FS, }, [IORING_OP_RECVMSG] = { .needs_file = 1, @@ -911,29 +904,23 @@ static const struct io_op_def io_op_defs[] = { .buffer_select = 1, .needs_async_data = 1, .async_size = sizeof(struct io_async_msghdr), - .work_flags = IO_WQ_WORK_MM | IO_WQ_WORK_BLKCG | - IO_WQ_WORK_FS, }, [IORING_OP_TIMEOUT] = { .needs_async_data = 1, .async_size = sizeof(struct io_timeout_data), - .work_flags = IO_WQ_WORK_MM, }, [IORING_OP_TIMEOUT_REMOVE] = { /* used by timeout updates' prep() */ - .work_flags = IO_WQ_WORK_MM, }, [IORING_OP_ACCEPT] = { .needs_file = 1, .unbound_nonreg_file = 1, .pollin = 1, - .work_flags = IO_WQ_WORK_MM | IO_WQ_WORK_FILES, }, [IORING_OP_ASYNC_CANCEL] = {}, [IORING_OP_LINK_TIMEOUT] = { .needs_async_data = 1, .async_size = sizeof(struct io_timeout_data), - .work_flags = IO_WQ_WORK_MM, }, [IORING_OP_CONNECT] = { .needs_file = 1, @@ -941,26 +928,14 @@ static const struct io_op_def io_op_defs[] = { .pollout = 1, .needs_async_data = 1, .async_size = sizeof(struct io_async_connect), - .work_flags = IO_WQ_WORK_MM, }, [IORING_OP_FALLOCATE] = { .needs_file = 1, - .work_flags = IO_WQ_WORK_BLKCG | IO_WQ_WORK_FSIZE, - }, - [IORING_OP_OPENAT] = { - .work_flags = IO_WQ_WORK_FILES | IO_WQ_WORK_BLKCG | - IO_WQ_WORK_FS | IO_WQ_WORK_MM, - }, - [IORING_OP_CLOSE] = { - .work_flags = IO_WQ_WORK_FILES | IO_WQ_WORK_BLKCG, - }, - [IORING_OP_FILES_UPDATE] = { - .work_flags = IO_WQ_WORK_FILES | IO_WQ_WORK_MM, - }, - [IORING_OP_STATX] = { - .work_flags = IO_WQ_WORK_FILES | IO_WQ_WORK_MM | - IO_WQ_WORK_FS | IO_WQ_WORK_BLKCG, }, + [IORING_OP_OPENAT] = {}, + [IORING_OP_CLOSE] = {}, + [IORING_OP_FILES_UPDATE] = {}, + [IORING_OP_STATX] = {}, [IORING_OP_READ] = { .needs_file = 1, .unbound_nonreg_file = 1, @@ -968,7 +943,6 @@ static const struct io_op_def io_op_defs[] = { .buffer_select = 1, .plug = 1, .async_size = sizeof(struct io_async_rw), - .work_flags = IO_WQ_WORK_MM | IO_WQ_WORK_BLKCG, }, [IORING_OP_WRITE] = { .needs_file = 1, @@ -976,42 +950,31 @@ static const struct io_op_def io_op_defs[] = { .pollout = 1, .plug = 1, .async_size = sizeof(struct io_async_rw), - .work_flags = IO_WQ_WORK_MM | IO_WQ_WORK_BLKCG | - IO_WQ_WORK_FSIZE, }, [IORING_OP_FADVISE] = { .needs_file = 1, - .work_flags = IO_WQ_WORK_BLKCG, - }, - [IORING_OP_MADVISE] = { - .work_flags = IO_WQ_WORK_MM | IO_WQ_WORK_BLKCG, }, + [IORING_OP_MADVISE] = {}, [IORING_OP_SEND] = { .needs_file = 1, .unbound_nonreg_file = 1, .pollout = 1, - .work_flags = IO_WQ_WORK_MM | IO_WQ_WORK_BLKCG, }, [IORING_OP_RECV] = { .needs_file = 1, .unbound_nonreg_file = 1, .pollin = 1, .buffer_select = 1, - .work_flags = IO_WQ_WORK_MM | IO_WQ_WORK_BLKCG, }, [IORING_OP_OPENAT2] = { - .work_flags = IO_WQ_WORK_FILES | IO_WQ_WORK_FS | - IO_WQ_WORK_BLKCG | IO_WQ_WORK_MM, }, [IORING_OP_EPOLL_CTL] = { .unbound_nonreg_file = 1, - .work_flags = IO_WQ_WORK_FILES, }, [IORING_OP_SPLICE] = { .needs_file = 1, .hash_reg_file = 1, .unbound_nonreg_file = 1, - .work_flags = IO_WQ_WORK_BLKCG, }, [IORING_OP_PROVIDE_BUFFERS] = {}, [IORING_OP_REMOVE_BUFFERS] = {}, @@ -1023,19 +986,14 @@ static const struct io_op_def io_op_defs[] = { [IORING_OP_SHUTDOWN] = { .needs_file = 1, }, - [IORING_OP_RENAMEAT] = { - .work_flags = IO_WQ_WORK_MM | IO_WQ_WORK_FILES | - IO_WQ_WORK_FS | IO_WQ_WORK_BLKCG, - }, - [IORING_OP_UNLINKAT] = { - .work_flags = IO_WQ_WORK_MM | IO_WQ_WORK_FILES | - IO_WQ_WORK_FS | IO_WQ_WORK_BLKCG, - }, + [IORING_OP_RENAMEAT] = {}, + [IORING_OP_UNLINKAT] = {}, }; static void io_uring_try_cancel_requests(struct io_ring_ctx *ctx, struct task_struct *task, struct files_struct *files); +static void io_uring_cancel_sqpoll(struct io_ring_ctx *ctx); static void destroy_fixed_rsrc_ref_node(struct fixed_rsrc_ref_node *ref_node); static struct fixed_rsrc_ref_node *alloc_fixed_rsrc_ref_node( struct io_ring_ctx *ctx); @@ -1126,161 +1084,18 @@ static bool io_match_task(struct io_kiocb *head, continue; if (req->file && req->file->f_op == &io_uring_fops) return true; - if ((req->work.flags & IO_WQ_WORK_FILES) && - req->work.identity->files == files) + if (req->task->files == files) return true; } return false; } -static void io_sq_thread_drop_mm_files(void) -{ - struct files_struct *files = current->files; - struct mm_struct *mm = current->mm; - - if (mm) { - kthread_unuse_mm(mm); - mmput(mm); - current->mm = NULL; - } - if (files) { - struct nsproxy *nsproxy = current->nsproxy; - - task_lock(current); - current->files = NULL; - current->nsproxy = NULL; - task_unlock(current); - put_files_struct(files); - put_nsproxy(nsproxy); - } -} - -static int __io_sq_thread_acquire_files(struct io_ring_ctx *ctx) -{ - if (!current->files) { - struct files_struct *files; - struct nsproxy *nsproxy; - - task_lock(ctx->sqo_task); - files = ctx->sqo_task->files; - if (!files) { - task_unlock(ctx->sqo_task); - return -EOWNERDEAD; - } - atomic_inc(&files->count); - get_nsproxy(ctx->sqo_task->nsproxy); - nsproxy = ctx->sqo_task->nsproxy; - task_unlock(ctx->sqo_task); - - task_lock(current); - current->files = files; - current->nsproxy = nsproxy; - task_unlock(current); - } - return 0; -} - -static int __io_sq_thread_acquire_mm(struct io_ring_ctx *ctx) -{ - struct mm_struct *mm; - - if (current->mm) - return 0; - - task_lock(ctx->sqo_task); - mm = ctx->sqo_task->mm; - if (unlikely(!mm || !mmget_not_zero(mm))) - mm = NULL; - task_unlock(ctx->sqo_task); - - if (mm) { - kthread_use_mm(mm); - return 0; - } - - return -EFAULT; -} - -static int __io_sq_thread_acquire_mm_files(struct io_ring_ctx *ctx, - struct io_kiocb *req) -{ - const struct io_op_def *def = &io_op_defs[req->opcode]; - int ret; - - if (def->work_flags & IO_WQ_WORK_MM) { - ret = __io_sq_thread_acquire_mm(ctx); - if (unlikely(ret)) - return ret; - } - - if (def->needs_file || (def->work_flags & IO_WQ_WORK_FILES)) { - ret = __io_sq_thread_acquire_files(ctx); - if (unlikely(ret)) - return ret; - } - - return 0; -} - -static inline int io_sq_thread_acquire_mm_files(struct io_ring_ctx *ctx, - struct io_kiocb *req) -{ - if (!(ctx->flags & IORING_SETUP_SQPOLL)) - return 0; - return __io_sq_thread_acquire_mm_files(ctx, req); -} - -static void io_sq_thread_associate_blkcg(struct io_ring_ctx *ctx, - struct cgroup_subsys_state **cur_css) - -{ -#ifdef CONFIG_BLK_CGROUP - /* puts the old one when swapping */ - if (*cur_css != ctx->sqo_blkcg_css) { - kthread_associate_blkcg(ctx->sqo_blkcg_css); - *cur_css = ctx->sqo_blkcg_css; - } -#endif -} - -static void io_sq_thread_unassociate_blkcg(void) -{ -#ifdef CONFIG_BLK_CGROUP - kthread_associate_blkcg(NULL); -#endif -} - static inline void req_set_fail_links(struct io_kiocb *req) { if ((req->flags & (REQ_F_LINK | REQ_F_HARDLINK)) == REQ_F_LINK) req->flags |= REQ_F_FAIL_LINK; } -/* - * None of these are dereferenced, they are simply used to check if any of - * them have changed. If we're under current and check they are still the - * same, we're fine to grab references to them for actual out-of-line use. - */ -static void io_init_identity(struct io_identity *id) -{ - id->files = current->files; - id->mm = current->mm; -#ifdef CONFIG_BLK_CGROUP - rcu_read_lock(); - id->blkcg_css = blkcg_css(); - rcu_read_unlock(); -#endif - id->creds = current_cred(); - id->nsproxy = current->nsproxy; - id->fs = current->fs; - id->fsize = rlimit(RLIMIT_FSIZE); -#ifdef CONFIG_AUDIT - id->loginuid = current->loginuid; - id->sessionid = current->sessionid; -#endif - refcount_set(&id->count, 1); -} - static inline void __io_req_init_async(struct io_kiocb *req) { memset(&req->work, 0, sizeof(req->work)); @@ -1293,17 +1108,10 @@ static inline void __io_req_init_async(struct io_kiocb *req) */ static inline void io_req_init_async(struct io_kiocb *req) { - struct io_uring_task *tctx = current->io_uring; - if (req->flags & REQ_F_WORK_INITIALIZED) return; __io_req_init_async(req); - - /* Grab a ref if this isn't our static identity */ - req->work.identity = tctx->identity; - if (tctx->identity != &tctx->__identity) - refcount_inc(&req->work.identity->count); } static void io_ring_ctx_ref_free(struct percpu_ref *ref) @@ -1388,40 +1196,14 @@ static bool req_need_defer(struct io_kiocb *req, u32 seq) return false; } -static void io_put_identity(struct io_uring_task *tctx, struct io_kiocb *req) -{ - if (req->work.identity == &tctx->__identity) - return; - if (refcount_dec_and_test(&req->work.identity->count)) - kfree(req->work.identity); -} - static void io_req_clean_work(struct io_kiocb *req) { if (!(req->flags & REQ_F_WORK_INITIALIZED)) return; - if (req->work.flags & IO_WQ_WORK_MM) - mmdrop(req->work.identity->mm); -#ifdef CONFIG_BLK_CGROUP - if (req->work.flags & IO_WQ_WORK_BLKCG) - css_put(req->work.identity->blkcg_css); -#endif - if (req->work.flags & IO_WQ_WORK_CREDS) - put_cred(req->work.identity->creds); - if (req->work.flags & IO_WQ_WORK_FS) { - struct fs_struct *fs = req->work.identity->fs; - - spin_lock(&req->work.identity->fs->lock); - if (--fs->users) - fs = NULL; - spin_unlock(&req->work.identity->fs->lock); - if (fs) - free_fs_struct(fs); - } - if (req->work.flags & IO_WQ_WORK_FILES) { - put_files_struct(req->work.identity->files); - put_nsproxy(req->work.identity->nsproxy); + if (req->work.creds) { + put_cred(req->work.creds); + req->work.creds = NULL; } if (req->flags & REQ_F_INFLIGHT) { struct io_ring_ctx *ctx = req->ctx; @@ -1437,54 +1219,6 @@ static void io_req_clean_work(struct io_kiocb *req) } req->flags &= ~REQ_F_WORK_INITIALIZED; - req->work.flags &= ~(IO_WQ_WORK_MM | IO_WQ_WORK_BLKCG | IO_WQ_WORK_FS | - IO_WQ_WORK_CREDS | IO_WQ_WORK_FILES); - io_put_identity(req->task->io_uring, req); -} - -/* - * Create a private copy of io_identity, since some fields don't match - * the current context. - */ -static bool io_identity_cow(struct io_kiocb *req) -{ - struct io_uring_task *tctx = current->io_uring; - const struct cred *creds = NULL; - struct io_identity *id; - - if (req->work.flags & IO_WQ_WORK_CREDS) - creds = req->work.identity->creds; - - id = kmemdup(req->work.identity, sizeof(*id), GFP_KERNEL); - if (unlikely(!id)) { - req->work.flags |= IO_WQ_WORK_CANCEL; - return false; - } - - /* - * We can safely just re-init the creds we copied Either the field - * matches the current one, or we haven't grabbed it yet. The only - * exception is ->creds, through registered personalities, so handle - * that one separately. - */ - io_init_identity(id); - if (creds) - id->creds = creds; - - /* add one for this request */ - refcount_inc(&id->count); - - /* drop tctx and req identity references, if needed */ - if (tctx->identity != &tctx->__identity && - refcount_dec_and_test(&tctx->identity->count)) - kfree(tctx->identity); - if (req->work.identity != &tctx->__identity && - refcount_dec_and_test(&req->work.identity->count)) - kfree(req->work.identity); - - req->work.identity = id; - tctx->identity = id; - return true; } static void io_req_track_inflight(struct io_kiocb *req) @@ -1501,79 +1235,6 @@ static void io_req_track_inflight(struct io_kiocb *req) } } -static bool io_grab_identity(struct io_kiocb *req) -{ - const struct io_op_def *def = &io_op_defs[req->opcode]; - struct io_identity *id = req->work.identity; - - if (def->work_flags & IO_WQ_WORK_FSIZE) { - if (id->fsize != rlimit(RLIMIT_FSIZE)) - return false; - req->work.flags |= IO_WQ_WORK_FSIZE; - } -#ifdef CONFIG_BLK_CGROUP - if (!(req->work.flags & IO_WQ_WORK_BLKCG) && - (def->work_flags & IO_WQ_WORK_BLKCG)) { - rcu_read_lock(); - if (id->blkcg_css != blkcg_css()) { - rcu_read_unlock(); - return false; - } - /* - * This should be rare, either the cgroup is dying or the task - * is moving cgroups. Just punt to root for the handful of ios. - */ - if (css_tryget_online(id->blkcg_css)) - req->work.flags |= IO_WQ_WORK_BLKCG; - rcu_read_unlock(); - } -#endif - if (!(req->work.flags & IO_WQ_WORK_CREDS)) { - if (id->creds != current_cred()) - return false; - get_cred(id->creds); - req->work.flags |= IO_WQ_WORK_CREDS; - } -#ifdef CONFIG_AUDIT - if (!uid_eq(current->loginuid, id->loginuid) || - current->sessionid != id->sessionid) - return false; -#endif - if (!(req->work.flags & IO_WQ_WORK_FS) && - (def->work_flags & IO_WQ_WORK_FS)) { - if (current->fs != id->fs) - return false; - spin_lock(&id->fs->lock); - if (!id->fs->in_exec) { - id->fs->users++; - req->work.flags |= IO_WQ_WORK_FS; - } else { - req->work.flags |= IO_WQ_WORK_CANCEL; - } - spin_unlock(¤t->fs->lock); - } - if (!(req->work.flags & IO_WQ_WORK_FILES) && - (def->work_flags & IO_WQ_WORK_FILES) && - !(req->flags & REQ_F_NO_FILE_TABLE)) { - if (id->files != current->files || - id->nsproxy != current->nsproxy) - return false; - atomic_inc(&id->files->count); - get_nsproxy(id->nsproxy); - req->work.flags |= IO_WQ_WORK_FILES; - io_req_track_inflight(req); - } - if (!(req->work.flags & IO_WQ_WORK_MM) && - (def->work_flags & IO_WQ_WORK_MM)) { - if (id->mm != current->mm) - return false; - mmgrab(id->mm); - req->work.flags |= IO_WQ_WORK_MM; - } - - return true; -} - static void io_prep_async_work(struct io_kiocb *req) { const struct io_op_def *def = &io_op_defs[req->opcode]; @@ -1591,17 +1252,8 @@ static void io_prep_async_work(struct io_kiocb *req) if (def->unbound_nonreg_file) req->work.flags |= IO_WQ_WORK_UNBOUND; } - - /* if we fail grabbing identity, we must COW, regrab, and retry */ - if (io_grab_identity(req)) - return; - - if (!io_identity_cow(req)) - return; - - /* can't fail at this point */ - if (!io_grab_identity(req)) - WARN_ON(1); + if (!req->work.creds) + req->work.creds = get_current_cred(); } static void io_prep_async_link(struct io_kiocb *req) @@ -1616,10 +1268,14 @@ static struct io_kiocb *__io_queue_async_work(struct io_kiocb *req) { struct io_ring_ctx *ctx = req->ctx; struct io_kiocb *link = io_prep_linked_timeout(req); + struct io_uring_task *tctx = req->task->io_uring; + + BUG_ON(!tctx); + BUG_ON(!tctx->io_wq); trace_io_uring_queue_async_work(ctx, io_wq_is_hashed(&req->work), req, &req->work, req->flags); - io_wq_enqueue(ctx->io_wq, &req->work); + io_wq_enqueue(tctx->io_wq, &req->work); return link; } @@ -2313,11 +1969,14 @@ static int io_req_task_work_add(struct io_kiocb *req) static void io_req_task_work_add_fallback(struct io_kiocb *req, task_work_func_t cb) { - struct task_struct *tsk = io_wq_get_task(req->ctx->io_wq); + struct io_ring_ctx *ctx = req->ctx; + struct callback_head *head; init_task_work(&req->task_work, cb); - task_work_add(tsk, &req->task_work, TWA_NONE); - wake_up_process(tsk); + do { + head = READ_ONCE(ctx->exit_task_work); + req->task_work.next = head; + } while (cmpxchg(&ctx->exit_task_work, head, &req->task_work) != head); } static void __io_req_task_cancel(struct io_kiocb *req, int error) @@ -2351,15 +2010,11 @@ static void __io_req_task_submit(struct io_kiocb *req) /* ctx stays valid until unlock, even if we drop all ours ctx->refs */ mutex_lock(&ctx->uring_lock); - if (!ctx->sqo_dead && !(current->flags & PF_EXITING) && - !io_sq_thread_acquire_mm_files(ctx, req)) + if (!ctx->sqo_dead && !(current->flags & PF_EXITING) && !current->in_execve) __io_queue_sqe(req); else __io_req_task_cancel(req, -EFAULT); mutex_unlock(&ctx->uring_lock); - - if (ctx->flags & IORING_SETUP_SQPOLL) - io_sq_thread_drop_mm_files(); } static void io_req_task_submit(struct callback_head *cb) @@ -2823,7 +2478,6 @@ static bool io_rw_reissue(struct io_kiocb *req) { #ifdef CONFIG_BLOCK umode_t mode = file_inode(req->file)->i_mode; - int ret; if (!S_ISBLK(mode) && !S_ISREG(mode)) return false; @@ -2839,9 +2493,7 @@ static bool io_rw_reissue(struct io_kiocb *req) lockdep_assert_held(&req->ctx->uring_lock); - ret = io_sq_thread_acquire_mm_files(req->ctx, req); - - if (!ret && io_resubmit_prep(req)) { + if (io_resubmit_prep(req)) { refcount_inc(&req->refs); io_queue_async_work(req); return true; @@ -5946,12 +5598,15 @@ static bool io_cancel_cb(struct io_wq_work *work, void *data) return req->user_data == (unsigned long) data; } -static int io_async_cancel_one(struct io_ring_ctx *ctx, void *sqe_addr) +static int io_async_cancel_one(struct io_uring_task *tctx, void *sqe_addr) { enum io_wq_cancel cancel_ret; int ret = 0; - cancel_ret = io_wq_cancel_cb(ctx->io_wq, io_cancel_cb, sqe_addr, false); + if (!tctx->io_wq) + return -ENOENT; + + cancel_ret = io_wq_cancel_cb(tctx->io_wq, io_cancel_cb, sqe_addr, false); switch (cancel_ret) { case IO_WQ_CANCEL_OK: ret = 0; @@ -5974,7 +5629,8 @@ static void io_async_find_and_cancel(struct io_ring_ctx *ctx, unsigned long flags; int ret; - ret = io_async_cancel_one(ctx, (void *) (unsigned long) sqe_addr); + ret = io_async_cancel_one(req->task->io_uring, + (void *) (unsigned long) sqe_addr); if (ret != -ENOENT) { spin_lock_irqsave(&ctx->completion_lock, flags); goto done; @@ -6563,10 +6219,9 @@ static void __io_queue_sqe(struct io_kiocb *req) const struct cred *old_creds = NULL; int ret; - if ((req->flags & REQ_F_WORK_INITIALIZED) && - (req->work.flags & IO_WQ_WORK_CREDS) && - req->work.identity->creds != current_cred()) - old_creds = override_creds(req->work.identity->creds); + if ((req->flags & REQ_F_WORK_INITIALIZED) && req->work.creds && + req->work.creds != current_cred()) + old_creds = override_creds(req->work.creds); ret = io_issue_sqe(req, IO_URING_F_NONBLOCK|IO_URING_F_COMPLETE_DEFER); @@ -6684,9 +6339,6 @@ static int io_init_req(struct io_ring_ctx *ctx, struct io_kiocb *req, if (unlikely(req->opcode >= IORING_OP_LAST)) return -EINVAL; - if (unlikely(io_sq_thread_acquire_mm_files(ctx, req))) - return -EFAULT; - if (unlikely(!io_check_restriction(ctx, req, sqe_flags))) return -EACCES; @@ -6696,17 +6348,11 @@ static int io_init_req(struct io_ring_ctx *ctx, struct io_kiocb *req, id = READ_ONCE(sqe->personality); if (id) { - struct io_identity *iod; - - iod = idr_find(&ctx->personality_idr, id); - if (unlikely(!iod)) - return -EINVAL; - refcount_inc(&iod->count); - __io_req_init_async(req); - get_cred(iod->creds); - req->work.identity = iod; - req->work.flags |= IO_WQ_WORK_CREDS; + req->work.creds = idr_find(&ctx->personality_idr, id); + if (unlikely(!req->work.creds)) + return -EINVAL; + get_cred(req->work.creds); } state = &ctx->submit_state; @@ -7008,71 +6654,97 @@ static void io_sqd_init_new(struct io_sq_data *sqd) io_sqd_update_thread_idle(sqd); } +static bool io_sq_thread_should_stop(struct io_sq_data *sqd) +{ + return test_bit(IO_SQ_THREAD_SHOULD_STOP, &sqd->state); +} + +static bool io_sq_thread_should_park(struct io_sq_data *sqd) +{ + return test_bit(IO_SQ_THREAD_SHOULD_PARK, &sqd->state); +} + +static void io_sq_thread_parkme(struct io_sq_data *sqd) +{ + for (;;) { + /* + * TASK_PARKED is a special state; we must serialize against + * possible pending wakeups to avoid store-store collisions on + * task->state. + * + * Such a collision might possibly result in the task state + * changin from TASK_PARKED and us failing the + * wait_task_inactive() in kthread_park(). + */ + set_special_state(TASK_PARKED); + if (!test_bit(IO_SQ_THREAD_SHOULD_PARK, &sqd->state)) + break; + + /* + * Thread is going to call schedule(), do not preempt it, + * or the caller of kthread_park() may spend more time in + * wait_task_inactive(). + */ + preempt_disable(); + complete(&sqd->completion); + schedule_preempt_disabled(); + preempt_enable(); + } + __set_current_state(TASK_RUNNING); +} + static int io_sq_thread(void *data) { - struct cgroup_subsys_state *cur_css = NULL; - struct files_struct *old_files = current->files; - struct nsproxy *old_nsproxy = current->nsproxy; - const struct cred *old_cred = NULL; struct io_sq_data *sqd = data; struct io_ring_ctx *ctx; unsigned long timeout = 0; + char buf[TASK_COMM_LEN]; DEFINE_WAIT(wait); - task_lock(current); - current->files = NULL; - current->nsproxy = NULL; - task_unlock(current); + sprintf(buf, "iou-sqp-%d", sqd->task_pid); + set_task_comm(current, buf); + sqd->thread = current; + current->pf_io_worker = NULL; + + if (sqd->sq_cpu != -1) + set_cpus_allowed_ptr(current, cpumask_of(sqd->sq_cpu)); + else + set_cpus_allowed_ptr(current, cpu_online_mask); + current->flags |= PF_NO_SETAFFINITY; + + complete(&sqd->completion); - while (!kthread_should_stop()) { + wait_for_completion(&sqd->startup); + + while (!io_sq_thread_should_stop(sqd)) { int ret; bool cap_entries, sqt_spin, needs_sched; /* * Any changes to the sqd lists are synchronized through the - * kthread parking. This synchronizes the thread vs users, + * thread parking. This synchronizes the thread vs users, * the users are synchronized on the sqd->ctx_lock. */ - if (kthread_should_park()) { - kthread_parkme(); - /* - * When sq thread is unparked, in case the previous park operation - * comes from io_put_sq_data(), which means that sq thread is going - * to be stopped, so here needs to have a check. - */ - if (kthread_should_stop()) - break; + if (io_sq_thread_should_park(sqd)) { + io_sq_thread_parkme(sqd); + continue; } - if (unlikely(!list_empty(&sqd->ctx_new_list))) { io_sqd_init_new(sqd); timeout = jiffies + sqd->sq_thread_idle; } - + if (fatal_signal_pending(current)) + break; sqt_spin = false; cap_entries = !list_is_singular(&sqd->ctx_list); list_for_each_entry(ctx, &sqd->ctx_list, sqd_list) { - if (current->cred != ctx->creds) { - if (old_cred) - revert_creds(old_cred); - old_cred = override_creds(ctx->creds); - } - io_sq_thread_associate_blkcg(ctx, &cur_css); -#ifdef CONFIG_AUDIT - current->loginuid = ctx->loginuid; - current->sessionid = ctx->sessionid; -#endif - ret = __io_sq_thread(ctx, cap_entries); if (!sqt_spin && (ret > 0 || !list_empty(&ctx->iopoll_list))) sqt_spin = true; - - io_sq_thread_drop_mm_files(); } if (sqt_spin || !time_after(jiffies, timeout)) { io_run_task_work(); - io_sq_thread_drop_mm_files(); cond_resched(); if (sqt_spin) timeout = jiffies + sqd->sq_thread_idle; @@ -7093,7 +6765,7 @@ static int io_sq_thread(void *data) } } - if (needs_sched && !kthread_should_park()) { + if (needs_sched && !io_sq_thread_should_park(sqd)) { list_for_each_entry(ctx, &sqd->ctx_list, sqd_list) io_ring_set_wakeup_flag(ctx); @@ -7106,22 +6778,25 @@ static int io_sq_thread(void *data) timeout = jiffies + sqd->sq_thread_idle; } - io_run_task_work(); - io_sq_thread_drop_mm_files(); - - if (cur_css) - io_sq_thread_unassociate_blkcg(); - if (old_cred) - revert_creds(old_cred); + list_for_each_entry(ctx, &sqd->ctx_list, sqd_list) + io_uring_cancel_sqpoll(ctx); - task_lock(current); - current->files = old_files; - current->nsproxy = old_nsproxy; - task_unlock(current); + io_run_task_work(); - kthread_parkme(); + /* + * Clear thread under lock so that concurrent parks work correctly + */ + complete_all(&sqd->completion); + mutex_lock(&sqd->lock); + sqd->thread = NULL; + list_for_each_entry(ctx, &sqd->ctx_list, sqd_list) { + ctx->sqo_exec = 1; + io_ring_set_wakeup_flag(ctx); + } + mutex_unlock(&sqd->lock); - return 0; + complete(&sqd->exited); + do_exit(0); } struct io_wait_queue { @@ -7413,20 +7088,74 @@ static int io_sqe_files_unregister(struct io_ring_ctx *ctx) return 0; } +static void io_sq_thread_unpark(struct io_sq_data *sqd) + __releases(&sqd->lock) +{ + if (!sqd->thread) + return; + if (sqd->thread == current) + return; + clear_bit(IO_SQ_THREAD_SHOULD_PARK, &sqd->state); + wake_up_state(sqd->thread, TASK_PARKED); + mutex_unlock(&sqd->lock); +} + +static bool io_sq_thread_park(struct io_sq_data *sqd) + __acquires(&sqd->lock) +{ + if (sqd->thread == current) + return true; + mutex_lock(&sqd->lock); + if (!sqd->thread) { + mutex_unlock(&sqd->lock); + return false; + } + set_bit(IO_SQ_THREAD_SHOULD_PARK, &sqd->state); + wake_up_process(sqd->thread); + wait_for_completion(&sqd->completion); + return true; +} + +static void io_sq_thread_stop(struct io_sq_data *sqd) +{ + if (!sqd->thread) + return; + + set_bit(IO_SQ_THREAD_SHOULD_STOP, &sqd->state); + WARN_ON_ONCE(test_bit(IO_SQ_THREAD_SHOULD_PARK, &sqd->state)); + wake_up_process(sqd->thread); + wait_for_completion(&sqd->exited); +} + static void io_put_sq_data(struct io_sq_data *sqd) { if (refcount_dec_and_test(&sqd->refs)) { - /* - * The park is a bit of a work-around, without it we get - * warning spews on shutdown with SQPOLL set and affinity - * set to a single CPU. - */ + io_sq_thread_stop(sqd); + kfree(sqd); + } +} + +static void io_sq_thread_finish(struct io_ring_ctx *ctx) +{ + struct io_sq_data *sqd = ctx->sq_data; + + if (sqd) { + complete(&sqd->startup); if (sqd->thread) { - kthread_park(sqd->thread); - kthread_stop(sqd->thread); + wait_for_completion(&ctx->sq_thread_comp); + io_sq_thread_park(sqd); } - kfree(sqd); + mutex_lock(&sqd->ctx_lock); + list_del(&ctx->sqd_list); + io_sqd_update_thread_idle(sqd); + mutex_unlock(&sqd->ctx_lock); + + if (sqd->thread) + io_sq_thread_unpark(sqd); + + io_put_sq_data(sqd); + ctx->sq_data = NULL; } } @@ -7473,68 +7202,12 @@ static struct io_sq_data *io_get_sq_data(struct io_uring_params *p) mutex_init(&sqd->ctx_lock); mutex_init(&sqd->lock); init_waitqueue_head(&sqd->wait); + init_completion(&sqd->startup); + init_completion(&sqd->completion); + init_completion(&sqd->exited); return sqd; } -static void io_sq_thread_unpark(struct io_sq_data *sqd) - __releases(&sqd->lock) -{ - if (!sqd->thread) - return; - kthread_unpark(sqd->thread); - mutex_unlock(&sqd->lock); -} - -static void io_sq_thread_park(struct io_sq_data *sqd) - __acquires(&sqd->lock) -{ - if (!sqd->thread) - return; - mutex_lock(&sqd->lock); - kthread_park(sqd->thread); -} - -static void io_sq_thread_stop(struct io_ring_ctx *ctx) -{ - struct io_sq_data *sqd = ctx->sq_data; - - if (sqd) { - if (sqd->thread) { - /* - * We may arrive here from the error branch in - * io_sq_offload_create() where the kthread is created - * without being waked up, thus wake it up now to make - * sure the wait will complete. - */ - wake_up_process(sqd->thread); - wait_for_completion(&ctx->sq_thread_comp); - - io_sq_thread_park(sqd); - } - - mutex_lock(&sqd->ctx_lock); - list_del(&ctx->sqd_list); - io_sqd_update_thread_idle(sqd); - mutex_unlock(&sqd->ctx_lock); - - if (sqd->thread) - io_sq_thread_unpark(sqd); - - io_put_sq_data(sqd); - ctx->sq_data = NULL; - } -} - -static void io_finish_async(struct io_ring_ctx *ctx) -{ - io_sq_thread_stop(ctx); - - if (ctx->io_wq) { - io_wq_destroy(ctx->io_wq); - ctx->io_wq = NULL; - } -} - #if defined(CONFIG_UNIX) /* * Ensure the UNIX gc is aware of our file set, so we are certain that @@ -7561,7 +7234,7 @@ static int __io_sqe_files_scm(struct io_ring_ctx *ctx, int nr, int offset) skb->sk = sk; nr_files = 0; - fpl->user = get_uid(ctx->user); + fpl->user = get_uid(current_user()); for (i = 0; i < nr; i++) { struct file *file = io_file_from_index(ctx, i + offset); @@ -8093,54 +7766,34 @@ static struct io_wq_work *io_free_work(struct io_wq_work *work) return req ? &req->work : NULL; } -static int io_init_wq_offload(struct io_ring_ctx *ctx, - struct io_uring_params *p) +static struct io_wq *io_init_wq_offload(struct io_ring_ctx *ctx) { + struct io_wq_hash *hash; struct io_wq_data data; - struct fd f; - struct io_ring_ctx *ctx_attach; unsigned int concurrency; - int ret = 0; - - data.user = ctx->user; - data.free_work = io_free_work; - data.do_work = io_wq_submit_work; - - if (!(p->flags & IORING_SETUP_ATTACH_WQ)) { - /* Do QD, or 4 * CPUS, whatever is smallest */ - concurrency = min(ctx->sq_entries, 4 * num_online_cpus()); - ctx->io_wq = io_wq_create(concurrency, &data); - if (IS_ERR(ctx->io_wq)) { - ret = PTR_ERR(ctx->io_wq); - ctx->io_wq = NULL; - } - return ret; + hash = ctx->hash_map; + if (!hash) { + hash = kzalloc(sizeof(*hash), GFP_KERNEL); + if (!hash) + return ERR_PTR(-ENOMEM); + refcount_set(&hash->refs, 1); + init_waitqueue_head(&hash->wait); + ctx->hash_map = hash; } - f = fdget(p->wq_fd); - if (!f.file) - return -EBADF; - - if (f.file->f_op != &io_uring_fops) { - ret = -EINVAL; - goto out_fput; - } + data.hash = hash; + data.free_work = io_free_work; + data.do_work = io_wq_submit_work; - ctx_attach = f.file->private_data; - /* @io_wq is protected by holding the fd */ - if (!io_wq_get(ctx_attach->io_wq, &data)) { - ret = -EINVAL; - goto out_fput; - } + /* Do QD, or 4 * CPUS, whatever is smallest */ + concurrency = min(ctx->sq_entries, 4 * num_online_cpus()); - ctx->io_wq = ctx_attach->io_wq; -out_fput: - fdput(f); - return ret; + return io_wq_create(concurrency, &data); } -static int io_uring_alloc_task_context(struct task_struct *task) +static int io_uring_alloc_task_context(struct task_struct *task, + struct io_ring_ctx *ctx) { struct io_uring_task *tctx; int ret; @@ -8155,13 +7808,19 @@ static int io_uring_alloc_task_context(struct task_struct *task) return ret; } + tctx->io_wq = io_init_wq_offload(ctx); + if (IS_ERR(tctx->io_wq)) { + ret = PTR_ERR(tctx->io_wq); + percpu_counter_destroy(&tctx->inflight); + kfree(tctx); + return ret; + } + xa_init(&tctx->xa); init_waitqueue_head(&tctx->wait); tctx->last = NULL; atomic_set(&tctx->in_idle, 0); tctx->sqpoll = false; - io_init_identity(&tctx->__identity); - tctx->identity = &tctx->__identity; task->io_uring = tctx; spin_lock_init(&tctx->task_lock); INIT_WQ_LIST(&tctx->task_list); @@ -8175,19 +7834,49 @@ void __io_uring_free(struct task_struct *tsk) struct io_uring_task *tctx = tsk->io_uring; WARN_ON_ONCE(!xa_empty(&tctx->xa)); - WARN_ON_ONCE(refcount_read(&tctx->identity->count) != 1); - if (tctx->identity != &tctx->__identity) - kfree(tctx->identity); percpu_counter_destroy(&tctx->inflight); kfree(tctx); tsk->io_uring = NULL; } +static int io_sq_thread_fork(struct io_sq_data *sqd, struct io_ring_ctx *ctx) +{ + int ret; + + clear_bit(IO_SQ_THREAD_SHOULD_STOP, &sqd->state); + reinit_completion(&sqd->completion); + ctx->sqo_dead = ctx->sqo_exec = 0; + sqd->task_pid = current->pid; + current->flags |= PF_IO_WORKER; + ret = io_wq_fork_thread(io_sq_thread, sqd); + current->flags &= ~PF_IO_WORKER; + if (ret < 0) { + sqd->thread = NULL; + return ret; + } + wait_for_completion(&sqd->completion); + return io_uring_alloc_task_context(sqd->thread, ctx); +} + static int io_sq_offload_create(struct io_ring_ctx *ctx, struct io_uring_params *p) { int ret; + /* Retain compatibility with failing for an invalid attach attempt */ + if ((ctx->flags & (IORING_SETUP_ATTACH_WQ | IORING_SETUP_SQPOLL)) == + IORING_SETUP_ATTACH_WQ) { + struct fd f; + + f = fdget(p->wq_fd); + if (!f.file) + return -ENXIO; + if (f.file->f_op != &io_uring_fops) { + fdput(f); + return -EINVAL; + } + fdput(f); + } if (ctx->flags & IORING_SETUP_SQPOLL) { struct io_sq_data *sqd; @@ -8213,7 +7902,7 @@ static int io_sq_offload_create(struct io_ring_ctx *ctx, ctx->sq_thread_idle = HZ; if (sqd->thread) - goto done; + return 0; if (p->flags & IORING_SETUP_SQ_AFF) { int cpu = p->sq_thread_cpu; @@ -8224,18 +7913,21 @@ static int io_sq_offload_create(struct io_ring_ctx *ctx, if (!cpu_online(cpu)) goto err; - sqd->thread = kthread_create_on_cpu(io_sq_thread, sqd, - cpu, "io_uring-sq"); + sqd->sq_cpu = cpu; } else { - sqd->thread = kthread_create(io_sq_thread, sqd, - "io_uring-sq"); + sqd->sq_cpu = -1; } - if (IS_ERR(sqd->thread)) { - ret = PTR_ERR(sqd->thread); + + sqd->task_pid = current->pid; + current->flags |= PF_IO_WORKER; + ret = io_wq_fork_thread(io_sq_thread, sqd); + current->flags &= ~PF_IO_WORKER; + if (ret < 0) { sqd->thread = NULL; goto err; } - ret = io_uring_alloc_task_context(sqd->thread); + wait_for_completion(&sqd->completion); + ret = io_uring_alloc_task_context(sqd->thread, ctx); if (ret) goto err; } else if (p->flags & IORING_SETUP_SQ_AFF) { @@ -8244,14 +7936,9 @@ static int io_sq_offload_create(struct io_ring_ctx *ctx, goto err; } -done: - ret = io_init_wq_offload(ctx, p); - if (ret) - goto err; - return 0; err: - io_finish_async(ctx); + io_sq_thread_finish(ctx); return ret; } @@ -8259,8 +7946,8 @@ static void io_sq_offload_start(struct io_ring_ctx *ctx) { struct io_sq_data *sqd = ctx->sq_data; - if ((ctx->flags & IORING_SETUP_SQPOLL) && sqd->thread) - wake_up_process(sqd->thread); + if (ctx->flags & IORING_SETUP_SQPOLL) + complete(&sqd->startup); } static inline void __io_unaccount_mem(struct user_struct *user, @@ -8290,7 +7977,7 @@ static inline int __io_account_mem(struct user_struct *user, static void io_unaccount_mem(struct io_ring_ctx *ctx, unsigned long nr_pages) { - if (ctx->limit_mem) + if (ctx->user) __io_unaccount_mem(ctx->user, nr_pages); if (ctx->mm_account) @@ -8301,7 +7988,7 @@ static int io_account_mem(struct io_ring_ctx *ctx, unsigned long nr_pages) { int ret; - if (ctx->limit_mem) { + if (ctx->user) { ret = __io_account_mem(ctx->user, nr_pages); if (ret) return ret; @@ -8730,21 +8417,14 @@ static void io_ring_ctx_free(struct io_ring_ctx *ctx) mutex_lock(&ctx->uring_lock); mutex_unlock(&ctx->uring_lock); - io_finish_async(ctx); + io_sq_thread_finish(ctx); io_sqe_buffers_unregister(ctx); - if (ctx->sqo_task) { - put_task_struct(ctx->sqo_task); - ctx->sqo_task = NULL; + if (ctx->mm_account) { mmdrop(ctx->mm_account); ctx->mm_account = NULL; } -#ifdef CONFIG_BLK_CGROUP - if (ctx->sqo_blkcg_css) - css_put(ctx->sqo_blkcg_css); -#endif - mutex_lock(&ctx->uring_lock); io_sqe_files_unregister(ctx); mutex_unlock(&ctx->uring_lock); @@ -8764,8 +8444,9 @@ static void io_ring_ctx_free(struct io_ring_ctx *ctx) percpu_ref_exit(&ctx->refs); free_uid(ctx->user); - put_cred(ctx->creds); io_req_caches_free(ctx, NULL); + if (ctx->hash_map) + io_wq_put_hash(ctx->hash_map); kfree(ctx->cancel_hash); kfree(ctx); } @@ -8812,13 +8493,11 @@ static int io_uring_fasync(int fd, struct file *file, int on) static int io_unregister_personality(struct io_ring_ctx *ctx, unsigned id) { - struct io_identity *iod; + const struct cred *creds; - iod = idr_remove(&ctx->personality_idr, id); - if (iod) { - put_cred(iod->creds); - if (refcount_dec_and_test(&iod->count)) - kfree(iod); + creds = idr_remove(&ctx->personality_idr, id); + if (creds) { + put_cred(creds); return 0; } @@ -8833,6 +8512,28 @@ static int io_remove_personalities(int id, void *p, void *data) return 0; } +static void io_run_ctx_fallback(struct io_ring_ctx *ctx) +{ + struct callback_head *work, *head, *next; + + do { + do { + head = NULL; + work = READ_ONCE(ctx->exit_task_work); + } while (cmpxchg(&ctx->exit_task_work, work, head) != work); + + if (!work) + break; + + do { + next = work->next; + work->func(work); + work = next; + cond_resched(); + } while (work); + } while (1); +} + static void io_ring_exit_work(struct work_struct *work) { struct io_ring_ctx *ctx = container_of(work, struct io_ring_ctx, @@ -8846,17 +8547,11 @@ static void io_ring_exit_work(struct work_struct *work) */ do { io_uring_try_cancel_requests(ctx, NULL, NULL); + io_run_ctx_fallback(ctx); } while (!wait_for_completion_timeout(&ctx->ref_comp, HZ/20)); io_ring_ctx_free(ctx); } -static bool io_cancel_ctx_cb(struct io_wq_work *work, void *data) -{ - struct io_kiocb *req = container_of(work, struct io_kiocb, work); - - return req->ctx == data; -} - static void io_ring_ctx_wait_and_kill(struct io_ring_ctx *ctx) { mutex_lock(&ctx->uring_lock); @@ -8875,9 +8570,6 @@ static void io_ring_ctx_wait_and_kill(struct io_ring_ctx *ctx) io_kill_timeouts(ctx, NULL, NULL); io_poll_remove_all(ctx, NULL, NULL); - if (ctx->io_wq) - io_wq_cancel_cb(ctx->io_wq, io_cancel_ctx_cb, ctx, true); - /* if we failed setting up the ctx, we might not have any rings */ io_iopoll_try_reap_events(ctx); @@ -8956,13 +8648,14 @@ static void io_uring_try_cancel_requests(struct io_ring_ctx *ctx, struct files_struct *files) { struct io_task_cancel cancel = { .task = task, .files = files, }; + struct io_uring_task *tctx = current->io_uring; while (1) { enum io_wq_cancel cret; bool ret = false; - if (ctx->io_wq) { - cret = io_wq_cancel_cb(ctx->io_wq, io_cancel_task_cb, + if (tctx && tctx->io_wq) { + cret = io_wq_cancel_cb(tctx->io_wq, io_cancel_task_cb, &cancel, true); ret |= (cret != IO_WQ_CANCEL_NOTFOUND); } @@ -9045,12 +8738,15 @@ static void io_uring_cancel_task_requests(struct io_ring_ctx *ctx, struct files_struct *files) { struct task_struct *task = current; + bool did_park = false; if ((ctx->flags & IORING_SETUP_SQPOLL) && ctx->sq_data) { io_disable_sqo_submit(ctx); - task = ctx->sq_data->thread; - atomic_inc(&task->io_uring->in_idle); - io_sq_thread_park(ctx->sq_data); + did_park = io_sq_thread_park(ctx->sq_data); + if (did_park) { + task = ctx->sq_data->thread; + atomic_inc(&task->io_uring->in_idle); + } } io_cancel_defer_files(ctx, task, files); @@ -9059,7 +8755,7 @@ static void io_uring_cancel_task_requests(struct io_ring_ctx *ctx, if (!files) io_uring_try_cancel_requests(ctx, task, NULL); - if ((ctx->flags & IORING_SETUP_SQPOLL) && ctx->sq_data) { + if (did_park) { atomic_dec(&task->io_uring->in_idle); io_sq_thread_unpark(ctx->sq_data); } @@ -9074,7 +8770,7 @@ static int io_uring_add_task_file(struct io_ring_ctx *ctx, struct file *file) int ret; if (unlikely(!tctx)) { - ret = io_uring_alloc_task_context(current); + ret = io_uring_alloc_task_context(current, ctx); if (unlikely(ret)) return ret; tctx = current->io_uring; @@ -9144,8 +8840,13 @@ void __io_uring_files_cancel(struct files_struct *files) io_uring_cancel_task_requests(file->private_data, files); atomic_dec(&tctx->in_idle); - if (files) + if (files) { io_uring_remove_task_files(tctx); + if (tctx->io_wq) { + io_wq_put(tctx->io_wq); + tctx->io_wq = NULL; + } + } } static s64 tctx_inflight(struct io_uring_task *tctx) @@ -9155,14 +8856,17 @@ static s64 tctx_inflight(struct io_uring_task *tctx) static void io_uring_cancel_sqpoll(struct io_ring_ctx *ctx) { + struct io_sq_data *sqd = ctx->sq_data; struct io_uring_task *tctx; s64 inflight; DEFINE_WAIT(wait); - if (!ctx->sq_data) + if (!sqd) return; - tctx = ctx->sq_data->thread->io_uring; io_disable_sqo_submit(ctx); + if (!io_sq_thread_park(sqd)) + return; + tctx = ctx->sq_data->thread->io_uring; atomic_inc(&tctx->in_idle); do { @@ -9183,6 +8887,7 @@ static void io_uring_cancel_sqpoll(struct io_ring_ctx *ctx) finish_wait(&tctx->wait, &wait); } while (1); atomic_dec(&tctx->in_idle); + io_sq_thread_unpark(sqd); } /* @@ -9236,11 +8941,17 @@ static int io_uring_flush(struct file *file, void *data) struct io_uring_task *tctx = current->io_uring; struct io_ring_ctx *ctx = file->private_data; + /* Ignore helper thread files exit */ + if (current->flags & PF_IO_WORKER) + return 0; + if (fatal_signal_pending(current) || (current->flags & PF_EXITING)) { io_uring_cancel_task_requests(ctx, NULL); io_req_caches_free(ctx, current); } + io_run_ctx_fallback(ctx); + if (!tctx) return 0; @@ -9439,6 +9150,12 @@ SYSCALL_DEFINE6(io_uring_enter, unsigned int, fd, u32, to_submit, if (ctx->flags & IORING_SETUP_SQPOLL) { io_cqring_overflow_flush(ctx, false, NULL, NULL); + if (unlikely(ctx->sqo_exec)) { + ret = io_sq_thread_fork(ctx->sq_data, ctx); + if (ret) + goto out; + ctx->sqo_exec = 0; + } ret = -EOWNERDEAD; if (unlikely(ctx->sqo_dead)) goto out; @@ -9495,8 +9212,7 @@ out_fput: #ifdef CONFIG_PROC_FS static int io_uring_show_cred(int id, void *p, void *data) { - struct io_identity *iod = p; - const struct cred *cred = iod->creds; + const struct cred *cred = p; struct seq_file *m = data; struct user_namespace *uns = seq_user_ns(m); struct group_info *gi; @@ -9541,8 +9257,11 @@ static void __io_uring_show_fdinfo(struct io_ring_ctx *ctx, struct seq_file *m) */ has_lock = mutex_trylock(&ctx->uring_lock); - if (has_lock && (ctx->flags & IORING_SETUP_SQPOLL)) + if (has_lock && (ctx->flags & IORING_SETUP_SQPOLL)) { sq = ctx->sq_data; + if (!sq->thread) + sq = NULL; + } seq_printf(m, "SqThread:\t%d\n", sq ? task_pid_nr(sq->thread) : -1); seq_printf(m, "SqThreadCpu:\t%d\n", sq ? task_cpu(sq->thread) : -1); @@ -9702,7 +9421,6 @@ static struct file *io_uring_get_file(struct io_ring_ctx *ctx) static int io_uring_create(unsigned entries, struct io_uring_params *p, struct io_uring_params __user *params) { - struct user_struct *user = NULL; struct io_ring_ctx *ctx; struct file *file; int ret; @@ -9744,22 +9462,13 @@ static int io_uring_create(unsigned entries, struct io_uring_params *p, p->cq_entries = 2 * p->sq_entries; } - user = get_uid(current_user()); - ctx = io_ring_ctx_alloc(p); - if (!ctx) { - free_uid(user); + if (!ctx) return -ENOMEM; - } ctx->compat = in_compat_syscall(); - ctx->limit_mem = !capable(CAP_IPC_LOCK); - ctx->user = user; - ctx->creds = get_current_cred(); -#ifdef CONFIG_AUDIT - ctx->loginuid = current->loginuid; - ctx->sessionid = current->sessionid; -#endif - ctx->sqo_task = get_task_struct(current); + if (!capable(CAP_IPC_LOCK)) + ctx->user = get_uid(current_user()); + ctx->sqo_task = current; /* * This is just grabbed for accounting purposes. When a process exits, @@ -9770,24 +9479,6 @@ static int io_uring_create(unsigned entries, struct io_uring_params *p, mmgrab(current->mm); ctx->mm_account = current->mm; -#ifdef CONFIG_BLK_CGROUP - /* - * The sq thread will belong to the original cgroup it was inited in. - * If the cgroup goes offline (e.g. disabling the io controller), then - * issued bios will be associated with the closest cgroup later in the - * block layer. - */ - rcu_read_lock(); - ctx->sqo_blkcg_css = blkcg_css(); - ret = css_tryget_online(ctx->sqo_blkcg_css); - rcu_read_unlock(); - if (!ret) { - /* don't init against a dying cgroup, have the user try again */ - ctx->sqo_blkcg_css = NULL; - ret = -ENODEV; - goto err; - } -#endif ret = io_allocate_scq_urings(ctx, p); if (ret) goto err; @@ -9821,7 +9512,7 @@ static int io_uring_create(unsigned entries, struct io_uring_params *p, IORING_FEAT_SUBMIT_STABLE | IORING_FEAT_RW_CUR_POS | IORING_FEAT_CUR_PERSONALITY | IORING_FEAT_FAST_POLL | IORING_FEAT_POLL_32BITS | IORING_FEAT_SQPOLL_NONFIXED | - IORING_FEAT_EXT_ARG; + IORING_FEAT_EXT_ARG | IORING_FEAT_NATIVE_WORKERS; if (copy_to_user(params, p, sizeof(*p))) { ret = -EFAULT; @@ -9927,21 +9618,15 @@ out: static int io_register_personality(struct io_ring_ctx *ctx) { - struct io_identity *id; + const struct cred *creds; int ret; - id = kmalloc(sizeof(*id), GFP_KERNEL); - if (unlikely(!id)) - return -ENOMEM; - - io_init_identity(id); - id->creds = get_current_cred(); + creds = get_current_cred(); - ret = idr_alloc_cyclic(&ctx->personality_idr, id, 1, USHRT_MAX, GFP_KERNEL); - if (ret < 0) { - put_cred(id->creds); - kfree(id); - } + ret = idr_alloc_cyclic(&ctx->personality_idr, (void *) creds, 1, + USHRT_MAX, GFP_KERNEL); + if (ret < 0) + put_cred(creds); return ret; } diff --git a/fs/iomap/buffered-io.c b/fs/iomap/buffered-io.c index 16a1e82e3aeb..7ffcd7ef33d4 100644 --- a/fs/iomap/buffered-io.c +++ b/fs/iomap/buffered-io.c @@ -278,14 +278,14 @@ iomap_readpage_actor(struct inode *inode, loff_t pos, loff_t length, void *data, if (!is_contig || bio_full(ctx->bio, plen)) { gfp_t gfp = mapping_gfp_constraint(page->mapping, GFP_KERNEL); gfp_t orig_gfp = gfp; - int nr_vecs = (length + PAGE_SIZE - 1) >> PAGE_SHIFT; + unsigned int nr_vecs = DIV_ROUND_UP(length, PAGE_SIZE); if (ctx->bio) submit_bio(ctx->bio); if (ctx->rac) /* same as readahead_gfp_mask */ gfp |= __GFP_NORETRY | __GFP_NOWARN; - ctx->bio = bio_alloc(gfp, min(BIO_MAX_PAGES, nr_vecs)); + ctx->bio = bio_alloc(gfp, bio_max_segs(nr_vecs)); /* * If the bio_alloc fails, try it again for a single page to * avoid having to deal with partial page reads. This emulates @@ -1459,13 +1459,6 @@ iomap_do_writepage(struct page *page, struct writeback_control *wbc, void *data) goto redirty; /* - * Given that we do not allow direct reclaim to call us, we should - * never be called in a recursive filesystem reclaim context. - */ - if (WARN_ON_ONCE(current->flags & PF_MEMALLOC_NOFS)) - goto redirty; - - /* * Is this page beyond the end of the file? * * The page index is less than the end_index, adjust the end_offset diff --git a/fs/jfs/super.c b/fs/jfs/super.c index b2dc4d1f9dcc..1f0ffabbde56 100644 --- a/fs/jfs/super.c +++ b/fs/jfs/super.c @@ -551,7 +551,6 @@ static int jfs_fill_super(struct super_block *sb, void *data, int silent) ret = -ENOMEM; goto out_unload; } - inode->i_ino = 0; inode->i_size = i_size_read(sb->s_bdev->bd_inode); inode->i_mapping->a_ops = &jfs_metapage_aops; inode_fake_hash(inode); diff --git a/fs/mpage.c b/fs/mpage.c index 830e6cc2a9e7..961234d68779 100644 --- a/fs/mpage.c +++ b/fs/mpage.c @@ -304,9 +304,7 @@ alloc_new: goto out; } args->bio = mpage_alloc(bdev, blocks[0] << (blkbits - 9), - min_t(int, args->nr_pages, - BIO_MAX_PAGES), - gfp); + bio_max_segs(args->nr_pages), gfp); if (args->bio == NULL) goto confused; } diff --git a/fs/namespace.c b/fs/namespace.c index 062fe984a697..56bb5a5fdc0d 100644 --- a/fs/namespace.c +++ b/fs/namespace.c @@ -381,50 +381,36 @@ int mnt_want_write(struct vfsmount *m) EXPORT_SYMBOL_GPL(mnt_want_write); /** - * mnt_clone_write - get write access to a mount - * @mnt: the mount on which to take a write - * - * This is effectively like mnt_want_write, except - * it must only be used to take an extra write reference - * on a mountpoint that we already know has a write reference - * on it. This allows some optimisation. - * - * After finished, mnt_drop_write must be called as usual to - * drop the reference. - */ -int mnt_clone_write(struct vfsmount *mnt) -{ - /* superblock may be r/o */ - if (__mnt_is_readonly(mnt)) - return -EROFS; - preempt_disable(); - mnt_inc_writers(real_mount(mnt)); - preempt_enable(); - return 0; -} -EXPORT_SYMBOL_GPL(mnt_clone_write); - -/** * __mnt_want_write_file - get write access to a file's mount * @file: the file who's mount on which to take a write * - * This is like __mnt_want_write, but it takes a file and can - * do some optimisations if the file is open for write already + * This is like __mnt_want_write, but if the file is already open for writing it + * skips incrementing mnt_writers (since the open file already has a reference) + * and instead only does the check for emergency r/o remounts. This must be + * paired with __mnt_drop_write_file. */ int __mnt_want_write_file(struct file *file) { - if (!(file->f_mode & FMODE_WRITER)) - return __mnt_want_write(file->f_path.mnt); - else - return mnt_clone_write(file->f_path.mnt); + if (file->f_mode & FMODE_WRITER) { + /* + * Superblock may have become readonly while there are still + * writable fd's, e.g. due to a fs error with errors=remount-ro + */ + if (__mnt_is_readonly(file->f_path.mnt)) + return -EROFS; + return 0; + } + return __mnt_want_write(file->f_path.mnt); } /** * mnt_want_write_file - get write access to a file's mount * @file: the file who's mount on which to take a write * - * This is like mnt_want_write, but it takes a file and can - * do some optimisations if the file is open for write already + * This is like mnt_want_write, but if the file is already open for writing it + * skips incrementing mnt_writers (since the open file already has a reference) + * and instead only does the freeze protection and the check for emergency r/o + * remounts. This must be paired with mnt_drop_write_file. */ int mnt_want_write_file(struct file *file) { @@ -470,7 +456,8 @@ EXPORT_SYMBOL_GPL(mnt_drop_write); void __mnt_drop_write_file(struct file *file) { - __mnt_drop_write(file->f_path.mnt); + if (!(file->f_mode & FMODE_WRITER)) + __mnt_drop_write(file->f_path.mnt); } void mnt_drop_write_file(struct file *file) diff --git a/fs/nfs/blocklayout/blocklayout.c b/fs/nfs/blocklayout/blocklayout.c index 1a96ce28efb0..fe860c538747 100644 --- a/fs/nfs/blocklayout/blocklayout.c +++ b/fs/nfs/blocklayout/blocklayout.c @@ -115,13 +115,13 @@ bl_submit_bio(struct bio *bio) return NULL; } -static struct bio * -bl_alloc_init_bio(int npg, struct block_device *bdev, sector_t disk_sector, +static struct bio *bl_alloc_init_bio(unsigned int npg, + struct block_device *bdev, sector_t disk_sector, bio_end_io_t end_io, struct parallel_io *par) { struct bio *bio; - npg = min(npg, BIO_MAX_PAGES); + npg = bio_max_segs(npg); bio = bio_alloc(GFP_NOIO, npg); if (bio) { bio->bi_iter.bi_sector = disk_sector; diff --git a/fs/proc/self.c b/fs/proc/self.c index a4012154e109..72cd69bcaf4a 100644 --- a/fs/proc/self.c +++ b/fs/proc/self.c @@ -16,13 +16,6 @@ static const char *proc_self_get_link(struct dentry *dentry, pid_t tgid = task_tgid_nr_ns(current, ns); char *name; - /* - * Not currently supported. Once we can inherit all of struct pid, - * we can allow this. - */ - if (current->flags & PF_IO_WORKER) - return ERR_PTR(-EOPNOTSUPP); - if (!tgid) return ERR_PTR(-ENOENT); /* max length of unsigned int in decimal + NULL term */ diff --git a/fs/proc/thread_self.c b/fs/proc/thread_self.c index d56681d86d28..a553273fbd41 100644 --- a/fs/proc/thread_self.c +++ b/fs/proc/thread_self.c @@ -17,13 +17,6 @@ static const char *proc_thread_self_get_link(struct dentry *dentry, pid_t pid = task_pid_nr_ns(current, ns); char *name; - /* - * Not currently supported. Once we can inherit all of struct pid, - * we can allow this. - */ - if (current->flags & PF_IO_WORKER) - return ERR_PTR(-EOPNOTSUPP); - if (!pid) return ERR_PTR(-ENOENT); name = kmalloc(10 + 6 + 10 + 1, dentry ? GFP_KERNEL : GFP_ATOMIC); diff --git a/fs/xfs/libxfs/xfs_btree.c b/fs/xfs/libxfs/xfs_btree.c index b56ff451adce..5b6fcb9b44e2 100644 --- a/fs/xfs/libxfs/xfs_btree.c +++ b/fs/xfs/libxfs/xfs_btree.c @@ -2805,7 +2805,7 @@ xfs_btree_split_worker( struct xfs_btree_split_args *args = container_of(work, struct xfs_btree_split_args, work); unsigned long pflags; - unsigned long new_pflags = PF_MEMALLOC_NOFS; + unsigned long new_pflags = 0; /* * we are in a transaction context here, but may also be doing work @@ -2817,12 +2817,20 @@ xfs_btree_split_worker( new_pflags |= PF_MEMALLOC | PF_SWAPWRITE | PF_KSWAPD; current_set_flags_nested(&pflags, new_pflags); + xfs_trans_set_context(args->cur->bc_tp); args->result = __xfs_btree_split(args->cur, args->level, args->ptrp, args->key, args->curp, args->stat); - complete(args->done); + xfs_trans_clear_context(args->cur->bc_tp); current_restore_flags_nested(&pflags, new_pflags); + + /* + * Do not access args after complete() has run here. We don't own args + * and the owner may run and free args before we return here. + */ + complete(args->done); + } /* diff --git a/fs/xfs/xfs_aops.c b/fs/xfs/xfs_aops.c index 4304c6416fbb..b4186d666157 100644 --- a/fs/xfs/xfs_aops.c +++ b/fs/xfs/xfs_aops.c @@ -62,7 +62,7 @@ xfs_setfilesize_trans_alloc( * We hand off the transaction to the completion thread now, so * clear the flag here. */ - current_restore_flags_nested(&tp->t_pflags, PF_MEMALLOC_NOFS); + xfs_trans_clear_context(tp); return 0; } @@ -125,7 +125,7 @@ xfs_setfilesize_ioend( * thus we need to mark ourselves as being in a transaction manually. * Similarly for freeze protection. */ - current_set_flags_nested(&tp->t_pflags, PF_MEMALLOC_NOFS); + xfs_trans_set_context(tp); __sb_writers_acquired(VFS_I(ip)->i_sb, SB_FREEZE_FS); /* we abort the update if there was an IO error */ @@ -568,6 +568,12 @@ xfs_vm_writepage( { struct xfs_writepage_ctx wpc = { }; + if (WARN_ON_ONCE(current->journal_info)) { + redirty_page_for_writepage(wbc, page); + unlock_page(page); + return 0; + } + return iomap_writepage(page, wbc, &wpc.ctx, &xfs_writeback_ops); } @@ -578,6 +584,13 @@ xfs_vm_writepages( { struct xfs_writepage_ctx wpc = { }; + /* + * Writing back data in a transaction context can result in recursive + * transactions. This is bad, so issue a warning and get out of here. + */ + if (WARN_ON_ONCE(current->journal_info)) + return 0; + xfs_iflags_clear(XFS_I(mapping->host), XFS_ITRUNCATED); return iomap_writepages(mapping, wbc, &wpc.ctx, &xfs_writeback_ops); } diff --git a/fs/xfs/xfs_bio_io.c b/fs/xfs/xfs_bio_io.c index e2148f2d5d6b..17f36db2f792 100644 --- a/fs/xfs/xfs_bio_io.c +++ b/fs/xfs/xfs_bio_io.c @@ -6,7 +6,7 @@ static inline unsigned int bio_max_vecs(unsigned int count) { - return min_t(unsigned, howmany(count, PAGE_SIZE), BIO_MAX_PAGES); + return bio_max_segs(howmany(count, PAGE_SIZE)); } int diff --git a/fs/xfs/xfs_buf.c b/fs/xfs/xfs_buf.c index f6e5235df7c9..37a1d12762d8 100644 --- a/fs/xfs/xfs_buf.c +++ b/fs/xfs/xfs_buf.c @@ -1480,7 +1480,7 @@ xfs_buf_ioapply_map( int op) { int page_index; - int total_nr_pages = bp->b_page_count; + unsigned int total_nr_pages = bp->b_page_count; int nr_pages; struct bio *bio; sector_t sector = bp->b_maps[map].bm_bn; @@ -1505,7 +1505,7 @@ xfs_buf_ioapply_map( next_chunk: atomic_inc(&bp->b_io_remaining); - nr_pages = min(total_nr_pages, BIO_MAX_PAGES); + nr_pages = bio_max_segs(total_nr_pages); bio = bio_alloc(GFP_NOIO, nr_pages); bio_set_dev(bio, bp->b_target->bt_bdev); diff --git a/fs/xfs/xfs_extent_busy.c b/fs/xfs/xfs_extent_busy.c index 3991e59cfd18..ef17c1f6db32 100644 --- a/fs/xfs/xfs_extent_busy.c +++ b/fs/xfs/xfs_extent_busy.c @@ -344,7 +344,6 @@ xfs_extent_busy_trim( ASSERT(*len > 0); spin_lock(&args->pag->pagb_lock); -restart: fbno = *bno; flen = *len; rbp = args->pag->pagb_tree.rb_node; @@ -363,19 +362,6 @@ restart: continue; } - /* - * If this is a metadata allocation, try to reuse the busy - * extent instead of trimming the allocation. - */ - if (!(args->datatype & XFS_ALLOC_USERDATA) && - !(busyp->flags & XFS_EXTENT_BUSY_DISCARDED)) { - if (!xfs_extent_busy_update_extent(args->mp, args->pag, - busyp, fbno, flen, - false)) - goto restart; - continue; - } - if (bbno <= fbno) { /* start overlap */ diff --git a/fs/xfs/xfs_sysctl.c b/fs/xfs/xfs_sysctl.c index 145e06c47744..546a6cd96729 100644 --- a/fs/xfs/xfs_sysctl.c +++ b/fs/xfs/xfs_sysctl.c @@ -51,7 +51,7 @@ xfs_panic_mask_proc_handler( #endif /* CONFIG_PROC_FS */ STATIC int -xfs_deprecate_irix_sgid_inherit_proc_handler( +xfs_deprecated_dointvec_minmax( struct ctl_table *ctl, int write, void *buffer, @@ -59,24 +59,8 @@ xfs_deprecate_irix_sgid_inherit_proc_handler( loff_t *ppos) { if (write) { - printk_once(KERN_WARNING - "XFS: " "%s sysctl option is deprecated.\n", - ctl->procname); - } - return proc_dointvec_minmax(ctl, write, buffer, lenp, ppos); -} - -STATIC int -xfs_deprecate_irix_symlink_mode_proc_handler( - struct ctl_table *ctl, - int write, - void *buffer, - size_t *lenp, - loff_t *ppos) -{ - if (write) { - printk_once(KERN_WARNING - "XFS: " "%s sysctl option is deprecated.\n", + printk_ratelimited(KERN_WARNING + "XFS: %s sysctl option is deprecated.\n", ctl->procname); } return proc_dointvec_minmax(ctl, write, buffer, lenp, ppos); @@ -88,7 +72,7 @@ static struct ctl_table xfs_table[] = { .data = &xfs_params.sgid_inherit.val, .maxlen = sizeof(int), .mode = 0644, - .proc_handler = xfs_deprecate_irix_sgid_inherit_proc_handler, + .proc_handler = xfs_deprecated_dointvec_minmax, .extra1 = &xfs_params.sgid_inherit.min, .extra2 = &xfs_params.sgid_inherit.max }, @@ -97,7 +81,7 @@ static struct ctl_table xfs_table[] = { .data = &xfs_params.symlink_mode.val, .maxlen = sizeof(int), .mode = 0644, - .proc_handler = xfs_deprecate_irix_symlink_mode_proc_handler, + .proc_handler = xfs_deprecated_dointvec_minmax, .extra1 = &xfs_params.symlink_mode.min, .extra2 = &xfs_params.symlink_mode.max }, @@ -201,6 +185,15 @@ static struct ctl_table xfs_table[] = { .extra1 = &xfs_params.blockgc_timer.min, .extra2 = &xfs_params.blockgc_timer.max, }, + { + .procname = "speculative_cow_prealloc_lifetime", + .data = &xfs_params.blockgc_timer.val, + .maxlen = sizeof(int), + .mode = 0644, + .proc_handler = xfs_deprecated_dointvec_minmax, + .extra1 = &xfs_params.blockgc_timer.min, + .extra2 = &xfs_params.blockgc_timer.max, + }, /* please keep this the last entry */ #ifdef CONFIG_PROC_FS { diff --git a/fs/xfs/xfs_trans.c b/fs/xfs/xfs_trans.c index 44f72c09c203..b22a09e9daee 100644 --- a/fs/xfs/xfs_trans.c +++ b/fs/xfs/xfs_trans.c @@ -72,6 +72,7 @@ xfs_trans_free( xfs_extent_busy_clear(tp->t_mountp, &tp->t_busy, false); trace_xfs_trans_free(tp, _RET_IP_); + xfs_trans_clear_context(tp); if (!(tp->t_flags & XFS_TRANS_NO_WRITECOUNT)) sb_end_intwrite(tp->t_mountp->m_super); xfs_trans_free_dqinfo(tp); @@ -123,7 +124,8 @@ xfs_trans_dup( ntp->t_rtx_res = tp->t_rtx_res - tp->t_rtx_res_used; tp->t_rtx_res = tp->t_rtx_res_used; - ntp->t_pflags = tp->t_pflags; + + xfs_trans_switch_context(tp, ntp); /* move deferred ops over to the new tp */ xfs_defer_move(ntp, tp); @@ -157,9 +159,6 @@ xfs_trans_reserve( int error = 0; bool rsvd = (tp->t_flags & XFS_TRANS_RESERVE) != 0; - /* Mark this thread as being in a transaction */ - current_set_flags_nested(&tp->t_pflags, PF_MEMALLOC_NOFS); - /* * Attempt to reserve the needed disk blocks by decrementing * the number needed from the number available. This will @@ -167,10 +166,8 @@ xfs_trans_reserve( */ if (blocks > 0) { error = xfs_mod_fdblocks(mp, -((int64_t)blocks), rsvd); - if (error != 0) { - current_restore_flags_nested(&tp->t_pflags, PF_MEMALLOC_NOFS); + if (error != 0) return -ENOSPC; - } tp->t_blk_res += blocks; } @@ -244,9 +241,6 @@ undo_blocks: xfs_mod_fdblocks(mp, (int64_t)blocks, rsvd); tp->t_blk_res = 0; } - - current_restore_flags_nested(&tp->t_pflags, PF_MEMALLOC_NOFS); - return error; } @@ -260,6 +254,7 @@ xfs_trans_alloc( struct xfs_trans **tpp) { struct xfs_trans *tp; + bool want_retry = true; int error; /* @@ -267,9 +262,11 @@ xfs_trans_alloc( * GFP_NOFS allocation context so that we avoid lockdep false positives * by doing GFP_KERNEL allocations inside sb_start_intwrite(). */ +retry: tp = kmem_cache_zalloc(xfs_trans_zone, GFP_KERNEL | __GFP_NOFAIL); if (!(flags & XFS_TRANS_NO_WRITECOUNT)) sb_start_intwrite(mp->m_super); + xfs_trans_set_context(tp); /* * Zero-reservation ("empty") transactions can't modify anything, so @@ -289,7 +286,9 @@ xfs_trans_alloc( tp->t_firstblock = NULLFSBLOCK; error = xfs_trans_reserve(tp, resp, blocks, rtextents); - if (error == -ENOSPC) { + if (error == -ENOSPC && want_retry) { + xfs_trans_cancel(tp); + /* * We weren't able to reserve enough space for the transaction. * Flush the other speculative space allocations to free space. @@ -297,8 +296,11 @@ xfs_trans_alloc( * other locks. */ error = xfs_blockgc_free_space(mp, NULL); - if (!error) - error = xfs_trans_reserve(tp, resp, blocks, rtextents); + if (error) + return error; + + want_retry = false; + goto retry; } if (error) { xfs_trans_cancel(tp); @@ -893,7 +895,6 @@ __xfs_trans_commit( xfs_log_commit_cil(mp, tp, &commit_lsn, regrant); - current_restore_flags_nested(&tp->t_pflags, PF_MEMALLOC_NOFS); xfs_trans_free(tp); /* @@ -925,7 +926,6 @@ out_unreserve: xfs_log_ticket_ungrant(mp->m_log, tp->t_ticket); tp->t_ticket = NULL; } - current_restore_flags_nested(&tp->t_pflags, PF_MEMALLOC_NOFS); xfs_trans_free_items(tp, !!error); xfs_trans_free(tp); @@ -985,9 +985,6 @@ xfs_trans_cancel( tp->t_ticket = NULL; } - /* mark this thread as no longer being in a transaction */ - current_restore_flags_nested(&tp->t_pflags, PF_MEMALLOC_NOFS); - xfs_trans_free_items(tp, dirty); xfs_trans_free(tp); } diff --git a/fs/xfs/xfs_trans.h b/fs/xfs/xfs_trans.h index 8b03fbfe9a1b..9dd745cf77c9 100644 --- a/fs/xfs/xfs_trans.h +++ b/fs/xfs/xfs_trans.h @@ -281,4 +281,34 @@ int xfs_trans_alloc_ichange(struct xfs_inode *ip, struct xfs_dquot *udqp, struct xfs_dquot *gdqp, struct xfs_dquot *pdqp, bool force, struct xfs_trans **tpp); +static inline void +xfs_trans_set_context( + struct xfs_trans *tp) +{ + ASSERT(current->journal_info == NULL); + tp->t_pflags = memalloc_nofs_save(); + current->journal_info = tp; +} + +static inline void +xfs_trans_clear_context( + struct xfs_trans *tp) +{ + if (current->journal_info == tp) { + memalloc_nofs_restore(tp->t_pflags); + current->journal_info = NULL; + } +} + +static inline void +xfs_trans_switch_context( + struct xfs_trans *old_tp, + struct xfs_trans *new_tp) +{ + ASSERT(current->journal_info == old_tp); + new_tp->t_pflags = old_tp->t_pflags; + old_tp->t_pflags = 0; + current->journal_info = new_tp; +} + #endif /* __XFS_TRANS_H__ */ diff --git a/include/linux/bio.h b/include/linux/bio.h index 5b468f2242ff..983ed2fe7c85 100644 --- a/include/linux/bio.h +++ b/include/linux/bio.h @@ -20,7 +20,12 @@ #define BIO_BUG_ON #endif -#define BIO_MAX_PAGES 256 +#define BIO_MAX_PAGES 256U + +static inline unsigned int bio_max_segs(unsigned int nr_segs) +{ + return min(nr_segs, BIO_MAX_PAGES); +} #define bio_prio(bio) (bio)->bi_ioprio #define bio_set_prio(bio, prio) ((bio)->bi_ioprio = prio) diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h index 69035e9f632b..c032cfe133c7 100644 --- a/include/linux/blkdev.h +++ b/include/linux/blkdev.h @@ -462,7 +462,6 @@ struct request_queue { #ifdef CONFIG_PM struct device *dev; enum rpm_status rpm_status; - unsigned int nr_pending; #endif /* diff --git a/include/linux/blktrace_api.h b/include/linux/blktrace_api.h index 05556573b896..a083e15df608 100644 --- a/include/linux/blktrace_api.h +++ b/include/linux/blktrace_api.h @@ -23,8 +23,6 @@ struct blk_trace { u32 pid; u32 dev; struct dentry *dir; - struct dentry *dropped_file; - struct dentry *msg_file; struct list_head running_list; atomic_t dropped; }; @@ -119,7 +117,7 @@ struct compat_blk_user_trace_setup { #endif -extern void blk_fill_rwbs(char *rwbs, unsigned int op, int bytes); +void blk_fill_rwbs(char *rwbs, unsigned int op); static inline sector_t blk_rq_trace_sector(struct request *rq) { diff --git a/include/linux/cpuhotplug.h b/include/linux/cpuhotplug.h index 9129ba231423..f14adb882338 100644 --- a/include/linux/cpuhotplug.h +++ b/include/linux/cpuhotplug.h @@ -186,6 +186,7 @@ enum cpuhp_state { CPUHP_AP_PERF_POWERPC_TRACE_IMC_ONLINE, CPUHP_AP_PERF_POWERPC_HV_24x7_ONLINE, CPUHP_AP_PERF_POWERPC_HV_GPCI_ONLINE, + CPUHP_AP_PERF_CSKY_ONLINE, CPUHP_AP_WATCHDOG_ONLINE, CPUHP_AP_WORKQUEUE_ONLINE, CPUHP_AP_RCUTREE_ONLINE, diff --git a/include/linux/io_uring.h b/include/linux/io_uring.h index 2eb6d19de336..51ede771cd99 100644 --- a/include/linux/io_uring.h +++ b/include/linux/io_uring.h @@ -5,23 +5,6 @@ #include <linux/sched.h> #include <linux/xarray.h> -struct io_identity { - struct files_struct *files; - struct mm_struct *mm; -#ifdef CONFIG_BLK_CGROUP - struct cgroup_subsys_state *blkcg_css; -#endif - const struct cred *creds; - struct nsproxy *nsproxy; - struct fs_struct *fs; - unsigned long fsize; -#ifdef CONFIG_AUDIT - kuid_t loginuid; - unsigned int sessionid; -#endif - refcount_t count; -}; - struct io_wq_work_node { struct io_wq_work_node *next; }; @@ -36,9 +19,8 @@ struct io_uring_task { struct xarray xa; struct wait_queue_head wait; struct file *last; + void *io_wq; struct percpu_counter inflight; - struct io_identity __identity; - struct io_identity *identity; atomic_t in_idle; bool sqpoll; @@ -61,7 +43,7 @@ static inline void io_uring_task_cancel(void) } static inline void io_uring_files_cancel(struct files_struct *files) { - if (current->io_uring && !xa_empty(¤t->io_uring->xa)) + if (current->io_uring) __io_uring_files_cancel(files); } static inline void io_uring_free(struct task_struct *tsk) diff --git a/include/linux/mount.h b/include/linux/mount.h index 161f4419db6c..5d92a7e1a742 100644 --- a/include/linux/mount.h +++ b/include/linux/mount.h @@ -86,7 +86,6 @@ struct path; extern int mnt_want_write(struct vfsmount *mnt); extern int mnt_want_write_file(struct file *file); -extern int mnt_clone_write(struct vfsmount *mnt); extern void mnt_drop_write(struct vfsmount *mnt); extern void mnt_drop_write_file(struct file *file); extern void mntput(struct vfsmount *mnt); diff --git a/include/linux/net.h b/include/linux/net.h index 9e2324efc26a..ba736b457a06 100644 --- a/include/linux/net.h +++ b/include/linux/net.h @@ -42,8 +42,6 @@ struct net; #define SOCK_PASSCRED 3 #define SOCK_PASSSEC 4 -#define PROTO_CMSG_DATA_ONLY 0x0001 - #ifndef ARCH_HAS_SOCKET_TYPES /** * enum sock_type - Socket types @@ -138,7 +136,6 @@ typedef int (*sk_read_actor_t)(read_descriptor_t *, struct sk_buff *, struct proto_ops { int family; - unsigned int flags; struct module *owner; int (*release) (struct socket *sock); int (*bind) (struct socket *sock, diff --git a/include/linux/sched.h b/include/linux/sched.h index 26f499810dfa..ef00bb22164c 100644 --- a/include/linux/sched.h +++ b/include/linux/sched.h @@ -895,6 +895,9 @@ struct task_struct { /* CLONE_CHILD_CLEARTID: */ int __user *clear_child_tid; + /* PF_IO_WORKER */ + void *pf_io_worker; + u64 utime; u64 stime; #ifdef CONFIG_ARCH_HAS_SCALED_CPUTIME diff --git a/include/scsi/libiscsi.h b/include/scsi/libiscsi.h index b3bbd10eb3f0..02f966e9358f 100644 --- a/include/scsi/libiscsi.h +++ b/include/scsi/libiscsi.h @@ -187,7 +187,7 @@ struct iscsi_conn { struct iscsi_task *task; /* xmit task in progress */ /* xmit */ - spinlock_t taskqueuelock; /* protects the next three lists */ + /* items must be added/deleted under frwd lock */ struct list_head mgmtqueue; /* mgmt (control) xmit queue */ struct list_head cmdqueue; /* data-path cmd queue */ struct list_head requeue; /* tasks needing another run */ @@ -332,7 +332,7 @@ struct iscsi_session { * cmdsn, queued_cmdsn * * session resources: * * - cmdpool kfifo_out , * - * - mgmtpool, */ + * - mgmtpool, queues */ spinlock_t back_lock; /* protects cmdsn_exp * * cmdsn_max, * * cmdpool kfifo_in */ @@ -395,6 +395,8 @@ extern struct Scsi_Host *iscsi_host_alloc(struct scsi_host_template *sht, extern void iscsi_host_remove(struct Scsi_Host *shost); extern void iscsi_host_free(struct Scsi_Host *shost); extern int iscsi_target_alloc(struct scsi_target *starget); +extern int iscsi_host_get_max_scsi_cmds(struct Scsi_Host *shost, + uint16_t requested_cmds_max); /* * session management diff --git a/include/target/target_core_backend.h b/include/target/target_core_backend.h index 6336780d83a7..ce2fba49c95d 100644 --- a/include/target/target_core_backend.h +++ b/include/target/target_core_backend.h @@ -72,6 +72,7 @@ int transport_backend_register(const struct target_backend_ops *); void target_backend_unregister(const struct target_backend_ops *); void target_complete_cmd(struct se_cmd *, u8); +void target_set_cmd_data_length(struct se_cmd *, int); void target_complete_cmd_with_length(struct se_cmd *, u8, int); void transport_copy_sense_to_cmd(struct se_cmd *, unsigned char *); diff --git a/include/trace/events/bcache.h b/include/trace/events/bcache.h index e41c611d6d3b..899fdacf57b9 100644 --- a/include/trace/events/bcache.h +++ b/include/trace/events/bcache.h @@ -28,7 +28,7 @@ DECLARE_EVENT_CLASS(bcache_request, __entry->sector = bio->bi_iter.bi_sector; __entry->orig_sector = bio->bi_iter.bi_sector - 16; __entry->nr_sector = bio->bi_iter.bi_size >> 9; - blk_fill_rwbs(__entry->rwbs, bio->bi_opf, bio->bi_iter.bi_size); + blk_fill_rwbs(__entry->rwbs, bio->bi_opf); ), TP_printk("%d,%d %s %llu + %u (from %d,%d @ %llu)", @@ -102,7 +102,7 @@ DECLARE_EVENT_CLASS(bcache_bio, __entry->dev = bio_dev(bio); __entry->sector = bio->bi_iter.bi_sector; __entry->nr_sector = bio->bi_iter.bi_size >> 9; - blk_fill_rwbs(__entry->rwbs, bio->bi_opf, bio->bi_iter.bi_size); + blk_fill_rwbs(__entry->rwbs, bio->bi_opf); ), TP_printk("%d,%d %s %llu + %u", @@ -137,7 +137,7 @@ TRACE_EVENT(bcache_read, __entry->dev = bio_dev(bio); __entry->sector = bio->bi_iter.bi_sector; __entry->nr_sector = bio->bi_iter.bi_size >> 9; - blk_fill_rwbs(__entry->rwbs, bio->bi_opf, bio->bi_iter.bi_size); + blk_fill_rwbs(__entry->rwbs, bio->bi_opf); __entry->cache_hit = hit; __entry->bypass = bypass; ), @@ -168,7 +168,7 @@ TRACE_EVENT(bcache_write, __entry->inode = inode; __entry->sector = bio->bi_iter.bi_sector; __entry->nr_sector = bio->bi_iter.bi_size >> 9; - blk_fill_rwbs(__entry->rwbs, bio->bi_opf, bio->bi_iter.bi_size); + blk_fill_rwbs(__entry->rwbs, bio->bi_opf); __entry->writeback = writeback; __entry->bypass = bypass; ), @@ -238,7 +238,7 @@ TRACE_EVENT(bcache_journal_write, __entry->sector = bio->bi_iter.bi_sector; __entry->nr_sector = bio->bi_iter.bi_size >> 9; __entry->nr_keys = keys; - blk_fill_rwbs(__entry->rwbs, bio->bi_opf, bio->bi_iter.bi_size); + blk_fill_rwbs(__entry->rwbs, bio->bi_opf); ), TP_printk("%d,%d %s %llu + %u keys %u", diff --git a/include/trace/events/block.h b/include/trace/events/block.h index 0d782663a005..cc5ab96a7471 100644 --- a/include/trace/events/block.h +++ b/include/trace/events/block.h @@ -89,7 +89,7 @@ TRACE_EVENT(block_rq_requeue, __entry->sector = blk_rq_trace_sector(rq); __entry->nr_sector = blk_rq_trace_nr_sectors(rq); - blk_fill_rwbs(__entry->rwbs, rq->cmd_flags, blk_rq_bytes(rq)); + blk_fill_rwbs(__entry->rwbs, rq->cmd_flags); __get_str(cmd)[0] = '\0'; ), @@ -133,7 +133,7 @@ TRACE_EVENT(block_rq_complete, __entry->nr_sector = nr_bytes >> 9; __entry->error = error; - blk_fill_rwbs(__entry->rwbs, rq->cmd_flags, nr_bytes); + blk_fill_rwbs(__entry->rwbs, rq->cmd_flags); __get_str(cmd)[0] = '\0'; ), @@ -166,7 +166,7 @@ DECLARE_EVENT_CLASS(block_rq, __entry->nr_sector = blk_rq_trace_nr_sectors(rq); __entry->bytes = blk_rq_bytes(rq); - blk_fill_rwbs(__entry->rwbs, rq->cmd_flags, blk_rq_bytes(rq)); + blk_fill_rwbs(__entry->rwbs, rq->cmd_flags); __get_str(cmd)[0] = '\0'; memcpy(__entry->comm, current->comm, TASK_COMM_LEN); ), @@ -196,7 +196,7 @@ DEFINE_EVENT(block_rq, block_rq_insert, /** * block_rq_issue - issue pending block IO request operation to device driver - * @rq: block IO operation operation request + * @rq: block IO operation request * * Called when block operation request @rq from queue @q is sent to a * device driver for processing. @@ -210,7 +210,7 @@ DEFINE_EVENT(block_rq, block_rq_issue, /** * block_rq_merge - merge request with another one in the elevator - * @rq: block IO operation operation request + * @rq: block IO operation request * * Called when block operation request @rq from queue @q is merged to another * request queued in the elevator. @@ -249,7 +249,7 @@ TRACE_EVENT(block_bio_complete, __entry->sector = bio->bi_iter.bi_sector; __entry->nr_sector = bio_sectors(bio); __entry->error = blk_status_to_errno(bio->bi_status); - blk_fill_rwbs(__entry->rwbs, bio->bi_opf, bio->bi_iter.bi_size); + blk_fill_rwbs(__entry->rwbs, bio->bi_opf); ), TP_printk("%d,%d %s %llu + %u [%d]", @@ -276,7 +276,7 @@ DECLARE_EVENT_CLASS(block_bio, __entry->dev = bio_dev(bio); __entry->sector = bio->bi_iter.bi_sector; __entry->nr_sector = bio_sectors(bio); - blk_fill_rwbs(__entry->rwbs, bio->bi_opf, bio->bi_iter.bi_size); + blk_fill_rwbs(__entry->rwbs, bio->bi_opf); memcpy(__entry->comm, current->comm, TASK_COMM_LEN); ), @@ -433,7 +433,7 @@ TRACE_EVENT(block_split, __entry->dev = bio_dev(bio); __entry->sector = bio->bi_iter.bi_sector; __entry->new_sector = new_sector; - blk_fill_rwbs(__entry->rwbs, bio->bi_opf, bio->bi_iter.bi_size); + blk_fill_rwbs(__entry->rwbs, bio->bi_opf); memcpy(__entry->comm, current->comm, TASK_COMM_LEN); ), @@ -474,7 +474,7 @@ TRACE_EVENT(block_bio_remap, __entry->nr_sector = bio_sectors(bio); __entry->old_dev = dev; __entry->old_sector = from; - blk_fill_rwbs(__entry->rwbs, bio->bi_opf, bio->bi_iter.bi_size); + blk_fill_rwbs(__entry->rwbs, bio->bi_opf); ), TP_printk("%d,%d %s %llu + %u <- (%d,%d) %llu", @@ -518,7 +518,7 @@ TRACE_EVENT(block_rq_remap, __entry->old_dev = dev; __entry->old_sector = from; __entry->nr_bios = blk_rq_count_bios(rq); - blk_fill_rwbs(__entry->rwbs, rq->cmd_flags, blk_rq_bytes(rq)); + blk_fill_rwbs(__entry->rwbs, rq->cmd_flags); ), TP_printk("%d,%d %s %llu + %u <- (%d,%d) %llu %u", diff --git a/include/uapi/linux/io_uring.h b/include/uapi/linux/io_uring.h index ac4e1738a9af..2514eb6b1cf2 100644 --- a/include/uapi/linux/io_uring.h +++ b/include/uapi/linux/io_uring.h @@ -262,6 +262,7 @@ struct io_uring_params { #define IORING_FEAT_POLL_32BITS (1U << 6) #define IORING_FEAT_SQPOLL_NONFIXED (1U << 7) #define IORING_FEAT_EXT_ARG (1U << 8) +#define IORING_FEAT_NATIVE_WORKERS (1U << 9) /* * io_uring_register(2) opcodes and arguments diff --git a/kernel/audit_fsnotify.c b/kernel/audit_fsnotify.c index 5b3f01da172b..60739d5e3373 100644 --- a/kernel/audit_fsnotify.c +++ b/kernel/audit_fsnotify.c @@ -84,7 +84,7 @@ struct audit_fsnotify_mark *audit_alloc_mark(struct audit_krule *krule, char *pa dentry = kern_path_locked(pathname, &path); if (IS_ERR(dentry)) - return (void *)dentry; /* returning an error */ + return ERR_CAST(dentry); /* returning an error */ inode = path.dentry->d_inode; inode_unlock(inode); diff --git a/kernel/ptrace.c b/kernel/ptrace.c index 61db50f7ca86..821cf1723814 100644 --- a/kernel/ptrace.c +++ b/kernel/ptrace.c @@ -375,7 +375,7 @@ static int ptrace_attach(struct task_struct *task, long request, audit_ptrace(task); retval = -EPERM; - if (unlikely(task->flags & PF_KTHREAD)) + if (unlikely(task->flags & (PF_KTHREAD | PF_IO_WORKER))) goto out; if (same_thread_group(task, current)) goto out; diff --git a/kernel/signal.c b/kernel/signal.c index 5ad8566534e7..ba4d1ef39a9e 100644 --- a/kernel/signal.c +++ b/kernel/signal.c @@ -91,7 +91,7 @@ static bool sig_task_ignored(struct task_struct *t, int sig, bool force) return true; /* Only allow kernel generated signals to this kthread */ - if (unlikely((t->flags & PF_KTHREAD) && + if (unlikely((t->flags & (PF_KTHREAD | PF_IO_WORKER)) && (handler == SIG_KTHREAD_KERNEL) && !force)) return true; @@ -1096,7 +1096,7 @@ static int __send_signal(int sig, struct kernel_siginfo *info, struct task_struc /* * Skip useless siginfo allocation for SIGKILL and kernel threads. */ - if ((sig == SIGKILL) || (t->flags & PF_KTHREAD)) + if ((sig == SIGKILL) || (t->flags & (PF_KTHREAD | PF_IO_WORKER))) goto out_set; /* diff --git a/kernel/trace/blktrace.c b/kernel/trace/blktrace.c index c286c13bd31a..c221e4c3f625 100644 --- a/kernel/trace/blktrace.c +++ b/kernel/trace/blktrace.c @@ -312,8 +312,6 @@ record_it: static void blk_trace_free(struct blk_trace *bt) { - debugfs_remove(bt->msg_file); - debugfs_remove(bt->dropped_file); relay_close(bt->rchan); debugfs_remove(bt->dir); free_percpu(bt->sequence); @@ -545,10 +543,8 @@ static int do_blk_trace_setup(struct request_queue *q, char *name, dev_t dev, INIT_LIST_HEAD(&bt->running_list); ret = -EIO; - bt->dropped_file = debugfs_create_file("dropped", 0444, dir, bt, - &blk_dropped_fops); - - bt->msg_file = debugfs_create_file("msg", 0222, dir, bt, &blk_msg_fops); + debugfs_create_file("dropped", 0444, dir, bt, &blk_dropped_fops); + debugfs_create_file("msg", 0222, dir, bt, &blk_msg_fops); bt->rchan = relay_open("trace", dir, buts->buf_size, buts->buf_nr, &blk_relay_callbacks, bt); @@ -1868,7 +1864,17 @@ void blk_trace_remove_sysfs(struct device *dev) #ifdef CONFIG_EVENT_TRACING -void blk_fill_rwbs(char *rwbs, unsigned int op, int bytes) +/** + * blk_fill_rwbs - Fill the buffer rwbs by mapping op to character string. + * @rwbs: buffer to be filled + * @op: REQ_OP_XXX for the tracepoint + * + * Description: + * Maps the REQ_OP_XXX to character and fills the buffer provided by the + * caller with resulting string. + * + **/ +void blk_fill_rwbs(char *rwbs, unsigned int op) { int i = 0; diff --git a/net/ipv4/af_inet.c b/net/ipv4/af_inet.c index a02ce89b56b5..1355e6c0d567 100644 --- a/net/ipv4/af_inet.c +++ b/net/ipv4/af_inet.c @@ -1021,7 +1021,6 @@ static int inet_compat_ioctl(struct socket *sock, unsigned int cmd, unsigned lon const struct proto_ops inet_stream_ops = { .family = PF_INET, - .flags = PROTO_CMSG_DATA_ONLY, .owner = THIS_MODULE, .release = inet_release, .bind = inet_bind, diff --git a/net/ipv6/af_inet6.c b/net/ipv6/af_inet6.c index 1fb75f01756c..802f5111805a 100644 --- a/net/ipv6/af_inet6.c +++ b/net/ipv6/af_inet6.c @@ -665,7 +665,6 @@ int inet6_recvmsg(struct socket *sock, struct msghdr *msg, size_t size, const struct proto_ops inet6_stream_ops = { .family = PF_INET6, - .flags = PROTO_CMSG_DATA_ONLY, .owner = THIS_MODULE, .release = inet6_release, .bind = inet6_bind, diff --git a/net/socket.c b/net/socket.c index 23c7842389de..84a8049c2b09 100644 --- a/net/socket.c +++ b/net/socket.c @@ -2413,10 +2413,6 @@ static int ___sys_sendmsg(struct socket *sock, struct user_msghdr __user *msg, long __sys_sendmsg_sock(struct socket *sock, struct msghdr *msg, unsigned int flags) { - /* disallow ancillary data requests from this path */ - if (msg->msg_control || msg->msg_controllen) - return -EINVAL; - return ____sys_sendmsg(sock, msg, flags, NULL, 0); } @@ -2625,12 +2621,6 @@ long __sys_recvmsg_sock(struct socket *sock, struct msghdr *msg, struct user_msghdr __user *umsg, struct sockaddr __user *uaddr, unsigned int flags) { - if (msg->msg_control || msg->msg_controllen) { - /* disallow ancillary data reqs unless cmsg is plain data */ - if (!(sock->ops->flags & PROTO_CMSG_DATA_ONLY)) - return -EINVAL; - } - return ____sys_recvmsg(sock, msg, umsg, uaddr, flags, 0); } |