This is a note to let you know that I've just added the patch titled x86/entry/64: Remove thread_struct::sp0 to the 4.14-stable tree which can be found at: http://www.kernel.org/git/?p=linux/kernel/git/stable/stable-queue.git;a=summary The filename of the patch is: x86-entry-64-remove-thread_struct-sp0.patch and it can be found in the queue-4.14 subdirectory. If you, or anyone else, feels it should not be added to the stable tree, please let <stable@xxxxxxxxxxxxxxx> know about it. >From d375cf1530595e33961a8844192cddab913650e3 Mon Sep 17 00:00:00 2001 From: Andy Lutomirski <luto@xxxxxxxxxx> Date: Thu, 2 Nov 2017 00:59:16 -0700 Subject: x86/entry/64: Remove thread_struct::sp0 From: Andy Lutomirski <luto@xxxxxxxxxx> commit d375cf1530595e33961a8844192cddab913650e3 upstream. On x86_64, we can easily calculate sp0 when needed instead of storing it in thread_struct. On x86_32, a similar cleanup would be possible, but it would require cleaning up the vm86 code first, and that can wait for a later cleanup series. Signed-off-by: Andy Lutomirski <luto@xxxxxxxxxx> Cc: Borislav Petkov <bpetkov@xxxxxxx> Cc: Brian Gerst <brgerst@xxxxxxxxx> Cc: Dave Hansen <dave.hansen@xxxxxxxxx> Cc: Linus Torvalds <torvalds@xxxxxxxxxxxxxxxxxxxx> Cc: Peter Zijlstra <peterz@xxxxxxxxxxxxx> Cc: Thomas Gleixner <tglx@xxxxxxxxxxxxx> Link: http://lkml.kernel.org/r/719cd9c66c548c4350d98a90f050aee8b17f8919.1509609304.git.luto@xxxxxxxxxx Signed-off-by: Ingo Molnar <mingo@xxxxxxxxxx> Signed-off-by: Greg Kroah-Hartman <gregkh@xxxxxxxxxxxxxxxxxxx> --- arch/x86/include/asm/compat.h | 1 + arch/x86/include/asm/processor.h | 28 +++++++++------------------- arch/x86/include/asm/switch_to.h | 6 ++++++ arch/x86/kernel/process_64.c | 1 - 4 files changed, 16 insertions(+), 20 deletions(-) --- a/arch/x86/include/asm/compat.h +++ b/arch/x86/include/asm/compat.h @@ -7,6 +7,7 @@ */ #include <linux/types.h> #include <linux/sched.h> +#include <linux/sched/task_stack.h> #include <asm/processor.h> #include <asm/user32.h> #include <asm/unistd.h> --- a/arch/x86/include/asm/processor.h +++ b/arch/x86/include/asm/processor.h @@ -431,7 +431,9 @@ typedef struct { struct thread_struct { /* Cached TLS descriptors: */ struct desc_struct tls_array[GDT_ENTRY_TLS_ENTRIES]; +#ifdef CONFIG_X86_32 unsigned long sp0; +#endif unsigned long sp; #ifdef CONFIG_X86_32 unsigned long sysenter_cs; @@ -798,6 +800,13 @@ static inline void spin_lock_prefetch(co #define task_top_of_stack(task) ((unsigned long)(task_pt_regs(task) + 1)) +#define task_pt_regs(task) \ +({ \ + unsigned long __ptr = (unsigned long)task_stack_page(task); \ + __ptr += THREAD_SIZE - TOP_OF_KERNEL_STACK_PADDING; \ + ((struct pt_regs *)__ptr) - 1; \ +}) + #ifdef CONFIG_X86_32 /* * User space process size: 3GB (default). @@ -817,23 +826,6 @@ static inline void spin_lock_prefetch(co .addr_limit = KERNEL_DS, \ } -/* - * TOP_OF_KERNEL_STACK_PADDING reserves 8 bytes on top of the ring0 stack. - * This is necessary to guarantee that the entire "struct pt_regs" - * is accessible even if the CPU haven't stored the SS/ESP registers - * on the stack (interrupt gate does not save these registers - * when switching to the same priv ring). - * Therefore beware: accessing the ss/esp fields of the - * "struct pt_regs" is possible, but they may contain the - * completely wrong values. - */ -#define task_pt_regs(task) \ -({ \ - unsigned long __ptr = (unsigned long)task_stack_page(task); \ - __ptr += THREAD_SIZE - TOP_OF_KERNEL_STACK_PADDING; \ - ((struct pt_regs *)__ptr) - 1; \ -}) - #define KSTK_ESP(task) (task_pt_regs(task)->sp) #else @@ -867,11 +859,9 @@ static inline void spin_lock_prefetch(co #define STACK_TOP_MAX TASK_SIZE_MAX #define INIT_THREAD { \ - .sp0 = TOP_OF_INIT_STACK, \ .addr_limit = KERNEL_DS, \ } -#define task_pt_regs(tsk) ((struct pt_regs *)(tsk)->thread.sp0 - 1) extern unsigned long KSTK_ESP(struct task_struct *task); #endif /* CONFIG_X86_64 */ --- a/arch/x86/include/asm/switch_to.h +++ b/arch/x86/include/asm/switch_to.h @@ -2,6 +2,8 @@ #ifndef _ASM_X86_SWITCH_TO_H #define _ASM_X86_SWITCH_TO_H +#include <linux/sched/task_stack.h> + struct task_struct; /* one of the stranger aspects of C forward declarations */ struct task_struct *__switch_to_asm(struct task_struct *prev, @@ -88,7 +90,11 @@ static inline void refresh_sysenter_cs(s /* This is used when switching tasks or entering/exiting vm86 mode. */ static inline void update_sp0(struct task_struct *task) { +#ifdef CONFIG_X86_32 load_sp0(task->thread.sp0); +#else + load_sp0(task_top_of_stack(task)); +#endif } #endif /* _ASM_X86_SWITCH_TO_H */ --- a/arch/x86/kernel/process_64.c +++ b/arch/x86/kernel/process_64.c @@ -274,7 +274,6 @@ int copy_thread_tls(unsigned long clone_ struct inactive_task_frame *frame; struct task_struct *me = current; - p->thread.sp0 = (unsigned long)task_stack_page(p) + THREAD_SIZE; childregs = task_pt_regs(p); fork_frame = container_of(childregs, struct fork_frame, regs); frame = &fork_frame->frame; Patches currently in stable-queue which might be from luto@xxxxxxxxxx are queue-4.14/x86-entry-64-simplify-reg-restore-code-in-the-standard-iret-paths.patch queue-4.14/x86-cpufeatures-fix-various-details-in-the-feature-definitions.patch queue-4.14/x86-entry-64-move-the-ist-stacks-into-struct-cpu_entry_area.patch queue-4.14/x86-dumpstack-add-get_stack_info-support-for-the-sysenter-stack.patch queue-4.14/x86-asm-don-t-use-the-confusing-.ifeq-directive.patch queue-4.14/selftests-x86-ldt_gdt-add-infrastructure-to-test-set_thread_area.patch queue-4.14/x86-entry-remap-the-tss-into-the-cpu-entry-area.patch queue-4.14/x86-entry-64-paravirt-use-paravirt-safe-macro-to-access-eflags.patch queue-4.14/x86-mm-fixmap-generalize-the-gdt-fixmap-mechanism-introduce-struct-cpu_entry_area.patch queue-4.14/x86-paravirt-dont-patch-flush_tlb_single.patch queue-4.14/x86-dumpstack-handle-stack-overflow-on-all-stacks.patch queue-4.14/x86-entry-64-use-pop-instead-of-movq-in-syscall_return_via_sysret.patch queue-4.14/x86-entry-64-return-to-userspace-from-the-trampoline-stack.patch queue-4.14/x86-paravirt-provide-a-way-to-check-for-hypervisors.patch queue-4.14/xen-x86-entry-64-add-xen-nmi-trap-entry.patch queue-4.14/x86-entry-64-remove-the-restore_c_regs_and_iret-label.patch queue-4.14/x86-entry-64-create-a-per-cpu-syscall-entry-trampoline.patch queue-4.14/x86-entry-64-remove-the-restore_..._regs-infrastructure.patch queue-4.14/x86-xen-64-x86-entry-64-clean-up-sp-code-in-cpu_initialize_context.patch queue-4.14/x86-entry-add-task_top_of_stack-to-find-the-top-of-a-task-s-stack.patch queue-4.14/x86-entry-64-remove-all-remaining-direct-thread_struct-sp0-reads.patch queue-4.14/x86-boot-relocate-definition-of-the-initial-state-of-cr0.patch queue-4.14/x86-entry-64-stop-initializing-tss.sp0-at-boot.patch queue-4.14/x86-entry-64-de-xen-ify-our-nmi-code.patch queue-4.14/objtool-don-t-report-end-of-section-error-after-an-empty-unwind-hint.patch queue-4.14/x86-entry-64-pass-sp0-directly-to-load_sp0.patch queue-4.14/x86-entry-64-use-a-per-cpu-trampoline-stack-for-idt-entries.patch queue-4.14/x86-entry-64-move-swapgs-into-the-common-iret-to-usermode-path.patch queue-4.14/x86-entry-64-shorten-test-instructions.patch queue-4.14/x86-cpufeature-add-user-mode-instruction-prevention-definitions.patch queue-4.14/x86-cpufeatures-make-cpu-bugs-sticky.patch queue-4.14/x86-espfix-64-stop-assuming-that-pt_regs-is-on-the-entry-stack.patch queue-4.14/x86-traps-use-a-new-on_thread_stack-helper-to-clean-up-an-assertion.patch queue-4.14/x86-xen-fix-xen-head-elf-annotations.patch queue-4.14/x86-entry-64-remove-thread_struct-sp0.patch queue-4.14/x86-entry-move-sysenter_stack-to-the-beginning-of-struct-tss_struct.patch queue-4.14/x86-entry-64-allocate-and-enable-the-sysenter-stack.patch queue-4.14/x86-unwinder-orc-dont-bail-on-stack-overflow.patch queue-4.14/x86-head-fix-head-elf-function-annotations.patch queue-4.14/selftests-x86-ldt_gdt-run-most-existing-ldt-test-cases-against-the-gdt-as-well.patch queue-4.14/x86-entry-32-pull-the-msr_ia32_sysenter_cs-update-code-out-of-native_load_sp0.patch queue-4.14/x86-entry-64-split-the-iret-to-user-and-iret-to-kernel-paths.patch queue-4.14/x86-entry-64-shrink-paranoid_exit_restore-and-make-labels-local.patch queue-4.14/x86-head-add-unwind-hint-annotations.patch queue-4.14/x86-head-remove-unused-bad_address-code.patch queue-4.14/x86-xen-add-unwind-hint-annotations.patch queue-4.14/x86-kasan-64-teach-kasan-about-the-cpu_entry_area.patch queue-4.14/x86-head-remove-confusing-comment.patch queue-4.14/x86-entry-64-remove-the-sysenter-stack-canary.patch queue-4.14/x86-mm-kasan-don-t-use-vmemmap_populate-to-initialize-shadow.patch queue-4.14/x86-entry-gdt-put-per-cpu-gdt-remaps-in-ascending-order.patch queue-4.14/x86-entry-fix-assumptions-that-the-hw-tss-is-at-the-beginning-of-cpu_tss.patch queue-4.14/x86-cpufeatures-re-tabulate-the-x86_feature-definitions.patch queue-4.14/x86-entry-32-fix-cpu_current_top_of_stack-initialization-at-boot.patch queue-4.14/ptrace-x86-make-user_64bit_mode-available-to-32-bit-builds.patch queue-4.14/x86-entry-64-make-cpu_entry_area.tss-read-only.patch queue-4.14/x86-mm-relocate-page-fault-error-codes-to-traps.h.patch queue-4.14/x86-unwinder-handle-stack-overflows-more-gracefully.patch queue-4.14/x86-entry-64-use-pop-instead-of-mov-to-restore-regs-on-nmi-return.patch queue-4.14/x86-irq-64-print-the-offending-ip-in-the-stack-overflow-warning.patch queue-4.14/x86-entry-clean-up-the-sysenter_stack-code.patch queue-4.14/x86-entry-64-merge-the-fast-and-slow-sysret-paths.patch queue-4.14/x86-entry-64-separate-cpu_current_top_of_stack-from-tss.sp0.patch queue-4.14/x86-boot-annotate-verify_cpu-as-a-callable-function.patch queue-4.14/x86-irq-remove-an-old-outdated-comment-about-context-tracking-races.patch