From 3386bc8aed825e9f1f65ce38df4b109b2019b71a Mon Sep 17 00:00:00 2001 From: Andy Lutomirski Date: Mon, 4 Dec 2017 15:07:25 +0100 Subject: [PATCH] x86/entry/64: Create a per-CPU SYSCALL entry trampoline Handling SYSCALL is tricky: the SYSCALL handler is entered with every single register (except FLAGS), including RSP, live. It somehow needs to set RSP to point to a valid stack, which means it needs to save the user RSP somewhere and find its own stack pointer. The canonical way to do this is with SWAPGS, which lets us access percpu data using the %gs prefix. With PAGE_TABLE_ISOLATION-like pagetable switching, this is problematic. Without a scratch register, switching CR3 is impossible, so %gs-based percpu memory would need to be mapped in the user pagetables. Doing that without information leaks is difficult or impossible. Instead, use a different sneaky trick. Map a copy of the first part of the SYSCALL asm at a different address for each CPU. Now RIP varies depending on the CPU, so we can use RIP-relative memory access to access percpu memory. By putting the relevant information (one scratch slot and the stack address) at a constant offset relative to RIP, we can make SYSCALL work without relying on %gs. A nice thing about this approach is that we can easily switch it on and off if we want pagetable switching to be configurable. The compat variant of SYSCALL doesn't have this problem in the first place -- there are plenty of scratch registers, since we don't care about preserving r8-r15. This patch therefore doesn't touch SYSCALL32 at all. This patch actually seems to be a small speedup. With this patch, SYSCALL touches an extra cache line and an extra virtual page, but the pipeline no longer stalls waiting for SWAPGS. It seems that, at least in a tight loop, the latter outweights the former. Thanks to David Laight for an optimization tip. Signed-off-by: Andy Lutomirski Signed-off-by: Thomas Gleixner Reviewed-by: Thomas Gleixner Reviewed-by: Borislav Petkov Cc: Boris Ostrovsky Cc: Borislav Petkov Cc: Brian Gerst Cc: Dave Hansen Cc: Dave Hansen Cc: David Laight Cc: Denys Vlasenko Cc: Eduardo Valentin Cc: Greg KH Cc: H. Peter Anvin Cc: Josh Poimboeuf Cc: Juergen Gross Cc: Linus Torvalds Cc: Peter Zijlstra Cc: Rik van Riel Cc: Will Deacon Cc: aliguori@amazon.com Cc: daniel.gruss@iaik.tugraz.at Cc: hughd@google.com Cc: keescook@google.com Link: https://lkml.kernel.org/r/20171204150606.403607157@linutronix.de Signed-off-by: Ingo Molnar --- arch/x86/entry/entry_64.S | 58 +++++++++++++++++++++++++++++++++++ arch/x86/include/asm/fixmap.h | 2 ++ arch/x86/kernel/asm-offsets.c | 1 + arch/x86/kernel/cpu/common.c | 15 ++++++++- arch/x86/kernel/vmlinux.lds.S | 9 ++++++ 5 files changed, 84 insertions(+), 1 deletion(-) diff --git a/arch/x86/entry/entry_64.S b/arch/x86/entry/entry_64.S index 42a9379f7acb7..2582984ffb4b2 100644 --- a/arch/x86/entry/entry_64.S +++ b/arch/x86/entry/entry_64.S @@ -136,6 +136,64 @@ END(native_usergs_sysret64) * with them due to bugs in both AMD and Intel CPUs. */ + .pushsection .entry_trampoline, "ax" + +/* + * The code in here gets remapped into cpu_entry_area's trampoline. This means + * that the assembler and linker have the wrong idea as to where this code + * lives (and, in fact, it's mapped more than once, so it's not even at a + * fixed address). So we can't reference any symbols outside the entry + * trampoline and expect it to work. + * + * Instead, we carefully abuse %rip-relative addressing. + * _entry_trampoline(%rip) refers to the start of the remapped) entry + * trampoline. We can thus find cpu_entry_area with this macro: + */ + +#define CPU_ENTRY_AREA \ + _entry_trampoline - CPU_ENTRY_AREA_entry_trampoline(%rip) + +/* The top word of the SYSENTER stack is hot and is usable as scratch space. */ +#define RSP_SCRATCH CPU_ENTRY_AREA_tss + CPU_TSS_SYSENTER_stack + \ + SIZEOF_SYSENTER_stack - 8 + CPU_ENTRY_AREA + +ENTRY(entry_SYSCALL_64_trampoline) + UNWIND_HINT_EMPTY + swapgs + + /* Stash the user RSP. */ + movq %rsp, RSP_SCRATCH + + /* Load the top of the task stack into RSP */ + movq CPU_ENTRY_AREA_tss + TSS_sp1 + CPU_ENTRY_AREA, %rsp + + /* Start building the simulated IRET frame. */ + pushq $__USER_DS /* pt_regs->ss */ + pushq RSP_SCRATCH /* pt_regs->sp */ + pushq %r11 /* pt_regs->flags */ + pushq $__USER_CS /* pt_regs->cs */ + pushq %rcx /* pt_regs->ip */ + + /* + * x86 lacks a near absolute jump, and we can't jump to the real + * entry text with a relative jump. We could push the target + * address and then use retq, but this destroys the pipeline on + * many CPUs (wasting over 20 cycles on Sandy Bridge). Instead, + * spill RDI and restore it in a second-stage trampoline. + */ + pushq %rdi + movq $entry_SYSCALL_64_stage2, %rdi + jmp *%rdi +END(entry_SYSCALL_64_trampoline) + + .popsection + +ENTRY(entry_SYSCALL_64_stage2) + UNWIND_HINT_EMPTY + popq %rdi + jmp entry_SYSCALL_64_after_hwframe +END(entry_SYSCALL_64_stage2) + ENTRY(entry_SYSCALL_64) UNWIND_HINT_EMPTY /* diff --git a/arch/x86/include/asm/fixmap.h b/arch/x86/include/asm/fixmap.h index 84558b611ad33..6a699474c2c7d 100644 --- a/arch/x86/include/asm/fixmap.h +++ b/arch/x86/include/asm/fixmap.h @@ -61,6 +61,8 @@ struct cpu_entry_area { * of the TSS region. */ struct tss_struct tss; + + char entry_trampoline[PAGE_SIZE]; }; #define CPU_ENTRY_AREA_PAGES (sizeof(struct cpu_entry_area) / PAGE_SIZE) diff --git a/arch/x86/kernel/asm-offsets.c b/arch/x86/kernel/asm-offsets.c index 55858b277cf62..61b1af88ac070 100644 --- a/arch/x86/kernel/asm-offsets.c +++ b/arch/x86/kernel/asm-offsets.c @@ -101,4 +101,5 @@ void common(void) { /* Layout info for cpu_entry_area */ OFFSET(CPU_ENTRY_AREA_tss, cpu_entry_area, tss); + OFFSET(CPU_ENTRY_AREA_entry_trampoline, cpu_entry_area, entry_trampoline); } diff --git a/arch/x86/kernel/cpu/common.c b/arch/x86/kernel/cpu/common.c index 57968880e39b6..430f950b0b7fa 100644 --- a/arch/x86/kernel/cpu/common.c +++ b/arch/x86/kernel/cpu/common.c @@ -486,6 +486,8 @@ DEFINE_PER_CPU(struct cpu_entry_area *, cpu_entry_area); static inline void setup_cpu_entry_area(int cpu) { #ifdef CONFIG_X86_64 + extern char _entry_trampoline[]; + /* On 64-bit systems, we use a read-only fixmap GDT. */ pgprot_t gdt_prot = PAGE_KERNEL_RO; #else @@ -532,6 +534,11 @@ static inline void setup_cpu_entry_area(int cpu) #ifdef CONFIG_X86_32 this_cpu_write(cpu_entry_area, get_cpu_entry_area(cpu)); #endif + +#ifdef CONFIG_X86_64 + __set_fixmap(get_cpu_entry_area_index(cpu, entry_trampoline), + __pa_symbol(_entry_trampoline), PAGE_KERNEL_RX); +#endif } /* Load the original GDT from the per-cpu structure */ @@ -1395,10 +1402,16 @@ static DEFINE_PER_CPU_PAGE_ALIGNED(char, exception_stacks /* May not be marked __init: used by software suspend */ void syscall_init(void) { + extern char _entry_trampoline[]; + extern char entry_SYSCALL_64_trampoline[]; + int cpu = smp_processor_id(); + unsigned long SYSCALL64_entry_trampoline = + (unsigned long)get_cpu_entry_area(cpu)->entry_trampoline + + (entry_SYSCALL_64_trampoline - _entry_trampoline); wrmsr(MSR_STAR, 0, (__USER32_CS << 16) | __KERNEL_CS); - wrmsrl(MSR_LSTAR, (unsigned long)entry_SYSCALL_64); + wrmsrl(MSR_LSTAR, SYSCALL64_entry_trampoline); #ifdef CONFIG_IA32_EMULATION wrmsrl(MSR_CSTAR, (unsigned long)entry_SYSCALL_compat); diff --git a/arch/x86/kernel/vmlinux.lds.S b/arch/x86/kernel/vmlinux.lds.S index a4009fb9be872..d2a8b5a24a44a 100644 --- a/arch/x86/kernel/vmlinux.lds.S +++ b/arch/x86/kernel/vmlinux.lds.S @@ -107,6 +107,15 @@ SECTIONS SOFTIRQENTRY_TEXT *(.fixup) *(.gnu.warning) + +#ifdef CONFIG_X86_64 + . = ALIGN(PAGE_SIZE); + _entry_trampoline = .; + *(.entry_trampoline) + . = ALIGN(PAGE_SIZE); + ASSERT(. - _entry_trampoline == PAGE_SIZE, "entry trampoline is too big"); +#endif + /* End of text section */ _etext = .; } :text = 0x9090 -- 2.39.5