mirror of
https://github.com/RPCS3/llvm-mirror.git
synced 2024-12-05 10:17:36 +00:00
2818e607ab
Each hwasan check requires emitting a small piece of code like this: https://clang.llvm.org/docs/HardwareAssistedAddressSanitizerDesign.html#memory-accesses The problem with this is that these code blocks typically bloat code size significantly. An obvious solution is to outline these blocks of code. In fact, this has already been implemented under the -hwasan-instrument-with-calls flag. However, as currently implemented this has a number of problems: - The functions use the same calling convention as regular C functions. This means that the backend must spill all temporary registers as required by the platform's C calling convention, even though the check only needs two registers on the hot path. - The functions take the address to be checked in a fixed register, which increases register pressure. Both of these factors can diminish the code size effect and increase the performance hit of -hwasan-instrument-with-calls. The solution that this patch implements is to involve the aarch64 backend in outlining the checks. An intrinsic and pseudo-instruction are created to represent a hwasan check. The pseudo-instruction is register allocated like any other instruction, and we allow the register allocator to select almost any register for the address to check. A particular combination of (register selection, type of check) triggers the creation in the backend of a function to handle the check for specifically that pair. The resulting functions are deduplicated by the linker. The pseudo-instruction (really the function) is specified to preserve all registers except for the registers that the AAPCS specifies may be clobbered by a call. To measure the code size and performance effect of this change, I took a number of measurements using Chromium for Android on aarch64, comparing a browser with inlined checks (the baseline) against a browser with outlined checks. Code size: Size of .text decreases from 243897420 to 171619972 bytes, or a 30% decrease. Performance: Using Chromium's blink_perf.layout microbenchmarks I measured a median performance regression of 6.24%. The fact that a perf/size tradeoff is evident here suggests that we might want to make the new behaviour conditional on -Os/-Oz. But for now I've enabled it unconditionally, my reasoning being that hwasan users typically expect a relatively large perf hit, and ~6% isn't really adding much. We may want to revisit this decision in the future, though. I also tried experimenting with varying the number of registers selectable by the hwasan check pseudo-instruction (which would result in fewer variants being created), on the hypothesis that creating fewer variants of the function would expose another perf/size tradeoff by reducing icache pressure from the check functions at the cost of register pressure. Although I did observe a code size increase with fewer registers, I did not observe a strong correlation between the number of registers and the performance of the resulting browser on the microbenchmarks, so I conclude that we might as well use ~all registers to get the maximum code size improvement. My results are below: Regs | .text size | Perf hit -----+------------+--------- ~all | 171619972 | 6.24% 16 | 171765192 | 7.03% 8 | 172917788 | 5.82% 4 | 177054016 | 6.89% Differential Revision: https://reviews.llvm.org/D56954 llvm-svn: 351920
64 lines
1.9 KiB
LLVM
64 lines
1.9 KiB
LLVM
; RUN: llc < %s | FileCheck %s
|
|
|
|
target triple = "aarch64--linux-android"
|
|
|
|
define i8* @f1(i8* %x0, i8* %x1) {
|
|
; CHECK: f1:
|
|
; CHECK: str x30, [sp, #-16]!
|
|
; CHECK-NEXT: .cfi_def_cfa_offset 16
|
|
; CHECK-NEXT: .cfi_offset w30, -16
|
|
; CHECK-NEXT: mov x9, x0
|
|
; CHECK-NEXT: bl __hwasan_check_x1_123
|
|
; CHECK-NEXT: mov x0, x1
|
|
; CHECK-NEXT: ldr x30, [sp], #16
|
|
; CHECK-NEXT: ret
|
|
call void @llvm.hwasan.check.memaccess(i8* %x0, i8* %x1, i32 123)
|
|
ret i8* %x1
|
|
}
|
|
|
|
define i8* @f2(i8* %x0, i8* %x1) {
|
|
; CHECK: f2:
|
|
; CHECK: str x30, [sp, #-16]!
|
|
; CHECK-NEXT: .cfi_def_cfa_offset 16
|
|
; CHECK-NEXT: .cfi_offset w30, -16
|
|
; CHECK-NEXT: mov x9, x1
|
|
; CHECK-NEXT: bl __hwasan_check_x0_456
|
|
; CHECK-NEXT: ldr x30, [sp], #16
|
|
; CHECK-NEXT: ret
|
|
call void @llvm.hwasan.check.memaccess(i8* %x1, i8* %x0, i32 456)
|
|
ret i8* %x0
|
|
}
|
|
|
|
declare void @llvm.hwasan.check.memaccess(i8*, i8*, i32)
|
|
|
|
; CHECK: .section .text.hot,"axG",@progbits,__hwasan_check_x0_456,comdat
|
|
; CHECK-NEXT: .type __hwasan_check_x0_456,@function
|
|
; CHECK-NEXT: .weak __hwasan_check_x0_456
|
|
; CHECK-NEXT: .hidden __hwasan_check_x0_456
|
|
; CHECK-NEXT: __hwasan_check_x0_456:
|
|
; CHECK-NEXT: ubfx x16, x0, #4, #52
|
|
; CHECK-NEXT: ldrb w16, [x9, x16]
|
|
; CHECK-NEXT: lsr x17, x0, #56
|
|
; CHECK-NEXT: cmp w16, w17
|
|
; CHECK-NEXT: b.ne .Ltmp0
|
|
; CHECK-NEXT: ret
|
|
; CHECK-NEXT: .Ltmp0:
|
|
; CHECK-NEXT: mov x1, #456
|
|
; CHECK-NEXT: b __hwasan_tag_mismatch
|
|
|
|
; CHECK: .section .text.hot,"axG",@progbits,__hwasan_check_x1_123,comdat
|
|
; CHECK-NEXT: .type __hwasan_check_x1_123,@function
|
|
; CHECK-NEXT: .weak __hwasan_check_x1_123
|
|
; CHECK-NEXT: .hidden __hwasan_check_x1_123
|
|
; CHECK-NEXT: __hwasan_check_x1_123:
|
|
; CHECK-NEXT: ubfx x16, x1, #4, #52
|
|
; CHECK-NEXT: ldrb w16, [x9, x16]
|
|
; CHECK-NEXT: lsr x17, x1, #56
|
|
; CHECK-NEXT: cmp w16, w17
|
|
; CHECK-NEXT: b.ne .Ltmp1
|
|
; CHECK-NEXT: ret
|
|
; CHECK-NEXT: .Ltmp1:
|
|
; CHECK-NEXT: mov x0, x1
|
|
; CHECK-NEXT: mov x1, #123
|
|
; CHECK-NEXT: b __hwasan_tag_mismatch
|