mirror of
https://github.com/RPCS3/llvm-mirror.git
synced 2024-10-21 03:53:04 +02:00
01cf1a3805
The primary motivation of this change is to bring the code more closely in sync behavior wise with the assembler's version of nop emission. I'd like to eventually factor them into one, but that's hard to do when one has features the other doesn't. The longest encodeable nop on x86 is 15 bytes, but many processors - for instance all intel chips - can't decode the 15 byte form efficiently. On those processors, it's better to use either a 10 byte or 11 byte sequence depending.
52 lines
2.5 KiB
LLVM
52 lines
2.5 KiB
LLVM
; RUN: llc -mtriple=x86_64-apple-darwin -mcpu=znver1 -frame-pointer=all -filetype=obj %s -o - | llvm-objdump -d - | FileCheck %s
|
|
; RUN: llc -mtriple=x86_64-apple-darwin -mcpu=znver1 -frame-pointer=all -filetype=asm %s -o - | llvm-mc -triple=x86_64-apple-darwin -mcpu=znver1 -filetype=obj - | llvm-objdump -d - | FileCheck %s
|
|
|
|
define void @nop_test() {
|
|
entry:
|
|
; CHECK: 0: 55
|
|
; CHECK: 1: 48 89 e5
|
|
|
|
; CHECK: 4: 90
|
|
; CHECK: 5: 66 90
|
|
; CHECK: 7: 0f 1f 00
|
|
; CHECK: a: 0f 1f 40 08
|
|
; CHECK: e: 0f 1f 44 00 08
|
|
; CHECK: 13: 66 0f 1f 44 00 08
|
|
; CHECK: 19: 0f 1f 80 00 02 00 00
|
|
; CHECK: 20: 0f 1f 84 00 00 02 00 00
|
|
; CHECK: 28: 66 0f 1f 84 00 00 02 00 00
|
|
; CHECK: 31: 2e 66 0f 1f 84 00 00 02 00 00
|
|
; CHECK: 3b: 66 2e 66 0f 1f 84 00 00 02 00 00
|
|
; CHECK: 46: 66 66 2e 66 0f 1f 84 00 00 02 00 00
|
|
; CHECK: 52: 66 66 66 2e 66 0f 1f 84 00 00 02 00 00
|
|
; CHECK: 5f: 66 66 66 66 2e 66 0f 1f 84 00 00 02 00 00
|
|
; CHECK: 6d: 66 66 66 66 66 2e 66 0f 1f 84 00 00 02 00 00
|
|
|
|
; CHECK: 7c: 5d
|
|
; CHECK: 7d: c3
|
|
|
|
tail call void (i64, i32, ...) @llvm.experimental.stackmap(i64 0, i32 0)
|
|
tail call void (i64, i32, ...) @llvm.experimental.stackmap(i64 1, i32 1)
|
|
tail call void (i64, i32, ...) @llvm.experimental.stackmap(i64 2, i32 2)
|
|
tail call void (i64, i32, ...) @llvm.experimental.stackmap(i64 3, i32 3)
|
|
tail call void (i64, i32, ...) @llvm.experimental.stackmap(i64 4, i32 4)
|
|
tail call void (i64, i32, ...) @llvm.experimental.stackmap(i64 5, i32 5)
|
|
tail call void (i64, i32, ...) @llvm.experimental.stackmap(i64 6, i32 6)
|
|
tail call void (i64, i32, ...) @llvm.experimental.stackmap(i64 7, i32 7)
|
|
tail call void (i64, i32, ...) @llvm.experimental.stackmap(i64 8, i32 8)
|
|
tail call void (i64, i32, ...) @llvm.experimental.stackmap(i64 9, i32 9)
|
|
tail call void (i64, i32, ...) @llvm.experimental.stackmap(i64 10, i32 10)
|
|
tail call void (i64, i32, ...) @llvm.experimental.stackmap(i64 11, i32 11)
|
|
tail call void (i64, i32, ...) @llvm.experimental.stackmap(i64 12, i32 12)
|
|
tail call void (i64, i32, ...) @llvm.experimental.stackmap(i64 13, i32 13)
|
|
tail call void (i64, i32, ...) @llvm.experimental.stackmap(i64 14, i32 14)
|
|
tail call void (i64, i32, ...) @llvm.experimental.stackmap(i64 15, i32 15)
|
|
; Add an extra stackmap with a zero-length shadow to thwart the shadow
|
|
; optimization. This will force all 15 bytes of the previous shadow to be
|
|
; padded with nops.
|
|
tail call void (i64, i32, ...) @llvm.experimental.stackmap(i64 16, i32 0)
|
|
ret void
|
|
}
|
|
|
|
declare void @llvm.experimental.stackmap(i64, i32, ...)
|