maxjhandsome
/
llvm-project
forked from OSchip/llvm-project
1
0
Fork 0
Code Issues Pull Requests Packages Releases Wiki Activity
llvm-project/llvm/test/CodeGen/X86/patchpoint.ll

88 lines
3.6 KiB
LLVM
Raw Normal View History

[Stackmaps] Make ithe frame-pointer required for stackmaps. Do not eliminate the frame pointer if there is a stackmap or patchpoint in the function. All stackmap references should be FP relative. This fixes PR21107. llvm-svn: 218920
2014-10-03 06:21:49 +08:00
; RUN: llc -mtriple=x86_64-apple-darwin -mcpu=corei7 < %s | FileCheck %s
Change the fast-isel-abort option from bool to int to enable "levels" Summary: Currently fast-isel-abort will only abort for regular instructions, and just warn for function calls, terminators, function arguments. There is already fast-isel-abort-args but nothing for calls and terminators. This change turns the fast-isel-abort options into an integer option, so that multiple levels of strictness can be defined. This will help no being surprised when the "abort" option indeed does not abort, and enables the possibility to write test that verifies that no intrinsics are forgotten by fast-isel. Reviewers: resistor, echristo Subscribers: jfb, llvm-commits Differential Revision: http://reviews.llvm.org/D7941 From: Mehdi Amini <mehdi.amini@apple.com> llvm-svn: 230775
2015-02-28 02:32:11 +08:00
; RUN: llc -mtriple=x86_64-apple-darwin -mcpu=corei7 -fast-isel -fast-isel-abort=1 < %s | FileCheck %s
Add support for stack map generation in the X86 backend. Originally implemented by Lang Hames. llvm-svn: 193811
2013-11-01 06:11:56 +08:00
; Trivial patchpoint codegen
;
define i64 @trivial_patchpoint_codegen(i64 %p1, i64 %p2, i64 %p3, i64 %p4) {
entry:
[Stackmap] Materialize the jump address within the patchpoint noop slide. This patch moves the jump address materialization inside the noop slide. This enables patching of the materialization itself or its complete removal. This patch also adds the ability to define scratch registers that can be used safely by the code called from the patchpoint intrinsic. At least one scratch register is required, because that one is used for the materialization of the jump address. This patch depends on D2009. Differential Revision: http://llvm-reviews.chandlerc.com/D2074 Reviewed by Andy llvm-svn: 194306
2013-11-09 09:51:33 +08:00
; CHECK-LABEL: trivial_patchpoint_codegen:
; CHECK: movabsq $-559038736, %r11
; CHECK-NEXT: callq *%r11
[Stackmap] Emit multi-byte nops for X86. llvm-svn: 196334
2013-12-04 08:39:08 +08:00
; CHECK-NEXT: xchgw %ax, %ax
Add support for stack map generation in the X86 backend. Originally implemented by Lang Hames. llvm-svn: 193811
2013-11-01 06:11:56 +08:00
; CHECK: movq %rax, %[[REG:r.+]]
[Stackmap] Materialize the jump address within the patchpoint noop slide. This patch moves the jump address materialization inside the noop slide. This enables patching of the materialization itself or its complete removal. This patch also adds the ability to define scratch registers that can be used safely by the code called from the patchpoint intrinsic. At least one scratch register is required, because that one is used for the materialization of the jump address. This patch depends on D2009. Differential Revision: http://llvm-reviews.chandlerc.com/D2074 Reviewed by Andy llvm-svn: 194306
2013-11-09 09:51:33 +08:00
; CHECK: callq *%r11
[Stackmap] Emit multi-byte nops for X86. llvm-svn: 196334
2013-12-04 08:39:08 +08:00
; CHECK-NEXT: xchgw %ax, %ax
Add support for stack map generation in the X86 backend. Originally implemented by Lang Hames. llvm-svn: 193811
2013-11-01 06:11:56 +08:00
; CHECK: movq %[[REG]], %rax
; CHECK: ret
%resolveCall2 = inttoptr i64 -559038736 to i8*
Grow the stackmap/patchpoint format to hold 64-bit IDs. llvm-svn: 197255
2013-12-14 02:37:10 +08:00
%result = tail call i64 (i64, i32, i8*, i32, ...)* @llvm.experimental.patchpoint.i64(i64 2, i32 15, i8* %resolveCall2, i32 4, i64 %p1, i64 %p2, i64 %p3, i64 %p4)
Add support for stack map generation in the X86 backend. Originally implemented by Lang Hames. llvm-svn: 193811
2013-11-01 06:11:56 +08:00
%resolveCall3 = inttoptr i64 -559038737 to i8*
Grow the stackmap/patchpoint format to hold 64-bit IDs. llvm-svn: 197255
2013-12-14 02:37:10 +08:00
tail call void (i64, i32, i8*, i32, ...)* @llvm.experimental.patchpoint.void(i64 3, i32 15, i8* %resolveCall3, i32 2, i64 %p1, i64 %result)
Add support for stack map generation in the X86 backend. Originally implemented by Lang Hames. llvm-svn: 193811
2013-11-01 06:11:56 +08:00
ret i64 %result
}
; Caller frame metadata with stackmaps. This should not be optimized
; as a leaf function.
;
These test cases for experimental features are a bit too darwin-specific still. Use a triple. llvm-svn: 193820
2013-11-01 06:46:51 +08:00
; CHECK-LABEL: caller_meta_leaf
Platform proof a test case. llvm-svn: 194788
2013-11-15 13:52:56 +08:00
; CHECK: subq $32, %rsp
Add support for stack map generation in the X86 backend. Originally implemented by Lang Hames. llvm-svn: 193811
2013-11-01 06:11:56 +08:00
; CHECK: Ltmp
Platform proof a test case. llvm-svn: 194788
2013-11-15 13:52:56 +08:00
; CHECK: addq $32, %rsp
Add support for stack map generation in the X86 backend. Originally implemented by Lang Hames. llvm-svn: 193811
2013-11-01 06:11:56 +08:00
; CHECK: ret
define void @caller_meta_leaf() {
entry:
%metadata = alloca i64, i32 3, align 8
store i64 11, i64* %metadata
store i64 12, i64* %metadata
store i64 13, i64* %metadata
Grow the stackmap/patchpoint format to hold 64-bit IDs. llvm-svn: 197255
2013-12-14 02:37:10 +08:00
call void (i64, i32, ...)* @llvm.experimental.stackmap(i64 4, i32 0, i64* %metadata)
Add support for stack map generation in the X86 backend. Originally implemented by Lang Hames. llvm-svn: 193811
2013-11-01 06:11:56 +08:00
ret void
}
Slightly change the way stackmap and patchpoint intrinsics are lowered. MorphNodeTo is not safe to call during DAG building. It eagerly deletes dependent DAG nodes which invalidates the NodeMap. We could expose a safe interface for morphing nodes, but I don't think it's worth it. Just create a new MachineNode and replaceAllUsesWith. My understaning of the SD design has been that we want to support early target opcode selection. That isn't very well supported, but generally works. It seems reasonable to rely on this feature even if it isn't widely used. llvm-svn: 194102
2013-11-06 06:44:04 +08:00
; Test patchpoints reusing the same TargetConstant.
; <rdar:15390785> Assertion failed: (CI.getNumArgOperands() >= NumArgs + 4)
; There is no way to verify this, since it depends on memory allocation.
; But I think it's useful to include as a working example.
define i64 @testLowerConstant(i64 %arg, i64 %tmp2, i64 %tmp10, i64* %tmp33, i64 %tmp79) {
entry:
%tmp80 = add i64 %tmp79, -16
%tmp81 = inttoptr i64 %tmp80 to i64*
[opaque pointer type] Add textual IR support for explicit type parameter to load instruction Essentially the same as the GEP change in r230786. A similar migration script can be used to update test cases, though a few more test case improvements/changes were required this time around: (r229269-r229278) import fileinput import sys import re pat = re.compile(r"((?:=|:|^)\s*load (?:atomic )?(?:volatile )?(.*?))(| addrspace\(\d+\) *)\*($| *(?:%|@|null|undef|blockaddress|getelementptr|addrspacecast|bitcast|inttoptr|\[\[[a-zA-Z]|\{\{).*$)") for line in sys.stdin: sys.stdout.write(re.sub(pat, r"\1, \2\3*\4", line)) Reviewers: rafael, dexonsmith, grosser Differential Revision: http://reviews.llvm.org/D7649 llvm-svn: 230794
2015-02-28 05:17:42 +08:00
%tmp82 = load i64, i64* %tmp81, align 8
Grow the stackmap/patchpoint format to hold 64-bit IDs. llvm-svn: 197255
2013-12-14 02:37:10 +08:00
tail call void (i64, i32, ...)* @llvm.experimental.stackmap(i64 14, i32 5, i64 %arg, i64 %tmp2, i64 %tmp10, i64 %tmp82)
tail call void (i64, i32, i8*, i32, ...)* @llvm.experimental.patchpoint.void(i64 15, i32 30, i8* null, i32 3, i64 %arg, i64 %tmp10, i64 %tmp82)
[opaque pointer type] Add textual IR support for explicit type parameter to load instruction Essentially the same as the GEP change in r230786. A similar migration script can be used to update test cases, though a few more test case improvements/changes were required this time around: (r229269-r229278) import fileinput import sys import re pat = re.compile(r"((?:=|:|^)\s*load (?:atomic )?(?:volatile )?(.*?))(| addrspace\(\d+\) *)\*($| *(?:%|@|null|undef|blockaddress|getelementptr|addrspacecast|bitcast|inttoptr|\[\[[a-zA-Z]|\{\{).*$)") for line in sys.stdin: sys.stdout.write(re.sub(pat, r"\1, \2\3*\4", line)) Reviewers: rafael, dexonsmith, grosser Differential Revision: http://reviews.llvm.org/D7649 llvm-svn: 230794
2015-02-28 05:17:42 +08:00
%tmp83 = load i64, i64* %tmp33, align 8
Slightly change the way stackmap and patchpoint intrinsics are lowered. MorphNodeTo is not safe to call during DAG building. It eagerly deletes dependent DAG nodes which invalidates the NodeMap. We could expose a safe interface for morphing nodes, but I don't think it's worth it. Just create a new MachineNode and replaceAllUsesWith. My understaning of the SD design has been that we want to support early target opcode selection. That isn't very well supported, but generally works. It seems reasonable to rely on this feature even if it isn't widely used. llvm-svn: 194102
2013-11-06 06:44:04 +08:00
%tmp84 = add i64 %tmp83, -24
%tmp85 = inttoptr i64 %tmp84 to i64*
[opaque pointer type] Add textual IR support for explicit type parameter to load instruction Essentially the same as the GEP change in r230786. A similar migration script can be used to update test cases, though a few more test case improvements/changes were required this time around: (r229269-r229278) import fileinput import sys import re pat = re.compile(r"((?:=|:|^)\s*load (?:atomic )?(?:volatile )?(.*?))(| addrspace\(\d+\) *)\*($| *(?:%|@|null|undef|blockaddress|getelementptr|addrspacecast|bitcast|inttoptr|\[\[[a-zA-Z]|\{\{).*$)") for line in sys.stdin: sys.stdout.write(re.sub(pat, r"\1, \2\3*\4", line)) Reviewers: rafael, dexonsmith, grosser Differential Revision: http://reviews.llvm.org/D7649 llvm-svn: 230794
2015-02-28 05:17:42 +08:00
%tmp86 = load i64, i64* %tmp85, align 8
Grow the stackmap/patchpoint format to hold 64-bit IDs. llvm-svn: 197255
2013-12-14 02:37:10 +08:00
tail call void (i64, i32, ...)* @llvm.experimental.stackmap(i64 17, i32 5, i64 %arg, i64 %tmp10, i64 %tmp86)
tail call void (i64, i32, i8*, i32, ...)* @llvm.experimental.patchpoint.void(i64 18, i32 30, i8* null, i32 3, i64 %arg, i64 %tmp10, i64 %tmp86)
Slightly change the way stackmap and patchpoint intrinsics are lowered. MorphNodeTo is not safe to call during DAG building. It eagerly deletes dependent DAG nodes which invalidates the NodeMap. We could expose a safe interface for morphing nodes, but I don't think it's worth it. Just create a new MachineNode and replaceAllUsesWith. My understaning of the SD design has been that we want to support early target opcode selection. That isn't very well supported, but generally works. It seems reasonable to rely on this feature even if it isn't widely used. llvm-svn: 194102
2013-11-06 06:44:04 +08:00
ret i64 10
}
Minor extension to llvm.experimental.patchpoint: don't require a call. If a null call target is provided, don't emit a dummy call. This allows the runtime to reserve as little nop space as it needs without the requirement of emitting a call. llvm-svn: 194676
2013-11-14 14:54:10 +08:00
; Test small patchpoints that don't emit calls.
define void @small_patchpoint_codegen(i64 %p1, i64 %p2, i64 %p3, i64 %p4) {
entry:
; CHECK-LABEL: small_patchpoint_codegen:
; CHECK: Ltmp
[Stackmap] Emit multi-byte nops for X86. llvm-svn: 196334
2013-12-04 08:39:08 +08:00
; CHECK: nopl 8(%rax,%rax)
Minor extension to llvm.experimental.patchpoint: don't require a call. If a null call target is provided, don't emit a dummy call. This allows the runtime to reserve as little nop space as it needs without the requirement of emitting a call. llvm-svn: 194676
2013-11-14 14:54:10 +08:00
; CHECK-NEXT: popq
; CHECK-NEXT: ret
Grow the stackmap/patchpoint format to hold 64-bit IDs. llvm-svn: 197255
2013-12-14 02:37:10 +08:00
%result = tail call i64 (i64, i32, i8*, i32, ...)* @llvm.experimental.patchpoint.i64(i64 5, i32 5, i8* null, i32 2, i64 %p1, i64 %p2)
Minor extension to llvm.experimental.patchpoint: don't require a call. If a null call target is provided, don't emit a dummy call. This allows the runtime to reserve as little nop space as it needs without the requirement of emitting a call. llvm-svn: 194676
2013-11-14 14:54:10 +08:00
ret void
}
[FastISel] Fix the patchpoint intrinsic lowering in FastISel for large target addresses. This fixes a mistake where I accidentially dropped the upper 32bit of a 64bit pointer during FastISel lowering of the patchpoint intrinsic. llvm-svn: 214367
2014-07-31 08:11:16 +08:00
; Test large target address.
define i64 @large_target_address_patchpoint_codegen() {
entry:
; CHECK-LABEL: large_target_address_patchpoint_codegen:
; CHECK: movabsq $6153737369414576827, %r11
; CHECK-NEXT: callq *%r11
%resolveCall2 = inttoptr i64 6153737369414576827 to i8*
%result = tail call i64 (i64, i32, i8*, i32, ...)* @llvm.experimental.patchpoint.i64(i64 2, i32 15, i8* %resolveCall2, i32 0)
ret i64 %result
}
Grow the stackmap/patchpoint format to hold 64-bit IDs. llvm-svn: 197255
2013-12-14 02:37:10 +08:00
declare void @llvm.experimental.stackmap(i64, i32, ...)
declare void @llvm.experimental.patchpoint.void(i64, i32, i8*, i32, ...)
declare i64 @llvm.experimental.patchpoint.i64(i64, i32, i8*, i32, ...)