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/AMDGPU/pal-simple-indirect-call.ll

Ignoring revisions in .git-blame-ignore-revs. Click here to bypass and see the normal blame view.

77 lines
3.1 KiB
LLVM
Raw Normal View History

[AMDGPU] Deduce attributes with the Attributor This patch introduces a pass that uses the Attributor to deduce AMDGPU specific attributes. Reviewed By: jdoerfert, arsenm Differential Revision: https://reviews.llvm.org/D104997
2021-06-28 01:41:56 +08:00
; NOTE: Assertions have been autogenerated by utils/update_test_checks.py UTC_ARGS: --function-signature --check-globals
[AMDGPU] Do not annotate features for graphics SITargetLowering::LowerFormalArguments asserts that none of these features are used for graphics calling conventions, so AnnotateKernelFeatures should not add them. Differential Revision: https://reviews.llvm.org/D101534
2021-05-03 16:33:11 +08:00
; Check that no attributes are added to graphics functions
[AMDGPU] Deduce attributes with the Attributor This patch introduces a pass that uses the Attributor to deduce AMDGPU specific attributes. Reviewed By: jdoerfert, arsenm Differential Revision: https://reviews.llvm.org/D104997
2021-06-28 01:41:56 +08:00
; RUN: opt -S -mtriple=amdgcn-amd-amdpal -amdgpu-annotate-kernel-features %s | FileCheck -check-prefixes=AKF_GCN %s
; RUN: opt -S -mtriple=amdgcn-amd-amdpal -amdgpu-attributor %s | FileCheck -check-prefixes=ATTRIBUTOR_GCN %s
[AMDGPU] Do not annotate features for graphics SITargetLowering::LowerFormalArguments asserts that none of these features are used for graphics calling conventions, so AnnotateKernelFeatures should not add them. Differential Revision: https://reviews.llvm.org/D101534
2021-05-03 16:33:11 +08:00
; Check that it doesn't crash
AMDGPU: Correct const_index_stride for wave 32 for PAL ABI Retrying after revert and fix (removed implicit def flag from operand). Now passes with expensive_checks enabled. Since there is a single scratch resource descriptor for all shaders, if there is a wave32 and a wave64 shader (for instance for VsFs pairs) then the const_index_stride will be incorrect for wave32 shaders. Differential Revision: https://reviews.llvm.org/D101830 Change-Id: Ie3b8b2921237968caca91527dd0c97b1b0cc0360
2020-02-25 05:19:15 +08:00
; RUN: llc -mtriple=amdgcn-amd-amdpal -mcpu=gfx900 < %s | FileCheck -check-prefixes=GFX9 %s
; RUN: llc -mtriple=amdgcn-amd-amdpal -mcpu=gfx1010 < %s | FileCheck -check-prefixes=GFX10 %s
[AMDGPU][GlobalISel] Allow amdgpu_gfx calling conv Calling functions from shaders already works with the SelectionDAG. Differential Revision: https://reviews.llvm.org/D103183
2021-05-27 00:50:19 +08:00
; RUN: llc -mtriple=amdgcn-amd-amdpal -mcpu=gfx1010 -global-isel < %s | FileCheck -check-prefixes=GFX10 %s
[AMDGPU] Do not annotate features for graphics SITargetLowering::LowerFormalArguments asserts that none of these features are used for graphics calling conventions, so AnnotateKernelFeatures should not add them. Differential Revision: https://reviews.llvm.org/D101534
2021-05-03 16:33:11 +08:00
target datalayout = "A5"
[AMDGPU] Deduce attributes with the Attributor This patch introduces a pass that uses the Attributor to deduce AMDGPU specific attributes. Reviewed By: jdoerfert, arsenm Differential Revision: https://reviews.llvm.org/D104997
2021-06-28 01:41:56 +08:00
[AMDGPU] Do not annotate features for graphics SITargetLowering::LowerFormalArguments asserts that none of these features are used for graphics calling conventions, so AnnotateKernelFeatures should not add them. Differential Revision: https://reviews.llvm.org/D101534
2021-05-03 16:33:11 +08:00
define amdgpu_cs void @test_simple_indirect_call() {
[AMDGPU] Deduce attributes with the Attributor This patch introduces a pass that uses the Attributor to deduce AMDGPU specific attributes. Reviewed By: jdoerfert, arsenm Differential Revision: https://reviews.llvm.org/D104997
2021-06-28 01:41:56 +08:00
; AKF_GCN-LABEL: define {{[^@]+}}@test_simple_indirect_call() {
; AKF_GCN-NEXT: [[PC:%.*]] = call i64 @llvm.amdgcn.s.getpc()
; AKF_GCN-NEXT: [[FUN:%.*]] = inttoptr i64 [[PC]] to void ()*
; AKF_GCN-NEXT: call amdgpu_gfx void [[FUN]]()
; AKF_GCN-NEXT: ret void
;
; ATTRIBUTOR_GCN-LABEL: define {{[^@]+}}@test_simple_indirect_call
; ATTRIBUTOR_GCN-SAME: () #[[ATTR0:[0-9]+]] {
; ATTRIBUTOR_GCN-NEXT: [[PC:%.*]] = call i64 @llvm.amdgcn.s.getpc()
; ATTRIBUTOR_GCN-NEXT: [[FUN:%.*]] = inttoptr i64 [[PC]] to void ()*
; ATTRIBUTOR_GCN-NEXT: call amdgpu_gfx void [[FUN]]()
; ATTRIBUTOR_GCN-NEXT: ret void
;
; Attributor adds work-group-size attribute. This should be ok.
[AMDGPU] Do not annotate features for graphics SITargetLowering::LowerFormalArguments asserts that none of these features are used for graphics calling conventions, so AnnotateKernelFeatures should not add them. Differential Revision: https://reviews.llvm.org/D101534
2021-05-03 16:33:11 +08:00
; GFX9-LABEL: test_simple_indirect_call:
; GFX9: ; %bb.0:
[AMDGPU] Changes the AMDGPU_Gfx calling convention by making the SGPRs 4..29 callee-save. This is to avoid superfluous s_movs when executing amdgpu_gfx function calls as the callee is likely not going to change the argument values. This patch changes the AMDGPU_Gfx calling convention. It defines the SGPR registers s[4:29] as callee-save and leaves some SGPRs usable for callers. The intention is to avoid unneccessary s_mov instructions for arguments the caller would otherwise save and restore in these registers. Reviewed By: sebastian-ne Differential Revision: https://reviews.llvm.org/D111637
2021-10-25 16:33:55 +08:00
; GFX9-NEXT: s_getpc_b64 s[8:9]
; GFX9-NEXT: s_mov_b32 s8, s0
; GFX9-NEXT: s_load_dwordx4 s[8:11], s[8:9], 0x10
AMDGPU: Correct const_index_stride for wave 32 for PAL ABI Retrying after revert and fix (removed implicit def flag from operand). Now passes with expensive_checks enabled. Since there is a single scratch resource descriptor for all shaders, if there is a wave32 and a wave64 shader (for instance for VsFs pairs) then the const_index_stride will be incorrect for wave32 shaders. Differential Revision: https://reviews.llvm.org/D101830 Change-Id: Ie3b8b2921237968caca91527dd0c97b1b0cc0360
2020-02-25 05:19:15 +08:00
; GFX9-NEXT: s_getpc_b64 s[4:5]
[AMDGPU] Do not annotate features for graphics SITargetLowering::LowerFormalArguments asserts that none of these features are used for graphics calling conventions, so AnnotateKernelFeatures should not add them. Differential Revision: https://reviews.llvm.org/D101534
2021-05-03 16:33:11 +08:00
; GFX9-NEXT: s_mov_b32 s32, 0
; GFX9-NEXT: s_waitcnt lgkmcnt(0)
[AMDGPU] Changes the AMDGPU_Gfx calling convention by making the SGPRs 4..29 callee-save. This is to avoid superfluous s_movs when executing amdgpu_gfx function calls as the callee is likely not going to change the argument values. This patch changes the AMDGPU_Gfx calling convention. It defines the SGPR registers s[4:29] as callee-save and leaves some SGPRs usable for callers. The intention is to avoid unneccessary s_mov instructions for arguments the caller would otherwise save and restore in these registers. Reviewed By: sebastian-ne Differential Revision: https://reviews.llvm.org/D111637
2021-10-25 16:33:55 +08:00
; GFX9-NEXT: s_add_u32 s8, s8, s0
; GFX9-NEXT: s_addc_u32 s9, s9, 0
; GFX9-NEXT: s_mov_b64 s[0:1], s[8:9]
; GFX9-NEXT: s_mov_b64 s[2:3], s[10:11]
[AMDGPU] Do not annotate features for graphics SITargetLowering::LowerFormalArguments asserts that none of these features are used for graphics calling conventions, so AnnotateKernelFeatures should not add them. Differential Revision: https://reviews.llvm.org/D101534
2021-05-03 16:33:11 +08:00
; GFX9-NEXT: s_swappc_b64 s[30:31], s[4:5]
; GFX9-NEXT: s_endpgm
[AMDGPU] Changes the AMDGPU_Gfx calling convention by making the SGPRs 4..29 callee-save. This is to avoid superfluous s_movs when executing amdgpu_gfx function calls as the callee is likely not going to change the argument values. This patch changes the AMDGPU_Gfx calling convention. It defines the SGPR registers s[4:29] as callee-save and leaves some SGPRs usable for callers. The intention is to avoid unneccessary s_mov instructions for arguments the caller would otherwise save and restore in these registers. Reviewed By: sebastian-ne Differential Revision: https://reviews.llvm.org/D111637
2021-10-25 16:33:55 +08:00
;
AMDGPU: Correct const_index_stride for wave 32 for PAL ABI Retrying after revert and fix (removed implicit def flag from operand). Now passes with expensive_checks enabled. Since there is a single scratch resource descriptor for all shaders, if there is a wave32 and a wave64 shader (for instance for VsFs pairs) then the const_index_stride will be incorrect for wave32 shaders. Differential Revision: https://reviews.llvm.org/D101830 Change-Id: Ie3b8b2921237968caca91527dd0c97b1b0cc0360
2020-02-25 05:19:15 +08:00
; GFX10-LABEL: test_simple_indirect_call:
; GFX10: ; %bb.0:
[AMDGPU] Changes the AMDGPU_Gfx calling convention by making the SGPRs 4..29 callee-save. This is to avoid superfluous s_movs when executing amdgpu_gfx function calls as the callee is likely not going to change the argument values. This patch changes the AMDGPU_Gfx calling convention. It defines the SGPR registers s[4:29] as callee-save and leaves some SGPRs usable for callers. The intention is to avoid unneccessary s_mov instructions for arguments the caller would otherwise save and restore in these registers. Reviewed By: sebastian-ne Differential Revision: https://reviews.llvm.org/D111637
2021-10-25 16:33:55 +08:00
; GFX10-NEXT: s_getpc_b64 s[8:9]
; GFX10-NEXT: s_mov_b32 s8, s0
AMDGPU: Correct const_index_stride for wave 32 for PAL ABI Retrying after revert and fix (removed implicit def flag from operand). Now passes with expensive_checks enabled. Since there is a single scratch resource descriptor for all shaders, if there is a wave32 and a wave64 shader (for instance for VsFs pairs) then the const_index_stride will be incorrect for wave32 shaders. Differential Revision: https://reviews.llvm.org/D101830 Change-Id: Ie3b8b2921237968caca91527dd0c97b1b0cc0360
2020-02-25 05:19:15 +08:00
; GFX10-NEXT: s_getpc_b64 s[4:5]
[AMDGPU] Changes the AMDGPU_Gfx calling convention by making the SGPRs 4..29 callee-save. This is to avoid superfluous s_movs when executing amdgpu_gfx function calls as the callee is likely not going to change the argument values. This patch changes the AMDGPU_Gfx calling convention. It defines the SGPR registers s[4:29] as callee-save and leaves some SGPRs usable for callers. The intention is to avoid unneccessary s_mov instructions for arguments the caller would otherwise save and restore in these registers. Reviewed By: sebastian-ne Differential Revision: https://reviews.llvm.org/D111637
2021-10-25 16:33:55 +08:00
; GFX10-NEXT: s_load_dwordx4 s[8:11], s[8:9], 0x10
AMDGPU: Correct const_index_stride for wave 32 for PAL ABI Retrying after revert and fix (removed implicit def flag from operand). Now passes with expensive_checks enabled. Since there is a single scratch resource descriptor for all shaders, if there is a wave32 and a wave64 shader (for instance for VsFs pairs) then the const_index_stride will be incorrect for wave32 shaders. Differential Revision: https://reviews.llvm.org/D101830 Change-Id: Ie3b8b2921237968caca91527dd0c97b1b0cc0360
2020-02-25 05:19:15 +08:00
; GFX10-NEXT: s_mov_b32 s32, 0
; GFX10-NEXT: s_waitcnt lgkmcnt(0)
[AMDGPU] Changes the AMDGPU_Gfx calling convention by making the SGPRs 4..29 callee-save. This is to avoid superfluous s_movs when executing amdgpu_gfx function calls as the callee is likely not going to change the argument values. This patch changes the AMDGPU_Gfx calling convention. It defines the SGPR registers s[4:29] as callee-save and leaves some SGPRs usable for callers. The intention is to avoid unneccessary s_mov instructions for arguments the caller would otherwise save and restore in these registers. Reviewed By: sebastian-ne Differential Revision: https://reviews.llvm.org/D111637
2021-10-25 16:33:55 +08:00
; GFX10-NEXT: s_bitset0_b32 s11, 21
; GFX10-NEXT: s_add_u32 s8, s8, s0
; GFX10-NEXT: s_addc_u32 s9, s9, 0
; GFX10-NEXT: s_mov_b64 s[0:1], s[8:9]
; GFX10-NEXT: s_mov_b64 s[2:3], s[10:11]
AMDGPU: Correct const_index_stride for wave 32 for PAL ABI Retrying after revert and fix (removed implicit def flag from operand). Now passes with expensive_checks enabled. Since there is a single scratch resource descriptor for all shaders, if there is a wave32 and a wave64 shader (for instance for VsFs pairs) then the const_index_stride will be incorrect for wave32 shaders. Differential Revision: https://reviews.llvm.org/D101830 Change-Id: Ie3b8b2921237968caca91527dd0c97b1b0cc0360
2020-02-25 05:19:15 +08:00
; GFX10-NEXT: s_swappc_b64 s[30:31], s[4:5]
; GFX10-NEXT: s_endpgm
[AMDGPU] Do not annotate features for graphics SITargetLowering::LowerFormalArguments asserts that none of these features are used for graphics calling conventions, so AnnotateKernelFeatures should not add them. Differential Revision: https://reviews.llvm.org/D101534
2021-05-03 16:33:11 +08:00
%pc = call i64 @llvm.amdgcn.s.getpc()
%fun = inttoptr i64 %pc to void()*
call amdgpu_gfx void %fun()
ret void
}
; Function Attrs: nounwind readnone speculatable willreturn
declare i64 @llvm.amdgcn.s.getpc() #0
attributes #0 = { nounwind readnone speculatable willreturn }
[AMDGPU] Deduce attributes with the Attributor This patch introduces a pass that uses the Attributor to deduce AMDGPU specific attributes. Reviewed By: jdoerfert, arsenm Differential Revision: https://reviews.llvm.org/D104997
2021-06-28 01:41:56 +08:00
;.
; AKF_GCN: attributes #[[ATTR0:[0-9]+]] = { nounwind readnone speculatable willreturn }
;.
; ATTRIBUTOR_GCN: attributes #[[ATTR0]] = { "uniform-work-group-size"="false" }
AMDGPU: Stop attributor adding attributes to intrinsic declarations
2021-08-13 03:19:54 +08:00
; ATTRIBUTOR_GCN: attributes #[[ATTR1:[0-9]+]] = { nounwind readnone speculatable willreturn }
[AMDGPU] Deduce attributes with the Attributor This patch introduces a pass that uses the Attributor to deduce AMDGPU specific attributes. Reviewed By: jdoerfert, arsenm Differential Revision: https://reviews.llvm.org/D104997
2021-06-28 01:41:56 +08:00
;.