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/insert-waitcnts-exp.mir

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

64 lines
2.6 KiB
Plaintext
Raw Normal View History

[AMDGPU][Waitcnt] Update a few tests to use default waitcnt pass (si-insert-waitcnts) rather than old pass (si-insert-waits); this is a small step towards the overall goal of removing the old waitcnt pass, which is no longer maintained. Differential Revision: https://reviews.llvm.org/D46154 llvm-svn: 331062
2018-04-28 01:59:15 +08:00
# RUN: llc -march=amdgcn -verify-machineinstrs -run-pass si-insert-waitcnts -o - %s | FileCheck %s
AMDGPU: Refactor exp instructions Structure the definitions a bit more like the other classes. The main change here is to split EXP with the done bit set to a separate opcode, so we can set mayLoad = 1 so that it won't be reordered before the other exp stores, since this has the special constraint that if the done bit is set then this should be the last exp in she shader. Previously all exp instructions were inferred to have unmodeled side effects. llvm-svn: 288695
2016-12-06 04:23:10 +08:00
--- |
AMDGPU: Remove some uses of llvm.SI.export in tests Merge some of the old, smaller tests into more complete versions. llvm-svn: 295792
2017-02-22 08:02:21 +08:00
define amdgpu_ps <4 x float> @exp_done_waitcnt(<4 x i32> inreg, <4 x
i32> inreg, i32 inreg %w, float %v) #0 {
AMDGPU: Refactor exp instructions Structure the definitions a bit more like the other classes. The main change here is to split EXP with the done bit set to a separate opcode, so we can set mayLoad = 1 so that it won't be reordered before the other exp stores, since this has the special constraint that if the done bit is set then this should be the last exp in she shader. Previously all exp instructions were inferred to have unmodeled side effects. llvm-svn: 288695
2016-12-06 04:23:10 +08:00
%a = load volatile float, float addrspace(1)* undef
%b = load volatile float, float addrspace(1)* undef
%c = load volatile float, float addrspace(1)* undef
%d = load volatile float, float addrspace(1)* undef
AMDGPU: Remove some uses of llvm.SI.export in tests Merge some of the old, smaller tests into more complete versions. llvm-svn: 295792
2017-02-22 08:02:21 +08:00
call void @llvm.amdgcn.exp.f32(i32 15, i32 1, float %a, float %b, float %c, float %d, i1 true, i1 false)
AMDGPU: Refactor exp instructions Structure the definitions a bit more like the other classes. The main change here is to split EXP with the done bit set to a separate opcode, so we can set mayLoad = 1 so that it won't be reordered before the other exp stores, since this has the special constraint that if the done bit is set then this should be the last exp in she shader. Previously all exp instructions were inferred to have unmodeled side effects. llvm-svn: 288695
2016-12-06 04:23:10 +08:00
ret <4 x float> <float 5.000000e-01, float 1.000000e+00, float 2.000000e+00, float 4.000000e+00>
}
AMDGPU: Remove some uses of llvm.SI.export in tests Merge some of the old, smaller tests into more complete versions. llvm-svn: 295792
2017-02-22 08:02:21 +08:00
declare void @llvm.amdgcn.exp.f32(i32, i32, float, float, float, float, i1, i1) #0
AMDGPU: Refactor exp instructions Structure the definitions a bit more like the other classes. The main change here is to split EXP with the done bit set to a separate opcode, so we can set mayLoad = 1 so that it won't be reordered before the other exp stores, since this has the special constraint that if the done bit is set then this should be the last exp in she shader. Previously all exp instructions were inferred to have unmodeled side effects. llvm-svn: 288695
2016-12-06 04:23:10 +08:00
AMDGPU: Remove some uses of llvm.SI.export in tests Merge some of the old, smaller tests into more complete versions. llvm-svn: 295792
2017-02-22 08:02:21 +08:00
attributes #0 = { nounwind }
AMDGPU: Refactor exp instructions Structure the definitions a bit more like the other classes. The main change here is to split EXP with the done bit set to a separate opcode, so we can set mayLoad = 1 so that it won't be reordered before the other exp stores, since this has the special constraint that if the done bit is set then this should be the last exp in she shader. Previously all exp instructions were inferred to have unmodeled side effects. llvm-svn: 288695
2016-12-06 04:23:10 +08:00
...
---
# CHECK-LABEL: name: exp_done_waitcnt{{$}}
# CHECK: EXP_DONE
# CHECK-NEXT: S_WAITCNT 3855
Followup on Proposal to move MIR physical register namespace to '$' sigil. Discussed here: http://lists.llvm.org/pipermail/llvm-dev/2018-January/120320.html In preparation for adding support for named vregs we are changing the sigil for physical registers in MIR to '$' from '%'. This will prevent name clashes of named physical register with named vregs. llvm-svn: 323922
2018-02-01 06:04:26 +08:00
# CHECK: $vgpr0 = V_MOV_B32
# CHECK: $vgpr1 = V_MOV_B32
# CHECK: $vgpr2 = V_MOV_B32
# CHECK: $vgpr3 = V_MOV_B32
AMDGPU: Refactor exp instructions Structure the definitions a bit more like the other classes. The main change here is to split EXP with the done bit set to a separate opcode, so we can set mayLoad = 1 so that it won't be reordered before the other exp stores, since this has the special constraint that if the done bit is set then this should be the last exp in she shader. Previously all exp instructions were inferred to have unmodeled side effects. llvm-svn: 288695
2016-12-06 04:23:10 +08:00
name: exp_done_waitcnt
[Alignment] Use llvm::Align in MachineFunction and TargetLowering - fixes mir parsing Summary: This catches malformed mir files which specify alignment as log2 instead of pow2. See https://reviews.llvm.org/D65945 for reference, This is patch is part of a series to introduce an Alignment type. See this thread for context: http://lists.llvm.org/pipermail/llvm-dev/2019-July/133851.html See this patch for the introduction of the type: https://reviews.llvm.org/D64790 Reviewers: courbet Subscribers: MatzeB, qcolombet, dschuff, arsenm, sdardis, nemanjai, jvesely, nhaehnle, hiraditya, kbarton, asb, rbar, johnrusso, simoncook, apazos, sabuasal, niosHD, jrtc27, MaskRay, zzheng, edward-jones, atanasyan, rogfer01, MartinMosbeck, brucehoult, the_o, PkmX, jocewei, jsji, Petar.Avramovic, asbirlea, s.egerton, pzheng, llvm-commits Tags: #llvm Differential Revision: https://reviews.llvm.org/D67433 llvm-svn: 371608
2019-09-11 19:16:48 +08:00
alignment: 1
AMDGPU: Refactor exp instructions Structure the definitions a bit more like the other classes. The main change here is to split EXP with the done bit set to a separate opcode, so we can set mayLoad = 1 so that it won't be reordered before the other exp stores, since this has the special constraint that if the done bit is set then this should be the last exp in she shader. Previously all exp instructions were inferred to have unmodeled side effects. llvm-svn: 288695
2016-12-06 04:23:10 +08:00
exposesReturnsTwice: false
legalized: false
regBankSelected: false
selected: false
tracksRegLiveness: true
frameInfo:
isFrameAddressTaken: false
isReturnAddressTaken: false
hasStackMap: false
hasPatchPoint: false
stackSize: 0
offsetAdjustment: 0
maxAlignment: 0
adjustsStack: false
hasCalls: false
maxCallFrameSize: 0
hasOpaqueSPAdjustment: false
hasVAStart: false
hasMustTailInVarArgFunc: false
body: |
bb.0 (%ir-block.2):
Followup on Proposal to move MIR physical register namespace to '$' sigil. Discussed here: http://lists.llvm.org/pipermail/llvm-dev/2018-January/120320.html In preparation for adding support for named vregs we are changing the sigil for physical registers in MIR to '$' from '%'. This will prevent name clashes of named physical register with named vregs. llvm-svn: 323922
2018-02-01 06:04:26 +08:00
$sgpr3 = S_MOV_B32 61440
$sgpr2 = S_MOV_B32 -1
[AMDGPU] Extend buffer intrinsics with swizzling Summary: Extend cachepolicy operand in the new VMEM buffer intrinsics to supply information whether the buffer data is swizzled. Also, propagate this information to MIR. Intrinsics updated: int_amdgcn_raw_buffer_load int_amdgcn_raw_buffer_load_format int_amdgcn_raw_buffer_store int_amdgcn_raw_buffer_store_format int_amdgcn_raw_tbuffer_load int_amdgcn_raw_tbuffer_store int_amdgcn_struct_buffer_load int_amdgcn_struct_buffer_load_format int_amdgcn_struct_buffer_store int_amdgcn_struct_buffer_store_format int_amdgcn_struct_tbuffer_load int_amdgcn_struct_tbuffer_store Furthermore, disable merging of VMEM buffer instructions in SI Load/Store optimizer, if the "swizzled" bit on the instruction is on. The default value of the bit is 0, meaning that data in buffer is linear and buffer instructions can be merged. There is no difference in the generated code with this commit. However, in the future it will be expected that front-ends use buffer intrinsics with correct "swizzled" bit set. Reviewers: arsenm, nhaehnle, tpr Reviewed By: nhaehnle Subscribers: arsenm, kzhuravl, jvesely, wdng, nhaehnle, yaxunl, dstuttard, tpr, t-tye, arphaman, jfb, Petar.Avramovic, llvm-commits Tags: #llvm Differential Revision: https://reviews.llvm.org/D68200 llvm-svn: 373491
2019-10-03 01:22:36 +08:00
$vgpr0 = BUFFER_LOAD_DWORD_OFFSET $sgpr0_sgpr1_sgpr2_sgpr3, 0, 0, 0, 0, 0, 0, 0, implicit $exec :: (volatile load 4 from `float addrspace(1)* undef`)
$vgpr1 = BUFFER_LOAD_DWORD_OFFSET $sgpr0_sgpr1_sgpr2_sgpr3, 0, 0, 0, 0, 0, 0, 0, implicit $exec :: (volatile load 4 from `float addrspace(1)* undef`)
$vgpr2 = BUFFER_LOAD_DWORD_OFFSET $sgpr0_sgpr1_sgpr2_sgpr3, 0, 0, 0, 0, 0, 0, 0, implicit $exec :: (volatile load 4 from `float addrspace(1)* undef`)
$vgpr3 = BUFFER_LOAD_DWORD_OFFSET killed $sgpr0_sgpr1_sgpr2_sgpr3, 0, 0, 0, 0, 0, 0, 0, implicit $exec :: (volatile load 4 from `float addrspace(1)* undef`)
Followup on Proposal to move MIR physical register namespace to '$' sigil. Discussed here: http://lists.llvm.org/pipermail/llvm-dev/2018-January/120320.html In preparation for adding support for named vregs we are changing the sigil for physical registers in MIR to '$' from '%'. This will prevent name clashes of named physical register with named vregs. llvm-svn: 323922
2018-02-01 06:04:26 +08:00
EXP_DONE 0, killed $vgpr0, killed $vgpr1, killed $vgpr2, killed $vgpr3, -1, -1, 15, implicit $exec
$vgpr0 = V_MOV_B32_e32 1056964608, implicit $exec
$vgpr1 = V_MOV_B32_e32 1065353216, implicit $exec
$vgpr2 = V_MOV_B32_e32 1073741824, implicit $exec
$vgpr3 = V_MOV_B32_e32 1082130432, implicit $exec
SI_RETURN_TO_EPILOG killed $vgpr0, killed $vgpr1, killed $vgpr2, killed $vgpr3
AMDGPU: Refactor exp instructions Structure the definitions a bit more like the other classes. The main change here is to split EXP with the done bit set to a separate opcode, so we can set mayLoad = 1 so that it won't be reordered before the other exp stores, since this has the special constraint that if the done bit is set then this should be the last exp in she shader. Previously all exp instructions were inferred to have unmodeled side effects. llvm-svn: 288695
2016-12-06 04:23:10 +08:00
...