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/spill-before-exec.mir

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

87 lines
3.0 KiB
Plaintext
Raw Normal View History

[RegAllocGreedy] avoid using physreg candidates that cannot be correctly spilled For the AMDGPU target if a MBB contains exec mask restore preamble, SplitEditor may get state when it cannot insert a spill instruction. E.g. for a MIR bb.100: %1 = S_OR_SAVEEXEC_B64 %2, implicit-def $exec, implicit-def $scc, implicit $exec and if the regalloc will try to allocate a virtreg to the physreg already assigned to virtreg %1, it should insert spill instruction before the S_OR_SAVEEXEC_B64 instruction. But it is not possible since can generate incorrect code in terms of exec mask. The change makes regalloc to ignore such physreg candidates. Reviewed By: rampitec Differential Revision: https://reviews.llvm.org/D52052 llvm-svn: 343004
2018-09-26 02:37:38 +08:00
# REQUIRES: asserts
# RUN: llc -mtriple=amdgcn--- -verify-machineinstrs -debug-only=regalloc -run-pass=greedy -o /dev/null %s 2>&1 | FileCheck %s
---
# Check that physreg candidate is not used since cannot be spilled in a block,
# e.g. before exec mask preamble
# CHECK: , cannot spill all interferences.
name: foo
tracksRegLiveness: true
MIR: Allow targets to serialize MachineFunctionInfo This has been a very painful missing feature that has made producing reduced testcases difficult. In particular the various registers determined for stack access during function lowering were necessary to avoid undefined register errors in a large percentage of cases. Implement a subset of the important fields that need to be preserved for AMDGPU. Most of the changes are to support targets parsing register fields and properly reporting errors. The biggest sort-of bug remaining is for fields that can be initialized from the IR section will be overwritten by a default initialized machineFunctionInfo section. Another remaining bug is the machineFunctionInfo section is still printed even if empty. llvm-svn: 356215
2019-03-15 06:54:43 +08:00
machineFunctionInfo:
scratchRSrcReg: $sgpr0_sgpr1_sgpr2_sgpr3
stackPtrOffsetReg: $sgpr32
[RegAllocGreedy] avoid using physreg candidates that cannot be correctly spilled For the AMDGPU target if a MBB contains exec mask restore preamble, SplitEditor may get state when it cannot insert a spill instruction. E.g. for a MIR bb.100: %1 = S_OR_SAVEEXEC_B64 %2, implicit-def $exec, implicit-def $scc, implicit $exec and if the regalloc will try to allocate a virtreg to the physreg already assigned to virtreg %1, it should insert spill instruction before the S_OR_SAVEEXEC_B64 instruction. But it is not possible since can generate incorrect code in terms of exec mask. The change makes regalloc to ignore such physreg candidates. Reviewed By: rampitec Differential Revision: https://reviews.llvm.org/D52052 llvm-svn: 343004
2018-09-26 02:37:38 +08:00
body: |
bb.0:
liveins: $sgpr96_sgpr97, $sgpr98_sgpr99, $sgpr100_sgpr101, $sgpr102_sgpr103
AMDGPU: Cleanup MIR test Remove registers section and compact block/register numbers
2020-10-23 00:10:36 +08:00
[RegAllocGreedy] avoid using physreg candidates that cannot be correctly spilled For the AMDGPU target if a MBB contains exec mask restore preamble, SplitEditor may get state when it cannot insert a spill instruction. E.g. for a MIR bb.100: %1 = S_OR_SAVEEXEC_B64 %2, implicit-def $exec, implicit-def $scc, implicit $exec and if the regalloc will try to allocate a virtreg to the physreg already assigned to virtreg %1, it should insert spill instruction before the S_OR_SAVEEXEC_B64 instruction. But it is not possible since can generate incorrect code in terms of exec mask. The change makes regalloc to ignore such physreg candidates. Reviewed By: rampitec Differential Revision: https://reviews.llvm.org/D52052 llvm-svn: 343004
2018-09-26 02:37:38 +08:00
%0:sreg_64 = COPY $sgpr102_sgpr103
AMDGPU: Cleanup MIR test Remove registers section and compact block/register numbers
2020-10-23 00:10:36 +08:00
%1:sgpr_128 = COPY $sgpr100_sgpr101_sgpr102_sgpr103
%2:sgpr_128 = COPY %1
%3:sgpr_128 = COPY %1
%4:sgpr_128 = COPY %1
%5:sgpr_128 = COPY %1
%6:sgpr_128 = COPY %1
%7:sgpr_128 = COPY %1
%8:sgpr_128 = COPY %1
%9:sgpr_128 = COPY %1
%10:sgpr_128 = COPY %1
%11:sgpr_128 = COPY %1
%12:sgpr_128 = COPY %1
%13:sgpr_128 = COPY %1
%14:sgpr_128 = COPY %1
%15:sgpr_128 = COPY %1
%16:sgpr_128 = COPY %1
%17:sgpr_128 = COPY %1
%18:sgpr_128 = COPY %1
%19:sgpr_128 = COPY %1
%20:sgpr_128 = COPY %1
%21:sgpr_128 = COPY %1
%22:sgpr_128 = COPY %1
%23:sgpr_128 = COPY %1
%24:sgpr_128 = COPY %1
%25:sgpr_128 = COPY %1
%26:sgpr_128 = COPY %1
[RegAllocGreedy] avoid using physreg candidates that cannot be correctly spilled For the AMDGPU target if a MBB contains exec mask restore preamble, SplitEditor may get state when it cannot insert a spill instruction. E.g. for a MIR bb.100: %1 = S_OR_SAVEEXEC_B64 %2, implicit-def $exec, implicit-def $scc, implicit $exec and if the regalloc will try to allocate a virtreg to the physreg already assigned to virtreg %1, it should insert spill instruction before the S_OR_SAVEEXEC_B64 instruction. But it is not possible since can generate incorrect code in terms of exec mask. The change makes regalloc to ignore such physreg candidates. Reviewed By: rampitec Differential Revision: https://reviews.llvm.org/D52052 llvm-svn: 343004
2018-09-26 02:37:38 +08:00
S_BRANCH %bb.1
bb.1:
liveins: $sgpr96_sgpr97, $sgpr98_sgpr99, $sgpr102_sgpr103
AMDGPU: Cleanup MIR test Remove registers section and compact block/register numbers
2020-10-23 00:10:36 +08:00
%0:sreg_64 = S_OR_SAVEEXEC_B64 $sgpr96_sgpr97, implicit-def $exec, implicit-def $scc, implicit $exec
[RegAllocGreedy] avoid using physreg candidates that cannot be correctly spilled For the AMDGPU target if a MBB contains exec mask restore preamble, SplitEditor may get state when it cannot insert a spill instruction. E.g. for a MIR bb.100: %1 = S_OR_SAVEEXEC_B64 %2, implicit-def $exec, implicit-def $scc, implicit $exec and if the regalloc will try to allocate a virtreg to the physreg already assigned to virtreg %1, it should insert spill instruction before the S_OR_SAVEEXEC_B64 instruction. But it is not possible since can generate incorrect code in terms of exec mask. The change makes regalloc to ignore such physreg candidates. Reviewed By: rampitec Differential Revision: https://reviews.llvm.org/D52052 llvm-svn: 343004
2018-09-26 02:37:38 +08:00
$exec = S_XOR_B64_term $exec, %0, implicit-def $scc
AMDGPU: Cleanup MIR test Remove registers section and compact block/register numbers
2020-10-23 00:10:36 +08:00
SI_MASK_BRANCH %bb.3, implicit $exec
[RegAllocGreedy] avoid using physreg candidates that cannot be correctly spilled For the AMDGPU target if a MBB contains exec mask restore preamble, SplitEditor may get state when it cannot insert a spill instruction. E.g. for a MIR bb.100: %1 = S_OR_SAVEEXEC_B64 %2, implicit-def $exec, implicit-def $scc, implicit $exec and if the regalloc will try to allocate a virtreg to the physreg already assigned to virtreg %1, it should insert spill instruction before the S_OR_SAVEEXEC_B64 instruction. But it is not possible since can generate incorrect code in terms of exec mask. The change makes regalloc to ignore such physreg candidates. Reviewed By: rampitec Differential Revision: https://reviews.llvm.org/D52052 llvm-svn: 343004
2018-09-26 02:37:38 +08:00
S_BRANCH %bb.2
bb.2:
liveins: $sgpr98_sgpr99, $sgpr102_sgpr103
AMDGPU: Cleanup MIR test Remove registers section and compact block/register numbers
2020-10-23 00:10:36 +08:00
[RegAllocGreedy] avoid using physreg candidates that cannot be correctly spilled For the AMDGPU target if a MBB contains exec mask restore preamble, SplitEditor may get state when it cannot insert a spill instruction. E.g. for a MIR bb.100: %1 = S_OR_SAVEEXEC_B64 %2, implicit-def $exec, implicit-def $scc, implicit $exec and if the regalloc will try to allocate a virtreg to the physreg already assigned to virtreg %1, it should insert spill instruction before the S_OR_SAVEEXEC_B64 instruction. But it is not possible since can generate incorrect code in terms of exec mask. The change makes regalloc to ignore such physreg candidates. Reviewed By: rampitec Differential Revision: https://reviews.llvm.org/D52052 llvm-svn: 343004
2018-09-26 02:37:38 +08:00
%0:sreg_64 = S_OR_SAVEEXEC_B64 $sgpr98_sgpr99, implicit-def $exec, implicit-def $scc, implicit $exec
$exec = S_XOR_B64_term $exec, %0, implicit-def $scc
AMDGPU: Cleanup MIR test Remove registers section and compact block/register numbers
2020-10-23 00:10:36 +08:00
SI_MASK_BRANCH %bb.3, implicit $exec
S_BRANCH %bb.4
[RegAllocGreedy] avoid using physreg candidates that cannot be correctly spilled For the AMDGPU target if a MBB contains exec mask restore preamble, SplitEditor may get state when it cannot insert a spill instruction. E.g. for a MIR bb.100: %1 = S_OR_SAVEEXEC_B64 %2, implicit-def $exec, implicit-def $scc, implicit $exec and if the regalloc will try to allocate a virtreg to the physreg already assigned to virtreg %1, it should insert spill instruction before the S_OR_SAVEEXEC_B64 instruction. But it is not possible since can generate incorrect code in terms of exec mask. The change makes regalloc to ignore such physreg candidates. Reviewed By: rampitec Differential Revision: https://reviews.llvm.org/D52052 llvm-svn: 343004
2018-09-26 02:37:38 +08:00
AMDGPU: Cleanup MIR test Remove registers section and compact block/register numbers
2020-10-23 00:10:36 +08:00
bb.3:
[RegAllocGreedy] avoid using physreg candidates that cannot be correctly spilled For the AMDGPU target if a MBB contains exec mask restore preamble, SplitEditor may get state when it cannot insert a spill instruction. E.g. for a MIR bb.100: %1 = S_OR_SAVEEXEC_B64 %2, implicit-def $exec, implicit-def $scc, implicit $exec and if the regalloc will try to allocate a virtreg to the physreg already assigned to virtreg %1, it should insert spill instruction before the S_OR_SAVEEXEC_B64 instruction. But it is not possible since can generate incorrect code in terms of exec mask. The change makes regalloc to ignore such physreg candidates. Reviewed By: rampitec Differential Revision: https://reviews.llvm.org/D52052 llvm-svn: 343004
2018-09-26 02:37:38 +08:00
liveins: $sgpr102_sgpr103
AMDGPU: Cleanup MIR test Remove registers section and compact block/register numbers
2020-10-23 00:10:36 +08:00
[RegAllocGreedy] avoid using physreg candidates that cannot be correctly spilled For the AMDGPU target if a MBB contains exec mask restore preamble, SplitEditor may get state when it cannot insert a spill instruction. E.g. for a MIR bb.100: %1 = S_OR_SAVEEXEC_B64 %2, implicit-def $exec, implicit-def $scc, implicit $exec and if the regalloc will try to allocate a virtreg to the physreg already assigned to virtreg %1, it should insert spill instruction before the S_OR_SAVEEXEC_B64 instruction. But it is not possible since can generate incorrect code in terms of exec mask. The change makes regalloc to ignore such physreg candidates. Reviewed By: rampitec Differential Revision: https://reviews.llvm.org/D52052 llvm-svn: 343004
2018-09-26 02:37:38 +08:00
%0:sreg_64 = S_OR_SAVEEXEC_B64 $sgpr102_sgpr103, implicit-def $exec, implicit-def $scc, implicit $exec
$exec = S_XOR_B64_term $exec, %0, implicit-def $scc
AMDGPU: Cleanup MIR test Remove registers section and compact block/register numbers
2020-10-23 00:10:36 +08:00
S_BRANCH %bb.4
[RegAllocGreedy] avoid using physreg candidates that cannot be correctly spilled For the AMDGPU target if a MBB contains exec mask restore preamble, SplitEditor may get state when it cannot insert a spill instruction. E.g. for a MIR bb.100: %1 = S_OR_SAVEEXEC_B64 %2, implicit-def $exec, implicit-def $scc, implicit $exec and if the regalloc will try to allocate a virtreg to the physreg already assigned to virtreg %1, it should insert spill instruction before the S_OR_SAVEEXEC_B64 instruction. But it is not possible since can generate incorrect code in terms of exec mask. The change makes regalloc to ignore such physreg candidates. Reviewed By: rampitec Differential Revision: https://reviews.llvm.org/D52052 llvm-svn: 343004
2018-09-26 02:37:38 +08:00
AMDGPU: Cleanup MIR test Remove registers section and compact block/register numbers
2020-10-23 00:10:36 +08:00
bb.4:
S_CMP_EQ_U64 %1.sub0_sub1, %2.sub2_sub3, implicit-def $scc
S_CMP_EQ_U64 %3.sub0_sub1, %4.sub2_sub3, implicit-def $scc
S_CMP_EQ_U64 %5.sub0_sub1, %6.sub2_sub3, implicit-def $scc
S_CMP_EQ_U64 %7.sub0_sub1, %8.sub2_sub3, implicit-def $scc
S_CMP_EQ_U64 %9.sub0_sub1, %10.sub2_sub3, implicit-def $scc
S_CMP_EQ_U64 %11.sub0_sub1, %12.sub2_sub3, implicit-def $scc
S_CMP_EQ_U64 %13.sub0_sub1, %14.sub2_sub3, implicit-def $scc
S_CMP_EQ_U64 %15.sub0_sub1, %16.sub2_sub3, implicit-def $scc
S_CMP_EQ_U64 %17.sub0_sub1, %18.sub2_sub3, implicit-def $scc
S_CMP_EQ_U64 %19.sub0_sub1, %20.sub2_sub3, implicit-def $scc
S_CMP_EQ_U64 %21.sub0_sub1, %22.sub2_sub3, implicit-def $scc
S_CMP_EQ_U64 %23.sub0_sub1, %24.sub2_sub3, implicit-def $scc
S_CMP_EQ_U64 %25.sub0_sub1, %26.sub2_sub3, implicit-def $scc
[RegAllocGreedy] avoid using physreg candidates that cannot be correctly spilled For the AMDGPU target if a MBB contains exec mask restore preamble, SplitEditor may get state when it cannot insert a spill instruction. E.g. for a MIR bb.100: %1 = S_OR_SAVEEXEC_B64 %2, implicit-def $exec, implicit-def $scc, implicit $exec and if the regalloc will try to allocate a virtreg to the physreg already assigned to virtreg %1, it should insert spill instruction before the S_OR_SAVEEXEC_B64 instruction. But it is not possible since can generate incorrect code in terms of exec mask. The change makes regalloc to ignore such physreg candidates. Reviewed By: rampitec Differential Revision: https://reviews.llvm.org/D52052 llvm-svn: 343004
2018-09-26 02:37:38 +08:00
$vgpr0 = V_MOV_B32_e32 0, implicit $exec
AMDGPU: Fix test failing since r365512 llvm-svn: 365521
2019-07-10 01:54:34 +08:00
S_SETPC_B64_return undef $sgpr30_sgpr31, implicit %0, implicit $vgpr0
[RegAllocGreedy] avoid using physreg candidates that cannot be correctly spilled For the AMDGPU target if a MBB contains exec mask restore preamble, SplitEditor may get state when it cannot insert a spill instruction. E.g. for a MIR bb.100: %1 = S_OR_SAVEEXEC_B64 %2, implicit-def $exec, implicit-def $scc, implicit $exec and if the regalloc will try to allocate a virtreg to the physreg already assigned to virtreg %1, it should insert spill instruction before the S_OR_SAVEEXEC_B64 instruction. But it is not possible since can generate incorrect code in terms of exec mask. The change makes regalloc to ignore such physreg candidates. Reviewed By: rampitec Differential Revision: https://reviews.llvm.org/D52052 llvm-svn: 343004
2018-09-26 02:37:38 +08:00
...