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/rename-independent-subregs.mir

70 lines
2.0 KiB
Plaintext
Raw Normal View History

RenameIndependentSubregs: Fix liveness query in rewriteOperands() rewriteOperands() always performed liveness queries at the base index rather than the RegSlot/Base as apropriate for the machine operand. This could lead to illegal rewriting in some cases. llvm-svn: 277661
2016-08-04 06:37:47 +08:00
# RUN: llc -march=amdgcn -verify-machineinstrs -run-pass simple-register-coalescing,rename-independent-subregs -o - %s | FileCheck %s
CodeGen: Refactor renameDisconnectedComponents() as a pass Refactor LiveIntervals::renameDisconnectedComponents() to be a pass. Also change the name to "RenameIndependentSubregs": - renameDisconnectedComponents() worked on a MachineFunction at a time so it is a natural candidate for a machine function pass. - The algorithm is testable with a .mir test now. - This also fixes a problem where the lazy renaming as part of the MachineScheduler introduced IMPLICIT_DEF instructions after the number of a nodes in a region were counted leading to a mismatch. Differential Revision: http://reviews.llvm.org/D20507 llvm-svn: 271345
2016-06-01 06:38:06 +08:00
--- |
AMDGPU: Mark all unspecified CC functions in tests as amdgpu_kernel Currently the default C calling convention functions are treated the same as compute kernels. Make this explicit so the default calling convention can be changed to a non-kernel. Converted with perl -pi -e 's/define void/define amdgpu_kernel void/' on the relevant test directories (and undoing in one place that actually wanted a non-kernel). llvm-svn: 298444
2017-03-22 05:39:51 +08:00
define amdgpu_kernel void @test0() { ret void }
define amdgpu_kernel void @test1() { ret void }
CodeGen: Refactor renameDisconnectedComponents() as a pass Refactor LiveIntervals::renameDisconnectedComponents() to be a pass. Also change the name to "RenameIndependentSubregs": - renameDisconnectedComponents() worked on a MachineFunction at a time so it is a natural candidate for a machine function pass. - The algorithm is testable with a .mir test now. - This also fixes a problem where the lazy renaming as part of the MachineScheduler introduced IMPLICIT_DEF instructions after the number of a nodes in a region were counted leading to a mismatch. Differential Revision: http://reviews.llvm.org/D20507 llvm-svn: 271345
2016-06-01 06:38:06 +08:00
...
---
# In the test below we have two independent def+use pairs of subregister1 which
# can be moved to a new virtual register. The third def of sub1 however is used
# in combination with sub0 and needs to stay with the original vreg.
# CHECK-LABEL: name: test0
MIRParser: Use dot instead of colon to mark subregisters Change the syntax to use `%0.sub8` to denote a subregister. This seems like a more natural fit to denote subregisters; I also plan to introduce a new ":classname" syntax in upcoming patches to denote the register class of a vreg. Note that this commit disallows plain identifiers to start with a '.' character. This shouldn't affect anything as external names/IR references are all prefixed with '$'/'%', plain identifiers are only used for instruction names, register mask names and subreg indexes. Differential Revision: https://reviews.llvm.org/D22390 llvm-svn: 276815
2016-07-27 05:49:34 +08:00
# CHECK: S_NOP 0, implicit-def undef %0.sub0
# CHECK: S_NOP 0, implicit-def undef %2.sub1
# CHECK: S_NOP 0, implicit %2.sub1
# CHECK: S_NOP 0, implicit-def undef %1.sub1
# CHECK: S_NOP 0, implicit %1.sub1
# CHECK: S_NOP 0, implicit-def %0.sub1
CodeGen: Refactor renameDisconnectedComponents() as a pass Refactor LiveIntervals::renameDisconnectedComponents() to be a pass. Also change the name to "RenameIndependentSubregs": - renameDisconnectedComponents() worked on a MachineFunction at a time so it is a natural candidate for a machine function pass. - The algorithm is testable with a .mir test now. - This also fixes a problem where the lazy renaming as part of the MachineScheduler introduced IMPLICIT_DEF instructions after the number of a nodes in a region were counted leading to a mismatch. Differential Revision: http://reviews.llvm.org/D20507 llvm-svn: 271345
2016-06-01 06:38:06 +08:00
# CHECK: S_NOP 0, implicit %0
name: test0
registers:
- { id: 0, class: sreg_128 }
body: |
bb.0:
MIRParser: Use dot instead of colon to mark subregisters Change the syntax to use `%0.sub8` to denote a subregister. This seems like a more natural fit to denote subregisters; I also plan to introduce a new ":classname" syntax in upcoming patches to denote the register class of a vreg. Note that this commit disallows plain identifiers to start with a '.' character. This shouldn't affect anything as external names/IR references are all prefixed with '$'/'%', plain identifiers are only used for instruction names, register mask names and subreg indexes. Differential Revision: https://reviews.llvm.org/D22390 llvm-svn: 276815
2016-07-27 05:49:34 +08:00
S_NOP 0, implicit-def undef %0.sub0
S_NOP 0, implicit-def %0.sub1
S_NOP 0, implicit %0.sub1
S_NOP 0, implicit-def %0.sub1
S_NOP 0, implicit %0.sub1
S_NOP 0, implicit-def %0.sub1
CodeGen: Refactor renameDisconnectedComponents() as a pass Refactor LiveIntervals::renameDisconnectedComponents() to be a pass. Also change the name to "RenameIndependentSubregs": - renameDisconnectedComponents() worked on a MachineFunction at a time so it is a natural candidate for a machine function pass. - The algorithm is testable with a .mir test now. - This also fixes a problem where the lazy renaming as part of the MachineScheduler introduced IMPLICIT_DEF instructions after the number of a nodes in a region were counted leading to a mismatch. Differential Revision: http://reviews.llvm.org/D20507 llvm-svn: 271345
2016-06-01 06:38:06 +08:00
S_NOP 0, implicit %0
...
RenameIndependentSubregs: Fix liveness query in rewriteOperands() rewriteOperands() always performed liveness queries at the base index rather than the RegSlot/Base as apropriate for the machine operand. This could lead to illegal rewriting in some cases. llvm-svn: 277661
2016-08-04 06:37:47 +08:00
---
# Test for a bug where we would incorrectly query liveness at the instruction
# index in rewriteOperands(). This should pass the verifier afterwards.
# CHECK-LABEL: test1
# CHECK: bb.0
# CHECK: S_NOP 0, implicit-def undef %2.sub2
# CHECK: bb.1
# CHECK: S_NOP 0, implicit-def %2.sub1
# CHECK-NEXT: S_NOP 0, implicit-def %2.sub3
# CHECK-NEXT: S_NOP 0, implicit %2
# CHECK-NEXT: S_NOP 0, implicit-def undef %0.sub0
# CHECK-NEXT: S_NOP 0, implicit %2.sub1
# CHECK-NEXT: S_NOP 0, implicit %0.sub0
# CHECK: bb.2
# CHECK: S_NOP 0, implicit %2.sub
name: test1
registers:
- { id: 0, class: sreg_128 }
- { id: 1, class: sreg_128 }
body: |
bb.0:
S_NOP 0, implicit-def undef %0.sub2
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
S_CBRANCH_VCCNZ %bb.1, implicit undef $vcc
RenameIndependentSubregs: Fix liveness query in rewriteOperands() rewriteOperands() always performed liveness queries at the base index rather than the RegSlot/Base as apropriate for the machine operand. This could lead to illegal rewriting in some cases. llvm-svn: 277661
2016-08-04 06:37:47 +08:00
S_BRANCH %bb.2
bb.1:
S_NOP 0, implicit-def %0.sub1
S_NOP 0, implicit-def %0.sub3
%1 = COPY %0
S_NOP 0, implicit %1
S_NOP 0, implicit-def %1.sub0
S_NOP 0, implicit %1.sub1
S_NOP 0, implicit %1.sub0
bb.2:
S_NOP 0, implicit %0.sub2
...