[llvm] [AMDGPU] SIFixSgprCopies should not process twice VGPR to SGPR copies inserted by PHI preprocessing. (PR #134153)

[LLVM Continuous Integration via llvm-commits]

llvm-ci wrote:

LLVM Buildbot has detected a new failure on builder `premerge-monolithic-linux` running on `premerge-linux-1` while building `llvm` at step 7 "test-build-unified-tree-check-all".

Full details are available at: https://lab.llvm.org/buildbot/#/builders/153/builds/28432

<details>
<summary>Here is the relevant piece of the build log for the reference</summary>

```
Step 7 (test-build-unified-tree-check-all) failure: test (failure)
******************** TEST 'LLVM :: CodeGen/AMDGPU/fcopysign.f16.ll' FAILED ********************
Exit Code: 1

Command Output (stderr):
--
/build/buildbot/premerge-monolithic-linux/build/bin/llc -mtriple=amdgcn -mcpu=tahiti < /build/buildbot/premerge-monolithic-linux/llvm-project/llvm/test/CodeGen/AMDGPU/fcopysign.f16.ll | /build/buildbot/premerge-monolithic-linux/build/bin/FileCheck -enable-var-scope --check-prefixes=SI /build/buildbot/premerge-monolithic-linux/llvm-project/llvm/test/CodeGen/AMDGPU/fcopysign.f16.ll # RUN: at line 2
+ /build/buildbot/premerge-monolithic-linux/build/bin/llc -mtriple=amdgcn -mcpu=tahiti
+ /build/buildbot/premerge-monolithic-linux/build/bin/FileCheck -enable-var-scope --check-prefixes=SI /build/buildbot/premerge-monolithic-linux/llvm-project/llvm/test/CodeGen/AMDGPU/fcopysign.f16.ll
/build/buildbot/premerge-monolithic-linux/build/bin/llc -mtriple=amdgcn -mcpu=tonga < /build/buildbot/premerge-monolithic-linux/llvm-project/llvm/test/CodeGen/AMDGPU/fcopysign.f16.ll | /build/buildbot/premerge-monolithic-linux/build/bin/FileCheck -enable-var-scope --check-prefixes=VI /build/buildbot/premerge-monolithic-linux/llvm-project/llvm/test/CodeGen/AMDGPU/fcopysign.f16.ll # RUN: at line 3
+ /build/buildbot/premerge-monolithic-linux/build/bin/llc -mtriple=amdgcn -mcpu=tonga
+ /build/buildbot/premerge-monolithic-linux/build/bin/FileCheck -enable-var-scope --check-prefixes=VI /build/buildbot/premerge-monolithic-linux/llvm-project/llvm/test/CodeGen/AMDGPU/fcopysign.f16.ll
/build/buildbot/premerge-monolithic-linux/build/bin/llc -mtriple=amdgcn -mcpu=gfx900 < /build/buildbot/premerge-monolithic-linux/llvm-project/llvm/test/CodeGen/AMDGPU/fcopysign.f16.ll | /build/buildbot/premerge-monolithic-linux/build/bin/FileCheck -enable-var-scope --check-prefixes=GFX9 /build/buildbot/premerge-monolithic-linux/llvm-project/llvm/test/CodeGen/AMDGPU/fcopysign.f16.ll # RUN: at line 4
+ /build/buildbot/premerge-monolithic-linux/build/bin/llc -mtriple=amdgcn -mcpu=gfx900
+ /build/buildbot/premerge-monolithic-linux/build/bin/FileCheck -enable-var-scope --check-prefixes=GFX9 /build/buildbot/premerge-monolithic-linux/llvm-project/llvm/test/CodeGen/AMDGPU/fcopysign.f16.ll
/build/buildbot/premerge-monolithic-linux/build/bin/llc -mtriple=amdgcn -mcpu=gfx1100 -mattr=+real-true16 < /build/buildbot/premerge-monolithic-linux/llvm-project/llvm/test/CodeGen/AMDGPU/fcopysign.f16.ll | /build/buildbot/premerge-monolithic-linux/build/bin/FileCheck -enable-var-scope --check-prefixes=GFX11,GFX11-TRUE16 /build/buildbot/premerge-monolithic-linux/llvm-project/llvm/test/CodeGen/AMDGPU/fcopysign.f16.ll # RUN: at line 5
+ /build/buildbot/premerge-monolithic-linux/build/bin/llc -mtriple=amdgcn -mcpu=gfx1100 -mattr=+real-true16
+ /build/buildbot/premerge-monolithic-linux/build/bin/FileCheck -enable-var-scope --check-prefixes=GFX11,GFX11-TRUE16 /build/buildbot/premerge-monolithic-linux/llvm-project/llvm/test/CodeGen/AMDGPU/fcopysign.f16.ll
/build/buildbot/premerge-monolithic-linux/llvm-project/llvm/test/CodeGen/AMDGPU/fcopysign.f16.ll:1931:22: error: GFX11-TRUE16-NEXT: expected string not found in input
; GFX11-TRUE16-NEXT: s_delay_alu instid0(VALU_DEP_2) | instskip(SKIP_3) | instid1(SALU_CYCLE_1)
                     ^
<stdin>:1435:28: note: scanning from here
 v_readfirstlane_b32 s3, v0
                           ^
<stdin>:1437:2: note: possible intended match here
 s_delay_alu instid0(VALU_DEP_3) | instskip(SKIP_2) | instid1(SALU_CYCLE_1)
 ^

Input file: <stdin>
Check file: /build/buildbot/premerge-monolithic-linux/llvm-project/llvm/test/CodeGen/AMDGPU/fcopysign.f16.ll

-dump-input=help explains the following input dump.

Input was:
<<<<<<
             .
             .
             .
          1430:  v_cndmask_b32_e64 v0, 0, 1, s2 
          1431:  s_bfe_u32 s2, s3, 0xb0014 
          1432:  s_sub_i32 s3, 0x3f1, s2 
          1433:  s_delay_alu instid0(SALU_CYCLE_1) | instskip(NEXT) | instid1(VALU_DEP_2) 
          1434:  v_med3_i32 v1, s3, 0, 13 
          1435:  v_readfirstlane_b32 s3, v0 
next:1931'0                                X error: no match found
          1436:  v_mov_b32_e32 v0, s4 
next:1931'0     ~~~~~~~~~~~~~~~~~~~~~~
          1437:  s_delay_alu instid0(VALU_DEP_3) | instskip(SKIP_2) | instid1(SALU_CYCLE_1) 
next:1931'0     ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
next:1931'1      ?                                                                           possible intended match
          1438:  v_readfirstlane_b32 s6, v1 
...

```

</details>

https://github.com/llvm/llvm-project/pull/134153