[llvm] [AMDGPU][True16][CodeGen] uaddsat/usubsat true16 selection in gisel (PR #128233)

[LLVM Continuous Integration via llvm-commits]

llvm-ci wrote:

LLVM Buildbot has detected a new failure on builder `llvm-x86_64-debian-dylib` running on `gribozavr4` while building `llvm` at step 7 "test-build-unified-tree-check-llvm".

Full details are available at: https://lab.llvm.org/buildbot/#/builders/60/builds/20458

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

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

Command Output (stderr):
--
RUN: at line 2: /b/1/llvm-x86_64-debian-dylib/build/bin/llc -mtriple=amdgcn -mcpu=gfx1200 -mattr=+real-true16 -verify-machineinstrs < /b/1/llvm-x86_64-debian-dylib/llvm-project/llvm/test/CodeGen/AMDGPU/minimummaximum.ll | /b/1/llvm-x86_64-debian-dylib/build/bin/FileCheck -check-prefixes=GFX12,SDAG,SDAG-TRUE16 /b/1/llvm-x86_64-debian-dylib/llvm-project/llvm/test/CodeGen/AMDGPU/minimummaximum.ll
+ /b/1/llvm-x86_64-debian-dylib/build/bin/FileCheck -check-prefixes=GFX12,SDAG,SDAG-TRUE16 /b/1/llvm-x86_64-debian-dylib/llvm-project/llvm/test/CodeGen/AMDGPU/minimummaximum.ll
+ /b/1/llvm-x86_64-debian-dylib/build/bin/llc -mtriple=amdgcn -mcpu=gfx1200 -mattr=+real-true16 -verify-machineinstrs
RUN: at line 3: /b/1/llvm-x86_64-debian-dylib/build/bin/llc -mtriple=amdgcn -mcpu=gfx1200 -mattr=-real-true16 -verify-machineinstrs < /b/1/llvm-x86_64-debian-dylib/llvm-project/llvm/test/CodeGen/AMDGPU/minimummaximum.ll | /b/1/llvm-x86_64-debian-dylib/build/bin/FileCheck -check-prefixes=GFX12,SDAG,SDAG-FAKE16 /b/1/llvm-x86_64-debian-dylib/llvm-project/llvm/test/CodeGen/AMDGPU/minimummaximum.ll
+ /b/1/llvm-x86_64-debian-dylib/build/bin/FileCheck -check-prefixes=GFX12,SDAG,SDAG-FAKE16 /b/1/llvm-x86_64-debian-dylib/llvm-project/llvm/test/CodeGen/AMDGPU/minimummaximum.ll
+ /b/1/llvm-x86_64-debian-dylib/build/bin/llc -mtriple=amdgcn -mcpu=gfx1200 -mattr=-real-true16 -verify-machineinstrs
RUN: at line 4: /b/1/llvm-x86_64-debian-dylib/build/bin/llc -global-isel -mtriple=amdgcn -mcpu=gfx1200 -mattr=+real-true16 -verify-machineinstrs < /b/1/llvm-x86_64-debian-dylib/llvm-project/llvm/test/CodeGen/AMDGPU/minimummaximum.ll | /b/1/llvm-x86_64-debian-dylib/build/bin/FileCheck -check-prefixes=GFX12,GISEL,GISEL-TRUE16 /b/1/llvm-x86_64-debian-dylib/llvm-project/llvm/test/CodeGen/AMDGPU/minimummaximum.ll
+ /b/1/llvm-x86_64-debian-dylib/build/bin/FileCheck -check-prefixes=GFX12,GISEL,GISEL-TRUE16 /b/1/llvm-x86_64-debian-dylib/llvm-project/llvm/test/CodeGen/AMDGPU/minimummaximum.ll
+ /b/1/llvm-x86_64-debian-dylib/build/bin/llc -global-isel -mtriple=amdgcn -mcpu=gfx1200 -mattr=+real-true16 -verify-machineinstrs
/b/1/llvm-x86_64-debian-dylib/llvm-project/llvm/test/CodeGen/AMDGPU/minimummaximum.ll:187:15: error: GISEL-NEXT: is not on the line after the previous match
; GISEL-NEXT: s_mov_b32 s6, s3
              ^
<stdin>:495:2: note: 'next' match was here
 s_mov_b32 s6, s3
 ^
<stdin>:493:26: note: previous match ended here
 s_maximum_f16 s0, s0, s1
                         ^
<stdin>:494:1: note: non-matching line after previous match is here
 v_mov_b32_e32 v1, 0
^

Input file: <stdin>
Check file: /b/1/llvm-x86_64-debian-dylib/llvm-project/llvm/test/CodeGen/AMDGPU/minimummaximum.ll

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

Input was:
<<<<<<
          .
          .
          .
        490:  .type s_test_minmax_f16, at function 
        491: s_test_minmax_f16: ; @s_test_minmax_f16 
        492: ; %bb.0: 
        493:  s_maximum_f16 s0, s0, s1 
        494:  v_mov_b32_e32 v1, 0 
        495:  s_mov_b32 s6, s3 
next:187      !~~~~~~~~~~~~~~~  error: match on wrong line
        496:  s_mov_b32 s7, s4 
        497:  s_minimum_f16 s0, s0, s2 
        498:  s_delay_alu instid0(SALU_CYCLE_3) 
        499:  v_mov_b16_e32 v0.l, s0 
        500:  global_store_b16 v1, v0, s[6:7] 
          .
...

```

</details>

https://github.com/llvm/llvm-project/pull/128233