[AMDGPU][True16][CodeGen] update test for buildbot failure #131028
Merged
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
|
@llvm/pr-subscribers-backend-amdgpu Author: Brox Chen (broxigarchen) ChangesThis is a NFC patch #103366 hit a buildbot failure with i1-to-bf16.ll. Update the test to fix the build. Also remove duplicated comments added in #103366 Full diff: https://github.com/llvm/llvm-project/pull/131028.diff 2 Files Affected:
diff --git a/llvm/lib/Target/AMDGPU/GCNPreRAOptimizations.cpp b/llvm/lib/Target/AMDGPU/GCNPreRAOptimizations.cpp
index 1a00b5a846b2f..4deb2a9485e4d 100644
--- a/llvm/lib/Target/AMDGPU/GCNPreRAOptimizations.cpp
+++ b/llvm/lib/Target/AMDGPU/GCNPreRAOptimizations.cpp
@@ -25,9 +25,6 @@
/// This pass also adds register allocation hints to COPY.
/// The hints will be post-processed by SIRegisterInfo::getRegAllocationHints.
/// When using True16, we often see COPY moving a 16-bit value between a VGPR_32
-/// This pass also adds register allocation hints to COPY.
-/// The hints will be post-processed by SIRegisterInfo::getRegAllocationHints.
-/// When using True16, we often see COPY moving a 16-bit value between a VGPR_32
/// and a VGPR_16. If we use the VGPR_16 that corresponds to the lo16 bits of
/// the VGPR_32, the COPY can be completely eliminated.
///
diff --git a/llvm/test/CodeGen/AMDGPU/i1-to-bf16.ll b/llvm/test/CodeGen/AMDGPU/i1-to-bf16.ll
index a46d629c02b85..59b0ba2469a20 100644
--- a/llvm/test/CodeGen/AMDGPU/i1-to-bf16.ll
+++ b/llvm/test/CodeGen/AMDGPU/i1-to-bf16.ll
@@ -108,13 +108,14 @@ define amdgpu_ps i32 @s_uitofp_i1_to_bf16(i1 inreg %num) {
; GFX11-NEXT: v_cndmask_b32_e64 v0, 0, 1.0, s0
; GFX11-NEXT: v_readfirstlane_b32 s0, v0
; GFX11-NEXT: v_cmp_u_f32_e32 vcc_lo, v0, v0
+; GFX11-NEXT: s_delay_alu instid0(VALU_DEP_2) | instskip(NEXT) | instid1(SALU_CYCLE_1)
; GFX11-NEXT: s_bfe_u32 s1, s0, 0x10010
-; GFX11-NEXT: s_delay_alu instid0(SALU_CYCLE_1) | instskip(SKIP_4) | instid1(SALU_CYCLE_1)
; GFX11-NEXT: s_add_i32 s1, s1, s0
; GFX11-NEXT: s_bitset1_b32 s0, 22
; GFX11-NEXT: s_addk_i32 s1, 0x7fff
; GFX11-NEXT: s_and_b32 s2, vcc_lo, exec_lo
; GFX11-NEXT: s_cselect_b32 s0, s0, s1
+; GFX11-NEXT: s_delay_alu instid0(SALU_CYCLE_1)
; GFX11-NEXT: s_lshr_b32 s0, s0, 16
; GFX11-NEXT: ; return to shader part epilog
;
@@ -125,6 +126,7 @@ define amdgpu_ps i32 @s_uitofp_i1_to_bf16(i1 inreg %num) {
; GFX12-NEXT: s_delay_alu instid0(SALU_CYCLE_1) | instskip(NEXT) | instid1(VALU_DEP_1)
; GFX12-NEXT: v_cndmask_b32_e64 v0, 0, 1.0, s0
; GFX12-NEXT: v_readfirstlane_b32 s0, v0
+; GFX12-NEXT: s_delay_alu instid0(VALU_DEP_1)
; GFX12-NEXT: s_bfe_u32 s1, s0, 0x10010
; GFX12-NEXT: s_or_b32 s2, s0, 0x400000
; GFX12-NEXT: s_wait_alu 0xfffe
@@ -305,10 +307,11 @@ define amdgpu_ps <2 x i32> @s_uitofp_v2i1_to_v2bf16(<2 x i1> inreg %num) {
; GFX11-NEXT: v_cndmask_b32_e64 v1, 0, 1.0, s0
; GFX11-NEXT: v_readfirstlane_b32 s2, v0
; GFX11-NEXT: v_cmp_u_f32_e64 s1, v0, v0
-; GFX11-NEXT: s_delay_alu instid0(VALU_DEP_3) | instskip(SKIP_2) | instid1(SALU_CYCLE_1)
+; GFX11-NEXT: s_delay_alu instid0(VALU_DEP_3) | instskip(SKIP_1) | instid1(VALU_DEP_2)
; GFX11-NEXT: v_readfirstlane_b32 s0, v1
; GFX11-NEXT: v_cmp_u_f32_e32 vcc_lo, v1, v1
; GFX11-NEXT: s_bfe_u32 s3, s0, 0x10010
+; GFX11-NEXT: s_delay_alu instid0(SALU_CYCLE_1)
; GFX11-NEXT: s_add_i32 s3, s3, s0
; GFX11-NEXT: s_bitset1_b32 s0, 22
; GFX11-NEXT: s_addk_i32 s3, 0x7fff
@@ -338,6 +341,7 @@ define amdgpu_ps <2 x i32> @s_uitofp_v2i1_to_v2bf16(<2 x i1> inreg %num) {
; GFX12-NEXT: s_delay_alu instid0(VALU_DEP_2) | instskip(NEXT) | instid1(VALU_DEP_2)
; GFX12-NEXT: v_readfirstlane_b32 s2, v0
; GFX12-NEXT: v_readfirstlane_b32 s0, v1
+; GFX12-NEXT: s_delay_alu instid0(VALU_DEP_1)
; GFX12-NEXT: s_bfe_u32 s1, s0, 0x10010
; GFX12-NEXT: s_or_b32 s3, s0, 0x400000
; GFX12-NEXT: s_wait_alu 0xfffe
@@ -1161,13 +1165,14 @@ define amdgpu_ps i32 @s_sitofp_i1_to_bf16(i1 inreg %num) {
; GFX11-NEXT: v_cndmask_b32_e64 v0, 0, -1.0, s0
; GFX11-NEXT: v_readfirstlane_b32 s0, v0
; GFX11-NEXT: v_cmp_u_f32_e32 vcc_lo, v0, v0
+; GFX11-NEXT: s_delay_alu instid0(VALU_DEP_2) | instskip(NEXT) | instid1(SALU_CYCLE_1)
; GFX11-NEXT: s_bfe_u32 s1, s0, 0x10010
-; GFX11-NEXT: s_delay_alu instid0(SALU_CYCLE_1) | instskip(SKIP_4) | instid1(SALU_CYCLE_1)
; GFX11-NEXT: s_add_i32 s1, s1, s0
; GFX11-NEXT: s_bitset1_b32 s0, 22
; GFX11-NEXT: s_addk_i32 s1, 0x7fff
; GFX11-NEXT: s_and_b32 s2, vcc_lo, exec_lo
; GFX11-NEXT: s_cselect_b32 s0, s0, s1
+; GFX11-NEXT: s_delay_alu instid0(SALU_CYCLE_1)
; GFX11-NEXT: s_ashr_i32 s0, s0, 16
; GFX11-NEXT: ; return to shader part epilog
;
@@ -1178,6 +1183,7 @@ define amdgpu_ps i32 @s_sitofp_i1_to_bf16(i1 inreg %num) {
; GFX12-NEXT: s_delay_alu instid0(SALU_CYCLE_1) | instskip(NEXT) | instid1(VALU_DEP_1)
; GFX12-NEXT: v_cndmask_b32_e64 v0, 0, -1.0, s0
; GFX12-NEXT: v_readfirstlane_b32 s0, v0
+; GFX12-NEXT: s_delay_alu instid0(VALU_DEP_1)
; GFX12-NEXT: s_bfe_u32 s1, s0, 0x10010
; GFX12-NEXT: s_or_b32 s2, s0, 0x400000
; GFX12-NEXT: s_wait_alu 0xfffe
@@ -1358,10 +1364,11 @@ define amdgpu_ps <2 x i32> @s_sitofp_v2i1_to_v2bf16(<2 x i1> inreg %num) {
; GFX11-NEXT: v_cndmask_b32_e64 v1, 0, -1.0, s0
; GFX11-NEXT: v_readfirstlane_b32 s2, v0
; GFX11-NEXT: v_cmp_u_f32_e64 s0, v0, v0
-; GFX11-NEXT: s_delay_alu instid0(VALU_DEP_3) | instskip(SKIP_2) | instid1(SALU_CYCLE_1)
+; GFX11-NEXT: s_delay_alu instid0(VALU_DEP_3) | instskip(SKIP_1) | instid1(VALU_DEP_2)
; GFX11-NEXT: v_readfirstlane_b32 s1, v1
; GFX11-NEXT: v_cmp_u_f32_e32 vcc_lo, v1, v1
; GFX11-NEXT: s_bfe_u32 s3, s1, 0x10010
+; GFX11-NEXT: s_delay_alu instid0(SALU_CYCLE_1)
; GFX11-NEXT: s_add_i32 s3, s3, s1
; GFX11-NEXT: s_bitset1_b32 s1, 22
; GFX11-NEXT: s_addk_i32 s3, 0x7fff
@@ -1391,6 +1398,7 @@ define amdgpu_ps <2 x i32> @s_sitofp_v2i1_to_v2bf16(<2 x i1> inreg %num) {
; GFX12-NEXT: s_delay_alu instid0(VALU_DEP_2) | instskip(NEXT) | instid1(VALU_DEP_2)
; GFX12-NEXT: v_readfirstlane_b32 s2, v0
; GFX12-NEXT: v_readfirstlane_b32 s0, v1
+; GFX12-NEXT: s_delay_alu instid0(VALU_DEP_1)
; GFX12-NEXT: s_bfe_u32 s1, s0, 0x10010
; GFX12-NEXT: s_or_b32 s3, s0, 0x400000
; GFX12-NEXT: s_wait_alu 0xfffe
|
|
LLVM Buildbot has detected a new failure on builder Full details are available at: https://lab.llvm.org/buildbot/#/builders/73/builds/14480 Here is the relevant piece of the build log for the reference |
frederik-h
pushed a commit
to frederik-h/llvm-project
that referenced
this pull request
Mar 18, 2025
This is a NFC patch llvm#103366 hit a buildbot failure with i1-to-bf16.ll. Update the test to fix the build. Also remove duplicated comments added in llvm#103366
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
This is a NFC patch
#103366 hit a buildbot failure with i1-to-bf16.ll. Update the test to fix the build.
Also remove duplicated comments added in #103366