Reapply "[AMDGPU] gfx1250 trans instructions bf16 codegen tests update. NFC (#155310)" #155515
+100
−0
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
This stack of pull requests is managed by Graphite. Learn more about stacking. |
|
@llvm/pr-subscribers-backend-amdgpu Author: Matt Arsenault (arsenm) ChangesThis reverts commit 665da0a. Regenerate checks after 4454152 landed Full diff: https://github.com/llvm/llvm-project/pull/155515.diff 3 Files Affected:
diff --git a/llvm/test/CodeGen/AMDGPU/llvm.amdgcn.rcp.bf16.ll b/llvm/test/CodeGen/AMDGPU/llvm.amdgcn.rcp.bf16.ll
index 199494d1c3473..1e44a09ddadf9 100644
--- a/llvm/test/CodeGen/AMDGPU/llvm.amdgcn.rcp.bf16.ll
+++ b/llvm/test/CodeGen/AMDGPU/llvm.amdgcn.rcp.bf16.ll
@@ -31,5 +31,71 @@ define amdgpu_kernel void @rcp_bf16(ptr addrspace(1) %out, bfloat %src) #1 {
ret void
}
+define amdgpu_kernel void @rcp_bf16_constant_4(ptr addrspace(1) %out) #1 {
+; SDAG-TRUE16-LABEL: rcp_bf16_constant_4:
+; SDAG-TRUE16: ; %bb.0:
+; SDAG-TRUE16-NEXT: s_load_b64 s[0:1], s[4:5], 0x0
+; SDAG-TRUE16-NEXT: v_mov_b16_e32 v0.l, 0x3e80
+; SDAG-TRUE16-NEXT: v_mov_b32_e32 v1, 0
+; SDAG-TRUE16-NEXT: s_wait_kmcnt 0x0
+; SDAG-TRUE16-NEXT: flat_store_b16 v1, v0, s[0:1]
+; SDAG-TRUE16-NEXT: s_endpgm
+;
+; SDAG-FAKE16-LABEL: rcp_bf16_constant_4:
+; SDAG-FAKE16: ; %bb.0:
+; SDAG-FAKE16-NEXT: s_load_b64 s[0:1], s[4:5], 0x0
+; SDAG-FAKE16-NEXT: v_dual_mov_b32 v0, 0 :: v_dual_mov_b32 v1, 0x3e80
+; SDAG-FAKE16-NEXT: s_wait_kmcnt 0x0
+; SDAG-FAKE16-NEXT: global_store_b16 v0, v1, s[0:1]
+; SDAG-FAKE16-NEXT: s_endpgm
+ %rcp = call bfloat @llvm.amdgcn.rcp.bf16(bfloat 4.0) #0
+ store bfloat %rcp, ptr addrspace(1) %out, align 2
+ ret void
+}
+
+define amdgpu_kernel void @rcp_bf16_constant_100(ptr addrspace(1) %out) #1 {
+; SDAG-TRUE16-LABEL: rcp_bf16_constant_100:
+; SDAG-TRUE16: ; %bb.0:
+; SDAG-TRUE16-NEXT: s_load_b64 s[0:1], s[4:5], 0x0
+; SDAG-TRUE16-NEXT: v_mov_b16_e32 v0.l, 0x3c24
+; SDAG-TRUE16-NEXT: v_mov_b32_e32 v1, 0
+; SDAG-TRUE16-NEXT: s_wait_kmcnt 0x0
+; SDAG-TRUE16-NEXT: flat_store_b16 v1, v0, s[0:1]
+; SDAG-TRUE16-NEXT: s_endpgm
+;
+; SDAG-FAKE16-LABEL: rcp_bf16_constant_100:
+; SDAG-FAKE16: ; %bb.0:
+; SDAG-FAKE16-NEXT: s_load_b64 s[0:1], s[4:5], 0x0
+; SDAG-FAKE16-NEXT: v_dual_mov_b32 v0, 0 :: v_dual_mov_b32 v1, 0x3c24
+; SDAG-FAKE16-NEXT: s_wait_kmcnt 0x0
+; SDAG-FAKE16-NEXT: global_store_b16 v0, v1, s[0:1]
+; SDAG-FAKE16-NEXT: s_endpgm
+ %rcp = call bfloat @llvm.amdgcn.rcp.bf16(bfloat 100.0) #0
+ store bfloat %rcp, ptr addrspace(1) %out, align 2
+ ret void
+}
+
+define amdgpu_kernel void @rcp_undef_bf16(ptr addrspace(1) %out) #1 {
+; SDAG-TRUE16-LABEL: rcp_undef_bf16:
+; SDAG-TRUE16: ; %bb.0:
+; SDAG-TRUE16-NEXT: s_load_b64 s[0:1], s[4:5], 0x0
+; SDAG-TRUE16-NEXT: v_mov_b16_e32 v0.l, 0x7fc0
+; SDAG-TRUE16-NEXT: v_mov_b32_e32 v1, 0
+; SDAG-TRUE16-NEXT: s_wait_kmcnt 0x0
+; SDAG-TRUE16-NEXT: flat_store_b16 v1, v0, s[0:1]
+; SDAG-TRUE16-NEXT: s_endpgm
+;
+; SDAG-FAKE16-LABEL: rcp_undef_bf16:
+; SDAG-FAKE16: ; %bb.0:
+; SDAG-FAKE16-NEXT: s_load_b64 s[0:1], s[4:5], 0x0
+; SDAG-FAKE16-NEXT: v_dual_mov_b32 v0, 0 :: v_dual_mov_b32 v1, 0x7fc0
+; SDAG-FAKE16-NEXT: s_wait_kmcnt 0x0
+; SDAG-FAKE16-NEXT: global_store_b16 v0, v1, s[0:1]
+; SDAG-FAKE16-NEXT: s_endpgm
+ %rcp = call bfloat @llvm.amdgcn.rcp.bf16(bfloat undef)
+ store bfloat %rcp, ptr addrspace(1) %out, align 2
+ ret void
+}
+
attributes #0 = { nounwind readnone }
attributes #1 = { nounwind }
diff --git a/llvm/test/CodeGen/AMDGPU/llvm.cos.bf16.ll b/llvm/test/CodeGen/AMDGPU/llvm.cos.bf16.ll
index 1015b753607ef..0948530abdf1b 100644
--- a/llvm/test/CodeGen/AMDGPU/llvm.cos.bf16.ll
+++ b/llvm/test/CodeGen/AMDGPU/llvm.cos.bf16.ll
@@ -6,6 +6,23 @@
declare bfloat @llvm.cos.bf16(bfloat) #0
+define amdgpu_kernel void @cos_bf16(ptr addrspace(1) %out, bfloat %src) #1 {
+; GCN-LABEL: cos_bf16:
+; GCN: ; %bb.0:
+; GCN-NEXT: s_load_b96 s[0:2], s[4:5], 0x0
+; GCN-NEXT: s_mov_b32 s3, 0x3e230000
+; GCN-NEXT: v_mov_b32_e32 v1, 0
+; GCN-NEXT: s_wait_kmcnt 0x0
+; GCN-NEXT: v_fma_mixlo_bf16 v0, s2, s3, 0 op_sel_hi:[1,0,0]
+; GCN-NEXT: s_delay_alu instid0(VALU_DEP_1)
+; GCN-NEXT: v_cos_bf16_e32 v0, v0
+; GCN-NEXT: global_store_b16 v1, v0, s[0:1]
+; GCN-NEXT: s_endpgm
+ %cos = call bfloat @llvm.cos.bf16(bfloat %src) #0
+ store bfloat %cos, ptr addrspace(1) %out, align 2
+ ret void
+}
+
define amdgpu_kernel void @cos_bf16_constant_4(ptr addrspace(1) %out) #1 {
; GCN-LABEL: cos_bf16_constant_4:
; GCN: ; %bb.0:
diff --git a/llvm/test/CodeGen/AMDGPU/llvm.sin.bf16.ll b/llvm/test/CodeGen/AMDGPU/llvm.sin.bf16.ll
index 701f54b468ce2..fe8ace52c06bf 100644
--- a/llvm/test/CodeGen/AMDGPU/llvm.sin.bf16.ll
+++ b/llvm/test/CodeGen/AMDGPU/llvm.sin.bf16.ll
@@ -6,6 +6,23 @@
declare bfloat @llvm.sin.bf16(bfloat) #0
+define amdgpu_kernel void @sin_bf16(ptr addrspace(1) %out, bfloat %src) #1 {
+; GCN-LABEL: sin_bf16:
+; GCN: ; %bb.0:
+; GCN-NEXT: s_load_b96 s[0:2], s[4:5], 0x0
+; GCN-NEXT: s_mov_b32 s3, 0x3e230000
+; GCN-NEXT: v_mov_b32_e32 v1, 0
+; GCN-NEXT: s_wait_kmcnt 0x0
+; GCN-NEXT: v_fma_mixlo_bf16 v0, s2, s3, 0 op_sel_hi:[1,0,0]
+; GCN-NEXT: s_delay_alu instid0(VALU_DEP_1)
+; GCN-NEXT: v_sin_bf16_e32 v0, v0
+; GCN-NEXT: global_store_b16 v1, v0, s[0:1]
+; GCN-NEXT: s_endpgm
+ %sin = call bfloat @llvm.sin.bf16(bfloat %src) #0
+ store bfloat %sin, ptr addrspace(1) %out, align 2
+ ret void
+}
+
define amdgpu_kernel void @sin_bf16_constant_4(ptr addrspace(1) %out) #1 {
; GCN-LABEL: sin_bf16_constant_4:
; GCN: ; %bb.0:
|
You can test this locally with the following command:git diff -U0 --pickaxe-regex -S '([^a-zA-Z0-9#_-]undef[^a-zA-Z0-9_-]|UndefValue::get)' 'HEAD~1' HEAD llvm/test/CodeGen/AMDGPU/llvm.amdgcn.rcp.bf16.ll llvm/test/CodeGen/AMDGPU/llvm.cos.bf16.ll llvm/test/CodeGen/AMDGPU/llvm.sin.bf16.llThe following files introduce new uses of undef:
Undef is now deprecated and should only be used in the rare cases where no replacement is possible. For example, a load of uninitialized memory yields In tests, avoid using For example, this is considered a bad practice: define void @fn() {
...
br i1 undef, ...
}Please use the following instead: define void @fn(i1 %cond) {
...
br i1 %cond, ...
}Please refer to the Undefined Behavior Manual for more information. |
shiltian
approved these changes
Aug 27, 2025
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 reverts commit 665da0a.
Regenerate checks after 4454152 landed
after adding these tests.