[AMDGPU][NFC] Add occupancy checks for gfx950 and gfx1250 #170392
Merged
+82 −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
Member
| @llvm/pr-subscribers-backend-amdgpu Author: Changpeng Fang (changpeng) ChangesFull diff: https://github.com/llvm/llvm-project/pull/170392.diff 1 Files Affected:
diff --git a/llvm/test/CodeGen/AMDGPU/occupancy-levels.ll b/llvm/test/CodeGen/AMDGPU/occupancy-levels.ll index d1ab92e1d48ff..7b6ed77e06e46 100644 --- a/llvm/test/CodeGen/AMDGPU/occupancy-levels.ll +++ b/llvm/test/CodeGen/AMDGPU/occupancy-levels.ll @@ -1,4 +1,5 @@ ; RUN: llc -mtriple=amdgcn -mcpu=gfx900 -mattr=-xnack < %s | FileCheck --check-prefixes=GCN,GFX9 %s +; RUN: llc -mtriple=amdgcn -mcpu=gfx950 -mattr=-xnack < %s | FileCheck --check-prefixes=GCN,GFX950 %s ; RUN: llc -mtriple=amdgcn -mcpu=gfx1010 -mattr=-xnack < %s | FileCheck --check-prefixes=GCN,GFX10,GFX10W32,GFX1010,GFX1010W32 %s ; RUN: llc -mtriple=amdgcn -mcpu=gfx1010 -mattr=-xnack -mattr=+wavefrontsize64 < %s | FileCheck --check-prefixes=GCN,GFX10,GFX10W64,GFX1010,GFX1010W64 %s ; RUN: llc -mtriple=amdgcn -mcpu=gfx1030 < %s | FileCheck --check-prefixes=GCN,GFX10,GFX10W32,GFX1030,GFX1030W32 %s @@ -19,49 +20,60 @@ ; RUN: llc -mtriple=amdgcn -mcpu=gfx1153 -mattr=+wavefrontsize64 < %s | FileCheck --check-prefixes=GCN,GFX1030,GFX1030W64 %s ; RUN: llc -mtriple=amdgcn -mcpu=gfx1200 < %s | FileCheck --check-prefixes=GCN,GFX1100,GFX1100W32 %s ; RUN: llc -mtriple=amdgcn -mcpu=gfx1200 -mattr=+wavefrontsize64 < %s | FileCheck --check-prefixes=GCN,GFX1100,GFX1100W64 %s +; RUN: llc -mtriple=amdgcn -mcpu=gfx1250 < %s | FileCheck --check-prefixes=GCN,GFX1250 %s ; GCN-LABEL: {{^}}max_occupancy: ; GFX9: ; Occupancy: 10 +; GFX950: ; Occupancy: 8 ; GFX1010: ; Occupancy: 20 ; GFX1030: ; Occupancy: 16 ; GFX1100: ; Occupancy: 16 +; GFX1250: ; Occupancy: 16 define amdgpu_kernel void @max_occupancy() #10 { ret void } ; GCN-LABEL: {{^}}limited_occupancy_3: ; GFX9: ; Occupancy: 3 +; GFX950: ; Occupancy: 3 ; GFX10W64: ; Occupancy: 3 ; GFX10W32: ; Occupancy: 4 ; GFX1100W64: ; Occupancy: 3 ; GFX1100W32: ; Occupancy: 5 +; GFX1250: ; Occupancy: 3 define amdgpu_kernel void @limited_occupancy_3() #0 { ret void } ; GCN-LABEL: {{^}}limited_occupancy_18: ; GFX9: ; Occupancy: 10 +; GFX950: ; Occupancy: 8 ; GFX1010: ; Occupancy: 18 ; GFX1030: ; Occupancy: 16 ; GFX1100: ; Occupancy: 16 +; GFX1250: ; Occupancy: 16 define amdgpu_kernel void @limited_occupancy_18() #1 { ret void } ; GCN-LABEL: {{^}}limited_occupancy_19: ; GFX9: ; Occupancy: 10 +; GFX950: ; Occupancy: 8 ; GFX1010: ; Occupancy: 20 ; GFX1030: ; Occupancy: 16 ; GFX1100: ; Occupancy: 16 +; GFX1250: ; Occupancy: 16 define amdgpu_kernel void @limited_occupancy_19() #2 { ret void } ; GCN-LABEL: {{^}}used_24_vgprs: ; GFX9: ; Occupancy: 10 +; GFX950: ; Occupancy: 8 ; GFX1010: ; Occupancy: 20 ; GFX1030: ; Occupancy: 16 ; GFX1100: ; Occupancy: 16 +; GFX1250: ; Occupancy: 16 define amdgpu_kernel void @used_24_vgprs() #10 { call void asm sideeffect "", "~{v23}" () ret void @@ -69,10 +81,12 @@ define amdgpu_kernel void @used_24_vgprs() #10 { ; GCN-LABEL: {{^}}used_28_vgprs: ; GFX9: ; Occupancy: 9 +; GFX950: ; Occupancy: 8 ; GFX1010W64: ; Occupancy: 18 ; GFX1010W32: ; Occupancy: 20 ; GFX1030: ; Occupancy: 16 ; GFX1100: ; Occupancy: 16 +; GFX1250: ; Occupancy: 16 define amdgpu_kernel void @used_28_vgprs() #10 { call void asm sideeffect "", "~{v27}" () ret void @@ -80,10 +94,12 @@ define amdgpu_kernel void @used_28_vgprs() #10 { ; GCN-LABEL: {{^}}used_32_vgprs: ; GFX9: ; Occupancy: 8 +; GFX950: ; Occupancy: 8 ; GFX10W64: ; Occupancy: 16 ; GFX1010W32: ; Occupancy: 20 ; GFX1030W32: ; Occupancy: 16 ; GFX1100: ; Occupancy: 16 +; GFX1250: ; Occupancy: 16 define amdgpu_kernel void @used_32_vgprs() #10 { call void asm sideeffect "", "~{v31}" () ret void @@ -91,11 +107,13 @@ define amdgpu_kernel void @used_32_vgprs() #10 { ; GCN-LABEL: {{^}}used_36_vgprs: ; GFX9: ; Occupancy: 7 +; GFX950: ; Occupancy: 8 ; GFX1010W64: ; Occupancy: 14 ; GFX1010W32: ; Occupancy: 20 ; GFX1030W64: ; Occupancy: 12 ; GFX1030W32: ; Occupancy: 16 ; GFX1100: ; Occupancy: 16 +; GFX1250: ; Occupancy: 16 define amdgpu_kernel void @used_36_vgprs() #10 { call void asm sideeffect "", "~{v35}" () ret void @@ -103,10 +121,12 @@ define amdgpu_kernel void @used_36_vgprs() #10 { ; GCN-LABEL: {{^}}used_40_vgprs: ; GFX9: ; Occupancy: 6 +; GFX950: ; Occupancy: 8 ; GFX10W64: ; Occupancy: 12 ; GFX1010W32: ; Occupancy: 20 ; GFX1030W32: ; Occupancy: 16 ; GFX1100: ; Occupancy: 16 +; GFX1250: ; Occupancy: 16 define amdgpu_kernel void @used_40_vgprs() #10 { call void asm sideeffect "", "~{v39}" () ret void @@ -114,11 +134,13 @@ define amdgpu_kernel void @used_40_vgprs() #10 { ; GCN-LABEL: {{^}}used_44_vgprs: ; GFX9: ; Occupancy: 5 +; GFX950: ; Occupancy: 8 ; GFX1010W64: ; Occupancy: 11 ; GFX1010W32: ; Occupancy: 20 ; GFX1030W64: ; Occupancy: 10 ; GFX1030W32: ; Occupancy: 16 ; GFX1100: ; Occupancy: 16 +; GFX1250: ; Occupancy: 16 define amdgpu_kernel void @used_44_vgprs() #10 { call void asm sideeffect "", "~{v43}" () ret void @@ -126,10 +148,12 @@ define amdgpu_kernel void @used_44_vgprs() #10 { ; GCN-LABEL: {{^}}used_48_vgprs: ; GFX9: ; Occupancy: 5 +; GFX950: ; Occupancy: 8 ; GFX10W64: ; Occupancy: 10 ; GFX1010W32: ; Occupancy: 20 ; GFX1030W32: ; Occupancy: 16 ; GFX1100: ; Occupancy: 16 +; GFX1250: ; Occupancy: 16 define amdgpu_kernel void @used_48_vgprs() #10 { call void asm sideeffect "", "~{v47}" () ret void @@ -137,11 +161,13 @@ define amdgpu_kernel void @used_48_vgprs() #10 { ; GCN-LABEL: {{^}}used_56_vgprs: ; GFX9: ; Occupancy: 4 +; GFX950: ; Occupancy: 8 ; GFX10W64: ; Occupancy: 9 ; GFX1010W32: ; Occupancy: 18 ; GFX1030W32: ; Occupancy: 16 ; GFX1100W64: ; Occupancy: 12 ; GFX1100W32: ; Occupancy: 16 +; GFX1250: ; Occupancy: 16 define amdgpu_kernel void @used_56_vgprs() #10 { call void asm sideeffect "", "~{v55}" () ret void @@ -149,10 +175,12 @@ define amdgpu_kernel void @used_56_vgprs() #10 { ; GCN-LABEL: {{^}}used_64_vgprs: ; GFX9: ; Occupancy: 4 +; GFX950: ; Occupancy: 8 ; GFX10W64: ; Occupancy: 8 ; GFX10W32: ; Occupancy: 16 ; GFX1100W64: ; Occupancy: 10 ; GFX1100W32: ; Occupancy: 16 +; GFX1250: ; Occupancy: 16 define amdgpu_kernel void @used_64_vgprs() #10 { call void asm sideeffect "", "~{v63}" () ret void @@ -160,11 +188,13 @@ define amdgpu_kernel void @used_64_vgprs() #10 { ; GCN-LABEL: {{^}}used_72_vgprs: ; GFX9: ; Occupancy: 3 +; GFX950: ; Occupancy: 7 ; GFX10W64: ; Occupancy: 7 ; GFX1010W32: ; Occupancy: 14 ; GFX1030W32: ; Occupancy: 12 ; GFX1100W64: ; Occupancy: 10 ; GFX1100W32: ; Occupancy: 16 +; GFX1250: ; Occupancy: 12 define amdgpu_kernel void @used_72_vgprs() #10 { call void asm sideeffect "", "~{v71}" () ret void @@ -172,10 +202,12 @@ define amdgpu_kernel void @used_72_vgprs() #10 { ; GCN-LABEL: {{^}}used_80_vgprs: ; GFX9: ; Occupancy: 3 +; GFX950: ; Occupancy: 6 ; GFX10W64: ; Occupancy: 6 ; GFX10W32: ; Occupancy: 12 ; GFX1100W64: ; Occupancy: 9 ; GFX1100W32: ; Occupancy: 16 +; GFX1250: ; Occupancy: 12 define amdgpu_kernel void @used_80_vgprs() #10 { call void asm sideeffect "", "~{v79}" () ret void @@ -183,12 +215,14 @@ define amdgpu_kernel void @used_80_vgprs() #10 { ; GCN-LABEL: {{^}}used_84_vgprs: ; GFX9: ; Occupancy: 3 +; GFX950: ; Occupancy: 5 ; GFX1010W64: ; Occupancy: 6 ; GFX1010W32: ; Occupancy: 11 ; GFX1030W64: ; Occupancy: 5 ; GFX1030W32: ; Occupancy: 10 ; GFX1100W64: ; Occupancy: 9 ; GFX1100W32: ; Occupancy: 16 +; GFX1250: ; Occupancy: 10 define amdgpu_kernel void @used_84_vgprs() #10 { call void asm sideeffect "", "~{v83}" () ret void @@ -196,11 +230,13 @@ define amdgpu_kernel void @used_84_vgprs() #10 { ; GCN-LABEL: {{^}}used_88_vgprs: ; GFX9: ; Occupancy: 2 +; GFX950: ; Occupancy: 5 ; GFX10W64: ; Occupancy: 5 ; GFX1010W32: ; Occupancy: 11 ; GFX1030W32: ; Occupancy: 10 ; GFX1100W64: ; Occupancy: 8 ; GFX1100W32: ; Occupancy: 16 +; GFX1250: ; Occupancy: 10 define amdgpu_kernel void @used_88_vgprs() #10 { call void asm sideeffect "", "~{v87}" () ret void @@ -208,10 +244,12 @@ define amdgpu_kernel void @used_88_vgprs() #10 { ; GCN-LABEL: {{^}}used_96_vgprs: ; GFX9: ; Occupancy: 2 +; GFX950: ; Occupancy: 5 ; GFX10W64: ; Occupancy: 5 ; GFX10W32: ; Occupancy: 10 ; GFX1100W64: ; Occupancy: 8 ; GFX1100W32: ; Occupancy: 16 +; GFX1250: ; Occupancy: 10 define amdgpu_kernel void @used_96_vgprs() #10 { call void asm sideeffect "", "~{v95}" () ret void @@ -219,11 +257,13 @@ define amdgpu_kernel void @used_96_vgprs() #10 { ; GCN-LABEL: {{^}}used_100_vgprs: ; GFX9: ; Occupancy: 2 +; GFX950: ; Occupancy: 4 ; GFX1010W64: ; Occupancy: 5 ; GFX1030W64: ; Occupancy: 4 ; GFX10W32: ; Occupancy: 9 ; GFX1100W64: ; Occupancy: 7 ; GFX1100W32: ; Occupancy: 12 +; GFX1250: ; Occupancy: 9 define amdgpu_kernel void @used_100_vgprs() #10 { call void asm sideeffect "", "~{v99}" () ret void @@ -231,10 +271,12 @@ define amdgpu_kernel void @used_100_vgprs() #10 { ; GCN-LABEL: {{^}}used_112_vgprs: ; GFX9: ; Occupancy: 2 +; GFX950: ; Occupancy: 4 ; GFX10W64: ; Occupancy: 4 ; GFX10W32: ; Occupancy: 9 ; GFX1100W64: ; Occupancy: 6 ; GFX1100W32: ; Occupancy: 12 +; GFX1250: ; Occupancy: 9 define amdgpu_kernel void @used_112_vgprs() #10 { call void asm sideeffect "", "~{v111}" () ret void @@ -242,10 +284,12 @@ define amdgpu_kernel void @used_112_vgprs() #10 { ; GCN-LABEL: {{^}}used_128_vgprs: ; GFX9: ; Occupancy: 2 +; GFX950: ; Occupancy: 4 ; GFX10W64: ; Occupancy: 4 ; GFX10W32: ; Occupancy: 8 ; GFX1100W64: ; Occupancy: 5 ; GFX1100W32: ; Occupancy: 10 +; GFX1250: ; Occupancy: 8 define amdgpu_kernel void @used_128_vgprs() #10 { call void asm sideeffect "", "~{v127}" () ret void @@ -253,10 +297,12 @@ define amdgpu_kernel void @used_128_vgprs() #10 { ; GCN-LABEL: {{^}}used_144_vgprs: ; GFX9: ; Occupancy: 1 +; GFX950: ; Occupancy: 3 ; GFX10W64: ; Occupancy: 3 ; GFX10W32: ; Occupancy: 7 ; GFX1100W64: ; Occupancy: 5 ; GFX1100W32: ; Occupancy: 10 +; GFX1250: ; Occupancy: 7 define amdgpu_kernel void @used_144_vgprs() #10 { call void asm sideeffect "", "~{v143}" () ret void @@ -264,11 +310,13 @@ define amdgpu_kernel void @used_144_vgprs() #10 { ; GCN-LABEL: {{^}}used_168_vgprs: ; GFX9: ; Occupancy: 1 +; GFX950: ; Occupancy: 3 ; GFX10W64: ; Occupancy: 3 ; GFX1010W32: ; Occupancy: 6 ; GFX1030W32: ; Occupancy: 5 ; GFX1100W64: ; Occupancy: 4 ; GFX1100W32: ; Occupancy: 9 +; GFX1250: ; Occupancy: 5 define amdgpu_kernel void @used_168_vgprs() #10 { call void asm sideeffect "", "~{v167}" () ret void @@ -276,11 +324,13 @@ define amdgpu_kernel void @used_168_vgprs() #10 { ; GCN-LABEL: {{^}}used_200_vgprs: ; GFX9: ; Occupancy: 1 +; GFX950: ; Occupancy: 2 ; GFX10W64: ; Occupancy: 2 ; GFX1010W32: ; Occupancy: 5 ; GFX1030W32: ; Occupancy: 4 ; GFX1100W64: ; Occupancy: 3 ; GFX1100W32: ; Occupancy: 7 +; GFX1250: ; Occupancy: 4 define amdgpu_kernel void @used_200_vgprs() #10 { call void asm sideeffect "", "~{v199}" () ret void @@ -288,10 +338,12 @@ define amdgpu_kernel void @used_200_vgprs() #10 { ; GCN-LABEL: {{^}}used_256_vgprs: ; GFX9: ; Occupancy: 1 +; GFX950: ; Occupancy: 2 ; GFX10W64: ; Occupancy: 2 ; GFX10W32: ; Occupancy: 4 ; GFX1100W64: ; Occupancy: 2 ; GFX1100W32: ; Occupancy: 5 +; GFX1250: ; Occupancy: 4 define amdgpu_kernel void @used_256_vgprs() #10 { call void asm sideeffect "", "~{v255}" () ret void @@ -299,9 +351,11 @@ define amdgpu_kernel void @used_256_vgprs() #10 { ; GCN-LABEL: {{^}}used_80_sgprs: ; GFX9: ; Occupancy: 10 +; GFX950: ; Occupancy: 8 ; GFX1010: ; Occupancy: 20 ; GFX1030: ; Occupancy: 16 ; GFX1100: ; Occupancy: 16 +; GFX1250: ; Occupancy: 16 define amdgpu_kernel void @used_80_sgprs() #10 { call void asm sideeffect "", "~{s79}" () ret void @@ -309,9 +363,11 @@ define amdgpu_kernel void @used_80_sgprs() #10 { ; GCN-LABEL: {{^}}used_88_sgprs: ; GFX9: ; Occupancy: 9 +; GFX950: ; Occupancy: 8 ; GFX1010: ; Occupancy: 20 ; GFX1030: ; Occupancy: 16 ; GFX1100: ; Occupancy: 16 +; GFX1250: ; Occupancy: 16 define amdgpu_kernel void @used_88_sgprs() #10 { call void asm sideeffect "", "~{s87}" () ret void @@ -319,9 +375,11 @@ define amdgpu_kernel void @used_88_sgprs() #10 { ; GCN-LABEL: {{^}}used_100_sgprs: ; GFX9: ; Occupancy: 8 +; GFX950: ; Occupancy: 7 ; GFX1010: ; Occupancy: 20 ; GFX1030: ; Occupancy: 16 ; GFX1100: ; Occupancy: 16 +; GFX1250: ; Occupancy: 16 define amdgpu_kernel void @used_100_sgprs() #10 { call void asm sideeffect "", "~{s99}" () ret void @@ -329,9 +387,11 @@ define amdgpu_kernel void @used_100_sgprs() #10 { ; GCN-LABEL: {{^}}used_101_sgprs: ; GFX9: ; Occupancy: 7 +; GFX950: ; Occupancy: 7 ; GFX1010: ; Occupancy: 20 ; GFX1030: ; Occupancy: 16 ; GFX1100: ; Occupancy: 16 +; GFX1250: ; Occupancy: 16 define amdgpu_kernel void @used_101_sgprs() #10 { call void asm sideeffect "", "~{s100}" () ret void @@ -339,10 +399,12 @@ define amdgpu_kernel void @used_101_sgprs() #10 { ; GCN-LABEL: {{^}}used_lds_6552: ; GFX9: ; Occupancy: 8 +; GFX950: ; Occupancy: 8 ; GFX1010W64: ; Occupancy: 20 ; GFX1030W64: ; Occupancy: 16 ; GFX10W32: ; Occupancy: 16 ; GFX1100: ; Occupancy: 16 +; GFX1250: ; Occupancy: 16 @lds6552 = internal addrspace(3) global [6552 x i8] poison, align 4 define amdgpu_kernel void @used_lds_6552() { store volatile i8 1, ptr addrspace(3) @lds6552 @@ -351,10 +413,12 @@ define amdgpu_kernel void @used_lds_6552() { ; GCN-LABEL: {{^}}used_lds_6556: ; GFX9: ; Occupancy: 8 +; GFX950: ; Occupancy: 8 ; GFX1010W64: ; Occupancy: 20 ; GFX1030W64: ; Occupancy: 16 ; GFX10W32: ; Occupancy: 16 ; GFX1100: ; Occupancy: 16 +; GFX1250: ; Occupancy: 16 @lds6556 = internal addrspace(3) global [6556 x i8] poison, align 4 define amdgpu_kernel void @used_lds_6556() { store volatile i8 1, ptr addrspace(3) @lds6556 @@ -363,10 +427,12 @@ define amdgpu_kernel void @used_lds_6556() { ; GCN-LABEL: {{^}}used_lds_13112: ; GFX9: ; Occupancy: 8 +; GFX950: ; Occupancy: 8 ; GFX1010W64: ; Occupancy: 20 ; GFX1030W64: ; Occupancy: 16 ; GFX10W32: ; Occupancy: 16 ; GFX1100: ; Occupancy: 16 +; GFX1250: ; Occupancy: 16 @lds13112 = internal addrspace(3) global [13112 x i8] poison, align 4 define amdgpu_kernel void @used_lds_13112() { store volatile i8 1, ptr addrspace(3) @lds13112 @@ -375,10 +441,12 @@ define amdgpu_kernel void @used_lds_13112() { ; GCN-LABEL: {{^}}used_lds_8252_max_group_size_64: ; GFX9: ; Occupancy: 2{{$}} +; GFX950: ; Occupancy: 5{{$}} ; GFX10W64: ; Occupancy: 4{{$}} ; GFX10W32: ; Occupancy: 8{{$}} ; GFX1100W64: ; Occupancy: 4{{$}} ; GFX1100W32: ; Occupancy: 8{{$}} +; GFX1250: ; Occupancy: 10 @lds8252 = internal addrspace(3) global [8252 x i8] poison, align 4 define amdgpu_kernel void @used_lds_8252_max_group_size_64() #3 { store volatile i8 1, ptr addrspace(3) @lds8252 @@ -387,10 +455,12 @@ define amdgpu_kernel void @used_lds_8252_max_group_size_64() #3 { ; GCN-LABEL: {{^}}used_lds_8252_max_group_size_96: ; GFX9: ; Occupancy: 4{{$}} +; GFX950: ; Occupancy: 8{{$}} ; GFX10W64: ; Occupancy: 8{{$}} ; GFX10W32: ; Occupancy: 12{{$}} ; GFX1100W64: ; Occupancy: 8{{$}} ; GFX1100W32: ; Occupancy: 12{{$}} +; GFX1250: ; Occupancy: 12 define amdgpu_kernel void @used_lds_8252_max_group_size_96() #4 { store volatile i8 1, ptr addrspace(3) @lds8252 ret void @@ -398,10 +468,12 @@ define amdgpu_kernel void @used_lds_8252_max_group_size_96() #4 { ; GCN-LABEL: {{^}}used_lds_8252_max_group_size_128: ; GFX9: ; Occupancy: 4{{$}} +; GFX950: ; Occupancy: 8{{$}} ; GFX10W64: ; Occupancy: 8{{$}} ; GFX10W32: ; Occupancy: 15{{$}} ; GFX1100W64: ; Occupancy: 8{{$}} ; GFX1100W32: ; Occupancy: 15{{$}} +; GFX1250: ; Occupancy: 16 define amdgpu_kernel void @used_lds_8252_max_group_size_128() #5 { store volatile i8 1, ptr addrspace(3) @lds8252 ret void @@ -409,11 +481,13 @@ define amdgpu_kernel void @used_lds_8252_max_group_size_128() #5 { ; GCN-LABEL: {{^}}used_lds_8252_max_group_size_192: ; GFX9: ; Occupancy: 6{{$}} +; GFX950: ; Occupancy: 8{{$}} ; GFX10W64: ; Occupancy: 12{{$}} ; GFX1010W32: ; Occupancy: 20{{$}} ; GFX1030W32: ; Occupancy: 15{{$}} ; GFX1100W64: ; Occupancy: 12{{$}} ; GFX1100W32: ; Occupancy: 15{{$}} +; GFX1250: ; Occupancy: 15 define amdgpu_kernel void @used_lds_8252_max_group_size_192() #6 { store volatile i8 1, ptr addrspace(3) @lds8252 ret void @@ -421,11 +495,13 @@ define amdgpu_kernel void @used_lds_8252_max_group_size_192() #6 { ; GCN-LABEL: {{^}}used_lds_8252_max_group_size_256: ; GFX9: ; Occupancy: 7{{$}} +; GFX950: ; Occupancy: 8{{$}} ; GFX10W64: ; Occupancy: 15{{$}} ; GFX1010W32: ; Occupancy: 20{{$}} ; GFX1030W32: ; Occupancy: 16{{$}} ; GFX1100W64: ; Occupancy: 15{{$}} ; GFX1100W32: ; Occupancy: 16{{$}} +; GFX1250: ; Occupancy: 16 define amdgpu_kernel void @used_lds_8252_max_group_size_256() #7 { store volatile i8 1, ptr addrspace(3) @lds8252 ret void @@ -433,9 +509,11 @@ define amdgpu_kernel void @used_lds_8252_max_group_size_256() #7 { ; GCN-LABEL: {{^}}used_lds_8252_max_group_size_512: ; GFX9: ; Occupancy: 10{{$}} +; GFX950: ; Occupancy: 8{{$}} ; GFX1010: ; Occupancy: 20{{$}} ; GFX1030: ; Occupancy: 16{{$}} ; GFX1100: ; Occupancy: 16{{$}} +; GFX1250: ; Occupancy: 16 define amdgpu_kernel void @used_lds_8252_max_group_size_512() #8 { store volatile i8 1, ptr addrspace(3) @lds8252 ret void @@ -443,10 +521,12 @@ define amdgpu_kernel void @used_lds_8252_max_group_size_512() #8 { ; GCN-LABEL: {{^}}used_lds_8252_max_group_size_1024: ; GFX9: ; Occupancy: 8{{$}} +; GFX950: ; Occupancy: 8{{$}} ; GFX1010W32: ; Occupancy: 16{{$}} ; GFX1010W64: ; Occupancy: 20{{$}} ; GFX1030: ; Occupancy: 16{{$}} ; GFX1100: ; Occupancy: 16{{$}} +; GFX1250: ; Occupancy: 16 define amdgpu_kernel void @used_lds_8252_max_group_size_1024() #9 { store volatile i8 1, ptr addrspace(3) @lds8252 ret void @@ -454,8 +534,10 @@ define amdgpu_kernel void @used_lds_8252_max_group_size_1024() #9 { ; GCN-LABEL: {{^}}used_lds_8252_max_group_size_32: ; GFX9: ; Occupancy: 2{{$}} +; GFX950: ; Occupancy: 5{{$}} ; GFX10: ; Occupancy: 4{{$}} ; GFX1100: ; Occupancy: 4{{$}} +; GFX1250: ; Occupancy: 10 define amdgpu_kernel void @used_lds_8252_max_group_size_32() #10 { store volatile i8 1, ptr addrspace(3) @lds8252 ret void |
rampitec approved these changes Dec 2, 2025
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
3 participants
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.
No description provided.