Skip to content

[AMDGPU] Add support for v_tanh_f16 on gfx1250#149439

Merged
shiltian merged 1 commit intomainfrom
users/shiltian/v_tanh_f16
Jul 18, 2025
Merged

[AMDGPU] Add support for v_tanh_f16 on gfx1250#149439
shiltian merged 1 commit intomainfrom
users/shiltian/v_tanh_f16

Conversation

@shiltian
Copy link
Copy Markdown
Contributor

@shiltian shiltian commented Jul 18, 2025

Co-authored-by: Mekhanoshin, Stanislav Stanislav.Mekhanoshin@amd.com

@shiltian @rampitec
[AMDGPU] Add support for v_tanh_f16 on gfx1250
9c6e6a6 
Co-authored-by: Mekhanoshin, Stanislav <Stanislav.Mekhanoshin@amd.com>
@shiltian shiltian requested review from changpeng and rampitec July 18, 2025 02:11
@llvmbot llvmbot added clang Clang issues not falling into any other category backend:AMDGPU clang:frontend Language frontend issues, e.g. anything involving "Sema" clang:codegen IR generation bugs: mangling, exceptions, etc. llvm:mc Machine (object) code labels Jul 18, 2025
@shiltian Graphite App
Copy link
Copy Markdown
Contributor Author

shiltian commented Jul 18, 2025

This stack of pull requests is managed by Graphite. Learn more about stacking.

@llvmbot
Copy link
Copy Markdown
Member

llvmbot commented Jul 18, 2025 •
edited
Loading

@llvm/pr-subscribers-mc
@llvm/pr-subscribers-backend-amdgpu

@llvm/pr-subscribers-clang

Author: Shilei Tian (shiltian)

Changes

Co-authored-by: Mekhanoshin, Stanislav <Stanislav.Mekhanoshin@amd.com>

Patch is 70.80 KiB, truncated to 20.00 KiB below, full version: https://github.com/llvm/llvm-project/pull/149439.diff

23 Files Affected:

  • (modified) clang/include/clang/Basic/BuiltinsAMDGPU.def (+1)
  • (modified) clang/lib/CodeGen/TargetBuiltins/AMDGPU.cpp (+1)
  • (modified) clang/test/CodeGenOpenCL/builtins-amdgcn-gfx1250.cl (+20)
  • (modified) llvm/lib/Target/AMDGPU/VOP1Instructions.td (+5)
  • (modified) llvm/test/CodeGen/AMDGPU/llvm.amdgcn.tanh.ll (+83)
  • (modified) llvm/test/MC/AMDGPU/gfx1250_asm_vop1-fake16.s (+45)
  • (modified) llvm/test/MC/AMDGPU/gfx1250_asm_vop1.s (+48)
  • (modified) llvm/test/MC/AMDGPU/gfx1250_asm_vop1_dpp16-fake16.s (+56)
  • (modified) llvm/test/MC/AMDGPU/gfx1250_asm_vop1_dpp16.s (+60)
  • (modified) llvm/test/MC/AMDGPU/gfx1250_asm_vop1_dpp8-fake16.s (+12)
  • (modified) llvm/test/MC/AMDGPU/gfx1250_asm_vop1_dpp8.s (+16)
  • (modified) llvm/test/MC/AMDGPU/gfx1250_asm_vop3_from_vop1-fake16.s (+45)
  • (modified) llvm/test/MC/AMDGPU/gfx1250_asm_vop3_from_vop1.s (+48)
  • (modified) llvm/test/MC/AMDGPU/gfx1250_asm_vop3_from_vop1_dpp16-fake16.s (+56)
  • (modified) llvm/test/MC/AMDGPU/gfx1250_asm_vop3_from_vop1_dpp16.s (+60)
  • (modified) llvm/test/MC/AMDGPU/gfx1250_asm_vop3_from_vop1_dpp8-fake16.s (+16)
  • (modified) llvm/test/MC/AMDGPU/gfx1250_asm_vop3_from_vop1_dpp8.s (+20)
  • (modified) llvm/test/MC/Disassembler/AMDGPU/gfx1250_dasm_vop1.txt (+63)
  • (modified) llvm/test/MC/Disassembler/AMDGPU/gfx1250_dasm_vop1_dpp16.txt (+59)
  • (modified) llvm/test/MC/Disassembler/AMDGPU/gfx1250_dasm_vop1_dpp8.txt (+16)
  • (modified) llvm/test/MC/Disassembler/AMDGPU/gfx1250_dasm_vop3_from_vop1.txt (+64)
  • (modified) llvm/test/MC/Disassembler/AMDGPU/gfx1250_dasm_vop3_from_vop1_dpp16.txt (+62-2)
  • (modified) llvm/test/MC/Disassembler/AMDGPU/gfx1250_dasm_vop3_from_vop1_dpp8.txt (+20)
diff --git a/clang/include/clang/Basic/BuiltinsAMDGPU.def b/clang/include/clang/Basic/BuiltinsAMDGPU.def
index 4111837d962b5..ed51f1d5de447 100644
--- a/clang/include/clang/Basic/BuiltinsAMDGPU.def
+++ b/clang/include/clang/Basic/BuiltinsAMDGPU.def
@@ -670,6 +670,7 @@ TARGET_BUILTIN(__builtin_amdgcn_s_wait_asynccnt, "vIUs", "n", "gfx1250-insts")
 TARGET_BUILTIN(__builtin_amdgcn_s_wait_tensorcnt, "vIUs", "n", "gfx1250-insts")
 
 TARGET_BUILTIN(__builtin_amdgcn_tanhf, "ff", "nc", "tanh-insts")
+TARGET_BUILTIN(__builtin_amdgcn_tanhh, "hh", "nc", "tanh-insts")
 TARGET_BUILTIN(__builtin_amdgcn_tanh_bf16, "yy", "nc", "bf16-trans-insts")
 TARGET_BUILTIN(__builtin_amdgcn_rcp_bf16, "yy", "nc", "bf16-trans-insts")
 TARGET_BUILTIN(__builtin_amdgcn_rsq_bf16, "yy", "nc", "bf16-trans-insts")
diff --git a/clang/lib/CodeGen/TargetBuiltins/AMDGPU.cpp b/clang/lib/CodeGen/TargetBuiltins/AMDGPU.cpp
index bcdb488f11639..a7d796ecccc61 100644
--- a/clang/lib/CodeGen/TargetBuiltins/AMDGPU.cpp
+++ b/clang/lib/CodeGen/TargetBuiltins/AMDGPU.cpp
@@ -504,6 +504,7 @@ Value *CodeGenFunction::EmitAMDGPUBuiltinExpr(unsigned BuiltinID,
     return Builder.CreateCall(F, { Src });
   }
   case AMDGPU::BI__builtin_amdgcn_tanhf:
+  case AMDGPU::BI__builtin_amdgcn_tanhh:
   case AMDGPU::BI__builtin_amdgcn_tanh_bf16:
     return emitBuiltinWithOneOverloadedType<1>(*this, E,
                                                Intrinsic::amdgcn_tanh);
diff --git a/clang/test/CodeGenOpenCL/builtins-amdgcn-gfx1250.cl b/clang/test/CodeGenOpenCL/builtins-amdgcn-gfx1250.cl
index e120a46c6327b..738b7ab7f2b75 100644
--- a/clang/test/CodeGenOpenCL/builtins-amdgcn-gfx1250.cl
+++ b/clang/test/CodeGenOpenCL/builtins-amdgcn-gfx1250.cl
@@ -61,6 +61,26 @@ void test_tanh_f32(global float* out, float a)
   *out = __builtin_amdgcn_tanhf(a);
 }
 
+// CHECK-LABEL: @test_tanh_f16(
+// CHECK-NEXT:  entry:
+// CHECK-NEXT:    [[OUT_ADDR:%.*]] = alloca ptr addrspace(1), align 8, addrspace(5)
+// CHECK-NEXT:    [[A_ADDR:%.*]] = alloca ptr addrspace(1), align 8, addrspace(5)
+// CHECK-NEXT:    [[OUT_ADDR_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[OUT_ADDR]] to ptr
+// CHECK-NEXT:    [[A_ADDR_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[A_ADDR]] to ptr
+// CHECK-NEXT:    store ptr addrspace(1) [[OUT:%.*]], ptr [[OUT_ADDR_ASCAST]], align 8
+// CHECK-NEXT:    store ptr addrspace(1) [[A:%.*]], ptr [[A_ADDR_ASCAST]], align 8
+// CHECK-NEXT:    [[TMP0:%.*]] = load ptr addrspace(1), ptr [[A_ADDR_ASCAST]], align 8
+// CHECK-NEXT:    [[TMP1:%.*]] = load half, ptr addrspace(1) [[TMP0]], align 2
+// CHECK-NEXT:    [[TMP2:%.*]] = call half @llvm.amdgcn.tanh.f16(half [[TMP1]])
+// CHECK-NEXT:    [[TMP3:%.*]] = load ptr addrspace(1), ptr [[OUT_ADDR_ASCAST]], align 8
+// CHECK-NEXT:    store half [[TMP2]], ptr addrspace(1) [[TMP3]], align 2
+// CHECK-NEXT:    ret void
+//
+void test_tanh_f16(global half* out, global half* a)
+{
+  *out = __builtin_amdgcn_tanhh(*a);
+}
+
 // CHECK-LABEL: @test_tanh_bf16(
 // CHECK-NEXT:  entry:
 // CHECK-NEXT:    [[OUT_ADDR:%.*]] = alloca ptr addrspace(1), align 8, addrspace(5)
diff --git a/llvm/lib/Target/AMDGPU/VOP1Instructions.td b/llvm/lib/Target/AMDGPU/VOP1Instructions.td
index 8c35fea8259f4..1bbbb610305e9 100644
--- a/llvm/lib/Target/AMDGPU/VOP1Instructions.td
+++ b/llvm/lib/Target/AMDGPU/VOP1Instructions.td
@@ -530,6 +530,10 @@ defm V_EXP_F16 : VOP1Inst_t16 <"v_exp_f16", VOP_F16_F16, AMDGPUexpf16>;
 defm V_SIN_F16 : VOP1Inst_t16 <"v_sin_f16", VOP_F16_F16, AMDGPUsin>;
 defm V_COS_F16 : VOP1Inst_t16 <"v_cos_f16", VOP_F16_F16, AMDGPUcos>;
 
+let SubtargetPredicate = HasTanhInsts in {
+defm V_TANH_F16  : VOP1Inst_t16 <"v_tanh_f16",  VOP_F16_F16, int_amdgcn_tanh>;
+}
+
 let SubtargetPredicate = HasBF16TransInsts in {
 defm V_TANH_BF16 : VOP1Inst_t16 <"v_tanh_bf16", VOP_BF16_BF16, int_amdgcn_tanh>;
 defm V_RCP_BF16  : VOP1Inst_t16 <"v_rcp_bf16",  VOP_BF16_BF16, AMDGPUrcp>;
@@ -1142,6 +1146,7 @@ defm V_CVT_F32_F16           : VOP1_Real_FULL_t16_and_fake16_gfx11_gfx12<0x00b>;
 defm V_MOV_B64 : VOP1_Real_FULL <GFX1250Gen, 0x1d>;
 
 defm V_TANH_F32              : VOP1_Real_FULL<GFX1250Gen, 0x01e>;
+defm V_TANH_F16              : VOP1_Real_FULL_t16_and_fake16_gfx1250<0x01f>;
 defm V_TANH_BF16             : VOP1_Real_FULL_t16_and_fake16_gfx1250<0x04a>;
 defm V_CVT_F32_BF16          : VOP1_Real_FULL_t16_and_fake16_gfx1250<0x072, "v_cvt_f32_bf16", "V_CVT_F32_BF16_gfx1250">;
 defm V_CVT_PK_F16_FP8        : VOP1_Real_FULL_t16_and_fake16_gfx1250<0x075>;
diff --git a/llvm/test/CodeGen/AMDGPU/llvm.amdgcn.tanh.ll b/llvm/test/CodeGen/AMDGPU/llvm.amdgcn.tanh.ll
index 81db7354757d9..dd89f80a54949 100644
--- a/llvm/test/CodeGen/AMDGPU/llvm.amdgcn.tanh.ll
+++ b/llvm/test/CodeGen/AMDGPU/llvm.amdgcn.tanh.ll
@@ -7,6 +7,7 @@
 ; FIXME: GlobalISel does not work with bf16
 
 declare float @llvm.amdgcn.tanh.f32(float) #0
+declare half @llvm.amdgcn.tanh.f16(half) #0
 declare bfloat @llvm.amdgcn.tanh.bf16(bfloat) #0
 
 define amdgpu_kernel void @tanh_f32(ptr addrspace(1) %out, float %src) #1 {
@@ -92,6 +93,88 @@ define amdgpu_kernel void @tanh_undef_f32(ptr addrspace(1) %out) #1 {
   ret void
 }
 
+define amdgpu_kernel void @tanh_f16(ptr addrspace(1) %out, half %src) #1 {
+; SDAG-REAL16-LABEL: tanh_f16:
+; SDAG-REAL16:       ; %bb.0:
+; SDAG-REAL16-NEXT:    s_load_b96 s[0:2], s[4:5], 0x0
+; SDAG-REAL16-NEXT:    v_mov_b32_e32 v1, 0
+; SDAG-REAL16-NEXT:    s_wait_kmcnt 0x0
+; SDAG-REAL16-NEXT:    v_tanh_f16_e32 v0.l, s2
+; SDAG-REAL16-NEXT:    flat_store_b16 v1, v0, s[0:1]
+; SDAG-REAL16-NEXT:    s_endpgm
+;
+; SDAG-FAKE16-LABEL: tanh_f16:
+; SDAG-FAKE16:       ; %bb.0:
+; SDAG-FAKE16-NEXT:    s_load_b96 s[0:2], s[4:5], 0x0
+; SDAG-FAKE16-NEXT:    v_mov_b32_e32 v1, 0
+; SDAG-FAKE16-NEXT:    s_wait_kmcnt 0x0
+; SDAG-FAKE16-NEXT:    v_tanh_f16_e32 v0, s2
+; SDAG-FAKE16-NEXT:    global_store_b16 v1, v0, s[0:1]
+; SDAG-FAKE16-NEXT:    s_endpgm
+  %tanh = call half @llvm.amdgcn.tanh.f16(half %src) #0
+  store half %tanh, ptr addrspace(1) %out, align 2
+  ret void
+}
+
+define amdgpu_kernel void @tanh_f16_constant_4.0(ptr addrspace(1) %out) #1 {
+; SDAG-REAL16-LABEL: tanh_f16_constant_4.0:
+; SDAG-REAL16:       ; %bb.0:
+; SDAG-REAL16-NEXT:    s_load_b64 s[0:1], s[4:5], 0x0
+; SDAG-REAL16-NEXT:    v_tanh_f16_e32 v0.l, 4.0
+; SDAG-REAL16-NEXT:    v_mov_b32_e32 v1, 0
+; SDAG-REAL16-NEXT:    s_wait_kmcnt 0x0
+; SDAG-REAL16-NEXT:    flat_store_b16 v1, v0, s[0:1]
+; SDAG-REAL16-NEXT:    s_endpgm
+;
+; SDAG-FAKE16-LABEL: tanh_f16_constant_4.0:
+; SDAG-FAKE16:       ; %bb.0:
+; SDAG-FAKE16-NEXT:    s_load_b64 s[0:1], s[4:5], 0x0
+; SDAG-FAKE16-NEXT:    v_tanh_f16_e32 v0, 4.0
+; SDAG-FAKE16-NEXT:    v_mov_b32_e32 v1, 0
+; SDAG-FAKE16-NEXT:    s_wait_kmcnt 0x0
+; SDAG-FAKE16-NEXT:    global_store_b16 v1, v0, s[0:1]
+; SDAG-FAKE16-NEXT:    s_endpgm
+  %tanh = call half @llvm.amdgcn.tanh.f16(half 4.0) #0
+  store half %tanh, ptr addrspace(1) %out, align 2
+  ret void
+}
+
+define amdgpu_kernel void @tanh_f16_constant_100.0(ptr addrspace(1) %out) #1 {
+; SDAG-REAL16-LABEL: tanh_f16_constant_100.0:
+; SDAG-REAL16:       ; %bb.0:
+; SDAG-REAL16-NEXT:    s_load_b64 s[0:1], s[4:5], 0x0
+; SDAG-REAL16-NEXT:    v_tanh_f16_e32 v0.l, 0x5640
+; SDAG-REAL16-NEXT:    v_mov_b32_e32 v1, 0
+; SDAG-REAL16-NEXT:    s_wait_kmcnt 0x0
+; SDAG-REAL16-NEXT:    flat_store_b16 v1, v0, s[0:1]
+; SDAG-REAL16-NEXT:    s_endpgm
+;
+; SDAG-FAKE16-LABEL: tanh_f16_constant_100.0:
+; SDAG-FAKE16:       ; %bb.0:
+; SDAG-FAKE16-NEXT:    s_load_b64 s[0:1], s[4:5], 0x0
+; SDAG-FAKE16-NEXT:    v_tanh_f16_e32 v0, 0x5640
+; SDAG-FAKE16-NEXT:    v_mov_b32_e32 v1, 0
+; SDAG-FAKE16-NEXT:    s_wait_kmcnt 0x0
+; SDAG-FAKE16-NEXT:    global_store_b16 v1, v0, s[0:1]
+; SDAG-FAKE16-NEXT:    s_endpgm
+  %tanh = call half @llvm.amdgcn.tanh.f16(half 100.0) #0
+  store half %tanh, ptr addrspace(1) %out, align 2
+  ret void
+}
+
+define amdgpu_kernel void @tanh_undef_f16(ptr addrspace(1) %out) #1 {
+; SDAG-REAL16-LABEL: tanh_undef_f16:
+; SDAG-REAL16:       ; %bb.0:
+; SDAG-REAL16-NEXT:    s_endpgm
+;
+; SDAG-FAKE16-LABEL: tanh_undef_f16:
+; SDAG-FAKE16:       ; %bb.0:
+; SDAG-FAKE16-NEXT:    s_endpgm
+  %tanh = call half @llvm.amdgcn.tanh.f16(half undef)
+  store half %tanh, ptr addrspace(1) %out, align 2
+  ret void
+}
+
 define amdgpu_kernel void @tanh_bf16(ptr addrspace(1) %out, bfloat %src) #1 {
 ; SDAG-REAL16-LABEL: tanh_bf16:
 ; SDAG-REAL16:       ; %bb.0:
diff --git a/llvm/test/MC/AMDGPU/gfx1250_asm_vop1-fake16.s b/llvm/test/MC/AMDGPU/gfx1250_asm_vop1-fake16.s
index f9e217d1f0361..279bb262bff04 100644
--- a/llvm/test/MC/AMDGPU/gfx1250_asm_vop1-fake16.s
+++ b/llvm/test/MC/AMDGPU/gfx1250_asm_vop1-fake16.s
@@ -73,6 +73,51 @@ v_tanh_f32 v5, src_scc
 v_tanh_f32 v255, 0xaf123456
 // GFX1250: v_tanh_f32_e32 v255, 0xaf123456         ; encoding: [0xff,0x3c,0xfe,0x7f,0x56,0x34,0x12,0xaf]
 
+v_tanh_f16 v5, v1
+// GFX1250: v_tanh_f16_e32 v5, v1                   ; encoding: [0x01,0x3f,0x0a,0x7e]
+
+v_tanh_f16 v5, v127
+// GFX1250: v_tanh_f16_e32 v5, v127                 ; encoding: [0x7f,0x3f,0x0a,0x7e]
+
+v_tanh_f16 v5, s1
+// GFX1250: v_tanh_f16_e32 v5, s1                   ; encoding: [0x01,0x3e,0x0a,0x7e]
+
+v_tanh_f16 v5, s105
+// GFX1250: v_tanh_f16_e32 v5, s105                 ; encoding: [0x69,0x3e,0x0a,0x7e]
+
+v_tanh_f16 v5, vcc_lo
+// GFX1250: v_tanh_f16_e32 v5, vcc_lo               ; encoding: [0x6a,0x3e,0x0a,0x7e]
+
+v_tanh_f16 v5, vcc_hi
+// GFX1250: v_tanh_f16_e32 v5, vcc_hi               ; encoding: [0x6b,0x3e,0x0a,0x7e]
+
+v_tanh_f16 v5, ttmp15
+// GFX1250: v_tanh_f16_e32 v5, ttmp15               ; encoding: [0x7b,0x3e,0x0a,0x7e]
+
+v_tanh_f16 v5, m0
+// GFX1250: v_tanh_f16_e32 v5, m0                   ; encoding: [0x7d,0x3e,0x0a,0x7e]
+
+v_tanh_f16 v5, exec_lo
+// GFX1250: v_tanh_f16_e32 v5, exec_lo              ; encoding: [0x7e,0x3e,0x0a,0x7e]
+
+v_tanh_f16 v5, exec_hi
+// GFX1250: v_tanh_f16_e32 v5, exec_hi              ; encoding: [0x7f,0x3e,0x0a,0x7e]
+
+v_tanh_f16 v5, null
+// GFX1250: v_tanh_f16_e32 v5, null                 ; encoding: [0x7c,0x3e,0x0a,0x7e]
+
+v_tanh_f16 v5, -1
+// GFX1250: v_tanh_f16_e32 v5, -1                   ; encoding: [0xc1,0x3e,0x0a,0x7e]
+
+v_tanh_f16 v5, 0.5
+// GFX1250: v_tanh_f16_e32 v5, 0.5                  ; encoding: [0xf0,0x3e,0x0a,0x7e]
+
+v_tanh_f16 v5, src_scc
+// GFX1250: v_tanh_f16_e32 v5, src_scc              ; encoding: [0xfd,0x3e,0x0a,0x7e]
+
+v_tanh_f16 v127, 0x8000
+// GFX1250: v_tanh_f16_e32 v127, 0x8000             ; encoding: [0xff,0x3e,0xfe,0x7e,0x00,0x80,0x00,0x00]
+
 v_tanh_bf16 v5, v1
 // GFX1250: v_tanh_bf16_e32 v5, v1                  ; encoding: [0x01,0x95,0x0a,0x7e]
 
diff --git a/llvm/test/MC/AMDGPU/gfx1250_asm_vop1.s b/llvm/test/MC/AMDGPU/gfx1250_asm_vop1.s
index d51ef68bf1e19..76272d25d92d4 100644
--- a/llvm/test/MC/AMDGPU/gfx1250_asm_vop1.s
+++ b/llvm/test/MC/AMDGPU/gfx1250_asm_vop1.s
@@ -73,6 +73,54 @@ v_tanh_f32 v5, src_scc
 v_tanh_f32 v255, 0xaf123456
 // GFX1250: v_tanh_f32_e32 v255, 0xaf123456         ; encoding: [0xff,0x3c,0xfe,0x7f,0x56,0x34,0x12,0xaf]
 
+v_tanh_f16 v5, v1
+// GFX1250: v_tanh_f16_e32 v5, v1                   ; encoding: [0x01,0x3f,0x0a,0x7e]
+
+v_tanh_f16 v5, v127
+// GFX1250: v_tanh_f16_e32 v5, v127                 ; encoding: [0x7f,0x3f,0x0a,0x7e]
+
+v_tanh_f16 v5, s1
+// GFX1250: v_tanh_f16_e32 v5, s1                   ; encoding: [0x01,0x3e,0x0a,0x7e]
+
+v_tanh_f16 v5, s105
+// GFX1250: v_tanh_f16_e32 v5, s105                 ; encoding: [0x69,0x3e,0x0a,0x7e]
+
+v_tanh_f16 v5, vcc_lo
+// GFX1250: v_tanh_f16_e32 v5, vcc_lo               ; encoding: [0x6a,0x3e,0x0a,0x7e]
+
+v_tanh_f16 v5, vcc_hi
+// GFX1250: v_tanh_f16_e32 v5, vcc_hi               ; encoding: [0x6b,0x3e,0x0a,0x7e]
+
+v_tanh_f16 v5, ttmp15
+// GFX1250: v_tanh_f16_e32 v5, ttmp15               ; encoding: [0x7b,0x3e,0x0a,0x7e]
+
+v_tanh_f16 v5, m0
+// GFX1250: v_tanh_f16_e32 v5, m0                   ; encoding: [0x7d,0x3e,0x0a,0x7e]
+
+v_tanh_f16 v5, exec_lo
+// GFX1250: v_tanh_f16_e32 v5, exec_lo              ; encoding: [0x7e,0x3e,0x0a,0x7e]
+
+v_tanh_f16 v5, exec_hi
+// GFX1250: v_tanh_f16_e32 v5, exec_hi              ; encoding: [0x7f,0x3e,0x0a,0x7e]
+
+v_tanh_f16 v5, null
+// GFX1250: v_tanh_f16_e32 v5, null                 ; encoding: [0x7c,0x3e,0x0a,0x7e]
+
+v_tanh_f16 v5, -1
+// GFX1250: v_tanh_f16_e32 v5, -1                   ; encoding: [0xc1,0x3e,0x0a,0x7e]
+
+v_tanh_f16 v5, 0.5
+// GFX1250: v_tanh_f16_e32 v5, 0.5                  ; encoding: [0xf0,0x3e,0x0a,0x7e]
+
+v_tanh_f16 v5, src_scc
+// GFX1250: v_tanh_f16_e32 v5, src_scc              ; encoding: [0xfd,0x3e,0x0a,0x7e]
+
+v_tanh_f16 v127, 0x8000
+// GFX1250: v_tanh_f16_e32 v127, 0x8000             ; encoding: [0xff,0x3e,0xfe,0x7e,0x00,0x80,0x00,0x00]
+
+v_tanh_f16 v5.h, v1.h
+// GFX1250: v_tanh_f16_e32 v5.h, v1.h               ; encoding: [0x81,0x3f,0x0a,0x7f]
+
 v_tanh_bf16 v5, v1
 // GFX1250: v_tanh_bf16_e32 v5, v1                  ; encoding: [0x01,0x95,0x0a,0x7e]
 
diff --git a/llvm/test/MC/AMDGPU/gfx1250_asm_vop1_dpp16-fake16.s b/llvm/test/MC/AMDGPU/gfx1250_asm_vop1_dpp16-fake16.s
index ae22f68e54835..0a8ee84561d33 100644
--- a/llvm/test/MC/AMDGPU/gfx1250_asm_vop1_dpp16-fake16.s
+++ b/llvm/test/MC/AMDGPU/gfx1250_asm_vop1_dpp16-fake16.s
@@ -58,6 +58,62 @@ v_tanh_f32 v255, -|v255| row_xmask:15 row_mask:0x3 bank_mask:0x0 bound_ctrl:0 fi
 // GFX1250: v_tanh_f32_dpp v255, -|v255| row_xmask:15 row_mask:0x3 bank_mask:0x0 fi:1 ; encoding: [0xfa,0x3c,0xfe,0x7f,0xff,0x6f,0x35,0x30]
 // GFX12-ERR: :[[@LINE-2]]:1: error: instruction not supported on this GPU
 
+v_tanh_f16 v5, v1 quad_perm:[3,2,1,0]
+// GFX1250: v_tanh_f16_dpp v5, v1 quad_perm:[3,2,1,0] row_mask:0xf bank_mask:0xf ; encoding: [0xfa,0x3e,0x0a,0x7e,0x01,0x1b,0x00,0xff]
+// GFX12-ERR: :[[@LINE-2]]:1: error: instruction not supported on this GPU
+
+v_tanh_f16 v5, v1 quad_perm:[0,1,2,3]
+// GFX1250: v_tanh_f16_dpp v5, v1 quad_perm:[0,1,2,3] row_mask:0xf bank_mask:0xf ; encoding: [0xfa,0x3e,0x0a,0x7e,0x01,0xe4,0x00,0xff]
+// GFX12-ERR: :[[@LINE-2]]:1: error: instruction not supported on this GPU
+
+v_tanh_f16 v5, v1 row_mirror
+// GFX1250: v_tanh_f16_dpp v5, v1 row_mirror row_mask:0xf bank_mask:0xf ; encoding: [0xfa,0x3e,0x0a,0x7e,0x01,0x40,0x01,0xff]
+// GFX12-ERR: :[[@LINE-2]]:1: error: instruction not supported on this GPU
+
+v_tanh_f16 v5, v1 row_half_mirror
+// GFX1250: v_tanh_f16_dpp v5, v1 row_half_mirror row_mask:0xf bank_mask:0xf ; encoding: [0xfa,0x3e,0x0a,0x7e,0x01,0x41,0x01,0xff]
+// GFX12-ERR: :[[@LINE-2]]:1: error: instruction not supported on this GPU
+
+v_tanh_f16 v5, v1 row_shl:1
+// GFX1250: v_tanh_f16_dpp v5, v1 row_shl:1 row_mask:0xf bank_mask:0xf ; encoding: [0xfa,0x3e,0x0a,0x7e,0x01,0x01,0x01,0xff]
+// GFX12-ERR: :[[@LINE-2]]:1: error: instruction not supported on this GPU
+
+v_tanh_f16 v5, v1 row_shl:15
+// GFX1250: v_tanh_f16_dpp v5, v1 row_shl:15 row_mask:0xf bank_mask:0xf ; encoding: [0xfa,0x3e,0x0a,0x7e,0x01,0x0f,0x01,0xff]
+// GFX12-ERR: :[[@LINE-2]]:1: error: instruction not supported on this GPU
+
+v_tanh_f16 v5, v1 row_shr:1
+// GFX1250: v_tanh_f16_dpp v5, v1 row_shr:1 row_mask:0xf bank_mask:0xf ; encoding: [0xfa,0x3e,0x0a,0x7e,0x01,0x11,0x01,0xff]
+// GFX12-ERR: :[[@LINE-2]]:1: error: instruction not supported on this GPU
+
+v_tanh_f16 v5, v1 row_shr:15
+// GFX1250: v_tanh_f16_dpp v5, v1 row_shr:15 row_mask:0xf bank_mask:0xf ; encoding: [0xfa,0x3e,0x0a,0x7e,0x01,0x1f,0x01,0xff]
+// GFX12-ERR: :[[@LINE-2]]:1: error: instruction not supported on this GPU
+
+v_tanh_f16 v5, v1 row_ror:1
+// GFX1250: v_tanh_f16_dpp v5, v1 row_ror:1 row_mask:0xf bank_mask:0xf ; encoding: [0xfa,0x3e,0x0a,0x7e,0x01,0x21,0x01,0xff]
+// GFX12-ERR: :[[@LINE-2]]:1: error: instruction not supported on this GPU
+
+v_tanh_f16 v5, v1 row_ror:15
+// GFX1250: v_tanh_f16_dpp v5, v1 row_ror:15 row_mask:0xf bank_mask:0xf ; encoding: [0xfa,0x3e,0x0a,0x7e,0x01,0x2f,0x01,0xff]
+// GFX12-ERR: :[[@LINE-2]]:1: error: instruction not supported on this GPU
+
+v_tanh_f16 v5, v1 row_share:0 row_mask:0xf bank_mask:0xf
+// GFX1250: v_tanh_f16_dpp v5, v1 row_share:0 row_mask:0xf bank_mask:0xf ; encoding: [0xfa,0x3e,0x0a,0x7e,0x01,0x50,0x01,0xff]
+// GFX12-ERR: :[[@LINE-2]]:1: error: instruction not supported on this GPU
+
+v_tanh_f16 v5, v1 row_share:15 row_mask:0x0 bank_mask:0x1
+// GFX1250: v_tanh_f16_dpp v5, v1 row_share:15 row_mask:0x0 bank_mask:0x1 ; encoding: [0xfa,0x3e,0x0a,0x7e,0x01,0x5f,0x01,0x01]
+// GFX12-ERR: :[[@LINE-2]]:1: error: instruction not supported on this GPU
+
+v_tanh_f16 v5, v1 row_xmask:0 row_mask:0x1 bank_mask:0x3 bound_ctrl:1 fi:0
+// GFX1250: v_tanh_f16_dpp v5, v1 row_xmask:0 row_mask:0x1 bank_mask:0x3 bound_ctrl:1 ; encoding: [0xfa,0x3e,0x0a,0x7e,0x01,0x60,0x09,0x13]
+// GFX12-ERR: :[[@LINE-2]]:1: error: instruction not supported on this GPU
+
+v_tanh_f16 v127, -|v127| row_xmask:15 row_mask:0x3 bank_mask:0x0 bound_ctrl:0 fi:1
+// GFX1250: v_tanh_f16_dpp v127, -|v127| row_xmask:15 row_mask:0x3 bank_mask:0x0 fi:1 ; encoding: [0xfa,0x3e,0xfe,0x7e,0x7f,0x6f,0x35,0x30]
+// GFX12-ERR: :[[@LINE-2]]:1: error: instruction not supported on this GPU
+
 v_tanh_bf16 v5, v1 quad_perm:[3,2,1,0]
 // GFX1250: v_tanh_bf16_dpp v5, v1 quad_perm:[3,2,1,0] row_mask:0xf bank_mask:0xf ; encoding: [0xfa,0x94,0x0a,0x7e,0x01,0x1b,0x00,0xff]
 // GFX12-ERR: :[[@LINE-2]]:1: error: instruction not supported on this GPU
diff --git a/llvm/test/MC/AMDGPU/gfx1250_asm_vop1_dpp16.s b/llvm/test/MC/AMDGPU/gfx1250_asm_vop1_dpp16.s
index 37ecb66bfe809..d4afb9d9b2d9a 100644
--- a/llvm/test/MC/AMDGPU/gfx1250_asm_vop1_dpp16.s
+++ b/llvm/test/MC/AMDGPU/gfx1250_asm_vop1_dpp16.s
@@ -58,6 +58,66 @@ v_tanh_f32 v255, -|v255| row_xmask:15 row_mask:0x3 bank_mask:0x0 bound_ctrl:0 fi
 // GFX1250: v_tanh_f32_dpp v255, -|v255| row_xmask:15 row_mask:0x3 bank_mask:0x0 fi:1 ; encoding: [0xfa,0x3c,0xfe,0x7f,0xff,0x6f,0x35,0x30]
 // GFX12-ERR: :[[@LINE-2]]:1: error: instruction not supported on this GPU
 
+v_tanh_f16 v5, v1 quad_perm:[3,2,1,0]
+// GFX1250: v_tanh_f16_dpp v5, v1 quad_perm:[3,2,1,0] row_mask:0xf bank_mask:0xf ; encoding: [0xfa,0x3e,0x0a,0x7e,0x01,0x1b,0x00,0xff]
+// GFX12-ERR: :[[@LINE-2]]:1: error: instruction not supported on this GPU
+
+v_tanh_f16 v5, v1 quad_perm:[0,1,2,3]
+// GFX1250: v_tanh_f16_dpp v5, v1 quad_perm:[0,1,2,3] row_mask:0xf bank_mask:0xf ; encoding: [0xfa,0x3e,0x0a,0x7e,0x01,0xe4,0x00,0xff]
+// GFX12-ERR: :[[@LINE-2]]:1: error: instruction not supported on this GPU
+
+v_tanh_f16 v5, v1 row_mirror
+// GFX1250: v_tanh_f16_dpp v5, v1 row_mirror row_mask:0xf bank_mask:0xf ; encoding: [0xfa,0x3e,0x0a,0x7e,0x01,0x40,0x01,0xff]
+// GFX12-ERR: :[[@LINE-2]]:1: error: instruction not supported on this GPU
+
+v_tanh_f16 v5, v1 row_half_mirror
+// GFX1250: v_tanh_f16_dpp v5, v1 row_half_mirror row_mask:0xf bank_mask:0xf ; encoding: [0xfa,0x3e,0x0a,0x7e,0x01,0x41,0x01,0xff]
+// GFX12-ERR: :[[@LINE-2]]:1: error: instruction not supported on this GPU
+
+v_tanh_f16 v5, v1 row_shl:1
+// GFX1250: v_tanh_f16_dpp v5, v1 row_shl:1 row_mask:0xf bank_mask:0xf ; encoding: [0xfa,0x3e,0x0a,0x7e,0x01,0x01,0x01,0xff]
+// GFX12-ERR: :[[@LINE-2]]:1: error: instruction not supported on this GPU
+
+v_tanh_f16 v5, v1 row_shl:15
+// GFX1250: v_tanh_f16_dpp v5, v1 row_shl:15 row_mask:0xf bank_mask:0xf ; encoding: [0xfa,0x3e,0x0a,0x7e,0x01,0x0f,0x01,0xff]
+// GFX12-ERR: :[[@LINE-2]]:1: error: instruction not supported on this GPU
+
+v_tanh_f16 v5, v1 row_shr:1
+// GFX1250: v_tanh_f16_dpp v5, v1 row_shr:1 row_mask:0xf bank_mask:0xf ; encoding: [0xfa,0x3e,0x0a,0x7e,0x01,0x11,0x01,0xff]
+// GFX12-ERR: :[[@LINE-2]]:1: error: instruction not supported on this GPU
+
+v_tanh_f16 v5, v1 row_shr:15
+// GFX1250: v_tanh_f16_dpp v5, v1 row_shr:15 row_mask:0xf bank_mask:0xf ; encoding: [0xfa,0x3e,0x0a,0x7e,0x01,0x1f,0x01,0xff]
+// GFX12-ERR: :[[@LINE-2]]:1: error: instruction not supported on this GPU
+
+v_tanh_f16 v5, v1 row_ror:1
+// GFX1250: v_tanh_f16_dpp v5, v1 row_ror:1 row_mask:0xf bank_mask:0xf ; encoding: [0xfa,0x3e,0x0a,0x7e,0x01,0x21,0x01,0xff]
+// GFX12-ERR: :[[@LINE-2]]:1: error: instruction not supported on this GPU
+
+v_tanh_f16 v5, v1 row_ror:15
+// GFX1250: v_tanh_f16_dpp v5, v1 row_ror:15 row_mask:0xf bank_mask:0xf ; encoding: [0xfa,0x3e,0x0a,0x7e,0x01,0x2f,0x01,0xff]
+// GFX12-ERR: :[[@LINE-2]]:1: error: instruction not supported on this GPU
+
+v_tanh_f16 v5, v1 row_share:0 row_mask:0xf bank_mask:0xf
+// GFX1250: v_tanh_f16_dpp v5, v1 row_share:0 row_mask:0xf bank_mask:0xf ; encoding: [0xfa,0x3e,0x0a,0x7e,0x01,0x50,0x01,0xff]
+// GFX12-ERR: :[[@LINE-2]]:1: error: instruction not suppor...
[truncated]

@github-actions
Copy link
Copy Markdown

github-actions bot commented Jul 18, 2025

⚠️ undef deprecator found issues in your code. ⚠️

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 clang/lib/CodeGen/TargetBuiltins/AMDGPU.cpp llvm/test/CodeGen/AMDGPU/llvm.amdgcn.tanh.ll

The following files introduce new uses of undef:

  • llvm/test/CodeGen/AMDGPU/llvm.amdgcn.tanh.ll

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 undef. You should use poison values for placeholders instead.

In tests, avoid using undef and having tests that trigger undefined behavior. If you need an operand with some unimportant value, you can add a new argument to the function and use that instead.

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.

@changpeng
Copy link
Copy Markdown
Contributor

changpeng commented Jul 18, 2025

Is the "Undef" in the test to be taken care of?

@shiltian
Copy link
Copy Markdown
Contributor Author

shiltian commented Jul 18, 2025

Is the "Undef" in the test to be taken care of?

I ignored all of them. Plan to do it after all upstreaming of VOP1 instructions to avoid creating any unnecessary conflicts during downstream merge.

arsenm
arsenm reviewed Jul 18, 2025
View reviewed changes
clang/lib/CodeGen/TargetBuiltins/AMDGPU.cpp
return Builder.CreateCall(F, { Src });
}
case AMDGPU::BI__builtin_amdgcn_tanhf:
case AMDGPU::BI__builtin_amdgcn_tanhh:
Copy link
Copy Markdown
Contributor

@arsenm arsenm Jul 18, 2025

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

can we just use llvm.tanh since it exists now?

Copy link
Copy Markdown
Contributor Author

@shiltian shiltian Jul 18, 2025

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I have a TODO list that will go over all newly added VOP1 instructions again and add missing tests and support accordingly.

@shiltian shiltian merged commit aecd448 into main Jul 18, 2025
14 of 15 checks passed
@shiltian shiltian deleted the users/shiltian/v_tanh_f16 branch July 18, 2025 04:21
This was referenced Jul 23, 2025
Closed
Closed
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Reviewers

@arsenm arsenm arsenm left review comments

@changpeng changpeng changpeng approved these changes

@rampitec rampitec Awaiting requested review from rampitec

Assignees

No one assigned

Labels

backend:AMDGPU clang:codegen IR generation bugs: mangling, exceptions, etc. clang:frontend Language frontend issues, e.g. anything involving "Sema" clang Clang issues not falling into any other category llvm:mc Machine (object) code

Projects

None yet

Milestone

No milestone

Development

Successfully merging this pull request may close these issues.

4 participants

@shiltian @llvmbot @changpeng @arsenm