[clang] [llvm] [RFC][WIP][AMDGPU] Use `bf16` instead of `i16` for bfloat (PR #80908)

https://github.com/shiltian updated https://github.com/llvm/llvm-project/pull/80908 >From d14668fdfeef603624af520d11f5b66aa19da7be Mon Sep 17 00:00:00 2001 From: Shilei Tian Date: Thu, 8 Feb 2024 12:12:48 -0500 Subject: [PATCH] [RFC][WIP][AMDGPU] Use `bf16` instead of `i16` for bfloat Currentl

[clang] [llvm] [RFC][WIP][AMDGPU] Use `bf16` instead of `i16` for bfloat (PR #80908)

https://github.com/shiltian updated https://github.com/llvm/llvm-project/pull/80908 >From 672fd3cf584480eb4769ccdb5f86acbc03865ec9 Mon Sep 17 00:00:00 2001 From: Shilei Tian Date: Thu, 8 Feb 2024 11:17:25 -0500 Subject: [PATCH] [RFC][WIP][AMDGPU] Use `bf16` instead of `i16` for bfloat Currentl

[clang] [llvm] [RFC][WIP][AMDGPU] Use `bf16` instead of `i16` for bfloat (PR #80908)

@@ -2819,11 +2819,11 @@ def int_amdgcn_fdot2_f16_f16 : def int_amdgcn_fdot2_bf16_bf16 : ClangBuiltin<"__builtin_amdgcn_fdot2_bf16_bf16">, DefaultAttrsIntrinsic< -[llvm_i16_ty], // %r +[llvm_bfloat_ty], // %r arsenm wrote: Changing the clang bui

[clang] [llvm] [RFC][WIP][AMDGPU] Use `bf16` instead of `i16` for bfloat (PR #80908)

@@ -2835,8 +2835,8 @@ def int_amdgcn_fdot2_f32_bf16 : DefaultAttrsIntrinsic< [llvm_float_ty], // %r [ - llvm_v2i16_ty, // %a - llvm_v2i16_ty, // %b + llvm_v2bf16_ty, // %a + llvm_v2bf16_ty, // %b arsenm wrote: For potential revert

[clang] [llvm] [RFC][WIP][AMDGPU] Use `bf16` instead of `i16` for bfloat (PR #80908)

@@ -1562,8 +1562,9 @@ bool IRTranslator::translateBitCast(const User &U, bool IRTranslator::translateCast(unsigned Opcode, const User &U, MachineIRBuilder &MIRBuilder) { - if (U.getType()->getScalarType()->isBFloatTy() || - U.getOperand(0

[clang] [llvm] [RFC][WIP][AMDGPU] Use `bf16` instead of `i16` for bfloat (PR #80908)

@@ -0,0 +1,8 @@ +// RUN: llvm-mc -arch=amdgcn -mcpu=gfx1100 -show-encoding %s | FileCheck %s +// RUN: llvm-mc -arch=amdgcn -mcpu=gfx1200 -show-encoding %s | FileCheck %s + +v_dot2_bf16_bf16 v5, v1, v2, 100.0 arsenm wrote: does this help with #79369 at all? http

[clang] [llvm] [RFC][WIP][AMDGPU] Use `bf16` instead of `i16` for bfloat (PR #80908)

@@ -1562,8 +1562,9 @@ bool IRTranslator::translateBitCast(const User &U, bool IRTranslator::translateCast(unsigned Opcode, const User &U, MachineIRBuilder &MIRBuilder) { - if (U.getType()->getScalarType()->isBFloatTy() || - U.getOperand(0

[clang] [llvm] [RFC][WIP][AMDGPU] Use `bf16` instead of `i16` for bfloat (PR #80908)

https://github.com/shiltian updated https://github.com/llvm/llvm-project/pull/80908 >From 6a2bacee940d95abc53bcff2332b0d9aa0f1073f Mon Sep 17 00:00:00 2001 From: Shilei Tian Date: Wed, 7 Feb 2024 23:09:33 -0500 Subject: [PATCH] [RFC][WIP][AMDGPU] Use `bf16` instead of `i16` for bfloat Currentl

[clang] [llvm] [RFC][WIP][AMDGPU] Use `bf16` instead of `i16` for bfloat (PR #80908)

@@ -2819,11 +2819,11 @@ def int_amdgcn_fdot2_f16_f16 : def int_amdgcn_fdot2_bf16_bf16 : ClangBuiltin<"__builtin_amdgcn_fdot2_bf16_bf16">, DefaultAttrsIntrinsic< -[llvm_i16_ty], // %r +[llvm_bfloat_ty], // %r shiltian wrote: Does it make sense i

[clang] [llvm] [RFC][WIP][AMDGPU] Use `bf16` instead of `i16` for bfloat (PR #80908)

@@ -1562,8 +1562,9 @@ bool IRTranslator::translateBitCast(const User &U, bool IRTranslator::translateCast(unsigned Opcode, const User &U, MachineIRBuilder &MIRBuilder) { - if (U.getType()->getScalarType()->isBFloatTy() || - U.getOperand(0

[clang] [llvm] [RFC][WIP][AMDGPU] Use `bf16` instead of `i16` for bfloat (PR #80908)

https://github.com/shiltian updated https://github.com/llvm/llvm-project/pull/80908 >From a535bf3e8cd9b10d87281f94fed68fc300f3e24c Mon Sep 17 00:00:00 2001 From: Shilei Tian Date: Wed, 7 Feb 2024 14:43:24 -0500 Subject: [PATCH] [RFC][WIP][AMDGPU] Use `bf16` instead of `i16` for bfloat Currentl