forked from OSchip/llvm-project
AMDGPU: Reduce code size with fcanonicalize (fneg x)
When fcanonicalize is lowered to a mul, we can use -1.0 for free and avoid the cost of the bigger encoding for source modifers. llvm-svn: 338244
This commit is contained in:
Matt Arsenault
parent
f3c9a34def
commit
de496c32a4
|
|
@ -567,6 +567,7 @@ int PI = 0x40490fdb;
|
|||
int TWO_PI_INV = 0x3e22f983;
|
||||
int FP_UINT_MAX_PLUS_1 = 0x4f800000; // 1 << 32 in floating point encoding
|
||||
int FP16_ONE = 0x3C00;
|
||||
int FP16_NEG_ONE = 0xBC00;
|
||||
int V2FP16_ONE = 0x3C003C00;
|
||||
int FP32_ONE = 0x3f800000;
|
||||
int FP32_NEG_ONE = 0xbf800000;
|
||||
|
|
|
|||
|
|
@ -1386,6 +1386,11 @@ def : GCNPat<
|
|||
(V_MUL_F16_e64 0, (i32 CONST.FP16_ONE), $src_mods, $src, 0, 0)
|
||||
>;
|
||||
|
||||
def : GCNPat<
|
||||
(fcanonicalize (f16 (fneg (VOP3Mods f16:$src, i32:$src_mods)))),
|
||||
(V_MUL_F16_e64 0, (i32 CONST.FP16_NEG_ONE), $src_mods, $src, 0, 0)
|
||||
>;
|
||||
|
||||
def : GCNPat<
|
||||
(fcanonicalize (v2f16 (VOP3PMods v2f16:$src, i32:$src_mods))),
|
||||
(V_PK_MUL_F16 0, (i32 CONST.V2FP16_ONE), $src_mods, $src, DSTCLAMP.NONE)
|
||||
|
|
@ -1411,6 +1416,11 @@ def : GCNPat<
|
|||
(fcanonicalize (f32 (VOP3Mods f32:$src, i32:$src_mods))),
|
||||
(V_MUL_F32_e64 0, (i32 CONST.FP32_ONE), $src_mods, $src, 0, 0)
|
||||
>;
|
||||
|
||||
def : GCNPat<
|
||||
(fcanonicalize (f32 (fneg (VOP3Mods f32:$src, i32:$src_mods)))),
|
||||
(V_MUL_F32_e64 0, (i32 CONST.FP32_NEG_ONE), $src_mods, $src, 0, 0)
|
||||
>;
|
||||
}
|
||||
|
||||
let OtherPredicates = [FP32Denormals] in {
|
||||
|
|
|
|||
|
|
@ -225,7 +225,7 @@ define amdgpu_kernel void @test_fold_canonicalize_fpround_value_v2f16_v2f32(<2 x
|
|||
}
|
||||
|
||||
; GCN-LABEL: test_no_fold_canonicalize_fneg_value_f32:
|
||||
; GCN-FLUSH: v_mul_f32_e64 v{{[0-9]+}}, 1.0, -v{{[0-9]+}}
|
||||
; GCN-FLUSH: v_mul_f32_e32 v{{[0-9]+}}, -1.0, v{{[0-9]+}}
|
||||
; GCN-DENORM: v_max_f32_e64 v{{[0-9]+}}, -v{{[0-9]+}}, -v{{[0-9]+}}
|
||||
define amdgpu_kernel void @test_no_fold_canonicalize_fneg_value_f32(float addrspace(1)* %arg) {
|
||||
%id = tail call i32 @llvm.amdgcn.workitem.id.x()
|
||||
|
|
|
|||
|
|
@ -1,5 +1,5 @@
|
|||
; RUN: llc -march=amdgcn -mcpu=tonga -mattr=-flat-for-global -verify-machineinstrs < %s | FileCheck -check-prefix=GCN -check-prefix=VI %s
|
||||
; RUN: llc -march=amdgcn -mcpu=gfx900 -mattr=-flat-for-global -verify-machineinstrs < %s | FileCheck -check-prefix=GCN -check-prefix=GFX9 %s
|
||||
; RUN: llc -march=amdgcn -mcpu=tonga -verify-machineinstrs < %s | FileCheck -enable-var-scope -check-prefixes=GCN,VI %s
|
||||
; RUN: llc -march=amdgcn -mcpu=gfx900 -verify-machineinstrs < %s | FileCheck -enable-var-scope -check-prefixes=GCN,GFX9 %s
|
||||
|
||||
declare half @llvm.fabs.f16(half) #0
|
||||
declare half @llvm.canonicalize.f16(half) #0
|
||||
|
|
@ -10,7 +10,7 @@ declare i32 @llvm.amdgcn.workitem.id.x() #0
|
|||
|
||||
; GCN-LABEL: {{^}}v_test_canonicalize_var_f16:
|
||||
; GCN: v_max_f16_e32 [[REG:v[0-9]+]], {{v[0-9]+}}, {{v[0-9]+}}
|
||||
; GCN: buffer_store_short [[REG]]
|
||||
; GCN: {{flat|global}}_store_short v{{\[[0-9]+:[0-9]+\]}}, [[REG]]
|
||||
define amdgpu_kernel void @v_test_canonicalize_var_f16(half addrspace(1)* %out) #1 {
|
||||
%val = load half, half addrspace(1)* %out
|
||||
%canonicalized = call half @llvm.canonicalize.f16(half %val)
|
||||
|
|
@ -20,7 +20,7 @@ define amdgpu_kernel void @v_test_canonicalize_var_f16(half addrspace(1)* %out)
|
|||
|
||||
; GCN-LABEL: {{^}}s_test_canonicalize_var_f16:
|
||||
; GCN: v_max_f16_e64 [[REG:v[0-9]+]], {{s[0-9]+}}, {{s[0-9]+}}
|
||||
; GCN: buffer_store_short [[REG]]
|
||||
; GCN: {{flat|global}}_store_short v{{\[[0-9]+:[0-9]+\]}}, [[REG]]
|
||||
define amdgpu_kernel void @s_test_canonicalize_var_f16(half addrspace(1)* %out, i16 zeroext %val.arg) #1 {
|
||||
%val = bitcast i16 %val.arg to half
|
||||
%canonicalized = call half @llvm.canonicalize.f16(half %val)
|
||||
|
|
@ -30,7 +30,7 @@ define amdgpu_kernel void @s_test_canonicalize_var_f16(half addrspace(1)* %out,
|
|||
|
||||
; GCN-LABEL: {{^}}v_test_canonicalize_fabs_var_f16:
|
||||
; GCN: v_max_f16_e64 [[REG:v[0-9]+]], |{{v[0-9]+}}|, |{{v[0-9]+}}|
|
||||
; GCN: buffer_store_short [[REG]]
|
||||
; GCN: {{flat|global}}_store_short v{{\[[0-9]+:[0-9]+\]}}, [[REG]]
|
||||
define amdgpu_kernel void @v_test_canonicalize_fabs_var_f16(half addrspace(1)* %out) #1 {
|
||||
%val = load half, half addrspace(1)* %out
|
||||
%val.fabs = call half @llvm.fabs.f16(half %val)
|
||||
|
|
@ -41,7 +41,7 @@ define amdgpu_kernel void @v_test_canonicalize_fabs_var_f16(half addrspace(1)* %
|
|||
|
||||
; GCN-LABEL: {{^}}v_test_canonicalize_fneg_fabs_var_f16:
|
||||
; GCN: v_max_f16_e64 [[REG:v[0-9]+]], -|{{v[0-9]+}}|, -|{{v[0-9]+}}|
|
||||
; GCN: buffer_store_short [[REG]]
|
||||
; GCN: {{flat|global}}_store_short v{{\[[0-9]+:[0-9]+\]}}, [[REG]]
|
||||
define amdgpu_kernel void @v_test_canonicalize_fneg_fabs_var_f16(half addrspace(1)* %out) #1 {
|
||||
%val = load half, half addrspace(1)* %out
|
||||
%val.fabs = call half @llvm.fabs.f16(half %val)
|
||||
|
|
@ -53,7 +53,7 @@ define amdgpu_kernel void @v_test_canonicalize_fneg_fabs_var_f16(half addrspace(
|
|||
|
||||
; GCN-LABEL: {{^}}v_test_canonicalize_fneg_var_f16:
|
||||
; GCN: v_max_f16_e64 [[REG:v[0-9]+]], -{{v[0-9]+}}, -{{v[0-9]+}}
|
||||
; GCN: buffer_store_short [[REG]]
|
||||
; GCN: {{flat|global}}_store_short v{{\[[0-9]+:[0-9]+\]}}, [[REG]]
|
||||
define amdgpu_kernel void @v_test_canonicalize_fneg_var_f16(half addrspace(1)* %out) #1 {
|
||||
%val = load half, half addrspace(1)* %out
|
||||
%val.fneg = fsub half -0.0, %val
|
||||
|
|
@ -62,9 +62,32 @@ define amdgpu_kernel void @v_test_canonicalize_fneg_var_f16(half addrspace(1)* %
|
|||
ret void
|
||||
}
|
||||
|
||||
; GCN-LABEL: {{^}}v_test_no_denormals_canonicalize_fneg_var_f16:
|
||||
; GCN: v_mul_f16_e32 [[REG:v[0-9]+]], -1.0, v{{[0-9]+}}
|
||||
; GCN: {{flat|global}}_store_short v{{\[[0-9]+:[0-9]+\]}}, [[REG]]
|
||||
define amdgpu_kernel void @v_test_no_denormals_canonicalize_fneg_var_f16(half addrspace(1)* %out) #2 {
|
||||
%val = load half, half addrspace(1)* %out
|
||||
%val.fneg = fsub half -0.0, %val
|
||||
%canonicalized = call half @llvm.canonicalize.f16(half %val.fneg)
|
||||
store half %canonicalized, half addrspace(1)* %out
|
||||
ret void
|
||||
}
|
||||
|
||||
; GCN-LABEL: {{^}}v_test_no_denormals_canonicalize_fneg_fabs_var_f16:
|
||||
; GCN: v_mul_f16_e64 [[REG:v[0-9]+]], -1.0, |v{{[0-9]+}}|
|
||||
; GCN: {{flat|global}}_store_short v{{\[[0-9]+:[0-9]+\]}}, [[REG]]
|
||||
define amdgpu_kernel void @v_test_no_denormals_canonicalize_fneg_fabs_var_f16(half addrspace(1)* %out) #2 {
|
||||
%val = load half, half addrspace(1)* %out
|
||||
%val.fabs = call half @llvm.fabs.f16(half %val)
|
||||
%val.fabs.fneg = fsub half -0.0, %val.fabs
|
||||
%canonicalized = call half @llvm.canonicalize.f16(half %val.fabs.fneg)
|
||||
store half %canonicalized, half addrspace(1)* %out
|
||||
ret void
|
||||
}
|
||||
|
||||
; GCN-LABEL: {{^}}test_fold_canonicalize_p0_f16:
|
||||
; GCN: v_mov_b32_e32 [[REG:v[0-9]+]], 0{{$}}
|
||||
; GCN: buffer_store_short [[REG]]
|
||||
; GCN: {{flat|global}}_store_short v{{\[[0-9]+:[0-9]+\]}}, [[REG]]
|
||||
define amdgpu_kernel void @test_fold_canonicalize_p0_f16(half addrspace(1)* %out) #1 {
|
||||
%canonicalized = call half @llvm.canonicalize.f16(half 0.0)
|
||||
store half %canonicalized, half addrspace(1)* %out
|
||||
|
|
@ -73,7 +96,7 @@ define amdgpu_kernel void @test_fold_canonicalize_p0_f16(half addrspace(1)* %out
|
|||
|
||||
; GCN-LABEL: {{^}}test_fold_canonicalize_n0_f16:
|
||||
; GCN: v_mov_b32_e32 [[REG:v[0-9]+]], 0xffff8000{{$}}
|
||||
; GCN: buffer_store_short [[REG]]
|
||||
; GCN: {{flat|global}}_store_short v{{\[[0-9]+:[0-9]+\]}}, [[REG]]
|
||||
define amdgpu_kernel void @test_fold_canonicalize_n0_f16(half addrspace(1)* %out) #1 {
|
||||
%canonicalized = call half @llvm.canonicalize.f16(half -0.0)
|
||||
store half %canonicalized, half addrspace(1)* %out
|
||||
|
|
@ -82,7 +105,7 @@ define amdgpu_kernel void @test_fold_canonicalize_n0_f16(half addrspace(1)* %out
|
|||
|
||||
; GCN-LABEL: {{^}}test_fold_canonicalize_p1_f16:
|
||||
; GCN: v_mov_b32_e32 [[REG:v[0-9]+]], 0x3c00{{$}}
|
||||
; GCN: buffer_store_short [[REG]]
|
||||
; GCN: {{flat|global}}_store_short v{{\[[0-9]+:[0-9]+\]}}, [[REG]]
|
||||
define amdgpu_kernel void @test_fold_canonicalize_p1_f16(half addrspace(1)* %out) #1 {
|
||||
%canonicalized = call half @llvm.canonicalize.f16(half 1.0)
|
||||
store half %canonicalized, half addrspace(1)* %out
|
||||
|
|
@ -91,7 +114,7 @@ define amdgpu_kernel void @test_fold_canonicalize_p1_f16(half addrspace(1)* %out
|
|||
|
||||
; GCN-LABEL: {{^}}test_fold_canonicalize_n1_f16:
|
||||
; GCN: v_mov_b32_e32 [[REG:v[0-9]+]], 0xffffbc00{{$}}
|
||||
; GCN: buffer_store_short [[REG]]
|
||||
; GCN: {{flat|global}}_store_short v{{\[[0-9]+:[0-9]+\]}}, [[REG]]
|
||||
define amdgpu_kernel void @test_fold_canonicalize_n1_f16(half addrspace(1)* %out) #1 {
|
||||
%canonicalized = call half @llvm.canonicalize.f16(half -1.0)
|
||||
store half %canonicalized, half addrspace(1)* %out
|
||||
|
|
@ -100,7 +123,7 @@ define amdgpu_kernel void @test_fold_canonicalize_n1_f16(half addrspace(1)* %out
|
|||
|
||||
; GCN-LABEL: {{^}}test_fold_canonicalize_literal_f16:
|
||||
; GCN: v_mov_b32_e32 [[REG:v[0-9]+]], 0x4c00{{$}}
|
||||
; GCN: buffer_store_short [[REG]]
|
||||
; GCN: {{flat|global}}_store_short v{{\[[0-9]+:[0-9]+\]}}, [[REG]]
|
||||
define amdgpu_kernel void @test_fold_canonicalize_literal_f16(half addrspace(1)* %out) #1 {
|
||||
%canonicalized = call half @llvm.canonicalize.f16(half 16.0)
|
||||
store half %canonicalized, half addrspace(1)* %out
|
||||
|
|
@ -109,7 +132,7 @@ define amdgpu_kernel void @test_fold_canonicalize_literal_f16(half addrspace(1)*
|
|||
|
||||
; GCN-LABEL: {{^}}test_default_denormals_fold_canonicalize_denormal0_f16:
|
||||
; GCN: v_mov_b32_e32 [[REG:v[0-9]+]], 0x3ff{{$}}
|
||||
; GCN: buffer_store_short [[REG]]
|
||||
; GCN: {{flat|global}}_store_short v{{\[[0-9]+:[0-9]+\]}}, [[REG]]
|
||||
define amdgpu_kernel void @test_default_denormals_fold_canonicalize_denormal0_f16(half addrspace(1)* %out) #1 {
|
||||
%canonicalized = call half @llvm.canonicalize.f16(half 0xH03FF)
|
||||
store half %canonicalized, half addrspace(1)* %out
|
||||
|
|
@ -118,7 +141,7 @@ define amdgpu_kernel void @test_default_denormals_fold_canonicalize_denormal0_f1
|
|||
|
||||
; GCN-LABEL: {{^}}test_denormals_fold_canonicalize_denormal0_f16:
|
||||
; GCN: v_mov_b32_e32 [[REG:v[0-9]+]], 0x3ff{{$}}
|
||||
; GCN: buffer_store_short [[REG]]
|
||||
; GCN: {{flat|global}}_store_short v{{\[[0-9]+:[0-9]+\]}}, [[REG]]
|
||||
define amdgpu_kernel void @test_denormals_fold_canonicalize_denormal0_f16(half addrspace(1)* %out) #3 {
|
||||
%canonicalized = call half @llvm.canonicalize.f16(half 0xH03FF)
|
||||
store half %canonicalized, half addrspace(1)* %out
|
||||
|
|
@ -127,7 +150,7 @@ define amdgpu_kernel void @test_denormals_fold_canonicalize_denormal0_f16(half a
|
|||
|
||||
; GCN-LABEL: {{^}}test_default_denormals_fold_canonicalize_denormal1_f16:
|
||||
; GCN: v_mov_b32_e32 [[REG:v[0-9]+]], 0xffff83ff{{$}}
|
||||
; GCN: buffer_store_short [[REG]]
|
||||
; GCN: {{flat|global}}_store_short v{{\[[0-9]+:[0-9]+\]}}, [[REG]]
|
||||
define amdgpu_kernel void @test_default_denormals_fold_canonicalize_denormal1_f16(half addrspace(1)* %out) #1 {
|
||||
%canonicalized = call half @llvm.canonicalize.f16(half 0xH83FF)
|
||||
store half %canonicalized, half addrspace(1)* %out
|
||||
|
|
@ -136,7 +159,7 @@ define amdgpu_kernel void @test_default_denormals_fold_canonicalize_denormal1_f1
|
|||
|
||||
; GCN-LABEL: {{^}}test_denormals_fold_canonicalize_denormal1_f16:
|
||||
; GCN: v_mov_b32_e32 [[REG:v[0-9]+]], 0xffff83ff{{$}}
|
||||
; GCN: buffer_store_short [[REG]]
|
||||
; GCN: {{flat|global}}_store_short v{{\[[0-9]+:[0-9]+\]}}, [[REG]]
|
||||
define amdgpu_kernel void @test_denormals_fold_canonicalize_denormal1_f16(half addrspace(1)* %out) #3 {
|
||||
%canonicalized = call half @llvm.canonicalize.f16(half 0xH83FF)
|
||||
store half %canonicalized, half addrspace(1)* %out
|
||||
|
|
@ -145,7 +168,7 @@ define amdgpu_kernel void @test_denormals_fold_canonicalize_denormal1_f16(half a
|
|||
|
||||
; GCN-LABEL: {{^}}test_fold_canonicalize_qnan_f16:
|
||||
; GCN: v_mov_b32_e32 [[REG:v[0-9]+]], 0x7c00{{$}}
|
||||
; GCN: buffer_store_short [[REG]]
|
||||
; GCN: {{flat|global}}_store_short v{{\[[0-9]+:[0-9]+\]}}, [[REG]]
|
||||
define amdgpu_kernel void @test_fold_canonicalize_qnan_f16(half addrspace(1)* %out) #1 {
|
||||
%canonicalized = call half @llvm.canonicalize.f16(half 0xH7C00)
|
||||
store half %canonicalized, half addrspace(1)* %out
|
||||
|
|
@ -154,7 +177,7 @@ define amdgpu_kernel void @test_fold_canonicalize_qnan_f16(half addrspace(1)* %o
|
|||
|
||||
; GCN-LABEL: {{^}}test_fold_canonicalize_qnan_value_neg1_f16:
|
||||
; GCN: v_mov_b32_e32 [[REG:v[0-9]+]], 0x7e00{{$}}
|
||||
; GCN: buffer_store_short [[REG]]
|
||||
; GCN: {{flat|global}}_store_short v{{\[[0-9]+:[0-9]+\]}}, [[REG]]
|
||||
define amdgpu_kernel void @test_fold_canonicalize_qnan_value_neg1_f16(half addrspace(1)* %out) #1 {
|
||||
%canonicalized = call half @llvm.canonicalize.f16(half bitcast (i16 -1 to half))
|
||||
store half %canonicalized, half addrspace(1)* %out
|
||||
|
|
@ -163,7 +186,7 @@ define amdgpu_kernel void @test_fold_canonicalize_qnan_value_neg1_f16(half addrs
|
|||
|
||||
; GCN-LABEL: {{^}}test_fold_canonicalize_qnan_value_neg2_f16:
|
||||
; GCN: v_mov_b32_e32 [[REG:v[0-9]+]], 0x7e00{{$}}
|
||||
; GCN: buffer_store_short [[REG]]
|
||||
; GCN: {{flat|global}}_store_short v{{\[[0-9]+:[0-9]+\]}}, [[REG]]
|
||||
define amdgpu_kernel void @test_fold_canonicalize_qnan_value_neg2_f16(half addrspace(1)* %out) #1 {
|
||||
%canonicalized = call half @llvm.canonicalize.f16(half bitcast (i16 -2 to half))
|
||||
store half %canonicalized, half addrspace(1)* %out
|
||||
|
|
@ -172,7 +195,7 @@ define amdgpu_kernel void @test_fold_canonicalize_qnan_value_neg2_f16(half addrs
|
|||
|
||||
; GCN-LABEL: {{^}}test_fold_canonicalize_snan0_value_f16:
|
||||
; GCN: v_mov_b32_e32 [[REG:v[0-9]+]], 0x7e00{{$}}
|
||||
; GCN: buffer_store_short [[REG]]
|
||||
; GCN: {{flat|global}}_store_short v{{\[[0-9]+:[0-9]+\]}}, [[REG]]
|
||||
define amdgpu_kernel void @test_fold_canonicalize_snan0_value_f16(half addrspace(1)* %out) #1 {
|
||||
%canonicalized = call half @llvm.canonicalize.f16(half 0xH7C01)
|
||||
store half %canonicalized, half addrspace(1)* %out
|
||||
|
|
@ -181,7 +204,7 @@ define amdgpu_kernel void @test_fold_canonicalize_snan0_value_f16(half addrspace
|
|||
|
||||
; GCN-LABEL: {{^}}test_fold_canonicalize_snan1_value_f16:
|
||||
; GCN: v_mov_b32_e32 [[REG:v[0-9]+]], 0x7e00{{$}}
|
||||
; GCN: buffer_store_short [[REG]]
|
||||
; GCN: {{flat|global}}_store_short v{{\[[0-9]+:[0-9]+\]}}, [[REG]]
|
||||
define amdgpu_kernel void @test_fold_canonicalize_snan1_value_f16(half addrspace(1)* %out) #1 {
|
||||
%canonicalized = call half @llvm.canonicalize.f16(half 0xH7DFF)
|
||||
store half %canonicalized, half addrspace(1)* %out
|
||||
|
|
@ -190,7 +213,7 @@ define amdgpu_kernel void @test_fold_canonicalize_snan1_value_f16(half addrspace
|
|||
|
||||
; GCN-LABEL: {{^}}test_fold_canonicalize_snan2_value_f16:
|
||||
; GCN: v_mov_b32_e32 [[REG:v[0-9]+]], 0x7e00{{$}}
|
||||
; GCN: buffer_store_short [[REG]]
|
||||
; GCN: {{flat|global}}_store_short v{{\[[0-9]+:[0-9]+\]}}, [[REG]]
|
||||
define amdgpu_kernel void @test_fold_canonicalize_snan2_value_f16(half addrspace(1)* %out) #1 {
|
||||
%canonicalized = call half @llvm.canonicalize.f16(half 0xHFDFF)
|
||||
store half %canonicalized, half addrspace(1)* %out
|
||||
|
|
@ -199,7 +222,7 @@ define amdgpu_kernel void @test_fold_canonicalize_snan2_value_f16(half addrspace
|
|||
|
||||
; GCN-LABEL: {{^}}test_fold_canonicalize_snan3_value_f16:
|
||||
; GCN: v_mov_b32_e32 [[REG:v[0-9]+]], 0x7e00{{$}}
|
||||
; GCN: buffer_store_short [[REG]]
|
||||
; GCN: {{flat|global}}_store_short v{{\[[0-9]+:[0-9]+\]}}, [[REG]]
|
||||
define amdgpu_kernel void @test_fold_canonicalize_snan3_value_f16(half addrspace(1)* %out) #1 {
|
||||
%canonicalized = call half @llvm.canonicalize.f16(half 0xHFC01)
|
||||
store half %canonicalized, half addrspace(1)* %out
|
||||
|
|
@ -212,7 +235,7 @@ define amdgpu_kernel void @test_fold_canonicalize_snan3_value_f16(half addrspace
|
|||
; VI-NOT: v_and_b32
|
||||
|
||||
; GFX9: v_pk_max_f16 [[REG:v[0-9]+]], {{v[0-9]+}}, {{v[0-9]+$}}
|
||||
; GFX9: buffer_store_dword [[REG]]
|
||||
; GFX9: {{flat|global}}_store_dword v{{\[[0-9]+:[0-9]+\]}}, [[REG]]
|
||||
define amdgpu_kernel void @v_test_canonicalize_var_v2f16(<2 x half> addrspace(1)* %out) #1 {
|
||||
%tid = call i32 @llvm.amdgcn.workitem.id.x()
|
||||
%gep = getelementptr <2 x half>, <2 x half> addrspace(1)* %out, i32 %tid
|
||||
|
|
@ -230,7 +253,7 @@ define amdgpu_kernel void @v_test_canonicalize_var_v2f16(<2 x half> addrspace(1)
|
|||
|
||||
; GFX9: v_and_b32_e32 [[ABS:v[0-9]+]], 0x7fff7fff, v{{[0-9]+}}
|
||||
; GFX9: v_pk_max_f16 [[REG:v[0-9]+]], [[ABS]], [[ABS]]{{$}}
|
||||
; GCN: buffer_store_dword
|
||||
; GCN: {{flat|global}}_store_dword
|
||||
define amdgpu_kernel void @v_test_canonicalize_fabs_var_v2f16(<2 x half> addrspace(1)* %out) #1 {
|
||||
%tid = call i32 @llvm.amdgcn.workitem.id.x()
|
||||
%gep = getelementptr <2 x half>, <2 x half> addrspace(1)* %out, i32 %tid
|
||||
|
|
@ -248,7 +271,7 @@ define amdgpu_kernel void @v_test_canonicalize_fabs_var_v2f16(<2 x half> addrspa
|
|||
|
||||
; GFX9: v_and_b32_e32 [[ABS:v[0-9]+]], 0x7fff7fff, v{{[0-9]+}}
|
||||
; GFX9: v_pk_max_f16 [[REG:v[0-9]+]], [[ABS]], [[ABS]] neg_lo:[1,1] neg_hi:[1,1]{{$}}
|
||||
; GCN: buffer_store_dword
|
||||
; GCN: {{flat|global}}_store_dword
|
||||
define amdgpu_kernel void @v_test_canonicalize_fneg_fabs_var_v2f16(<2 x half> addrspace(1)* %out) #1 {
|
||||
%tid = call i32 @llvm.amdgcn.workitem.id.x()
|
||||
%gep = getelementptr <2 x half>, <2 x half> addrspace(1)* %out, i32 %tid
|
||||
|
|
@ -266,7 +289,7 @@ define amdgpu_kernel void @v_test_canonicalize_fneg_fabs_var_v2f16(<2 x half> ad
|
|||
; VI-NOT: 0xffff
|
||||
|
||||
; GFX9: v_pk_max_f16 [[REG:v[0-9]+]], {{v[0-9]+}}, {{v[0-9]+}} neg_lo:[1,1] neg_hi:[1,1]{{$}}
|
||||
; GFX9: buffer_store_dword [[REG]]
|
||||
; GFX9: {{flat|global}}_store_dword v{{\[[0-9]+:[0-9]+\]}}, [[REG]]
|
||||
define amdgpu_kernel void @v_test_canonicalize_fneg_var_v2f16(<2 x half> addrspace(1)* %out) #1 {
|
||||
%tid = call i32 @llvm.amdgcn.workitem.id.x()
|
||||
%gep = getelementptr <2 x half>, <2 x half> addrspace(1)* %out, i32 %tid
|
||||
|
|
@ -283,7 +306,7 @@ define amdgpu_kernel void @v_test_canonicalize_fneg_var_v2f16(<2 x half> addrspa
|
|||
; VI-NOT: v_and_b32
|
||||
|
||||
; GFX9: v_pk_max_f16 [[REG:v[0-9]+]], {{s[0-9]+}}, {{s[0-9]+$}}
|
||||
; GFX9: buffer_store_dword [[REG]]
|
||||
; GFX9: {{flat|global}}_store_dword v{{\[[0-9]+:[0-9]+\]}}, [[REG]]
|
||||
define amdgpu_kernel void @s_test_canonicalize_var_v2f16(<2 x half> addrspace(1)* %out, i32 zeroext %val.arg) #1 {
|
||||
%val = bitcast i32 %val.arg to <2 x half>
|
||||
%canonicalized = call <2 x half> @llvm.canonicalize.v2f16(<2 x half> %val)
|
||||
|
|
@ -293,7 +316,7 @@ define amdgpu_kernel void @s_test_canonicalize_var_v2f16(<2 x half> addrspace(1)
|
|||
|
||||
; GCN-LABEL: {{^}}test_fold_canonicalize_p0_v2f16:
|
||||
; GCN: v_mov_b32_e32 [[REG:v[0-9]+]], 0{{$}}
|
||||
; GCN: buffer_store_dword [[REG]]
|
||||
; GCN: {{flat|global}}_store_dword v{{\[[0-9]+:[0-9]+\]}}, [[REG]]
|
||||
define amdgpu_kernel void @test_fold_canonicalize_p0_v2f16(<2 x half> addrspace(1)* %out) #1 {
|
||||
%canonicalized = call <2 x half> @llvm.canonicalize.v2f16(<2 x half> zeroinitializer)
|
||||
store <2 x half> %canonicalized, <2 x half> addrspace(1)* %out
|
||||
|
|
@ -302,7 +325,7 @@ define amdgpu_kernel void @test_fold_canonicalize_p0_v2f16(<2 x half> addrspace(
|
|||
|
||||
; GCN-LABEL: {{^}}test_fold_canonicalize_n0_v2f16:
|
||||
; GCN: v_mov_b32_e32 [[REG:v[0-9]+]], 0x80008000{{$}}
|
||||
; GCN: buffer_store_dword [[REG]]
|
||||
; GCN: {{flat|global}}_store_dword v{{\[[0-9]+:[0-9]+\]}}, [[REG]]
|
||||
define amdgpu_kernel void @test_fold_canonicalize_n0_v2f16(<2 x half> addrspace(1)* %out) #1 {
|
||||
%canonicalized = call <2 x half> @llvm.canonicalize.v2f16(<2 x half> <half -0.0, half -0.0>)
|
||||
store <2 x half> %canonicalized, <2 x half> addrspace(1)* %out
|
||||
|
|
@ -311,7 +334,7 @@ define amdgpu_kernel void @test_fold_canonicalize_n0_v2f16(<2 x half> addrspace(
|
|||
|
||||
; GCN-LABEL: {{^}}test_fold_canonicalize_p1_v2f16:
|
||||
; GCN: v_mov_b32_e32 [[REG:v[0-9]+]], 0x3c003c00{{$}}
|
||||
; GCN: buffer_store_dword [[REG]]
|
||||
; GCN: {{flat|global}}_store_dword v{{\[[0-9]+:[0-9]+\]}}, [[REG]]
|
||||
define amdgpu_kernel void @test_fold_canonicalize_p1_v2f16(<2 x half> addrspace(1)* %out) #1 {
|
||||
%canonicalized = call <2 x half> @llvm.canonicalize.v2f16(<2 x half> <half 1.0, half 1.0>)
|
||||
store <2 x half> %canonicalized, <2 x half> addrspace(1)* %out
|
||||
|
|
@ -320,7 +343,7 @@ define amdgpu_kernel void @test_fold_canonicalize_p1_v2f16(<2 x half> addrspace(
|
|||
|
||||
; GCN-LABEL: {{^}}test_fold_canonicalize_n1_v2f16:
|
||||
; GCN: v_mov_b32_e32 [[REG:v[0-9]+]], 0xbc00bc00{{$}}
|
||||
; GCN: buffer_store_dword [[REG]]
|
||||
; GCN: {{flat|global}}_store_dword v{{\[[0-9]+:[0-9]+\]}}, [[REG]]
|
||||
define amdgpu_kernel void @test_fold_canonicalize_n1_v2f16(<2 x half> addrspace(1)* %out) #1 {
|
||||
%canonicalized = call <2 x half> @llvm.canonicalize.v2f16(<2 x half> <half -1.0, half -1.0>)
|
||||
store <2 x half> %canonicalized, <2 x half> addrspace(1)* %out
|
||||
|
|
@ -329,7 +352,7 @@ define amdgpu_kernel void @test_fold_canonicalize_n1_v2f16(<2 x half> addrspace(
|
|||
|
||||
; GCN-LABEL: {{^}}test_fold_canonicalize_literal_v2f16:
|
||||
; GCN: v_mov_b32_e32 [[REG:v[0-9]+]], 0x4c004c00{{$}}
|
||||
; GCN: buffer_store_dword [[REG]]
|
||||
; GCN: {{flat|global}}_store_dword v{{\[[0-9]+:[0-9]+\]}}, [[REG]]
|
||||
define amdgpu_kernel void @test_fold_canonicalize_literal_v2f16(<2 x half> addrspace(1)* %out) #1 {
|
||||
%canonicalized = call <2 x half> @llvm.canonicalize.v2f16(<2 x half> <half 16.0, half 16.0>)
|
||||
store <2 x half> %canonicalized, <2 x half> addrspace(1)* %out
|
||||
|
|
@ -338,7 +361,7 @@ define amdgpu_kernel void @test_fold_canonicalize_literal_v2f16(<2 x half> addrs
|
|||
|
||||
; GCN-LABEL: {{^}}test_no_denormals_fold_canonicalize_denormal0_v2f16:
|
||||
; GCN: v_mov_b32_e32 [[REG:v[0-9]+]], 0x3ff03ff{{$}}
|
||||
; GCN: buffer_store_dword [[REG]]
|
||||
; GCN: {{flat|global}}_store_dword v{{\[[0-9]+:[0-9]+\]}}, [[REG]]
|
||||
define amdgpu_kernel void @test_no_denormals_fold_canonicalize_denormal0_v2f16(<2 x half> addrspace(1)* %out) #1 {
|
||||
%canonicalized = call <2 x half> @llvm.canonicalize.v2f16(<2 x half> <half 0xH03FF, half 0xH03FF>)
|
||||
store <2 x half> %canonicalized, <2 x half> addrspace(1)* %out
|
||||
|
|
@ -347,7 +370,7 @@ define amdgpu_kernel void @test_no_denormals_fold_canonicalize_denormal0_v2f16(<
|
|||
|
||||
; GCN-LABEL: {{^}}test_denormals_fold_canonicalize_denormal0_v2f16:
|
||||
; GCN: v_mov_b32_e32 [[REG:v[0-9]+]], 0x3ff03ff{{$}}
|
||||
; GCN: buffer_store_dword [[REG]]
|
||||
; GCN: {{flat|global}}_store_dword v{{\[[0-9]+:[0-9]+\]}}, [[REG]]
|
||||
define amdgpu_kernel void @test_denormals_fold_canonicalize_denormal0_v2f16(<2 x half> addrspace(1)* %out) #3 {
|
||||
%canonicalized = call <2 x half> @llvm.canonicalize.v2f16(<2 x half> <half 0xH03FF, half 0xH03FF>)
|
||||
store <2 x half> %canonicalized, <2 x half> addrspace(1)* %out
|
||||
|
|
@ -356,7 +379,7 @@ define amdgpu_kernel void @test_denormals_fold_canonicalize_denormal0_v2f16(<2 x
|
|||
|
||||
; GCN-LABEL: {{^}}test_no_denormals_fold_canonicalize_denormal1_v2f16:
|
||||
; GCN: v_mov_b32_e32 [[REG:v[0-9]+]], 0x83ff83ff{{$}}
|
||||
; GCN: buffer_store_dword [[REG]]
|
||||
; GCN: {{flat|global}}_store_dword v{{\[[0-9]+:[0-9]+\]}}, [[REG]]
|
||||
define amdgpu_kernel void @test_no_denormals_fold_canonicalize_denormal1_v2f16(<2 x half> addrspace(1)* %out) #1 {
|
||||
%canonicalized = call <2 x half> @llvm.canonicalize.v2f16(<2 x half> <half 0xH83FF, half 0xH83FF>)
|
||||
store <2 x half> %canonicalized, <2 x half> addrspace(1)* %out
|
||||
|
|
@ -365,7 +388,7 @@ define amdgpu_kernel void @test_no_denormals_fold_canonicalize_denormal1_v2f16(<
|
|||
|
||||
; GCN-LABEL: {{^}}test_denormals_fold_canonicalize_denormal1_v2f16:
|
||||
; GCN: v_mov_b32_e32 [[REG:v[0-9]+]], 0x83ff83ff{{$}}
|
||||
; GCN: buffer_store_dword [[REG]]
|
||||
; GCN: {{flat|global}}_store_dword v{{\[[0-9]+:[0-9]+\]}}, [[REG]]
|
||||
define amdgpu_kernel void @test_denormals_fold_canonicalize_denormal1_v2f16(<2 x half> addrspace(1)* %out) #3 {
|
||||
%canonicalized = call <2 x half> @llvm.canonicalize.v2f16(<2 x half> <half 0xH83FF, half 0xH83FF>)
|
||||
store <2 x half> %canonicalized, <2 x half> addrspace(1)* %out
|
||||
|
|
@ -374,7 +397,7 @@ define amdgpu_kernel void @test_denormals_fold_canonicalize_denormal1_v2f16(<2 x
|
|||
|
||||
; GCN-LABEL: {{^}}test_fold_canonicalize_qnan_v2f16:
|
||||
; GCN: v_mov_b32_e32 [[REG:v[0-9]+]], 0x7c007c00{{$}}
|
||||
; GCN: buffer_store_dword [[REG]]
|
||||
; GCN: {{flat|global}}_store_dword v{{\[[0-9]+:[0-9]+\]}}, [[REG]]
|
||||
define amdgpu_kernel void @test_fold_canonicalize_qnan_v2f16(<2 x half> addrspace(1)* %out) #1 {
|
||||
%canonicalized = call <2 x half> @llvm.canonicalize.v2f16(<2 x half> <half 0xH7C00, half 0xH7C00>)
|
||||
store <2 x half> %canonicalized, <2 x half> addrspace(1)* %out
|
||||
|
|
@ -383,7 +406,7 @@ define amdgpu_kernel void @test_fold_canonicalize_qnan_v2f16(<2 x half> addrspac
|
|||
|
||||
; GCN-LABEL: {{^}}test_fold_canonicalize_qnan_value_neg1_v2f16:
|
||||
; GCN: v_mov_b32_e32 [[REG:v[0-9]+]], 0x7e007e00{{$}}
|
||||
; GCN: buffer_store_dword [[REG]]
|
||||
; GCN: {{flat|global}}_store_dword v{{\[[0-9]+:[0-9]+\]}}, [[REG]]
|
||||
define amdgpu_kernel void @test_fold_canonicalize_qnan_value_neg1_v2f16(<2 x half> addrspace(1)* %out) #1 {
|
||||
%canonicalized = call <2 x half> @llvm.canonicalize.v2f16(<2 x half> bitcast (i32 -1 to <2 x half>))
|
||||
store <2 x half> %canonicalized, <2 x half> addrspace(1)* %out
|
||||
|
|
@ -392,7 +415,7 @@ define amdgpu_kernel void @test_fold_canonicalize_qnan_value_neg1_v2f16(<2 x hal
|
|||
|
||||
; GCN-LABEL: {{^}}test_fold_canonicalize_qnan_value_neg2_v2f16:
|
||||
; GCN: v_mov_b32_e32 [[REG:v[0-9]+]], 0x7e007e00{{$}}
|
||||
; GCN: buffer_store_dword [[REG]]
|
||||
; GCN: {{flat|global}}_store_dword v{{\[[0-9]+:[0-9]+\]}}, [[REG]]
|
||||
define amdgpu_kernel void @test_fold_canonicalize_qnan_value_neg2_v2f16(<2 x half> addrspace(1)* %out) #1 {
|
||||
%canonicalized = call <2 x half> @llvm.canonicalize.v2f16(<2 x half> <half bitcast (i16 -2 to half), half bitcast (i16 -2 to half)>)
|
||||
store <2 x half> %canonicalized, <2 x half> addrspace(1)* %out
|
||||
|
|
@ -401,7 +424,7 @@ define amdgpu_kernel void @test_fold_canonicalize_qnan_value_neg2_v2f16(<2 x hal
|
|||
|
||||
; GCN-LABEL: {{^}}test_fold_canonicalize_snan0_value_v2f16:
|
||||
; GCN: v_mov_b32_e32 [[REG:v[0-9]+]], 0x7e007e00{{$}}
|
||||
; GCN: buffer_store_dword [[REG]]
|
||||
; GCN: {{flat|global}}_store_dword v{{\[[0-9]+:[0-9]+\]}}, [[REG]]
|
||||
define amdgpu_kernel void @test_fold_canonicalize_snan0_value_v2f16(<2 x half> addrspace(1)* %out) #1 {
|
||||
%canonicalized = call <2 x half> @llvm.canonicalize.v2f16(<2 x half> <half 0xH7C01, half 0xH7C01>)
|
||||
store <2 x half> %canonicalized, <2 x half> addrspace(1)* %out
|
||||
|
|
@ -410,7 +433,7 @@ define amdgpu_kernel void @test_fold_canonicalize_snan0_value_v2f16(<2 x half> a
|
|||
|
||||
; GCN-LABEL: {{^}}test_fold_canonicalize_snan1_value_v2f16:
|
||||
; GCN: v_mov_b32_e32 [[REG:v[0-9]+]], 0x7e007e00{{$}}
|
||||
; GCN: buffer_store_dword [[REG]]
|
||||
; GCN: {{flat|global}}_store_dword v{{\[[0-9]+:[0-9]+\]}}, [[REG]]
|
||||
define amdgpu_kernel void @test_fold_canonicalize_snan1_value_v2f16(<2 x half> addrspace(1)* %out) #1 {
|
||||
%canonicalized = call <2 x half> @llvm.canonicalize.v2f16(<2 x half> <half 0xH7DFF, half 0xH7DFF>)
|
||||
store <2 x half> %canonicalized, <2 x half> addrspace(1)* %out
|
||||
|
|
@ -419,7 +442,7 @@ define amdgpu_kernel void @test_fold_canonicalize_snan1_value_v2f16(<2 x half> a
|
|||
|
||||
; GCN-LABEL: {{^}}test_fold_canonicalize_snan2_value_v2f16:
|
||||
; GCN: v_mov_b32_e32 [[REG:v[0-9]+]], 0x7e007e00{{$}}
|
||||
; GCN: buffer_store_dword [[REG]]
|
||||
; GCN: {{flat|global}}_store_dword v{{\[[0-9]+:[0-9]+\]}}, [[REG]]
|
||||
define amdgpu_kernel void @test_fold_canonicalize_snan2_value_v2f16(<2 x half> addrspace(1)* %out) #1 {
|
||||
%canonicalized = call <2 x half> @llvm.canonicalize.v2f16(<2 x half> <half 0xHFDFF, half 0xHFDFF>)
|
||||
store <2 x half> %canonicalized, <2 x half> addrspace(1)* %out
|
||||
|
|
@ -428,7 +451,7 @@ define amdgpu_kernel void @test_fold_canonicalize_snan2_value_v2f16(<2 x half> a
|
|||
|
||||
; GCN-LABEL: {{^}}test_fold_canonicalize_snan3_value_v2f16:
|
||||
; GCN: v_mov_b32_e32 [[REG:v[0-9]+]], 0x7e007e00{{$}}
|
||||
; GCN: buffer_store_dword [[REG]]
|
||||
; GCN: {{flat|global}}_store_dword v{{\[[0-9]+:[0-9]+\]}}, [[REG]]
|
||||
define amdgpu_kernel void @test_fold_canonicalize_snan3_value_v2f16(<2 x half> addrspace(1)* %out) #1 {
|
||||
%canonicalized = call <2 x half> @llvm.canonicalize.v2f16(<2 x half> <half 0xHFC01, half 0xHFC01>)
|
||||
store <2 x half> %canonicalized, <2 x half> addrspace(1)* %out
|
||||
|
|
|
|||
|
|
@ -40,7 +40,7 @@ define amdgpu_kernel void @v_test_canonicalize_fabs_var_f32(float addrspace(1)*
|
|||
}
|
||||
|
||||
; GCN-LABEL: {{^}}v_test_canonicalize_fneg_fabs_var_f32:
|
||||
; GCN: v_mul_f32_e64 [[REG:v[0-9]+]], 1.0, -|{{v[0-9]+}}|
|
||||
; GCN: v_mul_f32_e64 [[REG:v[0-9]+]], -1.0, |{{v[0-9]+}}|
|
||||
; GCN: buffer_store_dword [[REG]]
|
||||
define amdgpu_kernel void @v_test_canonicalize_fneg_fabs_var_f32(float addrspace(1)* %out) #1 {
|
||||
%val = load float, float addrspace(1)* %out
|
||||
|
|
@ -52,7 +52,7 @@ define amdgpu_kernel void @v_test_canonicalize_fneg_fabs_var_f32(float addrspace
|
|||
}
|
||||
|
||||
; GCN-LABEL: {{^}}v_test_canonicalize_fneg_var_f32:
|
||||
; GCN: v_mul_f32_e64 [[REG:v[0-9]+]], 1.0, -{{v[0-9]+}}
|
||||
; GCN: v_mul_f32_e32 [[REG:v[0-9]+]], -1.0, {{v[0-9]+}}
|
||||
; GCN: buffer_store_dword [[REG]]
|
||||
define amdgpu_kernel void @v_test_canonicalize_fneg_var_f32(float addrspace(1)* %out) #1 {
|
||||
%val = load float, float addrspace(1)* %out
|
||||
|
|
|
|||
|
|
@ -1730,7 +1730,7 @@ define amdgpu_kernel void @v_fneg_nearbyint_f32(float addrspace(1)* %out, float
|
|||
|
||||
; GCN-LABEL: {{^}}v_fneg_canonicalize_f32:
|
||||
; GCN: {{buffer|flat}}_load_dword [[A:v[0-9]+]]
|
||||
; GCN: v_mul_f32_e64 [[RESULT:v[0-9]+]], 1.0, -[[A]]
|
||||
; GCN: v_mul_f32_e32 [[RESULT:v[0-9]+]], -1.0, [[A]]
|
||||
; GCN: buffer_store_dword [[RESULT]]
|
||||
define amdgpu_kernel void @v_fneg_canonicalize_f32(float addrspace(1)* %out, float addrspace(1)* %a.ptr) #0 {
|
||||
%tid = call i32 @llvm.amdgcn.workitem.id.x()
|
||||
|
|
|
|||
Loading…
Reference in New Issue