aboutsummaryrefslogtreecommitdiff
path: root/test/CodeGen/AMDGPU/udiv.ll
diff options
context:
space:
mode:
Diffstat (limited to 'test/CodeGen/AMDGPU/udiv.ll')
-rw-r--r--test/CodeGen/AMDGPU/udiv.ll72
1 files changed, 56 insertions, 16 deletions
diff --git a/test/CodeGen/AMDGPU/udiv.ll b/test/CodeGen/AMDGPU/udiv.ll
index da88d2a8e8cb..2874a0cdbc05 100644
--- a/test/CodeGen/AMDGPU/udiv.ll
+++ b/test/CodeGen/AMDGPU/udiv.ll
@@ -1,22 +1,27 @@
; RUN: llc -march=amdgcn -mcpu=verde -verify-machineinstrs < %s | FileCheck -check-prefix=SI -check-prefix=FUNC %s
-; RUN: llc -march=amdgcn -mcpu=tonga -mattr=-flat-for-global -verify-machineinstrs < %s | FileCheck -check-prefix=SI -check-prefix=FUNC %s
+; RUN: llc -march=amdgcn -mcpu=tonga -mattr=-flat-for-global -verify-machineinstrs -mattr=-fp32-denormals < %s | FileCheck -check-prefix=SI -check-prefix=FUNC -check-prefix=VI %s
+
+; RUN: llc -mtriple=amdgcn--amdhsa -mcpu=fiji -mattr=+fp32-denormals < %s | FileCheck -check-prefix=GCN -check-prefix=VI %s
+
; RUN: llc -march=r600 -mcpu=redwood < %s | FileCheck -check-prefix=EG -check-prefix=FUNC %s
; FUNC-LABEL: {{^}}udiv_i32:
; EG-NOT: SETGE_INT
; EG: CF_END
-define void @udiv_i32(i32 addrspace(1)* %out, i32 addrspace(1)* %in) {
+
+; SI: v_rcp_iflag_f32_e32
+define amdgpu_kernel void @udiv_i32(i32 addrspace(1)* %out, i32 addrspace(1)* %in) {
%b_ptr = getelementptr i32, i32 addrspace(1)* %in, i32 1
- %a = load i32, i32 addrspace(1) * %in
- %b = load i32, i32 addrspace(1) * %b_ptr
+ %a = load i32, i32 addrspace(1)* %in
+ %b = load i32, i32 addrspace(1)* %b_ptr
%result = udiv i32 %a, %b
store i32 %result, i32 addrspace(1)* %out
ret void
}
; FUNC-LABEL: {{^}}s_udiv_i32:
-
-define void @s_udiv_i32(i32 addrspace(1)* %out, i32 %a, i32 %b) {
+; SI: v_rcp_iflag_f32_e32
+define amdgpu_kernel void @s_udiv_i32(i32 addrspace(1)* %out, i32 %a, i32 %b) {
%result = udiv i32 %a, %b
store i32 %result, i32 addrspace(1)* %out
ret void
@@ -30,8 +35,10 @@ define void @s_udiv_i32(i32 addrspace(1)* %out, i32 %a, i32 %b) {
; FUNC-LABEL: {{^}}udiv_v2i32:
; EG: CF_END
+; SI: v_rcp_iflag_f32_e32
+; SI: v_rcp_iflag_f32_e32
; SI: s_endpgm
-define void @udiv_v2i32(<2 x i32> addrspace(1)* %out, <2 x i32> addrspace(1)* %in) {
+define amdgpu_kernel void @udiv_v2i32(<2 x i32> addrspace(1)* %out, <2 x i32> addrspace(1)* %in) {
%b_ptr = getelementptr <2 x i32>, <2 x i32> addrspace(1)* %in, i32 1
%a = load <2 x i32>, <2 x i32> addrspace(1) * %in
%b = load <2 x i32>, <2 x i32> addrspace(1) * %b_ptr
@@ -43,7 +50,7 @@ define void @udiv_v2i32(<2 x i32> addrspace(1)* %out, <2 x i32> addrspace(1)* %i
; FUNC-LABEL: {{^}}udiv_v4i32:
; EG: CF_END
; SI: s_endpgm
-define void @udiv_v4i32(<4 x i32> addrspace(1)* %out, <4 x i32> addrspace(1)* %in) {
+define amdgpu_kernel void @udiv_v4i32(<4 x i32> addrspace(1)* %out, <4 x i32> addrspace(1)* %in) {
%b_ptr = getelementptr <4 x i32>, <4 x i32> addrspace(1)* %in, i32 1
%a = load <4 x i32>, <4 x i32> addrspace(1) * %in
%b = load <4 x i32>, <4 x i32> addrspace(1) * %b_ptr
@@ -56,7 +63,7 @@ define void @udiv_v4i32(<4 x i32> addrspace(1)* %out, <4 x i32> addrspace(1)* %i
; SI: buffer_load_dword [[VAL:v[0-9]+]]
; SI: v_lshrrev_b32_e32 [[RESULT:v[0-9]+]], 4, [[VAL]]
; SI: buffer_store_dword [[RESULT]]
-define void @udiv_i32_div_pow2(i32 addrspace(1)* %out, i32 addrspace(1)* %in) {
+define amdgpu_kernel void @udiv_i32_div_pow2(i32 addrspace(1)* %out, i32 addrspace(1)* %in) {
%b_ptr = getelementptr i32, i32 addrspace(1)* %in, i32 1
%a = load i32, i32 addrspace(1)* %in
%result = udiv i32 %a, 16
@@ -70,7 +77,7 @@ define void @udiv_i32_div_pow2(i32 addrspace(1)* %out, i32 addrspace(1)* %in) {
; SI: v_mul_hi_u32 [[MULHI:v[0-9]+]], [[K]], [[VAL]]
; SI: v_lshrrev_b32_e32 [[RESULT:v[0-9]+]], 25, [[MULHI]]
; SI: buffer_store_dword [[RESULT]]
-define void @udiv_i32_div_k_even(i32 addrspace(1)* %out, i32 addrspace(1)* %in) {
+define amdgpu_kernel void @udiv_i32_div_k_even(i32 addrspace(1)* %out, i32 addrspace(1)* %in) {
%b_ptr = getelementptr i32, i32 addrspace(1)* %in, i32 1
%a = load i32, i32 addrspace(1)* %in
%result = udiv i32 %a, 34259182
@@ -84,7 +91,7 @@ define void @udiv_i32_div_k_even(i32 addrspace(1)* %out, i32 addrspace(1)* %in)
; SI: v_mul_hi_u32 [[MULHI:v[0-9]+]], [[K]], [[VAL]]
; SI: v_lshrrev_b32_e32 [[RESULT:v[0-9]+]], 24, [[MULHI]]
; SI: buffer_store_dword [[RESULT]]
-define void @udiv_i32_div_k_odd(i32 addrspace(1)* %out, i32 addrspace(1)* %in) {
+define amdgpu_kernel void @udiv_i32_div_k_odd(i32 addrspace(1)* %out, i32 addrspace(1)* %in) {
%b_ptr = getelementptr i32, i32 addrspace(1)* %in, i32 1
%a = load i32, i32 addrspace(1)* %in
%result = udiv i32 %a, 34259183
@@ -96,7 +103,7 @@ define void @udiv_i32_div_k_odd(i32 addrspace(1)* %out, i32 addrspace(1)* %in) {
; SI: v_rcp_f32
; SI: v_and_b32_e32 [[TRUNC:v[0-9]+]], 0xff, v{{[0-9]+}}
; SI: buffer_store_dword [[TRUNC]]
-define void @v_udiv_i8(i32 addrspace(1)* %out, i8 addrspace(1)* %in) {
+define amdgpu_kernel void @v_udiv_i8(i32 addrspace(1)* %out, i8 addrspace(1)* %in) {
%den_ptr = getelementptr i8, i8 addrspace(1)* %in, i8 1
%num = load i8, i8 addrspace(1) * %in
%den = load i8, i8 addrspace(1) * %den_ptr
@@ -110,7 +117,7 @@ define void @v_udiv_i8(i32 addrspace(1)* %out, i8 addrspace(1)* %in) {
; SI: v_rcp_f32
; SI: v_and_b32_e32 [[TRUNC:v[0-9]+]], 0xffff, v{{[0-9]+}}
; SI: buffer_store_dword [[TRUNC]]
-define void @v_udiv_i16(i32 addrspace(1)* %out, i16 addrspace(1)* %in) {
+define amdgpu_kernel void @v_udiv_i16(i32 addrspace(1)* %out, i16 addrspace(1)* %in) {
%den_ptr = getelementptr i16, i16 addrspace(1)* %in, i16 1
%num = load i16, i16 addrspace(1) * %in
%den = load i16, i16 addrspace(1) * %den_ptr
@@ -124,7 +131,7 @@ define void @v_udiv_i16(i32 addrspace(1)* %out, i16 addrspace(1)* %in) {
; SI: v_rcp_f32
; SI: v_and_b32_e32 [[TRUNC:v[0-9]+]], 0x7fffff, v{{[0-9]+}}
; SI: buffer_store_dword [[TRUNC]]
-define void @v_udiv_i23(i32 addrspace(1)* %out, i23 addrspace(1)* %in) {
+define amdgpu_kernel void @v_udiv_i23(i32 addrspace(1)* %out, i23 addrspace(1)* %in) {
%den_ptr = getelementptr i23, i23 addrspace(1)* %in, i23 1
%num = load i23, i23 addrspace(1) * %in
%den = load i23, i23 addrspace(1) * %den_ptr
@@ -136,7 +143,7 @@ define void @v_udiv_i23(i32 addrspace(1)* %out, i23 addrspace(1)* %in) {
; FUNC-LABEL: {{^}}v_udiv_i24:
; SI-NOT: v_rcp_f32
-define void @v_udiv_i24(i32 addrspace(1)* %out, i24 addrspace(1)* %in) {
+define amdgpu_kernel void @v_udiv_i24(i32 addrspace(1)* %out, i24 addrspace(1)* %in) {
%den_ptr = getelementptr i24, i24 addrspace(1)* %in, i24 1
%num = load i24, i24 addrspace(1) * %in
%den = load i24, i24 addrspace(1) * %den_ptr
@@ -152,9 +159,42 @@ define void @v_udiv_i24(i32 addrspace(1)* %out, i24 addrspace(1)* %in) {
; SI: v_mul_hi_u32
; SI: v_mul_hi_u32
-define void @scalarize_mulhu_4xi32(<4 x i32> addrspace(1)* nocapture readonly %in, <4 x i32> addrspace(1)* nocapture %out) {
+define amdgpu_kernel void @scalarize_mulhu_4xi32(<4 x i32> addrspace(1)* nocapture readonly %in, <4 x i32> addrspace(1)* nocapture %out) {
%1 = load <4 x i32>, <4 x i32> addrspace(1)* %in, align 16
%2 = udiv <4 x i32> %1, <i32 53668, i32 53668, i32 53668, i32 53668>
store <4 x i32> %2, <4 x i32> addrspace(1)* %out, align 16
ret void
}
+
+; FUNC-LABEL: {{^}}test_udiv2:
+; SI: s_lshr_b32 s{{[0-9]}}, s{{[0-9]}}, 1
+define amdgpu_kernel void @test_udiv2(i32 %p) {
+ %i = udiv i32 %p, 2
+ store volatile i32 %i, i32 addrspace(1)* undef
+ ret void
+}
+
+; FUNC-LABEL: {{^}}test_udiv_3_mulhu:
+; SI: v_mov_b32_e32 v{{[0-9]+}}, 0xaaaaaaab
+; SI: v_mul_hi_u32 v0, {{v[0-9]+}}, {{s[0-9]+}}
+; SI-NEXT: v_lshrrev_b32_e32 v0, 1, v0
+define amdgpu_kernel void @test_udiv_3_mulhu(i32 %p) {
+ %i = udiv i32 %p, 3
+ store volatile i32 %i, i32 addrspace(1)* undef
+ ret void
+}
+
+; GCN-LABEL: {{^}}fdiv_test_denormals
+; VI: v_mad_f32 v{{[0-9]+}}, -v{{[0-9]+}}, v{{[0-9]+}}, v{{[0-9]+}}
+define amdgpu_kernel void @fdiv_test_denormals(i8 addrspace(1)* nocapture readonly %arg) {
+bb:
+ %tmp = load i8, i8 addrspace(1)* null, align 1
+ %tmp1 = sext i8 %tmp to i32
+ %tmp2 = getelementptr inbounds i8, i8 addrspace(1)* %arg, i64 undef
+ %tmp3 = load i8, i8 addrspace(1)* %tmp2, align 1
+ %tmp4 = sext i8 %tmp3 to i32
+ %tmp5 = sdiv i32 %tmp1, %tmp4
+ %tmp6 = trunc i32 %tmp5 to i8
+ store i8 %tmp6, i8 addrspace(1)* null, align 1
+ ret void
+}
generated by cgit v1.2.3 (git 2.25.1) at 2025-03-06 21:38:49 +0000