mirror of
https://github.com/RPCS3/llvm-mirror.git
synced 2024-11-26 04:32:44 +01:00
38b9cda6d0
The previous implementation was incorrect, and based off incorrect instruction definitions. Unfortunately we can't match natural addressing in a lot of cases due to the shift/scale applied in getelementptrs. This relies on reducing the 64-bit shift to 32-bits.
275 lines
11 KiB
LLVM
275 lines
11 KiB
LLVM
; RUN: llc -march=amdgcn -mcpu=tahiti -denormal-fp-math-f32=preserve-sign -verify-machineinstrs < %s | FileCheck -check-prefix=GCN -check-prefix=SI %s
|
|
; RUN: llc -march=amdgcn -mcpu=fiji -mattr=-flat-for-global -denormal-fp-math-f32=preserve-sign -verify-machineinstrs < %s | FileCheck -check-prefix=GCN -check-prefix=GFX8_9_10 %s
|
|
; RUN: llc -march=amdgcn -mcpu=fiji -mattr=-flat-for-global -denormal-fp-math-f32=preserve-sign -verify-machineinstrs < %s | FileCheck -check-prefix=GCN -check-prefix=GFX8_9_10 %s
|
|
; RUN: llc -march=amdgcn -mcpu=gfx900 -mattr=-flat-for-global -denormal-fp-math-f32=preserve-sign -verify-machineinstrs < %s | FileCheck -check-prefix=GCN -check-prefix=GFX8_9_10 %s
|
|
; RUN: llc -march=amdgcn -mcpu=gfx1010 -mattr=-flat-for-global -denormal-fp-math-f32=preserve-sign -verify-machineinstrs < %s | FileCheck -check-prefix=GCN -check-prefix=GFX8_9_10 %s
|
|
|
|
; Make sure fdiv is promoted to f32.
|
|
|
|
; GCN-LABEL: {{^}}v_fdiv_f16
|
|
; SI: v_cvt_f32_f16
|
|
; SI: v_cvt_f32_f16
|
|
; SI: v_div_scale_f32
|
|
; SI-DAG: v_div_scale_f32
|
|
; SI-DAG: v_rcp_f32
|
|
; SI: v_fma_f32
|
|
; SI: v_fma_f32
|
|
; SI: v_mul_f32
|
|
; SI: v_fma_f32
|
|
; SI: v_fma_f32
|
|
; SI: v_fma_f32
|
|
; SI: v_div_fmas_f32
|
|
; SI: v_div_fixup_f32
|
|
; SI: v_cvt_f16_f32
|
|
|
|
; GFX8_9_10: {{flat|global}}_load_ushort [[LHS:v[0-9]+]]
|
|
; GFX8_9_10: {{flat|global}}_load_ushort [[RHS:v[0-9]+]]
|
|
|
|
; GFX8_9_10-DAG: v_cvt_f32_f16_e32 [[CVT_LHS:v[0-9]+]], [[LHS]]
|
|
; GFX8_9_10-DAG: v_cvt_f32_f16_e32 [[CVT_RHS:v[0-9]+]], [[RHS]]
|
|
|
|
; GFX8_9_10-DAG: v_rcp_f32_e32 [[RCP_RHS:v[0-9]+]], [[CVT_RHS]]
|
|
; GFX8_9_10: v_mul_f32_e32 [[MUL:v[0-9]+]], [[CVT_LHS]], [[RCP_RHS]]
|
|
; GFX8_9_10: v_cvt_f16_f32_e32 [[CVT_BACK:v[0-9]+]], [[MUL]]
|
|
; GFX8_9_10: v_div_fixup_f16 [[RESULT:v[0-9]+]], [[CVT_BACK]], [[RHS]], [[LHS]]
|
|
; GFX8_9_10: {{flat|global}}_store_short v{{.+}}, [[RESULT]]
|
|
define amdgpu_kernel void @v_fdiv_f16(
|
|
half addrspace(1)* %r,
|
|
half addrspace(1)* %a,
|
|
half addrspace(1)* %b) #0 {
|
|
entry:
|
|
%tid = call i32 @llvm.amdgcn.workitem.id.x()
|
|
%tid.ext = sext i32 %tid to i64
|
|
%gep.a = getelementptr inbounds half, half addrspace(1)* %a, i64 %tid.ext
|
|
%gep.b = getelementptr inbounds half, half addrspace(1)* %b, i64 %tid.ext
|
|
%gep.r = getelementptr inbounds half, half addrspace(1)* %r, i64 %tid.ext
|
|
%a.val = load volatile half, half addrspace(1)* %gep.a
|
|
%b.val = load volatile half, half addrspace(1)* %gep.b
|
|
%r.val = fdiv half %a.val, %b.val
|
|
store half %r.val, half addrspace(1)* %gep.r
|
|
ret void
|
|
}
|
|
|
|
; GCN-LABEL: {{^}}v_rcp_f16:
|
|
; GFX8_9_10: {{flat|global}}_load_ushort [[VAL:v[0-9]+]]
|
|
; GFX8_9_10-NOT: [[VAL]]
|
|
; GFX8_9_10: v_rcp_f16_e32 [[RESULT:v[0-9]+]], [[VAL]]
|
|
; GFX8_9_10-NOT: [[RESULT]]
|
|
; GFX8_9_10: {{flat|global}}_store_short v{{.+}}, [[RESULT]]
|
|
define amdgpu_kernel void @v_rcp_f16(half addrspace(1)* %r, half addrspace(1)* %b) #0 {
|
|
entry:
|
|
%tid = call i32 @llvm.amdgcn.workitem.id.x()
|
|
%tid.ext = sext i32 %tid to i64
|
|
%gep.b = getelementptr inbounds half, half addrspace(1)* %b, i64 %tid.ext
|
|
%gep.r = getelementptr inbounds half, half addrspace(1)* %r, i64 %tid.ext
|
|
%b.val = load volatile half, half addrspace(1)* %gep.b
|
|
%r.val = fdiv half 1.0, %b.val, !fpmath !0
|
|
store half %r.val, half addrspace(1)* %gep.r
|
|
ret void
|
|
}
|
|
|
|
; GCN-LABEL: {{^}}v_rcp_f16_abs:
|
|
; GFX8_9_10: {{flat|global}}_load_ushort [[VAL:v[0-9]+]]
|
|
; GFX8_9_10-NOT: [[VAL]]
|
|
; GFX8_9_10: v_rcp_f16_e64 [[RESULT:v[0-9]+]], |[[VAL]]|
|
|
; GFX8_9_10-NOT: [RESULT]]
|
|
; GFX8_9_10: {{flat|global}}_store_short v{{.+}}, [[RESULT]]
|
|
define amdgpu_kernel void @v_rcp_f16_abs(half addrspace(1)* %r, half addrspace(1)* %b) #0 {
|
|
entry:
|
|
%tid = call i32 @llvm.amdgcn.workitem.id.x()
|
|
%tid.ext = sext i32 %tid to i64
|
|
%gep.b = getelementptr inbounds half, half addrspace(1)* %b, i64 %tid.ext
|
|
%gep.r = getelementptr inbounds half, half addrspace(1)* %r, i64 %tid.ext
|
|
%b.val = load volatile half, half addrspace(1)* %gep.b
|
|
%b.abs = call half @llvm.fabs.f16(half %b.val)
|
|
%r.val = fdiv half 1.0, %b.abs, !fpmath !0
|
|
store half %r.val, half addrspace(1)* %gep.r
|
|
ret void
|
|
}
|
|
|
|
; We could not do 1/b -> rcp_f16(b) under !fpmath < 1ulp.
|
|
|
|
; GCN-LABEL: {{^}}reciprocal_f16_rounded:
|
|
; GFX8_9_10: {{flat|global}}_load_ushort [[VAL16:v[0-9]+]], v{{.+}}
|
|
; GFX8_9_10: v_cvt_f32_f16_e32 [[CVT_TO32:v[0-9]+]], [[VAL16]]
|
|
; GFX8_9_10: v_rcp_f32_e32 [[RCP32:v[0-9]+]], [[CVT_TO32]]
|
|
; GFX8_9_10: v_cvt_f16_f32_e32 [[CVT_BACK16:v[0-9]+]], [[RCP32]]
|
|
; GFX8_9_10: v_div_fixup_f16 [[RESULT:v[0-9]+]], [[CVT_BACK16]], [[VAL16]], 1.0
|
|
; GFX8_9_10: {{flat|global}}_store_short v{{.+}}, [[RESULT]]
|
|
define amdgpu_kernel void @reciprocal_f16_rounded(half addrspace(1)* %r, half addrspace(1)* %b) #0 {
|
|
entry:
|
|
%tid = call i32 @llvm.amdgcn.workitem.id.x()
|
|
%tid.ext = sext i32 %tid to i64
|
|
%gep.b = getelementptr inbounds half, half addrspace(1)* %b, i64 %tid.ext
|
|
%gep.r = getelementptr inbounds half, half addrspace(1)* %r, i64 %tid.ext
|
|
%b.val = load volatile half, half addrspace(1)* %gep.b
|
|
%r.val = fdiv half 1.0, %b.val
|
|
store half %r.val, half addrspace(1)* %gep.r
|
|
ret void
|
|
}
|
|
|
|
; GCN-LABEL: {{^}}v_rcp_f16_afn:
|
|
; GFX8_9_10: {{flat|global}}_load_ushort [[VAL:v[0-9]+]]
|
|
; GFX8_9_10-NOT: [[VAL]]
|
|
; GFX8_9_10: v_rcp_f16_e32 [[RESULT:v[0-9]+]], [[VAL]]
|
|
; GFX8_9_10-NOT: [[RESULT]]
|
|
; GFX8_9_10: {{flat|global}}_store_short v{{.+}}, [[RESULT]]
|
|
define amdgpu_kernel void @v_rcp_f16_afn(half addrspace(1)* %r, half addrspace(1)* %b) #0 {
|
|
entry:
|
|
%tid = call i32 @llvm.amdgcn.workitem.id.x()
|
|
%tid.ext = sext i32 %tid to i64
|
|
%gep.b = getelementptr inbounds half, half addrspace(1)* %b, i64 %tid.ext
|
|
%gep.r = getelementptr inbounds half, half addrspace(1)* %r, i64 %tid.ext
|
|
%b.val = load volatile half, half addrspace(1)* %gep.b
|
|
%r.val = fdiv afn half 1.0, %b.val, !fpmath !0
|
|
store half %r.val, half addrspace(1)* %gep.r
|
|
ret void
|
|
}
|
|
|
|
; GCN-LABEL: {{^}}v_rcp_f16_neg:
|
|
; GFX8_9_10: {{flat|global}}_load_ushort [[VAL:v[0-9]+]]
|
|
; GFX8_9_10-NOT: [[VAL]]
|
|
; GFX8_9_10: v_rcp_f16_e64 [[RESULT:v[0-9]+]], -[[VAL]]
|
|
; GFX8_9_10-NOT: [RESULT]]
|
|
; GFX8_9_10: {{flat|global}}_store_short v{{.+}}, [[RESULT]]
|
|
define amdgpu_kernel void @v_rcp_f16_neg(half addrspace(1)* %r, half addrspace(1)* %b) #0 {
|
|
entry:
|
|
%tid = call i32 @llvm.amdgcn.workitem.id.x()
|
|
%tid.ext = sext i32 %tid to i64
|
|
%gep.b = getelementptr inbounds half, half addrspace(1)* %b, i64 %tid.ext
|
|
%gep.r = getelementptr inbounds half, half addrspace(1)* %r, i64 %tid.ext
|
|
%b.val = load volatile half, half addrspace(1)* %gep.b
|
|
%r.val = fdiv half -1.0, %b.val, !fpmath !0
|
|
store half %r.val, half addrspace(1)* %gep.r
|
|
ret void
|
|
}
|
|
|
|
; GCN-LABEL: {{^}}v_rsq_f16:
|
|
; GFX8_9_10: {{flat|global}}_load_ushort [[VAL:v[0-9]+]]
|
|
; GFX8_9_10-NOT: [[VAL]]
|
|
; GFX8_9_10: v_rsq_f16_e32 [[RESULT:v[0-9]+]], [[VAL]]
|
|
; GFX8_9_10-NOT: [RESULT]]
|
|
; GFX8_9_10: {{flat|global}}_store_short v{{.+}}, [[RESULT]]
|
|
define amdgpu_kernel void @v_rsq_f16(half addrspace(1)* %r, half addrspace(1)* %b) #0 {
|
|
entry:
|
|
%tid = call i32 @llvm.amdgcn.workitem.id.x()
|
|
%tid.ext = sext i32 %tid to i64
|
|
%gep.b = getelementptr inbounds half, half addrspace(1)* %b, i64 %tid.ext
|
|
%gep.r = getelementptr inbounds half, half addrspace(1)* %r, i64 %tid.ext
|
|
%b.val = load volatile half, half addrspace(1)* %gep.b
|
|
%b.sqrt = call half @llvm.sqrt.f16(half %b.val)
|
|
%r.val = fdiv half 1.0, %b.sqrt, !fpmath !0
|
|
store half %r.val, half addrspace(1)* %gep.r
|
|
ret void
|
|
}
|
|
|
|
; GCN-LABEL: {{^}}v_rsq_f16_neg:
|
|
; GFX8_9_10: {{flat|global}}_load_ushort [[VAL:v[0-9]+]]
|
|
; GFX8_9_10-NOT: [[VAL]]
|
|
; GFX8_9_10: v_sqrt_f16_e32 [[SQRT:v[0-9]+]], [[VAL]]
|
|
; GFX8_9_10-NEXT: v_rcp_f16_e64 [[RESULT:v[0-9]+]], -[[SQRT]]
|
|
; GFX8_9_10-NOT: [RESULT]]
|
|
; GFX8_9_10: {{flat|global}}_store_short v{{.+}}, [[RESULT]]
|
|
define amdgpu_kernel void @v_rsq_f16_neg(half addrspace(1)* %r, half addrspace(1)* %b) #0 {
|
|
entry:
|
|
%tid = call i32 @llvm.amdgcn.workitem.id.x()
|
|
%tid.ext = sext i32 %tid to i64
|
|
%gep.b = getelementptr inbounds half, half addrspace(1)* %b, i64 %tid.ext
|
|
%gep.r = getelementptr inbounds half, half addrspace(1)* %r, i64 %tid.ext
|
|
%b.val = load volatile half, half addrspace(1)* %gep.b
|
|
%b.sqrt = call half @llvm.sqrt.f16(half %b.val)
|
|
%r.val = fdiv half -1.0, %b.sqrt, !fpmath !0
|
|
store half %r.val, half addrspace(1)* %gep.r
|
|
ret void
|
|
}
|
|
|
|
; GCN-LABEL: {{^}}v_fdiv_f16_afn:
|
|
; GFX8_9_10: {{flat|global}}_load_ushort [[LHS:v[0-9]+]]
|
|
; GFX8_9_10: {{flat|global}}_load_ushort [[RHS:v[0-9]+]]
|
|
|
|
; GFX8_9_10: v_rcp_f16_e32 [[RCP:v[0-9]+]], [[RHS]]
|
|
; GFX8_9_10: v_mul_f16_e32 [[RESULT:v[0-9]+]], [[LHS]], [[RCP]]
|
|
|
|
; GFX8_9_10: {{flat|global}}_store_short v{{.+}}, [[RESULT]]
|
|
define amdgpu_kernel void @v_fdiv_f16_afn(half addrspace(1)* %r, half addrspace(1)* %a, half addrspace(1)* %b) #0 {
|
|
entry:
|
|
%tid = call i32 @llvm.amdgcn.workitem.id.x()
|
|
%tid.ext = sext i32 %tid to i64
|
|
%gep.a = getelementptr inbounds half, half addrspace(1)* %a, i64 %tid.ext
|
|
%gep.b = getelementptr inbounds half, half addrspace(1)* %b, i64 %tid.ext
|
|
%gep.r = getelementptr inbounds half, half addrspace(1)* %r, i64 %tid.ext
|
|
%a.val = load volatile half, half addrspace(1)* %gep.a
|
|
%b.val = load volatile half, half addrspace(1)* %gep.b
|
|
%r.val = fdiv afn half %a.val, %b.val
|
|
store half %r.val, half addrspace(1)* %gep.r
|
|
ret void
|
|
}
|
|
|
|
; GCN-LABEL: {{^}}v_fdiv_f16_unsafe:
|
|
; GFX8_9_10: {{flat|global}}_load_ushort [[LHS:v[0-9]+]]
|
|
; GFX8_9_10: {{flat|global}}_load_ushort [[RHS:v[0-9]+]]
|
|
|
|
; GFX8_9_10: v_rcp_f16_e32 [[RCP:v[0-9]+]], [[RHS]]
|
|
; GFX8_9_10: v_mul_f16_e32 [[RESULT:v[0-9]+]], [[LHS]], [[RCP]]
|
|
|
|
; GFX8_9_10: {{flat|global}}_store_short v{{.+}}, [[RESULT]]
|
|
define amdgpu_kernel void @v_fdiv_f16_unsafe(half addrspace(1)* %r, half addrspace(1)* %a, half addrspace(1)* %b) #2 {
|
|
entry:
|
|
%tid = call i32 @llvm.amdgcn.workitem.id.x()
|
|
%tid.ext = sext i32 %tid to i64
|
|
%gep.a = getelementptr inbounds half, half addrspace(1)* %a, i64 %tid.ext
|
|
%gep.b = getelementptr inbounds half, half addrspace(1)* %b, i64 %tid.ext
|
|
%gep.r = getelementptr inbounds half, half addrspace(1)* %r, i64 %tid.ext
|
|
%a.val = load volatile half, half addrspace(1)* %gep.a
|
|
%b.val = load volatile half, half addrspace(1)* %gep.b
|
|
%r.val = fdiv half %a.val, %b.val
|
|
store half %r.val, half addrspace(1)* %gep.r
|
|
ret void
|
|
}
|
|
|
|
; SI-LABEL: {{^}}div_afn_2_x_pat_f16:
|
|
; SI: v_mul_f32_e32 v{{[0-9]+}}, 0.5, v{{[0-9]+}}
|
|
|
|
; GFX8_9_10: v_mul_f16_e32 [[MUL:v[0-9]+]], 0.5, v{{[0-9]+}}
|
|
; GFX8_9_10: buffer_store_short [[MUL]]
|
|
define amdgpu_kernel void @div_afn_2_x_pat_f16(half addrspace(1)* %out) #0 {
|
|
%x = load half, half addrspace(1)* undef
|
|
%rcp = fdiv afn half %x, 2.0
|
|
store half %rcp, half addrspace(1)* %out, align 4
|
|
ret void
|
|
}
|
|
|
|
; SI-LABEL: {{^}}div_afn_k_x_pat_f16:
|
|
; SI: v_mul_f32_e32 v{{[0-9]+}}, 0x3dcccccd, v{{[0-9]+}}
|
|
|
|
; GFX8_9_10: v_mul_f16_e32 [[MUL:v[0-9]+]], 0x2e66, v{{[0-9]+}}
|
|
; GFX8_9_10: buffer_store_short [[MUL]]
|
|
define amdgpu_kernel void @div_afn_k_x_pat_f16(half addrspace(1)* %out) #0 {
|
|
%x = load half, half addrspace(1)* undef
|
|
%rcp = fdiv afn half %x, 10.0
|
|
store half %rcp, half addrspace(1)* %out, align 4
|
|
ret void
|
|
}
|
|
|
|
; SI-LABEL: {{^}}div_afn_neg_k_x_pat_f16:
|
|
; SI: v_mul_f32_e32 v{{[0-9]+}}, 0xbdcccccd, v{{[0-9]+}}
|
|
|
|
; GFX8_9_10: v_mul_f16_e32 [[MUL:v[0-9]+]], 0xae66, v{{[0-9]+}}
|
|
; GFX8_9_10: buffer_store_short [[MUL]]
|
|
define amdgpu_kernel void @div_afn_neg_k_x_pat_f16(half addrspace(1)* %out) #0 {
|
|
%x = load half, half addrspace(1)* undef
|
|
%rcp = fdiv afn half %x, -10.0
|
|
store half %rcp, half addrspace(1)* %out, align 4
|
|
ret void
|
|
}
|
|
|
|
declare i32 @llvm.amdgcn.workitem.id.x() #2
|
|
declare half @llvm.sqrt.f16(half) #2
|
|
declare half @llvm.fabs.f16(half) #2
|
|
|
|
attributes #0 = { nounwind }
|
|
attributes #1 = { nounwind readnone }
|
|
attributes #2 = { nounwind "unsafe-fp-math"="true" }
|
|
|
|
!0 = !{float 2.500000e+00}
|