1
0
mirror of https://github.com/RPCS3/llvm-mirror.git synced 2024-11-23 11:13:28 +01:00
llvm-mirror/test/CodeGen/AMDGPU/or.ll
Matt Arsenault 2b0231f519 AMDGPU: Add pass to lower kernel arguments to loads
This replaces most argument uses with loads, but for
now not all.

The code in SelectionDAG for calling convention lowering
is actively harmful for amdgpu_kernel. It attempts to
split the argument types into register legal types, which
results in low quality code for arbitary types. Since
all kernel arguments are passed in memory, we just want the
raw types.

I've tried a couple of methods of mitigating this in SelectionDAG,
but it's easier to just bypass this problem alltogether. It's
possible to hack around the problem in the initial lowering,
but the real problem is the DAG then expects to be able to use
CopyToReg/CopyFromReg for uses of the arguments outside the block.

Exposing the argument loads in the IR also has the advantage
that the LoadStoreVectorizer can merge them.

I'm not sure the best approach to dealing with the IR
argument list is. The patch as-is just leaves the IR arguments
in place, so all the existing code will still compute the same
kernarg size and pointlessly lowers the arguments.

Arguably the frontend should emit kernels with an empty argument
list in the first place. Alternatively a dummy array could be
inserted as a single argument just to reserve space.

This does have some disadvantages. Local pointer kernel arguments can
no longer have AssertZext placed  on them as the equivalent !range
metadata is not valid on pointer  typed loads. This is mostly bad
for SI which needs to know about the known bits in order to use the
DS instruction offset, so in this case this is not done.

More importantly, this skips noalias arguments since this pass
does not yet convert this to the equivalent !alias.scope and !noalias
metadata. Producing this metadata correctly seems to be tricky,
although this logically is the same as inlining into a function which
doesn't exist. Additionally, exposing these loads to the vectorizer
may result in degraded aliasing information if a pointer load is
merged with another argument load.

I'm also not entirely sure this is preserving the current clover
ABI, although I would greatly prefer if it would stop widening
arguments and match the HSA ABI. As-is I think it is extending
< 4-byte arguments to 4-bytes but doesn't align them to 4-bytes.

llvm-svn: 335650
2018-06-26 19:10:00 +00:00

273 lines
10 KiB
LLVM

; RUN: llc -amdgpu-scalarize-global-loads=false -march=amdgcn -mcpu=verde -verify-machineinstrs < %s | FileCheck -check-prefix=SI -check-prefix=FUNC %s
; RUN: llc -amdgpu-scalarize-global-loads=false -march=amdgcn -mcpu=tonga -mattr=-flat-for-global -verify-machineinstrs < %s | FileCheck -check-prefix=SI -check-prefix=FUNC %s
; RUN: llc -amdgpu-scalarize-global-loads=false -march=r600 -mcpu=redwood < %s | FileCheck -check-prefix=EG -check-prefix=FUNC %s
; FUNC-LABEL: {{^}}or_v2i32:
; EG: OR_INT {{\*? *}}T{{[0-9]+\.[XYZW], T[0-9]+\.[XYZW], T[0-9]+\.[XYZW]}}
; EG: OR_INT {{\*? *}}T{{[0-9]+\.[XYZW], T[0-9]+\.[XYZW], T[0-9]+\.[XYZW]}}
; SI: v_or_b32_e32 v{{[0-9]+, v[0-9]+, v[0-9]+}}
; SI: v_or_b32_e32 v{{[0-9]+, v[0-9]+, v[0-9]+}}
define amdgpu_kernel void @or_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
%result = or <2 x i32> %a, %b
store <2 x i32> %result, <2 x i32> addrspace(1)* %out
ret void
}
; FUNC-LABEL: {{^}}or_v4i32:
; EG: OR_INT {{\*? *}}T{{[0-9]+\.[XYZW], T[0-9]+\.[XYZW], T[0-9]+\.[XYZW]}}
; EG: OR_INT {{\*? *}}T{{[0-9]+\.[XYZW], T[0-9]+\.[XYZW], T[0-9]+\.[XYZW]}}
; EG: OR_INT {{\*? *}}T{{[0-9]+\.[XYZW], T[0-9]+\.[XYZW], T[0-9]+\.[XYZW]}}
; EG: OR_INT {{\*? *}}T{{[0-9]+\.[XYZW], T[0-9]+\.[XYZW], T[0-9]+\.[XYZW]}}
; SI: v_or_b32_e32 v{{[0-9]+, v[0-9]+, v[0-9]+}}
; SI: v_or_b32_e32 v{{[0-9]+, v[0-9]+, v[0-9]+}}
; SI: v_or_b32_e32 v{{[0-9]+, v[0-9]+, v[0-9]+}}
; SI: v_or_b32_e32 v{{[0-9]+, v[0-9]+, v[0-9]+}}
define amdgpu_kernel void @or_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
%result = or <4 x i32> %a, %b
store <4 x i32> %result, <4 x i32> addrspace(1)* %out
ret void
}
; FUNC-LABEL: {{^}}scalar_or_i32:
; SI: s_or_b32
define amdgpu_kernel void @scalar_or_i32(i32 addrspace(1)* %out, i32 %a, i32 %b) {
%or = or i32 %a, %b
store i32 %or, i32 addrspace(1)* %out
ret void
}
; FUNC-LABEL: {{^}}vector_or_i32:
; SI: v_or_b32_e32 v{{[0-9]}}
define amdgpu_kernel void @vector_or_i32(i32 addrspace(1)* %out, i32 addrspace(1)* %a, i32 %b) {
%loada = load i32, i32 addrspace(1)* %a
%or = or i32 %loada, %b
store i32 %or, i32 addrspace(1)* %out
ret void
}
; FUNC-LABEL: {{^}}scalar_or_literal_i32:
; SI: s_or_b32 s{{[0-9]+}}, s{{[0-9]+}}, 0x1869f
define amdgpu_kernel void @scalar_or_literal_i32(i32 addrspace(1)* %out, i32 %a) {
%or = or i32 %a, 99999
store i32 %or, i32 addrspace(1)* %out, align 4
ret void
}
; FUNC-LABEL: {{^}}scalar_or_literal_i64:
; SI: s_load_dwordx2 s{{\[}}[[LO:[0-9]+]]:[[HI:[0-9]+]]{{\]}}, s{{\[[0-9]+:[0-9]+\]}}, {{0x13|0x4c}}
; SI-DAG: s_or_b32 s[[RES_HI:[0-9]+]], s[[HI]], 0xf237b
; SI-DAG: s_or_b32 s[[RES_LO:[0-9]+]], s[[LO]], 0x3039
; SI-DAG: v_mov_b32_e32 v{{[0-9]+}}, s[[RES_LO]]
; SI-DAG: v_mov_b32_e32 v{{[0-9]+}}, s[[RES_HI]]
define amdgpu_kernel void @scalar_or_literal_i64(i64 addrspace(1)* %out, [8 x i32], i64 %a) {
%or = or i64 %a, 4261135838621753
store i64 %or, i64 addrspace(1)* %out
ret void
}
; FUNC-LABEL: {{^}}scalar_or_literal_multi_use_i64:
; SI: s_load_dwordx2 s{{\[}}[[LO:[0-9]+]]:[[HI:[0-9]+]]{{\]}}, s{{\[[0-9]+:[0-9]+\]}}, {{0x13|0x4c}}
; SI-DAG: s_mov_b32 s[[K_HI:[0-9]+]], 0xf237b
; SI-DAG: s_movk_i32 s[[K_LO:[0-9]+]], 0x3039
; SI: s_or_b64 s{{\[[0-9]+:[0-9]+\]}}, s{{\[[0-9]+:[0-9]+\]}}, s{{\[}}[[K_LO]]:[[K_HI]]{{\]}}
; SI: s_add_u32 s{{[0-9]+}}, s{{[0-9]+}}, s[[K_LO]]
; SI: s_addc_u32 s{{[0-9]+}}, s{{[0-9]+}}, s[[K_HI]]
define amdgpu_kernel void @scalar_or_literal_multi_use_i64(i64 addrspace(1)* %out, [8 x i32], i64 %a, [8 x i32], i64 %b) {
%or = or i64 %a, 4261135838621753
store i64 %or, i64 addrspace(1)* %out
%foo = add i64 %b, 4261135838621753
store volatile i64 %foo, i64 addrspace(1)* undef
ret void
}
; FUNC-LABEL: {{^}}scalar_or_inline_imm_i64:
; SI: s_load_dwordx2 s{{\[}}[[VAL_LO:[0-9]+]]:[[VAL_HI:[0-9]+]]{{\]}}, s{{\[[0-9]+:[0-9]+\]}}, {{0x13|0x4c}}
; SI-NOT: or_b32
; SI: s_or_b32 s[[VAL_LO]], s[[VAL_LO]], 63
; SI-NOT: or_b32
; SI: v_mov_b32_e32 v[[VLO:[0-9]+]], s[[VAL_LO]]
; SI-NOT: or_b32
; SI: v_mov_b32_e32 v[[VHI:[0-9]+]], s[[VAL_HI]]
; SI-NOT: or_b32
; SI: buffer_store_dwordx2 v{{\[}}[[VLO]]:[[VHI]]{{\]}}
define amdgpu_kernel void @scalar_or_inline_imm_i64(i64 addrspace(1)* %out, [8 x i32], i64 %a) {
%or = or i64 %a, 63
store i64 %or, i64 addrspace(1)* %out
ret void
}
; FUNC-LABEL: {{^}}scalar_or_inline_imm_multi_use_i64:
; SI-NOT: or_b32
; SI: s_or_b32 s{{[0-9]+}}, s{{[0-9]+}}, 63
; SI-NOT: or_b32
define amdgpu_kernel void @scalar_or_inline_imm_multi_use_i64(i64 addrspace(1)* %out, i64 %a, i64 %b) {
%or = or i64 %a, 63
store i64 %or, i64 addrspace(1)* %out
%foo = add i64 %b, 63
store volatile i64 %foo, i64 addrspace(1)* undef
ret void
}
; FUNC-LABEL: {{^}}scalar_or_neg_inline_imm_i64:
; SI-DAG: s_load_dword [[VAL:s[0-9]+]]
; SI-DAG: s_or_b32 [[VAL]], [[VAL]], -8
; SI-DAG: v_mov_b32_e32 v[[V_HI:[0-9]+]], -1{{$}}
; SI-DAG: v_mov_b32_e32 v[[V_LO:[0-9]+]], [[VAL]]
; SI: buffer_store_dwordx2 v{{\[}}[[V_LO]]:[[V_HI]]{{\]}}
define amdgpu_kernel void @scalar_or_neg_inline_imm_i64(i64 addrspace(1)* %out, [8 x i32], i64 %a) {
%or = or i64 %a, -8
store i64 %or, i64 addrspace(1)* %out
ret void
}
; FUNC-LABEL: {{^}}vector_or_literal_i32:
; SI: v_or_b32_e32 v{{[0-9]+}}, 0xffff, v{{[0-9]+}}
define amdgpu_kernel void @vector_or_literal_i32(i32 addrspace(1)* %out, i32 addrspace(1)* %a, i32 addrspace(1)* %b) {
%loada = load i32, i32 addrspace(1)* %a, align 4
%or = or i32 %loada, 65535
store i32 %or, i32 addrspace(1)* %out, align 4
ret void
}
; FUNC-LABEL: {{^}}vector_or_inline_immediate_i32:
; SI: v_or_b32_e32 v{{[0-9]+}}, 4, v{{[0-9]+}}
define amdgpu_kernel void @vector_or_inline_immediate_i32(i32 addrspace(1)* %out, i32 addrspace(1)* %a, i32 addrspace(1)* %b) {
%loada = load i32, i32 addrspace(1)* %a, align 4
%or = or i32 %loada, 4
store i32 %or, i32 addrspace(1)* %out, align 4
ret void
}
; FUNC-LABEL: {{^}}scalar_or_i64:
; EG-DAG: OR_INT * T{{[0-9]\.[XYZW]}}, KC0[2].W, KC0[3].Y
; EG-DAG: OR_INT * T{{[0-9]\.[XYZW]}}, KC0[3].X, KC0[3].Z
; SI: s_or_b64
define amdgpu_kernel void @scalar_or_i64(i64 addrspace(1)* %out, i64 %a, i64 %b) {
%or = or i64 %a, %b
store i64 %or, i64 addrspace(1)* %out
ret void
}
; FUNC-LABEL: {{^}}vector_or_i64:
; SI: v_or_b32_e32 v{{[0-9]}}
; SI: v_or_b32_e32 v{{[0-9]}}
define amdgpu_kernel void @vector_or_i64(i64 addrspace(1)* %out, i64 addrspace(1)* %a, i64 addrspace(1)* %b) {
%loada = load i64, i64 addrspace(1)* %a, align 8
%loadb = load i64, i64 addrspace(1)* %b, align 8
%or = or i64 %loada, %loadb
store i64 %or, i64 addrspace(1)* %out
ret void
}
; FUNC-LABEL: {{^}}scalar_vector_or_i64:
; SI: v_or_b32_e32 v{{[0-9]}}
; SI: v_or_b32_e32 v{{[0-9]}}
define amdgpu_kernel void @scalar_vector_or_i64(i64 addrspace(1)* %out, i64 addrspace(1)* %a, i64 %b) {
%loada = load i64, i64 addrspace(1)* %a
%or = or i64 %loada, %b
store i64 %or, i64 addrspace(1)* %out
ret void
}
; FUNC-LABEL: {{^}}vector_or_i64_loadimm:
; SI-DAG: buffer_load_dwordx2 v{{\[}}[[LO_VREG:[0-9]+]]:[[HI_VREG:[0-9]+]]{{\]}},
; SI-DAG: v_or_b32_e32 {{v[0-9]+}}, 0xdf77987f, v[[LO_VREG]]
; SI-DAG: v_or_b32_e32 {{v[0-9]+}}, 0x146f, v[[HI_VREG]]
; SI: s_endpgm
define amdgpu_kernel void @vector_or_i64_loadimm(i64 addrspace(1)* %out, i64 addrspace(1)* %a, i64 addrspace(1)* %b) {
%loada = load i64, i64 addrspace(1)* %a, align 8
%or = or i64 %loada, 22470723082367
store i64 %or, i64 addrspace(1)* %out
ret void
}
; FIXME: The or 0 should really be removed.
; FUNC-LABEL: {{^}}vector_or_i64_imm:
; SI: buffer_load_dwordx2 v{{\[}}[[LO_VREG:[0-9]+]]:[[HI_VREG:[0-9]+]]{{\]}},
; SI: v_or_b32_e32 v[[LO_RESULT:[0-9]+]], 8, v[[LO_VREG]]
; SI-NOT: v_or_b32_e32 {{v[0-9]+}}, 0
; SI: buffer_store_dwordx2 v{{\[}}[[LO_RESULT]]:[[HI_VREG]]{{\]}}
; SI: s_endpgm
define amdgpu_kernel void @vector_or_i64_imm(i64 addrspace(1)* %out, i64 addrspace(1)* %a, i64 addrspace(1)* %b) {
%loada = load i64, i64 addrspace(1)* %a, align 8
%or = or i64 %loada, 8
store i64 %or, i64 addrspace(1)* %out
ret void
}
; FUNC-LABEL: {{^}}vector_or_i64_neg_inline_imm:
; SI-DAG: buffer_load_dword v[[LO_VREG:[0-9]+]]
; SI-DAG: v_or_b32_e32 v[[RES_LO:[0-9]+]], -8, v[[LO_VREG]]
; SI-DAG: v_mov_b32_e32 v[[RES_HI:[0-9]+]], -1{{$}}
; SI: buffer_store_dwordx2 v{{\[}}[[RES_LO]]:[[RES_HI]]{{\]}}
; SI: s_endpgm
define amdgpu_kernel void @vector_or_i64_neg_inline_imm(i64 addrspace(1)* %out, i64 addrspace(1)* %a, i64 addrspace(1)* %b) {
%loada = load i64, i64 addrspace(1)* %a, align 8
%or = or i64 %loada, -8
store i64 %or, i64 addrspace(1)* %out
ret void
}
; FUNC-LABEL: {{^}}vector_or_i64_neg_literal:
; SI-DAG: buffer_load_dword v[[LO_VREG:[0-9]+]]
; SI-DAG: v_mov_b32_e32 v{{[0-9]+}}, -1{{$}}
; SI-DAG: v_or_b32_e32 {{v[0-9]+}}, 0xffffff38, v[[LO_VREG]]
; SI: buffer_store_dwordx2
; SI: s_endpgm
define amdgpu_kernel void @vector_or_i64_neg_literal(i64 addrspace(1)* %out, i64 addrspace(1)* %a, i64 addrspace(1)* %b) {
%loada = load i64, i64 addrspace(1)* %a, align 8
%or = or i64 %loada, -200
store i64 %or, i64 addrspace(1)* %out
ret void
}
; FUNC-LABEL: {{^}}trunc_i64_or_to_i32:
; SI: s_load_dword s[[SREG0:[0-9]+]]
; SI: s_load_dword s[[SREG1:[0-9]+]]
; SI: s_or_b32 s[[SRESULT:[0-9]+]], s[[SREG1]], s[[SREG0]]
; SI: v_mov_b32_e32 [[VRESULT:v[0-9]+]], s[[SRESULT]]
; SI: buffer_store_dword [[VRESULT]],
define amdgpu_kernel void @trunc_i64_or_to_i32(i32 addrspace(1)* %out, [8 x i32], i64 %a, [8 x i32], i64 %b) {
%add = or i64 %b, %a
%trunc = trunc i64 %add to i32
store i32 %trunc, i32 addrspace(1)* %out, align 8
ret void
}
; FUNC-LABEL: {{^}}or_i1:
; EG: OR_INT * {{\** *}}T{{[0-9]+\.[XYZW], PS, PV\.[XYZW]}}
; SI: s_or_b64 s[{{[0-9]+:[0-9]+}}], s[{{[0-9]+:[0-9]+}}], vcc
define amdgpu_kernel void @or_i1(i32 addrspace(1)* %out, float addrspace(1)* %in0, float addrspace(1)* %in1) {
%a = load float, float addrspace(1)* %in0
%b = load float, float addrspace(1)* %in1
%acmp = fcmp oge float %a, 0.000000e+00
%bcmp = fcmp oge float %b, 0.000000e+00
%or = or i1 %acmp, %bcmp
%result = zext i1 %or to i32
store i32 %result, i32 addrspace(1)* %out
ret void
}
; FUNC-LABEL: {{^}}s_or_i1:
; SI: s_or_b64 s[{{[0-9]+:[0-9]+}}], vcc, s[{{[0-9]+:[0-9]+}}]
define amdgpu_kernel void @s_or_i1(i1 addrspace(1)* %out, i32 %a, i32 %b, i32 %c, i32 %d) {
%cmp0 = icmp eq i32 %a, %b
%cmp1 = icmp eq i32 %c, %d
%or = or i1 %cmp0, %cmp1
store i1 %or, i1 addrspace(1)* %out
ret void
}