mirror of
https://github.com/RPCS3/llvm-mirror.git
synced 2024-11-22 02:33:06 +01:00
[AMDGPU] Fix MMO for raw/struct buffer access with non-constant offset
Codegen for the raw/struct buffer access intrinsics would update the offset in the MMO to reflect the combined offset, if it was known to be constant. If the combined offset was not known to be constant, or if there was an index, it would set the offset in the MMO to 0. This is unsafe because it makes it look like the access does not alias with another access with a fixed non-zero offset. Fix these cases by setting the pointer in the MMO to null, to reflect the fact that we do not have any known IR value pointer + constant offset for the access. Differential Revision: https://reviews.llvm.org/D106284
This commit is contained in:
Jay Foad
parent
1cb5f9c5d7
commit
7f0f3d6b7b
@ -6780,20 +6780,25 @@ SDValue SITargetLowering::LowerINTRINSIC_WO_CHAIN(SDValue Op,
|
||||
}
|
||||
}
|
||||
|
||||
/// Update \p MMO based on the offset inputs to an intrinsic. If any of the
|
||||
/// offsets are non-constant or if \p VIndex is non-zero then this function does
|
||||
/// nothing. Otherwise, it sets the MMO offset to the sum of \p VOffset, \p
|
||||
/// SOffset, and \p Offset.
|
||||
/// Update \p MMO based on the offset inputs to an intrinsic.
|
||||
static void updateBufferMMO(MachineMemOperand *MMO, SDValue VOffset,
|
||||
SDValue SOffset, SDValue Offset,
|
||||
SDValue VIndex = SDValue()) {
|
||||
if (!isa<ConstantSDNode>(VOffset) || !isa<ConstantSDNode>(SOffset) ||
|
||||
!isa<ConstantSDNode>(Offset))
|
||||
!isa<ConstantSDNode>(Offset)) {
|
||||
// The combined offset is not known to be constant, so we cannot represent
|
||||
// it in the MMO. Give up.
|
||||
MMO->setValue((Value *)nullptr);
|
||||
return;
|
||||
}
|
||||
|
||||
if (VIndex && (!isa<ConstantSDNode>(VIndex) ||
|
||||
!cast<ConstantSDNode>(VIndex)->isNullValue()))
|
||||
!cast<ConstantSDNode>(VIndex)->isNullValue())) {
|
||||
// The strided index component of the address is not known to be zero, so we
|
||||
// cannot represent it in the MMO. Give up.
|
||||
MMO->setValue((Value *)nullptr);
|
||||
return;
|
||||
}
|
||||
|
||||
MMO->setOffset(cast<ConstantSDNode>(VOffset)->getSExtValue() +
|
||||
cast<ConstantSDNode>(SOffset)->getSExtValue() +
|
||||
|
||||
@ -12,29 +12,29 @@ define amdgpu_cs void @mmo_offsets0(<4 x i32> addrspace(6)* inreg noalias derefe
|
||||
; GCN: [[S_LOAD_DWORDX4_IMM:%[0-9]+]]:sgpr_128 = S_LOAD_DWORDX4_IMM killed [[REG_SEQUENCE]], 0, 0 :: (dereferenceable invariant load (s128) from %ir.arg0, addrspace 6)
|
||||
; GCN: [[BUFFER_LOAD_DWORDX4_OFFSET:%[0-9]+]]:vreg_128 = BUFFER_LOAD_DWORDX4_OFFSET [[S_LOAD_DWORDX4_IMM]], [[S_MOV_B32_]], 16, 0, 0, 0, implicit $exec :: (dereferenceable load (s128) from custom "BufferResource" + 16, align 1, addrspace 4)
|
||||
; GCN: [[V_MOV_B32_e32_:%[0-9]+]]:vgpr_32 = V_MOV_B32_e32 1, implicit $exec
|
||||
; GCN: [[BUFFER_LOAD_DWORDX4_IDXEN:%[0-9]+]]:vreg_128 = BUFFER_LOAD_DWORDX4_IDXEN [[V_MOV_B32_e32_]], [[S_LOAD_DWORDX4_IMM]], [[S_MOV_B32_]], 16, 0, 0, 0, implicit $exec :: (dereferenceable load (s128) from custom "BufferResource", align 1, addrspace 4)
|
||||
; GCN: [[BUFFER_LOAD_DWORDX4_OFFEN:%[0-9]+]]:vreg_128 = BUFFER_LOAD_DWORDX4_OFFEN [[COPY]], [[S_LOAD_DWORDX4_IMM]], [[S_MOV_B32_]], 0, 0, 0, 0, implicit $exec :: (dereferenceable load (s128) from custom "BufferResource", align 1, addrspace 4)
|
||||
; GCN: [[BUFFER_LOAD_DWORDX4_IDXEN1:%[0-9]+]]:vreg_128 = BUFFER_LOAD_DWORDX4_IDXEN [[COPY]], [[S_LOAD_DWORDX4_IMM]], [[S_MOV_B32_]], 16, 0, 0, 0, implicit $exec :: (dereferenceable load (s128) from custom "BufferResource", align 1, addrspace 4)
|
||||
; GCN: [[BUFFER_LOAD_DWORDX4_IDXEN:%[0-9]+]]:vreg_128 = BUFFER_LOAD_DWORDX4_IDXEN [[V_MOV_B32_e32_]], [[S_LOAD_DWORDX4_IMM]], [[S_MOV_B32_]], 16, 0, 0, 0, implicit $exec :: (dereferenceable load (s128), align 1, addrspace 4)
|
||||
; GCN: [[BUFFER_LOAD_DWORDX4_OFFEN:%[0-9]+]]:vreg_128 = BUFFER_LOAD_DWORDX4_OFFEN [[COPY]], [[S_LOAD_DWORDX4_IMM]], [[S_MOV_B32_]], 0, 0, 0, 0, implicit $exec :: (dereferenceable load (s128), align 1, addrspace 4)
|
||||
; GCN: [[BUFFER_LOAD_DWORDX4_IDXEN1:%[0-9]+]]:vreg_128 = BUFFER_LOAD_DWORDX4_IDXEN [[COPY]], [[S_LOAD_DWORDX4_IMM]], [[S_MOV_B32_]], 16, 0, 0, 0, implicit $exec :: (dereferenceable load (s128), align 1, addrspace 4)
|
||||
; GCN: INLINEASM &"", 1 /* sideeffect attdialect */
|
||||
; GCN: BUFFER_STORE_DWORDX4_OFFSET_exact killed [[BUFFER_LOAD_DWORDX4_OFFSET]], [[S_LOAD_DWORDX4_IMM]], [[S_MOV_B32_]], 32, 0, 0, 0, implicit $exec :: (dereferenceable store (s128) into custom "BufferResource" + 32, align 1, addrspace 4)
|
||||
; GCN: BUFFER_STORE_DWORDX4_OFFEN_exact killed [[BUFFER_LOAD_DWORDX4_OFFEN]], [[COPY]], [[S_LOAD_DWORDX4_IMM]], [[S_MOV_B32_]], 0, 0, 0, 0, implicit $exec :: (dereferenceable store (s128) into custom "BufferResource", align 1, addrspace 4)
|
||||
; GCN: BUFFER_STORE_DWORDX4_IDXEN_exact killed [[BUFFER_LOAD_DWORDX4_IDXEN]], [[V_MOV_B32_e32_]], [[S_LOAD_DWORDX4_IMM]], [[S_MOV_B32_]], 32, 0, 0, 0, implicit $exec :: (dereferenceable store (s128) into custom "BufferResource", align 1, addrspace 4)
|
||||
; GCN: BUFFER_STORE_DWORDX4_IDXEN_exact killed [[BUFFER_LOAD_DWORDX4_IDXEN1]], [[COPY]], [[S_LOAD_DWORDX4_IMM]], [[S_MOV_B32_]], 32, 0, 0, 0, implicit $exec :: (dereferenceable store (s128) into custom "BufferResource", align 1, addrspace 4)
|
||||
; GCN: BUFFER_STORE_DWORDX4_OFFEN_exact killed [[BUFFER_LOAD_DWORDX4_OFFEN]], [[COPY]], [[S_LOAD_DWORDX4_IMM]], [[S_MOV_B32_]], 0, 0, 0, 0, implicit $exec :: (dereferenceable store (s128), align 1, addrspace 4)
|
||||
; GCN: BUFFER_STORE_DWORDX4_IDXEN_exact killed [[BUFFER_LOAD_DWORDX4_IDXEN]], [[V_MOV_B32_e32_]], [[S_LOAD_DWORDX4_IMM]], [[S_MOV_B32_]], 32, 0, 0, 0, implicit $exec :: (dereferenceable store (s128), align 1, addrspace 4)
|
||||
; GCN: BUFFER_STORE_DWORDX4_IDXEN_exact killed [[BUFFER_LOAD_DWORDX4_IDXEN1]], [[COPY]], [[S_LOAD_DWORDX4_IMM]], [[S_MOV_B32_]], 32, 0, 0, 0, implicit $exec :: (dereferenceable store (s128), align 1, addrspace 4)
|
||||
; GCN: INLINEASM &"", 1 /* sideeffect attdialect */
|
||||
; GCN: [[BUFFER_LOAD_FORMAT_XYZW_OFFSET:%[0-9]+]]:vreg_128 = BUFFER_LOAD_FORMAT_XYZW_OFFSET [[S_LOAD_DWORDX4_IMM]], [[S_MOV_B32_]], 48, 0, 0, 0, implicit $exec :: (dereferenceable load (s128) from custom "BufferResource" + 48, align 1, addrspace 4)
|
||||
; GCN: [[BUFFER_LOAD_FORMAT_XYZW_IDXEN:%[0-9]+]]:vreg_128 = BUFFER_LOAD_FORMAT_XYZW_IDXEN [[V_MOV_B32_e32_]], [[S_LOAD_DWORDX4_IMM]], [[S_MOV_B32_]], 48, 0, 0, 0, implicit $exec :: (dereferenceable load (s128) from custom "BufferResource", align 1, addrspace 4)
|
||||
; GCN: [[BUFFER_LOAD_FORMAT_XYZW_OFFEN:%[0-9]+]]:vreg_128 = BUFFER_LOAD_FORMAT_XYZW_OFFEN [[COPY]], [[S_LOAD_DWORDX4_IMM]], [[S_MOV_B32_]], 0, 0, 0, 0, implicit $exec :: (dereferenceable load (s128) from custom "BufferResource", align 1, addrspace 4)
|
||||
; GCN: [[BUFFER_LOAD_FORMAT_XYZW_IDXEN1:%[0-9]+]]:vreg_128 = BUFFER_LOAD_FORMAT_XYZW_IDXEN [[COPY]], [[S_LOAD_DWORDX4_IMM]], [[S_MOV_B32_]], 48, 0, 0, 0, implicit $exec :: (dereferenceable load (s128) from custom "BufferResource", align 1, addrspace 4)
|
||||
; GCN: [[BUFFER_LOAD_FORMAT_XYZW_IDXEN:%[0-9]+]]:vreg_128 = BUFFER_LOAD_FORMAT_XYZW_IDXEN [[V_MOV_B32_e32_]], [[S_LOAD_DWORDX4_IMM]], [[S_MOV_B32_]], 48, 0, 0, 0, implicit $exec :: (dereferenceable load (s128), align 1, addrspace 4)
|
||||
; GCN: [[BUFFER_LOAD_FORMAT_XYZW_OFFEN:%[0-9]+]]:vreg_128 = BUFFER_LOAD_FORMAT_XYZW_OFFEN [[COPY]], [[S_LOAD_DWORDX4_IMM]], [[S_MOV_B32_]], 0, 0, 0, 0, implicit $exec :: (dereferenceable load (s128), align 1, addrspace 4)
|
||||
; GCN: [[BUFFER_LOAD_FORMAT_XYZW_IDXEN1:%[0-9]+]]:vreg_128 = BUFFER_LOAD_FORMAT_XYZW_IDXEN [[COPY]], [[S_LOAD_DWORDX4_IMM]], [[S_MOV_B32_]], 48, 0, 0, 0, implicit $exec :: (dereferenceable load (s128), align 1, addrspace 4)
|
||||
; GCN: INLINEASM &"", 1 /* sideeffect attdialect */
|
||||
; GCN: BUFFER_STORE_FORMAT_XYZW_OFFSET_exact killed [[BUFFER_LOAD_FORMAT_XYZW_OFFSET]], [[S_LOAD_DWORDX4_IMM]], [[S_MOV_B32_]], 64, 0, 0, 0, implicit $exec :: (dereferenceable store (s128) into custom "BufferResource" + 64, align 1, addrspace 4)
|
||||
; GCN: BUFFER_STORE_FORMAT_XYZW_OFFEN_exact killed [[BUFFER_LOAD_FORMAT_XYZW_OFFEN]], [[COPY]], [[S_LOAD_DWORDX4_IMM]], [[S_MOV_B32_]], 0, 0, 0, 0, implicit $exec :: (dereferenceable store (s128) into custom "BufferResource", align 1, addrspace 4)
|
||||
; GCN: BUFFER_STORE_FORMAT_XYZW_IDXEN_exact killed [[BUFFER_LOAD_FORMAT_XYZW_IDXEN]], [[V_MOV_B32_e32_]], [[S_LOAD_DWORDX4_IMM]], [[S_MOV_B32_]], 64, 0, 0, 0, implicit $exec :: (dereferenceable store (s128) into custom "BufferResource", align 1, addrspace 4)
|
||||
; GCN: BUFFER_STORE_FORMAT_XYZW_IDXEN_exact killed [[BUFFER_LOAD_FORMAT_XYZW_IDXEN1]], [[COPY]], [[S_LOAD_DWORDX4_IMM]], [[S_MOV_B32_]], 64, 0, 0, 0, implicit $exec :: (dereferenceable store (s128) into custom "BufferResource", align 1, addrspace 4)
|
||||
; GCN: BUFFER_STORE_FORMAT_XYZW_OFFEN_exact killed [[BUFFER_LOAD_FORMAT_XYZW_OFFEN]], [[COPY]], [[S_LOAD_DWORDX4_IMM]], [[S_MOV_B32_]], 0, 0, 0, 0, implicit $exec :: (dereferenceable store (s128), align 1, addrspace 4)
|
||||
; GCN: BUFFER_STORE_FORMAT_XYZW_IDXEN_exact killed [[BUFFER_LOAD_FORMAT_XYZW_IDXEN]], [[V_MOV_B32_e32_]], [[S_LOAD_DWORDX4_IMM]], [[S_MOV_B32_]], 64, 0, 0, 0, implicit $exec :: (dereferenceable store (s128), align 1, addrspace 4)
|
||||
; GCN: BUFFER_STORE_FORMAT_XYZW_IDXEN_exact killed [[BUFFER_LOAD_FORMAT_XYZW_IDXEN1]], [[COPY]], [[S_LOAD_DWORDX4_IMM]], [[S_MOV_B32_]], 64, 0, 0, 0, implicit $exec :: (dereferenceable store (s128), align 1, addrspace 4)
|
||||
; GCN: INLINEASM &"", 1 /* sideeffect attdialect */
|
||||
; GCN: BUFFER_ATOMIC_ADD_OFFSET [[COPY]], [[S_LOAD_DWORDX4_IMM]], [[S_MOV_B32_]], 80, 0, implicit $exec :: (volatile dereferenceable load store (s32) on custom "BufferResource" + 80, align 1, addrspace 4)
|
||||
; GCN: BUFFER_ATOMIC_ADD_OFFEN [[COPY]], [[COPY]], [[S_LOAD_DWORDX4_IMM]], [[S_MOV_B32_]], 0, 0, implicit $exec :: (volatile dereferenceable load store (s32) on custom "BufferResource", align 1, addrspace 4)
|
||||
; GCN: BUFFER_ATOMIC_ADD_IDXEN [[COPY]], [[V_MOV_B32_e32_]], [[S_LOAD_DWORDX4_IMM]], [[S_MOV_B32_]], 80, 0, implicit $exec :: (volatile dereferenceable load store (s32) on custom "BufferResource", align 1, addrspace 4)
|
||||
; GCN: BUFFER_ATOMIC_ADD_IDXEN [[COPY]], [[COPY]], [[S_LOAD_DWORDX4_IMM]], [[S_MOV_B32_]], 80, 0, implicit $exec :: (volatile dereferenceable load store (s32) on custom "BufferResource", align 1, addrspace 4)
|
||||
; GCN: BUFFER_ATOMIC_ADD_OFFEN [[COPY]], [[COPY]], [[S_LOAD_DWORDX4_IMM]], [[S_MOV_B32_]], 0, 0, implicit $exec :: (volatile dereferenceable load store (s32), align 1, addrspace 4)
|
||||
; GCN: BUFFER_ATOMIC_ADD_IDXEN [[COPY]], [[V_MOV_B32_e32_]], [[S_LOAD_DWORDX4_IMM]], [[S_MOV_B32_]], 80, 0, implicit $exec :: (volatile dereferenceable load store (s32), align 1, addrspace 4)
|
||||
; GCN: BUFFER_ATOMIC_ADD_IDXEN [[COPY]], [[COPY]], [[S_LOAD_DWORDX4_IMM]], [[S_MOV_B32_]], 80, 0, implicit $exec :: (volatile dereferenceable load store (s32), align 1, addrspace 4)
|
||||
; GCN: INLINEASM &"", 1 /* sideeffect attdialect */
|
||||
; GCN: [[REG_SEQUENCE1:%[0-9]+]]:vreg_64 = REG_SEQUENCE [[COPY]], %subreg.sub0, [[COPY]], %subreg.sub1
|
||||
; GCN: [[DEF:%[0-9]+]]:vreg_64 = IMPLICIT_DEF
|
||||
@ -42,50 +42,50 @@ define amdgpu_cs void @mmo_offsets0(<4 x i32> addrspace(6)* inreg noalias derefe
|
||||
; GCN: [[COPY2:%[0-9]+]]:vreg_64 = COPY [[DEF]]
|
||||
; GCN: [[COPY3:%[0-9]+]]:vgpr_32 = COPY [[COPY2]].sub0
|
||||
; GCN: [[DEF1:%[0-9]+]]:vreg_64 = IMPLICIT_DEF
|
||||
; GCN: BUFFER_ATOMIC_CMPSWAP_OFFEN [[REG_SEQUENCE1]], [[COPY]], [[S_LOAD_DWORDX4_IMM]], [[S_MOV_B32_]], 0, 1, implicit $exec :: (volatile dereferenceable load store (s32) on custom "BufferResource", align 1, addrspace 4)
|
||||
; GCN: BUFFER_ATOMIC_CMPSWAP_OFFEN [[REG_SEQUENCE1]], [[COPY]], [[S_LOAD_DWORDX4_IMM]], [[S_MOV_B32_]], 0, 1, implicit $exec :: (volatile dereferenceable load store (s32), align 1, addrspace 4)
|
||||
; GCN: [[COPY4:%[0-9]+]]:vreg_64 = COPY [[DEF1]]
|
||||
; GCN: [[COPY5:%[0-9]+]]:vgpr_32 = COPY [[COPY4]].sub0
|
||||
; GCN: [[DEF2:%[0-9]+]]:vreg_64 = IMPLICIT_DEF
|
||||
; GCN: BUFFER_ATOMIC_CMPSWAP_IDXEN [[REG_SEQUENCE1]], [[V_MOV_B32_e32_]], [[S_LOAD_DWORDX4_IMM]], [[S_MOV_B32_]], 96, 1, implicit $exec :: (volatile dereferenceable load store (s32) on custom "BufferResource", align 1, addrspace 4)
|
||||
; GCN: BUFFER_ATOMIC_CMPSWAP_IDXEN [[REG_SEQUENCE1]], [[V_MOV_B32_e32_]], [[S_LOAD_DWORDX4_IMM]], [[S_MOV_B32_]], 96, 1, implicit $exec :: (volatile dereferenceable load store (s32), align 1, addrspace 4)
|
||||
; GCN: [[COPY6:%[0-9]+]]:vreg_64 = COPY [[DEF2]]
|
||||
; GCN: [[COPY7:%[0-9]+]]:vgpr_32 = COPY [[COPY6]].sub0
|
||||
; GCN: [[DEF3:%[0-9]+]]:vreg_64 = IMPLICIT_DEF
|
||||
; GCN: BUFFER_ATOMIC_CMPSWAP_IDXEN [[REG_SEQUENCE1]], [[COPY]], [[S_LOAD_DWORDX4_IMM]], [[S_MOV_B32_]], 96, 1, implicit $exec :: (volatile dereferenceable load store (s32) on custom "BufferResource", align 1, addrspace 4)
|
||||
; GCN: BUFFER_ATOMIC_CMPSWAP_IDXEN [[REG_SEQUENCE1]], [[COPY]], [[S_LOAD_DWORDX4_IMM]], [[S_MOV_B32_]], 96, 1, implicit $exec :: (volatile dereferenceable load store (s32), align 1, addrspace 4)
|
||||
; GCN: [[COPY8:%[0-9]+]]:vreg_64 = COPY [[DEF3]]
|
||||
; GCN: [[COPY9:%[0-9]+]]:vgpr_32 = COPY [[COPY8]].sub0
|
||||
; GCN: INLINEASM &"", 1 /* sideeffect attdialect */
|
||||
; GCN: [[V_MOV_B32_e32_1:%[0-9]+]]:vgpr_32 = V_MOV_B32_e32 1065353216, implicit $exec
|
||||
; GCN: BUFFER_ATOMIC_ADD_F32_OFFSET [[V_MOV_B32_e32_1]], [[S_LOAD_DWORDX4_IMM]], [[S_MOV_B32_]], 112, 0, implicit $exec :: (volatile dereferenceable load store (s32) on custom "BufferResource" + 112, align 1, addrspace 4)
|
||||
; GCN: BUFFER_ATOMIC_ADD_F32_OFFEN [[V_MOV_B32_e32_1]], [[COPY]], [[S_LOAD_DWORDX4_IMM]], [[S_MOV_B32_]], 0, 0, implicit $exec :: (volatile dereferenceable load store (s32) on custom "BufferResource", align 1, addrspace 4)
|
||||
; GCN: BUFFER_ATOMIC_ADD_F32_IDXEN [[V_MOV_B32_e32_1]], [[V_MOV_B32_e32_]], [[S_LOAD_DWORDX4_IMM]], [[S_MOV_B32_]], 112, 0, implicit $exec :: (volatile dereferenceable load store (s32) on custom "BufferResource", align 1, addrspace 4)
|
||||
; GCN: BUFFER_ATOMIC_ADD_F32_IDXEN [[V_MOV_B32_e32_1]], [[COPY]], [[S_LOAD_DWORDX4_IMM]], [[S_MOV_B32_]], 112, 0, implicit $exec :: (volatile dereferenceable load store (s32) on custom "BufferResource", align 1, addrspace 4)
|
||||
; GCN: BUFFER_ATOMIC_ADD_F32_OFFEN [[V_MOV_B32_e32_1]], [[COPY]], [[S_LOAD_DWORDX4_IMM]], [[S_MOV_B32_]], 0, 0, implicit $exec :: (volatile dereferenceable load store (s32), align 1, addrspace 4)
|
||||
; GCN: BUFFER_ATOMIC_ADD_F32_IDXEN [[V_MOV_B32_e32_1]], [[V_MOV_B32_e32_]], [[S_LOAD_DWORDX4_IMM]], [[S_MOV_B32_]], 112, 0, implicit $exec :: (volatile dereferenceable load store (s32), align 1, addrspace 4)
|
||||
; GCN: BUFFER_ATOMIC_ADD_F32_IDXEN [[V_MOV_B32_e32_1]], [[COPY]], [[S_LOAD_DWORDX4_IMM]], [[S_MOV_B32_]], 112, 0, implicit $exec :: (volatile dereferenceable load store (s32), align 1, addrspace 4)
|
||||
; GCN: INLINEASM &"", 1 /* sideeffect attdialect */
|
||||
; GCN: [[BUFFER_LOAD_DWORDX4_OFFSET1:%[0-9]+]]:vreg_128 = BUFFER_LOAD_DWORDX4_OFFSET [[S_LOAD_DWORDX4_IMM]], [[S_MOV_B32_]], 128, 0, 0, 0, implicit $exec :: (dereferenceable load (s128) from custom "BufferResource" + 128, align 1, addrspace 4)
|
||||
; GCN: [[S_MOV_B32_1:%[0-9]+]]:sreg_32 = S_MOV_B32 64
|
||||
; GCN: [[BUFFER_LOAD_DWORDX4_OFFSET2:%[0-9]+]]:vreg_128 = BUFFER_LOAD_DWORDX4_OFFSET [[S_LOAD_DWORDX4_IMM]], killed [[S_MOV_B32_1]], 64, 0, 0, 0, implicit $exec :: (dereferenceable load (s128) from custom "BufferResource" + 128, align 1, addrspace 4)
|
||||
; GCN: [[S_MOV_B32_2:%[0-9]+]]:sreg_32 = S_MOV_B32 128
|
||||
; GCN: [[BUFFER_LOAD_DWORDX4_OFFSET3:%[0-9]+]]:vreg_128 = BUFFER_LOAD_DWORDX4_OFFSET [[S_LOAD_DWORDX4_IMM]], [[S_MOV_B32_2]], 0, 0, 0, 0, implicit $exec :: (dereferenceable load (s128) from custom "BufferResource" + 128, align 1, addrspace 4)
|
||||
; GCN: [[BUFFER_LOAD_DWORDX4_OFFEN1:%[0-9]+]]:vreg_128 = BUFFER_LOAD_DWORDX4_OFFEN [[COPY]], [[S_LOAD_DWORDX4_IMM]], [[S_MOV_B32_2]], 0, 0, 0, 0, implicit $exec :: (dereferenceable load (s128) from custom "BufferResource", align 1, addrspace 4)
|
||||
; GCN: [[BUFFER_LOAD_DWORDX4_OFFEN1:%[0-9]+]]:vreg_128 = BUFFER_LOAD_DWORDX4_OFFEN [[COPY]], [[S_LOAD_DWORDX4_IMM]], [[S_MOV_B32_2]], 0, 0, 0, 0, implicit $exec :: (dereferenceable load (s128), align 1, addrspace 4)
|
||||
; GCN: [[COPY10:%[0-9]+]]:sreg_32 = COPY [[COPY]]
|
||||
; GCN: [[BUFFER_LOAD_DWORDX4_OFFSET4:%[0-9]+]]:vreg_128 = BUFFER_LOAD_DWORDX4_OFFSET [[S_LOAD_DWORDX4_IMM]], [[COPY10]], 128, 0, 0, 0, implicit $exec :: (dereferenceable load (s128) from custom "BufferResource", align 1, addrspace 4)
|
||||
; GCN: [[BUFFER_LOAD_DWORDX4_OFFSET4:%[0-9]+]]:vreg_128 = BUFFER_LOAD_DWORDX4_OFFSET [[S_LOAD_DWORDX4_IMM]], [[COPY10]], 128, 0, 0, 0, implicit $exec :: (dereferenceable load (s128), align 1, addrspace 4)
|
||||
; GCN: INLINEASM &"", 1 /* sideeffect attdialect */
|
||||
; GCN: [[BUFFER_LOAD_FORMAT_XYZW_OFFSET1:%[0-9]+]]:vreg_128 = BUFFER_LOAD_FORMAT_XYZW_OFFSET [[S_LOAD_DWORDX4_IMM]], [[S_MOV_B32_]], 144, 0, 0, 0, implicit $exec :: (dereferenceable load (s128) from custom "BufferResource" + 144, align 1, addrspace 4)
|
||||
; GCN: [[S_MOV_B32_3:%[0-9]+]]:sreg_32 = S_MOV_B32 72
|
||||
; GCN: [[BUFFER_LOAD_FORMAT_XYZW_OFFSET2:%[0-9]+]]:vreg_128 = BUFFER_LOAD_FORMAT_XYZW_OFFSET [[S_LOAD_DWORDX4_IMM]], killed [[S_MOV_B32_3]], 72, 0, 0, 0, implicit $exec :: (dereferenceable load (s128) from custom "BufferResource" + 144, align 1, addrspace 4)
|
||||
; GCN: [[S_MOV_B32_4:%[0-9]+]]:sreg_32 = S_MOV_B32 144
|
||||
; GCN: [[BUFFER_LOAD_FORMAT_XYZW_OFFSET3:%[0-9]+]]:vreg_128 = BUFFER_LOAD_FORMAT_XYZW_OFFSET [[S_LOAD_DWORDX4_IMM]], [[S_MOV_B32_4]], 0, 0, 0, 0, implicit $exec :: (dereferenceable load (s128) from custom "BufferResource" + 144, align 1, addrspace 4)
|
||||
; GCN: [[BUFFER_LOAD_FORMAT_XYZW_OFFEN1:%[0-9]+]]:vreg_128 = BUFFER_LOAD_FORMAT_XYZW_OFFEN [[COPY]], [[S_LOAD_DWORDX4_IMM]], [[S_MOV_B32_4]], 0, 0, 0, 0, implicit $exec :: (dereferenceable load (s128) from custom "BufferResource", align 1, addrspace 4)
|
||||
; GCN: [[BUFFER_LOAD_FORMAT_XYZW_OFFEN1:%[0-9]+]]:vreg_128 = BUFFER_LOAD_FORMAT_XYZW_OFFEN [[COPY]], [[S_LOAD_DWORDX4_IMM]], [[S_MOV_B32_4]], 0, 0, 0, 0, implicit $exec :: (dereferenceable load (s128), align 1, addrspace 4)
|
||||
; GCN: [[COPY11:%[0-9]+]]:sreg_32 = COPY [[COPY]]
|
||||
; GCN: [[BUFFER_LOAD_FORMAT_XYZW_OFFSET4:%[0-9]+]]:vreg_128 = BUFFER_LOAD_FORMAT_XYZW_OFFSET [[S_LOAD_DWORDX4_IMM]], [[COPY11]], 144, 0, 0, 0, implicit $exec :: (dereferenceable load (s128) from custom "BufferResource", align 1, addrspace 4)
|
||||
; GCN: [[BUFFER_LOAD_FORMAT_XYZW_OFFSET4:%[0-9]+]]:vreg_128 = BUFFER_LOAD_FORMAT_XYZW_OFFSET [[S_LOAD_DWORDX4_IMM]], [[COPY11]], 144, 0, 0, 0, implicit $exec :: (dereferenceable load (s128), align 1, addrspace 4)
|
||||
; GCN: INLINEASM &"", 1 /* sideeffect attdialect */
|
||||
; GCN: BUFFER_ATOMIC_ADD_OFFSET [[COPY]], [[S_LOAD_DWORDX4_IMM]], [[S_MOV_B32_]], 160, 0, implicit $exec :: (volatile dereferenceable load store (s32) on custom "BufferResource" + 160, align 1, addrspace 4)
|
||||
; GCN: [[S_MOV_B32_5:%[0-9]+]]:sreg_32 = S_MOV_B32 80
|
||||
; GCN: BUFFER_ATOMIC_ADD_OFFSET [[COPY]], [[S_LOAD_DWORDX4_IMM]], killed [[S_MOV_B32_5]], 80, 0, implicit $exec :: (volatile dereferenceable load store (s32) on custom "BufferResource" + 160, align 1, addrspace 4)
|
||||
; GCN: [[S_MOV_B32_6:%[0-9]+]]:sreg_32 = S_MOV_B32 160
|
||||
; GCN: BUFFER_ATOMIC_ADD_OFFSET [[COPY]], [[S_LOAD_DWORDX4_IMM]], [[S_MOV_B32_6]], 0, 0, implicit $exec :: (volatile dereferenceable load store (s32) on custom "BufferResource" + 160, align 1, addrspace 4)
|
||||
; GCN: BUFFER_ATOMIC_ADD_OFFEN [[COPY]], [[COPY]], [[S_LOAD_DWORDX4_IMM]], [[S_MOV_B32_6]], 0, 0, implicit $exec :: (volatile dereferenceable load store (s32) on custom "BufferResource", align 1, addrspace 4)
|
||||
; GCN: BUFFER_ATOMIC_ADD_OFFEN [[COPY]], [[COPY]], [[S_LOAD_DWORDX4_IMM]], [[S_MOV_B32_6]], 0, 0, implicit $exec :: (volatile dereferenceable load store (s32), align 1, addrspace 4)
|
||||
; GCN: [[COPY12:%[0-9]+]]:sreg_32 = COPY [[COPY]]
|
||||
; GCN: BUFFER_ATOMIC_ADD_OFFSET [[COPY]], [[S_LOAD_DWORDX4_IMM]], [[COPY12]], 160, 0, implicit $exec :: (volatile dereferenceable load store (s32) on custom "BufferResource", align 1, addrspace 4)
|
||||
; GCN: BUFFER_ATOMIC_ADD_OFFSET [[COPY]], [[S_LOAD_DWORDX4_IMM]], [[COPY12]], 160, 0, implicit $exec :: (volatile dereferenceable load store (s32), align 1, addrspace 4)
|
||||
; GCN: INLINEASM &"", 1 /* sideeffect attdialect */
|
||||
; GCN: [[DEF4:%[0-9]+]]:vreg_64 = IMPLICIT_DEF
|
||||
; GCN: BUFFER_ATOMIC_CMPSWAP_OFFSET [[REG_SEQUENCE1]], [[S_LOAD_DWORDX4_IMM]], [[S_MOV_B32_]], 176, 1, implicit $exec :: (volatile dereferenceable load store (s32) on custom "BufferResource" + 176, align 1, addrspace 4)
|
||||
@ -102,12 +102,12 @@ define amdgpu_cs void @mmo_offsets0(<4 x i32> addrspace(6)* inreg noalias derefe
|
||||
; GCN: [[COPY17:%[0-9]+]]:vreg_64 = COPY [[DEF6]]
|
||||
; GCN: [[COPY18:%[0-9]+]]:vgpr_32 = COPY [[COPY17]].sub0
|
||||
; GCN: [[DEF7:%[0-9]+]]:vreg_64 = IMPLICIT_DEF
|
||||
; GCN: BUFFER_ATOMIC_CMPSWAP_OFFEN [[REG_SEQUENCE1]], [[COPY]], [[S_LOAD_DWORDX4_IMM]], [[S_MOV_B32_8]], 0, 1, implicit $exec :: (volatile dereferenceable load store (s32) on custom "BufferResource", align 1, addrspace 4)
|
||||
; GCN: BUFFER_ATOMIC_CMPSWAP_OFFEN [[REG_SEQUENCE1]], [[COPY]], [[S_LOAD_DWORDX4_IMM]], [[S_MOV_B32_8]], 0, 1, implicit $exec :: (volatile dereferenceable load store (s32), align 1, addrspace 4)
|
||||
; GCN: [[COPY19:%[0-9]+]]:vreg_64 = COPY [[DEF7]]
|
||||
; GCN: [[COPY20:%[0-9]+]]:vgpr_32 = COPY [[COPY19]].sub0
|
||||
; GCN: [[COPY21:%[0-9]+]]:sreg_32 = COPY [[COPY]]
|
||||
; GCN: [[DEF8:%[0-9]+]]:vreg_64 = IMPLICIT_DEF
|
||||
; GCN: BUFFER_ATOMIC_CMPSWAP_OFFSET [[REG_SEQUENCE1]], [[S_LOAD_DWORDX4_IMM]], [[COPY21]], 176, 1, implicit $exec :: (volatile dereferenceable load store (s32) on custom "BufferResource", align 1, addrspace 4)
|
||||
; GCN: BUFFER_ATOMIC_CMPSWAP_OFFSET [[REG_SEQUENCE1]], [[S_LOAD_DWORDX4_IMM]], [[COPY21]], 176, 1, implicit $exec :: (volatile dereferenceable load store (s32), align 1, addrspace 4)
|
||||
; GCN: [[COPY22:%[0-9]+]]:vreg_64 = COPY [[DEF8]]
|
||||
; GCN: [[COPY23:%[0-9]+]]:vgpr_32 = COPY [[COPY22]].sub0
|
||||
; GCN: INLINEASM &"", 1 /* sideeffect attdialect */
|
||||
@ -116,18 +116,18 @@ define amdgpu_cs void @mmo_offsets0(<4 x i32> addrspace(6)* inreg noalias derefe
|
||||
; GCN: BUFFER_STORE_DWORDX4_OFFSET_exact killed [[BUFFER_LOAD_DWORDX4_OFFSET2]], [[S_LOAD_DWORDX4_IMM]], killed [[S_MOV_B32_9]], 96, 0, 0, 0, implicit $exec :: (dereferenceable store (s128) into custom "BufferResource" + 192, align 1, addrspace 4)
|
||||
; GCN: [[S_MOV_B32_10:%[0-9]+]]:sreg_32 = S_MOV_B32 192
|
||||
; GCN: BUFFER_STORE_DWORDX4_OFFSET_exact killed [[BUFFER_LOAD_DWORDX4_OFFSET3]], [[S_LOAD_DWORDX4_IMM]], [[S_MOV_B32_10]], 0, 0, 0, 0, implicit $exec :: (dereferenceable store (s128) into custom "BufferResource" + 192, align 1, addrspace 4)
|
||||
; GCN: BUFFER_STORE_DWORDX4_OFFEN_exact killed [[BUFFER_LOAD_DWORDX4_OFFEN1]], [[COPY]], [[S_LOAD_DWORDX4_IMM]], [[S_MOV_B32_10]], 0, 0, 0, 0, implicit $exec :: (dereferenceable store (s128) into custom "BufferResource", align 1, addrspace 4)
|
||||
; GCN: BUFFER_STORE_DWORDX4_OFFEN_exact killed [[BUFFER_LOAD_DWORDX4_OFFEN1]], [[COPY]], [[S_LOAD_DWORDX4_IMM]], [[S_MOV_B32_10]], 0, 0, 0, 0, implicit $exec :: (dereferenceable store (s128), align 1, addrspace 4)
|
||||
; GCN: [[COPY24:%[0-9]+]]:sreg_32 = COPY [[COPY]]
|
||||
; GCN: BUFFER_STORE_DWORDX4_OFFSET_exact killed [[BUFFER_LOAD_DWORDX4_OFFSET4]], [[S_LOAD_DWORDX4_IMM]], [[COPY24]], 192, 0, 0, 0, implicit $exec :: (dereferenceable store (s128) into custom "BufferResource", align 1, addrspace 4)
|
||||
; GCN: BUFFER_STORE_DWORDX4_OFFSET_exact killed [[BUFFER_LOAD_DWORDX4_OFFSET4]], [[S_LOAD_DWORDX4_IMM]], [[COPY24]], 192, 0, 0, 0, implicit $exec :: (dereferenceable store (s128), align 1, addrspace 4)
|
||||
; GCN: INLINEASM &"", 1 /* sideeffect attdialect */
|
||||
; GCN: BUFFER_STORE_FORMAT_XYZW_OFFSET_exact killed [[BUFFER_LOAD_FORMAT_XYZW_OFFSET1]], [[S_LOAD_DWORDX4_IMM]], [[S_MOV_B32_]], 208, 0, 0, 0, implicit $exec :: (dereferenceable store (s128) into custom "BufferResource" + 208, align 1, addrspace 4)
|
||||
; GCN: [[S_MOV_B32_11:%[0-9]+]]:sreg_32 = S_MOV_B32 104
|
||||
; GCN: BUFFER_STORE_FORMAT_XYZW_OFFSET_exact killed [[BUFFER_LOAD_FORMAT_XYZW_OFFSET2]], [[S_LOAD_DWORDX4_IMM]], killed [[S_MOV_B32_11]], 104, 0, 0, 0, implicit $exec :: (dereferenceable store (s128) into custom "BufferResource" + 208, align 1, addrspace 4)
|
||||
; GCN: [[S_MOV_B32_12:%[0-9]+]]:sreg_32 = S_MOV_B32 208
|
||||
; GCN: BUFFER_STORE_FORMAT_XYZW_OFFSET_exact killed [[BUFFER_LOAD_FORMAT_XYZW_OFFSET3]], [[S_LOAD_DWORDX4_IMM]], [[S_MOV_B32_12]], 0, 0, 0, 0, implicit $exec :: (dereferenceable store (s128) into custom "BufferResource" + 208, align 1, addrspace 4)
|
||||
; GCN: BUFFER_STORE_FORMAT_XYZW_OFFEN_exact killed [[BUFFER_LOAD_FORMAT_XYZW_OFFEN1]], [[COPY]], [[S_LOAD_DWORDX4_IMM]], [[S_MOV_B32_12]], 0, 0, 0, 0, implicit $exec :: (dereferenceable store (s128) into custom "BufferResource", align 1, addrspace 4)
|
||||
; GCN: BUFFER_STORE_FORMAT_XYZW_OFFEN_exact killed [[BUFFER_LOAD_FORMAT_XYZW_OFFEN1]], [[COPY]], [[S_LOAD_DWORDX4_IMM]], [[S_MOV_B32_12]], 0, 0, 0, 0, implicit $exec :: (dereferenceable store (s128), align 1, addrspace 4)
|
||||
; GCN: [[COPY25:%[0-9]+]]:sreg_32 = COPY [[COPY]]
|
||||
; GCN: BUFFER_STORE_FORMAT_XYZW_OFFSET_exact killed [[BUFFER_LOAD_FORMAT_XYZW_OFFSET4]], [[S_LOAD_DWORDX4_IMM]], [[COPY25]], 208, 0, 0, 0, implicit $exec :: (dereferenceable store (s128) into custom "BufferResource", align 1, addrspace 4)
|
||||
; GCN: BUFFER_STORE_FORMAT_XYZW_OFFSET_exact killed [[BUFFER_LOAD_FORMAT_XYZW_OFFSET4]], [[S_LOAD_DWORDX4_IMM]], [[COPY25]], 208, 0, 0, 0, implicit $exec :: (dereferenceable store (s128), align 1, addrspace 4)
|
||||
; GCN: INLINEASM &"", 1 /* sideeffect attdialect */
|
||||
; GCN: [[COPY26:%[0-9]+]]:vgpr_32 = COPY [[S_MOV_B32_]]
|
||||
; GCN: [[BUFFER_LOAD_DWORDX4_IDXEN2:%[0-9]+]]:vreg_128 = BUFFER_LOAD_DWORDX4_IDXEN [[COPY26]], [[S_LOAD_DWORDX4_IMM]], [[S_MOV_B32_]], 224, 0, 0, 0, implicit $exec :: (dereferenceable load (s128) from custom "BufferResource" + 224, align 1, addrspace 4)
|
||||
@ -138,12 +138,12 @@ define amdgpu_cs void @mmo_offsets0(<4 x i32> addrspace(6)* inreg noalias derefe
|
||||
; GCN: [[COPY28:%[0-9]+]]:vgpr_32 = COPY [[S_MOV_B32_]]
|
||||
; GCN: [[BUFFER_LOAD_DWORDX4_IDXEN4:%[0-9]+]]:vreg_128 = BUFFER_LOAD_DWORDX4_IDXEN [[COPY28]], [[S_LOAD_DWORDX4_IMM]], [[S_MOV_B32_14]], 0, 0, 0, 0, implicit $exec :: (dereferenceable load (s128) from custom "BufferResource" + 224, align 1, addrspace 4)
|
||||
; GCN: [[REG_SEQUENCE2:%[0-9]+]]:vreg_64 = REG_SEQUENCE [[S_MOV_B32_]], %subreg.sub0, [[COPY]], %subreg.sub1
|
||||
; GCN: [[BUFFER_LOAD_DWORDX4_BOTHEN:%[0-9]+]]:vreg_128 = BUFFER_LOAD_DWORDX4_BOTHEN [[REG_SEQUENCE2]], [[S_LOAD_DWORDX4_IMM]], [[S_MOV_B32_14]], 0, 0, 0, 0, implicit $exec :: (dereferenceable load (s128) from custom "BufferResource", align 1, addrspace 4)
|
||||
; GCN: [[BUFFER_LOAD_DWORDX4_BOTHEN:%[0-9]+]]:vreg_128 = BUFFER_LOAD_DWORDX4_BOTHEN [[REG_SEQUENCE2]], [[S_LOAD_DWORDX4_IMM]], [[S_MOV_B32_14]], 0, 0, 0, 0, implicit $exec :: (dereferenceable load (s128), align 1, addrspace 4)
|
||||
; GCN: [[COPY29:%[0-9]+]]:vgpr_32 = COPY [[S_MOV_B32_]]
|
||||
; GCN: [[COPY30:%[0-9]+]]:sreg_32 = COPY [[COPY]]
|
||||
; GCN: [[BUFFER_LOAD_DWORDX4_IDXEN5:%[0-9]+]]:vreg_128 = BUFFER_LOAD_DWORDX4_IDXEN [[COPY29]], [[S_LOAD_DWORDX4_IMM]], [[COPY30]], 224, 0, 0, 0, implicit $exec :: (dereferenceable load (s128) from custom "BufferResource", align 1, addrspace 4)
|
||||
; GCN: [[BUFFER_LOAD_DWORDX4_IDXEN6:%[0-9]+]]:vreg_128 = BUFFER_LOAD_DWORDX4_IDXEN [[V_MOV_B32_e32_]], [[S_LOAD_DWORDX4_IMM]], [[S_MOV_B32_]], 224, 0, 0, 0, implicit $exec :: (dereferenceable load (s128) from custom "BufferResource", align 1, addrspace 4)
|
||||
; GCN: [[BUFFER_LOAD_DWORDX4_IDXEN7:%[0-9]+]]:vreg_128 = BUFFER_LOAD_DWORDX4_IDXEN [[COPY]], [[S_LOAD_DWORDX4_IMM]], [[S_MOV_B32_]], 224, 0, 0, 0, implicit $exec :: (dereferenceable load (s128) from custom "BufferResource", align 1, addrspace 4)
|
||||
; GCN: [[BUFFER_LOAD_DWORDX4_IDXEN5:%[0-9]+]]:vreg_128 = BUFFER_LOAD_DWORDX4_IDXEN [[COPY29]], [[S_LOAD_DWORDX4_IMM]], [[COPY30]], 224, 0, 0, 0, implicit $exec :: (dereferenceable load (s128), align 1, addrspace 4)
|
||||
; GCN: [[BUFFER_LOAD_DWORDX4_IDXEN6:%[0-9]+]]:vreg_128 = BUFFER_LOAD_DWORDX4_IDXEN [[V_MOV_B32_e32_]], [[S_LOAD_DWORDX4_IMM]], [[S_MOV_B32_]], 224, 0, 0, 0, implicit $exec :: (dereferenceable load (s128), align 1, addrspace 4)
|
||||
; GCN: [[BUFFER_LOAD_DWORDX4_IDXEN7:%[0-9]+]]:vreg_128 = BUFFER_LOAD_DWORDX4_IDXEN [[COPY]], [[S_LOAD_DWORDX4_IMM]], [[S_MOV_B32_]], 224, 0, 0, 0, implicit $exec :: (dereferenceable load (s128), align 1, addrspace 4)
|
||||
; GCN: INLINEASM &"", 1 /* sideeffect attdialect */
|
||||
; GCN: [[COPY31:%[0-9]+]]:vgpr_32 = COPY [[S_MOV_B32_]]
|
||||
; GCN: [[BUFFER_LOAD_FORMAT_XYZW_IDXEN2:%[0-9]+]]:vreg_128 = BUFFER_LOAD_FORMAT_XYZW_IDXEN [[COPY31]], [[S_LOAD_DWORDX4_IMM]], [[S_MOV_B32_]], 240, 0, 0, 0, implicit $exec :: (dereferenceable load (s128) from custom "BufferResource" + 240, align 1, addrspace 4)
|
||||
@ -153,12 +153,12 @@ define amdgpu_cs void @mmo_offsets0(<4 x i32> addrspace(6)* inreg noalias derefe
|
||||
; GCN: [[S_MOV_B32_16:%[0-9]+]]:sreg_32 = S_MOV_B32 240
|
||||
; GCN: [[COPY33:%[0-9]+]]:vgpr_32 = COPY [[S_MOV_B32_]]
|
||||
; GCN: [[BUFFER_LOAD_FORMAT_XYZW_IDXEN4:%[0-9]+]]:vreg_128 = BUFFER_LOAD_FORMAT_XYZW_IDXEN [[COPY33]], [[S_LOAD_DWORDX4_IMM]], [[S_MOV_B32_16]], 0, 0, 0, 0, implicit $exec :: (dereferenceable load (s128) from custom "BufferResource" + 240, align 1, addrspace 4)
|
||||
; GCN: [[BUFFER_LOAD_FORMAT_XYZW_BOTHEN:%[0-9]+]]:vreg_128 = BUFFER_LOAD_FORMAT_XYZW_BOTHEN [[REG_SEQUENCE2]], [[S_LOAD_DWORDX4_IMM]], [[S_MOV_B32_16]], 0, 0, 0, 0, implicit $exec :: (dereferenceable load (s128) from custom "BufferResource", align 1, addrspace 4)
|
||||
; GCN: [[BUFFER_LOAD_FORMAT_XYZW_BOTHEN:%[0-9]+]]:vreg_128 = BUFFER_LOAD_FORMAT_XYZW_BOTHEN [[REG_SEQUENCE2]], [[S_LOAD_DWORDX4_IMM]], [[S_MOV_B32_16]], 0, 0, 0, 0, implicit $exec :: (dereferenceable load (s128), align 1, addrspace 4)
|
||||
; GCN: [[COPY34:%[0-9]+]]:vgpr_32 = COPY [[S_MOV_B32_]]
|
||||
; GCN: [[COPY35:%[0-9]+]]:sreg_32 = COPY [[COPY]]
|
||||
; GCN: [[BUFFER_LOAD_FORMAT_XYZW_IDXEN5:%[0-9]+]]:vreg_128 = BUFFER_LOAD_FORMAT_XYZW_IDXEN [[COPY34]], [[S_LOAD_DWORDX4_IMM]], [[COPY35]], 240, 0, 0, 0, implicit $exec :: (dereferenceable load (s128) from custom "BufferResource", align 1, addrspace 4)
|
||||
; GCN: [[BUFFER_LOAD_FORMAT_XYZW_IDXEN6:%[0-9]+]]:vreg_128 = BUFFER_LOAD_FORMAT_XYZW_IDXEN [[V_MOV_B32_e32_]], [[S_LOAD_DWORDX4_IMM]], [[S_MOV_B32_]], 240, 0, 0, 0, implicit $exec :: (dereferenceable load (s128) from custom "BufferResource", align 1, addrspace 4)
|
||||
; GCN: [[BUFFER_LOAD_FORMAT_XYZW_IDXEN7:%[0-9]+]]:vreg_128 = BUFFER_LOAD_FORMAT_XYZW_IDXEN [[COPY]], [[S_LOAD_DWORDX4_IMM]], [[S_MOV_B32_]], 240, 0, 0, 0, implicit $exec :: (dereferenceable load (s128) from custom "BufferResource", align 1, addrspace 4)
|
||||
; GCN: [[BUFFER_LOAD_FORMAT_XYZW_IDXEN5:%[0-9]+]]:vreg_128 = BUFFER_LOAD_FORMAT_XYZW_IDXEN [[COPY34]], [[S_LOAD_DWORDX4_IMM]], [[COPY35]], 240, 0, 0, 0, implicit $exec :: (dereferenceable load (s128), align 1, addrspace 4)
|
||||
; GCN: [[BUFFER_LOAD_FORMAT_XYZW_IDXEN6:%[0-9]+]]:vreg_128 = BUFFER_LOAD_FORMAT_XYZW_IDXEN [[V_MOV_B32_e32_]], [[S_LOAD_DWORDX4_IMM]], [[S_MOV_B32_]], 240, 0, 0, 0, implicit $exec :: (dereferenceable load (s128), align 1, addrspace 4)
|
||||
; GCN: [[BUFFER_LOAD_FORMAT_XYZW_IDXEN7:%[0-9]+]]:vreg_128 = BUFFER_LOAD_FORMAT_XYZW_IDXEN [[COPY]], [[S_LOAD_DWORDX4_IMM]], [[S_MOV_B32_]], 240, 0, 0, 0, implicit $exec :: (dereferenceable load (s128), align 1, addrspace 4)
|
||||
; GCN: INLINEASM &"", 1 /* sideeffect attdialect */
|
||||
; GCN: [[COPY36:%[0-9]+]]:vgpr_32 = COPY [[S_MOV_B32_]]
|
||||
; GCN: BUFFER_ATOMIC_ADD_IDXEN [[COPY]], [[COPY36]], [[S_LOAD_DWORDX4_IMM]], [[S_MOV_B32_]], 256, 0, implicit $exec :: (volatile dereferenceable load store (s32) on custom "BufferResource" + 256, align 1, addrspace 4)
|
||||
@ -167,12 +167,12 @@ define amdgpu_cs void @mmo_offsets0(<4 x i32> addrspace(6)* inreg noalias derefe
|
||||
; GCN: [[S_MOV_B32_17:%[0-9]+]]:sreg_32 = S_MOV_B32 256
|
||||
; GCN: [[COPY38:%[0-9]+]]:vgpr_32 = COPY [[S_MOV_B32_]]
|
||||
; GCN: BUFFER_ATOMIC_ADD_IDXEN [[COPY]], [[COPY38]], [[S_LOAD_DWORDX4_IMM]], [[S_MOV_B32_17]], 0, 0, implicit $exec :: (volatile dereferenceable load store (s32) on custom "BufferResource" + 256, align 1, addrspace 4)
|
||||
; GCN: BUFFER_ATOMIC_ADD_BOTHEN [[COPY]], [[REG_SEQUENCE2]], [[S_LOAD_DWORDX4_IMM]], [[S_MOV_B32_17]], 0, 0, implicit $exec :: (volatile dereferenceable load store (s32) on custom "BufferResource", align 1, addrspace 4)
|
||||
; GCN: BUFFER_ATOMIC_ADD_BOTHEN [[COPY]], [[REG_SEQUENCE2]], [[S_LOAD_DWORDX4_IMM]], [[S_MOV_B32_17]], 0, 0, implicit $exec :: (volatile dereferenceable load store (s32), align 1, addrspace 4)
|
||||
; GCN: [[COPY39:%[0-9]+]]:vgpr_32 = COPY [[S_MOV_B32_]]
|
||||
; GCN: [[COPY40:%[0-9]+]]:sreg_32 = COPY [[COPY]]
|
||||
; GCN: BUFFER_ATOMIC_ADD_IDXEN [[COPY]], [[COPY39]], [[S_LOAD_DWORDX4_IMM]], [[COPY40]], 256, 0, implicit $exec :: (volatile dereferenceable load store (s32) on custom "BufferResource", align 1, addrspace 4)
|
||||
; GCN: BUFFER_ATOMIC_ADD_IDXEN [[COPY]], [[V_MOV_B32_e32_]], [[S_LOAD_DWORDX4_IMM]], [[S_MOV_B32_]], 256, 0, implicit $exec :: (volatile dereferenceable load store (s32) on custom "BufferResource", align 1, addrspace 4)
|
||||
; GCN: BUFFER_ATOMIC_ADD_IDXEN [[COPY]], [[COPY]], [[S_LOAD_DWORDX4_IMM]], [[S_MOV_B32_]], 256, 0, implicit $exec :: (volatile dereferenceable load store (s32) on custom "BufferResource", align 1, addrspace 4)
|
||||
; GCN: BUFFER_ATOMIC_ADD_IDXEN [[COPY]], [[COPY39]], [[S_LOAD_DWORDX4_IMM]], [[COPY40]], 256, 0, implicit $exec :: (volatile dereferenceable load store (s32), align 1, addrspace 4)
|
||||
; GCN: BUFFER_ATOMIC_ADD_IDXEN [[COPY]], [[V_MOV_B32_e32_]], [[S_LOAD_DWORDX4_IMM]], [[S_MOV_B32_]], 256, 0, implicit $exec :: (volatile dereferenceable load store (s32), align 1, addrspace 4)
|
||||
; GCN: BUFFER_ATOMIC_ADD_IDXEN [[COPY]], [[COPY]], [[S_LOAD_DWORDX4_IMM]], [[S_MOV_B32_]], 256, 0, implicit $exec :: (volatile dereferenceable load store (s32), align 1, addrspace 4)
|
||||
; GCN: INLINEASM &"", 1 /* sideeffect attdialect */
|
||||
; GCN: [[COPY41:%[0-9]+]]:vgpr_32 = COPY [[S_MOV_B32_]]
|
||||
; GCN: [[DEF9:%[0-9]+]]:vreg_64 = IMPLICIT_DEF
|
||||
@ -192,21 +192,21 @@ define amdgpu_cs void @mmo_offsets0(<4 x i32> addrspace(6)* inreg noalias derefe
|
||||
; GCN: [[COPY48:%[0-9]+]]:vreg_64 = COPY [[DEF11]]
|
||||
; GCN: [[COPY49:%[0-9]+]]:vgpr_32 = COPY [[COPY48]].sub0
|
||||
; GCN: [[DEF12:%[0-9]+]]:vreg_64 = IMPLICIT_DEF
|
||||
; GCN: BUFFER_ATOMIC_CMPSWAP_BOTHEN [[REG_SEQUENCE1]], [[REG_SEQUENCE2]], [[S_LOAD_DWORDX4_IMM]], [[S_MOV_B32_19]], 0, 1, implicit $exec :: (volatile dereferenceable load store (s32) on custom "BufferResource", align 1, addrspace 4)
|
||||
; GCN: BUFFER_ATOMIC_CMPSWAP_BOTHEN [[REG_SEQUENCE1]], [[REG_SEQUENCE2]], [[S_LOAD_DWORDX4_IMM]], [[S_MOV_B32_19]], 0, 1, implicit $exec :: (volatile dereferenceable load store (s32), align 1, addrspace 4)
|
||||
; GCN: [[COPY50:%[0-9]+]]:vreg_64 = COPY [[DEF12]]
|
||||
; GCN: [[COPY51:%[0-9]+]]:vgpr_32 = COPY [[COPY50]].sub0
|
||||
; GCN: [[COPY52:%[0-9]+]]:vgpr_32 = COPY [[S_MOV_B32_]]
|
||||
; GCN: [[COPY53:%[0-9]+]]:sreg_32 = COPY [[COPY]]
|
||||
; GCN: [[DEF13:%[0-9]+]]:vreg_64 = IMPLICIT_DEF
|
||||
; GCN: BUFFER_ATOMIC_CMPSWAP_IDXEN [[REG_SEQUENCE1]], [[COPY52]], [[S_LOAD_DWORDX4_IMM]], [[COPY53]], 272, 1, implicit $exec :: (volatile dereferenceable load store (s32) on custom "BufferResource", align 1, addrspace 4)
|
||||
; GCN: BUFFER_ATOMIC_CMPSWAP_IDXEN [[REG_SEQUENCE1]], [[COPY52]], [[S_LOAD_DWORDX4_IMM]], [[COPY53]], 272, 1, implicit $exec :: (volatile dereferenceable load store (s32), align 1, addrspace 4)
|
||||
; GCN: [[COPY54:%[0-9]+]]:vreg_64 = COPY [[DEF13]]
|
||||
; GCN: [[COPY55:%[0-9]+]]:vgpr_32 = COPY [[COPY54]].sub0
|
||||
; GCN: [[DEF14:%[0-9]+]]:vreg_64 = IMPLICIT_DEF
|
||||
; GCN: BUFFER_ATOMIC_CMPSWAP_IDXEN [[REG_SEQUENCE1]], [[V_MOV_B32_e32_]], [[S_LOAD_DWORDX4_IMM]], [[S_MOV_B32_]], 272, 1, implicit $exec :: (volatile dereferenceable load store (s32) on custom "BufferResource", align 1, addrspace 4)
|
||||
; GCN: BUFFER_ATOMIC_CMPSWAP_IDXEN [[REG_SEQUENCE1]], [[V_MOV_B32_e32_]], [[S_LOAD_DWORDX4_IMM]], [[S_MOV_B32_]], 272, 1, implicit $exec :: (volatile dereferenceable load store (s32), align 1, addrspace 4)
|
||||
; GCN: [[COPY56:%[0-9]+]]:vreg_64 = COPY [[DEF14]]
|
||||
; GCN: [[COPY57:%[0-9]+]]:vgpr_32 = COPY [[COPY56]].sub0
|
||||
; GCN: [[DEF15:%[0-9]+]]:vreg_64 = IMPLICIT_DEF
|
||||
; GCN: BUFFER_ATOMIC_CMPSWAP_IDXEN [[REG_SEQUENCE1]], [[COPY]], [[S_LOAD_DWORDX4_IMM]], [[S_MOV_B32_]], 272, 1, implicit $exec :: (volatile dereferenceable load store (s32) on custom "BufferResource", align 1, addrspace 4)
|
||||
; GCN: BUFFER_ATOMIC_CMPSWAP_IDXEN [[REG_SEQUENCE1]], [[COPY]], [[S_LOAD_DWORDX4_IMM]], [[S_MOV_B32_]], 272, 1, implicit $exec :: (volatile dereferenceable load store (s32), align 1, addrspace 4)
|
||||
; GCN: [[COPY58:%[0-9]+]]:vreg_64 = COPY [[DEF15]]
|
||||
; GCN: [[COPY59:%[0-9]+]]:vgpr_32 = COPY [[COPY58]].sub0
|
||||
; GCN: INLINEASM &"", 1 /* sideeffect attdialect */
|
||||
@ -217,12 +217,12 @@ define amdgpu_cs void @mmo_offsets0(<4 x i32> addrspace(6)* inreg noalias derefe
|
||||
; GCN: [[S_MOV_B32_20:%[0-9]+]]:sreg_32 = S_MOV_B32 288
|
||||
; GCN: [[COPY62:%[0-9]+]]:vgpr_32 = COPY [[S_MOV_B32_]]
|
||||
; GCN: BUFFER_STORE_DWORDX4_IDXEN_exact killed [[BUFFER_LOAD_DWORDX4_IDXEN4]], [[COPY62]], [[S_LOAD_DWORDX4_IMM]], [[S_MOV_B32_20]], 0, 0, 0, 0, implicit $exec :: (dereferenceable store (s128) into custom "BufferResource" + 288, align 1, addrspace 4)
|
||||
; GCN: BUFFER_STORE_DWORDX4_BOTHEN_exact killed [[BUFFER_LOAD_DWORDX4_BOTHEN]], [[REG_SEQUENCE2]], [[S_LOAD_DWORDX4_IMM]], [[S_MOV_B32_20]], 0, 0, 0, 0, implicit $exec :: (dereferenceable store (s128) into custom "BufferResource", align 1, addrspace 4)
|
||||
; GCN: BUFFER_STORE_DWORDX4_BOTHEN_exact killed [[BUFFER_LOAD_DWORDX4_BOTHEN]], [[REG_SEQUENCE2]], [[S_LOAD_DWORDX4_IMM]], [[S_MOV_B32_20]], 0, 0, 0, 0, implicit $exec :: (dereferenceable store (s128), align 1, addrspace 4)
|
||||
; GCN: [[COPY63:%[0-9]+]]:vgpr_32 = COPY [[S_MOV_B32_]]
|
||||
; GCN: [[COPY64:%[0-9]+]]:sreg_32 = COPY [[COPY]]
|
||||
; GCN: BUFFER_STORE_DWORDX4_IDXEN_exact killed [[BUFFER_LOAD_DWORDX4_IDXEN5]], [[COPY63]], [[S_LOAD_DWORDX4_IMM]], [[COPY64]], 288, 0, 0, 0, implicit $exec :: (dereferenceable store (s128) into custom "BufferResource", align 1, addrspace 4)
|
||||
; GCN: BUFFER_STORE_DWORDX4_IDXEN_exact killed [[BUFFER_LOAD_DWORDX4_IDXEN6]], [[V_MOV_B32_e32_]], [[S_LOAD_DWORDX4_IMM]], [[S_MOV_B32_]], 288, 0, 0, 0, implicit $exec :: (dereferenceable store (s128) into custom "BufferResource", align 1, addrspace 4)
|
||||
; GCN: BUFFER_STORE_DWORDX4_IDXEN_exact killed [[BUFFER_LOAD_DWORDX4_IDXEN7]], [[COPY]], [[S_LOAD_DWORDX4_IMM]], [[S_MOV_B32_]], 288, 0, 0, 0, implicit $exec :: (dereferenceable store (s128) into custom "BufferResource", align 1, addrspace 4)
|
||||
; GCN: BUFFER_STORE_DWORDX4_IDXEN_exact killed [[BUFFER_LOAD_DWORDX4_IDXEN5]], [[COPY63]], [[S_LOAD_DWORDX4_IMM]], [[COPY64]], 288, 0, 0, 0, implicit $exec :: (dereferenceable store (s128), align 1, addrspace 4)
|
||||
; GCN: BUFFER_STORE_DWORDX4_IDXEN_exact killed [[BUFFER_LOAD_DWORDX4_IDXEN6]], [[V_MOV_B32_e32_]], [[S_LOAD_DWORDX4_IMM]], [[S_MOV_B32_]], 288, 0, 0, 0, implicit $exec :: (dereferenceable store (s128), align 1, addrspace 4)
|
||||
; GCN: BUFFER_STORE_DWORDX4_IDXEN_exact killed [[BUFFER_LOAD_DWORDX4_IDXEN7]], [[COPY]], [[S_LOAD_DWORDX4_IMM]], [[S_MOV_B32_]], 288, 0, 0, 0, implicit $exec :: (dereferenceable store (s128), align 1, addrspace 4)
|
||||
; GCN: INLINEASM &"", 1 /* sideeffect attdialect */
|
||||
; GCN: [[COPY65:%[0-9]+]]:vgpr_32 = COPY [[S_MOV_B32_]]
|
||||
; GCN: BUFFER_STORE_FORMAT_XYZW_IDXEN_exact killed [[BUFFER_LOAD_FORMAT_XYZW_IDXEN2]], [[COPY65]], [[S_LOAD_DWORDX4_IMM]], [[S_MOV_B32_]], 304, 0, 0, 0, implicit $exec :: (dereferenceable store (s128) into custom "BufferResource" + 304, align 1, addrspace 4)
|
||||
@ -232,12 +232,12 @@ define amdgpu_cs void @mmo_offsets0(<4 x i32> addrspace(6)* inreg noalias derefe
|
||||
; GCN: [[S_MOV_B32_22:%[0-9]+]]:sreg_32 = S_MOV_B32 304
|
||||
; GCN: [[COPY67:%[0-9]+]]:vgpr_32 = COPY [[S_MOV_B32_]]
|
||||
; GCN: BUFFER_STORE_FORMAT_XYZW_IDXEN_exact killed [[BUFFER_LOAD_FORMAT_XYZW_IDXEN4]], [[COPY67]], [[S_LOAD_DWORDX4_IMM]], [[S_MOV_B32_22]], 0, 0, 0, 0, implicit $exec :: (dereferenceable store (s128) into custom "BufferResource" + 304, align 1, addrspace 4)
|
||||
; GCN: BUFFER_STORE_FORMAT_XYZW_BOTHEN_exact killed [[BUFFER_LOAD_FORMAT_XYZW_BOTHEN]], [[REG_SEQUENCE2]], [[S_LOAD_DWORDX4_IMM]], [[S_MOV_B32_22]], 0, 0, 0, 0, implicit $exec :: (dereferenceable store (s128) into custom "BufferResource", align 1, addrspace 4)
|
||||
; GCN: BUFFER_STORE_FORMAT_XYZW_BOTHEN_exact killed [[BUFFER_LOAD_FORMAT_XYZW_BOTHEN]], [[REG_SEQUENCE2]], [[S_LOAD_DWORDX4_IMM]], [[S_MOV_B32_22]], 0, 0, 0, 0, implicit $exec :: (dereferenceable store (s128), align 1, addrspace 4)
|
||||
; GCN: [[COPY68:%[0-9]+]]:vgpr_32 = COPY [[S_MOV_B32_]]
|
||||
; GCN: [[COPY69:%[0-9]+]]:sreg_32 = COPY [[COPY]]
|
||||
; GCN: BUFFER_STORE_FORMAT_XYZW_IDXEN_exact killed [[BUFFER_LOAD_FORMAT_XYZW_IDXEN5]], [[COPY68]], [[S_LOAD_DWORDX4_IMM]], [[COPY69]], 304, 0, 0, 0, implicit $exec :: (dereferenceable store (s128) into custom "BufferResource", align 1, addrspace 4)
|
||||
; GCN: BUFFER_STORE_FORMAT_XYZW_IDXEN_exact killed [[BUFFER_LOAD_FORMAT_XYZW_IDXEN6]], [[V_MOV_B32_e32_]], [[S_LOAD_DWORDX4_IMM]], [[S_MOV_B32_]], 304, 0, 0, 0, implicit $exec :: (dereferenceable store (s128) into custom "BufferResource", align 1, addrspace 4)
|
||||
; GCN: BUFFER_STORE_FORMAT_XYZW_IDXEN_exact killed [[BUFFER_LOAD_FORMAT_XYZW_IDXEN7]], [[COPY]], [[S_LOAD_DWORDX4_IMM]], [[S_MOV_B32_]], 304, 0, 0, 0, implicit $exec :: (dereferenceable store (s128) into custom "BufferResource", align 1, addrspace 4)
|
||||
; GCN: BUFFER_STORE_FORMAT_XYZW_IDXEN_exact killed [[BUFFER_LOAD_FORMAT_XYZW_IDXEN5]], [[COPY68]], [[S_LOAD_DWORDX4_IMM]], [[COPY69]], 304, 0, 0, 0, implicit $exec :: (dereferenceable store (s128), align 1, addrspace 4)
|
||||
; GCN: BUFFER_STORE_FORMAT_XYZW_IDXEN_exact killed [[BUFFER_LOAD_FORMAT_XYZW_IDXEN6]], [[V_MOV_B32_e32_]], [[S_LOAD_DWORDX4_IMM]], [[S_MOV_B32_]], 304, 0, 0, 0, implicit $exec :: (dereferenceable store (s128), align 1, addrspace 4)
|
||||
; GCN: BUFFER_STORE_FORMAT_XYZW_IDXEN_exact killed [[BUFFER_LOAD_FORMAT_XYZW_IDXEN7]], [[COPY]], [[S_LOAD_DWORDX4_IMM]], [[S_MOV_B32_]], 304, 0, 0, 0, implicit $exec :: (dereferenceable store (s128), align 1, addrspace 4)
|
||||
; GCN: S_ENDPGM 0
|
||||
bb.0:
|
||||
%tmp0 = load <4 x i32>, <4 x i32> addrspace(6)* %arg0, align 16, !invariant.load !0
|
||||
|
||||
@ -44,11 +44,9 @@ define amdgpu_cs void @_amdgpu_cs_main(<3 x i32> inreg %arg3, <3 x i32> %arg5) {
|
||||
ret void
|
||||
}
|
||||
|
||||
; FIXME: the first load and store should not be reordered because they might
|
||||
; alias depending on the value of %off
|
||||
; GCN-LABEL: {{^}}test1:
|
||||
; GCN: buffer_load_dword
|
||||
; GCN: buffer_store_dword
|
||||
; GCN: buffer_load_dword
|
||||
; GCN: buffer_store_dword
|
||||
define amdgpu_cs void @test1(<4 x i32> inreg %buf, i32 %off) {
|
||||
.entry:
|
||||
|
||||
@ -12,7 +12,7 @@ define amdgpu_hs void @main([0 x i8] addrspace(6)* inreg %arg) {
|
||||
; GCN: [[DEF:%[0-9]+]]:sreg_32 = IMPLICIT_DEF
|
||||
; GCN: [[COPY:%[0-9]+]]:vgpr_32 = COPY [[DEF]]
|
||||
; GCN: [[DEF1:%[0-9]+]]:sgpr_128 = IMPLICIT_DEF
|
||||
; GCN: [[BUFFER_LOAD_DWORDX4_OFFEN:%[0-9]+]]:vreg_128 = BUFFER_LOAD_DWORDX4_OFFEN [[COPY]], [[DEF1]], [[S_MOV_B32_]], 0, 0, 0, 0, implicit $exec :: (dereferenceable load (s128) from custom "BufferResource", align 1, addrspace 4)
|
||||
; GCN: [[BUFFER_LOAD_DWORDX4_OFFEN:%[0-9]+]]:vreg_128 = BUFFER_LOAD_DWORDX4_OFFEN [[COPY]], [[DEF1]], [[S_MOV_B32_]], 0, 0, 0, 0, implicit $exec :: (dereferenceable load (s128), align 1, addrspace 4)
|
||||
; GCN: [[COPY1:%[0-9]+]]:sgpr_32 = COPY [[BUFFER_LOAD_DWORDX4_OFFEN]].sub2
|
||||
; GCN: [[COPY2:%[0-9]+]]:sgpr_32 = COPY [[BUFFER_LOAD_DWORDX4_OFFEN]].sub1
|
||||
; GCN: [[COPY3:%[0-9]+]]:sgpr_32 = COPY [[BUFFER_LOAD_DWORDX4_OFFEN]].sub0
|
||||
@ -21,7 +21,7 @@ define amdgpu_hs void @main([0 x i8] addrspace(6)* inreg %arg) {
|
||||
; GCN: [[DEF2:%[0-9]+]]:sreg_32 = IMPLICIT_DEF
|
||||
; GCN: [[COPY5:%[0-9]+]]:vgpr_32 = COPY [[DEF2]]
|
||||
; GCN: [[DEF3:%[0-9]+]]:sgpr_128 = IMPLICIT_DEF
|
||||
; GCN: BUFFER_STORE_DWORDX3_OFFEN_exact killed [[COPY4]], [[COPY5]], [[DEF3]], [[S_MOV_B32_]], 0, 0, 0, 0, implicit $exec :: (dereferenceable store (s96) into custom "BufferResource", align 1, addrspace 4)
|
||||
; GCN: BUFFER_STORE_DWORDX3_OFFEN_exact killed [[COPY4]], [[COPY5]], [[DEF3]], [[S_MOV_B32_]], 0, 0, 0, 0, implicit $exec :: (dereferenceable store (s96), align 1, addrspace 4)
|
||||
; GCN: S_ENDPGM 0
|
||||
main_body:
|
||||
%tmp25 = call <4 x float> @llvm.amdgcn.raw.buffer.load.v4f32(<4 x i32> undef, i32 undef, i32 0, i32 0)
|
||||
|
||||
Loading…
Reference in New Issue
Block a user