2021-01-06 12:23:39 +08:00
|
|
|
;RUN: llc < %s -march=amdgcn -mcpu=verde -verify-machineinstrs | FileCheck -check-prefixes=GCN,PREGFX10 %s
|
2019-05-02 00:32:58 +08:00
|
|
|
;RUN: llc < %s -march=amdgcn -mcpu=tonga -verify-machineinstrs | FileCheck -check-prefixes=GCN,PREGFX10 %s
|
|
|
|
|
;RUN: llc < %s -march=amdgcn -mcpu=gfx1010 -verify-machineinstrs | FileCheck -check-prefixes=GCN,GFX10 %s
|
[AMDGPU] New tbuffer intrinsics
Summary:
This commit adds new intrinsics
llvm.amdgcn.raw.tbuffer.load
llvm.amdgcn.struct.tbuffer.load
llvm.amdgcn.raw.tbuffer.store
llvm.amdgcn.struct.tbuffer.store
with the following changes from the llvm.amdgcn.tbuffer.* intrinsics:
* there are separate raw and struct versions: raw does not have an index
arg and sets idxen=0 in the instruction, and struct always sets
idxen=1 in the instruction even if the index is 0, to allow for the
fact that gfx9 does bounds checking differently depending on whether
idxen is set;
* there is a combined format arg (dfmt+nfmt)
* there is a combined cachepolicy arg (glc+slc)
* there are now only two offset args: one for the offset that is
included in bounds checking and swizzling, to be split between the
instruction's voffset and immoffset fields, and one for the offset
that is excluded from bounds checking and swizzling, to go into the
instruction's soffset field.
The AMDISD::TBUFFER_* SD nodes always have an index operand, all three
offset operands, combined format operand, combined cachepolicy operand,
and an extra idxen operand.
The tbuffer pseudo- and real instructions now also have a combined
format operand.
The obsolescent llvm.amdgcn.tbuffer.* and llvm.SI.tbuffer.store
intrinsics continue to work.
V2: Separate raw and struct intrinsics.
V3: Moved extract_glc and extract_slc defs to a more sensible place.
V4: Rebased on D49995.
V5: Only two separate offset args instead of three.
V6: Pseudo- and real instructions have joint format operand.
V7: Restored optionality of dfmt and nfmt in assembler.
V8: Addressed minor review comments.
Subscribers: arsenm, kzhuravl, wdng, nhaehnle, yaxunl, dstuttard, t-tye, llvm-commits
Differential Revision: https://reviews.llvm.org/D49026
Change-Id: If22ad77e349fac3a5d2f72dda53c010377d470d4
llvm-svn: 340268
2018-08-21 19:06:05 +08:00
|
|
|
|
|
|
|
|
; GCN-LABEL: {{^}}tbuffer_store:
|
[AMDGPU][MC] Added support of SP3 syntax for MTBUF format modifier
Currently supported LLVM MTBUF syntax is shown below. It is not compatible with SP3.
op dst, addr, rsrc, FORMAT, soffset
This change adds support for SP3 syntax:
op dst, addr, rsrc, soffset SP3FORMAT
In addition to being compatible with SP3, this syntax allows using symbolic names for data, numeric and unified formats. Below is a list of added syntax variants.
format:<expression>
format:[<numeric-format-name>,<data-format-name>]
format:[<data-format-name>,<numeric-format-name>]
format:[<data-format-name>]
format:[<numeric-format-name>]
format:[<unified-format-name>]
The last syntax variant is supported for GFX10 only.
See llvm bug 37738
Reviewers: arsenm, rampitec, vpykhtin
Differential Revision: https://reviews.llvm.org/D84026
2020-07-24 21:39:42 +08:00
|
|
|
; PREGFX10: tbuffer_store_format_xyzw v[0:3], off, s[0:3], 0 format:[BUF_DATA_FORMAT_16_16_16_16,BUF_NUM_FORMAT_USCALED]
|
|
|
|
|
; PREGFX10: tbuffer_store_format_xyzw v[4:7], off, s[0:3], 0 format:[BUF_DATA_FORMAT_32_32_32,BUF_NUM_FORMAT_SSCALED] glc
|
|
|
|
|
; PREGFX10: tbuffer_store_format_xyzw v[8:11], off, s[0:3], 0 format:[BUF_DATA_FORMAT_32_32_32_32,BUF_NUM_FORMAT_UINT] slc
|
|
|
|
|
; PREGFX10: tbuffer_store_format_xyzw v[8:11], off, s[0:3], 0 format:[BUF_DATA_FORMAT_32_32_32_32,BUF_NUM_FORMAT_UINT] glc
|
|
|
|
|
; GFX10: tbuffer_store_format_xyzw v[0:3], off, s[0:3], 0 format:[BUF_FMT_10_10_10_2_UNORM]
|
|
|
|
|
; GFX10: tbuffer_store_format_xyzw v[4:7], off, s[0:3], 0 format:[BUF_FMT_8_8_8_8_SINT] glc
|
|
|
|
|
; GFX10: tbuffer_store_format_xyzw v[8:11], off, s[0:3], 0 format:78 slc
|
|
|
|
|
; GFX10: tbuffer_store_format_xyzw v[8:11], off, s[0:3], 0 format:78 glc dlc
|
[AMDGPU] New tbuffer intrinsics
Summary:
This commit adds new intrinsics
llvm.amdgcn.raw.tbuffer.load
llvm.amdgcn.struct.tbuffer.load
llvm.amdgcn.raw.tbuffer.store
llvm.amdgcn.struct.tbuffer.store
with the following changes from the llvm.amdgcn.tbuffer.* intrinsics:
* there are separate raw and struct versions: raw does not have an index
arg and sets idxen=0 in the instruction, and struct always sets
idxen=1 in the instruction even if the index is 0, to allow for the
fact that gfx9 does bounds checking differently depending on whether
idxen is set;
* there is a combined format arg (dfmt+nfmt)
* there is a combined cachepolicy arg (glc+slc)
* there are now only two offset args: one for the offset that is
included in bounds checking and swizzling, to be split between the
instruction's voffset and immoffset fields, and one for the offset
that is excluded from bounds checking and swizzling, to go into the
instruction's soffset field.
The AMDISD::TBUFFER_* SD nodes always have an index operand, all three
offset operands, combined format operand, combined cachepolicy operand,
and an extra idxen operand.
The tbuffer pseudo- and real instructions now also have a combined
format operand.
The obsolescent llvm.amdgcn.tbuffer.* and llvm.SI.tbuffer.store
intrinsics continue to work.
V2: Separate raw and struct intrinsics.
V3: Moved extract_glc and extract_slc defs to a more sensible place.
V4: Rebased on D49995.
V5: Only two separate offset args instead of three.
V6: Pseudo- and real instructions have joint format operand.
V7: Restored optionality of dfmt and nfmt in assembler.
V8: Addressed minor review comments.
Subscribers: arsenm, kzhuravl, wdng, nhaehnle, yaxunl, dstuttard, t-tye, llvm-commits
Differential Revision: https://reviews.llvm.org/D49026
Change-Id: If22ad77e349fac3a5d2f72dda53c010377d470d4
llvm-svn: 340268
2018-08-21 19:06:05 +08:00
|
|
|
define amdgpu_ps void @tbuffer_store(<4 x i32> inreg, <4 x float>, <4 x float>, <4 x float>) {
|
|
|
|
|
main_body:
|
|
|
|
|
%in1 = bitcast <4 x float> %1 to <4 x i32>
|
|
|
|
|
%in2 = bitcast <4 x float> %2 to <4 x i32>
|
|
|
|
|
%in3 = bitcast <4 x float> %3 to <4 x i32>
|
|
|
|
|
call void @llvm.amdgcn.raw.tbuffer.store.v4i32(<4 x i32> %in1, <4 x i32> %0, i32 0, i32 0, i32 44, i32 0)
|
|
|
|
|
call void @llvm.amdgcn.raw.tbuffer.store.v4i32(<4 x i32> %in2, <4 x i32> %0, i32 0, i32 0, i32 61, i32 1)
|
|
|
|
|
call void @llvm.amdgcn.raw.tbuffer.store.v4i32(<4 x i32> %in3, <4 x i32> %0, i32 0, i32 0, i32 78, i32 2)
|
2019-05-02 00:32:58 +08:00
|
|
|
call void @llvm.amdgcn.raw.tbuffer.store.v4f32(<4 x float> %3, <4 x i32> %0, i32 0, i32 0, i32 78, i32 5)
|
[AMDGPU] New tbuffer intrinsics
Summary:
This commit adds new intrinsics
llvm.amdgcn.raw.tbuffer.load
llvm.amdgcn.struct.tbuffer.load
llvm.amdgcn.raw.tbuffer.store
llvm.amdgcn.struct.tbuffer.store
with the following changes from the llvm.amdgcn.tbuffer.* intrinsics:
* there are separate raw and struct versions: raw does not have an index
arg and sets idxen=0 in the instruction, and struct always sets
idxen=1 in the instruction even if the index is 0, to allow for the
fact that gfx9 does bounds checking differently depending on whether
idxen is set;
* there is a combined format arg (dfmt+nfmt)
* there is a combined cachepolicy arg (glc+slc)
* there are now only two offset args: one for the offset that is
included in bounds checking and swizzling, to be split between the
instruction's voffset and immoffset fields, and one for the offset
that is excluded from bounds checking and swizzling, to go into the
instruction's soffset field.
The AMDISD::TBUFFER_* SD nodes always have an index operand, all three
offset operands, combined format operand, combined cachepolicy operand,
and an extra idxen operand.
The tbuffer pseudo- and real instructions now also have a combined
format operand.
The obsolescent llvm.amdgcn.tbuffer.* and llvm.SI.tbuffer.store
intrinsics continue to work.
V2: Separate raw and struct intrinsics.
V3: Moved extract_glc and extract_slc defs to a more sensible place.
V4: Rebased on D49995.
V5: Only two separate offset args instead of three.
V6: Pseudo- and real instructions have joint format operand.
V7: Restored optionality of dfmt and nfmt in assembler.
V8: Addressed minor review comments.
Subscribers: arsenm, kzhuravl, wdng, nhaehnle, yaxunl, dstuttard, t-tye, llvm-commits
Differential Revision: https://reviews.llvm.org/D49026
Change-Id: If22ad77e349fac3a5d2f72dda53c010377d470d4
llvm-svn: 340268
2018-08-21 19:06:05 +08:00
|
|
|
ret void
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
; GCN-LABEL: {{^}}tbuffer_store_immoffs:
|
[AMDGPU][MC] Added support of SP3 syntax for MTBUF format modifier
Currently supported LLVM MTBUF syntax is shown below. It is not compatible with SP3.
op dst, addr, rsrc, FORMAT, soffset
This change adds support for SP3 syntax:
op dst, addr, rsrc, soffset SP3FORMAT
In addition to being compatible with SP3, this syntax allows using symbolic names for data, numeric and unified formats. Below is a list of added syntax variants.
format:<expression>
format:[<numeric-format-name>,<data-format-name>]
format:[<data-format-name>,<numeric-format-name>]
format:[<data-format-name>]
format:[<numeric-format-name>]
format:[<unified-format-name>]
The last syntax variant is supported for GFX10 only.
See llvm bug 37738
Reviewers: arsenm, rampitec, vpykhtin
Differential Revision: https://reviews.llvm.org/D84026
2020-07-24 21:39:42 +08:00
|
|
|
; PREGFX10: tbuffer_store_format_xyzw v[0:3], off, s[0:3], 0 format:[BUF_DATA_FORMAT_16_16,BUF_NUM_FORMAT_FLOAT] offset:42
|
|
|
|
|
; GFX10: tbuffer_store_format_xyzw v[0:3], off, s[0:3], 0 format:117 offset:42
|
[AMDGPU] New tbuffer intrinsics
Summary:
This commit adds new intrinsics
llvm.amdgcn.raw.tbuffer.load
llvm.amdgcn.struct.tbuffer.load
llvm.amdgcn.raw.tbuffer.store
llvm.amdgcn.struct.tbuffer.store
with the following changes from the llvm.amdgcn.tbuffer.* intrinsics:
* there are separate raw and struct versions: raw does not have an index
arg and sets idxen=0 in the instruction, and struct always sets
idxen=1 in the instruction even if the index is 0, to allow for the
fact that gfx9 does bounds checking differently depending on whether
idxen is set;
* there is a combined format arg (dfmt+nfmt)
* there is a combined cachepolicy arg (glc+slc)
* there are now only two offset args: one for the offset that is
included in bounds checking and swizzling, to be split between the
instruction's voffset and immoffset fields, and one for the offset
that is excluded from bounds checking and swizzling, to go into the
instruction's soffset field.
The AMDISD::TBUFFER_* SD nodes always have an index operand, all three
offset operands, combined format operand, combined cachepolicy operand,
and an extra idxen operand.
The tbuffer pseudo- and real instructions now also have a combined
format operand.
The obsolescent llvm.amdgcn.tbuffer.* and llvm.SI.tbuffer.store
intrinsics continue to work.
V2: Separate raw and struct intrinsics.
V3: Moved extract_glc and extract_slc defs to a more sensible place.
V4: Rebased on D49995.
V5: Only two separate offset args instead of three.
V6: Pseudo- and real instructions have joint format operand.
V7: Restored optionality of dfmt and nfmt in assembler.
V8: Addressed minor review comments.
Subscribers: arsenm, kzhuravl, wdng, nhaehnle, yaxunl, dstuttard, t-tye, llvm-commits
Differential Revision: https://reviews.llvm.org/D49026
Change-Id: If22ad77e349fac3a5d2f72dda53c010377d470d4
llvm-svn: 340268
2018-08-21 19:06:05 +08:00
|
|
|
define amdgpu_ps void @tbuffer_store_immoffs(<4 x i32> inreg, <4 x float>) {
|
|
|
|
|
main_body:
|
|
|
|
|
%in1 = bitcast <4 x float> %1 to <4 x i32>
|
|
|
|
|
call void @llvm.amdgcn.raw.tbuffer.store.v4i32(<4 x i32> %in1, <4 x i32> %0, i32 42, i32 0, i32 117, i32 0)
|
|
|
|
|
ret void
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
; GCN-LABEL: {{^}}tbuffer_store_scalar_and_imm_offs:
|
[AMDGPU][MC] Added support of SP3 syntax for MTBUF format modifier
Currently supported LLVM MTBUF syntax is shown below. It is not compatible with SP3.
op dst, addr, rsrc, FORMAT, soffset
This change adds support for SP3 syntax:
op dst, addr, rsrc, soffset SP3FORMAT
In addition to being compatible with SP3, this syntax allows using symbolic names for data, numeric and unified formats. Below is a list of added syntax variants.
format:<expression>
format:[<numeric-format-name>,<data-format-name>]
format:[<data-format-name>,<numeric-format-name>]
format:[<data-format-name>]
format:[<numeric-format-name>]
format:[<unified-format-name>]
The last syntax variant is supported for GFX10 only.
See llvm bug 37738
Reviewers: arsenm, rampitec, vpykhtin
Differential Revision: https://reviews.llvm.org/D84026
2020-07-24 21:39:42 +08:00
|
|
|
; PREGFX10: tbuffer_store_format_xyzw v[0:3], off, s[0:3], {{s[0-9]+}} format:[BUF_DATA_FORMAT_16_16,BUF_NUM_FORMAT_FLOAT] offset:42
|
|
|
|
|
; GFX10: tbuffer_store_format_xyzw v[0:3], off, s[0:3], {{s[0-9]+}} format:117 offset:42
|
[AMDGPU] New tbuffer intrinsics
Summary:
This commit adds new intrinsics
llvm.amdgcn.raw.tbuffer.load
llvm.amdgcn.struct.tbuffer.load
llvm.amdgcn.raw.tbuffer.store
llvm.amdgcn.struct.tbuffer.store
with the following changes from the llvm.amdgcn.tbuffer.* intrinsics:
* there are separate raw and struct versions: raw does not have an index
arg and sets idxen=0 in the instruction, and struct always sets
idxen=1 in the instruction even if the index is 0, to allow for the
fact that gfx9 does bounds checking differently depending on whether
idxen is set;
* there is a combined format arg (dfmt+nfmt)
* there is a combined cachepolicy arg (glc+slc)
* there are now only two offset args: one for the offset that is
included in bounds checking and swizzling, to be split between the
instruction's voffset and immoffset fields, and one for the offset
that is excluded from bounds checking and swizzling, to go into the
instruction's soffset field.
The AMDISD::TBUFFER_* SD nodes always have an index operand, all three
offset operands, combined format operand, combined cachepolicy operand,
and an extra idxen operand.
The tbuffer pseudo- and real instructions now also have a combined
format operand.
The obsolescent llvm.amdgcn.tbuffer.* and llvm.SI.tbuffer.store
intrinsics continue to work.
V2: Separate raw and struct intrinsics.
V3: Moved extract_glc and extract_slc defs to a more sensible place.
V4: Rebased on D49995.
V5: Only two separate offset args instead of three.
V6: Pseudo- and real instructions have joint format operand.
V7: Restored optionality of dfmt and nfmt in assembler.
V8: Addressed minor review comments.
Subscribers: arsenm, kzhuravl, wdng, nhaehnle, yaxunl, dstuttard, t-tye, llvm-commits
Differential Revision: https://reviews.llvm.org/D49026
Change-Id: If22ad77e349fac3a5d2f72dda53c010377d470d4
llvm-svn: 340268
2018-08-21 19:06:05 +08:00
|
|
|
define amdgpu_ps void @tbuffer_store_scalar_and_imm_offs(<4 x i32> inreg, <4 x float> %vdata, i32 inreg %soffset) {
|
|
|
|
|
main_body:
|
|
|
|
|
%in1 = bitcast <4 x float> %vdata to <4 x i32>
|
|
|
|
|
call void @llvm.amdgcn.raw.tbuffer.store.v4i32(<4 x i32> %in1, <4 x i32> %0, i32 42, i32 %soffset, i32 117, i32 0)
|
|
|
|
|
ret void
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
; GCN-LABEL: {{^}}buffer_store_ofs:
|
[AMDGPU][MC] Added support of SP3 syntax for MTBUF format modifier
Currently supported LLVM MTBUF syntax is shown below. It is not compatible with SP3.
op dst, addr, rsrc, FORMAT, soffset
This change adds support for SP3 syntax:
op dst, addr, rsrc, soffset SP3FORMAT
In addition to being compatible with SP3, this syntax allows using symbolic names for data, numeric and unified formats. Below is a list of added syntax variants.
format:<expression>
format:[<numeric-format-name>,<data-format-name>]
format:[<data-format-name>,<numeric-format-name>]
format:[<data-format-name>]
format:[<numeric-format-name>]
format:[<unified-format-name>]
The last syntax variant is supported for GFX10 only.
See llvm bug 37738
Reviewers: arsenm, rampitec, vpykhtin
Differential Revision: https://reviews.llvm.org/D84026
2020-07-24 21:39:42 +08:00
|
|
|
; PREGFX10: tbuffer_store_format_xyzw v[0:3], v4, s[0:3], 0 format:[BUF_DATA_FORMAT_8_8,BUF_NUM_FORMAT_FLOAT] offen
|
|
|
|
|
; GFX10: tbuffer_store_format_xyzw v[0:3], v4, s[0:3], 0 format:115 offen
|
[AMDGPU] New tbuffer intrinsics
Summary:
This commit adds new intrinsics
llvm.amdgcn.raw.tbuffer.load
llvm.amdgcn.struct.tbuffer.load
llvm.amdgcn.raw.tbuffer.store
llvm.amdgcn.struct.tbuffer.store
with the following changes from the llvm.amdgcn.tbuffer.* intrinsics:
* there are separate raw and struct versions: raw does not have an index
arg and sets idxen=0 in the instruction, and struct always sets
idxen=1 in the instruction even if the index is 0, to allow for the
fact that gfx9 does bounds checking differently depending on whether
idxen is set;
* there is a combined format arg (dfmt+nfmt)
* there is a combined cachepolicy arg (glc+slc)
* there are now only two offset args: one for the offset that is
included in bounds checking and swizzling, to be split between the
instruction's voffset and immoffset fields, and one for the offset
that is excluded from bounds checking and swizzling, to go into the
instruction's soffset field.
The AMDISD::TBUFFER_* SD nodes always have an index operand, all three
offset operands, combined format operand, combined cachepolicy operand,
and an extra idxen operand.
The tbuffer pseudo- and real instructions now also have a combined
format operand.
The obsolescent llvm.amdgcn.tbuffer.* and llvm.SI.tbuffer.store
intrinsics continue to work.
V2: Separate raw and struct intrinsics.
V3: Moved extract_glc and extract_slc defs to a more sensible place.
V4: Rebased on D49995.
V5: Only two separate offset args instead of three.
V6: Pseudo- and real instructions have joint format operand.
V7: Restored optionality of dfmt and nfmt in assembler.
V8: Addressed minor review comments.
Subscribers: arsenm, kzhuravl, wdng, nhaehnle, yaxunl, dstuttard, t-tye, llvm-commits
Differential Revision: https://reviews.llvm.org/D49026
Change-Id: If22ad77e349fac3a5d2f72dda53c010377d470d4
llvm-svn: 340268
2018-08-21 19:06:05 +08:00
|
|
|
define amdgpu_ps void @buffer_store_ofs(<4 x i32> inreg, <4 x float> %vdata, i32 %voffset) {
|
|
|
|
|
main_body:
|
|
|
|
|
%in1 = bitcast <4 x float> %vdata to <4 x i32>
|
|
|
|
|
call void @llvm.amdgcn.raw.tbuffer.store.v4i32(<4 x i32> %in1, <4 x i32> %0, i32 %voffset, i32 0, i32 115, i32 0)
|
|
|
|
|
ret void
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
; GCN-LABEL: {{^}}buffer_store_x1:
|
[AMDGPU][MC] Added support of SP3 syntax for MTBUF format modifier
Currently supported LLVM MTBUF syntax is shown below. It is not compatible with SP3.
op dst, addr, rsrc, FORMAT, soffset
This change adds support for SP3 syntax:
op dst, addr, rsrc, soffset SP3FORMAT
In addition to being compatible with SP3, this syntax allows using symbolic names for data, numeric and unified formats. Below is a list of added syntax variants.
format:<expression>
format:[<numeric-format-name>,<data-format-name>]
format:[<data-format-name>,<numeric-format-name>]
format:[<data-format-name>]
format:[<numeric-format-name>]
format:[<unified-format-name>]
The last syntax variant is supported for GFX10 only.
See llvm bug 37738
Reviewers: arsenm, rampitec, vpykhtin
Differential Revision: https://reviews.llvm.org/D84026
2020-07-24 21:39:42 +08:00
|
|
|
; PREGFX10: tbuffer_store_format_x v0, off, s[0:3], 0 format:[BUF_DATA_FORMAT_32_32_32,BUF_NUM_FORMAT_FLOAT]
|
|
|
|
|
; GFX10: tbuffer_store_format_x v0, off, s[0:3], 0 format:125
|
[AMDGPU] New tbuffer intrinsics
Summary:
This commit adds new intrinsics
llvm.amdgcn.raw.tbuffer.load
llvm.amdgcn.struct.tbuffer.load
llvm.amdgcn.raw.tbuffer.store
llvm.amdgcn.struct.tbuffer.store
with the following changes from the llvm.amdgcn.tbuffer.* intrinsics:
* there are separate raw and struct versions: raw does not have an index
arg and sets idxen=0 in the instruction, and struct always sets
idxen=1 in the instruction even if the index is 0, to allow for the
fact that gfx9 does bounds checking differently depending on whether
idxen is set;
* there is a combined format arg (dfmt+nfmt)
* there is a combined cachepolicy arg (glc+slc)
* there are now only two offset args: one for the offset that is
included in bounds checking and swizzling, to be split between the
instruction's voffset and immoffset fields, and one for the offset
that is excluded from bounds checking and swizzling, to go into the
instruction's soffset field.
The AMDISD::TBUFFER_* SD nodes always have an index operand, all three
offset operands, combined format operand, combined cachepolicy operand,
and an extra idxen operand.
The tbuffer pseudo- and real instructions now also have a combined
format operand.
The obsolescent llvm.amdgcn.tbuffer.* and llvm.SI.tbuffer.store
intrinsics continue to work.
V2: Separate raw and struct intrinsics.
V3: Moved extract_glc and extract_slc defs to a more sensible place.
V4: Rebased on D49995.
V5: Only two separate offset args instead of three.
V6: Pseudo- and real instructions have joint format operand.
V7: Restored optionality of dfmt and nfmt in assembler.
V8: Addressed minor review comments.
Subscribers: arsenm, kzhuravl, wdng, nhaehnle, yaxunl, dstuttard, t-tye, llvm-commits
Differential Revision: https://reviews.llvm.org/D49026
Change-Id: If22ad77e349fac3a5d2f72dda53c010377d470d4
llvm-svn: 340268
2018-08-21 19:06:05 +08:00
|
|
|
define amdgpu_ps void @buffer_store_x1(<4 x i32> inreg %rsrc, float %data) {
|
|
|
|
|
main_body:
|
|
|
|
|
%data.i = bitcast float %data to i32
|
|
|
|
|
call void @llvm.amdgcn.raw.tbuffer.store.i32(i32 %data.i, <4 x i32> %rsrc, i32 0, i32 0, i32 125, i32 0)
|
|
|
|
|
ret void
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
; GCN-LABEL: {{^}}buffer_store_x2:
|
[AMDGPU][MC] Added support of SP3 syntax for MTBUF format modifier
Currently supported LLVM MTBUF syntax is shown below. It is not compatible with SP3.
op dst, addr, rsrc, FORMAT, soffset
This change adds support for SP3 syntax:
op dst, addr, rsrc, soffset SP3FORMAT
In addition to being compatible with SP3, this syntax allows using symbolic names for data, numeric and unified formats. Below is a list of added syntax variants.
format:<expression>
format:[<numeric-format-name>,<data-format-name>]
format:[<data-format-name>,<numeric-format-name>]
format:[<data-format-name>]
format:[<numeric-format-name>]
format:[<unified-format-name>]
The last syntax variant is supported for GFX10 only.
See llvm bug 37738
Reviewers: arsenm, rampitec, vpykhtin
Differential Revision: https://reviews.llvm.org/D84026
2020-07-24 21:39:42 +08:00
|
|
|
; PREGFX10: tbuffer_store_format_xy v[0:1], off, s[0:3], 0 format:[BUF_NUM_FORMAT_USCALED]
|
|
|
|
|
; GFX10: tbuffer_store_format_xy v[0:1], off, s[0:3], 0 format:[BUF_FMT_10_11_11_SSCALED]
|
[AMDGPU] New tbuffer intrinsics
Summary:
This commit adds new intrinsics
llvm.amdgcn.raw.tbuffer.load
llvm.amdgcn.struct.tbuffer.load
llvm.amdgcn.raw.tbuffer.store
llvm.amdgcn.struct.tbuffer.store
with the following changes from the llvm.amdgcn.tbuffer.* intrinsics:
* there are separate raw and struct versions: raw does not have an index
arg and sets idxen=0 in the instruction, and struct always sets
idxen=1 in the instruction even if the index is 0, to allow for the
fact that gfx9 does bounds checking differently depending on whether
idxen is set;
* there is a combined format arg (dfmt+nfmt)
* there is a combined cachepolicy arg (glc+slc)
* there are now only two offset args: one for the offset that is
included in bounds checking and swizzling, to be split between the
instruction's voffset and immoffset fields, and one for the offset
that is excluded from bounds checking and swizzling, to go into the
instruction's soffset field.
The AMDISD::TBUFFER_* SD nodes always have an index operand, all three
offset operands, combined format operand, combined cachepolicy operand,
and an extra idxen operand.
The tbuffer pseudo- and real instructions now also have a combined
format operand.
The obsolescent llvm.amdgcn.tbuffer.* and llvm.SI.tbuffer.store
intrinsics continue to work.
V2: Separate raw and struct intrinsics.
V3: Moved extract_glc and extract_slc defs to a more sensible place.
V4: Rebased on D49995.
V5: Only two separate offset args instead of three.
V6: Pseudo- and real instructions have joint format operand.
V7: Restored optionality of dfmt and nfmt in assembler.
V8: Addressed minor review comments.
Subscribers: arsenm, kzhuravl, wdng, nhaehnle, yaxunl, dstuttard, t-tye, llvm-commits
Differential Revision: https://reviews.llvm.org/D49026
Change-Id: If22ad77e349fac3a5d2f72dda53c010377d470d4
llvm-svn: 340268
2018-08-21 19:06:05 +08:00
|
|
|
define amdgpu_ps void @buffer_store_x2(<4 x i32> inreg %rsrc, <2 x float> %data) {
|
|
|
|
|
main_body:
|
|
|
|
|
%data.i = bitcast <2 x float> %data to <2 x i32>
|
|
|
|
|
call void @llvm.amdgcn.raw.tbuffer.store.v2i32(<2 x i32> %data.i, <4 x i32> %rsrc, i32 0, i32 0, i32 33, i32 0)
|
|
|
|
|
ret void
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
declare void @llvm.amdgcn.raw.tbuffer.store.i32(i32, <4 x i32>, i32, i32, i32, i32) #0
|
|
|
|
|
declare void @llvm.amdgcn.raw.tbuffer.store.v2i32(<2 x i32>, <4 x i32>, i32, i32, i32, i32) #0
|
|
|
|
|
declare void @llvm.amdgcn.raw.tbuffer.store.v4i32(<4 x i32>, <4 x i32>, i32, i32, i32, i32) #0
|
|
|
|
|
declare void @llvm.amdgcn.raw.tbuffer.store.v4f32(<4 x float>, <4 x i32>, i32, i32, i32, i32) #0
|
|
|
|
|
|
|
|
|
|
attributes #0 = { nounwind }
|
|
|
|
|
attributes #1 = { nounwind readonly }
|