2015-01-07 02:00:21 +08:00
|
|
|
; RUN: llc -march=amdgcn -mcpu=bonaire -verify-machineinstrs -mattr=+load-store-opt -enable-misched < %s | FileCheck -strict-whitespace -check-prefix=SI %s
|
2014-10-11 06:01:59 +08:00
|
|
|
|
|
|
|
|
; FIXME: We don't get cases where the address was an SGPR because we
|
|
|
|
|
; get a copy to the address register for each one.
|
|
|
|
|
|
2014-11-14 03:56:13 +08:00
|
|
|
@lds = addrspace(3) global [512 x float] undef, align 4
|
|
|
|
|
@lds.f64 = addrspace(3) global [512 x double] undef, align 8
|
2014-10-11 06:01:59 +08:00
|
|
|
|
|
|
|
|
; SI-LABEL: @simple_read2_f32
|
2015-04-08 09:09:19 +08:00
|
|
|
; SI: ds_read2_b32 v{{\[}}[[LO_VREG:[0-9]+]]:[[HI_VREG:[0-9]+]]{{\]}}, v{{[0-9]+}} offset1:8
|
2014-11-05 22:50:53 +08:00
|
|
|
; SI: s_waitcnt lgkmcnt(0)
|
|
|
|
|
; SI: v_add_f32_e32 [[RESULT:v[0-9]+]], v[[HI_VREG]], v[[LO_VREG]]
|
|
|
|
|
; SI: buffer_store_dword [[RESULT]]
|
|
|
|
|
; SI: s_endpgm
|
2014-10-11 06:01:59 +08:00
|
|
|
define void @simple_read2_f32(float addrspace(1)* %out) #0 {
|
|
|
|
|
%x.i = tail call i32 @llvm.r600.read.tidig.x() #1
|
[opaque pointer type] Add textual IR support for explicit type parameter to getelementptr instruction
One of several parallel first steps to remove the target type of pointers,
replacing them with a single opaque pointer type.
This adds an explicit type parameter to the gep instruction so that when the
first parameter becomes an opaque pointer type, the type to gep through is
still available to the instructions.
* This doesn't modify gep operators, only instructions (operators will be
handled separately)
* Textual IR changes only. Bitcode (including upgrade) and changing the
in-memory representation will be in separate changes.
* geps of vectors are transformed as:
getelementptr <4 x float*> %x, ...
->getelementptr float, <4 x float*> %x, ...
Then, once the opaque pointer type is introduced, this will ultimately look
like:
getelementptr float, <4 x ptr> %x
with the unambiguous interpretation that it is a vector of pointers to float.
* address spaces remain on the pointer, not the type:
getelementptr float addrspace(1)* %x
->getelementptr float, float addrspace(1)* %x
Then, eventually:
getelementptr float, ptr addrspace(1) %x
Importantly, the massive amount of test case churn has been automated by
same crappy python code. I had to manually update a few test cases that
wouldn't fit the script's model (r228970,r229196,r229197,r229198). The
python script just massages stdin and writes the result to stdout, I
then wrapped that in a shell script to handle replacing files, then
using the usual find+xargs to migrate all the files.
update.py:
import fileinput
import sys
import re
ibrep = re.compile(r"(^.*?[^%\w]getelementptr inbounds )(((?:<\d* x )?)(.*?)(| addrspace\(\d\)) *\*(|>)(?:$| *(?:%|@|null|undef|blockaddress|getelementptr|addrspacecast|bitcast|inttoptr|\[\[[a-zA-Z]|\{\{).*$))")
normrep = re.compile( r"(^.*?[^%\w]getelementptr )(((?:<\d* x )?)(.*?)(| addrspace\(\d\)) *\*(|>)(?:$| *(?:%|@|null|undef|blockaddress|getelementptr|addrspacecast|bitcast|inttoptr|\[\[[a-zA-Z]|\{\{).*$))")
def conv(match, line):
if not match:
return line
line = match.groups()[0]
if len(match.groups()[5]) == 0:
line += match.groups()[2]
line += match.groups()[3]
line += ", "
line += match.groups()[1]
line += "\n"
return line
for line in sys.stdin:
if line.find("getelementptr ") == line.find("getelementptr inbounds"):
if line.find("getelementptr inbounds") != line.find("getelementptr inbounds ("):
line = conv(re.match(ibrep, line), line)
elif line.find("getelementptr ") != line.find("getelementptr ("):
line = conv(re.match(normrep, line), line)
sys.stdout.write(line)
apply.sh:
for name in "$@"
do
python3 `dirname "$0"`/update.py < "$name" > "$name.tmp" && mv "$name.tmp" "$name"
rm -f "$name.tmp"
done
The actual commands:
From llvm/src:
find test/ -name *.ll | xargs ./apply.sh
From llvm/src/tools/clang:
find test/ -name *.mm -o -name *.m -o -name *.cpp -o -name *.c | xargs -I '{}' ../../apply.sh "{}"
From llvm/src/tools/polly:
find test/ -name *.ll | xargs ./apply.sh
After that, check-all (with llvm, clang, clang-tools-extra, lld,
compiler-rt, and polly all checked out).
The extra 'rm' in the apply.sh script is due to a few files in clang's test
suite using interesting unicode stuff that my python script was throwing
exceptions on. None of those files needed to be migrated, so it seemed
sufficient to ignore those cases.
Reviewers: rafael, dexonsmith, grosser
Differential Revision: http://reviews.llvm.org/D7636
llvm-svn: 230786
2015-02-28 03:29:02 +08:00
|
|
|
%arrayidx0 = getelementptr inbounds [512 x float], [512 x float] addrspace(3)* @lds, i32 0, i32 %x.i
|
2015-02-28 05:17:42 +08:00
|
|
|
%val0 = load float, float addrspace(3)* %arrayidx0, align 4
|
2014-10-11 06:01:59 +08:00
|
|
|
%add.x = add nsw i32 %x.i, 8
|
[opaque pointer type] Add textual IR support for explicit type parameter to getelementptr instruction
One of several parallel first steps to remove the target type of pointers,
replacing them with a single opaque pointer type.
This adds an explicit type parameter to the gep instruction so that when the
first parameter becomes an opaque pointer type, the type to gep through is
still available to the instructions.
* This doesn't modify gep operators, only instructions (operators will be
handled separately)
* Textual IR changes only. Bitcode (including upgrade) and changing the
in-memory representation will be in separate changes.
* geps of vectors are transformed as:
getelementptr <4 x float*> %x, ...
->getelementptr float, <4 x float*> %x, ...
Then, once the opaque pointer type is introduced, this will ultimately look
like:
getelementptr float, <4 x ptr> %x
with the unambiguous interpretation that it is a vector of pointers to float.
* address spaces remain on the pointer, not the type:
getelementptr float addrspace(1)* %x
->getelementptr float, float addrspace(1)* %x
Then, eventually:
getelementptr float, ptr addrspace(1) %x
Importantly, the massive amount of test case churn has been automated by
same crappy python code. I had to manually update a few test cases that
wouldn't fit the script's model (r228970,r229196,r229197,r229198). The
python script just massages stdin and writes the result to stdout, I
then wrapped that in a shell script to handle replacing files, then
using the usual find+xargs to migrate all the files.
update.py:
import fileinput
import sys
import re
ibrep = re.compile(r"(^.*?[^%\w]getelementptr inbounds )(((?:<\d* x )?)(.*?)(| addrspace\(\d\)) *\*(|>)(?:$| *(?:%|@|null|undef|blockaddress|getelementptr|addrspacecast|bitcast|inttoptr|\[\[[a-zA-Z]|\{\{).*$))")
normrep = re.compile( r"(^.*?[^%\w]getelementptr )(((?:<\d* x )?)(.*?)(| addrspace\(\d\)) *\*(|>)(?:$| *(?:%|@|null|undef|blockaddress|getelementptr|addrspacecast|bitcast|inttoptr|\[\[[a-zA-Z]|\{\{).*$))")
def conv(match, line):
if not match:
return line
line = match.groups()[0]
if len(match.groups()[5]) == 0:
line += match.groups()[2]
line += match.groups()[3]
line += ", "
line += match.groups()[1]
line += "\n"
return line
for line in sys.stdin:
if line.find("getelementptr ") == line.find("getelementptr inbounds"):
if line.find("getelementptr inbounds") != line.find("getelementptr inbounds ("):
line = conv(re.match(ibrep, line), line)
elif line.find("getelementptr ") != line.find("getelementptr ("):
line = conv(re.match(normrep, line), line)
sys.stdout.write(line)
apply.sh:
for name in "$@"
do
python3 `dirname "$0"`/update.py < "$name" > "$name.tmp" && mv "$name.tmp" "$name"
rm -f "$name.tmp"
done
The actual commands:
From llvm/src:
find test/ -name *.ll | xargs ./apply.sh
From llvm/src/tools/clang:
find test/ -name *.mm -o -name *.m -o -name *.cpp -o -name *.c | xargs -I '{}' ../../apply.sh "{}"
From llvm/src/tools/polly:
find test/ -name *.ll | xargs ./apply.sh
After that, check-all (with llvm, clang, clang-tools-extra, lld,
compiler-rt, and polly all checked out).
The extra 'rm' in the apply.sh script is due to a few files in clang's test
suite using interesting unicode stuff that my python script was throwing
exceptions on. None of those files needed to be migrated, so it seemed
sufficient to ignore those cases.
Reviewers: rafael, dexonsmith, grosser
Differential Revision: http://reviews.llvm.org/D7636
llvm-svn: 230786
2015-02-28 03:29:02 +08:00
|
|
|
%arrayidx1 = getelementptr inbounds [512 x float], [512 x float] addrspace(3)* @lds, i32 0, i32 %add.x
|
2015-02-28 05:17:42 +08:00
|
|
|
%val1 = load float, float addrspace(3)* %arrayidx1, align 4
|
2014-10-11 06:01:59 +08:00
|
|
|
%sum = fadd float %val0, %val1
|
[opaque pointer type] Add textual IR support for explicit type parameter to getelementptr instruction
One of several parallel first steps to remove the target type of pointers,
replacing them with a single opaque pointer type.
This adds an explicit type parameter to the gep instruction so that when the
first parameter becomes an opaque pointer type, the type to gep through is
still available to the instructions.
* This doesn't modify gep operators, only instructions (operators will be
handled separately)
* Textual IR changes only. Bitcode (including upgrade) and changing the
in-memory representation will be in separate changes.
* geps of vectors are transformed as:
getelementptr <4 x float*> %x, ...
->getelementptr float, <4 x float*> %x, ...
Then, once the opaque pointer type is introduced, this will ultimately look
like:
getelementptr float, <4 x ptr> %x
with the unambiguous interpretation that it is a vector of pointers to float.
* address spaces remain on the pointer, not the type:
getelementptr float addrspace(1)* %x
->getelementptr float, float addrspace(1)* %x
Then, eventually:
getelementptr float, ptr addrspace(1) %x
Importantly, the massive amount of test case churn has been automated by
same crappy python code. I had to manually update a few test cases that
wouldn't fit the script's model (r228970,r229196,r229197,r229198). The
python script just massages stdin and writes the result to stdout, I
then wrapped that in a shell script to handle replacing files, then
using the usual find+xargs to migrate all the files.
update.py:
import fileinput
import sys
import re
ibrep = re.compile(r"(^.*?[^%\w]getelementptr inbounds )(((?:<\d* x )?)(.*?)(| addrspace\(\d\)) *\*(|>)(?:$| *(?:%|@|null|undef|blockaddress|getelementptr|addrspacecast|bitcast|inttoptr|\[\[[a-zA-Z]|\{\{).*$))")
normrep = re.compile( r"(^.*?[^%\w]getelementptr )(((?:<\d* x )?)(.*?)(| addrspace\(\d\)) *\*(|>)(?:$| *(?:%|@|null|undef|blockaddress|getelementptr|addrspacecast|bitcast|inttoptr|\[\[[a-zA-Z]|\{\{).*$))")
def conv(match, line):
if not match:
return line
line = match.groups()[0]
if len(match.groups()[5]) == 0:
line += match.groups()[2]
line += match.groups()[3]
line += ", "
line += match.groups()[1]
line += "\n"
return line
for line in sys.stdin:
if line.find("getelementptr ") == line.find("getelementptr inbounds"):
if line.find("getelementptr inbounds") != line.find("getelementptr inbounds ("):
line = conv(re.match(ibrep, line), line)
elif line.find("getelementptr ") != line.find("getelementptr ("):
line = conv(re.match(normrep, line), line)
sys.stdout.write(line)
apply.sh:
for name in "$@"
do
python3 `dirname "$0"`/update.py < "$name" > "$name.tmp" && mv "$name.tmp" "$name"
rm -f "$name.tmp"
done
The actual commands:
From llvm/src:
find test/ -name *.ll | xargs ./apply.sh
From llvm/src/tools/clang:
find test/ -name *.mm -o -name *.m -o -name *.cpp -o -name *.c | xargs -I '{}' ../../apply.sh "{}"
From llvm/src/tools/polly:
find test/ -name *.ll | xargs ./apply.sh
After that, check-all (with llvm, clang, clang-tools-extra, lld,
compiler-rt, and polly all checked out).
The extra 'rm' in the apply.sh script is due to a few files in clang's test
suite using interesting unicode stuff that my python script was throwing
exceptions on. None of those files needed to be migrated, so it seemed
sufficient to ignore those cases.
Reviewers: rafael, dexonsmith, grosser
Differential Revision: http://reviews.llvm.org/D7636
llvm-svn: 230786
2015-02-28 03:29:02 +08:00
|
|
|
%out.gep = getelementptr inbounds float, float addrspace(1)* %out, i32 %x.i
|
2014-10-11 06:01:59 +08:00
|
|
|
store float %sum, float addrspace(1)* %out.gep, align 4
|
|
|
|
|
ret void
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
; SI-LABEL: @simple_read2_f32_max_offset
|
2015-04-08 09:09:19 +08:00
|
|
|
; SI: ds_read2_b32 v{{\[}}[[LO_VREG:[0-9]+]]:[[HI_VREG:[0-9]+]]{{\]}}, v{{[0-9]+}} offset1:255
|
2014-11-05 22:50:53 +08:00
|
|
|
; SI: s_waitcnt lgkmcnt(0)
|
|
|
|
|
; SI: v_add_f32_e32 [[RESULT:v[0-9]+]], v[[HI_VREG]], v[[LO_VREG]]
|
|
|
|
|
; SI: buffer_store_dword [[RESULT]]
|
|
|
|
|
; SI: s_endpgm
|
2014-10-11 06:01:59 +08:00
|
|
|
define void @simple_read2_f32_max_offset(float addrspace(1)* %out) #0 {
|
|
|
|
|
%x.i = tail call i32 @llvm.r600.read.tidig.x() #1
|
[opaque pointer type] Add textual IR support for explicit type parameter to getelementptr instruction
One of several parallel first steps to remove the target type of pointers,
replacing them with a single opaque pointer type.
This adds an explicit type parameter to the gep instruction so that when the
first parameter becomes an opaque pointer type, the type to gep through is
still available to the instructions.
* This doesn't modify gep operators, only instructions (operators will be
handled separately)
* Textual IR changes only. Bitcode (including upgrade) and changing the
in-memory representation will be in separate changes.
* geps of vectors are transformed as:
getelementptr <4 x float*> %x, ...
->getelementptr float, <4 x float*> %x, ...
Then, once the opaque pointer type is introduced, this will ultimately look
like:
getelementptr float, <4 x ptr> %x
with the unambiguous interpretation that it is a vector of pointers to float.
* address spaces remain on the pointer, not the type:
getelementptr float addrspace(1)* %x
->getelementptr float, float addrspace(1)* %x
Then, eventually:
getelementptr float, ptr addrspace(1) %x
Importantly, the massive amount of test case churn has been automated by
same crappy python code. I had to manually update a few test cases that
wouldn't fit the script's model (r228970,r229196,r229197,r229198). The
python script just massages stdin and writes the result to stdout, I
then wrapped that in a shell script to handle replacing files, then
using the usual find+xargs to migrate all the files.
update.py:
import fileinput
import sys
import re
ibrep = re.compile(r"(^.*?[^%\w]getelementptr inbounds )(((?:<\d* x )?)(.*?)(| addrspace\(\d\)) *\*(|>)(?:$| *(?:%|@|null|undef|blockaddress|getelementptr|addrspacecast|bitcast|inttoptr|\[\[[a-zA-Z]|\{\{).*$))")
normrep = re.compile( r"(^.*?[^%\w]getelementptr )(((?:<\d* x )?)(.*?)(| addrspace\(\d\)) *\*(|>)(?:$| *(?:%|@|null|undef|blockaddress|getelementptr|addrspacecast|bitcast|inttoptr|\[\[[a-zA-Z]|\{\{).*$))")
def conv(match, line):
if not match:
return line
line = match.groups()[0]
if len(match.groups()[5]) == 0:
line += match.groups()[2]
line += match.groups()[3]
line += ", "
line += match.groups()[1]
line += "\n"
return line
for line in sys.stdin:
if line.find("getelementptr ") == line.find("getelementptr inbounds"):
if line.find("getelementptr inbounds") != line.find("getelementptr inbounds ("):
line = conv(re.match(ibrep, line), line)
elif line.find("getelementptr ") != line.find("getelementptr ("):
line = conv(re.match(normrep, line), line)
sys.stdout.write(line)
apply.sh:
for name in "$@"
do
python3 `dirname "$0"`/update.py < "$name" > "$name.tmp" && mv "$name.tmp" "$name"
rm -f "$name.tmp"
done
The actual commands:
From llvm/src:
find test/ -name *.ll | xargs ./apply.sh
From llvm/src/tools/clang:
find test/ -name *.mm -o -name *.m -o -name *.cpp -o -name *.c | xargs -I '{}' ../../apply.sh "{}"
From llvm/src/tools/polly:
find test/ -name *.ll | xargs ./apply.sh
After that, check-all (with llvm, clang, clang-tools-extra, lld,
compiler-rt, and polly all checked out).
The extra 'rm' in the apply.sh script is due to a few files in clang's test
suite using interesting unicode stuff that my python script was throwing
exceptions on. None of those files needed to be migrated, so it seemed
sufficient to ignore those cases.
Reviewers: rafael, dexonsmith, grosser
Differential Revision: http://reviews.llvm.org/D7636
llvm-svn: 230786
2015-02-28 03:29:02 +08:00
|
|
|
%arrayidx0 = getelementptr inbounds [512 x float], [512 x float] addrspace(3)* @lds, i32 0, i32 %x.i
|
2015-02-28 05:17:42 +08:00
|
|
|
%val0 = load float, float addrspace(3)* %arrayidx0, align 4
|
2014-10-11 06:01:59 +08:00
|
|
|
%add.x = add nsw i32 %x.i, 255
|
[opaque pointer type] Add textual IR support for explicit type parameter to getelementptr instruction
One of several parallel first steps to remove the target type of pointers,
replacing them with a single opaque pointer type.
This adds an explicit type parameter to the gep instruction so that when the
first parameter becomes an opaque pointer type, the type to gep through is
still available to the instructions.
* This doesn't modify gep operators, only instructions (operators will be
handled separately)
* Textual IR changes only. Bitcode (including upgrade) and changing the
in-memory representation will be in separate changes.
* geps of vectors are transformed as:
getelementptr <4 x float*> %x, ...
->getelementptr float, <4 x float*> %x, ...
Then, once the opaque pointer type is introduced, this will ultimately look
like:
getelementptr float, <4 x ptr> %x
with the unambiguous interpretation that it is a vector of pointers to float.
* address spaces remain on the pointer, not the type:
getelementptr float addrspace(1)* %x
->getelementptr float, float addrspace(1)* %x
Then, eventually:
getelementptr float, ptr addrspace(1) %x
Importantly, the massive amount of test case churn has been automated by
same crappy python code. I had to manually update a few test cases that
wouldn't fit the script's model (r228970,r229196,r229197,r229198). The
python script just massages stdin and writes the result to stdout, I
then wrapped that in a shell script to handle replacing files, then
using the usual find+xargs to migrate all the files.
update.py:
import fileinput
import sys
import re
ibrep = re.compile(r"(^.*?[^%\w]getelementptr inbounds )(((?:<\d* x )?)(.*?)(| addrspace\(\d\)) *\*(|>)(?:$| *(?:%|@|null|undef|blockaddress|getelementptr|addrspacecast|bitcast|inttoptr|\[\[[a-zA-Z]|\{\{).*$))")
normrep = re.compile( r"(^.*?[^%\w]getelementptr )(((?:<\d* x )?)(.*?)(| addrspace\(\d\)) *\*(|>)(?:$| *(?:%|@|null|undef|blockaddress|getelementptr|addrspacecast|bitcast|inttoptr|\[\[[a-zA-Z]|\{\{).*$))")
def conv(match, line):
if not match:
return line
line = match.groups()[0]
if len(match.groups()[5]) == 0:
line += match.groups()[2]
line += match.groups()[3]
line += ", "
line += match.groups()[1]
line += "\n"
return line
for line in sys.stdin:
if line.find("getelementptr ") == line.find("getelementptr inbounds"):
if line.find("getelementptr inbounds") != line.find("getelementptr inbounds ("):
line = conv(re.match(ibrep, line), line)
elif line.find("getelementptr ") != line.find("getelementptr ("):
line = conv(re.match(normrep, line), line)
sys.stdout.write(line)
apply.sh:
for name in "$@"
do
python3 `dirname "$0"`/update.py < "$name" > "$name.tmp" && mv "$name.tmp" "$name"
rm -f "$name.tmp"
done
The actual commands:
From llvm/src:
find test/ -name *.ll | xargs ./apply.sh
From llvm/src/tools/clang:
find test/ -name *.mm -o -name *.m -o -name *.cpp -o -name *.c | xargs -I '{}' ../../apply.sh "{}"
From llvm/src/tools/polly:
find test/ -name *.ll | xargs ./apply.sh
After that, check-all (with llvm, clang, clang-tools-extra, lld,
compiler-rt, and polly all checked out).
The extra 'rm' in the apply.sh script is due to a few files in clang's test
suite using interesting unicode stuff that my python script was throwing
exceptions on. None of those files needed to be migrated, so it seemed
sufficient to ignore those cases.
Reviewers: rafael, dexonsmith, grosser
Differential Revision: http://reviews.llvm.org/D7636
llvm-svn: 230786
2015-02-28 03:29:02 +08:00
|
|
|
%arrayidx1 = getelementptr inbounds [512 x float], [512 x float] addrspace(3)* @lds, i32 0, i32 %add.x
|
2015-02-28 05:17:42 +08:00
|
|
|
%val1 = load float, float addrspace(3)* %arrayidx1, align 4
|
2014-10-11 06:01:59 +08:00
|
|
|
%sum = fadd float %val0, %val1
|
[opaque pointer type] Add textual IR support for explicit type parameter to getelementptr instruction
One of several parallel first steps to remove the target type of pointers,
replacing them with a single opaque pointer type.
This adds an explicit type parameter to the gep instruction so that when the
first parameter becomes an opaque pointer type, the type to gep through is
still available to the instructions.
* This doesn't modify gep operators, only instructions (operators will be
handled separately)
* Textual IR changes only. Bitcode (including upgrade) and changing the
in-memory representation will be in separate changes.
* geps of vectors are transformed as:
getelementptr <4 x float*> %x, ...
->getelementptr float, <4 x float*> %x, ...
Then, once the opaque pointer type is introduced, this will ultimately look
like:
getelementptr float, <4 x ptr> %x
with the unambiguous interpretation that it is a vector of pointers to float.
* address spaces remain on the pointer, not the type:
getelementptr float addrspace(1)* %x
->getelementptr float, float addrspace(1)* %x
Then, eventually:
getelementptr float, ptr addrspace(1) %x
Importantly, the massive amount of test case churn has been automated by
same crappy python code. I had to manually update a few test cases that
wouldn't fit the script's model (r228970,r229196,r229197,r229198). The
python script just massages stdin and writes the result to stdout, I
then wrapped that in a shell script to handle replacing files, then
using the usual find+xargs to migrate all the files.
update.py:
import fileinput
import sys
import re
ibrep = re.compile(r"(^.*?[^%\w]getelementptr inbounds )(((?:<\d* x )?)(.*?)(| addrspace\(\d\)) *\*(|>)(?:$| *(?:%|@|null|undef|blockaddress|getelementptr|addrspacecast|bitcast|inttoptr|\[\[[a-zA-Z]|\{\{).*$))")
normrep = re.compile( r"(^.*?[^%\w]getelementptr )(((?:<\d* x )?)(.*?)(| addrspace\(\d\)) *\*(|>)(?:$| *(?:%|@|null|undef|blockaddress|getelementptr|addrspacecast|bitcast|inttoptr|\[\[[a-zA-Z]|\{\{).*$))")
def conv(match, line):
if not match:
return line
line = match.groups()[0]
if len(match.groups()[5]) == 0:
line += match.groups()[2]
line += match.groups()[3]
line += ", "
line += match.groups()[1]
line += "\n"
return line
for line in sys.stdin:
if line.find("getelementptr ") == line.find("getelementptr inbounds"):
if line.find("getelementptr inbounds") != line.find("getelementptr inbounds ("):
line = conv(re.match(ibrep, line), line)
elif line.find("getelementptr ") != line.find("getelementptr ("):
line = conv(re.match(normrep, line), line)
sys.stdout.write(line)
apply.sh:
for name in "$@"
do
python3 `dirname "$0"`/update.py < "$name" > "$name.tmp" && mv "$name.tmp" "$name"
rm -f "$name.tmp"
done
The actual commands:
From llvm/src:
find test/ -name *.ll | xargs ./apply.sh
From llvm/src/tools/clang:
find test/ -name *.mm -o -name *.m -o -name *.cpp -o -name *.c | xargs -I '{}' ../../apply.sh "{}"
From llvm/src/tools/polly:
find test/ -name *.ll | xargs ./apply.sh
After that, check-all (with llvm, clang, clang-tools-extra, lld,
compiler-rt, and polly all checked out).
The extra 'rm' in the apply.sh script is due to a few files in clang's test
suite using interesting unicode stuff that my python script was throwing
exceptions on. None of those files needed to be migrated, so it seemed
sufficient to ignore those cases.
Reviewers: rafael, dexonsmith, grosser
Differential Revision: http://reviews.llvm.org/D7636
llvm-svn: 230786
2015-02-28 03:29:02 +08:00
|
|
|
%out.gep = getelementptr inbounds float, float addrspace(1)* %out, i32 %x.i
|
2014-10-11 06:01:59 +08:00
|
|
|
store float %sum, float addrspace(1)* %out.gep, align 4
|
|
|
|
|
ret void
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
; SI-LABEL: @simple_read2_f32_too_far
|
2014-11-05 22:50:53 +08:00
|
|
|
; SI-NOT ds_read2_b32
|
|
|
|
|
; SI: ds_read_b32 v{{[0-9]+}}, v{{[0-9]+}}
|
|
|
|
|
; SI: ds_read_b32 v{{[0-9]+}}, v{{[0-9]+}} offset:1028
|
|
|
|
|
; SI: s_endpgm
|
2014-10-11 06:01:59 +08:00
|
|
|
define void @simple_read2_f32_too_far(float addrspace(1)* %out) #0 {
|
|
|
|
|
%x.i = tail call i32 @llvm.r600.read.tidig.x() #1
|
[opaque pointer type] Add textual IR support for explicit type parameter to getelementptr instruction
One of several parallel first steps to remove the target type of pointers,
replacing them with a single opaque pointer type.
This adds an explicit type parameter to the gep instruction so that when the
first parameter becomes an opaque pointer type, the type to gep through is
still available to the instructions.
* This doesn't modify gep operators, only instructions (operators will be
handled separately)
* Textual IR changes only. Bitcode (including upgrade) and changing the
in-memory representation will be in separate changes.
* geps of vectors are transformed as:
getelementptr <4 x float*> %x, ...
->getelementptr float, <4 x float*> %x, ...
Then, once the opaque pointer type is introduced, this will ultimately look
like:
getelementptr float, <4 x ptr> %x
with the unambiguous interpretation that it is a vector of pointers to float.
* address spaces remain on the pointer, not the type:
getelementptr float addrspace(1)* %x
->getelementptr float, float addrspace(1)* %x
Then, eventually:
getelementptr float, ptr addrspace(1) %x
Importantly, the massive amount of test case churn has been automated by
same crappy python code. I had to manually update a few test cases that
wouldn't fit the script's model (r228970,r229196,r229197,r229198). The
python script just massages stdin and writes the result to stdout, I
then wrapped that in a shell script to handle replacing files, then
using the usual find+xargs to migrate all the files.
update.py:
import fileinput
import sys
import re
ibrep = re.compile(r"(^.*?[^%\w]getelementptr inbounds )(((?:<\d* x )?)(.*?)(| addrspace\(\d\)) *\*(|>)(?:$| *(?:%|@|null|undef|blockaddress|getelementptr|addrspacecast|bitcast|inttoptr|\[\[[a-zA-Z]|\{\{).*$))")
normrep = re.compile( r"(^.*?[^%\w]getelementptr )(((?:<\d* x )?)(.*?)(| addrspace\(\d\)) *\*(|>)(?:$| *(?:%|@|null|undef|blockaddress|getelementptr|addrspacecast|bitcast|inttoptr|\[\[[a-zA-Z]|\{\{).*$))")
def conv(match, line):
if not match:
return line
line = match.groups()[0]
if len(match.groups()[5]) == 0:
line += match.groups()[2]
line += match.groups()[3]
line += ", "
line += match.groups()[1]
line += "\n"
return line
for line in sys.stdin:
if line.find("getelementptr ") == line.find("getelementptr inbounds"):
if line.find("getelementptr inbounds") != line.find("getelementptr inbounds ("):
line = conv(re.match(ibrep, line), line)
elif line.find("getelementptr ") != line.find("getelementptr ("):
line = conv(re.match(normrep, line), line)
sys.stdout.write(line)
apply.sh:
for name in "$@"
do
python3 `dirname "$0"`/update.py < "$name" > "$name.tmp" && mv "$name.tmp" "$name"
rm -f "$name.tmp"
done
The actual commands:
From llvm/src:
find test/ -name *.ll | xargs ./apply.sh
From llvm/src/tools/clang:
find test/ -name *.mm -o -name *.m -o -name *.cpp -o -name *.c | xargs -I '{}' ../../apply.sh "{}"
From llvm/src/tools/polly:
find test/ -name *.ll | xargs ./apply.sh
After that, check-all (with llvm, clang, clang-tools-extra, lld,
compiler-rt, and polly all checked out).
The extra 'rm' in the apply.sh script is due to a few files in clang's test
suite using interesting unicode stuff that my python script was throwing
exceptions on. None of those files needed to be migrated, so it seemed
sufficient to ignore those cases.
Reviewers: rafael, dexonsmith, grosser
Differential Revision: http://reviews.llvm.org/D7636
llvm-svn: 230786
2015-02-28 03:29:02 +08:00
|
|
|
%arrayidx0 = getelementptr inbounds [512 x float], [512 x float] addrspace(3)* @lds, i32 0, i32 %x.i
|
2015-02-28 05:17:42 +08:00
|
|
|
%val0 = load float, float addrspace(3)* %arrayidx0, align 4
|
2014-10-11 06:12:32 +08:00
|
|
|
%add.x = add nsw i32 %x.i, 257
|
[opaque pointer type] Add textual IR support for explicit type parameter to getelementptr instruction
One of several parallel first steps to remove the target type of pointers,
replacing them with a single opaque pointer type.
This adds an explicit type parameter to the gep instruction so that when the
first parameter becomes an opaque pointer type, the type to gep through is
still available to the instructions.
* This doesn't modify gep operators, only instructions (operators will be
handled separately)
* Textual IR changes only. Bitcode (including upgrade) and changing the
in-memory representation will be in separate changes.
* geps of vectors are transformed as:
getelementptr <4 x float*> %x, ...
->getelementptr float, <4 x float*> %x, ...
Then, once the opaque pointer type is introduced, this will ultimately look
like:
getelementptr float, <4 x ptr> %x
with the unambiguous interpretation that it is a vector of pointers to float.
* address spaces remain on the pointer, not the type:
getelementptr float addrspace(1)* %x
->getelementptr float, float addrspace(1)* %x
Then, eventually:
getelementptr float, ptr addrspace(1) %x
Importantly, the massive amount of test case churn has been automated by
same crappy python code. I had to manually update a few test cases that
wouldn't fit the script's model (r228970,r229196,r229197,r229198). The
python script just massages stdin and writes the result to stdout, I
then wrapped that in a shell script to handle replacing files, then
using the usual find+xargs to migrate all the files.
update.py:
import fileinput
import sys
import re
ibrep = re.compile(r"(^.*?[^%\w]getelementptr inbounds )(((?:<\d* x )?)(.*?)(| addrspace\(\d\)) *\*(|>)(?:$| *(?:%|@|null|undef|blockaddress|getelementptr|addrspacecast|bitcast|inttoptr|\[\[[a-zA-Z]|\{\{).*$))")
normrep = re.compile( r"(^.*?[^%\w]getelementptr )(((?:<\d* x )?)(.*?)(| addrspace\(\d\)) *\*(|>)(?:$| *(?:%|@|null|undef|blockaddress|getelementptr|addrspacecast|bitcast|inttoptr|\[\[[a-zA-Z]|\{\{).*$))")
def conv(match, line):
if not match:
return line
line = match.groups()[0]
if len(match.groups()[5]) == 0:
line += match.groups()[2]
line += match.groups()[3]
line += ", "
line += match.groups()[1]
line += "\n"
return line
for line in sys.stdin:
if line.find("getelementptr ") == line.find("getelementptr inbounds"):
if line.find("getelementptr inbounds") != line.find("getelementptr inbounds ("):
line = conv(re.match(ibrep, line), line)
elif line.find("getelementptr ") != line.find("getelementptr ("):
line = conv(re.match(normrep, line), line)
sys.stdout.write(line)
apply.sh:
for name in "$@"
do
python3 `dirname "$0"`/update.py < "$name" > "$name.tmp" && mv "$name.tmp" "$name"
rm -f "$name.tmp"
done
The actual commands:
From llvm/src:
find test/ -name *.ll | xargs ./apply.sh
From llvm/src/tools/clang:
find test/ -name *.mm -o -name *.m -o -name *.cpp -o -name *.c | xargs -I '{}' ../../apply.sh "{}"
From llvm/src/tools/polly:
find test/ -name *.ll | xargs ./apply.sh
After that, check-all (with llvm, clang, clang-tools-extra, lld,
compiler-rt, and polly all checked out).
The extra 'rm' in the apply.sh script is due to a few files in clang's test
suite using interesting unicode stuff that my python script was throwing
exceptions on. None of those files needed to be migrated, so it seemed
sufficient to ignore those cases.
Reviewers: rafael, dexonsmith, grosser
Differential Revision: http://reviews.llvm.org/D7636
llvm-svn: 230786
2015-02-28 03:29:02 +08:00
|
|
|
%arrayidx1 = getelementptr inbounds [512 x float], [512 x float] addrspace(3)* @lds, i32 0, i32 %add.x
|
2015-02-28 05:17:42 +08:00
|
|
|
%val1 = load float, float addrspace(3)* %arrayidx1, align 4
|
2014-10-11 06:01:59 +08:00
|
|
|
%sum = fadd float %val0, %val1
|
[opaque pointer type] Add textual IR support for explicit type parameter to getelementptr instruction
One of several parallel first steps to remove the target type of pointers,
replacing them with a single opaque pointer type.
This adds an explicit type parameter to the gep instruction so that when the
first parameter becomes an opaque pointer type, the type to gep through is
still available to the instructions.
* This doesn't modify gep operators, only instructions (operators will be
handled separately)
* Textual IR changes only. Bitcode (including upgrade) and changing the
in-memory representation will be in separate changes.
* geps of vectors are transformed as:
getelementptr <4 x float*> %x, ...
->getelementptr float, <4 x float*> %x, ...
Then, once the opaque pointer type is introduced, this will ultimately look
like:
getelementptr float, <4 x ptr> %x
with the unambiguous interpretation that it is a vector of pointers to float.
* address spaces remain on the pointer, not the type:
getelementptr float addrspace(1)* %x
->getelementptr float, float addrspace(1)* %x
Then, eventually:
getelementptr float, ptr addrspace(1) %x
Importantly, the massive amount of test case churn has been automated by
same crappy python code. I had to manually update a few test cases that
wouldn't fit the script's model (r228970,r229196,r229197,r229198). The
python script just massages stdin and writes the result to stdout, I
then wrapped that in a shell script to handle replacing files, then
using the usual find+xargs to migrate all the files.
update.py:
import fileinput
import sys
import re
ibrep = re.compile(r"(^.*?[^%\w]getelementptr inbounds )(((?:<\d* x )?)(.*?)(| addrspace\(\d\)) *\*(|>)(?:$| *(?:%|@|null|undef|blockaddress|getelementptr|addrspacecast|bitcast|inttoptr|\[\[[a-zA-Z]|\{\{).*$))")
normrep = re.compile( r"(^.*?[^%\w]getelementptr )(((?:<\d* x )?)(.*?)(| addrspace\(\d\)) *\*(|>)(?:$| *(?:%|@|null|undef|blockaddress|getelementptr|addrspacecast|bitcast|inttoptr|\[\[[a-zA-Z]|\{\{).*$))")
def conv(match, line):
if not match:
return line
line = match.groups()[0]
if len(match.groups()[5]) == 0:
line += match.groups()[2]
line += match.groups()[3]
line += ", "
line += match.groups()[1]
line += "\n"
return line
for line in sys.stdin:
if line.find("getelementptr ") == line.find("getelementptr inbounds"):
if line.find("getelementptr inbounds") != line.find("getelementptr inbounds ("):
line = conv(re.match(ibrep, line), line)
elif line.find("getelementptr ") != line.find("getelementptr ("):
line = conv(re.match(normrep, line), line)
sys.stdout.write(line)
apply.sh:
for name in "$@"
do
python3 `dirname "$0"`/update.py < "$name" > "$name.tmp" && mv "$name.tmp" "$name"
rm -f "$name.tmp"
done
The actual commands:
From llvm/src:
find test/ -name *.ll | xargs ./apply.sh
From llvm/src/tools/clang:
find test/ -name *.mm -o -name *.m -o -name *.cpp -o -name *.c | xargs -I '{}' ../../apply.sh "{}"
From llvm/src/tools/polly:
find test/ -name *.ll | xargs ./apply.sh
After that, check-all (with llvm, clang, clang-tools-extra, lld,
compiler-rt, and polly all checked out).
The extra 'rm' in the apply.sh script is due to a few files in clang's test
suite using interesting unicode stuff that my python script was throwing
exceptions on. None of those files needed to be migrated, so it seemed
sufficient to ignore those cases.
Reviewers: rafael, dexonsmith, grosser
Differential Revision: http://reviews.llvm.org/D7636
llvm-svn: 230786
2015-02-28 03:29:02 +08:00
|
|
|
%out.gep = getelementptr inbounds float, float addrspace(1)* %out, i32 %x.i
|
2014-10-11 06:01:59 +08:00
|
|
|
store float %sum, float addrspace(1)* %out.gep, align 4
|
|
|
|
|
ret void
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
; SI-LABEL: @simple_read2_f32_x2
|
2015-04-08 09:09:19 +08:00
|
|
|
; SI: ds_read2_b32 v{{\[[0-9]+:[0-9]+\]}}, [[BASEADDR:v[0-9]+]] offset1:8
|
2014-11-05 22:50:53 +08:00
|
|
|
; SI: ds_read2_b32 v{{\[[0-9]+:[0-9]+\]}}, [[BASEADDR]] offset0:11 offset1:27
|
|
|
|
|
; SI: s_endpgm
|
2014-10-11 06:01:59 +08:00
|
|
|
define void @simple_read2_f32_x2(float addrspace(1)* %out) #0 {
|
|
|
|
|
%tid.x = tail call i32 @llvm.r600.read.tidig.x() #1
|
|
|
|
|
%idx.0 = add nsw i32 %tid.x, 0
|
[opaque pointer type] Add textual IR support for explicit type parameter to getelementptr instruction
One of several parallel first steps to remove the target type of pointers,
replacing them with a single opaque pointer type.
This adds an explicit type parameter to the gep instruction so that when the
first parameter becomes an opaque pointer type, the type to gep through is
still available to the instructions.
* This doesn't modify gep operators, only instructions (operators will be
handled separately)
* Textual IR changes only. Bitcode (including upgrade) and changing the
in-memory representation will be in separate changes.
* geps of vectors are transformed as:
getelementptr <4 x float*> %x, ...
->getelementptr float, <4 x float*> %x, ...
Then, once the opaque pointer type is introduced, this will ultimately look
like:
getelementptr float, <4 x ptr> %x
with the unambiguous interpretation that it is a vector of pointers to float.
* address spaces remain on the pointer, not the type:
getelementptr float addrspace(1)* %x
->getelementptr float, float addrspace(1)* %x
Then, eventually:
getelementptr float, ptr addrspace(1) %x
Importantly, the massive amount of test case churn has been automated by
same crappy python code. I had to manually update a few test cases that
wouldn't fit the script's model (r228970,r229196,r229197,r229198). The
python script just massages stdin and writes the result to stdout, I
then wrapped that in a shell script to handle replacing files, then
using the usual find+xargs to migrate all the files.
update.py:
import fileinput
import sys
import re
ibrep = re.compile(r"(^.*?[^%\w]getelementptr inbounds )(((?:<\d* x )?)(.*?)(| addrspace\(\d\)) *\*(|>)(?:$| *(?:%|@|null|undef|blockaddress|getelementptr|addrspacecast|bitcast|inttoptr|\[\[[a-zA-Z]|\{\{).*$))")
normrep = re.compile( r"(^.*?[^%\w]getelementptr )(((?:<\d* x )?)(.*?)(| addrspace\(\d\)) *\*(|>)(?:$| *(?:%|@|null|undef|blockaddress|getelementptr|addrspacecast|bitcast|inttoptr|\[\[[a-zA-Z]|\{\{).*$))")
def conv(match, line):
if not match:
return line
line = match.groups()[0]
if len(match.groups()[5]) == 0:
line += match.groups()[2]
line += match.groups()[3]
line += ", "
line += match.groups()[1]
line += "\n"
return line
for line in sys.stdin:
if line.find("getelementptr ") == line.find("getelementptr inbounds"):
if line.find("getelementptr inbounds") != line.find("getelementptr inbounds ("):
line = conv(re.match(ibrep, line), line)
elif line.find("getelementptr ") != line.find("getelementptr ("):
line = conv(re.match(normrep, line), line)
sys.stdout.write(line)
apply.sh:
for name in "$@"
do
python3 `dirname "$0"`/update.py < "$name" > "$name.tmp" && mv "$name.tmp" "$name"
rm -f "$name.tmp"
done
The actual commands:
From llvm/src:
find test/ -name *.ll | xargs ./apply.sh
From llvm/src/tools/clang:
find test/ -name *.mm -o -name *.m -o -name *.cpp -o -name *.c | xargs -I '{}' ../../apply.sh "{}"
From llvm/src/tools/polly:
find test/ -name *.ll | xargs ./apply.sh
After that, check-all (with llvm, clang, clang-tools-extra, lld,
compiler-rt, and polly all checked out).
The extra 'rm' in the apply.sh script is due to a few files in clang's test
suite using interesting unicode stuff that my python script was throwing
exceptions on. None of those files needed to be migrated, so it seemed
sufficient to ignore those cases.
Reviewers: rafael, dexonsmith, grosser
Differential Revision: http://reviews.llvm.org/D7636
llvm-svn: 230786
2015-02-28 03:29:02 +08:00
|
|
|
%arrayidx0 = getelementptr inbounds [512 x float], [512 x float] addrspace(3)* @lds, i32 0, i32 %idx.0
|
2015-02-28 05:17:42 +08:00
|
|
|
%val0 = load float, float addrspace(3)* %arrayidx0, align 4
|
2014-10-11 06:01:59 +08:00
|
|
|
|
|
|
|
|
%idx.1 = add nsw i32 %tid.x, 8
|
[opaque pointer type] Add textual IR support for explicit type parameter to getelementptr instruction
One of several parallel first steps to remove the target type of pointers,
replacing them with a single opaque pointer type.
This adds an explicit type parameter to the gep instruction so that when the
first parameter becomes an opaque pointer type, the type to gep through is
still available to the instructions.
* This doesn't modify gep operators, only instructions (operators will be
handled separately)
* Textual IR changes only. Bitcode (including upgrade) and changing the
in-memory representation will be in separate changes.
* geps of vectors are transformed as:
getelementptr <4 x float*> %x, ...
->getelementptr float, <4 x float*> %x, ...
Then, once the opaque pointer type is introduced, this will ultimately look
like:
getelementptr float, <4 x ptr> %x
with the unambiguous interpretation that it is a vector of pointers to float.
* address spaces remain on the pointer, not the type:
getelementptr float addrspace(1)* %x
->getelementptr float, float addrspace(1)* %x
Then, eventually:
getelementptr float, ptr addrspace(1) %x
Importantly, the massive amount of test case churn has been automated by
same crappy python code. I had to manually update a few test cases that
wouldn't fit the script's model (r228970,r229196,r229197,r229198). The
python script just massages stdin and writes the result to stdout, I
then wrapped that in a shell script to handle replacing files, then
using the usual find+xargs to migrate all the files.
update.py:
import fileinput
import sys
import re
ibrep = re.compile(r"(^.*?[^%\w]getelementptr inbounds )(((?:<\d* x )?)(.*?)(| addrspace\(\d\)) *\*(|>)(?:$| *(?:%|@|null|undef|blockaddress|getelementptr|addrspacecast|bitcast|inttoptr|\[\[[a-zA-Z]|\{\{).*$))")
normrep = re.compile( r"(^.*?[^%\w]getelementptr )(((?:<\d* x )?)(.*?)(| addrspace\(\d\)) *\*(|>)(?:$| *(?:%|@|null|undef|blockaddress|getelementptr|addrspacecast|bitcast|inttoptr|\[\[[a-zA-Z]|\{\{).*$))")
def conv(match, line):
if not match:
return line
line = match.groups()[0]
if len(match.groups()[5]) == 0:
line += match.groups()[2]
line += match.groups()[3]
line += ", "
line += match.groups()[1]
line += "\n"
return line
for line in sys.stdin:
if line.find("getelementptr ") == line.find("getelementptr inbounds"):
if line.find("getelementptr inbounds") != line.find("getelementptr inbounds ("):
line = conv(re.match(ibrep, line), line)
elif line.find("getelementptr ") != line.find("getelementptr ("):
line = conv(re.match(normrep, line), line)
sys.stdout.write(line)
apply.sh:
for name in "$@"
do
python3 `dirname "$0"`/update.py < "$name" > "$name.tmp" && mv "$name.tmp" "$name"
rm -f "$name.tmp"
done
The actual commands:
From llvm/src:
find test/ -name *.ll | xargs ./apply.sh
From llvm/src/tools/clang:
find test/ -name *.mm -o -name *.m -o -name *.cpp -o -name *.c | xargs -I '{}' ../../apply.sh "{}"
From llvm/src/tools/polly:
find test/ -name *.ll | xargs ./apply.sh
After that, check-all (with llvm, clang, clang-tools-extra, lld,
compiler-rt, and polly all checked out).
The extra 'rm' in the apply.sh script is due to a few files in clang's test
suite using interesting unicode stuff that my python script was throwing
exceptions on. None of those files needed to be migrated, so it seemed
sufficient to ignore those cases.
Reviewers: rafael, dexonsmith, grosser
Differential Revision: http://reviews.llvm.org/D7636
llvm-svn: 230786
2015-02-28 03:29:02 +08:00
|
|
|
%arrayidx1 = getelementptr inbounds [512 x float], [512 x float] addrspace(3)* @lds, i32 0, i32 %idx.1
|
2015-02-28 05:17:42 +08:00
|
|
|
%val1 = load float, float addrspace(3)* %arrayidx1, align 4
|
2014-10-11 06:01:59 +08:00
|
|
|
%sum.0 = fadd float %val0, %val1
|
|
|
|
|
|
|
|
|
|
%idx.2 = add nsw i32 %tid.x, 11
|
[opaque pointer type] Add textual IR support for explicit type parameter to getelementptr instruction
One of several parallel first steps to remove the target type of pointers,
replacing them with a single opaque pointer type.
This adds an explicit type parameter to the gep instruction so that when the
first parameter becomes an opaque pointer type, the type to gep through is
still available to the instructions.
* This doesn't modify gep operators, only instructions (operators will be
handled separately)
* Textual IR changes only. Bitcode (including upgrade) and changing the
in-memory representation will be in separate changes.
* geps of vectors are transformed as:
getelementptr <4 x float*> %x, ...
->getelementptr float, <4 x float*> %x, ...
Then, once the opaque pointer type is introduced, this will ultimately look
like:
getelementptr float, <4 x ptr> %x
with the unambiguous interpretation that it is a vector of pointers to float.
* address spaces remain on the pointer, not the type:
getelementptr float addrspace(1)* %x
->getelementptr float, float addrspace(1)* %x
Then, eventually:
getelementptr float, ptr addrspace(1) %x
Importantly, the massive amount of test case churn has been automated by
same crappy python code. I had to manually update a few test cases that
wouldn't fit the script's model (r228970,r229196,r229197,r229198). The
python script just massages stdin and writes the result to stdout, I
then wrapped that in a shell script to handle replacing files, then
using the usual find+xargs to migrate all the files.
update.py:
import fileinput
import sys
import re
ibrep = re.compile(r"(^.*?[^%\w]getelementptr inbounds )(((?:<\d* x )?)(.*?)(| addrspace\(\d\)) *\*(|>)(?:$| *(?:%|@|null|undef|blockaddress|getelementptr|addrspacecast|bitcast|inttoptr|\[\[[a-zA-Z]|\{\{).*$))")
normrep = re.compile( r"(^.*?[^%\w]getelementptr )(((?:<\d* x )?)(.*?)(| addrspace\(\d\)) *\*(|>)(?:$| *(?:%|@|null|undef|blockaddress|getelementptr|addrspacecast|bitcast|inttoptr|\[\[[a-zA-Z]|\{\{).*$))")
def conv(match, line):
if not match:
return line
line = match.groups()[0]
if len(match.groups()[5]) == 0:
line += match.groups()[2]
line += match.groups()[3]
line += ", "
line += match.groups()[1]
line += "\n"
return line
for line in sys.stdin:
if line.find("getelementptr ") == line.find("getelementptr inbounds"):
if line.find("getelementptr inbounds") != line.find("getelementptr inbounds ("):
line = conv(re.match(ibrep, line), line)
elif line.find("getelementptr ") != line.find("getelementptr ("):
line = conv(re.match(normrep, line), line)
sys.stdout.write(line)
apply.sh:
for name in "$@"
do
python3 `dirname "$0"`/update.py < "$name" > "$name.tmp" && mv "$name.tmp" "$name"
rm -f "$name.tmp"
done
The actual commands:
From llvm/src:
find test/ -name *.ll | xargs ./apply.sh
From llvm/src/tools/clang:
find test/ -name *.mm -o -name *.m -o -name *.cpp -o -name *.c | xargs -I '{}' ../../apply.sh "{}"
From llvm/src/tools/polly:
find test/ -name *.ll | xargs ./apply.sh
After that, check-all (with llvm, clang, clang-tools-extra, lld,
compiler-rt, and polly all checked out).
The extra 'rm' in the apply.sh script is due to a few files in clang's test
suite using interesting unicode stuff that my python script was throwing
exceptions on. None of those files needed to be migrated, so it seemed
sufficient to ignore those cases.
Reviewers: rafael, dexonsmith, grosser
Differential Revision: http://reviews.llvm.org/D7636
llvm-svn: 230786
2015-02-28 03:29:02 +08:00
|
|
|
%arrayidx2 = getelementptr inbounds [512 x float], [512 x float] addrspace(3)* @lds, i32 0, i32 %idx.2
|
2015-02-28 05:17:42 +08:00
|
|
|
%val2 = load float, float addrspace(3)* %arrayidx2, align 4
|
2014-10-11 06:01:59 +08:00
|
|
|
|
|
|
|
|
%idx.3 = add nsw i32 %tid.x, 27
|
[opaque pointer type] Add textual IR support for explicit type parameter to getelementptr instruction
One of several parallel first steps to remove the target type of pointers,
replacing them with a single opaque pointer type.
This adds an explicit type parameter to the gep instruction so that when the
first parameter becomes an opaque pointer type, the type to gep through is
still available to the instructions.
* This doesn't modify gep operators, only instructions (operators will be
handled separately)
* Textual IR changes only. Bitcode (including upgrade) and changing the
in-memory representation will be in separate changes.
* geps of vectors are transformed as:
getelementptr <4 x float*> %x, ...
->getelementptr float, <4 x float*> %x, ...
Then, once the opaque pointer type is introduced, this will ultimately look
like:
getelementptr float, <4 x ptr> %x
with the unambiguous interpretation that it is a vector of pointers to float.
* address spaces remain on the pointer, not the type:
getelementptr float addrspace(1)* %x
->getelementptr float, float addrspace(1)* %x
Then, eventually:
getelementptr float, ptr addrspace(1) %x
Importantly, the massive amount of test case churn has been automated by
same crappy python code. I had to manually update a few test cases that
wouldn't fit the script's model (r228970,r229196,r229197,r229198). The
python script just massages stdin and writes the result to stdout, I
then wrapped that in a shell script to handle replacing files, then
using the usual find+xargs to migrate all the files.
update.py:
import fileinput
import sys
import re
ibrep = re.compile(r"(^.*?[^%\w]getelementptr inbounds )(((?:<\d* x )?)(.*?)(| addrspace\(\d\)) *\*(|>)(?:$| *(?:%|@|null|undef|blockaddress|getelementptr|addrspacecast|bitcast|inttoptr|\[\[[a-zA-Z]|\{\{).*$))")
normrep = re.compile( r"(^.*?[^%\w]getelementptr )(((?:<\d* x )?)(.*?)(| addrspace\(\d\)) *\*(|>)(?:$| *(?:%|@|null|undef|blockaddress|getelementptr|addrspacecast|bitcast|inttoptr|\[\[[a-zA-Z]|\{\{).*$))")
def conv(match, line):
if not match:
return line
line = match.groups()[0]
if len(match.groups()[5]) == 0:
line += match.groups()[2]
line += match.groups()[3]
line += ", "
line += match.groups()[1]
line += "\n"
return line
for line in sys.stdin:
if line.find("getelementptr ") == line.find("getelementptr inbounds"):
if line.find("getelementptr inbounds") != line.find("getelementptr inbounds ("):
line = conv(re.match(ibrep, line), line)
elif line.find("getelementptr ") != line.find("getelementptr ("):
line = conv(re.match(normrep, line), line)
sys.stdout.write(line)
apply.sh:
for name in "$@"
do
python3 `dirname "$0"`/update.py < "$name" > "$name.tmp" && mv "$name.tmp" "$name"
rm -f "$name.tmp"
done
The actual commands:
From llvm/src:
find test/ -name *.ll | xargs ./apply.sh
From llvm/src/tools/clang:
find test/ -name *.mm -o -name *.m -o -name *.cpp -o -name *.c | xargs -I '{}' ../../apply.sh "{}"
From llvm/src/tools/polly:
find test/ -name *.ll | xargs ./apply.sh
After that, check-all (with llvm, clang, clang-tools-extra, lld,
compiler-rt, and polly all checked out).
The extra 'rm' in the apply.sh script is due to a few files in clang's test
suite using interesting unicode stuff that my python script was throwing
exceptions on. None of those files needed to be migrated, so it seemed
sufficient to ignore those cases.
Reviewers: rafael, dexonsmith, grosser
Differential Revision: http://reviews.llvm.org/D7636
llvm-svn: 230786
2015-02-28 03:29:02 +08:00
|
|
|
%arrayidx3 = getelementptr inbounds [512 x float], [512 x float] addrspace(3)* @lds, i32 0, i32 %idx.3
|
2015-02-28 05:17:42 +08:00
|
|
|
%val3 = load float, float addrspace(3)* %arrayidx3, align 4
|
2014-10-11 06:01:59 +08:00
|
|
|
%sum.1 = fadd float %val2, %val3
|
|
|
|
|
|
|
|
|
|
%sum = fadd float %sum.0, %sum.1
|
[opaque pointer type] Add textual IR support for explicit type parameter to getelementptr instruction
One of several parallel first steps to remove the target type of pointers,
replacing them with a single opaque pointer type.
This adds an explicit type parameter to the gep instruction so that when the
first parameter becomes an opaque pointer type, the type to gep through is
still available to the instructions.
* This doesn't modify gep operators, only instructions (operators will be
handled separately)
* Textual IR changes only. Bitcode (including upgrade) and changing the
in-memory representation will be in separate changes.
* geps of vectors are transformed as:
getelementptr <4 x float*> %x, ...
->getelementptr float, <4 x float*> %x, ...
Then, once the opaque pointer type is introduced, this will ultimately look
like:
getelementptr float, <4 x ptr> %x
with the unambiguous interpretation that it is a vector of pointers to float.
* address spaces remain on the pointer, not the type:
getelementptr float addrspace(1)* %x
->getelementptr float, float addrspace(1)* %x
Then, eventually:
getelementptr float, ptr addrspace(1) %x
Importantly, the massive amount of test case churn has been automated by
same crappy python code. I had to manually update a few test cases that
wouldn't fit the script's model (r228970,r229196,r229197,r229198). The
python script just massages stdin and writes the result to stdout, I
then wrapped that in a shell script to handle replacing files, then
using the usual find+xargs to migrate all the files.
update.py:
import fileinput
import sys
import re
ibrep = re.compile(r"(^.*?[^%\w]getelementptr inbounds )(((?:<\d* x )?)(.*?)(| addrspace\(\d\)) *\*(|>)(?:$| *(?:%|@|null|undef|blockaddress|getelementptr|addrspacecast|bitcast|inttoptr|\[\[[a-zA-Z]|\{\{).*$))")
normrep = re.compile( r"(^.*?[^%\w]getelementptr )(((?:<\d* x )?)(.*?)(| addrspace\(\d\)) *\*(|>)(?:$| *(?:%|@|null|undef|blockaddress|getelementptr|addrspacecast|bitcast|inttoptr|\[\[[a-zA-Z]|\{\{).*$))")
def conv(match, line):
if not match:
return line
line = match.groups()[0]
if len(match.groups()[5]) == 0:
line += match.groups()[2]
line += match.groups()[3]
line += ", "
line += match.groups()[1]
line += "\n"
return line
for line in sys.stdin:
if line.find("getelementptr ") == line.find("getelementptr inbounds"):
if line.find("getelementptr inbounds") != line.find("getelementptr inbounds ("):
line = conv(re.match(ibrep, line), line)
elif line.find("getelementptr ") != line.find("getelementptr ("):
line = conv(re.match(normrep, line), line)
sys.stdout.write(line)
apply.sh:
for name in "$@"
do
python3 `dirname "$0"`/update.py < "$name" > "$name.tmp" && mv "$name.tmp" "$name"
rm -f "$name.tmp"
done
The actual commands:
From llvm/src:
find test/ -name *.ll | xargs ./apply.sh
From llvm/src/tools/clang:
find test/ -name *.mm -o -name *.m -o -name *.cpp -o -name *.c | xargs -I '{}' ../../apply.sh "{}"
From llvm/src/tools/polly:
find test/ -name *.ll | xargs ./apply.sh
After that, check-all (with llvm, clang, clang-tools-extra, lld,
compiler-rt, and polly all checked out).
The extra 'rm' in the apply.sh script is due to a few files in clang's test
suite using interesting unicode stuff that my python script was throwing
exceptions on. None of those files needed to be migrated, so it seemed
sufficient to ignore those cases.
Reviewers: rafael, dexonsmith, grosser
Differential Revision: http://reviews.llvm.org/D7636
llvm-svn: 230786
2015-02-28 03:29:02 +08:00
|
|
|
%out.gep = getelementptr inbounds float, float addrspace(1)* %out, i32 %idx.0
|
2014-10-11 06:01:59 +08:00
|
|
|
store float %sum, float addrspace(1)* %out.gep, align 4
|
|
|
|
|
ret void
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
; Make sure there is an instruction between the two sets of reads.
|
|
|
|
|
; SI-LABEL: @simple_read2_f32_x2_barrier
|
2015-04-08 09:09:19 +08:00
|
|
|
; SI: ds_read2_b32 v{{\[[0-9]+:[0-9]+\]}}, [[BASEADDR:v[0-9]+]] offset1:8
|
2014-11-05 22:50:53 +08:00
|
|
|
; SI: s_barrier
|
|
|
|
|
; SI: ds_read2_b32 v{{\[[0-9]+:[0-9]+\]}}, [[BASEADDR]] offset0:11 offset1:27
|
|
|
|
|
; SI: s_endpgm
|
2014-10-11 06:01:59 +08:00
|
|
|
define void @simple_read2_f32_x2_barrier(float addrspace(1)* %out) #0 {
|
|
|
|
|
%tid.x = tail call i32 @llvm.r600.read.tidig.x() #1
|
|
|
|
|
%idx.0 = add nsw i32 %tid.x, 0
|
[opaque pointer type] Add textual IR support for explicit type parameter to getelementptr instruction
One of several parallel first steps to remove the target type of pointers,
replacing them with a single opaque pointer type.
This adds an explicit type parameter to the gep instruction so that when the
first parameter becomes an opaque pointer type, the type to gep through is
still available to the instructions.
* This doesn't modify gep operators, only instructions (operators will be
handled separately)
* Textual IR changes only. Bitcode (including upgrade) and changing the
in-memory representation will be in separate changes.
* geps of vectors are transformed as:
getelementptr <4 x float*> %x, ...
->getelementptr float, <4 x float*> %x, ...
Then, once the opaque pointer type is introduced, this will ultimately look
like:
getelementptr float, <4 x ptr> %x
with the unambiguous interpretation that it is a vector of pointers to float.
* address spaces remain on the pointer, not the type:
getelementptr float addrspace(1)* %x
->getelementptr float, float addrspace(1)* %x
Then, eventually:
getelementptr float, ptr addrspace(1) %x
Importantly, the massive amount of test case churn has been automated by
same crappy python code. I had to manually update a few test cases that
wouldn't fit the script's model (r228970,r229196,r229197,r229198). The
python script just massages stdin and writes the result to stdout, I
then wrapped that in a shell script to handle replacing files, then
using the usual find+xargs to migrate all the files.
update.py:
import fileinput
import sys
import re
ibrep = re.compile(r"(^.*?[^%\w]getelementptr inbounds )(((?:<\d* x )?)(.*?)(| addrspace\(\d\)) *\*(|>)(?:$| *(?:%|@|null|undef|blockaddress|getelementptr|addrspacecast|bitcast|inttoptr|\[\[[a-zA-Z]|\{\{).*$))")
normrep = re.compile( r"(^.*?[^%\w]getelementptr )(((?:<\d* x )?)(.*?)(| addrspace\(\d\)) *\*(|>)(?:$| *(?:%|@|null|undef|blockaddress|getelementptr|addrspacecast|bitcast|inttoptr|\[\[[a-zA-Z]|\{\{).*$))")
def conv(match, line):
if not match:
return line
line = match.groups()[0]
if len(match.groups()[5]) == 0:
line += match.groups()[2]
line += match.groups()[3]
line += ", "
line += match.groups()[1]
line += "\n"
return line
for line in sys.stdin:
if line.find("getelementptr ") == line.find("getelementptr inbounds"):
if line.find("getelementptr inbounds") != line.find("getelementptr inbounds ("):
line = conv(re.match(ibrep, line), line)
elif line.find("getelementptr ") != line.find("getelementptr ("):
line = conv(re.match(normrep, line), line)
sys.stdout.write(line)
apply.sh:
for name in "$@"
do
python3 `dirname "$0"`/update.py < "$name" > "$name.tmp" && mv "$name.tmp" "$name"
rm -f "$name.tmp"
done
The actual commands:
From llvm/src:
find test/ -name *.ll | xargs ./apply.sh
From llvm/src/tools/clang:
find test/ -name *.mm -o -name *.m -o -name *.cpp -o -name *.c | xargs -I '{}' ../../apply.sh "{}"
From llvm/src/tools/polly:
find test/ -name *.ll | xargs ./apply.sh
After that, check-all (with llvm, clang, clang-tools-extra, lld,
compiler-rt, and polly all checked out).
The extra 'rm' in the apply.sh script is due to a few files in clang's test
suite using interesting unicode stuff that my python script was throwing
exceptions on. None of those files needed to be migrated, so it seemed
sufficient to ignore those cases.
Reviewers: rafael, dexonsmith, grosser
Differential Revision: http://reviews.llvm.org/D7636
llvm-svn: 230786
2015-02-28 03:29:02 +08:00
|
|
|
%arrayidx0 = getelementptr inbounds [512 x float], [512 x float] addrspace(3)* @lds, i32 0, i32 %idx.0
|
2015-02-28 05:17:42 +08:00
|
|
|
%val0 = load float, float addrspace(3)* %arrayidx0, align 4
|
2014-10-11 06:01:59 +08:00
|
|
|
|
|
|
|
|
%idx.1 = add nsw i32 %tid.x, 8
|
[opaque pointer type] Add textual IR support for explicit type parameter to getelementptr instruction
One of several parallel first steps to remove the target type of pointers,
replacing them with a single opaque pointer type.
This adds an explicit type parameter to the gep instruction so that when the
first parameter becomes an opaque pointer type, the type to gep through is
still available to the instructions.
* This doesn't modify gep operators, only instructions (operators will be
handled separately)
* Textual IR changes only. Bitcode (including upgrade) and changing the
in-memory representation will be in separate changes.
* geps of vectors are transformed as:
getelementptr <4 x float*> %x, ...
->getelementptr float, <4 x float*> %x, ...
Then, once the opaque pointer type is introduced, this will ultimately look
like:
getelementptr float, <4 x ptr> %x
with the unambiguous interpretation that it is a vector of pointers to float.
* address spaces remain on the pointer, not the type:
getelementptr float addrspace(1)* %x
->getelementptr float, float addrspace(1)* %x
Then, eventually:
getelementptr float, ptr addrspace(1) %x
Importantly, the massive amount of test case churn has been automated by
same crappy python code. I had to manually update a few test cases that
wouldn't fit the script's model (r228970,r229196,r229197,r229198). The
python script just massages stdin and writes the result to stdout, I
then wrapped that in a shell script to handle replacing files, then
using the usual find+xargs to migrate all the files.
update.py:
import fileinput
import sys
import re
ibrep = re.compile(r"(^.*?[^%\w]getelementptr inbounds )(((?:<\d* x )?)(.*?)(| addrspace\(\d\)) *\*(|>)(?:$| *(?:%|@|null|undef|blockaddress|getelementptr|addrspacecast|bitcast|inttoptr|\[\[[a-zA-Z]|\{\{).*$))")
normrep = re.compile( r"(^.*?[^%\w]getelementptr )(((?:<\d* x )?)(.*?)(| addrspace\(\d\)) *\*(|>)(?:$| *(?:%|@|null|undef|blockaddress|getelementptr|addrspacecast|bitcast|inttoptr|\[\[[a-zA-Z]|\{\{).*$))")
def conv(match, line):
if not match:
return line
line = match.groups()[0]
if len(match.groups()[5]) == 0:
line += match.groups()[2]
line += match.groups()[3]
line += ", "
line += match.groups()[1]
line += "\n"
return line
for line in sys.stdin:
if line.find("getelementptr ") == line.find("getelementptr inbounds"):
if line.find("getelementptr inbounds") != line.find("getelementptr inbounds ("):
line = conv(re.match(ibrep, line), line)
elif line.find("getelementptr ") != line.find("getelementptr ("):
line = conv(re.match(normrep, line), line)
sys.stdout.write(line)
apply.sh:
for name in "$@"
do
python3 `dirname "$0"`/update.py < "$name" > "$name.tmp" && mv "$name.tmp" "$name"
rm -f "$name.tmp"
done
The actual commands:
From llvm/src:
find test/ -name *.ll | xargs ./apply.sh
From llvm/src/tools/clang:
find test/ -name *.mm -o -name *.m -o -name *.cpp -o -name *.c | xargs -I '{}' ../../apply.sh "{}"
From llvm/src/tools/polly:
find test/ -name *.ll | xargs ./apply.sh
After that, check-all (with llvm, clang, clang-tools-extra, lld,
compiler-rt, and polly all checked out).
The extra 'rm' in the apply.sh script is due to a few files in clang's test
suite using interesting unicode stuff that my python script was throwing
exceptions on. None of those files needed to be migrated, so it seemed
sufficient to ignore those cases.
Reviewers: rafael, dexonsmith, grosser
Differential Revision: http://reviews.llvm.org/D7636
llvm-svn: 230786
2015-02-28 03:29:02 +08:00
|
|
|
%arrayidx1 = getelementptr inbounds [512 x float], [512 x float] addrspace(3)* @lds, i32 0, i32 %idx.1
|
2015-02-28 05:17:42 +08:00
|
|
|
%val1 = load float, float addrspace(3)* %arrayidx1, align 4
|
2014-10-11 06:01:59 +08:00
|
|
|
%sum.0 = fadd float %val0, %val1
|
|
|
|
|
|
|
|
|
|
call void @llvm.AMDGPU.barrier.local() #2
|
|
|
|
|
|
|
|
|
|
%idx.2 = add nsw i32 %tid.x, 11
|
[opaque pointer type] Add textual IR support for explicit type parameter to getelementptr instruction
One of several parallel first steps to remove the target type of pointers,
replacing them with a single opaque pointer type.
This adds an explicit type parameter to the gep instruction so that when the
first parameter becomes an opaque pointer type, the type to gep through is
still available to the instructions.
* This doesn't modify gep operators, only instructions (operators will be
handled separately)
* Textual IR changes only. Bitcode (including upgrade) and changing the
in-memory representation will be in separate changes.
* geps of vectors are transformed as:
getelementptr <4 x float*> %x, ...
->getelementptr float, <4 x float*> %x, ...
Then, once the opaque pointer type is introduced, this will ultimately look
like:
getelementptr float, <4 x ptr> %x
with the unambiguous interpretation that it is a vector of pointers to float.
* address spaces remain on the pointer, not the type:
getelementptr float addrspace(1)* %x
->getelementptr float, float addrspace(1)* %x
Then, eventually:
getelementptr float, ptr addrspace(1) %x
Importantly, the massive amount of test case churn has been automated by
same crappy python code. I had to manually update a few test cases that
wouldn't fit the script's model (r228970,r229196,r229197,r229198). The
python script just massages stdin and writes the result to stdout, I
then wrapped that in a shell script to handle replacing files, then
using the usual find+xargs to migrate all the files.
update.py:
import fileinput
import sys
import re
ibrep = re.compile(r"(^.*?[^%\w]getelementptr inbounds )(((?:<\d* x )?)(.*?)(| addrspace\(\d\)) *\*(|>)(?:$| *(?:%|@|null|undef|blockaddress|getelementptr|addrspacecast|bitcast|inttoptr|\[\[[a-zA-Z]|\{\{).*$))")
normrep = re.compile( r"(^.*?[^%\w]getelementptr )(((?:<\d* x )?)(.*?)(| addrspace\(\d\)) *\*(|>)(?:$| *(?:%|@|null|undef|blockaddress|getelementptr|addrspacecast|bitcast|inttoptr|\[\[[a-zA-Z]|\{\{).*$))")
def conv(match, line):
if not match:
return line
line = match.groups()[0]
if len(match.groups()[5]) == 0:
line += match.groups()[2]
line += match.groups()[3]
line += ", "
line += match.groups()[1]
line += "\n"
return line
for line in sys.stdin:
if line.find("getelementptr ") == line.find("getelementptr inbounds"):
if line.find("getelementptr inbounds") != line.find("getelementptr inbounds ("):
line = conv(re.match(ibrep, line), line)
elif line.find("getelementptr ") != line.find("getelementptr ("):
line = conv(re.match(normrep, line), line)
sys.stdout.write(line)
apply.sh:
for name in "$@"
do
python3 `dirname "$0"`/update.py < "$name" > "$name.tmp" && mv "$name.tmp" "$name"
rm -f "$name.tmp"
done
The actual commands:
From llvm/src:
find test/ -name *.ll | xargs ./apply.sh
From llvm/src/tools/clang:
find test/ -name *.mm -o -name *.m -o -name *.cpp -o -name *.c | xargs -I '{}' ../../apply.sh "{}"
From llvm/src/tools/polly:
find test/ -name *.ll | xargs ./apply.sh
After that, check-all (with llvm, clang, clang-tools-extra, lld,
compiler-rt, and polly all checked out).
The extra 'rm' in the apply.sh script is due to a few files in clang's test
suite using interesting unicode stuff that my python script was throwing
exceptions on. None of those files needed to be migrated, so it seemed
sufficient to ignore those cases.
Reviewers: rafael, dexonsmith, grosser
Differential Revision: http://reviews.llvm.org/D7636
llvm-svn: 230786
2015-02-28 03:29:02 +08:00
|
|
|
%arrayidx2 = getelementptr inbounds [512 x float], [512 x float] addrspace(3)* @lds, i32 0, i32 %idx.2
|
2015-02-28 05:17:42 +08:00
|
|
|
%val2 = load float, float addrspace(3)* %arrayidx2, align 4
|
2014-10-11 06:01:59 +08:00
|
|
|
|
|
|
|
|
%idx.3 = add nsw i32 %tid.x, 27
|
[opaque pointer type] Add textual IR support for explicit type parameter to getelementptr instruction
One of several parallel first steps to remove the target type of pointers,
replacing them with a single opaque pointer type.
This adds an explicit type parameter to the gep instruction so that when the
first parameter becomes an opaque pointer type, the type to gep through is
still available to the instructions.
* This doesn't modify gep operators, only instructions (operators will be
handled separately)
* Textual IR changes only. Bitcode (including upgrade) and changing the
in-memory representation will be in separate changes.
* geps of vectors are transformed as:
getelementptr <4 x float*> %x, ...
->getelementptr float, <4 x float*> %x, ...
Then, once the opaque pointer type is introduced, this will ultimately look
like:
getelementptr float, <4 x ptr> %x
with the unambiguous interpretation that it is a vector of pointers to float.
* address spaces remain on the pointer, not the type:
getelementptr float addrspace(1)* %x
->getelementptr float, float addrspace(1)* %x
Then, eventually:
getelementptr float, ptr addrspace(1) %x
Importantly, the massive amount of test case churn has been automated by
same crappy python code. I had to manually update a few test cases that
wouldn't fit the script's model (r228970,r229196,r229197,r229198). The
python script just massages stdin and writes the result to stdout, I
then wrapped that in a shell script to handle replacing files, then
using the usual find+xargs to migrate all the files.
update.py:
import fileinput
import sys
import re
ibrep = re.compile(r"(^.*?[^%\w]getelementptr inbounds )(((?:<\d* x )?)(.*?)(| addrspace\(\d\)) *\*(|>)(?:$| *(?:%|@|null|undef|blockaddress|getelementptr|addrspacecast|bitcast|inttoptr|\[\[[a-zA-Z]|\{\{).*$))")
normrep = re.compile( r"(^.*?[^%\w]getelementptr )(((?:<\d* x )?)(.*?)(| addrspace\(\d\)) *\*(|>)(?:$| *(?:%|@|null|undef|blockaddress|getelementptr|addrspacecast|bitcast|inttoptr|\[\[[a-zA-Z]|\{\{).*$))")
def conv(match, line):
if not match:
return line
line = match.groups()[0]
if len(match.groups()[5]) == 0:
line += match.groups()[2]
line += match.groups()[3]
line += ", "
line += match.groups()[1]
line += "\n"
return line
for line in sys.stdin:
if line.find("getelementptr ") == line.find("getelementptr inbounds"):
if line.find("getelementptr inbounds") != line.find("getelementptr inbounds ("):
line = conv(re.match(ibrep, line), line)
elif line.find("getelementptr ") != line.find("getelementptr ("):
line = conv(re.match(normrep, line), line)
sys.stdout.write(line)
apply.sh:
for name in "$@"
do
python3 `dirname "$0"`/update.py < "$name" > "$name.tmp" && mv "$name.tmp" "$name"
rm -f "$name.tmp"
done
The actual commands:
From llvm/src:
find test/ -name *.ll | xargs ./apply.sh
From llvm/src/tools/clang:
find test/ -name *.mm -o -name *.m -o -name *.cpp -o -name *.c | xargs -I '{}' ../../apply.sh "{}"
From llvm/src/tools/polly:
find test/ -name *.ll | xargs ./apply.sh
After that, check-all (with llvm, clang, clang-tools-extra, lld,
compiler-rt, and polly all checked out).
The extra 'rm' in the apply.sh script is due to a few files in clang's test
suite using interesting unicode stuff that my python script was throwing
exceptions on. None of those files needed to be migrated, so it seemed
sufficient to ignore those cases.
Reviewers: rafael, dexonsmith, grosser
Differential Revision: http://reviews.llvm.org/D7636
llvm-svn: 230786
2015-02-28 03:29:02 +08:00
|
|
|
%arrayidx3 = getelementptr inbounds [512 x float], [512 x float] addrspace(3)* @lds, i32 0, i32 %idx.3
|
2015-02-28 05:17:42 +08:00
|
|
|
%val3 = load float, float addrspace(3)* %arrayidx3, align 4
|
2014-10-11 06:01:59 +08:00
|
|
|
%sum.1 = fadd float %val2, %val3
|
|
|
|
|
|
|
|
|
|
%sum = fadd float %sum.0, %sum.1
|
[opaque pointer type] Add textual IR support for explicit type parameter to getelementptr instruction
One of several parallel first steps to remove the target type of pointers,
replacing them with a single opaque pointer type.
This adds an explicit type parameter to the gep instruction so that when the
first parameter becomes an opaque pointer type, the type to gep through is
still available to the instructions.
* This doesn't modify gep operators, only instructions (operators will be
handled separately)
* Textual IR changes only. Bitcode (including upgrade) and changing the
in-memory representation will be in separate changes.
* geps of vectors are transformed as:
getelementptr <4 x float*> %x, ...
->getelementptr float, <4 x float*> %x, ...
Then, once the opaque pointer type is introduced, this will ultimately look
like:
getelementptr float, <4 x ptr> %x
with the unambiguous interpretation that it is a vector of pointers to float.
* address spaces remain on the pointer, not the type:
getelementptr float addrspace(1)* %x
->getelementptr float, float addrspace(1)* %x
Then, eventually:
getelementptr float, ptr addrspace(1) %x
Importantly, the massive amount of test case churn has been automated by
same crappy python code. I had to manually update a few test cases that
wouldn't fit the script's model (r228970,r229196,r229197,r229198). The
python script just massages stdin and writes the result to stdout, I
then wrapped that in a shell script to handle replacing files, then
using the usual find+xargs to migrate all the files.
update.py:
import fileinput
import sys
import re
ibrep = re.compile(r"(^.*?[^%\w]getelementptr inbounds )(((?:<\d* x )?)(.*?)(| addrspace\(\d\)) *\*(|>)(?:$| *(?:%|@|null|undef|blockaddress|getelementptr|addrspacecast|bitcast|inttoptr|\[\[[a-zA-Z]|\{\{).*$))")
normrep = re.compile( r"(^.*?[^%\w]getelementptr )(((?:<\d* x )?)(.*?)(| addrspace\(\d\)) *\*(|>)(?:$| *(?:%|@|null|undef|blockaddress|getelementptr|addrspacecast|bitcast|inttoptr|\[\[[a-zA-Z]|\{\{).*$))")
def conv(match, line):
if not match:
return line
line = match.groups()[0]
if len(match.groups()[5]) == 0:
line += match.groups()[2]
line += match.groups()[3]
line += ", "
line += match.groups()[1]
line += "\n"
return line
for line in sys.stdin:
if line.find("getelementptr ") == line.find("getelementptr inbounds"):
if line.find("getelementptr inbounds") != line.find("getelementptr inbounds ("):
line = conv(re.match(ibrep, line), line)
elif line.find("getelementptr ") != line.find("getelementptr ("):
line = conv(re.match(normrep, line), line)
sys.stdout.write(line)
apply.sh:
for name in "$@"
do
python3 `dirname "$0"`/update.py < "$name" > "$name.tmp" && mv "$name.tmp" "$name"
rm -f "$name.tmp"
done
The actual commands:
From llvm/src:
find test/ -name *.ll | xargs ./apply.sh
From llvm/src/tools/clang:
find test/ -name *.mm -o -name *.m -o -name *.cpp -o -name *.c | xargs -I '{}' ../../apply.sh "{}"
From llvm/src/tools/polly:
find test/ -name *.ll | xargs ./apply.sh
After that, check-all (with llvm, clang, clang-tools-extra, lld,
compiler-rt, and polly all checked out).
The extra 'rm' in the apply.sh script is due to a few files in clang's test
suite using interesting unicode stuff that my python script was throwing
exceptions on. None of those files needed to be migrated, so it seemed
sufficient to ignore those cases.
Reviewers: rafael, dexonsmith, grosser
Differential Revision: http://reviews.llvm.org/D7636
llvm-svn: 230786
2015-02-28 03:29:02 +08:00
|
|
|
%out.gep = getelementptr inbounds float, float addrspace(1)* %out, i32 %idx.0
|
2014-10-11 06:01:59 +08:00
|
|
|
store float %sum, float addrspace(1)* %out.gep, align 4
|
|
|
|
|
ret void
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
; For some reason adding something to the base address for the first
|
|
|
|
|
; element results in only folding the inner pair.
|
|
|
|
|
|
|
|
|
|
; SI-LABEL: @simple_read2_f32_x2_nonzero_base
|
2014-11-05 22:50:53 +08:00
|
|
|
; SI: ds_read2_b32 v{{\[[0-9]+:[0-9]+\]}}, [[BASEADDR:v[0-9]+]] offset0:2 offset1:8
|
|
|
|
|
; SI: ds_read2_b32 v{{\[[0-9]+:[0-9]+\]}}, [[BASEADDR]] offset0:11 offset1:27
|
|
|
|
|
; SI: s_endpgm
|
2014-10-11 06:01:59 +08:00
|
|
|
define void @simple_read2_f32_x2_nonzero_base(float addrspace(1)* %out) #0 {
|
|
|
|
|
%tid.x = tail call i32 @llvm.r600.read.tidig.x() #1
|
|
|
|
|
%idx.0 = add nsw i32 %tid.x, 2
|
[opaque pointer type] Add textual IR support for explicit type parameter to getelementptr instruction
One of several parallel first steps to remove the target type of pointers,
replacing them with a single opaque pointer type.
This adds an explicit type parameter to the gep instruction so that when the
first parameter becomes an opaque pointer type, the type to gep through is
still available to the instructions.
* This doesn't modify gep operators, only instructions (operators will be
handled separately)
* Textual IR changes only. Bitcode (including upgrade) and changing the
in-memory representation will be in separate changes.
* geps of vectors are transformed as:
getelementptr <4 x float*> %x, ...
->getelementptr float, <4 x float*> %x, ...
Then, once the opaque pointer type is introduced, this will ultimately look
like:
getelementptr float, <4 x ptr> %x
with the unambiguous interpretation that it is a vector of pointers to float.
* address spaces remain on the pointer, not the type:
getelementptr float addrspace(1)* %x
->getelementptr float, float addrspace(1)* %x
Then, eventually:
getelementptr float, ptr addrspace(1) %x
Importantly, the massive amount of test case churn has been automated by
same crappy python code. I had to manually update a few test cases that
wouldn't fit the script's model (r228970,r229196,r229197,r229198). The
python script just massages stdin and writes the result to stdout, I
then wrapped that in a shell script to handle replacing files, then
using the usual find+xargs to migrate all the files.
update.py:
import fileinput
import sys
import re
ibrep = re.compile(r"(^.*?[^%\w]getelementptr inbounds )(((?:<\d* x )?)(.*?)(| addrspace\(\d\)) *\*(|>)(?:$| *(?:%|@|null|undef|blockaddress|getelementptr|addrspacecast|bitcast|inttoptr|\[\[[a-zA-Z]|\{\{).*$))")
normrep = re.compile( r"(^.*?[^%\w]getelementptr )(((?:<\d* x )?)(.*?)(| addrspace\(\d\)) *\*(|>)(?:$| *(?:%|@|null|undef|blockaddress|getelementptr|addrspacecast|bitcast|inttoptr|\[\[[a-zA-Z]|\{\{).*$))")
def conv(match, line):
if not match:
return line
line = match.groups()[0]
if len(match.groups()[5]) == 0:
line += match.groups()[2]
line += match.groups()[3]
line += ", "
line += match.groups()[1]
line += "\n"
return line
for line in sys.stdin:
if line.find("getelementptr ") == line.find("getelementptr inbounds"):
if line.find("getelementptr inbounds") != line.find("getelementptr inbounds ("):
line = conv(re.match(ibrep, line), line)
elif line.find("getelementptr ") != line.find("getelementptr ("):
line = conv(re.match(normrep, line), line)
sys.stdout.write(line)
apply.sh:
for name in "$@"
do
python3 `dirname "$0"`/update.py < "$name" > "$name.tmp" && mv "$name.tmp" "$name"
rm -f "$name.tmp"
done
The actual commands:
From llvm/src:
find test/ -name *.ll | xargs ./apply.sh
From llvm/src/tools/clang:
find test/ -name *.mm -o -name *.m -o -name *.cpp -o -name *.c | xargs -I '{}' ../../apply.sh "{}"
From llvm/src/tools/polly:
find test/ -name *.ll | xargs ./apply.sh
After that, check-all (with llvm, clang, clang-tools-extra, lld,
compiler-rt, and polly all checked out).
The extra 'rm' in the apply.sh script is due to a few files in clang's test
suite using interesting unicode stuff that my python script was throwing
exceptions on. None of those files needed to be migrated, so it seemed
sufficient to ignore those cases.
Reviewers: rafael, dexonsmith, grosser
Differential Revision: http://reviews.llvm.org/D7636
llvm-svn: 230786
2015-02-28 03:29:02 +08:00
|
|
|
%arrayidx0 = getelementptr inbounds [512 x float], [512 x float] addrspace(3)* @lds, i32 0, i32 %idx.0
|
2015-02-28 05:17:42 +08:00
|
|
|
%val0 = load float, float addrspace(3)* %arrayidx0, align 4
|
2014-10-11 06:01:59 +08:00
|
|
|
|
|
|
|
|
%idx.1 = add nsw i32 %tid.x, 8
|
[opaque pointer type] Add textual IR support for explicit type parameter to getelementptr instruction
One of several parallel first steps to remove the target type of pointers,
replacing them with a single opaque pointer type.
This adds an explicit type parameter to the gep instruction so that when the
first parameter becomes an opaque pointer type, the type to gep through is
still available to the instructions.
* This doesn't modify gep operators, only instructions (operators will be
handled separately)
* Textual IR changes only. Bitcode (including upgrade) and changing the
in-memory representation will be in separate changes.
* geps of vectors are transformed as:
getelementptr <4 x float*> %x, ...
->getelementptr float, <4 x float*> %x, ...
Then, once the opaque pointer type is introduced, this will ultimately look
like:
getelementptr float, <4 x ptr> %x
with the unambiguous interpretation that it is a vector of pointers to float.
* address spaces remain on the pointer, not the type:
getelementptr float addrspace(1)* %x
->getelementptr float, float addrspace(1)* %x
Then, eventually:
getelementptr float, ptr addrspace(1) %x
Importantly, the massive amount of test case churn has been automated by
same crappy python code. I had to manually update a few test cases that
wouldn't fit the script's model (r228970,r229196,r229197,r229198). The
python script just massages stdin and writes the result to stdout, I
then wrapped that in a shell script to handle replacing files, then
using the usual find+xargs to migrate all the files.
update.py:
import fileinput
import sys
import re
ibrep = re.compile(r"(^.*?[^%\w]getelementptr inbounds )(((?:<\d* x )?)(.*?)(| addrspace\(\d\)) *\*(|>)(?:$| *(?:%|@|null|undef|blockaddress|getelementptr|addrspacecast|bitcast|inttoptr|\[\[[a-zA-Z]|\{\{).*$))")
normrep = re.compile( r"(^.*?[^%\w]getelementptr )(((?:<\d* x )?)(.*?)(| addrspace\(\d\)) *\*(|>)(?:$| *(?:%|@|null|undef|blockaddress|getelementptr|addrspacecast|bitcast|inttoptr|\[\[[a-zA-Z]|\{\{).*$))")
def conv(match, line):
if not match:
return line
line = match.groups()[0]
if len(match.groups()[5]) == 0:
line += match.groups()[2]
line += match.groups()[3]
line += ", "
line += match.groups()[1]
line += "\n"
return line
for line in sys.stdin:
if line.find("getelementptr ") == line.find("getelementptr inbounds"):
if line.find("getelementptr inbounds") != line.find("getelementptr inbounds ("):
line = conv(re.match(ibrep, line), line)
elif line.find("getelementptr ") != line.find("getelementptr ("):
line = conv(re.match(normrep, line), line)
sys.stdout.write(line)
apply.sh:
for name in "$@"
do
python3 `dirname "$0"`/update.py < "$name" > "$name.tmp" && mv "$name.tmp" "$name"
rm -f "$name.tmp"
done
The actual commands:
From llvm/src:
find test/ -name *.ll | xargs ./apply.sh
From llvm/src/tools/clang:
find test/ -name *.mm -o -name *.m -o -name *.cpp -o -name *.c | xargs -I '{}' ../../apply.sh "{}"
From llvm/src/tools/polly:
find test/ -name *.ll | xargs ./apply.sh
After that, check-all (with llvm, clang, clang-tools-extra, lld,
compiler-rt, and polly all checked out).
The extra 'rm' in the apply.sh script is due to a few files in clang's test
suite using interesting unicode stuff that my python script was throwing
exceptions on. None of those files needed to be migrated, so it seemed
sufficient to ignore those cases.
Reviewers: rafael, dexonsmith, grosser
Differential Revision: http://reviews.llvm.org/D7636
llvm-svn: 230786
2015-02-28 03:29:02 +08:00
|
|
|
%arrayidx1 = getelementptr inbounds [512 x float], [512 x float] addrspace(3)* @lds, i32 0, i32 %idx.1
|
2015-02-28 05:17:42 +08:00
|
|
|
%val1 = load float, float addrspace(3)* %arrayidx1, align 4
|
2014-10-11 06:01:59 +08:00
|
|
|
%sum.0 = fadd float %val0, %val1
|
|
|
|
|
|
|
|
|
|
%idx.2 = add nsw i32 %tid.x, 11
|
[opaque pointer type] Add textual IR support for explicit type parameter to getelementptr instruction
One of several parallel first steps to remove the target type of pointers,
replacing them with a single opaque pointer type.
This adds an explicit type parameter to the gep instruction so that when the
first parameter becomes an opaque pointer type, the type to gep through is
still available to the instructions.
* This doesn't modify gep operators, only instructions (operators will be
handled separately)
* Textual IR changes only. Bitcode (including upgrade) and changing the
in-memory representation will be in separate changes.
* geps of vectors are transformed as:
getelementptr <4 x float*> %x, ...
->getelementptr float, <4 x float*> %x, ...
Then, once the opaque pointer type is introduced, this will ultimately look
like:
getelementptr float, <4 x ptr> %x
with the unambiguous interpretation that it is a vector of pointers to float.
* address spaces remain on the pointer, not the type:
getelementptr float addrspace(1)* %x
->getelementptr float, float addrspace(1)* %x
Then, eventually:
getelementptr float, ptr addrspace(1) %x
Importantly, the massive amount of test case churn has been automated by
same crappy python code. I had to manually update a few test cases that
wouldn't fit the script's model (r228970,r229196,r229197,r229198). The
python script just massages stdin and writes the result to stdout, I
then wrapped that in a shell script to handle replacing files, then
using the usual find+xargs to migrate all the files.
update.py:
import fileinput
import sys
import re
ibrep = re.compile(r"(^.*?[^%\w]getelementptr inbounds )(((?:<\d* x )?)(.*?)(| addrspace\(\d\)) *\*(|>)(?:$| *(?:%|@|null|undef|blockaddress|getelementptr|addrspacecast|bitcast|inttoptr|\[\[[a-zA-Z]|\{\{).*$))")
normrep = re.compile( r"(^.*?[^%\w]getelementptr )(((?:<\d* x )?)(.*?)(| addrspace\(\d\)) *\*(|>)(?:$| *(?:%|@|null|undef|blockaddress|getelementptr|addrspacecast|bitcast|inttoptr|\[\[[a-zA-Z]|\{\{).*$))")
def conv(match, line):
if not match:
return line
line = match.groups()[0]
if len(match.groups()[5]) == 0:
line += match.groups()[2]
line += match.groups()[3]
line += ", "
line += match.groups()[1]
line += "\n"
return line
for line in sys.stdin:
if line.find("getelementptr ") == line.find("getelementptr inbounds"):
if line.find("getelementptr inbounds") != line.find("getelementptr inbounds ("):
line = conv(re.match(ibrep, line), line)
elif line.find("getelementptr ") != line.find("getelementptr ("):
line = conv(re.match(normrep, line), line)
sys.stdout.write(line)
apply.sh:
for name in "$@"
do
python3 `dirname "$0"`/update.py < "$name" > "$name.tmp" && mv "$name.tmp" "$name"
rm -f "$name.tmp"
done
The actual commands:
From llvm/src:
find test/ -name *.ll | xargs ./apply.sh
From llvm/src/tools/clang:
find test/ -name *.mm -o -name *.m -o -name *.cpp -o -name *.c | xargs -I '{}' ../../apply.sh "{}"
From llvm/src/tools/polly:
find test/ -name *.ll | xargs ./apply.sh
After that, check-all (with llvm, clang, clang-tools-extra, lld,
compiler-rt, and polly all checked out).
The extra 'rm' in the apply.sh script is due to a few files in clang's test
suite using interesting unicode stuff that my python script was throwing
exceptions on. None of those files needed to be migrated, so it seemed
sufficient to ignore those cases.
Reviewers: rafael, dexonsmith, grosser
Differential Revision: http://reviews.llvm.org/D7636
llvm-svn: 230786
2015-02-28 03:29:02 +08:00
|
|
|
%arrayidx2 = getelementptr inbounds [512 x float], [512 x float] addrspace(3)* @lds, i32 0, i32 %idx.2
|
2015-02-28 05:17:42 +08:00
|
|
|
%val2 = load float, float addrspace(3)* %arrayidx2, align 4
|
2014-10-11 06:01:59 +08:00
|
|
|
|
|
|
|
|
%idx.3 = add nsw i32 %tid.x, 27
|
[opaque pointer type] Add textual IR support for explicit type parameter to getelementptr instruction
One of several parallel first steps to remove the target type of pointers,
replacing them with a single opaque pointer type.
This adds an explicit type parameter to the gep instruction so that when the
first parameter becomes an opaque pointer type, the type to gep through is
still available to the instructions.
* This doesn't modify gep operators, only instructions (operators will be
handled separately)
* Textual IR changes only. Bitcode (including upgrade) and changing the
in-memory representation will be in separate changes.
* geps of vectors are transformed as:
getelementptr <4 x float*> %x, ...
->getelementptr float, <4 x float*> %x, ...
Then, once the opaque pointer type is introduced, this will ultimately look
like:
getelementptr float, <4 x ptr> %x
with the unambiguous interpretation that it is a vector of pointers to float.
* address spaces remain on the pointer, not the type:
getelementptr float addrspace(1)* %x
->getelementptr float, float addrspace(1)* %x
Then, eventually:
getelementptr float, ptr addrspace(1) %x
Importantly, the massive amount of test case churn has been automated by
same crappy python code. I had to manually update a few test cases that
wouldn't fit the script's model (r228970,r229196,r229197,r229198). The
python script just massages stdin and writes the result to stdout, I
then wrapped that in a shell script to handle replacing files, then
using the usual find+xargs to migrate all the files.
update.py:
import fileinput
import sys
import re
ibrep = re.compile(r"(^.*?[^%\w]getelementptr inbounds )(((?:<\d* x )?)(.*?)(| addrspace\(\d\)) *\*(|>)(?:$| *(?:%|@|null|undef|blockaddress|getelementptr|addrspacecast|bitcast|inttoptr|\[\[[a-zA-Z]|\{\{).*$))")
normrep = re.compile( r"(^.*?[^%\w]getelementptr )(((?:<\d* x )?)(.*?)(| addrspace\(\d\)) *\*(|>)(?:$| *(?:%|@|null|undef|blockaddress|getelementptr|addrspacecast|bitcast|inttoptr|\[\[[a-zA-Z]|\{\{).*$))")
def conv(match, line):
if not match:
return line
line = match.groups()[0]
if len(match.groups()[5]) == 0:
line += match.groups()[2]
line += match.groups()[3]
line += ", "
line += match.groups()[1]
line += "\n"
return line
for line in sys.stdin:
if line.find("getelementptr ") == line.find("getelementptr inbounds"):
if line.find("getelementptr inbounds") != line.find("getelementptr inbounds ("):
line = conv(re.match(ibrep, line), line)
elif line.find("getelementptr ") != line.find("getelementptr ("):
line = conv(re.match(normrep, line), line)
sys.stdout.write(line)
apply.sh:
for name in "$@"
do
python3 `dirname "$0"`/update.py < "$name" > "$name.tmp" && mv "$name.tmp" "$name"
rm -f "$name.tmp"
done
The actual commands:
From llvm/src:
find test/ -name *.ll | xargs ./apply.sh
From llvm/src/tools/clang:
find test/ -name *.mm -o -name *.m -o -name *.cpp -o -name *.c | xargs -I '{}' ../../apply.sh "{}"
From llvm/src/tools/polly:
find test/ -name *.ll | xargs ./apply.sh
After that, check-all (with llvm, clang, clang-tools-extra, lld,
compiler-rt, and polly all checked out).
The extra 'rm' in the apply.sh script is due to a few files in clang's test
suite using interesting unicode stuff that my python script was throwing
exceptions on. None of those files needed to be migrated, so it seemed
sufficient to ignore those cases.
Reviewers: rafael, dexonsmith, grosser
Differential Revision: http://reviews.llvm.org/D7636
llvm-svn: 230786
2015-02-28 03:29:02 +08:00
|
|
|
%arrayidx3 = getelementptr inbounds [512 x float], [512 x float] addrspace(3)* @lds, i32 0, i32 %idx.3
|
2015-02-28 05:17:42 +08:00
|
|
|
%val3 = load float, float addrspace(3)* %arrayidx3, align 4
|
2014-10-11 06:01:59 +08:00
|
|
|
%sum.1 = fadd float %val2, %val3
|
|
|
|
|
|
|
|
|
|
%sum = fadd float %sum.0, %sum.1
|
[opaque pointer type] Add textual IR support for explicit type parameter to getelementptr instruction
One of several parallel first steps to remove the target type of pointers,
replacing them with a single opaque pointer type.
This adds an explicit type parameter to the gep instruction so that when the
first parameter becomes an opaque pointer type, the type to gep through is
still available to the instructions.
* This doesn't modify gep operators, only instructions (operators will be
handled separately)
* Textual IR changes only. Bitcode (including upgrade) and changing the
in-memory representation will be in separate changes.
* geps of vectors are transformed as:
getelementptr <4 x float*> %x, ...
->getelementptr float, <4 x float*> %x, ...
Then, once the opaque pointer type is introduced, this will ultimately look
like:
getelementptr float, <4 x ptr> %x
with the unambiguous interpretation that it is a vector of pointers to float.
* address spaces remain on the pointer, not the type:
getelementptr float addrspace(1)* %x
->getelementptr float, float addrspace(1)* %x
Then, eventually:
getelementptr float, ptr addrspace(1) %x
Importantly, the massive amount of test case churn has been automated by
same crappy python code. I had to manually update a few test cases that
wouldn't fit the script's model (r228970,r229196,r229197,r229198). The
python script just massages stdin and writes the result to stdout, I
then wrapped that in a shell script to handle replacing files, then
using the usual find+xargs to migrate all the files.
update.py:
import fileinput
import sys
import re
ibrep = re.compile(r"(^.*?[^%\w]getelementptr inbounds )(((?:<\d* x )?)(.*?)(| addrspace\(\d\)) *\*(|>)(?:$| *(?:%|@|null|undef|blockaddress|getelementptr|addrspacecast|bitcast|inttoptr|\[\[[a-zA-Z]|\{\{).*$))")
normrep = re.compile( r"(^.*?[^%\w]getelementptr )(((?:<\d* x )?)(.*?)(| addrspace\(\d\)) *\*(|>)(?:$| *(?:%|@|null|undef|blockaddress|getelementptr|addrspacecast|bitcast|inttoptr|\[\[[a-zA-Z]|\{\{).*$))")
def conv(match, line):
if not match:
return line
line = match.groups()[0]
if len(match.groups()[5]) == 0:
line += match.groups()[2]
line += match.groups()[3]
line += ", "
line += match.groups()[1]
line += "\n"
return line
for line in sys.stdin:
if line.find("getelementptr ") == line.find("getelementptr inbounds"):
if line.find("getelementptr inbounds") != line.find("getelementptr inbounds ("):
line = conv(re.match(ibrep, line), line)
elif line.find("getelementptr ") != line.find("getelementptr ("):
line = conv(re.match(normrep, line), line)
sys.stdout.write(line)
apply.sh:
for name in "$@"
do
python3 `dirname "$0"`/update.py < "$name" > "$name.tmp" && mv "$name.tmp" "$name"
rm -f "$name.tmp"
done
The actual commands:
From llvm/src:
find test/ -name *.ll | xargs ./apply.sh
From llvm/src/tools/clang:
find test/ -name *.mm -o -name *.m -o -name *.cpp -o -name *.c | xargs -I '{}' ../../apply.sh "{}"
From llvm/src/tools/polly:
find test/ -name *.ll | xargs ./apply.sh
After that, check-all (with llvm, clang, clang-tools-extra, lld,
compiler-rt, and polly all checked out).
The extra 'rm' in the apply.sh script is due to a few files in clang's test
suite using interesting unicode stuff that my python script was throwing
exceptions on. None of those files needed to be migrated, so it seemed
sufficient to ignore those cases.
Reviewers: rafael, dexonsmith, grosser
Differential Revision: http://reviews.llvm.org/D7636
llvm-svn: 230786
2015-02-28 03:29:02 +08:00
|
|
|
%out.gep = getelementptr inbounds float, float addrspace(1)* %out, i32 %idx.0
|
2014-10-11 06:01:59 +08:00
|
|
|
store float %sum, float addrspace(1)* %out.gep, align 4
|
|
|
|
|
ret void
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
; Be careful of vectors of pointers. We don't know if the 2 pointers
|
|
|
|
|
; in the vectors are really the same base, so this is not safe to
|
|
|
|
|
; merge.
|
|
|
|
|
; Base pointers come from different subregister of same super
|
|
|
|
|
; register. We can't safely merge this.
|
|
|
|
|
|
|
|
|
|
; SI-LABEL: @read2_ptr_is_subreg_arg_f32
|
2014-11-05 22:50:53 +08:00
|
|
|
; SI-NOT: ds_read2_b32
|
|
|
|
|
; SI: ds_read_b32
|
|
|
|
|
; SI: ds_read_b32
|
|
|
|
|
; SI: s_endpgm
|
2014-10-11 06:01:59 +08:00
|
|
|
define void @read2_ptr_is_subreg_arg_f32(float addrspace(1)* %out, <2 x float addrspace(3)*> %lds.ptr) #0 {
|
|
|
|
|
%x.i = tail call i32 @llvm.r600.read.tidig.x() #1
|
|
|
|
|
%index.0 = insertelement <2 x i32> undef, i32 %x.i, i32 0
|
|
|
|
|
%index.1 = insertelement <2 x i32> %index.0, i32 8, i32 0
|
[opaque pointer type] Add textual IR support for explicit type parameter to getelementptr instruction
One of several parallel first steps to remove the target type of pointers,
replacing them with a single opaque pointer type.
This adds an explicit type parameter to the gep instruction so that when the
first parameter becomes an opaque pointer type, the type to gep through is
still available to the instructions.
* This doesn't modify gep operators, only instructions (operators will be
handled separately)
* Textual IR changes only. Bitcode (including upgrade) and changing the
in-memory representation will be in separate changes.
* geps of vectors are transformed as:
getelementptr <4 x float*> %x, ...
->getelementptr float, <4 x float*> %x, ...
Then, once the opaque pointer type is introduced, this will ultimately look
like:
getelementptr float, <4 x ptr> %x
with the unambiguous interpretation that it is a vector of pointers to float.
* address spaces remain on the pointer, not the type:
getelementptr float addrspace(1)* %x
->getelementptr float, float addrspace(1)* %x
Then, eventually:
getelementptr float, ptr addrspace(1) %x
Importantly, the massive amount of test case churn has been automated by
same crappy python code. I had to manually update a few test cases that
wouldn't fit the script's model (r228970,r229196,r229197,r229198). The
python script just massages stdin and writes the result to stdout, I
then wrapped that in a shell script to handle replacing files, then
using the usual find+xargs to migrate all the files.
update.py:
import fileinput
import sys
import re
ibrep = re.compile(r"(^.*?[^%\w]getelementptr inbounds )(((?:<\d* x )?)(.*?)(| addrspace\(\d\)) *\*(|>)(?:$| *(?:%|@|null|undef|blockaddress|getelementptr|addrspacecast|bitcast|inttoptr|\[\[[a-zA-Z]|\{\{).*$))")
normrep = re.compile( r"(^.*?[^%\w]getelementptr )(((?:<\d* x )?)(.*?)(| addrspace\(\d\)) *\*(|>)(?:$| *(?:%|@|null|undef|blockaddress|getelementptr|addrspacecast|bitcast|inttoptr|\[\[[a-zA-Z]|\{\{).*$))")
def conv(match, line):
if not match:
return line
line = match.groups()[0]
if len(match.groups()[5]) == 0:
line += match.groups()[2]
line += match.groups()[3]
line += ", "
line += match.groups()[1]
line += "\n"
return line
for line in sys.stdin:
if line.find("getelementptr ") == line.find("getelementptr inbounds"):
if line.find("getelementptr inbounds") != line.find("getelementptr inbounds ("):
line = conv(re.match(ibrep, line), line)
elif line.find("getelementptr ") != line.find("getelementptr ("):
line = conv(re.match(normrep, line), line)
sys.stdout.write(line)
apply.sh:
for name in "$@"
do
python3 `dirname "$0"`/update.py < "$name" > "$name.tmp" && mv "$name.tmp" "$name"
rm -f "$name.tmp"
done
The actual commands:
From llvm/src:
find test/ -name *.ll | xargs ./apply.sh
From llvm/src/tools/clang:
find test/ -name *.mm -o -name *.m -o -name *.cpp -o -name *.c | xargs -I '{}' ../../apply.sh "{}"
From llvm/src/tools/polly:
find test/ -name *.ll | xargs ./apply.sh
After that, check-all (with llvm, clang, clang-tools-extra, lld,
compiler-rt, and polly all checked out).
The extra 'rm' in the apply.sh script is due to a few files in clang's test
suite using interesting unicode stuff that my python script was throwing
exceptions on. None of those files needed to be migrated, so it seemed
sufficient to ignore those cases.
Reviewers: rafael, dexonsmith, grosser
Differential Revision: http://reviews.llvm.org/D7636
llvm-svn: 230786
2015-02-28 03:29:02 +08:00
|
|
|
%gep = getelementptr inbounds float, <2 x float addrspace(3)*> %lds.ptr, <2 x i32> %index.1
|
2014-10-11 06:01:59 +08:00
|
|
|
%gep.0 = extractelement <2 x float addrspace(3)*> %gep, i32 0
|
|
|
|
|
%gep.1 = extractelement <2 x float addrspace(3)*> %gep, i32 1
|
2015-02-28 05:17:42 +08:00
|
|
|
%val0 = load float, float addrspace(3)* %gep.0, align 4
|
|
|
|
|
%val1 = load float, float addrspace(3)* %gep.1, align 4
|
2014-10-11 06:01:59 +08:00
|
|
|
%add.x = add nsw i32 %x.i, 8
|
|
|
|
|
%sum = fadd float %val0, %val1
|
[opaque pointer type] Add textual IR support for explicit type parameter to getelementptr instruction
One of several parallel first steps to remove the target type of pointers,
replacing them with a single opaque pointer type.
This adds an explicit type parameter to the gep instruction so that when the
first parameter becomes an opaque pointer type, the type to gep through is
still available to the instructions.
* This doesn't modify gep operators, only instructions (operators will be
handled separately)
* Textual IR changes only. Bitcode (including upgrade) and changing the
in-memory representation will be in separate changes.
* geps of vectors are transformed as:
getelementptr <4 x float*> %x, ...
->getelementptr float, <4 x float*> %x, ...
Then, once the opaque pointer type is introduced, this will ultimately look
like:
getelementptr float, <4 x ptr> %x
with the unambiguous interpretation that it is a vector of pointers to float.
* address spaces remain on the pointer, not the type:
getelementptr float addrspace(1)* %x
->getelementptr float, float addrspace(1)* %x
Then, eventually:
getelementptr float, ptr addrspace(1) %x
Importantly, the massive amount of test case churn has been automated by
same crappy python code. I had to manually update a few test cases that
wouldn't fit the script's model (r228970,r229196,r229197,r229198). The
python script just massages stdin and writes the result to stdout, I
then wrapped that in a shell script to handle replacing files, then
using the usual find+xargs to migrate all the files.
update.py:
import fileinput
import sys
import re
ibrep = re.compile(r"(^.*?[^%\w]getelementptr inbounds )(((?:<\d* x )?)(.*?)(| addrspace\(\d\)) *\*(|>)(?:$| *(?:%|@|null|undef|blockaddress|getelementptr|addrspacecast|bitcast|inttoptr|\[\[[a-zA-Z]|\{\{).*$))")
normrep = re.compile( r"(^.*?[^%\w]getelementptr )(((?:<\d* x )?)(.*?)(| addrspace\(\d\)) *\*(|>)(?:$| *(?:%|@|null|undef|blockaddress|getelementptr|addrspacecast|bitcast|inttoptr|\[\[[a-zA-Z]|\{\{).*$))")
def conv(match, line):
if not match:
return line
line = match.groups()[0]
if len(match.groups()[5]) == 0:
line += match.groups()[2]
line += match.groups()[3]
line += ", "
line += match.groups()[1]
line += "\n"
return line
for line in sys.stdin:
if line.find("getelementptr ") == line.find("getelementptr inbounds"):
if line.find("getelementptr inbounds") != line.find("getelementptr inbounds ("):
line = conv(re.match(ibrep, line), line)
elif line.find("getelementptr ") != line.find("getelementptr ("):
line = conv(re.match(normrep, line), line)
sys.stdout.write(line)
apply.sh:
for name in "$@"
do
python3 `dirname "$0"`/update.py < "$name" > "$name.tmp" && mv "$name.tmp" "$name"
rm -f "$name.tmp"
done
The actual commands:
From llvm/src:
find test/ -name *.ll | xargs ./apply.sh
From llvm/src/tools/clang:
find test/ -name *.mm -o -name *.m -o -name *.cpp -o -name *.c | xargs -I '{}' ../../apply.sh "{}"
From llvm/src/tools/polly:
find test/ -name *.ll | xargs ./apply.sh
After that, check-all (with llvm, clang, clang-tools-extra, lld,
compiler-rt, and polly all checked out).
The extra 'rm' in the apply.sh script is due to a few files in clang's test
suite using interesting unicode stuff that my python script was throwing
exceptions on. None of those files needed to be migrated, so it seemed
sufficient to ignore those cases.
Reviewers: rafael, dexonsmith, grosser
Differential Revision: http://reviews.llvm.org/D7636
llvm-svn: 230786
2015-02-28 03:29:02 +08:00
|
|
|
%out.gep = getelementptr inbounds float, float addrspace(1)* %out, i32 %x.i
|
2014-10-11 06:01:59 +08:00
|
|
|
store float %sum, float addrspace(1)* %out.gep, align 4
|
|
|
|
|
ret void
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
; Apply a constant scalar offset after the pointer vector extract. We
|
|
|
|
|
; are rejecting merges that have the same, constant 0 offset, so make
|
|
|
|
|
; sure we are really rejecting it because of the different
|
|
|
|
|
; subregisters.
|
|
|
|
|
|
|
|
|
|
; SI-LABEL: @read2_ptr_is_subreg_arg_offset_f32
|
2014-11-05 22:50:53 +08:00
|
|
|
; SI-NOT: ds_read2_b32
|
|
|
|
|
; SI: ds_read_b32
|
|
|
|
|
; SI: ds_read_b32
|
|
|
|
|
; SI: s_endpgm
|
2014-10-11 06:01:59 +08:00
|
|
|
define void @read2_ptr_is_subreg_arg_offset_f32(float addrspace(1)* %out, <2 x float addrspace(3)*> %lds.ptr) #0 {
|
|
|
|
|
%x.i = tail call i32 @llvm.r600.read.tidig.x() #1
|
|
|
|
|
%index.0 = insertelement <2 x i32> undef, i32 %x.i, i32 0
|
|
|
|
|
%index.1 = insertelement <2 x i32> %index.0, i32 8, i32 0
|
[opaque pointer type] Add textual IR support for explicit type parameter to getelementptr instruction
One of several parallel first steps to remove the target type of pointers,
replacing them with a single opaque pointer type.
This adds an explicit type parameter to the gep instruction so that when the
first parameter becomes an opaque pointer type, the type to gep through is
still available to the instructions.
* This doesn't modify gep operators, only instructions (operators will be
handled separately)
* Textual IR changes only. Bitcode (including upgrade) and changing the
in-memory representation will be in separate changes.
* geps of vectors are transformed as:
getelementptr <4 x float*> %x, ...
->getelementptr float, <4 x float*> %x, ...
Then, once the opaque pointer type is introduced, this will ultimately look
like:
getelementptr float, <4 x ptr> %x
with the unambiguous interpretation that it is a vector of pointers to float.
* address spaces remain on the pointer, not the type:
getelementptr float addrspace(1)* %x
->getelementptr float, float addrspace(1)* %x
Then, eventually:
getelementptr float, ptr addrspace(1) %x
Importantly, the massive amount of test case churn has been automated by
same crappy python code. I had to manually update a few test cases that
wouldn't fit the script's model (r228970,r229196,r229197,r229198). The
python script just massages stdin and writes the result to stdout, I
then wrapped that in a shell script to handle replacing files, then
using the usual find+xargs to migrate all the files.
update.py:
import fileinput
import sys
import re
ibrep = re.compile(r"(^.*?[^%\w]getelementptr inbounds )(((?:<\d* x )?)(.*?)(| addrspace\(\d\)) *\*(|>)(?:$| *(?:%|@|null|undef|blockaddress|getelementptr|addrspacecast|bitcast|inttoptr|\[\[[a-zA-Z]|\{\{).*$))")
normrep = re.compile( r"(^.*?[^%\w]getelementptr )(((?:<\d* x )?)(.*?)(| addrspace\(\d\)) *\*(|>)(?:$| *(?:%|@|null|undef|blockaddress|getelementptr|addrspacecast|bitcast|inttoptr|\[\[[a-zA-Z]|\{\{).*$))")
def conv(match, line):
if not match:
return line
line = match.groups()[0]
if len(match.groups()[5]) == 0:
line += match.groups()[2]
line += match.groups()[3]
line += ", "
line += match.groups()[1]
line += "\n"
return line
for line in sys.stdin:
if line.find("getelementptr ") == line.find("getelementptr inbounds"):
if line.find("getelementptr inbounds") != line.find("getelementptr inbounds ("):
line = conv(re.match(ibrep, line), line)
elif line.find("getelementptr ") != line.find("getelementptr ("):
line = conv(re.match(normrep, line), line)
sys.stdout.write(line)
apply.sh:
for name in "$@"
do
python3 `dirname "$0"`/update.py < "$name" > "$name.tmp" && mv "$name.tmp" "$name"
rm -f "$name.tmp"
done
The actual commands:
From llvm/src:
find test/ -name *.ll | xargs ./apply.sh
From llvm/src/tools/clang:
find test/ -name *.mm -o -name *.m -o -name *.cpp -o -name *.c | xargs -I '{}' ../../apply.sh "{}"
From llvm/src/tools/polly:
find test/ -name *.ll | xargs ./apply.sh
After that, check-all (with llvm, clang, clang-tools-extra, lld,
compiler-rt, and polly all checked out).
The extra 'rm' in the apply.sh script is due to a few files in clang's test
suite using interesting unicode stuff that my python script was throwing
exceptions on. None of those files needed to be migrated, so it seemed
sufficient to ignore those cases.
Reviewers: rafael, dexonsmith, grosser
Differential Revision: http://reviews.llvm.org/D7636
llvm-svn: 230786
2015-02-28 03:29:02 +08:00
|
|
|
%gep = getelementptr inbounds float, <2 x float addrspace(3)*> %lds.ptr, <2 x i32> %index.1
|
2014-10-11 06:01:59 +08:00
|
|
|
%gep.0 = extractelement <2 x float addrspace(3)*> %gep, i32 0
|
|
|
|
|
%gep.1 = extractelement <2 x float addrspace(3)*> %gep, i32 1
|
|
|
|
|
|
|
|
|
|
; Apply an additional offset after the vector that will be more obviously folded.
|
[opaque pointer type] Add textual IR support for explicit type parameter to getelementptr instruction
One of several parallel first steps to remove the target type of pointers,
replacing them with a single opaque pointer type.
This adds an explicit type parameter to the gep instruction so that when the
first parameter becomes an opaque pointer type, the type to gep through is
still available to the instructions.
* This doesn't modify gep operators, only instructions (operators will be
handled separately)
* Textual IR changes only. Bitcode (including upgrade) and changing the
in-memory representation will be in separate changes.
* geps of vectors are transformed as:
getelementptr <4 x float*> %x, ...
->getelementptr float, <4 x float*> %x, ...
Then, once the opaque pointer type is introduced, this will ultimately look
like:
getelementptr float, <4 x ptr> %x
with the unambiguous interpretation that it is a vector of pointers to float.
* address spaces remain on the pointer, not the type:
getelementptr float addrspace(1)* %x
->getelementptr float, float addrspace(1)* %x
Then, eventually:
getelementptr float, ptr addrspace(1) %x
Importantly, the massive amount of test case churn has been automated by
same crappy python code. I had to manually update a few test cases that
wouldn't fit the script's model (r228970,r229196,r229197,r229198). The
python script just massages stdin and writes the result to stdout, I
then wrapped that in a shell script to handle replacing files, then
using the usual find+xargs to migrate all the files.
update.py:
import fileinput
import sys
import re
ibrep = re.compile(r"(^.*?[^%\w]getelementptr inbounds )(((?:<\d* x )?)(.*?)(| addrspace\(\d\)) *\*(|>)(?:$| *(?:%|@|null|undef|blockaddress|getelementptr|addrspacecast|bitcast|inttoptr|\[\[[a-zA-Z]|\{\{).*$))")
normrep = re.compile( r"(^.*?[^%\w]getelementptr )(((?:<\d* x )?)(.*?)(| addrspace\(\d\)) *\*(|>)(?:$| *(?:%|@|null|undef|blockaddress|getelementptr|addrspacecast|bitcast|inttoptr|\[\[[a-zA-Z]|\{\{).*$))")
def conv(match, line):
if not match:
return line
line = match.groups()[0]
if len(match.groups()[5]) == 0:
line += match.groups()[2]
line += match.groups()[3]
line += ", "
line += match.groups()[1]
line += "\n"
return line
for line in sys.stdin:
if line.find("getelementptr ") == line.find("getelementptr inbounds"):
if line.find("getelementptr inbounds") != line.find("getelementptr inbounds ("):
line = conv(re.match(ibrep, line), line)
elif line.find("getelementptr ") != line.find("getelementptr ("):
line = conv(re.match(normrep, line), line)
sys.stdout.write(line)
apply.sh:
for name in "$@"
do
python3 `dirname "$0"`/update.py < "$name" > "$name.tmp" && mv "$name.tmp" "$name"
rm -f "$name.tmp"
done
The actual commands:
From llvm/src:
find test/ -name *.ll | xargs ./apply.sh
From llvm/src/tools/clang:
find test/ -name *.mm -o -name *.m -o -name *.cpp -o -name *.c | xargs -I '{}' ../../apply.sh "{}"
From llvm/src/tools/polly:
find test/ -name *.ll | xargs ./apply.sh
After that, check-all (with llvm, clang, clang-tools-extra, lld,
compiler-rt, and polly all checked out).
The extra 'rm' in the apply.sh script is due to a few files in clang's test
suite using interesting unicode stuff that my python script was throwing
exceptions on. None of those files needed to be migrated, so it seemed
sufficient to ignore those cases.
Reviewers: rafael, dexonsmith, grosser
Differential Revision: http://reviews.llvm.org/D7636
llvm-svn: 230786
2015-02-28 03:29:02 +08:00
|
|
|
%gep.1.offset = getelementptr float, float addrspace(3)* %gep.1, i32 8
|
2014-10-11 06:01:59 +08:00
|
|
|
|
2015-02-28 05:17:42 +08:00
|
|
|
%val0 = load float, float addrspace(3)* %gep.0, align 4
|
|
|
|
|
%val1 = load float, float addrspace(3)* %gep.1.offset, align 4
|
2014-10-11 06:01:59 +08:00
|
|
|
%add.x = add nsw i32 %x.i, 8
|
|
|
|
|
%sum = fadd float %val0, %val1
|
[opaque pointer type] Add textual IR support for explicit type parameter to getelementptr instruction
One of several parallel first steps to remove the target type of pointers,
replacing them with a single opaque pointer type.
This adds an explicit type parameter to the gep instruction so that when the
first parameter becomes an opaque pointer type, the type to gep through is
still available to the instructions.
* This doesn't modify gep operators, only instructions (operators will be
handled separately)
* Textual IR changes only. Bitcode (including upgrade) and changing the
in-memory representation will be in separate changes.
* geps of vectors are transformed as:
getelementptr <4 x float*> %x, ...
->getelementptr float, <4 x float*> %x, ...
Then, once the opaque pointer type is introduced, this will ultimately look
like:
getelementptr float, <4 x ptr> %x
with the unambiguous interpretation that it is a vector of pointers to float.
* address spaces remain on the pointer, not the type:
getelementptr float addrspace(1)* %x
->getelementptr float, float addrspace(1)* %x
Then, eventually:
getelementptr float, ptr addrspace(1) %x
Importantly, the massive amount of test case churn has been automated by
same crappy python code. I had to manually update a few test cases that
wouldn't fit the script's model (r228970,r229196,r229197,r229198). The
python script just massages stdin and writes the result to stdout, I
then wrapped that in a shell script to handle replacing files, then
using the usual find+xargs to migrate all the files.
update.py:
import fileinput
import sys
import re
ibrep = re.compile(r"(^.*?[^%\w]getelementptr inbounds )(((?:<\d* x )?)(.*?)(| addrspace\(\d\)) *\*(|>)(?:$| *(?:%|@|null|undef|blockaddress|getelementptr|addrspacecast|bitcast|inttoptr|\[\[[a-zA-Z]|\{\{).*$))")
normrep = re.compile( r"(^.*?[^%\w]getelementptr )(((?:<\d* x )?)(.*?)(| addrspace\(\d\)) *\*(|>)(?:$| *(?:%|@|null|undef|blockaddress|getelementptr|addrspacecast|bitcast|inttoptr|\[\[[a-zA-Z]|\{\{).*$))")
def conv(match, line):
if not match:
return line
line = match.groups()[0]
if len(match.groups()[5]) == 0:
line += match.groups()[2]
line += match.groups()[3]
line += ", "
line += match.groups()[1]
line += "\n"
return line
for line in sys.stdin:
if line.find("getelementptr ") == line.find("getelementptr inbounds"):
if line.find("getelementptr inbounds") != line.find("getelementptr inbounds ("):
line = conv(re.match(ibrep, line), line)
elif line.find("getelementptr ") != line.find("getelementptr ("):
line = conv(re.match(normrep, line), line)
sys.stdout.write(line)
apply.sh:
for name in "$@"
do
python3 `dirname "$0"`/update.py < "$name" > "$name.tmp" && mv "$name.tmp" "$name"
rm -f "$name.tmp"
done
The actual commands:
From llvm/src:
find test/ -name *.ll | xargs ./apply.sh
From llvm/src/tools/clang:
find test/ -name *.mm -o -name *.m -o -name *.cpp -o -name *.c | xargs -I '{}' ../../apply.sh "{}"
From llvm/src/tools/polly:
find test/ -name *.ll | xargs ./apply.sh
After that, check-all (with llvm, clang, clang-tools-extra, lld,
compiler-rt, and polly all checked out).
The extra 'rm' in the apply.sh script is due to a few files in clang's test
suite using interesting unicode stuff that my python script was throwing
exceptions on. None of those files needed to be migrated, so it seemed
sufficient to ignore those cases.
Reviewers: rafael, dexonsmith, grosser
Differential Revision: http://reviews.llvm.org/D7636
llvm-svn: 230786
2015-02-28 03:29:02 +08:00
|
|
|
%out.gep = getelementptr inbounds float, float addrspace(1)* %out, i32 %x.i
|
2014-10-11 06:01:59 +08:00
|
|
|
store float %sum, float addrspace(1)* %out.gep, align 4
|
|
|
|
|
ret void
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
; We should be able to merge in this case, but probably not worth the effort.
|
2014-11-05 22:50:53 +08:00
|
|
|
; SI-NOT: ds_read2_b32
|
|
|
|
|
; SI: ds_read_b32
|
|
|
|
|
; SI: ds_read_b32
|
|
|
|
|
; SI: s_endpgm
|
2014-10-11 06:01:59 +08:00
|
|
|
define void @read2_ptr_is_subreg_f32(float addrspace(1)* %out) #0 {
|
|
|
|
|
%x.i = tail call i32 @llvm.r600.read.tidig.x() #1
|
|
|
|
|
%ptr.0 = insertelement <2 x [512 x float] addrspace(3)*> undef, [512 x float] addrspace(3)* @lds, i32 0
|
|
|
|
|
%ptr.1 = insertelement <2 x [512 x float] addrspace(3)*> %ptr.0, [512 x float] addrspace(3)* @lds, i32 1
|
|
|
|
|
%x.i.v.0 = insertelement <2 x i32> undef, i32 %x.i, i32 0
|
|
|
|
|
%x.i.v.1 = insertelement <2 x i32> %x.i.v.0, i32 %x.i, i32 1
|
|
|
|
|
%idx = add <2 x i32> %x.i.v.1, <i32 0, i32 8>
|
[opaque pointer type] Add textual IR support for explicit type parameter to getelementptr instruction
One of several parallel first steps to remove the target type of pointers,
replacing them with a single opaque pointer type.
This adds an explicit type parameter to the gep instruction so that when the
first parameter becomes an opaque pointer type, the type to gep through is
still available to the instructions.
* This doesn't modify gep operators, only instructions (operators will be
handled separately)
* Textual IR changes only. Bitcode (including upgrade) and changing the
in-memory representation will be in separate changes.
* geps of vectors are transformed as:
getelementptr <4 x float*> %x, ...
->getelementptr float, <4 x float*> %x, ...
Then, once the opaque pointer type is introduced, this will ultimately look
like:
getelementptr float, <4 x ptr> %x
with the unambiguous interpretation that it is a vector of pointers to float.
* address spaces remain on the pointer, not the type:
getelementptr float addrspace(1)* %x
->getelementptr float, float addrspace(1)* %x
Then, eventually:
getelementptr float, ptr addrspace(1) %x
Importantly, the massive amount of test case churn has been automated by
same crappy python code. I had to manually update a few test cases that
wouldn't fit the script's model (r228970,r229196,r229197,r229198). The
python script just massages stdin and writes the result to stdout, I
then wrapped that in a shell script to handle replacing files, then
using the usual find+xargs to migrate all the files.
update.py:
import fileinput
import sys
import re
ibrep = re.compile(r"(^.*?[^%\w]getelementptr inbounds )(((?:<\d* x )?)(.*?)(| addrspace\(\d\)) *\*(|>)(?:$| *(?:%|@|null|undef|blockaddress|getelementptr|addrspacecast|bitcast|inttoptr|\[\[[a-zA-Z]|\{\{).*$))")
normrep = re.compile( r"(^.*?[^%\w]getelementptr )(((?:<\d* x )?)(.*?)(| addrspace\(\d\)) *\*(|>)(?:$| *(?:%|@|null|undef|blockaddress|getelementptr|addrspacecast|bitcast|inttoptr|\[\[[a-zA-Z]|\{\{).*$))")
def conv(match, line):
if not match:
return line
line = match.groups()[0]
if len(match.groups()[5]) == 0:
line += match.groups()[2]
line += match.groups()[3]
line += ", "
line += match.groups()[1]
line += "\n"
return line
for line in sys.stdin:
if line.find("getelementptr ") == line.find("getelementptr inbounds"):
if line.find("getelementptr inbounds") != line.find("getelementptr inbounds ("):
line = conv(re.match(ibrep, line), line)
elif line.find("getelementptr ") != line.find("getelementptr ("):
line = conv(re.match(normrep, line), line)
sys.stdout.write(line)
apply.sh:
for name in "$@"
do
python3 `dirname "$0"`/update.py < "$name" > "$name.tmp" && mv "$name.tmp" "$name"
rm -f "$name.tmp"
done
The actual commands:
From llvm/src:
find test/ -name *.ll | xargs ./apply.sh
From llvm/src/tools/clang:
find test/ -name *.mm -o -name *.m -o -name *.cpp -o -name *.c | xargs -I '{}' ../../apply.sh "{}"
From llvm/src/tools/polly:
find test/ -name *.ll | xargs ./apply.sh
After that, check-all (with llvm, clang, clang-tools-extra, lld,
compiler-rt, and polly all checked out).
The extra 'rm' in the apply.sh script is due to a few files in clang's test
suite using interesting unicode stuff that my python script was throwing
exceptions on. None of those files needed to be migrated, so it seemed
sufficient to ignore those cases.
Reviewers: rafael, dexonsmith, grosser
Differential Revision: http://reviews.llvm.org/D7636
llvm-svn: 230786
2015-02-28 03:29:02 +08:00
|
|
|
%gep = getelementptr inbounds [512 x float], <2 x [512 x float] addrspace(3)*> %ptr.1, <2 x i32> <i32 0, i32 0>, <2 x i32> %idx
|
2014-10-11 06:01:59 +08:00
|
|
|
%gep.0 = extractelement <2 x float addrspace(3)*> %gep, i32 0
|
|
|
|
|
%gep.1 = extractelement <2 x float addrspace(3)*> %gep, i32 1
|
2015-02-28 05:17:42 +08:00
|
|
|
%val0 = load float, float addrspace(3)* %gep.0, align 4
|
|
|
|
|
%val1 = load float, float addrspace(3)* %gep.1, align 4
|
2014-10-11 06:01:59 +08:00
|
|
|
%add.x = add nsw i32 %x.i, 8
|
|
|
|
|
%sum = fadd float %val0, %val1
|
[opaque pointer type] Add textual IR support for explicit type parameter to getelementptr instruction
One of several parallel first steps to remove the target type of pointers,
replacing them with a single opaque pointer type.
This adds an explicit type parameter to the gep instruction so that when the
first parameter becomes an opaque pointer type, the type to gep through is
still available to the instructions.
* This doesn't modify gep operators, only instructions (operators will be
handled separately)
* Textual IR changes only. Bitcode (including upgrade) and changing the
in-memory representation will be in separate changes.
* geps of vectors are transformed as:
getelementptr <4 x float*> %x, ...
->getelementptr float, <4 x float*> %x, ...
Then, once the opaque pointer type is introduced, this will ultimately look
like:
getelementptr float, <4 x ptr> %x
with the unambiguous interpretation that it is a vector of pointers to float.
* address spaces remain on the pointer, not the type:
getelementptr float addrspace(1)* %x
->getelementptr float, float addrspace(1)* %x
Then, eventually:
getelementptr float, ptr addrspace(1) %x
Importantly, the massive amount of test case churn has been automated by
same crappy python code. I had to manually update a few test cases that
wouldn't fit the script's model (r228970,r229196,r229197,r229198). The
python script just massages stdin and writes the result to stdout, I
then wrapped that in a shell script to handle replacing files, then
using the usual find+xargs to migrate all the files.
update.py:
import fileinput
import sys
import re
ibrep = re.compile(r"(^.*?[^%\w]getelementptr inbounds )(((?:<\d* x )?)(.*?)(| addrspace\(\d\)) *\*(|>)(?:$| *(?:%|@|null|undef|blockaddress|getelementptr|addrspacecast|bitcast|inttoptr|\[\[[a-zA-Z]|\{\{).*$))")
normrep = re.compile( r"(^.*?[^%\w]getelementptr )(((?:<\d* x )?)(.*?)(| addrspace\(\d\)) *\*(|>)(?:$| *(?:%|@|null|undef|blockaddress|getelementptr|addrspacecast|bitcast|inttoptr|\[\[[a-zA-Z]|\{\{).*$))")
def conv(match, line):
if not match:
return line
line = match.groups()[0]
if len(match.groups()[5]) == 0:
line += match.groups()[2]
line += match.groups()[3]
line += ", "
line += match.groups()[1]
line += "\n"
return line
for line in sys.stdin:
if line.find("getelementptr ") == line.find("getelementptr inbounds"):
if line.find("getelementptr inbounds") != line.find("getelementptr inbounds ("):
line = conv(re.match(ibrep, line), line)
elif line.find("getelementptr ") != line.find("getelementptr ("):
line = conv(re.match(normrep, line), line)
sys.stdout.write(line)
apply.sh:
for name in "$@"
do
python3 `dirname "$0"`/update.py < "$name" > "$name.tmp" && mv "$name.tmp" "$name"
rm -f "$name.tmp"
done
The actual commands:
From llvm/src:
find test/ -name *.ll | xargs ./apply.sh
From llvm/src/tools/clang:
find test/ -name *.mm -o -name *.m -o -name *.cpp -o -name *.c | xargs -I '{}' ../../apply.sh "{}"
From llvm/src/tools/polly:
find test/ -name *.ll | xargs ./apply.sh
After that, check-all (with llvm, clang, clang-tools-extra, lld,
compiler-rt, and polly all checked out).
The extra 'rm' in the apply.sh script is due to a few files in clang's test
suite using interesting unicode stuff that my python script was throwing
exceptions on. None of those files needed to be migrated, so it seemed
sufficient to ignore those cases.
Reviewers: rafael, dexonsmith, grosser
Differential Revision: http://reviews.llvm.org/D7636
llvm-svn: 230786
2015-02-28 03:29:02 +08:00
|
|
|
%out.gep = getelementptr inbounds float, float addrspace(1)* %out, i32 %x.i
|
2014-10-11 06:01:59 +08:00
|
|
|
store float %sum, float addrspace(1)* %out.gep, align 4
|
|
|
|
|
ret void
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
; SI-LABEL: @simple_read2_f32_volatile_0
|
2014-11-05 22:50:53 +08:00
|
|
|
; SI-NOT ds_read2_b32
|
|
|
|
|
; SI: ds_read_b32 v{{[0-9]+}}, v{{[0-9]+}}
|
|
|
|
|
; SI: ds_read_b32 v{{[0-9]+}}, v{{[0-9]+}} offset:32
|
|
|
|
|
; SI: s_endpgm
|
2014-10-11 06:01:59 +08:00
|
|
|
define void @simple_read2_f32_volatile_0(float addrspace(1)* %out) #0 {
|
|
|
|
|
%x.i = tail call i32 @llvm.r600.read.tidig.x() #1
|
[opaque pointer type] Add textual IR support for explicit type parameter to getelementptr instruction
One of several parallel first steps to remove the target type of pointers,
replacing them with a single opaque pointer type.
This adds an explicit type parameter to the gep instruction so that when the
first parameter becomes an opaque pointer type, the type to gep through is
still available to the instructions.
* This doesn't modify gep operators, only instructions (operators will be
handled separately)
* Textual IR changes only. Bitcode (including upgrade) and changing the
in-memory representation will be in separate changes.
* geps of vectors are transformed as:
getelementptr <4 x float*> %x, ...
->getelementptr float, <4 x float*> %x, ...
Then, once the opaque pointer type is introduced, this will ultimately look
like:
getelementptr float, <4 x ptr> %x
with the unambiguous interpretation that it is a vector of pointers to float.
* address spaces remain on the pointer, not the type:
getelementptr float addrspace(1)* %x
->getelementptr float, float addrspace(1)* %x
Then, eventually:
getelementptr float, ptr addrspace(1) %x
Importantly, the massive amount of test case churn has been automated by
same crappy python code. I had to manually update a few test cases that
wouldn't fit the script's model (r228970,r229196,r229197,r229198). The
python script just massages stdin and writes the result to stdout, I
then wrapped that in a shell script to handle replacing files, then
using the usual find+xargs to migrate all the files.
update.py:
import fileinput
import sys
import re
ibrep = re.compile(r"(^.*?[^%\w]getelementptr inbounds )(((?:<\d* x )?)(.*?)(| addrspace\(\d\)) *\*(|>)(?:$| *(?:%|@|null|undef|blockaddress|getelementptr|addrspacecast|bitcast|inttoptr|\[\[[a-zA-Z]|\{\{).*$))")
normrep = re.compile( r"(^.*?[^%\w]getelementptr )(((?:<\d* x )?)(.*?)(| addrspace\(\d\)) *\*(|>)(?:$| *(?:%|@|null|undef|blockaddress|getelementptr|addrspacecast|bitcast|inttoptr|\[\[[a-zA-Z]|\{\{).*$))")
def conv(match, line):
if not match:
return line
line = match.groups()[0]
if len(match.groups()[5]) == 0:
line += match.groups()[2]
line += match.groups()[3]
line += ", "
line += match.groups()[1]
line += "\n"
return line
for line in sys.stdin:
if line.find("getelementptr ") == line.find("getelementptr inbounds"):
if line.find("getelementptr inbounds") != line.find("getelementptr inbounds ("):
line = conv(re.match(ibrep, line), line)
elif line.find("getelementptr ") != line.find("getelementptr ("):
line = conv(re.match(normrep, line), line)
sys.stdout.write(line)
apply.sh:
for name in "$@"
do
python3 `dirname "$0"`/update.py < "$name" > "$name.tmp" && mv "$name.tmp" "$name"
rm -f "$name.tmp"
done
The actual commands:
From llvm/src:
find test/ -name *.ll | xargs ./apply.sh
From llvm/src/tools/clang:
find test/ -name *.mm -o -name *.m -o -name *.cpp -o -name *.c | xargs -I '{}' ../../apply.sh "{}"
From llvm/src/tools/polly:
find test/ -name *.ll | xargs ./apply.sh
After that, check-all (with llvm, clang, clang-tools-extra, lld,
compiler-rt, and polly all checked out).
The extra 'rm' in the apply.sh script is due to a few files in clang's test
suite using interesting unicode stuff that my python script was throwing
exceptions on. None of those files needed to be migrated, so it seemed
sufficient to ignore those cases.
Reviewers: rafael, dexonsmith, grosser
Differential Revision: http://reviews.llvm.org/D7636
llvm-svn: 230786
2015-02-28 03:29:02 +08:00
|
|
|
%arrayidx0 = getelementptr inbounds [512 x float], [512 x float] addrspace(3)* @lds, i32 0, i32 %x.i
|
2015-02-28 05:17:42 +08:00
|
|
|
%val0 = load volatile float, float addrspace(3)* %arrayidx0, align 4
|
2014-10-11 06:01:59 +08:00
|
|
|
%add.x = add nsw i32 %x.i, 8
|
[opaque pointer type] Add textual IR support for explicit type parameter to getelementptr instruction
One of several parallel first steps to remove the target type of pointers,
replacing them with a single opaque pointer type.
This adds an explicit type parameter to the gep instruction so that when the
first parameter becomes an opaque pointer type, the type to gep through is
still available to the instructions.
* This doesn't modify gep operators, only instructions (operators will be
handled separately)
* Textual IR changes only. Bitcode (including upgrade) and changing the
in-memory representation will be in separate changes.
* geps of vectors are transformed as:
getelementptr <4 x float*> %x, ...
->getelementptr float, <4 x float*> %x, ...
Then, once the opaque pointer type is introduced, this will ultimately look
like:
getelementptr float, <4 x ptr> %x
with the unambiguous interpretation that it is a vector of pointers to float.
* address spaces remain on the pointer, not the type:
getelementptr float addrspace(1)* %x
->getelementptr float, float addrspace(1)* %x
Then, eventually:
getelementptr float, ptr addrspace(1) %x
Importantly, the massive amount of test case churn has been automated by
same crappy python code. I had to manually update a few test cases that
wouldn't fit the script's model (r228970,r229196,r229197,r229198). The
python script just massages stdin and writes the result to stdout, I
then wrapped that in a shell script to handle replacing files, then
using the usual find+xargs to migrate all the files.
update.py:
import fileinput
import sys
import re
ibrep = re.compile(r"(^.*?[^%\w]getelementptr inbounds )(((?:<\d* x )?)(.*?)(| addrspace\(\d\)) *\*(|>)(?:$| *(?:%|@|null|undef|blockaddress|getelementptr|addrspacecast|bitcast|inttoptr|\[\[[a-zA-Z]|\{\{).*$))")
normrep = re.compile( r"(^.*?[^%\w]getelementptr )(((?:<\d* x )?)(.*?)(| addrspace\(\d\)) *\*(|>)(?:$| *(?:%|@|null|undef|blockaddress|getelementptr|addrspacecast|bitcast|inttoptr|\[\[[a-zA-Z]|\{\{).*$))")
def conv(match, line):
if not match:
return line
line = match.groups()[0]
if len(match.groups()[5]) == 0:
line += match.groups()[2]
line += match.groups()[3]
line += ", "
line += match.groups()[1]
line += "\n"
return line
for line in sys.stdin:
if line.find("getelementptr ") == line.find("getelementptr inbounds"):
if line.find("getelementptr inbounds") != line.find("getelementptr inbounds ("):
line = conv(re.match(ibrep, line), line)
elif line.find("getelementptr ") != line.find("getelementptr ("):
line = conv(re.match(normrep, line), line)
sys.stdout.write(line)
apply.sh:
for name in "$@"
do
python3 `dirname "$0"`/update.py < "$name" > "$name.tmp" && mv "$name.tmp" "$name"
rm -f "$name.tmp"
done
The actual commands:
From llvm/src:
find test/ -name *.ll | xargs ./apply.sh
From llvm/src/tools/clang:
find test/ -name *.mm -o -name *.m -o -name *.cpp -o -name *.c | xargs -I '{}' ../../apply.sh "{}"
From llvm/src/tools/polly:
find test/ -name *.ll | xargs ./apply.sh
After that, check-all (with llvm, clang, clang-tools-extra, lld,
compiler-rt, and polly all checked out).
The extra 'rm' in the apply.sh script is due to a few files in clang's test
suite using interesting unicode stuff that my python script was throwing
exceptions on. None of those files needed to be migrated, so it seemed
sufficient to ignore those cases.
Reviewers: rafael, dexonsmith, grosser
Differential Revision: http://reviews.llvm.org/D7636
llvm-svn: 230786
2015-02-28 03:29:02 +08:00
|
|
|
%arrayidx1 = getelementptr inbounds [512 x float], [512 x float] addrspace(3)* @lds, i32 0, i32 %add.x
|
2015-02-28 05:17:42 +08:00
|
|
|
%val1 = load float, float addrspace(3)* %arrayidx1, align 4
|
2014-10-11 06:01:59 +08:00
|
|
|
%sum = fadd float %val0, %val1
|
[opaque pointer type] Add textual IR support for explicit type parameter to getelementptr instruction
One of several parallel first steps to remove the target type of pointers,
replacing them with a single opaque pointer type.
This adds an explicit type parameter to the gep instruction so that when the
first parameter becomes an opaque pointer type, the type to gep through is
still available to the instructions.
* This doesn't modify gep operators, only instructions (operators will be
handled separately)
* Textual IR changes only. Bitcode (including upgrade) and changing the
in-memory representation will be in separate changes.
* geps of vectors are transformed as:
getelementptr <4 x float*> %x, ...
->getelementptr float, <4 x float*> %x, ...
Then, once the opaque pointer type is introduced, this will ultimately look
like:
getelementptr float, <4 x ptr> %x
with the unambiguous interpretation that it is a vector of pointers to float.
* address spaces remain on the pointer, not the type:
getelementptr float addrspace(1)* %x
->getelementptr float, float addrspace(1)* %x
Then, eventually:
getelementptr float, ptr addrspace(1) %x
Importantly, the massive amount of test case churn has been automated by
same crappy python code. I had to manually update a few test cases that
wouldn't fit the script's model (r228970,r229196,r229197,r229198). The
python script just massages stdin and writes the result to stdout, I
then wrapped that in a shell script to handle replacing files, then
using the usual find+xargs to migrate all the files.
update.py:
import fileinput
import sys
import re
ibrep = re.compile(r"(^.*?[^%\w]getelementptr inbounds )(((?:<\d* x )?)(.*?)(| addrspace\(\d\)) *\*(|>)(?:$| *(?:%|@|null|undef|blockaddress|getelementptr|addrspacecast|bitcast|inttoptr|\[\[[a-zA-Z]|\{\{).*$))")
normrep = re.compile( r"(^.*?[^%\w]getelementptr )(((?:<\d* x )?)(.*?)(| addrspace\(\d\)) *\*(|>)(?:$| *(?:%|@|null|undef|blockaddress|getelementptr|addrspacecast|bitcast|inttoptr|\[\[[a-zA-Z]|\{\{).*$))")
def conv(match, line):
if not match:
return line
line = match.groups()[0]
if len(match.groups()[5]) == 0:
line += match.groups()[2]
line += match.groups()[3]
line += ", "
line += match.groups()[1]
line += "\n"
return line
for line in sys.stdin:
if line.find("getelementptr ") == line.find("getelementptr inbounds"):
if line.find("getelementptr inbounds") != line.find("getelementptr inbounds ("):
line = conv(re.match(ibrep, line), line)
elif line.find("getelementptr ") != line.find("getelementptr ("):
line = conv(re.match(normrep, line), line)
sys.stdout.write(line)
apply.sh:
for name in "$@"
do
python3 `dirname "$0"`/update.py < "$name" > "$name.tmp" && mv "$name.tmp" "$name"
rm -f "$name.tmp"
done
The actual commands:
From llvm/src:
find test/ -name *.ll | xargs ./apply.sh
From llvm/src/tools/clang:
find test/ -name *.mm -o -name *.m -o -name *.cpp -o -name *.c | xargs -I '{}' ../../apply.sh "{}"
From llvm/src/tools/polly:
find test/ -name *.ll | xargs ./apply.sh
After that, check-all (with llvm, clang, clang-tools-extra, lld,
compiler-rt, and polly all checked out).
The extra 'rm' in the apply.sh script is due to a few files in clang's test
suite using interesting unicode stuff that my python script was throwing
exceptions on. None of those files needed to be migrated, so it seemed
sufficient to ignore those cases.
Reviewers: rafael, dexonsmith, grosser
Differential Revision: http://reviews.llvm.org/D7636
llvm-svn: 230786
2015-02-28 03:29:02 +08:00
|
|
|
%out.gep = getelementptr inbounds float, float addrspace(1)* %out, i32 %x.i
|
2014-10-11 06:01:59 +08:00
|
|
|
store float %sum, float addrspace(1)* %out.gep, align 4
|
|
|
|
|
ret void
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
; SI-LABEL: @simple_read2_f32_volatile_1
|
2014-11-05 22:50:53 +08:00
|
|
|
; SI-NOT ds_read2_b32
|
|
|
|
|
; SI: ds_read_b32 v{{[0-9]+}}, v{{[0-9]+}}
|
|
|
|
|
; SI: ds_read_b32 v{{[0-9]+}}, v{{[0-9]+}} offset:32
|
|
|
|
|
; SI: s_endpgm
|
2014-10-11 06:01:59 +08:00
|
|
|
define void @simple_read2_f32_volatile_1(float addrspace(1)* %out) #0 {
|
|
|
|
|
%x.i = tail call i32 @llvm.r600.read.tidig.x() #1
|
[opaque pointer type] Add textual IR support for explicit type parameter to getelementptr instruction
One of several parallel first steps to remove the target type of pointers,
replacing them with a single opaque pointer type.
This adds an explicit type parameter to the gep instruction so that when the
first parameter becomes an opaque pointer type, the type to gep through is
still available to the instructions.
* This doesn't modify gep operators, only instructions (operators will be
handled separately)
* Textual IR changes only. Bitcode (including upgrade) and changing the
in-memory representation will be in separate changes.
* geps of vectors are transformed as:
getelementptr <4 x float*> %x, ...
->getelementptr float, <4 x float*> %x, ...
Then, once the opaque pointer type is introduced, this will ultimately look
like:
getelementptr float, <4 x ptr> %x
with the unambiguous interpretation that it is a vector of pointers to float.
* address spaces remain on the pointer, not the type:
getelementptr float addrspace(1)* %x
->getelementptr float, float addrspace(1)* %x
Then, eventually:
getelementptr float, ptr addrspace(1) %x
Importantly, the massive amount of test case churn has been automated by
same crappy python code. I had to manually update a few test cases that
wouldn't fit the script's model (r228970,r229196,r229197,r229198). The
python script just massages stdin and writes the result to stdout, I
then wrapped that in a shell script to handle replacing files, then
using the usual find+xargs to migrate all the files.
update.py:
import fileinput
import sys
import re
ibrep = re.compile(r"(^.*?[^%\w]getelementptr inbounds )(((?:<\d* x )?)(.*?)(| addrspace\(\d\)) *\*(|>)(?:$| *(?:%|@|null|undef|blockaddress|getelementptr|addrspacecast|bitcast|inttoptr|\[\[[a-zA-Z]|\{\{).*$))")
normrep = re.compile( r"(^.*?[^%\w]getelementptr )(((?:<\d* x )?)(.*?)(| addrspace\(\d\)) *\*(|>)(?:$| *(?:%|@|null|undef|blockaddress|getelementptr|addrspacecast|bitcast|inttoptr|\[\[[a-zA-Z]|\{\{).*$))")
def conv(match, line):
if not match:
return line
line = match.groups()[0]
if len(match.groups()[5]) == 0:
line += match.groups()[2]
line += match.groups()[3]
line += ", "
line += match.groups()[1]
line += "\n"
return line
for line in sys.stdin:
if line.find("getelementptr ") == line.find("getelementptr inbounds"):
if line.find("getelementptr inbounds") != line.find("getelementptr inbounds ("):
line = conv(re.match(ibrep, line), line)
elif line.find("getelementptr ") != line.find("getelementptr ("):
line = conv(re.match(normrep, line), line)
sys.stdout.write(line)
apply.sh:
for name in "$@"
do
python3 `dirname "$0"`/update.py < "$name" > "$name.tmp" && mv "$name.tmp" "$name"
rm -f "$name.tmp"
done
The actual commands:
From llvm/src:
find test/ -name *.ll | xargs ./apply.sh
From llvm/src/tools/clang:
find test/ -name *.mm -o -name *.m -o -name *.cpp -o -name *.c | xargs -I '{}' ../../apply.sh "{}"
From llvm/src/tools/polly:
find test/ -name *.ll | xargs ./apply.sh
After that, check-all (with llvm, clang, clang-tools-extra, lld,
compiler-rt, and polly all checked out).
The extra 'rm' in the apply.sh script is due to a few files in clang's test
suite using interesting unicode stuff that my python script was throwing
exceptions on. None of those files needed to be migrated, so it seemed
sufficient to ignore those cases.
Reviewers: rafael, dexonsmith, grosser
Differential Revision: http://reviews.llvm.org/D7636
llvm-svn: 230786
2015-02-28 03:29:02 +08:00
|
|
|
%arrayidx0 = getelementptr inbounds [512 x float], [512 x float] addrspace(3)* @lds, i32 0, i32 %x.i
|
2015-02-28 05:17:42 +08:00
|
|
|
%val0 = load float, float addrspace(3)* %arrayidx0, align 4
|
2014-10-11 06:01:59 +08:00
|
|
|
%add.x = add nsw i32 %x.i, 8
|
[opaque pointer type] Add textual IR support for explicit type parameter to getelementptr instruction
One of several parallel first steps to remove the target type of pointers,
replacing them with a single opaque pointer type.
This adds an explicit type parameter to the gep instruction so that when the
first parameter becomes an opaque pointer type, the type to gep through is
still available to the instructions.
* This doesn't modify gep operators, only instructions (operators will be
handled separately)
* Textual IR changes only. Bitcode (including upgrade) and changing the
in-memory representation will be in separate changes.
* geps of vectors are transformed as:
getelementptr <4 x float*> %x, ...
->getelementptr float, <4 x float*> %x, ...
Then, once the opaque pointer type is introduced, this will ultimately look
like:
getelementptr float, <4 x ptr> %x
with the unambiguous interpretation that it is a vector of pointers to float.
* address spaces remain on the pointer, not the type:
getelementptr float addrspace(1)* %x
->getelementptr float, float addrspace(1)* %x
Then, eventually:
getelementptr float, ptr addrspace(1) %x
Importantly, the massive amount of test case churn has been automated by
same crappy python code. I had to manually update a few test cases that
wouldn't fit the script's model (r228970,r229196,r229197,r229198). The
python script just massages stdin and writes the result to stdout, I
then wrapped that in a shell script to handle replacing files, then
using the usual find+xargs to migrate all the files.
update.py:
import fileinput
import sys
import re
ibrep = re.compile(r"(^.*?[^%\w]getelementptr inbounds )(((?:<\d* x )?)(.*?)(| addrspace\(\d\)) *\*(|>)(?:$| *(?:%|@|null|undef|blockaddress|getelementptr|addrspacecast|bitcast|inttoptr|\[\[[a-zA-Z]|\{\{).*$))")
normrep = re.compile( r"(^.*?[^%\w]getelementptr )(((?:<\d* x )?)(.*?)(| addrspace\(\d\)) *\*(|>)(?:$| *(?:%|@|null|undef|blockaddress|getelementptr|addrspacecast|bitcast|inttoptr|\[\[[a-zA-Z]|\{\{).*$))")
def conv(match, line):
if not match:
return line
line = match.groups()[0]
if len(match.groups()[5]) == 0:
line += match.groups()[2]
line += match.groups()[3]
line += ", "
line += match.groups()[1]
line += "\n"
return line
for line in sys.stdin:
if line.find("getelementptr ") == line.find("getelementptr inbounds"):
if line.find("getelementptr inbounds") != line.find("getelementptr inbounds ("):
line = conv(re.match(ibrep, line), line)
elif line.find("getelementptr ") != line.find("getelementptr ("):
line = conv(re.match(normrep, line), line)
sys.stdout.write(line)
apply.sh:
for name in "$@"
do
python3 `dirname "$0"`/update.py < "$name" > "$name.tmp" && mv "$name.tmp" "$name"
rm -f "$name.tmp"
done
The actual commands:
From llvm/src:
find test/ -name *.ll | xargs ./apply.sh
From llvm/src/tools/clang:
find test/ -name *.mm -o -name *.m -o -name *.cpp -o -name *.c | xargs -I '{}' ../../apply.sh "{}"
From llvm/src/tools/polly:
find test/ -name *.ll | xargs ./apply.sh
After that, check-all (with llvm, clang, clang-tools-extra, lld,
compiler-rt, and polly all checked out).
The extra 'rm' in the apply.sh script is due to a few files in clang's test
suite using interesting unicode stuff that my python script was throwing
exceptions on. None of those files needed to be migrated, so it seemed
sufficient to ignore those cases.
Reviewers: rafael, dexonsmith, grosser
Differential Revision: http://reviews.llvm.org/D7636
llvm-svn: 230786
2015-02-28 03:29:02 +08:00
|
|
|
%arrayidx1 = getelementptr inbounds [512 x float], [512 x float] addrspace(3)* @lds, i32 0, i32 %add.x
|
2015-02-28 05:17:42 +08:00
|
|
|
%val1 = load volatile float, float addrspace(3)* %arrayidx1, align 4
|
2014-10-11 06:01:59 +08:00
|
|
|
%sum = fadd float %val0, %val1
|
[opaque pointer type] Add textual IR support for explicit type parameter to getelementptr instruction
One of several parallel first steps to remove the target type of pointers,
replacing them with a single opaque pointer type.
This adds an explicit type parameter to the gep instruction so that when the
first parameter becomes an opaque pointer type, the type to gep through is
still available to the instructions.
* This doesn't modify gep operators, only instructions (operators will be
handled separately)
* Textual IR changes only. Bitcode (including upgrade) and changing the
in-memory representation will be in separate changes.
* geps of vectors are transformed as:
getelementptr <4 x float*> %x, ...
->getelementptr float, <4 x float*> %x, ...
Then, once the opaque pointer type is introduced, this will ultimately look
like:
getelementptr float, <4 x ptr> %x
with the unambiguous interpretation that it is a vector of pointers to float.
* address spaces remain on the pointer, not the type:
getelementptr float addrspace(1)* %x
->getelementptr float, float addrspace(1)* %x
Then, eventually:
getelementptr float, ptr addrspace(1) %x
Importantly, the massive amount of test case churn has been automated by
same crappy python code. I had to manually update a few test cases that
wouldn't fit the script's model (r228970,r229196,r229197,r229198). The
python script just massages stdin and writes the result to stdout, I
then wrapped that in a shell script to handle replacing files, then
using the usual find+xargs to migrate all the files.
update.py:
import fileinput
import sys
import re
ibrep = re.compile(r"(^.*?[^%\w]getelementptr inbounds )(((?:<\d* x )?)(.*?)(| addrspace\(\d\)) *\*(|>)(?:$| *(?:%|@|null|undef|blockaddress|getelementptr|addrspacecast|bitcast|inttoptr|\[\[[a-zA-Z]|\{\{).*$))")
normrep = re.compile( r"(^.*?[^%\w]getelementptr )(((?:<\d* x )?)(.*?)(| addrspace\(\d\)) *\*(|>)(?:$| *(?:%|@|null|undef|blockaddress|getelementptr|addrspacecast|bitcast|inttoptr|\[\[[a-zA-Z]|\{\{).*$))")
def conv(match, line):
if not match:
return line
line = match.groups()[0]
if len(match.groups()[5]) == 0:
line += match.groups()[2]
line += match.groups()[3]
line += ", "
line += match.groups()[1]
line += "\n"
return line
for line in sys.stdin:
if line.find("getelementptr ") == line.find("getelementptr inbounds"):
if line.find("getelementptr inbounds") != line.find("getelementptr inbounds ("):
line = conv(re.match(ibrep, line), line)
elif line.find("getelementptr ") != line.find("getelementptr ("):
line = conv(re.match(normrep, line), line)
sys.stdout.write(line)
apply.sh:
for name in "$@"
do
python3 `dirname "$0"`/update.py < "$name" > "$name.tmp" && mv "$name.tmp" "$name"
rm -f "$name.tmp"
done
The actual commands:
From llvm/src:
find test/ -name *.ll | xargs ./apply.sh
From llvm/src/tools/clang:
find test/ -name *.mm -o -name *.m -o -name *.cpp -o -name *.c | xargs -I '{}' ../../apply.sh "{}"
From llvm/src/tools/polly:
find test/ -name *.ll | xargs ./apply.sh
After that, check-all (with llvm, clang, clang-tools-extra, lld,
compiler-rt, and polly all checked out).
The extra 'rm' in the apply.sh script is due to a few files in clang's test
suite using interesting unicode stuff that my python script was throwing
exceptions on. None of those files needed to be migrated, so it seemed
sufficient to ignore those cases.
Reviewers: rafael, dexonsmith, grosser
Differential Revision: http://reviews.llvm.org/D7636
llvm-svn: 230786
2015-02-28 03:29:02 +08:00
|
|
|
%out.gep = getelementptr inbounds float, float addrspace(1)* %out, i32 %x.i
|
2014-10-11 06:01:59 +08:00
|
|
|
store float %sum, float addrspace(1)* %out.gep, align 4
|
|
|
|
|
ret void
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
; Can't fold since not correctly aligned.
|
|
|
|
|
; XXX: This isn't really testing anything useful now. I think CI
|
|
|
|
|
; allows unaligned LDS accesses, which would be a problem here.
|
|
|
|
|
; SI-LABEL: @unaligned_read2_f32
|
2014-11-05 22:50:53 +08:00
|
|
|
; SI-NOT: ds_read2_b32
|
|
|
|
|
; SI: s_endpgm
|
2014-10-11 06:01:59 +08:00
|
|
|
define void @unaligned_read2_f32(float addrspace(1)* %out, float addrspace(3)* %lds) #0 {
|
|
|
|
|
%x.i = tail call i32 @llvm.r600.read.tidig.x() #1
|
[opaque pointer type] Add textual IR support for explicit type parameter to getelementptr instruction
One of several parallel first steps to remove the target type of pointers,
replacing them with a single opaque pointer type.
This adds an explicit type parameter to the gep instruction so that when the
first parameter becomes an opaque pointer type, the type to gep through is
still available to the instructions.
* This doesn't modify gep operators, only instructions (operators will be
handled separately)
* Textual IR changes only. Bitcode (including upgrade) and changing the
in-memory representation will be in separate changes.
* geps of vectors are transformed as:
getelementptr <4 x float*> %x, ...
->getelementptr float, <4 x float*> %x, ...
Then, once the opaque pointer type is introduced, this will ultimately look
like:
getelementptr float, <4 x ptr> %x
with the unambiguous interpretation that it is a vector of pointers to float.
* address spaces remain on the pointer, not the type:
getelementptr float addrspace(1)* %x
->getelementptr float, float addrspace(1)* %x
Then, eventually:
getelementptr float, ptr addrspace(1) %x
Importantly, the massive amount of test case churn has been automated by
same crappy python code. I had to manually update a few test cases that
wouldn't fit the script's model (r228970,r229196,r229197,r229198). The
python script just massages stdin and writes the result to stdout, I
then wrapped that in a shell script to handle replacing files, then
using the usual find+xargs to migrate all the files.
update.py:
import fileinput
import sys
import re
ibrep = re.compile(r"(^.*?[^%\w]getelementptr inbounds )(((?:<\d* x )?)(.*?)(| addrspace\(\d\)) *\*(|>)(?:$| *(?:%|@|null|undef|blockaddress|getelementptr|addrspacecast|bitcast|inttoptr|\[\[[a-zA-Z]|\{\{).*$))")
normrep = re.compile( r"(^.*?[^%\w]getelementptr )(((?:<\d* x )?)(.*?)(| addrspace\(\d\)) *\*(|>)(?:$| *(?:%|@|null|undef|blockaddress|getelementptr|addrspacecast|bitcast|inttoptr|\[\[[a-zA-Z]|\{\{).*$))")
def conv(match, line):
if not match:
return line
line = match.groups()[0]
if len(match.groups()[5]) == 0:
line += match.groups()[2]
line += match.groups()[3]
line += ", "
line += match.groups()[1]
line += "\n"
return line
for line in sys.stdin:
if line.find("getelementptr ") == line.find("getelementptr inbounds"):
if line.find("getelementptr inbounds") != line.find("getelementptr inbounds ("):
line = conv(re.match(ibrep, line), line)
elif line.find("getelementptr ") != line.find("getelementptr ("):
line = conv(re.match(normrep, line), line)
sys.stdout.write(line)
apply.sh:
for name in "$@"
do
python3 `dirname "$0"`/update.py < "$name" > "$name.tmp" && mv "$name.tmp" "$name"
rm -f "$name.tmp"
done
The actual commands:
From llvm/src:
find test/ -name *.ll | xargs ./apply.sh
From llvm/src/tools/clang:
find test/ -name *.mm -o -name *.m -o -name *.cpp -o -name *.c | xargs -I '{}' ../../apply.sh "{}"
From llvm/src/tools/polly:
find test/ -name *.ll | xargs ./apply.sh
After that, check-all (with llvm, clang, clang-tools-extra, lld,
compiler-rt, and polly all checked out).
The extra 'rm' in the apply.sh script is due to a few files in clang's test
suite using interesting unicode stuff that my python script was throwing
exceptions on. None of those files needed to be migrated, so it seemed
sufficient to ignore those cases.
Reviewers: rafael, dexonsmith, grosser
Differential Revision: http://reviews.llvm.org/D7636
llvm-svn: 230786
2015-02-28 03:29:02 +08:00
|
|
|
%arrayidx0 = getelementptr inbounds float, float addrspace(3)* %lds, i32 %x.i
|
2015-02-28 05:17:42 +08:00
|
|
|
%val0 = load float, float addrspace(3)* %arrayidx0, align 1
|
2014-10-11 06:01:59 +08:00
|
|
|
%add.x = add nsw i32 %x.i, 8
|
[opaque pointer type] Add textual IR support for explicit type parameter to getelementptr instruction
One of several parallel first steps to remove the target type of pointers,
replacing them with a single opaque pointer type.
This adds an explicit type parameter to the gep instruction so that when the
first parameter becomes an opaque pointer type, the type to gep through is
still available to the instructions.
* This doesn't modify gep operators, only instructions (operators will be
handled separately)
* Textual IR changes only. Bitcode (including upgrade) and changing the
in-memory representation will be in separate changes.
* geps of vectors are transformed as:
getelementptr <4 x float*> %x, ...
->getelementptr float, <4 x float*> %x, ...
Then, once the opaque pointer type is introduced, this will ultimately look
like:
getelementptr float, <4 x ptr> %x
with the unambiguous interpretation that it is a vector of pointers to float.
* address spaces remain on the pointer, not the type:
getelementptr float addrspace(1)* %x
->getelementptr float, float addrspace(1)* %x
Then, eventually:
getelementptr float, ptr addrspace(1) %x
Importantly, the massive amount of test case churn has been automated by
same crappy python code. I had to manually update a few test cases that
wouldn't fit the script's model (r228970,r229196,r229197,r229198). The
python script just massages stdin and writes the result to stdout, I
then wrapped that in a shell script to handle replacing files, then
using the usual find+xargs to migrate all the files.
update.py:
import fileinput
import sys
import re
ibrep = re.compile(r"(^.*?[^%\w]getelementptr inbounds )(((?:<\d* x )?)(.*?)(| addrspace\(\d\)) *\*(|>)(?:$| *(?:%|@|null|undef|blockaddress|getelementptr|addrspacecast|bitcast|inttoptr|\[\[[a-zA-Z]|\{\{).*$))")
normrep = re.compile( r"(^.*?[^%\w]getelementptr )(((?:<\d* x )?)(.*?)(| addrspace\(\d\)) *\*(|>)(?:$| *(?:%|@|null|undef|blockaddress|getelementptr|addrspacecast|bitcast|inttoptr|\[\[[a-zA-Z]|\{\{).*$))")
def conv(match, line):
if not match:
return line
line = match.groups()[0]
if len(match.groups()[5]) == 0:
line += match.groups()[2]
line += match.groups()[3]
line += ", "
line += match.groups()[1]
line += "\n"
return line
for line in sys.stdin:
if line.find("getelementptr ") == line.find("getelementptr inbounds"):
if line.find("getelementptr inbounds") != line.find("getelementptr inbounds ("):
line = conv(re.match(ibrep, line), line)
elif line.find("getelementptr ") != line.find("getelementptr ("):
line = conv(re.match(normrep, line), line)
sys.stdout.write(line)
apply.sh:
for name in "$@"
do
python3 `dirname "$0"`/update.py < "$name" > "$name.tmp" && mv "$name.tmp" "$name"
rm -f "$name.tmp"
done
The actual commands:
From llvm/src:
find test/ -name *.ll | xargs ./apply.sh
From llvm/src/tools/clang:
find test/ -name *.mm -o -name *.m -o -name *.cpp -o -name *.c | xargs -I '{}' ../../apply.sh "{}"
From llvm/src/tools/polly:
find test/ -name *.ll | xargs ./apply.sh
After that, check-all (with llvm, clang, clang-tools-extra, lld,
compiler-rt, and polly all checked out).
The extra 'rm' in the apply.sh script is due to a few files in clang's test
suite using interesting unicode stuff that my python script was throwing
exceptions on. None of those files needed to be migrated, so it seemed
sufficient to ignore those cases.
Reviewers: rafael, dexonsmith, grosser
Differential Revision: http://reviews.llvm.org/D7636
llvm-svn: 230786
2015-02-28 03:29:02 +08:00
|
|
|
%arrayidx1 = getelementptr inbounds float, float addrspace(3)* %lds, i32 %add.x
|
2015-02-28 05:17:42 +08:00
|
|
|
%val1 = load float, float addrspace(3)* %arrayidx1, align 1
|
2014-10-11 06:01:59 +08:00
|
|
|
%sum = fadd float %val0, %val1
|
[opaque pointer type] Add textual IR support for explicit type parameter to getelementptr instruction
One of several parallel first steps to remove the target type of pointers,
replacing them with a single opaque pointer type.
This adds an explicit type parameter to the gep instruction so that when the
first parameter becomes an opaque pointer type, the type to gep through is
still available to the instructions.
* This doesn't modify gep operators, only instructions (operators will be
handled separately)
* Textual IR changes only. Bitcode (including upgrade) and changing the
in-memory representation will be in separate changes.
* geps of vectors are transformed as:
getelementptr <4 x float*> %x, ...
->getelementptr float, <4 x float*> %x, ...
Then, once the opaque pointer type is introduced, this will ultimately look
like:
getelementptr float, <4 x ptr> %x
with the unambiguous interpretation that it is a vector of pointers to float.
* address spaces remain on the pointer, not the type:
getelementptr float addrspace(1)* %x
->getelementptr float, float addrspace(1)* %x
Then, eventually:
getelementptr float, ptr addrspace(1) %x
Importantly, the massive amount of test case churn has been automated by
same crappy python code. I had to manually update a few test cases that
wouldn't fit the script's model (r228970,r229196,r229197,r229198). The
python script just massages stdin and writes the result to stdout, I
then wrapped that in a shell script to handle replacing files, then
using the usual find+xargs to migrate all the files.
update.py:
import fileinput
import sys
import re
ibrep = re.compile(r"(^.*?[^%\w]getelementptr inbounds )(((?:<\d* x )?)(.*?)(| addrspace\(\d\)) *\*(|>)(?:$| *(?:%|@|null|undef|blockaddress|getelementptr|addrspacecast|bitcast|inttoptr|\[\[[a-zA-Z]|\{\{).*$))")
normrep = re.compile( r"(^.*?[^%\w]getelementptr )(((?:<\d* x )?)(.*?)(| addrspace\(\d\)) *\*(|>)(?:$| *(?:%|@|null|undef|blockaddress|getelementptr|addrspacecast|bitcast|inttoptr|\[\[[a-zA-Z]|\{\{).*$))")
def conv(match, line):
if not match:
return line
line = match.groups()[0]
if len(match.groups()[5]) == 0:
line += match.groups()[2]
line += match.groups()[3]
line += ", "
line += match.groups()[1]
line += "\n"
return line
for line in sys.stdin:
if line.find("getelementptr ") == line.find("getelementptr inbounds"):
if line.find("getelementptr inbounds") != line.find("getelementptr inbounds ("):
line = conv(re.match(ibrep, line), line)
elif line.find("getelementptr ") != line.find("getelementptr ("):
line = conv(re.match(normrep, line), line)
sys.stdout.write(line)
apply.sh:
for name in "$@"
do
python3 `dirname "$0"`/update.py < "$name" > "$name.tmp" && mv "$name.tmp" "$name"
rm -f "$name.tmp"
done
The actual commands:
From llvm/src:
find test/ -name *.ll | xargs ./apply.sh
From llvm/src/tools/clang:
find test/ -name *.mm -o -name *.m -o -name *.cpp -o -name *.c | xargs -I '{}' ../../apply.sh "{}"
From llvm/src/tools/polly:
find test/ -name *.ll | xargs ./apply.sh
After that, check-all (with llvm, clang, clang-tools-extra, lld,
compiler-rt, and polly all checked out).
The extra 'rm' in the apply.sh script is due to a few files in clang's test
suite using interesting unicode stuff that my python script was throwing
exceptions on. None of those files needed to be migrated, so it seemed
sufficient to ignore those cases.
Reviewers: rafael, dexonsmith, grosser
Differential Revision: http://reviews.llvm.org/D7636
llvm-svn: 230786
2015-02-28 03:29:02 +08:00
|
|
|
%out.gep = getelementptr inbounds float, float addrspace(1)* %out, i32 %x.i
|
2014-10-11 06:01:59 +08:00
|
|
|
store float %sum, float addrspace(1)* %out.gep, align 4
|
|
|
|
|
ret void
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
; SI-LABEL: @misaligned_2_simple_read2_f32
|
2014-11-05 22:50:53 +08:00
|
|
|
; SI-NOT: ds_read2_b32
|
|
|
|
|
; SI: s_endpgm
|
2014-10-11 06:01:59 +08:00
|
|
|
define void @misaligned_2_simple_read2_f32(float addrspace(1)* %out, float addrspace(3)* %lds) #0 {
|
|
|
|
|
%x.i = tail call i32 @llvm.r600.read.tidig.x() #1
|
[opaque pointer type] Add textual IR support for explicit type parameter to getelementptr instruction
One of several parallel first steps to remove the target type of pointers,
replacing them with a single opaque pointer type.
This adds an explicit type parameter to the gep instruction so that when the
first parameter becomes an opaque pointer type, the type to gep through is
still available to the instructions.
* This doesn't modify gep operators, only instructions (operators will be
handled separately)
* Textual IR changes only. Bitcode (including upgrade) and changing the
in-memory representation will be in separate changes.
* geps of vectors are transformed as:
getelementptr <4 x float*> %x, ...
->getelementptr float, <4 x float*> %x, ...
Then, once the opaque pointer type is introduced, this will ultimately look
like:
getelementptr float, <4 x ptr> %x
with the unambiguous interpretation that it is a vector of pointers to float.
* address spaces remain on the pointer, not the type:
getelementptr float addrspace(1)* %x
->getelementptr float, float addrspace(1)* %x
Then, eventually:
getelementptr float, ptr addrspace(1) %x
Importantly, the massive amount of test case churn has been automated by
same crappy python code. I had to manually update a few test cases that
wouldn't fit the script's model (r228970,r229196,r229197,r229198). The
python script just massages stdin and writes the result to stdout, I
then wrapped that in a shell script to handle replacing files, then
using the usual find+xargs to migrate all the files.
update.py:
import fileinput
import sys
import re
ibrep = re.compile(r"(^.*?[^%\w]getelementptr inbounds )(((?:<\d* x )?)(.*?)(| addrspace\(\d\)) *\*(|>)(?:$| *(?:%|@|null|undef|blockaddress|getelementptr|addrspacecast|bitcast|inttoptr|\[\[[a-zA-Z]|\{\{).*$))")
normrep = re.compile( r"(^.*?[^%\w]getelementptr )(((?:<\d* x )?)(.*?)(| addrspace\(\d\)) *\*(|>)(?:$| *(?:%|@|null|undef|blockaddress|getelementptr|addrspacecast|bitcast|inttoptr|\[\[[a-zA-Z]|\{\{).*$))")
def conv(match, line):
if not match:
return line
line = match.groups()[0]
if len(match.groups()[5]) == 0:
line += match.groups()[2]
line += match.groups()[3]
line += ", "
line += match.groups()[1]
line += "\n"
return line
for line in sys.stdin:
if line.find("getelementptr ") == line.find("getelementptr inbounds"):
if line.find("getelementptr inbounds") != line.find("getelementptr inbounds ("):
line = conv(re.match(ibrep, line), line)
elif line.find("getelementptr ") != line.find("getelementptr ("):
line = conv(re.match(normrep, line), line)
sys.stdout.write(line)
apply.sh:
for name in "$@"
do
python3 `dirname "$0"`/update.py < "$name" > "$name.tmp" && mv "$name.tmp" "$name"
rm -f "$name.tmp"
done
The actual commands:
From llvm/src:
find test/ -name *.ll | xargs ./apply.sh
From llvm/src/tools/clang:
find test/ -name *.mm -o -name *.m -o -name *.cpp -o -name *.c | xargs -I '{}' ../../apply.sh "{}"
From llvm/src/tools/polly:
find test/ -name *.ll | xargs ./apply.sh
After that, check-all (with llvm, clang, clang-tools-extra, lld,
compiler-rt, and polly all checked out).
The extra 'rm' in the apply.sh script is due to a few files in clang's test
suite using interesting unicode stuff that my python script was throwing
exceptions on. None of those files needed to be migrated, so it seemed
sufficient to ignore those cases.
Reviewers: rafael, dexonsmith, grosser
Differential Revision: http://reviews.llvm.org/D7636
llvm-svn: 230786
2015-02-28 03:29:02 +08:00
|
|
|
%arrayidx0 = getelementptr inbounds float, float addrspace(3)* %lds, i32 %x.i
|
2015-02-28 05:17:42 +08:00
|
|
|
%val0 = load float, float addrspace(3)* %arrayidx0, align 2
|
2014-10-11 06:01:59 +08:00
|
|
|
%add.x = add nsw i32 %x.i, 8
|
[opaque pointer type] Add textual IR support for explicit type parameter to getelementptr instruction
One of several parallel first steps to remove the target type of pointers,
replacing them with a single opaque pointer type.
This adds an explicit type parameter to the gep instruction so that when the
first parameter becomes an opaque pointer type, the type to gep through is
still available to the instructions.
* This doesn't modify gep operators, only instructions (operators will be
handled separately)
* Textual IR changes only. Bitcode (including upgrade) and changing the
in-memory representation will be in separate changes.
* geps of vectors are transformed as:
getelementptr <4 x float*> %x, ...
->getelementptr float, <4 x float*> %x, ...
Then, once the opaque pointer type is introduced, this will ultimately look
like:
getelementptr float, <4 x ptr> %x
with the unambiguous interpretation that it is a vector of pointers to float.
* address spaces remain on the pointer, not the type:
getelementptr float addrspace(1)* %x
->getelementptr float, float addrspace(1)* %x
Then, eventually:
getelementptr float, ptr addrspace(1) %x
Importantly, the massive amount of test case churn has been automated by
same crappy python code. I had to manually update a few test cases that
wouldn't fit the script's model (r228970,r229196,r229197,r229198). The
python script just massages stdin and writes the result to stdout, I
then wrapped that in a shell script to handle replacing files, then
using the usual find+xargs to migrate all the files.
update.py:
import fileinput
import sys
import re
ibrep = re.compile(r"(^.*?[^%\w]getelementptr inbounds )(((?:<\d* x )?)(.*?)(| addrspace\(\d\)) *\*(|>)(?:$| *(?:%|@|null|undef|blockaddress|getelementptr|addrspacecast|bitcast|inttoptr|\[\[[a-zA-Z]|\{\{).*$))")
normrep = re.compile( r"(^.*?[^%\w]getelementptr )(((?:<\d* x )?)(.*?)(| addrspace\(\d\)) *\*(|>)(?:$| *(?:%|@|null|undef|blockaddress|getelementptr|addrspacecast|bitcast|inttoptr|\[\[[a-zA-Z]|\{\{).*$))")
def conv(match, line):
if not match:
return line
line = match.groups()[0]
if len(match.groups()[5]) == 0:
line += match.groups()[2]
line += match.groups()[3]
line += ", "
line += match.groups()[1]
line += "\n"
return line
for line in sys.stdin:
if line.find("getelementptr ") == line.find("getelementptr inbounds"):
if line.find("getelementptr inbounds") != line.find("getelementptr inbounds ("):
line = conv(re.match(ibrep, line), line)
elif line.find("getelementptr ") != line.find("getelementptr ("):
line = conv(re.match(normrep, line), line)
sys.stdout.write(line)
apply.sh:
for name in "$@"
do
python3 `dirname "$0"`/update.py < "$name" > "$name.tmp" && mv "$name.tmp" "$name"
rm -f "$name.tmp"
done
The actual commands:
From llvm/src:
find test/ -name *.ll | xargs ./apply.sh
From llvm/src/tools/clang:
find test/ -name *.mm -o -name *.m -o -name *.cpp -o -name *.c | xargs -I '{}' ../../apply.sh "{}"
From llvm/src/tools/polly:
find test/ -name *.ll | xargs ./apply.sh
After that, check-all (with llvm, clang, clang-tools-extra, lld,
compiler-rt, and polly all checked out).
The extra 'rm' in the apply.sh script is due to a few files in clang's test
suite using interesting unicode stuff that my python script was throwing
exceptions on. None of those files needed to be migrated, so it seemed
sufficient to ignore those cases.
Reviewers: rafael, dexonsmith, grosser
Differential Revision: http://reviews.llvm.org/D7636
llvm-svn: 230786
2015-02-28 03:29:02 +08:00
|
|
|
%arrayidx1 = getelementptr inbounds float, float addrspace(3)* %lds, i32 %add.x
|
2015-02-28 05:17:42 +08:00
|
|
|
%val1 = load float, float addrspace(3)* %arrayidx1, align 2
|
2014-10-11 06:01:59 +08:00
|
|
|
%sum = fadd float %val0, %val1
|
[opaque pointer type] Add textual IR support for explicit type parameter to getelementptr instruction
One of several parallel first steps to remove the target type of pointers,
replacing them with a single opaque pointer type.
This adds an explicit type parameter to the gep instruction so that when the
first parameter becomes an opaque pointer type, the type to gep through is
still available to the instructions.
* This doesn't modify gep operators, only instructions (operators will be
handled separately)
* Textual IR changes only. Bitcode (including upgrade) and changing the
in-memory representation will be in separate changes.
* geps of vectors are transformed as:
getelementptr <4 x float*> %x, ...
->getelementptr float, <4 x float*> %x, ...
Then, once the opaque pointer type is introduced, this will ultimately look
like:
getelementptr float, <4 x ptr> %x
with the unambiguous interpretation that it is a vector of pointers to float.
* address spaces remain on the pointer, not the type:
getelementptr float addrspace(1)* %x
->getelementptr float, float addrspace(1)* %x
Then, eventually:
getelementptr float, ptr addrspace(1) %x
Importantly, the massive amount of test case churn has been automated by
same crappy python code. I had to manually update a few test cases that
wouldn't fit the script's model (r228970,r229196,r229197,r229198). The
python script just massages stdin and writes the result to stdout, I
then wrapped that in a shell script to handle replacing files, then
using the usual find+xargs to migrate all the files.
update.py:
import fileinput
import sys
import re
ibrep = re.compile(r"(^.*?[^%\w]getelementptr inbounds )(((?:<\d* x )?)(.*?)(| addrspace\(\d\)) *\*(|>)(?:$| *(?:%|@|null|undef|blockaddress|getelementptr|addrspacecast|bitcast|inttoptr|\[\[[a-zA-Z]|\{\{).*$))")
normrep = re.compile( r"(^.*?[^%\w]getelementptr )(((?:<\d* x )?)(.*?)(| addrspace\(\d\)) *\*(|>)(?:$| *(?:%|@|null|undef|blockaddress|getelementptr|addrspacecast|bitcast|inttoptr|\[\[[a-zA-Z]|\{\{).*$))")
def conv(match, line):
if not match:
return line
line = match.groups()[0]
if len(match.groups()[5]) == 0:
line += match.groups()[2]
line += match.groups()[3]
line += ", "
line += match.groups()[1]
line += "\n"
return line
for line in sys.stdin:
if line.find("getelementptr ") == line.find("getelementptr inbounds"):
if line.find("getelementptr inbounds") != line.find("getelementptr inbounds ("):
line = conv(re.match(ibrep, line), line)
elif line.find("getelementptr ") != line.find("getelementptr ("):
line = conv(re.match(normrep, line), line)
sys.stdout.write(line)
apply.sh:
for name in "$@"
do
python3 `dirname "$0"`/update.py < "$name" > "$name.tmp" && mv "$name.tmp" "$name"
rm -f "$name.tmp"
done
The actual commands:
From llvm/src:
find test/ -name *.ll | xargs ./apply.sh
From llvm/src/tools/clang:
find test/ -name *.mm -o -name *.m -o -name *.cpp -o -name *.c | xargs -I '{}' ../../apply.sh "{}"
From llvm/src/tools/polly:
find test/ -name *.ll | xargs ./apply.sh
After that, check-all (with llvm, clang, clang-tools-extra, lld,
compiler-rt, and polly all checked out).
The extra 'rm' in the apply.sh script is due to a few files in clang's test
suite using interesting unicode stuff that my python script was throwing
exceptions on. None of those files needed to be migrated, so it seemed
sufficient to ignore those cases.
Reviewers: rafael, dexonsmith, grosser
Differential Revision: http://reviews.llvm.org/D7636
llvm-svn: 230786
2015-02-28 03:29:02 +08:00
|
|
|
%out.gep = getelementptr inbounds float, float addrspace(1)* %out, i32 %x.i
|
2014-10-11 06:01:59 +08:00
|
|
|
store float %sum, float addrspace(1)* %out.gep, align 4
|
|
|
|
|
ret void
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
; SI-LABEL: @simple_read2_f64
|
2014-11-05 22:50:53 +08:00
|
|
|
; SI: v_lshlrev_b32_e32 [[VPTR:v[0-9]+]], 3, {{v[0-9]+}}
|
2015-04-08 09:09:19 +08:00
|
|
|
; SI: ds_read2_b64 v{{\[}}[[LO_VREG:[0-9]+]]:[[HI_VREG:[0-9]+]]{{\]}}, [[VPTR]] offset1:8
|
2014-11-05 22:50:53 +08:00
|
|
|
; SI: v_add_f64 [[RESULT:v\[[0-9]+:[0-9]+\]]], v{{\[}}[[LO_VREG]]:{{[0-9]+\]}}, v{{\[[0-9]+}}:[[HI_VREG]]{{\]}}
|
|
|
|
|
; SI: buffer_store_dwordx2 [[RESULT]]
|
|
|
|
|
; SI: s_endpgm
|
2014-10-11 06:01:59 +08:00
|
|
|
define void @simple_read2_f64(double addrspace(1)* %out) #0 {
|
|
|
|
|
%x.i = tail call i32 @llvm.r600.read.tidig.x() #1
|
[opaque pointer type] Add textual IR support for explicit type parameter to getelementptr instruction
One of several parallel first steps to remove the target type of pointers,
replacing them with a single opaque pointer type.
This adds an explicit type parameter to the gep instruction so that when the
first parameter becomes an opaque pointer type, the type to gep through is
still available to the instructions.
* This doesn't modify gep operators, only instructions (operators will be
handled separately)
* Textual IR changes only. Bitcode (including upgrade) and changing the
in-memory representation will be in separate changes.
* geps of vectors are transformed as:
getelementptr <4 x float*> %x, ...
->getelementptr float, <4 x float*> %x, ...
Then, once the opaque pointer type is introduced, this will ultimately look
like:
getelementptr float, <4 x ptr> %x
with the unambiguous interpretation that it is a vector of pointers to float.
* address spaces remain on the pointer, not the type:
getelementptr float addrspace(1)* %x
->getelementptr float, float addrspace(1)* %x
Then, eventually:
getelementptr float, ptr addrspace(1) %x
Importantly, the massive amount of test case churn has been automated by
same crappy python code. I had to manually update a few test cases that
wouldn't fit the script's model (r228970,r229196,r229197,r229198). The
python script just massages stdin and writes the result to stdout, I
then wrapped that in a shell script to handle replacing files, then
using the usual find+xargs to migrate all the files.
update.py:
import fileinput
import sys
import re
ibrep = re.compile(r"(^.*?[^%\w]getelementptr inbounds )(((?:<\d* x )?)(.*?)(| addrspace\(\d\)) *\*(|>)(?:$| *(?:%|@|null|undef|blockaddress|getelementptr|addrspacecast|bitcast|inttoptr|\[\[[a-zA-Z]|\{\{).*$))")
normrep = re.compile( r"(^.*?[^%\w]getelementptr )(((?:<\d* x )?)(.*?)(| addrspace\(\d\)) *\*(|>)(?:$| *(?:%|@|null|undef|blockaddress|getelementptr|addrspacecast|bitcast|inttoptr|\[\[[a-zA-Z]|\{\{).*$))")
def conv(match, line):
if not match:
return line
line = match.groups()[0]
if len(match.groups()[5]) == 0:
line += match.groups()[2]
line += match.groups()[3]
line += ", "
line += match.groups()[1]
line += "\n"
return line
for line in sys.stdin:
if line.find("getelementptr ") == line.find("getelementptr inbounds"):
if line.find("getelementptr inbounds") != line.find("getelementptr inbounds ("):
line = conv(re.match(ibrep, line), line)
elif line.find("getelementptr ") != line.find("getelementptr ("):
line = conv(re.match(normrep, line), line)
sys.stdout.write(line)
apply.sh:
for name in "$@"
do
python3 `dirname "$0"`/update.py < "$name" > "$name.tmp" && mv "$name.tmp" "$name"
rm -f "$name.tmp"
done
The actual commands:
From llvm/src:
find test/ -name *.ll | xargs ./apply.sh
From llvm/src/tools/clang:
find test/ -name *.mm -o -name *.m -o -name *.cpp -o -name *.c | xargs -I '{}' ../../apply.sh "{}"
From llvm/src/tools/polly:
find test/ -name *.ll | xargs ./apply.sh
After that, check-all (with llvm, clang, clang-tools-extra, lld,
compiler-rt, and polly all checked out).
The extra 'rm' in the apply.sh script is due to a few files in clang's test
suite using interesting unicode stuff that my python script was throwing
exceptions on. None of those files needed to be migrated, so it seemed
sufficient to ignore those cases.
Reviewers: rafael, dexonsmith, grosser
Differential Revision: http://reviews.llvm.org/D7636
llvm-svn: 230786
2015-02-28 03:29:02 +08:00
|
|
|
%arrayidx0 = getelementptr inbounds [512 x double], [512 x double] addrspace(3)* @lds.f64, i32 0, i32 %x.i
|
2015-02-28 05:17:42 +08:00
|
|
|
%val0 = load double, double addrspace(3)* %arrayidx0, align 8
|
2014-10-11 06:01:59 +08:00
|
|
|
%add.x = add nsw i32 %x.i, 8
|
[opaque pointer type] Add textual IR support for explicit type parameter to getelementptr instruction
One of several parallel first steps to remove the target type of pointers,
replacing them with a single opaque pointer type.
This adds an explicit type parameter to the gep instruction so that when the
first parameter becomes an opaque pointer type, the type to gep through is
still available to the instructions.
* This doesn't modify gep operators, only instructions (operators will be
handled separately)
* Textual IR changes only. Bitcode (including upgrade) and changing the
in-memory representation will be in separate changes.
* geps of vectors are transformed as:
getelementptr <4 x float*> %x, ...
->getelementptr float, <4 x float*> %x, ...
Then, once the opaque pointer type is introduced, this will ultimately look
like:
getelementptr float, <4 x ptr> %x
with the unambiguous interpretation that it is a vector of pointers to float.
* address spaces remain on the pointer, not the type:
getelementptr float addrspace(1)* %x
->getelementptr float, float addrspace(1)* %x
Then, eventually:
getelementptr float, ptr addrspace(1) %x
Importantly, the massive amount of test case churn has been automated by
same crappy python code. I had to manually update a few test cases that
wouldn't fit the script's model (r228970,r229196,r229197,r229198). The
python script just massages stdin and writes the result to stdout, I
then wrapped that in a shell script to handle replacing files, then
using the usual find+xargs to migrate all the files.
update.py:
import fileinput
import sys
import re
ibrep = re.compile(r"(^.*?[^%\w]getelementptr inbounds )(((?:<\d* x )?)(.*?)(| addrspace\(\d\)) *\*(|>)(?:$| *(?:%|@|null|undef|blockaddress|getelementptr|addrspacecast|bitcast|inttoptr|\[\[[a-zA-Z]|\{\{).*$))")
normrep = re.compile( r"(^.*?[^%\w]getelementptr )(((?:<\d* x )?)(.*?)(| addrspace\(\d\)) *\*(|>)(?:$| *(?:%|@|null|undef|blockaddress|getelementptr|addrspacecast|bitcast|inttoptr|\[\[[a-zA-Z]|\{\{).*$))")
def conv(match, line):
if not match:
return line
line = match.groups()[0]
if len(match.groups()[5]) == 0:
line += match.groups()[2]
line += match.groups()[3]
line += ", "
line += match.groups()[1]
line += "\n"
return line
for line in sys.stdin:
if line.find("getelementptr ") == line.find("getelementptr inbounds"):
if line.find("getelementptr inbounds") != line.find("getelementptr inbounds ("):
line = conv(re.match(ibrep, line), line)
elif line.find("getelementptr ") != line.find("getelementptr ("):
line = conv(re.match(normrep, line), line)
sys.stdout.write(line)
apply.sh:
for name in "$@"
do
python3 `dirname "$0"`/update.py < "$name" > "$name.tmp" && mv "$name.tmp" "$name"
rm -f "$name.tmp"
done
The actual commands:
From llvm/src:
find test/ -name *.ll | xargs ./apply.sh
From llvm/src/tools/clang:
find test/ -name *.mm -o -name *.m -o -name *.cpp -o -name *.c | xargs -I '{}' ../../apply.sh "{}"
From llvm/src/tools/polly:
find test/ -name *.ll | xargs ./apply.sh
After that, check-all (with llvm, clang, clang-tools-extra, lld,
compiler-rt, and polly all checked out).
The extra 'rm' in the apply.sh script is due to a few files in clang's test
suite using interesting unicode stuff that my python script was throwing
exceptions on. None of those files needed to be migrated, so it seemed
sufficient to ignore those cases.
Reviewers: rafael, dexonsmith, grosser
Differential Revision: http://reviews.llvm.org/D7636
llvm-svn: 230786
2015-02-28 03:29:02 +08:00
|
|
|
%arrayidx1 = getelementptr inbounds [512 x double], [512 x double] addrspace(3)* @lds.f64, i32 0, i32 %add.x
|
2015-02-28 05:17:42 +08:00
|
|
|
%val1 = load double, double addrspace(3)* %arrayidx1, align 8
|
2014-10-11 06:01:59 +08:00
|
|
|
%sum = fadd double %val0, %val1
|
[opaque pointer type] Add textual IR support for explicit type parameter to getelementptr instruction
One of several parallel first steps to remove the target type of pointers,
replacing them with a single opaque pointer type.
This adds an explicit type parameter to the gep instruction so that when the
first parameter becomes an opaque pointer type, the type to gep through is
still available to the instructions.
* This doesn't modify gep operators, only instructions (operators will be
handled separately)
* Textual IR changes only. Bitcode (including upgrade) and changing the
in-memory representation will be in separate changes.
* geps of vectors are transformed as:
getelementptr <4 x float*> %x, ...
->getelementptr float, <4 x float*> %x, ...
Then, once the opaque pointer type is introduced, this will ultimately look
like:
getelementptr float, <4 x ptr> %x
with the unambiguous interpretation that it is a vector of pointers to float.
* address spaces remain on the pointer, not the type:
getelementptr float addrspace(1)* %x
->getelementptr float, float addrspace(1)* %x
Then, eventually:
getelementptr float, ptr addrspace(1) %x
Importantly, the massive amount of test case churn has been automated by
same crappy python code. I had to manually update a few test cases that
wouldn't fit the script's model (r228970,r229196,r229197,r229198). The
python script just massages stdin and writes the result to stdout, I
then wrapped that in a shell script to handle replacing files, then
using the usual find+xargs to migrate all the files.
update.py:
import fileinput
import sys
import re
ibrep = re.compile(r"(^.*?[^%\w]getelementptr inbounds )(((?:<\d* x )?)(.*?)(| addrspace\(\d\)) *\*(|>)(?:$| *(?:%|@|null|undef|blockaddress|getelementptr|addrspacecast|bitcast|inttoptr|\[\[[a-zA-Z]|\{\{).*$))")
normrep = re.compile( r"(^.*?[^%\w]getelementptr )(((?:<\d* x )?)(.*?)(| addrspace\(\d\)) *\*(|>)(?:$| *(?:%|@|null|undef|blockaddress|getelementptr|addrspacecast|bitcast|inttoptr|\[\[[a-zA-Z]|\{\{).*$))")
def conv(match, line):
if not match:
return line
line = match.groups()[0]
if len(match.groups()[5]) == 0:
line += match.groups()[2]
line += match.groups()[3]
line += ", "
line += match.groups()[1]
line += "\n"
return line
for line in sys.stdin:
if line.find("getelementptr ") == line.find("getelementptr inbounds"):
if line.find("getelementptr inbounds") != line.find("getelementptr inbounds ("):
line = conv(re.match(ibrep, line), line)
elif line.find("getelementptr ") != line.find("getelementptr ("):
line = conv(re.match(normrep, line), line)
sys.stdout.write(line)
apply.sh:
for name in "$@"
do
python3 `dirname "$0"`/update.py < "$name" > "$name.tmp" && mv "$name.tmp" "$name"
rm -f "$name.tmp"
done
The actual commands:
From llvm/src:
find test/ -name *.ll | xargs ./apply.sh
From llvm/src/tools/clang:
find test/ -name *.mm -o -name *.m -o -name *.cpp -o -name *.c | xargs -I '{}' ../../apply.sh "{}"
From llvm/src/tools/polly:
find test/ -name *.ll | xargs ./apply.sh
After that, check-all (with llvm, clang, clang-tools-extra, lld,
compiler-rt, and polly all checked out).
The extra 'rm' in the apply.sh script is due to a few files in clang's test
suite using interesting unicode stuff that my python script was throwing
exceptions on. None of those files needed to be migrated, so it seemed
sufficient to ignore those cases.
Reviewers: rafael, dexonsmith, grosser
Differential Revision: http://reviews.llvm.org/D7636
llvm-svn: 230786
2015-02-28 03:29:02 +08:00
|
|
|
%out.gep = getelementptr inbounds double, double addrspace(1)* %out, i32 %x.i
|
2014-10-11 06:01:59 +08:00
|
|
|
store double %sum, double addrspace(1)* %out.gep, align 8
|
|
|
|
|
ret void
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
; SI-LABEL: @simple_read2_f64_max_offset
|
2015-04-08 09:09:19 +08:00
|
|
|
; SI: ds_read2_b64 {{v\[[0-9]+:[0-9]+\]}}, v{{[0-9]+}} offset1:255
|
2014-11-05 22:50:53 +08:00
|
|
|
; SI: s_endpgm
|
2014-10-11 06:01:59 +08:00
|
|
|
define void @simple_read2_f64_max_offset(double addrspace(1)* %out) #0 {
|
|
|
|
|
%x.i = tail call i32 @llvm.r600.read.tidig.x() #1
|
[opaque pointer type] Add textual IR support for explicit type parameter to getelementptr instruction
One of several parallel first steps to remove the target type of pointers,
replacing them with a single opaque pointer type.
This adds an explicit type parameter to the gep instruction so that when the
first parameter becomes an opaque pointer type, the type to gep through is
still available to the instructions.
* This doesn't modify gep operators, only instructions (operators will be
handled separately)
* Textual IR changes only. Bitcode (including upgrade) and changing the
in-memory representation will be in separate changes.
* geps of vectors are transformed as:
getelementptr <4 x float*> %x, ...
->getelementptr float, <4 x float*> %x, ...
Then, once the opaque pointer type is introduced, this will ultimately look
like:
getelementptr float, <4 x ptr> %x
with the unambiguous interpretation that it is a vector of pointers to float.
* address spaces remain on the pointer, not the type:
getelementptr float addrspace(1)* %x
->getelementptr float, float addrspace(1)* %x
Then, eventually:
getelementptr float, ptr addrspace(1) %x
Importantly, the massive amount of test case churn has been automated by
same crappy python code. I had to manually update a few test cases that
wouldn't fit the script's model (r228970,r229196,r229197,r229198). The
python script just massages stdin and writes the result to stdout, I
then wrapped that in a shell script to handle replacing files, then
using the usual find+xargs to migrate all the files.
update.py:
import fileinput
import sys
import re
ibrep = re.compile(r"(^.*?[^%\w]getelementptr inbounds )(((?:<\d* x )?)(.*?)(| addrspace\(\d\)) *\*(|>)(?:$| *(?:%|@|null|undef|blockaddress|getelementptr|addrspacecast|bitcast|inttoptr|\[\[[a-zA-Z]|\{\{).*$))")
normrep = re.compile( r"(^.*?[^%\w]getelementptr )(((?:<\d* x )?)(.*?)(| addrspace\(\d\)) *\*(|>)(?:$| *(?:%|@|null|undef|blockaddress|getelementptr|addrspacecast|bitcast|inttoptr|\[\[[a-zA-Z]|\{\{).*$))")
def conv(match, line):
if not match:
return line
line = match.groups()[0]
if len(match.groups()[5]) == 0:
line += match.groups()[2]
line += match.groups()[3]
line += ", "
line += match.groups()[1]
line += "\n"
return line
for line in sys.stdin:
if line.find("getelementptr ") == line.find("getelementptr inbounds"):
if line.find("getelementptr inbounds") != line.find("getelementptr inbounds ("):
line = conv(re.match(ibrep, line), line)
elif line.find("getelementptr ") != line.find("getelementptr ("):
line = conv(re.match(normrep, line), line)
sys.stdout.write(line)
apply.sh:
for name in "$@"
do
python3 `dirname "$0"`/update.py < "$name" > "$name.tmp" && mv "$name.tmp" "$name"
rm -f "$name.tmp"
done
The actual commands:
From llvm/src:
find test/ -name *.ll | xargs ./apply.sh
From llvm/src/tools/clang:
find test/ -name *.mm -o -name *.m -o -name *.cpp -o -name *.c | xargs -I '{}' ../../apply.sh "{}"
From llvm/src/tools/polly:
find test/ -name *.ll | xargs ./apply.sh
After that, check-all (with llvm, clang, clang-tools-extra, lld,
compiler-rt, and polly all checked out).
The extra 'rm' in the apply.sh script is due to a few files in clang's test
suite using interesting unicode stuff that my python script was throwing
exceptions on. None of those files needed to be migrated, so it seemed
sufficient to ignore those cases.
Reviewers: rafael, dexonsmith, grosser
Differential Revision: http://reviews.llvm.org/D7636
llvm-svn: 230786
2015-02-28 03:29:02 +08:00
|
|
|
%arrayidx0 = getelementptr inbounds [512 x double], [512 x double] addrspace(3)* @lds.f64, i32 0, i32 %x.i
|
2015-02-28 05:17:42 +08:00
|
|
|
%val0 = load double, double addrspace(3)* %arrayidx0, align 8
|
2014-10-11 06:01:59 +08:00
|
|
|
%add.x = add nsw i32 %x.i, 255
|
[opaque pointer type] Add textual IR support for explicit type parameter to getelementptr instruction
One of several parallel first steps to remove the target type of pointers,
replacing them with a single opaque pointer type.
This adds an explicit type parameter to the gep instruction so that when the
first parameter becomes an opaque pointer type, the type to gep through is
still available to the instructions.
* This doesn't modify gep operators, only instructions (operators will be
handled separately)
* Textual IR changes only. Bitcode (including upgrade) and changing the
in-memory representation will be in separate changes.
* geps of vectors are transformed as:
getelementptr <4 x float*> %x, ...
->getelementptr float, <4 x float*> %x, ...
Then, once the opaque pointer type is introduced, this will ultimately look
like:
getelementptr float, <4 x ptr> %x
with the unambiguous interpretation that it is a vector of pointers to float.
* address spaces remain on the pointer, not the type:
getelementptr float addrspace(1)* %x
->getelementptr float, float addrspace(1)* %x
Then, eventually:
getelementptr float, ptr addrspace(1) %x
Importantly, the massive amount of test case churn has been automated by
same crappy python code. I had to manually update a few test cases that
wouldn't fit the script's model (r228970,r229196,r229197,r229198). The
python script just massages stdin and writes the result to stdout, I
then wrapped that in a shell script to handle replacing files, then
using the usual find+xargs to migrate all the files.
update.py:
import fileinput
import sys
import re
ibrep = re.compile(r"(^.*?[^%\w]getelementptr inbounds )(((?:<\d* x )?)(.*?)(| addrspace\(\d\)) *\*(|>)(?:$| *(?:%|@|null|undef|blockaddress|getelementptr|addrspacecast|bitcast|inttoptr|\[\[[a-zA-Z]|\{\{).*$))")
normrep = re.compile( r"(^.*?[^%\w]getelementptr )(((?:<\d* x )?)(.*?)(| addrspace\(\d\)) *\*(|>)(?:$| *(?:%|@|null|undef|blockaddress|getelementptr|addrspacecast|bitcast|inttoptr|\[\[[a-zA-Z]|\{\{).*$))")
def conv(match, line):
if not match:
return line
line = match.groups()[0]
if len(match.groups()[5]) == 0:
line += match.groups()[2]
line += match.groups()[3]
line += ", "
line += match.groups()[1]
line += "\n"
return line
for line in sys.stdin:
if line.find("getelementptr ") == line.find("getelementptr inbounds"):
if line.find("getelementptr inbounds") != line.find("getelementptr inbounds ("):
line = conv(re.match(ibrep, line), line)
elif line.find("getelementptr ") != line.find("getelementptr ("):
line = conv(re.match(normrep, line), line)
sys.stdout.write(line)
apply.sh:
for name in "$@"
do
python3 `dirname "$0"`/update.py < "$name" > "$name.tmp" && mv "$name.tmp" "$name"
rm -f "$name.tmp"
done
The actual commands:
From llvm/src:
find test/ -name *.ll | xargs ./apply.sh
From llvm/src/tools/clang:
find test/ -name *.mm -o -name *.m -o -name *.cpp -o -name *.c | xargs -I '{}' ../../apply.sh "{}"
From llvm/src/tools/polly:
find test/ -name *.ll | xargs ./apply.sh
After that, check-all (with llvm, clang, clang-tools-extra, lld,
compiler-rt, and polly all checked out).
The extra 'rm' in the apply.sh script is due to a few files in clang's test
suite using interesting unicode stuff that my python script was throwing
exceptions on. None of those files needed to be migrated, so it seemed
sufficient to ignore those cases.
Reviewers: rafael, dexonsmith, grosser
Differential Revision: http://reviews.llvm.org/D7636
llvm-svn: 230786
2015-02-28 03:29:02 +08:00
|
|
|
%arrayidx1 = getelementptr inbounds [512 x double], [512 x double] addrspace(3)* @lds.f64, i32 0, i32 %add.x
|
2015-02-28 05:17:42 +08:00
|
|
|
%val1 = load double, double addrspace(3)* %arrayidx1, align 8
|
2014-10-11 06:01:59 +08:00
|
|
|
%sum = fadd double %val0, %val1
|
[opaque pointer type] Add textual IR support for explicit type parameter to getelementptr instruction
One of several parallel first steps to remove the target type of pointers,
replacing them with a single opaque pointer type.
This adds an explicit type parameter to the gep instruction so that when the
first parameter becomes an opaque pointer type, the type to gep through is
still available to the instructions.
* This doesn't modify gep operators, only instructions (operators will be
handled separately)
* Textual IR changes only. Bitcode (including upgrade) and changing the
in-memory representation will be in separate changes.
* geps of vectors are transformed as:
getelementptr <4 x float*> %x, ...
->getelementptr float, <4 x float*> %x, ...
Then, once the opaque pointer type is introduced, this will ultimately look
like:
getelementptr float, <4 x ptr> %x
with the unambiguous interpretation that it is a vector of pointers to float.
* address spaces remain on the pointer, not the type:
getelementptr float addrspace(1)* %x
->getelementptr float, float addrspace(1)* %x
Then, eventually:
getelementptr float, ptr addrspace(1) %x
Importantly, the massive amount of test case churn has been automated by
same crappy python code. I had to manually update a few test cases that
wouldn't fit the script's model (r228970,r229196,r229197,r229198). The
python script just massages stdin and writes the result to stdout, I
then wrapped that in a shell script to handle replacing files, then
using the usual find+xargs to migrate all the files.
update.py:
import fileinput
import sys
import re
ibrep = re.compile(r"(^.*?[^%\w]getelementptr inbounds )(((?:<\d* x )?)(.*?)(| addrspace\(\d\)) *\*(|>)(?:$| *(?:%|@|null|undef|blockaddress|getelementptr|addrspacecast|bitcast|inttoptr|\[\[[a-zA-Z]|\{\{).*$))")
normrep = re.compile( r"(^.*?[^%\w]getelementptr )(((?:<\d* x )?)(.*?)(| addrspace\(\d\)) *\*(|>)(?:$| *(?:%|@|null|undef|blockaddress|getelementptr|addrspacecast|bitcast|inttoptr|\[\[[a-zA-Z]|\{\{).*$))")
def conv(match, line):
if not match:
return line
line = match.groups()[0]
if len(match.groups()[5]) == 0:
line += match.groups()[2]
line += match.groups()[3]
line += ", "
line += match.groups()[1]
line += "\n"
return line
for line in sys.stdin:
if line.find("getelementptr ") == line.find("getelementptr inbounds"):
if line.find("getelementptr inbounds") != line.find("getelementptr inbounds ("):
line = conv(re.match(ibrep, line), line)
elif line.find("getelementptr ") != line.find("getelementptr ("):
line = conv(re.match(normrep, line), line)
sys.stdout.write(line)
apply.sh:
for name in "$@"
do
python3 `dirname "$0"`/update.py < "$name" > "$name.tmp" && mv "$name.tmp" "$name"
rm -f "$name.tmp"
done
The actual commands:
From llvm/src:
find test/ -name *.ll | xargs ./apply.sh
From llvm/src/tools/clang:
find test/ -name *.mm -o -name *.m -o -name *.cpp -o -name *.c | xargs -I '{}' ../../apply.sh "{}"
From llvm/src/tools/polly:
find test/ -name *.ll | xargs ./apply.sh
After that, check-all (with llvm, clang, clang-tools-extra, lld,
compiler-rt, and polly all checked out).
The extra 'rm' in the apply.sh script is due to a few files in clang's test
suite using interesting unicode stuff that my python script was throwing
exceptions on. None of those files needed to be migrated, so it seemed
sufficient to ignore those cases.
Reviewers: rafael, dexonsmith, grosser
Differential Revision: http://reviews.llvm.org/D7636
llvm-svn: 230786
2015-02-28 03:29:02 +08:00
|
|
|
%out.gep = getelementptr inbounds double, double addrspace(1)* %out, i32 %x.i
|
2014-10-11 06:01:59 +08:00
|
|
|
store double %sum, double addrspace(1)* %out.gep, align 8
|
|
|
|
|
ret void
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
; SI-LABEL: @simple_read2_f64_too_far
|
2014-11-05 22:50:53 +08:00
|
|
|
; SI-NOT ds_read2_b64
|
|
|
|
|
; SI: ds_read_b64 {{v\[[0-9]+:[0-9]+\]}}, v{{[0-9]+}}
|
|
|
|
|
; SI: ds_read_b64 {{v\[[0-9]+:[0-9]+\]}}, v{{[0-9]+}} offset:2056
|
|
|
|
|
; SI: s_endpgm
|
2014-10-11 06:01:59 +08:00
|
|
|
define void @simple_read2_f64_too_far(double addrspace(1)* %out) #0 {
|
|
|
|
|
%x.i = tail call i32 @llvm.r600.read.tidig.x() #1
|
[opaque pointer type] Add textual IR support for explicit type parameter to getelementptr instruction
One of several parallel first steps to remove the target type of pointers,
replacing them with a single opaque pointer type.
This adds an explicit type parameter to the gep instruction so that when the
first parameter becomes an opaque pointer type, the type to gep through is
still available to the instructions.
* This doesn't modify gep operators, only instructions (operators will be
handled separately)
* Textual IR changes only. Bitcode (including upgrade) and changing the
in-memory representation will be in separate changes.
* geps of vectors are transformed as:
getelementptr <4 x float*> %x, ...
->getelementptr float, <4 x float*> %x, ...
Then, once the opaque pointer type is introduced, this will ultimately look
like:
getelementptr float, <4 x ptr> %x
with the unambiguous interpretation that it is a vector of pointers to float.
* address spaces remain on the pointer, not the type:
getelementptr float addrspace(1)* %x
->getelementptr float, float addrspace(1)* %x
Then, eventually:
getelementptr float, ptr addrspace(1) %x
Importantly, the massive amount of test case churn has been automated by
same crappy python code. I had to manually update a few test cases that
wouldn't fit the script's model (r228970,r229196,r229197,r229198). The
python script just massages stdin and writes the result to stdout, I
then wrapped that in a shell script to handle replacing files, then
using the usual find+xargs to migrate all the files.
update.py:
import fileinput
import sys
import re
ibrep = re.compile(r"(^.*?[^%\w]getelementptr inbounds )(((?:<\d* x )?)(.*?)(| addrspace\(\d\)) *\*(|>)(?:$| *(?:%|@|null|undef|blockaddress|getelementptr|addrspacecast|bitcast|inttoptr|\[\[[a-zA-Z]|\{\{).*$))")
normrep = re.compile( r"(^.*?[^%\w]getelementptr )(((?:<\d* x )?)(.*?)(| addrspace\(\d\)) *\*(|>)(?:$| *(?:%|@|null|undef|blockaddress|getelementptr|addrspacecast|bitcast|inttoptr|\[\[[a-zA-Z]|\{\{).*$))")
def conv(match, line):
if not match:
return line
line = match.groups()[0]
if len(match.groups()[5]) == 0:
line += match.groups()[2]
line += match.groups()[3]
line += ", "
line += match.groups()[1]
line += "\n"
return line
for line in sys.stdin:
if line.find("getelementptr ") == line.find("getelementptr inbounds"):
if line.find("getelementptr inbounds") != line.find("getelementptr inbounds ("):
line = conv(re.match(ibrep, line), line)
elif line.find("getelementptr ") != line.find("getelementptr ("):
line = conv(re.match(normrep, line), line)
sys.stdout.write(line)
apply.sh:
for name in "$@"
do
python3 `dirname "$0"`/update.py < "$name" > "$name.tmp" && mv "$name.tmp" "$name"
rm -f "$name.tmp"
done
The actual commands:
From llvm/src:
find test/ -name *.ll | xargs ./apply.sh
From llvm/src/tools/clang:
find test/ -name *.mm -o -name *.m -o -name *.cpp -o -name *.c | xargs -I '{}' ../../apply.sh "{}"
From llvm/src/tools/polly:
find test/ -name *.ll | xargs ./apply.sh
After that, check-all (with llvm, clang, clang-tools-extra, lld,
compiler-rt, and polly all checked out).
The extra 'rm' in the apply.sh script is due to a few files in clang's test
suite using interesting unicode stuff that my python script was throwing
exceptions on. None of those files needed to be migrated, so it seemed
sufficient to ignore those cases.
Reviewers: rafael, dexonsmith, grosser
Differential Revision: http://reviews.llvm.org/D7636
llvm-svn: 230786
2015-02-28 03:29:02 +08:00
|
|
|
%arrayidx0 = getelementptr inbounds [512 x double], [512 x double] addrspace(3)* @lds.f64, i32 0, i32 %x.i
|
2015-02-28 05:17:42 +08:00
|
|
|
%val0 = load double, double addrspace(3)* %arrayidx0, align 8
|
2014-10-11 06:12:32 +08:00
|
|
|
%add.x = add nsw i32 %x.i, 257
|
[opaque pointer type] Add textual IR support for explicit type parameter to getelementptr instruction
One of several parallel first steps to remove the target type of pointers,
replacing them with a single opaque pointer type.
This adds an explicit type parameter to the gep instruction so that when the
first parameter becomes an opaque pointer type, the type to gep through is
still available to the instructions.
* This doesn't modify gep operators, only instructions (operators will be
handled separately)
* Textual IR changes only. Bitcode (including upgrade) and changing the
in-memory representation will be in separate changes.
* geps of vectors are transformed as:
getelementptr <4 x float*> %x, ...
->getelementptr float, <4 x float*> %x, ...
Then, once the opaque pointer type is introduced, this will ultimately look
like:
getelementptr float, <4 x ptr> %x
with the unambiguous interpretation that it is a vector of pointers to float.
* address spaces remain on the pointer, not the type:
getelementptr float addrspace(1)* %x
->getelementptr float, float addrspace(1)* %x
Then, eventually:
getelementptr float, ptr addrspace(1) %x
Importantly, the massive amount of test case churn has been automated by
same crappy python code. I had to manually update a few test cases that
wouldn't fit the script's model (r228970,r229196,r229197,r229198). The
python script just massages stdin and writes the result to stdout, I
then wrapped that in a shell script to handle replacing files, then
using the usual find+xargs to migrate all the files.
update.py:
import fileinput
import sys
import re
ibrep = re.compile(r"(^.*?[^%\w]getelementptr inbounds )(((?:<\d* x )?)(.*?)(| addrspace\(\d\)) *\*(|>)(?:$| *(?:%|@|null|undef|blockaddress|getelementptr|addrspacecast|bitcast|inttoptr|\[\[[a-zA-Z]|\{\{).*$))")
normrep = re.compile( r"(^.*?[^%\w]getelementptr )(((?:<\d* x )?)(.*?)(| addrspace\(\d\)) *\*(|>)(?:$| *(?:%|@|null|undef|blockaddress|getelementptr|addrspacecast|bitcast|inttoptr|\[\[[a-zA-Z]|\{\{).*$))")
def conv(match, line):
if not match:
return line
line = match.groups()[0]
if len(match.groups()[5]) == 0:
line += match.groups()[2]
line += match.groups()[3]
line += ", "
line += match.groups()[1]
line += "\n"
return line
for line in sys.stdin:
if line.find("getelementptr ") == line.find("getelementptr inbounds"):
if line.find("getelementptr inbounds") != line.find("getelementptr inbounds ("):
line = conv(re.match(ibrep, line), line)
elif line.find("getelementptr ") != line.find("getelementptr ("):
line = conv(re.match(normrep, line), line)
sys.stdout.write(line)
apply.sh:
for name in "$@"
do
python3 `dirname "$0"`/update.py < "$name" > "$name.tmp" && mv "$name.tmp" "$name"
rm -f "$name.tmp"
done
The actual commands:
From llvm/src:
find test/ -name *.ll | xargs ./apply.sh
From llvm/src/tools/clang:
find test/ -name *.mm -o -name *.m -o -name *.cpp -o -name *.c | xargs -I '{}' ../../apply.sh "{}"
From llvm/src/tools/polly:
find test/ -name *.ll | xargs ./apply.sh
After that, check-all (with llvm, clang, clang-tools-extra, lld,
compiler-rt, and polly all checked out).
The extra 'rm' in the apply.sh script is due to a few files in clang's test
suite using interesting unicode stuff that my python script was throwing
exceptions on. None of those files needed to be migrated, so it seemed
sufficient to ignore those cases.
Reviewers: rafael, dexonsmith, grosser
Differential Revision: http://reviews.llvm.org/D7636
llvm-svn: 230786
2015-02-28 03:29:02 +08:00
|
|
|
%arrayidx1 = getelementptr inbounds [512 x double], [512 x double] addrspace(3)* @lds.f64, i32 0, i32 %add.x
|
2015-02-28 05:17:42 +08:00
|
|
|
%val1 = load double, double addrspace(3)* %arrayidx1, align 8
|
2014-10-11 06:01:59 +08:00
|
|
|
%sum = fadd double %val0, %val1
|
[opaque pointer type] Add textual IR support for explicit type parameter to getelementptr instruction
One of several parallel first steps to remove the target type of pointers,
replacing them with a single opaque pointer type.
This adds an explicit type parameter to the gep instruction so that when the
first parameter becomes an opaque pointer type, the type to gep through is
still available to the instructions.
* This doesn't modify gep operators, only instructions (operators will be
handled separately)
* Textual IR changes only. Bitcode (including upgrade) and changing the
in-memory representation will be in separate changes.
* geps of vectors are transformed as:
getelementptr <4 x float*> %x, ...
->getelementptr float, <4 x float*> %x, ...
Then, once the opaque pointer type is introduced, this will ultimately look
like:
getelementptr float, <4 x ptr> %x
with the unambiguous interpretation that it is a vector of pointers to float.
* address spaces remain on the pointer, not the type:
getelementptr float addrspace(1)* %x
->getelementptr float, float addrspace(1)* %x
Then, eventually:
getelementptr float, ptr addrspace(1) %x
Importantly, the massive amount of test case churn has been automated by
same crappy python code. I had to manually update a few test cases that
wouldn't fit the script's model (r228970,r229196,r229197,r229198). The
python script just massages stdin and writes the result to stdout, I
then wrapped that in a shell script to handle replacing files, then
using the usual find+xargs to migrate all the files.
update.py:
import fileinput
import sys
import re
ibrep = re.compile(r"(^.*?[^%\w]getelementptr inbounds )(((?:<\d* x )?)(.*?)(| addrspace\(\d\)) *\*(|>)(?:$| *(?:%|@|null|undef|blockaddress|getelementptr|addrspacecast|bitcast|inttoptr|\[\[[a-zA-Z]|\{\{).*$))")
normrep = re.compile( r"(^.*?[^%\w]getelementptr )(((?:<\d* x )?)(.*?)(| addrspace\(\d\)) *\*(|>)(?:$| *(?:%|@|null|undef|blockaddress|getelementptr|addrspacecast|bitcast|inttoptr|\[\[[a-zA-Z]|\{\{).*$))")
def conv(match, line):
if not match:
return line
line = match.groups()[0]
if len(match.groups()[5]) == 0:
line += match.groups()[2]
line += match.groups()[3]
line += ", "
line += match.groups()[1]
line += "\n"
return line
for line in sys.stdin:
if line.find("getelementptr ") == line.find("getelementptr inbounds"):
if line.find("getelementptr inbounds") != line.find("getelementptr inbounds ("):
line = conv(re.match(ibrep, line), line)
elif line.find("getelementptr ") != line.find("getelementptr ("):
line = conv(re.match(normrep, line), line)
sys.stdout.write(line)
apply.sh:
for name in "$@"
do
python3 `dirname "$0"`/update.py < "$name" > "$name.tmp" && mv "$name.tmp" "$name"
rm -f "$name.tmp"
done
The actual commands:
From llvm/src:
find test/ -name *.ll | xargs ./apply.sh
From llvm/src/tools/clang:
find test/ -name *.mm -o -name *.m -o -name *.cpp -o -name *.c | xargs -I '{}' ../../apply.sh "{}"
From llvm/src/tools/polly:
find test/ -name *.ll | xargs ./apply.sh
After that, check-all (with llvm, clang, clang-tools-extra, lld,
compiler-rt, and polly all checked out).
The extra 'rm' in the apply.sh script is due to a few files in clang's test
suite using interesting unicode stuff that my python script was throwing
exceptions on. None of those files needed to be migrated, so it seemed
sufficient to ignore those cases.
Reviewers: rafael, dexonsmith, grosser
Differential Revision: http://reviews.llvm.org/D7636
llvm-svn: 230786
2015-02-28 03:29:02 +08:00
|
|
|
%out.gep = getelementptr inbounds double, double addrspace(1)* %out, i32 %x.i
|
2014-10-11 06:01:59 +08:00
|
|
|
store double %sum, double addrspace(1)* %out.gep, align 8
|
|
|
|
|
ret void
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
; Alignment only 4
|
|
|
|
|
; SI-LABEL: @misaligned_read2_f64
|
2015-04-08 09:09:19 +08:00
|
|
|
; SI: ds_read2_b32 v{{\[[0-9]+:[0-9]+\]}}, {{v[0-9]+}} offset1:1
|
2014-11-05 22:50:53 +08:00
|
|
|
; SI: ds_read2_b32 v{{\[[0-9]+:[0-9]+\]}}, {{v[0-9]+}} offset0:14 offset1:15
|
|
|
|
|
; SI: s_endpgm
|
2014-10-11 06:01:59 +08:00
|
|
|
define void @misaligned_read2_f64(double addrspace(1)* %out, double addrspace(3)* %lds) #0 {
|
|
|
|
|
%x.i = tail call i32 @llvm.r600.read.tidig.x() #1
|
[opaque pointer type] Add textual IR support for explicit type parameter to getelementptr instruction
One of several parallel first steps to remove the target type of pointers,
replacing them with a single opaque pointer type.
This adds an explicit type parameter to the gep instruction so that when the
first parameter becomes an opaque pointer type, the type to gep through is
still available to the instructions.
* This doesn't modify gep operators, only instructions (operators will be
handled separately)
* Textual IR changes only. Bitcode (including upgrade) and changing the
in-memory representation will be in separate changes.
* geps of vectors are transformed as:
getelementptr <4 x float*> %x, ...
->getelementptr float, <4 x float*> %x, ...
Then, once the opaque pointer type is introduced, this will ultimately look
like:
getelementptr float, <4 x ptr> %x
with the unambiguous interpretation that it is a vector of pointers to float.
* address spaces remain on the pointer, not the type:
getelementptr float addrspace(1)* %x
->getelementptr float, float addrspace(1)* %x
Then, eventually:
getelementptr float, ptr addrspace(1) %x
Importantly, the massive amount of test case churn has been automated by
same crappy python code. I had to manually update a few test cases that
wouldn't fit the script's model (r228970,r229196,r229197,r229198). The
python script just massages stdin and writes the result to stdout, I
then wrapped that in a shell script to handle replacing files, then
using the usual find+xargs to migrate all the files.
update.py:
import fileinput
import sys
import re
ibrep = re.compile(r"(^.*?[^%\w]getelementptr inbounds )(((?:<\d* x )?)(.*?)(| addrspace\(\d\)) *\*(|>)(?:$| *(?:%|@|null|undef|blockaddress|getelementptr|addrspacecast|bitcast|inttoptr|\[\[[a-zA-Z]|\{\{).*$))")
normrep = re.compile( r"(^.*?[^%\w]getelementptr )(((?:<\d* x )?)(.*?)(| addrspace\(\d\)) *\*(|>)(?:$| *(?:%|@|null|undef|blockaddress|getelementptr|addrspacecast|bitcast|inttoptr|\[\[[a-zA-Z]|\{\{).*$))")
def conv(match, line):
if not match:
return line
line = match.groups()[0]
if len(match.groups()[5]) == 0:
line += match.groups()[2]
line += match.groups()[3]
line += ", "
line += match.groups()[1]
line += "\n"
return line
for line in sys.stdin:
if line.find("getelementptr ") == line.find("getelementptr inbounds"):
if line.find("getelementptr inbounds") != line.find("getelementptr inbounds ("):
line = conv(re.match(ibrep, line), line)
elif line.find("getelementptr ") != line.find("getelementptr ("):
line = conv(re.match(normrep, line), line)
sys.stdout.write(line)
apply.sh:
for name in "$@"
do
python3 `dirname "$0"`/update.py < "$name" > "$name.tmp" && mv "$name.tmp" "$name"
rm -f "$name.tmp"
done
The actual commands:
From llvm/src:
find test/ -name *.ll | xargs ./apply.sh
From llvm/src/tools/clang:
find test/ -name *.mm -o -name *.m -o -name *.cpp -o -name *.c | xargs -I '{}' ../../apply.sh "{}"
From llvm/src/tools/polly:
find test/ -name *.ll | xargs ./apply.sh
After that, check-all (with llvm, clang, clang-tools-extra, lld,
compiler-rt, and polly all checked out).
The extra 'rm' in the apply.sh script is due to a few files in clang's test
suite using interesting unicode stuff that my python script was throwing
exceptions on. None of those files needed to be migrated, so it seemed
sufficient to ignore those cases.
Reviewers: rafael, dexonsmith, grosser
Differential Revision: http://reviews.llvm.org/D7636
llvm-svn: 230786
2015-02-28 03:29:02 +08:00
|
|
|
%arrayidx0 = getelementptr inbounds double, double addrspace(3)* %lds, i32 %x.i
|
2015-02-28 05:17:42 +08:00
|
|
|
%val0 = load double, double addrspace(3)* %arrayidx0, align 4
|
2014-10-11 06:01:59 +08:00
|
|
|
%add.x = add nsw i32 %x.i, 7
|
[opaque pointer type] Add textual IR support for explicit type parameter to getelementptr instruction
One of several parallel first steps to remove the target type of pointers,
replacing them with a single opaque pointer type.
This adds an explicit type parameter to the gep instruction so that when the
first parameter becomes an opaque pointer type, the type to gep through is
still available to the instructions.
* This doesn't modify gep operators, only instructions (operators will be
handled separately)
* Textual IR changes only. Bitcode (including upgrade) and changing the
in-memory representation will be in separate changes.
* geps of vectors are transformed as:
getelementptr <4 x float*> %x, ...
->getelementptr float, <4 x float*> %x, ...
Then, once the opaque pointer type is introduced, this will ultimately look
like:
getelementptr float, <4 x ptr> %x
with the unambiguous interpretation that it is a vector of pointers to float.
* address spaces remain on the pointer, not the type:
getelementptr float addrspace(1)* %x
->getelementptr float, float addrspace(1)* %x
Then, eventually:
getelementptr float, ptr addrspace(1) %x
Importantly, the massive amount of test case churn has been automated by
same crappy python code. I had to manually update a few test cases that
wouldn't fit the script's model (r228970,r229196,r229197,r229198). The
python script just massages stdin and writes the result to stdout, I
then wrapped that in a shell script to handle replacing files, then
using the usual find+xargs to migrate all the files.
update.py:
import fileinput
import sys
import re
ibrep = re.compile(r"(^.*?[^%\w]getelementptr inbounds )(((?:<\d* x )?)(.*?)(| addrspace\(\d\)) *\*(|>)(?:$| *(?:%|@|null|undef|blockaddress|getelementptr|addrspacecast|bitcast|inttoptr|\[\[[a-zA-Z]|\{\{).*$))")
normrep = re.compile( r"(^.*?[^%\w]getelementptr )(((?:<\d* x )?)(.*?)(| addrspace\(\d\)) *\*(|>)(?:$| *(?:%|@|null|undef|blockaddress|getelementptr|addrspacecast|bitcast|inttoptr|\[\[[a-zA-Z]|\{\{).*$))")
def conv(match, line):
if not match:
return line
line = match.groups()[0]
if len(match.groups()[5]) == 0:
line += match.groups()[2]
line += match.groups()[3]
line += ", "
line += match.groups()[1]
line += "\n"
return line
for line in sys.stdin:
if line.find("getelementptr ") == line.find("getelementptr inbounds"):
if line.find("getelementptr inbounds") != line.find("getelementptr inbounds ("):
line = conv(re.match(ibrep, line), line)
elif line.find("getelementptr ") != line.find("getelementptr ("):
line = conv(re.match(normrep, line), line)
sys.stdout.write(line)
apply.sh:
for name in "$@"
do
python3 `dirname "$0"`/update.py < "$name" > "$name.tmp" && mv "$name.tmp" "$name"
rm -f "$name.tmp"
done
The actual commands:
From llvm/src:
find test/ -name *.ll | xargs ./apply.sh
From llvm/src/tools/clang:
find test/ -name *.mm -o -name *.m -o -name *.cpp -o -name *.c | xargs -I '{}' ../../apply.sh "{}"
From llvm/src/tools/polly:
find test/ -name *.ll | xargs ./apply.sh
After that, check-all (with llvm, clang, clang-tools-extra, lld,
compiler-rt, and polly all checked out).
The extra 'rm' in the apply.sh script is due to a few files in clang's test
suite using interesting unicode stuff that my python script was throwing
exceptions on. None of those files needed to be migrated, so it seemed
sufficient to ignore those cases.
Reviewers: rafael, dexonsmith, grosser
Differential Revision: http://reviews.llvm.org/D7636
llvm-svn: 230786
2015-02-28 03:29:02 +08:00
|
|
|
%arrayidx1 = getelementptr inbounds double, double addrspace(3)* %lds, i32 %add.x
|
2015-02-28 05:17:42 +08:00
|
|
|
%val1 = load double, double addrspace(3)* %arrayidx1, align 4
|
2014-10-11 06:01:59 +08:00
|
|
|
%sum = fadd double %val0, %val1
|
[opaque pointer type] Add textual IR support for explicit type parameter to getelementptr instruction
One of several parallel first steps to remove the target type of pointers,
replacing them with a single opaque pointer type.
This adds an explicit type parameter to the gep instruction so that when the
first parameter becomes an opaque pointer type, the type to gep through is
still available to the instructions.
* This doesn't modify gep operators, only instructions (operators will be
handled separately)
* Textual IR changes only. Bitcode (including upgrade) and changing the
in-memory representation will be in separate changes.
* geps of vectors are transformed as:
getelementptr <4 x float*> %x, ...
->getelementptr float, <4 x float*> %x, ...
Then, once the opaque pointer type is introduced, this will ultimately look
like:
getelementptr float, <4 x ptr> %x
with the unambiguous interpretation that it is a vector of pointers to float.
* address spaces remain on the pointer, not the type:
getelementptr float addrspace(1)* %x
->getelementptr float, float addrspace(1)* %x
Then, eventually:
getelementptr float, ptr addrspace(1) %x
Importantly, the massive amount of test case churn has been automated by
same crappy python code. I had to manually update a few test cases that
wouldn't fit the script's model (r228970,r229196,r229197,r229198). The
python script just massages stdin and writes the result to stdout, I
then wrapped that in a shell script to handle replacing files, then
using the usual find+xargs to migrate all the files.
update.py:
import fileinput
import sys
import re
ibrep = re.compile(r"(^.*?[^%\w]getelementptr inbounds )(((?:<\d* x )?)(.*?)(| addrspace\(\d\)) *\*(|>)(?:$| *(?:%|@|null|undef|blockaddress|getelementptr|addrspacecast|bitcast|inttoptr|\[\[[a-zA-Z]|\{\{).*$))")
normrep = re.compile( r"(^.*?[^%\w]getelementptr )(((?:<\d* x )?)(.*?)(| addrspace\(\d\)) *\*(|>)(?:$| *(?:%|@|null|undef|blockaddress|getelementptr|addrspacecast|bitcast|inttoptr|\[\[[a-zA-Z]|\{\{).*$))")
def conv(match, line):
if not match:
return line
line = match.groups()[0]
if len(match.groups()[5]) == 0:
line += match.groups()[2]
line += match.groups()[3]
line += ", "
line += match.groups()[1]
line += "\n"
return line
for line in sys.stdin:
if line.find("getelementptr ") == line.find("getelementptr inbounds"):
if line.find("getelementptr inbounds") != line.find("getelementptr inbounds ("):
line = conv(re.match(ibrep, line), line)
elif line.find("getelementptr ") != line.find("getelementptr ("):
line = conv(re.match(normrep, line), line)
sys.stdout.write(line)
apply.sh:
for name in "$@"
do
python3 `dirname "$0"`/update.py < "$name" > "$name.tmp" && mv "$name.tmp" "$name"
rm -f "$name.tmp"
done
The actual commands:
From llvm/src:
find test/ -name *.ll | xargs ./apply.sh
From llvm/src/tools/clang:
find test/ -name *.mm -o -name *.m -o -name *.cpp -o -name *.c | xargs -I '{}' ../../apply.sh "{}"
From llvm/src/tools/polly:
find test/ -name *.ll | xargs ./apply.sh
After that, check-all (with llvm, clang, clang-tools-extra, lld,
compiler-rt, and polly all checked out).
The extra 'rm' in the apply.sh script is due to a few files in clang's test
suite using interesting unicode stuff that my python script was throwing
exceptions on. None of those files needed to be migrated, so it seemed
sufficient to ignore those cases.
Reviewers: rafael, dexonsmith, grosser
Differential Revision: http://reviews.llvm.org/D7636
llvm-svn: 230786
2015-02-28 03:29:02 +08:00
|
|
|
%out.gep = getelementptr inbounds double, double addrspace(1)* %out, i32 %x.i
|
2014-10-11 06:01:59 +08:00
|
|
|
store double %sum, double addrspace(1)* %out.gep, align 4
|
|
|
|
|
ret void
|
|
|
|
|
}
|
|
|
|
|
|
2014-11-14 03:56:13 +08:00
|
|
|
@foo = addrspace(3) global [4 x i32] undef, align 4
|
2014-10-15 01:21:19 +08:00
|
|
|
|
|
|
|
|
; SI-LABEL: @load_constant_adjacent_offsets
|
2014-11-05 22:50:53 +08:00
|
|
|
; SI: v_mov_b32_e32 [[ZERO:v[0-9]+]], 0{{$}}
|
2015-04-08 09:09:19 +08:00
|
|
|
; SI: ds_read2_b32 v{{\[[0-9]+:[0-9]+\]}}, [[ZERO]] offset1:1
|
2014-10-15 01:21:19 +08:00
|
|
|
define void @load_constant_adjacent_offsets(i32 addrspace(1)* %out) {
|
2015-03-14 02:20:45 +08:00
|
|
|
%val0 = load i32, i32 addrspace(3)* getelementptr inbounds ([4 x i32], [4 x i32] addrspace(3)* @foo, i32 0, i32 0), align 4
|
|
|
|
|
%val1 = load i32, i32 addrspace(3)* getelementptr inbounds ([4 x i32], [4 x i32] addrspace(3)* @foo, i32 0, i32 1), align 4
|
2014-10-15 01:21:19 +08:00
|
|
|
%sum = add i32 %val0, %val1
|
|
|
|
|
store i32 %sum, i32 addrspace(1)* %out, align 4
|
|
|
|
|
ret void
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
; SI-LABEL: @load_constant_disjoint_offsets
|
2014-11-05 22:50:53 +08:00
|
|
|
; SI: v_mov_b32_e32 [[ZERO:v[0-9]+]], 0{{$}}
|
2015-04-08 09:09:19 +08:00
|
|
|
; SI: ds_read2_b32 v{{\[[0-9]+:[0-9]+\]}}, [[ZERO]] offset1:2
|
2014-10-15 01:21:19 +08:00
|
|
|
define void @load_constant_disjoint_offsets(i32 addrspace(1)* %out) {
|
2015-03-14 02:20:45 +08:00
|
|
|
%val0 = load i32, i32 addrspace(3)* getelementptr inbounds ([4 x i32], [4 x i32] addrspace(3)* @foo, i32 0, i32 0), align 4
|
|
|
|
|
%val1 = load i32, i32 addrspace(3)* getelementptr inbounds ([4 x i32], [4 x i32] addrspace(3)* @foo, i32 0, i32 2), align 4
|
2014-10-15 01:21:19 +08:00
|
|
|
%sum = add i32 %val0, %val1
|
|
|
|
|
store i32 %sum, i32 addrspace(1)* %out, align 4
|
|
|
|
|
ret void
|
|
|
|
|
}
|
|
|
|
|
|
2014-11-14 03:56:13 +08:00
|
|
|
@bar = addrspace(3) global [4 x i64] undef, align 4
|
2014-10-16 02:06:43 +08:00
|
|
|
|
|
|
|
|
; SI-LABEL: @load_misaligned64_constant_offsets
|
2014-11-05 22:50:53 +08:00
|
|
|
; SI: v_mov_b32_e32 [[ZERO:v[0-9]+]], 0{{$}}
|
2015-04-08 09:09:19 +08:00
|
|
|
; SI: ds_read2_b32 v{{\[[0-9]+:[0-9]+\]}}, [[ZERO]] offset1:1
|
2014-11-05 22:50:53 +08:00
|
|
|
; SI: ds_read2_b32 v{{\[[0-9]+:[0-9]+\]}}, [[ZERO]] offset0:2 offset1:3
|
2014-10-16 02:06:43 +08:00
|
|
|
define void @load_misaligned64_constant_offsets(i64 addrspace(1)* %out) {
|
2015-03-14 02:20:45 +08:00
|
|
|
%val0 = load i64, i64 addrspace(3)* getelementptr inbounds ([4 x i64], [4 x i64] addrspace(3)* @bar, i32 0, i32 0), align 4
|
|
|
|
|
%val1 = load i64, i64 addrspace(3)* getelementptr inbounds ([4 x i64], [4 x i64] addrspace(3)* @bar, i32 0, i32 1), align 4
|
2014-10-16 02:06:43 +08:00
|
|
|
%sum = add i64 %val0, %val1
|
|
|
|
|
store i64 %sum, i64 addrspace(1)* %out, align 8
|
|
|
|
|
ret void
|
|
|
|
|
}
|
|
|
|
|
|
2014-11-14 03:56:13 +08:00
|
|
|
@bar.large = addrspace(3) global [4096 x i64] undef, align 4
|
2014-10-16 02:06:43 +08:00
|
|
|
|
|
|
|
|
; SI-LABEL: @load_misaligned64_constant_large_offsets
|
2014-11-05 22:50:53 +08:00
|
|
|
; SI-DAG: v_mov_b32_e32 [[BASE0:v[0-9]+]], 0x7ff8{{$}}
|
|
|
|
|
; SI-DAG: v_mov_b32_e32 [[BASE1:v[0-9]+]], 0x4000
|
2015-04-08 09:09:19 +08:00
|
|
|
; SI-DAG: ds_read2_b32 v{{\[[0-9]+:[0-9]+\]}}, [[BASE0]] offset1:1
|
|
|
|
|
; SI-DAG: ds_read2_b32 v{{\[[0-9]+:[0-9]+\]}}, [[BASE1]] offset1:1
|
2014-11-05 22:50:53 +08:00
|
|
|
; SI: s_endpgm
|
2014-10-16 02:06:43 +08:00
|
|
|
define void @load_misaligned64_constant_large_offsets(i64 addrspace(1)* %out) {
|
2015-03-14 02:20:45 +08:00
|
|
|
%val0 = load i64, i64 addrspace(3)* getelementptr inbounds ([4096 x i64], [4096 x i64] addrspace(3)* @bar.large, i32 0, i32 2048), align 4
|
|
|
|
|
%val1 = load i64, i64 addrspace(3)* getelementptr inbounds ([4096 x i64], [4096 x i64] addrspace(3)* @bar.large, i32 0, i32 4095), align 4
|
2014-10-16 02:06:43 +08:00
|
|
|
%sum = add i64 %val0, %val1
|
|
|
|
|
store i64 %sum, i64 addrspace(1)* %out, align 8
|
|
|
|
|
ret void
|
|
|
|
|
}
|
|
|
|
|
|
2014-11-14 03:56:13 +08:00
|
|
|
@sgemm.lA = internal unnamed_addr addrspace(3) global [264 x float] undef, align 4
|
|
|
|
|
@sgemm.lB = internal unnamed_addr addrspace(3) global [776 x float] undef, align 4
|
2014-10-11 06:01:59 +08:00
|
|
|
|
|
|
|
|
define void @sgemm_inner_loop_read2_sequence(float addrspace(1)* %C, i32 %lda, i32 %ldb) #0 {
|
|
|
|
|
%x.i = tail call i32 @llvm.r600.read.tgid.x() #1
|
|
|
|
|
%y.i = tail call i32 @llvm.r600.read.tidig.y() #1
|
[opaque pointer type] Add textual IR support for explicit type parameter to getelementptr instruction
One of several parallel first steps to remove the target type of pointers,
replacing them with a single opaque pointer type.
This adds an explicit type parameter to the gep instruction so that when the
first parameter becomes an opaque pointer type, the type to gep through is
still available to the instructions.
* This doesn't modify gep operators, only instructions (operators will be
handled separately)
* Textual IR changes only. Bitcode (including upgrade) and changing the
in-memory representation will be in separate changes.
* geps of vectors are transformed as:
getelementptr <4 x float*> %x, ...
->getelementptr float, <4 x float*> %x, ...
Then, once the opaque pointer type is introduced, this will ultimately look
like:
getelementptr float, <4 x ptr> %x
with the unambiguous interpretation that it is a vector of pointers to float.
* address spaces remain on the pointer, not the type:
getelementptr float addrspace(1)* %x
->getelementptr float, float addrspace(1)* %x
Then, eventually:
getelementptr float, ptr addrspace(1) %x
Importantly, the massive amount of test case churn has been automated by
same crappy python code. I had to manually update a few test cases that
wouldn't fit the script's model (r228970,r229196,r229197,r229198). The
python script just massages stdin and writes the result to stdout, I
then wrapped that in a shell script to handle replacing files, then
using the usual find+xargs to migrate all the files.
update.py:
import fileinput
import sys
import re
ibrep = re.compile(r"(^.*?[^%\w]getelementptr inbounds )(((?:<\d* x )?)(.*?)(| addrspace\(\d\)) *\*(|>)(?:$| *(?:%|@|null|undef|blockaddress|getelementptr|addrspacecast|bitcast|inttoptr|\[\[[a-zA-Z]|\{\{).*$))")
normrep = re.compile( r"(^.*?[^%\w]getelementptr )(((?:<\d* x )?)(.*?)(| addrspace\(\d\)) *\*(|>)(?:$| *(?:%|@|null|undef|blockaddress|getelementptr|addrspacecast|bitcast|inttoptr|\[\[[a-zA-Z]|\{\{).*$))")
def conv(match, line):
if not match:
return line
line = match.groups()[0]
if len(match.groups()[5]) == 0:
line += match.groups()[2]
line += match.groups()[3]
line += ", "
line += match.groups()[1]
line += "\n"
return line
for line in sys.stdin:
if line.find("getelementptr ") == line.find("getelementptr inbounds"):
if line.find("getelementptr inbounds") != line.find("getelementptr inbounds ("):
line = conv(re.match(ibrep, line), line)
elif line.find("getelementptr ") != line.find("getelementptr ("):
line = conv(re.match(normrep, line), line)
sys.stdout.write(line)
apply.sh:
for name in "$@"
do
python3 `dirname "$0"`/update.py < "$name" > "$name.tmp" && mv "$name.tmp" "$name"
rm -f "$name.tmp"
done
The actual commands:
From llvm/src:
find test/ -name *.ll | xargs ./apply.sh
From llvm/src/tools/clang:
find test/ -name *.mm -o -name *.m -o -name *.cpp -o -name *.c | xargs -I '{}' ../../apply.sh "{}"
From llvm/src/tools/polly:
find test/ -name *.ll | xargs ./apply.sh
After that, check-all (with llvm, clang, clang-tools-extra, lld,
compiler-rt, and polly all checked out).
The extra 'rm' in the apply.sh script is due to a few files in clang's test
suite using interesting unicode stuff that my python script was throwing
exceptions on. None of those files needed to be migrated, so it seemed
sufficient to ignore those cases.
Reviewers: rafael, dexonsmith, grosser
Differential Revision: http://reviews.llvm.org/D7636
llvm-svn: 230786
2015-02-28 03:29:02 +08:00
|
|
|
%arrayidx44 = getelementptr inbounds [264 x float], [264 x float] addrspace(3)* @sgemm.lA, i32 0, i32 %x.i
|
2015-02-28 05:17:42 +08:00
|
|
|
%tmp16 = load float, float addrspace(3)* %arrayidx44, align 4
|
2014-10-11 06:01:59 +08:00
|
|
|
%add47 = add nsw i32 %x.i, 1
|
[opaque pointer type] Add textual IR support for explicit type parameter to getelementptr instruction
One of several parallel first steps to remove the target type of pointers,
replacing them with a single opaque pointer type.
This adds an explicit type parameter to the gep instruction so that when the
first parameter becomes an opaque pointer type, the type to gep through is
still available to the instructions.
* This doesn't modify gep operators, only instructions (operators will be
handled separately)
* Textual IR changes only. Bitcode (including upgrade) and changing the
in-memory representation will be in separate changes.
* geps of vectors are transformed as:
getelementptr <4 x float*> %x, ...
->getelementptr float, <4 x float*> %x, ...
Then, once the opaque pointer type is introduced, this will ultimately look
like:
getelementptr float, <4 x ptr> %x
with the unambiguous interpretation that it is a vector of pointers to float.
* address spaces remain on the pointer, not the type:
getelementptr float addrspace(1)* %x
->getelementptr float, float addrspace(1)* %x
Then, eventually:
getelementptr float, ptr addrspace(1) %x
Importantly, the massive amount of test case churn has been automated by
same crappy python code. I had to manually update a few test cases that
wouldn't fit the script's model (r228970,r229196,r229197,r229198). The
python script just massages stdin and writes the result to stdout, I
then wrapped that in a shell script to handle replacing files, then
using the usual find+xargs to migrate all the files.
update.py:
import fileinput
import sys
import re
ibrep = re.compile(r"(^.*?[^%\w]getelementptr inbounds )(((?:<\d* x )?)(.*?)(| addrspace\(\d\)) *\*(|>)(?:$| *(?:%|@|null|undef|blockaddress|getelementptr|addrspacecast|bitcast|inttoptr|\[\[[a-zA-Z]|\{\{).*$))")
normrep = re.compile( r"(^.*?[^%\w]getelementptr )(((?:<\d* x )?)(.*?)(| addrspace\(\d\)) *\*(|>)(?:$| *(?:%|@|null|undef|blockaddress|getelementptr|addrspacecast|bitcast|inttoptr|\[\[[a-zA-Z]|\{\{).*$))")
def conv(match, line):
if not match:
return line
line = match.groups()[0]
if len(match.groups()[5]) == 0:
line += match.groups()[2]
line += match.groups()[3]
line += ", "
line += match.groups()[1]
line += "\n"
return line
for line in sys.stdin:
if line.find("getelementptr ") == line.find("getelementptr inbounds"):
if line.find("getelementptr inbounds") != line.find("getelementptr inbounds ("):
line = conv(re.match(ibrep, line), line)
elif line.find("getelementptr ") != line.find("getelementptr ("):
line = conv(re.match(normrep, line), line)
sys.stdout.write(line)
apply.sh:
for name in "$@"
do
python3 `dirname "$0"`/update.py < "$name" > "$name.tmp" && mv "$name.tmp" "$name"
rm -f "$name.tmp"
done
The actual commands:
From llvm/src:
find test/ -name *.ll | xargs ./apply.sh
From llvm/src/tools/clang:
find test/ -name *.mm -o -name *.m -o -name *.cpp -o -name *.c | xargs -I '{}' ../../apply.sh "{}"
From llvm/src/tools/polly:
find test/ -name *.ll | xargs ./apply.sh
After that, check-all (with llvm, clang, clang-tools-extra, lld,
compiler-rt, and polly all checked out).
The extra 'rm' in the apply.sh script is due to a few files in clang's test
suite using interesting unicode stuff that my python script was throwing
exceptions on. None of those files needed to be migrated, so it seemed
sufficient to ignore those cases.
Reviewers: rafael, dexonsmith, grosser
Differential Revision: http://reviews.llvm.org/D7636
llvm-svn: 230786
2015-02-28 03:29:02 +08:00
|
|
|
%arrayidx48 = getelementptr inbounds [264 x float], [264 x float] addrspace(3)* @sgemm.lA, i32 0, i32 %add47
|
2015-02-28 05:17:42 +08:00
|
|
|
%tmp17 = load float, float addrspace(3)* %arrayidx48, align 4
|
2014-10-11 06:01:59 +08:00
|
|
|
%add51 = add nsw i32 %x.i, 16
|
[opaque pointer type] Add textual IR support for explicit type parameter to getelementptr instruction
One of several parallel first steps to remove the target type of pointers,
replacing them with a single opaque pointer type.
This adds an explicit type parameter to the gep instruction so that when the
first parameter becomes an opaque pointer type, the type to gep through is
still available to the instructions.
* This doesn't modify gep operators, only instructions (operators will be
handled separately)
* Textual IR changes only. Bitcode (including upgrade) and changing the
in-memory representation will be in separate changes.
* geps of vectors are transformed as:
getelementptr <4 x float*> %x, ...
->getelementptr float, <4 x float*> %x, ...
Then, once the opaque pointer type is introduced, this will ultimately look
like:
getelementptr float, <4 x ptr> %x
with the unambiguous interpretation that it is a vector of pointers to float.
* address spaces remain on the pointer, not the type:
getelementptr float addrspace(1)* %x
->getelementptr float, float addrspace(1)* %x
Then, eventually:
getelementptr float, ptr addrspace(1) %x
Importantly, the massive amount of test case churn has been automated by
same crappy python code. I had to manually update a few test cases that
wouldn't fit the script's model (r228970,r229196,r229197,r229198). The
python script just massages stdin and writes the result to stdout, I
then wrapped that in a shell script to handle replacing files, then
using the usual find+xargs to migrate all the files.
update.py:
import fileinput
import sys
import re
ibrep = re.compile(r"(^.*?[^%\w]getelementptr inbounds )(((?:<\d* x )?)(.*?)(| addrspace\(\d\)) *\*(|>)(?:$| *(?:%|@|null|undef|blockaddress|getelementptr|addrspacecast|bitcast|inttoptr|\[\[[a-zA-Z]|\{\{).*$))")
normrep = re.compile( r"(^.*?[^%\w]getelementptr )(((?:<\d* x )?)(.*?)(| addrspace\(\d\)) *\*(|>)(?:$| *(?:%|@|null|undef|blockaddress|getelementptr|addrspacecast|bitcast|inttoptr|\[\[[a-zA-Z]|\{\{).*$))")
def conv(match, line):
if not match:
return line
line = match.groups()[0]
if len(match.groups()[5]) == 0:
line += match.groups()[2]
line += match.groups()[3]
line += ", "
line += match.groups()[1]
line += "\n"
return line
for line in sys.stdin:
if line.find("getelementptr ") == line.find("getelementptr inbounds"):
if line.find("getelementptr inbounds") != line.find("getelementptr inbounds ("):
line = conv(re.match(ibrep, line), line)
elif line.find("getelementptr ") != line.find("getelementptr ("):
line = conv(re.match(normrep, line), line)
sys.stdout.write(line)
apply.sh:
for name in "$@"
do
python3 `dirname "$0"`/update.py < "$name" > "$name.tmp" && mv "$name.tmp" "$name"
rm -f "$name.tmp"
done
The actual commands:
From llvm/src:
find test/ -name *.ll | xargs ./apply.sh
From llvm/src/tools/clang:
find test/ -name *.mm -o -name *.m -o -name *.cpp -o -name *.c | xargs -I '{}' ../../apply.sh "{}"
From llvm/src/tools/polly:
find test/ -name *.ll | xargs ./apply.sh
After that, check-all (with llvm, clang, clang-tools-extra, lld,
compiler-rt, and polly all checked out).
The extra 'rm' in the apply.sh script is due to a few files in clang's test
suite using interesting unicode stuff that my python script was throwing
exceptions on. None of those files needed to be migrated, so it seemed
sufficient to ignore those cases.
Reviewers: rafael, dexonsmith, grosser
Differential Revision: http://reviews.llvm.org/D7636
llvm-svn: 230786
2015-02-28 03:29:02 +08:00
|
|
|
%arrayidx52 = getelementptr inbounds [264 x float], [264 x float] addrspace(3)* @sgemm.lA, i32 0, i32 %add51
|
2015-02-28 05:17:42 +08:00
|
|
|
%tmp18 = load float, float addrspace(3)* %arrayidx52, align 4
|
2014-10-11 06:01:59 +08:00
|
|
|
%add55 = add nsw i32 %x.i, 17
|
[opaque pointer type] Add textual IR support for explicit type parameter to getelementptr instruction
One of several parallel first steps to remove the target type of pointers,
replacing them with a single opaque pointer type.
This adds an explicit type parameter to the gep instruction so that when the
first parameter becomes an opaque pointer type, the type to gep through is
still available to the instructions.
* This doesn't modify gep operators, only instructions (operators will be
handled separately)
* Textual IR changes only. Bitcode (including upgrade) and changing the
in-memory representation will be in separate changes.
* geps of vectors are transformed as:
getelementptr <4 x float*> %x, ...
->getelementptr float, <4 x float*> %x, ...
Then, once the opaque pointer type is introduced, this will ultimately look
like:
getelementptr float, <4 x ptr> %x
with the unambiguous interpretation that it is a vector of pointers to float.
* address spaces remain on the pointer, not the type:
getelementptr float addrspace(1)* %x
->getelementptr float, float addrspace(1)* %x
Then, eventually:
getelementptr float, ptr addrspace(1) %x
Importantly, the massive amount of test case churn has been automated by
same crappy python code. I had to manually update a few test cases that
wouldn't fit the script's model (r228970,r229196,r229197,r229198). The
python script just massages stdin and writes the result to stdout, I
then wrapped that in a shell script to handle replacing files, then
using the usual find+xargs to migrate all the files.
update.py:
import fileinput
import sys
import re
ibrep = re.compile(r"(^.*?[^%\w]getelementptr inbounds )(((?:<\d* x )?)(.*?)(| addrspace\(\d\)) *\*(|>)(?:$| *(?:%|@|null|undef|blockaddress|getelementptr|addrspacecast|bitcast|inttoptr|\[\[[a-zA-Z]|\{\{).*$))")
normrep = re.compile( r"(^.*?[^%\w]getelementptr )(((?:<\d* x )?)(.*?)(| addrspace\(\d\)) *\*(|>)(?:$| *(?:%|@|null|undef|blockaddress|getelementptr|addrspacecast|bitcast|inttoptr|\[\[[a-zA-Z]|\{\{).*$))")
def conv(match, line):
if not match:
return line
line = match.groups()[0]
if len(match.groups()[5]) == 0:
line += match.groups()[2]
line += match.groups()[3]
line += ", "
line += match.groups()[1]
line += "\n"
return line
for line in sys.stdin:
if line.find("getelementptr ") == line.find("getelementptr inbounds"):
if line.find("getelementptr inbounds") != line.find("getelementptr inbounds ("):
line = conv(re.match(ibrep, line), line)
elif line.find("getelementptr ") != line.find("getelementptr ("):
line = conv(re.match(normrep, line), line)
sys.stdout.write(line)
apply.sh:
for name in "$@"
do
python3 `dirname "$0"`/update.py < "$name" > "$name.tmp" && mv "$name.tmp" "$name"
rm -f "$name.tmp"
done
The actual commands:
From llvm/src:
find test/ -name *.ll | xargs ./apply.sh
From llvm/src/tools/clang:
find test/ -name *.mm -o -name *.m -o -name *.cpp -o -name *.c | xargs -I '{}' ../../apply.sh "{}"
From llvm/src/tools/polly:
find test/ -name *.ll | xargs ./apply.sh
After that, check-all (with llvm, clang, clang-tools-extra, lld,
compiler-rt, and polly all checked out).
The extra 'rm' in the apply.sh script is due to a few files in clang's test
suite using interesting unicode stuff that my python script was throwing
exceptions on. None of those files needed to be migrated, so it seemed
sufficient to ignore those cases.
Reviewers: rafael, dexonsmith, grosser
Differential Revision: http://reviews.llvm.org/D7636
llvm-svn: 230786
2015-02-28 03:29:02 +08:00
|
|
|
%arrayidx56 = getelementptr inbounds [264 x float], [264 x float] addrspace(3)* @sgemm.lA, i32 0, i32 %add55
|
2015-02-28 05:17:42 +08:00
|
|
|
%tmp19 = load float, float addrspace(3)* %arrayidx56, align 4
|
[opaque pointer type] Add textual IR support for explicit type parameter to getelementptr instruction
One of several parallel first steps to remove the target type of pointers,
replacing them with a single opaque pointer type.
This adds an explicit type parameter to the gep instruction so that when the
first parameter becomes an opaque pointer type, the type to gep through is
still available to the instructions.
* This doesn't modify gep operators, only instructions (operators will be
handled separately)
* Textual IR changes only. Bitcode (including upgrade) and changing the
in-memory representation will be in separate changes.
* geps of vectors are transformed as:
getelementptr <4 x float*> %x, ...
->getelementptr float, <4 x float*> %x, ...
Then, once the opaque pointer type is introduced, this will ultimately look
like:
getelementptr float, <4 x ptr> %x
with the unambiguous interpretation that it is a vector of pointers to float.
* address spaces remain on the pointer, not the type:
getelementptr float addrspace(1)* %x
->getelementptr float, float addrspace(1)* %x
Then, eventually:
getelementptr float, ptr addrspace(1) %x
Importantly, the massive amount of test case churn has been automated by
same crappy python code. I had to manually update a few test cases that
wouldn't fit the script's model (r228970,r229196,r229197,r229198). The
python script just massages stdin and writes the result to stdout, I
then wrapped that in a shell script to handle replacing files, then
using the usual find+xargs to migrate all the files.
update.py:
import fileinput
import sys
import re
ibrep = re.compile(r"(^.*?[^%\w]getelementptr inbounds )(((?:<\d* x )?)(.*?)(| addrspace\(\d\)) *\*(|>)(?:$| *(?:%|@|null|undef|blockaddress|getelementptr|addrspacecast|bitcast|inttoptr|\[\[[a-zA-Z]|\{\{).*$))")
normrep = re.compile( r"(^.*?[^%\w]getelementptr )(((?:<\d* x )?)(.*?)(| addrspace\(\d\)) *\*(|>)(?:$| *(?:%|@|null|undef|blockaddress|getelementptr|addrspacecast|bitcast|inttoptr|\[\[[a-zA-Z]|\{\{).*$))")
def conv(match, line):
if not match:
return line
line = match.groups()[0]
if len(match.groups()[5]) == 0:
line += match.groups()[2]
line += match.groups()[3]
line += ", "
line += match.groups()[1]
line += "\n"
return line
for line in sys.stdin:
if line.find("getelementptr ") == line.find("getelementptr inbounds"):
if line.find("getelementptr inbounds") != line.find("getelementptr inbounds ("):
line = conv(re.match(ibrep, line), line)
elif line.find("getelementptr ") != line.find("getelementptr ("):
line = conv(re.match(normrep, line), line)
sys.stdout.write(line)
apply.sh:
for name in "$@"
do
python3 `dirname "$0"`/update.py < "$name" > "$name.tmp" && mv "$name.tmp" "$name"
rm -f "$name.tmp"
done
The actual commands:
From llvm/src:
find test/ -name *.ll | xargs ./apply.sh
From llvm/src/tools/clang:
find test/ -name *.mm -o -name *.m -o -name *.cpp -o -name *.c | xargs -I '{}' ../../apply.sh "{}"
From llvm/src/tools/polly:
find test/ -name *.ll | xargs ./apply.sh
After that, check-all (with llvm, clang, clang-tools-extra, lld,
compiler-rt, and polly all checked out).
The extra 'rm' in the apply.sh script is due to a few files in clang's test
suite using interesting unicode stuff that my python script was throwing
exceptions on. None of those files needed to be migrated, so it seemed
sufficient to ignore those cases.
Reviewers: rafael, dexonsmith, grosser
Differential Revision: http://reviews.llvm.org/D7636
llvm-svn: 230786
2015-02-28 03:29:02 +08:00
|
|
|
%arrayidx60 = getelementptr inbounds [776 x float], [776 x float] addrspace(3)* @sgemm.lB, i32 0, i32 %y.i
|
2015-02-28 05:17:42 +08:00
|
|
|
%tmp20 = load float, float addrspace(3)* %arrayidx60, align 4
|
2014-10-11 06:01:59 +08:00
|
|
|
%add63 = add nsw i32 %y.i, 1
|
[opaque pointer type] Add textual IR support for explicit type parameter to getelementptr instruction
One of several parallel first steps to remove the target type of pointers,
replacing them with a single opaque pointer type.
This adds an explicit type parameter to the gep instruction so that when the
first parameter becomes an opaque pointer type, the type to gep through is
still available to the instructions.
* This doesn't modify gep operators, only instructions (operators will be
handled separately)
* Textual IR changes only. Bitcode (including upgrade) and changing the
in-memory representation will be in separate changes.
* geps of vectors are transformed as:
getelementptr <4 x float*> %x, ...
->getelementptr float, <4 x float*> %x, ...
Then, once the opaque pointer type is introduced, this will ultimately look
like:
getelementptr float, <4 x ptr> %x
with the unambiguous interpretation that it is a vector of pointers to float.
* address spaces remain on the pointer, not the type:
getelementptr float addrspace(1)* %x
->getelementptr float, float addrspace(1)* %x
Then, eventually:
getelementptr float, ptr addrspace(1) %x
Importantly, the massive amount of test case churn has been automated by
same crappy python code. I had to manually update a few test cases that
wouldn't fit the script's model (r228970,r229196,r229197,r229198). The
python script just massages stdin and writes the result to stdout, I
then wrapped that in a shell script to handle replacing files, then
using the usual find+xargs to migrate all the files.
update.py:
import fileinput
import sys
import re
ibrep = re.compile(r"(^.*?[^%\w]getelementptr inbounds )(((?:<\d* x )?)(.*?)(| addrspace\(\d\)) *\*(|>)(?:$| *(?:%|@|null|undef|blockaddress|getelementptr|addrspacecast|bitcast|inttoptr|\[\[[a-zA-Z]|\{\{).*$))")
normrep = re.compile( r"(^.*?[^%\w]getelementptr )(((?:<\d* x )?)(.*?)(| addrspace\(\d\)) *\*(|>)(?:$| *(?:%|@|null|undef|blockaddress|getelementptr|addrspacecast|bitcast|inttoptr|\[\[[a-zA-Z]|\{\{).*$))")
def conv(match, line):
if not match:
return line
line = match.groups()[0]
if len(match.groups()[5]) == 0:
line += match.groups()[2]
line += match.groups()[3]
line += ", "
line += match.groups()[1]
line += "\n"
return line
for line in sys.stdin:
if line.find("getelementptr ") == line.find("getelementptr inbounds"):
if line.find("getelementptr inbounds") != line.find("getelementptr inbounds ("):
line = conv(re.match(ibrep, line), line)
elif line.find("getelementptr ") != line.find("getelementptr ("):
line = conv(re.match(normrep, line), line)
sys.stdout.write(line)
apply.sh:
for name in "$@"
do
python3 `dirname "$0"`/update.py < "$name" > "$name.tmp" && mv "$name.tmp" "$name"
rm -f "$name.tmp"
done
The actual commands:
From llvm/src:
find test/ -name *.ll | xargs ./apply.sh
From llvm/src/tools/clang:
find test/ -name *.mm -o -name *.m -o -name *.cpp -o -name *.c | xargs -I '{}' ../../apply.sh "{}"
From llvm/src/tools/polly:
find test/ -name *.ll | xargs ./apply.sh
After that, check-all (with llvm, clang, clang-tools-extra, lld,
compiler-rt, and polly all checked out).
The extra 'rm' in the apply.sh script is due to a few files in clang's test
suite using interesting unicode stuff that my python script was throwing
exceptions on. None of those files needed to be migrated, so it seemed
sufficient to ignore those cases.
Reviewers: rafael, dexonsmith, grosser
Differential Revision: http://reviews.llvm.org/D7636
llvm-svn: 230786
2015-02-28 03:29:02 +08:00
|
|
|
%arrayidx64 = getelementptr inbounds [776 x float], [776 x float] addrspace(3)* @sgemm.lB, i32 0, i32 %add63
|
2015-02-28 05:17:42 +08:00
|
|
|
%tmp21 = load float, float addrspace(3)* %arrayidx64, align 4
|
2014-10-11 06:01:59 +08:00
|
|
|
%add67 = add nsw i32 %y.i, 32
|
[opaque pointer type] Add textual IR support for explicit type parameter to getelementptr instruction
One of several parallel first steps to remove the target type of pointers,
replacing them with a single opaque pointer type.
This adds an explicit type parameter to the gep instruction so that when the
first parameter becomes an opaque pointer type, the type to gep through is
still available to the instructions.
* This doesn't modify gep operators, only instructions (operators will be
handled separately)
* Textual IR changes only. Bitcode (including upgrade) and changing the
in-memory representation will be in separate changes.
* geps of vectors are transformed as:
getelementptr <4 x float*> %x, ...
->getelementptr float, <4 x float*> %x, ...
Then, once the opaque pointer type is introduced, this will ultimately look
like:
getelementptr float, <4 x ptr> %x
with the unambiguous interpretation that it is a vector of pointers to float.
* address spaces remain on the pointer, not the type:
getelementptr float addrspace(1)* %x
->getelementptr float, float addrspace(1)* %x
Then, eventually:
getelementptr float, ptr addrspace(1) %x
Importantly, the massive amount of test case churn has been automated by
same crappy python code. I had to manually update a few test cases that
wouldn't fit the script's model (r228970,r229196,r229197,r229198). The
python script just massages stdin and writes the result to stdout, I
then wrapped that in a shell script to handle replacing files, then
using the usual find+xargs to migrate all the files.
update.py:
import fileinput
import sys
import re
ibrep = re.compile(r"(^.*?[^%\w]getelementptr inbounds )(((?:<\d* x )?)(.*?)(| addrspace\(\d\)) *\*(|>)(?:$| *(?:%|@|null|undef|blockaddress|getelementptr|addrspacecast|bitcast|inttoptr|\[\[[a-zA-Z]|\{\{).*$))")
normrep = re.compile( r"(^.*?[^%\w]getelementptr )(((?:<\d* x )?)(.*?)(| addrspace\(\d\)) *\*(|>)(?:$| *(?:%|@|null|undef|blockaddress|getelementptr|addrspacecast|bitcast|inttoptr|\[\[[a-zA-Z]|\{\{).*$))")
def conv(match, line):
if not match:
return line
line = match.groups()[0]
if len(match.groups()[5]) == 0:
line += match.groups()[2]
line += match.groups()[3]
line += ", "
line += match.groups()[1]
line += "\n"
return line
for line in sys.stdin:
if line.find("getelementptr ") == line.find("getelementptr inbounds"):
if line.find("getelementptr inbounds") != line.find("getelementptr inbounds ("):
line = conv(re.match(ibrep, line), line)
elif line.find("getelementptr ") != line.find("getelementptr ("):
line = conv(re.match(normrep, line), line)
sys.stdout.write(line)
apply.sh:
for name in "$@"
do
python3 `dirname "$0"`/update.py < "$name" > "$name.tmp" && mv "$name.tmp" "$name"
rm -f "$name.tmp"
done
The actual commands:
From llvm/src:
find test/ -name *.ll | xargs ./apply.sh
From llvm/src/tools/clang:
find test/ -name *.mm -o -name *.m -o -name *.cpp -o -name *.c | xargs -I '{}' ../../apply.sh "{}"
From llvm/src/tools/polly:
find test/ -name *.ll | xargs ./apply.sh
After that, check-all (with llvm, clang, clang-tools-extra, lld,
compiler-rt, and polly all checked out).
The extra 'rm' in the apply.sh script is due to a few files in clang's test
suite using interesting unicode stuff that my python script was throwing
exceptions on. None of those files needed to be migrated, so it seemed
sufficient to ignore those cases.
Reviewers: rafael, dexonsmith, grosser
Differential Revision: http://reviews.llvm.org/D7636
llvm-svn: 230786
2015-02-28 03:29:02 +08:00
|
|
|
%arrayidx68 = getelementptr inbounds [776 x float], [776 x float] addrspace(3)* @sgemm.lB, i32 0, i32 %add67
|
2015-02-28 05:17:42 +08:00
|
|
|
%tmp22 = load float, float addrspace(3)* %arrayidx68, align 4
|
2014-10-11 06:01:59 +08:00
|
|
|
%add71 = add nsw i32 %y.i, 33
|
[opaque pointer type] Add textual IR support for explicit type parameter to getelementptr instruction
One of several parallel first steps to remove the target type of pointers,
replacing them with a single opaque pointer type.
This adds an explicit type parameter to the gep instruction so that when the
first parameter becomes an opaque pointer type, the type to gep through is
still available to the instructions.
* This doesn't modify gep operators, only instructions (operators will be
handled separately)
* Textual IR changes only. Bitcode (including upgrade) and changing the
in-memory representation will be in separate changes.
* geps of vectors are transformed as:
getelementptr <4 x float*> %x, ...
->getelementptr float, <4 x float*> %x, ...
Then, once the opaque pointer type is introduced, this will ultimately look
like:
getelementptr float, <4 x ptr> %x
with the unambiguous interpretation that it is a vector of pointers to float.
* address spaces remain on the pointer, not the type:
getelementptr float addrspace(1)* %x
->getelementptr float, float addrspace(1)* %x
Then, eventually:
getelementptr float, ptr addrspace(1) %x
Importantly, the massive amount of test case churn has been automated by
same crappy python code. I had to manually update a few test cases that
wouldn't fit the script's model (r228970,r229196,r229197,r229198). The
python script just massages stdin and writes the result to stdout, I
then wrapped that in a shell script to handle replacing files, then
using the usual find+xargs to migrate all the files.
update.py:
import fileinput
import sys
import re
ibrep = re.compile(r"(^.*?[^%\w]getelementptr inbounds )(((?:<\d* x )?)(.*?)(| addrspace\(\d\)) *\*(|>)(?:$| *(?:%|@|null|undef|blockaddress|getelementptr|addrspacecast|bitcast|inttoptr|\[\[[a-zA-Z]|\{\{).*$))")
normrep = re.compile( r"(^.*?[^%\w]getelementptr )(((?:<\d* x )?)(.*?)(| addrspace\(\d\)) *\*(|>)(?:$| *(?:%|@|null|undef|blockaddress|getelementptr|addrspacecast|bitcast|inttoptr|\[\[[a-zA-Z]|\{\{).*$))")
def conv(match, line):
if not match:
return line
line = match.groups()[0]
if len(match.groups()[5]) == 0:
line += match.groups()[2]
line += match.groups()[3]
line += ", "
line += match.groups()[1]
line += "\n"
return line
for line in sys.stdin:
if line.find("getelementptr ") == line.find("getelementptr inbounds"):
if line.find("getelementptr inbounds") != line.find("getelementptr inbounds ("):
line = conv(re.match(ibrep, line), line)
elif line.find("getelementptr ") != line.find("getelementptr ("):
line = conv(re.match(normrep, line), line)
sys.stdout.write(line)
apply.sh:
for name in "$@"
do
python3 `dirname "$0"`/update.py < "$name" > "$name.tmp" && mv "$name.tmp" "$name"
rm -f "$name.tmp"
done
The actual commands:
From llvm/src:
find test/ -name *.ll | xargs ./apply.sh
From llvm/src/tools/clang:
find test/ -name *.mm -o -name *.m -o -name *.cpp -o -name *.c | xargs -I '{}' ../../apply.sh "{}"
From llvm/src/tools/polly:
find test/ -name *.ll | xargs ./apply.sh
After that, check-all (with llvm, clang, clang-tools-extra, lld,
compiler-rt, and polly all checked out).
The extra 'rm' in the apply.sh script is due to a few files in clang's test
suite using interesting unicode stuff that my python script was throwing
exceptions on. None of those files needed to be migrated, so it seemed
sufficient to ignore those cases.
Reviewers: rafael, dexonsmith, grosser
Differential Revision: http://reviews.llvm.org/D7636
llvm-svn: 230786
2015-02-28 03:29:02 +08:00
|
|
|
%arrayidx72 = getelementptr inbounds [776 x float], [776 x float] addrspace(3)* @sgemm.lB, i32 0, i32 %add71
|
2015-02-28 05:17:42 +08:00
|
|
|
%tmp23 = load float, float addrspace(3)* %arrayidx72, align 4
|
2014-10-11 06:01:59 +08:00
|
|
|
%add75 = add nsw i32 %y.i, 64
|
[opaque pointer type] Add textual IR support for explicit type parameter to getelementptr instruction
One of several parallel first steps to remove the target type of pointers,
replacing them with a single opaque pointer type.
This adds an explicit type parameter to the gep instruction so that when the
first parameter becomes an opaque pointer type, the type to gep through is
still available to the instructions.
* This doesn't modify gep operators, only instructions (operators will be
handled separately)
* Textual IR changes only. Bitcode (including upgrade) and changing the
in-memory representation will be in separate changes.
* geps of vectors are transformed as:
getelementptr <4 x float*> %x, ...
->getelementptr float, <4 x float*> %x, ...
Then, once the opaque pointer type is introduced, this will ultimately look
like:
getelementptr float, <4 x ptr> %x
with the unambiguous interpretation that it is a vector of pointers to float.
* address spaces remain on the pointer, not the type:
getelementptr float addrspace(1)* %x
->getelementptr float, float addrspace(1)* %x
Then, eventually:
getelementptr float, ptr addrspace(1) %x
Importantly, the massive amount of test case churn has been automated by
same crappy python code. I had to manually update a few test cases that
wouldn't fit the script's model (r228970,r229196,r229197,r229198). The
python script just massages stdin and writes the result to stdout, I
then wrapped that in a shell script to handle replacing files, then
using the usual find+xargs to migrate all the files.
update.py:
import fileinput
import sys
import re
ibrep = re.compile(r"(^.*?[^%\w]getelementptr inbounds )(((?:<\d* x )?)(.*?)(| addrspace\(\d\)) *\*(|>)(?:$| *(?:%|@|null|undef|blockaddress|getelementptr|addrspacecast|bitcast|inttoptr|\[\[[a-zA-Z]|\{\{).*$))")
normrep = re.compile( r"(^.*?[^%\w]getelementptr )(((?:<\d* x )?)(.*?)(| addrspace\(\d\)) *\*(|>)(?:$| *(?:%|@|null|undef|blockaddress|getelementptr|addrspacecast|bitcast|inttoptr|\[\[[a-zA-Z]|\{\{).*$))")
def conv(match, line):
if not match:
return line
line = match.groups()[0]
if len(match.groups()[5]) == 0:
line += match.groups()[2]
line += match.groups()[3]
line += ", "
line += match.groups()[1]
line += "\n"
return line
for line in sys.stdin:
if line.find("getelementptr ") == line.find("getelementptr inbounds"):
if line.find("getelementptr inbounds") != line.find("getelementptr inbounds ("):
line = conv(re.match(ibrep, line), line)
elif line.find("getelementptr ") != line.find("getelementptr ("):
line = conv(re.match(normrep, line), line)
sys.stdout.write(line)
apply.sh:
for name in "$@"
do
python3 `dirname "$0"`/update.py < "$name" > "$name.tmp" && mv "$name.tmp" "$name"
rm -f "$name.tmp"
done
The actual commands:
From llvm/src:
find test/ -name *.ll | xargs ./apply.sh
From llvm/src/tools/clang:
find test/ -name *.mm -o -name *.m -o -name *.cpp -o -name *.c | xargs -I '{}' ../../apply.sh "{}"
From llvm/src/tools/polly:
find test/ -name *.ll | xargs ./apply.sh
After that, check-all (with llvm, clang, clang-tools-extra, lld,
compiler-rt, and polly all checked out).
The extra 'rm' in the apply.sh script is due to a few files in clang's test
suite using interesting unicode stuff that my python script was throwing
exceptions on. None of those files needed to be migrated, so it seemed
sufficient to ignore those cases.
Reviewers: rafael, dexonsmith, grosser
Differential Revision: http://reviews.llvm.org/D7636
llvm-svn: 230786
2015-02-28 03:29:02 +08:00
|
|
|
%arrayidx76 = getelementptr inbounds [776 x float], [776 x float] addrspace(3)* @sgemm.lB, i32 0, i32 %add75
|
2015-02-28 05:17:42 +08:00
|
|
|
%tmp24 = load float, float addrspace(3)* %arrayidx76, align 4
|
2014-10-11 06:01:59 +08:00
|
|
|
%add79 = add nsw i32 %y.i, 65
|
[opaque pointer type] Add textual IR support for explicit type parameter to getelementptr instruction
One of several parallel first steps to remove the target type of pointers,
replacing them with a single opaque pointer type.
This adds an explicit type parameter to the gep instruction so that when the
first parameter becomes an opaque pointer type, the type to gep through is
still available to the instructions.
* This doesn't modify gep operators, only instructions (operators will be
handled separately)
* Textual IR changes only. Bitcode (including upgrade) and changing the
in-memory representation will be in separate changes.
* geps of vectors are transformed as:
getelementptr <4 x float*> %x, ...
->getelementptr float, <4 x float*> %x, ...
Then, once the opaque pointer type is introduced, this will ultimately look
like:
getelementptr float, <4 x ptr> %x
with the unambiguous interpretation that it is a vector of pointers to float.
* address spaces remain on the pointer, not the type:
getelementptr float addrspace(1)* %x
->getelementptr float, float addrspace(1)* %x
Then, eventually:
getelementptr float, ptr addrspace(1) %x
Importantly, the massive amount of test case churn has been automated by
same crappy python code. I had to manually update a few test cases that
wouldn't fit the script's model (r228970,r229196,r229197,r229198). The
python script just massages stdin and writes the result to stdout, I
then wrapped that in a shell script to handle replacing files, then
using the usual find+xargs to migrate all the files.
update.py:
import fileinput
import sys
import re
ibrep = re.compile(r"(^.*?[^%\w]getelementptr inbounds )(((?:<\d* x )?)(.*?)(| addrspace\(\d\)) *\*(|>)(?:$| *(?:%|@|null|undef|blockaddress|getelementptr|addrspacecast|bitcast|inttoptr|\[\[[a-zA-Z]|\{\{).*$))")
normrep = re.compile( r"(^.*?[^%\w]getelementptr )(((?:<\d* x )?)(.*?)(| addrspace\(\d\)) *\*(|>)(?:$| *(?:%|@|null|undef|blockaddress|getelementptr|addrspacecast|bitcast|inttoptr|\[\[[a-zA-Z]|\{\{).*$))")
def conv(match, line):
if not match:
return line
line = match.groups()[0]
if len(match.groups()[5]) == 0:
line += match.groups()[2]
line += match.groups()[3]
line += ", "
line += match.groups()[1]
line += "\n"
return line
for line in sys.stdin:
if line.find("getelementptr ") == line.find("getelementptr inbounds"):
if line.find("getelementptr inbounds") != line.find("getelementptr inbounds ("):
line = conv(re.match(ibrep, line), line)
elif line.find("getelementptr ") != line.find("getelementptr ("):
line = conv(re.match(normrep, line), line)
sys.stdout.write(line)
apply.sh:
for name in "$@"
do
python3 `dirname "$0"`/update.py < "$name" > "$name.tmp" && mv "$name.tmp" "$name"
rm -f "$name.tmp"
done
The actual commands:
From llvm/src:
find test/ -name *.ll | xargs ./apply.sh
From llvm/src/tools/clang:
find test/ -name *.mm -o -name *.m -o -name *.cpp -o -name *.c | xargs -I '{}' ../../apply.sh "{}"
From llvm/src/tools/polly:
find test/ -name *.ll | xargs ./apply.sh
After that, check-all (with llvm, clang, clang-tools-extra, lld,
compiler-rt, and polly all checked out).
The extra 'rm' in the apply.sh script is due to a few files in clang's test
suite using interesting unicode stuff that my python script was throwing
exceptions on. None of those files needed to be migrated, so it seemed
sufficient to ignore those cases.
Reviewers: rafael, dexonsmith, grosser
Differential Revision: http://reviews.llvm.org/D7636
llvm-svn: 230786
2015-02-28 03:29:02 +08:00
|
|
|
%arrayidx80 = getelementptr inbounds [776 x float], [776 x float] addrspace(3)* @sgemm.lB, i32 0, i32 %add79
|
2015-02-28 05:17:42 +08:00
|
|
|
%tmp25 = load float, float addrspace(3)* %arrayidx80, align 4
|
2014-10-11 06:01:59 +08:00
|
|
|
%sum.0 = fadd float %tmp16, %tmp17
|
|
|
|
|
%sum.1 = fadd float %sum.0, %tmp18
|
|
|
|
|
%sum.2 = fadd float %sum.1, %tmp19
|
|
|
|
|
%sum.3 = fadd float %sum.2, %tmp20
|
|
|
|
|
%sum.4 = fadd float %sum.3, %tmp21
|
|
|
|
|
%sum.5 = fadd float %sum.4, %tmp22
|
|
|
|
|
%sum.6 = fadd float %sum.5, %tmp23
|
|
|
|
|
%sum.7 = fadd float %sum.6, %tmp24
|
|
|
|
|
%sum.8 = fadd float %sum.7, %tmp25
|
|
|
|
|
store float %sum.8, float addrspace(1)* %C, align 4
|
|
|
|
|
ret void
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
define void @misaligned_read2_v2i32(<2 x i32> addrspace(1)* %out, <2 x i32> addrspace(3)* %in) #0 {
|
2015-02-28 05:17:42 +08:00
|
|
|
%load = load <2 x i32>, <2 x i32> addrspace(3)* %in, align 4
|
2014-10-11 06:01:59 +08:00
|
|
|
store <2 x i32> %load, <2 x i32> addrspace(1)* %out, align 8
|
|
|
|
|
ret void
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
define void @misaligned_read2_i64(i64 addrspace(1)* %out, i64 addrspace(3)* %in) #0 {
|
2015-02-28 05:17:42 +08:00
|
|
|
%load = load i64, i64 addrspace(3)* %in, align 4
|
2014-10-11 06:01:59 +08:00
|
|
|
store i64 %load, i64 addrspace(1)* %out, align 8
|
|
|
|
|
ret void
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
; Function Attrs: nounwind readnone
|
|
|
|
|
declare i32 @llvm.r600.read.tgid.x() #1
|
|
|
|
|
|
|
|
|
|
; Function Attrs: nounwind readnone
|
|
|
|
|
declare i32 @llvm.r600.read.tgid.y() #1
|
|
|
|
|
|
|
|
|
|
; Function Attrs: nounwind readnone
|
|
|
|
|
declare i32 @llvm.r600.read.tidig.x() #1
|
|
|
|
|
|
|
|
|
|
; Function Attrs: nounwind readnone
|
|
|
|
|
declare i32 @llvm.r600.read.tidig.y() #1
|
|
|
|
|
|
|
|
|
|
; Function Attrs: noduplicate nounwind
|
|
|
|
|
declare void @llvm.AMDGPU.barrier.local() #2
|
|
|
|
|
|
|
|
|
|
attributes #0 = { nounwind "less-precise-fpmad"="false" "no-frame-pointer-elim"="false" "no-infs-fp-math"="false" "no-nans-fp-math"="false" "no-realign-stack" "stack-protector-buffer-size"="8" "unsafe-fp-math"="false" "use-soft-float"="false" }
|
|
|
|
|
attributes #1 = { nounwind readnone }
|
|
|
|
|
attributes #2 = { noduplicate nounwind }
|