2014-09-11 01:58:16 +08:00
|
|
|
; RUN: opt < %s -loop-vectorize -force-vector-interleave=1 -force-vector-width=4 -dce -instcombine -S | FileCheck %s
|
2012-10-20 07:05:40 +08:00
|
|
|
|
|
|
|
|
target datalayout = "e-p:64:64:64-i1:8:8-i8:8:8-i16:16:16-i32:32:32-i64:64:64-f32:32:32-f64:64:64-v64:64:64-v128:128:128-a0:0:64-s0:64:64-f80:128:128-n8:16:32:64-S128"
|
|
|
|
|
|
2013-07-14 09:42:54 +08:00
|
|
|
;CHECK-LABEL: @reduction_sum(
|
2012-10-20 07:05:40 +08:00
|
|
|
;CHECK: phi <4 x i32>
|
|
|
|
|
;CHECK: load <4 x i32>
|
|
|
|
|
;CHECK: add <4 x i32>
|
LoopVectorize: Emit reductions as log2(vectorsize) shuffles + vector ops instead of scalar operations.
For example on x86 with SSE4.2 a <8 x i8> add reduction becomes
movdqa %xmm0, %xmm1
movhlps %xmm1, %xmm1 ## xmm1 = xmm1[1,1]
paddw %xmm0, %xmm1
pshufd $1, %xmm1, %xmm0 ## xmm0 = xmm1[1,0,0,0]
paddw %xmm1, %xmm0
phaddw %xmm0, %xmm0
pextrb $0, %xmm0, %edx
instead of
pextrb $2, %xmm0, %esi
pextrb $0, %xmm0, %edx
addb %sil, %dl
pextrb $4, %xmm0, %esi
addb %dl, %sil
pextrb $6, %xmm0, %edx
addb %sil, %dl
pextrb $8, %xmm0, %esi
addb %dl, %sil
pextrb $10, %xmm0, %edi
pextrb $14, %xmm0, %edx
addb %sil, %dil
pextrb $12, %xmm0, %esi
addb %dil, %sil
addb %sil, %dl
llvm-svn: 170439
2012-12-19 02:40:20 +08:00
|
|
|
;CHECK: shufflevector <4 x i32> %{{.*}}, <4 x i32> undef, <4 x i32> <i32 2, i32 3, i32 undef, i32 undef>
|
|
|
|
|
;CHECK: add <4 x i32>
|
|
|
|
|
;CHECK: shufflevector <4 x i32> %{{.*}}, <4 x i32> undef, <4 x i32> <i32 1, i32 undef, i32 undef, i32 undef>
|
|
|
|
|
;CHECK: add <4 x i32>
|
|
|
|
|
;CHECK: extractelement <4 x i32> %{{.*}}, i32 0
|
2012-10-20 07:05:40 +08:00
|
|
|
;CHECK: ret i32
|
|
|
|
|
define i32 @reduction_sum(i32 %n, i32* noalias nocapture %A, i32* noalias nocapture %B) nounwind uwtable readonly noinline ssp {
|
|
|
|
|
%1 = icmp sgt i32 %n, 0
|
|
|
|
|
br i1 %1, label %.lr.ph, label %._crit_edge
|
|
|
|
|
|
|
|
|
|
.lr.ph: ; preds = %0, %.lr.ph
|
|
|
|
|
%indvars.iv = phi i64 [ %indvars.iv.next, %.lr.ph ], [ 0, %0 ]
|
|
|
|
|
%sum.02 = phi i32 [ %9, %.lr.ph ], [ 0, %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
|
|
|
%2 = getelementptr inbounds i32, i32* %A, i64 %indvars.iv
|
2015-02-28 05:17:42 +08:00
|
|
|
%3 = load i32, i32* %2, 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
|
|
|
%4 = getelementptr inbounds i32, i32* %B, i64 %indvars.iv
|
2015-02-28 05:17:42 +08:00
|
|
|
%5 = load i32, i32* %4, align 4
|
2012-10-20 07:05:40 +08:00
|
|
|
%6 = trunc i64 %indvars.iv to i32
|
|
|
|
|
%7 = add i32 %sum.02, %6
|
|
|
|
|
%8 = add i32 %7, %3
|
|
|
|
|
%9 = add i32 %8, %5
|
|
|
|
|
%indvars.iv.next = add i64 %indvars.iv, 1
|
|
|
|
|
%lftr.wideiv = trunc i64 %indvars.iv.next to i32
|
|
|
|
|
%exitcond = icmp eq i32 %lftr.wideiv, %n
|
|
|
|
|
br i1 %exitcond, label %._crit_edge, label %.lr.ph
|
|
|
|
|
|
|
|
|
|
._crit_edge: ; preds = %.lr.ph, %0
|
|
|
|
|
%sum.0.lcssa = phi i32 [ 0, %0 ], [ %9, %.lr.ph ]
|
|
|
|
|
ret i32 %sum.0.lcssa
|
|
|
|
|
}
|
|
|
|
|
|
2013-07-14 09:42:54 +08:00
|
|
|
;CHECK-LABEL: @reduction_prod(
|
2012-10-20 07:05:40 +08:00
|
|
|
;CHECK: phi <4 x i32>
|
|
|
|
|
;CHECK: load <4 x i32>
|
|
|
|
|
;CHECK: mul <4 x i32>
|
LoopVectorize: Emit reductions as log2(vectorsize) shuffles + vector ops instead of scalar operations.
For example on x86 with SSE4.2 a <8 x i8> add reduction becomes
movdqa %xmm0, %xmm1
movhlps %xmm1, %xmm1 ## xmm1 = xmm1[1,1]
paddw %xmm0, %xmm1
pshufd $1, %xmm1, %xmm0 ## xmm0 = xmm1[1,0,0,0]
paddw %xmm1, %xmm0
phaddw %xmm0, %xmm0
pextrb $0, %xmm0, %edx
instead of
pextrb $2, %xmm0, %esi
pextrb $0, %xmm0, %edx
addb %sil, %dl
pextrb $4, %xmm0, %esi
addb %dl, %sil
pextrb $6, %xmm0, %edx
addb %sil, %dl
pextrb $8, %xmm0, %esi
addb %dl, %sil
pextrb $10, %xmm0, %edi
pextrb $14, %xmm0, %edx
addb %sil, %dil
pextrb $12, %xmm0, %esi
addb %dil, %sil
addb %sil, %dl
llvm-svn: 170439
2012-12-19 02:40:20 +08:00
|
|
|
;CHECK: shufflevector <4 x i32> %{{.*}}, <4 x i32> undef, <4 x i32> <i32 2, i32 3, i32 undef, i32 undef>
|
|
|
|
|
;CHECK: mul <4 x i32>
|
|
|
|
|
;CHECK: shufflevector <4 x i32> %{{.*}}, <4 x i32> undef, <4 x i32> <i32 1, i32 undef, i32 undef, i32 undef>
|
|
|
|
|
;CHECK: mul <4 x i32>
|
|
|
|
|
;CHECK: extractelement <4 x i32> %{{.*}}, i32 0
|
2012-10-20 07:05:40 +08:00
|
|
|
;CHECK: ret i32
|
|
|
|
|
define i32 @reduction_prod(i32 %n, i32* noalias nocapture %A, i32* noalias nocapture %B) nounwind uwtable readonly noinline ssp {
|
|
|
|
|
%1 = icmp sgt i32 %n, 0
|
|
|
|
|
br i1 %1, label %.lr.ph, label %._crit_edge
|
|
|
|
|
|
|
|
|
|
.lr.ph: ; preds = %0, %.lr.ph
|
|
|
|
|
%indvars.iv = phi i64 [ %indvars.iv.next, %.lr.ph ], [ 0, %0 ]
|
|
|
|
|
%prod.02 = phi i32 [ %9, %.lr.ph ], [ 1, %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
|
|
|
%2 = getelementptr inbounds i32, i32* %A, i64 %indvars.iv
|
2015-02-28 05:17:42 +08:00
|
|
|
%3 = load i32, i32* %2, 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
|
|
|
%4 = getelementptr inbounds i32, i32* %B, i64 %indvars.iv
|
2015-02-28 05:17:42 +08:00
|
|
|
%5 = load i32, i32* %4, align 4
|
2012-10-20 07:05:40 +08:00
|
|
|
%6 = trunc i64 %indvars.iv to i32
|
|
|
|
|
%7 = mul i32 %prod.02, %6
|
|
|
|
|
%8 = mul i32 %7, %3
|
|
|
|
|
%9 = mul i32 %8, %5
|
|
|
|
|
%indvars.iv.next = add i64 %indvars.iv, 1
|
|
|
|
|
%lftr.wideiv = trunc i64 %indvars.iv.next to i32
|
|
|
|
|
%exitcond = icmp eq i32 %lftr.wideiv, %n
|
|
|
|
|
br i1 %exitcond, label %._crit_edge, label %.lr.ph
|
|
|
|
|
|
|
|
|
|
._crit_edge: ; preds = %.lr.ph, %0
|
|
|
|
|
%prod.0.lcssa = phi i32 [ 1, %0 ], [ %9, %.lr.ph ]
|
|
|
|
|
ret i32 %prod.0.lcssa
|
|
|
|
|
}
|
|
|
|
|
|
2013-07-14 09:42:54 +08:00
|
|
|
;CHECK-LABEL: @reduction_mix(
|
2012-10-20 07:05:40 +08:00
|
|
|
;CHECK: phi <4 x i32>
|
|
|
|
|
;CHECK: load <4 x i32>
|
2012-11-01 05:40:39 +08:00
|
|
|
;CHECK: mul nsw <4 x i32>
|
LoopVectorize: Emit reductions as log2(vectorsize) shuffles + vector ops instead of scalar operations.
For example on x86 with SSE4.2 a <8 x i8> add reduction becomes
movdqa %xmm0, %xmm1
movhlps %xmm1, %xmm1 ## xmm1 = xmm1[1,1]
paddw %xmm0, %xmm1
pshufd $1, %xmm1, %xmm0 ## xmm0 = xmm1[1,0,0,0]
paddw %xmm1, %xmm0
phaddw %xmm0, %xmm0
pextrb $0, %xmm0, %edx
instead of
pextrb $2, %xmm0, %esi
pextrb $0, %xmm0, %edx
addb %sil, %dl
pextrb $4, %xmm0, %esi
addb %dl, %sil
pextrb $6, %xmm0, %edx
addb %sil, %dl
pextrb $8, %xmm0, %esi
addb %dl, %sil
pextrb $10, %xmm0, %edi
pextrb $14, %xmm0, %edx
addb %sil, %dil
pextrb $12, %xmm0, %esi
addb %dil, %sil
addb %sil, %dl
llvm-svn: 170439
2012-12-19 02:40:20 +08:00
|
|
|
;CHECK: shufflevector <4 x i32> %{{.*}}, <4 x i32> undef, <4 x i32> <i32 2, i32 3, i32 undef, i32 undef>
|
|
|
|
|
;CHECK: add <4 x i32>
|
|
|
|
|
;CHECK: shufflevector <4 x i32> %{{.*}}, <4 x i32> undef, <4 x i32> <i32 1, i32 undef, i32 undef, i32 undef>
|
|
|
|
|
;CHECK: add <4 x i32>
|
|
|
|
|
;CHECK: extractelement <4 x i32> %{{.*}}, i32 0
|
2012-10-20 07:05:40 +08:00
|
|
|
;CHECK: ret i32
|
|
|
|
|
define i32 @reduction_mix(i32 %n, i32* noalias nocapture %A, i32* noalias nocapture %B) nounwind uwtable readonly noinline ssp {
|
|
|
|
|
%1 = icmp sgt i32 %n, 0
|
|
|
|
|
br i1 %1, label %.lr.ph, label %._crit_edge
|
|
|
|
|
|
|
|
|
|
.lr.ph: ; preds = %0, %.lr.ph
|
|
|
|
|
%indvars.iv = phi i64 [ %indvars.iv.next, %.lr.ph ], [ 0, %0 ]
|
|
|
|
|
%sum.02 = phi i32 [ %9, %.lr.ph ], [ 0, %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
|
|
|
%2 = getelementptr inbounds i32, i32* %A, i64 %indvars.iv
|
2015-02-28 05:17:42 +08:00
|
|
|
%3 = load i32, i32* %2, 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
|
|
|
%4 = getelementptr inbounds i32, i32* %B, i64 %indvars.iv
|
2015-02-28 05:17:42 +08:00
|
|
|
%5 = load i32, i32* %4, align 4
|
2012-10-20 07:05:40 +08:00
|
|
|
%6 = mul nsw i32 %5, %3
|
|
|
|
|
%7 = trunc i64 %indvars.iv to i32
|
|
|
|
|
%8 = add i32 %sum.02, %7
|
|
|
|
|
%9 = add i32 %8, %6
|
|
|
|
|
%indvars.iv.next = add i64 %indvars.iv, 1
|
|
|
|
|
%lftr.wideiv = trunc i64 %indvars.iv.next to i32
|
|
|
|
|
%exitcond = icmp eq i32 %lftr.wideiv, %n
|
|
|
|
|
br i1 %exitcond, label %._crit_edge, label %.lr.ph
|
|
|
|
|
|
|
|
|
|
._crit_edge: ; preds = %.lr.ph, %0
|
|
|
|
|
%sum.0.lcssa = phi i32 [ 0, %0 ], [ %9, %.lr.ph ]
|
|
|
|
|
ret i32 %sum.0.lcssa
|
|
|
|
|
}
|
|
|
|
|
|
2013-07-14 09:42:54 +08:00
|
|
|
;CHECK-LABEL: @reduction_mul(
|
2012-10-21 13:52:51 +08:00
|
|
|
;CHECK: mul <4 x i32>
|
LoopVectorize: Emit reductions as log2(vectorsize) shuffles + vector ops instead of scalar operations.
For example on x86 with SSE4.2 a <8 x i8> add reduction becomes
movdqa %xmm0, %xmm1
movhlps %xmm1, %xmm1 ## xmm1 = xmm1[1,1]
paddw %xmm0, %xmm1
pshufd $1, %xmm1, %xmm0 ## xmm0 = xmm1[1,0,0,0]
paddw %xmm1, %xmm0
phaddw %xmm0, %xmm0
pextrb $0, %xmm0, %edx
instead of
pextrb $2, %xmm0, %esi
pextrb $0, %xmm0, %edx
addb %sil, %dl
pextrb $4, %xmm0, %esi
addb %dl, %sil
pextrb $6, %xmm0, %edx
addb %sil, %dl
pextrb $8, %xmm0, %esi
addb %dl, %sil
pextrb $10, %xmm0, %edi
pextrb $14, %xmm0, %edx
addb %sil, %dil
pextrb $12, %xmm0, %esi
addb %dil, %sil
addb %sil, %dl
llvm-svn: 170439
2012-12-19 02:40:20 +08:00
|
|
|
;CHECK: shufflevector <4 x i32> %{{.*}}, <4 x i32> undef, <4 x i32> <i32 2, i32 3, i32 undef, i32 undef>
|
|
|
|
|
;CHECK: mul <4 x i32>
|
|
|
|
|
;CHECK: shufflevector <4 x i32> %{{.*}}, <4 x i32> undef, <4 x i32> <i32 1, i32 undef, i32 undef, i32 undef>
|
|
|
|
|
;CHECK: mul <4 x i32>
|
|
|
|
|
;CHECK: extractelement <4 x i32> %{{.*}}, i32 0
|
2012-10-20 07:05:40 +08:00
|
|
|
;CHECK: ret i32
|
2012-10-21 13:52:51 +08:00
|
|
|
define i32 @reduction_mul(i32 %n, i32* noalias nocapture %A, i32* noalias nocapture %B) nounwind uwtable readonly noinline ssp {
|
2012-10-20 07:05:40 +08:00
|
|
|
%1 = icmp sgt i32 %n, 0
|
|
|
|
|
br i1 %1, label %.lr.ph, label %._crit_edge
|
|
|
|
|
|
|
|
|
|
.lr.ph: ; preds = %0, %.lr.ph
|
|
|
|
|
%indvars.iv = phi i64 [ %indvars.iv.next, %.lr.ph ], [ 0, %0 ]
|
2012-10-21 13:52:51 +08:00
|
|
|
%sum.02 = phi i32 [ %9, %.lr.ph ], [ 19, %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
|
|
|
%2 = getelementptr inbounds i32, i32* %A, i64 %indvars.iv
|
2015-02-28 05:17:42 +08:00
|
|
|
%3 = load i32, i32* %2, 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
|
|
|
%4 = getelementptr inbounds i32, i32* %B, i64 %indvars.iv
|
2015-02-28 05:17:42 +08:00
|
|
|
%5 = load i32, i32* %4, align 4
|
2012-10-20 07:05:40 +08:00
|
|
|
%6 = trunc i64 %indvars.iv to i32
|
|
|
|
|
%7 = add i32 %3, %6
|
|
|
|
|
%8 = add i32 %7, %5
|
|
|
|
|
%9 = mul i32 %8, %sum.02
|
|
|
|
|
%indvars.iv.next = add i64 %indvars.iv, 1
|
|
|
|
|
%lftr.wideiv = trunc i64 %indvars.iv.next to i32
|
|
|
|
|
%exitcond = icmp eq i32 %lftr.wideiv, %n
|
|
|
|
|
br i1 %exitcond, label %._crit_edge, label %.lr.ph
|
|
|
|
|
|
|
|
|
|
._crit_edge: ; preds = %.lr.ph, %0
|
|
|
|
|
%sum.0.lcssa = phi i32 [ 0, %0 ], [ %9, %.lr.ph ]
|
|
|
|
|
ret i32 %sum.0.lcssa
|
|
|
|
|
}
|
2012-10-21 13:52:51 +08:00
|
|
|
|
2013-07-14 09:42:54 +08:00
|
|
|
;CHECK-LABEL: @start_at_non_zero(
|
2012-10-21 13:52:51 +08:00
|
|
|
;CHECK: phi <4 x i32>
|
|
|
|
|
;CHECK: <i32 120, i32 0, i32 0, i32 0>
|
LoopVectorize: Emit reductions as log2(vectorsize) shuffles + vector ops instead of scalar operations.
For example on x86 with SSE4.2 a <8 x i8> add reduction becomes
movdqa %xmm0, %xmm1
movhlps %xmm1, %xmm1 ## xmm1 = xmm1[1,1]
paddw %xmm0, %xmm1
pshufd $1, %xmm1, %xmm0 ## xmm0 = xmm1[1,0,0,0]
paddw %xmm1, %xmm0
phaddw %xmm0, %xmm0
pextrb $0, %xmm0, %edx
instead of
pextrb $2, %xmm0, %esi
pextrb $0, %xmm0, %edx
addb %sil, %dl
pextrb $4, %xmm0, %esi
addb %dl, %sil
pextrb $6, %xmm0, %edx
addb %sil, %dl
pextrb $8, %xmm0, %esi
addb %dl, %sil
pextrb $10, %xmm0, %edi
pextrb $14, %xmm0, %edx
addb %sil, %dil
pextrb $12, %xmm0, %esi
addb %dil, %sil
addb %sil, %dl
llvm-svn: 170439
2012-12-19 02:40:20 +08:00
|
|
|
;CHECK: shufflevector <4 x i32> %{{.*}}, <4 x i32> undef, <4 x i32> <i32 2, i32 3, i32 undef, i32 undef>
|
|
|
|
|
;CHECK: add <4 x i32>
|
|
|
|
|
;CHECK: shufflevector <4 x i32> %{{.*}}, <4 x i32> undef, <4 x i32> <i32 1, i32 undef, i32 undef, i32 undef>
|
|
|
|
|
;CHECK: add <4 x i32>
|
|
|
|
|
;CHECK: extractelement <4 x i32> %{{.*}}, i32 0
|
2012-10-21 13:52:51 +08:00
|
|
|
;CHECK: ret i32
|
|
|
|
|
define i32 @start_at_non_zero(i32* nocapture %in, i32* nocapture %coeff, i32* nocapture %out, i32 %n) nounwind uwtable readonly ssp {
|
|
|
|
|
entry:
|
|
|
|
|
%cmp7 = icmp sgt i32 %n, 0
|
|
|
|
|
br i1 %cmp7, label %for.body, label %for.end
|
|
|
|
|
|
|
|
|
|
for.body: ; preds = %entry, %for.body
|
|
|
|
|
%indvars.iv = phi i64 [ %indvars.iv.next, %for.body ], [ 0, %entry ]
|
|
|
|
|
%sum.09 = phi i32 [ %add, %for.body ], [ 120, %entry ]
|
[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
|
|
|
%arrayidx = getelementptr inbounds i32, i32* %in, i64 %indvars.iv
|
2015-02-28 05:17:42 +08:00
|
|
|
%0 = load i32, i32* %arrayidx, 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
|
|
|
%arrayidx2 = getelementptr inbounds i32, i32* %coeff, i64 %indvars.iv
|
2015-02-28 05:17:42 +08:00
|
|
|
%1 = load i32, i32* %arrayidx2, align 4
|
2012-10-21 13:52:51 +08:00
|
|
|
%mul = mul nsw i32 %1, %0
|
|
|
|
|
%add = add nsw i32 %mul, %sum.09
|
|
|
|
|
%indvars.iv.next = add i64 %indvars.iv, 1
|
|
|
|
|
%lftr.wideiv = trunc i64 %indvars.iv.next to i32
|
|
|
|
|
%exitcond = icmp eq i32 %lftr.wideiv, %n
|
|
|
|
|
br i1 %exitcond, label %for.end, label %for.body
|
|
|
|
|
|
|
|
|
|
for.end: ; preds = %for.body, %entry
|
|
|
|
|
%sum.0.lcssa = phi i32 [ 120, %entry ], [ %add, %for.body ]
|
|
|
|
|
ret i32 %sum.0.lcssa
|
|
|
|
|
}
|
|
|
|
|
|
2013-07-14 09:42:54 +08:00
|
|
|
;CHECK-LABEL: @reduction_and(
|
2012-10-31 02:12:36 +08:00
|
|
|
;CHECK: <i32 -1, i32 -1, i32 -1, i32 -1>
|
2015-09-02 18:15:39 +08:00
|
|
|
;CHECK: and <4 x i32>
|
LoopVectorize: Emit reductions as log2(vectorsize) shuffles + vector ops instead of scalar operations.
For example on x86 with SSE4.2 a <8 x i8> add reduction becomes
movdqa %xmm0, %xmm1
movhlps %xmm1, %xmm1 ## xmm1 = xmm1[1,1]
paddw %xmm0, %xmm1
pshufd $1, %xmm1, %xmm0 ## xmm0 = xmm1[1,0,0,0]
paddw %xmm1, %xmm0
phaddw %xmm0, %xmm0
pextrb $0, %xmm0, %edx
instead of
pextrb $2, %xmm0, %esi
pextrb $0, %xmm0, %edx
addb %sil, %dl
pextrb $4, %xmm0, %esi
addb %dl, %sil
pextrb $6, %xmm0, %edx
addb %sil, %dl
pextrb $8, %xmm0, %esi
addb %dl, %sil
pextrb $10, %xmm0, %edi
pextrb $14, %xmm0, %edx
addb %sil, %dil
pextrb $12, %xmm0, %esi
addb %dil, %sil
addb %sil, %dl
llvm-svn: 170439
2012-12-19 02:40:20 +08:00
|
|
|
;CHECK: shufflevector <4 x i32> %{{.*}}, <4 x i32> undef, <4 x i32> <i32 2, i32 3, i32 undef, i32 undef>
|
|
|
|
|
;CHECK: and <4 x i32>
|
|
|
|
|
;CHECK: shufflevector <4 x i32> %{{.*}}, <4 x i32> undef, <4 x i32> <i32 1, i32 undef, i32 undef, i32 undef>
|
|
|
|
|
;CHECK: and <4 x i32>
|
|
|
|
|
;CHECK: extractelement <4 x i32> %{{.*}}, i32 0
|
2012-10-25 08:08:41 +08:00
|
|
|
;CHECK: ret i32
|
|
|
|
|
define i32 @reduction_and(i32 %n, i32* nocapture %A, i32* nocapture %B) nounwind uwtable readonly {
|
|
|
|
|
entry:
|
|
|
|
|
%cmp7 = icmp sgt i32 %n, 0
|
|
|
|
|
br i1 %cmp7, label %for.body, label %for.end
|
|
|
|
|
|
|
|
|
|
for.body: ; preds = %entry, %for.body
|
|
|
|
|
%indvars.iv = phi i64 [ %indvars.iv.next, %for.body ], [ 0, %entry ]
|
|
|
|
|
%result.08 = phi i32 [ %and, %for.body ], [ -1, %entry ]
|
[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
|
|
|
%arrayidx = getelementptr inbounds i32, i32* %A, i64 %indvars.iv
|
2015-02-28 05:17:42 +08:00
|
|
|
%0 = load i32, i32* %arrayidx, 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
|
|
|
%arrayidx2 = getelementptr inbounds i32, i32* %B, i64 %indvars.iv
|
2015-02-28 05:17:42 +08:00
|
|
|
%1 = load i32, i32* %arrayidx2, align 4
|
2012-10-25 08:08:41 +08:00
|
|
|
%add = add nsw i32 %1, %0
|
|
|
|
|
%and = and i32 %add, %result.08
|
|
|
|
|
%indvars.iv.next = add i64 %indvars.iv, 1
|
|
|
|
|
%lftr.wideiv = trunc i64 %indvars.iv.next to i32
|
|
|
|
|
%exitcond = icmp eq i32 %lftr.wideiv, %n
|
|
|
|
|
br i1 %exitcond, label %for.end, label %for.body
|
2012-10-21 13:52:51 +08:00
|
|
|
|
2012-10-25 08:08:41 +08:00
|
|
|
for.end: ; preds = %for.body, %entry
|
|
|
|
|
%result.0.lcssa = phi i32 [ -1, %entry ], [ %and, %for.body ]
|
|
|
|
|
ret i32 %result.0.lcssa
|
|
|
|
|
}
|
|
|
|
|
|
2013-07-14 09:42:54 +08:00
|
|
|
;CHECK-LABEL: @reduction_or(
|
2012-10-25 08:08:41 +08:00
|
|
|
;CHECK: or <4 x i32>
|
LoopVectorize: Emit reductions as log2(vectorsize) shuffles + vector ops instead of scalar operations.
For example on x86 with SSE4.2 a <8 x i8> add reduction becomes
movdqa %xmm0, %xmm1
movhlps %xmm1, %xmm1 ## xmm1 = xmm1[1,1]
paddw %xmm0, %xmm1
pshufd $1, %xmm1, %xmm0 ## xmm0 = xmm1[1,0,0,0]
paddw %xmm1, %xmm0
phaddw %xmm0, %xmm0
pextrb $0, %xmm0, %edx
instead of
pextrb $2, %xmm0, %esi
pextrb $0, %xmm0, %edx
addb %sil, %dl
pextrb $4, %xmm0, %esi
addb %dl, %sil
pextrb $6, %xmm0, %edx
addb %sil, %dl
pextrb $8, %xmm0, %esi
addb %dl, %sil
pextrb $10, %xmm0, %edi
pextrb $14, %xmm0, %edx
addb %sil, %dil
pextrb $12, %xmm0, %esi
addb %dil, %sil
addb %sil, %dl
llvm-svn: 170439
2012-12-19 02:40:20 +08:00
|
|
|
;CHECK: shufflevector <4 x i32> %{{.*}}, <4 x i32> undef, <4 x i32> <i32 2, i32 3, i32 undef, i32 undef>
|
|
|
|
|
;CHECK: or <4 x i32>
|
|
|
|
|
;CHECK: shufflevector <4 x i32> %{{.*}}, <4 x i32> undef, <4 x i32> <i32 1, i32 undef, i32 undef, i32 undef>
|
|
|
|
|
;CHECK: or <4 x i32>
|
|
|
|
|
;CHECK: extractelement <4 x i32> %{{.*}}, i32 0
|
2012-10-25 08:08:41 +08:00
|
|
|
;CHECK: ret i32
|
|
|
|
|
define i32 @reduction_or(i32 %n, i32* nocapture %A, i32* nocapture %B) nounwind uwtable readonly {
|
|
|
|
|
entry:
|
|
|
|
|
%cmp7 = icmp sgt i32 %n, 0
|
|
|
|
|
br i1 %cmp7, label %for.body, label %for.end
|
|
|
|
|
|
|
|
|
|
for.body: ; preds = %entry, %for.body
|
|
|
|
|
%indvars.iv = phi i64 [ %indvars.iv.next, %for.body ], [ 0, %entry ]
|
|
|
|
|
%result.08 = phi i32 [ %or, %for.body ], [ 0, %entry ]
|
[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
|
|
|
%arrayidx = getelementptr inbounds i32, i32* %A, i64 %indvars.iv
|
2015-02-28 05:17:42 +08:00
|
|
|
%0 = load i32, i32* %arrayidx, 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
|
|
|
%arrayidx2 = getelementptr inbounds i32, i32* %B, i64 %indvars.iv
|
2015-02-28 05:17:42 +08:00
|
|
|
%1 = load i32, i32* %arrayidx2, align 4
|
2012-10-25 08:08:41 +08:00
|
|
|
%add = add nsw i32 %1, %0
|
|
|
|
|
%or = or i32 %add, %result.08
|
|
|
|
|
%indvars.iv.next = add i64 %indvars.iv, 1
|
|
|
|
|
%lftr.wideiv = trunc i64 %indvars.iv.next to i32
|
|
|
|
|
%exitcond = icmp eq i32 %lftr.wideiv, %n
|
|
|
|
|
br i1 %exitcond, label %for.end, label %for.body
|
|
|
|
|
|
|
|
|
|
for.end: ; preds = %for.body, %entry
|
|
|
|
|
%result.0.lcssa = phi i32 [ 0, %entry ], [ %or, %for.body ]
|
|
|
|
|
ret i32 %result.0.lcssa
|
|
|
|
|
}
|
|
|
|
|
|
2013-07-14 09:42:54 +08:00
|
|
|
;CHECK-LABEL: @reduction_xor(
|
2012-10-25 08:08:41 +08:00
|
|
|
;CHECK: xor <4 x i32>
|
LoopVectorize: Emit reductions as log2(vectorsize) shuffles + vector ops instead of scalar operations.
For example on x86 with SSE4.2 a <8 x i8> add reduction becomes
movdqa %xmm0, %xmm1
movhlps %xmm1, %xmm1 ## xmm1 = xmm1[1,1]
paddw %xmm0, %xmm1
pshufd $1, %xmm1, %xmm0 ## xmm0 = xmm1[1,0,0,0]
paddw %xmm1, %xmm0
phaddw %xmm0, %xmm0
pextrb $0, %xmm0, %edx
instead of
pextrb $2, %xmm0, %esi
pextrb $0, %xmm0, %edx
addb %sil, %dl
pextrb $4, %xmm0, %esi
addb %dl, %sil
pextrb $6, %xmm0, %edx
addb %sil, %dl
pextrb $8, %xmm0, %esi
addb %dl, %sil
pextrb $10, %xmm0, %edi
pextrb $14, %xmm0, %edx
addb %sil, %dil
pextrb $12, %xmm0, %esi
addb %dil, %sil
addb %sil, %dl
llvm-svn: 170439
2012-12-19 02:40:20 +08:00
|
|
|
;CHECK: shufflevector <4 x i32> %{{.*}}, <4 x i32> undef, <4 x i32> <i32 2, i32 3, i32 undef, i32 undef>
|
|
|
|
|
;CHECK: xor <4 x i32>
|
|
|
|
|
;CHECK: shufflevector <4 x i32> %{{.*}}, <4 x i32> undef, <4 x i32> <i32 1, i32 undef, i32 undef, i32 undef>
|
|
|
|
|
;CHECK: xor <4 x i32>
|
|
|
|
|
;CHECK: extractelement <4 x i32> %{{.*}}, i32 0
|
2012-10-25 08:08:41 +08:00
|
|
|
;CHECK: ret i32
|
|
|
|
|
define i32 @reduction_xor(i32 %n, i32* nocapture %A, i32* nocapture %B) nounwind uwtable readonly {
|
|
|
|
|
entry:
|
|
|
|
|
%cmp7 = icmp sgt i32 %n, 0
|
|
|
|
|
br i1 %cmp7, label %for.body, label %for.end
|
|
|
|
|
|
|
|
|
|
for.body: ; preds = %entry, %for.body
|
|
|
|
|
%indvars.iv = phi i64 [ %indvars.iv.next, %for.body ], [ 0, %entry ]
|
|
|
|
|
%result.08 = phi i32 [ %xor, %for.body ], [ 0, %entry ]
|
[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
|
|
|
%arrayidx = getelementptr inbounds i32, i32* %A, i64 %indvars.iv
|
2015-02-28 05:17:42 +08:00
|
|
|
%0 = load i32, i32* %arrayidx, 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
|
|
|
%arrayidx2 = getelementptr inbounds i32, i32* %B, i64 %indvars.iv
|
2015-02-28 05:17:42 +08:00
|
|
|
%1 = load i32, i32* %arrayidx2, align 4
|
2012-10-25 08:08:41 +08:00
|
|
|
%add = add nsw i32 %1, %0
|
|
|
|
|
%xor = xor i32 %add, %result.08
|
|
|
|
|
%indvars.iv.next = add i64 %indvars.iv, 1
|
|
|
|
|
%lftr.wideiv = trunc i64 %indvars.iv.next to i32
|
|
|
|
|
%exitcond = icmp eq i32 %lftr.wideiv, %n
|
|
|
|
|
br i1 %exitcond, label %for.end, label %for.body
|
|
|
|
|
|
|
|
|
|
for.end: ; preds = %for.body, %entry
|
|
|
|
|
%result.0.lcssa = phi i32 [ 0, %entry ], [ %xor, %for.body ]
|
|
|
|
|
ret i32 %result.0.lcssa
|
|
|
|
|
}
|
2013-01-05 06:10:16 +08:00
|
|
|
|
2013-01-05 09:15:47 +08:00
|
|
|
; In this code the subtracted variable is on the RHS and this is not an induction variable.
|
2013-07-14 09:42:54 +08:00
|
|
|
;CHECK-LABEL: @reduction_sub_rhs(
|
2013-01-05 06:10:16 +08:00
|
|
|
;CHECK-NOT: phi <4 x i32>
|
|
|
|
|
;CHECK-NOT: sub nsw <4 x i32>
|
|
|
|
|
;CHECK: ret i32
|
2013-01-05 09:15:47 +08:00
|
|
|
define i32 @reduction_sub_rhs(i32 %n, i32* noalias nocapture %A) nounwind uwtable readonly {
|
2013-01-05 06:10:16 +08:00
|
|
|
entry:
|
|
|
|
|
%cmp4 = icmp sgt i32 %n, 0
|
|
|
|
|
br i1 %cmp4, label %for.body, label %for.end
|
|
|
|
|
|
|
|
|
|
for.body: ; preds = %entry, %for.body
|
|
|
|
|
%indvars.iv = phi i64 [ %indvars.iv.next, %for.body ], [ 0, %entry ]
|
|
|
|
|
%x.05 = phi i32 [ %sub, %for.body ], [ 0, %entry ]
|
[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
|
|
|
%arrayidx = getelementptr inbounds i32, i32* %A, i64 %indvars.iv
|
2015-02-28 05:17:42 +08:00
|
|
|
%0 = load i32, i32* %arrayidx, align 4
|
2013-01-05 06:10:16 +08:00
|
|
|
%sub = sub nsw i32 %0, %x.05
|
|
|
|
|
%indvars.iv.next = add i64 %indvars.iv, 1
|
|
|
|
|
%lftr.wideiv = trunc i64 %indvars.iv.next to i32
|
|
|
|
|
%exitcond = icmp eq i32 %lftr.wideiv, %n
|
|
|
|
|
br i1 %exitcond, label %for.end, label %for.body
|
|
|
|
|
|
|
|
|
|
for.end: ; preds = %for.body, %entry
|
|
|
|
|
%x.0.lcssa = phi i32 [ 0, %entry ], [ %sub, %for.body ]
|
|
|
|
|
ret i32 %x.0.lcssa
|
|
|
|
|
}
|
2013-01-05 09:15:47 +08:00
|
|
|
|
|
|
|
|
|
|
|
|
|
; In this test the reduction variable is on the LHS and we can vectorize it.
|
2013-07-14 09:42:54 +08:00
|
|
|
;CHECK-LABEL: @reduction_sub_lhs(
|
2013-01-05 09:15:47 +08:00
|
|
|
;CHECK: phi <4 x i32>
|
|
|
|
|
;CHECK: sub nsw <4 x i32>
|
|
|
|
|
;CHECK: ret i32
|
|
|
|
|
define i32 @reduction_sub_lhs(i32 %n, i32* noalias nocapture %A) nounwind uwtable readonly {
|
|
|
|
|
entry:
|
|
|
|
|
%cmp4 = icmp sgt i32 %n, 0
|
|
|
|
|
br i1 %cmp4, label %for.body, label %for.end
|
|
|
|
|
|
|
|
|
|
for.body: ; preds = %entry, %for.body
|
|
|
|
|
%indvars.iv = phi i64 [ %indvars.iv.next, %for.body ], [ 0, %entry ]
|
|
|
|
|
%x.05 = phi i32 [ %sub, %for.body ], [ 0, %entry ]
|
[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
|
|
|
%arrayidx = getelementptr inbounds i32, i32* %A, i64 %indvars.iv
|
2015-02-28 05:17:42 +08:00
|
|
|
%0 = load i32, i32* %arrayidx, align 4
|
2013-01-05 09:15:47 +08:00
|
|
|
%sub = sub nsw i32 %x.05, %0
|
|
|
|
|
%indvars.iv.next = add i64 %indvars.iv, 1
|
|
|
|
|
%lftr.wideiv = trunc i64 %indvars.iv.next to i32
|
|
|
|
|
%exitcond = icmp eq i32 %lftr.wideiv, %n
|
|
|
|
|
br i1 %exitcond, label %for.end, label %for.body
|
|
|
|
|
|
|
|
|
|
for.end: ; preds = %for.body, %entry
|
|
|
|
|
%x.0.lcssa = phi i32 [ 0, %entry ], [ %sub, %for.body ]
|
|
|
|
|
ret i32 %x.0.lcssa
|
|
|
|
|
}
|
2013-05-08 05:55:37 +08:00
|
|
|
|
|
|
|
|
; We can vectorize conditional reductions with multi-input phis.
|
|
|
|
|
; CHECK: reduction_conditional
|
|
|
|
|
; CHECK: fadd <4 x float>
|
|
|
|
|
|
|
|
|
|
define float @reduction_conditional(float* %A, float* %B, float* %C, float %S) {
|
|
|
|
|
entry:
|
|
|
|
|
br label %for.body
|
|
|
|
|
|
|
|
|
|
for.body:
|
|
|
|
|
%indvars.iv = phi i64 [ 0, %entry ], [ %indvars.iv.next, %for.inc ]
|
|
|
|
|
%sum.033 = phi float [ %S, %entry ], [ %sum.1, %for.inc ]
|
[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
|
|
|
%arrayidx = getelementptr inbounds float, float* %A, i64 %indvars.iv
|
2015-02-28 05:17:42 +08:00
|
|
|
%0 = load float, float* %arrayidx, 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
|
|
|
%arrayidx2 = getelementptr inbounds float, float* %B, i64 %indvars.iv
|
2015-02-28 05:17:42 +08:00
|
|
|
%1 = load float, float* %arrayidx2, align 4
|
2013-05-08 05:55:37 +08:00
|
|
|
%cmp3 = fcmp ogt float %0, %1
|
|
|
|
|
br i1 %cmp3, label %if.then, label %for.inc
|
|
|
|
|
|
|
|
|
|
if.then:
|
|
|
|
|
%cmp6 = fcmp ogt float %1, 1.000000e+00
|
|
|
|
|
br i1 %cmp6, label %if.then8, label %if.else
|
|
|
|
|
|
|
|
|
|
if.then8:
|
|
|
|
|
%add = fadd fast float %sum.033, %0
|
|
|
|
|
br label %for.inc
|
|
|
|
|
|
|
|
|
|
if.else:
|
|
|
|
|
%cmp14 = fcmp ogt float %0, 2.000000e+00
|
|
|
|
|
br i1 %cmp14, label %if.then16, label %for.inc
|
|
|
|
|
|
|
|
|
|
if.then16:
|
|
|
|
|
%add19 = fadd fast float %sum.033, %1
|
|
|
|
|
br label %for.inc
|
|
|
|
|
|
|
|
|
|
for.inc:
|
|
|
|
|
%sum.1 = phi float [ %add, %if.then8 ], [ %add19, %if.then16 ], [ %sum.033, %if.else ], [ %sum.033, %for.body ]
|
|
|
|
|
%indvars.iv.next = add i64 %indvars.iv, 1
|
|
|
|
|
%lftr.wideiv = trunc i64 %indvars.iv.next to i32
|
|
|
|
|
%exitcond = icmp ne i32 %lftr.wideiv, 128
|
|
|
|
|
br i1 %exitcond, label %for.body, label %for.end
|
|
|
|
|
|
|
|
|
|
for.end:
|
|
|
|
|
%sum.1.lcssa = phi float [ %sum.1, %for.inc ]
|
|
|
|
|
ret float %sum.1.lcssa
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
; We can't vectorize reductions with phi inputs from outside the reduction.
|
|
|
|
|
; CHECK: noreduction_phi
|
|
|
|
|
; CHECK-NOT: fadd <4 x float>
|
|
|
|
|
define float @noreduction_phi(float* %A, float* %B, float* %C, float %S) {
|
|
|
|
|
entry:
|
|
|
|
|
br label %for.body
|
|
|
|
|
|
|
|
|
|
for.body:
|
|
|
|
|
%indvars.iv = phi i64 [ 0, %entry ], [ %indvars.iv.next, %for.inc ]
|
|
|
|
|
%sum.033 = phi float [ %S, %entry ], [ %sum.1, %for.inc ]
|
[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
|
|
|
%arrayidx = getelementptr inbounds float, float* %A, i64 %indvars.iv
|
2015-02-28 05:17:42 +08:00
|
|
|
%0 = load float, float* %arrayidx, 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
|
|
|
%arrayidx2 = getelementptr inbounds float, float* %B, i64 %indvars.iv
|
2015-02-28 05:17:42 +08:00
|
|
|
%1 = load float, float* %arrayidx2, align 4
|
2013-05-08 05:55:37 +08:00
|
|
|
%cmp3 = fcmp ogt float %0, %1
|
|
|
|
|
br i1 %cmp3, label %if.then, label %for.inc
|
|
|
|
|
|
|
|
|
|
if.then:
|
|
|
|
|
%cmp6 = fcmp ogt float %1, 1.000000e+00
|
|
|
|
|
br i1 %cmp6, label %if.then8, label %if.else
|
|
|
|
|
|
|
|
|
|
if.then8:
|
|
|
|
|
%add = fadd fast float %sum.033, %0
|
|
|
|
|
br label %for.inc
|
|
|
|
|
|
|
|
|
|
if.else:
|
|
|
|
|
%cmp14 = fcmp ogt float %0, 2.000000e+00
|
|
|
|
|
br i1 %cmp14, label %if.then16, label %for.inc
|
|
|
|
|
|
|
|
|
|
if.then16:
|
|
|
|
|
%add19 = fadd fast float %sum.033, %1
|
|
|
|
|
br label %for.inc
|
|
|
|
|
|
|
|
|
|
for.inc:
|
|
|
|
|
%sum.1 = phi float [ %add, %if.then8 ], [ %add19, %if.then16 ], [ 0.000000e+00, %if.else ], [ %sum.033, %for.body ]
|
|
|
|
|
%indvars.iv.next = add i64 %indvars.iv, 1
|
|
|
|
|
%lftr.wideiv = trunc i64 %indvars.iv.next to i32
|
|
|
|
|
%exitcond = icmp ne i32 %lftr.wideiv, 128
|
|
|
|
|
br i1 %exitcond, label %for.body, label %for.end
|
|
|
|
|
|
|
|
|
|
for.end:
|
|
|
|
|
%sum.1.lcssa = phi float [ %sum.1, %for.inc ]
|
|
|
|
|
ret float %sum.1.lcssa
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
; We can't vectorize reductions that feed another header PHI.
|
|
|
|
|
; CHECK: noredux_header_phi
|
|
|
|
|
; CHECK-NOT: fadd <4 x float>
|
|
|
|
|
|
|
|
|
|
define float @noredux_header_phi(float* %A, float* %B, float* %C, float %S) {
|
|
|
|
|
entry:
|
|
|
|
|
br label %for.body
|
|
|
|
|
|
|
|
|
|
for.body:
|
|
|
|
|
%indvars.iv = phi i64 [ 0, %entry ], [ %indvars.iv.next, %for.body ]
|
|
|
|
|
%sum2.09 = phi float [ 0.000000e+00, %entry ], [ %add1, %for.body ]
|
|
|
|
|
%sum.08 = phi float [ %S, %entry ], [ %add, %for.body ]
|
[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
|
|
|
%arrayidx = getelementptr inbounds float, float* %B, i64 %indvars.iv
|
2015-02-28 05:17:42 +08:00
|
|
|
%0 = load float, float* %arrayidx, align 4
|
2013-05-08 05:55:37 +08:00
|
|
|
%add = fadd fast float %sum.08, %0
|
|
|
|
|
%add1 = fadd fast float %sum2.09, %add
|
|
|
|
|
%indvars.iv.next = add i64 %indvars.iv, 1
|
|
|
|
|
%lftr.wideiv = trunc i64 %indvars.iv.next to i32
|
|
|
|
|
%exitcond = icmp ne i32 %lftr.wideiv, 128
|
|
|
|
|
br i1 %exitcond, label %for.body, label %for.end
|
|
|
|
|
|
|
|
|
|
for.end:
|
|
|
|
|
%add1.lcssa = phi float [ %add1, %for.body ]
|
|
|
|
|
%add.lcssa = phi float [ %add, %for.body ]
|
|
|
|
|
%add2 = fadd fast float %add.lcssa, %add1.lcssa
|
|
|
|
|
ret float %add2
|
|
|
|
|
}
|
LoopVectorizer: Disallow reductions whose header phi is used outside the loop
If an outside loop user of the reduction value uses the header phi node we
cannot just reduce the vectorized phi value in the vector code epilog because
we would loose VF-1 reductions.
lp:
p = phi (0, lv)
lv = lv + 1
...
brcond , lp, outside
outside:
usr = add 0, p
(Say the loop iterates two times, the value of p coming out of the loop is one).
We cannot just transform this to:
vlp:
p = phi (<0,0>, lv)
lv = lv + <1,1>
..
brcond , lp, outside
outside:
p_reduced = p[0] + [1];
usr = add 0, p_reduced
(Because the original loop iterated two times the vectorized loop would iterate
one time, but p_reduced ends up being zero instead of one).
We would have to execute VF-1 iterations in the scalar remainder loop in such
cases. For now, just disable vectorization.
PR16522
llvm-svn: 186256
2013-07-14 03:09:29 +08:00
|
|
|
|
|
|
|
|
|
|
|
|
|
; When vectorizing a reduction whose loop header phi value is used outside the
|
|
|
|
|
; loop special care must be taken. Otherwise, the reduced value feeding into the
|
|
|
|
|
; outside user misses a few iterations (VF-1) of the loop.
|
|
|
|
|
; PR16522
|
|
|
|
|
|
2013-07-14 09:42:54 +08:00
|
|
|
; CHECK-LABEL: @phivalueredux(
|
LoopVectorizer: Disallow reductions whose header phi is used outside the loop
If an outside loop user of the reduction value uses the header phi node we
cannot just reduce the vectorized phi value in the vector code epilog because
we would loose VF-1 reductions.
lp:
p = phi (0, lv)
lv = lv + 1
...
brcond , lp, outside
outside:
usr = add 0, p
(Say the loop iterates two times, the value of p coming out of the loop is one).
We cannot just transform this to:
vlp:
p = phi (<0,0>, lv)
lv = lv + <1,1>
..
brcond , lp, outside
outside:
p_reduced = p[0] + [1];
usr = add 0, p_reduced
(Because the original loop iterated two times the vectorized loop would iterate
one time, but p_reduced ends up being zero instead of one).
We would have to execute VF-1 iterations in the scalar remainder loop in such
cases. For now, just disable vectorization.
PR16522
llvm-svn: 186256
2013-07-14 03:09:29 +08:00
|
|
|
; CHECK-NOT: x i32>
|
|
|
|
|
|
|
|
|
|
define i32 @phivalueredux(i32 %p) {
|
|
|
|
|
entry:
|
|
|
|
|
br label %for.body
|
|
|
|
|
|
|
|
|
|
for.body:
|
|
|
|
|
%t.03 = phi i32 [ 0, %entry ], [ %inc, %for.body ]
|
|
|
|
|
%p.addr.02 = phi i32 [ %p, %entry ], [ %xor, %for.body ]
|
|
|
|
|
%xor = xor i32 %p.addr.02, -1
|
|
|
|
|
%inc = add nsw i32 %t.03, 1
|
|
|
|
|
%exitcond = icmp eq i32 %inc, 16
|
|
|
|
|
br i1 %exitcond, label %for.end, label %for.body
|
|
|
|
|
|
|
|
|
|
for.end:
|
|
|
|
|
ret i32 %p.addr.02
|
|
|
|
|
}
|
2013-10-08 05:05:43 +08:00
|
|
|
|
|
|
|
|
; Don't vectorize a reduction value that is not the last in a reduction cyle. We
|
|
|
|
|
; would loose iterations (VF-1) on the operations after that use.
|
|
|
|
|
; PR17498
|
|
|
|
|
|
|
|
|
|
; CHECK-LABEL: not_last_operation
|
|
|
|
|
; CHECK-NOT: x i32>
|
|
|
|
|
define i32 @not_last_operation(i32 %p, i32 %val) {
|
|
|
|
|
entry:
|
|
|
|
|
%tobool = icmp eq i32 %p, 0
|
|
|
|
|
br label %for.body
|
|
|
|
|
|
|
|
|
|
for.body:
|
|
|
|
|
%inc613.1 = phi i32 [ 0, %entry ], [ %inc6.1, %for.body ]
|
|
|
|
|
%inc511.1 = phi i32 [ %val, %entry ], [ %inc5.1, %for.body ]
|
|
|
|
|
%0 = zext i1 %tobool to i32
|
|
|
|
|
%inc4.1 = xor i32 %0, 1
|
|
|
|
|
%inc511.1.inc4.1 = add nsw i32 %inc511.1, %inc4.1
|
|
|
|
|
%inc5.1 = add nsw i32 %inc511.1.inc4.1, 1
|
|
|
|
|
%inc6.1 = add nsw i32 %inc613.1, 1
|
|
|
|
|
%exitcond.1 = icmp eq i32 %inc6.1, 22
|
|
|
|
|
br i1 %exitcond.1, label %exit, label %for.body
|
|
|
|
|
|
|
|
|
|
exit:
|
|
|
|
|
%inc.2 = add nsw i32 %inc511.1.inc4.1, 2
|
|
|
|
|
ret i32 %inc.2
|
|
|
|
|
}
|
2017-01-19 03:02:52 +08:00
|
|
|
|
|
|
|
|
;CHECK-LABEL: @reduction_sum_multiuse(
|
|
|
|
|
;CHECK: phi <4 x i32>
|
|
|
|
|
;CHECK: load <4 x i32>
|
|
|
|
|
;CHECK: add <4 x i32>
|
|
|
|
|
;CHECK: shufflevector <4 x i32> %{{.*}}, <4 x i32> undef, <4 x i32> <i32 2, i32 3, i32 undef, i32 undef>
|
|
|
|
|
;CHECK: add <4 x i32>
|
|
|
|
|
;CHECK: shufflevector <4 x i32> %{{.*}}, <4 x i32> undef, <4 x i32> <i32 1, i32 undef, i32 undef, i32 undef>
|
|
|
|
|
;CHECK: add <4 x i32>
|
|
|
|
|
;CHECK: extractelement <4 x i32> %{{.*}}, i32 0
|
|
|
|
|
;CHECK: %sum.lcssa = phi i32 [ %[[SCALAR:.*]], %.lr.ph ], [ %[[VECTOR:.*]], %middle.block ]
|
|
|
|
|
;CHECK: %sum.copy = phi i32 [ %[[SCALAR]], %.lr.ph ], [ %[[VECTOR]], %middle.block ]
|
|
|
|
|
;CHECK: ret i32
|
|
|
|
|
define i32 @reduction_sum_multiuse(i32 %n, i32* noalias nocapture %A, i32* noalias nocapture %B) {
|
|
|
|
|
%1 = icmp sgt i32 %n, 0
|
|
|
|
|
br i1 %1, label %.lr.ph.preheader, label %end
|
|
|
|
|
.lr.ph.preheader: ; preds = %0
|
|
|
|
|
br label %.lr.ph
|
|
|
|
|
|
|
|
|
|
.lr.ph: ; preds = %0, %.lr.ph
|
|
|
|
|
%indvars.iv = phi i64 [ %indvars.iv.next, %.lr.ph ], [ 0, %.lr.ph.preheader ]
|
|
|
|
|
%sum.02 = phi i32 [ %9, %.lr.ph ], [ 0, %.lr.ph.preheader ]
|
|
|
|
|
%2 = getelementptr inbounds i32, i32* %A, i64 %indvars.iv
|
|
|
|
|
%3 = load i32, i32* %2, align 4
|
|
|
|
|
%4 = getelementptr inbounds i32, i32* %B, i64 %indvars.iv
|
|
|
|
|
%5 = load i32, i32* %4, align 4
|
|
|
|
|
%6 = trunc i64 %indvars.iv to i32
|
|
|
|
|
%7 = add i32 %sum.02, %6
|
|
|
|
|
%8 = add i32 %7, %3
|
|
|
|
|
%9 = add i32 %8, %5
|
|
|
|
|
%indvars.iv.next = add i64 %indvars.iv, 1
|
|
|
|
|
%lftr.wideiv = trunc i64 %indvars.iv.next to i32
|
|
|
|
|
%exitcond = icmp eq i32 %lftr.wideiv, %n
|
|
|
|
|
br i1 %exitcond, label %._crit_edge, label %.lr.ph
|
|
|
|
|
|
|
|
|
|
._crit_edge: ; preds = %.lr.ph, %0
|
|
|
|
|
%sum.lcssa = phi i32 [ %9, %.lr.ph ]
|
|
|
|
|
%sum.copy = phi i32 [ %9, %.lr.ph ]
|
|
|
|
|
br label %end
|
|
|
|
|
|
|
|
|
|
end:
|
|
|
|
|
%f1 = phi i32 [ 0, %0 ], [ %sum.lcssa, %._crit_edge ]
|
|
|
|
|
%f2 = phi i32 [ 0, %0 ], [ %sum.copy, %._crit_edge ]
|
|
|
|
|
%final = add i32 %f1, %f2
|
|
|
|
|
ret i32 %final
|
|
|
|
|
}
|