2014-04-03 16:01:44 +00:00
|
|
|
; RUN: llc -mtriple=arm-eabi -mattr=+v7 %s -o - | FileCheck %s
|
2010-04-22 23:24:18 +00:00
|
|
|
|
|
|
|
|
define i32 @sbfx1(i32 %a) {
|
|
|
|
|
; CHECK: sbfx1
|
|
|
|
|
; CHECK: sbfx r0, r0, #7, #11
|
|
|
|
|
%t1 = lshr i32 %a, 7
|
|
|
|
|
%t2 = trunc i32 %t1 to i11
|
|
|
|
|
%t3 = sext i11 %t2 to i32
|
|
|
|
|
ret i32 %t3
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
define i32 @ubfx1(i32 %a) {
|
|
|
|
|
; CHECK: ubfx1
|
|
|
|
|
; CHECK: ubfx r0, r0, #7, #11
|
|
|
|
|
%t1 = lshr i32 %a, 7
|
|
|
|
|
%t2 = trunc i32 %t1 to i11
|
|
|
|
|
%t3 = zext i11 %t2 to i32
|
|
|
|
|
ret i32 %t3
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
define i32 @ubfx2(i32 %a) {
|
|
|
|
|
; CHECK: ubfx2
|
|
|
|
|
; CHECK: ubfx r0, r0, #7, #11
|
|
|
|
|
%t1 = lshr i32 %a, 7
|
|
|
|
|
%t2 = and i32 %t1, 2047
|
|
|
|
|
ret i32 %t2
|
|
|
|
|
}
|
|
|
|
|
|
LLVM sdisel normalize bit extraction of the form:
((x & 0xff00) >> 8) << 2
to
(x >> 6) & 0x3fc
This is general goodness since it folds a left shift into the mask. However,
the trailing zeros in the mask prevents the ARM backend from using the bit
extraction instructions. And worse since the mask materialization may require
an addition instruction. This comes up fairly frequently when the result of
the bit twiddling is used as memory address. e.g.
= ptr[(x & 0xFF0000) >> 16]
We want to generate:
ubfx r3, r1, #16, #8
ldr.w r3, [r0, r3, lsl #2]
vs.
mov.w r9, #1020
and.w r2, r9, r1, lsr #14
ldr r2, [r0, r2]
Add a late ARM specific isel optimization to
ARMDAGToDAGISel::PreprocessISelDAG(). It folds the left shift to the
'base + offset' address computation; change the mask to one which doesn't have
trailing zeros and enable the use of ubfx.
Note the optimization has to be done late since it's target specific and we
don't want to change the DAG normalization. It's also fairly restrictive
as shifter operands are not always free. It's only done for lsh 1 / 2. It's
known to be free on some cpus and they are most common for address
computation.
This is a slight win for blowfish, rijndael, etc.
rdar://12870177
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@170581 91177308-0d34-0410-b5e6-96231b3b80d8
2012-12-19 20:16:09 +00:00
|
|
|
; rdar://12870177
|
|
|
|
|
define i32 @ubfx_opt(i32* nocapture %ctx, i32 %x) nounwind readonly ssp {
|
|
|
|
|
entry:
|
|
|
|
|
; CHECK: ubfx_opt
|
|
|
|
|
; CHECK: lsr [[REG1:(lr|r[0-9]+)]], r1, #24
|
|
|
|
|
; CHECK: ldr {{lr|r[0-9]+}}, [r0, [[REG1]], lsl #2]
|
|
|
|
|
; CHECK: ubfx [[REG2:(lr|r[0-9]+)]], r1, #16, #8
|
|
|
|
|
; CHECK: ldr {{lr|r[0-9]+}}, [r0, [[REG2]], lsl #2]
|
|
|
|
|
; CHECK: ubfx [[REG3:(lr|r[0-9]+)]], r1, #8, #8
|
|
|
|
|
; CHECK: ldr {{lr|r[0-9]+}}, [r0, [[REG3]], lsl #2]
|
|
|
|
|
%and = lshr i32 %x, 8
|
|
|
|
|
%shr = and i32 %and, 255
|
|
|
|
|
%and1 = lshr i32 %x, 16
|
|
|
|
|
%shr2 = and i32 %and1, 255
|
|
|
|
|
%shr4 = lshr i32 %x, 24
|
[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
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@230786 91177308-0d34-0410-b5e6-96231b3b80d8
2015-02-27 19:29:02 +00:00
|
|
|
%arrayidx = getelementptr inbounds i32, i32* %ctx, i32 %shr4
|
2015-02-27 21:17:42 +00:00
|
|
|
%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
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@230786 91177308-0d34-0410-b5e6-96231b3b80d8
2015-02-27 19:29:02 +00:00
|
|
|
%arrayidx5 = getelementptr inbounds i32, i32* %ctx, i32 %shr2
|
2015-02-27 21:17:42 +00:00
|
|
|
%1 = load i32, i32* %arrayidx5, align 4
|
LLVM sdisel normalize bit extraction of the form:
((x & 0xff00) >> 8) << 2
to
(x >> 6) & 0x3fc
This is general goodness since it folds a left shift into the mask. However,
the trailing zeros in the mask prevents the ARM backend from using the bit
extraction instructions. And worse since the mask materialization may require
an addition instruction. This comes up fairly frequently when the result of
the bit twiddling is used as memory address. e.g.
= ptr[(x & 0xFF0000) >> 16]
We want to generate:
ubfx r3, r1, #16, #8
ldr.w r3, [r0, r3, lsl #2]
vs.
mov.w r9, #1020
and.w r2, r9, r1, lsr #14
ldr r2, [r0, r2]
Add a late ARM specific isel optimization to
ARMDAGToDAGISel::PreprocessISelDAG(). It folds the left shift to the
'base + offset' address computation; change the mask to one which doesn't have
trailing zeros and enable the use of ubfx.
Note the optimization has to be done late since it's target specific and we
don't want to change the DAG normalization. It's also fairly restrictive
as shifter operands are not always free. It's only done for lsh 1 / 2. It's
known to be free on some cpus and they are most common for address
computation.
This is a slight win for blowfish, rijndael, etc.
rdar://12870177
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@170581 91177308-0d34-0410-b5e6-96231b3b80d8
2012-12-19 20:16:09 +00:00
|
|
|
%add = add i32 %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
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@230786 91177308-0d34-0410-b5e6-96231b3b80d8
2015-02-27 19:29:02 +00:00
|
|
|
%arrayidx6 = getelementptr inbounds i32, i32* %ctx, i32 %shr
|
2015-02-27 21:17:42 +00:00
|
|
|
%2 = load i32, i32* %arrayidx6, align 4
|
LLVM sdisel normalize bit extraction of the form:
((x & 0xff00) >> 8) << 2
to
(x >> 6) & 0x3fc
This is general goodness since it folds a left shift into the mask. However,
the trailing zeros in the mask prevents the ARM backend from using the bit
extraction instructions. And worse since the mask materialization may require
an addition instruction. This comes up fairly frequently when the result of
the bit twiddling is used as memory address. e.g.
= ptr[(x & 0xFF0000) >> 16]
We want to generate:
ubfx r3, r1, #16, #8
ldr.w r3, [r0, r3, lsl #2]
vs.
mov.w r9, #1020
and.w r2, r9, r1, lsr #14
ldr r2, [r0, r2]
Add a late ARM specific isel optimization to
ARMDAGToDAGISel::PreprocessISelDAG(). It folds the left shift to the
'base + offset' address computation; change the mask to one which doesn't have
trailing zeros and enable the use of ubfx.
Note the optimization has to be done late since it's target specific and we
don't want to change the DAG normalization. It's also fairly restrictive
as shifter operands are not always free. It's only done for lsh 1 / 2. It's
known to be free on some cpus and they are most common for address
computation.
This is a slight win for blowfish, rijndael, etc.
rdar://12870177
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@170581 91177308-0d34-0410-b5e6-96231b3b80d8
2012-12-19 20:16:09 +00:00
|
|
|
%add7 = add i32 %add, %2
|
|
|
|
|
ret i32 %add7
|
|
|
|
|
}
|