maxjhandsome
/
llvm-project
forked from OSchip/llvm-project
1
0
Fork 0
Code Issues Pull Requests Packages Releases Wiki Activity
llvm-project/llvm/test/CodeGen/X86/memcpy-2.ll

210 lines
5.9 KiB
LLVM
Raw Normal View History

When lowering memcpys to loads and stores, make sure we don't promote alignments past the natural stack alignment. llvm-svn: 174085
2013-02-01 04:23:43 +08:00
; RUN: llc < %s -mattr=+sse2 -mtriple=i686-apple-darwin -mcpu=core2 | FileCheck %s -check-prefix=SSE2-Darwin
; RUN: llc < %s -mattr=+sse2 -mtriple=i686-pc-mingw32 -mcpu=core2 | FileCheck %s -check-prefix=SSE2-Mingw32
Revert r122955. It seems using movups to lower memcpy can cause massive regression (even on Nehalem) in edge cases. I also didn't see any real performance benefit. llvm-svn: 123015
2011-01-08 03:35:30 +08:00
; RUN: llc < %s -mattr=+sse,-sse2 -mtriple=i686-apple-darwin -mcpu=core2 | FileCheck %s -check-prefix=SSE1
Add -mcpu to memcpy / memset tests to ensure they behave the same on all hosts / targets. llvm-svn: 100101
2010-04-01 16:25:26 +08:00
; RUN: llc < %s -mattr=-sse -mtriple=i686-apple-darwin -mcpu=core2 | FileCheck %s -check-prefix=NOSSE
In 64-bit mode, use i64 to lower memcpy / memset instead of f64. llvm-svn: 100137
2010-04-02 04:27:45 +08:00
; RUN: llc < %s -mtriple=x86_64-apple-darwin -mcpu=core2 | FileCheck %s -check-prefix=X86-64
make current codegen visible in the checks, so we can decide if it's right llvm-svn: 245120
2015-08-15 07:03:01 +08:00
; RUN: llc < %s -mtriple=x86_64-apple-darwin -mcpu=nehalem | FileCheck %s -check-prefix=NHM_64
Add a couple of test cases. llvm-svn: 51441
2008-05-23 05:19:19 +08:00
Avoid using f64 to lower memcpy from constant string. It's cheaper to use i32 store of immediates. llvm-svn: 100751
2010-04-08 15:37:57 +08:00
@.str = internal constant [25 x i8] c"image\00\00\00\00\00\00\00\00\00\00\00\00\00\00\00\00\00\00\00\00"
@.str2 = internal constant [30 x i8] c"xxxxxxxxxxxxxxxxxxxxxxxxxxxxx\00", align 4
Add a couple of test cases. llvm-svn: 51441
2008-05-23 05:19:19 +08:00
Fix sdisel memcpy, memset, memmove lowering: 1. Makes it possible to lower with floating point loads and stores. 2. Avoid unaligned loads / stores unless it's fast. 3. Fix some memcpy lowering logic bug related to when to optimize a load from constant string into a constant. 4. Adjust x86 memcpy lowering threshold to make it more sane. 5. Fix x86 target hook so it uses vector and floating point memory ops more effectively. rdar://7774704 llvm-svn: 100090
2010-04-01 14:04:33 +08:00
define void @t1(i32 %argc, i8** %argv) nounwind {
Add a couple of test cases. llvm-svn: 51441
2008-05-23 05:19:19 +08:00
entry:
Mass update to CodeGen tests to use CHECK-LABEL for labels corresponding to function definitions for more informative error messages. No functionality change and all updated tests passed locally. This update was done with the following bash script: find test/CodeGen -name "*.ll" | \ while read NAME; do echo "$NAME" if ! grep -q "^; *RUN: *llc.*debug" $NAME; then TEMP=`mktemp -t temp` cp $NAME $TEMP sed -n "s/^define [^@]*@\([A-Za-z0-9_]*\)(.*$/\1/p" < $NAME | \ while read FUNC; do sed -i '' "s/;\(.*\)\([A-Za-z0-9_-]*\):\( *\)$FUNC: *\$/;\1\2-LABEL:\3$FUNC:/g" $TEMP done sed -i '' "s/;\(.*\)-LABEL-LABEL:/;\1-LABEL:/" $TEMP sed -i '' "s/;\(.*\)-NEXT-LABEL:/;\1-NEXT:/" $TEMP sed -i '' "s/;\(.*\)-NOT-LABEL:/;\1-NOT:/" $TEMP sed -i '' "s/;\(.*\)-DAG-LABEL:/;\1-DAG:/" $TEMP mv $TEMP $NAME fi done llvm-svn: 186280
2013-07-14 14:24:09 +08:00
; SSE2-Darwin-LABEL: t1:
When lowering memcpys to loads and stores, make sure we don't promote alignments past the natural stack alignment. llvm-svn: 174085
2013-02-01 04:23:43 +08:00
; SSE2-Darwin: movsd _.str+16, %xmm0
; SSE2-Darwin: movsd %xmm0, 16(%esp)
; SSE2-Darwin: movaps _.str, %xmm0
; SSE2-Darwin: movaps %xmm0
; SSE2-Darwin: movb $0, 24(%esp)
Mass update to CodeGen tests to use CHECK-LABEL for labels corresponding to function definitions for more informative error messages. No functionality change and all updated tests passed locally. This update was done with the following bash script: find test/CodeGen -name "*.ll" | \ while read NAME; do echo "$NAME" if ! grep -q "^; *RUN: *llc.*debug" $NAME; then TEMP=`mktemp -t temp` cp $NAME $TEMP sed -n "s/^define [^@]*@\([A-Za-z0-9_]*\)(.*$/\1/p" < $NAME | \ while read FUNC; do sed -i '' "s/;\(.*\)\([A-Za-z0-9_-]*\):\( *\)$FUNC: *\$/;\1\2-LABEL:\3$FUNC:/g" $TEMP done sed -i '' "s/;\(.*\)-LABEL-LABEL:/;\1-LABEL:/" $TEMP sed -i '' "s/;\(.*\)-NEXT-LABEL:/;\1-NEXT:/" $TEMP sed -i '' "s/;\(.*\)-NOT-LABEL:/;\1-NOT:/" $TEMP sed -i '' "s/;\(.*\)-DAG-LABEL:/;\1-DAG:/" $TEMP mv $TEMP $NAME fi done llvm-svn: 186280
2013-07-14 14:24:09 +08:00
; SSE2-Mingw32-LABEL: t1:
When lowering memcpys to loads and stores, make sure we don't promote alignments past the natural stack alignment. llvm-svn: 174085
2013-02-01 04:23:43 +08:00
; SSE2-Mingw32: movsd _.str+16, %xmm0
; SSE2-Mingw32: movsd %xmm0, 16(%esp)
; SSE2-Mingw32: movaps _.str, %xmm0
; SSE2-Mingw32: movups %xmm0
; SSE2-Mingw32: movb $0, 24(%esp)
Fix sdisel memcpy, memset, memmove lowering: 1. Makes it possible to lower with floating point loads and stores. 2. Avoid unaligned loads / stores unless it's fast. 3. Fix some memcpy lowering logic bug related to when to optimize a load from constant string into a constant. 4. Adjust x86 memcpy lowering threshold to make it more sane. 5. Fix x86 target hook so it uses vector and floating point memory ops more effectively. rdar://7774704 llvm-svn: 100090
2010-04-01 14:04:33 +08:00
Mass update to CodeGen tests to use CHECK-LABEL for labels corresponding to function definitions for more informative error messages. No functionality change and all updated tests passed locally. This update was done with the following bash script: find test/CodeGen -name "*.ll" | \ while read NAME; do echo "$NAME" if ! grep -q "^; *RUN: *llc.*debug" $NAME; then TEMP=`mktemp -t temp` cp $NAME $TEMP sed -n "s/^define [^@]*@\([A-Za-z0-9_]*\)(.*$/\1/p" < $NAME | \ while read FUNC; do sed -i '' "s/;\(.*\)\([A-Za-z0-9_-]*\):\( *\)$FUNC: *\$/;\1\2-LABEL:\3$FUNC:/g" $TEMP done sed -i '' "s/;\(.*\)-LABEL-LABEL:/;\1-LABEL:/" $TEMP sed -i '' "s/;\(.*\)-NEXT-LABEL:/;\1-NEXT:/" $TEMP sed -i '' "s/;\(.*\)-NOT-LABEL:/;\1-NOT:/" $TEMP sed -i '' "s/;\(.*\)-DAG-LABEL:/;\1-DAG:/" $TEMP mv $TEMP $NAME fi done llvm-svn: 186280
2013-07-14 14:24:09 +08:00
; SSE1-LABEL: t1:
Revert r122955. It seems using movups to lower memcpy can cause massive regression (even on Nehalem) in edge cases. I also didn't see any real performance benefit. llvm-svn: 123015
2011-01-08 03:35:30 +08:00
; SSE1: movaps _.str, %xmm0
; SSE1: movaps %xmm0
Some enhancements for memcpy / memset inline expansion. 1. Teach it to use overlapping unaligned load / store to copy / set the trailing bytes. e.g. On 86, use two pairs of movups / movaps for 17 - 31 byte copies. 2. Use f64 for memcpy / memset on targets where i64 is not legal but f64 is. e.g. x86 and ARM. 3. When memcpy from a constant string, do *not* replace the load with a constant if it's not possible to materialize an integer immediate with a single instruction (required a new target hook: TLI.isIntImmLegal()). 4. Use unaligned load / stores more aggressively if target hooks indicates they are "fast". 5. Update ARM target hooks to use unaligned load / stores. e.g. vld1.8 / vst1.8. Also increase the threshold to something reasonable (8 for memset, 4 pairs for memcpy). This significantly improves Dhrystone, up to 50% on ARM iOS devices. rdar://12760078 llvm-svn: 169791
2012-12-11 07:21:26 +08:00
; SSE1: movb $0, 24(%esp)
Avoid using lossy load / stores for memcpy / memset expansion. e.g. f64 load / store on non-SSE2 x86 targets. llvm-svn: 169944
2012-12-12 08:42:09 +08:00
; SSE1: movl $0, 20(%esp)
; SSE1: movl $0, 16(%esp)
Revert r122955. It seems using movups to lower memcpy can cause massive regression (even on Nehalem) in edge cases. I also didn't see any real performance benefit. llvm-svn: 123015
2011-01-08 03:35:30 +08:00
Mass update to CodeGen tests to use CHECK-LABEL for labels corresponding to function definitions for more informative error messages. No functionality change and all updated tests passed locally. This update was done with the following bash script: find test/CodeGen -name "*.ll" | \ while read NAME; do echo "$NAME" if ! grep -q "^; *RUN: *llc.*debug" $NAME; then TEMP=`mktemp -t temp` cp $NAME $TEMP sed -n "s/^define [^@]*@\([A-Za-z0-9_]*\)(.*$/\1/p" < $NAME | \ while read FUNC; do sed -i '' "s/;\(.*\)\([A-Za-z0-9_-]*\):\( *\)$FUNC: *\$/;\1\2-LABEL:\3$FUNC:/g" $TEMP done sed -i '' "s/;\(.*\)-LABEL-LABEL:/;\1-LABEL:/" $TEMP sed -i '' "s/;\(.*\)-NEXT-LABEL:/;\1-NEXT:/" $TEMP sed -i '' "s/;\(.*\)-NOT-LABEL:/;\1-NOT:/" $TEMP sed -i '' "s/;\(.*\)-DAG-LABEL:/;\1-DAG:/" $TEMP mv $TEMP $NAME fi done llvm-svn: 186280
2013-07-14 14:24:09 +08:00
; NOSSE-LABEL: t1:
Fix sdisel memcpy, memset, memmove lowering: 1. Makes it possible to lower with floating point loads and stores. 2. Avoid unaligned loads / stores unless it's fast. 3. Fix some memcpy lowering logic bug related to when to optimize a load from constant string into a constant. 4. Adjust x86 memcpy lowering threshold to make it more sane. 5. Fix x86 target hook so it uses vector and floating point memory ops more effectively. rdar://7774704 llvm-svn: 100090
2010-04-01 14:04:33 +08:00
; NOSSE: movb $0
; NOSSE: movl $0
; NOSSE: movl $0
; NOSSE: movl $0
; NOSSE: movl $0
; NOSSE: movl $101
; NOSSE: movl $1734438249
In 64-bit mode, use i64 to lower memcpy / memset instead of f64. llvm-svn: 100137
2010-04-02 04:27:45 +08:00
Mass update to CodeGen tests to use CHECK-LABEL for labels corresponding to function definitions for more informative error messages. No functionality change and all updated tests passed locally. This update was done with the following bash script: find test/CodeGen -name "*.ll" | \ while read NAME; do echo "$NAME" if ! grep -q "^; *RUN: *llc.*debug" $NAME; then TEMP=`mktemp -t temp` cp $NAME $TEMP sed -n "s/^define [^@]*@\([A-Za-z0-9_]*\)(.*$/\1/p" < $NAME | \ while read FUNC; do sed -i '' "s/;\(.*\)\([A-Za-z0-9_-]*\):\( *\)$FUNC: *\$/;\1\2-LABEL:\3$FUNC:/g" $TEMP done sed -i '' "s/;\(.*\)-LABEL-LABEL:/;\1-LABEL:/" $TEMP sed -i '' "s/;\(.*\)-NEXT-LABEL:/;\1-NEXT:/" $TEMP sed -i '' "s/;\(.*\)-NOT-LABEL:/;\1-NOT:/" $TEMP sed -i '' "s/;\(.*\)-DAG-LABEL:/;\1-DAG:/" $TEMP mv $TEMP $NAME fi done llvm-svn: 186280
2013-07-14 14:24:09 +08:00
; X86-64-LABEL: t1:
Avoid using f64 to lower memcpy from constant string. It's cheaper to use i32 store of immediates. llvm-svn: 100751
2010-04-08 15:37:57 +08:00
; X86-64: movaps _.str(%rip), %xmm0
In 64-bit mode, use i64 to lower memcpy / memset instead of f64. llvm-svn: 100137
2010-04-02 04:27:45 +08:00
; X86-64: movaps %xmm0
; X86-64: movb $0
; X86-64: movq $0
Avoid using f64 to lower memcpy from constant string. It's cheaper to use i32 store of immediates. llvm-svn: 100751
2010-04-08 15:37:57 +08:00
%tmp1 = alloca [25 x i8]
%tmp2 = bitcast [25 x i8]* %tmp1 to i8*
[opaque pointer type] Add textual IR support for explicit type parameter to gep operator Similar to gep (r230786) and load (r230794) changes. Similar migration script can be used to update test cases, which successfully migrated all of LLVM and Polly, but about 4 test cases needed manually changes in Clang. (this script will read the contents of stdin and massage it into stdout - wrap it in the 'apply.sh' script shown in previous commits + xargs to apply it over a large set of test cases) import fileinput import sys import re rep = re.compile(r"(getelementptr(?:\s+inbounds)?\s*\()((<\d*\s+x\s+)?([^@]*?)(|\s*addrspace\(\d+\))\s*\*(?(3)>)\s*)(?=$|%|@|null|undef|blockaddress|getelementptr|addrspacecast|bitcast|inttoptr|zeroinitializer|<|\[\[[a-zA-Z]|\{\{)", re.MULTILINE | re.DOTALL) def conv(match): line = match.group(1) line += match.group(4) line += ", " line += match.group(2) return line line = sys.stdin.read() off = 0 for match in re.finditer(rep, line): sys.stdout.write(line[off:match.start()]) sys.stdout.write(conv(match)) off = match.end() sys.stdout.write(line[off:]) llvm-svn: 232184
2015-03-14 02:20:45 +08:00
call void @llvm.memcpy.p0i8.p0i8.i32(i8* %tmp2, i8* getelementptr inbounds ([25 x i8], [25 x i8]* @.str, i32 0, i32 0), i32 25, i32 1, i1 false)
Avoid using f64 to lower memcpy from constant string. It's cheaper to use i32 store of immediates. llvm-svn: 100751
2010-04-08 15:37:57 +08:00
unreachable
Add a couple of test cases. llvm-svn: 51441
2008-05-23 05:19:19 +08:00
}
Fix sdisel memcpy, memset, memmove lowering: 1. Makes it possible to lower with floating point loads and stores. 2. Avoid unaligned loads / stores unless it's fast. 3. Fix some memcpy lowering logic bug related to when to optimize a load from constant string into a constant. 4. Adjust x86 memcpy lowering threshold to make it more sane. 5. Fix x86 target hook so it uses vector and floating point memory ops more effectively. rdar://7774704 llvm-svn: 100090
2010-04-01 14:04:33 +08:00
;rdar://7774704
%struct.s0 = type { [2 x double] }
define void @t2(%struct.s0* nocapture %a, %struct.s0* nocapture %b) nounwind ssp {
entry:
Mass update to CodeGen tests to use CHECK-LABEL for labels corresponding to function definitions for more informative error messages. No functionality change and all updated tests passed locally. This update was done with the following bash script: find test/CodeGen -name "*.ll" | \ while read NAME; do echo "$NAME" if ! grep -q "^; *RUN: *llc.*debug" $NAME; then TEMP=`mktemp -t temp` cp $NAME $TEMP sed -n "s/^define [^@]*@\([A-Za-z0-9_]*\)(.*$/\1/p" < $NAME | \ while read FUNC; do sed -i '' "s/;\(.*\)\([A-Za-z0-9_-]*\):\( *\)$FUNC: *\$/;\1\2-LABEL:\3$FUNC:/g" $TEMP done sed -i '' "s/;\(.*\)-LABEL-LABEL:/;\1-LABEL:/" $TEMP sed -i '' "s/;\(.*\)-NEXT-LABEL:/;\1-NEXT:/" $TEMP sed -i '' "s/;\(.*\)-NOT-LABEL:/;\1-NOT:/" $TEMP sed -i '' "s/;\(.*\)-DAG-LABEL:/;\1-DAG:/" $TEMP mv $TEMP $NAME fi done llvm-svn: 186280
2013-07-14 14:24:09 +08:00
; SSE2-Darwin-LABEL: t2:
Enable MI Sched for x86. This changes the SelectionDAG scheduling preference to source order. Soon, the SelectionDAG scheduler can be bypassed saving a nice chunk of compile time. Performance differences that result from this change are often a consequence of register coalescing. The register coalescer is far from perfect. Bugs can be filed for deficiencies. On x86 SandyBridge/Haswell, the source order schedule is often preserved, particularly for small blocks. Register pressure is generally improved over the SD scheduler's ILP mode. However, we are still able to handle large blocks that require latency hiding, unlike the SD scheduler's BURR mode. MI scheduler also attempts to discover the critical path in single-block loops and adjust heuristics accordingly. The MI scheduler relies on the new machine model. This is currently unimplemented for AVX, so we may not be generating the best code yet. Unit tests are updated so they don't depend on SD scheduling heuristics. llvm-svn: 192750
2013-10-16 07:33:07 +08:00
; SSE2-Darwin: movaps (%ecx), %xmm0
When lowering memcpys to loads and stores, make sure we don't promote alignments past the natural stack alignment. llvm-svn: 174085
2013-02-01 04:23:43 +08:00
; SSE2-Darwin: movaps %xmm0, (%eax)
Mass update to CodeGen tests to use CHECK-LABEL for labels corresponding to function definitions for more informative error messages. No functionality change and all updated tests passed locally. This update was done with the following bash script: find test/CodeGen -name "*.ll" | \ while read NAME; do echo "$NAME" if ! grep -q "^; *RUN: *llc.*debug" $NAME; then TEMP=`mktemp -t temp` cp $NAME $TEMP sed -n "s/^define [^@]*@\([A-Za-z0-9_]*\)(.*$/\1/p" < $NAME | \ while read FUNC; do sed -i '' "s/;\(.*\)\([A-Za-z0-9_-]*\):\( *\)$FUNC: *\$/;\1\2-LABEL:\3$FUNC:/g" $TEMP done sed -i '' "s/;\(.*\)-LABEL-LABEL:/;\1-LABEL:/" $TEMP sed -i '' "s/;\(.*\)-NEXT-LABEL:/;\1-NEXT:/" $TEMP sed -i '' "s/;\(.*\)-NOT-LABEL:/;\1-NOT:/" $TEMP sed -i '' "s/;\(.*\)-DAG-LABEL:/;\1-DAG:/" $TEMP mv $TEMP $NAME fi done llvm-svn: 186280
2013-07-14 14:24:09 +08:00
; SSE2-Mingw32-LABEL: t2:
Enable MI Sched for x86. This changes the SelectionDAG scheduling preference to source order. Soon, the SelectionDAG scheduler can be bypassed saving a nice chunk of compile time. Performance differences that result from this change are often a consequence of register coalescing. The register coalescer is far from perfect. Bugs can be filed for deficiencies. On x86 SandyBridge/Haswell, the source order schedule is often preserved, particularly for small blocks. Register pressure is generally improved over the SD scheduler's ILP mode. However, we are still able to handle large blocks that require latency hiding, unlike the SD scheduler's BURR mode. MI scheduler also attempts to discover the critical path in single-block loops and adjust heuristics accordingly. The MI scheduler relies on the new machine model. This is currently unimplemented for AVX, so we may not be generating the best code yet. Unit tests are updated so they don't depend on SD scheduling heuristics. llvm-svn: 192750
2013-10-16 07:33:07 +08:00
; SSE2-Mingw32: movaps (%ecx), %xmm0
When lowering memcpys to loads and stores, make sure we don't promote alignments past the natural stack alignment. llvm-svn: 174085
2013-02-01 04:23:43 +08:00
; SSE2-Mingw32: movaps %xmm0, (%eax)
Fix sdisel memcpy, memset, memmove lowering: 1. Makes it possible to lower with floating point loads and stores. 2. Avoid unaligned loads / stores unless it's fast. 3. Fix some memcpy lowering logic bug related to when to optimize a load from constant string into a constant. 4. Adjust x86 memcpy lowering threshold to make it more sane. 5. Fix x86 target hook so it uses vector and floating point memory ops more effectively. rdar://7774704 llvm-svn: 100090
2010-04-01 14:04:33 +08:00
Mass update to CodeGen tests to use CHECK-LABEL for labels corresponding to function definitions for more informative error messages. No functionality change and all updated tests passed locally. This update was done with the following bash script: find test/CodeGen -name "*.ll" | \ while read NAME; do echo "$NAME" if ! grep -q "^; *RUN: *llc.*debug" $NAME; then TEMP=`mktemp -t temp` cp $NAME $TEMP sed -n "s/^define [^@]*@\([A-Za-z0-9_]*\)(.*$/\1/p" < $NAME | \ while read FUNC; do sed -i '' "s/;\(.*\)\([A-Za-z0-9_-]*\):\( *\)$FUNC: *\$/;\1\2-LABEL:\3$FUNC:/g" $TEMP done sed -i '' "s/;\(.*\)-LABEL-LABEL:/;\1-LABEL:/" $TEMP sed -i '' "s/;\(.*\)-NEXT-LABEL:/;\1-NEXT:/" $TEMP sed -i '' "s/;\(.*\)-NOT-LABEL:/;\1-NOT:/" $TEMP sed -i '' "s/;\(.*\)-DAG-LABEL:/;\1-DAG:/" $TEMP mv $TEMP $NAME fi done llvm-svn: 186280
2013-07-14 14:24:09 +08:00
; SSE1-LABEL: t2:
Enable MI Sched for x86. This changes the SelectionDAG scheduling preference to source order. Soon, the SelectionDAG scheduler can be bypassed saving a nice chunk of compile time. Performance differences that result from this change are often a consequence of register coalescing. The register coalescer is far from perfect. Bugs can be filed for deficiencies. On x86 SandyBridge/Haswell, the source order schedule is often preserved, particularly for small blocks. Register pressure is generally improved over the SD scheduler's ILP mode. However, we are still able to handle large blocks that require latency hiding, unlike the SD scheduler's BURR mode. MI scheduler also attempts to discover the critical path in single-block loops and adjust heuristics accordingly. The MI scheduler relies on the new machine model. This is currently unimplemented for AVX, so we may not be generating the best code yet. Unit tests are updated so they don't depend on SD scheduling heuristics. llvm-svn: 192750
2013-10-16 07:33:07 +08:00
; SSE1: movaps (%ecx), %xmm0
Revert r122955. It seems using movups to lower memcpy can cause massive regression (even on Nehalem) in edge cases. I also didn't see any real performance benefit. llvm-svn: 123015
2011-01-08 03:35:30 +08:00
; SSE1: movaps %xmm0, (%eax)
Mass update to CodeGen tests to use CHECK-LABEL for labels corresponding to function definitions for more informative error messages. No functionality change and all updated tests passed locally. This update was done with the following bash script: find test/CodeGen -name "*.ll" | \ while read NAME; do echo "$NAME" if ! grep -q "^; *RUN: *llc.*debug" $NAME; then TEMP=`mktemp -t temp` cp $NAME $TEMP sed -n "s/^define [^@]*@\([A-Za-z0-9_]*\)(.*$/\1/p" < $NAME | \ while read FUNC; do sed -i '' "s/;\(.*\)\([A-Za-z0-9_-]*\):\( *\)$FUNC: *\$/;\1\2-LABEL:\3$FUNC:/g" $TEMP done sed -i '' "s/;\(.*\)-LABEL-LABEL:/;\1-LABEL:/" $TEMP sed -i '' "s/;\(.*\)-NEXT-LABEL:/;\1-NEXT:/" $TEMP sed -i '' "s/;\(.*\)-NOT-LABEL:/;\1-NOT:/" $TEMP sed -i '' "s/;\(.*\)-DAG-LABEL:/;\1-DAG:/" $TEMP mv $TEMP $NAME fi done llvm-svn: 186280
2013-07-14 14:24:09 +08:00
; NOSSE-LABEL: t2:
Fix sdisel memcpy, memset, memmove lowering: 1. Makes it possible to lower with floating point loads and stores. 2. Avoid unaligned loads / stores unless it's fast. 3. Fix some memcpy lowering logic bug related to when to optimize a load from constant string into a constant. 4. Adjust x86 memcpy lowering threshold to make it more sane. 5. Fix x86 target hook so it uses vector and floating point memory ops more effectively. rdar://7774704 llvm-svn: 100090
2010-04-01 14:04:33 +08:00
; NOSSE: movl
; NOSSE: movl
; NOSSE: movl
; NOSSE: movl
; NOSSE: movl
; NOSSE: movl
; NOSSE: movl
; NOSSE: movl
; NOSSE: movl
; NOSSE: movl
In 64-bit mode, use i64 to lower memcpy / memset instead of f64. llvm-svn: 100137
2010-04-02 04:27:45 +08:00
Mass update to CodeGen tests to use CHECK-LABEL for labels corresponding to function definitions for more informative error messages. No functionality change and all updated tests passed locally. This update was done with the following bash script: find test/CodeGen -name "*.ll" | \ while read NAME; do echo "$NAME" if ! grep -q "^; *RUN: *llc.*debug" $NAME; then TEMP=`mktemp -t temp` cp $NAME $TEMP sed -n "s/^define [^@]*@\([A-Za-z0-9_]*\)(.*$/\1/p" < $NAME | \ while read FUNC; do sed -i '' "s/;\(.*\)\([A-Za-z0-9_-]*\):\( *\)$FUNC: *\$/;\1\2-LABEL:\3$FUNC:/g" $TEMP done sed -i '' "s/;\(.*\)-LABEL-LABEL:/;\1-LABEL:/" $TEMP sed -i '' "s/;\(.*\)-NEXT-LABEL:/;\1-NEXT:/" $TEMP sed -i '' "s/;\(.*\)-NOT-LABEL:/;\1-NOT:/" $TEMP sed -i '' "s/;\(.*\)-DAG-LABEL:/;\1-DAG:/" $TEMP mv $TEMP $NAME fi done llvm-svn: 186280
2013-07-14 14:24:09 +08:00
; X86-64-LABEL: t2:
In 64-bit mode, use i64 to lower memcpy / memset instead of f64. llvm-svn: 100137
2010-04-02 04:27:45 +08:00
; X86-64: movaps (%rsi), %xmm0
; X86-64: movaps %xmm0, (%rdi)
Fix sdisel memcpy, memset, memmove lowering: 1. Makes it possible to lower with floating point loads and stores. 2. Avoid unaligned loads / stores unless it's fast. 3. Fix some memcpy lowering logic bug related to when to optimize a load from constant string into a constant. 4. Adjust x86 memcpy lowering threshold to make it more sane. 5. Fix x86 target hook so it uses vector and floating point memory ops more effectively. rdar://7774704 llvm-svn: 100090
2010-04-01 14:04:33 +08:00
%tmp2 = bitcast %struct.s0* %a to i8* ; <i8*> [#uses=1]
%tmp3 = bitcast %struct.s0* %b to i8* ; <i8*> [#uses=1]
rip out a ton of intrinsic modernization logic from AutoUpgrade.cpp, which is for pre-2.9 bitcode files. We keep x86 unaligned loads, movnt, crc32, and the target indep prefetch change. As usual, updating the testsuite is a PITA. llvm-svn: 133337
2011-06-18 14:05:24 +08:00
tail call void @llvm.memcpy.p0i8.p0i8.i32(i8* %tmp2, i8* %tmp3, i32 16, i32 16, i1 false)
Fix sdisel memcpy, memset, memmove lowering: 1. Makes it possible to lower with floating point loads and stores. 2. Avoid unaligned loads / stores unless it's fast. 3. Fix some memcpy lowering logic bug related to when to optimize a load from constant string into a constant. 4. Adjust x86 memcpy lowering threshold to make it more sane. 5. Fix x86 target hook so it uses vector and floating point memory ops more effectively. rdar://7774704 llvm-svn: 100090
2010-04-01 14:04:33 +08:00
ret void
}
define void @t3(%struct.s0* nocapture %a, %struct.s0* nocapture %b) nounwind ssp {
entry:
Mass update to CodeGen tests to use CHECK-LABEL for labels corresponding to function definitions for more informative error messages. No functionality change and all updated tests passed locally. This update was done with the following bash script: find test/CodeGen -name "*.ll" | \ while read NAME; do echo "$NAME" if ! grep -q "^; *RUN: *llc.*debug" $NAME; then TEMP=`mktemp -t temp` cp $NAME $TEMP sed -n "s/^define [^@]*@\([A-Za-z0-9_]*\)(.*$/\1/p" < $NAME | \ while read FUNC; do sed -i '' "s/;\(.*\)\([A-Za-z0-9_-]*\):\( *\)$FUNC: *\$/;\1\2-LABEL:\3$FUNC:/g" $TEMP done sed -i '' "s/;\(.*\)-LABEL-LABEL:/;\1-LABEL:/" $TEMP sed -i '' "s/;\(.*\)-NEXT-LABEL:/;\1-NEXT:/" $TEMP sed -i '' "s/;\(.*\)-NOT-LABEL:/;\1-NOT:/" $TEMP sed -i '' "s/;\(.*\)-DAG-LABEL:/;\1-DAG:/" $TEMP mv $TEMP $NAME fi done llvm-svn: 186280
2013-07-14 14:24:09 +08:00
; SSE2-Darwin-LABEL: t3:
Enable MI Sched for x86. This changes the SelectionDAG scheduling preference to source order. Soon, the SelectionDAG scheduler can be bypassed saving a nice chunk of compile time. Performance differences that result from this change are often a consequence of register coalescing. The register coalescer is far from perfect. Bugs can be filed for deficiencies. On x86 SandyBridge/Haswell, the source order schedule is often preserved, particularly for small blocks. Register pressure is generally improved over the SD scheduler's ILP mode. However, we are still able to handle large blocks that require latency hiding, unlike the SD scheduler's BURR mode. MI scheduler also attempts to discover the critical path in single-block loops and adjust heuristics accordingly. The MI scheduler relies on the new machine model. This is currently unimplemented for AVX, so we may not be generating the best code yet. Unit tests are updated so they don't depend on SD scheduling heuristics. llvm-svn: 192750
2013-10-16 07:33:07 +08:00
; SSE2-Darwin: movsd (%ecx), %xmm0
; SSE2-Darwin: movsd 8(%ecx), %xmm1
When lowering memcpys to loads and stores, make sure we don't promote alignments past the natural stack alignment. llvm-svn: 174085
2013-02-01 04:23:43 +08:00
; SSE2-Darwin: movsd %xmm1, 8(%eax)
; SSE2-Darwin: movsd %xmm0, (%eax)
Mass update to CodeGen tests to use CHECK-LABEL for labels corresponding to function definitions for more informative error messages. No functionality change and all updated tests passed locally. This update was done with the following bash script: find test/CodeGen -name "*.ll" | \ while read NAME; do echo "$NAME" if ! grep -q "^; *RUN: *llc.*debug" $NAME; then TEMP=`mktemp -t temp` cp $NAME $TEMP sed -n "s/^define [^@]*@\([A-Za-z0-9_]*\)(.*$/\1/p" < $NAME | \ while read FUNC; do sed -i '' "s/;\(.*\)\([A-Za-z0-9_-]*\):\( *\)$FUNC: *\$/;\1\2-LABEL:\3$FUNC:/g" $TEMP done sed -i '' "s/;\(.*\)-LABEL-LABEL:/;\1-LABEL:/" $TEMP sed -i '' "s/;\(.*\)-NEXT-LABEL:/;\1-NEXT:/" $TEMP sed -i '' "s/;\(.*\)-NOT-LABEL:/;\1-NOT:/" $TEMP sed -i '' "s/;\(.*\)-DAG-LABEL:/;\1-DAG:/" $TEMP mv $TEMP $NAME fi done llvm-svn: 186280
2013-07-14 14:24:09 +08:00
; SSE2-Mingw32-LABEL: t3:
Enable MI Sched for x86. This changes the SelectionDAG scheduling preference to source order. Soon, the SelectionDAG scheduler can be bypassed saving a nice chunk of compile time. Performance differences that result from this change are often a consequence of register coalescing. The register coalescer is far from perfect. Bugs can be filed for deficiencies. On x86 SandyBridge/Haswell, the source order schedule is often preserved, particularly for small blocks. Register pressure is generally improved over the SD scheduler's ILP mode. However, we are still able to handle large blocks that require latency hiding, unlike the SD scheduler's BURR mode. MI scheduler also attempts to discover the critical path in single-block loops and adjust heuristics accordingly. The MI scheduler relies on the new machine model. This is currently unimplemented for AVX, so we may not be generating the best code yet. Unit tests are updated so they don't depend on SD scheduling heuristics. llvm-svn: 192750
2013-10-16 07:33:07 +08:00
; SSE2-Mingw32: movsd (%ecx), %xmm0
; SSE2-Mingw32: movsd 8(%ecx), %xmm1
When lowering memcpys to loads and stores, make sure we don't promote alignments past the natural stack alignment. llvm-svn: 174085
2013-02-01 04:23:43 +08:00
; SSE2-Mingw32: movsd %xmm1, 8(%eax)
; SSE2-Mingw32: movsd %xmm0, (%eax)
Revert r122955. It seems using movups to lower memcpy can cause massive regression (even on Nehalem) in edge cases. I also didn't see any real performance benefit. llvm-svn: 123015
2011-01-08 03:35:30 +08:00
Mass update to CodeGen tests to use CHECK-LABEL for labels corresponding to function definitions for more informative error messages. No functionality change and all updated tests passed locally. This update was done with the following bash script: find test/CodeGen -name "*.ll" | \ while read NAME; do echo "$NAME" if ! grep -q "^; *RUN: *llc.*debug" $NAME; then TEMP=`mktemp -t temp` cp $NAME $TEMP sed -n "s/^define [^@]*@\([A-Za-z0-9_]*\)(.*$/\1/p" < $NAME | \ while read FUNC; do sed -i '' "s/;\(.*\)\([A-Za-z0-9_-]*\):\( *\)$FUNC: *\$/;\1\2-LABEL:\3$FUNC:/g" $TEMP done sed -i '' "s/;\(.*\)-LABEL-LABEL:/;\1-LABEL:/" $TEMP sed -i '' "s/;\(.*\)-NEXT-LABEL:/;\1-NEXT:/" $TEMP sed -i '' "s/;\(.*\)-NOT-LABEL:/;\1-NOT:/" $TEMP sed -i '' "s/;\(.*\)-DAG-LABEL:/;\1-DAG:/" $TEMP mv $TEMP $NAME fi done llvm-svn: 186280
2013-07-14 14:24:09 +08:00
; SSE1-LABEL: t3:
Revert r122955. It seems using movups to lower memcpy can cause massive regression (even on Nehalem) in edge cases. I also didn't see any real performance benefit. llvm-svn: 123015
2011-01-08 03:35:30 +08:00
; SSE1: movl
; SSE1: movl
; SSE1: movl
; SSE1: movl
; SSE1: movl
; SSE1: movl
; SSE1: movl
; SSE1: movl
; SSE1: movl
; SSE1: movl
Fix sdisel memcpy, memset, memmove lowering: 1. Makes it possible to lower with floating point loads and stores. 2. Avoid unaligned loads / stores unless it's fast. 3. Fix some memcpy lowering logic bug related to when to optimize a load from constant string into a constant. 4. Adjust x86 memcpy lowering threshold to make it more sane. 5. Fix x86 target hook so it uses vector and floating point memory ops more effectively. rdar://7774704 llvm-svn: 100090
2010-04-01 14:04:33 +08:00
Mass update to CodeGen tests to use CHECK-LABEL for labels corresponding to function definitions for more informative error messages. No functionality change and all updated tests passed locally. This update was done with the following bash script: find test/CodeGen -name "*.ll" | \ while read NAME; do echo "$NAME" if ! grep -q "^; *RUN: *llc.*debug" $NAME; then TEMP=`mktemp -t temp` cp $NAME $TEMP sed -n "s/^define [^@]*@\([A-Za-z0-9_]*\)(.*$/\1/p" < $NAME | \ while read FUNC; do sed -i '' "s/;\(.*\)\([A-Za-z0-9_-]*\):\( *\)$FUNC: *\$/;\1\2-LABEL:\3$FUNC:/g" $TEMP done sed -i '' "s/;\(.*\)-LABEL-LABEL:/;\1-LABEL:/" $TEMP sed -i '' "s/;\(.*\)-NEXT-LABEL:/;\1-NEXT:/" $TEMP sed -i '' "s/;\(.*\)-NOT-LABEL:/;\1-NOT:/" $TEMP sed -i '' "s/;\(.*\)-DAG-LABEL:/;\1-DAG:/" $TEMP mv $TEMP $NAME fi done llvm-svn: 186280
2013-07-14 14:24:09 +08:00
; NOSSE-LABEL: t3:
Fix sdisel memcpy, memset, memmove lowering: 1. Makes it possible to lower with floating point loads and stores. 2. Avoid unaligned loads / stores unless it's fast. 3. Fix some memcpy lowering logic bug related to when to optimize a load from constant string into a constant. 4. Adjust x86 memcpy lowering threshold to make it more sane. 5. Fix x86 target hook so it uses vector and floating point memory ops more effectively. rdar://7774704 llvm-svn: 100090
2010-04-01 14:04:33 +08:00
; NOSSE: movl
; NOSSE: movl
; NOSSE: movl
; NOSSE: movl
; NOSSE: movl
; NOSSE: movl
; NOSSE: movl
; NOSSE: movl
; NOSSE: movl
; NOSSE: movl
In 64-bit mode, use i64 to lower memcpy / memset instead of f64. llvm-svn: 100137
2010-04-02 04:27:45 +08:00
Mass update to CodeGen tests to use CHECK-LABEL for labels corresponding to function definitions for more informative error messages. No functionality change and all updated tests passed locally. This update was done with the following bash script: find test/CodeGen -name "*.ll" | \ while read NAME; do echo "$NAME" if ! grep -q "^; *RUN: *llc.*debug" $NAME; then TEMP=`mktemp -t temp` cp $NAME $TEMP sed -n "s/^define [^@]*@\([A-Za-z0-9_]*\)(.*$/\1/p" < $NAME | \ while read FUNC; do sed -i '' "s/;\(.*\)\([A-Za-z0-9_-]*\):\( *\)$FUNC: *\$/;\1\2-LABEL:\3$FUNC:/g" $TEMP done sed -i '' "s/;\(.*\)-LABEL-LABEL:/;\1-LABEL:/" $TEMP sed -i '' "s/;\(.*\)-NEXT-LABEL:/;\1-NEXT:/" $TEMP sed -i '' "s/;\(.*\)-NOT-LABEL:/;\1-NOT:/" $TEMP sed -i '' "s/;\(.*\)-DAG-LABEL:/;\1-DAG:/" $TEMP mv $TEMP $NAME fi done llvm-svn: 186280
2013-07-14 14:24:09 +08:00
; X86-64-LABEL: t3:
Revert r122955. It seems using movups to lower memcpy can cause massive regression (even on Nehalem) in edge cases. I also didn't see any real performance benefit. llvm-svn: 123015
2011-01-08 03:35:30 +08:00
; X86-64: movq (%rsi), %rax
; X86-64: movq 8(%rsi), %rcx
; X86-64: movq %rcx, 8(%rdi)
; X86-64: movq %rax, (%rdi)
Fix sdisel memcpy, memset, memmove lowering: 1. Makes it possible to lower with floating point loads and stores. 2. Avoid unaligned loads / stores unless it's fast. 3. Fix some memcpy lowering logic bug related to when to optimize a load from constant string into a constant. 4. Adjust x86 memcpy lowering threshold to make it more sane. 5. Fix x86 target hook so it uses vector and floating point memory ops more effectively. rdar://7774704 llvm-svn: 100090
2010-04-01 14:04:33 +08:00
%tmp2 = bitcast %struct.s0* %a to i8* ; <i8*> [#uses=1]
%tmp3 = bitcast %struct.s0* %b to i8* ; <i8*> [#uses=1]
rip out a ton of intrinsic modernization logic from AutoUpgrade.cpp, which is for pre-2.9 bitcode files. We keep x86 unaligned loads, movnt, crc32, and the target indep prefetch change. As usual, updating the testsuite is a PITA. llvm-svn: 133337
2011-06-18 14:05:24 +08:00
tail call void @llvm.memcpy.p0i8.p0i8.i32(i8* %tmp2, i8* %tmp3, i32 16, i32 8, i1 false)
Fix sdisel memcpy, memset, memmove lowering: 1. Makes it possible to lower with floating point loads and stores. 2. Avoid unaligned loads / stores unless it's fast. 3. Fix some memcpy lowering logic bug related to when to optimize a load from constant string into a constant. 4. Adjust x86 memcpy lowering threshold to make it more sane. 5. Fix x86 target hook so it uses vector and floating point memory ops more effectively. rdar://7774704 llvm-svn: 100090
2010-04-01 14:04:33 +08:00
ret void
}
Avoid using f64 to lower memcpy from constant string. It's cheaper to use i32 store of immediates. llvm-svn: 100751
2010-04-08 15:37:57 +08:00
define void @t4() nounwind {
entry:
Mass update to CodeGen tests to use CHECK-LABEL for labels corresponding to function definitions for more informative error messages. No functionality change and all updated tests passed locally. This update was done with the following bash script: find test/CodeGen -name "*.ll" | \ while read NAME; do echo "$NAME" if ! grep -q "^; *RUN: *llc.*debug" $NAME; then TEMP=`mktemp -t temp` cp $NAME $TEMP sed -n "s/^define [^@]*@\([A-Za-z0-9_]*\)(.*$/\1/p" < $NAME | \ while read FUNC; do sed -i '' "s/;\(.*\)\([A-Za-z0-9_-]*\):\( *\)$FUNC: *\$/;\1\2-LABEL:\3$FUNC:/g" $TEMP done sed -i '' "s/;\(.*\)-LABEL-LABEL:/;\1-LABEL:/" $TEMP sed -i '' "s/;\(.*\)-NEXT-LABEL:/;\1-NEXT:/" $TEMP sed -i '' "s/;\(.*\)-NOT-LABEL:/;\1-NOT:/" $TEMP sed -i '' "s/;\(.*\)-DAG-LABEL:/;\1-DAG:/" $TEMP mv $TEMP $NAME fi done llvm-svn: 186280
2013-07-14 14:24:09 +08:00
; SSE2-Darwin-LABEL: t4:
When lowering memcpys to loads and stores, make sure we don't promote alignments past the natural stack alignment. llvm-svn: 174085
2013-02-01 04:23:43 +08:00
; SSE2-Darwin: movw $120
; SSE2-Darwin: movl $2021161080
; SSE2-Darwin: movl $2021161080
; SSE2-Darwin: movl $2021161080
; SSE2-Darwin: movl $2021161080
; SSE2-Darwin: movl $2021161080
; SSE2-Darwin: movl $2021161080
; SSE2-Darwin: movl $2021161080
Mass update to CodeGen tests to use CHECK-LABEL for labels corresponding to function definitions for more informative error messages. No functionality change and all updated tests passed locally. This update was done with the following bash script: find test/CodeGen -name "*.ll" | \ while read NAME; do echo "$NAME" if ! grep -q "^; *RUN: *llc.*debug" $NAME; then TEMP=`mktemp -t temp` cp $NAME $TEMP sed -n "s/^define [^@]*@\([A-Za-z0-9_]*\)(.*$/\1/p" < $NAME | \ while read FUNC; do sed -i '' "s/;\(.*\)\([A-Za-z0-9_-]*\):\( *\)$FUNC: *\$/;\1\2-LABEL:\3$FUNC:/g" $TEMP done sed -i '' "s/;\(.*\)-LABEL-LABEL:/;\1-LABEL:/" $TEMP sed -i '' "s/;\(.*\)-NEXT-LABEL:/;\1-NEXT:/" $TEMP sed -i '' "s/;\(.*\)-NOT-LABEL:/;\1-NOT:/" $TEMP sed -i '' "s/;\(.*\)-DAG-LABEL:/;\1-DAG:/" $TEMP mv $TEMP $NAME fi done llvm-svn: 186280
2013-07-14 14:24:09 +08:00
; SSE2-Mingw32-LABEL: t4:
When lowering memcpys to loads and stores, make sure we don't promote alignments past the natural stack alignment. llvm-svn: 174085
2013-02-01 04:23:43 +08:00
; SSE2-Mingw32: movw $120
; SSE2-Mingw32: movl $2021161080
; SSE2-Mingw32: movl $2021161080
; SSE2-Mingw32: movl $2021161080
; SSE2-Mingw32: movl $2021161080
; SSE2-Mingw32: movl $2021161080
; SSE2-Mingw32: movl $2021161080
; SSE2-Mingw32: movl $2021161080
Revert r122955. It seems using movups to lower memcpy can cause massive regression (even on Nehalem) in edge cases. I also didn't see any real performance benefit. llvm-svn: 123015
2011-01-08 03:35:30 +08:00
Mass update to CodeGen tests to use CHECK-LABEL for labels corresponding to function definitions for more informative error messages. No functionality change and all updated tests passed locally. This update was done with the following bash script: find test/CodeGen -name "*.ll" | \ while read NAME; do echo "$NAME" if ! grep -q "^; *RUN: *llc.*debug" $NAME; then TEMP=`mktemp -t temp` cp $NAME $TEMP sed -n "s/^define [^@]*@\([A-Za-z0-9_]*\)(.*$/\1/p" < $NAME | \ while read FUNC; do sed -i '' "s/;\(.*\)\([A-Za-z0-9_-]*\):\( *\)$FUNC: *\$/;\1\2-LABEL:\3$FUNC:/g" $TEMP done sed -i '' "s/;\(.*\)-LABEL-LABEL:/;\1-LABEL:/" $TEMP sed -i '' "s/;\(.*\)-NEXT-LABEL:/;\1-NEXT:/" $TEMP sed -i '' "s/;\(.*\)-NOT-LABEL:/;\1-NOT:/" $TEMP sed -i '' "s/;\(.*\)-DAG-LABEL:/;\1-DAG:/" $TEMP mv $TEMP $NAME fi done llvm-svn: 186280
2013-07-14 14:24:09 +08:00
; SSE1-LABEL: t4:
Revert r122955. It seems using movups to lower memcpy can cause massive regression (even on Nehalem) in edge cases. I also didn't see any real performance benefit. llvm-svn: 123015
2011-01-08 03:35:30 +08:00
; SSE1: movw $120
; SSE1: movl $2021161080
; SSE1: movl $2021161080
; SSE1: movl $2021161080
; SSE1: movl $2021161080
; SSE1: movl $2021161080
; SSE1: movl $2021161080
; SSE1: movl $2021161080
Avoid using f64 to lower memcpy from constant string. It's cheaper to use i32 store of immediates. llvm-svn: 100751
2010-04-08 15:37:57 +08:00
Mass update to CodeGen tests to use CHECK-LABEL for labels corresponding to function definitions for more informative error messages. No functionality change and all updated tests passed locally. This update was done with the following bash script: find test/CodeGen -name "*.ll" | \ while read NAME; do echo "$NAME" if ! grep -q "^; *RUN: *llc.*debug" $NAME; then TEMP=`mktemp -t temp` cp $NAME $TEMP sed -n "s/^define [^@]*@\([A-Za-z0-9_]*\)(.*$/\1/p" < $NAME | \ while read FUNC; do sed -i '' "s/;\(.*\)\([A-Za-z0-9_-]*\):\( *\)$FUNC: *\$/;\1\2-LABEL:\3$FUNC:/g" $TEMP done sed -i '' "s/;\(.*\)-LABEL-LABEL:/;\1-LABEL:/" $TEMP sed -i '' "s/;\(.*\)-NEXT-LABEL:/;\1-NEXT:/" $TEMP sed -i '' "s/;\(.*\)-NOT-LABEL:/;\1-NOT:/" $TEMP sed -i '' "s/;\(.*\)-DAG-LABEL:/;\1-DAG:/" $TEMP mv $TEMP $NAME fi done llvm-svn: 186280
2013-07-14 14:24:09 +08:00
; NOSSE-LABEL: t4:
Avoid using f64 to lower memcpy from constant string. It's cheaper to use i32 store of immediates. llvm-svn: 100751
2010-04-08 15:37:57 +08:00
; NOSSE: movw $120
; NOSSE: movl $2021161080
; NOSSE: movl $2021161080
; NOSSE: movl $2021161080
; NOSSE: movl $2021161080
; NOSSE: movl $2021161080
; NOSSE: movl $2021161080
; NOSSE: movl $2021161080
[x86] fix allowsMisalignedMemoryAccesses() for 8-byte and smaller accesses This is a continuation of the fix from: http://reviews.llvm.org/D10662 and discussion in: http://reviews.llvm.org/D12154 Here, we distinguish slow unaligned SSE (128-bit) accesses from slow unaligned scalar (64-bit and under) accesses. Other lowering (eg, getOptimalMemOpType) assumes that unaligned scalar accesses are always ok, so this changes allowsMisalignedMemoryAccesses() to match that behavior. Differential Revision: http://reviews.llvm.org/D12543 llvm-svn: 246658
2015-09-02 23:42:49 +08:00
;;; TODO: (1) Some of the loads and stores are certainly unaligned and (2) the first load and first
;;; store overlap with the second load and second store respectively.
;;;
;;; Is either of the sequences ideal?
Mass update to CodeGen tests to use CHECK-LABEL for labels corresponding to function definitions for more informative error messages. No functionality change and all updated tests passed locally. This update was done with the following bash script: find test/CodeGen -name "*.ll" | \ while read NAME; do echo "$NAME" if ! grep -q "^; *RUN: *llc.*debug" $NAME; then TEMP=`mktemp -t temp` cp $NAME $TEMP sed -n "s/^define [^@]*@\([A-Za-z0-9_]*\)(.*$/\1/p" < $NAME | \ while read FUNC; do sed -i '' "s/;\(.*\)\([A-Za-z0-9_-]*\):\( *\)$FUNC: *\$/;\1\2-LABEL:\3$FUNC:/g" $TEMP done sed -i '' "s/;\(.*\)-LABEL-LABEL:/;\1-LABEL:/" $TEMP sed -i '' "s/;\(.*\)-NEXT-LABEL:/;\1-NEXT:/" $TEMP sed -i '' "s/;\(.*\)-NOT-LABEL:/;\1-NOT:/" $TEMP sed -i '' "s/;\(.*\)-DAG-LABEL:/;\1-DAG:/" $TEMP mv $TEMP $NAME fi done llvm-svn: 186280
2013-07-14 14:24:09 +08:00
; X86-64-LABEL: t4:
[x86] fix allowsMisalignedMemoryAccesses() for 8-byte and smaller accesses This is a continuation of the fix from: http://reviews.llvm.org/D10662 and discussion in: http://reviews.llvm.org/D12154 Here, we distinguish slow unaligned SSE (128-bit) accesses from slow unaligned scalar (64-bit and under) accesses. Other lowering (eg, getOptimalMemOpType) assumes that unaligned scalar accesses are always ok, so this changes allowsMisalignedMemoryAccesses() to match that behavior. Differential Revision: http://reviews.llvm.org/D12543 llvm-svn: 246658
2015-09-02 23:42:49 +08:00
; X86-64: movabsq $33909456017848440, %rax ## imm = 0x78787878787878
; X86-64: movq %rax, -10(%rsp)
; X86-64: movabsq $8680820740569200760, %rax ## imm = 0x7878787878787878
; X86-64: movq %rax, -16(%rsp)
; X86-64: movq %rax, -24(%rsp)
; X86-64: movq %rax, -32(%rsp)
make current codegen visible in the checks, so we can decide if it's right llvm-svn: 245120
2015-08-15 07:03:01 +08:00
; NHM_64-LABEL: t4:
; NHM_64: movups _.str2+14(%rip), %xmm0
; NHM_64: movups %xmm0, -26(%rsp)
; NHM_64: movups _.str2(%rip), %xmm0
; NHM_64: movaps %xmm0, -40(%rsp)
Avoid using f64 to lower memcpy from constant string. It's cheaper to use i32 store of immediates. llvm-svn: 100751
2010-04-08 15:37:57 +08:00
%tmp1 = alloca [30 x i8]
%tmp2 = bitcast [30 x i8]* %tmp1 to i8*
[opaque pointer type] Add textual IR support for explicit type parameter to gep operator Similar to gep (r230786) and load (r230794) changes. Similar migration script can be used to update test cases, which successfully migrated all of LLVM and Polly, but about 4 test cases needed manually changes in Clang. (this script will read the contents of stdin and massage it into stdout - wrap it in the 'apply.sh' script shown in previous commits + xargs to apply it over a large set of test cases) import fileinput import sys import re rep = re.compile(r"(getelementptr(?:\s+inbounds)?\s*\()((<\d*\s+x\s+)?([^@]*?)(|\s*addrspace\(\d+\))\s*\*(?(3)>)\s*)(?=$|%|@|null|undef|blockaddress|getelementptr|addrspacecast|bitcast|inttoptr|zeroinitializer|<|\[\[[a-zA-Z]|\{\{)", re.MULTILINE | re.DOTALL) def conv(match): line = match.group(1) line += match.group(4) line += ", " line += match.group(2) return line line = sys.stdin.read() off = 0 for match in re.finditer(rep, line): sys.stdout.write(line[off:match.start()]) sys.stdout.write(conv(match)) off = match.end() sys.stdout.write(line[off:]) llvm-svn: 232184
2015-03-14 02:20:45 +08:00
call void @llvm.memcpy.p0i8.p0i8.i32(i8* %tmp2, i8* getelementptr inbounds ([30 x i8], [30 x i8]* @.str2, i32 0, i32 0), i32 30, i32 1, i1 false)
Avoid using f64 to lower memcpy from constant string. It's cheaper to use i32 store of immediates. llvm-svn: 100751
2010-04-08 15:37:57 +08:00
unreachable
}
rip out a ton of intrinsic modernization logic from AutoUpgrade.cpp, which is for pre-2.9 bitcode files. We keep x86 unaligned loads, movnt, crc32, and the target indep prefetch change. As usual, updating the testsuite is a PITA. llvm-svn: 133337
2011-06-18 14:05:24 +08:00
declare void @llvm.memcpy.p0i8.p0i8.i32(i8* nocapture, i8* nocapture, i32, i32, i1) nounwind