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/PowerPC/split-index-tc.ll

83 lines
3.3 KiB
LLVM
Raw Normal View History

Enable splitting indexing from loads with TargetConstants When I recommitted r208640 (in r216898) I added an exclusion for TargetConstant offsets, as there is no guarantee that a backend can handle them on generic ADDs (even if it generates them during address-mode matching) -- and, specifically, applying this transformation directly with TargetConstants caused a self-hosting failure on PPC64. Ignoring all TargetConstants, however, is less than ideal. Instead, for non-opaque constants, we can convert them into regular constants for use with the generated ADD (or SUB). llvm-svn: 216908
2014-09-03 00:05:23 +08:00
; RUN: llc -mcpu=pwr7 < %s | FileCheck %s
target datalayout = "E-m:e-i64:64-n32:64"
target triple = "powerpc64-unknown-linux-gnu"
%"class.llvm::MachineOperand" = type { i8, [3 x i8], i64, i64*, i64 }
; Function Attrs: nounwind
define void @_ZN4llvm17ScheduleDAGInstrs14addPhysRegDepsEPNS_5SUnitEj() #0 align 2 {
; If we were able to split out the indexing, the load with update should be
; removed (resulting in a nearly-empty output).
; CHECK-LABEL: @_ZN4llvm17ScheduleDAGInstrs14addPhysRegDepsEPNS_5SUnitEj
; CHECK-NOT: lhzu
entry:
[opaque pointer type] Add textual IR support for explicit type parameter to load instruction Essentially the same as the GEP change in r230786. A similar migration script can be used to update test cases, though a few more test case improvements/changes were required this time around: (r229269-r229278) import fileinput import sys import re pat = re.compile(r"((?:=|:|^)\s*load (?:atomic )?(?:volatile )?(.*?))(| addrspace\(\d+\) *)\*($| *(?:%|@|null|undef|blockaddress|getelementptr|addrspacecast|bitcast|inttoptr|\[\[[a-zA-Z]|\{\{).*$)") for line in sys.stdin: sys.stdout.write(re.sub(pat, r"\1, \2\3*\4", line)) Reviewers: rafael, dexonsmith, grosser Differential Revision: http://reviews.llvm.org/D7649 llvm-svn: 230794
2015-02-28 05:17:42 +08:00
%0 = load %"class.llvm::MachineOperand"*, %"class.llvm::MachineOperand"** undef, align 8
Enable splitting indexing from loads with TargetConstants When I recommitted r208640 (in r216898) I added an exclusion for TargetConstant offsets, as there is no guarantee that a backend can handle them on generic ADDs (even if it generates them during address-mode matching) -- and, specifically, applying this transformation directly with TargetConstants caused a self-hosting failure on PPC64. Ignoring all TargetConstants, however, is less than ideal. Instead, for non-opaque constants, we can convert them into regular constants for use with the generated ADD (or SUB). llvm-svn: 216908
2014-09-03 00:05:23 +08:00
br i1 undef, label %_ZNK4llvm14MachineOperand6getRegEv.exit, label %cond.false.i123
cond.false.i123: ; preds = %_ZN4llvm12MachineInstr10getOperandEj.exit
unreachable
_ZNK4llvm14MachineOperand6getRegEv.exit: ; preds = %_ZN4llvm12MachineInstr10getOperandEj.exit
[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
%IsDef.i = getelementptr inbounds %"class.llvm::MachineOperand", %"class.llvm::MachineOperand"* %0, i64 undef, i32 1
Enable splitting indexing from loads with TargetConstants When I recommitted r208640 (in r216898) I added an exclusion for TargetConstant offsets, as there is no guarantee that a backend can handle them on generic ADDs (even if it generates them during address-mode matching) -- and, specifically, applying this transformation directly with TargetConstants caused a self-hosting failure on PPC64. Ignoring all TargetConstants, however, is less than ideal. Instead, for non-opaque constants, we can convert them into regular constants for use with the generated ADD (or SUB). llvm-svn: 216908
2014-09-03 00:05:23 +08:00
%1 = bitcast [3 x i8]* %IsDef.i to i24*
[opaque pointer type] Add textual IR support for explicit type parameter to load instruction Essentially the same as the GEP change in r230786. A similar migration script can be used to update test cases, though a few more test case improvements/changes were required this time around: (r229269-r229278) import fileinput import sys import re pat = re.compile(r"((?:=|:|^)\s*load (?:atomic )?(?:volatile )?(.*?))(| addrspace\(\d+\) *)\*($| *(?:%|@|null|undef|blockaddress|getelementptr|addrspacecast|bitcast|inttoptr|\[\[[a-zA-Z]|\{\{).*$)") for line in sys.stdin: sys.stdout.write(re.sub(pat, r"\1, \2\3*\4", line)) Reviewers: rafael, dexonsmith, grosser Differential Revision: http://reviews.llvm.org/D7649 llvm-svn: 230794
2015-02-28 05:17:42 +08:00
%bf.load.i = load i24, i24* %1, align 1
Enable splitting indexing from loads with TargetConstants When I recommitted r208640 (in r216898) I added an exclusion for TargetConstant offsets, as there is no guarantee that a backend can handle them on generic ADDs (even if it generates them during address-mode matching) -- and, specifically, applying this transformation directly with TargetConstants caused a self-hosting failure on PPC64. Ignoring all TargetConstants, however, is less than ideal. Instead, for non-opaque constants, we can convert them into regular constants for use with the generated ADD (or SUB). llvm-svn: 216908
2014-09-03 00:05:23 +08:00
%2 = and i24 %bf.load.i, 128
br i1 undef, label %for.cond.cleanup, label %for.body.lr.ph
for.body.lr.ph: ; preds = %_ZNK4llvm14MachineOperand6getRegEv.exit
%3 = zext i24 %2 to i32
br i1 undef, label %cond.false.i134, label %_ZNK4llvm18MCRegAliasIteratordeEv.exit
for.cond.cleanup: ; preds = %_ZNK4llvm14MachineOperand6getRegEv.exit
br i1 undef, label %_ZNK4llvm14MachineOperand5isDefEv.exit, label %cond.false.i129
cond.false.i129: ; preds = %for.cond.cleanup
unreachable
_ZNK4llvm14MachineOperand5isDefEv.exit: ; preds = %for.cond.cleanup
br i1 undef, label %_ZNK4llvm14MachineOperand6getRegEv.exit247, label %cond.false.i244
cond.false.i134: ; preds = %for.body.lr.ph
unreachable
_ZNK4llvm18MCRegAliasIteratordeEv.exit: ; preds = %for.body.lr.ph
unreachable
cond.false.i244: ; preds = %_ZNK4llvm14MachineOperand5isDefEv.exit
unreachable
_ZNK4llvm14MachineOperand6getRegEv.exit247: ; preds = %_ZNK4llvm14MachineOperand5isDefEv.exit
br i1 undef, label %if.then53, label %if.end55
if.then53: ; preds = %_ZNK4llvm14MachineOperand6getRegEv.exit247
unreachable
if.end55: ; preds = %_ZNK4llvm14MachineOperand6getRegEv.exit247
br i1 undef, label %_ZNK4llvm14MachineOperand6isDeadEv.exit262, label %cond.false.i257
cond.false.i257: ; preds = %if.end55
unreachable
_ZNK4llvm14MachineOperand6isDeadEv.exit262: ; preds = %if.end55
[opaque pointer type] Add textual IR support for explicit type parameter to load instruction Essentially the same as the GEP change in r230786. A similar migration script can be used to update test cases, though a few more test case improvements/changes were required this time around: (r229269-r229278) import fileinput import sys import re pat = re.compile(r"((?:=|:|^)\s*load (?:atomic )?(?:volatile )?(.*?))(| addrspace\(\d+\) *)\*($| *(?:%|@|null|undef|blockaddress|getelementptr|addrspacecast|bitcast|inttoptr|\[\[[a-zA-Z]|\{\{).*$)") for line in sys.stdin: sys.stdout.write(re.sub(pat, r"\1, \2\3*\4", line)) Reviewers: rafael, dexonsmith, grosser Differential Revision: http://reviews.llvm.org/D7649 llvm-svn: 230794
2015-02-28 05:17:42 +08:00
%bf.load.i259 = load i24, i24* %1, align 1
Enable splitting indexing from loads with TargetConstants When I recommitted r208640 (in r216898) I added an exclusion for TargetConstant offsets, as there is no guarantee that a backend can handle them on generic ADDs (even if it generates them during address-mode matching) -- and, specifically, applying this transformation directly with TargetConstants caused a self-hosting failure on PPC64. Ignoring all TargetConstants, however, is less than ideal. Instead, for non-opaque constants, we can convert them into regular constants for use with the generated ADD (or SUB). llvm-svn: 216908
2014-09-03 00:05:23 +08:00
br i1 undef, label %if.then57, label %if.else59
if.then57: ; preds = %_ZNK4llvm14MachineOperand6isDeadEv.exit262
unreachable
if.else59: ; preds = %_ZNK4llvm14MachineOperand6isDeadEv.exit262
br i1 undef, label %if.end89, label %if.then62
if.then62: ; preds = %if.else59
unreachable
if.end89: ; preds = %if.else59
unreachable
}
attributes #0 = { nounwind }