llvm-project/llvm/test/CodeGen/AMDGPU/combine_vloads.ll

43 lines
2.0 KiB
LLVM
Raw Normal View History

; RUN: llc -march=r600 -mcpu=cypress < %s | FileCheck -check-prefix=EG %s
;
; kernel void combine_vloads(global char8* src, global char8* result) {
; for (int i = 0; i < 1024; ++i)
; result[i] = src[0] + src[1] + src[2] + src[3];
; }
;
; 128-bit loads instead of many 8-bit
; EG-LABEL: {{^}}combine_vloads:
; EG: VTX_READ_128
; EG: VTX_READ_128
define amdgpu_kernel void @combine_vloads(<8 x i8> addrspace(1)* nocapture %src, <8 x i8> addrspace(1)* nocapture %result) nounwind {
entry:
br label %for.body
for.exit: ; preds = %for.body
ret void
for.body: ; preds = %for.body, %entry
%i.01 = phi i32 [ 0, %entry ], [ %tmp19, %for.body ]
%arrayidx_v4 = bitcast <8 x i8> addrspace(1)* %src to <32 x i8> addrspace(1)*
%0 = bitcast <32 x i8> addrspace(1)* %arrayidx_v4 to <8 x i32> addrspace(1)*
%vecload2 = load <8 x i32>, <8 x i32> addrspace(1)* %0, align 32
%1 = bitcast <8 x i32> %vecload2 to <32 x i8>
%tmp5 = shufflevector <32 x i8> %1, <32 x i8> undef, <8 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7>
%tmp8 = shufflevector <32 x i8> %1, <32 x i8> undef, <8 x i32> <i32 8, i32 9, i32 10, i32 11, i32 12, i32 13, i32 14, i32 15>
%tmp9 = add nsw <8 x i8> %tmp5, %tmp8
%tmp12 = shufflevector <32 x i8> %1, <32 x i8> undef, <8 x i32> <i32 16, i32 17, i32 18, i32 19, i32 20, i32 21, i32 22, i32 23>
%tmp13 = add nsw <8 x i8> %tmp9, %tmp12
%tmp16 = shufflevector <32 x i8> %1, <32 x i8> undef, <8 x i32> <i32 24, i32 25, i32 26, i32 27, i32 28, i32 29, i32 30, i32 31>
%tmp17 = add nsw <8 x i8> %tmp13, %tmp16
[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
%scevgep = getelementptr <8 x i8>, <8 x i8> addrspace(1)* %result, i32 %i.01
%2 = bitcast <8 x i8> %tmp17 to <2 x i32>
%3 = bitcast <8 x i8> addrspace(1)* %scevgep to <2 x i32> addrspace(1)*
store <2 x i32> %2, <2 x i32> addrspace(1)* %3, align 8
%tmp19 = add nsw i32 %i.01, 1
%exitcond = icmp eq i32 %tmp19, 1024
br i1 %exitcond, label %for.exit, label %for.body
}