mirror of
https://github.com/c64scene-ar/llvm-6502.git
synced 2024-12-15 04:30:12 +00:00
198d8baafb
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
181 lines
4.6 KiB
LLVM
181 lines
4.6 KiB
LLVM
; Test additions between an i64 and a zero-extended i32.
|
|
;
|
|
; RUN: llc < %s -mtriple=s390x-linux-gnu | FileCheck %s
|
|
|
|
declare i64 @foo()
|
|
|
|
; Check ALGFR.
|
|
define i64 @f1(i64 %a, i32 %b) {
|
|
; CHECK-LABEL: f1:
|
|
; CHECK: algfr %r2, %r3
|
|
; CHECK: br %r14
|
|
%bext = zext i32 %b to i64
|
|
%add = add i64 %a, %bext
|
|
ret i64 %add
|
|
}
|
|
|
|
; Check ALGF with no displacement.
|
|
define i64 @f2(i64 %a, i32 *%src) {
|
|
; CHECK-LABEL: f2:
|
|
; CHECK: algf %r2, 0(%r3)
|
|
; CHECK: br %r14
|
|
%b = load i32 *%src
|
|
%bext = zext i32 %b to i64
|
|
%add = add i64 %a, %bext
|
|
ret i64 %add
|
|
}
|
|
|
|
; Check the high end of the aligned ALGF range.
|
|
define i64 @f3(i64 %a, i32 *%src) {
|
|
; CHECK-LABEL: f3:
|
|
; CHECK: algf %r2, 524284(%r3)
|
|
; CHECK: br %r14
|
|
%ptr = getelementptr i32, i32 *%src, i64 131071
|
|
%b = load i32 *%ptr
|
|
%bext = zext i32 %b to i64
|
|
%add = add i64 %a, %bext
|
|
ret i64 %add
|
|
}
|
|
|
|
; Check the next word up, which needs separate address logic.
|
|
; Other sequences besides this one would be OK.
|
|
define i64 @f4(i64 %a, i32 *%src) {
|
|
; CHECK-LABEL: f4:
|
|
; CHECK: agfi %r3, 524288
|
|
; CHECK: algf %r2, 0(%r3)
|
|
; CHECK: br %r14
|
|
%ptr = getelementptr i32, i32 *%src, i64 131072
|
|
%b = load i32 *%ptr
|
|
%bext = zext i32 %b to i64
|
|
%add = add i64 %a, %bext
|
|
ret i64 %add
|
|
}
|
|
|
|
; Check the high end of the negative aligned ALGF range.
|
|
define i64 @f5(i64 %a, i32 *%src) {
|
|
; CHECK-LABEL: f5:
|
|
; CHECK: algf %r2, -4(%r3)
|
|
; CHECK: br %r14
|
|
%ptr = getelementptr i32, i32 *%src, i64 -1
|
|
%b = load i32 *%ptr
|
|
%bext = zext i32 %b to i64
|
|
%add = add i64 %a, %bext
|
|
ret i64 %add
|
|
}
|
|
|
|
; Check the low end of the ALGF range.
|
|
define i64 @f6(i64 %a, i32 *%src) {
|
|
; CHECK-LABEL: f6:
|
|
; CHECK: algf %r2, -524288(%r3)
|
|
; CHECK: br %r14
|
|
%ptr = getelementptr i32, i32 *%src, i64 -131072
|
|
%b = load i32 *%ptr
|
|
%bext = zext i32 %b to i64
|
|
%add = add i64 %a, %bext
|
|
ret i64 %add
|
|
}
|
|
|
|
; Check the next word down, which needs separate address logic.
|
|
; Other sequences besides this one would be OK.
|
|
define i64 @f7(i64 %a, i32 *%src) {
|
|
; CHECK-LABEL: f7:
|
|
; CHECK: agfi %r3, -524292
|
|
; CHECK: algf %r2, 0(%r3)
|
|
; CHECK: br %r14
|
|
%ptr = getelementptr i32, i32 *%src, i64 -131073
|
|
%b = load i32 *%ptr
|
|
%bext = zext i32 %b to i64
|
|
%add = add i64 %a, %bext
|
|
ret i64 %add
|
|
}
|
|
|
|
; Check that ALGF allows an index.
|
|
define i64 @f8(i64 %a, i64 %src, i64 %index) {
|
|
; CHECK-LABEL: f8:
|
|
; CHECK: algf %r2, 524284({{%r4,%r3|%r3,%r4}})
|
|
; CHECK: br %r14
|
|
%add1 = add i64 %src, %index
|
|
%add2 = add i64 %add1, 524284
|
|
%ptr = inttoptr i64 %add2 to i32 *
|
|
%b = load i32 *%ptr
|
|
%bext = zext i32 %b to i64
|
|
%add = add i64 %a, %bext
|
|
ret i64 %add
|
|
}
|
|
|
|
; Check that additions of spilled values can use ALGF rather than ALGFR.
|
|
define i64 @f9(i32 *%ptr0) {
|
|
; CHECK-LABEL: f9:
|
|
; CHECK: brasl %r14, foo@PLT
|
|
; CHECK: algf %r2, 16{{[04]}}(%r15)
|
|
; CHECK: br %r14
|
|
%ptr1 = getelementptr i32, i32 *%ptr0, i64 2
|
|
%ptr2 = getelementptr i32, i32 *%ptr0, i64 4
|
|
%ptr3 = getelementptr i32, i32 *%ptr0, i64 6
|
|
%ptr4 = getelementptr i32, i32 *%ptr0, i64 8
|
|
%ptr5 = getelementptr i32, i32 *%ptr0, i64 10
|
|
%ptr6 = getelementptr i32, i32 *%ptr0, i64 12
|
|
%ptr7 = getelementptr i32, i32 *%ptr0, i64 14
|
|
%ptr8 = getelementptr i32, i32 *%ptr0, i64 16
|
|
%ptr9 = getelementptr i32, i32 *%ptr0, i64 18
|
|
|
|
%val0 = load i32 *%ptr0
|
|
%val1 = load i32 *%ptr1
|
|
%val2 = load i32 *%ptr2
|
|
%val3 = load i32 *%ptr3
|
|
%val4 = load i32 *%ptr4
|
|
%val5 = load i32 *%ptr5
|
|
%val6 = load i32 *%ptr6
|
|
%val7 = load i32 *%ptr7
|
|
%val8 = load i32 *%ptr8
|
|
%val9 = load i32 *%ptr9
|
|
|
|
%frob0 = add i32 %val0, 100
|
|
%frob1 = add i32 %val1, 100
|
|
%frob2 = add i32 %val2, 100
|
|
%frob3 = add i32 %val3, 100
|
|
%frob4 = add i32 %val4, 100
|
|
%frob5 = add i32 %val5, 100
|
|
%frob6 = add i32 %val6, 100
|
|
%frob7 = add i32 %val7, 100
|
|
%frob8 = add i32 %val8, 100
|
|
%frob9 = add i32 %val9, 100
|
|
|
|
store i32 %frob0, i32 *%ptr0
|
|
store i32 %frob1, i32 *%ptr1
|
|
store i32 %frob2, i32 *%ptr2
|
|
store i32 %frob3, i32 *%ptr3
|
|
store i32 %frob4, i32 *%ptr4
|
|
store i32 %frob5, i32 *%ptr5
|
|
store i32 %frob6, i32 *%ptr6
|
|
store i32 %frob7, i32 *%ptr7
|
|
store i32 %frob8, i32 *%ptr8
|
|
store i32 %frob9, i32 *%ptr9
|
|
|
|
%ret = call i64 @foo()
|
|
|
|
%ext0 = zext i32 %frob0 to i64
|
|
%ext1 = zext i32 %frob1 to i64
|
|
%ext2 = zext i32 %frob2 to i64
|
|
%ext3 = zext i32 %frob3 to i64
|
|
%ext4 = zext i32 %frob4 to i64
|
|
%ext5 = zext i32 %frob5 to i64
|
|
%ext6 = zext i32 %frob6 to i64
|
|
%ext7 = zext i32 %frob7 to i64
|
|
%ext8 = zext i32 %frob8 to i64
|
|
%ext9 = zext i32 %frob9 to i64
|
|
|
|
%add0 = add i64 %ret, %ext0
|
|
%add1 = add i64 %add0, %ext1
|
|
%add2 = add i64 %add1, %ext2
|
|
%add3 = add i64 %add2, %ext3
|
|
%add4 = add i64 %add3, %ext4
|
|
%add5 = add i64 %add4, %ext5
|
|
%add6 = add i64 %add5, %ext6
|
|
%add7 = add i64 %add6, %ext7
|
|
%add8 = add i64 %add7, %ext8
|
|
%add9 = add i64 %add8, %ext9
|
|
|
|
ret i64 %add9
|
|
}
|