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/AMDGPU/lds-output-queue.ll

100 lines
3.6 KiB
LLVM
Raw Normal View History

AMDGPU/R600: Replace barrier intrinsics llvm-svn: 275870
2016-07-19 02:34:59 +08:00
; RUN: llc -march=r600 -mcpu=redwood -verify-machineinstrs < %s | FileCheck %s
R600: Fix scheduling of instructions that use the LDS output queue The LDS output queue is accessed via the OQAP register. The OQAP register cannot be live across clauses, so if value is written to the output queue, it must be retrieved before the end of the clause. With the machine scheduler, we cannot statisfy this constraint, because it lacks proper alias analysis and it will mark some LDS accesses as having a chain dependency on vertex fetches. Since vertex fetches require a new clauses, the dependency may end up spiltting OQAP uses and defs so the end up in different clauses. See the lds-output-queue.ll test for a more detailed explanation. To work around this issue, we now combine the LDS read and the OQAP copy into one instruction and expand it after register allocation. This patch also adds some checks to the EmitClauseMarker pass, so that it doesn't end a clause with a value still in the output queue and removes AR.X and OQAP handling from the scheduler (AR.X uses and defs were already being expanded post-RA, so the scheduler will never see them). Reviewed-by: Vincent Lejeune <vljn at ovi.com> llvm-svn: 194755
2013-11-15 08:12:45 +08:00
;
; This test checks that the lds input queue will is empty at the end of
; the ALU clause.
R600: Call EmitFunctionHeader() in the AsmPrinter to populate the ELF symbol table llvm-svn: 218776
2014-10-02 01:15:17 +08:00
; CHECK-LABEL: {{^}}lds_input_queue:
R600: Fix scheduling of instructions that use the LDS output queue The LDS output queue is accessed via the OQAP register. The OQAP register cannot be live across clauses, so if value is written to the output queue, it must be retrieved before the end of the clause. With the machine scheduler, we cannot statisfy this constraint, because it lacks proper alias analysis and it will mark some LDS accesses as having a chain dependency on vertex fetches. Since vertex fetches require a new clauses, the dependency may end up spiltting OQAP uses and defs so the end up in different clauses. See the lds-output-queue.ll test for a more detailed explanation. To work around this issue, we now combine the LDS read and the OQAP copy into one instruction and expand it after register allocation. This patch also adds some checks to the EmitClauseMarker pass, so that it doesn't end a clause with a value still in the output queue and removes AR.X and OQAP handling from the scheduler (AR.X uses and defs were already being expanded post-RA, so the scheduler will never see them). Reviewed-by: Vincent Lejeune <vljn at ovi.com> llvm-svn: 194755
2013-11-15 08:12:45 +08:00
; CHECK: LDS_READ_RET * OQAP
; CHECK-NOT: ALU clause
; CHECK: MOV * T{{[0-9]\.[XYZW]}}, OQAP
R600: Error on initializer for LDS. Also give a proper error for other address spaces. llvm-svn: 221917
2014-11-14 03:56:13 +08:00
@local_mem = internal unnamed_addr addrspace(3) global [2 x i32] undef, align 4
R600: Fix scheduling of instructions that use the LDS output queue The LDS output queue is accessed via the OQAP register. The OQAP register cannot be live across clauses, so if value is written to the output queue, it must be retrieved before the end of the clause. With the machine scheduler, we cannot statisfy this constraint, because it lacks proper alias analysis and it will mark some LDS accesses as having a chain dependency on vertex fetches. Since vertex fetches require a new clauses, the dependency may end up spiltting OQAP uses and defs so the end up in different clauses. See the lds-output-queue.ll test for a more detailed explanation. To work around this issue, we now combine the LDS read and the OQAP copy into one instruction and expand it after register allocation. This patch also adds some checks to the EmitClauseMarker pass, so that it doesn't end a clause with a value still in the output queue and removes AR.X and OQAP handling from the scheduler (AR.X uses and defs were already being expanded post-RA, so the scheduler will never see them). Reviewed-by: Vincent Lejeune <vljn at ovi.com> llvm-svn: 194755
2013-11-15 08:12:45 +08:00
AMDGPU: Mark all unspecified CC functions in tests as amdgpu_kernel Currently the default C calling convention functions are treated the same as compute kernels. Make this explicit so the default calling convention can be changed to a non-kernel. Converted with perl -pi -e 's/define void/define amdgpu_kernel void/' on the relevant test directories (and undoing in one place that actually wanted a non-kernel). llvm-svn: 298444
2017-03-22 05:39:51 +08:00
define amdgpu_kernel void @lds_input_queue(i32 addrspace(1)* %out, i32 addrspace(1)* %in, i32 %index) {
R600: Fix scheduling of instructions that use the LDS output queue The LDS output queue is accessed via the OQAP register. The OQAP register cannot be live across clauses, so if value is written to the output queue, it must be retrieved before the end of the clause. With the machine scheduler, we cannot statisfy this constraint, because it lacks proper alias analysis and it will mark some LDS accesses as having a chain dependency on vertex fetches. Since vertex fetches require a new clauses, the dependency may end up spiltting OQAP uses and defs so the end up in different clauses. See the lds-output-queue.ll test for a more detailed explanation. To work around this issue, we now combine the LDS read and the OQAP copy into one instruction and expand it after register allocation. This patch also adds some checks to the EmitClauseMarker pass, so that it doesn't end a clause with a value still in the output queue and removes AR.X and OQAP handling from the scheduler (AR.X uses and defs were already being expanded post-RA, so the scheduler will never see them). Reviewed-by: Vincent Lejeune <vljn at ovi.com> llvm-svn: 194755
2013-11-15 08:12:45 +08:00
entry:
[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
%0 = getelementptr inbounds [2 x i32], [2 x i32] addrspace(3)* @local_mem, i32 0, i32 %index
[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
%1 = load i32, i32 addrspace(3)* %0
AMDGPU/R600: Replace barrier intrinsics llvm-svn: 275870
2016-07-19 02:34:59 +08:00
call void @llvm.r600.group.barrier()
R600: Fix scheduling of instructions that use the LDS output queue The LDS output queue is accessed via the OQAP register. The OQAP register cannot be live across clauses, so if value is written to the output queue, it must be retrieved before the end of the clause. With the machine scheduler, we cannot statisfy this constraint, because it lacks proper alias analysis and it will mark some LDS accesses as having a chain dependency on vertex fetches. Since vertex fetches require a new clauses, the dependency may end up spiltting OQAP uses and defs so the end up in different clauses. See the lds-output-queue.ll test for a more detailed explanation. To work around this issue, we now combine the LDS read and the OQAP copy into one instruction and expand it after register allocation. This patch also adds some checks to the EmitClauseMarker pass, so that it doesn't end a clause with a value still in the output queue and removes AR.X and OQAP handling from the scheduler (AR.X uses and defs were already being expanded post-RA, so the scheduler will never see them). Reviewed-by: Vincent Lejeune <vljn at ovi.com> llvm-svn: 194755
2013-11-15 08:12:45 +08:00
; This will start a new clause for the vertex fetch
[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
%2 = load i32, i32 addrspace(1)* %in
R600: Fix scheduling of instructions that use the LDS output queue The LDS output queue is accessed via the OQAP register. The OQAP register cannot be live across clauses, so if value is written to the output queue, it must be retrieved before the end of the clause. With the machine scheduler, we cannot statisfy this constraint, because it lacks proper alias analysis and it will mark some LDS accesses as having a chain dependency on vertex fetches. Since vertex fetches require a new clauses, the dependency may end up spiltting OQAP uses and defs so the end up in different clauses. See the lds-output-queue.ll test for a more detailed explanation. To work around this issue, we now combine the LDS read and the OQAP copy into one instruction and expand it after register allocation. This patch also adds some checks to the EmitClauseMarker pass, so that it doesn't end a clause with a value still in the output queue and removes AR.X and OQAP handling from the scheduler (AR.X uses and defs were already being expanded post-RA, so the scheduler will never see them). Reviewed-by: Vincent Lejeune <vljn at ovi.com> llvm-svn: 194755
2013-11-15 08:12:45 +08:00
%3 = add i32 %1, %2
store i32 %3, i32 addrspace(1)* %out
ret void
}
AMDGPU/R600: Replace barrier intrinsics llvm-svn: 275870
2016-07-19 02:34:59 +08:00
declare void @llvm.r600.group.barrier() nounwind convergent
R600: Fix scheduling of instructions that use the LDS output queue The LDS output queue is accessed via the OQAP register. The OQAP register cannot be live across clauses, so if value is written to the output queue, it must be retrieved before the end of the clause. With the machine scheduler, we cannot statisfy this constraint, because it lacks proper alias analysis and it will mark some LDS accesses as having a chain dependency on vertex fetches. Since vertex fetches require a new clauses, the dependency may end up spiltting OQAP uses and defs so the end up in different clauses. See the lds-output-queue.ll test for a more detailed explanation. To work around this issue, we now combine the LDS read and the OQAP copy into one instruction and expand it after register allocation. This patch also adds some checks to the EmitClauseMarker pass, so that it doesn't end a clause with a value still in the output queue and removes AR.X and OQAP handling from the scheduler (AR.X uses and defs were already being expanded post-RA, so the scheduler will never see them). Reviewed-by: Vincent Lejeune <vljn at ovi.com> llvm-svn: 194755
2013-11-15 08:12:45 +08:00
; The machine scheduler does not do proper alias analysis and assumes that
; loads from global values (Note that a global value is different that a
; value from global memory. A global value is a value that is declared
; outside of a function, it can reside in any address space) alias with
; all other loads.
;
; This is a problem for scheduling the reads from the local data share (lds).
; These reads are implemented using two instructions. The first copies the
; data from lds into the lds output queue, and the second moves the data from
; the input queue into main memory. These two instructions don't have to be
; scheduled one after the other, but they do need to be scheduled in the same
; clause. The aliasing problem mentioned above causes problems when there is a
; load from global memory which immediately follows a load from a global value that
; has been declared in the local memory space:
;
[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
; %0 = getelementptr inbounds [2 x i32], [2 x i32] addrspace(3)* @local_mem, i32 0, i32 %index
[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
; %1 = load i32, i32 addrspace(3)* %0
; %2 = load i32, i32 addrspace(1)* %in
R600: Fix scheduling of instructions that use the LDS output queue The LDS output queue is accessed via the OQAP register. The OQAP register cannot be live across clauses, so if value is written to the output queue, it must be retrieved before the end of the clause. With the machine scheduler, we cannot statisfy this constraint, because it lacks proper alias analysis and it will mark some LDS accesses as having a chain dependency on vertex fetches. Since vertex fetches require a new clauses, the dependency may end up spiltting OQAP uses and defs so the end up in different clauses. See the lds-output-queue.ll test for a more detailed explanation. To work around this issue, we now combine the LDS read and the OQAP copy into one instruction and expand it after register allocation. This patch also adds some checks to the EmitClauseMarker pass, so that it doesn't end a clause with a value still in the output queue and removes AR.X and OQAP handling from the scheduler (AR.X uses and defs were already being expanded post-RA, so the scheduler will never see them). Reviewed-by: Vincent Lejeune <vljn at ovi.com> llvm-svn: 194755
2013-11-15 08:12:45 +08:00
;
; The instruction selection phase will generate ISA that looks like this:
[CodeGen] Print register names in lowercase in both MIR and debug output As part of the unification of the debug format and the MIR format, always print registers as lowercase. * Only debug printing is affected. It now follows MIR. Differential Revision: https://reviews.llvm.org/D40417 llvm-svn: 319187
2017-11-29 01:15:09 +08:00
; %oqap = LDS_READ_RET
[CodeGen] Print "%vreg0" as "%0" in both MIR and debug output As part of the unification of the debug format and the MIR format, avoid printing "vreg" for virtual registers (which is one of the current MIR possibilities). Basically: * find . \( -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E "s/%vreg([0-9]+)/%\1/g" * grep -nr '%vreg' . and fix if needed * find . \( -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E "s/ vreg([0-9]+)/ %\1/g" * grep -nr 'vreg[0-9]\+' . and fix if needed Differential Revision: https://reviews.llvm.org/D40420 llvm-svn: 319427
2017-11-30 20:12:19 +08:00
; %0 = MOV %oqap
; %1 = VTX_READ_32
; %2 = ADD_INT %1, %0
R600: Fix scheduling of instructions that use the LDS output queue The LDS output queue is accessed via the OQAP register. The OQAP register cannot be live across clauses, so if value is written to the output queue, it must be retrieved before the end of the clause. With the machine scheduler, we cannot statisfy this constraint, because it lacks proper alias analysis and it will mark some LDS accesses as having a chain dependency on vertex fetches. Since vertex fetches require a new clauses, the dependency may end up spiltting OQAP uses and defs so the end up in different clauses. See the lds-output-queue.ll test for a more detailed explanation. To work around this issue, we now combine the LDS read and the OQAP copy into one instruction and expand it after register allocation. This patch also adds some checks to the EmitClauseMarker pass, so that it doesn't end a clause with a value still in the output queue and removes AR.X and OQAP handling from the scheduler (AR.X uses and defs were already being expanded post-RA, so the scheduler will never see them). Reviewed-by: Vincent Lejeune <vljn at ovi.com> llvm-svn: 194755
2013-11-15 08:12:45 +08:00
;
; The bottom scheduler will schedule the two ALU instructions first:
;
; UNSCHEDULED:
[CodeGen] Print register names in lowercase in both MIR and debug output As part of the unification of the debug format and the MIR format, always print registers as lowercase. * Only debug printing is affected. It now follows MIR. Differential Revision: https://reviews.llvm.org/D40417 llvm-svn: 319187
2017-11-29 01:15:09 +08:00
; %oqap = LDS_READ_RET
[CodeGen] Print "%vreg0" as "%0" in both MIR and debug output As part of the unification of the debug format and the MIR format, avoid printing "vreg" for virtual registers (which is one of the current MIR possibilities). Basically: * find . \( -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E "s/%vreg([0-9]+)/%\1/g" * grep -nr '%vreg' . and fix if needed * find . \( -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E "s/ vreg([0-9]+)/ %\1/g" * grep -nr 'vreg[0-9]\+' . and fix if needed Differential Revision: https://reviews.llvm.org/D40420 llvm-svn: 319427
2017-11-30 20:12:19 +08:00
; %1 = VTX_READ_32
R600: Fix scheduling of instructions that use the LDS output queue The LDS output queue is accessed via the OQAP register. The OQAP register cannot be live across clauses, so if value is written to the output queue, it must be retrieved before the end of the clause. With the machine scheduler, we cannot statisfy this constraint, because it lacks proper alias analysis and it will mark some LDS accesses as having a chain dependency on vertex fetches. Since vertex fetches require a new clauses, the dependency may end up spiltting OQAP uses and defs so the end up in different clauses. See the lds-output-queue.ll test for a more detailed explanation. To work around this issue, we now combine the LDS read and the OQAP copy into one instruction and expand it after register allocation. This patch also adds some checks to the EmitClauseMarker pass, so that it doesn't end a clause with a value still in the output queue and removes AR.X and OQAP handling from the scheduler (AR.X uses and defs were already being expanded post-RA, so the scheduler will never see them). Reviewed-by: Vincent Lejeune <vljn at ovi.com> llvm-svn: 194755
2013-11-15 08:12:45 +08:00
;
; SCHEDULED:
;
[CodeGen] Print "%vreg0" as "%0" in both MIR and debug output As part of the unification of the debug format and the MIR format, avoid printing "vreg" for virtual registers (which is one of the current MIR possibilities). Basically: * find . \( -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E "s/%vreg([0-9]+)/%\1/g" * grep -nr '%vreg' . and fix if needed * find . \( -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E "s/ vreg([0-9]+)/ %\1/g" * grep -nr 'vreg[0-9]\+' . and fix if needed Differential Revision: https://reviews.llvm.org/D40420 llvm-svn: 319427
2017-11-30 20:12:19 +08:00
; %0 = MOV %oqap
; %2 = ADD_INT %1, %2
R600: Fix scheduling of instructions that use the LDS output queue The LDS output queue is accessed via the OQAP register. The OQAP register cannot be live across clauses, so if value is written to the output queue, it must be retrieved before the end of the clause. With the machine scheduler, we cannot statisfy this constraint, because it lacks proper alias analysis and it will mark some LDS accesses as having a chain dependency on vertex fetches. Since vertex fetches require a new clauses, the dependency may end up spiltting OQAP uses and defs so the end up in different clauses. See the lds-output-queue.ll test for a more detailed explanation. To work around this issue, we now combine the LDS read and the OQAP copy into one instruction and expand it after register allocation. This patch also adds some checks to the EmitClauseMarker pass, so that it doesn't end a clause with a value still in the output queue and removes AR.X and OQAP handling from the scheduler (AR.X uses and defs were already being expanded post-RA, so the scheduler will never see them). Reviewed-by: Vincent Lejeune <vljn at ovi.com> llvm-svn: 194755
2013-11-15 08:12:45 +08:00
;
; The lack of proper aliasing results in the local memory read (LDS_READ_RET)
; to consider the global memory read (VTX_READ_32) has a chain dependency, so
; the global memory read will always be scheduled first. This will give us a
; final program which looks like this:
;
; Alu clause:
[CodeGen] Print register names in lowercase in both MIR and debug output As part of the unification of the debug format and the MIR format, always print registers as lowercase. * Only debug printing is affected. It now follows MIR. Differential Revision: https://reviews.llvm.org/D40417 llvm-svn: 319187
2017-11-29 01:15:09 +08:00
; %oqap = LDS_READ_RET
R600: Fix scheduling of instructions that use the LDS output queue The LDS output queue is accessed via the OQAP register. The OQAP register cannot be live across clauses, so if value is written to the output queue, it must be retrieved before the end of the clause. With the machine scheduler, we cannot statisfy this constraint, because it lacks proper alias analysis and it will mark some LDS accesses as having a chain dependency on vertex fetches. Since vertex fetches require a new clauses, the dependency may end up spiltting OQAP uses and defs so the end up in different clauses. See the lds-output-queue.ll test for a more detailed explanation. To work around this issue, we now combine the LDS read and the OQAP copy into one instruction and expand it after register allocation. This patch also adds some checks to the EmitClauseMarker pass, so that it doesn't end a clause with a value still in the output queue and removes AR.X and OQAP handling from the scheduler (AR.X uses and defs were already being expanded post-RA, so the scheduler will never see them). Reviewed-by: Vincent Lejeune <vljn at ovi.com> llvm-svn: 194755
2013-11-15 08:12:45 +08:00
; VTX clause:
[CodeGen] Print "%vreg0" as "%0" in both MIR and debug output As part of the unification of the debug format and the MIR format, avoid printing "vreg" for virtual registers (which is one of the current MIR possibilities). Basically: * find . \( -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E "s/%vreg([0-9]+)/%\1/g" * grep -nr '%vreg' . and fix if needed * find . \( -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E "s/ vreg([0-9]+)/ %\1/g" * grep -nr 'vreg[0-9]\+' . and fix if needed Differential Revision: https://reviews.llvm.org/D40420 llvm-svn: 319427
2017-11-30 20:12:19 +08:00
; %1 = VTX_READ_32
R600: Fix scheduling of instructions that use the LDS output queue The LDS output queue is accessed via the OQAP register. The OQAP register cannot be live across clauses, so if value is written to the output queue, it must be retrieved before the end of the clause. With the machine scheduler, we cannot statisfy this constraint, because it lacks proper alias analysis and it will mark some LDS accesses as having a chain dependency on vertex fetches. Since vertex fetches require a new clauses, the dependency may end up spiltting OQAP uses and defs so the end up in different clauses. See the lds-output-queue.ll test for a more detailed explanation. To work around this issue, we now combine the LDS read and the OQAP copy into one instruction and expand it after register allocation. This patch also adds some checks to the EmitClauseMarker pass, so that it doesn't end a clause with a value still in the output queue and removes AR.X and OQAP handling from the scheduler (AR.X uses and defs were already being expanded post-RA, so the scheduler will never see them). Reviewed-by: Vincent Lejeune <vljn at ovi.com> llvm-svn: 194755
2013-11-15 08:12:45 +08:00
; Alu clause:
[CodeGen] Print "%vreg0" as "%0" in both MIR and debug output As part of the unification of the debug format and the MIR format, avoid printing "vreg" for virtual registers (which is one of the current MIR possibilities). Basically: * find . \( -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E "s/%vreg([0-9]+)/%\1/g" * grep -nr '%vreg' . and fix if needed * find . \( -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E "s/ vreg([0-9]+)/ %\1/g" * grep -nr 'vreg[0-9]\+' . and fix if needed Differential Revision: https://reviews.llvm.org/D40420 llvm-svn: 319427
2017-11-30 20:12:19 +08:00
; %0 = MOV %oqap
; %2 = ADD_INT %1, %2
R600: Fix scheduling of instructions that use the LDS output queue The LDS output queue is accessed via the OQAP register. The OQAP register cannot be live across clauses, so if value is written to the output queue, it must be retrieved before the end of the clause. With the machine scheduler, we cannot statisfy this constraint, because it lacks proper alias analysis and it will mark some LDS accesses as having a chain dependency on vertex fetches. Since vertex fetches require a new clauses, the dependency may end up spiltting OQAP uses and defs so the end up in different clauses. See the lds-output-queue.ll test for a more detailed explanation. To work around this issue, we now combine the LDS read and the OQAP copy into one instruction and expand it after register allocation. This patch also adds some checks to the EmitClauseMarker pass, so that it doesn't end a clause with a value still in the output queue and removes AR.X and OQAP handling from the scheduler (AR.X uses and defs were already being expanded post-RA, so the scheduler will never see them). Reviewed-by: Vincent Lejeune <vljn at ovi.com> llvm-svn: 194755
2013-11-15 08:12:45 +08:00
;
[CodeGen] Print register names in lowercase in both MIR and debug output As part of the unification of the debug format and the MIR format, always print registers as lowercase. * Only debug printing is affected. It now follows MIR. Differential Revision: https://reviews.llvm.org/D40417 llvm-svn: 319187
2017-11-29 01:15:09 +08:00
; This is an illegal program because the oqap def and use know occur in
R600: Fix scheduling of instructions that use the LDS output queue The LDS output queue is accessed via the OQAP register. The OQAP register cannot be live across clauses, so if value is written to the output queue, it must be retrieved before the end of the clause. With the machine scheduler, we cannot statisfy this constraint, because it lacks proper alias analysis and it will mark some LDS accesses as having a chain dependency on vertex fetches. Since vertex fetches require a new clauses, the dependency may end up spiltting OQAP uses and defs so the end up in different clauses. See the lds-output-queue.ll test for a more detailed explanation. To work around this issue, we now combine the LDS read and the OQAP copy into one instruction and expand it after register allocation. This patch also adds some checks to the EmitClauseMarker pass, so that it doesn't end a clause with a value still in the output queue and removes AR.X and OQAP handling from the scheduler (AR.X uses and defs were already being expanded post-RA, so the scheduler will never see them). Reviewed-by: Vincent Lejeune <vljn at ovi.com> llvm-svn: 194755
2013-11-15 08:12:45 +08:00
; different ALU clauses.
;
; This test checks this scenario and makes sure it doesn't result in an
; illegal program. For now, we have fixed this issue by merging the
; LDS_READ_RET and MOV together during instruction selection and then
; expanding them after scheduling. Once the scheduler has better alias
; analysis, we should be able to keep these instructions sparate before
; scheduling.
;
R600: Call EmitFunctionHeader() in the AsmPrinter to populate the ELF symbol table llvm-svn: 218776
2014-10-02 01:15:17 +08:00
; CHECK-LABEL: {{^}}local_global_alias:
R600: Fix scheduling of instructions that use the LDS output queue The LDS output queue is accessed via the OQAP register. The OQAP register cannot be live across clauses, so if value is written to the output queue, it must be retrieved before the end of the clause. With the machine scheduler, we cannot statisfy this constraint, because it lacks proper alias analysis and it will mark some LDS accesses as having a chain dependency on vertex fetches. Since vertex fetches require a new clauses, the dependency may end up spiltting OQAP uses and defs so the end up in different clauses. See the lds-output-queue.ll test for a more detailed explanation. To work around this issue, we now combine the LDS read and the OQAP copy into one instruction and expand it after register allocation. This patch also adds some checks to the EmitClauseMarker pass, so that it doesn't end a clause with a value still in the output queue and removes AR.X and OQAP handling from the scheduler (AR.X uses and defs were already being expanded post-RA, so the scheduler will never see them). Reviewed-by: Vincent Lejeune <vljn at ovi.com> llvm-svn: 194755
2013-11-15 08:12:45 +08:00
; CHECK: LDS_READ_RET
; CHECK-NOT: ALU clause
Fix broken CHECK lines. llvm-svn: 199016
2014-01-12 05:06:00 +08:00
; CHECK: MOV * T{{[0-9]\.[XYZW]}}, OQAP
AMDGPU: Mark all unspecified CC functions in tests as amdgpu_kernel Currently the default C calling convention functions are treated the same as compute kernels. Make this explicit so the default calling convention can be changed to a non-kernel. Converted with perl -pi -e 's/define void/define amdgpu_kernel void/' on the relevant test directories (and undoing in one place that actually wanted a non-kernel). llvm-svn: 298444
2017-03-22 05:39:51 +08:00
define amdgpu_kernel void @local_global_alias(i32 addrspace(1)* %out, i32 addrspace(1)* %in) {
R600: Fix scheduling of instructions that use the LDS output queue The LDS output queue is accessed via the OQAP register. The OQAP register cannot be live across clauses, so if value is written to the output queue, it must be retrieved before the end of the clause. With the machine scheduler, we cannot statisfy this constraint, because it lacks proper alias analysis and it will mark some LDS accesses as having a chain dependency on vertex fetches. Since vertex fetches require a new clauses, the dependency may end up spiltting OQAP uses and defs so the end up in different clauses. See the lds-output-queue.ll test for a more detailed explanation. To work around this issue, we now combine the LDS read and the OQAP copy into one instruction and expand it after register allocation. This patch also adds some checks to the EmitClauseMarker pass, so that it doesn't end a clause with a value still in the output queue and removes AR.X and OQAP handling from the scheduler (AR.X uses and defs were already being expanded post-RA, so the scheduler will never see them). Reviewed-by: Vincent Lejeune <vljn at ovi.com> llvm-svn: 194755
2013-11-15 08:12:45 +08:00
entry:
[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
%0 = getelementptr inbounds [2 x i32], [2 x i32] addrspace(3)* @local_mem, i32 0, i32 0
[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
%1 = load i32, i32 addrspace(3)* %0
%2 = load i32, i32 addrspace(1)* %in
R600: Fix scheduling of instructions that use the LDS output queue The LDS output queue is accessed via the OQAP register. The OQAP register cannot be live across clauses, so if value is written to the output queue, it must be retrieved before the end of the clause. With the machine scheduler, we cannot statisfy this constraint, because it lacks proper alias analysis and it will mark some LDS accesses as having a chain dependency on vertex fetches. Since vertex fetches require a new clauses, the dependency may end up spiltting OQAP uses and defs so the end up in different clauses. See the lds-output-queue.ll test for a more detailed explanation. To work around this issue, we now combine the LDS read and the OQAP copy into one instruction and expand it after register allocation. This patch also adds some checks to the EmitClauseMarker pass, so that it doesn't end a clause with a value still in the output queue and removes AR.X and OQAP handling from the scheduler (AR.X uses and defs were already being expanded post-RA, so the scheduler will never see them). Reviewed-by: Vincent Lejeune <vljn at ovi.com> llvm-svn: 194755
2013-11-15 08:12:45 +08:00
%3 = add i32 %2, %1
store i32 %3, i32 addrspace(1)* %out
ret void
}