aboutsummaryrefslogtreecommitdiff
path: root/llvm/lib/IR/AutoUpgrade.cpp
AgeCommit message (Collapse)AuthorFilesLines
2016-09-13[X86] Remove masked shufpd/shufps intrinsics and autoupgrade to native ↵Craig Topper1-0/+26
vector shuffles. They were removed from clang previously but accidentally left in the backend. llvm-svn: 281300
2016-09-04[AVX-512] Remove 128-bit and 256-bit masked floating point add/sub/mul/div ↵Craig Topper1-0/+44
intrinsics and upgrade to native IR. llvm-svn: 280633
2016-09-04[AVX-512] Remove masked integer add/sub/mull intrinsics and upgrade to ↵Craig Topper1-0/+15
native IR. llvm-svn: 280611
2016-09-03[X86] Combine some of the strings in autoupgrade code.Craig Topper1-35/+7
llvm-svn: 280603
2016-09-02[AVX-512] Remove floating point logical operation instrinsics and replace ↵Craig Topper1-0/+37
them with native IR. llvm-svn: 280466
2016-08-13Revert "Revert "Invariant start/end intrinsics overloaded for address space""Mehdi Amini1-1/+27
This reverts commit 32fc6488e48eafc0ca1bac1bd9cbf0008224d530. llvm-svn: 278609
2016-08-13Revert "Invariant start/end intrinsics overloaded for address space"Mehdi Amini1-27/+1
This reverts commit r276447. llvm-svn: 278608
2016-08-11Use range algorithms instead of unpacking begin/endDavid Majnemer1-1/+1
No functionality change is intended. llvm-svn: 278417
2016-08-10[x86] Fix a bug in the auto-upgrade from r276416 where we failed to giveChandler Carruth1-1/+1
a sufficiently low alignment for the IR load created. There is no test case because we don't have any test cases for the *IR* produced by the autoupgrade, only the x86 assembly, and it happens that the x86 assembly for this intrinsic as it is tested in the autoupgrade path just happens to not produce a separate load instruction where we might have observed the alignment. I'm going to follow up on the original commit to suggest getting IR-level testing in addition to the asm level testing here so that we can see and test these kinds of issues. We might never get an x86 instruction out with an alignment constraint, but we could stil miscompile code by folding against the alignment marked on (or inferred for in this case) the load. llvm-svn: 278203
2016-07-22Invariant start/end intrinsics overloaded for address spaceAnna Thomas1-1/+27
Summary: The llvm.invariant.start and llvm.invariant.end intrinsics currently support specifying invariant memory objects only in the default address space. With this change, these intrinsics are overloaded for any adddress space for memory objects and we can use these llvm invariant intrinsics in non-default address spaces. Example: llvm.invariant.start.p1i8(i64 4, i8 addrspace(1)* %ptr) This overloaded intrinsic is needed for representing final or invariant memory in managed languages. Reviewers: apilipenko, reames Subscribers: llvm-commits llvm-svn: 276447
2016-07-22[X86][AVX] Added support for lowering to VBROADCASTF128/VBROADCASTI128 ↵Simon Pilgrim1-7/+14
(reapplied) As reported on PR26235, we don't currently make use of the VBROADCASTF128/VBROADCASTI128 instructions (or the AVX512 equivalents) to load+splat a 128-bit vector to both lanes of a 256-bit vector. This patch enables lowering from subvector insertion/concatenation patterns and auto-upgrades the llvm.x86.avx.vbroadcastf128.pd.256 / llvm.x86.avx.vbroadcastf128.ps.256 intrinsics to match. We could possibly investigate using VBROADCASTF128/VBROADCASTI128 to load repeated constants as well (similar to how we already do for scalar broadcasts). Reapplied with fix for PR28657 - removed intrinsic definitions (clang companion patch to be be submitted shortly). Differential Revision: https://reviews.llvm.org/D22460 llvm-svn: 276416
2016-07-22Revert "[X86][AVX] Added support for lowering to VBROADCASTF128/VBROADCASTI128"Benjamin Kramer1-14/+7
It caused PR28657. This reverts commit r276281. llvm-svn: 276405
2016-07-21Revert "Invariant start/end intrinsics overloaded for address space"Anna Thomas1-27/+1
This reverts commit r276316. llvm-svn: 276320
2016-07-21Invariant start/end intrinsics overloaded for address spaceAnna Thomas1-1/+27
Summary: The llvm.invariant.start and llvm.invariant.end intrinsics currently support specifying invariant memory objects only in the default address space. With this change, these intrinsics are overloaded for any adddress space for memory objects and we can use these llvm invariant intrinsics in non-default address spaces. Example: llvm.invariant.start.p1i8(i64 4, i8 addrspace(1)* %ptr) This overloaded intrinsic is needed for representing final or invariant memory in managed languages. Reviewers: tstellarAMD, reames, apilipenko Subscribers: llvm-commits Differential Revision: https://reviews.llvm.org/D22519 llvm-svn: 276316
2016-07-21[X86][AVX] Added support for lowering to VBROADCASTF128/VBROADCASTI128Simon Pilgrim1-7/+14
As reported on PR26235, we don't currently make use of the VBROADCASTF128/VBROADCASTI128 instructions (or the AVX512 equivalents) to load+splat a 128-bit vector to both lanes of a 256-bit vector. This patch enables lowering from subvector insertion/concatenation patterns and auto-upgrades the llvm.x86.avx.vbroadcastf128.pd.256 / llvm.x86.avx.vbroadcastf128.ps.256 intrinsics to match. We could possibly investigate using VBROADCASTF128/VBROADCASTI128 to load repeated constants as well (similar to how we already do for scalar broadcasts). Differential Revision: https://reviews.llvm.org/D22460 llvm-svn: 276281
2016-07-19[X86][SSE] Reimplement SSE fp2si conversion intrinsics instead of using ↵Simon Pilgrim1-8/+0
generic IR D20859 and D20860 attempted to replace the SSE (V)CVTTPS2DQ and VCVTTPD2DQ truncating conversions with generic IR instead. It turns out that the behaviour of these intrinsics is different enough from generic IR that this will cause problems, INF/NAN/out of range values are guaranteed to result in a 0x80000000 value - which plays havoc with constant folding which converts them to either zero or UNDEF. This is also an issue with the scalar implementations (which were already generic IR and what I was trying to match). This patch changes both scalar and packed versions back to using x86-specific builtins. It also deals with the other scalar conversion cases that are runtime rounding mode dependent and can have similar issues with constant folding. A companion clang patch is at D22105 Differential Revision: https://reviews.llvm.org/D22106 llvm-svn: 275981
2016-07-12[AVX512] Remove masked logic op intrinsics and autoupgrade them to native IR.Craig Topper1-0/+21
llvm-svn: 275155
2016-07-12[X86,IR] Remove unnecessary or unused LLVMContext parameter from some of the ↵Craig Topper1-17/+16
X86 intrinsic upgrade functions. llvm-svn: 275138
2016-07-09[X86] Remove and autoupgrade 512-bit non-temporal store intrinsics.Craig Topper1-2/+6
llvm-svn: 274966
2016-07-08Move setName after accessing NameEric Liu1-5/+2
llvm-svn: 274862
2016-07-08Make a std::string copy of StringRef Name so that it remains valid when the ↵Eric Liu1-1/+4
original Name is overridden. Summary: lib/IR/AutoUpgrade.cpp:348 and lib/IR/AutoUpgrade.cpp:350 upset sanitizer. Reviewers: bkramer Subscribers: llvm-commits Differential Revision: http://reviews.llvm.org/D22140 llvm-svn: 274861
2016-07-08[AVX512] Remove and autoupgrade a duplicate set of 512-bit masked shift ↵Craig Topper1-1/+24
intrinsics. I'm not sure if clang ever used these builtin names or not. llvm-svn: 274827
2016-07-07[AVX512] Zero extend the result of vpcmpeq/vpcmpgt and similar intrinsics in ↵Craig Topper1-2/+3
the autoupgrade code. This currently results in worse codegen but is needed for correctness. llvm-svn: 274736
2016-07-05[X86][AVX512] Autoupgrade the BROADCAST intrinsicsSimon Pilgrim1-1/+9
llvm-svn: 274550
2016-07-04[IR,X86] Remove some intrinsic prefixes earlier in the auto-upgrade code so ↵Craig Topper1-214/+226
we can shorten the length of the comparison strings and avoid repeatedly comparing the common prefix. No functional change intended. llvm-svn: 274522
2016-07-04[X86][AVX512] Autoupgrade the VPERMPD/VPERMQ intrinsicsSimon Pilgrim1-0/+18
llvm-svn: 274506
2016-07-04[X86][AVX512] Autoupgrade the VPERMILPD/VPERMILPS intrinsicsSimon Pilgrim1-1/+3
llvm-svn: 274498
2016-07-02[X86][AVX512] Autoupgrade the MOVDDUP/MOVSLDUP/MOVSHDUP intrinsicsSimon Pilgrim1-0/+25
llvm-svn: 274439
2016-06-28Support arbitrary addrspace pointers in masked load/store intrinsicsArtur Pilipenko1-0/+34
This is a resubmittion of 263158 change after fixing the existing problem with intrinsics mangling (see LTO and intrinsics mangling llvm-dev thread for details). This patch fixes the problem which occurs when loop-vectorize tries to use @llvm.masked.load/store intrinsic for a non-default addrspace pointer. It fails with "Calling a function with a bad signature!" assertion in CallInst constructor because it tries to pass a non-default addrspace pointer to the pointer argument which has default addrspace. The fix is to add pointer type as another overloaded type to @llvm.masked.load/store intrinsics. Reviewed By: reames Differential Revision: http://reviews.llvm.org/D17270 llvm-svn: 274043
2016-06-27Revert -r273892 "Support arbitrary addrspace pointers in masked load/store ↵Artur Pilipenko1-34/+0
intrinsics" since some of the clang tests don't expect to see the updated signatures. llvm-svn: 273895
2016-06-27Support arbitrary addrspace pointers in masked load/store intrinsicsArtur Pilipenko1-0/+34
This is a resubmittion of 263158 change after fixing the existing problem with intrinsics mangling (see LTO and intrinsics mangling llvm-dev thread for details). This patch fixes the problem which occurs when loop-vectorize tries to use @llvm.masked.load/store intrinsic for a non-default addrspace pointer. It fails with "Calling a function with a bad signature!" assertion in CallInst constructor because it tries to pass a non-default addrspace pointer to the pointer argument which has default addrspace. The fix is to add pointer type as another overloaded type to @llvm.masked.load/store intrinsics. Reviewed By: reames Differential Revision: http://reviews.llvm.org/D17270 llvm-svn: 273892
2016-06-23[AVX512] Remove masked unpack intrinsics and autoupgrade to vectorshuffle ↵Craig Topper1-0/+36
and selects. llvm-svn: 273543
2016-06-21[AVX512] Remove the masked vpcmpeq/vcmpgt intrinsics and autoupgrade them to ↵Craig Topper1-0/+30
native icmps. llvm-svn: 273240
2016-06-18[X86][SSE4A] Autoupgrade and remove MOVNTSD/MOVNTSS intrinsicsSimon Pilgrim1-0/+25
Required better annotation of the instruction defs upon removal of the builtin intrinsic pattern. llvm-svn: 273077
2016-06-16[x86] autoupgrade and remove AVX2 integer min/max intrinsicsSanjay Patel1-4/+10
This will (hopefully very temporarily) break clang. The clang side of this should be the next commit. llvm-svn: 272932
2016-06-16[x86] autoupgrade and remove SSE2/SSE41 integer min/max intrinsicsSanjay Patel1-0/+36
Follow-up to: http://reviews.llvm.org/rL272806 http://reviews.llvm.org/rL272807 llvm-svn: 272907
2016-06-15fix comments; NFCSanjay Patel1-6/+3
llvm-svn: 272848
2016-06-13[AVX512] Remove maksed pshufd, pshuflw, and phufhw intrinsics and ↵Craig Topper1-3/+21
autoupgrade them to selects and shufflevector. llvm-svn: 272527
2016-06-13[X86] Refactor some of the X86 autoupgrade code to split mask vector and ↵Craig Topper1-74/+69
select generation into routines that can be reused for future intrinsic upgrades. NFC llvm-svn: 272526
2016-06-12[X86] Remove sse2 pshufd/pshuflw/pshufhw intrinsics and upgrade them to ↵Craig Topper1-1/+31
shufflevector. llvm-svn: 272510
2016-06-12[X86] Pre-allocate some of the shuffle mask SmallVectors in the auto upgrade ↵Craig Topper1-20/+12
code instead of calling push_back in a loop. This removes the need to check if the vector needs to grow on each iteration. llvm-svn: 272501
2016-06-12[X86] Greatly simplify the llvm.x86.avx.vpermil.* auto-upgrade code. We can ↵Craig Topper1-36/+18
fully derive everything using types of the intrinsic arguments rather than writing separate loops for each intrinsic. NFC llvm-svn: 272496
2016-06-12[X86,IR] Make use of the CreateShuffleVector form that takes an ↵Craig Topper1-20/+19
ArrayRef<uint32_t> to avoid the need to manually create a bunch of Constants and a ConstantVector. NFC llvm-svn: 272493
2016-06-12[IR] Require ArrayRef of 'uint32_t' instead of 'int' for the mask argument ↵Craig Topper1-10/+11
for one of the signatures of CreateShuffleVector. This better emphasises that you can't use it for the -1 as undef behavior. llvm-svn: 272491
2016-06-09[X86][AVX512] Dropped avx512 VPSLLDQ/VPSRLDQ intrinsicsSimon Pilgrim1-10/+14
Auto-upgrade to generic shuffles like sse/avx2 implementations now that we can lower to VPSLLDQ/VPSRLDQ llvm-svn: 272308
2016-06-06[AVX512] Remove masked palignr intrinsics and auto-upgrade them to native IR ↵Craig Topper1-0/+54
of vector shuffle and select. llvm-svn: 271872
2016-06-03[X86][XOP] Support for VPERMIL2PD/VPERMIL2PS 2-input shuffle instructionsSimon Pilgrim1-0/+35
This patch begins adding support for lowering to the XOP VPERMIL2PD/VPERMIL2PS shuffle instructions - adding the X86ISD::VPERMIL2 opcode and cleaning up the usage. The internal llvm intrinsics were assuming the shuffle mask operand was the same type as the float/double input operands (I guess to simplify the intrinsic definitions in X86InstrXOP.td to a single value type). These needed changing to integer types (matching the clang builtin and the AMD intrinsics definitions), an auto upgrade path is added to convert old calls. Mask decoding/target shuffle support will be added in future patches. Differential Revision: http://reviews.llvm.org/D20049 llvm-svn: 271633
2016-06-02[X86][SSE] Replace (V)CVTTPS2DQ and VCVTTPD2DQ truncating (round to zero) ↵Simon Pilgrim1-0/+8
f32/f64 to i32 with generic IR (llvm) This patch removes the llvm intrinsics (V)CVTTPS2DQ and VCVTTPD2DQ truncation (round to zero) conversions and auto-upgrades to FP_TO_SINT calls instead. Note: I looked at updating CVTTPD2DQ as well but this still requires a lot more work to correctly lower. Differential Revision: http://reviews.llvm.org/D20860 llvm-svn: 271510
2016-06-02[AVX512] Remove masked load intrinsics. Clang now emits generic masked load ↵Craig Topper1-2/+62
intrinsics instead. The intrinsics will be autoupgraded to the same generic masked loads. llvm-svn: 271478
2016-06-01Revert r271362 "[AVX512] Remove masked load intrinsics. Clang now emits ↵Craig Topper1-62/+2
generic masked load intrinsics instead." Looks like something isn't quite right still. Also forgot to move the test cases to an autoupgrade test. llvm-svn: 271363
generated by cgit v1.1 at 2025-10-14 05:56:51 +0000