This patch is in 2 parts:
1 - replace combineBT's use of SimplifyDemandedBits (hasOneUse only) with SelectionDAG::GetDemandedBits to more aggressively determine the lower bits used by BT.
2 - update SelectionDAG::GetDemandedBits to support ANY_EXTEND - if the demanded bits are only in the non-extended portion, then peek through and demand from the source value and then ANY_EXTEND that if we found a match.
Differential Revision: https://reviews.llvm.org/D35896
llvm-svn: 309486
The X86 tail call eligibility logic was correct when it was written, but
the addition of inalloca and argument copy elision broke its
assumptions. It was assuming that fixed stack objects were immutable.
Currently, we aim to emit a tail call if no arguments have to be
re-arranged in memory. This code would trace the outgoing argument
values back to check if they are loads from an incoming stack object.
If the stack argument is immutable, then we won't need to store it back
to the stack when we tail call.
Fortunately, stack objects track their mutability, so we can just make
the obvious check to fix the bug.
This was http://crbug.com/749826
llvm-svn: 309343
Like r309323, X86 had a typo where it passed the wrong flags to TLO.
Found by inspection; I haven't been able to tickle this into having
observable behavior. I don't think it does, given that X86 doesn't have
custom demanded bits logic, and the generic logic doesn't have a lot of
exposure to illegal constructs.
llvm-svn: 309325
Assign all concat elements to zero and then just replace the first element, instead of setting them all to null and copying everything in.
llvm-svn: 309261
Changing mask argument type from const SmallVectorImpl<int>& to
ArrayRef<int>.
This came up in D35700 where a mask is received as an ArrayRef<int> and
we want to pass it to TargetLowering::isShuffleMaskLegal().
Also saves a few lines of code.
llvm-svn: 309085
splitting patch D34601 into two part. This part changes the location of two functions.
The second part will be based on that patch. This was requested by @RKSimon.
Reviewers:
1. dorit
2. Farhana
3. RKSimon
4. guyblank
5. DavidKreitzer
llvm-svn: 309084
D35067/rL308322 attempted to support up to 4 load pairs for memcmp inlining which resulted in regressions for some optimized libc memcmp implementations (PR33914).
Until we can match these more optimal cases, this patch reduces the memcmp expansion to a maximum of 2 load pairs (which matches what we do for -Os).
This patch should be considered for the 5.0.0 release branch as well
Differential Revision: https://reviews.llvm.org/D35830
llvm-svn: 308986
This patch makes LSR generate better code for SystemZ in the cases of memory
intrinsics, Load->Store pairs or comparison of immediate with memory.
In order to achieve this, the following common code changes were made:
* New TTI hook: LSRWithInstrQueries(), which defaults to false. Controls if
LSR should do instruction-based addressing evaluations by calling
isLegalAddressingMode() with the Instruction pointers.
* In LoopStrengthReduce: handle address operands of memset, memmove and memcpy
as address uses, and call isFoldableMemAccessOffset() for any LSRUse::Address,
not just loads or stores.
SystemZ changes:
* isLSRCostLess() implemented with Insns first, and without ImmCost.
* New function supportedAddressingMode() that is a helper for TTI methods
looking at Instructions passed via pointers.
Review: Ulrich Weigand, Quentin Colombet
https://reviews.llvm.org/D35262https://reviews.llvm.org/D35049
llvm-svn: 308729
Currently we only support (i32 bitcast(v32i1)) using the AVX2 VPMOVMSKB ymm instruction.
This patch adds support for splitting pre-AVX2 targets into 2 x (V)PMOVMSKB xmm instructions and merging the integer results.
In future we could probably generalize this to handle more cases.
Differential Revision: https://reviews.llvm.org/D35303
llvm-svn: 308723
It should be a win to avoid going out to the system lib for all small memcmp() calls using scalar ops. For x86 32-bit, this means most everything up to 16 bytes. For 64-bit, that doubles because we can do 8-byte loads.
Notes:
Reduced from 4 to 2 loads for -Os behavior, which might not be optimal in all cases. It's effectively a question of how much do we trust the system implementation. Linux and macOS (and Windows I assume, but did not test) have optimized memcmp() code for x86, so it's probably not bad either way? PPC is using 8/4 for defaults on these. We do not expand at all for -Oz.
There are still potential improvements to make for the CGP expansion IR and/or lowering such as avoiding select-of-constants (D34904) and not doing zexts to the max load type before doing a compare.
We have special-case SSE/AVX codegen for (memcmp(x, y, 16/32) == 0) that will no longer be produced after this patch. I've shown the experimental justification for that change in PR33329:
https://bugs.llvm.org/show_bug.cgi?id=33329#c12
TLDR: While the vector code is a likely winner, we can't guarantee that it's a winner in all cases on all CPUs, so I'm willing to sacrifice it for the greater good of expanding all small memcmp(). If we want to resurrect that codegen, it can be done by adjusting the CGP params or poking a hole to let those fall-through the CGP expansion.
Committed on behalf of Sanjay Patel
Differential Revision: https://reviews.llvm.org/D35067
llvm-svn: 308322
This isn't legal code, but we shouldn't crash on it. Now we just don't convert the gather intrinsic if the scale isn't constant and let it go through to isel where we'll report an isel failure.
Fixes PR33772.
llvm-svn: 308267
Rename the enum value from X86_64_Win64 to plain Win64.
The symbol exposed in the textual IR is changed from 'x86_64_win64cc'
to 'win64cc', but the numeric value is kept, keeping support for
old bitcode.
Differential Revision: https://reviews.llvm.org/D34474
llvm-svn: 308208
OpenCL 2.0 introduces the notion of memory scopes in atomic operations to
global and local memory. These scopes restrict how synchronization is
achieved, which can result in improved performance.
This change extends existing notion of synchronization scopes in LLVM to
support arbitrary scopes expressed as target-specific strings, in addition to
the already defined scopes (single thread, system).
The LLVM IR and MIR syntax for expressing synchronization scopes has changed
to use *syncscope("<scope>")*, where <scope> can be "singlethread" (this
replaces *singlethread* keyword), or a target-specific name. As before, if
the scope is not specified, it defaults to CrossThread/System scope.
Implementation details:
- Mapping from synchronization scope name/string to synchronization scope id
is stored in LLVM context;
- CrossThread/System and SingleThread scopes are pre-defined to efficiently
check for known scopes without comparing strings;
- Synchronization scope names are stored in SYNC_SCOPE_NAMES_BLOCK in
the bitcode.
Differential Revision: https://reviews.llvm.org/D21723
llvm-svn: 307722
x86 scalar select-of-constants (Cond ? C1 : C2) combining/lowering is a mess
with missing optimizations. We handle some patterns, but miss logical variants.
To clean that up, we should convert all select-of-constants to logic/math and
enhance the combining for the expected patterns from that. Selecting 0 or -1
needs extra attention to produce the optimal code as shown here.
Attempt to verify that all of these IR forms are logically equivalent:
http://rise4fun.com/Alive/plxs
Earlier steps in this series:
rL306040
rL306072
rL307404 (D34652)
As acknowledged in the earlier review, there's a possibility that some Intel
uarch would prefer to produce an xor to clear the fake register operand with
sbb %eax, %eax. This will likely need to be addressed in a separate pass.
llvm-svn: 307471
x86 scalar select-of-constants (Cond ? C1 : C2) combining/lowering is a mess
with missing optimizations. We handle some patterns, but miss logical variants.
To clean that up, we should convert all select-of-constants to logic/math and
enhance the combining for the expected patterns from that. DAGCombiner already
has the foundation to allow the transforms, so we just need to fill in the holes
for x86 math op lowering. Selecting 0 or -1 needs extra attention to produce the
optimal code as shown here.
Attempt to verify that all of these IR forms are logically equivalent:
http://rise4fun.com/Alive/plxs
Earlier steps in this series:
rL306040
rL306072
Differential Revision: https://reviews.llvm.org/D34652
llvm-svn: 307404
We are combining shuffles to bit shifts before unary permutes, which means we can't fold loads plus the destination register is destructive
llvm-svn: 306978
We are combining shuffles to bit shifts before unary permutes, which means we can't fold loads plus the destination register is destructive
The 32-bit shuffles are a bit tricky and will be dealt with in a later patch
llvm-svn: 306977
[X86][AVX512] Improve lowering of AVX512 compare intrinsics (remove redundant shift left+right instructions).
AVX512 compare instructions return v*i1 types.
In cases where the number of elements in the returned value are less than 8, clang adds zeroes to get a mask of v8i1 type.
Later on it's replaced with CONCAT_VECTORS, which then is lowered to many DAG nodes including insert/extract element and shift right/left nodes.
The fact that AVX512 compare instructions put the result in a k register and zeroes all its upper bits allows us to remove the extra nodes simply by copying the result to the required register class.
When lowering, identify these cases and transform them into an INSERT_SUBVECTOR node (marked legal), then catch this pattern in instructions selection phase and transform it into one avx512 cmp instruction.
Differential Revision: https://reviews.llvm.org/D33188
llvm-svn: 306402
Convert vector increment or decrement to sub/add with an all-ones constant:
add X, <1, 1...> --> sub X, <-1, -1...>
sub X, <1, 1...> --> add X, <-1, -1...>
The all-ones vector constant can be materialized using a pcmpeq instruction that is
commonly recognized as an idiom (has no register dependency), so that's better than
loading a splat 1 constant.
AVX512 uses 'vpternlogd' for 512-bit vectors because there is apparently no better
way to produce 512 one-bits.
The general advantages of this lowering are:
1. pcmpeq has lower latency than a memop on every uarch I looked at in Agner's tables,
so in theory, this could be better for perf, but...
2. That seems unlikely to affect any OOO implementation, and I can't measure any real
perf difference from this transform on Haswell or Jaguar, but...
3. It doesn't look like it from the diffs, but this is an overall size win because we
eliminate 16 - 64 constant bytes in the case of a vector load. If we're broadcasting
a scalar load (which might itself be a bug), then we're replacing a scalar constant
load + broadcast with a single cheap op, so that should always be smaller/better too.
4. This makes the DAG/isel output more consistent - we use pcmpeq already for padd x, -1
and psub x, -1, so we should use that form for +1 too because we can. If there's some
reason to favor a constant load on some CPU, let's make the reverse transform for all
of these cases (either here in the DAG or in a later machine pass).
This should fix:
https://bugs.llvm.org/show_bug.cgi?id=33483
Differential Revision: https://reviews.llvm.org/D34336
llvm-svn: 306289
This is very similar to the transform in:
https://reviews.llvm.org/rL306040
...but in this case, we use cmp X, 1 to set the carry bit as needed.
Again, we can show that all of these are logically equivalent (although
InstCombine currently canonicalizes to a form not seen here), and if
we believe IACA, then this is the smallest/fastest code. Eg, with SNB:
| Num Of | Ports pressure in cycles | |
| Uops | 0 - DV | 1 | 2 - D | 3 - D | 4 | 5 | |
---------------------------------------------------------------------
| 1 | 1.0 | | | | | | | cmp edi, 0x1
| 2 | | 1.0 | | | | 1.0 | CP | sbb eax, eax
The larger motivation is to clean up all select-of-constants combining/lowering
because we're missing some common cases.
llvm-svn: 306072
Our handling of select-of-constants is lumpy in IR (https://reviews.llvm.org/D24480),
lumpy in DAGCombiner, and lumpy in X86ISelLowering. That's why we only had the 'sbb'
codegen in 1 out of the 4 tests. This is a step towards smoothing that out.
First, show that all of these IR forms are equivalent:
http://rise4fun.com/Alive/mx
Second, show that the 'sbb' version is faster/smaller. IACA output for SandyBridge
(later Intel and AMD chips are similar based on Agner's tables):
This is the "obvious" x86 codegen (what gcc appears to produce currently):
| Num Of | Ports pressure in cycles | |
| Uops | 0 - DV | 1 | 2 - D | 3 - D | 4 | 5 | |
---------------------------------------------------------------------
| 1* | | | | | | | | xor eax, eax
| 1 | 1.0 | | | | | | CP | test edi, edi
| 1 | | | | | | 1.0 | CP | setnz al
| 1 | | 1.0 | | | | | CP | neg eax
This is the adc version:
| 1* | | | | | | | | xor eax, eax
| 1 | 1.0 | | | | | | CP | cmp edi, 0x1
| 2 | | 1.0 | | | | 1.0 | CP | adc eax, 0xffffffff
And this is sbb:
| 1 | 1.0 | | | | | | | neg edi
| 2 | | 1.0 | | | | 1.0 | CP | sbb eax, eax
If IACA is trustworthy, then sbb became a single uop in Broadwell, so this will be
clearly better than the alternatives going forward.
llvm-svn: 306040
Masked gather for vector length 2 is lowered incorrectly for element type i32.
The type <2 x i32> was automatically extended to <2 x i64> and we generated VPGATHERQQ instead of VPGATHERQD.
The type <2 x float> is extended to <4 x float>, so there is no bug for this type, but the sequence may be more optimal.
In this patch I'm fixing <2 x i32>bug and optimizing <2 x float> sequence for GATHERs only. The same fix should be done for Scatters as well.
Differential revision: https://reviews.llvm.org/D34343
llvm-svn: 305987
There are a couple of potential improvements as seen in the IR and asm:
1. We're unnecessarily extending to a larger type to compare values.
2. The codegen for (select cond, 1, -1) could avoid a cmov.
(or we could change the order of the compares, so we have a select with 0 operand)
llvm-svn: 305802
Target shuffle combining now supports the matching of INSERT_VECTOR_ELT/PINSRW/PINSRB for merging multiple insertions into shuffles/bitmasks.
llvm-svn: 305788
AVX512 compare instructions return v*i1 types.
In cases where the number of elements in the returned value are less than 8, clang adds zeroes to get a mask of v8i1 type.
Later on it's replaced with CONCAT_VECTORS, which then is lowered to many DAG nodes including insert/extract element and shift right/left nodes.
The fact that AVX512 compare instructions put the result in a k register and zeroes all its upper bits allows us to remove the extra nodes simply by copying the result to the required register class.
When lowering, identify these cases and transform them into an INSERT_SUBVECTOR node (marked legal), then catch this pattern in instructions selection phase and transform it into one avx512 cmp instruction.
Differential Revision: https://reviews.llvm.org/D33188
llvm-svn: 305465
We know that shuffle masks are power-of-2 sizes, but there's no way (?) for LLVM to know that,
so hack combineX86ShufflesRecursively() to be much faster by replacing div/rem with shift/mask.
This makes the motivating compile-time bug in PR32037 ( https://bugs.llvm.org/show_bug.cgi?id=32037 )
about 9% faster overall.
Differential Revision: https://reviews.llvm.org/D34174
llvm-svn: 305398
Much of PR32037's compile time regression is due to getTargetConstantBitsFromNode always creating large (>64bit) APInts during the bitcasting from the source data to the destination bitwidth.
This commit avoids this bitcast stage if the data is already the correct bitwidth.
llvm-svn: 305284
This step is just intended to reduce code duplication rather than change any functionality.
A follow-up would be to replace PPCTargetLowering::spliceIntoChain() usage with this new helper.
Differential Revision: https://reviews.llvm.org/D33649
llvm-svn: 305192
I was looking closer at the x86 test diffs in D33866, and the first change seems like it
shouldn't happen in the first place. So this patch will resolve that.
Using Agner's tables and AMD docs, vperm2f128 and vinsertf128 have identical timing for
any given CPU model, so we should be able to interchange those without affecting perf.
But as we can see in some of the diffs here, using vperm2f128 allows load folding, so
we should take that opportunity to reduce code size and register pressure.
A secondary advantage is making AVX1 and AVX2 codegen more similar. Given that vperm2f128
was introduced with AVX1, we should be selecting it in all of the same situations that we
would with AVX2. If there's some reason that an AVX1 CPU would not want to use this
instruction, that should be fixed up in a later pass.
Differential Revision: https://reviews.llvm.org/D33938
llvm-svn: 305171
If we know that both operands of an unsigned integer vector comparison are non-negative,
then it's safe to directly use a signed-compare-greater-than instruction (the only non-equality
integer vector compare predicate provided by SSE/AVX).
We're intentionally not changing the condition code to signed in order to preserve the
existing transforms that use min/max/psubus below here.
This should solve PR33276:
https://bugs.llvm.org/show_bug.cgi?id=33276
Differential Revision: https://reviews.llvm.org/D33862
llvm-svn: 304909
We currently generate BUILD_VECTOR as a tree of UNPCKL shuffles of the same type:
e.g. for v4f32:
Step 1: unpcklps 0, 2 ==> X: <?, ?, 2, 0>
: unpcklps 1, 3 ==> Y: <?, ?, 3, 1>
Step 2: unpcklps X, Y ==> <3, 2, 1, 0>
The issue is because we are not placing sequential vector elements together early enough, we fail to recognise many combinable patterns - consecutive scalar loads, extractions etc.
Instead, this patch unpacks progressively larger sequential vector elements together:
e.g. for v4f32:
Step 1: unpcklps 0, 2 ==> X: <?, ?, 1, 0>
: unpcklps 1, 3 ==> Y: <?, ?, 3, 2>
Step 2: unpcklpd X, Y ==> <3, 2, 1, 0>
This does mean that we are creating UNPCKL shuffle of different value types, but the relevant combines that benefit from this are quite capable of handling the additional BITCASTs that are now included in the shuffle tree.
Differential Revision: https://reviews.llvm.org/D33864
llvm-svn: 304688
Since r288804, we try to lower build_vectors on AVX using broadcasts of
float/double. However, when we broadcast integer values that happen to
have a NaN float bitpattern, we lose the NaN payload, thereby changing
the integer value being broadcast.
This is caused by ConstantFP::get, to which we pass the splat i32 as
a float (by bitcasting it using bitsToFloat). ConstantFP::get takes
a double parameter, so we end up lossily converting a single-precision
NaN to double-precision.
Instead, avoid any kinds of conversions by directly building an APFloat
from the splatted APInt.
Note that this also fixes another piece of code (broadcast of
subvectors), that currently isn't susceptible to the same problem.
Also note that we could really just use APInt and ConstantInt
throughout: the constant pool type doesn't matter much. Still, for
consistency, use the appropriate type.
llvm-svn: 304590
This might give a few better opportunities to optimize these to memcpy
rather than loops - also a few minor cleanups (StringRef-izing,
templating (to avoid std::function indirection), etc).
The SmallVector::assign(iter, iter) could be improved with the use of
SFINAE, but the (iter, iter) ctor and append(iter, iter) need it to and
don't have it - so, workaround it for now rather than bothering with the
added complexity.
(also, as noted in the added FIXME, these assign ops could potentially
be optimized better at least for non-trivially-copyable types)
llvm-svn: 304566
Summary:
Add an early combine to match patterns such as:
(i16 bitcast (v16i1 x))
->
(i16 movmsk (v16i8 sext (v16i1 x)))
This combine needs to happen early enough before
type-legalization scalarizes the result of the setcc.
Reviewers: igorb, craig.topper, RKSimon
Subscribers: delena, llvm-commits
Differential Revision: https://reviews.llvm.org/D33311
llvm-svn: 304406
Summary:
This is a continuation of the work started in D29872 . Passing the carry down as a value rather than as a glue allows for further optimizations. Introducing setcccarry makes the use of addc/subc unecessary and we can start the removal process.
This patch only introduce the optimization strictly required to get the same level of optimization as was available before nothing more.
Reviewers: jyknight, nemanjai, mkuper, spatel, RKSimon, zvi, bkramer
Subscribers: llvm-commits
Differential Revision: https://reviews.llvm.org/D33374
llvm-svn: 304404
Summary:
Currently FPOWI defaults to Legal and LegalizeDAG.cpp turns Legal into Expand for this opcode because Legal is a "lie".
This patch changes the default for this opcode to Expand and removes the hack from LegalizeDAG.cpp. It also removes all the code in the targets that set this opcode to Expand themselves since they can just rely on the default.
Reviewers: spatel, RKSimon, efriedma
Reviewed By: RKSimon
Subscribers: jfb, dschuff, sbc100, jgravelle-google, nemanjai, javed.absar, andrew.w.kaylor, llvm-commits
Differential Revision: https://reviews.llvm.org/D33530
llvm-svn: 304215
AVX512_VPOPCNTDQ is a new feature set that was published by Intel.
The patch represents the LLVM side of the addition of two new intrinsic based instructions (vpopcntd and vpopcntq).
Differential Revision: https://reviews.llvm.org/D33169
llvm-svn: 303858
This patch defines the i1 type as illegal in the X86 backend for AVX512.
For DAG operations on <N x i1> types (build vector, extract vector element, ...) i8 is used, and should be truncated/extended.
This should produce better scalar code for i1 types since GPRs will be used instead of mask registers.
Differential Revision: https://reviews.llvm.org/D32273
llvm-svn: 303421
- '-verify-mahcineinstrs' starts to complain allocatable live-in physical
registers on non-entry or non-landing-pad basic blocks.
- Refactor the XBEGIN translation to define EAX on a dedicated fallback code
path due to XABORT. Add a pseudo instruction to define EAX explicitly to
avoid add physical register live-in.
Differential Revision: https://reviews.llvm.org/D33168
llvm-svn: 303306
This function gives the wrong answer on some non-ELF platforms in some
cases. The function that does the right thing lives in Mangler.h. To try to
discourage people from using this function, give it a different name.
Differential Revision: https://reviews.llvm.org/D33162
llvm-svn: 303134
Replace SelectionDAG::getNode(ISD::SELECT, ...)
and SelectionDAG::getNode(ISD::VSELECT, ...)
with SelectionDAG::getSelect(...)
Saves a few lines of code and in some cases saves the need to explicitly
check the type of the desired node.
llvm-svn: 303024
This patch adds min/max population count, leading/trailing zero/one bit counting methods.
The min methods return answers based on bits that are known without considering unknown bits. The max methods give answers taking into account the largest count that unknown bits could give.
Differential Revision: https://reviews.llvm.org/D32931
llvm-svn: 302925
Avoid using report_fatal_error, because it will ask the user to file a
bug. If the user attempts to disable SSE on x86_64 and them use floating
point, that's a bug in their code, not a bug in the compiler.
This is just a start. There are other ways to crash the backend in this
configuration, but they should be updated to follow this pattern.
Differential Revision: https://reviews.llvm.org/D27522
llvm-svn: 302835
manages to form a VSELECT with a non-i1 element type condition. Those
are technically allowed in SDAG (at least, the generic type legalization
logic will form them and I wouldn't want to try to audit everything te
preclude forming them) so we need to be able to lower them.
This isn't too hard to implement. We mark VSELECT as custom so we get
a chance in C++, add a fast path for i1 conditions to get directly
handled by the patterns, and a fallback when we need to manually force
the condition to be an i1 that uses the vptestm instruction to turn
a non-mask into a mask.
This, unsurprisingly, generates awful code. But it at least doesn't
crash. This was actually impacting open source packages built with LLVM
for AVX-512 in the wild, so quickly landing a patch that at least stops
the immediate bleeding.
I think I've found where to fix the codegen quality issue, but less
confident of that change so separating it out from the thing that
doesn't change the result of any existing test case but causes mine to
not crash.
llvm-svn: 302785
Use variadic templates instead of relying on <cstdarg> + sentinel.
This enforces better type checking and makes code more readable.
Differential Revision: https://reviews.llvm.org/D32541
llvm-svn: 302571
for scalar masked instructions only the lower bit of the mask is relevant. so for constant masks we should either do an unmasked operation or no operation, depending on the value of the lower bit.
This patch handles cases where the lower bit is '1'.
Differential Revision: https://reviews.llvm.org/D32805
llvm-svn: 302546
Using arguments with attribute inalloca creates problems for verification
of machine representation. This attribute instructs the backend that the
argument is prepared in stack prior to CALLSEQ_START..CALLSEQ_END
sequence (see http://llvm.org/docs/InAlloca.htm for details). Frame size
stored in CALLSEQ_START in this case does not count the size of this
argument. However CALLSEQ_END still keeps total frame size, as caller can
be responsible for cleanup of entire frame. So CALLSEQ_START and
CALLSEQ_END keep different frame size and the difference is treated by
MachineVerifier as stack error. Currently there is no way to distinguish
this case from actual errors.
This patch adds additional argument to CALLSEQ_START and its
target-specific counterparts to keep size of stack that is set up prior to
the call frame sequence. This argument allows MachineVerifier to calculate
actual frame size associated with frame setup instruction and correctly
process the case of inalloca arguments.
The changes made by the patch are:
- Frame setup instructions get the second mandatory argument. It
affects all targets that use frame pseudo instructions and touched many
files although the changes are uniform.
- Access to frame properties are implemented using special instructions
rather than calls getOperand(N).getImm(). For X86 and ARM such
replacement was made previously.
- Changes that reflect appearance of additional argument of frame setup
instruction. These involve proper instruction initialization and
methods that access instruction arguments.
- MachineVerifier retrieves frame size using method, which reports sum of
frame parts initialized inside frame instruction pair and outside it.
The patch implements approach proposed by Quentin Colombet in
https://bugs.llvm.org/show_bug.cgi?id=27481#c1.
It fixes 9 tests failed with machine verifier enabled and listed
in PR27481.
Differential Revision: https://reviews.llvm.org/D32394
llvm-svn: 302527
Similar to what we do for vXi8 ASHR(X, 7), use SSE42's PCMPGTQ to splat the sign instead of using the PSRAD+PSHUFD.
Avoiding bitcasts this improves combines that utilize computeNumSignBits, permits memory folding and reduces pipe pressure. Although it does require a second register, given that this is a (cheap) zero register the impact is minimal.
Differential Revision: https://reviews.llvm.org/D32973
llvm-svn: 302525
Currently combineLogicBlendIntoPBLENDV can only match ASHR to detect sign splatting of a bit mask, this patch generalises this to use computeNumSignBits instead.
This is a first step in several things we can do to improve PBLENDV support:
* Better matching of X86ISD::ANDNP patterns.
* Handle floating point cases.
* Better vector and bitcast support in computeNumSignBits.
* Recognise that PBLENDV only uses the sign bit of the mask, we should be able strip away sign splats (ASHR, PCMPGT isNeg tests etc.).
Differential Revision: https://reviews.llvm.org/D32953
llvm-svn: 302424
This patch introduces an LLVM intrinsic and a target opcode for custom event
logging in XRay. Initially, its use case will be to allow users of XRay to log
some type of string ("poor man's printf"). The target opcode compiles to a noop
sled large enough to enable calling through to a runtime-determined relative
function call. At runtime, when X-Ray is enabled, the sled is replaced by
compiler-rt with a trampoline to the logic for creating the custom log entries.
Future patches will implement the compiler-rt parts and clang-side support for
emitting the IR corresponding to this intrinsic.
Reviewers: timshen, dberris
Subscribers: igorb, pelikan, rSerge, timshen, echristo, dberris, llvm-commits
Differential Revision: https://reviews.llvm.org/D27503
llvm-svn: 302405
rL294581 broke unnecessary register dependencies on partial v16i8/v8i16 BUILD_VECTORs, but on SSE41 we (currently) use insertion for full BUILD_VECTORs as well. By allowing full insertion to occur on SSE41 targets we can break register dependencies here as well.
llvm-svn: 302355
This adds routines for reseting KnownBits to unknown, making the value all zeros or all ones. It also adds methods for querying if the value is zero, all ones or unknown.
Differential Revision: https://reviews.llvm.org/D32637
llvm-svn: 302262
This patch adds zext, sext, and trunc methods to KnownBits and uses them where possible.
Differential Revision: https://reviews.llvm.org/D32784
llvm-svn: 302088
This patch adds support for the the LightWeight Profiling (LWP) instructions which are available on all AMD Bulldozer class CPUs (bdver1 to bdver4).
Reapplied - this time without changing line endings of existing files.
Differential Revision: https://reviews.llvm.org/D32769
llvm-svn: 302041
This patch adds support for the the LightWeight Profiling (LWP) instructions which are available on all AMD Bulldozer class CPUs (bdver1 to bdver4).
Differential Revision: https://reviews.llvm.org/D32769
llvm-svn: 302028
Summary: As per discution on how to get better codegen an large int legalization, it became clear that using a glue for the carry was preventing several desirable optimizations. Passing the carry down as a value allow for more flexibility.
Reviewers: jyknight, nemanjai, mkuper, spatel, RKSimon, zvi, bkramer
Subscribers: igorb, llvm-commits
Differential Revision: https://reviews.llvm.org/D29872
llvm-svn: 301775
Adds a new method finalizeLowering to TargetLoweringBase. This is in
preparation for an upcoming commit.
This function is meant for target specific adjustments to
MachineFrameInfo or register reservations.
Move the freezeRegisters() and the hasCopyImplyingStackAdjustment()
handling into the new function to prove the concept. As an added bonus
GlobalISel no longer missed the hasCopyImplyingStackAdjustment()
handling with this.
Differential Revision: https://reviews.llvm.org/D32621
llvm-svn: 301679
This patch replaces the separate APInts for KnownZero/KnownOne with a single KnownBits struct. This is similar to what was done to ValueTracking's version recently.
This is largely a mechanical transformation from KnownZero to Known.Zero.
Differential Revision: https://reviews.llvm.org/D32569
llvm-svn: 301620
This patch uses various APInt methods to reduce the number of temporary APInts. These were all found while working through converting SelectionDAG's computeKnownBits to also use the KnownBits struct recently added to the ValueTracking version.
llvm-svn: 301618
This reverts commit r301105, 4, 3 and 1, as a follow up of the previous
revert, which broke even more bots.
For reference:
Revert "[APInt] Use operator<<= where possible. NFC"
Revert "[APInt] Use operator<<= instead of shl where possible. NFC"
Revert "[APInt] Use ashInPlace where possible."
PR32754.
llvm-svn: 301111
immediate operands.
This commit adds an AArch64 dag-combine that optimizes code generation
for logical instructions taking immediate operands. The optimization
uses demanded bits to change a logical instruction's immediate operand
so that the immediate can be folded into the immediate field of the
instruction.
This recommits r300932 and r300930, which was causing dag-combine to
loop forever. The problem was that optimizeLogicalImm was returning
true even when there was no change to the immediate node (which happened
when the immediate was all zeros or ones), which caused dag-combine to
push and pop the same node to the work list over and over again without
making any progress.
This commit fixes the bug by returning false early in optimizeLogicalImm
if the immediate is all zeros or ones. Also, it changes the code to
compare the immediate with 0 or Mask rather than calling
countPopulation.
rdar://problem/18231627
Differential Revision: https://reviews.llvm.org/D5591
llvm-svn: 301019
It seems that r300930 was creating an infinite loop in dag-combine when
compling the following file:
MultiSource/Benchmarks/MiBench/consumer-typeset/z21.c
llvm-svn: 300940
immediate operands.
This commit adds an AArch64 dag-combine that optimizes code generation
for logical instructions taking immediate operands. The optimization
uses demanded bits to change a logical instruction's immediate operand
so that the immediate can be folded into the immediate field of the
instruction.
This recommits r300913, which broke bots because I didn't fix a call to
ShrinkDemandedConstant in SIISelLowering.cpp after changing the APIs of
TargetLoweringOpt and TargetLowering.
rdar://problem/18231627
Differential Revision: https://reviews.llvm.org/D5591
llvm-svn: 300930
immediate operands.
This commit adds an AArch64 dag-combine that optimizes code generation
for logical instructions taking immediate operands. The optimization
uses demanded bits to change a logical instruction's immediate operand
so that the immediate can be folded into the immediate field of the
instruction.
rdar://problem/18231627
Differential Revision: https://reviews.llvm.org/D5591
llvm-svn: 300913
getSignBit is a static function that creates an APInt with only the sign bit set. getSignMask seems like a better name to convey its functionality. In fact several places use it and then store in an APInt named SignMask.
Differential Revision: https://reviews.llvm.org/D32108
llvm-svn: 300856
This patch uses lshrInPlace to replace code where the object that lshr is called on is being overwritten with the result.
This adds an lshrInPlace(const APInt &) version as well.
Differential Revision: https://reviews.llvm.org/D32155
llvm-svn: 300566
Our 16 bit support is assembler-only + the terrible hack that is
.code16gcc. Simply using 32 bit registers does the right thing for the
latter.
Fixes PR32681.
llvm-svn: 300429
Summary:
In PR32594, inline assembly using the 'A' constraint on x86_64 causes
llvm to crash with a "Cannot select" stack trace. This is because
`X86TargetLowering::getRegForInlineAsmConstraint` hardcodes that 'A'
means the EAX and EDX registers.
However, on x86_64 it means the RAX and RDX registers, and on 16-bit x86
(ia16?) it means the old AX and DX registers.
Add new register classes in `X86RegisterInfo.td` to support these cases,
and amend the logic in `getRegForInlineAsmConstraint` to cope with
different subtargets. Also add a test case, derived from PR32594.
Reviewers: craig.topper, qcolombet, RKSimon, ab
Reviewed By: ab
Subscribers: ab, emaste, royger, llvm-commits
Differential Revision: https://reviews.llvm.org/D31902
llvm-svn: 300404
From a user prospective, it forces the use of an annoying nullptr to mark the end of the vararg, and there's not type checking on the arguments.
The variadic template is an obvious solution to both issues.
Differential Revision: https://reviews.llvm.org/D31070
llvm-svn: 299949
Module::getOrInsertFunction is using C-style vararg instead of
variadic templates.
From a user prospective, it forces the use of an annoying nullptr
to mark the end of the vararg, and there's not type checking on the
arguments. The variadic template is an obvious solution to both
issues.
llvm-svn: 299925
Module::getOrInsertFunction is using C-style vararg instead of
variadic templates.
From a user prospective, it forces the use of an annoying nullptr
to mark the end of the vararg, and there's not type checking on the
arguments. The variadic template is an obvious solution to both
issues.
Patch by: Serge Guelton <serge.guelton@telecom-bretagne.eu>
Differential Revision: https://reviews.llvm.org/D31070
llvm-svn: 299699
Before r294774, there was a problem when lowering broadcasts to use
128-bit subvectors.
When we looked through a bitcast to find the broadcast input, we'd keep
using the original type, so you'd end up with things like:
(v8f32 (broadcast
(v4f32 (extract_subvector
(v8i32 V),
...))
))
r294774 fixed it to always emit subvectors with the scalar type of the
original source.
It also introduced some asserts, to check that we use scalars with
the same size, and vectors with the same number of elements.
The scalar size equality is checked earlier when looking through bitcasts,
and is a useful assert.
However, the number of elements don't have to be identical: we're always
going to extract a 128-bit subvector, and we can have different size
inputs if we looked through a concat_vector to find a 256-bit source.
Relax the overzealous assert.
Replace it with a check of the original source vector being 256 or 512
bits. If it's 128 bits, we can't extract_subvector from it.
Fixes PR32371.
llvm-svn: 299490
https://reviews.llvm.org/D30537 / https://reviews.llvm.org/rL296977 added these transforms
and other related transforms to the generic DAGCombiner (with a hook that x86 sets to true),
so these patterns should not exist by the time we reach the target-specific combiner hook.
llvm-svn: 299448
PSADBW pattern currently supports the 32 bit IR pattern and only GLT (greather than) comparison.
The patch extends the pattern to catch also 64 bit IR pattern and includes all other comparison types (not only GLT).
Differential Revision: https://reviews.llvm.org/D31577
llvm-svn: 299425
It can be costly to transfer from the gprs to the xmm registers and can prevent loads merging.
This patch splits vXi16/vXi32/vXi64 BUILD_VECTORS that use the same operand in multiple elements into a BUILD_VECTOR with only a single insertion of each of those elements and then performs an unary shuffle to duplicate the values.
There are a couple of minor regressions this patch unearths due to some missing MOVDDUP/BROADCAST folds that I will address in a future patch.
Note: Now that vector shuffle lowering and combining is pretty good we should be reusing that instead of duplicating so much in LowerBUILD_VECTOR - this is the first of several patches to address this.
Differential Revision: https://reviews.llvm.org/D31373
llvm-svn: 299387
The x86_64 ABI requires that the stack is 16 byte aligned on function calls. Thus, the 8-byte error code, which is pushed by the CPU for certain exceptions, leads to a misaligned stack. This results in bugs such as Bug 26413, where misaligned movaps instructions are generated.
This commit fixes the misalignment by adjusting the stack pointer in these cases. The adjustment is done at the beginning of the prologue generation by subtracting another 8 bytes from the stack pointer. These additional bytes are popped again in the function epilogue.
Fixes Bug 26413
Patch by Philipp Oppermann.
Differential Revision: https://reviews.llvm.org/D30049
llvm-svn: 299383
This moves the isMask and isShiftedMask functions to be class methods. They now use the MathExtras.h function for single word size and leading/trailing zeros/ones or countPopulation for the multiword size. The previous implementation made multiple temorary memory allocations to do the bitwise arithmetic operations to match the MathExtras.h implementation.
Differential Revision: https://reviews.llvm.org/D31565
llvm-svn: 299362
Our _MM_HINT_T0/T1 constant values are 3/2 which matches gcc, but not icc or Intel documentation. Interestingly gcc had this same bug on their implementation of the gather/scatter builtins at one point too.
Fixes PR32411.
llvm-svn: 299234
Currently ComputeNumSignBits returns the minimum number of sign bits for all elements of vector data, when we may only be interested in one/some of the elements.
This patch adds a DemandedElts argument that allows us to specify the elements we actually care about. The original ComputeNumSignBits implementation calls with a DemandedElts demanding all elements to match current behaviour. Scalar types set this to 1.
I've only added support for BUILD_VECTOR and EXTRACT_VECTOR_ELT so far, all others will default to demanding all elements but can be updated in due course.
Followup to D25691.
Differential Revision: https://reviews.llvm.org/D31311
llvm-svn: 299219
Follow up to D25691, this sets up the plumbing necessary to support vector demanded elements support in known bits calculations in target nodes.
Differential Revision: https://reviews.llvm.org/D31249
llvm-svn: 299201
We currently perform the various fp_to_sint XMM conversion and then transfer to the MMX register (on 32-bit via the stack).
This patch improves support for MOVDQ2Q XMM to MMX transfers and adds the XMM->MMX fp_to_sint direct conversion patterns. The SSE2 specifications are the same as for XMM->XMM and XMM->MMX rounding/exceptions/etc.
Differential Revision: https://reviews.llvm.org/D30868
llvm-svn: 298943
This is a patch for an on-going bugzilla bug 21281 on the generated X86 code for a matrix transpose8x8 subroutine which requires vector interleaving. The generated code in AVX2 is currently non-optimal and requires 60 instructions as opposed to only 40 instructions generated for AVX1.
The patch includes a fix for the AVX2 case where vector unpack instructions use less operations than the vector blend operations available in AVX2.
In this case using vector unpack instructions is more efficient.
Reviewers:
zvi
delena
igorb
craig.topper
guyblank
eladcohen
m_zuckerman
aymanmus
RKSimon
llvm-svn: 298840
Patch to generalize combinePCMPAnd1 (for handling SETCC + ZEXT cases) to work for any input that has zero/all bits set masked with an 'all low bits' mask.
Replaced the implicit assumption of shift availability with a call to SupportedVectorShiftWithImm.
Part 1 of 3.
Differential Revision: https://reviews.llvm.org/D31347
llvm-svn: 298779
This is the payoff for D31156 - if a target has efficient comparison instructions for vector-sized equality,
we can replace memcmp calls with inline code that is both smaller and faster.
Differential Revision: https://reviews.llvm.org/D31290
llvm-svn: 298775
Up until now, vpmovm2 instruction described its destination operand size
by the source operand size. This patch adds new pattern for the vpmovm2
instruction. The node describes new expansion of the destination (from
{128|256} to 512).
Differential Revision: https://reviews.llvm.org/D30654
llvm-svn: 298586
Summary:
This class is a list of AttributeSetNodes corresponding the function
prototype of a call or function declaration. This class used to be
called ParamAttrListPtr, then AttrListPtr, then AttributeSet. It is
typically accessed by parameter and return value index, so
"AttributeList" seems like a more intuitive name.
Rename AttributeSetImpl to AttributeListImpl to follow suit.
It's useful to rename this class so that we can rename AttributeSetNode
to AttributeSet later. AttributeSet is the set of attributes that apply
to a single function, argument, or return value.
Reviewers: sanjoy, javed.absar, chandlerc, pete
Reviewed By: pete
Subscribers: pete, jholewinski, arsenm, dschuff, mehdi_amini, jfb, nhaehnle, sbc100, void, llvm-commits
Differential Revision: https://reviews.llvm.org/D31102
llvm-svn: 298393
We could do better by splitting any oversized type into whatever vector size the target supports,
but I left that for future work if it ever comes up. The motivating case is memcmp() calls on 16-byte
structs, so I think we can wire that up with a TLI hook that feeds into this.
Differential Revision: https://reviews.llvm.org/D31156
llvm-svn: 298376
Make x86_64-fuchsia targets under -mcmodel=kernel use %gs rather
than %fs to access ABI slots for stack-protector and safe-stack
Patch by Roland McGrath.
Differential Revision: https://reviews.llvm.org/D30870
llvm-svn: 298302
Summary:
Currently we handle these intrinsics at isel with special patterns. But as they just map to normal logic operations, we should just handle them at lowering. This will expose them to DAG combine optimizations. Right now the kor-sequence test generates a bunch of regclass copies between GR16 and VK16 that the peephole optimizer and/or register coallescing are removing to keep everything in the mask domain. By handling the logic op intrinsics earlier, these copies become bitcasts in the DAG and get removed by DAG combine which seems more robust.
This should help enable my plan to stop copying between K registers and GR8/GR16. The peephole optimizer can't remove a chain of copies between K and GR32 with insert_subreg/extract_subreg present in the chain so the kor-sequence test break. But this patch should dodge the problem entirely.
Reviewers: zvi, delena, RKSimon, igorb
Reviewed By: igorb
Subscribers: llvm-commits
Differential Revision: https://reviews.llvm.org/D31056
llvm-svn: 298228
The MIR printer dumps a string that describe the register mask of a function.
A static predefined list of register masks matches a static list of strings.
However when the register mask is not from the static predefined list, there is no descriptor string and the printer fails.
This patch adds support to custom register mask printing and dumping.
Also the list of callee saved registers (describing the registers that must be preserved for the caller) might be dynamic.
As such this data needs to be dumped and parsed back to the Machine Register Info.
Differential Revision: https://reviews.llvm.org/D30971
llvm-svn: 298207
As noted in the comment, we might want to account for this case,
but I didn't look at what that would mean for the asm.
I'm also not sure why this only reproduces with avx512, but I'm
putting a conservative fix in for now to avoid the crash.
Also, if both sides of an add are zexted, shouldn't we shrink that add?
https://bugs.llvm.org/show_bug.cgi?id=32316
llvm-svn: 298107
Reduced version of D26357 - based on the discussion on llvm-dev about canonicalization of UMIN/UMAX/SMIN/SMAX as well as ABS I've reduced that patch to just the ABS ISD node (with x86/sse support) to improve basic combines and lowering.
ARM/AArch64, Hexagon, PowerPC and NVPTX all have similar instructions allowing us to make this a generic opcode and move away from the hard coded tablegen patterns which makes it tricky to match more complex patterns.
At the moment this patch doesn't attempt legalization as we only create an ABS node if its legal/custom.
Differential Revision: https://reviews.llvm.org/D29639
llvm-svn: 297780
Each Calling convention (CC) defines a static list of registers that should be preserved by a callee function. All other registers should be saved by the caller.
Some CCs use additional condition: If the register is used for passing/returning arguments – the caller needs to save it - even if it is part of the Callee Saved Registers (CSR) list.
The current LLVM implementation doesn’t support it. It will save a register if it is part of the static CSR list and will not care if the register is passed/returned by the callee.
The solution is to dynamically allocate the CSR lists (Only for these CCs). The lists will be updated with actual registers that should be saved by the callee.
Since we need the allocated lists to live as long as the function exists, the list should reside inside the Machine Register Info (MRI) which is a property of the Machine Function and managed by it (and has the same life span).
The lists should be saved in the MRI and populated upon LowerCall and LowerFormalArguments.
The patch will also assist to implement future no_caller_saved_regsiters attribute intended for interrupt handler CC.
Differential Revision: https://reviews.llvm.org/D28566
llvm-svn: 297715
For AVX-512 we force the input to zero if the input is undef or the mask is all ones to break an execution dependency. This patch brings the same behavior to AVX2.
llvm-svn: 297652
We were already forcing undef inputs to become a zero vector, this now catches an all ones mask too.
Ideally we'd use undef and let execution dep fix handle picking the best register/clearance for the undef, but I don't think it can handle the early clobber today.
llvm-svn: 297651
The immediate should be 1 or 2, not 0 or 1. This was found while adding bounds checking to clang. In fact the existing clang builtin test failed if we ran it all the way to assembly.
llvm-svn: 297591
I noticed unnecessary 'sbb' instructions in D30472 and while looking at 'ptest' codegen recently.
This happens because we were transforming any 'setb' - even when we only wanted a single-bit result.
This patch moves those transforms under visitAdd/visitSub, so we we're only creating sbb/adc when it
is a win. I don't know why we need a SETCC_CARRY node type, but I'm not proposing to change that
existing behavior in this patch.
Also, I'm skeptical that sbb/adc are a win for all micro-arches, so I added comments to the test files
where this transform still fires.
The test changes here are all cases where we no longer produce sbb/adc. Avoiding partial register
stalls (generating an xor to clear a register) is not handled in some cases, but that's a separate
issue.
Differential Revision: https://reviews.llvm.org/D30611
llvm-svn: 297586
Without SSE41 (pextrb) we currently extract byte elements from a vector by spilling to stack and reloading the byte.
This patch is an initial attempt at using MOVD/PEXTRW to extract the relevant DWORD/WORD from the vector and then shift+truncate to collect the correct byte.
Extraction of multiple bytes this way would result in code bloat, but as explained in the patch we could probably afford to be more aggressive with the supported extractions before again falling back on spilling - possibly through counting the number of extracts and which DWORD/WORD they originate?
Differential Revision: https://reviews.llvm.org/D29841
llvm-svn: 297568
We currently have to insert bits via a temporary variable of the same size as the target with various shift/mask stages, resulting in further temporary variables, all of which require the allocation of memory for large APInts (MaskSizeInBits > 64).
This is another of the compile time issues identified in PR32037 (see also D30265).
This patch adds the APInt::insertBits() helper method which avoids the temporary memory allocation and masks/inserts the raw bits directly into the target.
Differential Revision: https://reviews.llvm.org/D30780
llvm-svn: 297458
Summary:
Loop alignment can cause a significant change of
the perfromance for short loops.
To be able to evaluate the impact of loop alignment this change
introduces the new option x86-experimental-pref-loop-alignment.
The alignment will be 2^Value bytes, the default value is 4.
Patch by Serguei Katkov!
Reviewers: craig.topper
Reviewed By: craig.topper
Subscribers: sanjoy, llvm-commits
Differential Revision: https://reviews.llvm.org/D30391
llvm-svn: 297178
Use the store size of the argument type, which will be a byte-sized
quantity, rather than dividing the size in bits by 8.
Fixes PR32136 and re-enables copy elision from i64 arguments.
Reverts the workaround in from r296950.
llvm-svn: 297045
As described on PR31712, we miss a variety of legalization combines because we lower these to X86ISD::VSEXT/VZEXT despite them having the same functionality. This patch makes 128-bit (SSE41) SIGN/ZERO_EXTEND_VECTOR_IN_REG ops legal, adds the necessary tablegen plumbing and uses a helper 'getExtendInVec' to decide when to use SIGN/ZERO_EXTEND_VECTOR_IN_REG or VSEXT/VZEXT.
We're missing a couple of shuffle combines that will be added in a future patch for review.
Later patches can then support the AVX2 cases as a mixture of SIGN/ZERO_EXTEND and SIGN/ZERO_EXTEND_VECTOR_IN_REG, and then finally deal with the AVX512 cases.
Differential Revision: https://reviews.llvm.org/D30549
llvm-svn: 296985
The larger goal is to move the ADC/SBB transforms currently in
combineX86SetCC() to combineAddOrSubToADCOrSBB() because we're
creating ADC/SBB in lots of places where we shouldn't.
This was intended to be an NFC change, but avx-512 has something
strange going on. It doesn't seem like any of the affected tests
should really be using SET+TEST or ADC; a simple ADD could replace
several instructions. But that's another bug...
llvm-svn: 296978
Long ago (2010 according to svn blame), combineShuffle probably needed to prevent the accidental creation of illegal i64 types but there doesn't appear to be any combines that can cause this any more as they all have their own legality checks.
Differential Revision: https://reviews.llvm.org/D30213
llvm-svn: 296966
This fixes cases where i1 types were not properly legalized yet and lead
to the creating of 0-sized stack slots.
This fixes http://llvm.org/PR32136
llvm-svn: 296950
The comments were wrong, and this is not an obvious transform.
This hopefully makes it clearer that we're missing the commuted
patterns for adds. It's less clear that this is actually a good
transform for all micro-arch.
This is prep work for trying to clean up the current adc/sbb
codegen because it's definitely not happening optimally.
llvm-svn: 296918
MMX extraction often ends up as extract_i32(bitcast_v2i32(extract_i64(bitcast_v1i64(x86mmx v), 0)), 0) which fails to simplify on 32-bit targets as i64 isn't legal
llvm-svn: 296782
Summary:
Avoids tons of prologue boilerplate when arguments are passed in memory
and left in memory. This can happen in a debug build or in a release
build when an argument alloca is escaped. This will dramatically affect
the code size of x86 debug builds, because X86 fast isel doesn't handle
arguments passed in memory at all. It only handles the x86_64 case of up
to 6 basic register parameters.
This is implemented by analyzing the entry block before ISel to identify
copy elision candidates. A copy elision candidate is an argument that is
used to fully initialize an alloca before any other possibly escaping
uses of that alloca. If an argument is a copy elision candidate, we set
a flag on the InputArg. If the the target generates loads from a fixed
stack object that matches the size and alignment requirements of the
alloca, the SelectionDAG builder will delete the stack object created
for the alloca and replace it with the fixed stack object. The load is
left behind to satisfy any remaining uses of the argument value. The
store is now dead and is therefore elided. The fixed stack object is
also marked as mutable, as it may now be modified by the user, and it
would be invalid to rematerialize the initial load from it.
Supersedes D28388
Fixes PR26328
Reviewers: chandlerc, MatzeB, qcolombet, inglorion, hans
Subscribers: igorb, llvm-commits
Differential Revision: https://reviews.llvm.org/D29668
llvm-svn: 296683
DAGCombiner already supports peeking thorough shuffles to improve vector element extraction, but legalization often leaves us in situations where we need to extract vector elements after shuffles have already been lowered.
This patch adds support for VECTOR_EXTRACT_ELEMENT/PEXTRW/PEXTRB instructions to attempt to handle target shuffles as well. I've covered some basic scenarios including handling shuffle mask scaling and the implicit zero-extension of PEXTRW/PEXTRB, there is more that could be done here (that I've mentioned in TODOs) but I haven't found many cases where its worth it.
Differential Revision: https://reviews.llvm.org/D30176
llvm-svn: 296381
Summary:
SmallBitVector uses a malloc for more than 58 bits on a 64-bit target and more than 27 bits on a 32-bit target. Some of the vector types we deal with here use more than those number of elements and therefore cause a malloc.
APInt on the other hand supports up to 64 bits without a malloc. That's the maximum number of bits we need here so we can avoid a malloc for all cases by using APInt.
Reviewers: RKSimon
Reviewed By: RKSimon
Subscribers: llvm-commits
Differential Revision: https://reviews.llvm.org/D30392
llvm-svn: 296355
Summary:
SmallBitVector uses a malloc for more than 58 bits on a 64-bit target and more than 27 bits on a 32-bit target. Some of the vector types we deal with here use more than those number of elements and therefore cause a malloc.
APInt on the other hand supports up to 64 bits without a malloc. That's the maximum number of bits we need here so we can avoid a malloc for all cases by using APInt.
Reviewers: RKSimon
Reviewed By: RKSimon
Subscribers: llvm-commits
Differential Revision: https://reviews.llvm.org/D30390
llvm-svn: 296354
The current pattern for extract bits in range is typically:
Mask.lshr(BitOffset).trunc(SubSizeInBits);
Which can be particularly slow for large APInts (MaskSizeInBits > 64) as they require the allocation of memory for the temporary variable.
This is another of the compile time issues identified in PR32037 (see also D30265).
This patch adds the APInt::extractBits() helper method which avoids the temporary memory allocation.
Differential Revision: https://reviews.llvm.org/D30336
llvm-svn: 296272
The current pattern for extract bits in range is typically:
Mask.lshr(BitOffset).trunc(SubSizeInBits);
Which can be particularly slow for large APInts (MaskSizeInBits > 64) as they require the allocation of memory for the temporary variable.
This is another of the compile time issues identified in PR32037 (see also D30265).
This patch adds the APInt::extractBits() helper method which avoids the temporary memory allocation.
Differential Revision: https://reviews.llvm.org/D30336
llvm-svn: 296147
The current pattern for extract bits in range is typically:
Mask.lshr(BitOffset).trunc(SubSizeInBits);
Which can be particularly slow for large APInts (MaskSizeInBits > 64) as they require the allocation of memory for the temporary variable.
This is another of the compile time issues identified in PR32037 (see also D30265).
This patch adds the APInt::extractBits() helper method which avoids the temporary memory allocation.
Differential Revision: https://reviews.llvm.org/D30336
llvm-svn: 296141
Noticed while profiling PR32037, the target shuffle ops were being stored in SmallVector<*,8> types but the combiner could store as many as 16 ops at maximum depth (2 per depth).
llvm-svn: 296130
The current pattern for setting bits in range is typically:
Mask |= APInt::getBitsSet(MaskSizeInBits, LoPos, HiPos);
Which can be particularly slow for large APInts (MaskSizeInBits > 64) as they require the allocation memory for the temporary variable.
This is one of the key compile time issues identified in PR32037.
This patch adds the APInt::setBits() helper method which avoids the temporary memory allocation completely, this first implementation uses setBit() internally instead but already significantly reduces the regression in PR32037 (~10% drop). Additional optimization may be possible.
I investigated whether there is need for APInt::clearBits() and APInt::flipBits() equivalents but haven't seen these patterns to be particularly common, but reusing the code would be trivial.
Differential Revision: https://reviews.llvm.org/D30265
llvm-svn: 296102
The Fuchsia ABI defines slots from the thread pointer where the
stack-guard value for stack-protector, and the unsafe stack pointer
for safe-stack, are stored. This parallels the Android ABI support.
Patch by Roland McGrath
Differential Revision: https://reviews.llvm.org/D30237
llvm-svn: 296081
Minor optimization, don't create temporary mask APInts that are just going to be OR'd into the accumulate masks - insert directly instead.
llvm-svn: 295848
This patch introduces new X86ISD::FMAXS and X86ISD::FMINS opcodes. The legacy intrinsics now lower to this node. As do the AVX-512 masked intrinsics when the rounding mode is CUR_DIRECTION.
I've merged a copy of the tablegen multiclass avx512_fp_scalar into avx512_fp_scalar_sae. avx512_fp_scalar still needs to support CUR_DIRECTION appearing as a rounding mode for X86ISD::FADD_ROUND and others.
Differential revision: https://reviews.llvm.org/D30186
llvm-svn: 295810
Summary:
Rework the code that was sinking/duplicating (icmp and, 0) sequences
into blocks where they were being used by conditional branches to form
more tbz instructions on AArch64. The new code is more general in that
it just looks for 'and's that have all icmp 0's as users, with a target
hook used to select which subset of 'and' instructions to consider.
This change also enables 'and' sinking for X86, where it is more widely
beneficial than on AArch64.
The 'and' sinking/duplicating code is moved into the optimizeInst phase
of CodeGenPrepare, where it can take advantage of the fact the
OptimizeCmpExpression has already sunk/duplicated any icmps into the
blocks where they are used. One minor complication from this change is
that optimizeLoadExt needed to be updated to always mark 'and's it has
determined should be in the same block as their feeding load in the
InsertedInsts set to avoid an infinite loop of hoisting and sinking the
same 'and'.
This change fixes a regression on X86 in the tsan runtime caused by
moving GVNHoist to a later place in the optimization pipeline (see
PR31382).
Reviewers: t.p.northover, qcolombet, MatzeB
Subscribers: aemerson, mcrosier, sebpop, llvm-commits
Differential Revision: https://reviews.llvm.org/D28813
llvm-svn: 295746
This matches what is already done during shuffle lowering and helps prevent the need for a zero-vector in cases where shuffles match both patterns.
llvm-svn: 295723
Pull out repeated code for extraction index operand and source vector value type.
Use isNullConstant helper to check for zero extraction index.
llvm-svn: 295670
Its more profitable to go through memory (1 cycles throughput)
than using VMOVD + VPERMV/PSHUFB sequence ( 2/3 cycles throughput) to implement EXTRACT_VECTOR_ELT with variable index.
IACA tool was used to get performace estimation (https://software.intel.com/en-us/articles/intel-architecture-code-analyzer)
For example for var_shuffle_v16i8_v16i8_xxxxxxxxxxxxxxxx_i8 test from vector-shuffle-variable-128.ll I get 26 cycles vs 79 cycles.
Removing the VINSERT node, we don't need it any more.
Differential Revision: https://reviews.llvm.org/D29690
llvm-svn: 295660
Replaces existing approach that could only search BUILD_VECTOR nodes.
Requires getTargetConstantBitsFromNode to discriminate cases with all/partial UNDEF bits in each element - this should also be useful when we get around to supporting getTargetShuffleMaskIndices with UNDEF elements.
llvm-svn: 295613
As discussed on D27692, this permits another domain to be used to combine a shuffle at high depths.
We currently set the required depth at 4 or more combined shuffles, this is probably too high for most targets but is a good starting point and already helps avoid a number of costly variable shuffles.
llvm-svn: 295608
Add the infrastructure to flag whether float and/or int domains are permitable.
A future patch will enable domain crossing based off shuffle depth and the value types of the source vectors.
llvm-svn: 295604
Minor performance speedup - if any call to getShuffleScalarElt fails to get a result, don't both calling for the remaining elements as EltsFromConsecutiveLoads will fail anyhow.
llvm-svn: 295235
Summary:
We don't seem to have great rules on what a valid VBROADCAST node looks like. And as a consequence we end up with a lot of patterns to try to catch everything. We have patterns with scalar inputs, 128-bit vector inputs, 256-bit vector inputs, and 512-bit vector inputs.
As you can see from the things improved here we are currently missing patterns for 128-bit loads being extended to 256-bit before the vbroadcast.
I'd like to propose that VBROADCAST should always take a 128-bit vector type as input. As a first step towards that this patch adds an EXTRACT_SUBVECTOR in front of VBROADCAST when the input is 256 or 512-bits. In the future I would like to add scalar_to_vector around all the scalar operations. And maybe we should consider adding a VBROADCAST+load node to avoid separating loads from the broadcasting operation when the load itself isn't foldable.
This requires an additional change in target shuffle combining to look for the extract subvector and look through it to find the original operand. I'm sure this change isn't perfect but was enough to fix a few test failures that were being caused.
Another interesting thing I noticed is that the changes in masked_gather_scatter.ll show cases were we don't remove a useless insert into element 1 before broadcasting element 0.
Reviewers: delena, RKSimon, zvi
Reviewed By: zvi
Subscribers: igorb, llvm-commits
Differential Revision: https://reviews.llvm.org/D28747
llvm-svn: 295155
We now detect that both the extract and insert indices are non-zero and convert to a shuffle. This will be lowered as a blend for 256-bit vectors or as a vshuf operations for 512-bit vectors.
llvm-svn: 294931
This results in the simplifications inside of getNode running while we're legalizing nodes popped off the worklist during the final DAG combine. This basically makes a DAG combine like operation occur during this legalize step, but we don't handle something quite the same way. I think we don't recursively added the removed nodes to the DAG combiner worklist.
llvm-svn: 294929
The target shuffle match function arguments were using the term 'Ops' but the function names referred to them as 'Inputs' - use 'Inputs' consistently.
llvm-svn: 294900
Initial 256-bit vector support - 512-bit support requires extra checks for AVX512BW support (PMOVZXBW) that will be handled in a future patch.
llvm-svn: 294896
Removes duplicate constant extraction code in getTargetShuffleMaskIndices.
getTargetConstantBitsFromNode - adds support for VZEXT_MOVL(SCALAR_TO_VECTOR) and fail if the caller doesn't support undef bits.
llvm-svn: 294856
Since r274013, we've been looking through bitcasts on broadcast inputs.
In the scalar-folding case (from a load, build_vector, or sc2vec),
the input type didn't matter, as we'd simply bitcast the resulting
scalar back.
However, when broadcasting a 128-bit-lane-aligned element, we create an
EXTRACT_SUBVECTOR. Use proper types, by creating an extract_subvector
of the original input type.
llvm-svn: 294774
In some cases we call getTargetConstantBitsFromNode for nodes that haven't been lowered from BUILD_VECTOR yet
Note: We're getting very close to being able to move most of the constant extraction code from getTargetShuffleMaskIndices into getTargetConstantBitsFromNode
llvm-svn: 294746
In combineOrCmpEqZeroToCtlzSrl, replace "getConstantOperand == 0" by "isNullConstant" to account for floating point constants.
Differential Revision: https://reviews.llvm.org/D29756
llvm-svn: 294588
LowerBuildVectorv16i8/LowerBuildVectorv8i16 insert values into a UNDEF vector if the build vector doesn't contain any zero elements, resulting in register dependencies with a previous use of the register.
This patch attempts to break the register dependency by either always zeroing the vector before hand or (if we're inserting to the 0'th element) by using VZEXT_MOVL(SCALAR_TO_VECTOR(i32 AEXT(Elt))) which lowers to (V)MOVD and performs a similar function. Additionally (V)MOVD is a shorter instruction than PINSRB/PINSRW. We already do something similar for SSE41 PINSRD.
On pre-SSE41 LowerBuildVectorv16i8 we go a little further and use VZEXT_MOVL(SCALAR_TO_VECTOR(i32 ZEXT(Elt))) if the build vector contains zeros to avoid the vector zeroing at the cost of a scalar zero extension, which can probably be brought over to the other cases in a future patch in some cases (load folding etc.)
Differential Revision: https://reviews.llvm.org/D29720
llvm-svn: 294581
This patch does the following.
1. Adds an Intrinsic int_x86_clzero which works with __builtin_ia32_clzero
2. Identifies clzero feature using cpuid info. (Function:8000_0008, Checks if EBX[0]=1)
3. Adds the clzero feature under znver1 architecture.
4. The custom inserter is added in Lowering.
5. A testcase is added to check the intrinsic.
6. The clzero instruction is added to assembler test.
Patch by Ganesh Gopalasubramanian with a couple formatting tweaks, a disassembler test, and using update_llc_test.py from me.
Differential revision: https://reviews.llvm.org/D29385
llvm-svn: 294558
vXi8/vXi64 vector shifts are often shifted as vYi16/vYi32 types but we weren't always remembering to bitcast the input.
Tested with a new assert as we don't currently manipulate these shifts enough for test cases to catch them.
llvm-svn: 294308
Currently we only combine shuffle nodes if they have a single user to prevent us from causing code bloat by splitting the shuffles into several different combines.
We don't take into account that in some cases we will already have combined all the users during recursively calling up the shuffle tree.
This patch keeps a list of all the shuffle nodes that have been combined so far and permits combining of further shuffle nodes if all its users are in that list.
Differential Revision: https://reviews.llvm.org/D29399
llvm-svn: 294183
Similar to what we already do for zero elt insertion, we can quickly rematerialize 'allbits' vectors so to avoid a unnecessary gpr value and insertion into a vector
llvm-svn: 294162
Similar was already done for several other shuffles in this function.
The test changes are because the old code used explicity zeroing for elements that could have been undef.
While I was here I also changed other shuffle vectors in the same function to use the same input twice instead of creating UNDEF nodes. getVectorShuffle can create the UNDEF for us.
llvm-svn: 294130
On one test this seems to have given more chance for DAG combine to do other INSERT_SUBVECTOR/EXTRACT_SUBVECTOR combines before the BLENDI was created. Looks like we can still improve more by teaching DAG combine to optimize INSERT_SUBVECTOR/EXTRACT_SUBVECTOR with BLENDI.
llvm-svn: 293944
This is a first attempt at using the MOVMSK instructions to replace all_of/any_of reduction patterns (i.e. an and/or + shuffle chain).
So far this only matches patterns where we are reducing an all/none bits source vector (i.e. a comparison result) but we should be able to expand on this in conjunction with improvements to 'bool vector' handling both in the x86 backend as well as the vectorizers etc.
Differential Revision: https://reviews.llvm.org/D28810
llvm-svn: 293880
This moves creation of SUBV_BROADCAST and merging of adjacent loads that are being inserted together.
This is a step towards removing legalizing of INSERT_SUBVECTOR except for vXi1 cases.
llvm-svn: 293872
Also add the ability to recognise PINSR(Vex, 0, Idx).
Targets shuffle combines won't replace multiple insertions with a bit mask until a depth of 3 or more, so we avoid codesize bloat.
The unnecessary vpblendw in clearupper8xi16a will be fixed in an upcoming patch.
llvm-svn: 293627
combineX86ShufflesRecursively can still only handle a maximum of 2 shuffle inputs but everything before it now supports any number of shuffle inputs.
This will be necessary for combining OR(SHUFFLE, SHUFFLE) patterns.
llvm-svn: 293560
Previously this test case fired an assertion in getNode because we tried to create an insert_subvector with both input types the same size and the index pointing to half the vector width.
llvm-svn: 293446
Replaces an xor+movd/movq with an xorps which will be shorter in codesize, avoid an int-fpu transfer, allow modern cores to fast path the result during decode and helps other combines recognise an all-zero vector.
The only reason I can think of that we'd want to keep scalar_to_vector in this case is to help recognise the upper elts are undef but this doesn't seem to be a problem.
Differential Revision: https://reviews.llvm.org/D29097
llvm-svn: 293438
PACKUSWB converts Signed word to Unsigned byte, (the same about DW) and it can't be used for umin+truncate pattern.
AVX-512 VPMOVUS* instructions fit the pattern since they convert Unsigned to Unsigned.
See https://llvm.org/bugs/show_bug.cgi?id=31773
Differential Revision: https://reviews.llvm.org/D29196
llvm-svn: 293431
Pulled out code that removed unused inputs from a target shuffle mask into a helper function to allow it to be reused in a future commit.
llvm-svn: 293175
As discussed on D28219 - it is profitable to combine trunc(binop (s/zext(x), s/zext(y)) to binop(trunc(s/zext(x)), trunc(s/zext(y))) assuming the trunc(ext()) will simplify further
llvm-svn: 292493
Summary:
Currently we expand and scalarize these operations, but I think we should be able to implement ADD/SUB with KXOR and MUL with KAND.
We already do this for scalar i1 operations so I just extended it to vectors of i1.
Reviewers: zvi, delena
Reviewed By: delena
Subscribers: guyblank, llvm-commits
Differential Revision: https://reviews.llvm.org/D28888
llvm-svn: 292474
r291670 doesn't crash on the original testcase from PR31589,
but it crashes on a slightly more complex one.
PR31589 has the new reproducer.
llvm-svn: 292444
This patch improves the mul instruction combine function (combineMul)
by adding new layer of logic.
In this patch, we are adding the ability to fold (mul x, -((1 << c) -1))
or (mul x, -((1 << c) +1)) into (neg(X << c) -x) or (neg((x << c) + x) respective.
Differential Revision: https://reviews.llvm.org/D28232
llvm-svn: 292358
Even with the fix from r291630, this still causes problems. I get
widespread assertion failures in the Swift runtime's WeakRefCount::increment()
function. I sent a reduced testcase in reply to the commit.
llvm-svn: 292242
Use v8i64 variable ASHR instructions if we don't have VLX.
This is a reduced version of D28537 that just adds support for variable shifts - I'll continue with that patch (for just constant/uniform shifts) once I've fixed the type legalization issue in avx512-cvt.ll.
Differential Revision: https://reviews.llvm.org/D28604
llvm-svn: 291901
Rename from addOperand to just add, to match the other method that has been
added to MachineInstrBuilder for adding more than just 1 operand.
See https://reviews.llvm.org/D28057 for the whole discussion.
Differential Revision: https://reviews.llvm.org/D28556
llvm-svn: 291891
Some shuffles can be lowered to blend mask instruction (VPBLENDMB/VPBLENDMW/VPBLENDMD/VPBLENDMQ) .
In this patch, I added new pattern match for this case.
Reviewers:
1. craig.topper
2. guyblank
3. RKSimon
4. igorb
Differential Revision: https://reviews.llvm.org/D28483
llvm-svn: 291888
Emit SHRQ/SHLQ instead of ANDQ with a 64 bit constant mask if the result
is unused and the mask has only higher/lower bits set. For example, with
this patch LLVM emits
shrq $41, %rdi
je
instead of
movabsq $0xFFFFFE0000000000, %rcx
testq %rcx, %rdi
je
This reduces number of instructions, code size and register pressure.
The transformation is applied only for cases where the mask cannot be
encoded as an immediate value within TESTQ instruction.
Differential Revision: https://reviews.llvm.org/D28198
llvm-svn: 291806
64-bit integer division in Intel CPUs is extremely slow, much slower
than 32-bit division. On the other hand, 8-bit and 16-bit divisions
aren't any faster. The only important exception is Atom where DIV8
is fastest. Because of that, the patch
1) Enables bypassing of 64-bit division for Atom, Silvermont and
all big cores.
2) Modifies 64-bit bypassing to use 32-bit division instead of
16-bit one. This doesn't make the shorter division slower but
increases chances of taking it. Moreover, it's much more likely
to prove at compile-time that a value fits 32 bits and doesn't
require a run-time check (e.g. zext i32 to i64).
Differential Revision: https://reviews.llvm.org/D28196
llvm-svn: 291800
r289653 added a case where `vselect <cond> <vector1> <all-zeros>`
is transformed to:
`vselect xor(cond, DAG.getConstant(1, DL, CondVT) <all-zeros> <vector1>`
This was not aimed to catch cases where Cond is not a vXi1
mask but it does. Moreover, when Cond type is VxiN (N > 1)
then xor(cond, DAG.getConstant(1, DL, CondVT) != NOT(cond).
This patch changes the above to xor with allones, and avoids
entering the case for non-mask Conds.
llvm-svn: 291745
DAG patterns optimization: truncate + unsigned saturation supported by VPMOVUS* instructions in AVX-512.
And VPACKUS* instructions on SEE* targets.
Differential Revision: https://reviews.llvm.org/D28216
llvm-svn: 291670
This was reverted because it would miscompile code where the cmp had
multiple uses. That was due to a deficiency in the existing code, which
was fixed in r291630 (see the PR for details).
This re-commit includes an extra test for the kind of code that got
miscompiled: @test_sub_1_setcc_jcc.
llvm-svn: 291640
We would miscompile the following:
void g(int);
int f(volatile long long *p) {
bool b = __atomic_fetch_add(p, 1, __ATOMIC_SEQ_CST) < 0;
g(b ? 12 : 34);
return b ? 56 : 78;
}
into
pushq %rax
lock incq (%rdi)
movl $12, %eax
movl $34, %edi
cmovlel %eax, %edi
callq g(int)
testq %rax, %rax <---- Bad.
movl $56, %ecx
movl $78, %eax
cmovsl %ecx, %eax
popq %rcx
retq
because the code failed to take into account that the cmp has multiple
uses, replaced one of them, and left the other one comparing garbage.
llvm-svn: 291630
This patch fix PR31351: https://llvm.org/bugs/show_bug.cgi?id=31351
1. This patch adds new type of shuffle lowering
2. We can use the expand instruction, When the shuffle pattern is as following:
{ 0*a[0]0*a[1]...0*a[n] , n >=0 where a[] elements in a ascending order}.
Reviewers: 1. igorb
2. guyblank
3. craig.topper
4. RKSimon
Differential Revision: https://reviews.llvm.org/D28352
llvm-svn: 291584
Use the existing AVX2 v8i16 vector shift lowering for v16i8 (extending to v16i32) on AVX512 targets and v32i8 (extending to v32i16) on AVX512BW targets.
Cost model updates to follow.
llvm-svn: 291451
Use the existing AVX2 v8i16 vector shift lowering for v16i16 on AVX512 targets (AVX512BW will have already have lowered with vpsravw).
Cost model updates to follow.
llvm-svn: 291445
I noticed this problem as part of the ongoing attempt to canonicalize min/max ops in IR.
The debug output shows nodes like this:
t4: i32 = xor t2, Constant:i32<-1>
t21: i8 = setcc t4, Constant:i32<0>, setlt:ch
t14: i32 = select t21, t4, Constant:i32<-1>
And because the select is holding onto the t4 (xor) node while EmitTest creates a new
x86-specific xor node, the lowering results in:
t4: i32 = xor t2, Constant:i32<-1>
t25: i32,i32 = X86ISD::XOR t2, Constant:i32<-1>
t28: i32,glue = X86ISD::CMOV Constant:i32<-1>, t4, Constant:i8<15>, t25:1
Differential Revision: https://reviews.llvm.org/D28374
llvm-svn: 291392
Summary:
For instructions such as PSLLW/PSLLD/PSLLQ a variable shift amount may be passed in an XMM register.
The lower 64-bits of the register are evaluated to determine the shift amount.
This patch improves the construction of the vector containing the shift amount.
Reviewers: craig.topper, delena, RKSimon
Subscribers: llvm-commits
Differential Revision: https://reviews.llvm.org/D28353
llvm-svn: 291120
In some cases its more efficient to combine TRUNC( BINOP( X, Y ) ) --> BINOP( TRUNC( X ), TRUNC( Y ) ) if the binop is legal for the truncated types.
This is true for vector integer multiplication (especially vXi64), as well as ADD/AND/XOR/OR in cases where we only need to truncate one of the inputs at runtime (e.g. a duplicated input or an one use constant we can fold).
Further work could be done here - scalar cases (especially i64) could often benefit (if we avoid partial registers etc.), other opcodes, and better analysis of when truncating the inputs reduces costs.
I have considered implementing this for all targets within the DAGCombiner but wasn't sure we could devise a suitable cost model system that would give us the range we need.
Differential Revision: https://reviews.llvm.org/D28219
llvm-svn: 290947
Konstantin Belochapka