a bunch more improvements.
Non-lane-crossing is fine, the key is that lane merging only makes sense
for single-input shuffles. Not sure why I got so turned around here. The
code all works, I was just using the wrong model for it.
This only updates v4 and v8 lowering. The v16 and v32 lowering requires
restructuring the entire check sequence.
llvm-svn: 222537
lanes.
By special casing these we can often either reduce the total number of
shuffles significantly or reduce the number of (high latency on Haswell)
AVX2 shuffles that potentially cross 128-bit lanes. Even when these
don't actually cross lanes, they have much higher latency to support
that. Doing two of them and a blend is worse than doing a single insert
across the 128-bit lanes to blend and then doing a single interleaved
shuffle.
While this seems like a narrow case, it kept cropping up on me and the
difference is *huge* as you can see in many of the test cases. I first
hit this trying to perfectly fix the interleaving shuffle patterns used
by Halide for AVX2.
llvm-svn: 222533
Windows itanium targets the MSVCRT, and the stack probe symbol is provided by
MSVCRT. This corrects the emission of stack probes on i686-windows-itanium.
llvm-svn: 222439
This patch improves the lowering of v4f32 and v4i32 build_vector dag nodes
that are known to have at least two non-zero elements.
With this patch, a build_vector that performs a blend with zero is
converted into a shuffle. This is done to let the shuffle legalizer expand
the dag node in a optimal way. For example, if we know that a build_vector
performs a blend with zero, we can try to lower it as a movq/blend instead of
always selecting an insertps.
This patch also improves the logic that lowers a build_vector into a insertps
with zero masking. See for example the extra test cases added to test sse41.ll.
Differential Revision: http://reviews.llvm.org/D6311
llvm-svn: 222375
This patch builds on http://reviews.llvm.org/D5598 to perform byte rotation shuffles (lowerVectorShuffleAsByteRotate) on pre-SSSE3 (palignr) targets - pre-SSSE3 is only enabled on i8 and i16 vector targets where it is a more definite performance gain.
I've also added a separate byte shift shuffle (lowerVectorShuffleAsByteShift) that makes use of the ability of the SLLDQ/SRLDQ instructions to implicitly shift in zero bytes to avoid the need to create a zero register if we had used palignr.
Differential Revision: http://reviews.llvm.org/D5699
llvm-svn: 222340
Updated X86TargetLowering::isShuffleMaskLegal to match SHUFP masks with commuted inputs and PSHUFD masks that reference the second input.
As part of this I've refactored isPSHUFDMask to work in a more general manner and allow it to match against either the first or second input vector.
Differential Revision: http://reviews.llvm.org/D6287
llvm-svn: 222087
getTargetConstant should only be used when you can guarantee the instruction
selected will be able to cope with the raw value. BUILD_VECTOR is rather too
generic for this so we should use getConstant instead. In that case, an
instruction can still consume the constant, but if it doesn't it'll be
materialised through its own round of ISel.
Should fix PR21352.
llvm-svn: 221961
between splitting a vector into 128-bit lanes and recombining them vs.
decomposing things into single-input shuffles and a final blend.
This handles a large number of cases in AVX1 where the cross-lane
shuffles would be much more expensive to represent even though we end up
with a fast blend at the root. Instead, we can do a better job of
shuffling in a single lane and then inserting it into the other lanes.
This fixes the remaining bits of Halide's regression captured in PR21281
for AVX1. However, the bug persists in AVX2 because I've made this
change reasonably conservative. The cases where it makes sense in AVX2
to split into 128-bit lanes are much more rare because we can often do
full permutations across all elements of the 256-bit vector. However,
the particular test case in PR21281 is an example of one of the rare
cases where it is *always* better to work in a single 128-bit lane. I'm
going to try to teach the logic to detect and form the good code even in
AVX2 next, but it will need to use a separate heuristic.
Finally, there is one pesky regression here where we previously would
craftily use vpermilps in AVX1 to shuffle both high and low halves at
the same time. We no longer pull that off, and not for any really good
reason. Ultimately, I think this is just another missing nuance to the
selection heuristic that I'll try to add in afterward, but this change
already seems strictly worth doing considering the magnitude of the
improvements in common matrix math shuffle patterns.
As always, please let me know if this causes a surprising regression for
you.
llvm-svn: 221861
re-combining shuffles because nothing was available in the wider vector
type.
The key observation (which I've put in the comments for future
maintainers) is that at this point, no further combining is really
possible. And so even though these shuffles trivially could be combined,
we need to actually do that as we produce them when producing them this
late in the lowering.
This fixes another (huge) part of the Halide vector shuffle regressions.
As it happens, this was already well covered by the tests, but I hadn't
noticed how bad some of these got. The specific patterns that turn
directly into unpckl/h patterns were occurring *many* times in common
vector processing code.
There are still more problems here sadly, but trying to incrementally
tease them apart and it looks like this is the core of the problem in
the splitting logic.
There is some chance of regression here, you can see it in the test
changes. Specifically, where we stop forming pshufb in some cases, it is
possible that pshufb was in fact faster. Intel "says" that pshufb is
slower than the instruction sequences replacing it.
llvm-svn: 221852
This is a follow-on to r221706 and r221731 and discussed in more detail in PR21385.
This patch also loosens the testcase checking for btver2. We know that the "1.0" will be loaded, but
we can't tell exactly when, so replace the CHECK-NEXT specifiers with plain CHECKs. The CHECK-NEXT
sequence relied on a quirk of post-RA-scheduling that may change independently of anything in these tests.
llvm-svn: 221819
cases from Halide folks. This initial step was extracted from
a prototype change by Clay Wood to try and address regressions found
with Halide and the new vector shuffle lowering.
llvm-svn: 221779
This is a first step for generating SSE rcp instructions for reciprocal
calcs when fast-math allows it. This is very similar to the rsqrt optimization
enabled in D5658 ( http://reviews.llvm.org/rL220570 ).
For now, be conservative and only enable this for AMD btver2 where performance
improves significantly both in terms of latency and throughput.
We may never enable this codegen for Intel Core* chips because the divider circuits
are just too fast. On SandyBridge, divss can be as fast as 10 cycles versus the 21
cycle critical path for the rcp + mul + sub + mul + add estimate.
Follow-on patches may allow configuration of the number of Newton-Raphson refinement
steps, add AVX512 support, and enable the optimization for more chips.
More background here: http://llvm.org/bugs/show_bug.cgi?id=21385
Differential Revision: http://reviews.llvm.org/D6175
llvm-svn: 221706
The ISel lowering for global TLS access in PIC mode was creating a pseudo
instruction that is later expanded to a call, but the code was not
setting the hasCalls flag in the MachineFrameInfo alongside the adjustsStack
flag. This caused some functions to be mistakenly recognized as leaf functions,
and this in turn affected the decision to eliminate the frame pointer.
With the fix, hasCalls is properly set and the leaf frame pointer is correctly
preserved.
llvm-svn: 221695
AVX2 is available.
According to IACA, the new lowering has a throughput of 8 cycles instead of 13
with the previous one.
Althought this lowering kicks in some SPECs benchmarks, the performance
improvement was within the noise.
Correctness testing has been done for the whole range of uint32_t with the
following program:
uint4 v = (uint4) {0,1,2,3};
uint32_t i;
//Check correctness over entire range for uint4 -> float4 conversion
for( i = 0; i < 1U << (32-2); i++ )
{
float4 t = test(v);
float4 c = correct(v);
if( 0xf != _mm_movemask_ps( t == c ))
{
printf( "Error @ %vx: %vf vs. %vf\n", v, c, t);
return -1;
}
v += 4;
}
Where "correct" is the old lowering and "test" the new one.
The patch adds a test case for the two custom lowering instruction.
It also modifies the vector cost model, which is why cast.ll and uitofp.ll are
modified.
2009-02-26-MachineLICMBug.ll is also modified because we now hoist 7
instructions instead of 4 (3 more constant loads).
rdar://problem/18153096>
llvm-svn: 221657
condition to match a blend.
This prevents optimizations that work on VSELECT to perform invalid
transformations. Indeed, the optimized condition does not match the vector
boolean content that is expected and bad things may happen.
This patch yields the exact same code on the whole test-suite + specs (-O3 and
-O3 -march=core-avx2), it improves one test case (vector-blend.ll) and fixes a
bug reduced in vselect-avx.ll.
<rdar://problem/18819506>
llvm-svn: 221429
This patch improves the folding of vector AND nodes into blend operations for
targets that feature SSE4.1. A vector AND node where one of the operands is
a constant build_vector with elements that are either zero or all-ones can be
converted into a blend.
This allows for example to simplify the following code:
define <4 x i32> @test(<4 x i32> %A, <4 x i32> %B) {
%1 = and <4 x i32> %A, <i32 0, i32 0, i32 0, i32 -1>
%2 = and <4 x i32> %B, <i32 -1, i32 -1, i32 -1, i32 0>
%3 = or <4 x i32> %1, %2
ret <4 x i32> %3
}
Before this patch llc (-mcpu=corei7) generated:
andps LCPI1_0(%rip), %xmm0, %xmm0
andps LCPI1_1(%rip), %xmm1, %xmm1
orps %xmm1, %xmm0, %xmm0
retq
With this patch we generate a single 'vpblendw'.
llvm-svn: 221343
For 8-bit divrems where the remainder is used, we used to generate:
divb %sil
shrw $8, %ax
movzbl %al, %eax
That was to avoid an H-reg access, which is problematic mainly because
it isn't possible in REX-prefixed instructions.
This patch optimizes that to:
divb %sil
movzbl %ah, %eax
To do that, we explicitly extend AH, and extract the L-subreg in the
resulting register. The extension is done using the NOREX variants of
MOVZX. To support signed operations, MOVSX_NOREX is also added.
Further, this introduces a new SDNode type, [us]divrem_ext_hreg, which is
then lowered to a sequence containing a single zext (rather than 2).
Differential Revision: http://reviews.llvm.org/D6064
llvm-svn: 221176
"[x86] Simplify vector selection if condition value type matches vselect value type and true value is all ones or false value is all zeros."
llvm-svn: 221028
This transformation worked if selector is produced by SETCC, however SETCC is needed only if we consider to swap operands. So I replaced SETCC check for this case.
Added tests for vselect of <X x i1> values.
llvm-svn: 220777
Ffter commit at rev219046 512-bit broadcasts lowering become non-optimal. Most of tests on broadcasting and embedded broadcasting were changed and they doesn’t produce efficient code.
Example below is from commit changes (it’s the first test from test/CodeGen/X86/avx512-vbroadcast.ll):
define <16 x i32> @_inreg16xi32(i32 %a) {
; CHECK-LABEL: _inreg16xi32:
; CHECK: ## BB#0:
-; CHECK-NEXT: vpbroadcastd %edi, %zmm0
+; CHECK-NEXT: vmovd %edi, %xmm0
+; CHECK-NEXT: vpbroadcastd %xmm0, %ymm0
+; CHECK-NEXT: vinserti64x4 $1, %ymm0, %zmm0, %zmm0
; CHECK-NEXT: retq
%b = insertelement <16 x i32> undef, i32 %a, i32 0
%c = shufflevector <16 x i32> %b, <16 x i32> undef, <16 x i32> zeroinitializer
ret <16 x i32> %c
}
Here, 256-bit broadcast was generated instead of 512-bit one.
In this patch
1) I added vector-shuffle lowering through broadcasts
2) Removed asserts and branches likes because this is incorrect
- assert(Subtarget->hasDQI() && "We can only lower v8i64 with AVX-512-DQI");
3) Fixed lowering tests
llvm-svn: 220774
Minor patch to fix an issue in XFormVExtractWithShuffleIntoLoad where a load is unary shuffled, then bitcast (to a type with the same number of elements) before extracting an element.
An undef was created for the second shuffle operand using the original (post-bitcasted) vector type instead of the pre-bitcasted type like the rest of the shuffle node - this was then causing an assertion on the different types later on inside SelectionDAG::getVectorShuffle.
Differential Revision: http://reviews.llvm.org/D5917
llvm-svn: 220592
This is a first step for generating SSE rsqrt instructions for
reciprocal square root calcs when fast-math is allowed.
For now, be conservative and only enable this for AMD btver2
where performance improves significantly - for example, 29%
on llvm/projects/test-suite/SingleSource/Benchmarks/BenchmarkGame/n-body.c
(if we convert the data type to single-precision float).
This patch adds a two constant version of the Newton-Raphson
refinement algorithm to DAGCombiner that can be selected by any target
via a parameter returned by getRsqrtEstimate()..
See PR20900 for more details:
http://llvm.org/bugs/show_bug.cgi?id=20900
Differential Revision: http://reviews.llvm.org/D5658
llvm-svn: 220570
Currently, @llvm.smul.with.overflow.i8 expands to 9 instructions, where
3 are really needed.
This adds X86ISD::UMUL8/SMUL8 SD nodes, and custom lowers them to
MUL8/IMUL8 + SETO.
i8 is a special case because there is no two/three operand variants of
(I)MUL8, so the first operand and return value need to go in AL/AX.
Also, we can't write patterns for these instructions: TableGen refuses
patterns where output operands don't match SDNode results. In this case,
instructions where the output operand is an implicitly defined register.
A related special case (and FIXME) exists for MUL8 (X86InstrArith.td):
// FIXME: Used for 8-bit mul, ignore result upper 8 bits.
// This probably ought to be moved to a def : Pat<> if the
// syntax can be accepted.
[(set AL, (mul AL, GR8:$src)), (implicit EFLAGS)]
Ideally, these go away with UMUL8, but we still need to improve TableGen
support of implicit operands in patterns.
Before this change:
movsbl %sil, %eax
movsbl %dil, %ecx
imull %eax, %ecx
movb %cl, %al
sarb $7, %al
movzbl %al, %eax
movzbl %ch, %esi
cmpl %eax, %esi
setne %al
After:
movb %dil, %al
imulb %sil
seto %al
Also, remove a made-redundant testcase for PR19858, and enable more FastISel
ALU-overflow tests for SelectionDAG too.
Differential Revision: http://reviews.llvm.org/D5809
llvm-svn: 220516
X86 code to lower VSELECT messed a bit with the bits set in the mask of VSELECT
when it knows it can be lowered into BLEND. Indeed, only the high bits need to be
set for those and it optimizes those accordingly.
However, when the mask is a compile time constant, the lowering will be handled
by the generic optimizer and those modifications will generate bad code in the
generic optimizer.
This patch fixes that by preventing the optimization if the VSELECT will be
handled by the generic optimizer.
<rdar://problem/18675020>
llvm-svn: 220242
new vector shuffle lowering.
This is loosely based on a patch by Marius Wachtler to the PR (thanks!).
I refactored it a bi to use std::count_if and a mutable array ref but
the core idea was exactly right. I also added some direct testing of
this case.
I believe PR21137 is now the only remaining regression.
llvm-svn: 219081
shuffles using AVX and AVX2 instructions. This fixes PR21138, one of the
few remaining regressions impacting benchmarks from the new vector
shuffle lowering.
You may note that it "regresses" many of the vperm2x128 test cases --
these were actually "improved" by the naive lowering that the new
shuffle lowering previously did. This regression gave me fits. I had
this patch ready-to-go about an hour after flipping the switch but
wasn't sure how to have the best of both worlds here and thought the
correct solution might be a completely different approach to lowering
these vector shuffles.
I'm now convinced this is the correct lowering and the missed
optimizations shown in vperm2x128 are actually due to missing
target-independent DAG combines. I've even written most of the needed
DAG combine and will submit it shortly, but this part is ready and
should help some real-world benchmarks out.
llvm-svn: 219079
Update the entire regression test suite for the new shuffles. Remove
most of the old testing which was devoted to the old shuffle lowering
path and is no longer relevant really. Also remove a few other random
tests that only really exercised shuffles and only incidently or without
any interesting aspects to them.
Benchmarking that I have done shows a few small regressions with this on
LNT, zero measurable regressions on real, large applications, and for
several benchmarks where the loop vectorizer fires in the hot path it
shows 5% to 40% improvements for SSE2 and SSE3 code running on Sandy
Bridge machines. Running on AMD machines shows even more dramatic
improvements.
When using newer ISA vector extensions the gains are much more modest,
but the code is still better on the whole. There are a few regressions
being tracked (PR21137, PR21138, PR21139) but by and large this is
expected to be a win for x86 generated code performance.
It is also more correct than the code it replaces. I have fuzz tested
this extensively with ISA extensions up through AVX2 and found no
crashes or miscompiles (yet...). The old lowering had a few miscompiles
and crashers after a somewhat smaller amount of fuzz testing.
There is one significant area where the new code path lags behind and
that is in AVX-512 support. However, there was *extremely little*
support for that already and so this isn't a significant step backwards
and the new framework will probably make it easier to implement lowering
that uses the full power of AVX-512's table-based shuffle+blend (IMO).
Many thanks to Quentin, Andrea, Robert, and others for benchmarking
assistance. Thanks to Adam and others for help with AVX-512. Thanks to
Hal, Eric, and *many* others for answering my incessant questions about
how the backend actually works. =]
I will leave the old code path in the tree until the 3 PRs above are at
least resolved to folks' satisfaction. Then I will rip it (and 1000s of
lines of code) out. =] I don't expect this flag to stay around for very
long. It may not survive next week.
llvm-svn: 219046
It turns out this combine was always somewhat flawed -- there are cases
where nested VZEXT nodes *can't* be combined: if their types have
a mismatch that can be observed in the result. While none of these show
up in currently, once I switch to the new vector shuffle lowering a few
test cases actually form such nested VZEXT nodes. I've not come up with
any IR pattern that I can sensible write to exercise this, but it will
be covered by tests once I flip the switch.
llvm-svn: 219044
nodes to the DAG combining of them.
This will allow the combine to fire on both old vector shuffle lowering
and the new vector shuffle lowering and generally seems like a cleaner
design. I've trimmed down the code a bit and tried to make it and the
surrounding combine fairly clean while moving it around.
llvm-svn: 219042
the various ways in which blends can be used to do vector element
insertion for lowering with the scalar math instruction forms that
effectively re-blend with the high elements after performing the
operation.
This then allows me to bail on the element insertion lowering path when
we have SSE4.1 and are going to be doing a normal blend, which in turn
restores the last of the blends lost from the new vector shuffle
lowering when I got it to prioritize insertion in other cases (for
example when we don't *have* a blend instruction).
Without the patterns, using blends here would have regressed
sse-scalar-fp-arith.ll *completely* with the new vector shuffle
lowering. For completeness, I've added RUN-lines with the new lowering
here. This is somewhat superfluous as I'm about to flip the default, but
hey, it shows that this actually significantly changed behavior.
The patterns I've added are just ridiculously repetative. Suggestions on
making them better very much welcome. In particular, handling the
commuted form of the v2f64 patterns is somewhat obnoxious.
llvm-svn: 219033
and MOVSD nodes for single element vector inserts.
This is particularly important because a number of patterns in the
backend detect these patterns and leverage them to simplify things. It
also fixes quite a few of the insertion bad code examples. However, it
regresses a specific area: when available, blendps and blendpd are
*dramatically* faster than movss and movsd respectively. But it doesn't
really work to form the blend logic first because the blends *aren't* as
crazy efficient when the data is coming from memory anyways, and thus
will have a movss or movsd regardless. Also, doing that would block
a bunch of the patterns that this is designed to hit.
So my plan is to go into the patterns for lowering MOVSS and MOVSD and
lower them via blends when available. However that's a pretty invasive
restructuring so it will need to be a follow-up patch.
I have already gone into the patterns to lower MOVSS and MOVSD from
memory using MOVLPD, etc. Without that, several of the test cases
I already have regress.
llvm-svn: 218985
lowering to handle the potential mirroring of 2-element vectors (because
we can't reliably sort them one way) in the caller rather than in the
insertion logic.
This will simplify things considerably as more ways to fail to match the
insertion are added because now we have a nice try and retry point.
llvm-svn: 218980
lowering to match VZEXT_MOVL patterns.
I hadn't realized that these had sufficient pattern smarts in the
backend to lower zext-ing from the low element of a vector without it
being a scalar_to_vector node. They do, and this is how to match a bunch
of patterns for movq, movss, etc.
There is a weird propensity to end up using pshufd to place the element
afterward even though it means domain crossing (or rather, to use
xorps+movss to zext the element rather than movq) but that's an
orthogonal problem with VZEXT_MOVL that someone should probably look at.
llvm-svn: 218977
element types to form illegal vector types.
I've added a special SSE1 test case here that makes sure we don't break
this going forward.
llvm-svn: 218974
elements as well as integer elements in order to form simpler shuffle
patterns.
This is the primary reason why we were failing to match some of the
2-and-2 floating point shuffles such as PR21140. Even after fixing this
we need to support some extra patterns in the backend in order to match
the resulting X86ISD::UNPCKL nodes into the correct instructions. This
commit should fix PR21140 and includes more comprehensive testing of
insertion patterns in v4 shuffles.
Not all of the added tests are beautiful. For example, we don't have
clever instructions to insert-via-load in the integer domain. There are
also some places where we aren't sufficiently cunning with our use of
movq and movd, but that's future work.
llvm-svn: 218911
matching and lowering 64-bit insertions.
The first problem was that we weren't looking through bitcasts to
discover that we *could* lower as insertions. Once fixed, we in turn
weren't looking through bitcasts to discover that we could fold a load
into the lowering. Once fixed, we weren't forming a SCALAR_TO_VECTOR
node around the inserted element and instead were passing a scalar to
a DAG node that expected a vector. It turns out there are some patterns
that will "lower" this into the correct asm, but the rest of the X86
backend is very unhappy with such antics.
This should fix a few more edge case regressions I've spotted going
through the regression test suite to enable the new vector shuffle
lowering.
llvm-svn: 218839
Negative FABS of either a scalar or vector should be handled the same way
on x86 with SSE/AVX: a single OR instruction of the FP operand with a
constant to light up the sign bit(s).
http://llvm.org/bugs/show_bug.cgi?id=20578
Differential Revision: http://reviews.llvm.org/D5201
llvm-svn: 218822
that keep cropping up in the regression test suite.
This also addresses one of the issues raised on the mailing list with
failing to form 'movsd' in as many cases as we realistically should.
There will be corresponding patches forthcoming for v4f32 at least. This
was a lot of fuss for a relatively small gain, but all the fuss was on
my end trying different ways of holding the pieces of the x86 fragment
patterns *just right*. Now that it works, the code is reasonably simple.
In the new test cases I'm adding here, v2i64 sticks out as just plain
horrible. I've not come up with any great ideas here other than that it
would be nice to recognize when we're *going* to take a domain crossing
hit and cross earlier to get the decent instructions. At least with AVX
it is slightly less silly....
llvm-svn: 218756
Nothing was relying on this and there are potentially some edge cases
that it would not be correct under. Removing it seems better than trying
to "fix" it as nothing was relying on it.
llvm-svn: 218755
in exposing the scalar value to the broadcast DAG fragment so that we
can catch even reloads and fold them into the broadcast.
This is somewhat magical I'm afraid but seems to work. It is also what
the old lowering did, and I've switched an old test to run both
lowerings demonstrating that we get the same result.
Unlike the old code, I'm not lowering f32 or f64 scalars through this
path when we only have AVX1. The target patterns include pretty heinous
code to re-cast those as shuffles when the scalar happens to not be
spilled because AVX1 provides no broadcast mechanism from registers
what-so-ever. This is terribly brittle. I'd much rather go through our
generic lowering code to get this. If needed, we can add a peephole to
get even more opportunities to broadcast-from-spill-slots that are
exposed post-RA, but my suspicion is this just doesn't matter that much.
llvm-svn: 218734
the same speed as pshufd but we can fold loads into the pmovzx
instructions.
This fixes some regressions that came up in the regression test suite
for the new vector shuffle lowering.
llvm-svn: 218733
VPBROADCAST.
This has the somewhat expected pervasive impact. I don't know why
I forgot about this. Everything seems good with lots of significant
improvements in the tests.
llvm-svn: 218724
Fixed lowering of this intrinsics in case when mask is v2i1 and v4i1.
Now cmp intrinsics lower in the following way:
(i8 (int_x86_avx512_mask_pcmpeq_q_128
(v2i64 %a), (v2i64 %b), (i8 %mask))) ->
(i8 (bitcast
(v8i1 (insert_subvector undef,
(v2i1 (and (PCMPEQM %a, %b),
(extract_subvector
(v8i1 (bitcast %mask)), 0))), 0))))
llvm-svn: 218669
a flawed direction and causing miscompiles. Read on for details.
Fundamentally, the premise of this patch series was to map
VECTOR_SHUFFLE DAG nodes into VSELECT DAG nodes for all blends because
we are going to *have* to lower to VSELECT nodes for some blends to
trigger the instruction selection patterns of variable blend
instructions. This doesn't actually work out so well.
In order to match performance with the existing VECTOR_SHUFFLE
lowering code, we would need to re-slice the blend in order to fit it
into either the integer or floating point blends available on the ISA.
When coming from VECTOR_SHUFFLE (or other vNi1 style VSELECT sources)
this works well because the X86 backend ensures that these types of
operands to VSELECT get sign extended into '-1' and '0' for true and
false, allowing us to re-slice the bits in whatever granularity without
changing semantics.
However, if the VSELECT condition comes from some other source, for
example code lowering vector comparisons, it will likely only have the
required bit set -- the high bit. We can't blindly slice up this style
of VSELECT. Reid found some code using Halide that triggers this and I'm
hopeful to eventually get a test case, but I don't need it to understand
why this is A Bad Idea.
There is another aspect that makes this approach flawed. When in
VECTOR_SHUFFLE form, we have very distilled information that represents
the *constant* blend mask. Converting back to a VSELECT form actually
can lose this information, and so I think now that it is better to treat
this as VECTOR_SHUFFLE until the very last moment and only use VSELECT
nodes for instruction selection purposes.
My plan is to:
1) Clean up and formalize the target pre-legalization DAG combine that
converts a VSELECT with a constant condition operand into
a VECTOR_SHUFFLE.
2) Remove any fancy lowering from VSELECT during *legalization* relying
entirely on the DAG combine to catch cases where we can match to an
immediate-controlled blend instruction.
One additional step that I'm not planning on but would be interested in
others' opinions on: we could add an X86ISD::VSELECT or X86ISD::BLENDV
which encodes a fully legalized VSELECT node. Then it would be easy to
write isel patterns only in terms of this to ensure VECTOR_SHUFFLE
legalization only ever forms the fully legalized construct and we can't
cycle between it and VSELECT combining.
llvm-svn: 218658
nodes, and rely exclusively on its logic. This removes a ton of
duplication from the blend lowering and centralizes it in one place.
One downside is that it requires a bunch of hacks to make this work with
the current legalization framework. We have to manually speculate one
aspect of legalizing VSELECT nodes to get everything to work nicely
because the existing legalization framework isn't *actually* bottom-up.
The other grossness is that we somewhat duplicate the analysis of
constant blends. I'm on the fence here. If reviewers thing this would
look better with VSELECT when it has constant operands dumping over tho
VECTOR_SHUFFLE, we could go that way. But it would be a substantial
change because currently all of the actual blend instructions are
matched via patterns in the TD files based around VSELECT nodes (despite
them not being perfect fits for that). Suggestions welcome, but at least
this removes the rampant duplication in the backend.
llvm-svn: 218600
X86 target-specific DAG combining that tried to convert VSELECT nodes
into VECTOR_SHUFFLE nodes that it "knew" would lower into
immediate-controlled blend nodes.
Turns out, we have perfectly good lowering of all these VSELECT nodes,
and indeed that lowering already knows how to handle lowering through
BLENDI to immediate-controlled blend nodes. The code just wasn't getting
used much because this thing forced the world to go through the vector
shuffle lowering. Yuck.
This also exposes that I was too aggressive in avoiding domain crossing
in v218588 with that lowering -- when the other option is to expand into
two 128-bit vectors, it is worth domain crossing. Restore that behavior
now that we have nice tests covering it.
The test updates here fall into two camps. One is where previously we
ended up with an unsigned encoding of the blend operand and now we get
a signed encoding. In most of those places there were elaborate comments
explaining exactly what these operands really mean. Rather than that,
just switch these tests to use the nicely decoded comments that make it
obvious that the final shuffle matches.
The other updates are just removing pointless domain crossing by
blending integers with PBLENDW rather than BLENDPS.
llvm-svn: 218589
crossing and generally work more like the blend emission code in the new
vector shuffle lowering.
My goal is to have the new vector shuffle lowering just produce VSELECT
nodes that are either matched here to BLENDI or are legal and matched in
the .td files to specific blend instructions. That seems much cleaner as
there are other ways to produce a VSELECT anyways. =]
No *observable* functionality changed yet, mostly because this code
appears to be near-dead. The behavior of this lowering routine did
change though. This code being mostly dead and untestable will change
with my next commit which will also point some new tests at it.
llvm-svn: 218588
AVX-512.
There is no interesting logic yet. Everything ends up eventually
delegating to the generic code to split the vector and shuffle the
halves. Interestingly, that logic does a significantly better job of
lowering all of these types than the generic vector expansion code does.
Mostly, it lets most of the cases fall back to nice AVX2 code rather
than all the way back to SSE code paths.
Step 2 of basic AVX-512 support in the new vector shuffle lowering. Next
up will be to incrementally add direct support for the basic instruction
set to each type (adding tests first).
llvm-svn: 218585
assertion, making the name generic, and improving the documentation.
Step 1 in adding very primitive support for AVX-512. No functionality
changed yet.
llvm-svn: 218584
vectors.
Someone will need to build the AVX512 lowering, which should follow
AVX1 and AVX2 *very* closely for AVX512F and AVX512BW resp. I've added
a dummy test which is a port of the v8f32 and v8i32 tests from AVX and
AVX2 to v8f64 and v8i64 tests for AVX512F and AVX512BW. Hopefully this
is enough information for someone to implement proper lowering here. If
not, I'll be happy to help, but right now the AVX-512 support isn't
a priority for me.
llvm-svn: 218583
lowerings.
This was hopelessly broken. First, the x86 backend wants '-1' to be the
element value representing true in a boolean vector, and second the
operand order for VSELECT is backwards from the actual x86 instructions.
To make matters worse, the backend is just using '-1' as the true value
to get the high bit to be set. It doesn't actually symbolically map the
'-1' to anything. But on x86 this isn't quite how it works: there *only*
the high bit is relevant. As a consequence weird non-'-1' values like
0x80 actually "work" once you flip the operands to be backwards.
Anyways, thanks to Hal for helping me sort out what these *should* be.
llvm-svn: 218582
new vector shuffle target DAG combines -- it helps to actually test for
the value you want rather than just using an integer in a boolean
context.
Have I mentioned that I loathe implicit conversions recently? :: sigh ::
llvm-svn: 218576
of widening masks.
We can't widen a zeroing mask unless both elements that would be merged
are either zeroed or undef. This is the only way to widen a mask if it
has a zeroed element.
Also clean up the code here by ordering the checks in a more logical way
and by using the symoblic values for undef and zero. I'm actually torn
on using the symbolic values because the existing code is littered with
the assumption that -1 is undef, and moreover that entries '< 0' are the
special entries. While that works with the values given to these
constants, using the symbolic constants actually makes it a bit more
opaque why this is the case.
llvm-svn: 218575
I spotted this by inspection when debugging something else, so I have no
test case what-so-ever, and am not even sure it is possible to
realistically trigger the bug. But this is what was intended here.
llvm-svn: 218565
and in the target shuffle combining when trying to widen vector
elements.
Previously only one of these was correct, and we didn't correctly
propagate zeroing target shuffle masks (which have a different sentinel
value from undef in non- target shuffle masks now). This isn't just
a missed optimization, this caused us to drop zeroing shuffles on the
floor and miscompile code. The added test case is one example of that.
There are other fixes to the test suite as a consequence of this as well
as restoring the undef elements in some of the masks that were lost when
I brought sanity to the actual *value* of the undef and zero sentinels.
I've also just cleaned up some of the PSHUFD and PSHUFLW and PSHUFHW
combining code, but that code really needs to go. It was a nice initial
attempt, but it isn't very principled and the recursive shuffle combiner
is much more powerful.
llvm-svn: 218562
that managed to elude all of my fuzz testing historically. =/
Something changed to allow this code path to actually be exercised and
it was doing bad things. It is especially heavily exercised by the
patterns that emerge when doing AVX shuffles that end up lowered through
the 128-bit code path.
llvm-svn: 218540
layer of tie-breaking sorting, it really helps to check that you're in
a tie first. =] Otherwise the whole thing cycles infinitely. Test case
added, another one found through fuzz testing.
llvm-svn: 218523
AVX support.
New test cases included. Note that none of the existing test cases
covered these buggy code paths. =/ Also, it is clear from this that
SHUFPS and SHUFPD are the most bug prone shuffle instructions in x86. =[
These were all detected by fuzz-testing. (I <3 fuzz testing.)
llvm-svn: 218522
Summary:
I originally tried doing this specifically for X86 in the backend in D5091,
but it was rather brittle and generally running too late to be general.
Furthermore, other targets may want to implement similar optimizations.
So I reimplemented it at the IR-level, fitting it into AtomicExpandPass
as it interacts with that pass (which could not be cleanly done before
at the backend level).
This optimization relies on a new target hook, which is only used by X86
for now, as the correctness of the optimization on other targets remains
an open question. If it is found correct on other targets, it should be
trivial to enable for them.
Details of the optimization are discussed in D5091.
Test Plan: make check-all + a new test
Reviewers: jfb
Subscribers: llvm-commits
Differential Revision: http://reviews.llvm.org/D5422
llvm-svn: 218455
v4f64 and v8f32 shuffles when they are lane-crossing. We have fully
general lane-crossing permutation functions in AVX2 that make this easy.
Part of this also changes exactly when and how these vectors are split
up when we don't have AVX2. This isn't always a win but it usually is
a win, so on the balance I think its better. The primary regressions are
all things that just need to be fixed anyways such as modeling when
a blend can be completely accomplished via VINSERTF128, etc.
Also, this highlights one of the few remaining big features: we do
a really poor job of inserting elements into AVX registers efficiently.
This completes almost all of the big tricks I have in mind for AVX2. The
only things left that I plan to add:
1) element insertion smarts
2) palignr and other fairly specialized lowerings when they happen to
apply
llvm-svn: 218449
256-bit vectors with lane-crossing.
Rather than immediately decomposing to 128-bit vectors, try flipping the
256-bit vector lanes, shuffling them and blending them together. This
reduces our worst case shuffle by a pretty significant margin across the
board.
llvm-svn: 218446
lowering where it only used the mask of the low 128-bit lane rather than
the entire mask.
This allows the new lowering to correctly match the unpack patterns for
v8i32 vectors.
For reference, the reason that we check for the the entire mask rather
than checking the repeated mask is because the repeated masks don't
abide by all of the invariants of normal masks. As a consequence, it is
safer to use the full mask with functions like the generic equivalence
test.
llvm-svn: 218442
reduce the amount of checking we do here.
The first realization is that only non-crossing cases between 128-bit
lanes are handled by almost the entire function. It makes more sense to
handle the crossing cases first.
THe second is that until we actually are going to generate fancy shared
lowering strategies that use the repeated semantics of the v8i16
lowering, we should waste time checking for repeated masks. It is
simplest to directly test for the entire unpck masks anyways, so we
gained nothing from this.
This also matches the structure of v32i8 more closely.
No functionality changed here.
llvm-svn: 218441
lowering.
This completes the basic AVX2 feature support, but there are still some
improvements I'd like to do to really get the last mile of performance
here.
llvm-svn: 218440
for this now.
Should prevent folks from running afoul of this and not knowing why
their code won't instruction select the way I just did...
llvm-svn: 218436
missing test cases for it.
Unsurprisingly, without test cases, there were bugs here. Surprisingly,
this bug wasn't caught at compile time. Yep, there is an X86ISD::BLENDV.
It isn't wired to anything. Oops. I'll fix than next.
llvm-svn: 218434
lowering.
This also implements the fancy blend lowering for v16i16 using AVX2 and
teaches the X86 backend to print shuffle masks for 256-bit PSHUFB
and PBLENDW instructions. It also makes the mask decoding correct for
PBLENDW instructions. The yaks, they are legion.
Tests are updated accordingly. There are some missing tests for the
VBLENDVB lowering, but I'll add those in a follow-up as this commit has
accumulated enough cruft already.
llvm-svn: 218430
into unblended shuffles and a blend.
This is the consistent fallback for the lowering paths that have fast
blend operations available, and its getting quite repetitive.
No functionality changed.
llvm-svn: 218399
the native AVX2 instructions.
Note that the test case is really frustrating here because VPERMD
requires the mask to be in the register input and we don't produce
a comment looking through that to the constant pool. I'm going to
attempt to improve this in a subsequent commit, but not sure if I will
succeed.
llvm-svn: 218347
detection. It was incorrectly handling undef lanes by actually treating
an undef lane in the first 128-bit lane as a *numeric* shuffle value.
Fortunately, this almost always DTRT and disabled detecting repeated
patterns. But not always. =/ This patch introduces a much more
principled approach and fixes the miscompiles I spotted by inspection
previously.
llvm-svn: 218346
shuffles using the AVX2 instructions. This is the first step of cutting
in real AVX2 support.
Note that I have spotted at least one bug in the test cases already, but
I suspect it was already present and just is getting surfaced. Will
investigate next.
llvm-svn: 218338
add VPBLENDD to the InstPrinter's comment generation so we get nice
comments everywhere.
Now that we have the nice comments, I can see the bug introduced by
a silly typo in the commit that enabled VPBLENDD, and have fixed it. Yay
tests that are easy to inspect.
llvm-svn: 218335
Summary:
AtomicExpand already had logic for expanding wide loads and stores on LL/SC
architectures, and for expanding wide stores on CmpXchg architectures, but
not for wide loads on CmpXchg architectures. This patch fills this hole,
and makes use of this new feature in the X86 backend.
Only one functionnal change: we now lose the SynchScope attribute.
It is regrettable, but I have another patch that I will submit soon that will
solve this for all of AtomicExpand (it seemed better to split it apart as it
is a different concern).
Test Plan: make check-all (lots of tests for this functionality already exist)
Reviewers: jfb
Subscribers: llvm-commits
Differential Revision: http://reviews.llvm.org/D5404
llvm-svn: 218332
VPBLENDD where appropriate even on 128-bit vectors.
According to Agner's tables, this instruction is significantly higher
throughput (can execute on any port) on Haswell chips so we should
aggressively try to form it when available.
Sadly, this loses our delightful shuffle comments. I'll add those back
for VPBLENDD next.
llvm-svn: 218322
undef in the shuffle mask. This shows up when we're printing comments
during lowering and we still have an IR-level constant hanging around
that models undef.
A nice consequence of this is *much* prettier test cases where the undef
lanes actually show up as undef rather than as a particular set of
values. This also allows us to print shuffle comments in cases that use
undef such as the recently added variable VPERMILPS lowering. Now those
test cases have nice shuffle comments attached with their details.
The shuffle lowering for PSHUFB has been augmented to use undef, and the
shuffle combining has been augmented to comprehend it.
llvm-svn: 218301
trick that I missed.
VPERMILPS has a non-immediate memory operand mode that allows it to do
asymetric shuffles in the two 128-bit lanes. Use this rather than two
shuffles and a blend.
However, it turns out the variable shuffle path to VPERMILPS (and
VPERMILPD, although that one offers no functional differenc from the
immediate operand other than variability) wasn't even plumbed through
codegen. Do such plumbing so that we can reasonably emit
a variable-masked VPERMILP instruction. Also plumb basic comment parsing
and printing through so that the tests are reasonable.
There are still a few tests which don't show the shuffle pattern. These
are tests with undef lanes. I'll teach the shuffle decoding and printing
to handle undef mask entries in a follow-up. I've looked at the masks
and they seem reasonable.
llvm-svn: 218300
td pattern). Currently we only model the immediate operand variation of
VPERMILPS and VPERMILPD, we should make that clear in the pseudos used.
Will be adding support for the variable mask variant in my next commit.
llvm-svn: 218282
We generate broadcast instructions on CPUs with AVX2 to load some constant splat vectors.
This patch should preserve all existing behavior with regular optimization levels,
but also use splats whenever possible when optimizing for *size* on any CPU with AVX or AVX2.
The tradeoff is up to 5 extra instruction bytes for the broadcast instruction to save
at least 8 bytes (up to 31 bytes) of constant pool data.
Differential Revision: http://reviews.llvm.org/D5347
llvm-svn: 218263
Summary:
Update segmented-stacks*.ll tests with x32 target case and make
corresponding changes to make them pass.
Test Plan: tests updated with x32 target
Reviewers: nadav, rafael, dschuff
Subscribers: llvm-commits, zinovy.nis
Differential Revision: http://reviews.llvm.org/D5245
llvm-svn: 218247
a more sane approach to AVX2 support.
Fundamentally, there is no useful way to lower integer vectors in AVX.
None. We always end up with a VINSERTF128 in the end, so we might as
well eagerly switch to the floating point domain and do everything
there. This cleans up lots of weird and unlikely to be correct
differences between integer and floating point shuffles when we only
have AVX1.
The other nice consequence is that by doing things this way we will make
it much easier to write the integer lowering routines as we won't need
to duplicate the logic to check for AVX vs. AVX2 in each one -- if we
actually try to lower a 256-bit vector as an integer vector, we have
AVX2 and can rely on it. I think this will make the code much simpler
and more comprehensible.
Currently, I've disabled *all* support for AVX2 so that we always fall
back to AVX. This keeps everything working rather than asserting. That
will go away with the subsequent series of patches that provide
a baseline AVX2 implementation.
Please note, I'm going to implement AVX2 *without access to hardware*.
That means I cannot correctness test this path. I will be relying on
those with access to AVX2 hardware to do correctness testing and fix
bugs here, but as a courtesy I'm trying to sketch out the framework for
the new-style vector shuffle lowering in the context of the AVX2 ISA.
llvm-svn: 218228
input v8f32 shuffles which are not 128-bit lane crossing but have
different shuffle patterns in the low and high lanes. This removes most
of the extract/insert traffic that was unnecessary and is particularly
good at lowering cases where only one of the two lanes is shuffled at
all.
I've also added a collection of test cases with undef lanes because this
lowering is somewhat more sensitive to undef lanes than others.
llvm-svn: 218226
lowering when it can use a symmetric SHUFPS across both 128-bit lanes.
This required making the SHUFPS lowering tolerant of other vector types,
and adjusting our canonicalization to canonicalize harder.
This is the last of the clever uses of symmetry I've thought of for
v8f32. The rest of the tricks I'm aware of here are to work around
assymetry in the mask.
llvm-svn: 218216
a generic vector shuffle mask into a helper that isn't specific to the
other things that influence which choice is made or the specific types
used with the instruction.
No functionality changed.
llvm-svn: 218215
of a single element into a zero vector for v4f64 and v4i64 in AVX.
Ironically, there is less to see here because xor+blend is so crazy fast
that we can't really beat that to zero the high 128-bit lane.
llvm-svn: 218214
UNPCKHPS with AVX vectors by recognizing those patterns when they are
repeated for both 128-bit lanes.
With this, we now generate the exact same (really nice) code for
Quentin's avx_test_case.ll which was the most significant regression
reported for the new shuffle lowering. In fact, I'm out of specific test
cases for AVX lowering, the rest were AVX2 I think. However, there are
a bunch of pretty obvious remaining things to improve with AVX...
llvm-svn: 218213
important bits of cleverness: to detect and lower repeated shuffle
patterns between the two 128-bit lanes with a single instruction.
This patch just teaches it how to lower single-input shuffles that fit
this model using VPERMILPS. =] There is more that needs to happen here.
llvm-svn: 218211
v8f32 shuffles in the new vector shuffle lowering code.
This is very cheap to do and makes it much more clear that anything more
expensive but overlapping with this lowering should be selected
afterward (for example using AVX2's VPERMPS). However, no functionality
changed here as without this code we would fall through to create no-op
shuffles of each input and a blend. =]
llvm-svn: 218209
VBLENDPD over using VSHUFPD. While the 256-bit variant of VBLENDPD slows
down to the same speed as VSHUFPD on Sandy Bridge CPUs, it has twice the
reciprocal throughput on Ivy Bridge CPUs much like it does everywhere
for 128-bits. There isn't a downside, so just eagerly use this
instruction when it suffices.
llvm-svn: 218208
awkward conditions. The readability improvement of this will be even
more important as I generalize it to handle more types.
No functionality changed.
llvm-svn: 218205
128-bit lane crossings, not 'half' crossings. This came up in code
review ages ago, but I hadn't really addresesd it. Also added some
documentation for the helper.
No functionality changed.
llvm-svn: 218203
actual support for complex AVX shuffling tricks. We can do independent
blends of the low and high 128-bit lanes of an avx vector, so shuffle
the inputs into place and then do the blend at 256 bits. This will in
many cases remove one blend instruction.
The next step is to permute the low and high halves in-place rather than
extracting them and re-inserting them.
llvm-svn: 218202
single-input shuffles with doubles. This allows them to fold memory
operands into the shuffle, etc. This is just the analog to the v4f32
case in my prior commit.
llvm-svn: 218193
instruction for single-vector floating point shuffles. This in turn
allows the shuffles to fold a load into the instruction which is one of
the common regressions hit with the new shuffle lowering.
llvm-svn: 218190
tricky case of single-element insertion into the zero lane of a zero
vector.
We can't just use the same pattern here as we do in every other vector
type because the general insertion logic can handle insertion into the
non-zero lane of the vector. However, in SSE4.1 with v4f32 vectors we
have INSERTPS that is a much better choice than the generic one for such
lowerings. But INSERTPS can do lots of other lowerings as well so
factoring its logic into the general insertion logic doesn't work very
well. We also can't just extract the core common part of the general
insertion logic that is faster (forming VZEXT_MOVL synthetic nodes that
lower to MOVSS when they can) because VZEXT_MOVL is often *faster* than
a blend while INSERTPS is slower! So instead we do a restrictive
condition on attempting to use the generic insertion logic to narrow it
to those cases where VZEXT_MOVL won't need a shuffle afterward and thus
will do better than INSERTPS. Then we try blending. Then we go back to
INSERTPS.
This still doesn't generate perfect code for some silly reasons that can
be fixed by tweaking the td files for lowering VZEXT_MOVL to use
XORPS+BLENDPS when available rather than XORPS+MOVSS when the input ends
up in a register rather than a load from memory -- BLENDPSrr has twice
the reciprocal throughput of MOVSSrr. Don't you love this ISA?
llvm-svn: 218177
analysis used elsewhere. This removes the last duplicate of this logic.
Also simplify the code here quite a bit. No functionality changed.
llvm-svn: 218176
floating point types and use it for both v2f64 and v2i64 single-element
insertion lowering.
This fixes the last non-AVX performance regression test case I've gotten
of for the new vector shuffle lowering. There is obvious analogous
lowering for v4f32 that I'll add in a follow-up patch (because with
INSERTPS, v4f32 requires special treatment). After that, its AVX stuff.
llvm-svn: 218175
vector lanes can be modeled as zero with a call to the new function that
computes a bit-vector representing that information.
No functionality changed here, but will allow doing more clever things
with the zero-test.
llvm-svn: 218174
lowering to support both anyext and zext and to custom lower for many
different microarchitectures.
Using this allows us to get *exactly* the right code for zext and anyext
shuffles in all the vector sizes. For v16i8, the improvement is *huge*.
The new SSE2 test case added I refused to add before this because it was
sooooo muny instructions.
llvm-svn: 218143
to undef lanes as well as defined widenable lanes. This dramatically
improves the lowering we use for undef-shuffles in a zext-ish pattern
for SSE2.
llvm-svn: 218115
shuffles that are zext-ing.
Not a lot to see here; the undef lane variant is better handled with
pshufd, but this improves the actual zext pattern.
llvm-svn: 218112
to the new vector shuffle lowering code.
This allows us to emit PMOVZX variants consistently for patterns where
it is a viable lowering. This instruction is both fast and allows us to
fold loads into it. This only hooks the new lowering up for i16 and i8
element widths, mostly so I could manage the change to the tests. I'll
add the i32 one next, although it is significantly less interesting.
One thing to note is that we already had some tests for these patterns
but those tests had far less horrible instructions. The problem is that
those tests weren't checking the strict start and end of the instruction
sequence. =[ As a consequence something changed in the lowering making
us generate *TERRIBLE* code for these patterns in SSE2 through SSSE3.
I've consolidated all of the tests and spelled out the madness that we
currently emit for these shuffles. I'm going to try to figure out what
has gone wrong here.
llvm-svn: 218102
There is no purpose in using it for single-input shuffles as
pshufd is just as fast and doesn't tie the two operands. This removes
a substantial amount of wrong-domain blend operations in SSSE3 mode. It
also completes the usage of PALIGNR for integer shuffles and addresses
one of the test cases Quentin hit with the new vector shuffle lowering.
There is still the question of whether and when to use this for floating
point shuffles. It is faster than shufps or shufpd but in the integer
domain. I don't yet really have a good heuristic here for when to use
this instruction for floating point vectors.
llvm-svn: 218038
PALIGNR. This just adds it to the v8i16 and v16i8 lowering steps where
it is completely unmatched. It also introduces the logic for detecting
rotation shuffle masks even in the presence of single input or blend
masks and arbitrarily undef lanes.
I've added fairly comprehensive tests for the matching logic in v8i16
because the tests at that size are much easier to write and manage.
I've not checked the SSE2 code generated for these tests because the
code is *horrible*. It is absolute madness. Testing it will just make
the test brittle without giving any interesting improvements in the
correctness confidence.
llvm-svn: 218013
This required a new hook called hasLoadLinkedStoreConditional to know whether
to expand atomics to LL/SC (ARM, AArch64, in a future patch Power) or to
CmpXchg (X86).
Apart from that, the new code in AtomicExpandPass is mostly moved from
X86AtomicExpandPass. The main result of this patch is to get rid of that
pass, which had lots of code duplicated with AtomicExpandPass.
llvm-svn: 217928
the blend that is matched by this are "used" in any sense, and so any
build_vector or other nodes feeding these will already drop other lanes.
llvm-svn: 217855
matching. This design just fundamentally didn't work because ADDSUB is
available prior to any legal lowerings of BLENDI nodes. Instead, we have
a dedicated ADDSUB synthetic ISD node which is pattern matched trivially
into the instructions. These nodes are then recognized by both the
existing and a trivial new lowering combine in the backend. Removing
these patterns required adding 2 missing shuffle masks to the DAG
combine, without which tests would have failed. Added the masks and
a helpful assert as well to catch if anything ever goes wrong here.
llvm-svn: 217851
that we don't use VSELECT and directly emit an addsub synthetic node.
Also remove a stale comment referencing VSELECT.
The test case is updated to use 'core2' which only has SSE3, not SSE4.1,
and it still passes. Previously it would not because we lacked
sufficient blend support to legalize the VSELECT.
llvm-svn: 217849
ADDSUBPD nodes out of blends of adds and subs.
This allows us to actually form these instructions with SSE3 rather than
only forming them when we had both SSE3 for the ADDSUB instructions and
SSE4.1 for the blend instructions. ;] Kind-of important.
I've adjusted the CPU requirements on one of the tests to demonstrate
this kicking in nicely for an SSE3 cpu configuration.
llvm-svn: 217848
introducing a synthetic X86 ISD node representing this generic
operation.
The relevant patterns for mapping these nodes into the concrete
instructions are also added, and a gnarly bit of C++ code in the
target-specific DAG combiner is replaced with simple code emitting this
primitive.
The next step is to generically combine blends of adds and subs into
this node so that we can drop the reliance on an SSE4.1 ISD node
(BLENDI) when matching an SSE3 feature (ADDSUB).
llvm-svn: 217819
when SSE4.1 is available.
This removes a ton of domain crossing from blend code paths that were
ending up in the floating point code path.
This is just the tip of the iceberg though. The real switch is for
integer blend lowering to more actively rely on this instruction being
available so we don't hit shufps at all any longer. =] That will come in
a follow-up patch.
Another place where we need better support is for using PBLENDVB when
doing so avoids the need to have two complementary PSHUFB masks.
llvm-svn: 217767
instructions from the relevant shuffle patterns.
This is the last tweak I'm aware of to generate essentially perfect
v4f32 and v2f64 shuffles with the new vector shuffle lowering up through
SSE4.1. I'm sure I've missed some and it'd be nice to check since v4f32
is amenable to exhaustive exploration, but this is all of the tricks I'm
aware of.
With AVX there is a new trick to use the VPERMILPS instruction, that's
coming up in a subsequent patch.
llvm-svn: 217761
instructions when it finds an appropriate pattern.
These are lovely instructions, and its a shame to not use them. =] They
are fast, and can hand loads folded into their operands, etc.
I've also plumbed the comment shuffle decoding through the various
layers so that the test cases are printed nicely.
llvm-svn: 217758
AVX is available, and generally tidy up things surrounding UNPCK
formation.
Originally, I was thinking that the only advantage of PSHUFD over UNPCK
instruction variants was its free copy, and otherwise we should use the
shorter encoding UNPCK instructions. This isn't right though, there is
a larger advantage of being able to fold a load into the operand of
a PSHUFD. For UNPCK, the operand *must* be in a register so it can be
the second input.
This removes the UNPCK formation in the target-specific DAG combine for
v4i32 shuffles. It also lifts the v8 and v16 cases out of the
AVX-specific check as they are potentially replacing multiple
instructions with a single instruction and so should always be valuable.
The floating point checks are simplified accordingly.
This also adjusts the formation of PSHUFD instructions to attempt to
match the shuffle mask to one which would fit an UNPCK instruction
variant. This was originally motivated to allow it to match the UNPCK
instructions in the combiner, but clearly won't now.
Eventually, we should add a MachineCombiner pass that can form UNPCK
instructions post-RA when the operand is known to be in a register and
thus there is no loss.
llvm-svn: 217755
'punpckhwd' instructions when suitable rather than falling back to the
generic algorithm.
While we could canonicalize to these patterns late in the process, that
wouldn't help when the freedom to use them is only visible during
initial lowering when undef lanes are well understood. This, it turns
out, is very important for matching the shuffle patterns that are used
to lower sign extension. Fixes a small but relevant regression in
gcc-loops with the new lowering.
When I changed this I noticed that several 'pshufd' lowerings became
unpck variants. This is bad because it removes the ability to freely
copy in the same instruction. I've adjusted the widening test to handle
undef lanes correctly and now those will correctly continue to use
'pshufd' to lower. However, this caused a bunch of churn in the test
cases. No functional change, just churn.
Both of these changes are part of addressing a general weakness in the
new lowering -- it doesn't sufficiently leverage undef lanes. I've at
least a couple of patches that will help there at least in an academic
sense.
llvm-svn: 217752
These are super simple. They even take precedence over crazy
instructions like INSERTPS because they have very high throughput on
modern x86 chips.
I still have to teach the integer shuffle variants about this to avoid
so many domain crossings. However, due to the particular instructions
available, that's a touch more complex and so a separate patch.
Also, the backend doesn't seem to realize it can commute blend
instructions by negating the mask. That would help remove a number of
copies here. Suggestions on how to do this welcome, it's an area I'm
less familiar with.
llvm-svn: 217744
support transforming the forms from the new vector shuffle lowering to
use 'movddup' when appropriate.
A bunch of the cases where we actually form 'movddup' don't actually
show up in the test results because something even later than DAG
legalization maps them back to 'unpcklpd'. If this shows back up as
a performance problem, I'll probably chase it down, but it is at least
an encoded size loss. =/
To make this work, also always do this canonicalizing step for floating
point vectors where the baseline shuffle instructions don't provide any
free copies of their inputs. This also causes us to canonicalize
unpck[hl]pd into mov{hl,lh}ps (resp.) which is a nice encoding space
win.
There is one test which is "regressed" by this: extractelement-load.
There, the test case where the optimization it is testing *fails*, the
exact instruction pattern which results is slightly different. This
should probably be fixed by having the appropriate extract formed
earlier in the DAG, but that would defeat the purpose of the test.... If
this test case is critically important for anyone, please let me know
and I'll try to work on it. The prior behavior was actually contrary to
the comment in the test case and seems likely to have been an accident.
llvm-svn: 217738
r189189 implemented AVX512 unpack by essentially performing a 256-bit unpack
between the low and the high 256 bits of src1 into the low part of the
destination and another unpack of the low and high 256 bits of src2 into the
high part of the destination.
I don't think that's how unpack works. AVX512 unpack simply has more 128-bit
lanes but other than it works the same way as AVX. So in each 128-bit lane,
we're always interleaving certain parts of both operands rather different
parts of one of the operands.
E.g. for this:
__v16sf a = { 0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15 };
__v16sf b = { 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31 };
__v16sf c = __builtin_shufflevector(a, b, 0, 8, 1, 9, 4, 12, 5, 13, 16,
24, 17, 25, 20, 28, 21, 29);
we generated punpcklps (notice how the elements of a and b are not interleaved
in the shuffle). In turn, c was set to this:
0 16 1 17 4 20 5 21 8 24 9 25 12 28 13 29
Obviously this should have just returned the mask vector of the shuffle
vector.
I mostly reverted this change and made sure the original AVX code worked
for 512-bit vectors as well.
Also updated the tests because they matched the logic from the code.
llvm-svn: 217602
When compiling without SSE2, isTruncStoreLegal(F64, F32) would return Legal, whereas with SSE2 it would return Expand. And since the Target doesn't seem to actually handle a truncstore for double -> float, it would just output a store of a full double in the space for a float hence overwriting other bits on the stack.
Patch by Luqman Aden!
llvm-svn: 217410
support for MOVDDUP which is really important for matrix multiply style
operations that do lots of non-vector-aligned load and splats.
The original motivation was to add support for MOVDDUP as the lack of it
regresses matmul_f64_4x4 by 5% or so. However, all of the rules here
were somewhat suspicious.
First, we should always be using the floating point domain shuffles,
regardless of how many copies we have to make as a movapd is *crazy*
faster than the domain switching cost on some chips. (Mostly because
movapd is crazy cheap.) Because SHUFPD can't do the copy-for-free trick
of the PSHUF instructions, there is no need to avoid canonicalizing on
UNPCK variants, so do that canonicalizing. This also ensures we have the
chance to form MOVDDUP. =]
Second, we assume SSE2 support when doing any vector lowering, and given
that we should just use UNPCKLPD and UNPCKHPD as they can operate on
registers or memory. If vectors get spilled or come from memory at all
this is going to allow the load to be folded into the operation. If we
want to optimize for encoding size (the only difference, and only
a 2 byte difference) it should be done *much* later, likely after RA.
llvm-svn: 217332
computation was totally wrong, but somehow it didn't really show up with
llc.
I've added an assert that triggers on multiple existing test cases and
updated one of them to show the correct value.
There appear to still be more bugs lurking around insertps's mask. =/
However, note that this only really impacts the new vector shuffle
lowering.
llvm-svn: 217289
shuffle lowering for integer vectors and share it from v4i32, v8i16, and
v16i8 code paths.
Ironically, the SSE2 v16i8 code for this is now better than the SSSE3!
=] Will have to fix the SSSE3 code next to just using a single pshufb.
llvm-svn: 217240
vzext patterns and insert-element patterns that for SSE4 have dedicated
instructions.
With this we can enable the experimental mode in a regression test that
happens to cover some of the past set of issues. You can see that the
new logic does significantly better here on the floating point cases.
A follow-up to this change and the previous ones will hoist the logic
into helpers so it can be shared across element type sizes as in this
particular case it generalizes cleanly.
llvm-svn: 217136
abilities of INSERTPS which are really powerful and come up in very
important contexts such as forming diagonal matrices, etc.
With this I ended up being able to remove the somewhat weird helper
I added for INSERTPS because we can collapse the entire state to a no-op
mask. Added a bunch of tests for inserting into a zero-ish vector.
llvm-svn: 217117
'insertps' patterns.
This replaces two shuffles with a single insertps in very common cases.
My next patch will extend this to leverage the zeroing capabilities of
insertps which will allow it to be used in a much wider set of cases.
llvm-svn: 217100
We duplicate ~30 lines of code to lower FABS and FNEG for x86, so this patch combines them into one function.
No functional change intended, so no additional test cases. Test-suite behavior is unchanged.
Differential Revision: http://reviews.llvm.org/D5064
llvm-svn: 216942
This change will ease refactoring LowerFABS() and LowerFNEG()
since they have a lot of overlap.
Remove the creation of a floating point constant from an integer
because it's going to be used for a bitwise integer op anyway.
No change to codegen expected, but the verbose comment string
for asm output may change from float values to hex (integer),
depending on whether the constant already exists or not.
Differential Revision: http://reviews.llvm.org/D5052
llvm-svn: 216889
Summary:
If a variadic function body contains a musttail call, then we copy all
of the remaining register parameters into virtual registers in the
function prologue. We track the virtual registers through the function
body, and add them as additional registers to pass to the call. Because
this is all done in virtual registers, the register allocator usually
gives us good code. If the function does a call, however, it will have
to spill and reload all argument registers (ew).
Forwarding regparms on x86_32 is not implemented because most compilers
don't support varargs in 32-bit with regparms.
Reviewers: majnemer
Subscribers: aemerson, llvm-commits
Differential Revision: http://reviews.llvm.org/D5060
llvm-svn: 216780
We've rejected these kinds of functions since r28405 in 2006 because
it's impossible to lower the return of a callee cleanup varargs
function. However there are lots of legal ways to leave such a function
without returning, such as aborting. Today we can leave a function with
a musttail call to another function with the correct prototype, and
everything works out.
I'm removing the verifier check declaring that a normal return from such
a function is UB.
Reviewed By: nlewycky
Differential Revision: http://reviews.llvm.org/D5059
llvm-svn: 216779
functionality changed.
Separating this into two functions wasn't helping. There was a decent
amount of boilerplate duplicated, and some subsequent refactorings here
will pull even more common code out.
llvm-svn: 216644
we stopped efficiently lowering sextload using the SSE41 instructions
for that operation.
This is a consequence of a bad predicate I used thinking of the memory
access needs. The code actually handles the cases where the predicate
doesn't apply, and handles them much better. =] Simple fix and a test
case added. Fixes PR20767.
llvm-svn: 216538
This combine is essentially combining target-specific nodes back into target
independent nodes that it "knows" will be combined yet again by a target
independent DAG combine into a different set of target-independent nodes that
are legal (not custom though!) and thus "ok". This seems... deeply flawed. The
crux of the problem is that we don't combine un-legalized shuffles that are
introduced by legalizing other operations, and thus we don't see a very
profitable combine opportunity. So the backend just forces the input to that
combine to re-appear.
However, for this to work, the conditions detected to re-form the unlegalized
nodes must be *exactly* right. Previously, failing this would have caused poor
code (if you're lucky) or a crasher when we failed to select instructions.
After r215611 we would fall back into the legalizer. In some cases, this just
"fixed" the crasher by produces bad code. But in the test case added it caused
the legalizer and the dag combiner to iterate forever.
The fix is to make the alignment checking in the x86 side of things match the
alignment checking in the generic DAG combine exactly. This isn't really a
satisfying or principled fix, but it at least make the code work as intended.
It also highlights that it would be nice to detect the availability of under
aligned loads for a given type rather than bailing on this optimization. I've
left a FIXME to document this.
Original commit message for r215611 which covers the rest of the chang:
[SDAG] Fix a case where we would iteratively legalize a node during
combining by replacing it with something else but not re-process the
node afterward to remove it.
In a truly remarkable stroke of bad luck, this would (in the test case
attached) end up getting some other node combined into it without ever
getting re-processed. By adding it back on to the worklist, in addition
to deleting the dead nodes more quickly we also ensure that if it
*stops* being dead for any reason it makes it back through the
legalizer. Without this, the test case will end up failing during
instruction selection due to an and node with a type we don't have an
instruction pattern for.
It took many million runs of the shuffle fuzz tester to find this.
llvm-svn: 216537
This actually was caught by existing tests but those tests were disabled
with an XFAIL because of PR20736. While working on fixing that,
I noticed the test failure, and tracked it down to this.
We even have a really nice Clang warning that would have caught this but
it isn't enabled in LLVM! =[ I may look at enabling it.
llvm-svn: 216391
these DAG combines.
The DAG auto-CSE thing is truly terrible. Due to it, when RAUW-ing
a node with its operand, you can cause its uses to CSE to itself, which
then causes their uses to become your uses which causes them to be
picked up by the RAUW. For nodes that are determined to be "no-ops",
this is "fine". But if the RAUW is one of several steps to enact
a transformation, this causes the DAG to really silently eat an discard
nodes that you would never expect. It took days for me to actually
pinpoint a test case triggering this and a really frustrating amount of
time to even comprehend the bug because I never even thought about the
ability of RAUW to iteratively consume nodes due to CSE-ing them into
itself.
To fix this, we have to build up a brand-new chain of operations any
time we are combining across (potentially) intervening nodes. But once
the logic is added to do this, another issue surfaces: CombineTo eagerly
deletes the one node combined, *but no others*. This is... really
frustrating. If deleting it makes its operands become dead, those
operand nodes often won't go onto the worklist in the
order you would want -- they're already on it and not near the top. That
means things higher on the worklist will get combined prior to these
dead nodes being GCed out of the worklist, and if the chain is long, the
immediate users won't be enough to re-detect where the root of the chain
is that became single-use again after deleting the dead nodes. The
better way to do this is to never immediately delete nodes, and instead
to just enqueue them so we can recursively delete them. The
combined-from node is typically not on the worklist anyways by virtue of
having been popped off.... But that in turn breaks other tests that
*require* CombineTo to delete unused nodes. :: sigh ::
Fortunately, there is a better way. This whole routine should have been
returning the replacement rather than using CombineTo which is quite
hacky. Switch to that, and all the pieces fall together.
I suspect the same kind of miscompile is possible in the half-shuffle
folding code, and potentially the recursive folding code. I'll be
switching those over to a pattern more like this one for safety's sake
even though I don't immediately have any test cases for them. Note that
the only way I got a test case for this instance was with *heavily* DAG
combined 256-bit shuffle sequences generated by my fuzzer. ;]
llvm-svn: 216319
There's no need to do this if the user doesn't call va_start. In the
future, we're going to have thunks that forward these register
parameters with musttail calls, and they won't need these spills for
handling va_start.
Most of the test suite changes are adding va_start calls to existing
tests to keep things working.
llvm-svn: 216294
This (mostly) reverts commit r216119.
Somewhere during the review Reid committed r214980 which fixed this
another way, and I neglected to check that the testcase still failed
before committing.
I've left test/CodeGen/X86/aligned-variadic.ll around in case it adds
extra coverage.
llvm-svn: 216246
Fix for PR20648 - http://llvm.org/bugs/show_bug.cgi?id=20648
This patch checks the operands of a vselect to see if all values are constants.
If yes, bail out of any further attempts to create a blend or shuffle because
SelectionDAGLegalize knows how to turn this kind of vselect into a single load.
This already happens for machines without SSE4.1, so the added checks just send
more targets down that path.
Differential Revision: http://reviews.llvm.org/D4934
llvm-svn: 216121
The goal of the patch is to implement section 3.2.3 of the AMD64 ABI
correctly. The controlling sentence is, "The size of each argument gets
rounded up to eightbytes. Therefore the stack will always be eightbyte
aligned." The equivalent sentence in the i386 ABI page 37 says, "At all
times, the stack pointer should point to a word-aligned area." For both
architectures, the stack pointer is not being rounded up to the nearest
eightbyte or word between the last normal argument and the first
variadic argument.
Patch by Thomas Jablin!
llvm-svn: 216119
Summary: This fixes http://llvm.org/bugs/show_bug.cgi?id=19530.
The problem is that X86ISelLowering erroneously thought the third call
was eligible for tail call elimination.
It would have been if it's return value was actually the one returned
by the calling function, but here that is not the case and
additional values are being returned.
Test Plan: Test case from the original bug report is included.
Reviewers: rafael
Reviewed By: rafael
Subscribers: rafael, llvm-commits
Differential Revision: http://reviews.llvm.org/D4968
llvm-svn: 216117
It should remove dosens of lines in handling instrinsics (in a huge switch) and give an easy way to add new intrinsics.
I did not completed to move al intrnsics to the table, I'll do this in the upcomming commits.
llvm-svn: 215826
MSVC gives this awesome diagnostic:
..\lib\Target\X86\X86ISelLowering.cpp(7085) : error C2971: 'llvm::VariadicFunction1' : template parameter 'Func' : 'isShuffleEquivalentImpl' : a local variable cannot be used as a non-type argument
..\include\llvm/ADT/VariadicFunction.h(153) : see declaration of 'llvm::VariadicFunction1'
..\lib\Target\X86\X86ISelLowering.cpp(7061) : see declaration of 'isShuffleEquivalentImpl'
Using an anonymous namespace makes the problem go away.
llvm-svn: 215744
the new shuffle lowering and an implementation for v4 shuffles.
This allows us to handle non-half-crossing shuffles directly for v4
shuffles, both integer and floating point. This currently misses places
where we could perform the blend via UNPCK instructions, but otherwise
generates equally good or better code for the test cases included to the
existing vector shuffle lowering. There are a few cases that are
entertainingly better. ;]
llvm-svn: 215702
target-specific shuffl DAG combines.
We were recognizing the paired shuffles backwards. This code needs to be
replaced anyways as we have the same functionality elsewhere, but I'll
do the refactoring in a follow-up, this is the minimal fix to the
behavior.
In addition to fixing miscompiles with the new vector shuffle lowering,
it also causes the canonicalization to kick in much better, selecting
the smaller encoding variants in lots of places in the new AVX path.
This still isn't quite ideal as we don't need both the shufpd and the
punpck instructions, but that'll get fixed in a follow-up patch.
llvm-svn: 215690
broken logic for merging shuffle masks in the face of SM_SentinelZero
mask operands.
While these are '-1' they don't mean 'undef' the way '-1' means in the
pre-legalized shuffle masks. Instead, they mean that the shuffle
operation is forcibly zeroing that lane. Reflect this and explicitly
handle it in a bunch of places. In one place the effect is equivalent
but much more clear. In the rest it was really weirdly broken.
Also, rewrite the entire merging thing to be a more directy operation
with a single loop and just doing math to map the indices through the
various masks.
Also add a bunch of asserts to try to make in extremely clear what the
different masks can possibly look like.
Finally, add some comments to clarify that we're merging shuffle masks
*up* here rather than *down* as we do everywhere else, and thus the
logic is quite confusing.
Thanks to several different people for sending test cases, and for
Robert Khasanov for an initial attempt at fixing.
llvm-svn: 215687
lowering scheme.
Currently, this just directly bails to the fallback path of splitting
the 256-bit vector into two 128-bit vectors, operating there, and then
joining the results back together. While the results are far from
perfect, they are *shockingly* good for what we're doing here. I'll be
layering the rest of the functionality on top of this piece by piece and
updating tests as I go.
Note that 256-bit vectors in this mode are still somewhat WIP. While
I think the code paths that I'm adding here are clean and good-to-go,
there are still a lot of 128-bit assumptions that I'll need to stomp out
as I march through the functional spread here.
llvm-svn: 215637
one pesky test case correctly.
This test case caused the old code to infloop occilating between solving
the low-half and the high-half. The 'side balancing' part of
single-input v8 shuffle lowering didn't handle the one pattern which can
cause it to occilate. Fortunately the fuzz testing found this case.
Unfortuately it was *terrible* to handle. I'm really sorry for the
amount and density of the code here, I'd love suggestions on how to
simplify it. I feel like there *must* be a simpler form here, but after
a lot of days I've not found it. This is the only one I've found that
even works. I've added the one pesky test case along with some nice
comments explaining the core problem that we have to solve here.
So far this has survived approximately 32k test cases. More strenuous
fuzzing commencing.
llvm-svn: 215519
I think that this will scale better in most cases than adding a Pat<> for each
mapping from the intrinsic DAG to the intruction (i.e. rri, rrik, rrikz). We
can just lower to the SDNode and have the resulting DAG be matches by the DAG
patterns.
Alternatively (long term), we could keep the Pat<>s but generate them via the
new AVX512_masking multiclass. The difficulty is that in order to formulate
that we would have to concatenate DAGs. Currently this is only supported if
the operators of the input DAGs are identical.
llvm-svn: 215473
shuffle lowering.
This is closely related to the previous one. Here we failed to use the
source offset when swapping in the other case -- where we end up
swapping the *final* shuffle. The cause of this bug is a bit different:
I simply wasn't thinking about the fact that this mask is actually
a slice of a wide mask and thus has numbers that need SourceOffset
applied. Simple fix. Would be even more simple with an algorithm-y thing
to use here, but correctness first. =]
llvm-svn: 215095
via the fuzz tester.
Here I missed an offset when round-tripping a value through a shuffle
mask. I got it right 2 lines below. See a problem? I do. ;] I'll
probably be adding a little "swap" algorithm which accepts a range and
two values and swaps those values where they occur in the range. Don't
really have a name for it, let me know if you do.
llvm-svn: 215094
through the new fuzzer.
This one is great: bad operator precedence led the modulus to happen at
the wrong point. All the asserts didn't fire because there were usually
the right values past the end of the 4 element region we were looking
at. Probably could have gotten a crash here with ASan + fuzzing, but the
correctness tests pinpointed this really nicely.
llvm-svn: 215092
Summary:
Since pointers are 32-bit on x32 we can use ebp and esp as frame and stack
pointer. Some operations like PUSH/POP and CFI_INSTRUCTION still
require 64-bit register, so using 64-bit MachineFramePtr where required.
X86_64 NaCl uses 64-bit frame/stack pointers, however it's been found that
both isTarget64BitLP64 and isTarget64BitILP32 are true for NaCl. Addressing
this issue here as well by making isTarget64BitLP64 false.
Also mark hasReservedSpillSlot unreachable on X86. See inlined comments.
Test Plan: Add one new simple test and upgrade 2 existing with x32 target case.
Reviewers: nadav, dschuff
Subscribers: llvm-commits, zinovy.nis
Differential Revision: http://reviews.llvm.org/D4617
llvm-svn: 215091
fuzz testing.
The function which tested for adjacency did what it said on the tin, but
when I called it, I wanted it to do something more thorough: I wanted to
know if the *pairs* of shuffle elements were adjacent and started at
0 mod 2. In one place I had the decency to try to test for this, but in
the other it was completely skipped, miscompiling this test case. Fix
this by making the helper actually do what I wanted it to do everywhere
I called it (and removing the now redundant code in one place).
I *really* dislike the name "canWidenShuffleElements" for this
predicate. If anyone can come up with a better name, please let me know.
The other name I thought about was "canWidenShuffleMask" but is it
really widening the mask to reduce the number of lanes shuffled? I don't
know. Naming things is hard.
llvm-svn: 215089
to get the subtarget and that's accessible from the MachineFunction
now. This helps clear the way for smaller changes where we getting
a subtarget will require passing in a MachineFunction/Function as
well.
llvm-svn: 214988
test case to actually generate correct code.
The primary miscompile fixed here is that we weren't correctly handling
in-place elements in one half of a single-input v8i16 shuffle when
moving a dword of elements from that half to the other half. Some times,
we would clobber the in-place elements in forming the dword to move
across halves.
The fix to this involves forcibly marking the in-place inputs even when
there is no need to gather them into a dword, and to much more carefully
re-arrange the elements when grouping them into a dword to move across
halves. With these two changes we would generate correct shuffles for
the test case, but found another miscompile. There are also some random
perturbations of the generated shuffle pattern in SSE2. It looks like
a wash; more instructions in some cases fewer in others.
The second miscompile would corrupt the results into nonsense. This is
a buggy pattern in one of the added DAG combines. Mapping elements
through a PSHUFD when pairing redundant half-shuffles is *much* harder
than this code makes it out to be -- it requires reasoning about *all*
of where the input is used in the PSHUFD, not just one part of where it
is used. Plus, we can't combine a half shuffle *into* a PSHUFD but the
code didn't guard against it. I think this was just a bad idea and I've
just removed that aspect of the combine. No tests regress as
a consequence so seems OK.
llvm-svn: 214954
not corrupting the mask by mutating it more times than intended. No
functionality changed (the results were non-overlapping so the old
version "worked" but was non-obvious).
llvm-svn: 214953
a test case.
We also miscompile this test case which is showing a serious flaw in the
single-input v8i16 shuffle code. I've left the specific instruction
checks FIXME-ed out until I can address the bug in the single-input
code, but I wanted to separate out a significant functionality change to
produce correct code from a very simple and targeted crasher fix.
The miscompile problem stems from keeping track of inputs by value
rather than by index. As a consequence of doing this, we can't reliably
update those inputs because they might swap and we can't detect this
without copying the mask.
The blend code now uses indices for the input lists and this seems
strictly better. It also should make it easier to sort things and do
other cleanups. I think the time has come to simplify The Great Lambda
here.
llvm-svn: 214914
This was currently part of lowering to PALIGNR with some special-casing to
make interlane shifting work. Since AVX512F has interlane alignr (valignd/q)
and AVX512BW has vpalignr we need to support both of these *at the same time*,
e.g. for SKX.
This patch breaks out the common code and then add support to check both of
these lowering options from LowerVECTOR_SHUFFLE.
I also added some FIXMEs where I think the AVX512BW and AVX512VL additions
should probably go.
llvm-svn: 214888
They have different semantics (valign is interlane while palingr is intralane)
and palingr is still needed even in the AVX512 context. According to the
latest spec AVX512BW provides these.
llvm-svn: 214887
found by a single test reduced out of a failure on llvm-stress.
The start of the problem (and the crash) came when we tried to use
a find of a non-used slot in the move-to half of the move-mask as the
target for two bad-half inputs. While if lucky this will be the first of
a pair of slots which we can place the bad-half inputs into, it isn't
actually guaranteed. This really isn't surprising, not sure what I was
thinking. The correct way to find the two unused slots is to look for
one of the *used* slots. We know it isn't that pair, and we can use some
modular arithmetic to find the other pair by masking off the odd bit and
adding 2 modulo 4. With this, we reliably found a viable pair of slots
for the bad-half inputs.
Sadly, that wasn't enough. We also had a wrong code bug that surfaced
when I reduced the test case for this where we would use the same slot
twice for the two bad inputs. This is because both of the bad inputs
could be in odd slots originally and thus the mod-2 mapping would
actually be the same. The whole point of the weird indexing into the
pair of empty slots was to try to leverage when the end result needed
the two bad-half inputs to be paired in a dword and pre-pair them in the
correct orrientation. This is less important with the powerful combining
we're now doing, and also easier and more reliable to achieve be noting
that we add the bad-half inputs in order. Thus, if they are in a dword
pair, the low part of that will be the first input in the sequence.
Always putting that in the low element will just do the right thing in
addition to computing the correct result.
Test case added. =]
llvm-svn: 214849
shorter/easier and have the DAG use that to do the same lookup. This
can be used in the future for TargetMachine based caching lookups from
the MachineFunction easily.
Update the MIPS subtarget switching machinery to update this pointer
at the same time it runs.
llvm-svn: 214838
use of PACKUS. It's cleaner that way.
I looked at implementing clever combine-based folding of PACKUS chains
into PSHUFB but it is quite hard and doesn't seem likely to be worth it.
The most annoying part would be detecting that the correct masking had
been done to use PACKUS-style instructions as a blend operation rather
than there being any saturating as is indicated by its name. We generate
really nice code for what few test cases I've come up with that aren't
completely contrived for this by just directly prefering PSHUFB and so
let's go with that strategy for now. =]
llvm-svn: 214707
patterns of v16i8 shuffles.
This implements one of the more important FIXMEs for the SSE2 support in
the new shuffle lowering. We now generate the optimal shuffle sequence
for truncate-derived shuffles which show up essentially everywhere.
Unfortunately, this exposes a weakness in other parts of the shuffle
logic -- we can no longer form PSHUFB here. I'll add the necessary
support for that and other things in a subsequent commit.
llvm-svn: 214702
I spent some time looking into a better or more principled way to handle
this. For example, by detecting arbitrary "unneeded" ORs... But really,
there wasn't any point. We just shouldn't build blatantly wrong code so
late in the pipeline rather than adding more stages and logic later on
to fix it. Avoiding this is just too simple.
llvm-svn: 214680
lowering with a small addition to it and adding PSHUFB combining.
There is one obvious place in the new vector shuffle lowering where we
should form PSHUFBs directly: when without them we will unpack a vector
of i8s across two different registers and do a potentially 4-way blend
as i16s only to re-pack them into i8s afterward. This is the crazy
expensive fallback path for i8 shuffles and we can just directly use
pshufb here as it will always be cheaper (the unpack and pack are
two instructions so even a single shuffle between them hits our
three instruction limit for forming PSHUFB).
However, this doesn't generate very good code in many cases, and it
leaves a bunch of common patterns not using PSHUFB. So this patch also
adds support for extracting a shuffle mask from PSHUFB in the X86
lowering code, and uses it to handle PSHUFBs in the recursive shuffle
combining. This allows us to combine through them, combine multiple ones
together, and generally produce sufficiently high quality code.
Extracting the PSHUFB mask is annoyingly complex because it could be
either pre-legalization or post-legalization. At least this doesn't have
to deal with re-materialized constants. =] I've added decode routines to
handle the different patterns that show up at this level and we dispatch
through them as appropriate.
The two primary test cases are updated. For the v16 test case there is
still a lot of room for improvement. Since I was going through it
systematically I left behind a bunch of FIXME lines that I'm hoping to
turn into ALL lines by the end of this.
llvm-svn: 214628
of normally binary shuffle instructions like PUNPCKL and MOVLHPS.
This detects cases where a single register is used for both operands
making the shuffle behave in a unary way. We detect this and adjust the
mask to use the unary form which allows the existing DAG combine for
shuffle instructions to actually work at all.
As a consequence, this uncovered a number of obvious bugs in the
existing DAG combine which are fixed. It also now canonicalizes several
shuffles even with the existing lowering. These typically are trying to
match the shuffle to the domain of the input where before we only really
modeled them with the floating point variants. All of the cases which
change to an integer shuffle here have something in the integer domain, so
there are no more or fewer domain crosses here AFAICT. Technically, it
might be better to go from a GPR directly to the floating point domain,
but detecting floating point *outputs* despite integer inputs is a lot
more code and seems unlikely to be worthwhile in practice. If folks are
seeing domain-crossing regressions here though, let me know and I can
hack something up to fix it.
Also as a consequence, a bunch of missed opportunities to form pshufb
now can be formed. Notably, splats of i8s now form pshufb.
Interestingly, this improves the existing splat lowering too. We go from
3 instructions to 1. Yes, we may tie up a register, but it seems very
likely to be worth it, especially if splatting the 0th byte (the
common case) as then we can use a zeroed register as the mask.
llvm-svn: 214625
Stop using ST registers for function returns and inline-asm instructions and use
FP registers instead. This allows removing a large amount of code in the
stackifier pass that was needed to track register liveness and handle copies
between ST and FP registers and function calls returning floating point values.
It also fixes a bug which manifests when an ST register defined by an
inline-asm instruction was live across another inline-asm instruction, as shown
in the following sequence of machine instructions:
1. INLINEASM <es:frndint> $0:[regdef], %ST0<imp-def,tied5>
2. INLINEASM <es:fldcw $0>
3. %FP0<def> = COPY %ST0
<rdar://problem/16952634>
llvm-svn: 214580
Currently when DAGCombine converts loads feeding a switch into a switch of
addresses feeding a load the new load inherits the isInvariant flag of the left
side. This is incorrect since invariant loads can be reordered in cases where it
is illegal to reoarder normal loads.
This patch adds an isInvariant parameter to getExtLoad() and updates all call
sites to pass in the data if they have it or false if they don't. It also
changes the DAGCombine to use that data to make the right decision when
creating the new load.
llvm-svn: 214449
Rename to allowsMisalignedMemoryAccess.
On R600, 8 and 16 byte accesses are mostly OK with 4-byte alignment,
and don't need to be split into multiple accesses. Vector loads with
an alignment of the element type are not uncommon in OpenCL code.
llvm-svn: 214055
instructions in the legalized DAG, and leverage it to combine long
sequences of instructions to PSHUFB.
Eventually, the other x86-instruction-specific shuffle combines will
probably all be driven out of this routine. But the real motivation is
to detect after we have fully legalized and optimized a shuffle to the
minimal number of x86 instructions whether it is profitable to replace
the chain with a fully generic PSHUFB instruction even though doing so
requires either a load from a constant pool or tying up a register with
the mask.
While the Intel manuals claim it should be used when it replaces 5 or
more instructions (!!!!) my experience is that it is actually very fast
on modern chips, and so I've gon with a much more aggressive model of
replacing any sequence of 3 or more instructions.
I've also taught it to do some basic canonicalization to special-purpose
instructions which have smaller encodings than their generic
counterparts.
There are still quite a few FIXMEs here, and I've not yet implemented
support for lowering blends with PSHUFB (where its power really shines
due to being able to zero out lanes), but this starts implementing real
PSHUFB support even when using the new, fancy shuffle lowering. =]
llvm-svn: 214042
The tale starts with r212808 which attempted to fix inversion of the low
and high bits when lowering MUL_LOHI. Sadly, that commit did not include
any positive test cases, and just removed some operations from a test
case where the actual logic being changed isn't fully visible from the
test.
What this commit did was two things. First, it reversed the low and high
results in the formation of the MERGE_VALUES node for the multiple
results. This is entirely correct.
Second it changed the shuffles for extracting the low and high
components from the i64 results of the multiplies to extract them
assuming a big-endian-style encoding of the multiply results. This
second change is wrong. There is no big-endian encoding in x86, the
results of the multiplies are normal v2i64s: when cast to v4i32, the low
i32s are at offsets 0 and 2, and the high i32s are at offsets 1 and 3.
However, the first change wasn't enough to actually fix the bug, which
is (I assume) why the second change was also made. There was another bug
in the MERGE_VALUES formation: we weren't using a VTList, and so were
getting a single result node! When grabbing the *second* result from the
node, we got... well.. colud be anything. I think this *appeared* to
invert things, but had to be causing other problems as well.
Fortunately, I fixed the MERGE_VALUES issue in r213931, so we should
have been fine, right? NOOOPE! Because the core bug was never addressed,
the test in vector-idiv failed when I fixed the MERGE_VALUES node.
Because there are essentially no docs for this node, I had to guess at
how to fix it and tried swapping the operands, restoring the order of
the original code before r212808. While this "fixed" the test case (in
that we produced the write instructions) we were still extracting the
wrong elements of the i64s, and thus PR20355 was still broken.
This commit essentially reverts the big-endian-style extraction part of
r212808 and goes back to the original masks which were correct. Now that
the MERGE_VALUES node formation is also correct, everything works. I've
also included a more detailed test from PR20355 to make sure this stays
fixed.
llvm-svn: 214011
The clever way to implement signed multiplication with unsigned *is
already implemented* and tested and working correctly. The bug is
somewhere else. Re-investigating.
This will teach me to not scroll far enough to read the code that did
what I thought needed to be done.
llvm-svn: 214009
signed multiplication is requested. While there is not a difference in
the *low* half of the result, the *high* half (used specifically to
implement the signed division by these constants) certainly is used. The
test case I've nuked was actively asserting wrong code.
There is a delightful solution to doing signed multiplication even when
we don't have it that Richard Smith has crafted, but I'll add the
machinery back and implement that in a follow-up patch. This at least
restores correctness.
llvm-svn: 214007
address of the stack guard was being spilled to the stack.
Previously the address of the stack guard would get spilled to the stack if it
was impossible to keep it in a register. This patch introduces a new target
independent node and pseudo instruction which gets expanded post-RA to a
sequence of instructions that load the stack guard value. Register allocator
can now just remat the value when it can't keep it in a register.
<rdar://problem/12475629>
llvm-svn: 213967
SDValues, fixing the two bugs left in the regression suite.
The key for both of these was the use a single value type rather than
a VTList which caused an unintentionally single-result merge-value node.
Fix this by getting the appropriate VTList in place.
Doing this exposed that the comments in x86's code abouth how MUL_LOHI
operands are handle is wrong. The bug with the use of out-of-range
result numbers was hiding the bug about the order of operands here (as
best i can tell). There are more places where the code appears to get
this backwards still...
llvm-svn: 213931
vector operation legalization with support for custom target lowering
and fallback to expand when it fails, and use this to implement sext and
anyext load lowering for x86 in a more principled way.
Previously, the x86 backend relied on a target DAG combine to "combine
away" sextload and extload nodes prior to legalization, or would expand
them during legalization with terrible code. This is particularly
problematic because the DAG combine relies on running over non-canonical
DAG nodes at just the right time to match several common and important
patterns. It used a combine rather than lowering because we didn't have
good lowering support, and to expose some tricks being employed to more
combine phases.
With this change it becomes a proper lowering operation, the backend
marks that it can lower these nodes, and I've added support for handling
the canonical forms that don't have direct legal representations such as
sextload of a v4i8 -> v4i64 on AVX1. With this change, our test cases
for this behavior continue to pass even after the DAG combiner beigns
running more systematically over every node.
There is some noise caused by this in the test suite where we actually
use vector extends instead of subregister extraction. This doesn't
really seem like the right thing to do, but is unlikely to be a critical
regression. We do regress in one case where by lowering to the
target-specific patterns early we were able to combine away extraneous
legal math nodes. However, this regression is completely addressed by
switching to a widening based legalization which is what I'm working
toward anyways, so I've just switched the test to that mode.
Differential Revision: http://reviews.llvm.org/D4654
llvm-svn: 213897
GCC 4.8 detected a signed compare [-Wsign-compare]. Add a cast for the
destination index. Add an assert to catch a potential overflow however unlikely
it may be.
llvm-svn: 213878
When we had a vector_shuffle where we had an input from each vector, we
could miscompile it because we were assuming the input from V2 wouldn't
be moved from where it was on the vector.
Added a test case.
llvm-svn: 213826
The transform to constant fold unary operations with an AND across a
vector comparison applies when the constant is not a splat of a scalar
as well.
llvm-svn: 213800
The folding of unary operations through a vector compare and mask operation
is only safe if the unary operation result is of the same size as its input.
For example, it's not safe for [su]itofp from v4i32 to v4f64.
llvm-svn: 213799
This chang fully reverts r211771.
That revision added a canonicalization rule which has the potential to causes a
combine-cycle in the target-independent canonicalizing DAG combine.
The plan is to move the logic that forms target specific addsub nodes as part of
the lowering of shuffles.
llvm-svn: 213736
This patch removes function 'CommuteVectorShuffle' from X86ISelLowering.cpp
and moves its logic into SelectionDAG.cpp as method 'getCommutedVectorShuffles'.
This refactoring is in preperation of an upcoming change to the DAGCombiner.
llvm-svn: 213503
Since the result of a SETCC for X86 is 0 or -1 in each lane, we can
move unary operations, in this case [su]int_to_fp through the mask
operation and constant fold the operation away. Generally speaking:
UNARYOP(AND(VECTOR_CMP(x,y), constant))
--> AND(VECTOR_CMP(x,y), constant2)
where constant2 is UNARYOP(constant).
This implements the transform where UNARYOP is [su]int_to_fp.
For example, consider the simple function:
define <4 x float> @foo(<4 x float> %val, <4 x float> %test) nounwind {
%cmp = fcmp oeq <4 x float> %val, %test
%ext = zext <4 x i1> %cmp to <4 x i32>
%result = sitofp <4 x i32> %ext to <4 x float>
ret <4 x float> %result
}
Before this change, the SSE code is generated as:
LCPI0_0:
.long 1 ## 0x1
.long 1 ## 0x1
.long 1 ## 0x1
.long 1 ## 0x1
.section __TEXT,__text,regular,pure_instructions
.globl _foo
.align 4, 0x90
_foo: ## @foo
cmpeqps %xmm1, %xmm0
andps LCPI0_0(%rip), %xmm0
cvtdq2ps %xmm0, %xmm0
retq
After, the code is improved to:
LCPI0_0:
.long 1065353216 ## float 1.000000e+00
.long 1065353216 ## float 1.000000e+00
.long 1065353216 ## float 1.000000e+00
.long 1065353216 ## float 1.000000e+00
.section __TEXT,__text,regular,pure_instructions
.globl _foo
.align 4, 0x90
_foo: ## @foo
cmpeqps %xmm1, %xmm0
andps LCPI0_0(%rip), %xmm0
retq
The cvtdq2ps has been constant folded away and the floating point 1.0f
vector lanes are materialized directly via the ModRM operand of andps.
llvm-svn: 213342
Previously we asserted on this code. Currently compiler-rt doesn't
actually implement any of these new libcalls, but external help is
pretty much the only viable option for LLVM.
I've followed the much more generic "__truncST2" naming, as opposed to
the odd name for f32 -> f16 truncation. This can obviously be changed
later, or overridden by any targets that need to.
llvm-svn: 213252
x86 has no native ability to extend an f16 to f64, but the same result
is obtained if we expand it into two separate extensions: f16 -> f32
-> f64.
Unfortunately the same is not true for truncate, so that still results
in a compilation failure.
llvm-svn: 213251
This makes the two intrinsics @llvm.convert.from.f16 and
@llvm.convert.to.f16 accept types other than simple "float". This is
only strictly needed for the truncate operation, since otherwise
double rounding occurs and there's no way to represent the strict IEEE
conversion. However, for symmetry we allow larger types in the extend
too.
During legalization, we can expand an "fp16_to_double" operation into
two extends for convenience, but abort when the truncate isn't legal. A new
libcall is probably needed here.
Even after this commit, various target tweaks are needed to actually use the
extended intrinsics. I've put these into separate commits for clarity, so there
are no actual tests of f64 conversion here.
llvm-svn: 213248
It turns out that in most cases (the main exception being i1-related
types) once these operations are formed we cannot separate them and
the targets end up having to deal with them whether they want to or
not.
This is not a good situation, and a more reasonable default can be
formed by ackowledging this and having targets leave them as Legal.
Only x86 seems to be affected (other targets don't even try marking
the operation Expand).
Mostly there's no visible change here yet, but it will be useful to
have truly expanded EXTLOADS for MVT::f16 softening support.
llvm-svn: 213162
Before this change, method 'isShuffleMaskLegal' didn't know that shuffles
implementing a 'movhlps' operation were perfectly legal for SSE targets.
This patch adds the missing check for 'isMOVHLPSMask' inside method
'isShuffleMaskLegal' to fix the problem.
The reason why it is important to do this is because the DAGCombiner
conservatively avoids combining a pair of shuffles if the resulting shuffle
node has an illegal mask. Before this patch, shuffles with a MOVHLPS mask were
wrongly considered not to be legal. This was the root cause of some poor-code
generation bugs.
llvm-svn: 213137
shuffle lowering: match shuffle patterns equivalent to an unpcklwd or
unpckhwd instruction.
This allows us to use generic lowering code for v8i16 shuffles and match
the unpack pattern late.
llvm-svn: 212705
combine into half-shuffles through unpack instructions that expand the
half to a whole vector without messing with the dword lanes.
This fixes some redundant instructions in splat-like lowerings for
v16i8, which are now getting to be *really* nice.
llvm-svn: 212695
that splat i8s into i16s.
Previously, we would try much too hard to arrange a sequence of i8s in
one half of the input such that we could unpack them into i16s and
shuffle those into place. This isn't always going to be a cheaper i8
shuffle than our other strategies. The case where it is always going to
be cheaper is when we can arrange all the necessary inputs into one half
using just i16 shuffles. It happens that viewing the problem this way
also makes it much easier to produce an efficient set of shuffles to
move the inputs into one half and then unpack them.
With this, our splat code gets one step closer to being not terrible
with the new experimental lowering strategy. It also exposes two
combines missing which I will add next.
llvm-svn: 212692
shuffles specifically for cases where a small subset of the elements in
the input vector are actually used.
This is specifically targetted at improving the shuffles generated for
trunc operations, but also helps out splat-like operations.
There is still some really low-hanging fruit here that I want to address
but this is a huge step in the right direction.
llvm-svn: 212680
don't need to set it manually.
This is based on feedback from Tom who pointed out that if every target
needs to handle this we need to reach out to those maintainers. In fact,
it doesn't make sense to duplicate everything when anything other than
expand seems unlikely at this stage.
llvm-svn: 212661
Turns out my trick of using the same masks for SSE4.1 and AVX2 didn't work out
as we have to blend two vectors. While there remove unecessary cross-lane moves
from the shuffles so the backend can lower it to palignr instead of vperm.
Fixes PR20118, a miscompilation of vector sdiv by constant on AVX2.
llvm-svn: 212611
vector types to be legal and a ZERO_EXTEND node is encountered.
When we use widening to legalize vector types, extend nodes are a real
challenge. Either the input or output is likely to be legal, but in many
cases not both. As a consequence, we don't really have any way to
represent this situation and the prior code in the widening legalization
framework would just scalarize the extend operation completely.
This patch introduces a new DAG node to represent doing a zero extend of
a vector "in register". The core of the idea is to allow legal but
different vector types in the input and output. The output vector must
have fewer lanes but wider elements. The operation is defined to zero
extend the low elements of the input to the size of the output elements,
and drop all of the high elements which don't have a corresponding lane
in the output vector.
It also includes generic expansion of this node in terms of blending
a zero vector into the high elements of the vector and bitcasting
across. This in turn yields extremely nice code for x86 SSE2 when we use
the new widening legalization logic in conjunction with the new shuffle
lowering logic.
There is still more to do here. We need to support sign extension, any
extension, and potentially int-to-float conversions. My current plan is
to continue using similar synthetic nodes to model each of these
transitions with generic lowering code for each one.
However, with this patch LLVM already reaches performance parity with
GCC for the core C loops of the x264 code (assuming you disable the
hand-written assembly versions) when compiling for SSE2 and SSE3
architectures and enabling the new widening and lowering logic for
vectors.
Differential Revision: http://reviews.llvm.org/D4405
llvm-svn: 212610
has settled without incident, removing the x86-specific and overly
strict 'isVectorSplat' routine in favor of generic and more powerful
splat detection.
The primary motivation and result of this is that the x86 backend can
now see through splats which contain undef elements. This is essential
if we are using a widening form of legalization and I've updated a test
case to also run in that mode as before this change the generated code
for the test case was completely scalarized.
This version of the patch much more carefully handles the undef lanes.
- We aren't overly conservative about them in the shift lowering
(where we will never use the splat itself).
- One place where the splat would have been re-used by the existing code
now explicitly constructs a new constant splat that will be safe.
- The broadcast lowering is much more reasonable with undefs by doing
a correct check of whether the splat is the only user of a loaded
value, checking that the splat actually crosses multiple lanes before
using a broadcast, and handling broadcasts of non-constant splats.
As a consequence of the last bullet, the weird usage of vpshufd instead
of vbroadcast is gone, and we actually can lower an AVX splat with
vbroadcastss where before we emitted a really strange pattern of
a vector load and a manual splat across the vector.
llvm-svn: 212602
aggressively from the x86 shuffle lowering to the generic SDAG vector
shuffle formation code.
This code already tried to fold away shuffles of splats! It just had
lots of bugs and couldn't handle the case my new x86 shuffle lowering
needed.
First, it failed to correctly compute whether N2 was undef because it
pre-computed this, then did transformations which could *make* N2 undef,
then failed to ever re-consider the precomputed state.
Second, it didn't look through bitcasts at all, even in the safe cases
where they are just element-type bitcasts with no change to the number
of elements.
Third, it didn't handle all-zero bit casts nicely the way my code in the
x86 side of things did, which is essential to getting good zext-shuffle
lowerings.
But all of these are generic. I just ported the code down to this layer
and fixed the surrounding bugs. Tests exercising this in the x86 backend
still pass and some silly code in widen_cast-6.ll gets better. I updated
that test to be a bit more precise but it's still pretty unclear what
the value of the test is in this day and age.
llvm-svn: 212517
When combining a sequence of two PSHUFD dag nodes into a single PSHUFD,
make sure that we assign the correct type to the resulting PSHUFD.
X86ISD::PSHUFD dag nodes can be either MVT::v4i32 or MVT::v4f32.
Before this change, an assertion failure was triggered in method
'DAGCombinerInfo::CombineTo' when trying to combine the shuffles from the test
below into a single PSHUFD.
define <4 x float> @test1(<4 x float> %V) {
%1 = shufflevector <4 x float> %V, <4 x float> undef, <4 x i32> <i32 3, i32 0, i32 2, i32 1>
%2 = shufflevector <4 x float> %1, <4 x float> undef, <4 x i32> <i32 3, i32 0, i32 2, i32 1>
ret <4 x float> %2
}
llvm-svn: 212498
lanes in vector splats.
The core problem here is that undef lanes can't *unilaterally* be
considered to contribute to splats. Their handling needs to be more
cautious. There is also a reported failure of the nightly testers
(thanks Tobias!) that may well stem from the same core issue. I'm going
to fix this theoretical issue, factor the APIs a bit better, and then
verify that I don't see anything bad with Tobias's reduction from the
test suite before recommitting.
Original commit message for r212324:
[x86] Generalize BuildVectorSDNode::getConstantSplatValue to work for
any constant, constant FP, or undef splat and to tolerate any undef
lanes in a splat, then replace all uses of isSplatVector in X86's
lowering with it.
This fixes issues where undef lanes in an otherwise splat vector would
prevent the splat logic from firing. It is a touch more awkward to use
this interface, but it is much more accurate. Suggestions for better
interface structuring welcome.
With this fix, the code generated with the widening legalization
strategy for widen_cast-4.ll is *dramatically* improved as the special
lowering strategies for a v16i8 SRA kick in even though the high lanes
are undef.
We also get a slightly different choice for broadcasting an aligned
memory location, and use vpshufd instead of vbroadcastss. This looks
like a minor win for pipelining and domain crossing, but a minor loss
for the number of micro-ops. I suspect its a wash, but folks can
easily tweak the lowering if they want.
llvm-svn: 212475
essentially a DAG combine that never gets a chance to run.
We might typically expect DAG combining to remove shuffles-of-splats and
other similar patterns, but we don't get a chance to run the DAG
combiner when we recursively form sub-shuffles during the lowering of
a shuffle. So instead hand-roll a really important combine directly into
the lowering code to detect shuffles-of-splats, especially shuffles of
an all-zero splat which needn't even have the same element width, etc.
This lets the new vector shuffle lowering handle shuffles which
implement things like zero-extension really nicely. This will become
even more important when I wire the legalization of zero-extension to
vector shuffles with the new widening legalization strategy.
llvm-svn: 212444
any constant, constant FP, or undef splat and to tolerate any undef
lanes in a splat, then replace all uses of isSplatVector in X86's
lowering with it.
This fixes issues where undef lanes in an otherwise splat vector would
prevent the splat logic from firing. It is a touch more awkward to use
this interface, but it is much more accurate. Suggestions for better
interface structuring welcome.
With this fix, the code generated with the widening legalization
strategy for widen_cast-4.ll is *dramatically* improved as the special
lowering strategies for a v16i8 SRA kick in even though the high lanes
are undef.
We also get a slightly different choice for broadcasting an aligned
memory location, and use vpshufd instead of vbroadcastss. This looks
like a minor win for pipelining and domain crossing, but a minor loss
for the number of micro-ops. I suspect its a wash, but folks can easily
tweak the lowering if they want.
llvm-svn: 212324
This patch adds tablegen patterns to select F16C float-to-half-float
conversion instructions from 'f32_to_f16' and 'f16_to_f32' dag nodes.
If the target doesn't have F16C, then 'f32_to_f16' and 'f16_to_f32'
are expanded into library calls.
llvm-svn: 212293
mode.
This also runs the test in that mode which would reproduce the crash.
What I love is that *every single FIXME* in the test is addressed by
switching to widening.
llvm-svn: 212254
Finkel, Eric Christopher, and a bunch of other people I'm probably
forgetting (sorry), add an option to the x86 backend to widen vectors
during type legalization rather than promote them.
This still would promote vNi1 vectors to get the masks right, but would
widen other vectors. A lot of experiments are piling up right now
showing that widening should probably be the default legalization
strategy outside of vNi1 cases, but it is very hard to test the
rammifications of that and fix bugs in widening-based legalization
without an option that enables it. I'll be checking in tests shortly
that use this option to exercise cases where widening doesn't work well
and hopefully we'll be able to switch fully to this soon.
llvm-svn: 212249
The argument list vector is never used after it has been passed to the
CallLoweringInfo and moving it to the CallLoweringInfo is cleaner and
pretty much as cheap as keeping a pointer to it.
llvm-svn: 212135
On targets without cmpxchg16b or cmpxchg8b, the borderline atomic
operations were slipping through the gaps.
X86AtomicExpand.cpp was delegating to ISelLowering. Generic
ISelLowering was delegating to X86ISelLowering and X86ISelLowering was
asserting. The correct behaviour is to expand to a libcall, preferably
in generic ISelLowering.
This can be achieved by X86ISelLowering deciding it doesn't want the
faff after all.
llvm-svn: 212134
The logic for expanding atomics that aren't natively supported in
terms of cmpxchg loops is much simpler to express at the IR level. It
also allows the normal optimisations and CodeGen improvements to help
out with atomics, instead of using a limited set of possible
instructions..
rdar://problem/13496295
llvm-svn: 212119
This patch adds support for a new builtin instruction called
__builtin_ia32_rdpmc.
Builtin '__builtin_ia32_rdpmc' is defined as a 'GCC builtin'; on X86, it can
be used to read performance monitoring counters. It takes as input the index
of the performance counter to read, and returns the value of the specified
performance counter as a 64-bit number.
Calls to this new builtin will map to instruction RDPMC.
The index in input to the builtin call is moved to register %ECX. The result
of the builtin call is the value of the specified performance counter (RDPMC
would return that quantity in registers RDX:RAX).
This patch:
- Adds builtin int_x86_rdpmc as a GCCBuiltin;
- Adds a new x86 DAG node called 'RDPMC_DAG';
- Teaches how to lower this new builtin;
- Adds an ISel pattern to select instruction RDPMC;
- Fixes the definition of instruction RDPMC adding %RAX and %RDX as
implicit definitions, and adding %ECX as implicit use;
- Adds a LLVM test to verify that the new builtin is correctly selected.
llvm-svn: 212049
lowering for v16i8.
ASan and some bots caught this bug with existing test cases. Fixing it
even fixed a miscompile with one of the test cases. I'm still a bit
suspicious of this test case as I've not taken a proper amount of time
to think about it, but the fix here is strict goodness.
llvm-svn: 211976
These show up really frequently, not the least with actual splats. =] We
lowered these quite badly before. The new code path tries to widen i8
shuffles to i16 shuffles in a splat-like way. There are still some
inefficiencies in our i16 splat logic though, so we aren't really done
here.
Also, for certain patterns (bit of a gather-and-splat) we still
generate pretty silly code, and I've left a fixme for addressing it.
However, I'm not actually worried about this code pattern as much. The
old shuffle lowering generates a 29 instruction monstrosity for it that
should execute much more slowly.
llvm-svn: 211974
lowering.
For maximum irony, I had already discovered this bug, diagnosed it, and
left FIXMEs about it in the test cases. =[ I just failed to go back over
those until after i had reduced a bootstrap miscompile down to a single
TU, stared at the assembly for an hour, and figured out the bug. Again.
Oh well.
llvm-svn: 211955
a bootstrap.
I managed to mis-remember how PACKUS worked on x86, and was using undef
for the high bytes instead of zero. The fix is fairly obvious.
llvm-svn: 211922
Summary:
This allows it to fold pshufd instructions across intervening
half-shuffles and other noise. This pattern actually shows up in the
generic lowering tests, but I've also added direct tests using
intrinsics to make sure that the specific desired functionality is
working even if the lowering stuff changes in the future.
Differential Revision: http://reviews.llvm.org/D4292
llvm-svn: 211892
half-shuffles, even looking through intervening instructions in a chain.
Summary:
This doesn't happen to show up with any test cases I've found for the current
shuffle lowering, but previous attempts would benefit from this and it seems
generally useful. I've tested it directly using intrinsics, which also shows
that it will work with hand vectorized code as well.
Note that even though pshufd isn't directly used in these tests, it gets
exercised because we combine some of the half shuffles into a pshufd
first, and then merge them.
Differential Revision: http://reviews.llvm.org/D4291
llvm-svn: 211890
trivially redundant.
This fixes several cases in the new vector shuffle lowering algorithm
which would generate redundant shuffle instructions for the sake of
simplicity.
I'm also deleting a testcase which was somewhat ridiculous. It was
checking for a bug in 2007 about incorrectly transforming shuffles by
looking for the string "-86" in the output of a pretty substantial
function. This test case doesn't seem to have any value at this point.
Differential Revision: http://reviews.llvm.org/D4240
llvm-svn: 211889
x86 backend.
This sketches out a new code path for vector lowering, hidden behind an
off-by-default flag while it is under development. The fundamental idea
behind the new code path is to aggressively break down the problem space
in ways that ease selecting the odd set of instructions available on
x86, and carefully avoid scalarizing code even when forced to use older
ISAs. Notably, this starts off restricting itself to SSE2 and implements
the complete vector shuffle and blend space for 128-bit vectors in SSE2
without scalarizing. The plan is to layer on top of this ISA extensions
where we can bail out of the complex SSE2 lowering and opt for
a cheaper, specialized instruction (or set of instructions). It also
needs to be generalized to AVX and AVX512 vector widths.
Currently, this does a decent but not perfect job for SSE2. There are
some specific shortcomings that I plan to address:
- We need a peephole combine to fold together shuffles where possible.
There are cases where a previous shuffle could be modified slightly to
arrange for elements to be in the correct position and a later shuffle
eliminated. Doing this eagerly added quite a bit of complexity, and
so my plan is to combine away these redundancies afterward.
- There are a lot more clever ways to use unpck and pack that need to be
added. This is essential for real world shuffles as it turns out...
Once SSE2 is polished a bit I should be able to get interesting numbers
on performance improvements on benchmarks conducive to vectorization.
All of this will be off by default until it is functionally equivalent
of course.
Differential Revision: http://reviews.llvm.org/D4225
llvm-svn: 211888
This patch teaches the backend how to canonicalize a shuffle vectors
according to the rule:
- (shuffle (FADD A, B), (FSUB A, B), Mask) ->
(shuffle (FSUB A, -B), (FADD A, -B), Mask)
Where 'Mask' is:
<0,5,2,7> ;; for v4f32 and v4f64 shuffles.
<0,3> ;; for v2f64 shuffles.
<0,9,2,11,4,13,6,15> ;; for v8f32 shuffles.
In general, ISel only knows how to pattern-match a canonical
'fadd + fsub + blendi' dag node sequence into an ADDSUB instruction.
This new rule allows to convert a non-canonical dag sequence into a
canonical one that will be matched by a single ADDSUB at ISel stage.
The idea of converting a non-canonical ADDSUB into a canonical one by
swapping the first two operands of the shuffle, and then negating the
second operand of the FADD and FSUB, was originally proposed by Hal Finkel.
llvm-svn: 211771
This patch teaches method 'LowerVECTOR_SHUFFLE' to give higher precedence to
the check for 'isBlendMask'; the idea is that, when possible, we should firstly
check if a shuffle performs a blend, and in case, try to lower it into a BLENDI
instead of selecting a SHUFP or (worse) a VPERM2X128.
In general:
- AVX VBLENDPS/D always have better latency and throughput than VPERM2F128;
- BLENDPS/D instructions tend to always have better 'reciprocal throughput'
than the equivalent SHUFPS/D;
- Both BLENDPS/D and SHUFPS/D are often decoded into the same number of
m-ops; however, a m-op obtained from a BLENDPS/D can be scheduled to more
than one execution port.
This patch:
- Moves the check for 'isBlendMask' immediately before the check for
'isSHUFPMask' within method 'LowerVECTOR_SHUFFLE';
- Updates existing tests for sse/avx shuffle/blend instructions to verify
that we select (v)blendps/d when possible (instead of (v)shufps/d or
vperm2f128).
llvm-svn: 211720
--
This patch enables LLVM to emit Win64-native unwind info rather than
DWARF CFI. It handles all corner cases (I hope), including stack
realignment.
Because the unwind info is not flexible enough to describe stack frames
with a gap of unknown size in the middle, such as the one caused by
stack realignment, I modified register spilling code to place all spills
into the fixed frame slots, so that they can be accessed relative to the
frame pointer.
Patch by Vadim Chugunov!
Reviewed By: rnk
Differential Revision: http://reviews.llvm.org/D4081
llvm-svn: 211691
This patch teaches the backend how to combine a build_vector that implements
an 'addsub' between packed float vectors into a sequence of vector add
and vector sub followed by a VSELECT.
The new VSELECT is expected to be lowered into a BLENDI.
At ISel stage, the sequence 'vector add + vector sub + BLENDI' is
pattern-matched against ISel patterns added at r211427 to select
'addsub' instructions.
Added three more ISel patterns for ADDSUB.
Added test sse3-avx-addsub-2.ll to verify that we correctly emit 'addsub'
instructions.
llvm-svn: 211679
This patch enables LLVM to emit Win64-native unwind info rather than
DWARF CFI. It handles all corner cases (I hope), including stack
realignment.
Because the unwind info is not flexible enough to describe stack frames
with a gap of unknown size in the middle, such as the one caused by
stack realignment, I modified register spilling code to place all spills
into the fixed frame slots, so that they can be accessed relative to the
frame pointer.
Patch by Vadim Chugunov!
Reviewed By: rnk
Differential Revision: http://reviews.llvm.org/D4081
llvm-svn: 211399
instructions available as synthetic SDNodes PACKSS and PACKUS that will
select to the correct instruction variants based on the return type.
This allows us to use these rather important instructions when lowering
vector shuffles.
Also moves the relevant instruction definitions to be split out from
the fully generic multiclasses to allow them to match these new SDNodes
in the same way that the UNPCK instructions do.
No functionality should actually be changed here.
llvm-svn: 211332
Before this change, the backend was unable to fold a build_vector dag
node with UNDEF operands into a single horizontal add/sub.
This patch teaches how to combine a build_vector with UNDEF operands into a
horizontal add/sub when possible. The algorithm conservatively avoids to combine
a build_vector with only a single non-UNDEF operand.
Added test haddsub-undef.ll to verify that we correctly fold horizontal binop
even in the presence of UNDEFs.
llvm-svn: 211265
Lowering this new node allows us to fold the almost universal
comparison for success before it's even formed. Instead we can create
a copy from EFLAGS and an X86ISD::SETCC operation since all "cmpxchg"
instructions set the zero-flag to the correct value.
rdar://problem/13201607
llvm-svn: 210923
This commit adds a weak variant of the cmpxchg operation, as described
in C++11. A cmpxchg instruction with this modifier is permitted to
fail to store, even if the comparison indicated it should.
As a result, cmpxchg instructions must return a flag indicating
success in addition to their original iN value loaded. Thus, for
uniformity *all* cmpxchg instructions now return "{ iN, i1 }". The
second flag is 1 when the store succeeded.
At the DAG level, a new ATOMIC_CMP_SWAP_WITH_SUCCESS node has been
added as the natural representation for the new cmpxchg instructions.
It is a strong cmpxchg.
By default this gets Expanded to the existing ATOMIC_CMP_SWAP during
Legalization, so existing backends should see no change in behaviour.
If they wish to deal with the enhanced node instead, they can call
setOperationAction on it. Beware: as a node with 2 results, it cannot
be selected from TableGen.
Currently, no use is made of the extra information provided in this
patch. Test updates are almost entirely adapting the input IR to the
new scheme.
Summary for out of tree users:
------------------------------
+ Legacy Bitcode files are upgraded during read.
+ Legacy assembly IR files will be invalid.
+ Front-ends must adapt to different type for "cmpxchg".
+ Backends should be unaffected by default.
llvm-svn: 210903
When I originally added node RDTSCP_DAG (r207127) I forgot to add
a string name for it in method 'getTargetNodeName'.
No functional change intended.
llvm-svn: 210769
This patch adds target combine rules to match:
- [AVX] Horizontal add/sub of packed single/double precision floating point
values from 256-bit vectors;
- [AVX2] Horizontal add/sub of packed integer values from 256-bit vectors.
llvm-svn: 210761
This patch moves part of the logic implemented by the target specific
combine rules added at r210477 to a separate helper function.
This should make easier to add more rules for matching AVX/AVX2 horizontal
adds/subs.
This patch also fixes a problem caused by a wrong check performed on indices
of extract_vector_elt dag nodes in input to the scalar adds/subs.
New tests have been added to verify that we correctly check indices of
extract_vector_elt dag nodes when selecting a horizontal operation.
llvm-svn: 210644
This patch slightly changes the algorithm introduced at revision 210477
to fix a problem where the algorithm was producing incorrect code for
the VEX.256 encoded versions of horizontal add/sub.
For these cases, we now try to split the two 256-bit vectors into
128-bit chunks before emitting horizontal add/sub dag nodes.
Added a new test case into haddsub-2.ll.
llvm-svn: 210545
The SelectionDAG bad a special case for ISD::SELECT_CC, where it would
allow targets to specify:
setOperationAction(ISD::SELECT_CC, MVT::Other, Expand);
to indicate that they wanted to expand ISD::SELECT_CC for all types.
This wasn't applied correctly everywhere, and it makes writing new
DAG patterns with ISD::SELECT_CC difficult.
llvm-svn: 210541
The C++ and C semantics of the compare_and_swap operations actually
require us to return a boolean "success" value. In LLVM terms this
means a second comparison of the output of "cmpxchg" against the input
desired value.
However, x86's "cmpxchg" instruction sets all flags for the comparison
formed, so we can skip any secondary comparison. (N.b. this isn't true
for cmpxchg8b/16b, which only set ZF).
rdar://problem/13201607
llvm-svn: 210523
This patch adds new target specific combine rules to identify horizontal
add/sub idioms from BUILD_VECTOR dag nodes.
This patch also teaches the DAGCombiner how to canonicalize sequences of
insert_vector_elt dag nodes according to the following rule:
(insert_vector_elt (insert_vector_elt A, I0), I1) ->
(insert_vecto_elt (insert_vector_elt A, I1), I0)
This new canonicalization rule only triggers if the inner insert_vector
dag node has exactly one use; also, both indices must be known constants,
and I1 < I0.
This last rule made it possible to write a simpler algorithm to identify
horizontal add/sub patterns because now we don't have to worry about the
ordering of insert_vector_elt dag nodes.
llvm-svn: 210477
This patch teaches the backend how to check for the 'NoSignedWrap' flag on
binary operations to improve the emission of 'test' instructions.
If the result of a binary operation is known not to overflow we know that
resetting the Overflow flag is unnecessary and so we can avoid emitting
the test instruction.
Patch by Marcello Maggioni.
llvm-svn: 210468
According to Intel Software Optimization Manual
on Silvermont INC or DEC instructions require
an additional uop to merge the flags.
As a result, a branch instruction depending
on an INC or a DEC instruction incurs a 1 cycle penalty.
Differential Revision: http://reviews.llvm.org/D3990
llvm-svn: 210466
Summary:
We were being too strict and not accounting for undefs.
Added a test case and fixed another one where we improved codegen.
Reviewers: grosbach, nadav, delena
Subscribers: llvm-commits
Differential Revision: http://reviews.llvm.org/D4039
llvm-svn: 210361
When lowering a ISD::BRCOND into a test+branch, make sure that we
always use the correct condition code to emit the test operation.
This fixes PR19858: "i8 checked mul is wrong on x86".
Patch by Keno Fisher!
llvm-svn: 210032
This patch teaches the backend how to simplify/canonicalize dag node
sequences normally introduced by the backend when promoting certain dag nodes
with illegal vector type.
This patch adds two new combine rules:
1) fold (shuffle (bitcast (BINOP A, B)), Undef, <Mask>) ->
(shuffle (BINOP (bitcast A), (bitcast B)), Undef, <Mask>)
2) fold (BINOP (shuffle (A, Undef, <Mask>)), (shuffle (B, Undef, <Mask>))) ->
(shuffle (BINOP A, B), Undef, <Mask>).
Both rules are only triggered on the type-legalized DAG.
In particular, rule 1. is a target specific combine rule that attempts
to sink a bitconvert into the operands of a binary operation.
Rule 2. is a target independet rule that attempts to move a shuffle
immediately after a binary operation.
llvm-svn: 209930
Summary:
Separate the check for blend shuffle_vector masks into isBlendMask.
This function will also be used to check if a vector shuffle is legal. No
change in functionality was intended, but we ended up improving codegen on
two tests, which were being (more) optimized only if the resulting shuffle
was legal.
Reviewers: nadav, delena, andreadb
Subscribers: llvm-commits
Differential Revision: http://reviews.llvm.org/D3964
llvm-svn: 209923
This patch teaches the x86 backend how to efficiently lower ISD::BITCAST dag
nodes from MVT::f64 to MVT::v4i16 (and vice versa), and from MVT::f64 to
MVT::v8i8 (and vice versa).
This patch extends the logic from revision 208107 to also handle MVT::v4i16
and MVT::v8i8. Also, this patch correctly propagates Undef values when
performing the widening of a vector (example: when widening from v2i32 to
v4i32, the upper 64bits of the resulting vector are 'undef').
llvm-svn: 209451
ISD::VSELECT mask uses 1 to identify the first argument and 0 to identify the
second argument.
On the other hand, BLENDI uses 0 to identify the first argument and 1 to
identify the second argument.
Fix the generation of the blend mask to account for this difference.
The bug did not show up with r209043, because we were not checking for the
actual arguments of the blend instruction!
This commit also fixes the test cases.
Note: The same mask works for the BLENDr variant because the arguments are
swapped during instruction selection (see the BLENDXXrr patterns).
<rdar://problem/16975435>
llvm-svn: 209324
Summary:
When inserting an element that's coming from a vector load or a broadcast
of a vector (or scalar) load, combine the load into the insertps
instruction.
Added PerformINSERTPSCombine for the case where we need to fix the load
(load of a vector + insertps with a non-zero CountS).
Added patterns for the broadcasts.
Also added tests for SSE4.1, AVX, and AVX2.
Reviewers: delena, nadav, craig.topper
Subscribers: llvm-commits
Differential Revision: http://reviews.llvm.org/D3581
llvm-svn: 209156
- On ARM/ARM64 we get a vrev because the shuffle matching code is really smart. We still unroll anything that's not v4i32 though.
- On X86 we get a pshufb with SSSE3. Required more cleverness in isShuffleMaskLegal.
- On PPC we get a vperm for v8i16 and v4i32. v2i64 is unrolled.
llvm-svn: 209123
This is mostly a mechanical change changing all the call sites to the newer
chained-function construction pattern. This removes the horrible 15-parameter
constructor for the CallLoweringInfo in favour of setting properties of the call
via chained functions. No functional change beyond the removal of the old
constructors are intended.
llvm-svn: 209082
were added in SSE2, no SSSE3. Found this while auditing all uses of
SSSE3 in the X86 target. I don't actually expect this to make
a significant difference on anything and I don't have any detailed test
cases but I updated the existing test cases that already covered some of
this code path.
llvm-svn: 209056
vselects with constant masks, after legalization, will get turned into
specialized shuffle_vectors so they can be matched to blend+imm
instructions.
Fixed some tests.
llvm-svn: 209044
LowerVSELECT will, if possible, generate a X86ISD::BLENDI DAG node if the
condition is constant and we can emit that instruction, given the
subtarget.
This is not enough for all cases. An additional SELECTCombine optimization
will be committed.
Fixed tests that were expecting variable blends but where a blend+imm can
be generated.
Added test where we can't emit blend+immediate.
Added avx2 blend+imm tests.
llvm-svn: 209043
No functionality change intended. The types that previously were set to
lower as Expand or Legal are doing the same thing with this lowering
function.
llvm-svn: 209042
Added target specific combine rules to fold blend intrinsics according
to the following rules:
1) fold(blend A, A, Mask) -> A;
2) fold(blend A, B, <allZeros>) -> A;
3) fold(blend A, B, <allOnes>) -> B.
Added two new tests to verify that the new folding rules work for all
the optimized blend intrinsics.
llvm-svn: 208895
r208453 added support for having sret on the second parameter. In that
change, the code for copying sret into a virtual register was hoisted
into the loop that lowers formal parameters. This caused a "Wrong
topological sorting" assertion failure during scheduling when a
parameter is passed in memory. This change undoes that by creating a
second loop that deals with sret.
I'm worried that this fix is incomplete. I don't fully understand the
dependence issues. However, with this change we produce the same DAGs
we used to produce, so if they are broken, they are just as broken as
they have always been.
llvm-svn: 208637
1) Changed gather and scatter intrinsics. Now they are aligned with GCC built-ins. There is no more non-masked form. Masked intrinsic receives -1 if all lanes are executed.
2) I changed the function that works with intrinsics inside X86ISelLowering.cpp. I put all intrinsics in one table. I did it for INTRINSICS_W_CHAIN and plan to put all intrinsics from WO_CHAIN set to the same table in order to avoid the long-long "switch". (I wanted to use static map initialization that allowed by C++11 but I wasn't able to compile it on VS2012).
3) I added gather/scatter prefetch intrinsics.
4) I fixed MRMm encoding for masked instructions.
llvm-svn: 208522
We must validate the value type in TLI::getRegisterByName, because if we
don't and the wrong type was used with the IR intrinsic, then we'll assert
(because we won't be able to find a valid register class with which to
construct the requested copy operation). For PPC64, additionally, the type
information is necessary to decide between the 64-bit register and the 32-bit
subregister.
No functionality change.
llvm-svn: 208508
When lowering build_vector to an insertps, we would still lower it, even
if the source vectors weren't v4x32. This would break on avx if the source
was a v8x32. We now check the type of the source vectors.
llvm-svn: 208487
MSVC always places the implicit sret parameter after the implicit this
parameter of instance methods. We used to handle this for
x86_thiscallcc by allocating the sret parameter on the stack and leaving
the this pointer in ecx, but that doesn't handle alternative calling
conventions like cdecl, stdcall, fastcall, or the win64 convention.
Instead, change the verifier to allow sret on the second parameter.
This also requires changing the Mips and X86 backends to return the
argument with the sret parameter, instead of assuming that the sret
parameter comes first.
The Sparc backend also returns sret parameters in a register, but I
wasn't able to update it to handle secondary sret parameters. It
currently calls report_fatal_error if you feed it an sret in the second
parameter.
Reviewers: rafael.espindola, majnemer
Differential Revision: http://reviews.llvm.org/D3617
llvm-svn: 208453
Summary:
Also ran clang-format on the function. The code added is the last else
if block.
Reviewers: nadav, craig.topper, delena
Subscribers: llvm-commits
Differential Revision: http://reviews.llvm.org/D3518
llvm-svn: 208372
This patch teaches the backend how to combine packed SSE2/AVX2 arithmetic shift
intrinsics.
The rules are:
- Always fold a packed arithmetic shift by zero to its first operand;
- Convert a packed arithmetic shift intrinsic dag node into a ISD::SRA only if
the shift count is known to be smaller than the vector element size.
This patch also teaches to function 'getTargetVShiftByConstNode' how fold
target specific vector shifts by zero.
Added two new tests to verify that the DAGCombiner is able to fold
sequences of SSE2/AVX2 packed arithmetic shift calls.
llvm-svn: 208342
Summary:
Vectors built with zeros and elements in the same order as another
(source) vector are optimized to be built using a single insertps
instruction.
Also optimize when we move one element in a vector to a different place
in that vector while zeroing out some of the other elements.
Further optimizations are possible, described in TODO comments.
I will be implementing at least some of them in the near future.
Added some tests for different cases where this optimization triggers.
Reviewers: nadav, delena, craig.topper
Subscribers: llvm-commits
Differential Revision: http://reviews.llvm.org/D3521
llvm-svn: 208271
Before this patch, the backend always emitted a store+load sequence to
bitconvert from f64 to i64 the input operand of a ISD::BITCAST dag node that
performed a bitconvert from type MVT::f64 to type MVT::v2i32. The resulting
i64 node was then used to build a v2i32 vector.
With this patch, the backend now produces a cheaper SCALAR_TO_VECTOR from
MVT::f64 to MVT::v2f64. That SCALAR_TO_VECTOR is then followed by a "free"
bitcast to type MVT::v4i32. The elements of the resulting
v4i32 are then extracted to build a v2i32 vector (which is illegal and
therefore promoted to MVT::v2i64).
This is in general cheaper than emitting a stack store+load sequence
to bitconvert the operand from type f64 to type i64.
llvm-svn: 208107
This patch implements the infrastructure to use named register constructs in
programs that need access to specific registers (bare metal, kernels, etc).
So far, only the stack pointer is supported as a technology preview, but as it
is, the intrinsic can already support all non-allocatable registers from any
architecture.
llvm-svn: 208104
The Win64 docs are very clear that anything larger than 8 bytes is
passed by reference, and GCC MinGW64 honors that for __modti3 and
friends.
Patch by Jameson Nash!
llvm-svn: 208029
Summary:
Also ran clang-format on the function. The code added is the last else
if block.
Reviewers: nadav, craig.topper
Subscribers: llvm-commits
Differential Revision: http://reviews.llvm.org/D3518
llvm-svn: 207992
Currently, musttail codegen is relying on sibcall optimization, and
reporting a fatal error if fails. Sibcall optimization fails when stack
arguments need to be modified, which is insufficient for musttail.
The logic for moving arguments in memory safely is already implemented
for GuaranteedTailCallOpt. This change merely arranges for musttail
calls to use it.
No functional change for GuaranteedTailCallOpt.
Reviewers: espindola
Differential Revision: http://reviews.llvm.org/D3493
llvm-svn: 207598
Scaling factors are not free on X86 because every "complex" addressing mode
breaks the related instruction into 2 allocations instead of 1.
<rdar://problem/16730541>
llvm-svn: 207301
Summary:
If we're doing a v4f32/v4i32 shuffle on x86 with SSE4.1, we can lower
certain shufflevectors to an insertps instruction:
When most of the shufflevector result's elements come from one vector (and
keep their index), and one element comes from another vector or a memory
operand.
Added tests for insertps optimizations on shufflevector.
Added support and tests for v4i32 vector optimization.
Reviewers: nadav
Subscribers: llvm-commits
Differential Revision: http://reviews.llvm.org/D3475
llvm-svn: 207291
This is similar to the 'tail' marker, except that it guarantees that
tail call optimization will occur. It also comes with convervative IR
verification rules that ensure that tail call optimization is possible.
Reviewers: nicholas
Differential Revision: http://llvm-reviews.chandlerc.com/D3240
llvm-svn: 207143
This patch:
- Adds two new X86 builtin intrinsics ('int_x86_rdtsc' and
'int_x86_rdtscp') as GCCBuiltin intrinsics;
- Teaches the backend how to lower the two new builtins;
- Introduces a common function to lower READCYCLECOUNTER dag nodes
and the two new rdtsc/rdtscp intrinsics;
- Improves (and extends) the existing x86 test 'rdtsc.ll'; now test 'rdtsc.ll'
correctly verifies that both READCYCLECOUNTER and the two new intrinsics
work fine for both 64bit and 32bit Subtargets.
llvm-svn: 207127
This allows us to compile
return (mask & 0x8 ? a : b);
into
testb $8, %dil
cmovnel %edx, %esi
instead of
andl $8, %edi
shrl $3, %edi
cmovnel %edx, %esi
which we formed previously because dag combiner canonicalizes setcc of and into shift.
llvm-svn: 207088
Generating BZHI in the variable mask case, i.e. (and X, (sub (shl 1, N), 1)),
was already supported, but we were missing the constant-mask case. This patch
fixes that.
<rdar://problem/15480077>
llvm-svn: 206738
For a 256-bit BUILD_VECTOR consisting mostly of shuffles of 256-bit vectors,
both the BUILD_VECTOR and its operands may need to be legalized in multiple
steps. Consider:
(v8f32 (BUILD_VECTOR (extract_vector_elt (v8f32 %vreg0,) Constant<1>),
(extract_vector_elt %vreg0, Constant<2>),
(extract_vector_elt %vreg0, Constant<3>),
(extract_vector_elt %vreg0, Constant<4>),
(extract_vector_elt %vreg0, Constant<5>),
(extract_vector_elt %vreg0, Constant<6>),
(extract_vector_elt %vreg0, Constant<7>),
%vreg1))
a. We can't build a 256-bit vector efficiently so, we need to split it into
two 128-bit vecs and combine them with VINSERTX128.
b. Operands like (extract_vector_elt (v8f32 %vreg0), Constant<7>) needs to be
split into a VEXTRACTX128 and a further extract_vector_elt from the
resulting 128-bit vector.
c. The extract_vector_elt from b. is lowered into a shuffle to the first
element and a movss.
Depending on the order in which we legalize the BUILD_VECTOR and its
operands[1], buildFromShuffleMostly may be faced with:
(v4f32 (BUILD_VECTOR (extract_vector_elt
(vector_shuffle<1,u,u,u> (extract_subvector %vreg0, Constant<4>), undef),
Constant<0>),
(extract_vector_elt
(vector_shuffle<2,u,u,u> (extract_subvector %vreg0, Constant<4>), undef),
Constant<0>),
(extract_vector_elt
(vector_shuffle<3,u,u,u> (extract_subvector %vreg0, Constant<4>), undef),
Constant<0>),
%vreg1))
In order to figure out the underlying vector and their identity we need to see
through the shuffles.
[1] Note that the order in which operations and their operands are legalized is
only guaranteed in the first iteration of LegalizeDAG.
Fixes <rdar://problem/16296956>
llvm-svn: 206634
This patch teaches the backend how to efficiently lower logical and
arithmetic packed shifts on both SSE and AVX/AVX2 machines.
When possible, instead of scalarizing a vector shift, the backend should try
to expand the shift into a sequence of two packed shifts by immedate count
followed by a MOVSS/MOVSD.
Example
(v4i32 (srl A, (build_vector < X, Y, Y, Y>)))
Can be rewritten as:
(v4i32 (MOVSS (srl A, <Y,Y,Y,Y>), (srl A, <X,X,X,X>)))
[with X and Y ConstantInt]
The advantage is that the two new shifts from the example would be lowered into
X86ISD::VSRLI nodes. This is always cheaper than scalarizing the vector into
four scalar shifts plus four pairs of vector insert/extract.
llvm-svn: 206316
I found this from a particular GDB test suite case of inlining
(something similar is provided as a test case) but came across a few
other related cases (other callers of the same functions, and one other
instance of the same coding mistake in a separate function).
I'm not sure what the best way to test this is (let alone to cover the
other cases I discovered), so hopefully this sufficies - open to ideas.
llvm-svn: 206130
This removes the -segmented-stacks command line flag in favor of a
per-function "split-stack" attribute.
Patch by Luqman Aden and Alex Crichton!
llvm-svn: 205997
and isTargetCygwin() to isTargetWindowsCygwin() to be consistent with the
four Windows environments in Triple.h.
Suggestion by Saleem Abdulrasool!
llvm-svn: 205393
This adds back r204781.
Original message:
Aliases are just another name for a position in a file. As such, the
regular symbol resolutions are not applied. For example, given
define void @my_func() {
ret void
}
@my_alias = alias weak void ()* @my_func
@my_alias2 = alias void ()* @my_alias
We produce without this patch:
.weak my_alias
my_alias = my_func
.globl my_alias2
my_alias2 = my_alias
That is, in the resulting ELF file my_alias, my_func and my_alias are
just 3 names pointing to offset 0 of .text. That is *not* the
semantics of IR linking. For example, linking in a
@my_alias = alias void ()* @other_func
would require the strong my_alias to override the weak one and
my_alias2 would end up pointing to other_func.
There is no way to represent that with aliases being just another
name, so the best solution seems to be to just disallow it, converting
a miscompile into an error.
llvm-svn: 204934
This reverts commit r204781.
I will follow up to with msan folks to see what is what they
were trying to do with aliases to weak aliases.
llvm-svn: 204784
Aliases are just another name for a position in a file. As such, the
regular symbol resolutions are not applied. For example, given
define void @my_func() {
ret void
}
@my_alias = alias weak void ()* @my_func
@my_alias2 = alias void ()* @my_alias
We produce without this patch:
.weak my_alias
my_alias = my_func
.globl my_alias2
my_alias2 = my_alias
That is, in the resulting ELF file my_alias, my_func and my_alias are
just 3 names pointing to offset 0 of .text. That is *not* the
semantics of IR linking. For example, linking in a
@my_alias = alias void ()* @other_func
would require the strong my_alias to override the weak one and
my_alias2 would end up pointing to other_func.
There is no way to represent that with aliases being just another
name, so the best solution seems to be to just disallow it, converting
a miscompile into an error.
llvm-svn: 204781
This used to resort to splitting the 256-bit operation into two 128-bit
shuffles and then recombining the results.
Fixes <rdar://problem/16167303>
llvm-svn: 204735
I found three implementations of this. This splits it out into a new function
and uses it from the three places.
My plan is to add a fourth use when lowering a vector_shuffle:v16i16.
Compared the assembly output of test/CodeGen/X86 before and after.
The only change is due to how the first PSHUFB was generated in
LowerVECTOR_SHUFFLEv8i16. If the shuffle mask specified undef (i.e. -1), the
old implementation would write -1 * 2 and -1 * 2 + 1 (254 and 255) in the
control mask. Now we write 0x80. These are of course interchangeable since
bit 7 decides if a constant zero is written in the result byte. The other
instances of this code use 0x80 consistently.
Related to <rdar://problem/16167303>
llvm-svn: 204734
This can be observed with the old testcase of CodeGen/X86/pr12312.ll:
47c47
< vorps %ymm0, %ymm1, %ymm0
---
> vorps %ymm1, %ymm0, %ymm0
97c97
< vorps %ymm1, %ymm0, %ymm0
---
> vorps %ymm0, %ymm1, %ymm0
The vector VecIns is populated with all the values from VecInMap. This is done
while iterating VecInMap. VecInMap uses a hash of pointer values so the
resulting order can vary depending on the memory layout.
The fix is to populate the vector VecIns earlier as VecInMap is populated.
This is done in DAG traversal order.
Fixes <rdar://problem/16398806>
llvm-svn: 204623
Rather than LegalizeAction::Expand, this needs LegalizeAction::Promote to get
promoted to fp_to_sint v8f32->v8i32. This is a legal operation on AVX.
For that to work properly, we also need to teach the legalizer about the
specific promotion required here. The default vector promotion uses
bitcasting to a vector type of the same total size. We want to promote the
vector element type, effectively widening the operation and then truncating
the result. This is analogous to the current logic of how int_to_fp is
promoted.
The change also factors out some code from the int_to_fp promotion code to
ValueType::widenIntegerVectorElementType. This is now shared between
int_to_fp and fp_to_int.
There is no longer need for the custom lowering of fp_to_sint f32->v8i16 in
X86. It can now go through the new target-independent fp_to_*int promotion
logic.
I also checked that no other target uses Promote for these ops yet, so there
shouldn't be any unexpected change in behavior.
Fixes <rdar://problem/16202247>
llvm-svn: 204058
operator* on the by-operand iterators to return a MachineOperand& rather than
a MachineInstr&. At this point they almost behave like normal iterators!
Again, this requires making some existing loops more verbose, but should pave
the way for the big range-based for-loop cleanups in the future.
llvm-svn: 203865
This fixes the bug where we would bitcast the 64-bit floating point result
of cmpneqsd to a 64-bit integer even on 32-bit targets.
Differential Revision: http://llvm-reviews.chandlerc.com/D3009
llvm-svn: 203581
The syntax for "cmpxchg" should now look something like:
cmpxchg i32* %addr, i32 42, i32 3 acquire monotonic
where the second ordering argument gives the required semantics in the case
that no exchange takes place. It should be no stronger than the first ordering
constraint and cannot be either "release" or "acq_rel" (since no store will
have taken place).
rdar://problem/15996804
llvm-svn: 203559
When the MOVBE instructions are available, use them for 16-bit endian
swapping as well as for 32 and 64 bit.
The patterns were already present on the instructions, but weren't being
matched because the operation was unconditionally marked to 'Expand.'
Change that to be conditional on whether the MOVBE instructions are
available. Use 'rolw' to implement the in-register version (32 and 64
bit have the dedicated 'bswap' instruction for that).
Patch by Louis Gerbarg <lgg@apple.com>.
rdar://15479984
llvm-svn: 203524
The current approach to lower a vsetult is to flip the sign bit of the
operands, swap the operands and then use a (signed) pcmpgt. psubus (unsigned
saturating subtract) can be used to emulate a vsetult more efficiently:
+ case ISD::SETULT: {
+ // If the comparison is against a constant we can turn this into a
+ // setule. With psubus, setule does not require a swap. This is
+ // beneficial because the constant in the register is no longer
+ // destructed as the destination so it can be hoisted out of a loop.
I also enable lowering via psubus in a few other cases where it's clearly
beneficial: setule and setuge if minu/maxu cannot be used.
rdar://problem/14338765
Patch by Adam Nemet <anemet@apple.com>.
llvm-svn: 202301
On x86, shifting a vector by a scalar is significantly cheaper than shifting a
vector by another fully general vector. Unfortunately, because SelectionDAG
operates on just one basic block at a time, the shufflevector instruction that
reveals whether the right-hand side of a shift *is* really a scalar is often
not visible to CodeGen when it's needed.
This adds another handler to CodeGenPrepare, to sink any useful shufflevector
instructions down to the basic block where they're used, predicated on a target
hook (since on other architectures, doing so will often just introduce extra
real work).
rdar://problem/16063505
llvm-svn: 201655
Instead of expanding a packed shift into a sequence of scalar shifts,
the backend now tries (when possible) to convert the vector shift into a
vector multiply.
Before this change, a shift of a MVT::v8i16 vector by a
build_vector of constants was always scalarized into a long sequence of "vector
extracts + scalar shifts + vector insert".
With this change, if there is SSE2 support, we emit a single vector multiply.
This change also affects SSE4.1, AVX, AVX2 shifts:
- A shift of a MVT::v4i32 vector by a build_vector of non uniform constants
is now lowered when possible into a single SSE4.1 vector multiply.
- Packed v16i16 shift left by constant build_vector are now expanded when
possible into a single AVX2 vpmullw.
This change also improves the lowering of AVX512f vector shifts.
Added test CodeGen/X86/vec_shift6.ll with some code examples that are affected
by this change.
llvm-svn: 201271
I believe VZEXT_MOVL means "zero all vector elements except the first" (and
should have identical input & output types) whereas VZEXT means "zero extend
each element of a vector (discarding higher elements if necessary)".
For example:
(v4i32 (vzext (v16i8 ...)))
should zero extend the low 4 bytes of the incoming vector to 32-bits,
discarding higher bytes.
However, somewhere in the past, these two concepts had become confused, even
leading to a nonsensical VSEXT_MOVL.
This re-merges the nodes where appropriate (all VSEXT_MOVL -> VSEXT, VZEXT_MOVL
-> VZEXT when it's an actual extension).
rdar://problem/15981990
llvm-svn: 200918
Calls with inalloca are lowered by skipping all stores for arguments
passed in memory and the initial stack adjustment to allocate argument
memory.
Now the frontend is responsible for the memory layout, and the backend
doesn't have to do any work. As a result these changes are pretty
minimal.
Reviewers: echristo
Differential Revision: http://llvm-reviews.chandlerc.com/D2637
llvm-svn: 200596
Before this patch we used getIntImmCost from TargetTransformInfo to determine if
a load of a constant should be converted to just a constant, but the threshold
for this was set to an arbitrary value. This value works well for the two
targets (X86 and ARM) that implement this target-hook, but it isn't
target-independent at all.
Now targets have the possibility to decide directly if this optimization should
be performed. The default value is set to false to preserve the current
behavior. The target hook has been moved to TargetLowering, which removed the
last use and need of TargetTransformInfo in SelectionDAG.
llvm-svn: 200271
This commit teaches the X86 backend to create the same X86 instructions when it
lowers an sadd/ssub with overflow intrinsic and a conditional branch that uses
that overflow result. This allows SelectionDAG to recognize and remove one of
the redundant operations.
This fixes <rdar://problem/15874016> and <rdar://problem/15661073>.
Reviewed by Nadav
llvm-svn: 199976
Add target specific rules for combining vselect dag nodes into movss/movsd
when possible.
If the vector type of the vselect dag node in input is either MVT::v4i13 or
MVT::v4f32, then try to fold according to rules:
1) fold (vselect (build_vector (0, -1, -1, -1)), A, B) -> (movss A, B)
2) fold (vselect (build_vector (-1, 0, 0, 0)), A, B) -> (movss B, A)
If the vector type of the vselect dag node in input is either MVT::v2i64 or
MVT::v2f64 (and we have SSE2), then try to fold according to rules:
3) fold (vselect (build_vector (0, -1)), A, B) -> (movsd A, B)
4) fold (vselect (build_vector (-1, 0)), A, B) -> (movsd B, A)
llvm-svn: 199683
MSVC on x64 requires that we create image relative symbol
references to refer to RTTI data. Seeing as how there is no way to
explicitly make reference to a given relocation type in LLVM IR, pattern
match expressions of the form &foo - &__ImageBase.
Differential Revision: http://llvm-reviews.chandlerc.com/D2523
llvm-svn: 199312
promotion code, Tablegen will now select FPExt for floating point promotions
(previously it had returned AExt, which is not valid for floating point types).
Any out-of-tree targets that were relying on AExt being returned for FP
promotions will need to update their code check for FPExt instead.
llvm-svn: 199252
Representing dllexport/dllimport as distinct linkage types prevents using
these attributes on templates and inline functions.
Instead of introducing further mixed linkage types to include linkonce and
weak ODR, the old import/export linkage types are replaced with a new
separate visibility-like specifier:
define available_externally dllimport void @f() {}
@Var = dllexport global i32 1, align 4
Linkage for dllexported globals and functions is now equal to their linkage
without dllexport. Imported globals and functions must be either
declarations with external linkage, or definitions with
AvailableExternallyLinkage.
llvm-svn: 199218
This fixes a regression intruced by r198113.
Revision r198113 introduced an algorithm that tries to fold a vector shift
by immediate count into a build_vector if the input vector is a known vector
of constants.
However the algorithm only worked under the assumption that the input vector
type and the shift type are exactly the same.
This patch disables the folding of vector shift by immediate count if the
input vector type and the shift value type are not the same.
llvm-svn: 199213
Representing dllexport/dllimport as distinct linkage types prevents using
these attributes on templates and inline functions.
Instead of introducing further mixed linkage types to include linkonce and
weak ODR, the old import/export linkage types are replaced with a new
separate visibility-like specifier:
define available_externally dllimport void @f() {}
@Var = dllexport global i32 1, align 4
Linkage for dllexported globals and functions is now equal to their linkage
without dllexport. Imported globals and functions must be either
declarations with external linkage, or definitions with
AvailableExternallyLinkage.
llvm-svn: 199204
This moves the check up into the parent class so that all targets can use it
without having to copy (and keep in sync) the same error message.
llvm-svn: 198579
__builtin_returnaddress requires that the value passed into is be a constant.
However, at -O0 even a constant expression may not be converted to a constant.
Emit an error message intead of crashing.
llvm-svn: 198531
vector shift by immedate count (VSHLI/VSRLI/VSRAI) into a build_vector when
the vector in input to the shift is a build_vector of all constants or UNDEFs.
Target specific nodes for packed shifts by immediate count are in
general introduced by function 'getTargetVShiftByConstNode' (in
X86ISelLowering.cpp) when lowering shift operations, SSE/AVX immediate
shift intrinsics and (only in very few cases) SIGN_EXTEND_INREG dag
nodes.
This patch adds extra rules for simplifying vector shifts inside
function 'getTargetVShiftByConstNode'.
Added file test/CodeGen/X86/vec_shift5.ll to verify that packed
shifts by immediate are correctly folded into a build_vector when the
input vector to the shift dag node is a vector of constants or undefs.
llvm-svn: 198113
This reverts commit r197481, recommiting r197469 with an extra fix.
The vastart_save_xmm_regs pseudo-instruction expands to a test and a
branch, so it modifies EFLAGS. Mark it so, or else the scheduler might
place it in the middle of another test+branch.
This fixes a bug exposed by r192750, which changed the initial scheduler
to source-order as part of enabling the MI Scheduler for X86.
This re-commit changes the VASTART_SAVE_XMM_REGS custom inserter not to
try to save %flags, and adds a test that catches the bad behavior of
r197469.
<rdar://problem/15627766>
llvm-svn: 197503
http://llvm.org/bugs/show_bug.cgi?id=18045
Short issue description:
For X86 machines with sse < sse4.1 we got failures for some
particular load/store vector sequences:
$ clang-trunk -m32 -O2 test-case.c
fatal error: error in backend: Cannot select: 0x4200920: v4i32,ch = load 0x41d6ab0, 0x4205850,
0x41dcb10<LD16[getelementptr inbounds ([4 x i32]* @e, i32 0, i32 0)](align=4)> [ORD=82]
[ID=58]
0x4205850: i32 = X86ISD::Wrapper 0x41d5490 [ORD=26] [ID=43]
0x41d5490: i32 = TargetGlobalAddress<[4 x i32]* @e> 0 [ORD=26] [ID=23]
0x41dcb10: i32 = undef [ID=2]
The reason is that EltsFromConsecutiveLoads could emit such load instruction
both before and after legalize stage. Though this instruction is not legal for
machines with SSSE3 and lower.
The fix: In EltsFromConsecutiveLoads, if we have passed legalize stage, we
check whether nodes it emits are legal.
P.S.: If you get failure in time from 12:00 and till 22:00 (UTC-8),
perhaps I'll slow with response, so you better reject this commit. Thanks!
llvm-svn: 197492
Added scalar compare VCMPSS, VCMPSD.
Implemented LowerSELECT for scalar FP operations.
I replaced FSETCCss, FSETCCsd with one node type FSETCCs.
Node extract_vector_elt(v16i1/v8i1, idx) returns an element of type i1.
llvm-svn: 197384
While it's safe for the X86-specific shift nodes, dag combining will
kill generic nodes. Insert an AND to make it safe, isel will nuke it
as x86's shift instructions have an implicit AND.
Fixes PR16108, which contains a contraption to hit this case in between
constant folders.
llvm-svn: 197228
Most users would be surprised if "isCOFF" and "isMachO" were simultaneously
true, unless they'd put the compiler in a box with a gun attached to a photon
detector.
This makes sure precisely one of the three formats is true for any triple and
simplifies some target logic based on that.
llvm-svn: 196934
target independent.
Most of the x86 specific stackmap/patchpoint handling was necessitated by the
use of the native address-mode format for frame index operands. PEI has now
been modified to treat stackmap/patchpoint similarly to DEBUG_INFO, allowing
us to use a simple, platform independent register/offset pair for frame
indexes on stackmap/patchpoints.
Notes:
- Folding is now platform independent and automatically supported.
- Emiting patchpoints with direct memory references now just involves calling
the TargetLoweringBase::emitPatchPoint utility method from the target's
XXXTargetLowering::EmitInstrWithCustomInserter method. (See
X86TargetLowering for an example).
- No more ugly platform-specific operand parsers.
This patch shouldn't change the generated output for X86.
llvm-svn: 195944
- Fix bug in (vsext (vzext x)) -> (vsext x) in SIGN_EXTEND_IN_REG
lowering where we need to check whether x is a vector type (in-reg
type) of i8, i16 or i32; otherwise, that optimization is not valid.
llvm-svn: 195779
A Direct stack map location records the address of frame index. This
address is itself the value that the runtime requested. This differs
from IndirectMemRefOp locations, which refer to a stack locations from
which the requested values must be loaded. Direct locations can
directly communicate the address if an alloca, while IndirectMemRefOp
handle register spills.
For example:
entry:
%a = alloca i64...
llvm.experimental.stackmap(i32 <ID>, i32 <shadowBytes>, i64* %a)
Since both the alloca and stackmap intrinsic are in the entry block,
and the intrinsic takes the address of the alloca, the runtime can
assume that LLVM will not substitute alloca with any intervening
value. This must be verified by the runtime by checking that the stack
map's location is a Direct location type. The runtime can then
determine the alloca's relative location on the stack immediately after
compilation, or at any time thereafter. This differs from Register and
Indirect locations, because the runtime can only read the values in
those locations when execution reaches the instruction address of the
stack map.
llvm-svn: 195712
Utilizing the 8 and 16 bit comparison instructions, even when an input can
be folded into the comparison instruction itself, is typically not worth it.
There are too many partial register stalls as a result, leading to significant
slowdowns. By always performing comparisons on at least 32-bit
registers, performance of the calculation chain leading to the
comparison improves. Continue to use the smaller comparisons when
minimizing size, as that allows better folding of loads into the
comparison instructions.
rdar://15386341
llvm-svn: 195496
- When simplifying the mask generation for BLEND, check whether that mask is
also consumed by other non-BLEND insns. If true, skip that simplification.
llvm-svn: 195476
AMD's processors family K7, K8, K10, K12, K15 and K16 are known to have SHLD/SHRD instructions with very poor latency. Optimization guides for these processors recommend using an alternative sequence of instructions. For these AMD's processors, I disabled folding (or (x << c) | (y >> (64 - c))) when we are not optimizing for size.
It might be beneficial to disable this folding for some of the Intel's processors. However, since I couldn't find specific recommendations regarding using SHLD/SHRD instructions on Intel's processors, I haven't disabled this peephole for Intel.
llvm-svn: 195383
clang optimizes tail calls, as in this example:
int foo(void);
int bar(void) {
return foo();
}
where the call is transformed to:
calll .L0$pb
.L0$pb:
popl %eax
.Ltmp0:
addl $_GLOBAL_OFFSET_TABLE_+(.Ltmp0-.L0$pb), %eax
movl foo@GOT(%eax), %eax
popl %ebp
jmpl *%eax # TAILCALL
However, the GOT references must all be resolved at dlopen() time, and so this
approach cannot be used with lazy dynamic linking (e.g. using RTLD_LAZY), which
usually populates the PLT with stubs that perform the actual resolving.
This patch changes X86TargetLowering::LowerCall() to skip tail call
optimization, if the called function is a global or external symbol.
Patch by Dimitry Andric!
PR15086
llvm-svn: 195318
This patch reapplies r193676 with an additional fix for the Hexagon backend. The
SystemZ backend has already been fixed by r194148.
The Type Legalizer recognizes that VSELECT needs to be split, because the type
is to wide for the given target. The same does not always apply to SETCC,
because less space is required to encode the result of a comparison. As a result
VSELECT is split and SETCC is unrolled into scalar comparisons.
This commit fixes the issue by checking for VSELECT-SETCC patterns in the DAG
Combiner. If a matching pattern is found, then the result mask of SETCC is
promoted to the expected vector mask type for the given target. Now the type
legalizer will split both VSELECT and SETCC.
This allows the following X86 DAG Combine code to sucessfully detect the MIN/MAX
pattern. This fixes PR16695, PR17002, and <rdar://problem/14594431>.
Reviewed by Nadav
llvm-svn: 194542
This patch moves the jump address materialization inside the noop slide. This
enables patching of the materialization itself or its complete removal. This
patch also adds the ability to define scratch registers that can be used safely
by the code called from the patchpoint intrinsic. At least one scratch register
is required, because that one is used for the materialization of the jump
address. This patch depends on D2009.
Differential Revision: http://llvm-reviews.chandlerc.com/D2074
Reviewed by Andy
llvm-svn: 194306
The idea of the AnyReg Calling Convention is to provide the call arguments in
registers, but not to force them to be placed in a paticular order into a
specified set of registers. Instead it is up tp the register allocator to assign
any register as it sees fit. The same applies to the return value (if
applicable).
Differential Revision: http://llvm-reviews.chandlerc.com/D2009
Reviewed by Andy
llvm-svn: 194293
The Type Legalizer recognizes that VSELECT needs to be split, because the type
is to wide for the given target. The same does not always apply to SETCC,
because less space is required to encode the result of a comparison. As a result
VSELECT is split and SETCC is unrolled into scalar comparisons.
This commit fixes the issue by checking for VSELECT-SETCC patterns in the DAG
Combiner. If a matching pattern is found, then the result mask of SETCC is
promoted to the expected vector mask type for the given target. This mask has
usually the same size as the VSELECT return type (except for Intel KNL). Now the
type legalizer will split both VSELECT and SETCC.
This allows the following X86 DAG Combine code to sucessfully detect the MIN/MAX
pattern. This fixes PR16695, PR17002, and <rdar://problem/14594431>.
Reviewed by Nadav
llvm-svn: 193676
This optimization is not SSE specific so I am moving it to DAGco.
The new scalar_to_vector dag node exposed a missing pattern in the AArch64 target that I needed to add.
llvm-svn: 193393
Calling _chkstk is required on ELF as well as COFF on Windows. Without
_chkstk, functions requiring large stack crash in initialization code.
Previous code tested for COFF format but not Mach-O and this patch modifies
the code to test for Windows OS (both Windows target and MingW target)
but not Mach-O object format: Looks like macho environment was used to
build some EFI code.
Credits to Andrew MacPherson.
llvm-svn: 193289
Without _chkstk functions requiring large stack crash in
initialization code. Previous code tested for COFF format but
not Mach-O and this patch modifies the code to test for Windows.
Credits to Andrew MacPherson.
llvm-svn: 193263
On sandy bridge (PR17654) we now get
vpxor %xmm1, %xmm1, %xmm1
vpunpckhbw %xmm1, %xmm0, %xmm2
vpunpcklbw %xmm1, %xmm0, %xmm0
vinsertf128 $1, %xmm2, %ymm0, %ymm0
On haswell it's a simple
vpmovzxbw %xmm0, %ymm0
There is a maze of duplicated and dead transforms and patterns in this
area. Remove the dead custom lowering of zext v8i16 to v8i32, that's
already handled by LowerAVXExtend.
llvm-svn: 193262
the instruction defenitions and ISEL reflect this.
Prior to this patch these instructions took an i32i8imm, and the high bits were
dropped during encoding. This led to incorrect behavior for shifts by
immediates higher than 255. This patch fixes that issue by detecting large
immediate shifts and returning constant zero (for logical shifts) or capping
the shift amount at an encodable value (for arithmetic shifts).
Fixes <rdar://problem/14968098>
llvm-svn: 193096
Consider the following:
typedef unsigned short ushort4U __attribute__((ext_vector_type(4),
aligned(2)));
typedef unsigned short ushort4 __attribute__((ext_vector_type(4)));
typedef unsigned short ushort8 __attribute__((ext_vector_type(8)));
typedef int int4 __attribute__((ext_vector_type(4)));
int4 __bbase_cvt_int(ushort4 v) {
ushort8 a;
a.lo = v;
return _mm_cvtepu16_epi32(a);
}
This generates the, not unreasonable, IR:
define <4 x i32> @foo0(double %v.coerce) nounwind ssp {
%tmp = bitcast double %v.coerce to <4 x i16>
%tmp1 = shufflevector <4 x i16> %tmp, <4 x i16> undef, <8 x i32> <i32
%0, i32 1, i32 2, i32 3, i32 undef, i32 undef, i32 undef, i32 undef>
%tmp2 = tail call <4 x i32> @llvm.x86.sse41.pmovzxwd(<8 x i16> %tmp1)
ret <4 x i32> %tmp2
}
The problem is when type legalization gets hold of the v4i16. It
legalizes that by spilling to the stack, then doing a zero-extending
load. Things go even more silly from there, ending up with something
like:
_foo0:
movsd %xmm0, -8(%rsp) <== Spill to the stack.
movq -8(%rsp), %xmm0 <== Reload it right back out.
pmovzxwd %xmm0, %xmm1 <== Here's what we actually asked for.
pblendw $1, %xmm1, %xmm0 <== We don't need this at all
pmovzxwd %xmm0, %xmm0 <== We already did this
ret
The v8i8 to v8i16 zext intrinsic gives even worse results, with two
table lookups via pshufb instructions(!!).
To avoid all that, we can move the bitcasting until after we've formed
the wider (legal) vector type. Then our normal codegen flows along
nicely and we get the expected:
_foo0:
pmovzxwd %xmm0, %xmm0
ret
rdar://15245794
llvm-svn: 192866
- Type of index used in extract_vector_elt or insert_vector_elt supposes
to be TLI.getVectorIdxTy() which is pointer type on most targets. It'd
better to truncate (or zero-extend in case it's changed later) it to
mask element type to guarantee they are matching instead of asserting
that.
llvm-svn: 192722
- Lower signed division by constant powers-of-2 to target-independent
DAG operators instead of target-dependent ones to support them better
on targets where vector types are legal but shift operators on that
types are illegal. E.g., on AVX, PSRAW is only available on <8 x i16>
though <16 x i16> is a legal type.
llvm-svn: 192721
In AVX 256bit vectors are valid vectors and therefore the Type Legalizer doesn't
split the VSELECT and SETCC nodes. AVX only supports MIN/MAX on 128bit vectors
and this fix enables vector splitting for this special case in the X86 DAG
Combiner.
This fix is related to PR16695, PR17002, and <rdar://problem/14594431>.
llvm-svn: 191131
The Type Legalizer recognizes that VSELECT needs to be split, because the type
is to wide for the given target. The same does not always apply to SETCC,
because less space is required to encode the result of a comparison. As a result
VSELECT is split and SETCC is unrolled into scalar comparisons.
This commit fixes the issue by checking for VSELECT-SETCC patterns in the DAG
Combiner. If a matching pattern is found, then the result mask of SETCC is
promoted to the expected vector mask for the given target. This mask has usually
te same size as the VSELECT return type (except for Intel KNL). Now the type
legalizer will split both VSELECT and SETCC.
This allows the following X86 DAG Combine code to sucessfully detect the MIN/MAX
pattern. This fixes PR16695, PR17002, and <rdar://problem/14594431>.
llvm-svn: 191130
If the DAG already has only legal types, then the second round of DAG combines
is skipped. In this case VSELECT+SETCC patterns that match a more efficient
instruction (e.g. min/max) are never recognized.
This fix allows VSELECT+SETCC combines if the types are already legal before DAG
type legalization.
Reviewer: Nadav
llvm-svn: 190105
This change came about primarily because of two issues in the existing code.
Niether of:
define i64 @test1(i64 %val) {
%in = trunc i64 %val to i32
tail call i32 @ret32(i32 returned %in)
ret i64 %val
}
define i64 @test2(i64 %val) {
tail call i32 @ret32(i32 returned undef)
ret i32 42
}
should be tail calls, and the function sameNoopInput is responsible. The main
problem is that it is completely symmetric in the "tail call" and "ret" value,
but in reality different things are allowed on each side.
For these cases:
1. Any truncation should lead to a larger value being generated by "tail call"
than needed by "ret".
2. Undef should only be allowed as a source for ret, not as a result of the
call.
Along the way I noticed that a mismatch between what this function treats as a
valid truncation and what the backends see can lead to invalid calls as well
(see x86-32 test case).
This patch refactors the code so that instead of being based primarily on
values which it recurses into when necessary, it starts by inspecting the type
and considers each fundamental slot that the backend will see in turn. For
example, given a pathological function that returned {{}, {{}, i32, {}}, i32}
we would consider each "real" i32 in turn, and ask if it passes through
unchanged. This is much closer to what the backend sees as a result of
ComputeValueVTs.
Aside from the bug fixes, this eliminates the recursion that's going on and, I
believe, makes the bulk of the code significantly easier to understand. The
trade-off is the nasty iterators needed to find the real types inside a
returned value.
llvm-svn: 187787
Due to the weird and wondeful usual arithmetic conversions, some
calculations involving negative values were getting performed in
uint32_t and then promoted to int64_t, which is really not a good
idea.
Patch by Katsuhiro Ueno.
llvm-svn: 187703
All insertf*/extractf* functions replaced with insert/extract since we have insertf and inserti forms.
Added lowering for INSERT_VECTOR_ELT / EXTRACT_VECTOR_ELT for 512-bit vectors.
Added lowering for EXTRACT/INSERT subvector for 512-bit vectors.
Added a test.
llvm-svn: 187491
CustomLowerNode was not being called during SplitVectorOperand,
meaning custom legalization could not be used by targets.
This also adds a test case for NVPTX that depends on this custom
legalization.
Differential Revision: http://llvm-reviews.chandlerc.com/D1195
Attempt to fix the buildbots by making the X86 test I just added platform independent
llvm-svn: 187202
This reverts commit 187198. It broke the bots.
The soft float test probably needs a -triple because of name differences.
On the hard float test I am getting a "roundss $1, %xmm0, %xmm0", instead of
"vroundss $1, %xmm0, %xmm0, %xmm0".
llvm-svn: 187201
CustomLowerNode was not being called during SplitVectorOperand,
meaning custom legalization could not be used by targets.
This also adds a test case for NVPTX that depends on this custom
legalization.
Differential Revision: http://llvm-reviews.chandlerc.com/D1195
llvm-svn: 187198
Use PMIN/PMAX for UGE/ULE vector comparions to reduce the number of required
instructions. This trick also works for UGT/ULT, but there is no advantage in
doing so. It wouldn't reduce the number of instructions and it would actually
reduce performance.
Reviewer: Ben
radar:5972691
llvm-svn: 186432
Summary:
This patch adds explicit calling convention types for the Win64 and
System V/x86-64 ABIs. This allows code to override the default, and use
the Win64 convention on a target that wants to use SysV (and
vice-versa). This is needed to implement the `ms_abi` and `sysv_abi` GNU
attributes.
Reviewers:
CC:
llvm-svn: 186144
in-tree implementations of TargetLoweringBase::isFMAFasterThanMulAndAdd in
order to resolve the following issues with fmuladd (i.e. optional FMA)
intrinsics:
1. On X86(-64) targets, ISD::FMA nodes are formed when lowering fmuladd
intrinsics even if the subtarget does not support FMA instructions, leading
to laughably bad code generation in some situations.
2. On AArch64 targets, ISD::FMA nodes are formed for operations on fp128,
resulting in a call to a software fp128 FMA implementation.
3. On PowerPC targets, FMAs are not generated from fmuladd intrinsics on types
like v2f32, v8f32, v4f64, etc., even though they promote, split, scalarize,
etc. to types that support hardware FMAs.
The function has also been slightly renamed for consistency and to force a
merge/build conflict for any out-of-tree target implementing it. To resolve,
see comments and fixed in-tree examples.
llvm-svn: 185956
Fixes PR16146: gdb.base__call-ar-st.exp fails after
pre-RA-sched=source fixes.
Patch by Xiaoyi Guo!
This also fixes an unsupported dbg.value test case. Codegen was
previously incorrect but the test was passing by luck.
llvm-svn: 182885
Shuffles that only move an element into position 0 of the vector are common in
the output of the loop vectorizer and often generate suboptimal code when SSSE3
is not available. Lower them to vector shifts if possible.
We still prefer palignr over psrldq because it has higher throughput on
sandybridge.
llvm-svn: 182102
X86ISelLowering has support to treat:
(icmp ne (and (xor %flags, -1), (shl 1, flag)), 0)
as if it were actually:
(icmp eq (and %flags, (shl 1, flag)), 0)
However, r179386 has code at the InstCombine level to handle this.
llvm-svn: 181145
I think it's almost impossible to fold atomic fences profitably under
LLVM/C++11 semantics. As a result, this is now unused and just
cluttering up the target interface.
llvm-svn: 179940
As packed comparisons in AVX/SSE produce all 0s or all 1s in each SIMD lane,
vector select could be simplified to AND/OR or removed if one or both values
being selected is all 0s or all 1s.
llvm-svn: 179267
This patch is revised based on patch from Victor Umansky
<victor.umansky@intel.com>. More cases are handled in X86's bool
simplification, i.e.
- SETCC_CARRY
- value is truncated to i1 with AND
As a by-product, PR5443 is also fixed.
llvm-svn: 179265
During LTO, the target options on functions within the same Module may
change. This would necessitate resetting some of the back-end. Do this for X86,
because it's a Friday afternoon.
llvm-svn: 178917
- RDRAND always clears the destination value when a random value is not
available (i.e. CF == 0). This value is truncated or zero-extended as
the false boolean value to be returned. Boolean simplification needs
to skip this 'zext' or 'trunc' node.
llvm-svn: 178312
To enable a load of a call address to be folded with that call, this
load is moved from outside of callseq into callseq. Such a moving
adds a non-glued node (that load) into a glued sequence. This non-glue
load is only removed when DAG selection folds them into a memory form
call instruction. When such instruction selection is disabled, it breaks
DAG schedule.
To prevent that, such moving is disabled when target favors register
indirect call.
Previous workaround disabling CALL32m/CALL64m insn selection is removed.
llvm-svn: 178308
indirect through a memory address is to load the memory address into
a register and then call indirect through the register.
This patch implements this improvement by modifying SelectionDAG to
force a function address which is a memory reference to be loaded
into a virtual register.
Patch by Sriram Murali.
llvm-svn: 178171
- It's still considered aligned when the specified alignment is larger
than the natural alignment;
- The new alignment for the high 128-bit vector should be min(16,
alignment) as the pointer is advanced by 16, a power-of-2 offset.
llvm-svn: 177947
MinGW is almost completely compatible to MSVC, with the exception of the _tls_array global not being available.
Patch by David Nadlinger!
llvm-svn: 177257
LegalizeDAG.cpp uses the value of the comparison operands when checking
the legality of BR_CC, so DAGCombiner should do the same.
v2:
- Expand more BR_CC value types for NVPTX
v3:
- Expand correct BR_CC value types for Hexagon, Mips, and XCore.
llvm-svn: 176694
That can usually be lowered efficiently and is common in sandybridge code.
It would be nice to do this in DAGCombiner but we can't insert arbitrary
BUILD_VECTORs this late.
Fixes PR15462.
llvm-svn: 176634
- Phi nodes should be replaced/updated after lowering CMOV into branch
because 'mainMBB' updating operand in Phi node is changed.
- Add EFLAGS in livein before lowering the 2nd CMOV. It's necessary as
we will reuse the EFLAGS generated before the 1st lowered CMOV, which
won't clobber EFLAGS. However, we need explicitly specify that.
- '-attr=-cmov' test case are added.
llvm-svn: 176598
- Clear 'mayStore' flag when loading from the atomic variable before the
spin loop
- Clear kill flag from one use to multiple use in registers forming the
address to that atomic variable
- don't use a physical register as live-in register in BB (neither entry
nor landing pad.) by copying it into virtual register
(patch by Cameron Zwarich)
llvm-svn: 176538
* Only apply divide bypass optimization when not optimizing for size.
* Fixed bug caused by constant for 0 value of type Int32,
used dividend type to generate the constant instead.
* For atom x86-64 apply the divide bypass to use 16-bit divides instead of
64-bit divides when operand values are small enough.
* Added lit tests for 64-bit divide bypass.
Patch by Tyler Nowicki!
llvm-svn: 176442
- ISD::SHL/SRL/SRA must have either both scalar or both vector operands
but TLI.getShiftAmountTy() so far only return scalar type. As a
result, backend logic assuming that breaks.
- Rename the original TLI.getShiftAmountTy() to
TLI.getScalarShiftAmountTy() and re-define TLI.getShiftAmountTy() to
return target-specificed scalar type or the same vector type as the
1st operand.
- Fix most TICG logic assuming TLI.getShiftAmountTy() a simple scalar
type.
llvm-svn: 176364
sext <4 x i1> to <4 x i64>
sext <4 x i8> to <4 x i64>
sext <4 x i16> to <4 x i64>
I'm running Combine on SIGN_EXTEND_IN_REG and revert SEXT patterns:
(sext_in_reg (v4i64 anyext (v4i32 x )), ExtraVT) -> (v4i64 sext (v4i32 sext_in_reg (v4i32 x , ExtraVT)))
The sext_in_reg (v4i32 x) may be lowered to shl+sar operations.
The "sar" does not exist on 64-bit operation, so lowering sext_in_reg (v4i64 x) has no vector solution.
I also added a cost of this operations to the AVX costs table.
llvm-svn: 175619
conditions are met:
1. They share the same operand and are in the same BB.
2. Both outputs are used.
3. The target has a native instruction that maps to ISD::FSINCOS node or
the target provides a sincos library call.
Implemented the generic optimization in sdisel and enabled it for
Mac OSX. Also added an additional optimization for x86_64 Mac OSX by
using an alternative entry point __sincos_stret which returns the two
results in xmm0 / xmm1.
rdar://13087969
PR13204
llvm-svn: 173755
This catches many cases where we can emit a more efficient shuffle for a
specific mask or when the mask contains undefs. Once the splat is lowered to
unpacks we can't do that anymore.
There is a possibility of moving the promotion after pshufb matching, but I'm
not sure if pshufb with a mask loaded from memory is faster than 3 shuffles, so
I avoided that for now.
llvm-svn: 173569
(defined by the x32 ABI) mode, in which case its pointers are 32-bits
in size. This knowledge is also added to X86RegisterInfo that now
returns the appropriate registers in getPointerRegClass.
There are many outcomes to this change. In order to keep the patches
separate and manageable, we start by focusing on some simple testable
cases. The patch adds a test with passing a pointer to a function -
focusing on the difference between the two data models for x86-64.
Another test is added for handling of 'sret' arguments (and
functionality is added in X86ISelLowering to make it work).
A note on naming: the "x32 ABI" document refers to the AMD64
architecture (in LLVM it's distinguished by being is64Bits() in the
x86 subtarget) with two variations: the LP64 (default) data model, and
the ILP32 data model. This patch adds predicates to the subtarget
which are consistent with this naming scheme.
llvm-svn: 173503
- Add list of physical registers clobbered in pseudo atomic insts
Physical registers are clobbered when pseudo atomic instructions are
expanded. Add them in clobber list to prevent DAG scheduler to
mis-schedule them after these insns are declared side-effect free.
- Add test case from Michael Kuperstein <michael.m.kuperstein@intel.com>
llvm-svn: 173200
Previously we tried to infer it from the bit width size, with an added
IsIEEE argument for the PPC/IEEE 128-bit case, which had a default
value. This default value allowed bugs to creep in, where it was
inappropriate.
llvm-svn: 173138
The optimization handles esoteric cases but adds a lot of complexity both to the X86 backend and to other backends.
This optimization disables an important canonicalization of chains of SEXT nodes and makes SEXT and ZEXT asymmetrical.
Disabling the canonicalization of consecutive SEXT nodes into a single node disables other DAG optimizations that assume
that there is only one SEXT node. The AVX mask optimizations is one example. Additionally this optimization does not update the cost model.
llvm-svn: 172968
PR 14848. The lowered sequence is based on the existing sequence the target-independent
DAG Combiner creates for the scalar case.
Patch by Zvi Rackover.
llvm-svn: 171953
a TargetMachine to construct (and thus isn't always available), to an
analysis group that supports layered implementations much like
AliasAnalysis does. This is a pretty massive change, with a few parts
that I was unable to easily separate (sorry), so I'll walk through it.
The first step of this conversion was to make TargetTransformInfo an
analysis group, and to sink the nonce implementations in
ScalarTargetTransformInfo and VectorTargetTranformInfo into
a NoTargetTransformInfo pass. This allows other passes to add a hard
requirement on TTI, and assume they will always get at least on
implementation.
The TargetTransformInfo analysis group leverages the delegation chaining
trick that AliasAnalysis uses, where the base class for the analysis
group delegates to the previous analysis *pass*, allowing all but tho
NoFoo analysis passes to only implement the parts of the interfaces they
support. It also introduces a new trick where each pass in the group
retains a pointer to the top-most pass that has been initialized. This
allows passes to implement one API in terms of another API and benefit
when some other pass above them in the stack has more precise results
for the second API.
The second step of this conversion is to create a pass that implements
the TargetTransformInfo analysis using the target-independent
abstractions in the code generator. This replaces the
ScalarTargetTransformImpl and VectorTargetTransformImpl classes in
lib/Target with a single pass in lib/CodeGen called
BasicTargetTransformInfo. This class actually provides most of the TTI
functionality, basing it upon the TargetLowering abstraction and other
information in the target independent code generator.
The third step of the conversion adds support to all TargetMachines to
register custom analysis passes. This allows building those passes with
access to TargetLowering or other target-specific classes, and it also
allows each target to customize the set of analysis passes desired in
the pass manager. The baseline LLVMTargetMachine implements this
interface to add the BasicTTI pass to the pass manager, and all of the
tools that want to support target-aware TTI passes call this routine on
whatever target machine they end up with to add the appropriate passes.
The fourth step of the conversion created target-specific TTI analysis
passes for the X86 and ARM backends. These passes contain the custom
logic that was previously in their extensions of the
ScalarTargetTransformInfo and VectorTargetTransformInfo interfaces.
I separated them into their own file, as now all of the interface bits
are private and they just expose a function to create the pass itself.
Then I extended these target machines to set up a custom set of analysis
passes, first adding BasicTTI as a fallback, and then adding their
customized TTI implementations.
The fourth step required logic that was shared between the target
independent layer and the specific targets to move to a different
interface, as they no longer derive from each other. As a consequence,
a helper functions were added to TargetLowering representing the common
logic needed both in the target implementation and the codegen
implementation of the TTI pass. While technically this is the only
change that could have been committed separately, it would have been
a nightmare to extract.
The final step of the conversion was just to delete all the old
boilerplate. This got rid of the ScalarTargetTransformInfo and
VectorTargetTransformInfo classes, all of the support in all of the
targets for producing instances of them, and all of the support in the
tools for manually constructing a pass based around them.
Now that TTI is a relatively normal analysis group, two things become
straightforward. First, we can sink it into lib/Analysis which is a more
natural layer for it to live. Second, clients of this interface can
depend on it *always* being available which will simplify their code and
behavior. These (and other) simplifications will follow in subsequent
commits, this one is clearly big enough.
Finally, I'm very aware that much of the comments and documentation
needs to be updated. As soon as I had this working, and plausibly well
commented, I wanted to get it committed and in front of the build bots.
I'll be doing a few passes over documentation later if it sticks.
Commits to update DragonEgg and Clang will be made presently.
llvm-svn: 171681
1. Add code to estimate register pressure.
2. Add code to select the unroll factor based on register pressure.
3. Add bits to TargetTransformInfo to provide the number of registers.
llvm-svn: 171469
In order to cost subvector insertion and extraction, we need to know
the type of the subvector being extracted.
No functionality change.
llvm-svn: 171453
into their new header subdirectory: include/llvm/IR. This matches the
directory structure of lib, and begins to correct a long standing point
of file layout clutter in LLVM.
There are still more header files to move here, but I wanted to handle
them in separate commits to make tracking what files make sense at each
layer easier.
The only really questionable files here are the target intrinsic
tablegen files. But that's a battle I'd rather not fight today.
I've updated both CMake and Makefile build systems (I think, and my
tests think, but I may have missed something).
I've also re-sorted the includes throughout the project. I'll be
committing updates to Clang, DragonEgg, and Polly momentarily.
llvm-svn: 171366
directly.
This is in preparation for removing the use of the 'Attribute' class as a
collection of attributes. That will shift to the AttributeSet class instead.
llvm-svn: 171253
register. In most cases we actually compare or select YMM-sized registers
and mixing the two types creates horrible code. This commit optimizes
some of the transition sequences.
PR14657.
llvm-svn: 171148
The vector truncs were scalarized during LegalizeVectorOps, later vectorized again by some DAGCombine optimization
and finally, lowered by a dagcombing optimization. Now, they are properly lowered during LegalizeVectorOps.
No new testcase because the original testcases still work.
llvm-svn: 171146
pmuludq is slow, but it turns out that all the unpacking and packing of the
scalarized mul is even slower. 10% speedup on loop-vectorized paq8p.
llvm-svn: 170985
We match the pattern "x >= y ? x-y : 0" into "subus x, y" and two special cases
if y is a constant. DAGCombiner canonicalizes those so we first have to undo the
canonicalization for those cases. The pattern occurs in gzip when the loop
vectorizer is enabled. Part of PR14613.
llvm-svn: 170273
mention the inline memcpy / memset expansion code is a mess?
This patch split the ZeroOrLdSrc argument into two: IsMemset and ZeroMemset.
The first indicates whether it is expanding a memset or a memcpy / memmove.
The later is whether the memset is a memset of zero. It's totally possible
(likely even) that targets may want to do different things for memcpy and
memset of zero.
llvm-svn: 169959
Also added more comments to explain why it is generally ok to return true.
- Rename getOptimalMemOpType argument IsZeroVal to ZeroOrLdSrc. It's meant to
be true for loaded source (memcpy) or zero constants (memset). The poor name
choice is probably some kind of legacy issue.
llvm-svn: 169954
1. Teach it to use overlapping unaligned load / store to copy / set the trailing
bytes. e.g. On 86, use two pairs of movups / movaps for 17 - 31 byte copies.
2. Use f64 for memcpy / memset on targets where i64 is not legal but f64 is. e.g.
x86 and ARM.
3. When memcpy from a constant string, do *not* replace the load with a constant
if it's not possible to materialize an integer immediate with a single
instruction (required a new target hook: TLI.isIntImmLegal()).
4. Use unaligned load / stores more aggressively if target hooks indicates they
are "fast".
5. Update ARM target hooks to use unaligned load / stores. e.g. vld1.8 / vst1.8.
Also increase the threshold to something reasonable (8 for memset, 4 pairs
for memcpy).
This significantly improves Dhrystone, up to 50% on ARM iOS devices.
rdar://12760078
llvm-svn: 169791
There are still bugs in this pass, as well as other issues that are
being worked on, but the bugs are crashers that occur pretty easily in
the wild. Test cases have been sent to the original commit's review
thread.
This reverts the commits:
r169671: Fix a logic error.
r169604: Move the popcnt tests to an X86 subdirectory.
r168931: Initial commit adding the pass.
llvm-svn: 169683
understand target implementation of any_extend / extload, just generate
zero_extend in place of any_extend for liveouts when the target knows the
zero_extend will be implicit (e.g. ARM ldrb / ldrh) or folded (e.g. x86 movz).
rdar://12771555
llvm-svn: 169536
Sooooo many of these had incorrect or strange main module includes.
I have manually inspected all of these, and fixed the main module
include to be the nearest plausible thing I could find. If you own or
care about any of these source files, I encourage you to take some time
and check that these edits were sensible. I can't have broken anything
(I strictly added headers, and reordered them, never removed), but they
may not be the headers you'd really like to identify as containing the
API being implemented.
Many forward declarations and missing includes were added to a header
files to allow them to parse cleanly when included first. The main
module rule does in fact have its merits. =]
llvm-svn: 169131
This revision attempts to recognize following population-count pattern:
while(a) { c++; ... ; a &= a - 1; ... },
where <c> and <a>could be used multiple times in the loop body.
TODO: On X8664 and ARM, __buildin_ctpop() are not expanded to a efficent
instruction sequence, which need to be improved in the following commits.
Reviewed by Nadav, really appreciate!
llvm-svn: 168931
The stack realignment code was fixed to work when there is stack realignment and
a dynamic alloca is present so this shouldn't cause correctness issues anymore.
Note that this also enables generation of AVX instructions for memset
under the assumptions:
- Unaligned loads/stores are always fast on CPUs supporting AVX
- AVX is not slower than SSE
We may need some tweaked heuristics if one of those assumptions turns out not to
be true.
Effectively reverts r58317. Part of PR2962.
llvm-svn: 167967
chain is correctly setup.
As an example, if the original load must happen before later stores, we need
to make sure the constructed VZEXT_LOAD is constrained to be before the stores.
rdar://12684358
llvm-svn: 167859
- Add RTM code generation support throught 3 X86 intrinsics:
xbegin()/xend() to start/end a transaction region, and xabort() to abort a
tranaction region
llvm-svn: 167573
r165941: Resubmit the changes to llvm core to update the functions to
support different pointer sizes on a per address space basis.
Despite this commit log, this change primarily changed stuff outside of
VMCore, and those changes do not carry any tests for correctness (or
even plausibility), and we have consistently found questionable or flat
out incorrect cases in these changes. Most of them are probably correct,
but we need to devise a system that makes it more clear when we have
handled the address space concerns correctly, and ideally each pass that
gets updated would receive an accompanying test case that exercises that
pass specificaly w.r.t. alternate address spaces.
However, from this commit, I have retained the new C API entry points.
Those were an orthogonal change that probably should have been split
apart, but they seem entirely good.
In several places the changes were very obvious cleanups with no actual
multiple address space code added; these I have not reverted when
I spotted them.
In a few other places there were merge conflicts due to a cleaner
solution being implemented later, often not using address spaces at all.
In those cases, I've preserved the new code which isn't address space
dependent.
This is part of my ongoing effort to clean out the partial address space
code which carries high risk and low test coverage, and not likely to be
finished before the 3.2 release looms closer. Duncan and I would both
like to see the above issues addressed before we return to these
changes.
llvm-svn: 167222
The adc/sbb optimization is to able to convert following expression
into a single adc/sbb instruction:
(ult) ... = x + 1 // where the ult is unsigned-less-than comparison
(ult) ... = x - 1
This change is to flip the "x >u y" (i.e. ugt comparison) in order
to expose the adc/sbb opportunity.
llvm-svn: 167180
%0 = load <8 x i16>* %dest
%1 = shufflevector <8 x i16> %0, <8 x i16> %in,
<8 x i32> < i32 0, i32 1, i32 2, i32 3, i32 13, i32 undef, i32 14, i32 14>
store <8 x i16> %1, <8 x i16>* %dest
We get:
vmovlpd (%eax), %xmm0, %xmm0
instead of:
vmovaps (%eax), %xmm1
vmovsd %xmm1, %xmm0, %xmm0
No extra test-case is added. I just fixed the existing one
(also it uses FileCheck now).
llvm-svn: 166971
- As there's no 64-bit GPRs in 32-bit mode, a custom conversion from v2u32 to
v2f32 is added to improve the efficiency of the code generated.
llvm-svn: 166545
- Replace v4i8/v8i8 -> v8f32 DAG combine with custom lowering to reduce
DAG combine overhead.
- Extend the support to v4i16/v8i16 as well.
llvm-svn: 166487
which is supposed to consistently raise SIGTRAP across all systems. In contrast,
__builtin_trap() behave differently on different systems. e.g. it raises SIGTRAP on ARM, and
SIGILL on X86. The purpose of __builtin_debugtrap() is to consistently provide "trap"
functionality, in the mean time preserve the compatibility with on gcc on __builtin_trap().
The X86 backend is already able to handle debugtrap(). This patch is to:
1) make front-end recognize "__builtin_debugtrap()" (emboddied in the one-line change to Clang).
2) In DAG legalization phase, by default, "debugtrap" will be replaced with "trap", which
make the __builtin_debugtrap() "available" to all existing ports without the hassle of
changing their code.
3) If trap-function is specified (via -trap-func=xyz to llc), both __builtin_debugtrap() and
__builtin_trap() will be expanded into the function call of the specified trap function.
This behavior may need change in the future.
The provided testing-case is to make sure 2) and 3) are working for ARM port, and we
already have a testing case for x86.
llvm-svn: 166300
- If INSERT_VECTOR_ELT is supported (above SSE2, either by custom
sequence of legal insn), transform BUILD_VECTOR into SHUFFLE +
INSERT_VECTOR_ELT if most of elements could be built from SHUFFLE with few
(so far 1) elements being inserted.
llvm-svn: 166288
- MBB address is only valid as an immediate value in Small & Static
code/relocation models. On other models, LEA is needed to load IP address of
the restore MBB.
- A minor fix of MBB in MC lowering is added as well to enable target
relocation flag being propagated into MC.
llvm-svn: 166084
- Add custom FP_TO_SINT on v8i16 (and v8i8 which is legalized as v8i16 due to
vector element-wise widening) to reduce DAG combiner and its overhead added
in X86 backend.
llvm-svn: 166036
Original message:
The attached is the fix to radar://11663049. The optimization can be outlined by following rules:
(select (x != c), e, c) -> select (x != c), e, x),
(select (x == c), c, e) -> select (x == c), x, e)
where the <c> is an integer constant.
The reason for this change is that : on x86, conditional-move-from-constant needs two instructions;
however, conditional-move-from-register need only one instruction.
While the LowerSELECT() sounds to be the most convenient place for this optimization, it turns out to be a bad place. The reason is that by replacing the constant <c> with a symbolic value, it obscure some instruction-combining opportunities which would otherwise be very easy to spot. For that reason, I have to postpone the change to last instruction-combining phase.
The change passes the test of "make check-all -C <build-root/test" and "make -C project/test-suite/SingleSource".
Original message since r165661:
My previous change has a bug: I negated the condition code of a CMOV, and go ahead creating a new CMOV using the *ORIGINAL* condition code.
llvm-svn: 166017
Chandler Carruth