of architectures: SSE2, SSSE3, SSE4.1, AVX, and AVX2.
Unfortunately, this exposses the absolute horror of the code we generate
for many of these patterns. Anyone wanting to familiarize themselves
with the x86 backend and improve performance could do a lot of good
sitting down and making these test cases not look so terrible. While the
new vector shuffle code I'm working on well help some, it won't fix all
of the crimes here.
llvm-svn: 218807
These tests are far and away the best sext and zext tests we have for
vectors. I'm going to merge the other similar tests into them and expand
the ISA coverage.
llvm-svn: 218800
script to make them nice and predictable. This will ease updating them
for the new vector shuffle lowering and seeing the delta if any.
llvm-svn: 218795
avx-sext.ll using my new script.
Also add an AVX2 mode to this test.
Part of cleaning up the test suite before enabling the new vector
shuffle lowering. This also highlights some of the abysmal failures of
the old shuffle lowering. Check out those 'pinsrw' and 'pextrw'
sequences!
llvm-svn: 218794
argument of the llvm.dbg.declare/llvm.dbg.value intrinsics.
Previously, DIVariable was a variable-length field that has an optional
reference to a Metadata array consisting of a variable number of
complex address expressions. In the case of OpPiece expressions this is
wasting a lot of storage in IR, because when an aggregate type is, e.g.,
SROA'd into all of its n individual members, the IR will contain n copies
of the DIVariable, all alike, only differing in the complex address
reference at the end.
By making the complex address into an extra argument of the
dbg.value/dbg.declare intrinsics, all of the pieces can reference the
same variable and the complex address expressions can be uniqued across
the CU, too.
Down the road, this will allow us to move other flags, such as
"indirection" out of the DIVariable, too.
The new intrinsics look like this:
declare void @llvm.dbg.declare(metadata %storage, metadata %var, metadata %expr)
declare void @llvm.dbg.value(metadata %storage, i64 %offset, metadata %var, metadata %expr)
This patch adds a new LLVM-local tag to DIExpressions, so we can detect
and pretty-print DIExpression metadata nodes.
What this patch doesn't do:
This patch does not touch the "Indirect" field in DIVariable; but moving
that into the expression would be a natural next step.
http://reviews.llvm.org/D4919
rdar://problem/17994491
Thanks to dblaikie and dexonsmith for reviewing this patch!
Note: I accidentally committed a bogus older version of this patch previously.
llvm-svn: 218787
argument of the llvm.dbg.declare/llvm.dbg.value intrinsics.
Previously, DIVariable was a variable-length field that has an optional
reference to a Metadata array consisting of a variable number of
complex address expressions. In the case of OpPiece expressions this is
wasting a lot of storage in IR, because when an aggregate type is, e.g.,
SROA'd into all of its n individual members, the IR will contain n copies
of the DIVariable, all alike, only differing in the complex address
reference at the end.
By making the complex address into an extra argument of the
dbg.value/dbg.declare intrinsics, all of the pieces can reference the
same variable and the complex address expressions can be uniqued across
the CU, too.
Down the road, this will allow us to move other flags, such as
"indirection" out of the DIVariable, too.
The new intrinsics look like this:
declare void @llvm.dbg.declare(metadata %storage, metadata %var, metadata %expr)
declare void @llvm.dbg.value(metadata %storage, i64 %offset, metadata %var, metadata %expr)
This patch adds a new LLVM-local tag to DIExpressions, so we can detect
and pretty-print DIExpression metadata nodes.
What this patch doesn't do:
This patch does not touch the "Indirect" field in DIVariable; but moving
that into the expression would be a natural next step.
http://reviews.llvm.org/D4919
rdar://problem/17994491
Thanks to dblaikie and dexonsmith for reviewing this patch!
llvm-svn: 218778
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
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
cases.
While clearly we don't need the AVX vector width, these ISA extensions
often cause us to select different instructions and we should cover them
even with the narrow vector width.
Also, while here, nuke the stress_test2 contents. There is no reason to
try to FileCheck this entire body when it is mostly a test for
successfully surviving the code generator.
llvm-svn: 218710
shuffle tests to match that used in the script I posted and now used
consistently in 128-bit tests.
Nothing interesting changing here, just using the label name as the
FileCheck label and a slightly more general comment marker consumption
strategy.
llvm-svn: 218709
updating script so that they are more thorough and consistent.
Specific fixes here include:
- Actually test VEX-encoded AVX mnemonics.
- Actually use an SSE 4.1 run to test SSE 4.1 features!
- Correctly check instructions sequences from the start of the function.
- Elide the shuffle operands and comment designator in a consistent way.
- Test all of the architectures instead of just the ones I was motivated
to manually author.
I've gone back through and fixed up any egregious issues I spotted. Let
me know if I missed something you really dislike.
One downside to this is that we're now not as diligently using FileCheck
variables for registers. I would be much more concerned with this if we
had larger register usage, but there just aren't that interesting of
register choices here and most of the registers are constrained by the
ABI. Ultimately, I don't think this is likely to be the maintenance
burden for these tests and updating them again should be staright
forward.
llvm-svn: 218707
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
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
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
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
to significantly more sane sentinels. Notably, everywhere else in the
backend's representation of shuffles uses '-1' to represent undef. The
target shuffle masks really shouldn't diverge from that, especially as
in a few places they are manipulated by shared code.
This causes us to lose some undef lanes in various test masks. I want to
get these back, but technically it isn't invalid and there are a *lot*
of bugs here so I want to try to establish a saner baseline for fixing
some of the bugs by aligning the specific senitnel values used.
llvm-svn: 218561
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
Machine Sink uses loop depth information to select between successors BBs to
sink machine instructions into, where BBs within smaller loop depths are
preferable. This patch adds support for choosing between successors by using
profile information from BlockFrequencyInfo instead, whenever the information
is available.
Tested it under SPEC2006 train (average of 30 runs for each program); ~1.5%
execution speedup in average on x86-64 darwin.
<rdar://problem/18021659>
llvm-svn: 218472
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
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
I made a mistake in the previous commit and produced the wrong pattern.
Fix that. Also make one more shuffle pattern byte-based rather than
word-based, and add two more blend patterns.
llvm-svn: 218439
shuffles rather than word shuffles.
As you might guess, these were built starting from the word shuffle test
cases and I failed to properly port a bunch of them and left them as
widened word shuffle test cases. We still have a couple of tests that
check our ability to widen shuffles, but now we will test the actual
byte shuffle quite a bit better.
llvm-svn: 218438
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
pool data being loaded into a vector register.
The comments take the form of:
# ymm0 = [a,b,c,d,...]
# xmm1 = <x,y,z...>
The []s are used for generic sequential data and the <>s are used for
specifically ConstantVector loads. Undef elements are printed as the
letter 'u', integers in decimal, and floating point values as floating
point values. Suggestions on improving the formatting or other aspects
of the display are very welcome.
My primary use case for this is to be able to FileCheck test masks
passed to vector shuffle instructions in-register. It isn't fantastic
for that (no decoding special zeroing semantics or other tricks), but it
at least puts the mask onto an instruction line that could reasonably be
checked. I've updated many of the new vector shuffle lowering tests to
leverage this in their test cases so that we're actually checking the
shuffle masks remain as expected.
Before implementing this, I tried a *bunch* of different approaches.
I looked into teaching the MCInstLower code to scan up the basic block
and find a definition of a register used in a shuffle instruction and
then decode that, but this seems incredibly brittle and complex.
I talked to Hal a lot about the "right" way to do this: attach the raw
shuffle mask to the instruction itself in some form of unencoded
operands, and then use that to emit the comments. I still think that's
the optimal solution here, but it proved to be beyond what I'm up for
here. In particular, it seems likely best done by completing the
plumbing of metadata through these layers and attaching the shuffle mask
in metadata which could have fully automatic dropping when encoding an
actual instruction.
llvm-svn: 218377
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
These are just test cases, no actual code yet. This establishes the
baseline fallback strategy we're starting from on AVX2 and the expected
lowering we use on AVX1.
Also, these test cases are very much generated. I've manually crafted
the specific pattern set that I'm hoping will be useful at exercising
the lowering code, but I've not (and could not) manually verify *all* of
these. I've spot checked and they seem legit to me.
As with the rest of vector shuffling, at a certain point the only really
useful way to check the correctness of this stuff is through fuzz
testing.
llvm-svn: 218267
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
The PSHUFB mask decode routine used to assert if the mask index was out of
range (<0 or greater than the size of the vector). The problem is, we can
legitimately have a PSHUFB with a large index using intrinsics. The
instruction only uses the least significant 4 bits. This change removes the
assert and masks the index to match the instruction behaviour.
llvm-svn: 218242
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
in the high and low 128-bit lanes of a v8f32 vector.
No functionality change yet, but wanted to set up the baseline for my
next patch which will make these quite a bit better. =]
llvm-svn: 218224
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
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
generating the test cases to format things more consistently and
actually catch all the operand sequences that should be elided in favor
of the asm comments. No actual changes here.
llvm-svn: 218210
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
This expands the integer cases to cover the fact that AVX2 moves their
lane-crossing shuffles into the integer domain. It also adds proper
support for AVX2 run lines and the "ALL" group when it doesn't matter.
llvm-svn: 218206
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
under AVX.
This really just documents the current state of the world. I'm going to
try to flesh it out to cover any test cases I plan to improve prior to
improving them so that the delta made by changes is actually visible to
code reviewers.
This is made easier by the fact that I now have a script to automate the
process of producing test cases including the check lines. =]
llvm-svn: 218199
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
duplication of check lines. The idea is to have broad sets of
compilation modes that will frequently diverge without having to always
and immediately explode to the precise ISA feature set.
While this already helps due to VEX encoded differences, it will help
much more as I teach the new shuffle lowering about more of the new VEX
encoded instructions which can still be used to implement 128-bit
shuffles.
llvm-svn: 218188
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
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
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
Not sure why I only did SSSE3 here. Also, I've left out some of the SSE2
ones because the shuffles are so absurd it's not worth transcribing
them. Will try to fix them to be sane and then check them.
llvm-svn: 218114
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
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
On MachO, and MachO only, we cannot have a truly empty function since that
breaks the linker logic for atomizing the section.
When we are emitting a frame pointer, the presence of an unreachable will
create a cfi instruction pointing past the last instruction. This is perfectly
fine. The FDE information encodes the pc range it applies to. If some tool
cannot handle this, we should explicitly say which bug we are working around
and only work around it when it is actually relevant (not for ELF for example).
Given the unreachable we could omit the .cfi_def_cfa_register, but then
again, we could also omit the entire function prologue if we wanted to.
llvm-svn: 217801
introduced in r217629.
We were returning the old sext instead of the new zext as the promoted instruction!
Thanks Joerg Sonnenberger for the test case.
llvm-svn: 217800
Peephole optimization was folding MOVSDrm, which is a zero-extending double
precision floating point load, into ADDPDrr, which is a SIMD add of two packed
double precision floating point values.
(before)
%vreg21<def> = MOVSDrm <fi#0>, 1, %noreg, 0, %noreg; mem:LD8[%7](align=16)(tbaa=<badref>) VR128:%vreg21
%vreg23<def,tied1> = ADDPDrr %vreg20<tied0>, %vreg21; VR128:%vreg23,%vreg20,%vreg21
(after)
%vreg23<def,tied1> = ADDPDrm %vreg20<tied0>, <fi#0>, 1, %noreg, 0, %noreg; mem:LD8[%7](align=16)(tbaa=<badref>) VR128:%vreg23,%vreg20
X86InstrInfo::foldMemoryOperandImpl already had the logic that prevented this
from happening. However the check wasn't being conducted for loads from stack
objects. This commit factors out the logic into a new function and uses it for
checking loads from stack slots are not zero-extending loads.
rdar://problem/18236850
llvm-svn: 217799
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
missing specific checks.
While there is a lot of redundancy here where all-but-one mode use the
same code generation, I'd rather have each variant spelled out and
checked so that readers aren't misled by an omission in the test suite.
llvm-svn: 217765
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
Do
(shl (add x, c1), c2) -> (add (shl x, c2), c1 << c2)
This is already done for multiplies, but since multiplies
by powers of two are turned into shifts, we also need
to handle it here.
This might want checks for isLegalAddImmediate to avoid
transforming an add of a legal immediate with one that isn't.
llvm-svn: 217610
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
This is an extension of the change made with r215820:
http://llvm.org/viewvc/llvm-project?view=revision&revision=215820
That patch allowed combining of splatted vector FP constants that are multiplied.
This patch allows combining non-uniform vector FP constants too by relaxing the
check on the type of vector. Also, canonicalize a vector fmul in the
same way that we already do for scalars - if only one operand of the fmul is a
constant, make it operand 1. Otherwise, we miss potential folds.
This fold is also done by -instcombine, but it's possible that extra
fmuls may have been generated during lowering.
Differential Revision: http://reviews.llvm.org/D5254
llvm-svn: 217599
Summary:
In AT&T annotation for both x86_64 and x32 calls should be printed as
callq in assembly. It's only a matter of correct mnemonic, object output
is ok.
Test Plan: trivial test added
Reviewers: nadav, dschuff, craig.topper
Subscribers: llvm-commits, zinovy.nis
Differential Revision: http://reviews.llvm.org/D5213
llvm-svn: 217435
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
Previously, fast-isel would not clean up after failing to select a call
instruction, because it would have called flushLocalValueMap() which moves
the insertion point, making SavedInsertPt in selectInstruction() invalid.
Fixing this by making SavedInsertPt a member variable, and having
flushLocalValueMap() update it.
This removes some redundant code at -O0, and more importantly fixes PR20863.
Differential Revision: http://reviews.llvm.org/D5249
llvm-svn: 217401
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
parsing (and latent bug in the instruction definitions).
This is effectively a revert of r136287 which tried to address
a specific and narrow case of immediate operands failing to be accepted
by x86 instructions with a pretty heavy hammer: it introduced a new kind
of operand that behaved differently. All of that is removed with this
commit, but the test cases are both preserved and enhanced.
The core problem that r136287 and this commit are trying to handle is
that gas accepts both of the following instructions:
insertps $192, %xmm0, %xmm1
insertps $-64, %xmm0, %xmm1
These will encode to the same byte sequence, with the immediate
occupying an 8-bit entry. The first form was fixed by r136287 but that
broke the prior handling of the second form! =[ Ironically, we would
still emit the second form in some cases and then be unable to
re-assemble the output.
The reason why the first instruction failed to be handled is because
prior to r136287 the operands ere marked 'i32i8imm' which forces them to
be sign-extenable. Clearly, that won't work for 192 in a single byte.
However, making thim zero-extended or "unsigned" doesn't really address
the core issue either because it breaks negative immediates. The correct
fix is to make these operands 'i8imm' reflecting that they can be either
signed or unsigned but must be 8-bit immediates. This patch backs out
r136287 and then changes those places as well as some others to use
'i8imm' rather than one of the extended variants.
Naturally, this broke something else. The custom DAG nodes had to be
updated to have a much more accurate type constraint of an i8 node, and
a bunch of Pat immediates needed to be specified as i8 values.
The fallout didn't end there though. We also then ceased to be able to
match the instruction-specific intrinsics to the instructions so
modified. Digging, this is because they too used i32 rather than i8 in
their signature. So I've also switched those intrinsics to i8 arguments
in line with the instructions.
In order to make the intrinsic adjustments of course, I also had to add
auto upgrading for the intrinsics.
I suspect that the intrinsic argument types may have led everything down
this rabbit hole. Pretty happy with the result.
llvm-svn: 217310
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
This problem is bigger than just fsub, but this is the minimum fix to solve
fneg for PR20556 ( http://llvm.org/bugs/show_bug.cgi?id=20556 ), and we solve
zero subtraction with the same change.
llvm-svn: 217286
This reverts commit r217211.
Both the bfd ld and gold outputs were valid. They were using a Rela relocation,
so the value present in the relocated location was not used, which caused me
to misread the output.
llvm-svn: 217264
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
This CL replaces the constant DarwinX86AsmBackend.PushInstrSize with a method
that lets the backend account for different sizes of "push %reg" instruction
sizes.
llvm-svn: 217020
The only valid lowering of atomic stores in the X86 backend was mov from
register to memory. As a result, storing an immediate required a useless copy
of the immediate in a register. Now these can be compiled as a simple mov.
Similarily, adding/and-ing/or-ing/xor-ing an
immediate to an atomic location (but through an atomic_store/atomic_load,
not a fetch_whatever intrinsic) can now make use of an 'add $imm, x(%rip)'
instead of using a register. And the same applies to inc/dec.
This second point matches the first issue identified in
http://llvm.org/bugs/show_bug.cgi?id=17281
llvm-svn: 216980
We have been using .init-array for most systems for quiet some time,
but tools like llc are still defaulting to .ctors because the old
option was never changed.
This patch makes llc default to .init-array and changes the option to
be -use-ctors.
Clang is not affected by this. It has its own fancier logic.
llvm-svn: 216905
Summary:
Fixes a FIXME in MachineSinking. Instead of using the simple heuristics
in isPostDominatedBy, use the real MachinePostDominatorTree. The old
heuristics caused instructions to sink unnecessarily, and might create
register pressure.
Test Plan:
Added a NVPTX codegen test to verify that our change is in effect. It also
shows the unnecessary register pressure caused by over-sinking. Updated
affected tests in AArch64 and X86.
Reviewers: eliben, meheff, Jiangning
Reviewed By: Jiangning
Subscribers: jholewinski, aemerson, mcrosier, llvm-commits
Differential Revision: http://reviews.llvm.org/D4814
llvm-svn: 216862
