Summary:
I was testing using this expansion logic in other cases besides
NVPTX, and found some runtime failures due to the lack of a check
for a zero length memcpy/memset before the loop. There is already
such a check in the memmove expansion code though.
Reviewers: hfinkel
Subscribers: jholewinski, wdng, llvm-commits
Differential Revision: https://reviews.llvm.org/D34707
llvm-svn: 306541
CFI instructions that set appropriate cfa offset and cfa register are now
inserted in emitEpilogue() in X86FrameLowering.
Majority of the changes in this patch:
1. Ensure that CFI instructions do not affect code generation.
2. Enable maintaining correct information about cfa offset and cfa register
in a function when basic blocks are reordered, merged, split, duplicated.
These changes are target independent and described below.
Changed CFI instructions so that they:
1. are duplicable
2. are not counted as instructions when tail duplicating or tail merging
3. can be compared as equal
Add information to each MachineBasicBlock about cfa offset and cfa register
that are valid at its entry and exit (incoming and outgoing CFI info). Add
support for updating this information when basic blocks are merged, split,
duplicated, created. Add a verification pass (CFIInfoVerifier) that checks
that outgoing cfa offset and register of predecessor blocks match incoming
values of their successors.
Incoming and outgoing CFI information is used by a late pass
(CFIInstrInserter) that corrects CFA calculation rule for a basic block if
needed. That means that additional CFI instructions get inserted at basic
block beginning to correct the rule for calculating CFA. Having CFI
instructions in function epilogue can cause incorrect CFA calculation rule
for some basic blocks. This can happen if, due to basic block reordering,
or the existence of multiple epilogue blocks, some of the blocks have wrong
cfa offset and register values set by the epilogue block above them.
Patch by Violeta Vukobrat.
Differential Revision: https://reviews.llvm.org/D18046
llvm-svn: 306529
The original patch was an improvement to IR ValueTracking on non-negative
integers. It has been checked in to trunk (D18777, r284022). But was disabled by
default due to performance regressions.
Perf impact has improved. The patch would be enabled by default.
Reviewers: reames
Differential Revision: https://reviews.llvm.org/D34101
Patch by: Olga Chupina <olga.chupina@intel.com>
llvm-svn: 306528
Summary:
This commit allows matchSelectPattern to recognize clamp of float
arguments in the presence of FMF the same way as already done for
integers.
This case is a little different though. With integers, given the
min/max pattern is recognized, DAGBuilder starts selecting MIN/MAX
"automatically". That is not the case for float, because for them only
full FMINNAN/FMINNUM/FMAXNAN/FMAXNUM ISD nodes exist and they do care
about NaNs. On the other hand, some backends (e.g. X86) have only
FMIN/FMAX nodes that do not care about NaNS and the former NAN/NUM
nodes are illegal thus selection is not happening. So I decided to do
such kind of transformation in IR (InstCombiner) instead of
complicating the logic in the backend.
Reviewers: spatel, jmolloy, majnemer, efriedma, craig.topper
Reviewed By: efriedma
Subscribers: hiraditya, javed.absar, n.bozhenov, llvm-commits
Patch by Andrei Elovikov <andrei.elovikov@intel.com>
Differential Revision: https://reviews.llvm.org/D33186
llvm-svn: 306525
Summary:
This commit adds the tests for clamp pattern as a prerequisite of
D33186 to make the impact of that fix more clear and also to document
current behavior.
Reviewers: spatel, jmolloy
Reviewed By: spatel
Subscribers: n.bozhenov, llvm-commits
Patch by Andrei Elovikov <andrei.elovikov@intel.com>
Differential Revision: https://reviews.llvm.org/D34350
llvm-svn: 306524
The benchmarking summarized in
http://lists.llvm.org/pipermail/llvm-dev/2017-May/113525.html showed
this is beneficial for a wide range of cores.
As is to be expected, quite a few small adaptations are needed to the
regressions tests, as the difference in scheduling results in:
- Quite a few small instruction schedule differences.
- A few changes in register allocation decisions caused by different
instruction schedules.
- A few changes in IfConversion decisions, due to a difference in
instruction schedule and/or the estimated cost of a branch mispredict.
llvm-svn: 306514
That is pretty common for clang to produce code like
(shl %x, (and %amt, 31)). In this situation we can still perform
trunc (shl) into shl (trunc) conversion given the known value
range of shift amount.
Differential Revision: https://reviews.llvm.org/D34723
llvm-svn: 306499
Summary:
This is the llvm part of the initial implementation to support Windows ARM64 COFF format.
I will gradually add more functionality in subsequent patches.
Reviewers: ruiu, rnk, t.p.northover, compnerd
Reviewed By: ruiu, compnerd
Subscribers: aemerson, mgorny, javed.absar, llvm-commits, kristof.beyls
Differential Revision: https://reviews.llvm.org/D34705
llvm-svn: 306490
As noted in D34071, there are some IR optimization opportunities that could be
handled by normal IR passes if this expansion wasn't happening so late in CGP.
Regardless of that, it seems wasteful to knowingly produce suboptimal IR here,
so I'm proposing this change:
%s = sub i32 %x, %y
%r = icmp ne %s, 0
=>
%r = icmp ne %x, %y
Changing the predicate to 'eq' mimics what InstCombine would do, so that's just
an efficiency improvement if we decide this expansion should happen sooner.
The fact that the PowerPC backend doesn't eliminate the 'subf.' might be
something for PPC folks to investigate separately.
Differential Revision: https://reviews.llvm.org/D34416
llvm-svn: 306471
Without this check, COPY instructions can actually be one of the generic casts
in disguise. That's confusing and bad.
At some point during ISel this restriction has to be relaxed since the fully
selected instructions will usually use COPY for those purposes. Right now I
think it's possible that relaxation occurs during RegBankSelect (hence the
change there). I'm not convinced that's where it belongs long-term though.
llvm-svn: 306470
The overal size of the data section (including BSS)
is otherwise not included in the wasm binary.
Differential Revision: https://reviews.llvm.org/D34657
llvm-svn: 306459
The check to see if we can propagate the nsw flag used m_ConstantInt(uint64_t*&) which doesn't work with splat vectors and has a restriction that the bitwidth of the ConstantInt must be 64-bits are less.
This patch changes it to use m_APInt to remove both these issues
Differential Revision: https://reviews.llvm.org/D34699
llvm-svn: 306457
BlockAddress are only valid within their function context, which does not
interact well with CodeExtractor. Detect this case and prevent it.
Differential Revision: https://reviews.llvm.org/D33839
llvm-svn: 306448
Depending on the compare code that can be either an argument of
sext or negate of it. This helps to avoid v_cndmask_b64 instruction
for sext. A reversed value can be further simplified and folded into
its parent comparison if possible.
Differential Revision: https://reviews.llvm.org/D34545
llvm-svn: 306446
Apparently this replacement can really be substituting the
same as the original register. Avoid restarting the loop
when there's been no change in the register uses.
llvm-svn: 306441
SROA assumes alloca address space is 0, which causes assertion. This patch fixes that.
Differential Revision: https://reviews.llvm.org/D34104
llvm-svn: 306440
Also factored out function to check if a boolean is an already
deserialized value which does not require v_cndmask_b32 to be
loaded. Added binary logical operators to its check.
Differential Revision: https://reviews.llvm.org/D34500
llvm-svn: 306439
This canonicalization was suggested in D33172 as a way to make InstCombine behavior more uniform.
We have this transform for icmp+br, so unless there's some reason that icmp+select should be
treated differently, we should do the same thing here.
The benefit comes from increasing the chances of creating identical instructions. This is shown in
the tests in logical-select.ll (PR32791). InstCombine doesn't fold those directly, but EarlyCSE
can simplify the identical cmps, and then InstCombine can fold the selects together.
The possible regression for the tests in select.ll raises questions about poison/undef:
http://lists.llvm.org/pipermail/llvm-dev/2017-May/113261.html
...but that transform is just as likely to be triggered by this canonicalization as it is to be
missed, so we're just pointing out a commutation deficiency in the pattern matching:
https://reviews.llvm.org/rL228409
Differential Revision: https://reviews.llvm.org/D34242
llvm-svn: 306435
Not sure why this restriction existed, but it seems like we should support any size Constant here.
The particular pattern in the tests is not the only use of this matcher in the tree. There's one in CodeGenPrepare and one in InstSimplify as well.
Differential Revision: https://reviews.llvm.org/D34666
llvm-svn: 306417
•static latency
•number of uOps from which the instructions consists
•all ports used by the instruction
Reviewers:
RKSimon
zvi
aymanmus
m_zuckerman
Differential Revision: https://reviews.llvm.org/D33897
llvm-svn: 306414
Summary:
1. Instruction V_CVT_U32_F32 allow omod operand (see SIInstrInfo.td:1435). In fact this operand shouldn't be allowed here. This fix checks if SDWA pseudo instruction has OMod operand and then copy it.
2. There were several problems with support of VOPC instructions in SDWA peephole pass.
Reviewers: tstellar, arsenm, vpykhtin, airlied, kzhuravl
Subscribers: wdng, nhaehnle, yaxunl, dstuttard, tpr, sarnex, t-tye
Differential Revision: https://reviews.llvm.org/D34626
llvm-svn: 306413
This patch modifies the conditional compares pass so that it keeps successor
probabilities up-to-date after the conversion. Previously, successor
probabilities were being normalized to a uniform distribution, even though they
may have been heavily biased prior to the conversion (e.g., if one of the edges
was the back edge of a loop). This loss of information affected passes later in
the pipeline.
Differential Revision: https://reviews.llvm.org/D34109
llvm-svn: 306412
Add the instruction aliases for ds(r|l)l for the two operand alias
of ds(r|l)lv and the aliases ds(r|l)l with the three register operands.
llvm-svn: 306405
When SelectionDAG merges consecutive stores and loads in MergeConsecutiveStores, it does not set dereferenceable flag for a created load instruction. This results in an assertion failure if SelectionDAG commonizes this load instruction with other load instructions, as well as it may miss optimization opportunities.
This patch sat dereferenceable flag for the newly created load instruction if all the load instructions to be merged are dereferenceable.
Differential Revision: https://reviews.llvm.org/D34679
llvm-svn: 306404
[X86][AVX512] Improve lowering of AVX512 compare intrinsics (remove redundant shift left+right instructions).
AVX512 compare instructions return v*i1 types.
In cases where the number of elements in the returned value are less than 8, clang adds zeroes to get a mask of v8i1 type.
Later on it's replaced with CONCAT_VECTORS, which then is lowered to many DAG nodes including insert/extract element and shift right/left nodes.
The fact that AVX512 compare instructions put the result in a k register and zeroes all its upper bits allows us to remove the extra nodes simply by copying the result to the required register class.
When lowering, identify these cases and transform them into an INSERT_SUBVECTOR node (marked legal), then catch this pattern in instructions selection phase and transform it into one avx512 cmp instruction.
Differential Revision: https://reviews.llvm.org/D33188
llvm-svn: 306402
Summary:
After this patch, we finally have test cases that require multiple
instruction emission.
Depends on D33590
Reviewers: ab, qcolombet, t.p.northover, rovka, kristof.beyls
Subscribers: javed.absar, llvm-commits, igorb
Differential Revision: https://reviews.llvm.org/D33596
llvm-svn: 306388
* Mark as legal for (s32, i1, s32, s32)
* Map everything into GPRs
* Select to two instructions: a CMP of the condition against 0, to set
the flags, and a MOVCCr to select between the two inputs based on the
flags that we've just set
llvm-svn: 306382
This is based heavily on the work done ni D34285. I mostly wanted to do
test cleanup for the author to save them some time, but I had a really
hard time understanding why it was so hard to write better test cases
for these issues.
The problem is that because SROA does a second rewrite of the loads and
because we *don't* propagate !nonnull for non-pointer loads, we first
introduced invalid !nonnull metadata and then stripped it back off just
in time to avoid most ways of this PR manifesting. Moving to the more
careful utility only fixes this by changing the predicate to look at the
new load's type rather than the target type. However, that *does* fix
the bug, and the utility is much nicer including adding range metadata
to model the nonnull property after a conversion to an integer.
However, we have bigger problems because we don't actually propagate
*range* metadata, and the utility to do this extracted from instcombine
isn't really in good shape to do this currently. It *only* handles the
case of copying range metadata from an integer load to a pointer load.
It doesn't even handle the trivial cases of propagating from one integer
load to another when they are the same width! This utility will need to
be beefed up prior to using in this location to get the metadata to
fully survive.
And even then, we need to go and teach things to turn the range metadata
into an assume the way we do with nonnull so that when we *promote* an
integer we don't lose the information.
All of this will require a new test case that looks kind-of like
`preserve-nonnull.ll` does here but focuses on range metadata. It will
also likely require more testing because it needs to correctly handle
changes to the integer width, especially as SROA actively tries to
change the integer width!
Last but not least, I'm a little worried about hooking the range
metadata up here because the instcombine logic for converting from
a range metadata *to* a nonnull metadata node seems broken in the face
of non-zero address spaces where null is not mapped to the integer `0`.
So that probably needs to get fixed with test cases both in SROA and in
instcombine to cover it.
But this *does* extract the core PR fix from D34285 of preventing the
!nonnull metadata from being propagated in a broken state just long
enough to feed into promotion and crash value tracking.
On D34285 there is some discussion of zero-extend handling because it
isn't necessary. First, the new load size covers all of the non-undef
(ie, possibly initialized) bits. This may even extend past the original
alloca if loading those bits could produce valid data. The only way its
valid for us to zero-extend an integer load in SROA is if the original
code had a zero extend or those bits were undef. And we get to assume
things like undef *never* satifies nonnull, so non undef bits can
participate here. No need to special case the zero-extend handling, it
just falls out correctly.
The original credit goes to Ariel Ben-Yehuda! I'm mostly landing this to
save a few rounds of trivial edits fixing style issues and test case
formulation.
Differental Revision: D34285
llvm-svn: 306379
Summary:
With scalar stores, M0 is clobbered and therefore marked as implicitly
defined. However, it is also dead.
This fixes an assertion when the Greedy Register Allocator decides to
optimize a spill/restore pair away again (via tryHintsRecoloring).
Reviewers: arsenm
Subscribers: qcolombet, kzhuravl, wdng, yaxunl, dstuttard, tpr, t-tye, llvm-commits
Differential Revision: https://reviews.llvm.org/D33319
llvm-svn: 306375
Summary:
EraseInst didn't report that it made IR changes through MadeChange.
It is essential that changes to the IR are reported correctly,
since for example ReassociatePass::run() will indicate that all
analyses are preserved otherwise.
And the CGPassManager determines if the CallGraph is up-to-date
based on status from InstructionCombiningPass::runOnFunction().
Reviewers: craig.topper, rnk, davide
Reviewed By: rnk, davide
Subscribers: llvm-commits
Differential Revision: https://reviews.llvm.org/D34616
llvm-svn: 306368
PowerPC backend does not pass the current optimization level to SelectionDAGISel and so SelectionDAGISel works with the default optimization level regardless of the current optimization level.
This patch makes the PowerPC backend set the optimization level correctly.
Differential Revision: https://reviews.llvm.org/D34615
llvm-svn: 306367
Remove invalid shortcut in fixupKills(): A register needs to be marked
live even when we are not adding a kill flag. This is because a
partially live register must not get a kill flags, but it still needs to
be fully marked live when walking backwards.
llvm-svn: 306352
Summary:
vectorizer-maximize-bandwidth is generally useful in terms of performance. I've tested the impact of changing this to default on speccpu benchmarks on sandybridge machines. The result shows non-negative impact:
spec/2006/fp/C++/444.namd 26.84 -0.31%
spec/2006/fp/C++/447.dealII 46.19 +0.89%
spec/2006/fp/C++/450.soplex 42.92 -0.44%
spec/2006/fp/C++/453.povray 38.57 -2.25%
spec/2006/fp/C/433.milc 24.54 -0.76%
spec/2006/fp/C/470.lbm 41.08 +0.26%
spec/2006/fp/C/482.sphinx3 47.58 -0.99%
spec/2006/int/C++/471.omnetpp 22.06 +1.87%
spec/2006/int/C++/473.astar 22.65 -0.12%
spec/2006/int/C++/483.xalancbmk 33.69 +4.97%
spec/2006/int/C/400.perlbench 33.43 +1.70%
spec/2006/int/C/401.bzip2 23.02 -0.19%
spec/2006/int/C/403.gcc 32.57 -0.43%
spec/2006/int/C/429.mcf 40.35 +0.27%
spec/2006/int/C/445.gobmk 26.96 +0.06%
spec/2006/int/C/456.hmmer 24.4 +0.19%
spec/2006/int/C/458.sjeng 27.91 -0.08%
spec/2006/int/C/462.libquantum 57.47 -0.20%
spec/2006/int/C/464.h264ref 46.52 +1.35%
geometric mean +0.29%
The regression on 453.povray seems real, but is due to secondary effects as all hot functions are bit-identical with and without the flag.
I started this patch to consult upstream opinions on this. It will be greatly appreciated if the community can help test the performance impact of this change on other architectures so that we can decided if this should be target-dependent.
Reviewers: hfinkel, mkuper, davidxl, chandlerc
Reviewed By: chandlerc
Subscribers: rengolin, sanjoy, javed.absar, bjope, dorit, magabari, RKSimon, llvm-commits, mzolotukhin
Differential Revision: https://reviews.llvm.org/D33341
llvm-svn: 306336
Fixes bug 33597.
Use of substituteRegister in the tied operand case messes
up the register use iterator, causing some uses to be left
unprocessed.
llvm-svn: 306333
When we forward a stored value to a load and eliminate it entirely we need to
make sure the liveness of the register is maintained all the way to its use.
Previously we only cleared liveness on the store doing the forwarding, but
there could be other killing uses in between.
We already do the right thing when the load has to be converted into something
else, it was just this one path that skipped it.
llvm-svn: 306318
The recommit fixes three bugs: The first one is to use CurrentBlock instead of
PREInstr's Parent as param of performScalarPREInsertion because the Parent
of a clone instruction may be uninitialized. The second one is stop PRE when
CurrentBlock to its predecessor is a backedge and an operand of CurInst is
defined inside of CurrentBlock. The same value defined inside of loop in last
iteration can not be regarded as available. The third one is an out-of-bound
array access in a flipped if guard.
Right now scalarpre doesn't have phi-translate support, so it will miss some
simple pre opportunities. Like the following testcase, current scalarpre cannot
recognize the last "a * b" is fully redundent because a and b used by the last
"a * b" expr are both defined by phis.
long a[100], b[100], g1, g2, g3;
__attribute__((pure)) long goo();
void foo(long a, long b, long c, long d) {
g1 = a * b;
if (__builtin_expect(g2 > 3, 0)) {
a = c;
b = d;
g2 = a * b;
}
g3 = a * b; // fully redundant.
}
The patch adds phi-translate support in scalarpre. This is only a temporary
solution before the newpre based on newgvn is available.
llvm-svn: 306313
We sometimes need emergency spill slots for the register scavenger.
This may be the case when code needs to access a stack slot that
has an offset of 4096 or more relative to the stack pointer.
To make that determination, processFunctionBeforeFrameFinalized
currently simply checks the total stack frame size of the current
function. But this is not enough, since code may need to access
stack slots in the caller's stack frame as well, in particular
incoming arguments stored on the stack.
This commit fixes the problem by taking argument slots into account.
llvm-svn: 306305
The non-AVX-512 behavior was changed in r248266 to match N1778
(C bindings for IEEE-754 (2008)), which defined the four functions
to not raise the inexact exception ("rint" is still defined as raising
it).
Update the AVX-512 lowering of these functions to match that: it should
not be different.
llvm-svn: 306299
Convert vector increment or decrement to sub/add with an all-ones constant:
add X, <1, 1...> --> sub X, <-1, -1...>
sub X, <1, 1...> --> add X, <-1, -1...>
The all-ones vector constant can be materialized using a pcmpeq instruction that is
commonly recognized as an idiom (has no register dependency), so that's better than
loading a splat 1 constant.
AVX512 uses 'vpternlogd' for 512-bit vectors because there is apparently no better
way to produce 512 one-bits.
The general advantages of this lowering are:
1. pcmpeq has lower latency than a memop on every uarch I looked at in Agner's tables,
so in theory, this could be better for perf, but...
2. That seems unlikely to affect any OOO implementation, and I can't measure any real
perf difference from this transform on Haswell or Jaguar, but...
3. It doesn't look like it from the diffs, but this is an overall size win because we
eliminate 16 - 64 constant bytes in the case of a vector load. If we're broadcasting
a scalar load (which might itself be a bug), then we're replacing a scalar constant
load + broadcast with a single cheap op, so that should always be smaller/better too.
4. This makes the DAG/isel output more consistent - we use pcmpeq already for padd x, -1
and psub x, -1, so we should use that form for +1 too because we can. If there's some
reason to favor a constant load on some CPU, let's make the reverse transform for all
of these cases (either here in the DAG or in a later machine pass).
This should fix:
https://bugs.llvm.org/show_bug.cgi?id=33483
Differential Revision: https://reviews.llvm.org/D34336
llvm-svn: 306289
Csmith discovered that this function can be called with a zero argument,
in which case an assert for this triggered.
This patch also adds a guard before the other call to this function since
it was missing, although the test only covers the case where it was
discovered.
Reduced test case attached as CodeGen/SystemZ/int-cmp-54.ll.
Review: Ulrich Weigand
llvm-svn: 306287
This is a last fix for the corner case of PR32214. Actually this is not really corner case in general.
We should not do a loop rotation if we create an additional branch due to it.
Consider the case where we have a loop chain H, M, B, C , where
H is header with viable fallthrough from pre-header and exit from the loop
M - some middle block
B - backedge to Header but with exit from the loop also.
C - some cold block of the loop.
Let's H is determined as a best exit. If we do a loop rotation M, B, C, H we can introduce the extra branch.
Let's compute the change in number of branches:
+1 branch from pre-header to header
-1 branch from header to exit
+1 branch from header to middle block if there is such
-1 branch from cold bock to header if there is one
So if C is not a predecessor of H then we introduce extra branch.
This change actually prohibits rotation of the loop if both true
1) Best Exit has next element in chain as successor.
2) Last element in chain is not a predecessor of first element of chain.
Reviewers: iteratee, xur
Reviewed By: iteratee
Subscribers: llvm-commits
Differential Revision: https://reviews.llvm.org/D34271
llvm-svn: 306272
This should not be treated as a different version of
private_segment_buffer. These are distinct things with
different uses and register classes, and requires the
function argument info to have more context about the
function's type and environment.
Also add missing test coverage for the intrinsic, and
emit an error for HSA. This also encovers that the intrinsic
is broken unless there happen to be stack objects.
llvm-svn: 306264
This was reverted in r306252, but I already had the bug fixed and was
just trying to form a test case.
The original commit factored the logic for forming dedicated exits
inside of LoopSimplify into a helper that could be used elsewhere and
with an approach that required fewer intermediate data structures. See
that commit for full details including the change to the statistic, etc.
The code looked fine to me and my reviewers, but in fact didn't handle
indirectbr correctly -- it left the 'InLoopPredecessors' vector dirty.
If you have code that looks *just* right, you can end up leaking these
predecessors into a subsequent rewrite, and crash deep down when trying
to update PHI nodes for predecessors that don't exist.
I've added an assert that makes the bug much more obvious, and then
changed the code to reliably clear the vector so we don't get this bug
again in some other form as the code changes.
I've also added a test case that *does* manage to catch this while also
giving some nice positive coverage in the face of indirectbr.
The real code that found this came out of what I think is CPython's
interpreter loop, but any code with really "creative" interpreter loops
mixing indirectbr and other exit paths could manage to tickle the bug.
I was hard to reduce the original test case because in addition to
having a particular pattern of IR, the whole thing depends on the order
of the predecessors which is in turn depends on use list order. The test
case added here was designed so that in multiple different predecessor
orderings it should always end up going down the same path and tripping
the same bug. I hope. At least, it tripped it for me without
manipulating the use list order which is better than anything bugpoint
could do...
llvm-svn: 306257
I did some basic testing while looking for a bug in my recent change to
loop simplify and even though it didn't find the bug it seems like
a useful improvement anyways.
llvm-svn: 306256
http://rise4fun.com/Alive/i8Q
A narrow bitwise logic op is obviously better than math for value tracking,
and zext is better than sext. Typically, the 'not' will be folded into an
icmp predicate.
The IR difference would even survive through codegen for x86, so we would see
worse code:
https://godbolt.org/g/C14HMF
one_or_zero(int, int): # @one_or_zero(int, int)
xorl %eax, %eax
cmpl %esi, %edi
setle %al
retq
one_or_zero_alt(int, int): # @one_or_zero_alt(int, int)
xorl %ecx, %ecx
cmpl %esi, %edi
setg %cl
movl $1, %eax
subl %ecx, %eax
retq
llvm-svn: 306243
The compiler fails with assertion during legalization of SETCC for <3 x i8> operands.
The result is extended to <4 x i8> and then truncated <4 x i1>. It does not happen on AVX2, because the final result of SETCC is <4 x i32>.
Differential Revision: https://reviews.llvm.org/D34503
llvm-svn: 306242
Summary:
Support vector type G_EXTRACT selection. For now G_EXTRACT marked as legal for any type, so nothing to do in legalizer.
Split from https://reviews.llvm.org/D33665
Reviewers: qcolombet, t.p.northover, zvi, guyblank
Reviewed By: guyblank
Subscribers: guyblank, rovka, llvm-commits, kristof.beyls
Differential Revision: https://reviews.llvm.org/D33957
llvm-svn: 306240
The cost of an interleaved access was only implemented for AVX512. For other
X86 targets an overly conservative Base cost was returned, resulting in
avoiding vectorization where it is actually profitable to vectorize.
This patch starts to add costs for AVX2 for most prominent cases of
interleaved accesses (stride 3,4 chars, for now).
Note1: Improvements of up to ~4x were observed in some of EEMBC's rgb
workloads; There is also a known issue of 15-30% degradations on some of these
workloads, associated with an interleaved access followed by type
promotion/widening; the resulting shuffle sequence is currently inefficient and
will be improved by a series of patches that extend the X86InterleavedAccess pass
(such as D34601 and more to follow).
Note 2: The costs in this patch do not reflect port pressure penalties which can
be very dominant in the case of interleaved accesses since most of the shuffle
operations are restricted to a single port. Further tuning, that may incorporate
these considerations, will be done on top of the upcoming improved shuffle
sequences (that is, along with the abovementioned work to extend
X86InterleavedAccess pass).
Differential Revision: https://reviews.llvm.org/D34023
llvm-svn: 306238
When SelectionDAG expands memcpy (or memmove) call into a sequence of load and store instructions, it disregards dereferenceable flag even the source pointer is known to be dereferenceable.
This results in an assertion failure if SelectionDAG commonizes a load instruction generated for memcpy with another load instruction for the source pointer.
This patch makes SelectionDAG to set the dereferenceable flag for the load instructions properly to avoid the assertion failure.
Differential Revision: https://reviews.llvm.org/D34467
llvm-svn: 306209
Summary:
InstCombine replaces large allocas with small globals consts causing buffer overflows
on valid code, see PR33372.
This fix permits this optimization only if the global is dereference for alloca size.
Fixes PR33372
Reviewers: eugenis, majnemer, chandlerc
Subscribers: llvm-commits
Differential Revision: https://reviews.llvm.org/D34311
llvm-svn: 306194
processFixupValue is called on every relaxation iteration. applyFixup
is only called once at the very end. applyFixup is then the correct
place to do last minute changes and value checks.
While here, do proper range checks again for fixup_arm_thumb_bl. We
used to do it, but dropped because of thumb2. We now do it again, but
use the thumb2 range.
llvm-svn: 306177
After fixing (r306173) a failing test in the lld test suite (r306173),
reland r306095.
Original commit message:
[mips] Fix register positions in the aui/daui instructions
Swapped the position of the rt and rs register in the aui/daui
instructions for mips32r6 and mips64r6. With this change, the format of
the generated instructions complies with specifications and GCC.
Patch by Milos Stojanovic.
llvm-svn: 306174
We would return an error in getVaPtr if the RVA table being dumped was
the last data in the .rdata section. Avoid the issue by subtracting one
from the offset and adding it back to get an open interval again.
llvm-svn: 306171
Summary:
Without this patch some types have incorrect size and/or alignment
according to the MSP430 EABI.
Reviewers: asl, awygle
Reviewed By: asl
Subscribers: llvm-commits
Differential Revision: https://reviews.llvm.org/D34561
llvm-svn: 306159
This is useful when you want to look at a specific chunk of a
stream or look for discontinuities, and you need to know the
list of blocks occupied by a stream.
llvm-svn: 306150
This patch dumps the raw bytes of the pdb name map which contains
the mapping of stream name to stream index for the string table
and other reserved streams.
llvm-svn: 306148
Normally we can only make sense of the content of a PDB in terms
of streams and blocks, but in some cases it may be useful to dump
bytes at a specific absolute file offset. For example, if you
know that some interesting data is at a particular location and
you want to see some surrounding data.
llvm-svn: 306146
This patch contains a pass that transforms CBZ/CBNZ/TBZ/TBNZ instructions into a
conditional branch (Bcc), when the NZCV flags can be set for "free". This is
preferred on targets that have more flexibility when scheduling Bcc
instructions as compared to CBZ/CBNZ/TBZ/TBNZ (assuming all other variables are
equal). This can reduce register pressure and is also the default behavior for
GCC.
A few examples:
add w8, w0, w1 -> cmn w0, w1 ; CMN is an alias of ADDS.
cbz w8, .LBB_2 -> b.eq .LBB0_2 ; single def/use of w8 removed.
add w8, w0, w1 -> adds w8, w0, w1 ; w8 has multiple uses.
cbz w8, .LBB1_2 -> b.eq .LBB1_2
sub w8, w0, w1 -> subs w8, w0, w1 ; w8 has multiple uses.
tbz w8, #31, .LBB6_2 -> b.ge .LBB6_2
In looking at all current sub-target machine descriptions, this transformation
appears to be either positive or neutral.
Differential Revision: https://reviews.llvm.org/D34220.
llvm-svn: 306144
The goal here is to make it possible to display absolute
file offsets when dumping byets from an MSF. The problem is
that when dumping bytes from an MSF, often the bytes will
cross a block boundary and encounter a discontinuity. We
can't use the normal formatBinary() function for this because
this would just treat the sequence as entirely ascending, and
not account out-of-order blocks.
This patch adds a formatMsfData() function to our printer, and
then uses this function to improve the output of the -stream-data
command line option for dumping bytes from a particular stream.
Test coverage is also expanded to make sure to include all possible
scenarios of offsets, sizes, and crossing block boundaries.
llvm-svn: 306141
It was trying to do too many things. The basic lumping together of values for
legalization purposes is now handled by G_MERGE_VALUES. More complex things
involving gaps and odd sizes are handled by G_INSERT sequences.
llvm-svn: 306120
G_SEQUENCE is going away soon so as a first step the MachineIRBuilder needs to
be taught how to emulate it with alternatives. We use G_MERGE_VALUES where
possible, and a sequence of G_INSERTs if not.
llvm-svn: 306119
Summary: visitSwitchInst should not take INT_MAX when Cost is negative. Instead of INT_MAX , we also use a valid upperbound cost when overflow occurs in Cost.
Reviewers: hans, echristo, dmgreen
Reviewed By: dmgreen
Subscribers: mcrosier, javed.absar, llvm-commits, eraman
Differential Revision: https://reviews.llvm.org/D34436
llvm-svn: 306118
This reverts the use of TargetLowering::prepareVolatileOrAtomicLoad
introduced by r196905. Nothing in the semantics of the "volatile"
keyword or the definition of the z/Architecture actually requires
that volatile loads are preceded by a serialization operation, and
no other compiler on the platform actually implements this.
Since we've now seen a use case where this additional serialization
causes noticable performance degradation, this patch removes it.
The patch still leaves in the serialization before atomic loads,
which is now implemented directly in lowerATOMIC_LOAD. (This also
seems overkill, but that can be addressed separately.)
llvm-svn: 306117
The command-line params override the target setting in the file itself, so delete that.
Also, remove the cpu and arch because those don't matter and neither does the OS specification in the triple.
llvm-svn: 306109
The isBarrier/isTerminator flags have been removed from the SystemZ trap
instructions, so that tests do not fail with EXPENSIVE_CHECKS. This was just
an issue at -O0 and did not affect code output on benchmarks.
(Like Eli pointed out: "targets are split over whether they consider their
"trap" a terminator; x86, AArch64, and NVPTX don't, but ARM, MIPS, PPC, and
SystemZ do. We should probably try to be consistent here.". This is still the
case, although SystemZ has switched sides).
SystemZ now returns true in isMachineVerifierClean() :-)
These Generic tests have been modified so that they can be run with or without
EXPENSIVE_CHECKS: CodeGen/Generic/llc-start-stop.ll and
CodeGen/Generic/print-machineinstrs.ll
Review: Ulrich Weigand, Simon Pilgrim, Eli Friedman
https://bugs.llvm.org/show_bug.cgi?id=33047https://reviews.llvm.org/D34143
llvm-svn: 306106
Summary:
Many languages have a three way comparison idiom where comparing two values
produces not a boolean, but a tri-state value. Typical values (e.g. as used in
the lcmp/fcmp bytecodes from Java) are -1 for less than, 0 for equality, and +1
for greater than.
We actually do a great job already of converting three way comparisons into
binary comparisons when the result produced has one a single use. Unfortunately,
such values can have more than one use, and in that case, our existing
optimizations break down.
The patch adds a peephole which converts a three-way compare + test idiom into a
binary comparison on the original inputs. It focused on replacing the test on
the result of the three way compare and does nothing about removing the three
way compare itself. That's left to other optimizations (which do actually kick
in commonly.)
We currently recognize one idiom on signed integer compare. In the future, we
plan to recognize and simplify other comparison idioms on
other signed/unsigned datatypes such as floats, vectors etc.
This is a resurrection of Philip Reames' original patch:
https://reviews.llvm.org/D19452
Reviewers: majnemer, apilipenko, reames, sanjoy, mkazantsev
Reviewed by: mkazantsev
Subscribers: llvm-commits
Differential Revision: https://reviews.llvm.org/D34278
llvm-svn: 306100
ELF/mips-plt-r6.s in lld-test is failing. Reverting the change.
Original commit message:
[mips] Fix register positions in the aui/daui instructions
Swapped the position of the rt and rs register in the aut/daui
instructions for mips32r6 and mips64r6. With this change, the format of
the generated instructions complies with specifications and GCC.
Patch by Milos Stojanovic.
llvm-svn: 306099
Swapped the position of the rt and rs register in the aut/daui instructions
for mips32r6 and mips64r6. With this change, the format of the generated
instructions complies with specifications and GCC.
Patch by Milos Stojanovic.
Differential Revision: https://reviews.llvm.org/D33988
llvm-svn: 306095
Before this change, it was always the first element of a vector that got splatted since the lower 6 bits of vshf.d $wd were always zero for little endian.
Additionally, masking has been performed for vshf via which splat.d is created.
Vshf has a property where if its first operand's elements have either bit 6 or 7 set, destination element is set to zero.
Initially masked with 63 to avoid this property, which would result in generation of and.v + vshf.d in all cases.
Masking with one results in generating a single splati.d instruction when possible.
Differential Revision: https://reviews.llvm.org/D32216
llvm-svn: 306090
X86_64 COFF only has support for 32 bit pcrel relocations. Produce an
error on all others.
Note that gnu as has extended the relocation values to support
this. It is not clear if we should support the gnu extension.
llvm-svn: 306082
I want to use the same logic as LoopSimplify to form dedicated exits in
another pass (SimpleLoopUnswitch) so I wanted to factor it out here.
I also noticed that there is a pretty significantly more efficient way
to implement this than the way the code in LoopSimplify worked. We don't
need to actually retain the set of unique exit blocks, we can just
rewrite them as we find them and use only a set to deduplicate.
This did require changing one part of LoopSimplify to not re-use the
unique set of exits, but it only used it to check that there was
a single unique exit. That part of the code is about to walk the exiting
blocks anyways, so it seemed better to rewrite it to use those exiting
blocks to compute this property on-demand.
I also had to ditch a statistic, but it doesn't seem terribly valuable.
Differential Revision: https://reviews.llvm.org/D34049
llvm-svn: 306081
Summary: LVI can reason about an AND of icmps on the true dest of a branch. I believe we can do similar for the false dest of ORs. This allows us to get the same answer for the demorganed versions of some of the AND test cases as you can see.
Reviewers: anna, reames
Reviewed By: reames
Subscribers: llvm-commits
Differential Revision: https://reviews.llvm.org/D34431
llvm-svn: 306076
This is very similar to the transform in:
https://reviews.llvm.org/rL306040
...but in this case, we use cmp X, 1 to set the carry bit as needed.
Again, we can show that all of these are logically equivalent (although
InstCombine currently canonicalizes to a form not seen here), and if
we believe IACA, then this is the smallest/fastest code. Eg, with SNB:
| Num Of | Ports pressure in cycles | |
| Uops | 0 - DV | 1 | 2 - D | 3 - D | 4 | 5 | |
---------------------------------------------------------------------
| 1 | 1.0 | | | | | | | cmp edi, 0x1
| 2 | | 1.0 | | | | 1.0 | CP | sbb eax, eax
The larger motivation is to clean up all select-of-constants combining/lowering
because we're missing some common cases.
llvm-svn: 306072
Also document the attribute, since "probe-stack" already is.
Reviewed By: majnemer
Differential Revision: https://reviews.llvm.org/D34528
llvm-svn: 306069
For whatever reason, when processing
.globl foo
foo:
.data
bar:
.long foo-bar
llvm-mc creates a relocation with the section:
0x0 IMAGE_REL_I386_REL32 .text
This is different than when the relocation is relative from the
beginning. For example, a file with
call foo
produces
0x0 IMAGE_REL_I386_REL32 foo
I would like to refactor the logic for converting "foo - ." into a
relative relocation so that it is shared with ELF. This is the first
step and just changes the coff implementation to match what ELF (and
COFF in the case of calls) does.
llvm-svn: 306063
The feeder instruction will be moved to right before the compare, so
the updating code should not be looking for kills past the compare.
llvm-svn: 306059
There's nothing incorrect about emitting such relocations against
symbols defined in other objects. The code in EmitCOFFSec* was missing
the visitUsedExpr part of MCStreamer::EmitValueImpl, so these symbols
were not being registered with the object file assembler.
This will be used to make reduced test cases for LLD.
llvm-svn: 306057
This idea originally came about when I was doing some deep
investigation of why certain bytes in a PDB that we round-tripped
differed from their original bytes in the source PDB. I found
myself having to hack up the code in many places to dump the
bytes of this substream, or that record. It would be nice if
we could just do this for every possible stream, substream,
debug chunk type, etc.
It doesn't make sense to put this under dump because there's just
so many options that would detract from the more common use case
of just dumping deserialized records. So making a new subcommand
seems like the most logical course of action. In doing so, we
already have two command line options that are suitable for this
new subcommand, so start out by moving them there.
llvm-svn: 306056
Now you run llvm-pdbutil dump <options>. This is a followup
after having renamed the tool, whereas before raw was obviously
just the style of dumping, whereas now "dump" is the action to
perform with the "util".
llvm-svn: 306055
Summary:
Currently, we incorrectly update exit blocks of loops when there are multiple
edges from a single exiting block to the exit block. This can happen when we
have switches as the terminator of the exiting blocks.
The fix here is to correctly update the phi nodes in the exit block, and remove
all incoming values *except* for one which is from the preheader.
Note: Currently, this error can manifest only while deleting non-executed loops. However, it
is possible to trigger this error in invariant loops, once we enhance the logic
around the exit conditions for the loop check.
Reviewers: chandlerc, dberlin, sanjoy, efriedma
Reviewed by: efriedma
Subscribers: mzolotukhin, llvm-commits
Differential Revision: https://reviews.llvm.org/D34516
llvm-svn: 306048
Summary:
These intrinsics aren't used by clang and haven't been for a while.
There's some really terrible codegen in the 32-bit target for avx512bw due to i64 not being legal. But as I said these intrinsics aren't used by clang even before this patch so this codegen reflects our clang behavior today.
Reviewers: spatel, RKSimon, zvi, igorb
Reviewed By: RKSimon
Subscribers: llvm-commits
Differential Revision: https://reviews.llvm.org/D34389
llvm-svn: 306047
Our handling of select-of-constants is lumpy in IR (https://reviews.llvm.org/D24480),
lumpy in DAGCombiner, and lumpy in X86ISelLowering. That's why we only had the 'sbb'
codegen in 1 out of the 4 tests. This is a step towards smoothing that out.
First, show that all of these IR forms are equivalent:
http://rise4fun.com/Alive/mx
Second, show that the 'sbb' version is faster/smaller. IACA output for SandyBridge
(later Intel and AMD chips are similar based on Agner's tables):
This is the "obvious" x86 codegen (what gcc appears to produce currently):
| Num Of | Ports pressure in cycles | |
| Uops | 0 - DV | 1 | 2 - D | 3 - D | 4 | 5 | |
---------------------------------------------------------------------
| 1* | | | | | | | | xor eax, eax
| 1 | 1.0 | | | | | | CP | test edi, edi
| 1 | | | | | | 1.0 | CP | setnz al
| 1 | | 1.0 | | | | | CP | neg eax
This is the adc version:
| 1* | | | | | | | | xor eax, eax
| 1 | 1.0 | | | | | | CP | cmp edi, 0x1
| 2 | | 1.0 | | | | 1.0 | CP | adc eax, 0xffffffff
And this is sbb:
| 1 | 1.0 | | | | | | | neg edi
| 2 | | 1.0 | | | | 1.0 | CP | sbb eax, eax
If IACA is trustworthy, then sbb became a single uop in Broadwell, so this will be
clearly better than the alternatives going forward.
llvm-svn: 306040
Intrinsic already existed for llvm.SI.tbuffer.store
Needed tbuffer.load and also re-implementing the intrinsic as llvm.amdgcn.tbuffer.*
Added CodeGen tests for the 2 new variants added.
Left the original llvm.SI.tbuffer.store implementation to avoid issues with existing code
Subscribers: arsenm, kzhuravl, wdng, nhaehnle, yaxunl, tony-tye, tpr
Differential Revision: https://reviews.llvm.org/D30687
llvm-svn: 306031
Summary:
InstCombine likes to turn (icmp eq (and X, C1), 0) into (icmp slt (trunc (X)), 0) sometimes. This breaks foldSelectICmpAndOr's ability to recognize (select (icmp eq (and X, C1), 0), Y, (or Y, C2))->(or (shl (and X, C1), C3), y).
This patch tries to recover this. I had to flip around some of the early out checks so that I could create a new And instruction during the compare processing without it possibly never getting used.
Reviewers: spatel, majnemer, davide
Reviewed By: spatel
Subscribers: llvm-commits
Differential Revision: https://reviews.llvm.org/D34184
llvm-svn: 306029
If the components of the and/or had multiple uses, this transform created an additional instruction.
This patch makes sure we remove one of the components.
Differential Revision: https://reviews.llvm.org/D34498
llvm-svn: 306027
There are 2 parts to this patch made simultaneously to avoid a regression.
We're reversing the canonicalization that moves bitwise vector ops before bitcasts.
We're moving bitwise vector ops *after* bitcasts instead. That's the 1st and 3rd hunks
of the patch. The motivation is that there's only one fold that currently depends on
the existing canonicalization (see next), but there are many folds that would
automatically benefit from the new canonicalization.
PR33138 ( https://bugs.llvm.org/show_bug.cgi?id=33138 ) shows why/how we have these
patterns in IR.
There's an or(and,andn) pattern that requires an adjustment in order to continue matching
to 'select' because the bitcast changes position. This match is unfortunately complicated
because it requires 4 logic ops with optional bitcast and sext ops.
Test diffs:
1. The bitcast.ll and bitcast-bigendian.ll changes show the most basic difference -
bitcast comes before logic.
2. There are also tests with no diffs in bitcast.ll that verify that we're still doing
folds that were enabled by the previous canonicalization.
3. icmp-xor-signbit.ll shows the payoff. We don't need to adjust existing icmp patterns
to look through bitcasts.
4. logical-select.ll contains several tests for the or(and,andn) --> select fold to
verify that we are still handling those cases. The lone diff shows the movement of
the bitcast from the new canonicalization rule.
Differential Revision: https://reviews.llvm.org/D33517
llvm-svn: 306011
Summary:
The ARM ELF ABI requires the linker to do interworking for wide
conditional branches from Thumb code to ARM code.
That was pointed out by @peter.smith in the comments for D33436.
Reviewers: rafael, peter.smith, echristo
Reviewed By: peter.smith
Subscribers: aemerson, javed.absar, kristof.beyls, llvm-commits, peter.smith
Differential Revision: https://reviews.llvm.org/D34447
llvm-svn: 306009
This patch allows $AT to be used as a register name in assembly files.
Currently only $at is recognized as a valid register name.
Patch by Stanislav Ocovaj.
Differential Revision: https://reviews.llvm.org/D34348
llvm-svn: 306007
Summary:
Despite that this instructions are listed in VOP2, they are treated as VOP3 in specs. They should not support SDWA.
There are no real instructions for them, but there are pseudo instructions.
Reviewers: arsenm, vpykhtin, cfang
Subscribers: kzhuravl, wdng, nhaehnle, yaxunl, dstuttard, tpr, t-tye
Differential Revision: https://reviews.llvm.org/D34403
llvm-svn: 305999
This has been deprecated since ARMARM v7-AR, release C.b, published back
in 2012.
This also removes test/CodeGen/Thumb2/ifcvt-neon.ll that originally was
introduced to check that conditionalization of Neon instructions did
happen when generating Thumb2. However, the test had evolved and was no
longer testing that. Rather than trying to adapt that test, this commit
introduces test/CodeGen/Thumb2/ifcvt-neon-deprecated.mir, since we can
now use the MIR framework to write nicer/more maintainable tests.
llvm-svn: 305998
Rather than creating a separate ".rdata" section distinct from the
customary ".rodata" in ELF, ".rdata" switches to the ".rodata" section.
This patch relands r305949 and r305950 with the correct commit message
and addresses nit raised during review.
Patch By: John Baldwin!
Differential Revision: https://reviews.llvm.org/D34452
llvm-svn: 305995
Summary:
This patch adds a macro fusion using CodeGen/MacroFusion.cpp to pair AES
instructions back to back and adds FeatureFuseAES to enable the feature.
Reviewers: evandro, javed.absar, rengolin, t.p.northover
Reviewed By: javed.absar
Subscribers: aemerson, mgorny, kristof.beyls, llvm-commits
Differential Revision: https://reviews.llvm.org/D34142
llvm-svn: 305988
Masked gather for vector length 2 is lowered incorrectly for element type i32.
The type <2 x i32> was automatically extended to <2 x i64> and we generated VPGATHERQQ instead of VPGATHERQD.
The type <2 x float> is extended to <4 x float>, so there is no bug for this type, but the sequence may be more optimal.
In this patch I'm fixing <2 x i32>bug and optimizing <2 x float> sequence for GATHERs only. The same fix should be done for Scatters as well.
Differential revision: https://reviews.llvm.org/D34343
llvm-svn: 305987
Summary:
Added support based on merged SDWA pseudo instructions. Now peephole allow one scalar operand, omod and clamp modifiers.
Added several subtarget features for GFX9 SDWA.
This diff also contains changes from D34026.
Depends D34026
Reviewers: vpykhtin, rampitec, arsenm
Subscribers: kzhuravl, wdng, nhaehnle, yaxunl, dstuttard, tpr, t-tye
Differential Revision: https://reviews.llvm.org/D34241
llvm-svn: 305986
This includes the safe SEH tables and the control flow guard function
table. LLD will emit the guard table soon, and I need a tool that dumps
them for testing.
llvm-svn: 305979
Summary:
This fixes a bug where we always treat APSInts in Codeview as
signed when writing them to YAML. One symptom of this problem is that
llvm-pdbdump raw would show Enumerator Values that differ between the
original PDB and a PDB that has been round-tripped through YAML.
Reviewers: zturner
Reviewed By: zturner
Subscribers: llvm-commits, fhahn
Differential Revision: https://reviews.llvm.org/D34013
llvm-svn: 305965
If one of the arguments of adde/sube is zero we can fold another
add/sub into it.
Differential Revision: https://reviews.llvm.org/D34374
llvm-svn: 305964
This simplification allows to avoid generating v_cndmask_b32
to serialize condition code between compare and use.
Differential Revision: https://reviews.llvm.org/D34300
llvm-svn: 305962
Summary:
vectorizer-maximize-bandwidth is generally useful in terms of performance. I've tested the impact of changing this to default on speccpu benchmarks on sandybridge machines. The result shows non-negative impact:
spec/2006/fp/C++/444.namd 26.84 -0.31%
spec/2006/fp/C++/447.dealII 46.19 +0.89%
spec/2006/fp/C++/450.soplex 42.92 -0.44%
spec/2006/fp/C++/453.povray 38.57 -2.25%
spec/2006/fp/C/433.milc 24.54 -0.76%
spec/2006/fp/C/470.lbm 41.08 +0.26%
spec/2006/fp/C/482.sphinx3 47.58 -0.99%
spec/2006/int/C++/471.omnetpp 22.06 +1.87%
spec/2006/int/C++/473.astar 22.65 -0.12%
spec/2006/int/C++/483.xalancbmk 33.69 +4.97%
spec/2006/int/C/400.perlbench 33.43 +1.70%
spec/2006/int/C/401.bzip2 23.02 -0.19%
spec/2006/int/C/403.gcc 32.57 -0.43%
spec/2006/int/C/429.mcf 40.35 +0.27%
spec/2006/int/C/445.gobmk 26.96 +0.06%
spec/2006/int/C/456.hmmer 24.4 +0.19%
spec/2006/int/C/458.sjeng 27.91 -0.08%
spec/2006/int/C/462.libquantum 57.47 -0.20%
spec/2006/int/C/464.h264ref 46.52 +1.35%
geometric mean +0.29%
The regression on 453.povray seems real, but is due to secondary effects as all hot functions are bit-identical with and without the flag.
I started this patch to consult upstream opinions on this. It will be greatly appreciated if the community can help test the performance impact of this change on other architectures so that we can decided if this should be target-dependent.
Reviewers: hfinkel, mkuper, davidxl, chandlerc
Reviewed By: chandlerc
Subscribers: rengolin, sanjoy, javed.absar, bjope, dorit, magabari, RKSimon, llvm-commits, mzolotukhin
Differential Revision: https://reviews.llvm.org/D33341
llvm-svn: 305960
Michael Zuckerman