The isel patterns for these use a bitcast and load/store, but
DAG combine should have canonicalized those away.
For the purposes of the memory folding table these opcodes can be
replaced by the MOVSSrm_alt/MOVSDrm_alt and MOVSSmr/MOVSDmr opcodes.
llvm-svn: 363644
Rename the old versions that use FR32/FR64 to MOVSSrm_alt/MOVSDrm_alt.
Use the new versions in patterns that previously used a COPY_TO_REGCLASS
to VR128. These patterns expect the upper bits to be zero. The
current set up appears to work, but I'm not sure we should be
enforcing upper bits being zero through a COPY_TO_REGCLASS.
I wanted to flip the arrangement and use a COPY_TO_REGCLASS to
FR32/FR64 for the patterns that need an f32/f64 result, but that
complicated fastisel and globalisel.
I've been doing some experiments with reducing some isel patterns
and ended up in a situation where I had a
(SUBREG_TO_REG (COPY_TO_RECLASS (VMOVSSrm), VR128)) and our
post-isel peephole was unable to avoid using an instruction for
the SUBREG_TO_REG due to the COPY_TO_REGCLASS. Having a VR128
instruction removes the COPY_TO_REGCLASS that was breaking this.
llvm-svn: 363643
This was ignoring the flag on fneg, and using the source instruction's
flags. Also fixes tests missing from r358702.
Note the expansion itself isn't correct without nnan, but that should
be fixed separately.
llvm-svn: 363637
The goal is to improve hwasan's error reporting for stack use-after-return by
recording enough information to allow the specific variable that was accessed
to be identified based on the pointer's tag. Currently we record the PC and
lower bits of SP for each stack frame we create (which will eventually be
enough to derive the base tag used by the stack frame) but that's not enough
to determine the specific tag for each variable, which is the stack frame's
base tag XOR a value (the "tag offset") that is unique for each variable in
a function.
In IR, the tag offset is most naturally represented as part of a location
expression on the llvm.dbg.declare instruction. However, the presence of the
tag offset in the variable's actual location expression is likely to confuse
debuggers which won't know about tag offsets, and moreover the tag offset
is not required for a debugger to determine the location of the variable on
the stack, so at the DWARF level it is represented as an attribute so that
it will be ignored by debuggers that don't know about it.
Differential Revision: https://reviews.llvm.org/D63119
llvm-svn: 363635
Inter-block localization is the same as what currently happens, except now it
only runs on the entry block because that's where the problematic constants with
long live ranges come from.
The second phase is a new intra-block localization phase which attempts to
re-sink the already localized instructions further right before one of the
multiple uses.
One additional change is to also localize G_GLOBAL_VALUE as they're constants
too. However, on some targets like arm64 it takes multiple instructions to
materialize the value, so some additional heuristics with a TTI hook have been
introduced attempt to prevent code size regressions when localizing these.
Overall, these changes improve CTMark code size on arm64 by 1.2%.
Full code size results:
Program baseline new diff
------------------------------------------------------------------------------
test-suite...-typeset/consumer-typeset.test 1249984 1217216 -2.6%
test-suite...:: CTMark/ClamAV/clamscan.test 1264928 1232152 -2.6%
test-suite :: CTMark/SPASS/SPASS.test 1394092 1361316 -2.4%
test-suite...Mark/mafft/pairlocalalign.test 731320 714928 -2.2%
test-suite :: CTMark/lencod/lencod.test 1340592 1324200 -1.2%
test-suite :: CTMark/kimwitu++/kc.test 3853512 3820420 -0.9%
test-suite :: CTMark/Bullet/bullet.test 3406036 3389652 -0.5%
test-suite...ark/tramp3d-v4/tramp3d-v4.test 8017000 8016992 -0.0%
test-suite...TMark/7zip/7zip-benchmark.test 2856588 2856588 0.0%
test-suite...:: CTMark/sqlite3/sqlite3.test 765704 765704 0.0%
Geomean difference -1.2%
Differential Revision: https://reviews.llvm.org/D63303
llvm-svn: 363632
Summary: This case is related to D63405 in that we need to be propagating FMF on negates.
Reviewers: volkan, spatel, arsenm
Reviewed By: arsenm
Subscribers: wdng, javed.absar
Differential Revision: https://reviews.llvm.org/D63458
llvm-svn: 363631
Summary:
Change the way we deal with iterator invalidation in the extload combines as it
was still possible to neglect to visit a use. Even worse, it happened in the
in-tree test cases and the checks weren't good enough to detect it.
We now take a cheap copy of the use list before iterating over it. This
prevents iterator invalidation from occurring and has the nice side effect
of making the existing schedule-for-erase/schedule-for-insert mechanism
moot.
Reviewers: aditya_nandakumar
Reviewed By: aditya_nandakumar
Subscribers: rovka, kristof.beyls, javed.absar, volkan, Petar.Avramovic, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D61813
llvm-svn: 363616
AMDGPUPropagateAttributes will not work on function bitcatsts,
so move AMDGPUFixFunctionBitcasts before it.
Differential Revision: https://reviews.llvm.org/D63455
llvm-svn: 363614
Summary:
The purpose of the padding is to guard against stale code being
fetched into the instruction cache by the lowest level prefetching.
We're generating relocatable ELF here, and so the padding should
arguably be added by the linker. This is in fact what Mesa does.
This also fixes multi-part shaders for Mesa.
Change-Id: I6bfede58f20e9f337762ccf39ef9e0e263e69e82
Reviewers: arsenm, rampitec, t-tye
Subscribers: kzhuravl, jvesely, wdng, yaxunl, dstuttard, tpr, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D63427
llvm-svn: 363602
Summary:
This is related to the changes to the groupstaticsize intrinsic in
D61494 which would otherwise make the related tests in these files
fail or much less useful.
Note that for some reason, SOPK generation is less effective in the
amdhsa OS, which is why I chose PAL. I haven't investigated this
deeper.
Change-Id: I6bb99569338f7a433c28b4c9eb1e3e036b00d166
Reviewers: arsenm
Subscribers: kzhuravl, jvesely, wdng, yaxunl, dstuttard, tpr, t-tye, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D63392
llvm-svn: 363600
Basically porting over the behaviour in AArch64ISelLowering to GISel. See
emitComparison for reference.
When we have something like this:
```
lhs = G_SUB 0, y
...
G_ICMP lhs, rhs
```
We can fold away the G_SUB and produce a cmn instead, given that we produce
the same value in NZCV.
Add a test showing that the transformation works, and also showing that we
don't perform the transformation when it's unsafe.
Also factor out the CSet emission into emitCSetForICMP.
Differential Revision: https://reviews.llvm.org/D63163
llvm-svn: 363596
If a XMM non-temporal store has less than natural alignment, scalarize the vector - with SSE4A we can stay on the vector and use MOVNTSD(f64), else we must move to GPRs and use MOVNTI(i32/i64).
llvm-svn: 363592
The pass works in two modes:
Mode 1: Just set attributes starting from kernels. This can work at
the very beginning of opt and llc pipeline, but cannot clone functions
because it must be a function pass.
Mode 2: Actually clone functions for new attributes. This can only work
after all function passes in the opt pipeline because it has to be a
module pass.
Differential Revision: https://reviews.llvm.org/D63208
llvm-svn: 363586
If a YMM/ZMM non-temporal store has less than natural alignment, split the vector - either they will be satisfactorily aligned or will continue to be split until they are XMMs - at which point the legalizer will scalarize it.
llvm-svn: 363582
A target intrinsic may be defined as possibly reading memory, but the
call site may have additional knowledge that it doesn't read
memory. The intrinsic lowering will expect the pessimistic assumption
of the intrinsic definition, so the chain should still be used.
I fixed the same bug in SelectionDAG in r287593.
llvm-svn: 363580
I keep using the wrong instruction when manually writing tests. This
really needs to check the number of operands, but I don't see an easy
way to do that right now.
llvm-svn: 363579
For loads, pre-SSE41 we can't perform NT loads at all, and after that we can only perform vector aligned loads, so if the alignment is less than for a xmm we'll just end up using the regular unaligned vector loads anyway.
First step towards fixing PR42026 - the next step for stores will be to use SSE4A movntsd where possible and to avoid the stack spill on SSE2 targets.
Differential Revision: https://reviews.llvm.org/D63246
llvm-svn: 363564
Some GEPs were not being split, presumably because that split would just be
undone by the DAGCombiner. Not performing those splits can prevent important
optimizations, such as preventing the element indices / member offsets from
being (partially) folded into load/store instruction immediates. This patch:
- Makes the splits also occur in the cases where the base address and the GEP
are in the same BB.
- Ensures that the DAGCombiner doesn't reassociate them back again.
Differential Revision: https://reviews.llvm.org/D60294
llvm-svn: 363544
This is already done in DAGCombiner::visitINSERT_SUBVECTOR, but this helps a number of shuffles across different vector widths recognise when they come from the same source.
llvm-svn: 363542
Second functional change following on from rL362687. Pass the
NoWrapFlags from the MulExpr to InsertBinop when we're generating a
shl or mul.
Differential Revision: https://reviews.llvm.org/D61934
llvm-svn: 363540
This patch changes MIR stack-id from an integer to an enum,
and adds printing/parsing support for this in MIR files. The default
stack-id '0' is now renamed to 'default'.
This should make MIR tests that have stack objects with different stack-ids
more descriptive. It also clarifies code operating on StackID.
Reviewers: arsenm, thegameg, qcolombet
Reviewed By: arsenm
Differential Revision: https://reviews.llvm.org/D60137
llvm-svn: 363533
Third time's the charm.
This was reverted in r363220 due to being suspected of an internal benchmark
regression and a test failure, none of which turned out to be caused by this.
llvm-svn: 363529
Summary:
SPE passes doubles the same as soft-float, in register pairs as i32
types. This is all handled by the target-independent layer. However,
this is not optimal when splitting or reforming the doubles, as it
pushes to the stack and loads from, on either side.
For instance, to pass a double argument to a function, assuming the
double value is in r5, the sequence currently looks like this:
evstdd 5, X(1)
lwz 3, X(1)
lwz 4, X+4(1)
Likewise, to form a double into r5 from args in r3 and r4:
stw 3, X(1)
stw 4, X+4(1)
evldd 5, X(1)
This optimizes the fence to use SPE instructions. Now, to pass a double
to a function:
mr 4, 5
evmergehi 3, 5, 5
And to form a double into r5 from args in r3 and r4:
evmergelo 5, 3, 4
This is comparable to the way that gcc generates the double splits.
This also fixes a bug with expanding builtins to libcalls, where the
LowerCallTo() code path was generating intermediate illegal type nodes.
Reviewers: nemanjai, hfinkel, joerg
Subscribers: kbarton, jfb, jsji, llvm-commits
Differential Revision: https://reviews.llvm.org/D54583
llvm-svn: 363526
Summary:
Instead of encoding a high-word of 0 using a fake TargetGlobalAddress,
just use a literal target constant. This simplifies some subsequent changes.
The generated assembly is now more explicit about the kind of relocation
that is to be used.
Change-Id: I066835202d23b5941fa7a358eb4b89e9b71ab6f8
Reviewers: arsenm, rampitec
Subscribers: kzhuravl, jvesely, wdng, yaxunl, dstuttard, tpr, t-tye, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D61491
llvm-svn: 363516
This is based on the example/discussion in PR37428:
https://bugs.llvm.org/show_bug.cgi?id=37428
Proper vector shift instructions don't appear until AVX2, so we may generate several
extra instructions within a loop trying to compensate for that. It's difficult to
recover from that shift expansion later than this, so use the existing TLI hook and
splat analysis to enable better codegen.
This extends CGP functionality introduced with:
rL201655
Differential Revision: https://reviews.llvm.org/D63233
llvm-svn: 363511
This is similar logic/motivation to the select splitting in D62969.
In D63233, the pattern changes so that we no longer have an extract_subvector of vselect,
but the operands of the select are still being concatenated.
The closest case is represented in either the first or last test diffs here - we have an
extra instruction, but we converted 3-4 ymm instructions into 4-5 xmm instructions.
I think that's the right trade-off for most AVX1 targets.
In the example based on PR37428:
https://bugs.llvm.org/show_bug.cgi?id=37428
...this makes the loop about 30% faster (tested on Haswell by compiling with -mavx).
Differential Revision: https://reviews.llvm.org/D63364
llvm-svn: 363508
Summary:
If the nested loop is an innermost loop, prefer to a 32-byte alignment, so that
we can decrease cache misses and branch-prediction misses. Actual alignment of
the loop will depend on the hotness check and other logic in alignBlocks.
The old code will only align hot loop to 32 bytes when the LoopSize larger than
16 bytes and smaller than 32 bytes, this patch will align the innermost hot loop
to 32 bytes not only for the hot loop whose size is 16~32 bytes.
Reviewed By: steven.zhang, jsji
Differential Revision: https://reviews.llvm.org/D61228
llvm-svn: 363495
This reverts rL363474. -debug-only=isel was added to some tests that
don't specify `REQUIRES: asserts`. This causes failures on
-DLLVM_ENABLE_ASSERTIONS=off builds.
I chose to revert instead of fixing the tests because I'm not sure
whether we should add `REQUIRES: asserts` to more tests.
llvm-svn: 363482
