Some more refactoring, like registering the IT Block pass, less cryptic
variable names, and some simplification of loops.
Differential Revision: https://reviews.llvm.org/D63419
llvm-svn: 363666
Recommit of D32530 with a few small changes:
- Stopped recursively walking through aggregates in
the verifier, so that we don't impose too much
overhead on large modules under LTO (see PR42210).
- Changed tests to match; the errors are slightly
different since they only report the array or
struct that actually contains a scalable vector,
rather than all aggregates which contain one in
a nested member.
- Corrected an older comment
Reviewers: thakis, rengolin, sdesmalen
Reviewed By: sdesmalen
Differential Revision: https://reviews.llvm.org/D63321
llvm-svn: 363658
First step toward addressing the vector-reduce-mul-widen.ll regression in D63281 - we should replace ANY_EXTEND/ANY_EXTEND_VECTOR_INREG in X86ISelDAGToDAG to avoid having to add duplicate patterns when treating any extensions as legal.
In future patches this will also allow us to keep any extension nodes around a lot longer in the DAG, which should mean that we can keep better track of undef elements that otherwise become zeros that we think we have to keep......
Differential Revision: https://reviews.llvm.org/D63326
llvm-svn: 363655
Using the new SwitchInstProfUpdateWrapper this patch
simplifies 3 places of prof branch_weights handling.
Differential Revision: https://reviews.llvm.org/D62123
llvm-svn: 363652
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
Summary:
All the GlobalISel passes are initialized when the target calls
initializeGlobalISel(), so we don't need to call the initializers
from the pass constructors.
Reviewers: qcolombet, t.p.northover, paquette, dsanders, aemerson, aditya_nandakumar
Reviewed By: aemerson
Subscribers: rovka, kristof.beyls, hiraditya, volkan, Petar.Avramovic, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D63235
llvm-svn: 363642
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
This saves roughly 32 bytes of instructions per function with stack objects
and causes us to preserve enough information that we can recover the original
tags of all stack variables.
Now that stack tags are deterministic, we no longer need to pass
-hwasan-generate-tags-with-calls during check-hwasan. This also means that
the new stack tag generation mechanism is exercised by check-hwasan.
Differential Revision: https://reviews.llvm.org/D63360
llvm-svn: 363636
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
This is part of the approved D63204 pending parent revision.
This small change is in fact a part of the VOP2b legalization which
does not technically belong to wave32 support, so extracted
separately.
llvm-svn: 363625
This patch really contains two pieces:
Teach SCEV how to fold a phi in the header of a loop to the value on the backedge when a) the backedge is known to execute at least once, and b) the value is safe to use globally within the scope dominated by the original phi.
Teach IndVarSimplify's rewriteLoopExitValues to allow loop invariant expressions which already exist (and thus don't need new computation inserted) even in loops where we can't optimize away other uses.
Differential Revision: https://reviews.llvm.org/D63224
llvm-svn: 363619
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
Recommit r363289 with a bug fix for crash identified in pr42279. Issue was that a loop exit test does not have to be an icmp, leading to a null dereference crash when new logic was exercised for that case. Test case previously committed in r363601.
Original commit comment follows:
This contains fixes for two cases where we might invalidate inbounds and leave it stale in the IR (a miscompile). Case 1 is when switching to an IV with no dynamically live uses, and case 2 is when doing pre-to-post conversion on the same pointer type IV.
The basic scheme used is to prove that using the given IV (pre or post increment forms) would have to already trigger UB on the path to the test we're modifying. As such, our potential UB triggering use does not change the semantics of the original program.
As was pointed out in the review thread by Nikita, this is defending against a separate issue from the hasConcreteDef case. This is about poison, that's about undef. Unfortunately, the two are different, see Nikita's comment for a fuller explanation, he explains it well.
(Note: I'm going to address Nikita's last style comment in a separate commit just to minimize chance of subtle bugs being introduced due to typos.)
Differential Revision: https://reviews.llvm.org/D62939
llvm-svn: 363613
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:
Update compare normalization in SimpleValue hashing to break ties (when
the same value is being compared to itself) by switching to the swapped
predicate if it has a lower numerical value. This brings the hashing in
line with isEqual, which already recognizes the self-compares with
swapped predicates as equal.
Fixes PR 42280.
Reviewers: spatel, efriedma, nikic, fhahn, uabelho
Reviewed By: nikic
Subscribers: hiraditya, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D63349
llvm-svn: 363598
Summary:
LoopRotate doesn't create a faithful clone of an instruction, it may
simplify it beforehand. Hence the clone of an instruction that has a
MemoryDef associated may not be a definition, but a use or not a memory
alternig instruction.
Don't rely on the template when the clone may be simplified.
Reviewers: george.burgess.iv
Subscribers: jlebar, Prazek, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D63355
llvm-svn: 363597
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
We don't know if its safe to unfold if we're in 32-bit mode.
This is simlar to what was done to some load opcodes in r363523.
I think its pretty unlikely we will try to unfold these anyway so
I don't think this is testable.
llvm-svn: 363595
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
Summary:
Add all MemoryPhis in IDF before filling in their incomign values.
Otherwise, a new Phi can be added that needs to become the incoming
value of another Phi.
Test fails the verification in verifyPrevDefInPhis.
Reviewers: george.burgess.iv
Subscribers: jlebar, Prazek, zzheng, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D63353
llvm-svn: 363590
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
When considering a loop containing nontemporal stores or loads for
vectorization, suppress the vectorization if the corresponding
vectorized store or load with the aligment of the original scaler
memory op is not supported with the nontemporal hint on the target.
This adds two new functions:
bool isLegalNTStore(Type *DataType, unsigned Alignment) const;
bool isLegalNTLoad(Type *DataType, unsigned Alignment) const;
to TTI, leaving the target independent default implementation as
returning true, but with overriding implementations for X86 that
check the legality based on available Subtarget features.
This fixes https://llvm.org/PR40759
Differential Revision: https://reviews.llvm.org/D61764
llvm-svn: 363581
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
Use -fsave-optimization-record=<format> to specify a different format
than the default, which is YAML.
For now, only YAML is supported.
llvm-svn: 363573
This is currently only used for ymm->xmm splitting but we shouldn't hardcode the offsets/alignment.
This is necessary for an upcoming patch to split under-aligned non-temporal vector loads.
llvm-svn: 363570
Summary:
There is PHINode::getBasicBlockIndex() and PHINode::setIncomingValue()
but no function to replace incoming value for a specified BasicBlock*
predecessor.
Clearly, there are a lot of places that could use that functionality.
Reviewer: craig.topper, lebedev.ri, Meinersbur, kbarton, fhahn
Reviewed By: Meinersbur, fhahn
Subscribers: fhahn, hiraditya, zzheng, jsji, llvm-commits
Tag: LLVM
Differential Revision: https://reviews.llvm.org/D63338
llvm-svn: 363566
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
If an addrspacecast needed to be inserted again, this was creating a
clone of the original cast for each user. Just use the original, which
also saves losing the value name.
llvm-svn: 363562
The HardwareLoops pass finds exit blocks with a scevable exit count.
If the target specifies to update the loop counter in a register,
through a phi, we need to ensure that the exit block is a latch so
that we can insert the phi with the correct value for the incoming
edge.
Differential Revision: https://reviews.llvm.org/D63336
llvm-svn: 363556
Summary:
Avoid that loop vectorizer creates loads/stores of vectors
with "irregular" types when interleaving. An example of
an irregular type is x86_fp80 that is 80 bits, but that
may have an allocation size that is 96 bits. So an array
of x86_fp80 is not bitcast compatible with a vector
of the same type.
Not sure if interleavedAccessCanBeWidened is the best
place for this check, but it solves the problem seen
in the added test case. And it is the same kind of check
that already exists in memoryInstructionCanBeWidened.
Reviewers: fhahn, Ayal, craig.topper
Reviewed By: fhahn
Subscribers: hiraditya, rkruppe, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D63386
llvm-svn: 363547
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
Create the ARMBasicBlockUtils class for tracking and querying basic
blocks sizes so we can use them when generating low-overhead loops.
Differential Revision: https://reviews.llvm.org/D63265
llvm-svn: 363530
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
SimplifyCFG has a bug that results in inconsistent prof branch_weights metadata
if unreachable switch cases are removed. This patch fixes this bug by making use
of the newly introduced SwitchInstProfUpdateWrapper class (see patch D62122).
A new test is created.
Differential Revision: https://reviews.llvm.org/D62186
llvm-svn: 363527
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
It would not be safe to unfold the memory form the register form
without checking that we are compiling for 64-bit mode.
This probaby isn't a real functional issue since we are unlikely
to unfold any of these instructions since they don't have any
tied registers, aren't commutable, and don't have any inputs
other than the address.
llvm-svn: 363523
Fix folds of addo and subo with an undef operand to be:
`@llvm.{u,s}{add,sub}.with.overflow` all fold to `{ undef, false }`,
as per LLVM undef rules.
Same for commuted variants.
Based on the original version of the patch by @nikic.
Fixes [[ https://bugs.llvm.org/show_bug.cgi?id=42209 | PR42209 ]]
Differential Revision: https://reviews.llvm.org/D63065
llvm-svn: 363522
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:
With Split DWARF the resulting object file (then called skeleton CU)
contains the file name of another ("DWO") file with the debug info.
This can be a problem for remote compilation, as it will contain the
name of the file on the compilation server, not on the client.
To use Split DWARF with remote compilation, one needs to either
* make sure only relative paths are used, and mirror the build directory
structure of the client on the server,
* inject the desired file name on the client directly.
Since llc already supports the latter solution, we're just copying that
over. We allow setting the actual output filename separately from the
value of the DW_AT_[GNU_]dwo_name attribute in the skeleton CU.
Fixes PR40276.
Reviewers: dblaikie, echristo, tejohnson
Reviewed By: dblaikie
Differential Revision: https://reviews.llvm.org/D59673
llvm-svn: 363496
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
Summary:
When using ConstantExpr we often need the result of the expression to be kept in the AST. Currently this is done on a by the node that needs the result and has been done multiple times for enumerator, for constexpr variables... . This patch adds to ConstantExpr the ability to store the result of evaluating the expression. no functional changes expected.
Changes:
- Add trailling object to ConstantExpr that can hold an APValue or an uint64_t. the uint64_t is here because most ConstantExpr yield integral values so there is an optimized layout for integral values.
- Add basic* serialization support for the trailing result.
- Move conversion functions from an enum to a fltSemantics from clang::FloatingLiteral to llvm::APFloatBase. this change is to make it usable for serializing APValues.
- Add basic* Import support for the trailing result.
- ConstantExpr created in CheckConvertedConstantExpression now stores the result in the ConstantExpr Node.
- Adapt AST dump to print the result when present.
basic* : None, Indeterminate, Int, Float, FixedPoint, ComplexInt, ComplexFloat,
the result is not yet used anywhere but for -ast-dump.
Reviewers: rsmith, martong, shafik
Reviewed By: rsmith
Subscribers: rnkovacs, hiraditya, dexonsmith, cfe-commits, llvm-commits
Tags: #clang, #llvm
Differential Revision: https://reviews.llvm.org/D62399
llvm-svn: 363493
Based on D59959, this switches SCEV to use unsigned/signed range
intersection based on the sign hint. This will prefer non-wrapping
ranges in the relevant domain. I've left the one intersection in
getRangeForAffineAR() to use the smallest intersection heuristic,
as there doesn't seem to be any obvious preference there.
Differential Revision: https://reviews.llvm.org/D60035
llvm-svn: 363490
If we can detect that saturating math that depends on an IV cannot
overflow, replace it with simple math. This is similar to the CVP
optimization from D62703, just based on a different underlying
analysis (SCEV vs LVI) that catches different cases.
Differential Revision: https://reviews.llvm.org/D62792
llvm-svn: 363489
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
Current findBestLoopTop can find and move one kind of block to top, a latch block has one successor. Another common case is:
* a latch block
* it has two successors, one is loop header, another is exit
* it has more than one predecessors
If it is below one of its predecessors P, only P can fall through to it, all other predecessors need a jump to it, and another conditional jump to loop header. If it is moved before loop header, all its predecessors jump to it, then fall through to loop header. So all its predecessors except P can reduce one taken branch.
Differential Revision: https://reviews.llvm.org/D43256
llvm-svn: 363471
with 'objc_arc_inert'
Those calls are no-ops, so they can be safely deleted.
rdar://problem/49839633
Differential Revision: https://reviews.llvm.org/D62433
llvm-svn: 363468
Currently you get extra waits, because waits are inserted for the
register dependencies of the call, and the function prolog waits on
everything.
Currently waits are still inserted on returns. It may make sense to
not do this, and wait in the caller instead.
llvm-svn: 363465
This patch allows clang users to print out a list of supported CPU models using
clang [--target=<target triple>] --print-supported-cpus
Then, users can select the CPU model to compile to using
clang --target=<triple> -mcpu=<model> a.c
It is a handy feature to help cross compilation.
llvm-svn: 363464
There is a circular dependency between SROA and InferAddressSpaces
today that requires running both multiple times in order to be able to
eliminate all simple allocas and addrspacecasts. InferAddressSpaces
can't remove addrspacecasts when written to memory, and SROA helps
move pointers out of memory.
This should avoid inserting new commuting addrspacecasts with GEPs,
since there are unresolved questions about pointer wrapping between
different address spaces.
For now, don't replace volatile operations that don't match the alloca
addrspace, as it would change the address space of the access. It may
be still OK to insert an addrspacecast from the new alloca, but be
more conservative for now.
llvm-svn: 363462
The way SelectionDAG treats memory operands is very frustrating, and
by default drops them unless a property is set on the pattern. There
is no pattern for manually selected instructions, so this requires
manually setting them.
llvm-svn: 363455
This is a branch opcode that takes a jump table pointer, jump table index and an
index into the table to do an indirect branch.
We pass both the table pointer and JTI to allow targets like ARM64 to more
easily use the existing jump table compression optimization without having to
walk up the block to find a paired G_JUMP_TABLE.
Differential Revision: https://reviews.llvm.org/D63159
llvm-svn: 363434
and replace with an equilivent countTrailingZeros.
GCD is much more expensive than this, with repeated division.
This depends on D60823
Differential Revision: https://reviews.llvm.org/D61151
llvm-svn: 363422
I recently discovered a bug on the x86 platform: The fp80 type was not handled well by x86 for constrained floating point nodes, as their regular counterparts are replaced by extending loads and truncating stores during the preprocess phase. Normally, platforms don't have this issue, as they don't typically attempt to perform such legalizations during instruction selection preprocessing. Before this change, strict_fp nodes survived until they were mutated to normal nodes, which happened shortly after preprocessing on other platforms. This modification lowers these nodes at the same phase while properly utilizing the chain.5
Submitted by: Drew Wock <drew.wock@sas.com>
Reviewed by: Craig Topper, Kevin P. Neal
Approved by: Craig Topper
Differential Revision: https://reviews.llvm.org/D63271
llvm-svn: 363417
Earlier commit has added AMDGPUOperand::isBoolReg(). Turns out
gcc issues warning about unused function since D63204 is not
yet submitted.
Added NFC part of D63204 to have a use of that function and
mute the warning.
llvm-svn: 363416
* Add a common function to setup opt-remarks
* Rename common options to the same names
* Add error types to distinguish between file errors and regex errors
llvm-svn: 363415
Avoid producing illegal register bank copies for reg_sequence and
phi. The default implementation assumes it is possible to pick any
operand's bank and use that for the result, introducing a copy for
operands with a different bank. This does not check for illegal
copies. It is not legal to introduce a VGPR->SGPR copy, so any VGPR
operand requires the result to be a VGPR.
The changes in getInstrMappingImpl aren't strictly necessary, since
AMDGPU now just bypasses this for reg_sequence/phi. This could be
replaced with an assert in case other targets run into this. It is
currently responsible for producing the error for unsatisfiable
copies, but this will be better served with a verifier check.
For phis, for now assume any undetermined operands must be
VGPRs. Eventually, this needs to be able to defer mapping these
operations. This also does not yet have a way to check for whether the
block is in a divergent region.
llvm-svn: 363410
I'm not 100% sure about this, since I'm worried about IR transforms
that might end up introducing divergence downstream once replaced with
a constant, but I haven't come up with an example yet.
llvm-svn: 363406
This is the family of vector instructions that combine all the lanes
in their input vector(s), and output a value in one or two GPRs.
Differential Revision: https://reviews.llvm.org/D62670
llvm-svn: 363403
r363261 caused test failure on 32-bit ARM buildbot,
because of unsigned integer overflow. This patch
fixes it changing offset type from size_t to uint64_t.
llvm-svn: 363393
Avoid a check for valid and a set of redundant asserts. The place
InstructionMapping is constructed asserts all of the default fields
are passed anyway for an invalid mapping, so don't overcomplicate
this.
llvm-svn: 363391
In debug mode, the tool also raises a warning and prints out a message which
helps identify the problematic MCWriteProcResEntry from the scheduling class.
This message would have been useful to have when triaging PR42282.
llvm-svn: 363387
Initial commit of a new pass to create vector predication blocks, called VPT
blocks, that are supported by the Armv8.1-M MVE architecture.
This is a first naive implementation. I.e., for 2 consecutive predicated
instructions I1 and I2, for example, it will generate 2 VPT blocks:
VPST
I1
VPST
I2
A more optimal implementation would obviously put instructions in the same VPT
block when they are predicated on the same condition and when it is allowed to
do this:
VPTT
I1
I2
We will address this optimisation with follow up patches when the groundwork is
in. Creating VPT Blocks is very similar to IT Blocks, which is the reason I
added this to Thumb2ITBlocks.cpp. This allows reuse of the def use analysis
that we need for the more optimal implementation.
VPT blocks cannot be nested in IT blocks, and vice versa, and so these 2 passes
cannot interact with each other. Instructions allowed in VPT blocks must
be MVE instructions that are marked as VPT compatible.
Differential Revision: https://reviews.llvm.org/D63247
llvm-svn: 363370
With this patch we get ability to set any flags we want
for implicit sections defined in YAML.
Differential revision: https://reviews.llvm.org/D63136
llvm-svn: 363367
InsertBinop now accepts NoWrapFlags, so pass them through when
expanding a simple add expression.
This is the first re-commit of the functional changes from rL362687,
which was previously reverted.
Differential Revision: https://reviews.llvm.org/D61934
llvm-svn: 363364
This is consistent with GCC's behavior (which is the defacto standard
for pubnames). Though I find the presence of enumerators from enum
classes to be a bit confusing, possibly a bug on GCC's end (since they
can't be named unqualified, unlike the other names - and names nested in
classes don't go in pubnames, for instance - presumably because one must
name the class first & that's enough to limit the scope of the search)
llvm-svn: 363349
It looks like an older version of gcc can't figure out that it needs to
move a unique_ptr while implicitly constructing an Expected object.
llvm-svn: 363342
Summary:
Before it was using the fully qualified name only for static data members.
Now it does for all variable names to match MSVC.
Reviewers: rnk
Subscribers: hiraditya, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D63012
llvm-svn: 363335
The only caller of SymbolizableObjectFile::create passes a non-null
DebugInfoContext and asserts that they do so. Move the assert into
SymbolizableObjectFile::create and remove null checks.
Differential Revision: https://reviews.llvm.org/D63298
llvm-svn: 363334
Constants, including G_GLOBAL_VALUE, are all emitted into the entry block which
lets us use the vreg def assuming it dominates all other users. However, it can
cause jumpy debug behaviour since the DebugLoc attached to these MIs are from
a user instruction that could be in a different block.
Fixes PR40887.
Differential Revision: https://reviews.llvm.org/D63286
llvm-svn: 363331
Merging the two bits shrinks the context table from 16384 bytes to 8192 bytes.
Remove the ATTRIBUTE_BITS macro and just create an enum directly. Then fix the ATTR_max define to be 8192 to reflect the table size so we stop hardcoding it separately.
llvm-svn: 363330
This was exposed by PowerPC target enablement.
In ScheduleDAG, if we haven't seen any uses in this scheduling region,
we will create a dependence edge to ExitSU to model the live-out latency.
This is required for vreg defs with no in-region use, and prefetches with
no vreg def.
When we build NodeOrder in Scheduler, we ignore these boundary nodes.
However, when we check Succs in checkValidNodeOrder, we did not skip
them, so we still assume all the nodes have been sorted and in order in
Indices array. So when we call lower_bound() for ExitSU, it will return
Indices.end(), causing memory issues in following Node access.
Differential Revision: https://reviews.llvm.org/D63282
llvm-svn: 363329
* Add a common function to setup opt-remarks
* Rename common options to the same names
* Add error types to distinguish between file errors and regex errors
llvm-svn: 363328
Support loading code coverage data from regular archives, thin archives,
and from MachO universal binaries which contain archives.
Testing: check-llvm, check-profile (with {A,UB}San enabled)
rdar://51538999
Differential Revision: https://reviews.llvm.org/D63232
llvm-svn: 363325
notifyResolved/notifyEmitted.
The 'notify' prefix better describes what these methods do: they update the JIT
symbol states and notify any pending queries that the 'resolved' and 'emitted'
states have been reached (rather than actually performing the resolution or
emission themselves). Since new states are going to be introduced in the near
future (to track symbol registration/initialization) it's worth changing the
convention pre-emptively to avoid further confusion.
llvm-svn: 363322
I find the current documentation of poison somewhat confusing,
mainly because its use of "undefined behavior" doesn't seem to
align with our usual interpretation (of immediate UB). Especially
the sentence "any instruction that has a dependence on a poison
value has undefined behavior" is very confusing.
Clarify poison semantics by:
* Replacing the introductory paragraph with the standard rationale
for having poison values.
* Spelling out that instructions depending on poison return poison.
* Spelling out how we go from a poison value to immediate undefined
behavior and give the two examples we currently use in ValueTracking.
* Spelling out that side effects depending on poison are UB.
Differential Revision: https://reviews.llvm.org/D63044
llvm-svn: 363320
I recently got this wrong (again), and I'm sure I'm not the only one. Put a comment in the logical place someone would look to "fix" the obvious "missed optimization" which arrises based on the common misunderstanding. Hopefully, this will save others time. :)
llvm-svn: 363318
As pointed out by Nikita in D62625, BackedgeTakenCount is generally used to refer to the backedge taken count of the loop. A conditional backedge taken count - one which only applies if a particular exit is taken - is called a ExitCount in SCEV code, so be consistent here.
llvm-svn: 363293
This contains fixes for two cases where we might invalidate inbounds and leave it stale in the IR (a miscompile). Case 1 is when switching to an IV with no dynamically live uses, and case 2 is when doing pre-to-post conversion on the same pointer type IV.
The basic scheme used is to prove that using the given IV (pre or post increment forms) would have to already trigger UB on the path to the test we're modifying. As such, our potential UB triggering use does not change the semantics of the original program.
As was pointed out in the review thread by Nikita, this is defending against a separate issue from the hasConcreteDef case. This is about poison, that's about undef. Unfortunately, the two are different, see Nikita's comment for a fuller explanation, he explains it well.
(Note: I'm going to address Nikita's last style comment in a separate commit just to minimize chance of subtle bugs being introduced due to typos.)
Differential Revision: https://reviews.llvm.org/D62939
llvm-svn: 363289
Add an AssumptionCache callback to the InlineFuntionInfo used for the
AlwaysInlinerPass to match codegen of the AlwaysInlinerLegacyPass to generate
llvm.assume. This fixes CodeGen/builtin-movdir.c when new PM is enabled by
default.
Differential Revision: https://reviews.llvm.org/D63170
llvm-svn: 363287
Summary:
I found the following case having tail blocks with no successors merging opportunities after block placement.
Before block placement:
bb0:
...
bne a0, 0, bb2:
bb1:
mv a0, 1
ret
bb2:
...
bb3:
mv a0, 1
ret
bb4:
mv a0, -1
ret
The conditional branch bne in bb0 is opposite to beq.
After block placement:
bb0:
...
beq a0, 0, bb1
bb2:
...
bb4:
mv a0, -1
ret
bb1:
mv a0, 1
ret
bb3:
mv a0, 1
ret
After block placement, that appears new tail merging opportunity, bb1 and bb3 can be merged as one block. So the conditional constraint for merging tail blocks with no successors should be removed. In my experiment for RISC-V, it decreases code size.
Author of original patch: Jim Lin
Reviewers: haicheng, aheejin, craig.topper, rnk, RKSimon, Jim, dmgreen
Reviewed By: Jim, dmgreen
Subscribers: xbolva00, dschuff, javed.absar, sbc100, jgravelle-google, aheejin, kito-cheng, dmgreen, PkmX, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D54411
llvm-svn: 363284
Summary:
The logic in EarlyCSE that looks through 'not' operations in the
predicate recognizes e.g. that `select (not (cmp sgt X, Y)), X, Y` is
equivalent to `select (cmp sgt X, Y), Y, X`. Without this change,
however, only the latter is recognized as a form of `smin X, Y`, so the
two expressions receive different hash codes. This leads to missed
optimization opportunities when the quadratic probing for the two hashes
doesn't happen to collide, and assertion failures when probing doesn't
collide on insertion but does collide on a subsequent table grow
operation.
This change inverts the order of some of the pattern matching, checking
first for the optional `not` and then for the min/max/abs patterns, so
that e.g. both expressions above are recognized as a form of `smin X, Y`.
It also adds an assertion to isEqual verifying that it implies equal
hash codes; this fires when there's a collision during insertion, not
just grow, and so will make it easier to notice if these functions fall
out of sync again. A new flag --earlycse-debug-hash is added which can
be used when changing the hash function; it forces hash collisions so
that any pair of values inserted which compare as equal but hash
differently will be caught by the isEqual assertion.
Reviewers: spatel, nikic
Reviewed By: spatel, nikic
Subscribers: lebedev.ri, arsenm, craig.topper, efriedma, hiraditya, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D62644
llvm-svn: 363274
Previously it copied over MachineMemOperands verbatim which caused MOV32rm to have store flags set, and MOV32mr to have load flags set. This fixes some assertions being thrown with EXPENSIVE_CHECKS on.
Committed on behalf of @luke (Luke Lau)
Differential Revision: https://reviews.llvm.org/D62726
llvm-svn: 363268
Summary:
Relate bug: https://bugs.llvm.org/show_bug.cgi?id=37472
The shrink wrapping pass prematurally restores the stack, at a point where the stack might still be accessed.
Taking an exception can cause the stack to be corrupted.
As a first approach, this patch is overly conservative, assuming that any instruction that may load or store could access
the stack.
Reviewers: dmgreen, qcolombet
Reviewed By: qcolombet
Subscribers: simpal01, efriedma, eli.friedman, javed.absar, llvm-commits, eugenis, chill, carwil, thegameg
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D63152
llvm-svn: 363265
This commit prepares the way to start adding the main collection of
MVE instructions, which operate on the 128-bit vector registers.
The most obvious thing that's needed, and the simplest, is to add the
MQPR register class, which is like the existing QPR except that it has
fewer registers in it.
The more complicated part: MVE defines a system of vector predication,
in which instructions operating on 128-bit vector registers can be
constrained to operate on only a subset of the lanes, using a system
of prefix instructions similar to the existing Thumb IT, in that you
have one prefix instruction which designates up to 4 following
instructions as subject to predication, and within that sequence, the
predicate can be inverted by means of T/E suffixes ('Then' / 'Else').
To support instructions of this type, we've added two new Tablegen
classes `vpred_n` and `vpred_r` for standard clusters of MC operands
to add to a predicated instruction. Both include a flag indicating how
the instruction is predicated at all (options are T, E and 'not
predicated'), and an input register field for the register controlling
the set of active lanes. They differ from each other in that `vpred_r`
also includes an input operand for the previous value of the output
register, for instructions that leave inactive lanes unchanged.
`vpred_n` lacks that extra operand; it will be used for instructions
that don't preserve inactive lanes in their output register (either
because inactive lanes are zeroed, as the MVE load instructions do, or
because the output register isn't a vector at all).
This commit also adds the family of prefix instructions themselves
(VPT / VPST), and all the machinery needed to work with them in
assembly and disassembly (e.g. generating the 't' and 'e' mnemonic
suffixes on disassembled instructions within a predicated block)
I've added a couple of demo instructions that derive from the new
Tablegen base classes and use those two operand clusters. The bulk of
the vector instructions will come in followup commits small enough to
be manageable. (One exception is that I've added the full version of
`isMnemonicVPTPredicable` in the AsmParser, because it seemed
pointless to carefully split it up.)
Reviewers: dmgreen, samparker, SjoerdMeijer, t.p.northover
Subscribers: javed.absar, kristof.beyls, hiraditya, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D62669
llvm-svn: 363258
This patch makes the LiveDebugValues pass consider fragments when propagating
DBG_VALUE insts between blocks, fixing PR41979. Fragment info for a variable
location is added to the open-ranges key, which allows distinct fragments to be
tracked separately. To handle overlapping fragments things become slightly
funkier. To avoid excessive searching for overlaps in the data-flow part of
LiveDebugValues, this patch:
* Pre-computes pairings of fragments that overlap, for each DILocalVariable
* During data-flow, whenever something happens that causes an open range to
be terminated (via erase), any fragments pre-determined to overlap are
also terminated.
The effect of which is that when encountering a DBG_VALUE fragment that
overlaps others, the overlapped fragments do not get propagated to other
blocks. We still rely on later location-list building to correctly handle
overlapping fragments within blocks.
It's unclear whether a mixture of DBG_VALUEs with and without fragmented
expressions are legitimate. To avoid suprises, this patch interprets a
DBG_VALUE with no fragment as overlapping any DBG_VALUE _with_ a fragment.
Differential Revision: https://reviews.llvm.org/D62904
llvm-svn: 363256
Since the DebugLocEntry::Value is used as part of DwarfDebug and
DebugLocEntry make it as the separate class.
Reviewers: aprantl, dstenb
Reviewed By: aprantl
Differential Revision: https://reviews.llvm.org/D63213
llvm-svn: 363246
We aim to ignore changes in variable locations during the prologue and
epilogue of functions, to avoid using space documenting location changes
that aren't visible. However in D61940 / r362951 this got ripped out as
the previous implementation was unsound.
Instead, use the FrameDestroy flag to identify when we're in the epilogue
of a function, and ignore variable location changes accordingly. This fits
in with existing code that examines the FrameSetup flag.
Some variable locations get shuffled in modified tests as they now cover
greater ranges, which is what would be expected. Some additional
single-location variables are generated too. Two tests are un-xfailed,
they were only xfailed due to r362951 deleting functionality they depended
on.
Apparently some out-of-tree backends don't accurately maintain FrameDestroy
flags -- if you're an out-of-tree maintainer and see changes in variable
locations disappear due to a faulty FrameDestroy flag, it's safe to back
this change out. The impact is just slightly more debug info than necessary.
Differential Revision: https://reviews.llvm.org/D62314
llvm-svn: 363245
During assembly, the mask operand to an IT instruction (storing the
sequence of T/E for 'Then' and 'Else') is parsed out of the mnemonic
into a representation that encodes 'Then' and 'Else' in the same way
regardless of the condition code. At some point during encoding it has
to be converted into the instruction encoding used in the
architecture, in which the mask encodes a sequence of replacement
low-order bits for the condition code, so that which bit value means
'then' and which 'else' depends on whether the original condition code
had its low bit set.
Previously, that transformation was done by processInstruction(), half
way through assembly. So an MCOperand storing an IT mask would
sometimes store it in one format, and sometimes in the other,
depending on where in the assembly pipeline you were. You can see this
in diagnostics from `llvm-mc -debug -triple=thumbv8a -show-inst`, for
example: if you give it an instruction such as `itete eq`, you'd see
an `<MCOperand Imm:5>` in a diagnostic become `<MCOperand Imm:11>` in
the final output.
Having the same data structure store values with time-dependent
semantics is confusing already, and it will get more confusing when we
introduce the MVE VPT instruction which reuses the Then/Else bitmask
idea in a different context. So I'm refactoring: now, all `ARMOperand`
and `MCOperand` representations of an IT mask work exactly the same
way, namely, 0 means 'Then' and 1 means 'Else', regardless of what
original predicate is being referred to. The architectural encoding of
IT that depends on the original condition is now constructed at the
point when we turn the `MCOperand` into the final instruction bit
pattern, and decoded similarly in the disassembler.
The previous condition-independent parse-time format used 0 for Else
and 1 for Then. I've taken the opportunity to flip the sense of it
while I'm changing all of this anyway, because it seems to me more
natural to use 0 for 'leave the starting condition unchanged' and 1
for 'invert it', as if those bits were an XOR mask.
Reviewers: ostannard
Subscribers: javed.absar, kristof.beyls, hiraditya, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D63219
llvm-svn: 363244
This patch uses the mechanism from D62995 to strengthen the
definitions of the reduction intrinsics by letting the scalar
result/accumulator type be overloaded from the vector element type.
For example:
; The LLVM LangRef specifies that the scalar result must equal the
; vector element type, but this is not checked/enforced by LLVM.
declare i32 @llvm.experimental.vector.reduce.or.i32.v4i32(<4 x i32> %a)
This patch changes that into:
declare i32 @llvm.experimental.vector.reduce.or.v4i32(<4 x i32> %a)
Which has the type-constraint more explicit and causes LLVM to check
the result type with the vector element type.
Reviewers: RKSimon, arsenm, rnk, greened, aemerson
Reviewed By: arsenm
Differential Revision: https://reviews.llvm.org/D62996
llvm-svn: 363240
TTI should report that it's not profitable to generate a hardware loop
if it, or one of its child loops, has already been converted.
Differential Revision: https://reviews.llvm.org/D63212
llvm-svn: 363234
Extend the mechanism to overload intrinsic arguments by using either
backward or forward references to the overloadable arguments.
In for example:
def int_something : Intrinsic<[LLVMPointerToElt<0>],
[llvm_anyvector_ty], []>;
LLVMPointerToElt<0> is a forward reference to the overloadable operand
of type 'llvm_anyvector_ty' and would allow intrinsics such as:
declare i32* @llvm.something.v4i32(<4 x i32>);
declare i64* @llvm.something.v2i64(<2 x i64>);
where the result pointer type is deduced from the element type of the
first argument.
If the returned pointer is not a pointer to the element type, LLVM will
give an error:
Intrinsic has incorrect return type!
i64* (<4 x i32>)* @llvm.something.v4i32
Reviewers: RKSimon, arsenm, rnk, greened
Reviewed By: arsenm
Differential Revision: https://reviews.llvm.org/D62995
llvm-svn: 363233
This reverts 363226 and 363227, both NFC intended
I swear I fixed the test case that is failing, and ran
the tests, but I will look into it again.
llvm-svn: 363229
and replace with an equilivent countTrailingZeros.
GCD is much more expensive than this, with repeated division.
This depends on D60823
Differential Revision: https://reviews.llvm.org/D61151
llvm-svn: 363227
Summary:
- Remove redundant initializations from pass constructors that were
already being initialized by LLVMInitializeX86Target().
- Add initialization function for the FPS pass.
Reviewers: craig.topper
Reviewed By: craig.topper
Subscribers: hiraditya, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D63218
llvm-svn: 363221
We have observed some failures with internal builds with this revision.
- Performance regressions:
- llvm's SingleSource/Misc evalloop shows performance regressions (although these may be red herrings).
- Benchmarks for Abseil's SwissTable.
- Correctness:
- Failures for particular libicu tests when building the Google AppEngine SDK (for PHP).
hwennborg has already been notified, and is aware of reproducer failures.
llvm-svn: 363220
Add support for s.d instruction for Mips1 which expands into two swc1
instructions.
Patch by Mirko Brkusanin.
Differential Revision: https://reviews.llvm.org/D63199
llvm-svn: 363184
SCEV does not propagate arguments through one-input Phis so as to make it easy for the SCEV expander (and related code) to preserve LCSSA. It's not entirely clear this restriction is neccessary, but for the moment it exists. For this reason, we don't analyze single-entry phi inputs. However it is possible that when an this input leaves the loop through LCSSA Phi, it is a provable constant. Missing that results in an order of optimization issue in loop exit value rewriting where we miss some oppurtunities based on order in which we visit sibling loops.
This patch teaches computeSCEVAtScope about this case. We can generalize it later, but so far we can only replace LCSSA Phis with their constant loop-exiting values. We should probably also add similiar logic directly in the SCEV construction path itself.
Patch by: mkazantsev (with revised commit message by me)
Differential Revision: https://reviews.llvm.org/D58113
llvm-svn: 363180
As discussed on D62910, we need to check whether particular types of memory access are allowed, not just their alignment/address-space.
This NFC patch adds a MachineMemOperand::Flags argument to allowsMemoryAccess and allowsMisalignedMemoryAccesses, and wires up calls to pass the relevant flags to them.
If people are happy with this approach I can then update X86TargetLowering::allowsMisalignedMemoryAccesses to handle misaligned NT load/stores.
Differential Revision: https://reviews.llvm.org/D63075
llvm-svn: 363179
This was using its own, outdated list of possible captures. This was
at minimum not catching cmpxchg and addrspacecast captures.
One change is now any volatile access is treated as capturing. The
test coverage for this pass is quite inadequate, but this required
removing volatile in the lifetime capture test.
Also fixes some infrastructure issues to allow running just the IR
pass.
Fixes bug 42238.
llvm-svn: 363169
Without this fix clang 3.6 complains with:
../lib/Target/ARM/ARMAsmPrinter.cpp:1473:18: error: variable 'BranchTarget' is used uninitialized whenever 'if' condition is false [-Werror,-Wsometimes-uninitialized]
} else if (MI->getOperand(1).isSymbol()) {
^~~~~~~~~~~~~~~~~~~~~~~~~~~~
../lib/Target/ARM/ARMAsmPrinter.cpp:1479:22: note: uninitialized use occurs here
MCInst.addExpr(BranchTarget);
^~~~~~~~~~~~
../lib/Target/ARM/ARMAsmPrinter.cpp:1473:14: note: remove the 'if' if its condition is always true
} else if (MI->getOperand(1).isSymbol()) {
^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
../lib/Target/ARM/ARMAsmPrinter.cpp:1465:33: note: initialize the variable 'BranchTarget' to silence this warning
const MCExpr *BranchTarget;
^
= nullptr
1 error generated.
Discussed here:
http://lists.llvm.org/pipermail/llvm-commits/Week-of-Mon-20190610/661417.html
llvm-svn: 363166
This changes the standalone pass only. Arguably the utility class
itself should assert there are no convergent calls. However, a target
pass with additional context may still be able to version a loop if
all of the dynamic conditions are sufficiently uniform.
llvm-svn: 363165
Summary:
Fix hoisting to basic block which are not legal for hoisting cause
it can be terminated by exception or it is return block.
Reviewers: john.brawn, RKSimon, MatzeB
Subscribers: hiraditya, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D63148
llvm-svn: 363164
This case is slightly tricky, because loop distribution should be
allowed in some cases, and not others. As long as runtime dependency
checks don't need to be introduced, this should be OK. This is further
complicated by the fact that LoopDistribute partially ignores if LAA
says that vectorization is safe, and then does its own runtime pointer
legality checks.
Note this pass still does not handle noduplicate correctly, as this
should always be forbidden with it. I'm not going to bother trying to
fix it, as it would require more effort and I think noduplicate should
be removed.
https://reviews.llvm.org/D62607
llvm-svn: 363160
The capture was added in the first commit of https://reviews.llvm.org/D61934
when it was used. In the reland, the use was removed but the capture
wasn't removed.
llvm-svn: 363155
Implement the backend target hook to drive the HardwareLoops pass.
The low-overhead branch extension for Arm M-class cores is flexible
enough that we don't have to ensure correctness at this point, except
checking that the loop counter variable can be stored in LR - a
32-bit register. For it to be profitable, we want to avoid loops that
contain function calls, or any other instruction that alters the PC.
This implementation uses TargetLoweringInfo, to query type and
operation actions, looks at intrinsic calls and also performs some
manual checks for remainder/division and FP operations.
I think this should be a good base to start and extra details can be
filled out later.
Differential Revision: https://reviews.llvm.org/D62907
llvm-svn: 363149
'Use wrap flags in InsertBinop' (rL362687) was reverted due to
miscompiles. This patch introduces the previous change to pass
no-wrap flags but now only FlagAnyWrap is passed.
Differential Revision: https://reviews.llvm.org/D61934
llvm-svn: 363147
r363016 let lld-link and llvm-lib share the /machine: parsing code.
This lets llvm-cvtres share it as well.
Making llvm-cvtres depend on llvm-lib seemed a bit strange (it doesn't
need llvm-lib's dependencies on BinaryFormat and BitReader) and I
couldn't find a good place to put this code. Since it's just a few
lines, put it in lib/Object for now.
Differential Revision: https://reviews.llvm.org/D63120
llvm-svn: 363144
Noticed in D63075 - there was a allowsMisalignedMemoryAccesses call to check for unaligned loads and a check for aligned legal type loads - which is exactly what allowsMemoryAccess does.
llvm-svn: 363141
Dependent libraries support for the legacy api was committed in a
broken state (see: https://reviews.llvm.org/D60274). This was missed
due to the painful nature of having to integrate the changes into a
linker in order to test. This change implements support for dependent
libraries in the legacy LTO api:
- I have removed the current api function, which returns a single
string, and added functions to access each dependent library
specifier individually.
- To reduce the testing pain, I have made the api functions as thin as
possible to maximize coverage from llvm-lto.
- When doing ThinLTO the system linker will load the modules lazily
when scanning the input files. Unfortunately, when modules are
lazily loaded there is no access to module level named metadata. To
fix this I have added api functions that allow querying the IRSymtab
for the dependent libraries. I hope to expand the api in the future
so that, eventually, all the information needed by a client linker
during scan can be retrieved from the IRSymtab.
Differential Revision: https://reviews.llvm.org/D62935
llvm-svn: 363140
Noticed in D63075 - there was a allowsMisalignedMemoryAccesses call to check for unaligned loads and a check for aligned legal type loads - which is exactly what allowsMemoryAccess does.
llvm-svn: 363137
Extern global merging is good for code-size. There's definitely potential for
performance too, but there's one regression in a benchmark that needs
investigating, so that's why we enable it only when we optimise for size for
now.
Patch by Ramakota Reddy and Sjoerd Meijer.
Differential Revision: https://reviews.llvm.org/D61947
llvm-svn: 363130
The non-masked versions are already in there. I'm having some
trouble coming up with a way to test this right now. Most load
folding should happen during isel so I'm not sure how to get
peephole pass to do it.
llvm-svn: 363125
In order to generate correct debug frame information, it needs to
generate CFI information in prologue and epilog.
Differential Revision: https://reviews.llvm.org/D61773
llvm-svn: 363120
The issue is that if we have a loop with multiple predecessors outside the loop, the code was expecting to merge them and only return if equal, but instead returned the first one seen.
I have no idea if this actually tripped anywhere. I noticed it by accident when reading the code and have no idea how to go about constructing a test case.
llvm-svn: 363112
We were only matching RHS being a loop invariant value, not the inverse. Since there's nothing which appears to canonicalize loop invariant values to RHS, this means we missed cases.
Differential Revision: https://reviews.llvm.org/D63112
llvm-svn: 363108
We have the related getSplatValue() already in IR (see code just above the proposed addition).
But sometimes we only need to know that the value is a splat rather than capture the splatted
scalar value. Also, we have an isSplatValue() function already in SDAG.
Motivation - recent bugs that would potentially benefit from improved splat analysis in IR:
https://bugs.llvm.org/show_bug.cgi?id=37428https://bugs.llvm.org/show_bug.cgi?id=42174
Differential Revision: https://reviews.llvm.org/D63138
llvm-svn: 363106
This opcode generates a pointer to the address of the jump table
specified by the source operand, which is a jump table index.
It will be used in conjunction with an upcoming G_BRJT opcode to support
jump table codegen with GlobalISel.
Differential Revision: https://reviews.llvm.org/D63111
llvm-svn: 363096
Summary: After applying a set of insert updates, there may be trivial Phis left over. Clean them up.
Reviewers: george.burgess.iv
Subscribers: jlebar, Prazek, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D63033
llvm-svn: 363094
Summary:
The method `getLoopPassPreservedAnalyses` should not mark MemorySSA as
preserved, because it's being called in a lot of passes that do not
preserve MemorySSA.
Instead, mark the MemorySSA analysis as preserved by each pass that does
preserve it.
These changes only affect the new pass mananger.
Reviewers: chandlerc
Subscribers: mehdi_amini, jlebar, Prazek, george.burgess.iv, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D62536
llvm-svn: 363091
Summary: Deduplicate S_CONSTANTS when linking, if they have the same value.
Reviewers: rnk
Subscribers: hiraditya, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D63151
llvm-svn: 363089
Implement necessary target hooks to enable MachinePipeliner for P9 only.
The pass is off by default, can be enabled with -ppc-enable-pipeliner for P9.
Differential Revision: https://reviews.llvm.org/D62164
llvm-svn: 363085
When moving a temp file, explicitly set the file descriptor to -1 so we
can never accidentally close the moved-from TempFile.
Differential revision: https://reviews.llvm.org/D63087
llvm-svn: 363083
Users are exepcted to pass all .res files to the linker, which then
merges all the resource in all .res files into a tree structure and then
converts the final tree structure to a .obj file with .rsrc$01 and
.rsrc$02 sections and then links that.
If the user instead passes several .obj files containing such resources,
the correct thing to do would be to have custom code to merge the trees
in the resource sections instead of doing normal section merging -- but
link.exe rejects if multiple resource obj files are passed in with
LNK4078, so let lld-link do that too instead of silently writing broken
.rsrc sections in that case.
The only real way to run into this is if users manually convert .res
files to .obj files by running cvtres and then handing the resulting
.obj files to lld-link instead, which in practice likely never happens.
(lld-link is slightly stricter than link.exe now: If link.exe is passed
one .obj file created by cvtres, and a .res file, for some reason it
just emits a warning instead of an error and outputs strange looking
data. lld-link now errors out on mixed input like this.)
One way users could accidentally run into this is the following
scenario: If a .res file is passed to lib.exe, then lib.exe calls
cvtres.exe on the .res file before putting it in the output .lib.
(llvm-lib currently doesn't do this.)
link.exe's /wholearchive seems to only add obj files referenced from the
static library index, but lld-link current really adds all files in the
archive. So if lld-link /wholearchive is used with .lib files produced
by lib.exe and .res files were among the files handed to lib.exe, we
previously silently produced invalid output, but now we error out.
link.exe's /wholearchive semantics on the other hand mean that it
wouldn't load the resource object files from the .lib file at all.
Since this scenario is probably still an unlikely corner case,
the difference in behavior here seems fine -- and lld-link might have to
change to use link.exe's /wholearchive semantics in the future anyways.
Vaguely related to PR42180.
Differential Revision: https://reviews.llvm.org/D63109
llvm-svn: 363078
This patch allows lowering of PIC addresses by using PC-relative
addressing for DSO-local symbols and accessing the address through the
global offset table for non-DSO-local symbols.
Differential Revision: https://reviews.llvm.org/D55303
llvm-svn: 363058
This validates and lowers arguments to inline asm nodes which have the
constraints I, J & K, with the following semantics (equivalent to GCC):
I: Any 12-bit signed immediate.
J: Immediate integer zero only.
K: Any 5-bit unsigned immediate.
Differential Revision: https://reviews.llvm.org/D54093
llvm-svn: 363054
This introduces a new decoding table for MVE instructions, and starts
by adding the family of scalar shift instructions that are part of the
MVE architecture extension: saturating shifts within a single GPR, and
long shifts across a pair of GPRs (both saturating and normal).
Some of these shift instructions have only 3-bit register fields in
the encoding, with the low bit fixed. So they can only address an odd
or even numbered GPR (depending on the operand), and therefore I add
two new register classes, GPREven and GPROdd.
Differential Revision: https://reviews.llvm.org/D62668
Change-Id: Iad95d5f83d26aef70c674027a184a6b1e0098d33
llvm-svn: 363051
For lld, pass in Config->Timestamp (which is set based on lld's
/timestamp: and /Brepro flags). Since the writeWindowsResourceCOFF()
data is only used in-memory by LLD and the obj's timestamp isn't used
for anything in the output, this doesn't change behavior.
For llvm-cvtres, add an optional /timestamp: parameter, and use the
current behavior of calling time() if the parameter is not passed in.
This doesn't really change observable behavior (unless someone passes
/timestamp: to llvm-cvtres, which wasn't possible before), but it
removes the last unqualified call to time() from llvm/lib, which seems
like a good thing.
Differential Revision: https://reviews.llvm.org/D63116
llvm-svn: 363050
As suggested by @arsenm on D63075 - this adds a TargetLowering::allowsMemoryAccess wrapper that takes a Load/Store node's MachineMemOperand to handle the AddressSpace/Alignment arguments and will also implicitly handle the MachineMemOperand::Flags change in D63075.
llvm-svn: 363048
Summary:
Bug: https://bugs.llvm.org/show_bug.cgi?id=39024
The bug reports that a vectorized loop is stepped through 4 times and each step through the loop seemed to show a different path. I found two problems here:
A) An incorrect line number on a preheader block (for.body.preheader) instruction causes a step into the loop before it begins.
B) Instructions in the middle block have different line numbers which give the impression of another iteration.
In this patch I give all of the middle block instructions the line number of the scalar loop latch terminator branch. This seems to provide the smoothest debugging experience because the vectorized loops will always end on this line before dropping into the scalar loop. To solve problem A I have altered llvm::SplitBlockPredecessors to accommodate loop header blocks.
I have set up a separate review D61933 for a fix which is required for this patch.
Reviewers: samsonov, vsk, aprantl, probinson, anemet, hfinkel, jmorse
Reviewed By: hfinkel, jmorse
Subscribers: jmorse, javed.absar, eraman, kcc, bjope, jmellorcrummey, hfinkel, gbedwell, hiraditya, zzheng, llvm-commits
Tags: #llvm, #debug-info
Differential Revision: https://reviews.llvm.org/D60831
llvm-svn: 363046
The variable `OffsetMask` is currently only used in an assertion, so
if assertions are compiled out and -Werror is enabled, it becomes a
build failure.
llvm-svn: 363043
This adds support for the new family of conditional selection /
increment / negation instructions; the low-overhead branch
instructions (e.g. BF, WLS, DLS); the CLRM instruction to zero a whole
list of registers at once; the new VMRS/VMSR and VLDR/VSTR
instructions to get data in and out of 8.1-M system registers,
particularly including the new VPR register used by MVE vector
predication.
To support this, we also add a register name 'zr' (used by the CSEL
family to force one of the inputs to the constant 0), and operand
types for lists of registers that are also allowed to include APSR or
VPR (used by CLRM). The VLDR/VSTR instructions also need a new
addressing mode.
The low-overhead branch instructions exist in their own separate
architecture extension, which we treat as enabled by default, but you
can say -mattr=-lob or equivalent to turn it off.
Reviewers: dmgreen, samparker, SjoerdMeijer, t.p.northover
Reviewed By: samparker
Subscribers: miyuki, javed.absar, kristof.beyls, hiraditya, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D62667
llvm-svn: 363039
This patch changes how LLVM handles the accumulator/start value
in the reduction, by never ignoring it regardless of the presence of
fast-math flags on callsites. This change introduces the following
new intrinsics to replace the existing ones:
llvm.experimental.vector.reduce.fadd -> llvm.experimental.vector.reduce.v2.fadd
llvm.experimental.vector.reduce.fmul -> llvm.experimental.vector.reduce.v2.fmul
and adds functionality to auto-upgrade existing LLVM IR and bitcode.
Reviewers: RKSimon, greened, dmgreen, nikic, simoll, aemerson
Reviewed By: nikic
Differential Revision: https://reviews.llvm.org/D60261
llvm-svn: 363035
This reverts r362990 (git commit 374571301d)
This was causing linker warnings on Darwin:
ld: warning: direct access in function 'llvm::initializeEvexToVexInstPassPass(llvm::PassRegistry&)'
from file '../../lib/libLLVMX86CodeGen.a(X86EvexToVex.cpp.o)' to global weak symbol
'void std::__1::__call_once_proxy<std::__1::tuple<void* (&)(llvm::PassRegistry&),
std::__1::reference_wrapper<llvm::PassRegistry>&&> >(void*)' from file '../../lib/libLLVMCore.a(Verifier.cpp.o)'
means the weak symbol cannot be overridden at runtime. This was likely caused by different translation
units being compiled with different visibility settings.
llvm-svn: 363028
This makes the interface simpler and more consistent with the interface for
.dSYM files and fixes a bug where llvm-symbolizer would not read the dwp if
it was asked to symbolize data before symbolizing code.
Differential Revision: https://reviews.llvm.org/D63114
llvm-svn: 363025
And share some code with lld-link.
While here, also add a FIXME about PR42180 and merge r360150 to llvm-lib.
Differential Revision: https://reviews.llvm.org/D63021
llvm-svn: 363016
Returns "cortex-a73" for 3rd and 4th gen Kryo; not precisely correct,
but close enough.
Differential Revision: https://reviews.llvm.org/D63099
llvm-svn: 363013
An earlier fix of a subtle iterator invalidation bug had uncovered a
nondeterminism that was present in the MultiUsers bag. Problem was that
MultiUsers was being looked up using pointers.
This patch is an NFC change that numbers each multiuser and processes each in
numbered order. This fixes the test failure on netbsd and will likely fix the
green-dragon bot too.
llvm-svn: 363012
Previous detection relied upon an arbitrary hard coded limit of 21
response files, which some code bases were running up against.
The new detection maintains a stack of processing response files and
explicitly checks if a newly encountered file is in the current stack.
Some bookkeeping data is necessary in order to detect when to pop the
stack.
Patch by Chris Glover.
Differential Revision: https://reviews.llvm.org/D62798
llvm-svn: 363005
As shown in PR41279, some basic blocks (such as catchswitch) cannot be
instrumented. This patch filters out these BBs in PGO instrumentation.
It also sets the profile count to the fail-to-instrument edge, so that we
can propagate the counts in the CFG.
Differential Revision: https://reviews.llvm.org/D62700
llvm-svn: 362995
Summary:
For builds with LLVM_BUILD_LLVM_DYLIB=ON and BUILD_SHARED_LIBS=OFF
this change makes all symbols in the target specific libraries hidden
by default.
A new macro called LLVM_EXTERNAL_VISIBILITY has been added to mark symbols in these
libraries public, which is mainly needed for the definitions of the
LLVMInitialize* functions.
This patch reduces the number of public symbols in libLLVM.so by about
25%. This should improve load times for the dynamic library and also
make abi checker tools, like abidiff require less memory when analyzing
libLLVM.so
One side-effect of this change is that for builds with
LLVM_BUILD_LLVM_DYLIB=ON and LLVM_LINK_LLVM_DYLIB=ON some unittests that
access symbols that are no longer public will need to be statically linked.
Before and after public symbol counts (using gcc 8.2.1, ld.bfd 2.31.1):
nm before/libLLVM-9svn.so | grep ' [A-Zuvw] ' | wc -l
36221
nm after/libLLVM-9svn.so | grep ' [A-Zuvw] ' | wc -l
26278
Reviewers: chandlerc, beanz, mgorny, rnk, hans
Reviewed By: rnk, hans
Subscribers: Jim, hiraditya, michaelplatings, chapuni, jholewinski, arsenm, dschuff, jyknight, dylanmckay, sdardis, nemanjai, jvesely, nhaehnle, javed.absar, sbc100, jgravelle-google, aheejin, kbarton, fedor.sergeev, asb, rbar, johnrusso, simoncook, apazos, sabuasal, niosHD, jrtc27, zzheng, edward-jones, mgrang, atanasyan, rogfer01, MartinMosbeck, brucehoult, the_o, PkmX, jocewei, kristina, jsji, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D54439
llvm-svn: 362990
If the source is undef, then just don't do anything.
This matches SelectionDAG's behaviour in SelectionDAG.cpp.
Also add a test showing that we do the right thing here.
(irtranslator-memfunc-undef.ll)
Differential Revision: https://reviews.llvm.org/D63095
llvm-svn: 362989
This was discussed as part of D62880. The basic thought is that computing BE taken count after widening should produce (on average) an equally good backedge taken count as the one before widening. Since there's only one test in the suite which is impacted by this change, and it's essentially equivelent codegen, that seems to be a reasonable assertion. This change was separated from r362971 so that if this turns out to be problematic, the triggering piece is obvious and easily revertable.
For the nestedIV example from elim-extend.ll, we end up with the following BE counts:
BEFORE: (-2 + (-1 * %innercount) + %limit)
AFTER: (-1 + (sext i32 (-1 + %limit) to i64) + (-1 * (sext i32 %innercount to i64))<nsw>)
Note that before is an i32 type, and the after is an i64. Truncating the i64 produces the i32.
llvm-svn: 362975
This was found during HTM cleanup.
Adding a test for builtin_ttest would expose following issue.
*** Bad machine code: Illegal physical register for instruction ***
- function: test10
- basic block: %bb.0 entry (0xf0e57497b58)
- instruction: %5:crrc0 = TABORTWCI 0, $zero, 0
- operand 2: $zero
$zero is not a GPRC register.
LLVM ERROR: Found 1 machine code errors.
Differential Revision: https://reviews.llvm.org/D63079
llvm-svn: 362974
This change does the plumbing to wire an ExitingBB parameter through the LFTR implementation, and reorganizes the code to work in terms of a set of individual loop exits. Most of it is fairly obvious, but there's one key complexity which makes it worthy of consideration. The actual multi-exit LFTR patch is in D62625 for context.
Specifically, it turns out the existing code uses the backedge taken count from before a IV is widened. Oddly, we can end up with a different (more expensive, but semantically equivelent) BE count for the loop when requerying after widening. For the nestedIV example from elim-extend, we end up with the following BE counts:
BEFORE: (-2 + (-1 * %innercount) + %limit)
AFTER: (-1 + (sext i32 (-1 + %limit) to i64) + (-1 * (sext i32 %innercount to i64))<nsw>)
This is the only test in tree which seems sensitive to this difference. The actual result of using the wider BETC on this example is that we actually produce slightly better code. :)
In review, we decided to accept that test change. This patch is structured to preserve the old behavior, but a separate change will immediate follow with the behavior change. (I wanted it separate for problem attribution purposes.)
Differential Revision: https://reviews.llvm.org/D62880
llvm-svn: 362971
This behavior was added in r130928 for both FastISel and SD, and then
disabled in r131156 for FastISel.
This re-enables it for FastISel with the corresponding fix.
This is triggered only when FastISel can't lower the arguments and falls
back to SelectionDAG for it.
FastISel contains a map of "register fixups" where at the end of the
selection phase it replaces all uses of a register with another
register that FastISel sometimes pre-assigned. Code at the end of
SelectionDAGISel::runOnMachineFunction is doing the replacement at the
very end of the function, while other pieces that come in before that
look through the MachineFunction and assume everything is done. In this
case, the real issue is that the code emitting COPY instructions for the
liveins (physreg to vreg) (EmitLiveInCopies) is checking if the vreg
assigned to the physreg is used, and if it's not, it will skip the COPY.
If a register wasn't replaced with its assigned fixup yet, the copy will
be skipped and we'll end up with uses of undefined registers.
This fix moves the replacement of registers before the emission of
copies for the live-ins.
The initial motivation for this fix is to enable tail calls for
swiftself functions, which were blocked because we couldn't prove that
the swiftself argument (which is callee-save) comes from a function
argument (live-in), because there was an extra copy (vreg to vreg).
A few tests are affected by this:
* llvm/test/CodeGen/AArch64/swifterror.ll: we used to spill x21
(callee-save) but never reload it because it's attached to the return.
We now don't even spill it anymore.
* llvm/test/CodeGen/*/swiftself.ll: we tail-call now.
* llvm/test/CodeGen/AMDGPU/mubuf-legalize-operands.ll: I believe this
test was not really testing the right thing, but it worked because the
same registers were re-used.
* llvm/test/CodeGen/ARM/cmpxchg-O0.ll: regalloc changes
* llvm/test/CodeGen/ARM/swifterror.ll: get rid of a copy
* llvm/test/CodeGen/Mips/*: get rid of spills and copies
* llvm/test/CodeGen/SystemZ/swift-return.ll: smaller stack
* llvm/test/CodeGen/X86/atomic-unordered.ll: smaller stack
* llvm/test/CodeGen/X86/swifterror.ll: same as AArch64
* llvm/test/DebugInfo/X86/dbg-declare-arg.ll: stack size changed
Differential Revision: https://reviews.llvm.org/D62361
llvm-svn: 362963
These caused a build failure because I managed not to notice they
depended on a later unpushed commit in my current stack. Sorry about
that.
llvm-svn: 362956
This should have been part of r362953, but I had a finger-trouble
incident and committed the old rather than new version of the patch.
Sorry.
llvm-svn: 362955
This adds support for the new family of conditional selection /
increment / negation instructions; the low-overhead branch
instructions (e.g. BF, WLS, DLS); the CLRM instruction to zero a whole
list of registers at once; the new VMRS/VMSR and VLDR/VSTR
instructions to get data in and out of 8.1-M system registers,
particularly including the new VPR register used by MVE vector
predication.
To support this, we also add a register name 'zr' (used by the CSEL
family to force one of the inputs to the constant 0), and operand
types for lists of registers that are also allowed to include APSR or
VPR (used by CLRM). The VLDR/VSTR instructions also need some new
addressing modes.
The low-overhead branch instructions exist in their own separate
architecture extension, which we treat as enabled by default, but you
can say -mattr=-lob or equivalent to turn it off.
Reviewers: dmgreen, samparker, SjoerdMeijer, t.p.northover
Reviewed By: samparker
Subscribers: miyuki, javed.absar, kristof.beyls, hiraditya, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D62667
llvm-svn: 362953
This commit reapplies r359426 (which was reverted in r360301 due to
performance problems) and rolls in D61940 to address the performance problem.
I've combined the two to avoid creating a span of slow-performance, and to
ease reverting if more problems crop up.
The summary of D61940: This patch removes the "ChangingRegs" facility in
DbgEntityHistoryCalculator, as its overapproximate nature can produce incorrect
variable locations. An unchanging register doesn't mean a variable doesn't
change its location.
The patch kills off everything that calculates the ChangingRegs vector.
Previously ChangingRegs spotted epilogues and marked registers as unchanging if
they weren't modified outside the epilogue, increasing the chance that we can
emit a single-location variable record. Without this feature,
debug-loc-offset.mir and pr19307.mir become temporarily XFAIL. They'll be
re-enabled by D62314, using the FrameDestroy flag to identify epilogues, I've
split this into two steps as FrameDestroy isn't necessarily supported by all
backends.
The logic for terminating variable locations at the end of a basic block now
becomes much more enjoyably simple: we just terminate them all.
Other test changes: inlined-argument.ll becomes XFAIL, but for a longer term.
The current algorithm for detecting that a variable has a single-location
doesn't work in this scenario (inlined function in multiple blocks), only other
bugs were making this test work. fission-ranges.ll gets slightly refreshed too,
as the location of "p" is now correctly determined to be a single location.
Differential Revision: https://reviews.llvm.org/D61940
llvm-svn: 362951
Similar to rL362909:
This isn't the ideal fix (use FMF on the select), but it's still an
improvement until we have better FMF propagation to selects and other
FP math operators.
I don't think there's much risk of regression from this change by
not including the FMF on the fcmp any more. The nsz/nnan FMF
should be the same on the fcmp and the fsub because they have the
same operand.
llvm-svn: 362943
Arm v8.1-M supports the VMOV instructions that move a half-precision
value to and from a GPR, but not if the GPR is SP or PC.
To fix this, I've changed those instructions to use the rGPR register
class instead of GPR. rGPR always excludes PC, and it excludes SP
except in the presence of the HasV8Ops target feature (i.e. Arm v8-A).
So the effect is that VMOV.F16 to and from PC is now illegal
everywhere, but VMOV.F16 to and from SP is illegal only on non-v8-A
cores (which I believe is all as it should be).
Reviewers: dmgreen, samparker, SjoerdMeijer, ostannard
Reviewed By: ostannard
Subscribers: ostannard, javed.absar, kristof.beyls, hiraditya, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D60704
llvm-svn: 362942
This is a follow-up for D62809.
Content and Size fields should be optional as was discussed in comments
of the D62809's thread. With that, we can describe a specific string table and
symbol table sections in a more correct way and also show appropriate errors.
The patch adds lots of test cases where the behavior is described in details.
Differential revision: https://reviews.llvm.org/D62957
llvm-svn: 362931
This option allows loops with small max trip counts to be fully unrolled. This
can help with code like the remainder loops from manually unrolled loops like
those that appear in the cmsis dsp library. We would apparently previously
runtime unroll them with the default unroll count (4).
Differential Revision: https://reviews.llvm.org/D63064
llvm-svn: 362928
Variable's stack location can stretch longer than it should. If a
variable is placed at the stack in a some nested basic block its range
can be calculated to be up to the next occurrence of the variable's
DBG_VALUE, or up to the end of the function, thus covering a basic
blocks that should not be included in the variable’s location range.
This happens because the DbgEntityHistoryCalculator ends register
locations at the end of a basic block only if the variable’s location
register has been changed throughout the function, which is not the
case for the register used to reference stack objects.
This patch also tries to produce a single value location if the location
list builder managed to merge all the locations into one.
Reviewers: aprantl, dstenb, jmorse
Reviewed By: aprantl, dstenb, jmorse
Subscribers: djtodoro, ivanbaev, asowda
Tags: #debug-info
Differential Revision: https://reviews.llvm.org/D61600
llvm-svn: 362923
This opportunity is found from spec 2017 557.xz_r. And it is used by the sha encrypt/decrypt. See sha-2/sha512.c
static void store64(u64 x, unsigned char* y)
{
for(int i = 0; i != 8; ++i)
y[i] = (x >> ((7-i) * 8)) & 255;
}
static u64 load64(const unsigned char* y)
{
u64 res = 0;
for(int i = 0; i != 8; ++i)
res |= (u64)(y[i]) << ((7-i) * 8);
return res;
}
The load64 has been implemented by https://reviews.llvm.org/D26149
This patch is trying to implement the store pattern.
Match a pattern where a wide type scalar value is stored by several narrow
stores. Fold it into a single store or a BSWAP and a store if the targets
supports it.
Assuming little endian target:
i8 *p = ...
i32 val = ...
p[0] = (val >> 0) & 0xFF;
p[1] = (val >> 8) & 0xFF;
p[2] = (val >> 16) & 0xFF;
p[3] = (val >> 24) & 0xFF;
>
*((i32)p) = val;
i8 *p = ...
i32 val = ...
p[0] = (val >> 24) & 0xFF;
p[1] = (val >> 16) & 0xFF;
p[2] = (val >> 8) & 0xFF;
p[3] = (val >> 0) & 0xFF;
>
*((i32)p) = BSWAP(val);
Differential Revision: https://reviews.llvm.org/D62897
llvm-svn: 362921
Summary:
Our default behavior is to use sign_extend for signed comparisons and zero_extend for everything else. But for equality we have the freedom to use either extension. If we can prove the input has been truncated from something with enough sign bits, we can use sign_extend instead and let DAG combine optimize it out. A similar rule is used by type legalization in LegalizeIntegerTypes.
This gets rid of the movzx in PR42189. The immediate will still take 4 bytes instead of the 2 bytes plus 0x66 prefix a cmp di, 32767 would get, but it avoids a length changing prefix.
Reviewers: RKSimon, spatel, xbolva00
Reviewed By: xbolva00
Subscribers: hiraditya, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D63032
llvm-svn: 362920
Summary:
We can only use the memory form of cvtss2sd under optsize due to a partial register update. So previously we were emitting 2 instructions for extload when optimizing for speed. Also due to a late optimization in preprocessiseldag we had to handle (fpextend (loadf32)) under optsize.
This patch forces extload to expand so that it will always be in the (fpextend (loadf32)) form during isel. And when optimizing for speed we can just let each of those pieces select an instruction independently.
Reviewers: spatel, RKSimon
Reviewed By: RKSimon
Subscribers: hiraditya, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D62710
llvm-svn: 362919
Previously we did the equivalent operation in isel patterns with
COPY_TO_REGCLASS operations to transition. By inserting
scalar_to_vetors and extract_vector_elts before isel we can
allow each piece to be selected individually and accomplish the
same final result.
I ideally we'd use vector operations earlier in lowering/combine,
but that looks to be more difficult.
The scalar-fp-to-i64.ll changes are because we have a pattern for
using movlpd for store+extract_vector_elt. While an f64 store
uses movsd. The encoding sizes are the same.
llvm-svn: 362914
This isn't the ideal fix (use FMF on the select), but it's still an
improvement until we have better FMF propagation to selects and other
FP math operators.
I don't think there's much risk of regression from this change by
not including the FMF on the fcmp any more. The nsz/nnan FMF
should be the same on the fcmp and the fneg (fsub) because they
have the same operand.
This works around the most glaring FMF logical inconsistency cited
in PR38086:
https://bugs.llvm.org/show_bug.cgi?id=38086
llvm-svn: 362909
This is another step towards correcting our usage of fast-math-flags when applied on an fcmp.
In this case, we are checking for 'nnan' on the fcmp itself rather than the operand of
the fcmp. But I'm leaving that clause in until we're more confident that we can stop
relying on fcmp's FMF.
By using the more general "isKnownNeverNaN()", we gain a simplification shown on the
tests with 'uitofp' regardless of the FMF on the fcmp (uitofp never produces a NaN).
On the tests with 'fabs', we are now relying on the FMF for the call fabs instruction
in addition to the FMF on the fcmp.
This is a continuation of D62979 / rL362879.
llvm-svn: 362903
This is the second part of the commit fixing PR38917 (hoisting
partitially redundant machine instruction). Most of PRE (partitial
redundancy elimination) and CSE work is done on LLVM IR, but some of
redundancy arises during DAG legalization. Machine CSE is not enough
to deal with it. This simple PRE implementation works a little bit
intricately: it passes before CSE, looking for partitial redundancy
and transforming it to fully redundancy, anticipating that the next
CSE step will eliminate this created redundancy. If CSE doesn't
eliminate this, than created instruction will remain dead and eliminated
later by Remove Dead Machine Instructions pass.
The third part of the commit is supposed to refactor MachineCSE,
to make it more clear and to merge MachinePRE with MachineCSE,
so one need no rely on further Remove Dead pass to clear instrs
not eliminated by CSE.
First step: https://reviews.llvm.org/D54839
Fixes llvm.org/PR38917
This is fixed recommit of r361356 after PowerPC64 multistage build failure.
llvm-svn: 362901
Pointers that are in-bounds (either through dereferenceable_or_null or
thorough a getelementptr inbounds) cannot be captured with a comparison
against null. There is no way to construct a pointer that is still in
bounds but also NULL.
This helps safe languages that insert null checks before load/store
instructions. Without this patch, almost all pointers would be
considered captured even for simple loads. With this patch, an icmp with
null will not be seen as escaping as long as certain conditions are met.
There was a lot of discussion about this patch. See the Phabricator
thread for detals.
Differential Revision: https://reviews.llvm.org/D60047
llvm-svn: 362900
This is 1 step towards correcting our usage of fast-math-flags when applied on an fcmp.
In this case, we are checking for 'nnan' on the fcmp itself rather than the operand of
the fcmp. But I'm leaving that clause in until we're more confident that we can stop
relying on fcmp's FMF.
By using the more general "isKnownNeverNaN()", we gain a simplification shown on the
tests with 'uitofp' regardless of the FMF on the fcmp (uitofp never produces a NaN).
On the tests with 'fabs', we are now relying on the FMF for the call fabs instruction
in addition to the FMF on the fcmp.
I'll update the 'ult' case below here as a follow-up assuming no problems here.
Differential Revision: https://reviews.llvm.org/D62979
llvm-svn: 362879
Types such as float and i64's do not have legal loads in Thumb1, but will still
be loaded with a LDR (or potentially multiple LDR's). As such we can treat the
cost of addressing mode calculations the same as an i32 and get some optimisation
benefits.
Differential Revision: https://reviews.llvm.org/D62968
llvm-svn: 362874
Now with MVE being added, we can add the vector addressing mode costs for it.
These are generally imm7 multiplied by the size of the type being loaded /
stored.
Differential Revision: https://reviews.llvm.org/D62967
llvm-svn: 362873
The fp16 version of VLDR takes a imm8 multiplied by 2. This updates the costs
to account for those, and adds extra testing. It is dependant upon hasFPRegs16
as this is what the load/store instructions require.
Differential Revision: https://reviews.llvm.org/D62966
llvm-svn: 362872
We are starting to add an entirely separate vector architecture to the ARM
backend. To do that we need at least some separation between the existing NEON
and the new MVE code. This patch just goes through the Neon patterns and
ensures that they are predicated on HasNEON, giving MVE a stable place to start
from.
No tests yet as this is largely an NFC, and we don't have the other target that
will treat any of these intructions as legal.
Differential Revision: https://reviews.llvm.org/D62945
llvm-svn: 362870
This patch aims to reduce spilling and register moves by using the 3-address
versions of instructions per default instead of the 2-address equivalent
ones. It seems that both spilling and register moves are improved noticeably
generally.
Regalloc hints are passed to increase conversions to 2-address instructions
which are done in SystemZShortenInst.cpp (after regalloc).
Since the SystemZ reg/mem instructions are 2-address (dst and lhs regs are
the same), foldMemoryOperandImpl() can no longer trivially fold a spilled
source register since the reg/reg instruction is now 3-address. In order to
remedy this, new 3-address pseudo memory instructions are used to perform the
folding only when the dst and lhs virtual registers are known to be allocated
to the same physreg. In order to not let MachineCopyPropagation run and
change registers on these transformed instructions (making it 3-address), a
new target pass called SystemZPostRewrite.cpp is run just after
VirtRegRewriter, that immediately lowers the pseudo to a target instruction.
If it would have been possibe to insert a COPY instruction and change a
register operand (convert to 2-address) in foldMemoryOperandImpl() while
trusting that the caller (e.g. InlineSpiller) would update/repair the
involved LiveIntervals, the solution involving pseudo instructions would not
have been needed. This is perhaps a potential improvement (see Phabricator
post).
Common code changes:
* A new hook TargetPassConfig::addPostRewrite() is utilized to be able to run a
target pass immediately before MachineCopyPropagation.
* VirtRegMap is passed as an argument to foldMemoryOperand().
Review: Ulrich Weigand, Quentin Colombet
https://reviews.llvm.org/D60888
llvm-svn: 362868
In order for GlobalISel to re-use the significant amount of analysis and
optimization code in SDAG's switch lowering, we first have to extract it and
create an interface to be used by both frameworks.
No test changes as it's NFC.
Differential Revision: https://reviews.llvm.org/D62745
llvm-svn: 362857
Summary: Move some code around, in preparation for later fixes
to the non-integral addrspace handling (D59661)
Patch By Jameson Nash <jameson@juliacomputing.com>
Reviewed By: reames, loladiro
Differential Revision: https://reviews.llvm.org/D59729
llvm-svn: 362853
Summary:
The cleanup in D62751 introduced a compile-time regression due to the way DT updates are performed.
Add all insert edges then all delete edges in DTU to match the previous compile time.
Compile time on the test provided by @mstorsjo before and after this patch on my machine:
113.046s vs 35.649s
Repro: clang -target x86_64-w64-mingw32 -c -O3 glew-preproc.c; on https://martin.st/temp/glew-preproc.c.
Reviewers: kuhar, NutshellySima, mstorsjo
Subscribers: jlebar, mstorsjo, dmgreen, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D62981
llvm-svn: 362839
rather than two callbacks.
The asynchronous lookup API (which the synchronous lookup API wraps for
convenience) used to take two callbacks: OnResolved (called once all requested
symbols had an address assigned) and OnReady to be called once all requested
symbols were safe to access). This patch updates the asynchronous lookup API to
take a single 'OnComplete' callback and a required state (SymbolState) to
determine when the callback should be made. This simplifies the common use case
(where the client is interested in a specific state) and will generalize neatly
as new states are introduced to track runtime initialization of symbols.
Clients who were making use of both callbacks in a single query will now need to
issue two queries (one for SymbolState::Resolved and another for
SymbolState::Ready). Synchronous lookup API clients who were explicitly passing
the WaitOnReady argument will now need neeed to pass a SymbolState instead (for
'WaitOnReady == true' use SymbolState::Ready, for 'WaitOnReady == false' use
SymbolState::Resolved). Synchronous lookup API clients who were using default
arugment values should see no change.
llvm-svn: 362832
When we call checkResourceLimit in bumpCycle or bumpNode, and we
know the resource count has just reached the limit (the equations
are equal). We should return true to mark that we are resource
limited for next schedule, or else we might continue to schedule
in favor of latency for 1 more schedule and create a schedule that
actually overbook the resource.
When we call checkResourceLimit to estimate the resource limite before
scheduling, we don't need to return true even if the equations are
equal, as it shouldn't limit the schedule for it .
Differential Revision: https://reviews.llvm.org/D62345
llvm-svn: 362805
This could fail, which looked concerning. However nothing was actually
using the results of this. I assume this was intended to use the
anti-feature of analyzeBranch of removing instructions, but wasn't
actually calling it with AllowModify = true.
Fixes bug 42162.
llvm-svn: 362800
lib.exe doesn't allow creating .lib files with object files that have
differing machine types. Update llvm-lib to match.
The motivation is to make it possible to infer the machine type of a
.lib file in lld, so that it can warn when e.g. a 32-bit .lib file is
passed to a 64-bit link (PR38965).
Fixes PR38782.
Differential Revision: https://reviews.llvm.org/D62913
llvm-svn: 362798
This is a potentially large perf win for AVX1 targets because of the way we
auto-vectorize to 256-bit but then expect the backend to legalize/optimize
for the half-implemented AVX1 ISA.
On the motivating example from PR37428 (even though this patch doesn't solve
the vector shift issue):
https://bugs.llvm.org/show_bug.cgi?id=37428
...there's a 16% speedup when compiling with "-mavx" (perf tested on Haswell)
because we eliminate the remaining 256-bit vblendv ops.
I added comments on a couple of tests that require further work. If we have
256-bit logic ops separating the vselect and extract, we should probably narrow
everything to 128-bit, but that requires a larger pattern match.
Differential Revision: https://reviews.llvm.org/D62969
llvm-svn: 362797
Change D60691 caused some knock-on failures that weren't caught by the
existing tests. Firstly, selecting a CPU that should have had a
restricted FPU (e.g. `-mcpu=cortex-m4`, which should have 16 d-regs
and no double precision) could give the unrestricted version, because
`ARM::getFPUFeatures` returned a list of features including subtracted
ones (here `-fp64`,`-d32`), but `ARMTargetInfo::initFeatureMap` threw
away all the ones that didn't start with `+`. Secondly, the
preprocessor macros didn't reliably match the actual compilation
settings: for example, `-mfpu=softvfp` could still set `__ARM_FP` as
if hardware FP was available, because the list of features on the cc1
command line would include things like `+vfp4`,`-vfp4d16` and clang
didn't realise that one of those cancelled out the other.
I've fixed both of these issues by rewriting `ARM::getFPUFeatures` so
that it returns a list that enables every FP-related feature
compatible with the selected FPU and disables every feature not
compatible, which is more verbose but means clang doesn't have to
understand the dependency relationships between the backend features.
Meanwhile, `ARMTargetInfo::handleTargetFeatures` is testing for all
the various forms of the FP feature names, so that it won't miss cases
where it should have set `HW_FP` to feed into feature test macros.
That in turn caused an ordering problem when handling `-mcpu=foo+bar`
together with `-mfpu=something_that_turns_off_bar`. To fix that, I've
arranged that the `+bar` suffixes on the end of `-mcpu` and `-march`
cause feature names to be put into a separate vector which is
concatenated after the output of `getFPUFeatures`.
Another side effect of all this is to fix a bug where `clang -target
armv8-eabi` by itself would fail to set `__ARM_FEATURE_FMA`, even
though `armv8` (aka Arm v8-A) implies FP-Armv8 which has FMA. That was
because `HW_FP` was being set to a value including only the `FPARMV8`
bit, but that feature test macro was testing only the `VFP4FPU` bit.
Now `HW_FP` ends up with all the bits set, so it gives the right
answer.
Changes to tests included in this patch:
* `arm-target-features.c`: I had to change basically all the expected
results. (The Cortex-M4 test in there should function as a
regression test for the accidental double-precision bug.)
* `arm-mfpu.c`, `armv8.1m.main.c`: switched to using `CHECK-DAG`
everywhere so that those tests are no longer sensitive to the order
of cc1 feature options on the command line.
* `arm-acle-6.5.c`: been updated to expect the right answer to that
FMA test.
* `Preprocessor/arm-target-features.c`: added a regression test for
the `mfpu=softvfp` issue.
Reviewers: SjoerdMeijer, dmgreen, ostannard, samparker, JamesNagurne
Reviewed By: ostannard
Subscribers: srhines, javed.absar, kristof.beyls, hiraditya, cfe-commits, llvm-commits
Tags: #clang, #llvm
Differential Revision: https://reviews.llvm.org/D62998
llvm-svn: 362791
Summary:
This allows some integer bitwise operations to instead be performed by
hardware fp instructions. This is correct because the RISC-V spec
requires the F and D extensions to use the IEEE-754 standard
representation, and fp register loads and stores to be bit-preserving.
This is tested against the soft-float ABI, but with hardware float
extensions enabled, so that the tests also ensure the optimisation also
fires in this case.
Reviewers: asb, luismarques
Reviewed By: asb
Subscribers: hiraditya, rbar, johnrusso, simoncook, apazos, sabuasal, niosHD, kito-cheng, shiva0217, jrtc27, zzheng, edward-jones, rogfer01, MartinMosbeck, brucehoult, the_o, rkruppe, PkmX, jocewei, psnobl, benna, Jim, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D62900
llvm-svn: 362790
Summary:
This patch fixes a bug in the assembler that permitted a type suffix on
predicate registers when not expected. For instance, the following was
previously valid:
faddv h0, p0.q, z1.h
This bug was present in all SVE instructions containing predicates with
no type suffix and no predication form qualifier, i.e. /z or /m. The
latter instructions are already caught with an appropiate error message
by the assembler, e.g.:
.text
<stdin>:1:13: error: not expecting size suffix
cmpne p1.s, p0.b/z, z2.s, 0
^
A similar issue for SVE vector registers was fixed in:
https://reviews.llvm.org/D59636
Reviewed By: SjoerdMeijer
Differential Revision: https://reviews.llvm.org/D62942
llvm-svn: 362780
Patch which introduces a target-independent framework for generating
hardware loops at the IR level. Most of the code has been taken from
PowerPC CTRLoops and PowerPC has been ported over to use this generic
pass. The target dependent parts have been moved into
TargetTransformInfo, via isHardwareLoopProfitable, with
HardwareLoopInfo introduced to transfer information from the backend.
Three generic intrinsics have been introduced:
- void @llvm.set_loop_iterations
Takes as a single operand, the number of iterations to be executed.
- i1 @llvm.loop_decrement(anyint)
Takes the maximum number of elements processed in an iteration of
the loop body and subtracts this from the total count. Returns
false when the loop should exit.
- anyint @llvm.loop_decrement_reg(anyint, anyint)
Takes the number of elements remaining to be processed as well as
the maximum numbe of elements processed in an iteration of the loop
body. Returns the updated number of elements remaining.
llvm-svn: 362774
In r356860, the legalization logic for BSWAP was modified to ISD::ROTL,
rather than the old ISD::{SHL, SRL, OR} nodes.
This works fine on AVR for 8-bit rotations, but 16-bit rotations are
currently unimplemented - they always trigger an assertion error in the
AVRExpandPseudoInsts pass ("RORW unimplemented").
This patch instructions the legalizer to expand 16-bit rotations into
the previous SHL, SRL, OR pattern it did previously.
This fixes the 'issue-cannot-select-bswap.ll' test. Interestingly, this
test failure seems flaky - it passes successfully on the avr-build-01
buildbot, but fails locally on my Arch Linux install.
llvm-svn: 362773
We should keep the symbol type (STT_GNU_IFUNC) for a local ifunc because
it may result in an IRELATIVE reloc that the dynamic loader will use to
resolve the address at startup time.
There is another problem that is not fixed by this patch: a PC relative
relocation should also create a relocation with the ifunc symbol.
llvm-svn: 362767
Use the PPC vector min/max instructions for computing the corresponding
operation as these should be faster than the compare/select sequences
we currently emit.
Differential revision: https://reviews.llvm.org/D47332
llvm-svn: 362759
SIInsertSkips really doesn't understand the control flow, and makes
very stupid assumptions about the block layout. This was able to get
away with not skipping return blocks, since usually after
structurization there is only one placed at the end of the
function. Tail duplication can break this assumption.
llvm-svn: 362754
Incorrect Debug Variable Range was calculated while "COMPUTING LIVE DEBUG VARIABLES" stage.
Range for Debug Variable("i") computed according to current state of instructions
inside of basic block. But Register Allocator creates new instructions which were not taken
into account when Live Debug Variables computed. In the result DBG_VALUE instruction for
the "i" variable was put after these newly inserted instructions. This is incorrect.
Debug Value for the loop counter should be inserted before any loop instruction.
Differential Revision: https://reviews.llvm.org/D62650
llvm-svn: 362750
"Divergence driven ISel. Assign register class for cross block values
according to the divergence."
that discovered the design flaw leading to several issues that
required to be solved before.
This change reverts AMDGPU specific changes and keeps common part
unaffected.
llvm-svn: 362749
This primarily affects add/fadd/mul/fmul/and/or/xor/pmuludq/pmuldq/max/min/fmaxc/fminc/pmaddwd/pavg.
We already commuted the unmasked and zero masked versions.
I've added 512-bit stack folding tests for most of the instructions
affected. I've tested needing commuting and not commuting across
unmasked, merged masked, and zero masked. The 128/256 bit instructions
should behave similarly.
llvm-svn: 362746
A function for loop vectorization illegality reporting has been
introduced:
void LoopVectorizationLegality::reportVectorizationFailure(
const StringRef DebugMsg, const StringRef OREMsg,
const StringRef ORETag, Instruction * const I) const;
The function prints a debug message when the debug for the compilation
unit is enabled as well as invokes the optimization report emitter to
generate a message with a specified tag. The function doesn't cover any
complicated logic when a custom lambda should be passed to the emitter,
only generating a message with a tag is supported.
The function always prints the instruction `I` after the debug message
whenever the instruction is specified, otherwise the debug message
ends with a dot: 'LV: Not vectorizing: Disabled/already vectorized.'
Patch by Pavel Samolysov <samolisov@gmail.com>
llvm-svn: 362736
Summary:
(1) Function descriptor on AIX
On AIX, a called routine may have 2 distinct symbols associated with it:
* A function descriptor (Name)
* A function entry point (.Name)
The descriptor structure on AIX is the same as those in the ELF V1 ABI:
* The address of the entry point of the function.
* The TOC base address for the function.
* The environment pointer.
The descriptor symbol uses the same name as the source level function in C.
The function entry point is analogous to the symbol we would generate for a
function in a non-descriptor-based ABI, except that it is renamed by
prepending a ".".
Which symbol gets referenced depends on the context:
* Taking the address of the function references the descriptor symbol.
* Calling the function references the entry point symbol.
(2) Speaking of implementation on AIX, for direct function call target, we
create proper MCSymbol SDNode(e.g . ".foo") while constructing SDAG to
replace original TargetGlobalAddress SDNode. Then down the path, we can
take advantage of this MCSymbol.
Patch by: Xiangling_L
Reviewed by: sfertile, hubert.reinterpretcast, jasonliu, syzaara
Differential Revision: https://reviews.llvm.org/D62532
llvm-svn: 362735
This adds support for unary fneg based on the implementation of BinaryOperator without the soft float FP cost.
Previously we would just delegate to visitUnaryInstruction. I think the only real change is that we will pass the FastMath flags to SimplifyFNeg now.
Differential Revision: https://reviews.llvm.org/D62699
llvm-svn: 362732
This is a really silly bug that even a simple test w/an unconditional latch would have caught. I tried to guard against the case, but put it in the wrong if check. Oops.
llvm-svn: 362727
This patch is the first step towards ensuring MergeConsecutiveStores correctly handles non-temporal loads\stores:
1 - When merging load\stores we must ensure that they all have the same non-temporal flag. This is unlikely to occur, but can in strange cases where we're storing at the end of one page and the beginning of another.
2 - The merged load\store node must retain the non-temporal flag.
Differential Revision: https://reviews.llvm.org/D62910
llvm-svn: 362723
Summary: Dependence Analysis performs static checks to confirm validity
of delinearization. These checks often fail for 64-bit targets due to
type conversions and integer wrapping that prevent simplification of the
SCEV expressions. These checks would also fail at compile-time if the
lower bound of the loops are compile-time unknown.
For example:
void foo(int n, int m, int a[][m]) {
for (int i = 0; i < n; ++i)
for (int j = 0; j < m; ++j) {
a[i][j] = a[i+1][j-2];
}
}
opt -mem2reg -instcombine -indvars -loop-simplify -loop-rotate -inline
-pass-remarks=.* -debug-pass=Arguments
-da-permissive-validity-checks=false k3.ll -analyze -da
will produce the following by default:
da analyze - anti [* *|<]!
but will produce the following expected dependence vector if the
validity checks are disabled:
da analyze - consistent anti [1 -2]!
This revision will introduce a debug option that will leave the validity
checks in place by default, but allow them to be turned off. New tests
are added for cases where it cannot be proven at compile-time that the
individual subscripts stay in-bound with respect to a particular
dimension of an array. These tests enable the option to provide user
guarantee that the subscripts do not over/under-flow into other
dimensions, thereby producing more accurate dependence vectors.
For prior discussion on this topic, leading to this change, please see
the following thread:
http://lists.llvm.org/pipermail/llvm-dev/2019-May/132372.html
Reviewers: Meinersbur, jdoerfert, kbarton, dmgreen, fhahn
Reviewed By: Meinersbur, jdoerfert, dmgreen
Subscribers: fhahn, hiraditya, javed.absar, llvm-commits, Whitney,
etiotto
Tag: LLVM
Differential Revision: https://reviews.llvm.org/D62610
llvm-svn: 362711
Summary:
This patch implements SDAG call lowering on AIX for functions
which only have parameters that could fit into GPRs.
Reviewers: hubert.reinterpretcast, syzaara
Differential Revision: https://reviews.llvm.org/D62823
llvm-svn: 362708
Summary:
This patch is part of a patch series to add support for FileCheck
numeric expressions. This specific patch introduces support for defining
numeric variable in a CHECK directive.
This commit introduces support for defining numeric variable from a
litteral value in the input text. Numeric expressions can then use the
variable provided it is on a later line.
Copyright:
- Linaro (changes up to diff 183612 of revision D55940)
- GraphCore (changes in later versions of revision D55940 and
in new revision created off D55940)
Reviewers: jhenderson, chandlerc, jdenny, probinson, grimar, arichardson, rnk
Subscribers: hiraditya, llvm-commits, probinson, dblaikie, grimar, arichardson, tra, rnk, kristina, hfinkel, rogfer01, JonChesterfield
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D60386
llvm-svn: 362705
This patch is a follow up for D62018 to add lrint/llrint
support for float16.
Reviewed By: SjoerdMeijer
Differential Revision: https://reviews.llvm.org/D62863
llvm-svn: 362700
This patch is a follow up for D61391 to add lround/llround
support for float16.
Reviewed By: SjoerdMeijer
Differential Revision: https://reviews.llvm.org/D62861
llvm-svn: 362698
If the given SCEVExpr has no (un)signed flags attached to it, transfer
these to the resulting instruction or use them to find an existing
instruction.
Differential Revision: https://reviews.llvm.org/D61934
llvm-svn: 362687
We already get support for G_ZEXTLOAD to s32 from the importer, but it can't
deal with the SUBREG_TO_REG in the pattern. Tweaking the existing manual
selection code for G_LOAD to handle an additional SUBREG_TO_REG when dealing
with G_ZEXTLOAD isn't much work.
Also add tests to check the imported pattern selections to s32 work.
llvm-svn: 362681
This is intended to enable the use of an immediate blend or
more optimal instruction. But if the passthru is zero we don't
need any additional instructions.
llvm-svn: 362675
When looking through copies, make sure to not try to find the vreg def of a physreg.
Normally getVRegDef will return nullptr in this case, but if there happens to be
multiple defs then it will assert.
This fixes PR42129.
llvm-svn: 362666
This forced the caller to be aware of this, which is an ugly ABI
feature.
Partially reverts r295877. The original reasons for doing this are
mostly fixed. Alloca is now in a non-0 address space, so it should be
OK to have 0 as a valid pointer. Since we treat the absolute address
as the pointer value, this part only really needed to apply to
kernels.
Since r357093, we avoid the need to increment/decrement the offset
register in more cases, and since r354816 the scavenger can fail
without spilling, so it's less critical that we try to avoid an offset
that fits in the MUBUF offset.
Restrict to callable functions for now to split this into 2 steps to
limit thte number of test updates and in case anything breaks.
llvm-svn: 362665
The ISD::STRICT_ nodes used to implement the constrained floating-point
intrinsics are currently never passed to the target back-end, which makes
it impossible to handle them correctly (e.g. mark instructions are depending
on a floating-point status and control register, or mark instructions as
possibly trapping).
This patch allows the target to use setOperationAction to switch the action
on ISD::STRICT_ nodes to Legal. If this is done, the SelectionDAG common code
will stop converting the STRICT nodes to regular floating-point nodes, but
instead pass the STRICT nodes to the target using normal SelectionDAG
matching rules.
To avoid having the back-end duplicate all the floating-point instruction
patterns to handle both strict and non-strict variants, we make the MI
codegen explicitly aware of the floating-point exceptions by introducing
two new concepts:
- A new MCID flag "mayRaiseFPException" that the target should set on any
instruction that possibly can raise FP exception according to the
architecture definition.
- A new MI flag FPExcept that CodeGen/SelectionDAG will set on any MI
instruction resulting from expansion of any constrained FP intrinsic.
Any MI instruction that is *both* marked as mayRaiseFPException *and*
FPExcept then needs to be considered as raising exceptions by MI-level
codegen (e.g. scheduling).
Setting those two new flags is straightforward. The mayRaiseFPException
flag is simply set via TableGen by marking all relevant instruction
patterns in the .td files.
The FPExcept flag is set in SDNodeFlags when creating the STRICT_ nodes
in the SelectionDAG, and gets inherited in the MachineSDNode nodes created
from it during instruction selection. The flag is then transfered to an
MIFlag when creating the MI from the MachineSDNode. This is handled just
like fast-math flags like no-nans are handled today.
This patch includes both common code changes required to implement the
new features, and the SystemZ implementation.
Reviewed By: andrew.w.kaylor
Differential Revision: https://reviews.llvm.org/D55506
llvm-svn: 362663
Since the beginning, the offset of a frame index has been consistently
interpreted backwards. It was treating it as an offset from the
scratch wave offset register as a frame register. The correct
interpretation is the offset from the SP on entry to the function,
before the prolog. Frame index elimination then should select either
SP or another register as an FP.
Treat the scratch wave offset on kernel entry as the pre-incremented
SP. Rely more heavily on the standard hasFP and frame pointer
elimination logic, and clean up the private reservation code. This
saves a copy in most callee functions.
The kernel prolog emission code is still kind of a mess relying on
checking the uses of physical registers, which I would prefer to
eliminate.
Currently selection directly emits MUBUF instructions, which require
using a reference to some register. Use the register chosen for SP,
and then ignore this later. This should probably be cleaned up to use
pseudos that don't refer to any specific base register until frame
index elimination.
Add a workaround for shaders using large numbers of SGPRs. I'm not
sure these cases were ever working correctly, since as far as I can
tell the logic for figuring out which SGPR is the scratch wave offset
doesn't match up with the shader input initialization in the shader
programming guide.
llvm-svn: 362661
Summary:
This change only unifies the API previous API pair accepting
CallInst and InvokeInst, thus making it easier to refactor
inliner pass ode to CallBase. The implementation of the unified
API still relies on the CallSite implementation.
Reviewers: eraman, chandlerc, jdoerfert
Reviewed By: jdoerfert
Subscribers: jdoerfert, hiraditya, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D62283
llvm-svn: 362656
The AllConstant check needs to be moved out of the if/else if chain to
avoid a test regression. The "there is no SimplifyZExt" comment
puzzles me, since there is SimplifyCastInst. Additionally, the
Simplify* calls seem to not see the operand as constant, so this needs
to be tried if the simplify failed.
llvm-svn: 362653
One of the sources controls the pass through value for the upper bits
of the result so we can't really commute it.
In practice this problem isn't a functional issue because we would
only try to commute this instruction in order to fold a load. But
we can't do embedded rounding and fold a load at the same time. So
the load fold would never succeed so I don't think we would ever
commute or at least keep the version after commuting.
llvm-svn: 362647
When the byval attribute has a type, it must match the pointee type of
any parameter; but InstCombine was not updating the attribute when
folding casts of various kinds away.
llvm-svn: 362643
Most parts of LLVM don't care whether the byval type is derived from an
explicit Attribute or from the parameter's pointee type, so it makes
sense for the main access function to just return the right value.
The very few users who do care (only BitcodeReader so far) can find out
how it's specified by accessing the Attribute directly.
llvm-svn: 362642
The current PIC support currently only works with Emscripten, so
disable it for other targets.
This is the PIC portion of https://reviews.llvm.org/D62542.
Reviewed By: dschuff, sbc100
llvm-svn: 362638
As far as I know these should be freely reassociatable just like
the floating point MAXC/MINC instructions.
The *reduce* test changes are largely regressions and caused by
the "generic" CPU we default to not having a scheduler model.
The machine-combiner-int-vec.ll test shows the positive benefits
of this change.
Differential Revision: https://reviews.llvm.org/D62787
llvm-svn: 362629
As suggested in D62498 - collectConcatOps() matches both
concat_vectors and insert_subvector patterns, and we see
more test improvements by using the more general match.
llvm-svn: 362620
This patch fixes a regression caused by the operand reordering refactoring patch https://reviews.llvm.org/D59973 .
The fix changes the strategy to Splat instead of Opcode, if broadcast opportunities are found.
Please see the lit test for some examples.
Committed on behalf of @vporpo (Vasileios Porpodas)
Differential Revision: https://reviews.llvm.org/D62427
llvm-svn: 362613
Instead of passing around fast-math-flags as a parameter, we can set those
using an IRBuilder guard object. This is no-functional-change-intended.
The motivation is to eventually fix the vectorizers to use and set the
correct fast-math-flags for reductions. Examples of that not behaving as
expected are:
https://bugs.llvm.org/show_bug.cgi?id=23116 (should be able to reduce with less than 'fast')
https://bugs.llvm.org/show_bug.cgi?id=35538 (possible miscompile for -0.0)
D61802 (should be able to reduce with IR-level FMF)
Differential Revision: https://reviews.llvm.org/D62272
llvm-svn: 362612
This is the LLVM part of this change, the Clang part contains the full
description in its commit message.
Differential Revision: https://reviews.llvm.org/D60697
llvm-svn: 362600
We already handle the case where we combine shuffle(extractsubvector(x),extractsubvector(x)), this relaxes the requirement to permit different sources as long as they have the same value type.
This causes a couple of cases where the VPERMV3 binary shuffles occur at a wider width than before, which I intend to improve in future commits - but as only the subvector's mask indices are defined, these will broadcast so we don't see any increase in constant size.
llvm-svn: 362599
@jdoerfert Looks like these are placeholders for incoming abstract attributes patches so I've just commented the code out, even though this is usually frowned upon.
llvm-svn: 362592
This reverts commit 5b32f60ec3.
The fix is in commit 4f9e68148b.
This patch fixes the CorrelatedValuePropagation pass to keep
prof branch_weights metadata of SwitchInst consistent.
It makes use of SwitchInstProfUpdateWrapper.
New tests are added.
Reviewed By: nikic
Differential Revision: https://reviews.llvm.org/D62126
llvm-svn: 362583
NOTE: Note that no attributes are derived yet. This patch will not go in
alone but only with others that derive attributes. The framework is
split for review purposes.
This commit introduces the Attributor pass infrastructure and fixpoint
iteration framework. Further patches will introduce abstract attributes
into this framework.
In a nutshell, the Attributor will update instances of abstract
arguments until a fixpoint, or a "timeout", is reached. Communication
between the Attributor and the abstract attributes that are derived is
restricted to the AbstractState and AbstractAttribute interfaces.
Please see the file comment in Attributor.h for detailed information
including design decisions and typical use case. Also consider the class
documentation for Attributor, AbstractState, and AbstractAttribute.
Reviewers: chandlerc, homerdin, hfinkel, fedor.sergeev, sanjoy, spatel, nlopes, nicholas, reames
Subscribers: mehdi_amini, mgorny, hiraditya, bollu, steven_wu, dexonsmith, dang, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D59918
llvm-svn: 362578
Generally speaking, we lower to an optimal rotate sequence for nodes visible in
the SDAG. However, there are instances where the two rotates are not visible at
ISEL time - most notably those in a very common sequence when lowering switch
statements to jump tables.
A common situation is a switch on a 32-bit integer. This value has to have the
upper 32 bits cleared and because jump table offsets are word offsets, the value
needs to be shifted left by 2 bits. We currently emit the clear and the left
shift as two separate instructions, but this is not needed as we can lower it to
a single RLDIC.
This patch just cleans that up.
Differential revision: https://reviews.llvm.org/D60402
llvm-svn: 362576
In glibc, DT_PPC_GOT indicates that PowerPC32 Secure PLT ABI is used.
I plan to use it in D62464.
DT_PPC_OPT currently indicates if a TLSDESC inspired TLS optimization is
enabled.
Reviewed By: grimar, jhenderson, rupprecht
Differential Revision: https://reviews.llvm.org/D62851
llvm-svn: 362569
-Use early returns to reduce indentation
-Replace multipe ifs with a switch.
-Replace an assert with an llvm_unreachable default in the switch.
-Check that the FP type we're going to use for the
X86ISD::FAND/FOR/FXOR is legal rather than checking that the
integer type matches the width of a legal scalar fp type. This all
runs after legalization so it shouldn't really matter, but making
sure we're using a valid type in the X86ISD node is really
whats important.
llvm-svn: 362565
Although we had the support in the prelegalizer combiner to generate the
G_SEXTLOAD or G_ZEXTLOAD ops, the legalizer definitions for arm64 had them as
lowering back to separate ops.
llvm-svn: 362553
Summary:
Adjusts the index and adds a bitcast around the vector operand of
EXTRACT_VECTOR_ELT so that its lane type matches the source type of
its parent sext_inreg. Without this bitcast the ISel patterns do not
match and ISel fails.
Reviewers: aheejin
Subscribers: dschuff, sbc100, jgravelle-google, hiraditya, sunfish, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D62646
llvm-svn: 362547
Summary:
An argument that is return by a function but bit-casted before can still
be annotated as "returned". Make sure we do not crash for this case.
Reviewers: sunfish, stephenwlin, niravd, arsenm
Subscribers: wdng, hiraditya, bollu, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D59917
llvm-svn: 362546
This patch allows current users of Value::stripPointerCasts() to force
the result of the function to have the same representation as the value
it was called on. This is useful in various cases, e.g., (non-)null
checks.
In this patch only a single call site was adjusted to fix an existing
misuse that would cause nonnull where they may be wrong. Uses in
attribute deduction and other areas, e.g., D60047, are to be expected.
For a discussion on this topic, please see [0].
[0] http://lists.llvm.org/pipermail/llvm-dev/2018-December/128423.html
Reviewers: hfinkel, arsenm, reames
Subscribers: wdng, hiraditya, bollu, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D61607
llvm-svn: 362545
- The loop in demangleFunctionParameterList() only exits
on Error, @, and Z. All 3 cases were handled, so the
rest of the function is DEMANGLE_UNREACHABLE.
- The loop in demangleTemplateParameterList() always returns
on Error, so there's no need to check for that in the loop
header and after the loop.
- Add test cases for invalid function parameter manglings.
- Add a (redundant) test case for a simple template parameter
list mangling.
- Add a test case pointing out that varargs functions aren't
demangled correctly.
llvm-svn: 362540
Summary:
Following the cleanup in D48202, method foldBlockIntoPredecessor has the
same behavior. Replace its uses with MergeBlockIntoPredecessor.
Remove foldBlockIntoPredecessor.
Reviewers: chandlerc, dmgreen
Subscribers: jlebar, javed.absar, zzheng, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D62751
llvm-svn: 362538
We already need to have patterns for X86ISD::RNDSCALE to support software intrinsics. But we currently have 5 sets of patterns for the 5 rounding operations. For of these 6 patterns we have to support 3 vectors widths, 2 element sizes, sse/vex/evex encodings, load folding, and broadcast load folding. This results in a fair amount of bytes in the isel table.
This patch adds code to PreProcessIselDAG to morph the fceil/ffloor/ftrunc/fnearbyint/frint to X86ISD::RNDSCALE. This way we can remove everything, but the intrinsic pattern while still allowing the operations to be considered Legal for DAGCombine and Legalization. This shrinks the DAGISel by somewhere between 9K and 10K.
There is one complication to this, the STRICT versions of these nodes are currently mutated to their none strict equivalents at isel time when the node is visited. This won't be true in the future since that loses the chain ordering information. For now I've also added support for the non-STRICT nodes to Select so we can change the STRICT versions there after they've been mutated to their non-STRICT versions. We'll probably need a STRICT version of RNDSCALE or something to handle this in the future. Which will take us back to needing 2 sets of patterns for strict and non-strict, but that's still better than the 11 or 12 sets of patterns we'd need.
We can probably do something similar for scalar, but I haven't looked at it yet.
Differential Revision: https://reviews.llvm.org/D62757
llvm-svn: 362535
This is a special case of a more general transform (not (sub Y, X)) -> (add X, ~Y). InstCombine knows the general form. I've restricted to the special case to fix the motivating case PR42118. I tried handling any case where Y was constant, but got some changes on some Mips tests that I couldn't quickly prove where beneficial.
Fixes PR42118
Differential Revision: https://reviews.llvm.org/D62828
llvm-svn: 362533
This shows up as a side issue to the main problem for the AVX target example from PR37428:
https://bugs.llvm.org/show_bug.cgi?id=37428 - https://godbolt.org/z/7tpRa3
But as we can see in the pile of existing test diffs, it's actually a widespread problem
that affects any AVX or later target. Apart from a couple of oddballs, I think these are
all improvements for the reasons stated in the code comment: we do not want to enable YMM
unnecessarily (avoid vzeroupper and frequency throttling) and some cores split 256-bit
stores anyway.
We could say that MergeConsecutiveStores() is going overboard on some of these examples,
but that won't solve the problem completely. But that is a reason I'm proposing this as
a lowering rather than a combine: we will infinite loop fighting the merge code if we try
this earlier.
Differential Revision: https://reviews.llvm.org/D62498
llvm-svn: 362524
Arm Architecture v8.5a introduces Branch Target Identification (BTI). When
enabled all indirect branches must target a bti instruction of the
appropriate form. As PLT sequences may sometimes be the target of an
indirect branch and PLT[0] always is, a static linker may need to generate
PLT sequences that contain "bti c" as the first instruction. In effect:
bti c
adrp x16, page offset to .got.plt
...
Instead of:
adrp x16, page offset to .got.plt
...
At present the PLT decoding assumes the adrp will always be the first
instruction. This patch adds support for a single "bti c" to prefix it. A
test binary has been uploaded with such a PLT sequence. A forthcoming LLD
patch will make heavy use of the PLT decoding code.
Differential Revision: https://reviews.llvm.org/D62598
llvm-svn: 362523
- For error returns in demangleSpecialTableNode(),
demangleLocalStaticGuard(), RTTITypeDescriptor,
demangleRttiBaseClassDescriptorNode(), demangleUnsigned(),
demangleUntypedVariable() (via RttiBaseClassArray)
- For ?_A and ?_P which are handled at early levels of the
demangler but are not implemented in a later stage; this
is now more obvious
- Replace a "default:" with an explicit list of cases, to
get -Wswitch check we list all cases
llvm-svn: 362520
The underlying ConstantRange functionality has been added in D60952,
D61207 and D61238, this just exposes it for LVI.
I'm switching the code from using a whitelist to a blacklist, as
we're down to one unsupported operation here (xor) and writing it
this way seems more obvious :)
Differential Revision: https://reviews.llvm.org/D62822
llvm-svn: 362519
- Add test coverage around invalid anon namespaces and
for error paths in demanglePrimitiveType() and in
demangleFullyQualifiedTypeName()
- Use DEMANGLE_UNREACHABLE in two more unreachable places
llvm-svn: 362514
This is to address some of the problems in existing P9 resource modeling,
especially about the dispatching rules.
Instead of using a hypothetical DISPATCHER , we try to use the number of
actual dispatch slots, and define SchedWriteRes to model dispatch rules,
then update instruction classes according to dispatch rules.
All the dispatch rules and instruction classes update are made according
to POWER9 User Manual.
Differential Revision: https://reviews.llvm.org/D61873
llvm-svn: 362509
The proposal in D62498 showed that x86 would benefit from vector
store splitting, but that may conflict with the generic DAG
combiner's store merging transforms.
Add memory type to the existing TLI hook that enables the merging
transforms, so we can limit those changes to scalars only for x86.
llvm-svn: 362507
- Replace `Error = true` in a few branches that are truly unreachable
with DEMANGLE_UNREACHABLE
- Remove early return early in startsWithLocalScopePattern() because
it's redundant with the next two early returns
- Remove unreachable `case '0'` (it's handled in the branch below)
- Remove an unused bool return
- Add test coverage for several early error returns, mostly in
array type parsing
llvm-svn: 362506
As discussed on D62777 - we should be able to use this in more SSE41+ cases as well but that requires us to separate it from the OR(AND(),ANDN()) matcher.
llvm-svn: 362504
This matches APInt's versions of these functions, and there is no need for these to be size_t.
(as well as __builtin_clzll())
Differential Revision: https://reviews.llvm.org/D60823
llvm-svn: 362503
ELF for the 64-bit Arm Architecture defines two processor-specific dynamic
tags:
DT_AARCH64_BTI_PLT 0x70000001, d_val
DT_AARCH64_PAC_PLT 0x70000003, d_val
These presence of these tags indicate that PLT sequences have been
protected using Branch Target Identification and Pointer Authentication
respectively. The presence of both indicates that the PLT sequences have
been protected with both Branch Target Identification and Pointer
Authentication.
This patch adds the tags and tests for llvm-readobj and yaml2obj.
As some of the processor specific dynamic tags overlap, this patch splits
them up, keeping their original default value if they were not previously
mentioned explicitly in a switch case.
Differential Revision: https://reviews.llvm.org/D62596
llvm-svn: 362493
Summary:
This *might* be the last fold for `sink-addsub-of-const.ll`, but i'm not sure yet.
As far as i can tell, there are no regressions here (ignoring x86-32),
all changes are either good or neutral.
This, almost surprisingly to me, fixes the motivational tests (in `shift-amount-mod.ll`)
`@reg32_lshr_by_sub_from_negated` from [[ https://bugs.llvm.org/show_bug.cgi?id=41952 | PR41952 ]].
https://rise4fun.com/Alive/vMd3
Reviewers: RKSimon, t.p.northover, craig.topper, spatel, efriedma
Reviewed By: RKSimon
Subscribers: sdardis, javed.absar, arichardson, kristof.beyls, jrtc27, atanasyan, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D62774
llvm-svn: 362488
As I mentioned on D61887 we don't get many hits on ComputeNumSignBits as we did on computeKnownBits.
The case we do get is interesting though - it allows us to use the 'ConditionalNegate' combine in combineLogicBlendIntoPBLENDV to remove a select.
It comes too late for SSE41 (BLENDV) cases, but SSE2 tests can hit it now. We should probably try to make use of this for SSE41+ targets as well - avoiding variable blends is usually a good idea. I'll investigate as a followup.
Differential Revision: https://reviews.llvm.org/D62777
llvm-svn: 362486
Includes a fix for an introduced build failure due to a post c++11 use of std::mismatch.
This fixes some thin archive relative path issues, paths are shortened where possible and paths are output correctly when using the display table command.
Differential Revision: https://reviews.llvm.org/D59491
llvm-svn: 362484
This change adds two FP16 extraction and two insertion patterns
(one per possible vector length).
Extractions are handled by copying a Q/D register into one of VFP2
class registers, where single FP32 sub-registers can be accessed. Then
the extraction of even lanes are simple sub-register extractions
(because we don't care about the top parts of registers for FP16
operations). Odd lanes need an additional VMOVX instruction.
Unfortunately, insertions cannot be handled in the same way, because:
* There is no instruction to insert FP16 into an even lane (VINS only
works with odd lanes)
* The patterns for odd lanes will have a form of a DAG (not a tree),
and will not be implementable in pure tablegen
Because of this insertions are handled in the same way as 16-bit
integer insertions (with conversions between FP registers and GPRs
using VMOVHR instructions).
Without these patterns the ARM backend would sometimes fail during
instruction selection.
This patch also adds patterns which combine:
* an FP16 element extraction and a store into a single VST1
instruction
* an FP16 load and insertion into a single VLD1 instruction
Differential Revision: https://reviews.llvm.org/D62651
llvm-svn: 362482
While prof branch_weights inconsistencies are being fixed patch
by patch (pass by pass) we need SwitchInstProfUpdateWrapper to
be safe with respect to inconsistent metadata that can come from
passes that have not been fixed yet. See the bug found by @nikic
in https://reviews.llvm.org/D62126.
This patch introduces one more state (called Invalid) to the
wrapper class that allows users to work with the underlying
SwitchInst ignoring the prof metadata changes.
Created a unit test for the SwitchInstProfUpdateWrapper class.
Reviewers: davidx, nikic, eraman, reames, chandlerc
Reviewed By: davidx
Differential Revision: https://reviews.llvm.org/D62656
llvm-svn: 362473
This opportunity is found from spec 2017 557.xz_r. And it is used by the sha encrypt/decrypt. See sha-2/sha512.c
static void store64(u64 x, unsigned char* y)
{
for(int i = 0; i != 8; ++i)
y[i] = (x >> ((7-i) * 8)) & 255;
}
static u64 load64(const unsigned char* y)
{
u64 res = 0;
for(int i = 0; i != 8; ++i)
res |= (u64)(y[i]) << ((7-i) * 8);
return res;
}
The load64 has been implemented by https://reviews.llvm.org/D26149
This patch is trying to implement the store pattern.
Match a pattern where a wide type scalar value is stored by several narrow
stores. Fold it into a single store or a BSWAP and a store if the targets
supports it.
Assuming little endian target:
i8 *p = ...
i32 val = ...
p[0] = (val >> 0) & 0xFF;
p[1] = (val >> 8) & 0xFF;
p[2] = (val >> 16) & 0xFF;
p[3] = (val >> 24) & 0xFF;
>
*((i32)p) = val;
i8 *p = ...
i32 val = ...
p[0] = (val >> 24) & 0xFF;
p[1] = (val >> 16) & 0xFF;
p[2] = (val >> 8) & 0xFF;
p[3] = (val >> 0) & 0xFF;
>
*((i32)p) = BSWAP(val);
Differential Revision: https://reviews.llvm.org/D61843
llvm-svn: 362472
The family of 32-bit Thumb instruction encodings that include t2ORR,
t2AND and t2EOR are all listed in the ArmARM as having (0) in bit 15.
The Tablegen descriptions of those instructions listed them as ?. This
change tightens that up by making them into 0 + Unpredictable.
In the specific case of t2ORR, we tighten it up still further by
making the zero bit mandatory. This change comes from Arm v8.1-M, in
which encodings with that bit equal to 1 will now be used for
different instructions.
Reviewers: dmgreen, samparker, SjoerdMeijer, efriedma
Reviewed By: dmgreen, efriedma
Subscribers: efriedma, javed.absar, kristof.beyls, hiraditya, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D60705
llvm-svn: 362470
This is something of a workaround, and the state of stack realignment
controls is kind of a mess. Ideally, we would be able to specify the
stack is infinitely aligned on entry to a kernel.
TargetFrameLowering provides multiple controls which apply at
different points. The StackRealignable field is used during
SelectionDAG, and for some reason distinct from this
hook. StackAlignment is a single field not dependent on the
function. It would probably be better to make that dependent on the
calling convention, and the maximum value for kernels.
Currently this doesn't really change anything, since the frame
lowering mostly does its own thing. This helps avoid regressions in a
future change which will rely more heavily on hasFP.
llvm-svn: 362447
Instead of emitting all of the test stuff for a compare when it's only used by
a select, instead, just emit the compare + select. The select will use the
value of NZCV correctly, so we don't need to emit all of the test instructions
etc.
For now, only support fp selects which use G_FCMP. Also only support condition
codes which will only require one select to represent.
Also add a test.
Differential Revision: https://reviews.llvm.org/D62695
llvm-svn: 362446
This looks like an oversight as all the other binary operators are present.
Accidentally noticed while auditing places that need FNeg handling.
No test because as noted in the review it would be contrived and amount to "don't crash"
Differential Revision: https://reviews.llvm.org/D62790
llvm-svn: 362441
r362199 fixed it for zero masking, but not zero masking. The load
folding in the peephole pass hid the bug. This patch turns off
the peephole pass on the relevant test to ensure coverage.
llvm-svn: 362440
Summary: This change facilitates propagating fmf which was placed on setcc from fcmp through folds with selects so that back ends can model this path for arithmetic folds on selects in SDAG.
Reviewers: qcolombet, spatel
Reviewed By: qcolombet
Subscribers: nemanjai, jsji
Differential Revision: https://reviews.llvm.org/D62552
llvm-svn: 362439
We currently miss the opportunities for optmizing comparisons in the peephole
optimizer if the input is the result of a COPY since we look for record-form
versions of the producing instruction.
This patch simply lets the optimization peek through copies.
Differential revision: https://reviews.llvm.org/D59633
llvm-svn: 362438
For some reason multiple places need to do this, and the variant the
loop unroller and inliner use was not handling it.
Also, introduce a new wrapper to be slightly more precise, since on
AMDGPU some addrspacecasts are free, but not no-ops.
llvm-svn: 362436
The implementation is conceptually simple: We separate the LHS and
RHS into positive and negative components and then also compute the
positive and negative components of the result, taking into account
that e.g. only pos/pos and neg/neg will give a positive result.
However, there's one significant complication: SignedMin / -1 is UB
for sdiv, and we can't just ignore it, because the APInt result of
SignedMin would break the sign segregation. Instead we drop SignedMin
or -1 from the corresponding ranges, taking into account some edge
cases with wrapped ranges.
Because of the sign segregation, the implementation ends up being
nearly fully precise even for wrapped ranges (the remaining
imprecision is due to ranges that are both signed and unsigned
wrapping and are divided by a trivial divisor like 1). This means
that the testing cannot just check the signed envelope as we
usually do. Instead we collect all possible results in a bitvector
and construct a better sign wrapped range (than the full envelope).
Differential Revision: https://reviews.llvm.org/D61238
llvm-svn: 362430
This patch fixes a problem that occurs in LowerSwitch when a switch statement has a PHI node as its condition, and the PHI node only has two incoming blocks, and one of those incoming blocks is through an unreachable default in the switch statement. When this condition occurs, LowerSwitch holds a pointer to the condition value, but removes the switch block as a predecessor of the PHI block, causing the PHI node to be replaced. LowerSwitch then tries to use its stale pointer to the original condition value, causing a crash.
Differential Revision: https://reviews.llvm.org/D62560
llvm-svn: 362427
This reverts commit r362407. It broke compilation of
llvm/lib/Object/ArchiveWriter.cpp:
error: type 'llvm::sys::path::const_iterator' does not provide a call
operator
llvm-svn: 362413
The big endian PPC buildbots are all failing now due to calls to cache
invalidation in unit tests on data that has only the PROT_EXEC flag set.
This has been an issue all along on FreeBSD but it can affect Linux machines
depending on configuration.
This patch mitigates the issue the same way it is mitigated on FreeBSD.
Since this is needed to bring the buildbots back to green, I plan to commit this
and allow for post-commit review, but I thought I would also post it here for
ease of access/readability.
Differential revision: https://reviews.llvm.org/D62741
llvm-svn: 362412
This fixes some thin archive relative path issues, paths are shortened where possible and paths are output correctly when using the display table command.
Differential Revision: https://reviews.llvm.org/D59491
llvm-svn: 362407
We were missing this fold in the DAG, which I've copied directly from llvm::ConstantFoldCastInstruction
Differential Revision: https://reviews.llvm.org/D62807
llvm-svn: 362397
LanaiMCCodeEmitter.cpp was not using any APIs from Lanai.h, and was only
including it for transitive dependencies. Doing so is problematic from
include-what-you-use perspective, but it is also a layering issue (it
creates a dependency cycle between the primary Lanai target library and
the MCTargetDesc library).
llvm-svn: 362394
HexagonInstPrinter.cpp was not using any APIs from HexagonAsmPrinter.h.
Doing so is problematic from include-what-you-use perspective, but it is
also a layering issue (it creates a dependency cycle between the primary
Hexagon target library and the MCTargetDesc library).
llvm-svn: 362389
HexagonMCInstrInfo.cpp was not using any APIs from Hexagon.h. Doing so
is problematic from include-what-you-use perspective, but it is also a
layering issue (it creates a dependency cycle between the primary
Hexagon target library and the MCTargetDesc library).
llvm-svn: 362387
HexagonMCCodeEmitter.cpp was not using any APIs from Hexagon.h. Doing
so is problematic from include-what-you-use perspective, but it is also
a layering issue (it creates a dependency cycle between the primary
Hexagon target library and the MCTargetDesc library).
llvm-svn: 362386
HexagonMCCompound.cpp was not using any APIs from Hexagon.h. Doing so
is problematic from include-what-you-use perspective, but it is also a
layering issue (it creates a dependency cycle between the primary
Hexagon target library and the MCTargetDesc library).
llvm-svn: 362385
HexagonShuffler.cpp was not using any APIs from Hexagon.h, and was only
including it for transitive dependencies. Doing so is problematic from
include-what-you-use perspective, but it is also a layering issue (it
creates a dependency cycle between the primary Hexagon target library
and the MCTargetDesc library).
llvm-svn: 362384
HexagonMCChecker.cpp was not using any APIs from Hexagon.h. Doing so is
problematic from include-what-you-use perspective, but it is also a
layering issue (it creates a dependency cycle between the primary
Hexagon target library and the MCTargetDesc library).
llvm-svn: 362383
HexagonMCTargetDesc.cpp was not using any APIs from Hexagon.h. Doing so
is problematic from include-what-you-use perspective, but it is also a
layering issue (it creates a dependency cycle between the primary
Hexagon target library and the MCTargetDesc library).
llvm-svn: 362382
HexagonMCShuffler.cpp was not using any APIs from Hexagon.h. Doing so
is problematic from include-what-you-use perspective, but it is also a
layering issue (it creates a dependency cycle between the primary
Hexagon target library and the MCTargetDesc library).
llvm-svn: 362381
The recent change D60691 introduced a bug in clang when handling
option combinations such as `-mcpu=cortex-m4 -mfpu=none`. Those
options together should select Cortex-M4 but disable all use of
hardware FP, but in fact, now hardware FP instructions can still be
generated in that mode.
The reason is because the handling of FPUVersion::NONE disables all
the same feature names it used to, of which the base one is `vfp2`.
But now there are further features below that, like `vfp2d16fp` and
(following D60694) `fpregs`, which also need to be turned off to
disable hardware FP completely.
Added a tiny test which double-checks that compiling a simple FP
function doesn't access the FP registers.
Reviewers: SjoerdMeijer, dmgreen
Reviewed By: dmgreen
Subscribers: lebedev.ri, javed.absar, kristof.beyls, hiraditya, cfe-commits, llvm-commits
Tags: #clang, #llvm
Differential Revision: https://reviews.llvm.org/D62729
llvm-svn: 362380
HexagonELFObjectWriter.cpp was not using any APIs from Hexagon.h, and
was only including it for transitive dependencies. Doing so is
problematic from include-what-you-use perspective, but it is also a
layering issue (it creates a dependency cycle between the primary
Hexagon target library and the MCTargetDesc library).
llvm-svn: 362376
When LiveDebugValues deduces new variable's location from spill, restore or
register copy instruction it should close old variable's location. Otherwise
we can have multiple block output locations for same variable. That could lead
to inserting two DBG_VALUEs for same variable to the beginning of the successor
block which results to ignoring of first DBG_VALUE.
Reviewers: aprantl, jmorse, wolfgangp, dstenb
Reviewed By: aprantl
Subscribers: probinson, asowda, ivanbaev, petarj, djtodoro
Tags: #debug-info
Differential Revision: https://reviews.llvm.org/D62196
llvm-svn: 362373
HexagonAsmBackend.cpp was not using any APIs from Hexagon.h. Doing so
is problematic from include-what-you-use perspective, but it is also a
layering issue (it creates a dependency cycle between the primary
Hexagon target library and the MCTargetDesc library).
llvm-svn: 362372
HexagonAsmParser.cpp was not using any APIs from Hexagon.h. Doing so is
problematic from include-what-you-use perspective, but it is also a
layering issue (it creates a dependency cycle between the primary
Hexagon target library and the AsmParser library).
llvm-svn: 362370
HexagonShuffler.h was not using any APIs from Hexagon.h, and was only
including it for transitive dependencies. Doing so is problematic from
include-what-you-use perspective, but it is also a layering issue (it
creates a dependency cycle between the primary Hexagon target library
and the MCTargetDesc library).
llvm-svn: 362369
BPFMCTargetDesc.cpp was not using any APIs from BPF.h. Doing so is
problematic from include-what-you-use perspective, but it is also a
layering issue (it creates a dependency cycle between the primary
BPF target library and the MCTargetDesc library).
llvm-svn: 362368
Summary:
- pr42062
When compiling for MinSize,
ARMTargetLowering::LowerCall decides to indirect
multiple calls to a same function. However,
it disconsiders the limitation that thumb1
indirect calls require the callee to be in a
register from r0 to r3 (llvm limiation).
If all those registers are used by arguments, the
compiler dies with "error: run out of registers
during register allocation".
This patch tells the function
IsEligibleForTailCallOptimization if we intend to
perform indirect calls, as to avoid tail call
optimization.
Reviewers: dmgreen, efriedma
Reviewed By: efriedma
Subscribers: javed.absar, kristof.beyls, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D62683
llvm-svn: 362366
Summary:
LDWRdPtr would be expanded to ld+ldd. ldd only accepts the pointer register is Y or Z.
So the register class of pointer of LDWRdPtr should be PTRDISPREGS instead of PTRREGS.
Reviewers: dylanmckay
Reviewed By: dylanmckay
Subscribers: dylanmckay, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D62300
llvm-svn: 362351
If we hit the limit, we do expand the outstanding tokenfactors.
Otherwise, we might drop nodes with users in the unexpanded
tokenfactors. This fixes the crashes reported by Jordan Rupprecht.
Reviewers: niravd, spatel, craig.topper, rupprecht
Reviewed By: niravd
Differential Revision: https://reviews.llvm.org/D62633
llvm-svn: 362350
Also add two FC_Far that seem to be missing, by symmetry from
the public and protected cases. (But FC_Far isn't really a thing
anymore, so this doesn't really have an observable effect.)
llvm-svn: 362344
Move this combine from x86 into generic DAGCombine, which currently only manages cases where the bitcast is between types of the same scalarsize.
Differential Revision: https://reviews.llvm.org/D59188
llvm-svn: 362324
Add (opt-in) support for implicit truncation to isConstOrConstSplat, which allows us to match truncated 'all ones' cases in isBitwiseNot.
PR41020 compares against using ISD::isBuildVectorAllOnes() instead, but that predicate silently accepts any UNDEF elements in the build vector which might not be what we want in isBitwiseNot - so I've added an opt-in 'AllowUndefs' flag that is set to false by default but will allow us to enable it on individual cases where its safe.
Differential Revision: https://reviews.llvm.org/D62783
llvm-svn: 362323
The results of the dyn_casts were immediately dereferenced on the next line
so they had better not be null.
I don't think there's any way for these dyn_casts to fail, so use a cast
of adding null check.
llvm-svn: 362315
Over a year ago, MachineInstr gained a fourth boolean parameter that occurs
before the TII pointer. When this happened, several places started accidentally
passing TII into this boolean parameter instead of the TII parameter.
llvm-svn: 362312
Extract a willNotOverflow() helper function that is shared between
eliminateOverflowIntrinsic() and strengthenOverflowingOperation().
Use WithOverflowInst for the former.
We'll be able to reuse the same code for saturating intrinsics as
well.
llvm-svn: 362305
Summary: Fneg can be implemented with an xor rather than a function call so we don't need to add the function call overhead. This was pointed out in D62699
Reviewers: efriedma, cameron.mcinally
Reviewed By: efriedma
Subscribers: javed.absar, eraman, hiraditya, haicheng, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D62747
llvm-svn: 362304
The `cfcmsa` and `ctcmsa` instructions accept index of MSA control
register. The MIPS64 SIMD Architecture define eight MSA control
registers. But register index for `cfcmsa` and `ctcmsa` instructions
might be any number in 0..31 range. If the index is greater then 7,
`cfcmsa` writes zero to the destination registers and `ctcmsa` does
nothing [1].
[1] MIPS Architecture for Programmers Volume IV-j:
The MIPS64 SIMD Architecture Module
https://www.mips.com/?do-download=the-mips64-simd-architecture-module
Differential Revision: https://reviews.llvm.org/D62597
llvm-svn: 362299
If we would allow register coalescing on PTRDISPREGS class then register
allocator can lock Z register to some virtual register. Larger instructions
requiring a memory acces then fail during the register allocation phase since
there is no available register to hold a pointer if Y register was already
taken for a stack frame. This patch prevents it by keeping Z register
spillable. It does it by not allowing coalescer to lock it.
Original discussion on https://github.com/avr-rust/rust/issues/128.
llvm-svn: 362298
Fix for https://bugs.llvm.org/show_bug.cgi?id=31181 and partial fix
for LFTR poison handling issues in general.
When LFTR moves a condition from pre-inc to post-inc, it may now
depend on value that is poison due to nowrap flags. To avoid this,
we clear any nowrap flag that SCEV cannot prove for the post-inc
addrec.
Additionally, LFTR may switch to a different IV that is dynamically
dead and as such may be arbitrarily poison. This patch will correct
nowrap flags in some but not all cases where this happens. This is
related to the adoption of IR nowrap flags for the pre-inc addrec.
(See some of the switch_to_different_iv tests, where flags are not
dropped or insufficiently dropped.)
Finally, there are likely similar issues with the handling of GEP
inbounds, but we don't have a test case for this yet.
Differential Revision: https://reviews.llvm.org/D60935
llvm-svn: 362292
This allows the DWARFExpression class to handle addresses without
crashing on targets with 16-bit pointers like AVR.
This is required in order to generate assembly from clang via the '-S'
flag.
This fixes an error with the following message:
clang: llvm/include/llvm/DebugInfo/DWARF/DWARFExpression.h:132: llvm::DWARFExpression::DWARFExpression(llvm::DataExtractor, uint16_t, uint8_t):
Assertion `AddressSize == 8 || AddressSize == 4' failed.
llvm-svn: 362290
Summary:
This was flagged in https://www.viva64.com/en/b/0629/ under "Snippet No.
33".
It seems that this statement is doing the standard bitwise trick for
adjusting a value to have a specific alignment.
The issue is that getStubAlignment() returns an unsigned, while DataSize
is declared a uint64_t. The right hand side of the expression is not
extended to 64b before bitwise negation, resulting in the top half of
the mask being 0s, which is not correct for realignment.
Reviewers: lhames, MaskRay
Reviewed By: MaskRay
Subscribers: RKSimon, MaskRay, hiraditya, llvm-commits, srhines
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D62227
llvm-svn: 362286
At the moment, LoopPredication completely bails out if it sees a latch of the form:
%cmp = icmp ne %iv, %N
br i1 %cmp, label %loop, label %exit
OR
%cmp = icmp ne %iv.next, %NPlus1
br i1 %cmp, label %loop, label %exit
This is unfortunate since this is exactly the form that LFTR likes to produce. So, go ahead and recognize simple cases where we can.
For pre-increment loops, we leverage the fact that LFTR likes canonical counters (i.e. those starting at zero) and a (presumed) range fact on RHS to discharge the check trivially.
For post-increment forms, the key insight is in remembering that LFTR had to insert a (N+1) for the RHS. CVP can hopefully prove that add nsw/nuw (if there's appropriate range on N to start with). This leaves us both with the post-inc IV and the RHS involving an nsw/nuw add, and SCEV can discharge that with no problem.
This does still need to be extended to handle non-one steps, or other harder patterns of variable (but range restricted) starting values. That'll come later.
Differential Revision: https://reviews.llvm.org/D62748
llvm-svn: 362282
We were hashing the string pointer, not the string, so two instructions
could be identical (isIdenticalTo), but have different hash codes.
This showed up as a very rare, non-deterministic assertion failure
rehashing a DenseMap constructed by MachineOutliner. So there's no
"real" testcase, just a unittest which checks that the hash function
behaves correctly.
I'm a little scared fixing this is going to cause a regression in
outlining or MachineCSE, but hopefully we won't run into any issues.
Differential Revision: https://reviews.llvm.org/D61975
llvm-svn: 362281
CodeView has its own register map which is defined in cvconst.h. Missing this
mapping before saving register to CodeView causes debugger to show incorrect
value for all register based variables, like variables in register and local
variables addressed by register (stack pointer + offset).
This change added mapping between LLVM register and CodeView register so the
correct register number will be stored to CodeView/PDB, it aso fixed the
mapping from CodeView register number to register name based on current
CPUType but print PDB to yaml still assumes X86 CPU and needs to be fixed.
Differential Revision: https://reviews.llvm.org/D62608
llvm-svn: 362280
Summary:
It looks like since INLINEASM_BR was created off of INLINEASM (r353563),
a few checks for INLINEASM needed to be updated to check for either
case.
pr/41999
Reviewers: hfinkel
Reviewed By: hfinkel
Subscribers: nemanjai, hiraditya, kbarton, jsji, llvm-commits, craig.topper, srhines
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D62403
llvm-svn: 362278
Testing with debuggers shows that our previous behavior was correct.
The reason I thought MSVC did things differently is that MSVC prefers to
use the 0xB combined code offset and code length update opcode when
inline sites are discontiguous.
Keep the test changes, and update the llvm-pdbutil inline line table
dumper to account for this new interpretation of the opcodes.
llvm-svn: 362277
When the object size argument is -1, no checking can be done, so calling the
_chk variant is unnecessary. We already did this for a bunch of these
functions.
rdar://50797197
Differential revision: https://reviews.llvm.org/D62358
llvm-svn: 362272
Just copy all of the operands except the chain and call MorphNode on that.
This removes the IsUnary and IsTernary flags.
Also always get the result type from the result type of the original
nodes. Previously we got it from the operand except for two nodes
where that didn't work.
llvm-svn: 362269
Summary:
Fixes a warning produced from scan-build (llvm.org/reports/scan-build/),
further warnings found by annotation isMoveInstr [[nodiscard]].
isMoveInstr potentially does not assign to its parameters, so if they
were uninitialized, they will potentially stay uninitialized. It seems
most call sites pass references to uninitialized values, then use them
without checking the return value.
Reviewers: wmi
Reviewed By: wmi
Subscribers: MatzeB, qcolombet, hiraditya, tpr, llvm-commits, srhines
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D62109
llvm-svn: 362265
After improving the inline line table dumper in llvm-pdbutil and looking
at MSVC's inline line tables, it is clear that setting the length of the
inlined code region does not update the code offset. This means that the
delta to the beginning of a new discontiguous inlined code region should
be calculated relative to the last code offset, excluding the length.
Implementing this is a one line fix for MC: simply don't update
LastLabel.
While I'm updating these test cases, switch them to use llvm-objdump -d
and llvm-pdbutil. This allows us to show offsets of each instruction and
correlate the line table offsets to the actual code.
llvm-svn: 362264
If we can determine that a saturating add/sub will not overflow based
on range analysis, convert it into a simple binary operation. This is
a sibling transform to the existing with.overflow handling.
Reapplying this with an additional check that the saturating intrinsic
has integer type, as LVI currently does not support vector types.
Differential Revision: https://reviews.llvm.org/D62703
llvm-svn: 362263
Noticed on D62703. LVI only handles plain integers, not vectors of
integers. This was previously not an issue, because vector support
for with.overflow is only a relatively recent addition.
llvm-svn: 362261
We don't want to create vregs if there is nothing to use them for. That causes
verifier errors.
Differential Revision: https://reviews.llvm.org/D62740
llvm-svn: 362247
If we can determine that a saturating add/sub will not overflow
based on range analysis, convert it into a simple binary operation.
This is a sibling transform to the existing with.overflow handling.
Differential Revision: https://reviews.llvm.org/D62703
llvm-svn: 362242
[FPEnv] Added a special UnrollVectorOp method to deal with the chain on StrictFP opcodes
This change creates UnrollVectorOp_StrictFP. The purpose of this is to address a failure that consistently occurs when calling StrictFP functions on vectors whose number of elements is 3 + 2n on most platforms, such as PowerPC or SystemZ. The old UnrollVectorOp method does not expect that the vector that it will unroll will have a chain, so it has an assert that prevents it from running if this is the case. This new StrictFP version of the method deals with the chain while unrolling the vector. With this new function in place during vector widending, llc can run vector-constrained-fp-intrinsics.ll for SystemZ successfully.
Submitted by: Drew Wock <drew.wock@sas.com>
Reviewed by: Cameron McInally, Kevin P. Neal
Approved by: Cameron McInally
Differential Revision: https://reviews.llvm.org/D62546
llvm-svn: 362241
AMDGPU uses multiplier 9 for the inline cost. It is taken into account
everywhere except for inline hint threshold. As a result we are penalizing
functions with the inline hint making them less probable to be inlined
than those without the hint. Defaults are 225 for a normal function and
325 for a function with an inline hint. Currently we have effective
threshold 225 * 9 = 2025 for normal functions and just 325 for those with
the hint. That is fixed by this patch.
Differential Revision: https://reviews.llvm.org/D62707
llvm-svn: 362239
In PPCReduceCRLogicals after splitting the original MBB into 2, the 2 impacted branches still use original branch probability. This is unreasonable. Suppose we have following code, and the probability of each successor is 50%.
condc = conda || condb
br condc, label %target, label %fallthrough
It can be transformed to following,
br conda, label %target, label %newbb
newbb:
br condb, label %target, label %fallthrough
Since each branch has a probability of 50% to each successor, the total probability to %fallthrough is 25% now, and the total probability to %target is 75%. This actually changed the original profiling data. A more reasonable probability can be set to 70% to the false side for each branch instruction, so the total probability to %fallthrough is close to 50%.
This patch assumes the branch target with two incoming edges have same edge frequency and computes new probability fore each target, and keep the total probability to original targets unchanged.
Differential Revision: https://reviews.llvm.org/D62430
llvm-svn: 362237
These can take a significant amount of time in some builds.
Suggested by Andrea Di Biagio.
Differential Revision: https://reviews.llvm.org/D62666
llvm-svn: 362219
It looks this fold was already partially happening, indirectly
via some other folds, but with one-use limitation.
No other fold here has that restriction.
https://rise4fun.com/Alive/ftR
llvm-svn: 362217
Summary:
A three sources variant of the TBL instruction is added to the existing
SVE instruction in SVE2. This is implemented with minor changes to the
existing TableGen class. TBX is a new instruction with its own
definition.
The specification can be found here:
https://developer.arm.com/docs/ddi0602/latest
Reviewed By: chill
Differential Revision: https://reviews.llvm.org/D62600
llvm-svn: 362214
Handle position independent code for MIPS32.
When callee is global address, lower call will emit callee
as G_GLOBAL_VALUE and add target flag if needed.
Support $gp in getRegBankFromRegClass().
Select G_GLOBAL_VALUE, specially handle case when
there are target flags attached by lowerCall.
Differential Revision: https://reviews.llvm.org/D62589
llvm-svn: 362210
Move initGlobalBaseReg from MipsSEDAGToDAGISel to MipsFunctionInfo.
This way functions used for handling position independent code during
instruction selection, getGlobalBaseReg and initGlobalBaseReg,
end up in same class.
Differential Revision: https://reviews.llvm.org/D62586
llvm-svn: 362206
Lower call for callee that is register for MIPS32.
Register should contain callee function address.
Differential Revision: https://reviews.llvm.org/D62585
llvm-svn: 362204
These patterns can incorrectly narrow a volatile load from 128-bits to 64-bits.
Similar to PR42079.
Switch to using (v4i32 (bitcast (v2i64 (scalar_to_vector (loadi64))))) as the
load pattern used in the instructions.
This probably still has issues in 32-bit mode where loadi64 isn't legal. Maybe
we should use VZMOVL for widened loads even when we don't need the upper bits
as zeroes?
llvm-svn: 362203
DAG combine will usually fold fpextend+load to an fp extload anyway. So the
256 and 512 patterns were probably unnecessary. The 128 bit pattern was special
in that it looked for a v4f32 load, but then used it in an instruction that
only loads 64-bits. This is bad if the load happens to be volatile. We could
probably make the patterns volatile aware, but that's more work for something
that's probably rare. The peephole pass might kick in and save us anyway. We
might also be able to fix this with some additional DAG combines.
This also adds patterns for vselect+extload to enabled masked vcvtps2pd to be
used. Previously we looked for the unlikely vselect+fpextend+load.
llvm-svn: 362199
This consolidates the vreg skip code into one function (SkipVRegs()).
SkipVRegs() now knows if it should skip as if it is the first initialization or
subsequent skips.
The first skip is also done the first time createVirtualRegister is called by
the cursor instead of by the cursor's constructor. This prevents verifier
errors on machine functions that have no vregs (where the verifier will
complain that there are vregs when the function uses none).
Differential Revision: https://reviews.llvm.org/D62717
llvm-svn: 362195
This makes the 5 address operands come first. And the data operand comes last.
This matches the operand order the instruction is created with. It's also the
expected order in X86MCInstLower. So everything appeared to work, but the
operands didn't match their declared type.
Fixes a -verify-machineinstrs failure.
Also remove the isel patterns from these instructions since they should only
be used for stack spills and reloads. I'm not even sure what types the patterns
were looking for to match.
llvm-svn: 362193
This is am almost NFC, it does the following:
- If there is no register class for a COPY's src or dst, bail.
- Fixes uses iterator invalidation bug.
Differential Revision: https://reviews.llvm.org/D62713
llvm-svn: 362191
The result types aren't mentioned in the pattern name so really shouldn't be in the PatFrags.
The users of these either have their own type constraint or rely on the type constranit system to realize the only legal extend would be to f64.
llvm-svn: 362175
The LoadExt table defaults to all combinations being Legal. For
vector types, only src VTs with an i1 element type were ever changed.
So we don't need to mark them legal manually.
llvm-svn: 362170
Separate the remark serialization to YAML from the LLVM Diagnostics.
This adds a new serialization abstraction: remarks::Serializer. It's
completely independent from lib/IR and it provides an easy way to
replace YAML by providing a new remarks::Serializer.
Differential Revision: https://reviews.llvm.org/D62632
llvm-svn: 362160
I don't have a test case for these, but there is a test case for D62266
where, even after all the constant-folding patches, we still end up
with endless combine loop. Which makes sense, since we don't constant
fold for opaque constants.
llvm-svn: 362156
Previously, this used a statement like this:
Map[A] = Map[B];
This is equivalent to the following:
const auto &Src = Map[B];
auto &Dest = Map[A];
Dest = Src;
The second statement, "auto &Dest = Map[A];" can insert a new
element into the DenseMap, which can potentially grow and reallocate
the DenseMap's internal storage, which will invalidate the existing
reference to the source. When doing the actual assignment,
the Src reference is dereferenced, accessing memory that was
freed when the DenseMap grew.
This issue hasn't shown up when LLVM was built with Clang, because
the right hand side ended up dereferenced before evaulating the
left hand side. (If the value type is a larger data type, Clang doesn't
do this but behaves like GCC.)
With GCC, a cast to Value* isn't enough to make it dereference the
right hand side reference before invoking operator[] (while that is
enough to make Clang/LLVM do the right thing for larger types), but
storing it in an intermediate variable in a separate statement works.
This fixes PR42065.
Differential Revision: https://reviews.llvm.org/D62624
llvm-svn: 362150
Summary:
Only vector tests are being affected here,
since subtraction by scalar constant is rewritten
as addition by negated constant.
No surprising test changes.
https://rise4fun.com/Alive/pbT
This is a recommit, originally committed in rL361852, but reverted
to investigate test-suite compile-time hangs.
Reviewers: RKSimon, craig.topper, spatel
Reviewed By: RKSimon
Subscribers: javed.absar, kristof.beyls, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D62257
llvm-svn: 362146
Summary:
Again only vectors affected. Frustrating. Let me take a look into that..
https://rise4fun.com/Alive/AAq
This is a recommit, originally committed in rL361852, but reverted
to investigate test-suite compile-time hangs, and then reverted in
rL362109 to fix missing constant folds that were causing
endless combine loops.
Reviewers: RKSimon, craig.topper, spatel
Reviewed By: RKSimon
Subscribers: javed.absar, JDevlieghere, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D62294
llvm-svn: 362145
Summary:
This prevents regressions in next patch,
and somewhat recovers from the regression to AMDGPU test in D62223.
It is indeed not great that we leave vector decrement,
don't transform it into vector add all-ones..
https://rise4fun.com/Alive/ZRl
This is a recommit, originally committed in rL361852, but reverted
to investigate test-suite compile-time hangs, and then reverted in
rL362109 to fix missing constant folds that were causing
endless combine loops.
Reviewers: RKSimon, craig.topper, spatel, arsenm
Reviewed By: RKSimon, arsenm
Subscribers: kzhuravl, jvesely, wdng, nhaehnle, yaxunl, javed.absar, dstuttard, tpr, t-tye, kristof.beyls, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D62263
llvm-svn: 362144
Summary:
Direct sibling of D62223 patch.
While i don't have a direct motivational pattern for this,
it would seem to make sense to handle both patterns (or none),
for symmetry?
The aarch64 changes look neutral;
sparc and systemz look like improvement (one less instruction each);
x86 changes - 32bit case improves, 64bit case shows that LEA no longer
gets constructed, which may be because that whole test is `-mattr=+slow-lea,+slow-3ops-lea`
https://rise4fun.com/Alive/ffh
This is a recommit, originally committed in rL361852, but reverted
to investigate test-suite compile-time hangs, and then reverted in
rL362109 to fix missing constant folds that were causing
endless combine loops.
Reviewers: RKSimon, craig.topper, spatel, t.p.northover
Reviewed By: t.p.northover
Subscribers: t.p.northover, jyknight, javed.absar, kristof.beyls, fedor.sergeev, jrtc27, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D62252
llvm-svn: 362143
Summary:
The main motivation is shown by all these `neg` instructions that are now created.
In particular, the `@reg32_lshr_by_negated_unfolded_sub_b` test.
AArch64 test changes all look good (`neg` created), or neutral.
X86 changes look neutral (vectors), or good (`neg` / `xor eax, eax` created).
I'm not sure about `X86/ragreedy-hoist-spill.ll`, it looks like the spill
is now hoisted into preheader (which should still be good?),
2 4-byte reloads become 1 8-byte reload, and are elsewhere,
but i'm not sure how that affects that loop.
I'm unable to interpret AMDGPU change, looks neutral-ish?
This is hopefully a step towards solving [[ https://bugs.llvm.org/show_bug.cgi?id=41952 | PR41952 ]].
https://rise4fun.com/Alive/pkdq (we are missing more patterns, i'll submit them later)
This is a recommit, originally committed in rL361852, but reverted
to investigate test-suite compile-time hangs, and then reverted in
rL362109 to fix missing constant folds that were causing
endless combine loops.
Reviewers: craig.topper, RKSimon, spatel, arsenm
Reviewed By: RKSimon
Subscribers: bjope, qcolombet, kzhuravl, jvesely, wdng, nhaehnle, yaxunl, javed.absar, dstuttard, tpr, t-tye, kristof.beyls, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D62223
llvm-svn: 362142
increase the MachO/x86-64 stub alignment to 8.
Stub alignment should be guaranteed for any section containing RuntimeDyld
stubs/GOT-entries. To do this we should pad and align all sections containing
stubs, not just code sections.
This commit also bumps the MachO/x86-64 stub alignment to 8, so that GOT entries
will be aligned.
llvm-svn: 362139
Matt Arsenault