Musl treats PowerPC SPE as a soft-float target (as the PowerPC SPE ABI
is soft-float compatible).
Reviewed By: jhibbits, MaskRay
Differential Revision: https://reviews.llvm.org/D105869
Fix r285764: there is no guarantee that Out::first is placed before other
static data members of `struct Out`. After `bufferStart` was introduced, this
out-of-bounds write is destined in many compilers. It is likely benign, though.
And move `Out::elfHeader->size` assignment beside `Out::elfHeader->sectionIndex`
As reported in https://bugs.freebsd.org/260078, the gnutls Makefiles
pass -Wa,-march=all to compile a number of assembly files. Clang does
not support this -march value, but because of a mistake in handling
the arguments, an unitialized Arg pointer is dereferenced, which can
cause a segfault.
Work around this by adding a check if the local WaMArch variable is
initialized, and if so, using its value in the diagnostic message.
Reviewed By: tschuett
Differential Revision: https://reviews.llvm.org/D114677
I'm not sure whether this can cause any actual non-determinism,
but at least it makes the DSE debug log non-deterministic, which
makes it harder to debug other non-determinism issues.
For -z separate-code and -z separate-loadable-segments:
When RW is present, the RX to RW transition is aligned with max-page-size.
When RW is absent, the RX to non-SHF_ALLOC transition should use max-page-size as well.
The code in widenGEP has already been transitioned to only rely on
information provided by VPWidenGEPRecipe directly.
Moving the code directly to VPWidenGEPRecipe::execute completes
the transition for the recipe.
It provides the following advantages:
1. Less indirection, easier to see what's going on.
2. Removes accesses to fields of ILV.
2) in particular ensures that no dependencies on
fields in ILV for GEP code generation are re-introduced.
Reviewed By: Ayal
Differential Revision: https://reviews.llvm.org/D114321
After 04f30795f1, -Wreturn-type has an effect on functions that
contain @try/@catch statements. CheckFallThrough() was missing
a case for ObjCAtTryStmts, leading to a false positive.
(What about the other two places in CheckFallThrough() that handle
CXXTryStmt but not ObjCAtTryStmts?
- I think the last use of CXXTryStmt is dead in practice: 04c6851cd made it so
that calls never add edges to try bodies, and the CFG block for a try
statement is always an empty block containing just the try element itself as
terminator (the try body itself is part of the normal flow of the function
and not connected to the block for the try statement itself. The try statment
cfg block is only connected to the catch bodies, and only reachable from
throw expressions within the try body.)
- The first use of CXXTryStmt might be important. It looks similar to
the code that adds all cfg blocks for try statements as roots of
the reachability graph for the reachability warnings, but I can't
find a way to trigger it. So I'm omitting it for now. The CXXTryStmt
code path seems to only be hit by try statements that are function
bodies without a surrounding compound statements
(`f() try { ... } catch ...`), and those don't exist for ObjC
@try statements.
)
Fixes PR52473.
Differential Revfision: https://reviews.llvm.org/D114660
This is a follow-on suggested in D112634.
Two folds that were added with that patch are subsumed in the call to
decomposeBitTestICmp, and two other folds are potentially inverted.
The deleted folds were very specialized by instcombine standards
because they were restricted to legal integer types based on the data
layout. This generalizes the canonical form independent of target/types.
This change has a reasonable chance of exposing regressions either in
IR or codegen, but I don't have any evidence for either of those yet.
A spot check of asm across several in-tree targets shows variations
that I expect are mostly neutral.
We have one improvement in an existing IR test that I noted with a
comment. Using mask ops might also make more code match with D114272.
Differential Revision: https://reviews.llvm.org/D114386
I encountered this while reviewing an unrelated patch. Will land after
the CI passes.
Reviewed By: #libc, Mordante
Differential Revision: https://reviews.llvm.org/D114673
These benchmarks will be used to test the performance inpact of the next
set of optimization patches.
Reviewed By: ldionne, #libc
Differential Revision: https://reviews.llvm.org/D110501
collectLoopScalars should only add non-uniform nodes to the list if they
are used by a load/store instruction that is marked as CM_Scalarize.
Before this patch, the LV incorrectly marked pointer induction variables
as 'scalar' when they required to be widened by something else,
such as a compare instruction, and weren't used by a node marked as
'CM_Scalarize'. This case is covered by sve-widen-phi.ll.
This change also allows removing some code where the LV tried to
widen the PHI nodes with a stepvector, even though it was marked as
'scalarAfterVectorization'. Now that this code is more careful about
marking instructions that need widening as 'scalar', this code has
become redundant.
Differential Revision: https://reviews.llvm.org/D114373
Also remove `TooFewDims` test which tried to create an invalid AffineMap.
The creation of an invalid AffineMap is rejected by `willBeValidAffineMap`,
as a result we can deprecate the test.
Reviewed By: bondhugula
Differential Revision: https://reviews.llvm.org/D114657
We sometimes end up generating CMOV with constant operands that can be
simplified to CSINC or CSINV under Arm-8.1m. This adds some simple
patterns for them.
Differential Revision: https://reviews.llvm.org/D114349
Codegen from expanded vector operations can end up with unnecessary
CMPZ/CSINC, of the form:
CSXYZ A, B, C1 (CMPZ (CSINC 0, 0, C2, D), 0)
These can be converted to remove the CMPZ and CSINC, depending on the
condition.
if C1==NE -> CSXYZ A, B, C2, D
if C1==EQ -> CSXYZ A, B, NOT(C2), D
Differential Revision: https://reviews.llvm.org/D114013
No functional changes intended.
Before this patch DwarfCompileUnit::createScopeChildrenDIE() and
DwarfCompileUnit::createAndAddScopeChildrenDIE() used to emit child subtrees
and then when all the children get created, attach them to a parent scope DIE.
However, when a DIE doesn't have a parent, all the requests for its unit DIE
fail.
Currently, this is not a big issue since it isn't usually needed to know unit DIE
for a local (function-scoped) entity. But once we introduce lexical blocks as
a valid scope for global variables (static locals) and type DIEs, any requests
for a unit DIE need to be guarded against local scope due to the potential
absence of the DIE's parent.
To avoid the aforementioned issue, this patch refactors a few DwarfCompileUnit
methods to support the idea of attaching a DIE to its parent as close to the
creation of this DIE as possible.
Reviewed By: ellis
Differential Revision: https://reviews.llvm.org/D114350
With the recently introduced tracking as well as D113349, we can
greatly simplify forgetSymbolicName(). In fact, we can simply
replace it with forgetMemoizedResults().
What forgetSymbolicName() used to do is to walk the IR use-def
chain to find all SCEVs that mention the SymbolicName. However,
thanks to use tracking, we can now determine the relevant SCEVs
in a more direct way. D113349 is needed to also clear out the
actual IR to SCEV mapping in ValueExprMap.
Differential Revision: https://reviews.llvm.org/D114263
After backedge taken counts have been calculated, we want to
invalidate all addrecs and dependent expressions in the loop,
because we might compute better results with the newly available
backedge taken counts. Previously this was done with a forgetLoop()
style use-def walk. With recent improvements to SCEV invalidation,
we can instead directly invalidate any SCEVs using addrecs in this
loop. This requires a great deal less subtlety to avoid invalidating
more than necessary, and in particular gets rid of the hack from
D113349. The change is similar to D114263 in spirit.
This patch extends the code that walks memory defs upwards to find
clobbering accesses to also try to optimize the clobbering defining
access.
We should be able to find set the optimized access of our starting def
(KillingDef), if the following holds:
1. It is the first call of getDomMemoryDef for KillingDef (so Current
== KillingDef->getDefiningAccess().
2. No potentially aliasing defs are skipped.
Then if a (partly) aliasing def is encountered, it can be used as
optimized access for KillingDef. No further optimizations can be
applied to KillingDef.
I'd appreciate a careful look, as the existing documentation is not too
clear on what is expected for optimized accesses.
The motivation for this patch is to use the optimized accesses to cover
more cases of redundant stores as follow-up to D111727.
Reviewed By: nikic
Differential Revision: https://reviews.llvm.org/D112313
Instead of removing globals from a module, we, it seems, can just override
AsmPrinter::emitGlobalVariable() to do nothing as NVPTXAsmPrinter already
emitted globals by this time and we don't want to do it twice.
Reviewed By: tra
Differential Revision: https://reviews.llvm.org/D113653
Relative to the previous landing attempt, this introduces an additional
flag on forgetMemoizedResults() to not remove SCEVUnknown phis from
the value map. The invalidation after BECount calculation wants to
leave these alone and skips them in its own use-def walk, but we can
still end up invalidating them via forgetMemoizedResults() if there
is another IR value with the same SCEV. This is intended as a temporary
workaround only, and the need for this should go away once the
getBackedgeTakenInfo() invalidation is refactored in the spirit of
D114263.
-----
This adds validation for consistency of ValueExprMap and
ExprValueMap, and fixes identified issues:
* Addrec construction directly wrote to ValueExprMap in a few places,
without updating ExprValueMap. Add a helper to ensures they stay
consistent. The adjustment in forgetSymbolicName() explicitly
drops the old value from the map, so that we don't rely on it
being overwritten.
* forgetMemoizedResultsImpl() was dropping the SCEV from
ExprValueMap, but not dropping the corresponding entries from
ValueExprMap.
Differential Revision: https://reviews.llvm.org/D113349
Currently, LLD does not support the complete set of ARM group relocations.
Given that I intend to start using these in the Linux kernel [0], let's add
support for these.
This implements the group processing as documented in the ELF psABI. Notably,
this means support is dropped for very far symbol references that also carry a
small component, where the immediate is rotated in such a way that only part of
it wraps to the other end of the 32-bit word. To me, it seems unlikely that
this is something anyone could be relying on, but of course I could be wrong.
[0] https://lore.kernel.org/r/20211122092816.2865873-8-ardb@kernel.org/
Reviewed By: peter.smith, MaskRay
Differential Revision: https://reviews.llvm.org/D114172
Simply use the new PM syntax instead of the deprecated legacy PM
syntax when specifying the opt pipeline in some regression tests.
Differential Revision: https://reviews.llvm.org/D114518
The UpdateTestChecks tool itself does not care about which pass
manager that is used in the opt invocation. So the lit tests that
are verifying the behavior of the UpdateTestChecks tool is updated
to use the new-PM syntax (-passes=) when specifying the pass pipeline
in the test cases that are used for verifying the UpdateTestChecks
tool.
Differential Revision: https://reviews.llvm.org/D114517
The legacy PM is deprecated, so update a bunch of lit tests running
opt to use the new PM syntax when specifying the pipeline.
In this patch focus has been put on test cases for ConstantMerge,
ConstraintElimination, CorrelatedValuePropagation, GlobalDCE,
GlobalOpt, SCCP, TailCallElim and PredicateInfo.
Differential Revision: https://reviews.llvm.org/D114516
VLENB is the length of a vector register in bytes. We use
<vscale x 64 bits> to represent one vector register. The dwarf offset is
VLENB * scalable_offset / 8.
For the mask vector, it occupies one vector register.
Differential Revision: https://reviews.llvm.org/D107432
This diff fixes broken build caused by D108550. Under GCC 5, auto lambdas that capture this require `this->` for member calls.
Reviewed By: bondhugula
Differential Revision: https://reviews.llvm.org/D114659
Add missing tests to improve associative containers code coverage:
- Tests for key_comp() and value_comp() observers
- Tests for std::map and std::multimap value_compare member class
Reviewed by: ldionne, rarutyun, #libc
Differential Revision: https://reviews.llvm.org/D113998
Larger blocks are more convenient for compressions.
Blocks are allocated with MmapNoReserveOrDie to save some memory.
Also it's 15% faster on StackDepotBenchmarkSuite
Depends on D114464.
Reviewed By: morehouse
Differential Revision: https://reviews.llvm.org/D114488
Patrick Oppenlander