Summary:
Use X86ISD::VFPROUND in the instruction isel patterns. Add new patterns for ISD::FP_ROUND to maintain support for fptrunc in IR.
In the process I found a couple duplicate isel patterns which I also deleted in this patch.
Reviewers: RKSimon, spatel
Reviewed By: RKSimon
Subscribers: llvm-commits
Differential Revision: https://reviews.llvm.org/D56991
llvm-svn: 351762
Summary:
For compress, a select node doesn't semantically reflect the behavior of the instruction. The mask would have holes in it, but the resulting write is to contiguous elements at the bottom of the vector.
Furthermore, as far as the compressing and expanding is concerned the behavior is depended on the mask. You can't just have an expand/compress node that only reads the input vector. That node would have no meaning by itself.
This all only works because we pattern match the compress/expand+select back to the instruction. But conceivably an optimization of the select could break the pattern and leave something meaningless.
This patch modifies the expand and compress node to take the mask and passthru as additional inputs and gets rid of the select all together.
Reviewers: RKSimon, spatel
Reviewed By: RKSimon
Subscribers: llvm-commits
Differential Revision: https://reviews.llvm.org/D57002
llvm-svn: 351761
Fixes two problems with GCNHazardRecognizer:
1. It only scans up to 5 instructions emitted earlier.
2. It does not take control flow into account. An earlier instruction
from the previous basic block is not necessarily a predecessor.
At the same time a real predecessor block is not scanned.
The patch provides a way to distinguish between scheduler and
hazard recognizer mode. It is OK to work with emitted instructions
in the scheduler because we do not really know what will be emitted
later and its order. However, when pass works as a hazard recognizer
the schedule is already finalized, and we have full access to the
instructions for the whole function, so we can properly traverse
predecessors and their instructions.
Differential Revision: https://reviews.llvm.org/D56923
llvm-svn: 351759
This is a remnant from before the gn build had a working config.h.
Defining LLVM_LIBXML2_ENABLED only for targets that depend on build/libs/xml is
nice in that only some of the codebase needs to be rebuilt when
llvm_enable_libxml2 changes -- but config.h already defines it and defining it
there and then redundantly a second time for some targets is worse than having
it just in config.h.
No behavior change.
Differential Revision: https://reviews.llvm.org/D56908
llvm-svn: 351758
This version of gcc seems to be having issues with raw literals inside macro
arguments. I change the string to use regular string literals instead.
llvm-svn: 351756
D56777 added +1cy local forwarding penalty for horizontal operations, but this penalty only affects sse2/xmm variants, the mmx variants don't suffer the penalty.
Confirmed with @andreadb
llvm-svn: 351755
These are copied from the sibling x86 file. I'm not sure which
of the current outputs (if any) is considered optimal, but
someone more familiar with AArch may want to take a look.
llvm-svn: 351754
The regression test is reduced from the example shown in D56281.
This does raise a question as noted in the test file: do we want
to handle this pattern? I don't have a motivating example for
that on x86 yet, but it seems like we could have that pattern
there too, so we could avoid the back-and-forth using a shuffle.
llvm-svn: 351753
Pass -ccc-install-dir explicitly as the compilation database code does
not pass argv[0] to getMainExecutable(), while some systems require it
to return the correct path. Since the relevant code is apparently only
applicable to Darwin, just pass correct -ccc-install-dir to make
the tests pass on *BSD systems.
Differential Revision: https://reviews.llvm.org/D56976
llvm-svn: 351752
The readability-else-after-return check should be smarter about cases where the
variable defined in the condition is used in the `else` branch. This patch makes
it just ignore such cases, but alternative solutions may be better (added a
FIXME).
llvm-svn: 351751
Methods can now be qualified with address spaces to prevent
undesirable conversions to generic or to provide custom
implementation to be used if the object is located in certain
memory segments.
This commit extends parsing and standard C++ overloading to
work for an address space of a method (i.e. implicit 'this'
parameter).
Differential Revision: https://reviews.llvm.org/D55850
llvm-svn: 351747
Summary:
This patch adds two unit-tests that are the result of reducing a crashing TU
when running ExprMutAnalyzer over it. They are added only to ensure the regression
that has been fixed with https://reviews.llvm.org/D56444 don't creep back.
Reviewers: aaron.ballman, sammccall, rsmith, george.karpenkov
Reviewed By: sammccall
Subscribers: baloghadamsoftware, a.sidorin, Szelethus, donat.nagy, dkrupp, cfe-commits
Differential Revision: https://reviews.llvm.org/D56917
llvm-svn: 351743
r327630 introduced new write definitions for float/vector loads.
Before that revision, WriteLoad was used by both integer/float (scalar/vector)
load. So, WriteLoad had to conservatively declare a latency to 5cy. That is
because the load-to-use latency for float/vector load is 5cy.
Now that we have dedicated writes for float/vector loads, there is no reason why
we should keep the latency of WriteLoad to 5cy. At the moment, WriteLoad is only
used by scalar integer loads only; we can assume an optimstic 3cy latency for
them.
This patch changes that latency from 5cy to 3cy, and regenerates the affected
scheduling/mca tests.
Differential Revision: https://reviews.llvm.org/D56922
llvm-svn: 351742
The ACLE states that 64-bit crc32, wsr, rsr and rbit operands are
uint64_t so we should have the clang builtin match this description
- which is what we already do for AArch32.
Differential Revision: https://reviews.llvm.org/D56852
llvm-svn: 351740
Summary: See rC351531 for the introduction of getStripPluginsAdjuster.
Reviewers: alexfh
Subscribers: xazax.hun, cfe-commits
Differential Revision: https://reviews.llvm.org/D56902
llvm-svn: 351738
all missed!
Thanks to Alex Bradbury for pointing this out, and the fact that I never
added the intended `legacy` anchor to the developer policy. Add that
anchor too. With hope, this will cause the links to all resolve
successfully.
llvm-svn: 351731
Reports correct size and tags when either size is not power of two
or offset to bad granule is not zero.
Differential revision: https://reviews.llvm.org/D56603
llvm-svn: 351730
This patch introduces the field `ExpressionSize` in SCEV. This field is
calculated only once on SCEV creation, and it represents the complexity of
this SCEV from arithmetical point of view (not from the point of the number
of actual different SCEV nodes that are used in the expression). Roughly
saying, it is the number of operands and operations symbols when we print this
SCEV.
A formal definition is following: if SCEV `X` has operands
`Op1`, `Op2`, ..., `OpN`,
then
Size(X) = 1 + Size(Op1) + Size(Op2) + ... + Size(OpN).
Size of SCEVConstant and SCEVUnknown is one.
Expression size may be used as a universal way to limit SCEV transformations
for huge SCEVs. Currently, we have a bunch of options that represents various
limits (such as recursion depth limit) that may not make any sense from the
point of view of a LLVM users who is not familiar with SCEV internals, and all
these different options pursue one goal. A more general rule that may
potentially allow us to get rid of this redundancy in options is "do not make
transformations with SCEVs of huge size". It can apply to all SCEV traversals
and transformations that may need to visit a SCEV node more than once, hence
they are prone to combinatorial explosions.
This patch only introduces SCEV sizes calculation as NFC, its utilization will
be introduced in follow-up patches.
Differential Revision: https://reviews.llvm.org/D35989
Reviewed By: reames
llvm-svn: 351725
Patch rL322965 changed how intrinsics for memset and memzero
were defined. This causes a regression in LLGO making it no
longer buidable. In addition to that one pass was renamed,
so this also addresses the pass naming disparity.
I removed all split stack related bits from this patch as
as asked.
Differential Revision: https://reviews.llvm.org/D56638
llvm-svn: 351722
This updates the AVR Select8/Select16 expansion code so that, when
inserting the two basic blocks for true and false conditions, any
existing fallthrough on the previous block is preserved.
Prior to this patch, if the block before the Select pseudo fell through
to the subsequent block, two new basic blocks would be inserted at the
prior fallthrough point, changing the fallthrough destination.
The predecessor or successor lists were not updated, causing the
BranchFolding pass at -O1 and above the rearrange basic blocks, causing
an infinite loop. Not to mention the unconditional fallthrough to the
true block is incorrect in of itself.
This patch modifies the Select8/16 expansion so that, if inserting true
and false basic blocks at a fallthrough point, the implicit branch is
preserved by means of an explicit, unconditional branch to the previous
fallthrough destination.
Thanks to Carl Peto for reporting this bug.
This fixes avr-rust bug https://github.com/avr-rust/rust/issues/123.
llvm-svn: 351721
Prior to this, the code was missing AVR-specific relocation logic in
RelocVisitor.h.
This patch teaches RelocVisitor about R_AVR_16 and R_AVR_32.
Debug information is emitted in the final object file, and understood by
'avr-readelf --debug-dump' from AVR-GCC.
llvm-dwarfdump is yet to understand how to dump AVR DWARF symbols.
llvm-svn: 351720
This reverts commit r351718.
Carl pointed out that the unit test could be improved.
This patch will be recommitted once the test is made more resilient.
llvm-svn: 351719
This updates the AVR Select8/Select16 expansion code so that, when
inserting the two basic blocks for true and false conditions, any
existing fallthrough on the previous block is preserved.
Prior to this patch, if the block before the Select pseudo fell through
to the subsequent block, two new basic blocks would be inserted at the
prior fallthrough point, changing the fallthrough destination.
The predecessor or successor lists were not updated, causing the
BranchFolding pass at -O1 and above the rearrange basic blocks, causing
an infinite loop. Not to mention the unconditional fallthrough to the
true block is incorrect in of itself.
This patch modifies the Select8/16 expansion so that, if inserting true
and false basic blocks at a fallthrough point, the implicit branch is
preserved by means of an explicit, unconditional branch to the previous
fallthrough destination.
Thanks to Carl Peto for reporting this bug.
This fixes avr-rust bug https://github.com/avr-rust/rust/issues/123.
llvm-svn: 351718
Summary:
Make Sanitizer Coverage work when compiled work when compiler-rt
is compiled with MSVC.
The previous solution did not work for MSVC because MSVC tried to
align the .SCOV$CZ section even though we used
__declspec(align(1)) on its only symbol:
__stop___sancov_cntrs.
Because the counter array is composed
of 1 byte elements, it does not always end on an 8 or 4 byte
boundary. This means that padding was sometimes added to
added to align the next section, .SCOV$CZ.
Use a different strategy now: instead of only instructing
the compiler not to align the symbol, make the section
one byte long by making its only symbol a uint8_t, so that
the linker won't try to align it.
Reviewers: morehouse, rnk
Reviewed By: rnk
Subscribers: kubamracek
Differential Revision: https://reviews.llvm.org/D56866
llvm-svn: 351714
This implements `mi_extra_init(...)` for the ASan allocator on
Darwin and uses the `__lsan::GetAllocatorGlobalRange(...)` function
to retrieve the allocator pointer and size.
rdar://problem/45284065
llvm-svn: 351713
`sanitizer_malloc_introspection_t` and initialize them to zero.
We allow sanitizer implementations to perform different initialization
by defining `COMMON_MALLOC_HAS_EXTRA_INTROSPECTION_INIT` to be `1`
and providing an implementation of `mi_extra_init(...)`.
We use these changes in future patches to implement malloc zone enumeration.
rdar://problem/45284065
llvm-svn: 351712
enumerator.
This is done by defining `COMMON_MALLOC_HAS_ZONE_ENUMERATOR` to `1` and
then by providing an implementation of the `mi_enumerator(...)` function.
If a custom implementation isn't desired the macro is set to `0` which
causes a stub version (that fails) to be used.
Currently all Darwin sanitizers that have malloc implementations define
this to be `0` so there is no functionality change.
rdar://problem/45284065
llvm-svn: 351711
-DNDEBUG is no longer needed now that we don't enable assertions,
modules should improve build times for the second stage.
Differential Revision: https://reviews.llvm.org/D56972
llvm-svn: 351709
For some reason we were missing tests for several unmasked conversion intrinsics, but had their mask form.
Also use a non-default rounding mode on some tests to provide better coverage for a future patch.
llvm-svn: 351708
Craig Topper