Jason Molenda [Mon, 11 Jun 2018 21:35:36 +0000 (21:35 +0000)]
Add DebugNamesDWARFIndex.cpp.
llvm-svn: 334441
Matt Arsenault [Mon, 11 Jun 2018 21:27:41 +0000 (21:27 +0000)]
DAG: Fix extract_subvector combine for a single element
This would fail before because 1x vectors aren't legal,
so instead just use the scalar type.
Avoids regressions in a future AMDGPU commit to add
v4i16/v4f16 as legal types.
Test update is just the one test that this triggers
on in tree now. It wasn't checking anything before.
The result is completely changed since the selects
are eliminated. Not sure if it's considered better
or not.
llvm-svn: 334440
Leonard Mosescu [Mon, 11 Jun 2018 21:19:26 +0000 (21:19 +0000)]
Add a new SBTarget::LoadCore() overload which surfaces errors if the load fails
There was no way to find out what's wrong if SBProcess SBTarget::LoadCore(const char *core_file) failed.
Additionally, the implementation was unconditionally setting sb_process, so it wasn't even possible to check if the return SBProcess is valid.
This change adds a new overload which surfaces the errors and also returns a valid SBProcess only if the core load succeeds:
SBProcess SBTarget::LoadCore(const char *core_file, SBError &error);
Differential Revision: https://reviews.llvm.org/D48049
llvm-svn: 334439
Alex Shlyapnikov [Mon, 11 Jun 2018 21:04:24 +0000 (21:04 +0000)]
[Sanitizers] %tool_options -> %env_tool_opts in allocator_returns_null.cc
Use proper substitution in the common allocator_returns_null.cc test.
llvm-svn: 334438
Petr Hosek [Mon, 11 Jun 2018 20:59:31 +0000 (20:59 +0000)]
[CMake] Allow specifying extra dependencies of bootstrap stage
This allows adding additional bootstrap dependencies to the bootstrap
compiler that may be needed by later stages.
Differential Revision: https://reviews.llvm.org/D47355
llvm-svn: 334437
Craig Topper [Mon, 11 Jun 2018 20:50:58 +0000 (20:50 +0000)]
[X86] Push some variable declarations down into the individual switch cases that need them. NFC
All of the cases are already wrapped in curly braces so declaring a variable there isn't an issue. And the variables aren't assigned or used in the larger scope.
llvm-svn: 334436
Farhana Aleen [Mon, 11 Jun 2018 20:29:31 +0000 (20:29 +0000)]
[SLP] Add testcases of min/max reduction pattern for AMDGPU.
Author: FarhanaAleen
llvm-svn: 334435
Tim Shen [Mon, 11 Jun 2018 19:56:12 +0000 (19:56 +0000)]
Fix incorrect CHECK-LABEL
llvm-svn: 334434
Alex Shlyapnikov [Mon, 11 Jun 2018 19:45:59 +0000 (19:45 +0000)]
[Sanitizers] Move allocator_returns_null.cc test to common.
Summary:
Add allocator_returns_null.cc test to sanitizer_common and
remove all sanitizer-specific ones except:
- HWASan is not covered by sanitizer_common
- TSan allocator does not have comprehensive error reporting yet
Reviewers: vitalybuka
Subscribers: kubamracek, delcypher, #sanitizers, llvm-commits
Differential Revision: https://reviews.llvm.org/D47971
llvm-svn: 334433
Fangrui Song [Mon, 11 Jun 2018 19:42:57 +0000 (19:42 +0000)]
[ELF] Fix copy relocation when two symbols share the same Symbol instance.
In glibc libc.so.6, the multiple versions of sys_errlist share the same Symbol instance. When sys_errlist is copy relocated, we would replace SharedSymbol with Defined in the first iteration of the following loop:
for (SharedSymbol *Sym : getSymbolsAt<ELFT>(SS))
Then in the second iteration, we think the symbol (which has been changed to Defined) is still SharedSymbol and screw up (the address ends up in the `Size` field).
llvm-svn: 334432
Volodymyr Sapsai [Mon, 11 Jun 2018 19:42:27 +0000 (19:42 +0000)]
Mark the test using <experimental/memory_resource> to require c++experimental.
When built against the old libc++ version the test was causing linker error
Undefined symbols for architecture x86_64:
"std::experimental::fundamentals_v1::pmr::new_delete_resource()", referenced from:
void test_evil<WidgetV0, WidgetV0>() in construct_piecewise_pair_evil.pass.cpp.o
void test_evil<WidgetV0, WidgetV1>() in construct_piecewise_pair_evil.pass.cpp.o
void test_evil<WidgetV0, WidgetV2>() in construct_piecewise_pair_evil.pass.cpp.o
void test_evil<WidgetV0, WidgetV3>() in construct_piecewise_pair_evil.pass.cpp.o
void test_evil<WidgetV1, WidgetV0>() in construct_piecewise_pair_evil.pass.cpp.o
void test_evil<WidgetV1, WidgetV1>() in construct_piecewise_pair_evil.pass.cpp.o
void test_evil<WidgetV1, WidgetV2>() in construct_piecewise_pair_evil.pass.cpp.o
...
llvm-svn: 334431
Craig Topper [Mon, 11 Jun 2018 19:20:15 +0000 (19:20 +0000)]
[X86] Reorder some type constraints to force things to be vectors and integer/fp before forcing them to be the same size.
This may be needed by another patch that I'm working on. It should have no effect on any of the generated outputs.
llvm-svn: 334430
Justin Lebar [Mon, 11 Jun 2018 18:57:58 +0000 (18:57 +0000)]
[SCEV] Add transform zext((A * B * ...)<nuw>) --> (zext(A) * zext(B) * ...)<nuw>.
Reviewers: sanjoy
Subscribers: hiraditya, llvm-commits
Differential Revision: https://reviews.llvm.org/D48041
llvm-svn: 334429
Justin Lebar [Mon, 11 Jun 2018 18:57:42 +0000 (18:57 +0000)]
[SCEV] Add nuw/nsw to mul ops in StrengthenNoWrapFlags where safe.
Summary:
Previously we would add them for adds, but not multiplies.
Reviewers: sanjoy
Subscribers: llvm-commits, hiraditya
Differential Revision: https://reviews.llvm.org/D48038
llvm-svn: 334428
Justin Lebar [Mon, 11 Jun 2018 18:57:27 +0000 (18:57 +0000)]
Fix indentation in ScalarEvolution.cpp.
Whitespace-only change. (clang-formatted the whole block.)
llvm-svn: 334427
Krzysztof Parzyszek [Mon, 11 Jun 2018 18:45:52 +0000 (18:45 +0000)]
[Hexagon] Late predicate producers cannot be used as dot-new sources
llvm-svn: 334426
Tim Shen [Mon, 11 Jun 2018 18:44:58 +0000 (18:44 +0000)]
[SCEV] Canonicalize "A /u C1 /u C2" to "A /u (C1*C2)".
Summary: FWIW InstCombine already folds this. Also avoid the case where C1*C2 overflows.
Reviewers: sunfish, sanjoy
Subscribers: hiraditya, bixia, llvm-commits
Differential Revision: https://reviews.llvm.org/D47965
llvm-svn: 334425
Alex Shlyapnikov [Mon, 11 Jun 2018 17:33:53 +0000 (17:33 +0000)]
[Sanitizers] Move pvalloc overflow tests to common.
Summary:
Now all sanitizers with improved allocator error reporting are covered
by these common tests.
Also, add pvalloc-specific checks to LSan.
HWASan is not covered by sanitizer_common, hence its own pvalloc
and other allocator tests.
Reviewers: vitalybuka
Subscribers: srhines, kubamracek, delcypher, #sanitizers, llvm-commits
Differential Revision: https://reviews.llvm.org/D47970
llvm-svn: 334424
Simon Pilgrim [Mon, 11 Jun 2018 17:28:00 +0000 (17:28 +0000)]
[X86][AVX512] Tag AVX5124FMAPS/AVX5124VNNIW with missing scheduler classes
Necessary for D46276 as even though btver2 doesn't use these instructions, its now flagged as complete so complains if ANY instruction isn't tagged.....
UnsupportedFeatures wouldn't help here as these instructions don't appear to have a feature predicate (like a lot of AVX512).
llvm-svn: 334423
Craig Topper [Mon, 11 Jun 2018 17:06:01 +0000 (17:06 +0000)]
[X86] Fix operand order in the shuffle created for blend builtins.
This was broken when the builtin was added in r334249.
llvm-svn: 334422
Matt Morehouse [Mon, 11 Jun 2018 17:05:45 +0000 (17:05 +0000)]
[clang-fuzzer] Modified protobuf and converter to add new signature, remove conditionals.
Changed the function signature and removed conditionals from loop body.
Patch By: emmettneyman
Differential Revision: https://reviews.llvm.org/D47964
llvm-svn: 334421
Stanislav Mekhanoshin [Mon, 11 Jun 2018 16:50:49 +0000 (16:50 +0000)]
[AMDGPU] Do not consider indirect acces through phi for wave limiter
Rational: if there is indirect access that is usually an issue
because load is not ready by the use. However, if use is inside a
loop and load is outside that is potentially an issue for a first
iteration only.
Differential Revision: https://reviews.llvm.org/D47740
llvm-svn: 334420
Aleksandar Beserminji [Mon, 11 Jun 2018 16:50:28 +0000 (16:50 +0000)]
[mips] Fix spill slot for mips3, n64 abi
When program is compiled for mips3 with n64 abi, wrong register class
is used for creating an emergency spill slot. This patch fixes the
correct register class to be chosen.
This patch resolves PR35859.
Thanks to John Baldwin for reporting the issue!
Differential Revision: https://reviews.llvm.org/D47938
llvm-svn: 334419
Reid Kleckner [Mon, 11 Jun 2018 16:50:07 +0000 (16:50 +0000)]
Enable crash recovery tests on Windows, globs work in the lit internal shell now
llvm-svn: 334418
Reid Kleckner [Mon, 11 Jun 2018 16:49:43 +0000 (16:49 +0000)]
[MS] Use mangled names and comdats for string merging with ASan
This should reduce the binary size penalty of ASan on Windows. After
r334313, ASan will add red zones to globals in comdats, so we will still
find OOB accesses to string literals.
llvm-svn: 334417
Craig Topper [Mon, 11 Jun 2018 16:34:10 +0000 (16:34 +0000)]
[X86] Properly account for the immediate being multiplied by 8 in the immediate range checking for BI__builtin_ia32_psrldqi128 and friends.
The limit was set to 1023 which only up to 127*8. It needs to be 2047 to allow 255*8.
llvm-svn: 334416
Martin Probst [Mon, 11 Jun 2018 16:20:13 +0000 (16:20 +0000)]
clang-format: [JS] strict prop init annotation.
Summary:
TypeScript uses the `!` token for strict property initialization
assertions, as in:
class X {
strictPropAsserted!: string;
}
Previously, clang-format would wrap between the `!` and the `:` for
overly long lines. This patch fixes that by generally preventing the
wrap in that location.
Reviewers: krasimir
Subscribers: cfe-commits
Differential Revision: https://reviews.llvm.org/D48030
llvm-svn: 334415
Mikhail Maltsev [Mon, 11 Jun 2018 16:10:06 +0000 (16:10 +0000)]
[Driver] Add aliases for -Qn/-Qy
This patch adds aliases for -Qn (-fno-ident) and -Qy (-fident) which
look less cryptic than -Qn/-Qy. The aliases are compatible with GCC.
Differential Revision: https://reviews.llvm.org/D48021
llvm-svn: 334414
Tobias Grosser [Mon, 11 Jun 2018 15:11:57 +0000 (15:11 +0000)]
Drop unnecessary whitespace [NFCI]
llvm-svn: 334413
Tobias Grosser [Mon, 11 Jun 2018 14:59:28 +0000 (14:59 +0000)]
[ScopBuilder] Slightly improve code structure [NFCI]
First build the surrounding loops and then build up the polyhedral
structures. Before r326664 we had to mix these updates, clean this
up to improve readability (slightly).
llvm-svn: 334412
Pavel Labath [Mon, 11 Jun 2018 14:52:52 +0000 (14:52 +0000)]
Fix tuple getter in std unique pointer pretty-printer
Summary: Check case when _M_t child member is not present.
Reviewers: labath, tberghammer
Reviewed By: labath, tberghammer
Differential Revision: https://reviews.llvm.org/D47932
Patch by Aleksandr Urakov <aleksandr.urakov@jetbrains.com>.
llvm-svn: 334411
Kostya Kortchinsky [Mon, 11 Jun 2018 14:50:31 +0000 (14:50 +0000)]
[scudo] Improve the scalability of the shared TSD model
Summary:
The shared TSD model in its current form doesn't scale. Here is an example of
rpc2-benchmark (with default parameters, which is threading heavy) on a 72-core
machines (defaulting to a `CompactSizeClassMap` and no Quarantine):
- with tcmalloc: 337K reqs/sec, peak RSS of 338MB;
- with scudo (exclusive): 321K reqs/sec, peak RSS of 637MB;
- with scudo (shared): 241K reqs/sec, peak RSS of 324MB.
This isn't great, since the exclusive model uses a lot of memory, while the
shared model doesn't even come close to be competitive.
This is mostly due to the fact that we are consistently scanning the TSD pool
starting at index 0 for an available TSD, which can result in a lot of failed
lock attempts, and touching some memory that needs not be touched.
This CL attempts to make things better in most situations:
- first, use a thread local variable on Linux (intead of pthread APIs) to store
the current TSD in the shared model;
- move the locking boolean out of the TSD: this allows the compiler to use a
register and potentially optimize out a branch instead of reading it from the
TSD everytime (we also save a tiny bit of memory per TSD);
- 64-bit atomic operations on 32-bit ARM platforms happen to be expensive: so
store the `Precedence` in a `uptr` instead of a `u64`. We lose some
nanoseconds of precision and we'll wrap around at some point, but the benefit
is worth it;
- change a `CHECK` to a `DCHECK`: this should never happen, but if something is
ever terribly wrong, we'll crash on a near null AV if the TSD happens to be
null;
- based on an idea by dvyukov@, we are implementing a bound random scan for
an available TSD. This requires computing the coprimes for the number of TSDs,
and attempting to lock up to 4 TSDs in an random order before falling back to
the current one. This is obviously slightly more expansive when we have just
2 TSDs (barely noticeable) but is otherwise beneficial. The `Precedence` still
basically corresponds to the moment of the first contention on a TSD. To seed
on random choice, we use the precedence of the current TSD since it is very
likely to be non-zero (since we are in the slow path after a failed `tryLock`)
With those modifications, the benchmark yields to:
- with scudo (shared): 330K reqs/sec, peak RSS of 327MB.
So the shared model for this specific situation not only becomes competitive but
outperforms the exclusive model. I experimented with some values greater than 4
for the number of TSDs to attempt to lock and it yielded a decrease in QPS. Just
sticking with the current TSD is also a tad slower. Numbers on platforms with
less cores (eg: Android) remain similar.
Reviewers: alekseyshl, dvyukov, javed.absar
Reviewed By: alekseyshl, dvyukov
Subscribers: srhines, kristof.beyls, delcypher, llvm-commits, #sanitizers
Differential Revision: https://reviews.llvm.org/D47289
llvm-svn: 334410
Dylan McKay [Mon, 11 Jun 2018 14:46:48 +0000 (14:46 +0000)]
[AVR] Set trackLivenessAfterRegAlloc
This sets trackLivenessAfterRegAlloc on AVRRegisterInfo.
Most existing targets set this flag. Without it, specific IR inputs
cause LLVM to fail with:
Assertion failed: (getParent()->getProperties().hasProperty( MachineFunctionProperties::Property::TracksLiveness) &&
"Liveness information is accurate"), function livein_begin
file MachineBasicBlock.cpp, line 1354.
With this commit, this no longer happens.
Patch by Peter Nimmervoll.
llvm-svn: 334409
Francois Ferrand [Mon, 11 Jun 2018 14:41:26 +0000 (14:41 +0000)]
clang-format: Introduce BreakInheritanceList option
Summary:
This option replaces the BreakBeforeInheritanceComma option with an
enum, thus introducing a mode where the colon stays on the same line as
constructor declaration:
// When it fits on line:
class A : public B, public C {
...
};
// When it does not fit:
class A :
public B,
public C {
...
};
This matches the behavior of the `BreakConstructorInitializers` option,
introduced in https://reviews.llvm.org/D32479.
Reviewers: djasper, klimek
Reviewed By: djasper
Subscribers: mzeren-vmw, cfe-commits
Differential Revision: https://reviews.llvm.org/D43015
llvm-svn: 334408
Clement Courbet [Mon, 11 Jun 2018 14:37:53 +0000 (14:37 +0000)]
[X86] Fix skylake server scheduling info.
Summary:
This fixes most of the scheduling info for SKX vector operations.
I had to split a lot of the YMM/ZMM classes into separate classes for YMM and ZMM.
The before/after llvm-exegesis analysis are in the phabricator diff.
Subscribers: llvm-commits
Differential Revision: https://reviews.llvm.org/D47721
llvm-svn: 334407
Tobias Grosser [Mon, 11 Jun 2018 14:25:42 +0000 (14:25 +0000)]
Update isl to
isl-0.19-185-g8e9f55ce
This is mainly a maintenance update.
llvm-svn: 334406
Guillaume Chatelet [Mon, 11 Jun 2018 14:10:10 +0000 (14:10 +0000)]
[llvm-exegesis] Fix unhandled error.
Summary: Fixing an unhandled error when calling writeYaml.
Reviewers: courbet
Subscribers: tschuett, llvm-commits
Differential Revision: https://reviews.llvm.org/D48022
llvm-svn: 334405
Sanjay Patel [Mon, 11 Jun 2018 13:51:34 +0000 (13:51 +0000)]
[x86] add scalar cvtt intrinsic tests; NFC
More coverage for the problem noted in D47993 (although these shouldn't be affected by that patch).
llvm-svn: 334404
Pavel Labath [Mon, 11 Jun 2018 13:30:47 +0000 (13:30 +0000)]
Fix build errors on some configurations
It's been reported
<http://lists.llvm.org/pipermail/llvm-commits/Week-of-Mon-
20180611/559616.html>
that template argument deduction for RetryAfterSignal fails if open is
not prefixed with "::".
This should help us build correctly on those platforms and explicitly
specifying the namespace is more correct anyway.
llvm-svn: 334403
Pavel Labath [Mon, 11 Jun 2018 13:22:31 +0000 (13:22 +0000)]
DWARFDebugNames: Fix lookup in dwo files
The getDIESectionOffset function is not correct for split dwarf files
(and will probably be removed in D48009).
This patch implements correct section offset computation for split and
non-split compile units -- we first need to check if the referenced unit
is a skeleton unit, and if it is, we add the die offset to the full unit
base offset (as the full unit is the one which contains the die).
llvm-svn: 334402
Krasimir Georgiev [Mon, 11 Jun 2018 12:53:25 +0000 (12:53 +0000)]
[clang-format] text protos: put entries on separate lines if there is a submessage
Summary:
This patch updates clang-format text protos to put entries of a submessage into separate lines if the submessage contains at least two entries and contains at least one submessage entry.
For example, the entries here are kept on separate lines even if putting them on a single line would be under the column limit:
```
message: {
entry: 1
submessage: { key: value }
}
```
Messages containing a single submessage or several scalar entries can still be put on one line if they fit:
```
message { submessage { key: value } }
message { x: 1 y: 2 z: 3 }
```
Reviewers: sammccall
Reviewed By: sammccall
Subscribers: klimek, cfe-commits
Differential Revision: https://reviews.llvm.org/D46757
llvm-svn: 334401
Alexander Kornienko [Mon, 11 Jun 2018 12:46:48 +0000 (12:46 +0000)]
Add support for arrays in performance-implicit-conversion-in-loop
Summary:
Add support for arrays (and structure that use naked pointers for their iterator, like std::array) in performance-implicit-conversion-in-loop
Reviewers: alexfh
Reviewed By: alexfh
Subscribers: cfe-commits
Patch by Alex Pilkiewicz.
Differential Revision: https://reviews.llvm.org/D47945
llvm-svn: 334400
Pavel Labath [Mon, 11 Jun 2018 10:28:04 +0000 (10:28 +0000)]
Move VersionTuple from clang/Basic to llvm/Support
Summary:
This kind of functionality is useful to other project apart from clang.
LLDB works with version numbers a lot, but it does not have a convenient
abstraction for this. Moving this class to a lower level library allows
it to be freely used within LLDB.
Since this class is used in a lot of places in clang, and it used to be
in the clang namespace, it seemed appropriate to add it to the list of
adopted classes in LLVM.h to avoid prefixing all uses with "llvm::".
Also, I didn't find any tests specific for this class, so I wrote a
couple of quick ones for the more interesting bits of functionality.
Reviewers: zturner, erik.pilkington
Subscribers: mgorny, cfe-commits, llvm-commits
Differential Revision: https://reviews.llvm.org/D47887
llvm-svn: 334399
Roman Lebedev [Mon, 11 Jun 2018 10:21:10 +0000 (10:21 +0000)]
[NFC][AMDGPU] Add tests for all the various IR patterns equivalent to extracting low bits.
Summary:
The idiom recognition seems rather poor.
Only the `@bzhi32_d0` produces `v_bfe_u32`.
But they all should.
This needs to be fixed before D47980 can be re-landed.
Reviewers: mareko, bogner, rampitec, arsenm, tstellar, nhaehnle
Reviewed By: nhaehnle
Subscribers: kzhuravl, wdng, nhaehnle, yaxunl, dstuttard, tpr, t-tye, llvm-commits
Tags: #amdgpu
Differential Revision: https://reviews.llvm.org/D48005
llvm-svn: 334398
Pavel Labath [Mon, 11 Jun 2018 09:32:58 +0000 (09:32 +0000)]
Editline: make #include <codecvt> conditional
My previous patch made this include unconditional. However, it seems it
is not available everywhere. This patch makes us include it only in
configurations we really need it, which should be enough to unblock the
bots.
llvm-svn: 334397
Roman Lebedev [Mon, 11 Jun 2018 09:20:21 +0000 (09:20 +0000)]
[Utils] update_llc_test_checks.py: support AMDGPU backend: AMDGCN, r600 triples
Summary:
Lack of that support has taken me by surprise.
I need to add (or at least look at) some tests for https://reviews.llvm.org/D47980#
1127615,
and i don't really fancy doing that by hand.
The asm pattern is quite similar to that of x86:
https://godbolt.org/g/hfgeds
just with `#` replaced with `;`
Reviewers: spatel, RKSimon, MaskRay, tstellar, arsenm
Reviewed By: arsenm
Subscribers: arsenm, kzhuravl, wdng, yaxunl, dstuttard, tpr, t-tye, rampitec, bogner, mareko, llvm-commits
Tags: #amdgpu
Differential Revision: https://reviews.llvm.org/D48001
llvm-svn: 334396
Guillaume Chatelet [Mon, 11 Jun 2018 09:18:01 +0000 (09:18 +0000)]
[llvm-exegesis] Program should succeed if benchmark returns StringError.
Summary: Fix for https://bugs.llvm.org/show_bug.cgi?id=37759.
Reviewers: courbet
Subscribers: tschuett, llvm-commits
Differential Revision: https://reviews.llvm.org/D48004
llvm-svn: 334395
Mikhail Maltsev [Mon, 11 Jun 2018 09:15:37 +0000 (09:15 +0000)]
[Unittests] Change linker flags of dynamic library tests
A recent change https://reviews.llvm.org/D46898 which had no intended
behavior change, actually modified the linker flags used when linking
the dynamic libraries used by the DynamicLibraryTests unit test. This
made the test fail in our testing environment which runs the tests
from an NFS share. Prior to D46898 the two libraries used by the test
were different (because the library name used to be embedded into the
binary), and after the change they became bit-to-bit identical. This
causes dlopen to return the same handle when these two libraries are
loaded from an NFS share, and the test expects two different handles.
This patch reverts the part of D46898 that is responsible for
changing the linker flags.
Differential Revision: https://reviews.llvm.org/D47469
llvm-svn: 334394
Pavel Labath [Mon, 11 Jun 2018 09:14:26 +0000 (09:14 +0000)]
[cmake] Detect presence of wide-char libedit at build time
Summary:
Instead of hardcoding a list of platforms where libedit is known to have
wide char support we detect this in cmake. The main motivation for this
is attempting to improve compatibility with different versions of
libedit, as the interface of non-wide-char functions varies slightly
between versions.
Reviewers: krytarowski, uweigand, jankratochvil, timshen, beanz
Subscribers: mgorny, lldb-commits
Differential Revision: https://reviews.llvm.org/D47625
llvm-svn: 334393
Simon Atanasyan [Mon, 11 Jun 2018 08:37:19 +0000 (08:37 +0000)]
[ELF] Pass a pointer to InputFile to the getRelocTargetVA to escape dereferencing of nullptr. NFC
llvm-svn: 334392
Clement Courbet [Mon, 11 Jun 2018 07:33:08 +0000 (07:33 +0000)]
[ExynosM1][Sched] Fix resource usage in scheduling model.
This is part of https://reviews.llvm.org/D46356.
llvm-svn: 334391
Simon Atanasyan [Mon, 11 Jun 2018 07:24:31 +0000 (07:24 +0000)]
[ELF][MIPS] Multi-GOT implementation
Almost all entries inside MIPS GOT are referenced by signed 16-bit
index. Zero entry lies approximately in the middle of the GOT. So the
total number of GOT entries cannot exceed ~16384 for 32-bit architecture
and ~8192 for 64-bit architecture. This limitation makes impossible to
link rather large application like for example LLVM+Clang. There are two
workaround for this problem. The first one is using the -mxgot
compiler's flag. It enables using a 32-bit index to access GOT entries.
But each access requires two assembly instructions two load GOT entry
index to a register. Another workaround is multi-GOT. This patch
implements it.
Here is a brief description of multi-GOT for detailed one see the
following link https://dmz-portal.mips.com/wiki/MIPS_Multi_GOT.
If the sum of local, global and tls entries is less than 64K only single
got is enough. Otherwise, multi-got is created. Series of primary and
multiple secondary GOTs have the following layout:
```
- Primary GOT
Header
Local entries
Global entries
Relocation only entries
TLS entries
- Secondary GOT
Local entries
Global entries
TLS entries
...
```
All GOT entries required by relocations from a single input file
entirely belong to either primary or one of secondary GOTs. To reference
GOT entries each GOT has its own _gp value points to the "middle" of the
GOT. In the code this value loaded to the register which is used for GOT
access.
MIPS 32 function's prologue:
```
lui v0,0x0
0: R_MIPS_HI16 _gp_disp
addiu v0,v0,0
4: R_MIPS_LO16 _gp_disp
```
MIPS 64 function's prologue:
```
lui at,0x0
14: R_MIPS_GPREL16 main
```
Dynamic linker does not know anything about secondary GOTs and cannot
use a regular MIPS mechanism for GOT entries initialization. So we have
to use an approach accepted by other architectures and create dynamic
relocations R_MIPS_REL32 to initialize global entries (and local in case
of PIC code) in secondary GOTs. But ironically MIPS dynamic linker
requires GOT entries and correspondingly ordered dynamic symbol table
entries to deal with dynamic relocations. To handle this problem
relocation-only section in the primary GOT contains entries for all
symbols referenced in global parts of secondary GOTs. Although the sum
of local and normal global entries of the primary got should be less
than 64K, the size of the primary got (including relocation-only entries
can be greater than 64K, because parts of the primary got that overflow
the 64K limit are used only by the dynamic linker at dynamic link-time
and not by 16-bit gp-relative addressing at run-time.
The patch affects common LLD code in the following places:
- Added new hidden -mips-got-size flag. This flag required to set low
maximum size of a single GOT to be able to test the implementation using
small test cases.
- Added InputFile argument to the getRelocTargetVA function. The same
symbol referenced by GOT relocation from different input file might be
allocated in different GOT. So result of relocation depends on the file.
- Added new ctor to the DynamicReloc class. This constructor records
settings of dynamic relocation which used to adjust address of 64kb page
lies inside a specific output section.
With the patch LLD is able to link all LLVM+Clang+LLD applications and
libraries for MIPS 32/64 targets.
Differential revision: https://reviews.llvm.org/D31528
llvm-svn: 334390
Clement Courbet [Mon, 11 Jun 2018 07:00:08 +0000 (07:00 +0000)]
[X86] Explicitly mark unsupported classes in scheduling models.
Summary: In preparation for D47721. HSW and SNB still define unsupported
classes as they are used by KNL and generic models respectively.
Reviewers: RKSimon
Subscribers: llvm-commits
Differential Revision: https://reviews.llvm.org/D47763
llvm-svn: 334389
Hans Wennborg [Mon, 11 Jun 2018 06:54:23 +0000 (06:54 +0000)]
[MS ABI] Mangle unnamed empty enums (PR37723)
Differential Revision: https://reviews.llvm.org/D47875
llvm-svn: 334388
Craig Topper [Mon, 11 Jun 2018 06:22:41 +0000 (06:22 +0000)]
[X86] Add encoding tests for avx5124fmaps and avx5124vnni instructions.
I forgot to git add these in r333812
llvm-svn: 334387
Craig Topper [Mon, 11 Jun 2018 06:20:24 +0000 (06:20 +0000)]
[X86] Add test files for upgrade of vbmi2 expand load and compress store intrinsics that was done in r334381.
llvm-svn: 334386
Craig Topper [Mon, 11 Jun 2018 06:18:29 +0000 (06:18 +0000)]
[X86] Remove masking from dbpsadbw builtins, use select builtin instead.
llvm-svn: 334385
Craig Topper [Mon, 11 Jun 2018 06:18:22 +0000 (06:18 +0000)]
[X86] Remove masking from dbpsadbw intrinsics, use select in IR instead.
llvm-svn: 334384
Daniel Cederman [Mon, 11 Jun 2018 05:50:08 +0000 (05:50 +0000)]
[Sparc] Add support for 13-bit PIC
Summary: When compiling with -fpic, in contrast to -fPIC, use only the
immediate field to index into the GOT. This saves space if the GOT is
known to be small. The linker will warn if the GOT is too large for
this method.
Reviewers: jyknight, venkatra
Reviewed By: jyknight
Subscribers: brad, fedor.sergeev, jrtc27, llvm-commits
Differential Revision: https://reviews.llvm.org/D47136
llvm-svn: 334383
Brock Wyma [Mon, 11 Jun 2018 01:39:34 +0000 (01:39 +0000)]
[CodeView] Omit forward references for unnamed structs and unions
Codeview references to unnamed structs and unions are expected to refer to the
complete type definition instead of a forward reference so Visual Studio can
resolve the type properly.
Differential Revision: https://reviews.llvm.org/D32498
llvm-svn: 334382
Craig Topper [Mon, 11 Jun 2018 01:25:22 +0000 (01:25 +0000)]
[X86] Remove and autoupgrade the expandload and compressstore intrinsics.
We use the target independent intrinsics now.
llvm-svn: 334381
Craig Topper [Sun, 10 Jun 2018 23:15:49 +0000 (23:15 +0000)]
[TableGen] Make better use of std::map::emplace and emplace construct the object in the map rather than moving it into it. Remove a use std::map::find by remembering the return from emplace.
llvm-svn: 334380
Craig Topper [Sun, 10 Jun 2018 23:15:48 +0000 (23:15 +0000)]
[TableGen] Combine two constructors by taking vectors by value instead of trying to support combininations for rvalue and lvalue references.
llvm-svn: 334379
Sanjay Patel [Sun, 10 Jun 2018 23:09:50 +0000 (23:09 +0000)]
[DAGCombiner] match vector compare and select sizes with extload operand (PR37427)
This patch started off much more general and ambitious, but it's been a nightmare
seeing all the ways x86 vector codegen can go wrong.
So the code is still structured to allow extending easily, but it's currently
limited in several ways:
1. Only handle cases with an extending load.
2. Only handle cases with a zero constant compare.
3. Ignore setcc with vector bitmask (SetCCWidth != 1) - so AVX512 should be unaffected.
The motivating case from PR37427:
https://bugs.llvm.org/show_bug.cgi?id=37427
...is the 1st test, and that shows the expected win - we eliminated the unnecessary
intermediate cast.
There's a clear regression in the last test (sgt_zero_fp_select) because we longer
recognize a 'SHRUNKBLEND' opportunity. I think that general problem is also present
in sgt_zero, so I'll try to fix that in a follow-up. We need to match a sign-bit
setcc from a sign-extended operand and remove it.
Differential Revision: https://reviews.llvm.org/D47330
llvm-svn: 334378
Craig Topper [Sun, 10 Jun 2018 21:48:24 +0000 (21:48 +0000)]
[X86] Miscellaneous fixes to get the load folding table generator to work again.
llvm-svn: 334377
Florian Hahn [Sun, 10 Jun 2018 21:06:24 +0000 (21:06 +0000)]
Revert r334374 [TableGen] Move some shared_ptrs to avoid unnecessary copies (NFC).
This breaks some builders.
llvm-svn: 334376
Zachary Turner [Sun, 10 Jun 2018 20:57:14 +0000 (20:57 +0000)]
Attempt 3: Resubmit "[Support] Expose flattenWindowsCommandLine."
I took some liberties and quoted fewer characters than before,
based on an article from MSDN which says that only certain characters
cause an arg to require quoting. This seems to be incorrect, though,
and worse it seems to be a difference in Windows version. The bot
that fails is Windows 7, and I can't reproduce the failure on Win
10. But it's definitely related to quoting and special characters,
because both tests that fail have a * in the argument, which is one
of the special characters that would cause an argument to be quoted
before but not any longer after the new patch.
Since I don't have Win 7, all I can do is just guess that I need to
restore the old quoting rules. So this patch does that in hopes that
it fixes the problem on Windows 7.
llvm-svn: 334375
Florian Hahn [Sun, 10 Jun 2018 20:38:52 +0000 (20:38 +0000)]
[TableGen] Move some shared_ptrs to avoid unnecessary copies (NFC).
Those changes were suggested post-commit for D47463.
llvm-svn: 334374
Roman Lebedev [Sun, 10 Jun 2018 20:32:03 +0000 (20:32 +0000)]
Revert rL334371 / D47980: "[InstCombine] Fold (x << y) >> y -> x & (-1 >> y)"
test/Transforms/InstCombine/AMDGPU/amdgcn-intrinsics.ll broke,
and i did not notice because i did not build that backend.
llvm-svn: 334373
Roman Lebedev [Sun, 10 Jun 2018 20:10:13 +0000 (20:10 +0000)]
[InstCombine] Fold (x >> y) << y -> x & (-1 << y)
Summary:
We already do it for matching splat constants, but not just values.
Further improvements for non-matching splat constants, as noted in
https://reviews.llvm.org/D46760#
1123713 will be needed,
but i'd prefer to do that as a follow-up.
https://bugs.llvm.org/show_bug.cgi?id=37603
https://rise4fun.com/Alive/cplX
https://rise4fun.com/Alive/0HF
Reviewers: spatel, craig.topper
Reviewed By: spatel
Subscribers: llvm-commits
Differential Revision: https://reviews.llvm.org/D47981
llvm-svn: 334372
Roman Lebedev [Sun, 10 Jun 2018 20:10:06 +0000 (20:10 +0000)]
[InstCombine] Fold (x << y) >> y -> x & (-1 >> y)
Summary:
We already do it for splat constants, but not just values.
Also, undef cases are mostly non-functional.
https://bugs.llvm.org/show_bug.cgi?id=37603
https://rise4fun.com/Alive/cplX
Reviewers: spatel, craig.topper
Reviewed By: spatel
Subscribers: llvm-commits
Differential Revision: https://reviews.llvm.org/D47980
llvm-svn: 334371
Roman Lebedev [Sun, 10 Jun 2018 20:10:00 +0000 (20:10 +0000)]
[NFC][InstCombine] Revisit tests for D47980 / D47981 once more.
llvm-svn: 334370
Ahmed Bougacha [Sun, 10 Jun 2018 19:38:26 +0000 (19:38 +0000)]
[debuginfo-tests] Always use the system python to invoke llgdb.py.
/usr/bin/env is recommended as a cross-platform way to find python. But:
- we're only using lldb on darwin, where we know python (or at least,
the xcrun-style shortcut) is in /usr/bin/
- the python interpreter in LLDB comes from /S/L/F:
$ otool -L Contents/SharedFrameworks/LLDB.framework/LLDB | grep Python
/System/Library/Frameworks/Python.framework/Versions/2.7/Python
so when we use the lldb python module, it calls into the swig/python
support in the lldb framework, and if there's a mismatch between the
interpreter and the linked python, weird things occur.
In theory, I believe this should be done by:
- looking for the LLDB framework (llgdb.py does some of that)
- finding the binary inside the framework
- looking for the Python it was linked against (otool -L)
- finding the interpreter executable inside the Python.framework
But in practice, that's only different if we use a custom LLDB
framework/pythonpath when running these tests, and AFAIK nobody does
that right now, so the code would be dead anyway.
Don't pretend we can use any arbitrary python: just use the system one.
Differential Revision: https://reviews.llvm.org/D47967
llvm-svn: 334369
Craig Topper [Sun, 10 Jun 2018 18:55:37 +0000 (18:55 +0000)]
[X86] Add expandload and compresstore fast-isel tests for avx512f and avx512vl. Update existing tests for avx512vbmi2 to use target independent intrinsics.
llvm-svn: 334368
Sanjay Patel [Sun, 10 Jun 2018 17:42:12 +0000 (17:42 +0000)]
[x86] add tests for potentially miscompiling cvttp2si (PR37751); NFC
llvm-svn: 334367
Craig Topper [Sun, 10 Jun 2018 17:27:05 +0000 (17:27 +0000)]
[X86] Use target independent masked expandload and compressstore intrinsics to implement expandload/compressstore builtins.
Summary: We've had these target independent intrinsics for at least a year and a half. Looks like they do exactly what we need here and the backend already supports them.
Reviewers: RKSimon, delena, spatel, GBuella
Reviewed By: RKSimon
Subscribers: cfe-commits, llvm-commits
Differential Revision: https://reviews.llvm.org/D47693
llvm-svn: 334366
Craig Topper [Sun, 10 Jun 2018 17:27:00 +0000 (17:27 +0000)]
[X86] Remove GCCBuiltin from some intrinsics so we can custom handle them in clang.
llvm-svn: 334365
Alexander Polyakov [Sun, 10 Jun 2018 14:58:29 +0000 (14:58 +0000)]
[lldb-mi] Re-implement MI -exec-step command.
Summary: Now -exec-step uses SB API instead of HandleCommand hack.
Reviewers: aprantl, clayborg, labath, stella.stamenova
Reviewed By: aprantl
Subscribers: ki.stfu, lldb-commits
Differential Revision: https://reviews.llvm.org/D47838
llvm-svn: 334364
Peter Wu [Sun, 10 Jun 2018 11:17:47 +0000 (11:17 +0000)]
[ASAN] Fix crash on i?86-linux (32-bit) against glibc 2.27 and later
Summary:
Running sanitized 32-bit x86 programs on glibc 2.27 crashes at startup, with:
ERROR: AddressSanitizer: SEGV on unknown address 0xf7a8a250 (pc 0xf7f807f4 bp 0xff969fc8 sp 0xff969f7c T16777215)
The signal is caused by a WRITE memory access.
#0 0xf7f807f3 in _dl_get_tls_static_info (/lib/ld-linux.so.2+0x127f3)
#1 0xf7a92599 (/lib/libasan.so.5+0x112599)
#2 0xf7a80737 (/lib/libasan.so.5+0x100737)
#3 0xf7f7e14f in _dl_init (/lib/ld-linux.so.2+0x1014f)
#4 0xf7f6eb49 (/lib/ld-linux.so.2+0xb49)
The problem is that glibc changed the calling convention for the GLIBC_PRIVATE
symbol that sanitizer uses (even when it should not, GLIBC_PRIVATE is exactly
for symbols that can change at any time, be removed etc.), see
https://sourceware.org/ml/libc-alpha/2017-08/msg00497.html
Fixes https://github.com/google/sanitizers/issues/954
Patch By: Jakub Jelinek
Reviewed By: vitalybuka, Lekensteyn
Differential Revison: https://reviews.llvm.org/D44623
llvm-svn: 334363
Ivan A. Kosarev [Sun, 10 Jun 2018 09:28:10 +0000 (09:28 +0000)]
[NEON] Support VST1xN intrinsics in AArch32 mode (Clang part)
We currently support them only in AArch64. The NEON Reference,
however, says they are 'ARMv7, ARMv8' intrinsics.
Differential Revision: https://reviews.llvm.org/D47446
llvm-svn: 334362
Ivan A. Kosarev [Sun, 10 Jun 2018 09:27:27 +0000 (09:27 +0000)]
[NEON] Support VST1xN intrinsics in AArch32 mode (LLVM part)
We currently support them only in AArch64. The NEON Reference,
however, says they are 'ARMv7, ARMv8' intrinsics.
Differential Revision: https://reviews.llvm.org/D47447
llvm-svn: 334361
Craig Topper [Sun, 10 Jun 2018 06:43:48 +0000 (06:43 +0000)]
[X86] Fix forward declaration in a test case that was messed up in r334358
llvm-svn: 334360
Craig Topper [Sun, 10 Jun 2018 06:01:42 +0000 (06:01 +0000)]
[X86] Remove masking from the 512-bit packed floating point add/sub/mul/div builtins. Use select in IR instead.
llvm-svn: 334359
Craig Topper [Sun, 10 Jun 2018 06:01:36 +0000 (06:01 +0000)]
[X86] Remove masking from the 512-bit masked floating point add/sub/mul/div intrinsics. Use a select in IR instead.
llvm-svn: 334358
Fangrui Song [Sun, 10 Jun 2018 04:53:14 +0000 (04:53 +0000)]
Cleanup. NFC
llvm-svn: 334357
Zachary Turner [Sun, 10 Jun 2018 03:16:25 +0000 (03:16 +0000)]
Revert "Resubmit "[Support] Expose flattenWindowsCommandLine.""
This reverts commit
65243b6d19143cb7a03f68df0169dcb63e8b4632.
Seems like it's not a flake. It might have something to do with
the '*' character being in a command line.
llvm-svn: 334356
Zachary Turner [Sun, 10 Jun 2018 02:46:11 +0000 (02:46 +0000)]
Resubmit "[Support] Expose flattenWindowsCommandLine."
There were a few linux compilation failures, but other than that
I think this was just a flake that caused the tests to fail. I'm
going to resubmit and see if the failures go away, if not I'll
revert again.
llvm-svn: 334355
Zachary Turner [Sat, 9 Jun 2018 23:07:39 +0000 (23:07 +0000)]
Revert "[Support] Expose flattenWindowsCommandLine."
This reverts commit
10d2e88e87150a35dc367ba30716189d2af26774.
This is causing some test failures for some reason, reverting
while I investigate.
llvm-svn: 334354
Zachary Turner [Sat, 9 Jun 2018 22:44:44 +0000 (22:44 +0000)]
[Support] Expose flattenWindowsCommandLine.
This function was internal to Program.inc, but I've needed this
on several occasions when I've had to use CreateProcess without
llvm's sys::Execute functions. In doing so, I noticed that the
function was written using unsafe C-string access and was pretty
hard to understand / make sense of, so I've also re-written the
functions to use more modern LLVM constructs.
llvm-svn: 334353
Reka Kovacs [Sat, 9 Jun 2018 21:08:27 +0000 (21:08 +0000)]
[analyzer] Clean up the program state map of DanglingInternalBufferChecker.
Symbols are cleaned up from the program state map when they go out of scope.
Memory regions are cleaned up when the corresponding object is destroyed, and
additionally in 'checkDeadSymbols' in case destructor modeling was incomplete.
Differential Revision: https://reviews.llvm.org/D47416
llvm-svn: 334352
Simon Pilgrim [Sat, 9 Jun 2018 16:08:25 +0000 (16:08 +0000)]
[CostModel][X86] Add 'select' style shuffle costs tests (PR33744)
llvm-svn: 334351
Alexander Polyakov [Sat, 9 Jun 2018 15:11:37 +0000 (15:11 +0000)]
[lldb, lldb-mi] Re-implement MI -exec-continue command.
Summary: Now -exec-continue command uses SB API to resume target's process.
Reviewers: aprantl, clayborg, labath
Reviewed By: clayborg
Subscribers: apolyakov, labath, ki.stfu, llvm-commits, lldb-commits
Differential Revision: https://reviews.llvm.org/D47415
llvm-svn: 334350
Roman Lebedev [Sat, 9 Jun 2018 14:01:46 +0000 (14:01 +0000)]
[NFC][InstCombine] More tests for (x >> y) << y -> x & (-1 << y) fold.
Followup for rL334347.
The fold is also valid for ashr.
https://rise4fun.com/Alive/0HF
https://bugs.llvm.org/show_bug.cgi?id=37603
https://reviews.llvm.org/D46760#
1123713
https://rise4fun.com/Alive/cplX
llvm-svn: 334349
Reka Kovacs [Sat, 9 Jun 2018 13:03:49 +0000 (13:03 +0000)]
[analyzer] Add dangling internal buffer check.
This check will mark raw pointers to C++ standard library container internal
buffers 'released' when the objects themselves are destroyed. Such information
can be used by MallocChecker to warn about use-after-free problems.
In this first version, 'std::basic_string's are supported.
Differential Revision: https://reviews.llvm.org/D47135
llvm-svn: 334348
Roman Lebedev [Sat, 9 Jun 2018 09:27:43 +0000 (09:27 +0000)]
[NFC][InstCombine] Tests for (x >> y) << y -> x & (-1 << y) fold.
We already do it for splat constants, but not just values.
Also, undef cases are mostly non-functional.
https://bugs.llvm.org/show_bug.cgi?id=37603
https://reviews.llvm.org/D46760#
1123713
https://rise4fun.com/Alive/cplX
llvm-svn: 334347
Roman Lebedev [Sat, 9 Jun 2018 09:27:39 +0000 (09:27 +0000)]
[NFC][InstCombine] Tests for (x << y) >> y -> x & (-1 >> y) fold.
We already do it for splat constants, but not just values.
Also, undef cases are mostly non-functional.
https://bugs.llvm.org/show_bug.cgi?id=37603
https://rise4fun.com/Alive/cplX
llvm-svn: 334346
Gabor Buella [Sat, 9 Jun 2018 09:19:40 +0000 (09:19 +0000)]
[X86] NFC Use member initialization in X86Subtarget
The separate initializeEnvironment function was sort of
useless since r217071.
ARM did this move already with r273556.
llvm-svn: 334345
Serge Pavlov [Sat, 9 Jun 2018 05:19:45 +0000 (05:19 +0000)]
Use uniform mechanism for OOM errors handling
This is a recommit of r333506, which was reverted in r333518.
The original commit message is below.
In r325551 many calls of malloc/calloc/realloc were replaces with calls of
their safe counterparts defined in the namespace llvm. There functions
generate crash if memory cannot be allocated, such behavior facilitates
handling of out of memory errors on Windows.
If the result of *alloc function were checked for success, the function was
not replaced with the safe variant. In these cases the calling function made
the error handling, like:
T *NewElts = static_cast<T*>(malloc(NewCapacity*sizeof(T)));
if (NewElts == nullptr)
report_bad_alloc_error("Allocation of SmallVector element failed.");
Actually knowledge about the function where OOM occurred is useless. Moreover
having a single entry point for OOM handling is convenient for investigation
of memory problems. This change removes custom OOM errors handling and
replaces them with calls to functions `llvm::safe_*alloc`.
Declarations of `safe_*alloc` are moved to a separate include file, to avoid
cyclic dependency in SmallVector.h
Differential Revision: https://reviews.llvm.org/D47440
llvm-svn: 334344
Craig Topper [Sat, 9 Jun 2018 05:04:20 +0000 (05:04 +0000)]
Use SmallPtrSet instead of SmallSet in places where we iterate over the set.
SmallSet forwards to SmallPtrSet for pointer types. SmallPtrSet supports iteration, but a normal SmallSet doesn't. So if it wasn't for the forwarding, this wouldn't work.
These places were found by hiding the begin/end methods in the SmallSet forwarding
llvm-svn: 334343
Rui Ueyama [Sat, 9 Jun 2018 00:58:05 +0000 (00:58 +0000)]
Fix build broken by r334341.
llvm-svn: 334342