George Rimar [Tue, 17 Jul 2018 08:04:27 +0000 (08:04 +0000)]
[ELF] - Check we are able to report errors when DW_AT_decl_file is invalid.
We did not try to support this intentionally but have
an error handling and reporting logic that can take care
of that and hence needs a test.
llvm-svn: 337250
Roman Lebedev [Tue, 17 Jul 2018 07:12:08 +0000 (07:12 +0000)]
Harden/relax clang/test/CodeGen/opt-record-MIR.c test
Summary:
If the build path is short, `Line` field can end up fitting on the same line as `File`,
but the `{{.*}}` would consume it. Keeping in mind rL293149, i think we can fix it,
while keeping it working when there are and there are not any quotations.
At least this fixes this test for me.
Reviewers: anemet, aaron.ballman, hfinkel
Reviewed By: anemet
Subscribers: cfe-commits, llvm-commits
Differential Revision: https://reviews.llvm.org/D49348
llvm-svn: 337249
Tobias Grosser [Tue, 17 Jul 2018 07:08:06 +0000 (07:08 +0000)]
[DependenceInfo] Use isl++ to replace foreach_set with for loop
llvm-svn: 337248
Tobias Grosser [Tue, 17 Jul 2018 07:08:01 +0000 (07:08 +0000)]
[IslNodeBuilder] Use isl++ to replace foreach_set with for loop
llvm-svn: 337247
Tobias Grosser [Tue, 17 Jul 2018 06:41:20 +0000 (06:41 +0000)]
[ScopInfo] Replace isl foreach calls with for loops
llvm-svn: 337246
Tobias Grosser [Tue, 17 Jul 2018 06:33:41 +0000 (06:33 +0000)]
[ZoneAlgo] Replace isl foreach calls with for loops
llvm-svn: 337245
Tobias Grosser [Tue, 17 Jul 2018 06:33:37 +0000 (06:33 +0000)]
[FlattenSchedule] Replace isl foreach calls with for loops
llvm-svn: 337244
Tobias Grosser [Tue, 17 Jul 2018 06:33:34 +0000 (06:33 +0000)]
[MaximalStaticExpansion] Replace isl foreach calls with for loops
llvm-svn: 337243
Tobias Grosser [Tue, 17 Jul 2018 06:33:31 +0000 (06:33 +0000)]
[ForwardOpTree] Replace isl foreach calls with for loops
llvm-svn: 337242
Tobias Grosser [Tue, 17 Jul 2018 06:33:26 +0000 (06:33 +0000)]
[Simplify] Replace isl foreach calls with for loops
llvm-svn: 337241
Craig Topper [Tue, 17 Jul 2018 06:24:16 +0000 (06:24 +0000)]
[X86] Properly qualify some MOVSS/MOVSD patterns with OptSize.
These are integer versions of patterns that I already fixed for floating point.
llvm-svn: 337240
Tobias Grosser [Tue, 17 Jul 2018 06:16:58 +0000 (06:16 +0000)]
[FlattenAlgo] Replace more isl foreach calls with for loops
This time we replace for loops where the return isl::stat::error has
been used to carry status information.
There are still two uses of foreach remaining as we do not have a
corresponding for implementation for pw_aff functions.
llvm-svn: 337239
Tobias Grosser [Tue, 17 Jul 2018 06:11:53 +0000 (06:11 +0000)]
[FlattenAlgo] Replace some isl foreach calls with for loops
Replace foreach calls which only return 'ok' with for loops.
llvm-svn: 337238
Tobias Grosser [Tue, 17 Jul 2018 06:08:04 +0000 (06:08 +0000)]
[unittest/DeLICM] Replace isl foreach calls with for loops
llvm-svn: 337237
Daniel Cederman [Tue, 17 Jul 2018 05:49:33 +0000 (05:49 +0000)]
[Sparc] Do not depend on icc for ta 1
The ta instruction will always trap, regardless of the value
of the integer condition codes. TRAPri is marked as using icc,
so we cannot use a pattern for TRAPri to implement ta 1, as
verify-machineinstrs can complain that icc is not defined.
Instead we implement ta 1 the same way as ta 5.
llvm-svn: 337236
Eric Fiselier [Tue, 17 Jul 2018 05:48:48 +0000 (05:48 +0000)]
Address "always inline function is not always inlinable" warning with GCC.
When an always_inline function is used prior to the functions definition,
the compiler may not be able to inline it as requested by the attribute.
GCC flags the `basic_string(CharT const*)` function as one such example.
This patch supresses the warning, and the problem, by moving the
definition of the string constructor to the inline declaration.
This ensures the body is available when it is first ODR used.
llvm-svn: 337235
Craig Topper [Tue, 17 Jul 2018 05:48:48 +0000 (05:48 +0000)]
[X86] Add full set of patterns for turning ceil/floor/trunc/rint/nearbyint into rndscale with loads, broadcast, and masking.
This amounts to pretty ridiculous number of patterns. Ideally we'd canonicalize the X86ISD::VRNDSCALE earlier to reuse those patterns. I briefly looked into doing that, but some strict FP operations could still get converted to rint and nearbyint during isel. It's probably still worthwhile to look into. This patch is meant as a starting point to work from.
llvm-svn: 337234
Craig Topper [Tue, 17 Jul 2018 05:48:46 +0000 (05:48 +0000)]
[X86] Add test cases for selecting floor/ceil/trunc/rint/nearbyint to rndscale with masking, loading, and broadcasting.
llvm-svn: 337233
Jacob Bandes-Storch [Tue, 17 Jul 2018 04:56:22 +0000 (04:56 +0000)]
[Driver] Add -fno-digraphs
Summary: Add a flag `-fno-digraphs` to disable digraphs in the lexer, similar to `-fno-operator-names` which disables alternative names for C++ operators.
Reviewers: rsmith
Reviewed By: rsmith
Subscribers: rsmith, cfe-commits
Differential Revision: https://reviews.llvm.org/D48266
llvm-svn: 337232
Artem Dergachev [Tue, 17 Jul 2018 01:39:25 +0000 (01:39 +0000)]
[analyzer] Fix size_t in tests.
Should fix a buildbot. No functional change intended.
llvm-svn: 337231
Chen Zheng [Tue, 17 Jul 2018 01:04:41 +0000 (01:04 +0000)]
[testcases] move testcases to right place - NFC
Differential Revision: https://reviews.llvm.org/D49409
llvm-svn: 337230
Artem Dergachev [Tue, 17 Jul 2018 00:57:57 +0000 (00:57 +0000)]
[CFG] [analyzer] Allow elidable copies to have more than one arguments.
Copy-constructors and move-constructors may have default arguments. It is
incorrect to assert that they only have one argument, i.e. the reference to the
object being copied or moved. Remove the assertion.
Differential Revision: https://reviews.llvm.org/D49215
llvm-svn: 337229
Artem Dergachev [Tue, 17 Jul 2018 00:42:35 +0000 (00:42 +0000)]
[analyzer] pr37802: Fix symbolic-pointer-to-boolean casts during load.
The canonical representation of pointer &SymRegion{$x} casted to boolean is
"$x != 0", not "$x". Assertion added in r337227 catches that.
Differential Revision: https://reviews.llvm.org/D48232
llvm-svn: 337228
Artem Dergachev [Tue, 17 Jul 2018 00:22:27 +0000 (00:22 +0000)]
[analyzer] Assert that nonloc::SymbolVal always wraps a non-Loc-type symbol.
In the current SVal hierarchy there are multiple ways of representing certain
values but few are actually used and expected to be seen by the code.
In particular, a value of a symbolic pointer is always represented by a
loc::MemRegionVal that wraps a SymbolicRegion that wraps the pointer symbol
and never by a nonloc::SymbolVal that wraps that symbol directly.
Assert the aforementioned fact. Fix one minor violation of it.
Differential Revision: https://reviews.llvm.org/D48205
llvm-svn: 337227
Richard Smith [Tue, 17 Jul 2018 00:11:41 +0000 (00:11 +0000)]
Restructure checking for, and warning on, lifetime extension.
This change implements C++ DR1696, which makes initialization of a
reference member of a class from a temporary object ill-formed. The
standard wording here is imprecise, but we interpret it as meaning that
any time a mem-initializer would result in lifetime extension, the
program is ill-formed.
llvm-svn: 337226
Matt Davis [Mon, 16 Jul 2018 23:50:53 +0000 (23:50 +0000)]
[llvm-mca][docs] Add notes about cycle and resource callbacks. NFC.
llvm-svn: 337225
Matt Morehouse [Mon, 16 Jul 2018 23:22:54 +0000 (23:22 +0000)]
[libFuzzer] Avoid STL in MSan test.
Summary:
STL can cause MSan false positives if lib[std]c++ isn't instrumented
with MSan.
Reviewers: kcc
Reviewed By: kcc
Subscribers: Dor1s, llvm-commits
Differential Revision: https://reviews.llvm.org/D49404
llvm-svn: 337224
Craig Topper [Mon, 16 Jul 2018 23:10:58 +0000 (23:10 +0000)]
[X86] Add a missing FMA3 scalar intrinsic pattern.
This allows us to use 231 form to fold an insertelement on the add input to the fma. There is technically no software intrinsic that can use this until AVX512F, but it can be manually built up from other intrinsics.
llvm-svn: 337223
Sam Clegg [Mon, 16 Jul 2018 23:09:29 +0000 (23:09 +0000)]
[WebAssembly] Remove ELF file support.
This support was partial and temporary. Now that we have
wasm object file support its no longer needed.
Differential Revision: https://reviews.llvm.org/D48744
llvm-svn: 337222
Sanjay Patel [Mon, 16 Jul 2018 22:59:31 +0000 (22:59 +0000)]
[Intrinsics] define funnel shift IR intrinsics + DAG builder support
As discussed here:
http://lists.llvm.org/pipermail/llvm-dev/2018-May/123292.html
http://lists.llvm.org/pipermail/llvm-dev/2018-July/124400.html
We want to add rotate intrinsics because the IR expansion of that pattern is 4+ instructions,
and we can lose pieces of the pattern before it gets to the backend. Generalizing the operation
by allowing 2 different input values (plus the 3rd shift/rotate amount) gives us a "funnel shift"
operation which may also be a single hardware instruction.
Initially, I thought we needed to define new DAG nodes for these ops, and I spent time working
on that (much larger patch), but then I concluded that we don't need it. At least as a first
step, we have all of the backend support necessary to match these ops...because it was required.
And shepherding these through the IR optimizer is the primary concern, so the IR intrinsics are
likely all that we'll ever need.
There was also a question about converting the intrinsics to the existing ROTL/ROTR DAG nodes
(along with improving the oversized shift documentation). Again, I don't think that's strictly
necessary (as the test results here prove). That can be an efficiency improvement as a small
follow-up patch.
So all we're left with is documentation, definition of the IR intrinsics, and DAG builder support.
Differential Revision: https://reviews.llvm.org/D49242
llvm-svn: 337221
Puyan Lotfi [Mon, 16 Jul 2018 22:17:05 +0000 (22:17 +0000)]
[NFC][llvm-objcopy] Make helper functions static
Anywhere in tools/llvm-objcopy where functions or classes are not referenced
outside of a given file, we change things to make the function or class static
or put inside an anonymous namespace.
llvm-svn: 337220
Matt Davis [Mon, 16 Jul 2018 21:42:58 +0000 (21:42 +0000)]
[llvm-mca][docs] Initial description of mca internals. NFC
This patch introduces a brief description of the components of MCA. The main
focus is on Views. This is a work in progress, and more descriptions will be
introduced later. I want to flesh-out the Views section more and provide a
detailed description of eventing in MCA. Eventually a brief code example of a
View should accompany the description.
Also, we should consider moving the MCA internals guide elsewhere at some point.
llvm-svn: 337219
Zachary Turner [Mon, 16 Jul 2018 21:34:25 +0000 (21:34 +0000)]
Add missing includes.
llvm-svn: 337218
Zachary Turner [Mon, 16 Jul 2018 21:24:03 +0000 (21:24 +0000)]
[LLVMDemangle] Move some utility classes to header files.
In a followup I'm looking to add a Microsoft demangler. Doing
so needs a lot of the same utility classes and feature test
macros which are already implemented in ItaniumDemangle.cpp.
So move all of these things into header files so that they
can be re-used by a new demangler.
Differential Revision: https://reviews.llvm.org/D49399
llvm-svn: 337217
Reka Kovacs [Mon, 16 Jul 2018 20:47:45 +0000 (20:47 +0000)]
[analyzer] Make checkEndFunction() give access to the return statement.
Differential Revision: https://reviews.llvm.org/D49387
llvm-svn: 337215
George Karpenkov [Mon, 16 Jul 2018 20:42:37 +0000 (20:42 +0000)]
[ASTMatchers] Quickfix for tests.
llvm-svn: 337214
George Karpenkov [Mon, 16 Jul 2018 20:33:25 +0000 (20:33 +0000)]
[analyzer] Bugfix for an overly eager suppression for null pointer return from macros.
Only suppress those cases where the null which came from the macro is
relevant to the bug, and was not overwritten in between.
rdar://
41497323
Differential Revision: https://reviews.llvm.org/D48856
llvm-svn: 337213
George Karpenkov [Mon, 16 Jul 2018 20:32:57 +0000 (20:32 +0000)]
[analyzer] Fix GCDAntipatternChecker to only fire when the semaphore is initialized to zero
Initializing a semaphore with a different constant most likely signals a different intent
rdar://
41802552
Differential Revision: https://reviews.llvm.org/D48911
llvm-svn: 337212
George Karpenkov [Mon, 16 Jul 2018 20:32:32 +0000 (20:32 +0000)]
[analyzer] Provide a symmetric method for generating a PathDiagnosticLocation from Decl
Differential Revision: https://reviews.llvm.org/D49166
llvm-svn: 337211
Yunlian Jiang [Mon, 16 Jul 2018 20:28:48 +0000 (20:28 +0000)]
remove temporal file directory before testing in thinlto-debug-fission.ll
Summary:
This removes the %T/dwo directory before calling ld.lld in ELF/lto/thinlto-debug-fission.ll so that
files aren't left over from previous runs.
Reviewers: espindola, pcc
Reviewed By: pcc
Subscribers: inglorion, emaste, arichardson, eraman, steven_wu, dexonsmith, llvm-commits
Differential Revision: https://reviews.llvm.org/D49390
llvm-svn: 337210
George Karpenkov [Mon, 16 Jul 2018 20:22:12 +0000 (20:22 +0000)]
[ASTMatchers] Introduce Objective-C matchers `hasReceiver` and `isInstanceMessage` for ObjCMessageExpr
Differential Revision: https://reviews.llvm.org/D49333
llvm-svn: 337209
Roman Lebedev [Mon, 16 Jul 2018 20:10:46 +0000 (20:10 +0000)]
[NFC][InstCombine] Fine-tune 'check for [no] signed truncation' tests
We are using i8 for these tests, and shifting by 4,
which is exactly the half of i8.
But as it is seen from the proofs https://rise4fun.com/Alive/mgu
KeptBits = bitwidth(%x) - MaskedBits,
so with using shifts by 4, we are not really testing that
we actually properly handle the other cases with shifts not by half...
llvm-svn: 337208
Alexey Bataev [Mon, 16 Jul 2018 20:05:25 +0000 (20:05 +0000)]
[OPENMP] Fix checks for declare target link entries.
If the declare target link entries are created but not used, the
compiler will produce an error message. Patch improves handling of such
situations + improves checks for possibly lost declare target variables.
llvm-svn: 337207
Max Moroz [Mon, 16 Jul 2018 20:05:18 +0000 (20:05 +0000)]
Revert r337194 (https://reviews.llvm.org/D48891) due to compilation errors.
llvm-svn: 337206
Eric Fiselier [Mon, 16 Jul 2018 20:01:59 +0000 (20:01 +0000)]
Fix PR38160 - init_priority attribute not supported by GCC on Apple.
This patch guards the use of __attribute__((init_priority(101)))
within memory_resource.cpp when building with compilers that don't
support it. Specifically GCC on Apple platforms, and MSVC.
llvm-svn: 337205
Jake Ehrlich [Mon, 16 Jul 2018 19:48:52 +0000 (19:48 +0000)]
[llvm-objcopy] Add support for large indexes
This patch is an update of an older patch that never landed
(see here: https://reviews.llvm.org/D42516)
Recently various users have run into this issue and it just 100%
has to be solved at this point. The main difference in this patch
is that I use gunzip instead of unzip which should hopefully allow
tests to pass. Please review this as if it is a new patch however.
I found some issues along the way and made some minor modifications.
The binary used in this patch for testing (a zip file to make it small)
can be found here:
https://drive.google.com/file/d/1UjsnTO9edLttZibbr-2T1bJl92KEQFAO/view?usp=sharing
Differential Revision: https://reviews.llvm.org/D49206
llvm-svn: 337204
Alex Lorenz [Mon, 16 Jul 2018 19:41:49 +0000 (19:41 +0000)]
[cmake][libFuzzer] fixup r337193 to ensure msan/dfsan are not added
to test deps for libfuzzer when they're not supported by the platform
llvm-svn: 337203
Alex Langford [Mon, 16 Jul 2018 19:19:56 +0000 (19:19 +0000)]
[CMake] Give lldb tools functional install targets when building LLDB.framework
Summary:
This change makes the install targets for lldb tools functional when
building for the framework.
I am currently working on the install rules for lldb-framework and this will
let me make `install-lldb-framework` rely on `install-lldb-argdumper` for
instance. This is especially important for `install-lldb-framework-stripped`. It
is much better for `install-lldb-framework-stripped` to rely on
`install-lldb-argdumper-stripped` than to copy and strip lldb-argdumper
manually.
Differential Revision: https://reviews.llvm.org/D49038
llvm-svn: 337202
Tobias Grosser [Mon, 16 Jul 2018 19:04:16 +0000 (19:04 +0000)]
[ScopInfo] Replace isl foreach calls with for loops
After Philip added support for range-based for loops to our C++
bindings, we now convert another bunch of foreach calls to range-for loops.
This improves general readability of the code.
llvm-svn: 337201
Fangrui Song [Mon, 16 Jul 2018 18:51:40 +0000 (18:51 +0000)]
[CodeGen] Fix inconsistent declaration parameter name
llvm-svn: 337200
Petr Hosek [Mon, 16 Jul 2018 18:48:20 +0000 (18:48 +0000)]
[CMake] Use cxx-headers as a depedency for C++ headers
We no longer pass CLANG_DEFAULT_CXX_STDLIB to the runtimes build
as it was causing issues so we can no longer use this variable. We
instead use cxx-headers as a dependency whenever this is available
since both XRay and libFuzzer are built as static libraries so this
is sufficient.
Differential Revision: https://reviews.llvm.org/D49346
llvm-svn: 337199
Farhana Aleen [Mon, 16 Jul 2018 18:19:59 +0000 (18:19 +0000)]
[AMDGPU] [AMDGPU] Support a fdot2 pattern.
Summary: Optimize fma((float)S0.x, (float)S1.x fma((float)S0.y, (float)S1.y, z))
-> fdot2((v2f16)S0, (v2f16)S1, (float)z)
Author: FarhanaAleen
Reviewed By: rampitec, b-sumner
Subscribers: AMDGPU
Differential Revision: https://reviews.llvm.org/D49146
llvm-svn: 337198
Alexey Bataev [Mon, 16 Jul 2018 18:12:18 +0000 (18:12 +0000)]
[OPENMP] Fix syntactic errors in error messages.
Fixed spelling of the offloading error messages.
llvm-svn: 337196
Yunlian Jiang [Mon, 16 Jul 2018 17:55:48 +0000 (17:55 +0000)]
Support option -plugin-opt=dwo_dir=
Summary:
This adds support to option -plugin-opt=dwo_dir=${DIR}. This option is used to specify the directory to store the .dwo files when LTO and debug fission is used
at the same time.
Reviewers: ruiu, espindola, pcc
Reviewed By: pcc
Subscribers: eraman, dexonsmith, mehdi_amini, emaste, arichardson, steven_wu, llvm-commits
Differential Revision: https://reviews.llvm.org/D47904
llvm-svn: 337195
Max Moroz [Mon, 16 Jul 2018 17:50:46 +0000 (17:50 +0000)]
[libFuzzer] Mutation tracking and logging implemented.
Summary:
Code now exists to track number of mutations that are used in fuzzing in total
and ones that produce new coverage. The stats are currently being dumped to the
command line.
Patch by Kodé Williams (@kodewilliams).
Reviewers: metzman, Dor1s, morehouse, kcc
Reviewed By: Dor1s, morehouse, kcc
Subscribers: delcypher, kubamracek, kcc, morehouse, llvm-commits, #sanitizers, mgorny
Differential Revision: https://reviews.llvm.org/D48054
llvm-svn: 337194
Matt Morehouse [Mon, 16 Jul 2018 17:37:15 +0000 (17:37 +0000)]
[libFuzzer] Add msan and dfsan to test deps.
Required now that we have tests using MSan and DFSan.
llvm-svn: 337193
Mandeep Singh Grang [Mon, 16 Jul 2018 17:26:37 +0000 (17:26 +0000)]
[llvm] Change 2 instances of std::sort to llvm::sort
llvm-svn: 337192
Alexey Bataev [Mon, 16 Jul 2018 16:49:20 +0000 (16:49 +0000)]
[OPENMP, NVPTX] Globalize only captured variables.
Sometimes we can try to globalize non-variable declarations, which may
lead to compiler crash.
llvm-svn: 337191
Roman Lebedev [Mon, 16 Jul 2018 16:45:42 +0000 (16:45 +0000)]
[InstCombine] Fold 'check for [no] signed truncation' pattern
Summary:
[[ https://bugs.llvm.org/show_bug.cgi?id=38149 | PR38149 ]]
As discussed in https://reviews.llvm.org/D49179#1158957 and later,
the IR for 'check for [no] signed truncation' pattern can be improved:
https://rise4fun.com/Alive/gBf
^ that pattern will be produced by Implicit Integer Truncation sanitizer,
https://reviews.llvm.org/D48958 https://bugs.llvm.org/show_bug.cgi?id=21530
in signed case, therefore it is probably a good idea to improve it.
Proofs for this transform: https://rise4fun.com/Alive/mgu
This transform is surprisingly frustrating.
This does not deal with non-splat shift amounts, or with undef shift amounts.
I've outlined what i think the solution should be:
```
// Potential handling of non-splats: for each element:
// * if both are undef, replace with constant 0.
// Because (1<<0) is OK and is 1, and ((1<<0)>>1) is also OK and is 0.
// * if both are not undef, and are different, bailout.
// * else, only one is undef, then pick the non-undef one.
```
The DAGCombine will reverse this transform, see
https://reviews.llvm.org/D49266
Reviewers: spatel, craig.topper
Reviewed By: spatel
Subscribers: JDevlieghere, rkruppe, llvm-commits
Differential Revision: https://reviews.llvm.org/D49320
llvm-svn: 337190
Raphael Isemann [Mon, 16 Jul 2018 16:38:30 +0000 (16:38 +0000)]
Fix some crashes and deadlocks in FormatAnsiTerminalCodes
Summary:
This patch fixes a few problems with the FormatAnsiTerminalCodes function:
* It does an infinite loop on an unknown color value.
* It crashes when the color value is at the end of the string.
* It deletes the first character behind the color token.
Also added a few tests that reproduce those problems (and test some other corner cases).
Reviewers: davide, labath
Reviewed By: labath
Subscribers: labath, lldb-commits, mgorny
Differential Revision: https://reviews.llvm.org/D49307
llvm-svn: 337189
Pavel Labath [Mon, 16 Jul 2018 16:18:52 +0000 (16:18 +0000)]
Fix typo in find-basic-function test
Wrong FileCheck header meant that we were not matching what we should.
This allows us to get rid of the -allow-deprecated-dag-overlap flag in
the test.
llvm-svn: 337188
Max Moroz [Mon, 16 Jul 2018 16:01:31 +0000 (16:01 +0000)]
[libFuzzer] Implement stat::stability_rate based on the percentage of unstable edges.
Summary:
Created a -print_unstable_stats flag.
When -print_unstable_stats=1, we run it 2 more times on interesting inputs poisoning unstable edges in an array.
On program termination, we run PrintUnstableStats() which will print a line with a stability percentage like AFL does.
Patch by Kyungtak Woo (@kevinwkt).
Reviewers: metzman, Dor1s, kcc, morehouse
Reviewed By: metzman, Dor1s, morehouse
Subscribers: delcypher, llvm-commits, #sanitizers, kcc, morehouse, Dor1s
Differential Revision: https://reviews.llvm.org/D49212
llvm-svn: 337187
Wei Mi [Mon, 16 Jul 2018 15:42:20 +0000 (15:42 +0000)]
[RegAlloc] Skip global splitting if the live range is huge and its spill is
trivially rematerializable.
We run into a case where machineLICM hoists a large number of live ranges
outside of a big loop because it thinks those live ranges are trivially
rematerializable. In regalloc, global splitting is tried out first for those
live ranges before they are spilled and rematerialized. Because the global
splitting algorithm is quadratic, increasing a lot of global splitting
candidates causes huge compile time increase (50s to 1400s on my local
machine when compiling a module).
However, we think for live ranges which are very large and are trivially
rematerialiable, it is better to just skip global splitting so as to save
compile time with little chance of sacrificing performance. We uses the
segment size of live range to indirectly evaluate whether the global
splitting of the live range can introduce high cost, and use an option
as a knob to adjust the size limit threshold.
Differential Revision: https://reviews.llvm.org/D49353
llvm-svn: 337186
Teresa Johnson [Mon, 16 Jul 2018 15:30:36 +0000 (15:30 +0000)]
Restore "[ThinLTO] Ensure we always select the same function copy to import"
This reverts commit r337082, restoring r337051, since the LLVM side
patch has been restored.
llvm-svn: 337185
Teresa Johnson [Mon, 16 Jul 2018 15:30:27 +0000 (15:30 +0000)]
Restore "[ThinLTO] Ensure we always select the same function copy to import"
This reverts commit r337081, therefore restoring r337050 (and fix in
r337059), with test fix for bot failure described after the original
description below.
In order to always import the same copy of a linkonce function,
even when encountering it with different thresholds (a higher one then a
lower one), keep track of the summary we decided to import.
This ensures that the backend only gets a single definition to import
for each GUID, so that it doesn't need to choose one.
Move the largest threshold the GUID was considered for import into the
current module out of the ImportMap (which is part of a larger map
maintained across the whole index), and into a new map just maintained
for the current module we are computing imports for. This saves some
memory since we no longer have the thresholds maintained across the
whole index (and throughout the in-process backends when doing a normal
non-distributed ThinLTO build), at the cost of some additional
information being maintained for each invocation of ComputeImportForModule
(the selected summary pointer for each import).
There is an additional map lookup for each callee being considered for
importing, however, this was able to subsume a map lookup in the
Worklist iteration that invokes computeImportForFunction. We also are
able to avoid calling selectCallee if we already failed to import at the
same or higher threshold.
I compared the run time and peak memory for the SPEC2006 471.omnetpp
benchmark (running in-process ThinLTO backends), as well as for a large
internal benchmark with a distributed ThinLTO build (so just looking at
the thin link time/memory). Across a number of runs with and without
this change there was no significant change in the time and memory.
(I tried a few other variations of the change but they also didn't
improve time or peak memory).
The new commit removes a test that no longer makes sense
(Transforms/FunctionImport/hotness_based_import2.ll), as exposed by the
reverse-iteration bot. The test depends on the order of processing the
summary call edges, and actually depended on the old problematic
behavior of selecting more than one summary for a given GUID when
encountered with different thresholds. There was no guarantee even
before that we would eventually pick the linkonce copy with the hottest
call edges, it just happened to work with the test and the old code, and
there was no guarantee that we would end up importing the selected
version of the copy that had the hottest call edges (since the backend
would effectively import only one of the selected copies).
Reviewers: davidxl
Subscribers: mehdi_amini, inglorion, llvm-commits
Differential Revision: https://reviews.llvm.org/D48670
llvm-svn: 337184
George Rimar [Mon, 16 Jul 2018 15:29:35 +0000 (15:29 +0000)]
[ELF] - Eliminate ObjFile<ELFT>::getLineInfo. NFC.
Flow is the same, but a bit shorter after this change.
llvm-svn: 337183
George Rimar [Mon, 16 Jul 2018 15:27:53 +0000 (15:27 +0000)]
[ELF] - Cover ObjFile<ELFT>::getLineInfo() with a test.
There was no test for the first `return`.
llvm-svn: 337182
Joel Galenson [Mon, 16 Jul 2018 15:26:44 +0000 (15:26 +0000)]
[cfi-verify] Abort on unsupported targets
As suggested in the review for r337007, this makes cfi-verify abort on unsupported targets instead of producing incorrect results. It also updates the design document to reflect this.
Differential Revision: https://reviews.llvm.org/D49304
llvm-svn: 337181
Max Moroz [Mon, 16 Jul 2018 15:15:34 +0000 (15:15 +0000)]
Revert r337175 (https://reviews.llvm.org/D49212) due to unintentional format changes.
llvm-svn: 337180
Chen Zheng [Mon, 16 Jul 2018 15:06:42 +0000 (15:06 +0000)]
[InstrSimplify] add testcases for fold sdiv if two operands are negatived and non-overflow
Differential Revision: https://reviews.llvm.org/D49365
llvm-svn: 337179
Chandler Carruth [Mon, 16 Jul 2018 15:01:26 +0000 (15:01 +0000)]
Fix warnings in a static assert added to CFI.
llvm-svn: 337178
Chandler Carruth [Mon, 16 Jul 2018 14:58:32 +0000 (14:58 +0000)]
[x86/SLH] Completely rework how we sink post-load hardening past data
invariant instructions to be both more correct and much more powerful.
While testing, I continued to find issues with sinking post-load
hardening. Unfortunately, it was amazingly hard to create any useful
tests of this because we were mostly sinking across copies and other
loading instructions. The fact that we couldn't sink past normal
arithmetic was really a big oversight.
So first, I've ported roughly the same set of instructions from the data
invariant loads to also have their non-loading varieties understood to
be data invariant. I've also added a few instructions that came up so
often it again made testing complicated: inc, dec, and lea.
With this, I was able to shake out a few nasty bugs in the validity
checking. We need to restrict to hardening single-def instructions with
defined registers that match a particular form: GPRs that don't have
a NOREX constraint directly attached to their register class.
The (tiny!) test case included catches all of the issues I was seeing
(once we can sink the hardening at all) except for the NOREX issue. The
only test I have there is horrible. It is large, inexplicable, and
doesn't even produce an error unless you try to emit encodings. I can
keep looking for a way to test it, but I'm out of ideas really.
Thanks to Ben for giving me at least a sanity-check review. I'll follow
up with Craig to go over this more thoroughly post-commit, but without
it SLH crashes everywhere so landing it for now.
Differential Revision: https://reviews.llvm.org/D49378
llvm-svn: 337177
Henry Zhu [Mon, 16 Jul 2018 14:54:29 +0000 (14:54 +0000)]
Test commit
llvm-svn: 337176
Max Moroz [Mon, 16 Jul 2018 14:54:23 +0000 (14:54 +0000)]
[libFuzzer] Implement stat::stability_rate based on the percentage of unstable edges.
Summary:
Created a -print_unstable_stats flag.
When -print_unstable_stats=1, we run it 2 more times on interesting inputs poisoning unstable edges in an array.
On program termination, we run PrintUnstableStats() which will print a line with a stability percentage like AFL does.
Patch by Kyungtak Woo (@kevinwkt).
Reviewers: metzman, Dor1s, kcc, morehouse
Reviewed By: metzman, Dor1s, morehouse
Subscribers: delcypher, llvm-commits, #sanitizers, kcc, morehouse, Dor1s
Differential Revision: https://reviews.llvm.org/D49212
llvm-svn: 337175
Marco Castelluccio [Mon, 16 Jul 2018 14:40:33 +0000 (14:40 +0000)]
[gcov] Add a test showing differences in line counts when building with or without exceptions enabled.
Test for https://bugs.llvm.org/show_bug.cgi?id=38066.
llvm-svn: 337174
Pavel Labath [Mon, 16 Jul 2018 14:37:58 +0000 (14:37 +0000)]
Fix TestDataFormatterUnordered for older libc++ versions
clang recently started diagnosing "exception specification in
declaration does not match previous declaration" errors. Unfortunately
old libc++ versions had a bug, where they violated this rule, which
means that tests using this library version now fail due to build
errors.
Since it was easy to work around the bug by compiling this test with
-fno-exceptions, I do that here. If supporting old libc++ versions
becomes a burden, we'll have to revisit this.
llvm-svn: 337173
Balazs Keri [Mon, 16 Jul 2018 14:05:18 +0000 (14:05 +0000)]
[ASTImporter] Changed constant int to unsigned int in test code.
llvm-svn: 337172
Simon Atanasyan [Mon, 16 Jul 2018 13:52:41 +0000 (13:52 +0000)]
[mips] Eliminate the usage of hasStdEnc in MipsPat.
Instead, the pattern is tagged with the correct predicate when
it is declared. Some patterns have been duplicated as necessary.
Patch by Simon Dardis.
Differential revision: https://reviews.llvm.org/D48365
llvm-svn: 337171
Filipe Cabecinhas [Mon, 16 Jul 2018 13:41:54 +0000 (13:41 +0000)]
[cfi] Don't pass a uint16_t to memset. Make sure the 16-bit constant is appropriate for us.
Reviewers: eugenis, pcc, kcc
Subscribers: delcypher, #sanitizers, llvm-commits
Differential Revision: https://reviews.llvm.org/D49252
llvm-svn: 337170
Mikhail R. Gadelha [Mon, 16 Jul 2018 13:32:22 +0000 (13:32 +0000)]
[analyzer] Fix the Z3 backend always generating unsigned APSInt
Summary:
In `toAPSInt`, the Z3 backend was not checking the variable `Int`'s type and was always generating unsigned `APSInt`s.
This was found by accident when I removed:
```
llvm::APSInt ConvertedLHS, ConvertedRHS;
QualType LTy, RTy;
std::tie(ConvertedLHS, LTy) = fixAPSInt(*LHS);
std::tie(ConvertedRHS, RTy) = fixAPSInt(*RHS);
- doIntTypePromotion<llvm::APSInt, Z3ConstraintManager::castAPSInt>(
- ConvertedLHS, LTy, ConvertedRHS, RTy);
return BVF.evalAPSInt(BSE->getOpcode(), ConvertedLHS, ConvertedRHS);
```
And the `BasicValueFactory` started to complain about different `signedness`.
Reviewers: george.karpenkov, NoQ, ddcc
Reviewed By: ddcc
Subscribers: xazax.hun, szepet, a.sidorin
Differential Revision: https://reviews.llvm.org/D49305
llvm-svn: 337169
Petar Jovanovic [Mon, 16 Jul 2018 13:29:32 +0000 (13:29 +0000)]
[MIPS GlobalISel] Select instructions to load and store i32 on stack
Add code for selection of G_LOAD, G_STORE, G_GEP, G_FRAMEINDEX and
G_CONSTANT. Support loads and stores of i32 values.
Patch by Petar Avramovic.
Differential Revision: https://reviews.llvm.org/D48957
llvm-svn: 337168
Mikhail R. Gadelha [Mon, 16 Jul 2018 13:14:46 +0000 (13:14 +0000)]
[analyzer] Fix constraint being dropped when analyzing a program without taint tracking enabled
Summary:
This patch removes the constraint dropping when taint tracking is disabled.
It also voids the crash reported in D28953 by treating a SymSymExpr with non pointer symbols as an opaque expression.
Updated the regressions and verifying the big projects now; I'll update here when they're done.
Based on the discussion on the mailing list and the patches by @ddcc.
Reviewers: george.karpenkov, NoQ, ddcc, baloghadamsoftware
Reviewed By: george.karpenkov
Subscribers: delcypher, llvm-commits, rnkovacs, xazax.hun, szepet, a.sidorin, ddcc
Differential Revision: https://reviews.llvm.org/D48650
llvm-svn: 337167
Roman Lebedev [Mon, 16 Jul 2018 12:44:10 +0000 (12:44 +0000)]
[X86][AArch64][DAGCombine] Unfold 'check for [no] signed truncation' pattern
Summary:
[[ https://bugs.llvm.org/show_bug.cgi?id=38149 | PR38149 ]]
As discussed in https://reviews.llvm.org/D49179#1158957 and later,
the IR for 'check for [no] signed truncation' pattern can be improved:
https://rise4fun.com/Alive/gBf
^ that pattern will be produced by Implicit Integer Truncation sanitizer,
https://reviews.llvm.org/D48958 https://bugs.llvm.org/show_bug.cgi?id=21530
in signed case, therefore it is probably a good idea to improve it.
But the IR-optimal patter does not lower efficiently, so we want to undo it..
This handles the simple pattern.
There is a second pattern with predicate and constants inverted.
NOTE: we do not check uses here. we always do the transform.
Reviewers: spatel, craig.topper, RKSimon, javed.absar
Reviewed By: spatel
Subscribers: kristof.beyls, llvm-commits
Differential Revision: https://reviews.llvm.org/D49266
llvm-svn: 337166
Daniel Cederman [Mon, 16 Jul 2018 12:28:26 +0000 (12:28 +0000)]
[Sparc] Use the correct encoding for ta 3
Summary: The old encoding generated a "tn %g1 + 3" instruction instead
of the expected "ta 3".
Reviewers: venkatra, jyknight
Reviewed By: jyknight
Subscribers: fedor.sergeev, jrtc27, llvm-commits
Differential Revision: https://reviews.llvm.org/D49171
llvm-svn: 337165
Daniel Cederman [Mon, 16 Jul 2018 12:22:08 +0000 (12:22 +0000)]
[Sparc] Use the names .rem and .urem instead of __modsi3 and __umodsi3
Summary: These are the names used in libgcc.
Reviewers: venkatra, jyknight, ekedaigle
Reviewed By: jyknight
Subscribers: joerg, fedor.sergeev, jrtc27, llvm-commits
Differential Revision: https://reviews.llvm.org/D48915
llvm-svn: 337164
Daniel Cederman [Mon, 16 Jul 2018 12:16:53 +0000 (12:16 +0000)]
[Sparc] Generate ta 1 for the @llvm.debugtrap intrinsic
Summary: Software trap number one is the trap used for breakpoints
in the Sparc ABI.
Reviewers: jyknight, venkatra
Reviewed By: jyknight
Subscribers: fedor.sergeev, jrtc27, llvm-commits
Differential Revision: https://reviews.llvm.org/D48637
llvm-svn: 337163
Balazs Keri [Mon, 16 Jul 2018 12:16:39 +0000 (12:16 +0000)]
[ASTImporter] Import implicit methods of existing class.
Summary:
When an already existing class is encountered during import,
check if it has implicit methods that are missing in the existing one,
and import these.
The to-be-imported code may use the same class in different way than the
existing (before the import) code. This may result in that there are
implicit methods that are not generated for the existing code.
Reviewers: a.sidorin, a_sidorin
Reviewed By: a_sidorin
Subscribers: a_sidorin, martong, cfe-commits
Differential Revision: https://reviews.llvm.org/D49245
llvm-svn: 337162
Daniel Cederman [Mon, 16 Jul 2018 12:14:17 +0000 (12:14 +0000)]
Avoid losing Hi part when expanding VAARG nodes on big endian machines
Summary:
If the high part of the load is not used the offset to the next element
will not be set correctly.
For example, on Sparc V8, the following code will read val2 from offset 4
instead of 8.
```
int val = __builtin_va_arg(va, long long);
int val2 = __builtin_va_arg(va, int);
```
Reviewers: jyknight
Reviewed By: jyknight
Subscribers: fedor.sergeev, jrtc27, llvm-commits
Differential Revision: https://reviews.llvm.org/D48595
llvm-svn: 337161
Chandler Carruth [Mon, 16 Jul 2018 11:38:48 +0000 (11:38 +0000)]
[x86/SLH] Fix a bug where we would try to post-load harden non-GPRs.
Found cases that hit the assert I added. This patch factors the validity
checking into a nice helper routine and calls it when deciding to harden
post-load, and asserts it when doing so later.
I've added tests for the various ways of loading a floating point type,
as well as loading all vector permutations. Even though many of these go
to identical instructions, it seems good to somewhat comprehensively
test them.
I'm confident there will be more fixes needed here, I'll try to add
tests each time as I get this predicate adjusted.
llvm-svn: 337160
Alexander Potapenko [Mon, 16 Jul 2018 10:57:19 +0000 (10:57 +0000)]
MSan: minor fixes, NFC
- remove an extra space after |ID| declaration
- drop the unused |FirstInsn| parameter in getShadowOriginPtrUserspace()
llvm-svn: 337159
Jonas Devlieghere [Mon, 16 Jul 2018 10:52:27 +0000 (10:52 +0000)]
[AccelTable] Provide DWARF5AccelTableStaticData for dsymutil.
For dsymutil we want to store offsets in the accelerator table entries
rather than DIE pointers. In addition, we need a way to communicate
which CU a DIE belongs to. This patch provides support for both of these
issues.
Differential revision: https://reviews.llvm.org/D49102
llvm-svn: 337158
Chandler Carruth [Mon, 16 Jul 2018 10:46:16 +0000 (10:46 +0000)]
[x86/SLH] Extract another small helper function, add better comments and
use better terminology. NFC.
llvm-svn: 337157
Mark Searles [Mon, 16 Jul 2018 10:21:36 +0000 (10:21 +0000)]
[AMDGPU][Waitcnt] Re-apply fix "comparison of integers of different signs" build error"
Re-apply "[AMDGPU][Waitcnt] fix "comparison of integers of different signs" build error""
(
fe0a456510131f268e388c4a18a92f575c0db183 ), which was inadvertantly reverted via
2b2ee080f0164485562593b1b87291a48cea4a9a .
llvm-svn: 337156
Alexander Potapenko [Mon, 16 Jul 2018 10:03:30 +0000 (10:03 +0000)]
[MSan] factor userspace-specific declarations into createUserspaceApi(). NFC
This patch introduces createUserspaceApi() that creates function/global
declarations for symbols used by MSan in the userspace.
This is a step towards the upcoming KMSAN implementation patch.
Reviewed at https://reviews.llvm.org/D49292
llvm-svn: 337155
Mark Searles [Mon, 16 Jul 2018 10:02:41 +0000 (10:02 +0000)]
run post-RA hazard recognizer pass late
Memory legalizer, waitcnt, and shrink passes can perturb the instructions,
which means that the post-RA hazard recognizer pass should run after them.
Otherwise, one of those passes may invalidate the work done by the hazard
recognizer. Note that this has adverse side-effect that any consecutive
S_NOP 0's, emitted by the hazard recognizer, will not be shrunk into a
single S_NOP <N>. This should be addressed in a follow-on patch.
Differential Revision: https://reviews.llvm.org/D49288
llvm-svn: 337154
Mark Searles [Mon, 16 Jul 2018 10:02:40 +0000 (10:02 +0000)]
Revert "[AMDGPU][Waitcnt] fix "comparison of integers of different signs" build error"
This reverts commit
fe0a456510131f268e388c4a18a92f575c0db183.
llvm-svn: 337153
Benjamin Kramer [Mon, 16 Jul 2018 09:52:02 +0000 (09:52 +0000)]
[Sema] Reword warning for constant captures that are not required
This is one of the darker corners of C++, make it clear that this is
about constants and rephrase it a bit.
Before: lambda capture 'i' is not required to be captured for this use
After: lambda capture of constant 'i' is not required for this use
llvm-svn: 337152
Adam Balogh [Mon, 16 Jul 2018 09:27:27 +0000 (09:27 +0000)]
[Analyzer] Mark `SymbolData` parts of iterator position as live in program state maps
Marking a symbolic expression as live is non-recursive. In our checkers we
either use conjured symbols or conjured symbols plus/minus integers to
represent abstract position of iterators, so in this latter case we also
must mark the `SymbolData` part of these symbolic expressions as live to
prevent them from getting reaped.
Differential Revision: https://reviews.llvm.org/D48764
llvm-svn: 337151
Marco Castelluccio [Mon, 16 Jul 2018 09:13:46 +0000 (09:13 +0000)]
Add a test with __gcov_flush called before terminating the program.
Test for https://bugs.llvm.org/show_bug.cgi?id=38067.
llvm-svn: 337150
Alexandros Lamprineas [Mon, 16 Jul 2018 07:51:27 +0000 (07:51 +0000)]
[MemorySSAUpdater] Remove deleted trivial Phis from active workset
Bug fix for PR37808. The regression test is a reduced version of the
original reproducer attached to the bug report. As stated in the report,
the problem was that InsertedPHIs was keeping dangling pointers to
deleted Memory-Phis. MemoryPhis are created eagerly and sometimes get
zapped shortly afterwards. I've used WeakVH instead of an expensive
removal operation from the active workset.
Differential Revision: https://reviews.llvm.org/D48372
llvm-svn: 337149