Hristo Hristov [Sun, 11 Jun 2023 19:45:16 +0000 (22:45 +0300)]
[libc++][spaceship][NFC] P1612R2: Mark some items "remove `operator!=` from `*_iterator`" as "Complete"
Several parts of P1612R2 are already implemented:
- `common_iterator` (new in C++20) https://reviews.llvm.org/D103335
- `unreachable_sentinel_t` (new in C++20) https://reviews.llvm.org/D107920
- `istream_iterator` (removed) https://reviews.llvm.org/D119620
- `istreambuf_iterator` (removed) https://reviews.llvm.org/D119620
Reviewed By: #libc, Mordante
Differential Revision: https://reviews.llvm.org/D152656
Jacques Pienaar [Wed, 7 Jun 2023 22:42:03 +0000 (15:42 -0700)]
[mlir][docgen] Enable custom prefix strip for op name.
The fully qualified name gets long in the TOC (especially on mobile),
enable specifying a prefix to be stripped.
Differential Revision: https://reviews.llvm.org/D152404
Ryan Prichard [Tue, 13 Jun 2023 03:43:00 +0000 (20:43 -0700)]
[libc++] Android temp dir is /data/local/tmp, enable Windows test
[libc++] Android temp dir is /data/local/tmp, enable Windows test
On Android, std::filesystem::temp_directory_path() should fall back to
/data/local/tmp when no environment variable is set. There is no /tmp
directory. Most apps can't access /data/local/tmp, but they do have a
"cache dir" (Context#getCacheDir()) that is usable for temporary files.
However, there is no obvious and reliable way for libc++ to query this
directory in contexts where it is available. The global fallback
/data/local/tmp is available for "adb shell", making it useful for test
suites.
On Windows, temp_directory_path falls back to the Windows directory
(e.g. "C:\Windows"), so call GetWindowsDirectoryW to do the test.
Reviewed By: ldionne, #libc, enh
Differential Revision: https://reviews.llvm.org/D137131
Kun Wu [Tue, 13 Jun 2023 02:10:16 +0000 (02:10 +0000)]
[mlir][sparse][GPU] add 2:4 integration test
Reviewed By: aartbik
Differential Revision: https://reviews.llvm.org/D152287
Matt Arsenault [Mon, 12 Jun 2023 20:25:51 +0000 (16:25 -0400)]
AMDGPU: Correct constants used in fast math log expansion
The division between float constants was done with less
precision. Performing the divide in double and truncating to float
provides the same value as used in the library fast math expansion.
Matt Arsenault [Sat, 10 Jun 2023 20:39:32 +0000 (16:39 -0400)]
AMDGPU: Modernize log codegen tests
Matt Arsenault [Sun, 11 Jun 2023 00:03:16 +0000 (20:03 -0400)]
AMDGPU: Add basic folds for llvm.amdgcn.log
Matt Arsenault [Sun, 11 Jun 2023 16:35:28 +0000 (12:35 -0400)]
AMDGPU: Add baseline instcombine test for llvm.amdgcn.log
Matt Arsenault [Sun, 11 Jun 2023 16:25:30 +0000 (12:25 -0400)]
AMDGPU: Add llvm.amdgcn.log intrinsic
This will map directly to the hardware instruction which does not
handle denormals for f32. This will allow moving the generic intrinsic
to be lowered correctly. Also handles selecting the f16 version, but
there's no reason to use it over the generic intrinsic.
Matt Arsenault [Sat, 10 Jun 2023 16:20:43 +0000 (12:20 -0400)]
LowerMemIntrinsics: Handle inserting addrspacecast for memmove lowering
We're missing a trivial non-AA way to check for non-aliasing address
spaces.
Fangrui Song [Tue, 13 Jun 2023 00:58:40 +0000 (17:58 -0700)]
[Driver] Default -fsanitize-address-globals-dead-stripping to true for ELF
-fsanitize-address-globals-dead-stripping is the default for non-ELF
platforms. For ELF, we disabled it to work around an ancient gold 2.26
bug. However, some platforms (Fuchsia and PS) default the option to
true.
This patch changes -fsanitize-address-globals-dead-stripping to true for all ELF
platforms. Without specifying -fdata-sections (non-default for most ELF
platforms), `asan_globals` can only be GCed if the monolithic .data/.bss section
is GCed, which makes it less effective.
However, I think this simplified rule is better than making the
-fsanitize-address-globals-dead-stripping default dependent on another option.
Related: D120394
Close https://github.com/llvm/llvm-project/issues/63127
Reviewed By: #sanitizers, eugenis, phosek
Differential Revision: https://reviews.llvm.org/D152604
Christopher Ferris [Mon, 12 Jun 2023 23:21:24 +0000 (16:21 -0700)]
[scudo] Check support for pvalloc/valloc tests.
Some platforms do not support pvalloc/valloc so add checks so
that these tests are only run on the appropriate platforms.
Reviewed By: Chia-hungDuan
Differential Revision: https://reviews.llvm.org/D152756
Krzysztof Parzyszek [Fri, 2 Jun 2023 16:07:49 +0000 (09:07 -0700)]
[RDF] Define short type names: NodeAddr<XyzNode*> -> Xyz, NFC
Remove all instances of 'using namespace rdf' from files in CodeGen,
all implementation code is explicitly enclosed in 'namespace rdf'.
Jin Xin Ng [Mon, 12 Jun 2023 22:33:34 +0000 (22:33 +0000)]
[tsan] Check for nullptr on user_alloc_usable_size_fast
MBlock could be null (in the context of a call from RunFreeHooks)
if a static object was runtime initialized before tsan finished
initializing and that object later did a free().
While having the check isn't strictly required by
__sanitizer_get_allocated_size_fast's contract, a user's static object
would expect ptrs returned from malloc to be valid inputs.
Differential Revision: https://reviews.llvm.org/D152755
Maksim Panchenko [Mon, 12 Jun 2023 20:16:09 +0000 (13:16 -0700)]
[BOLT] Fix --max-funcs=<N> option
Fix off-by-one error while handling of the --max-funcs=<N> option.
We used to process N+1 functions when N was requested.
Reviewed By: Amir
Differential Revision: https://reviews.llvm.org/D152751
Aart Bik [Mon, 12 Jun 2023 18:43:27 +0000 (11:43 -0700)]
[mlir][sparse][gpu] guard matvec COO AoS
Reviewed By: K-Wu
Differential Revision: https://reviews.llvm.org/D152738
Kun Wu [Mon, 12 Jun 2023 23:24:08 +0000 (23:24 +0000)]
[mlir][sparse][gpu]fix various cusparseLt bugs
Reviewed By: aartbik
Differential Revision: https://reviews.llvm.org/D152489
Nikolas Klauser [Fri, 9 Jun 2023 19:34:07 +0000 (12:34 -0700)]
[libc++][PSTL] Update papers and add a status page for the algorithms
Reviewed By: #libc, ldionne
Spies: libcxx-commits, mgrang, arphaman
Differential Revision: https://reviews.llvm.org/D152578
Alex Langford [Fri, 9 Jun 2023 20:47:32 +0000 (13:47 -0700)]
[lldb] Change return type of UnixSignals::GetShortName
The short names of each signal name and alias only exist as ConstStrings
in this one scenario. For example, GetShortName("SIGHUP") will just give
you "HUP". There's not a good reason the string "HUP" needs to be in the
ConstString StringPool, and that's true for just about every signal
name.
Differential Revision: https://reviews.llvm.org/D152582
Fangrui Song [Mon, 12 Jun 2023 22:50:29 +0000 (15:50 -0700)]
[Driver] Remove misused NoXarchOption from some Windows options
If an option has the `NoXarchOption` flag, ClangDriver will emit an error if the
option is used after `-Xarch_*` (originally for universal macOS binary, reused
by `-Xarch_host`/etc). The error checking only applies to a small set of options
(e.g. `-o`) and is not very useful for most options, but `NoXarchOption` was
traditionally named `DriverOption` (commit
aabb0b11a3c1d8a6bb859db80400cffdcc9b336f) and lured some contributors to add
`NoXarchOption` to options that should not have the flag.
Daniel Thornburgh [Mon, 12 Jun 2023 22:43:27 +0000 (15:43 -0700)]
[Fuchsia] Pass CursesAndPanel_ROOT through to stage2
Eric Gouriou [Mon, 12 Jun 2023 22:31:18 +0000 (15:31 -0700)]
[RISC-V] Zvk update to 0.9.7, Zvknc/Zvksc
Update the RISC-V Zvk (vector cryptography) extension support from 0.5
to version 0.9.7 (2023-05-31), per
<https://github.com/riscv/riscv-crypto/releases/download/v20230531/riscv-crypto-spec-vector.pdf>
Differences:
- Zvbc is dropped from Zvkn and Zvks, and by extension
from Zvkng and Zvksg;
- new combo extensions Zvknc and Zvksc are introduced,
adding Zvbc to Zvkn and Zvks;
- the experimentatl extensions are tagged as "0.9",
from the earlier "0.5".
Reviewed By: 4vtomat
Differential Revision: https://reviews.llvm.org/D152117
Vitaly Buka [Mon, 12 Jun 2023 22:35:24 +0000 (15:35 -0700)]
[NFC][sanitizer] Move ArrayRef into own header
Vitaly Buka [Mon, 12 Jun 2023 22:32:47 +0000 (15:32 -0700)]
[NFC][sanitizer] Rename sanitizer_common_range -> sanitizer_range
Craig Topper [Mon, 12 Jun 2023 22:29:03 +0000 (15:29 -0700)]
[RISCV] Prevent overflowing the small size of InstSeq in generateInstSeq.
The small size is 8 which is the worst case of the core recursive
algorithm.
The special cases use the core algorithm and append additonal
instructions. We were pushing the extra instructions before checking
the profitability. This could lead to 9 and maybe 10 instructions
in the sequence which overflows the small size.
This patch does the profitability check before inserting the
extra instructions so that we don't create 9 or 10 insruction
sequences.
Alternative we could bump the small size to 9 or 10, but then
we're pushing things that are never going be used.
Reviewed By: asb
Differential Revision: https://reviews.llvm.org/D152308
Vitaly Buka [Mon, 12 Jun 2023 22:23:30 +0000 (15:23 -0700)]
[NFC][sanitizer] Move ArrayRef into own header
Stanislav Mekhanoshin [Sat, 10 Jun 2023 08:10:16 +0000 (01:10 -0700)]
[AMDGPU] Remove integer division in VOPD checks
There is no way any compiler can simplify this division, while
the check is done rather often.
Differential Revision: https://reviews.llvm.org/D152613
Alex Langford [Wed, 24 May 2023 21:08:33 +0000 (14:08 -0700)]
[lldb] Remove lldb's DWARFAbbreviationDeclarationSet in favor of llvm's
lldb's and llvm's implementations of DWARFAbbreviationDeclarationSet are
now close enough (almost the same, actually) to replace lldb's with
llvm's wholesale. llvm's is also tested against the same kinds of
scenarios that lldb's is tested against so we can remove lldb's tests
here. (see: llvm/unittests/DebugInfo/DWARF/DWARFDebugAbbrevTest.cpp).
Differential Revision: https://reviews.llvm.org/D152476
Adrian Prantl [Mon, 12 Jun 2023 21:51:10 +0000 (14:51 -0700)]
Speculative fix for windows test
Aart Bik [Mon, 12 Jun 2023 18:50:48 +0000 (11:50 -0700)]
[mlir][sparse][gpu] fixed whitespace in tbgen
Reviewed By: K-Wu
Differential Revision: https://reviews.llvm.org/D152739
Augie Fackler [Wed, 5 Apr 2023 22:03:14 +0000 (18:03 -0400)]
CodeGen: hand two tests to the care of update_cc_test_checks
These two tests were pretty challenging to manage by hand, and in
rebasing the next change in my stack I finally ran out of patience and
decided to just embrace automation. This change _only_ changes the tests
so that it's obvious what's new in the next change.
Differential Revision: https://reviews.llvm.org/D152400
Kirill Stoimenov [Sat, 10 Jun 2023 00:07:53 +0000 (00:07 +0000)]
[Sanitizer] Add munmap interceptor in sanitizer_common
Reviewed By: vitalybuka
Differential Revision: https://reviews.llvm.org/D152598
Ivan Kelarev [Mon, 12 Jun 2023 21:21:43 +0000 (14:21 -0700)]
[NFC][SLP] Fix a few minor formatting issues
Differential Revision: https://reviews.llvm.org/D152395
Fangrui Song [Mon, 12 Jun 2023 21:15:03 +0000 (14:15 -0700)]
[sanitizer][test] Remove no-op REQUIRES:
Some patches around D109843 added `REQUIRES: freebsd` but they have no effects
due to sanitizer_common/TestCases/Linux/lit.local.cfg.py . I informed the author
but don't plan to move the tests.
Krzysztof Parzyszek [Thu, 1 Jun 2023 19:00:47 +0000 (12:00 -0700)]
[RDF] Allow RegisterRef to contain register unit
Recommit with the fix for incorrect `constexpr` with `hash` from a later
commit, and a fix for "std::swap(std::less<RegisterRef>...)" issue with
MSVC's `std::map`.
James Knight [Mon, 12 Jun 2023 20:05:35 +0000 (16:05 -0400)]
[bazel] Repair clang_headers_gen when run on macOS.
The antique version of bash (3.2.57, from 2007) which is available on
macOS cannot deal with quoted slashes in a `${x/...}`
substitution. Since only prefix-removal is required here, switch to a
`${x#...}` substitution instead.
(E.g. `src="foo/bar/baz.h"; echo ${src/"foo/bar"}` echos `bar/bar/baz.h`
instead of `/baz.h` on old bash versions).
Originally broken by
459420c33a2c80525bc73ad1ff89fdf4dd079d6c.
Fixes #63222
Luke Lau [Mon, 12 Jun 2023 16:12:28 +0000 (16:12 +0000)]
[RISCV] Use TU vmv.v.v instead of vslideup.vi N,M,0 for insert vector_shuffles
This applies the same technique in https://reviews.llvm.org/D152565 to
vslideups when lowering vector_shuffles that have an insert pattern in
their mask.
Reviewed By: reames
Differential Revision: https://reviews.llvm.org/D152724
Joseph Huber [Mon, 12 Jun 2023 19:17:56 +0000 (14:17 -0500)]
[libc] Fix floating point test failing to build on the GPU
A patch enabled this test which uses that `add_fp_unittest`.
Unfortunately we do not support these on the GPU because it attempts to
link in the floating point utils which are not built supporting
hermetic tests. This was attempted to be fixed in D151123 but that had
to be reverted. For now disable these so the tests pass.
Reviewed By: tra
Differential Revision: https://reviews.llvm.org/D152742
Fangrui Song [Mon, 12 Jun 2023 20:04:29 +0000 (13:04 -0700)]
[docs] Improve UndefinedBehaviorSanitizer.rst
* Mention that -fsanitize= and -fno-sanitize= apply to check groups.
* Mention "all" can be used as a check group.
* Mention that -fsanitize-trap= and -fsanitize-recover= lead to no unused command line option warning.
* Mention that trap mode typically causes the program to terminate due to a `SIGILL` or `SIGTRAP` signal.
Reviewed By: #sanitizers, vitalybuka
Differential Revision: https://reviews.llvm.org/D152650
Fangrui Song [Mon, 12 Jun 2023 19:47:33 +0000 (12:47 -0700)]
[Driver] Mark m_x86_Features_Group options as TargetSpecific
so that they get an error on non-x86 targets.
Follow-up to D151590.
As a workaround for https://github.com/llvm/llvm-project/issues/63270, we don't
report an error for -msse4.2.
Sami Tolvanen [Mon, 12 Jun 2023 16:24:54 +0000 (16:24 +0000)]
[KCFI] Fix hash offset calculation in Thumb mode
ARM stores the Thumb state in the least significant bit of the
function pointers. When compiling for ARM or Thumb, as all
instructions are at least 16-bit aligned, ignore the LSB when
computing the prefix hash location, so we can support both
pure Thumb and mixed ARM/Thumb binaries.
Fixes #62936
Reviewed By: MaskRay, simon_tatham
Differential Revision: https://reviews.llvm.org/D152484
Louis Dionne [Mon, 12 Jun 2023 19:43:04 +0000 (12:43 -0700)]
[libc++][NFC] Refactor filesystem namespace test
Fangrui Song [Mon, 12 Jun 2023 19:34:08 +0000 (12:34 -0700)]
[Driver] Mark many target-specific driver-only options as TargetSpecific
so that they get an error on other targets. This change uses let statements to
apply `Flags = [TargetSpecific]` to options (mostly -m*) without specifying `Flags`.
Follow-up to D151590.
For some options, e.g. -mdefault-build-attributes (D31813), -mbranch-likely
(D38168), -mfpu=/-mabi= (
6890b9b71e525020ab58d436336664beede71575), a warning
seems desired in at least certain cases. This is not the best practice, but this
change works around them by not applying `Flags = [TargetSpecific]`.
Some m_x86_Features_Group options are popular. We make them TargetSpecific in a
future patch.
(
For Intel CPU errata -malign-branch= family options, we also drop the unneeded
NotXarchOption flag (was confuslingly named DriverOption). This flag reports an
error if the option is used with -Xarch_*. This error reporting does not seem
correct/useful.
)
Hristo Hristov [Mon, 12 Jun 2023 11:29:12 +0000 (14:29 +0300)]
[libc++][spaceship] P1614R2: Removed `operator!=` from `weekday`s
Implements parts of P1614R2. Removed `operator!=` from:
- `weekday`
- `weekday_indexed`
- `weekday_last`
- `month_weekday`
- `month_weekday_last`
- `year_month_weekday`
- `year_month_weekday_last`
Note these operators were added and removed in C++20.
Reviewed By: #libc, Mordante
Differential Revision: https://reviews.llvm.org/D152699
Kazu Hirata [Mon, 12 Jun 2023 19:21:05 +0000 (12:21 -0700)]
[RISCV] Fix an unused variable warning
This patch fixes:
llvm/lib/Target/RISCV/RISCVISelDAGToDAG.cpp:3214:12: error: unused
variable 'TSFlags' [-Werror,-Wunused-variable]
Craig Topper [Mon, 12 Jun 2023 18:57:44 +0000 (11:57 -0700)]
[RISCV] Remove HasDummyMask from tablegen files.
This isn't needed after D152050.
Reviewed By: reames
Differential Revision: https://reviews.llvm.org/D152218
Craig Topper [Mon, 12 Jun 2023 18:57:40 +0000 (11:57 -0700)]
[RISCV] Begin removing hasDummyMask.
This was used to know if we need to insert a dummy operand during
MCInstLowering. We can use the operand info from MCInstrDesc to
figure this out without needing a separate flag.
I'll remove the tablegen bits if there is consensus this is a good
idea.
Differential Revision: https://reviews.llvm.org/D152050
Noah Goldstein [Mon, 12 Jun 2023 18:18:31 +0000 (13:18 -0500)]
[InstCombine] Factor out power of 2 comparison patterns to helper; NFC
Differential Revision: https://reviews.llvm.org/D152728
Noah Goldstein [Mon, 12 Jun 2023 18:17:27 +0000 (13:17 -0500)]
[DAGCombine] Use `IsKnownNeverZero` to see if we need zero-check in is_pow2 setcc patern
`ctpop(X) eq/ne 1` is checking if X is a non-zero power of 2. Power of
2 check including zero is `(X & (X-1)) eq/ne 0` and unfortunately
there is no good pattern for checking a power of 2 while excluding
zero. So, when lowering `ctpop(X) eq/ne 1`, explicitly check
`IsKnownNeverZero(X)` to maybe be able to optimize out the extra zero
check.
We need this explicitly as DAGCombiner does not re-analyze provable
setcc nodes, and the middle-end never finds it beneficially to broaden
`ctpop(X) eq/ne 1` -> `ctpop(X) ule/ugt 1` (power of 2 including
zero).
Reviewed By: RKSimon
Differential Revision: https://reviews.llvm.org/D152675
Noah Goldstein [Mon, 12 Jun 2023 03:25:58 +0000 (22:25 -0500)]
[X86] Add tests for optimizing out zero-check in is_pow2 setcc pattern; NFC
Differential Revision: https://reviews.llvm.org/D152674
Fangrui Song [Mon, 12 Jun 2023 18:38:29 +0000 (11:38 -0700)]
[CMake] Remove unused COMPILER_RT_HAS_MSSE3_FLAG and COMPILER_RT_HAS_WNON_VIRTUAL_DTOR_FLAG
D106948 and commit
9c31e12609e1935eb84a2497ac08a49e3139859a removed the last
user, respectively.
Yitzhak Mandelbaum [Mon, 12 Jun 2023 17:38:31 +0000 (17:38 +0000)]
[clang][dataflow] Support limits on the SAT solver to force timeouts.
This patch allows the client of a `WatchedLiteralsSolver` to specify a
computation limit on the use of the solver. After the limit is exhausted, the
SAT solver times out.
Fixes issues #60265.
Differential Revision: https://reviews.llvm.org/D152732
Michael Buch [Mon, 12 Jun 2023 18:31:24 +0000 (19:31 +0100)]
[lldb][test] incomplete-tag-type.cpp: fix expected error message
Follow up to
133c3eaac0a532380c3d6ad21a60da1490f51fb8
Differential Revision: https://reviews.llvm.org/D152590
David Carlier [Mon, 12 Jun 2023 18:25:43 +0000 (19:25 +0100)]
sanitizers: enable pthread_getaffinity_np interception on freebsd.
Reviewers: mevler
Reviewed-By: mevler
Differential Revision: https://reviews.llvm.org/D152735
Mikhail Gudim [Mon, 12 Jun 2023 15:15:15 +0000 (11:15 -0400)]
[RISCV] Fold binary op into select if profitable.
Consider the following pattern `binOp (select cond, x, c0), c1`.
Where `c0` and `c1` are constants.
We can transform it to `select cond, binOp(x, c1), binOp(c0, c1)`.
If `binOp(c0, c1)` ends up being `0` or `-1` we can turn the select into
a more profitable sequence.
Michael Jones [Fri, 9 Jun 2023 22:12:25 +0000 (15:12 -0700)]
[libc][docs] Add undefined behavior doc to site
This document is based on the RFC posted to discourse:
https://discourse.llvm.org/t/rfc-defining-undefined-behavior-in-libc/
Reviewed By: sivachandra, lntue
Differential Revision: https://reviews.llvm.org/D152588
Michael Jones [Thu, 8 Jun 2023 21:04:55 +0000 (14:04 -0700)]
[libc] Add qsort_r
This patch adds the reentrent qsort entrypoint, qsort_r. This is done by
extending the qsort functionality and moving it to a shared utility
header. For this reason the qsort_r tests focus mostly on the places
where it differs from qsort, since they share the same sorting code.
Reviewed By: sivachandra, lntue
Differential Revision: https://reviews.llvm.org/D152467
Vitaly Buka [Mon, 12 Jun 2023 18:03:09 +0000 (11:03 -0700)]
Revert "[SCCP] Replace new value's value state with removed value's"
Breaks all sanitizers bootstrap bots:
https://lab.llvm.org/buildbot/#/waterfall?tags=sanitizer
This reverts commit
cf79773a9006a7e22f3919268b7db381ddcb3abc.
Valentin Clement [Mon, 12 Jun 2023 17:59:05 +0000 (10:59 -0700)]
[flang][openacc] Add parser support for the force modifier in the collapse clause
This patch adds parser support for the force modifier on the collapse clause
introduced in OpenACC 3.3.
Lowering will currently hit a TODO as the MLIR representation of the acc.loop
might need some update.
Reviewed By: kiranchandramohan
Differential Revision: https://reviews.llvm.org/D151974
Vitaly Buka [Mon, 12 Jun 2023 17:58:03 +0000 (10:58 -0700)]
[test][HWASAN] Trim leading zeros from hex value
Match like [[#T2]] does not include leading zeroes.
Maksim Panchenko [Wed, 7 Jun 2023 01:25:46 +0000 (18:25 -0700)]
[BOLT] Fix handling of code references from unmodified code
In lite mode (default for X86), BOLT optimizes and relocates functions
with profile. The rest of the code is preserved, but if it references
relocated code such references have to be updated. The update is handled
by scanExternalRefs() function. Note that we cannot solely rely on
relocations written by the linker, as not all code references are
exposed to the linker. Additionally, the linker can modify certain
instructions and relocations will no longer match the code.
With this change, start using symbolic disassembler for scanning code
for references in scanExternalRefs(). Unlike the previous approach, the
symbolizer properly detects and creates references for instructions with
multiple/ambiguous symbolic operands and handles cases where a
relocation doesn't match any operand. See test cases for examples.
Reviewed By: Amir
Differential Revision: https://reviews.llvm.org/D152631
Krzysztof Parzyszek [Tue, 6 Jun 2023 19:14:49 +0000 (12:14 -0700)]
[RDF] Stop looking when reached code node in getNextRef with NextOnly
NextOnly tells getNextRef to only check the next node in the chain, and
stop iterating. The loop didn't stop though when the next link pointed
back to the beginning of the circular list.
Recommit this one right now, because it fixes a latent bug.
LLVM GN Syncbot [Mon, 12 Jun 2023 17:33:51 +0000 (17:33 +0000)]
[gn build] Port
11e5a0d290ba
Adrian Prantl [Fri, 9 Jun 2023 22:29:28 +0000 (15:29 -0700)]
Streamline expression parser error messages.
Currently the expression parser prints a mostly useless generic error before printing the compiler error:
(lldb) p 1+x)
error: expression failed to parse:
error: <user expression 18>:1:3: use of undeclared identifier 'x'
1+x)
^
This is distracting and as far as I can tell only exists to work
around the fact that the first "error: " is unconditionally injected
by CommandReturnObject. The solution is not very elegant, but the
result looks much better.
(Partially addresses rdar://
110492710)
Differential Revision: https://reviews.llvm.org/D152590
Craig Topper [Mon, 12 Jun 2023 17:27:08 +0000 (10:27 -0700)]
[RISCV] Use tail undisturbed vmv.v.v instead of vadd.vi with 0 for vp.merge with all ones mask.
No idea what I was thinking when I suggested vadd.vi.
Reviewed By: reames, frasercrmck, fakepaper56
Differential Revision: https://reviews.llvm.org/D152553
Jonas Devlieghere [Mon, 12 Jun 2023 17:21:07 +0000 (10:21 -0700)]
[dsymutil] Use a regex in the static-archive-collision.test
Windows and Unix use different epochs: 1970-01-01 00:00:00.
000000000 and
1969-12-31 16:00:00.
000000000 respectively.
Giuseppe Rossini [Mon, 12 Jun 2023 16:56:08 +0000 (16:56 +0000)]
[AMDGPU] Add basic support for gfx11xx
This patch fixes a minor issue in AMDGPUToROCDL to add gfx11 support in MLIR
Reviewed By: krzysz00
Differential Revision: https://reviews.llvm.org/D152450
Craig Topper [Mon, 12 Jun 2023 17:01:34 +0000 (10:01 -0700)]
[RISCV] Use tail undisturbed vmv.v.v instead of vslideup.vi vN, vM, 0 for subvector insertion
vslideup has a vector overlap constraint that vmv.v.v doesn't.
vmv.v.v is also a simpler instruction so may have better
throughput and/or latency in some CPUs.
This is an alternative to D152298, D152368, and D152496.
Reviewed By: luke, reames
Differential Revision: https://reviews.llvm.org/D152565
Jonas Devlieghere [Mon, 12 Jun 2023 16:57:01 +0000 (09:57 -0700)]
[dsymutil] Skip duplicates files with identical time stamps in the debug map
Static archives can contain multiple files with the same file name, in
which case the timestamp is used to disambiguate. Because timestamps are
expressed in seconds since epoch timestamp collisions are far from
impossible. Furthermore, to facilitate reproducible builds, the static
linker can be told to emit no timestamps at all.
dsymutil already detects timestamp mismatches between the debug map and
the object files. However, it does not handle timestamp collisions
within the debug maps (STABS). Currently, we arbitrarily pick the first
debug map entry and ignore the rest. This is incorrect: if a symbol
exists in multiple object files, the linker might not have picked the
one from the first object file. This also results in missing symbol
warnings for all the symbols not defined in the first object file.
Given that in this scenario, dsymutil does not have enough information
to disambiguate, it should print a single informative warning and skip
the ambiguous debug map objects.
rdar://
110374836
Differential revision: https://reviews.llvm.org/D152585
Alfred Persson Forsberg [Mon, 12 Jun 2023 16:58:29 +0000 (17:58 +0100)]
[libc] Fix argument types for {f,}truncate specs
The current argument types are currently switched around for ftruncate
and truncate. Currently passes tests because the internal definitions
inside the __llvm_libc namespace are fine.
Reviewed By: michaelrj, thesamesam, sivachandra
Differential Revision: https://reviews.llvm.org/D152664
Mark de Wever [Sun, 4 Jun 2023 12:37:20 +0000 (14:37 +0200)]
[libc++][format] Make public functions nodiscard.
This is an extension and only adds the functions that are a considered a
but when called and ignoring the result.
Drive-by sort all nodiscard extensions in the documentation.
Reviewed By: #libc, philnik
Differential Revision: https://reviews.llvm.org/D152097
Amir Ayupov [Mon, 12 Jun 2023 16:54:42 +0000 (09:54 -0700)]
[BOLT][NFC] Const-ify getDynamicRelocationAt
Reviewed By: #bolt, maksfb
Differential Revision: https://reviews.llvm.org/D152662
Hristo Hristov [Sun, 11 Jun 2023 07:54:14 +0000 (10:54 +0300)]
[libc++][spaceship] P1614R2: Added `operator==` to `file_status`
Implemented parts of P1614R2:
- Added `operator==` to `file_status`
Reviewed By: #libc, Mordante
Differential Revision: https://reviews.llvm.org/D152647
Craig Topper [Mon, 12 Jun 2023 16:52:35 +0000 (09:52 -0700)]
[RISCV] Don't use (ADD (SLLI C, 32), C) to materialize constants when optimizing for size.
Addresses post-commit feedback from D152236.
Reviewed By: asb, dtcxzyw
Differential Revision: https://reviews.llvm.org/D152602
Hristo Hristov [Mon, 12 Jun 2023 06:56:15 +0000 (09:56 +0300)]
[libc++][spaceship] P1614R2: Removed `operator!=` from `scoped_allocator_adaptor`
Implements parts of P1614R2
- Removed `operator!=` from `scoped_allocator_adaptor`
Reviewed By: #libc, Mordante
Differential Revision: https://reviews.llvm.org/D152678
Hristo Hristov [Mon, 12 Jun 2023 12:17:02 +0000 (15:17 +0300)]
[libc++][spaceship] P1614R2: Removed `operator!=` from `functional`
Implements parts of P1614R2:
- Removed `operator!=` from `functional`
Reviewed By: #libc, Mordante
Differential Revision: https://reviews.llvm.org/D152704
Vladislav Dzhidzhoev [Mon, 12 Jun 2023 14:01:18 +0000 (16:01 +0200)]
[DebugMetadata] Simplify handling subprogram's retainedNodes field. NFCI (1/7)
RFC https://discourse.llvm.org/t/rfc-dwarfdebug-fix-and-improve-handling-imported-entities-types-and-static-local-in-subprogram-and-lexical-block-scopes/68544
Currently, `retainedNodes` tracks function-local variables and labels.
To support function-local import, types and static variables (which are globals
in LLVM IR), subsequent patches use the same field. So this patch makes
preliminary refactoring of the code tracking local entities to apply future
functional changes lucidly and cleanly.
No functional changes intended.
Differential Revision: https://reviews.llvm.org/D143984
Authored-by: Kristina Bessonova <kbessonova@accesssoftek.com>
Craig Topper [Mon, 12 Jun 2023 16:36:57 +0000 (09:36 -0700)]
[RISCV] Merge RISCVMCInstLower.cpp into RISCVAsmPrinter.cpp.
The separation here doesn't make much sense. I think it's a
leftover from the creation of the MC layer that has been
replicated to new targets.
By merging them we can avoid passing the AsmPrinter to the
MCInstLowering functions. We can make them member functions instead.
I think we can still do more integration of lowerSymbolOperand
and lowerRISCVVMachineInstrToMCInst, but I wanted to get feedback
on the direction first.
Reviewed By: asb, barannikov88
Differential Revision: https://reviews.llvm.org/D152311
Aart Bik [Fri, 9 Jun 2023 19:48:45 +0000 (12:48 -0700)]
[mlir][sparse] add support for direct prod/and/min/max reductions
We recently fixed a bug in "sparsifying" such reductions, since
it incorrectly changed this into reductions over stored elements
only , which only works for add/sub/or/xor. However, we still want
to be able to "sparsify" the reductions even in the general case,
and this is a first step by rewriting them into a custom reduction
that feeds in the implicit zeros. NOTE HOWEVER, that in the long run
we want to do this better and feed in any implicit zero only ONCE
for efficiency.
Reviewed By: Peiming
Differential Revision: https://reviews.llvm.org/D152580
Anna Thomas [Mon, 12 Jun 2023 15:55:05 +0000 (11:55 -0400)]
[SelectionDAG][AArch64] Legalize FMAXIMUM/FMINIMUM
The missing legalization in SelectionDAG was identified when adding the
intrinsic support for vector reduction for maximum/minimum (D152370).
Fixes part of PR: https://github.com/llvm/llvm-project/issues/63267
Differential Revision: https://reviews.llvm.org/D152718
OCHyams [Mon, 12 Jun 2023 16:06:50 +0000 (17:06 +0100)]
[DebugInfo] Add DW_ATE_complex_float case to assert in isUnsignedDIType
Without this patch a `DW_ATE_complex_float` encoding trips an assertion in
`DebugHandlerBase::isUnsignedDIType` with the message `"Unsupported
encoding"`.
By adding a case to the `assert` for `DW_ATE_complex_float` it becomes
supported, behaving in the same way as the already supported `DW_ATE_float`
type (return false).
Note: For the reported reproducer:
#include <complex.h>
int main() {
long double complex r1;
}
The assertion isn't tripped without assignment tracking because instcombine
deletes everything, including the `dbg.declare`, without recovering any
location information. Whereas with assignment tracking we track a zeroing
memset that is emitted by clang.
Reviewed By: probinson
Differential Revision: https://reviews.llvm.org/D151795
Florian Hahn [Mon, 12 Jun 2023 16:07:01 +0000 (17:07 +0100)]
[ConstraintElim] Extend monotonic phi test coverage.
Add additional tests which exposed some crashes in upcoming patches and
test for and-implied-by-operands handling.
Nico Weber [Mon, 12 Jun 2023 15:32:29 +0000 (11:32 -0400)]
Revert "[RDF] Allow RegisterRef to contain register unit" and follow-ups
This reverts commit
859b05b02d3fd9ab6b77f2bed8df6902fe704806.
Also reverts these follow-ups:
Revert "[RDF] Remove `constexpr` from `hash"
This reverts commit
621507ce20ad8eef2986be2712631165e53b7d91.
Revert "[RDF] Do not use trailing return type after all, NFC"
This reverts commit
46e19e3a2c45e7fb5f501bdb983a7151c158304f.
Revert "[RDF] Stop looking when reached code node in getNextRef with NextOnly"
This reverts commit
a049ce9d1bd5a7c1c4fcccc6a801b72b00ea8e0f.
Revert "[RDF] Use trailing return type syntax, NFC"
This reverts commit
d3b34b7f3a7cbfc96aea897419f167b5ee19e61a.
Revert "[RDF] Define short type names: NodeAddr<XyzNode*> -> Xyz, NFC"
This reverts commit
f8ed60b56d1948422dda924fcf450560591e8a19.
Hans Wennborg [Mon, 12 Jun 2023 15:02:37 +0000 (17:02 +0200)]
Revert "[Driver] Mark many target-specific driver-only options as TargetSpecific"
This broke building the TSan runtime on Mac, see comment on
https://github.com/llvm/llvm-project/commit/
5548843d692a92a7840f14002debc3cebcb3cdc3
> so that they get an error on other targets. This change uses let statements to
> apply `Flags = [TargetSpecific]` to options (mostly -m*) without specifying `Flags`.
> Follow-up to D151590.
>
> For some options, e.g. -mdefault-build-attributes (D31813), -mbranch-likely
> (D38168), -mfpu=/-mabi= (
6890b9b71e525020ab58d436336664beede71575), a warning
> seems desired in at least certain cases. This is not the best practice, but this
> change works around them by not applying `Flags = [TargetSpecific]`.
>
> (
> For Intel CPU errata -malign-branch= family options, we also drop the unneeded
> NotXarchOption flag. This flag reports an error if the option is used with
> -Xarch_*. This error reporting does not seem very useful.
> )
This reverts commit
5548843d692a92a7840f14002debc3cebcb3cdc3.
Justin Cady [Mon, 12 Jun 2023 14:58:38 +0000 (10:58 -0400)]
[test][ASAN] Fix incorrect REQUIRES directive for scandir.c
The missing colon caused the test to run on unintended platforms.
Reviewed By: emaste
Differential Revision: https://reviews.llvm.org/D152711
Kazu Hirata [Mon, 12 Jun 2023 15:11:01 +0000 (08:11 -0700)]
[FlowSensitive] Use {DenseMapBase,StringMap}::lookup (NFC)
Kazu Hirata [Mon, 12 Jun 2023 15:10:59 +0000 (08:10 -0700)]
[IR] Remove getABITypeAlignment
The last use of getABITypeAlignment was removed by:
commit
26bd6476c61f08fc8c01895caa02b938d6a37221
Author: Guillaume Chatelet <gchatelet@google.com>
Date: Fri Jan 13 15:05:24 2023 +0000
Differential Revision: https://reviews.llvm.org/D152670
Kazu Hirata [Mon, 12 Jun 2023 15:06:46 +0000 (08:06 -0700)]
[AMDGPU] Fix resource-usage-pal.ll
Akash Banerjee [Mon, 12 Jun 2023 14:00:17 +0000 (15:00 +0100)]
[OpenMP][OMPIRBuilder] Migrate emitOffloadingArrays and EmitNonContiguousDescriptor from Clang
This patch migrates the emitOffloadingArrays and EmitNonContiguousDescriptor functions from Clang codegen to OpenMPIRBuilder.
Reviewed By: jdoerfert
Differential Revision: https://reviews.llvm.org/D149872
Jie Fu [Mon, 12 Jun 2023 14:34:08 +0000 (22:34 +0800)]
Remove unused variable 's32' in X86LegalizerInfo.cpp (NFC)
/data/llvm-project/llvm/lib/Target/X86/X86LegalizerInfo.cpp:470:13: error: unused variable 's32' [-Werror,-Wunused-variable]
const LLT s32 = LLT::scalar(32);
^
1 error generated.
Simon Pilgrim [Mon, 12 Jun 2023 14:26:05 +0000 (15:26 +0100)]
[GlobalIsel][X86] Move G_SITOFP/G_FPTOSI getActionDefinitionsBuilder out of setLegalizerInfo64bit and add basic 32-bit support
We were using x86_64-only support as a SSE2 proxy - vector support is still missing.
Andrzej Warzynski [Fri, 2 Jun 2023 07:45:16 +0000 (08:45 +0100)]
[mlir][linalg] Add scalar broadcast load case to the vectoriser
This patch extends the Linalg vectoriser so that scalar loads are
correctly identified as scalar rather than gather loads. Below is an
example of a scalar load (note that both indices are loop invariant):
```
func.func @example(%arg0: tensor<80x16xf32>, %arg2: tensor<1x4xf32>) -> tensor<1x4xf32> {
%c8 = arith.constant 8 : index
%c16 = arith.constant 16 : index
%1 = linalg.generic {
indexing_maps = [affine_map<(d0, d1) -> (d0, d1)>],
iterator_types = ["parallel", "parallel"]
} outs(%arg2 : tensor<1x4xf32>) {
^bb0(%out: f32):
%2 = linalg.index 0 : index
%extracted = tensor.extract %arg0[%2, %c16] : tensor<80x16xf32>
linalg.yield %extracted : f32
} -> tensor<1x4xf32>
return %1 : tensor<1x4xf32>
}
```
This patch also makes sure that these scalar loads are indeed lowered to
a scalar load followed by a broadcast:
```
%extracted = tensor.extract %arg0[%1, %c16] : tensor<80x16xf32>
%2 = vector.broadcast %extracted : f32 to vector<1x4xf32>
```
Differential Revision: https://reviews.llvm.org/D149678
Baptiste [Mon, 12 Jun 2023 14:12:54 +0000 (10:12 -0400)]
[AMDGPU] Do not assume stack size for PAL code object indirect calls
There is no need to set a big default stack size for PAL code object indirect
calls. The driver knows the max recursion depth, so it can compute a more
accurate value from the minimum scratch size.
Reviewed By: arsenm
Differential Revision: https://reviews.llvm.org/D150609
Leonard Grey [Fri, 9 Jun 2023 22:04:48 +0000 (18:04 -0400)]
[gn] Mac support for check-asan target
Differential Revision: https://reviews.llvm.org/D152591
Guillaume Chatelet [Mon, 12 Jun 2023 13:30:50 +0000 (13:30 +0000)]
[libc] Improve memcmp latency and codegen
This is based on ideas from @nafi to:
- use a branchless version of 'cmp' for 'uint32_t',
- completely resolve the lexicographic comparison through vector
operations when wide types are available. We also get rid of byte
reloads and serializing '__builtin_ctzll'.
I did not include the suggestion to replace comparisons of 'uint16_t'
with two 'uint8_t' as it did not seem to help the codegen. This can
be revisited in sub-sequent patches.
The code been rewritten to reduce nested function calls, making the
job of the inliner easier and preventing harmful code duplication.
Reviewed By: nafi3000
Differential Revision: https://reviews.llvm.org/D148717
Timm Bäder [Mon, 12 Jun 2023 13:41:34 +0000 (15:41 +0200)]
[clang][tests] Replace a %clang++ with %clang
%clang++ does not work on windows, see e.g.:
https://lab.llvm.org/staging/#/builders/204/builds/2141
oontvoo [Mon, 12 Jun 2023 12:16:08 +0000 (08:16 -0400)]
[MC][MachO]Do not emit DWARF for no-personality case
Detail: Follow up to D144999, where we emitted DWARF for non-canonical personality.
Reviewed By: jyknight
Differential Revision: https://reviews.llvm.org/D152540
Tue Ly [Tue, 6 Jun 2023 15:50:26 +0000 (11:50 -0400)]
[libc] Add platform independent floating point rounding mode checks.
Many math functions need to check for floating point rounding modes to
return correct values. Currently most of them use the internal implementation
of `fegetround`, which is platform-dependent and blocking math functions to be
enabled on platforms with unimplemented `fegetround`. In this change, we add
platform independent rounding mode checks and switching math functions to use
them instead. https://github.com/llvm/llvm-project/issues/63016
Reviewed By: sivachandra
Differential Revision: https://reviews.llvm.org/D152280
Nikita Popov [Mon, 12 Jun 2023 13:14:27 +0000 (15:14 +0200)]
[GVN] Fix verifyRemoved() verification
Fix the verification failure reported in
https://reviews.llvm.org/D141712#4413647. We need to remove the
load from the VN table as well, not just the leader table.
Also make sure that this verification always runs when assertions
are enabled, rather than only when -debug is passed.
Nikita Popov [Mon, 12 Jun 2023 12:49:02 +0000 (14:49 +0200)]
[LoopVectorize] Convert test to opaque pointers (NFC)
I'm keeping the bitcast in the input here, because without it
we end up introducing a stride 1 assumption and end up testing
a different case.