All of these found by grepping through IR from the builtin tests for extra trunc and zext/sext instructions that shouldn't have been there.
Some of these were real bugs where we lost bits from the user input:
_mm512_mask_broadcast_f32x8
_mm512_maskz_broadcast_f32x8
_mm512_mask_broadcast_i32x8
_mm512_maskz_broadcast_i32x8
_mm256_mask_cvtusepi16_storeu_epi8
llvm-svn: 336042
Summary:
Changes to some clang side tests to go with the summary parsing patch.
Depends on D47905.
Reviewers: pcc, dexonsmith, mehdi_amini
Subscribers: inglorion, eraman, cfe-commits, steven_wu
Differential Revision: https://reviews.llvm.org/D47906
llvm-svn: 335618
The WebAssembly backend in particular benefits from being
able to distinguish between varargs functions (...) and prototype-less
C functions.
Differential Revision: https://reviews.llvm.org/D48443
llvm-svn: 335510
With MSVC, PCH files are created along with an object file that needs to
be linked into the final library or executable. That object file
contains the code generated when building the headers. In particular, it
will include definitions of inline dllexport functions, and because they
are emitted in this object file, other files using the PCH do not need
to emit them. See the bug for an example.
This patch makes clang-cl match MSVC's behaviour in this regard, causing
significant compile-time savings when building dlls using precompiled
headers.
For example, in a 64-bit optimized shared library build of Chromium with
PCH, it reduces the binary size and compile time of
stroke_opacity_custom.obj from 9315564 bytes to 3659629 bytes and 14.6
to 6.63 s. The wall-clock time of building blink_core.dll goes from
38m41s to 22m33s. ("user" time goes from 1979m to 1142m).
Differential Revision: https://reviews.llvm.org/D48426
llvm-svn: 335466
Since we are now producing a summary also for regular LTO builds, we
need to run the NameAnonGlobals pass in those cases as well (the
summary cannot handle anonymous globals).
See https://reviews.llvm.org/D34156 for details on the original change.
This reverts commit 6c9ee4a4a438a8059aacc809b2dd57128fccd6b3.
llvm-svn: 335385
dead code.
This is important for C++ templates that essentially compute the valid
input in a way that is constant and will cause all the invalid cases to
be dead code that is deleted. Code in the wild actually does this and
GCC also accepts these kinds of patterns so it is important to support
it.
To make this work, we provide a non-error path to diagnose these issues,
and use a default-error warning instead. This keeps the relatively
strict handling but prevents nastiness like SFINAE on these errors. It
also allows us to safely use the system to diagnose this only when it
occurs at runtime (in emitted code).
Entertainingly, this required fixing the syntax in various other ways
for the x86 test because we never bothered to diagnose that the returns
were invalid.
Since debugging these compile failures was super confusing, I've also
improved the diagnostic to actually say what the value was. Most of the
checks I've made ignore this to simplify maintenance, but I've checked
it in a few places to make sure the diagnsotic is working.
Depends on D48462. Without that, we might actually crash some part of
the compiler after bypassing the error here.
Thanks to Richard, Ben Kramer, and especially Craig Topper for all the
help here.
Differential Revision: https://reviews.llvm.org/D48464
llvm-svn: 335309
D48464 contains changes that will loosen some of the range checks in SemaChecking to a DefaultError warning that can be disabled.
This patch adds explicit masking to avoid using the upper bits of immediates to gracefully handle the warning being disabled.
llvm-svn: 335308
This is breaking a couple of buildbots. We need to run the
NameAnonGlobal pass for regular LTO now as well (since we're producing a
summary). I'll post a separate patch for review to make this happen and
then re-commit.
This reverts commit c0759b7b1f4a81ff9021b952aa38a222d5fa4dfd.
llvm-svn: 335291
Summary:
With D33921, we gained the ability to have module summaries in regular
LTO modules without triggering ThinLTO compilation. Module summaries in
regular LTO allow garbage collection (dead stripping) before LTO
compilation and thus open up additional optimization opportunities.
This patch enables summary emission in regular LTO for all targets
except ld64-based ones (which use the legacy LTO API).
Reviewers: pcc, tejohnson, mehdi_amini
Subscribers: inglorion, eraman, cfe-commits
Differential Revision: https://reviews.llvm.org/D34156
llvm-svn: 335284
Summary:
This test is a strip down version of a function inside the
amalgamated sqlite source. When converted to IR clang produces
a phi instruction without debug location.
This patch fixes the above issue.
Differential Revision: https://reviews.llvm.org/D47720
llvm-svn: 335255
Similar to what was done to max/min recently.
These already reduced the vector width to 256 and 128 bit as we go unlike the original max/min code.
Differential Revision: https://reviews.llvm.org/D48346
llvm-svn: 335253
The SPIR target currently allows for half precision floating point types to be
emitted using the LLVM intrinsic functions which convert half types to floats
and doubles. However, this is illegal in SPIR as the only intrinsic allowed by
SPIR is memcpy, as per section 3 of the SPIR specification. Currently this is
leading to an assert being hit in the Clang CodeGen when attempting to emit a
constant or literal _Float16 type in a comparison operation on a SPIR or SPIR64
target. This assert stems from the CodeGen attempting to emit a constant half
value as an integer because the backend has specified that it is using these
half conversion intrinsics (which represents half as i16). This patch prevents
SPIR targets from using these intrinsics by overloading the responsible target
info method, marks SPIR targets as having a legal half type and provides
additional regression testing for the _Float16 type on SPIR targets.
Patch by: Stephen McGroarty
Differential Revision: https://reviews.llvm.org/D48188
llvm-svn: 335111
We only need to use 512 bit vectors all the way through v8i64 reductions since those max instructions are new to avx512f and only available in 512 bits until SKX.
For v16i32 and floating point we have legacy 128/256 bit instructions we can use.
I've tried to use other intrinsics to reduce the verbosity of the code and avoid having to mention all the shuffles. I've also removed all the -1 shuffle indices so the output sequence is fully specified and not left to backend optimization.
Differential Revision: https://reviews.llvm.org/D47401
llvm-svn: 335070
Clang/LLVM doesn't have a way to pass an HLE hint through to the X86 backend to emit HLE prefixed instructions. So this is a good short term fix.
Differential Revision: https://reviews.llvm.org/D47672
llvm-svn: 334751
Summary: These intrinsics result in hint instructions. They are provided here for MSVC ARM64 compatibility.
Reviewers: mstorsjo, compnerd, javed.absar
Reviewed By: mstorsjo
Subscribers: kristof.beyls, chrib, cfe-commits
Differential Revision: https://reviews.llvm.org/D48132
llvm-svn: 334639
https://bugs.llvm.org/show_bug.cgi?id=37778
...shows a miscompile resulting from marking nan builtins as 'const'.
The nan libcalls/builtins take a pointer argument:
http://www.cplusplus.com/reference/cmath/nan-function/
...and the chars dereferenced by that arg are used to fill in the NaN constant payload bits.
"const" means that the pointer argument isn't dereferenced. That's translated to "readnone" in LLVM.
"pure" means that the pointer argument may be dereferenced. That's translated to "readonly" in LLVM.
This change prevents the IR optimizer from killing the lead-up to the nan call here:
double a() {
char buf[4];
buf[0] = buf[1] = buf[2] = '9';
buf[3] = '\0';
return __builtin_nan(buf);
}
...the optimizer isn't currently able to simplify this to a constant as we might hope,
but this patch should solve the miscompile.
Differential Revision: https://reviews.llvm.org/D48134
llvm-svn: 334628
Summary:
This fixes the ranges for the vcvth family of FP16 intrinsics in the clang front end. Previously it was accepting incorrect ranges
-Changed builtin range checking in SemaChecking
-added tests SemaCheck changes - included in their own file since no similar one exists
-modified existing tests to reflect new ranges
Reviewers: SjoerdMeijer, javed.absar
Reviewed By: SjoerdMeijer
Subscribers: kristof.beyls, cfe-commits
Differential Revision: https://reviews.llvm.org/D47592
llvm-svn: 334489
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
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
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
Test changes are due to differences in how we generate undef elements now. We also changed the types used for extractf128_si256/insertf128_si256 to match the signature of the builtin that previously existed which this patch resurrects. This also matches gcc.
llvm-svn: 334261
The windows-msvc target is used for MSVC ABI compatibility, including
the exceptions model. It doesn't make sense to pair a windows-msvc
target with a non-MSVC exception model. This would previously cause an
assertion failure; explicitly error out for it in the frontend instead.
This also allows us to reduce the matrix of target/exception models a
bit (see the modified tests), and we can possibly simplify some of the
personality code in a follow-up.
Differential Revision: https://reviews.llvm.org/D47853
llvm-svn: 334243
Adds support for these intrinsics, which are ARM and ARM64 only:
_interlockedbittestandreset_acq
_interlockedbittestandreset_rel
_interlockedbittestandreset_nf
_interlockedbittestandset_acq
_interlockedbittestandset_rel
_interlockedbittestandset_nf
Refactor the bittest intrinsic handling to decompose each intrinsic into
its action, its width, and its atomicity.
llvm-svn: 334239
We still lower them to native shuffle IR, but we do it in CGBuiltin.cpp now. This allows us to check the target feature and ensure the immediate fits in 8 bits.
This also improves our -O0 codegen slightly because we're able to see the zeroinitializer in the shuffle. It looks like it got lost behind a store+load previously.
llvm-svn: 334208
Summary:
When requirement imposed by __target__ attributes on functions
are not satisfied, prefer printing those requirements, which
are explicitly mentioned in the attributes.
This makes such messages more useful, e.g. printing avx512f instead of avx2
in the following scenario:
```
$ cat foo.c
static inline void __attribute__((__always_inline__, __target__("avx512f")))
x(void)
{
}
int main(void)
{
x();
}
$ clang foo.c
foo.c:7:2: error: always_inline function 'x' requires target feature 'avx2', but would be inlined into function 'main' that is compiled without support for 'avx2'
x();
^
1 error generated.
```
bugzilla: https://bugs.llvm.org/show_bug.cgi?id=37338
Reviewers: craig.topper, echristo, dblaikie
Reviewed By: craig.topper, echristo
Differential Revision: https://reviews.llvm.org/D46541
llvm-svn: 334174
Summary:
We recently switch to using a selects in the intrinsics header files for FMA instructions. But the 512-bit versions support flavors with rounding mode which must be an Integer Constant Expression. This has forced those intrinsics to be implemented as macros. As it stands now the mask and mask3 intrinsics evaluate one of their macro arguments twice. If that argument itself is another intrinsic macro, we can end up over expanding macros. Or if its something we can CSE later it would show up multiple times when it shouldn't.
I tried adding __extension__ around the macro and making it an expression statement and declaring a local variable. But whatever name you choose for the local variable can never be used as the name of an input to the macro in user code. If that happens you would end up with the same name on the LHS and RHS of an assignment after expansion. We might be safe if we use __ in front of the variable names because those names are reserved and user code shouldn't use that, but I wasn't sure I wanted to make that claim.
The other option which I've chosen here, is to add back _mask, _maskz, and _mask3 flavors of the builtin which we will expand in CGBuiltin.cpp to replicate the argument as needed and insert any fneg needed on the third operand to make a subtract. The _maskz isn't truly necessary if we have an unmasked version or if we use the masked version with a -1 mask and wrap a select around it. But I've chosen to make things more uniform.
I separated out the scalar builtin handling to avoid too many things going on in EmitX86FMAExpr. It was different enough due to the extract and insert that the minor duplication of the CreateCall was probably worth it.
Reviewers: tkrupa, RKSimon, spatel, GBuella
Reviewed By: tkrupa
Subscribers: cfe-commits
Differential Revision: https://reviews.llvm.org/D47724
llvm-svn: 334159
Factor out the common setjmp call emission code.
Based on a patch by Chris January
Differential Revision: https://reviews.llvm.org/D47784
llvm-svn: 334112
I tested these locally on an x86 machine by disabling the inline asm
codepath and confirming that it does the same bitflips as we do with the
inline asm.
Addresses code review feedback.
llvm-svn: 334059
Previously we were just using extended vector operations in the header file.
This unfortunately allowed non-constant indices to be used with the intrinsics. This is incompatible with gcc, icc, and MSVC. It also introduces a different performance characteristic because non-constant index gets lowered to a vector store and an element sized load.
By adding the builtins we can check for the index to be a constant and ensure its in range of the vector element count.
User code still has the option to use extended vector operations themselves if they need non-constant indexing.
llvm-svn: 334057
We need to implement _interlockedbittestandset as a builtin for
windows.h, so we might as well do the whole family. It reduces code
duplication anyway.
Fixes PR33188, a long standing bug in our bittest implementation
encountered by Chakra.
llvm-svn: 333978
Summary:
Adds testing of combined index summary entries in disassembly format
to CFI tests that were already testing the bitcode format.
Depends on D46699.
Reviewers: pcc
Subscribers: mehdi_amini, inglorion, eraman, cfe-commits
Differential Revision: https://reviews.llvm.org/D46700
llvm-svn: 333966
Adding __attribute__((aligned(32))) to __m256 breaks the implementation
of _mm256_loadu_ps on Windows. On Windows, alignment attributes have
higher precedence than packing attributes.
We also might want to carefully consider the consequences of changing
our vector typedefs, since many users copy them and invent their own
new, non-Intel specific vector type names.
llvm-svn: 333958
This is more consistent with other usages of builtin_shufflevector. Later optimization passes or codegen will detect the duplicate vector and replace it with undef. Using _mm_undefined just puts a zeroinitializer that still needs to be optimized out later.
llvm-svn: 333944
We currently support them only in AArch64. The NEON Reference,
however, says they are 'ARMv7, ARMv8' intrinsics.
Differential Revision: https://reviews.llvm.org/D47121
llvm-svn: 333829
This fixes two major problems:
- We were not capping vector alignment as desired on 32-bit ARM.
- We were using different alignments based on the AVX settings on
Intel, so we did not have a consistent ABI.
This is an ABI break, but we think we can get away with it because
vectors tend to be used mostly in inline code (which is why not having
a consistent ABI has not proven disastrous on Intel).
Intel's AVX types are specified as having 32-byte / 64-byte alignment,
so align them explicitly instead of relying on the base ABI rule.
Note that this sort of attribute is stripped from template arguments
in template substitution, so there's a possibility that code templated
over vectors will produce inadequately-aligned objects. The right
long-term solution for this is for alignment attributes to be
interpreted as true qualifiers and thus preserved in the canonical type.
llvm-svn: 333791
The WebAssembly committee has decided on the names `memory.size` and
`memory.grow` for the memory intrinsics, so update the clang builtin
functions to follow those names, keeping both sets of old names in place
for compatibility.
llvm-svn: 333712
I think this is a holdover from when we used to declare variables inside the macros. And then its been copy and pasted forward for years every time a new macro intrinsic gets added.
Interestingly this caused some tests for IRGen to be slightly more optimized. We now return a zeroinitializer directly instead of going through a store+load.
It also removed a bogus error message on another test.
llvm-svn: 333613
We don't need the insertion back into the original vector at the end. The builtin already understands that.
This is different than _mm_sqrt_sd which takes two arguments and we do need to insert.
llvm-svn: 333572
We had quite a few for different element sizes of integers sometimes with strange target features attached to them.
We only need a single version for each of _m128i, _m256i, and _m512i with the target feature that first introduced those types.
llvm-svn: 333568
This patch replaces all packed (and scalar without rounding
mode) fused intrinsics with fmadd/fmaddsub variations.
Then fmadd/fmaddsub are lowered to native IR.
Patch by tkrupa
Reviewers: craig.topper, sroland, spatel, RKSimon
Reviewed By: craig.topper
Differential Revision: https://reviews.llvm.org/D47444
llvm-svn: 333555
These intrinsics are used by MSVC's header files on AArch64 Windows as
well as AArch32, so we should support them for both targets. I've
factored them out of CodeGenFunction::EmitARMBuiltinExpr into separate
functions that EmitAArch64BuiltinExpr can call as well.
Reviewers: javed.absar, mstorsjo
Reviewed By: mstorsjo
Subscribers: kristof.beyls, cfe-commits
Differential Revision: https://reviews.llvm.org/D47476
llvm-svn: 333513
Mostly this fixes the names of all the 128-bit intrinsics to start with _mm_ instead of _mm128_ as is the convention and what the Intel docs say.
This also fixes the name of the bitshuffle intrinsics to say epi64 for 128 and 256 bit versions.
llvm-svn: 333497
Summary:
We only need to use 512 bit vectors all the way through v8i64 reductions since those max instructions are new to avx512f and only available in 512 bits until SKX.
For v16i32 and floating point we have legacy 128/256 bit instructions we can use.
I've tried to use other intrinsics to reduce the verbosity of the code and avoid having to mention all the shuffles. I've also removed all the -1 shuffle indices so the output sequence is fully specified and not left to backend optimization.
Reviewers: RKSimon, spatel, GBuella
Subscribers: cfe-commits
Differential Revision: https://reviews.llvm.org/D47401
llvm-svn: 333347
The checksum will not reflect the real source, so there's no clear
reason to include them in the debug info. Also this was causing a
crash on the DWARF side.
Differential Revision: https://reviews.llvm.org/D47260
llvm-svn: 333311
An intrinsic for an old instruction, as described in the Intel SDM.
Reviewers: craig.topper, rnk
Reviewed By: craig.topper, rnk
Differential Revision: https://reviews.llvm.org/D47142
llvm-svn: 333256
to checking for attributes on the call site - and fix up builtin
functions that we were testing for but not ensuring wouldn't be
optimized by the backend.
Leave one set of asm tests to make sure that we're also communicating
builtin-ness to TLI.
llvm-svn: 333154
It caused asserts, see PR37560.
> Use zeroinitializer for (trailing zero portion of) large array initializers
> more reliably.
>
> Clang has two different ways it emits array constants (from InitListExprs and
> from APValues), and both had some ability to emit zeroinitializer, but neither
> was able to catch all cases where we could use zeroinitializer reliably. In
> particular, emitting from an APValue would fail to notice if all the explicit
> array elements happened to be zero. In addition, for large arrays where only an
> initial portion has an explicit initializer, we would emit the complete
> initializer (which could be huge) rather than emitting only the non-zero
> portion. With this change, when the element would have a suffix of more than 8
> zero elements, we emit the array constant as a packed struct of its initial
> portion followed by a zeroinitializer constant for the trailing zero portion.
>
> In passing, I found a bug where SemaInit would sometimes walk the entire array
> when checking an initializer that only covers the first few elements; that's
> fixed here to unblock testing of the rest.
>
> Differential Revision: https://reviews.llvm.org/D47166
llvm-svn: 333067
Previously we negated the whole vector after splatting infinity. But its better to negate the infinity before splatting. This generates IR with the negate already folded with the infinity constant.
llvm-svn: 333062
more reliably.
Clang has two different ways it emits array constants (from InitListExprs and
from APValues), and both had some ability to emit zeroinitializer, but neither
was able to catch all cases where we could use zeroinitializer reliably. In
particular, emitting from an APValue would fail to notice if all the explicit
array elements happened to be zero. In addition, for large arrays where only an
initial portion has an explicit initializer, we would emit the complete
initializer (which could be huge) rather than emitting only the non-zero
portion. With this change, when the element would have a suffix of more than 8
zero elements, we emit the array constant as a packed struct of its initial
portion followed by a zeroinitializer constant for the trailing zero portion.
In passing, I found a bug where SemaInit would sometimes walk the entire array
when checking an initializer that only covers the first few elements; that's
fixed here to unblock testing of the rest.
Differential Revision: https://reviews.llvm.org/D47166
llvm-svn: 333044
The clang builtins have the same semantics as the stdlib functions.
The stdlib functions are defined in section 7.20.6.1 of the C standard with:
"If the result cannot be represented, the behavior is undefined."
That lets us mark the negation with 'nsw' because "sub i32 0, INT_MIN" would
be UB/poison.
Differential Revision: https://reviews.llvm.org/D47202
llvm-svn: 333038
This change will help Visual Studio resolve forward references to C++ lambda
routines used by captured variables.
Differential Revision: https://reviews.llvm.org/D45438
llvm-svn: 332975
Because the intrinsics in the headers are implemented as macros, we can't just use a select builtin and pternlog builtin. This would require one of the macro arguments to be used twice. Depending on what was passed to the macro we could expand an expression twice leading to weird behavior. We could maybe declare our local variable in the macro, but that would need to worry about name collisions.
To avoid that just generate IR directly in CGBuiltin.cpp.
Differential Revision: https://reviews.llvm.org/D47125
llvm-svn: 332891
I believe this is safe assuming default default FP environment. The conversion might be inexact, but it can never overflow the FP type so this shouldn't be undefined behavior for the uitofp/sitofp instructions.
We already do something similar for scalar conversions.
Differential Revision: https://reviews.llvm.org/D46863
llvm-svn: 332882
If a variable has an initializer, codegen tries to build its value. If
the variable is large in size, building its value requires substantial
resources. It causes strange behavior from user viewpoint: compilation
of huge zero initialized arrays like:
char data_1[2147483648u] = { 0 };
consumes enormous amount of time and memory.
With this change codegen tries to determine if variable initializer is
equivalent to zero initializer. In this case variable value is not
constructed.
This change fixes PR18978.
Differential Revision: https://reviews.llvm.org/D46241
llvm-svn: 332847
in gcc by https://gcc.gnu.org/ml/gcc-cvs/2018-04/msg00534.html.
The -mibt feature flag is being removed, and the -fcf-protection
option now also defines a CET macro and causes errors when used
on non-X86 targets, while X86 targets no longer check for -mibt
and -mshstk to determine if -fcf-protection is supported. -mshstk
is now used only to determine availability of shadow stack intrinsics.
Comes with an LLVM patch (D46882).
Patch by mike.dvoretsky
Differential Revision: https://reviews.llvm.org/D46881
llvm-svn: 332704
The existing test for the AArch64 inline assembly constraint S uses the
A and L modifiers. These modifiers were implemented in the original
AArch64 backend but were not carried forward to the merged backend. The
A is associated with ADRP and does nothing, the L is associated with
:lo12: . Given that A and L are not supported by GCC and not supported
by the new implementation of constraint S in LLVM (see D46745) I've
altered the test to put :lo12: directly in the string so that A and L
are not needed.
Differential Revision: https://reviews.llvm.org/D46932
llvm-svn: 332606
As long as the destination type is a 256 or 128 bit vector with the same number of elements we can use __builtin_convertvector to directly generate trunc IR instruction which will be handled natively by the backend.
Differential Revision: https://reviews.llvm.org/D46742
llvm-svn: 332266
If we're using default rounding mode we can let __builtin_convertvector to generate an fpextend. This matches 128 and 256 bit.
If we're using the version that takes an explicit rounding mode argument we would need to look at the immediate to see if its CUR_DIRECTION.
llvm-svn: 332210
We can use direct C code for these that will use uitofp and insertelement instructions.
For the versions that take an explicit rounding mode we can't do this.
llvm-svn: 332203
These intrinsics work exactly as all other atomic_fetch_* intrinsics and allow to create *atomicrmw* with ordering.
Updated the clang-extensions document.
Differential Revision: https://reviews.llvm.org/D46386
llvm-svn: 332193
This is unnecessary for AVX512VL supporting CPUs like SKX. We can just emit a 128-bit masked load/store here no matter what. The backend will widen it to 512-bits on KNL CPUs.
Fixes the frontend portion of PR37386. Need to fix the backend to optimize the new sequences well.
llvm-svn: 331958
Previously we emitted something like
rotl(x, n) {
n &= bitwidth-1;
return n != 0 ? ((x << n) | (x >> (bitwidth - n)) : x;
}
We use a select to avoid the undefined behavior on the (bitwidth - n) shift.
The middle and backend don't really recognize this as a rotate and end up emitting a cmov or control flow because of the select.
A better pattern is (x << (n & mask)) | (x << (-n & mask)) where mask is bitwidth - 1.
Fixes the main complaint in PR37387. There's still some work to be done if the user writes that sequence directly on a short or char where type promotion rules can prevent it from being recognized. The builtin is emitting direct IR with unpromoted types so that isn't a problem for it.
Differential Revision: https://reviews.llvm.org/D46656
llvm-svn: 331943
Summary:
This attribute tells clang to skip this function from stack protector
when -stack-protector option is passed.
GCC option for this is:
__attribute__((__optimize__("no-stack-protector"))) and the
equivalent clang syntax would be: __attribute__((no_stack_protector))
This is used in Linux kernel to selectively disable stack protector
in certain functions.
Reviewers: aaron.ballman, rsmith, rnk, probinson
Reviewed By: aaron.ballman
Subscribers: probinson, srhines, cfe-commits
Differential Revision: https://reviews.llvm.org/D46300
llvm-svn: 331925
It broke the Chromium build (see reply on the review).
> Generate DILabel metadata and call llvm.dbg.label after label
> statement to associate the metadata with the label.
>
> Differential Revision: https://reviews.llvm.org/D45045
>
> Patch by Hsiangkai Wang.
This doesn't revert the change to backend-unsupported-error.ll
that seems to correspond to an llvm-side change.
llvm-svn: 331861
Summary:
An _Atomic of an empty struct is pretty silly. In general we just widen empty
structs to hold a byte's worth of storage, and we represent size and alignment
as 0 internally and let LLVM figure out what to do. For _Atomic it's a bit
different: the memory model mandates concrete effects occur when atomic
operations occur, so in most cases actual instructions need to get emitted. It's
really not worth trying to optimize empty struct atomics by figuring out e.g.
that a fence would do, even though sane compilers should do optimize atomics.
Further, wg21.link/p0528 will fix C++20 atomics with padding bits so that
cmpxchg on them works, which means that we'll likely need to do the zero-init
song and dance for empty atomic structs anyways (and I think we shouldn't
special-case this behavior to C++20 because prior standards are just broken).
This patch therefore makes a minor change to r176658 "Promote atomic type sizes
up to a power of two": if the width of the atomic's value type is 0, just use 1
byte for width and leave alignment as-is (since it should never be zero, and
over-aligned zero-width structs are weird but fine).
This fixes an assertion:
(NumBits >= MIN_INT_BITS && "bitwidth too small"), function get, file ../lib/IR/Type.cpp, line 241.
It seems like this has run into other assertions before (namely the unreachable
Kind check in ImpCastExprToType), but I haven't reproduced that issue with
tip-of-tree.
<rdar://problem/39678063>
Reviewers: arphaman, rjmccall
Subscribers: aheejin, cfe-commits
Differential Revision: https://reviews.llvm.org/D46613
llvm-svn: 331845
Generate DILabel metadata and call llvm.dbg.label after label
statement to associate the metadata with the label.
Differential Revision: https://reviews.llvm.org/D45045
Patch by Hsiangkai Wang.
llvm-svn: 331843
The error turns out to be:
Assertion failed: (Target.isCompatibleDataLayout(getDataLayout()) && "Can't create a MachineFunction using a Module with a " "Target-incompatible DataLayout attached\n"), function init, file /Users/buildslave/jenkins/workspace/clang-stage1-cmake-RA-incremental/llvm/lib/CodeGen/MachineFunction.cpp, line 180.
Add -target to address this. Also re-enable the test I had temporarily
commented, and move it further down in case there is still a failure
(since it pipes stderr to FileCheck).
llvm-svn: 331597
Summary:
Passes down the necessary code ge options to the LTO Config to enable
-fdiagnostics-show-hotness and -fsave-optimization-record in the ThinLTO
backend for a distributed build.
Also, remove warning about not having PGO when the input is IR.
Reviewers: pcc
Subscribers: mehdi_amini, inglorion, eraman, cfe-commits
Differential Revision: https://reviews.llvm.org/D46464
llvm-svn: 331592
instrumentation codegeneration strategy of using a data structure and
a loop. Required some finesse to get the critical things being tested to
surface in a nice way for FileCheck but I think this preserves the
original intent of the test.
llvm-svn: 331411
Both sides of this #if #include the same file. Drop the #if, leaving only the #include.
Patch by Matt Glazar.
Differential Revision: https://reviews.llvm.org/D45779
llvm-svn: 331305
Teach AsmParser to check with Assembler for when evaluating constant
expressions. This improves the handing of preprocessor expressions
that must be resolved at parse time. This idiom can be found as
assembling-time assertion checks in source-level assemblers. Note that
this relies on the MCStreamer to keep sufficient tabs on Section /
Fragment information which the MCAsmStreamer does not. As a result the
textual output may fail where the equivalent object generation would
pass. This can most easily be resolved by folding the MCAsmStreamer
and MCObjectStreamer together which is planned for in a separate
patch.
Currently, this feature is only enabled for assembly input, keeping IR
compilation consistent between assembly and object generation.
Reviewers: echristo, rnk, probinson, espindola, peter.smith
Reviewed By: peter.smith
Subscribers: eraman, peter.smith, arichardson, jyknight, hiraditya, llvm-commits
Differential Revision: https://reviews.llvm.org/D45164
llvm-svn: 331218
As suggested in the post-commit thread for rL331056, we should match these
clang options with the established vocabulary of the corresponding sanitizer
option. Also, the use of 'strict' is well-known for these kinds of knobs,
and we can improve the descriptive text in the docs.
So this intends to match the logic of D46135 but only change the words.
Matching LLVM commit to match this spelling of the attribute to follow shortly.
Differential Revision: https://reviews.llvm.org/D46236
llvm-svn: 331209
As discussed in the post-commit thread for:
rL330437 ( http://lists.llvm.org/pipermail/llvm-commits/Week-of-Mon-20180423/545906.html )
We need a way to opt-out of a float-to-int-to-float cast optimization because too much
existing code relies on the platform-specific undefined result of those casts when the
float-to-int overflows.
The LLVM changes associated with adding this function attribute are here:
rL330947
rL330950
rL330951
Also as suggested, I changed the LLVM doc to mention the specific sanitizer flag that
catches this problem:
rL330958
Differential Revision: https://reviews.llvm.org/D46135
llvm-svn: 331041
The ACLE spec which describes these intrinsics hasn't been published yet, but
this is based on the final draft which will be published soon, and these have
already been implemented by GCC.
Differential revision: https://reviews.llvm.org/D46109
llvm-svn: 331039
On AVX512F targets we'll produce an emulated sequence using 3 pmuludqs with shifts and adds. On AVX512DQ we'll use vpmulld.
Fixes PR37140.
llvm-svn: 330923
Passing the features in random order will lead to unpredictable results
when some of the features are related (like the architecture-version
features on ARM).
It might be possible to fix this particular case in the ARM target code,
to avoid adding overlapping target features. But we should probably be
sorting in any case: the behavior shouldn't depend on StringMap's
hashing algorithm.
Differential Revision: https://reviews.llvm.org/D46030
llvm-svn: 330861
The unmasked versions already didn't have this restrction. I don't think gcc or icc limit these to 64-bit mode so we shouldn't either.
llvm-svn: 330681
If an atomic variable is misaligned (and that suspicion is why Clang emits
libcalls at all) the runtime support library will have to use a lock to safely
access it, with potentially very bad performance consequences. There's a very
good chance this is unintentional so it makes sense to issue a warning.
Also give it a named group so people can promote it to an error, or disable it
if they really don't care.
llvm-svn: 330566
Summary:
By default Clang outputs its version (including git commit hash, in
case of trunk builds) into object and assembly files. It might be
useful to have an option to disable this, especially for debugging
purposes.
This patch implements new command line flags -Qn and -Qy (the names
are chosen for compatibility with GCC). -Qn disables output of
the 'llvm.ident' metadata string and the 'producer' debug info. -Qy
(enabled by default) does the opposite.
Reviewers: faisalv, echristo, aprantl
Reviewed By: aprantl
Subscribers: aprantl, cfe-commits, JDevlieghere, rogfer01
Differential Revision: https://reviews.llvm.org/D45255
llvm-svn: 330442
This implements support for the previously ignored flag
`-falign-functions`. This allows the frontend to request alignment on
function definitions in the translation unit where they are not
explicitly requested in code. This is compatible with the GCC behaviour
and the ICC behaviour.
The scalar value passed to `-falign-functions` aligns functions to a
power-of-two boundary. If flag is used, the functions are aligned to
16-byte boundaries. If the scalar is specified, it must be an integer
less than or equal to 4096. If the value is not a power-of-two, the
driver will round it up to the nearest power of two.
llvm-svn: 330378
The force_align_arg_pointer attribute was using a hardcoded 16-byte
alignment value which in combination with -mstack-alignment=32 (or
larger) would produce a misaligned stack which could result in crashes
when accessing stack buffers using aligned AVX load/store instructions.
Fix the issue by using the "stackrealign" function attribute instead
of using a hardcoded 16-byte alignment.
Patch By: Gramner
Differential Revision: https://reviews.llvm.org/D45812
llvm-svn: 330331
This is the patch that lowers x86 intrinsics to native IR
in order to enable optimizations.
Patch by tkrupa
Differential Revision: https://reviews.llvm.org/D44786
llvm-svn: 330323
Summary:
A clang builtin for xray typed events. Differs from
__xray_customevent(...) by the presence of a type tag that is vended by
compiler-rt in typical usage. This allows xray handlers to expand logged
events with their type description and plugins to process traced events
based on type.
This change depends on D45633 for the intrinsic definition.
Reviewers: dberris, pelikan, rnk, eizan
Subscribers: cfe-commits, llvm-commits
Differential Revision: https://reviews.llvm.org/D45716
llvm-svn: 330220
Yaron Keren