Summary:
Clang -fpic defaults to -fno-semantic-interposition (GCC -fpic defaults
to -fsemantic-interposition).
Users need to specify -fsemantic-interposition to get semantic
interposition behavior.
Semantic interposition is currently a best-effort feature. There may
still be some cases where it is not handled well.
Reviewers: peter.smith, rnk, serge-sans-paille, sfertile, jfb, jdoerfert
Subscribers: dschuff, jyknight, dylanmckay, nemanjai, jvesely, kbarton, fedor.sergeev, asb, rbar, johnrusso, simoncook, sabuasal, niosHD, jrtc27, zzheng, edward-jones, atanasyan, rogfer01, MartinMosbeck, brucehoult, the_o, arphaman, PkmX, jocewei, jsji, Jim, lenary, s.egerton, pzheng, sameer.abuasal, apazos, luismarques, cfe-commits
Tags: #clang
Differential Revision: https://reviews.llvm.org/D73865
Summary:
In big-endian MVE, the simple vector load/store instructions (i.e.
both contiguous and non-widening) don't all store the bytes of a
register to memory in the same order: it matters whether you did a
VSTRB.8, VSTRH.16 or VSTRW.32. Put another way, the in-register
formats of different vector types relate to each other in a different
way from the in-memory formats.
So, if you want to 'bitcast' or 'reinterpret' one vector type as
another, you have to carefully specify which you mean: did you want to
reinterpret the //register// format of one type as that of the other,
or the //memory// format?
The ACLE `vreinterpretq` intrinsics are specified to reinterpret the
register format. But I had implemented them as LLVM IR bitcast, which
is specified for all types as a reinterpretation of the memory format.
So a `vreinterpretq` intrinsic, applied to values already in registers,
would code-generate incorrectly if compiled big-endian: instead of
emitting no code, it would emit a `vrev`.
To fix this, I've introduced a new IR intrinsic to perform a
register-format reinterpretation: `@llvm.arm.mve.vreinterpretq`. It's
implemented by a trivial isel pattern that expects the input in an
MQPR register, and just returns it unchanged.
In the clang codegen, I only emit this new intrinsic where it's
actually needed: I prefer a bitcast wherever it will have the right
effect, because LLVM understands bitcasts better. So we still generate
bitcasts in little-endian mode, and even in big-endian when you're
casting between two vector types with the same lane size.
For testing, I've moved all the codegen tests of vreinterpretq out
into their own file, so that they can have a different set of RUN
lines to check both big- and little-endian.
Reviewers: dmgreen, MarkMurrayARM, miyuki, ostannard
Reviewed By: dmgreen
Subscribers: kristof.beyls, hiraditya, cfe-commits, llvm-commits
Tags: #clang, #llvm
Differential Revision: https://reviews.llvm.org/D73786
Summary:
These instructions generate a vector of consecutive elements starting
from a given base value and incrementing by 1, 2, 4 or 8. The `wdup`
versions also wrap the values back to zero when they reach a given
limit value. The instruction updates the scalar base register so that
another use of the same instruction will continue the sequence from
where the previous one left off.
At the IR level, I've represented these instructions as a family of
target-specific intrinsics with two return values (the constructed
vector and the updated base). The user-facing ACLE API provides a set
of intrinsics that throw away the written-back base and another set
that receive it as a pointer so they can update it, plus the usual
predicated versions.
Because the intrinsics return two values (as do the underlying
instructions), the isel has to be done in C++.
This is the first family of MVE intrinsics that use the `imm_1248`
immediate type in the clang Tablegen framework, so naturally, I found
I'd given it the wrong C integer type. Also added some tests of the
check that the immediate has a legal value, because this is the first
time those particular checks have been exercised.
Finally, I also had to fix a bug in MveEmitter which failed an
assertion when I nested two `seq` nodes (the inner one used to extract
the two values from the pair returned by the IR intrinsic, and the
outer one put on by the predication multiclass).
Reviewers: dmgreen, MarkMurrayARM, miyuki, ostannard
Reviewed By: dmgreen
Subscribers: kristof.beyls, hiraditya, cfe-commits, llvm-commits
Tags: #clang, #llvm
Differential Revision: https://reviews.llvm.org/D73357
Summary:
The unpredicated case of this is trivial: the clang codegen just makes
a vector splat of the input, and LLVM isel is already prepared to
handle that. For the predicated version, I've generated a `select`
between the same vector splat and the `inactive` input parameter, and
added new Tablegen isel rules to match that pattern into a predicated
`MVE_VDUP` instruction.
Reviewers: dmgreen, MarkMurrayARM, miyuki, ostannard
Reviewed By: dmgreen
Subscribers: kristof.beyls, hiraditya, cfe-commits, llvm-commits
Tags: #clang, #llvm
Differential Revision: https://reviews.llvm.org/D73356
Driver errors if -fomit-frame-pointer is used together with -pg.
useFramePointerForTargetByDefault() returns true if -pg is specified.
=>
(!OmitFP && useFramePointerForTargetByDefault(Args, Triple)) is true
=>
We cannot get FramePointerKind::None
First attempt at implementing -fsemantic-interposition.
Rely on GlobalValue::isInterposable that already captures most of the expected
behavior.
Rely on a ModuleFlag to state whether we should respect SemanticInterposition or
not. The default remains no.
So this should be a no-op if -fsemantic-interposition isn't used, and if it is,
isInterposable being already used in most optimisation, they should honor it
properly.
Note that it only impacts architecture compiled with -fPIC and no pie.
Differential Revision: https://reviews.llvm.org/D72829
The constrained fcmp intrinsics don't allow the TRUE/FALSE predicates.
Using them will assert. To workaround this I'm emitting the old X86 specific intrinsics that were never removed from the backend when we switched to using fcmp in IR. We have no way to mark them as being strict, but that's true of all target specific intrinsics so doesn't seem like we need to solve that here.
I've also added support for selecting between signaling and quiet.
Still need to support SAE which will require using a target specific
intrinsic. Also need to fix masking to not use an AND instruction
after the compare.
Differential Revision: https://reviews.llvm.org/D72906
Summary:
Currently, sqdmulh_lane and friends from the ACLE (implemented in arm_neon.h),
are represented in LLVM IR as a (by vector) sqdmulh and a vector of (repeated)
indices, like so:
%shuffle = shufflevector <4 x i16> %v, <4 x i16> undef, <4 x i32> <i32 3, i32 3, i32 3, i32 3>
%vqdmulh2.i = tail call <4 x i16> @llvm.aarch64.neon.sqdmulh.v4i16(<4 x i16> %a, <4 x i16> %shuffle)
When %v's values are known, the shufflevector is optimized away and we are no
longer able to select the lane variant of sqdmulh in the backend.
This defeats a (hand-coded) optimization that packs several constants into a
single vector and uses the lane intrinsics to reduce register pressure and
trade-off materialising several constants for a single vector load from the
constant pool, like so:
int16x8_t v = {2,3,4,5,6,7,8,9};
a = vqdmulh_laneq_s16(a, v, 0);
b = vqdmulh_laneq_s16(b, v, 1);
c = vqdmulh_laneq_s16(c, v, 2);
d = vqdmulh_laneq_s16(d, v, 3);
[...]
In one microbenchmark from libjpeg-turbo this accounts for a 2.5% to 4%
performance difference.
We could teach the compiler to recover the lane variants, but this would likely
require its own pass. (Alternatively, "volatile" could be used on the constants
vector, but this is a bit ugly.)
This patch instead implements the following LLVM IR intrinsics for AArch64 to
maintain the original structure through IR optmization and into instruction
selection:
- sqdmulh_lane
- sqdmulh_laneq
- sqrdmulh_lane
- sqrdmulh_laneq.
These 'lane' variants need an additional register class. The second argument
must be in the lower half of the 64-bit NEON register file, but only when
operating on i16 elements.
Note that the existing patterns for shufflevector and sqdmulh into sqdmulh_lane
(etc.) remain, so code that does not rely on NEON intrinsics to generate these
instructions is not affected.
This patch also changes clang to emit these IR intrinsics for the corresponding
NEON intrinsics (AArch64 only).
Reviewers: SjoerdMeijer, dmgreen, t.p.northover, rovka, rengolin, efriedma
Reviewed By: efriedma
Subscribers: kristof.beyls, hiraditya, jdoerfert, cfe-commits, llvm-commits
Tags: #clang, #llvm
Differential Revision: https://reviews.llvm.org/D71469
It broke some Chromium tests, so let's revert until it can be fixed; see
https://crbug.com/1046362
This reverts commit 2af74e27ed7d0832cbdde9cb969aaca7a42e99f9.
Performing a cast where the result is ignored caused Clang to crash when
performing codegen for the conversion:
_Complex int a;
void fn1() { (_Complex double) a; }
This patch addresses the crash by not trying to emit the scalar conversions,
causing it to be a noop. Fixes PR44624.
This restores 59733525d37cf9ad88b5021b33ecdbaf2e18911c (D71913), along
with bot fix 19c76989bb505c3117730c47df85fd3800ea2767.
The bot failure should be fixed by D73418, committed as
af954e441a5170a75687699d91d85e0692929d43.
I also added a fix for non-x86 bot failures by requiring x86 in new test
lld/test/ELF/lto/devirt_vcall_vis_public.ll.
Summary:
Third part in series to support Safe Whole Program Devirtualization
Enablement, see RFC here:
http://lists.llvm.org/pipermail/llvm-dev/2019-December/137543.html
This patch adds type test metadata under -fwhole-program-vtables,
even for classes without hidden visibility. It then changes WPD to skip
devirtualization for a virtual function call when any of the compatible
vtables has public vcall visibility.
Additionally, internal LLVM options as well as lld and gold-plugin
options are added which enable upgrading all public vcall visibility
to linkage unit (hidden) visibility during LTO. This enables the more
aggressive WPD to kick in based on LTO time knowledge of the visibility
guarantees.
Support was added to all flavors of LTO WPD (regular, hybrid and
index-only), and to both the new and old LTO APIs.
Unfortunately it was not simple to split the first and second parts of
this part of the change (the unconditional emission of type tests and
the upgrading of the vcall visiblity) as I needed a way to upgrade the
public visibility on legacy WPD llvm assembly tests that don't include
linkage unit vcall visibility specifiers, to avoid a lot of test churn.
I also added a mechanism to LowerTypeTests that allows dropping type
test assume sequences we now aggressively insert when we invoke
distributed ThinLTO backends with null indexes, which is used in testing
mode, and which doesn't invoke the normal ThinLTO backend pipeline.
Depends on D71907 and D71911.
Reviewers: pcc, evgeny777, steven_wu, espindola
Subscribers: emaste, Prazek, inglorion, arichardson, hiraditya, MaskRay, dexonsmith, dang, davidxl, cfe-commits, llvm-commits
Tags: #clang, #llvm
Differential Revision: https://reviews.llvm.org/D71913
MSVC 2013 would refuse to pass highly aligned things (typically vectors
and aggregates) by value. Users would receive this error:
t.cpp(11) : error C2719: 'w': formal parameter with __declspec(align('32')) won't be aligned
t.cpp(11) : error C2719: 'q': formal parameter with __declspec(align('32')) won't be aligned
However, in MSVC 2015, this behavior was changed, and highly aligned
things are now passed indirectly. To avoid breaking backwards
incompatibility, objects that do not have a *required* high alignment
(i.e. double) are still passed directly, even though they are not
naturally aligned. This change implements the new behavior of passing
things indirectly.
The new behavior is:
- up to three vector parameters can be passed in [XYZ]MM0-2
- remaining arguments with required alignment greater than 4 bytes are
passed indirectly
Previously, MSVC never passed things truly indirectly, meaning clang
would always apply the byval attribute to indirect arguments. We had to
go to the trouble of adding inalloca so that non-trivially copyable C++
types could be passed in place without copying the object
representation. When inalloca was added, we asserted that all arguments
passed indirectly must use byval. With this change, that assert no
longer holds, and I had to update inalloca to handle that case. The
implicit sret pointer parameter was already handled this way, and this
change generalizes some of that logic to arguments.
There are two cases that this change leaves unfixed:
1. objects that are non-trivially copyable *and* overaligned
2. vectorcall + inalloca + vectors
For case 1, I need to touch C++ ABI code in MicrosoftCXXABI.cpp, so I
want to do it in a follow-up.
For case 2, my fix is one line, but it will require updating IR tests to
use lots of inreg, so I wanted to separate it out.
Related to D71915 and D72110
Fixes most of PR44395
Reviewed By: rjmccall, craig.topper, erichkeane
Differential Revision: https://reviews.llvm.org/D72114
Summary:
Much like with the previous patch (D73005) with `AssumeAlignedAttr`
handling, results in mildly more readable IR,
and will improve test coverage in upcoming patch.
Note that in `AllocAlignAttr`'s case, there is no requirement
for that alignment parameter to end up being an I-C-E.
Reviewers: erichkeane, jdoerfert, hfinkel, aaron.ballman, rsmith
Reviewed By: erichkeane
Subscribers: cfe-commits
Tags: #clang
Differential Revision: https://reviews.llvm.org/D73006
Summary:
This should be mostly NFC - we still lower the same alignment
knowledge to the IR. The main reasoning here is that
this somewhat improves readability of IR like this,
and will improve test coverage in upcoming patch.
Even though the alignment is guaranteed to always be an I-C-E,
we don't always materialize it as llvm's Alignment Attribute because:
1. There may be a non-zero offset
2. We may be sanitizing for alignment
Note that if there already was an IR alignment attribute
on return value, we union them, and thus the alignment
only ever rises.
Also, there is a second relevant clang attribute `AllocAlignAttr`,
so that is why `AbstractAssumeAlignedAttrEmitter` is templated.
Reviewers: erichkeane, jdoerfert, hfinkel, aaron.ballman, rsmith
Reviewed By: erichkeane
Subscribers: cfe-commits
Tags: #clang
Differential Revision: https://reviews.llvm.org/D73005
Summary:
Apparently nobody has tried this in months of development. It turns
out that `FunctionDecl::getBuiltinID` will never consider a function
to be a builtin if it is in C++ and not extern "C". So none of the
function declarations in <arm_mve.h> are recognized as builtins when
clang is compiling in C++ mode: it just emits calls to them as
ordinary functions, which then turn out not to exist at link time.
The trivial fix is to wrap most of arm_mve.h in an extern "C".
Added a test in clang/test/CodeGen/arm-mve-intrinsics which checks
basic functioning of the MVE header file in C++ mode. I've filled it
with copies of existing test functions from other files in that
directory, including a few moderately tricky cases of overloading (in
particular one that relies on the strict-polymorphism attribute added
in D72518).
(I considered making //every// test in that directory compile in both
C and C++ mode and check the code generation was identical. But I
think that would increase testing time by more than the value it adds,
and also update_cc_test_checks gets confused when the output function
name varies between RUN lines.)
Reviewers: LukeGeeson, MarkMurrayARM, miyuki, dmgreen
Reviewed By: MarkMurrayARM
Subscribers: kristof.beyls, cfe-commits
Tags: #clang
Differential Revision: https://reviews.llvm.org/D73268
Summary:
Immediate vmvnq is code-generated as a simple vector constant in IR,
and left to the backend to recognize that it can be created with an
MVE VMVN instruction. The predicated version is represented as a
select between the input and the same constant, and I've added a
Tablegen isel rule to turn that into a predicated VMVN. (That should
be better than the previous VMVN + VPSEL: it's the same number of
instructions but now it can fold into an adjacent VPT block.)
The unpredicated forms of VBIC and VORR are done by enabling the same
isel lowering as for NEON, recognizing appropriate immediates and
rewriting them as ARMISD::VBICIMM / ARMISD::VORRIMM SDNodes, which I
then instruction-select into the right MVE instructions (now that I've
also reworked those instructions to use the same MC operand encoding).
In order to do that, I had to promote the Tablegen SDNode instance
`NEONvorrImm` to a general `ARMvorrImm` available in MVE as well, and
similarly for `NEONvbicImm`.
The predicated forms of VBIC and VORR are represented as a vector
select between the original input vector and the output of the
unpredicated operation. The main convenience of this is that it still
lets me use the existing isel lowering for VBICIMM/VORRIMM, and not
have to write another copy of the operand encoding translation code.
This intrinsic family is the first to use the `imm_simd` system I put
into the MveEmitter tablegen backend. So, naturally, it showed up a
bug or two (emitting bogus range checks and the like). Fixed those,
and added a full set of tests for the permissible immediates in the
existing Sema test.
Also adjusted the isel pattern for `vmovlb.u8`, which stopped matching
because lowering started turning its input into a VBICIMM. Now it
recognizes the VBICIMM instead.
Reviewers: dmgreen, MarkMurrayARM, miyuki, ostannard
Reviewed By: dmgreen
Subscribers: kristof.beyls, hiraditya, cfe-commits, llvm-commits
Tags: #clang, #llvm
Differential Revision: https://reviews.llvm.org/D72934
Summary:
We shouldn't be just giving up if we find one of them
(like we currently do with `AssumeAlignedAttr`),
we should emit them all.
As the tests show, even if we materialized good knowledge
from `__attribute__((assume_aligned(32)`, it doesn't mean
`__attribute__((alloc_align([...])))` info won't be useful.
It might be, but that isn't given.
Reviewers: erichkeane, jdoerfert, aaron.ballman
Reviewed By: erichkeane
Subscribers: cfe-commits
Tags: #clang
Differential Revision: https://reviews.llvm.org/D72979
When constrained floating point is enabled the SystemZ-specific builtins
don't use constrained intrinsics in some cases. Fix that.
Differential Revision: https://reviews.llvm.org/D72722
Summary:
This patch resumes the work of D16586.
According to the AAPCS, volatile bit-fields should
be accessed using containers of the widht of their
declarative type. In such case:
```
struct S1 {
short a : 1;
}
```
should be accessed using load and stores of the width
(sizeof(short)), where now the compiler does only load
the minimum required width (char in this case).
However, as discussed in D16586,
that could overwrite non-volatile bit-fields, which
conflicted with C and C++ object models by creating
data race conditions that are not part of the bit-field,
e.g.
```
struct S2 {
short a;
int b : 16;
}
```
Accessing `S2.b` would also access `S2.a`.
The AAPCS Release 2019Q1.1
(https://static.docs.arm.com/ihi0042/g/aapcs32.pdf)
section 8.1 Data Types, page 35, "Volatile bit-fields -
preserving number and width of container accesses" has been
updated to avoid conflict with the C++ Memory Model.
Now it reads in the note:
```
This ABI does not place any restrictions on the access widths
of bit-fields where the container overlaps with a non-bit-field member.
This is because the C/C++ memory model defines these as being separate
memory locations, which can be accessed by two threads
simultaneously. For this reason, compilers must be permitted to use a
narrower memory access width (including splitting the access
into multiple instructions) to avoid writing to a different memory location.
```
I've updated the patch D16586 to follow such behavior by verifying that we
only change volatile bit-field access when:
- it won't overlap with any other non-bit-field member
- we only access memory inside the bounds of the record
Regarding the number of memory accesses, that should be preserved, that will
be implemented by D67399.
Reviewers: rsmith, rjmccall, eli.friedman, ostannard
Subscribers: ostannard, kristof.beyls, cfe-commits, carwil, olista01
Tags: #clang
Differential Revision: https://reviews.llvm.org/D72932
Extend -fxray-instrumentation-bundle to split function-entry and
function-exit into two separate options, so that it is possible to
instrument only function entry or only function exit. For use cases
that only care about one or the other this will save significant overhead
and code size.
Differential Revision: https://reviews.llvm.org/D72890
XRay allows tuning by minimum function size, but also always instruments
functions with loops in them. If the minimum function size is set to a
large value the loop instrumention ends up causing most functions to be
instrumented anyway. This adds a new flag, -fxray-ignore-loops, to disable
the loop detection logic.
Differential Revision: https://reviews.llvm.org/D72873
[this re-applies c0176916a4824812d25a5a22c4ff7c95857b0cd6
with the correct commit message and phabricator link]
This addresses point 1 of PR44213.
https://bugs.llvm.org/show_bug.cgi?id=44213
The DW_AT_LLVM_sysroot attribute is used for Clang module debug info,
to allow LLDB to import a Clang module from source. Currently it is
part of each DW_TAG_module, however, it is the same for all modules in
a compile unit. It is more efficient and less ambiguous to store it
once in the DW_TAG_compile_unit.
This should have no effect on DWARF consumers other than LLDB.
Differential Revision: https://reviews.llvm.org/D71732
Summary:
When compiling with -munwind-tables, the SEH filter funclet needs the uwtable
function attribute, which gets automatically added if we use
SetInternalFunctionAttributes. The filter funclet is internal so this seems
appropriate.
Reviewers: rnk
Subscribers: cfe-commits
Tags: #clang
Differential Revision: https://reviews.llvm.org/D72786
This is a purely cosmetic change that is NFC in terms of the binary
output. I bugs me that I called the attribute DW_AT_LLVM_isysroot
since the "i" is an artifact of GCC command line option syntax
(-isysroot is in the category of -i options) and doesn't carry any
useful information otherwise.
This attribute only appears in Clang module debug info.
Differential Revision: https://reviews.llvm.org/D71722
This reverts commit 3d210ed3d1880c615776b07d1916edb400c245a6.
See https://reviews.llvm.org/D71082 for the patch and discussion that make it
possible to reapply this patch.
This reverts commit 921f871ac438175ca8fcfcafdfcfac4d7ddf3905 because it
causes libc++ code to trigger __warn_memset_zero_len.
See https://reviews.llvm.org/D71082.
I noticed one bot failure due to
24a00ef2404104e9ca6fbd7eb523a8a340be9d99 because the wildcard matching
was not working as intended, fixed it to act similar to other checks of
CGSCCToFunctionPassAdaptor.
This restores 2af97be8027a0823b88d4b6a07fc5eedb440bc1f (reverted at
6288f86e870c7bb7fe47cc138320b9eb34c93941), with all the fixes I had
applied at the time, along with a new fix for non-determinism in the
ordering of a couple of passes due to being accessed as parameters on
the same call.
I've also added --dump-input=fail to the new tests so I can more
thoroughly fix any additional failures.
Summary:
Before this change, X86_32ABIInfo::classifyArgument would be called
twice on vector arguments to vectorcall functions. This function has
side effects to track GPR register usage, and this would lead to
incorrect GPR usage in some cases. The specific case I noticed is from
running out of XMM registers with mixed FP and vector arguments and no
aggregates of any kind. Consider this prototype:
void __vectorcall vectorcall_indirect_vec(
double xmm0, double xmm1, double xmm2, double xmm3, double xmm4,
__m128 xmm5,
__m128 ecx,
int edx,
__m128 mem);
classifyArgument has no effects when called on a plain FP type, but when
called on a vector type, it modifies FreeRegs to model GPR consumption.
However, this should not happen during the vector call first pass.
I refactored the code to unify vectorcall HVA logic with regcall HVA
logic. The conventions pass HVAs in registers differently (expanded vs.
not expanded), but if they do not fit in registers, they both pass them
indirectly by address.
Reviewers: erichkeane, craig.topper
Subscribers: cfe-commits
Tags: #clang
Differential Revision: https://reviews.llvm.org/D72110
