These two nodes were added by 69caef2b781130a7d0eeaf8898eb346b6423ae03 in 2005
and they are not used by PowerPC backend anymore. And the ISD::FMA is a prefer
way for VMADDFP if we really want to create that node. For VNMSUBFP, we will
also add a more generic node FNMSUB in D76585 if we really want it.
Reviewed By: qiucf
Differential Revision: https://reviews.llvm.org/D80429
Use getVectorElementCount() instead of getVectorNumElements().
The code changed in this patch is covered by an existing test:
CodeGen/AArch64/sve-intrinsics-contiguous-prefetches.ll
Differential Revision: https://reviews.llvm.org/D80615
New functions `lockFile`, `tryLockFile` and `unlockFile` implement
simple file locking. They lock or unlock entire file. This must be
enough to support simulataneous writes to log files in parallel builds.
Differential Revision: https://reviews.llvm.org/D78896
Explicitly set the exec mask for SGPR spills and reloads.
This fixes a bug where SGPR spills to memory could be incorrect
if the exec mask was 0 (or differed between spill and reload).
Additionally pack scalar subregisters (upto 16/32 per VGPR),
so that the majority of scalar types can be spilt or reloaded
with a simple memory access. This should amortize some of the
additional overhead of manipulating the exec mask.
Reviewed By: arsenm
Differential Revision: https://reviews.llvm.org/D80282
Port the code to recognize a zip1/zip2 shuffle mask from AArch64ISelLowering
and put it into the post-legalizer combiner.
Add G_ZIP1 and G_ZIP2 to AArch64InstrGISel.td and hook them up as equivalent
nodes to AArch64zip1 and AArch64zip2. This allows us to select them.
Minor code size improvements for SPECINT2000 at -O3 on 197.parser, 252.eon, and
186.crafty.
Differential Revision: https://reviews.llvm.org/D80969
Summary:
This patch simplifies FindMostPopularDest without changing the
functionality.
Given a list of jump threading destinations, the function finds the
most popular destination. To ensure determinism when there are
multiple destinations with the highest popularity, the function picks
the first one in the successor list with the highest popularity.
Without this patch:
- The function populates DestPopularity -- a histogram mapping
destinations to their respective occurrence counts.
- Then we iterate over DestPopularity, looking for the highest
popularity while building a vector of destinations with the highest
popularity.
- Finally, we iterate the successor list, looking for the destination
with the highest popularity.
With this patch:
- We implement DestPopularity with MapVector instead of DenseMap. We
populate the map with popularity 0 for all successors in the order
they appear in the successor list.
- We build the histogram in the same way as before.
- We simply use std::max_element on DestPopularity to find the most
popular destination. The use of MapVector ensures determinism.
Reviewers: wmi, efriedma
Reviewed By: wmi
Subscribers: hiraditya, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D81030
When sampleFDO is enabled, people may expect they can use
-fno-profile-sample-use to opt-out using sample profile for a certain file.
That could be either for debugging purpose or for performance tuning purpose.
However, when thinlto is enabled, if a function in file A compiled with
-fno-profile-sample-use is imported to another file B compiled with
-fprofile-sample-use, the inlined copy of the function in file B may still
get its profile annotated.
The inconsistency may even introduce profile unused warning because if the
target is not compiled with explicit debug information flag, the function
in file A won't have its debug information enabled (debug information will
be enabled implicitly only when -fprofile-sample-use is used). After it is
imported into file B which is compiled with -fprofile-sample-use, profile
annotation for the outline copy of the function will fail because the
function has no debug information, and that will trigger profile unused
warning.
We add a new attribute use-sample-profile to control whether a function
will use its sample profile no matter for its outline or inline copies.
That will make the behavior of -fno-profile-sample-use consistent.
Differential Revision: https://reviews.llvm.org/D79959
This lets us to remove !stack-safe metadata and
better controll when to perform StackSafety
analysis.
Reviewers: eugenis
Subscribers: hiraditya, steven_wu, dexonsmith, cfe-commits, llvm-commits
Tags: #clang, #llvm
Differential Revision: https://reviews.llvm.org/D80771
Summary:
An upgrade of LLVM for CrOS [0] containing [1] triggered a bunch of
errors related to writing to reserved registers for a Linux kernel's
arm64 compat vdso (which is a aarch32 image).
After a discussion on LKML [2], it was determined that
-f{no-}omit-frame-pointer was not being specified. Comparing GCC and
Clang [3], it becomes apparent that GCC defaults to omitting the frame
pointer implicitly when optimizations are enabled, and Clang does not.
ie. setting -O1 (or above) implies -fomit-frame-pointer. Clang was
defaulting to -fno-omit-frame-pointer implicitly unless -fomit-frame-pointer
was set explicitly.
Why this becomes a problem is that the Linux kernel's arm64 compat vdso
contains code that uses r7. r7 is used sometimes for the frame pointer
(for example, when targeting thumb (-mthumb)). See useR7AsFramePointer()
in llvm/llvm-project/llvm/lib/Target/ARM/ARMSubtarget.h. This is mostly
for legacy/compatibility reasons, and the 2019 Q4 revision of the ARM
AAPCS looks to standardize r11 as the frame pointer for aarch32, though
this is not yet implemented in LLVM.
Users that are reliant on the implicit value if unspecified when
optimizations are enabled should explicitly choose -fomit-frame-pointer
(new behavior) or -fno-omit-frame-pointer (old behavior).
[0] https://bugs.chromium.org/p/chromium/issues/detail?id=1084372
[1] https://reviews.llvm.org/D76848
[2] https://lore.kernel.org/lkml/20200526173117.155339-1-ndesaulniers@google.com/
[3] https://godbolt.org/z/0oY39t
Reviewers: kristof.beyls, psmith, danalbert, srhines, MaskRay, ostannard, efriedma
Reviewed By: psmith, danalbert, srhines, MaskRay, efriedma
Subscribers: efriedma, olista01, MaskRay, vhscampos, cfe-commits, llvm-commits, manojgupta, llozano, glider, hctim, eugenis, pcc, peter.smith, srhines
Tags: #clang, #llvm
Differential Revision: https://reviews.llvm.org/D80828
-Fix one place where we had a X86vzload64 but should have had
X86vzload32.
-Make sure all patterns that have scalar_to_vector+loadi64 also
have scalar_to_vector+f64 to match 32-bit codegen.
-Add some bitcasts that were missing from patterns.
-Make sure that if we have a scalar_to_vector+load pattern
we also have a vzload pattern.
We probably need some better canonicalization to avoid having
so many patterns.
This patch implements a target independent DAG combine to produce multiply-high
instructions from shifts. This DAG combine will combine shifts for any type as
long as the MULH on the narrow type is legal.
For now, it is enabled on PowerPC as PowerPC is the only target that has an
implementation of the isMulhCheaperThanMulShift TLI hook introduced in
D78271.
Moreover, this DAG combine focuses on catching the pattern:
(shift (mul (ext <narrow_type>:$a to <wide_type>), (ext <narrow_type>:$b to <wide_type>)), <narrow_width>)
to produce mulhs when we have a sign-extend, and mulhu when we have
a zero-extend.
The patch performs the following checks:
- Operation is a right shift arithmetic (sra) or logical (srl)
- Input to the shift is a multiply
- Both operands to the shift are sext/zext nodes
- The extends into the multiply are both the same
- The narrow type is half the width of the wide type
- The shift amount is the width of the narrow type
- The respective mulh operation is legal
Differential Revision: https://reviews.llvm.org/D78272
Summary:
Jump tables for most targets cannot handle out of range indices by
themselves, so LLVM emits range checks to guard the jump
tables. WebAssembly, on the other hand, implements jump tables using
the br_table instruction, which takes a default branch target as an
operand, making the range checks redundant. This patch introduces a
new MachineFunction pass in the WebAssembly backend to find and
eliminate the redundant range checks.
Reviewers: aheejin, dschuff
Subscribers: mgorny, sbc100, jgravelle-google, hiraditya, sunfish, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D80863
'git push' command, without any other arguments, can do different
things depending on the local configuration of Git. This patch
updates the 'git push' command with extra arguments to be more
resilient to any local configuration.
Reviewed By: mehdi_amini
Differential Revision: https://reviews.llvm.org/D79964
Some of the --debug-* options can take an optional offset. Although the
man page does a good job of making that clear, it's much harder to
discover from the help output.
Currently the only reference to this is the following sentence:
> Where applicable these parameters take an optional =<offset> argument
> to dump only the entry at the specified offset.
This patch changes the help output from to print [=<offset>] after the
options that take an offset.
--debug-info[=<offset>] - Dump the .debug_info section
rdar://problem/63150066
Differential revision: https://reviews.llvm.org/D80959
These are scalar instructions that change vector instructions, so they
should not be executed without any active lanes.
The implementation of -amdgpu-skip-threshold also seem to be backwards
from expected, since decreasing it prevents removal.
Summary:
Added initial codegen for 'affinity' clauses on task directives.
Emits next code:
```
kmp_task_affinity_info_t affs[<num_elems>];
void *td = __kmpc_task_alloc(..);
affs[<i>].base = &data_i;
affs[<i>].size = sizeof(data_i);
__kmpc_omp_reg_task_with_affinity(&loc, <gtid>, td, <num_elems>, affs);
```
The result returned by the call of `__kmpc_omp_reg_task_with_affinity`
function is ignored currently sincethe runtime currently ignores args
and returns 0 uncoditionally.
Reviewers: jdoerfert
Subscribers: yaxunl, guansong, sstefan1, llvm-commits, cfe-commits, caomhin
Tags: #clang, #llvm
Differential Revision: https://reviews.llvm.org/D80240
This teaches yaml2obj to allocate file space for a no-bits section
when there is a non-nobits section in the same segment that follows it.
It was discussed in D78005 thread and matches GNU linkers and LLD behavior.
Differential revision: https://reviews.llvm.org/D80629
Instead of using a fake call and metadata to temporarily represent a probed
static alloca, use a pseudo instruction.
This is inspired by the SystemZ approach proposed in https://reviews.llvm.org/D78717.
Differential Revision: https://reviews.llvm.org/D80641
The pass which infers when it's legal to load a global address space
as SMRD was only considering amdgpu_kernel, and ignoring the shader
entry type calling conventions.
The collectCallSiteParameters() method searches for instructions
which load values into registers used for parameters passing.
Previously, interpretation of those values, loaded by one such
instruction, was implemented inside collectCallSiteParameters() method.
This patch moves the interpretation code from collectCallSiteParameters()
method into a separate static method named interpretValue. New method is
called from collectCallSiteParameters() to process each instruction from
targeted instruction scope.
The collectCallSiteParameters() searches for loaded parameter value
among instructions which precede the call instruction, inside the same
basic block. When needed, new method (interpretValue) could be used for
searching any instruction scope.
This is preparation for search of parameter value, loaded inside call
delay slot.
Patch by Nikola Tesic
Differential revision: https://reviews.llvm.org/D78106
