As suggested on the bug, to help (but not completely....) stop folded instructions crossing the inline asm barriers used for llvm-mca analysis, we should recommend tagging with memory captures/attributes.
Differential Revision: https://reviews.llvm.org/D117788
If the bias is zero, we can remove it from the image instruction.
Also copy other image optimizations (l->lz, mip->nomip) to IR combines.
Differential Revision: https://reviews.llvm.org/D116042
As the codegen fix in D111754, the LOD bias needs to be converted to 16
bits. Fix this in the combine.
Differential Revision: https://reviews.llvm.org/D116038
Add documentation around:
* Removing JITDylib from the session
* Add support for custom program representation
Reviewed By: lhames
Differential Revision: https://reviews.llvm.org/D116476
- Avoid using <iterator> for std::end on a plain array (using <array> instead)
- Avoid using <algorithm> for std::min and std::equal (using alternate logic and std::strcmp instead)
As an hint to the impact of the cleanup, running
clang++ -E -Iinclude -I../llvm/include ../llvm/lib/Demangle/*.cpp -std=c++14 -fno-rtti -fno-exceptions | wc -l
before: 203965 lines
after: 169704 lines
This patch adds VPWidenIntOrFpInductionRecipe::isCanonical to check if
an induction recipe is canonical. The code is also updated to use it
instead of isCanonicalID.
Reviewed By: Ayal
Differential Revision: https://reviews.llvm.org/D117551
As an hint to the impact of the cleanup, running
clang++ -E -Iinclude -I../llvm/include ../llvm/lib/Demangle/*.cpp -std=c++14 -fno-rtti -fno-exceptions | wc -l
before: 208053 lines
after: 203965 lines
Peculiarly, the necessary code to handle pointers (including the
check for non-integral address spaces) is already in place,
because we were already allowing vectors of pointers here, just
not plain pointers.
In some cases, fusion can produce illegal operations if after fusion
the range of some of the loops cannot be computed from shapes of its
operands. Check for this case and abort the fusion if this happens.
Differential Revision: https://reviews.llvm.org/D117602
`CppHashInfo.Filename` is a `StringRef` that references a part of the
source file and it is not null-terminated at the end of the file name.
`AsmParser::parseAndMatchAndEmitTargetInstruction()` passes it to
`getStreamer().emitDwarfFileDirective()`, and it eventually comes to
`isRootFile()`. The comparison fails because `FileName.data()` is not
properly terminated.
In addition, the old code might cause a significant speed degradation
for long source files. The `operator!=()` for `std::string` and
`const char *` can be implemented in a way that it finds the length of
the second argument first, which slows the comparison for long data.
`parseAndMatchAndEmitTargetInstruction()` calls
`emitDwarfFileDirective()` every time if `CppHashInfo.Filename` is not
empty. As a result, the longer the source file is, the slower the
compilation wend, and for a very long file, it might take hours instead
of a couple of seconds normally.
Differential Revision: https://reviews.llvm.org/D117785
After D86836, we can define multiple cost values for
different cost models. So here we set CostPerUse to
1 iff RVC is enabled to avoid potential impact on RA.
Reviewed By: craig.topper
Differential Revision: https://reviews.llvm.org/D117741
There are static and dynamic TLS address lowering in DAG stage according to different TLS model.
It needs PseudoTLSLA32 pseudo to get address of TLS-related entry which resides in constant pool.
This is superseded by the same method on OpAsmOpInterface, which is
available on the Dialect through the Fallback mechanism,
Reviewed By: rriddle
Differential Revision: https://reviews.llvm.org/D117750
This extends dense attribute element access to support 8b and 16b ints.
Also extends the corresponding parts of the C api.
Reviewed By: ftynse
Differential Revision: https://reviews.llvm.org/D117731
This allows to pipe sequences of `mlir-opt -split-input-file | mlir-opt -split-input-file`.
Depends On D117750
Reviewed By: rriddle
Differential Revision: https://reviews.llvm.org/D117756
Zbkb supports some encodings of the general grevi, shfli, and
unshfli instructions legal, so we added separate instructions for
those encodings to improve the diagnostics for assembler and
disassembler. To be consistent we should always use these separate
instructions whenever those specific encodings of grevi/shfli/unshfli
occur. So this patch adds specific isel patterns to override the generic
isel patterns for these cases. Similar was done for rev8 and zext.h
for Zbb previously.
This commit add instructions supports of `zbkb` which defined in scalar cryptography extension version v1.0.0 (has been ratified already).
Most of the zbkb directives reuse parts of the zbp and zbb directives, so this patch just modified some of the inst aliases and predicates.
Reviewed By: craig.topper
Differential Revision: https://reviews.llvm.org/D117640
Intel's CET/IBT requires every indirect branch target to be an ENDBR instruction. Because of that, the compiler needs to correctly emit these instruction on function's prologues. Because this is a security feature, it is desirable that only actual indirect-branch-targeted functions are emitted with ENDBRs. While it is possible to identify address-taken functions through LTO, minimizing these ENDBR instructions remains a hard task for user-space binaries because exported functions may end being reachable through PLT entries, that will use an indirect branch for such. Because this cannot be determined during compilation-time, the compiler currently emits ENDBRs to every non-local-linkage function.
Despite the challenge presented for user-space, the kernel landscape is different as no PLTs are used. With the intent of providing the most fit ENDBR emission for the kernel, kernel developers proposed an optimization named "ibt-seal" which replaces the ENDBRs for NOPs directly in the binary. The discussion of this feature can be seen in [1].
This diff brings the enablement of the flag -mibt-seal, which in combination with LTO enforces a different policy for ENDBR placement in when the code-model is set to "kernel". In this scenario, the compiler will only emit ENDBRs to address taken functions, ignoring non-address taken functions that are don't have local linkage.
A comparison between an LTO-compiled kernel binaries without and with the -mibt-seal feature enabled shows that when -mibt-seal was used, the number of ENDBRs in the vmlinux.o binary patched by objtool decreased from 44383 to 33192, and that the number of superfluous ENDBR instructions nopped-out decreased from 11730 to 540.
The 540 missed superfluous ENDBRs need to be investigated further, but hypotheses are: assembly code not being taken care of by the compiler, kernel exported symbols mechanisms creating bogus address taken situations or even these being removed due to other binary optimizations like kernel's static_calls. For now, I assume that the large drop in the number of ENDBR instructions already justifies the feature being merged.
[1] - https://lkml.org/lkml/2021/11/22/591
Reviewed By: xiangzhangllvm
Differential Revision: https://reviews.llvm.org/D116070
If `CMAKE_INSTALL_BINDIR` is a different absolute path per project, as
it is with NixOS when we install every package to its own prefix, the
old way fails when the absolute path gets prepended.
There are still some issues with dowstream packages using `LLVM_TOOLS_INSTALL_DIR` which also may be absolute and just for LLVM proper, but that will be addressed in a future commit.
Differential Revision: https://reviews.llvm.org/D101070
This patch changes the visibility for all construct in the new device
RTL to be hidden by default. This is done after the changes introduced
in D117806 changed the visibility from being hidden by default for all
device compilations. This asserts that the visibility for the device
runtime library will be hidden except for the internal environment
variable. This is done to aid optimization and linking of the device
library.
Reviewed By: JonChesterfield
Differential Revision: https://reviews.llvm.org/D117807
This patch changes the special-case handling of visibility when
compiling for an OpenMP target offloading device. This was orignally
added as a precaution against the bug encountered in PR41826 when
symbols in the device were being preempted by shared library symbols.
This should instead be done by making the visibility protected by default.
With protected visibility we are asserting that the symbols on the device
will never be preempted or preempt another symbol pending a shared library
load.
Reviewed By: JonChesterfield
Differential Revision: https://reviews.llvm.org/D117806
Instrument the SB API with signposts on Darwin. This gives us a time
profile on whose behalf LLDB spends time (particularly when run via the
SBAPI from an IDE).
Differential revision: https://reviews.llvm.org/D117632
Remove the last remaining references to the reproducers from the
instrumentation. This patch renames the relevant files and macros.
Differential revision: https://reviews.llvm.org/D117712
I revived lldb-instr to update the macros for D117712. I think the new
macros are simple enough that we add them by hand, but this tool can do
it automatically for you.
Differential revision: https://reviews.llvm.org/D117748
Originally, hasRVVFrameObject() will scan all the stack objects to check
whether if there is any scalable vector object on the stack or not.
However, it causes errors in the register allocator. In issue 53016, it
returns false before RA because there is no RVV stack objects. After RA,
it returns true because there are spilling slots for RVV values during RA.
The compiler will not reserve BP during register allocation and generate BP
access in the PEI pass due to the inconsistent behavior of the function.
The function is changed to use hasStdExtV() as the return value. It is
not precise, but it can make the register allocation correct.
Refer to https://github.com/llvm/llvm-project/issues/53016.
Differential Revision: https://reviews.llvm.org/D117663
Implements a near-universal extension in which NAMELIST
input will skip over unrelated namelist groups in the
input stream until the group with the requested name appears.
Differential Revision: https://reviews.llvm.org/D117843
Array subscripts and substring limits in NAMELIST input are
allowed to bear an explicit plus sign.
Differential Revision: https://reviews.llvm.org/D117818
This is needed to properly limit fractional LMULs for Zve32.
Add new RUN Zve32 RUN lines to the existing tests for the
-riscv-v-fixed-length-vector-elen-max command line option.
Nikita Popov