Using the "raw" Elf64_Dyn or Elf32_Dyn structures in
DynamicSection<ELFT>::writeTo does not correctly handle mixed-Endian
situations. Instead, use the corresponding llvm::object::* structures which
have Endian-converting members (like the rest of the code).
This fixes all currently-failing elf2 tests when running on big-Endian
PPC64/Linux (I've added a big-Endian test case which should fail on
little-Endian machines in the same way that test/elf2/shared.s failed on
big-Endian machines prior to this change).
llvm-svn: 249150
This patch implements the outlining for offloading functions for code
annotated with the OpenMP target directive. It uses a temporary naming
of the outlined functions that will have to be updated later on once
target side codegen and registration of offloading libraries is
implemented - the naming needs to be made unique in the produced
library.
llvm-svn: 249148
This patch teaches FastIsel the following two things:
1) On SSE2, no instructions are needed for bitcasts between 128-bit vector types;
2) On AVX, no instructions are needed for bitcasts between 256-bit vector types.
Example:
%1 = bitcast <4 x i31> %V to <2 x i64>
Before (-fast-isel -fast-isel-abort=1):
FastIsel miss: %1 = bitcast <4 x i31> %V to <2 x i64>
Now we don't fall back to SelectionDAG and we correctly fold that computation
propagating the register associated to %V.
Originally reviewed here: http://reviews.llvm.org/D13347
llvm-svn: 249147
Found while testing a FreeBSD base system build with lld. Ignored for
now while we continue to identify missing options and functionality.
llvm-svn: 249144
Reapply r248935.
Usually, when using LTO with a clang installation newer than the
system's one, there's a libLTO.dylib version mismatch and LTO fails. One
solution to this is to make ld point to the right libLTO.dylib by
changing DYLD_LIBRARY_PATH.
However, ld64 supports specifying the complete path to the desired
libLTO.dylib through the -lto_library option. This commit adds support
for the clang driver to use this option whenever it's capable of finding
a libLTO.dylib in clang's installed library directory. This way, we
don't need to rely on DYLD_LIBRARY_PATH nor get caught by version
mismatches.
Differential Revision: http://reviews.llvm.org/D13117
rdar://problem/7363476
llvm-svn: 249143
Currently FastISel doesn't know how to select vector bitcasts.
During instruction selection, fast-isel always falls back to SelectionDAG
every time it encounters a vector bitcast.
As a consequence of this, all the 'packed vector shift by immedate count'
test cases in avx2-builtins.c are optimized by the DAGCombiner.
In particular, the DAGCombiner would always fold trivial stack loads of
constant shift counts into the operands of packed shift builtins.
This behavior would start changing as soon as I reapply revision 249121.
That revision would teach x86 fast-isel how to select bitcasts between vector
types of the same size.
As a consequence of that change, fast-isel would less often fall back to
SelectionDAG. More importantly, DAGCombiner would no longer be able to
simplify the code by folding the stack reload of a constant.
No functional change.
llvm-svn: 249142
We have to skip accesses in non-affine subregions during hoisting as
they might not be executed under the same condition as the entry of
the non-affine subregion.
llvm-svn: 249139
Summary:
This new toolchain uses primarily LLVM-based tools, eg. compiler-rt, lld,
libcxx, etc. Because of this, it doesn't require neither an existing GCC
installation nor a GNU environment. Ideally, in a follow-up patch we
would like to add a new --{llvm|clang}-toolchain option (similar to
--gcc-toolchain) in order to allow the use of this toolchain with
independent Clang builds. For the time being, we use the --sysroot
option just to test the correctness of the paths generated by the
driver.
Reviewers: atanasyan, dsanders, rsmith
Subscribers: jfb, tberghammer, danalbert, srhines, dschuff, cfe-commits
Differential Revision: http://reviews.llvm.org/D13340
llvm-svn: 249137
This moves the construction of ScopStmt to the beginning of the
ScopInfo pass. The late creation was a result of the earlier separation
of ScopInfo and TempScopInfo. This will avoid introducing more
ScopStmt-like maps in future commits. The AccFuncMap will also be
removed in some future commit. DomainMap might also be included into
ScopStmt.
The order in which ScopStmt are created changes and initially creates
empty statements that are removed in a simplification.
Differential Revision: http://reviews.llvm.org/D13341
llvm-svn: 249132
their associated .cpp file.
Previous refactorings long long ago had split out the above categories
of classes from Stmt.h into their own header files, but failed to also
split the Stmt.cpp implementation file similarly. Do so for
readability's sake.
llvm-svn: 249131
If a value is globally mapped (IslNodeBuilder::ValueMap) and
referenced in the code that will be put into a subfunction, we hand
down the new value to the subfunction.
This patch also removes code that handed down all invariant loads to
the subfunction. Instead, only needed invariant loads are given to the
subfunction. There are two possible reasons for an invariant load to
be handed down:
1) The invariant load is used in a block that is placed in the
subfunction but which is not the parent of the load. In this
case, the scalar access that will read the loaded value, will
cause its base pointer (the preloaded value) to be handed down to
the subfunction.
2) The invariant load is defined and used in a block that is placed
in the subfunction. With this patch we will hand down the
preloaded value to the subfunction as the invariant load is
globally mapped to that value.
llvm-svn: 249126
This patch teaches FastIsel the following two things:
1) On SSE2, no instructions are needed for bitcasts between 128-bit vector types;
2) On AVX, no instructions are needed for bitcasts between 256-bit vector types.
Example:
%1 = bitcast <4 x i31> %V to <2 x i64>
Before (-fast-isel -fast-isel-abort=1):
FastIsel miss: %1 = bitcast <4 x i31> %V to <2 x i64>
Now we don't fall back to SelectionDAG and we correctly fold that computation
propagating the register associated to %V.
Differential Revision: http://reviews.llvm.org/D13347
llvm-svn: 249121
* Use .ARM.exidx as a fallback unwind plan for non-call site when the
instruction emulation based unwind failed.
* Work around an old compiler issue where the compiler isn't sort the
entries in .ARM.exidx based on their address.
* Fix unwind info parsing when the virtual file address >= 0x80000000
* Fix bug in unwind info parsing when neither lr nor pc is explicitly
restored.
Differential revision: http://reviews.llvm.org/D13380
llvm-svn: 249119
partial specialization can perform conversions on the argument. Be sure we
start again from the original argument when checking each possible template.
llvm-svn: 249114
When error blocks are not terminated by an unreachable they have successors
that might only be reachable via error blocks. Additionally, branches in
error blocks are not checked during SCoP detection, thus we might not be able
to handle them. With this patch we do not try to model error block exit
conditions. Anything that is only reachable via error blocks is ignored too,
as it will not be executed in the optimized version of the SCoP anyway.
llvm-svn: 249099
The user can provide function names with
-polly-error-functions=name1,name2,name3
that will be treated as error functions. Any call to them is assumed
not to be executed.
This feature is mainly for developers to play around with the new
"error block" feature.
llvm-svn: 249098
test that our intrinsics behave the same under -fsigned-char and
-funsigned-char.
This further testing uncovered that AVX-2 has a broken cmpgt for 8-bit
elements, and has for a long time. This is fixed in the same way as
SSE4 handles the case.
The other ISA extensions currently work correctly because they use
specific instruction intrinsics. As soon as they are rewritten in terms
of generic IR, they will need to add these special casts. I've added the
necessary testing to catch this however, so we shouldn't have to chase
it down again.
I considered changing the core typedef to be signed, but that seems like
a bad idea. Notably, it would be an ABI break if anyone is reaching into
the innards of the intrinsic headers and passing __v16qi on an API
boundary. I can't be completely confident that this wouldn't happen due
to a macro expanding in a lambda, etc., so it seems much better to leave
it alone. It also matches GCC's behavior exactly.
A fun side note is that for both GCC and Clang, -funsigned-char really
does change the semantics of __v16qi. To observe this, consider:
% cat x.cc
#include <smmintrin.h>
#include <iostream>
int main() {
__v16qi a = { 1, -1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0};
__v16qi b = _mm_set1_epi8(-1);
std::cout << (int)(a / b)[0] << ", " << (int)(a / b)[1] << '\n';
}
% clang++ -o x x.cc && ./x
-1, 1
% clang++ -funsigned-char -o x x.cc && ./x
0, 1
However, while this may be surprising, both Clang and GCC agree.
Differential Revision: http://reviews.llvm.org/D13324
llvm-svn: 249097
The ClangExpressionVariable::CreateVariableInList functions looked cute, but
caused more confusion than they solved. I removed them, and instead made sure
that there are adequate facilities for easily adding newly-constructed
ExpressionVariables to lists.
I also made some of the constructors that are common be generic, so that it's
possible to construct expression variables from generic places (like the ABI and
ValueObject) without having to know the specifics about the class.
llvm-svn: 249095
Yaron Keren