Previously we eliminated the randomized scheme for finding memory when the
underlying process cannot allocate memory, and replaced it with an algorithm
that starts the allocations at 00x.
This was more determinstic, but runs into problems on embedded targets where the
pages near 0x0 are in fact interesting memory. To deal with those cases, this
patch does two things:
- It makes the default fallback be an address that is less likely than 0x0 to
contain interesting information.
- Before falling back to this, it adds an algorithm that consults the
GetMemoryRegionInfo() API to see if it can find an unmapped area.
This should eliminate the randomness (and unpredictable memory accesseas) of the
previous scheme while making expressions more likely to return correct results.
<rdar://problem/25545573>
llvm-svn: 272301
512-bit VPSLLDQ/VPSRLDQ can only be used for avx512bw targets so lowerVectorShuffleAsShift had to be adjusted to include the subtarget
llvm-svn: 272300
Summary: Clang changes to make use of the LLVM intrinsics added in D21160.
Reviewers: tra
Subscribers: jholewinski, cfe-commits
Differential Revision: http://reviews.llvm.org/D21162
llvm-svn: 272299
Summary:
Currently clang emits these instructions via inline (volatile) asm in
the CUDA headers. Switching to intrinsics will let the optimizer reason
across calls to these intrinsics.
Reviewers: tra
Subscribers: llvm-commits, jholewinski
Differential Revision: http://reviews.llvm.org/D21160
llvm-svn: 272298
These ExprWithCleanups are added for holding a RunCleanupsScope not
for destructor calls; rather, they are for lifetime marks. This requires
ExprWithCleanups to keep a bit to indicate whether it have cleanups with
side effects (e.g. dtor calls).
Differential Revision: http://reviews.llvm.org/D20498
llvm-svn: 272296
Summary:
__syncthreads, which corresponds to bar.sync 0, is already convergent.
This makes the more general bar.sync n likewise convergent.
Reviewers: tra
Subscribers: llvm-commits, jholewinski
Differential Revision: http://reviews.llvm.org/D21161
llvm-svn: 272295
Use os.devnull instead of tempfiles when substituting '/dev/null' on
Windows machines. This should make the bots just a bit speedier.
Thanks to Yunzhong Gao for testing this patch on Windows!
Differential Revision: http://reviews.llvm.org/D20549
llvm-svn: 272290
Refactored __kmp_execute_tasks_template to shorten and remove code redundancy.
The original code for __kmp_execute_tasks_template was very redundant with
large sections of repeated code that needed to be kept consistent, and goto
statements that made the control flow difficult to discern. This refactoring
removes all gotos and redundancy.
Patch by Terry Wilmarth
Differential Revision: http://reviews.llvm.org/D20879
llvm-svn: 272286
Previously, we materialized secondary vector IVs from the primary scalar IV,
by offseting the primary to match the correct start value, and then broadcasting
it - inside the loop body. Instead, we can use a real vector IV, like we do for
the primary.
This enables using vector IVs for secondary integer IVs whose type matches the
type of the primary.
Differential Revision: http://reviews.llvm.org/D20932
llvm-svn: 272283
We can now use this to decide whether to emit a verneed during the final
pass over the symbols. We were previously wrongly creating a verneed entry
in the case where all references to a DSO's symbols were weak.
In a future change we may also want to use the used bit to control whether
shared symbols are preemptible and appear in the dynsym. This seems a little
tricky to do at the moment because isNeeded() is templated.
The only other functional change here is that we emit a DT_NEEDED for DSOs
whose symbols are all preempted by objects that appear later in the link. But
that doesn't seem too important to me.
Differential Revision: http://reviews.llvm.org/D21171
llvm-svn: 272282
This reverts r272275. This actually wasn't the right way to fix the problem. The correct solution is in r272279.
Applying the fix to LLVM as done in r272279, means this fix will get picked up by all projects building out of tree using LLVM's CMake modules. As opposed to the fix I had in r272275, which would require each project to change.
llvm-svn: 272280
This is the more-correct fix to out-of-tree building. AddLLVM.cmake relies on this variable being set, so we should make sure it is set in LLVMConfig.cmake.
llvm-svn: 272279
In order to make this happen, I have added permissions to sections so that we can know what the permissions are for a given section, and modified both core file plug-ins to override Process::GetMemoryRegionInfo() and answer things correctly.
llvm-svn: 272276
The out-of-tree build needs to read LLVM_TOOLS_INSTALL_DIR out of TOOLS_BINARY_DIR because LLVM_TOOLS_INSTALL_DIR is used by AddLLVM.cmake
llvm-svn: 272275
MSVC doesn't allow std::atomic<>s in a union since they don't have trivial
copy constructor. Replacing them with e.g. std::atomic_int works, but that
breaks the GCC build on Linux, because then calls to e.g. std::atomic_load_explicit
fail, as they expect a real std::atomic<> pointer.
Fixing this with an #ifdef to unbreak the build for now.
llvm-svn: 272271
This reapplies commit r271930, r271915, r271923. They hit a bug in
Thumb which is fixed in r272258 now.
The original message:
The code layout that TailMerging (inside BranchFolding) works on is not the
final layout optimized based on the branch probability. Generally, after
BlockPlacement, many new merging opportunities emerge.
This patch calls Tail Merging after MBP and calls MBP again if Tail Merging
merges anything.
llvm-svn: 272267
This enables use of the 'S' constraint for inline ASM operands on
SystemZ, which allows for a memory reference with a signed 20-bit
immediate displacement. This patch includes corresponding documentation
and test case updates.
I've changed the 'T' constraint to match the new behavior for 'S', as
'T' also uses a long displacement (though index constraints are still
not implemented). I also changed 'm' to match the behavior for 'S' as
this will allow for a wider range of displacements for 'm', though
correct me if that's not the right decision.
Author: colpell
Differential Revision: http://reviews.llvm.org/D21097
llvm-svn: 272266
Summary:
we want to exclude friend declaration, but the `DeclContext` of a
friend function declaration is not the class in which it is declared, so we need
to explicitly check if the parent is a `friendDecl`.
Reviewers: bkramer
Subscribers: cfe-commits
Differential Revision: http://reviews.llvm.org/D21175
llvm-svn: 272261
ReplaceTailWithBranchTo assumed that if an instruction is predicated, it must be part of an IT block. This is not correct for conditional branches.
No testcase as this was triggered by the reverted patch r272017 - test coverage will occur when that patch is re-reverted and there is no known way to trigger this in the meantime.
llvm-svn: 272258
This is made possible by removing an assert in llc that assumed
MIRParser::parseLLVMModule would exit on error. MIRParser's documentation states
that it returns null if a parsing error occurs, so there's no reason to assert.
We can instead just fall through to where the check for a module is performed
and exit if it is null.
This commit is part of the clean-up after r269655.
Fixes PR27770
Differential Revision: http://reviews.llvm.org/D20371
llvm-svn: 272254
If an immediate is only used in an AND node, it is possible that the immediate can be more optimally materialized when negated. If this is the case, we can negate the immediate and use a BIC instead;
int i(int a) {
return a & 0xfffffeec;
}
Used to produce:
ldr r1, [CONSTPOOL]
ands r0, r1
CONSTPOOL: 0xfffffeec
And now produces:
movs r1, #255
adds r1, #20 ; Less costly immediate generation
bics r0, r1
llvm-svn: 272251
Rafael Espindola