* Only apply divide bypass optimization when not optimizing for size.
* Fixed bug caused by constant for 0 value of type Int32,
used dividend type to generate the constant instead.
* For atom x86-64 apply the divide bypass to use 16-bit divides instead of
64-bit divides when operand values are small enough.
* Added lit tests for 64-bit divide bypass.
Patch by Tyler Nowicki!
llvm-svn: 176442
Instead of passing NULL on to sub-matcher, just return false. Updated tests and
regenerated docs.
Author: Tareq A Siraj <tareq.a.siraj@intel.com>
llvm-svn: 176441
LoopConvert isn't properly handling iterators whose dereference operator
returns by value. This test case demonstrates the failure.
See PR15437.
llvm-svn: 176437
With [] marking the selected range, clang-format invoked on
[ ] int a;
Would so far not reformat anything. With this patch, it formats a
line if its leading whitespace is touched.
llvm-svn: 176435
with both -static-libgcc and -static on the commandline.
Fix a warning in the latter case due to a backwards short circuiting ||
operator in the driver. No real functionality changed here, just allows
the driver to properly consume -static-libgcc when -static is also
specified.
llvm-svn: 176429
- Generate atomicrmw operations in most of the cases when it's sensible to do
so.
- Don't crash in several common cases (and hopefully don't crash in more of
them).
- Add some better tests.
We now generate significantly better code for things like:
_Atomic(int) x;
...
x++;
On MIPS, this now generates a 4-instruction ll/sc loop, where previously it
generated about 30 instructions in two nested loops. On x86-64, we generate a
single lock incl, instead of a lock cmpxchgl loop (one instruction instead of
ten).
llvm-svn: 176420
That is where the documentation gets built to.
I'm pretty much clueless about SVN; I would appreciate if someone more
knowledgeable about SVN made the equivalent change.
llvm-svn: 176417
The VDUP instruction source register doesn't allow a non-constant lane
index, so make sure we don't construct a ARM::VDUPLANE node asking it to
do so.
rdar://13328063
http://llvm.org/bugs/show_bug.cgi?id=13963
llvm-svn: 176413
This adds minimalistic support for PHI nodes to llvm.objectsize() evaluation
fingers crossed so that it does break clang boostrap again..
llvm-svn: 176408
in a core file if it didn't start at the beginning of a memory segment.
I added more sophisticated kernel location code to DynamicLoaderDarwinKernel
and removed the simple one in ProcessMachCore. Unfortunately the kernel
DynamicLoader doesn't get a chance to search around in memory unless there's
a hint that this might be a kernel debug session. It was easy ot make the
kernel location code static in DynamicLoaderDarwinKernel and call it from
ProcessMachCore on the start of the session, so that's what I did.
<rdar://problem/13326647>
llvm-svn: 176405
This matters for example in following matrix multiply:
int **mmult(int rows, int cols, int **m1, int **m2, int **m3) {
int i, j, k, val;
for (i=0; i<rows; i++) {
for (j=0; j<cols; j++) {
val = 0;
for (k=0; k<cols; k++) {
val += m1[i][k] * m2[k][j];
}
m3[i][j] = val;
}
}
return(m3);
}
Taken from the test-suite benchmark Shootout.
We estimate the cost of the multiply to be 2 while we generate 9 instructions
for it and end up being quite a bit slower than the scalar version (48% on my
machine).
Also, properly differentiate between avx1 and avx2. On avx-1 we still split the
vector into 2 128bits and handle the subvector muls like above with 9
instructions.
Only on avx-2 will we have a cost of 9 for v4i64.
I changed the test case in test/Transforms/LoopVectorize/X86/avx1.ll to use an
add instead of a mul because with a mul we now no longer vectorize. I did
verify that the mul would be indeed more expensive when vectorized with 3
kernels:
for (i ...)
r += a[i] * 3;
for (i ...)
m1[i] = m1[i] * 3; // This matches the test case in avx1.ll
and a matrix multiply.
In each case the vectorized version was considerably slower.
radar://13304919
llvm-svn: 176403
Inlining brought a few "null pointer use" false positives, which occur because
the callee defensively checks if a pointer is NULL, whereas the caller knows
that the pointer cannot be NULL in the context of the given call.
This is a first attempt to silence these warnings by tracking the symbolic value
along the execution path in the BugReporter. The new visitor finds the node
in which the symbol was first constrained to NULL. If the node belongs to
a function on the active stack, the warning is reported, otherwise, it is
suppressed.
There are several areas for follow up work, for example:
- How do we differentiate the cases where the first check is followed by
another one, which does happen on the active stack?
Also, this only silences a fraction of null pointer use warnings. For example, it
does not do anything for the cases where NULL was assigned inside a callee.
llvm-svn: 176402
The LoopVectorizer often runs multiple times on the same function due to inlining.
When this happens the loop vectorizer often vectorizes the same loops multiple times, increasing code size and adding unneeded branches.
With this patch, the vectorizer during vectorization puts metadata on scalar loops and marks them as 'already vectorized' so that it knows to ignore them when it sees them a second time.
PR14448.
llvm-svn: 176399
Fariborz Jahanian