We now no longer need alias analysis - the cases that alias analysis would
handle are now handled as accesses with a large dependence distance.
We can now vectorize loops with simple constant dependence distances.
for (i = 8; i < 256; ++i) {
a[i] = a[i+4] * a[i+8];
}
for (i = 8; i < 256; ++i) {
a[i] = a[i-4] * a[i-8];
}
We would be able to vectorize about 200 more loops (in many cases the cost model
instructs us no to) in the test suite now. Results on x86-64 are a wash.
I have seen one degradation in ammp. Interestingly, the function in which we
now vectorize a loop is never executed so we probably see some instruction
cache effects. There is a 2% improvement in h264ref. There is one or the other
TSCV loop kernel that speeds up.
radar://13681598
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@184685 91177308-0d34-0410-b5e6-96231b3b80d8
vectorizing loops with memory accesses to non-zero address spaces. It
simply dropped the AS info. Fixes PR16306.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@184103 91177308-0d34-0410-b5e6-96231b3b80d8
We check that instructions in the loop don't have outside users (except if
they are reduction values). Unfortunately, we skipped this check for
if-convertable PHIs.
Fixes PR16184.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@183035 91177308-0d34-0410-b5e6-96231b3b80d8
- llvm.loop.parallel metadata has been renamed to llvm.loop to be more generic
by making the root of additional loop metadata.
- Loop::isAnnotatedParallel now looks for llvm.loop and associated
llvm.mem.parallel_loop_access
- document llvm.loop and update llvm.mem.parallel_loop_access
- add support for llvm.vectorizer.width and llvm.vectorizer.unroll
- document llvm.vectorizer.* metadata
- add utility class LoopVectorizerHints for getting/setting loop metadata
- use llvm.vectorizer.width=1 to indicate already vectorized instead of
already_vectorized
- update existing tests that used llvm.loop.parallel and
llvm.vectorizer.already_vectorized
Reviewed by: Nadav Rotem
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@182802 91177308-0d34-0410-b5e6-96231b3b80d8
The Value pointers we store in the induction variable list can be RAUW'ed by a
call to SCEVExpander::expandCodeFor, use a TrackingVH instead. Do the same thing
in some other places where we store pointers that could potentially be RAUW'ed.
Fixes PR16073.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@182485 91177308-0d34-0410-b5e6-96231b3b80d8
InstCombine can be uncooperative to vectorization and sink loads into
conditional blocks. This prevents vectorization.
Undo this optimization if there are unconditional memory accesses to the same
addresses in the loop.
radar://13815763
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@181860 91177308-0d34-0410-b5e6-96231b3b80d8
We used to give up if we saw two integer inductions. After this patch, we base
further induction variables on the chosen one like we do in the reverse
induction and pointer induction case.
Fixes PR15720.
radar://13851975
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@181746 91177308-0d34-0410-b5e6-96231b3b80d8
Use the widest induction type encountered for the cannonical induction variable.
We used to turn the following loop into an empty loop because we used i8 as
induction variable type and truncated 1024 to 0 as trip count.
int a[1024];
void fail() {
int reverse_induction = 1023;
unsigned char forward_induction = 0;
while ((reverse_induction) >= 0) {
forward_induction++;
a[reverse_induction] = forward_induction;
--reverse_induction;
}
}
radar://13862901
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@181667 91177308-0d34-0410-b5e6-96231b3b80d8
A computable loop exit count does not imply the presence of an induction
variable. Scalar evolution can return a value for an infinite loop.
Fixes PR15926.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@181495 91177308-0d34-0410-b5e6-96231b3b80d8
The two nested loops were confusing and also conservative in identifying
reduction variables. This patch replaces them by a worklist based approach.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@181369 91177308-0d34-0410-b5e6-96231b3b80d8
We were passing an i32 to ConstantInt::get where an i64 was needed and we must
also pass the sign if we pass negatives numbers. The start index passed to
getConsecutiveVector must also be signed.
Should fix PR15882.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@181286 91177308-0d34-0410-b5e6-96231b3b80d8
Add support for min/max reductions when "no-nans-float-math" is enabled. This
allows us to assume we have ordered floating point math and treat ordered and
unordered predicates equally.
radar://13723044
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@181144 91177308-0d34-0410-b5e6-96231b3b80d8
We can just use the initial element that feeds the reduction.
max(max(x, y), z) == max(max(x,y), max(x,z))
radar://13723044
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@181141 91177308-0d34-0410-b5e6-96231b3b80d8
By supporting the vectorization of PHINodes with more than two incoming values we can increase the complexity of nested if statements.
We can now vectorize this loop:
int foo(int *A, int *B, int n) {
for (int i=0; i < n; i++) {
int x = 9;
if (A[i] > B[i]) {
if (A[i] > 19) {
x = 3;
} else if (B[i] < 4 ) {
x = 4;
} else {
x = 5;
}
}
A[i] = x;
}
}
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@181037 91177308-0d34-0410-b5e6-96231b3b80d8
This patch disables memory-instruction vectorization for types that need padding
bytes, e.g., x86_fp80 has 10 bytes store size with 6 bytes padding in darwin on
x86_64. Because the load/store vectorization is performed by the bit casting to
a packed vector, which has incompatible memory layout due to the lack of padding
bytes, the present vectorizer produces inconsistent result for memory
instructions of those types.
This patch checks an equality of the AllocSize of a scalar type and allocated
size for each vector element, to ensure that there is no padding bytes and the
array can be read/written using vector operations.
Patch by Daisuke Takahashi!
Fixes PR15758.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@180196 91177308-0d34-0410-b5e6-96231b3b80d8
even if erroneously annotated with the parallel loop metadata.
Fixes Bug 15794:
"Loop Vectorizer: Crashes with the use of llvm.loop.parallel metadata"
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@180081 91177308-0d34-0410-b5e6-96231b3b80d8
A min/max operation is represented by a select(cmp(lt/le/gt/ge, X, Y), X, Y)
sequence in LLVM. If we see such a sequence we can treat it just as any other
commutative binary instruction and reduce it.
This appears to help bzip2 by about 1.5% on an imac12,2.
radar://12960601
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@179773 91177308-0d34-0410-b5e6-96231b3b80d8
Pass down the fact that an operand is going to be a vector of constants.
This should bring the performance of MultiSource/Benchmarks/PAQ8p/paq8p on x86
back. It had degraded to scalar performance due to my pervious shift cost change
that made all shifts expensive on x86.
radar://13576547
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@178809 91177308-0d34-0410-b5e6-96231b3b80d8
We generate a select with a vectorized condition argument when the condition is
NOT loop invariant. Not the other way around.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@177098 91177308-0d34-0410-b5e6-96231b3b80d8
We want vectorization to happen at -g. Ignore calls to the dbg.value intrinsic
and don't transfer them to the vectorized code.
radar://13378964
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@176768 91177308-0d34-0410-b5e6-96231b3b80d8
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
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@176403 91177308-0d34-0410-b5e6-96231b3b80d8
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.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@176399 91177308-0d34-0410-b5e6-96231b3b80d8
This properly asks TargetLibraryInfo if a call is available and if it is, it
can be translated into the corresponding LLVM builtin. We don't vectorize sqrt()
yet because I'm not sure about the semantics for negative numbers. The other
intrinsic should be exact equivalents to the libm functions.
Differential Revision: http://llvm-reviews.chandlerc.com/D465
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@176188 91177308-0d34-0410-b5e6-96231b3b80d8
Storing the load/store instructions with the values
and inspect them using Alias Analysis to make sure
they don't alias, since the GEP pointer operand doesn't
take the offset into account.
Trying hard to not add any extra cost to loads and stores
that don't overlap on global values, AA is *only* calculated
if all of the previous attempts failed.
Using biggest vector register size as the stride for the
vectorization access, as we're being conservative and
the cost model (which calculates the real vectorization
factor) is only run after the legalization phase.
We might re-think this relationship in the future, but
for now, I'd rather be safe than sorry.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@175818 91177308-0d34-0410-b5e6-96231b3b80d8
In the loop vectorizer cost model, we used to ignore stores/loads of a pointer
type when computing the widest type within a loop. This meant that if we had
only stores/loads of pointers in a loop we would return a widest type of 8bits
(instead of 32 or 64 bit) and therefore a vector factor that was too big.
Now, if we see a consecutive store/load of pointers we use the size of a pointer
(from data layout).
This problem occured in SingleSource/Benchmarks/Shootout-C++/hash.cpp (reduced
test case is the first test in vector_ptr_load_store.ll).
radar://13139343
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@174377 91177308-0d34-0410-b5e6-96231b3b80d8
