llvm/test/Transforms/LoopVectorize, branch stablehttp://llvm.orghttps://git.amat.us/llvm/atom/test/Transforms/LoopVectorize?h=stable2013-04-01T10:23:49ZX86TTI: Add accurate costs for itofp operations, based on the actual instruction counts.2013-04-01T10:23:49ZBenjamin Kramerbenny.kra@googlemail.com2013-04-01T10:23:49Zurn:sha1:13497b3aa7589fc4f9e924f850a7e5151e9ddd2f
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@178459 91177308-0d34-0410-b5e6-96231b3b80d8
LoopVectorizer: Insert some white space to make test case more readable2013-03-14T21:31:09ZArnold Schwaighoferaschwaighofer@apple.com2013-03-14T21:31:09Zurn:sha1:c184a5f4cad9415ee93ff6d24da2ed676573b61c
Also remove some unneeded function attributes.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@177114 91177308-0d34-0410-b5e6-96231b3b80d8
Add missing asserts flag to test - it uses debug flags2013-03-14T19:01:58ZArnold Schwaighoferaschwaighofer@apple.com2013-03-14T19:01:58Zurn:sha1:e2188d9c43d9e81407d2c7b7ef887e27f028a8ff
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@177102 91177308-0d34-0410-b5e6-96231b3b80d8
LoopVectorize: Invert case when we use a vector cmp value to query select cost2013-03-14T18:54:36ZArnold Schwaighoferaschwaighofer@apple.com2013-03-14T18:54:36Zurn:sha1:d517da33b712b5d8d687ee5e5974056a4787ec4f
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
Test case hygiene.2013-03-09T18:25:40ZBenjamin Kramerbenny.kra@googlemail.com2013-03-09T18:25:40Zurn:sha1:1cb47b9afe77231d6b87b8445dfec475fb5e59cd
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@176772 91177308-0d34-0410-b5e6-96231b3b80d8
LoopVectorizer: Ignore dbg.value instructions2013-03-09T15:56:34ZArnold Schwaighoferaschwaighofer@apple.com2013-03-09T15:56:34Zurn:sha1:56ee544a3af2b019329f06422c00e8a3645b895c
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
Force cpu in test.2013-03-08T17:01:18ZBenjamin Kramerbenny.kra@googlemail.com2013-03-08T17:01:18Zurn:sha1:35a4a0ca510feb21aeffb27982b9610f65f55e61
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@176702 91177308-0d34-0410-b5e6-96231b3b80d8
Insert the reduction start value into the first bypass block to preserve domination.2013-03-08T16:58:37ZBenjamin Kramerbenny.kra@googlemail.com2013-03-08T16:58:37Zurn:sha1:f22d9cfa6d145ce26930804cc3ac54340f49c38b
Fixes PR15344.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@176701 91177308-0d34-0410-b5e6-96231b3b80d8
X86 cost model: Adjust cost for custom lowered vector multiplies2013-03-02T04:02:52ZArnold Schwaighoferaschwaighofer@apple.com2013-03-02T04:02:52Zurn:sha1:5f0d9dbdf48a9efe16bfadf88e5335f7b9a8ec3f
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
PR14448 - prevent the loop vectorizer from vectorizing the same loop twice.2013-03-02T01:33:49ZNadav Rotemnrotem@apple.com2013-03-02T01:33:49Zurn:sha1:5290baacb8ca4fb75d798e873a441cad11cbfb2c
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