6502/llvm-6502
1
0
Fork 0
mirror of https://github.com/c64scene-ar/llvm-6502.git synced 2025-05-18 22:38:56 +00:00
Code Issues Projects Releases Wiki Activity

Fix vectorizer docs.

Browse Source 
This example is not vectorized because LLVM does not prove no-wrapping of
"a[i*7] += ...".

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@203734 91177308-0d34-0410-b5e6-96231b3b80d8
This commit is contained in:
Arnold Schwaighofer 2014-03-12 23:23:44 +00:00
parent 9a4d525b7d
commit 6a7d263e8b
1 changed files with 6 additions and 3 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

9
docs/Vectorizers.rst
View File

@ -182,11 +182,14 @@ that scatter/gathers memory.
.. code-block:: c++ .. code-block:: c++
int foo(int *A, int *B, int n, int k) { int foo(int * A, int * B, int n) {
for (int i = 0; i < n; ++i) for (intptr_t i = 0; i < n; ++i)
A[i*7] += B[i*k]; A[i] += B[i*4];
} }
In many situations the cost model will inform LLVM that this is not beneficial
and LLVM will only vectorize such code if forced with "-mllvm -force-vector-width=#".
Vectorization of Mixed Types Vectorization of Mixed Types
^^^^^^^^^^^^^^^^^^^^^^^^^^^^ ^^^^^^^^^^^^^^^^^^^^^^^^^^^^