forked from OSchip/llvm-project
Brag about function call vectorization in the docs.
llvm-svn: 176292
This commit is contained in:
parent
d9916eaecc
commit
a87d5128dd
|
@ -245,6 +245,17 @@ See the table below for a list of these functions.
|
|||
| | | fmuladd |
|
||||
+-----+-----+---------+
|
||||
|
||||
The loop vectorizer knows about special instructions on the target and will
|
||||
vectorize a loop containing a function call that maps to the instructions. For
|
||||
example, the loop below will be vectorized on Intel x86 if the SSE4.1 roundps
|
||||
instruction is available.
|
||||
|
||||
.. code-block:: c++
|
||||
|
||||
void foo(float *f) {
|
||||
for (int i = 0; i != 1024; ++i)
|
||||
f[i] = floorf(f[i]);
|
||||
}
|
||||
|
||||
Partial unrolling during vectorization
|
||||
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
|
||||
|
|
Loading…
Reference in New Issue