[PATCH v3] aarch64: thunderx2 memcpy branches reordering

Anton Youdkevitch anton.youdkevitch@bell-sw.com
Fri Mar 22 20:15:00 GMT 2019


The "ext" chunk changes are:

1. Always taken conditional branch at the beginning is
removed.
2. Epilogue code is placed after the end of the loop to
reduce the number of branches.
2. The redundant "mov" instruction inside the loop is
removed.
3. Invariant code in the loop epilogue is no more
repeated for each chunk.

make check showed no regressions.

-------------- next part --------------
A non-text attachment was scrubbed...
Name: memcpy-reorder-ext-v3.patch
Type: text/x-diff
Size: 2177 bytes
Desc: not available
URL: <http://sourceware.org/pipermail/libc-alpha/attachments/20190322/d1a08530/attachment.bin>


More information about the Libc-alpha mailing list