ef1313deaf
Add a VMX optimised xor, used primarily for RAID5. On a POWER7 blade this is a decent win: 32regs : 17932.800 MB/sec altivec : 19724.800 MB/sec The bigger gain is when the same test is run in SMT4 mode, as it would if there was a lot of work going on: 8regs : 8377.600 MB/sec altivec : 15801.600 MB/sec I tested this against an array created without the patch, and also verified it worked as expected on a little endian kernel. [ Fix !CONFIG_ALTIVEC build -- BenH ] Signed-off-by: Anton Blanchard <anton@samba.org> Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org> |
||
|---|---|---|
| .. | ||
| alloc.c | ||
| checksum_32.S | ||
| checksum_64.S | ||
| checksum_wrappers_64.c | ||
| code-patching.c | ||
| copy_32.S | ||
| copypage_64.S | ||
| copypage_power7.S | ||
| copyuser_64.S | ||
| copyuser_power7.S | ||
| crtsavres.S | ||
| devres.c | ||
| div64.S | ||
| feature-fixups-test.S | ||
| feature-fixups.c | ||
| hweight_64.S | ||
| ldstfp.S | ||
| locks.c | ||
| Makefile | ||
| mem_64.S | ||
| memcpy_64.S | ||
| memcpy_power7.S | ||
| rheap.c | ||
| sstep.c | ||
| string.S | ||
| string_64.S | ||
| usercopy_64.c | ||
| vmx-helper.c | ||
| xor_vmx.c | ||