Changes from v2 to v3:
* Unit testing. This includes having x86 attempt all versions of
the accelerator that will run on the hardware. Thus an avx2 host
will run the basic test 5 times (1.5sec on my laptop).
* Drop the ppc and aarch64 specializations. I have improved the
basic integer version to the point that those vectorized versions
are not a win.
In the case of my aarch64 mustang, the integer version is 4 times
faster than the neon version that I delete. With effort I was
able to rewrite the neon version to come to within a factor of 1.1,
but it remained slower than the integer. To be fair, gcc6 makes
very good use of ldp, so the integer path is *also* loading 16 bytes
per insn.
I can forward my standalone aarch64 benchmark if anyone is interested.
Note however that at least the avx2 acceleration is still very much
a win, being about 3 times faster on my laptop. Of course, it's
handling 4 times as much data per loop as the integer version, so
one can still see the overhead caused by using vector insns.
For grins I wrote an avx512 version, if someone has a skylake upon
which to test and benchmark. That requires additional configure
checks, so I didn't bother to include it here.
r~
Richard Henderson (9):
cutils: Move buffer_is_zero and subroutines to a new file
cutils: Remove SPLAT macro
cutils: Export only buffer_is_zero
cutils: Rearrange buffer_is_zero acceleration
cutils: Add test for buffer_is_zero
cutils: Add generic prefetch
cutils: Rewrite x86 buffer zero checking
cutils: Remove aarch64 buffer zero checking
cutils: Remove ppc buffer zero checking
configure | 21 +--
include/qemu/cutils.h | 3 +-
migration/ram.c | 2 +-
migration/rdma.c | 5 +-
tests/Makefile.include | 3 +
tests/test-bufferiszero.c | 78 +++++++++++
util/Makefile.objs | 1 +
util/bufferiszero.c | 332 ++++++++++++++++++++++++++++++++++++++++++++++
util/cutils.c | 244 ----------------------------------
9 files changed, 423 insertions(+), 266 deletions(-)
create mode 100644 tests/test-bufferiszero.c
create mode 100644 util/bufferiszero.c
--
2.7.4
