This is the mail archive of the
libc-alpha@sourceware.org
mailing list for the glibc project.
Re: [PATCH] Improve string benchtest timing
Siddhesh,
On 22.5.2019 14:27 , Siddhesh Poyarekar wrote:
On 22/05/19 4:41 PM, Wilco Dijkstra wrote:
Well if I force the copies to be mutually unaligned, there is only about 1% difference
for a few of the memcpy implementations compared to them being always aligned
The others show identical performance whether aligned or not. This is not too
surprising since the test is basically waiting for DRAM most of the time.
That's a good point. Is that the case for thunderx as well? IIRC they
perform particularly badly with misaligned code but I don't know if
they're bad enough to be significant in the face of DRAM waits.
They were. The difference was up to 50% in some cases. But this is the
data for bench-memcpy. Reports for bench-memcpy-walk and
bench-memcpy-random do not have alignment info in them.
So, unaligned accesses can be much slower that aligned ones. Or, at least,
this is how the benchmarks measure them. We already know that the results
are not very stable, though.
P.S. To be precise I'm speaking about TX2 here.
--
Thanks,
Anton