Is asm code selection optimal?
tg at gmplib.org
Wed May 17 01:01:05 UTC 2017
tg at gmplib.org (Torbjörn Granlund) writes:
And in the case you quote I'd assume the default is actually the C code,
i.e., that the config did not pick up any asm.
Alas, that's not the case.
I don't know what's going on. For a normalised divisor, timings match
I fixed some problems with the tables, the result should start
apppearing with next set of runs.
The C code you suggested is still missing.
There will be some duplicates related too include_mpn. Suppressing
duplicates isn't as easy as one might think as just excluding
include_mpn files misses many things we do want to measure.
Please encrypt, key id 0xC8601622
More information about the gmp-devel