posix_memalign performance regression in 2.38?

DJ Delorie dj@redhat.com
Mon Aug 7 20:07:37 GMT 2023


Noah Goldstein <goldstein.w.n@gmail.com> writes:
>> The old code was terrible wrt memory usage.
>
> How is the perf compared to malloc?

The old code *always* used at least 2X the memory as unaligned malloc,
and never reused free'd chunks, even if they were originally alloc'd
with a memalign.



More information about the Libc-alpha mailing list