| |||||||||||||
Glibc change exposing bugsGlibc change exposing bugsPosted Nov 11, 2010 5:01 UTC (Thu) by nicooo (guest, #69134)In reply to: Glibc change exposing bugs by jwb Parent article: Glibc change exposing bugs
This change is starting to sound more and more pointless.
(Log in to post comments)
Glibc change exposing bugs Posted Nov 11, 2010 6:54 UTC (Thu) by madscientist (subscriber, #16861) [Link]
Well, I have a core2 and I'd LOVE to get 4x speed improvement on memcpy(), which is probably the single most widely used function in all of C programming.
So it doesn't sound useless to me.
Glibc change exposing bugs Posted Nov 11, 2010 9:42 UTC (Thu) by dgm (subscriber, #49227) [Link]
Also, any improvement to Atom performance can only be welcome.
Glibc change exposing bugs Posted Nov 11, 2010 12:00 UTC (Thu) by alankila (subscriber, #47141) [Link]
I took a simple oprofile trace of me using my desktop for a while. Sad to say, but in that trace glibc's memcpy() used 0.24 % of time.
It would probably be better idea for majority of systems to just remove memcpy() and just replace it with memmove() which showed up with 0.17 %. Together, that would add up to 0.5 % at most, I suppose.
|
|||||||||||||