mh, yet another comment on zerocopy. I wish there was a reply to my question on the usb-multiple-interface issue ;)
On Sun, May 01, 2005 at 10:25:15PM +0200, Stephane Fillod wrote: > "Premature optimization is the root of all evil", Donald Knuth. Mh, that's my common mistake. I always over-optimize in the beginning, and software takes ages until it's actually released. Still I can't help. > Before implementing, verify that this is the biggest spot of > optimization. You can use great tools like OProfile for that purpose. It just hurts to see waste valuable cpu time, memory bandwidth and cache pollution in a case that at least appears to be very easily optimizied. I was probably spending man-months to see how to optimize kernel networking code, where we can't do all those neat optimizations that seem perfectly possible with USB. > Do you(or somebody else) know a more clever kernel/user interface > with zero-copy for the USB subsystem? there is none. That's what my point was about. Implementing mmap() on the filedescriptor in usbdevfs. > I guess now we're more in functional development, but it's always > open on who has spare time ;-) I don't have the time, but the problem seems interesting enough to find some time for it. -- - Harald Welte <[EMAIL PROTECTED]> http://gnumonks.org/ ============================================================================ "Privacy in residential applications is a desirable marketing option." (ETSI EN 300 175-7 Ch. A6)
pgpe87B5ElQ03.pgp
Description: PGP signature
_______________________________________________ Discuss-gnuradio mailing list Discuss-gnuradio@gnu.org http://lists.gnu.org/mailman/listinfo/discuss-gnuradio
