> Under Solaris 7, when the number of idle sockets was increased from > 100 to 10000, the time to check for active sockets with poll() > increased by a factor of only 6.5. That's a sublinear increase in time, > pretty spiffy. Under Solaris 7, when the number of idle sockets was decreased from 10,000 to 100, the time to check for active sockets with poll() decreased by a factor of only 6.5. Shouldn't it be more like 100? Sounds like Solaris' poll implementation is horribly broken for the most common case. DS - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] Please read the FAQ at http://www.tux.org/lkml/
- Re: Linux's implementatio... Gideon Glass
- Re: Linux's implementatio... Simon Kirby
- Re: Linux's implementatio... Linus Torvalds
- Re: Linux's implementatio... Simon Kirby
- Re: Linux's implementatio... Miquel van Smoorenburg
- Re: Linux's implementatio... Dan Maas
- Re: Linux's implementatio... Kai Henningsen
- async IO events Ingo Molnar
- Re: Linux's implementatio... Andi Kleen
- Re: Linux's implementation of poll() n... Felix von Leitner
- Re: Linux's implementation of poll() not scalable? David Schwartz
- Re: Linux's implementation of poll() not scalable? Dan Kegel
- Re: Linux's implementation of poll() not scala... Jordan Mendelson
- Re: Linux's implementation of poll() not s... Dan Kegel
- Re: Linux's implementation of poll() n... Dan Kegel
- Re: Linux's implementation of poll... Linus Torvalds
- Linux C++ safe header project Linux Kernel Developer
- RE: Linux's implementation of poll... Mike Jagdis
- Re: Linux's implementation of... Dan Kegel
- Re: Linux's implementation of poll() n... Linus Torvalds
- Re: Linux's implementation of poll... Dan Kegel