dean gaudet <[EMAIL PROTECTED]> writes:
> there should be no reason to tweak TCP_NODELAY except at socket creation
> time.
on at least some levels of Linux TCP_NODELAY and TCP_CORK are mutually
exclusive...
2.2.5:
case TCP_NODELAY:
...
if (sk->nonagle == 2)
return -EINVAL;
sk->nonagle = (val == 0) ? 0 : 1;
return 0;
case TCP_CORK:
...
if (sk->nonagle == 1)
return -EINVAL;
--
Jeff Trawick | [EMAIL PROTECTED] | PGP public key at web site:
http://www.geocities.com/SiliconValley/Park/9289/
Born in Roswell... married an alien...
- Re: mod_file_cache performa... Victor J. Orlikowski
- Re: mod_file_cache performa... Sander van Zoest
- Re: mod_file_cache performa... dean gaudet
- Re: mod_file_cache performa... Cliff Woolley
- Re: mod_file_cache performa... Bill Stoddard
- Re: mod_file_cache performa... dean gaudet
- Re: mod_file_cache performa... Bill Stoddard
- Re: mod_file_cache performa... rbb
- Re: mod_file_cache performa... Brian Pane
- Re: mod_file_cache performa... dean gaudet
- Re: mod_file_cache performa... Jeff Trawick
- Re: mod_file_cache performa... dean gaudet
- Re: mod_file_cache performance dean gaudet
- Re: mod_file_cache performance Ian Holsman
- Re: mod_file_cache performance Cliff Woolley
- RE: mod_file_cache performance Gonyou, Austin
- RE: mod_file_cache performance Cliff Woolley
- RE: mod_file_cache performance Gonyou, Austin
- RE: mod_file_cache performance Gonyou, Austin
- RE: mod_file_cache performance Cliff Woolley
- RE: mod_file_cache performance Gonyou, Austin
