Messages in this thread | | | Date | Tue, 13 Apr 1999 06:36:26 -0400 (EDT) | From | jamal <> | Subject | Linux TCP Fixing everyones problems? WAS(Re: TCP push sometimes missing under 2.2.5? |
| |
cw> So why put in the fix? Are we actually doing anything wrong? And as cw> you point out, the register pressure might not be a good thing cw> considering how many register the x86 has...
I have never understood the logic behind some of the fixes that seem to go into the kernel TCP code. Some OS fucks up by either not conforming to the standards (Sunos was the last one i remember being discussed here) or by doing something unusual as in this case with the Push -- and a fix comes around for it in Linux, and it seems OK for it to go into the kernel. Should these fixes be put on some unofficial patch web page somewhere like linuxmama.com? Why in hell should i be running a few extra opcodes per TCP packet (in the fast path of all places) when i dont talk to Apples?
cheers, jamal
- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.rutgers.edu Please read the FAQ at http://www.tux.org/lkml/
| |