[lkml]   [2000]   [Nov]   [4]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
Messages in this thread
SubjectRe: [PATCH] Re: Negative scalability by removal of lock_kernel()?(Was:Strange performance behavior of 2.4.0-test9)
On Sat, 4 Nov 2000, Andrew Morton wrote:

> Dean,
> neither flock() nor fcntl() serialisation are effective
> on linux 2.2 or linux 2.4.

i have to admit the last time i timed any of the methods on linux was in
2.0.x days. thanks for the updated data!

> For kernel 2.2 I recommend that Apache consider using
> sysv semaphores for serialisation. They use wake-one.

sysv semaphores have a very unfortunate negative feature -- if the admin
kill -9's the server (impatient admins do this all the time) then you end
up leaving a semaphore lying around. sysvsem don't have the usual unix
unlink semantics. actually flock has the same problem... which is why i
generally preferred fcntl whenever it was a performance wash, as it was
back in 2.0.x days.

however given the vast performance difference i think it warrants the
change. i'll include your results with the commit.

> For kernel 2.4 I recommend that Apache use unserialised
> accept.

per linus' request i'll unserialise 2.2 as well.

i'll leave 2.0.x settings alone.

(oh yeah, and compile-time only detection.)


To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to
Please read the FAQ at

 \ /
  Last update: 2005-03-22 12:45    [W:0.210 / U:0.576 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site