Re: X much slower in 2.0.24 than in 1.2.13

Kevin Buhr (buhr@stat.wisc.edu)
05 Nov 1996 11:42:27 -0600

alan@lxorguk.ukuu.org.uk (Alan Cox) writes:
|
| > The 2.0.24 changes to "struct sock" in "linux/include/net/sock.h" push
| > its size, in my kernel, up to 0x1fc. This new size, plus the
| > "kmalloc"'s "block_header" is just big enough to push "struct sock"
| > allocations to the next allocation order.
|
| We hardly ever allocate a struct sock

I wasn't thinking of the actual *allocation* overhead; I was thinking
of the working set of (unswappable) pages needed to play with a bunch
of existing "struct sock"s. Perhaps this isn't as much of an issue as
I first assumed, but I see all that extra time spent in "free_pages"
and "_get_free_pages", and I imagine the kernel working twice as hard
to find memory for everything else simply because its pool of "struct
sock"s is twice as big as it used to be.

Kevin <buhr@stat.wisc.edu>