>>but this also means that you should be *very* careful where you put your
>>"touch_buffer(bh)" because you don't want that in a performance path like
>>find_buffer(). leaving touch_buffer() in bread() and brw_page() might be
>
>Hmm, yes probably this is a good idea, I'll think about that tomorrow.
I think to have fixed the lru-list-mkyoung performance problem in
find_buffer/find_page (and looking at the ext2 code the only place where
we must touch a buffer is getblk and not get_hashtable or find_buffer!). I
am releasing a new 2.2.5_arca10.bz2 that should perform better under
no-VM-kernel intensive operation (like a kernel compile), but will be
_equally_ good (or better!) in preserving the working set under heavy
swap. With the latest changes I did today _here_ again responsiveness
under swap is _impressive_ and I suggest _everybody_ to try it out.
ftp://e-mind.com/pub/linux/arca-tree/2.2.5_arca10.bz2
Thanks to Chuck for the good hints.
Andrea Arcangeli
PS. with the help of some no-i386 guy this last arca-patch should compile
also on sparc and alpha (and maybe other archs).
-
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/