"Roy Huang" <[EMAIL PROTECTED]> writes:
> A patch provide a interface to limit total page cache in
> /proc/sys/vm/pagecache_ratio. The default value is 90 percent. Any
> feedback is appreciated.
Anything except a default value of 100% will change the behavior
and probably reduce the performance on most systems.
> -Roy
>
> diff -urp a/include/linux/sysctl.h b/include/linux/sysctl.h
> --- a/include/linux/sysctl.h 2007-01-15 17:18:46.000000000 +0800
> +++ b/include/linux/sysctl.h 2007-01-15 17:03:09.000000000 +0800
> @@ -202,6 +202,7 @@ enum
> VM_PANIC_ON_OOM=33, /* panic at out-of-memory */
> VM_VDSO_ENABLED=34, /* map VDSO into new processes? */
> VM_MIN_SLAB=35, /* Percent pages ignored by zone reclaim */
> + VM_PAGECACHE_RATIO=36, /* Percent memory is used as page cache */
> };
>
>
> diff -urp a/kernel/sysctl.c b/kernel/sysctl.c
> --- a/kernel/sysctl.c 2007-01-15 17:18:46.000000000 +0800
> +++ b/kernel/sysctl.c 2007-01-15 17:03:09.000000000 +0800
> @@ -1035,6 +1035,15 @@ static ctl_table vm_table[] = {
> .extra1 = &zero,
> },
> #endif
> + {
> + .ctl_name = VM_PAGECACHE_RATIO,
> + .procname = "pagecache_ratio",
> + .data = &pagecache_ratio,
> + .maxlen = sizeof(pagecache_ratio),
> + .mode = 0644,
> + .proc_handler = &pagecache_ratio_sysctl_handler,
> + .strategy = &sysctl_intvec,
> + },
> { .ctl_name = 0 }
> };
This is broken.
You have allocated a binary number for use with sys_sysctl but
did not test it.
If you need a special proc_handler to take action when the
value is changed you need a special strategy routine.
So since you aren't going to test the binary interface and don't
care about it please don't allocate a number for it and just
use CTL_UNNUMBERED.
And of course please read the top of linux/sysctl.h
Thank you.
Eric
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to [EMAIL PROTECTED]
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
