>> On a home PC with 2 IDE hard disks, max of around 16 partitions
>> total, and a maximum useful simultaneous mount count of around
>> 16-32 ever being used in the forseeable future...
>>
>> Would it be any form of an optimization to reduce super-max via
>> /proc/sys? I was thinking it might reduce some overhead, perhaps
>> freeing up a page or two of kernel memory or something.
>
>Looking at the source (*cough*), you can see that it wouldn't. super-max
>just maps to max_super_blocks in fs/super.c, and the system won't create a
>new super if that number is >= nr_super_blocks. The actual structures are
>kmalloc()d and put in a linked list.
Well the source for this is over my head, which is why I looked
at the docs... I'm just trying to understand all the proc items
to determine which ones can legitimately be used for localized
system optimization on workstations, servers, busy websites,
etc.. and which are just fluff, or whatnot.
Thanks for pointing out this one for me! Do you have any
suggestions as to which proc entries might actually do something
optimal if tweaked? The vm stuff I figured would, but I'm not
sure what is safe to touch, and what is not, and the
documentation says that you can hose stuff if you aren't sure...
IMHO it should protect against situations like that, but I can't
complain running a free kernel. ;o)
Thanks for the info!
TTYL
-- Mike A. Harris Linux advocate Computer Consultant GNU advocate Capslock Consulting Open Source advocate[insert witty random tagline or cool URL here]
- 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/