In message <[EMAIL PROTECTED]>, Poul-Henning Kamp
writes:
>phk 2002/12/27 03:05:05 PST
>
> Modified files:
> sys/ufs/ffs ffs_vfsops.c
> Log:
> Use three UMA zones for FFS/UFS inodes instead of malloc space.
> Since inodes are currently 144 bytes, this will save 112 bytes per
> inode. This can amount to up to 10MByte on large systems.
All things considered, I have a hard time telling if this will
result in a performance improvement (apart from the memory savings)
or not. There are a lot of indications which point in opposite
directions so I think it will be a wash.
I considered making the zones per mountpoint, but decided against
it for now, so for now all FFS/UFS filesystms share these three
zones.
You can monitor these zones with:
syv# sysctl vm.zone | sed -n -e 2p -e /FFS/p
ITEM SIZE LIMIT USED FREE REQUESTS
FFS2 dinode: 256, 0, 17200, 95, 97431
FFS1 dinode: 128, 0, 0, 0, 0
FFS inode: 144, 0, 17200, 104, 97431
But be aware that the USED count is number of items in the zone,
not number of items allocated from the zone.
--
Poul-Henning Kamp | UNIX since Zilog Zeus 3.20
[EMAIL PROTECTED] | TCP/IP since RFC 956
FreeBSD committer | BSD since 4.3-tahoe
Never attribute to malice what can adequately be explained by incompetence.
To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-current" in the body of the message
