Relative performance of swap-backed MFS vs. regular UFS?

2004-10-22 Thread Clifton Royston
I have seen some conflicting information posted about this in the past, and I figure this is the best place to get an authoritative answer. For a large temporary file system which must hold short-lived files, mostly small but occasionally several very large ones (e.g. 100MB+), is it better for

Re: Messed up my partition

2004-10-22 Thread Anish Mistry
On Friday 22 October 2004 02:55 pm, Erik Udo wrote: > My boot loader wansn't working properly and i didn't find my 5.2.1 > install cd, so i grabbed 4.4(or is that 4.6) install cd, and chose fdisk, > marked the freebsd partition bootable(S), and wrote changes. > You should be able to use gpart on a

Re: Messed up my partition

2004-10-22 Thread Erik Udo
Seems like my paste from sysinstall's fdisk is messy, here is the output of fdisk: koti# fdisk *** Working on device /dev/ad0 *** parameters extracted from in-core disklabel are: cylinders=395136 heads=16 sectors/track=63 (1008 blks/cyl) Figures below won't work with BIOS for partitions not

Messed up my partition

2004-10-22 Thread Erik Udo
My boot loader wansn't working properly and i didn't find my 5.2.1 install cd, so i grabbed 4.4(or is that 4.6) install cd, and chose fdisk, marked the freebsd partition bootable(S), and wrote changes. I didn't notice that it had made my about 150GB partition as 100GB(ad0s3) and 60GB(unused) partit

Re: [PATCH] Re: Linksys PCM200

2004-10-22 Thread Brooks Davis
On Fri, Oct 22, 2004 at 02:16:54AM -0600, [EMAIL PROTECTED] wrote: > > > Could some people with dc(4) devices please test the following patch? > > It compiles for me and is trivial, but a quick test is probalby in order > > before I commit it. > > It's rather necessary to test it well, because I

Re: ServerWorks and PM/SMB

2004-10-22 Thread Stewart Morgan
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 Hello again, An update: I found mention of the ServerWorks OSB4 chips having a quirk with the SMBus mapping register that has been fixed in -current. Adding an entry to pci.c to apply the same quirk to the CSB5 chip enables intpm(4) to successf

fix to a data corruption problem

2004-10-22 Thread Mike Tancsa
We are trying to fix a bug in the uhid interrupt handler/driver http://www.freebsd.org/cgi/query-pr.cgi?pr=kern/72935 where it can inadvertently walk into another handler working with clist data structures in sys/kern/tty_subr.c. Uhid is a "bio" class device, which means that its not masked when an

ServerWorks and PM/SMB

2004-10-22 Thread Stewart Morgan
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 Hello, I'm running a 4-STABLE kernel build from last night's CVS-sources. Trying to get some sort of power/temp/fan system monitoring running on an ASUS PR-DLSR533, dual Xeon based on the ServerWorks chipset. Attached is a dump of "pcico

[PATCH] Re: Linksys PCM200

2004-10-22 Thread soralx
> Could some people with dc(4) devices please test the following patch? > It compiles for me and is trivial, but a quick test is probalby in order > before I commit it. It's rather necessary to test it well, because I didn't actually remove the card's case to see the chip; I relied on my own test

help with poppassd

2004-10-22 Thread Ed Stover
Problems with poppassd on 4.10, I can talk to poppassd 106 from telnet both on and off localhost. but for some odd reason i can't get it to change my password. ok, here are the steps that i have taken all ready cd /usr/ports/mail/poppassd/ && make install (it install perfectly) (added this line t