Re: Question about file system checks
On Thursday 27 March 2008 14:45:49 Marian Hettwer wrote: > On Thu, 27 Mar 2008 13:39:55 +, Matthew Seaman > > <[EMAIL PROTECTED]> wrote: > > -BEGIN PGP SIGNED MESSAGE- > > Hash: RIPEMD160 > > > > Jared Carlson wrote: > >> Hi I have a question about startup scripts for BSD distributions. > >> Can you turn off the file system check that occurs every 30 boots, > >> etc? I recall this being the case on a BSD platform, although my Mac > >> OS X doesn't (to my knowledge) do a file system check that often at > >> all. > > > > You are thinking of the Linux ext2/ext3 filesystem. > > Although this is OT, does anybody have a clue why ext2/ext3 filesystems > behave like that? > I wouldn't like to trust a filesystem which thinks a fsck is worth it, > although it always was a clean shutdown. > Any clue?! :) ext2/3 is mounted async by default, I reckon most linux distros expect some fs damage to occur because of that over time maybe. Or it's a relic of the days when that was necessary, maybe it's not really necessary now anymore. Perhaps it also does some defrag while fsck'ing. Generally I can say that with freebsd even if you pull the plug and then let it reboot and do the automatical background fsck you'll likely loose only that one file you might have been editing while (or just before) you unplugged the box. Dan > cheers, > Marian > > ___ > freebsd-stable@freebsd.org mailing list > http://lists.freebsd.org/mailman/listinfo/freebsd-stable > To unsubscribe, send any mail to "[EMAIL PROTECTED]" ___ freebsd-stable@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-stable To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: +rtfree: 0xffffff0003635780 has 1 refs
On Wednesday 19 March 2008 14:28:28 John Baldwin wrote: > On Tuesday 18 March 2008 06:09:46 pm Danny Pansters wrote: > > On Monday 17 March 2008 20:55:01 Stefan Lambrev wrote: > > > Greetings Dave, > > > > > > Dave Overton wrote: > > > > I am new to the AMD64 stable branch, so forgive me if this has been > > > > beat to death, but I can't find why this message keeps occurring over > > > > and over all day. FreeBSD 7.0 Stable on AMD x2. It works (or seems > > > > to) fine. > > > > > > > > +rtfree: 0xff0003635780 has 1 refs > > > > > > check google for rtfree() used when RTFREE() needed .. or something > > > like this :) > > > Those messages are annoying but harmless. > > > > Harmless perhaps, but it still should be fixed, so if you don't see any > > similar PR already I'd suggest sending one. > > > > Has to do with certain variables being of one type but used as if it were > > another (e.g. int vs long) which on 64bit platforms as a band-aid > > gets "MSB-filled" with 0xf's to the proper size. So such warning pretty > > much means "fix your code". > > No. The value printed is a pointer and kernel pointers on amd64 are in the > upper range of the address space. The warning above has to do with code > that calls rtfree() vs. the RTFREE() macro. The macro inlines the the > common case (refs > 1) so in theory is cheaper than always doing a function > call. I stand corrected! Dan ___ freebsd-stable@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-stable To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: +rtfree: 0xffffff0003635780 has 1 refs
On Monday 17 March 2008 20:55:01 Stefan Lambrev wrote: > Greetings Dave, > > Dave Overton wrote: > > I am new to the AMD64 stable branch, so forgive me if this has been beat > > to death, but I can't find why this message keeps occurring over and over > > all day. FreeBSD 7.0 Stable on AMD x2. It works (or seems to) fine. > > > > +rtfree: 0xff0003635780 has 1 refs > > check google for rtfree() used when RTFREE() needed .. or something like > this :) > Those messages are annoying but harmless. Harmless perhaps, but it still should be fixed, so if you don't see any similar PR already I'd suggest sending one. Has to do with certain variables being of one type but used as if it were another (e.g. int vs long) which on 64bit platforms as a band-aid gets "MSB-filled" with 0xf's to the proper size. So such warning pretty much means "fix your code". HTH, Dan > > its a kernel "bold" on the terminal, and would scare me to death if I > > just knew what it meant... > > > > Should I be worried about something? I hate bold white text > > > > > > > > Dave Overton, Owner > > SYIX.COM > > > > [EMAIL PROTECTED] > > (530) 755-1751 x101 > > Fax (530) 751-8871 > > 800-988-SYIX > > > > ___ > > freebsd-stable@freebsd.org mailing list > > http://lists.freebsd.org/mailman/listinfo/freebsd-stable > > To unsubscribe, send any mail to "[EMAIL PROTECTED]" > > ___ > freebsd-stable@freebsd.org mailing list > http://lists.freebsd.org/mailman/listinfo/freebsd-stable > To unsubscribe, send any mail to "[EMAIL PROTECTED]" ___ freebsd-stable@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-stable To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: kld regression
On Thursday 31 January 2008 16:05:57 Andriy Gapon wrote: > on 31/01/2008 14:39 Andriy Gapon said the following: > > on 31/01/2008 13:07 John Baldwin said the following: > >> On Wednesday 30 January 2008 12:39:14 pm Andriy Gapon wrote: > >>> The problem is as follows: > >>> 1. put udf_load="YES" in loader.conf > >>> 2. you can mount and unmount udf filesystems > >>> 3. you can kldunload udf if no udf filesystems are mounted > >>> 4. now mount udf fs while udf.ko is unloaded > >>> 5. udf is auto loaded and fs is mounted > >>> 6. unmount fs > >>> 7. try to kldunload udf > >>> kldunload: can't unload file: Device busy > >>> kernel message: kldunload: attempt to unload file that was loaded by > >>> the kernel > >>> > >>> Yeah, it was loaded by kernel indeed, but WTF - what is the difference > >>> from manual/loader.conf loading and why I can not manage my modules as > >>> I wish? > >> > >> Hmm, the relevant code (vfs_init.c) hasn't changed in 6.x since 6.0. > >> There were some changes in 7.0, but this should work in both branches. > >> What is the previous release that this worked on? > > > > Maybe I was wrong when I called this regression, but this was very > > surprising behavior for me. And in 5.X I did a lot of udf > > debugging/experimenting and never encountered such a problem. Maybe I > > always did kldload before mount, I can't tell now. > > Anyway, this seems like an annoyance at the very least, pinning a kernel > > module without any important reasons. > > Hmm, I found one difference with previous setups: in step 1 I also have > udf_iconv_load="YES" and udf_iconv.ko module is what seems to prevent > udf.ko from unloading in step 7. I can actually unload udf_iconv and > then I am able again to unload udf. > > Still don't understand what is a big difference here. > > And if I had UDF_ICONV built into kernel then I wouldn't have this > work-around. The way I understand it, there are two different things: 1. kldload will load "child" modules on demand, but kldunload does not attempt to unload any other modules than the one you ask for. I don't think it's material whether the kldload was done via the loader (before the kernel kmod gets loaded, kernel is also a kmod), or after. It's also possible to have modules who need to access the filesystem (other than /boot partition) when kldloaded, and such modules can obviously not be loaded via the loader at all. 2. There may be modules (such as bktr) that allocate memory in kernel space. These can not be unloaded, because that memory may not be deallocated while the kernel is up and running (the latter is my assumption). Anyhow, unless you're very tight on RAM, it hardly matters if you have some kmods loaded but left unused. Kmods are small, typically 10-100 kB. If all your kmods are built into the kernel, you have the footprint of all of them and you can't disable any at runtime. Feel free to correct me, anyone, if you think the above is not correct or complete. Dan ___ freebsd-stable@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-stable To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: [kde-freebsd] problem hal - k3b ?
On Thursday 19 April 2007 22:47:08 Richard Kuhns wrote: > On Thursday 19 April 2007 12:12, Michael Nottebrock wrote: > > On Wednesday, 18. April 2007, Beni wrote: > > > Hi List, > > > > > > I think I have a problem with hal(d) and k3b (version 1.0 from ports) : > > > my whole system freezes when starting up k3b. I get the splash screen > > > and then it all stops and a ctrl-alt-del is the only way out. > > > > Other people have reported kernel panics. It looks to me like k3b's > > device probing and hald's device probing at the same time manages to > > tickle a bug in ata(4). > > > > Ref: > > http://lists.freebsd.org/pipermail/freebsd-current/2007-April/070753.html > > http://lists.freebsd.org/pipermail/freebsd-stable/2007-April/034486.html > > > > I'm afraid a true kernel hacker will have to inconvenince themselves with > > running k3b and hal in order to have this one fixed. FWIW, I haven't seen > > in happening on 5.5. > > > > > > Cheers, > > FWIW: I've been running with a kernel from Jan 22 with no problem. I built > and installed a new kernel/world today and can't run k3b as described > above. I've since reverted to my old kernel and everything's fine again. > - Rich I haven't used k3b for a while but I do have problems with kscd, kaudiocreator, and amarok CD playing. I'm running -stable from Feb 26 and latest kde and CD contents are only properly seen if you first insert the cd, then wait, and then choose the app from the "hal" menu (which takes several seconds to pop up). But even then ripping seems to deadlock, it just hangs and does nothing after you queue and start, and kscd/amarok don't play anything. Amarok merely shows the titles briefly. Both with self-made CDs and official audio CDs. It's possible that my DVD player/recorder is just dying but I'm afraid we're having a more general regression here. Dan ___ freebsd-stable@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-stable To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: upgrade from 5.4-STABLE -> 6 questions
On Sunday 20 November 2005 01:28, you wrote: > On Sun, Nov 20, 2005 at 12:43:06AM +0000, Danny Pansters wrote: > > On Saturday 19 November 2005 23:37, Jason wrote: > > > 1. after my upgrade, my serial ports are not accessible > > > monsterjam# ls -al /dev/cuaa* > > > ls: No match. > > > > I got surprised by this too, when trying to use my camera through serial > > cable. They've become /dev/cuad and /dev/ttyd for outgoing and incoming > > respectively. See sio(4). (I needed /dev/ttyd0 for my camera) > > Isn't this also in the release notes? > > Kris From a quick scan (off the web page) the sio change appears only briefly at "2.2 Kernel Changes". This was probably already in 5.4, no? Perhaps that's why it's not very pronounced. I went from 5.X to 6.0 through cvsup and UPDATING doesn't seem to have it either. It was no problem for me though, just a matter of reading the man page. I haven't checked the Handbook on this, but if nothing about sio is there perhaps someone would want to update. Dan ___ freebsd-stable@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-stable To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: upgrade from 5.4-STABLE -> 6 questions
On Saturday 19 November 2005 23:37, Jason wrote: > 1. after my upgrade, my serial ports are not accessible > monsterjam# ls -al /dev/cuaa* > ls: No match. I got surprised by this too, when trying to use my camera through serial cable. They've become /dev/cuad and /dev/ttyd for outgoing and incoming respectively. See sio(4). (I needed /dev/ttyd0 for my camera) HTH, Dan ___ freebsd-stable@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-stable To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: mplayer + bktr
On Monday 14 November 2005 01:02, Daniel O'Connor wrote: > On Mon, 14 Nov 2005 07:13, Miguel Mendez wrote: > > > If you add immediatemode=0 to the tv args it will do it. > > > > I've just tried that and the speed drops to something like 5-6 fps. > > Interesting problem. > > Yeah, I suspect there is something wrong with the audio sampling, but I > haven't looked at it properly. I myself am not really doing anything with capturing, but some ideas that may be helpful: FWIW, from what I know in ring capture mode the video gets synch'd by the audio by having enough frames per audio sample. So if audio sample size or expected speed or expected kHz is somehow wrong... Also the code for ring capture mode (as opposed to immediate which does not do audio but does give a video one could capture at 25 fps) has its own timing (perhaps it uses rtc down the line, I dunno, is rtc.ko alright?). You may get into a worst-worst-worst-even worst scenario where the software timer degrades on and on possibly. Maybe capturing only works well if you use immediate (case 2 in bktr(4) IIRC) and you should capture audio seperately and later merge them to frames. > > > I don't use either of those, but a small program I wrote which captures > > > YUV frames and uses the Xv extension doesn't show the problem. > > Capturing only pictures at 25 fps (with mplayer vo) can be handled easily. Dan ___ freebsd-stable@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-stable To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: Problems with PCI-express video card
[ top posting for a change ;-) ] Is this is a brand new board/box (eg first time trying FreeBSD with it)? If so, try physically installing first the RAM then the pci-x card. They want a memory adress pool allocated and that should be "above" your RAM allocated addresses. You don't need agp in your kernel, in fact better not if you use nvidia-driver. You might also have a basket case situation where your allocated addresses for the nvidia overlap with those for another PCI card, but I think usually the OS will handle this. My first bet would be on the "RAM chipping" that might occur as explained above. Also note (perhaps superfluous) that you shouldn't be running X when (re)loading nvidia.ko. But perhaps it just isn't supported. Does using X' nv work or not? HTH, Dan On Thursday 07 April 2005 04:12, Stephen Montgomery-Smith wrote: > I have a nvidia PCI-express 6600 video card on an Intel SE7525GP2 > motherboard. I am running a fairly recent version of RELENG_5. > > I am having great difficulty getting the video card to work. On an > older version of Xorg it would try to use the vesa driver. It did put > up some video, but it was illegible. > > The most recent version of Xorg tries to use the nv driver. I have also > tried out the very recent nvidia driver that just came out on the nvidia > website a few days ago. In these cases startx simply doesn't work, and > I get the error messages: > > (EE) NVIDIA(0): No valid FB address in PCI config space > > I might add that when the nvidia.ko is loading, it puts out messages > like this: > > pci3: on pcib3 > nvidia0: mem > 0xdd00-0xddff,0xf000-0x,0xd8 > 00-0xdbff at device 0.0 on pci3 > nvidia0: NVRM: NVIDIA MEM resource alloc failed. > nvidia0: NVRM: NVIDIA driver alloc failed. > device_attach: nvidia0 attach returned 6 > > I have tried also messing with the AGP settings to no avail, but I guess > that is irrelevant anyway (is AGP totally unrelated to PCI-express?) > > I don't know if this is a video card issue, or a motherboard issue, or > what. It works beautifully with Windows XP. I might try it out with > Linux also, but I haven't got that far yet. > > Thanks, Stephen > ___ > freebsd-stable@freebsd.org mailing list > http://lists.freebsd.org/mailman/listinfo/freebsd-stable > To unsubscribe, send any mail to "[EMAIL PROTECTED]" ___ freebsd-stable@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-stable To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: Problems with AMD64 and 8 GB RAM? (solved)
This may not be related at all, but I've been following this thread and at certain times I wanted to drop in and stress that things like pci express cards should be re-inserted after the RAM has been (re)inserted. Perhaps the BIOSes are really that stupid that they only rely on memory adresses assigned by when you plugged in what. Anyway, it may not matter at all for you but knowing the kind of inquiring mind you have, and considering that this was one of the "sorta-would-it-matter" things hanging over this discussion, perhaps it's a point of interest also. This is how I understand it: FreeBSD 5.4-PRERELEASE #0: Fri Apr 1 01:16:43 CEST 2005 root@:/usr/obj/usr/src/sys/GENERIC Timecounter "i8254" frequency 1193182 Hz quality 0 CPU: Intel(R) Pentium(R) 4 CPU 3.40GHz (3400.12-MHz 686-class CPU) Origin = "GenuineIntel" Id = 0xf34 Stepping = 4 Features=0xbfebfbff Hyperthreading: 2 logical CPUs real memory = 1071837184 (1022 MB) avail memory = 1035120640 (987 MB) mptable_probe: MP Config Table has bad signature: \M^?\M^?\M^?\M^? ACPI APIC Table: [part of dmesg with this board] I have a rather new uniprocessor Intel board and the manual specifically stressed (in the light of the 4 BG limit and the &*^*&^:-o things BIOSes might do to overcome that -- the bios will want to have that reserved adress pool just below the 4 GB and depending on the bios it can map it above 4GB perhaps) that any pci express cards (my graphics card) must be inserted AFTER the RAM. Also, about the pairing of DDR2 RAM: you must alight the ram in both channels in principle (that is both in A not B), but if you don't it works if you have the same totals. Again, mobo docs should have all the quirky rules. I didn't want to disturb the discussion but in retrospective, I feel that perhaps some of these points might matter. My observations are on i386 not amd64 but the same kind of logic seems to apply AFAICT. HTH, (in retrospective of course ;-) Dan ___ freebsd-stable@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-stable To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Errata - 5.4-BETA sysinstall seems to want to write to its own / instead of the one on disk
s/fsck/fdisk (and disklabel) Sorry, typo. ___ freebsd-stable@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-stable To unsubscribe, send any mail to "[EMAIL PROTECTED]"
5.4-BETA sysinstall seems to want to write to its own / instead of the one on disk
I've just put together a new box and tried the 5.4-BETA1 bootonly iso. The box has one SATA drive ATM, it has a Intel D925XECV2 mobo with onboard SATA. It does newfsk the disk but when it wants to write to it either when setting up users/groups or installing packages or distribution sets it folds: /: write failed, filesystem is full It seems to be trying to write to its RO memory disk? FYI I set up the hard disk pretty normally with an /, a swap, a /tmp, a /var and /usr and /home partitions. It did complain about geometry but I get that often. I could newfs and fsck it through sysinstall, so I think it's not a device problem. Installing 5.3 then cvsup to 5.4 and make world worked fine. I can test stuff on this box if folks want to. Dan ___ freebsd-stable@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-stable To unsubscribe, send any mail to "[EMAIL PROTECTED]"
rpm or rpm2cpio broken?? (4.3-RC)
Hello,
I just cvsup'ed and upgraded my desktop box from 4.2 stable (12202000) to
what is now 4.3RC (04192001) including all ports I wanted to use. I used
pkg_version to find out which ports needed recompiling/installation. Almost
everything works, even KDE2.1, building the parts one by one. However I ran
into two weird problems which look to be related:
When trying to compile/install kdeutils-2.1.1 from the ports, it threw me a
syntax error when compiling the rpm bits of kpackage (rpm: redhat package
manager). I have pasted the output below. Looks like the RPM headers used are
wrong?
When trying to install the Linux RealPlayer 8 which is to be fectched
manually as a RPM package it appeared to either have trouble with RPM or with
rpm2cpio. I know from older posts in -questions that rpm2cpio lives in
/compat/linux/bin as well as in /usr/local/bin. As the Makefile for
RealPlayer seems to use the "native" one I did not attempt to change this.
After all, why would they use it if it don't work? I thought that the rpm and
the rpm2cpio ports were basically unrelated anyway? Error messages are below.
When reinstalling rpm2cpio (hey, you never know :-) I noticed that I had to
use NO_CHECKSUM=yes, as if its version number is not correct or something.
I'm confused! Can someone help me out?
Many thanks in advance,
--
Danny Pansters
http://www.ricin.com
ERROR OUTPUT:
** kdeadmin install fail:
gmake[3]: Entering directory
/usr/ports/sysutils/kdeadmin/work/kdeadmin-2.1.1/kpackage'
c++ -DHAVE_CONFIG_H -I. -I. -I.. -I/usr/local/include
-I/usr/X11R6/include/qt2 -
I/usr/X11R6/include -I/usr/local/include/rpm -I/usr/include -D_GETOPT_H
-D_PTH
_H_ -D_PTH_PTHREAD_H_ -I/usr/local/include -I/usr/local/include
-I/usr/X11R6/in
clude/qt2 -O2 -fno-exceptions -fno-check-new -Wall -pedantic -W
-Wpointer-arith
-Wmissing-prototypes -Wwrite-strings -Wno-long-long -Wnon-virtual-dtor
-fno-bui
ltin -O -pipe -DQT_CLEAN_NAMESPACE -DQT_NO_COMPAT -DQT_NO_ASCII_CAST -c
main.cpp
In file included from main.cpp:32: rpmInterface.h:16: syntax error before ;'
gmake[3]: *** [main.o] Error 1
gmake[3]: Leaving directory
/usr/ports/sysutils/kdeadmin/work/kdeadmin-2.1.1/kpackage'
gmake[2]: *** [all-recursive] Error 1
gmake[2]: Leaving directory
/usr/ports/sysutils/kdeadmin/work/kdeadmin-2.1.1/kpackage'
gmake[1]: *** [all-recursive] Error 1
gmake[1]: Leaving directory
/usr/ports/sysutils/kdeadmin/work/kdeadmin-2.1.1'
** /usr/ports/sysutils/work/kdeadmin-2.1.1/kpackage/rpmInterface.h:16:
[...]
#ifdef HAVE_RPM
#include
/* stupid RPM headers! */
typedef __off64_t off64_t; <-- line 16
#include
#else
typedef struct Header{
};
#endif
[...]
** linux-realplayer8 install fail:
ricin # make NO_CHECKSUM=yes install
===> Extracting for linux-realplayer-8.cs1
===> linux-realplayer-8.cs1 depends on executable: rpm2cpio - found
CPIO archive found!
gzip: stdin: unexpected end of file
cpio: premature end of file
*** Error code 1
To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-stable" in the body of the message
