Re: [Cooker] Re: kernel-2.4.18.10mdk-1-1mdk
Ural Khassanov <[EMAIL PROTECTED]> writes: > When called from root, pinfo tries to use nonexistent group nobody. > To fix: s/SAFE-GROUP=nobody/SAFE-GROUP=nogroup/ in etc/pinforc 2mdk is on the way, just wait :-) thanks for the bug report. -- "il a ete brule au 28e degre" (the naheulbeuk witch) "c curieux, gcc fonctionne" (gwenole)
Re: [Cooker] Re: kernel-2.4.18.10mdk-1-1mdk
On Tue, Apr 23, 2002 at 10:17:43AM +0200, Juan Quintela wrote:
> > "brian" == Brian J Murrell <[EMAIL PROTECTED]> writes:
>
> brian> On Wed, Apr 10, 2002 at 06:00:19PM +0200, Juan Quintela wrote:
> >> --=-=-=
> >> Name: kernel-2.4.18.10mdk Relocations: (not relocateable)
> >> Version : 1 Vendor: MandrakeSoft
> >> Release : 1mdk Build Date: Wed Apr 10 17:04:06 2002
>
> brian> This kernel is hanging my box. Here is what happens:
>
> brian> I have built the v4l2 and bttv 0.8.33 (would be nice for these to be
> brian> in the Mandrake kernel as they provide v4l tools with much better
> brian> information that the stock v4l does) modules and installed them and
> brian> did the depmod stuff and all that. These are steps which have worked
> brian> time and time again, and are working currently, right now with:
>
> brian> $ uname -r
> brian> 2.4.18-2mdk
>
> brian> But with 2.4.18.10mdk-1-1mdk, when I try to run xawtv the box hangs
> brian> and the CAPS LOCK and SCROLL LOCK LEDs are blinking in unison. I have
> brian> to reboot.
>
> brian> Any ideas? Anything more I can tell you? Any tests you want me to
> brian> run?
>
> Sorry for the delay, you entered kdb (I released that kernel by error
> with kdb enabled), could you test with a later kernel?
Just tried with 2.4.18-12mdk and got an oops trying to load my freshly
built v4l2/bttv-0.8.33 modules. Much nicer to see the oops than the
flashing LEDs from KDB with no promt to do a backtrace with or
anything. :-)
Here is my oops processed through ksymoops:
ksymoops 2.4.5 on i686 2.4.18-12mdk. Options used
-V (default)
-k /proc/ksyms (default)
-l /proc/modules (default)
-o /lib/modules/2.4.18-12mdk/ (default)
-m /boot/System.map-2.4.18-12mdk (default)
Warning: You did not tell me where to find symbol information. I will
assume that the log matches the kernel and modules that are running
right now and I'll use the default options above for symbol resolution.
If the current kernel and/or modules do not match the log, you can get
more accurate output by telling me the kernel version and where to find
map, modules, ksyms etc. ksymoops -h explains the options.
Warning (compare_ksyms_lsmod): module ext3 is in lsmod but not in ksyms, probably no
symbols exported
Unable to handle kernel NULL pointer dereference at virtual address
d8962ecb
*pde =
Oops:
CPU:0
EIP:0010:[]Tainted: P
Using defaults from ksymoops -t elf32-i386 -a i386
EFLAGS: 00010283
eax: ebx: ecx: d89431c0 edx: 0040
esi: edi: d8966534 ebp: b12c5d98 esp: c02f9e60
ds: 0018 es: 0018 ss: 0018
Process insmod (pid: 3500, stackpage=c02f9000)
Stack: d8942a3b d8966534 d8943068 8000 0db0 d8966360
d8966534 d8966360 d895b0bc d8966534 c1439000
d895b613 d8966360 d8962559 d8966360 d8966360 4ea1
Call Trace: [] [] [] [] []
[] [] [] [] [] []
[] [] [call_console_drivers+86/256] [] []
[]
[] [] [] [] [] []
[] [] [] [] [] []
[] [sys_init_module+1346/1568] [] []
[system_call+51/64]
[] [] [] [] []
Code: d8 00 00 00 00 00 00 00 00 50 9a 95 d8 f0 23 94 d8 c0 9c 95
>>EIP; d8962ecb <[bttv]bttv_fops+b/60> <=
>>ecx; d89431c0 <[videodev]videodev_lock+0/0>
>>edi; d8966534 <[bttv]bttvs+1d4/2e30>
>>ebp; b12c5d98 Before first symbol
>>esp; c02f9e60 <_end+d5ec/185157dc>
Trace; d8942a3b <[videodev]video_register_device+ab/1f0>
Trace; d8966534 <[bttv]bttvs+1d4/2e30>
Trace; d8943068 <[videodev].text.end+389/4e1>
Trace; d8966360 <[bttv]bttvs+0/2e30>
Trace; d8966534 <[bttv]bttvs+1d4/2e30>
Trace; d8966360 <[bttv]bttvs+0/2e30>
Trace; d895b0bc <[bttv]bttv_register_video+1c/f0>
Trace; d8966534 <[bttv]bttvs+1d4/2e30>
Trace; d895b613 <[bttv]bttv_probe+473/530>
Trace; d8966360 <[bttv]bttvs+0/2e30>
Trace; d8962559 <[bttv].rodata.end+267a/2e21>
Trace; d8966360 <[bttv]bttvs+0/2e30>
Trace; d8966360 <[bttv]bttvs+0/2e30>
Trace; d8966360 <[bttv]bttvs+0/2e30>
Trace; d8966360 <[bttv]bttvs+0/2e30>
Trace; c011a546
Trace; d8966370 <[bttv]bttvs+10/2e30>
Trace; d89631d0 <[bttv]bttv_pci_tbl+38/8c>
Trace; d8963240 <[bttv]bttv_pci_driver+0/40>
Trace; c01af7de
Trace; d89631d0 <[bttv]bttv_pci_tbl+38/8c>
Trace; d8963240 <[bttv]bttv_pci_driver+0/40>
Trace; c01af857
Trace; d8963240 <[bttv]bttv_pci_driver+0/40>
Trace; d895b8c2 <[bttv]bttv_init_module+b2/e0>
Trace; d8963240 <[bttv]bttv_pci_driver+0/40>
Trace; d8963240 <[bttv]bttv_pci_driver+0/40>
Trace; c011b422
Trace; d8962cb0 <[bttv].rodata.end+2dd1/2e21>
Trace; d8956060 <[bttv]check_alloc_btres+0/6>
Trace; c0108b73
Code; d8962ecb <[bttv]bttv_fops+b/60>
<_EIP>:
Code; d8962ecb <[bttv]bttv_fops+b/60> <=
0: d8 00 fadds (%eax) <=
Code; d8962ecd <[bttv]bttv_fops+d/60>
2: 00 00 add%al,(%eax)
Code; d8962ecf <[bttv]bt
Re: [Cooker] Re: kernel-2.4.18.10mdk-1-1mdk
On Mon, Apr 15, 2002 at 11:52:40AM -0700, Todd Lyons wrote: > > All it's doing is dropping you into the kernel debugger. From the man > page: A! You see, I knew the 2 blinking LEDs was meaningful. > So the command "go" should restart execution. But what is interesting is that there is no kdb prompt anywhere. Even if I am on a console when it happens, no kdb> (or any such kind of prompt) anywhere. > The fact that it started > kdb should show that something is amiss though and it will probably fail > shortly thereafter. Indeed. > If I'm saying things that you already know, my apologies. You seemed to > be asking what the two blinking LED's mean. Indeed, that was one of the points of the question. Much thanks for this information. b. -- Brian J. Murrell msg62645/pgp0.pgp Description: PGP signature
Re: [Cooker] Re: kernel-2.4.18.10mdk-1-1mdk
Thierry Vignaud wrote: >"Brian J. Murrell" <[EMAIL PROTECTED]> writes: > >>>falling back to bttv-0.7.9x ... >>> >>Was this a serious solution? >> >>It is unacceptable. If bttv-0.8.33 caused problems with prior >>kernels then I could understand it but it hasn't -- I have been >>using bttv-0.8.x for my last half dozen MDK kernels at least and it >>has worked fine. It is obvious that something new introduced >>between 2.4.18.2mdk-1-1mdk and 2.4.18.10mdk-1-1mdk has caused a >>problem. >> >>bttv-0.8.x is much superior to bttv-0.7.9x. >> >even you reports some problems. > >so don't come back crying on mandrake because we release stable >drivers whereas you want support for unstable drivers you compiled >yourself. > >we don't support unstable softwares. >we offer support for releases. >cooker isn't a support mailling list, but a development list. >regarding this, i'll better spent my time on v4l userland apps rather >than on bttv. > I ran into the same problem with ATI GATOS experimental driver (km) as well. When I tried to capture video, all I got was a freeze and blinking CAPS + SCROLL lock. Kernel magic SysRq keys worked, though. I tried it on kernel-2.4.18.11-1-1mdk So this may not be related to bttv driver at all, rather v4l issue in general. Michal Bukovjan
Re: [Cooker] Re: kernel-2.4.18.10mdk-1-1mdk
"Brian J. Murrell" <[EMAIL PROTECTED]> writes: > > falling back to bttv-0.7.9x ... > > Was this a serious solution? > > It is unacceptable. If bttv-0.8.33 caused problems with prior > kernels then I could understand it but it hasn't -- I have been > using bttv-0.8.x for my last half dozen MDK kernels at least and it > has worked fine. It is obvious that something new introduced > between 2.4.18.2mdk-1-1mdk and 2.4.18.10mdk-1-1mdk has caused a > problem. > > bttv-0.8.x is much superior to bttv-0.7.9x. the design may be much superior ... but not yet the implementation. > Anyone who has tried to capture any video to a file and play it back > knows this. i've regularly tested it, and on my card, bttv-0.7.xx works smoothly whereas 0.8.x doesn't. what's more, the 0.7.x branch is quite stable whereas 0.8.x not. i'd reports of freeze, black screens, ... you can even look at last davej patch where i fix some compile problems... even you reports some problems. so don't come back crying on mandrake because we release stable drivers whereas you want support for unstable drivers you compiled yourself. we don't support unstable softwares. we offer support for releases. cooker isn't a support mailling list, but a development list. regarding this, i'll better spent my time on v4l userland apps rather than on bttv. -- "la marmotte est toute puissante" (gwenole)
Re: [Cooker] Re: kernel-2.4.18.10mdk-1-1mdk
On Mon, Apr 15, 2002 at 11:36:09AM +0200, Thierry Vignaud wrote: > > falling back to bttv-0.7.9x ... Was this a serious solution? It is unacceptable. If bttv-0.8.33 caused problems with prior kernels then I could understand it but it hasn't -- I have been using bttv-0.8.x for my last half dozen MDK kernels at least and it has worked fine. It is obvious that something new introduced between 2.4.18.2mdk-1-1mdk and 2.4.18.10mdk-1-1mdk has caused a problem. bttv-0.8.x is much superior to bttv-0.7.9x. Anyone who has tried to capture any video to a file and play it back knows this. b. -- Brian J. Murrell msg62613/pgp0.pgp Description: PGP signature
