Re: audio cd / cdrom drive in ti-book
On Sun, 25 Aug 2002 17:00:56 +1000, Toby Sargeant <[EMAIL PROTECTED]> wrote: > On Sat, Aug 24, 2002 at 05:02:33AM +0200, malte wrote: > > Joss...you're a wonderful person! =) > > that was it...i think its the oddest thing. > > but pointing #open file# right at /dev/scd0 did the > > trick. dont know why...but i hear audio cd from my tibook this second > > The difference here is that xmms is reading CDDA off the disc, and > playing it as an audio sample. Playing the audio cd directly does > not work, and this is (as far as I'm aware) a limitation of the sound > driver. It's the hardware. Unlike in a desktop, which usually has a thin cable running through the EMF-noise-filled environment of the inside of your case from the CD-ROM to the sound card, the T/iBooks don't have a direct audio connection. -- |>|\/|< /----------\ |David M. Cooke http://arbutus.physics.mcmaster.ca/cookedm/ |[EMAIL PROTECTED]
Re: Automatic power-on
On Sun, Sep 01, 2002 at 06:26:53PM +0200, Michel Lanners wrote: > On 31 Aug, this message from Eric C. Cooper echoed through cyberspace: > > On CUDA-based PowerMacs, this will set it to "file server mode" (i.e., > > automatically reboot upon restoration of power): > > echo -ne '\001\023\001' > /dev/adb > > > > This Apple tech note was helpful in finding the magic '\023' that > > enables "server mode": > > http://developer.apple.com/technotes/tn/tn1079.html > > Emm, wouldn't that be 0x13? According to my copy of that technote, 0x23 > sets wakeup mode (i.e enable wakeup by the GPi in of the Geoport serial > ports). Hex vs. octal escapes (\023 == \0x13) -- |>|\/|< /--\ |David M. Cooke http://arbutus.physics.mcmaster.ca/cookedm/ |[EMAIL PROTECTED]
Re: ibook 2 : several question
On Mon, Sep 02, 2002 at 06:02:42PM +0200, Pierre Gambarotto wrote: > I just bought an ibook 700Mhz/14" LCD. > > So far, I succeeded in installing debian in dual boot with Mac OS X, and > it runs pretty well :-) > > Several remaining problems anyway : > _ I read pmud's manual : is it right that sleep mode for ibook 700Mhz > is not supported ? It's not clear in the manual, but the /etc/power/pwrctl > script say so. Can anybody confirm ? I haven't had problems. (I have the same, but with the 12" screen.) Occasionally, I'll compile a kernel that won't sleep properly, but that's probably my fault for fiddling with the kernel config too much :-) > _ DRI : the dri module is loaded (thanks to src from > http://www.penguinppc.org/~daenzer/). > Glxinfo reports dri is ok : > ... > direct rendering: Yes > ... > > But any GL app freezes X (I use Xfree4.2 from > http://people.debian.org/~branden/) > Someone with more luck than me ? Nope :-) Since I don't really need OpenGL stuff at the moment, I've set LIBGL_ALWAYS_INDIRECT in my .xsession so that direct rendering isn't used. Then GL apps won't freeze X (but will run slower :0) -- |>|\/|< /--\ |David M. Cooke http://arbutus.physics.mcmaster.ca/cookedm/ |[EMAIL PROTECTED]
Re: ibook 2 : several question
On Mon, Sep 02, 2002 at 02:24:24PM -0400, Francis J. Lacoste wrote: > On lun, 2002-09-02 at 14:10, Jesus Climent wrote: > > On Mon, Sep 02, 2002 at 02:07:05PM -0400, Francis J. Lacoste wrote: > > > I have a iBook2 700Mhz (12" screen) with the Radeon M6 and sleeping is > > > not working properly. Well, sleeping works but it never wakes up. I have > > > ran several kernels and it never work. > > > > Here it was working until I rebuilt the kernel without ide-cdrom (since > > the ide-scsi parameter on boot did not work: it was rejected by the > > kernel saying that it was not an existing kernel parameter, and it > > happened between rc4-benh0 to 2.4.19-benh0 final). > > > > After that it never wakes up, so I maight have to decide wetween > > sleep/wake up or cd recording. > > > > My kernel was compiled with IDE CDROM support. (SCSI emulation is > compiled as a module and is not working either.) > > CONFIG_BLK_DEV_IDECD=y > CONFIG_BLK_DEV_IDESCSI=m > > On boot, I have the following messages: > > Kernel command line: root=/dev/hda10 ro hdb=scsi > ide_setup: hdb=scsi -- BAD OPTION I had similiar problems, but it's working now. I had to disable IDE CDROM support, and compile SCSI emulation in. Since several people seem to have problems, I've put the deb for my kernel (2.4.20-pre5-ben0), along with the kernel config file and a deb for daenzer's drm-trunk-module at http://arbutus.physics.mcmaster.ca/cookedm/ibook/ Currently, CDROM support with SCSI emulation, sleeping/waking up, and sound work. I haven't tested the CD-RW, but cdrecord -scanbus works. -- |>|\/|< /--\ |David M. Cooke http://arbutus.physics.mcmaster.ca/cookedm/ |[EMAIL PROTECTED]
Re: ibook 2 : several question
On Mon, Sep 02, 2002 at 10:06:27PM +0200, Vincent Bernat wrote: > OoO Pendant le journal télévisé du lundi 02 septembre 2002, vers > 20:53, [EMAIL PROTECTED] (David M. Cooke) disait: > > > Since several people seem to have problems, I've put the deb for my > > kernel (2.4.20-pre5-ben0), along with the kernel config file and a deb > > for daenzer's drm-trunk-module at > > > http://arbutus.physics.mcmaster.ca/cookedm/ibook/ > > What kind of iBook do you have to use the drm-trunk stuff ? They're compiled from http://penguinppc.org/~daenzer/debian/dri-trunk/ They should work with any iBook, but the kernel is likely iBook2 only Direct rendering in OpenGL with the Radeon M6 on mine still doesn't work. -- |>|\/|< /----------\ |David M. Cooke http://arbutus.physics.mcmaster.ca/cookedm/ |[EMAIL PROTECTED]
Re: ibook 2 : several question
On Mon, Sep 02, 2002 at 10:24:29PM +0200, Vincent Bernat wrote: > OoO En cette soirée bien amorcée du lundi 02 septembre 2002, vers > 22:18, [EMAIL PROTECTED] (David M. Cooke) disait: > > > They should work with any iBook, but the kernel is likely iBook2 only > > > Direct rendering in OpenGL with the Radeon M6 on mine still doesn't > > work. > > Does DRM is still useful without using OpenGL ? Good question. I think so :-) Actually, I believe stuff like video overlays use it. -- |>|\/|< /----------\ |David M. Cooke http://arbutus.physics.mcmaster.ca/cookedm/ |[EMAIL PROTECTED]
Re: Sleeping weirdness (was Re: ibook 2 : several question)
On Tue, Sep 03, 2002 at 07:52:31PM -0400, Francis J. Lacoste wrote: > On lun, 2002-09-02 at 14:53, David M. Cooke wrote: > > > > I had similiar problems, but it's working now. I had to disable IDE > > CDROM support, and compile SCSI emulation in. > > Indeed, this configuration works. Compiling both as modules also works. > > > > > Since several people seem to have problems, I've put the deb for my > > kernel (2.4.20-pre5-ben0), along with the kernel config file and a deb > > for daenzer's drm-trunk-module at > > > > http://arbutus.physics.mcmaster.ca/cookedm/ibook/ > > > > Currently, CDROM support with SCSI emulation, sleeping/waking up, and > > sound work. I haven't tested the CD-RW, but cdrecord -scanbus works. > > > > I've tried your kernel and with it sleeping now works... most of the > time. Once, it didn't wake up and another time it woke up and locked up > after. Actually, that happened to me tonight -- closed the lid, looked like it went to sleep, then, later, turns out it was on and drained the battery. [Might have been me -- I was using it as a pillow, probably not the best of ideas] But, for previous kernels, sleeping would either not work at all, or it would work fine. > Something even weirder is that I can't compile a kernel where sleeping > works. I've compiled a kernel using exactly you configuration file > (using the 2.4.20-pre5-benh0 rsync source) and kernel-package but it > never works. Except that BenH is continually working on his source, so your source is not necessarily my source...[I believe] I'll put up a source package (which I /think/ is the same as what I used) tomorrow, when I get into work. > I noticed that you compiled your kernel using gcc 3.2 pre-release so I > thought it might be the problem. I've compiled the kernel (always using > your configuration and doing a make-kpkg clean between run) using gcc > 2.95, gcc 3.0 and gcc 3.2 and sleeping never works on the produced > kernel. With 2.4.18-something, I noticed that using 2.95 made a difference (it worked as opposed to not), but doesn't seem to now. > A thing that I noticed is that the iBook doesn't seems to sleep > correctly on the failing kernels: the light on the front side of the > iBook doesn't do it's usual sleeping heart beat. It does with the kernel > you compiled. I think when the heart beat doesn't go the iBook is actually off. > I'm including here the packages related to my build environment and also > a diff of the dmesg between your kernel and the last one I've compiled > using 3.2. > > How is your build environment different than mine? Any insights in this > matter are welcome. Your build environment looks the same as mine. -- |>|\/|< /--\ |David M. Cooke http://arbutus.physics.mcmaster.ca/cookedm/ |[EMAIL PROTECTED]
Re: Sleeping weirdness (was Re: ibook 2 : several question)
On Wed, Sep 04, 2002 at 03:49:27PM +0200, Vincent Bernat wrote: > > I have another little problem (but I have not investigated) : now, > when I close the lid, the screen goes off, then off and on again (I > could see this because of the apple). Does someone else get this > behaviour ? Yep. Always have. Don't know why. Seems to be a problem with X (try closing the lid while in the console; didn't happen for me just now). -- |>|\/|< /----------\ |David M. Cooke http://arbutus.physics.mcmaster.ca/cookedm/ |[EMAIL PROTECTED]
Re: Screen poweroff with an ibook
On Wed, Sep 18, 2002 at 03:05:44AM +0200, J. Volkmann wrote: > Hello, > > as I use my laptop in classes, I sometimes don't use it for 20 or more > Minutes. Is there a way to turn the screen off manually perhaps with a > key shortcut? If you turn the screen brightness down to zero with the F1/F2 keys, it will turn the screen off. > My laptop is an ibook2 with the radeon (2x USB, 700 MHz). > > Another thing: Is the ibook able to do a hot-switch of batteries? You mean, without being connected to mains? There's only one battery, so if it isn't getting power from somewhere, it goes off when you remove it. Considering that you can easily get >4 hours off of one battery, I haven't found that to be a problem :-) -- |>|\/|< /------\ |David M. Cooke http://arbutus.physics.mcmaster.ca/cookedm/ |[EMAIL PROTECTED]
Re: xdm v gdm
On Thu, Nov 21, 2002 at 07:18:52PM +0100, Claas Langbehn wrote: > Hi, > > > I've recently installed woody (8100/80) and during the installation > > selected xdm as the window manager. Gnome doesn't seem to like xdm. > Well, that is not true. Here it likes xdm :) > > > How do I switch from xdm to gdm and what is the difference between them? > This is being controlled by the startup-scripts in /etc/rc?.d/ . > Just disable xdm and enable gdm. Read the manpage of "update-rc.d" . easier: dpkg-reconfigure xdm (or gdm, kdm, or wdm) The display managers in Debian cooperate on choosing the default. -- |>|\/|< /----------\ |David M. Cooke http://arbutus.physics.mcmaster.ca/cookedm/ |[EMAIL PROTECTED]
Re: Does sungem support MII?
On Sat, Feb 08, 2003 at 05:31:14PM +0100, ?ukasz Studzi?ski wrote:
> I would like to use laptop-net package on ibook3 (Nov2002), but I am not
> quite sure if its NIC (which is sungem) support MII (Media Independent
> Interface). I had some trouble with network when a driver was configured
> as a module.
> I move the computer to many locations, so since suspending works with no
> problem now I would like to automatize network configuring. Maybe some
> of you have some experience on this subject?
I do this in /etc/power/pwrctl-local, by testing for the network when
the laptop wakes up:
function is_network_up () {
timeout=${1-10}
# get default gateway machine
gateway=`ip route list | awk '/^default via [0-9.]+/{print $3}'`
if [ -n "$gateway" ] ; then
# ping for a maximum of $timeout s, until at least one packet is
returned
ping -w${timeout} -c1 -n $gateway > /dev/null
return $?
else
return 1
fi
}
case $1 in
wakeup)
if ! is_network_up ; then
ifdown eth0 ppp0
ifup eth0 ppp0
fi
;;
esac
Then, I do magic in /etc/network/interfaces (which controls how ifup and
ifdown do their stuff). I use guessnet to choose the right network,
which can then be mapped to how to set it up (using dhcp, static IP,
etc.).
This works quite well for me.
--
|>|\/|<
/----------\
|David M. Cooke http://arbutus.physics.mcmaster.ca/dmc/
|[EMAIL PROTECTED]
Re: ibook trackpad
On Sun, May 11, 2003 at 09:00:13AM +1200, David Friggens wrote: [clipped - talking about trackpad and sensitivity] > It worked fine when I lowered the sensitivity in KDE. I assume there's > some X setting for outside of KDE but I haven't gotten around to > looking in to it just yet. Hopefully someone else has done so and can > save us the search ... :-) > > David I assume KDE (and GNOME for that matter) do the exact same thing with respect to mouse settings as the 'xset' command does [1]. xset is standard with X. [1] except for button mappings (like right/left-handed mouse). You'll have to use xmodmap for that. Usage: $ xset q shows you current settings (not just mouse; keyboard, font path, etc.). $ xset show usage. From it: To set mouse acceleration and threshold: m [acc_mult[/acc_div] [thr]]m default My current settings, for instance, would be set with $ xset m 2/1 4 Fiddle with that until you're happy, and put it in your ~/.xsession -- |>|\/|< /----------\ |David M. Cooke http://arbutus.physics.mcmaster.ca/dmc/ |[EMAIL PROTECTED]
Re: Newbie problems
On Wed, May 14, 2003 at 09:05:56AM -0400, Jule Slootbeek wrote: > Hmm when i put in an audio CD and open KsCD it refuses to play, could it have > anything to do with the fact hat i disabled artsd at startsup? since it gave > me this error? On the ibook2, there is no direct analog connection between the CDROM and the sound hardware (on a PC, say, there is a four-wire cable running from the CDROM to the sound card). You'll have to use something that reads the CD digitially, such as xmms-cdread. AFAIK, that's the only thing in Debian that will do this (unless you cobble something together with cdda2wav). -- |>|\/|< /------\ |David M. Cooke http://arbutus.physics.mcmaster.ca/dmc/ |[EMAIL PROTECTED]
Re: python on powerpc
On Mon, Jun 16, 2003 at 11:03:34AM -0400, christophe barbe wrote: > Hi, > > I was wondering if you shared my impression that most python programs > are partially broken on powerpc. The last three programs written in > python that I have tried are gdeskcal, straw and eroaster. > The first one has significant display problems (the purpose of this program > is to display a calendar in the background of the desktop). The second one > is unable to fetch new blog entries (again the main purpose of this > soft). And the third one, a burner GUI doesn't display the burning > progress. "Most" != 3. "Most" python programs I use under powerpc, I have no problems with. Some fiddling with numbers from apt-cache shows ~70 packages that depend on python2.2 that aren't python modules. Note that the three programs you mention share also python2.2-gtk2 as a dependency; I'd look there for GUI problems. The corruption in gdeskcal is likely an endian issue (which is not an uncommon problem when programs are ported from x86 to ppc). > I have not tested but I guess at least gdeskcal and straw work well on > x86. Bold statement :) [actually, they do.] -- |>|\/|< /------\ |David M. Cooke http://arbutus.physics.mcmaster.ca/dmc/ |[EMAIL PROTECTED]
Re: python on powerpc
On Tue, Jun 17, 2003 at 10:23:46AM -0400, christophe barbe wrote: > On Mon, Jun 16, 2003 at 11:45:13PM -0400, David M. Cooke wrote: > > On Mon, Jun 16, 2003 at 11:03:34AM -0400, christophe barbe wrote: > > > Hi, > > > > > > I was wondering if you shared my impression that most python programs > > > are partially broken on powerpc. The last three programs written in > > > python that I have tried are gdeskcal, straw and eroaster. > > > > "Most" != 3. "Most" python programs I use under powerpc, I have no > > problems with. Some fiddling with numbers from apt-cache shows ~70 > > packages that depend on python2.2 that aren't python modules. > > "Most" != "impression that most" > And comparing 'most' to a number is just plain stupid. As you point out, > ~70 packages depend on python2.2 (including dia, gnumeric, ... that are > NOT written in python) and 3 is a significant subset of it. A bit more drilling gives 35 packages that depend on python2.2, w/o python in the package name, and don't depend on libc6. There are ~3 python libraries in that list. 3 is still much smaller than 35 :-) > > Note that the three programs you mention share also python2.2-gtk2 as a > > dependency; I'd look there for GUI problems. The corruption in gdeskcal > > is likely an endian issue (which is not an uncommon problem when > > programs are ported from x86 to ppc). > > Interesting. Do you really mean that software written in an interpreted > language need to be ported? Or you just don't know what you are talking > about? I don't think you can program anything (large and) interesting in a pure cross-platform manner -- at some point you're making some assumption about the system (just read the os module documentation for the "Availability" comments). But pure python running on two different linux systems should be fine; although there are still places the different architectures make an appearance. For instance, the struct module, and I've been bitten by pickle (a while ago) when moving a binary pickle from an x86 to an Alpha. C extensions (like python2.2-gtk2) are a different matter from pure Python -- they'll have the same problems that regular C programs have when moving between architectures. Endian issues seems to bite the most with programs that manipulate images or audio. > straw network related problem is not caused by python2.2-gtk2, neither > is eroaster (parsing bug). But straw i18n support is broken and this is > certainly related to python2.2-gtk2. Looking at eroaster's bugs, it appears it's a problem on x86 also. -- |>|\/|< /--\ |David M. Cooke http://arbutus.physics.mcmaster.ca/dmc/ |[EMAIL PROTECTED]
Re: mol
On Sun, Jun 22, 2003 at 12:45:45AM -0500, [EMAIL PROTECTED] wrote: > Hi! > > if i want to run mol as any user it tells me that the program most be > suid, so: > > $chmod u+s startmol (link and target) > > but steal said me the same. Any ideas? # chmod u+x /usr/lib/mol/bin/mol -- |>|\/|< /----------\ |David M. Cooke http://arbutus.physics.mcmaster.ca/dmc/ |[EMAIL PROTECTED]
Re: mol
On Sun, Jun 22, 2003 at 12:17:35PM +0200, Jens Schmalzing wrote: > Hi, > > David M. Cooke writes: > > > > if i want to run mol as any user it tells me that the program most > > > be suid, so: > > > # chmod u+x /usr/lib/mol/bin/mol > > Nah. First, you probably meant u+s, and second, any changes you make to > files from Debian packages without telling dpkg will be clobbered on > the next upgrade. To let anybody run mol, do > > # dpkg-statoverride --update --add root root 4755 /usr/lib/mol/bin/mol Whoops, you're right. And looking into my setup, that's what I had done :) > Then again, you may not really want to allow anybody to run mol, but > only a few selected users (to use the example from upstream, I will > call them samuel and tux). In addition to adding allow and deny lines > to /etc/mol/session.map, do the following: > > # addgroup --system mol > # adduser samuel mol > # adduser tux mol > # dpkg-statoverride --update --add root mol 4710 /usr/lib/mol/bin/mol You can also edit /etc/mol/session.map and add your users there. This sgid approach would be more secure, though. -- |>|\/|< /--\ |David M. Cooke http://arbutus.physics.mcmaster.ca/dmc/ |[EMAIL PROTECTED]
Re: WARNING: libc6 in unstable
On Sat, Jul 12, 2003 at 11:03:32AM -0700, Chris Tillman wrote: [...] > Since re-installing the base system overwrites the status file, my > machine had no idea what software had been installed. I wonder if a > utility for apt would be useful to reconstruct the status file from > what's on disk. Does it overwrite the backup of dpkg.status in /var/backups? (There's a cron job that backs up key config files every day). -- |>|\/|< /----------\ |David M. Cooke http://arbutus.physics.mcmaster.ca/dmc/ |[EMAIL PROTECTED]
Re: ibook2
On Wed, Jul 30, 2003 at 05:06:00PM +0200, Vincent Bernat wrote: > OoO Vers la fin de l'après-midi du mercredi 30 juillet 2003, vers > 16:21, Michel Dänzer <[EMAIL PROTECTED]> disait: [...] > > Many thanks again for your help, having a working clone mode and > without the need to reboot will be very useful ! That was a long thread -- could you summarise your final solution? I got lost :-) -- |>|\/|< /----------\ |David M. Cooke http://arbutus.physics.mcmaster.ca/dmc/ |[EMAIL PROTECTED]
Re: swiss german keyboard config needed (xev results)
On Tue, Aug 19, 2003 at 10:33:17PM +0200, Michel Dänzer wrote: > On Tue, 2003-08-19 at 22:22, Frank Murphy wrote: > > > > It seems that we need a new key for Multi_key as well. Any ideas? > > I proposed earlier what I've been using for a while: > > 108 0xff7e (Mode_switch)0xff20 (Multi_key) > > Or does anybody _really_ need that stupid enter key? :) Makes a great mouse button. -- |>|\/|< /----------\ |David M. Cooke http://arbutus.physics.mcmaster.ca/dmc/ |[EMAIL PROTECTED]
Re: Kernel build, auto-off
On Sun, Sep 21, 2003 at 11:16:04PM +0200, Paul van Tilburg wrote: > Hi! > > Two short questions of a TiBook 15" user, hope anyone can help. > [question I can't help with snipped] > * I've configured pmud not to take action on lid close, since that's what > I'm used to (close lid to save screen and prevent peeking, walk away, > come back after 20mins). But it seems then when AC is offline, my > TiBook turns itself off 5mins after the lid closed! My OS, the fs and I > obviously don't like that... But can't find how to turn that off, is it > possible anyway? Are you also using pbbuttonsd? By default, it replaces pmud in handling power (I thought this was rather high-handed when I found out; why else would I be running pmud?!!). Check /etc/pbbuttonsd.conf (IIRC), and look for a something pmud option. (Don't have my laptop handy so I can't check.) -- |>|\/|< /------\ |David M. Cooke http://arbutus.physics.mcmaster.ca/dmc/ |[EMAIL PROTECTED]
