On Mon, Jan 24, 2000, Kevin Puetz <[EMAIL PROTECTED]> wrote:
>> That's understood - I was looking for a way to detect potential input
>> devices hooked up to USB even when there's no OF supported USB. But I
>> figure that's a rare corner case, we should better focus on handling
>> the primary inpu
[EMAIL PROTECTED] said:
> That's understood - I was looking for a way to detect potential input
> devices hooked up to USB even when there's no OF supported USB. But I
> figure that's a rare corner case, we should better focus on handling
> the primary input devices properly.
Fine with me, as lo
"der.hans" wrote:
> will show you the exact output of thescript between the
> parens. Presumably this include a newline, however. If that's the case do
> "thescript >outfile", then vi the outfile and do ":set list", that puts a
> $ at the end of the line, e.g. "Arch ist: powerppc/mac $" means the
"der.hans" wrote:
> will show you the exact output of thescript between the
> parens. Presumably this include a newline, however. If that's the case do
> "thescript >outfile", then vi the outfile and do ":set list", that puts a
> $ at the end of the line, e.g. "Arch ist: powerppc/mac $" means the
On Sun, 23 Jan 2000, Nelson Abramson wrote:
> Okay, well, it gives me the same error. When I run it by itself, it
> spits out "Arch ist: powerppc/mac", but I couldn't tell you if there
> was an extra space at the end. But when I stick it in
> console-data.config, it just gives me the same error.
Hartmut Koptein wrote:
> >
> > Thanks for posting it, but it still fails on me :( Atleast it's a
> > different error: no keymaps described
> > for architecture powerpc/mac
> i assume this ^^^ is correct!?
yeah, i'm on a powerppc/mac.
>
> > Was there another part of the source I w
> > Replace this part in the console-data.config file.
> > (/var/lib/dpkg/info/console-data.config).
>
> Thanks for posting it, but it still fails on me :( Atleast it's a different
> error: no keymaps described
> for architecture powerpc/mac
i assume this ^^^ is correct!?
Hartmut Koptein wrote:
> [snip]
> Replace this part in the console-data.config file.
> (/var/lib/dpkg/info/console-data.config).
Thanks for posting it, but it still fails on me :( Atleast it's a different
error: no keymaps described
for architecture powerpc/mac
Was there another part of the
> > That's right, does this card show up on /proc/pci ? My builtin USB
> > shows up ...
>
> Yes, it does show up in /proc/pci. It also has /dev nodes (that I created by
> hand). But it's not going to show up in the /proc/device-tree, as OF knows
> nothing about it.
That's understood - I was lo
> > Seems Hartmut found another way, no reason to worry.
>
> Sorry, not for usb; the architecture/subarch detections was broken for
> console-data. The console-* sets the kbd-layout, not the device from /dev
> (adb,
> ps2 or usb).
That still leaves the problem of newer Apple machines with USB
> > usb is not available on every powerpc machine. We fixed console-data
> > yesterday, lets hope this will doit what we want for potato.
>
> Could you please post the definitive fix? I tried the fix posted yesterday,
> and
> it still gave me errorsor are you just going to update the consol
On Fri, Jan 21, 2000, Kevin Puetz <[EMAIL PROTECTED]> wrote:
>Yes, it does show up in /proc/pci. It also has /dev nodes (that I created by
>hand). But it's not going to show up in the /proc/device-tree, as OF knows
>nothing about it.
>
/proc/device-tree should, with a few rare exceptions, show
[EMAIL PROTECTED] said:
> This would be a poor test - I have a 3rd party USB card in my beige
> G3, and it
> works fine. However, my OF doesn't know about it, and there are no
> such
> devices.
> That's right, does this card show up on /proc/pci ? My builtin USB
> shows up ...
Yes, it does sh
Hi,
> That's right, does this card show up on /proc/pci ? My builtin USB shows
> up ...
>
> > So at least leave a way to bypass whatever suggestion is going to be made
> > because of this test.
>
> Seems Hartmut found another way, no reason to worry.
Sorry, not for usb; the architecture/subar
Hartmut Koptein wrote:
> usb is not available on every powerpc machine. We fixed console-data
> yesterday, lets hope this will doit what we want for potato.
Could you please post the definitive fix? I tried the fix posted yesterday, and
it still gave me errorsor are you just going to update
> [EMAIL PROTECTED] said:
> > The absence of the USB option would be indicated by the absence of the
> > usb-* entries in /proc/device-tree/aliases (the fact that there just
> > isn't any USB hardware actually connected at boot time results in
> > empty files there).
>
> This would be a poor tes
[EMAIL PROTECTED] said:
> The absence of the USB option would be indicated by the absence of the
> usb-* entries in /proc/device-tree/aliases (the fact that there just
> isn't any USB hardware actually connected at boot time results in
> empty files there).
This would be a poor test - I have a
> > More info in /proc/device-tree/aliases (see adb-mouse, adb-keyboard,
> > usb-mouse, usb-keyboard; empty files mean device not present).
> > Hartmut: which of these would break on PPC models other than Powermacs?
>
> usb is not available on every powerpc machine. We fixed console-data
> yest
> /proc/hardware has a lot more information in it than /proc/pci, on
> my powerpc/apus system. but all info from /proc/cpuinfo seems to be
> included into /proc/hardware.
Same on m68k. The function for /proc/hardware in the kernel calls the
one for /proc/cpuinfo.
Roman
On Fri, Jan 21, 2000 at 09:47:50AM +0100, Michael Schmitz wrote:
> > > Configuring packages ...
> > > Please report this bug: no keymaps described for architecture m68k/
> > amiga 2000
> >
> > This is a known problem. Fixing it as well.
> >
> > So now it seems that /proc/hardware on ppc
> No machine: line here. I suggest the PowerPC port add /proc/hardware to
> the kernel (the machine: stuff isn't strictly info about the CPU). Or read
> /proc/device-tree/model for the model ID (on PowerMac models at least).
> More info in /proc/device-tree/aliases (see adb-mouse, adb-keyboard,
> u
> > Configuring packages ...
> > Please report this bug: no keymaps described for architecture m68k/ amiga
> 2000
>
> This is a known problem. Fixing it as well.
>
> So now it seems that /proc/hardware on ppc is not useful. Is there
> also a "machine" field on m68k ? That would be nic
OK, I borrowed a machine for tonight (mine is still down). I'll try
to close as much issues as possible. If any of you is online right
now (21:30 GMT+1) and willing to test live before the upload, just
tell me.
Hartmut Koptein writes:
> Powerc and m68k need some tests, i've not changed any of
On Wed, Jan 19, 2000 at 02:37:14PM +0100, Christian T. Steigies wrote:
> Moin Haggie,
> On Wed, Jan 19, 2000 at 01:12:22PM +0100, Hartmut Koptein wrote:
> > as you all may know, the console-data arch-detection routine is broken.
> >
> > Powerc and m68k need some tests, i've not changed any of the
Moin Haggie,
On Wed, Jan 19, 2000 at 01:12:22PM +0100, Hartmut Koptein wrote:
> as you all may know, the console-data arch-detection routine is broken.
>
> Powerc and m68k need some tests, i've not changed any of the m68k code, it
> is the source from console-data.config.
what do you want us to d
Hi all,
as you all may know, the console-data arch-detection routine is broken.
Powerc and m68k need some tests, i've not changed any of the m68k code, it
is the source from console-data.config.
For the tests:
#!/usr/bin/perl -w
my $arch = `dpkg --print-architecture`;
$arch =~ s/\n//;
[EMAIL PROTECTED] said:
> > I think that usb-keyboard have the same layout as adb-keyboards,
> its
> > only a different connector. Same as serial and ps/2.
To my knowledge, this is due to the kernel translating the USB keycodes to
ADB-style ones, so that all the old keymaps work. I don't t
Hartmut Koptein writes:
> > > The debian arch-name is powerpc, the kernel name it 'ppc'.
> > > apus, chrp, pmac, prep are the Subarchs.
> >
> > I only use "ppc" as a column-saver here ;)
>
> Thats your choice, but any (external) reference is 'powerpc' as for example
> /usr/lib/perl5/
On Mon, 22 Nov 1999, Geert Uytterhoeven wrote:
> On Mon, 22 Nov 1999, Juri Jensen wrote:
> > To add to this discussion about keymaps, here's some of my experiences:
> >
> > I'm from Denmark and I'm using both i386 and PPC Debian Linux. Getting a
> > working keymap on the i386 machine hasn't been a
On Mon, Nov 22, 1999 at 06:57:26PM +0100, Geert Uytterhoeven wrote:
> On Mon, 22 Nov 1999, Juri Jensen wrote:
> > To add to this discussion about keymaps, here's some of my experiences:
> >
> > I'm from Denmark and I'm using both i386 and PPC Debian Linux. Getting a
> > working keymap on the i386
Hartmut Koptein writes:
> # powerpc/apusamiga "Model: Amiga"
> # powerpc/chrpi386,mac"Model: CHRP"
> # powerpc/pmacmac "Model:
> PowerMac|[Pp]ower[Bb]ook*|Power|iMac*|PowerMac1*"
> # powerpc/prepi386"Model: PReP"
> # powerpc/{bbox,mbx,ppc64,82xx,8xx}
On Mon, 22 Nov 1999, Juri Jensen wrote:
> To add to this discussion about keymaps, here's some of my experiences:
>
> I'm from Denmark and I'm using both i386 and PPC Debian Linux. Getting a
> working keymap on the i386 machine hasn't been a big problem, but
> getting one on the PPC indeed has; I'
> * Is there no powerpc atari's ?
No.
Roman
Hi,
To add to this discussion about keymaps, here's some of my experiences:
I'm from Denmark and I'm using both i386 and PPC Debian Linux. Getting a
working keymap on the i386 machine hasn't been a big problem, but
getting one on the PPC indeed has; I've tried using the same as on the
i386 result
> > # powerpc/apusamiga "Model: Amiga"
> > # powerpc/chrpi386,mac"Model: CHRP"
> > # powerpc/pmacmac "Model:
> PowerMac|[Pp]ower[Bb]ook*|Power|iMac*|PowerMac1*"
> > # powerpc/prepi386"Model: PReP"
> > # powerpc/{bbox,mbx,ppc64,82xx,8xx} Not yet suppor
On Fri, 19 Nov 1999, Yann Dirson wrote:
> # ppc/mac i386,mac,usb"Model: PowerMac|[Pp]ower[Bb]ook*|Power"
AFAIK Macs cannot have PC (isn't that better than i386?) keyboards
> # ppc/apusi386,mac,amiga "Model: Amiga"
APUS can have an Amiga keyboard only
> # ppc/prep?
Thus spake Geert Uytterhoeven ([EMAIL PROTECTED]):
> If you don't use XKB, XFree86 queries the kernel for the current keymap (it
> traverses all keycodes and modifiers and asks for their mapping). Hence the
> keymap will automagically be correct (assumed it's correct in the console).
>
> That's on
On Fri, Nov 19, 1999 at 10:20:35AM +0100, Christian T. Steigies wrote:
> As you see, I put the @ also on the Q, like on PC keyboard (does that work
> out for french keyboards as well?). There are some other signs which are
Just for you to know, i don't see any @ on my french pc keyboard. NOr
natur
On Thu, Nov 18, 1999 at 11:11:59PM +0100, Yann Dirson wrote:
> Christian T. Steigies writes:
> > Where does X get the information about the keyboard? Its not from
> > the console tools, is it?
>
> Generated XF86Config use an Xkb definition. If none is in XF86Config,
> I guess it uses /etc/X11/X
On Thu, Nov 18, 1999 at 11:16:40PM +0100, Yann Dirson wrote:
> Sven LUTHER writes:
> > Also, why doies the debconf stuff ask questions
> >about the keymap every time a > package is upgraded, i suppose most of
> >the time it will stay the same between > two apt-get upgrade, ...
>
> Because new key
On Thu, Nov 18, 1999 at 11:11:59PM +0100, Yann Dirson wrote:
> Christian T. Steigies writes:
> > Where does X get the information about the keyboard? Its not from
> > the console tools, is it?
>
> Generated XF86Config use an Xkb definition. If none is in XF86Config,
> I guess it uses /etc/X11/X
> # Subarches
> # FIXME: not done
> # Issues:
> # * ppc issues not fully asserted yet
> # * /proc/hardware, which only exists on m68k and ppc, seems to be
> # an OPTION while configuring the kernel !
> # Any fallback when it's not there ?
> # * USB set of keymap should contain what ? mac ? i386
OK, so here is the info I gathered from the various mails I recieved,
as I summarize them in the config script. I did not srite any code
for this for now, as I still miss some info, esp. about ppc boxes.
I'll issue -8 without 68k/ppc support - there are many bugs to fix in
the meantime.
Please t
Joel Klecker writes:
> Loading /usr/share/keymaps/i386/qwerty/gr.kmap.gz
> findfile(): timeout waiting for undead child(ren) ?
Ha yes, this is surely a bug in console-tools. My own code. :(
Submitted through BTS so that I think of investigating.
> Loading /usr/share/keymaps/mac/mac-de-latin1
Christian T. Steigies writes:
> Where does X get the information about the keyboard? Its not from
> the console tools, is it?
Generated XF86Config use an Xkb definition. If none is in XF86Config,
I guess it uses /etc/X11/Xmodmap (that is, it uses the old mechanism
instead of newer Xkb).
Sven
Sven LUTHER writes:
> Also, why doies the debconf stuff ask questions
>about the keymap every time a > package is upgraded, i suppose most of
>the time it will stay the same between > two apt-get upgrade, ...
Because new keymaps have been added - the user must be given a chance
to select one of t
> If you don't use XKB, XFree86 queries the kernel for the current keymap (it
> traverses all keycodes and modifiers and asks for their mapping). Hence the
> keymap will automagically be correct (assumed it's correct in the console).
>
> That's one reason to not use XKB :-)
That's the theory, b
On Thu, 18 Nov 1999, Sven LUTHER wrote:
> On Tue, Nov 16, 1999 at 09:40:44PM +0100, Yann Dirson wrote:
> > > A, somehow, related question. How to get a correct keymap under X?
> >
> > Hm, X keymap setup is currently quite a distinct thing. I hope to add
> > one day support for reading Xkb defs i
> > I seem to remember some stuff about X using the same keymap as the console,
> > but i am not sure anymore about it. I did not do any config file editing
> > when
> > i installed X and it took my azerty keymap without major trouble (just the
> > Delete/BackSpace stuff, but i guess it is not key
On Thu, Nov 18, 1999 at 03:19:03PM +0100, Sven LUTHER wrote:
> On the french keymap, the @ and # are mapped to the shifted * and - of the
> keypad, i think. I painted the corresponding symbols on said keys, and i
> worked fine since then.
Excellant idea, so instead of finding a xmodmap entry we wil
On Thu, Nov 18, 1999 at 02:02:59PM +0100, Christian T. Steigies wrote:
> On Thu, Nov 18, 1999 at 12:35:33PM +0100, Sven LUTHER wrote:
> > On Tue, Nov 16, 1999 at 09:40:44PM +0100, Yann Dirson wrote:
> > > > A, somehow, related question. How to get a correct keymap under X?
> > >
> > > Hm, X keyma
On Thu, Nov 18, 1999 at 12:35:33PM +0100, Sven LUTHER wrote:
> On Tue, Nov 16, 1999 at 09:40:44PM +0100, Yann Dirson wrote:
> > > A, somehow, related question. How to get a correct keymap under X?
> >
> > Hm, X keymap setup is currently quite a distinct thing. I hope to add
> > one day support f
On Tue, Nov 16, 1999 at 09:40:44PM +0100, Yann Dirson wrote:
> > A, somehow, related question. How to get a correct keymap under X?
>
> Hm, X keymap setup is currently quite a distinct thing. I hope to add
> one day support for reading Xkb defs in console-tools, but there are
> so much things to
On Tue, Nov 16, 1999 at 03:29:09PM -0800, Joel Klecker wrote:
> At 22:02 +0100 1999-11-16, Hartmut Koptein wrote:
> >> And what about CHRP, PreP ? I guess some of them may only exist
> >> as serial-console ?
> >>
> > > #elif #cpu (powerpc)
> > > struct d_choices keymaps[]={
> > > { "i38
On Tue, Nov 16, 1999 at 08:56:43AM +0100, Hartmut Koptein wrote:
> > As you may have seen from recent bug reports ;), I'm currently adding
> > support for debconf-driven keymap choice to console-data and friends,
> > but need the following help for ppc/m68k machines:
> >
> > * what sub-architectur
On Mon, Nov 15, 1999 at 10:36:12PM +0100, Yann Dirson wrote:
> Hello,
>
> As you may have seen from recent bug reports ;), I'm currently adding
> support for debconf-driven keymap choice to console-data and friends,
> but need the following help for ppc/m68k machines:
>
> * what sub-architectures
On Wed, 17 Nov 1999, Roman Hodek wrote:
> > Anyone knows about what keyboards BVME* use ? Same as MVME ?
>
> I think so, yes.
Do these boxes have a keyboard interface? Aren't they just rack-mounted boards
with serial consoles?
> > ... ??! Do all PPC's have any use of i386 keyamps ? Or only MVME
> Hm... I'd prefer to keep layout as "qwerty", "azerty" and such :)
> Subarches will just be arches as far as I'm concerned.
Ah... The layout is IMHO country specific, i.e. US maps should always
be qwerty, German ones qwertz and French azerty, and so on... This is
also true for m68k machines :-)
At 22:02 +0100 1999-11-16, Hartmut Koptein wrote:
And what about CHRP, PreP ? I guess some of them may only exist
as serial-console ?
> #elif #cpu (powerpc)
> struct d_choices keymaps[]={
> { "i386/qwerty/us", "U.S. English (QWERTY)" , KBD_C},
...
> { "mac/mac-de-latin
> And what about CHRP, PreP ? I guess some of them may only exist
> as serial-console ?
>
> #elif #cpu (powerpc)
> struct d_choices keymaps[]={
> { "i386/qwerty/us", "U.S. English (QWERTY)" , KBD_C},
...
> { "mac/mac-de-latin1-nodeadkeys", "Germany (Mac)" , 0},
> { "m
Roman Hodek writes:
> > * meaningful descriptions for those keymaps I do not use... For i386,
> >I classified them by Layout/Country_name/Variant_name.
>
> This should be easy done for the m68k maps, too. The layout is the
> subarch, the country name is obvious, and variants (nodeadkeys
On Tue, Nov 16, 1999 at 08:56:43AM +0100, Hartmut Koptein wrote:
> >
> > * what sub-architectures are there for each arch ?
> > My guess: m68k -> atari, amiga, sun, next(, mac) ; ppc -> mac, RS/6000
>
> m68k: amiga, atari, mac, bvme*
> powerpc: apus, chrp, pmac, prep (mbx)
APUS is a
> As you may have seen from recent bug reports ;), I'm currently adding
> support for debconf-driven keymap choice to console-data and friends,
> but need the following help for ppc/m68k machines:
>
> * what sub-architectures are there for each arch ?
> My guess: m68k -> atari, amiga, sun, next(
Well, I forgot an important point that will be needed as well:
* meaningful descriptions for those keymaps I do not use... For i386,
I classified them by Layout/Country_name/Variant_name.
--
Yann Dirson<[EMAIL PROTECTED]> |Why make M$-Bill richer & richer ?
debian-email: <[EMAIL PR
Hello,
As you may have seen from recent bug reports ;), I'm currently adding
support for debconf-driven keymap choice to console-data and friends,
but need the following help for ppc/m68k machines:
* what sub-architectures are there for each arch ?
My guess: m68k -> atari, amiga, sun, next(, ma
65 matches
Mail list logo
