Am Donnerstag, 21. Oktober 2004 20:15 schrieb Roland Scheidegger:
> Jon Smirl wrote:
> > I fixed up every error path that I can think of in the radeon i2c
> > code. It's checked into CVS. Let me know if I've fixed the problem. I
> > still haven't figure out how to reproduce it so I'm just fixing
>
Jon Smirl wrote:
I fixed up every error path that I can think of in the radeon i2c
code. It's checked into CVS. Let me know if I've fixed the problem. I
still haven't figure out how to reproduce it so I'm just fixing
everything that could possibly be wrong.
Yes, that seems to have fixed it. I can n
I fixed up every error path that I can think of in the radeon i2c
code. It's checked into CVS. Let me know if I've fixed the problem. I
still haven't figure out how to reproduce it so I'm just fixing
everything that could possibly be wrong.
Don't use the new radeonfb with it, it is likely broken
Roland Scheidegger wrote:
Jon Smirl wrote:
I checked in a fix for freeing uninitialized i2c channels.
Works fine (tested with drm-core), thanks.
Actually, it doesn't.
I can rmmod the radeon module just fine now, but it causes oopses when
removing other i2c drivers.
Oct 21 11:52:44 ZakTower kernel
>
> Do you mean the difference between linux-core and linux-2.6 (and related) ?
> One of the differences is that linux-core contains R300 microcode, so the
> version number was bumped to make checking for this easy.
I'll fix up the older one in the next day or two.. I'm trying to get some
time to
Vladimir Dergachev wrote:
On Wed, 20 Oct 2004, Roland Scheidegger wrote:
Jon Smirl wrote:
I checked in a fix for freeing uninitialized i2c channels.
Works fine (tested with drm-core), thanks.
I just noticed though that the radeon minor version of the core and
non-core version is not the same, whi
On Wed, 20 Oct 2004, Roland Scheidegger wrote:
Jon Smirl wrote:
I checked in a fix for freeing uninitialized i2c channels.
Works fine (tested with drm-core), thanks.
I just noticed though that the radeon minor version of the core and non-core
version is not the same, which is irritating (well I n
Jon Smirl wrote:
I checked in a fix for freeing uninitialized i2c channels.
Works fine (tested with drm-core), thanks.
I just noticed though that the radeon minor version of the core and
non-core version is not the same, which is irritating (well I need to
add a new packet for emitting, and thus
I checked in a fix for freeing uninitialized i2c channels.
--
Jon Smirl
[EMAIL PROTECTED]
---
This SF.net email is sponsored by: IT Product Guide on ITManagersJournal
Use IT products in your business? Tell us what you think of them. Give us
Yo
On Wed, 20 Oct 2004 14:18:58 +0200, Roland Scheidegger
<[EMAIL PROTECTED]> wrote:
> I've just noticed though I can't rmmod the radeon module (neither the
> old linux-2.6 version nor drm-cor), maybe it's trying to unregister the
> i2c bus it couldn't register previously?
I need to add code to track
Jon Smirl wrote:
I checked in a fix that switches from
dev->pdev->driver->name
to
drm->driver->pci_driver.name
This should fix the segfaults in stealth mode when the pci driver is
set to the wrong driver (fbdev) or no driver (vesafb).
Please give it a try.
Yes, that works here (with vesafb).
I've
I checked in a fix that switches from
dev->pdev->driver->name
to
drm->driver->pci_driver.name
This should fix the segfaults in stealth mode when the pci driver is
set to the wrong driver (fbdev) or no driver (vesafb).
Please give it a try.
--
Jon Smirl
[EMAIL PROTECTED]
-
On Tue, 19 Oct 2004 12:24:23 -0400, Jon Smirl <[EMAIL PROTECTED]> wrote:
> On Tue, 19 Oct 2004 18:14:32 +0200, Roland Scheidegger
> <[EMAIL PROTECTED]> wrote:
> > > Another posibility is that the monid port on your card is just
> > > broken, but you never noticed since you don't have a monitor plug
On Tue, 19 Oct 2004 18:14:32 +0200, Roland Scheidegger
<[EMAIL PROTECTED]> wrote:
> > Another posibility is that the monid port on your card is just
> > broken, but you never noticed since you don't have a monitor plugged
> > in that needs it.
> What exactly are the different ports good for? Gettin
Roland Scheidegger wrote:
Jon Smirl wrote:
A fix is in CVS now to make missing/broken I2C buses a non-fatal
error. Fixed in linux-2.6 and linux-core.
Ok, thanks. That gets around that problem, though it still doesn't work
(linux-core, I've not yet tried the linux-2.6 version).
Forgot to mention,
Jon Smirl wrote:
A fix is in CVS now to make missing/broken I2C buses a non-fatal
error. Fixed in linux-2.6 and linux-core.
Ok, thanks. That gets around that problem, though it still doesn't work
(linux-core, I've not yet tried the linux-2.6 version).
When the X server starts and vesafb is active,
Jon Smirl wrote:
On Tue, 19 Oct 2004 03:12:36 +0200, Roland Scheidegger
<[EMAIL PROTECTED]> wrote:
Vladimir Dergachev wrote:
On Tue, 19 Oct 2004, Roland Scheidegger wrote:
Jon Smirl wrote:
Does the new radeonfb driver in the kernel load? It uses the
same I2C initialization code.
excuse my ignora
A fix is in CVS now to make missing/broken I2C buses a non-fatal
error. Fixed in linux-2.6 and linux-core.
--
Jon Smirl
[EMAIL PROTECTED]
---
This SF.net email is sponsored by: IT Product Guide on ITManagersJournal
Use IT products in your busi
On Tue, 19 Oct 2004 03:12:36 +0200, Roland Scheidegger
<[EMAIL PROTECTED]> wrote:
> Vladimir Dergachev wrote:
> > On Tue, 19 Oct 2004, Roland Scheidegger wrote:
> >> Jon Smirl wrote:
> >>> Does the new radeonfb driver in the kernel load? It uses the same I2C
> >>> initialization code.
> >>
> >> exc
I fixed this segfault. It was an error path that hadn't been hit before.
It is checked into CVS.
On Tue, 19 Oct 2004 02:43:02 +0200, Roland Scheidegger
<[EMAIL PROTECTED]> wrote:
> Jon Smirl wrote:
> > Does the new radeonfb driver in the kernel load? It uses the same I2C
> > initialization code.
>
On Tue, 19 Oct 2004 03:02:45 +0200, Roland Scheidegger
<[EMAIL PROTECTED]> wrote:
> Jon Smirl wrote:
> > What Radeon chip is is it? Maybe it doesn't have an i2c port on it. It
> > is the i2c conflict that is causing the module load to fail.
> It's a rv250 (radeon 9000pro).
I have a radeon 9000pro
Vladimir Dergachev wrote:
On Tue, 19 Oct 2004, Roland Scheidegger wrote:
Jon Smirl wrote:
Does the new radeonfb driver in the kernel load? It uses the same I2C
initialization code.
excuse my ignorance, but where would I get that new radeonfb driver?
Try latest 2.6.x kernel. You might need to app
On Tue, 19 Oct 2004, Roland Scheidegger wrote:
Jon Smirl wrote:
Does the new radeonfb driver in the kernel load? It uses the same I2C
initialization code.
excuse my ignorance, but where would I get that new radeonfb driver?
Try latest 2.6.x kernel. You might need to apply the rc4 patch.
There is a
Jon Smirl wrote:
Does the new radeonfb driver in the kernel load? It uses the same I2C
initialization code.
excuse my ignorance, but where would I get that new radeonfb driver?
btw I've also tried the linux-core version. Just the same:
Oct 19 02:26:26 ZakTower kernel: [drm] Initialized drm 1.0.0 20
Does the new radeonfb driver in the kernel load? It uses the same I2C
initialization code.
--
Jon Smirl
[EMAIL PROTECTED]
---
This SF.net email is sponsored by: IT Product Guide on ITManagersJournal
Use IT products in your business? Tell us wh
What Radeon chip is is it? Maybe it doesn't have an i2c port on it. It
is the i2c conflict that is causing the module load to fail.
The code knows about VESAfb and works around it so that should be ok.
I believe both the new and old radeonfb drivers should work too.
--
Jon Smirl
[EMAIL PROTECTED
I don't load fglrx at the same time of course, I just want to be able to
load it from time to time (it used to garble the framebuffer console
completely if you used radeonfb, might even have caused lockups, can't
remember exactly).
I can of course try radeonfb instead of vesa with the new radeon
you can't have the fglrx driver loaded with the newer DRM's. They are
both trying to control the same piece of hardware.
On Mon, 18 Oct 2004 22:52:45 +0200, Roland Scheidegger
<[EMAIL PROTECTED]> wrote:
> Jon Smirl wrote:
> > Do you have the new radeon framebuffer driver loaded? If so, try
> > loa
Jon Smirl wrote:
Do you have the new radeon framebuffer driver loaded? If so, try
loading the old one. Let me know and I'll check it out more here.
No, I'm using the old vesa framebuffer (compiled in). I'll have radeonfb
compiled as a module, but it's not loaded.
(I'm basically using vesa due to e
On Mon, 18 Oct 2004 22:05:17 +0200, Roland Scheidegger
<[EMAIL PROTECTED]> wrote:
> I can't get the drm module (radeon.ko) from cvs, linux-2.6 directory to
> work. I'm not sure since when it no longer works, I haven't updated it for ages...
You should also update to current cvs there have been bug
Do you have the new radeon framebuffer driver loaded? If so, try
loading the old one. Let me know and I'll check it out more here.
--
Jon Smirl
[EMAIL PROTECTED]
---
This SF.net email is sponsored by: IT Product Guide on ITManagersJournal
Use
I can't get the drm module (radeon.ko) from cvs, linux-2.6 directory to
work.
I'm not sure since when it no longer works, I haven't updated it for ages...
Starting Xorg shows up in the kernel log as follows:
Oct 18 21:47:48 ZakTower kernel: mtrr: 0xe800,0x400 overlaps
existing 0xe800
32 matches
Mail list logo
