lubuntu-qa team mailing list archive

[Paul Gorski <paul@xxxxxxxxxxx>]

 As mentioned previously, startx via at system prompt following G3 imac live cd boot failure yields a:

fglrx error, failure to load module.

This text from th fglrx.py file on my eMac

"def available(self):
        # we don't offer fglrx in a life CD environment, as we will run out of
        # RAM trying to download and install all the packages in the RAM disk.
        if os.path.isdir('/rofs'):
            logging.debug('Disabling fglrx driver on live system')
            return False

Yet, the Lubuntu live CD is trying to load fglrx.
As a matter of fact, via the error log it is trying to load:

fglrx (0)
ati (1)
r128 (2) 

Perhaps we could take the fglrx out of that instal cycle?



 

Sincerely,
Paul Gorski
 

 

-----Original Message-----
From: Phill Whiteside <PhillW@xxxxxxxxxx>
To: Jim Price <d1version@xxxxxxxxxxx>
Cc: lubuntu-qa <lubuntu-qa@xxxxxxxxxxxxxxxxxxx>
Sent: Tue, Mar 6, 2012 6:27 pm
Subject: Re: [Lubuntu-qa] Mac G3 testing - desktop .iso


most interesting,


the last long standing bug was when Apple renamed themselves internally on the system, so the the install routine failed to spot the new name! This may have some milage, as they patched for that one on the installer a couple of years ago :)


Do have a play with it, we have a far better chance of a fix if we can say what is needed to fix it :)


regards,


Phill.


On 7 March 2012 00:13, Jim Price <d1version@xxxxxxxxxxx> wrote:

Some more details:

I tried to identify the graphics hardware on the iMac G3 using dmesg (which is quick compared to something like lshw) and got the following:

$ dmesg |grep -i rage
[    0.654002] Using unsupported 1024x768 ATY,Rage128k at 94008000, depth=8, pitch=1024
[    0.686744] fb0: Open Firmware frame buffer device on /pci@f0000000/ATY,Rage128k@10

This looks to me like ATY has been used instead of ATI! I need to investigate further to see what the implications of this are, but my guess at this point is that the second of those two lines would be an indication that the frame buffer mods in the xorg.conf we have been using would be needed  to make things work. I don't like the "unsupported" in the first line either. It also looks like this is a pretty longstanding bug if the googling I've done is any indication. In the meantime, it's late here so I'm off to bed.

-- 
JimP


On 06/03/12 23:14, Jim Price wrote:

Phill,

I would think that the argument would be that we are not changing any features in the installed distribution, but changing the install CD to cope with some hardware issues discovered in testing. The xorg.conf included would only end up on the user's machine if they have an iMac G3, and it will just fail to work without it (or similar). We're only in feature freeze, and having a working X is a pretty important feature to fix, not a new one we're adding. I'll just check to see if I can come up with a way of determining that the machine is a G3 iMac, and that could be a way of presenting the fix for inclusion in an install script somewhere rather than being a bolt-on after the user is dumped to the command prompt.




-- 


Mailing list: https://launchpad.net/~lubuntu-qa
Post to     : lubuntu-qa@xxxxxxxxxxxxxxxxxxx
Unsubscribe : https://launchpad.net/~lubuntu-qa
More help   : https://help.launchpad.net/ListHelp







-- 
https://wiki.ubuntu.com/phillw

 
-- 
Mailing list: https://launchpad.net/~lubuntu-qa
Post to     : lubuntu-qa@xxxxxxxxxxxxxxxxxxx
Unsubscribe : https://launchpad.net/~lubuntu-qa
More help   : https://help.launchpad.net/ListHelp