Bug 7463

Please post your xorg.conf and your log file. 
Also are you using the radeon framebuffer driver?

Created attachment 6164 [details]
xorg.conf

Created attachment 6165 [details]
xorg.log

This problem appears whether i'm using a framebuffer (vesafb) or not.
Please don't tell me that's because i'm using a gentoo system :-)

Are you experiencing the same issue with the fglrx driver?

no I don't.

Since i'm maybe the only one with this problem, I'm playing with the old/new
files of the xf86-video-ati drivers. The problem is located in radeon_video.c
With the old file it runs silently.
More news soon.

Ok, I found the problem, but I don't know how to solve it.
Fans go crazy just after this call in radeon_video.c

/* Initialize VIP bus */
    RADEONVIP_init(pScrn, pPriv);

Deeper in radeon_vip.c, it goes wrong just after this:
void RADEONVIP_init(ScrnInfoPtr pScrn, RADEONPortPrivPtr pPriv)
{
    pPriv->VIP=xcalloc(1,sizeof(GENERIC_BUS_Rec)); <=====at this line
    pPriv->VIP->scrnIndex=pScrn->scrnIndex;
    pPriv->VIP->DriverPrivate.ptr=pPriv;
    pPriv->VIP->ioctl=RADEONVIP_ioctl;
    pPriv->VIP->read=RADEONVIP_read;
    pPriv->VIP->write=RADEONVIP_write;
    pPriv->VIP->fifo_read=RADEONVIP_fifo_read;
    pPriv->VIP->fifo_write=RADEONVIP_fifo_write;
   
    RADEONVIP_reset(pScrn, pPriv);
}

Any suggestion?

Ok, now finally :
pPriv->VIP=xcalloc(1,sizeof(GENERIC_BUS_Rec)); just makes that pPriv->VIP is not
NULL. It's not the problem itself.
The problem comes from RADEONVIP_reset. I don't know what it talks about so my
contribution on this problem stops here. I commented out this function, and it's
OK for me. I stay at your disposal if you wish to change anything in the code
and if you need data you don't have.
sincerely

(In reply to comment #9)
> Ok, now finally :
> pPriv->VIP=xcalloc(1,sizeof(GENERIC_BUS_Rec)); just makes that pPriv->VIP is not
> NULL. It's not the problem itself.
> The problem comes from RADEONVIP_reset. I don't know what it talks about so my
> contribution on this problem stops here. I commented out this function, and it's
> OK for me. I stay at your disposal if you wish to change anything in the code
> and if you need data you don't have.
> sincerely
> 
> 

That should be fine. That function just resets the VIP bus on the radeon (used
for tv-in stuff).  Not sure exactly how it would affect the fan though.  Perhaps
your card (no tv-in?) doesn't care for having the VIP bus active.

Well, thank you Alex for giving me your opinion.
You're right. My card has no tv input.
The problem is present with the recent xf86-video-ati-6.6.3 drivers too and I
applied the same solution. I don't know why it changes the fan speed, but maybe
that Asus applied some no conventional tricks to use the radeon in this laptop...
I don't know if i have to close this bug or leave it open.

(In reply to comment #11)
> Well, thank you Alex for giving me your opinion.
> You're right. My card has no tv input.
> The problem is present with the recent xf86-video-ati-6.6.3 drivers too and I
> applied the same solution. I don't know why it changes the fan speed, but maybe
> that Asus applied some no conventional tricks to use the radeon in this laptop...
> I don't know if i have to close this bug or leave it open.
> 
> 

Leave it open for now.  I may revisit this if I get a chance at some point in
the future.

fix pushed to git.