Re: i810 BUG summary. Was Re: [SoftwareSuspend-devel] [Announce] 2.1.7 for 2.6.11-rc4

From: Nigel Cunningham (ncunningham_at_cyclades.com)
Date: 02/17/05

  • Next message: Theodore Ts'o: "Re: Pty is losing bytes" 
    To: mhf@berlios.de, FrameBuffer Devel <linux-fbdev-devel@lists.sourceforge.net>
Date:	Thu, 17 Feb 2005 15:39:25 +1100

    Hi Michael.

    Perhaps this belongs with the framebuffer devel guys? I'll copy them
    now. Antonio et al, have I called it right?

    Regards,

    Nigel

    On Thu, 2005-02-17 at 15:22, mhf@berlios.de wrote:
    > Hi Nigel!
    >
    > On Wednesday 16 February 2005 05:22, Nigel Cunningham wrote:
    > > Hi Michael.
    > >
    > > On Tue, 2005-02-15 at 22:32, mhf@berlios.de wrote:
    > > > - Removing intel-agp as well fixes the problem and X
    > > > can be started alright after resume.
    > >
    > > Did you see Karol's patch? It looks like it should
    > > improve your situation.
    >
    > It did :-)
    >
    > >
    > > Did you get any joy out of the posting to LKML?
    > >
    >
    > Yes, Andrew suggested privately thqt screen should come back
    > as prior to S3 and Pavel sees the setfont issue as a bug.
    >
    > 2.6.11-rc4 i810 bugs:
    >
    > 1) Screen comes back in 25 line mode but scroll remains at
    > last set linecount. (also on all other displays tested like
    > savage or sis or via.
    >
    > 2) setfont dies on resume from S3.
    >
    > 3) A scroll problem in console mode with gentoo portage and
    > likely othe apps which scroll lines of random length
    >
    > 4) With X wo DRM random colored bands flash over the screen
    > when running some programs and shellscripts as if it would
    > send the data bus instead of video data to the DAC. A
    > swsusp2 cycles fixes it until the next S3 cycle.
    >
    > 5) Then DRM (20050211 snapshot) dies on resume.
    >
    > Guess it's time to find an i810 data*** and go digging.
    >
    > Otherwise, No issues seen wrt swsusp 2.1.7 on 2.6.11-rc4.
    >
    > Good news is that S3 seems to be quite stable also when
    > interleaved with swsusp2. There were no failures.
    >
    > Lastly, I'd like to see the ACPI alarm function work to do
    > stress testing in a loop:
    >
    > 1) set alarm to current time + xx sec
    > 2) enter S3 or S4/swsusp2
    > 3) on resume triggered by alarm continue from step 1.
    >
    > Regards
    > Michael 

    -- 
Nigel Cunningham
Software Engineer, Canberra, Australia
http://www.cyclades.com
Ph: +61 (2) 6292 8028      Mob: +61 (417) 100 574
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/
  • Next message: Theodore Ts'o: "Re: Pty is losing bytes"