[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index][Old Index]
Re: xsrc/42262: Thinkpad T500 Intel GM45 Express X doesn't like i915drm
The following reply was made to PR xsrc/42262; it has been noted by GNATS.
From: Matthias Pfaller <leo%marco.de@localhost>
To: Jeremy Morse <jeremy.morse%gmail.com@localhost>
Subject: Re: xsrc/42262: Thinkpad T500 Intel GM45 Express X doesn't like
Date: Mon, 11 Jan 2010 21:31:54 +0100
Jeremy Morse wrote:
> When debugging I found all the
> [drm:pid0:i915_driver_irq_handler] [drm:pid0:i915_driver_irq_handler]
> iir = 0x00000040, pipestats a = 0x00020203, b = 0x00000000
> messages slowed userland to a crawl - it might be worth commenting that
> portion of i915_irq.c (line 218) out. (Plus the output would be easier
> to read).
> [drm:pid649:i915_wait_irq] [drm:pid649:i915_wait_irq] irq_nr=1 breadcrumb=0
> [drm:pid649:drm_ioctl] [drm:pid649:drm_ioctl] returning 35
> Suggests this is still the hang experienced before. Yuck.
> If you have the time to build xsrc/external/mit/xf86-video-intel
> manually, there's an "intel_statuspage" program built which dumps the
> status page contents. If you could run that and post the output (while X
> is busy hanging, if possible) we can work out whether this is hardware
> not writing to the status page (contents will all be zero), or if it's
> the status page being improperly mapped.
The contents of the status page is all zeros.
> Could you also check that bus-mastering is enabled on the device (pcictl
> /dev/pci0 dump -d <devnum>). According to  this recent commit, -5
> xorg doesn't explicitly ensure that it's turned on.
>  http://mail-index.netbsd.org/source-changes/2010/01/09/msg005178.htm
Looks like dma is off. I have attached my pcictl list and dump outputs
as well as the dmesg output. What is this device 2 function 1 thing?
Because this has dma turned on. I'm using modular-xorg-server. Probably
I need to rebuild (and patch).
Main Index |
Thread Index |