Bisected regression on RV610

[Alex Deucher]

On Fri, Dec 21, 2012 at 10:33 AM, Pavel Roskin <proski at gnu.org> wrote:
> Hello!
>
> The system running the current mainline Linux shows black screen for me
> reliably if I run the unshield program on xterm.  Linux 3.7 is OK.
>
> The bisection found the first bad commit:
>
> deab48f140d28d788cb2b5705761a92b02e3440d
> Alex Deucher <alexander.deucher at amd.com>
>
>     drm/radeon: add dma engine support for vm pt updates on si (v2)
>
>     Async DMA has a special packet for contiguous pt updates
>     which saves overhead.
>
>     v2: rebase

That patch only affects SI chips which yours is not, so it's probably
not what's causing problems for you.

>
> It can be reverted on top on the mainline Linux, but I still get the black
> screen.  I guess it's the first commit that is bad for my video card, but
> not the only one.
>
> While bisecting, I noticed another manifestation of the problem - unshield
> would hang for a few seconds and continue.  Following appears in the kernel
> log:
>
> [   31.476089] radeon 0000:01:00.0: GPU lockup CP stall for more than
> 10000msec
> [   31.476094] radeon 0000:01:00.0: GPU lockup (waiting for
> 0x0000000000000006 last fence id 0x0000000000000005)
>
> # lspci -vnn -s 01:00.0
> 01:00.0 VGA compatible controller [0300]: ATI Technologies Inc RV610 video
> device [Radeon HD 2400 PRO] [1002:94c3] (prog-if 00 [VGA controller])
>         Subsystem: ASUSTeK Computer Inc. Device [1043:01c2]
>         Flags: bus master, fast devsel, latency 0, IRQ 46
>         Memory at e0000000 (64-bit, prefetchable) [size=256M]
>         Memory at f5000000 (64-bit, non-prefetchable) [size=64K]
>         I/O ports at b000 [size=256]
>         [virtual] Expansion ROM at f4000000 [disabled] [size=128K]
>         Capabilities: [50] Power Management version 3
>         Capabilities: [58] Express Legacy Endpoint, MSI 00
>         Capabilities: [a0] MSI: Enable+ Count=1/1 Maskable- 64bit+
>         Capabilities: [100] Vendor Specific Information: ID=0001 Rev=1
> Len=010 <?>
>         Kernel driver in use: radeon
>

This looks like a duplicate of this issue:
http://lists.freedesktop.org/archives/dri-devel/2012-December/032139.html
http://lists.freedesktop.org/archives/dri-devel/2012-December/032291.html

Some of the ttm changes in 3.8 seem problematic.

Alex