Bug 33648 - Black line in Lightsmark with Z compression enabled (RV530)
Summary: Black line in Lightsmark with Z compression enabled (RV530)
Status: RESOLVED FIXED
Alias: None
Product: Mesa
Classification: Unclassified
Component: Drivers/Gallium/r300 (show other bugs)
Version: git
Hardware: Other All
: medium minor
Assignee: Default DRI bug account
QA Contact:
URL: http://dee.cz/lightsmark/Lightsmark20...
Whiteboard:
Keywords:
Depends on:
Blocks:
 
Reported: 2011-01-28 05:53 UTC by Pavel Ondračka
Modified: 2012-12-02 18:04 UTC (History)
1 user (show)

See Also:
i915 platform:
i915 features:


Attachments
screenshot (187.47 KB, image/jpeg)
2011-01-28 05:54 UTC, Pavel Ondračka
Details
Lightsmark with upper part of the screen corrupt (238.78 KB, image/jpeg)
2011-03-11 08:50 UTC, Sven Arvidsson
Details
ut2004-zmask-corruption.jpg (213.23 KB, image/jpeg)
2011-03-22 06:34 UTC, Alan Swanson
Details

Description Pavel Ondračka 2011-01-28 05:53:18 UTC
After recent zbuffer compression rework Lightsmark is the only remaining app I've tested which is still not working properly here on RV530. There is a thick black line across the screen in some scenes, particularly: "no radiosity","soft shadows", "penumbra shadows" and the last scene.  Maybe this is a known issue and there are many open bugs about HyperZ, however according to db299a9f8244d53d9041fcdbd396a77ebe1f9e3e comment RV530 should just work. So I'm reporting it isn't. Also this is not a regression, it was much worse before and this is the only remaining issue here.

To be sure this isn't HiZ issue I've changed line 367 in r300_chipset.c to "caps->hiz_ram = 0;" and it haven't changed anything.
r300: AA compression: NO, Z compression: YES, HiZ: NO
Comment 1 Pavel Ondračka 2011-01-28 05:54:33 UTC
Created attachment 42642 [details]
screenshot
Comment 2 Marek Olšák 2011-03-07 14:14:43 UTC
*** Bug 35095 has been marked as a duplicate of this bug. ***
Comment 3 Marek Olšák 2011-03-07 14:15:31 UTC
Chris Rankin said it doesn't show up on RV350, so I guess it's RV530-specific.
Comment 4 Marek Olšák 2011-03-07 14:17:46 UTC
Nope, the black line shows up on R580 too (4 pipes).
Comment 5 Sven Arvidsson 2011-03-11 08:50:41 UTC
Created attachment 44363 [details]
Lightsmark with upper part of the screen corrupt

RV570 here, and Lightsmark seems to be the last app with hyper z.

The upper part of the screen is corrupt when hz is enabled. 

When RADEON_DEBUG=nozmask is used, the problem almost goes away, there's still some corruption in the upper right corner though.

With RADEON_DEBUG=nohiz I'm only getting the black line you guys described.
Comment 6 Alan Swanson 2011-03-22 06:34:29 UTC
Created attachment 44712 [details]
ut2004-zmask-corruption.jpg

Similarly on RV530 I'm seeing frequent (but relatively minor) corruption in UT2004 with a number of maps which appears to be triangular blocks of old scenes (not textures AFAICT). However these were not from previously rendered frames at same location.

Running "RADEON_DEBUG=nozmask" removes the corruption. Example screenshot has;
- Under green DM_Junkyard label is thin line of another texture.
- Under green ONS_ArticStronghold labels there is sky textures, which when standing on base platform appear in front or behind of alpha transparent textures.
Comment 7 almos 2011-05-27 07:10:29 UTC
(In reply to comment #6)
> Similarly on RV530 I'm seeing frequent (but relatively minor) corruption in
> UT2004 with a number of maps which appears to be triangular blocks of old
> scenes (not textures AFAICT). However these were not from previously rendered
> frames at same location.
The skybox is rendered into those tiles (perhaps Z order is wrong?). It's very apparent on CTF-FaceClassic for example.
Comment 8 Marek Olšák 2012-12-02 06:09:41 UTC
Lightsmark was fixed by the commit 0222b2bd4107b9e5cabfbc. Closing.
Comment 9 Alan Swanson 2012-12-02 18:04:25 UTC
To confirm, this also fixes the UT2004 zmask issues reported here too. Thanks.


Use of freedesktop.org services, including Bugzilla, is subject to our Code of Conduct. How we collect and use information is described in our Privacy Policy.