Bug 614452 - Dual monitor and cursor problems with Quadro NVS295
Summary: Dual monitor and cursor problems with Quadro NVS295
Keywords:
Status: CLOSED ERRATA
Alias: None
Product: Fedora
Classification: Fedora
Component: xorg-x11-drv-nouveau
Version: 13
Hardware: All
OS: Linux
low
medium
Target Milestone: ---
Assignee: Ben Skeggs
QA Contact: Fedora Extras Quality Assurance
URL:
Whiteboard:
Depends On:
Blocks:
TreeView+ depends on / blocked
 
Reported: 2010-07-14 14:23 UTC by Adam Huffman
Modified: 2010-08-30 18:24 UTC (History)
3 users (show)

Fixed In Version: kernel-2.6.34.6-47.fc13
Clone Of:
Environment:
Last Closed: 2010-08-30 18:24:42 UTC
Type: ---
Embargoed:


Attachments (Terms of Use)
Xorg log when the problem is happening (92.01 KB, text/plain)
2010-07-14 14:23 UTC, Adam Huffman
no flags Details
lspci -v output (10.44 KB, text/plain)
2010-07-14 14:27 UTC, Adam Huffman
no flags Details
dmesg output while the problem is happening (71.06 KB, text/plain)
2010-07-14 14:28 UTC, Adam Huffman
no flags Details
dmesg with extra detail, as requested (310.41 KB, text/plain)
2010-07-15 10:26 UTC, Adam Huffman
no flags Details
dmesg with full nouveau debug setting (322.66 KB, text/plain)
2010-07-16 09:50 UTC, Adam Huffman
no flags Details
dmesg with suggested test kernel (578.17 KB, text/plain)
2010-08-25 17:26 UTC, Adam Huffman
no flags Details

Description Adam Huffman 2010-07-14 14:23:05 UTC
Created attachment 431788 [details]
Xorg log when the problem is happening

Description of problem:

I saw the new testing kernel kernel-2.6.34.1-11.fc13 which had some nouveau fixes and so tried it on a Dell Precision T7500 machine here that's recently had some instability, seemingly caused by the video drivers.

With that kernel, the boot sequence proceeds normally.  However, when X starts, only one of the two monitors switches to "graphics" mode - the other one is on but only shows a static cursor in the top left-hand corner.

The monitors applet shows the second monitor as being on.

The mouse cursor on the other monitor (the one that's working normally) is also hardly visible at all.

Version-Release number of selected component (if applicable):
xorg-x11-drv-nouveau-0.0.16-7.20100423git13c1043.fc13.x86_64

How reproducible:
Every time

Steps to Reproduce:
1. startx with 2 monitors connected via DP-DVI adapters
2.
3.
  
Actual results:


Expected results:


Additional info:

Comment 1 Adam Huffman 2010-07-14 14:27:40 UTC
Created attachment 431790 [details]
lspci -v output

Comment 2 Adam Huffman 2010-07-14 14:28:59 UTC
Created attachment 431791 [details]
dmesg output while the problem is happening

Comment 3 Adam Huffman 2010-07-14 14:29:47 UTC
I should add that I have another very similar machine that isn't in production yet and so can be used for extensive testing, if necessary.

Comment 4 Ben Skeggs 2010-07-14 22:45:44 UTC
Ah, this is something I couldn't test.  My DP->DVI adaptors are too chunky to fit two of them at once.  Can you boot with "log_buf_len=1M drm.debug=14 nouveau.reg_debug=0x0200" and re-post your dmesg, I'll see what I can tell from that.

Thanks!

Comment 5 Adam Huffman 2010-07-15 10:26:58 UTC
Created attachment 432019 [details]
dmesg with extra detail, as requested

Here's the dmesg output with those boot parameters

Comment 6 Adam Huffman 2010-07-15 10:28:27 UTC
In case it's useful, the second monitor showed some of the text console output when I switched back to X.

Comment 7 Ben Skeggs 2010-07-16 00:52:55 UTC
Ah, I think "nouveau.reg_debug=0x0200" got missed off your boot options :)

Comment 8 Adam Huffman 2010-07-16 09:49:43 UTC
Oops.  Actually, it was there but on the next line in grub.conf (owing to the way I pasted it) and so was probably ignored.

New version attached.  Sorry about that.

Comment 9 Adam Huffman 2010-07-16 09:50:26 UTC
Created attachment 432341 [details]
dmesg with full nouveau debug setting

Comment 10 Adam Huffman 2010-07-28 15:04:19 UTC
Is there anything else I can try?  The user who had the original problem is back on the Nvidia driver because it's currently the only way of using both his monitors.  However, it's crashing unpredictably, necessitating a reboot...

I do have an almost identical machine available for testing, the only different being it has 56xx CPUs instead of 55xx - the graphics card is the same.

Comment 11 Ben Skeggs 2010-08-17 06:04:30 UTC
Can you give this build a try: http://koji.fedoraproject.org/koji/taskinfo?taskID=2405849

Comment 12 Adam Huffman 2010-08-25 17:24:12 UTC
Just tried it.  There's something strange going on as even though the monitors detect a signal at boot, none of the BIOS screens show up.  Anyway, the Plymouth boot screen showed up on the right-hand monitor, which is connected to the right-hand card socket (looking from the front of the machine).  However, logging in seemingly didn't complete - there's just a black screen with the X cursor.

Comment 13 Adam Huffman 2010-08-25 17:26:14 UTC
Created attachment 441006 [details]
dmesg with suggested test kernel

Comment 14 Fedora Update System 2010-08-27 11:24:35 UTC
kernel-2.6.34.6-47.fc13 has been submitted as an update for Fedora 13.
https://admin.fedoraproject.org/updates/kernel-2.6.34.6-47.fc13

Comment 15 Fedora Update System 2010-08-30 18:23:21 UTC
kernel-2.6.34.6-47.fc13 has been pushed to the Fedora 13 stable repository.  If problems still persist, please make note of it in this bug report.


Note You need to log in before you can comment on or make changes to this bug.