Bug 23086 - TNT2 doesn't start with current nouveau
Summary: TNT2 doesn't start with current nouveau
Status: RESOLVED DUPLICATE of bug 23593
Alias: None
Product: xorg
Classification: Unclassified
Component: Driver/nouveau (show other bugs)
Version: git
Hardware: x86 (IA32) Linux (All)
: medium normal
Assignee: Nouveau Project
QA Contact: Xorg Project Team
URL:
Whiteboard:
Keywords:
Depends on:
Blocks:
 
Reported: 2009-08-02 01:32 UTC by Andrew Randrianasulu
Modified: 2009-09-17 04:07 UTC (History)
0 users

See Also:
i915 platform:
i915 features:


Attachments
full X log (19.72 KB, text/plain)
2009-08-02 01:33 UTC, Andrew Randrianasulu
no flags Details
xorg.conf (10.38 KB, text/plain)
2009-08-02 01:34 UTC, Andrew Randrianasulu
no flags Details
newer X log (19.72 KB, text/plain)
2009-08-03 00:42 UTC, Andrew Randrianasulu
no flags Details
last 4000 lines from /var/log/debug (332.71 KB, text/plain)
2009-08-03 00:42 UTC, Andrew Randrianasulu
no flags Details

Description Andrew Randrianasulu 2009-08-02 01:32:42 UTC
Hardware:

01:00.0 VGA compatible controller: nVidia Corporation NV5 [RIVA TNT2/TNT2 Pro] (rev 15) (prog-if 00 [VGA controller])
        Subsystem: ASUSTeK Computer Inc. AGP-V3800PRO
        Flags: bus master, 66MHz, medium devsel, latency 64, IRQ 16
        Memory at d6000000 (32-bit, non-prefetchable) [size=16M]
        Memory at d2000000 (32-bit, prefetchable) [size=32M]
        Expansion ROM at d7ef0000 [disabled] [size=64K]
        Capabilities: [60] Power Management version 1
        Capabilities: [44] AGP version 2.0
        Kernel modules: nouveau

Software:

kernel nouveau/linux-2.6 master commit 29af57c65904deced7a431623c8fea55e24e5c83 (drm/nouveau: block channel creation without ctxprog)

libdrm - master commit  5a73f066ba149816cc0fc2de4b97ec4714cf8ebc (nouveau: don't overwrite user-specified bo size needlessly)

pixman - git master, commit 9dec2e352b24bdccaac4f570b8cf12e61a9194ee (Various updates to the CODING_STYLE document)

xserver - git master , commit  ae65daab080c6269f0eadc6be1a48bf27866c542 (XQuartz: Unify how we set our bitmasks for visuals)

xf86-video-nouveau - git master, commit 1ecd4358c851b3d7bebfb00f5f7ac3c7c151bc20 ( Recalculate displayWidth if we do a late NoAccel fallback)

X itself starts, but if i run E16 1.0.0 WM on top of it - i get only movable  cursor, my background, pager, and  at this point X segfaults:

(**) Option "xkb_options" "grp:ctrl_shift_toggle"
[mi] EQ overflowing. The server is probably stuck in an infinite loop.

Backtrace:
0: /usr/X11/bin/X(xorg_backtrace+0x34) [0x80a2134]
1: /usr/X11/bin/X(mieqEnqueue+0x221) [0x809a3a1]
2: /usr/X11/bin/X(xf86PostMotionEventP+0xcf) [0x80ba3ff]
3: /usr/X11R7/lib/xorg/modules/input/evdev_drv.so [0xb5ea12f2]
4: /usr/X11/bin/X [0x80b002b]
5: /usr/X11/bin/X [0x815b742]
6: [0xffffe400]
7: /usr/X11R7/lib/libdrm.so.2(drmCommandWrite+0x35) [0xb7afbaa5]
8: /usr/X11R7/lib/libdrm_nouveau.so.1 [0xb7b1d4f1]
9: /usr/X11R7/lib/libdrm_nouveau.so.1(nouveau_bo_map_range+0xf3) [0xb7b1d723]
10: /usr/X11R7/lib/libdrm_nouveau.so.1(nouveau_bo_map+0x26) [0xb7b1d7d6]
11: /usr/X11R7/lib/xorg/modules/drivers/nouveau_drv.so [0xb7ab7db3]
12: /usr/X11R7/lib/xorg/modules/drivers/nouveau_drv.so [0xb7ab8eef]
13: /usr/X11R7/lib/xorg/modules/libexa.so [0xb68c552c]
14: /usr/X11R7/lib/xorg/modules/libexa.so [0xb68c59b6]
15: /usr/X11R7/lib/xorg/modules/libexa.so [0xb68c5a28]
16: /usr/X11R7/lib/xorg/modules/libexa.so [0xb68c64d7]
17: /usr/X11R7/lib/xorg/modules/libexa.so [0xb68c0ae8]
18: /usr/X11R7/lib/xorg/modules/libexa.so [0xb68c0b45]
19: /usr/X11R7/lib/xorg/modules/libexa.so [0xb68cabed]
20: /usr/X11R7/lib/xorg/modules/libexa.so [0xb68c8796]
21: /usr/X11/bin/X [0x8113e62]
22: /usr/X11/bin/X(CompositePicture+0x249) [0x8107da9]
23: /usr/X11/bin/X(miGlyphs+0x4c0) [0x81a9210]
24: /usr/X11/bin/X [0x8114115]
25: /usr/X11/bin/X(CompositeGlyphs+0x7d) [0x81a931d]
26: /usr/X11/bin/X [0x810f683]
27: /usr/X11/bin/X [0x810a89f]
28: /usr/X11/bin/X [0x806fd31]
29: /usr/X11/bin/X [0x8063544]
30: /lib/libc.so.6(__libc_start_main+0xe5) [0xb7ba66a5]
31: /usr/X11/bin/X [0x8063101]
Comment 1 Andrew Randrianasulu 2009-08-02 01:33:55 UTC
Created attachment 28254 [details]
full X log
Comment 2 Andrew Randrianasulu 2009-08-02 01:34:27 UTC
Created attachment 28255 [details]
xorg.conf
Comment 3 Andrew Randrianasulu 2009-08-02 01:37:05 UTC
Bug present with both KMS and non-KMS cases. Option NoAccel "1" gives me different segfault:



(II) Module shadowfb: vendor="X.Org Foundation"
        compiled for 1.6.99.1, module version = 1.0.0
        ABI class: X.Org ANSI C Emulation, version 0.4
(--) Depth 24 pixmap format is 32 bpp
(II) NOUVEAU(0): Saving VGA fonts
(II) NOUVEAU(0): Saving crtcs
(II) NOUVEAU(0): Saving encoders
(==) NOUVEAU(0): Backing store disabled
(==) NOUVEAU(0): Silken mouse enabled
(II) NOUVEAU(0): NVEnterVT is called.
(II) NOUVEAU(0): Saving VGA fonts
(II) NOUVEAU(0): Saving crtcs
(II) NOUVEAU(0): Saving encoders

Backtrace:
0: /usr/X11/bin/X(xorg_backtrace+0x34) [0x80a2134]
1: /usr/X11/bin/X [0x80a5614]
2: [0xffffe410]
3: /usr/X11R7/lib/xorg/modules/drivers/nouveau_drv.so [0xb7941fb2]
4: /usr/X11R7/lib/xorg/modules/drivers/nouveau_drv.so [0xb7942b50]
5: /usr/X11/bin/X(AddScreen+0x1a4) [0x8069784]
6: /usr/X11/bin/X(InitOutput+0x222) [0x80b4072]
7: /usr/X11/bin/X [0x806335c]
8: /lib/libc.so.6(__libc_start_main+0xe5) [0xb7a1e6a5]
9: /usr/X11/bin/X [0x8063101]
Segmentation fault at address 0x8

So, i'm using vesa driver right now.
Comment 4 Pekka Paalanen 2009-08-02 08:23:24 UTC
Just making sure: did you recompile all X drivers after updating xorg-server?
Especially evdev has been quite segfaulty in the past.

Based on the backtrace it looks like it would be stuck in an ioctl call in kernel. I'm guessing a kernel log with drm.debug=1 could be useful (and very verbose, watch out for kernel message buffer overruns). We're mainly interested in the KMS case with acceleration.
Comment 5 Andrew Randrianasulu 2009-08-02 12:56:52 UTC
Yes, i recompiled evdev and bug still here. Once i even run eterm sucessfully, but it segfaults very fast afterwards.

i'll add drm debug log tonight.
Comment 6 Andrew Randrianasulu 2009-08-03 00:42:01 UTC
Created attachment 28283 [details]
newer X log
Comment 7 Andrew Randrianasulu 2009-08-03 00:42:58 UTC
Created attachment 28284 [details]
last 4000 lines from /var/log/debug
Comment 8 Pekka Paalanen 2009-08-08 02:35:15 UTC
(In reply to comment #7)
> Created an attachment (id=28284) [details]
> last 4000 lines from /var/log/debug

Right, so for some reason it is spinning in DRM_NOUVEAU_GEM_CPU_PREP ioctl, which never completes.
Comment 9 Francisco Jerez 2009-09-17 04:07:09 UTC

*** This bug has been marked as a duplicate of bug 23593 ***


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.