Nouveau - Dec 2009 - [Bug 25453] New: nouveau + multiple xorg servers + suspend2ram trashes root partition

http://bugs.freedesktop.org/show_bug.cgi?id=25453

           Summary: nouveau + multiple xorg servers + suspend2ram trashes
                    root partition
           Product: xorg
           Version: unspecified
          Platform: x86-64 (AMD64)
        OS/Version: Linux (All)
            Status: NEW
          Severity: critical
          Priority: high
         Component: Driver/nouveau
        AssignedTo: nouveau at lists.freedesktop.org
        ReportedBy: nouveau at genoetigt.de
         QAContact: xorg-team at lists.x.org


Created an attachment (id=31752)
 --> (http://bugs.freedesktop.org/attachment.cgi?id=31752)
output of dmesg after resume

One note before: the following only happens with nouveau, there are no problems
with the proprietary nvidia drivers.

I use 2 Xorg servers with nouveau as video graphics driver. Everything works
quite nice, at least I didn't experience any strange behaviour.
I also use suspend2ram. Also this works fine, at least, if I only use 1 X
server.

If I have 2 X servers started and suspend my machine, everything is fine until
the machine wakes up. Sometimes (I haven't yet figured out when exactly
"sometimes" is, but at latest after 2 suspends the problem occurs.) I
get some
ugly messages in my dmesg(more in the attached dmesg log):

...
sata_nv 0000:00:07.0: PCI-DMA: Out of IOMMU space for 4096 bytes
sata_nv 0000:00:07.0: PCI-DMA: Out of IOMMU space for 4096 bytes
ata2.00: exception Emask 0x0 SAct 0x0 SErr 0x0 action 0x0
ata2.00: cmd ca/00:08:43:bf:91/00:00:00:00:00/e0 tag 0 dma 4096 out
         res 50/00:00:00:00:00/00:00:00:00:00/e0 Emask 0x40 (internal error)
ata2.00: status: { DRDY }
ata2.00: configured for UDMA/133
ata2: EH complete
...

and my root partition gets mounted read-only.

Here is some information of my system:

tobi at nyx ~ $ sudo lspci -v -s 05:00.0

05:00.0 VGA compatible controller: nVidia Corporation G96 [GeForce 9400 GT]
(rev a1) (prog-if 00 [VGA controller])
        Subsystem: nVidia Corporation Device 0551
        Flags: bus master, fast devsel, latency 0, IRQ 18
        Memory at fd000000 (32-bit, non-prefetchable) [size=16M]
        Memory at d0000000 (64-bit, prefetchable) [size=256M]
        Memory at fa000000 (64-bit, non-prefetchable) [size=32M]
        I/O ports at ec00 [size=128]
        [virtual] Expansion ROM at feb80000 [disabled] [size=512K]
        Capabilities: [60] Power Management version 3
        Capabilities: [68] MSI: Enable- Count=1/1 Maskable- 64bit+
        Capabilities: [78] Express Endpoint, MSI 00
        Capabilities: [b4] Vendor Specific Information <?>
        Capabilities: [100] Virtual Channel <?>
        Capabilities: [128] Power Budgeting <?>
        Capabilities: [600] Vendor Specific Information <?>
        Kernel driver in use: nvidia
        Kernel modules: nvidia

Ignore the Kernel driver/modules lines, I got this information while running
the nvidia drivers. If lspci output while running nouveau is needed, ask.

tobi at nyx ~ $ uname -a
Linux nyx 2.6.31-gentoo-r6-nyx #2 SMP PREEMPT Mon Nov 30 23:18:53 CET 2009
x86_64 AMD Athlon(tm) 64 X2 Dual Core Processor 6000+ AuthenticAMD GNU/Linux

The following gentoo packages were installed (on Wed Dec 2 2009, about 20:00
GMT) from the x11-overlay:

x11-drivers/xf86-video-nouveau
x11-libs/libdrm
x11-base/nouveau-drm
x11-base/xorg-server

The proprietary nvidia drivers were of course uninstalled.

If more logfiles or system information is needed, just ask. ;-)


-- 
Configure bugmail: http://bugs.freedesktop.org/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the assignee for the bug.

https://bugs.freedesktop.org/show_bug.cgi?id=25453

--- Comment #1 from Marcin Slusarz <marcin.slusarz at gmail.com>
2010-06-01 10:07:46 PDT ---
3 drivers for the same hardware loaded (nvidiafb, the blob and nouveau) - this
cannot work...

-- 
Configure bugmail: https://bugs.freedesktop.org/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the assignee for the bug.

https://bugs.freedesktop.org/show_bug.cgi?id=25453

--- Comment #2 from Tobias Hommel <nouveau at genoetigt.de> 2010-06-03
13:12:41 PDT ---
Created an attachment (id=36036)
 --> (https://bugs.freedesktop.org/attachment.cgi?id=36036)
output of dmesg after resume (linux 2.6.34)

-- 
Configure bugmail: https://bugs.freedesktop.org/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the assignee for the bug.

https://bugs.freedesktop.org/show_bug.cgi?id=25453

--- Comment #3 from Tobias Hommel <nouveau at genoetigt.de> 2010-06-03
13:13:32 PDT ---
So I tried again, this time no nvidiafb(wonder why it was loaded last time).
Only nouveau, fb, fbcon kernel modules were loaded.

Again after 3 or 4 suspends/resumes with 2 xservers I get this strange harddisk
errors, which have definitely nothing to do with the harddrive itself, as it
also happens on a mirrored disk. And as already mentioned it only occurs with
nouveau, not with the binary drivers.

I attached the dmesg output for the current situation.

-- 
Configure bugmail: https://bugs.freedesktop.org/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the assignee for the bug.

https://bugs.freedesktop.org/show_bug.cgi?id=25453

--- Comment #4 from Marcin Slusarz <marcin.slusarz at gmail.com>
2010-06-04 09:25:19 PDT ---
Worrying number of hardware/BIOS bugs...

---
AMI BIOS detected: BIOS may corrupt low RAM, working around it.

mtrr: your BIOS has configured an incorrect mask, fixing it.

Node 0: aperture @ 20000000 size 32 MB
Aperture pointing to e820 RAM. Ignoring.
Your BIOS doesn't leave a aperture memory hole
Please enable the IOMMU option in the BIOS setup
This costs you 64 MB of RAM

ACPI Warning: Incorrect checksum in table [OEMB] - E9, should be E8
(20100121/tbutils-314)
ACPI: No dock devices found.
PCI: Ignoring host bridge windows from ACPI; if necessary, use
"pci=use_crs"
and report a bug

mtrr: your BIOS has configured an incorrect mask, fixing it.
mtrr: your BIOS has configured an incorrect mask, fixing it.

k8temp 0000:00:18.3: Temperature readouts might be wrong - check erratum #141

pci 0000:00:02.0: OHCI: BIOS handoff failed (BIOS bug?) 000007b4
---

Maybe there's a BIOS update for your machine?

-- 
Configure bugmail: https://bugs.freedesktop.org/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the assignee for the bug.

https://bugs.freedesktop.org/show_bug.cgi?id=25453

--- Comment #5 from Tobias Hommel <nouveau at genoetigt.de> 2010-06-05
04:38:16 PDT ---
(In reply to comment #4)
> Worrying number of hardware/BIOS bugs...
I know, but a graphics driver shouldn't touch filesystem related stuff
anyway.
Never!
I suppose I'm not the only one with a buggy BIOS, at least the ACPI tables
are
broken on nearly every board and probably the other stuff too. So the cause for
this problem should be found. Or is it that unusual to use more than one x
server and suspend?
As soon as I can spare some time, I'll do some further tests.
> .....
> 
> Maybe there's a BIOS update for your machine?
Negative on that, unfortunately.

-- 
Configure bugmail: https://bugs.freedesktop.org/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the assignee for the bug.

https://bugs.freedesktop.org/show_bug.cgi?id=25453

--- Comment #6 from Marcin Slusarz <marcin.slusarz at gmail.com>
2010-06-05 14:37:21 PDT ---
"Worrying number of hardware/BIOS bugs" == "hardware is very
buggy, so there
might be some other hard-to-discover bugs which might affect us"

BUT

I just noticed that some driver is leaking IOMMU space (which is used both by
nouveau and your SATA controller), so maybe enabling CONFIG_IOMMU_LEAK in
kernel config will help figure it out.

Why do you have iommu=force in kernel command line?
Did you enable IOMMU in BIOS?

-- 
Configure bugmail: https://bugs.freedesktop.org/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the assignee for the bug.

https://bugs.freedesktop.org/show_bug.cgi?id=25453

Ilia Mirkin <imirkin at alum.mit.edu> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|---                         |INVALID

--- Comment #7 from Ilia Mirkin <imirkin at alum.mit.edu> ---
It appears that this bug report has laid dormant for quite a while. Sorry we
haven't gotten to it. Since we fix bugs all the time, chances are pretty
good
that your issue has been fixed with the latest software. Please give it a shot.
(Linux kernel 3.10.7, xf86-video-nouveau 1.0.9, mesa 9.1.6, or their git
versions.) If upgrading to the latest isn't an option for you, your
distro's
bugzilla is probably the right destination for your bug report.

In an effort to clean up our bug list, we're pre-emptively closing all bugs
that haven't seen updates since 2011. If the original issue remains, please
make sure to provide fresh info, see http://nouveau.freedesktop.org/wiki/Bugs/
for what we need to see, and re-open this one.

Thanks,

The Nouveau Team

-- 
You are receiving this mail because:
You are the assignee for the bug.
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
<http://lists.freedesktop.org/archives/nouveau/attachments/20130818/59255108/attachment-0001.html>