Bug 4988 - [UPDATE REQUEST] kernel 3.14.33
: [UPDATE REQUEST] kernel 3.14.33
Status: VERIFIED FIXED
Product: Desktop Bugs
Classification: ROSA Desktop
Component: Main Packages
: Fresh
: All Linux
: Normal normal
: ---
Assigned To: ROSA Linux Bugs
: ROSA Linux Bugs
:
Depends on:
Blocks:
  Show dependency treegraph
 
Reported: 2015-02-02 13:40 MSK by Eugene Shatokhin
Modified: 2015-02-19 17:26 MSK (History)
4 users (show)

See Also:
RPM Package: kernel
ISO-related:
Bad POT generating:
Upstream:
vladimir.potapov: qa_verified+
denis.silakov: published+


Attachments

Note You need to log in before you can comment on or make changes to this bug.
Description Eugene Shatokhin 2015-02-02 13:40:41 MSK
Updated the kernel to version 3.14.31.
Comment 1 Eugene Shatokhin 2015-02-02 13:43:40 MSK
Advisory:
The kernel was updated to version 3.14.31 with lots of fixes all around. Besides, a patch requested in bug #4791 was added.

Build lists:
i586:
https://abf.io/build_lists/2404330

x86_64:
https://abf.io/build_lists/2404331
Comment 2 Eugene Shatokhin 2015-02-02 13:45:46 MSK
From the kernel log, changes in 3.14.31 compared to 3.14.25:
http://pastebin.com/hnfsQfHc
Comment 3 FirstLevel 2015-02-03 14:52:18 MSK
For virtual machines in VB update goes well.
Comment 4 FirstLevel 2015-02-03 22:48:42 MSK
On my HP 6465b with AMD A4 I such messages in the dmesg
[sergei@r5-64 ~]$ dmesg | grep radeo
[    2.444587] [drm] VGACON disable radeon kernel modesetting.
[    2.444593] [drm:radeon_init] *ERROR* No UMS support in radeon module!
[    5.468594] [drm] VGACON disable radeon kernel modesetting.
[    5.468600] [drm:radeon_init] *ERROR* No UMS support in radeon module!
[sergei@r5-64 ~]$
Comment 5 FirstLevel 2015-02-03 22:51:59 MSK
(In reply to comment #4)
> On my HP 6465b with AMD A4 I such messages in the dmesg
> [sergei@r5-64 ~]$ dmesg | grep radeo
> [    2.444587] [drm] VGACON disable radeon kernel modesetting.
> [    2.444593] [drm:radeon_init] *ERROR* No UMS support in radeon module!
> [    5.468594] [drm] VGACON disable radeon kernel modesetting.
> [    5.468600] [drm:radeon_init] *ERROR* No UMS support in radeon module!
> [sergei@r5-64 ~]$

It is for new kernel.

For old kernel I have no such strings. Only these:
[sergei@r5-64 ~]$ dmesg | grep rade
[    3.202975] [drm] VGACON disable radeon kernel modesetting.
[    3.202980] [drm] radeon userspace modesetting enabled.
[    3.204543] [drm] Initialized radeon 1.34.0 20080528 for 0000:00:01.0 on minor 0
[sergei@r5-64 ~]$
Comment 6 Eugene Shatokhin 2015-02-04 10:43:48 MSK
I suppose, neither 'nomodeset' nor 'radeon.modeset=0' are set in the boot options.

The messages might be the result of userspace modesetting support in radeon being deprecated for quite some time now (http://lists.freedesktop.org/archives/dri-devel/2013-January/032986.html), it has been finally turned off in this kernel.

Anyway, I'd like to cancel this update request: kernel 3.14.32 will be released in a day or two and it provides some very important fixes (e.g., memory corruption in i915 and errors in the workqueue subsystem).

I'll see about that issue with UMS too.

Thanks for testing!
Comment 7 Eugene Shatokhin 2015-02-05 16:43:55 MSK
(In reply to comment #4)
> On my HP 6465b with AMD A4 I such messages in the dmesg
> [sergei@r5-64 ~]$ dmesg | grep radeo
> [    2.444587] [drm] VGACON disable radeon kernel modesetting.
> [    2.444593] [drm:radeon_init] *ERROR* No UMS support in radeon module!
> [    5.468594] [drm] VGACON disable radeon kernel modesetting.
> [    5.468600] [drm:radeon_init] *ERROR* No UMS support in radeon module!
> [sergei@r5-64 ~]$

By the way, I could not reproduce the problem on available systems when neither 'nomodeset' nor 'radeon.modeset=0' were set. 

If the system used, fglrx, for example, and 'nomodeset' was there in the boot options, then yes, such messages appeared. However, if the system uses fglrx, the errors from the free driver can be ignored as long as the system works OK as a whole.

If there are no system stability issues, I am inclined to keep UMS support for radeon turned off. It is deprecated and might be buggy as it seems to have little maintenance. Ubuntu 14.10 and Fedora 21 turned it off, BTW.
Comment 8 Eugene Shatokhin 2015-02-11 16:55:04 MSK
Advisory:
The kernel was updated to version 3.14.33 with lots of fixes all around. Besides, the following changes were made:
* a patch requested in bug #4791 was added;
* revisited the configuration of the kernel: https://abf.io/soft/kernel-patches-and-configs/commit/02dbdd960cd8cacff2add1c3903be722b8642413

Build lists:
i586:
https://abf.io/build_lists/2424195
x86_64:
https://abf.io/build_lists/2424196
Comment 9 Eugene Shatokhin 2015-02-11 17:08:00 MSK
Notes and hints for testing:

1. Unfortunately, no fixes for the problems with brightness in Lenovo G570 yet. Work in progress.

2. Please pay attention to how suspend/resume works, see if there are any regressions there. I tweaked some ACPI-related options that might affect this.

3. On the systems with AMD graphics with fglrx installed and used, the kernel may complain about missing UMS support in 'radeon'. This is acceptable as long as the graphics works OK.

4. Please pay attention to how Intel graphics works.

5. Do some stress-tests for the file system (ext4).

6. If possible, please check if there are any improvements/regressions in RAID support. I made some related stuff built-in and enabled some kind of device detection in the kernel.
Comment 10 FirstLevel 2015-02-12 09:09:44 MSK
After update to 3.14.33 on my laptop HP 6465b with AMD A4 I have such errors in dmesg output:
[sergei-32@r5-32 ~]$ dmesg | grep ail
[    0.000000] 1736MB HIGHMEM available.
[    0.000000] 819MB LOWMEM available.
[    0.000000] e820: [mem 0x9fc00000-0xdfffffff] available for PCI devices
[    0.000000] Memory: 2555168K/2605176K available (6195K kernel code, 594K rwdata, 2744K rodata, 712K init, 752K bss, 50008K reserved, 1765900K highmem)
[    0.584718] ACPI Error: [\_SB_.PCI0.SATA.ODDZ.ZPOD] Namespace lookup failure, AE_NOT_FOUND (20131218/psargs-359)
[    0.584725] ACPI Error: Method parse/execution failed [\_SB_.PCI0.SATA.PRT1._STA] (Node f082dcc0), AE_NOT_FOUND (20131218/psparse-536)
[    0.584740] ACPI Error: [\_SB_.PCI0.SATA.ODDZ.ZPOD] Namespace lookup failure, AE_NOT_FOUND (20131218/psargs-359)
[    0.584745] ACPI Error: Method parse/execution failed [\_SB_.PCI0.SATA.PRT1._STA] (Node f082dcc0), AE_NOT_FOUND (20131218/psparse-536)
[    0.584769] ACPI Error: [\_SB_.PCI0.SATA.ODDZ.ZPOD] Namespace lookup failure, AE_NOT_FOUND (20131218/psargs-359)
[    0.584773] ACPI Error: Method parse/execution failed [\_SB_.PCI0.SATA.PRT1._STA] (Node f082dcc0), AE_NOT_FOUND (20131218/psparse-536)
[    0.588989] acpi PNP0A08:00: _OSC failed (AE_SUPPORT); disabling ASPM
[    0.602509] ACPI Error: [\_SB_.PCI0.SATA.ODDZ.ZPOD] Namespace lookup failure, AE_NOT_FOUND (20131218/psargs-359)
[    0.602516] ACPI Error: Method parse/execution failed [\_SB_.PCI0.SATA.PRT1._STA] (Node f082dcc0), AE_NOT_FOUND (20131218/psparse-536)
[    0.629569] ACPI Error: [\_SB_.PCI0.SATA.ODDZ.ZPOD] Namespace lookup failure, AE_NOT_FOUND (20131218/psargs-359)
[    0.629576] ACPI Error: Method parse/execution failed [\_SB_.PCI0.SATA.PRT1._STA] (Node f082dcc0), AE_NOT_FOUND (20131218/psparse-536)
[    1.310104] EDD information not available.
[    2.313158] random: systemd urandom read with 38 bits of entropy available
[    9.851145] ACPI: If an ACPI driver is available for this device, you should use it instead of the native driver
[   25.601513] ACPI Error: [\_SB_.PCI0.SATA.ODDZ.ZPOD] Namespace lookup failure, AE_NOT_FOUND (20131218/psargs-359)
[   25.601524] ACPI Error: Method parse/execution failed [\_SB_.PCI0.SATA.PRT1._STA] (Node f082dcc0), AE_NOT_FOUND (20131218/psparse-536)
[   25.675028] ACPI Error: [\_SB_.PCI0.SATA.ODDZ.ZPOD] Namespace lookup failure, AE_NOT_FOUND (20131218/psargs-359)
[   25.675049] ACPI Error: Method parse/execution failed [\_SB_.PCI0.SATA.PRT1._STA] (Node f082dcc0), AE_NOT_FOUND (20131218/psparse-536)
[sergei-32@r5-32 ~]$ 


and 

[sergei-32@r5-32 ~]$ dmesg | grep rro
[    0.584718] ACPI Error: [\_SB_.PCI0.SATA.ODDZ.ZPOD] Namespace lookup failure, AE_NOT_FOUND (20131218/psargs-359)
[    0.584725] ACPI Error: Method parse/execution failed [\_SB_.PCI0.SATA.PRT1._STA] (Node f082dcc0), AE_NOT_FOUND (20131218/psparse-536)
[    0.584740] ACPI Error: [\_SB_.PCI0.SATA.ODDZ.ZPOD] Namespace lookup failure, AE_NOT_FOUND (20131218/psargs-359)
[    0.584745] ACPI Error: Method parse/execution failed [\_SB_.PCI0.SATA.PRT1._STA] (Node f082dcc0), AE_NOT_FOUND (20131218/psparse-536)
[    0.584769] ACPI Error: [\_SB_.PCI0.SATA.ODDZ.ZPOD] Namespace lookup failure, AE_NOT_FOUND (20131218/psargs-359)
[    0.584773] ACPI Error: Method parse/execution failed [\_SB_.PCI0.SATA.PRT1._STA] (Node f082dcc0), AE_NOT_FOUND (20131218/psparse-536)
[    0.602509] ACPI Error: [\_SB_.PCI0.SATA.ODDZ.ZPOD] Namespace lookup failure, AE_NOT_FOUND (20131218/psargs-359)
[    0.602516] ACPI Error: Method parse/execution failed [\_SB_.PCI0.SATA.PRT1._STA] (Node f082dcc0), AE_NOT_FOUND (20131218/psparse-536)
[    0.629569] ACPI Error: [\_SB_.PCI0.SATA.ODDZ.ZPOD] Namespace lookup failure, AE_NOT_FOUND (20131218/psargs-359)
[    0.629576] ACPI Error: Method parse/execution failed [\_SB_.PCI0.SATA.PRT1._STA] (Node f082dcc0), AE_NOT_FOUND (20131218/psparse-536)
[   25.601513] ACPI Error: [\_SB_.PCI0.SATA.ODDZ.ZPOD] Namespace lookup failure, AE_NOT_FOUND (20131218/psargs-359)
[   25.601524] ACPI Error: Method parse/execution failed [\_SB_.PCI0.SATA.PRT1._STA] (Node f082dcc0), AE_NOT_FOUND (20131218/psparse-536)
[   25.675028] ACPI Error: [\_SB_.PCI0.SATA.ODDZ.ZPOD] Namespace lookup failure, AE_NOT_FOUND (20131218/psargs-359)
[   25.675049] ACPI Error: Method parse/execution failed [\_SB_.PCI0.SATA.PRT1._STA] (Node f082dcc0), AE_NOT_FOUND (20131218/psparse-536)
[   67.193052] ERROR @wl_inform_single_bss : cfg80211_inform_bss_frame error
[  110.171208] ERROR @wl_inform_single_bss : cfg80211_inform_bss_frame error
[sergei-32@r5-32 ~]$
Comment 11 FirstLevel 2015-02-12 09:11:33 MSK
And this:
[root@r5-32 ~]# dmesg | grep RRO
[    2.721464] [drm:radeon_init] *ERROR* No UMS support in radeon module!
[    9.537144] [drm:radeon_init] *ERROR* No UMS support in radeon module!
[   67.193052] ERROR @wl_inform_single_bss : cfg80211_inform_bss_frame error
[  110.171208] ERROR @wl_inform_single_bss : cfg80211_inform_bss_frame error
[root@r5-32 ~]#
Comment 12 FirstLevel 2015-02-12 09:30:03 MSK
It is strange but I see such messages in my old kernel :(
[sergei@r5-64 ~]$ dmesg | grep ail
[    0.000000] e820: [mem 0x9fc00000-0xdfffffff] available for PCI devices
[    0.000000] Calgary: Unable to locate Rio Grande table in EBDA - bailing!
[    0.000000] Memory: 5516976K/5734520K available (6605K kernel code, 958K rwdata, 3128K rodata, 1252K init, 944K bss, 217544K reserved)
[    0.471291] ACPI Error: [\_SB_.PCI0.SATA.ODDZ.ZPOD] Namespace lookup failure, AE_NOT_FOUND (20131218/psargs-359)
[    0.471298] ACPI Error: Method parse/execution failed [\_SB_.PCI0.SATA.PRT1._STA] (Node ffff8801b8836000), AE_NOT_FOUND (20131218/psparse-536)
[    0.471311] ACPI Error: [\_SB_.PCI0.SATA.ODDZ.ZPOD] Namespace lookup failure, AE_NOT_FOUND (20131218/psargs-359)
[    0.471315] ACPI Error: Method parse/execution failed [\_SB_.PCI0.SATA.PRT1._STA] (Node ffff8801b8836000), AE_NOT_FOUND (20131218/psparse-536)
[    0.471334] ACPI Error: [\_SB_.PCI0.SATA.ODDZ.ZPOD] Namespace lookup failure, AE_NOT_FOUND (20131218/psargs-359)
[    0.471338] ACPI Error: Method parse/execution failed [\_SB_.PCI0.SATA.PRT1._STA] (Node ffff8801b8836000), AE_NOT_FOUND (20131218/psparse-536)
[    0.474747] acpi PNP0A08:00: _OSC failed (AE_SUPPORT); disabling ASPM
[    0.486311] ACPI Error: [\_SB_.PCI0.SATA.ODDZ.ZPOD] Namespace lookup failure, AE_NOT_FOUND (20131218/psargs-359)
[    0.486317] ACPI Error: Method parse/execution failed [\_SB_.PCI0.SATA.PRT1._STA] (Node ffff8801b8836000), AE_NOT_FOUND (20131218/psparse-536)
[    0.512099] ACPI Error: [\_SB_.PCI0.SATA.ODDZ.ZPOD] Namespace lookup failure, AE_NOT_FOUND (20131218/psargs-359)
[    0.512106] ACPI Error: Method parse/execution failed [\_SB_.PCI0.SATA.PRT1._STA] (Node ffff8801b8836000), AE_NOT_FOUND (20131218/psparse-536)
[    1.720805] EDD information not available.
[    3.127094] random: systemd-udevd urandom read with 41 bits of entropy available
[    6.087015] AMD IOMMUv2 functionality not available on this system
[    7.200440] ACPI: If an ACPI driver is available for this device, you should use it instead of the native driver
[sergei@r5-64 ~]$
Comment 13 Eugene Shatokhin 2015-02-12 10:39:32 MSK
(In reply to comment #12)
> It is strange but I see such messages in my old kernel :(

1.
These warnings from ACPI seem to be harmless as long as suspend/resume, plugging and unplugging the power adapter, etc. are handled correctly. 

The kernel tries to find which ACPI features are supported and probes them. If a feature is not supported, such messages may appear.

2. 
> [drm:radeon_init] *ERROR* No UMS support in radeon module!

As I wrote before, such things may happen if the proprietary driver (fglrx) is used and 'nomodeset' is set. Again, this is OK, if the graphics works correctly then. 

One may blacklist 'radeon' module to silence these warnings but I would not recommend that: too easy to forget to un-blacklist it when the user decides to use the free driver again.

3.
> ERROR @wl_inform_single_bss : cfg80211_inform_bss_frame error

This might be related to a known problem with broadcom-wl driver: sometimes its components seem to be initialized in a not completely correct order. This happened with the previous kernel versions too. 

Still, WiFi has worked normally so far, even if such messages are written to the log. 

Does WiFi work on that your laptop too?
Comment 14 FirstLevel 2015-02-12 18:48:57 MSK
(In reply to comment #13)
> 
> Does WiFi work on that your laptop too?

Yes, wi-fi works well. And other devices too.
Comment 15 FirstLevel 2015-02-12 19:29:29 MSK
On my laptop HP 6465b AMD A4 after kernel update:
- wi-fi works well
- fglrx works well
- suspend to ram and suspend to disk woeks well
Comment 16 Eugene Shatokhin 2015-02-16 17:12:16 MSK
No regressions found on ASUS EeePC 1015PED with ROSA R5 i586 and nrj-desktop, nrj-desktop-pae, nrj-laptop and nrj-laptop-pae.

Probe with kernel-nrj-desktop-pae-3.14.33:
http://hw.rosalinux.ru/index.php?probe=b5de91619a

Wired networking, WiFi, sound, graphics, suspend/hibernate & resume work OK.
----------------

Testing the kernels on our test rack is in progress, by the way.
Comment 17 Andrey Ponomarenko 2015-02-16 18:28:24 MSK
test-desktop x86_64 + kernel 3.14.31: http://hw.rosalinux.ru/index.php?probe=8b621b2168

graphics, sound, wifi, bluetooth, ethernet, usb, browser, suspend, hibernate, hybrid-sleep - OK
Comment 18 Andrey Ponomarenko 2015-02-16 18:51:54 MSK
test-desktop x86_64 + kernel 3.14.33: http://hw.rosalinux.ru/index.php?probe=2ea161e13a

graphics, sound, wifi, bluetooth, ethernet, usb, browser, suspend, hibernate, hybrid-sleep - OK
Comment 19 Eugene Shatokhin 2015-02-16 19:15:40 MSK
HP Pavilion dv7 with ROSA R5 x64 - no regressions found.
http://hw.rosalinux.ru/index.php?probe=34622a2440&kind=debug
Comment 20 Eugene Shatokhin 2015-02-16 19:45:33 MSK
Test server ("test rack") with Radeon HD 6450 - OK
Hardware: http://hw.rosalinux.ru/index.php?probe=7bf547756d

Checked sound, graphics, wired networking with nrj-desktop and nrj-laptop kernels on x86-64 and with all 4 flavours (desktop / laptop, PAE / no PAE) on i586.

Suspend/resume works OK too.
Comment 21 Eugene Shatokhin 2015-02-16 19:54:10 MSK
Save hardware (incl. Radeon HD 6450) + fglrx also works fine.
http://hw.rosalinux.ru/index.php?probe=9cdefd4552&kind=debug
Comment 22 Eugene Shatokhin 2015-02-16 21:46:09 MSK
Tested kernel 3.14.33 on the same test server but with 
* GeForce GTX 480 (drivers: nouveau, nvidia340) 
* GeForce 7300 LE (drivers: nouveau, nvidia304)

Checked both i586 and x64_64.

No regressions found.
Comment 23 Andrey Ponomarenko 2015-02-17 14:37:39 MSK
ASUS X71SL x86_64 + R5 + kernel 3.14.33: http://hw.rosalinux.ru/index.php?probe=8a459d0d5a

graphics, sound, wifi, bluetooth, ethernet, usb, browser, suspend, hybrid-sleep - OK

hibernate doesn't work properly: doesn't turn off the PC completely (CPU fan continues to operate), but successfully saves the state of the system.

Probe after hibernate: http://hw.rosalinux.ru/index.php?probe=c54002bacf

The same problem is with kernel 3.14.25, so no regressions found.
Comment 24 Andrey Ponomarenko 2015-02-17 14:39:11 MSK
Dell E6530 + R5 + kernel 3.14.33: http://hw.rosalinux.ru/index.php?probe=41f0ca3341

graphics, sound, wifi, bluetooth, ethernet, usb, browser, suspend, hibernate, hybrid-sleep - OK

No regressions found.
Comment 25 Andrey Ponomarenko 2015-02-17 16:50:57 MSK
Lenovo G570 + R5 + kernel 3.14.33: http://hw.rosalinux.ru/index.php?probe=a3218abe0d

graphics, sound, wifi, ethernet, usb, browser, suspend, hibernate, hybrid-sleep - OK

No regressions found.
Comment 26 Andrey Ponomarenko 2015-02-17 16:53:27 MSK
Extended testing report - OK
Comment 27 Vladimir Potapov 2015-02-18 21:03:44 MSK
Sent to external testing
************************
Comment 28 Vladimir Potapov 2015-02-19 17:05:05 MSK
kernel-3.14.33-1
http://abf-downloads.rosalinux.ru/rosa2014.1/container/2424195/i586/main/release/
http://abf-downloads.rosalinux.ru/rosa2014.1/container/2424196/x86_64/main/release/
**************************** Advisory *********************
The kernel was updated to version 3.14.33 with lots of fixes all around. Besides, the following changes were made:
* a patch requested in bug #4791 was added;
* revisited the configuration of the kernel: https://abf.io/soft/kernel-patches-and-configs/commit/02dbdd960cd8cacff2add1c3903be722b8642413
************************************************************
QA Verified