Skip to contentSkip to conversation

2.6.25-rc3-git3: Reported regressions from 2.6.24

DISCUSSION Rafael J. Wysocki 12 participants 27 messages Mar 3, 2008 – Mar 6, 2008 PCI

lore
Share & tools
lore
#0 lore raw

This message contains a list of some regressions from 2.6.24 reported since 2.6.25-rc1 was released, for which there are no fixes in the mainline I know of.  If any of them have been fixed already, please let me know.

If you know of any other unresolved regressions from 2.6.24, please let me know either and I'll add them to the list.  Also, please let me know if any of the entries below are invalid.

Listed regressions statistics:

Date Total Pending Unresolved
----------------------------------------
2008-03-03 115 65 49
2008-02-25       90       51          39
  2008-02-17       61       45          37

Unresolved regressions
----------------------

Bug-Entry : http://bugzilla.kernel.org/show_bug.cgi?id=9938
Subject : building fails on lguest
Submitter : Cedric OLLIVIER <[email protected]>
Date : 2008-02-11 15:34

Bug-Entry : http://bugzilla.kernel.org/show_bug.cgi?id=9954
Subject : iwl3945: not only it periodically dies, it also BUG()s
Submitter : Pavel Machek <[email protected]>
Date : 2008-02-05 22:44
References : http://lkml.org/lkml/2008/2/5/453

Bug-Entry : http://bugzilla.kernel.org/show_bug.cgi?id=9958
Subject : parisc compile error
Submitter : Adrian Bunk <[email protected]>
Date : 2008-02-08 01:12
References : http://lkml.org/lkml/2008/2/7/572

Bug-Entry : http://bugzilla.kernel.org/show_bug.cgi?id=9962
Subject : mount: could not find filesystem
Submitter : Kamalesh Babulal <[email protected]>
Date : 2008-02-12 14:34
References : http://lkml.org/lkml/2008/2/12/91

Yinghai Lu <[email protected]>

Bug-Entry : http://bugzilla.kernel.org/show_bug.cgi?id=9966
Subject : kernel BUG at kernel/power/snapshot.c:464!
Submitter : Jeff Mahoney <[email protected]>
Date : 2008-02-08 20:03
References : http://lkml.org/lkml/2008/2/8/331

Bug-Entry : http://bugzilla.kernel.org/show_bug.cgi?id=9976
Subject : BUG: 2.6.25-rc1: iptables postrouting setup causes oops
Submitter : Ben Nizette <[email protected]>
Date : 2008-02-12 12:46
References : http://lkml.org/lkml/2008/2/12/148

Bug-Entry : http://bugzilla.kernel.org/show_bug.cgi?id=9978
Subject : 2.6.25-rc1: volanoMark 45% regression
Submitter : Zhang, Yanmin <[email protected]>
Date : 2008-02-13 10:30
References : http://lkml.org/lkml/2008/2/13/128

Balbir Singh <[email protected]>

Bug-Entry : http://bugzilla.kernel.org/show_bug.cgi?id=9980
Subject : 2.6.25-rc1 on Sun Ultra 40
Submitter : Jasper Bryant-Greene <[email protected]>
Date : 2008-02-13 12:25
References : http://lkml.org/lkml/2008/2/13/181

Bug-Entry : http://bugzilla.kernel.org/show_bug.cgi?id=9983
Subject : PROBLEM: 2.6.25-rc1-git2 freezes when accessing external USB hard disk (ehci-hcd)
Submitter : Linas Žvirblis <[email protected]>
Date : 2008-02-13 22:38
References : http://lkml.org/lkml/2008/2/13/566

Bug-Entry : http://bugzilla.kernel.org/show_bug.cgi?id=9984
Subject : problem with starting 2.6.25-rc1 and latest git
Submitter : Mariusz Kozlowski <[email protected]>
Date : 2008-02-13 23:16
References : http://lkml.org/lkml/2008/2/13/587

Bug-Entry : http://bugzilla.kernel.org/show_bug.cgi?id=9992
Subject : 2.6.24-git: kmap_atomic() WARN_ON()
Submitter : Thomas Gleixner <[email protected]>
Date : 2008-02-07 00:58
References : http://lkml.org/lkml/2008/2/6/451

		  http://lkml.org/lkml/2007/1/14/38

Bug-Entry : http://bugzilla.kernel.org/show_bug.cgi?id=9995
Subject : 2.6.25-rc1 regression - backlight controlls do not work - ThinkPad T61
Submitter : Lukas Hejtmanek <[email protected]>
Date : 2008-02-15 04:51

Bug-Entry : http://bugzilla.kernel.org/show_bug.cgi?id=10003
Subject : hda_intel: balance control does not work correctly
Submitter : Frans Pop <[email protected]>
Date : 2008-02-16 09:58

Bug-Entry : http://bugzilla.kernel.org/show_bug.cgi?id=10011
Subject : The computer is blocked when X is started
Submitter : François Valenduc <[email protected]>
Date : 2008-02-17 06:28

Bug-Entry : http://bugzilla.kernel.org/show_bug.cgi?id=10012
Subject : 2.6.24-git4+ regression
Submitter : Lukas Hejtmanek <[email protected]>
Date : 2008-01-30 14:56
References : http://lkml.org/lkml/2008/1/30/254

Srivatsa Vaddagiri <[email protected]>

Bug-Entry : http://bugzilla.kernel.org/show_bug.cgi?id=10015
Subject : [BUG] Linux 2.6.25-rc2 - Kernel Ooops while running dbench
Submitter : Kamalesh Babulal <[email protected]>
Date : 2008-02-16 06:44
References : http://lkml.org/lkml/2008/2/16/4

Bug-Entry : http://bugzilla.kernel.org/show_bug.cgi?id=10016
Subject : cobalt_btns.c &lt;-&gt; struct platform_device compile error
Submitter : Adrian Bunk <[email protected]>
Date : 2008-02-17 12:12
References : http://lkml.org/lkml/2008/2/17/293

Bug-Entry : http://bugzilla.kernel.org/show_bug.cgi?id=10017
Subject : cdev removal broke cobalt_btns.c compilation
Submitter : Adrian Bunk <[email protected]>
Date : 2008-02-17 12:14
References : http://lkml.org/lkml/2008/2/17/295

Bug-Entry : http://bugzilla.kernel.org/show_bug.cgi?id=10021
Subject : Linux 2.6.25-rc2 regression: LVM cannot find volume group
Submitter : Tilman Schmidt <[email protected]>
Date : 2008-02-16 20:14
References : http://lkml.org/lkml/2008/2/16/208

Jiri Slaby <[email protected]>

Bug-Entry : http://bugzilla.kernel.org/show_bug.cgi?id=10022
Subject : [2.6.25-rc1] jerky mouse cursor and randoooom key repeats
Submitter : Frans Pop <[email protected]>
Date : 2008-02-13 09:41
References : http://lkml.org/lkml/2008/2/13/82

		  http://lkml.org/lkml/2008/2/16/17

Mike Galbraith <[email protected]>

Bug-Entry : http://bugzilla.kernel.org/show_bug.cgi?id=10025
Subject : Current git very broken on the Dreamcast
Submitter : Adrian McMenamin <[email protected]>
Date : 2008-02-16 19:38
References : http://lkml.org/lkml/2008/2/16/196

Magnus Damm <[email protected]>

Bug-Entry : http://bugzilla.kernel.org/show_bug.cgi?id=10027
Subject : 2.6.25-rc[12] Video4Linux Bttv Regression
Submitter : Bongani Hlope <[email protected]>
Date : 2008-02-17 09:36
References : http://lkml.org/lkml/2008/2/17/55

Bug-Entry : http://bugzilla.kernel.org/show_bug.cgi?id=10033
Subject : mips yosemite_defconfig compile error
Submitter : Adrian Bunk <[email protected]>
Date : 2008-02-17 16:45
References : http://lkml.org/lkml/2008/2/17/383

Bug-Entry : http://bugzilla.kernel.org/show_bug.cgi?id=10041
Subject : 2.6.25-rc1/2 regression: first-time login into gnome fails
Submitter : Romano Giannetti <[email protected]>
Date : 2008-02-18 11:56
References : http://lkml.org/lkml/2008/2/18/145

Bug-Entry : http://bugzilla.kernel.org/show_bug.cgi?id=10051
Subject : Spurious messages at boot, eventually hangs the usb subsustem
Submitter : Jean-Luc Coulon <[email protected]>
Date : 2008-02-20 09:10

Bug-Entry : http://bugzilla.kernel.org/show_bug.cgi?id=10061
Subject : Hang in md5_resync
Submitter : Steinar H. Gunderson <[email protected]>
Date : 2008-02-21 13:13

Bug-Entry : http://bugzilla.kernel.org/show_bug.cgi?id=10065
Subject : 2.6.25-rc2 regression - hang on suspend
Submitter : Soeren Sonnenburg <[email protected]>
Date : 2008-02-19 12:59
References : http://lkml.org/lkml/2008/2/19/165

		  http://lkml.org/lkml/2008/2/17/381

Bug-Entry : http://bugzilla.kernel.org/show_bug.cgi?id=10067
Subject : TUNER_TDA8290=y, VIDEO_DEV=n build error
Submitter : Toralf Förster <[email protected]>
Date : 2008-02-22 10:36
References : http://lkml.org/lkml/2008/2/19/262

Bug-Entry : http://bugzilla.kernel.org/show_bug.cgi?id=10078
Subject : USB OOPS 2.6.25-rc2-git1
Submitter : Andre Tomt <[email protected]>
Date : 2008-02-19 16:19
References : http://lkml.org/lkml/2008/2/19/253

Alan Stern <[email protected]>

Bug-Entry : http://bugzilla.kernel.org/show_bug.cgi?id=10080
Subject : 2.6.25-rc2: ohci1394 problem
Submitter : Thomas Meyer <[email protected]>
Date : 2008-02-20 08:47
References : http://lkml.org/lkml/2008/2/20/58

Bug-Entry : http://bugzilla.kernel.org/show_bug.cgi?id=10082
Subject : [BUG] 2.6.25-rc2-git4 - Regression Kernel oops while running kernbench and tbench on powerpc
Submitter : Kamalesh Babulal <[email protected]>
Date : 2008-02-20 16:01
References : http://lkml.org/lkml/2008/2/20/218

		  http://lkml.org/lkml/2008/1/18/71

Bug-Entry : http://bugzilla.kernel.org/show_bug.cgi?id=10083
Subject : BUG?: &quot;Cannot map mmconfig aperture&quot;
Submitter : Diego Calleja <[email protected]>
Date : 2008-02-20 22:31
References : http://lkml.org/lkml/2008/2/20/551

Bug-Entry : http://bugzilla.kernel.org/show_bug.cgi?id=10084
Subject : 2.6.25-rc2-git4 BUG: sysfs_readdir
Submitter : Randy Dunlap <[email protected]>
Date : 2008-02-21 17:25
References : http://lkml.org/lkml/2008/2/21/212

Bug-Entry : http://bugzilla.kernel.org/show_bug.cgi?id=10086
Subject : 2.6.25-rc2 + smartd = hang
Submitter : Anders Eriksson <[email protected]>
Date : 2008-02-22 17:51
References : http://lkml.org/lkml/2008/2/22/239

Bug-Entry : http://bugzilla.kernel.org/show_bug.cgi?id=10093
Subject : 2.6.25-current-git hangs on boot
Submitter : Soeren Sonnenburg <[email protected]>
Date : 2008-02-23 18:55
References : http://lkml.org/lkml/2008/2/23/263

		  http://marc.info/?l=linux-acpi&amp;m=120387537018467&amp;w=4

Bug-Entry : http://bugzilla.kernel.org/show_bug.cgi?id=10097
Subject : SMP BUG in __nf_conntrack_find
Submitter : Christian Casteyde <[email protected]>
Date : 2008-02-25 10:44

Bug-Entry : http://bugzilla.kernel.org/show_bug.cgi?id=10100
Subject : 208c70a45624400fafd7511b96bc426bf01f8f5e breaks EC init
Submitter : Michael S. Tsirkin <[email protected]>
Date : 2008-02-25 20:19
References : http://lkml.org/lkml/2008/2/25/282

Bug-Entry : http://bugzilla.kernel.org/show_bug.cgi?id=10102
Subject : 2.6.25-rc2 Regression Thinkpad acpi
Submitter : Lukas Hejtmanek <[email protected]>
Date : 2008-02-25 12:47
References : http://lkml.org/lkml/2008/2/25/73

Bug-Entry : http://bugzilla.kernel.org/show_bug.cgi?id=10104
Subject : 2.6.25-rc3: WARNING: at arch/x86/mm/ioremap.c:137
Submitter : Phil Oester <[email protected]>
Date : 2008-02-25 03:09
References : http://lkml.org/lkml/2008/2/24/265

Bug-Entry : http://bugzilla.kernel.org/show_bug.cgi?id=10117
Subject : 2.6.25-current-git hangs on boot (pci=nommconf helps)
Submitter : Soeren Sonnenburg <[email protected]>
Date : 2008-02-23 18:55
References : http://lkml.org/lkml/2008/2/23/263

Bug-Entry : http://bugzilla.kernel.org/show_bug.cgi?id=10123
Subject : No power-off / reboot with 2.6.25-rcX (up to -rc3) kernels
Submitter : Guennadi Liakhovetski <[email protected]>
Date : 2008-02-27 08:15

Bug-Entry : http://bugzilla.kernel.org/show_bug.cgi?id=10133
Subject : INFO: possible circular locking in the resume
Submitter : Zdenek Kabelac <[email protected]>
Date : 2008-02-27
References : http://lkml.org/lkml/2008/2/26/479

Bug-Entry : http://bugzilla.kernel.org/show_bug.cgi?id=10140
Subject : Kernel blocked after &quot;ACPI: bus type pci registered&quot;
Submitter : François Valenduc <[email protected]>
Date : 2008-03-01 08:13
References : http://lkml.org/lkml/2008/2/28/153

Bug-Entry : http://bugzilla.kernel.org/show_bug.cgi?id=10146
Subject : 2.6.25-rc: complete lockup on boot/start of X (bisected)
Submitter : Marcin Slusarz <[email protected]>
Date : 2008-03-02 20:00
References : http://lkml.org/lkml/2008/3/2/91

Bug-Entry : http://bugzilla.kernel.org/show_bug.cgi?id=10147
Subject : sh r7780mp_defconfig compile error
Submitter : Adrian Bunk <[email protected]>
Date : 2008-03-02 13:42
References : http://lkml.org/lkml/2008/3/2/141

Bug-Entry : http://bugzilla.kernel.org/show_bug.cgi?id=10148
Subject : [BUG kernel 2.6.25-rc3 IPV6] ping6 -I eth0 ff02::1 causes system hang.
Submitter : Komuro <[email protected]>
Date : 2008-03-02 13:07
References : http://lkml.org/lkml/2008/3/2/32

Bug-Entry : http://bugzilla.kernel.org/show_bug.cgi?id=10149
Subject : bisected boot regression post 2.6.25-rc3.. please revert
Submitter : Arjan van de Ven <[email protected]>
Date : 2008-03-01 10:56
References : http://lkml.org/lkml/2008/3/1/155

Bug-Entry : http://bugzilla.kernel.org/show_bug.cgi?id=10152
Subject : Clocksource tsc is always unstable with 2.6.25-* kernels and CONFIG_NO_HZ=y on my box
Submitter : Gabriel C <[email protected]>
Date : 2008-02-24 01:31
References : http://lkml.org/lkml/2008/2/23/380

		  http://lkml.org/lkml/2008/2/24/281

Bug-Entry : http://bugzilla.kernel.org/show_bug.cgi?id=10155
Subject : Regression in 2.6.25-rc3: s2ram segfaults before suspending
Submitter : Klaus S. Madsen <[email protected]>
Date : 2008-02-27 23:10
References : http://lkml.org/lkml/2008/2/27/364

Ingo Molnar <[email protected]>

Bug-Entry : http://bugzilla.kernel.org/show_bug.cgi?id=10156
Subject : KVM &amp; Qemu crashed with infinite recursive kernel loop in the guest
Submitter : Zdenek Kabelac <[email protected]>
Date : 2008-02-28 11:25
References : http://lkml.org/lkml/2008/2/28/106

Regressionn with patches
------------------------

Bug-Entry : http://bugzilla.kernel.org/show_bug.cgi?id=9874
Subject : Undocking Lenovo ThinkPad T61 causes oops
Submitter : Lukas Hejtmanek <[email protected]>
Date : 2008-02-02 02:07

Patch : http://marc.info/?l=linux-acpi&amp;m=120389632114090&amp;w=2

Bug-Entry : http://bugzilla.kernel.org/show_bug.cgi?id=9969
Subject : 2.6.24-git15 Keyboard Issue?
Submitter : Chris Holvenstot <[email protected]>
Date : 2008-02-06 14:02
References : http://lkml.org/lkml/2008/2/6/100

		  http://lkml.org/lkml/2008/2/13/82

Patch : http://lkml.org/lkml/2008/2/15/343

Bug-Entry : http://bugzilla.kernel.org/show_bug.cgi?id=10013
Subject : tbench regression in 2.6.25-rc1
Submitter : Zhang, Yanmin <[email protected]>
Date : 2008-02-15 02:52
References : http://lkml.org/lkml/2008/2/14/546

David Miller <[email protected]>
Patch : http://lkml.org/lkml/2008/2/18/66

		  http://lkml.org/lkml/2008/2/18/117

Bug-Entry : http://bugzilla.kernel.org/show_bug.cgi?id=10020
Subject : mips SMBUS_PSC_BASE compile errors
Submitter : Adrian Bunk <[email protected]>
Date : 2008-02-17 12:21
References : http://lkml.org/lkml/2008/2/17/299

Patch : http://lkml.org/lkml/2008/2/18/122

Bug-Entry : http://bugzilla.kernel.org/show_bug.cgi?id=10024
Subject : 2.6.25-rc2 regression in rt61pci wireless driver
Submitter : Chris Clayton <[email protected]>
Date : 2008-02-16 13:06
References : http://lkml.org/lkml/2008/2/16/82

Stefano Brivio <[email protected]>
Patch : http://lkml.org/lkml/2008/3/2/26

Bug-Entry : http://bugzilla.kernel.org/show_bug.cgi?id=10030
Subject : Suspend doesn't work when SD card is inserted
Submitter : Zdenek Kabelac <[email protected]>
Date : 2008-02-17 12:00
References : http://lkml.org/lkml/2008/2/17/81

Patch : http://marc.info/?l=linux-acpi&amp;m=120389632114090&amp;w=2

Bug-Entry : http://bugzilla.kernel.org/show_bug.cgi?id=10031
Subject : [2.6.25-rc2] e100: Trying to free already-free IRQ 11 during suspend ...
Submitter : Andrey Borzenkov <[email protected]>
Date : 2008-02-16 13:36
References : http://lkml.org/lkml/2008/2/17/125

Patch : http://lkml.org/lkml/2008/2/21/259

Bug-Entry : http://bugzilla.kernel.org/show_bug.cgi?id=10071
Subject : kernel hang in inet_init
Submitter : Yannick Dirou <[email protected]>
Date : 2008-02-23 00:34

Patch : http://lkml.org/lkml/2008/2/2/11

Bug-Entry : http://bugzilla.kernel.org/show_bug.cgi?id=10075
Subject : regression: CD burning (k3b) went broke
Submitter : Mike Galbraith <[email protected]>
Date : 2008-02-21 09:42
References : http://lkml.org/lkml/2008/2/21/41

		  http://lkml.org/lkml/2008/3/2/131

Tejun Heo <[email protected]>
Patch : http://lkml.org/lkml/2008/2/24/278

		  http://article.gmane.org/gmane.linux.scsi/39425

Bug-Entry : http://bugzilla.kernel.org/show_bug.cgi?id=10079
Subject : 2.6.25-rc[1,2]: failed to setup dm-crypt key mapping
Submitter : Michael S. Tsirkin <[email protected]>
Date : 2008-02-20 08:43
References : http://lkml.org/lkml/2008/2/20/131

Patch : http://lkml.org/lkml/2008/2/22/127

Bug-Entry : http://bugzilla.kernel.org/show_bug.cgi?id=10081
Subject : build #365 issue for v2.6.25-rc2-342-g5d9c4a7 in ./arch/x86/kvm/kvm.ko
Submitter : Toralf Förster <[email protected]>
Date : 2008-02-20 14:11
References : http://lkml.org/lkml/2008/2/20/167

Randy Dunlap <[email protected]>
Patch : http://lkml.org/lkml/2008/2/20/195

		  http://lkml.org/lkml/2008/2/20/379

Bug-Entry : http://bugzilla.kernel.org/show_bug.cgi?id=10122
Subject : FIXED_PHY must depend on PHYLIB=y
Submitter : Olaf Hering <[email protected]>
Date : 2008-02-27 07:14
References : http://lkml.org/lkml/2008/2/27/90

Patch : http://lkml.org/lkml/2008/2/27/157

Bug-Entry : http://bugzilla.kernel.org/show_bug.cgi?id=10132
Subject : 2.6.25 git regression, oops on boot
Submitter : Jonathan McDowell <[email protected]>
Date : 2008-02-29 11:09
References : http://marc.info/?l=linux-kernel&amp;m=120423268404812&amp;w=2

		  http://lkml.org/lkml/2008/2/28/369

Lin Ming <[email protected]>
Patch : http://lkml.org/lkml/2008/2/29/49

Bug-Entry : http://bugzilla.kernel.org/show_bug.cgi?id=10145
Subject : apanel: fix kconfig dependencies
Submitter : Randy Dunlap <[email protected]>
Date : 2008-02-06 16:27
References : http://lkml.org/lkml/2008/2/6/492

Patch : http://lkml.org/lkml/2008/2/7/333

Bug-Entry : http://bugzilla.kernel.org/show_bug.cgi?id=10153
Subject : (regression) kernel/timeconst.h bugs with HZ=128
Submitter : David Brownell <[email protected]>
Date : 2008-02-26 19:32
References : http://lkml.org/lkml/2008/2/26/294

Patch : http://bugzilla.kernel.org/attachment.cgi?id=15114&amp;action=view

		  http://bugzilla.kernel.org/attachment.cgi?id=15115&amp;action=view

For details, please follow the links given in references.

As you can see, there is a Bugzilla entry for each of the listed regressions. There also is a Bugzilla entry used for tracking the regressions from 2.6.24, unresolved as well as resolved, at:

http://bugzilla.kernel.org/show_bug.cgi?id=9832

Please let me know if there are any Bugzilla entries that should be added to the list in there.

Thanks,
Rafael

#2 lore raw
Quoted 11 lines · Ingo Molnar
Quoted 7 lines · Rafael J. Wysocki

Unresolved regressions
----------------------

Bug-Entry : http://bugzilla.kernel.org/show_bug.cgi?id=9938
Subject : building fails on lguest
Submitter : Cedric OLLIVIER <[email protected]>
Date : 2008-02-11 15:34

Quoted 7 lines

this should be resolved by:

commit db342d216ba9e060d8c5501eefc1d0a789c9e711
Author: Tony Breeds <[email protected]>
Date: Tue Feb 19 08:16:03 2008 +0100

lguest: fix build breakage

Thanks, closed.

Rafael

#4 lore raw
In reply to Rafael J. Wysocki
Quoted 18 lines · Rafael J. Wysocki

This message contains a list of some regressions from 2.6.24 reported since 2.6.25-rc1 was released, for which there are no fixes in the mainline I know of.  If any of them have been fixed already, please let me know.

If you know of any other unresolved regressions from 2.6.24, please let me know either and I'll add them to the list.  Also, please let me know if any of the entries below are invalid.

Listed regressions statistics:

Date Total Pending Unresolved
----------------------------------------
2008-03-03 115 65 49
2008-02-25       90       51          39
  2008-02-17       61       45          37

Quoted 5 lines

Bug-Entry : http://bugzilla.kernel.org/show_bug.cgi?id=9995
Subject : 2.6.25-rc1 regression - backlight controlls do not work - ThinkPad T61
Submitter : Lukas Hejtmanek <[email protected]>
Date : 2008-02-15 04:51

Actually, I see this one too. Witch 2.6.25-rc3, brightness keys just do not work. I can still control brightness if I load acpi video module and do

root@amd:/sys/class/backlight/acpi_video0# echo 1 > brightness 

.

Quoted 2 lines

Bug-Entry : http://bugzilla.kernel.org/show_bug.cgi?id=10012
Subject : 2.6.24-git4+ regression

Better description would be "interactivity problems in -git4+".

Quoted 5 lines

Submitter : Lukas Hejtmanek <[email protected]>
Date : 2008-01-30 14:56
References : http://lkml.org/lkml/2008/1/30/254

Srivatsa Vaddagiri <[email protected]>

And he has GROUP_SCHED enabled, that is known broken.

Quoted 8 lines

Bug-Entry : http://bugzilla.kernel.org/show_bug.cgi?id=10022
Subject : [2.6.25-rc1] jerky mouse cursor and randoooom key repeats
Submitter : Frans Pop <[email protected]>
Date : 2008-02-13 09:41
References : http://lkml.org/lkml/2008/2/13/82

		  http://lkml.org/lkml/2008/2/16/17

Mike Galbraith <[email protected]>

This is duplicate of 10012.

Pavel

Signature · 2 lines
(english) http://www.livejournal.com/~pavelmachek
(cesky, pictures) http://atrey.karlin.mff.cuni.cz/~pavel/picture/horses/blog.html
#5 lore raw
Quoted 5 lines · Pavel Machek
Quoted 2 lines

Bug-Entry : http://bugzilla.kernel.org/show_bug.cgi?id=10012
Subject : 2.6.24-git4+ regression

Better description would be "interactivity problems in -git4+".

Quoted 7 lines
Quoted 5 lines

Submitter : Lukas Hejtmanek <[email protected]>
Date : 2008-01-30 14:56
References : http://lkml.org/lkml/2008/1/30/254

Srivatsa Vaddagiri <[email protected]>

And he has GROUP_SCHED enabled, that is known broken.

this should all be fixed in sched-devel.git (via a revert):

http://people.redhat.com/mingo/sched-devel.git/README

it's lined up for upstream pull.

Ingo

#6 lore raw
Quoted 14 lines · Ingo Molnar
Quoted 11 lines · Pavel Machek
Quoted 2 lines

Bug-Entry : http://bugzilla.kernel.org/show_bug.cgi?id=10012
Subject : 2.6.24-git4+ regression

Better description would be "interactivity problems in -git4+".

Quoted 5 lines

Submitter : Lukas Hejtmanek <[email protected]>
Date : 2008-01-30 14:56
References : http://lkml.org/lkml/2008/1/30/254

Srivatsa Vaddagiri <[email protected]>

Quoted 7 lines
Quoted 1 line

And he has GROUP_SCHED enabled, that is known broken.

this should all be fixed in sched-devel.git (via a revert):

http://people.redhat.com/mingo/sched-devel.git/README

it's lined up for upstream pull.

Is it:

commit 62fb185130e4d420f71a30ff59d8b16b74ef5d2b

Author: Peter Zijlstra <[email protected]>
Date:   Mon Feb 25 17:34:02 2008 +0100

sched: revert load_balance_monitor() changes

#7 lore raw
In reply to Rafael J. Wysocki
Quoted 8 lines

Bug-Entry : http://bugzilla.kernel.org/show_bug.cgi?id=9978
Subject : 2.6.25-rc1: volanoMark 45% regression
Submitter : Zhang, Yanmin <[email protected]>
Date : 2008-02-13 10:30
References : http://lkml.org/lkml/2008/2/13/128

Balbir Singh <[email protected]>

Peter's revert of the load balance patches should fix this one. Yanmin, could you please confirm if the patch at
http://lkml.org/lkml/2008/2/25/202 helps?

Signature · 2 lines
regards,
Dhaval
#8 lore raw
Quoted 9 lines · Dhaval Giani
Quoted 8 lines

Bug-Entry : http://bugzilla.kernel.org/show_bug.cgi?id=9978
Subject : 2.6.25-rc1: volanoMark 45% regression
Submitter : Zhang, Yanmin <[email protected]>
Date : 2008-02-13 10:30
References : http://lkml.org/lkml/2008/2/13/128

Balbir Singh <[email protected]>

Quoted 3 lines

Peter's revert of the load balance patches should fix this one. Yanmin, could you please confirm if the patch at
http://lkml.org/lkml/2008/2/25/202 helps?

I tested it against 2.6.25-rc3 on my 16-core tigerton machine. It really improves volano result although it doesn't recover all the result.
Comparing with 2.6.24, without the patch, volanoMark has about 50% regression with 2.6.25-rc3. With the patch, volanoMark has about 15% regression.

-yanmin

#9 lore raw
Quoted 9 lines · Yanmin
Quoted 8 lines · Dhaval Giani
Quoted 7 lines

Bug-Entry : http://bugzilla.kernel.org/show_bug.cgi?id=9978
Subject : 2.6.25-rc1: volanoMark 45% regression
Submitter : Zhang, Yanmin <[email protected]>
Date : 2008-02-13 10:30
References : http://lkml.org/lkml/2008/2/13/128

Balbir Singh <[email protected]>

Quoted 8 lines
Quoted 3 lines

Peter's revert of the load balance patches should fix this one. Yanmin, could you please confirm if the patch at
http://lkml.org/lkml/2008/2/25/202 helps?

I tested it against 2.6.25-rc3 on my 16-core tigerton machine. It really improves volano result although it doesn't recover all the result.
Comparing with 2.6.24, without the patch, volanoMark has about 50% regression with 2.6.25-rc3. With the patch, volanoMark has about 15% regression.

Have you had a chance to git-bisect the culprit after the revert?

Signature · 4 lines
	Warm Regards,
	Balbir Singh
	Linux Technology Center
	IBM, ISTL
#10 lore raw
Quoted 13 lines · Balbir Singh
Quoted 12 lines · Yanmin
Quoted 11 lines · Dhaval Giani
Quoted 7 lines

Bug-Entry : http://bugzilla.kernel.org/show_bug.cgi?id=9978
Subject : 2.6.25-rc1: volanoMark 45% regression
Submitter : Zhang, Yanmin <[email protected]>
Date : 2008-02-13 10:30
References : http://lkml.org/lkml/2008/2/13/128

Balbir Singh <[email protected]>

Peter's revert of the load balance patches should fix this one. Yanmin, could you please confirm if the patch at
http://lkml.org/lkml/2008/2/25/202 helps?

Quoted 7 lines
Quoted 5 lines

I tested it against 2.6.25-rc3 on my 16-core tigerton machine. It really improves volano result although it doesn't recover all the result.
Comparing with 2.6.24, without the patch, volanoMark has about 50% regression with 2.6.25-rc3. With the patch, volanoMark has about 15% regression.

Have you had a chance to git-bisect the culprit after the revert?

How to bisect it if the reverted patch is submitted after the culprit patch?

-yanmin

#11 lore raw
Quoted 14 lines · Yanmin
Quoted 13 lines · Balbir Singh
Quoted 12 lines · Yanmin
Quoted 11 lines · Dhaval Giani
Quoted 7 lines

Bug-Entry : http://bugzilla.kernel.org/show_bug.cgi?id=9978
Subject : 2.6.25-rc1: volanoMark 45% regression
Submitter : Zhang, Yanmin <[email protected]>
Date : 2008-02-13 10:30
References : http://lkml.org/lkml/2008/2/13/128

Balbir Singh <[email protected]>

Peter's revert of the load balance patches should fix this one. Yanmin, could you please confirm if the patch at
http://lkml.org/lkml/2008/2/25/202 helps?

Quoted 7 lines
Quoted 6 lines
Quoted 5 lines

I tested it against 2.6.25-rc3 on my 16-core tigerton machine. It really improves volano result although it doesn't recover all the result.
Comparing with 2.6.24, without the patch, volanoMark has about 50% regression with 2.6.25-rc3. With the patch, volanoMark has about 15% regression.

Have you had a chance to git-bisect the culprit after the revert?

How to bisect it if the reverted patch is submitted after the culprit patch?

Good question. What I would do is create a branch at the patches that caused the regression and apply all patches (except the reverted patches to it, solving conflicts if any) and run git-bisect that. But, I suspect it is a lot of work.

Signature · 4 lines
	Warm Regards,
	Balbir Singh
	Linux Technology Center
	IBM, ISTL
#12 lore raw
In reply to Zhang, Yanmin
Quoted 4 lines · Zhang, Yanmin
Quoted 1 line

Have you had a chance to git-bisect the culprit after the revert?

How to bisect it if the reverted patch is submitted after the culprit patch?

i do this by using quilt ontop of git-bisect.

I do something like this:

mkdir patches
echo revert.patch > patches/series
git-log -1 -p 62fb185130e4d420f > patches/revert.patch

git-bisect start
git-bisect bad v2.6.24-rc3
git-bisect good v2.6.24

quilt push # the revert is applied
[ test the kernel ]
quilt pop # revert is unapplied

git-bisect bad # if it's still bad

quilt push # apply the revert again
[ test the next kernel ]
quilt pop # undo the revert

git-bisect good # if it's good

etc. NOTE: if the "quilt push" fails, it's likely because you are in a point in the tree that does not have the reverted commits applied yet. In that case there's no need to push/pop, just test the bisection point.

Note, since there are _two_ guilty commits here:

commit 58e2d4ca581167c2a079f4ee02be2f0bc52e8729
Author: Srivatsa Vaddagiri <[email protected]>
Date: Fri Jan 25 21:08:00 2008 +0100
sched: group scheduling, change how cpu load is calculated

commit 6b2d7700266b9402e12824e11e0099ae6a4a6a79
Author: Srivatsa Vaddagiri <[email protected]>
Date: Fri Jan 25 21:08:00 2008 +0100
sched: group scheduler, fix fairness of cpu bandwidth allocation for task

make sure the bisection point is never "between" these two commits.

You can check whether a bisection point has the two guilty commits applied, via:

git-log | grep -E '58e2d4ca581167c2a0|6b2d7700266b9402e12'

if this comes up empty, the guilty commits are not applied.

Ingo

#13 lore raw
Quoted 8 lines · Ingo Molnar
Quoted 4 lines · Zhang, Yanmin
Quoted 1 line

Have you had a chance to git-bisect the culprit after the revert?

How to bisect it if the reverted patch is submitted after the culprit patch?

i do this by using quilt ontop of git-bisect.

Thanks for your kind information. My machines are buys in testing 2.6.25-rc4.

Let me find a timeslot to track this issue again.

-yanmin

Quoted 49 lines

I do something like this:

mkdir patches
echo revert.patch > patches/series
git-log -1 -p 62fb185130e4d420f > patches/revert.patch

git-bisect start
git-bisect bad v2.6.24-rc3
git-bisect good v2.6.24

quilt push # the revert is applied
[ test the kernel ]
quilt pop # revert is unapplied

git-bisect bad # if it's still bad

quilt push # apply the revert again
[ test the next kernel ]
quilt pop # undo the revert

git-bisect good # if it's good

etc. NOTE: if the "quilt push" fails, it's likely because you are in a point in the tree that does not have the reverted commits applied yet. In that case there's no need to push/pop, just test the bisection point.

Note, since there are _two_ guilty commits here:

commit 58e2d4ca581167c2a079f4ee02be2f0bc52e8729
Author: Srivatsa Vaddagiri <[email protected]>
Date: Fri Jan 25 21:08:00 2008 +0100
sched: group scheduling, change how cpu load is calculated

commit 6b2d7700266b9402e12824e11e0099ae6a4a6a79
Author: Srivatsa Vaddagiri <[email protected]>
Date: Fri Jan 25 21:08:00 2008 +0100
sched: group scheduler, fix fairness of cpu bandwidth allocation for task

make sure the bisection point is never "between" these two commits.

You can check whether a bisection point has the two guilty commits applied, via:

git-log | grep -E '58e2d4ca581167c2a0|6b2d7700266b9402e12'

if this comes up empty, the guilty commits are not applied.

Ingo

#14 lore raw
In reply to Zhang, Yanmin
Quoted 10 lines · Yanmin
Quoted 9 lines · Dhaval Giani
Quoted 8 lines

Bug-Entry : http://bugzilla.kernel.org/show_bug.cgi?id=9978
Subject : 2.6.25-rc1: volanoMark 45% regression
Submitter : Zhang, Yanmin <[email protected]>
Date : 2008-02-13 10:30
References : http://lkml.org/lkml/2008/2/13/128

Balbir Singh <[email protected]>

Quoted 7 lines
Quoted 3 lines

Peter's revert of the load balance patches should fix this one. Yanmin, could you please confirm if the patch at
http://lkml.org/lkml/2008/2/25/202 helps?

I tested it against 2.6.25-rc3 on my 16-core tigerton machine. It really improves volano result although it doesn't recover all the result.
Comparing with 2.6.24, without the patch, volanoMark has about 50% regression with 2.6.25-rc3. With the patch, volanoMark has about 15% regression.

One more update on the reverted patch: Comparing with 2.6.24, cpu2000-fp has about 4% regression with kernel 2.6.25-rc on my madison IPF machine. As you know, cpu2000-fp consists of many sub-testing. The most regression looks relevant to a couple of testing in the middle step. But if I ran the sub-testing manually, I couldn't see any regression. If I started kernel with boot parameter maxcpus=1, the regression becomes 1%.

If I apply Peter's revert patch to 2.6.25-rc3, the regression also becomes 1%.

I don't know what causes the last 1% regression.

-yanmin

#15 lore raw
In reply to Rafael J. Wysocki
Quoted 4 lines · Rafael J. Wysocki

Bug-Entry : http://bugzilla.kernel.org/show_bug.cgi?id=10123
Subject : No power-off / reboot with 2.6.25-rcX (up to -rc3) kernels
Submitter : Guennadi Liakhovetski <[email protected]>
Date : 2008-02-27 08:15

Guennadi bisected this down to:

commit fd7d1ced29e5beb88c9068801da7a362606d8273
PCI: make pci_bus a struct device

and it's suspected that Andrew's poweroff problems might be related as well. Guennadi, Andrew, find below a manual revert of this change - does it make any difference?

Ingo

---------------->
Subject: revert "PCI: make pci_bus a struct device"
From: Ingo Molnar <[email protected]>
Date: Thu Mar 06 08:12:56 CET 2008

revert:

commit fd7d1ced29e5beb88c9068801da7a362606d8273
PCI: make pci_bus a struct device

patch for testing purposes.

---
 drivers/pci/bus.c       |   19 +++-----------
 drivers/pci/pci-sysfs.c |    6 ++--
 drivers/pci/pci.h       |    2 -
 drivers/pci/probe.c     |   64 +++++++++++++++++++++++++++++-------------------
 drivers/pci/remove.c    |    6 +++-
 include/linux/pci.h     |    4 +--
 6 files changed, 53 insertions(+), 48 deletions(-)
drivers/pci/bus.c +4 -15
Index: linux/drivers/pci/bus.c
===================================================================
@@ -108,7 +108,6 @@ int pci_bus_add_device(struct pci_dev *d
 void pci_bus_add_devices(struct pci_bus *bus)
 {
 	struct pci_dev *dev;
-	struct pci_bus *child_bus;
 	int retval;
 
 	list_for_each_entry(dev, &bus->devices, bus_list) {
@@ -139,21 +138,11 @@ void pci_bus_add_devices(struct pci_bus 
 			       up_write(&pci_bus_sem);
 			}
 			pci_bus_add_devices(dev->subordinate);
-
-			/* register the bus with sysfs as the parent is now
-			 * properly registered. */
-			child_bus = dev->subordinate;
-			child_bus->dev.parent = child_bus->bridge;
-			retval = device_register(&child_bus->dev);
-			if (retval)
-				dev_err(&dev->dev, "Error registering pci_bus,"
-					" continuing...\n");
-			else
-				retval = device_create_file(&child_bus->dev,
-							&dev_attr_cpuaffinity);
+			retval = sysfs_create_link(&dev->subordinate->class_dev.kobj,
+						   &dev->dev.kobj, "bridge");
 			if (retval)
-				dev_err(&dev->dev, "Error creating cpuaffinity"
-					" file, continuing...\n");
+				dev_err(&dev->dev, "Error creating sysfs "
+					"bridge symlink, continuing...\n");
 		}
 	}
 }
drivers/pci/pci-sysfs.c +3 -3
Index: linux/drivers/pci/pci-sysfs.c
===================================================================
@@ -358,7 +358,7 @@ pci_read_legacy_io(struct kobject *kobj,
 		   char *buf, loff_t off, size_t count)
 {
         struct pci_bus *bus = to_pci_bus(container_of(kobj,
-                                                      struct device,
+                                                      struct class_device,
 						      kobj));
 
         /* Only support 1, 2 or 4 byte accesses */
@@ -383,7 +383,7 @@ pci_write_legacy_io(struct kobject *kobj
 		    char *buf, loff_t off, size_t count)
 {
         struct pci_bus *bus = to_pci_bus(container_of(kobj,
-						      struct device,
+						      struct class_device,
 						      kobj));
         /* Only support 1, 2 or 4 byte accesses */
         if (count != 1 && count != 2 && count != 4)
@@ -407,7 +407,7 @@ pci_mmap_legacy_mem(struct kobject *kobj
                     struct vm_area_struct *vma)
 {
         struct pci_bus *bus = to_pci_bus(container_of(kobj,
-                                                      struct device,
+                                                      struct class_device,
 						      kobj));
 
         return pci_mmap_legacy_page_range(bus, vma);
drivers/pci/pci.h +1 -1
Index: linux/drivers/pci/pci.h
===================================================================
@@ -64,7 +64,7 @@ static inline int pci_no_d1d2(struct pci
 }
 extern int pcie_mch_quirk;
 extern struct device_attribute pci_dev_attrs[];
-extern struct device_attribute dev_attr_cpuaffinity;
+extern struct class_device_attribute class_device_attr_cpuaffinity;
 
 /**
  * pci_match_one_device - Tell if a PCI device structure has a matching
drivers/pci/probe.c +39 -25

146 lines not on this page · raw message

drivers/pci/remove.c +4 -2
Index: linux/drivers/pci/remove.c
===================================================================
@@ -74,8 +74,10 @@ void pci_remove_bus(struct pci_bus *pci_
 	list_del(&pci_bus->node);
 	up_write(&pci_bus_sem);
 	pci_remove_legacy_files(pci_bus);
-	device_remove_file(&pci_bus->dev, &dev_attr_cpuaffinity);
-	device_unregister(&pci_bus->dev);
+	class_device_remove_file(&pci_bus->class_dev,
+		&class_device_attr_cpuaffinity);
+	sysfs_remove_link(&pci_bus->class_dev.kobj, "bridge");
+	class_device_unregister(&pci_bus->class_dev);
 }
 EXPORT_SYMBOL(pci_remove_bus);
 
include/linux/pci.h +2 -2
Index: linux/include/linux/pci.h
===================================================================
@@ -275,13 +275,13 @@ struct pci_bus {
 	unsigned short  bridge_ctl;	/* manage NO_ISA/FBB/et al behaviors */
 	pci_bus_flags_t bus_flags;	/* Inherited by child busses */
 	struct device		*bridge;
-	struct device		dev;
+	struct class_device	class_dev;
 	struct bin_attribute	*legacy_io; /* legacy I/O for this bus */
 	struct bin_attribute	*legacy_mem; /* legacy mem */
 };
 
 #define pci_bus_b(n)	list_entry(n, struct pci_bus, node)
-#define to_pci_bus(n)	container_of(n, struct pci_bus, dev)
+#define to_pci_bus(n)	container_of(n, struct pci_bus, class_dev)
 
 /*
  * Error values that may be returned by PCI functions.
#16 lore raw
Quoted 8 lines · Ingo Molnar
Quoted 4 lines · Rafael J. Wysocki

Bug-Entry : http://bugzilla.kernel.org/show_bug.cgi?id=10123
Subject : No power-off / reboot with 2.6.25-rcX (up to -rc3) kernels
Submitter : Guennadi Liakhovetski <[email protected]>
Date : 2008-02-27 08:15

Quoted 8 lines

Guennadi bisected this down to:

commit fd7d1ced29e5beb88c9068801da7a362606d8273
PCI: make pci_bus a struct device

and it's suspected that Andrew's poweroff problems might be related as well. Guennadi, Andrew, find below a manual revert of this change - does it make any difference?

Yes, this patch fixes both startup warnings and lets the system reboot and power off again!

Thanks
Guennadi

Quoted 291 lines

Ingo

---------------->
Subject: revert "PCI: make pci_bus a struct device"
From: Ingo Molnar <[email protected]>
Date: Thu Mar 06 08:12:56 CET 2008

revert:

commit fd7d1ced29e5beb88c9068801da7a362606d8273
PCI: make pci_bus a struct device

patch for testing purposes.

---
 drivers/pci/bus.c       |   19 +++-----------
 drivers/pci/pci-sysfs.c |    6 ++--
 drivers/pci/pci.h       |    2 -
 drivers/pci/probe.c     |   64 +++++++++++++++++++++++++++++-------------------
 drivers/pci/remove.c    |    6 +++-
 include/linux/pci.h     |    4 +--
 6 files changed, 53 insertions(+), 48 deletions(-)

Index: linux/drivers/pci/bus.c
===================================================================

drivers/pci/bus.c +4 -15
@@ -108,7 +108,6 @@ int pci_bus_add_device(struct pci_dev *d
 void pci_bus_add_devices(struct pci_bus *bus)
 {
 	struct pci_dev *dev;
-	struct pci_bus *child_bus;
 	int retval;
 
 	list_for_each_entry(dev, &bus->devices, bus_list) {
@@ -139,21 +138,11 @@ void pci_bus_add_devices(struct pci_bus 
 			       up_write(&pci_bus_sem);
 			}
 			pci_bus_add_devices(dev->subordinate);
-
-			/* register the bus with sysfs as the parent is now
-			 * properly registered. */
-			child_bus = dev->subordinate;
-			child_bus->dev.parent = child_bus->bridge;
-			retval = device_register(&child_bus->dev);
-			if (retval)
-				dev_err(&dev->dev, "Error registering pci_bus,"
-					" continuing...\n");
-			else
-				retval = device_create_file(&child_bus->dev,
-							&dev_attr_cpuaffinity);
+			retval = sysfs_create_link(&dev->subordinate->class_dev.kobj,
+						   &dev->dev.kobj, "bridge");
 			if (retval)
-				dev_err(&dev->dev, "Error creating cpuaffinity"
-					" file, continuing...\n");
+				dev_err(&dev->dev, "Error creating sysfs "
+					"bridge symlink, continuing...\n");
 		}
 	}
 }

Index: linux/drivers/pci/pci-sysfs.c
===================================================================

drivers/pci/pci-sysfs.c +3 -3
@@ -358,7 +358,7 @@ pci_read_legacy_io(struct kobject *kobj,
 		   char *buf, loff_t off, size_t count)
 {
         struct pci_bus *bus = to_pci_bus(container_of(kobj,
-                                                      struct device,
+                                                      struct class_device,
 						      kobj));
 
         /* Only support 1, 2 or 4 byte accesses */
@@ -383,7 +383,7 @@ pci_write_legacy_io(struct kobject *kobj
 		    char *buf, loff_t off, size_t count)
 {
         struct pci_bus *bus = to_pci_bus(container_of(kobj,
-						      struct device,
+						      struct class_device,
 						      kobj));
         /* Only support 1, 2 or 4 byte accesses */
         if (count != 1 && count != 2 && count != 4)
@@ -407,7 +407,7 @@ pci_mmap_legacy_mem(struct kobject *kobj
                     struct vm_area_struct *vma)
 {
         struct pci_bus *bus = to_pci_bus(container_of(kobj,
-                                                      struct device,
+                                                      struct class_device,
 						      kobj));
 
         return pci_mmap_legacy_page_range(bus, vma);

Index: linux/drivers/pci/pci.h
===================================================================

drivers/pci/pci.h +1 -1
@@ -64,7 +64,7 @@ static inline int pci_no_d1d2(struct pci
 }
 extern int pcie_mch_quirk;
 extern struct device_attribute pci_dev_attrs[];
-extern struct device_attribute dev_attr_cpuaffinity;
+extern struct class_device_attribute class_device_attr_cpuaffinity;
 
 /**
  * pci_match_one_device - Tell if a PCI device structure has a matching

Index: linux/drivers/pci/probe.c
===================================================================

drivers/pci/probe.c +39 -25
@@ -53,7 +53,7 @@ static void pci_create_legacy_files(stru
 		b->legacy_io->attr.mode = S_IRUSR | S_IWUSR;
 		b->legacy_io->read = pci_read_legacy_io;
 		b->legacy_io->write = pci_write_legacy_io;
-		device_create_bin_file(&b->dev, b->legacy_io);
+		class_device_create_bin_file(&b->class_dev, b->legacy_io);
 
 		/* Allocated above after the legacy_io struct */
 		b->legacy_mem = b->legacy_io + 1;
@@ -61,15 +61,15 @@ static void pci_create_legacy_files(stru
 		b->legacy_mem->size = 1024*1024;
 		b->legacy_mem->attr.mode = S_IRUSR | S_IWUSR;
 		b->legacy_mem->mmap = pci_mmap_legacy_mem;
-		device_create_bin_file(&b->dev, b->legacy_mem);
+		class_device_create_bin_file(&b->class_dev, b->legacy_mem);
 	}
 }
 
 void pci_remove_legacy_files(struct pci_bus *b)
 {
 	if (b->legacy_io) {
-		device_remove_bin_file(&b->dev, b->legacy_io);
-		device_remove_bin_file(&b->dev, b->legacy_mem);
+		class_device_remove_bin_file(&b->class_dev, b->legacy_io);
+		class_device_remove_bin_file(&b->class_dev, b->legacy_mem);
 		kfree(b->legacy_io); /* both are allocated here */
 	}
 }
@@ -81,27 +81,26 @@ void pci_remove_legacy_files(struct pci_
 /*
  * PCI Bus Class Devices
  */
-static ssize_t pci_bus_show_cpuaffinity(struct device *dev,
-					struct device_attribute *attr,
+static ssize_t pci_bus_show_cpuaffinity(struct class_device *class_dev,
 					char *buf)
 {
 	int ret;
 	cpumask_t cpumask;
 
-	cpumask = pcibus_to_cpumask(to_pci_bus(dev));
+	cpumask = pcibus_to_cpumask(to_pci_bus(class_dev));
 	ret = cpumask_scnprintf(buf, PAGE_SIZE, cpumask);
 	if (ret < PAGE_SIZE)
 		buf[ret++] = '\n';
 	return ret;
 }
-DEVICE_ATTR(cpuaffinity, S_IRUGO, pci_bus_show_cpuaffinity, NULL);
+CLASS_DEVICE_ATTR(cpuaffinity, S_IRUGO, pci_bus_show_cpuaffinity, NULL);
 
 /*
  * PCI Bus Class
  */
-static void release_pcibus_dev(struct device *dev)
+static void release_pcibus_dev(struct class_device *class_dev)
 {
-	struct pci_bus *pci_bus = to_pci_bus(dev);
+	struct pci_bus *pci_bus = to_pci_bus(class_dev);
 
 	if (pci_bus->bridge)
 		put_device(pci_bus->bridge);
@@ -110,7 +109,7 @@ static void release_pcibus_dev(struct de
 
 static struct class pcibus_class = {
 	.name		= "pci_bus",
-	.dev_release	= &release_pcibus_dev,
+	.release	= &release_pcibus_dev,
 };
 
 static int __init pcibus_class_init(void)
@@ -393,6 +392,7 @@ pci_alloc_child_bus(struct pci_bus *pare
 {
 	struct pci_bus *child;
 	int i;
+	int retval;
 
 	/*
 	 * Allocate a new bus, and inherit stuff from the parent..
@@ -408,12 +408,15 @@ pci_alloc_child_bus(struct pci_bus *pare
 	child->bus_flags = parent->bus_flags;
 	child->bridge = get_device(&bridge->dev);
 
-	/* initialize some portions of the bus device, but don't register it
-	 * now as the parent is not properly set up yet.  This device will get
-	 * registered later in pci_bus_add_devices()
-	 */
-	child->dev.class = &pcibus_class;
-	sprintf(child->dev.bus_id, "%04x:%02x", pci_domain_nr(child), busnr);
+	child->class_dev.class = &pcibus_class;
+	sprintf(child->class_dev.class_id, "%04x:%02x", pci_domain_nr(child), busnr);
+	retval = class_device_register(&child->class_dev);
+	if (retval)
+		goto error_register;
+	retval = class_device_create_file(&child->class_dev,
+					  &class_device_attr_cpuaffinity);
+	if (retval)
+		goto error_file_create;
 
 	/*
 	 * Set up the primary, secondary and subordinate
@@ -431,6 +434,12 @@ pci_alloc_child_bus(struct pci_bus *pare
 	bridge->subordinate = child;
 
 	return child;
+
+error_file_create:
+	class_device_unregister(&child->class_dev);
+error_register:
+	kfree(child);
+	return NULL;
 }
 
 struct pci_bus *__ref pci_add_new_bus(struct pci_bus *parent, struct pci_dev *dev, int busnr)
@@ -1083,27 +1092,32 @@ struct pci_bus * pci_create_bus(struct d
 		goto dev_reg_err;
 	b->bridge = get_device(dev);
 
-	b->dev.class = &pcibus_class;
-	b->dev.parent = b->bridge;
-	sprintf(b->dev.bus_id, "%04x:%02x", pci_domain_nr(b), bus);
-	error = device_register(&b->dev);
+	b->class_dev.class = &pcibus_class;
+	sprintf(b->class_dev.class_id, "%04x:%02x", pci_domain_nr(b), bus);
+	error = class_device_register(&b->class_dev);
 	if (error)
 		goto class_dev_reg_err;
-	error = device_create_file(&b->dev, &dev_attr_cpuaffinity);
+	error = class_device_create_file(&b->class_dev, &class_device_attr_cpuaffinity);
 	if (error)
-		goto dev_create_file_err;
+		goto class_dev_create_file_err;
 
 	/* Create legacy_io and legacy_mem files for this bus */
 	pci_create_legacy_files(b);
 
+	error = sysfs_create_link(&b->class_dev.kobj, &b->bridge->kobj, "bridge");
+	if (error)
+		goto sys_create_link_err;
+
 	b->number = b->secondary = bus;
 	b->resource[0] = &ioport_resource;
 	b->resource[1] = &iomem_resource;
 
 	return b;
 
-dev_create_file_err:
-	device_unregister(&b->dev);
+sys_create_link_err:
+	class_device_remove_file(&b->class_dev, &class_device_attr_cpuaffinity);
+class_dev_create_file_err:
+	class_device_unregister(&b->class_dev);
 class_dev_reg_err:
 	device_unregister(dev);
 dev_reg_err:

Index: linux/drivers/pci/remove.c
===================================================================

drivers/pci/remove.c +4 -2
@@ -74,8 +74,10 @@ void pci_remove_bus(struct pci_bus *pci_
 	list_del(&pci_bus->node);
 	up_write(&pci_bus_sem);
 	pci_remove_legacy_files(pci_bus);
-	device_remove_file(&pci_bus->dev, &dev_attr_cpuaffinity);
-	device_unregister(&pci_bus->dev);
+	class_device_remove_file(&pci_bus->class_dev,
+		&class_device_attr_cpuaffinity);
+	sysfs_remove_link(&pci_bus->class_dev.kobj, "bridge");
+	class_device_unregister(&pci_bus->class_dev);
 }
 EXPORT_SYMBOL(pci_remove_bus);
 

Index: linux/include/linux/pci.h
===================================================================

include/linux/pci.h +1 -1
@@ -275,13 +275,13 @@ struct pci_bus {
 	unsigned short  bridge_ctl;	/* manage NO_ISA/FBB/et al behaviors */
 	pci_bus_flags_t bus_flags;	/* Inherited by child busses */
 	struct device		*bridge;
-	struct device		dev;
+	struct class_device	class_dev;

include/linux/pci.h struct pci_bus { · line 287

Quoted 10 lines
include/linux/pci.h +1 -1
 	struct bin_attribute	*legacy_io; /* legacy I/O for this bus */
 	struct bin_attribute	*legacy_mem; /* legacy mem */
 };
 
 #define pci_bus_b(n)	list_entry(n, struct pci_bus, node)
-#define to_pci_bus(n)	container_of(n, struct pci_bus, dev)
+#define to_pci_bus(n)	container_of(n, struct pci_bus, class_dev)
 
 /*
  * Error values that may be returned by PCI functions.

---
Guennadi Liakhovetski

#17 lore raw
Quoted 15 lines · On Thu, 6 Mar 2008 20:55:25 +0100 (CET) Guennadi Liakhovetski…
Quoted 13 lines · Ingo Molnar
Quoted 4 lines · Rafael J. Wysocki

Bug-Entry : http://bugzilla.kernel.org/show_bug.cgi?id=10123
Subject : No power-off / reboot with 2.6.25-rcX (up to -rc3) kernels
Submitter : Guennadi Liakhovetski <[email protected]>
Date : 2008-02-27 08:15

Guennadi bisected this down to:

commit fd7d1ced29e5beb88c9068801da7a362606d8273
PCI: make pci_bus a struct device

Quoted 6 lines
Quoted 3 lines

and it's suspected that Andrew's poweroff problems might be related as well. Guennadi, Andrew, find below a manual revert of this change - does it make any difference?

Yes, this patch fixes both startup warnings and lets the system reboot and power off again!

Ingo's revert doesn't fix my machine-wont-power-off regression.

Reminder: what _does_ fix it is:

a) CONFIG_DETECT_SOFTLOCKUP=n or

b) This:

kernel/softlockup.c renamed from kernel/softlockup.c~softlockup-workaround +1 -0
@@ -289,6 +289,7 @@ cpu_callback(struct notifier_block *nfb,
 	case CPU_DEAD_FROZEN:
 		p = per_cpu(watchdog_task, hotcpu);
 		per_cpu(watchdog_task, hotcpu) = NULL;
+		msleep(1);
 		kthread_stop(p);
 		break;
 #endif /* CONFIG_HOTPLUG_CPU */

_

this one is unrelated to Greg's patch, I think. It's a timing thing.

#18 lore raw
Quoted 9 lines · On Thu, 6 Mar 2008 12:11:27 -0800 Andrew Morton <[email protected]>…

Reminder: what _does_ fix it is:

a) CONFIG_DETECT_SOFTLOCKUP=n or

b) This:

kernel/softlockup.c cpu_callback(struct notifier_block *nfb, · line 295

Quoted 7 lines
kernel/softlockup.c renamed from kernel/softlockup.c~softlockup-workaround +1 -0
@@ -289,6 +289,7 @@ cpu_callback(struct notifier_block *nfb,
 	case CPU_DEAD_FROZEN:
 		p = per_cpu(watchdog_task, hotcpu);
 		per_cpu(watchdog_task, hotcpu) = NULL;
+		msleep(1);
 		kthread_stop(p);
 		break;
 #endif /* CONFIG_HOTPLUG_CPU */

sysrq-t works: http://userweb.kernel.org/~akpm/x.txt

It shows that halt' is stuck in kthread_stop(), waiting for watchdog' to go away. But all the watchdog tasks are dreamily asleep, as if the wakeup didn't work.

I'd love to poke around in kgdb (what does kthread_stop_info.k point at?) but it seems that -mm's copy of kgdb got taken away when I wasn't looking. Can I have it back please?

(btw, it isn't compulsory that every cpu callback function be literally called "cpu_callback").

#19 lore raw
Quoted 3 lines · Andrew Morton

I'd love to poke around in kgdb (what does kthread_stop_info.k point at?) but it seems that -mm's copy of kgdb got taken away when I wasn't looking. Can I have it back please?

it's in the full x86.git or you can pick up the kgdb-light tree:

http://people.redhat.com/mingo/kgdb-light.git/README

Ingo

#20 lore raw
Quoted 11 lines · On Thu, 6 Mar 2008 21:59:51 +0100 Ingo Molnar <[email protected]> wrote:
Quoted 3 lines · Andrew Morton

I'd love to poke around in kgdb (what does kthread_stop_info.k point at?) but it seems that -mm's copy of kgdb got taken away when I wasn't looking. Can I have it back please?

it's in the full x86.git or you can pick up the kgdb-light tree:

http://people.redhat.com/mingo/kgdb-light.git/README

We'll see.

Meanwhile, further investigation show that cpu_callback() (the one in kernel/softlockup.c) is waiting on this thread:

watchdog/1 R running task 0 8 2 task_struct:ffff81025f1089e0
ffff81025f10deb0 0000000000000046 0000000000000000 0000000000000246 ffff81025f10de20 ffff81025f1089e0 ffff81025f1080c0 ffff81025f108d30 000000015f10de50 00000000ffff2adf ffffffffffffffff ffffffffffffffff

Call Trace:
 [<ffffffff80263290>] ? watchdog+0x0/0x1dc
 [<ffffffff802632d6>] watchdog+0x46/0x1dc
 [<ffffffff80263290>] ? watchdog+0x0/0x1dc
 [<ffffffff8024704d>] kthread+0x44/0x6b
 [<ffffffff8020cd88>] child_rip+0xa/0x12
 [<ffffffff80247009>] ? kthread+0x0/0x6b
 [<ffffffff8020cd7e>] ? child_rip+0x0/0x12

kthread_stop_info.k=ffff81025f1089e0

(gdb) l *0xffffffff802632d6
0xffffffff802632d6 is in watchdog (kernel/softlockup.c:229).
224 */
225 while (!kthread_should_stop()) {
226 touch_softlockup_watchdog();
227 schedule();
228
229 if (kthread_should_stop())
230 break;
231
232 if (this_cpu == check_cpu) {
233 if (sysctl_hung_task_timeout_secs)

so this watchdog thread seems to be runnable, but not running. What would cause this?

The only other runnable task is

events/2 R running task 0 29 2 task_struct:ffff81025f2da300
ffff81025f2e3ec0 0000000000000046 0000000000000286 ffff81000102ebc0 ffff81000102ebe0 ffff81025f2da300 ffff81025f147380 ffff81025f2da650 000000025f2e3e60 00000000ffff3310 ffff81000102ebc8 ffff81025f20e440

Call Trace:
 [<ffffffff802442ce>] ? worker_thread+0x0/0xe5
 [<ffffffff80244371>] worker_thread+0xa3/0xe5
 [<ffffffff80247364>] ? autoremove_wake_function+0x0/0x36
 [<ffffffff802442ce>] ? worker_thread+0x0/0xe5
 [<ffffffff8024704d>] kthread+0x44/0x6b
 [<ffffffff8020cd88>] child_rip+0xa/0x12
 [<ffffffff80247009>] ? kthread+0x0/0x6b
 [<ffffffff8020cd7e>] ? child_rip+0x0/0x12

which is on a different CPU.

#21 lore raw
Quoted 15 lines · On Thu, 6 Mar 2008 13:36:32 -0800 Andrew Morton <[email protected]>…
Quoted 11 lines · On Thu, 6 Mar 2008 21:59:51 +0100 Ingo Molnar <[email protected]> wrote:
Quoted 3 lines · Andrew Morton

I'd love to poke around in kgdb (what does kthread_stop_info.k point at?) but it seems that -mm's copy of kgdb got taken away when I wasn't looking. Can I have it back please?

it's in the full x86.git or you can pick up the kgdb-light tree:

http://people.redhat.com/mingo/kgdb-light.git/README

Quoted 6 lines

We'll see.

Meanwhile, further investigation show that cpu_callback() (the one in kernel/softlockup.c) is waiting on this thread:

watchdog/1 R running task 0 8 2 task_struct:ffff81025f1089e0

Note the "/1".

Quoted 27 lines

ffff81025f10deb0 0000000000000046 0000000000000000 0000000000000246 ffff81025f10de20 ffff81025f1089e0 ffff81025f1080c0 ffff81025f108d30 000000015f10de50 00000000ffff2adf ffffffffffffffff ffffffffffffffff

Call Trace:
 [<ffffffff80263290>] ? watchdog+0x0/0x1dc
 [<ffffffff802632d6>] watchdog+0x46/0x1dc
 [<ffffffff80263290>] ? watchdog+0x0/0x1dc
 [<ffffffff8024704d>] kthread+0x44/0x6b
 [<ffffffff8020cd88>] child_rip+0xa/0x12
 [<ffffffff80247009>] ? kthread+0x0/0x6b
 [<ffffffff8020cd7e>] ? child_rip+0x0/0x12

kthread_stop_info.k=ffff81025f1089e0

(gdb) l *0xffffffff802632d6
0xffffffff802632d6 is in watchdog (kernel/softlockup.c:229).
224 */
225 while (!kthread_should_stop()) {
226 touch_softlockup_watchdog();
227 schedule();
228
229 if (kthread_should_stop())
230 break;
231
232 if (this_cpu == check_cpu) {
233 if (sysctl_hung_task_timeout_secs)

Quoted 2 lines

so this watchdog thread seems to be runnable, but not running. What would cause this?

At the start of the sysrq-T trace we have:

sd 1:0:0:0: [sdb] Stopping disk
sd 0:0:0:0: [sda] Synchronizing SCSI cache sd 0:0:0:0: [sda] Stopping disk
ACPI: PCI interrupt for device 0000:05:00.1 disabled
ACPI: PCI interrupt for device 0000:05:00.0 disabled
ACPI: Preparing to enter system sleep state S5 Disabling non-boot CPUs ...
CPU 1 is now offline
SysRq : Show State
task PC stack pid father

So CPU 1 is offline. But the comatose watchdog thread is pinned to CPU 1. Could this be related to the problem? By what means is a task which is pinned to a going-away CPU handled? How is this guy supposed to ever run again?

#22 lore raw
Quoted 59 lines · Andrew Morton
Quoted 15 lines · On Thu, 6 Mar 2008 13:36:32 -0800 Andrew Morton <[email protected]>…
Quoted 11 lines · On Thu, 6 Mar 2008 21:59:51 +0100 Ingo Molnar <[email protected]> wrote:
Quoted 3 lines · Andrew Morton

I'd love to poke around in kgdb (what does kthread_stop_info.k point at?) but it seems that -mm's copy of kgdb got taken away when I wasn't looking. Can I have it back please?

it's in the full x86.git or you can pick up the kgdb-light tree:

http://people.redhat.com/mingo/kgdb-light.git/README

Quoted 6 lines

We'll see.

Meanwhile, further investigation show that cpu_callback() (the one in kernel/softlockup.c) is waiting on this thread:

watchdog/1 R running task 0 8 2 task_struct:ffff81025f1089e0

Note the "/1".

Quoted 27 lines

ffff81025f10deb0 0000000000000046 0000000000000000 0000000000000246 ffff81025f10de20 ffff81025f1089e0 ffff81025f1080c0 ffff81025f108d30 000000015f10de50 00000000ffff2adf ffffffffffffffff ffffffffffffffff

Call Trace:
 [<ffffffff80263290>] ? watchdog+0x0/0x1dc
 [<ffffffff802632d6>] watchdog+0x46/0x1dc
 [<ffffffff80263290>] ? watchdog+0x0/0x1dc
 [<ffffffff8024704d>] kthread+0x44/0x6b
 [<ffffffff8020cd88>] child_rip+0xa/0x12
 [<ffffffff80247009>] ? kthread+0x0/0x6b
 [<ffffffff8020cd7e>] ? child_rip+0x0/0x12

kthread_stop_info.k=ffff81025f1089e0

(gdb) l *0xffffffff802632d6
0xffffffff802632d6 is in watchdog (kernel/softlockup.c:229).
224 */
225 while (!kthread_should_stop()) {
226 touch_softlockup_watchdog();
227 schedule();
228
229 if (kthread_should_stop())
230 break;
231
232 if (this_cpu == check_cpu) {
233 if (sysctl_hung_task_timeout_secs)

Quoted 2 lines

so this watchdog thread seems to be runnable, but not running. What would cause this?

At the start of the sysrq-T trace we have:

Quoted 10 lines

sd 1:0:0:0: [sdb] Stopping disk
sd 0:0:0:0: [sda] Synchronizing SCSI cache sd 0:0:0:0: [sda] Stopping disk
ACPI: PCI interrupt for device 0000:05:00.1 disabled
ACPI: PCI interrupt for device 0000:05:00.0 disabled
ACPI: Preparing to enter system sleep state S5 Disabling non-boot CPUs ...
CPU 1 is now offline
SysRq : Show State
task PC stack pid father

I have been looking into a similar issue, which stops my system going into standy.

Quoted 5 lines

So CPU 1 is offline. But the comatose watchdog thread is pinned to CPU 1. Could this be related to the problem? By what means is a task which is pinned to a going-away CPU handled? How is this guy supposed to ever run again?

move_task_off_dead_cpu() should move that thread to another online cpu. But for some reason it isn't running.

thanks,
suresh

#23 lore raw
Quoted 11 lines · On Thu, 6 Mar 2008 15:13:31 -0800 Suresh Siddha <[email protected]>…
Quoted 10 lines

sd 1:0:0:0: [sdb] Stopping disk
sd 0:0:0:0: [sda] Synchronizing SCSI cache sd 0:0:0:0: [sda] Stopping disk
ACPI: PCI interrupt for device 0000:05:00.1 disabled
ACPI: PCI interrupt for device 0000:05:00.0 disabled
ACPI: Preparing to enter system sleep state S5 Disabling non-boot CPUs ...
CPU 1 is now offline
SysRq : Show State
task PC stack pid father

Quoted 2 lines

I have been looking into a similar issue, which stops my system going into standy.

OK.

Quoted 8 lines
Quoted 5 lines

So CPU 1 is offline. But the comatose watchdog thread is pinned to CPU 1. Could this be related to the problem? By what means is a task which is pinned to a going-away CPU handled? How is this guy supposed to ever run again?

move_task_off_dead_cpu() should move that thread to another online cpu. But for some reason it isn't running.

hm. What guarantees that kernel/sched.c:migration_call(CPU_DEAD) is called before kernel/softlockup.c:cpu_callback(CPU_DEAD)? Just the ordering in do_pre_smp_initcalls(), and notifier-chain behaviour, I guess.

#24 lore raw
In reply to Guennadi Liakhovetski
Quoted 15 lines · Guennadi Liakhovetski
Quoted 13 lines · Ingo Molnar
Quoted 4 lines · Rafael J. Wysocki

Bug-Entry : http://bugzilla.kernel.org/show_bug.cgi?id=10123
Subject : No power-off / reboot with 2.6.25-rcX (up to -rc3) kernels
Submitter : Guennadi Liakhovetski <[email protected]>
Date : 2008-02-27 08:15

Guennadi bisected this down to:

commit fd7d1ced29e5beb88c9068801da7a362606d8273
PCI: make pci_bus a struct device

Quoted 6 lines
Quoted 3 lines

and it's suspected that Andrew's poweroff problems might be related as well. Guennadi, Andrew, find below a manual revert of this change - does it make any difference?

Yes, this patch fixes both startup warnings and lets the system reboot and power off again!

Ok, I think we have two different problems here (as andrew showed.)

I'll work with Guennadi on his issue for now, as the pci patch shows there is still an issue here.

thanks,

greg k-h

#25 lore raw
In reply to Rafael J. Wysocki
Quoted 10 lines · On 03.03.2008 03:16 Rafael J. Wysocki wrote:

Unresolved regressions
----------------------

Bug-Entry : http://bugzilla.kernel.org/show_bug.cgi?id=10021
Subject : Linux 2.6.25-rc2 regression: LVM cannot find volume group
Submitter : Tilman Schmidt <[email protected]>
Date : 2008-02-16 20:14
References : http://lkml.org/lkml/2008/2/16/208

Jiri Slaby <[email protected]>

Fixed in 2.6.25-rc4 by the introduction of CONFIG_SYSFS_DEPRECATED_V2.

Thanks,
Tilman

Signature · 4 lines
Tilman Schmidt                          E-Mail: [email protected]
Bonn, Germany
Diese Nachricht besteht zu 100% aus wiederverwerteten Bits.
Ungeöffnet mindestens haltbar bis: (siehe Rückseite)
#26 lore raw
Quoted 13 lines · Tilman Schmidt
Quoted 10 lines · On 03.03.2008 03:16 Rafael J. Wysocki wrote:

Unresolved regressions
----------------------

Bug-Entry : http://bugzilla.kernel.org/show_bug.cgi?id=10021
Subject : Linux 2.6.25-rc2 regression: LVM cannot find volume group
Submitter : Tilman Schmidt <[email protected]>
Date : 2008-02-16 20:14
References : http://lkml.org/lkml/2008/2/16/208

Jiri Slaby <[email protected]>

Quoted 1 line

Fixed in 2.6.25-rc4 by the introduction of CONFIG_SYSFS_DEPRECATED_V2.

Already closed.

Thanks,
Rafael