Bug#622325: linux-image-2.6.38-2-orion5x: Problem With I2C
* Arnaud Ebalard a...@natisbad.org [2013-06-25 22:52]: Either wait for 3.11 or pick 4243fa0bad551b8c8d4ff7104e8fd557ae848845 (I2C: mv64xxx: fix race between FSM/interrupt and process context) from my for-next branch. (Note the other patches from Russell, probably all worth to pick). Would be nice if you could pass the info to the Debian bug as well. I can confirm that 4243fa0b is a valid fix for the bus i2c lock problem with marvell mv64xxx i2c controller driver (note that I also applied other patches from Russel, as this commit depends on some of those). Maybe you should give it a try and consider it a long term solution for #622325. Great, thanks for your report, Arnaud! Ben, or Arnaud (Patard), can you please add that fix to the wheezy kernel. -- Martin Michlmayr http://www.cyrius.com/ -- To UNSUBSCRIBE, email to debian-kernel-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/20130812192934.gf32...@jirafa.cyrius.com
Bug#622325: linux-image-2.6.38-2-orion5x: Problem With I2C
Hi guys, While debugging a similar mv64xxx i2c bus lock on a Netgear ReadyNAS 102 NAS (based on a Marvell Armada 370 SoC), I ended up on #622325 bug report. Reverting eda6bee6c7 makes the issue disappear for me and also for Simon (in Cc:) for another Marvell-based NAS with the same i2c controller (its issue [1] appears different but is fixed in the same way by the revert). Both on current 3.10-rc kernels (not tested previous one though). [1] http://www.spinics.net/lists/arm-kernel/msg254423.html Note that I also tested d295a86e but it does not fix things for me. I reported the issue to i2c kernel people and got the following suggestion from Wolfram in (CC:) [2]: does nothing fancy either. While debugging I ended up on Debian bug #622325 (http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=622325) which made the driver a little less suspect when a revert of commit eda6bee6c7 fixed the issue for me. Either wait for 3.11 or pick 4243fa0bad551b8c8d4ff7104e8fd557ae848845 (I2C: mv64xxx: fix race between FSM/interrupt and process context) from my for-next branch. (Note the other patches from Russell, probably all worth to pick). Would be nice if you could pass the info to the Debian bug as well. [2]: http://www.spinics.net/lists/arm-kernel/msg254677.html I can confirm that 4243fa0b is a valid fix for the bus i2c lock problem with marvell mv64xxx i2c controller driver (note that I also applied other patches from Russel, as this commit depends on some of those). Maybe you should give it a try and consider it a long term solution for #622325. Cheers, a+ -- To UNSUBSCRIBE, email to debian-kernel-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/8761x1oquj@natisbad.org
Bug#622325: linux-image-2.6.38-2-orion5x: Problem With I2C
* Roman Mamedov r...@romanrm.ru [2012-01-31 02:38]: Were you able to try backing out eda6bee6c7 (i2c-mv64xxx: send repeated START between messages in xfer) as Guenter suggested? I have finally managed to do that, and indeed, this problem is solved by rolling-back that commit. The same problem was observed with the current vanilla 3.2.2, and only after the roll-back it started working fine: Sounds like this patch might help: https://git.kernel.org/cgit/linux/kernel/git/torvalds/linux.git/commit/drivers/i2c/busses/i2c-mv64xxx.c?id=d295a86eab200b3f0c513e78dbe1f189fd32d397 Roman, can you put eda6bee6c7 back in and add this patch and see if that works? -- Martin Michlmayr http://www.cyrius.com/ -- To UNSUBSCRIBE, email to debian-kernel-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/20130604195629.ga1...@jirafa.cyrius.com
Bug#622325: linux-image-2.6.38-2-orion5x: Problem With I2C
On Mon, 23 Jan 2012 14:27:50 -0600 Jonathan Nieder jrnie...@gmail.com wrote: Were you able to try backing out eda6bee6c7 (i2c-mv64xxx: send repeated START between messages in xfer) as Guenter suggested? I have finally managed to do that, and indeed, this problem is solved by rolling-back that commit. The same problem was observed with the current vanilla 3.2.2, and only after the roll-back it started working fine: root@mahou:~# uname -a Linux mahou 3.2.2orion5x-rm1+ #3 Mon Jan 30 19:29:49 YEKT 2012 armv5tel GNU/Linux root@mahou:~# sensors g760a-i2c-0-3e Adapter: mv64xxx_i2c adapter fan1:4681 RPM lm75-i2c-0-48 Adapter: mv64xxx_i2c adapter temp1:+42.0°C (high = +80.0°C, hyst = +75.0°C) No errors in dmesg. -- With respect, Roman ~~~ Stallman had a printer, with code he could not see. So he began to tinker, and set the software free. signature.asc Description: PGP signature
Bug#622325: linux-image-2.6.38-2-orion5x: Problem With I2C
On Sun, 1 Jan 2012 01:42:36 -0600 Jonathan Nieder jrnie...@gmail.com wrote: This bug is still present for me in the latest kernel version in Debian testing (3.1.6). Passed upstream. But now I notice that there has been some upstream work in this area recently (v3.2-rc7~23^2~1, rtc: m41t80: Workaround broken alarm functionality, 2011-12-12). Please test v3.2-rc7 from experimental and let the upstream folks know how it fares (by replying-to-all on the other side thread). Hello, I have now upgraded to linux-image-3.1.0-1-orion5x (3.1.8-2) from testing. According to http://www.kernel.org/pub/linux/kernel/v3.x/ChangeLog-3.1.8 it has the rtc: m41t80: Workaround broken alarm functionality patch included. However, it does not solve the problem. The messages continue to appear: [ 34.442094] mv643xx_eth_port mv643xx_eth_port.0: eth0: link up, 1000 Mb/s, full duplex, flow control disabled [ 34.452167] ADDRCONF(NETDEV_CHANGE): eth0: link becomes ready [ 35.107852] i2c i2c-0: mv64xxx: I2C bus locked, block: 1, time_left: 0 [ 37.107743] i2c i2c-0: mv64xxx: I2C bus locked, block: 1, time_left: 0 [ 41.007743] i2c i2c-0: mv64xxx: I2C bus locked, block: 1, time_left: 0 [ 43.007755] i2c i2c-0: mv64xxx: I2C bus locked, block: 1, time_left: 0 [ 45.007747] i2c i2c-0: mv64xxx: I2C bus locked, block: 1, time_left: 0 [ 47.007744] i2c i2c-0: mv64xxx: I2C bus locked, block: 1, time_left: 0 [ 50.090809] mv643xx_eth_port mv643xx_eth_port.0: eth0: link up, 1000 Mb/s, full duplex, flow control disabled [ 85.637697] i2c i2c-0: mv64xxx: I2C bus locked, block: 1, time_left: 0 [ 87.637699] i2c i2c-0: mv64xxx: I2C bus locked, block: 1, time_left: 0 [ 89.637693] i2c i2c-0: mv64xxx: I2C bus locked, block: 1, time_left: 0 [ 91.647701] i2c i2c-0: mv64xxx: I2C bus locked, block: 1, time_left: 0 [ 93.647738] i2c i2c-0: mv64xxx: I2C bus locked, block: 1, time_left: 0 [ 95.647701] i2c i2c-0: mv64xxx: I2C bus locked, block: 1, time_left: 0 Fan does not rotate (verified visually) and the temperature sensor does not work: # sensors g760a-i2c-0-3e Adapter: mv64xxx_i2c adapter fan1: 0 RPM lm75-i2c-0-48 Adapter: mv64xxx_i2c adapter temp1: +0.0°C (high = +0.0°C, hyst = +0.0°C) -- With respect, Roman ~~~ Stallman had a printer, with code he could not see. So he began to tinker, and set the software free. signature.asc Description: PGP signature
Bug#622325: linux-image-2.6.38-2-orion5x: Problem With I2C
found 622325 linux-2.6/3.1.8-2 quit Roman Mamedov wrote: Jonathan Nieder jrnie...@gmail.com wrote: Passed upstream. But now I notice that there has been some upstream work in this area recently (v3.2-rc7~23^2~1, rtc: m41t80: Workaround broken alarm functionality, 2011-12-12). [...] I have now upgraded to linux-image-3.1.0-1-orion5x (3.1.8-2) from testing. According to http://www.kernel.org/pub/linux/kernel/v3.x/ChangeLog-3.1.8 it has the rtc: m41t80: Workaround broken alarm functionality patch included. However, it does not solve the problem. The messages continue to appear: Great. Please let upstream know by replying-to-all on the other side thread[1]. Were you able to try backing out eda6bee6c7 (i2c-mv64xxx: send repeated START between messages in xfer) as Guenter suggested? It works like this: 1. Build a stock upstream kernel: apt-get install git build-essential git clone \ git://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git cd linux cp /boot/config-$(uname -r) .config; # stock configuration make localmodconfig; # minimal configuration make deb-pkg; # can use -jn to build faster if you have multiple cores dpkg -i ../name of package; # install the resulting package reboot ... test test test ... If at this point it works, the bug was fixed or upstream or it is a Debian-specific or configuration-specific problem. Yay! 2. Back out the patch and compare results: git revert eda6bee6c7 make deb-pkg; # perhaps with -jn dpkg -i ../name of package; # install the resulting package reboot ... test test test ... Many thanks, and hope that helps. Sincerely, Jonathan [1] http://thread.gmane.org/gmane.linux.kernel/1234150 -- To UNSUBSCRIBE, email to debian-kernel-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/20120123202749.GD20833@burratino
Bug#622325: linux-image-2.6.38-2-orion5x: Problem With I2C
On Mon, 23 Jan 2012 14:27:50 -0600 Jonathan Nieder jrnie...@gmail.com wrote: Were you able to try backing out eda6bee6c7 (i2c-mv64xxx: send repeated START between messages in xfer) as Guenter suggested? It works like this: Hello, If this had been a regular x86 machine I would have done this long ago :) But I don't currently have a working ARM crosscompile environment [and the one I'd trust enough to create non-bricking kernels], and building a kernel on the device itself, with just 64 MB of RAM and 2GB of slow USB flash storage, might not work well. I'll see if I can solve either problem in one way or the other, and will post an update some time later. -- With respect, Roman ~~~ Stallman had a printer, with code he could not see. So he began to tinker, and set the software free. signature.asc Description: PGP signature
Bug#622325: linux-image-2.6.38-2-orion5x: Problem With I2C
Roman Mamedov wrote: Jonathan Nieder jrnie...@gmail.com wrote: Were you able to try backing out eda6bee6c7 (i2c-mv64xxx: send repeated START between messages in xfer) as Guenter suggested? It works like this: [...] If this had been a regular x86 machine I would have done this long ago :) But I don't currently have a working ARM crosscompile environment Ah, ok. My only hint is that you can cheat with qemu + debootstrap to make a chroot on an x86 machine. :) Not sure how slow that is, though; the folks on debian-arm@ or debian-embed...@lists.debian.org might be able to help. -- To UNSUBSCRIBE, email to debian-kernel-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/20120124044942.GB27596@burratino
Bug#622325: linux-image-2.6.38-2-orion5x: Problem With I2C
On Sun, 1 Jan 2012 01:42:36 -0600 Jonathan Nieder jrnie...@gmail.com wrote: Passed upstream. But now I notice that there has been some upstream work in this area recently (v3.2-rc7~23^2~1, rtc: m41t80: Workaround broken alarm functionality, 2011-12-12). Please test v3.2-rc7 from experimental and let the upstream folks know how it fares (by replying-to-all on the other side thread). Hello, Thank you for looking into this and for the information about the 3.2 possible fix. Unfortunately I can't try that kernel easily at the moment, as there doesn't seem to be an orion5x variant of the package in experimental. Some more info: - During my testing I have found that after booting with a 'bad' kernel, a full power-off of the device including disconnecting the power supply from the mains might be required, before a 'good' kernel will work properly. - This might have been mentioned, but the main reason this bug concerns me, is that the temperature sensor and the fan PWM control do not work on the 'bad' kernels (i.e. it's not just the periodic dmesg error). The fan does not turn on, which risks overheating the device and/or drives. -- With respect, Roman ~~~ Stallman had a printer, with code he could not see. So he began to tinker, and set the software free. signature.asc Description: PGP signature
Bug#622325: linux-image-2.6.38-2-orion5x: Problem With I2C
severity 622325 important quit Roman Mamedov wrote: Unfortunately I can't try that kernel easily at the moment, as there doesn't seem to be an orion5x variant of the package in experimental. Oh, right.[1] Let's wait to hear what the developers come up with, then. Thanks, Jonathan [1] https://buildd.debian.org/status/package.php?p=linux-2.6suite=experimental -- To UNSUBSCRIBE, email to debian-kernel-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/20120101091729.gc4...@elie.hsd1.il.comcast.net
Bug#622325: linux-image-2.6.38-2-orion5x: Problem With I2C
Hi again, Roman Mamedov wrote: This bug is still present for me in the latest kernel version in Debian testing (3.1.6). Passed upstream. But now I notice that there has been some upstream work in this area recently (v3.2-rc7~23^2~1, rtc: m41t80: Workaround broken alarm functionality, 2011-12-12). Please test v3.2-rc7 from experimental and let the upstream folks know how it fares (by replying-to-all on the other side thread). Hope that helps, Jonathan -- To UNSUBSCRIBE, email to debian-kernel-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/20120101074236.gc2...@elie.hsd1.il.comcast.net
Bug#622325: linux-image-2.6.38-2-orion5x: Problem With I2C
Hi, It should be fixed the 2.6.39 kernel uploaded to unstable. Can you test it, please ? Thanks, Arnaud -- To UNSUBSCRIBE, email to debian-kernel-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/87hb8iq0ta@lebrac.rtp-net.org
Bug#622325: linux-image-2.6.38-2-orion5x: Problem With I2C
Package: linux-2.6 Version: 2.6.38-3 Severity: normal syslog tell me every few seconds: i2c i2c-0: mv64xxx: I2C bus locked, block: 1, time_left: 0 I found a mailing list entry here but no result: http://comments.gmane.org/gmane.linux.ports.arm.kernel/111480 -- Package-specific info: ** Version: Linux version 2.6.38-2-orion5x (Debian 2.6.38-3) (b...@decadent.org.uk) (gcc version 4.4.5 (Debian 4.4.5-15) ) #1 Fri Apr 8 01:22:30 UTC 2011 ** Command line: root=/dev/ram console=ttyS0,115200 :::DB88FXX81:egiga0:none ** Not tainted ** Kernel log: [ 513.731669] i2c i2c-0: mv64xxx: I2C bus locked, block: 1, time_left: 0 [ 525.761632] i2c i2c-0: mv64xxx: I2C bus locked, block: 1, time_left: 0 [ 527.761615] i2c i2c-0: mv64xxx: I2C bus locked, block: 1, time_left: 0 [ 529.761600] i2c i2c-0: mv64xxx: I2C bus locked, block: 1, time_left: 0 [ 531.761594] i2c i2c-0: mv64xxx: I2C bus locked, block: 1, time_left: 0 [ 533.761600] i2c i2c-0: mv64xxx: I2C bus locked, block: 1, time_left: 0 [ 535.761649] i2c i2c-0: mv64xxx: I2C bus locked, block: 1, time_left: 0 [ 537.781583] i2c i2c-0: mv64xxx: I2C bus locked, block: 1, time_left: 0 [ 540.801565] i2c i2c-0: mv64xxx: I2C bus locked, block: 1, time_left: 0 [ 542.801528] i2c i2c-0: mv64xxx: I2C bus locked, block: 1, time_left: 0 [ 544.801538] i2c i2c-0: mv64xxx: I2C bus locked, block: 1, time_left: 0 [ 546.801533] i2c i2c-0: mv64xxx: I2C bus locked, block: 1, time_left: 0 [ 559.071550] i2c i2c-0: mv64xxx: I2C bus locked, block: 1, time_left: 0 [ 561.071476] i2c i2c-0: mv64xxx: I2C bus locked, block: 1, time_left: 0 [ 563.071426] i2c i2c-0: mv64xxx: I2C bus locked, block: 1, time_left: 0 [ 565.141411] i2c i2c-0: mv64xxx: I2C bus locked, block: 1, time_left: 0 [ 567.141452] i2c i2c-0: mv64xxx: I2C bus locked, block: 1, time_left: 0 [ 569.141382] i2c i2c-0: mv64xxx: I2C bus locked, block: 1, time_left: 0 [ 571.151412] i2c i2c-0: mv64xxx: I2C bus locked, block: 1, time_left: 0 [ 574.261367] i2c i2c-0: mv64xxx: I2C bus locked, block: 1, time_left: 0 [ 576.261347] i2c i2c-0: mv64xxx: I2C bus locked, block: 1, time_left: 0 [ 578.261366] i2c i2c-0: mv64xxx: I2C bus locked, block: 1, time_left: 0 [ 580.261343] i2c i2c-0: mv64xxx: I2C bus locked, block: 1, time_left: 0 [ 592.501221] i2c i2c-0: mv64xxx: I2C bus locked, block: 1, time_left: 0 [ 594.501222] i2c i2c-0: mv64xxx: I2C bus locked, block: 1, time_left: 0 [ 596.501199] i2c i2c-0: mv64xxx: I2C bus locked, block: 1, time_left: 0 [ 598.521202] i2c i2c-0: mv64xxx: I2C bus locked, block: 1, time_left: 0 [ 600.521200] i2c i2c-0: mv64xxx: I2C bus locked, block: 1, time_left: 0 [ 602.521164] i2c i2c-0: mv64xxx: I2C bus locked, block: 1, time_left: 0 [ 604.541156] i2c i2c-0: mv64xxx: I2C bus locked, block: 1, time_left: 0 [ 607.621131] i2c i2c-0: mv64xxx: I2C bus locked, block: 1, time_left: 0 [ 609.621184] i2c i2c-0: mv64xxx: I2C bus locked, block: 1, time_left: 0 [ 611.621117] i2c i2c-0: mv64xxx: I2C bus locked, block: 1, time_left: 0 [ 613.621119] i2c i2c-0: mv64xxx: I2C bus locked, block: 1, time_left: 0 [ 625.651007] i2c i2c-0: mv64xxx: I2C bus locked, block: 1, time_left: 0 [ 627.651014] i2c i2c-0: mv64xxx: I2C bus locked, block: 1, time_left: 0 [ 629.651044] i2c i2c-0: mv64xxx: I2C bus locked, block: 1, time_left: 0 [ 631.650988] i2c i2c-0: mv64xxx: I2C bus locked, block: 1, time_left: 0 [ 633.651001] i2c i2c-0: mv64xxx: I2C bus locked, block: 1, time_left: 0 [ 635.650996] i2c i2c-0: mv64xxx: I2C bus locked, block: 1, time_left: 0 [ 637.660944] i2c i2c-0: mv64xxx: I2C bus locked, block: 1, time_left: 0 [ 640.680941] i2c i2c-0: mv64xxx: I2C bus locked, block: 1, time_left: 0 [ 642.680922] i2c i2c-0: mv64xxx: I2C bus locked, block: 1, time_left: 0 [ 644.680915] i2c i2c-0: mv64xxx: I2C bus locked, block: 1, time_left: 0 [ 646.680911] i2c i2c-0: mv64xxx: I2C bus locked, block: 1, time_left: 0 [ 658.710827] i2c i2c-0: mv64xxx: I2C bus locked, block: 1, time_left: 0 [ 660.710832] i2c i2c-0: mv64xxx: I2C bus locked, block: 1, time_left: 0 [ 662.710805] i2c i2c-0: mv64xxx: I2C bus locked, block: 1, time_left: 0 [ 664.710788] i2c i2c-0: mv64xxx: I2C bus locked, block: 1, time_left: 0 [ 666.710780] i2c i2c-0: mv64xxx: I2C bus locked, block: 1, time_left: 0 [ 668.710762] i2c i2c-0: mv64xxx: I2C bus locked, block: 1, time_left: 0 [ 670.720749] i2c i2c-0: mv64xxx: I2C bus locked, block: 1, time_left: 0 [ 673.740754] i2c i2c-0: mv64xxx: I2C bus locked, block: 1, time_left: 0 [ 675.740752] i2c i2c-0: mv64xxx: I2C bus locked, block: 1, time_left: 0 [ 677.740701] i2c i2c-0: mv64xxx: I2C bus locked, block: 1, time_left: 0 [ 679.740725] i2c i2c-0: mv64xxx: I2C bus locked, block: 1, time_left: 0 [ 691.770638] i2c i2c-0: mv64xxx: I2C bus locked, block: 1, time_left: 0 [ 693.770617] i2c i2c-0: mv64xxx: I2C bus locked, block: 1, time_left: 0 [ 695.770630] i2c i2c-0: mv64xxx: I2C bus locked, block: 1, time_left: 0 [ 697.770587] i2c i2c-0: mv64xxx: I2C bus locked, block: 1, time_left: 0 [
Bug#622325: linux-image-2.6.38-2-orion5x: Problem With I2C
* Michael Ott mich...@king-coder.de [2011-04-12 09:00]: syslog tell me every few seconds: i2c i2c-0: mv64xxx: I2C bus locked, block: 1, time_left: 0 I found a mailing list entry here but no result: http://comments.gmane.org/gmane.linux.ports.arm.kernel/111480 Can one of you bisect the kernel to see when the bug was introduced? -- Martin Michlmayr http://www.cyrius.com/ -- To UNSUBSCRIBE, email to debian-kernel-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/20110412212128.gh7...@jirafa.cyrius.com
