Summary: | kernel 2.6.36-gentoo-r5 libata driver broken support of CD-R/RW media | ||
---|---|---|---|
Product: | Gentoo Linux | Reporter: | Sergey S. Starikoff <Ikonta> |
Component: | [OLD] Core system | Assignee: | Gentoo Kernel Bug Wranglers and Kernel Maintainers <kernel> |
Status: | RESOLVED CANTFIX | ||
Severity: | minor | ||
Priority: | Normal | ||
Version: | unspecified | ||
Hardware: | All | ||
OS: | Linux | ||
URL: | http://gentoo.ru/node/22590 | ||
Whiteboard: | |||
Package list: | Runtime testing required: | --- | |
Attachments: | 2.6.38-gentoo-r6 kernel config file |
Description
Sergey S. Starikoff
2011-04-14 07:45:37 UTC
It's more common issue. The same problem I've seen just noe on my new PC: Linux tux 2.6.37-gentoo-r4 #2 SMP Sat Apr 16 22:30:42 MSD 2011 x86_64 Intel(R) Core(TM) i5 CPU 661 @ 3.33GHz GenuineIntel GNU/Linux with 'Optiarc ' 'DVD RW AD-7263S ' '1.03' Removable CD-ROM SATA DVD recorder. The media was the same: high speed Verbatim CD-RW disk. Similiar issue with written won windows media (on another pc with win reads also normally). The atip of issued CD-R media is: ATIP info from disk: Indicated writing power: 5 Disk Is not unrestricted Disk Is not erasable Disk sub type: Medium Type B, low Beta category (B-) (4) ATIP start of lead in: -12369 (97:17/06) ATIP start of lead out: 359849 (79:59/74) Disk type: Short strategy type (Phthalocyanine or similar) Manuf. index: 69 Manufacturer: Moser Baer India Limited Manufacturer is guessed because of the orange forum embargo. The orange forum likes to get money for recent information. The information for this media may not be correct. Can you try gentoo-sources-2.6.38-r3? (In reply to comment #3) > Can you try gentoo-sources-2.6.38-r3? Currently I've checked with gentoo-sources-2.6.38-r4 on amd64 keyword system. The issue seems to have an easy workaround: $ eject (here put a disk in a drive) $ date && mount /mnt/cdrom/ Сбт Май 7 12:17:11 MSD 2011 mount: no medium found on /dev/sr0 $ date && mount /mnt/cdrom/ Сбт Май 7 12:17:16 MSD 2011 mount: no medium found on /dev/sr0 $ date && mount /mnt/cdrom/ Сбт Май 7 12:17:18 MSD 2011 mount: no medium found on /dev/sr0 $ date && mount /mnt/cdrom/ Сбт Май 7 12:17:19 MSD 2011 mount: no medium found on /dev/sr0 $ date && mount /mnt/cdrom/ Сбт Май 7 12:17:21 MSD 2011 mount: no medium found on /dev/sr0 $ date && mount /mnt/cdrom/ Сбт Май 7 12:17:22 MSD 2011 So, it's enough just to wait about 10-12 seconds after insertind and CD is mounted normally. But earlier (in a time AFAIR of gentoo-sources-2.6.32) it was successfully mount when inserting disk in a drive was invoked by mount command. (In reply to comment #4) > Currently I've checked with gentoo-sources-2.6.38-r4 on amd64 keyword system. > > The issue seems to have an easy workaround: > > ... > > So, it's enough just to wait about 10-12 seconds after insertind and CD is > mounted normally. > > But earlier (in a time AFAIR of gentoo-sources-2.6.32) it was successfully > mount when inserting disk in a drive was invoked by mount command. Similiar behaviour with 2.6.37-gentoo-r4 (SATA DVD/RW drive). Tomorrow I'll check it on installation with IDE drive (in libata mode). Can you post the dmesg output? Does it log anything interesting there? (In reply to comment #6) > Can you post the dmesg output? Does it log anything interesting there? I could. But there is really nothing. Start point (dmesg | tail -n 2): [11490.817570] ISO 9660 Extensions: Microsoft Joliet Level 3 [11490.855302] ISO 9660 Extensions: RRIP_1991A $ mount /mnt/cdrom/ mount: no medium found on /dev/sr0 Wait for a time, repeat mount attemp (this time successful): [11490.817570] ISO 9660 Extensions: Microsoft Joliet Level 3 [11490.855302] ISO 9660 Extensions: RRIP_1991A [15616.173769] ISO 9660 Extensions: Microsoft Joliet Level 3 [15616.188183] ISOFS: changing to secondary root (In reply to comment #5) > Tomorrow I'll check it on installation with IDE drive (in libata mode). 'TSSTcorp' 'CDW/DVD SH-M522C' 'TS04' Removable CD-ROM drive 2.6.36-gentoo-r5 kernel Similiar behaviour (insert disk into drive, wait about 1 minute, after that disk mountes successfully). I'm not sure if this should be reported upstream. Can you try booting more recent kernels (2.6.39 for example)? It might help too if we had the .configs, emerge --info, and the full dmesg output. Thanks. Created attachment 277197 [details] 2.6.38-gentoo-r6 kernel config file (In reply to comment #9) > I'm not sure if this should be reported upstream. > Can you try booting more recent kernels (2.6.39 for example)? Currently I've reproduced the bug with current stable (x86) kernel (2.6.38-gentoo-r6). If it's necessary, I'll check any (2.6 branch) kernel you want to test. > It might help too if we had the .configs, emerge --info, and the full dmesg > output. Current cernel config is attached. emerge --info was listed previously (or you need the most recent one?). Full dmesg output contains mostly useless for this ussue data. About mounting CD-ROM it tells the following: $ dmesg ... cdrom: sr0: mrw address space DMA selected cdrom: open: mrw_status 'not mrw' ISO 9660 Extensions: Microsoft Joliet Level 3 ISOFS: changing to secondary root What happens on the latest 2.6.39-r1 kernel? (In reply to comment #11) > What happens on the latest 2.6.39-r1 kernel? The same. And one more: the timeouts of handling eject utility calls seemed to be large (encreased in about two or three times). With kernel 2.6.39-gentoo-r2 bthaviour seems to be changed: $ mount /mnt/cdrom/ mount: no medium found on /dev/sr0 wait a few seconds (error changes): $ mount /mnt/cdrom/ mount: block device /dev/sr0 is write-protected, mounting read-only mount: /dev/sr0 already mounted or /mnt/cdrom busy wait some more seconds (success mount): $ mount /mnt/cdrom/ mount: block device /dev/sr0 is write-protected, mounting read-only Well, you could try with a 3.0.x kernel, just in case. Maybe stracing the mount command could provide some useful information. And, do you remember which was the last working kernel? Your initial bug report was for 2.6.36-gentoo-r5. Does previous 2.6.36 kernels work, or maybe 2.6.35? Thanks. (In reply to comment #14) > Well, you could try with a 3.0.x kernel, just in case. Currently with 2.6.39-gentoo-r3 the same issue. Trying to upgrade to 3.0.3 (with the same kernel conf) doesn't solve it. > Maybe stracing the mount command could provide some useful information. Trying to get strace of mount I've discovered that it completely breaks mount operability: using strace makes fail not only oryginally failed attemps but also oryginally (otherwise) successfull. Maybe it wiil be helpful: I still use manual mount via /etc/fstab: $ grep cdrom /etc/fstab /dev/cdrom /mnt/cdrom iso9660 noauto,user,iocharset=koi8-r,mode=0444 0 0 (my another workstation with the same issue is localized in utf8) > And, do you remember which was the last working kernel? Your initial bug report > was for 2.6.36-gentoo-r5. Does previous 2.6.36 kernels work, or maybe 2.6.35? That time I'vu used CD rather rare and it will be difficult to remember where it worked. May be about 2.6.32?.. Am I to try to downgrade kernel (at least on one of my workstations it should be enough easy)? If it's possible, yes, just to confirm that you can't reproduce the issue with eg 2.6.32 kernels. Thanks. (In reply to comment #16) > If it's possible, yes, just to confirm that you can't reproduce the issue with > eg 2.6.32 kernels. > > Thanks. With 2.6.32-gentoo-r29 the same issue. I've seen how thunar displays blank CD/DVD media. After that I've setup for test purposes automounting. And see that it works. It seems, that the pause is needed for udev to detect media status. So (remembering currant udev requirements to kernel versions) I think, that the issue should appear abount first required by current udev kernel (AFAIR 2.6.27). Where/how it should be fixed (or we need to update documentation) --- I don't see good solution. Is everything at this point? (In reply to comment #19) > Is everything at this point? It seems to be not a bug, but a feature. 3.2.1-gentoo-r2 generally the same behaviour (time-out from inserting media to mounting it). On my another PC I've set up automounter and see the similiar time-out. So, may be, it will be right to solve it not on kernel-side, but on mount utility side. Please reopen and reassign as appropriate if anything further is needed here. |