[Bug 233998] New: USB Device Detection for drive loops endlessly.
https://bugzilla.novell.com/show_bug.cgi?id=233998 Summary: USB Device Detection for drive loops endlessly. Product: openSUSE 10.2 Version: RC 5 Platform: i686 OS/Version: Linux Status: NEW Severity: Normal Priority: P5 - None Component: Kernel AssignedTo: gregkh@novell.com ReportedBy: eich@novell.com QAContact: qa@suse.de Using kernel 2.6.18.2-34-default the device detection of my USB drive loops endlessly with the messages: sd 11:0:0:0: Attached scsi disk sda sd 11:0:0:0: Attached scsi generic sg0 type 0 usb-storage: device scan complete usb 1-3: new high speed USB device using ehci_hcd and address 21 usb 1-3: device descriptor read/64, error -71 usb 1-3: device descriptor read/64, error -71 usb 1-3: new high speed USB device using ehci_hcd and address 22 usb 1-3: new device found, idVendor=07ab, idProduct=fccd usb 1-3: new device strings: Mfr=1, Product=2, SerialNumber=3 usb 1-3: Product: FHD-Classic usb 1-3: Manufacturer: Freecom Technologies usb 1-3: SerialNumber: FW436_0C5B16F342BBC59A usb 1-3: configuration #1 chosen from 1 choice scsi12 : SCSI emulation for USB Mass Storage devices usb-storage: device found at 22 usb-storage: waiting for device to settle before scanning Vendor: HDT72252 Model: 5DLAT80 Rev: V44O Type: Direct-Access ANSI SCSI revision: 00 SCSI device sda: 488397168 512-byte hdwr sectors (250059 MB) usb 1-3: reset high speed USB device using ehci_hcd and address 22 usb 1-3: failed to restore interface 0 altsetting 0 (error=-110) usb 1-3: USB disconnect, address 22 sda: Write Protect is off sda: Mode Sense: 03 00 00 00 sda: assuming drive cache: write through sda : READ CAPACITY failed. sda : status=0, message=00, host=1, driver=00 sda : sense not available. sda: Write Protect is off sda: Mode Sense: 00 00 00 00 sda: assuming drive cache: write through .. start over again The drive produced a similar problem before which I fixed as descrived in bug #135362. -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug, or are watching someone who is.
https://bugzilla.novell.com/show_bug.cgi?id=233998 gregkh@novell.com changed: What |Removed |Added ---------------------------------------------------------------------------- Status|NEW |NEEDINFO Info Provider| |eich@novell.com ------- Comment #1 from gregkh@novell.com 2007-01-15 12:20 MST ------- Your device is disconnecting itself from the USB bus. Perhaps you aren't providing enough power to it (is it bus powered, or external powered?) Could the cable be loose / bad? Can you try another cable? Does this device, on this machine, with this cable work on other versions of Linux or other operating systems? Does this device work on other machines properly? -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug, or are watching someone who is.
https://bugzilla.novell.com/show_bug.cgi?id=233998 eich@novell.com changed: What |Removed |Added ---------------------------------------------------------------------------- Status|NEEDINFO |NEW Info Provider|eich@novell.com | ------- Comment #2 from eich@novell.com 2007-01-15 13:49 MST ------- The device is externally powered. I've checked, it won't do anything if its power supply isn't connected. Since I use it on my laptop I even stuck an externally powered USB hub in between - still same results. On the same laptop the device worked well with 10.1 (although it sometimes needed a few reset cycles). On my desktop it works well with a 10.1. With a 10.2. (still Beta2 or 2.6.18.2-4) it hogs the entire USB subsystem. I was planning to dig thru the kernel tomorrow. Hints would be appreciated ;-) -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug, or are watching someone who is.
https://bugzilla.novell.com/show_bug.cgi?id=233998 ------- Comment #3 from eich@novell.com 2007-01-18 16:28 MST ------- Created an attachment (id=113779) --> (https://bugzilla.novell.com/attachment.cgi?id=113779&action=view) usbmon dump from non-working system This usbmon dump was done with the 2.6.18.2-34-default kernel. The detection routine loops forever as a reset happens after usb_stor_clear_halt() gets a result code of -71. -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug, or are watching someone who is.
https://bugzilla.novell.com/show_bug.cgi?id=233998 ------- Comment #4 from eich@novell.com 2007-01-18 16:33 MST ------- I've attached the wrong file in above attachment. This was the usbmon dump of the *working* system ie. running kernel 2.6.16.13-4-default. -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug, or are watching someone who is.
https://bugzilla.novell.com/show_bug.cgi?id=233998 eich@novell.com changed: What |Removed |Added ---------------------------------------------------------------------------- Attachment #113779|0 |1 is obsolete| | ------- Comment #5 from eich@novell.com 2007-01-18 16:36 MST ------- Created an attachment (id=113781) --> (https://bugzilla.novell.com/attachment.cgi?id=113781&action=view) usbmon dump of *non* woring system This dump was done on kernel 2.6.18.2-34-default. The detection routine loops forever as a reset happens after usb_stor_clear_halt() gets a result code of -71. -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug, or are watching someone who is.
https://bugzilla.novell.com/show_bug.cgi?id=233998 ------- Comment #6 from eich@novell.com 2007-01-18 16:41 MST ------- Created an attachment (id=113784) --> (https://bugzilla.novell.com/attachment.cgi?id=113784&action=view) usbmon dump of device initialization on a working system This dump was done on kernel 2.6.16.13-4-default. Here again the first attempt to init the drive fails however the second one succeeds. The first attempt again shows the result code of -71 in usb_stor_clear_halt(). The second one is successful, though. I've compared the communication sequence of the working and non-working init attempts - I don't see any striking difference here. -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug, or are watching someone who is.
https://bugzilla.novell.com/show_bug.cgi?id=233998 ------- Comment #7 from eich@novell.com 2007-01-19 06:07 MST ------- Created an attachment (id=113885) --> (https://bugzilla.novell.com/attachment.cgi?id=113885&action=view) Fix By comparing the 10.1 and 10.2 driver I was finally able to locate the problem. A very simple one line change introduced this problem on my hardware. Reducing the wait delay after a device scan may work on most hardware however not on all. I'm aware that this problem can be worked around - howerver introducing a regression for some broken piece of hardware and letting the user fix it thru some obscure option doesn't seem to be the thing to do as the user doesn't necessarily associate this change nor this option to his problems. If this is a problem that only occurs on my particular broken piece of hardware a blacklist entry should probably be created. -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug, or are watching someone who is.
https://bugzilla.novell.com/show_bug.cgi?id=233998 gregkh@novell.com changed: What |Removed |Added ---------------------------------------------------------------------------- Status|NEW |RESOLVED Resolution| |FIXED ------- Comment #8 from gregkh@novell.com 2007-01-19 14:51 MST ------- We changed the default from 5 to 1 due to the "slowness" for all other devices in the world that do follow the specs :) We were having lots of bug reports that people were not seeing their device show up quick enough (compared to other operating systems.) If you need to have this value changed, just add it to your modules.conf file, to change it there, no need to have a kernel patch for this. And as this is a simple fix, and can be done in a configuration file, I'll mark this closed. -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug, or are watching someone who is.
https://bugzilla.novell.com/show_bug.cgi?id=233998 ------- Comment #9 from eich@novell.com 2007-01-22 07:17 MST ------- I should try this device under another OS. Anyway, I'm not ahppy about this solution. I may be able to debug this - others will just see their 'broken' devices being turned into a paperweights. Adding an option may be easy - if you know about it - which does not seem to be all that obvious. Can't we add this to the blacklist flags - provided the device ID is known when the scanning delay happens? -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug, or are watching someone who is.
https://bugzilla.novell.com/show_bug.cgi?id=233998 eich@novell.com changed: What |Removed |Added ---------------------------------------------------------------------------- Attachment #113885|0 |1 is obsolete| | ------- Comment #10 from eich@novell.com 2007-01-22 22:32 MST ------- Created an attachment (id=114321) --> (https://bugzilla.novell.com/attachment.cgi?id=114321&action=view) Fix: add a blacklist flag and flag broken device Instead of letting each user wonder how to get his broken device working I propose to flag known to be broken devices. -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug, or are watching someone who is.
https://bugzilla.novell.com/show_bug.cgi?id=233998 gregkh@novell.com changed: What |Removed |Added ---------------------------------------------------------------------------- Status|REOPENED |ASSIGNED ------- Comment #12 from gregkh@novell.com 2007-01-23 17:50 MST ------- I like it, I'll add it to my queue. -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug, or are watching someone who is.
https://bugzilla.novell.com/show_bug.cgi?id=233998 oneukum@novell.com changed: What |Removed |Added ---------------------------------------------------------------------------- CC| |Ivory@SerNet.de ------- Comment #13 from oneukum@novell.com 2007-02-12 07:36 MST ------- *** Bug 240451 has been marked as a duplicate of this bug. *** -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug, or are watching someone who is.
https://bugzilla.novell.com/show_bug.cgi?id=233998 Ivory@SerNet.de changed: What |Removed |Added ---------------------------------------------------------------------------- Version|RC 5 |Final ------- Comment #14 from Ivory@SerNet.de 2007-02-12 08:06 MST ------- Since my bug has just been marked as duplicate to this one and you may be collecting device information for the blacklist, I'd like to emphasize this also happens with an external Plextor USB DVD Rewriter PX-608CU (more HW info in the attachments of Bug 240451). Since not only RC5 but all final releases of openSUSE 10.2 are affected, I am bumping up the "found in version". -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug, or are watching someone who is.
https://bugzilla.novell.com/show_bug.cgi?id=233998 sndirsch@novell.com changed: What |Removed |Added ---------------------------------------------------------------------------- CC| |mhopf@novell.com, sndirsch@novell.com ------- Comment #15 from sndirsch@novell.com 2007-05-12 04:42 MST ------- JFYI, Matthias. This is a bugreport, which is assigned to Egbert/me or with Egbert/me in CC or reported by Egbert/me. -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug, or are watching someone who is.
https://bugzilla.novell.com/show_bug.cgi?id=233998 sndirsch@novell.com changed: What |Removed |Added ---------------------------------------------------------------------------- CC|sndirsch@novell.com | Severity|Normal |Critical -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug, or are watching someone who is.
https://bugzilla.novell.com/show_bug.cgi?id=233998 ------- Comment #16 from eich@novell.com 2007-05-14 05:31 MST ------- Just a few more datapoints: Meanwile the drive in my device passed away unpeacefully. To salvage the electronics I pried the plastic case open and replaced the faulty HITACHI drive with a Seagate. With this new drive the delay problem went away. The device now works well even with an unpatched kernel for 10.2. This seems to indicate that the powerup delay depends very much on the drive used. Furthermore I tested the faulty drive with a different USB controller befor it failed completely. With the other controller I didn't need an extended delay either. Difference between the two controllers: the original controller only spins up the drive when the USB device is initialized while the other controller had the drive spinning as soon as it got power. This the delay seems to be needed for USB devices that only spin up the drive when initialized if a drive is used that's slightly slow on spinup. -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug, or are watching someone who is.
https://bugzilla.novell.com/show_bug.cgi?id=233998 gregkh@novell.com changed: What |Removed |Added ---------------------------------------------------------------------------- Status|ASSIGNED |RESOLVED Resolution| |INVALID ------- Comment #17 from gregkh@novell.com 2007-05-21 10:41 MST ------- As your hardware is now dead, and I have not seen any other reports of this kind of issue in bugzilla or on the mailing lists, I'll mark this up to bad hardware and we don't need to add a kernel work-around for it at this time. -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug, or are watching someone who is.
https://bugzilla.novell.com/show_bug.cgi?id=233998 Ivory@SerNet.de changed: What |Removed |Added ---------------------------------------------------------------------------- Status|RESOLVED |REOPENED Resolution|INVALID | ------- Comment #18 from Ivory@SerNet.de 2007-05-22 02:11 MST ------- As I mentioned in comment 14, my bug 240451 was marked and closed as a duplicate of this one and I have different hardware. Hardware info is attached to that bug. -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug, or are watching someone who is.
https://bugzilla.novell.com/show_bug.cgi?id=233998#c19
Greg Kroah-Hartman
participants (1)
-
bugzilla_noreply＠novell.com