Re: [linux-usb-devel] Netac OnlyDisk Mini incorrectly detected as write protected
Many thanks for the reply Alan! Well I finally had a chance to compile debugfs and usbmon into my 2.6.20 kernel just now. I also took out my hack of sd_read_write_protect_flag() in drivers/scsi/sd.c to reintroduce the write protect problem. I left the device in drivers/usb/storage/unusual_devs.h. And would you believe it, on rebooting and plugging the stick in, it correctly detected it as NOT write protected. So with the device added to unusual_devs.h with US_FL_IGNORE_RESIDUE, it works perfectly :) I think what happened is when I was originally testing I would remove the usb-storage module, make changes to unusual_devs.h, then modprobe the module again. SCSI sd was compiled into the kernel, and I didn't reboot the whole time I was testing. Therefore I think continually removing/reinserting the module with changes each time, combined with plugging/unplugging the stick, and not reloading the SCSI sd code, must have confused the kernel somehow, causing it to incorrectly detect it as write protected. Anyway, bottom line is it works perfectly now, but requires adding the device to unusual_devs.h I've attached the patch again that does this. Is it possible to get this added upstream? regards, Eddie On Tue, 2007-03-13 at 22:07 -0400, Alan Stern wrote: On Wed, 14 Mar 2007, Edward Chapman wrote: Hi, On vanilla kernel 2.6.20, Netac OnlyDisk Mini 512MB usb stick lets you write a file and copy it from the device until it is unmounted. Upon mounting the device again, attempts to copy the file written on last mount results in SCSI I/O errors and the cp command eventually failing (see error log at the bottom of this message). Those error messages aren't very informative. It would help to have a usbmon log. Instructions for usbmon are in the kernel source file Documentation/usb/usbmon.txt. Post a log showing what happens when you mount the drive, copy a short file (no more than 1 KB), unmount it, mount it again, and try to copy the file back. The good news is adding the device to drivers/usb/storage/unusual_devs.h with the US_FL_IGNORE_RESIDUE flag completely cures the problem :) I've attached a patch against 2.6.20 to add the device to this file. The bad news is adding the device to unusual_devs.h and using any of the flags (I tried many of them) causes the device to be incorrectly detected as write protected. I tried using US_FL_NO_WP_DETECT (does this disable write protect?) It doesn't disable write-protect -- the only way to do that is by setting the device's switch off. However it does prevent the kernel from asking the device whether it is write-protected. but to no avail. I've had to hack sd_read_write_protect_flag() in drivers/scsi/sd.c to disable write protect detection in order to use the device with it added to unusual_devs.h. Without adding the device to unusual_devs.h, write protect detection works fine. The device does have a write protect switch by the way, and yes I have it set to off. So my question is can anyone help me figure out the incorrect write protect detection problem? I can test, provide whatever info needed. I have over 40 of these devices so I'm desperate to iron out this last issue :) Post another usbmon log showing what happens with your unusual_devs entry present. Alan Stern --- linux-2.6.20/drivers/usb/storage/unusual_devs.h 2007-03-09 18:58:04.0 + +++ linux-2.6.20b/drivers/usb/storage/unusual_devs.h 2007-03-13 23:59:21.0 + @@ -1286,6 +1286,14 @@ US_SC_DEVICE, US_PR_DEVICE, NULL, US_FL_IGNORE_RESIDUE ), +/* Reported by Edward Chapman [EMAIL PROTECTED] + Netac OnlyDisk Mini U2CV2 512MB USB 2.0 Flash Drive */ +UNUSUAL_DEV( 0x0dd8, 0xd202, 0x, 0x, + Netac, + USB Flash Disk, + US_SC_DEVICE, US_PR_DEVICE, NULL, + US_FL_IGNORE_RESIDUE ), + /* Reported by Benjamin Schiller [EMAIL PROTECTED] * It is also sold by Easylite as DJ 20 */ UNUSUAL_DEV( 0x0ed1, 0x7636, 0x0103, 0x0103, - Take Surveys. Earn Cash. Influence the Future of IT Join SourceForge.net's Techsay panel and you'll get the chance to share your opinions on IT business topics through brief surveys-and earn cash http://www.techsay.com/default.php?page=join.phpp=sourceforgeCID=DEVDEV___ linux-usb-devel@lists.sourceforge.net To unsubscribe, use the last form field at: https://lists.sourceforge.net/lists/listinfo/linux-usb-devel
[linux-usb-devel] Netac OnlyDisk Mini incorrectly detected as write protected
Hi, On vanilla kernel 2.6.20, Netac OnlyDisk Mini 512MB usb stick lets you write a file and copy it from the device until it is unmounted. Upon mounting the device again, attempts to copy the file written on last mount results in SCSI I/O errors and the cp command eventually failing (see error log at the bottom of this message). The good news is adding the device to drivers/usb/storage/unusual_devs.h with the US_FL_IGNORE_RESIDUE flag completely cures the problem :) I've attached a patch against 2.6.20 to add the device to this file. The bad news is adding the device to unusual_devs.h and using any of the flags (I tried many of them) causes the device to be incorrectly detected as write protected. I tried using US_FL_NO_WP_DETECT (does this disable write protect?) but to no avail. I've had to hack sd_read_write_protect_flag() in drivers/scsi/sd.c to disable write protect detection in order to use the device with it added to unusual_devs.h. Without adding the device to unusual_devs.h, write protect detection works fine. The device does have a write protect switch by the way, and yes I have it set to off. So my question is can anyone help me figure out the incorrect write protect detection problem? I can test, provide whatever info needed. I have over 40 of these devices so I'm desperate to iron out this last issue :) many thanks, Eddie Errors WITHOUT device in unusual_devs.h: Mar 11 13:16:07 srv1 kernel: usbcore: registered new interface driver libusual Mar 11 13:16:07 srv1 kernel: Initializing USB Mass Storage driver... Mar 11 13:16:07 srv1 kernel: usbcore: registered new interface driver usb-storage Mar 11 13:16:07 srv1 kernel: USB Mass Storage support registered. Mar 11 13:16:23 srv1 kernel: usb 5-2.7: new high speed USB device using ehci_hcd and address 6 Mar 11 13:16:23 srv1 kernel: usb 5-2.7: configuration #1 chosen from 1 choice Mar 11 13:16:23 srv1 kernel: scsi2 : SCSI emulation for USB Mass Storage devices Mar 11 13:16:23 srv1 kernel: usb-storage: device found at 6 Mar 11 13:16:23 srv1 kernel: usb-storage: waiting for device to settle before scanning Mar 11 13:16:28 srv1 kernel: scsi 2:0:0:0: Direct-Access Netac OnlyDisk 1.10 PQ: 0 ANSI: 2 Mar 11 13:16:28 srv1 kernel: SCSI device sdc: 1024000 512-byte hdwr sectors (524 MB) Mar 11 13:16:28 srv1 kernel: sdc: Write Protect is off Mar 11 13:16:28 srv1 kernel: sdc: Mode Sense: 00 00 00 00 Mar 11 13:16:28 srv1 kernel: sdc: assuming drive cache: write through Mar 11 13:16:28 srv1 kernel: SCSI device sdc: 1024000 512-byte hdwr sectors (524 MB) Mar 11 13:16:28 srv1 kernel: sdc: Write Protect is off Mar 11 13:16:28 srv1 kernel: sdc: Mode Sense: 00 00 00 00 Mar 11 13:16:28 srv1 kernel: sdc: assuming drive cache: write through Mar 11 13:16:28 srv1 kernel: sdc: sdc1 Mar 11 13:16:28 srv1 kernel: sd 2:0:0:0: Attached scsi removable disk sdc Mar 11 13:16:28 srv1 kernel: sd 2:0:0:0: Attached scsi generic sg2 type 0 Mar 11 13:16:28 srv1 kernel: usb-storage: device scan complete Mar 11 13:19:00 srv1 kernel: sd 2:0:0:0: SCSI error: return code = 0x1007 Mar 11 13:19:00 srv1 kernel: end_request: I/O error, dev sdc, sector 8351 Mar 11 13:19:00 srv1 kernel: sd 2:0:0:0: SCSI error: return code = 0x1007 Mar 11 13:19:00 srv1 kernel: end_request: I/O error, dev sdc, sector 8479 Mar 11 13:19:01 srv1 kernel: sd 2:0:0:0: SCSI error: return code = 0x1007 Mar 11 13:19:01 srv1 kernel: end_request: I/O error, dev sdc, sector 8795 Mar 11 13:19:01 srv1 kernel: sd 2:0:0:0: SCSI error: return code = 0x1007 Mar 11 13:19:01 srv1 kernel: end_request: I/O error, dev sdc, sector 8995 Mar 11 13:19:01 srv1 kernel: sd 2:0:0:0: SCSI error: return code = 0x1007 Mar 11 13:19:01 srv1 kernel: end_request: I/O error, dev sdc, sector 9309 Mar 11 13:19:01 srv1 kernel: sd 2:0:0:0: SCSI error: return code = 0x1007 Mar 11 13:19:01 srv1 kernel: end_request: I/O error, dev sdc, sector 9509 Mar 11 13:19:01 srv1 kernel: sd 2:0:0:0: SCSI error: return code = 0x1007 - SNIP --- Mar 11 13:19:35 srv1 kernel: end_request: I/O error, dev sdc, sector 48061 Mar 11 13:19:35 srv1 kernel: sd 2:0:0:0: SCSI error: return code = 0x1007 Mar 11 13:19:35 srv1 kernel: end_request: I/O error, dev sdc, sector 48375 Mar 11 13:19:35 srv1 kernel: sd 2:0:0:0: SCSI error: return code = 0x1007 Mar 11 13:19:35 srv1 kernel: end_request: I/O error, dev sdc, sector 48575 Mar 11 13:19:36 srv1 kernel: sd 2:0:0:0: SCSI error: return code = 0x1007 Mar 11 13:19:36 srv1 kernel: end_request: I/O error, dev sdc, sector 48889 Mar 11 13:19:36 srv1 kernel: sd 2:0:0:0: SCSI error: return code = 0x1007 Mar 11 13:19:36 srv1 kernel: end_request: I/O error, dev sdc, sector 49731 Mar 11 13:19:36 srv1 kernel: Buffer I/O error on device sdc1, logical block 24836 Mar 11 13:19:36 srv1 kernel: Buffer I/O error on device sdc1, logical block 24837 Mar 11 13:19:36 srv1 kernel: sd 2:0:0:0: SCSI error: return code = 0x1007 Mar 11 13:19:36 srv1 kernel:
Re: [linux-usb-devel] Netac OnlyDisk Mini incorrectly detected as write protected
On Wed, 14 Mar 2007, Edward Chapman wrote: Hi, On vanilla kernel 2.6.20, Netac OnlyDisk Mini 512MB usb stick lets you write a file and copy it from the device until it is unmounted. Upon mounting the device again, attempts to copy the file written on last mount results in SCSI I/O errors and the cp command eventually failing (see error log at the bottom of this message). Those error messages aren't very informative. It would help to have a usbmon log. Instructions for usbmon are in the kernel source file Documentation/usb/usbmon.txt. Post a log showing what happens when you mount the drive, copy a short file (no more than 1 KB), unmount it, mount it again, and try to copy the file back. The good news is adding the device to drivers/usb/storage/unusual_devs.h with the US_FL_IGNORE_RESIDUE flag completely cures the problem :) I've attached a patch against 2.6.20 to add the device to this file. The bad news is adding the device to unusual_devs.h and using any of the flags (I tried many of them) causes the device to be incorrectly detected as write protected. I tried using US_FL_NO_WP_DETECT (does this disable write protect?) It doesn't disable write-protect -- the only way to do that is by setting the device's switch off. However it does prevent the kernel from asking the device whether it is write-protected. but to no avail. I've had to hack sd_read_write_protect_flag() in drivers/scsi/sd.c to disable write protect detection in order to use the device with it added to unusual_devs.h. Without adding the device to unusual_devs.h, write protect detection works fine. The device does have a write protect switch by the way, and yes I have it set to off. So my question is can anyone help me figure out the incorrect write protect detection problem? I can test, provide whatever info needed. I have over 40 of these devices so I'm desperate to iron out this last issue :) Post another usbmon log showing what happens with your unusual_devs entry present. Alan Stern - Take Surveys. Earn Cash. Influence the Future of IT Join SourceForge.net's Techsay panel and you'll get the chance to share your opinions on IT business topics through brief surveys-and earn cash http://www.techsay.com/default.php?page=join.phpp=sourceforgeCID=DEVDEV ___ linux-usb-devel@lists.sourceforge.net To unsubscribe, use the last form field at: https://lists.sourceforge.net/lists/listinfo/linux-usb-devel
