[sane-devel] CanoScan LiDE30 problem
Quoting Gerhard Jaeger gerhard at gjaeger.de: On Thursday 06 December 2007 09:18:36 Johannes Ranke wrote: Hi Gerhard and Gerard, (replying to Gerard because I didn't receive Gerhards mail since I am not on the list) Thanks for the suggestion with CONFIG_USB_SUSPEND! I just booted a self-compiled kernel without this option, and scanning works again from the graphical frontends. What I don't understand is why it works from the command line with scanimage, but not with the graphical frontends. The GUI frontends are checking for the device and the release it. This is the point, where the USB subsystem waits some time and decides to go to the suspend mode - which the scanners don't like. scanimage itself, opens the device and scans without giving the USB subsystem the chance to go to sleep... AFAIK, there's a workaround in more recent kernels, but I don't recall - maybe Julien? Another workaround came to my mind: use the scanbutton daemon, which checks each 200ms for a pressed scanner button. That way, the USB won't also fell asleep... Ciao, Gerhard -- sane-devel mailing list: sane-devel at lists.alioth.debian.org http://lists.alioth.debian.org/mailman/listinfo/sane-devel Unsubscribe: Send mail with subject unsubscribe your_password to sane-devel-request at lists.alioth.debian.org IIRC in SANE CVS there is also a patch for the suspend mode problem m.vr.gr. Gerard Klaver
[sane-devel] CanoScan LiDE30 problem
* G.J.M. KLAVER gerard at gkall.hobby.nl [071207 21:10]: Quoting Gerhard Jaeger gerhard at gjaeger.de: On Thursday 06 December 2007 09:18:36 Johannes Ranke wrote: Hi Gerhard and Gerard, (replying to Gerard because I didn't receive Gerhards mail since I am not on the list) Thanks for the suggestion with CONFIG_USB_SUSPEND! I just booted a self-compiled kernel without this option, and scanning works again from the graphical frontends. What I don't understand is why it works from the command line with scanimage, but not with the graphical frontends. The GUI frontends are checking for the device and the release it. This is the point, where the USB subsystem waits some time and decides to go to the suspend mode - which the scanners don't like. scanimage itself, opens the device and scans without giving the USB subsystem the chance to go to sleep... AFAIK, there's a workaround in more recent kernels, but I don't recall - maybe Julien? Another workaround came to my mind: use the scanbutton daemon, which checks each 200ms for a pressed scanner button. That way, the USB won't also fell asleep... Ciao, Gerhard -- sane-devel mailing list: sane-devel at lists.alioth.debian.org http://lists.alioth.debian.org/mailman/listinfo/sane-devel Unsubscribe: Send mail with subject unsubscribe your_password to sane-devel-request at lists.alioth.debian.org IIRC in SANE CVS there is also a patch for the suspend mode problem Ah, would this have worked with kernel 2.6.20 with USB_SUSPEND though? I am sure there are still people using such kernels (as I was), and when they occasionally want to scan they will not know how to work around their problem. I am a little hesitant to file bug reports against all the graphical frontends from xscanimage via quiteinsane, kooka and whoknowswhatelse in the Debian BTS. Johannes m.vr.gr. Gerard Klaver
[sane-devel] CanoScan LiDE30 problem
Hi Gerhard and Gerard, (replying to Gerard because I didn't receive Gerhards mail since I am not on the list) Thanks for the suggestion with CONFIG_USB_SUSPEND! I just booted a self-compiled kernel without this option, and scanning works again from the graphical frontends. What I don't understand is why it works from the command line with scanimage, but not with the graphical frontends. Regards, Johannes * Gerard Klaver gerard at gkall.hobby.nl [071203 18:30]: On Mon, 2007-12-03 at 11:08 +0100, Johannes Ranke wrote: Hi, My Canoscan LiDE30 used to work nicely under Debian unstable for more than a year. Since about half a year it makes problems: If I scan from a frontend like quiteinsane, I can select the device (I am using libusb), and the frontend pretends to scan, but the scanner does nothing, so I end up with a black image. The scanner works under Windows. Last week, after I tested under Windows, it worked under Linux, too. $ scanimage -L device `v4l:/dev/video0' is a Noname BT878 video (AVerMedia AVerTV D virtual device device `plustek:libusb:003:003' is a Canon CanoScan N1240U/LiDE30 flatbed scanner I discovered that I can scan from the command line as normal user: $ SANE_DEBUG_PLUSTEK=128 scanimage -d 'plustek:libusb:003:003' \ image.pnm \ 2 canoscanLiDE30.err I am attaching the output of this. Graphical frontends still don't work, even when specifying the device on the command line for xscanimage, or selecting it in quiteinsane. I also noticed the following dmesg output after trying to scan with xscanimage: usb 3-2: usbfs: interface 0 claimed by usbfs while 'xscanimage' sets config #1 I don't know which (combination of) software is reponsible for the problems, otherwise I would file a bug report in the Debian BTS. My kernel is: $ uname -a Linux stiller 2.6.20-1-amd64 #1 SMP Tue Apr 24 21:10:58 UTC 2007 x86_64 GNU/Linux Sane version is: $ scanimage -V scanimage (sane-backends) 1.0.18-cvs; backend version 1.0.18 Best regards, Johannes Add a # before the v4l line in /etc/sane.d/dll.conf this will disable the v4l backend and will prevent that your video device is selected. m.vr.gr. Gerard Klaver
[sane-devel] CanoScan LiDE30 problem
On Thursday 06 December 2007 09:18:36 Johannes Ranke wrote: Hi Gerhard and Gerard, (replying to Gerard because I didn't receive Gerhards mail since I am not on the list) Thanks for the suggestion with CONFIG_USB_SUSPEND! I just booted a self-compiled kernel without this option, and scanning works again from the graphical frontends. What I don't understand is why it works from the command line with scanimage, but not with the graphical frontends. The GUI frontends are checking for the device and the release it. This is the point, where the USB subsystem waits some time and decides to go to the suspend mode - which the scanners don't like. scanimage itself, opens the device and scans without giving the USB subsystem the chance to go to sleep... AFAIK, there's a workaround in more recent kernels, but I don't recall - maybe Julien? Another workaround came to my mind: use the scanbutton daemon, which checks each 200ms for a pressed scanner button. That way, the USB won't also fell asleep... Ciao, Gerhard
[sane-devel] CanoScan LiDE30 problem
* Gerhard Jaeger gerhard at gjaeger.de [071206 15:50]: On Thursday 06 December 2007 09:18:36 Johannes Ranke wrote: Hi Gerhard and Gerard, (replying to Gerard because I didn't receive Gerhards mail since I am not on the list) Thanks for the suggestion with CONFIG_USB_SUSPEND! I just booted a self-compiled kernel without this option, and scanning works again from the graphical frontends. What I don't understand is why it works from the command line with scanimage, but not with the graphical frontends. The GUI frontends are checking for the device and the release it. This is the point, where the USB subsystem waits some time and decides to go to the suspend mode - which the scanners don't like. OK. scanimage itself, opens the device and scans without giving the USB subsystem the chance to go to sleep... AFAIK, there's a workaround in more recent kernels, but I don't recall - maybe Julien? Yes, I just read in a long Ubuntu forum thread, that in more recent kernels, the possibility exists to exclude usb suspend to be active on certain devices via udev rules. Together with a new sane version this is supposed to fix the issue. Another workaround came to my mind: use the scanbutton daemon, which checks each 200ms for a pressed scanner button. That way, the USB won't also fell asleep... Yes, I just saw that this workaround was mentioned in the forum thread to be working, too. I guess this is THE workaround if you have a working older kernel with CONFIG_USB_SUSPEND=y and don't want to upgrade. Thanks a lot for your explanations! Johannes Ciao, Gerhard -- Dr. Johannes Ranke jranke at uni-bremen.de Key ID: F649AF90 UFT Bremen, Leobenerstr. 1 +49 421 218 63373 D-28359 Bremen http://www.uft.uni-bremen.de/chemie/ranke
[sane-devel] CanoScan LiDE30 problem
* Gerard Klaver gerard at gkall.hobby.nl [071203 18:30]: On Mon, 2007-12-03 at 11:08 +0100, Johannes Ranke wrote: Hi, My Canoscan LiDE30 used to work nicely under Debian unstable for more than a year. Since about half a year it makes problems: If I scan from a frontend like quiteinsane, I can select the device (I am using libusb), and the frontend pretends to scan, but the scanner does nothing, so I end up with a black image. The scanner works under Windows. Last week, after I tested under Windows, it worked under Linux, too. $ scanimage -L device `v4l:/dev/video0' is a Noname BT878 video (AVerMedia AVerTV D virtual device device `plustek:libusb:003:003' is a Canon CanoScan N1240U/LiDE30 flatbed scanner I discovered that I can scan from the command line as normal user: $ SANE_DEBUG_PLUSTEK=128 scanimage -d 'plustek:libusb:003:003' \ image.pnm \ 2 canoscanLiDE30.err I am attaching the output of this. Graphical frontends still don't work, even when specifying the device on the command line for xscanimage, or selecting it in quiteinsane. I also noticed the following dmesg output after trying to scan with xscanimage: usb 3-2: usbfs: interface 0 claimed by usbfs while 'xscanimage' sets config #1 I don't know which (combination of) software is reponsible for the problems, otherwise I would file a bug report in the Debian BTS. My kernel is: $ uname -a Linux stiller 2.6.20-1-amd64 #1 SMP Tue Apr 24 21:10:58 UTC 2007 x86_64 GNU/Linux Sane version is: $ scanimage -V scanimage (sane-backends) 1.0.18-cvs; backend version 1.0.18 Best regards, Johannes Add a # before the v4l line in /etc/sane.d/dll.conf this will disable the v4l backend and will prevent that your video device is selected. Danke, das habe ich probiert - es wird dann zwar nur noch der USB Scanner von scanimage -L aufgef?hrt, aber quiteinsane und xscanimage sind immer noch funktionslos und sprechen den Scanner nicht an. Gru?, Johannes Ranke m.vr.gr. Gerard Klaver -- Dr. Johannes Ranke jranke at uni-bremen.de Key ID: F649AF90 UFT Bremen, Leobenerstr. 1 +49 421 218 63373 D-28359 Bremen http://www.uft.uni-bremen.de/chemie/ranke
[sane-devel] CanoScan LiDE30 problem
Hi, My Canoscan LiDE30 used to work nicely under Debian unstable for more than a year. Since about half a year it makes problems: If I scan from a frontend like quiteinsane, I can select the device (I am using libusb), and the frontend pretends to scan, but the scanner does nothing, so I end up with a black image. The scanner works under Windows. Last week, after I tested under Windows, it worked under Linux, too. $ scanimage -L device `v4l:/dev/video0' is a Noname BT878 video (AVerMedia AVerTV D virtual device device `plustek:libusb:003:003' is a Canon CanoScan N1240U/LiDE30 flatbed scanner I discovered that I can scan from the command line as normal user: $ SANE_DEBUG_PLUSTEK=128 scanimage -d 'plustek:libusb:003:003' \ image.pnm \ 2 canoscanLiDE30.err I am attaching the output of this. Graphical frontends still don't work, even when specifying the device on the command line for xscanimage, or selecting it in quiteinsane. I also noticed the following dmesg output after trying to scan with xscanimage: usb 3-2: usbfs: interface 0 claimed by usbfs while 'xscanimage' sets config #1 I don't know which (combination of) software is reponsible for the problems, otherwise I would file a bug report in the Debian BTS. My kernel is: $ uname -a Linux stiller 2.6.20-1-amd64 #1 SMP Tue Apr 24 21:10:58 UTC 2007 x86_64 GNU/Linux Sane version is: $ scanimage -V scanimage (sane-backends) 1.0.18-cvs; backend version 1.0.18 Best regards, Johannes -- Dr. Johannes Ranke jranke at uni-bremen.de Key ID: F649AF90 UFT Bremen, Leobenerstr. 1 +49 421 218 63373 D-28359 Bremen http://www.uft.uni-bremen.de/chemie/ranke -- next part -- [sanei_debug] Setting debug level of plustek to 128. [plustek] Plustek backend V0.52-3, part of sane-backends 1.0.18-cvs [plustek] Retrieving all supported and conntected devices [plustek] Checking for 0x07b3-0x0010 [plustek] Checking for 0x07b3-0x0011 [plustek] Checking for 0x07b3-0x0017 [plustek] Checking for 0x07b3-0x0015 [plustek] Checking for 0x07b3-0x0015 [plustek] Checking for 0x07b3-0x0017 [plustek] Checking for 0x07b3-0x0013 [plustek] Checking for 0x07b3-0x0013 [plustek] Checking for 0x07b3-0x0011 [plustek] Checking for 0x07b3-0x0010 [plustek] Checking for 0x07b3-0x0014 [plustek] Checking for 0x07b3-0x0014 [plustek] Checking for 0x07b3-0x0016 [plustek] Checking for 0x07b3-0x0017 [plustek] Checking for 0x07b3-0x0017 [plustek] Checking for 0x07b3-0x0007 [plustek] Checking for 0x07b3-0x000f [plustek] Checking for 0x07b3-0x000f [plustek] Checking for 0x07b3-0x0005 [plustek] Checking for 0x07b3-0x0014 [plustek] Checking for 0x07b3-0x0012 [plustek] Checking for 0x0400-0x1000 [plustek] Checking for 0x0400-0x1001 [plustek] Checking for 0x0400-0x1001 [plustek] Checking for 0x0458-0x2007 [plustek] Checking for 0x0458-0x2008 [plustek] Checking for 0x0458-0x2009 [plustek] Checking for 0x0458-0x2013 [plustek] Checking for 0x0458-0x2015 [plustek] Checking for 0x0458-0x2016 [plustek] Checking for 0x03f0-0x0505 [plustek] Checking for 0x03f0-0x0605 [plustek] Checking for 0x04b8-0x010f [plustek] Checking for 0x04b8-0x011d [plustek] Checking for 0x1606-0x0050 [plustek] Checking for 0x1606-0x0060 [plustek] Checking for 0x1606-0x0160 [plustek] Checking for 0x049f-0x001a [plustek] Checking for 0x04a9-0x2206 [plustek] Checking for 0x04a9-0x2207 [plustek] Checking for 0x04a9-0x2208 [plustek] Checking for 0x04a9-0x220d [plustek] Checking for 0x04a9-0x220e [plustek] Checking for 0x04a9-0x2220 [plustek] Checking for 0x0a82-0x6620 [plustek] Checking for 0x0a53-0x1000 [plustek] Available and supported devices: [plustek] Device: libusb:003:003 - 0x04a9x0x220e [plustek] # Plustek-SANE Backend configuration file [plustek] # For use with LM9831/2/3 based USB scanners [plustek] # [plustek] [plustek] # each device needs at least two lines: [plustek] # - [usb] vendor-ID and product-ID [plustek] # - device devicename [plustek] # i.e. for Plustek (0x07B3) UT12/16/24 (0x0017) [plustek] # [usb] 0x07B3 0x0017 [plustek] # device /dev/usbscanner [plustek] # or [plustek] # device libusb:bbb:ddd [plustek] # where bbb is the busnumber and ddd the device number [plustek] # make sure that your user has access to /proc/bus/usb/bbb/ddd [plustek] # [plustek] # additionally you can specify some options [plustek] # warmup, lOffOnEnd, lampOff [plustek] # [plustek] # For autodetection use [plustek] # [usb] [plustek] # device /dev/usbscanner [plustek] # [plustek] # or simply [plustek] # [usb] [plustek] # [plustek] # or if you want a specific device but you have no idea about the [plustek] # device node or you use libusb, simply set vendor- and product-ID [plustek] # [usb] 0x07B3 0x0017 [plustek] # device auto [plustek] # [plustek] # NOTE: autodetection is safe, as it uses the info it got [plustek] # from the USB subsystem. If you're not using the [plustek] # autodetection, you MUST have attached that device
[sane-devel] CanoScan LiDE30 problem
On Mon, 2007-12-03 at 11:08 +0100, Johannes Ranke wrote: Hi, My Canoscan LiDE30 used to work nicely under Debian unstable for more than a year. Since about half a year it makes problems: If I scan from a frontend like quiteinsane, I can select the device (I am using libusb), and the frontend pretends to scan, but the scanner does nothing, so I end up with a black image. The scanner works under Windows. Last week, after I tested under Windows, it worked under Linux, too. $ scanimage -L device `v4l:/dev/video0' is a Noname BT878 video (AVerMedia AVerTV D virtual device device `plustek:libusb:003:003' is a Canon CanoScan N1240U/LiDE30 flatbed scanner I discovered that I can scan from the command line as normal user: $ SANE_DEBUG_PLUSTEK=128 scanimage -d 'plustek:libusb:003:003' \ image.pnm \ 2 canoscanLiDE30.err I am attaching the output of this. Graphical frontends still don't work, even when specifying the device on the command line for xscanimage, or selecting it in quiteinsane. I also noticed the following dmesg output after trying to scan with xscanimage: usb 3-2: usbfs: interface 0 claimed by usbfs while 'xscanimage' sets config #1 I don't know which (combination of) software is reponsible for the problems, otherwise I would file a bug report in the Debian BTS. My kernel is: $ uname -a Linux stiller 2.6.20-1-amd64 #1 SMP Tue Apr 24 21:10:58 UTC 2007 x86_64 GNU/Linux Sane version is: $ scanimage -V scanimage (sane-backends) 1.0.18-cvs; backend version 1.0.18 Best regards, Johannes Add a # before the v4l line in /etc/sane.d/dll.conf this will disable the v4l backend and will prevent that your video device is selected. m.vr.gr. Gerard Klaver
[sane-devel] CanoScan LiDE30 problem
Am Montag, 3. Dezember 2007 11:08:15 schrieb Johannes Ranke: Hi, My Canoscan LiDE30 used to work nicely under Debian unstable for more than a year. Since about half a year it makes problems: If I scan from a frontend like quiteinsane, I can select the device (I am using libusb), and the frontend pretends to scan, but the scanner does nothing, so I end up with a black image. The scanner works under Windows. Last week, after I tested under Windows, it worked under Linux, too. This is a know issue and relatet to the suspend feature on the kernel you are using, disable the CONFIG_USB_SUSPEND kernel config option and recompile the kernel. Thought this has been fixed in more recent versions. Maybe there are some motr better ideas around - or check this list for USB_SUSPEND... HTH Gerhard
