Thread: [Hp5400backend-devel] Trouble with driver: Segmentation fault
Status: Alpha
Brought to you by:
soumarmt
From: Lilith <luc...@nt...> - 2003-06-02 12:30:16
|
Hi, Thank you for for putting the work in on this driver. I've tried to list everything I think you'll need to help resolve this: I've complied sane-backends-1.0.12 on RH8, but my hp5400 scanner is not being recognised by scanimage or xsane: scanimage -L No scanners were identified. If you were expecting something different, check that the scanner is plugged in, turned on and detected by the sane-find-scanner tool (if appropriate). Please read the documentation which came with this software (README, FAQ, manpages). sane-find-scanner -q found USB scanner (vendor=0x03f0, product=0x1005) at /dev/usb/scanner0 found USB scanner (vendor=0x03f0 [Hewlett-Packard], product=0x1005 [HP Scanjet 5400C Series]) at libusb:002:002 dll.conf contains hp5400 hp5400.conf # HP 5400C usb 0x03F0 0x1005 # # Device filename to use for scanner access /dev/usb/scanner0 (I have tried commenting out 'usb 0x03F0 0x1005' and changing it to 'usb 0x3F0 0x1005) dmesg stuff hub.c: new USB device 00:1f.4-1, assigned address 2 usb.c: USB device 2 (vend/prod 0x3f0/0x1005) is not claimed by any active driver. Adding Swap: 3068404k swap-space (priority -1) usb.c: registered new driver usbscanner scanner.c: USB scanner device (0x03f0/0x1005) now attached to scanner0 scanner.c: 0.4.12:USB Scanner Driver scanner.c: probe_scanner: User specified USB scanner -- Vendor:Product - 3f0:1005 SANE_DEBUG_HP5400=255 scanimage -d hp5400 [sanei_debug] Setting debug level of hp5400 to 255. [hp5400] sane_init: SANE hp5400 backend version 1.0-2 (from sane-backends 1.0.12) [hp5400] Reading config file [hp5400] Trying to attach /dev/usb/scanner0 [hp5400] vendor/product 0x03F0-0x1005 opened [hp5400] Read: reqtype = 0xC0, req = 0x04, value = 1200 [hp5400] Read: reqtype = 0xC0, req = 0x04, value = C500 [hp5400] Command 12 verified [hp5400] Sorry, unknown scanner version. Attempted match on 'SilitekIBlizd C3 ScannerV0.84' and 'SilitekIBlizd C3 ScannerV0.86' [hp5400] Vesion is '' [hp5400] attach_one_device: couldn't attach /dev/usb/scanner0 [hp5400] Discarding line 15 [hp5400] sane_open: Segmentation fault modules.conf #added for HP5400 scanner alias /dev/usb/scanner0 scanner options scanner vendor=0x03f0 product=0x1005 (I've also tried vendor as 0x3f0) I can see that Thomas Soumarmon's solution here (https://sourceforge.net/mailarchive/message.php?msg_id=4094726) fits my problem: > You have to modify the C code in hp5400.c to be able to run the backend. > The scanner is queried about a version string which is typically : > 'SilitekIBlizd C3 ScannerV0.84' or 'SilitekIBlizd C3 ScannerV0.86' on the > scanner we tested. > You may disable this test, changing > > if( memcmp( szVersion+1, MatchVersion, sizeof(MatchVersion)-1 ) ) > { > if( memcmp( szVersion+1, MatchVersion2, sizeof(MatchVersion2)-1 ) ) > { > DBG(DBG_MSG, "Sorry, unknown scanner version. Attempted match on '%s' > and '%s'\n", MatchVersion, MatchVersion2 ); > DBG(DBG_MSG, "Vesion is '%s'", szVersion ); > ret = -1; > goto hp5400_close_exit; > } > } But I couldn't find the above in hp5400.c. I found it in hp5400_internal.c in this form: #ifndef NO_STRING_VERSION_MATCH if (memcmp (szVersion + 1, MatchVersion, sizeof (MatchVersion) - 1)) { if (memcmp (szVersion + 1, MatchVersion2, sizeof (MatchVersion2) - 1)) { DBG (DBG_MSG, "Sorry, unknown scanner version. Attempted match on '%s' and '%s'\n", MatchVersion, MatchVersion2); DBG (DBG_MSG, "Vesion is '%s'\n", szVersion); ret = -1; goto hp5400_close_exit; } } #else DBG (DBG_MSG, "Warning, Version match is disabled. Version is '%s'\n", szVersion); #endif /* NO_STRING_VERSION_MATCH */ Is that correct? Do I still need to do this: > into : > > #ifdef 0 > if( memcmp( szVersion+1, MatchVersion, sizeof(MatchVersion)-1 ) ) > { > if( memcmp( szVersion+1, MatchVersion2, sizeof(MatchVersion2)-1 ) ) > { > DBG(DBG_MSG, "Sorry, unknown scanner version. Attempted match on '%s' > and '%s'\n", MatchVersion, MatchVersion2 ); > DBG(DBG_MSG, "Vesion is '%s'", szVersion ); > ret = -1; > goto hp5400_close_exit; > } > } > #endif Also, do I need to alter this as well: #ifndef NO_STRING_VERSION_MATCH /* Match on everything except the version number */ if (memcmp (szVersion + 1, MatchVersion, sizeof (MatchVersion) - 4)) { if (memcmp (szVersion + 1, MatchVersion2, sizeof (MatchVersion2) - 4)) { DBG (DBG_MSG, "Sorry, unknown scanner version. Attempted match on '%s' and '%s'\n", MatchVersion, MatchVersion2); DBG (DBG_MSG, "Vesion is '%s'\n", szVersion); goto hp5400_close_exit; } } #else DBG (DBG_MSG, "Warning, Version match is disabled. Version is '%s'\n", szVersion); #endif /* NO_STRING_VERSION_MATCH */ Any ideas? Is there any other info you need from me? Thanks for your time. |
From: Lilith <luc...@nt...> - 2003-06-02 12:30:46
|
Hi, Thank you for for putting the work in on this driver. I've tried to list everything I think you'll need to help resolve this: I've complied sane-backends-1.0.12 on RH8, but my hp5400 scanner is not being recognised by scanimage or xsane: scanimage -L No scanners were identified. If you were expecting something different, check that the scanner is plugged in, turned on and detected by the sane-find-scanner tool (if appropriate). Please read the documentation which came with this software (README, FAQ, manpages). sane-find-scanner -q found USB scanner (vendor=0x03f0, product=0x1005) at /dev/usb/scanner0 found USB scanner (vendor=0x03f0 [Hewlett-Packard], product=0x1005 [HP Scanjet 5400C Series]) at libusb:002:002 dll.conf contains hp5400 hp5400.conf # HP 5400C usb 0x03F0 0x1005 # # Device filename to use for scanner access /dev/usb/scanner0 (I have tried commenting out 'usb 0x03F0 0x1005' and changing it to 'usb 0x3F0 0x1005) dmesg stuff hub.c: new USB device 00:1f.4-1, assigned address 2 usb.c: USB device 2 (vend/prod 0x3f0/0x1005) is not claimed by any active driver. Adding Swap: 3068404k swap-space (priority -1) usb.c: registered new driver usbscanner scanner.c: USB scanner device (0x03f0/0x1005) now attached to scanner0 scanner.c: 0.4.12:USB Scanner Driver scanner.c: probe_scanner: User specified USB scanner -- Vendor:Product - 3f0:1005 SANE_DEBUG_HP5400=255 scanimage -d hp5400 [sanei_debug] Setting debug level of hp5400 to 255. [hp5400] sane_init: SANE hp5400 backend version 1.0-2 (from sane-backends 1.0.12) [hp5400] Reading config file [hp5400] Trying to attach /dev/usb/scanner0 [hp5400] vendor/product 0x03F0-0x1005 opened [hp5400] Read: reqtype = 0xC0, req = 0x04, value = 1200 [hp5400] Read: reqtype = 0xC0, req = 0x04, value = C500 [hp5400] Command 12 verified [hp5400] Sorry, unknown scanner version. Attempted match on 'SilitekIBlizd C3 ScannerV0.84' and 'SilitekIBlizd C3 ScannerV0.86' [hp5400] Vesion is '' [hp5400] attach_one_device: couldn't attach /dev/usb/scanner0 [hp5400] Discarding line 15 [hp5400] sane_open: Segmentation fault modules.conf #added for HP5400 scanner alias /dev/usb/scanner0 scanner options scanner vendor=0x03f0 product=0x1005 (I've also tried vendor as 0x3f0) I can see that Thomas Soumarmon's solution here (https://sourceforge.net/mailarchive/message.php?msg_id=4094726) fits my problem: > You have to modify the C code in hp5400.c to be able to run the backend. > The scanner is queried about a version string which is typically : > 'SilitekIBlizd C3 ScannerV0.84' or 'SilitekIBlizd C3 ScannerV0.86' on the > scanner we tested. > You may disable this test, changing > > if( memcmp( szVersion+1, MatchVersion, sizeof(MatchVersion)-1 ) ) > { > if( memcmp( szVersion+1, MatchVersion2, sizeof(MatchVersion2)-1 ) ) > { > DBG(DBG_MSG, "Sorry, unknown scanner version. Attempted match on '%s' > and '%s'\n", MatchVersion, MatchVersion2 ); > DBG(DBG_MSG, "Vesion is '%s'", szVersion ); > ret = -1; > goto hp5400_close_exit; > } > } But I couldn't find the above in hp5400.c. I found it in hp5400_internal.c in this form: #ifndef NO_STRING_VERSION_MATCH if (memcmp (szVersion + 1, MatchVersion, sizeof (MatchVersion) - 1)) { if (memcmp (szVersion + 1, MatchVersion2, sizeof (MatchVersion2) - 1)) { DBG (DBG_MSG, "Sorry, unknown scanner version. Attempted match on '%s' and '%s'\n", MatchVersion, MatchVersion2); DBG (DBG_MSG, "Vesion is '%s'\n", szVersion); ret = -1; goto hp5400_close_exit; } } #else DBG (DBG_MSG, "Warning, Version match is disabled. Version is '%s'\n", szVersion); #endif /* NO_STRING_VERSION_MATCH */ Is that correct? Do I still need to do this: > into : > > #ifdef 0 > if( memcmp( szVersion+1, MatchVersion, sizeof(MatchVersion)-1 ) ) > { > if( memcmp( szVersion+1, MatchVersion2, sizeof(MatchVersion2)-1 ) ) > { > DBG(DBG_MSG, "Sorry, unknown scanner version. Attempted match on '%s' > and '%s'\n", MatchVersion, MatchVersion2 ); > DBG(DBG_MSG, "Vesion is '%s'", szVersion ); > ret = -1; > goto hp5400_close_exit; > } > } > #endif Also, do I need to alter this as well: #ifndef NO_STRING_VERSION_MATCH /* Match on everything except the version number */ if (memcmp (szVersion + 1, MatchVersion, sizeof (MatchVersion) - 4)) { if (memcmp (szVersion + 1, MatchVersion2, sizeof (MatchVersion2) - 4)) { DBG (DBG_MSG, "Sorry, unknown scanner version. Attempted match on '%s' and '%s'\n", MatchVersion, MatchVersion2); DBG (DBG_MSG, "Vesion is '%s'\n", szVersion); goto hp5400_close_exit; } } #else DBG (DBG_MSG, "Warning, Version match is disabled. Version is '%s'\n", szVersion); #endif /* NO_STRING_VERSION_MATCH */ Any ideas? Is there any other info you need from me? Thanks for your time. |
From: Thomas S. <tho...@co...> - 2003-06-02 15:08:24
|
Hi,=20 you may try adding the following line at the begiinning of the=20 hp5400_internal.c file : #define NO_STRING_VERSION_MATCH A cleaner thing would be to have the -DNO_STRING_VERSION_MATCH added to the= =20 compilation line, but I don't know how to do it with the SANE configure=20 program. Explanation : In the C compiler command (something like "cc OPTION= S=20 file.c"), the -DNO_STRING_VERSION_MATCH would define NO_STRING_VERSION_MATC= H=20 in the compilng process and then avoid compiling the version string match=20 process. The segmentation fault could be caused by following actions : * scanner is recognized by its USB identification * scanner is disconnected because it has not a known version string * scanner is accessed after disconnection =3D> seg fault It is just a guess. Tell us if it works after that modification. Thank you for your report, Thomas. Le Lundi 2 Juin 2003 14:30, Lilith a =E9crit : > Hi, > > Thank you for for putting the work in on this driver. I've tried to list > everything I think you'll need to help resolve this: > > I've complied sane-backends-1.0.12 on RH8, but my hp5400 scanner is not > being recognised by scanimage or xsane: > > scanimage -L > No scanners were identified. If you were expecting something different, > check that the scanner is plugged in, turned on and detected by the > sane-find-scanner tool (if appropriate). Please read the documentation > which came with this software (README, FAQ, manpages). > > sane-find-scanner -q > found USB scanner (vendor=3D0x03f0, product=3D0x1005) at /dev/usb/scanner0 > found USB scanner (vendor=3D0x03f0 [Hewlett-Packard], product=3D0x1005 [HP > Scanjet 5400C Series]) at libusb:002:002 > > dll.conf contains hp5400 > > hp5400.conf > # HP 5400C > usb 0x03F0 0x1005 > # > # Device filename to use for scanner access > /dev/usb/scanner0 > (I have tried commenting out 'usb 0x03F0 0x1005' and changing it to 'usb > 0x3F0 0x1005) > > dmesg stuff > hub.c: new USB device 00:1f.4-1, assigned address 2 > usb.c: USB device 2 (vend/prod 0x3f0/0x1005) is not claimed by any > active driver. > Adding Swap: 3068404k swap-space (priority -1) > usb.c: registered new driver usbscanner > scanner.c: USB scanner device (0x03f0/0x1005) now attached to scanner0 > scanner.c: 0.4.12:USB Scanner Driver > scanner.c: probe_scanner: User specified USB scanner -- Vendor:Product - > 3f0:1005 > > SANE_DEBUG_HP5400=3D255 scanimage -d hp5400 > [sanei_debug] Setting debug level of hp5400 to 255. > [hp5400] sane_init: SANE hp5400 backend version 1.0-2 (from > sane-backends 1.0.12) > [hp5400] Reading config file > [hp5400] Trying to attach /dev/usb/scanner0 > [hp5400] vendor/product 0x03F0-0x1005 opened > [hp5400] Read: reqtype =3D 0xC0, req =3D 0x04, value =3D 1200 > [hp5400] Read: reqtype =3D 0xC0, req =3D 0x04, value =3D C500 > [hp5400] Command 12 verified > [hp5400] Sorry, unknown scanner version. Attempted match on > 'SilitekIBlizd C3 ScannerV0.84' and 'SilitekIBlizd C3 ScannerV0.86' > [hp5400] Vesion is '' > [hp5400] attach_one_device: couldn't attach /dev/usb/scanner0 > [hp5400] Discarding line 15 > [hp5400] sane_open: > Segmentation fault > > modules.conf > #added for HP5400 scanner > alias /dev/usb/scanner0 scanner > options scanner vendor=3D0x03f0 product=3D0x1005 > (I've also tried vendor as 0x3f0) > > > I can see that Thomas Soumarmon's solution here > (https://sourceforge.net/mailarchive/message.php?msg_id=3D4094726) fits my > > problem: > > You have to modify the C code in hp5400.c to be able to run the > > backend. The scanner is queried about a version string which is > > typically : 'SilitekIBlizd C3 ScannerV0.84' or 'SilitekIBlizd C3 > > ScannerV0.86' > > on the > > > scanner we tested. > > You may disable this test, changing > > > > if( memcmp( szVersion+1, MatchVersion, sizeof(MatchVersion)-1 ) ) > > { > > if( memcmp( szVersion+1, MatchVersion2, sizeof(MatchVersion2)-1 )= ) > > { > > DBG(DBG_MSG, "Sorry, unknown scanner version. Attempted match > > on '%s' > > > and '%s'\n", MatchVersion, MatchVersion2 ); > > DBG(DBG_MSG, "Vesion is '%s'", szVersion ); > > ret =3D -1; > > goto hp5400_close_exit; > > } > > } > > But I couldn't find the above in hp5400.c. I found it in > hp5400_internal.c in this form: > > #ifndef NO_STRING_VERSION_MATCH > if (memcmp (szVersion + 1, MatchVersion, sizeof (MatchVersion) - 1)) > { > if (memcmp (szVersion + 1, MatchVersion2, sizeof (MatchVersion2) - > 1)) { > DBG (DBG_MSG, > "Sorry, unknown scanner version. Attempted match on '%s' and > '%s'\n", > MatchVersion, MatchVersion2); > DBG (DBG_MSG, "Vesion is '%s'\n", szVersion); > ret =3D -1; > goto hp5400_close_exit; > } > } > #else > DBG (DBG_MSG, "Warning, Version match is disabled. Version is '%s'\n", > szVersion); > #endif /* NO_STRING_VERSION_MATCH */ > > Is that correct? Do I still need to do this: > > into : > > > > #ifdef 0 > > if( memcmp( szVersion+1, MatchVersion, sizeof(MatchVersion)-1 ) ) > > { > > if( memcmp( szVersion+1, MatchVersion2, sizeof(MatchVersion2)-1 )= ) > > { > > DBG(DBG_MSG, "Sorry, unknown scanner version. Attempted match > > on '%s' > > > and '%s'\n", MatchVersion, MatchVersion2 ); > > DBG(DBG_MSG, "Vesion is '%s'", szVersion ); > > ret =3D -1; > > goto hp5400_close_exit; > > } > > } > > #endif > > Also, do I need to alter this as well: > > #ifndef NO_STRING_VERSION_MATCH > /* Match on everything except the version number */ > if (memcmp (szVersion + 1, MatchVersion, sizeof (MatchVersion) - 4)) > { > if (memcmp (szVersion + 1, MatchVersion2, sizeof (MatchVersion2) - > 4)) { > DBG (DBG_MSG, > "Sorry, unknown scanner version. Attempted match on '%s' and > '%s'\n", > MatchVersion, MatchVersion2); > DBG (DBG_MSG, "Vesion is '%s'\n", szVersion); > goto hp5400_close_exit; > } > } > #else > DBG (DBG_MSG, "Warning, Version match is disabled. Version is '%s'\n", > szVersion); > #endif /* NO_STRING_VERSION_MATCH */ > > > Any ideas? Is there any other info you need from me? > > Thanks for your time. > > > > ------------------------------------------------------- > This SF.net email is sponsored by: eBay > Get office equipment for less on eBay! > http://adfarm.mediaplex.com/ad/ck/711-11697-6916-5 > _______________________________________________ > Hp5400backend-devel mailing list > Hp5...@li... > https://lists.sourceforge.net/lists/listinfo/hp5400backend-devel |
From: Lilith <luc...@nt...> - 2003-06-02 23:44:07
|
I untarred a clean version of sane-backends-1.0.12 and then added '#define NO_STRING_VERSION_MATCH' to the beginning of hp5400_internal.c scanimage now recognises scanner: scanimage -L device `hp5400:/dev/usb/scanner0' is a Hewlett-Packard HP54xx Flatbed Scanner flatbed scanner device `hp5400:/dev/usb/scanner0' is a Hewlett-Packard HP54xx Flatbed Scanner flatbed scanner Tried scanning black and white document, resulting image has a blue tint, but is readable. Tried to run Xsane 0.84 (version packaged with RH8). Says no device found. Tried 'xsane hp5400:/dev/usb/scanner0', error message is 'Failure to open device 'hp5400:/dev/usb/scanner0'. Operation not supported.' Downloaded and compiled xsane-0.91.tar.gz <ftp://ftp.mostang.com/pub/sane/xsane/xsane-0.91.tar.gz> (available here: ftp://ftp.mostang.com/pub/sane/xsane ) Xsane now works, but lists the scanner twice under 'available device', just like it does with scanimage -L. (Will that cause any problems?) I'm very happy to have a working scanner now. Thanks for your help. Lilith Thomas Soumarmon wrote: >Hi, > >you may try adding the following line at the begiinning of the >hp5400_internal.c file : > >#define NO_STRING_VERSION_MATCH > >A cleaner thing would be to have the -DNO_STRING_VERSION_MATCH added to the >compilation line, but I don't know how to do it with the SANE configure >program. Explanation : In the C compiler command (something like "cc OPTIONS >file.c"), the -DNO_STRING_VERSION_MATCH would define NO_STRING_VERSION_MATCH >in the compilng process and then avoid compiling the version string match >process. > > >The segmentation fault could be caused by following actions : >* scanner is recognized by its USB identification >* scanner is disconnected because it has not a known version string >* scanner is accessed after disconnection => seg fault >It is just a guess. > >Tell us if it works after that modification. > >Thank you for your report, > >Thomas. > > |
From: Thomas S. <tho...@co...> - 2003-06-03 06:32:13
|
The problem of having twice the scanner as available could be due to a doub= le=20 declaration in hp5400.conf or a=20 Check that the hp5400.conf does not declare twice the scanner or that there= is=20 only one version of hp5400.conf on your computer. I never saw this problem and I can't think of any problem that could be cau= sed=20 by that. The problem of xsane 0.84 is probably that it could not talk with=20 sane-backends-1.0.12 but with an older version. Thomas. Le Mardi 3 Juin 2003 01:34, Lilith a =E9crit : > I untarred a clean version of sane-backends-1.0.12 and then added '#define > NO_STRING_VERSION_MATCH' to the beginning of hp5400_internal.c > > scanimage now recognises scanner: > > scanimage -L > device `hp5400:/dev/usb/scanner0' is a Hewlett-Packard HP54xx Flatbed > Scanner flatbed scanner device `hp5400:/dev/usb/scanner0' is a > Hewlett-Packard HP54xx Flatbed Scanner flatbed scanner > > Tried scanning black and white document, resulting image has a blue tint, > but is readable. > > Tried to run Xsane 0.84 (version packaged with RH8). Says no device found. > Tried 'xsane hp5400:/dev/usb/scanner0', error message is 'Failure to open > device 'hp5400:/dev/usb/scanner0'. Operation not supported.' > > Downloaded and compiled xsane-0.91.tar.gz > <ftp://ftp.mostang.com/pub/sane/xsane/xsane-0.91.tar.gz> (available here: > ftp://ftp.mostang.com/pub/sane/xsane ) > Xsane now works, but lists the scanner twice under 'available device', ju= st > like it does with scanimage -L. (Will that cause any problems?) > > I'm very happy to have a working scanner now. Thanks for your help. > > Lilith > <snip> |
From: Lilith <luc...@nt...> - 2003-06-03 11:03:34
|
Thomas Soumarmon wrote: >The problem of having twice the scanner as available could be due to a double >declaration in hp5400.conf or a >Check that the hp5400.conf does not declare twice the scanner or that there is >only one version of hp5400.conf on your computer. >I never saw this problem and I can't think of any problem that could be caused >by that. > I commented out everything in hp5400.conf and I also commented out in /etc/modules.conf : alias /dev/usb/scanner0 scanner options scanner vendor=0x3f0 product=0x1005 Scan still works after those modifications, but it's still being listed twice by scanimage and xsane. sane-find-scanner -q found USB scanner (vendor=0x03f0, product=0x1005) at /dev/usb/scanner0 found USB scanner (vendor=0x03f0, product=0x1005) at libusb:001:002 It seems as it's reading '/dev/usb/scanner0' and 'libusb:001:002' as two separate addresses. |
From: Thomas S. <tho...@co...> - 2003-06-03 11:49:00
|
<snip> > > I commented out everything in hp5400.conf and I also commented out in > /etc/modules.conf : > alias /dev/usb/scanner0 scanner > options scanner vendor=0x3f0 product=0x1005 removing the info from modules.conf is not taken into account before you reinit the modules. You may run rmmod scanner to unload the scanner module and have /dev/usb/scanner0 removed > > Scan still works after those modifications, but it's still being listed > twice by scanimage and xsane. > > sane-find-scanner -q > found USB scanner (vendor=0x03f0, product=0x1005) at /dev/usb/scanner0 > found USB scanner (vendor=0x03f0, product=0x1005) at libusb:001:002 It looks that the hp5400.conf that is used is not the one you commented out as sane still finds the 2 entries you mention. > > It seems as it's reading '/dev/usb/scanner0' and 'libusb:001:002' as two > separate addresses. Actually they are different because it is 2 different paths to the same hardware. /dev/usb/scanner0 is managed by scanner module libusb:001:002 is managed by libusb module You'd better leave one of the two lines in hp5400.conf : * if you let the scanner0, don't forget to put back the modules.conf line to have it connect at next boot. * if you let the libusb you would not need (for SANE) the scanner module. But maybe it could be used by another software, so I recommend you to put it back too. Hope it helps, Thomas. |
From: Lilith <luc...@nt...> - 2003-06-03 13:21:15
|
I deleted the hp5400.conf which was in the folder which I complied and installed from, altered hp5400.conf to: /dev/usb/scanner0 (left 'usb 0x03F0 0x1005' commented out) Reinstated the info in modules.conf and Xsane is now only detecting one scanner now and working nicely with the backend. Thank you for your help :-) Thomas Soumarmon wrote: ><snip> > > > >>I commented out everything in hp5400.conf and I also commented out in >>/etc/modules.conf : >>alias /dev/usb/scanner0 scanner >>options scanner vendor=0x3f0 product=0x1005 >> >> > >removing the info from modules.conf is not taken into account before you >reinit the modules. You may run >rmmod scanner >to unload the scanner module and have /dev/usb/scanner0 removed > > > >>Scan still works after those modifications, but it's still being listed >>twice by scanimage and xsane. >> >>sane-find-scanner -q >>found USB scanner (vendor=0x03f0, product=0x1005) at /dev/usb/scanner0 >>found USB scanner (vendor=0x03f0, product=0x1005) at libusb:001:002 >> >> > >It looks that the hp5400.conf that is used is not the one you commented out as >sane still finds the 2 entries you mention. > > > >>It seems as it's reading '/dev/usb/scanner0' and 'libusb:001:002' as two >>separate addresses. >> >> > >Actually they are different because it is 2 different paths to the same >hardware. > >/dev/usb/scanner0 is managed by scanner module >libusb:001:002 is managed by libusb module > > >You'd better leave one of the two lines in hp5400.conf : >* if you let the scanner0, don't forget to put back the modules.conf line to >have it connect at next boot. >* if you let the libusb you would not need (for SANE) the scanner module. But >maybe it could be used by another software, so I recommend you to put it back >too. > >Hope it helps, > >Thomas. > > > > > > >------------------------------------------------------- >This SF.net email is sponsored by: eBay >Get office equipment for less on eBay! >http://adfarm.mediaplex.com/ad/ck/711-11697-6916-5 >_______________________________________________ >Hp5400backend-devel mailing list >Hp5...@li... >https://lists.sourceforge.net/lists/listinfo/hp5400backend-devel > > > |