diff options
Diffstat (limited to 'doc/plustek')
| -rw-r--r-- | doc/plustek/FAQ | 385 | ||||
| -rw-r--r-- | doc/plustek/MakeModule.sh | 116 | ||||
| -rw-r--r-- | doc/plustek/Makefile.kernel24 | 251 | ||||
| -rw-r--r-- | doc/plustek/Makefile.kernel26 | 124 | ||||
| -rw-r--r-- | doc/plustek/Plustek-PARPORT-TODO.txt | 38 | ||||
| -rw-r--r-- | doc/plustek/Plustek-PARPORT.changes | 180 | ||||
| -rw-r--r-- | doc/plustek/Plustek-PARPORT.txt | 49 | ||||
| -rw-r--r-- | doc/plustek/Plustek-USB-TODO.txt | 53 | ||||
| -rw-r--r-- | doc/plustek/Plustek-USB.changes | 307 | ||||
| -rw-r--r-- | doc/plustek/Plustek-USB.txt | 457 |
10 files changed, 1960 insertions, 0 deletions
diff --git a/doc/plustek/FAQ b/doc/plustek/FAQ new file mode 100644 index 0000000..7d65f41 --- /dev/null +++ b/doc/plustek/FAQ @@ -0,0 +1,385 @@ +Plustek-Driver TROUBLESHOOTING/FAQ - 19.04.2001 +----------------------------------------------- + +The driver exists now since March 2000, this is the first trial to setup +a FAQ. This leads into the following: +First we have a part, which is more a troubleshooting section and the second +one is the FAQ. +To search through this text, search for "SYMPTOM" or "Q:". + +If you have some more FAQ entries, let me know <gerhard@gjaeger.de> + +THANKS to Jochen <jochen@puchalla-online.de>, who roamed through the +mailing-list and gathered the different questions. + + +******************************************************************************* +* TROUBLE-SHOOTING * +* * +* DRIVER TOPICS * +******************************************************************************* + + +SYMPTOM: "kernel-module version mismatch" +-------- +"When I try to "make load" I get an error" + +/sbin/modprobe pt_drv || exit 1 +/lib/modules/2.2.14-6.0.6/misc/pt_drv.o: kernel-module version mismatch +/lib/modules/2.2.14-6.0.1/misc/pt_drv.o was compiled +for kernel version 2.2.14-5.0 +while this kernel is version 2.2.14-6.0.1. + +PROBLEM: +-------- +You're using the wrong Kernel-header files. + +The path +/usr/include/linux is a link to +/usr/src/linux/include/linux and +/usr/src/linux is normally a link to your current +kernel, i.e: +/usr/src/linux-2.2.14 + +If you now upgrade your kernel and you have +forgotten to correct the link /usr/src/linux, then +every kernel-module you build outside the +kernelsource structure will include the wrong +version information. + +SOLUTION: +--------- +Set the link /usr/src/linux +to your current kernel-source tree before recompiling +kernel-modules like pt_drv, then your version mismatch +problem should be solved. + + + +SYMPTOM: "Device or resource busy" +-------- +Any attempt to "modprobe" or "make load" the driver leads +to this message. + +PROBLEM: +-------- +The driver refuses to load. During startup, the driver performs +some tests according to the parallel port and the connected +scanners. +If the parallel port is not supported or the scanner cannot be +found, the driver returns an error and the system will report +"device or resource busy" + +SOLUTION: +--------- +There's no way to provide a general solution. First of all +you should check your connections and the parallel port. +Also check power for your scanner. +If the problem still exists, enable the debug-messages of the +pt_drv module (see INSTALL). +Then check the messages after loading or send the output to +the mailing list. +Most of these problems are related to HW-problems. Before +giving up, check the scanner under Windows (not really a +good advice I know, but it helps to find HW-damage). +There was also one case were the internal plugs of the parport +connection were wrong... + + + +SYMPTOM: "Unresolved symbols" +-------- +"make load" or "modprobe pt_drv" produces this message: + +/sbin/modprobe pt_drv || exit 1 +/lib/modules/2.2.14-5.0/misc/pt_drv.o: unresolved symbol parport_unregister_device +/lib/modules/2.2.14-5.0/misc/pt_drv.o: unresolved symbol parport_enumerate +/lib/modules/2.2.14-5.0/misc/pt_drv.o: unresolved symbol parport_register_device +/lib/modules/2.2.14-5.0/misc/pt_drv.o: unresolved symbol parport_claim +/lib/modules/2.2.14-5.0/misc/pt_drv.o: unresolved symbol parport_release +/lib/modules/2.2.14-5.0/misc/pt_drv.o: insmod +/lib/modules/2.2.14-5.0/misc/pt_drv.o failed + + +PROBLEM: +-------- +The driver needs at least the parport_pc and the parport modules +to work. They MUST be loaded prior to pt_drv. If these modules +are not loaded you get this message. + +SOLUTION: +--------- +Simply load the parport module. You can add this dependency to +your /etc/modules.conf (or /etc/conf.modules) file (see INSTALL). +Then everytime you load pt_drv with modprobe, parport will be loaded +automatically. After modifying this file, call +depmod -va + + + +SYMPTOM: Scanner makes awful noise +-------- + +PROBLEM: +-------- +This can have two major causes: +- You have an ASIC96001/3 based scanner, then the sensor hits the + scanbed: !!! TURN OFF SCANNER POWER !!! + +- You have an ASIC98001 based scanner, then the motor control does + not work correctly --> hit the cancel button + +SOLUTION: +--------- +For the first case (ASIC 96001/3), there's no solution available. This +happens, when the driver can't keep track of the stepper motor. The image +you get is normally unusable. +The second case is often reported when the printer driver lp.o is +already loaded. So remove lp.o before loading pt_drv. + + + +SYMPTOM: Printer starts to print while scanning +-------- + +PROBLEM: +-------- +It has been reported, that some printers (esp. HP printers) start to +print, even during startup of the driver. This is a problem to printers +which use a bidirectional protocol for talking with the system. + + +SOLUTION: +--------- +There is no possibility to handle this problem inside the driver. +One possibility is to switch off the bidirectional communication of the +printer. The other one is to connect the printer to another parallel port. + + +******************************************************************************* +* SANE-topics * +******************************************************************************* + +Here you will not find how to setup xscanimage to work with GIMP or +how to compile sane to obtain xscanimage, for such questions, please +refer to the mailing lists at SANE (http://www.sane-project.org) + + +SYMPTOM: error in loading shared libraries: libsane.so.1: cannot open +-------- shared object file: No such file or directory + +PROBLEM: +-------- +The shared libraries for SANE have been installed to a +directory which is not in your "library path" + +SOLUTION: +--------- +SANE installs its shared libraries to PREFIX/lib/sane. +PREFIX is in general /usr or /usr/local. You should +either add the path to your /etc/ld.so.conf file or +set links to the libraries from a wellknown lib-path. +(i.e. ln -s /usr/lib/sane/libsane.so.1 /usr/lib/libsane.so.1) +In any case you should run "ldconfig -v" + + + +SYMPTOM: "no SANE devices found" +-------- +While testing with "scanimage -L" nothing happens. And when starting scanimage +this message is displayed. + +PROBLEM: +-------- +The frontend cannot find any sane-device, this might have several causes: +- the driver is not loaded +- you have no access to the driver +- the backend is not enabled +- the driver is not specified +- version conflict between the loaded driver and the backend + +SOLUTION: +--------- +Driver not loaded: check "cat /proc/pt_drv" if this does not exist, load +the driver (see INSTALL) + +No access to the driver: check "ls -l /dev/pt_drv*" and you should get + +crw-rw-r-- 1 root root 40, 0 Aug 29 15:18 /dev/pt_drv +crw-rw-r-- 1 root root 40, 0 Aug 29 15:18 /dev/pt_drv0 +crw-rw-r-- 1 root root 40, 1 Aug 29 15:18 /dev/pt_drv1 +crw-rw-r-- 1 root root 40, 2 Aug 29 15:18 /dev/pt_drv2 +crw-rw-r-- 1 root root 40, 3 Aug 29 15:18 /dev/pt_drv3 + +If not, call "make load" in the driver source directory. This call sometimes +fails, so you might create the entries by using mknod: +mknod -m 0664 /dev/pt_drv c 40 0 +mknod -m 0664 /dev/pt_drv0 c 40 0 +mknod -m 0664 /dev/pt_drv1 c 40 1 +mknod -m 0664 /dev/pt_drv2 c 40 2 +mknod -m 0664 /dev/pt_drv3 c 40 3 + +The backend is not enabled: Check the SANE configuration in /etc/sane.d, +/usr/etc/sane.d or /usr/local/etc/sane.d (depending on your installation). +Your dll.conf MUST include an enabled "plustek" entry (no "#" in front of +that line) + +The driver is not specified: Check the sane configuration (see above) +for the file plustek.conf and make sure that it contains at least +/dev/pt_drv as entry. + +Version conflict: After enabling the SANE debug messages you will +get something like upon starting scanimage: +ioctl PT_DRV_OPEN_DEVICE failed(-9019) +Version problem, please recompile driver! +In this case, you MUST recompile the backend AND the driver and reinstall +both. This feature makes sure, that both always will use the same ioctl +interface. + + +PROBLEM: +-------- +I have SANE-1.0.4 and there's no backend-directory after unpacking, so +what to do? + +SOLUTION: +--------- +You're wrong there is a backend directory!!! +Since SANE-1.0.4, there are two packages: +sane-backends +sane-frontends +For the installation, you need the backend tarball. This should give +you something like: +sane-1.0.4/sane-backends +and there below sane-backends is the backend directory. So you have +to unpack the plustek-sane...tgz in sane-backends. + + + +******************************************************************************* +* F A Q * +******************************************************************************* + +Q: What does "Turbo" mean ? + +A: Turbo does not mean, that the scanner is faster ;-) + There are currently three OpticPro9636P Scanners. + OP9636P + OP9636P+ + OP9636PTurbo + The + and the Turbo models are quite the same and the difference between + the Turbo and the non-Turbo is the scan-sensor. The Turbo uses the same + like the newer OP9636T/12000T and the non-Turbo uses the older ones + from the OP9630.... + + +Q: Is it possible to use the front-button of my Plustek 9630P ? + +A: It is possible since version 0.37. After loading the driver, + use the proc filesystem to get the button value: + cat /proc/pt_drv/device0/button0 + + +Q: Is it possible to use the scanner in gimp ? + +A: You should be able to use xscanimage with gimp: + 1. Login as root + 2. Go into the gimp plug-in directory + (on SuSE it is: /usr/lib/gimp/1.1/plug-ins/ ) + 3. Find xscanimage : which xscanimage + 4. Create a symbolic link to xscanimage : + ln -s "/path/to/"xscanimage xscanimage + 5. logout as root + Now call gimp, in the xtns menu you should find + now the Acquire Image menu entry. That's all. + If not, start gimp from an xterm and have a look at the output. + + +Q: Is there any way to load the plustek_driver at bootup-time? + +A: Normally, there is no need to do so, because once you + altered your /etc/conf.modules file (or /etc/modules.conf) + and added the suggested stuff (see INSTALL file), you won't + have to worry about loading the driver. It happens automagically. ;-) + The driver will be loaded when needed (and eventually + removed after a specific time of not using it). + + Nevertheless, another possibility is to do a modprobe pt_drv in the + /etc/rc.d/boot.local (using SuSE) or + /etc/rc.d/rc.local (using RedHat or Mandrake) file + So the driver will be loaded at bootup-time... + !!! YOUR SCANNER MUST BE ONLINE during power-up then !!! + + +Q: I wonder if Mandrake 7.1's sane rpm file has this support already built in. + My scanner light is coming on yet Xscanimage seems to report that there + is no scanner present. + +A: The backend code will be included in Version 1.0.3 of sane and + AFAIK Mandrake 7.1 only includes 1.0.2. + Anyway go to the plustek-driver download page at + http://home.t-online.de/home/g-jaeger/plustek.html + download and install the driver. + + +Q: Mandrake installs sane by default at install-time. + Is there a way to install the driver with the preinstalled sane + or should the preinstalled sane be removed first? + If so, how do I remove the preinstalled sane? + +A: I recommend to uninstall the previous one. + I think Mandrake uses RPM for managing the packages. To remove sane + do the following (as root) + + Ask for all installed packages and filter sane: + rpm -qa | grep sane + you should get something like this: + sane-1.0.1-94 + Now remove this package: + + rpm -e sane-1.0.1-94 + + After theses steps, sane should be removed. + I'm not sure, but you might have to remove the /etc/sane.d manually... + Now got to your sane-1.0.3 stuff and reinstall this + as described. Before using, you might have a look + at the dll.conf and enable only the plustek backend. + + +Q: Now that there's a sane 1.0.3, how do i uninstall sane 1.0.2? + Or can i install it over? + +A: In general you can install it over. BUT this has caused + quite a lot of trouble, so I recommend to remove the 1.0.2 + completely and install 1.0.3... + + try the rpm-tool: + Get a list of the installed packages and filter out sane: + + rpm -qa | grep sane + + You should get something like: + + sane-1.0.2-48 + + Then use rpm to remove sane: + + rpm -e sane-1.0.2-48 + + The other way is: + + rm -rf /etc/sane.d + rm -rf /usr/lib/sane + rm -rf /usr/lib/libsane* + rm -rf /usr/include/sane + rm -rf /usr/etc/sane.d + rm -rf /usr/share/sane* + rm -rf /usr/local/lib/sane + rm -rf /usr/local/lib/libsane* + rm -rf /usr/local/include/sane + rm -rf /usr/local/etc/sane.d + rm -rf /usr/local/share/sane* + + This should cover all the directories where maybe some + sane files are installed... + diff --git a/doc/plustek/MakeModule.sh b/doc/plustek/MakeModule.sh new file mode 100644 index 0000000..b684f0d --- /dev/null +++ b/doc/plustek/MakeModule.sh @@ -0,0 +1,116 @@ +#!/bin/bash +#****************************************************************************** +# +# Bash-Script to create Plustek-Scannerdriver modules for Kernel 2.4, 2.6 and 3.x +# out of the backend sources... +# + +BUILD_DIR=$PWD/build +SRC_DIR=$PWD/../../backend +MAKEFILE=$PWD/Makefile.kernel26 +KERNEL_V=`uname -r` +OSMINOR=`uname -r | cut -b 3` +OSMAJOR=`uname -r | cut -b 1` + +# +# some intro ;-) +# +echo "This script will try and build a suitable kernel-module for your system." +echo "If you'd like to make the module WITH debug output, restart this script" +echo "with as follows:" +echo "./MakeModule.sh DEBUG=y" +echo "Press <ENTER> to continue or <CTRL><C> to cancel." +read + +# +# we need to be root user... +# +echo -n "Check for root..." +if [ $EUID -ne 0 ]; then + echo -e "\b\b\b - failed" + echo "Please retry as root user." + exit -1 +fi +echo -e "\b\b\b - done." + +# +# Version checks... +# +echo -e "\nCheck for kernelversion:" +if [ "$OSMAJOR" == "3" ];then + echo "Using makefile for kernel 2.6.x - okay for kernel 3 as well..." + MAKEFILE=$PWD/Makefile.kernel26 +elif [ "$OSMINOR" == "6" ]; then + echo "Using makefile for kernel 2.6.x" + MAKEFILE=$PWD/Makefile.kernel26 +elif [ "$OSMINOR" == "4" ]; then + echo "Using makefile for kernel 2.4.x" + MAKEFILE=$PWD/Makefile.kernel24 +else + echo "Your kernelversion >"$OSMAJOR"."$OSMINOR"< is probably not supported" + exit -2 +fi + +# +# Setup... +# +echo -e "Build-directory: \n"$BUILD_DIR +echo -n "Removing build-directory..." +rm -rf $BUILD_DIR +echo -e "\b\b\b - done." + +echo -n "Creating build-directory..." +mkdir $BUILD_DIR +cd $BUILD_DIR +echo -e "\b\b\b - done.\n" + +echo -n "Linking source files..." +C_FILES=`ls $SRC_DIR/plustek-pp_*.c` +H_FILES=`ls $SRC_DIR/plustek-pp_*.h` + +for F in $C_FILES $H_FILES $SRC_DIR/plustek-pp.h $SRC_DIR/plustek_pp.c; do + ln -s $F . +done +echo -e "\b\b\b - done." + +echo -n "Copying Makefile to build-directory..." +cp $MAKEFILE Makefile +echo -e "\b\b\b - done." + +# +# Building the module... +# +echo "Making the module..." +if [ "$OSMAJOR" == "2" -a "$OSMINOR" == "4" ]; then + make all $1 +else + make -C /lib/modules/$KERNEL_V/build/ SUBDIRS=$BUILD_DIR modules $1 +fi +RES=$? +cd .. +if [ $RES != 0 ]; then + echo "There were some build errors..." + exit -1 +fi +echo "done." + +echo "Should I install the module?" +echo "Press <ENTER> to continue or <CTRL><C> to cancel." +read + +make -C $BUILD_DIR install + +echo "Should I try and load the module?" +echo "If this step fails, check the kernel-log." +echo "Press <ENTER> to continue or <CTRL><C> to cancel." +read + +make -C $BUILD_DIR load +echo "done." + +echo "Should I remove the build directory?" +echo "Press <ENTER> to continue or <CTRL><C> to cancel." +read + +rm -rf $BUILD_DIR +echo "done." diff --git a/doc/plustek/Makefile.kernel24 b/doc/plustek/Makefile.kernel24 new file mode 100644 index 0000000..207ef39 --- /dev/null +++ b/doc/plustek/Makefile.kernel24 @@ -0,0 +1,251 @@ +# Makefile for the plustek scanner driver (kernel-module) +# +############################################################################### +# +# define the directories +# +HOME_DIR := . +SRC_DIR := $(HOME_DIR) +INC_DIR := $(SRC_DIR) +OBJ_DIR := $(HOME_DIR)/obj +DOC_DIR := $(HOME_DIR)/doc +BACKEND := $(SRC_DIR) + +# +# define the used tools +# +MD = mkdir -p +CC = gcc +TAR = tar +REF = cxref + +# +# Comment/uncomment the following line to disable/enable debugging +# can also be set by commandline parameter: make all DEBUG=y +# +#DEBUG = y + +# +# common compiler options +# +OPT = -fomit-frame-pointer -D_PTDRV_VERSTR=\"$(VERSIONSTR)\" + +# +# cxref options +# +REFOPT = -xref-all -index-all -html32 + +# +# Comment out if you are not running SMP. Someone take this out of here +# when the SMP stuff gets moved out of the kernel Makefile. +# SMP = 1 +# SMP_PROF = 1 + +# +# add the following to get assembly listing +# -Wa,-alh,-L -g + +# +# get some version numbers +# +ifeq ($(LINUXVERSION),) + LINUXVERSION = $(shell uname -r) +endif + +VERSIONSTR = $(shell grep "define BACKEND_VERSION" $(SRC_DIR)/plustek_pp.c | cut -b25-50 ) + +# Change it here or specify it on the "make" commandline +ifeq ($(HEADER_PATH),) +MACHTYPE = $(shell env | grep debian-linux | wc -l | sed 's/ //g') +ifeq ($(MACHTYPE),1) +# debian + HEADER_PATH = /usr/src/kernel-headers-$(LINUXVERSION)/include +else +# redhat, slackware + HEADER_PATH = /usr/src/linux/include +endif +# HEADER_PATH = /usr/include +endif + +ifeq ($(DEBUG),y) + DEBFLAGS = -O -g -DDEBUG # "-O" is needed to expand inlines +else + DEBFLAGS = -O2 +endif + +# +# the new style reference +# +K24_HEADER_PATH = /lib/modules/$(LINUXVERSION)/build/include + +# +# try to autodetect if we can use the new style header include references +# +KERNEL_HEADERS = $(shell if test -d $(K24_HEADER_PATH); then \ + echo $(K24_HEADER_PATH); \ + else \ + echo $(HEADER_PATH); \ + fi; ) + +# +# seems to be necessary for kernels 2.4.x +# +MODVERFILE = $(shell if [ -e $(KERNEL_HEADERS)/linux/modversions.h ]; then \ + echo $(KERNEL_HEADERS)/linux/modversions.h ; \ + else \ + echo $(KERNEL_HEADERS)/linux/modsetver.h ; \ + fi ) + +MODFLAGS = -DMODULE + +# +# set MODVERSIONS if the kernel uses it +# +VERSUSED = $(shell grep 'define CONFIG_MODVERSIONS' \ + $(KERNEL_HEADERS)/linux/autoconf.h | wc -l | sed 's/ //g') +ifeq ($(VERSUSED),1) + MODFLAGS += -DMODVERSIONS -include $(MODVERFILE) +endif + + +WARNFLAGS = -Wall -Wstrict-prototypes +CFLAGS = $(WARNFLAGS) $(OPT) -D__KERNEL__ -I$(KERNEL_HEADERS) -I$(INC_DIR) -I$(BACKEND) $(DEBFLAGS) $(MODFLAGS) +MODLIB = /lib/modules/$(LINUXVERSION) + +ifdef SMP +CFLAGS += -D__SMP__ + +ifdef SMP_PROF +CFLAGS += -D__SMP_PROF__ +endif +endif + +TARGET = pt_drv + +OBJ = $(TARGET).o +NAMES := dac detect genericio image map misc models io procfs +NAMES := $(NAMES) motor p9636 ptdrv scale tpa p48xx p12 p12ccd +NAMES := $(addprefix plustek-pp_, $(NAMES)) +SRCS := $(addprefix $(SRC_DIR)/, $(NAMES)) +SRCS := $(addsuffix .c, $(SRCS)) +OBJS := $(addprefix $(OBJ_DIR)/, $(NAMES)) +OBJS := $(addsuffix .o, $(OBJS)) +INCS := scan dbg types scandata procs hwdefs sysdep +INCS := $(addsuffix .h, $(INCS)) +HDRS = $(addprefix $(INC_DIR)/plustek-pp_, $(INCS)) + +# +# the header files we need from the backend +# +BACKINCS := plustek-pp.h +BACKINCS := $(addprefix $(BACKEND)/, $(BACKINCS)) + +group = "root" +mode = "644" +INST_DIR = /lib/modules/$(LINUXVERSION)/kernel/drivers/char + +info: + @clear + @echo "Makefile to create the Plustek-Scanner kernel-module:" + @echo "all ... builds the module" + @echo "all DEBUG=y ... builds the module with debug-messages enabled" + @echo "clean ... cleans up the show" + @echo "install ... installs the module to the library path" + @echo "uninstall ... removes the module from the library path" + @echo "load ... tries to load the module and creates device nodes" + @echo "unload ... unloads the module" + + +all: .depend chkdir $(OBJ) + +# +# create object directory +# +.PHONY : chkdir +chkdir: + @-$(MD) $(OBJ_DIR) + @-$(MD) $(DOC_DIR) + +$(OBJ): $(OBJS) + $(LD) -r $^ -o $@ + +$(OBJS): Makefile $(HDRS) $(BACKINCS) + +$(OBJ_DIR)/%.o : $(SRC_DIR)/%.c + $(CC) $(CFLAGS) -c $< -o $@ + +$(OBJ_DIR)/$(OBJ): VERSION1 VERSION0 + +# +# copy the driver to the modules directory +# +install: + mkdir -p $(INST_DIR) + install -c -m $(mode) $(OBJ) $(INST_DIR) + /sbin/depmod -a + +# +# remove it +# +uninstall: + rm -f $(INST_DIR)/$(OBJ) + +# +# use modprobe to load the driver, remember to set the +# parameter in /etc/modules.conf (see sane-plustek_pp.man for more details) +# +load: $(INST_DIR)/$(OBJ) +# invoke modprobe with all arguments we got + /sbin/modprobe $(TARGET) || exit 1 + +# Remove stale nodes and replace them, then give gid and perms + rm -f /dev/$(TARGET)* + +# when using the devfs support, we check the /dev/scanner entries +# and only create links to the devfs nodes +# at least we create one link + @if [ -e /dev/scanner/$(TARGET)* ]; then \ + ln -s /dev/scanner/$(TARGET)0 /dev/$(TARGET); \ + for name in `ls /dev/scanner | grep $(TARGET)`; do \ + ln -s /dev/scanner/$$name /dev/$$name ; \ + done \ + else \ + mknod /dev/$(TARGET) c `cat /proc/devices | sed -ne "s/\([0-9]*\) pt_drv/\1/p"` 0; \ + mknod /dev/$(TARGET)0 c `cat /proc/devices | sed -ne "s/\([0-9]*\) pt_drv/\1/p"` 0; \ + mknod /dev/$(TARGET)1 c `cat /proc/devices | sed -ne "s/\([0-9]*\) pt_drv/\1/p"` 1; \ + mknod /dev/$(TARGET)2 c `cat /proc/devices | sed -ne "s/\([0-9]*\) pt_drv/\1/p"` 2; \ + mknod /dev/$(TARGET)3 c `cat /proc/devices | sed -ne "s/\([0-9]*\) pt_drv/\1/p"` 3; \ + \ + chgrp $(group) /dev/$(TARGET)*; \ + chmod $(mode) /dev/$(TARGET)*; \ + fi + +# +# unload the driver +# +unload: + /sbin/modprobe -r $(TARGET) || exit 1 + +# Remove stale nodes + rm -f /dev/$(TARGET)* + +# +# create reference docu +# +doc: chkdir + $(REF) $(REFOPT) $(INC_DIR)/*.h $(SRC_DIR)/*.c $(BACKEND)/plustek-share.h \ + -D__KERNEL__ -I$(KERNEL_HEADERS) -I$(INC_DIR) -I$(BACKEND) $(MODFLAGS) \ + -D_PTDRV_V1=$(VERSION1) -D_PTDRV_V0=$(VERSION0) -D_PTDRV_BUILD=$(BUILD) -O$(DOC_DIR) + +clean: + @-rm -f $(OBJ_DIR)/*.o .depend depend dep $(REF).* *.html $(TARGET).o + @-rm -rf $(OBJ_DIR) + @-rm -rf $(DOC_DIR) + +depend .depend dep: + $(CC) $(CFLAGS) -M $(SRCS) > $@ + +ifeq (.depend,$(wildcard .depend)) +#include .depend +endif + diff --git a/doc/plustek/Makefile.kernel26 b/doc/plustek/Makefile.kernel26 new file mode 100644 index 0000000..43a71ad --- /dev/null +++ b/doc/plustek/Makefile.kernel26 @@ -0,0 +1,124 @@ +# Makefile for the plustek scanner driver (kernel-module) +# +############################################################################### + +# +# retrieve the version numbers +# +ifeq ($(LINUXVERSION),) + LINUXVERSION = $(shell uname -r) +endif +LINUXRELEASE = $(shell uname -r | cut -d'.' -f3) + +ifeq ($(VERSIONSTR),) + ifeq ($(SUBDIRS),) + VERSIONSTR = $(shell grep "define BACKEND_VERSION" $(M)/plustek_pp.c | cut -b25-50 ) + else + VERSIONSTR = $(shell grep "define BACKEND_VERSION" $(SUBDIRS)/plustek_pp.c | cut -b25-50 ) + endif +endif + +# +# extra flags +# +EXTRA_CFLAGS += -D_PTDRV_VERSTR=\"$(VERSIONSTR)\" + +ifeq ($(DEBUG),y) + EXTRA_CFLAGS += -DDEBUG +endif + +# +# the module name +# +TARGET := pt_drv +MODULE := $(TARGET).ko + +# +# our files... +# +NAMES := dac detect genericio image map misc models io procfs +NAMES := $(NAMES) motor p9636 ptdrv scale tpa p48xx p12 p12ccd +NAMES := $(addprefix plustek-pp_, $(NAMES)) +OBJS := $(addsuffix .o, $(NAMES)) + +# +# now the kernel magic +# +ifneq ($(KERNELRELEASE),) +obj-m := $(TARGET).o + +$(TARGET)-objs := $(OBJS) + +else +KDIR := /lib/modules/$(shell uname -r)/build +PWD := $(shell pwd) + +default: + $(MAKE) -C $(KDIR) M=$(PWD) modules +endif + +# +# the installation stuff +# +group = "root" +mode = "644" +INST_DIR = /lib/modules/$(LINUXVERSION)/kernel/drivers/parport + +# +# copy the driver to the modules directory +# +install: + mkdir -p $(INST_DIR) + install -c -m $(mode) $(MODULE) $(INST_DIR) + /sbin/depmod -a + +# +# +# +uninstall: + rm -f $(INST_DIR)/$(MODULE) + +# +# use modprobe to load the driver, remember to set the +# parameter in /etc/conf.modules (see INSTALL for more details) +# +load: $(INST_DIR)/$(MODULE) +# invoke modprobe with all arguments we got + /sbin/modprobe $(TARGET) || exit 1 + +# Remove stale nodes and replace them, then give gid and perms + rm -f /dev/$(TARGET)* + +# when using the devfs support, we check the /dev/scanner entries +# and only create links to the devfs nodes +# at least we create one link + @if [ -e /dev/scanner/$(TARGET)* ]; then \ + ln -s /dev/scanner/$(TARGET)0 /dev/$(TARGET); \ + for name in `ls /dev/scanner | grep $(TARGET)`; do \ + ln -s /dev/scanner/$$name /dev/$$name ; \ + done \ + else \ + mknod /dev/$(TARGET) c `cat /proc/devices | sed -ne "s/\([0-9]*\) pt_drv/\1/p"` 0; \ + mknod /dev/$(TARGET)0 c `cat /proc/devices | sed -ne "s/\([0-9]*\) pt_drv/\1/p"` 0; \ + mknod /dev/$(TARGET)1 c `cat /proc/devices | sed -ne "s/\([0-9]*\) pt_drv/\1/p"` 1; \ + mknod /dev/$(TARGET)2 c `cat /proc/devices | sed -ne "s/\([0-9]*\) pt_drv/\1/p"` 2; \ + mknod /dev/$(TARGET)3 c `cat /proc/devices | sed -ne "s/\([0-9]*\) pt_drv/\1/p"` 3; \ + \ + chgrp $(group) /dev/$(TARGET)*; \ + chmod $(mode) /dev/$(TARGET)*; \ + fi + +# +# unload the driver +# +unload: + /sbin/modprobe -r $(TARGET) || exit 1 + +# Remove stale nodes + rm -f /dev/$(TARGET)* + +# +# cleanup the show +# +clean: + @-rm -f *.o .depend depend dep $(MODULE) $(TARGET).o $(TARGET).mod.c .*.cmd diff --git a/doc/plustek/Plustek-PARPORT-TODO.txt b/doc/plustek/Plustek-PARPORT-TODO.txt new file mode 100644 index 0000000..94ef5f7 --- /dev/null +++ b/doc/plustek/Plustek-PARPORT-TODO.txt @@ -0,0 +1,38 @@ +Plustek-PARPORT-TODO.txt (2003-29-10) Gerhard Jäger <gerhard@gjaeger.de> +======================================================================== + +TODO (in general): + - more documentation/comments on the different functions + - remove/check the CHECK marks + - add additional button handler + - adding adjustment of the scan area + - checking timeout conditions (MiscCheckTimer !) and handle them correctly + - let the 96001/3 based models work with the highest available + phys. resolution + - Create the modes out of the capability list of each device + +(USER-MODE): + - line by line feedback to the backend + + +KNOWN BUGS/Limitations: + +Model Description Workaround +------------------------------------------------------------------------------ + +OpticPro P9636 detected as OP12000 none +Genius VividPro II Film Halftone/Gray > 450 dpi didn't work none +OpticPro P4830 binary scanning does not work none +OpticPro A3I bad picture quality none +Genius HR6 negative scanning not okay none + +********************************* DONE *************************************** + +- PS/2 bidi does not work (libieee1284) +- Update the IOCTL interface and remove some old structs (CmdBlk) + +******************************** SKIPPED ************************************* + +better dither maps nobody needs halftone scanning +make the backtrace work for ASIC 96001/3 not possible due to ASIC limits +based models diff --git a/doc/plustek/Plustek-PARPORT.changes b/doc/plustek/Plustek-PARPORT.changes new file mode 100644 index 0000000..0e7d73b --- /dev/null +++ b/doc/plustek/Plustek-PARPORT.changes @@ -0,0 +1,180 @@ +Version 0.1 10/20/98 + 1.0 Initial release +Version 0.2 10/26/98 + 1.0 Fix Red/Green swap, optimize ppm output for black and white. +Version 0.3 10/28/98 + 1.0 Merge code with SANE + 2.0 Put GPL/SANE license info in all files +Version 0.4 10/31/98 + 1.0 Have driver report ASIC ID up to backend. + 2.0 Put NT stuff back in so that driver may be built under NT, try to + test build under NT but run out of memory in link. +Version 0.5 10/31/98 + 1.0 Fix parameters not getting reflected from crop. +Version 0.6 11/1/98 + 1.0 Take out unused stuff from ioctl_pt_drv.h. + 2.0 Put in VERSION info in Makefile and archiver. +Version 0.7 11/4/98 + 1.0 Fix model string when -L option is given. Add in driver version + string to model. + 2.0 Change Makefile so it builds without MODVERSIONS on distributions + that don't have it (Slackware). +Version 0.8 11/5/98 + 1.0 Change output when -L option is given to add status byte to try + to determine any differences between the 4830 and the 9630. +Version 0.9 11/6/98 + 1.0 Fix 4800 problem of inverse image when not in Binary mode. + 2.0 Put more debug stuff in, print out all read + reg's. +Version 0.10 11/6/98 + 1.0 Get halftone working. + 2.0 Put in hooks for kernel 2.1 +Version 0.11 11/17/98 + 1.0 Put in sysdep-2.1.h to Makefile distrib +Version 0.20 3/7/99 + 1.0 Lots of changes to support 9630/12000. +------------------------------------------------------------------------------- +Version 0.3 02/29/2000 + 1.0 Completely rewritten, does now support Plustek OpticPro 9636T scanner + +Version 0.31 03/20/2000 + 1.0 Fix a bug in the autodetection routine + 2.0 Enhanced backend, to reflect the different modes the OpticPro 9636T + supports + +Version 0.32 - never released + 1.0 Corrected warmup-behaviour + 2.0 Fixed "first picture corrupted"-bug for transparency mode + 3.0 Fixed a bug that causes seg-faults when using cat /dev/pt_drv + 4.0 Added SCALE.C + 5.0 Made 4830P work + +Version 0.33 - never released + 1.0 Now using parport-module info for par-mode detection + 2.0 Made 9630P work + +Version 0.34 05/10/2000 + 1.0 Data-recovery code for heavy load systems + +Version 0.35 05/19/2000 + 1.0 Made 12000P/96000P work + 2.0 Merged parallel port code from Kevin Sisson + 3.0 Added new parameter "legal" + +Version 0.36 07/22/2000 + 1.0 Improved data-recovery mechanism for ASIC96001/3 based models + 2.0 Added code for OP4800 shading + 3.0 Improved backend code to allow canceling the scan process + 4.0 OP600 and Primax 4800 Direct are working now + 5.0 Added new parameter "mov" (Model OVerride) + 6.0 Removed parameter "legal", use now "mov" + 7.0 Moved common headers for backend and driver to backend directory + and prepared the backend for SANE distribution + 8.0 OP6000 is working + 9.0 added man-page + 10.0 added multi-device support + 11.0 added support for 12-bit scans (SANE-1.0.3 and scanimage!!) + +Version 0.37 11/09/2000 + 1.0 Added file io.c for parallel port I/O + 2.0 Added file procfs.c for proc-filesystem support + 3.0 Code cleanup + 4.0 Added A3I support + 5.0 Added Michaels backend patches + +Version 0.38 11/30/2000 + 1.0 Started integration of ASIC98003 specific stuff + 2.0 renamed file transform.c tpa.c + 3.0 removed portmode stuff (I decided to support SPP/BPP and EPP, that's all) + 4.0 again code-cleanup + 5.0 PT12 is working + 6.0 Removed gray-mode scanning for TPA modes + 7.0 some minor changes according to kernel 2.4.x + +Version 0.39 08/11/2001 + 1.0 Added slowIO switch + 2.0 Fixed some minor bugs + 3.0 Added forceMode switch + 4.0 Fixed a problem in the PT12 code that causes I/O timeouts + 5.0 Added Genius Colorpage Vivivd III V2 stuff + 6.0 Added A3I stuff aquired from Plustek + 7.0 Fixed a bug, that causes some seg-faults + +Version 0.40 12/17/2001 + 1.0 The USB stuff has been included + 2.0 Fixed some minor problems + 3.0 Added Kernel 2.4.17 specific stuff + 4.0 Disabled A3I stuff as this refuses ASIC96003 based devices to work + +Version 0.41 01/10/2002 + 1.0 Changed IOCTL interface to the kernel module to allow adjustment of some + driver internal variables + 2.0 Changed plustek.conf configuration fileformat to reflect the enhanced + capabilites + +Version 0.42 02/12/2002 + 1.0 Added custom gamma table support + 2.0 Added configuration-file options to preset some stuff + 3.0 Changed all eMail entries to gerhard@gjaeger.de + +Version 0.42-11 09/17/2003 + 1.0 Some Kernel 2.6.x additions + 2.0 Added Primax 4800Direct 30Bit override switch + +Somewhere on the way to 0.43 I decided to split parport & usb +backends, so the parport backend now is also able to run complete +in userspace by using either direct port I/O or libieee1248 + +Version 0.43-2 04/15/2004 + 1.0 Fixed some kernel 2.6 issues + +Version 0.43-3 07/14/2004 + 1.0 User-space code has now the same version as the module code + +Version 0.43-4 07/15/2004 + 1.0 DevFS support for kernel 2.6 + 2.0 Removed floating point operations (Thanx to Rafal Rzepecki) + 3.0 More cleanup work + 4.0 bumped up build number + +Version 0.43-5 10/06/2004 + 1.0 Change some scanmode definitions to match the SANE predefined + ones + +Version 0.43-6 04/25/2005 + 1.0 Removed inline stuff from kernel module code, as GCC3.4 will + make some trouble + +Version 0.43-7 05/11/2005 + 1.0 Fixed timer issues in user-space (Thanx to Rod Roark) + +Version 0.43-8 08/10/2005 + 1.0 Fixed problem on ASIC96003/1 based devices, when scanning + lineart at 200 and 300 dpi + +Version 0.43-9 11/14/2005 + 1.0 Fixed sizeof(long) issue for 64bit platforms, see + bug #302195 + +Version 0.43-10 02/16/2006 + 1.0 Fixed bug, that prevents backend from working, when the + device is connected to parport1 or higher, parport0 works. + Bug spotted by Christoph Steinbruchel. + +Version 0.43-11 08/09/2006 + 1.0 Changed sane.type to "flatbed scanner" only + +Version 0.43-12 08/29/2006 + 1.0 Fixed "not homing" problem, the sensor did not + return when the driver gets the _IOCTL_STOP command + 2.0 Fixed compilation issue for kernels > 2.6.15 + 3.0 Fixed compiler warning conditions + +Version 0.43-13 06/18/2007 + 1.0 Removed DEVFS stuff for Kernels > 2.6.15 + 2.0 Added class stuff + +Version 0.44-1 02/13/2013 + 1.0 Fixed issues for Kernels > 2.6.35 + 2.0 Define ULong and friends to uint32_t to make the driver + work in a 64 bit environment
\ No newline at end of file diff --git a/doc/plustek/Plustek-PARPORT.txt b/doc/plustek/Plustek-PARPORT.txt new file mode 100644 index 0000000..e5b6097 --- /dev/null +++ b/doc/plustek/Plustek-PARPORT.txt @@ -0,0 +1,49 @@ +Plustek-PARPRORT.txt (2004-03-28) Gerhard Jäger <gerhard@gjaeger.de> +==================================================================== + + +Beginning with SANE-1.0.13, there's a backend called plustek_pp. +This is for controlling Plustek parallel-port scanner and compatible +devices. For a full listing, see plustek_pp.desc. + +This code formerly was available for creating the Linux kernelmodule +pt_drv. This should no longer be necessary. You should be able to use +the backend out of the box. + + +The kernel module +----------------- + +As it might be helpful to create and use the kernel-module, this way still +exists. It's possible to create this module out of the backend sources +in sane-backends/backend. Simply do (as root user) + +./MakeModule.sh + +Then the module should be compiled, installed and loaded. + +Add the following three lines to file /etc/modules.conf + +alias char-major-40 pt_drv +pre-install pt_drv modprobe -k parport +options pt_drv lampoff=180 warmup=15 port=0x378 lOffonEnd=0 mov=0 slowIO=1 + +See man page for sane-plustek_pp ("man sane-plustek_pp") for explanation of +these options. + +Now "scanimage -L" should show something like this: +device `plustek:/dev/pt_drv' is a Plustek 9630P flatbed scanner + + +Known Problems: +--------------- + +Sometimes it is necessary to change the ioctl-interface between the +driver and the backend, in this case the version number of the communication +protocol will be changed and newer drivers won't work with older backends +and vice versa. +In this case (error -9019 in the SANE debug output!!) you have to recompile SANE +AND the driver (have a look at the installation procedure above). + + + diff --git a/doc/plustek/Plustek-USB-TODO.txt b/doc/plustek/Plustek-USB-TODO.txt new file mode 100644 index 0000000..0ea7e8b --- /dev/null +++ b/doc/plustek/Plustek-USB-TODO.txt @@ -0,0 +1,53 @@ +Plustek-USB-TODO.txt (2007-12-13) Gerhard Jaeger <gerhard@gjaeger.de> +===================================================================== + +TODO (in general): +- more documentation/comments for the different functions +- replace mclk_fast stuff and use instead values from 75DPI grayscale + + +KNOWN BUGS/Limitations: + +Model Description Workaround +------------------------------------------------------------------------------ + +HP 2100c Straight Line Bug none +Bearpaw1200 works only once restart backend each time +All with TPA Negative Scanning quality bad none +CIS devices Skip calibration does not work none + +********************************* DONE *************************************** + +all add plustek.conf presettings to done + interface for frontends + +all Auto-warmup done +all Sensor movement speedup ffw/rev done + +Canon fine calibration does not work fixed +Canon 650 color modes >=600 dpi too dark fixed +Canon 660 add support done +Canon 1220 image quality not very good fixed +Canon 1240 gray mode too bright fixed + +HP 2100c hits against scan-bed fixed +Genius Devices not recognized fixed +EPSON Photo Permanently warmup on TPA scanning fixed +EPSON Photo Transparency mode does not work fixed + correctly +all binary mode does not work correctly done +Canon 16bit gray mode does not work done +EPSON 1250/60 Motor control settings are wrong fixed +CanoScan 1240 added +EPSON + HP vertical line bug fixed +HP blooming on 42bit >=400dpi fixed +UMAX5400 bad picture quality fixed +UMAX3400 Straight Line Bug fixed +all problems on ARM and XScale fixed + +Make the color modes > 8bit work correctly done +make mono/gray modes work on Canon scanners done +copy picture line by line to backend done +cancel function does not work correctly in large resolutions fixed +Move documents to its correct SANE place done + diff --git a/doc/plustek/Plustek-USB.changes b/doc/plustek/Plustek-USB.changes new file mode 100644 index 0000000..8ed5cc8 --- /dev/null +++ b/doc/plustek/Plustek-USB.changes @@ -0,0 +1,307 @@ +Plustek.changes - Gerhard Jaeger <gerhard@gjaeger.de> +===================================================== + +Note: All other not mentioned builds have never been released. + +V 0.52-11 (2013-02-13) +--------------------- +Added support for Q-Scan A6 scanner (see report #312073) + +V 0.52-10 (2012-03-19) +--------------------- +Fixed batch scanning + +V 0.52-9 (2009-04-27) +--------------------- +Tweaked highspeed settings for Epson 1260 + +V 0.52-8 (2008-08-25) +--------------------- +Added preliminary support for Visioneer XP100 +Tweaked entry for LM9831 version of Plustek UT12 + +V 0.52-7 (2007-12-13) +--------------------- +Fixed alignment issues for archs like XScale/ARM +Fixed dumppic function to get valid grayscale pictures + +V 0.52-6 (2007-11-23) +--------------------- +Tweaked TravelScan464 settings. +Improved AFE gain calculation for CIS devices. This should avoid +stripes in the scanned images. + +V 0.52-4 (2007-11-16) +--------------------- +Tweaked TravelScan464 settings +Added possibility to disable dark-calibration with lamp on +(touches devices like CanoScan1220 etc) +Use attribute packed for data access structs + +V 0.52-4 (2007-11-07) +--------------------- +Added support for Syscan TravelScan464 + +V 0.52-3 (2007-07-26) +--------------------- +Checked UMAX3400/3450 models +Force output bit set (MIO), when lamp is switched + +V 0.52-2 (2007-07-10) +--------------------- +Added flag to allow only 1- and 8-bit scanmodes. The Q-Scan does +not seem to support 14-bit modes. + +V 0.52-1 (2007-06-29) +--------------------- + +Added sheetfed device Q-Scan USB001 from Portable Peripherals +Fixed Mustek Bearpaw and did some speedup (bugreports #304343 and +#301763) +Fixed calibration for senororders other that RGB + +V 0.51-16 (2007-03-05) +---------------------- + +Added CanoScan to all Canon device strings. +Identified one more Plustek device as U24. +Fixed button handling for Plustek/KYE devices and +added some more debug messages. + +V 0.51-15 (2006-11-22) +---------------------- +Fixed option descriptors (bugreport #303786) + +V 0.51-14 (2006-08-11) +---------------------- +Lowered speed for LiDE20/30 models in low resolution mode (<75dpi) + +V 0.51-13 (2006-08-09) +---------------------- +Changed sane.type to "flatbed scanner" only +Fixed motorsettings for LiDE30 (bugreport #303722) +Fixed LiDE20 settings. Got some reports of non-working devs, although +my N670U is working fine. Lowered the max speeds and increased the +MCLK-div for fast movement. (bugreport #303695) + +V 0.51-12 (2006-06-30) +---------------------- +Fixed bit-depth selection + +V 0.51-11 (2006-06-08) +---------------------- +Fixed calibration: All modes will be coarse calibrated now, fine calibration +will be done only for 8 bit modes. +Fixed paper eject during startup. + +V 0.51-10 (2006-05-28) +--------------------- +Released + +V 0.51-8 (2006-04-09) +--------------------- +Added cache for fine calibration (CIS devices only) +Added support for sheet-fed scanner (CIS devices only) +Added support for Syscan TravelScan662 + +V 0.50-10 (2006-01-31) +--------------------- +Fixed CanoScan N1220U settings. + +V 0.50-9 (2006-01-19) +--------------------- +Added high-speed setting for HP2200. +Fixed constraint_type for OPT_BUTTON. + +V 0.50-8 (2006-01-18) +--------------------- +Fixed CanoScan N670U settings (bugreport #302433). + +V 0.50-7 (2005-10-28) +--------------------- +Changed high-speed setting for UMAX 3400, due to bugreport #302317. +Added disable speeup feature (as option). +Fixed CanoScan N650U settings (bugreport #302433). + +V 0.50-6 (2005-10-25) +--------------------- +Fixed segfault in fine calibration for CIS devices. +Tweaked LiDE25 settings (thanks to Gerd Cendelin for lending me +his LiDE25). +Let 1200DPI CIS devices use now the green-channel for gray scans. + +V 0.50-5 (2005-10-21) +--------------------- +Fixed high-speed feature of CanoScan D660U. + +V 0.50-4 (2005-10-02) +--------------------- +Updated motor settings for Canoscan LiDE25, thanks to +Stephan February <stephanf@singnet.com.sg> for providing these +values. + +V 0.50-3 (2005-08-22) +--------------------- +Fixed problem, when trying to scan at resolutions beyond the +optical one (sensor stops too early) + +V 0.50-2 (2005-08-19) +--------------------- +More cleanup. +Removed obsolete _WAF_BLACKFINE. +LiDE20 does not seem to have a reliable black calibration area, +so the devices now will switch off the lamp for dark calibration. +Fixed line statistics and added calibration data output. + +V 0.50-1 (2005-08-15) +--------------------- +Cleanup +Activated IPC between reader-process and parent +Added button support for Plustek/Genius devices +Did some whitespace cleanup and removed those *p typedefs +Tried to improve CIS calibration (somewhat back to 1.0.13) +Added _TWEAK_GAIN definition to allow CIS calibration increase +gain values during basic lamp setup phase +Added call to speedtest to CIS calibration +Fixed CanoScan startup problem (the way misc I/Os will be set +after power-up has done the trick) + +V 0.49-8 (2005-08-08) +--------------------- +Added LiDE25 settings + +V 0.49-7 (2005-07-21) +--------------------- +Bugfix release - fixed lampsettings for CIS calibration + +V 0.49-6 (2005-07-07) +--------------------- +Bugfix release - calibration and segfault :( + +V 0.49-5 (2005-07-04) +--------------------- +Added IPC between reader-process and parent process to +be pepared for sharing calibration info. +Added button support for backend + +V 0.49-4 (2005-06-27) +--------------------- +Fixed "double free" problem during sane_exit. + +V 0.49-3 (2005-04-18) +--------------------- +Fixed backtracking problem, when sensor-speedup is active. + +V 0.49-2 (2005-01-12) +--------------------- +Tweaked some device settings +Updated copyright info +Using now PhyDpi.y as selector for the motor MCLK range + +V 0.49-1 (2004-11-19) +--------------------- +Improved multi-device capability + +V 0.48-10 (2004-10-30) +---------------------- +- Fixed a bug in buffer calculation for CIS devices + +V 0.48-9 (2004-10-19) +--------------------- +- Use the same option names for RGB gain as umax_pp backend + +V 0.48-8 (2004-10-06) +--------------------- +- split scanmodes & bit-depths + +V 0.48-7 (2004-10-04) +--------------------- +- changed some strings according to bug #300963 + +V 0.48-6 (2004-09-07) +--------------------- +- fixed UMAX lamp-off bug + +V 0.48-4 (2004-07-26) +--------------------- +- added speedup parameters for UMAX3400/3450 +- added disableSpeedup option +- added UMAX3400/3450 TPA autodetection + +V 0.48-3 (2004-07-16) +--------------------- +- improved autowarmup feature +- improved speedup feature + +V 0.48-2 (2004-07-07) +--------------------- +- fixed binary mode for CanoScan D660U + +V 0.48-1 (2004-07-02) +--------------------- +- added more options to the frontend interface +- added UMAX 3400 with product ID 0x0050 +- added support for the UMAX 3450 +- added autowarmup for CFL (most CCD-scanner) +- added speedup for sensor for some devices +- fixed lamp switching problem for D660U +- fixed negative scanning lamp problem +- fixed Gray 16 mode for CanoScan D660U +- code cleanup + +V 0.47-11 (2004-04-21) +---------------------- +- minor fixes, esp. for MacOSX +- added lamp-off and warmup options to be configurable by backends + +V 0.47-8 - 0.47-10 +------------------ +- never released + +V 0.47-7 (2004-02-08) +--------------------- +- fixed bypass calibration function for CIS devices +- added lampoff stuff to config file + +V 0.47-6 (2004-01-21) +--------------------- +- Cleanup work +- Changed lamp off timer for non-setitimer systems + +V 0.47-5 (2004-01-09) +--------------------- +- Added big-endian code to support also PPC architectures +- Added mov (model override) to support various devices using + the same product ID (namely Mustek BearPaw 1200) +- Added per-model gamma values +- Fixed a bug in the coarse CIS-lamp calibration +- Cleanup work + +V 0.47-1 - 0.47-4 +----------------- +- never released + +V 0.46-9 (2003-11-03) +--------------------- +- Added alternate calibration for CIS devices +- Added coarse calibration data-cache +- Moved parallelport code to extra backend file plustek_pp +- Fixed some minor bugs +- Fixed problem with CIS devices using FreeBSD + +V 0.46-1 - 0.46-8 +----------------- +- never released + +V 0.45 (2003-08-05) +------------------- +- Starting this changelog +- Updates since V 0.44: + * better CIS device support + * added CanoScan N1220U, N1240U + * made CanoScan N650U and N670U work + * fixed EPSON 1260 motor killer-bug + * improved TPA scanning + * fixed TPA warmup bug on EPSON 1260/Photo + * fixed Genius device detection problem + * Added Compaq S4-100 (identical to UMAX3400) diff --git a/doc/plustek/Plustek-USB.txt b/doc/plustek/Plustek-USB.txt new file mode 100644 index 0000000..e1ee793 --- /dev/null +++ b/doc/plustek/Plustek-USB.txt @@ -0,0 +1,457 @@ +Plustek-USB.txt (2005-08-08) Gerhard Jäger <gerhard@gjaeger.de> +=============================================================== + +NOTE: +----- + +ALL YOU NEED TO RUN YOUR USB SCANNER IS ALREADY INCLUDED. THERE'S +NO NEED TO INSTALL THE KERNEL MODULE pt_drv. THIS ONE IS ONLY NEEDED +FOR THE PLUSTEK PARALLELPORT SCANNER. + + +List of all currently implemented devices +----------------------------------------- + +Quite a lot of them are not tested or even identified. Please send me the ids +of your device... + + +PLUSTEK +======= +Vendor ID: 0x07B3 + +Product ID Device-description and status +0x0005 not identified yet +0x0007 not identified yet +0x000F not identified yet +0x0010 U12 - working +0x0011 U24 (LM9831) - working +0x0012 not identified yet +0x0013 UT12 (LM9831) - working +0x0014 not identified yet +0x0015 U24 (LM9832) - working +0x0016 not identified yet +0x0017 UT12 - working, UT16 - working, UT24 - working + +MUSTEK +====== +BearPaw vendor ID: 0x0400 (They use the NationalSemiconductors ID!!!) + +Product ID Device-description and status +0x1000 BearPaw 1200 (LM9831) - working +0x1001 BearPaw 1200 (LM9832) - not tested +0x1001 BearPaw 2400 (LM9832) - mostly working + +KYE (Genius) +============ +Vendor ID: 0x0458 + +Product ID Device-description and status +0x2007 ColorPage-HR6 V2 - working +0x2008 ColorPage-HR6 V2 - not tested +0x2009 ColorPage-HR6A - not tested +0x2013 ColorPage-HR7 - working +0x2015 ColorPage-HR7LE - not tested +0x2016 ColorPage-HR6X - not tested + +Hewlett Packard +=============== +Vendor ID: 0x03F0 + +Product ID Device-description and status +0x0505 HP Scanjet 2100c - working +0x0605 HP Scanjet 2200c - working + +EPSON +===== +Vendor ID: 0x04b8 + +Product ID Device-description and status +0x010F EPSON Perfection 1250/Photo - working +0x011D EPSON Perfection 1260/Photo - working + +UMAX +==== +Vendor ID: 0x1606 + +Product ID Device-description and status +0x0050 UMAX 3400 - working +0x0060 UMAX 3400/3450 - working +0x0160 UMAX 5400 - working + +COMPAQ +====== +Vendor ID: 0x049F + +Product ID Device-description and status +0x001A S4-100 - working, identical with UMAX 3400 + +CANON +===== +Vendor ID: 0x04A9 + +Product ID Device-description and status +0x???? FB620U - not integrated +0x2206 N650U - working +0x2207 N1220U - working +0x2208 D660U - working +0x220D N670U/LiDE20 - working +0x2220 LiDE25 - working +0x220E N1240U/LiDE30 - working + + +How to use and configure the Plustek USB backend +------------------------------------------------ + +Please note, that the following is only needed, if you need to upgrade +a SANE version... + + +Preparations +------------ + +What do we need ? + +a SANE backends archive (i.e. sane-backends-1.0.9.tar.gz) +a driver archive (i.e. plustek-sane-0.45-1.tar.gz) + +The latest SANE archive can be obtained at: +http://www.sane-project.org +and the latest backend at: +http://www.gjaeger.de/scanner/plustek.html + +Assumptions +----------- + +Our starting point is your home-directory: + +:~> + +The packages (here sane-backends-1.0.9.tar.gz und plustek-sane-0.45-1.tar.gz) +are in the /tmp directory. + +Let's go: +--------- + +Change to your home directory and create a sane directory +cd ~ +mkdir sane + +unpack your sane tar-ball (here "sane-backends-1.0.9.tar.gz") +to this "sane" directory + +cd sane +tar xvzf /tmp/sane-backends-1.0.9.tar.gz + +Now unpack your plustek-sane tarball (here "plustek-sane-0.45-1.tar.gz") +to the backends directory: + +cd sane-backends-1.0.9 +tar xvzf /tmp/plustek-sane-0.45-1.tar.gz + +Now do the ./configure step... +Especially for SuSE with a preinstalled SANE-RPM: +./configure --prefix=/usr --sysconfdir=/etc --mandir=/usr/share/man --infodir=/usr/share/info + +Then do the make step and after that as root user perform the make install +step. + +That's all! + +Before using, make sure that the USB scanner device driver is loaded: +modprobe scanner +or +modprobe scanner vendor=0x7b3 product=0x17 + +You might need to add the following line to /etc/modules.conf, if the scanner is not +supported directly - vendor and product must match your device! +options scanner vendor=0x7b3 product=0x17 + +If you're not sure about the vendor and product id of your device, simply load +the USB subsystem and plug in your scanner. Then do a +cat /proc/bus/usb/devices +and look for the scanner + +Now you have to configure the backend. Edit the file /etc/sane.d/plustek.conf +and fill in the appropriate vendor and product id (see there for examples) + +This is it... + + +Autoloading scanner.o +--------------------- + +To perform an automatic load of the scanner module, you might add the modprobe +line to your boot.local file. +i.e. on SuSE system > 7.x +/etc/init.d/boot.local +An alternative way is to use the hotplug utilities. + + +Using hotplug utilities and libusb +---------------------------------- + +When using libusb with SANE, then you should also use the hotplug utilities to +automatically setup your device nodes (at least the permissions) and prevent +scanner.o from loading. + +Assuming, that these utilites are properly installed on your box, you have +to tweak and add some files. + +In directory (where of course the config files reside): +/etc/hotplug + +Append the line +scanner +to file +blacklist + +This prevents the scanner module from the usb-subsystem to be loaded. + +Next thing is to add a new line in +usb.usermap: + +usbscanner 0x0003 0x1606 0x0160 0x0000 0x0000 0x00 0x00 0x00 0x00 0x00 0x00 0x00 000000 + +The example line shows the vendor ID of UMAX 0x1606 and the product ID of the 5400 +0x0160 - these values must match the ones of your scanner. + +Last step is adding a script in +/etc/hotplug/usb/ +It is called "usbscanner" (don't forget the executable rights) + +------------------------------------------------- +# !/bin/bash + +if [ "${ACTION}" = add ] && [ -f "${DEVICE}" ] +then + chgrp users "${DEVICE}" + chmod ug+rw "${DEVICE}" +fi +------------------------------------------------- + +This script will correct the access right to your scanner device when the device +is plugged in. + + +Debugging your USB scanner (written by Kev Green) +------------------------------------------------- + +Firstly, are you running the latest version of SANE, and the Plustek USB driver? + +Double check at http://www.gjaeger.de/scanner/plustek.html +(Plustek USB driver) and http://www.sane-project.org (SANE Suite) to +make sure you are. + +Now, in order to test and utilise your scanner with the Scanner Access Now Easy +(SANE) system, there are basically two programs that you will need to use from +the sane-frontends (versions post 1.0.3) or sane (versions pre 1.0.3) packages. + +Firstly, the sane-find-scanner program will allow you to locate your scanner, +and help you work out if sane knows where it is. This is not definite however, +as (you'll see this too when using sane-find-scanner) it will find all the +scanners on your USB and SCSI bus, whereas sane will only work when you have +the relevant sane backend installed and configured correctly for the relevant +scanner. + +So, if sane-find-scanner can find your scanner, but scanimage doesn't work, +then you need to check, double check, and even triple-check your SANE +configuration file for plustek scanners (plustek.conf, usually in +/etc/sane.d +or maybe in /usr/local/sane/ or somewhere, a "find" will be able to tell you +where. It may take a while though!). + +If sane-find-scanner doesn't find your scanner, then you should check to see if +the kernel recognises it at all, which you can do with the following command, +as mentioned above: + +cat /proc/bus/usb/devices + +If your scanner doesn't appear in there, then the kernel has not recognised it. +There may however be a crypic-looking entry in there which doesn't name itself +as plustek, in which case it is recognised as being there by the kernel, but is +not know to the kernel's USB device database, in which case this should be +mentioned on the plustek list (<plustek@linuxhacker.org>), from where +the "powers that be" will deal with it. + +If that file does not exist in the /proc filesystem, then you don't have the +"Preliminary USB filesystem" option set in your kernel, and you will need to +recompile your kernel to allow you to do this step in debugging. If it +exists, but is empty, you will have to make sure (use the kernel configure +help information and the details of your motherboard to ascertain this!) +you have the right one of UHCI or OHCI USB modules installed or compiled into +your kernel. + +Once you've established that the kernel has recognised your scanner, you can +start pointing the finger at SANE, or simply the "scanner" module. You will +need to have selected the "USB Scanner" option in your kernel compilation as +a module, or compiled into the kernel. If you have done neither, then SANE will +simply not be able to recognise your scanner. Rectify that, if you are missing +it. + +If you have carried out all of the above steps, then sane-find-scanner should +be able to recognise your scanner correctly. + +sane-find-scanner probes all of the devices on the SCSI and USB busses, and +so you may find that it outputs "unable to get minor data" errors or similar +to your terminal, or to your error logs, you can safely ignore these as long +as it does that, and has recognised your scanner. + +Once you have found your scanner okay using sane-find-scanner, then you are +ready to start messing around with the actual scanimage program to attempt +to scan an image in. + +Obviously (although this may turn out to be premature) you should now have +something in your scanner ready to scan for verification. + +At this stage, you should begin to be warey, because while the USB stuff for +Plustek scanners is in development it may (like any kernel/module related software) +crash your system with a kernel panic, or simply just segfault, so for your own sake, +close down all the applications you are running and ONLY use text console, rather +than X-Windows at this point, as you will certainly want to avoid any +potential filesystem corruption. + +At this point it's probably also good to ensure that you have selected the +"Magic SysRq Key" option in the kernel hacking section of the kernel config, +and done: + +echo 1 > /proc/sys/kernel/sysrq + +And of course read the readme for that in /usr/src/linux/Documentation, as that +should allow you to avoid filesystem corruption during any crashes that might +happen. + +Now, before running scanimage, you should enable the maximum levels of +debugging possible in both the SANE core and in the Plustek scanner +backend. + +To do this you should do: + +export SANE_DEBUG_PLUSTEK=12 +export SANE_DEBUG_DLL=12 + +Now, if you run scanimage, you should be able to see the maximum debugging +messages. + +If those messages don't tell you what is wrong, then take note of your scanner +type, the contents of /proc/bus/usb/devices, and the contents of your error +log, as well as (where possible) the output of the scanimage command (if +you can't capture it directly, an as-accurate-as-possible description is +MUCH better than nothing!), and try and get all of the activity that +your scanner did as well (light +came on? didnt? motors came on? etc?) and email that to the plustek list +(<plustek@linuxhacker.org>). + +For the particularly adventurous only... + +You might want to try running the scanimage program through strace or +gdb to see if you can go some or all of the way to debugging the problem +yourself, and post the relevant (ie. last!) parts of those utilities +output to the plustek-help list. If you don't know what strace or gdb +are, then you should probably not try that. + +It's probably safe to run your scanner in a normal operating environment +under Linux once you have got it working once in a text console. + + +How to add a new LM9831/2 based device description +-------------------------------------------------- + +If you have a LM9831 or LM9832 based scanner and your vendor and product id did +not appear in the list above and you're willing to do some experiments, then +simply add your device to the file plustek-devs.c at the end of the list. +See the list at the end of the file how this works... + + +Thanx guys for helping: +----------------------- + +Henning Meier-Geinitz henning@meier-geinitz.de for Mustek Bearpaw testing, ideas and patches +Stefan Nilsen stefan.nilsen@telia.com for HP2200c testing and patches +Kev Green kyrian@ore.org for Documentation, UT12 testing and RPMS. +Holger Bischof bischof@cs.tu-berlin.de for OpticPro U12 testing +Abhijit Sovakar a.sovakar@gmx.de for OpticPro UT24 testing +Peter Koellner peter@mezzo.net for OpticPro U24 testing +Gene Heskett gene_heskett@iolinc.net for EPSON 1250 testing and patches +Tasnim Ahmed tasnim_ahmed@yahoo.com for KYE Colorpage HR6 testing +Reinhard Max max@suse.de for EPSON 1250 testing and patches +Allan N. Hessenflow allan@kallisti.com for UMAX 3400 testing and patches +Craig Smoothey craig@smoothey.org for HP2100c testing and patches +Till Kamppeter till.kamppeter@gmx.net for EPSON1260 testing +Christopher Montgomery monty@xiph.org for CanoScan calibration works +Olaf Leidinger leidola@newcon.de for CanoScan LiDE25 testing + +and a lot of other guys not named here... + + +Some equations which are used to calculate some stuff +----------------------------------------------------- + +The stuff is used in plustek_usbscan.c to get the correct register settings +out of the hardware descriptions: + +Crystalfrequency is 48MHz + + +Pixel Period (seconds/pixel) +---------------------------- + + MCLK_DIV x 8 x CM +PixelPeriod = ----------------- + 48MHz + + +Integration Time (seconds) +-------------------------- + +Tint = PixelPeriod X LineLength + + MCLK_DIV x 8 x CM X LineLength +Tint = ------------------------------ + 48MHz + + +Ideal MCLK Divider +------------------ + 48MHz (DataPixelsEnd - DataPixelStart) +MCLK_DIV = --------------------------- x -------------------------------- + HostIORate x 8 x LineLength HDPI_ADJ x PP + + 48MHz (DataPixelsEnd - DataPixelStart) CM + = ------------------------------- x -------------------------------- x ---- + BytesPerSecond x 8 x LineLength HDPI_ADJ x PP CM + + 48MHz 1 + = --------------------------- x Tint x --------- + 8 x LineLength CM + + 48MHz (Tint * 1000) + = ----------------------- x -------------------- + 1000 x 8 x LineLength CM + + +Ideal Scan Speed (inches/second) + + 1 1 1 +ScanSpeed = ----------- x -------- x -------- + PixelPeriod FSPI x 4 StepSize + + 48MHz 1 1 +ScanSpeed = ----------------- x -------- x -------- + MCLK_DIV x 8 x CM FSPI x 4 StepSize + + + 48MHz +MCLK_DIV = ------------------------------------------ + ScanSpeed x 8 x CM x FSPI x 4 x StepSize + + + +---------------- +Symbols: + +CM - Color Mode (unitless), 3 for channel pixel rate color, else 1 + + +------------------------- + |