primax: some tests with Primax USB 19200
Subject:
some tests with Primax USB 19200
From:
Lars Lindner ####@####.####
Date:
26 Jan 2003 17:03:13 -0000
Message-Id: <1043600023.1343.123.camel@llando.workgroup>
Hi!
As an owner of the Primax USB 19200 I read trough
this mailing list... and my hopes faded away :)
I've installed the E3 driver v0.6 with sane 1.09
with the scanner_erik_2.4.17 patch on 2.4.19 kernel.
Steven Ellis wrote (http://schwarz.thueday.de/primax/0451.html):
"Plus I have now had confirmation that the driver
supports the Primax 19200 USB scanner."
Does that mean its IDs are recognized, or its
really working? I'm wondering because I found now
solution for the double IN endpoint problem.
Maybe this is already known, but if anyone is interested
I experimented a little and these are my results:
I've tried the IBM Ideascan 2000 driver in
the following combinations and all of them failed
(the init program of the driver reported a lot
of unexpected answers: code 84 and the scan
program did hang).
ep1 ep2 ep3
---------------
OUT IN %
OUT % IN
OUT IN INTR
OUT INTR IN
% means endpoint unused
I tried the same combinations using the E3 driver,
but "scanimage -L" always did a segfault. An
strace showed this:
old_mmap(0x40291000, 1266688, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_FI
ONYMOUS, -1, 0) = 0x40291000
close(3) = 0
open("./viceo.conf", O_RDONLY) = -1 ENOENT (No such file or d
open("/usr/local/etc/sane.d/viceo.conf", O_RDONLY) = 3
fstat64(3, {st_mode=S_IFREG|0644, st_size=22, ...}) = 0
old_mmap(NULL, 4096, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_ANONYMOUS,
x4001a000
read(3, "usb /dev/usb/scanner0\n", 4096) = 22
ioctl(0, 0x48, 0xbfffdc20) = -1 EINVAL (Invalid argument)
open("/dev/usb/scanner0", O_RDWR) = 4
ioctl(4, 0x49, 0xbfffd790) = -1 ETIMEDOUT (Connection tim
ioctl(4, 0x48, 0xbfffdc00) = -1 ETIMEDOUT (Connection tim
close(4) = 0
open("/dev/usb/scanner0", O_RDWR) = 4
ioctl(4, 0x49, 0xbfffd740) = -1 ETIMEDOUT (Connection tim
ioctl(4, 0x48, 0xbfffdbb0) = -1 ETIMEDOUT (Connection tim
ioctl(4, 0x47, 0xbfffdbd0) = -1 ETIMEDOUT (Connection tim
--- SIGSEGV (Segmentation fault) ---
+++ killed by SIGSEGV +++
On a win98 installation I used USBsniff to trace
scanner initialization and preview scan, the results
show the only the OUT and the second IN endpoints
ared used in the whole log.
If someone is interested in the logs:
http://llando.homeunix.org/doc/19200/index.htm
Finally: It would be nice if someone can help with the
segfault or other solutions for the 19200.
Lars Lindner
