Hi Ludovic,
I really appreciate your quick feedback. I only hope I could have that
kind of quick response for bugreports against my own free software projects.
On Thu, Mar 24, 2011 at 10:59:36PM +0100, Ludovic Rousseau wrote:
> >So, as it seems, somehow, if openct is installed (and not running), pcscd
> >will
> >not try to use its internal CCID driver? Is this intentional?
>
> openct also provides a ifd-handler for pcscd.
yes.
> So the problem may be the openct handler causing a NULL devpath. I
> will have a look.
actually, the devpath is resolved from libudev _before_ the ifdhandler is
even called. So I would be surprised if openct somehow causes the problem.
> >After applying your patch _and_ removing openct (no
> >/usr/lib/pcsc/drivers/openct-ifd.bundle/Contents/Linux/openct-ifd.so present
> >anymore), I was able to make it work:
>
> Can you also try _without_ the patch?
without openct installed, and without the patch, as expected, pcscd still
crashes at exactly the same location:
(gdb) run -f -a -d
Starting program: /usr/sbin/pcscd -f -a -d
[Thread debugging using libthread_db enabled]
00000000 debuglog.c:277:DebugLogSetLevel() debug level=debug
00000329 configfile.l:245:DBGetReaderListDir() Parsing conf directory:
/etc/reader.conf.d
00000028 configfile.l:287:DBGetReaderList() Parsing conf file:
/etc/reader.conf.d/libccidtwin
00000059 configfile.l:287:DBGetReaderList() Parsing conf file:
/etc/reader.conf.d/0comments
00000047 pcscdaemon.c:550:main() pcsc-lite 1.7.0 daemon ready.
00001384 hotplug_libudev.c:258:get_driver() Looking for a driver for VID:
0x1D6B, PID: 0x0001, path: (null)
00000157 hotplug_libudev.c:258:get_driver() Looking for a driver for VID:
0x1D6B, PID: 0x0001, path: (null)
00000159 hotplug_libudev.c:258:get_driver() Looking for a driver for VID:
0x1D6B, PID: 0x0002, path: (null)
00000155 hotplug_libudev.c:258:get_driver() Looking for a driver for VID:
0x1D6B, PID: 0x0001, path: (null)
00000137 hotplug_libudev.c:258:get_driver() Looking for a driver for VID:
0x1D6B, PID: 0x0001, path: (null)
00000137 hotplug_libudev.c:258:get_driver() Looking for a driver for VID:
0x1D6B, PID: 0x0001, path: (null)
00000136 hotplug_libudev.c:258:get_driver() Looking for a driver for VID:
0x1D6B, PID: 0x0001, path: (null)
00000142 hotplug_libudev.c:258:get_driver() Looking for a driver for VID:
0x1D6B, PID: 0x0002, path: (null)
[New Thread 0x7f59e74ba700 (LWP 20165)]
04034678 hotplug_libudev.c:258:get_driver() Looking for a driver for VID:
0x076B, PID: 0x3021, path: (null)
00000042 hotplug_libudev.c:309:HPAddDevice() Adding USB device: OmniKey CardMan
3121
Program received signal SIGSEGV, Segmentation fault.
[Switching to Thread 0x7f59e74ba700 (LWP 20165)]
__strlen_sse2 () at ../sysdeps/x86_64/multiarch/../strlen.S:31
31 ../sysdeps/x86_64/multiarch/../strlen.S: No such file or directory.
in ../sysdeps/x86_64/multiarch/../strlen.S
(gdb) bt full
#0 __strlen_sse2 () at ../sysdeps/x86_64/multiarch/../strlen.S:31
No locals.
#1 0x00007f59e7536856 in __strdup (s=0x0) at strdup.c:42
len = <value optimized out>
#2 0x000000000040d502 in HPAddDevice (dev=<value optimized out>, parent=<value
optimized out>, devpath=0x0)
at hotplug_libudev.c:368
i = 0
deviceName = "usb:076b/3021:libudev:0:(null)", '\000' <repeats 26
times>"\200,
\064'\350Y\177\000\000\260\235K\347Y\177\000\000ȝK\347Y\177\000\000\000\000\000\000\001\000\000\000(\024@",
'\000' <repeats 85 times>"\200,
#\304\347Y\177\000\000p\t'\350Y\177\000\000\000\000\000\000\000\000\000\000\377\377\377\377",
'\000' <repeats 12 times>, "\005\000\000\000\000\000\000\000\030\235a", '\000'
<repeats 13 times>, "\020\327F\001\000\000\000\000\003\000\000\000\000\000"
fullname = "OmniKey CardMan 3121", '\000' <repeats 20 times>"\200,
#\304\347Y\177", '\000' <repeats 11 times>,
"n+z\022\302\027\213\a\000\000\000\000\000\000\000\060\333F\001\000\000\000\000\060\333F\001\000\000\000\000&\\A\000\000\000\000\000
\fG\001"
driver = 0x14752b8
sSerialNumber = <value optimized out>
sInterfaceName = 0x0
ret = <value optimized out>
bInterfaceNumber = 0
__FUNCTION__ = "HPAddDevice"
#3 0x000000000040d9bc in HPEstablishUSBNotifications (udev=0x1470c20) at
hotplug_libudev.c:551
dev = 0x146d710
parent = 0x146db30
action = <value optimized out>
devpath = 0x146de50 "OmniKey CardMan 3121"
udev_monitor = <value optimized out>
r = <value optimized out>
i = <value optimized out>
__FUNCTION__ = "HPEstablishUSBNotifications"
#4 0x00007f59e78238ba in start_thread (arg=<value optimized out>) at
pthread_create.c:300
__res = <value optimized out>
pd = 0x7f59e74ba700
unwind_buf = {cancel_jmp_buf = {{jmp_buf = {140024109311744,
486541036462382269, 140736994234784,
140024109311744, 140024123695168, 3, -544015359995437891,
-544014735935431491},
mask_was_saved = 0}}, priv = {pad = {0x0, 0x0, 0x0, 0x0}, data =
{prev = 0x0, cleanup = 0x0,
canceltype = 0}}}
not_first_call = <value optimized out>
freesize = <value optimized out>
__PRETTY_FUNCTION__ = "start_thread"
#5 0x00007f59e758a3cd in clone () at
../sysdeps/unix/sysv/linux/x86_64/clone.S:112
No locals.
#6 0x0000000000000000 in ?? ()
No symbol table info available.
> >So the patch seems to fix the segfault, but still the question remains: Why
> >is
> >pcscd not using its internal CCID driver if OpenCT is installed (but not even
> >running) ?
>
> pcscd has no _internal_ CCID driver. pcscd has no internal driver at all.
well I was meaning to say: "the CCID driver / ifd_handler that is shipping as
part of pcsc-lite"
So as a summary, I think there are still two separate, problems:
1) the fact that pcscd does not handle the case that devpath == NULL,
which can be worked around using your patch
2) the question why pcscd is trying to use _only_ the OpenCT ifd_handler,
even when openct is not started (which it detects!)
and maybe even a third one:
3) why does libudev sometimes not provide a devpath in the first place.
Regards,
Harald
--
- Harald Welte <lafo...@gnumonks.org> http://laforge.gnumonks.org/
============================================================================
"Privacy in residential applications is a desirable marketing option."
(ETSI EN 300 175-7 Ch. A6)
--
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
