Ticket #38 (new defect)

Opened 8 years ago

Last modified 5 years ago

segfault when record with BUZ driver

Reported by: ga@… Owned by: cinelerra@…
Priority: high Milestone:
Component: Recording Version: 1.2.0
Severity: normal Keywords:
Cc:

Description

I get a segfault when i try to record from a Pinnacle DC10+ using the BUZ driver
(v4l does not work)

#0 0x409f19bf in memcpy () from /lib/libc.so.6
#1 0x0829d0e2 in VDeviceBUZInput::run ()
#2 0x402498a5 in Thread::entrypoint () from /usr/lib/libguicast.so.1
#3 0x406e4e51 in pthread_start_thread () from /lib/libpthread.so.0
#4 0x406e4ecf in pthread_start_thread_event () from /lib/libpthread.so.0
#5 0x40a5164a in clone () from /lib/libc.so.6

Change History

Changed 8 years ago by baver@…

  • status changed from new to assigned

Does this still exist in 1.2?

Changed 8 years ago by ga@…

yeah, unfortunately :(

Changed 8 years ago by herman@…

  • version changed from 1.1.9 to 1.2.0

Is Richard the right guy to fix this bug?

Changed 7 years ago by ga@…

More info about this bug:

==7565== Thread 43:
==7565== Invalid read of size 1
==7565== at 0x1B9018D0: memcpy (mac_replace_strmem.c:394)
==7565== by 0x8286450: VDeviceBUZInput::run() (vdevicebuz.C:113)
==7565== by 0x1BD4F053: Thread::entrypoint(void*) (thread.C:46)
==7565== by 0x1C23CCFC: start_thread (in /lib/tls/libpthread-2.3.5.so)
==7565== by 0x1C425F1D: clone (in /lib/tls/libc-2.3.5.so)
==7565== Address 0xFFFFFFFF is not stack'd, malloc'd or (recently) free'd

device->input_buffer == 0xFFFFFFFF

I don't remember since when I get this segfault. I guess since the 1.2.0 version
but I'm not sure.

Changed 5 years ago by baver@…

  • owner changed from baver@… to cinelerra@…
  • status changed from assigned to new
Note: See TracTickets for help on using tickets.