Follow-up Comment #7, bug #68223 (group findutils):

Line 26619 of raf's tracelooks very suspicious:


fstatat64(0x0, 0x0, 0x0)                 = 0 0


I wonder if p->fts_accpath is becoming NULL somehow, though it is even more
strange to see the third argument being NULL.  There are a lot of calls to
fstatat() in gnulib.   Hard to guess by inspection, and I have no Mac OS
machine (if somebody would like to send one of course, I would be glad to
investigate).  

raf, could you run gfind under a debugger and set a conditional breakpoint on
fstatat64() which triggers when the second argument is NULL OR the third
argument is NULL, then post a backtrace?

If for any reason that doesn't seem to shed any light, and your debugger has
hardware watchpoints, could you use a watchpoint to see when the FTS_STOP bit
in fts_info gets set, and post a backtrace for that?

Not sure if your debugger's backtrace will show function arguments, but of
course those would be useful, too.



    _______________________________________________________

Reply to this item at:

  <https://savannah.gnu.org/bugs/?68223>

_______________________________________________
Message sent via Savannah
https://savannah.gnu.org/

Attachment: signature.asc
Description: PGP signature

Reply via email to