------- You are receiving this mail because: -------
You are on the CC list for the bug.
http://bugs.exim.org/show_bug.cgi?id=1615
--- Comment #4 from Greg Hazel <ghazel@???> 2015-04-17 00:49:58 ---
I do not have a small C repro. Trying to isolate it has proven difficult.
Using malloc instead changes the Valgrind error:
==47830== Invalid read of size 1
==47830== at 0x41EC25C: ???
==47830== by 0x6E6EFF: ??? (in /path/to/myapp)
==47830== by 0x41AF09A: ???
==47830== Address 0x535cf9 is not stack'd, malloc'd or (recently) free'd
==47830==
==47830==
==47830== Process terminating with default action of signal 11 (SIGSEGV):
dumping core
==47830== Access not within mapped region at address 0x535CF9
==47830== at 0x41EC25C: ???
==47830== by 0x6E6EFF: ??? (in /path/to/myapp)
==47830== by 0x41AF09A: ???
Using PCRE 8.37-RC1 (with the stack allocation) triggers similar Valgrind
errors:
==41034== Conditional jump or move depends on uninitialised value(s)
==41034== at 0x41EC170: ???
==41034== by 0xFFEFFF2BF: ???
==41034== by 0xFFEFFF2BF: ???
==41034== by 0xFFEFFF2D1: ???
==41034== by 0xFFEFFF21F: ???
==41034== by 0xA426145F: ???
==41034== Uninitialised value was created by a stack allocation
==41034== at 0x40FC03: myfunc (myfile.c:64)
--
Configure bugmail:
http://bugs.exim.org/userprefs.cgi?tab=email
