This is the mail archive of the cygwin mailing list for the Cygwin project.

Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

gdb Not reporting line where crash occurred during segfault crash on XP machine

I'm running a Win XP machine with Cygwin 1.5.18-1. I run gdb as follows (either from the command line in a standard terminal or the xterm, or in emacs-X11):
gdb segfaulting-app.exe
run app-args

gdb returns:
Program recieved signal SIGSEGV, Segmentation fault.
0x004c4dcb in ?? ()

Now here's the even stranger (to me, anyway) part- if I then run the gdb command 'where' I get
#0 0x004c4dcb in ?? ()
#1 0x0022ed88 in ?? ()
#2 0x00402aa2 in appFunction (virtualAddress=something, accessType = 604276992) at sourceCodeFileA.c:427
#3 0x.... in ?? ()
#4 0x... in ?? () etc.

In other words in sourceCodeFileA.c at line 427, I call functioun that calls a long series of functions in sourceCodeFileB.c, but cygwin only reports the line of the higher level function call (the one in File A), not the specific function where the crash occurss in File B.
I looked back over the mailing list, and saw that in the past (~5 years ago) there was a lot of discussion about gdb problems, and even a year before that there was some mention of a similar problem to mine: that gdb doesn't report the line where the crash actually occurred. But I haven't seen recent mention of these problems, or a mention on of this as a known flaw, so I'm hoping/assuming gdb works for some people.

If this tells anyone anything, the 'step' function still works if I insert an appropriate breakpoint, but there's still >100 steps to step through with each crash to try to track where the segfault occurrs in different situations.

Thanks for any (polite) suggestions,
Rob Anderson
rfa03 at

Unsubscribe info:
Problem reports:

Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]