On Wed, Nov 01, 2017 at 09:21:46AM +0100, dieter wrote: > >> It points to a memory corruption. > > The i386/x64 architecture supports memory access breakpoints > and GDB, too, has support for this. You know the address which > gets corrupted. Thus, the following apporach could have a chance > to succeed: > > Put a "memory write" breakpoint on the address which gets corrupted. > this should stop the program each time this address is written; > Check then the backtrace. As the address forms part of the > address block prologue, it should only be accessed from > Python's "malloc" (and "free") implementation. Any other access > indicates bad behaviour.
I understand. Thank you for the explanation. This may seem like a dumb question: the actual address that gets corrupted varies from run to run (it may be the same "place" in the code but that place gets put at a different address each run). Since I don't know the address of a given run, how do I set a break point on that address ? It seems to me I first need to map the "place" to its address before the SIGABRT happens. How do I find out out which "place" needs to be mapped from the output I already have ? (the "place" is the memory block you refer to) Other than that I understand what you are getting at and am willing to try. Thanks, Karsten > Should your program get stopped too often (because the memory > block is reused too often), you must find a good point in your > program to activate the memory access breakpoint in a delayed way. > You could use the Python debugger for this: execute significant > parts of your program in the Python debugger; switching to > GDB, check whether the address is already corrupted - if > so, restart and refine the checks above in the portion of your program > which caused the corruption. If the part in your program > is sufficiently small, you can activate the memory access breakpoint. > This approach may also be feasible, should you use a CPU > without memory access breakpoints. -- GPG key ID E4071346 @ eu.pool.sks-keyservers.net E167 67FD A291 2BEA 73BD 4537 78B9 A9F9 E407 1346 -- https://mail.python.org/mailman/listinfo/python-list
