Follow @Openwall on Twitter for new release announcements and other news
[<prev] [next>] [thread-next>] [day] [month] [year] [list]
Date: Sun, 12 Aug 2012 00:41:11 +0800
From: myrice <>
Subject: Separate bitmaps report


I am debugging 4 bitmaps for raw-md5-opencl. It works on my laptop but
caused memory corruption on bull. The report is as follows:

[19:53:18 myrice] run $ rm john.pot -f ; ./john -fo=raw-md5-opencl
-wo=md5.dic md5.hash -pla=0
OpenCL platform 0: NVIDIA CUDA, 1 device(s).
Using device 0: GeForce GTX 570
Local work size (LWS) 128, Global work size (GWS) 65536
Loaded 1 password hash (Raw MD5 [OpenCL])
end self test
*** glibc detected *** ./john: malloc(): memory corruption:
0x0000000009b27e20 ***

It happened either at the end of run or stuck at copying bitmaps from
CPU to GPU. Even I only changed previous defined cl_uint* bitmaps; to
cl_uint* bitmaps[1]; and changed every correspond place, the error
still exists. This code is in attachment.

I don't very understand this. Maybe I misuse array pointer in my code?


Download attachment "john-PG-test-opencl-rawmd5-bitmaps.diff" of type "application/octet-stream" (2636 bytes)

Powered by blists - more mailing lists

Confused about mailing lists and their use? Read about mailing lists on Wikipedia and check out these guidelines on proper formatting of your messages.