cudaHashcat-lite32.bin v0.10, sha1($pass.$salt): ERROR: cuLaunchKernel() 701
#1
Thanks again for hashcat. I believe I found a bug in cudaHashcat-lite32.bin v0.10

The following command results in "ERROR: cuLaunchKernel() 701"
Code:
$cudaHashcat-lite32.bin $(head -c40 ~/Desktop/password.key):9ba76d19acb737e9 -m110  --pw-min=6 --pw-max=6 -1 ?l ?1?1?1?1?1?1

The same command works in cudaHashcat-lite32.bin v0.09

A similiar attack on the same pwd/salt in v0.10, this time with an md5 pwd succeeds:
Code:
$cudaHashcat-lite32.bin $(tail -c32 ~/Desktop/password.key):9ba76d19acb737e9 -m10  --pw-min=6 --pw-max=6 -1 ?l ?1?1?1?1?1?1

Interestingly, the md5 attack errs with "ERROR: to much combinations possible":
Code:
$ cudaHashcat-lite32.bin $(tail -c32 ~/Desktop/password.key):9ba76d19acb737e9 -m10  --pw-min=6 --pw-max=6 -1 ?a ?1?1?1?1?1?1

But this worked in v0.09 with:
Code:
$ cudaHashcat-lite32.bin $(tail -c32 ~/Desktop/password.key):9ba76d19acb737e9 -m10  --pw-min=6 --pw-max=6 -1 ?l?u?s?d ?1?1?1?1?1?1

Details:
Password lengths range: 6 - 6
Watchdog: Temperature abort trigger set to 90c
Watchdog: Temperature retain trigger set to 80c
Device #1: GeForce 9500 GT, 1023MB, 1375Mhz, 4MCU

Thanks
#2
ERROR: cuLaunchKernel() 701 is usually a driver problem. I found out it might be also related to a framebuffer running. You can test that by logging in from remote via SSH and try again. Otherwise its a driver problem.
#3
(10-26-2012, 10:12 AM)atom Wrote: ERROR: cuLaunchKernel() 701 is usually a driver problem. I found out it might be also related to a framebuffer running. You can test that by logging in from remote via SSH and try again. Otherwise its a driver problem.

That may be, but the same driver that works in the previous version of lite fails in the current version. I believe that makes it a bug in the program, not a driver.