ERROR: cuStreamSynchronize() 700
#1
Hello.

After I installed the .09, I attempted to crack a hash (1800 = sha512crypt, SHA512(Unix)) and received a cuStreamSynchronize() 700 error as shown below. I have updated my Nvidia drivers as suggested in other threads but I am still seeing the issue. It should be noted that before I updated the drivers, I was having problems cracking type 1800 on the .09 release no matter the wordlist size that I used...but then after I updated the drivers I was able to run it against smaller wordlists lists without any problem but still having problems (as shown below) with large wordlists.

Any help would be much appreciated.

NVRM version: NVIDIA UNIX x86_64 Kernel Module 304.43

Large wordlist - Error
Hashes: 1 total, 1 unique salts, 1 unique digests
Bitmaps: 8 bits, 256 entries, 0x000000ff mask, 1024 bytes
Rules: 1
Workload: 16 loops, 2 accel (I played around with the accels and it made no difference)
Watchdog: Temperature abort trigger set to 90c
Watchdog: Temperature retain trigger set to 80c
Device #1: GeForce GTX 580, 1535MB, 1564Mhz, 16MCU
Device #2: GeForce GTX 580, 1535MB, 1564Mhz, 16MCU
Device #3: GeForce GTX 580, 1535MB, 1564Mhz, 16MCU
Device #4: GeForce GTX 580, 1535MB, 1564Mhz, 16MCU
Device #5: GeForce GTX 580, 1535MB, 1564Mhz, 16MCU
Device #6: GeForce GTX 580, 1535MB, 1564Mhz, 16MCU
Device #7: GeForce GTX 580, 1535MB, 1564Mhz, 16MCU
Device #8: GeForce GTX 580, 1535MB, 1564Mhz, 16MCU
Device #1: Kernel ./kernels/4318/m1800.sm_20.ptx
Device #2: Kernel ./kernels/4318/m1800.sm_20.ptx
Device #3: Kernel ./kernels/4318/m1800.sm_20.ptx
Device #4: Kernel ./kernels/4318/m1800.sm_20.ptx
Device #5: Kernel ./kernels/4318/m1800.sm_20.ptx
Device #6: Kernel ./kernels/4318/m1800.sm_20.ptx
Device #7: Kernel ./kernels/4318/m1800.sm_20.ptx
Device #8: Kernel ./kernels/4318/m1800.sm_20.ptx

Scanned dictionary /media/Data/wordlists/rockyou.txt: 139921497 bytes, 14344391 words, 14344390 keyspace, starting attack...

[s]tatus [p]ause [r]esume [b]ypass [q]uit => ERROR: cuStreamSynchronize() 700


Small wordlist - No Errors
Hashes: 1 total, 1 unique salts, 1 unique digests
Bitmaps: 8 bits, 256 entries, 0x000000ff mask, 1024 bytes
Rules: 1
Workload: 16 loops, 2 accel
Watchdog: Temperature abort trigger set to 90c
Watchdog: Temperature retain trigger set to 80c
Device #1: GeForce GTX 580, 1535MB, 1564Mhz, 16MCU
Device #2: GeForce GTX 580, 1535MB, 1564Mhz, 16MCU
Device #3: GeForce GTX 580, 1535MB, 1564Mhz, 16MCU
Device #4: GeForce GTX 580, 1535MB, 1564Mhz, 16MCU
Device #5: GeForce GTX 580, 1535MB, 1564Mhz, 16MCU
Device #6: GeForce GTX 580, 1535MB, 1564Mhz, 16MCU
Device #7: GeForce GTX 580, 1535MB, 1564Mhz, 16MCU
Device #8: GeForce GTX 580, 1535MB, 1564Mhz, 16MCU
Device #1: Kernel ./kernels/4318/m1800.sm_20.ptx
Device #2: Kernel ./kernels/4318/m1800.sm_20.ptx
Device #3: Kernel ./kernels/4318/m1800.sm_20.ptx
Device #4: Kernel ./kernels/4318/m1800.sm_20.ptx
Device #5: Kernel ./kernels/4318/m1800.sm_20.ptx
Device #6: Kernel ./kernels/4318/m1800.sm_20.ptx
Device #7: Kernel ./kernels/4318/m1800.sm_20.ptx
Device #8: Kernel ./kernels/4318/m1800.sm_20.ptx

Scanned dictionary /media/Data/wordlists/all.txt: 327914 bytes, 32761 words, 32761 keyspace, starting attack...

Status.......: Exhausted
Input.Mode...: File (/media/Data/wordlists/all.txt)
Hash.Type....: sha512crypt, SHA512(Unix)
Time.Running.: 10 secs
Time.Left....: 0 secs
Time.Util....: 10045.6ms/0.0ms Real/CPU, 0.0% idle
Speed........: 3261 c/s Real, 262 c/s GPU
Recovered....: 0/1 Digests, 0/1 Salts
Progress.....: 32761/32761 (100.00%)
Rejected.....: 0/32761 (0.00%)
HWMon.GPU.#1.: -1% Util, 43c Temp, 40% Fan
HWMon.GPU.#2.: -1% Util, 49c Temp, 42% Fan
HWMon.GPU.#3.: -1% Util, 42c Temp, 40% Fan
HWMon.GPU.#4.: -1% Util, 43c Temp, 40% Fan
HWMon.GPU.#5.: -1% Util, 43c Temp, 40% Fan
HWMon.GPU.#6.: -1% Util, 48c Temp, 42% Fan
HWMon.GPU.#7.: -1% Util, 39c Temp, 40% Fan
HWMon.GPU.#8.: -1% Util, 42c Temp, 40% Fan

Started: Sat Sep 8 01:52:41 2012
Stopped: Sat Sep 8 01:53:02 2012
#2
I had this as well on a 580 and it turned out to be a bad card. Since I stopped using that card the 700 errors have gone away.
#3
@radix - When you experienced your bad card problem was your 700 error consistent? What I mean by that is did you always see that error no matter the wordlist size and no matter the hash?

I just re-ran my example above but with the hash type of NTLM instead of sha512crypt, SHA512(Unix) using both a big and small wordlists and I received no errors.
#4
No it was only on large lists, or highly iterated algos.
#5
Thanks for the suggestion

I went through each of my 8 cards by using the "-d" option and found out the following...

3 of my 8 cards throw an error when used with that specific hash type.

Device 1 - No Error
Device 2 - No Error
Device 3 - No Error
Device 4 - Error
Device 5 - Error
Device 6 - No Error
Device 7 - Error
Device 8 - No Error

Thanks again for your help
#6
It could be the pcie slot as well. I would double check this by swapping cards around.
#7
I have same error.
Windows Xp 32 bit, GeForce GTX 560 Ti, nvidia driver v306.02
Earlier driver version 304.xx also don't work. -> ERROR: cuStreamSynchronize() 700
Even on gpu-accel=1 | single hash | simple alg - md4,md5 | different cores -a 0,3,7 |
small wordlists - 50 lines, and all this together
Any ideas how to overcome this? Can cuda-memcheck be used to find the problem?
#8
I am using a 560Ti for development, too, and its not happening. You can try -n 1 --gpu-loops 1 and then increase till it occours.
#9
I tried just now:
d:\hashcat>cudaHashcat-plus32.exe -n 1 --gpu-loops 1 example0.hash example.dict
cudaHashcat-plus v0.09 by atom starting...

Hashes: 1 total, 1 unique salts, 1 unique digests
Bitmaps: 8 bits, 256 entries, 0x000000ff mask, 1024 bytes
Rules: 1
Workload: 1 loops, 1 accel
Watchdog: Temperature abort trigger set to 90c
Watchdog: Temperature retain trigger set to 80c
Device #1: GeForce GTX 560 Ti, 1023MB, 1800Mhz, 8MCU
Device #1: Kernel ./kernels/4318/m0000_a0.sm_21.ptx

Scanned dictionary example.dict: 90 bytes, 10 words, 10 keyspace, starting attack...

[s]tatus [p]ause [r]esume [b]ypass [q]uit => ERROR: cuStreamSynchronize() 700

I will try to update to windows 7 x64.
Thank you for this awesome release!