What am I missing?

[pragmatic]

So I'm running some bruteforce attacks with a system using 8 x 7970s and I can't seem to get my utilization up to full speed.

For instance:

$ ./oclHashcat-plus64.bin -a 3 -m 1000 -n 160 --gpu-loops=1024 --remove tobecracked ?a?a?a?a?a?a?a?a
oclHashcat-plus v0.09 by atom starting...

Hashes: 4 total, 1 unique salts, 4 unique digests
Bitmaps: 8 bits, 256 entries, 0x000000ff mask, 1024 bytes
Workload: 1024 loops, 160 accel
Watchdog: Temperature abort trigger set to 90c
Watchdog: Temperature retain trigger set to 80c
Device #1: Tahiti, 2928MB, 1000Mhz, 32MCU
Device #2: Tahiti, 2933MB, 1000Mhz, 32MCU
Device #3: Tahiti, 2933MB, 1000Mhz, 32MCU
Device #4: Tahiti, 2933MB, 1000Mhz, 32MCU
Device #5: Tahiti, 2933MB, 1000Mhz, 32MCU
Device #6: Tahiti, 2933MB, 1000Mhz, 32MCU
Device #7: Tahiti, 2933MB, 1000Mhz, 32MCU
Device #8: Tahiti, 2933MB, 1000Mhz, 32MCU
Device #1: Kernel ./kernels/4098/m1000_a3.Tahiti_1016.4_(VM).kernel (348640 bytes)
Device #2: Kernel ./kernels/4098/m1000_a3.Tahiti_1016.4_(VM).kernel (348640 bytes)
Device #3: Kernel ./kernels/4098/m1000_a3.Tahiti_1016.4_(VM).kernel (348640 bytes)
Device #4: Kernel ./kernels/4098/m1000_a3.Tahiti_1016.4_(VM).kernel (348640 bytes)
Device #5: Kernel ./kernels/4098/m1000_a3.Tahiti_1016.4_(VM).kernel (348640 bytes)
Device #6: Kernel ./kernels/4098/m1000_a3.Tahiti_1016.4_(VM).kernel (348640 bytes)
Device #7: Kernel ./kernels/4098/m1000_a3.Tahiti_1016.4_(VM).kernel (348640 bytes)
Device #8: Kernel ./kernels/4098/m1000_a3.Tahiti_1016.4_(VM).kernel (348640 bytes)

[s]tatus [p]ause [r]esume [b]ypass [q]uit => s
Status.......: Running
Input.Mode...: Mask (?a?a?a?a?a?a?a?a)
Hash.Target..: File (tobecracked)
Hash.Type....: NTLM
Time.Running.: 55 secs
Time.Left....: 1 day, 8 hours
Time.Util....: 55394.6ms/355.5ms Real/CPU, 0.6% idle
Speed........: 57243.4M c/s Real, 58087.3M c/s GPU
Recovered....: 0/4 Digests, 0/1 Salts
Progress.....: 3170975088640/6634204312890625 (0.05%)
Rejected.....: 0/3170975088640 (0.00%)
HWMon.GPU.#1.: 77% Util, 66c Temp, 31% Fan
HWMon.GPU.#2.: 77% Util, 64c Temp, 30% Fan
HWMon.GPU.#3.: 77% Util, 67c Temp, 33% Fan
HWMon.GPU.#4.: 77% Util, 64c Temp, 30% Fan
HWMon.GPU.#5.: 77% Util, 64c Temp, 30% Fan
HWMon.GPU.#6.: 77% Util, 65c Temp, 30% Fan
HWMon.GPU.#7.: 77% Util, 59c Temp, 24% Fan
HWMon.GPU.#8.: 77% Util, 65c Temp, 30% Fan

Using -lite gets them all going 99% just fine, but for plus it seems to ceiling out around 77-82%.

Any thoughts/suggestions?

[epixoip]

looks like you're using the wrong driver. need to use catalyst 12.8. if you're going to use catalyst 12.10, then you need to stop x11 and run hashcat as root.

[pragmatic]

looks like you're using the wrong driver. need to use catalyst 12.8. if you're going to use catalyst 12.10, then you need to stop x11 and run hashcat as root.

Thanks, I'll try it!

[pragmatic]

So I'm trying to run 12.10 without x11 as root but now I'm getting the dreaded "ERROR: clGetDeviceIDs() -1"

# ./oclHashcat-lite64.bin -n 800 --gpu-loops 1024 -1 ?l?d?s?u --gpu-temp-disable --outfile-watch 0 --restore-timer 0 --runtime 10 --pw-min 8 -m 0 ffffffffffffffffffffffffffffffff ?1?1?1?1?1?1?1?1?1
ERROR: clGetDeviceIDs() -1

aticonfig --lsa correctly shows all of my cards.

Any suggestions?

[epixoip]

Is fglrx loaded? What does clinfo show?

[pragmatic]

# dmesg | grep module.*fglrx
[fglrx] module loaded - fglrx 9.0.2 [Sep 28 2012] with 8 minors

# clinfo | egrep 'Name|Type|e ID'
Platform Name: AMD Accelerated Parallel Processing
Platform Name: AMD Accelerated Parallel Processing
Device Type: CL_DEVICE_TYPE_GPU
Device ID: 4098
Name: Tahiti
Device Type: CL_DEVICE_TYPE_GPU
Device ID: 4098
Name: Tahiti
Device Type: CL_DEVICE_TYPE_GPU
Device ID: 4098
Name: Tahiti
Device Type: CL_DEVICE_TYPE_GPU
Device ID: 4098
Name: Tahiti
Device Type: CL_DEVICE_TYPE_GPU
Device ID: 4098
Name: Tahiti
Device Type: CL_DEVICE_TYPE_GPU
Device ID: 4098
Name: Tahiti
Device Type: CL_DEVICE_TYPE_GPU
Device ID: 4098
Name: Tahiti
Device Type: CL_DEVICE_TYPE_GPU
Device ID: 4098
Name: Tahiti
Device Type: CL_DEVICE_TYPE_CPU
Device ID: 4098
Name: Intel(R) Xeon(R) CPU X5650 @ 2.67GHz

[epixoip]

grepping through dmesg actually only proves that it was loaded at one time since last reboot, it doesn't show that it's still loaded. for future reference, ''grep fglrx /proc/modules'' or ''lsmod | grep fglrx'' would have been preferred. but since clinfo is showing your devices, it looks like it is in fact loaded.

and oclhashcat still won't run? sounds like you may have more than one libOpenCL installed. do you by chance have vcl installed on the system as well?

[pragmatic]

Sorry, I haven't been able to work on this for a while so I couldn't get more information until now.

Just for the sake of completeness:

# grep fglrx /proc/modules
fglrx 4740767 0 - Live 0xffffffffa023a000 (P)

And I don't have vcl installed.

Any help would be greatly appreciated, and can now be acted on quickly.

Other info:

# ls /etc/OpenCL/vendors/
amdocl32.icd amdocl64.icd

# cat /etc/OpenCL/vendors/amdocl64.icd
libamdocl64.so

# ls -l /usr/lib64/libamd*
-rw-r--r--. 1 root root 38937110 Oct 23 20:38 /usr/lib64/libamdocl64.so

# runlevel
N 3

# ./oclHashcat-lite64.bin -n 800 --gpu-loops 1024 --outfile-watch 0 --restore-timer 0 --runtime 14400 --pw-min 8 -m 0 ffffffffffffffffffffffffffffffff ?a?a?a?a?a?a?a?a?a?a
ERROR: clGetDeviceIDs() -1

[atom]

Please do not use any other version than 12.8. Everything above is more or less broken.

[pragmatic]

Will do. I'm excited about getting away from X but obviously having it work is more important. Thanks!