hashcat Forum

Hi so basically Im on windows 10 and Im using 4 GPUs AMD Radeon R9 200 / HD 7900 Series and one CPU Intel(R) Core(TM) i7-3930K CPU  @ 3.20GHz 6 Cores and i am having the following error when benchmarking:
PS C:\Users\Crack\Downloads\hashcat-5.1.0> .\hashcat64.exe -b
hashcat (v5.1.0) starting in benchmark mode...

Benchmarking uses hand-optimized kernel code by default.
You can use it in your cracking session by setting the -O option.
Note: Using optimized kernel code limits the maximum supported password length.
To disable the optimized kernel code in benchmark mode, use the -w option.

OpenCL Platform #1: Advanced Micro Devices, Inc.
================================================
* Device #1: Tahiti, 2393/3072 MB allocatable, 32MCU
* Device #2: Tahiti, 2393/3072 MB allocatable, 32MCU
* Device #3: Tahiti, 2393/3072 MB allocatable, 32MCU
* Device #4: Tahiti, 2393/3072 MB allocatable, 32MCU

Benchmark relevant options:
===========================
* --optimized-kernel-enable

Hashmode: 0 - MD5

Speed.#1.........:  7739.6 MH/s (67.69ms) @ Accel:128 Loops:512 Thr:256 Vec:1
Speed.#2.........:  7740.0 MH/s (67.69ms) @ Accel:128 Loops:512 Thr:256 Vec:1
Speed.#3.........:  7740.3 MH/s (67.69ms) @ Accel:128 Loops:512 Thr:256 Vec:1
Speed.#4.........:  7740.7 MH/s (67.69ms) @ Accel:128 Loops:512 Thr:256 Vec:1
Speed.#*.........: 30960.6 MH/s

Hashmode: 100 - SHA1

clCreateBuffer(): CL_MEM_OBJECT_ALLOCATION_FAILURE

Started: Thu Feb 20 13:14:06 2020
Stopped: Thu Feb 20 13:14:18 2020


I just updated the drivers so I dont know what is wrong.

try with the latest beta version https://hashcat.net/beta/

(02-20-2020, 06:08 PM)undeath Wrote: [ -> ]try with the latest beta version https://hashcat.net/beta/

I was facing the same issue and using the latest beta version fixed it for some hash functions, see below:
I have two R9 280X in Crossfire using the Adrenaline 20.2.2 driver on windows 10.

hashcat-5.1.0+1774\hashcat-5.1.0>hashcat.exe -b
hashcat (v5.1.0-1774-gf96594ef) starting in benchmark mode...

Benchmarking uses hand-optimized kernel code by default.
You can use it in your cracking session by setting the -O option.
Note: Using optimized kernel code limits the maximum supported password length.
To disable the optimized kernel code in benchmark mode, use the -w option.

OpenCL API (OpenCL 2.1 AMD-APP (3004.8)) - Platform #1 [Advanced Micro Devices, Inc.]
=====================================================================================
* Device #1: Tahiti, 3008/3072 MB (2393 MB allocatable), 32MCU
* Device #2: Tahiti, 3008/3072 MB (2393 MB allocatable), 32MCU

Benchmark relevant options:
===========================
* --optimized-kernel-enable

Hashmode: 0 - MD5

Speed.#1.........: 7409.1 MH/s (71.66ms) @ Accel:1024 Loops:256 Thr:64 Vec:1
Speed.#2.........: 8695.2 MH/s (60.91ms) @ Accel:1024 Loops:256 Thr:64 Vec:1
Speed.#*.........: 16104.3 MH/s

Hashmode: 100 - SHA1

Speed.#1.........: 2361.6 MH/s (56.00ms) @ Accel:512 Loops:128 Thr:64 Vec:1
Speed.#2.........: 2770.1 MH/s (47.61ms) @ Accel:512 Loops:128 Thr:64 Vec:1
Speed.#*.........: 5131.7 MH/s

Hashmode: 1400 - SHA2-256

Speed.#1.........: 1000.2 MH/s (66.44ms) @ Accel:256 Loops:128 Thr:64 Vec:1
Speed.#2.........: 1190.3 MH/s (55.72ms) @ Accel:256 Loops:128 Thr:64 Vec:1
Speed.#*.........: 2190.4 MH/s

Hashmode: 1700 - SHA2-512

Speed.#1.........: 328.9 MH/s (50.51ms) @ Accel:128 Loops:64 Thr:64 Vec:1
Speed.#2.........: 387.1 MH/s (42.85ms) @ Accel:128 Loops:64 Thr:64 Vec:1
Speed.#*.........: 716.0 MH/s

Hashmode: 22000 - WPA-PBKDF2-PMKID+EAPOL (Iterations: 4095)

clCreateBuffer(): CL_MEM_OBJECT_ALLOCATION_FAILURE

Started: Sat May 02 13:08:29 2020
Stopped: Sat May 02 13:09:18 2020

TLDR: I managed to get it working by manually setting loop and acceleration values with the -n and -u options like so:

hashcat.exe -b -m 16800 -n 800 -u 1024 --force

-m 16800 -> My objective is to crack PMKID hashes, so I focused my efforts there.
--force -> If you run without the --force, execution will halt with a warning suggesting to use workload profile (-w option). However, none of them worked for me so I use --force to ignore warnings.

I am not sure if 800 and 1024 values are the best for this, I am quite ignorant on what they do internally (links welcome) so I picked them by experimenting and watching if number go up.

Using the stable version did not work, I had to go use the beta version (hashcat-5.1.0+1774).

hashcat.exe -b -m 16800 -n 800 -u 1024 --force
hashcat (v5.1.0-1774-gf96594ef) starting in benchmark mode...

Benchmarking uses hand-optimized kernel code by default.
You can use it in your cracking session by setting the -O option.
Note: Using optimized kernel code limits the maximum supported password length.
To disable the optimized kernel code in benchmark mode, use the -w option.

You have enabled --force to bypass dangerous warnings and errors!
This can hide serious problems and should only be done when debugging.
Do not report hashcat issues encountered when using --force.
./OpenCL/m16800-optimized.cl: Optimized kernel requested but not needed - falling back to pure kernel
OpenCL API (OpenCL 2.1 AMD-APP (3004.8)) - Platform #1 [Advanced Micro Devices, Inc.]
=====================================================================================
* Device #1: Tahiti, 3008/3072 MB (2393 MB allocatable), 32MCU
* Device #2: Tahiti, 3008/3072 MB (2393 MB allocatable), 32MCU

Benchmark relevant options:
===========================
* --force
* --optimized-kernel-enable
* --kernel-accel=800

Hashmode: 16800 - WPA-PMKID-PBKDF2 (Iterations: 4095)

Speed.#1.........: 144.5 kH/s (2826.31ms) @ Accel:800 Loops:1024 Thr:64 Vec:1
Speed.#2.........: 145.0 kH/s (2801.45ms) @ Accel:800 Loops:1024 Thr:64 Vec:1
Speed.#*.........: 289.5 kH/s

Started: Sun May 03 18:39:56 2020
Stopped: Sun May 03 18:40:16 2020

Thanks.