troubles upgrading to v3.1

[martin.po21]

Hello everyone, I would like to ask for an advice...

I tried to run hashcat 3.1 on one of my older machines (proliant dl580 g5), but got " clCreateContext(): CL_UNKNOWN_ERROR" error as shown in output posted below.
Since I did not recall having problems with ocl runtime, i tried old v2.0 version of hashcat, and it runs as it is supposed to, as shown in output posted below.

Machine has OpenCL™ Runtime 16.1.1 for Intel® Core™ and Intel® Xeon® Processors installed.
I also tried with older 14.2 version, but error stays the same.

Is it possible that this error is not ocl runtime related, considering v 2.0 of hashcat running fine?
Is there something I am missing?

hashcat 3.1 output:

Code:

hashcat (v3.10) starting in benchmark-mode...

OpenCL Platform #1: Intel(R) Corporation
========================================
- Device #1: Intel(R) Xeon(R) CPU           E7450  @ 2.40GHz, 8191/32764 MB allocatable, 24MCU



ERROR: clCreateContext(): CL_UNKNOWN_ERROR

hashcat 2.0 output:

Code:

Initializing hashcat v2.00 with 24 threads and 32mb segment-size...

Device...........: Intel(R) Xeon(R) CPU           E7450  @ 2.40GHz
Instruction set..: x86_64
Number of threads: 24

Hash type: bcrypt, Blowfish(OpenBSD)
Speed/sec: 14.08k words

Thank you in advance

Martin

[kiara]

try to uninstall any amd/nvidia drivers ( u can use https://www.guru3d.com/files-details/disp...nload.html for that. or do it manual) , and uninstall any openCL .
afterwards, download the newst GPU driver and openCL , if the newst gpu driver did not work download the one before the latest

[martin.po21]

Thanks for reply.

I have tried changing gpu driver to older, and then to latest- the problem stays the same in both cases.
I would sey that GPU driver does not affect it, as even without it, hashcat 2.0 works and 3.1 does not- same error as mentioned above.

Perhaps I should mention that I am trying to get it running on machine CPUs, not GPU (old integrated ATI ES1000)
(I tought this would be visible from posted output)

Also, OS is windows serer 2008 R2, if this means anything

As a side note, Hashcat v3.0 returns a bit different error than v3.1:

Code:

ERROR: clCreateContext() : -2 : CL_UNKNOWN_ERROR

[atom]

Looks like your opencl runtime is broken. The OpenCL™ Runtime 16.1.1 for Intel® Core™ and Intel® Xeon® Processors installed is the correct one, but if you had a different opencl runtime installed before (eventually installed by your gpu driver) this may be incompatible to the Intel OpenCL runtime. Therefore you need to make sure to uninstall the old one first. This also involves deleting some files by hand. Here's a writeup: https://hashcat.net/wiki/doku.php?id=fre...hould_i_do

[martin.po21]

Thank you for reply atom. I'll post back my result when I mannage to perform what you advised, somewhere at end of this week.

Great product with great support!

[martin.po21]

Looks like your opencl runtime is broken. The OpenCL™ Runtime 16.1.1 for Intel® Core™ and Intel® Xeon® Processors installed is the correct one, but if you had a different opencl runtime installed before (eventually installed by your gpu driver) this may be incompatible to the Intel OpenCL runtime. Therefore you need to make sure to uninstall the old one first. This also involves deleting some files by hand. Here's a writeup: https://hashcat.net/wiki/doku.php?id=fre...hould_i_do

After performing this, problem stays the same.
I went even further and format the system, followed by installing only intel opencl runtime 16.1.1.
Once again pre v3 hashcat works and v3.X returns same errors as mentioned above.

I'll give linux a try and report back

[atom]

I've tried on a XEON box, just to be sure, here's a log:

[atom@super hashcat]$ ./hashcat -b -m 1000 -d 5 -D 1
hashcat (v3.10-141-gfd70fde) starting in benchmark-mode...

OpenCL Platform #1: Advanced Micro Devices, Inc.
================================================
- Device #1: Juniper, skipped
- Device #2: Tahiti, skipped
- Device #3: Tahiti, skipped
- Device #4: WARNING: Not a native Intel OpenCL runtime, expect massive speed loss
You can use --force to override this but do not post error reports if you do so
- Device #4: Intel(R) Xeon(R) CPU E5-2670 0 @ 2.60GHz, skipped

OpenCL Platform #2: Intel(R) Corporation
========================================
- Device #5: Intel(R) Xeon(R) CPU E5-2670 0 @ 2.60GHz, 32257/129028 MB allocatable, 32MCU
- Device #6: Intel(R) Many Integrated Core Acceleration Card, skipped

OpenCL Platform #3: NVIDIA Corporation
======================================
- Device #7: GeForce GTX TITAN X, skipped
- Device #8: GeForce GTX TITAN, skipped

Hashtype: NTLM

Speed.Dev.#5.: 1697.0 MH/s (19.71ms)

So, Xeon works fine.

[martin.po21]

FInnaly, I got it working on centos.
Not from binaries, but build went smooth, with little to no trouble.

Thank you again for support