[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Date: Sun, 26 Jul 2015 12:23:53 +0200
From: MPB <pbmajorpb@...glemail.com>
To: john-users@...ts.openwall.com
Subject: Re: OpenCL with AMD FirePro on OS X
With --list=opencl-devices I get
**
Platform #0 name: Apple
Platform version: OpenCL 1.2 (Dec 14 2014 22:29:47)
Device #0 (0) name: Intel(R) Xeon(R) CPU E5-1620 v2 @ 3.70GHz
Device vendor: Intel
Device type: CPU (LE)
Device version: OpenCL 1.2
Driver version: 1.1
Native vector widths: char 16, short 8, int 4, long 2
Preferred vector width: char 16, short 8, int 4, long 2
Global Memory: 16.0 GB
Global Memory Cache: 64.0 B
Local Memory: 32.0 KB (Global)
Max memory alloc. size: 4.0 GB
Max clock (MHz): 4294970996
Profiling timer res.: 1 ns
Max Work Group Size: 1024
Parallel compute cores: 8
Device #1 (1) name: AMD Radeon HD - FirePro D300 Compute Engine
Device vendor: AMD
Device type: GPU (LE)
Device version: OpenCL 1.2
Driver version: 1.2 (Dec 23 2014 00:17:17)
Native vector widths: char 4, short 2, int 1, long 1
Preferred vector width: char 4, short 2, int 1, long 1
Global Memory: 2.0 GB
Local Memory: 32.0 KB (Local)
Max memory alloc. size: 512.5 MB
Max clock (MHz): 850
Profiling timer res.: 37 ns
Max Work Group Size: 256
Parallel compute cores: 20
Stream processors: 1280 (20 x 64)
Device #2 (2) name: AMD Radeon HD - FirePro D300 Compute Engine
Device vendor: AMD
Device type: GPU (LE)
Device version: OpenCL 1.2
Driver version: 1.2 (Dec 23 2014 00:17:17)
Native vector widths: char 4, short 2, int 1, long 1
Preferred vector width: char 4, short 2, int 1, long 1
Global Memory: 2.0 GB
Local Memory: 32.0 KB (Local)
Max memory alloc. size: 512.5 MB
Max clock (MHz): 150
Profiling timer res.: 37 ns
Max Work Group Size: 256
Parallel compute cores: 20
Stream processors: 1280 (20 x 64)
**
I tried using --device=1 and --device=2. Both fail with the same FAILED (cmp_all(1)) status. By the way: Not directly related to my main question but: Can I use both FirePro GPUs in order to double the performance?
> Am 26.07.2015 um 12:12 schrieb magnum <john.magnum@...hmail.com>:
>
> On 2015-07-25 18:06, Philip Bleicker wrote:
>> I tried to build John 1.8.0 Jumbo 1 with configure and build (with OpenMPI enabled). Every CPU based approach does work. Every OpenCL algorithm fails. Let's take this example:
>>
>> ./john --test --format=agilekeychain-opencl
>>
>> It gives me back:
>>
>> Device 1: AMD Radeon HD - FirePro D300 Compute Engine Local worksize (LWS) 64, global worksize (GWS) 524288 Benchmarking: agilekeychain-opencl, 1Password Agile Keychain [PBKDF2-SHA1 OpenCL AES]... FAILED (cmp_all(1))
>>
>> I tried building JTR with the legacy Makefile but it doesn't work neither.
>>
>> Did I overlook something? The hardware is a Mac Pro with two FirePro D300 running OS X 10.10.2.
>
> On my MBPr running Yosemite, iirc all Jumbo-1 OpenCL formats pass using any of CPU, HD4000 and nvidia GT650M.
>
> If really *all* OpenCL formats fail, there's likely some major problem with the driver or the build. I'm not sure what could cause it. What if you try --device=0 (CPU) or --device=3 (if that's some Intel graphics engine)?
>
> Please post the output of --list=opencl-devices
>
> magnum
>
Powered by blists - more mailing lists
Confused about mailing lists and their use?
Read about mailing lists on Wikipedia
and check out these
guidelines on proper formatting of your messages.
