[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Date: Sun, 24 May 2015 20:20:16 +0200
From: Agnieszka Bielec <bielecagnieszka8@...il.com>
To: john-dev@...ts.openwall.com
Subject: Re: PHC: Parallel in OpenCL
2015-05-24 4:57 GMT+02:00 Solar Designer <solar@...nwall.com>:
> On Sat, May 23, 2015 at 01:50:57PM +0200, Agnieszka Bielec wrote:
>> I was trying to optimize sha512block in parallel but when i was
>> copying this functions from various Claudio Andre's source codes the
>> result wasn't better, was one case when the speed was greater on super
>> for --dev=5 from 35k to 40k and on my laptop from 26k to 28k but on
>> --dev=1 the speed decreased from 40k to 20k or 10k, also I was
>> modifying slightly two versions of Claudio Andre's sha512 but the
>> speed didn't increase
>
> Please note that Claudio has a separate OpenCL kernel for GCN, such as
> --dev=1 on super:
I increased the speed slightly:
[a@...er run]$ ./john --test --format=parallel-opencl --dev=1 --cost=0:0
Device 1: Tahiti [AMD Radeon HD 7900 Series]
Building the kernel, this could take a while
Build log: LOOP UNROLL: pragma unroll (line 106)
Unrolled as requested!
LOOP UNROLL: pragma unroll (line 87)
Unrolled as requested!
Benchmarking: parallel-opencl, parallel SHA-512 [ ]... DONE
Speed for cost 1 (N) of 0
Many salts: 45936 c/s real, 4915K c/s virtual
Only one salt: 46369 c/s real, 4915K c/s virtual
[a@...er run]$ ./john --test --format=parallel-opencl --dev=5 --cost=0:0
Device 5: GeForce GTX TITAN
Benchmarking: parallel-opencl, parallel SHA-512 [ ]... DONE
Speed for cost 1 (N) of 0
Many salts: 38102 c/s real, 38325 c/s virtual
Only one salt: 38325 c/s real, 38325 c/s virtual
960m:
Local worksize (LWS) 64, global worksize (GWS) 16384
Benchmarking: parallel-opencl, parallel SHA-512 [ ]... DONE
Speed for cost 1 (N) of 0
Many salts: 28248 c/s real, 28248 c/s virtual
Only one salt: 28248 c/s real, 28248 c/s virtual
previously
dev=1
Speed for cost 1 (N) of 0
Many salts: 41478 c/s real, 6553K c/s virtual
Only one salt: 41217 c/s real, 6553K c/s virtual
dev=5
DONE
Speed for cost 1 (N) of 0
Many salts: 35617 c/s real, 35812 c/s virtual
Only one salt: 35812 c/s real, 35812 c/s virtual
960m:
26k
> I guess you took code from cryptsha512_kernel_GPU.cl?
cryptsha512_kernel_GPU.cl and sha512_kernel.cl which has the same
speed on --dev=1 and --dev=5
btw. https://github.com/Sc00bz/Parallel is incomplete so I'm not
implementing optimizations on CPU so far
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.
