MESSAGE
| DATE | 2016-10-31 |
| FROM | Ruben Safir
|
| SUBJECT | Re: [Learn] cuda kernels
|
From learn-bounces-at-nylxs.com Mon Oct 31 18:06:39 2016
Return-Path:
X-Original-To: archive-at-mrbrklyn.com
Delivered-To: archive-at-mrbrklyn.com
Received: from www.mrbrklyn.com (www.mrbrklyn.com [96.57.23.82])
by mrbrklyn.com (Postfix) with ESMTP id 382BB161312;
Mon, 31 Oct 2016 18:06:39 -0400 (EDT)
X-Original-To: learn-at-nylxs.com
Delivered-To: learn-at-nylxs.com
Received: from mail-qk0-f179.google.com (mail-qk0-f179.google.com
[209.85.220.179]) by mrbrklyn.com (Postfix) with ESMTP id B30C1160E77
for ; Mon, 31 Oct 2016 18:06:35 -0400 (EDT)
Received: by mail-qk0-f179.google.com with SMTP id o68so180059457qkf.3
for ; Mon, 31 Oct 2016 15:06:34 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=my.liu.edu; s=google;
h=subject:to:references:from:message-id:date:user-agent:mime-version
:in-reply-to:content-transfer-encoding;
bh=DNWNOSTEJqsjWIEpntU0M2aAVJNU/7IB/BDjpfXBYaQ=;
b=cxTPOXOzhKrnOz385S/LY1ybOpeGLQ5QAei4phlNSH42wWo1xRC00Bks3HYz0HlRAt
nCpO+VY3GNa4GowYpkRzeqv4Z2oZu4UVdAfp3G1mM6RS/zORcQDJJNDenFJMH4ysevBi
lzjfj0C4/lQK32vQ/lYmcB9ozRDvLbg2yw6tc=
X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed;
d=1e100.net; s=20130820;
h=x-gm-message-state:subject:to:references:from:message-id:date
:user-agent:mime-version:in-reply-to:content-transfer-encoding;
bh=DNWNOSTEJqsjWIEpntU0M2aAVJNU/7IB/BDjpfXBYaQ=;
b=Bgcd3UevaR8C+4+Kcyxu+h1DP/dDXVXIV+QXJgC7vOkCYTnqx7I+tVj8AjnnNmXqJ2
ZjyPKiGVYjyI0lWXzR/7kXiGZZoFR3E2zPhkXiFT3V6Xru23/Ym8kXZwcMZQYzBnwz4R
QoYuy5xzODd6l9xcHQELtnUceojn9khIH49ZF0rkVt0wpzm6wgZZ/k9/BlWH23gFozYY
BFQJ478nJVzoj4t05KeN11HuOlYBisgmmLdKD1pvdvWJSM2Mp/nd/D2/PxcfkyWmQcBt
jczb5MjMD1H6TWZ+jU+mZr1+dYHQyyxF5Q/QSYTwfrv3HeoDoPPY2QWx1swMTpCwZ7n4
iScg==
X-Gm-Message-State: ABUngve21UUoANnj+nCgwcSYD2TzpZ0Qouh+rLt+6NKom2pwwou1cU+2MupU1ug6IbjYIzKy
X-Received: by 10.55.27.226 with SMTP id m95mr29445042qkh.143.1477951254214;
Mon, 31 Oct 2016 15:00:54 -0700 (PDT)
Received: from [10.0.0.62] (www.mrbrklyn.com. [96.57.23.82])
by smtp.googlemail.com with ESMTPSA id j2sm14201094qtb.46.2016.10.31.15.00.53
for
(version=TLS1_2 cipher=ECDHE-RSA-AES128-GCM-SHA256 bits=128/128);
Mon, 31 Oct 2016 15:00:53 -0700 (PDT)
To: learn-at-nylxs.com
References: <9570d66b-7c02-4e2d-3225-20e008158f0d-at-mrbrklyn.com>
<87lgx48apw.fsf-at-contrapunctus.net>
From: Ruben Safir
Message-ID: <9195064a-00e1-19a4-cc27-d57f4cf82220-at-my.liu.edu>
Date: Mon, 31 Oct 2016 18:00:53 -0400
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:45.0) Gecko/20100101
Thunderbird/45.4.0
MIME-Version: 1.0
In-Reply-To: <87lgx48apw.fsf-at-contrapunctus.net>
Subject: Re: [Learn] cuda kernels
X-BeenThere: learn-at-nylxs.com
X-Mailman-Version: 2.1.17
Precedence: list
List-Id:
List-Unsubscribe: ,
List-Archive:
List-Post:
List-Help:
List-Subscribe: ,
Content-Type: text/plain; charset="windows-1252"
Content-Transfer-Encoding: quoted-printable
Errors-To: learn-bounces-at-nylxs.com
Sender: "Learn"
On 10/31/2016 05:33 PM, Christopher League wrote:
> Ruben Safir ruben-at-mrbrklyn.com writes:
> =
> The video says that the below lines ask for 64 copys of the kernel
> on 64 threads? I don=92t see that
> =
> |cudaMemcpy(d_in, h_in, ARRAY_BYTES, cudaMemcpyHostToDevice);
> //launch kernel square<<<1, ARRAY_SIZE>>>(d_in, d_out);
> cudaMemcpy(h_out, d_out, ARRAY_BYTES, cudaMemcpyDeviceToHost);|
> =
> I don=92t really know anything about CUDA yet, but ARRAY_SIZE was 64. And
> this syntax: |square<<<1, ARRAY_SIZE>>>| is NOT valid C/C++ syntax. So I
> assume that the specialized GPU compiler interprets that in a way that
> it distributes the computation of the |square| function across 64 GPU
> threads=85?
> =
> BTW, I=92m pretty sure =93kernel=94 here doesn=92t refer to =93the Linux =
kernel.=94
No this is someones idea of how to teach parallel programming which is
to use a NVIDEO GPU and CUDA. WHy they couldn't come up with some idea
of what to call it other than "kernel" (maybe thread_instance or maybe a
closure ;) )
Actually, it is like a closure in that it is run through a thread with a
loop on each block? See the linked code for an example.
http://www.nylxs.com/docs/grad_school/parallel/src/cuda_second.c
ruben is me
let.me.in lets me in to most places
Maybe I am not understanding this yet. I promoised Samir I would catch
up with this class by Monday and it is Monday night. I need to do this
online class from udacity and UC Davis
I hate these video classes. I took one on Coursera for paelontology and
these things are watered down to the DUMBASS level. They talk to you
like your an idiot and they lull me to sleep. I'd give my right arm for
a textbook
https://www.udacity.com/course/intro-to-parallel-programming--cs344
These things are WORSE than the US Army video on venereal disease or how
to strip an M16.
It is killing me to do this and I'm not learning much for the time
spent. One thing that really is puzzling me is how memory is being
allocated and freed. See this line of code
cudaFree(d_out);
cudaMalloc(( void **) &d_out, ARRAY_BYTES );
cube<<<1, ARRAY_SIZE>>>(d_in, d_out);
cudaMemcpy(h_out, d_out, ARRAY_BYTES, cudaMemcpyDeviceToHost); //h_out
i s a static array. Can you memcpy like this?
for( int i =3D 0; i < ARRAY_SIZE; i++)
{
printf("Output =3D> %f ", h_out[i]);
}
//Free GPU
cudaFree(d_in);
cudaFree(d_out);
I'm trying to reuse an array for two different kernel calls and it dawns
on me, yah know, the are allocating dynamic memory to a built in
array... like WTF, can you do that?
Ruben
> =
> CL
> =
> =
> =
> _______________________________________________
> Learn mailing list
> Learn-at-nylxs.com
> http://lists.mrbrklyn.com/mailman/listinfo/learn
> =
-- =
So many immigrant groups have swept through our town
that Brooklyn, like Atlantis, reaches mythological
proportions in the mind of the world - RI Safir 1998
http://www.mrbrklyn.com
DRM is THEFT - We are the STAKEHOLDERS - RI Safir 2002
http://www.nylxs.com - Leadership Development in Free Software
http://www2.mrbrklyn.com/resources - Unpublished Archive
http://www.coinhangout.com - coins!
http://www.brooklyn-living.com
Being so tracked is for FARM ANIMALS and and extermination camps,
but incompatible with living as a free human being. -RI Safir 2013
_______________________________________________
Learn mailing list
Learn-at-nylxs.com
http://lists.mrbrklyn.com/mailman/listinfo/learn
|
|
