discuss-gnuradio
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [Discuss-gnuradio] Google Summer of Code 2014 applicant : Optimizati


From: West, Nathan
Subject: Re: [Discuss-gnuradio] Google Summer of Code 2014 applicant : Optimization with VOLK
Date: Mon, 10 Mar 2014 17:22:12 -0500

On Mon, Mar 10, 2014 at 2:32 PM, Abhishek Bhowmick
<address@hidden> wrote:
>
>
>
> On Mon, Mar 10, 2014 at 10:04 PM, West, Nathan <address@hidden>
> wrote:
>>
>> On Mon, Mar 10, 2014 at 9:33 AM, Abhishek Bhowmick
>> <address@hidden> wrote:
>> > Hello,
>> > I would like to clarify some things :
>> >
>> > 1. I feel it is tough to beat spiral implementations through manual
>> > vectorization, performance wise. If so, is readability the prime and
>> > only reason for using intrinsics manually, and hence of value to the
>> > community ?
>>
>> Spiral is very good, and it would be hard to beat for the areas that
>> Spiral covers. However, Spiral only covers DFTs and Viterbi decoders.
>>
>> >
>> > 2. What is currently the state of adding support for sse4, neon in
>> > stock volk kernels (project ideas page mentions some work is under
>> > way) ? Would be great if someone who is working on this already shares
>> > his branch, so that I may know how much/if any work is needed in this
>> > before moving on to avx. Of course, new kernels will need support for
>> > all.
>>
>> The different versions of SSE are very similar and mostly don't really
>> replace each other, but are more complementary to each other. As for
>> NEON, I've been doing work on adding NEON support for many of the
>> currently existing kernels. It will be public (hopefully) around May. I'd
>> like to see focus on new kernels at this point, with AVX2 support for
>> some existing kernels thrown in if time permits.
>>
>> >
>> > 3. How feasible/useful does it sound to incorporate the newly added
>> > idea of 'turbo equalizer' within the ofdm system ? Are the
>> > requirements of the proposed equalizer overkill for the ofdm blocks?
>>
>> I'm glad Martin answered that one :-) I'm pretty excited about a potential
>> OFDM frame detection kernel.
>
>
> I am afraid my question didn't come out correctly. I was referring to the
> latest
> project idea addition to the wiki page on 'turbo equalizer', supposedly by
> openairinterfaces. I am not sure if I can incorporate that into my original
> proposal, or does that go as a separate proposal for openair...
>
> And if yes, whether the throughput requirements for their applications are
> much
> higher than what we can get away with on in-tree ofdm.
>
>>
>>
>> >
>> > Abhishek
>> >
>>
>> Nathan
>
>
> --
> Regards;
> Abhishek Bhowmick,
> Senior Undergraduate,
> Department of Electrical Engineering,
> IIT Bombay.

Ah! So there was a slight miscommunication. Yes, porting the OpenAirInterfaces
SIMD code to VOLK is a good option as well. The turbo channel coder/decoder
is part of that. I've *briefly* looked at the code to see what is
currently there, and
it's my understanding that the work involved will be to write generic
C implementations
of vectorized code where the generic version does not exist. Beyond
that porting to
newer/different ISAs (AVX or NEON depending on your preference and hardware
availability). I think Florian is on the gr-discuss mailing list, but
I've CCed him to
hopefully provide more details as he's more familiar with the original
code base.



reply via email to

[Prev in Thread] Current Thread [Next in Thread]