[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [lwip-devel] Paper: A Comparison of TCP Implementations, Linux vs. l
From: |
Richard Siegfried |
Subject: |
Re: [lwip-devel] Paper: A Comparison of TCP Implementations, Linux vs. lwIP |
Date: |
Mon, 22 May 2017 03:31:42 +0200 |
On 19/05/17 09:45, Simon Goldschmidt wrote:
> Richard Siegfried wrote:
>> Some months ago I wrote a paper on a Comparison of TCP Implementations.
>> (Features, Code Quality, Data Structures, etc.)
>
> Why did you compare lwIP to Linux? Skipping throught the paper, I did not see
> an answer to this. Nor did I see an introduction to what is lwIP (Linux
> should be known well enough to not need this...).
>
>> I'm thankfull for any feedback, corrections or opinions about the
>> conclusions I found.
>
> Conclusions for lwIP are mainly:
> 1) less features than linux
> 2) not suited for multithreading
> 3) the tcp_receive() function is too complex
>
> As to 1 & 2: this is well-known and intentional.
> As to 3: reading (or debugging) the sources, this is obvious :-) Sadly, noone
> has volunteered to imrpove it, yet :-(
Oh, so refactoring would be merged.
Well, good to know, perhaps, some day if I got some spare time.
Another question which always surprised me (not part of the paper I think):
Is there a reason lwIP doesn't use inline functions for performance?
Linux usess them a lot.
(The inline keyword is part of C99 standards, or as a gcc extension
before that. http://www.greenend.org.uk/rjk/tech/inline.html)
>
> Oh, and I was both glad and amused to read that CC seems to serve no real
> need!
Oh yeah, lol, my bad, sorry
Thanks
-- Richard
signature.asc
Description: OpenPGP digital signature