Re: [lwip-users] SOCK_RAW

[Kieran Mansley]

On Fri, 27 Feb 2004, Ed Sutter wrote:
> Assuming the sluggishness is in fact due to the sockets layer, has
> anyone considered improving it or does the underlying design of the
> sequential API just make it difficult to implement an efficient
> sockets layer on top of it?

There isn't a simple fix that we could do that would just make it all work
faster.  As you've noticed many people have made changes, but these are
generally fairly radical (I think - mine certainly were!), and involve a
change in the structure of the underlying system.  They also often make
assumptions about the driver/hardware and so aren't general purpose.
There are also ways it could be made faster, but would have side effects
on its correctness, and I think it's more important for most that it works
correctly rather than fast.

If, of course, anyone does have suggestions for fixes that could be made
that don't do any of that then let us know and we'll see what we can do.

Kieran

PS. If you're interested, I've gone into more detail in the past about the
changes I made to get a fast sockets layer, so you should still be able to
find them in the mailing list archives.