Re: [lwip-devel] IPH_V() and IPH_HL(hdr) macro endianess dependence?

[Simon Goldschmidt]

Leon Woestenberg <address@hidden> wrote:
> since when is nibble order dependent on byte endianess?

I guess that slipped in because using the endianness dependency was an 
intermediate step to getting rid of ntohs in these macros. And because our 
platforms are all little endian, I don't easily catch such a bug :(

> Sorry, but I do not get how this code gets into HEAD?

Well, by committing it of course. HEAD is currently a development tree for 
improvements, not for always having a stable ready-to-release code. And during 
development, bugs happen (even if I'd like them not to happen).

> This cost me hours of debugging finding this code and is rather frustrating.

I see that and I'm sorry for that!

> Proposal: I would like to open the 1.4.0 release branch for bug fixes
> only, This is a not a playground. If a regression comes in today,
> tomorrow tens of people will suffer from it.

I'm not against that, but I'm not the one to decide. The bugs that would need 
to be fixed in that tree are noted in CHANGELOG, so it's not too late to open 
such a tree.

However, we do need someone taking care of that tree. If you are willing to 
step forward here, Kieren might be easily convinced...

Simon
-- 
NEU: FreePhone - 0ct/min Handyspartarif mit Geld-zurück-Garantie!               
Jetzt informieren: http://www.gmx.net/de/go/freephone