Re: Accepted orbit2 1:2.14.19-0.3 (source amd64)
On Fri, 22 Nov 2013, Josselin Mouette wrote:
> Given your recent behavior on Debian mailing lists, I don’t think you
> are in the best place to call people names.
This just proves that you removed the patch just to spite me
as you know I’m working with the m68k porters.
> > Please revert this immediately.
>
> Please provide a patch that keeps the code the same for all
> architectures, without breaking ABI, and I’ll think about it.
Adding the 3-byte dummy field actually *keeps* ABI because
it makes the alignment assumptions explicit.
{ char c; int i; }
powerpc: [c] [p] [p] [p] [i] [i] [i] [i]
m68k: [c] [p] [i] [i] [i] [i]
Or, with __attribute__((__packed__)) on any platform:
[c] [i] [i] [i] [i]
This is legal in C. (Here, p is implied padding.)
So that part of the patch, at least, brings m68k into
ABI compatibility with other (“natural-alignment”)
architectures.
I can’t comment on the rest of the patch, you may want
to talk to its author instead. Some of us have a dayjob
to pay rent, food, etc.
bye,
//mirabilos
--
Sometimes they [people] care too much: pretty printers [and syntax highligh-
ting, d.A.] mechanically produce pretty output that accentuates irrelevant
detail in the program, which is as sensible as putting all the prepositions
in English text in bold font. -- Rob Pike in "Notes on Programming in C"
Reply to: