[PATCH] connect's addrlen should be sizeof(local), not the contained path's length. Permits layer23 to work on OS X.

David Wilson dw at botanicus.net
Thu Dec 9 13:35:27 UTC 2010

That's indeed the case, on OS X connect() will fail with ENOENT unless
the correct size is used. I just noticed I've changed the formatting
in that patch. Is there a particular code style adhered to?

With this patch, the pl2303 driver from
http://www.prolific.com.tw/eng/downloads.asp?id=31 ,
/dev/tty.usbserial , MacPorts arm-elf-gcc and RANLIB=arm-elf-ranlib,
things appear to just work on OS X. I will try to send another patch
for the ranlib change (system ranlib on OS X is Mach O).

Finally is there a good default set of git format-patch flags to use?
That was my first ever attempt, and in particular the git log message
didn't make it into the body at all.


On 9 December 2010 13:23, Harald Welte <laforge at gnumonks.org> wrote:
> Hi Zecke,
> On Thu, Dec 09, 2010 at 08:05:41AM +0100, Holger Hans Peter Freyther wrote:
>> On 12/09/2010 04:33 AM, David Wilson wrote:
>> any more verbose commit message? What issue did you face? What is your
>> reference?
> my suggestion is that he was using a non-linux platform (specifically MacOS)
> which has slightly different layout of the socket address structure.
> In any case, sizeof() the entire structure is correct, not size of the two
> members (which is ignoring any padding, etc.)
> nonetheless, a more detailed description would be helpful so we don't have
> to guess...
> --
> - Harald Welte <laforge at gnumonks.org>           http://laforge.gnumonks.org/
> ============================================================================
> "Privacy in residential applications is a desirable marketing option."
>                                                  (ETSI EN 300 175-7 Ch. A6)

More information about the baseband-devel mailing list