PLEASE TEST: New implementation of blocking socket I/O

[Corinna Vinschen]

On Mar 31 10:55, Pierre A. Humblet wrote:
> Is there an explanation for that invalid pointer?
> It's the cause of the "Bad address" in
> "cvs [update aborted]: error writing to server: Bad address".

Sorry, but... how should I be able to explain that invalid address if I
can't even reproduce the problem?  writev obviously checks the incoming
buffers.  Perhaps a glitch in cvs which doesn't expect writev to return
a value bigger than the number of bytes given as argument.


Corinna

-- 
Corinna Vinschen                  Please, send mails regarding Cygwin to
Cygwin Developer                                mailto:cygwin@cygwin.com
Red Hat, Inc.