rsync release planning

[Jim Kleckner]

jw schultz wrote:

> On Tue, Jul 08, 2003 at 09:51:16AM -0700, Wayne Davison wrote:
> 
>>On Sat, Jul 05, 2003 at 10:30:13PM -0700, jw schultz wrote:
...
>>>I'm also wondering about the craigb-perf patch.  Anyone know
>>>how widely it has been tested?
>>>
>>I've been wondering about it as well.  I've looked over the whole patch
>>and the comments and it certainly looks reasonable, but I am slightly
>>concerned about potential deadlock issues, though.  I haven't tested it
>>yet, nor have I tried to think through the I/O issues in depth yet.  The
>>
> 
> My grasp of what is going on in the client-server I/O just
> isn't firm enough in the light of apparent deadlocks in
> history for me to be comfortable with any change this
> central but otherwise it does indeed look a reasonable and
> good idea.  That is why i ask about testing.  My inclination
> is to commit it early in a release cycle with a back-out
> patch held ready.

...

For what it is worth, people wanting to use cygwin
with rsync for any meaningfully large work have
been forced to use the craigb-perf patch to work around
a non-blocking IO bug.  This bug was just identified
with a patch on 4-July 2003 but has not yet gotten
tested and integrated.  One would expect the buffering
improvements of craigb-perf to continue to make a good
contribution and would still be very valuable.

So the craigb-perf patch has gotten a fair bit of
testing using cygwin as the platform normally with
Linux on the server side (but not always).

See the following for the patch to cygwin stimulated
by the use of rsync:
  http://sources.redhat.com/ml/cygwin/2003-07/msg00254.html

Jim