This is the mail archive of the binutils@sources.redhat.com mailing list for the binutils project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

RE: [m68k]: convert mode 5 addressing w/zero offset into mode2


> -----Original Message-----
> From: binutils-owner On Behalf Of Nick Clifton
> Sent: 27 May 2004 17:21

> Hi Peter,
> 
> > The patch includes a new testcase (mode5.[sd]), and since I need to
> > submit a patch for the new file, I had to put it together 
> from a diff -c3pN
> > of the new files, so my apologies in advance if this is an 
> incorrect approach.
> > 
> > Any suggestions on a definitive procedure for generating a 
> patch for a
> > newly create file in a CVS tree would be most appreciated.
> 
> Actually using the -N switch is the recommended way.


  Say, didn't we just have this discussion?  Or was it on the gcc list?  

  Anyway, one point which came out of it last time is that that won't work
if the file hasn't already been "cvs add"ed, because cvs won't even run diff
on it.  So you can only use "cvs diff -N" to get a patch for a new file in
the window between cvs adding a file and cvs committing it.  

  And if you don't have write access to the repository in question, you
can't do the cvs add and so have to manually edit your CVS metadata files.
(IIRC; the add gets blocked rather than the later attempt to commit, even
though cvs add is only a local change to your CVS/Entries file).


    cheers, 
      DaveK
-- 
Can't think of a witty .sigline today....


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]