quandary with pthreads

Christopher Faylor cgf@redhat.com
Fri Dec 13 21:11:00 GMT 2002

On Thu, Dec 12, 2002 at 06:08:59PM +0100, Thomas Pfaff wrote:
>On Thu, 12 Dec 2002, Christopher Faylor wrote:
>>Nope.  I'm not going to look at this.  One or both of you need to
>>clarify what is going on here.  Sending a diff where nearly every line
>>is changed (thanks to gratuitous formatting difference?) is not a help.
>You may notice that most of the changes were tabs in your source and
>spaces in the other.  Attached is a whitespace ignored diff.

The theory is that the person submitting the patch should try, as much
as possible, to reduce the barrier to entry to getting it approved.  If
you noticed this, then you could have saved a step and eliminated the
white space differences and helped smooth out this process.

However, AFAICT, the changes were more than just the addition of one
function, complete with volatile.  There were other changes too.  It
doesn't look like there will be any further information on those
forthcoming, however.

>>I certainly appreciate the effort involved in tracking down the
>>problem.  I would appreciate a little more effort in showing what the
>>problem actually was.
>The most important problem was that you didn't create an
>InterlockedCompareEchange function.  If you did i never had any
>problems, because thats were i failed.  After creating some versions i
>had one that worked when not inlined.  I didn't realize the missing

I took a stab at modifying winbase.h based on the spirit of the changes.
I didn't see any reason to reindent anything, though, so I've kept the
previous whitespace usage.

These inline replacements are now on by default.  I'm generating a
snapshot now.

Again, I appreciate the effort that you and Gary made towards getting
working versions of these functions into cygwin.


More information about the Cygwin-developers mailing list