[PATCH]: Updated new sparse bitmap patch

Daniel Berlin dberlin@dberlin.org
Fri Mar 23 07:14:00 GMT 2007


Here is the updated patch, with all the fixes for things mentioned in
this thread.

For the moment, it uses the pos += fprintf.  I have no aversion to
changing that if rth (or anyone else) wants it done :)


On 3/20/07, Daniel Berlin <dberlin@dberlin.org> wrote:
> On 3/20/07, Seongbae Park <seongbae.park@gmail.com> wrote:
> > On 3/20/07, Daniel Berlin <dberlin@dberlin.org> wrote:
> > ...
> > > Copied from sbitmap.c so it matched exactly the output style for sbitmaps.  :)
> > > I've discovered that GCC  people are less likely to use new code if
> > > the debugging dumps don't look like what they are used to.
> > > Of course I personally have no aversion to changing it to whatever people want.
> >
> > Sounds like we should at least take Chris's suggestion
> > and fix both your ebitmap.c and the existing sbitmap.c.
> Sure.
>
> >
> > We can then spend the rest of our email bandwidth
> > on an endless discussion of which output format is better :)
> >
> > More seriously,
> > I wish we have a way to unittest the basic datastructure implementations
> > without building the whole thing.
> I agree.
> The reason the header list is so small is because i built it
> separately for a long time while working on it.
>
> > Then we should be able to do away with the internal checking code in ebitmap.c
> > as well as having much better testing...
> I agree.
>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: ebitmap.diff
Type: text/x-patch
Size: 34289 bytes
Desc: not available
URL: <http://gcc.gnu.org/pipermail/gcc-patches/attachments/20070323/51571ac0/attachment.bin>


More information about the Gcc-patches mailing list