I think Dmitri has a good point. Diagnostics are not "dynamic", tablegen's purpose is to emit the necessary static data so that we save on the cost of calculating the same data on every run and the management of the data is simplified (just static tables). I think it is worth modifying tablegen to do it.
On Dec 12, 2012, at 9:24 AM, Matthieu Monrocq <[email protected]> wrote: > > > On Wed, Dec 12, 2012 at 2:07 PM, Manuel Klimek <[email protected]> wrote: > > > As far as I understand, the information contained in the cache should be > known to TableGen during Clang compile-time? If yes -- why not just generate > this cache with TableGen? > > I never used tablegen - on IRC Benjamin said that knowing the size of the > holes at tablegen-time is not trivial. > > http://llvm-reviews.chandlerc.com/D197 > _______________________________________________ > cfe-commits mailing list > [email protected] > http://lists.cs.uiuc.edu/mailman/listinfo/cfe-commits > > The not trivial to know because the diagnostics are split into several files > and each file is processed separately by tablegen (last I fiddled with it, at > least). > > -- Matthieu > _______________________________________________ > cfe-commits mailing list > [email protected] > http://lists.cs.uiuc.edu/mailman/listinfo/cfe-commits
_______________________________________________ cfe-commits mailing list [email protected] http://lists.cs.uiuc.edu/mailman/listinfo/cfe-commits
