On Tue, 2018-08-28 at 14:26 +0200, Jakub Jelinek wrote: > On Tue, Aug 28, 2018 at 08:43:59AM +0200, Jakub Jelinek wrote: > > On Mon, Aug 27, 2018 at 08:32:11PM -0400, David Malcolm wrote: > > > gcc/ChangeLog: > > > * dumpfile.h (ATTRIBUTE_GCC_DUMP_PRINTF): Change version check > > > on > > > GCC_VERSION for usage of "__gcc_dump_printf__" format from > > > >= 3005 to >= 9000. > > > > Ok, thanks. > > Another option would be to use __gcc_tdiag__ for GCC_VERSION >= 3005 > and < > 9000, that differs from __gcc_dump_printf__ only in %E (tree vs. > gimple *) > and %D/%F/%V/%K added to __gcc_tdiag__, so worst case one would get a > couple > of rare warnings when %E is used somewhere; certainly better than > having a > warning on every TU that includes dumpfile.h.
Maybe, but I think that trying to decipher warnings based on something that's close to the rules but not the same as them could get confusing (especially if we add some new codes to __gcc_dump_printf__). I've gone ahead with the simpler fix, as posted above, to stop the flood of warnings (r263920). Sorry again. Dave
