On Tue, Jun 12, 2018 at 06:09:27PM +0100, Ramsay Jones wrote: > > > On 12/06/18 17:43, Luc Van Oostenryck wrote: > > On Tue, Jun 12, 2018 at 04:25:39PM +0100, Ramsay Jones wrote: > >> > >> > >> On 11/06/18 03:00, Luc Van Oostenryck wrote: > >>> When using sparse it's common to compile a file and directly > >>> run sparse on the same file, like it is done for the kernel. > >>> In this case, error messages from sparse are interspersed with > >>> those from the compiler. It's thus not always easy to know from > >>> which tools they come. > >>> > >>> Fix this by allowing to prefix all the diagnostic messages > >>> by some configurable string, by default "sparse". More exactly, > >>> an error message that was emitted like: > >>> file.c:<line>:<col>: error: this is invalid code > >>> can now be emitted as: > >>> file.c:<line>:<col>: sparse: error: this is invalid code > >>> > >>> Signed-off-by: Luc Van Oostenryck <luc.vanoostenryck@xxxxxxxxx> > >>> --- > >>> lib.c | 22 +++++++++++++++++++--- > >>> sparse.1 | 6 ++++++ > >>> validation/fdiag-prefix.c | 11 +++++++++++ > >>> 3 files changed, 36 insertions(+), 3 deletions(-) > >>> create mode 100644 validation/fdiag-prefix.c > >> > >> Don't you need to filter this out in cgcc, so that you don't > >> pass this option on to gcc? (ie, add this to the regex in the > >> 'check_only_option' subroutine). > > > > Maybe yes but OTOH, I'm not really sure about it. > > I thought that cgcc was theer to use sparse transparently: > > Indeed. > > > instead of using gcc you use cgcc et voila. In this case > > the options given to cgcc are the same ypu would pass to gcc > > I only have quick access to gcc-5 documentation, but I am not > seeing -fdiagnostic-prefix as an option supported by gcc. Absolutely. I had hoped that GCC would have such an option but it wasn't the case. > > itself and thus won't contains sparse-specific options. > > when using cgcc (without -no-compile), sparse is run over the > source (with all given options - sparse and gcc options) and > then (assuming no errors etc.,) gcc is then run _minus_ the > sparse specific options. The filtering of the sparse specific > options is done in the 'check_only_options' subroutine. > > [Yes, I've been meaning to change that to 'sparse_only_options'] Yes, but I meant that I thought that people would not use non-gcc options with cgcc because given thatcgcc is supposed to be a transparent replacement of gcc people would use cgcc with exactly the same options that they would use with gcc itself. But I realize that it is already not the case for git's selfcheck which contains -Wno-memcpy-max-count. So, I indeed need to filter out this option and I need to check the other options I've added the last months. Once again, many thanks for noticing this. -- Luc -- To unsubscribe from this list: send the line "unsubscribe linux-sparse" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
