On January 14, 2015 12:58:26 AM CST, Sebastian Huber <sebastian.hu...@embedded-brains.de> wrote: >As it turned out in the discussion of > >https://gcc.gnu.org/bugzilla/show_bug.cgi?id=64576 > >in GCC 5 we have a change of the default language > >https://gcc.gnu.org/gcc-5/changes.html > >Lets see if this was the only problem.
I left a build of all BSPs against what GCC from the head I have last night. GCC doesn't complete building for some targets so we will have no info on those. But this was the only set if failures I recall from the last build. >On 13/01/15 16:52, Joel Sherrill wrote: >> On 1/13/2015 12:11 AM, Sebastian Huber wrote: >>> >Since this worked for a long time, this might be a CPP regression, >but a >>> >%1 parameter doesn't really look like a C construct. >> I filed this as GCC PR 64576 and am doing a git bisect now. >> >> I would like to get a ruling and/or suggestion for it. This code >> is 15 years old so gcc has accepted it a LONG time. --joel _______________________________________________ devel mailing list devel@rtems.org http://lists.rtems.org/mailman/listinfo/devel
