On Fri, 2013-01-18 14:21:44 -0500, Paul Koning <paulkoning%comcast.net@localhost> wrote: > On Jan 18, 2013, at 1:10 PM, Holm Tiffe wrote: > > romanis%gmx.ch@localhost wrote: > > > in order to track down a internal compiler error, i've tried to > > > simplify a source file, until i have a minimal piece of code. > > > > > > unfortunatly it's not as easy as i expected :-( > > > > > > the source code i use will generate a internal compiler error, if > > > i compile it with -O1. But if i remove another line (which can be > > > completly irrelevant for the logic) it will compile cleanly. > > > > > > i've uploaded my attempt to: > > > http://ec2-54-228-13-95.eu-test-1.compute.amazonaws.com/test.c > > > > [..] > > > > A guy from the netherlands reported me that this link is dead > > (server not found) and asked if I can report this... > > I think the best way to report problems like this is to follow the > problem reporting procedure for GCC (see gcc.gnu.org). That means > logging the bug in the GCC bugz, with a .i file (preprocessor > output) that reproduces the issue. That's much more reliable than > pointing to random external web sites, and helps the maintainer > trying to find and fix the problem. I already had a quick look at it. The compiler in use is quite old (some 4.1.x variant) and the testcase doesn't ICE with a current GCC version (from upstream SCM.) So I asked him if he'd give a freshly compiled GCC a try. MfG, JBG -- Jan-Benedict Glaw jbglaw%lug-owl.de@localhost +49-172-7608481 Signature of: Lauf nicht vor Deinem Glück davon: the second : Es könnte hinter Dir stehen!
Description: Digital signature