On Tue, Oct 23, 2007 at 12:57:48AM +0200, Bernd Zeimetz wrote: > Hi Dann, > > thanks for the report. > Interestingly this was not reported by gcc-4.2.3 while test-building the > package here:
Built on amd64, I assume? > 0 [EMAIL PROTECTED]:~/workspace/debian/viking$ grep 'implicit' 0.9.3-build.log > viktrwlayer.c-implicit-decl-fix not applied to ./ . > dems.c-implicit-decl-fix not applied to ./ . > applying patch dems.c-implicit-decl-fix to ./ ... ok. > applying patch viktrwlayer.c-implicit-decl-fix to ./ ... ok. > 0 [EMAIL PROTECTED]:~/workspace/debian/viking$ > I don't have a pbuilder log anymore, though. > Two other implicit declarations were in there before, but the one which > is shown in the buildd's logs didn't appear. > > So out of curiosity: Which gcc version is installed on the buildd? You can see the versions of the toolchain here: http://buildd.debian.org/fetch.cgi?&pkg=viking&ver=0.9.3-1&arch=amd64&stamp=1193080142&file=log > > > Our automated buildd log filter[1] detected a problem that is likely to > > cause your package to segfault on architectures where the size of a > > pointer is greater than the size of an integer, such as ia64 and amd64. > > Didn't manage to segfault it on amd64 at least. This issue is often masked on amd64 because the top half of the address usually happens to be 0's. So, truncating it then 0-padding it results in the correct address. This isn't true on ia64, and I'm not sure about alpha these days. -- dann frazier -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
