On Tue, 2007-04-17 at 20:26 +0200, Jan Kiszka wrote: > Jan Kiszka wrote: > > RAKOTOSALAMA, Nirilanto wrote: > >> Hello everybody. > >> > >> I managed to compile a stable kernel (the Scalability O(1) scheduler > >> options seems to make my system freeze) > > > > That's not good. What precisely did you do? Applied the .config you > > posted, just enabling CONFIG_XENO_OPT_SCALABLE_SCHED? And what test > > case, latency? > > Hmm, maybe related: Philippe, I once posted you an oops under > qemu-system-x86_64 which we explained with a probably broken toolchain. > > I just pulled out that config again, updated everything > (crosstools-0.43: gcc-4.1.1-glibc-2.3.6-tls + tiny patch [1], Xenomai > trunk, adeos-ipipe-2.6.19-x86_64-1.0-06) - and still got that crash on > "latency" startup. The oops still happens around [2]. Interesting is > that turning off CONFIG_XENO_OPT_SCALABLE_SCHED makes it go away. >
I'm still having the same breakage on qemu with the scalable scheduling option disabled. Looking at the generated x86_64 code, there's really something fishy with both 3.4.5 and 4.1.0 stock. > Given that I now tested my setup with gcc4 (instead of the earlier gcc3 > run), there may be some code that "confuses" compilers (or it's simply > broken). I failed to reproduce the issue on x86, and cross debugging > somehow still don't work for me, so I can't tell more. > I have only reproduced it on qemu x86_64; the issue goes away with 3.4.6 20060404 (Red Hat 3.4.6-3) on real hw. > Jan > > [1]https://www.x86-64.org/pipermail/discuss/2006-November/009546.html > [2]http://www.rts.uni-hannover.de/xenomai/lxr/source/ksrc/nucleus/shadow.c?v=SVN-trunk#225 > > -- Philippe. _______________________________________________ Xenomai-help mailing list [email protected] https://mail.gna.org/listinfo/xenomai-help
