Hi all, Backports of the staircase deadline cpu scheduler for 2.6.18.8 and 2.6.19.7 kernels are available at http://linux-dev.qc.ec.gc.ca/.
Also, Debian Etch 4.0 kernels for both i686 and x86_64 are available. Fedora Core 6 kernels might also be available later today/tomorrow. Both patches looks quite stable. I've been able to fix the 2.6.19 backport by reversing this patch: "[PATCH] move_task_off_dead_cpu() should be called with disabled ints" (see http://git.kernel.org/?p=linux/kernel/git/stable/linux-2.6.20.y.git;a=co mmit;h=054b9108e01ef27e2e6b32b4226abb6024626f06). I don't know if this bug (see http://linux-dev.qc.ec.gc.ca/kernel/debian/etch/x86_64/2.6.19/bug-2.6.19 .7-rsdl-033.jpg) can also affect a Vanilla 2.6.19.7 kernel or even the actual mainline? Duno if it would be worth investigating a bit further? Although, here is a top of a 2.6.19.7 + SD 0.36 patch in action while building 3 kernels at once and running a glxgears in foreground: top - 12:39:39 up 17 min, 2 users, load average: 5.84, 4.37, 2.20 Tasks: 162 total, 5 running, 157 sleeping, 0 stopped, 0 zombie Cpu0 : 87.0%us, 13.0%sy, 0.0%ni, 0.0%id, 0.0%wa, 0.0%hi, 0.0%si, 0.0%st Cpu1 : 88.4%us, 11.6%sy, 0.0%ni, 0.0%id, 0.0%wa, 0.0%hi, 0.0%si, 0.0%st Mem: 2059940k total, 1249476k used, 810464k free, 137112k buffers Swap: 1959888k total, 0k used, 1959888k free, 620552k cached PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ COMMAND 15970 root 33 0 45016 36m 4520 R 18 1.8 0:00.54 cc1 16025 root 1 0 40904 32m 4444 R 18 1.6 0:00.53 cc1 16015 root 33 0 40620 32m 4496 R 17 1.6 0:00.52 cc1 13401 th0ma7 1 0 42356 9420 6700 R 6 0.5 1:16.13 glxgears 16589 root 1 0 3756 400 192 S 1 0.0 0:01.04 faked-sysv 5012 root 5 0 91208 46m 7200 S 0 2.3 0:11.57 Xorg Note that my KDE session/firefox/kontact/etc. was still pretty responsive! Vincent Fortier Informatique Environnement Canada - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
