On Thu, 20 Jan 2011 18:41:54 +0200 Alexander Nezhinsky <alexandern@xxxxxxxxxxxx> wrote: > Added timerfd-backed work scheduler where the kernel supports it. > Otherwise signal-based timer mechanism is used. > Using timerfd leads to less cpu usage compared with the signal-based scheduler. > > This patch needs to be applied on the top of: > http://lists.wpkg.org/pipermail/stgt/2011-January/004380.html > > Signed-off-by: Alexander Nezhinsky <alexandern@xxxxxxxxxxxx> > Signed-off-by: FUJITA Tomonori <fujita.tomonori@xxxxxxxxxxxxx> > --- > usr/Makefile | 4 ++ > usr/work.c | 145 +++++++++++++++++++++++++++++++++++----------------------- > usr/work.h | 33 ++++++++++++- > 3 files changed, 123 insertions(+), 59 deletions(-) (snip) > + timer_fd[0] = __timerfd_create(WORK_TIMER_INT_USEC); > + if (timer_fd[0] > 0) I think that zero is a valid fd. > -#include <sys/time.h> > +#if defined(__NR_timerfd) && defined(USE_TIMERFD) There is no __NR_timerfd. You need to use __NR_timerfd_create, etc. I fixed both and applied. Please check if the latest git works for you. Thanks, -- To unsubscribe from this list: send the line "unsubscribe stgt" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html