Anthony Liguori wrote:
This patch reworks the IO thread to use signalfd() instead of sigtimedwait().
This will eliminate the need to use SIGIO everywhere. In this version of the
patch, we use signalfd() when it's available. When it isn't available, we
create a separate thread and use
Anthony Liguori wrote:
Please break the SIGUSR1 changes into a separate patch. Ditto with
*fd syscall compat.
Done. I didn't make the syscall compat stuff separate patches because
that would break bisect on older hosts. However, I did split it up
logically between the remove sigusr1
This patch reworks the IO thread to use signalfd() instead of sigtimedwait().
This will eliminate the need to use SIGIO everywhere. In this version of the
patch, we use signalfd() when it's available. When it isn't available, we
create a separate thread and use sigwaitinfo() to simulate