On Mon, Jul 29, 2013 at 5:22 PM, Stefan Hajnoczi <stefa...@gmail.com> wrote: > On Mon, Jul 29, 2013 at 11:16:03AM +0800, Liu Ping Fan wrote: >> summary of the model: >> Three qemu-wide clock source allowed in system. And each AioContext has >> three corresponding timer list to run timer against clocksource. >> >> rfcv2: >> drop patches about alarm-timer(if timeout of poll will not satisfy, will >> come back to it) >> fix qemu_clock_enable sync problem (Thanks for Jan and Stefan) >> fix process=true when aio_poll run timers (Thanks for Alex) >> >> Liu Ping Fan (5): >> timer: protect timers_state with lock >> timer: pick out timer list info from QemuClock >> timer: make qemu_clock_enable sync between disable and timer's cb >> timer: associate three timerlists with AioContext >> timer: run timers on aio_poll >> >> aio-posix.c | 2 + >> async.c | 9 +++ >> cpus.c | 26 ++++++-- >> include/block/aio.h | 13 ++++ >> include/qemu/timer.h | 24 ++++++- >> main-loop.c | 2 - >> qemu-timer.c | 184 >> ++++++++++++++++++++++++++++++++++++--------------- >> 7 files changed, 198 insertions(+), 62 deletions(-) > > The potential for overlap with Alex Bligh's series is large. Can you > base your patches on his v4? > Re-check the patches and Alex' patches can meet my need for further step to make hpet run in dedicated thread. So I will rebase the sync part of my patches onto his.
> It seems the difference is that you make clock sources to be available > globally while Alex's series uses rt_clock (no synchronization > necessary). > > Stefan
