Re: [Qemu-devel] [6391] Fix nographic mode and VNC

[Anthony Liguori <anthony@codemonkey.ws>]

Paul Brook wrote:
>> Some other way, like SIGIO or IO worker thread, may work too. It may
>> also be a big change.
>>     
>
> Using a side effect of the the gui refresh timer to implement IO polling is 
> absolutely the wrong way to fix things.
>
> All IO should be event driven. If it isn't then then it needs fixing. The 
> polling timeout is determined by timeout = 5000 in vl.c.
>   

No, this is a fundamentally broken aspect of TCG (and dyngen).

cpu_exec() spins in a loop checking for 'event_pending'.  This is set 
whenever the host_alarm_timer callback signals.  Alternatively, a halted 
guest will cause cpu_exec() to exit.  Only then do you do a select() on 
pending IO.

If you are using dynticks as your clock, the guest has a slow clock, and 
the guest is not using halt instructions, then you will not poll IO 
until the next guest timer tick b/c your guest is just going to spin.

There are three possible solutions.  Set SIGIO on every file descriptor 
so that TCG breaks whenever IO is pending.  Besides ugliness, this fails 
because not every type of file descriptor supports SIGIO.

A second solution is to use a polling select() in cpu_exec.  Since 
you're adding a system call (and a rather heavy one) in the fast path, 
this is going to likely hurt TCG performance.

The third possibility is to have the select() run in a separate thread 
from the TCG cpu_exec() loop.  cpu_exec() would do an atomic read of 
'event_pending' and the IO thread would do an atomic write of 
'event_pending' whenever select() returned a writable file descriptor.

Regards,

Anthony Liguori

> Paul
>
>
>