RE: [Qemu-devel] [PATCH 1/5] single step with no IRQs and timers

[Wessel, Jason]

Fabrice,

I will defer to you on the real time timers.  I had not examined the
code thoroughly enough to understand if there was more to the real time
timers.  I had wanted to guard against any kind of interrupts or
triggers that might fire off.  Just call it a lack of total
understanding of the complete QEMU infrastructure.  I would say the real
time timers can run if you feel the dependencies for precise single
instruction execution are not violated.

I found gdb against a linux kernel to be next to useless for any real
run control debugging with out the patch.  Deterministic execution
sounds great.  I would be happy to add debugger hook controls to further
control it at runtime with GDB, if your deterministic execution controls
are not already there at time.  I have found QEMU to be very useful for
the analysis of strange run control issues which is where the patch came
from in the first place.

Just let me know if you want me to re-gen the patch or if it will
ultimately go to the bit bucket in favor of something better.  It is a
great stopgap in the in mean time.

Thanks,
Jason.

> -----Original Message-----
> From: 
> address@hidden 
> [mailto:address@hidden
> rg] On Behalf Of Fabrice Bellard
> Sent: Monday, May 22, 2006 4:37 PM
> To: address@hidden
> Subject: Re: [Qemu-devel] [PATCH 1/5] single step with no 
> IRQs and timers
> 
> Hi,
> 
> I agree that the feature is useful, but I am not sure yet if 
> I will apply as is. I am adding support in QEMU for 
> deterministic execution and your patch won't be needed once 
> this new feature is released.
> 
> BTW, why do you stop real time timers ? These timers are only 
> used for things which do not interfere with the VM (such as 
> VGA refresh).
> 
> Fabrice.