On Wed, 2005-11-23 at 17:05 -0500, Fabrice Bellard wrote:
> CVSROOT: /cvsroot/qemu
> Module name: qemu
> Branch:
> Changes by: Fabrice Bellard <[EMAIL PROTECTED]> 05/11/23 22:05:28
>
> Modified files:
> . : cpu-exec.c monitor.c
> target-ppc : helper.c translate.c
>
> Log message:
> halt state support for ppc
>
> CVSWeb URLs:
> http://savannah.gnu.org/cgi-bin/viewcvs/qemu/qemu/cpu-exec.c.diff?tr1=1.63&tr2=1.64&r1=text&r2=text
> http://savannah.gnu.org/cgi-bin/viewcvs/qemu/qemu/monitor.c.diff?tr1=1.42&tr2=1.43&r1=text&r2=text
> http://savannah.gnu.org/cgi-bin/viewcvs/qemu/qemu/target-ppc/helper.c.diff?tr1=1.28&tr2=1.29&r1=text&r2=text
> http://savannah.gnu.org/cgi-bin/viewcvs/qemu/qemu/target-ppc/translate.c.diff?tr1=1.39&tr2=1.40&r1=text&r2=text
One should beware before merging this patch in its local repository.
It was rejected in the past and is broken for many reasons:
- MSR_POW bit _does not_ mean "enter halt state". It means "enable power
management".
What the PowerPC core would do when this bit is set is implementation
dependant.
For example, PowerPC 750 (the one currently emulated by Qemu targets)
will enter halt
state only if some bits in the HID0 SPR are set.
- the cpu state cannot be changed outside of the PowerPC emulation code.
Setting the MSR_POW bit to zero is buggy because we loose useful
information to decide
wether the CPU has to go back in running state (if it's really halted)
or has to ignore the
raised exception.
--
J. Mayer <[EMAIL PROTECTED]>
Never organized
___
Qemu-devel mailing list
Qemu-devel@nongnu.org
http://lists.nongnu.org/mailman/listinfo/qemu-devel
