Stephen Rothwell wrote:
>
> Hi Garst,
>
> "Garst R. Reese" <reese@isn.net> writes:
> >
> > Stephen,
> > I sent the attached patch to apm linuxcare that you seem to think has
> > already been applied to the kernel. I don't think so. I never sent it to
> > Linus or the list, and have not seen it in any test6prex.
> > The problem I addressed was that there was a cli() instead of an
> > APM_DO_CLI
> > in int suspend(void);
> >
> > I've been running with the patch I submitted to fix this problem for
> > some time.
> > Your solution to do an sti() in APM_DO_CLI seems quite ugly. I grant
> > that we do not want to do a cli before suspending, but I could not find
> > anybody else doing that, so just not doing is probably sufficient.
>
> I did NOT assume that your patch had been applied (I am the person behind
> apm@linuxcare.com.au). What my patch does is to ensure that interrupts
> are explicitly enabled whenever we do a BIOS call (including the BIOS call
> to suspend the machine) rather than just assuming that interrupts were
> enabled. This does work in my Thinkpad (admittedly a 600E).
>
> I do not think that my patch is ugly and, in fact, implements the
> semantics of the CONFIG option even better than before. :-)
>
> Please try my patch ...
Hmm, how does changing APM_DO_CLI fix the fact that in test5-pre2, an
explicit cli() was issued? (in the function suspend at line 914).
I do not understand how you avoid that cli.
Garst
-- beauty is in the eye ...- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.rutgers.edu Please read the FAQ at http://www.tux.org/lkml/
This archive was generated by hypermail 2b29 : Tue Aug 15 2000 - 21:00:17 EST