On Tue, 06 Mar 2018 23:15:45 +1100
Michael Ellerman <m...@ellerman.id.au> wrote:
> Mauricio Faria de Oliveira <mauri...@linux.vnet.ibm.com> writes:
>
> > Hi Michael, Michal,
> >
> > I got back from vacation. Checking this one.
>
> Yeah it got stuck in a rut.
>
> > On 02/20/2018 02:06 PM, Michal Suchánek wrote:
> >>> I did it the way I did because otherwise we waste memory on every
> >>> system on earth just to support a use case that we don't actually
> >>> intend for anyone to ever use - ie. migrating from a patched
> >>> machine to an unpatched machine.
> >
> > If this thread eventually closes in 'ok, so that memory has to be
> > reserved/wasted anyway', that can be done only in pseries, right?
> >
> > It seems not so much memory for this particular platform/hardware.
>
> Yes and no. We certainly have pseries machines with 10s of TBs of
> memory, but pseries is also used when running on KVM/Openstack where
> some people are interested in running small VMs - so as usual it's not
> so simple :)
>
> But it's not *that* much memory so we should probably stop worrying
> about it and just always allocate the fallback on pseries.
>
> I *think* the patch below is all we need, as well as some tweaking of
> patch 2, are you able to test and repost?
>
> cheers
>
> diff --git a/arch/powerpc/platforms/pseries/setup.c
> b/arch/powerpc/platforms/pseries/setup.c index
> 1a527625acf7..9116824bd7c5 100644 ---
> a/arch/powerpc/platforms/pseries/setup.c +++
> b/arch/powerpc/platforms/pseries/setup.c @@ -468,26 +468,18 @@ static
> void pseries_setup_rfi_flush(void)
> /* Enable by default */
> enable = true;
> + types = L1D_FLUSH_FALLBACK;
>
> rc = plpar_get_cpu_characteristics(&result);
> if (rc == H_SUCCESS) {
> - types = L1D_FLUSH_NONE;
> -
> if (result.character & H_CPU_CHAR_L1D_FLUSH_TRIG2)
> types |= L1D_FLUSH_MTTRIG;
> if (result.character & H_CPU_CHAR_L1D_FLUSH_ORI30)
> types |= L1D_FLUSH_ORI;
>
> - /* Use fallback if nothing set in hcall */
> - if (types == L1D_FLUSH_NONE)
> - types = L1D_FLUSH_FALLBACK;
> -
> if ((!(result.behaviour & H_CPU_BEHAV_L1D_FLUSH_PR))
> || (!(result.behaviour & H_CPU_BEHAV_FAVOUR_SECURITY)))
> enable = false;
> - } else {
> - /* Default to fallback if case hcall is not
> available */
> - types = L1D_FLUSH_FALLBACK;
> }
>
> setup_rfi_flush(types, enable);
>
Enabling the fallback flush always looks a bit dodgy but
do_rfi_flush_fixups will overwrite the jump so long any other fixup is
enabled.
Reviewed-by: Michal Suchánek <msucha...@suse.de>
Thanks
Michal
