On Wed, Nov 29, 2023 at 08:13:52PM -0600, Kalra, Ashish wrote:
> It surely seems hard to follow up, so i am anyway going to clean it up by:
>
> Adding the "probe" parameter to sev_platform_init() where the parameter
> being true indicates that we only want to do SNP initialization on probe,
> the same parameter will get passed on to
> __sev_platform_init_locked().
That's exactly what you should *not* do - the probe parameter controls
whether
if (psp_init_on_probe)
__sev_platform_init_locked(error);
and so on should get executed or not.
If it is unclear, lemme know and I'll do a diff to show you what I mean.
> > > + /* Prepare for first SNP guest launch after INIT */
> > > + wbinvd_on_all_cpus();
> >
> > Why is that WBINVD needed?
>
> As the comment above mentions, WBINVD + DF_FLUSH is needed before the first
> guest launch.
Lemme see if I get this straight. The correct order is:
WBINVD
SNP_INIT_*
WBINVD
DF_FLUSH
If so, do a comment which goes like this:
/*
* The order of commands to execute before the first guest
* launch is the following:
*
* bla...
*/
> In the latest code base, _sev_snp_shutdown_locked() is called from
> __sev_firmware_shutdown().
Then carve that function out only when needed - do not do changes
preemptively. This is not helping during review.
Thx.
--
Regards/Gruss,
Boris.
https://people.kernel.org/tglx/notes-about-netiquette
