Re: [PATCH v3 08/10] x86/mtrr: let cache_aps_delayed_init replace mtrr_aps_delayed_init

From: Juergen Gross <jgross@suse.com>
To: Borislav Petkov <bp@alien8.de>
Cc: xen-devel@lists.xenproject.org, x86@kernel.org,
	linux-kernel@vger.kernel.org,
	Thomas Gleixner <tglx@linutronix.de>,
	Ingo Molnar <mingo@redhat.com>,
	Dave Hansen <dave.hansen@linux.intel.com>,
	"H. Peter Anvin" <hpa@zytor.com>
Subject: Re: [PATCH v3 08/10] x86/mtrr: let cache_aps_delayed_init replace mtrr_aps_delayed_init
Date: Wed, 28 Sep 2022 13:14:11 +0200	[thread overview]
Message-ID: <f8da6988-afa3-1e85-b47d-d91fc4113803@suse.com> (raw)
In-Reply-To: <YzQmeh50ne8dyR2P@zn.tnic>

[-- Attachment #1.1.1: Type: text/plain, Size: 1016 bytes --]

On 28.09.22 12:48, Borislav Petkov wrote:
> On Wed, Sep 28, 2022 at 08:16:53AM +0200, Juergen Gross wrote:
>>> Are sure the hotplug notifier doesn't get called in the boot and in the
> 
> It doesn't because it gets registered after smp_init()...
> 
>>> resume cases?
> 
> ... but it gets called during resume because by that time the notifier
> has been registered already. See my notes at the end of this mail of
> what the code does currently.
> 
>> In case my suspicion is correct: this can still be solved by adding the
>> hotplug notifier only in mtrr_aps_init(), and removing it again in
>> arch_thaw_secondary_cpus_begin().
> 
> Pretty much. Yeah, we still need a bool. ;-(

No, we don't.

Using basically your patch, but with

+	mtrr_online = cpuhp_setup_state_nocalls(CPUHP_AP_ONLINE_DYN,
+						"x86/mtrr:online",
+						mtrr_ap_init, NULL);

moved to the end of mtrr_aps_init(), and:

+void mtrr_aps_thaw(void)
+{
+	cpuhp_remove_state_nocalls(mtrr_online);
+}

Juergen

[-- Attachment #1.1.2: OpenPGP public key --]
[-- Type: application/pgp-keys, Size: 3149 bytes --]

[-- Attachment #2: OpenPGP digital signature --]
[-- Type: application/pgp-signature, Size: 495 bytes --]