mirror of https://lore.kernel.org/lkml/
 help / color / mirror / Atom feed
From: Thomas Gleixner <tglx@linutronix.de>
To: LKML <linux-kernel@vger.kernel.org>
Cc: x86@kernel.org, Borislav Petkov <bp@alien8.de>,
	Ashok Raj <ashok.raj@intel.com>,
	Arjan van de Ven <arjan@linux.intel.com>
Subject: [patch 27/30] x86/apic: Provide apic_force_nmi_on_cpu()
Date: Thu, 10 Aug 2023 20:38:06 +0200 (CEST)	[thread overview]
Message-ID: <20230810160806.506010083@linutronix.de> (raw)
In-Reply-To: <20230810153317.850017756@linutronix.de>

From: Thomas Gleixner <tglx@linutronix.de>

When SMT siblings are soft-offlined and parked in one of the play_dead()
variants they still react on NMI, which is problematic on affected Intel
CPUs. The default play_dead() variant uses MWAIT on modern CPUs, which is
not guaranteed to be safe when updated concurrently.

Right now late loading is prevented when not all SMT siblings are online,
but as they still react on NMI, it is possible to bring them out of their
park position into a trivial rendevouz handler.

Provide a function which allows to do that. I does sanity checks whether
the target is in the cpus_booted_once_mask and whether the APIC driver
supports it.

Mark X2APIC and XAPIC as capable, but exclude 32bit and the UV and NUMACHIP
variants as that needs feedback from the relevant experts.

Signed-off-by: Thomas Gleixner <tglx@linutronix.de>

---
 arch/x86/include/asm/apic.h           |  5 +++++
 arch/x86/kernel/apic/apic_flat_64.c   |  2 ++
 arch/x86/kernel/apic/ipi.c            |  9 ++++++++-
 arch/x86/kernel/apic/x2apic_cluster.c |  1 +
 arch/x86/kernel/apic/x2apic_phys.c    |  1 +
 5 files changed, 17 insertions(+), 1 deletion(-)
---
diff --git a/arch/x86/include/asm/apic.h b/arch/x86/include/asm/apic.h
index 98c32aa5963a..e219e6c62138 100644
--- a/arch/x86/include/asm/apic.h
+++ b/arch/x86/include/asm/apic.h
@@ -301,6 +301,9 @@ struct apic {
 	enum apic_delivery_modes delivery_mode;
 	bool	dest_mode_logical;
 
+	/* Allows to send an NMI to an "offline" CPU which hangs in *play_dead() */
+	bool	nmi_to_offline_cpu;
+
 	u32	(*calc_dest_apicid)(unsigned int cpu);
 
 	/* ICR related functions */
@@ -505,6 +508,8 @@ extern void default_ioapic_phys_id_map(physid_mask_t *phys_map, physid_mask_t *r
 extern int default_cpu_present_to_apicid(int mps_cpu);
 extern int default_check_phys_apicid_present(int phys_apicid);
 
+void apic_send_nmi_to_offline_cpu(unsigned int cpu);
+
 #endif /* CONFIG_X86_LOCAL_APIC */
 
 #ifdef CONFIG_SMP
diff --git a/arch/x86/kernel/apic/apic_flat_64.c b/arch/x86/kernel/apic/apic_flat_64.c
index 8f72b4351c9f..4340f471e6a6 100644
--- a/arch/x86/kernel/apic/apic_flat_64.c
+++ b/arch/x86/kernel/apic/apic_flat_64.c
@@ -138,6 +138,7 @@ static struct apic apic_flat __ro_after_init = {
 	.send_IPI_allbutself		= default_send_IPI_allbutself,
 	.send_IPI_all			= default_send_IPI_all,
 	.send_IPI_self			= default_send_IPI_self,
+	.nmi_to_offline_cpu		= true,
 
 	.inquire_remote_apic		= default_inquire_remote_apic,
 
@@ -229,6 +230,7 @@ static struct apic apic_physflat __ro_after_init = {
 	.send_IPI_allbutself		= default_send_IPI_allbutself,
 	.send_IPI_all			= default_send_IPI_all,
 	.send_IPI_self			= default_send_IPI_self,
+	.nmi_to_offline_cpu		= true,
 
 	.inquire_remote_apic		= default_inquire_remote_apic,
 
diff --git a/arch/x86/kernel/apic/ipi.c b/arch/x86/kernel/apic/ipi.c
index 2a6509e8c840..6ee6cce4423a 100644
--- a/arch/x86/kernel/apic/ipi.c
+++ b/arch/x86/kernel/apic/ipi.c
@@ -95,8 +95,15 @@ void native_send_call_func_ipi(const struct cpumask *mask)
 	apic->send_IPI_mask(mask, CALL_FUNCTION_VECTOR);
 }
 
+void apic_send_nmi_to_offline_cpu(unsigned int cpu)
+{
+	if (WARN_ON_ONCE(!apic->nmi_to_offline_cpu))
+		return;
+	if (WARN_ON_ONCE(!cpumask_test_cpu(cpu, &cpus_booted_once_mask)))
+		return;
+	apic->send_IPI(cpu, NMI_VECTOR);
+}
 #endif /* CONFIG_SMP */
-
 static inline int __prepare_ICR2(unsigned int mask)
 {
 	return SET_XAPIC_DEST_FIELD(mask);
diff --git a/arch/x86/kernel/apic/x2apic_cluster.c b/arch/x86/kernel/apic/x2apic_cluster.c
index b2b2b7f3e03f..685437a98463 100644
--- a/arch/x86/kernel/apic/x2apic_cluster.c
+++ b/arch/x86/kernel/apic/x2apic_cluster.c
@@ -264,6 +264,7 @@ static struct apic apic_x2apic_cluster __ro_after_init = {
 	.send_IPI_allbutself		= x2apic_send_IPI_allbutself,
 	.send_IPI_all			= x2apic_send_IPI_all,
 	.send_IPI_self			= x2apic_send_IPI_self,
+	.nmi_to_offline_cpu		= true,
 
 	.inquire_remote_apic		= NULL,
 
diff --git a/arch/x86/kernel/apic/x2apic_phys.c b/arch/x86/kernel/apic/x2apic_phys.c
index 896bc41cb2ba..d5e44cb7e15f 100644
--- a/arch/x86/kernel/apic/x2apic_phys.c
+++ b/arch/x86/kernel/apic/x2apic_phys.c
@@ -188,6 +188,7 @@ static struct apic apic_x2apic_phys __ro_after_init = {
 	.send_IPI_allbutself		= x2apic_send_IPI_allbutself,
 	.send_IPI_all			= x2apic_send_IPI_all,
 	.send_IPI_self			= x2apic_send_IPI_self,
+	.nmi_to_offline_cpu		= true,
 
 	.inquire_remote_apic		= NULL,
 


  parent reply	other threads:[~2023-08-10 18:39 UTC|newest]

Thread overview: 53+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2023-08-10 18:37 [patch 00/30] x86/microcode: Cleanup and late loading enhancements Thomas Gleixner
2023-08-10 18:37 ` [patch 01/30] x86/mm: Remove unused microcode.h include Thomas Gleixner
2023-08-10 18:37 ` [patch 02/30] x86/microcode: Hide the config knob Thomas Gleixner
2023-08-10 20:41   ` Ashok Raj
2023-08-11 11:22     ` Borislav Petkov
2023-08-10 18:37 ` [patch 03/30] x86/microcode/intel: Move microcode functions out of cpu/intel.c Thomas Gleixner
2023-08-10 18:37 ` [patch 04/30] x86/microcode: Include vendor headers into microcode.h Thomas Gleixner
2023-08-10 18:37 ` [patch 05/30] x86/microcode: Make reload_early_microcode() static Thomas Gleixner
2023-08-10 18:37 ` [patch 06/30] x86/microcode/intel: Rename get_datasize() since its used externally Thomas Gleixner
2023-08-10 18:37 ` [patch 07/30] x86/microcode: Move core specific defines to local header Thomas Gleixner
2023-08-10 18:37 ` [patch 08/30] x86/microcode/intel: Rip out mixed stepping support for Intel CPUs Thomas Gleixner
2023-08-11 22:25   ` Borislav Petkov
2023-08-12  7:16     ` Thomas Gleixner
2023-08-10 18:37 ` [patch 09/30] x86/microcode/intel: Remove debug code Thomas Gleixner
2023-08-11 10:45   ` Nikolay Borisov
2023-08-10 18:37 ` [patch 10/30] x86/microcode: Remove pointless mutex Thomas Gleixner
2023-08-10 18:37 ` [patch 11/30] x86/microcode/intel: Simplify scan_microcode() Thomas Gleixner
2023-08-10 18:37 ` [patch 12/30] x86/microcode/intel: Simplify and rename generic_load_microcode() Thomas Gleixner
2023-08-10 18:37 ` [patch 13/30] x86/microcode/intel: Cleanup code further Thomas Gleixner
2023-08-11 11:01   ` Nikolay Borisov
2023-08-10 18:37 ` [patch 14/30] x86/microcode/intel: Simplify early loading Thomas Gleixner
2023-08-10 18:37 ` [patch 15/30] x86/microcode/intel: Save the microcode only after a successful late-load Thomas Gleixner
2023-08-10 18:37 ` [patch 16/30] x86/microcode/intel: Switch to kvmalloc() Thomas Gleixner
2023-08-10 18:37 ` [patch 17/30] x86/microcode/intel: Unify microcode apply() functions Thomas Gleixner
2023-08-10 18:37 ` [patch 18/30] x86/microcode: Handle "nosmt" correctly Thomas Gleixner
2023-08-10 18:37 ` [patch 19/30] x86/microcode: Clarify the late load logic Thomas Gleixner
2023-08-10 18:37 ` [patch 20/30] x86/microcode: Sanitize __wait_for_cpus() Thomas Gleixner
2023-08-10 18:37 ` [patch 21/30] x86/microcode: Add per CPU result state Thomas Gleixner
2023-08-10 18:37 ` [patch 22/30] x86/microcode: Add per CPU control field Thomas Gleixner
2023-08-10 18:38 ` [patch 23/30] x86/microcode: Provide new control functions Thomas Gleixner
2023-08-10 20:25   ` Peter Zijlstra
2023-08-10 18:38 ` [patch 24/30] x86/microcode: Replace the all in one rendevouz handler Thomas Gleixner
2023-08-10 18:38 ` [patch 25/30] x86/microcode: Rendezvous and load in NMI Thomas Gleixner
2023-08-10 18:38 ` [patch 26/30] x86/microcode: Protect against instrumentation Thomas Gleixner
2023-08-10 20:36   ` Peter Zijlstra
2023-08-11  9:19     ` Thomas Gleixner
2023-08-10 18:38 ` Thomas Gleixner [this message]
2023-08-10 18:38 ` [patch 28/30] x86/microcode: Handle "offline" CPUs correctly Thomas Gleixner
2023-08-10 20:46   ` Peter Zijlstra
2023-08-10 20:50     ` Ashok Raj
2023-08-10 21:05       ` Peter Zijlstra
2023-08-10 21:49         ` Ashok Raj
2023-08-10 22:29           ` Peter Zijlstra
2023-08-10 23:02             ` Ashok Raj
2023-08-11  1:19               ` Thomas Gleixner
2023-08-11  1:31                 ` Thomas Gleixner
2023-08-11  9:36     ` Thomas Gleixner
2023-08-12 16:47       ` Thomas Gleixner
2023-08-10 18:38 ` [patch 29/30] x86/microcode: Prepare for minimal revision check Thomas Gleixner
2023-08-10 20:54   ` Peter Zijlstra
2023-08-11  9:38     ` Thomas Gleixner
2023-08-11 14:20     ` Arjan van de Ven
2023-08-10 18:38 ` [patch 30/30] x86/microcode/intel: Add a minimum required revision for late-loads Thomas Gleixner

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20230810160806.506010083@linutronix.de \
    --to=tglx@linutronix.de \
    --cc=arjan@linux.intel.com \
    --cc=ashok.raj@intel.com \
    --cc=bp@alien8.de \
    --cc=linux-kernel@vger.kernel.org \
    --cc=x86@kernel.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox

Powered by JetHome