From: Thomas Gleixner <tglx@linutronix.de>
To: LKML <linux-kernel@vger.kernel.org>
Cc: x86@kernel.org, Borislav Petkov <bp@alien8.de>
Subject: [patch V5 36/39] x86/apic: Provide apic_force_nmi_on_cpu()
Date: Tue, 17 Oct 2023 23:24:13 +0200 (CEST) [thread overview]
Message-ID: <20231017211723.968077065@linutronix.de> (raw)
In-Reply-To: 20231017200758.877560658@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 rendezvous 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 | 8 ++++++++
arch/x86/kernel/apic/x2apic_cluster.c | 1 +
arch/x86/kernel/apic/x2apic_phys.c | 1 +
5 files changed, 16 insertions(+), 1 deletion(-)
---
--- a/arch/x86/include/asm/apic.h
+++ b/arch/x86/include/asm/apic.h
@@ -276,7 +276,8 @@ struct apic {
u32 disable_esr : 1,
dest_mode_logical : 1,
- x2apic_set_max_apicid : 1;
+ x2apic_set_max_apicid : 1,
+ nmi_to_offline_cpu : 1;
u32 (*calc_dest_apicid)(unsigned int cpu);
@@ -542,6 +543,8 @@ extern bool default_check_apicid_used(ph
extern void default_ioapic_phys_id_map(physid_mask_t *phys_map, physid_mask_t *retmap);
extern int default_cpu_present_to_apicid(int mps_cpu);
+void apic_send_nmi_to_offline_cpu(unsigned int cpu);
+
#else /* CONFIG_X86_LOCAL_APIC */
static inline unsigned int read_apic_id(void) { return 0; }
--- a/arch/x86/kernel/apic/apic_flat_64.c
+++ b/arch/x86/kernel/apic/apic_flat_64.c
@@ -103,6 +103,7 @@ static struct apic apic_flat __ro_after_
.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,
.read = native_apic_mem_read,
.write = native_apic_mem_write,
@@ -175,6 +176,7 @@ static struct apic apic_physflat __ro_af
.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,
.read = native_apic_mem_read,
.write = native_apic_mem_write,
--- a/arch/x86/kernel/apic/ipi.c
+++ b/arch/x86/kernel/apic/ipi.c
@@ -97,6 +97,14 @@ void native_send_call_func_ipi(const str
__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)
--- a/arch/x86/kernel/apic/x2apic_cluster.c
+++ b/arch/x86/kernel/apic/x2apic_cluster.c
@@ -251,6 +251,7 @@ static struct apic apic_x2apic_cluster _
.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,
.read = native_apic_msr_read,
.write = native_apic_msr_write,
--- a/arch/x86/kernel/apic/x2apic_phys.c
+++ b/arch/x86/kernel/apic/x2apic_phys.c
@@ -166,6 +166,7 @@ static struct apic apic_x2apic_phys __ro
.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,
.read = native_apic_msr_read,
.write = native_apic_msr_write,
next prev parent reply other threads:[~2023-10-17 21:49 UTC|newest]
Thread overview: 70+ messages / expand[flat|nested] mbox.gz Atom feed top
2023-10-17 21:23 [patch V5 00/39] x86/microcode: Cleanup and late loading enhancements Thomas Gleixner
2023-10-17 21:23 ` [patch V5 01/39] x86/boot/32: Disable stackprotector and tracing for mk_early_pgtbl_32() Thomas Gleixner
2023-10-17 21:23 ` [patch V5 02/39] x86/boot: Use __pa_nodebug() in mk_early_pgtbl_32() Thomas Gleixner
2023-10-20 11:38 ` [tip: x86/microcode] " tip-bot2 for Thomas Gleixner
2023-10-17 21:23 ` [patch V5 03/39] x86/boot/32: De-uglify the 2/3 level paging difference " Thomas Gleixner
2023-10-18 10:00 ` Borislav Petkov
2023-10-18 13:20 ` Thomas Gleixner
2023-10-18 16:28 ` Borislav Petkov
2023-10-20 11:38 ` [tip: x86/microcode] " tip-bot2 for Thomas Gleixner
2023-10-17 21:23 ` [patch V5 04/39] x86/boot/32: Restructure mk_early_pgtbl_32() Thomas Gleixner
2023-10-20 11:38 ` [tip: x86/microcode] " tip-bot2 for Thomas Gleixner
2023-10-17 21:23 ` [patch V5 05/39] x86/microcode: Provide CONFIG_MICROCODE_INITRD32 Thomas Gleixner
2023-10-20 11:38 ` [tip: x86/microcode] " tip-bot2 for Thomas Gleixner
2023-10-17 21:23 ` [patch V5 06/39] x86/boot/32: Temporarily map initrd for microcode loading Thomas Gleixner
2023-10-20 11:38 ` [tip: x86/microcode] " tip-bot2 for Thomas Gleixner
2023-10-17 21:23 ` [patch V5 07/39] x86/microcode/32: Move early loading after paging enable Thomas Gleixner
2023-10-20 11:38 ` [tip: x86/microcode] " tip-bot2 for Thomas Gleixner
2023-10-17 21:23 ` [patch V5 08/39] x86/microcode/intel: Rip out mixed stepping support for Intel CPUs Thomas Gleixner
2023-10-20 11:38 ` [tip: x86/microcode] " tip-bot2 for Ashok Raj
2023-10-17 21:23 ` [patch V5 09/39] x86/microcode/intel: Simplify scan_microcode() Thomas Gleixner
2023-10-17 21:23 ` [patch V5 10/39] x86/microcode/intel: Simplify and rename generic_load_microcode() Thomas Gleixner
2023-10-17 21:23 ` [patch V5 11/39] x86/microcode/intel: Cleanup code further Thomas Gleixner
2023-10-17 21:23 ` [patch V5 12/39] x86/microcode/intel: Simplify early loading Thomas Gleixner
2023-10-20 11:38 ` [tip: x86/microcode] " tip-bot2 for Thomas Gleixner
2023-10-24 13:21 ` tip-bot2 for Thomas Gleixner
2023-10-17 21:23 ` [patch V5 13/39] x86/microcode/intel: Save the microcode only after a successful late-load Thomas Gleixner
2023-10-17 21:23 ` [patch V5 14/39] x86/microcode/intel: Switch to kvmalloc() Thomas Gleixner
2023-10-17 21:23 ` [patch V5 15/39] x86/microcode/intel: Unify microcode apply() functions Thomas Gleixner
2023-10-20 11:37 ` [tip: x86/microcode] " tip-bot2 for Thomas Gleixner
2023-10-24 13:21 ` tip-bot2 for Thomas Gleixner
2023-10-17 21:23 ` [patch V5 16/39] x86/microcode/intel: Rework intel_cpu_collect_info() Thomas Gleixner
2023-10-20 11:37 ` [tip: x86/microcode] " tip-bot2 for Thomas Gleixner
2023-10-24 13:21 ` tip-bot2 for Thomas Gleixner
2023-10-17 21:23 ` [patch V5 17/39] x86/microcode/intel: Reuse intel_cpu_collect_info() Thomas Gleixner
2023-10-17 21:23 ` [patch V5 18/39] x86/microcode/intel: Rework intel_find_matching_signature() Thomas Gleixner
2023-10-17 21:23 ` [patch V5 19/39] x86/microcode: Remove pointless apply() invocation Thomas Gleixner
2023-10-20 11:37 ` [tip: x86/microcode] " tip-bot2 for Thomas Gleixner
2023-10-24 13:21 ` tip-bot2 for Thomas Gleixner
2023-10-17 21:23 ` [patch V5 20/39] x86/microcode/amd: Use correct per CPU ucode_cpu_info Thomas Gleixner
2023-10-17 21:23 ` [patch V5 21/39] x86/microcode/amd: Cache builtin microcode too Thomas Gleixner
2023-10-17 21:23 ` [patch V5 22/39] x86/microcode/amd: Cache builtin/initrd microcode early Thomas Gleixner
2023-10-20 11:37 ` [tip: x86/microcode] " tip-bot2 for Thomas Gleixner
2023-10-24 13:21 ` tip-bot2 for Thomas Gleixner
2023-10-17 21:23 ` [patch V5 23/39] x86/microcode/amd: Use cached microcode for AP load Thomas Gleixner
2023-10-20 11:37 ` [tip: x86/microcode] " tip-bot2 for Thomas Gleixner
2023-10-24 13:21 ` tip-bot2 for Thomas Gleixner
2023-10-17 21:23 ` [patch V5 24/39] x86/microcode: Mop up early loading leftovers Thomas Gleixner
2023-10-20 11:37 ` [tip: x86/microcode] " tip-bot2 for Thomas Gleixner
2023-10-24 13:20 ` tip-bot2 for Thomas Gleixner
2023-10-17 21:23 ` [patch V5 25/39] x86/microcode: Get rid of the schedule work indirection Thomas Gleixner
2023-10-20 11:37 ` [tip: x86/microcode] " tip-bot2 for Thomas Gleixner
2023-10-24 13:20 ` tip-bot2 for Thomas Gleixner
2023-10-17 21:23 ` [patch V5 26/39] x86/microcode: Clean up mc_cpu_down_prep() Thomas Gleixner
2023-10-17 21:24 ` [patch V5 27/39] x86/microcode: Handle "nosmt" correctly Thomas Gleixner
2023-10-17 21:24 ` [patch V5 28/39] x86/microcode: Clarify the late load logic Thomas Gleixner
2023-10-17 21:24 ` [patch V5 29/39] x86/microcode: Sanitize __wait_for_cpus() Thomas Gleixner
2023-10-17 21:24 ` [patch V5 30/39] x86/microcode: Add per CPU result state Thomas Gleixner
2023-10-20 11:37 ` [tip: x86/microcode] " tip-bot2 for Thomas Gleixner
2023-10-24 13:20 ` tip-bot2 for Thomas Gleixner
2023-10-17 21:24 ` [patch V5 31/39] x86/microcode: Add per CPU control field Thomas Gleixner
2023-10-17 21:24 ` [patch V5 32/39] x86/microcode: Provide new control functions Thomas Gleixner
2023-10-17 21:24 ` [patch V5 33/39] x86/microcode: Replace the all-in-one rendevous handler Thomas Gleixner
2023-10-17 21:24 ` [patch V5 34/39] x86/microcode: Rendezvous and load in NMI Thomas Gleixner
2023-10-17 21:24 ` [patch V5 35/39] x86/microcode: Protect against instrumentation Thomas Gleixner
2023-10-17 21:24 ` Thomas Gleixner [this message]
2023-10-17 21:24 ` [patch V5 37/39] x86/microcode: Handle "offline" CPUs correctly Thomas Gleixner
2023-10-17 21:24 ` [patch V5 38/39] x86/microcode: Prepare for minimal revision check Thomas Gleixner
2023-10-20 11:37 ` [tip: x86/microcode] " tip-bot2 for Thomas Gleixner
2023-10-24 13:20 ` tip-bot2 for Thomas Gleixner
2023-10-17 21:24 ` [patch V5 39/39] x86/microcode/intel: Add a minimum required revision for late loading 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=20231017211723.968077065@linutronix.de \
--to=tglx@linutronix.de \
--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