linux-kernel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: tip-bot for Thomas Gleixner <tipbot@zytor.com>
To: linux-tip-commits@vger.kernel.org
Cc: tglx@linutronix.de, hpa@zytor.com, peterz@infradead.org,
	linux-kernel@vger.kernel.org, mingo@kernel.org,
	yinghai@kernel.org, bp@alien8.de
Subject: [tip:x86/timers] x86/tsc: Verify TSC_ADJUST from idle
Date: Tue, 29 Nov 2016 08:47:47 -0800	[thread overview]
Message-ID: <tip-cea7d48122b134e64d9617cc0a0ad5d53a407ea9@git.kernel.org> (raw)
In-Reply-To: <20161119134017.732180441@linutronix.de>

Commit-ID:  cea7d48122b134e64d9617cc0a0ad5d53a407ea9
Gitweb:     http://git.kernel.org/tip/cea7d48122b134e64d9617cc0a0ad5d53a407ea9
Author:     Thomas Gleixner <tglx@linutronix.de>
AuthorDate: Sat, 19 Nov 2016 13:47:37 +0000
Committer:  Thomas Gleixner <tglx@linutronix.de>
CommitDate: Tue, 29 Nov 2016 17:29:47 +0100

x86/tsc: Verify TSC_ADJUST from idle

When entering idle, it's a good oportunity to verify that the TSC_ADJUST
MSR has not been tampered with (BIOS hiding SMM cycles). If tampering is
detected, emit a warning and restore it to the previous value.

This is especially important for machines, which mark the TSC reliable
because there is no watchdog clocksource available (SoCs).

This is not sufficient for HPC (NOHZ_FULL) situations where a CPU never
goes idle, but adding a timer to do the check periodically is not an option
either. On a machine, which has this issue, the check triggeres right
during boot, so there is a decent chance that the sysadmin will notice.

Rate limit the check to once per second and warn only once per cpu.

Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Reviewed-by: Ingo Molnar <mingo@kernel.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Yinghai Lu <yinghai@kernel.org>
Cc: Borislav Petkov <bp@alien8.de>
Link: http://lkml.kernel.org/r/20161119134017.732180441@linutronix.de
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>

---
 arch/x86/include/asm/tsc.h |  2 ++
 arch/x86/kernel/process.c  |  1 +
 arch/x86/kernel/tsc_sync.c | 37 +++++++++++++++++++++++++++++++++++--
 3 files changed, 38 insertions(+), 2 deletions(-)

diff --git a/arch/x86/include/asm/tsc.h b/arch/x86/include/asm/tsc.h
index 1ec0595..b896e9e 100644
--- a/arch/x86/include/asm/tsc.h
+++ b/arch/x86/include/asm/tsc.h
@@ -50,8 +50,10 @@ extern void check_tsc_sync_target(void);
 
 #ifdef CONFIG_X86_TSC
 extern void tsc_store_and_check_tsc_adjust(void);
+extern void tsc_verify_tsc_adjust(void);
 #else
 static inline void tsc_store_and_check_tsc_adjust(void) { }
+static inline void tsc_verify_tsc_adjust(void) { }
 #endif
 
 extern int notsc_setup(char *);
diff --git a/arch/x86/kernel/process.c b/arch/x86/kernel/process.c
index 0888a87..4fe5dc8 100644
--- a/arch/x86/kernel/process.c
+++ b/arch/x86/kernel/process.c
@@ -277,6 +277,7 @@ void exit_idle(void)
 
 void arch_cpu_idle_enter(void)
 {
+	tsc_verify_tsc_adjust();
 	local_touch_nmi();
 	enter_idle();
 }
diff --git a/arch/x86/kernel/tsc_sync.c b/arch/x86/kernel/tsc_sync.c
index e4b2c04..f9c291e 100644
--- a/arch/x86/kernel/tsc_sync.c
+++ b/arch/x86/kernel/tsc_sync.c
@@ -22,12 +22,42 @@
 #include <asm/tsc.h>
 
 struct tsc_adjust {
-	s64	bootval;
-	s64	adjusted;
+	s64		bootval;
+	s64		adjusted;
+	unsigned long	nextcheck;
+	bool		warned;
 };
 
 static DEFINE_PER_CPU(struct tsc_adjust, tsc_adjust);
 
+void tsc_verify_tsc_adjust(void)
+{
+	struct tsc_adjust *adj = this_cpu_ptr(&tsc_adjust);
+	s64 curval;
+
+	if (!boot_cpu_has(X86_FEATURE_TSC_ADJUST))
+		return;
+
+	/* Rate limit the MSR check */
+	if (time_before(jiffies, adj->nextcheck))
+		return;
+
+	adj->nextcheck = jiffies + HZ;
+
+	rdmsrl(MSR_IA32_TSC_ADJUST, curval);
+	if (adj->adjusted == curval)
+		return;
+
+	/* Restore the original value */
+	wrmsrl(MSR_IA32_TSC_ADJUST, adj->adjusted);
+
+	if (!adj->warned) {
+		pr_warn(FW_BUG "TSC ADJUST differs: CPU%u %lld --> %lld. Restoring\n",
+			smp_processor_id(), adj->adjusted, curval);
+		adj->warned = true;
+	}
+}
+
 #ifndef CONFIG_SMP
 void __init tsc_store_and_check_tsc_adjust(void)
 {
@@ -40,6 +70,7 @@ void __init tsc_store_and_check_tsc_adjust(void)
 	rdmsrl(MSR_IA32_TSC_ADJUST, bootval);
 	cur->bootval = bootval;
 	cur->adjusted = bootval;
+	cur->nextcheck = jiffies + HZ;
 	pr_info("TSC ADJUST: Boot CPU%u: %lld\n",cpu, bootval);
 }
 
@@ -59,6 +90,8 @@ void tsc_store_and_check_tsc_adjust(void)
 
 	rdmsrl(MSR_IA32_TSC_ADJUST, bootval);
 	cur->bootval = bootval;
+	cur->nextcheck = jiffies + HZ;
+	cur->warned = false;
 
 	/*
 	 * Check whether this CPU is the first in a package to come up. In

  parent reply	other threads:[~2016-11-29 16:52 UTC|newest]

Thread overview: 36+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2016-11-19 13:47 [patch 0/8] x86/tsc: Utilize TSC_ADJUST MSR Thomas Gleixner
2016-11-19 13:47 ` [patch 1/8] x86/tsc: Use X86_FEATURE_TSC_ADJUST in detect_art() Thomas Gleixner
2016-11-29 16:46   ` [tip:x86/timers] " tip-bot for Thomas Gleixner
2016-11-29 18:28   ` tip-bot for Thomas Gleixner
2016-11-19 13:47 ` [patch 2/8] x86/tsc: Detect random warps Thomas Gleixner
2016-11-29 16:46   ` [tip:x86/timers] " tip-bot for Thomas Gleixner
2016-11-29 18:28   ` tip-bot for Thomas Gleixner
2016-11-19 13:47 ` [patch 3/8] x86/tsc: Store and check TSC ADJUST MSR Thomas Gleixner
2016-11-29 16:47   ` [tip:x86/timers] " tip-bot for Thomas Gleixner
2016-11-29 18:29   ` tip-bot for Thomas Gleixner
2017-01-27 13:36   ` [3/8] " Henning Schild
2017-01-30 10:20     ` Thomas Gleixner
2017-01-30 12:13       ` Henning Schild
2017-01-30 13:04         ` Thomas Gleixner
2017-01-30 15:58           ` Borislav Petkov
2016-11-19 13:47 ` [patch 4/8] x86/tsc: Verify TSC_ADJUST from idle Thomas Gleixner
2016-11-20 13:10   ` Peter Zijlstra
2016-11-21  8:16     ` Thomas Gleixner
2016-11-21 11:06       ` Peter Zijlstra
2016-11-29  9:17         ` Thomas Gleixner
2016-11-29 14:25           ` Ingo Molnar
2016-11-21 22:57   ` Andi Kleen
2016-11-29 16:47   ` tip-bot for Thomas Gleixner [this message]
2016-11-29 18:29   ` [tip:x86/timers] " tip-bot for Thomas Gleixner
2016-11-19 13:47 ` [patch 5/8] x86/tsc: Sync test only for the first cpu in a package Thomas Gleixner
2016-11-29 16:48   ` [tip:x86/timers] " tip-bot for Thomas Gleixner
2016-11-29 18:30   ` tip-bot for Thomas Gleixner
2016-11-19 13:47 ` [patch 6/8] x86/tsc: Move sync cleanup to a safe place Thomas Gleixner
2016-11-29 16:48   ` [tip:x86/timers] " tip-bot for Thomas Gleixner
2016-11-29 18:30   ` tip-bot for Thomas Gleixner
2016-11-19 13:47 ` [patch 7/8] x86/tsc: Prepare warp test for TSC adjustment Thomas Gleixner
2016-11-29 16:49   ` [tip:x86/timers] " tip-bot for Thomas Gleixner
2016-11-29 18:31   ` tip-bot for Thomas Gleixner
2016-11-19 13:47 ` [patch 8/8] x86/tsc: Try to adjust TSC if sync test fails Thomas Gleixner
2016-11-29 16:49   ` [tip:x86/timers] " tip-bot for Thomas Gleixner
2016-11-29 18:31   ` tip-bot for 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=tip-cea7d48122b134e64d9617cc0a0ad5d53a407ea9@git.kernel.org \
    --to=tipbot@zytor.com \
    --cc=bp@alien8.de \
    --cc=hpa@zytor.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-tip-commits@vger.kernel.org \
    --cc=mingo@kernel.org \
    --cc=peterz@infradead.org \
    --cc=tglx@linutronix.de \
    --cc=yinghai@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;
as well as URLs for NNTP newsgroup(s).