lists.openwall.net   lists  /  announce  owl-users  owl-dev  john-users  john-dev  passwdqc-users  yescrypt  popa3d-users  /  oss-security  kernel-hardening  musl  sabotage  tlsify  passwords  /  crypt-dev  xvendor  /  Bugtraq  Full-Disclosure  linux-kernel  linux-netdev  linux-ext4  linux-hardening  linux-cve-announce  PHC 
Open Source and information security mailing list archives
 
Hash Suite: Windows password security audit tool. GUI, reports in PDF.
[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <20190731065949.GB1756@jagdpanzerIV>
Date:   Wed, 31 Jul 2019 15:59:49 +0900
From:   Sergey Senozhatsky <sergey.senozhatsky.work@...il.com>
To:     Petr Mladek <pmladek@...e.com>
Cc:     Konstantin Khlebnikov <koct9i@...il.com>,
        Sergey Senozhatsky <sergey.senozhatsky@...il.com>,
        Steven Rostedt <rostedt@...dmis.org>,
        Peter Zijlstra <peterz@...radead.org>, x86@...nel.org,
        John Ogness <john.ogness@...utronix.de>,
        Thomas Gleixner <tglx@...utronix.de>,
        Andrew Morton <akpm@...ux-foundation.org>,
        Petr Tesarik <ptesarik@...e.cz>, linux-kernel@...r.kernel.org,
        Sergey Senozhatsky <sergey.senozhatsky.work@...il.com>
Subject: Re: [PATCH 1/2] printk/panic: Access the main printk log in panic()
 only when safe

On (07/31/19 15:08), Sergey Senozhatsky wrote:
> When you have a chance, mind to take a look at the patch below?
> Doesn't look very difficult (half of it are white-spaces and
> comments, I believe).

I'm very sorry for annoyance.

Updated version:
-- passes !PRINTK build
-- moved WRITE_ONCE(console_waiter, false) in printk_enter_panic_mode()
-- added panic_in_progress_on_other_cpu() to console_trylock_spinning()

No more updates this week. Will wait for feedback.

Once again, sorry!

---
 include/linux/printk.h |  5 ++++
 kernel/panic.c         |  9 +++++-
 kernel/printk/printk.c | 64 ++++++++++++++++++++++++++++++++++++++++++
 3 files changed, 77 insertions(+), 1 deletion(-)

diff --git a/include/linux/printk.h b/include/linux/printk.h
index 57c9473f4a81..8293156d8243 100644
--- a/include/linux/printk.h
+++ b/include/linux/printk.h
@@ -245,6 +245,7 @@ extern asmlinkage void dump_stack(void) __cold;
 extern void printk_safe_init(void);
 extern void printk_safe_flush(void);
 extern void printk_safe_flush_on_panic(void);
+extern void printk_enter_panic_mode(int cpu);
 #else
 static inline __printf(1, 0)
 int vprintk(const char *s, va_list args)
@@ -320,6 +321,10 @@ static inline void printk_safe_flush(void)
 static inline void printk_safe_flush_on_panic(void)
 {
 }
+
+static inline void printk_enter_panic_mode(int cpu)
+{
+}
 #endif
 
 extern int kptr_restrict;
diff --git a/kernel/panic.c b/kernel/panic.c
index d1ece4c363b9..85fac975a90f 100644
--- a/kernel/panic.c
+++ b/kernel/panic.c
@@ -254,13 +254,20 @@ void panic(const char *fmt, ...)
 		crash_smp_send_stop();
 	}
 
+	/* Misbehaving secondary CPUs cannot printk() to the main logbuf now */
+	printk_enter_panic_mode(this_cpu);
+
 	/*
 	 * Run any panic handlers, including those that might need to
 	 * add information to the kmsg dump output.
 	 */
 	atomic_notifier_call_chain(&panic_notifier_list, 0, buf);
 
-	/* Call flush even twice. It tries harder with a single online CPU */
+	/*
+	 * Call flush even twice. It tries harder with a single online CPU.
+	 * Even if we failed to stop some of secondary CPUs we have printk
+	 * locks re-initialized and keep secondary CPUs off printk().
+	 */
 	printk_safe_flush_on_panic();
 	kmsg_dump(KMSG_DUMP_PANIC);
 
diff --git a/kernel/printk/printk.c b/kernel/printk/printk.c
index f0bc37a511a7..cd51aa7d08a9 100644
--- a/kernel/printk/printk.c
+++ b/kernel/printk/printk.c
@@ -1625,6 +1625,57 @@ static DEFINE_RAW_SPINLOCK(console_owner_lock);
 static struct task_struct *console_owner;
 static bool console_waiter;
 
+/*
+ * When system is in panic() this is used to permit printk() calls only from
+ * panic_cpu.
+ */
+static atomic_t __read_mostly printk_panic_cpu = ATOMIC_INIT(PANIC_CPU_INVALID);
+
+static int panic_in_progress_on_other_cpu(void)
+{
+	int cpu = atomic_read(&printk_panic_cpu);
+
+	return cpu != PANIC_CPU_INVALID && cpu != smp_processor_id();
+}
+
+void printk_enter_panic_mode(int cpu)
+{
+	unsigned long timeout;
+
+	cpu = atomic_cmpxchg(&printk_panic_cpu, PANIC_CPU_INVALID, cpu);
+	/* printk can enter panic mode only once */
+	if (cpu != PANIC_CPU_INVALID)
+		return;
+
+	WRITE_ONCE(console_waiter, false);
+
+	/*
+	 * Wait for active secondary CPUs (if there are any) to leave
+	 * console_unlock() printing loop (for up to one second).
+	 */
+	if (num_online_cpus() > 1) {
+		timeout = USEC_PER_SEC;
+		while (num_online_cpus() > 1 && timeout--)
+			udelay(1);
+	}
+
+	debug_locks_off();
+	/*
+	 * On some platforms crash_smp_send_stop() can kill CPUs via NMI
+	 * vector. Re-init printk() locks just in case if any of those killed
+	 * CPUs held any of printk() locks. On platforms which don't support
+	 * NMI stop, misbehaving secondary CPUs will be handled by
+	 * panic_in_progress_on_other_cpu() test.
+	 *
+	 * We re-init only printk() locks here. oops_in_progress is expected
+	 * to be set by now, so good console drivers are in lockless mode,
+	 * bad console drivers, however, can deadlock.
+	 */
+	raw_spin_lock_init(&logbuf_lock);
+	sema_init(&console_sem, 1);
+	raw_spin_lock_init(&console_owner_lock);
+}
+
 /**
  * console_lock_spinning_enable - mark beginning of code where another
  *	thread might safely busy wait
@@ -1739,6 +1790,10 @@ static int console_trylock_spinning(void)
 	spin_release(&console_owner_dep_map, 1, _THIS_IP_);
 
 	printk_safe_exit_irqrestore(flags);
+
+	if (panic_in_progress_on_other_cpu())
+		return 0;
+
 	/*
 	 * The owner passed the console lock to us.
 	 * Since we did not spin on console lock, annotate
@@ -1900,6 +1955,9 @@ int vprintk_store(int facility, int level,
 	size_t text_len;
 	enum log_flags lflags = 0;
 
+	if (panic_in_progress_on_other_cpu())
+		return 0;
+
 	/*
 	 * The printf needs to come first; we need the syslog
 	 * prefix which might be passed-in as a parameter.
@@ -2076,6 +2134,7 @@ static ssize_t msg_print_ext_body(char *buf, size_t size,
 				  char *text, size_t text_len) { return 0; }
 static void console_lock_spinning_enable(void) { }
 static int console_lock_spinning_disable_and_check(void) { return 0; }
+static int panic_in_progress_on_other_cpu(void) { return 0; }
 static void call_console_drivers(const char *ext_text, size_t ext_len,
 				 const char *text, size_t len) {}
 static size_t msg_print_text(const struct printk_log *msg, bool syslog,
@@ -2468,6 +2527,11 @@ void console_unlock(void)
 			return;
 		}
 
+		if (panic_in_progress_on_other_cpu()) {
+			printk_safe_exit_irqrestore(flags);
+			return;
+		}
+
 		printk_safe_exit_irqrestore(flags);
 
 		if (do_cond_resched)
-- 
2.22.0

Powered by blists - more mailing lists

Powered by Openwall GNU/*/Linux Powered by OpenVZ