kernel_optimize_test/kernel/printk/printk_safe.c

292 lines
7.6 KiB
C
Raw Normal View History

printk/nmi: generic solution for safe printk in NMI printk() takes some locks and could not be used a safe way in NMI context. The chance of a deadlock is real especially when printing stacks from all CPUs. This particular problem has been addressed on x86 by the commit a9edc8809328 ("x86/nmi: Perform a safe NMI stack trace on all CPUs"). The patchset brings two big advantages. First, it makes the NMI backtraces safe on all architectures for free. Second, it makes all NMI messages almost safe on all architectures (the temporary buffer is limited. We still should keep the number of messages in NMI context at minimum). Note that there already are several messages printed in NMI context: WARN_ON(in_nmi()), BUG_ON(in_nmi()), anything being printed out from MCE handlers. These are not easy to avoid. This patch reuses most of the code and makes it generic. It is useful for all messages and architectures that support NMI. The alternative printk_func is set when entering and is reseted when leaving NMI context. It queues IRQ work to copy the messages into the main ring buffer in a safe context. __printk_nmi_flush() copies all available messages and reset the buffer. Then we could use a simple cmpxchg operations to get synchronized with writers. There is also used a spinlock to get synchronized with other flushers. We do not longer use seq_buf because it depends on external lock. It would be hard to make all supported operations safe for a lockless use. It would be confusing and error prone to make only some operations safe. The code is put into separate printk/nmi.c as suggested by Steven Rostedt. It needs a per-CPU buffer and is compiled only on architectures that call nmi_enter(). This is achieved by the new HAVE_NMI Kconfig flag. The are MN10300 and Xtensa architectures. We need to clean up NMI handling there first. Let's do it separately. The patch is heavily based on the draft from Peter Zijlstra, see https://lkml.org/lkml/2015/6/10/327 [arnd@arndb.de: printk-nmi: use %zu format string for size_t] [akpm@linux-foundation.org: min_t->min - all types are size_t here] Signed-off-by: Petr Mladek <pmladek@suse.com> Suggested-by: Peter Zijlstra <peterz@infradead.org> Suggested-by: Steven Rostedt <rostedt@goodmis.org> Cc: Jan Kara <jack@suse.cz> Acked-by: Russell King <rmk+kernel@arm.linux.org.uk> [arm part] Cc: Daniel Thompson <daniel.thompson@linaro.org> Cc: Jiri Kosina <jkosina@suse.com> Cc: Ingo Molnar <mingo@redhat.com> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: Ralf Baechle <ralf@linux-mips.org> Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org> Cc: Martin Schwidefsky <schwidefsky@de.ibm.com> Cc: David Miller <davem@davemloft.net> Cc: Daniel Thompson <daniel.thompson@linaro.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-05-21 08:00:33 +08:00
/*
* printk_safe.c - Safe printk in NMI context
printk/nmi: generic solution for safe printk in NMI printk() takes some locks and could not be used a safe way in NMI context. The chance of a deadlock is real especially when printing stacks from all CPUs. This particular problem has been addressed on x86 by the commit a9edc8809328 ("x86/nmi: Perform a safe NMI stack trace on all CPUs"). The patchset brings two big advantages. First, it makes the NMI backtraces safe on all architectures for free. Second, it makes all NMI messages almost safe on all architectures (the temporary buffer is limited. We still should keep the number of messages in NMI context at minimum). Note that there already are several messages printed in NMI context: WARN_ON(in_nmi()), BUG_ON(in_nmi()), anything being printed out from MCE handlers. These are not easy to avoid. This patch reuses most of the code and makes it generic. It is useful for all messages and architectures that support NMI. The alternative printk_func is set when entering and is reseted when leaving NMI context. It queues IRQ work to copy the messages into the main ring buffer in a safe context. __printk_nmi_flush() copies all available messages and reset the buffer. Then we could use a simple cmpxchg operations to get synchronized with writers. There is also used a spinlock to get synchronized with other flushers. We do not longer use seq_buf because it depends on external lock. It would be hard to make all supported operations safe for a lockless use. It would be confusing and error prone to make only some operations safe. The code is put into separate printk/nmi.c as suggested by Steven Rostedt. It needs a per-CPU buffer and is compiled only on architectures that call nmi_enter(). This is achieved by the new HAVE_NMI Kconfig flag. The are MN10300 and Xtensa architectures. We need to clean up NMI handling there first. Let's do it separately. The patch is heavily based on the draft from Peter Zijlstra, see https://lkml.org/lkml/2015/6/10/327 [arnd@arndb.de: printk-nmi: use %zu format string for size_t] [akpm@linux-foundation.org: min_t->min - all types are size_t here] Signed-off-by: Petr Mladek <pmladek@suse.com> Suggested-by: Peter Zijlstra <peterz@infradead.org> Suggested-by: Steven Rostedt <rostedt@goodmis.org> Cc: Jan Kara <jack@suse.cz> Acked-by: Russell King <rmk+kernel@arm.linux.org.uk> [arm part] Cc: Daniel Thompson <daniel.thompson@linaro.org> Cc: Jiri Kosina <jkosina@suse.com> Cc: Ingo Molnar <mingo@redhat.com> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: Ralf Baechle <ralf@linux-mips.org> Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org> Cc: Martin Schwidefsky <schwidefsky@de.ibm.com> Cc: David Miller <davem@davemloft.net> Cc: Daniel Thompson <daniel.thompson@linaro.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-05-21 08:00:33 +08:00
*
* This program is free software; you can redistribute it and/or
* modify it under the terms of the GNU General Public License
* as published by the Free Software Foundation; either version 2
* of the License, or (at your option) any later version.
*
* This program is distributed in the hope that it will be useful,
* but WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
* GNU General Public License for more details.
*
* You should have received a copy of the GNU General Public License
* along with this program; if not, see <http://www.gnu.org/licenses/>.
*/
#include <linux/preempt.h>
#include <linux/spinlock.h>
printk/nmi: flush NMI messages on the system panic In NMI context, printk() messages are stored into per-CPU buffers to avoid a possible deadlock. They are normally flushed to the main ring buffer via an IRQ work. But the work is never called when the system calls panic() in the very same NMI handler. This patch tries to flush NMI buffers before the crash dump is generated. In this case it does not risk a double release and bails out when the logbuf_lock is already taken. The aim is to get the messages into the main ring buffer when possible. It makes them better accessible in the vmcore. Then the patch tries to flush the buffers second time when other CPUs are down. It might be more aggressive and reset logbuf_lock. The aim is to get the messages available for the consequent kmsg_dump() and console_flush_on_panic() calls. The patch causes vprintk_emit() to be called even in NMI context again. But it is done via printk_deferred() so that the console handling is skipped. Consoles use internal locks and we could not prevent a deadlock easily. They are explicitly called later when the crash dump is not generated, see console_flush_on_panic(). Signed-off-by: Petr Mladek <pmladek@suse.com> Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org> Cc: Daniel Thompson <daniel.thompson@linaro.org> Cc: David Miller <davem@davemloft.net> Cc: Ingo Molnar <mingo@redhat.com> Cc: Jan Kara <jack@suse.cz> Cc: Jiri Kosina <jkosina@suse.com> Cc: Martin Schwidefsky <schwidefsky@de.ibm.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Ralf Baechle <ralf@linux-mips.org> Cc: Russell King <rmk+kernel@arm.linux.org.uk> Cc: Steven Rostedt <rostedt@goodmis.org> Cc: Thomas Gleixner <tglx@linutronix.de> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-05-21 08:00:42 +08:00
#include <linux/debug_locks.h>
printk/nmi: generic solution for safe printk in NMI printk() takes some locks and could not be used a safe way in NMI context. The chance of a deadlock is real especially when printing stacks from all CPUs. This particular problem has been addressed on x86 by the commit a9edc8809328 ("x86/nmi: Perform a safe NMI stack trace on all CPUs"). The patchset brings two big advantages. First, it makes the NMI backtraces safe on all architectures for free. Second, it makes all NMI messages almost safe on all architectures (the temporary buffer is limited. We still should keep the number of messages in NMI context at minimum). Note that there already are several messages printed in NMI context: WARN_ON(in_nmi()), BUG_ON(in_nmi()), anything being printed out from MCE handlers. These are not easy to avoid. This patch reuses most of the code and makes it generic. It is useful for all messages and architectures that support NMI. The alternative printk_func is set when entering and is reseted when leaving NMI context. It queues IRQ work to copy the messages into the main ring buffer in a safe context. __printk_nmi_flush() copies all available messages and reset the buffer. Then we could use a simple cmpxchg operations to get synchronized with writers. There is also used a spinlock to get synchronized with other flushers. We do not longer use seq_buf because it depends on external lock. It would be hard to make all supported operations safe for a lockless use. It would be confusing and error prone to make only some operations safe. The code is put into separate printk/nmi.c as suggested by Steven Rostedt. It needs a per-CPU buffer and is compiled only on architectures that call nmi_enter(). This is achieved by the new HAVE_NMI Kconfig flag. The are MN10300 and Xtensa architectures. We need to clean up NMI handling there first. Let's do it separately. The patch is heavily based on the draft from Peter Zijlstra, see https://lkml.org/lkml/2015/6/10/327 [arnd@arndb.de: printk-nmi: use %zu format string for size_t] [akpm@linux-foundation.org: min_t->min - all types are size_t here] Signed-off-by: Petr Mladek <pmladek@suse.com> Suggested-by: Peter Zijlstra <peterz@infradead.org> Suggested-by: Steven Rostedt <rostedt@goodmis.org> Cc: Jan Kara <jack@suse.cz> Acked-by: Russell King <rmk+kernel@arm.linux.org.uk> [arm part] Cc: Daniel Thompson <daniel.thompson@linaro.org> Cc: Jiri Kosina <jkosina@suse.com> Cc: Ingo Molnar <mingo@redhat.com> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: Ralf Baechle <ralf@linux-mips.org> Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org> Cc: Martin Schwidefsky <schwidefsky@de.ibm.com> Cc: David Miller <davem@davemloft.net> Cc: Daniel Thompson <daniel.thompson@linaro.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-05-21 08:00:33 +08:00
#include <linux/smp.h>
#include <linux/cpumask.h>
#include <linux/irq_work.h>
#include <linux/printk.h>
#include "internal.h"
/*
* printk() could not take logbuf_lock in NMI context. Instead,
* it uses an alternative implementation that temporary stores
* the strings into a per-CPU buffer. The content of the buffer
* is later flushed into the main ring buffer via IRQ work.
*
* The alternative implementation is chosen transparently
* via @printk_func per-CPU variable.
*
* The implementation allows to flush the strings also from another CPU.
* There are situations when we want to make sure that all buffers
* were handled or when IRQs are blocked.
*/
DEFINE_PER_CPU(printk_func_t, printk_func) = vprintk_default;
static int printk_safe_irq_ready;
atomic_t nmi_message_lost;
printk/nmi: generic solution for safe printk in NMI printk() takes some locks and could not be used a safe way in NMI context. The chance of a deadlock is real especially when printing stacks from all CPUs. This particular problem has been addressed on x86 by the commit a9edc8809328 ("x86/nmi: Perform a safe NMI stack trace on all CPUs"). The patchset brings two big advantages. First, it makes the NMI backtraces safe on all architectures for free. Second, it makes all NMI messages almost safe on all architectures (the temporary buffer is limited. We still should keep the number of messages in NMI context at minimum). Note that there already are several messages printed in NMI context: WARN_ON(in_nmi()), BUG_ON(in_nmi()), anything being printed out from MCE handlers. These are not easy to avoid. This patch reuses most of the code and makes it generic. It is useful for all messages and architectures that support NMI. The alternative printk_func is set when entering and is reseted when leaving NMI context. It queues IRQ work to copy the messages into the main ring buffer in a safe context. __printk_nmi_flush() copies all available messages and reset the buffer. Then we could use a simple cmpxchg operations to get synchronized with writers. There is also used a spinlock to get synchronized with other flushers. We do not longer use seq_buf because it depends on external lock. It would be hard to make all supported operations safe for a lockless use. It would be confusing and error prone to make only some operations safe. The code is put into separate printk/nmi.c as suggested by Steven Rostedt. It needs a per-CPU buffer and is compiled only on architectures that call nmi_enter(). This is achieved by the new HAVE_NMI Kconfig flag. The are MN10300 and Xtensa architectures. We need to clean up NMI handling there first. Let's do it separately. The patch is heavily based on the draft from Peter Zijlstra, see https://lkml.org/lkml/2015/6/10/327 [arnd@arndb.de: printk-nmi: use %zu format string for size_t] [akpm@linux-foundation.org: min_t->min - all types are size_t here] Signed-off-by: Petr Mladek <pmladek@suse.com> Suggested-by: Peter Zijlstra <peterz@infradead.org> Suggested-by: Steven Rostedt <rostedt@goodmis.org> Cc: Jan Kara <jack@suse.cz> Acked-by: Russell King <rmk+kernel@arm.linux.org.uk> [arm part] Cc: Daniel Thompson <daniel.thompson@linaro.org> Cc: Jiri Kosina <jkosina@suse.com> Cc: Ingo Molnar <mingo@redhat.com> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: Ralf Baechle <ralf@linux-mips.org> Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org> Cc: Martin Schwidefsky <schwidefsky@de.ibm.com> Cc: David Miller <davem@davemloft.net> Cc: Daniel Thompson <daniel.thompson@linaro.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-05-21 08:00:33 +08:00
#define SAFE_LOG_BUF_LEN ((1 << CONFIG_PRINTK_SAFE_LOG_BUF_SHIFT) - \
sizeof(atomic_t) - sizeof(struct irq_work))
printk/nmi: generic solution for safe printk in NMI printk() takes some locks and could not be used a safe way in NMI context. The chance of a deadlock is real especially when printing stacks from all CPUs. This particular problem has been addressed on x86 by the commit a9edc8809328 ("x86/nmi: Perform a safe NMI stack trace on all CPUs"). The patchset brings two big advantages. First, it makes the NMI backtraces safe on all architectures for free. Second, it makes all NMI messages almost safe on all architectures (the temporary buffer is limited. We still should keep the number of messages in NMI context at minimum). Note that there already are several messages printed in NMI context: WARN_ON(in_nmi()), BUG_ON(in_nmi()), anything being printed out from MCE handlers. These are not easy to avoid. This patch reuses most of the code and makes it generic. It is useful for all messages and architectures that support NMI. The alternative printk_func is set when entering and is reseted when leaving NMI context. It queues IRQ work to copy the messages into the main ring buffer in a safe context. __printk_nmi_flush() copies all available messages and reset the buffer. Then we could use a simple cmpxchg operations to get synchronized with writers. There is also used a spinlock to get synchronized with other flushers. We do not longer use seq_buf because it depends on external lock. It would be hard to make all supported operations safe for a lockless use. It would be confusing and error prone to make only some operations safe. The code is put into separate printk/nmi.c as suggested by Steven Rostedt. It needs a per-CPU buffer and is compiled only on architectures that call nmi_enter(). This is achieved by the new HAVE_NMI Kconfig flag. The are MN10300 and Xtensa architectures. We need to clean up NMI handling there first. Let's do it separately. The patch is heavily based on the draft from Peter Zijlstra, see https://lkml.org/lkml/2015/6/10/327 [arnd@arndb.de: printk-nmi: use %zu format string for size_t] [akpm@linux-foundation.org: min_t->min - all types are size_t here] Signed-off-by: Petr Mladek <pmladek@suse.com> Suggested-by: Peter Zijlstra <peterz@infradead.org> Suggested-by: Steven Rostedt <rostedt@goodmis.org> Cc: Jan Kara <jack@suse.cz> Acked-by: Russell King <rmk+kernel@arm.linux.org.uk> [arm part] Cc: Daniel Thompson <daniel.thompson@linaro.org> Cc: Jiri Kosina <jkosina@suse.com> Cc: Ingo Molnar <mingo@redhat.com> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: Ralf Baechle <ralf@linux-mips.org> Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org> Cc: Martin Schwidefsky <schwidefsky@de.ibm.com> Cc: David Miller <davem@davemloft.net> Cc: Daniel Thompson <daniel.thompson@linaro.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-05-21 08:00:33 +08:00
struct printk_safe_seq_buf {
printk/nmi: generic solution for safe printk in NMI printk() takes some locks and could not be used a safe way in NMI context. The chance of a deadlock is real especially when printing stacks from all CPUs. This particular problem has been addressed on x86 by the commit a9edc8809328 ("x86/nmi: Perform a safe NMI stack trace on all CPUs"). The patchset brings two big advantages. First, it makes the NMI backtraces safe on all architectures for free. Second, it makes all NMI messages almost safe on all architectures (the temporary buffer is limited. We still should keep the number of messages in NMI context at minimum). Note that there already are several messages printed in NMI context: WARN_ON(in_nmi()), BUG_ON(in_nmi()), anything being printed out from MCE handlers. These are not easy to avoid. This patch reuses most of the code and makes it generic. It is useful for all messages and architectures that support NMI. The alternative printk_func is set when entering and is reseted when leaving NMI context. It queues IRQ work to copy the messages into the main ring buffer in a safe context. __printk_nmi_flush() copies all available messages and reset the buffer. Then we could use a simple cmpxchg operations to get synchronized with writers. There is also used a spinlock to get synchronized with other flushers. We do not longer use seq_buf because it depends on external lock. It would be hard to make all supported operations safe for a lockless use. It would be confusing and error prone to make only some operations safe. The code is put into separate printk/nmi.c as suggested by Steven Rostedt. It needs a per-CPU buffer and is compiled only on architectures that call nmi_enter(). This is achieved by the new HAVE_NMI Kconfig flag. The are MN10300 and Xtensa architectures. We need to clean up NMI handling there first. Let's do it separately. The patch is heavily based on the draft from Peter Zijlstra, see https://lkml.org/lkml/2015/6/10/327 [arnd@arndb.de: printk-nmi: use %zu format string for size_t] [akpm@linux-foundation.org: min_t->min - all types are size_t here] Signed-off-by: Petr Mladek <pmladek@suse.com> Suggested-by: Peter Zijlstra <peterz@infradead.org> Suggested-by: Steven Rostedt <rostedt@goodmis.org> Cc: Jan Kara <jack@suse.cz> Acked-by: Russell King <rmk+kernel@arm.linux.org.uk> [arm part] Cc: Daniel Thompson <daniel.thompson@linaro.org> Cc: Jiri Kosina <jkosina@suse.com> Cc: Ingo Molnar <mingo@redhat.com> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: Ralf Baechle <ralf@linux-mips.org> Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org> Cc: Martin Schwidefsky <schwidefsky@de.ibm.com> Cc: David Miller <davem@davemloft.net> Cc: Daniel Thompson <daniel.thompson@linaro.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-05-21 08:00:33 +08:00
atomic_t len; /* length of written data */
struct irq_work work; /* IRQ work that flushes the buffer */
unsigned char buffer[SAFE_LOG_BUF_LEN];
printk/nmi: generic solution for safe printk in NMI printk() takes some locks and could not be used a safe way in NMI context. The chance of a deadlock is real especially when printing stacks from all CPUs. This particular problem has been addressed on x86 by the commit a9edc8809328 ("x86/nmi: Perform a safe NMI stack trace on all CPUs"). The patchset brings two big advantages. First, it makes the NMI backtraces safe on all architectures for free. Second, it makes all NMI messages almost safe on all architectures (the temporary buffer is limited. We still should keep the number of messages in NMI context at minimum). Note that there already are several messages printed in NMI context: WARN_ON(in_nmi()), BUG_ON(in_nmi()), anything being printed out from MCE handlers. These are not easy to avoid. This patch reuses most of the code and makes it generic. It is useful for all messages and architectures that support NMI. The alternative printk_func is set when entering and is reseted when leaving NMI context. It queues IRQ work to copy the messages into the main ring buffer in a safe context. __printk_nmi_flush() copies all available messages and reset the buffer. Then we could use a simple cmpxchg operations to get synchronized with writers. There is also used a spinlock to get synchronized with other flushers. We do not longer use seq_buf because it depends on external lock. It would be hard to make all supported operations safe for a lockless use. It would be confusing and error prone to make only some operations safe. The code is put into separate printk/nmi.c as suggested by Steven Rostedt. It needs a per-CPU buffer and is compiled only on architectures that call nmi_enter(). This is achieved by the new HAVE_NMI Kconfig flag. The are MN10300 and Xtensa architectures. We need to clean up NMI handling there first. Let's do it separately. The patch is heavily based on the draft from Peter Zijlstra, see https://lkml.org/lkml/2015/6/10/327 [arnd@arndb.de: printk-nmi: use %zu format string for size_t] [akpm@linux-foundation.org: min_t->min - all types are size_t here] Signed-off-by: Petr Mladek <pmladek@suse.com> Suggested-by: Peter Zijlstra <peterz@infradead.org> Suggested-by: Steven Rostedt <rostedt@goodmis.org> Cc: Jan Kara <jack@suse.cz> Acked-by: Russell King <rmk+kernel@arm.linux.org.uk> [arm part] Cc: Daniel Thompson <daniel.thompson@linaro.org> Cc: Jiri Kosina <jkosina@suse.com> Cc: Ingo Molnar <mingo@redhat.com> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: Ralf Baechle <ralf@linux-mips.org> Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org> Cc: Martin Schwidefsky <schwidefsky@de.ibm.com> Cc: David Miller <davem@davemloft.net> Cc: Daniel Thompson <daniel.thompson@linaro.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-05-21 08:00:33 +08:00
};
static DEFINE_PER_CPU(struct printk_safe_seq_buf, nmi_print_seq);
printk/nmi: generic solution for safe printk in NMI printk() takes some locks and could not be used a safe way in NMI context. The chance of a deadlock is real especially when printing stacks from all CPUs. This particular problem has been addressed on x86 by the commit a9edc8809328 ("x86/nmi: Perform a safe NMI stack trace on all CPUs"). The patchset brings two big advantages. First, it makes the NMI backtraces safe on all architectures for free. Second, it makes all NMI messages almost safe on all architectures (the temporary buffer is limited. We still should keep the number of messages in NMI context at minimum). Note that there already are several messages printed in NMI context: WARN_ON(in_nmi()), BUG_ON(in_nmi()), anything being printed out from MCE handlers. These are not easy to avoid. This patch reuses most of the code and makes it generic. It is useful for all messages and architectures that support NMI. The alternative printk_func is set when entering and is reseted when leaving NMI context. It queues IRQ work to copy the messages into the main ring buffer in a safe context. __printk_nmi_flush() copies all available messages and reset the buffer. Then we could use a simple cmpxchg operations to get synchronized with writers. There is also used a spinlock to get synchronized with other flushers. We do not longer use seq_buf because it depends on external lock. It would be hard to make all supported operations safe for a lockless use. It would be confusing and error prone to make only some operations safe. The code is put into separate printk/nmi.c as suggested by Steven Rostedt. It needs a per-CPU buffer and is compiled only on architectures that call nmi_enter(). This is achieved by the new HAVE_NMI Kconfig flag. The are MN10300 and Xtensa architectures. We need to clean up NMI handling there first. Let's do it separately. The patch is heavily based on the draft from Peter Zijlstra, see https://lkml.org/lkml/2015/6/10/327 [arnd@arndb.de: printk-nmi: use %zu format string for size_t] [akpm@linux-foundation.org: min_t->min - all types are size_t here] Signed-off-by: Petr Mladek <pmladek@suse.com> Suggested-by: Peter Zijlstra <peterz@infradead.org> Suggested-by: Steven Rostedt <rostedt@goodmis.org> Cc: Jan Kara <jack@suse.cz> Acked-by: Russell King <rmk+kernel@arm.linux.org.uk> [arm part] Cc: Daniel Thompson <daniel.thompson@linaro.org> Cc: Jiri Kosina <jkosina@suse.com> Cc: Ingo Molnar <mingo@redhat.com> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: Ralf Baechle <ralf@linux-mips.org> Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org> Cc: Martin Schwidefsky <schwidefsky@de.ibm.com> Cc: David Miller <davem@davemloft.net> Cc: Daniel Thompson <daniel.thompson@linaro.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-05-21 08:00:33 +08:00
/*
* Safe printk() for NMI context. It uses a per-CPU buffer to
* store the message. NMIs are not nested, so there is always only
* one writer running. But the buffer might get flushed from another
* CPU, so we need to be careful.
*/
static int vprintk_nmi(const char *fmt, va_list args)
printk/nmi: generic solution for safe printk in NMI printk() takes some locks and could not be used a safe way in NMI context. The chance of a deadlock is real especially when printing stacks from all CPUs. This particular problem has been addressed on x86 by the commit a9edc8809328 ("x86/nmi: Perform a safe NMI stack trace on all CPUs"). The patchset brings two big advantages. First, it makes the NMI backtraces safe on all architectures for free. Second, it makes all NMI messages almost safe on all architectures (the temporary buffer is limited. We still should keep the number of messages in NMI context at minimum). Note that there already are several messages printed in NMI context: WARN_ON(in_nmi()), BUG_ON(in_nmi()), anything being printed out from MCE handlers. These are not easy to avoid. This patch reuses most of the code and makes it generic. It is useful for all messages and architectures that support NMI. The alternative printk_func is set when entering and is reseted when leaving NMI context. It queues IRQ work to copy the messages into the main ring buffer in a safe context. __printk_nmi_flush() copies all available messages and reset the buffer. Then we could use a simple cmpxchg operations to get synchronized with writers. There is also used a spinlock to get synchronized with other flushers. We do not longer use seq_buf because it depends on external lock. It would be hard to make all supported operations safe for a lockless use. It would be confusing and error prone to make only some operations safe. The code is put into separate printk/nmi.c as suggested by Steven Rostedt. It needs a per-CPU buffer and is compiled only on architectures that call nmi_enter(). This is achieved by the new HAVE_NMI Kconfig flag. The are MN10300 and Xtensa architectures. We need to clean up NMI handling there first. Let's do it separately. The patch is heavily based on the draft from Peter Zijlstra, see https://lkml.org/lkml/2015/6/10/327 [arnd@arndb.de: printk-nmi: use %zu format string for size_t] [akpm@linux-foundation.org: min_t->min - all types are size_t here] Signed-off-by: Petr Mladek <pmladek@suse.com> Suggested-by: Peter Zijlstra <peterz@infradead.org> Suggested-by: Steven Rostedt <rostedt@goodmis.org> Cc: Jan Kara <jack@suse.cz> Acked-by: Russell King <rmk+kernel@arm.linux.org.uk> [arm part] Cc: Daniel Thompson <daniel.thompson@linaro.org> Cc: Jiri Kosina <jkosina@suse.com> Cc: Ingo Molnar <mingo@redhat.com> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: Ralf Baechle <ralf@linux-mips.org> Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org> Cc: Martin Schwidefsky <schwidefsky@de.ibm.com> Cc: David Miller <davem@davemloft.net> Cc: Daniel Thompson <daniel.thompson@linaro.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-05-21 08:00:33 +08:00
{
struct printk_safe_seq_buf *s = this_cpu_ptr(&nmi_print_seq);
printk/nmi: generic solution for safe printk in NMI printk() takes some locks and could not be used a safe way in NMI context. The chance of a deadlock is real especially when printing stacks from all CPUs. This particular problem has been addressed on x86 by the commit a9edc8809328 ("x86/nmi: Perform a safe NMI stack trace on all CPUs"). The patchset brings two big advantages. First, it makes the NMI backtraces safe on all architectures for free. Second, it makes all NMI messages almost safe on all architectures (the temporary buffer is limited. We still should keep the number of messages in NMI context at minimum). Note that there already are several messages printed in NMI context: WARN_ON(in_nmi()), BUG_ON(in_nmi()), anything being printed out from MCE handlers. These are not easy to avoid. This patch reuses most of the code and makes it generic. It is useful for all messages and architectures that support NMI. The alternative printk_func is set when entering and is reseted when leaving NMI context. It queues IRQ work to copy the messages into the main ring buffer in a safe context. __printk_nmi_flush() copies all available messages and reset the buffer. Then we could use a simple cmpxchg operations to get synchronized with writers. There is also used a spinlock to get synchronized with other flushers. We do not longer use seq_buf because it depends on external lock. It would be hard to make all supported operations safe for a lockless use. It would be confusing and error prone to make only some operations safe. The code is put into separate printk/nmi.c as suggested by Steven Rostedt. It needs a per-CPU buffer and is compiled only on architectures that call nmi_enter(). This is achieved by the new HAVE_NMI Kconfig flag. The are MN10300 and Xtensa architectures. We need to clean up NMI handling there first. Let's do it separately. The patch is heavily based on the draft from Peter Zijlstra, see https://lkml.org/lkml/2015/6/10/327 [arnd@arndb.de: printk-nmi: use %zu format string for size_t] [akpm@linux-foundation.org: min_t->min - all types are size_t here] Signed-off-by: Petr Mladek <pmladek@suse.com> Suggested-by: Peter Zijlstra <peterz@infradead.org> Suggested-by: Steven Rostedt <rostedt@goodmis.org> Cc: Jan Kara <jack@suse.cz> Acked-by: Russell King <rmk+kernel@arm.linux.org.uk> [arm part] Cc: Daniel Thompson <daniel.thompson@linaro.org> Cc: Jiri Kosina <jkosina@suse.com> Cc: Ingo Molnar <mingo@redhat.com> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: Ralf Baechle <ralf@linux-mips.org> Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org> Cc: Martin Schwidefsky <schwidefsky@de.ibm.com> Cc: David Miller <davem@davemloft.net> Cc: Daniel Thompson <daniel.thompson@linaro.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-05-21 08:00:33 +08:00
int add = 0;
size_t len;
again:
len = atomic_read(&s->len);
printk/NMI: fix up handling of the full nmi log buffer vsnprintf() adds the trailing '\0' but it does not count it into the number of printed characters. The result is that there is one byte less space for the real characters in the buffer. The broken check for the free space might cause that we will repeatedly try to print 1 character into the buffer, never reach the full buffer, and do not count the messages as missed. Also vsnprintf() returns the number of characters that would be printed if the buffer was big enough. As a result, s->len might be bigger than the size of the buffer[*]. And the printk() function might return bigger len than it really printed. Both problems are fixed by using vscnprintf() instead. Note that I though about increasing the number of missed messages even when the message was shrunken. But it made the code even more complicated. I think that it is not worth it. Shrunken messages are usually easy to recognize. And it should be a corner case. [*] The overflown s->len value is crazy and unexpected. I "made a mistake" and reported this situation as an internal error when fixed handling of PR_CONT headers in some other patch. Link: http://lkml.kernel.org/r/20161208174912.GA17042@linux.suse Signed-off-by: Petr Mladek <pmladek@suse.com> CcL Sergey Senozhatsky <sergey.senozhatsky.work@gmail.com> Cc: Chris Mason <clm@fb.com> Cc: David Sterba <dsterba@suse.com> Cc: Jason Wessel <jason.wessel@windriver.com> Cc: Josef Bacik <jbacik@fb.com> Cc: Joe Perches <joe@perches.com> Cc: Jaroslav Kysela <perex@perex.cz> Cc: Steven Rostedt <rostedt@goodmis.org> Cc: Takashi Iwai <tiwai@suse.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-12-13 08:45:40 +08:00
/* The trailing '\0' is not counted into len. */
if (len >= sizeof(s->buffer) - 1) {
atomic_inc(&nmi_message_lost);
printk/nmi: generic solution for safe printk in NMI printk() takes some locks and could not be used a safe way in NMI context. The chance of a deadlock is real especially when printing stacks from all CPUs. This particular problem has been addressed on x86 by the commit a9edc8809328 ("x86/nmi: Perform a safe NMI stack trace on all CPUs"). The patchset brings two big advantages. First, it makes the NMI backtraces safe on all architectures for free. Second, it makes all NMI messages almost safe on all architectures (the temporary buffer is limited. We still should keep the number of messages in NMI context at minimum). Note that there already are several messages printed in NMI context: WARN_ON(in_nmi()), BUG_ON(in_nmi()), anything being printed out from MCE handlers. These are not easy to avoid. This patch reuses most of the code and makes it generic. It is useful for all messages and architectures that support NMI. The alternative printk_func is set when entering and is reseted when leaving NMI context. It queues IRQ work to copy the messages into the main ring buffer in a safe context. __printk_nmi_flush() copies all available messages and reset the buffer. Then we could use a simple cmpxchg operations to get synchronized with writers. There is also used a spinlock to get synchronized with other flushers. We do not longer use seq_buf because it depends on external lock. It would be hard to make all supported operations safe for a lockless use. It would be confusing and error prone to make only some operations safe. The code is put into separate printk/nmi.c as suggested by Steven Rostedt. It needs a per-CPU buffer and is compiled only on architectures that call nmi_enter(). This is achieved by the new HAVE_NMI Kconfig flag. The are MN10300 and Xtensa architectures. We need to clean up NMI handling there first. Let's do it separately. The patch is heavily based on the draft from Peter Zijlstra, see https://lkml.org/lkml/2015/6/10/327 [arnd@arndb.de: printk-nmi: use %zu format string for size_t] [akpm@linux-foundation.org: min_t->min - all types are size_t here] Signed-off-by: Petr Mladek <pmladek@suse.com> Suggested-by: Peter Zijlstra <peterz@infradead.org> Suggested-by: Steven Rostedt <rostedt@goodmis.org> Cc: Jan Kara <jack@suse.cz> Acked-by: Russell King <rmk+kernel@arm.linux.org.uk> [arm part] Cc: Daniel Thompson <daniel.thompson@linaro.org> Cc: Jiri Kosina <jkosina@suse.com> Cc: Ingo Molnar <mingo@redhat.com> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: Ralf Baechle <ralf@linux-mips.org> Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org> Cc: Martin Schwidefsky <schwidefsky@de.ibm.com> Cc: David Miller <davem@davemloft.net> Cc: Daniel Thompson <daniel.thompson@linaro.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-05-21 08:00:33 +08:00
return 0;
}
printk/nmi: generic solution for safe printk in NMI printk() takes some locks and could not be used a safe way in NMI context. The chance of a deadlock is real especially when printing stacks from all CPUs. This particular problem has been addressed on x86 by the commit a9edc8809328 ("x86/nmi: Perform a safe NMI stack trace on all CPUs"). The patchset brings two big advantages. First, it makes the NMI backtraces safe on all architectures for free. Second, it makes all NMI messages almost safe on all architectures (the temporary buffer is limited. We still should keep the number of messages in NMI context at minimum). Note that there already are several messages printed in NMI context: WARN_ON(in_nmi()), BUG_ON(in_nmi()), anything being printed out from MCE handlers. These are not easy to avoid. This patch reuses most of the code and makes it generic. It is useful for all messages and architectures that support NMI. The alternative printk_func is set when entering and is reseted when leaving NMI context. It queues IRQ work to copy the messages into the main ring buffer in a safe context. __printk_nmi_flush() copies all available messages and reset the buffer. Then we could use a simple cmpxchg operations to get synchronized with writers. There is also used a spinlock to get synchronized with other flushers. We do not longer use seq_buf because it depends on external lock. It would be hard to make all supported operations safe for a lockless use. It would be confusing and error prone to make only some operations safe. The code is put into separate printk/nmi.c as suggested by Steven Rostedt. It needs a per-CPU buffer and is compiled only on architectures that call nmi_enter(). This is achieved by the new HAVE_NMI Kconfig flag. The are MN10300 and Xtensa architectures. We need to clean up NMI handling there first. Let's do it separately. The patch is heavily based on the draft from Peter Zijlstra, see https://lkml.org/lkml/2015/6/10/327 [arnd@arndb.de: printk-nmi: use %zu format string for size_t] [akpm@linux-foundation.org: min_t->min - all types are size_t here] Signed-off-by: Petr Mladek <pmladek@suse.com> Suggested-by: Peter Zijlstra <peterz@infradead.org> Suggested-by: Steven Rostedt <rostedt@goodmis.org> Cc: Jan Kara <jack@suse.cz> Acked-by: Russell King <rmk+kernel@arm.linux.org.uk> [arm part] Cc: Daniel Thompson <daniel.thompson@linaro.org> Cc: Jiri Kosina <jkosina@suse.com> Cc: Ingo Molnar <mingo@redhat.com> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: Ralf Baechle <ralf@linux-mips.org> Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org> Cc: Martin Schwidefsky <schwidefsky@de.ibm.com> Cc: David Miller <davem@davemloft.net> Cc: Daniel Thompson <daniel.thompson@linaro.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-05-21 08:00:33 +08:00
/*
* Make sure that all old data have been read before the buffer was
* reseted. This is not needed when we just append data.
*/
if (!len)
smp_rmb();
printk/NMI: fix up handling of the full nmi log buffer vsnprintf() adds the trailing '\0' but it does not count it into the number of printed characters. The result is that there is one byte less space for the real characters in the buffer. The broken check for the free space might cause that we will repeatedly try to print 1 character into the buffer, never reach the full buffer, and do not count the messages as missed. Also vsnprintf() returns the number of characters that would be printed if the buffer was big enough. As a result, s->len might be bigger than the size of the buffer[*]. And the printk() function might return bigger len than it really printed. Both problems are fixed by using vscnprintf() instead. Note that I though about increasing the number of missed messages even when the message was shrunken. But it made the code even more complicated. I think that it is not worth it. Shrunken messages are usually easy to recognize. And it should be a corner case. [*] The overflown s->len value is crazy and unexpected. I "made a mistake" and reported this situation as an internal error when fixed handling of PR_CONT headers in some other patch. Link: http://lkml.kernel.org/r/20161208174912.GA17042@linux.suse Signed-off-by: Petr Mladek <pmladek@suse.com> CcL Sergey Senozhatsky <sergey.senozhatsky.work@gmail.com> Cc: Chris Mason <clm@fb.com> Cc: David Sterba <dsterba@suse.com> Cc: Jason Wessel <jason.wessel@windriver.com> Cc: Josef Bacik <jbacik@fb.com> Cc: Joe Perches <joe@perches.com> Cc: Jaroslav Kysela <perex@perex.cz> Cc: Steven Rostedt <rostedt@goodmis.org> Cc: Takashi Iwai <tiwai@suse.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-12-13 08:45:40 +08:00
add = vscnprintf(s->buffer + len, sizeof(s->buffer) - len, fmt, args);
printk/nmi: generic solution for safe printk in NMI printk() takes some locks and could not be used a safe way in NMI context. The chance of a deadlock is real especially when printing stacks from all CPUs. This particular problem has been addressed on x86 by the commit a9edc8809328 ("x86/nmi: Perform a safe NMI stack trace on all CPUs"). The patchset brings two big advantages. First, it makes the NMI backtraces safe on all architectures for free. Second, it makes all NMI messages almost safe on all architectures (the temporary buffer is limited. We still should keep the number of messages in NMI context at minimum). Note that there already are several messages printed in NMI context: WARN_ON(in_nmi()), BUG_ON(in_nmi()), anything being printed out from MCE handlers. These are not easy to avoid. This patch reuses most of the code and makes it generic. It is useful for all messages and architectures that support NMI. The alternative printk_func is set when entering and is reseted when leaving NMI context. It queues IRQ work to copy the messages into the main ring buffer in a safe context. __printk_nmi_flush() copies all available messages and reset the buffer. Then we could use a simple cmpxchg operations to get synchronized with writers. There is also used a spinlock to get synchronized with other flushers. We do not longer use seq_buf because it depends on external lock. It would be hard to make all supported operations safe for a lockless use. It would be confusing and error prone to make only some operations safe. The code is put into separate printk/nmi.c as suggested by Steven Rostedt. It needs a per-CPU buffer and is compiled only on architectures that call nmi_enter(). This is achieved by the new HAVE_NMI Kconfig flag. The are MN10300 and Xtensa architectures. We need to clean up NMI handling there first. Let's do it separately. The patch is heavily based on the draft from Peter Zijlstra, see https://lkml.org/lkml/2015/6/10/327 [arnd@arndb.de: printk-nmi: use %zu format string for size_t] [akpm@linux-foundation.org: min_t->min - all types are size_t here] Signed-off-by: Petr Mladek <pmladek@suse.com> Suggested-by: Peter Zijlstra <peterz@infradead.org> Suggested-by: Steven Rostedt <rostedt@goodmis.org> Cc: Jan Kara <jack@suse.cz> Acked-by: Russell King <rmk+kernel@arm.linux.org.uk> [arm part] Cc: Daniel Thompson <daniel.thompson@linaro.org> Cc: Jiri Kosina <jkosina@suse.com> Cc: Ingo Molnar <mingo@redhat.com> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: Ralf Baechle <ralf@linux-mips.org> Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org> Cc: Martin Schwidefsky <schwidefsky@de.ibm.com> Cc: David Miller <davem@davemloft.net> Cc: Daniel Thompson <daniel.thompson@linaro.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-05-21 08:00:33 +08:00
/*
* Do it once again if the buffer has been flushed in the meantime.
* Note that atomic_cmpxchg() is an implicit memory barrier that
* makes sure that the data were written before updating s->len.
*/
if (atomic_cmpxchg(&s->len, len, len + add) != len)
goto again;
/* Get flushed in a more safe context. */
if (add && printk_safe_irq_ready) {
printk/nmi: generic solution for safe printk in NMI printk() takes some locks and could not be used a safe way in NMI context. The chance of a deadlock is real especially when printing stacks from all CPUs. This particular problem has been addressed on x86 by the commit a9edc8809328 ("x86/nmi: Perform a safe NMI stack trace on all CPUs"). The patchset brings two big advantages. First, it makes the NMI backtraces safe on all architectures for free. Second, it makes all NMI messages almost safe on all architectures (the temporary buffer is limited. We still should keep the number of messages in NMI context at minimum). Note that there already are several messages printed in NMI context: WARN_ON(in_nmi()), BUG_ON(in_nmi()), anything being printed out from MCE handlers. These are not easy to avoid. This patch reuses most of the code and makes it generic. It is useful for all messages and architectures that support NMI. The alternative printk_func is set when entering and is reseted when leaving NMI context. It queues IRQ work to copy the messages into the main ring buffer in a safe context. __printk_nmi_flush() copies all available messages and reset the buffer. Then we could use a simple cmpxchg operations to get synchronized with writers. There is also used a spinlock to get synchronized with other flushers. We do not longer use seq_buf because it depends on external lock. It would be hard to make all supported operations safe for a lockless use. It would be confusing and error prone to make only some operations safe. The code is put into separate printk/nmi.c as suggested by Steven Rostedt. It needs a per-CPU buffer and is compiled only on architectures that call nmi_enter(). This is achieved by the new HAVE_NMI Kconfig flag. The are MN10300 and Xtensa architectures. We need to clean up NMI handling there first. Let's do it separately. The patch is heavily based on the draft from Peter Zijlstra, see https://lkml.org/lkml/2015/6/10/327 [arnd@arndb.de: printk-nmi: use %zu format string for size_t] [akpm@linux-foundation.org: min_t->min - all types are size_t here] Signed-off-by: Petr Mladek <pmladek@suse.com> Suggested-by: Peter Zijlstra <peterz@infradead.org> Suggested-by: Steven Rostedt <rostedt@goodmis.org> Cc: Jan Kara <jack@suse.cz> Acked-by: Russell King <rmk+kernel@arm.linux.org.uk> [arm part] Cc: Daniel Thompson <daniel.thompson@linaro.org> Cc: Jiri Kosina <jkosina@suse.com> Cc: Ingo Molnar <mingo@redhat.com> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: Ralf Baechle <ralf@linux-mips.org> Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org> Cc: Martin Schwidefsky <schwidefsky@de.ibm.com> Cc: David Miller <davem@davemloft.net> Cc: Daniel Thompson <daniel.thompson@linaro.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-05-21 08:00:33 +08:00
/* Make sure that IRQ work is really initialized. */
smp_rmb();
irq_work_queue(&s->work);
}
return add;
}
static void printk_safe_flush_line(const char *text, int len)
printk/nmi: generic solution for safe printk in NMI printk() takes some locks and could not be used a safe way in NMI context. The chance of a deadlock is real especially when printing stacks from all CPUs. This particular problem has been addressed on x86 by the commit a9edc8809328 ("x86/nmi: Perform a safe NMI stack trace on all CPUs"). The patchset brings two big advantages. First, it makes the NMI backtraces safe on all architectures for free. Second, it makes all NMI messages almost safe on all architectures (the temporary buffer is limited. We still should keep the number of messages in NMI context at minimum). Note that there already are several messages printed in NMI context: WARN_ON(in_nmi()), BUG_ON(in_nmi()), anything being printed out from MCE handlers. These are not easy to avoid. This patch reuses most of the code and makes it generic. It is useful for all messages and architectures that support NMI. The alternative printk_func is set when entering and is reseted when leaving NMI context. It queues IRQ work to copy the messages into the main ring buffer in a safe context. __printk_nmi_flush() copies all available messages and reset the buffer. Then we could use a simple cmpxchg operations to get synchronized with writers. There is also used a spinlock to get synchronized with other flushers. We do not longer use seq_buf because it depends on external lock. It would be hard to make all supported operations safe for a lockless use. It would be confusing and error prone to make only some operations safe. The code is put into separate printk/nmi.c as suggested by Steven Rostedt. It needs a per-CPU buffer and is compiled only on architectures that call nmi_enter(). This is achieved by the new HAVE_NMI Kconfig flag. The are MN10300 and Xtensa architectures. We need to clean up NMI handling there first. Let's do it separately. The patch is heavily based on the draft from Peter Zijlstra, see https://lkml.org/lkml/2015/6/10/327 [arnd@arndb.de: printk-nmi: use %zu format string for size_t] [akpm@linux-foundation.org: min_t->min - all types are size_t here] Signed-off-by: Petr Mladek <pmladek@suse.com> Suggested-by: Peter Zijlstra <peterz@infradead.org> Suggested-by: Steven Rostedt <rostedt@goodmis.org> Cc: Jan Kara <jack@suse.cz> Acked-by: Russell King <rmk+kernel@arm.linux.org.uk> [arm part] Cc: Daniel Thompson <daniel.thompson@linaro.org> Cc: Jiri Kosina <jkosina@suse.com> Cc: Ingo Molnar <mingo@redhat.com> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: Ralf Baechle <ralf@linux-mips.org> Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org> Cc: Martin Schwidefsky <schwidefsky@de.ibm.com> Cc: David Miller <davem@davemloft.net> Cc: Daniel Thompson <daniel.thompson@linaro.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-05-21 08:00:33 +08:00
{
printk/nmi: flush NMI messages on the system panic In NMI context, printk() messages are stored into per-CPU buffers to avoid a possible deadlock. They are normally flushed to the main ring buffer via an IRQ work. But the work is never called when the system calls panic() in the very same NMI handler. This patch tries to flush NMI buffers before the crash dump is generated. In this case it does not risk a double release and bails out when the logbuf_lock is already taken. The aim is to get the messages into the main ring buffer when possible. It makes them better accessible in the vmcore. Then the patch tries to flush the buffers second time when other CPUs are down. It might be more aggressive and reset logbuf_lock. The aim is to get the messages available for the consequent kmsg_dump() and console_flush_on_panic() calls. The patch causes vprintk_emit() to be called even in NMI context again. But it is done via printk_deferred() so that the console handling is skipped. Consoles use internal locks and we could not prevent a deadlock easily. They are explicitly called later when the crash dump is not generated, see console_flush_on_panic(). Signed-off-by: Petr Mladek <pmladek@suse.com> Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org> Cc: Daniel Thompson <daniel.thompson@linaro.org> Cc: David Miller <davem@davemloft.net> Cc: Ingo Molnar <mingo@redhat.com> Cc: Jan Kara <jack@suse.cz> Cc: Jiri Kosina <jkosina@suse.com> Cc: Martin Schwidefsky <schwidefsky@de.ibm.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Ralf Baechle <ralf@linux-mips.org> Cc: Russell King <rmk+kernel@arm.linux.org.uk> Cc: Steven Rostedt <rostedt@goodmis.org> Cc: Thomas Gleixner <tglx@linutronix.de> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-05-21 08:00:42 +08:00
/*
* The buffers are flushed in NMI only on panic. The messages must
* go only into the ring buffer at this stage. Consoles will get
* explicitly called later when a crashdump is not generated.
*/
if (in_nmi())
printk_deferred("%.*s", len, text);
printk/nmi: flush NMI messages on the system panic In NMI context, printk() messages are stored into per-CPU buffers to avoid a possible deadlock. They are normally flushed to the main ring buffer via an IRQ work. But the work is never called when the system calls panic() in the very same NMI handler. This patch tries to flush NMI buffers before the crash dump is generated. In this case it does not risk a double release and bails out when the logbuf_lock is already taken. The aim is to get the messages into the main ring buffer when possible. It makes them better accessible in the vmcore. Then the patch tries to flush the buffers second time when other CPUs are down. It might be more aggressive and reset logbuf_lock. The aim is to get the messages available for the consequent kmsg_dump() and console_flush_on_panic() calls. The patch causes vprintk_emit() to be called even in NMI context again. But it is done via printk_deferred() so that the console handling is skipped. Consoles use internal locks and we could not prevent a deadlock easily. They are explicitly called later when the crash dump is not generated, see console_flush_on_panic(). Signed-off-by: Petr Mladek <pmladek@suse.com> Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org> Cc: Daniel Thompson <daniel.thompson@linaro.org> Cc: David Miller <davem@davemloft.net> Cc: Ingo Molnar <mingo@redhat.com> Cc: Jan Kara <jack@suse.cz> Cc: Jiri Kosina <jkosina@suse.com> Cc: Martin Schwidefsky <schwidefsky@de.ibm.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Ralf Baechle <ralf@linux-mips.org> Cc: Russell King <rmk+kernel@arm.linux.org.uk> Cc: Steven Rostedt <rostedt@goodmis.org> Cc: Thomas Gleixner <tglx@linutronix.de> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-05-21 08:00:42 +08:00
else
printk("%.*s", len, text);
printk/nmi: generic solution for safe printk in NMI printk() takes some locks and could not be used a safe way in NMI context. The chance of a deadlock is real especially when printing stacks from all CPUs. This particular problem has been addressed on x86 by the commit a9edc8809328 ("x86/nmi: Perform a safe NMI stack trace on all CPUs"). The patchset brings two big advantages. First, it makes the NMI backtraces safe on all architectures for free. Second, it makes all NMI messages almost safe on all architectures (the temporary buffer is limited. We still should keep the number of messages in NMI context at minimum). Note that there already are several messages printed in NMI context: WARN_ON(in_nmi()), BUG_ON(in_nmi()), anything being printed out from MCE handlers. These are not easy to avoid. This patch reuses most of the code and makes it generic. It is useful for all messages and architectures that support NMI. The alternative printk_func is set when entering and is reseted when leaving NMI context. It queues IRQ work to copy the messages into the main ring buffer in a safe context. __printk_nmi_flush() copies all available messages and reset the buffer. Then we could use a simple cmpxchg operations to get synchronized with writers. There is also used a spinlock to get synchronized with other flushers. We do not longer use seq_buf because it depends on external lock. It would be hard to make all supported operations safe for a lockless use. It would be confusing and error prone to make only some operations safe. The code is put into separate printk/nmi.c as suggested by Steven Rostedt. It needs a per-CPU buffer and is compiled only on architectures that call nmi_enter(). This is achieved by the new HAVE_NMI Kconfig flag. The are MN10300 and Xtensa architectures. We need to clean up NMI handling there first. Let's do it separately. The patch is heavily based on the draft from Peter Zijlstra, see https://lkml.org/lkml/2015/6/10/327 [arnd@arndb.de: printk-nmi: use %zu format string for size_t] [akpm@linux-foundation.org: min_t->min - all types are size_t here] Signed-off-by: Petr Mladek <pmladek@suse.com> Suggested-by: Peter Zijlstra <peterz@infradead.org> Suggested-by: Steven Rostedt <rostedt@goodmis.org> Cc: Jan Kara <jack@suse.cz> Acked-by: Russell King <rmk+kernel@arm.linux.org.uk> [arm part] Cc: Daniel Thompson <daniel.thompson@linaro.org> Cc: Jiri Kosina <jkosina@suse.com> Cc: Ingo Molnar <mingo@redhat.com> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: Ralf Baechle <ralf@linux-mips.org> Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org> Cc: Martin Schwidefsky <schwidefsky@de.ibm.com> Cc: David Miller <davem@davemloft.net> Cc: Daniel Thompson <daniel.thompson@linaro.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-05-21 08:00:33 +08:00
}
printk/NMI: handle continuous lines and missing newline Commit 4bcc595ccd80 ("printk: reinstate KERN_CONT for printing continuation lines") added back KERN_CONT message header. As a result it might appear in the middle of the line when the parts are squashed via the temporary NMI buffer. A reasonable solution seems to be to split the text in the NNI temporary not only by newlines but also by the message headers. Another solution would be to filter out KERN_CONT when writing to the temporary buffer. But this would complicate the lockless handling. Also it would not solve problems with a missing newline that was there even before the KERN_CONT stuff. This patch moves the temporary buffer handling into separate function. I played with it and it seems that using the char pointers make the code easier to read. Also it prints the final newline as a continuous line. Finally, it moves handling of the s->len overflow into the paranoid check. And allows to recover from the disaster. Link: http://lkml.kernel.org/r/1478695291-12169-2-git-send-email-pmladek@suse.com Signed-off-by: Petr Mladek <pmladek@suse.com> Reviewed-by: Sergey Senozhatsky <sergey.senozhatsky@gmail.com> Cc: Joe Perches <joe@perches.com> Cc: Steven Rostedt <rostedt@goodmis.org> Cc: Jason Wessel <jason.wessel@windriver.com> Cc: Jaroslav Kysela <perex@perex.cz> Cc: Takashi Iwai <tiwai@suse.com> Cc: Chris Mason <clm@fb.com> Cc: Josef Bacik <jbacik@fb.com> Cc: David Sterba <dsterba@suse.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-12-13 08:45:44 +08:00
/* printk part of the temporary buffer line by line */
static int printk_safe_flush_buffer(const char *start, size_t len)
{
printk/NMI: handle continuous lines and missing newline Commit 4bcc595ccd80 ("printk: reinstate KERN_CONT for printing continuation lines") added back KERN_CONT message header. As a result it might appear in the middle of the line when the parts are squashed via the temporary NMI buffer. A reasonable solution seems to be to split the text in the NNI temporary not only by newlines but also by the message headers. Another solution would be to filter out KERN_CONT when writing to the temporary buffer. But this would complicate the lockless handling. Also it would not solve problems with a missing newline that was there even before the KERN_CONT stuff. This patch moves the temporary buffer handling into separate function. I played with it and it seems that using the char pointers make the code easier to read. Also it prints the final newline as a continuous line. Finally, it moves handling of the s->len overflow into the paranoid check. And allows to recover from the disaster. Link: http://lkml.kernel.org/r/1478695291-12169-2-git-send-email-pmladek@suse.com Signed-off-by: Petr Mladek <pmladek@suse.com> Reviewed-by: Sergey Senozhatsky <sergey.senozhatsky@gmail.com> Cc: Joe Perches <joe@perches.com> Cc: Steven Rostedt <rostedt@goodmis.org> Cc: Jason Wessel <jason.wessel@windriver.com> Cc: Jaroslav Kysela <perex@perex.cz> Cc: Takashi Iwai <tiwai@suse.com> Cc: Chris Mason <clm@fb.com> Cc: Josef Bacik <jbacik@fb.com> Cc: David Sterba <dsterba@suse.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-12-13 08:45:44 +08:00
const char *c, *end;
bool header;
c = start;
end = start + len;
header = true;
/* Print line by line. */
while (c < end) {
if (*c == '\n') {
printk_safe_flush_line(start, c - start + 1);
printk/NMI: handle continuous lines and missing newline Commit 4bcc595ccd80 ("printk: reinstate KERN_CONT for printing continuation lines") added back KERN_CONT message header. As a result it might appear in the middle of the line when the parts are squashed via the temporary NMI buffer. A reasonable solution seems to be to split the text in the NNI temporary not only by newlines but also by the message headers. Another solution would be to filter out KERN_CONT when writing to the temporary buffer. But this would complicate the lockless handling. Also it would not solve problems with a missing newline that was there even before the KERN_CONT stuff. This patch moves the temporary buffer handling into separate function. I played with it and it seems that using the char pointers make the code easier to read. Also it prints the final newline as a continuous line. Finally, it moves handling of the s->len overflow into the paranoid check. And allows to recover from the disaster. Link: http://lkml.kernel.org/r/1478695291-12169-2-git-send-email-pmladek@suse.com Signed-off-by: Petr Mladek <pmladek@suse.com> Reviewed-by: Sergey Senozhatsky <sergey.senozhatsky@gmail.com> Cc: Joe Perches <joe@perches.com> Cc: Steven Rostedt <rostedt@goodmis.org> Cc: Jason Wessel <jason.wessel@windriver.com> Cc: Jaroslav Kysela <perex@perex.cz> Cc: Takashi Iwai <tiwai@suse.com> Cc: Chris Mason <clm@fb.com> Cc: Josef Bacik <jbacik@fb.com> Cc: David Sterba <dsterba@suse.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-12-13 08:45:44 +08:00
start = ++c;
header = true;
continue;
}
/* Handle continuous lines or missing new line. */
if ((c + 1 < end) && printk_get_level(c)) {
if (header) {
c = printk_skip_level(c);
continue;
}
printk_safe_flush_line(start, c - start);
printk/NMI: handle continuous lines and missing newline Commit 4bcc595ccd80 ("printk: reinstate KERN_CONT for printing continuation lines") added back KERN_CONT message header. As a result it might appear in the middle of the line when the parts are squashed via the temporary NMI buffer. A reasonable solution seems to be to split the text in the NNI temporary not only by newlines but also by the message headers. Another solution would be to filter out KERN_CONT when writing to the temporary buffer. But this would complicate the lockless handling. Also it would not solve problems with a missing newline that was there even before the KERN_CONT stuff. This patch moves the temporary buffer handling into separate function. I played with it and it seems that using the char pointers make the code easier to read. Also it prints the final newline as a continuous line. Finally, it moves handling of the s->len overflow into the paranoid check. And allows to recover from the disaster. Link: http://lkml.kernel.org/r/1478695291-12169-2-git-send-email-pmladek@suse.com Signed-off-by: Petr Mladek <pmladek@suse.com> Reviewed-by: Sergey Senozhatsky <sergey.senozhatsky@gmail.com> Cc: Joe Perches <joe@perches.com> Cc: Steven Rostedt <rostedt@goodmis.org> Cc: Jason Wessel <jason.wessel@windriver.com> Cc: Jaroslav Kysela <perex@perex.cz> Cc: Takashi Iwai <tiwai@suse.com> Cc: Chris Mason <clm@fb.com> Cc: Josef Bacik <jbacik@fb.com> Cc: David Sterba <dsterba@suse.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-12-13 08:45:44 +08:00
start = c++;
header = true;
continue;
}
header = false;
c++;
}
printk/NMI: handle continuous lines and missing newline Commit 4bcc595ccd80 ("printk: reinstate KERN_CONT for printing continuation lines") added back KERN_CONT message header. As a result it might appear in the middle of the line when the parts are squashed via the temporary NMI buffer. A reasonable solution seems to be to split the text in the NNI temporary not only by newlines but also by the message headers. Another solution would be to filter out KERN_CONT when writing to the temporary buffer. But this would complicate the lockless handling. Also it would not solve problems with a missing newline that was there even before the KERN_CONT stuff. This patch moves the temporary buffer handling into separate function. I played with it and it seems that using the char pointers make the code easier to read. Also it prints the final newline as a continuous line. Finally, it moves handling of the s->len overflow into the paranoid check. And allows to recover from the disaster. Link: http://lkml.kernel.org/r/1478695291-12169-2-git-send-email-pmladek@suse.com Signed-off-by: Petr Mladek <pmladek@suse.com> Reviewed-by: Sergey Senozhatsky <sergey.senozhatsky@gmail.com> Cc: Joe Perches <joe@perches.com> Cc: Steven Rostedt <rostedt@goodmis.org> Cc: Jason Wessel <jason.wessel@windriver.com> Cc: Jaroslav Kysela <perex@perex.cz> Cc: Takashi Iwai <tiwai@suse.com> Cc: Chris Mason <clm@fb.com> Cc: Josef Bacik <jbacik@fb.com> Cc: David Sterba <dsterba@suse.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-12-13 08:45:44 +08:00
/* Check if there was a partial line. Ignore pure header. */
if (start < end && !header) {
static const char newline[] = KERN_CONT "\n";
printk_safe_flush_line(start, end - start);
printk_safe_flush_line(newline, strlen(newline));
printk/NMI: handle continuous lines and missing newline Commit 4bcc595ccd80 ("printk: reinstate KERN_CONT for printing continuation lines") added back KERN_CONT message header. As a result it might appear in the middle of the line when the parts are squashed via the temporary NMI buffer. A reasonable solution seems to be to split the text in the NNI temporary not only by newlines but also by the message headers. Another solution would be to filter out KERN_CONT when writing to the temporary buffer. But this would complicate the lockless handling. Also it would not solve problems with a missing newline that was there even before the KERN_CONT stuff. This patch moves the temporary buffer handling into separate function. I played with it and it seems that using the char pointers make the code easier to read. Also it prints the final newline as a continuous line. Finally, it moves handling of the s->len overflow into the paranoid check. And allows to recover from the disaster. Link: http://lkml.kernel.org/r/1478695291-12169-2-git-send-email-pmladek@suse.com Signed-off-by: Petr Mladek <pmladek@suse.com> Reviewed-by: Sergey Senozhatsky <sergey.senozhatsky@gmail.com> Cc: Joe Perches <joe@perches.com> Cc: Steven Rostedt <rostedt@goodmis.org> Cc: Jason Wessel <jason.wessel@windriver.com> Cc: Jaroslav Kysela <perex@perex.cz> Cc: Takashi Iwai <tiwai@suse.com> Cc: Chris Mason <clm@fb.com> Cc: Josef Bacik <jbacik@fb.com> Cc: David Sterba <dsterba@suse.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-12-13 08:45:44 +08:00
}
return len;
}
printk/nmi: generic solution for safe printk in NMI printk() takes some locks and could not be used a safe way in NMI context. The chance of a deadlock is real especially when printing stacks from all CPUs. This particular problem has been addressed on x86 by the commit a9edc8809328 ("x86/nmi: Perform a safe NMI stack trace on all CPUs"). The patchset brings two big advantages. First, it makes the NMI backtraces safe on all architectures for free. Second, it makes all NMI messages almost safe on all architectures (the temporary buffer is limited. We still should keep the number of messages in NMI context at minimum). Note that there already are several messages printed in NMI context: WARN_ON(in_nmi()), BUG_ON(in_nmi()), anything being printed out from MCE handlers. These are not easy to avoid. This patch reuses most of the code and makes it generic. It is useful for all messages and architectures that support NMI. The alternative printk_func is set when entering and is reseted when leaving NMI context. It queues IRQ work to copy the messages into the main ring buffer in a safe context. __printk_nmi_flush() copies all available messages and reset the buffer. Then we could use a simple cmpxchg operations to get synchronized with writers. There is also used a spinlock to get synchronized with other flushers. We do not longer use seq_buf because it depends on external lock. It would be hard to make all supported operations safe for a lockless use. It would be confusing and error prone to make only some operations safe. The code is put into separate printk/nmi.c as suggested by Steven Rostedt. It needs a per-CPU buffer and is compiled only on architectures that call nmi_enter(). This is achieved by the new HAVE_NMI Kconfig flag. The are MN10300 and Xtensa architectures. We need to clean up NMI handling there first. Let's do it separately. The patch is heavily based on the draft from Peter Zijlstra, see https://lkml.org/lkml/2015/6/10/327 [arnd@arndb.de: printk-nmi: use %zu format string for size_t] [akpm@linux-foundation.org: min_t->min - all types are size_t here] Signed-off-by: Petr Mladek <pmladek@suse.com> Suggested-by: Peter Zijlstra <peterz@infradead.org> Suggested-by: Steven Rostedt <rostedt@goodmis.org> Cc: Jan Kara <jack@suse.cz> Acked-by: Russell King <rmk+kernel@arm.linux.org.uk> [arm part] Cc: Daniel Thompson <daniel.thompson@linaro.org> Cc: Jiri Kosina <jkosina@suse.com> Cc: Ingo Molnar <mingo@redhat.com> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: Ralf Baechle <ralf@linux-mips.org> Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org> Cc: Martin Schwidefsky <schwidefsky@de.ibm.com> Cc: David Miller <davem@davemloft.net> Cc: Daniel Thompson <daniel.thompson@linaro.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-05-21 08:00:33 +08:00
/*
* Flush data from the associated per_CPU buffer. The function
* can be called either via IRQ work or independently.
*/
static void __printk_safe_flush(struct irq_work *work)
printk/nmi: generic solution for safe printk in NMI printk() takes some locks and could not be used a safe way in NMI context. The chance of a deadlock is real especially when printing stacks from all CPUs. This particular problem has been addressed on x86 by the commit a9edc8809328 ("x86/nmi: Perform a safe NMI stack trace on all CPUs"). The patchset brings two big advantages. First, it makes the NMI backtraces safe on all architectures for free. Second, it makes all NMI messages almost safe on all architectures (the temporary buffer is limited. We still should keep the number of messages in NMI context at minimum). Note that there already are several messages printed in NMI context: WARN_ON(in_nmi()), BUG_ON(in_nmi()), anything being printed out from MCE handlers. These are not easy to avoid. This patch reuses most of the code and makes it generic. It is useful for all messages and architectures that support NMI. The alternative printk_func is set when entering and is reseted when leaving NMI context. It queues IRQ work to copy the messages into the main ring buffer in a safe context. __printk_nmi_flush() copies all available messages and reset the buffer. Then we could use a simple cmpxchg operations to get synchronized with writers. There is also used a spinlock to get synchronized with other flushers. We do not longer use seq_buf because it depends on external lock. It would be hard to make all supported operations safe for a lockless use. It would be confusing and error prone to make only some operations safe. The code is put into separate printk/nmi.c as suggested by Steven Rostedt. It needs a per-CPU buffer and is compiled only on architectures that call nmi_enter(). This is achieved by the new HAVE_NMI Kconfig flag. The are MN10300 and Xtensa architectures. We need to clean up NMI handling there first. Let's do it separately. The patch is heavily based on the draft from Peter Zijlstra, see https://lkml.org/lkml/2015/6/10/327 [arnd@arndb.de: printk-nmi: use %zu format string for size_t] [akpm@linux-foundation.org: min_t->min - all types are size_t here] Signed-off-by: Petr Mladek <pmladek@suse.com> Suggested-by: Peter Zijlstra <peterz@infradead.org> Suggested-by: Steven Rostedt <rostedt@goodmis.org> Cc: Jan Kara <jack@suse.cz> Acked-by: Russell King <rmk+kernel@arm.linux.org.uk> [arm part] Cc: Daniel Thompson <daniel.thompson@linaro.org> Cc: Jiri Kosina <jkosina@suse.com> Cc: Ingo Molnar <mingo@redhat.com> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: Ralf Baechle <ralf@linux-mips.org> Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org> Cc: Martin Schwidefsky <schwidefsky@de.ibm.com> Cc: David Miller <davem@davemloft.net> Cc: Daniel Thompson <daniel.thompson@linaro.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-05-21 08:00:33 +08:00
{
static raw_spinlock_t read_lock =
__RAW_SPIN_LOCK_INITIALIZER(read_lock);
struct printk_safe_seq_buf *s =
container_of(work, struct printk_safe_seq_buf, work);
printk/nmi: generic solution for safe printk in NMI printk() takes some locks and could not be used a safe way in NMI context. The chance of a deadlock is real especially when printing stacks from all CPUs. This particular problem has been addressed on x86 by the commit a9edc8809328 ("x86/nmi: Perform a safe NMI stack trace on all CPUs"). The patchset brings two big advantages. First, it makes the NMI backtraces safe on all architectures for free. Second, it makes all NMI messages almost safe on all architectures (the temporary buffer is limited. We still should keep the number of messages in NMI context at minimum). Note that there already are several messages printed in NMI context: WARN_ON(in_nmi()), BUG_ON(in_nmi()), anything being printed out from MCE handlers. These are not easy to avoid. This patch reuses most of the code and makes it generic. It is useful for all messages and architectures that support NMI. The alternative printk_func is set when entering and is reseted when leaving NMI context. It queues IRQ work to copy the messages into the main ring buffer in a safe context. __printk_nmi_flush() copies all available messages and reset the buffer. Then we could use a simple cmpxchg operations to get synchronized with writers. There is also used a spinlock to get synchronized with other flushers. We do not longer use seq_buf because it depends on external lock. It would be hard to make all supported operations safe for a lockless use. It would be confusing and error prone to make only some operations safe. The code is put into separate printk/nmi.c as suggested by Steven Rostedt. It needs a per-CPU buffer and is compiled only on architectures that call nmi_enter(). This is achieved by the new HAVE_NMI Kconfig flag. The are MN10300 and Xtensa architectures. We need to clean up NMI handling there first. Let's do it separately. The patch is heavily based on the draft from Peter Zijlstra, see https://lkml.org/lkml/2015/6/10/327 [arnd@arndb.de: printk-nmi: use %zu format string for size_t] [akpm@linux-foundation.org: min_t->min - all types are size_t here] Signed-off-by: Petr Mladek <pmladek@suse.com> Suggested-by: Peter Zijlstra <peterz@infradead.org> Suggested-by: Steven Rostedt <rostedt@goodmis.org> Cc: Jan Kara <jack@suse.cz> Acked-by: Russell King <rmk+kernel@arm.linux.org.uk> [arm part] Cc: Daniel Thompson <daniel.thompson@linaro.org> Cc: Jiri Kosina <jkosina@suse.com> Cc: Ingo Molnar <mingo@redhat.com> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: Ralf Baechle <ralf@linux-mips.org> Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org> Cc: Martin Schwidefsky <schwidefsky@de.ibm.com> Cc: David Miller <davem@davemloft.net> Cc: Daniel Thompson <daniel.thompson@linaro.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-05-21 08:00:33 +08:00
unsigned long flags;
printk/NMI: handle continuous lines and missing newline Commit 4bcc595ccd80 ("printk: reinstate KERN_CONT for printing continuation lines") added back KERN_CONT message header. As a result it might appear in the middle of the line when the parts are squashed via the temporary NMI buffer. A reasonable solution seems to be to split the text in the NNI temporary not only by newlines but also by the message headers. Another solution would be to filter out KERN_CONT when writing to the temporary buffer. But this would complicate the lockless handling. Also it would not solve problems with a missing newline that was there even before the KERN_CONT stuff. This patch moves the temporary buffer handling into separate function. I played with it and it seems that using the char pointers make the code easier to read. Also it prints the final newline as a continuous line. Finally, it moves handling of the s->len overflow into the paranoid check. And allows to recover from the disaster. Link: http://lkml.kernel.org/r/1478695291-12169-2-git-send-email-pmladek@suse.com Signed-off-by: Petr Mladek <pmladek@suse.com> Reviewed-by: Sergey Senozhatsky <sergey.senozhatsky@gmail.com> Cc: Joe Perches <joe@perches.com> Cc: Steven Rostedt <rostedt@goodmis.org> Cc: Jason Wessel <jason.wessel@windriver.com> Cc: Jaroslav Kysela <perex@perex.cz> Cc: Takashi Iwai <tiwai@suse.com> Cc: Chris Mason <clm@fb.com> Cc: Josef Bacik <jbacik@fb.com> Cc: David Sterba <dsterba@suse.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-12-13 08:45:44 +08:00
size_t len;
int i;
printk/nmi: generic solution for safe printk in NMI printk() takes some locks and could not be used a safe way in NMI context. The chance of a deadlock is real especially when printing stacks from all CPUs. This particular problem has been addressed on x86 by the commit a9edc8809328 ("x86/nmi: Perform a safe NMI stack trace on all CPUs"). The patchset brings two big advantages. First, it makes the NMI backtraces safe on all architectures for free. Second, it makes all NMI messages almost safe on all architectures (the temporary buffer is limited. We still should keep the number of messages in NMI context at minimum). Note that there already are several messages printed in NMI context: WARN_ON(in_nmi()), BUG_ON(in_nmi()), anything being printed out from MCE handlers. These are not easy to avoid. This patch reuses most of the code and makes it generic. It is useful for all messages and architectures that support NMI. The alternative printk_func is set when entering and is reseted when leaving NMI context. It queues IRQ work to copy the messages into the main ring buffer in a safe context. __printk_nmi_flush() copies all available messages and reset the buffer. Then we could use a simple cmpxchg operations to get synchronized with writers. There is also used a spinlock to get synchronized with other flushers. We do not longer use seq_buf because it depends on external lock. It would be hard to make all supported operations safe for a lockless use. It would be confusing and error prone to make only some operations safe. The code is put into separate printk/nmi.c as suggested by Steven Rostedt. It needs a per-CPU buffer and is compiled only on architectures that call nmi_enter(). This is achieved by the new HAVE_NMI Kconfig flag. The are MN10300 and Xtensa architectures. We need to clean up NMI handling there first. Let's do it separately. The patch is heavily based on the draft from Peter Zijlstra, see https://lkml.org/lkml/2015/6/10/327 [arnd@arndb.de: printk-nmi: use %zu format string for size_t] [akpm@linux-foundation.org: min_t->min - all types are size_t here] Signed-off-by: Petr Mladek <pmladek@suse.com> Suggested-by: Peter Zijlstra <peterz@infradead.org> Suggested-by: Steven Rostedt <rostedt@goodmis.org> Cc: Jan Kara <jack@suse.cz> Acked-by: Russell King <rmk+kernel@arm.linux.org.uk> [arm part] Cc: Daniel Thompson <daniel.thompson@linaro.org> Cc: Jiri Kosina <jkosina@suse.com> Cc: Ingo Molnar <mingo@redhat.com> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: Ralf Baechle <ralf@linux-mips.org> Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org> Cc: Martin Schwidefsky <schwidefsky@de.ibm.com> Cc: David Miller <davem@davemloft.net> Cc: Daniel Thompson <daniel.thompson@linaro.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-05-21 08:00:33 +08:00
/*
* The lock has two functions. First, one reader has to flush all
* available message to make the lockless synchronization with
* writers easier. Second, we do not want to mix messages from
* different CPUs. This is especially important when printing
* a backtrace.
*/
raw_spin_lock_irqsave(&read_lock, flags);
i = 0;
more:
len = atomic_read(&s->len);
/*
* This is just a paranoid check that nobody has manipulated
* the buffer an unexpected way. If we printed something then
printk/NMI: handle continuous lines and missing newline Commit 4bcc595ccd80 ("printk: reinstate KERN_CONT for printing continuation lines") added back KERN_CONT message header. As a result it might appear in the middle of the line when the parts are squashed via the temporary NMI buffer. A reasonable solution seems to be to split the text in the NNI temporary not only by newlines but also by the message headers. Another solution would be to filter out KERN_CONT when writing to the temporary buffer. But this would complicate the lockless handling. Also it would not solve problems with a missing newline that was there even before the KERN_CONT stuff. This patch moves the temporary buffer handling into separate function. I played with it and it seems that using the char pointers make the code easier to read. Also it prints the final newline as a continuous line. Finally, it moves handling of the s->len overflow into the paranoid check. And allows to recover from the disaster. Link: http://lkml.kernel.org/r/1478695291-12169-2-git-send-email-pmladek@suse.com Signed-off-by: Petr Mladek <pmladek@suse.com> Reviewed-by: Sergey Senozhatsky <sergey.senozhatsky@gmail.com> Cc: Joe Perches <joe@perches.com> Cc: Steven Rostedt <rostedt@goodmis.org> Cc: Jason Wessel <jason.wessel@windriver.com> Cc: Jaroslav Kysela <perex@perex.cz> Cc: Takashi Iwai <tiwai@suse.com> Cc: Chris Mason <clm@fb.com> Cc: Josef Bacik <jbacik@fb.com> Cc: David Sterba <dsterba@suse.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-12-13 08:45:44 +08:00
* @len must only increase. Also it should never overflow the
* buffer size.
printk/nmi: generic solution for safe printk in NMI printk() takes some locks and could not be used a safe way in NMI context. The chance of a deadlock is real especially when printing stacks from all CPUs. This particular problem has been addressed on x86 by the commit a9edc8809328 ("x86/nmi: Perform a safe NMI stack trace on all CPUs"). The patchset brings two big advantages. First, it makes the NMI backtraces safe on all architectures for free. Second, it makes all NMI messages almost safe on all architectures (the temporary buffer is limited. We still should keep the number of messages in NMI context at minimum). Note that there already are several messages printed in NMI context: WARN_ON(in_nmi()), BUG_ON(in_nmi()), anything being printed out from MCE handlers. These are not easy to avoid. This patch reuses most of the code and makes it generic. It is useful for all messages and architectures that support NMI. The alternative printk_func is set when entering and is reseted when leaving NMI context. It queues IRQ work to copy the messages into the main ring buffer in a safe context. __printk_nmi_flush() copies all available messages and reset the buffer. Then we could use a simple cmpxchg operations to get synchronized with writers. There is also used a spinlock to get synchronized with other flushers. We do not longer use seq_buf because it depends on external lock. It would be hard to make all supported operations safe for a lockless use. It would be confusing and error prone to make only some operations safe. The code is put into separate printk/nmi.c as suggested by Steven Rostedt. It needs a per-CPU buffer and is compiled only on architectures that call nmi_enter(). This is achieved by the new HAVE_NMI Kconfig flag. The are MN10300 and Xtensa architectures. We need to clean up NMI handling there first. Let's do it separately. The patch is heavily based on the draft from Peter Zijlstra, see https://lkml.org/lkml/2015/6/10/327 [arnd@arndb.de: printk-nmi: use %zu format string for size_t] [akpm@linux-foundation.org: min_t->min - all types are size_t here] Signed-off-by: Petr Mladek <pmladek@suse.com> Suggested-by: Peter Zijlstra <peterz@infradead.org> Suggested-by: Steven Rostedt <rostedt@goodmis.org> Cc: Jan Kara <jack@suse.cz> Acked-by: Russell King <rmk+kernel@arm.linux.org.uk> [arm part] Cc: Daniel Thompson <daniel.thompson@linaro.org> Cc: Jiri Kosina <jkosina@suse.com> Cc: Ingo Molnar <mingo@redhat.com> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: Ralf Baechle <ralf@linux-mips.org> Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org> Cc: Martin Schwidefsky <schwidefsky@de.ibm.com> Cc: David Miller <davem@davemloft.net> Cc: Daniel Thompson <daniel.thompson@linaro.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-05-21 08:00:33 +08:00
*/
printk/NMI: handle continuous lines and missing newline Commit 4bcc595ccd80 ("printk: reinstate KERN_CONT for printing continuation lines") added back KERN_CONT message header. As a result it might appear in the middle of the line when the parts are squashed via the temporary NMI buffer. A reasonable solution seems to be to split the text in the NNI temporary not only by newlines but also by the message headers. Another solution would be to filter out KERN_CONT when writing to the temporary buffer. But this would complicate the lockless handling. Also it would not solve problems with a missing newline that was there even before the KERN_CONT stuff. This patch moves the temporary buffer handling into separate function. I played with it and it seems that using the char pointers make the code easier to read. Also it prints the final newline as a continuous line. Finally, it moves handling of the s->len overflow into the paranoid check. And allows to recover from the disaster. Link: http://lkml.kernel.org/r/1478695291-12169-2-git-send-email-pmladek@suse.com Signed-off-by: Petr Mladek <pmladek@suse.com> Reviewed-by: Sergey Senozhatsky <sergey.senozhatsky@gmail.com> Cc: Joe Perches <joe@perches.com> Cc: Steven Rostedt <rostedt@goodmis.org> Cc: Jason Wessel <jason.wessel@windriver.com> Cc: Jaroslav Kysela <perex@perex.cz> Cc: Takashi Iwai <tiwai@suse.com> Cc: Chris Mason <clm@fb.com> Cc: Josef Bacik <jbacik@fb.com> Cc: David Sterba <dsterba@suse.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-12-13 08:45:44 +08:00
if ((i && i >= len) || len > sizeof(s->buffer)) {
const char *msg = "printk_safe_flush: internal error\n";
printk_safe_flush_line(msg, strlen(msg));
printk/NMI: handle continuous lines and missing newline Commit 4bcc595ccd80 ("printk: reinstate KERN_CONT for printing continuation lines") added back KERN_CONT message header. As a result it might appear in the middle of the line when the parts are squashed via the temporary NMI buffer. A reasonable solution seems to be to split the text in the NNI temporary not only by newlines but also by the message headers. Another solution would be to filter out KERN_CONT when writing to the temporary buffer. But this would complicate the lockless handling. Also it would not solve problems with a missing newline that was there even before the KERN_CONT stuff. This patch moves the temporary buffer handling into separate function. I played with it and it seems that using the char pointers make the code easier to read. Also it prints the final newline as a continuous line. Finally, it moves handling of the s->len overflow into the paranoid check. And allows to recover from the disaster. Link: http://lkml.kernel.org/r/1478695291-12169-2-git-send-email-pmladek@suse.com Signed-off-by: Petr Mladek <pmladek@suse.com> Reviewed-by: Sergey Senozhatsky <sergey.senozhatsky@gmail.com> Cc: Joe Perches <joe@perches.com> Cc: Steven Rostedt <rostedt@goodmis.org> Cc: Jason Wessel <jason.wessel@windriver.com> Cc: Jaroslav Kysela <perex@perex.cz> Cc: Takashi Iwai <tiwai@suse.com> Cc: Chris Mason <clm@fb.com> Cc: Josef Bacik <jbacik@fb.com> Cc: David Sterba <dsterba@suse.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-12-13 08:45:44 +08:00
len = 0;
}
printk/nmi: generic solution for safe printk in NMI printk() takes some locks and could not be used a safe way in NMI context. The chance of a deadlock is real especially when printing stacks from all CPUs. This particular problem has been addressed on x86 by the commit a9edc8809328 ("x86/nmi: Perform a safe NMI stack trace on all CPUs"). The patchset brings two big advantages. First, it makes the NMI backtraces safe on all architectures for free. Second, it makes all NMI messages almost safe on all architectures (the temporary buffer is limited. We still should keep the number of messages in NMI context at minimum). Note that there already are several messages printed in NMI context: WARN_ON(in_nmi()), BUG_ON(in_nmi()), anything being printed out from MCE handlers. These are not easy to avoid. This patch reuses most of the code and makes it generic. It is useful for all messages and architectures that support NMI. The alternative printk_func is set when entering and is reseted when leaving NMI context. It queues IRQ work to copy the messages into the main ring buffer in a safe context. __printk_nmi_flush() copies all available messages and reset the buffer. Then we could use a simple cmpxchg operations to get synchronized with writers. There is also used a spinlock to get synchronized with other flushers. We do not longer use seq_buf because it depends on external lock. It would be hard to make all supported operations safe for a lockless use. It would be confusing and error prone to make only some operations safe. The code is put into separate printk/nmi.c as suggested by Steven Rostedt. It needs a per-CPU buffer and is compiled only on architectures that call nmi_enter(). This is achieved by the new HAVE_NMI Kconfig flag. The are MN10300 and Xtensa architectures. We need to clean up NMI handling there first. Let's do it separately. The patch is heavily based on the draft from Peter Zijlstra, see https://lkml.org/lkml/2015/6/10/327 [arnd@arndb.de: printk-nmi: use %zu format string for size_t] [akpm@linux-foundation.org: min_t->min - all types are size_t here] Signed-off-by: Petr Mladek <pmladek@suse.com> Suggested-by: Peter Zijlstra <peterz@infradead.org> Suggested-by: Steven Rostedt <rostedt@goodmis.org> Cc: Jan Kara <jack@suse.cz> Acked-by: Russell King <rmk+kernel@arm.linux.org.uk> [arm part] Cc: Daniel Thompson <daniel.thompson@linaro.org> Cc: Jiri Kosina <jkosina@suse.com> Cc: Ingo Molnar <mingo@redhat.com> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: Ralf Baechle <ralf@linux-mips.org> Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org> Cc: Martin Schwidefsky <schwidefsky@de.ibm.com> Cc: David Miller <davem@davemloft.net> Cc: Daniel Thompson <daniel.thompson@linaro.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-05-21 08:00:33 +08:00
if (!len)
goto out; /* Someone else has already flushed the buffer. */
/* Make sure that data has been written up to the @len */
smp_rmb();
i += printk_safe_flush_buffer(s->buffer + i, len - i);
printk/nmi: generic solution for safe printk in NMI printk() takes some locks and could not be used a safe way in NMI context. The chance of a deadlock is real especially when printing stacks from all CPUs. This particular problem has been addressed on x86 by the commit a9edc8809328 ("x86/nmi: Perform a safe NMI stack trace on all CPUs"). The patchset brings two big advantages. First, it makes the NMI backtraces safe on all architectures for free. Second, it makes all NMI messages almost safe on all architectures (the temporary buffer is limited. We still should keep the number of messages in NMI context at minimum). Note that there already are several messages printed in NMI context: WARN_ON(in_nmi()), BUG_ON(in_nmi()), anything being printed out from MCE handlers. These are not easy to avoid. This patch reuses most of the code and makes it generic. It is useful for all messages and architectures that support NMI. The alternative printk_func is set when entering and is reseted when leaving NMI context. It queues IRQ work to copy the messages into the main ring buffer in a safe context. __printk_nmi_flush() copies all available messages and reset the buffer. Then we could use a simple cmpxchg operations to get synchronized with writers. There is also used a spinlock to get synchronized with other flushers. We do not longer use seq_buf because it depends on external lock. It would be hard to make all supported operations safe for a lockless use. It would be confusing and error prone to make only some operations safe. The code is put into separate printk/nmi.c as suggested by Steven Rostedt. It needs a per-CPU buffer and is compiled only on architectures that call nmi_enter(). This is achieved by the new HAVE_NMI Kconfig flag. The are MN10300 and Xtensa architectures. We need to clean up NMI handling there first. Let's do it separately. The patch is heavily based on the draft from Peter Zijlstra, see https://lkml.org/lkml/2015/6/10/327 [arnd@arndb.de: printk-nmi: use %zu format string for size_t] [akpm@linux-foundation.org: min_t->min - all types are size_t here] Signed-off-by: Petr Mladek <pmladek@suse.com> Suggested-by: Peter Zijlstra <peterz@infradead.org> Suggested-by: Steven Rostedt <rostedt@goodmis.org> Cc: Jan Kara <jack@suse.cz> Acked-by: Russell King <rmk+kernel@arm.linux.org.uk> [arm part] Cc: Daniel Thompson <daniel.thompson@linaro.org> Cc: Jiri Kosina <jkosina@suse.com> Cc: Ingo Molnar <mingo@redhat.com> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: Ralf Baechle <ralf@linux-mips.org> Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org> Cc: Martin Schwidefsky <schwidefsky@de.ibm.com> Cc: David Miller <davem@davemloft.net> Cc: Daniel Thompson <daniel.thompson@linaro.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-05-21 08:00:33 +08:00
/*
* Check that nothing has got added in the meantime and truncate
* the buffer. Note that atomic_cmpxchg() is an implicit memory
* barrier that makes sure that the data were copied before
* updating s->len.
*/
if (atomic_cmpxchg(&s->len, len, 0) != len)
goto more;
out:
raw_spin_unlock_irqrestore(&read_lock, flags);
}
/**
* printk_safe_flush - flush all per-cpu nmi buffers.
printk/nmi: generic solution for safe printk in NMI printk() takes some locks and could not be used a safe way in NMI context. The chance of a deadlock is real especially when printing stacks from all CPUs. This particular problem has been addressed on x86 by the commit a9edc8809328 ("x86/nmi: Perform a safe NMI stack trace on all CPUs"). The patchset brings two big advantages. First, it makes the NMI backtraces safe on all architectures for free. Second, it makes all NMI messages almost safe on all architectures (the temporary buffer is limited. We still should keep the number of messages in NMI context at minimum). Note that there already are several messages printed in NMI context: WARN_ON(in_nmi()), BUG_ON(in_nmi()), anything being printed out from MCE handlers. These are not easy to avoid. This patch reuses most of the code and makes it generic. It is useful for all messages and architectures that support NMI. The alternative printk_func is set when entering and is reseted when leaving NMI context. It queues IRQ work to copy the messages into the main ring buffer in a safe context. __printk_nmi_flush() copies all available messages and reset the buffer. Then we could use a simple cmpxchg operations to get synchronized with writers. There is also used a spinlock to get synchronized with other flushers. We do not longer use seq_buf because it depends on external lock. It would be hard to make all supported operations safe for a lockless use. It would be confusing and error prone to make only some operations safe. The code is put into separate printk/nmi.c as suggested by Steven Rostedt. It needs a per-CPU buffer and is compiled only on architectures that call nmi_enter(). This is achieved by the new HAVE_NMI Kconfig flag. The are MN10300 and Xtensa architectures. We need to clean up NMI handling there first. Let's do it separately. The patch is heavily based on the draft from Peter Zijlstra, see https://lkml.org/lkml/2015/6/10/327 [arnd@arndb.de: printk-nmi: use %zu format string for size_t] [akpm@linux-foundation.org: min_t->min - all types are size_t here] Signed-off-by: Petr Mladek <pmladek@suse.com> Suggested-by: Peter Zijlstra <peterz@infradead.org> Suggested-by: Steven Rostedt <rostedt@goodmis.org> Cc: Jan Kara <jack@suse.cz> Acked-by: Russell King <rmk+kernel@arm.linux.org.uk> [arm part] Cc: Daniel Thompson <daniel.thompson@linaro.org> Cc: Jiri Kosina <jkosina@suse.com> Cc: Ingo Molnar <mingo@redhat.com> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: Ralf Baechle <ralf@linux-mips.org> Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org> Cc: Martin Schwidefsky <schwidefsky@de.ibm.com> Cc: David Miller <davem@davemloft.net> Cc: Daniel Thompson <daniel.thompson@linaro.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-05-21 08:00:33 +08:00
*
* The buffers are flushed automatically via IRQ work. This function
* is useful only when someone wants to be sure that all buffers have
* been flushed at some point.
*/
void printk_safe_flush(void)
printk/nmi: generic solution for safe printk in NMI printk() takes some locks and could not be used a safe way in NMI context. The chance of a deadlock is real especially when printing stacks from all CPUs. This particular problem has been addressed on x86 by the commit a9edc8809328 ("x86/nmi: Perform a safe NMI stack trace on all CPUs"). The patchset brings two big advantages. First, it makes the NMI backtraces safe on all architectures for free. Second, it makes all NMI messages almost safe on all architectures (the temporary buffer is limited. We still should keep the number of messages in NMI context at minimum). Note that there already are several messages printed in NMI context: WARN_ON(in_nmi()), BUG_ON(in_nmi()), anything being printed out from MCE handlers. These are not easy to avoid. This patch reuses most of the code and makes it generic. It is useful for all messages and architectures that support NMI. The alternative printk_func is set when entering and is reseted when leaving NMI context. It queues IRQ work to copy the messages into the main ring buffer in a safe context. __printk_nmi_flush() copies all available messages and reset the buffer. Then we could use a simple cmpxchg operations to get synchronized with writers. There is also used a spinlock to get synchronized with other flushers. We do not longer use seq_buf because it depends on external lock. It would be hard to make all supported operations safe for a lockless use. It would be confusing and error prone to make only some operations safe. The code is put into separate printk/nmi.c as suggested by Steven Rostedt. It needs a per-CPU buffer and is compiled only on architectures that call nmi_enter(). This is achieved by the new HAVE_NMI Kconfig flag. The are MN10300 and Xtensa architectures. We need to clean up NMI handling there first. Let's do it separately. The patch is heavily based on the draft from Peter Zijlstra, see https://lkml.org/lkml/2015/6/10/327 [arnd@arndb.de: printk-nmi: use %zu format string for size_t] [akpm@linux-foundation.org: min_t->min - all types are size_t here] Signed-off-by: Petr Mladek <pmladek@suse.com> Suggested-by: Peter Zijlstra <peterz@infradead.org> Suggested-by: Steven Rostedt <rostedt@goodmis.org> Cc: Jan Kara <jack@suse.cz> Acked-by: Russell King <rmk+kernel@arm.linux.org.uk> [arm part] Cc: Daniel Thompson <daniel.thompson@linaro.org> Cc: Jiri Kosina <jkosina@suse.com> Cc: Ingo Molnar <mingo@redhat.com> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: Ralf Baechle <ralf@linux-mips.org> Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org> Cc: Martin Schwidefsky <schwidefsky@de.ibm.com> Cc: David Miller <davem@davemloft.net> Cc: Daniel Thompson <daniel.thompson@linaro.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-05-21 08:00:33 +08:00
{
int cpu;
for_each_possible_cpu(cpu)
__printk_safe_flush(&per_cpu(nmi_print_seq, cpu).work);
printk/nmi: generic solution for safe printk in NMI printk() takes some locks and could not be used a safe way in NMI context. The chance of a deadlock is real especially when printing stacks from all CPUs. This particular problem has been addressed on x86 by the commit a9edc8809328 ("x86/nmi: Perform a safe NMI stack trace on all CPUs"). The patchset brings two big advantages. First, it makes the NMI backtraces safe on all architectures for free. Second, it makes all NMI messages almost safe on all architectures (the temporary buffer is limited. We still should keep the number of messages in NMI context at minimum). Note that there already are several messages printed in NMI context: WARN_ON(in_nmi()), BUG_ON(in_nmi()), anything being printed out from MCE handlers. These are not easy to avoid. This patch reuses most of the code and makes it generic. It is useful for all messages and architectures that support NMI. The alternative printk_func is set when entering and is reseted when leaving NMI context. It queues IRQ work to copy the messages into the main ring buffer in a safe context. __printk_nmi_flush() copies all available messages and reset the buffer. Then we could use a simple cmpxchg operations to get synchronized with writers. There is also used a spinlock to get synchronized with other flushers. We do not longer use seq_buf because it depends on external lock. It would be hard to make all supported operations safe for a lockless use. It would be confusing and error prone to make only some operations safe. The code is put into separate printk/nmi.c as suggested by Steven Rostedt. It needs a per-CPU buffer and is compiled only on architectures that call nmi_enter(). This is achieved by the new HAVE_NMI Kconfig flag. The are MN10300 and Xtensa architectures. We need to clean up NMI handling there first. Let's do it separately. The patch is heavily based on the draft from Peter Zijlstra, see https://lkml.org/lkml/2015/6/10/327 [arnd@arndb.de: printk-nmi: use %zu format string for size_t] [akpm@linux-foundation.org: min_t->min - all types are size_t here] Signed-off-by: Petr Mladek <pmladek@suse.com> Suggested-by: Peter Zijlstra <peterz@infradead.org> Suggested-by: Steven Rostedt <rostedt@goodmis.org> Cc: Jan Kara <jack@suse.cz> Acked-by: Russell King <rmk+kernel@arm.linux.org.uk> [arm part] Cc: Daniel Thompson <daniel.thompson@linaro.org> Cc: Jiri Kosina <jkosina@suse.com> Cc: Ingo Molnar <mingo@redhat.com> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: Ralf Baechle <ralf@linux-mips.org> Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org> Cc: Martin Schwidefsky <schwidefsky@de.ibm.com> Cc: David Miller <davem@davemloft.net> Cc: Daniel Thompson <daniel.thompson@linaro.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-05-21 08:00:33 +08:00
}
printk/nmi: flush NMI messages on the system panic In NMI context, printk() messages are stored into per-CPU buffers to avoid a possible deadlock. They are normally flushed to the main ring buffer via an IRQ work. But the work is never called when the system calls panic() in the very same NMI handler. This patch tries to flush NMI buffers before the crash dump is generated. In this case it does not risk a double release and bails out when the logbuf_lock is already taken. The aim is to get the messages into the main ring buffer when possible. It makes them better accessible in the vmcore. Then the patch tries to flush the buffers second time when other CPUs are down. It might be more aggressive and reset logbuf_lock. The aim is to get the messages available for the consequent kmsg_dump() and console_flush_on_panic() calls. The patch causes vprintk_emit() to be called even in NMI context again. But it is done via printk_deferred() so that the console handling is skipped. Consoles use internal locks and we could not prevent a deadlock easily. They are explicitly called later when the crash dump is not generated, see console_flush_on_panic(). Signed-off-by: Petr Mladek <pmladek@suse.com> Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org> Cc: Daniel Thompson <daniel.thompson@linaro.org> Cc: David Miller <davem@davemloft.net> Cc: Ingo Molnar <mingo@redhat.com> Cc: Jan Kara <jack@suse.cz> Cc: Jiri Kosina <jkosina@suse.com> Cc: Martin Schwidefsky <schwidefsky@de.ibm.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Ralf Baechle <ralf@linux-mips.org> Cc: Russell King <rmk+kernel@arm.linux.org.uk> Cc: Steven Rostedt <rostedt@goodmis.org> Cc: Thomas Gleixner <tglx@linutronix.de> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-05-21 08:00:42 +08:00
/**
* printk_safe_flush_on_panic - flush all per-cpu nmi buffers when the system
printk/nmi: flush NMI messages on the system panic In NMI context, printk() messages are stored into per-CPU buffers to avoid a possible deadlock. They are normally flushed to the main ring buffer via an IRQ work. But the work is never called when the system calls panic() in the very same NMI handler. This patch tries to flush NMI buffers before the crash dump is generated. In this case it does not risk a double release and bails out when the logbuf_lock is already taken. The aim is to get the messages into the main ring buffer when possible. It makes them better accessible in the vmcore. Then the patch tries to flush the buffers second time when other CPUs are down. It might be more aggressive and reset logbuf_lock. The aim is to get the messages available for the consequent kmsg_dump() and console_flush_on_panic() calls. The patch causes vprintk_emit() to be called even in NMI context again. But it is done via printk_deferred() so that the console handling is skipped. Consoles use internal locks and we could not prevent a deadlock easily. They are explicitly called later when the crash dump is not generated, see console_flush_on_panic(). Signed-off-by: Petr Mladek <pmladek@suse.com> Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org> Cc: Daniel Thompson <daniel.thompson@linaro.org> Cc: David Miller <davem@davemloft.net> Cc: Ingo Molnar <mingo@redhat.com> Cc: Jan Kara <jack@suse.cz> Cc: Jiri Kosina <jkosina@suse.com> Cc: Martin Schwidefsky <schwidefsky@de.ibm.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Ralf Baechle <ralf@linux-mips.org> Cc: Russell King <rmk+kernel@arm.linux.org.uk> Cc: Steven Rostedt <rostedt@goodmis.org> Cc: Thomas Gleixner <tglx@linutronix.de> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-05-21 08:00:42 +08:00
* goes down.
*
* Similar to printk_safe_flush() but it can be called even in NMI context when
printk/nmi: flush NMI messages on the system panic In NMI context, printk() messages are stored into per-CPU buffers to avoid a possible deadlock. They are normally flushed to the main ring buffer via an IRQ work. But the work is never called when the system calls panic() in the very same NMI handler. This patch tries to flush NMI buffers before the crash dump is generated. In this case it does not risk a double release and bails out when the logbuf_lock is already taken. The aim is to get the messages into the main ring buffer when possible. It makes them better accessible in the vmcore. Then the patch tries to flush the buffers second time when other CPUs are down. It might be more aggressive and reset logbuf_lock. The aim is to get the messages available for the consequent kmsg_dump() and console_flush_on_panic() calls. The patch causes vprintk_emit() to be called even in NMI context again. But it is done via printk_deferred() so that the console handling is skipped. Consoles use internal locks and we could not prevent a deadlock easily. They are explicitly called later when the crash dump is not generated, see console_flush_on_panic(). Signed-off-by: Petr Mladek <pmladek@suse.com> Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org> Cc: Daniel Thompson <daniel.thompson@linaro.org> Cc: David Miller <davem@davemloft.net> Cc: Ingo Molnar <mingo@redhat.com> Cc: Jan Kara <jack@suse.cz> Cc: Jiri Kosina <jkosina@suse.com> Cc: Martin Schwidefsky <schwidefsky@de.ibm.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Ralf Baechle <ralf@linux-mips.org> Cc: Russell King <rmk+kernel@arm.linux.org.uk> Cc: Steven Rostedt <rostedt@goodmis.org> Cc: Thomas Gleixner <tglx@linutronix.de> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-05-21 08:00:42 +08:00
* the system goes down. It does the best effort to get NMI messages into
* the main ring buffer.
*
* Note that it could try harder when there is only one CPU online.
*/
void printk_safe_flush_on_panic(void)
printk/nmi: flush NMI messages on the system panic In NMI context, printk() messages are stored into per-CPU buffers to avoid a possible deadlock. They are normally flushed to the main ring buffer via an IRQ work. But the work is never called when the system calls panic() in the very same NMI handler. This patch tries to flush NMI buffers before the crash dump is generated. In this case it does not risk a double release and bails out when the logbuf_lock is already taken. The aim is to get the messages into the main ring buffer when possible. It makes them better accessible in the vmcore. Then the patch tries to flush the buffers second time when other CPUs are down. It might be more aggressive and reset logbuf_lock. The aim is to get the messages available for the consequent kmsg_dump() and console_flush_on_panic() calls. The patch causes vprintk_emit() to be called even in NMI context again. But it is done via printk_deferred() so that the console handling is skipped. Consoles use internal locks and we could not prevent a deadlock easily. They are explicitly called later when the crash dump is not generated, see console_flush_on_panic(). Signed-off-by: Petr Mladek <pmladek@suse.com> Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org> Cc: Daniel Thompson <daniel.thompson@linaro.org> Cc: David Miller <davem@davemloft.net> Cc: Ingo Molnar <mingo@redhat.com> Cc: Jan Kara <jack@suse.cz> Cc: Jiri Kosina <jkosina@suse.com> Cc: Martin Schwidefsky <schwidefsky@de.ibm.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Ralf Baechle <ralf@linux-mips.org> Cc: Russell King <rmk+kernel@arm.linux.org.uk> Cc: Steven Rostedt <rostedt@goodmis.org> Cc: Thomas Gleixner <tglx@linutronix.de> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-05-21 08:00:42 +08:00
{
/*
* Make sure that we could access the main ring buffer.
* Do not risk a double release when more CPUs are up.
*/
if (in_nmi() && raw_spin_is_locked(&logbuf_lock)) {
if (num_online_cpus() > 1)
return;
debug_locks_off();
raw_spin_lock_init(&logbuf_lock);
}
printk_safe_flush();
printk/nmi: flush NMI messages on the system panic In NMI context, printk() messages are stored into per-CPU buffers to avoid a possible deadlock. They are normally flushed to the main ring buffer via an IRQ work. But the work is never called when the system calls panic() in the very same NMI handler. This patch tries to flush NMI buffers before the crash dump is generated. In this case it does not risk a double release and bails out when the logbuf_lock is already taken. The aim is to get the messages into the main ring buffer when possible. It makes them better accessible in the vmcore. Then the patch tries to flush the buffers second time when other CPUs are down. It might be more aggressive and reset logbuf_lock. The aim is to get the messages available for the consequent kmsg_dump() and console_flush_on_panic() calls. The patch causes vprintk_emit() to be called even in NMI context again. But it is done via printk_deferred() so that the console handling is skipped. Consoles use internal locks and we could not prevent a deadlock easily. They are explicitly called later when the crash dump is not generated, see console_flush_on_panic(). Signed-off-by: Petr Mladek <pmladek@suse.com> Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org> Cc: Daniel Thompson <daniel.thompson@linaro.org> Cc: David Miller <davem@davemloft.net> Cc: Ingo Molnar <mingo@redhat.com> Cc: Jan Kara <jack@suse.cz> Cc: Jiri Kosina <jkosina@suse.com> Cc: Martin Schwidefsky <schwidefsky@de.ibm.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Ralf Baechle <ralf@linux-mips.org> Cc: Russell King <rmk+kernel@arm.linux.org.uk> Cc: Steven Rostedt <rostedt@goodmis.org> Cc: Thomas Gleixner <tglx@linutronix.de> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-05-21 08:00:42 +08:00
}
void __init printk_safe_init(void)
printk/nmi: generic solution for safe printk in NMI printk() takes some locks and could not be used a safe way in NMI context. The chance of a deadlock is real especially when printing stacks from all CPUs. This particular problem has been addressed on x86 by the commit a9edc8809328 ("x86/nmi: Perform a safe NMI stack trace on all CPUs"). The patchset brings two big advantages. First, it makes the NMI backtraces safe on all architectures for free. Second, it makes all NMI messages almost safe on all architectures (the temporary buffer is limited. We still should keep the number of messages in NMI context at minimum). Note that there already are several messages printed in NMI context: WARN_ON(in_nmi()), BUG_ON(in_nmi()), anything being printed out from MCE handlers. These are not easy to avoid. This patch reuses most of the code and makes it generic. It is useful for all messages and architectures that support NMI. The alternative printk_func is set when entering and is reseted when leaving NMI context. It queues IRQ work to copy the messages into the main ring buffer in a safe context. __printk_nmi_flush() copies all available messages and reset the buffer. Then we could use a simple cmpxchg operations to get synchronized with writers. There is also used a spinlock to get synchronized with other flushers. We do not longer use seq_buf because it depends on external lock. It would be hard to make all supported operations safe for a lockless use. It would be confusing and error prone to make only some operations safe. The code is put into separate printk/nmi.c as suggested by Steven Rostedt. It needs a per-CPU buffer and is compiled only on architectures that call nmi_enter(). This is achieved by the new HAVE_NMI Kconfig flag. The are MN10300 and Xtensa architectures. We need to clean up NMI handling there first. Let's do it separately. The patch is heavily based on the draft from Peter Zijlstra, see https://lkml.org/lkml/2015/6/10/327 [arnd@arndb.de: printk-nmi: use %zu format string for size_t] [akpm@linux-foundation.org: min_t->min - all types are size_t here] Signed-off-by: Petr Mladek <pmladek@suse.com> Suggested-by: Peter Zijlstra <peterz@infradead.org> Suggested-by: Steven Rostedt <rostedt@goodmis.org> Cc: Jan Kara <jack@suse.cz> Acked-by: Russell King <rmk+kernel@arm.linux.org.uk> [arm part] Cc: Daniel Thompson <daniel.thompson@linaro.org> Cc: Jiri Kosina <jkosina@suse.com> Cc: Ingo Molnar <mingo@redhat.com> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: Ralf Baechle <ralf@linux-mips.org> Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org> Cc: Martin Schwidefsky <schwidefsky@de.ibm.com> Cc: David Miller <davem@davemloft.net> Cc: Daniel Thompson <daniel.thompson@linaro.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-05-21 08:00:33 +08:00
{
int cpu;
for_each_possible_cpu(cpu) {
struct printk_safe_seq_buf *s = &per_cpu(nmi_print_seq, cpu);
printk/nmi: generic solution for safe printk in NMI printk() takes some locks and could not be used a safe way in NMI context. The chance of a deadlock is real especially when printing stacks from all CPUs. This particular problem has been addressed on x86 by the commit a9edc8809328 ("x86/nmi: Perform a safe NMI stack trace on all CPUs"). The patchset brings two big advantages. First, it makes the NMI backtraces safe on all architectures for free. Second, it makes all NMI messages almost safe on all architectures (the temporary buffer is limited. We still should keep the number of messages in NMI context at minimum). Note that there already are several messages printed in NMI context: WARN_ON(in_nmi()), BUG_ON(in_nmi()), anything being printed out from MCE handlers. These are not easy to avoid. This patch reuses most of the code and makes it generic. It is useful for all messages and architectures that support NMI. The alternative printk_func is set when entering and is reseted when leaving NMI context. It queues IRQ work to copy the messages into the main ring buffer in a safe context. __printk_nmi_flush() copies all available messages and reset the buffer. Then we could use a simple cmpxchg operations to get synchronized with writers. There is also used a spinlock to get synchronized with other flushers. We do not longer use seq_buf because it depends on external lock. It would be hard to make all supported operations safe for a lockless use. It would be confusing and error prone to make only some operations safe. The code is put into separate printk/nmi.c as suggested by Steven Rostedt. It needs a per-CPU buffer and is compiled only on architectures that call nmi_enter(). This is achieved by the new HAVE_NMI Kconfig flag. The are MN10300 and Xtensa architectures. We need to clean up NMI handling there first. Let's do it separately. The patch is heavily based on the draft from Peter Zijlstra, see https://lkml.org/lkml/2015/6/10/327 [arnd@arndb.de: printk-nmi: use %zu format string for size_t] [akpm@linux-foundation.org: min_t->min - all types are size_t here] Signed-off-by: Petr Mladek <pmladek@suse.com> Suggested-by: Peter Zijlstra <peterz@infradead.org> Suggested-by: Steven Rostedt <rostedt@goodmis.org> Cc: Jan Kara <jack@suse.cz> Acked-by: Russell King <rmk+kernel@arm.linux.org.uk> [arm part] Cc: Daniel Thompson <daniel.thompson@linaro.org> Cc: Jiri Kosina <jkosina@suse.com> Cc: Ingo Molnar <mingo@redhat.com> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: Ralf Baechle <ralf@linux-mips.org> Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org> Cc: Martin Schwidefsky <schwidefsky@de.ibm.com> Cc: David Miller <davem@davemloft.net> Cc: Daniel Thompson <daniel.thompson@linaro.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-05-21 08:00:33 +08:00
init_irq_work(&s->work, __printk_safe_flush);
printk/nmi: generic solution for safe printk in NMI printk() takes some locks and could not be used a safe way in NMI context. The chance of a deadlock is real especially when printing stacks from all CPUs. This particular problem has been addressed on x86 by the commit a9edc8809328 ("x86/nmi: Perform a safe NMI stack trace on all CPUs"). The patchset brings two big advantages. First, it makes the NMI backtraces safe on all architectures for free. Second, it makes all NMI messages almost safe on all architectures (the temporary buffer is limited. We still should keep the number of messages in NMI context at minimum). Note that there already are several messages printed in NMI context: WARN_ON(in_nmi()), BUG_ON(in_nmi()), anything being printed out from MCE handlers. These are not easy to avoid. This patch reuses most of the code and makes it generic. It is useful for all messages and architectures that support NMI. The alternative printk_func is set when entering and is reseted when leaving NMI context. It queues IRQ work to copy the messages into the main ring buffer in a safe context. __printk_nmi_flush() copies all available messages and reset the buffer. Then we could use a simple cmpxchg operations to get synchronized with writers. There is also used a spinlock to get synchronized with other flushers. We do not longer use seq_buf because it depends on external lock. It would be hard to make all supported operations safe for a lockless use. It would be confusing and error prone to make only some operations safe. The code is put into separate printk/nmi.c as suggested by Steven Rostedt. It needs a per-CPU buffer and is compiled only on architectures that call nmi_enter(). This is achieved by the new HAVE_NMI Kconfig flag. The are MN10300 and Xtensa architectures. We need to clean up NMI handling there first. Let's do it separately. The patch is heavily based on the draft from Peter Zijlstra, see https://lkml.org/lkml/2015/6/10/327 [arnd@arndb.de: printk-nmi: use %zu format string for size_t] [akpm@linux-foundation.org: min_t->min - all types are size_t here] Signed-off-by: Petr Mladek <pmladek@suse.com> Suggested-by: Peter Zijlstra <peterz@infradead.org> Suggested-by: Steven Rostedt <rostedt@goodmis.org> Cc: Jan Kara <jack@suse.cz> Acked-by: Russell King <rmk+kernel@arm.linux.org.uk> [arm part] Cc: Daniel Thompson <daniel.thompson@linaro.org> Cc: Jiri Kosina <jkosina@suse.com> Cc: Ingo Molnar <mingo@redhat.com> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: Ralf Baechle <ralf@linux-mips.org> Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org> Cc: Martin Schwidefsky <schwidefsky@de.ibm.com> Cc: David Miller <davem@davemloft.net> Cc: Daniel Thompson <daniel.thompson@linaro.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-05-21 08:00:33 +08:00
}
/* Make sure that IRQ works are initialized before enabling. */
smp_wmb();
printk_safe_irq_ready = 1;
printk/nmi: generic solution for safe printk in NMI printk() takes some locks and could not be used a safe way in NMI context. The chance of a deadlock is real especially when printing stacks from all CPUs. This particular problem has been addressed on x86 by the commit a9edc8809328 ("x86/nmi: Perform a safe NMI stack trace on all CPUs"). The patchset brings two big advantages. First, it makes the NMI backtraces safe on all architectures for free. Second, it makes all NMI messages almost safe on all architectures (the temporary buffer is limited. We still should keep the number of messages in NMI context at minimum). Note that there already are several messages printed in NMI context: WARN_ON(in_nmi()), BUG_ON(in_nmi()), anything being printed out from MCE handlers. These are not easy to avoid. This patch reuses most of the code and makes it generic. It is useful for all messages and architectures that support NMI. The alternative printk_func is set when entering and is reseted when leaving NMI context. It queues IRQ work to copy the messages into the main ring buffer in a safe context. __printk_nmi_flush() copies all available messages and reset the buffer. Then we could use a simple cmpxchg operations to get synchronized with writers. There is also used a spinlock to get synchronized with other flushers. We do not longer use seq_buf because it depends on external lock. It would be hard to make all supported operations safe for a lockless use. It would be confusing and error prone to make only some operations safe. The code is put into separate printk/nmi.c as suggested by Steven Rostedt. It needs a per-CPU buffer and is compiled only on architectures that call nmi_enter(). This is achieved by the new HAVE_NMI Kconfig flag. The are MN10300 and Xtensa architectures. We need to clean up NMI handling there first. Let's do it separately. The patch is heavily based on the draft from Peter Zijlstra, see https://lkml.org/lkml/2015/6/10/327 [arnd@arndb.de: printk-nmi: use %zu format string for size_t] [akpm@linux-foundation.org: min_t->min - all types are size_t here] Signed-off-by: Petr Mladek <pmladek@suse.com> Suggested-by: Peter Zijlstra <peterz@infradead.org> Suggested-by: Steven Rostedt <rostedt@goodmis.org> Cc: Jan Kara <jack@suse.cz> Acked-by: Russell King <rmk+kernel@arm.linux.org.uk> [arm part] Cc: Daniel Thompson <daniel.thompson@linaro.org> Cc: Jiri Kosina <jkosina@suse.com> Cc: Ingo Molnar <mingo@redhat.com> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: Ralf Baechle <ralf@linux-mips.org> Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org> Cc: Martin Schwidefsky <schwidefsky@de.ibm.com> Cc: David Miller <davem@davemloft.net> Cc: Daniel Thompson <daniel.thompson@linaro.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-05-21 08:00:33 +08:00
/* Flush pending messages that did not have scheduled IRQ works. */
printk_safe_flush();
printk/nmi: generic solution for safe printk in NMI printk() takes some locks and could not be used a safe way in NMI context. The chance of a deadlock is real especially when printing stacks from all CPUs. This particular problem has been addressed on x86 by the commit a9edc8809328 ("x86/nmi: Perform a safe NMI stack trace on all CPUs"). The patchset brings two big advantages. First, it makes the NMI backtraces safe on all architectures for free. Second, it makes all NMI messages almost safe on all architectures (the temporary buffer is limited. We still should keep the number of messages in NMI context at minimum). Note that there already are several messages printed in NMI context: WARN_ON(in_nmi()), BUG_ON(in_nmi()), anything being printed out from MCE handlers. These are not easy to avoid. This patch reuses most of the code and makes it generic. It is useful for all messages and architectures that support NMI. The alternative printk_func is set when entering and is reseted when leaving NMI context. It queues IRQ work to copy the messages into the main ring buffer in a safe context. __printk_nmi_flush() copies all available messages and reset the buffer. Then we could use a simple cmpxchg operations to get synchronized with writers. There is also used a spinlock to get synchronized with other flushers. We do not longer use seq_buf because it depends on external lock. It would be hard to make all supported operations safe for a lockless use. It would be confusing and error prone to make only some operations safe. The code is put into separate printk/nmi.c as suggested by Steven Rostedt. It needs a per-CPU buffer and is compiled only on architectures that call nmi_enter(). This is achieved by the new HAVE_NMI Kconfig flag. The are MN10300 and Xtensa architectures. We need to clean up NMI handling there first. Let's do it separately. The patch is heavily based on the draft from Peter Zijlstra, see https://lkml.org/lkml/2015/6/10/327 [arnd@arndb.de: printk-nmi: use %zu format string for size_t] [akpm@linux-foundation.org: min_t->min - all types are size_t here] Signed-off-by: Petr Mladek <pmladek@suse.com> Suggested-by: Peter Zijlstra <peterz@infradead.org> Suggested-by: Steven Rostedt <rostedt@goodmis.org> Cc: Jan Kara <jack@suse.cz> Acked-by: Russell King <rmk+kernel@arm.linux.org.uk> [arm part] Cc: Daniel Thompson <daniel.thompson@linaro.org> Cc: Jiri Kosina <jkosina@suse.com> Cc: Ingo Molnar <mingo@redhat.com> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: Ralf Baechle <ralf@linux-mips.org> Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org> Cc: Martin Schwidefsky <schwidefsky@de.ibm.com> Cc: David Miller <davem@davemloft.net> Cc: Daniel Thompson <daniel.thompson@linaro.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-05-21 08:00:33 +08:00
}
void printk_nmi_enter(void)
{
this_cpu_write(printk_func, vprintk_nmi);
}
void printk_nmi_exit(void)
{
this_cpu_write(printk_func, vprintk_default);
}