linux-stable.git/kernel/printk, branch v6.1.78

printk: Consolidate console deferred printing

2023-09-23T09:11:07+00:00

[ Upstream commit 696ffaf50e1f8dbc66223ff614473f945f5fb8d8 ]

Printing to consoles can be deferred for several reasons:

- explicitly with printk_deferred()
- printk() in NMI context
- recursive printk() calls

The current implementation is not consistent. For printk_deferred(),
irq work is scheduled twice. For NMI und recursive, panic CPU
suppression and caller delays are not properly enforced.

Correct these inconsistencies by consolidating the deferred printing
code so that vprintk_deferred() is the top-level function for
deferred printing and vprintk_emit() will perform whichever irq_work
queueing is appropriate.

Also add kerneldoc for wake_up_klogd() and defer_console_output() to
clarify their differences and appropriate usage.

Signed-off-by: John Ogness 
Reviewed-by: Sergey Senozhatsky 
Reviewed-by: Petr Mladek 
Signed-off-by: Petr Mladek 
Link: https://lore.kernel.org/r/20230717194607.145135-6-john.ogness@linutronix.de
Signed-off-by: Sasha Levin

printk: Keep non-panic-CPUs out of console lock

2023-09-23T09:11:07+00:00

[ Upstream commit 51a1d258e50e03a0216bf42b6af9ff34ec402ac1 ]

When in a panic situation, non-panic CPUs should avoid holding the
console lock so as not to contend with the panic CPU. This is already
implemented with abandon_console_lock_in_panic(), which is checked
after each printed line. However, non-panic CPUs should also avoid
trying to acquire the console lock during a panic.

Modify console_trylock() to fail and console_lock() to block() when
called from a non-panic CPU during a panic.

Signed-off-by: John Ogness 
Reviewed-by: Sergey Senozhatsky 
Reviewed-by: Petr Mladek 
Signed-off-by: Petr Mladek 
Link: https://lore.kernel.org/r/20230717194607.145135-4-john.ogness@linutronix.de
Signed-off-by: Sasha Levin

printk: ringbuffer: Fix truncating buffer size min_t cast

2023-09-13T07:43:00+00:00

commit 53e9e33ede37a247d926db5e4a9e56b55204e66c upstream.

If an output buffer size exceeded U16_MAX, the min_t(u16, ...) cast in
copy_data() was causing writes to truncate. This manifested as output
bytes being skipped, seen as %NUL bytes in pstore dumps when the available
record size was larger than 65536. Fix the cast to no longer truncate
the calculation.

Cc: Petr Mladek 
Cc: Sergey Senozhatsky 
Cc: Steven Rostedt 
Cc: John Ogness 
Reported-by: Vijay Balakrishna 
Link: https://lore.kernel.org/lkml/d8bb1ec7-a4c5-43a2-9de0-9643a70b899f@linux.microsoft.com/
Fixes: b6cf8b3f3312 ("printk: add lockless ringbuffer")
Cc: stable@vger.kernel.org
Signed-off-by: Kees Cook 
Tested-by: Vijay Balakrishna 
Tested-by: Guilherme G. Piccoli  # Steam Deck
Reviewed-by: Tyler Hicks (Microsoft) 
Tested-by: Tyler Hicks (Microsoft) 
Reviewed-by: John Ogness 
Reviewed-by: Sergey Senozhatsky 
Reviewed-by: Petr Mladek 
Signed-off-by: Petr Mladek 
Link: https://lore.kernel.org/r/20230811054528.never.165-kees@kernel.org
Signed-off-by: Greg Kroah-Hartman

kernel/printk/index.c: fix memory leak with using debugfs_lookup()

2023-03-11T12:55:32+00:00

[ Upstream commit 55bf243c514553e907efcf2bda92ba090eca8c64 ]

When calling debugfs_lookup() the result must have dput() called on it,
otherwise the memory will leak over time.  To make things simpler, just
call debugfs_lookup_and_remove() instead which handles all of the logic
at once.

Cc: Chris Down 
Cc: Petr Mladek 
Cc: Sergey Senozhatsky 
Cc: Steven Rostedt 
Cc: John Ogness 
Cc: linux-kernel@vger.kernel.org
Signed-off-by: Greg Kroah-Hartman 
Reviewed-by: Sergey Senozhatsky 
Reviewed-by: John Ogness 
Reviewed-by: Petr Mladek 
Signed-off-by: Petr Mladek 
Link: https://lore.kernel.org/r/20230202151411.2308576-1-gregkh@linuxfoundation.org
Signed-off-by: Sasha Levin

printk: Mark printk percpu data ready ro_after_init

2022-09-29T13:20:52+00:00

This variable cannot change post boot.

Signed-off-by: Thomas Gleixner 
Signed-off-by: John Ogness 
Reviewed-by: Sergey Senozhatsky 
Reviewed-by: Petr Mladek 
Reviewed-by: Greg Kroah-Hartman 
Signed-off-by: Petr Mladek 
Link: https://lore.kernel.org/r/20220924000454.3319186-6-john.ogness@linutronix.de

printk: Remove bogus comment vs. boot consoles

2022-09-29T13:20:44+00:00

The comment about unregistering boot consoles is just not matching the
reality. Remove it.

Signed-off-by: Thomas Gleixner 
Signed-off-by: John Ogness 
Reviewed-by: Sergey Senozhatsky 
Reviewed-by: Petr Mladek 
Reviewed-by: Greg Kroah-Hartman 
Signed-off-by: Petr Mladek 
Link: https://lore.kernel.org/r/20220924000454.3319186-5-john.ogness@linutronix.de

printk: Remove write only variable nr_ext_console_drivers

2022-09-29T13:20:36+00:00

Commit a699449bb13b ("printk: refactor and rework printing logic")
removed the need for @nr_ext_console_drivers. Remove the unneeded
variable.

Signed-off-by: Thomas Gleixner 
Signed-off-by: John Ogness 
Reviewed-by: Sergey Senozhatsky 
Reviewed-by: Petr Mladek 
Reviewed-by: Greg Kroah-Hartman 
Signed-off-by: Petr Mladek 
Link: https://lore.kernel.org/r/20220924000454.3319186-4-john.ogness@linutronix.de

printk: Make pr_flush() static

2022-09-29T13:20:22+00:00

No user outside the printk code and no reason to export this.

Signed-off-by: Thomas Gleixner 
Signed-off-by: John Ogness 
Reviewed-by: Sergey Senozhatsky 
Reviewed-by: Petr Mladek 
Reviewed-by: Greg Kroah-Hartman 
Signed-off-by: Petr Mladek 
Link: https://lore.kernel.org/r/20220924000454.3319186-2-john.ogness@linutronix.de

printk: do not wait for consoles when suspended

2022-07-15T08:52:11+00:00

The console_stop() and console_start() functions call pr_flush().
When suspending, these functions are called by the serial subsystem
while the serial port is suspended. In this scenario, if there are
any pending messages, a call to pr_flush() will always result in a
timeout because the serial port cannot make forward progress. This
causes longer suspend and resume times.

Add a check in pr_flush() so that it will immediately timeout if
the consoles are suspended.

Fixes: 3b604ca81202 ("printk: add pr_flush()")
Reported-by: Todd Brandt 
Signed-off-by: John Ogness 
Tested-by: Todd Brandt 
Signed-off-by: Petr Mladek 
Link: https://lore.kernel.org/r/20220715061042.373640-2-john.ogness@linutronix.de

Revert "printk: add functions to prefer direct printing"

2022-06-23T16:41:40+00:00

This reverts commit 2bb2b7b57f81255c13f4395ea911d6bdc70c9fe2.

The testing of 5.19 release candidates revealed missing synchronization
between early and regular console functionality.

It would be possible to start the console kthreads later as a workaround.
But it is clear that console lock serialized console drivers between
each other. It opens a big area of possible problems that were not
considered by people involved in the development and review.

printk() is crucial for debugging kernel issues and console output is
very important part of it. The number of consoles is huge and a proper
review would take some time. As a result it need to be reverted for 5.19.

Link: https://lore.kernel.org/r/YrBdjVwBOVgLfHyb@alley
Signed-off-by: Petr Mladek 
Link: https://lore.kernel.org/r/20220623145157.21938-7-pmladek@suse.com