]> git.baikalelectronics.ru Git - kernel.git/commit
arm: add loglvl to unwind_backtrace()
authorDmitry Safonov <dima@arista.com>
Tue, 9 Jun 2020 04:30:10 +0000 (21:30 -0700)
committerLinus Torvalds <torvalds@linux-foundation.org>
Tue, 9 Jun 2020 16:39:10 +0000 (09:39 -0700)
commit8155dfffa423ebe874a018e1abcb2287f8540e89
tree45529dfb97899f69fea275cb5c0dd98c747288ba
parenta0a0f8aff7d11028ecdc05789f6b07951ba7b2f7
arm: add loglvl to unwind_backtrace()

Currently, the log-level of show_stack() depends on a platform
realization.  It creates situations where the headers are printed with
lower log level or higher than the stacktrace (depending on a platform or
user).

Furthermore, it forces the logic decision from user to an architecture
side.  In result, some users as sysrq/kdb/etc are doing tricks with
temporary rising console_loglevel while printing their messages.  And in
result it not only may print unwanted messages from other CPUs, but also
omit printing at all in the unlucky case where the printk() was deferred.

Introducing log-level parameter and KERN_UNSUPPRESSED [1] seems an easier
approach than introducing more printk buffers.  Also, it will consolidate
printings with headers.

Add log level argument to unwind_backtrace() as a preparation for
introducing show_stack_loglvl().

As a good side-effect arm_syscall() is now printing errors with the same
log level as the backtrace.

[1]: https://lore.kernel.org/lkml/20190528002412.1625-1-dima@arista.com/T/#u

Signed-off-by: Dmitry Safonov <dima@arista.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Cc: Russell King <linux@armlinux.org.uk>
Cc: Will Deacon <will@kernel.org>
Link: http://lkml.kernel.org/r/20200418201944.482088-6-dima@arista.com
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
arch/arm/include/asm/unwind.h
arch/arm/kernel/traps.c
arch/arm/kernel/unwind.c