linux-toradex.git/kernel/panic.c, branch v5.1-rc1

kernel/panic.c: taint: fix debugfs_simple_attr.cocci warnings

2019-03-08T02:31:59+00:00

Use DEFINE_DEBUGFS_ATTRIBUTE rather than DEFINE_SIMPLE_ATTRIBUTE for
debugfs files.

Semantic patch information:
Rationale: DEFINE_SIMPLE_ATTRIBUTE + debugfs_create_file()
imposes some significant overhead as compared to
DEFINE_DEBUGFS_ATTRIBUTE + debugfs_create_file_unsafe().

Generated by: scripts/coccinelle/api/debugfs/debugfs_simple_attr.cocci

The _unsafe() part suggests that some of them "safeness
responsibilities" are now panic.c responsibilities.  The patch is OK
since panic's clear_warn_once_fops struct file_operations is safe
against removal, so we don't have to use otherwise necessary
debugfs_file_get()/debugfs_file_put().

[sergey.senozhatsky.work@gmail.com: changelog addition]
Link: http://lkml.kernel.org/r/1545990861-158097-1-git-send-email-yuehaibing@huawei.com
Signed-off-by: YueHaibing 
Reviewed-by: Sergey Senozhatsky 
Cc: Kees Cook 
Cc: Borislav Petkov 
Cc: Steven Rostedt (VMware) 
Cc: Petr Mladek 
Signed-off-by: Andrew Morton 
Signed-off-by: Linus Torvalds

kernel/sysctl: add panic_print into sysctl

2019-01-04T21:13:47+00:00

So that we can also runtime chose to print out the needed system info
for panic, other than setting the kernel cmdline.

Link: http://lkml.kernel.org/r/1543398842-19295-3-git-send-email-feng.tang@intel.com
Signed-off-by: Feng Tang 
Suggested-by: Steven Rostedt 
Acked-by: Steven Rostedt (VMware) 
Cc: Thomas Gleixner 
Cc: John Stultz 
Cc: Ingo Molnar 
Cc: Peter Zijlstra 
Cc: Kees Cook 
Signed-off-by: Andrew Morton 
Signed-off-by: Linus Torvalds

panic: add options to print system info when panic happens

2019-01-04T21:13:47+00:00

Kernel panic issues are always painful to debug, partially because it's
not easy to get enough information of the context when panic happens.

And we have ramoops and kdump for that, while this commit tries to
provide a easier way to show the system info by adding a cmdline
parameter, referring some idea from sysrq handler.

Link: http://lkml.kernel.org/r/1543398842-19295-2-git-send-email-feng.tang@intel.com
Signed-off-by: Feng Tang 
Reviewed-by: Kees Cook 
Acked-by: Steven Rostedt (VMware) 
Cc: Thomas Gleixner 
Cc: John Stultz 
Cc: Ingo Molnar 
Cc: Peter Zijlstra 
Cc: Steven Rostedt 
Signed-off-by: Andrew Morton 
Signed-off-by: Linus Torvalds

Merge tag 'printk-for-4.21' of git://git.kernel.org/pub/scm/linux/kernel/git/pmladek/printk

2018-12-27T19:24:43+00:00

Pull printk updates from Petr Mladek:

 - Keep spinlocks busted until the end of panic()

 - Fix races between calculating number of messages that would fit into
   user space buffers, filling the buffers, and switching printk.time
   parameter

 - Some code clean up

* tag 'printk-for-4.21' of git://git.kernel.org/pub/scm/linux/kernel/git/pmladek/printk:
  printk: Remove print_prefix() calls with NULL buffer.
  printk: fix printk_time race.
  printk: Make printk_emit() local function.
  panic: avoid deadlocks in re-entrant console drivers

panic: avoid deadlocks in re-entrant console drivers

2018-11-22T13:07:43+00:00

From printk()/serial console point of view panic() is special, because
it may force CPU to re-enter printk() or/and serial console driver.
Therefore, some of serial consoles drivers are re-entrant. E.g. 8250:

serial8250_console_write()
{
	if (port->sysrq)
		locked = 0;
	else if (oops_in_progress)
		locked = spin_trylock_irqsave(&port->lock, flags);
	else
		spin_lock_irqsave(&port->lock, flags);
	...
}

panic() does set oops_in_progress via bust_spinlocks(1), so in theory
we should be able to re-enter serial console driver from panic():

	CPU0
	
	uart_console_write()
	serial8250_console_write()		// if (oops_in_progress)
						//    spin_trylock_irqsave()
	call_console_drivers()
	console_unlock()
	console_flush_on_panic()
	bust_spinlocks(1)			// oops_in_progress++
	panic()
	
	spin_lock_irqsave(&port->lock, flags)   // spin_lock_irqsave()
	serial8250_console_write()
	call_console_drivers()
	console_unlock()
	printk()
	...

However, this does not happen and we deadlock in serial console on
port->lock spinlock. And the problem is that console_flush_on_panic()
called after bust_spinlocks(0):

void panic(const char *fmt, ...)
{
	bust_spinlocks(1);
	...
	bust_spinlocks(0);
	console_flush_on_panic();
	...
}

bust_spinlocks(0) decrements oops_in_progress, so oops_in_progress
can go back to zero. Thus even re-entrant console drivers will simply
spin on port->lock spinlock. Given that port->lock may already be
locked either by a stopped CPU, or by the very same CPU we execute
panic() on (for instance, NMI panic() on printing CPU) the system
deadlocks and does not reboot.

Fix this by removing bust_spinlocks(0), so oops_in_progress is always
set in panic() now and, thus, re-entrant console drivers will trylock
the port->lock instead of spinning on it forever, when we call them
from console_flush_on_panic().

Link: http://lkml.kernel.org/r/20181025101036.6823-1-sergey.senozhatsky@gmail.com
Cc: Steven Rostedt 
Cc: Daniel Wang 
Cc: Peter Zijlstra 
Cc: Andrew Morton 
Cc: Linus Torvalds 
Cc: Greg Kroah-Hartman 
Cc: Alan Cox 
Cc: Jiri Slaby 
Cc: Peter Feiner 
Cc: linux-serial@vger.kernel.org
Cc: Sergey Senozhatsky 
Cc: stable@vger.kernel.org
Signed-off-by: Sergey Senozhatsky 
Signed-off-by: Petr Mladek

kernel/panic.c: filter out a potential trailing newline

2018-10-31T15:54:14+00:00

If a call to panic() terminates the string with a \n , the result puts the
closing brace ']---' on a newline because panic() itself adds \n too.

Now, if one goes and removes the newline chars from all panic()
invocations - and the stats right now look like this:

~300 calls with a \n
~500 calls without a \n

one is destined to a neverending game of whack-a-mole because the usual
thing to do is add a newline at the end of a string a function is supposed
to print.

Therefore, simply zap any \n at the end of the panic string to avoid
touching so many places in the kernel.

Link: http://lkml.kernel.org/r/20181009205019.2786-1-bp@alien8.de
Signed-off-by: Borislav Petkov 
Acked-by: Kees Cook 
Reviewed-by: Steven Rostedt (VMware) 
Acked-by: Ingo Molnar 
Signed-off-by: Andrew Morton 
Signed-off-by: Linus Torvalds

kernel/panic.c: do not append newline to the stack protector panic string

2018-10-31T15:54:14+00:00

... because panic() itself already does this. Otherwise you have
line-broken trailer:

  [    1.836965] ---[ end Kernel panic - not syncing: stack-protector: Kernel stack is corrupted in: pgd_alloc+0x29e/0x2a0
  [    1.836965]  ]---

Link: http://lkml.kernel.org/r/20181008202901.7894-1-bp@alien8.de
Signed-off-by: Borislav Petkov 
Acked-by: Kees Cook 
Cc: Masahiro Yamada 
Cc: "Steven Rostedt (VMware)" 
Signed-off-by: Andrew Morton 
Signed-off-by: Linus Torvalds

Kbuild: rename CC_STACKPROTECTOR[_STRONG] config variables

2018-06-14T03:21:18+00:00

The changes to automatically test for working stack protector compiler
support in the Kconfig files removed the special STACKPROTECTOR_AUTO
option that picked the strongest stack protector that the compiler
supported.

That was all a nice cleanup - it makes no sense to have the AUTO case
now that the Kconfig phase can just determine the compiler support
directly.

HOWEVER.

It also meant that doing "make oldconfig" would now _disable_ the strong
stackprotector if you had AUTO enabled, because in a legacy config file,
the sane stack protector configuration would look like

  CONFIG_HAVE_CC_STACKPROTECTOR=y
  # CONFIG_CC_STACKPROTECTOR_NONE is not set
  # CONFIG_CC_STACKPROTECTOR_REGULAR is not set
  # CONFIG_CC_STACKPROTECTOR_STRONG is not set
  CONFIG_CC_STACKPROTECTOR_AUTO=y

and when you ran this through "make oldconfig" with the Kbuild changes,
it would ask you about the regular CONFIG_CC_STACKPROTECTOR (that had
been renamed from CONFIG_CC_STACKPROTECTOR_REGULAR to just
CONFIG_CC_STACKPROTECTOR), but it would think that the STRONG version
used to be disabled (because it was really enabled by AUTO), and would
disable it in the new config, resulting in:

  CONFIG_HAVE_CC_STACKPROTECTOR=y
  CONFIG_CC_HAS_STACKPROTECTOR_NONE=y
  CONFIG_CC_STACKPROTECTOR=y
  # CONFIG_CC_STACKPROTECTOR_STRONG is not set
  CONFIG_CC_HAS_SANE_STACKPROTECTOR=y

That's dangerously subtle - people could suddenly find themselves with
the weaker stack protector setup without even realizing.

The solution here is to just rename not just the old RECULAR stack
protector option, but also the strong one.  This does that by just
removing the CC_ prefix entirely for the user choices, because it really
is not about the compiler support (the compiler support now instead
automatially impacts _visibility_ of the options to users).

This results in "make oldconfig" actually asking the user for their
choice, so that we don't have any silent subtle security model changes.
The end result would generally look like this:

  CONFIG_HAVE_CC_STACKPROTECTOR=y
  CONFIG_CC_HAS_STACKPROTECTOR_NONE=y
  CONFIG_STACKPROTECTOR=y
  CONFIG_STACKPROTECTOR_STRONG=y
  CONFIG_CC_HAS_SANE_STACKPROTECTOR=y

where the "CC_" versions really are about internal compiler
infrastructure, not the user selections.

Acked-by: Masahiro Yamada 
Signed-off-by: Linus Torvalds

taint: add taint for randstruct

2018-04-11T17:28:35+00:00

Since the randstruct plugin can intentionally produce extremely unusual
kernel structure layouts (even performance pathological ones), some
maintainers want to be able to trivially determine if an Oops is coming
from a randstruct-built kernel, so as to keep their sanity when
debugging.  This adds the new flag and initializes taint_mask
immediately when built with randstruct.

Link: http://lkml.kernel.org/r/1519084390-43867-4-git-send-email-keescook@chromium.org
Signed-off-by: Kees Cook 
Reviewed-by: Andrew Morton 
Cc: Al Viro 
Cc: Alexey Dobriyan 
Cc: Jonathan Corbet 
Cc: Randy Dunlap 
Signed-off-by: Andrew Morton 
Signed-off-by: Linus Torvalds

taint: consolidate documentation

2018-04-11T17:28:34+00:00

This consolidates the taint bit documentation into a single place with
both numeric and letter values.  Additionally adds the missing TAINT_AUX
documentation.

Link: http://lkml.kernel.org/r/1519084390-43867-3-git-send-email-keescook@chromium.org
Signed-off-by: Kees Cook 
Reviewed-by: Andrew Morton 
Cc: Al Viro 
Cc: Alexey Dobriyan 
Cc: Jonathan Corbet 
Cc: Randy Dunlap 
Signed-off-by: Andrew Morton 
Signed-off-by: Linus Torvalds