Skip to content

K
kernel_linux

OpenHarmony / kernel_linux
上一次同步 4 年多

通知 15
Star 8
Fork 2
K
kernel_linux
切换分支/标签
  1. 16 7月, 2009 1 次提交
  3. 23 6月, 2009 1 次提交
  5. 19 6月, 2009 1 次提交
  7. 12 6月, 2009 1 次提交
  9. 03 4月, 2009 1 次提交
  11. 24 3月, 2009 1 次提交
  13. 07 1月, 2009 2 次提交
  15. 18 11月, 2008 1 次提交
  17. 28 10月, 2008 2 次提交
  19. 19 7月, 2008 1 次提交
    • T
      nohz: prevent tick stop outside of the idle loop · b8f8c3cf
      由 Thomas Gleixner 提交于 
      Jack Ren and Eric Miao tracked down the following long standing
problem in the NOHZ code:

	scheduler switch to idle task
	enable interrupts

Window starts here

	----> interrupt happens (does not set NEED_RESCHED)
	      	irq_exit() stops the tick

	----> interrupt happens (does set NEED_RESCHED)

	return from schedule()
	
	cpu_idle(): preempt_disable();

Window ends here

The interrupts can happen at any point inside the race window. The
first interrupt stops the tick, the second one causes the scheduler to
rerun and switch away from idle again and we end up with the tick
disabled.

The fact that it needs two interrupts where the first one does not set
NEED_RESCHED and the second one does made the bug obscure and extremly
hard to reproduce and analyse. Kudos to Jack and Eric.

Solution: Limit the NOHZ functionality to the idle loop to make sure
that we can not run into such a situation ever again.

cpu_idle()
{
	preempt_disable();

	while(1) {
		 tick_nohz_stop_sched_tick(1); <- tell NOHZ code that we
		 			          are in the idle loop

		 while (!need_resched())
		       halt();

		 tick_nohz_restart_sched_tick(); <- disables NOHZ mode
		 preempt_enable_no_resched();
		 schedule();
		 preempt_disable();
	}
}

In hindsight we should have done this forever, but ... 

/me grabs a large brown paperbag.

Debugged-by: Jack Ren <jack.ren@marvell.com>, 
Debugged-by: Neric miao <eric.y.miao@gmail.com>
Signed-off-by: NThomas Gleixner <tglx@linutronix.de>
      b8f8c3cf
  21. 07 5月, 2008 1 次提交
  23. 24 4月, 2008 1 次提交
    • B
      [Blackfin] arch: This allows XIP to work with FD-PIC. · d5adb029
      由 Bernd Schmidt 提交于 
      Previously, init failed to do anything meaningful;
it turns out that the reason is that FD-PIC has a readonly data
section which can be located in the XIP filesystem, and various address checks
in the kernel reject such addresses for syscall arguments.  Hence, init's
  execve ("/bin/sh", ...)
failed with error code EFAULT.

There's room for improvement here: in case people want to have filesystems
on flash rather than in main memory, _access_ok should be modified to
allow this.

This bug fix is also dedicated to Michael Hennerich.
Signed-off-by: NBernd Schmidt <bernds_cb1@t-online.de>
Signed-off-by: NBryan Wu <cooloney@kernel.org>
      d5adb029
  25. 07 3月, 2008 1 次提交
  27. 29 2月, 2008 1 次提交
  29. 09 2月, 2008 1 次提交
  31. 21 11月, 2007 1 次提交
  33. 24 12月, 2007 1 次提交
  35. 23 12月, 2007 1 次提交
  37. 18 11月, 2007 1 次提交
  39. 12 8月, 2007 1 次提交
  41. 10 10月, 2007 1 次提交
  43. 11 10月, 2007 1 次提交
  45. 10 10月, 2007 1 次提交
  47. 12 7月, 2007 1 次提交
  49. 21 6月, 2007 1 次提交
    • B
      Blackfin arch: defines and provides entry points for certain user space... · 7adfb58f
      由 Bernd Schmidt 提交于 
      Blackfin arch: defines and provides entry points for certain user space functions at fixed addresses

This patch defines (and provides) entry points for certain user space functions
at fixed addresses.  The Blackfin has no usable atomic instructions, but we can
ensure that these code sequences appear atomic from a user space point of view
by detecting when we're in the process of executing them during the interrupt
handler return path.  This allows much more efficient pthread lock
implementations than the bfin_spinlock syscall we're currently using.

Also provided is a small sys_rt_sigreturn stub which can be used by the signal
handler setup code.  The signal.c part will be committed separately.
Signed-off-by: NBernd Schmidt <bernd.schmidt@analog.com>
Signed-off-by: NBryan Wu <bryan.wu@analog.com>
      7adfb58f
  51. 08 5月, 2007 1 次提交