Re: [Patch] don't spin with irq disabled

[Juergen Gross <juergen.gross@fujitsu-siemens.com>]

Jan Beulich wrote:
>>>> Juergen Gross <juergen.gross@fujitsu-siemens.com> 26.03.09 10:00 >>>
>> Attached patch reduces interrupt latency in lock handling.
>> spin_lock_irq and spin_lock_irqsave used to turn off IRQs and then tried to
>> get the lock. If the lock was already held, waiting for the lock was done with
>> IRQs still off.
>> The patch reenables IRQs during the wait loop.
> 
> This is wrong - you must not enable interrupts if they weren't enabled upon
> entry to these two functions.

spin_lock_irq disables always IRQs. spin_unlock_irq enables always IRQs. They
are always used in pairs, so IRQs should always be enabled on entry of
spin_lock_irq.
I'm not enabling IRQs unconditionally in spin_lock_irqsave, of course, but use
the flags value saved before...

Did I miss something? Or did you refer only to my inexact comment above?


Juergen

-- 
Juergen Gross                             Principal Developer
IP SW OS6                      Telephone: +49 (0) 89 636 47950
Fujitsu Siemens Computers         e-mail: juergen.gross@fujitsu-siemens.com
Otto-Hahn-Ring 6                Internet: www.fujitsu-siemens.com
D-81739 Muenchen         Company details: www.fujitsu-siemens.com/imprint.html