Re: [PATCH 00/11] V4: rwsem changes + down_read_critical() proposal

[Peter Zijlstra <peterz@infradead.org>]

On Mon, 2010-05-24 at 13:31 -0700, Michel Lespinasse wrote:
> This is version 4 of my rwsem changes.
> 
> Changes since V3:
> 
> - Split first x86 rwsem change in two smaller parts
> - Added David's rwsem_waiter_type enum suggestion into
>   'rwsem: down_read_critical infrastructure support'
> - Lots of minor style fixes and comments clarified
> 
> I would hope the entire series can be considered for inclusion;
> however if we can not agree on the down_read_critical() bits I would
> still like patches 1-7 to be independently considered as I think they
> still have merit on their own.
> 
> Michel Lespinasse (11):
>   x86 rwsem: stay on fast path when count>0 in __up_write()
>   x86 rwsem: minor cleanups
>   rwsem: fully separate code pathes to wake writers vs readers
>   rwsem: lighter active count checks when waking up readers
>   rwsem: let RWSEM_WAITING_BIAS represent any number of waiting threads
>   rwsem: wake queued readers when writer blocks on active read lock
>   rwsem: smaller wrappers around rwsem_down_failed_common
>   generic rwsem: implement down_read_critical() / up_read_critical()
>   rwsem: down_read_critical infrastructure support
>   x86 rwsem: down_read_critical implementation
>   Use down_read_critical() for /proc/<pid>/exe and /proc/<pid>/maps files


So what happened to those patches that dropped mmap_sem during I/O?

I really don't like people tinkering with the lock implementations like
this. Nor do I like the naming, stats are in no way _critical_.

I really think adding something like this utterly defeats the purpose of
having a fair lock.