In article <[email protected]> you wrote:
> On Sun, 24 Aug 2003 00:13:15 +0900
> TeJun Huh <[email protected]> wrote:
>> The race conditions I'm mentioning in this thread are not likely to
>> cause real troubles. The first one does not make any difference on
>> x86, and AFAIK bh isn't used extensively anymore so the second one
>> isn't very relevant either. Only the race condition mentioned in the
>> other thread is of relvance if there is any :-(.
> Are you sure? bh is used in fs subtree to my knowledge ...
Would someone care to spemd a moment to tell me what the spin_lock_bh
does that spin_lock alone does not do? (not just "local_bh_disable",
pleasse :-). I am chasing SMP oopses for filesystems mounted on nbd
which only seem to happen in association with high memory stress (and
possibly "high memory"), and I suspect I am going to be interested by
the answer.
There is no commentary that I can find in the source, beyond the
assembler code in local_bh_enable().
Peter