On Tue, 20 Aug 2013 20:36:58 -0700
Kent Overstreet wrote:
> I think we're mostly quibbling over semantics here, and it's not that
> big a deal - that said, I don't really get your distinction between a
> semaphore and a mutex; I'd distinguish them by saying a semaphore can
> count an arbitrary
On Tue, 20 Aug 2013 20:36:58 -0700
Kent Overstreet k...@daterainc.com wrote:
I think we're mostly quibbling over semantics here, and it's not that
big a deal - that said, I don't really get your distinction between a
semaphore and a mutex; I'd distinguish them by saying a semaphore can
count
On Tue, Aug 20, 2013 at 10:41:53PM -0400, Steven Rostedt wrote:
> > I get that there's a problem, but the bcache code REALLY IS USING THE
> > RWSEM AS A LOCK; the answer isn't to open code the lock!
>
> Actually, it is using it as a semaphore. The problem with Linux
> was that it only had spin
On Tue, 20 Aug 2013 19:19:00 -0700
Kent Overstreet wrote:
> On Tue, Aug 20, 2013 at 11:16:02AM -0400, Steven Rostedt wrote:
> >
> I _really_ disagree with this approach.
I had a feeling you would love it.
>
> I get that there's a problem, but the bcache code REALLY IS USING THE
> RWSEM AS A
On Tue, Aug 20, 2013 at 11:16:02AM -0400, Steven Rostedt wrote:
>
> The down/up_read_non_owner() is a nasty hack in the API of the rwsem
> operations. It was once removed, but then resurrected for use with
> bcache. Not only is the API an abomination to the rwsem API, it also
> prevents bcache
The down/up_read_non_owner() is a nasty hack in the API of the rwsem
operations. It was once removed, but then resurrected for use with
bcache. Not only is the API an abomination to the rwsem API, it also
prevents bcache from ever being compiled with PREEMPT_RT, as the RT
kernel requires all
The down/up_read_non_owner() is a nasty hack in the API of the rwsem
operations. It was once removed, but then resurrected for use with
bcache. Not only is the API an abomination to the rwsem API, it also
prevents bcache from ever being compiled with PREEMPT_RT, as the RT
kernel requires all
On Tue, Aug 20, 2013 at 11:16:02AM -0400, Steven Rostedt wrote:
The down/up_read_non_owner() is a nasty hack in the API of the rwsem
operations. It was once removed, but then resurrected for use with
bcache. Not only is the API an abomination to the rwsem API, it also
prevents bcache from
On Tue, 20 Aug 2013 19:19:00 -0700
Kent Overstreet k...@daterainc.com wrote:
On Tue, Aug 20, 2013 at 11:16:02AM -0400, Steven Rostedt wrote:
I _really_ disagree with this approach.
I had a feeling you would love it.
I get that there's a problem, but the bcache code REALLY IS USING THE
On Tue, Aug 20, 2013 at 10:41:53PM -0400, Steven Rostedt wrote:
I get that there's a problem, but the bcache code REALLY IS USING THE
RWSEM AS A LOCK; the answer isn't to open code the lock!
Actually, it is using it as a semaphore. The problem with Linux
was that it only had spin locks and
10 matches
Mail list logo