Re: [PATCH v2] introduce atomic_pointer to fix a race condition in cancelable mcs spinlocks

[Peter Zijlstra]

On Mon, Jun 02, 2014 at 01:22:10PM -0700, Linus Torvalds wrote:
> On Mon, Jun 2, 2014 at 1:05 PM, Peter Zijlstra <peterz@xxxxxxxxxxxxx> wrote:
> >
> > So the question is, do you prefer subtly broken code or hard compile
> > fails? Me, I go for the compile fail.
> 
> The thing is, parisc has a perfectly fine "cmpxchg" implementation in
> practice, and ACCESS_ONCE() and friends work fine too for reading.
> 
> What the "use a spinlock" approach cannot generally do is:
> 
>  - ACCESS_ONCE() to _write_ things doesn't work well. You really
> should use "atomic_set()".
> 
>  - you may not necessarily be able to mix partial updates (ie
> differently sized updates to the same thing) depending on just how the
> spinlock hashing works
> 
> but both of those are really rare issues and don't affect normal code.

Agreed on the second, although that would be fairly easy to fix by
masking out the lower few bits in the pointer address before hashing.

The first, you're probably, right, but seeing how its a completely
silent fail atm I'm not at all comfortable with it.

Attachment: pgpCRIpFQGMBr.pgp
Description: PGP signature