Lock Elision / Transaction ABI discussion

[Andi Kleen]

On Tue, Sep 17, 2013 at 08:49:42AM +0200, Dominik Vogt wrote:
> On Mon, Sep 16, 2013 at 06:15:38PM +0200, Andi Kleen wrote:
> > The Intel SOM currently has three:
> > 
> > _XABORT_LOCK_BUSY 0xff lock busy
> > _XABORT_ISLOCKED  0xfe lock is locked (not in pthread, but used elsewhere) 
> > _XABORT_TRYLOCK   0xfd nested trylock
> > 0xf0..0xfc reserved
> > 
> > I was considering to add a fourth:
> > 
> > _XABORT_TEMPORARY 0xfc temporary failure, do not adapt
> > 
> > e.g. to add to the dynamic linker, which currently aborts, to prevent
> > useless adaptation at program startup.
> 
> Does that mean, if you use the Xabort instruction, you have no way
> of controlling bit 1 of the abort status (the retry bit)?

That is correct, it's always 0.

-Andi