Re: [Qemu-devel] [RFC 01/10] target-arm: protect cpu_exclusive_*.

[Dr. David Alan Gilbert]

* Mark Burton (address@hidden) wrote:
> 
> > On 3 Mar 2015, at 16:32, Paolo Bonzini <address@hidden> wrote:
> > 
> > 
> > 
> > On 03/03/2015 16:29, Mark Burton wrote:
> >> 
> >> ps. on our bug - we believe somehow the STREX is being marked as
> >> failed, but actually succeeds to write something.  There are only 3
> >> ways the strex can fail: 1/ the address doesn't match - in which case
> >> no ST will be attempted 2/ the value doesn't match - which means - no
> >> ST attempted 3/ the store starts, but causes a TLB fill/exception???
> >> 
> >> The 3rd has 2 possibilities - the TLB is filled, and the store goes
> >> ahead totally normally - there should be no ???fail??? - or an exception
> >> is generated in which case we will long jump away and never return.
> > 
> > When do you release the lock?
> > 
> (Thanks Paolo!)
> 
> We release the lock in either
> a) the end of the strex
> or
> b) in the ???raise_exception???

That works for ARM where you have to terminate a ldrex with a strex or clrex,
but not all architectures have the equivalent of a clrex; most as I remember
just let you do an ldrex equivalent, decide you don't want to do the strex 
equivalent
and get on with life. (Was clrex originally in ARM when ldrex went in?)

Dave

> 
> Cheers
> 
> Mark.
> 
> > Paolo
> 
> 
>        +44 (0)20 7100 3485 x 210
>  +33 (0)5 33 52 01 77x 210
> 
>       +33 (0)603762104
>       mark.burton
> 
--
Dr. David Alan Gilbert / address@hidden / Manchester, UK