[RFC PATCH] riscv/locking: Strengthen spin_lock() and spin_unlock()

[Andrea Parri]

On Thu, Feb 22, 2018 at 02:40:04PM +0100, Peter Zijlstra wrote:
> On Thu, Feb 22, 2018 at 01:19:50PM +0100, Andrea Parri wrote:
> 
> > C unlock-lock-read-ordering
> > 
> > {}
> > /* s initially owned by P1 */
> > 
> > P0(int *x, int *y)
> > {
> > 	WRITE_ONCE(*x, 1);
> > 	smp_wmb();
> > 	WRITE_ONCE(*y, 1);
> > }
> > 
> > P1(int *x, int *y, spinlock_t *s)
> > {
> > 	int r0;
> > 	int r1;
> > 
> > 	r0 = READ_ONCE(*y);
> > 	spin_unlock(s);
> > 	spin_lock(s);
> > 	r1 = READ_ONCE(*y);
> > }
> > 
> > exists (1:r0=1 /\ 1:r1=0)
> > 
> > RISCV RISCV-unlock-lock-read-ordering
> > {
> > 0:x2=x; 0:x4=y;
> > 1:x2=y; 1:x4=x; 1:x6=s;
> > s=1;
> > }
> >  P0           |  P1                      ;
> >  ori x1,x0,1  | lw x1,0(x2)              ;
> >  sw x1,0(x2)  | amoswap.w.rl x0,x0,(x6)  ;
> >  fence w,w    | ori x5,x0,1              ;
> >  ori x3,x0,1  | amoswap.w.aq x0,x5,(x6)  ;
> >  sw x3,0(x4)  | lw x3,0(x4)              ;
> > exists
> > (1:x1=1 /\ 1:x3=0)
> 
> So I would indeed expect this to be forbidden. Could someone please
> explain how this could be allowed?

As mentioned in IRC, my understanding here is only based on the spec.
referred below and on its (available) formalizations.  I expect that
RISC-V people will be able to provide more information.


> 
> > C unlock-lock-write-ordering
> > 
> > {}
> > /* s initially owned by P0 */
> > 
> > P0(int *x, int *y, spinlock_t *s)
> > {
> > 	WRITE_ONCE(*x, 1);
> > 	spin_unlock(s);
> > 	spin_lock(s);
> > 	WRITE_ONCE(*y, 1);
> > }
> > 
> > P1(int *x, int *y)
> > {
> > 	int r0;
> > 	int r1;
> > 
> > 	r0 = READ_ONCE(*y);
> > 	smp_rmb();
> > 	r1 = READ_ONCE(*y);
> > }
> > 
> > exists (1:r0=1 /\ 1:r1=0)
> > 
> > RISCV RISCV-unlock-lock-write-ordering
> > {
> > 0:x2=x; 0:x4=y; 0:x6=s;
> > 1:x2=y; 1:x4=x;
> > s=1;
> > }
> >  P0                       | P1           ;
> >  ori x1,x0,1              | lw x1,0(x2)  ;
> >  sw x1,0(x2)              | fence r,r    ;
> >  amoswap.w.rl x0,x0,(x6)  | lw x3,0(x4)  ;
> >  ori x5,x0,1              |              ;
> >  amoswap.w.aq x0,x5,(x6)  |              ;
> >  ori x3,x0,1              |              ;
> >  sw x3,0(x4)              |              ;
> > exists
> > (1:x1=1 /\ 1:x3=0)
> 
> And here I think the RISCV conversion is flawed, there should be a ctrl
> dependency. The second store-word in P0 should depend on the result of
> amoswap.w.aq being 0.

You're right: AFAICT, this can be remedied by inserting "beq x0,x5,FAIL00"
right after amoswap.w.aq (and this label at the end of P0); this does not
change the verdict of the available formalizations reported above however.

(So, AFAICT, the above question remains valid/open.)

  Andrea


> 
> (strictly speaking there should be a ctrl-dep in the read example too,
> except it'd be pointless for ordering reads, so I accept it being left
> out)
> 
> Again, I cannot see how this could be allowed.
>