Re: [PATCH v2] tools/memory-model: Add extra ordering for locks and remove it for ordinary release/acquire

[Daniel Lustig <dlustig@nvidia.com>]

On 7/9/2018 1:01 PM, Alan Stern wrote:
> More than one kernel developer has expressed the opinion that the LKMM
> should enforce ordering of writes by locking.  In other words, given
> the following code:
> 
> 	WRITE_ONCE(x, 1);
> 	spin_unlock(&s):
> 	spin_lock(&s);
> 	WRITE_ONCE(y, 1);
> 
> the stores to x and y should be propagated in order to all other CPUs,
> even though those other CPUs might not access the lock s.  In terms of
> the memory model, this means expanding the cumul-fence relation.
> 
> Locks should also provide read-read (and read-write) ordering in a
> similar way.  Given:
> 
> 	READ_ONCE(x);
> 	spin_unlock(&s);
> 	spin_lock(&s);
> 	READ_ONCE(y);		// or WRITE_ONCE(y, 1);
> 
> the load of x should be executed before the load of (or store to) y.
> The LKMM already provides this ordering, but it provides it even in
> the case where the two accesses are separated by a release/acquire
> pair of fences rather than unlock/lock.  This would prevent
> architectures from using weakly ordered implementations of release and
> acquire, which seems like an unnecessary restriction.  The patch
> therefore removes the ordering requirement from the LKMM for that
> case.
> 
> All the architectures supported by the Linux kernel (including RISC-V)
> do provide this ordering for locks, albeit for varying reasons.
> Therefore this patch changes the model in accordance with the
> developers' wishes.
> 
> Signed-off-by: Alan Stern <stern@rowland.harvard.edu>
> 
> ---
> 
> v.2: Restrict the ordering to lock operations, not general release
> and acquire fences.
> 
> [as1871b]
> 
> 
>  tools/memory-model/Documentation/explanation.txt                           |  186 +++++++---
>  tools/memory-model/linux-kernel.cat                                        |    8 
>  tools/memory-model/litmus-tests/ISA2+pooncelock+pooncelock+pombonce.litmus |    5 
>  3 files changed, 149 insertions(+), 50 deletions(-)
> 
> Index: usb-4.x/tools/memory-model/linux-kernel.cat
> ===================================================================
> --- usb-4.x.orig/tools/memory-model/linux-kernel.cat
> +++ usb-4.x/tools/memory-model/linux-kernel.cat
> @@ -38,7 +38,7 @@ let strong-fence = mb | gp
>  (* Release Acquire *)
>  let acq-po = [Acquire] ; po ; [M]
>  let po-rel = [M] ; po ; [Release]
> -let rfi-rel-acq = [Release] ; rfi ; [Acquire]
> +let unlock-rf-lock-po = [UL] ; rf ; [LKR] ; po

It feels slightly weird that unlock-rf-lock-po is asymmetrical.  And in
fact, I think the current RISC-V solution we've been discussing (namely,
putting a fence.tso instead of a fence rw,w in front of the release)
may not even technically respect that particular sequence.  The
fence.tso solution really enforces "po; [UL]; rf; [LKR]", right?

Does something like "po; [UL]; rf; [LKR]; po" fit in with the rest
of the model?  If so, maybe that solves the asymmetry and also
legalizes the approach of putting fence.tso in front?

Or, other suggestions?

Dan

>  (**********************************)
>  (* Fundamental coherence ordering *)
> @@ -60,13 +60,13 @@ let dep = addr | data
>  let rwdep = (dep | ctrl) ; [W]
>  let overwrite = co | fr
>  let to-w = rwdep | (overwrite & int)
> -let to-r = addr | (dep ; rfi) | rfi-rel-acq
> +let to-r = addr | (dep ; rfi)
>  let fence = strong-fence | wmb | po-rel | rmb | acq-po
> -let ppo = to-r | to-w | fence
> +let ppo = to-r | to-w | fence | (unlock-rf-lock-po & int)
>  
>  (* Propagation: Ordering from release operations and strong fences. *)
>  let A-cumul(r) = rfe? ; r
> -let cumul-fence = A-cumul(strong-fence | po-rel) | wmb
> +let cumul-fence = A-cumul(strong-fence | po-rel) | wmb | unlock-rf-lock-po
>  let prop = (overwrite & ext)? ; cumul-fence* ; rfe?