Re: [RFC PATCH v5 1/3] printk-rb: new printk ringbuffer implementation (writer)

[Andrea Parri]

On Fri, Jan 03, 2020 at 11:24:20AM +0100, Petr Mladek wrote:
> On Mon 2019-12-23 17:01:00, John Ogness wrote:
> > Hi Andrea,
> > 
> > On 2019-12-21, Andrea Parri  wrote:
> > >> +*desc_out = READ_ONCE(*desc);
> > >> +
> > >> +/* Load data before re-checking state. */
> > >> +smp_rmb(); /* matches LMM_REF(desc_reserve:A) */
> > >
> > > I looked for a matching WRITE_ONCE() or some other type of marked write,
> > > but I could not find it.  What is the rationale?  Or what did I miss?
> 
> Good question. READ_ONCE() looks superfluous here because it is
> surrounded by two read barriers. In each case, there is no
> corresponding WRITE_ONCE().
> 
> Note that we are copying the entire struct prb_desc here. All values
> are written only when state_val is in desc_reserved state. It happens
> between two full write barriers:
> 
>   + A writer is allowed to modify the descriptor after successful
> cmpxchg in desc_reserve(), see LMM_TAG(desc_reserve:A).
> 
>   + The writer must not touch the descriptor after changing
> state_var to committed state, see
> LMM_TAG(prb_commit:A) in prb_commit().
> 
> These barriers are mentioned in the comments for the two
> read barriers here.

Thanks for these remarks.  As usual, I'd recommend to (try to) map those
comments into litmus tests and check with the LKMM simulator.


> BTW: Documentation/memory-barriers.txt describes various aspects of
> the memory barriers. It describes implicit barriers provided
> by spin locks, mutexes, semaphores, and various scheduler-related
> operations.
> 
> But I can't find any explanation of the various variants of the atomic
> operations: acquire, release, fetch, return, try, relaxed. I can find
> some clues here and there but it is hard to get the picture.

Documentation/atomic_t.txt could serve this purpose.  Please have a look
there and let me know if you have any comments.

Thanks,
  Andrea

___
kexec mailing list
kexec@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/kexec

Re: [RFC PATCH v5 1/3] printk-rb: new printk ringbuffer implementation (writer)

[Andrea Parri]

Hi John,

Sorry for the delay.

I don't have an overall understanding of the patch(-set) yet, so I limit
to a couple of general questions about the memory barriers introduced by
the path.  Please see inline comments.


> + *desc_out = READ_ONCE(*desc);
> +
> + /* Load data before re-checking state. */
> + smp_rmb(); /* matches LMM_REF(desc_reserve:A) */

I looked for a matching WRITE_ONCE() or some other type of marked write,
but I could not find it.  What is the rationale?  Or what did I miss?


> + do {
> + next_lpos = get_next_lpos(data_ring, begin_lpos, size);
> +
> + if (!data_push_tail(rb, data_ring,
> + next_lpos - DATA_SIZE(data_ring))) {
> + /* Failed to allocate, specify a data-less block. */
> + blk_lpos->begin = INVALID_LPOS;
> + blk_lpos->next = INVALID_LPOS;
> + return NULL;
> + }
> + } while (!atomic_long_try_cmpxchg(_ring->head_lpos, _lpos,
> +   next_lpos));
> +
> + /*
> +  * No barrier is needed here. The data validity is defined by
> +  * the state of the associated descriptor. They are marked as
> +  * invalid at the moment. And only the winner of the above
> +  * cmpxchg() could write here.
> +  */

The (successful) CMPXCHG provides a full barrier.  This comment suggests
that that could be somehow relaxed?  Or the comment could be improved?

(The patch introduces a number of CMPXCHG: similar questions would apply
to those other instances...)

Thanks,
  Andrea

P. S.  Please use my @gmail.com address for future communications.

___
kexec mailing list
kexec@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/kexec