lkml.org 
[lkml]   [2018]   [Feb]   [27]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
Date
From
SubjectRe: [RFC PATCH] riscv/locking: Strengthen spin_lock() and spin_unlock()
On Tue, Feb 27, 2018 at 01:06:35PM +0800, Boqun Feng wrote:
> On Mon, Feb 26, 2018 at 04:24:27PM +0000, Will Deacon wrote:
> > On Mon, Feb 26, 2018 at 08:06:59AM -0800, Linus Torvalds wrote:
> > > On Mon, Feb 26, 2018 at 6:21 AM, Luc Maranget <luc.maranget@inria.fr> wrote:
> > > >
> > > > That is, locks are not implemented from more basic primitive but are specified.
> > > > The specification can be described as behaving that way:
> > > > - A lock behaves as a read-modify-write. the read behaving as a read-acquire
> > >
> > > This is wrong, or perhaps just misleading.
> > >
> > > The *whole* r-m-w acts as an acquire. Not just the read part. The
> > > write is very much part of it.
> > >
> > > Maybe that's what you meant, but it read to me as "just the read part
> > > of the rmw behaves as a read-acquire".
> > >
> > > Because it is very important that the _write_ part of the rmw is also
> > > ordered wrt everything that is inside the spinlock.
> > >
> > > So doing a spinlock as
> > >
> > > (a) read-locked-acquire
> > > modify
> > > (c) write-conditional
> > >
> > > would be wrong, because the accesses inside the spinlock are ordered
> > > not just wrt the read-acquire, they have to be ordered wrt the write
> > > too.
> > >
> > > So it is closer to say that it's the _write_ of the r-m-w sequence
> > > that has the acquire semantics, not the read.
> >
> > Strictly speaking, that's not what we've got implemented on arm64: only
> > the read part of the RmW has Acquire semantics, but there is a total
> > order on the lock/unlock operations for the lock. For example, if one
> > CPU does:
> >
> > spin_lock(&lock);
> > WRITE_ONCE(foo, 42);
> >
> > then another CPU could do:
> >
> > if (smp_load_acquire(&foo) == 42)
> > BUG_ON(!spin_is_locked(&lock));
> >
>
> Hmm.. this is new to me. So the write part of spin_lock() and the
> WRITE_ONCE() will not get reordered? Could you explain more about this
> or point where I should look in the document? I understand the write
> part of spin_lock() must be committed earlier than the WRITE_ONCE()
> committed due to the ll/sc, but I thought the ordering of their arrivals
> in memory system is undefined/arbitary.
>
> Regards,
> Boqun
>
> > and that could fire. Is that relied on somewhere?
> >

My bad, I misunderstood your mail. So you were saying the BUG_ON() can
be triggered in currently implementations. Never mind my reply above.

Regards,
Boqun

> > Will


[unhandled content-type:application/pgp-signature]
\
 
 \ /
  Last update: 2018-02-27 11:13    [W:0.070 / U:2.692 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site