Message-ID: <1312996459.12609.1505761819605.JavaMail.zimbra@efficios.com>
Date:   Mon, 18 Sep 2017 19:10:19 +0000 (UTC)
From:   Mathieu Desnoyers <mathieu.desnoyers@...icios.com>
To:     Alan Stern <stern@...land.harvard.edu>
Cc:     "Paul E. McKenney" <paulmck@...ux.vnet.ibm.com>,
        Peter Zijlstra <peterz@...radead.org>,
        Will Deacon <will.deacon@....com>,
        Andy Lutomirski <luto@...nel.org>,
        Michael Ellerman <mpe@...erman.id.au>,
        linux-kernel <linux-kernel@...r.kernel.org>,
        linux-arch@...r.kernel.org, Dave Watson <davejwatson@...com>,
        maged michael <maged.michael@...il.com>
Subject: Re: Rough notes from sys_membarrier() lightning BoF

----- On Sep 18, 2017, at 3:04 PM, Alan Stern stern@...land.harvard.edu wrote:

> On Sun, 17 Sep 2017, Paul E. McKenney wrote:
> 
>> Hello!
>> 
>> Rough notes from our discussion last Thursday.  Please reply to the
>> group with any needed elaborations or corrections.
>> 
>> Adding Andy and Michael on CC since this most closely affects their
>> architectures.  Also adding Dave Watson and Maged Michael because
>> the preferred approach requires that processes wanting to use the
>> lightweight sys_membarrier() do a registration step.
>> 
>> 							Thanx, Paul
>> 
>> ------------------------------------------------------------------------
>> 
>> Problem:
>> 
>> 1.	The current sys_membarrier() introduces an smp_mb() that
>> 	is not otherwise required on powerpc.
>> 
>> 2.	The envisioned JIT variant of sys_membarrier() assumes that
>> 	the return-to-user instruction sequence handling any change
>> 	to the usermode instruction stream, and Andy Lutomirski's
>> 	upcoming changes invalidate this assumption.  It is believed
>> 	that powerpc has a similar issue.
> 
>> E.	Require that threads register before using sys_membarrier() for
>> 	private or JIT usage.  (The historical implementation using
>> 	synchronize_sched() would continue to -not- require registration,
>> 	both for compatibility and because there is no need to do so.)
>> 
>> 	For x86 and powerpc, this registration would set a TIF flag
>> 	on all of the current process's threads.  This flag would be
>> 	inherited by any later thread creation within that process, and
>> 	would be cleared by fork() and exec().	When this TIF flag is set,
> 
> Why a TIF flag, and why clear it during fork()?  If a process registers
> to use private expedited sys_membarrier, shouldn't that apply to
> threads it will create in the future just as much as to threads it has
> already created?

In my implementation posted today, I'm not clearing it on fork. The child
inherits from the parent.

Why TIF flag ? It appears to be a convenient way to add an architecture-specific
single-bit state for each thread. We also don't want to do too much pointer
chasing on the scheduler fast-path (current->mm->..).

> 
>> 	the return-to-user path would execute additional code that would
>> 	ensure that ordering and newly JITed code was handled correctly.
>> 	We believe that checks for these TIF flags could be combined with
>> 	existing checks to avoid adding any overhead in the common case
>> 	where the process was not using these sys_membarrier() features.
>> 
>> 	For all other architecture, the registration step would be
>> 	a no-op.
> 
> Don't we want to fail private expedited sys_membarrier calls if the
> process hasn't registered for them?  This requires the registration
> call to set a flag for the process, even on architectures where no
> additional memory barriers are actually needed.  It can't be a no-op.

My implementation posted today fails the private expedited command
if the process is not registered yet. We indeed add a new flag in
mm_struct for all architectures to do so.

So why not re-use this flag instead of the TIF on powerpc ? See my
pointer chasing on fast-path argument above.

Thanks,

Mathieu

> 
> Alan Stern

-- 
Mathieu Desnoyers
EfficiOS Inc.
http://www.efficios.com

Powered by blists - more mailing lists