From: Will Deacon <will.deacon@arm.com>
To: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Cc: "linux-arch@vger.kernel.org" <linux-arch@vger.kernel.org>,
"linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>,
"arnd@arndb.de" <arnd@arndb.de>,
"monstr@monstr.eu" <monstr@monstr.eu>,
"dhowells@redhat.com" <dhowells@redhat.com>,
"broonie@linaro.org" <broonie@linaro.org>,
"peterz@infradead.org" <peterz@infradead.org>,
"paulmck@linux.vnet.ibm.com" <paulmck@linux.vnet.ibm.com>
Subject: Re: [PATCH v2 00/18] Cross-architecture definitions of relaxed MMIO accessors
Date: Tue, 27 May 2014 20:32:19 +0100 [thread overview]
Message-ID: <20140527193219.GB30751@arm.com> (raw)
In-Reply-To: <1401054363.3958.28.camel@pasglop>
Hi Ben,
On Sun, May 25, 2014 at 10:46:03PM +0100, Benjamin Herrenschmidt wrote:
> On Thu, 2014-05-22 at 17:47 +0100, Will Deacon wrote:
> > A corollary to this is that mmiowb() probably needs rethinking. As it currently
> > stands, an mmiowb() is required to order MMIO writes to a device from multiple
> > CPUs, even if that device is protected by a lock. However, this isn't often used
> > in practice, leading to PowerPC implementing both mmiowb() *and* synchronising
> > I/O in spin_unlock.
> >
> > I would propose making the non-relaxed I/O accessors ordered with respect to
> > LOCK/UNLOCK, leaving mmiowb() to be used with the relaxed accessors, if
> > required, but would welcome thoughts/suggestions on this topic.
>
> I agree on the proposed semantics, though for us that does mean we still need
> that per-cpu flag tracking non-relaxed MMIO stores and corresponding added barrier
> in unlock. Eventually, if the use of the relaxed accessors becomes pervasive
> enough I suppose I can just make the ordered ones unconditionally do 2 barriers.
Why would you need two barriers? I would have though an mmiowb() inlined
into writel after the store operation would be sufficient. Or is this to
ensure a non-relaxed write is ordered with respect to a relaxed write?
Anyway, we may need something similar for other architectures with mmiowb
implementations:
blackfin
frv
ia64
mips
sh
so I'm anticipating some more discussion when I try to push that patch :)
Cheers,
Will
next prev parent reply other threads:[~2014-05-27 19:32 UTC|newest]
Thread overview: 43+ messages / expand[flat|nested] mbox.gz Atom feed top
2014-05-22 16:47 [PATCH v2 00/18] Cross-architecture definitions of relaxed MMIO accessors Will Deacon
2014-05-22 16:47 ` [PATCH v2 01/18] asm-generic: io: implement relaxed accessor macros as conditional wrappers Will Deacon
2014-05-22 16:47 ` [PATCH v2 02/18] microblaze: io: remove dummy relaxed accessor macros Will Deacon
2014-05-22 16:47 ` [PATCH v2 03/18] s390: io: remove dummy relaxed accessor macros for reads Will Deacon
2014-05-22 16:47 ` [PATCH v2 04/18] xtensa: " Will Deacon
2014-05-22 16:47 ` [PATCH v2 05/18] alpha: io: implement relaxed accessor macros for writes Will Deacon
2014-05-22 18:15 ` Richard Henderson
2014-05-22 16:47 ` [PATCH v2 06/18] frv: io: implement dummy " Will Deacon
2014-05-22 16:47 ` [PATCH v2 07/18] cris: " Will Deacon
2014-05-22 16:47 ` [PATCH v2 08/18] ia64: " Will Deacon
2014-05-22 16:47 ` [PATCH v2 09/18] m32r: " Will Deacon
2014-05-22 16:47 ` [PATCH v2 10/18] m68k: " Will Deacon
2014-05-22 16:47 ` [PATCH v2 11/18] mn10300: " Will Deacon
2014-05-22 16:47 ` [PATCH v2 12/18] parisc: " Will Deacon
2014-05-22 16:47 ` [PATCH v2 13/18] powerpc: " Will Deacon
2014-05-22 16:47 ` [PATCH v2 14/18] sparc: " Will Deacon
2014-05-22 18:18 ` Sam Ravnborg
2014-05-23 14:38 ` Will Deacon
2014-05-30 0:10 ` David Miller
2014-05-22 16:47 ` [PATCH v2 15/18] tile: " Will Deacon
2014-05-22 16:47 ` [PATCH v2 16/18] x86: " Will Deacon
2014-05-22 17:15 ` H. Peter Anvin
2014-05-23 14:46 ` Will Deacon
2014-05-23 14:53 ` H. Peter Anvin
2014-05-23 14:57 ` Will Deacon
2014-05-23 15:20 ` H. Peter Anvin
2014-05-23 15:34 ` Will Deacon
2014-05-23 15:43 ` H. Peter Anvin
2014-05-23 15:56 ` Peter Zijlstra
2014-05-23 16:12 ` H. Peter Anvin
2014-05-23 16:21 ` Peter Zijlstra
2014-05-23 16:31 ` Geert Uytterhoeven
2014-05-23 16:35 ` H. Peter Anvin
2014-05-22 16:47 ` [PATCH v2 17/18] documentation: memory-barriers: clarify relaxed io accessor semantics Will Deacon
2014-05-22 16:47 ` [PATCH v2 18/18] asm-generic: io: define relaxed accessor macros unconditionally Will Deacon
2014-05-25 21:46 ` [PATCH v2 00/18] Cross-architecture definitions of relaxed MMIO accessors Benjamin Herrenschmidt
2014-05-27 19:32 ` Will Deacon [this message]
2014-05-27 20:21 ` Benjamin Herrenschmidt
2014-05-27 20:32 ` Will Deacon
2014-05-25 21:47 ` Benjamin Herrenschmidt
2014-05-27 19:34 ` Will Deacon
2014-05-27 20:23 ` Benjamin Herrenschmidt
2014-05-27 20:34 ` Will Deacon
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20140527193219.GB30751@arm.com \
--to=will.deacon@arm.com \
--cc=arnd@arndb.de \
--cc=benh@kernel.crashing.org \
--cc=broonie@linaro.org \
--cc=dhowells@redhat.com \
--cc=linux-arch@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=monstr@monstr.eu \
--cc=paulmck@linux.vnet.ibm.com \
--cc=peterz@infradead.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).