public inbox for linux-kernel@vger.kernel.org
 help / color / mirror / Atom feed
From: Paul Walmsley <paul@pwsan.com>
To: linux-arm-kernel@lists.arm.linux.org.uk, linux-kernel@vger.kernel.org
Cc: linux-omap@vger.kernel.org, Paul Walmsley <paul@pwsan.com>,
	Tony Lindgren <tony@atomide.com>
Subject: [PATCH F 01/12] OMAP2/3 clock: don't use a barrier after clk_disable()
Date: Wed, 28 Jan 2009 12:35:01 -0700	[thread overview]
Message-ID: <20090128193456.2396.29131.stgit@localhost.localdomain> (raw)
In-Reply-To: <20090128193326.2396.9437.stgit@localhost.localdomain>

clk_disable() previously used an ARM barrier, wmb(), to try to ensure
that the hardware write completed before continuing.  There are some
problems with this approach.

The first problem is that wmb() only ensures that the write leaves the
ARM -- not that it actually reaches the endpoint device.  In this
case, the endpoint device - either the PRM, CM, or SCM - is three
interconnects away from the ARM, and the final interconnect is
low-speed.  And the OCP interconnects will post the write, who knows
how long that will take to complete.  So the wmb() is not really what
we want.

Worse, the wmb() is indiscriminate; it will cause the ARM to flush any
other unrelated buffered writes and wait for the local interconnect to
acknowledge them - potentially very expensive.

This first problem could be fixed by doing a readback of the same PRM/CM/SCM
register.  Since these devices use a single OCP thread, this will cause the
MPU to wait for the write to complete.

But the primary problem is a conceptual one: clk_disable() should not
need any kind of barrier.  clk_enable() needs one since device driver
code must not access a device until its clocks are known to be
enabled.  But clk_disable() has no such restriction.

Since blocking the MPU on a PRM/CM/SCM write can be a very
high-latency operation - several hundred MPU cycles - it's worth
avoiding this barrier if possible.

linux-omap source commit is f4aacad2c0ed1055622d5c1e910befece24ef0e2.

Signed-off-by: Paul Walmsley <paul@pwsan.com>
Signed-off-by: Tony Lindgren <tony@atomide.com>
---
 arch/arm/mach-omap2/clock.c |    2 +-
 1 files changed, 1 insertions(+), 1 deletions(-)

diff --git a/arch/arm/mach-omap2/clock.c b/arch/arm/mach-omap2/clock.c
index 07b6232..a1ccdcb 100644
--- a/arch/arm/mach-omap2/clock.c
+++ b/arch/arm/mach-omap2/clock.c
@@ -448,7 +448,7 @@ static void _omap2_clk_disable(struct clk *clk)
 	else
 		v &= ~(1 << clk->enable_bit);
 	_omap2_clk_write_reg(v, clk->enable_reg, clk);
-	wmb();
+	/* No OCP barrier needed here since it is a disable operation */
 }
 
 void omap2_clk_disable(struct clk *clk)



  reply	other threads:[~2009-01-28 20:48 UTC|newest]

Thread overview: 21+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2009-01-28 19:34 [PATCH F 00/12] OMAP clock, F of F: more clock cleanup Paul Walmsley
2009-01-28 19:35 ` Paul Walmsley [this message]
2009-01-28 19:35 ` [PATCH F 02/12] OMAP2xxx clock: consolidate DELAYED_APP clock commits; fix barrier Paul Walmsley
2009-01-28 19:35 ` [PATCH F 03/12] OMAP2/3 clock: convert remaining MPU barriers into OCP barriers Paul Walmsley
2009-01-28 19:35 ` [PATCH F 04/12] OMAP clock: drop clk_get_usecount() Paul Walmsley
2009-01-28 19:35 ` [PATCH F 05/12] OMAP clock: add OMAP chip family-specific clk_register() option Paul Walmsley
2009-01-28 19:35 ` [PATCH F 06/12] OMAP2/3 clock: every clock must have a clkdm Paul Walmsley
2009-01-31 14:08   ` Russell King - ARM Linux
2009-02-03  9:27     ` Paul Walmsley
2009-02-03 14:20       ` Woodruff, Richard
2009-02-05  9:03         ` Paul Walmsley
2009-01-28 19:35 ` [PATCH F 07/12] OMAP2/3 clock: omap2_clk_enable(): refactor usecount check Paul Walmsley
2009-01-28 19:35 ` [PATCH F 08/12] OMAP2/3 clock: omap2_clk_enable(): fix bugs in clockdomain handling Paul Walmsley
2009-01-28 19:35 ` [PATCH F 09/12] OMAP2/3 clock: omap2_clk_enable(): fix usecount decrement bug Paul Walmsley
2009-01-28 19:35 ` [PATCH F 10/12] OMAP2/3 clock: omap2_clk_enable(): fix logic Paul Walmsley
2009-01-31 10:58   ` Russell King - ARM Linux
2009-02-02  8:01     ` Paul Walmsley
2009-01-28 19:35 ` [PATCH F 11/12] OMAP2/3 clock: don't tinker with hardirqs when they are supposed to be disabled Paul Walmsley
2009-01-28 19:35 ` [PATCH F 12/12] OMAP2/3 McBSP: add temporary clockdomain fix for McBSP virtual clocks Paul Walmsley
2009-01-31 11:42   ` Russell King - ARM Linux
2009-02-02  8:09     ` Paul Walmsley

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=20090128193456.2396.29131.stgit@localhost.localdomain \
    --to=paul@pwsan.com \
    --cc=linux-arm-kernel@lists.arm.linux.org.uk \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-omap@vger.kernel.org \
    --cc=tony@atomide.com \
    /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