[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20150512070117.GV12671@pengutronix.de>
Date: Tue, 12 May 2015 09:01:17 +0200
From: Uwe Kleine-König
<u.kleine-koenig@...gutronix.de>
To: Philippe Reynes <tremyfr@...il.com>
Cc: Shawn Guo <shawnguo@...nel.org>, mark.rutland@....com,
devicetree@...r.kernel.org, linux@....linux.org.uk,
pawel.moll@....com, ijc+devicetree@...lion.org.uk,
linux-kernel@...r.kernel.org, robh+dt@...nel.org,
kernel@...gutronix.de, galak@...eaurora.org, shawn.guo@...aro.org,
linux-arm-kernel@...ts.infradead.org
Subject: Re: [PATCH] imx27: dt: only map 1 Kbyte for fec registers
Hello,
On Mon, May 11, 2015 at 11:25:08PM +0200, Philippe Reynes wrote:
> On 11/05/15 14:01, Shawn Guo wrote:
> >On Sat, May 09, 2015 at 10:54:30PM +0200, Philippe Reynes wrote:
> >>According to the imx27 documentation, fec has a 1 Kbyte
> >>memory space map, spitted in two regions of 512 bytes.
> >>The first one for control/status registers, and the
> >>second one for event/statistic registers. So, we don't
> >>need to map 16 Kbyte for registers, 1 Kbyte is enough.
> >>
> >>Signed-off-by: Philippe Reynes<tremyfr@...il.com>
> >>---
> >> arch/arm/boot/dts/imx27.dtsi | 2 +-
> >> 1 files changed, 1 insertions(+), 1 deletions(-)
> >>
> >>diff --git a/arch/arm/boot/dts/imx27.dtsi b/arch/arm/boot/dts/imx27.dtsi
> >>index 6951b66..56bb917 100644
> >>--- a/arch/arm/boot/dts/imx27.dtsi
> >>+++ b/arch/arm/boot/dts/imx27.dtsi
> >>@@ -533,7 +533,7 @@
> >>
> >> fec: ethernet@...2b000 {
> >> compatible = "fsl,imx27-fec";
> >>- reg =<0x1002b000 0x4000>;
> >>+ reg =<0x1002b000 0x400>;
> >
> >No. Per MCIMX27RM.pdf, Table 2-7. AIPI2 Memory Map, it should be 4KiB.
>
> I agree, this table show that 4KiB is reserved for fec registers.
> But, in paragraph 29.6.1, there is :
> "The FEC implementation requires a 1-Kbyte memory map space"
> So I've thought that 1 Kbye is enough for the register memory space.
The table 2-7 suggests that 4 KiB are routed to the fec even though the
fec module might only make use of the first 1 KiB. That's no
contradiction. The convention used in the device trees is that the first
amount is used.
> I'm pleased to understand that we're both agree that 16 Kbyte is too large.
>
> If you prefer 4 Kbyte, I'll send a v2 of this patch with this value.
Yes please. In the commit log you might want to point out that a length
of 16 KiB overlaps with the (currently unused?) Security Controller
(SCC) of the i.MX27.
Best regards
Uwe
--
Pengutronix e.K. | Uwe Kleine-König |
Industrial Linux Solutions | http://www.pengutronix.de/ |
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
Powered by blists - more mailing lists