[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <5956667B.6040903@mentor.com>
Date: Fri, 30 Jun 2017 07:55:55 -0700
From: Jiada Wang <jiada_wang@...tor.com>
To: Sascha Hauer <s.hauer@...gutronix.de>
CC: Mark Brown <broonie@...nel.org>, <festevam@...il.com>,
<linux-spi@...r.kernel.org>, <linux-kernel@...r.kernel.org>
Subject: Re: Applied "spi: imx: dynamic burst length adjust for PIO mode"
to the spi tree
Hello Sascha
On 06/30/2017 05:14 AM, Sascha Hauer wrote:
> Hi Mark,
>
> On Fri, Jun 30, 2017 at 01:00:22PM +0100, Mark Brown wrote:
>> The patch
>>
>> spi: imx: dynamic burst length adjust for PIO mode
>>
>> has been applied to the spi tree at
>>
>> git://git.kernel.org/pub/scm/linux/kernel/git/broonie/spi.git
> Argh, I wasn't fast enough. I just ran spi-loopback-test.ko with this
> patch and it doesn't work properly.
>
>> + if (spi_imx->count == spi_imx->remainder) {
>> + ctrl = readl(spi_imx->base + MX51_ECSPI_CTRL);
>> + ctrl&= ~MX51_ECSPI_CTRL_BL_MASK;
>> + if (spi_imx->count> MX51_ECSPI_CTRL_MAX_BURST) {
>> + spi_imx->remainder = spi_imx->count %
>> + MX51_ECSPI_CTRL_MAX_BURST;
>> + val = MX51_ECSPI_CTRL_MAX_BURST;
> This is wrong. MX51_ECSPI_CTRL_MAX_BURST contains the burst length in
> bytes, but the register 'val' is written to takes the burst length in
> bits - 1, so this should be:
>
> val = MX51_ECSPI_CTRL_MAX_BURST * 8 - 1;
>
> instead.
Thanks for pointing this out,
I will submit a fix patch.
Thanks,
Jiada
> Sascha
>
Powered by blists - more mailing lists