public inbox for linux-kernel@vger.kernel.org
 help / color / mirror / Atom feed
From: Robin Murphy <robin.murphy@arm.com>
To: "Mark Brown" <broonie@kernel.org>,
	"Andreas Färber" <afaerber@suse.de>,
	"Heiner Kallweit" <hkallweit1@gmail.com>,
	"Christian Hewitt" <christianshewitt@gmail.com>
Cc: linux-spi@vger.kernel.org, linux-kernel@vger.kernel.org
Subject: Re: [PATCH] spi: bitbang: Fix lsb-first Rx
Date: Wed, 3 Aug 2022 16:22:13 +0100	[thread overview]
Message-ID: <ae7e7d7b-bf90-7f23-8a27-e6358b3ff4f6@arm.com> (raw)
In-Reply-To: <YuqOBYJSP8Pm7Bz5@sirena.org.uk>

On 2022-08-03 16:02, Mark Brown wrote:
> On Wed, Aug 03, 2022 at 03:58:57PM +0100, Robin Murphy wrote:
>> Shifting the recieved bit by "bits" inserts it at the top of the
>> *currently remaining* Tx data, so we end up accumulating the whole
>> transfer into bit 0 of the output word. Oops.
>>
>> For the algorithm to work as intended, we need to remember where the
>> top of the *original* word was, and shift Rx to there.
> 
> So if this never worked we presumably have some systems out there which
> somehow rely on the old behaviour that we need to fix somehow - copying
> in everyone from the original change...

Hmm, I can't imagine anyone's relying too critically on data transfer 
corruption :/

Apparently these TM16x8 chips are the only thing so far to even want 
lsb-first bitbang support at all, so I strongly suspect I'm the first 
person to try the receive side of it (since Rockchip's SPI hardware 
doesn't seem to support 3-wire half-duplex). It actually took me quite a 
while to figure out that this wasn't my bug in trying to hook up the 
matrix-keymap stuff, since half the bytes I was expecting to get back do 
only have bit 0 set anyway.

Cheers,
Robin.

>> Fixes: 1847e3046c52 ("spi: gpio: Implement LSB First bitbang support")
>> Signed-off-by: Robin Murphy <robin.murphy@arm.com>
>> ---
>>   drivers/spi/spi-bitbang-txrx.h | 6 ++++--
>>   1 file changed, 4 insertions(+), 2 deletions(-)
>>
>> diff --git a/drivers/spi/spi-bitbang-txrx.h b/drivers/spi/spi-bitbang-txrx.h
>> index 267342dfa738..2dcbe166df63 100644
>> --- a/drivers/spi/spi-bitbang-txrx.h
>> +++ b/drivers/spi/spi-bitbang-txrx.h
>> @@ -116,6 +116,7 @@ bitbang_txrx_le_cpha0(struct spi_device *spi,
>>   {
>>   	/* if (cpol == 0) this is SPI_MODE_0; else this is SPI_MODE_2 */
>>   
>> +	u8 rxbit = bits - 1;
>>   	u32 oldbit = !(word & 1);
>>   	/* clock starts at inactive polarity */
>>   	for (; likely(bits); bits--) {
>> @@ -135,7 +136,7 @@ bitbang_txrx_le_cpha0(struct spi_device *spi,
>>   		/* sample LSB (from slave) on leading edge */
>>   		word >>= 1;
>>   		if ((flags & SPI_MASTER_NO_RX) == 0)
>> -			word |= getmiso(spi) << (bits - 1);
>> +			word |= getmiso(spi) << rxbit;
>>   		setsck(spi, cpol);
>>   	}
>>   	return word;
>> @@ -148,6 +149,7 @@ bitbang_txrx_le_cpha1(struct spi_device *spi,
>>   {
>>   	/* if (cpol == 0) this is SPI_MODE_1; else this is SPI_MODE_3 */
>>   
>> +	u8 rxbit = bits - 1;
>>   	u32 oldbit = !(word & 1);
>>   	/* clock starts at inactive polarity */
>>   	for (; likely(bits); bits--) {
>> @@ -168,7 +170,7 @@ bitbang_txrx_le_cpha1(struct spi_device *spi,
>>   		/* sample LSB (from slave) on trailing edge */
>>   		word >>= 1;
>>   		if ((flags & SPI_MASTER_NO_RX) == 0)
>> -			word |= getmiso(spi) << (bits - 1);
>> +			word |= getmiso(spi) << rxbit;
>>   	}
>>   	return word;
>>   }
>> -- 
>> 2.36.1.dirty
>>

  reply	other threads:[~2022-08-03 15:22 UTC|newest]

Thread overview: 5+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2022-08-03 14:58 [PATCH] spi: bitbang: Fix lsb-first Rx Robin Murphy
2022-08-03 15:02 ` Mark Brown
2022-08-03 15:22   ` Robin Murphy [this message]
2022-08-03 16:33     ` Mark Brown
2022-08-15 15:58 ` Mark Brown

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=ae7e7d7b-bf90-7f23-8a27-e6358b3ff4f6@arm.com \
    --to=robin.murphy@arm.com \
    --cc=afaerber@suse.de \
    --cc=broonie@kernel.org \
    --cc=christianshewitt@gmail.com \
    --cc=hkallweit1@gmail.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-spi@vger.kernel.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