Re: [PATCH linux-next v2 01/14] mtd: spi-nor: remove micron_quad_enable()

[Cyrille Pitchen]

Le 20/01/2016 06:13, Bean Huo 霍斌斌 (beanhuo) a écrit :
>> Message-ID:
>> 	<6177f7317fe11922e1d1b6dc4548afbaf0ccebdd.1452268345.git.cyrille.
>> pitchen@xxxxxxxxx>
>>
>> Content-Type: text/plain
>>
>> This patch remove the micron_quad_enable() function which force the Quad
>> SPI mode. However, once this mode is enabled, the Micron memory expect
>> ALL
>> commands to use the SPI 4-4-4 protocol. Hence a failure does occur when
>> calling spi_nor_wait_till_ready() right after the update of the Enhanced
>> Volatile Configuration Register (EVCR) in the micron_quad_enable() as
>> the SPI controller driver is not aware about the protocol change.
>> Since there is almost no performance increase using Fast Read 4-4-4
>> commands instead of Fast Read 1-1-4 commands, we rather keep on using
>> the
>> Extended SPI mode than enabling the Quad SPI mode.
>>
>> Let's take the example of the pretty standard use of 8 dummy cycles during
>> Fast Read operations on 64KB erase sectors:
>>
>> Fast Read 1-1-4 requires 8 cycles for the command, then 24 cycles for the
>> 3byte address followed by 8 dummy clock cycles and finally 65536*2 cycles
>> for the read data; so 131112 clock cycles.
>>
>> On the other hand the Fast Read 4-4-4 would require 2 cycles for the
>> command, then 6 cycles for the 3byte address followed by 8 dummy clock
>> cycles and finally 65536*2 cycles for the read data. So 131088 clock
>> cycles. The theorical bandwidth increase is 0.0%.
>>
>> Now using Fast Read operations on 512byte pages:
>> Fast Read 1-1-4 needs 8+24+8+(512*2) = 1064 clock cycles whereas Fast
>> Read 4-4-4 would requires 2+6+8+(512*2) = 1040 clock cycles. Hence the
>> theorical bandwidth increase is 2.3%.
>> Consecutive reads for non sequential pages is not a relevant use case so
>> The Quad SPI mode is not worth it.
>>
>> mtd_speedtest seems to confirm these figures.
>>
>> Signed-off-by: Cyrille Pitchen <cyrille.pitchen@xxxxxxxxx>
>> Fixes: 548cd3ab54da ("mtd: spi-nor: Add quad I/O support for Micron SPI
>> NOR")
>> ---
>>  drivers/mtd/spi-nor/spi-nor.c | 46 +------------------------------------------
>>  1 file changed, 1 insertion(+), 45 deletions(-)
>>
>> diff --git a/drivers/mtd/spi-nor/spi-nor.c b/drivers/mtd/spi-nor/spi-nor.c
>> index ed0c19c558b5..3028c06547c1 100644
>> --- a/drivers/mtd/spi-nor/spi-nor.c
>> +++ b/drivers/mtd/spi-nor/spi-nor.c
>> @@ -1100,45 +1100,6 @@ static int spansion_quad_enable(struct spi_nor
>> *nor)
>>  	return 0;
>>  }
>>
>> -static int micron_quad_enable(struct spi_nor *nor)
>> -{
>> -	int ret;
>> -	u8 val;
>> -
>> -	ret = nor->read_reg(nor, SPINOR_OP_RD_EVCR, &val, 1);
>> -	if (ret < 0) {
>> -		dev_err(nor->dev, "error %d reading EVCR\n", ret);
>> -		return ret;
>> -	}
>> -
>> -	write_enable(nor);
>> -
>> -	/* set EVCR, enable quad I/O */
>> -	nor->cmd_buf[0] = val & ~EVCR_QUAD_EN_MICRON;
>> -	ret = nor->write_reg(nor, SPINOR_OP_WD_EVCR, nor->cmd_buf, 1);
>> -	if (ret < 0) {
>> -		dev_err(nor->dev, "error while writing EVCR register\n");
>> -		return ret;
>> -	}
>> -
>> -	ret = spi_nor_wait_till_ready(nor);
>> -	if (ret)
>> -		return ret;
>> -
>> -	/* read EVCR and check it */
>> -	ret = nor->read_reg(nor, SPINOR_OP_RD_EVCR, &val, 1);
>> -	if (ret < 0) {
>> -		dev_err(nor->dev, "error %d reading EVCR\n", ret);
>> -		return ret;
>> -	}
>> -	if (val & EVCR_QUAD_EN_MICRON) {
>> -		dev_err(nor->dev, "Micron EVCR Quad bit not clear\n");
>> -		return -EINVAL;
>> -	}
>> -
>> -	return 0;
>> -}
>> -
>>  static int set_quad_mode(struct spi_nor *nor, const struct flash_info *info)
>>  {
>>  	int status;
>> @@ -1152,12 +1113,7 @@ static int set_quad_mode(struct spi_nor *nor,
>> const struct flash_info *info)
>>  		}
>>  		return status;
>>  	case SNOR_MFR_MICRON:
>> -		status = micron_quad_enable(nor);
>> -		if (status) {
>> -			dev_err(nor->dev, "Micron quad-read not enabled\n");
>> -			return -EINVAL;
>> -		}
>> -		return status;
>> +		return 0;
>>  	default:
>>  		status = spansion_quad_enable(nor);
>>  		if (status) {
>> --
>> 1.8.2.2
>>
>>
>>
>>
> 
> Hi, Cyrlle
> Micron_quad_enable() function is just for how to enable Micron spi nor Quad mode,
> does not force to use SPI nor quad mode. if does not need SPI NOR quad mode,
> why spi-nor.c need set_quad_mode(), according to your patch, this function also need to
> be removed?
> Another question, if there is user want to enable SPI nor quad mode, how to do?
> 

Hi Bean,

the set_quad_mode() function should not been removed from spi-nor.c as it is
needed for both Macronix and Spansion memories.
Indeed, what macronix_quad_enable() and spansion_quad_enable() do is to change
a dedicated bit in some status/configuration register to enable Quad SPI
operations. More precisely, this disables some features such as the Write
Protection (Spansion & Macronix) or the Reset (Macronix) / Hold (Spansion) so
the associated pins can be reassigned to IO2 and IO3 data lines... and that's
all!
What I mean is that once set_quad_mode() has been called, Macronix and Spansion
memories still reply to commands such as Read Status Register in protocol
SPI 1-1-1. Of course, they now also support Fast Read 1-1-4 (0x6b) or 1-4-4
(0xeb).
For Macronix memories, setting the Quad Enable bit in its Status Register is
NOT the same as entering its QPI mode. In QPI mode, ALL commands are expected
to use the SPI 4-4-4 protocol: this mode is not used by spi-nor.c

Then going back to the Micron case, micron_quad_enable() makes the memory
enter its Quad mode where ALL commands must use the SPI 4-4-4 protocol. So
it is definitely not the same thing as what either macronix_quad_enable() or
spansion_quad_enable() do.

Also the current code of micron_quad_enable() is broken:

[...]
Here, the memory still expects the SPI 1-1-1 to be used by write_reg() or
read_reg().

	/* set EVCR, enable quad I/O */
	nor->cmd_buf[0] = val & ~EVCR_QUAD_EN_MICRON;
	ret = nor->write_reg(nor, SPINOR_OP_WD_EVCR, nor->cmd_buf, 1);
	if (ret < 0) {
		dev_err(nor->dev, "error while writing EVCR register\n");
		return ret;
	}

However, once the Quad Enable bit has been cleared into the Enhanced Volatile
Configuration Register, all commands must use the SPI 4-4-4 protocol.
Especially, the later call of spi_nor_wait_till_ready(), which calls read_reg()
to read the Status Register, will fail because the QSPI controller has to mean
to know about the protocol switch hence keeps on sending the Read Status
Register command with the wrong SPI 1-1-1 protocol.
This is the issue I faced when using a Micron memory with the Atmel QSPI
controller embedded in a sama5d2 SoC. Also for the same reason, users of the
m25p80 driver won't be able to use Micron memories for QSPI operations too.

	ret = spi_nor_wait_till_ready(nor);
	if (ret)
		return ret;
[...]

So in its current state, micron_quad_enable() prevents us from using Micron
memories for Quad SPI operations. Some code is missing to make it work.



As I said in the patch description, using the SPI 4-4-4 protocol is not
relevant for MTD usage. It would be worth for XIP operations but this is not the
purpose of the spi-nor driver.
Also, the series of patches still support the Micron Quad SPI mode: the spi-nor
driver no longer enables this mode itself but it still can adapt itself whether
the Quad SPI mode had already been enabled by a previous bootloader or at
reset. This is the purpose of the patch focusing on Read JEDEC ID commands:
it is used to probe the current state of the memory and use the SPI 4-4-4 if
needed.

Best regards,

Cyrille