[PATCH v3] mmc: block: prevent propagating R1_OUT_OF_RANGE for open-ending mode

[PATCH v3] mmc: block: prevent propagating R1_OUT_OF_RANGE for open-ending mode

[Shawn Lin]

We to some extent should tolerate R1_OUT_OF_RANGE for open-ending
mode as it is expected behaviour and most of the backup partition
tables should be located near some of the last blocks which will
always make open-ending read exceed the capacity of cards.

Fixes: 9820a5b11101 ("mmc: core: for data errors, take response of stop cmd into account")
Fixes: a04e6bae9e6f ("mmc: core: check also R1 response for stop commands")
Signed-off-by: Shawn Lin <shawn.lin@rock-chips.com>

Tested-by: Shawn Guo <shawnguo@kernel.org>
Tested-by: Yoshihiro Shimoda <yoshihiro.shimoda.uh@renesas.com>
---

Changes in v3:
- add explanation for why don't check predefined method
- check brq->mrq.sbc for easier to read suggested by Wolfram

Changes in v2:
- fix a typo and introduce STOP_ERROR
- reword the comment and always include a description from the spec if possible

 drivers/mmc/core/block.c | 47 +++++++++++++++++++++++++++++++++++++++++------
 1 file changed, 41 insertions(+), 6 deletions(-)

diff --git a/drivers/mmc/core/block.c b/drivers/mmc/core/block.c
index a11bead..a5a375b 100644
--- a/drivers/mmc/core/block.c
+++ b/drivers/mmc/core/block.c
@@ -1369,12 +1369,44 @@ static inline void mmc_apply_rel_rw(struct mmc_blk_request *brq,
 	 R1_CC_ERROR |		/* Card controller error */		\
 	 R1_ERROR)		/* General/unknown error */
 
-static bool mmc_blk_has_cmd_err(struct mmc_command *cmd)
+static void mmc_blk_eval_resp_error(struct mmc_blk_request *brq)
 {
-	if (!cmd->error && cmd->resp[0] & CMD_ERRORS)
-		cmd->error = -EIO;
+	u32 val;
 
-	return cmd->error;
+	/*
+	 * Per the SD specification(physical layer version 4.10)[1],
+	 * section 4.3.3, it explicitly states that "When the last
+	 * block of user area is read using CMD18, the host should
+	 * ignore OUT_OF_RANGE error that may occur even the sequence
+	 * is correct". And JESD84-B51 for eMMC also has a similar
+	 * statement on section 6.8.3.
+	 *
+	 * Multiple block read/write could be done by either predefined
+	 * method, namely CMD23, or open-ending mode.
+	 *
+	 * For open-ending mode, we should ignore the OUT_OF_RANGE
+	 * error as it's normal behaviour.
+	 *
+	 * However the spec[1] doesn't tell us whether we should also
+	 * ignore that for predefined method. But per the spec[1], section
+	 * 4.15 Set Block Count Command, it says"If illegal block count
+	 * is set, out of range error will be indicated during read/write
+	 * operation (For example, data transfer is stopped at user area
+	 * boundary)." In another word, we could expect a out of range error
+	 * in the response for the following CMD18/25. And if argument of
+	 * CMD23 + the argument of CMD18/25 exceed the max number of blocks,
+	 * we could also expect to get a -ETIMEDOUT or any error number from
+	 * the host drivers due to missing data response(for write)/data(for
+	 * read), as the cards will stop the data transfer by itself per the
+	 * spec. So we only need to check R1_OUT_OF_RANGE for open-ending mode.
+	 */
+
+	if (!brq->stop.error) {
+		/* If there is no error yet, check R1 response */
+		val = brq->stop.resp[0] & CMD_ERRORS;
+		if (val && !(val & R1_OUT_OF_RANGE && !brq->mrq.sbc))
+			brq->stop.error = -EIO;
+	}
 }
 
 static enum mmc_blk_status mmc_blk_err_check(struct mmc_card *card,
@@ -1398,8 +1430,11 @@ static enum mmc_blk_status mmc_blk_err_check(struct mmc_card *card,
 	 * stop.error indicates a problem with the stop command.  Data
 	 * may have been transferred, or may still be transferring.
 	 */
-	if (brq->sbc.error || brq->cmd.error || mmc_blk_has_cmd_err(&brq->stop) ||
-	    brq->data.error) {
+
+	mmc_blk_eval_resp_error(brq);
+
+	if (brq->sbc.error || brq->cmd.error ||
+	    brq->stop.error || brq->data.error) {
 		switch (mmc_blk_cmd_recovery(card, req, brq, &ecc_err, &gen_err)) {
 		case ERR_RETRY:
 			return MMC_BLK_RETRY;
-- 
1.9.1

Re: [PATCH v3] mmc: block: prevent propagating R1_OUT_OF_RANGE for open-ending mode

[Wolfram Sang]

[-- Attachment #1: Type: text/plain, Size: 3262 bytes --]

Hi Shawn,

On Wed, Aug 16, 2017 at 09:11:41AM +0800, Shawn Lin wrote:
> We to some extent should tolerate R1_OUT_OF_RANGE for open-ending
> mode as it is expected behaviour and most of the backup partition
> tables should be located near some of the last blocks which will
> always make open-ending read exceed the capacity of cards.
> 
> Fixes: 9820a5b11101 ("mmc: core: for data errors, take response of stop cmd into account")
> Fixes: a04e6bae9e6f ("mmc: core: check also R1 response for stop commands")
> Signed-off-by: Shawn Lin <shawn.lin@rock-chips.com>
> 
> Tested-by: Shawn Guo <shawnguo@kernel.org>
> Tested-by: Yoshihiro Shimoda <yoshihiro.shimoda.uh@renesas.com>

I'd think these Tested-by should be dropped. We changed the patch quite
a bit meanwhile and I am quite sure Shimoda-san didn't test this new
version of the patch. Dunno about Shawn. Hopefully they have time to
re-test?

> +	/*
> +	 * Per the SD specification(physical layer version 4.10)[1],
> +	 * section 4.3.3, it explicitly states that "When the last
> +	 * block of user area is read using CMD18, the host should
> +	 * ignore OUT_OF_RANGE error that may occur even the sequence
> +	 * is correct". And JESD84-B51 for eMMC also has a similar
> +	 * statement on section 6.8.3.
> +	 *
> +	 * Multiple block read/write could be done by either predefined
> +	 * method, namely CMD23, or open-ending mode.
> +	 *

Very minor nit: I'd remove this empty line and merge the two paragraphs.

> +	 * For open-ending mode, we should ignore the OUT_OF_RANGE
> +	 * error as it's normal behaviour.
> +	 *
> +	 * However the spec[1] doesn't tell us whether we should also
> +	 * ignore that for predefined method. But per the spec[1], section
> +	 * 4.15 Set Block Count Command, it says"If illegal block count
> +	 * is set, out of range error will be indicated during read/write
> +	 * operation (For example, data transfer is stopped at user area
> +	 * boundary)." In another word, we could expect a out of range error
> +	 * in the response for the following CMD18/25. And if argument of
> +	 * CMD23 + the argument of CMD18/25 exceed the max number of blocks,
> +	 * we could also expect to get a -ETIMEDOUT or any error number from
> +	 * the host drivers due to missing data response(for write)/data(for
> +	 * read), as the cards will stop the data transfer by itself per the
> +	 * spec. So we only need to check R1_OUT_OF_RANGE for open-ending mode.
> +	 */

Very good! I like this a lot.

Also minor nit, but likely better readable:

> +
> +	if (!brq->stop.error) {
		bool OOR_with_open_end;

> +		/* If there is no error yet, check R1 response */
> +		val = brq->stop.resp[0] & CMD_ERRORS;

		OOR_with_open_end = val & R1_OUT_OF_RANGE && !brq->mrq.sbc;

		if (val && !OOR_with_open_end)
> +			brq->stop.error = -EIO;

...

> +	if (brq->sbc.error || brq->cmd.error ||
> +	    brq->stop.error || brq->data.error) {

I am not super strict with the 80 char limit and think one line is
better readable, but I leave that to you and/or Ulf.

Other than that minor stuff:

Reviewed-by: Wolfram Sang <wsa+renesas@sang-engineering.com>

Thanks for this collaboration! I liked it :)

Regards,

   Wolfram


[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 833 bytes --]

Re: [PATCH v3] mmc: block: prevent propagating R1_OUT_OF_RANGE for open-ending mode

[Shawn Guo]

On Thu, Aug 17, 2017 at 12:40:47PM +0200, Wolfram Sang wrote:
> Hi Shawn,
> 
> On Wed, Aug 16, 2017 at 09:11:41AM +0800, Shawn Lin wrote:
> > We to some extent should tolerate R1_OUT_OF_RANGE for open-ending
> > mode as it is expected behaviour and most of the backup partition
> > tables should be located near some of the last blocks which will
> > always make open-ending read exceed the capacity of cards.
> > 
> > Fixes: 9820a5b11101 ("mmc: core: for data errors, take response of stop cmd into account")
> > Fixes: a04e6bae9e6f ("mmc: core: check also R1 response for stop commands")
> > Signed-off-by: Shawn Lin <shawn.lin@rock-chips.com>
> > 
> > Tested-by: Shawn Guo <shawnguo@kernel.org>
> > Tested-by: Yoshihiro Shimoda <yoshihiro.shimoda.uh@renesas.com>
> 
> I'd think these Tested-by should be dropped. We changed the patch quite
> a bit meanwhile and I am quite sure Shimoda-san didn't test this new
> version of the patch. Dunno about Shawn. Hopefully they have time to
> re-test?

I just re-tested it on Hikey, so my Tested-by tag can be kept.

Shawn

Re: [PATCH v3] mmc: block: prevent propagating R1_OUT_OF_RANGE for open-ending mode

[Shawn Lin]

Hi Wolfram,

On 2017/8/17 18:40, Wolfram Sang wrote:
> Hi Shawn,
> 
> On Wed, Aug 16, 2017 at 09:11:41AM +0800, Shawn Lin wrote:
>> We to some extent should tolerate R1_OUT_OF_RANGE for open-ending
>> mode as it is expected behaviour and most of the backup partition
>> tables should be located near some of the last blocks which will
>> always make open-ending read exceed the capacity of cards.
>>
>> Fixes: 9820a5b11101 ("mmc: core: for data errors, take response of stop cmd into account")
>> Fixes: a04e6bae9e6f ("mmc: core: check also R1 response for stop commands")
>> Signed-off-by: Shawn Lin <shawn.lin@rock-chips.com>
>>
>> Tested-by: Shawn Guo <shawnguo@kernel.org>
>> Tested-by: Yoshihiro Shimoda <yoshihiro.shimoda.uh@renesas.com>
> 
> I'd think these Tested-by should be dropped. We changed the patch quite
> a bit meanwhile and I am quite sure Shimoda-san didn't test this new
> version of the patch. Dunno about Shawn. Hopefully they have time to
> re-test?
> 
>> +	/*
>> +	 * Per the SD specification(physical layer version 4.10)[1],
>> +	 * section 4.3.3, it explicitly states that "When the last
>> +	 * block of user area is read using CMD18, the host should
>> +	 * ignore OUT_OF_RANGE error that may occur even the sequence
>> +	 * is correct". And JESD84-B51 for eMMC also has a similar
>> +	 * statement on section 6.8.3.
>> +	 *
>> +	 * Multiple block read/write could be done by either predefined
>> +	 * method, namely CMD23, or open-ending mode.
>> +	 *
> 
> Very minor nit: I'd remove this empty line and merge the two paragraphs.

well do in v4

> 
>> +	 * For open-ending mode, we should ignore the OUT_OF_RANGE
>> +	 * error as it's normal behaviour.
>> +	 *
>> +	 * However the spec[1] doesn't tell us whether we should also
>> +	 * ignore that for predefined method. But per the spec[1], section
>> +	 * 4.15 Set Block Count Command, it says"If illegal block count
>> +	 * is set, out of range error will be indicated during read/write
>> +	 * operation (For example, data transfer is stopped at user area
>> +	 * boundary)." In another word, we could expect a out of range error
>> +	 * in the response for the following CMD18/25. And if argument of
>> +	 * CMD23 + the argument of CMD18/25 exceed the max number of blocks,
>> +	 * we could also expect to get a -ETIMEDOUT or any error number from
>> +	 * the host drivers due to missing data response(for write)/data(for
>> +	 * read), as the cards will stop the data transfer by itself per the
>> +	 * spec. So we only need to check R1_OUT_OF_RANGE for open-ending mode.
>> +	 */
> 
> Very good! I like this a lot.
> 
> Also minor nit, but likely better readable:
> 
>> +
>> +	if (!brq->stop.error) {
> 		bool OOR_with_open_end;
> 
>> +		/* If there is no error yet, check R1 response */
>> +		val = brq->stop.resp[0] & CMD_ERRORS;
> 
> 		OOR_with_open_end = val & R1_OUT_OF_RANGE && !brq->mrq.sbc;
> 
> 		if (val && !OOR_with_open_end)
>> +			brq->stop.error = -EIO;
> 
> ...
> 
>> +	if (brq->sbc.error || brq->cmd.error ||
>> +	    brq->stop.error || brq->data.error) {
> 

will do in v4 but slightly rename it to oor_with_open_end as
it doesn't seem general to name a variable with upper letter.

> I am not super strict with the 80 char limit and think one line is
> better readable, but I leave that to you and/or Ulf.

Make sense but I will keep it and leave that to Ulf too. :)

> 
> Other than that minor stuff:
> 
> Reviewed-by: Wolfram Sang <wsa+renesas@sang-engineering.com>
> 
> Thanks for this collaboration! I liked it :)
> 
> Regards,
> 
>     Wolfram
> 

Re: [PATCH v3] mmc: block: prevent propagating R1_OUT_OF_RANGE for open-ending mode

[Shawn Lin]

On 2017/8/17 21:34, Shawn Guo wrote:
> On Thu, Aug 17, 2017 at 12:40:47PM +0200, Wolfram Sang wrote:
>> Hi Shawn,
>>
>> On Wed, Aug 16, 2017 at 09:11:41AM +0800, Shawn Lin wrote:
>>> We to some extent should tolerate R1_OUT_OF_RANGE for open-ending
>>> mode as it is expected behaviour and most of the backup partition
>>> tables should be located near some of the last blocks which will
>>> always make open-ending read exceed the capacity of cards.
>>>
>>> Fixes: 9820a5b11101 ("mmc: core: for data errors, take response of stop cmd into account")
>>> Fixes: a04e6bae9e6f ("mmc: core: check also R1 response for stop commands")
>>> Signed-off-by: Shawn Lin <shawn.lin@rock-chips.com>
>>>
>>> Tested-by: Shawn Guo <shawnguo@kernel.org>
>>> Tested-by: Yoshihiro Shimoda <yoshihiro.shimoda.uh@renesas.com>
>>
>> I'd think these Tested-by should be dropped. We changed the patch quite
>> a bit meanwhile and I am quite sure Shimoda-san didn't test this new
>> version of the patch. Dunno about Shawn. Hopefully they have time to
>> re-test?
> 
> I just re-tested it on Hikey, so my Tested-by tag can be kept.

Great! Thanks for testing it again!

> 
> Shawn
> --
> To unsubscribe from this list: send the line "unsubscribe linux-mmc" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html
> 
> 
>