[RFC] NAND: Optimize NAND_ECC_HW_OOB_FIRST read

[RFC] NAND: Optimize NAND_ECC_HW_OOB_FIRST read

[PrasannaKumar Muralidharan]

From: Lars-Peter Clausen <lars@metafoo.de>

Avoid sending unnecessary READ commands to the chip.

Signed-off-by: Lars-Peter Clausen <lars@metafoo.de>
Signed-off-by: PrasannaKumar Muralidharan <prasannatsmkumar@gmail.com>
---
This patch is taken from git://projects.qi-hardware.com/qi-kernel.git
branch jz-3.16. From [1] and [2] it can be seen that the patch author
thinks this can be sent upstream but it never happened so far. This
patch is used in OpenWRT as seen from [3].

I have only compile tested the patch.

1. https://www.mail-archive.com/discussion@lists.en.qi-hardware.com/msg04635.html
2. https://www.mail-archive.com/discussion@lists.en.qi-hardware.com/msg04639.html
3. https://git.lede-project.org/?p=source.git;a=blob;f=target/linux/xburst/patches-3.18/002-NAND-Optimize-NAND_ECC_HW_OOB_FIRST-read.patch;h=046da51912de3cd4444779df5de13d2c1999719a;hb=c03d4317a6bc891cb4a5e89cbdd77f37c23aff86

 drivers/mtd/nand/nand_base.c | 16 ++++++++++++----
 1 file changed, 12 insertions(+), 4 deletions(-)

diff --git a/drivers/mtd/nand/nand_base.c b/drivers/mtd/nand/nand_base.c
index 12edaae..4bf3bdb 100644
--- a/drivers/mtd/nand/nand_base.c
+++ b/drivers/mtd/nand/nand_base.c
@@ -1680,9 +1680,15 @@ static int nand_read_page_hwecc_oob_first(struct mtd_info *mtd,
 	unsigned int max_bitflips = 0;
 
 	/* Read the OOB area first */
-	chip->cmdfunc(mtd, NAND_CMD_READOOB, 0, page);
-	chip->read_buf(mtd, chip->oob_poi, mtd->oobsize);
-	chip->cmdfunc(mtd, NAND_CMD_READ0, 0, page);
+	if (mtd->writesize > 512) {
+		chip->cmdfunc(mtd, NAND_CMD_READ0, mtd->writesize, page);
+		chip->read_buf(mtd, chip->oob_poi, mtd->oobsize);
+		chip->cmdfunc(mtd, NAND_CMD_RNDOUT, 0, -1);
+	} else {
+		chip->cmdfunc(mtd, NAND_CMD_READOOB, 0, page);
+		chip->read_buf(mtd, chip->oob_poi, mtd->oobsize);
+		chip->cmdfunc(mtd, NAND_CMD_READ0, 0, page);
+	}
 
 	ret = mtd_ooblayout_get_eccbytes(mtd, ecc_code, chip->oob_poi, 0,
 					 chip->ecc.total);
@@ -1902,8 +1908,10 @@ static int nand_do_read_ops(struct mtd_info *mtd, loff_t from,
 						 __func__, buf);
 
 read_retry:
-			if (nand_standard_page_accessors(&chip->ecc))
+			if (nand_standard_page_accessors(&chip->ecc) &&
+			    chip->ecc.mode != NAND_ECC_HW_OOB_FIRST) {
 				chip->cmdfunc(mtd, NAND_CMD_READ0, 0x00, page);
+			}
 
 			/*
 			 * Now read the page into the buffer.  Absent an error,
-- 
2.10.0

Re: [RFC] NAND: Optimize NAND_ECC_HW_OOB_FIRST read

[Boris Brezillon]

Hi PrasannaKumar,

On Sat, 28 Oct 2017 13:13:51 +0530
PrasannaKumar Muralidharan <prasannatsmkumar@gmail.com> wrote:

> From: Lars-Peter Clausen <lars@metafoo.de>
> 
> Avoid sending unnecessary READ commands to the chip.
> 
> Signed-off-by: Lars-Peter Clausen <lars@metafoo.de>
> Signed-off-by: PrasannaKumar Muralidharan <prasannatsmkumar@gmail.com>
> ---
> This patch is taken from git://projects.qi-hardware.com/qi-kernel.git
> branch jz-3.16. From [1] and [2] it can be seen that the patch author
> thinks this can be sent upstream but it never happened so far. This
> patch is used in OpenWRT as seen from [3].

Sounds reasonable, but it's likely to conflict with something I'd like
to queue for 4.16 [1]. I'll rebase this patch on nand/next once the
->exec_op() series is merged. Don't hesitate to ping me if I forget.

Regards,

Boris

[1]http://patchwork.ozlabs.org/project/linux-mtd/list/?series=8923

> 
> I have only compile tested the patch.
> 
> 1. https://www.mail-archive.com/discussion@lists.en.qi-hardware.com/msg04635.html
> 2. https://www.mail-archive.com/discussion@lists.en.qi-hardware.com/msg04639.html
> 3. https://git.lede-project.org/?p=source.git;a=blob;f=target/linux/xburst/patches-3.18/002-NAND-Optimize-NAND_ECC_HW_OOB_FIRST-read.patch;h=046da51912de3cd4444779df5de13d2c1999719a;hb=c03d4317a6bc891cb4a5e89cbdd77f37c23aff86
> 
>  drivers/mtd/nand/nand_base.c | 16 ++++++++++++----
>  1 file changed, 12 insertions(+), 4 deletions(-)
> 
> diff --git a/drivers/mtd/nand/nand_base.c b/drivers/mtd/nand/nand_base.c
> index 12edaae..4bf3bdb 100644
> --- a/drivers/mtd/nand/nand_base.c
> +++ b/drivers/mtd/nand/nand_base.c
> @@ -1680,9 +1680,15 @@ static int nand_read_page_hwecc_oob_first(struct mtd_info *mtd,
>  	unsigned int max_bitflips = 0;
>  
>  	/* Read the OOB area first */
> -	chip->cmdfunc(mtd, NAND_CMD_READOOB, 0, page);
> -	chip->read_buf(mtd, chip->oob_poi, mtd->oobsize);
> -	chip->cmdfunc(mtd, NAND_CMD_READ0, 0, page);
> +	if (mtd->writesize > 512) {
> +		chip->cmdfunc(mtd, NAND_CMD_READ0, mtd->writesize, page);
> +		chip->read_buf(mtd, chip->oob_poi, mtd->oobsize);
> +		chip->cmdfunc(mtd, NAND_CMD_RNDOUT, 0, -1);
> +	} else {
> +		chip->cmdfunc(mtd, NAND_CMD_READOOB, 0, page);
> +		chip->read_buf(mtd, chip->oob_poi, mtd->oobsize);
> +		chip->cmdfunc(mtd, NAND_CMD_READ0, 0, page);
> +	}
>  
>  	ret = mtd_ooblayout_get_eccbytes(mtd, ecc_code, chip->oob_poi, 0,
>  					 chip->ecc.total);
> @@ -1902,8 +1908,10 @@ static int nand_do_read_ops(struct mtd_info *mtd, loff_t from,
>  						 __func__, buf);
>  
>  read_retry:
> -			if (nand_standard_page_accessors(&chip->ecc))
> +			if (nand_standard_page_accessors(&chip->ecc) &&
> +			    chip->ecc.mode != NAND_ECC_HW_OOB_FIRST) {
>  				chip->cmdfunc(mtd, NAND_CMD_READ0, 0x00, page);
> +			}
>  
>  			/*
>  			 * Now read the page into the buffer.  Absent an error,

Re: [RFC] NAND: Optimize NAND_ECC_HW_OOB_FIRST read

[PrasannaKumar Muralidharan]

Hi Boris,

On 30 October 2017 at 14:04, Boris Brezillon
<boris.brezillon@free-electrons.com> wrote:
> Hi PrasannaKumar,
>
> On Sat, 28 Oct 2017 13:13:51 +0530
> PrasannaKumar Muralidharan <prasannatsmkumar@gmail.com> wrote:
>
>> From: Lars-Peter Clausen <lars@metafoo.de>
>>
>> Avoid sending unnecessary READ commands to the chip.
>>
>> Signed-off-by: Lars-Peter Clausen <lars@metafoo.de>
>> Signed-off-by: PrasannaKumar Muralidharan <prasannatsmkumar@gmail.com>
>> ---
>> This patch is taken from git://projects.qi-hardware.com/qi-kernel.git
>> branch jz-3.16. From [1] and [2] it can be seen that the patch author
>> thinks this can be sent upstream but it never happened so far. This
>> patch is used in OpenWRT as seen from [3].
>
> Sounds reasonable, but it's likely to conflict with something I'd like
> to queue for 4.16 [1]. I'll rebase this patch on nand/next once the
> ->exec_op() series is merged. Don't hesitate to ping me if I forget.
>
> Regards,
>
> Boris
>
> [1]http://patchwork.ozlabs.org/project/linux-mtd/list/?series=8923

Okay. I will remind you if required.

There is another patch used in OpenWRT at [1]. Do you mind to have a
look at it and take that if it is useful? I don't see a point in me
sending it as a RFC as I am not contributing anything to it other than
noticing its existence.

1. https://git.lede-project.org/?p=source.git;a=blob;f=target/linux/xburst/patches-3.18/003-NAND-Add-support-for-subpage-reads-for-NAND_ECC_HW_O.patch;h=974eb7a5db9a09347fa6137461f030575f9a4328;hb=c03d4317a6bc891cb4a5e89cbdd77f37c23aff86

Thanks,
PrasannaKumar

Re: [RFC] NAND: Optimize NAND_ECC_HW_OOB_FIRST read

[Boris Brezillon]

On Mon, 30 Oct 2017 18:17:50 +0530
PrasannaKumar Muralidharan <prasannatsmkumar@gmail.com> wrote:

> Hi Boris,
> 
> On 30 October 2017 at 14:04, Boris Brezillon
> <boris.brezillon@free-electrons.com> wrote:
> > Hi PrasannaKumar,
> >
> > On Sat, 28 Oct 2017 13:13:51 +0530
> > PrasannaKumar Muralidharan <prasannatsmkumar@gmail.com> wrote:
> >  
> >> From: Lars-Peter Clausen <lars@metafoo.de>
> >>
> >> Avoid sending unnecessary READ commands to the chip.
> >>
> >> Signed-off-by: Lars-Peter Clausen <lars@metafoo.de>
> >> Signed-off-by: PrasannaKumar Muralidharan <prasannatsmkumar@gmail.com>
> >> ---
> >> This patch is taken from git://projects.qi-hardware.com/qi-kernel.git
> >> branch jz-3.16. From [1] and [2] it can be seen that the patch author
> >> thinks this can be sent upstream but it never happened so far. This
> >> patch is used in OpenWRT as seen from [3].  
> >
> > Sounds reasonable, but it's likely to conflict with something I'd like
> > to queue for 4.16 [1]. I'll rebase this patch on nand/next once the  
> > ->exec_op() series is merged. Don't hesitate to ping me if I forget.  
> >
> > Regards,
> >
> > Boris
> >
> > [1]http://patchwork.ozlabs.org/project/linux-mtd/list/?series=8923  
> 
> Okay. I will remind you if required.
> 
> There is another patch used in OpenWRT at [1]. Do you mind to have a
> look at it and take that if it is useful? I don't see a point in me
> sending it as a RFC as I am not contributing anything to it other than
> noticing its existence.

Yep, it makes sense to have that one applied too, but this is pretty
much the same problem: it will conflict with the ->exec_op() rework.

Can you please send it to the ML so I can track it in patchwork and
respin it when ->exec_op() is out. I'll also need someone owning a HW
using NAND_ECC_HW_OOB_FIRST to test the changes. Do you have such HW?

> 
> 1. https://git.lede-project.org/?p=source.git;a=blob;f=target/linux/xburst/patches-3.18/003-NAND-Add-support-for-subpage-reads-for-NAND_ECC_HW_O.patch;h=974eb7a5db9a09347fa6137461f030575f9a4328;hb=c03d4317a6bc891cb4a5e89cbdd77f37c23aff86
> 
> Thanks,
> PrasannaKumar

Re: [RFC] NAND: Optimize NAND_ECC_HW_OOB_FIRST read

[PrasannaKumar Muralidharan]

Hi Boris,

On 30 October 2017 at 18:33, Boris Brezillon
<boris.brezillon@free-electrons.com> wrote:
> On Mon, 30 Oct 2017 18:17:50 +0530
> PrasannaKumar Muralidharan <prasannatsmkumar@gmail.com> wrote:
>
>> Hi Boris,
>>
>> On 30 October 2017 at 14:04, Boris Brezillon
>> <boris.brezillon@free-electrons.com> wrote:
>> > Hi PrasannaKumar,
>> >
>> > On Sat, 28 Oct 2017 13:13:51 +0530
>> > PrasannaKumar Muralidharan <prasannatsmkumar@gmail.com> wrote:
>> >
>> >> From: Lars-Peter Clausen <lars@metafoo.de>
>> >>
>> >> Avoid sending unnecessary READ commands to the chip.
>> >>
>> >> Signed-off-by: Lars-Peter Clausen <lars@metafoo.de>
>> >> Signed-off-by: PrasannaKumar Muralidharan <prasannatsmkumar@gmail.com>
>> >> ---
>> >> This patch is taken from git://projects.qi-hardware.com/qi-kernel.git
>> >> branch jz-3.16. From [1] and [2] it can be seen that the patch author
>> >> thinks this can be sent upstream but it never happened so far. This
>> >> patch is used in OpenWRT as seen from [3].
>> >
>> > Sounds reasonable, but it's likely to conflict with something I'd like
>> > to queue for 4.16 [1]. I'll rebase this patch on nand/next once the
>> > ->exec_op() series is merged. Don't hesitate to ping me if I forget.
>> >
>> > Regards,
>> >
>> > Boris
>> >
>> > [1]http://patchwork.ozlabs.org/project/linux-mtd/list/?series=8923
>>
>> Okay. I will remind you if required.
>>
>> There is another patch used in OpenWRT at [1]. Do you mind to have a
>> look at it and take that if it is useful? I don't see a point in me
>> sending it as a RFC as I am not contributing anything to it other than
>> noticing its existence.
>
> Yep, it makes sense to have that one applied too, but this is pretty
> much the same problem: it will conflict with the ->exec_op() rework.

True. I did realise that.

> Can you please send it to the ML so I can track it in patchwork and

Sure.

> respin it when ->exec_op() is out. I'll also need someone owning a HW
> using NAND_ECC_HW_OOB_FIRST to test the changes. Do you have such HW?

I am looking at OpenWRT and trying to post patches that I feel useful.
I do not have access to JZ4740.

Grepping over driver/mtd/nand I can find that only JZ4740 and DaVinci
supports this. When I last searched I could not find people with
access to Ingenic JZ4740 SoC (which is targeted by arch/mips/xburst in
OpenWRT). I can check with people who have newer SoC from Ingenic if
that has the required hardware.

Please let me know how to test that patch if I could find someone who
can test that patch.

Thanks,
PrasannaKumar

Re: [RFC] NAND: Optimize NAND_ECC_HW_OOB_FIRST read

[PrasannaKumar Muralidharan]

Hi Boris,

On 30 October 2017 at 19:04, PrasannaKumar Muralidharan
<prasannatsmkumar@gmail.com> wrote:
> Hi Boris,
>
> On 30 October 2017 at 18:33, Boris Brezillon
> <boris.brezillon@free-electrons.com> wrote:
>>
>> Yep, it makes sense to have that one applied too, but this is pretty
>> much the same problem: it will conflict with the ->exec_op() rework.
>
> True. I did realise that.
>
>> Can you please send it to the ML so I can track it in patchwork and
>
> Sure.

I could not get that patch compiled yet. I will revisit it next week
and send the patch.

Regards,
PrasannaKumar

Re: [RFC] NAND: Optimize NAND_ECC_HW_OOB_FIRST read

[PrasannaKumar Muralidharan]

Hi Boris,

On 30 October 2017 at 14:04, Boris Brezillon
<boris.brezillon@free-electrons.com> wrote:
> Hi PrasannaKumar,
>
> On Sat, 28 Oct 2017 13:13:51 +0530
> PrasannaKumar Muralidharan <prasannatsmkumar@gmail.com> wrote:
>
>> From: Lars-Peter Clausen <lars@metafoo.de>
>>
>> Avoid sending unnecessary READ commands to the chip.
>>
>> Signed-off-by: Lars-Peter Clausen <lars@metafoo.de>
>> Signed-off-by: PrasannaKumar Muralidharan <prasannatsmkumar@gmail.com>
>> ---
>> This patch is taken from git://projects.qi-hardware.com/qi-kernel.git
>> branch jz-3.16. From [1] and [2] it can be seen that the patch author
>> thinks this can be sent upstream but it never happened so far. This
>> patch is used in OpenWRT as seen from [3].
>
> Sounds reasonable, but it's likely to conflict with something I'd like
> to queue for 4.16 [1]. I'll rebase this patch on nand/next once the
> ->exec_op() series is merged. Don't hesitate to ping me if I forget.
>
> Regards,
>
> Boris
>
> [1]http://patchwork.ozlabs.org/project/linux-mtd/list/?series=8923
>
>>
>> I have only compile tested the patch.
>>
>> 1. https://www.mail-archive.com/discussion@lists.en.qi-hardware.com/msg04635.html
>> 2. https://www.mail-archive.com/discussion@lists.en.qi-hardware.com/msg04639.html
>> 3. https://git.lede-project.org/?p=source.git;a=blob;f=target/linux/xburst/patches-3.18/002-NAND-Optimize-NAND_ECC_HW_OOB_FIRST-read.patch;h=046da51912de3cd4444779df5de13d2c1999719a;hb=c03d4317a6bc891cb4a5e89cbdd77f37c23aff86
>>
>>  drivers/mtd/nand/nand_base.c | 16 ++++++++++++----
>>  1 file changed, 12 insertions(+), 4 deletions(-)
>>
>> diff --git a/drivers/mtd/nand/nand_base.c b/drivers/mtd/nand/nand_base.c
>> index 12edaae..4bf3bdb 100644
>> --- a/drivers/mtd/nand/nand_base.c
>> +++ b/drivers/mtd/nand/nand_base.c
>> @@ -1680,9 +1680,15 @@ static int nand_read_page_hwecc_oob_first(struct mtd_info *mtd,
>>       unsigned int max_bitflips = 0;
>>
>>       /* Read the OOB area first */
>> -     chip->cmdfunc(mtd, NAND_CMD_READOOB, 0, page);
>> -     chip->read_buf(mtd, chip->oob_poi, mtd->oobsize);
>> -     chip->cmdfunc(mtd, NAND_CMD_READ0, 0, page);
>> +     if (mtd->writesize > 512) {
>> +             chip->cmdfunc(mtd, NAND_CMD_READ0, mtd->writesize, page);
>> +             chip->read_buf(mtd, chip->oob_poi, mtd->oobsize);
>> +             chip->cmdfunc(mtd, NAND_CMD_RNDOUT, 0, -1);
>> +     } else {
>> +             chip->cmdfunc(mtd, NAND_CMD_READOOB, 0, page);
>> +             chip->read_buf(mtd, chip->oob_poi, mtd->oobsize);
>> +             chip->cmdfunc(mtd, NAND_CMD_READ0, 0, page);
>> +     }
>>
>>       ret = mtd_ooblayout_get_eccbytes(mtd, ecc_code, chip->oob_poi, 0,
>>                                        chip->ecc.total);
>> @@ -1902,8 +1908,10 @@ static int nand_do_read_ops(struct mtd_info *mtd, loff_t from,
>>                                                __func__, buf);
>>
>>  read_retry:
>> -                     if (nand_standard_page_accessors(&chip->ecc))
>> +                     if (nand_standard_page_accessors(&chip->ecc) &&
>> +                         chip->ecc.mode != NAND_ECC_HW_OOB_FIRST) {
>>                               chip->cmdfunc(mtd, NAND_CMD_READ0, 0x00, page);
>> +                     }
>>
>>                       /*
>>                        * Now read the page into the buffer.  Absent an error,
>

Can you please revisit this?

Thanks,
PrasannaKumar