* [PATCH 2/6] mtd: diskonchip: don't call nand_scan_bbt() directly
@ 2015-04-16 2:11 Peter Pan 潘栋 (peterpandong)
2015-04-22 17:51 ` Brian Norris
0 siblings, 1 reply; 4+ messages in thread
From: Peter Pan 潘栋 (peterpandong) @ 2015-04-16 2:11 UTC (permalink / raw)
To: dwmw2@infradead.org, Brian Norris
Cc: linux-mtd@lists.infradead.org, linux-kernel@vger.kernel.org
The diskonchip driver almost uses the default nand_base hooks as-is,
except that it provides custom on-flash BBT descriptors and avoids using
factory-marked bad blockers.
So let's refactor the BBT initialization code into a private 'late_init'
hook which handles all the private details. Note the usage of
NAND_SKIP_BBTSCAN, which allows us to defer the BBT scan until we've
prepared everything.
Signed-off-by: Brian Norris <computersforpeace@gmail.com>
Signed-off-by: Peter Pan <peterpandong@micron.com>
---
drivers/mtd/nand/diskonchip.c | 27 ++++++++++++++++-----------
1 file changed, 16 insertions(+), 11 deletions(-)
diff --git a/drivers/mtd/nand/diskonchip.c b/drivers/mtd/nand/diskonchip.c
index f68a7bc..e580014 100644
--- a/drivers/mtd/nand/diskonchip.c
+++ b/drivers/mtd/nand/diskonchip.c
@@ -69,6 +69,9 @@ struct doc_priv {
int mh0_page;
int mh1_page;
struct mtd_info *nextdoc;
+
+ /* Handle the last stage of initialization (BBT scan, partitioning) */
+ int (*late_init)(struct mtd_info *mtd);
};
/* This is the syndrome computed by the HW ecc generator upon reading an empty
@@ -1294,10 +1297,10 @@ static int __init nftl_scan_bbt(struct mtd_info *mtd)
this->bbt_md = NULL;
}
- /* It's safe to set bd=NULL below because NAND_BBT_CREATE is not set.
- At least as nand_bbt.c is currently written. */
- if ((ret = nand_scan_bbt(mtd, NULL)))
+ ret = this->scan_bbt(mtd);
+ if (ret)
return ret;
+
mtd_device_register(mtd, NULL, 0);
if (!no_autopart)
mtd_device_register(mtd, parts, numparts);
@@ -1344,10 +1347,10 @@ static int __init inftl_scan_bbt(struct mtd_info *mtd)
this->bbt_md->pattern = "TBB_SYSM";
}
- /* It's safe to set bd=NULL below because NAND_BBT_CREATE is not set.
- At least as nand_bbt.c is currently written. */
- if ((ret = nand_scan_bbt(mtd, NULL)))
+ ret = this->scan_bbt(mtd);
+ if (ret)
return ret;
+
memset((char *)parts, 0, sizeof(parts));
numparts = inftl_partscan(mtd, parts);
/* At least for now, require the INFTL Media Header. We could probably
@@ -1369,7 +1372,7 @@ static inline int __init doc2000_init(struct mtd_info *mtd)
this->read_byte = doc2000_read_byte;
this->write_buf = doc2000_writebuf;
this->read_buf = doc2000_readbuf;
- this->scan_bbt = nftl_scan_bbt;
+ doc->late_init = nftl_scan_bbt;
doc->CDSNControl = CDSN_CTRL_FLASH_IO | CDSN_CTRL_ECC_IO;
doc2000_count_chips(mtd);
@@ -1396,13 +1399,13 @@ static inline int __init doc2001_init(struct mtd_info *mtd)
can have multiple chips. */
doc2000_count_chips(mtd);
mtd->name = "DiskOnChip 2000 (INFTL Model)";
- this->scan_bbt = inftl_scan_bbt;
+ doc->late_init = inftl_scan_bbt;
return (4 * doc->chips_per_floor);
} else {
/* Bog-standard Millennium */
doc->chips_per_floor = 1;
mtd->name = "DiskOnChip Millennium";
- this->scan_bbt = nftl_scan_bbt;
+ doc->late_init = nftl_scan_bbt;
return 1;
}
}
@@ -1415,7 +1418,7 @@ static inline int __init doc2001plus_init(struct mtd_info *mtd)
this->read_byte = doc2001plus_read_byte;
this->write_buf = doc2001plus_writebuf;
this->read_buf = doc2001plus_readbuf;
- this->scan_bbt = inftl_scan_bbt;
+ doc->late_init = inftl_scan_bbt;
this->cmd_ctrl = NULL;
this->select_chip = doc2001plus_select_chip;
this->cmdfunc = doc2001plus_command;
@@ -1591,6 +1594,8 @@ static int __init doc_probe(unsigned long physadr)
nand->ecc.bytes = 6;
nand->ecc.strength = 2;
nand->bbt_options = NAND_BBT_USE_FLASH;
+ /* Skip the automatic BBT scan so we can run it manually */
+ nand->options |= NAND_SKIP_BBTSCAN;
doc->physadr = physadr;
doc->virtadr = virtadr;
@@ -1608,7 +1613,7 @@ static int __init doc_probe(unsigned long physadr)
else
numchips = doc2001_init(mtd);
- if ((ret = nand_scan(mtd, numchips))) {
+ if ((ret = nand_scan(mtd, numchips)) || (ret = doc->late_init(mtd))) {
/* DBB note: i believe nand_release is necessary here, as
buffers may have been allocated in nand_base. Check with
Thomas. FIX ME! */
--
1.9.1
^ permalink raw reply related [flat|nested] 4+ messages in thread
* Re: [PATCH 2/6] mtd: diskonchip: don't call nand_scan_bbt() directly
2015-04-16 2:11 [PATCH 2/6] mtd: diskonchip: don't call nand_scan_bbt() directly Peter Pan 潘栋 (peterpandong)
@ 2015-04-22 17:51 ` Brian Norris
2015-04-23 0:30 ` Peter Pan 潘栋 (peterpandong)
0 siblings, 1 reply; 4+ messages in thread
From: Brian Norris @ 2015-04-22 17:51 UTC (permalink / raw)
To: Peter Pan 潘栋 (peterpandong)
Cc: linux-mtd@lists.infradead.org, dwmw2@infradead.org,
linux-kernel@vger.kernel.org
On Thu, Apr 16, 2015 at 02:11:24AM +0000, Peter Pan 潘栋 (peterpandong) wrote:
> The diskonchip driver almost uses the default nand_base hooks as-is,
> except that it provides custom on-flash BBT descriptors and avoids using
> factory-marked bad blockers.
>
> So let's refactor the BBT initialization code into a private 'late_init'
> hook which handles all the private details. Note the usage of
> NAND_SKIP_BBTSCAN, which allows us to defer the BBT scan until we've
> prepared everything.
>
> Signed-off-by: Brian Norris <computersforpeace@gmail.com>
> Signed-off-by: Peter Pan <peterpandong@micron.com>
Why are you just resending my patch? You could Ack/Review/Test it
instead. (Did you test it?)
http://patchwork.ozlabs.org/patch/444605/
> ---
> drivers/mtd/nand/diskonchip.c | 27 ++++++++++++++++-----------
> 1 file changed, 16 insertions(+), 11 deletions(-)
>
> diff --git a/drivers/mtd/nand/diskonchip.c b/drivers/mtd/nand/diskonchip.c
> index f68a7bc..e580014 100644
> --- a/drivers/mtd/nand/diskonchip.c
> +++ b/drivers/mtd/nand/diskonchip.c
> @@ -69,6 +69,9 @@ struct doc_priv {
> int mh0_page;
> int mh1_page;
> struct mtd_info *nextdoc;
> +
> + /* Handle the last stage of initialization (BBT scan, partitioning) */
> + int (*late_init)(struct mtd_info *mtd);
> };
>
> /* This is the syndrome computed by the HW ecc generator upon reading an empty
> @@ -1294,10 +1297,10 @@ static int __init nftl_scan_bbt(struct mtd_info *mtd)
> this->bbt_md = NULL;
> }
>
> - /* It's safe to set bd=NULL below because NAND_BBT_CREATE is not set.
> - At least as nand_bbt.c is currently written. */
> - if ((ret = nand_scan_bbt(mtd, NULL)))
> + ret = this->scan_bbt(mtd);
> + if (ret)
> return ret;
> +
> mtd_device_register(mtd, NULL, 0);
> if (!no_autopart)
> mtd_device_register(mtd, parts, numparts);
> @@ -1344,10 +1347,10 @@ static int __init inftl_scan_bbt(struct mtd_info *mtd)
> this->bbt_md->pattern = "TBB_SYSM";
> }
>
> - /* It's safe to set bd=NULL below because NAND_BBT_CREATE is not set.
> - At least as nand_bbt.c is currently written. */
> - if ((ret = nand_scan_bbt(mtd, NULL)))
> + ret = this->scan_bbt(mtd);
> + if (ret)
> return ret;
> +
> memset((char *)parts, 0, sizeof(parts));
> numparts = inftl_partscan(mtd, parts);
> /* At least for now, require the INFTL Media Header. We could probably
> @@ -1369,7 +1372,7 @@ static inline int __init doc2000_init(struct mtd_info *mtd)
> this->read_byte = doc2000_read_byte;
> this->write_buf = doc2000_writebuf;
> this->read_buf = doc2000_readbuf;
> - this->scan_bbt = nftl_scan_bbt;
> + doc->late_init = nftl_scan_bbt;
>
> doc->CDSNControl = CDSN_CTRL_FLASH_IO | CDSN_CTRL_ECC_IO;
> doc2000_count_chips(mtd);
> @@ -1396,13 +1399,13 @@ static inline int __init doc2001_init(struct mtd_info *mtd)
> can have multiple chips. */
> doc2000_count_chips(mtd);
> mtd->name = "DiskOnChip 2000 (INFTL Model)";
> - this->scan_bbt = inftl_scan_bbt;
> + doc->late_init = inftl_scan_bbt;
> return (4 * doc->chips_per_floor);
> } else {
> /* Bog-standard Millennium */
> doc->chips_per_floor = 1;
> mtd->name = "DiskOnChip Millennium";
> - this->scan_bbt = nftl_scan_bbt;
> + doc->late_init = nftl_scan_bbt;
> return 1;
> }
> }
> @@ -1415,7 +1418,7 @@ static inline int __init doc2001plus_init(struct mtd_info *mtd)
> this->read_byte = doc2001plus_read_byte;
> this->write_buf = doc2001plus_writebuf;
> this->read_buf = doc2001plus_readbuf;
> - this->scan_bbt = inftl_scan_bbt;
> + doc->late_init = inftl_scan_bbt;
> this->cmd_ctrl = NULL;
> this->select_chip = doc2001plus_select_chip;
> this->cmdfunc = doc2001plus_command;
> @@ -1591,6 +1594,8 @@ static int __init doc_probe(unsigned long physadr)
> nand->ecc.bytes = 6;
> nand->ecc.strength = 2;
> nand->bbt_options = NAND_BBT_USE_FLASH;
> + /* Skip the automatic BBT scan so we can run it manually */
> + nand->options |= NAND_SKIP_BBTSCAN;
>
> doc->physadr = physadr;
> doc->virtadr = virtadr;
> @@ -1608,7 +1613,7 @@ static int __init doc_probe(unsigned long physadr)
> else
> numchips = doc2001_init(mtd);
>
> - if ((ret = nand_scan(mtd, numchips))) {
> + if ((ret = nand_scan(mtd, numchips)) || (ret = doc->late_init(mtd))) {
> /* DBB note: i believe nand_release is necessary here, as
> buffers may have been allocated in nand_base. Check with
> Thomas. FIX ME! */
> --
> 1.9.1
>
^ permalink raw reply [flat|nested] 4+ messages in thread
* RE: [PATCH 2/6] mtd: diskonchip: don't call nand_scan_bbt() directly
2015-04-22 17:51 ` Brian Norris
@ 2015-04-23 0:30 ` Peter Pan 潘栋 (peterpandong)
2015-04-23 7:50 ` Rafał Miłecki
0 siblings, 1 reply; 4+ messages in thread
From: Peter Pan 潘栋 (peterpandong) @ 2015-04-23 0:30 UTC (permalink / raw)
To: Brian Norris
Cc: linux-mtd@lists.infradead.org, dwmw2@infradead.org,
linux-kernel@vger.kernel.org
On Thu, Apr 23, 2015 at 01:51:27PM +0000, Brian Norris wrote:
>
> On Thu, Apr 16, 2015 at 02:11:24AM +0000, Peter Pan 潘栋 (peterpandong)
> wrote:
> > The diskonchip driver almost uses the default nand_base hooks as-is,
> > except that it provides custom on-flash BBT descriptors and avoids
> using
> > factory-marked bad blockers.
> >
> > So let's refactor the BBT initialization code into a private
> 'late_init'
> > hook which handles all the private details. Note the usage of
> > NAND_SKIP_BBTSCAN, which allows us to defer the BBT scan until we've
> > prepared everything.
> >
> > Signed-off-by: Brian Norris <computersforpeace@gmail.com>
> > Signed-off-by: Peter Pan <peterpandong@micron.com>
>
> Why are you just resending my patch? You could Ack/Review/Test it
> instead. (Did you test it?)
My work is in the 6th patch.
I already tested the patch with micron own nand controller. I don't have
platform with docg4 or diskonchip controller. So the BBT and nand_base.c
part is covered while docg4.c and diskonchip.c not.
> http://patchwork.ozlabs.org/patch/444605/
>
> > ---
> > drivers/mtd/nand/diskonchip.c | 27 ++++++++++++++++-----------
> > 1 file changed, 16 insertions(+), 11 deletions(-)
> >
> > diff --git a/drivers/mtd/nand/diskonchip.c
> b/drivers/mtd/nand/diskonchip.c
> > index f68a7bc..e580014 100644
> > --- a/drivers/mtd/nand/diskonchip.c
> > +++ b/drivers/mtd/nand/diskonchip.c
> > @@ -69,6 +69,9 @@ struct doc_priv {
> > int mh0_page;
> > int mh1_page;
> > struct mtd_info *nextdoc;
> > +
> > + /* Handle the last stage of initialization (BBT scan,
> partitioning) */
> > + int (*late_init)(struct mtd_info *mtd);
> > };
> >
> > /* This is the syndrome computed by the HW ecc generator upon
> reading an empty
> > @@ -1294,10 +1297,10 @@ static int __init nftl_scan_bbt(struct
> mtd_info *mtd)
> > this->bbt_md = NULL;
> > }
> >
> > - /* It's safe to set bd=NULL below because NAND_BBT_CREATE is not
> set.
> > - At least as nand_bbt.c is currently written. */
> > - if ((ret = nand_scan_bbt(mtd, NULL)))
> > + ret = this->scan_bbt(mtd);
> > + if (ret)
> > return ret;
> > +
> > mtd_device_register(mtd, NULL, 0);
> > if (!no_autopart)
> > mtd_device_register(mtd, parts, numparts);
> > @@ -1344,10 +1347,10 @@ static int __init inftl_scan_bbt(struct
> mtd_info *mtd)
> > this->bbt_md->pattern = "TBB_SYSM";
> > }
> >
> > - /* It's safe to set bd=NULL below because NAND_BBT_CREATE is not
> set.
> > - At least as nand_bbt.c is currently written. */
> > - if ((ret = nand_scan_bbt(mtd, NULL)))
> > + ret = this->scan_bbt(mtd);
> > + if (ret)
> > return ret;
> > +
> > memset((char *)parts, 0, sizeof(parts));
> > numparts = inftl_partscan(mtd, parts);
> > /* At least for now, require the INFTL Media Header. We could
> probably
> > @@ -1369,7 +1372,7 @@ static inline int __init doc2000_init(struct
> mtd_info *mtd)
> > this->read_byte = doc2000_read_byte;
> > this->write_buf = doc2000_writebuf;
> > this->read_buf = doc2000_readbuf;
> > - this->scan_bbt = nftl_scan_bbt;
> > + doc->late_init = nftl_scan_bbt;
> >
> > doc->CDSNControl = CDSN_CTRL_FLASH_IO | CDSN_CTRL_ECC_IO;
> > doc2000_count_chips(mtd);
> > @@ -1396,13 +1399,13 @@ static inline int __init doc2001_init(struct
> mtd_info *mtd)
> > can have multiple chips. */
> > doc2000_count_chips(mtd);
> > mtd->name = "DiskOnChip 2000 (INFTL Model)";
> > - this->scan_bbt = inftl_scan_bbt;
> > + doc->late_init = inftl_scan_bbt;
> > return (4 * doc->chips_per_floor);
> > } else {
> > /* Bog-standard Millennium */
> > doc->chips_per_floor = 1;
> > mtd->name = "DiskOnChip Millennium";
> > - this->scan_bbt = nftl_scan_bbt;
> > + doc->late_init = nftl_scan_bbt;
> > return 1;
> > }
> > }
> > @@ -1415,7 +1418,7 @@ static inline int __init
> doc2001plus_init(struct mtd_info *mtd)
> > this->read_byte = doc2001plus_read_byte;
> > this->write_buf = doc2001plus_writebuf;
> > this->read_buf = doc2001plus_readbuf;
> > - this->scan_bbt = inftl_scan_bbt;
> > + doc->late_init = inftl_scan_bbt;
> > this->cmd_ctrl = NULL;
> > this->select_chip = doc2001plus_select_chip;
> > this->cmdfunc = doc2001plus_command;
> > @@ -1591,6 +1594,8 @@ static int __init doc_probe(unsigned long
> physadr)
> > nand->ecc.bytes = 6;
> > nand->ecc.strength = 2;
> > nand->bbt_options = NAND_BBT_USE_FLASH;
> > + /* Skip the automatic BBT scan so we can run it manually */
> > + nand->options |= NAND_SKIP_BBTSCAN;
> >
> > doc->physadr = physadr;
> > doc->virtadr = virtadr;
> > @@ -1608,7 +1613,7 @@ static int __init doc_probe(unsigned long
> physadr)
> > else
> > numchips = doc2001_init(mtd);
> >
> > - if ((ret = nand_scan(mtd, numchips))) {
> > + if ((ret = nand_scan(mtd, numchips)) || (ret = doc-
> >late_init(mtd))) {
> > /* DBB note: i believe nand_release is necessary here, as
> > buffers may have been allocated in nand_base. Check
> with
> > Thomas. FIX ME! */
> > --
> > 1.9.1
> >
^ permalink raw reply [flat|nested] 4+ messages in thread
* Re: [PATCH 2/6] mtd: diskonchip: don't call nand_scan_bbt() directly
2015-04-23 0:30 ` Peter Pan 潘栋 (peterpandong)
@ 2015-04-23 7:50 ` Rafał Miłecki
0 siblings, 0 replies; 4+ messages in thread
From: Rafał Miłecki @ 2015-04-23 7:50 UTC (permalink / raw)
To: Peter Pan 潘栋 (peterpandong)
Cc: dwmw2@infradead.org, Brian Norris, linux-mtd@lists.infradead.org,
linux-kernel@vger.kernel.org
On 23 April 2015 at 02:30, Peter Pan 潘栋 (peterpandong)
<peterpandong@micron.com> wrote:
> On Thu, Apr 23, 2015 at 01:51:27PM +0000, Brian Norris wrote:
>>
>> On Thu, Apr 16, 2015 at 02:11:24AM +0000, Peter Pan 潘栋 (peterpandong)
>> wrote:
>> > The diskonchip driver almost uses the default nand_base hooks as-is,
>> > except that it provides custom on-flash BBT descriptors and avoids
>> using
>> > factory-marked bad blockers.
>> >
>> > So let's refactor the BBT initialization code into a private
>> 'late_init'
>> > hook which handles all the private details. Note the usage of
>> > NAND_SKIP_BBTSCAN, which allows us to defer the BBT scan until we've
>> > prepared everything.
>> >
>> > Signed-off-by: Brian Norris <computersforpeace@gmail.com>
>> > Signed-off-by: Peter Pan <peterpandong@micron.com>
>>
>> Why are you just resending my patch? You could Ack/Review/Test it
>> instead. (Did you test it?)
>
> My work is in the 6th patch.
> I already tested the patch with micron own nand controller. I don't have
> platform with docg4 or diskonchip controller. So the BBT and nand_base.c
> part is covered while docg4.c and diskonchip.c not.
You can use patch format's comment place to post sth like
"Depends on 5 XXX patches from Brian"
--
Rafał
^ permalink raw reply [flat|nested] 4+ messages in thread
end of thread, other threads:[~2015-04-23 7:50 UTC | newest]
Thread overview: 4+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2015-04-16 2:11 [PATCH 2/6] mtd: diskonchip: don't call nand_scan_bbt() directly Peter Pan 潘栋 (peterpandong)
2015-04-22 17:51 ` Brian Norris
2015-04-23 0:30 ` Peter Pan 潘栋 (peterpandong)
2015-04-23 7:50 ` Rafał Miłecki
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox