* [PATCH v2 0/1] fix OOB size decoding on SK Hynix H27UCG8T2BTR
@ 2018-06-24 20:53 Martin Blumenstingl
2018-06-24 20:53 ` [PATCH v2 1/1] mtd: rawnand: hynix: fix decoding the OOB size on H27UCG8T2BTR Martin Blumenstingl
2018-06-25 12:38 ` [PATCH v2 0/1] fix OOB size decoding on SK Hynix H27UCG8T2BTR Miquel Raynal
0 siblings, 2 replies; 5+ messages in thread
From: Martin Blumenstingl @ 2018-06-24 20:53 UTC (permalink / raw)
To: linux-mtd, boris.brezillon, richard
Cc: dwmw2, computersforpeace, marek.vasut, miquel.raynal, liang.yang,
yixun.lan, linux-amlogic, Martin Blumenstingl
The OOB size of the SK Hynix H27UCG8T2BTR is currently detected to be
640 bytes, while the datasheet states that it should be 1280 bytes.
I only have one single device with a SK Hynix NAND flash, so I cannot
say whether this fix may or may not break other chips. other datasheets
I have checked:
- H27UBG8T2B uses a page size of 8KiB -> not applicable
- H27UBG8T2A also uses a page size of 8KiB -> not applicable either
so I would appreciate if someone else could confirm (even if it's
"just" by looking at a datasheet) that this patch should work for other
SK Hynix NAND chips with 16KiB page size.
I decided not to add a "Fixes" tag because it seems H27UCG8T2BTR was
not supported properly at any point in the past (upstream at least).
changes since v1 at [0]:
- don't apply the fix for all Hynix chips with 16KiB page size as
suggested by Miquel Raynal
- added an inline comment what the new logic is supposed to do
- I asked Hynix for feedback on my initial patch, unfortunately it
seems that they cannot disclose any information from the datasheets
publicly
[0] https://patchwork.ozlabs.org/cover/916836/
Martin Blumenstingl (1):
mtd: rawnand: hynix: fix decoding the OOB size on H27UCG8T2BTR
drivers/mtd/nand/raw/nand_hynix.c | 13 +++++++++++++
1 file changed, 13 insertions(+)
--
2.18.0
^ permalink raw reply [flat|nested] 5+ messages in thread
* [PATCH v2 1/1] mtd: rawnand: hynix: fix decoding the OOB size on H27UCG8T2BTR
2018-06-24 20:53 [PATCH v2 0/1] fix OOB size decoding on SK Hynix H27UCG8T2BTR Martin Blumenstingl
@ 2018-06-24 20:53 ` Martin Blumenstingl
2018-06-24 21:03 ` Boris Brezillon
2018-06-25 12:38 ` [PATCH v2 0/1] fix OOB size decoding on SK Hynix H27UCG8T2BTR Miquel Raynal
1 sibling, 1 reply; 5+ messages in thread
From: Martin Blumenstingl @ 2018-06-24 20:53 UTC (permalink / raw)
To: linux-mtd, boris.brezillon, richard
Cc: dwmw2, computersforpeace, marek.vasut, miquel.raynal, liang.yang,
yixun.lan, linux-amlogic, Martin Blumenstingl
The datasheet of the H27UCG8T2BTR states that this chip has a page size
of "16,384 + 1,280(Spare) bytes". The description of the "4th Byte of
Device Identifier Description" indicates that bits 6, 3 and 2 are
encoding the "Redundant Area Size / 8KB", where 640 bytes is a value of
0x6 (110 in binary notation).
hynix_nand_extract_oobsize decodes an OOB size of 640 bytes for this
chip. Kernel boot log extract before this patch:
nand: Could not find valid ONFI parameter page; aborting
nand: device found, Manufacturer ID: 0xad, Chip ID: 0xde
nand: Hynix NAND 8GiB 3,3V 8-bit
nand: 8192 MiB, MLC, erase size: 4096 KiB, page size: 16384,
OOB size: 640
However, based on the description in the datasheet we need to multiply
the OOB size with 2, because it's "640 spare bytes per 8192 bytes page
size" and this NAND chip has a page size of 16384 (= 2 * 8192). After
this patch the kernel boot log reports:
nand: Could not find valid ONFI parameter page; aborting
nand: device found, Manufacturer ID: 0xad, Chip ID: 0xde
nand: Hynix NAND 8GiB 3,3V 8-bit
nand: 8192 MiB, MLC, erase size: 4096 KiB, page size: 16384,
OOB size: 1280
Signed-off-by: Martin Blumenstingl <martin.blumenstingl@googlemail.com>
---
drivers/mtd/nand/raw/nand_hynix.c | 13 +++++++++++++
1 file changed, 13 insertions(+)
diff --git a/drivers/mtd/nand/raw/nand_hynix.c b/drivers/mtd/nand/raw/nand_hynix.c
index d542908a0ebb..8cbe77f447c7 100644
--- a/drivers/mtd/nand/raw/nand_hynix.c
+++ b/drivers/mtd/nand/raw/nand_hynix.c
@@ -473,6 +473,19 @@ static void hynix_nand_extract_oobsize(struct nand_chip *chip,
WARN(1, "Invalid OOB size");
break;
}
+
+ /*
+ * The datasheet of H27UCG8T2BTR mentions that the "Redundant
+ * Area Size" is encoded "per 8KB" (page size). This chip uses
+ * a page size of 16KiB. The datasheet mentions an OOB size of
+ * 1.280 bytes, but the OOB size encoded in the ID bytes (using
+ * the existing logic above) is 640 bytes.
+ * Update the OOB size for this chip by taking the value
+ * determined above and scaling it to the actual page size (so
+ * the actual OOB size for this chip is: 640 * 16k / 8k).
+ */
+ if (chip->id.data[1] == 0xde)
+ mtd->oobsize *= mtd->writesize / SZ_8K;
}
}
--
2.18.0
^ permalink raw reply related [flat|nested] 5+ messages in thread
* Re: [PATCH v2 1/1] mtd: rawnand: hynix: fix decoding the OOB size on H27UCG8T2BTR
2018-06-24 20:53 ` [PATCH v2 1/1] mtd: rawnand: hynix: fix decoding the OOB size on H27UCG8T2BTR Martin Blumenstingl
@ 2018-06-24 21:03 ` Boris Brezillon
2018-06-24 21:16 ` Martin Blumenstingl
0 siblings, 1 reply; 5+ messages in thread
From: Boris Brezillon @ 2018-06-24 21:03 UTC (permalink / raw)
To: Martin Blumenstingl
Cc: linux-mtd, richard, yixun.lan, marek.vasut, liang.yang,
miquel.raynal, linux-amlogic, computersforpeace, dwmw2
On Sun, 24 Jun 2018 22:53:55 +0200
Martin Blumenstingl <martin.blumenstingl@googlemail.com> wrote:
> The datasheet of the H27UCG8T2BTR states that this chip has a page size
> of "16,384 + 1,280(Spare) bytes". The description of the "4th Byte of
> Device Identifier Description" indicates that bits 6, 3 and 2 are
> encoding the "Redundant Area Size / 8KB", where 640 bytes is a value of
> 0x6 (110 in binary notation).
>
> hynix_nand_extract_oobsize decodes an OOB size of 640 bytes for this
> chip. Kernel boot log extract before this patch:
> nand: Could not find valid ONFI parameter page; aborting
> nand: device found, Manufacturer ID: 0xad, Chip ID: 0xde
> nand: Hynix NAND 8GiB 3,3V 8-bit
> nand: 8192 MiB, MLC, erase size: 4096 KiB, page size: 16384,
> OOB size: 640
>
> However, based on the description in the datasheet we need to multiply
> the OOB size with 2, because it's "640 spare bytes per 8192 bytes page
> size" and this NAND chip has a page size of 16384 (= 2 * 8192). After
> this patch the kernel boot log reports:
> nand: Could not find valid ONFI parameter page; aborting
> nand: device found, Manufacturer ID: 0xad, Chip ID: 0xde
> nand: Hynix NAND 8GiB 3,3V 8-bit
> nand: 8192 MiB, MLC, erase size: 4096 KiB, page size: 16384,
> OOB size: 1280
>
> Signed-off-by: Martin Blumenstingl <martin.blumenstingl@googlemail.com>
LGTM.
Reviewed-by: Boris Brezillon <boris.brezillon@bootlin.com>
Just out of curiosity, what do you plan to put on your MLC chip? I
guess you already know that UBI/UBIFS are not supporting MLC chips. Do
you use another FS or an FTL + a block FS?
> ---
> drivers/mtd/nand/raw/nand_hynix.c | 13 +++++++++++++
> 1 file changed, 13 insertions(+)
>
> diff --git a/drivers/mtd/nand/raw/nand_hynix.c b/drivers/mtd/nand/raw/nand_hynix.c
> index d542908a0ebb..8cbe77f447c7 100644
> --- a/drivers/mtd/nand/raw/nand_hynix.c
> +++ b/drivers/mtd/nand/raw/nand_hynix.c
> @@ -473,6 +473,19 @@ static void hynix_nand_extract_oobsize(struct nand_chip *chip,
> WARN(1, "Invalid OOB size");
> break;
> }
> +
> + /*
> + * The datasheet of H27UCG8T2BTR mentions that the "Redundant
> + * Area Size" is encoded "per 8KB" (page size). This chip uses
> + * a page size of 16KiB. The datasheet mentions an OOB size of
> + * 1.280 bytes, but the OOB size encoded in the ID bytes (using
> + * the existing logic above) is 640 bytes.
> + * Update the OOB size for this chip by taking the value
> + * determined above and scaling it to the actual page size (so
> + * the actual OOB size for this chip is: 640 * 16k / 8k).
> + */
> + if (chip->id.data[1] == 0xde)
> + mtd->oobsize *= mtd->writesize / SZ_8K;
> }
> }
>
^ permalink raw reply [flat|nested] 5+ messages in thread
* Re: [PATCH v2 1/1] mtd: rawnand: hynix: fix decoding the OOB size on H27UCG8T2BTR
2018-06-24 21:03 ` Boris Brezillon
@ 2018-06-24 21:16 ` Martin Blumenstingl
0 siblings, 0 replies; 5+ messages in thread
From: Martin Blumenstingl @ 2018-06-24 21:16 UTC (permalink / raw)
To: boris.brezillon
Cc: linux-mtd, richard, yixun.lan, marek.vasut, liang.yang,
miquel.raynal, linux-amlogic, computersforpeace, dwmw2
Hi Boris,
On Sun, Jun 24, 2018 at 11:03 PM Boris Brezillon
<boris.brezillon@bootlin.com> wrote:
>
> On Sun, 24 Jun 2018 22:53:55 +0200
> Martin Blumenstingl <martin.blumenstingl@googlemail.com> wrote:
>
> > The datasheet of the H27UCG8T2BTR states that this chip has a page size
> > of "16,384 + 1,280(Spare) bytes". The description of the "4th Byte of
> > Device Identifier Description" indicates that bits 6, 3 and 2 are
> > encoding the "Redundant Area Size / 8KB", where 640 bytes is a value of
> > 0x6 (110 in binary notation).
> >
> > hynix_nand_extract_oobsize decodes an OOB size of 640 bytes for this
> > chip. Kernel boot log extract before this patch:
> > nand: Could not find valid ONFI parameter page; aborting
> > nand: device found, Manufacturer ID: 0xad, Chip ID: 0xde
> > nand: Hynix NAND 8GiB 3,3V 8-bit
> > nand: 8192 MiB, MLC, erase size: 4096 KiB, page size: 16384,
> > OOB size: 640
> >
> > However, based on the description in the datasheet we need to multiply
> > the OOB size with 2, because it's "640 spare bytes per 8192 bytes page
> > size" and this NAND chip has a page size of 16384 (= 2 * 8192). After
> > this patch the kernel boot log reports:
> > nand: Could not find valid ONFI parameter page; aborting
> > nand: device found, Manufacturer ID: 0xad, Chip ID: 0xde
> > nand: Hynix NAND 8GiB 3,3V 8-bit
> > nand: 8192 MiB, MLC, erase size: 4096 KiB, page size: 16384,
> > OOB size: 1280
> >
> > Signed-off-by: Martin Blumenstingl <martin.blumenstingl@googlemail.com>
>
> LGTM.
>
> Reviewed-by: Boris Brezillon <boris.brezillon@bootlin.com>
thank you for reviewing this!
> Just out of curiosity, what do you plan to put on your MLC chip? I
> guess you already know that UBI/UBIFS are not supporting MLC chips. Do
> you use another FS or an FTL + a block FS?
my first goal is to get something (preferably a full backup) *from*
the chip, rather than putting something onto it
this MLC chip is part of a TV box I have which uses an Amlogic S812
(Meson8m2) SoC
you reviewed the first version of Amlogic's NAND driver today
I may add support for older SoCs to this driver later on (should be
fairly simple)
apart from that I have no specific plans yet
Regards
Martin
^ permalink raw reply [flat|nested] 5+ messages in thread
* Re: [PATCH v2 0/1] fix OOB size decoding on SK Hynix H27UCG8T2BTR
2018-06-24 20:53 [PATCH v2 0/1] fix OOB size decoding on SK Hynix H27UCG8T2BTR Martin Blumenstingl
2018-06-24 20:53 ` [PATCH v2 1/1] mtd: rawnand: hynix: fix decoding the OOB size on H27UCG8T2BTR Martin Blumenstingl
@ 2018-06-25 12:38 ` Miquel Raynal
1 sibling, 0 replies; 5+ messages in thread
From: Miquel Raynal @ 2018-06-25 12:38 UTC (permalink / raw)
To: Martin Blumenstingl
Cc: linux-mtd, boris.brezillon, richard, dwmw2, computersforpeace,
marek.vasut, liang.yang, yixun.lan, linux-amlogic
Hi Martin,
On Sun, 24 Jun 2018 22:53:54 +0200, Martin Blumenstingl
<martin.blumenstingl@googlemail.com> wrote:
> The OOB size of the SK Hynix H27UCG8T2BTR is currently detected to be
> 640 bytes, while the datasheet states that it should be 1280 bytes.
>
> I only have one single device with a SK Hynix NAND flash, so I cannot
> say whether this fix may or may not break other chips. other datasheets
> I have checked:
> - H27UBG8T2B uses a page size of 8KiB -> not applicable
> - H27UBG8T2A also uses a page size of 8KiB -> not applicable either
>
> so I would appreciate if someone else could confirm (even if it's
> "just" by looking at a datasheet) that this patch should work for other
> SK Hynix NAND chips with 16KiB page size.
>
> I decided not to add a "Fixes" tag because it seems H27UCG8T2BTR was
> not supported properly at any point in the past (upstream at least).
>
>
> changes since v1 at [0]:
> - don't apply the fix for all Hynix chips with 16KiB page size as
> suggested by Miquel Raynal
> - added an inline comment what the new logic is supposed to do
> - I asked Hynix for feedback on my initial patch, unfortunately it
> seems that they cannot disclose any information from the datasheets
> publicly
>
>
> [0] https://patchwork.ozlabs.org/cover/916836/
>
>
> Martin Blumenstingl (1):
> mtd: rawnand: hynix: fix decoding the OOB size on H27UCG8T2BTR
>
> drivers/mtd/nand/raw/nand_hynix.c | 13 +++++++++++++
> 1 file changed, 13 insertions(+)
>
Applied to nand/next.
Thanks,
Miquèl
^ permalink raw reply [flat|nested] 5+ messages in thread
end of thread, other threads:[~2018-06-25 12:38 UTC | newest]
Thread overview: 5+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2018-06-24 20:53 [PATCH v2 0/1] fix OOB size decoding on SK Hynix H27UCG8T2BTR Martin Blumenstingl
2018-06-24 20:53 ` [PATCH v2 1/1] mtd: rawnand: hynix: fix decoding the OOB size on H27UCG8T2BTR Martin Blumenstingl
2018-06-24 21:03 ` Boris Brezillon
2018-06-24 21:16 ` Martin Blumenstingl
2018-06-25 12:38 ` [PATCH v2 0/1] fix OOB size decoding on SK Hynix H27UCG8T2BTR Miquel Raynal
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).