Skip to contentSkip to conversation

mmc: card: Skip secure option for MoviNAND.

PATCH In review IAN CHEN 3 participants 5 messages Aug 29, 2012 – Aug 30, 2012 MMC, SD and SDIO

lore
Share & tools
lore

IAN CHEN

author
#0 lore raw

PATCH v6 1/1 mmc: card: Skip secure option for MoviNAND.

For several MoviNAND, there are some known issue with secure option. For these specific MoviNAND device, we skip secure option.

You could refer this discussion in XDA developers.
http://forum.xda-developers.com/showthread.php?t=1644364
https://plus.google.com/111398485184813224730/posts/21pTYfTsCkB#111398485184813224730/posts/21pTYfTsCkB

---
 drivers/mmc/card/block.c |   24 +++++++++++++++++++++++-
 include/linux/mmc/card.h |    1 +
 2 files changed, 24 insertions(+), 1 deletions(-)
drivers/mmc/card/block.c +23 -1
index f1c84de..0377885 100644
@@ -1411,7 +1411,8 @@ static int mmc_blk_issue_rq(struct mmc_queue *mq, struct request *req)
 		/* complete ongoing async transfer before issuing discard */
 		if (card->host->areq)
 			mmc_blk_issue_rw_rq(mq, NULL);
-		if (req->cmd_flags & REQ_SECURE)
+		if (req->cmd_flags & REQ_SECURE &&
+			!(card->quirks & MMC_QUIRK_SEC_ERASE_TRIM_BROKEN))
 			ret = mmc_blk_issue_secdiscard_rq(mq, req);
 		else
 			ret = mmc_blk_issue_discard_rq(mq, req);
@@ -1716,6 +1717,7 @@ force_ro_fail:
 #define CID_MANFID_SANDISK	0x2
 #define CID_MANFID_TOSHIBA	0x11
 #define CID_MANFID_MICRON	0x13
+#define CID_MANFID_SAMSUNG	0x15
 
 static const struct mmc_fixup blk_fixups[] =
 {
@@ -1752,6 +1754,26 @@ static const struct mmc_fixup blk_fixups[] =
 	MMC_FIXUP(CID_NAME_ANY, CID_MANFID_MICRON, 0x200, add_quirk_mmc,
 		  MMC_QUIRK_LONG_READ_TIME),
 
+	/*
+	 * Some issue about secure erase/secure trim for Samsung MoviNAND
+	 */
+	MMC_FIXUP("M8G2FA", CID_MANFID_SAMSUNG, CID_OEMID_ANY, add_quirk_mmc,
+		  MMC_QUIRK_SEC_ERASE_TRIM_BROKEN),
+	MMC_FIXUP("MAG4FA", CID_MANFID_SAMSUNG, CID_OEMID_ANY, add_quirk_mmc,
+		  MMC_QUIRK_SEC_ERASE_TRIM_BROKEN),
+	MMC_FIXUP("MBG8FA", CID_MANFID_SAMSUNG, CID_OEMID_ANY, add_quirk_mmc,
+		  MMC_QUIRK_SEC_ERASE_TRIM_BROKEN),
+	MMC_FIXUP("MCGAFA", CID_MANFID_SAMSUNG, CID_OEMID_ANY, add_quirk_mmc,
+		  MMC_QUIRK_SEC_ERASE_TRIM_BROKEN),
+	MMC_FIXUP("VAL00M", CID_MANFID_SAMSUNG, CID_OEMID_ANY, add_quirk_mmc,
+		  MMC_QUIRK_SEC_ERASE_TRIM_BROKEN),
+	MMC_FIXUP("VYL00M", CID_MANFID_SAMSUNG, CID_OEMID_ANY, add_quirk_mmc,
+		  MMC_QUIRK_SEC_ERASE_TRIM_BROKEN),
+	MMC_FIXUP("KYL00M", CID_MANFID_SAMSUNG, CID_OEMID_ANY, add_quirk_mmc,
+		  MMC_QUIRK_SEC_ERASE_TRIM_BROKEN),
+	MMC_FIXUP("VZL00M", CID_MANFID_SAMSUNG, CID_OEMID_ANY, add_quirk_mmc,
+		  MMC_QUIRK_SEC_ERASE_TRIM_BROKEN),
+
 	END_FIXUP
 };
 
include/linux/mmc/card.h +1 -0
index 111aca5..4b27f9f 100644
@@ -239,6 +239,7 @@ struct mmc_card {
 #define MMC_QUIRK_BLK_NO_CMD23	(1<<7)		/* Avoid CMD23 for regular multiblock */
 #define MMC_QUIRK_BROKEN_BYTE_MODE_512 (1<<8)	/* Avoid sending 512 bytes in */
 #define MMC_QUIRK_LONG_READ_TIME (1<<9)		/* Data read time > CSD says */
+#define MMC_QUIRK_SEC_ERASE_TRIM_BROKEN (1<<10)	/* Skip secure for erase/trim */
 						/* byte mode */
 	unsigned int    poweroff_notify_state;	/* eMMC4.5 notify feature */
 #define MMC_NO_POWER_NOTIFICATION	0
Signature · 1 line
1.7.0.4
#1 lore raw

Hi Chris,

Could you merge this patch?
We want to prevent this problem at mainline kernel.

Best Regards,
Jaehoon Chung

Quoted 78 lines · IAN CHEN

For several MoviNAND, there are some known issue with secure option. For these specific MoviNAND device, we skip secure option.

You could refer this discussion in XDA developers.
http://forum.xda-developers.com/showthread.php?t=1644364
https://plus.google.com/111398485184813224730/posts/21pTYfTsCkB#111398485184813224730/posts/21pTYfTsCkB

---
 drivers/mmc/card/block.c |   24 +++++++++++++++++++++++-
 include/linux/mmc/card.h |    1 +
 2 files changed, 24 insertions(+), 1 deletions(-)
drivers/mmc/card/block.c +23 -1
index f1c84de..0377885 100644
@@ -1411,7 +1411,8 @@ static int mmc_blk_issue_rq(struct mmc_queue *mq, struct request *req)
 		/* complete ongoing async transfer before issuing discard */
 		if (card->host->areq)
 			mmc_blk_issue_rw_rq(mq, NULL);
-		if (req->cmd_flags & REQ_SECURE)
+		if (req->cmd_flags & REQ_SECURE &&
+			!(card->quirks & MMC_QUIRK_SEC_ERASE_TRIM_BROKEN))
 			ret = mmc_blk_issue_secdiscard_rq(mq, req);
 		else
 			ret = mmc_blk_issue_discard_rq(mq, req);
@@ -1716,6 +1717,7 @@ force_ro_fail:
 #define CID_MANFID_SANDISK	0x2
 #define CID_MANFID_TOSHIBA	0x11
 #define CID_MANFID_MICRON	0x13
+#define CID_MANFID_SAMSUNG	0x15
 
 static const struct mmc_fixup blk_fixups[] =
 {
@@ -1752,6 +1754,26 @@ static const struct mmc_fixup blk_fixups[] =
 	MMC_FIXUP(CID_NAME_ANY, CID_MANFID_MICRON, 0x200, add_quirk_mmc,
 		  MMC_QUIRK_LONG_READ_TIME),
 
+	/*
+	 * Some issue about secure erase/secure trim for Samsung MoviNAND
+	 */
+	MMC_FIXUP("M8G2FA", CID_MANFID_SAMSUNG, CID_OEMID_ANY, add_quirk_mmc,
+		  MMC_QUIRK_SEC_ERASE_TRIM_BROKEN),
+	MMC_FIXUP("MAG4FA", CID_MANFID_SAMSUNG, CID_OEMID_ANY, add_quirk_mmc,
+		  MMC_QUIRK_SEC_ERASE_TRIM_BROKEN),
+	MMC_FIXUP("MBG8FA", CID_MANFID_SAMSUNG, CID_OEMID_ANY, add_quirk_mmc,
+		  MMC_QUIRK_SEC_ERASE_TRIM_BROKEN),
+	MMC_FIXUP("MCGAFA", CID_MANFID_SAMSUNG, CID_OEMID_ANY, add_quirk_mmc,
+		  MMC_QUIRK_SEC_ERASE_TRIM_BROKEN),
+	MMC_FIXUP("VAL00M", CID_MANFID_SAMSUNG, CID_OEMID_ANY, add_quirk_mmc,
+		  MMC_QUIRK_SEC_ERASE_TRIM_BROKEN),
+	MMC_FIXUP("VYL00M", CID_MANFID_SAMSUNG, CID_OEMID_ANY, add_quirk_mmc,
+		  MMC_QUIRK_SEC_ERASE_TRIM_BROKEN),
+	MMC_FIXUP("KYL00M", CID_MANFID_SAMSUNG, CID_OEMID_ANY, add_quirk_mmc,
+		  MMC_QUIRK_SEC_ERASE_TRIM_BROKEN),
+	MMC_FIXUP("VZL00M", CID_MANFID_SAMSUNG, CID_OEMID_ANY, add_quirk_mmc,
+		  MMC_QUIRK_SEC_ERASE_TRIM_BROKEN),
+
 	END_FIXUP
 };
 
include/linux/mmc/card.h +1 -0
index 111aca5..4b27f9f 100644
@@ -239,6 +239,7 @@ struct mmc_card {
 #define MMC_QUIRK_BLK_NO_CMD23	(1<<7)		/* Avoid CMD23 for regular multiblock */
 #define MMC_QUIRK_BROKEN_BYTE_MODE_512 (1<<8)	/* Avoid sending 512 bytes in */
 #define MMC_QUIRK_LONG_READ_TIME (1<<9)		/* Data read time > CSD says */
+#define MMC_QUIRK_SEC_ERASE_TRIM_BROKEN (1<<10)	/* Skip secure for erase/trim */
 						/* byte mode */
 	unsigned int    poweroff_notify_state;	/* eMMC4.5 notify feature */
 #define MMC_NO_POWER_NOTIFICATION	0
#2 lore raw
In reply to IAN CHEN

Hi,

Quoted 11 lines · IAN CHEN

For several MoviNAND, there are some known issue with secure option. For these specific MoviNAND device, we skip secure option.

You could refer this discussion in XDA developers.
http://forum.xda-developers.com/showthread.php?t=1644364
https://plus.google.com/111398485184813224730/posts/21pTYfTsCkB#111398485184813224730/posts/21pTYfTsCkB

Thanks. We should explain the nature of the bug in the commit message, so that users and the stable@ maintainers know how bad the problem is by reading the patch. (This should go to stable@, right? Do you know which kernel versions are affected?)

I'll add something like this, unless you have a better proposal:

==
For several MoviNAND, there are some known issues with secure option. For these specific MoviNAND device, we skip secure option.

Specifically, there is a bug in the eMMC firmware that causes unrecoverable corruption when the MMC is erased with MMC_CAP_ERASE enabled.

You could refer this discussion in XDA developers.
http://forum.xda-developers.com/showthread.php?t=1644364
https://plus.google.com/111398485184813224730/posts/21pTYfTsCkB#111398485184813224730/posts/21pTYfTsCkB
==

- Chris.

Signature · 2 lines
Chris Ball   <[email protected]>   <http://printf.net/>
One Laptop Per Child
author
#3 lore raw

Hi,

Thanks for your help.
The effected kernel will be all the version with MMC_CAP_ERASE & SECURE option enabled.
As commit content mentioned, the problem is unrecoverable corruption after SECURE ERASE or SECURE TRIM.
For devices using defective eMMC with kernel supporting MMC_CAP_ERASE & SECURE option, this issue might cause damage.
Therefore, we skip secure for these known devices.

Regards,
Ian

-----Original Message-----
From: Chris Ball [mailto:[email protected]]
Sent: Wednesday, August 29, 2012 7:27 PM

Hi,

Quoted 6 lines · IAN CHEN

For several MoviNAND, there are some known issue with secure option. For these specific MoviNAND device, we skip secure option.

You could refer this discussion in XDA developers.
http://forum.xda-developers.com/showthread.php?t=1644364

https://plus.google.com/111398485184813224730/posts/21pTYfTsCkB#111398485184
813224730/posts/21pTYfTsCkB

Quoted 5 lines

Thanks. We should explain the nature of the bug in the commit message, so that users and the stable@ maintainers know how bad the problem is by reading the patch. (This should go to stable@, right? Do you know which kernel versions are affected?)

I'll add something like this, unless you have a better proposal:

==
For several MoviNAND, there are some known issues with secure option. For these specific MoviNAND device, we skip secure option.

Specifically, there is a bug in the eMMC firmware that causes unrecoverable corruption when the MMC is erased with MMC_CAP_ERASE enabled.

You could refer this discussion in XDA developers.
http://forum.xda-developers.com/showthread.php?t=1644364
https://plus.google.com/111398485184813224730/posts/21pTYfTsCkB#111398485184
813224730/posts/21pTYfTsCkB
==

- Chris.

Signature · 2 lines
Chris Ball   <[email protected]>   <http://printf.net/>
One Laptop Per Child
#4 lore raw

Hi,

Quoted 8 lines · IAN CHEN

Thanks for your help.
The effected kernel will be all the version with MMC_CAP_ERASE & SECURE option enabled.
As commit content mentioned, the problem is unrecoverable corruption after SECURE ERASE or SECURE TRIM.
For devices using defective eMMC with kernel supporting MMC_CAP_ERASE & SECURE option, this issue might cause damage.
Therefore, we skip secure for these known devices.

Thanks, pushed to mmc-next for 3.6 with a stable@ tag.

- Chris.

Signature · 2 lines
Chris Ball   <[email protected]>   <http://printf.net/>
One Laptop Per Child