Skip to contentSkip to conversation

sunxi_nand_spl: Be smarter about where to look for backup u-boot.bin

PATCH Superseded Hans de Goede 2 participants 6 messages Sep 20, 2015 – Sep 21, 2015

Subsystems (3)
lore
Share & tools
lore
#0 lore raw

PATCH v2 1/1 sunxi_nand_spl: Be smarter about where to look for backup u-boot.bin

We know when u-boot is written to its own partition, in this case the layout always is:

eb 0 spl
eb 1 spl-backup
eb 2 u-boot
eb 3 u-boot-backup

eb: erase-block

So if we cannot load u-boot from its primary offset we know exactly where to look for it.

---
Changes in v2:
-Add an eraseblock_size helper variable to make the calculation for finding
the backup u-boot easier to understand

---
 drivers/mtd/nand/sunxi_nand_spl.c | 26 ++++++++++++++++++++------
 1 file changed, 20 insertions(+), 6 deletions(-)
drivers/mtd/nand/sunxi_nand_spl.c +20 -6
index 5985534..b0e07aa 100644
@@ -356,18 +356,32 @@ static int nand_read_buffer(uint32_t offs, unsigned int size, void *dest,
 
 int nand_spl_load_image(uint32_t offs, unsigned int size, void *dest)
 {
+#if CONFIG_SYS_NAND_U_BOOT_OFFS == CONFIG_SPL_PAD_TO
+	/*
+	 * u-boot-dtb.bin appended to SPL, use syndrome (like the BROM does)
+	 * and try different erase block sizes to find the backup.
+	 */
 	const uint32_t boot_offsets[] = {
 		0 * 1024 * 1024 + CONFIG_SYS_NAND_U_BOOT_OFFS,
 		1 * 1024 * 1024 + CONFIG_SYS_NAND_U_BOOT_OFFS,
 		2 * 1024 * 1024 + CONFIG_SYS_NAND_U_BOOT_OFFS,
 		4 * 1024 * 1024 + CONFIG_SYS_NAND_U_BOOT_OFFS,
 	};
-	int i, syndrome;
-
-	if (CONFIG_SYS_NAND_U_BOOT_OFFS == CONFIG_SPL_PAD_TO)
-		syndrome = 1; /* u-boot-dtb.bin appended to SPL */
-	else
-		syndrome = 0; /* u-boot-dtb.bin on its own partition */
+	const int syndrome = 1;
+#else
+	/*
+	 * u-boot-dtb.bin on its own partition, do not use syndrome, u-boot
+	 * partition sits after 2 eraseblocks (spl, spl-backup), look for
+	 * backup u-boot 1 erase block further.
+	 */
+	const uint32_t eraseblock_size = CONFIG_SYS_NAND_U_BOOT_OFFS / 2;
+	const uint32_t boot_offsets[] = {
+		CONFIG_SYS_NAND_U_BOOT_OFFS,
+		CONFIG_SYS_NAND_U_BOOT_OFFS + eraseblock_size,
+	};
+	const int syndrome = 0;
+#endif
+	int i;
 
 	if (offs == CONFIG_SYS_NAND_U_BOOT_OFFS) {
 		for (i = 0; i < ARRAY_SIZE(boot_offsets); i++) {
Signature · 1 line
2.4.3
#1 lore raw

PATCH v2 1/1 sunxi_nand_spl: Be smarter about where to look for backup u-boot.bin

Quoted 8 lines · Hans de Goede

We know when u-boot is written to its own partition, in this case the layout always is:

eb 0 spl
eb 1 spl-backup
eb 2 u-boot
eb 3 u-boot-backup

Quoted 4 lines

eb: erase-block

So if we cannot load u-boot from its primary offset we know exactly where to look for it.

Is it worth noting here (or perhaps in a code comment?) that the code currently assumes that the first four ebs are of uniform size?

Quoted 59 lines

---
Changes in v2:
-Add an eraseblock_size helper variable to make the calculation for finding
the backup u-boot easier to understand

---
 drivers/mtd/nand/sunxi_nand_spl.c | 26 ++++++++++++++++++++------
 1 file changed, 20 insertions(+), 6 deletions(-)
drivers/mtd/nand/sunxi_nand_spl.c +10 -7
index 5985534..b0e07aa 100644
@@ -356,18 +356,32 @@ static int nand_read_buffer(uint32_t offs, unsigned
 int size, void *dest,
 
 int nand_spl_load_image(uint32_t offs, unsigned int size, void *dest)
 {
+#if CONFIG_SYS_NAND_U_BOOT_OFFS == CONFIG_SPL_PAD_TO
+	/*
+	 * u-boot-dtb.bin appended to SPL, use syndrome (like the BROM
 does)
+	 * and try different erase block sizes to find the backup.
+	 */
 	const uint32_t boot_offsets[] = {
 		0 * 1024 * 1024 + CONFIG_SYS_NAND_U_BOOT_OFFS,
 		1 * 1024 * 1024 + CONFIG_SYS_NAND_U_BOOT_OFFS,
 		2 * 1024 * 1024 + CONFIG_SYS_NAND_U_BOOT_OFFS,
 		4 * 1024 * 1024 + CONFIG_SYS_NAND_U_BOOT_OFFS,
 	};
-	int i, syndrome;
-
-	if (CONFIG_SYS_NAND_U_BOOT_OFFS == CONFIG_SPL_PAD_TO)
-		syndrome = 1; /* u-boot-dtb.bin appended to SPL */
-	else
-		syndrome = 0; /* u-boot-dtb.bin on its own partition */
+	const int syndrome = 1;
+#else
+	/*
+	 * u-boot-dtb.bin on its own partition, do not use syndrome, u
-boot
+	 * partition sits after 2 eraseblocks (spl, spl-backup), look

for

drivers/mtd/nand/sunxi_nand_spl.c +3 -0
+	 * backup u-boot 1 erase block further.
+	 */
+	const uint32_t eraseblock_size = CONFIG_SYS_NAND_U_BOOT_OFFS /

2;

drivers/mtd/nand/sunxi_nand_spl.c +7 -0
+	const uint32_t boot_offsets[] = {
+		CONFIG_SYS_NAND_U_BOOT_OFFS,
+		CONFIG_SYS_NAND_U_BOOT_OFFS + eraseblock_size,
+	};
+	const int syndrome = 0;
+#endif
+	int i;
 
 	if (offs == CONFIG_SYS_NAND_U_BOOT_OFFS) {
 		for (i = 0; i < ARRAY_SIZE(boot_offsets); i++) {
#2 lore raw

PATCH v2 1/1 sunxi_nand_spl: Be smarter about where to look for backup u-boot.bin

Hi,

Quoted 9 lines · Ian Campbell
Quoted 8 lines · Hans de Goede

We know when u-boot is written to its own partition, in this case the layout always is:

eb 0 spl
eb 1 spl-backup
eb 2 u-boot
eb 3 u-boot-backup

Quoted 7 lines
Quoted 4 lines

eb: erase-block

So if we cannot load u-boot from its primary offset we know exactly where to look for it.

Is it worth noting here (or perhaps in a code comment?) that the code currently assumes that the first four ebs are of uniform size?

The eraseblock size is a property of the nand, given a certain nand chip, all eraseblocks on that chip always have the same size.

Regards,

Hans

Quoted 60 lines
Quoted 60 lines

---
Changes in v2:
-Add an eraseblock_size helper variable to make the calculation for finding
the backup u-boot easier to understand

---
  drivers/mtd/nand/sunxi_nand_spl.c | 26 ++++++++++++++++++++------
  1 file changed, 20 insertions(+), 6 deletions(-)
drivers/mtd/nand/sunxi_nand_spl.c +10 -7
index 5985534..b0e07aa 100644
@@ -356,18 +356,32 @@ static int nand_read_buffer(uint32_t offs, unsigned
 int size, void *dest,
 
  int nand_spl_load_image(uint32_t offs, unsigned int size, void *dest)
  {
+#if CONFIG_SYS_NAND_U_BOOT_OFFS == CONFIG_SPL_PAD_TO
+	/*
+	 * u-boot-dtb.bin appended to SPL, use syndrome (like the BROM
 does)
+	 * and try different erase block sizes to find the backup.
+	 */
  	const uint32_t boot_offsets[] = {
  		0 * 1024 * 1024 + CONFIG_SYS_NAND_U_BOOT_OFFS,
  		1 * 1024 * 1024 + CONFIG_SYS_NAND_U_BOOT_OFFS,
  		2 * 1024 * 1024 + CONFIG_SYS_NAND_U_BOOT_OFFS,
  		4 * 1024 * 1024 + CONFIG_SYS_NAND_U_BOOT_OFFS,
  	};
-	int i, syndrome;
-
-	if (CONFIG_SYS_NAND_U_BOOT_OFFS == CONFIG_SPL_PAD_TO)
-		syndrome = 1; /* u-boot-dtb.bin appended to SPL */
-	else
-		syndrome = 0; /* u-boot-dtb.bin on its own partition */
+	const int syndrome = 1;
+#else
+	/*
+	 * u-boot-dtb.bin on its own partition, do not use syndrome, u
-boot
+	 * partition sits after 2 eraseblocks (spl, spl-backup), look

for

drivers/mtd/nand/sunxi_nand_spl.c +3 -0
+	 * backup u-boot 1 erase block further.
+	 */
+	const uint32_t eraseblock_size = CONFIG_SYS_NAND_U_BOOT_OFFS /

2;

drivers/mtd/nand/sunxi_nand_spl.c +7 -0
+	const uint32_t boot_offsets[] = {
+		CONFIG_SYS_NAND_U_BOOT_OFFS,
+		CONFIG_SYS_NAND_U_BOOT_OFFS + eraseblock_size,
+	};
+	const int syndrome = 0;
+#endif
+	int i;
 
  	if (offs == CONFIG_SYS_NAND_U_BOOT_OFFS) {
  		for (i = 0; i < ARRAY_SIZE(boot_offsets); i++) {
#3 lore raw

PATCH v2 1/1 sunxi_nand_spl: Be smarter about where to look for backup u-boot.bin

Quoted 18 lines · Hans de Goede

Hi,

Quoted 15 lines · Ian Campbell
Quoted 13 lines · Hans de Goede

We know when u-boot is written to its own partition, in this case the layout always is:

eb 0 spl
eb 1 spl-backup
eb 2 u-boot
eb 3 u-boot-backup

eb: erase-block

So if we cannot load u-boot from its primary offset we know exactly where
to look for it.

Quoted 5 lines
Quoted 2 lines

Is it worth noting here (or perhaps in a code comment?) that the code currently assumes that the first four ebs are of uniform size?

The eraseblock size is a property of the nand, given a certain nand chip, all eraseblocks on that chip always have the same size.

So they never have "boot erase zones" at either the start or end, which divide one of the erase zones into more fine-grained sizes?

e.g. it used to be the case with NOR that with, say, a 128KB device you would have zones of 3x32K, 1x16K, 2x8K or something like that rather than simply 4x32K.

They don't do this with NAND then?

Ian.

#4 lore raw

PATCH v2 1/1 sunxi_nand_spl: Be smarter about where to look for backup u-boot.bin

Hi,

Quoted 25 lines · Ian Campbell
Quoted 18 lines · Hans de Goede

Hi,

Quoted 15 lines · Ian Campbell
Quoted 13 lines · Hans de Goede

We know when u-boot is written to its own partition, in this case the layout always is:

eb 0 spl
eb 1 spl-backup
eb 2 u-boot
eb 3 u-boot-backup

eb: erase-block

So if we cannot load u-boot from its primary offset we know exactly where
to look for it.

Quoted 5 lines
Quoted 2 lines

Is it worth noting here (or perhaps in a code comment?) that the code currently assumes that the first four ebs are of uniform size?

The eraseblock size is a property of the nand, given a certain nand chip, all eraseblocks on that chip always have the same size.

Quoted 8 lines

So they never have "boot erase zones" at either the start or end, which divide one of the erase zones into more fine-grained sizes?

e.g. it used to be the case with NOR that with, say, a 128KB device you would have zones of 3x32K, 1x16K, 2x8K or something like that rather than simply 4x32K.

They don't do this with NAND then?

To the best of my knowledge no.

Regards,

Hans

#5 lore raw

PATCH v2 1/1 sunxi_nand_spl: Be smarter about where to look for backup u-boot.bin

Quoted 35 lines · Hans de Goede

Hi,

Quoted 32 lines · Ian Campbell
Quoted 20 lines · Hans de Goede

Hi,

Quoted 17 lines · Ian Campbell
Quoted 15 lines · Hans de Goede

We know when u-boot is written to its own partition, in this case the
layout always is:

eb 0 spl
eb 1 spl-backup
eb 2 u-boot
eb 3 u-boot-backup

eb: erase-block

So if we cannot load u-boot from its primary offset we know exactly
where
to look for it.

Quoted 7 lines
Quoted 3 lines

Is it worth noting here (or perhaps in a code comment?) that the code
currently assumes that the first four ebs are of uniform size?

The eraseblock size is a property of the nand, given a certain nand chip,
all eraseblocks on that chip always have the same size.

So they never have "boot erase zones" at either the start or end, which divide one of the erase zones into more fine-grained sizes?

Quoted 8 lines
Quoted 6 lines

e.g. it used to be the case with NOR that with, say, a 128KB device you would have zones of 3x32K, 1x16K, 2x8K or something like that rather than
simply 4x32K.

They don't do this with NAND then?

To the best of my knowledge no.

Thanks.

In which case: