Skip to contentSkip to conversation

mtd: bcm47xxpart: find boot partition by CFE magic

PATCH In review Rafał Miłecki 2 participants 4 messages Dec 21, 2013 – Jan 5, 2014 Memory technology devices (MTD)

lore
Share & tools
lore
#0 lore raw

PATCH 1/2 mtd: bcm47xxpart: find boot partition by CFE magic

Some devices have even nicer-to-recognize CFE thanks to the magic.

---
This patch adds one line wider than 80 chars, however this improves code readability thanks to keeping one rule per line. I hope this can be accepted in this case.
---
Both patches were verified with
./verify ../linux/l2-mtd/ mips-bcm47xx_defconfig < patch

---
 drivers/mtd/bcm47xxpart.c |    8 +++++---
 1 file changed, 5 insertions(+), 3 deletions(-)
drivers/mtd/bcm47xxpart.c +5 -3
index 931746d..b72a3ac 100644
@@ -23,10 +23,11 @@
  * Amount of bytes we read when analyzing each block of flash memory.
  * Set it big enough to allow detecting partition and reading important data.
  */
-#define BCM47XXPART_BYTES_TO_READ	0x404
+#define BCM47XXPART_BYTES_TO_READ	0x4e8
 
 /* Magics */
 #define BOARD_DATA_MAGIC		0x5246504D	/* MPFR */
+#define CFE_MAGIC			0x43464531	/* 1EFC */
 #define FACTORY_MAGIC			0x59544346	/* FCTY */
 #define POT_MAGIC1			0x54544f50	/* POTT */
 #define POT_MAGIC2			0x504f		/* OP */
@@ -102,8 +103,9 @@ static int bcm47xxpart_parse(struct mtd_info *master,
 			continue;
 		}
 
-		/* CFE has small NVRAM at 0x400 */
-		if (buf[0x400 / 4] == NVRAM_HEADER) {
+		/* Magic or small NVRAM at 0x400 */
+		if ((buf[0x4e0 / 4] == CFE_MAGIC && buf[0x4e4 / 4] == CFE_MAGIC) ||
+		    (buf[0x400 / 4] == NVRAM_HEADER)) {
 			bcm47xxpart_add_part(&parts[curr_part++], "boot",
 					     offset, MTD_WRITEABLE);
 			continue;
Signature · 1 line
1.7.10.4
#1 lore raw

PATCH 2/2 mtd: bcm47xxpart: alternative MAGIC for board_data partition

Some devices (like WNDR3700v3) have board_data without MPFR magic, some extra header or extra NVRAM around 0x100. In such case we have to look for another magic which is BD 0B 0D BD (BD probably stands for Board Data). It's located "far far away", so instead extending buffer add another mtd_read.

---
 drivers/mtd/bcm47xxpart.c |   16 ++++++++++++++++
 1 file changed, 16 insertions(+)
drivers/mtd/bcm47xxpart.c +16 -0
index b72a3ac..91f32f2 100644
@@ -27,6 +27,7 @@
 
 /* Magics */
 #define BOARD_DATA_MAGIC		0x5246504D	/* MPFR */
+#define BOARD_DATA_MAGIC2		0xBD0D0BBD
 #define CFE_MAGIC			0x43464531	/* 1EFC */
 #define FACTORY_MAGIC			0x59544346	/* FCTY */
 #define POT_MAGIC1			0x54544f50	/* POTT */
@@ -192,6 +193,21 @@ static int bcm47xxpart_parse(struct mtd_info *master,
 					     offset, 0);
 			continue;
 		}
+
+		/* Read middle of the block */
+		if (mtd_read(master, offset + 0x8000, 0x4,
+			     &bytes_read, (uint8_t *)buf) < 0) {
+			pr_err("mtd_read error while parsing (offset: 0x%X)!\n",
+			       offset);
+			continue;
+		}
+
+		/* Some devices (ex. WNDR3700v3) don't have a standard 'MPFR' */
+		if (buf[0x000 / 4] == BOARD_DATA_MAGIC2) {
+			bcm47xxpart_add_part(&parts[curr_part++], "board_data",
+					     offset, MTD_WRITEABLE);
+			continue;
+		}
 	}
 
 	/* Look for NVRAM at the end of the last block. */
Signature · 1 line
1.7.10.4
#2 lore raw
In reply to Rafał Miłecki

Hi Rafał,

Quoted 7 lines · Rafał Miłecki

Some devices have even nicer-to-recognize CFE thanks to the magic.

---
This patch adds one line wider than 80 chars, however this improves code readability thanks to keeping one rule per line. I hope this can be accepted in this case.

I tripped on the same checkpatch error, and I thought the same thing about readability. Probably worth an exception in this case. Forcing the '||' or half the rule onto the next line wouldn't be too pretty.

Quoted 2 lines

Both patches were verified with
./verify ../linux/l2-mtd/ mips-bcm47xx_defconfig < patch

Nice to know I'm not the only one running these tools :)

Both patches look good. Pushed to l2-mtd.git.

Brian

#3 lore raw

2014/1/5 Brian Norris <[email protected]>:

Quoted 4 lines
Quoted 2 lines

Both patches were verified with
./verify ../linux/l2-mtd/ mips-bcm47xx_defconfig < patch

Nice to know I'm not the only one running these tools :)

Artem taught me to! ;)

Signature · 1 line
Rafał