FAILED: patch "[PATCH] mtd: Avoid boot crash in RedBoot partition table parser" failed to apply to 5.10-stable tree

PATCH New [email protected] 2 participants 3 messages Mar 20

Subsystems (2)

Stable branch Memory technology devices (MTD)

lore

Share & tools

lore

[email protected]

author maintainer

2026-03-20 17:33 UTC #0 lore raw

The patch below does not apply to the 5.10-stable tree.
If someone wants it applied there, or to any other stable or longterm tree, then please email the backport, including the original git commit id to <[email protected]>.

To reproduce the conflict and resubmit, you may use the following commands:

git fetch https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/ linux-5.10.y
git checkout FETCH_HEAD
git cherry-pick -x 8e2f8020270af7777d49c2e7132260983e4fc566
# <resolve conflicts, build, test, etc.>
git commit -s
git send-email --to '<[email protected]>' --in-reply-to '2026032025-scanning-frying-8ead@gregkh' --subject-prefix 'PATCH 5.10.y' HEAD^..

Possible dependencies:

thanks,

greg k-h

------------------ original commit in Linus's tree ------------------

From 8e2f8020270af7777d49c2e7132260983e4fc566 Mon Sep 17 00:00:00 2001
From: Finn Thain <[email protected]>
Date: Mon, 16 Feb 2026 18:01:30 +1100
Subject: [PATCH] mtd: Avoid boot crash in RedBoot partition table parser

Given CONFIG_FORTIFY_SOURCE=y and a recent compiler,
commit 439a1bcac648 ("fortify: Use __builtin_dynamic_object_size() when available") produces the warning below and an oops.

    Searching for RedBoot partition table in 50000000.flash at offset 0x7e0000
    ------------[ cut here ]------------
    WARNING: lib/string_helpers.c:1035 at 0xc029e04c, CPU#0: swapper/0/1
    memcmp: detected buffer overflow: 15 byte read of buffer size 14
    Modules linked in:
    CPU: 0 UID: 0 PID: 1 Comm: swapper/0 Not tainted 6.19.0 #1 NONE

As Kees said, "'names' is pointing to the final 'namelen' many bytes of the allocation ... 'namelen' could be basically any length at all. This fortify warning looks legit to me -- this code used to be reading beyond the end of the allocation."

Since the size of the dynamic allocation is calculated with strlen() we can use strcmp() instead of memcmp() and remain within bounds.

Cc Kees Cook <[email protected]>

Cc [email protected]

Link https://lore.kernel.org/all/202602151911.AD092DFFCD@keescook/ open here

Fixes 1da177e4c3f4 ("Linux-2.6.12-rc2")

Suggested-by Kees Cook <[email protected]>

Signed-off-by Finn Thain <[email protected]>

Signed-off-by Miquel Raynal <[email protected]>

drivers/mtd/parsers/redboot.c +3 -3

index 558905160ddb..bf162c44eafe 100644
@@ -270,9 +270,9 @@ nogood:
 
 		strcpy(names, fl->img->name);
 #ifdef CONFIG_MTD_REDBOOT_PARTS_READONLY
-		if (!memcmp(names, "RedBoot", 8) ||
-		    !memcmp(names, "RedBoot config", 15) ||
-		    !memcmp(names, "FIS directory", 14)) {
+		if (!strcmp(names, "RedBoot") ||
+		    !strcmp(names, "RedBoot config") ||
+		    !strcmp(names, "FIS directory")) {
 			parts[i].mask_flags = MTD_WRITEABLE;
 		}
 #endif

Sasha Levin

maintainer

2026-03-20 21:55 UTC #1 lore raw

PATCH 1/2 mtd: partitions: redboot: fix style issues

From: Corentin Labbe <[email protected]>

[ Upstream commit eb1765c40530ccc8690b9dad88cec6aaa6bfb498 ]

This patch fixes easy checkpatch issues.

Signed-off-by Corentin Labbe <[email protected]>

Reviewed-by Linus Walleij <[email protected]>

Signed-off-by Miquel Raynal <[email protected]>

Link https://lore.kernel.org/linux-mtd/[email protected] open here

Stable-dep-of: 8e2f8020270a ("mtd: Avoid boot crash in RedBoot partition table parser")

Signed-off-by Sasha Levin <[email protected]>

---
 drivers/mtd/parsers/redboot.c | 69 +++++++++++++++++------------------
 1 file changed, 34 insertions(+), 35 deletions(-)

drivers/mtd/parsers/redboot.c +34 ~~-35~~

index 3351be6514732..3b55b676ca6b9 100644
@@ -17,15 +17,15 @@
 #include <linux/module.h>
 
 struct fis_image_desc {
-    unsigned char name[16];      // Null terminated name
-    uint32_t	  flash_base;    // Address within FLASH of image
-    uint32_t	  mem_base;      // Address in memory where it executes
-    uint32_t	  size;          // Length of image
-    uint32_t	  entry_point;   // Execution entry point
-    uint32_t	  data_length;   // Length of actual data
-    unsigned char _pad[256-(16+7*sizeof(uint32_t))];
-    uint32_t	  desc_cksum;    // Checksum over image descriptor
-    uint32_t	  file_cksum;    // Checksum over image data
+	unsigned char name[16];      // Null terminated name
+	u32	  flash_base;    // Address within FLASH of image
+	u32	  mem_base;      // Address in memory where it executes
+	u32	  size;          // Length of image
+	u32	  entry_point;   // Execution entry point
+	u32	  data_length;   // Length of actual data
+	unsigned char _pad[256 - (16 + 7 * sizeof(u32))];
+	u32	  desc_cksum;    // Checksum over image descriptor
+	u32	  file_cksum;    // Checksum over image data
 };
 
 struct fis_list {

@@ -91,12 +91,12 @@ static int parse_redboot_partitions(struct mtd_info *master,
 
 	parse_redboot_of(master);
 
-	if ( directory < 0 ) {
+	if (directory < 0) {
 		offset = master->size + directory * master->erasesize;
 		while (mtd_block_isbad(master, offset)) {
 			if (!offset) {
-			nogood:
-				printk(KERN_NOTICE "Failed to find a non-bad block to check for RedBoot partition table\n");
+nogood:
+				pr_notice("Failed to find a non-bad block to check for RedBoot partition table\n");
 				return -EIO;
 			}
 			offset -= master->erasesize;

@@ -114,8 +114,8 @@ static int parse_redboot_partitions(struct mtd_info *master,
 	if (!buf)
 		return -ENOMEM;
 
-	printk(KERN_NOTICE "Searching for RedBoot partition table in %s at offset 0x%lx\n",
-	       master->name, offset);
+	pr_notice("Searching for RedBoot partition table in %s at offset 0x%lx\n",
+		  master->name, offset);
 
 	ret = mtd_read(master, offset, master->erasesize, &retlen,
 		       (void *)buf);

@@ -151,14 +151,13 @@ static int parse_redboot_partitions(struct mtd_info *master,
 			     && swab32(buf[i].size) < master->erasesize)) {
 				int j;
 				/* Update numslots based on actual FIS directory size */
-				numslots = swab32(buf[i].size) / sizeof (struct fis_image_desc);
+				numslots = swab32(buf[i].size) / sizeof(struct fis_image_desc);
 				for (j = 0; j < numslots; ++j) {
-
 					/* A single 0xff denotes a deleted entry.
 					 * Two of them in a row is the end of the table.
 					 */
 					if (buf[j].name[0] == 0xff) {
-				  		if (buf[j].name[1] == 0xff) {
+						if (buf[j].name[1] == 0xff) {
 							break;
 						} else {
 							continue;

@@ -185,8 +184,8 @@ static int parse_redboot_partitions(struct mtd_info *master,
 	}
 	if (i == numslots) {
 		/* Didn't find it */
-		printk(KERN_NOTICE "No RedBoot partition table detected in %s\n",
-		       master->name);
+		pr_notice("No RedBoot partition table detected in %s\n",
+			  master->name);
 		ret = 0;
 		goto out;
 	}

@@ -205,7 +204,7 @@ static int parse_redboot_partitions(struct mtd_info *master,
 			break;
 
 		new_fl = kmalloc(sizeof(struct fis_list), GFP_KERNEL);
-		namelen += strlen(buf[i].name)+1;
+		namelen += strlen(buf[i].name) + 1;
 		if (!new_fl) {
 			ret = -ENOMEM;
 			goto out;

@@ -214,13 +213,13 @@ static int parse_redboot_partitions(struct mtd_info *master,
 		if (data && data->origin)
 			buf[i].flash_base -= data->origin;
 		else
-			buf[i].flash_base &= master->size-1;
+			buf[i].flash_base &= master->size - 1;
 
 		/* I'm sure the JFFS2 code has done me permanent damage.
 		 * I now think the following is _normal_
 		 */
 		prev = &fl;
-		while(*prev && (*prev)->img->flash_base < new_fl->img->flash_base)
+		while (*prev && (*prev)->img->flash_base < new_fl->img->flash_base)
 			prev = &(*prev)->next;
 		new_fl->next = *prev;
 		*prev = new_fl;

@@ -240,7 +239,7 @@ static int parse_redboot_partitions(struct mtd_info *master,
 		}
 	}
 #endif
-	parts = kzalloc(sizeof(*parts)*nrparts + nulllen + namelen, GFP_KERNEL);
+	parts = kzalloc(sizeof(*parts) * nrparts + nulllen + namelen, GFP_KERNEL);
 
 	if (!parts) {
 		ret = -ENOMEM;

@@ -249,23 +248,22 @@ static int parse_redboot_partitions(struct mtd_info *master,
 
 	nullname = (char *)&parts[nrparts];
 #ifdef CONFIG_MTD_REDBOOT_PARTS_UNALLOCATED
-	if (nulllen > 0) {
+	if (nulllen > 0)
 		strcpy(nullname, nullstring);
-	}
 #endif
 	names = nullname + nulllen;
 
-	i=0;
+	i = 0;
 
 #ifdef CONFIG_MTD_REDBOOT_PARTS_UNALLOCATED
 	if (fl->img->flash_base) {
-	       parts[0].name = nullname;
-	       parts[0].size = fl->img->flash_base;
-	       parts[0].offset = 0;
+		parts[0].name = nullname;
+		parts[0].size = fl->img->flash_base;
+		parts[0].offset = 0;
 		i++;
 	}
 #endif
-	for ( ; i<nrparts; i++) {
+	for ( ; i < nrparts; i++) {
 		parts[i].size = fl->img->size;
 		parts[i].offset = fl->img->flash_base;
 		parts[i].name = names;

@@ -273,17 +271,17 @@ static int parse_redboot_partitions(struct mtd_info *master,
 		strcpy(names, fl->img->name);
 #ifdef CONFIG_MTD_REDBOOT_PARTS_READONLY
 		if (!memcmp(names, "RedBoot", 8) ||
-				!memcmp(names, "RedBoot config", 15) ||
-				!memcmp(names, "FIS directory", 14)) {
+		    !memcmp(names, "RedBoot config", 15) ||
+		    !memcmp(names, "FIS directory", 14)) {
 			parts[i].mask_flags = MTD_WRITEABLE;
 		}
 #endif
-		names += strlen(names)+1;
+		names += strlen(names) + 1;
 
 #ifdef CONFIG_MTD_REDBOOT_PARTS_UNALLOCATED
-		if(fl->next && fl->img->flash_base + fl->img->size + master->erasesize <= fl->next->img->flash_base) {
+		if (fl->next && fl->img->flash_base + fl->img->size + master->erasesize <= fl->next->img->flash_base) {
 			i++;
-			parts[i].offset = parts[i-1].size + parts[i-1].offset;
+			parts[i].offset = parts[i - 1].size + parts[i - 1].offset;
 			parts[i].size = fl->next->img->flash_base - parts[i].offset;
 			parts[i].name = nullname;
 		}

@@ -297,6 +295,7 @@ static int parse_redboot_partitions(struct mtd_info *master,
  out:
 	while (fl) {
 		struct fis_list *old = fl;
+
 		fl = fl->next;
 		kfree(old);
 	}

Signature · 1 line

2.51.0

Sasha Levin

maintainer

2026-03-20 21:55 UTC #2 lore raw

PATCH 2/2 mtd: Avoid boot crash in RedBoot partition table parser

From: Finn Thain <[email protected]>

[ Upstream commit 8e2f8020270af7777d49c2e7132260983e4fc566 ]

Given CONFIG_FORTIFY_SOURCE=y and a recent compiler,
commit 439a1bcac648 ("fortify: Use __builtin_dynamic_object_size() when available") produces the warning below and an oops.

    Searching for RedBoot partition table in 50000000.flash at offset 0x7e0000
    ------------[ cut here ]------------
    WARNING: lib/string_helpers.c:1035 at 0xc029e04c, CPU#0: swapper/0/1
    memcmp: detected buffer overflow: 15 byte read of buffer size 14
    Modules linked in:
    CPU: 0 UID: 0 PID: 1 Comm: swapper/0 Not tainted 6.19.0 #1 NONE

Since the size of the dynamic allocation is calculated with strlen() we can use strcmp() instead of memcmp() and remain within bounds.

Cc Kees Cook <[email protected]>

Cc [email protected]

Link https://lore.kernel.org/all/202602151911.AD092DFFCD@keescook/ open here

Fixes 1da177e4c3f4 ("Linux-2.6.12-rc2")

Suggested-by Kees Cook <[email protected]>

Signed-off-by Finn Thain <[email protected]>

Signed-off-by Miquel Raynal <[email protected]>

Signed-off-by Sasha Levin <[email protected]>

---
 drivers/mtd/parsers/redboot.c | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

drivers/mtd/parsers/redboot.c +3 -3

index 3b55b676ca6b9..c06ba7a2a34b4 100644
@@ -270,9 +270,9 @@ static int parse_redboot_partitions(struct mtd_info *master,
 
 		strcpy(names, fl->img->name);
 #ifdef CONFIG_MTD_REDBOOT_PARTS_READONLY
-		if (!memcmp(names, "RedBoot", 8) ||
-		    !memcmp(names, "RedBoot config", 15) ||
-		    !memcmp(names, "FIS directory", 14)) {
+		if (!strcmp(names, "RedBoot") ||
+		    !strcmp(names, "RedBoot config") ||
+		    !strcmp(names, "FIS directory")) {
 			parts[i].mask_flags = MTD_WRITEABLE;
 		}
 #endif

Signature · 1 line

2.51.0