Skip to contentSkip to conversation

init/version.c: Replace strlcpy with strscpy

PATCH Applied Azeem Shaikh 3 participants 4 messages Aug 30, 2023 – Sep 15, 2023

lore
Share & tools
lore
#0 lore raw

PATCH 1/1 init/version.c: Replace strlcpy with strscpy

strlcpy() reads the entire source buffer first. This read may exceed the destination size limit. This is both inefficient and can lead to linear read overflows if a source string is not NUL-terminated [1]. In an effort to remove strlcpy() completely [2], replace strlcpy() here with strscpy().

Direct replacement is safe here since return value of -errno is used to check for truncation instead of sizeof(dest).

[1] https://www.kernel.org/doc/html/latest/process/deprecated.html#strlcpy
[2] https://github.com/KSPP/linux/issues/89

---
 init/version.c |    6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)
init/version.c +3 -3
index f117921811b4..94c96f6fbfe6 100644
@@ -21,10 +21,10 @@ static int __init early_hostname(char *arg)
 {
 	size_t bufsize = sizeof(init_uts_ns.name.nodename);
 	size_t maxlen  = bufsize - 1;
-	size_t arglen;
+	ssize_t arglen;
 
-	arglen = strlcpy(init_uts_ns.name.nodename, arg, bufsize);
-	if (arglen > maxlen) {
+	arglen = strscpy(init_uts_ns.name.nodename, arg, bufsize);
+	if (arglen < 0) {
 		pr_warn("hostname parameter exceeds %zd characters and will be truncated",
 			maxlen);
 	}

--
2.41.0.255.g8b1d071c50-goog

Reviewed-by on v1 1/1
#1 lore raw
Quoted 10 lines · Azeem Shaikh

strlcpy() reads the entire source buffer first. This read may exceed the destination size limit. This is both inefficient and can lead to linear read overflows if a source string is not NUL-terminated [1]. In an effort to remove strlcpy() completely [2], replace strlcpy() here with strscpy().

Direct replacement is safe here since return value of -errno is used to check for truncation instead of sizeof(dest).

Great patch! This is similar to [1] which aims to eradicate strncpy.

Quoted 5 lines

[1] https://www.kernel.org/doc/html/latest/process/deprecated.html#strlcpy
[2] https://github.com/KSPP/linux/issues/89

Quoted 23 lines
---
 init/version.c |    6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)
init/version.c +3 -4
index f117921811b4..94c96f6fbfe6 100644
@@ -21,10 +21,10 @@ static int __init early_hostname(char *arg)
 {
        size_t bufsize = sizeof(init_uts_ns.name.nodename);
        size_t maxlen  = bufsize - 1;
-       size_t arglen;
+       ssize_t arglen;
 
-       arglen = strlcpy(init_uts_ns.name.nodename, arg, bufsize);
-       if (arglen > maxlen) {
+       arglen = strscpy(init_uts_ns.name.nodename, arg, bufsize);
+       if (arglen < 0) {
                pr_warn("hostname parameter exceeds %zd characters and will be truncated",
                        maxlen);
        }
--
Quoted 3 lines

2.41.0.255.g8b1d071c50-goog

[1]: https://github.com/KSPP/linux/issues/90

Reviewed-by on v1 1/1
#2 lore raw
In reply to Azeem Shaikh
Quoted 23 lines · Azeem Shaikh

strlcpy() reads the entire source buffer first. This read may exceed the destination size limit. This is both inefficient and can lead to linear read overflows if a source string is not NUL-terminated [1]. In an effort to remove strlcpy() completely [2], replace strlcpy() here with strscpy().

Direct replacement is safe here since return value of -errno is used to check for truncation instead of sizeof(dest).

[1] https://www.kernel.org/doc/html/latest/process/deprecated.html#strlcpy
[2] https://github.com/KSPP/linux/issues/89

---
 init/version.c |    6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)
init/version.c +0 -0
index f117921811b4..94c96f6fbfe6 100644

init/version.c static int __init early_hostname(char *arg) · line 27

Quoted 10 lines
init/version.c +3 -3
@@ -21,10 +21,10 @@ static int __init early_hostname(char *arg)
 {
 	size_t bufsize = sizeof(init_uts_ns.name.nodename);
 	size_t maxlen  = bufsize - 1;
-	size_t arglen;
+	ssize_t arglen;
 
-	arglen = strlcpy(init_uts_ns.name.nodename, arg, bufsize);
-	if (arglen > maxlen) {
+	arglen = strscpy(init_uts_ns.name.nodename, arg, bufsize);
+	if (arglen < 0) {

nitpick: this is no longer "length of arg", it's the length of the destination string. Regardless:

Quoted 7 lines
init/version.c +0 -1
 		pr_warn("hostname parameter exceeds %zd characters and will be truncated",
 			maxlen);
 	}
--

2.41.0.255.g8b1d071c50-goog

Signature · 1 line
Kees Cook
Applied
#3 lore raw
In reply to Azeem Shaikh
Quoted 8 lines · Azeem Shaikh

strlcpy() reads the entire source buffer first. This read may exceed the destination size limit. This is both inefficient and can lead to linear read overflows if a source string is not NUL-terminated [1]. In an effort to remove strlcpy() completely [2], replace strlcpy() here with strscpy().

[...]

Applied to for-next/hardening, thanks!

[1/1] init/version.c: Replace strlcpy with strscpy

      https://git.kernel.org/kees/c/ec23bc09c1c0

Take care,

Signature · 1 line
Kees Cook