Skip to contentSkip to conversation

tst_kconfig: Avoid buffer overflow when parsing /proc/cmdline

PATCH Superseded Li Wang 2 participants 5 messages Jun 19, 2024 – Jun 20, 2024 Library code

v1 of 2 Superseded Latest is v2 →

lore
Share & tools
lore

Li Wang

author
#0 lore raw

PATCH 1/1 tst_kconfig: Avoid buffer overflow when parsing /proc/cmdline

When the test is run with a kernel booting with many parameters, the buffer size is often not large enough to store the complete command line. This results in a buffer overflow and the test complains with the following message:

tst_kconfig.c:609: TWARN: Buffer overflowed while parsing /proc/cmdline

---
 lib/tst_kconfig.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
lib/tst_kconfig.c +1 -1
index e16ca1400..be3842c2d 100644
@@ -569,7 +569,7 @@ char tst_kconfig_get(const char *confname)
 
 void tst_kcmdline_parse(struct tst_kcmdline_var params[], size_t params_len)
 {
-	char buf[128], line[512];
+	char buf[512], line[512];
 	size_t b_pos = 0,l_pos =0, i;
 	int var_id = -1;
 
Signature · 5 lines
2.45.2


-- 
Mailing list info: https://lists.linux.it/listinfo/ltp
Reviewed-by on v1 1/1
#1 lore raw
Quoted 6 lines

When the test is run with a kernel booting with many parameters, the buffer size is often not large enough to store the complete command line. This results in a buffer overflow and the test complains with the following message:

tst_kconfig.c:609: TWARN: Buffer overflowed while parsing /proc/cmdline

Thanks for the fix!

Fixes 180834982 ("kconfig: add funtion to parse /proc/cmdline")

NOTE tst_kconfig_read() has char line[128], also struct tst_kcmdline_var member has this length. The longest line on some on my systems is 109, it's still OK, hopefully it stays :).

Kind regards,
Petr

Signature · 1 line
Mailing list info: https://lists.linux.it/listinfo/ltp
author
#2 lore raw
Quoted 7 lines · Petr Vorel
Quoted 6 lines

When the test is run with a kernel booting with many parameters, the buffer size is often not large enough to store the complete command line. This results in a buffer overflow and the test complains with the following message:

tst_kconfig.c:609: TWARN: Buffer overflowed while parsing /proc/cmdline

Quoted 8 lines

Thanks for the fix!

Fixes 180834982 ("kconfig: add funtion to parse /proc/cmdline")

NOTE tst_kconfig_read() has char line[128], also struct tst_kcmdline_var member

Typically 128 is long enough for Linux kernel parameters, otherwise it hard for people to memorize the name and value.

Quoted 4 lines

has this length. The longest line on some on my systems is 109, it's still OK,
hopefully it stays :).

Hmm, good point. After thinking it over, seems too hasty to enlarge the buf[] size to 512.

We'd better keep the size same between 'tst_kcmdline_var.value' to the local 'buf[]'.
And the overflow is not a problem, it just drops some unused info[1] to reloop to
the head for saving our target parameter.

[1] BOOT_IMAGE=(hd0,gpt2)/ostree/centos-dd7415ed2c7cc3f65bdc4bf8f9a63b95bbb13ee0fee633f6b92a872944d1d6e4/vmlinuz-5.14.0-4xx.4xx.4495_1336583944.el9iv.x86_64

I'm thinking of lowering the priority info from TWARN to TINFO in that line:

    tst_res(TWARN, "Buffer overflowed while parsing /proc/cmdline");

I will send a patch V2 once get a better solution.

Signature · 5 lines
Regards,
Li Wang

-- 
Mailing list info: https://lists.linux.it/listinfo/ltp
#3 lore raw

Hi Li,

Quoted 14 lines · Petr Vorel
Quoted 12 lines
Quoted 7 lines

When the test is run with a kernel booting with many parameters, the buffer size is often not large enough to store the complete command line. This results in a buffer overflow and the test complains with the following message:

tst_kconfig.c:609: TWARN: Buffer overflowed while parsing /proc/cmdline

Thanks for the fix!

Fixes 180834982 ("kconfig: add funtion to parse /proc/cmdline")
Quoted 6 lines
Quoted 2 lines

NOTE tst_kconfig_read() has char line[128], also struct tst_kcmdline_var member

Typically 128 is long enough for Linux kernel parameters, otherwise it hard for people to memorize the name and value.

Sure, let's keep it for now, but examples we are getting close :). Sooner or later these configs, which are generated by toolchain will be longer than 128 chars:

$ awk 'length > max_length { max_length = length; longest_line = $0 } END { print longest_line }' /boot/config*
CONFIG_CC_VERSION_TEXT="gcc (SUSE Linux) 13.2.1 20240206 [revision 67ac78caf31f7cb3202177e6428a46d829b70f23]"
$ awk 'length > max_length { max_length = length; longest_line = $0 } END { print longest_line }' /boot/config* | wc -L
109

I was surprised even real config option was quite long:

$ awk 'length > max_length { max_length = length; longest_line = $0 } END { print longest_line }' /boot/config*
CONFIG_LSM="landlock,lockdown,yama,loadpin,safesetid,integrity,apparmor,selinux,smack,tomoyo,bpf"
$ awk 'length > max_length { max_length = length; longest_line = $0 } END { print longest_line }' /boot/config* | wc -L
97
Quoted 10 lines
Quoted 3 lines

has this length. The longest line on some on my systems is 109, it's still OK,
hopefully it stays :).

Hmm, good point. After thinking it over, seems too hasty to enlarge the buf[] size to 512.

We'd better keep the size same between 'tst_kcmdline_var.value' to the local 'buf[]'.

+1. I would even define a constant to keep these 2 sizes the same.

Quoted 5 lines

And the overflow is not a problem, it just drops some unused info[1] to reloop to
the head for saving our target parameter.

[1] BOOT_IMAGE=(hd0,gpt2)/ostree/centos-dd7415ed2c7cc3f65bdc4bf8f9a63b95bbb13ee0fee633f6b92a872944d1d6e4/vmlinuz-5.14.0-4xx.4xx.4495_1336583944.el9iv.x86_64

Ah, what a long parameter :). OTOH would it harm to enlarge both to 256 or even 512?

Quoted 2 lines

I'm thinking of lowering the priority info from TWARN to TINFO in that line:

    tst_res(TWARN, "Buffer overflowed while parsing /proc/cmdline");

Yes, but are you sure that occasional long parameter will be always irrelevant? I'd really increase the size.

Kind regards,
Petr

Quoted 1 line

I will send a patch V2 once get a better solution.

Signature · 1 line
Mailing list info: https://lists.linux.it/listinfo/ltp
author
#4 lore raw

Hi Petr,

Quoted 69 lines · Petr Vorel

Hi Li,

Quoted 8 lines · Petr Vorel
Quoted 7 lines
Quoted 7 lines

When the test is run with a kernel booting with many parameters, the buffer size is often not large enough to store the complete command line. This results in a buffer overflow and the test complains with the following message:

tst_kconfig.c:609: TWARN: Buffer overflowed while parsing

/proc/cmdline

Quoted 6 lines
Quoted 6 lines

Thanks for the fix!

Fixes 180834982 ("kconfig: add funtion to parse /proc/cmdline")

NOTE tst_kconfig_read() has char line[128], also struct

tst_kcmdline_var

Quoted 4 lines
Quoted 1 line

member

Typically 128 is long enough for Linux kernel parameters, otherwise it

hard

Quoted 1 line

for people to memorize the name and value.

Sure, let's keep it for now, but examples we are getting close :). Sooner or
later these configs, which are generated by toolchain will be longer than 128
chars:

$ awk 'length > max_length { max_length = length; longest_line = $0 } END
{ print longest_line }' /boot/config*
CONFIG_CC_VERSION_TEXT="gcc (SUSE Linux) 13.2.1 20240206 [revision
67ac78caf31f7cb3202177e6428a46d829b70f23]"
$ awk 'length > max_length { max_length = length; longest_line = $0 } END
{ print longest_line }' /boot/config* | wc -L
109

I was surprised even real config option was quite long:

$ awk 'length > max_length { max_length = length; longest_line = $0 } END
{ print longest_line }' /boot/config*

CONFIG_LSM="landlock,lockdown,yama,loadpin,safesetid,integrity,apparmor,selinux,smack,tomoyo,bpf"

$ awk 'length > max_length { max_length = length; longest_line = $0 } END
{ print longest_line }' /boot/config* | wc -L
97
Quoted 1 line
Quoted 1 line

has this length. The longest line on some on my systems is 109, it's

still

Quoted 9 lines
Quoted 2 lines

OK,
hopefully it stays :).

Hmm, good point. After thinking it over, seems too hasty to enlarge the buf[] size to 512.

We'd better keep the size same between 'tst_kcmdline_var.value' to the local 'buf[]'.

+1. I would even define a constant to keep these 2 sizes the same.

Quoted 3 lines

And the overflow is not a problem, it just drops some unused info[1] to reloop to
the head for saving our target parameter.

Quoted 7 lines
Quoted 1 line

[1]

BOOT_IMAGE=(hd0,gpt2)/ostree/centos-dd7415ed2c7cc3f65bdc4bf8f9a63b95bbb13ee0fee633f6b92a872944d1d6e4/vmlinuz-5.14.0-4xx.4xx.4495_1336583944.el9iv.x86_64

Ah, what a long parameter :). OTOH would it harm to enlarge both to 256 or even
512?

Yes, but to be honest, we always don't care about that long parameter, and mostly, people just parse something they pass into the kernel, it is always short and popular.

So I believe raise to 256 is enough size for LTP.

Quoted 9 lines
Quoted 2 lines

I'm thinking of lowering the priority info from TWARN to TINFO in that

line:

Quoted 1 line
    tst_res(TWARN, "Buffer overflowed while parsing /proc/cmdline");

Yes, but are you sure that occasional long parameter will be always irrelevant?
I'd really increase the size.

I guess yes, at least so far I don't see any LTP test parse long parameters, it is safe to drop them IMHO.

Signature · 5 lines
Regards,
Li Wang

-- 
Mailing list info: https://lists.linux.it/listinfo/ltp