Skip to contentSkip to conversation

modules: Fix corruption of /proc/kallsyms

PATCH Applied Adrian Hunter 3 participants 5 messages Jul 1, 2022 – Jul 11, 2022 Module support

lore
Share & tools
lore
#0 lore raw

PATCH 1/1 modules: Fix corruption of /proc/kallsyms

The commit 91fb02f31505 ("module: Move kallsyms support into a separate file") changed from using strlcpy() to using strscpy() which created a buffer overflow. That happened because:
1) an incorrect value was passed as the buffer length
2) strscpy() (unlike strlcpy()) may copy beyond the length of the input string when copying word-by-word.
The assumption was that because it was already known that the strings being copied would fit in the space available, it was not necessary to correctly set the buffer length. strscpy() breaks that assumption because although it will not touch bytes beyond the given buffer length it may write bytes beyond the input string length when writing word-by-word.

The result of the buffer overflow is to corrupt the symbol type information that follows. e.g.

 $ sudo cat -v /proc/kallsyms | grep '\^' | head
 ffffffffc0615000 ^@ rfcomm_session_get  [rfcomm]
 ffffffffc061c060 ^@ session_list        [rfcomm]
 ffffffffc06150d0 ^@ rfcomm_send_frame   [rfcomm]
 ffffffffc0615130 ^@ rfcomm_make_uih     [rfcomm]
 ffffffffc07ed58d ^@ bnep_exit   [bnep]
 ffffffffc07ec000 ^@ bnep_rx_control     [bnep]
 ffffffffc07ec1a0 ^@ bnep_session        [bnep]
 ffffffffc07e7000 ^@ input_leds_event    [input_leds]
 ffffffffc07e9000 ^@ input_leds_handler  [input_leds]
 ffffffffc07e7010 ^@ input_leds_disconnect       [input_leds]

Notably, the null bytes (represented above by ^@) can confuse tools.

Fix by correcting the buffer length.

Fixes 91fb02f31505 ("module: Move kallsyms support into a separate file")
---
 kernel/module/kallsyms.c | 15 ++++++++++++---
 1 file changed, 12 insertions(+), 3 deletions(-)
kernel/module/kallsyms.c +12 -3
index 3e11523bc6f6..18c23545b984 100644
@@ -137,6 +137,7 @@ void layout_symtab(struct module *mod, struct load_info *info)
 	info->symoffs = ALIGN(mod->data_layout.size, symsect->sh_addralign ?: 1);
 	info->stroffs = mod->data_layout.size = info->symoffs + ndst * sizeof(Elf_Sym);
 	mod->data_layout.size += strtab_size;
+	/* Note add_kallsyms() computes strtab_size as core_typeoffs - stroffs */
 	info->core_typeoffs = mod->data_layout.size;
 	mod->data_layout.size += ndst * sizeof(char);
 	mod->data_layout.size = strict_align(mod->data_layout.size);
@@ -169,6 +170,7 @@ void add_kallsyms(struct module *mod, const struct load_info *info)
 	Elf_Sym *dst;
 	char *s;
 	Elf_Shdr *symsec = &info->sechdrs[info->index.sym];
+	unsigned long strtab_size;
 
 	/* Set up to point into init section. */
 	mod->kallsyms = (void __rcu *)mod->init_layout.base +
@@ -190,19 +192,26 @@ void add_kallsyms(struct module *mod, const struct load_info *info)
 	mod->core_kallsyms.symtab = dst = mod->data_layout.base + info->symoffs;
 	mod->core_kallsyms.strtab = s = mod->data_layout.base + info->stroffs;
 	mod->core_kallsyms.typetab = mod->data_layout.base + info->core_typeoffs;
+	strtab_size = info->core_typeoffs - info->stroffs;
 	src = rcu_dereference_sched(mod->kallsyms)->symtab;
 	for (ndst = i = 0; i < rcu_dereference_sched(mod->kallsyms)->num_symtab; i++) {
 		rcu_dereference_sched(mod->kallsyms)->typetab[i] = elf_type(src + i, info);
 		if (i == 0 || is_livepatch_module(mod) ||
 		    is_core_symbol(src + i, info->sechdrs, info->hdr->e_shnum,
 				   info->index.pcpu)) {
+			ssize_t ret;
+
 			mod->core_kallsyms.typetab[ndst] =
 			    rcu_dereference_sched(mod->kallsyms)->typetab[i];
 			dst[ndst] = src[i];
 			dst[ndst++].st_name = s - mod->core_kallsyms.strtab;
-			s += strscpy(s,
-				     &rcu_dereference_sched(mod->kallsyms)->strtab[src[i].st_name],
-				     KSYM_NAME_LEN) + 1;
+			ret = strscpy(s,
+				      &rcu_dereference_sched(mod->kallsyms)->strtab[src[i].st_name],
+				      strtab_size);
+			if (ret < 0)
+				break;
+			s += ret + 1;
+			strtab_size -= ret + 1;
 		}
 	}
 	preempt_enable();
Signature · 1 line
2.25.1
maintainer Applied
#1 lore raw
Quoted 28 lines · Adrian Hunter

The commit 91fb02f31505 ("module: Move kallsyms support into a separate file") changed from using strlcpy() to using strscpy() which created a buffer overflow. That happened because:
1) an incorrect value was passed as the buffer length
2) strscpy() (unlike strlcpy()) may copy beyond the length of the input string when copying word-by-word.
The assumption was that because it was already known that the strings being copied would fit in the space available, it was not necessary to correctly set the buffer length. strscpy() breaks that assumption because although it will not touch bytes beyond the given buffer length it may write bytes beyond the input string length when writing word-by-word.

The result of the buffer overflow is to corrupt the symbol type information that follows. e.g.

 $ sudo cat -v /proc/kallsyms | grep '\^' | head
 ffffffffc0615000 ^@ rfcomm_session_get  [rfcomm]
 ffffffffc061c060 ^@ session_list        [rfcomm]
 ffffffffc06150d0 ^@ rfcomm_send_frame   [rfcomm]
 ffffffffc0615130 ^@ rfcomm_make_uih     [rfcomm]
 ffffffffc07ed58d ^@ bnep_exit   [bnep]
 ffffffffc07ec000 ^@ bnep_rx_control     [bnep]
 ffffffffc07ec1a0 ^@ bnep_session        [bnep]
 ffffffffc07e7000 ^@ input_leds_event    [input_leds]
 ffffffffc07e9000 ^@ input_leds_handler  [input_leds]
 ffffffffc07e7010 ^@ input_leds_disconnect       [input_leds]
Quoted 6 lines

Notably, the null bytes (represented above by ^@) can confuse tools.

Fix by correcting the buffer length.

Fixes 91fb02f31505 ("module: Move kallsyms support into a separate file")

Queued up thanks!

Luis

#2 lore raw
Quoted 31 lines · Luis Chamberlain
Quoted 30 lines · Adrian Hunter

The commit 91fb02f31505 ("module: Move kallsyms support into a separate file") changed from using strlcpy() to using strscpy() which created a buffer overflow. That happened because:
1) an incorrect value was passed as the buffer length
2) strscpy() (unlike strlcpy()) may copy beyond the length of the input string when copying word-by-word.
The assumption was that because it was already known that the strings being copied would fit in the space available, it was not necessary to correctly set the buffer length. strscpy() breaks that assumption because although it will not touch bytes beyond the given buffer length it may write bytes beyond the input string length when writing word-by-word.

The result of the buffer overflow is to corrupt the symbol type information that follows. e.g.

 $ sudo cat -v /proc/kallsyms | grep '\^' | head
 ffffffffc0615000 ^@ rfcomm_session_get  [rfcomm]
 ffffffffc061c060 ^@ session_list        [rfcomm]
 ffffffffc06150d0 ^@ rfcomm_send_frame   [rfcomm]
 ffffffffc0615130 ^@ rfcomm_make_uih     [rfcomm]
 ffffffffc07ed58d ^@ bnep_exit   [bnep]
 ffffffffc07ec000 ^@ bnep_rx_control     [bnep]
 ffffffffc07ec1a0 ^@ bnep_session        [bnep]
 ffffffffc07e7000 ^@ input_leds_event    [input_leds]
 ffffffffc07e9000 ^@ input_leds_handler  [input_leds]
 ffffffffc07e7010 ^@ input_leds_disconnect       [input_leds]

Notably, the null bytes (represented above by ^@) can confuse tools.

Quoted 8 lines
Quoted 4 lines

Fix by correcting the buffer length.

Fixes 91fb02f31505 ("module: Move kallsyms support into a separate file")

Queued up thanks!

Luis

Thanks for processing this.

I notice it is -rc6 and I do not see it in Linus' tree. This is a fix for a regression, shouldn't it be included in 5.19?

maintainer
#3 lore raw
Quoted 37 lines · Adrian Hunter
Quoted 36 lines · Luis Chamberlain
Quoted 34 lines · Adrian Hunter

The commit 91fb02f31505 ("module: Move kallsyms support into a separate file") changed from using strlcpy() to using strscpy() which created a buffer overflow. That happened because:
1) an incorrect value was passed as the buffer length
2) strscpy() (unlike strlcpy()) may copy beyond the length of the input string when copying word-by-word.
The assumption was that because it was already known that the strings being copied would fit in the space available, it was not necessary to correctly set the buffer length. strscpy() breaks that assumption because although it will not touch bytes beyond the given buffer length it may write bytes beyond the input string length when writing word-by-word.

The result of the buffer overflow is to corrupt the symbol type information that follows. e.g.

 $ sudo cat -v /proc/kallsyms | grep '\^' | head
 ffffffffc0615000 ^@ rfcomm_session_get  [rfcomm]
 ffffffffc061c060 ^@ session_list        [rfcomm]
 ffffffffc06150d0 ^@ rfcomm_send_frame   [rfcomm]
 ffffffffc0615130 ^@ rfcomm_make_uih     [rfcomm]
 ffffffffc07ed58d ^@ bnep_exit   [bnep]
 ffffffffc07ec000 ^@ bnep_rx_control     [bnep]
 ffffffffc07ec1a0 ^@ bnep_session        [bnep]
 ffffffffc07e7000 ^@ input_leds_event    [input_leds]
 ffffffffc07e9000 ^@ input_leds_handler  [input_leds]
 ffffffffc07e7010 ^@ input_leds_disconnect       [input_leds]

Notably, the null bytes (represented above by ^@) can confuse tools.

Fix by correcting the buffer length.

Fixes 91fb02f31505 ("module: Move kallsyms support into a separate file")
Quoted 8 lines
Quoted 3 lines

Queued up thanks!

Luis

Thanks for processing this.

I notice it is -rc6 and I do not see it in Linus' tree. This is a fix for a regression, shouldn't it be included in 5.19?

Yeah sorry this will get to Linus today.

Luis

#4 lore raw
In reply to Adrian Hunter
Quoted 30 lines · Adrian Hunter

The commit 91fb02f31505 ("module: Move kallsyms support into a separate file") changed from using strlcpy() to using strscpy() which created a buffer overflow. That happened because:
1) an incorrect value was passed as the buffer length
2) strscpy() (unlike strlcpy()) may copy beyond the length of the input string when copying word-by-word.
The assumption was that because it was already known that the strings being copied would fit in the space available, it was not necessary to correctly set the buffer length. strscpy() breaks that assumption because although it will not touch bytes beyond the given buffer length it may write bytes beyond the input string length when writing word-by-word.

The result of the buffer overflow is to corrupt the symbol type information that follows. e.g.

 $ sudo cat -v /proc/kallsyms | grep '\^' | head
 ffffffffc0615000 ^@ rfcomm_session_get  [rfcomm]
 ffffffffc061c060 ^@ session_list        [rfcomm]
 ffffffffc06150d0 ^@ rfcomm_send_frame   [rfcomm]
 ffffffffc0615130 ^@ rfcomm_make_uih     [rfcomm]
 ffffffffc07ed58d ^@ bnep_exit   [bnep]
 ffffffffc07ec000 ^@ bnep_rx_control     [bnep]
 ffffffffc07ec1a0 ^@ bnep_session        [bnep]
 ffffffffc07e7000 ^@ input_leds_event    [input_leds]
 ffffffffc07e9000 ^@ input_leds_handler  [input_leds]
 ffffffffc07e7010 ^@ input_leds_disconnect       [input_leds]

Notably, the null bytes (represented above by ^@) can confuse tools.

Quoted 7 lines

Fix by correcting the buffer length.

Fixes 91fb02f31505 ("module: Move kallsyms support into a separate file")
---
 kernel/module/kallsyms.c | 15 ++++++++++++---
 1 file changed, 12 insertions(+), 3 deletions(-)

Hi Adrian,

Thanks for following up!

Kind regards,

Signature · 1 line
Aaron Tomlin