This is a note to let you know that I've just added the patch titled
modules: fix longstanding /proc/kallsyms vs module insertion race.
to the 3.14-stable tree which can be found at:
http://www.kernel.org/git/?p=linux/kernel/git/stable/stable-queue.git;a=summary
The filename of the patch is:
modules-fix-longstanding-proc-kallsyms-vs-module-insertion-race.patch
and it can be found in the queue-3.14 subdirectory.
If you, or anyone else, feels it should not be added to the stable tree,
please let <stable@xxxxxxxxxxxxxxx> know about it.
>From 8244062ef1e54502ef55f54cced659913f244c3e Mon Sep 17 00:00:00 2001
From: Rusty Russell <rusty@xxxxxxxxxxxxxxx>
Date: Wed, 3 Feb 2016 16:55:26 +1030
Subject: modules: fix longstanding /proc/kallsyms vs module insertion race.
From: Rusty Russell <rusty@xxxxxxxxxxxxxxx>
commit 8244062ef1e54502ef55f54cced659913f244c3e upstream.
For CONFIG_KALLSYMS, we keep two symbol tables and two string tables.
There's one full copy, marked SHF_ALLOC and laid out at the end of the
module's init section. There's also a cut-down version that only
contains core symbols and strings, and lives in the module's core
section.
After module init (and before we free the module memory), we switch
the mod->symtab, mod->num_symtab and mod->strtab to point to the core
versions. We do this under the module_mutex.
However, kallsyms doesn't take the module_mutex: it uses
preempt_disable() and rcu tricks to walk through the modules, because
it's used in the oops path. It's also used in /proc/kallsyms.
There's nothing atomic about the change of these variables, so we can
get the old (larger!) num_symtab and the new symtab pointer; in fact
this is what I saw when trying to reproduce.
By grouping these variables together, we can use a
carefully-dereferenced pointer to ensure we always get one or the
other (the free of the module init section is already done in an RCU
callback, so that's safe). We allocate the init one at the end of the
module init section, and keep the core one inside the struct module
itself (it could also have been allocated at the end of the module
core, but that's probably overkill).
[ Rebased for 4.4-stable and older, because the following changes aren't
in the older trees:
- e0224418516b4d8a6c2160574bac18447c354ef0: adds arg to is_core_symbol
- 7523e4dc5057e157212b4741abd6256e03404cf1: module_init/module_core/init_size/core_size
become init_layout.base/core_layout.base/init_layout.size/core_layout.size.
Original commit: 8244062ef1e54502ef55f54cced659913f244c3e
]
Reported-by: Weilong Chen <chenweilong@xxxxxxxxxx>
Bugzilla: https://bugzilla.kernel.org/show_bug.cgi?id=111541
Signed-off-by: Rusty Russell <rusty@xxxxxxxxxxxxxxx>
Signed-off-by: Greg Kroah-Hartman <gregkh@xxxxxxxxxxxxxxxxxxx>
---
include/linux/module.h | 17 +++----
kernel/module.c | 111 ++++++++++++++++++++++++++++++-------------------
2 files changed, 78 insertions(+), 50 deletions(-)
--- a/include/linux/module.h
+++ b/include/linux/module.h
@@ -227,6 +227,12 @@ struct module_ref {
unsigned long decs;
} __attribute((aligned(2 * sizeof(unsigned long))));
+struct mod_kallsyms {
+ Elf_Sym *symtab;
+ unsigned int num_symtab;
+ char *strtab;
+};
+
struct module {
enum module_state state;
@@ -314,14 +320,9 @@ struct module {
#endif
#ifdef CONFIG_KALLSYMS
- /*
- * We keep the symbol and string tables for kallsyms.
- * The core_* fields below are temporary, loader-only (they
- * could really be discarded after module init).
- */
- Elf_Sym *symtab, *core_symtab;
- unsigned int num_symtab, core_num_syms;
- char *strtab, *core_strtab;
+ /* Protected by RCU and/or module_mutex: use rcu_dereference() */
+ struct mod_kallsyms *kallsyms;
+ struct mod_kallsyms core_kallsyms;
/* Section attributes */
struct module_sect_attrs *sect_attrs;
--- a/kernel/module.c
+++ b/kernel/module.c
@@ -180,6 +180,9 @@ struct load_info {
struct _ddebug *debug;
unsigned int num_debug;
bool sig_ok;
+#ifdef CONFIG_KALLSYMS
+ unsigned long mod_kallsyms_init_off;
+#endif
struct {
unsigned int sym, str, mod, vers, info, pcpu;
} index;
@@ -2320,8 +2323,20 @@ static void layout_symtab(struct module
strsect->sh_entsize = get_offset(mod, &mod->init_size, strsect,
info->index.str) | INIT_OFFSET_MASK;
pr_debug("\t%s\n", info->secstrings + strsect->sh_name);
+
+ /* We'll tack temporary mod_kallsyms on the end. */
+ mod->init_size = ALIGN(mod->init_size,
+ __alignof__(struct mod_kallsyms));
+ info->mod_kallsyms_init_off = mod->init_size;
+ mod->init_size += sizeof(struct mod_kallsyms);
+ mod->init_size = debug_align(mod->init_size);
}
+/*
+ * We use the full symtab and strtab which layout_symtab arranged to
+ * be appended to the init section. Later we switch to the cut-down
+ * core-only ones.
+ */
static void add_kallsyms(struct module *mod, const struct load_info *info)
{
unsigned int i, ndst;
@@ -2330,28 +2345,33 @@ static void add_kallsyms(struct module *
char *s;
Elf_Shdr *symsec = &info->sechdrs[info->index.sym];
- mod->symtab = (void *)symsec->sh_addr;
- mod->num_symtab = symsec->sh_size / sizeof(Elf_Sym);
+ /* Set up to point into init section. */
+ mod->kallsyms = mod->module_init + info->mod_kallsyms_init_off;
+
+ mod->kallsyms->symtab = (void *)symsec->sh_addr;
+ mod->kallsyms->num_symtab = symsec->sh_size / sizeof(Elf_Sym);
/* Make sure we get permanent strtab: don't use info->strtab. */
- mod->strtab = (void *)info->sechdrs[info->index.str].sh_addr;
+ mod->kallsyms->strtab = (void *)info->sechdrs[info->index.str].sh_addr;
/* Set types up while we still have access to sections. */
- for (i = 0; i < mod->num_symtab; i++)
- mod->symtab[i].st_info = elf_type(&mod->symtab[i], info);
-
- mod->core_symtab = dst = mod->module_core + info->symoffs;
- mod->core_strtab = s = mod->module_core + info->stroffs;
- src = mod->symtab;
- for (ndst = i = 0; i < mod->num_symtab; i++) {
+ for (i = 0; i < mod->kallsyms->num_symtab; i++)
+ mod->kallsyms->symtab[i].st_info
+ = elf_type(&mod->kallsyms->symtab[i], info);
+
+ /* Now populate the cut down core kallsyms for after init. */
+ mod->core_kallsyms.symtab = dst = mod->module_core + info->symoffs;
+ mod->core_kallsyms.strtab = s = mod->module_core + info->stroffs;
+ src = mod->kallsyms->symtab;
+ for (ndst = i = 0; i < mod->kallsyms->num_symtab; i++) {
if (i == 0 ||
is_core_symbol(src+i, info->sechdrs, info->hdr->e_shnum)) {
dst[ndst] = src[i];
- dst[ndst++].st_name = s - mod->core_strtab;
- s += strlcpy(s, &mod->strtab[src[i].st_name],
+ dst[ndst++].st_name = s - mod->core_kallsyms.strtab;
+ s += strlcpy(s, &mod->kallsyms->strtab[src[i].st_name],
KSYM_NAME_LEN) + 1;
}
}
- mod->core_num_syms = ndst;
+ mod->core_kallsyms.num_symtab = ndst;
}
#else
static inline void layout_symtab(struct module *mod, struct load_info *info)
@@ -3089,9 +3109,8 @@ static int do_init_module(struct module
module_put(mod);
trim_init_extable(mod);
#ifdef CONFIG_KALLSYMS
- mod->num_symtab = mod->core_num_syms;
- mod->symtab = mod->core_symtab;
- mod->strtab = mod->core_strtab;
+ /* Switch to core kallsyms now init is done: kallsyms may be walking! */
+ rcu_assign_pointer(mod->kallsyms, &mod->core_kallsyms);
#endif
unset_module_init_ro_nx(mod);
module_free(mod, mod->module_init);
@@ -3381,9 +3400,9 @@ static inline int is_arm_mapping_symbol(
&& (str[2] == '\0' || str[2] == '.');
}
-static const char *symname(struct module *mod, unsigned int symnum)
+static const char *symname(struct mod_kallsyms *kallsyms, unsigned int symnum)
{
- return mod->strtab + mod->symtab[symnum].st_name;
+ return kallsyms->strtab + kallsyms->symtab[symnum].st_name;
}
static const char *get_ksymbol(struct module *mod,
@@ -3393,6 +3412,7 @@ static const char *get_ksymbol(struct mo
{
unsigned int i, best = 0;
unsigned long nextval;
+ struct mod_kallsyms *kallsyms = rcu_dereference_sched(mod->kallsyms);
/* At worse, next value is at end of module */
if (within_module_init(addr, mod))
@@ -3402,32 +3422,32 @@ static const char *get_ksymbol(struct mo
/* Scan for closest preceding symbol, and next symbol. (ELF
starts real symbols at 1). */
- for (i = 1; i < mod->num_symtab; i++) {
- if (mod->symtab[i].st_shndx == SHN_UNDEF)
+ for (i = 1; i < kallsyms->num_symtab; i++) {
+ if (kallsyms->symtab[i].st_shndx == SHN_UNDEF)
continue;
/* We ignore unnamed symbols: they're uninformative
* and inserted at a whim. */
- if (*symname(mod, i) == '\0'
- || is_arm_mapping_symbol(symname(mod, i)))
+ if (*symname(kallsyms, i) == '\0'
+ || is_arm_mapping_symbol(symname(kallsyms, i)))
continue;
- if (mod->symtab[i].st_value <= addr
- && mod->symtab[i].st_value > mod->symtab[best].st_value)
+ if (kallsyms->symtab[i].st_value <= addr
+ && kallsyms->symtab[i].st_value > kallsyms->symtab[best].st_value)
best = i;
- if (mod->symtab[i].st_value > addr
- && mod->symtab[i].st_value < nextval)
- nextval = mod->symtab[i].st_value;
+ if (kallsyms->symtab[i].st_value > addr
+ && kallsyms->symtab[i].st_value < nextval)
+ nextval = kallsyms->symtab[i].st_value;
}
if (!best)
return NULL;
if (size)
- *size = nextval - mod->symtab[best].st_value;
+ *size = nextval - kallsyms->symtab[best].st_value;
if (offset)
- *offset = addr - mod->symtab[best].st_value;
- return symname(mod, best);
+ *offset = addr - kallsyms->symtab[best].st_value;
+ return symname(kallsyms, best);
}
/* For kallsyms to ask for address resolution. NULL means not found. Careful
@@ -3523,18 +3543,21 @@ int module_get_kallsym(unsigned int symn
preempt_disable();
list_for_each_entry_rcu(mod, &modules, list) {
+ struct mod_kallsyms *kallsyms;
+
if (mod->state == MODULE_STATE_UNFORMED)
continue;
- if (symnum < mod->num_symtab) {
- *value = mod->symtab[symnum].st_value;
- *type = mod->symtab[symnum].st_info;
- strlcpy(name, symname(mod, symnum), KSYM_NAME_LEN);
+ kallsyms = rcu_dereference_sched(mod->kallsyms);
+ if (symnum < kallsyms->num_symtab) {
+ *value = kallsyms->symtab[symnum].st_value;
+ *type = kallsyms->symtab[symnum].st_info;
+ strlcpy(name, symname(kallsyms, symnum), KSYM_NAME_LEN);
strlcpy(module_name, mod->name, MODULE_NAME_LEN);
*exported = is_exported(name, *value, mod);
preempt_enable();
return 0;
}
- symnum -= mod->num_symtab;
+ symnum -= kallsyms->num_symtab;
}
preempt_enable();
return -ERANGE;
@@ -3543,11 +3566,12 @@ int module_get_kallsym(unsigned int symn
static unsigned long mod_find_symname(struct module *mod, const char *name)
{
unsigned int i;
+ struct mod_kallsyms *kallsyms = rcu_dereference_sched(mod->kallsyms);
- for (i = 0; i < mod->num_symtab; i++)
- if (strcmp(name, symname(mod, i)) == 0 &&
- mod->symtab[i].st_info != 'U')
- return mod->symtab[i].st_value;
+ for (i = 0; i < kallsyms->num_symtab; i++)
+ if (strcmp(name, symname(kallsyms, i)) == 0 &&
+ kallsyms->symtab[i].st_info != 'U')
+ return kallsyms->symtab[i].st_value;
return 0;
}
@@ -3584,11 +3608,14 @@ int module_kallsyms_on_each_symbol(int (
int ret;
list_for_each_entry(mod, &modules, list) {
+ /* We hold module_mutex: no need for rcu_dereference_sched */
+ struct mod_kallsyms *kallsyms = mod->kallsyms;
+
if (mod->state == MODULE_STATE_UNFORMED)
continue;
- for (i = 0; i < mod->num_symtab; i++) {
- ret = fn(data, symname(mod, i),
- mod, mod->symtab[i].st_value);
+ for (i = 0; i < kallsyms->num_symtab; i++) {
+ ret = fn(data, symname(kallsyms, i),
+ mod, kallsyms->symtab[i].st_value);
if (ret != 0)
return ret;
}
Patches currently in stable-queue which might be from rusty@xxxxxxxxxxxxxxx are
queue-3.14/modules-fix-longstanding-proc-kallsyms-vs-module-insertion-race.patch
--
To unsubscribe from this list: send the line "unsubscribe stable" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html