Home » Mailing lists » Devel » [PATCH 5/5] Fix race between cat /proc/slab_allocators and rmmod
| [PATCH 5/5] Fix race between cat /proc/slab_allocators and rmmod [message #11683] |
Mon, 02 April 2007 14:58  |
Alexey Dobriyan
Messages: 195
Registered: August 2006
|
Senior Member |
|
|
Same story as with cat /proc/*/wchan race vs rmmod race, only
/proc/slab_allocators want more info than just symbol name.
Signed-off-by: Alexey Dobriyan <adobriyan@sw.ru>
---
include/linux/kallsyms.h | 6 ++++++
include/linux/module.h | 6 ++++++
kernel/kallsyms.c | 19 +++++++++++++++++++
kernel/module.c | 27 +++++++++++++++++++++++++++
mm/slab.c | 10 +++-------
5 files changed, 61 insertions(+), 7 deletions(-)
--- a/include/linux/kallsyms.h
+++ b/include/linux/kallsyms.h
@@ -23,6 +23,7 @@ const char *kallsyms_lookup(unsigned lon
char **modname, char *namebuf);
int lookup_symbol_name(unsigned long addr, char *symname);
+int lookup_symbol_attrs(unsigned long addr, unsigned long *size, unsigned long *offset, char *modname, char *name);
/* Replace "%s" in format with address, if found */
extern void __print_symbol(const char *fmt, unsigned long address);
@@ -54,6 +55,11 @@ static inline int lookup_symbol_name(uns
return -ERANGE;
}
+static inline int lookup_symbol_attrs(unsigned long addr, unsigned long *size, unsigned long *offset, char *modname, char *name)
+{
+ return -ERANGE;
+}
+
/* Stupid that this does nothing, but I didn't create this mess. */
#define __print_symbol(fmt, addr)
#endif /*CONFIG_KALLSYMS*/
--- a/include/linux/module.h
+++ b/include/linux/module.h
@@ -457,6 +457,7 @@ const char *module_address_lookup(unsign
unsigned long *offset,
char **modname);
int lookup_module_symbol_name(unsigned long addr, char *symname);
+int lookup_module_symbol_attrs(unsigned long addr, unsigned long *size, unsigned long *offset, char *modname, char *name);
/* For extable.c to search modules' exception tables. */
const struct exception_table_entry *search_module_extables(unsigned long addr);
@@ -533,6 +534,11 @@ static inline int lookup_module_symbol_n
return -ERANGE;
}
+static inline int lookup_module_symbol_attrs(unsigned long addr, unsigned long *size, unsigned long *offset, char *modname, char *name)
+{
+ return -ERANGE;
+}
+
static inline int module_get_kallsym(unsigned int symnum, unsigned long *value,
char *type, char *name,
char *module_name, int *exported)
--- a/kernel/kallsyms.c
+++ b/kernel/kallsyms.c
@@ -286,6 +286,25 @@ int lookup_symbol_name(unsigned long add
return lookup_module_symbol_name(addr, symname);
}
+int lookup_symbol_attrs(unsigned long addr, unsigned long *size,
+ unsigned long *offset, char *modname, char *name)
+{
+ name[0] = '\0';
+ name[KSYM_NAME_LEN] = '\0';
+
+ if (is_ksym_addr(addr)) {
+ unsigned long pos;
+
+ pos = get_symbol_pos(addr, size, offset);
+ /* Grab name */
+ kallsyms_expand_symbol(get_symbol_offset(pos), name);
+ modname[0] = '\0';
+ return 0;
+ }
+ /* see if it's in a module */
+ return lookup_module_symbol_attrs(addr, size, offset, modname, name);
+}
+
/* Replace "%s" in format with address, or returns -errno. */
void __print_symbol(const char *fmt, unsigned long address)
{
--- a/kernel/module.c
+++ b/kernel/module.c
@@ -2145,6 +2145,33 @@ out:
return -ERANGE;
}
+int lookup_module_symbol_attrs(unsigned long addr, unsigned long *size,
+ unsigned long *offset, char *modname, char *name)
+{
+ struct module *mod;
+
+ mutex_lock(&module_mutex);
+ list_for_each_entry(mod, &modules, list) {
+ if (within(addr, mod->module_init, mod->init_size) ||
+ within(addr, mod->module_core, mod->core_size)) {
+ const char *sym;
+
+ sym = get_ksymbol(mod, addr, size, offset);
+ if (!sym)
+ goto out;
+ if (modname)
+ strlcpy(modname, mod->name, MODULE_NAME_LEN + 1);
+ if (name)
+ strlcpy(name, sym, KSYM_NAME_LEN + 1);
+ mutex_unlock(&module_mutex);
+ return 0;
+ }
+ }
+out:
+ mutex_unlock(&module_mutex);
+ return -ERANGE;
+}
+
int module_get_kallsym(unsigned int symnum, unsigned long *value, char *type,
char *name, char *module_name, int *exported)
{
--- a/mm/slab.c
+++ b/mm/slab.c
@@ -4380,16 +4380,12 @@ static void handle_slab(unsigned long *n
static void show_symbol(struct seq_file *m, unsigned long address)
{
#ifdef CONFIG_KALLSYMS
- char *modname;
- const char *name;
unsigned long offset, size;
- char namebuf[KSYM_NAME_LEN+1];
-
- name = kallsyms_lookup(address, &size, &offset, &modname, namebuf);
+ char modname[MODULE_NAME_LEN + 1], name[KSYM_NAME_LEN + 1];
- if (name) {
+ if (lookup_symbol_attrs(address, &size, &offset, modname, name) == 0) {
seq_printf(m, "%s+%#lx/%#lx", name, offset, size);
- if (modname)
+ if (modname[0])
seq_printf(m, " [%s]", modname);
return;
}
|
|
|
|
|
|
|
|
| Re: [PATCH 5/5] Fix race between cat /proc/slab_allocators and rmmod [message #11727 is a reply to message #11683] |
Wed, 04 April 2007 00:08  |
Andrew Morton
Messages: 127
Registered: December 2005
|
Senior Member |
|
|
On Mon, 2 Apr 2007 19:03:16 +0400
Alexey Dobriyan <adobriyan@sw.ru> wrote:
> +int lookup_module_symbol_attrs(unsigned long addr, unsigned long *size,
> + unsigned long *offset, char *modname, char *name)
> +{
> + struct module *mod;
> +
> + mutex_lock(&module_mutex);
> + list_for_each_entry(mod, &modules, list) {
> + if (within(addr, mod->module_init, mod->init_size) ||
> + within(addr, mod->module_core, mod->core_size)) {
> + const char *sym;
> +
> + sym = get_ksymbol(mod, addr, size, offset);
> + if (!sym)
> + goto out;
> + if (modname)
> + strlcpy(modname, mod->name, MODULE_NAME_LEN + 1);
> + if (name)
> + strlcpy(name, sym, KSYM_NAME_LEN + 1);
> + mutex_unlock(&module_mutex);
> + return 0;
> + }
> + }
> +out:
> + mutex_unlock(&module_mutex);
> + return -ERANGE;
> +}
> +
The functions lookup_symbol_name() and lookup_symbol_attrs() are quite
general and are likely to be used for other applications in the future.
Could we please pick better names for them? The names you've chosen are
far too general-sounding - we have a lot of different types of "symbol" in
the kernel! Perhaps module_lookup_symbol_attrs() and
kallsyms_lookup_symbol() and kallsyms_lookup_attrs() or something. But
something prefixed with kallsyms_ and module_.
Also, as these general-use functions are exported to all of vmlinux, a little
bit of documentation for them would be good.
|
|
|
|
Goto Forum:
Current Time: Sun Jul 06 09:55:37 GMT 2025
Total time taken to generate the page: 0.02721 seconds
|
OpenVZ Forum
Members
Search
Help
Register
Login
Home
