2019-12-03 06:15:03

by Masami Hiramatsu

Subject: [PATCH] modules: lockdep: Suppress suspicious RCU usage warning

While running kprobe module test, find_module_all() caused
a suspicious RCU usage warning.

-----
=============================
WARNING: suspicious RCU usage
5.4.0-next-20191202+ #63 Not tainted
-----------------------------
kernel/module.c:619 RCU-list traversed in non-reader section!!

other info that might help us debug this:


rcu_scheduler_active = 2, debug_locks = 1
1 lock held by rmmod/642:
#0: ffffffff8227da80 (module_mutex){+.+.}, at: __x64_sys_delete_module+0x9a/0x230

stack backtrace:
CPU: 0 PID: 642 Comm: rmmod Not tainted 5.4.0-next-20191202+ #63
Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS rel-1.12.1-0-ga5cab58e9a3f-prebuilt.qemu.org 04/01/2014
Call Trace:
dump_stack+0x71/0xa0
find_module_all+0xc1/0xd0
__x64_sys_delete_module+0xac/0x230
? do_syscall_64+0x12/0x1f0
do_syscall_64+0x50/0x1f0
entry_SYSCALL_64_after_hwframe+0x49/0xbe
RIP: 0033:0x4b6d49
-----

This is because list_for_each_entry_rcu(modules) is called
without rcu_read_lock(). This is safe because the module_mutex
is locked.

Pass lockdep_is_held(&module_lock) to the list_for_each_entry_rcu()
to suppress this warning, This also fixes similar issue in
mod_find() and each_symbol_section().

Signed-off-by: Masami Hiramatsu <[email protected]>
---
kernel/module.c | 9 ++++++---
1 file changed, 6 insertions(+), 3 deletions(-)

diff --git a/kernel/module.c b/kernel/module.c
index cb6250be6ee9..38e5c6a7451b 100644
--- a/kernel/module.c
+++ b/kernel/module.c
@@ -214,7 +214,8 @@ static struct module *mod_find(unsigned long addr)
{
struct module *mod;

- list_for_each_entry_rcu(mod, &modules, list) {
+ list_for_each_entry_rcu(mod, &modules, list,
+ lockdep_is_held(&module_mutex)) {
if (within_module(addr, mod))
return mod;
}
@@ -448,7 +449,8 @@ bool each_symbol_section(bool (*fn)(const struct symsearch *arr,
if (each_symbol_in_section(arr, ARRAY_SIZE(arr), NULL, fn, data))
return true;

- list_for_each_entry_rcu(mod, &modules, list) {
+ list_for_each_entry_rcu(mod, &modules, list,
+ lockdep_is_held(&module_mutex)) {
struct symsearch arr[] = {
{ mod->syms, mod->syms + mod->num_syms, mod->crcs,
NOT_GPL_ONLY, false },
@@ -616,7 +618,8 @@ static struct module *find_module_all(const char *name, size_t len,

module_assert_mutex_or_preempt();

- list_for_each_entry_rcu(mod, &modules, list) {
+ list_for_each_entry_rcu(mod, &modules, list,
+ lockdep_is_held(&module_mutex)) {
if (!even_unformed && mod->state == MODULE_STATE_UNFORMED)
continue;
if (strlen(mod->name) == len && !memcmp(mod->name, name, len))

2019-12-05 19:45:05

by Jessica Yu

Subject: Re: [PATCH] modules: lockdep: Suppress suspicious RCU usage warning

+++ Masami Hiramatsu [03/12/19 15:14 +0900]:
>While running kprobe module test, find_module_all() caused
>a suspicious RCU usage warning.
>
>-----
> =============================
> WARNING: suspicious RCU usage
> 5.4.0-next-20191202+ #63 Not tainted
> -----------------------------
> kernel/module.c:619 RCU-list traversed in non-reader section!!
>
> other info that might help us debug this:
>
>
> rcu_scheduler_active = 2, debug_locks = 1
> 1 lock held by rmmod/642:
> #0: ffffffff8227da80 (module_mutex){+.+.}, at: __x64_sys_delete_module+0x9a/0x230
>
> stack backtrace:
> CPU: 0 PID: 642 Comm: rmmod Not tainted 5.4.0-next-20191202+ #63
> Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS rel-1.12.1-0-ga5cab58e9a3f-prebuilt.qemu.org 04/01/2014
> Call Trace:
> dump_stack+0x71/0xa0
> find_module_all+0xc1/0xd0
> __x64_sys_delete_module+0xac/0x230
> ? do_syscall_64+0x12/0x1f0
> do_syscall_64+0x50/0x1f0
> entry_SYSCALL_64_after_hwframe+0x49/0xbe
> RIP: 0033:0x4b6d49
>-----
>
>This is because list_for_each_entry_rcu(modules) is called
>without rcu_read_lock(). This is safe because the module_mutex
>is locked.
>
>Pass lockdep_is_held(&module_lock) to the list_for_each_entry_rcu()

s/module_lock/module_mutex/, but you don't have to respin the patch
just for this.

>to suppress this warning, This also fixes similar issue in
>mod_find() and each_symbol_section().
>
>Signed-off-by: Masami Hiramatsu <[email protected]>

Thanks Masami! This looks good. I'll queue this up shortly after the
merge window.

Jessica

>---
> kernel/module.c | 9 ++++++---
> 1 file changed, 6 insertions(+), 3 deletions(-)
>
>diff --git a/kernel/module.c b/kernel/module.c
>index cb6250be6ee9..38e5c6a7451b 100644
>--- a/kernel/module.c
>+++ b/kernel/module.c
>@@ -214,7 +214,8 @@ static struct module *mod_find(unsigned long addr)
> {
> struct module *mod;
>
>- list_for_each_entry_rcu(mod, &modules, list) {
>+ list_for_each_entry_rcu(mod, &modules, list,
>+ lockdep_is_held(&module_mutex)) {
> if (within_module(addr, mod))
> return mod;
> }
>@@ -448,7 +449,8 @@ bool each_symbol_section(bool (*fn)(const struct symsearch *arr,
> if (each_symbol_in_section(arr, ARRAY_SIZE(arr), NULL, fn, data))
> return true;
>
>- list_for_each_entry_rcu(mod, &modules, list) {
>+ list_for_each_entry_rcu(mod, &modules, list,
>+ lockdep_is_held(&module_mutex)) {
> struct symsearch arr[] = {
> { mod->syms, mod->syms + mod->num_syms, mod->crcs,
> NOT_GPL_ONLY, false },
>@@ -616,7 +618,8 @@ static struct module *find_module_all(const char *name, size_t len,
>
> module_assert_mutex_or_preempt();
>
>- list_for_each_entry_rcu(mod, &modules, list) {
>+ list_for_each_entry_rcu(mod, &modules, list,
>+ lockdep_is_held(&module_mutex)) {
> if (!even_unformed && mod->state == MODULE_STATE_UNFORMED)
> continue;
> if (strlen(mod->name) == len && !memcmp(mod->name, name, len))
>

2019-12-06 01:22:37

by Masami Hiramatsu

Subject: Re: [PATCH] modules: lockdep: Suppress suspicious RCU usage warning

On Thu, 5 Dec 2019 20:17:59 +0100
Jessica Yu <[email protected]> wrote:

> +++ Masami Hiramatsu [03/12/19 15:14 +0900]:
> >While running kprobe module test, find_module_all() caused
> >a suspicious RCU usage warning.
> >
> >-----
> > =============================
> > WARNING: suspicious RCU usage
> > 5.4.0-next-20191202+ #63 Not tainted
> > -----------------------------
> > kernel/module.c:619 RCU-list traversed in non-reader section!!
> >
> > other info that might help us debug this:
> >
> >
> > rcu_scheduler_active = 2, debug_locks = 1
> > 1 lock held by rmmod/642:
> > #0: ffffffff8227da80 (module_mutex){+.+.}, at: __x64_sys_delete_module+0x9a/0x230
> >
> > stack backtrace:
> > CPU: 0 PID: 642 Comm: rmmod Not tainted 5.4.0-next-20191202+ #63
> > Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS rel-1.12.1-0-ga5cab58e9a3f-prebuilt.qemu.org 04/01/2014
> > Call Trace:
> > dump_stack+0x71/0xa0
> > find_module_all+0xc1/0xd0
> > __x64_sys_delete_module+0xac/0x230
> > ? do_syscall_64+0x12/0x1f0
> > do_syscall_64+0x50/0x1f0
> > entry_SYSCALL_64_after_hwframe+0x49/0xbe
> > RIP: 0033:0x4b6d49
> >-----
> >
> >This is because list_for_each_entry_rcu(modules) is called
> >without rcu_read_lock(). This is safe because the module_mutex
> >is locked.
> >
> >Pass lockdep_is_held(&module_lock) to the list_for_each_entry_rcu()
>
> s/module_lock/module_mutex/, but you don't have to respin the patch
> just for this.

Oops.

>
> >to suppress this warning, This also fixes similar issue in
> >mod_find() and each_symbol_section().
> >
> >Signed-off-by: Masami Hiramatsu <[email protected]>
>
> Thanks Masami! This looks good. I'll queue this up shortly after the
> merge window.

Thank you for merging :)


>
> Jessica
>
> >---
> > kernel/module.c | 9 ++++++---
> > 1 file changed, 6 insertions(+), 3 deletions(-)
> >
> >diff --git a/kernel/module.c b/kernel/module.c
> >index cb6250be6ee9..38e5c6a7451b 100644
> >--- a/kernel/module.c
> >+++ b/kernel/module.c
> >@@ -214,7 +214,8 @@ static struct module *mod_find(unsigned long addr)
> > {
> > struct module *mod;
> >
> >- list_for_each_entry_rcu(mod, &modules, list) {
> >+ list_for_each_entry_rcu(mod, &modules, list,
> >+ lockdep_is_held(&module_mutex)) {
> > if (within_module(addr, mod))
> > return mod;
> > }
> >@@ -448,7 +449,8 @@ bool each_symbol_section(bool (*fn)(const struct symsearch *arr,
> > if (each_symbol_in_section(arr, ARRAY_SIZE(arr), NULL, fn, data))
> > return true;
> >
> >- list_for_each_entry_rcu(mod, &modules, list) {
> >+ list_for_each_entry_rcu(mod, &modules, list,
> >+ lockdep_is_held(&module_mutex)) {
> > struct symsearch arr[] = {
> > { mod->syms, mod->syms + mod->num_syms, mod->crcs,
> > NOT_GPL_ONLY, false },
> >@@ -616,7 +618,8 @@ static struct module *find_module_all(const char *name, size_t len,
> >
> > module_assert_mutex_or_preempt();
> >
> >- list_for_each_entry_rcu(mod, &modules, list) {
> >+ list_for_each_entry_rcu(mod, &modules, list,
> >+ lockdep_is_held(&module_mutex)) {
> > if (!even_unformed && mod->state == MODULE_STATE_UNFORMED)
> > continue;
> > if (strlen(mod->name) == len && !memcmp(mod->name, name, len))
> >


--
Masami Hiramatsu <[email protected]>