[PATCH v2] fork: initialize function graph state before copy_exec_state()

[PATCH v2] fork: initialize function graph state before copy_exec_state()

[Jérémy Jean]

dup_task_struct() copies the parent's task_struct, including ret_stack.
ftrace_graph_init_task() clears the copied function graph state, but it
currently runs after copy_exec_state().

For non-CLONE_VM forks, copy_exec_state() allocates a new task_exec_state.
If that allocation fails, copy_process() reaches bad_fork_free and
free_task() calls ftrace_graph_exit_task(). Since the child still carries
the parent's ret_stack pointer, the unwind frees the parent's active
function graph return stack. The parent subsequently accesses freed memory
from function_graph_enter_regs().

KASAN reports:

[   22.190920] ==================================================================
[   22.195899] BUG: KASAN: slab-use-after-free in function_graph_enter_regs+0xa76/0xb90
[   22.200747] Write of size 8 at addr ff110000054dc0a8 by task repro/1
[   22.205134]
[   22.210770] CPU: 0 UID: 0 PID: 1 Comm: repro Not tainted 7.2.0-07732-g9328b3b03bdc-dirty #3 PREEMPT(lazy)
[   22.212576] Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.16.3-debian-1.16.3-2 04/01/2014
[   22.213750] Call Trace:
[   22.215271]  <TASK>
[   22.216242]  ? ftrace_stub_direct_tramp+0x10/0x10
[   22.217774]  dump_stack_lvl+0x4e/0x70
[   22.220531]  print_report+0x157/0x4b4
[   22.223202]  ? fixup_red_left+0x9/0x30
[   22.224407]  ? complete_report_info+0x83/0x110
[   22.226679]  ? function_graph_enter_regs+0xa76/0xb90
[   22.228084]  kasan_report+0xce/0x100
[   22.230109]  ? function_graph_enter_regs+0xa76/0xb90
[   22.232860]  ? stack_trace_save+0x4/0xd0
[   22.234156]  function_graph_enter_regs+0xa76/0xb90
[   22.236090]  ? kasan_save_stack+0x30/0x50
[   22.237752]  ? __pfx_function_graph_enter_regs+0x10/0x10
[   22.238694]  ? ring_buffer_lock_reserve+0x345/0xf80
[   22.239628]  ? stack_trace_save+0x4/0xd0
[   22.242121]  ? stack_trace_save+0x4/0xd0
[   22.243588]  ftrace_graph_func+0xda/0x160
[   22.245362]  ? ftrace_stub_direct_tramp+0x10/0x10
[   22.246520]  0xffffffffa0000095
[   22.250528]  ? stack_trace_save+0x9/0xd0
[   22.251757]  ? ring_buffer_unlock_commit+0x11d/0x5c0
[   22.253152]  stack_trace_save+0x9/0xd0
[   22.254264]  kasan_save_stack+0x30/0x50
[   22.273631]  kasan_save_track+0x14/0x30
[   22.276763]  kasan_save_free_info+0x3b/0x70
[   22.278296]  __kasan_slab_free+0x43/0x70
[   22.280157]  kmem_cache_free+0xbf/0x3b0
[   22.282963]  ? ftrace_stub_direct_tramp+0x10/0x10
[   22.284001]  free_task+0xa2/0x160
[   22.285699]  ? ftrace_stub_direct_tramp+0x10/0x10
[   22.286752]  copy_process+0x2aae/0x7bc0

Initialize the child function graph state immediately after
dup_task_struct(), before the first fallible operation.

Fixes: 6b1c66c9cca9 ("exec_state: relocate dumpable information")
Cc: stable@vger.kernel.org
Assisted-by: Codex:gpt-5
Signed-off-by: Jérémy Jean <Jeremy.Jean@oss.cyber.gouv.fr>
---
Changes in v2:
- Add the KASAN report excerpt to the commit message.
- Add a comment documenting why ftrace_graph_init_task() must run before
  fallible initialization.

v1: https://lore.kernel.org/all/20260821102207.3626491-2-Jeremy.Jean@oss.cyber.gouv.fr/

 kernel/fork.c | 7 +++++--
 1 file changed, 5 insertions(+), 2 deletions(-)

diff --git a/kernel/fork.c b/kernel/fork.c
index 1e68404bd773..0d1ad92e2d33 100644
--- a/kernel/fork.c
+++ b/kernel/fork.c
@@ -2139,6 +2139,11 @@ __latent_entropy struct task_struct *copy_process(
 	p = dup_task_struct(current, node);
 	if (!p)
 		goto fork_out;
+	/*
+	 * Must run before the first fallible op, so error paths never
+	 * free the parent's ret_stack.
+	 */
+	ftrace_graph_init_task(p);
 	retval = copy_exec_state(clone_flags, p);
 	if (retval)
 		goto bad_fork_free;
@@ -2165,8 +2170,6 @@ __latent_entropy struct task_struct *copy_process(
 	 */
 	p->clear_child_tid = (clone_flags & CLONE_CHILD_CLEARTID) ? args->child_tid : NULL;
 
-	ftrace_graph_init_task(p);
-
 	rt_mutex_init_task(p);
 	raw_spin_lock_init(&p->blocked_lock);
 
-- 
2.47.3

Re: [PATCH v2] fork: initialize function graph state before copy_exec_state()

[Bradley Morgan]

On 22 August 2026 08:49:27 BST, Jérémy Jean
<Jeremy.Jean@oss.cyber.gouv.fr> wrote:
>From: Jérémy Jean <Jeremy.Jean@oss.cyber.gouv.fr>
>
>KASAN reports:
>
>[   22.195899] BUG: KASAN: slab-use-after-free in
>function_graph_enter_regs+0xa76/0xb90
>[   22.200747] Write of size 8 at addr ff110000054dc0a8 by task repro/1

Perfect!

>[   22.278296]  __kasan_slab_free+0x43/0x70
>[   22.280157]  kmem_cache_free+0xbf/0x3b0
>[   22.282963]  ? ftrace_stub_direct_tramp+0x10/0x10
>[   22.284001]  free_task+0xa2/0x160
>[   22.285699]  ? ftrace_stub_direct_tramp+0x10/0x10
>[   22.286752]  copy_process+0x2aae/0x7bc0

And the free stack is exactly the mechanism from the changelog,
free_task() under copy_process()'s error path dropping the stack, then
the parent writes into it afterwards. Stable folks will love this.

>diff --git a/kernel/fork.c b/kernel/fork.c
>+ /*
>+  * Must run before the first fallible op, so error paths never
>+  * free the parent's ret_stack.
>+  */
>+ ftrace_graph_init_task(p);

And the comment landed too. Nice!

Reviewed-by: Bradley Morgan <include@grrlz.net>

Thanks!

Re: [PATCH v2] fork: initialize function graph state before copy_exec_state()

[Bradley Morgan]

On 22 August 2026 09:49:27 BST, "Jérémy Jean"
<Jeremy.Jean@oss.cyber.gouv.fr> wrote:
>dup_task_struct() copies the parent's task_struct, including ret_stack.
>ftrace_graph_init_task() clears the copied function graph state, but it
>currently runs after copy_exec_state().
>
>For non-CLONE_VM forks, copy_exec_state() allocates a new task_exec_state.
>If that allocation fails, copy_process() reaches bad_fork_free and
>free_task() calls ftrace_graph_exit_task(). Since the child still carries
>the parent's ret_stack pointer, the unwind frees the parent's active
>function graph return stack. The parent subsequently accesses freed memory
>from function_graph_enter_regs().
>

Wow! Real life use case?


>KASAN reports:
>
>[   22.190920]
>==================================================================
>[   22.195899] BUG: KASAN: slab-use-after-free in
>function_graph_enter_regs+0xa76/0xb90
>[   22.200747] Write of size 8 at addr ff110000054dc0a8 by task repro/1
>[   22.205134]
>[   22.210770] CPU: 0 UID: 0 PID: 1 Comm: repro Not tainted
>7.2.0-07732-g9328b3b03bdc-dirty #3 PREEMPT(lazy)
>[   22.212576] Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS
>1.16.3-debian-1.16.3-2 04/01/2014
>[   22.213750] Call Trace:
>[   22.215271]  <TASK>
>[   22.216242]  ? ftrace_stub_direct_tramp+0x10/0x10
>[   22.217774]  dump_stack_lvl+0x4e/0x70
>[   22.220531]  print_report+0x157/0x4b4
>[   22.223202]  ? fixup_red_left+0x9/0x30
>[   22.224407]  ? complete_report_info+0x83/0x110
>[   22.226679]  ? function_graph_enter_regs+0xa76/0xb90
>[   22.228084]  kasan_report+0xce/0x100
>[   22.230109]  ? function_graph_enter_regs+0xa76/0xb90
>[   22.232860]  ? stack_trace_save+0x4/0xd0
>[   22.234156]  function_graph_enter_regs+0xa76/0xb90
>[   22.236090]  ? kasan_save_stack+0x30/0x50
>[   22.237752]  ? __pfx_function_graph_enter_regs+0x10/0x10
>[   22.238694]  ? ring_buffer_lock_reserve+0x345/0xf80
>[   22.239628]  ? stack_trace_save+0x4/0xd0
>[   22.242121]  ? stack_trace_save+0x4/0xd0
>[   22.243588]  ftrace_graph_func+0xda/0x160
>[   22.245362]  ? ftrace_stub_direct_tramp+0x10/0x10
>[   22.246520]  0xffffffffa0000095
>[   22.250528]  ? stack_trace_save+0x9/0xd0
>[   22.251757]  ? ring_buffer_unlock_commit+0x11d/0x5c0
>[   22.253152]  stack_trace_save+0x9/0xd0
>[   22.254264]  kasan_save_stack+0x30/0x50
>[   22.273631]  kasan_save_track+0x14/0x30
>[   22.276763]  kasan_save_free_info+0x3b/0x70
>[   22.278296]  __kasan_slab_free+0x43/0x70
>[   22.280157]  kmem_cache_free+0xbf/0x3b0
>[   22.282963]  ? ftrace_stub_direct_tramp+0x10/0x10
>[   22.284001]  free_task+0xa2/0x160
>[   22.285699]  ? ftrace_stub_direct_tramp+0x10/0x10
>[   22.286752]  copy_process+0x2aae/0x7bc0
>

I see.

>Initialize the child function graph state immediately after
>dup_task_struct(), before the first fallible operation.
>
>Fixes: 6b1c66c9cca9 ("exec_state: relocate dumpable information")

Ack.

>Cc: stable@vger.kernel.org
>Assisted-by: Codex:gpt-5

Good, declared AI.

Reviewed-by: Bradley Morgan <include@grrlz.net>


>Signed-off-by: Jérémy Jean <Jeremy.Jean@oss.cyber.gouv.fr>
>---
>Changes in v2:
>- Add the KASAN report excerpt to the commit message.
>- Add a comment documenting why ftrace_graph_init_task() must run before
>  fallible initialization.
>
>v1: https://lore.kernel.org/all/20260821102207.3626491-2-Jeremy.Jean@oss.cyber.gouv.fr/
>
> kernel/fork.c | 7 +++++--
> 1 file changed, 5 insertions(+), 2 deletions(-)
>
>diff --git a/kernel/fork.c b/kernel/fork.c
>index 1e68404bd773..0d1ad92e2d33 100644
>--- a/kernel/fork.c
>+++ b/kernel/fork.c
>@@ -2139,6 +2139,11 @@ __latent_entropy struct task_struct *copy_process(
> 	p = dup_task_struct(current, node);
> 	if (!p)
> 		goto fork_out;
>+	/*
>+	 * Must run before the first fallible op, so error paths never
>+	 * free the parent's ret_stack.
>+	 */

Used my suggestion, good.

↓ good fix.

>+	ftrace_graph_init_task(p);
> 	retval = copy_exec_state(clone_flags, p);
> 	if (retval)
> 		goto bad_fork_free;
>@@ -2165,8 +2170,6 @@ __latent_entropy struct task_struct *copy_process(
> 	 */
> 	p->clear_child_tid = (clone_flags & CLONE_CHILD_CLEARTID) ? args->child_tid : NULL;
> 
>-	ftrace_graph_init_task(p);
>-
> 	rt_mutex_init_task(p);
> 	raw_spin_lock_init(&p->blocked_lock);
> 
>

Thanks!

Re: [PATCH v2] fork: initialize function graph state before copy_exec_state()

[Steven Rostedt]

On Sat, 22 Aug 2026 08:49:27 +0000
Jérémy Jean <Jeremy.Jean@oss.cyber.gouv.fr> wrote:

> dup_task_struct() copies the parent's task_struct, including ret_stack.
> ftrace_graph_init_task() clears the copied function graph state, but it
> currently runs after copy_exec_state().
> 
> For non-CLONE_VM forks, copy_exec_state() allocates a new task_exec_state.
> If that allocation fails, copy_process() reaches bad_fork_free and
> free_task() calls ftrace_graph_exit_task(). Since the child still carries
> the parent's ret_stack pointer, the unwind frees the parent's active
> function graph return stack. The parent subsequently accesses freed memory
> from function_graph_enter_regs().
> 

As this effects function_graph tracing, please resend and Cc:
linux-trace-kernel@vger.kernel.org so that it gets to the proper
patchwork. It will need to go through my tree so that it gets the
proper testing.

Thanks,

-- Steve

[PATCH v2] fork: initialize function graph state before copy_exec_state()

[Jérémy Jean]

dup_task_struct() copies the parent's task_struct, including ret_stack.
ftrace_graph_init_task() clears the copied function graph state, but it
currently runs after copy_exec_state().

For non-CLONE_VM forks, copy_exec_state() allocates a new task_exec_state.
If that allocation fails, copy_process() reaches bad_fork_free and
free_task() calls ftrace_graph_exit_task(). Since the child still carries
the parent's ret_stack pointer, the unwind frees the parent's active
function graph return stack. The parent subsequently accesses freed memory
from function_graph_enter_regs().

KASAN reports:

[   22.190920] ==================================================================
[   22.195899] BUG: KASAN: slab-use-after-free in function_graph_enter_regs+0xa76/0xb90
[   22.200747] Write of size 8 at addr ff110000054dc0a8 by task repro/1
[   22.205134]
[   22.210770] CPU: 0 UID: 0 PID: 1 Comm: repro Not tainted 7.2.0-07732-g9328b3b03bdc-dirty #3 PREEMPT(lazy)
[   22.212576] Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.16.3-debian-1.16.3-2 04/01/2014
[   22.213750] Call Trace:
[   22.215271]  <TASK>
[   22.216242]  ? ftrace_stub_direct_tramp+0x10/0x10
[   22.217774]  dump_stack_lvl+0x4e/0x70
[   22.220531]  print_report+0x157/0x4b4
[   22.223202]  ? fixup_red_left+0x9/0x30
[   22.224407]  ? complete_report_info+0x83/0x110
[   22.226679]  ? function_graph_enter_regs+0xa76/0xb90
[   22.228084]  kasan_report+0xce/0x100
[   22.230109]  ? function_graph_enter_regs+0xa76/0xb90
[   22.232860]  ? stack_trace_save+0x4/0xd0
[   22.234156]  function_graph_enter_regs+0xa76/0xb90
[   22.236090]  ? kasan_save_stack+0x30/0x50
[   22.237752]  ? __pfx_function_graph_enter_regs+0x10/0x10
[   22.238694]  ? ring_buffer_lock_reserve+0x345/0xf80
[   22.239628]  ? stack_trace_save+0x4/0xd0
[   22.242121]  ? stack_trace_save+0x4/0xd0
[   22.243588]  ftrace_graph_func+0xda/0x160
[   22.245362]  ? ftrace_stub_direct_tramp+0x10/0x10
[   22.246520]  0xffffffffa0000095
[   22.250528]  ? stack_trace_save+0x9/0xd0
[   22.251757]  ? ring_buffer_unlock_commit+0x11d/0x5c0
[   22.253152]  stack_trace_save+0x9/0xd0
[   22.254264]  kasan_save_stack+0x30/0x50
[   22.273631]  kasan_save_track+0x14/0x30
[   22.276763]  kasan_save_free_info+0x3b/0x70
[   22.278296]  __kasan_slab_free+0x43/0x70
[   22.280157]  kmem_cache_free+0xbf/0x3b0
[   22.282963]  ? ftrace_stub_direct_tramp+0x10/0x10
[   22.284001]  free_task+0xa2/0x160
[   22.285699]  ? ftrace_stub_direct_tramp+0x10/0x10
[   22.286752]  copy_process+0x2aae/0x7bc0

Initialize the child function graph state immediately after
dup_task_struct(), before the first fallible operation.

Fixes: 6b1c66c9cca9 ("exec_state: relocate dumpable information")
Cc: stable@vger.kernel.org
Assisted-by: Codex:gpt-5
Signed-off-by: Jérémy Jean <Jeremy.Jean@oss.cyber.gouv.fr>
---
Resending v2 as requested by adding Cc: linux-trace-kernel@vger.kernel.org

Changes in v2:
- Add the KASAN report excerpt to the commit message.
- Add a comment documenting why ftrace_graph_init_task() must run before
  fallible initialization.

v1: https://lore.kernel.org/all/20260821102207.3626491-2-Jeremy.Jean@oss.cyber.gouv.fr/

 kernel/fork.c | 7 +++++--
 1 file changed, 5 insertions(+), 2 deletions(-)

diff --git a/kernel/fork.c b/kernel/fork.c
index 1e68404bd773..0d1ad92e2d33 100644
--- a/kernel/fork.c
+++ b/kernel/fork.c
@@ -2139,6 +2139,11 @@ __latent_entropy struct task_struct *copy_process(
 	p = dup_task_struct(current, node);
 	if (!p)
 		goto fork_out;
+	/*
+	 * Must run before the first fallible op, so error paths never
+	 * free the parent's ret_stack.
+	 */
+	ftrace_graph_init_task(p);
 	retval = copy_exec_state(clone_flags, p);
 	if (retval)
 		goto bad_fork_free;
@@ -2165,8 +2170,6 @@ __latent_entropy struct task_struct *copy_process(
 	 */
 	p->clear_child_tid = (clone_flags & CLONE_CHILD_CLEARTID) ? args->child_tid : NULL;
 
-	ftrace_graph_init_task(p);
-
 	rt_mutex_init_task(p);
 	raw_spin_lock_init(&p->blocked_lock);
 
-- 
2.47.3

Re: [PATCH v2] fork: initialize function graph state before copy_exec_state()

[Jérémy Jean]

On 2026-08-22 17:22, Steven Rostedt wrote:
> On Sat, 22 Aug 2026 08:49:27 +0000
> Jérémy Jean <Jeremy.Jean@oss.cyber.gouv.fr> wrote:
> 
>> dup_task_struct() copies the parent's task_struct, including 
>> ret_stack.
>> ftrace_graph_init_task() clears the copied function graph state, but 
>> it
>> currently runs after copy_exec_state().
>> 
>> For non-CLONE_VM forks, copy_exec_state() allocates a new 
>> task_exec_state.
>> If that allocation fails, copy_process() reaches bad_fork_free and
>> free_task() calls ftrace_graph_exit_task(). Since the child still 
>> carries
>> the parent's ret_stack pointer, the unwind frees the parent's active
>> function graph return stack. The parent subsequently accesses freed 
>> memory
>> from function_graph_enter_regs().
>> 
> 
> As this effects function_graph tracing, please resend and Cc:
> linux-trace-kernel@vger.kernel.org so that it gets to the proper
> patchwork. It will need to go through my tree so that it gets the
> proper testing.

Hello Steve,

Done. Thanks for the feedback.
Let me know if I can help more.

Regards,
Jérémy