[PATCH V10 RESEND 1/2] bpf: refactor stack map trace depth calculation into helper function

[PATCH V10 RESEND 1/2] bpf: refactor stack map trace depth calculation into helper function

[Arnaud Lecomte]

Extract the duplicated maximum allowed depth computation for stack
traces stored in BPF stacks from bpf_get_stackid() and __bpf_get_stack()
into a dedicated stack_map_calculate_max_depth() helper function.

This unifies the logic for:
- The max depth computation
- Enforcing the sysctl_perf_event_max_stack limit

No functional changes for existing code paths.

Acked-by: Yonghong Song <yonghong.song@linux.dev>
Acked-by: Song Liu <song@kernel.org>
Signed-off-by: Arnaud Lecomte <contact@arnaud-lcm.com>
---
Changes in v2:
 - Removed the checking 'map_size % map_elem_size' from
   stack_map_calculate_max_depth
 - Changed stack_map_calculate_max_depth params name to be more generic

Changes in v3:
 - Changed map size param to size in max depth helper

Changes in v4:
 - Fixed indentation in max depth helper for args

Changes in v5:
 - Bound back trace_nr to num_elem in __bpf_get_stack
 - Make a copy of sysctl_perf_event_max_stack
   in stack_map_calculate_max_depth

Changes in v6:
 - Restrained max_depth computation only when required
 - Additional cleanup from Song in __bpf_get_stack

Changes in v7:
 - Removed additional cleanup from v6

Changes in v9:
 - Fixed incorrect removal of num_elem in get stack

Changes in v10:
 - Squashed 2 previous patch 1 and 2

Link to v9:
https://lore.kernel.org/all/20250912233409.74900-1-contact@arnaud-lcm.com/
---
---
 kernel/bpf/stackmap.c | 47 +++++++++++++++++++++++++++++--------------
 1 file changed, 32 insertions(+), 15 deletions(-)

diff --git a/kernel/bpf/stackmap.c b/kernel/bpf/stackmap.c
index 4d53cdd1374c..5e9ad050333c 100644
--- a/kernel/bpf/stackmap.c
+++ b/kernel/bpf/stackmap.c
@@ -42,6 +42,28 @@ static inline int stack_map_data_size(struct bpf_map *map)
 		sizeof(struct bpf_stack_build_id) : sizeof(u64);
 }
 
+/**
+ * stack_map_calculate_max_depth - Calculate maximum allowed stack trace depth
+ * @size:  Size of the buffer/map value in bytes
+ * @elem_size:  Size of each stack trace element
+ * @flags:  BPF stack trace flags (BPF_F_USER_STACK, BPF_F_USER_BUILD_ID, ...)
+ *
+ * Return: Maximum number of stack trace entries that can be safely stored
+ */
+static u32 stack_map_calculate_max_depth(u32 size, u32 elem_size, u64 flags)
+{
+	u32 skip = flags & BPF_F_SKIP_FIELD_MASK;
+	u32 max_depth;
+	u32 curr_sysctl_max_stack = READ_ONCE(sysctl_perf_event_max_stack);
+
+	max_depth = size / elem_size;
+	max_depth += skip;
+	if (max_depth > curr_sysctl_max_stack)
+		return curr_sysctl_max_stack;
+
+	return max_depth;
+}
+
 static int prealloc_elems_and_freelist(struct bpf_stack_map *smap)
 {
 	u64 elem_size = sizeof(struct stack_map_bucket) +
@@ -300,20 +322,17 @@ static long __bpf_get_stackid(struct bpf_map *map,
 BPF_CALL_3(bpf_get_stackid, struct pt_regs *, regs, struct bpf_map *, map,
 	   u64, flags)
 {
-	u32 max_depth = map->value_size / stack_map_data_size(map);
-	u32 skip = flags & BPF_F_SKIP_FIELD_MASK;
+	u32 elem_size = stack_map_data_size(map);
 	bool user = flags & BPF_F_USER_STACK;
 	struct perf_callchain_entry *trace;
 	bool kernel = !user;
+	u32 max_depth;
 
 	if (unlikely(flags & ~(BPF_F_SKIP_FIELD_MASK | BPF_F_USER_STACK |
 			       BPF_F_FAST_STACK_CMP | BPF_F_REUSE_STACKID)))
 		return -EINVAL;
 
-	max_depth += skip;
-	if (max_depth > sysctl_perf_event_max_stack)
-		max_depth = sysctl_perf_event_max_stack;
-
+	max_depth = stack_map_calculate_max_depth(map->value_size, elem_size, flags);
 	trace = get_perf_callchain(regs, kernel, user, max_depth,
 				   false, false);
 
@@ -406,7 +425,7 @@ static long __bpf_get_stack(struct pt_regs *regs, struct task_struct *task,
 			    struct perf_callchain_entry *trace_in,
 			    void *buf, u32 size, u64 flags, bool may_fault)
 {
-	u32 trace_nr, copy_len, elem_size, num_elem, max_depth;
+	u32 trace_nr, copy_len, elem_size, max_depth;
 	bool user_build_id = flags & BPF_F_USER_BUILD_ID;
 	bool crosstask = task && task != current;
 	u32 skip = flags & BPF_F_SKIP_FIELD_MASK;
@@ -438,21 +457,20 @@ static long __bpf_get_stack(struct pt_regs *regs, struct task_struct *task,
 		goto clear;
 	}
 
-	num_elem = size / elem_size;
-	max_depth = num_elem + skip;
-	if (sysctl_perf_event_max_stack < max_depth)
-		max_depth = sysctl_perf_event_max_stack;
+	max_depth = stack_map_calculate_max_depth(size, elem_size, flags);
 
 	if (may_fault)
 		rcu_read_lock(); /* need RCU for perf's callchain below */
 
-	if (trace_in)
+	if (trace_in) {
 		trace = trace_in;
-	else if (kernel && task)
+		trace->nr = min_t(u32, trace->nr, max_depth);
+	} else if (kernel && task) {
 		trace = get_callchain_entry_for_task(task, max_depth);
-	else
+	} else {
 		trace = get_perf_callchain(regs, kernel, user, max_depth,
 					   crosstask, false);
+	}
 
 	if (unlikely(!trace) || trace->nr < skip) {
 		if (may_fault)
@@ -461,7 +479,6 @@ static long __bpf_get_stack(struct pt_regs *regs, struct task_struct *task,
 	}
 
 	trace_nr = trace->nr - skip;
-	trace_nr = (trace_nr <= num_elem) ? trace_nr : num_elem;
 	copy_len = trace_nr * elem_size;
 
 	ips = trace->ip + skip;
-- 
2.47.3

[PATCH V10 RESEND 2/2] bpf: fix stackmap overflow check in __bpf_get_stackid()

[Arnaud Lecomte]

Syzkaller reported a KASAN slab-out-of-bounds write in __bpf_get_stackid()
when copying stack trace data. The issue occurs when the perf trace
 contains more stack entries than the stack map bucket can hold,
 leading to an out-of-bounds write in the bucket's data array.

Reported-by: syzbot+c9b724fbb41cf2538b7b@syzkaller.appspotmail.com
Closes: https://syzkaller.appspot.com/bug?extid=c9b724fbb41cf2538b7b
Fixes: ee2a098851bf ("bpf: Adjust BPF stack helper functions to accommodate skip > 0")
Acked-by: Yonghong Song <yonghong.song@linux.dev>
Acked-by: Song Liu <song@kernel.org>
Signed-off-by: Arnaud Lecomte <contact@arnaud-lcm.com>
---
Changes in v2:
 - Fixed max_depth names across get stack id

Changes in v4:
 - Removed unnecessary empty line in __bpf_get_stackid

Changes in v6:
 - Added back trace_len computation in __bpf_get_stackid

Changes in v7:
 - Removed usefull trace->nr assignation in bpf_get_stackid_pe
 - Added restoration of trace->nr for both kernel and user traces
   in bpf_get_stackid_pe

Changes in v9:
 - Fixed variable declarations in bpf_get_stackid_pe
 - Added the missing truncate of trace_nr in __bpf_getstackid

Changes in v10:
 - Remove not required trace->nr = nr_kernel; in bpf_get_stackid_pe

Link to v9:
https://lore.kernel.org/all/20250912233558.75076-1-contact@arnaud-lcm.com/
---
---
 kernel/bpf/stackmap.c | 15 ++++++++-------
 1 file changed, 8 insertions(+), 7 deletions(-)

diff --git a/kernel/bpf/stackmap.c b/kernel/bpf/stackmap.c
index 5e9ad050333c..2365541c81dd 100644
--- a/kernel/bpf/stackmap.c
+++ b/kernel/bpf/stackmap.c
@@ -251,8 +251,8 @@ static long __bpf_get_stackid(struct bpf_map *map,
 {
 	struct bpf_stack_map *smap = container_of(map, struct bpf_stack_map, map);
 	struct stack_map_bucket *bucket, *new_bucket, *old_bucket;
+	u32 hash, id, trace_nr, trace_len, i, max_depth;
 	u32 skip = flags & BPF_F_SKIP_FIELD_MASK;
-	u32 hash, id, trace_nr, trace_len, i;
 	bool user = flags & BPF_F_USER_STACK;
 	u64 *ips;
 	bool hash_matches;
@@ -261,7 +261,8 @@ static long __bpf_get_stackid(struct bpf_map *map,
 		/* skipping more than usable stack trace */
 		return -EFAULT;
 
-	trace_nr = trace->nr - skip;
+	max_depth = stack_map_calculate_max_depth(map->value_size, stack_map_data_size(map), flags);
+	trace_nr = min_t(u32, trace->nr - skip, max_depth - skip);
 	trace_len = trace_nr * sizeof(u64);
 	ips = trace->ip + skip;
 	hash = jhash2((u32 *)ips, trace_len / sizeof(u32), 0);
@@ -390,15 +391,11 @@ BPF_CALL_3(bpf_get_stackid_pe, struct bpf_perf_event_data_kern *, ctx,
 		return -EFAULT;
 
 	nr_kernel = count_kernel_ip(trace);
+	__u64 nr = trace->nr; /* save original */
 
 	if (kernel) {
-		__u64 nr = trace->nr;
-
 		trace->nr = nr_kernel;
 		ret = __bpf_get_stackid(map, trace, flags);
-
-		/* restore nr */
-		trace->nr = nr;
 	} else { /* user */
 		u64 skip = flags & BPF_F_SKIP_FIELD_MASK;
 
@@ -409,6 +406,10 @@ BPF_CALL_3(bpf_get_stackid_pe, struct bpf_perf_event_data_kern *, ctx,
 		flags = (flags & ~BPF_F_SKIP_FIELD_MASK) | skip;
 		ret = __bpf_get_stackid(map, trace, flags);
 	}
+
+	/* restore nr */
+	trace->nr = nr;
+
 	return ret;
 }
 
-- 
2.47.3

Re: [PATCH V10 RESEND 1/2] bpf: refactor stack map trace depth calculation into helper function

[patchwork-bot+netdevbpf]

Hello:

This series was applied to bpf/bpf-next.git (master)
by Andrii Nakryiko <andrii@kernel.org>:

On Sat, 25 Oct 2025 19:28:58 +0000 you wrote:
> Extract the duplicated maximum allowed depth computation for stack
> traces stored in BPF stacks from bpf_get_stackid() and __bpf_get_stack()
> into a dedicated stack_map_calculate_max_depth() helper function.
> 
> This unifies the logic for:
> - The max depth computation
> - Enforcing the sysctl_perf_event_max_stack limit
> 
> [...]

Here is the summary with links:
  - [V10,RESEND,1/2] bpf: refactor stack map trace depth calculation into helper function
    https://git.kernel.org/bpf/bpf-next/c/e17d62fedd10
  - [V10,RESEND,2/2] bpf: fix stackmap overflow check in __bpf_get_stackid()
    https://git.kernel.org/bpf/bpf-next/c/23f852daa4ba

You are awesome, thank you!
-- 
Deet-doot-dot, I am a bot.
https://korg.docs.kernel.org/patchwork/pwbot.html

Re: [PATCH V10 RESEND 2/2] bpf: fix stackmap overflow check in __bpf_get_stackid()

[Guenter Roeck]

Hi,

On Sat, Oct 25, 2025 at 07:29:41PM +0000, Arnaud Lecomte wrote:
> Syzkaller reported a KASAN slab-out-of-bounds write in __bpf_get_stackid()
> when copying stack trace data. The issue occurs when the perf trace
>  contains more stack entries than the stack map bucket can hold,
>  leading to an out-of-bounds write in the bucket's data array.
> 
> Reported-by: syzbot+c9b724fbb41cf2538b7b@syzkaller.appspotmail.com
> Closes: https://syzkaller.appspot.com/bug?extid=c9b724fbb41cf2538b7b
> Fixes: ee2a098851bf ("bpf: Adjust BPF stack helper functions to accommodate skip > 0")
> Acked-by: Yonghong Song <yonghong.song@linux.dev>
> Acked-by: Song Liu <song@kernel.org>
> Signed-off-by: Arnaud Lecomte <contact@arnaud-lcm.com>
> ---
> Changes in v2:
>  - Fixed max_depth names across get stack id
> 
> Changes in v4:
>  - Removed unnecessary empty line in __bpf_get_stackid
> 
> Changes in v6:
>  - Added back trace_len computation in __bpf_get_stackid
> 
> Changes in v7:
>  - Removed usefull trace->nr assignation in bpf_get_stackid_pe
>  - Added restoration of trace->nr for both kernel and user traces
>    in bpf_get_stackid_pe
> 
> Changes in v9:
>  - Fixed variable declarations in bpf_get_stackid_pe
>  - Added the missing truncate of trace_nr in __bpf_getstackid
> 
> Changes in v10:
>  - Remove not required trace->nr = nr_kernel; in bpf_get_stackid_pe
> 
> Link to v9:
> https://lore.kernel.org/all/20250912233558.75076-1-contact@arnaud-lcm.com/
> ---
> ---
>  kernel/bpf/stackmap.c | 15 ++++++++-------
>  1 file changed, 8 insertions(+), 7 deletions(-)
> 
> diff --git a/kernel/bpf/stackmap.c b/kernel/bpf/stackmap.c
> index 5e9ad050333c..2365541c81dd 100644
> --- a/kernel/bpf/stackmap.c
> +++ b/kernel/bpf/stackmap.c
> @@ -251,8 +251,8 @@ static long __bpf_get_stackid(struct bpf_map *map,
>  {
>  	struct bpf_stack_map *smap = container_of(map, struct bpf_stack_map, map);
>  	struct stack_map_bucket *bucket, *new_bucket, *old_bucket;
> +	u32 hash, id, trace_nr, trace_len, i, max_depth;
>  	u32 skip = flags & BPF_F_SKIP_FIELD_MASK;
> -	u32 hash, id, trace_nr, trace_len, i;
>  	bool user = flags & BPF_F_USER_STACK;
>  	u64 *ips;
>  	bool hash_matches;
> @@ -261,7 +261,8 @@ static long __bpf_get_stackid(struct bpf_map *map,
>  		/* skipping more than usable stack trace */
>  		return -EFAULT;
>  
> -	trace_nr = trace->nr - skip;
> +	max_depth = stack_map_calculate_max_depth(map->value_size, stack_map_data_size(map), flags);
> +	trace_nr = min_t(u32, trace->nr - skip, max_depth - skip);

I stumbled over this patch while researching a bpf related crash.
I assume that the problem it tries to solve is to prevent the
"skip > max_depth" condition.

From the context, it is guaranteed that trace->nr > skip, so we know that
trace->nr - skip is positive. If skip <= max_depth, the above then
constraints trace_nr to the difference, and there is no problem. However,
if skip > max_depth, "max_depth - skip" will be a large positive number,
effectively making trace->nr - skip unrestricted.

Is the condition "max_depth >= skip" guaranteed somewhere ? skip itself
seems to be provided by userspace, and stack_map_calculate_max_depth()
returns at most curr_sysctl_max_stack. What happens if skip is larger
than curr_sysctl_max_stack ?

My apologies for the noise if I am completely missing the point.

Thanks,
Guenter