Skip to contentSkip to conversation

tools/perf: Fix out of bound access to affinity "sched_cpus"

PATCH In review Athira Rajeev 2 participants 3 messages Sep 5, 2022 – Sep 6, 2022 perf events

lore
Share & tools
lore
#0 lore raw

PATCH v2 1/2 tools/perf: Fix out of bound access to affinity "sched_cpus"

The affinity code in "affinity_set" function access array named "sched_cpus". The size for this array is allocated in affinity_setup function which is nothing but value from get_cpu_set_size. This is used to contain the cpumask value for each cpu. While setting bit for each cpu, it calls "set_bit" function which access index in sched_cpus array. If we provide a command-line option to -C which is more than the number of CPU's present in the system, the set_bit could access an array member which is out-of the array size. This is because currently, there is no boundary check for the CPU. This will result in seg fault:

<<>>
./perf stat -C 12323431 ls
Perf can support 2048 CPUs. Consider raising MAX_NR_CPUS Segmentation fault (core dumped)
<<>>

Fix this by adding boundary check for the array.

After the fix from powerpc system:

<<>>
./perf stat -C 12323431 ls 1>out
Perf can support 2048 CPUs. Consider raising MAX_NR_CPUS

Performance counter stats for 'CPU(s) 12323431':

<not supported> msec cpu-clock

   <not supported>      context-switches
   <not supported>      cpu-migrations
   <not supported>      page-faults
   <not supported>      cycles
   <not supported>      instructions
   <not supported>      branches
   <not supported>      branch-misses
       0.001192373 seconds time elapsed

<<>>

---
Changelog:
From v1 -> v2:
Addressed review comment from Jiri Olsa by changing condition check to directly use "cpu_set_size * 8" for comparing with the cpu number.

 tools/perf/util/affinity.c | 8 +++++++-
 1 file changed, 7 insertions(+), 1 deletion(-)
tools/perf/util/affinity.c +7 -1
index 4d216c0dc425..4ee96b3c755b 100644
@@ -49,8 +49,14 @@ void affinity__set(struct affinity *a, int cpu)
 {
 	int cpu_set_size = get_cpu_set_size();
 
-	if (cpu == -1)
+	/*
+	 * Return:
+	 * - if cpu is -1
+	 * - restrict out of bound access to sched_cpus
+	 */
+	if (cpu == -1 || ((cpu >= (cpu_set_size * 8))))
 		return;
+
 	a->changed = true;
 	set_bit(cpu, a->sched_cpus);
 	/*
Signature · 1 line
2.35.1
#1 lore raw

PATCH v2 2/2 tools/perf: Fix out of bound access to cpu mask array

The cpu mask init code in "record__mmap_cpu_mask_init" function access "bits" array part of "struct mmap_cpu_mask". The size of this array is the value from cpu__max_cpu().cpu. This array is used to contain the cpumask value for each cpu. While setting bit for each cpu, it calls "set_bit" function which access index in "bits" array. If we provide a command line option to -C which is greater than the number of CPU's present in the system, the set_bit could access an array member which is out-of the array size. This is because currently, there is no boundary check for the CPU. This will result in seg fault:

<<>>
./perf record -C 12341234 ls
Perf can support 2048 CPUs. Consider raising MAX_NR_CPUS Segmentation fault (core dumped)
<<>>

Debugging with gdb, points to function flow as below:

<<>>
set_bit
record__mmap_cpu_mask_init
record__init_thread_default_masks
record__init_thread_masks
cmd_record
<<>>

Fix this by adding boundary check for the array.

After the patch:
<<>>
./perf record -C 12341234 ls
Perf can support 2048 CPUs. Consider raising MAX_NR_CPUS Failed to initialize parallel data streaming masks <<>>

With this fix, if -C is given a non-exsiting CPU, perf record will fail with:

<<>>
./perf record -C 50 ls
Failed to initialize parallel data streaming masks <<>>

---
 tools/perf/builtin-record.c | 26 ++++++++++++++++++++------
 1 file changed, 20 insertions(+), 6 deletions(-)
tools/perf/builtin-record.c +20 -6
index 4713f0f3a6cf..09b68d76bbdc 100644
@@ -3358,16 +3358,22 @@ static struct option __record_options[] = {
 
 struct option *record_options = __record_options;
 
-static void record__mmap_cpu_mask_init(struct mmap_cpu_mask *mask, struct perf_cpu_map *cpus)
+static int record__mmap_cpu_mask_init(struct mmap_cpu_mask *mask, struct perf_cpu_map *cpus)
 {
 	struct perf_cpu cpu;
 	int idx;
 
 	if (cpu_map__is_dummy(cpus))
-		return;
+		return 0;
 
-	perf_cpu_map__for_each_cpu(cpu, idx, cpus)
+	perf_cpu_map__for_each_cpu(cpu, idx, cpus) {
+		/* Return ENODEV is input cpu is greater than max cpu */
+		if ((unsigned long)cpu.cpu > mask->nbits)
+			return -ENODEV;
 		set_bit(cpu.cpu, mask->bits);
+	}
+
+	return 0;
 }
 
 static int record__mmap_cpu_mask_init_spec(struct mmap_cpu_mask *mask, const char *mask_spec)
@@ -3379,7 +3385,9 @@ static int record__mmap_cpu_mask_init_spec(struct mmap_cpu_mask *mask, const cha
 		return -ENOMEM;
 
 	bitmap_zero(mask->bits, mask->nbits);
-	record__mmap_cpu_mask_init(mask, cpus);
+	if (record__mmap_cpu_mask_init(mask, cpus))
+		return -ENODEV;
+
 	perf_cpu_map__put(cpus);
 
 	return 0;
@@ -3461,7 +3469,12 @@ static int record__init_thread_masks_spec(struct record *rec, struct perf_cpu_ma
 		pr_err("Failed to allocate CPUs mask\n");
 		return ret;
 	}
-	record__mmap_cpu_mask_init(&cpus_mask, cpus);
+
+	ret = record__mmap_cpu_mask_init(&cpus_mask, cpus);
+	if (ret) {
+		pr_err("Failed to init cpu mask\n");
+		goto out_free_cpu_mask;
+	}
 
 	ret = record__thread_mask_alloc(&full_mask, cpu__max_cpu().cpu);
 	if (ret) {
@@ -3702,7 +3715,8 @@ static int record__init_thread_default_masks(struct record *rec, struct perf_cpu
 	if (ret)
 		return ret;
 
-	record__mmap_cpu_mask_init(&rec->thread_masks->maps, cpus);
+	if (record__mmap_cpu_mask_init(&rec->thread_masks->maps, cpus))
+		return -ENODEV;
 
 	rec->nr_threads = 1;
 
Signature · 1 line
2.35.1
#2 lore raw
In reply to Athira Rajeev

Em Mon, Sep 05, 2022 at 07:49:28PM +0530, Athira Rajeev escreveu:

Quoted 11 lines

The affinity code in "affinity_set" function access array named "sched_cpus". The size for this array is allocated in affinity_setup function which is nothing but value from get_cpu_set_size. This is used to contain the cpumask value for each cpu. While setting bit for each cpu, it calls "set_bit" function which access index in sched_cpus array. If we provide a command-line option to -C which is more than the number of CPU's present in the system, the set_bit could access an array member which is out-of the array size. This is because currently, there is no boundary check for the CPU. This will result in seg fault:

Thanks, tested, reproduced the problem before, and the fix after, applied.

- Arnaldo

Quoted 63 lines

<<>>
./perf stat -C 12323431 ls
Perf can support 2048 CPUs. Consider raising MAX_NR_CPUS Segmentation fault (core dumped)
<<>>

Fix this by adding boundary check for the array.

After the fix from powerpc system:

<<>>
./perf stat -C 12323431 ls 1>out
Perf can support 2048 CPUs. Consider raising MAX_NR_CPUS

Performance counter stats for 'CPU(s) 12323431':

<not supported> msec cpu-clock

   <not supported>      context-switches
   <not supported>      cpu-migrations
   <not supported>      page-faults
   <not supported>      cycles
   <not supported>      instructions
   <not supported>      branches
   <not supported>      branch-misses
       0.001192373 seconds time elapsed

<<>>

---
Changelog:
From v1 -> v2:
Addressed review comment from Jiri Olsa by changing condition check to directly use "cpu_set_size * 8" for comparing with the cpu number.

 tools/perf/util/affinity.c | 8 +++++++-
 1 file changed, 7 insertions(+), 1 deletion(-)
tools/perf/util/affinity.c +7 -1
index 4d216c0dc425..4ee96b3c755b 100644
@@ -49,8 +49,14 @@ void affinity__set(struct affinity *a, int cpu)
 {
 	int cpu_set_size = get_cpu_set_size();
 
-	if (cpu == -1)
+	/*
+	 * Return:
+	 * - if cpu is -1
+	 * - restrict out of bound access to sched_cpus
+	 */
+	if (cpu == -1 || ((cpu >= (cpu_set_size * 8))))
 		return;
+
 	a->changed = true;
 	set_bit(cpu, a->sched_cpus);
 	/*
Signature · 1 line
2.35.1
Signature · 1 line
- Arnaldo