summaryrefslogtreecommitdiffstats
path: root/kernel/bpf
diff options
context:
space:
mode:
authorDaniel Baumann <daniel.baumann@progress-linux.org>2024-05-04 10:15:33 +0000
committerDaniel Baumann <daniel.baumann@progress-linux.org>2024-05-04 10:15:33 +0000
commita097cf30108c0151cb2684afd962bd2678ff97d7 (patch)
treede1ac8afc7248eec4b2fb60496b8e9273452d9f7 /kernel/bpf
parentAdding upstream version 6.1.82. (diff)
downloadlinux-a097cf30108c0151cb2684afd962bd2678ff97d7.tar.xz
linux-a097cf30108c0151cb2684afd962bd2678ff97d7.zip
Adding upstream version 6.1.85.upstream/6.1.85
Signed-off-by: Daniel Baumann <daniel.baumann@progress-linux.org>
Diffstat (limited to 'kernel/bpf')
-rw-r--r--kernel/bpf/core.c7
-rw-r--r--kernel/bpf/cpumap.c3
-rw-r--r--kernel/bpf/devmap.c11
-rw-r--r--kernel/bpf/hashtab.c14
-rw-r--r--kernel/bpf/helpers.c4
-rw-r--r--kernel/bpf/stackmap.c9
-rw-r--r--kernel/bpf/verifier.c5
7 files changed, 37 insertions, 16 deletions
diff --git a/kernel/bpf/core.c b/kernel/bpf/core.c
index 76bf1de26..44abf88e1 100644
--- a/kernel/bpf/core.c
+++ b/kernel/bpf/core.c
@@ -857,7 +857,12 @@ static LIST_HEAD(pack_list);
* CONFIG_MMU=n. Use PAGE_SIZE in these cases.
*/
#ifdef PMD_SIZE
-#define BPF_PROG_PACK_SIZE (PMD_SIZE * num_possible_nodes())
+/* PMD_SIZE is really big for some archs. It doesn't make sense to
+ * reserve too much memory in one allocation. Hardcode BPF_PROG_PACK_SIZE to
+ * 2MiB * num_possible_nodes(). On most architectures PMD_SIZE will be
+ * greater than or equal to 2MB.
+ */
+#define BPF_PROG_PACK_SIZE (SZ_2M * num_possible_nodes())
#else
#define BPF_PROG_PACK_SIZE PAGE_SIZE
#endif
diff --git a/kernel/bpf/cpumap.c b/kernel/bpf/cpumap.c
index 050893704..806a7c1b3 100644
--- a/kernel/bpf/cpumap.c
+++ b/kernel/bpf/cpumap.c
@@ -306,6 +306,7 @@ static int cpu_map_bpf_prog_run(struct bpf_cpu_map_entry *rcpu, void **frames,
static int cpu_map_kthread_run(void *data)
{
struct bpf_cpu_map_entry *rcpu = data;
+ unsigned long last_qs = jiffies;
complete(&rcpu->kthread_running);
set_current_state(TASK_INTERRUPTIBLE);
@@ -331,10 +332,12 @@ static int cpu_map_kthread_run(void *data)
if (__ptr_ring_empty(rcpu->queue)) {
schedule();
sched = 1;
+ last_qs = jiffies;
} else {
__set_current_state(TASK_RUNNING);
}
} else {
+ rcu_softirq_qs_periodic(last_qs);
sched = cond_resched();
}
diff --git a/kernel/bpf/devmap.c b/kernel/bpf/devmap.c
index f9a87dcc5..e051cbb07 100644
--- a/kernel/bpf/devmap.c
+++ b/kernel/bpf/devmap.c
@@ -131,13 +131,14 @@ static int dev_map_init_map(struct bpf_dtab *dtab, union bpf_attr *attr)
bpf_map_init_from_attr(&dtab->map, attr);
if (attr->map_type == BPF_MAP_TYPE_DEVMAP_HASH) {
- dtab->n_buckets = roundup_pow_of_two(dtab->map.max_entries);
-
- if (!dtab->n_buckets) /* Overflow check */
+ /* hash table size must be power of 2; roundup_pow_of_two() can
+ * overflow into UB on 32-bit arches, so check that first
+ */
+ if (dtab->map.max_entries > 1UL << 31)
return -EINVAL;
- }
- if (attr->map_type == BPF_MAP_TYPE_DEVMAP_HASH) {
+ dtab->n_buckets = roundup_pow_of_two(dtab->map.max_entries);
+
dtab->dev_index_head = dev_map_create_hash(dtab->n_buckets,
dtab->map.numa_node);
if (!dtab->dev_index_head)
diff --git a/kernel/bpf/hashtab.c b/kernel/bpf/hashtab.c
index 88c71de0a..0c74cc901 100644
--- a/kernel/bpf/hashtab.c
+++ b/kernel/bpf/hashtab.c
@@ -495,7 +495,13 @@ static struct bpf_map *htab_map_alloc(union bpf_attr *attr)
num_possible_cpus());
}
- /* hash table size must be power of 2 */
+ /* hash table size must be power of 2; roundup_pow_of_two() can overflow
+ * into UB on 32-bit arches, so check that first
+ */
+ err = -E2BIG;
+ if (htab->map.max_entries > 1UL << 31)
+ goto free_htab;
+
htab->n_buckets = roundup_pow_of_two(htab->map.max_entries);
htab->elem_size = sizeof(struct htab_elem) +
@@ -505,10 +511,8 @@ static struct bpf_map *htab_map_alloc(union bpf_attr *attr)
else
htab->elem_size += round_up(htab->map.value_size, 8);
- err = -E2BIG;
- /* prevent zero size kmalloc and check for u32 overflow */
- if (htab->n_buckets == 0 ||
- htab->n_buckets > U32_MAX / sizeof(struct bucket))
+ /* check for u32 overflow */
+ if (htab->n_buckets > U32_MAX / sizeof(struct bucket))
goto free_htab;
err = -ENOMEM;
diff --git a/kernel/bpf/helpers.c b/kernel/bpf/helpers.c
index 83f8f67e9..758510b46 100644
--- a/kernel/bpf/helpers.c
+++ b/kernel/bpf/helpers.c
@@ -328,7 +328,7 @@ static inline void __bpf_spin_lock_irqsave(struct bpf_spin_lock *lock)
__this_cpu_write(irqsave_flags, flags);
}
-notrace BPF_CALL_1(bpf_spin_lock, struct bpf_spin_lock *, lock)
+NOTRACE_BPF_CALL_1(bpf_spin_lock, struct bpf_spin_lock *, lock)
{
__bpf_spin_lock_irqsave(lock);
return 0;
@@ -350,7 +350,7 @@ static inline void __bpf_spin_unlock_irqrestore(struct bpf_spin_lock *lock)
local_irq_restore(flags);
}
-notrace BPF_CALL_1(bpf_spin_unlock, struct bpf_spin_lock *, lock)
+NOTRACE_BPF_CALL_1(bpf_spin_unlock, struct bpf_spin_lock *, lock)
{
__bpf_spin_unlock_irqrestore(lock);
return 0;
diff --git a/kernel/bpf/stackmap.c b/kernel/bpf/stackmap.c
index f86db3cf7..f0fd936ce 100644
--- a/kernel/bpf/stackmap.c
+++ b/kernel/bpf/stackmap.c
@@ -94,11 +94,14 @@ static struct bpf_map *stack_map_alloc(union bpf_attr *attr)
} else if (value_size / 8 > sysctl_perf_event_max_stack)
return ERR_PTR(-EINVAL);
- /* hash table size must be power of 2 */
- n_buckets = roundup_pow_of_two(attr->max_entries);
- if (!n_buckets)
+ /* hash table size must be power of 2; roundup_pow_of_two() can overflow
+ * into UB on 32-bit arches, so check that first
+ */
+ if (attr->max_entries > 1UL << 31)
return ERR_PTR(-E2BIG);
+ n_buckets = roundup_pow_of_two(attr->max_entries);
+
cost = n_buckets * sizeof(struct stack_map_bucket *) + sizeof(*smap);
smap = bpf_map_area_alloc(cost, bpf_map_attr_numa_node(attr));
if (!smap)
diff --git a/kernel/bpf/verifier.c b/kernel/bpf/verifier.c
index 1a29ac4db..27cc6e3db 100644
--- a/kernel/bpf/verifier.c
+++ b/kernel/bpf/verifier.c
@@ -4965,6 +4965,11 @@ static int check_stack_access_within_bounds(
err = check_stack_slot_within_bounds(min_off, state, type);
if (!err && max_off > 0)
err = -EINVAL; /* out of stack access into non-negative offsets */
+ if (!err && access_size < 0)
+ /* access_size should not be negative (or overflow an int); others checks
+ * along the way should have prevented such an access.
+ */
+ err = -EFAULT; /* invalid negative access size; integer overflow? */
if (err) {
if (tnum_is_const(reg->var_off)) {