bpf: Skip detached progs in trampoline images that are still in use

bpf_tramp_image_put() makes sure a trampoline image is not freed while
a task may still be running in it, but nothing similar is done for the
progs called by that image. Detach drops the last prog reference right
away and the prog is freed after grace periods, on the basis that a
task still in the traced function skips the fexit progs once the nop at
ip_after_call is patched to a jump.

That leaves out a task sleeping in a sleepable prog that runs before
the detached one, which no grace period waits for:

  CPU 0                               CPU 1
  in image I, sleeping in prog S
                                      detach P from I's trampoline
                                       -> new image, bpf_tramp_image_put(I)
                                      bpf_prog_put(P), last ref
                                      grace periods, P freed
  back from S
  __bpf_prog_enter(P)
  call P->bpf_func

If S and P are fexit progs the task is already past the patched jump,
and fentry only images don't have one. On x86 this is an int3 in
poisoned bpf_prog_pack memory:

  Oops: int3: 0000 [#1] SMP NOPTI
  CPU: 18 UID: 0 PID: 94573 Comm: x169 Not tainted 6.18.44 #1 PREEMPT(lazy)
  RIP: 0010:0xffffffffc0601d8d
  Call Trace:
   <TASK>
   ? bpf_trampoline_6442515411+0x1a4/0x21b
   bpf_lsm_bprm_committed_creds+0x5/0x10
   security_bprm_committed_creds+0x5f/0x70
   begin_new_exec+0x2d6/0x410
   ...

We hit this in production when progs attached through trampolines got
detached while their hooks were busy, and it was independently found
with a fuzzer and KASAN.

Have the JITs emit a patchable nop in front of each prog call sequence
and record it in the image. When a prog is detached, patch its nop to a
jump over the call sequence. Progs that stay attached keep running for
the tasks that are in the image, and ip_after_call isn't needed anymore.

The task can be in any image that isn't freed yet, not only in the
current one. It sleeps in image I1 that calls S, P and Q, then P is
detached and the trampoline moves to image I2, then Q is detached and
its call is still in I1. So the trampoline keeps a list of its images
until they are freed, and detaching a prog patches its nop in all of
them. Images hold a reference on the trampoline for that long.

On riscv and loongarch a jump of any range takes several instructions,
and a task preempted in the middle of them could resume into half of the
new sequence. The nop is a single instruction there, patched to a near
branch through arch_bpf_trampoline_skip().

With the extra nops, BPF_MAX_TRAMP_LINKS progs no longer fit in a page
on x86 and arm64 (and already didn't on powerpc), so lower the limit
there like s390 does.

Fixes: e21aa341785c ("bpf: Fix fexit trampoline.")
Reported-by: Sechang Lim <rhkrqnwk98@gmail.com>
Suggested-by: Alexei Starovoitov <ast@kernel.org>
Signed-off-by: Florent Revest (Anthropic) <florent.revest@linux.dev>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Link: https://patch.msgid.link/20260926135605.1217928-3-florent.revest@linux.dev
Closes: https://lore.kernel.org/bpf/20260815071927.147049-1-zirajs7@gmail.com/

1 files changed, 40 insertions, 4 deletions

diff --git a/include/linux/bpf.h b/include/linux/bpf.h
index a8ec85127571..0ecb9418dfbd 100644
--- a/include/linux/bpf.h
+++ b/include/linux/bpf.h
@@ -1258,11 +1258,15 @@ struct btf_func_model {
 #define BPF_TRAMP_F_INDIRECT		BIT(8)
 
 /* Each call __bpf_prog_enter + call bpf_func + call __bpf_prog_exit is ~50
- * bytes on x86.
+ * bytes on x86. The trampoline image has to fit in PAGE_SIZE.
  */
 enum {
-#if defined(__s390x__)
+#if defined(__s390x__) || defined(__powerpc64__)
 	BPF_MAX_TRAMP_LINKS = 27,
+#elif defined(__x86_64__)
+	BPF_MAX_TRAMP_LINKS = 36,
+#elif defined(__aarch64__)
+	BPF_MAX_TRAMP_LINKS = 37,
 #else
 	BPF_MAX_TRAMP_LINKS = 38,
 #endif
@@ -1315,6 +1319,7 @@ int arch_prepare_bpf_trampoline(struct bpf_tramp_image *im, void *image, void *i
 void *arch_alloc_bpf_trampoline(unsigned int size);
 void arch_free_bpf_trampoline(void *image, unsigned int size);
 int __must_check arch_protect_bpf_trampoline(void *image, unsigned int size);
+int arch_bpf_trampoline_skip(void *nop, void *target);
 int arch_bpf_trampoline_size(const struct btf_func_model *m, u32 flags,
 			     struct bpf_tramp_nodes *tnodes, void *func_addr);
 
@@ -1363,19 +1368,48 @@ enum bpf_tramp_prog_type {
 	BPF_TRAMP_FSESSION,
 };
 
+/*
+ * Each prog call in a trampoline image is preceded by a nop. When the prog is
+ * detached, the nop is patched to a jump to target, right after the call, so
+ * that tasks still running in the image skip the prog.
+ */
+struct bpf_tramp_skip {
+	struct bpf_prog *prog;
+	void *nop;
+	void *target;
+};
+
 struct bpf_tramp_image {
 	void *image;
 	int size;
 	struct bpf_ksym ksym;
 	struct percpu_ref pcref;
-	void *ip_after_call;
-	void *ip_epilogue;
+	bool call_orig;
+	/* entry in tr->images, the image holds a reference on tr */
+	struct bpf_trampoline *tr;
+	struct list_head list;
+	int nr_skips;
+	struct bpf_tramp_skip *skips;
 	union {
 		struct rcu_head rcu;
 		struct work_struct work;
 	};
 };
 
+static inline void bpf_tramp_image_add_skip(struct bpf_tramp_image *im, struct bpf_prog *prog,
+					    void *nop, void *target)
+{
+	struct bpf_tramp_skip *skip;
+
+	/* struct_ops trampolines and arch_bpf_trampoline_size() have no image */
+	if (!im || !im->skips)
+		return;
+	skip = &im->skips[im->nr_skips++];
+	skip->prog = prog;
+	skip->nop = nop;
+	skip->target = target;
+}
+
 struct bpf_trampoline {
 	/* hlist for trampoline_key_table */
 	struct hlist_node hlist_key;
@@ -1402,6 +1436,8 @@ struct bpf_trampoline {
 	int progs_cnt[BPF_TRAMP_MAX];
 	/* Executable image of trampoline */
 	struct bpf_tramp_image *cur_image;
+	/* Images not freed yet, cur_image and older ones still in use */
+	struct list_head images;
 	/* Used as temporary old image storage for multi_attach */
 	struct {
 		struct bpf_tramp_image *old_image;