Merge branch 'support-bpf_kptr_xchg-into-local-kptr'

Amery Hung says:

====================
Support bpf_kptr_xchg into local kptr

This revision adds substaintial changes to patch 2 to support structures
with kptr as the only special btf type. The test is split into
local_kptr_stash and task_kfunc_success to remove dependencies on
bpf_testmod that would break veristat results.

This series allows stashing kptr into local kptr. Currently, kptrs are
only allowed to be stashed into map value with bpf_kptr_xchg(). A
motivating use case of this series is to enable adding referenced kptr to
bpf_rbtree or bpf_list by using allocated object as graph node and the
storage of referenced kptr. For example, a bpf qdisc [0] enqueuing a
referenced kptr to a struct sk_buff* to a bpf_list serving as a fifo:

    struct skb_node {
            struct sk_buff __kptr *skb;
            struct bpf_list_node node;
    };

    private(A) struct bpf_spin_lock fifo_lock;
    private(A) struct bpf_list_head fifo __contains(skb_node, node);

    /* In Qdisc_ops.enqueue */
    struct skb_node *skbn;

    skbn = bpf_obj_new(typeof(*skbn));
    if (!skbn)
        goto drop;

    /* skb is a referenced kptr to struct sk_buff acquired earilier
     * but not shown in this code snippet.
     */
    skb = bpf_kptr_xchg(&skbn->skb, skb);
    if (skb)
        /* should not happen; do something below releasing skb to
         * satisfy the verifier */
    	...

    bpf_spin_lock(&fifo_lock);
    bpf_list_push_back(&fifo, &skbn->node);
    bpf_spin_unlock(&fifo_lock);

The implementation first searches for BPF_KPTR when generating program
BTF. Then, we teach the verifier that the detination argument of
bpf_kptr_xchg() can be local kptr, and use the btf_record in program BTF
to check against the source argument.

This series is mostly developed by Dave, who kindly helped and sent me
the patchset. The selftests in bpf qdisc (WIP) relies on this series to
work.

[0] https://lore.kernel.org/netdev/20240714175130.4051012-10-amery.hung@bytedance.com/
---
v3 -> v4
  - Allow struct in prog btf w/ kptr as the only special field type
  - Split tests of stashing referenced kptr and local kptr
  - v3: https://lore.kernel.org/bpf/20240809005131.3916464-1-amery.hung@bytedance.com/

v2 -> v3
  - Fix prog btf memory leak
  - Test stashing kptr in prog btf
  - Test unstashing kptrs after stashing into local kptrs
  - v2: https://lore.kernel.org/bpf/20240803001145.635887-1-amery.hung@bytedance.com/

v1 -> v2
  - Fix the document for bpf_kptr_xchg()
  - Add a comment explaining changes in the verifier
  - v1: https://lore.kernel.org/bpf/20240728030115.3970543-1-amery.hung@bytedance.com/
====================

Link: https://lore.kernel.org/r/20240813212424.2871455-1-amery.hung@bytedance.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>

2 files changed, 53 insertions, 3 deletions

diff --git a/tools/testing/selftests/bpf/progs/local_kptr_stash.c b/tools/testing/selftests/bpf/progs/local_kptr_stash.c
index 75043ffc5dad..b092a72b2c9d 100644
--- a/tools/testing/selftests/bpf/progs/local_kptr_stash.c
+++ b/tools/testing/selftests/bpf/progs/local_kptr_stash.c
@@ -8,9 +8,12 @@
 #include "../bpf_experimental.h"
 #include "../bpf_testmod/bpf_testmod_kfunc.h"
 
+struct plain_local;
+
 struct node_data {
 	long key;
 	long data;
+	struct plain_local __kptr * stashed_in_local_kptr;
 	struct bpf_rb_node node;
 };
 
@@ -85,6 +88,7 @@ static bool less(struct bpf_rb_node *a, const struct bpf_rb_node *b)
 
 static int create_and_stash(int idx, int val)
 {
+	struct plain_local *inner_local_kptr;
 	struct map_value *mapval;
 	struct node_data *res;
 
@@ -92,11 +96,25 @@ static int create_and_stash(int idx, int val)
 	if (!mapval)
 		return 1;
 
+	inner_local_kptr = bpf_obj_new(typeof(*inner_local_kptr));
+	if (!inner_local_kptr)
+		return 2;
+
 	res = bpf_obj_new(typeof(*res));
-	if (!res)
-		return 1;
+	if (!res) {
+		bpf_obj_drop(inner_local_kptr);
+		return 3;
+	}
 	res->key = val;
 
+	inner_local_kptr = bpf_kptr_xchg(&res->stashed_in_local_kptr, inner_local_kptr);
+	if (inner_local_kptr) {
+		/* Should never happen, we just obj_new'd res */
+		bpf_obj_drop(inner_local_kptr);
+		bpf_obj_drop(res);
+		return 4;
+	}
+
 	res = bpf_kptr_xchg(&mapval->node, res);
 	if (res)
 		bpf_obj_drop(res);
@@ -169,6 +187,7 @@ long stash_local_with_root(void *ctx)
 SEC("tc")
 long unstash_rb_node(void *ctx)
 {
+	struct plain_local *inner_local_kptr = NULL;
 	struct map_value *mapval;
 	struct node_data *res;
 	long retval;
@@ -180,6 +199,13 @@ long unstash_rb_node(void *ctx)
 
 	res = bpf_kptr_xchg(&mapval->node, NULL);
 	if (res) {
+		inner_local_kptr = bpf_kptr_xchg(&res->stashed_in_local_kptr, inner_local_kptr);
+		if (!inner_local_kptr) {
+			bpf_obj_drop(res);
+			return 1;
+		}
+		bpf_obj_drop(inner_local_kptr);
+
 		retval = res->key;
 		bpf_obj_drop(res);
 		return retval;
diff --git a/tools/testing/selftests/bpf/progs/task_kfunc_success.c b/tools/testing/selftests/bpf/progs/task_kfunc_success.c
index 70df695312dc..3138bb689b0b 100644
--- a/tools/testing/selftests/bpf/progs/task_kfunc_success.c
+++ b/tools/testing/selftests/bpf/progs/task_kfunc_success.c
@@ -5,6 +5,7 @@
 #include <bpf/bpf_tracing.h>
 #include <bpf/bpf_helpers.h>
 
+#include "../bpf_experimental.h"
 #include "task_kfunc_common.h"
 
 char _license[] SEC("license") = "GPL";
@@ -143,7 +144,7 @@ SEC("tp_btf/task_newtask")
 int BPF_PROG(test_task_xchg_release, struct task_struct *task, u64 clone_flags)
 {
 	struct task_struct *kptr;
-	struct __tasks_kfunc_map_value *v;
+	struct __tasks_kfunc_map_value *v, *local;
 	long status;
 
 	if (!is_test_kfunc_task())
@@ -167,6 +168,29 @@ int BPF_PROG(test_task_xchg_release, struct task_struct *task, u64 clone_flags)
 		return 0;
 	}
 
+	local = bpf_obj_new(typeof(*local));
+	if (!local) {
+		err = 4;
+		bpf_task_release(kptr);
+		return 0;
+	}
+
+	kptr = bpf_kptr_xchg(&local->task, kptr);
+	if (kptr) {
+		err = 5;
+		bpf_obj_drop(local);
+		bpf_task_release(kptr);
+		return 0;
+	}
+
+	kptr = bpf_kptr_xchg(&local->task, NULL);
+	if (!kptr) {
+		err = 6;
+		bpf_obj_drop(local);
+		return 0;
+	}
+
+	bpf_obj_drop(local);
 	bpf_task_release(kptr);
 
 	return 0;