Merge branch 'BPF rbtree next-gen datastructure'

Dave Marchevsky says:

====================
This series adds a rbtree datastructure following the "next-gen
datastructure" precedent set by recently-added linked-list [0]. This is
a reimplementation of previous rbtree RFC [1] to use kfunc + kptr
instead of adding a new map type. This series adds a smaller set of API
functions than that RFC - just the minimum needed to support current
cgfifo example scheduler in ongoing sched_ext effort [2], namely:

  bpf_rbtree_add
  bpf_rbtree_remove
  bpf_rbtree_first

The meat of this series is bugfixes and verifier infra work to support
these API functions. Adding more rbtree kfuncs in future patches should
be straightforward as a result.

First, the series refactors and extends linked_list's release_on_unlock
logic. The concept of "reference to node that was added to data
structure" is formalized as "non-owning reference". From linked_list's
perspective this non-owning reference after
linked_list_push_{front,back} has same semantics as release_on_unlock,
with the addition of writes to such references being valid in the
critical section. Such references are no longer marked PTR_UNTRUSTED.
Patches 2 and 13 go into more detail.

The series then adds rbtree API kfuncs and necessary verifier support
for them - namely support for callback args to kfuncs and some
non-owning reference interactions that linked_list didn't need.

BPF rbtree uses struct rb_root_cached + existing rbtree lib under the
hood. From the BPF program writer's perspective, a BPF rbtree is very
similar to existing linked list. Consider the following example:

  struct node_data {
    long key;
    long data;
    struct bpf_rb_node node;
  }

  static bool less(struct bpf_rb_node *a, const struct bpf_rb_node *b)
  {
    struct node_data *node_a;
    struct node_data *node_b;

    node_a = container_of(a, struct node_data, node);
    node_b = container_of(b, struct node_data, node);

    return node_a->key < node_b->key;
  }

  private(A) struct bpf_spin_lock glock;
  private(A) struct bpf_rb_root groot __contains(node_data, node);

  /* ... in BPF program */
  struct node_data *n, *m;
  struct bpf_rb_node *res;

  n = bpf_obj_new(typeof(*n));
  if (!n)
    /* skip */
  n->key = 5;
  n->data = 10;

  bpf_spin_lock(&glock);
  bpf_rbtree_add(&groot, &n->node, less);
  bpf_spin_unlock(&glock);

  bpf_spin_lock(&glock);
  res = bpf_rbtree_first(&groot);
  if (!res)
    /* skip */
  res = bpf_rbtree_remove(&groot, res);
  if (!res)
    /* skip */
  bpf_spin_unlock(&glock);

  m = container_of(res, struct node_data, node);
  bpf_obj_drop(m);

Some obvious similarities:

  * Special bpf_rb_root and bpf_rb_node types have same semantics
    as bpf_list_head and bpf_list_node, respectively
  * __contains is used to associated node type with root
  * The spin_lock associated with a rbtree must be held when using
    rbtree API kfuncs
  * Nodes are allocated via bpf_obj_new and dropped via bpf_obj_drop
  * Rbtree takes ownership of node lifetime when a node is added.
    Removing a node gives ownership back to the program, requiring a
    bpf_obj_drop before program exit

Some new additions as well:

  * Support for callbacks in kfunc args is added to enable 'less'
    callback use above
  * bpf_rbtree_first is the first graph API function to return a
    non-owning reference instead of convering an arg from own->non-own
  * Because all references to nodes already added to the rbtree are
    non-owning, bpf_rbtree_remove must accept such a reference in order
    to remove it from the tree

Summary of patches:
  Patches 1 - 5 implement the meat of rbtree-specific support in this
  series, gradually building up to implemented kfuncs that verify as
  expected.

  Patch 6 adds the bpf_rbtree_{add,first,remove} to bpf_experimental.h.

  Patch 7 adds tests, Patch 9 adds documentation.

  [0]: lore.kernel.org/bpf/20221118015614.2013203-1-memxor@gmail.com
  [1]: lore.kernel.org/bpf/20220830172759.4069786-1-davemarchevsky@fb.com
  [2]: lore.kernel.org/bpf/20221130082313.3241517-1-tj@kernel.org

Changelog:

v5 -> v6: lore.kernel.org/bpf/20230212092715.1422619-1-davemarchevsky@fb.com/

Patch #'s below refer to the patch's number in v5 unless otherwise stated.

* General / Patch 1
  * Rebase onto latest bpf-next: "bpf: Migrate release_on_unlock logic to non-owning ref semantics"
  * This was Patch 1 of v4, was applied, not included in v6

* Patch 3 - "bpf: Add bpf_rbtree_{add,remove,first} kfuncs"
  * Use bpf_callback_t instead of plain-C fn ptr for bpf_rbtree_add. This
    necessitated having bpf_rbtree_add duplicate rbtree_add's functionality.
    Wrapper function was used w/ internal __bpf_rbtree_add helper so that
    bpf_experimental.h proto could continue to use plain-C fn ptr so BPF progs
    could benefit from typechecking (Alexei)

v4 -> v5: lore.kernel.org/bpf/20230209174144.3280955-1-davemarchevsky@fb.com/

Patch #'s below refer to the patch's number in v4 unless otherwise stated.

* General
  * Rebase onto latest bpf-next: "Merge branch 'bpf, mm: introduce cgroup.memory=nobpf'"

* Patches 1-3 are squashed into "bpf: Migrate release_on_unlock logic to non-owning ref semantics".
  * Added type_is_non_owning_ref helper (Alexei)
  * Use a NON_OWN_REF type flag instead of separate bool (Alexei)

* Patch 8 - "bpf: Special verifier handling for bpf_rbtree_{remove, first}"
  * When doing btf_parse_fields, reject structs with both bpf_list_node and
    bpf_rb_node fields. This is a temporary measure that can be removed after
    "collection identity" followup. See comment added in btf_parse_fields for
    more detail (Kumar, Alexei)
  * Add linked_list BTF test exercising check added to btf_parse_fields
  * Minor changes and moving around of some reg type checks due to NON_OWN_REF type flag
    introduction

* Patch 10 - "selftests/bpf: Add rbtree selftests"
  * Migrate failure tests to RUN_TESTS, __failure, __msg() framework (Alexei)

v3 -> v4: lore.kernel.org/bpf/20230131180016.3368305-1-davemarchevsky@fb.com/

Patch #'s below refer to the patch's number in v3 unless otherwise stated.

* General
  * Don't base this series on "bpf: Refactor release_regno searching logic",
    which was submitted separately as a refactor.
  * Rebase onto latest bpf-next: "samples/bpf: Add openat2() enter/exit tracepoint to syscall_tp sample"

* Patch 2 - "bpf: Improve bpf_reg_state space usage for non-owning ref lock"
  * print_verifier_state change was adding redundant comma after "non_own_ref",
    fix it to put comma in correct place
  * invalidate_non_owning_refs no longer needs to take bpf_active_lock param,
    since any non-owning ref reg in env's cur_state is assumed to use that
    state's active_lock (Alexei)
  * invalidate_non_owning_refs' reg loop should check that the reg being
    inspected is a PTR_TO_BTF_ID before checking reg->non_owning_ref_lock,
    since that field is part of a union and may be filled w/ meaningless bytes
    if reg != PTR_TO_BTF_ID (Alexei)

* Patch 3 - "selftests/bpf: Update linked_list tests for non-owning ref semantics"
  * Change the string searched for by the following tests:
    * linked_list/incorrect_node_off1
    * linked_list/double_push_front
    * linked_list/double_push_back

    necessary due to rebase / dropping of "release_regno searching logic" patch
    (see "General" changes)

* Patch 8 - "bpf: Special verifier handling for bpf_rbtree_{remove, first}"
  * Just call invalidate_non_owning_refs w/ env instead of env, lock. (see
    Patch 2 changes)

* Patch 11 - "bpf, documentation: Add graph documentation for non-owning refs"
  * Fix documentation formatting and improve content (David)
  * v3's version of patch 11 was missing some changes, v4's patch 11 is still
    addressing David's feedback from v2

v2 -> v3: lore.kernel.org/bpf/20221217082506.1570898-1-davemarchevsky@fb.com/

Patch #'s below refer to the patch's number in v2 unless otherwise stated.

* Patch 1 - "bpf: Support multiple arg regs w/ ref_obj_id for kfuncs"
  * No longer needed as v3 doesn't have multiple ref_obj_id arg regs
  * The refactoring pieces were submitted separately
    (https://lore.kernel.org/bpf/20230121002417.1684602-1-davemarchevsky@fb.com/)

* Patch 2 - "bpf: Migrate release_on_unlock logic to non-owning ref semantics"
  * Remove KF_RELEASE_NON_OWN flag from list API push methods, just match
    against specific kfuncs for now (Alexei, David)
  * Separate "release non owning reference" logic from KF_RELEASE logic
    (Alexei, David)
  * reg_find_field_offset now correctly tests 'rec' instead of 'reg' after
    calling reg_btf_record (Dan Carpenter)

* New patch added after Patch 2 - "bpf: Improve bpf_reg_state space usage for non-owning ref lock"
  * Eliminates extra bpf_reg_state memory usage by using a bool instead of
    copying lock identity

* Patch 4 - "bpf: rename list_head -> graph_root in field info types"
  * v2's version was applied to bpf-next, not including in respins

* Patch 6 - "bpf: Add bpf_rbtree_{add,remove,first} kfuncs"
  * Remove KF_RELEASE_NON_OWN flag from rbtree_add, just add it to specific
    kfunc matching (Alexei, David)

* Patch 9 - "bpf: Special verifier handling for bpf_rbtree_{remove, first}"
  * Remove KF_INVALIDATE_NON_OWN kfunc flag, just match against specific kfunc
    for now (Alexei, David)

* Patch 11 - "libbpf: Make BTF mandatory if program BTF has spin_lock or alloc_obj type"
  * Drop for now, will submit separately

* Patch 12 - "selftests/bpf: Add rbtree selftests"
  * Some expected-failure tests have different error messages due to "release
    non-owning reference logic" being separated from KF_RELEASE logic in Patch
    2 changes

* Patch 13 - "bpf, documentation: Add graph documentation for non-owning refs"
  * Fix documentation formatting and improve content (David)

v1 -> v2: lore.kernel.org/bpf/20221206231000.3180914-1-davemarchevsky@fb.com/

Series-wide changes:
  * Rename datastructure_{head,node,api} -> graph_{root,node,api} (Alexei)
  * "graph datastructure" in patch summaries to refer to linked_list + rbtree
    instead of "next-gen datastructure" (Alexei)
  * Move from hacky marking of non-owning references as PTR_UNTRUSTED to
    cleaner implementation (Alexei)
  * Add invalidation of non-owning refs to rbtree_remove (Kumar, Alexei)

Patch #'s below refer to the patch's number in v1 unless otherwise stated.

Note that in v1 most of the meaty verifier changes were in the latter half
of the series. Here, about half of that complexity has been moved to
"bpf: Migrate release_on_unlock logic to non-owning ref semantics" - was Patch
3 in v1.

* Patch 1 - "bpf: Loosen alloc obj test in verifier's reg_btf_record"
  * Was applied, dropped from further iterations

* Patch 2 - "bpf: map_check_btf should fail if btf_parse_fields fails"
  * Dropped in favor of verifier check-on-use: when some normal verifier
    checking expects the map to have btf_fields correctly parsed, it won't
    find any and verification will fail

* New patch added before Patch 3 - "bpf: Support multiple arg regs w/ ref_obj_id for kfuncs"
  * Addition of KF_RELEASE_NON_OWN flag, which requires KF_RELEASE, and tagging
    of bpf_list_push_{front,back} KF_RELEASE | KF_RELEASE_NON_OWN, means that
    list-in-list push_{front,back} will trigger "only one ref_obj_id arg reg"
    logic. This is because "head" arg to those functions can be a list-in-list,
    which itself can be an owning reference with ref_obj_id. So need to
    support multiple ref_obj_id for release kfuncs.

* Patch 3 - "bpf: Minor refactor of ref_set_release_on_unlock"
  * Now a major refactor w/ a rename to reflect this
    * "bpf: Migrate release_on_unlock logic to non-owning ref semantics"
  * Replaces release_on_unlock with active_lock logic as discussed in v1

* New patch added after Patch 3 - "selftests/bpf: Update linked_list tests for non_owning_ref logic"
  * Removes "write after push" linked_list failure tests - no longer failure
    scenarios.

* Patch 4 - "bpf: rename list_head -> datastructure_head in field info types"
  * rename to graph_root instead. Similar renamings across the series - see
    series-wide changes.

* Patch 5 - "bpf: Add basic bpf_rb_{root,node} support"
  * OWNER_FIELD_MASK -> GRAPH_ROOT_MASK, OWNEE_FIELD_MASK -> GRAPH_NODE_MASK,
    and change of "owner"/"ownee" in big btf_check_and_fixup_fields comment to
    "root"/"node" (Alexei)

* Patch 6 - "bpf: Add bpf_rbtree_{add,remove,first} kfuncs"
  * bpf_rbtree_remove can no longer return NULL. v2 continues v1's "use type
    system to prevent remove of node that isn't in a datastructure" approach,
    so rbtree_remove should never have been able to return NULL

* Patch 7 - "bpf: Add support for bpf_rb_root and bpf_rb_node in kfunc args"
  * is_bpf_datastructure_api_kfunc -> is_bpf_graph_api_kfunc (Alexei)

* Patch 8 - "bpf: Add callback validation to kfunc verifier logic"
  * Explicitly disallow rbtree_remove in rbtree callback
  * Explicitly disallow bpf_spin_{lock,unlock} call in rbtree callback,
    preventing possibility of "unbalanced" unlock (Alexei)

* Patch 10 - "bpf, x86: BPF_PROBE_MEM handling for insn->off < 0"
  * Now that non-owning refs aren't marked PTR_UNTRUSTED it's not necessary to
    include this patch as part of the series
  * After conversation w/ Alexei, did another pass and submitted as an
    independent series (lore.kernel.org/bpf/20221213182726.325137-1-davemarchevsky@fb.com/)

* Patch 13 - "selftests/bpf: Add rbtree selftests"
  * Since bpf_rbtree_remove can no longer return null, remove null checks
  * Remove test confirming that rbtree_first isn't allowed in callback. We want
    this to be possible
  * Add failure test confirming that rbtree_remove's new non-owning reference
    invalidation behavior behaves as expected
  * Add SEC("license") to rbtree_btf_fail__* progs. They were previously
    failing due to lack of this section. Now they're failing for correct
    reasons.
  * rbtree_btf_fail__add_wrong_type.c - add locking around rbtree_add, rename
    the bpf prog to something reasonable

* New patch added after patch 13 - "bpf, documentation: Add graph documentation for non-owning refs"
  * Summarizes details of owning and non-owning refs which we hashed out in
    v1
====================

Signed-off-by: Alexei Starovoitov <ast@kernel.org>

Documentation/bpf/graph_ds_impl.rst

@@ -0,0 +1,266 @@
+=========================
+BPF Graph Data Structures
+=========================
+
+This document describes implementation details of new-style "graph" data
+structures (linked_list, rbtree), with particular focus on the verifier's
+implementation of semantics specific to those data structures.
+
+Although no specific verifier code is referred to in this document, the document
+assumes that the reader has general knowledge of BPF verifier internals, BPF
+maps, and BPF program writing.
+
+Note that the intent of this document is to describe the current state of
+these graph data structures. **No guarantees** of stability for either
+semantics or APIs are made or implied here.
+
+.. contents::
+    :local:
+    :depth: 2
+
+Introduction
+------------
+
+The BPF map API has historically been the main way to expose data structures
+of various types for use within BPF programs. Some data structures fit naturally
+with the map API (HASH, ARRAY), others less so. Consequentially, programs
+interacting with the latter group of data structures can be hard to parse
+for kernel programmers without previous BPF experience.
+
+Luckily, some restrictions which necessitated the use of BPF map semantics are
+no longer relevant. With the introduction of kfuncs, kptrs, and the any-context
+BPF allocator, it is now possible to implement BPF data structures whose API
+and semantics more closely match those exposed to the rest of the kernel.
+
+Two such data structures - linked_list and rbtree - have many verification
+details in common. Because both have "root"s ("head" for linked_list) and
+"node"s, the verifier code and this document refer to common functionality
+as "graph_api", "graph_root", "graph_node", etc.
+
+Unless otherwise stated, examples and semantics below apply to both graph data
+structures.
+
+Unstable API
+------------
+
+Data structures implemented using the BPF map API have historically used BPF
+helper functions - either standard map API helpers like ``bpf_map_update_elem``
+or map-specific helpers. The new-style graph data structures instead use kfuncs
+to define their manipulation helpers. Because there are no stability guarantees
+for kfuncs, the API and semantics for these data structures can be evolved in
+a way that breaks backwards compatibility if necessary.
+
+Root and node types for the new data structures are opaquely defined in the
+``uapi/linux/bpf.h`` header.
+
+Locking
+-------
+
+The new-style data structures are intrusive and are defined similarly to their
+vanilla kernel counterparts:
+
+.. code-block:: c
+        struct node_data {
+          long key;
+          long data;
+          struct bpf_rb_node node;
+        };
+
+        struct bpf_spin_lock glock;
+        struct bpf_rb_root groot __contains(node_data, node);
+
+The "root" type for both linked_list and rbtree expects to be in a map_value
+which also contains a ``bpf_spin_lock`` - in the above example both global
+variables are placed in a single-value arraymap. The verifier considers this
+spin_lock to be associated with the ``bpf_rb_root`` by virtue of both being in
+the same map_value and will enforce that the correct lock is held when
+verifying BPF programs that manipulate the tree. Since this lock checking
+happens at verification time, there is no runtime penalty.
+
+Non-owning references
+---------------------
+
+**Motivation**
+
+Consider the following BPF code:
+
+.. code-block:: c
+
+        struct node_data *n = bpf_obj_new(typeof(*n)); /* ACQUIRED */
+
+        bpf_spin_lock(&lock);
+
+        bpf_rbtree_add(&tree, n); /* PASSED */
+
+        bpf_spin_unlock(&lock);
+
+From the verifier's perspective, the pointer ``n`` returned from ``bpf_obj_new``
+has type ``PTR_TO_BTF_ID | MEM_ALLOC``, with a ``btf_id`` of
+``struct node_data`` and a nonzero ``ref_obj_id``. Because it holds ``n``, the
+program has ownership of the pointee's (object pointed to by ``n``) lifetime.
+The BPF program must pass off ownership before exiting - either via
+``bpf_obj_drop``, which ``free``'s the object, or by adding it to ``tree`` with
+``bpf_rbtree_add``.
+
+(``ACQUIRED`` and ``PASSED`` comments in the example denote statements where
+"ownership is acquired" and "ownership is passed", respectively)
+
+What should the verifier do with ``n`` after ownership is passed off? If the
+object was ``free``'d with ``bpf_obj_drop`` the answer is obvious: the verifier
+should reject programs which attempt to access ``n`` after ``bpf_obj_drop`` as
+the object is no longer valid. The underlying memory may have been reused for
+some other allocation, unmapped, etc.
+
+When ownership is passed to ``tree`` via ``bpf_rbtree_add`` the answer is less
+obvious. The verifier could enforce the same semantics as for ``bpf_obj_drop``,
+but that would result in programs with useful, common coding patterns being
+rejected, e.g.:
+
+.. code-block:: c
+
+        int x;
+        struct node_data *n = bpf_obj_new(typeof(*n)); /* ACQUIRED */
+
+        bpf_spin_lock(&lock);
+
+        bpf_rbtree_add(&tree, n); /* PASSED */
+        x = n->data;
+        n->data = 42;
+
+        bpf_spin_unlock(&lock);
+
+Both the read from and write to ``n->data`` would be rejected. The verifier
+can do better, though, by taking advantage of two details:
+
+  * Graph data structure APIs can only be used when the ``bpf_spin_lock``
+    associated with the graph root is held
+
+  * Both graph data structures have pointer stability
+
+     * Because graph nodes are allocated with ``bpf_obj_new`` and
+       adding / removing from the root involves fiddling with the
+       ``bpf_{list,rb}_node`` field of the node struct, a graph node will
+       remain at the same address after either operation.
+
+Because the associated ``bpf_spin_lock`` must be held by any program adding
+or removing, if we're in the critical section bounded by that lock, we know
+that no other program can add or remove until the end of the critical section.
+This combined with pointer stability means that, until the critical section
+ends, we can safely access the graph node through ``n`` even after it was used
+to pass ownership.
+
+The verifier considers such a reference a *non-owning reference*. The ref
+returned by ``bpf_obj_new`` is accordingly considered an *owning reference*.
+Both terms currently only have meaning in the context of graph nodes and API.
+
+**Details**
+
+Let's enumerate the properties of both types of references.
+
+*owning reference*
+
+  * This reference controls the lifetime of the pointee
+
+  * Ownership of pointee must be 'released' by passing it to some graph API
+    kfunc, or via ``bpf_obj_drop``, which ``free``'s the pointee
+
+    * If not released before program ends, verifier considers program invalid
+
+  * Access to the pointee's memory will not page fault
+
+*non-owning reference*
+
+  * This reference does not own the pointee
+
+     * It cannot be used to add the graph node to a graph root, nor ``free``'d via
+       ``bpf_obj_drop``
+
+  * No explicit control of lifetime, but can infer valid lifetime based on
+    non-owning ref existence (see explanation below)
+
+  * Access to the pointee's memory will not page fault
+
+From verifier's perspective non-owning references can only exist
+between spin_lock and spin_unlock. Why? After spin_unlock another program
+can do arbitrary operations on the data structure like removing and ``free``-ing
+via bpf_obj_drop. A non-owning ref to some chunk of memory that was remove'd,
+``free``'d, and reused via bpf_obj_new would point to an entirely different thing.
+Or the memory could go away.
+
+To prevent this logic violation all non-owning references are invalidated by the
+verifier after a critical section ends. This is necessary to ensure the "will
+not page fault" property of non-owning references. So if the verifier hasn't
+invalidated a non-owning ref, accessing it will not page fault.
+
+Currently ``bpf_obj_drop`` is not allowed in the critical section, so
+if there's a valid non-owning ref, we must be in a critical section, and can
+conclude that the ref's memory hasn't been dropped-and- ``free``'d or
+dropped-and-reused.
+
+Any reference to a node that is in an rbtree _must_ be non-owning, since
+the tree has control of the pointee's lifetime. Similarly, any ref to a node
+that isn't in rbtree _must_ be owning. This results in a nice property:
+graph API add / remove implementations don't need to check if a node
+has already been added (or already removed), as the ownership model
+allows the verifier to prevent such a state from being valid by simply checking
+types.
+
+However, pointer aliasing poses an issue for the above "nice property".
+Consider the following example:
+
+.. code-block:: c
+
+        struct node_data *n, *m, *o, *p;
+        n = bpf_obj_new(typeof(*n));     /* 1 */
+
+        bpf_spin_lock(&lock);
+
+        bpf_rbtree_add(&tree, n);        /* 2 */
+        m = bpf_rbtree_first(&tree);     /* 3 */
+
+        o = bpf_rbtree_remove(&tree, n); /* 4 */
+        p = bpf_rbtree_remove(&tree, m); /* 5 */
+
+        bpf_spin_unlock(&lock);
+
+        bpf_obj_drop(o);
+        bpf_obj_drop(p); /* 6 */
+
+Assume the tree is empty before this program runs. If we track verifier state
+changes here using numbers in above comments:
+
+  1) n is an owning reference
+
+  2) n is a non-owning reference, it's been added to the tree
+
+  3) n and m are non-owning references, they both point to the same node
+
+  4) o is an owning reference, n and m non-owning, all point to same node
+
+  5) o and p are owning, n and m non-owning, all point to the same node
+
+  6) a double-free has occurred, since o and p point to same node and o was
+     ``free``'d in previous statement
+
+States 4 and 5 violate our "nice property", as there are non-owning refs to
+a node which is not in an rbtree. Statement 5 will try to remove a node which
+has already been removed as a result of this violation. State 6 is a dangerous
+double-free.
+
+At a minimum we should prevent state 6 from being possible. If we can't also
+prevent state 5 then we must abandon our "nice property" and check whether a
+node has already been removed at runtime.
+
+We prevent both by generalizing the "invalidate non-owning references" behavior
+of ``bpf_spin_unlock`` and doing similar invalidation after
+``bpf_rbtree_remove``. The logic here being that any graph API kfunc which:
+
+  * takes an arbitrary node argument
+
+  * removes it from the data structure
+
+  * returns an owning reference to the removed node
+
+May result in a state where some other non-owning reference points to the same
+node. So ``remove``-type kfuncs must be considered a non-owning reference
+invalidation point as well.

Documentation/bpf/other.rst

@@ -6,4 +6,5 @@ Other
    :maxdepth: 1
 
    ringbuf
-   llvm_reloc
+   llvm_reloc
+   graph_ds_impl

include/linux/bpf.h

@@ -181,7 +181,10 @@ enum btf_field_type {
 	BPF_KPTR       = BPF_KPTR_UNREF | BPF_KPTR_REF,
 	BPF_LIST_HEAD  = (1 << 4),
 	BPF_LIST_NODE  = (1 << 5),
-	BPF_GRAPH_NODE_OR_ROOT = BPF_LIST_NODE | BPF_LIST_HEAD,
+	BPF_RB_ROOT    = (1 << 6),
+	BPF_RB_NODE    = (1 << 7),
+	BPF_GRAPH_NODE_OR_ROOT = BPF_LIST_NODE | BPF_LIST_HEAD |
+				 BPF_RB_NODE | BPF_RB_ROOT,
 };
 
 struct btf_field_kptr {
@@ -285,6 +288,10 @@ static inline const char *btf_field_type_name(enum btf_field_type type)
 		return "bpf_list_head";
 	case BPF_LIST_NODE:
 		return "bpf_list_node";
+	case BPF_RB_ROOT:
+		return "bpf_rb_root";
+	case BPF_RB_NODE:
+		return "bpf_rb_node";
 	default:
 		WARN_ON_ONCE(1);
 		return "unknown";
@@ -305,6 +312,10 @@ static inline u32 btf_field_type_size(enum btf_field_type type)
 		return sizeof(struct bpf_list_head);
 	case BPF_LIST_NODE:
 		return sizeof(struct bpf_list_node);
+	case BPF_RB_ROOT:
+		return sizeof(struct bpf_rb_root);
+	case BPF_RB_NODE:
+		return sizeof(struct bpf_rb_node);
 	default:
 		WARN_ON_ONCE(1);
 		return 0;
@@ -325,6 +336,10 @@ static inline u32 btf_field_type_align(enum btf_field_type type)
 		return __alignof__(struct bpf_list_head);
 	case BPF_LIST_NODE:
 		return __alignof__(struct bpf_list_node);
+	case BPF_RB_ROOT:
+		return __alignof__(struct bpf_rb_root);
+	case BPF_RB_NODE:
+		return __alignof__(struct bpf_rb_node);
 	default:
 		WARN_ON_ONCE(1);
 		return 0;
@@ -435,6 +450,9 @@ void copy_map_value_locked(struct bpf_map *map, void *dst, void *src,
 void bpf_timer_cancel_and_free(void *timer);
 void bpf_list_head_free(const struct btf_field *field, void *list_head,
 			struct bpf_spin_lock *spin_lock);
+void bpf_rb_root_free(const struct btf_field *field, void *rb_root,
+		      struct bpf_spin_lock *spin_lock);
+
 
 int bpf_obj_name_cpy(char *dst, const char *src, unsigned int size);
 

include/uapi/linux/bpf.h

@@ -6917,6 +6917,17 @@ struct bpf_list_node {
 	__u64 :64;
 } __attribute__((aligned(8)));
 
+struct bpf_rb_root {
+	__u64 :64;
+	__u64 :64;
+} __attribute__((aligned(8)));
+
+struct bpf_rb_node {
+	__u64 :64;
+	__u64 :64;
+	__u64 :64;
+} __attribute__((aligned(8)));
+
 struct bpf_sysctl {
 	__u32	write;		/* Sysctl is being read (= 0) or written (= 1).
 				 * Allows 1,2,4-byte read, but no write.