Re: [PATCH bpf-next v3 04/16] bpf: Convert bpf_selem_unlink to failable
From: Martin KaFai Lau <martin.lau@linux.dev>
Date: 2026-01-09 18:16:19
Also in:
bpf
On 12/18/25 9:56 AM, Amery Hung wrote:
To prepare changing both bpf_local_storage_map_bucket::lock and bpf_local_storage::lock to rqspinlock, convert bpf_selem_unlink() to failable. It still always succeeds and returns 0 until the change happens. No functional change. For bpf_local_storage_map_free(), WARN_ON() for now as no real error will happen until we switch to rqspinlock. __must_check is added to the function declaration locally to make sure all callers are accounted for during the conversion.
I don't see __must_check. The same for patch 2.
quoted hunk ↗ jump to hunk
Signed-off-by: Amery Hung <redacted> --- include/linux/bpf_local_storage.h | 2 +- kernel/bpf/bpf_cgrp_storage.c | 3 +-- kernel/bpf/bpf_inode_storage.c | 4 +--- kernel/bpf/bpf_local_storage.c | 8 +++++--- kernel/bpf/bpf_task_storage.c | 4 +--- net/core/bpf_sk_storage.c | 4 +--- 6 files changed, 10 insertions(+), 15 deletions(-)diff --git a/include/linux/bpf_local_storage.h b/include/linux/bpf_local_storage.h index 6cabf5154cf6..a94e12ddd83d 100644 --- a/include/linux/bpf_local_storage.h +++ b/include/linux/bpf_local_storage.h@@ -176,7 +176,7 @@ int bpf_local_storage_map_check_btf(const struct bpf_map *map, void bpf_selem_link_storage_nolock(struct bpf_local_storage *local_storage, struct bpf_local_storage_elem *selem); -void bpf_selem_unlink(struct bpf_local_storage_elem *selem, bool reuse_now); +int bpf_selem_unlink(struct bpf_local_storage_elem *selem, bool reuse_now); int bpf_selem_link_map(struct bpf_local_storage_map *smap, struct bpf_local_storage_elem *selem);diff --git a/kernel/bpf/bpf_cgrp_storage.c b/kernel/bpf/bpf_cgrp_storage.c index 0687a760974a..8fef24fcac68 100644 --- a/kernel/bpf/bpf_cgrp_storage.c +++ b/kernel/bpf/bpf_cgrp_storage.c@@ -118,8 +118,7 @@ static int cgroup_storage_delete(struct cgroup *cgroup, struct bpf_map *map) if (!sdata) return -ENOENT; - bpf_selem_unlink(SELEM(sdata), false); - return 0; + return bpf_selem_unlink(SELEM(sdata), false); } static long bpf_cgrp_storage_delete_elem(struct bpf_map *map, void *key)diff --git a/kernel/bpf/bpf_inode_storage.c b/kernel/bpf/bpf_inode_storage.c index e54cce2b9175..cedc99184dad 100644 --- a/kernel/bpf/bpf_inode_storage.c +++ b/kernel/bpf/bpf_inode_storage.c@@ -110,9 +110,7 @@ static int inode_storage_delete(struct inode *inode, struct bpf_map *map) if (!sdata) return -ENOENT; - bpf_selem_unlink(SELEM(sdata), false); - - return 0; + return bpf_selem_unlink(SELEM(sdata), false); } static long bpf_fd_inode_storage_delete_elem(struct bpf_map *map, void *key)diff --git a/kernel/bpf/bpf_local_storage.c b/kernel/bpf/bpf_local_storage.c index 0e3fa5fbaaf3..fa629a180e9e 100644 --- a/kernel/bpf/bpf_local_storage.c +++ b/kernel/bpf/bpf_local_storage.c@@ -367,7 +367,7 @@ static void bpf_selem_link_map_nolock(struct bpf_local_storage_map *smap, hlist_add_head_rcu(&selem->map_node, &b->list); } -void bpf_selem_unlink(struct bpf_local_storage_elem *selem, bool reuse_now) +int bpf_selem_unlink(struct bpf_local_storage_elem *selem, bool reuse_now) { struct bpf_local_storage *local_storage; bool free_local_storage = false;@@ -377,7 +377,7 @@ void bpf_selem_unlink(struct bpf_local_storage_elem *selem, bool reuse_now) if (unlikely(!selem_linked_to_storage_lockless(selem))) /* selem has already been unlinked from sk */ - return; + return 0; local_storage = rcu_dereference_check(selem->local_storage, bpf_rcu_lock_held());@@ -402,6 +402,8 @@ void bpf_selem_unlink(struct bpf_local_storage_elem *selem, bool reuse_now) if (free_local_storage) bpf_local_storage_free(local_storage, reuse_now); + + return err;
err is not used in patch 3 and then becomes useful in patch 4. The ai-review discovered issue on err also. Squash patch 4 into patch 3. It will be easier to read.
quoted hunk ↗ jump to hunk
} void __bpf_local_storage_insert_cache(struct bpf_local_storage *local_storage,@@ -837,7 +839,7 @@ void bpf_local_storage_map_free(struct bpf_map *map, struct bpf_local_storage_elem, map_node))) { if (busy_counter) this_cpu_inc(*busy_counter); - bpf_selem_unlink(selem, true); + WARN_ON(bpf_selem_unlink(selem, true));
nit. I would add __must_check to the needed functions in a single patch when everything is ready instead of having an intermediate WARN_ON here and then removed it later.