From: Amery Hung <ameryhung@gmail.com>
To: bpf@vger.kernel.org
Cc: alexei.starovoitov@gmail.com, andrii@kernel.org,
daniel@iogearbox.net, eddyz87@gmail.com, memxor@gmail.com,
ameryhung@gmail.com, kernel-team@meta.com
Subject: [PATCH bpf-next v2 09/23] bpf: Set OBJ_RELEASE when generating kfunc argument types
Date: Fri, 11 Sep 2026 15:04:01 -0700 [thread overview]
Message-ID: <20260911220415.1396439-10-ameryhung@gmail.com> (raw)
In-Reply-To: <20260911220415.1396439-1-ameryhung@gmail.com>
check_kfunc_args() ORs OBJ_RELEASE into arg_type on every verification
of every argument of every call, from a comparison against
meta->release_regno. Neither side of that comparison depends on the
call site. bpf_fetch_kfunc_arg_meta() sets release_regno to BPF_REG_1
when the kfunc is KF_RELEASE and leaves it zero otherwise, and regno
is derived from the argument index, so the test is true exactly for
argument 0 of a KF_RELEASE kfunc.
Set the flag in get_kfunc_arg_type() instead, where the rest
of the classification is built, and drop the comparison. The release
handling in check_func_proto() is helper-only, so an OBJ_RELEASE in a
generated kfunc prototype does not feed back into meta->release_regno.
meta->release_regno itself stays: it is still what tells the referenced
PTR_TO_BTF_ID check and release_reg() which register to act on.
No functional change.
Signed-off-by: Amery Hung <ameryhung@gmail.com>
---
kernel/bpf/verifier.c | 9 +++++++--
1 file changed, 7 insertions(+), 2 deletions(-)
diff --git a/kernel/bpf/verifier.c b/kernel/bpf/verifier.c
index c7822addfc64..20e7da6674a1 100644
--- a/kernel/bpf/verifier.c
+++ b/kernel/bpf/verifier.c
@@ -12311,6 +12311,13 @@ get_kfunc_arg_type(struct bpf_verifier_env *env, struct bpf_call_arg_meta *meta,
if (is_kfunc_arg_nullable(meta->btf, &args[arg]))
arg_type |= PTR_MAYBE_NULL;
+ /*
+ * Only the first argument of a KF_RELEASE kfunc releases anything, and
+ * bpf_fetch_kfunc_arg_meta() only ever records BPF_REG_1 for it.
+ */
+ if (is_kfunc_release(meta) && arg == 0)
+ arg_type |= OBJ_RELEASE;
+
return arg_type;
}
@@ -12997,8 +13004,6 @@ static int check_kfunc_args(struct bpf_verifier_env *env, struct bpf_call_arg_me
ref_tname = btf_name_by_offset(btf, ref_t->name_off);
}
- if (regno == meta->release_regno)
- arg_type |= OBJ_RELEASE;
ret = check_func_arg_reg_off(env, reg, argno, arg_type);
if (ret < 0)
return ret;
--
2.52.0
next prev parent reply other threads:[~2026-09-11 22:04 UTC|newest]
Thread overview: 28+ messages / expand[flat|nested] mbox.gz Atom feed top
2026-09-11 22:03 [PATCH bpf-next v2 00/23] Unify helper and kfunc argument checks Amery Hung
2026-09-11 22:03 ` [PATCH bpf-next v2 01/23] bpf: Pass call metadata through shared " Amery Hung
2026-09-11 22:03 ` [PATCH bpf-next v2 02/23] bpf: Address check_func_arg() arguments by argno Amery Hung
2026-09-11 22:03 ` [PATCH bpf-next v2 03/23] bpf: Only compare func_id against BPF_FUNC_* for helper calls Amery Hung
2026-09-11 22:03 ` [PATCH bpf-next v2 04/23] bpf: Only compare func_id against kfunc BTF IDs for kfunc calls Amery Hung
2026-09-11 22:03 ` [PATCH bpf-next v2 05/23] bpf: Clarify unused and scalar function argument types Amery Hung
2026-09-11 22:19 ` sashiko-bot
2026-09-11 22:03 ` [PATCH bpf-next v2 06/23] bpf: Unify kfunc argument kinds with enum bpf_arg_type Amery Hung
2026-09-11 22:26 ` sashiko-bot
2026-09-11 22:03 ` [PATCH bpf-next v2 07/23] bpf: Classify kfunc arguments the verifier ignores Amery Hung
2026-09-11 22:04 ` [PATCH bpf-next v2 08/23] bpf: Align helper and kfunc ARG_PTR_TO_PROG_AUX handling Amery Hung
2026-09-11 22:04 ` Amery Hung [this message]
2026-09-11 22:04 ` [PATCH bpf-next v2 10/23] bpf: Set MEM_UNINIT and dynptr subtypes when generating kfunc arg types Amery Hung
2026-09-11 22:04 ` [PATCH bpf-next v2 11/23] bpf: Set MEM_RCU when generating kfunc argument types Amery Hung
2026-09-11 22:04 ` [PATCH bpf-next v2 12/23] bpf: Resolve BTF ID of ARG_PTR_TO_BTF_ID in kfunc bpf_func_proto Amery Hung
2026-09-11 22:04 ` [PATCH bpf-next v2 13/23] bpf: Resolve ARG_PTR_TO_MEM | MEM_FIXED_SIZE size " Amery Hung
2026-09-11 22:04 ` [PATCH bpf-next v2 14/23] bpf: Consolidate runtime argument type resolution Amery Hung
2026-09-11 22:04 ` [PATCH bpf-next v2 15/23] bpf: Consolidate nullable argument validation Amery Hung
2026-09-11 22:04 ` [PATCH bpf-next v2 16/23] bpf: Drop redundant BTF pointer helper write rejection Amery Hung
2026-09-11 22:04 ` [PATCH bpf-next v2 17/23] bpf: Consolidate helper and kfunc PTR_TO_BTF_ID argument matching Amery Hung
2026-09-11 22:04 ` [PATCH bpf-next v2 18/23] bpf: Admit kfunc argument registers through check_reg_type() Amery Hung
2026-09-11 22:04 ` [PATCH bpf-next v2 19/23] bpf: Consolidate function call pkt_access validation Amery Hung
2026-09-11 22:04 ` [PATCH bpf-next v2 20/23] selftests/bpf: Test kfunc packet memory direct writes Amery Hung
2026-09-11 22:36 ` sashiko-bot
2026-09-11 22:04 ` [PATCH bpf-next v2 21/23] bpf: Consolidate release argument validation Amery Hung
2026-09-11 22:04 ` [PATCH bpf-next v2 22/23] selftests/bpf: Test nullable per-CPU kptr identity after exchange Amery Hung
2026-09-11 22:04 ` [PATCH bpf-next v2 23/23] bpf: Check helper and kfunc arguments in one path Amery Hung
2026-09-12 3:20 ` [PATCH bpf-next v2 00/23] Unify helper and kfunc argument checks patchwork-bot+netdevbpf
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20260911220415.1396439-10-ameryhung@gmail.com \
--to=ameryhung@gmail.com \
--cc=alexei.starovoitov@gmail.com \
--cc=andrii@kernel.org \
--cc=bpf@vger.kernel.org \
--cc=daniel@iogearbox.net \
--cc=eddyz87@gmail.com \
--cc=kernel-team@meta.com \
--cc=memxor@gmail.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox