From: Qiliang Yuan <odys.yuan@gmail.com>
To: Alexei Starovoitov <ast@kernel.org>,
Daniel Borkmann <daniel@iogearbox.net>,
John Fastabend <john.fastabend@gmail.com>,
Andrii Nakryiko <andrii@kernel.org>,
Eduard Zingerman <eddyz87@gmail.com>,
Kumar Kartikeya Dwivedi <memxor@gmail.com>,
Martin KaFai Lau <martin.lau@linux.dev>,
Song Liu <song@kernel.org>,
Yonghong Song <yonghong.song@linux.dev>,
Jiri Olsa <jolsa@kernel.org>,
Emil Tsalapatis <emil@etsalapatis.com>,
Ihor Solodrai <ihor.solodrai@linux.dev>,
Shuah Khan <shuah@kernel.org>
Cc: bpf@vger.kernel.org, linux-kernel@vger.kernel.org,
linux-kselftest@vger.kernel.org,
Qiliang Yuan <odys.yuan@gmail.com>
Subject: [PATCH bpf-next v2 19/22] bpf: Blind constants through the patch list
Date: Fri, 30 Oct 2026 07:52:04 -0400 [thread overview]
Message-ID: <20261030-bpf-verifier-patch-batch-v2-19-55cd43439f7c@gmail.com> (raw)
In-Reply-To: <20261030-bpf-verifier-patch-batch-v2-0-55cd43439f7c@gmail.com>
bpf_jit_blind_constants() patches the program for every instruction
whose constant it blinds, with bpf_patch_insn_data() under the verifier
and bpf_patch_insn_single() otherwise.
Each of these patches costs O(prog->len). With bpf_jit_harden=2,
loading the program of a million instructions from test_verifier's
"ld_dw: xor semi-random 64 bit imms, test 5" doesn't finish in 15
minutes.
Queue the rewrites and commit them once, through the verifier env when
there is one and with bpf_patch_list_apply() otherwise. Add stubs of
the verifier env helpers that core.c calls, for builds without the
verifier.
With bpf_jit_harden=2, perf stat -r10 -- veristat -q <obj> averaged
over two runs, and test 5 timed once, x86_64 VM with 32 vCPUs, before
and after this patch:
before after
pyperf180 3.173 s 0.368 s
pyperf600 9.793 s 0.907 s
test 5 > 900 s 5 s
Signed-off-by: Qiliang Yuan <odys.yuan@gmail.com>
---
include/linux/bpf_verifier.h | 17 +++++++++++
kernel/bpf/core.c | 73 +++++++++++++++++++++++---------------------
2 files changed, 56 insertions(+), 34 deletions(-)
diff --git a/include/linux/bpf_verifier.h b/include/linux/bpf_verifier.h
index b133bbaf614ac..cf93d473e4f0c 100644
--- a/include/linux/bpf_verifier.h
+++ b/include/linux/bpf_verifier.h
@@ -2075,10 +2075,27 @@ int bpf_jit_subprogs(struct bpf_verifier_env *env);
int bpf_fixup_call_args(struct bpf_verifier_env *env);
int bpf_do_misc_fixups(struct bpf_verifier_env *env);
int bpf_insn_def32(struct bpf_prog *prog, struct bpf_insn *insn);
+#ifdef CONFIG_BPF_SYSCALL
struct bpf_insn *bpf_patch_list_add(struct bpf_verifier_env *env, u32 off,
const struct bpf_insn *patch, u32 len);
int bpf_patch_list_commit(struct bpf_verifier_env *env);
void bpf_patch_list_free(struct bpf_verifier_env *env);
+#else
+static inline struct bpf_insn *bpf_patch_list_add(struct bpf_verifier_env *env, u32 off,
+ const struct bpf_insn *patch, u32 len)
+{
+ return ERR_PTR(-EOPNOTSUPP);
+}
+
+static inline int bpf_patch_list_commit(struct bpf_verifier_env *env)
+{
+ return -EOPNOTSUPP;
+}
+
+static inline void bpf_patch_list_free(struct bpf_verifier_env *env)
+{
+}
+#endif
int bpf_patch_list_del(struct bpf_verifier_env *env, u32 off);
int bpf_patch_list_insert(struct bpf_verifier_env *env, u32 off,
const struct bpf_insn *insns, u32 cnt);
diff --git a/kernel/bpf/core.c b/kernel/bpf/core.c
index 10d2c37f81280..7d58c47b73f07 100644
--- a/kernel/bpf/core.c
+++ b/kernel/bpf/core.c
@@ -1949,11 +1949,11 @@ void bpf_jit_prog_release_other(struct bpf_prog *fp, struct bpf_prog *fp_other)
*/
struct bpf_prog *bpf_jit_blind_constants(struct bpf_verifier_env *env, struct bpf_prog *prog)
{
- struct bpf_insn insn_buff[16], aux[2];
+ struct bpf_insn insn_buff[16], aux[2], *last;
+ struct bpf_patch_list pl = {};
struct bpf_prog *clone, *tmp;
- int insn_delta, insn_cnt;
struct bpf_insn *insn;
- int i, rewritten;
+ int i, insn_cnt, rewritten, err = 0;
if (WARN_ON_ONCE(env && env->prog != prog))
return ERR_PTR(-EINVAL);
@@ -1962,7 +1962,7 @@ struct bpf_prog *bpf_jit_blind_constants(struct bpf_verifier_env *env, struct bp
if (!clone)
return ERR_PTR(-ENOMEM);
- /* make sure bpf_patch_insn_data() patches the correct prog */
+ /* make sure the patches go to the clone */
if (env)
env->prog = clone;
@@ -1993,41 +1993,46 @@ struct bpf_prog *bpf_jit_blind_constants(struct bpf_verifier_env *env, struct bp
if (!rewritten)
continue;
- if (rewritten < 0)
- tmp = ERR_PTR(rewritten);
- else if (env)
- tmp = bpf_patch_insn_data(env, i, insn_buff, rewritten);
- else
- tmp = bpf_patch_insn_single(clone, i, insn_buff, rewritten);
-
- if (IS_ERR_OR_NULL(tmp)) {
- if (env)
- /* restore the original prog */
- env->prog = prog;
- /* Patching may have repointed aux->prog during
- * realloc from the original one, so we need to
- * fix it up here on error.
- */
- bpf_jit_prog_release_other(prog, clone);
- if (env && fatal_signal_pending(current))
- return ERR_PTR(-EINTR);
- return IS_ERR(tmp) ? tmp : ERR_PTR(-ENOMEM);
+ if (rewritten < 0) {
+ err = rewritten;
+ break;
}
+ last = env ? bpf_patch_list_add(env, i, insn_buff, rewritten) :
+ __bpf_patch_list_add(&pl, clone, i, insn_buff, rewritten);
+ if (IS_ERR(last)) {
+ err = PTR_ERR(last);
+ break;
+ }
+ }
- clone = tmp;
- insn_delta = rewritten - 1;
-
- if (env)
- env->prog = clone;
+ if (!err && env) {
+ err = bpf_patch_list_commit(env);
+ tmp = env->prog;
+ } else if (!err) {
+ tmp = bpf_patch_list_apply(clone, &pl);
+ err = PTR_ERR_OR_ZERO(tmp);
+ }
- /* Walk new program and skip insns we just inserted. */
- insn = clone->insnsi + i + insn_delta;
- insn_cnt += insn_delta;
- i += insn_delta;
+ if (err) {
+ if (env) {
+ bpf_patch_list_free(env);
+ /* restore the original prog */
+ clone = env->prog;
+ env->prog = prog;
+ }
+ bpf_patch_list_release(&pl);
+ /* Patching may have repointed aux->prog during
+ * realloc from the original one, so we need to
+ * fix it up here on error.
+ */
+ bpf_jit_prog_release_other(prog, clone);
+ if (env && fatal_signal_pending(current))
+ return ERR_PTR(-EINTR);
+ return ERR_PTR(err);
}
- clone->blinded = 1;
- return clone;
+ tmp->blinded = 1;
+ return tmp;
}
bool bpf_insn_is_indirect_target(const struct bpf_verifier_env *env, const struct bpf_prog *prog,
--
2.43.0
next prev parent reply other threads:[~2026-10-30 11:53 UTC|newest]
Thread overview: 31+ messages / expand[flat|nested] mbox.gz Atom feed top
2026-10-30 11:51 [PATCH bpf-next v2 00/22] bpf: Replace bpf_patch_insn_data() with a list of deferred patches Qiliang Yuan
2026-10-30 11:51 ` [PATCH bpf-next v2 01/22] bpf: Add a list of deferred instruction patches Qiliang Yuan
2026-10-30 11:51 ` [PATCH bpf-next v2 02/22] bpf: Remap the verifier state through the patch list Qiliang Yuan
2026-10-30 12:44 ` bot+bpf-ci
2026-10-30 11:51 ` [PATCH bpf-next v2 03/22] bpf: Apply single-instruction patches of the verifier in place Qiliang Yuan
2026-10-30 11:51 ` [PATCH bpf-next v2 04/22] bpf: Queue the rewrites of bpf_do_misc_fixups() Qiliang Yuan
2026-10-30 11:51 ` [PATCH bpf-next v2 05/22] bpf: Expand a backward may_goto to gotol when patches push it out of range Qiliang Yuan
2026-10-30 12:59 ` bot+bpf-ci
2026-10-30 11:51 ` [PATCH bpf-next v2 06/22] bpf: Drop the constant delta from bpf_do_misc_fixups() Qiliang Yuan
2026-10-30 11:51 ` [PATCH bpf-next v2 07/22] bpf: Let the patch list remove instructions Qiliang Yuan
2026-10-30 12:11 ` sashiko-bot
2026-10-30 11:51 ` [PATCH bpf-next v2 08/22] bpf: Remove dead code through the patch list Qiliang Yuan
2026-10-30 11:51 ` [PATCH bpf-next v2 09/22] bpf: Strip nops " Qiliang Yuan
2026-10-30 12:45 ` bot+bpf-ci
2026-10-30 11:51 ` [PATCH bpf-next v2 10/22] bpf: Batch the patches of bpf_opt_subreg_zext_lo32_rnd_hi32() Qiliang Yuan
2026-10-30 11:51 ` [PATCH bpf-next v2 11/22] bpf: Allow patching an instruction twice through the patch list Qiliang Yuan
2026-10-30 11:51 ` [PATCH bpf-next v2 12/22] bpf: Teach the patch list to insert in front of an instruction Qiliang Yuan
2026-10-30 12:44 ` bot+bpf-ci
2026-10-30 11:51 ` [PATCH bpf-next v2 13/22] bpf: Allow inserting in front of a queued patch Qiliang Yuan
2026-10-30 11:51 ` [PATCH bpf-next v2 14/22] bpf: Move bpf_convert_ctx_accesses() to the patch list Qiliang Yuan
2026-10-30 12:24 ` sashiko-bot
2026-10-30 11:52 ` [PATCH bpf-next v2 15/22] bpf: Insert the may_goto counter setup through " Qiliang Yuan
2026-10-30 11:52 ` [PATCH bpf-next v2 16/22] bpf: Append the hidden subprog " Qiliang Yuan
2026-10-30 11:52 ` [PATCH bpf-next v2 17/22] bpf: Inline bpf_loop() calls " Qiliang Yuan
2026-10-30 11:52 ` [PATCH bpf-next v2 18/22] bpf: Apply the patch list without a verifier env Qiliang Yuan
2026-10-30 11:52 ` Qiliang Yuan [this message]
2026-10-30 12:44 ` [PATCH bpf-next v2 19/22] bpf: Blind constants through the patch list bot+bpf-ci
2026-10-30 11:52 ` [PATCH bpf-next v2 20/22] bpf: Remove the old instruction patching code Qiliang Yuan
2026-10-30 11:52 ` [PATCH bpf-next v2 21/22] selftests/bpf: Test jumps around patches of bpf_do_misc_fixups() Qiliang Yuan
2026-10-30 11:52 ` [PATCH bpf-next v2 22/22] selftests/bpf: Cover a backward may_goto pushed out of range by patches Qiliang Yuan
2026-10-30 12:13 ` sashiko-bot
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=20261030-bpf-verifier-patch-batch-v2-19-55cd43439f7c@gmail.com \
--to=odys.yuan@gmail.com \
--cc=andrii@kernel.org \
--cc=ast@kernel.org \
--cc=bpf@vger.kernel.org \
--cc=daniel@iogearbox.net \
--cc=eddyz87@gmail.com \
--cc=emil@etsalapatis.com \
--cc=ihor.solodrai@linux.dev \
--cc=john.fastabend@gmail.com \
--cc=jolsa@kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-kselftest@vger.kernel.org \
--cc=martin.lau@linux.dev \
--cc=memxor@gmail.com \
--cc=shuah@kernel.org \
--cc=song@kernel.org \
--cc=yonghong.song@linux.dev \
/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