From: Eduard Zingerman <eddyz87@gmail.com>
To: bpf@vger.kernel.org, ast@kernel.org, andrii@kernel.org
Cc: daniel@iogearbox.net, martin.lau@linux.dev, kernel-team@fb.com,
yonghong.song@linux.dev, eddyz87@gmail.com,
Alexei Starovoitov <alexei.starovoitov@gmail.com>
Subject: [PATCH bpf-next v2 2/2] bpf: use realloc in bpf_patch_insn_data
Date: Wed, 6 Aug 2025 18:02:05 -0700 [thread overview]
Message-ID: <20250807010205.3210608-3-eddyz87@gmail.com> (raw)
In-Reply-To: <20250807010205.3210608-1-eddyz87@gmail.com>
Avoid excessive vzalloc/vfree calls when patching instructions in
do_misc_fixups(). bpf_patch_insn_data() uses vzalloc to allocate new
memory for env->insn_aux_data for each patch as follows:
struct bpf_prog *bpf_patch_insn_data(env, ...)
{
...
new_data = vzalloc(... O(program size) ...);
...
adjust_insn_aux_data(env, new_data, ...);
...
}
void adjust_insn_aux_data(env, new_data, ...)
{
...
memcpy(new_data, env->insn_aux_data);
vfree(env->insn_aux_data);
env->insn_aux_data = new_data;
...
}
The vzalloc/vfree pair is hot in perf report collected for e.g.
pyperf180 test case. It can be replaced with a call to vrealloc in
order to reduce the number of actual memory allocations.
This is a stop-gap solution, as bpf_patch_insn_data is still hot in
the profile. More comprehansive solutions had been discussed before
e.g. as in [1].
[1] https://lore.kernel.org/bpf/CAEf4BzY_E8MSL4mD0UPuuiDcbJhh9e2xQo2=5w+ppRWWiYSGvQ@mail.gmail.com/
Suggested-by: Alexei Starovoitov <alexei.starovoitov@gmail.com>
Signed-off-by: Eduard Zingerman <eddyz87@gmail.com>
---
kernel/bpf/verifier.c | 29 +++++++++++++++--------------
1 file changed, 15 insertions(+), 14 deletions(-)
diff --git a/kernel/bpf/verifier.c b/kernel/bpf/verifier.c
index 69eb2b5c2218..a61d57996692 100644
--- a/kernel/bpf/verifier.c
+++ b/kernel/bpf/verifier.c
@@ -20699,12 +20699,11 @@ static void convert_pseudo_ld_imm64(struct bpf_verifier_env *env)
* [0, off) and [off, end) to new locations, so the patched range stays zero
*/
static void adjust_insn_aux_data(struct bpf_verifier_env *env,
- struct bpf_insn_aux_data *new_data,
struct bpf_prog *new_prog, u32 off, u32 cnt)
{
- struct bpf_insn_aux_data *old_data = env->insn_aux_data;
+ struct bpf_insn_aux_data *data = env->insn_aux_data;
struct bpf_insn *insn = new_prog->insnsi;
- u32 old_seen = old_data[off].seen;
+ u32 old_seen = data[off].seen;
u32 prog_len;
int i;
@@ -20712,22 +20711,20 @@ static void adjust_insn_aux_data(struct bpf_verifier_env *env,
* (cnt == 1) is taken or not. There is no guarantee INSN at OFF is the
* original insn at old prog.
*/
- old_data[off].zext_dst = insn_has_def32(insn + off + cnt - 1);
+ data[off].zext_dst = insn_has_def32(insn + off + cnt - 1);
if (cnt == 1)
return;
prog_len = new_prog->len;
- memcpy(new_data, old_data, sizeof(struct bpf_insn_aux_data) * off);
- memcpy(new_data + off + cnt - 1, old_data + off,
- sizeof(struct bpf_insn_aux_data) * (prog_len - off - cnt + 1));
+ memmove(data + off + cnt - 1, data + off,
+ sizeof(struct bpf_insn_aux_data) * (prog_len - off - cnt + 1));
+ memset(data + off, 0, sizeof(struct bpf_insn_aux_data) * (cnt - 1));
for (i = off; i < off + cnt - 1; i++) {
/* Expand insni[off]'s seen count to the patched range. */
- new_data[i].seen = old_seen;
- new_data[i].zext_dst = insn_has_def32(insn + i);
+ data[i].seen = old_seen;
+ data[i].zext_dst = insn_has_def32(insn + i);
}
- env->insn_aux_data = new_data;
- vfree(old_data);
}
static void adjust_subprog_starts(struct bpf_verifier_env *env, u32 off, u32 len)
@@ -20765,10 +20762,14 @@ static struct bpf_prog *bpf_patch_insn_data(struct bpf_verifier_env *env, u32 of
struct bpf_insn_aux_data *new_data = NULL;
if (len > 1) {
- new_data = vzalloc(array_size(env->prog->len + len - 1,
- sizeof(struct bpf_insn_aux_data)));
+ new_data = vrealloc(env->insn_aux_data,
+ array_size(env->prog->len + len - 1,
+ sizeof(struct bpf_insn_aux_data)),
+ GFP_KERNEL_ACCOUNT | __GFP_ZERO);
if (!new_data)
return NULL;
+
+ env->insn_aux_data = new_data;
}
new_prog = bpf_patch_insn_single(env->prog, off, patch, len);
@@ -20780,7 +20781,7 @@ static struct bpf_prog *bpf_patch_insn_data(struct bpf_verifier_env *env, u32 of
vfree(new_data);
return NULL;
}
- adjust_insn_aux_data(env, new_data, new_prog, off, len);
+ adjust_insn_aux_data(env, new_prog, off, len);
adjust_subprog_starts(env, off, len);
adjust_poke_descs(new_prog, off, len);
return new_prog;
--
2.47.3
next prev parent reply other threads:[~2025-08-07 1:02 UTC|newest]
Thread overview: 7+ messages / expand[flat|nested] mbox.gz Atom feed top
2025-08-07 1:02 [PATCH bpf-next v2 0/2] bpf: use vrealloc() in bpf_patch_insn_data() Eduard Zingerman
2025-08-07 1:02 ` [PATCH bpf-next v2 1/2] bpf: removed unused 'env' parameter from is_reg64 and insn_has_def32 Eduard Zingerman
2025-08-07 15:26 ` Kumar Kartikeya Dwivedi
2025-08-07 1:02 ` Eduard Zingerman [this message]
2025-08-07 16:00 ` [PATCH bpf-next v2 2/2] bpf: use realloc in bpf_patch_insn_data Kumar Kartikeya Dwivedi
2025-08-07 16:02 ` Anton Protopopov
2025-08-07 16:20 ` [PATCH bpf-next v2 0/2] bpf: use vrealloc() in bpf_patch_insn_data() 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=20250807010205.3210608-3-eddyz87@gmail.com \
--to=eddyz87@gmail.com \
--cc=alexei.starovoitov@gmail.com \
--cc=andrii@kernel.org \
--cc=ast@kernel.org \
--cc=bpf@vger.kernel.org \
--cc=daniel@iogearbox.net \
--cc=kernel-team@fb.com \
--cc=martin.lau@linux.dev \
--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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.