From: Yonghong Song <yonghong.song@linux.dev>
To: Alexei Starovoitov <alexei.starovoitov@gmail.com>, bpf@vger.kernel.org
Cc: Andrii Nakryiko <andrii@kernel.org>,
Daniel Borkmann <daniel@iogearbox.net>,
Eduard Zingerman <eddyz87@gmail.com>,
kernel-team@fb.com
Subject: Re: [PATCH bpf-next v3 10/11] bpf, x86: Give a kfunc's stack argument the 16-byte alignment it needs
Date: Fri, 16 Oct 2026 08:45:55 -0700 [thread overview]
Message-ID: <ace9e0d5-0357-4401-be7a-b77ddca86f0b@linux.dev> (raw)
In-Reply-To: <DM5X2Z5473A5.DNWH4JDJTYJF@gmail.com>
On 10/15/26 7:46 PM, Alexei Starovoitov wrote:
> On Tue, Oct 13, 2026 at 10:10 AM Yonghong Song <yonghong.song@linux.dev> wrote:
>> The JIT places a kfunc's stack arguments at the offsets from RSP that
>> the convention gives them, holes included, but RSP itself is only 8-byte
>> aligned in a BPF program, so a kfunc can get such an argument at an
>> address that is not 16-byte aligned.
> There are no kfuncs that take __int128 on the stack.
> The kernel is built with -mno-sse, so nothing but cmpxchg16b on &g
> cares about this alignment.
Indeed, with -mno-sse, the stack base whether it is 16-byte align or
not probably does not matter.
> Can we drop this patch?
> because...
>
>> + for (i = 0; i < n; i++)
>> + EMIT4(0xFF, 0x74, 0x24, disp); /* push qword ptr [rsp + disp] */
>> + skip[-1] = prog - skip;
> u64 kfunc(u64 a, u64 b, u64 c, u64 d, u64 e,
> __int128 x, __int128 y, __int128 z, u64 w)
> takes 133 bytes total. BPF_MAX_INSN_SIZE is 128.
> do_jit() prints "bpf_jit: fatal insn size error" and the prog
> fails to load. It was 92 bytes before this patch.
Yes, will drop this commit and the next self test which is to demonstrate
this commit.
>
> pw-bot: cr
next prev parent reply other threads:[~2026-10-16 15:46 UTC|newest]
Thread overview: 20+ messages / expand[flat|nested] mbox.gz Atom feed top
2026-10-13 17:09 [PATCH bpf-next v3 00/11] bpf: Follow up by-value struct and __int128 argument support Yonghong Song
2026-10-13 17:09 ` [PATCH bpf-next v3 01/11] bpf, x86: Pass a trampoline's stack arguments by the calling convention Yonghong Song
2026-10-13 17:09 ` [PATCH bpf-next v3 02/11] selftests/bpf: Test trampoline stack args in a program that can tail call Yonghong Song
2026-10-13 17:09 ` [PATCH bpf-next v3 03/11] selftests/bpf: Test by-value __int128 kfunc arguments at an odd slot Yonghong Song
2026-10-13 17:09 ` [PATCH bpf-next v3 04/11] bpf: Describe a BPF function to its trampoline as one u64 per slot Yonghong Song
2026-10-13 17:10 ` [PATCH bpf-next v3 05/11] bpf, x86: Place trampoline arguments by the x86-64 calling convention Yonghong Song
2026-10-13 17:10 ` [PATCH bpf-next v3 06/11] selftests/bpf: Test trampoline args around a 16-byte aligned stack arg Yonghong Song
2026-10-13 17:10 ` [PATCH bpf-next v3 07/11] selftests/bpf: Test trampoline args of a BPF function at an odd slot Yonghong Song
2026-10-13 17:10 ` [PATCH bpf-next v3 08/11] bpf: Refuse a trampoline on an unreliable static function with stack args Yonghong Song
2026-10-13 17:10 ` [PATCH bpf-next v3 09/11] selftests/bpf: Test trampolines on more BPF functions " Yonghong Song
2026-10-13 17:10 ` [PATCH bpf-next v3 10/11] bpf, x86: Give a kfunc's stack argument the 16-byte alignment it needs Yonghong Song
2026-10-16 2:46 ` Alexei Starovoitov
2026-10-16 15:45 ` Yonghong Song [this message]
2026-10-13 17:10 ` [PATCH bpf-next v3 11/11] selftests/bpf: Test that a kfunc gets its __int128 stack argument aligned Yonghong Song
2026-10-16 2:48 ` [PATCH bpf-next v3 00/11] bpf: Follow up by-value struct and __int128 argument support Alexei Starovoitov
2026-10-16 15:50 ` Yonghong Song
2026-10-16 16:00 ` Alexei Starovoitov
2026-10-16 16:22 ` Yonghong Song
2026-10-16 16:34 ` Alexei Starovoitov
2026-10-16 16:56 ` Yonghong Song
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=ace9e0d5-0357-4401-be7a-b77ddca86f0b@linux.dev \
--to=yonghong.song@linux.dev \
--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@fb.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