From: Alexei Starovoitov <alexei.starovoitov@gmail.com>
To: Song Liu <songliubraving@fb.com>
Cc: Ilya Leoshkevich <iii@linux.ibm.com>, Song Liu <song@kernel.org>,
bpf <bpf@vger.kernel.org>,
Network Development <netdev@vger.kernel.org>,
LKML <linux-kernel@vger.kernel.org>,
Alexei Starovoitov <ast@kernel.org>,
Daniel Borkmann <daniel@iogearbox.net>,
Andrii Nakryiko <andrii@kernel.org>,
Kernel Team <Kernel-team@fb.com>,
Peter Zijlstra <peterz@infradead.org>, X86 ML <x86@kernel.org>
Subject: Re: [PATCH v6 bpf-next 6/7] bpf: introduce bpf_prog_pack allocator
Date: Mon, 24 Jan 2022 21:21:28 -0800 [thread overview]
Message-ID: <CAADnVQKgdMMeONmjUhbq_3X39t9HNQWteDuyWVfcxmTerTnaMw@mail.gmail.com> (raw)
In-Reply-To: <2AAC8B8C-96F1-400F-AFA6-D4AF41EC82F4@fb.com>
On Mon, Jan 24, 2022 at 10:27 AM Song Liu <songliubraving@fb.com> wrote:
> >
> > Are arches expected to allocate rw buffers in different ways? If not,
> > I would consider putting this into the common code as well. Then
> > arch-specific code would do something like
> >
> > header = bpf_jit_binary_alloc_pack(size, &prg_buf, &prg_addr, ...);
> > ...
> > /*
> > * Generate code into prg_buf, the code should assume that its first
> > * byte is located at prg_addr.
> > */
> > ...
> > bpf_jit_binary_finalize_pack(header, prg_buf);
> >
> > where bpf_jit_binary_finalize_pack() would copy prg_buf to header and
> > free it.
It feels right, but bpf_jit_binary_finalize_pack() sounds 100% arch
dependent. The only thing it will do is perform a copy via text_poke.
What else?
> I think this should work.
>
> We will need an API like: bpf_arch_text_copy, which uses text_poke_copy()
> for x86_64 and s390_kernel_write() for x390. We will use bpf_arch_text_copy
> to
> 1) write header->size;
> 2) do finally copy in bpf_jit_binary_finalize_pack().
we can combine all text_poke operations into one.
Can we add an 'image' pointer into struct bpf_binary_header ?
Then do:
int bpf_jit_binary_alloc_pack(size, &ro_hdr, &rw_hdr);
ro_hdr->image would be the address used to compute offsets by JIT.
rw_hdr->image would point to kvmalloc-ed area for emitting insns.
rw_hdr->size would already be populated.
The JITs would write insns into rw_hdr->image including 'int 3' insns.
At the end the JIT will do text_poke_copy(ro_hdr, rw_hdr, rw_hdr->size);
That would be the only copy that will transfer everything into final
location.
Then kvfree(rw_hdr)
wdyt?
next prev parent reply other threads:[~2022-01-25 6:55 UTC|newest]
Thread overview: 31+ messages / expand[flat|nested] mbox.gz Atom feed top
2022-01-21 19:49 [PATCH v6 bpf-next 0/7] bpf_prog_pack allocator Song Liu
2022-01-21 19:49 ` [PATCH v6 bpf-next 1/7] x86/Kconfig: select HAVE_ARCH_HUGE_VMALLOC with HAVE_ARCH_HUGE_VMAP Song Liu
2022-01-21 19:49 ` [PATCH v6 bpf-next 2/7] bpf: use bytes instead of pages for bpf_jit_[charge|uncharge]_modmem Song Liu
2022-01-21 19:49 ` [PATCH v6 bpf-next 3/7] bpf: use size instead of pages in bpf_binary_header Song Liu
2022-01-21 19:49 ` [PATCH v6 bpf-next 4/7] bpf: add a pointer of bpf_binary_header to bpf_prog Song Liu
2022-01-21 19:49 ` [PATCH v6 bpf-next 5/7] x86/alternative: introduce text_poke_copy Song Liu
2022-01-21 19:49 ` [PATCH v6 bpf-next 6/7] bpf: introduce bpf_prog_pack allocator Song Liu
2022-01-21 23:55 ` Alexei Starovoitov
2022-01-22 0:23 ` Song Liu
2022-01-22 0:41 ` Alexei Starovoitov
2022-01-22 1:01 ` Song Liu
2022-01-22 1:12 ` Alexei Starovoitov
2022-01-22 1:30 ` Song Liu
2022-01-22 2:12 ` Alexei Starovoitov
2022-01-23 1:03 ` Song Liu
2022-01-24 12:29 ` Ilya Leoshkevich
2022-01-24 18:27 ` Song Liu
2022-01-25 5:21 ` Alexei Starovoitov [this message]
2022-01-25 7:21 ` Song Liu
2022-01-25 19:59 ` Alexei Starovoitov
2022-01-25 22:25 ` Song Liu
2022-01-25 22:48 ` Alexei Starovoitov
2022-01-25 23:09 ` Song Liu
2022-01-26 0:38 ` Alexei Starovoitov
2022-01-26 0:50 ` Song Liu
2022-01-26 1:20 ` Alexei Starovoitov
2022-01-26 1:28 ` Song Liu
2022-01-26 1:31 ` Song Liu
2022-01-26 1:34 ` Alexei Starovoitov
2022-01-24 12:45 ` Peter Zijlstra
2022-01-21 19:49 ` [PATCH v6 bpf-next 7/7] bpf, x86_64: use " Song Liu
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=CAADnVQKgdMMeONmjUhbq_3X39t9HNQWteDuyWVfcxmTerTnaMw@mail.gmail.com \
--to=alexei.starovoitov@gmail.com \
--cc=Kernel-team@fb.com \
--cc=andrii@kernel.org \
--cc=ast@kernel.org \
--cc=bpf@vger.kernel.org \
--cc=daniel@iogearbox.net \
--cc=iii@linux.ibm.com \
--cc=linux-kernel@vger.kernel.org \
--cc=netdev@vger.kernel.org \
--cc=peterz@infradead.org \
--cc=song@kernel.org \
--cc=songliubraving@fb.com \
--cc=x86@kernel.org \
/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;
as well as URLs for NNTP newsgroup(s).