bpf.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Andrii Nakryiko <andrii@kernel.org>
To: <bpf@vger.kernel.org>, <netdev@vger.kernel.org>, <ast@fb.com>,
	<daniel@iogearbox.net>
Cc: <andrii@kernel.org>, <kernel-team@fb.com>
Subject: [RFC PATCH bpf-next 4/7] bpftool: stop emitting static variables in BPF skeleton
Date: Thu, 6 May 2021 22:41:16 -0700	[thread overview]
Message-ID: <20210507054119.270888-5-andrii@kernel.org> (raw)
In-Reply-To: <20210507054119.270888-1-andrii@kernel.org>

As discussed in [0], stop emitting static variables in BPF skeletons to avoid
issues with name-conflicting static variables across multiple
statically-linked BPF object files.

Users using static variables to pass data between BPF programs and user-space
should do a trivial one-time switch according to the following simple rules:
  - read-only `static volatile const` variables should be converted to
    `volatile const`;
  - read/write `static volatile` variables should just drop `static volatile`
    modifiers to become global variables/symbols. To better handle older Clang
    versions, such newly converted global variables should be explicitly
    initialized with a specific value or `= 0`/`= {}`, whichever is
    appropriate.

  [0] https://lore.kernel.org/bpf/CAEf4BzZo7_r-hsNvJt3w3kyrmmBJj7ghGY8+k4nvKF0KLjma=w@mail.gmail.com/T/#m664d4b0d6b31ac8b2669360e0fc2d6962e9f5ec1

Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
---
 tools/bpf/bpftool/gen.c | 4 ++++
 1 file changed, 4 insertions(+)

diff --git a/tools/bpf/bpftool/gen.c b/tools/bpf/bpftool/gen.c
index 06fee4a2910a..27dceaf66ecb 100644
--- a/tools/bpf/bpftool/gen.c
+++ b/tools/bpf/bpftool/gen.c
@@ -131,6 +131,10 @@ static int codegen_datasec_def(struct bpf_object *obj,
 		int need_off = sec_var->offset, align_off, align;
 		__u32 var_type_id = var->type;
 
+		/* static variables are not exposed through BPF skeleton */
+		if (btf_var(var)->linkage == BTF_VAR_STATIC)
+			continue;
+
 		if (off > need_off) {
 			p_err("Something is wrong for %s's variable #%d: need offset %d, already at %d.\n",
 			      sec_name, i, need_off, off);
-- 
2.30.2


  parent reply	other threads:[~2021-05-07  5:41 UTC|newest]

Thread overview: 8+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-05-07  5:41 [RFC PATCH bpf-next 0/7] BPF static linker: global symbols visibility Andrii Nakryiko
2021-05-07  5:41 ` [RFC PATCH bpf-next 1/7] bpftool: strip const/volatile/restrict modifiers from .bss and .data vars Andrii Nakryiko
2021-05-07  5:41 ` [RFC PATCH bpf-next 2/7] libbpf: add per-file linker opts Andrii Nakryiko
2021-05-07  5:41 ` [RFC PATCH bpf-next 3/7] selftests/bpf: stop using static variables for passing data to/from user-space Andrii Nakryiko
2021-05-07  5:41 ` Andrii Nakryiko [this message]
2021-05-07  5:41 ` [RFC PATCH bpf-next 5/7] libbpf: fix ELF symbol visibility update logic Andrii Nakryiko
2021-05-07  5:41 ` [RFC PATCH bpf-next 6/7] libbpf: treat STV_INTERNAL same as STV_HIDDEN for functions Andrii Nakryiko
2021-05-07  5:41 ` [RFC PATCH bpf-next 7/7] libbpf: convert STV_HIDDEN symbols into STV_INTERNAL after linking Andrii Nakryiko

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=20210507054119.270888-5-andrii@kernel.org \
    --to=andrii@kernel.org \
    --cc=ast@fb.com \
    --cc=bpf@vger.kernel.org \
    --cc=daniel@iogearbox.net \
    --cc=kernel-team@fb.com \
    --cc=netdev@vger.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).