From: Andrii Nakryiko <andrii@kernel.org>
To: <dwarves@vger.kernel.org>, <acme@kernel.org>,
<bpf@vger.kernel.org>, <jolsa@kernel.org>
Cc: <andrii@kernel.org>, <kernel-team@fb.com>
Subject: [PATCH dwarves] btf_encoder: fix and complete filtering out zero-sized per-CPU variables
Date: Mon, 24 May 2021 16:42:22 -0700 [thread overview]
Message-ID: <20210524234222.278676-1-andrii@kernel.org> (raw)
btf_encoder is ignoring zero-sized per-CPU ELF symbols, but the same has to be
done for DWARF variables when matching them with ELF symbols. This is due to
zero-sized DWARF variables matching unrelated (non-zero-sized) variable that
happens to be allocated at the exact same address, leading to a lot of
confusion in BTF.
See [0] for when this causes big problems.
[0] https://lore.kernel.org/bpf/CAEf4BzZ0-sihSL-UAm21JcaCCY92CqfNxycHRZYXcoj8OYb=wA@mail.gmail.com/
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
---
btf_encoder.c | 13 ++++++++++---
1 file changed, 10 insertions(+), 3 deletions(-)
diff --git a/btf_encoder.c b/btf_encoder.c
index c711f124b31e..672b9943a4e2 100644
--- a/btf_encoder.c
+++ b/btf_encoder.c
@@ -538,6 +538,7 @@ int cu__encode_btf(struct cu *cu, int verbose, bool force,
cu__for_each_variable(cu, core_id, pos) {
uint32_t size, type, linkage;
const char *name, *dwarf_name;
+ const struct tag *tag;
uint64_t addr;
int id;
@@ -550,6 +551,7 @@ int cu__encode_btf(struct cu *cu, int verbose, bool force,
/* addr has to be recorded before we follow spec */
addr = var->ip.addr;
+ dwarf_name = variable__name(var, cu);
/* DWARF takes into account .data..percpu section offset
* within its segment, which for vmlinux is 0, but for kernel
@@ -582,11 +584,9 @@ int cu__encode_btf(struct cu *cu, int verbose, bool force,
* modules per-CPU data section has non-zero offset so all
* per-CPU symbols have non-zero values.
*/
- if (var->ip.addr == 0) {
- dwarf_name = variable__name(var, cu);
+ if (var->ip.addr == 0)
if (!dwarf_name || strcmp(dwarf_name, name))
continue;
- }
if (var->spec)
var = var->spec;
@@ -600,6 +600,13 @@ int cu__encode_btf(struct cu *cu, int verbose, bool force,
break;
}
+ tag = cu__type(cu, var->ip.tag.type);
+ if (tag__size(tag, cu) == 0) {
+ if (btf_elf__verbose)
+ fprintf(stderr, "Ignoring zero-sized per-CPU variable '%s'...\n", dwarf_name ?: "<missing name>");
+ continue;
+ }
+
type = var->ip.tag.type + type_id_off;
linkage = var->external ? BTF_VAR_GLOBAL_ALLOCATED : BTF_VAR_STATIC;
--
2.30.2
next reply other threads:[~2021-05-24 23:43 UTC|newest]
Thread overview: 10+ messages / expand[flat|nested] mbox.gz Atom feed top
2021-05-24 23:42 Andrii Nakryiko [this message]
2021-05-25 19:58 ` [PATCH dwarves] btf_encoder: fix and complete filtering out zero-sized per-CPU variables Arnaldo Carvalho de Melo
2021-05-27 14:55 ` Arnaldo Carvalho de Melo
2021-05-27 14:58 ` Arnaldo Carvalho de Melo
2021-05-27 15:27 ` Michal Suchánek
2021-05-27 15:38 ` Arnaldo Carvalho de Melo
2021-05-27 16:47 ` Arnaldo Carvalho de Melo
2021-05-28 6:24 ` Michal Suchánek
2021-05-27 16:45 ` Andrii Nakryiko
2021-05-27 16:43 ` 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=20210524234222.278676-1-andrii@kernel.org \
--to=andrii@kernel.org \
--cc=acme@kernel.org \
--cc=bpf@vger.kernel.org \
--cc=dwarves@vger.kernel.org \
--cc=jolsa@kernel.org \
--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;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).