public inbox for [email protected]
 help / color / mirror / Atom feed
From: Ammar Faizi <[email protected]>
To: GNU/Weeb Mailing List <[email protected]>
Cc: Ammar Faizi <[email protected]>, Arthur Lapz <[email protected]>
Subject: [PATCH teavpn2] server: entry: Fix a false-positive warning from GCC 12.1.1
Date: Thu,  1 Sep 2022 20:49:54 +0700	[thread overview]
Message-ID: <[email protected]> (raw)

@rlapz reported that he is seeing a dangling pointer warning when
building TeaVPN2 with GCC 12.1.1 [1]:

  src/teavpn2/server/entry.c: In function ‘run_server’:
  src/teavpn2/server/entry.c:455:18: warning: storing the address of \
  local variable ‘cfg’ in ‘data_dir’ [-Wdangling-pointer=]
    455 |         data_dir = cfg.sys.data_dir;
        |         ~~~~~~~~~^~~~~~~~~~~~~~~~~~

This is a false-positive warning. GCC doesn't really know the real
program flow at runtime. Reasoning why GCC is wrong:

  1) The lifetime of cfg.sys.data dir ends when run_server() in entry.c
     returns.

  2) @data_dir is only used in teavpn2_auth() in auth.c.

  3) teavpn2_auth() is only called in the run_server() call stack.

That being said, when @data_dir var is used, it never be a dangling
pointer. Just to make GCC happy on this, set @data_dir to NULL before
the function returns, this way GCC assumption about dangling pointer
should be proven wrong at compile time. @rlapz confirmed it fixes the
warning.

Link: https://t.me/GNUWeeb/662168 [1]
Reported-by: Arthur Lapz <[email protected]>
Tested-by: Arthur Lapz <[email protected]>
Signed-off-by: Ammar Faizi <[email protected]>
---
 src/teavpn2/server/entry.c | 8 ++++++--
 1 file changed, 6 insertions(+), 2 deletions(-)

diff --git a/src/teavpn2/server/entry.c b/src/teavpn2/server/entry.c
index fd9bc85..6f47ef9 100644
--- a/src/teavpn2/server/entry.c
+++ b/src/teavpn2/server/entry.c
@@ -455,9 +455,13 @@ __cold int run_server(int argc, char *argv[])
 	data_dir = cfg.sys.data_dir;
 	switch (cfg.sock.type) {
 	case SOCK_UDP:
-		return -teavpn2_server_udp_run(&cfg);
+		ret = -teavpn2_server_udp_run(&cfg);
+		break;
 	case SOCK_TCP:
 	default:
-		return ESOCKTNOSUPPORT;
+		ret = ESOCKTNOSUPPORT;
+		break;
 	}
+	data_dir = NULL;
+	return ret;
 }

base-commit: cfee418df11d3b1c67ed96da25807462f68b3c9c
-- 
Ammar Faizi


                 reply	other threads:[~2022-09-01 13:50 UTC|newest]

Thread overview: [no followups] expand[flat|nested]  mbox.gz  Atom feed

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 \
    [email protected] \
    [email protected] \
    [email protected] \
    [email protected] \
    /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