From: Florian Weimer <fweimer@redhat.com>
To: libc-alpha@sourceware.org
Cc: levon@movementarian.org
Subject: [PATCH] resolv: Track single-request fallback via _res._flags (bug 31476)
Date: Mon, 10 Jun 2024 11:57:01 +0200 [thread overview]
Message-ID: <87ikyhqfwy.fsf@oldenburg.str.redhat.com> (raw)
This avoids changing _res.options, which inteferes with change
detection as part of automatic reloading of /etc/resolv.conf.
Tested on x86_64-linux-gnu with the existing tests. Not sure how to
write a good regression test for this.
---
resolv/res_send.c | 12 +++++++-----
resolv/resolv-internal.h | 2 ++
2 files changed, 9 insertions(+), 5 deletions(-)
diff --git a/resolv/res_send.c b/resolv/res_send.c
index fb0217204a..ea7cf192b2 100644
--- a/resolv/res_send.c
+++ b/resolv/res_send.c
@@ -947,9 +947,11 @@ send_dg(res_state statp,
seconds /= statp->nscount;
if (seconds <= 0)
seconds = 1;
- bool single_request_reopen = (statp->options & RES_SNGLKUPREOP) != 0;
- bool single_request = (((statp->options & RES_SNGLKUP) != 0)
- | single_request_reopen);
+ bool single_request_reopen = ((statp->options & RES_SNGLKUPREOP)
+ || (statp->_flags & RES_F_SNGLKUPREOP));
+ bool single_request = ((statp->options & RES_SNGLKUP)
+ || (statp->_flags & RES_F_SNGLKUP)
+ || single_request_reopen);
int save_gotsomewhere = *gotsomewhere;
int retval;
@@ -1006,14 +1008,14 @@ send_dg(res_state statp,
have received the first answer. */
if (!single_request)
{
- statp->options |= RES_SNGLKUP;
+ statp->_flags |= RES_F_SNGLKUP;
single_request = true;
*gotsomewhere = save_gotsomewhere;
goto retry;
}
else if (!single_request_reopen)
{
- statp->options |= RES_SNGLKUPREOP;
+ statp->_flags |= RES_F_SNGLKUPREOP;
single_request_reopen = true;
*gotsomewhere = save_gotsomewhere;
__res_iclose (statp, false);
diff --git a/resolv/resolv-internal.h b/resolv/resolv-internal.h
index 24b164f6b5..944af3ee76 100644
--- a/resolv/resolv-internal.h
+++ b/resolv/resolv-internal.h
@@ -26,6 +26,8 @@
#define RES_F_VC 0x00000001 /* Socket is TCP. */
#define RES_F_CONN 0x00000002 /* Socket is connected. */
#define RES_F_EDNS0ERR 0x00000004 /* EDNS0 caused errors. */
+#define RES_F_SNGLKUP 0x00200000 /* Private version of RES_SNGLKUP. */
+#define RES_F_SNGLKUPREOP 0x00400000 /* Private version of RES_SNGLKUPREOP. */
/* The structure HEADER is normally aligned on a word boundary. In
some code, we need to access this structure when it may be aligned
base-commit: 54c1efdac55b756a4e2ea95590bcc3ba396a6568
next reply other threads:[~2024-06-10 9:57 UTC|newest]
Thread overview: 2+ messages / expand[flat|nested] mbox.gz Atom feed top
2024-06-10 9:57 Florian Weimer [this message]
2024-06-12 20:27 ` DJ Delorie
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=87ikyhqfwy.fsf@oldenburg.str.redhat.com \
--to=fweimer@redhat.com \
--cc=levon@movementarian.org \
--cc=libc-alpha@sourceware.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 read-only IMAP folder(s) and NNTP newsgroup(s).