Mirror of the gdb-patches mailing list
 help / color / mirror / Atom feed
From: Pedro Alves <pedro@palves.net>
To: gdb-patches@sourceware.org
Subject: [PATCH 3/4] scoped_ignore_signal: Use sigprocmask+sigtimedwait instead of signal
Date: Tue, 15 Jun 2021 12:14:28 +0100	[thread overview]
Message-ID: <20210615111429.1879286-4-pedro@palves.net> (raw)
In-Reply-To: <20210615111429.1879286-1-pedro@palves.net>

The problem with using signal(...) to temporarily ignore a signal, is
that that changes the the signal disposition for the whole process.
If multiple threads do it at the same time, you have a race.

Fix this by using sigprocmask + sigtimedwait to implement the ignoring
instead, if available, which I think probably means everywhere except
Windows nowadays.  This way, we only change the signal mask for the
current thread, so there's no race.

Change-Id: Idfe3fb08327ef8cae926f3de9ee81c56a83b1738

gdbsupport/ChangeLog:
yyyy-mm-dd  Pedro Alves  <pedro@palves.net>

	* scoped_ignore_signal.h
	(scoped_ignore_signal::scoped_ignore_signal)
	[HAVE_SIGPROCMASK]: Use sigprocmask to block the signal instead of
	changing the signal disposition for the whole process.
	(scoped_ignore_signal::~scoped_ignore_signal) [HAVE_SIGPROCMASK]:
	Use sigtimedwait and sigprocmask to flush and unblock the signal.
---
 gdbsupport/scoped_ignore_signal.h | 37 +++++++++++++++++++++++++++++--
 1 file changed, 35 insertions(+), 2 deletions(-)

diff --git a/gdbsupport/scoped_ignore_signal.h b/gdbsupport/scoped_ignore_signal.h
index cccd390529b..55a921cb332 100644
--- a/gdbsupport/scoped_ignore_signal.h
+++ b/gdbsupport/scoped_ignore_signal.h
@@ -22,7 +22,10 @@
 
 #include <signal.h>
 
-/* RAII class used to ignore a signal in a scope.  */
+/* RAII class used to ignore a signal in a scope.  If sigprocmask is
+   supported, then the signal is only ignored by the calling thread.
+   Otherwise, the signal disposition is set to SIG_IGN, which affects
+   the whole process.  */
 
 template <int Sig>
 class scoped_ignore_signal
@@ -30,18 +33,48 @@ class scoped_ignore_signal
 public:
   scoped_ignore_signal ()
   {
+#ifdef HAVE_SIGPROCMASK
+    sigset_t set, old_state;
+
+    sigemptyset (&set);
+    sigaddset (&set, Sig);
+    sigprocmask (SIG_BLOCK, &set, &old_state);
+    m_was_blocked = sigismember (&old_state, Sig);
+#else
     m_osig = signal (Sig, SIG_IGN);
+#endif
   }
 
   ~scoped_ignore_signal ()
   {
+#ifdef HAVE_SIGPROCMASK
+    if (!m_was_blocked)
+      {
+	sigset_t set;
+	const timespec zero_timeout = {};
+
+	sigemptyset (&set);
+	sigaddset (&set, Sig);
+
+	/* If we got a pending Sig signal, consume it before
+	   unblocking.  */
+	sigtimedwait (&set, nullptr, &zero_timeout);
+
+	sigprocmask (SIG_UNBLOCK, &set, nullptr);
+      }
+#else
     signal (Sig, m_osig);
+#endif
   }
 
   DISABLE_COPY_AND_ASSIGN (scoped_ignore_signal);
 
 private:
-  sighandler_t m_osig = nullptr;
+#ifdef HAVE_SIGPROCMASK
+  bool m_was_blocked;
+#else
+  sighandler_t m_osig;
+#endif
 };
 
 struct scoped_ignore_signal_nop
-- 
2.26.2


  parent reply	other threads:[~2021-06-15 11:15 UTC|newest]

Thread overview: 19+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-06-15 11:14 [PATCH 0/4] Introduce scoped_ignore_signal & make it thread safe Pedro Alves
2021-06-15 11:14 ` [PATCH 1/4] Move scoped_ignore_sigttou to gdbsupport/ Pedro Alves
2021-06-15 11:14 ` [PATCH 2/4] Introduce scoped_restore_signal Pedro Alves
2021-06-17 14:57   ` Pedro Alves
2021-06-15 11:14 ` Pedro Alves [this message]
2021-06-26 12:35   ` [PATCH 3/4] scoped_ignore_signal: Use sigprocmask+sigtimedwait instead of signal Simon Marchi via Gdb-patches
2021-06-26 13:41     ` John Baldwin
2021-06-26 19:10       ` Simon Marchi via Gdb-patches
2021-06-27 14:55       ` Pedro Alves
2021-06-27 19:01         ` Simon Marchi via Gdb-patches
2021-06-15 11:14 ` [PATCH 4/4] Add a unit test for scoped_ignore_sigpipe Pedro Alves
2021-06-15 22:26   ` Lancelot SIX via Gdb-patches
2021-06-17 13:00     ` Pedro Alves
2021-06-15 17:04 ` [PATCH 0/4] Introduce scoped_ignore_signal & make it thread safe John Baldwin
2021-06-17 14:38   ` Pedro Alves
2021-06-17 15:36     ` Pedro Alves
2021-06-17 16:45       ` John Baldwin
2021-06-17 18:44         ` [pushed] Don't call sigtimedwait for scoped_ignore_sigttou Pedro Alves
2021-06-15 20:16 ` [PATCH 0/4] Introduce scoped_ignore_signal & make it thread safe Tom Tromey

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=20210615111429.1879286-4-pedro@palves.net \
    --to=pedro@palves.net \
    --cc=gdb-patches@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