public inbox for linux-kernel@vger.kernel.org
 help / color / mirror / Atom feed
From: George Anzinger <george@mvista.com>
To: Oleg Nesterov <oleg@tv-sign.ru>
Cc: paulmck@us.ibm.com, Roland McGrath <roland@redhat.com>,
	akpm@osdl.org, linux-kernel@vger.kernel.org, dipankar@in.ibm.com,
	mingo@elte.hu, suzannew@cs.pdx.edu
Subject: [PATCH] sigaction should clear all signals on SIG_IGN, not just < 32
Date: Wed, 16 Nov 2005 15:26:21 -0800	[thread overview]
Message-ID: <437BC01D.60302@mvista.com> (raw)
In-Reply-To: <436E1401.920A83EE@tv-sign.ru>

[-- Attachment #1: Type: text/plain, Size: 488 bytes --]

While rooting aroung in the signal code trying to understand how to
fix the SIG_IGN	ploy (set sig handler to SIG_IGN and flood system with
high speed repeating timers) I came across what, I think, is a problem
in sigaction() in that when processing a SIG_IGN request it flushes
signals from 1 to SIGRTMIN and leaves the rest.  The attached patch is
an attempt to fix this.

-- 
George Anzinger   george@mvista.com
HRT (High-res-timers):  http://sourceforge.net/projects/high-res-timers/


[-- Attachment #2: sigaction-fix.patch --]
[-- Type: text/plain, Size: 3026 bytes --]

Source: MontaVista Software, Inc.
Type: Defect Fix 
Description:
It appeares that the sigaction system call, when processing a SIG_IGN or
a SIG_DFL, is removing only signals that appear in the first mask word.

Signed-off-by: George Anzinger <george@mvista.com>

 include/linux/signal.h |   16 ++++++++++++++++
 kernel/signal.c        |   34 ++++++++++++++++++++++++++++++++--
 2 files changed, 48 insertions(+), 2 deletions(-)

Index: linux-2.6.15-rc/kernel/signal.c
===================================================================
--- linux-2.6.15-rc.orig/kernel/signal.c
+++ linux-2.6.15-rc/kernel/signal.c
@@ -633,6 +633,33 @@ void signal_wake_up(struct task_struct *
  * Returns 1 if any signals were found.
  *
  * All callers must be holding the siglock.
+ *
+ * This version takes a sigset mask and looks at all signals,
+ * not just those in the first mask word.
+ */
+static int rm_from_queue_full(sigset_t *mask, struct sigpending *s)
+{
+	struct sigqueue *q, *n;
+	sigset_t m;
+
+	sigandsets(&m, mask, &s->signal);
+	if (sigisemptyset(&m))
+		return 0;
+
+	signandsets(&s->signal, &s->signal, mask);
+	list_for_each_entry_safe(q, n, &s->list, list) {
+		if (sigismember(mask, q->info.si_signo)) {
+			list_del_init(&q->list);
+			__sigqueue_free(q);
+		}
+	}
+	return 1;
+}
+/*
+ * Remove signals in mask from the pending set and queue.
+ * Returns 1 if any signals were found.
+ *
+ * All callers must be holding the siglock.
  */
 static int rm_from_queue(unsigned long mask, struct sigpending *s)
 {
@@ -2471,6 +2498,7 @@ int
 do_sigaction(int sig, const struct k_sigaction *act, struct k_sigaction *oact)
 {
 	struct k_sigaction *k;
+	sigset_t mask;
 
 	if (!valid_signal(sig) || sig < 1 || (act && sig_kernel_only(sig)))
 		return -EINVAL;
@@ -2518,9 +2546,11 @@ do_sigaction(int sig, const struct k_sig
 			*k = *act;
 			sigdelsetmask(&k->sa.sa_mask,
 				      sigmask(SIGKILL) | sigmask(SIGSTOP));
-			rm_from_queue(sigmask(sig), &t->signal->shared_pending);
+			sigemptyset(&mask);
+			sigaddset(&mask, sig);
+			rm_from_queue_full(&mask, &t->signal->shared_pending);
 			do {
-				rm_from_queue(sigmask(sig), &t->pending);
+				rm_from_queue_full(&mask, &t->pending);
 				recalc_sigpending_tsk(t);
 				t = next_thread(t);
 			} while (t != current);
Index: linux-2.6.15-rc/include/linux/signal.h
===================================================================
--- linux-2.6.15-rc.orig/include/linux/signal.h
+++ linux-2.6.15-rc/include/linux/signal.h
@@ -82,6 +82,22 @@ static inline int sigfindinword(unsigned
 
 #endif /* __HAVE_ARCH_SIG_BITOPS */
 
+static inline int sigisemptyset(sigset_t *set)
+{
+	extern void _NSIG_WORDS_is_unsupported_size(void);
+	switch (_NSIG_WORDS) {
+	case 4:
+		return (set->sig[3] | set->sig[2] |
+			set->sig[1] | set->sig[0]) == 0;
+	case 2:
+		return (set->sig[1] | set->sig[0]) == 0;
+	case 1:
+		return set->sig[0] == 0;
+	default:
+		_NSIG_WORDS_is_unsupported_size();
+	}
+}
+
 #define sigmask(sig)	(1UL << ((sig) - 1))
 
 #ifndef __HAVE_ARCH_SIG_SETOPS

  parent reply	other threads:[~2005-11-16 23:27 UTC|newest]

Thread overview: 16+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2005-11-05  1:36 [PATCH] Additional/catchup RCU signal fixes for -mm Paul E. McKenney
2005-11-05 16:32 ` Oleg Nesterov
2005-11-06  1:00   ` Paul E. McKenney
2005-11-06 14:17     ` Oleg Nesterov
2005-11-06 14:46       ` Oleg Nesterov
2005-11-06 23:02         ` Paul E. McKenney
2005-11-06 14:32     ` Posix timers vs exec problems Oleg Nesterov
2005-11-07 18:12       ` [PATCH] fix de_thread() vs send_group_sigqueue() race Oleg Nesterov
2005-11-08 20:36         ` Chris Wright
2005-11-08 20:55           ` Linus Torvalds
2005-11-16 23:26       ` George Anzinger [this message]
2005-11-22  1:09         ` Thread group exec race -> null pointer... HELP George Anzinger
2005-11-22 14:45           ` Oleg Nesterov
2005-11-23 20:30             ` George Anzinger
2005-11-25 15:03               ` Ingo Molnar
2005-11-22 19:20           ` [PATCH] fix do_wait() vs exec() race Oleg Nesterov

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=437BC01D.60302@mvista.com \
    --to=george@mvista.com \
    --cc=akpm@osdl.org \
    --cc=dipankar@in.ibm.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=mingo@elte.hu \
    --cc=oleg@tv-sign.ru \
    --cc=paulmck@us.ibm.com \
    --cc=roland@redhat.com \
    --cc=suzannew@cs.pdx.edu \
    /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