linux-kernel - Re: [RFC] seqlock,lockdep: Add lock primitives to read

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-Id: <201103290657.JEG17111.VOtOFJFLSQFMOH@I-love.SAKURA.ne.jp>
Date:	Tue, 29 Mar 2011 06:57:41 +0900
From:	Tetsuo Handa <penguin-kernel@...ove.SAKURA.ne.jp>
To:	rostedt@...dmis.org
Cc:	peterz@...radead.org, mingo@...hat.com,
	linux-kernel@...r.kernel.org
Subject: Re: [RFC] seqlock,lockdep: Add lock primitives to read_seqbegin().

Steven Rostedt wrote:
> This is where you confused me. Why would a pattern that was previously
> checked, suddenly break?  If we found A->B and then come across B->A,
> lockdep should surely catch it.

I don't know. This bug can be triggered by simply running

  while :; do newns /sbin/pivot_root /proc/ /proc/sys/; done

and

  while :; do /bin/ls -l /proc/*/exe; done

in parallel. However, lockdep can't catch it even after applying

--- linux-2.6.38.1.orig/include/linux/seqlock.h
+++ linux-2.6.38.1/include/linux/seqlock.h
@@ -86,6 +86,11 @@ static inline int write_tryseqlock(seqlo
 static __always_inline unsigned read_seqbegin(const seqlock_t *sl)
 {
 	unsigned ret;
+#ifdef CONFIG_PROVE_LOCKING
+	unsigned long flags;
+	spin_lock_irqsave(&((seqlock_t *) sl)->lock, flags);
+	spin_unlock_irqrestore(&((seqlock_t *) sl)->lock, flags);
+#endif
 
 repeat:
 	ret = sl->sequence;

. Maybe usage of spin_lock_irqsave()/spin_unlock_irqrestore() is bad.

> Maybe I don't understand exactly whay you are trying to say.

Just apply the patch and run the 2 loops above in parallel using 2.6.38.1
will show you. Source code for newns is shown below.

---------- newns.c start ----------
#include <stdio.h>
#include <sys/types.h>
#include <sys/wait.h>
#include <unistd.h>
#include <sched.h>
#include <errno.h>
#include <stdlib.h>

static int child(void *arg)
{
	char **argv = (char **) arg;
	argv++;
	execvp(argv[0], argv);
	_exit(1);
}

int main(int argc, char *argv[])
{
	char c = 0;
	char *stack = malloc(8192);
	const pid_t pid = clone(child, stack + (8192 / 2), CLONE_NEWNS,
				(void *) argv);
	while (waitpid(pid, NULL, __WALL) == EOF && errno == EINTR)
		c++; /* Dummy. */
	return 0;
}
---------- newns.c end ----------

Regards.
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/