linux-kernel - HR timers:

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>] [day] [month] [year] [list]

Date:	Fri, 11 Jul 2008 14:26:54 +0200 (CEST)
From:	Michal Simek <Monstr@...nam.cz>
To:	Linux Kernel list <linux-kernel@...r.kernel.org>,
	mingo@...e.hu
Subject: HR timers: 

Hi Ingo,

I debug Microblaze kernel and I found one big problem which comes with your patch.
Here is more detail about patch (latest patch linux/kernel/git/torvalds/linux-2.6.git/include/linux/seqlock.h)

author	Ingo Molnar <mingo@...e.hu>
	Thu, 3 Apr 2008 07:06:13 +0000 (09:06 +0200)
committer	Ingo Molnar <mingo@...e.hu>
	Thu, 24 Apr 2008 22:25:08 +0000 (00:25 +0200)
commit	88a411c07b6fedcfc97b8dc51ae18540bd2beda0

You add there checking mechanism to (read_seqbegin and read_seqcount_begin) where you are checking
odd value in ret variable. I don't know yet why I got there odd  value (0x3553). What does it mean the odd bit here?
I read any comment about tick_oneshot code. 

Could you explain me what is broken?

Thanks for your help,
Michal Simek

 static __always_inline unsigned read_seqbegin(const seqlock_t *sl)
 {
-       unsigned ret = sl->sequence;
+       unsigned ret;
+
+repeat:
+       ret = sl->sequence;
        smp_rmb();
+       if (unlikely(ret & 1)) {
+               cpu_relax();
+               goto repeat;
+       }
+
        return ret;
 }
--
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/