linux-kernel - preempt bug in set_pmd

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-next>] [day] [month] [year] [list]

Message-ID: <47CDBB87.8090906@goop.org>
Date:	Tue, 04 Mar 2008 13:13:43 -0800
From:	Jeremy Fitzhardinge <jeremy@...p.org>
To:	Ingo Molnar <mingo@...e.hu>, "H. Peter Anvin" <hpa@...or.com>,
	Andi Kleen <ak@...e.de>
CC:	Linux Kernel Mailing List <linux-kernel@...r.kernel.org>
Subject: preempt bug in set_pmd_pfn?

I think set_pmd_pfn, which is only called by __set_fixmap, might have a 
preempt bug in it.

It can be executed with preemption enabled, but what if it gets preempted

	set_pmd(pmd, pfn_pmd(pfn, flags));
	/*
	 * It's enough to flush this one mapping.
	 * (PGE mappings get flushed as well)
	 */
>here<
	__flush_tlb_one(vaddr);
}

?

Won't this leave a stale tlb on the old processor?

I noticed this because the Xen tlb flushing code effectively has a 
smp_processor_id(), which provokes a warning when preemption is 
enabled.  It seems to me that it never makes sense to be doing a tlb 
flush unless you know which processor you're actually running on...

    J
--
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/