linux-kernel - Re: [PATCH?] uprobes: change uprobe_write

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: <529E5E3A.9070401@zytor.com>
Date:	Tue, 03 Dec 2013 14:42:02 -0800
From:	"H. Peter Anvin" <hpa@...or.com>
To:	Oleg Nesterov <oleg@...hat.com>
CC:	Linus Torvalds <torvalds@...ux-foundation.org>,
	Hugh Dickins <hughd@...gle.com>,
	Peter Zijlstra <peterz@...radead.org>,
	Jiri Kosina <jkosina@...e.cz>,
	Andi Kleen <andi@...stfloor.org>,
	Linux Kernel Mailing List <linux-kernel@...r.kernel.org>,
	the arch/x86 maintainers <x86@...nel.org>,
	Andi Kleen <ak@...ux.intel.com>,
	Ingo Molnar <mingo@...nel.org>, Borislav Petkov <bp@...en8.de>
Subject: Re: [PATCH?] uprobes: change uprobe_write_opcode() to modify the
 page directly

On 12/03/2013 12:54 PM, Oleg Nesterov wrote:
> On 12/03, H. Peter Anvin wrote:
>>
>> There is no architecture-independent way to make code globally visible.
> 
> Then I probably misunderstood "globally". Or something else.
> 
> So do you think the patch I sent is wrong? Why?
> 

It is wrong in the sense that without an architecture-specific
synchronization at the end it is not guaranteed to work.  It will work
fine on x86 and presumably any other architecture where icache-dcache
coherency is enforced in hardware (I am *guessing* that includes all or
most SMP platforms, but it is definitely *not* true on all UP platforms.)

	-hpa


--
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/