linux-kernel - Re: [PATCH] mm/vmalloc: Fix regression caused by needless vmalloc_sync

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

Date:   Thu, 14 Nov 2019 18:14:43 +0100
From:   Joerg Roedel <jroedel@...e.de>
To:     Shile Zhang <shile.zhang@...ux.alibaba.com>
Cc:     Michal Hocko <mhocko@...nel.org>,
        Andrew Morton <akpm@...ux-foundation.org>, linux-mm@...ck.org,
        linux-kernel@...r.kernel.org
Subject: Re: [PATCH] mm/vmalloc: Fix regression caused by needless
 vmalloc_sync_all()

On Thu, Nov 14, 2019 at 10:40:22PM +0800, Shile Zhang wrote:
> Could you please help to recall the original issue you encountered before?

The original issue was data corruption because old mappings did not get
removed from the vmalloc area, and thus also new mappings did not get
faulted in. So depending on the page-table currently loaded one
vmalloc/ioremap area pointed to different (often already freed or
re-used) memory and caused the corruption.

Regards,

	Joerg