linux-kernel - Re: [PATCH 1/1] mm/page_owner: store page_owner's gfp

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: <24037235-2174-423f-9055-c6a49aa659e2@suse.cz>
Date:   Mon, 17 Jun 2019 15:51:32 +0200
From:   Vlastimil Babka <vbabka@...e.cz>
To:     Sai Charan Sane <s.charan@...sung.com>, akpm@...ux-foundation.org,
        mhocko@...e.com, tglx@...utronix.de, rppt@...ux.vnet.ibm.com,
        gregkh@...uxfoundation.org, joe@...ches.com,
        miles.chen@...iatek.com
Cc:     linux-mm@...ck.org, linux-kernel@...r.kernel.org,
        a.sahrawat@...sung.com, pankaj.m@...sung.com, v.narang@...sung.com
Subject: Re: [PATCH 1/1] mm/page_owner: store page_owner's gfp_mask in
 stackdepot itself

On 6/7/19 7:53 AM, Sai Charan Sane wrote:
> Memory overhead of 4MB is reduced by storing gfp_mask in stackdepot along
> with stacktrace. Stackdepot memory usage increased by ~100kb for 4GB of RAM.
> 
> Page owner logs from dmesg:
> 	Before patch:
> 		allocated 20971520 bytes of page_ext
> 	After patch:
> 		allocated 16777216 bytes of page_ext
> 
> Signed-off-by: Sai Charan Sane <s.charan@...sung.com>

I don't know, this looks like unneeded abuse to me. In the debug
scenario when someone boots a kernel with page_owner enabled, does 4MB
out of 4GB RAM really make a difference?