linux-kernel - Re: [PATCH 2/2] x86, boot: Optimize the elf header handling.

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]

Date:	Sun, 01 Jul 2012 08:04:33 -0700
From:	"H. Peter Anvin" <hpa@...or.com>
To:	"Eric W. Biederman" <ebiederm@...ssion.com>
CC:	hacklu <embedway.linux@...il.com>, linux-kernel@...r.kernel.org
Subject: Re: [PATCH 2/2] x86, boot: Optimize the elf header handling.

On 06/03/2012 01:32 PM, Eric W. Biederman wrote:
>
> Create a space for the elf headers at the begginng of the
> kernels image in memory.  Removing the need for an extra
> copy of the kernel during boot.  Making things faster
> and making vmlinux smaller.
>
> - Allow room for the elf headers in the vmlinux.
>    This removes the need to insert padding between
>    the elf headers and the start of text.  Reducing
>    the size of vmlinux by 2MB on x86_64 and removing
>    the need for parse_elf in boot.c to move the code.
>
> - Return the relocated entry point address from parse_elf
>    as the kernel's entry point is no long at a fixed address.
>
> - Remove the now unnecessary copies in arch/x86/compress/misc.c:parse_elf
>    The ELF headers are now guaranteed to not conflict with the program
>    data in the uncompressed image.
>
> - In cleanup_highmap keep all pages starting with __START_KERNEL_map
>    instead of _text.  Those values used to be the same but with the
>    insertion of the hole for the ELF headers they differ and cause us
>    to nuke our first 2MB of text ouch!  So use the __START_KERNEL_map
>    which includes the elf headers.
>

Sadly, this patch breaks the boot with of of my standard configurations 
(attached)... it goes into an infinite reboot loop.  This was the first 
config I tested, so it might very well break more.

	-hpa

-- 
H. Peter Anvin, Intel Open Source Technology Center
I work for Intel.  I don't speak on their behalf.




View attachment "config" of type "text/plain" (83418 bytes)