linux-kernel - Re: [PATCH v2] x86/setup: Merge {early_,}reserve

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 for Android: free password hash cracker in your pocket

[<prev] [next>] [<thread-prev] [day] [month] [year] [list]

Message-ID: <20160209121016.GD500@gmail.com>
Date:	Tue, 9 Feb 2016 13:10:16 +0100
From:	Ingo Molnar <mingo@...nel.org>
To:	Alexander Kuleshov <kuleshovmail@...il.com>
Cc:	Ingo Molnar <mingo@...hat.com>,
	Thomas Gleixner <tglx@...utronix.de>,
	"H . Peter Anvin" <hpa@...or.com>, Borislav Petkov <bp@...e.de>,
	Joerg Roedel <jroedel@...e.de>, Dave Young <dyoung@...hat.com>,
	Andrew Morton <akpm@...ux-foundation.org>,
	Jiri Kosina <jkosina@...e.cz>, Baoquan He <bhe@...hat.com>,
	Paolo Bonzini <pbonzini@...hat.com>,
	Mark Salter <msalter@...hat.com>,
	LKML <linux-kernel@...r.kernel.org>,
	the arch/x86 maintainers <x86@...nel.org>
Subject: Re: [PATCH v2] x86/setup: Merge {early_,}reserve_initrd() to one
 function


* Alexander Kuleshov <kuleshovmail@...il.com> wrote:

> Hello Ingo,
> 
> On Tue, Feb 9, 2016 at 3:16 PM, Ingo Molnar <mingo@...nel.org> wrote:
> >
> > So I don't think the code got easier to understand - in particular the
> > memblock_reserve()/free() pattern, depending on a flag value, is confusing.
> >
> > The duplication is there - but please factor it out into a helper structure
> > ('struct ramdisk') and a helper function that sets up the structure.
> 
> What if instead of `struct ramdisk`, we will move all definitions/check from
> the early_reserve_initrd to the setup_arch() and than will pass these values
> to the reserve_initrd()?

There's too many of them, putting them into 'struct ramdisk' cleans up and 
documents the whole code.

And yes, 'struct ramdisk' state should be defined in setup_arch(), that way it 
does not have to be calculated twice: so the patch becomes not just a code size 
reduction but a (small) runtime reduction as well.

Should be tested with a real ramdisk, to make sure everything still works fine.

Thanks,

	Ingo