Re: [patch 2/3] Add flags parameter to reserve_bootmem_generic()

From: Vivek Goyal <vgoyal@redhat.com>
To: Bernhard Walle <bwalle@suse.de>
Cc: kexec@lists.infradead.org, linux-kernel@vger.kernel.org,
	Johannes Weiner <hannes@saeurebad.de>,
	Andi Kleen <andi@firstfloor.org>,
	anderson@redhat.com, hpa@zytor.com, "Romer,
	Benjamin M" <Benjamin.Romer@unisys.com>,
	Amul Shah <amul.shah@unisys.com>
Subject: Re: [patch 2/3] Add flags parameter to reserve_bootmem_generic()
Date: Mon, 9 Jun 2008 17:00:55 -0400	[thread overview]
Message-ID: <20080609210055.GK3542@redhat.com> (raw)
In-Reply-To: <20080609225734.2bb9ee64@kopernikus.site>

On Mon, Jun 09, 2008 at 10:57:34PM +0200, Bernhard Walle wrote:
> * Vivek Goyal <vgoyal@redhat.com> [2008-06-09 16:54]:
> >
> > On Mon, Jun 09, 2008 at 10:42:11PM +0200, Bernhard Walle wrote:
> > > Hi Vivek,
> > > 
> > > * Vivek Goyal <vgoyal@redhat.com> [2008-06-09 16:29]:
> > > >
> > > > Can you please put some more explanation comment here to explain that
> > > > why it is ok to return with success, despite the fact that we never
> > > > reserved any memory.
> > > 
> > > Do you think that's ok?
> > > 
> > >     When booting the kdump kernel, the MP tables (for example,
> > >     or other firmware-reserved memory) of the BIOS are beyond
> > >     end_pfn. (kexec-tools adds exactmap parameters to the kernel
> > >     so that the E820 map is no longer used.) Therefore, it's ok
> > >     to return "success" here. For normal boot, the MP tables
> > >     must be reserved normally.
> > > 
> > 
> > Looks good. Thanks
> 
> Should I resend the whole patch series to Linus for inclusion?
> 

I am really not sure. :-). In the past, Andrew picked up the patches, kept
in his tree for couple of days and if something is important, he will push
it out to Linus.

Thanks
Vivek

_______________________________________________
kexec mailing list
kexec@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/kexec