linux-mips
[Top] [All Lists]

Re: [PATCH v2] MIPS: Kernel crashes on boot with SPARSEMEM + HIGHMEM ena

To: Michael Sundius <msundius@cisco.com>
Subject: Re: [PATCH v2] MIPS: Kernel crashes on boot with SPARSEMEM + HIGHMEM enabled
From: David Daney <ddaney@caviumnetworks.com>
Date: Fri, 01 Apr 2011 11:48:01 -0700
Cc: Kevin Cernekee <cernekee@gmail.com>, Ralf Baechle <ralf@linux-mips.org>, David VomLehn <dvomlehn@cisco.com>, Dave Hansen <dave@linux.vnet.ibm.com>, Andy Whitcroft <apw@shadowen.org>, Jon Fraser <jfraser@broadcom.com>, linux-mips@linux-mips.org, linux-kernel@vger.kernel.org, stable@kernel.org
In-reply-to: <4D961C6A.9070808@cisco.com>
Original-recipient: rfc822;linux-mips@linux-mips.org
References: <c300b67a7a723369872c0b9a023d0b2e@localhost> <4D9603D8.2010709@caviumnetworks.com> <4D961C6A.9070808@cisco.com>
Sender: linux-mips-bounce@linux-mips.org
User-agent: Mozilla/5.0 (X11; U; Linux x86_64; en-US; rv:1.9.1.15) Gecko/20101027 Fedora/3.0.10-1.fc12 Thunderbird/3.0.10
On 04/01/2011 11:41 AM, Michael Sundius wrote:
David Daney wrote:


I think this may do the same thing as my patch:

http://patchwork.linux-mips.org/patch/1988/

Although my patch had different motivations, and changes some other
things around too.

David Daney

I'm not really sure why your kernel or initrd would be in memory was not
within
the range that had been accounted for. are you saying its in high mem?


Well the memory initialization code has a bunch of weird rules built in that prevent some memory from being used.

For example if the kernel resides in a different SPARSE page than the rest of memory bad things happen because memory_present() was not called on something that is later freed (when init memory is released).

If I try to put an initrd at a high physical address, the memory below that is not usable.

My three patches try to make some sense out of the whole thing.

David Daney


<Prev in Thread] Current Thread [Next in Thread>