linux-kernel - Re: [PATCH] Prevent reserving RAM in the region already reserved by BIOS

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:	Wed, 19 May 2010 17:01:41 -0700
From:	Mathieu Rondonneau <mathieu.rondonneau@...il.com>
To:	"H. Peter Anvin" <hpa@...or.com>
Cc:	Yinghai <yinghai.lu@...cle.com>, linux-kernel@...r.kernel.org,
	linux-arch@...r.kernel.org, Ingo Molnar <mingo@...e.hu>,
	Linus Torvalds <torvalds@...ux-foundation.org>
Subject: Re: [PATCH] Prevent reserving RAM in the region already reserved by 
	BIOS

it' s not in ISA space.
once loading the NVIDIA driver, the warnign oops shows up.
4K starting at 0x9f800 (i.e. to 0xa007ff) overlap with the ISA space
(starting at 0xa0000).

When I don' t load the driver, no oops.
So I am assuming nvidia driver request 4K of memory, that happens to
be available in the bios area.

I think that there is a check missing somewhere to report that
available RAM buffer is already reserved (by BIOS) so we need to get
it from somewhere else.

Does it make sense?
-Mathieu

On Wed, May 19, 2010 at 4:00 PM, H. Peter Anvin <hpa@...or.com> wrote:
> On 05/19/2010 03:58 PM, Mathieu Rondonneau wrote:
>> Thanks for the feedback.
>>
>> No, I don' t see any changes in the /proc/iomem.
>>
>> I am trying to prevent a ioremap of a 4K size on a non aligned 4K
>> address that is below the ISA_START_ADDRESS.
>>
>> The problem generates a oops about overlapping.
>> I have a fix which instruct to not to do any re-map if the section
>> name is "reserved".
>> Which is not really clean.
>> I am looking for a clean way to tell the ioremap function to not remap
>> bios reserved memory.
>> That' s why I thought the e820 would be a good start.
>>
>> I will continue looking into this. It does not crash the systems. A
>> warning generates the oops.
>>
>
> Why are you mapping a fixed-address in ISA space to begin with?
>
> Requests to a fixed address (as opposed to dynamic allocation) have to
> be granted even in reserved space -- after all, that's what the address
> might be reserved for!
>
>        -hpa
>
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/