linux-kernel - Re: [PATCH] sisfb: limit POST memory test according to PCI resource length

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] [day] [month] [year] [list]

Message-ID: <20101110010334.GA28852@linux-sh.org>
Date:	Wed, 10 Nov 2010 10:03:34 +0900
From:	Paul Mundt <lethal@...ux-sh.org>
To:	Aaro Koskinen <aaro.koskinen@....fi>
Cc:	linux-kernel@...r.kernel.org, linux-fbdev@...r.kernel.org,
	Thomas Winischhofer <thomas@...ischhofer.net>
Subject: Re: [PATCH] sisfb: limit POST memory test according to PCI resource length

On Wed, Nov 10, 2010 at 12:25:04AM +0200, Aaro Koskinen wrote:
> diff --git a/drivers/video/sis/sis_main.c b/drivers/video/sis/sis_main.c
> index b52f8e4..8a528aa 100644
> --- a/drivers/video/sis/sis_main.c
> +++ b/drivers/video/sis/sis_main.c
> @@ -4514,7 +4514,7 @@ sisfb_post_sis300(struct pci_dev *pdev)
>  	} else {
>  #endif
>  		/* Need to map max FB size for finding out about RAM size */
> -		mapsize = 64 << 20;
> +		mapsize = ivideo->video_size;
>  		sisfb_post_map_vram(ivideo, &mapsize, 4);
>  
>  		if(ivideo->video_vbase) {
> @@ -4680,7 +4680,7 @@ sisfb_post_xgi_ramsize(struct sis_video_info *ivideo)
>  	orSISIDXREG(SISSR, 0x20, (0x80 | 0x04));
>  
>  	/* Need to map max FB size for finding out about RAM size */
> -	mapsize = 256 << 20;
> +	mapsize = ivideo->video_size;
>  	sisfb_post_map_vram(ivideo, &mapsize, 32);
>  
>  	if(!ivideo->video_vbase) {
> 
sisfb_post_map_vram() expects that the mapsize >= min, and falls back on
the default aperture size otherwise. If you're going to pass in a
variable size for video_size then this expectation may no longer hold
true, and you've then changed the behaviour if an invalid size succeeds
on the initial ioremap() attempt.

Simply inserting a:

	if (*mapsize < min)
		return;

sanity check prior to the ioremap() should preserve the existing
behaviour.
--
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/