Re: [BUG 2.6.31-rc1] HIGHMEM64G causes hang in PCI init on 32-bit x86

["H. Peter Anvin" <hpa@zytor.com>]

Yinghai Lu wrote:
>  			continue;
> @@ -1409,8 +1409,10 @@ void __init e820_reserve_resources_late(
>  		end = round_up(start, ram_alignment(start));
>  		if (start == end)
>  			continue;
> -		reserve_region_with_split(&iomem_resource, start,
> -						  end - 1, "RAM buffer");
> +		if (end != (resource_size_t)end)
> +			continue;
> +		reserve_region_with_split(&iomem_resource, (resource_size_t)start,
> +					  (resource_size_t)(end - 1), "RAM buffer");
>  	}
>  }
>  

That doesn't quite look right; for one thing it doesn't handle the
(admittedly somewhat unlikely) case of end pointing to the end of the
address space.

Something like:

	if (start > (resource_size_t)end-1)
		continue;

... should work better.

	-hpa

-- 
H. Peter Anvin, Intel Open Source Technology Center
I work for Intel.  I don't speak on their behalf.