Re: [PATCH 3/5] x86: Only direct map addresses that are marked as E820_RAM

[Jacob Shin <jacob.shin@amd.com>]

On Fri, Aug 24, 2012 at 05:30:21PM -0700, H. Peter Anvin wrote:
> On 08/24/2012 04:55 PM, Jacob Shin wrote:
> >+
> >+	for (i = 0; i < e820.nr_map; i++) {
> >+		struct e820entry *ei = &e820.map[i];
> >+		u64 start = ei->addr;
> >+		u64 end = ei->addr + ei->size;
> >+
> >+		/* we only map E820_RAM */
> >+		if (ei->type != E820_RAM)
> >+			continue;
> >+
> >+		if (end <= ISA_END_ADDRESS)
> >+			continue;
> >+
> >+		if (start <= ISA_END_ADDRESS)
> >+			start = 0;
> >+#ifdef CONFIG_X86_32
> >+		/* on 32 bit, we only map up to max_low_pfn */
> >+		if ((start >> PAGE_SHIFT) >= max_low_pfn)
> >+			continue;
> >+
> >+		if ((end >> PAGE_SHIFT) > max_low_pfn)
> >+			end = max_low_pfn << PAGE_SHIFT;
> >+#endif
> >+		/* the ISA range is always mapped regardless of holes */
> >+		if (!pfn_range_is_mapped(0, ISA_END_ADDRESS << PAGE_SHIFT) &&
> >+		    start != 0)
> >+			init_memory_mapping(0, ISA_END_ADDRESS);
> >+
> >+		init_memory_mapping(start, end);
> >+	}
> >+
> 
> The ISA range mapping doesn't really make sense *inside* the loop,
> no? It seems you could do that before you enter the loop and then
> simply have:
> 
> +		if (end <= ISA_END_ADDRESS)
> +			continue;
> +
> +		if (start <= ISA_END_ADDRESS)
> +			start = ISA_END_ADDRESS;
> 
> ... no?

Right, I think what I was attempting to do was to merge the 1MB
with E820_RAM right above 1MB:

So instead of:

init_memory_mapping(0, 1MB)
init_memory_mapping(1MB, 2GB)

It would be:

init_memory_mapping(0, 2GB)

While taking care of the odd case where there is a gap right after
1MB.

But if its not worth it, I can move it out of the loop.

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