Re: [PATCH v3] x86, kdump: Set crashkernel_low automatically

From: Yinghai Lu
Date: Mon Mar 18 2013 - 19:27:19 EST

On Mon, Mar 18, 2013 at 3:52 PM, H. Peter Anvin <hpa@xxxxxxxxx> wrote:
> On 03/18/2013 02:25 PM, Yinghai Lu wrote:
>> Current code does not set low range for crashkernel if the user
>> does not specify that.
>> That cause regressions on system that does not support intel_iommu
>> properly.
>> Chao said that his system does work well on 3.8 without extra parameter.
>> even iommu does not work with kdump.
>> Set crashkernel_low automatically if the user does not specify that.
>> For system that does support IOMMU with kdump properly, user could
>> specify crashkernel_low=0 to save that 72M low ram.
>> -v3: add swiotlb_size() according to Konrad.
>> Reported-by: WANG Chao <chaowang@xxxxxxxxxx>
>> Tested-by: WANG Chao <chaowang@xxxxxxxxxx>
>> Signed-off-by: Yinghai Lu <yinghai@xxxxxxxxxx>
> Can we get a bit more of an explanation instead of "and etc 8M"? At
> least a hint of what kind of objects would go in there...

now only have:
swiotlb overflow buffer

v_overflow_buffer = alloc_bootmem_low_pages_nopanic(

* When the IOMMU overflows we return a fallback buffer. This
sets the size.
static unsigned long io_tlb_overflow = 32*1024;

so it is 32K, and I round it to 8M.
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at
Please read the FAQ at