Re: [PATCH v4 1/4] iommu/iova: Avoid over-allocating when size-aligned

[David Woodhouse <dwmw2-wEGCiKHe2LqWVfeAwA7xHQ@public.gmane.org>]

[-- Attachment #1.1: Type: text/plain, Size: 1394 bytes --]

On Thu, 2015-07-16 at 19:40 +0100, Robin Murphy wrote:
> Currently, allocating a size-aligned IOVA region quietly adjusts the
> actual allocation size in the process, returning a rounded-up
> power-of-two-sized allocation. This results in mismatched behaviour in
> the IOMMU driver if the original size was not a power of two, where the
> original size is mapped, but the rounded-up IOVA size is unmapped.
> 
> Whilst some IOMMUs will happily unmap already-unmapped pages, others
> consider this an error, so fix it by computing the necessary alignment
> padding without altering the actual allocation size. Also clean up by
> making pad_size unsigned, since its callers always pass unsigned values
> and negative padding makes little sense here anyway.

Applied; thanks.

I'm not 100% sure we *need* the hunk in intel-iommu.c, we can probably
live without rounding the size up. It means we'll use huge pages less
often, but it's not clear that using them when we didn't *mean* to map
the full size of them was the right thing to do in the first place.

But it makes sense to apply the patch as-is, without changing the
effective behaviour, and ponder that more deeply later.

Thanks.

-- 
David Woodhouse                            Open Source Technology Centre
David.Woodhouse-ral2JQCrhuEAvxtiuMwx3w@public.gmane.org                              Intel Corporation

[-- Attachment #1.2: smime.p7s --]
[-- Type: application/x-pkcs7-signature, Size: 5691 bytes --]

[-- Attachment #2: Type: text/plain, Size: 0 bytes --]