Re: [Qemu-devel] [PATCH] vfio: Align iova also to IOMMU page size

[Alex Bennée]

Pavel Fedin <address@hidden> writes:

>  Hello!
>
>> > On some architectures TARGET_PAGE_ALIGN() is not enough to get the right
>> > alignment. For example on ARM TARGET_PAGE_BITS is 10 because some old CPUs
>> > support 1K page size, while minimum SMMU page size is 4K.
>> >
>> > This fixes problems like:
>> >
>> > 2015-11-17T07:37:42.892265Z qemu-system-aarch64: VFIO_MAP_DMA: -22
>> > 2015-11-17T07:37:42.892309Z qemu-system-aarch64: vfio_dma_map(0x223da230, 
>> > 0x80002f0400,
>> 0x10fc00, 0x7f89b40400) = -22 (Invalid
>> > argument)
>> > qemu: hardware error: vfio: DMA mapping failed, unable to continue
>
> [skip]
>
>> I don't understand how this is supposed to work, if we align to a larger
>> size than the processor, then there are processor size pages of RAM than
>> could be handed out as DMA targets for devices, but we can't map them
>> through the IOMMU.  Thus if the guest tries to use them, we get IOMMU
>> faults in the host and likely memory corruption in the guest because the
>> device can't read or write to the page it's supposed to.  This doesn't
>> seem like the right solution.
>
>  Well, this was my first try on the problem. I've got your idea. But i guess 
> we should discuss the proper solution then.
>  So, i've got this problem on ARM64. On ARM64 we actually can never have 1K 
> pages. This page size was supported only by old 32-bit ARM CPUs, up to ARMv5 
> IIRC, then it was dropped. Linux OS never even used it.
>  But, since qemu can emulate those ancient CPUs, TARGET_PAGE_BITS is defined 
> to 10 for ARM. And, ARM64 and ARM32 is actually the same target for qemu, so 
> this is why we still get it.
>  Perhaps, TARGET_PAGE_BITS should be a variable for ARM, and we should
>  set it according to the actual used CPU. Then this IOMMU alignment
>  problem would disappear automatically. What do you think?

Yes it should be. For one thing we pay a fairly high performance penalty
for using these smaller pages for no reason. What the best way to do
this remains to be seen as I think there a lot of fixed sized arrays
currently in the system based on various derivations of TARGET_PAGE_BITS.

>  Cc'ed Peter since he is the main ARM guy here.
>
> Kind regards,
> Pavel Fedin
> Expert Engineer
> Samsung Electronics Research center Russia


--
Alex Bennée