On 2025/10/20 16:03, Duan, Zhenzhong wrote:
-----Original Message-----
From: Liu, Yi L <[email protected]>
Subject: Re: [PATCH v2 6/8] intel_iommu: Fix unmap_bitmap failure with
legacy VFIO backend
On 2025/10/17 16:22, Zhenzhong Duan wrote:
If a VFIO device in guest switches from IOMMU domain to block domain,
vtd_address_space_unmap() is called to unmap whole address space.
If that happens during migration, migration fails with legacy VFIO
backend as below:
Status: failed (vfio_container_dma_unmap(0x561bbbd92d90,
0x100000000000, 0x100000000000) = -7 (Argument list too long))
Because legacy VFIO limits maximum bitmap size to 256MB which maps to
8TB on
4K page system, when 16TB sized UNMAP notification is sent,
unmap_bitmap
ioctl fails.
It would be great to add some words to note why vIOMMU can trigger this.
Hi Yi, I think the first sentence in description is explaining that?
"If a VFIO device in guest switches from IOMMU domain to block domain,
vtd_address_space_unmap() is called to unmap whole address space."
aha, yes. I was trying to mark it is NOT necessarily related to VM
memory size. Could you note that the address space is guest IOVA which
is not system memory?
Regards,
Yi Liu
