min_align_mask Re: [PATCH V3 13/13] HV/Storvsc: Add Isolation VM support for storvsc driver

* min_align_mask Re: [PATCH V3 13/13] HV/Storvsc: Add Isolation VM support for storvsc driver
       [not found]       ` <CY4PR21MB1586FEB6F6ADD592C04E541BD7C19@CY4PR21MB1586.namprd21.prod.outlook.com>
@ 2021-08-24  8:49         ` hch
  0 siblings, 0 replies; only message in thread
From: hch @ 2021-08-24  8:49 UTC (permalink / raw)
  To: Michael Kelley
  Cc: hch, Tianyu Lan, KY Srinivasan, Haiyang Zhang, Stephen Hemminger,
	wei.liu, Dexuan Cui, tglx, mingo, bp, x86, hpa, dave.hansen,
	luto, peterz, konrad.wilk, boris.ostrovsky, jgross, sstabellini,
	joro, will, davem, kuba, jejb, martin.petersen, arnd,
	m.szyprowski, robin.murphy, thomas.lendacky, brijesh.singh, ardb,
	Tianyu Lan, pgonda, martin.b.radev, akpm, kirill.shutemov, rppt,
	sfr, saravanand, krish.sadhukhan, aneesh.kumar, xen-devel,
	rientjes, hannes, tj, iommu, linux-arch, linux-hyperv,
	linux-kernel, linux-scsi, netdev, vkuznets, parri.andrea,
	dave.hansen, linux-rdma, mpi3mr-linuxdrv.pdl

On Fri, Aug 20, 2021 at 03:40:08PM +0000, Michael Kelley wrote:
> I see that the swiotlb code gets and uses the min_align_mask field.  But
> the NVME driver is the only driver that ever sets it, so the value is zero
> in all other cases.  Does swiotlb just use PAGE_SIZE in that that case?  I
> couldn't tell from a quick glance at the swiotlb code.

The encoding isn't all that common.  I only know it for the RDMA memory
registration format, and RDMA in general does not interact very well
with swiotlb (although the in-kernel drivers should work fine, it is
userspace RDMA that is the problem).  It seems recently a new driver
using the format (mpi3mr) also showed up.  All these should probably set
the min_align_mask.

^ permalink raw reply	[flat|nested] only message in thread