Re: [PATCH v3 19/20] PCI/P2PDMA: introduce pci_mmap_p2pmem()

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

To: Jason Gunthorpe <jgg@xxxxxxxx>
Subject: Re: [PATCH v3 19/20] PCI/P2PDMA: introduce pci_mmap_p2pmem()
From: Christian König <christian.koenig@xxxxxxx>
Date: Mon, 4 Oct 2021 16:54:03 +0200
Cc: Logan Gunthorpe <logang@xxxxxxxxxxxx>, Alistair Popple <apopple@xxxxxxxxxx>, Felix Kuehling <Felix.Kuehling@xxxxxxx>, Christoph Hellwig <hch@xxxxxx>, Dan Williams <dan.j.williams@xxxxxxxxx>, linux-kernel@xxxxxxxxxxxxxxx, linux-nvme@xxxxxxxxxxxxxxxxxxx, linux-block@xxxxxxxxxxxxxxx, linux-pci@xxxxxxxxxxxxxxx, linux-mm@xxxxxxxxx, iommu@xxxxxxxxxxxxxxxxxxxxxxxxxx, Stephen Bates <sbates@xxxxxxxxxxxx>, John Hubbard <jhubbard@xxxxxxxxxx>, Don Dutile <ddutile@xxxxxxxxxx>, Matthew Wilcox <willy@xxxxxxxxxxxxx>, Daniel Vetter <daniel.vetter@xxxxxxxx>, Jakowski Andrzej <andrzej.jakowski@xxxxxxxxx>, Minturn Dave B <dave.b.minturn@xxxxxxxxx>, Jason Ekstrand <jason@xxxxxxxxxxxxxx>, Dave Hansen <dave.hansen@xxxxxxxxxxxxxxx>, Xiong Jianxin <jianxin.xiong@xxxxxxxxx>, Bjorn Helgaas <helgaas@xxxxxxxxxx>, Ira Weiny <ira.weiny@xxxxxxxxx>, Robin Murphy <robin.murphy@xxxxxxx>, Martin Oliveira <martin.oliveira@xxxxxxxxxxxxx>, Chaitanya Kulkarni <ckulkarnilinux@xxxxxxxxx>
In-reply-to: <20211004132759.GX3544071@ziepe.ca>
References: <32ce26d7-86e9-f8d5-f0cf-40497946efe9@deltatee.com> <20210929233540.GF3544071@ziepe.ca> <f9a83402-3d66-7437-ca47-77bac4108424@deltatee.com> <20210930003652.GH3544071@ziepe.ca> <20211001134856.GN3544071@ziepe.ca> <4fdd337b-fa35-a909-5eee-823bfd1e9dc4@deltatee.com> <20211001174511.GQ3544071@ziepe.ca> <809be72b-efb2-752c-31a6-702c8a307ce7@amd.com> <20211004131102.GU3544071@ziepe.ca> <1e219386-7547-4f42-d090-2afd62a268d7@amd.com> <20211004132759.GX3544071@ziepe.ca>
User-agent: Mozilla/5.0 (X11; Linux x86_64; rv:78.0) Gecko/20100101 Thunderbird/78.13.0

Am 04.10.21 um 15:27 schrieb Jason Gunthorpe:

On Mon, Oct 04, 2021 at 03:22:22PM +0200, Christian König wrote:

That use case is completely unrelated to GUP and when this doesn't work we
have quite a problem.

My read is that unmap_mapping_range() guarentees the physical TLB
hardware is serialized across all CPUs upon return.


Thanks, that's what I wanted to make sure.

Christian.


It also guarentees GUP slow is serialized due to the page table
spinlocks.

Jason

References:
- Re: [PATCH v3 19/20] PCI/P2PDMA: introduce pci_mmap_p2pmem()
  - From: Logan Gunthorpe
- Re: [PATCH v3 19/20] PCI/P2PDMA: introduce pci_mmap_p2pmem()
  - From: Jason Gunthorpe
- Re: [PATCH v3 19/20] PCI/P2PDMA: introduce pci_mmap_p2pmem()
  - From: Logan Gunthorpe
- Re: [PATCH v3 19/20] PCI/P2PDMA: introduce pci_mmap_p2pmem()
  - From: Jason Gunthorpe
- Re: [PATCH v3 19/20] PCI/P2PDMA: introduce pci_mmap_p2pmem()
  - From: Jason Gunthorpe
- Re: [PATCH v3 19/20] PCI/P2PDMA: introduce pci_mmap_p2pmem()
  - From: Logan Gunthorpe
- Re: [PATCH v3 19/20] PCI/P2PDMA: introduce pci_mmap_p2pmem()
  - From: Jason Gunthorpe
- Re: [PATCH v3 19/20] PCI/P2PDMA: introduce pci_mmap_p2pmem()
  - From: Christian König
- Re: [PATCH v3 19/20] PCI/P2PDMA: introduce pci_mmap_p2pmem()
  - From: Jason Gunthorpe
- Re: [PATCH v3 19/20] PCI/P2PDMA: introduce pci_mmap_p2pmem()
  - From: Christian König
- Re: [PATCH v3 19/20] PCI/P2PDMA: introduce pci_mmap_p2pmem()
  - From: Jason Gunthorpe

Prev by Date: [PATCH 50/62] mm/slub: Convert slab_lock() and slab_unlock() to struct slab
Next by Date: [PATCH 53/62] mm/slub: Remove pfmemalloc_match_unsafe()
Previous by thread: Re: [PATCH v3 19/20] PCI/P2PDMA: introduce pci_mmap_p2pmem()
Next by thread: Re: [PATCH v3 19/20] PCI/P2PDMA: introduce pci_mmap_p2pmem()
Index(es):
- Date
- Thread

[Index of Archives] [Linux ARM Kernel] [Linux ARM] [Linux Omap] [Fedora ARM] [IETF Annouce] [Bugtraq] [Linux OMAP] [Linux MIPS] [eCos] [Asterisk Internet PBX] [Linux API]