From: Robin Murphy <robin.murphy@arm.com>
To: Logan Gunthorpe <logang@deltatee.com>,
linux-kernel@vger.kernel.org, linux-nvme@lists.infradead.org,
linux-block@vger.kernel.org, linux-pci@vger.kernel.org,
linux-mm@kvack.org, iommu@lists.linux-foundation.org
Cc: "Minturn Dave B" <dave.b.minturn@intel.com>,
"John Hubbard" <jhubbard@nvidia.com>,
"Dave Hansen" <dave.hansen@linux.intel.com>,
"Ira Weiny" <iweiny@intel.com>,
"Matthew Wilcox" <willy@infradead.org>,
"Christian König" <christian.koenig@amd.com>,
"Jason Gunthorpe" <jgg@ziepe.ca>,
"Jason Ekstrand" <jason@jlekstrand.net>,
"Daniel Vetter" <daniel.vetter@ffwll.ch>,
"Dan Williams" <dan.j.williams@intel.com>,
"Stephen Bates" <sbates@raithlin.com>,
"Jakowski Andrzej" <andrzej.jakowski@intel.com>,
"Christoph Hellwig" <hch@lst.de>,
"Xiong Jianxin" <jianxin.xiong@intel.com>
Subject: Re: [RFC PATCH v2 08/11] iommu/dma: Support PCI P2PDMA pages in dma-iommu map_sg
Date: Fri, 12 Mar 2021 15:52:57 +0000 [thread overview]
Message-ID: <accd4187-7a9d-a8fc-f216-98ec24e3411a@arm.com> (raw)
In-Reply-To: <20210311233142.7900-9-logang@deltatee.com>
On 2021-03-11 23:31, Logan Gunthorpe wrote:
> When a PCI P2PDMA page is seen, set the IOVA length of the segment
> to zero so that it is not mapped into the IOVA. Then, in finalise_sg(),
> apply the appropriate bus address to the segment. The IOVA is not
> created if the scatterlist only consists of P2PDMA pages.
This misled me at first, but I see the implementation does actually
appear to accomodate the case of working ACS where P2P *would* still
need to be mapped at the IOMMU.
> Similar to dma-direct, the sg_mark_pci_p2pdma() flag is used to
> indicate bus address segments. On unmap, P2PDMA segments are skipped
> over when determining the start and end IOVA addresses.
>
> With this change, the flags variable in the dma_map_ops is
> set to DMA_F_PCI_P2PDMA_SUPPORTED to indicate support for
> P2PDMA pages.
>
> Signed-off-by: Logan Gunthorpe <logang@deltatee.com>
> ---
> drivers/iommu/dma-iommu.c | 63 ++++++++++++++++++++++++++++++++-------
> 1 file changed, 53 insertions(+), 10 deletions(-)
>
> diff --git a/drivers/iommu/dma-iommu.c b/drivers/iommu/dma-iommu.c
> index af765c813cc8..c0821e9051a9 100644
> --- a/drivers/iommu/dma-iommu.c
> +++ b/drivers/iommu/dma-iommu.c
> @@ -20,6 +20,7 @@
> #include <linux/mm.h>
> #include <linux/mutex.h>
> #include <linux/pci.h>
> +#include <linux/pci-p2pdma.h>
> #include <linux/swiotlb.h>
> #include <linux/scatterlist.h>
> #include <linux/vmalloc.h>
> @@ -846,7 +847,7 @@ static void iommu_dma_unmap_page(struct device *dev, dma_addr_t dma_handle,
> * segment's start address to avoid concatenating across one.
> */
> static int __finalise_sg(struct device *dev, struct scatterlist *sg, int nents,
> - dma_addr_t dma_addr)
> + dma_addr_t dma_addr, unsigned long attrs)
> {
> struct scatterlist *s, *cur = sg;
> unsigned long seg_mask = dma_get_seg_boundary(dev);
> @@ -864,6 +865,20 @@ static int __finalise_sg(struct device *dev, struct scatterlist *sg, int nents,
> sg_dma_address(s) = DMA_MAPPING_ERROR;
> sg_dma_len(s) = 0;
>
> + if (is_pci_p2pdma_page(sg_page(s)) && !s_iova_len) {
> + if (i > 0)
> + cur = sg_next(cur);
> +
> + sg_dma_address(cur) = sg_phys(s) + s->offset -
Are you sure about that? ;)
> + pci_p2pdma_bus_offset(sg_page(s));
Can the bus offset make P2P addresses overlap with regions of mem space
that we might use for regular IOVA allocation? That would be very bad...
> + sg_dma_len(cur) = s->length;
> + sg_mark_pci_p2pdma(cur);
> +
> + count++;
> + cur_len = 0;
> + continue;
> + }
> +
> /*
> * Now fill in the real DMA data. If...
> * - there is a valid output segment to append to
> @@ -960,11 +975,12 @@ static int iommu_dma_map_sg(struct device *dev, struct scatterlist *sg,
> struct iommu_dma_cookie *cookie = domain->iova_cookie;
> struct iova_domain *iovad = &cookie->iovad;
> struct scatterlist *s, *prev = NULL;
> + struct dev_pagemap *pgmap = NULL;
> int prot = dma_info_to_prot(dir, dev_is_dma_coherent(dev), attrs);
> dma_addr_t iova;
> size_t iova_len = 0;
> unsigned long mask = dma_get_seg_boundary(dev);
> - int i;
> + int i, map = -1, ret = 0;
>
> if (static_branch_unlikely(&iommu_deferred_attach_enabled) &&
> iommu_deferred_attach(dev, domain))
> @@ -993,6 +1009,23 @@ static int iommu_dma_map_sg(struct device *dev, struct scatterlist *sg,
> s_length = iova_align(iovad, s_length + s_iova_off);
> s->length = s_length;
>
> + if (is_pci_p2pdma_page(sg_page(s))) {
> + if (sg_page(s)->pgmap != pgmap) {
> + pgmap = sg_page(s)->pgmap;
> + map = pci_p2pdma_dma_map_type(dev, pgmap);
> + }
> +
> + if (map < 0) {
It rather feels like it should be the job of whoever creates the list in
the first place not to put unusable pages in it, especially since the
p2pdma_map_type looks to be a fairly coarse-grained and static thing.
The DMA API isn't responsible for validating normal memory pages, so
what makes P2P special?
> + ret = -EREMOTEIO;
> + goto out_restore_sg;
> + }
> +
> + if (map) {
> + s->length = 0;
I'm not really thrilled about the idea of passing zero-length segments
to iommu_map_sg(). Yes, it happens to trick the concatenation logic in
the current implementation into doing what you want, but it feels fragile.
> + continue;
> + }
> + }
> +
> /*
> * Due to the alignment of our single IOVA allocation, we can
> * depend on these assumptions about the segment boundary mask:
> @@ -1015,6 +1048,9 @@ static int iommu_dma_map_sg(struct device *dev, struct scatterlist *sg,
> prev = s;
> }
>
> + if (!iova_len)
> + return __finalise_sg(dev, sg, nents, 0, attrs);
> +
> iova = iommu_dma_alloc_iova(domain, iova_len, dma_get_mask(dev), dev);
> if (!iova)
> goto out_restore_sg;
> @@ -1026,19 +1062,19 @@ static int iommu_dma_map_sg(struct device *dev, struct scatterlist *sg,
> if (iommu_map_sg_atomic(domain, iova, sg, nents, prot) < iova_len)
> goto out_free_iova;
>
> - return __finalise_sg(dev, sg, nents, iova);
> + return __finalise_sg(dev, sg, nents, iova, attrs);
>
> out_free_iova:
> iommu_dma_free_iova(cookie, iova, iova_len, NULL);
> out_restore_sg:
> __invalidate_sg(sg, nents);
> - return 0;
> + return ret;
> }
>
> static void iommu_dma_unmap_sg(struct device *dev, struct scatterlist *sg,
> int nents, enum dma_data_direction dir, unsigned long attrs)
> {
> - dma_addr_t start, end;
> + dma_addr_t end, start = DMA_MAPPING_ERROR;
> struct scatterlist *tmp;
> int i;
>
> @@ -1054,14 +1090,20 @@ static void iommu_dma_unmap_sg(struct device *dev, struct scatterlist *sg,
> * The scatterlist segments are mapped into a single
> * contiguous IOVA allocation, so this is incredibly easy.
> */
> - start = sg_dma_address(sg);
> - for_each_sg(sg_next(sg), tmp, nents - 1, i) {
> + for_each_sg(sg, tmp, nents, i) {
> + if (sg_is_pci_p2pdma(tmp))
Since the flag is associated with the DMA address which will no longer
be valid, shouldn't it be cleared? The circumstances in which leaving it
around could cause a problem are tenuous, but definitely possible.
Robin.
> + continue;
> if (sg_dma_len(tmp) == 0)
> break;
> - sg = tmp;
> +
> + if (start == DMA_MAPPING_ERROR)
> + start = sg_dma_address(tmp);
> +
> + end = sg_dma_address(tmp) + sg_dma_len(tmp);
> }
> - end = sg_dma_address(sg) + sg_dma_len(sg);
> - __iommu_dma_unmap(dev, start, end - start);
> +
> + if (start != DMA_MAPPING_ERROR)
> + __iommu_dma_unmap(dev, start, end - start);
> }
>
> static dma_addr_t iommu_dma_map_resource(struct device *dev, phys_addr_t phys,
> @@ -1254,6 +1296,7 @@ static unsigned long iommu_dma_get_merge_boundary(struct device *dev)
> }
>
> static const struct dma_map_ops iommu_dma_ops = {
> + .flags = DMA_F_PCI_P2PDMA_SUPPORTED,
> .alloc = iommu_dma_alloc,
> .free = iommu_dma_free,
> .alloc_pages = dma_common_alloc_pages,
>
_______________________________________________
Linux-nvme mailing list
Linux-nvme@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-nvme
next prev parent reply other threads:[~2021-03-12 15:53 UTC|newest]
Thread overview: 47+ messages / expand[flat|nested] mbox.gz Atom feed top
2021-03-11 23:31 [RFC PATCH v2 00/11] Add support to dma_map_sg for P2PDMA Logan Gunthorpe
2021-03-11 23:31 ` [RFC PATCH v2 01/11] PCI/P2PDMA: Pass gfp_mask flags to upstream_bridge_distance_warn() Logan Gunthorpe
2021-03-12 20:39 ` Bjorn Helgaas
2021-03-12 20:53 ` Logan Gunthorpe
2021-03-11 23:31 ` [RFC PATCH v2 02/11] PCI/P2PDMA: Avoid pci_get_slot() which sleeps Logan Gunthorpe
2021-03-12 20:57 ` Bjorn Helgaas
2021-03-12 21:37 ` Logan Gunthorpe
2021-03-11 23:31 ` [RFC PATCH v2 03/11] PCI/P2PDMA: Attempt to set map_type if it has not been set Logan Gunthorpe
2021-03-11 23:31 ` [RFC PATCH v2 04/11] PCI/P2PDMA: Introduce pci_p2pdma_should_map_bus() and pci_p2pdma_bus_offset() Logan Gunthorpe
2021-03-13 1:38 ` Ira Weiny
2021-03-15 16:27 ` Logan Gunthorpe
2021-03-24 17:21 ` Jason Gunthorpe
2021-03-24 18:34 ` Christian König
2021-03-13 2:32 ` Ira Weiny
2021-03-15 16:30 ` Logan Gunthorpe
2021-03-16 8:14 ` Christoph Hellwig
2021-03-11 23:31 ` [RFC PATCH v2 05/11] lib/scatterlist: Add flag for indicating P2PDMA segments in an SGL Logan Gunthorpe
2021-03-11 23:31 ` [RFC PATCH v2 06/11] dma-direct: Support PCI P2PDMA pages in dma-direct map_sg Logan Gunthorpe
2021-03-12 15:52 ` Robin Murphy
2021-03-12 16:24 ` Logan Gunthorpe
2021-03-12 18:11 ` Robin Murphy
2021-03-12 18:27 ` Logan Gunthorpe
2021-03-16 7:58 ` Christoph Hellwig
2021-03-16 15:54 ` Logan Gunthorpe
2021-03-16 7:56 ` Christoph Hellwig
2021-03-16 8:11 ` Christoph Hellwig
2021-03-11 23:31 ` [RFC PATCH v2 07/11] dma-mapping: Add flags to dma_map_ops to indicate PCI P2PDMA support Logan Gunthorpe
2021-03-13 2:36 ` Ira Weiny
2021-03-15 16:33 ` Logan Gunthorpe
2021-03-16 8:00 ` Christoph Hellwig
2021-03-16 8:15 ` Christoph Hellwig
2021-03-11 23:31 ` [RFC PATCH v2 08/11] iommu/dma: Support PCI P2PDMA pages in dma-iommu map_sg Logan Gunthorpe
2021-03-12 15:52 ` Robin Murphy [this message]
2021-03-12 17:03 ` Logan Gunthorpe
2021-03-12 19:47 ` Robin Murphy
2021-03-12 20:06 ` Logan Gunthorpe
2021-03-11 23:31 ` [RFC PATCH v2 09/11] block: Add BLK_STS_P2PDMA Logan Gunthorpe
2021-03-16 8:00 ` Christoph Hellwig
2021-03-16 16:02 ` Logan Gunthorpe
2021-03-11 23:31 ` [RFC PATCH v2 10/11] nvme-pci: Check DMA ops when indicating support for PCI P2PDMA Logan Gunthorpe
2021-03-11 23:31 ` [RFC PATCH v2 11/11] nvme-pci: Convert to using dma_map_sg for p2pdma pages Logan Gunthorpe
2021-03-11 23:59 ` Jason Gunthorpe
2021-03-12 1:37 ` Logan Gunthorpe
2021-03-12 15:51 ` [RFC PATCH v2 00/11] Add support to dma_map_sg for P2PDMA Robin Murphy
2021-03-12 16:18 ` Logan Gunthorpe
2021-03-12 17:46 ` Robin Murphy
2021-03-12 18:24 ` Logan Gunthorpe
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=accd4187-7a9d-a8fc-f216-98ec24e3411a@arm.com \
--to=robin.murphy@arm.com \
--cc=andrzej.jakowski@intel.com \
--cc=christian.koenig@amd.com \
--cc=dan.j.williams@intel.com \
--cc=daniel.vetter@ffwll.ch \
--cc=dave.b.minturn@intel.com \
--cc=dave.hansen@linux.intel.com \
--cc=hch@lst.de \
--cc=iommu@lists.linux-foundation.org \
--cc=iweiny@intel.com \
--cc=jason@jlekstrand.net \
--cc=jgg@ziepe.ca \
--cc=jhubbard@nvidia.com \
--cc=jianxin.xiong@intel.com \
--cc=linux-block@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=linux-nvme@lists.infradead.org \
--cc=linux-pci@vger.kernel.org \
--cc=logang@deltatee.com \
--cc=sbates@raithlin.com \
--cc=willy@infradead.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).