Intel devices do not have good support for P2P requests that span different
host bridges as the transactions will cross the QPI/UPI bus and this does
not perform well.
Therefore, enable support for these devices only if the host bridges match.
Add Intel devices that have been tested and are known to work. There are
likely many others out there that will need to be tested and added.
Link: https://lore.kernel.org/r/20190730163545.4915-8-logang@deltatee.com
Link: https://lore.kernel.org/r/20190812173048.9186-8-logang@deltatee.com
Signed-off-by: Logan Gunthorpe <logang@deltatee.com>
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Push both PCI devices into the whitelist checking function seeing some
hardware will require us ensuring they are on the same host bridge.
At the same time we rename root_complex_whitelist() to
host_bridge_whitelist() to match the terminology used in the code.
Link: https://lore.kernel.org/r/20190730163545.4915-7-logang@deltatee.com
Link: https://lore.kernel.org/r/20190812173048.9186-7-logang@deltatee.com
Signed-off-by: Logan Gunthorpe <logang@deltatee.com>
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
When a P2PDMA transfer is rejected due to ACS being set, we can also check
the whitelist and allow the transactions.
Do this by pushing the whitelist check into the upstream_bridge_distance()
function.
Link: https://lore.kernel.org/r/20190730163545.4915-6-logang@deltatee.com
Link: https://lore.kernel.org/r/20190812173048.9186-6-logang@deltatee.com
Signed-off-by: Logan Gunthorpe <logang@deltatee.com>
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Add constant flags to indicate how two devices will be mapped or if they
are unsupported. upstream_bridge_distance() will now return the
mapping type and the distance in a passed-by-reference argument.
This helps annotate the code better, but the main reason is so we can use
the information to store the required mapping method in an xarray.
Link: https://lore.kernel.org/r/20190730163545.4915-4-logang@deltatee.com
Link: https://lore.kernel.org/r/20190812173048.9186-4-logang@deltatee.com
Signed-off-by: Logan Gunthorpe <logang@deltatee.com>
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Fix typos in drivers/pci. Comment and whitespace changes only.
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
Reviewed-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
Reviewed-by: Randy Dunlap <rdunlap@infradead.org>
Drivers that use dma_virt_ops were meant to be rejected when testing
compatibility for P2PDMA.
This check got inadvertently dropped in one of the later versions of the
original patchset, so add it back.
Fixes: 52916982af ("PCI/P2PDMA: Support peer-to-peer memory")
Link: https://lore.kernel.org/r/20190702173544.21950-1-logang@deltatee.com
Signed-off-by: Logan Gunthorpe <logang@deltatee.com>
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
The functionality is identical to the one currently open coded in
p2pdma.c.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Ira Weiny <ira.weiny@intel.com>
Reviewed-by: Logan Gunthorpe <logang@deltatee.com>
Tested-by: Logan Gunthorpe <logang@deltatee.com>
Reviewed-by: Dan Williams <dan.j.williams@intel.com>
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
Passing the actual typed structure leads to more understandable code
vs just passing the ref member.
Reported-by: Logan Gunthorpe <logang@deltatee.com>
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Logan Gunthorpe <logang@deltatee.com>
Reviewed-by: Jason Gunthorpe <jgg@mellanox.com>
Reviewed-by: Dan Williams <dan.j.williams@intel.com>
Tested-by: Dan Williams <dan.j.williams@intel.com>
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
The dev_pagemap is a growing too many callbacks. Move them into a
separate ops structure so that they are not duplicated for multiple
instances, and an attacker can't easily overwrite them.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Logan Gunthorpe <logang@deltatee.com>
Reviewed-by: Jason Gunthorpe <jgg@mellanox.com>
Reviewed-by: Dan Williams <dan.j.williams@intel.com>
Tested-by: Dan Williams <dan.j.williams@intel.com>
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
-----BEGIN PGP SIGNATURE-----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=0Qnd
-----END PGP SIGNATURE-----
Merge tag 'pci-v5.2-fixes-1' of git://git.kernel.org/pub/scm/linux/kernel/git/helgaas/pci
Pull PCI fix from Bjorn Helgaas:
"If an IOMMU is present, ignore the P2PDMA whitelist we added for v5.2
because we don't yet know how to support P2PDMA in that case (Logan
Gunthorpe)"
* tag 'pci-v5.2-fixes-1' of git://git.kernel.org/pub/scm/linux/kernel/git/helgaas/pci:
PCI/P2PDMA: Ignore root complex whitelist when an IOMMU is present
Presently, there is no path to DMA map P2PDMA memory, so if a TLP targeting
this memory hits the root complex and an IOMMU is present, the IOMMU will
reject the transaction, even if the RC would support P2PDMA.
So until the kernel knows to map these DMA addresses in the IOMMU, we
should not enable the whitelist when an IOMMU is present.
Link: https://lore.kernel.org/linux-pci/20190522201252.2997-1-logang@deltatee.com/
Fixes: 0f97da8310 ("PCI/P2PDMA: Allow P2P DMA between any devices under AMD ZEN Root Complex")
Signed-off-by: Logan Gunthorpe <logang@deltatee.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
Cc: Christoph Hellwig <hch@lst.de>
Logan noticed that devm_memremap_pages_release() kills the percpu_ref
drops all the page references that were acquired at init and then
immediately proceeds to unplug, arch_remove_memory(), the backing pages
for the pagemap. If for some reason device shutdown actually collides
with a busy / elevated-ref-count page then arch_remove_memory() should
be deferred until after that reference is dropped.
As it stands the "wait for last page ref drop" happens *after*
devm_memremap_pages_release() returns, which is obviously too late and
can lead to crashes.
Fix this situation by assigning the responsibility to wait for the
percpu_ref to go idle to devm_memremap_pages() with a new ->cleanup()
callback. Implement the new cleanup callback for all
devm_memremap_pages() users: pmem, devdax, hmm, and p2pdma.
Link: http://lkml.kernel.org/r/155727339156.292046.5432007428235387859.stgit@dwillia2-desk3.amr.corp.intel.com
Fixes: 41e94a8513 ("add devm_memremap_pages")
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
Reported-by: Logan Gunthorpe <logang@deltatee.com>
Reviewed-by: Ira Weiny <ira.weiny@intel.com>
Reviewed-by: Logan Gunthorpe <logang@deltatee.com>
Cc: Bjorn Helgaas <bhelgaas@google.com>
Cc: "Jérôme Glisse" <jglisse@redhat.com>
Cc: Christoph Hellwig <hch@lst.de>
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Cc: "Rafael J. Wysocki" <rafael@kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
In preparation for fixing a race between devm_memremap_pages_release()
and the final put of a page from the device-page-map, allocate a
percpu-ref per p2pdma resource mapping.
Link: http://lkml.kernel.org/r/155727338646.292046.9922678317501435597.stgit@dwillia2-desk3.amr.corp.intel.com
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
Reviewed-by: Ira Weiny <ira.weiny@intel.com>
Reviewed-by: Logan Gunthorpe <logang@deltatee.com>
Cc: Bjorn Helgaas <bhelgaas@google.com>
Cc: Christoph Hellwig <hch@lst.de>
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Cc: "Jérôme Glisse" <jglisse@redhat.com>
Cc: "Rafael J. Wysocki" <rafael@kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
The pci_p2pdma_add_resource() implementation immediately frees the pgmap
if gen_pool_add_virt() fails. However, that means that when @dev
triggers a devres release devm_memremap_pages_release() will crash
trying to access the freed @pgmap.
Use the new devm_memunmap_pages() to manually free the mapping in the
error path.
Link: http://lkml.kernel.org/r/155727337603.292046.13101332703665246702.stgit@dwillia2-desk3.amr.corp.intel.com
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
Fixes: 52916982af ("PCI/P2PDMA: Support peer-to-peer memory")
Reviewed-by: Ira Weiny <ira.weiny@intel.com>
Acked-by: Bjorn Helgaas <bhelgaas@google.com>
Reviewed-by: Logan Gunthorpe <logang@deltatee.com>
Cc: Christoph Hellwig <hch@lst.de>
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Cc: "Jérôme Glisse" <jglisse@redhat.com>
Cc: "Rafael J. Wysocki" <rafael@kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
The PCI specs say that peer-to-peer DMA should be supported between any two
devices that have a common upstream PCI-to-PCI bridge. But devices under
different Root Ports don't share a common upstream bridge, and PCIe r4.0,
sec 1.3.1, says routing peer-to-peer transactions between Root Ports in the
same Root Complex is optional.
Many Root Complexes, including AMD ZEN, *do* support peer-to-peer DMA even
between Root Ports. Add a whitelist and allow peer-to-peer DMA if both
participants are attached to a Root Complex known to support it.
Link: https://lore.kernel.org/linux-pci/20190418115859.2394-1-christian.koenig@amd.com
Signed-off-by: Christian König <christian.koenig@amd.com>
[bhelgaas: changelog]
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
Reviewed-by: Logan Gunthorpe <logang@deltatee.com>
At Maintainer Summit, Greg brought up a topic I proposed around
EXPORT_SYMBOL_GPL usage. The motivation was considerations for when
EXPORT_SYMBOL_GPL is warranted and the criteria for taking the exceptional
step of reclassifying an existing export. Specifically, I wanted to make
the case that although the line is fuzzy and hard to specify in abstract
terms, it is nonetheless clear that devm_memremap_pages() and HMM
(Heterogeneous Memory Management) have crossed it. The
devm_memremap_pages() facility should have been EXPORT_SYMBOL_GPL from the
beginning, and HMM as a derivative of that functionality should have
naturally picked up that designation as well.
Contrary to typical rules, the HMM infrastructure was merged upstream with
zero in-tree consumers. There was a promise at the time that those users
would be merged "soon", but it has been over a year with no drivers
arriving. While the Nouveau driver is about to belatedly make good on
that promise it is clear that HMM was targeted first and foremost at an
out-of-tree consumer.
HMM is derived from devm_memremap_pages(), a facility Christoph and I
spearheaded to support persistent memory. It combines a device lifetime
model with a dynamically created 'struct page' / memmap array for any
physical address range. It enables coordination and control of the many
code paths in the kernel built to interact with memory via 'struct page'
objects. With HMM the integration goes even deeper by allowing device
drivers to hook and manipulate page fault and page free events.
One interpretation of when EXPORT_SYMBOL is suitable is when it is
exporting stable and generic leaf functionality. The
devm_memremap_pages() facility continues to see expanding use cases,
peer-to-peer DMA being the most recent, with no clear end date when it
will stop attracting reworks and semantic changes. It is not suitable to
export devm_memremap_pages() as a stable 3rd party driver API due to the
fact that it is still changing and manipulates core behavior. Moreover,
it is not in the best interest of the long term development of the core
memory management subsystem to permit any external driver to effectively
define its own system-wide memory management policies with no
encouragement to engage with upstream.
I am also concerned that HMM was designed in a way to minimize further
engagement with the core-MM. That, with these hooks in place,
device-drivers are free to implement their own policies without much
consideration for whether and how the core-MM could grow to meet that
need. Going forward not only should HMM be EXPORT_SYMBOL_GPL, but the
core-MM should be allowed the opportunity and stimulus to change and
address these new use cases as first class functionality.
Original changelog:
hmm_devmem_add(), and hmm_devmem_add_resource() duplicated
devm_memremap_pages() and are now simple now wrappers around the core
facility to inject a dev_pagemap instance into the global pgmap_radix and
hook page-idle events. The devm_memremap_pages() interface is base
infrastructure for HMM. HMM has more and deeper ties into the kernel
memory management implementation than base ZONE_DEVICE which is itself a
EXPORT_SYMBOL_GPL facility.
Originally, the HMM page structure creation routines copied the
devm_memremap_pages() code and reused ZONE_DEVICE. A cleanup to unify the
implementations was discussed during the initial review:
http://lkml.iu.edu/hypermail/linux/kernel/1701.2/00812.html Recent work to
extend devm_memremap_pages() for the peer-to-peer-DMA facility enabled
this cleanup to move forward.
In addition to the integration with devm_memremap_pages() HMM depends on
other GPL-only symbols:
mmu_notifier_unregister_no_release
percpu_ref
region_intersects
__class_create
It goes further to consume / indirectly expose functionality that is not
exported to any other driver:
alloc_pages_vma
walk_page_range
HMM is derived from devm_memremap_pages(), and extends deep core-kernel
fundamentals. Similar to devm_memremap_pages(), mark its entry points
EXPORT_SYMBOL_GPL().
[logang@deltatee.com: PCI/P2PDMA: match interface changes to devm_memremap_pages()]
Link: http://lkml.kernel.org/r/20181130225911.2900-1-logang@deltatee.com
Link: http://lkml.kernel.org/r/154275560565.76910.15919297436557795278.stgit@dwillia2-desk3.amr.corp.intel.com
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
Signed-off-by: Logan Gunthorpe <logang@deltatee.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Cc: Logan Gunthorpe <logang@deltatee.com>
Cc: "Jérôme Glisse" <jglisse@redhat.com>
Cc: Balbir Singh <bsingharora@gmail.com>,
Cc: Michal Hocko <mhocko@suse.com>
Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Cc: <stable@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Fix typos, spellos, and grammar in p2pdma.rst and p2pdma.c.
Fix return value(s) in function pci_p2pmem_alloc_sgl().
Signed-off-by: Randy Dunlap <rdunlap@infradead.org>
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
Acked-by: Logan Gunthorpe <logang@deltatee.com>
Cc: Jonathan Corbet <corbet@lwn.net>
Users of the P2PDMA infrastructure will typically need a way for the user
to tell the kernel to use P2P resources. Typically this will be a simple
on/off boolean operation but sometimes it may be desirable for the user to
specify the exact device to use for the P2P operation.
Add new helpers for attributes which take a boolean or a PCI device. Any
boolean as accepted by strtobool() turn P2P on or off (such as 'y', 'n',
'1', '0', etc). Specifying a full PCI device name/BDF will select the
specific device.
Signed-off-by: Logan Gunthorpe <logang@deltatee.com>
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
The DMA address used when mapping PCI P2P memory must be the PCI bus
address. Thus, introduce pci_p2pmem_map_sg() to map the correct addresses
when using P2P memory.
Memory mapped in this way does not need to be unmapped and thus if we
provided pci_p2pmem_unmap_sg() it would be empty. This breaks the expected
balance between map/unmap but was left out as an empty function doesn't
really provide any benefit. In the future, if this call becomes necessary
it can be added without much difficulty.
For this, we assume that an SGL passed to these functions contain all P2P
memory or no P2P memory.
Signed-off-by: Logan Gunthorpe <logang@deltatee.com>
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Add a sysfs group to display statistics about P2P memory that is registered
in each PCI device.
Attributes in the group display the total amount of P2P memory, the amount
available and whether it is published or not.
Signed-off-by: Logan Gunthorpe <logang@deltatee.com>
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Some PCI devices may have memory mapped in a BAR space that's intended for
use in peer-to-peer transactions. To enable such transactions the memory
must be registered with ZONE_DEVICE pages so it can be used by DMA
interfaces in existing drivers.
Add an interface for other subsystems to find and allocate chunks of P2P
memory as necessary to facilitate transfers between two PCI peers:
struct pci_dev *pci_p2pmem_find[_many]();
int pci_p2pdma_distance[_many]();
void *pci_alloc_p2pmem();
The new interface requires a driver to collect a list of client devices
involved in the transaction then call pci_p2pmem_find() to obtain any
suitable P2P memory. Alternatively, if the caller knows a device which
provides P2P memory, they can use pci_p2pdma_distance() to determine if it
is usable. With a suitable p2pmem device, memory can then be allocated
with pci_alloc_p2pmem() for use in DMA transactions.
Depending on hardware, using peer-to-peer memory may reduce the bandwidth
of the transfer but can significantly reduce pressure on system memory.
This may be desirable in many cases: for example a system could be designed
with a small CPU connected to a PCIe switch by a small number of lanes
which would maximize the number of lanes available to connect to NVMe
devices.
The code is designed to only utilize the p2pmem device if all the devices
involved in a transfer are behind the same PCI bridge. This is because we
have no way of knowing whether peer-to-peer routing between PCIe Root Ports
is supported (PCIe r4.0, sec 1.3.1). Additionally, the benefits of P2P
transfers that go through the RC is limited to only reducing DRAM usage
and, in some cases, coding convenience. The PCI-SIG may be exploring
adding a new capability bit to advertise whether this is possible for
future hardware.
This commit includes significant rework and feedback from Christoph
Hellwig.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Logan Gunthorpe <logang@deltatee.com>
[bhelgaas: fold in fix from Keith Busch <keith.busch@intel.com>:
https://lore.kernel.org/linux-pci/20181012155920.15418-1-keith.busch@intel.com,
to address comment from Dan Carpenter <dan.carpenter@oracle.com>, fold in
https://lore.kernel.org/linux-pci/20181017160510.17926-1-logang@deltatee.com]
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
