| Message ID | 20180830185352.3369-1-logang@deltatee.com (mailing list archive) |
|---|---|
| Headers | show |
| Series | Copy Offload in NVMe Fabrics with P2P PCI Memory | expand |
On Thu, Aug 30, 2018 at 12:53:39PM -0600, Logan Gunthorpe wrote: [...] > > When the PCI P2PDMA config option is selected the ACS bits in every > bridge port in the system are turned off to allow traffic to > pass freely behind the root port. At this time, the bit must be disabled > at boot so the IOMMU subsystem can correctly create the groups, though > this could be addressed in the future. There is no way to dynamically > disable the bit and alter the groups. Can you provide an example on how to test this ? Like kernel command line option, the doc patch does not have any such example. It would be nice to add. Maybe i have miss it in some of the patch. I just skimmed over for now. Cheers, Jérôme
On 30/08/18 01:20 PM, Jerome Glisse wrote: > On Thu, Aug 30, 2018 at 12:53:39PM -0600, Logan Gunthorpe wrote: > > [...] > >> >> When the PCI P2PDMA config option is selected the ACS bits in every >> bridge port in the system are turned off to allow traffic to >> pass freely behind the root port. At this time, the bit must be disabled >> at boot so the IOMMU subsystem can correctly create the groups, though >> this could be addressed in the future. There is no way to dynamically >> disable the bit and alter the groups. Oh, sorry this paragraph in the cover letter is wrong now. We now rely on the disable_acs_redir command line option introduced in aaca43fda742 ("PCI: Add "pci=disable_acs_redir=" parameter for peer-to-peer support") > Can you provide an example on how to test this ? Like kernel command > line option, the doc patch does not have any such example. It would be > nice to add. Do you mean to test the patchset or the ACS bits you quoted? Testing the patchset is a matter of having the right hardware (ie an RDMA NIC and CMB enabled NVMe behind a PCIe switch, with the ACS bits set correctly by the above command line option) and setting the p2pmem configfs attribute in an nvme-of port to 'yes'. Logan
Hi Everyone, Now that the patchset which creates a command line option to disable ACS redirection has landed it's time to revisit the P2P patchset for copy offoad in NVMe fabrics. I present version 5 wihch no longer does any magic with the ACS bits and instead will reject P2P transactions between devices that would be affected by them. A few other cleanups were done which are described in the changelog below. This version is based on v4.19-rc1 and a git repo is here: https://github.com/sbates130272/linux-p2pmem pci-p2p-v5 Thanks, Logan -- Changes in v5: * Rebased on v4.19-rc1 * Drop changing ACS settings in this patchset. Now, the code will only allow P2P transactions between devices whos downstream ports do not restrict P2P TLPs. * Drop the REQ_PCI_P2PDMA block flag and instead use is_pci_p2pdma_page() to tell if a request is P2P or not. In that case we check for queue support and enforce using REQ_NOMERGE. Per feedback from Christoph. * Drop the pci_p2pdma_unmap_sg() function as it was empty and only there for symmetry and compatibility with dma_unmap_sg. Per feedback from Christoph. * Split off the logic to handle enabling P2P in NVMe fabrics' configfs into specific helpers in the p2pdma code. Per feedback from Christoph. * A number of other minor cleanups and fixes as pointed out by Christoph and others. Changes in v4: * Change the original upstream_bridges_match() function to upstream_bridge_distance() which calculates the distance between two devices as long as they are behind the same root port. This should address Bjorn's concerns that the code was to focused on being behind a single switch. * The disable ACS function now disables ACS for all bridge ports instead of switch ports (ie. those that had two upstream_bridge ports). * Change the pci_p2pmem_alloc_sgl() and pci_p2pmem_free_sgl() API to be more like sgl_alloc() in that the alloc function returns the allocated scatterlist and nents is not required bythe free function. * Moved the new documentation into the driver-api tree as requested by Jonathan * Add SGL alloc and free helpers in the nvmet code so that the individual drivers can share the code that allocates P2P memory. As requested by Christoph. * Cleanup the nvmet_p2pmem_store() function as Christoph thought my first attempt was ugly. * Numerous commit message and comment fix-ups Changes in v3: * Many more fixes and minor cleanups that were spotted by Bjorn * Additional explanation of the ACS change in both the commit message and Kconfig doc. Also, the code that disables the ACS bits is surrounded explicitly by an #ifdef * Removed the flag we added to rdma_rw_ctx() in favour of using is_pci_p2pdma_page(), as suggested by Sagi. * Adjust pci_p2pmem_find() so that it prefers P2P providers that are closest to (or the same as) the clients using them. In cases of ties, the provider is randomly chosen. * Modify the NVMe Target code so that the PCI device name of the provider may be explicitly specified, bypassing the logic in pci_p2pmem_find(). (Note: it's still enforced that the provider must be behind the same switch as the clients). * As requested by Bjorn, added documentation for driver writers. Changes in v2: * Renamed everything to 'p2pdma' per the suggestion from Bjorn as well as a bunch of cleanup and spelling fixes he pointed out in the last series. * To address Alex's ACS concerns, we change to a simpler method of just disabling ACS behind switches for any kernel that has CONFIG_PCI_P2PDMA. * We also reject using devices that employ 'dma_virt_ops' which should fairly simply handle Jason's concerns that this work might break with the HFI, QIB and rxe drivers that use the virtual ops to implement their own special DMA operations. -- This is a continuation of our work to enable using Peer-to-Peer PCI memory in the kernel with initial support for the NVMe fabrics target subsystem. Many thanks go to Christoph Hellwig who provided valuable feedback to get these patches to where they are today. The concept here is to use memory that's exposed on a PCI BAR as data buffers in the NVMe target code such that data can be transferred from an RDMA NIC to the special memory and then directly to an NVMe device avoiding system memory entirely. The upside of this is better QoS for applications running on the CPU utilizing memory and lower PCI bandwidth required to the CPU (such that systems could be designed with fewer lanes connected to the CPU). Due to these trade-offs we've designed the system to only enable using the PCI memory in cases where the NIC, NVMe devices and memory are all behind the same PCI switch hierarchy. This will mean many setups that could likely work well will not be supported so that we can be more confident it will work and not place any responsibility on the user to understand their topology. (We chose to go this route based on feedback we received at the last LSF). Future work may enable these transfers using a white list of known good root complexes. However, at this time, there is no reliable way to ensure that Peer-to-Peer transactions are permitted between PCI Root Ports. In order to enable this functionality, we introduce a few new PCI functions such that a driver can register P2P memory with the system. Struct pages are created for this memory using devm_memremap_pages() and the PCI bus offset is stored in the corresponding pagemap structure. When the PCI P2PDMA config option is selected the ACS bits in every bridge port in the system are turned off to allow traffic to pass freely behind the root port. At this time, the bit must be disabled at boot so the IOMMU subsystem can correctly create the groups, though this could be addressed in the future. There is no way to dynamically disable the bit and alter the groups. Another set of functions allow a client driver to create a list of client devices that will be used in a given P2P transactions and then use that list to find any P2P memory that is supported by all the client devices. In the block layer, we also introduce a P2P request flag to indicate a given request targets P2P memory as well as a flag for a request queue to indicate a given queue supports targeting P2P memory. P2P requests will only be accepted by queues that support it. Also, P2P requests are marked to not be merged seeing a non-homogenous request would complicate the DMA mapping requirements. In the PCI NVMe driver, we modify the existing CMB support to utilize the new PCI P2P memory infrastructure and also add support for P2P memory in its request queue. When a P2P request is received it uses the pci_p2pmem_map_sg() function which applies the necessary transformation to get the corrent pci_bus_addr_t for the DMA transactions. In the RDMA core, we also adjust rdma_rw_ctx_init() and rdma_rw_ctx_destroy() to take a flags argument which indicates whether to use the PCI P2P mapping functions or not. To avoid odd RDMA devices that don't use the proper DMA infrastructure this code rejects using any device that employs the virt_dma_ops implementation. Finally, in the NVMe fabrics target port we introduce a new configuration boolean: 'allow_p2pmem'. When set, the port will attempt to find P2P memory supported by the RDMA NIC and all namespaces. If supported memory is found, it will be used in all IO transfers. And if a port is using P2P memory, adding new namespaces that are not supported by that memory will fail. These patches have been tested on a number of Intel based systems and for a variety of RDMA NICs (Mellanox, Broadcomm, Chelsio) and NVMe SSDs (Intel, Seagate, Samsung) and p2pdma devices (Eideticom, Microsemi, Chelsio and Everspin) using switches from both Microsemi and Broadcomm. Logan Gunthorpe (13): PCI/P2PDMA: Support peer-to-peer memory PCI/P2PDMA: Add sysfs group to display p2pmem stats PCI/P2PDMA: Add PCI p2pmem DMA mappings to adjust the bus offset PCI/P2PDMA: Introduce configfs/sysfs enable attribute helpers docs-rst: Add a new directory for PCI documentation PCI/P2PDMA: Add P2P DMA driver writer's documentation block: Add PCI P2P flag for request queue and check support for requests IB/core: Ensure we map P2P memory correctly in rdma_rw_ctx_[init|destroy]() nvme-pci: Use PCI p2pmem subsystem to manage the CMB nvme-pci: Add support for P2P memory in requests nvme-pci: Add a quirk for a pseudo CMB nvmet: Introduce helper functions to allocate and free request SGLs nvmet: Optionally use PCI P2P memory Documentation/ABI/testing/sysfs-bus-pci | 25 + Documentation/driver-api/index.rst | 2 +- Documentation/driver-api/pci/index.rst | 21 + Documentation/driver-api/pci/p2pdma.rst | 170 ++++++ Documentation/driver-api/{ => pci}/pci.rst | 0 block/blk-core.c | 14 + drivers/infiniband/core/rw.c | 11 +- drivers/nvme/host/core.c | 4 + drivers/nvme/host/nvme.h | 8 + drivers/nvme/host/pci.c | 121 ++-- drivers/nvme/target/configfs.c | 36 ++ drivers/nvme/target/core.c | 149 +++++ drivers/nvme/target/nvmet.h | 15 + drivers/nvme/target/rdma.c | 22 +- drivers/pci/Kconfig | 17 + drivers/pci/Makefile | 1 + drivers/pci/p2pdma.c | 941 +++++++++++++++++++++++++++++ include/linux/blkdev.h | 3 + include/linux/memremap.h | 6 + include/linux/mm.h | 18 + include/linux/pci-p2pdma.h | 124 ++++ include/linux/pci.h | 4 + 22 files changed, 1658 insertions(+), 54 deletions(-) create mode 100644 Documentation/driver-api/pci/index.rst create mode 100644 Documentation/driver-api/pci/p2pdma.rst rename Documentation/driver-api/{ => pci}/pci.rst (100%) create mode 100644 drivers/pci/p2pdma.c create mode 100644 include/linux/pci-p2pdma.h -- 2.11.0