diff options
| author | Linus Torvalds <torvalds@linux-foundation.org> | 2018-02-02 21:01:04 +0300 |
|---|---|---|
| committer | Linus Torvalds <torvalds@linux-foundation.org> | 2018-02-02 21:01:04 +0300 |
| commit | 03f51d4efa2287cc628bb20b0c032036d2a9e66a (patch) | |
| tree | ec7fb3b6624d53092e2768578f3ef887c8d77f22 /Documentation/accelerators | |
| parent | 367b0df173b0ebea5d18b6971c244e260b5feb17 (diff) | |
| parent | 015eb1b89e959c9349f0a01803fb8ed1ced36f09 (diff) | |
| download | linux-03f51d4efa2287cc628bb20b0c032036d2a9e66a.tar.xz | |
Merge tag 'powerpc-4.16-1' of git://git.kernel.org/pub/scm/linux/kernel/git/powerpc/linux
Pull powerpc updates from Michael Ellerman:
"Highlights:
- Enable support for memory protection keys aka "pkeys" on Power7/8/9
when using the hash table MMU.
- Extend our interrupt soft masking to support masking PMU interrupts
as well as "normal" interrupts, and then use that to implement
local_t for a ~4x speedup vs the current atomics-based
implementation.
- A new driver "ocxl" for "Open Coherent Accelerator Processor
Interface (OpenCAPI)" devices.
- Support for new device tree properties on PowerVM to describe
hotpluggable memory and devices.
- Add support for CLOCK_{REALTIME/MONOTONIC}_COARSE to the 64-bit
VDSO.
- Freescale updates from Scott: fixes for CPM GPIO and an FSL PCI
erratum workaround, plus a minor cleanup patch.
As well as quite a lot of other changes all over the place, and small
fixes and cleanups as always.
Thanks to: Alan Modra, Alastair D'Silva, Alexey Kardashevskiy,
Alistair Popple, Andreas Schwab, Andrew Donnellan, Aneesh Kumar K.V,
Anju T Sudhakar, Anshuman Khandual, Anton Blanchard, Arnd Bergmann,
Balbir Singh, Benjamin Herrenschmidt, Bhaktipriya Shridhar, Bryant G.
Ly, Cédric Le Goater, Christophe Leroy, Christophe Lombard, Cyril Bur,
David Gibson, Desnes A. Nunes do Rosario, Dmitry Torokhov, Frederic
Barrat, Geert Uytterhoeven, Guilherme G. Piccoli, Gustavo A. R. Silva,
Gustavo Romero, Ivan Mikhaylov, Joakim Tjernlund, Joe Perches, Josh
Poimboeuf, Juan J. Alvarez, Julia Cartwright, Kamalesh Babulal,
Madhavan Srinivasan, Mahesh Salgaonkar, Mathieu Malaterre, Michael
Bringmann, Michael Hanselmann, Michael Neuling, Nathan Fontenot,
Naveen N. Rao, Nicholas Piggin, Paul Mackerras, Philippe Bergheaud,
Ram Pai, Russell Currey, Santosh Sivaraj, Scott Wood, Seth Forshee,
Simon Guo, Stewart Smith, Sukadev Bhattiprolu, Thiago Jung Bauermann,
Vaibhav Jain, Vasyl Gomonovych"
* tag 'powerpc-4.16-1' of git://git.kernel.org/pub/scm/linux/kernel/git/powerpc/linux: (199 commits)
powerpc/mm/radix: Fix build error when RADIX_MMU=n
macintosh/ams-input: Use true and false for boolean values
macintosh: change some data types from int to bool
powerpc/watchdog: Print the NIP in soft_nmi_interrupt()
powerpc/watchdog: regs can't be null in soft_nmi_interrupt()
powerpc/watchdog: Tweak watchdog printks
powerpc/cell: Remove axonram driver
rtc-opal: Fix handling of firmware error codes, prevent busy loops
powerpc/mpc52xx_gpt: make use of raw_spinlock variants
macintosh/adb: Properly mark continued kernel messages
powerpc/pseries: Fix cpu hotplug crash with memoryless nodes
powerpc/numa: Ensure nodes initialized for hotplug
powerpc/numa: Use ibm,max-associativity-domains to discover possible nodes
powerpc/kernel: Block interrupts when updating TIDR
powerpc/powernv/idoa: Remove unnecessary pcidev from pci_dn
powerpc/mm/nohash: do not flush the entire mm when range is a single page
powerpc/pseries: Add Initialization of VF Bars
powerpc/pseries/pci: Associate PEs to VFs in configure SR-IOV
powerpc/eeh: Add EEH notify resume sysfs
powerpc/eeh: Add EEH operations to notify resume
...
Diffstat (limited to 'Documentation/accelerators')
| -rw-r--r-- | Documentation/accelerators/ocxl.rst | 160 |
1 files changed, 160 insertions, 0 deletions
diff --git a/Documentation/accelerators/ocxl.rst b/Documentation/accelerators/ocxl.rst new file mode 100644 index 000000000000..4f7af841d935 --- /dev/null +++ b/Documentation/accelerators/ocxl.rst @@ -0,0 +1,160 @@ +======================================================== +OpenCAPI (Open Coherent Accelerator Processor Interface) +======================================================== + +OpenCAPI is an interface between processors and accelerators. It aims +at being low-latency and high-bandwidth. The specification is +developed by the `OpenCAPI Consortium <http://opencapi.org/>`_. + +It allows an accelerator (which could be a FPGA, ASICs, ...) to access +the host memory coherently, using virtual addresses. An OpenCAPI +device can also host its own memory, that can be accessed from the +host. + +OpenCAPI is known in linux as 'ocxl', as the open, processor-agnostic +evolution of 'cxl' (the driver for the IBM CAPI interface for +powerpc), which was named that way to avoid confusion with the ISDN +CAPI subsystem. + + +High-level view +=============== + +OpenCAPI defines a Data Link Layer (DL) and Transaction Layer (TL), to +be implemented on top of a physical link. Any processor or device +implementing the DL and TL can start sharing memory. + +:: + + +-----------+ +-------------+ + | | | | + | | | Accelerated | + | Processor | | Function | + | | +--------+ | Unit | +--------+ + | |--| Memory | | (AFU) |--| Memory | + | | +--------+ | | +--------+ + +-----------+ +-------------+ + | | + +-----------+ +-------------+ + | TL | | TLX | + +-----------+ +-------------+ + | | + +-----------+ +-------------+ + | DL | | DLX | + +-----------+ +-------------+ + | | + | PHY | + +---------------------------------------+ + + + +Device discovery +================ + +OpenCAPI relies on a PCI-like configuration space, implemented on the +device. So the host can discover AFUs by querying the config space. + +OpenCAPI devices in Linux are treated like PCI devices (with a few +caveats). The firmware is expected to abstract the hardware as if it +was a PCI link. A lot of the existing PCI infrastructure is reused: +devices are scanned and BARs are assigned during the standard PCI +enumeration. Commands like 'lspci' can therefore be used to see what +devices are available. + +The configuration space defines the AFU(s) that can be found on the +physical adapter, such as its name, how many memory contexts it can +work with, the size of its MMIO areas, ... + + + +MMIO +==== + +OpenCAPI defines two MMIO areas for each AFU: + +* the global MMIO area, with registers pertinent to the whole AFU. +* a per-process MMIO area, which has a fixed size for each context. + + + +AFU interrupts +============== + +OpenCAPI includes the possibility for an AFU to send an interrupt to a +host process. It is done through a 'intrp_req' defined in the +Transaction Layer, specifying a 64-bit object handle which defines the +interrupt. + +The driver allows a process to allocate an interrupt and obtain its +64-bit object handle, that can be passed to the AFU. + + + +char devices +============ + +The driver creates one char device per AFU found on the physical +device. A physical device may have multiple functions and each +function can have multiple AFUs. At the time of this writing though, +it has only been tested with devices exporting only one AFU. + +Char devices can be found in /dev/ocxl/ and are named as: +/dev/ocxl/<AFU name>.<location>.<index> + +where <AFU name> is a max 20-character long name, as found in the +config space of the AFU. +<location> is added by the driver and can help distinguish devices +when a system has more than one instance of the same OpenCAPI device. +<index> is also to help distinguish AFUs in the unlikely case where a +device carries multiple copies of the same AFU. + + + +Sysfs class +=========== + +An ocxl class is added for the devices representing the AFUs. See +/sys/class/ocxl. The layout is described in +Documentation/ABI/testing/sysfs-class-ocxl + + + +User API +======== + +open +---- + +Based on the AFU definition found in the config space, an AFU may +support working with more than one memory context, in which case the +associated char device may be opened multiple times by different +processes. + + +ioctl +----- + +OCXL_IOCTL_ATTACH: + + Attach the memory context of the calling process to the AFU so that + the AFU can access its memory. + +OCXL_IOCTL_IRQ_ALLOC: + + Allocate an AFU interrupt and return an identifier. + +OCXL_IOCTL_IRQ_FREE: + + Free a previously allocated AFU interrupt. + +OCXL_IOCTL_IRQ_SET_FD: + + Associate an event fd to an AFU interrupt so that the user process + can be notified when the AFU sends an interrupt. + + +mmap +---- + +A process can mmap the per-process MMIO area for interactions with the +AFU. |