linux/Documentation/ntb.txt
<<
>>
Prefs
   1===========
   2NTB Drivers
   3===========
   4
   5NTB (Non-Transparent Bridge) is a type of PCI-Express bridge chip that connects
   6the separate memory systems of two or more computers to the same PCI-Express
   7fabric. Existing NTB hardware supports a common feature set: doorbell
   8registers and memory translation windows, as well as non common features like
   9scratchpad and message registers. Scratchpad registers are read-and-writable
  10registers that are accessible from either side of the device, so that peers can
  11exchange a small amount of information at a fixed address. Message registers can
  12be utilized for the same purpose. Additionally they are provided with with
  13special status bits to make sure the information isn't rewritten by another
  14peer. Doorbell registers provide a way for peers to send interrupt events.
  15Memory windows allow translated read and write access to the peer memory.
  16
  17NTB Core Driver (ntb)
  18=====================
  19
  20The NTB core driver defines an api wrapping the common feature set, and allows
  21clients interested in NTB features to discover NTB the devices supported by
  22hardware drivers.  The term "client" is used here to mean an upper layer
  23component making use of the NTB api.  The term "driver," or "hardware driver,"
  24is used here to mean a driver for a specific vendor and model of NTB hardware.
  25
  26NTB Client Drivers
  27==================
  28
  29NTB client drivers should register with the NTB core driver.  After
  30registering, the client probe and remove functions will be called appropriately
  31as ntb hardware, or hardware drivers, are inserted and removed.  The
  32registration uses the Linux Device framework, so it should feel familiar to
  33anyone who has written a pci driver.
  34
  35NTB Typical client driver implementation
  36----------------------------------------
  37
  38Primary purpose of NTB is to share some peace of memory between at least two
  39systems. So the NTB device features like Scratchpad/Message registers are
  40mainly used to perform the proper memory window initialization. Typically
  41there are two types of memory window interfaces supported by the NTB API:
  42inbound translation configured on the local ntb port and outbound translation
  43configured by the peer, on the peer ntb port. The first type is
  44depicted on the next figure::
  45
  46 Inbound translation:
  47
  48 Memory:              Local NTB Port:      Peer NTB Port:      Peer MMIO:
  49  ____________
  50 | dma-mapped |-ntb_mw_set_trans(addr)  |
  51 | memory     |        _v____________   |   ______________
  52 | (addr)     |<======| MW xlat addr |<====| MW base addr |<== memory-mapped IO
  53 |------------|       |--------------|  |  |--------------|
  54
  55So typical scenario of the first type memory window initialization looks:
  561) allocate a memory region, 2) put translated address to NTB config,
  573) somehow notify a peer device of performed initialization, 4) peer device
  58maps corresponding outbound memory window so to have access to the shared
  59memory region.
  60
  61The second type of interface, that implies the shared windows being
  62initialized by a peer device, is depicted on the figure::
  63
  64 Outbound translation:
  65
  66 Memory:        Local NTB Port:    Peer NTB Port:      Peer MMIO:
  67  ____________                      ______________
  68 | dma-mapped |                |   | MW base addr |<== memory-mapped IO
  69 | memory     |                |   |--------------|
  70 | (addr)     |<===================| MW xlat addr |<-ntb_peer_mw_set_trans(addr)
  71 |------------|                |   |--------------|
  72
  73Typical scenario of the second type interface initialization would be:
  741) allocate a memory region, 2) somehow deliver a translated address to a peer
  75device, 3) peer puts the translated address to NTB config, 4) peer device maps
  76outbound memory window so to have access to the shared memory region.
  77
  78As one can see the described scenarios can be combined in one portable
  79algorithm.
  80
  81 Local device:
  82  1) Allocate memory for a shared window
  83  2) Initialize memory window by translated address of the allocated region
  84     (it may fail if local memory window initialization is unsupported)
  85  3) Send the translated address and memory window index to a peer device
  86
  87 Peer device:
  88  1) Initialize memory window with retrieved address of the allocated
  89     by another device memory region (it may fail if peer memory window
  90     initialization is unsupported)
  91  2) Map outbound memory window
  92
  93In accordance with this scenario, the NTB Memory Window API can be used as
  94follows:
  95
  96 Local device:
  97  1) ntb_mw_count(pidx) - retrieve number of memory ranges, which can
  98     be allocated for memory windows between local device and peer device
  99     of port with specified index.
 100  2) ntb_get_align(pidx, midx) - retrieve parameters restricting the
 101     shared memory region alignment and size. Then memory can be properly
 102     allocated.
 103  3) Allocate physically contiguous memory region in compliance with
 104     restrictions retrieved in 2).
 105  4) ntb_mw_set_trans(pidx, midx) - try to set translation address of
 106     the memory window with specified index for the defined peer device
 107     (it may fail if local translated address setting is not supported)
 108  5) Send translated base address (usually together with memory window
 109     number) to the peer device using, for instance, scratchpad or message
 110     registers.
 111
 112 Peer device:
 113  1) ntb_peer_mw_set_trans(pidx, midx) - try to set received from other
 114     device (related to pidx) translated address for specified memory
 115     window. It may fail if retrieved address, for instance, exceeds
 116     maximum possible address or isn't properly aligned.
 117  2) ntb_peer_mw_get_addr(widx) - retrieve MMIO address to map the memory
 118     window so to have an access to the shared memory.
 119
 120Also it is worth to note, that method ntb_mw_count(pidx) should return the
 121same value as ntb_peer_mw_count() on the peer with port index - pidx.
 122
 123NTB Transport Client (ntb\_transport) and NTB Netdev (ntb\_netdev)
 124------------------------------------------------------------------
 125
 126The primary client for NTB is the Transport client, used in tandem with NTB
 127Netdev.  These drivers function together to create a logical link to the peer,
 128across the ntb, to exchange packets of network data.  The Transport client
 129establishes a logical link to the peer, and creates queue pairs to exchange
 130messages and data.  The NTB Netdev then creates an ethernet device using a
 131Transport queue pair.  Network data is copied between socket buffers and the
 132Transport queue pair buffer.  The Transport client may be used for other things
 133besides Netdev, however no other applications have yet been written.
 134
 135NTB Ping Pong Test Client (ntb\_pingpong)
 136-----------------------------------------
 137
 138The Ping Pong test client serves as a demonstration to exercise the doorbell
 139and scratchpad registers of NTB hardware, and as an example simple NTB client.
 140Ping Pong enables the link when started, waits for the NTB link to come up, and
 141then proceeds to read and write the doorbell scratchpad registers of the NTB.
 142The peers interrupt each other using a bit mask of doorbell bits, which is
 143shifted by one in each round, to test the behavior of multiple doorbell bits
 144and interrupt vectors.  The Ping Pong driver also reads the first local
 145scratchpad, and writes the value plus one to the first peer scratchpad, each
 146round before writing the peer doorbell register.
 147
 148Module Parameters:
 149
 150* unsafe - Some hardware has known issues with scratchpad and doorbell
 151        registers.  By default, Ping Pong will not attempt to exercise such
 152        hardware.  You may override this behavior at your own risk by setting
 153        unsafe=1.
 154* delay\_ms - Specify the delay between receiving a doorbell
 155        interrupt event and setting the peer doorbell register for the next
 156        round.
 157* init\_db - Specify the doorbell bits to start new series of rounds.  A new
 158        series begins once all the doorbell bits have been shifted out of
 159        range.
 160* dyndbg - It is suggested to specify dyndbg=+p when loading this module, and
 161        then to observe debugging output on the console.
 162
 163NTB Tool Test Client (ntb\_tool)
 164--------------------------------
 165
 166The Tool test client serves for debugging, primarily, ntb hardware and drivers.
 167The Tool provides access through debugfs for reading, setting, and clearing the
 168NTB doorbell, and reading and writing scratchpads.
 169
 170The Tool does not currently have any module parameters.
 171
 172Debugfs Files:
 173
 174* *debugfs*/ntb\_tool/*hw*/
 175        A directory in debugfs will be created for each
 176        NTB device probed by the tool.  This directory is shortened to *hw*
 177        below.
 178* *hw*/db
 179        This file is used to read, set, and clear the local doorbell.  Not
 180        all operations may be supported by all hardware.  To read the doorbell,
 181        read the file.  To set the doorbell, write `s` followed by the bits to
 182        set (eg: `echo 's 0x0101' > db`).  To clear the doorbell, write `c`
 183        followed by the bits to clear.
 184* *hw*/mask
 185        This file is used to read, set, and clear the local doorbell mask.
 186        See *db* for details.
 187* *hw*/peer\_db
 188        This file is used to read, set, and clear the peer doorbell.
 189        See *db* for details.
 190* *hw*/peer\_mask
 191        This file is used to read, set, and clear the peer doorbell
 192        mask.  See *db* for details.
 193* *hw*/spad
 194        This file is used to read and write local scratchpads.  To read
 195        the values of all scratchpads, read the file.  To write values, write a
 196        series of pairs of scratchpad number and value
 197        (eg: `echo '4 0x123 7 0xabc' > spad`
 198        # to set scratchpads `4` and `7` to `0x123` and `0xabc`, respectively).
 199* *hw*/peer\_spad
 200        This file is used to read and write peer scratchpads.  See
 201        *spad* for details.
 202
 203NTB Hardware Drivers
 204====================
 205
 206NTB hardware drivers should register devices with the NTB core driver.  After
 207registering, clients probe and remove functions will be called.
 208
 209NTB Intel Hardware Driver (ntb\_hw\_intel)
 210------------------------------------------
 211
 212The Intel hardware driver supports NTB on Xeon and Atom CPUs.
 213
 214Module Parameters:
 215
 216* b2b\_mw\_idx
 217        If the peer ntb is to be accessed via a memory window, then use
 218        this memory window to access the peer ntb.  A value of zero or positive
 219        starts from the first mw idx, and a negative value starts from the last
 220        mw idx.  Both sides MUST set the same value here!  The default value is
 221        `-1`.
 222* b2b\_mw\_share
 223        If the peer ntb is to be accessed via a memory window, and if
 224        the memory window is large enough, still allow the client to use the
 225        second half of the memory window for address translation to the peer.
 226* xeon\_b2b\_usd\_bar2\_addr64
 227        If using B2B topology on Xeon hardware, use
 228        this 64 bit address on the bus between the NTB devices for the window
 229        at BAR2, on the upstream side of the link.
 230* xeon\_b2b\_usd\_bar4\_addr64 - See *xeon\_b2b\_bar2\_addr64*.
 231* xeon\_b2b\_usd\_bar4\_addr32 - See *xeon\_b2b\_bar2\_addr64*.
 232* xeon\_b2b\_usd\_bar5\_addr32 - See *xeon\_b2b\_bar2\_addr64*.
 233* xeon\_b2b\_dsd\_bar2\_addr64 - See *xeon\_b2b\_bar2\_addr64*.
 234* xeon\_b2b\_dsd\_bar4\_addr64 - See *xeon\_b2b\_bar2\_addr64*.
 235* xeon\_b2b\_dsd\_bar4\_addr32 - See *xeon\_b2b\_bar2\_addr64*.
 236* xeon\_b2b\_dsd\_bar5\_addr32 - See *xeon\_b2b\_bar2\_addr64*.
 237