Linus Torvalds | 1da177e | 2005-04-16 15:20:36 -0700 | [diff] [blame] | 1 | USERSPACE MAD ACCESS |
| 2 | |
| 3 | Device files |
| 4 | |
| 5 | Each port of each InfiniBand device has a "umad" device and an |
| 6 | "issm" device attached. For example, a two-port HCA will have two |
| 7 | umad devices and two issm devices, while a switch will have one |
| 8 | device of each type (for switch port 0). |
| 9 | |
| 10 | Creating MAD agents |
| 11 | |
| 12 | A MAD agent can be created by filling in a struct ib_user_mad_reg_req |
| 13 | and then calling the IB_USER_MAD_REGISTER_AGENT ioctl on a file |
| 14 | descriptor for the appropriate device file. If the registration |
| 15 | request succeeds, a 32-bit id will be returned in the structure. |
| 16 | For example: |
| 17 | |
| 18 | struct ib_user_mad_reg_req req = { /* ... */ }; |
| 19 | ret = ioctl(fd, IB_USER_MAD_REGISTER_AGENT, (char *) &req); |
| 20 | if (!ret) |
| 21 | my_agent = req.id; |
| 22 | else |
| 23 | perror("agent register"); |
| 24 | |
| 25 | Agents can be unregistered with the IB_USER_MAD_UNREGISTER_AGENT |
| 26 | ioctl. Also, all agents registered through a file descriptor will |
| 27 | be unregistered when the descriptor is closed. |
| 28 | |
| 29 | Receiving MADs |
| 30 | |
Hal Rosenstock | 3f75dad | 2005-07-27 11:45:41 -0700 | [diff] [blame] | 31 | MADs are received using read(). The receive side now supports |
| 32 | RMPP. The buffer passed to read() must be at least one |
| 33 | struct ib_user_mad + 256 bytes. For example: |
Linus Torvalds | 1da177e | 2005-04-16 15:20:36 -0700 | [diff] [blame] | 34 | |
Hal Rosenstock | 3f75dad | 2005-07-27 11:45:41 -0700 | [diff] [blame] | 35 | If the buffer passed is not large enough to hold the received |
| 36 | MAD (RMPP), the errno is set to ENOSPC and the length of the |
| 37 | buffer needed is set in mad.length. |
| 38 | |
| 39 | Example for normal MAD (non RMPP) reads: |
| 40 | struct ib_user_mad *mad; |
| 41 | mad = malloc(sizeof *mad + 256); |
| 42 | ret = read(fd, mad, sizeof *mad + 256); |
| 43 | if (ret != sizeof mad + 256) { |
Linus Torvalds | 1da177e | 2005-04-16 15:20:36 -0700 | [diff] [blame] | 44 | perror("read"); |
Hal Rosenstock | 3f75dad | 2005-07-27 11:45:41 -0700 | [diff] [blame] | 45 | free(mad); |
| 46 | } |
| 47 | |
| 48 | Example for RMPP reads: |
| 49 | struct ib_user_mad *mad; |
| 50 | mad = malloc(sizeof *mad + 256); |
| 51 | ret = read(fd, mad, sizeof *mad + 256); |
| 52 | if (ret == -ENOSPC)) { |
| 53 | length = mad.length; |
| 54 | free(mad); |
| 55 | mad = malloc(sizeof *mad + length); |
| 56 | ret = read(fd, mad, sizeof *mad + length); |
| 57 | } |
| 58 | if (ret < 0) { |
| 59 | perror("read"); |
| 60 | free(mad); |
| 61 | } |
Linus Torvalds | 1da177e | 2005-04-16 15:20:36 -0700 | [diff] [blame] | 62 | |
| 63 | In addition to the actual MAD contents, the other struct ib_user_mad |
| 64 | fields will be filled in with information on the received MAD. For |
| 65 | example, the remote LID will be in mad.lid. |
| 66 | |
| 67 | If a send times out, a receive will be generated with mad.status set |
| 68 | to ETIMEDOUT. Otherwise when a MAD has been successfully received, |
| 69 | mad.status will be 0. |
| 70 | |
| 71 | poll()/select() may be used to wait until a MAD can be read. |
| 72 | |
| 73 | Sending MADs |
| 74 | |
| 75 | MADs are sent using write(). The agent ID for sending should be |
| 76 | filled into the id field of the MAD, the destination LID should be |
Hal Rosenstock | 3f75dad | 2005-07-27 11:45:41 -0700 | [diff] [blame] | 77 | filled into the lid field, and so on. The send side does support |
| 78 | RMPP so arbitrary length MAD can be sent. For example: |
Linus Torvalds | 1da177e | 2005-04-16 15:20:36 -0700 | [diff] [blame] | 79 | |
Hal Rosenstock | 3f75dad | 2005-07-27 11:45:41 -0700 | [diff] [blame] | 80 | struct ib_user_mad *mad; |
Linus Torvalds | 1da177e | 2005-04-16 15:20:36 -0700 | [diff] [blame] | 81 | |
Hal Rosenstock | 3f75dad | 2005-07-27 11:45:41 -0700 | [diff] [blame] | 82 | mad = malloc(sizeof *mad + mad_length); |
Linus Torvalds | 1da177e | 2005-04-16 15:20:36 -0700 | [diff] [blame] | 83 | |
Hal Rosenstock | 3f75dad | 2005-07-27 11:45:41 -0700 | [diff] [blame] | 84 | /* fill in mad->data */ |
| 85 | |
| 86 | mad->hdr.id = my_agent; /* req.id from agent registration */ |
| 87 | mad->hdr.lid = my_dest; /* in network byte order... */ |
Linus Torvalds | 1da177e | 2005-04-16 15:20:36 -0700 | [diff] [blame] | 88 | /* etc. */ |
| 89 | |
Hal Rosenstock | 3f75dad | 2005-07-27 11:45:41 -0700 | [diff] [blame] | 90 | ret = write(fd, &mad, sizeof *mad + mad_length); |
| 91 | if (ret != sizeof *mad + mad_length) |
Linus Torvalds | 1da177e | 2005-04-16 15:20:36 -0700 | [diff] [blame] | 92 | perror("write"); |
| 93 | |
| 94 | Setting IsSM Capability Bit |
| 95 | |
| 96 | To set the IsSM capability bit for a port, simply open the |
| 97 | corresponding issm device file. If the IsSM bit is already set, |
| 98 | then the open call will block until the bit is cleared (or return |
| 99 | immediately with errno set to EAGAIN if the O_NONBLOCK flag is |
| 100 | passed to open()). The IsSM bit will be cleared when the issm file |
| 101 | is closed. No read, write or other operations can be performed on |
| 102 | the issm file. |
| 103 | |
| 104 | /dev files |
| 105 | |
| 106 | To create the appropriate character device files automatically with |
| 107 | udev, a rule like |
| 108 | |
| 109 | KERNEL="umad*", NAME="infiniband/%k" |
| 110 | KERNEL="issm*", NAME="infiniband/%k" |
| 111 | |
| 112 | can be used. This will create device nodes named |
| 113 | |
| 114 | /dev/infiniband/umad0 |
| 115 | /dev/infiniband/issm0 |
| 116 | |
| 117 | for the first port, and so on. The InfiniBand device and port |
| 118 | associated with these devices can be determined from the files |
| 119 | |
| 120 | /sys/class/infiniband_mad/umad0/ibdev |
| 121 | /sys/class/infiniband_mad/umad0/port |
| 122 | |
| 123 | and |
| 124 | |
| 125 | /sys/class/infiniband_mad/issm0/ibdev |
| 126 | /sys/class/infiniband_mad/issm0/port |