1# To Build 2Need `meson` and `ninja`. Alternatively, source an OpenBMC ARM/x86 SDK. 3``` 4meson build && ninja -C build 5``` 6## To run unit tests 7Tests can be run in the CI docker container, or with an OpenBMC x86 sdk(see 8below for x86 steps). 9``` 10meson -Doe-sdk=enabled -Dtests=enabled build 11ninja -C build test 12``` 13 14# Code Organization 15At a high-level, code in this repository belongs to one of the following three 16components. 17 18## libpldm 19This is a library which deals with the encoding and decoding of PLDM messages. 20It should be possible to use this library by projects other than OpenBMC, and 21hence certain constraints apply to it: 22- keeping it light weight 23- implementation in C 24- minimal dynamic memory allocations 25- endian-safe 26- no OpenBMC specific dependencies 27 28Source files are named according to the PLDM Type, for eg base.[h/c], fru.[h/c], 29etc. 30 31Given a PLDM command "foo", the library will provide the following API: 32For the Requester function: 33``` 34encode_foo_req() - encode a foo request 35decode_foo_resp() - decode a response to foo 36``` 37For the Responder function: 38``` 39decode_foo_req() - decode a foo request 40encode_foo_resp() - encode a response to foo 41``` 42The library also provides API to pack and unpack PLDM headers. 43 44## libpldmresponder 45This library provides handlers for incoming PLDM request messages. It provides 46for a registration as well as a plug-in mechanism. The library is implemented in 47modern C++, and handles OpenBMC's platform specifics. 48 49The handlers are of the form 50``` 51Response handler(Request payload, size_t payloadLen) 52``` 53 54Source files are named according to the PLDM Type, for eg base.[hpp/cpp], 55fru.[hpp/cpp], etc. 56 57 58## OEM/vendor-specific functions 59This will support OEM or vendor-specific functions and semantic information. 60Following directory structure has to be used: 61``` 62 pldm repo 63 |---- oem 64 |----<oem_name> 65 |----libpldm 66 |----<oem based encoding and decoding files> 67 |----libpldmresponder 68 |---<oem based handler files> 69 70``` 71<oem_name> - This folder must be created with the name of the OEM/vendor 72in lower case. Folders named libpldm and libpldmresponder must be created under 73the folder <oem_name> 74 75Files having the oem functionality for the libpldm library should be placed 76under the folder oem/<oem_name>/libpldm. They must be adhering to the rules 77mentioned under the libpldm section above. 78 79Files having the oem functionality for the libpldmresponder library should be 80placed under the folder oem/<oem_name>/libpldmresponder. They must be adhering 81to the rules mentioned under the libpldmresponder section above. 82 83Once the above is done a meson option has to be created in 84`pldm/meson_options.txt` with its mapped compiler flag to enable conditional 85compilation. 86 87For consistency would recommend using "oem-<oem_name>". 88 89The `pldm/meson.build` and the corresponding source file(s) will need to 90incorporate the logic of adding its mapped compiler flag to allow conditional 91compilation of the code. 92 93## pldmtool 94For more information on pldmtool please refer to plmdtool/README.md. 95 96## TODO 97Consider hosting libpldm above in a repo of its own, probably even outside the 98OpenBMC project? A separate repo would enable something like git submodule. 99 100# Flows 101This section documents important code flow paths. 102 103## BMC as PLDM responder 104a) PLDM daemon receives PLDM request message from underlying transport (MCTP). 105 106b) PLDM daemon routes message to message handler, based on the PLDM command. 107 108c) Message handler decodes request payload into various field(s) of the request 109 message. It can make use of a decode_foo_req() API, and doesn't have to 110 perform deserialization of the request payload by itself. 111 112d) Message handler works with the request field(s) and generates response 113 field(s). 114 115e) Message handler prepares a response message. It can make use of an 116 encode_foo_resp() API, and doesn't have to perform the serialization of the 117 response field(s) by itself. 118 119f) The PLDM daemon sends the response message prepared at step e) to the remote 120 PLDM device. 121 122## BMC as PLDM requester 123a) A BMC PLDM requester app prepares a PLDM request message. There would be 124 several requester apps (based on functionality/PLDM remote device). Each of 125 them needn't bother with the serialization of request field(s), and can 126 instead make use of an encode_foo_req() API. 127 128b) BMC requester app requests PLDM daemon to send the request message to remote 129 PLDM device. 130 131c) Once the PLDM daemon receives a corresponding response message, it notifies 132 the requester app. 133 134d) The requester app has to work with the response field(s). It can make use of 135 a decode_foo_resp() API to deserialize the response message. 136 137# PDR Implementation 138While PLDM Platform Descriptor Records (PDRs) are mostly static information, 139they can vary across platforms and systems. For this reason, platform specific 140PDR information is encoded in platform specific JSON files. JSON files must be 141named based on the PDR type number. For example a state effecter PDR JSON file 142will be named 11.json. The JSON files may also include information to enable 143additional processing (apart from PDR creation) for specific PDR types, for eg 144mapping an effecter id to a D-Bus object. 145 146The PLDM responder implementation finds and parses PDR JSON files to create the 147PDR repository. Platform specific PDR modifications would likely just result in 148JSON updates. New PDR type support would require JSON updates as well as PDR 149generation code. The PDR generator is a map of PDR Type -> C++ lambda to create 150PDR entries for that type based on the JSON, and to update the central PDR repo. 151