aboutsummaryrefslogtreecommitdiffstats
path: root/host/lib
Commit message (Collapse)AuthorAgeFilesLines
* usrp: Add I/O service manager for DPDKCiro Nishiguchi2019-12-203-0/+93
|
* lib: Remove dpdk_zero_copy filesMartin Braun2019-12-2014-3420/+0
| | | | These were left here as a reference.
* x300,mpmd: Enable DPDKMartin Braun2019-12-2012-465/+216
| | | | | | | | | | | | | | x300: - Remove obsolete variables from x300_eth_mgr and X300 motherboard components - Added some documentation / comments - Use constrained device args in more places - Enables the use of use_dpdk=1 - Switches between regular (kernel-based) and DPDK UDP mpmd: - Merge link_if_ctrl for udp and dpdk_udp - Update cmake options
* dpdk: Add new DPDK stack to integrate with I/O servicesAlex Williams2019-12-2015-343/+2557
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | docs: Update DPDK docs with new parameters: Parameter names have had their hyphens changed to underscores, and the I/O CPU argument is now named after the lcores and reflects the naming used by DPDK. transport: Add new udp_dpdk_link, based atop the new APIs: This link is tightly coupled with the DPDK I/O service. The link class carries all the address information to communicate with the other host, and it can send packets directly through the DPDK NIC ports. However, for receiving packets, the I/O service must pull the packets from the DMA queue and attach them to the appropriate link object. The link object merely formats the frame_buff object underneath, which is embedded in the rte_mbuf container. For get_recv_buff, the link will pull buffers only from its internal queue (the one filled by the I/O service). transport: Add DPDK-specific I/O service: The I/O service is split into two parts, the user threads and the I/O worker threads. The user threads submit requests through various appropriate queues, and the I/O threads perform all the I/O on their behalf. This includes routing UDP packets to the correct receiver and getting the MAC address of a destination (by performing the ARP request and handling the ARP replies). The DPDK context stores I/O services. The context spawns all I/O services on init(), and I/O services can be fetched from the dpdk_ctx object by using a port ID. I/O service clients: The clients have two lockless ring buffers. One is to get a buffer from the I/O service; the other is to release a buffer back to the I/O service. Threads sleeping on buffer I/O are kept in a separate list from the service queue and are processed in the course of doing RX or TX. The list nodes are embedded in the dpdk_io_if, and the head of the list is on the dpdk_io_service. The I/O service will transfer the embedded wait_req to the list if it cannot acquire the mutex to complete the condition for waking. Co-authored-by: Martin Braun <martin.braun@ettus.com> Co-authored-by: Ciro Nishiguchi <ciro.nishiguchi@ni.com> Co-authored-by: Brent Stapleton <brent.stapleton@ettus.com>
* transport: Fix uninitialized offload I/O service _client_typeAlex Williams2019-12-201-5/+2
| | | | | | The client type is kept in _offload_thread_params, but a separate, unitialized field is checked in member functions. Remove the duplicate and switch the checks over. This fixes offload_io_srv_test.
* prefs: Output debug trace when config file cannot be located or loadedAaron Rossetto2019-12-172-8/+14
|
* ddc: fixup decimation propertyBrent Stapleton2019-12-051-2/+2
| | | | | | | | - Use coerce_decim(), not int(), to coerce the decimation rate - Fix typo Fixes: a36fc4604f9b ("rfnoc: DDC: Fix property propagation") Fixes: 36853315741a ("rfnoc: Add DDC block controller")
* prefs: Set init_done flag to true after loading config filesMartin Braun2019-12-051-0/+1
| | | | | | | The prefs API is supposed to load the config files once, and stash them away for the process to consume at will. Because the init_done is never set, it will read the config files every time it's asked for them. This is usually not a problem, but it causes the logging output to be messy.
* e31x: Fix filter bank and antenna switching for channel 0Sugandha Gupta2019-12-023-43/+60
| | | | | | The filter bank and antenna switches have different configuration for channel 0 and channel 1. This commit fixes the issue where channel 0 produces only noise due to incorrect switches.
* rfnoc: DDC/DUC: Fix fp-issues with samp_rate propertiesBrent Stapleton2019-11-272-6/+67
| | | | | | Only update DDC/DUC samp_rate properties if the number is substantially different (don't update for sub-1Hz property calculations). This fixes resolver exceptions for certain rates.
* ddc: Applying clang-formatBrent Stapleton2019-11-271-8/+7
|
* rfnoc: graph: Add logging to resolver errorsBrent Stapleton2019-11-271-1/+6
|
* utils: tasks: Use uhd::set_thread_name()Martin Braun2019-11-261-3/+1
| | | | | This is a more portable option to set thread names. References to pthreads are now limited to thread.cpp, where they belong.
* mpmd: Name the claimer task threadMartin Braun2019-11-261-13/+15
| | | | | We call set_thread_name() on the claimer loop so the thread can be identified using OS utilities.
* x300: pcie manager updatesVirendra Kakade2019-11-262-129/+93
| | | | Signed-off-by: Virendra Kakade <virendra.kakade@ni.com>
* rfnoc: fix packet swapping from U32 to U64Virendra Kakade2019-11-261-5/+8
| | | | Signed-off-by: Virendra Kakade <virendra.kakade@ni.com>
* rfnoc: fix incorrect long fifo wait timeoutsVirendra Kakade2019-11-261-2/+2
| | | | | | | This fixes behavior where we would get long 100 sec timeouts on fifo waits instead of 100 ms timeouts. Signed-off-by: Virendra Kakade <virendra.kakade@ni.com>
* mg: Turn the set-lock into a recursive mutexMartin Braun2019-11-262-15/+15
| | | | | Individual API calls might have to call each other (e.g., like set_rate() will call set_rx_frequency()), which would cause a deadlock.
* mg: Always set MCR on both daughterboardsMartin Braun2019-11-263-12/+33
| | | | | | | | | | | | | | | | | The N310 cannot set the MCR for its daughterboards separately. This patch modifies the radio block controller such that any block controller, when requested to change the master clock rate, will first change Radio 0, and then Radio 1. This fixes the following issues: - In multi_usrp, calling set_master_clock_rate() will not necessarily call set_rate() on the radios in any particular order, which will break when calling Radio 1 first - In RFNoC apps, it wasn't possible to run off of slot B alone without this change. Note: When calling set_rate() on one radio, the other radio is in an invalid state until its set_rate() is also called.
* lib: Add DPDK service queueAlex Williams2019-11-261-0/+237
| | | | | | | This is a data structure intended for use by the DPDK I/O service. It uses DPDK's lockless ring in multi-producer, single-consumer mode to allow clients to submit requests to the DPDK I/O service's worker thread. Clients can specify a timeout for the requests to be fulfilled.
* rfnoc: fix invalid access when sending packet with no samplesCiro Nishiguchi2019-11-261-1/+3
|
* mpmd: Fix get_mboard_name()Martin Braun2019-11-261-1/+1
| | | | | It used to produce the individual name of the USRP, but it should return a product name.
* n310: Fix GPIO registersMartin Braun2019-11-261-3/+6
| | | | | This enables the use of the dboard and FP GPIOs. The problem was that the register offset of 8 was not encoded.
* mg/rh/rfnoc: Harmonize peripheral registersMartin Braun2019-11-264-21/+29
| | | | | - Move the SPI addresses out of radio_control_impl - Fix the GPIO address spaces for N310/N300
* gpio_atr_3000: Fix return value for pin control registerMartin Braun2019-11-261-1/+3
| | | | | | | | | | | | | | | This fixes a bug where get_gpio_attr(bank, "CTRL") would return the inverted value of what was written. Reason is that the underlying register was an ATR disable register. The fix is to invert the cached values of the register. Now, the following Python code will work: >>> U = uhd.usrp.MultiUSRP("type=x300") >>> atr_enable = 0xF # Enable ATR on lower 4 pins, rest is GPIO >>> U.set_gpio_attr("FP0A", "CTRL", atr_enable) >>> U.get_gpio_attr("FP0A", "CTRL") == atr_enable True
* detail::graph: Add shutdown capabilityMartin Braun2019-11-263-14/+35
| | | | | | | | | | | | In the existing graph, when the shutdown was simply a release. However, any outstanding actions would trigger warnings (because released graphs aren't supposed to still have actions being passed around), which would sometimes be visible at the end of an application. This is a safer solution than simply releasing, because it explicitly sets a shutdown flag that all graph-affecting functions (property propagation and action handling) respect. Once the flag is set, the graph can no longer be booted up again.
* e310: Fix issues in MPM and UHDMartin Braun2019-11-261-7/+3
| | | | | | | | | | | - Remove superfluous INFO logging - Improve formatting in many places - Improve Pylint score in various places - Add tear_down to DB object - Simplify custom EEPROM code for E310 - Fix time source selection code - Remove references to GPS_CTRL and GPS_STATUS (are E320 only) - Move clock source control out of MboardRegs object
* mpm/mpmd: Remove token requirement for device infoMartin Braun2019-11-261-2/+2
| | | | | This removes the token requirement for get_proto_ver() and get_chdr_width().
* rfnoc: Rename thread affinity argsCiro Nishiguchi2019-11-263-73/+73
| | | | | | Rename thread affinity args such that they do not end with an integer. Arg names ending with an integer are interpreted as being targeted at a specific motherboard index in device_addr methods.
* uhd: Allow device args that have an integer in the nameCiro Nishiguchi2019-11-261-1/+5
| | | | | | Allow device args that contain an integer within the name, such as recv_offload_thread_0_cpu. Previously this would cause an invalid format exception.
* lib,tests: Remove old DPDK files from buildAlex Williams2019-11-263-21/+8
| | | | The DPDK files are left behind as a reference, for now.
* transport: Add new base for DPDK links, based on 18.11Alex Williams2019-11-263-0/+956
| | | | | | | | | | | | dpdk_ctx represents the central context and manager of all memory and threads allocated via the DPDK EAL. In this commit, it parses the user's arguments, configures all the ports, and brings them up. dpdk_port represents each DPDK NIC port's configuration, and it manages the allocation of individual queues and their flow rules. It also would provide access to an ARP table and functions for handling ARP requests and responses. The flow rules and ARP functions are not yet implemented.
* docs: Change DPDK version to 18.11 and make args use underscoresAlex Williams2019-11-262-2/+2
| | | | | Swap out hyphens for underscores in the DPDK args. Also update list of distributions with the correct DPDK version in the repos.
* rfnoc_graph: Modify find_blocks() to sort return valueMartin Braun2019-11-262-35/+7
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | With this patch, the elements of of the return value of find_blocks() are sorted lexicographically (specifically, using uhd::rfnoc::block_id_it::operator<()). The underlying block_container class stores the blocks in an unordered set, so the return value for find_blocks() was always sorted randomly. multi_usrp_rfnoc had to sort the return values every time find_blocks() was used to get a useful return value. Because find_blocks() had no contract for the order of returned blocks, this change simply sorts the return value before returning it. multi_usrp_rfnoc is modified to remove all the sorts that are now superfluous. A good way to see the change is to run uhd_usrp_probe, which will now contain content like this: | _____________________________________________________ | / | | RFNoC blocks on this device: | | | | * 0/DDC#0 | | * 0/DDC#1 | | * 0/DUC#0 | | * 0/DUC#1 | | * 0/DmaFIFO#0 | | * 0/Radio#0 | | * 0/Radio#1 Assuming the blocks don't change, the order of this list will always be the same following this patch. Note that the order is unrelated to the order on the control crossbar, which it never was.
* multi_usrp: Fix GPIO bank selectionMartin Braun2019-11-261-11/+68
| | | | | | | | | | When calling a multi_usrp object like this: usrp->set_gpio_attr("TXB", "CTRL", 0xFFFF); Previously, it would only be able to address daughterboard A. Now, there is a full backward-compatible solution (compatible with 3.15), that will address either daughterboard's GPIOs.
* liberio: porting to I/O servicesBrent Stapleton2019-11-261-11/+2
| | | | Fixes 85f0551 ("rfnoc: Add I/O service manager to X300 and MPMD")
* rfnoc: Make polling I/O service not block on flow controlCiro Nishiguchi2019-11-269-68/+162
| | | | | | Add a new method to io_service::send_io to check whether the destination is ready for data, to make it possible to poll send_io rather than block waiting for flow control credits.
* rfnoc: Make I/O services relinquish CPU while waitingCiro Nishiguchi2019-11-264-25/+140
|
* multi_usrp: Add set_rx_spp() callMartin Braun2019-11-262-2/+34
| | | | | | | | This API call is a more explicit way of setting the spp than passing in an spp value in the args of the stream args when creating streamers. For pre-RFNoC devices, this is done by injecting the spp arg back into the stream args. For RFNoC devices, the set_property() call on the radio is called.
* rfnoc: xports: Count FC bytes as multiples of CHDR widthMartin Braun2019-11-266-17/+43
| | | | | | | | SEPs on the FPGA can only occupy multiples of the CHDR width in their FIFOs, unlike SW, where buffers are stored in RAM and can be aligned anyhow. Therefore, we align the counting of bytes for FC purpose and count multiples of CHDR width instead of the true number of bytes per packet.
* rfnoc: Restrict to inline I/O service based on link restrictionsCiro Nishiguchi2019-11-263-1/+69
| | | | | For links that do not support releasing buffers out of order, restrict the I/O service manager to always select the inline I/O service.
* rfnoc: Use graph_utils in multi_usrp chan generationBrent Stapleton2019-11-261-113/+142
|
* rfnoc: Merge I/O service device args with stream argsCiro Nishiguchi2019-11-266-45/+131
| | | | | This makes it possible for users to put I/O service-related args in either the device args or stream args.
* rfnoc: Split up offload I/O service into multiple filesCiro Nishiguchi2019-11-264-297/+351
|
* uhd: Remove all usages of boost::tuple and friendsMartin Braun2019-11-268-41/+40
| | | | | | | | | | | | | This replaces all of the following with standard C++ features: - boost::tuple - boost::make_tuple - boost::tuple::get - #include <boost/tuple/tuple.hpp> All usages were replaced with search-and-replace scripts (the usages of get could be automatically replaced with a vim macro, the rest was straightforward search-and-replace).
* uhd: Replace all occurrences of boost::bind with std::bindMartin Braun2019-11-2652-530/+530
| | | | | | | | | | | | | | | | | | | | | Note: Replacing everything with a lambda would be even better, but that can't be easily scripted so we'll do this as a first step to reduce the Boost footprint. This also removes occurences of #include <boost/bind.hpp>, and makes sure all usages of std::bind have an #include <functional>. clang-format wasn't always applied to minimize the changeset in this commit, however, it was applied to the blocks of #includes. Due to conflicts with other Boost libraries, the placeholders _1, _2, etc. could not be directly used, but had to be explicitly called out (as std::placeholders::_1, etc.). This makes the use of std::bind even uglier, which serves as another reminder that using std::bind (and even more so, boost::bind) should be avoided. nirio/rpc/rpc_client.cpp still contains a reference to boost::bind. It was not possible to remove it by simply doing a search and replace, so it will be removed in a separate commit.
* rfnoc: Adding rfnoc_graph utilitiesBrent Stapleton2019-11-262-0/+134
| | | | | | | Adding graph_utils to keep rfnoc_graph utilities to contain helper function and commonly used algorithms for the rfnoc_graph. These functions aren't core to the rfnoc_graph's functionality, so we'll keep them out of its API.
* mpmd: Add skip_oldlog device argMartin Braun2019-11-261-0/+8
| | | | | | | | | | This device arg will avoid printing old log messages by dumping all log messages in the buffer immediately after claiming a device. This can be helpful because reading old log messages can be confusing during debug situations, or simply during situations where we expect errors to contain warnings or errors, but we want to hide them. skip_oldlog defaults to off, i.e., the previous behaviour.
* mpmd: Skip generation of mb_controller and mb_iface for skip_init=1Martin Braun2019-11-261-3/+6
| | | | | | | | | The mpmd_mb_controller requires access to the fully functional FPGA, which in turn assumes flawless initialization. skip_init doesn't init at all, so inializing an mb_controller may not be possible, and shouldn't be attempted. This skips init of both mb_iface and mb_controller for skip_init, leaving it to the other code to figure out if they are initialized.
* uhd: Introduce I/O service managerAaron Rossetto2019-11-2626-176/+1264
| | | | | | | | | | | | | | | | | | | | | | | | | | - Implement I/O service detach link methods - The I/O service manager instantiates new I/O services or connects links to existing I/O services based on options provided by the user in stream_args. - Add a streamer ID parameter to methods to create transports so that the I/O service manager can group transports appropriately when using offload threads. - Change X300 and MPMD to use I/O service manager to connect links to I/O services. - There is now a single I/O service manager per rfnoc_graph (and it is also stored in the graph) - The I/O service manager now also knows the device args for the rfnoc_graph it was created with, and can make decisions based upon those (e.g, use a specific I/O service for DPDK, share cores between streamers, etc.) - The I/O Service Manager does not get any decision logic with this commit, though - The MB ifaces for mpmd and x300 now access this global I/O service manager - Add configuration of link parameters with overrides Co-Authored-By: Martin Braun <martin.braun@ettus.com> Co-Authored-By: Aaron Rossetto <aaron.rossetto@ni.com>
generated by cgit v1.2.3 (git 2.39.1) at 2025-01-05 23:14:55 +0000