aboutsummaryrefslogtreecommitdiffstats
path: root/host/lib/include/uhdlib/transport
Commit message (Collapse)AuthorAgeFilesLines
* rfnoc: Clarify usage of MTU vs. max payload size, remove DEFAULT_SPPMartin Braun2021-12-022-10/+29
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | These two values where being mixed up in the code. To summarize: - The MTU is the max CHDR packet size, including header & timestamp. - The max payload is the total number of bytes regular payload plus metadata that can be fit into into a CHDR packet. It is strictly smaller than the MTU. For example, for 64-bit CHDR widths, if a timestamp is desired, the max payload is 16 bytes smaller than the MTU. The other issue was that we were using a magic constant (DEFAULT_SPP) which was causing conflicts with MTUs and max payloads. This constant was harmful in multiple ways: - The explanatory comment was incorrect (it stated it would cap packets to 1500 bytes, which it didn't) - It imposed random, hardcoded values that interfered with an 'spp discovery', i.e., the ability to derive a good spp value from MTUs - The current value capped packet sizes to 8000 bytes CHDR packets, even when we wanted to use bigger ones This patch changes the following: - noc_block_base now has improved docs for MTU, and additional APIs (get_max_payload_size(), get_chdr_hdr_len()) which return the current payload size given MTU and CHDR width, and the CHDR header length. - The internally used graph nodes for TX and RX streamers also get equipped with the same new two API calls. - The radio, siggen, and replay block all where doing different calculations for their spp/ipp values. Now, they all use the max payload value to calculate spp/ipp. Unit tests where adapted accordingly. Usage of DEFAULT_SPP was removed. - The replay block used a hardcoded 16 bytes for header lengths, which was replaced by get_chdr_hdr_len() - The TX and RX streamers where discarding the MTU value and using the max payload size as the MTU, which then propagated throughout the graph. Now, both values are stored and can be used where appropriate.
* uhd: update num_recv_frames calculation for ctrl linksAndrew Lynch2021-11-021-1/+4
|
* uhd: streamer: Restore original recv(0) semanticsAaron Rossetto2021-08-201-0/+32
| | | | | | | | | A recv() of zero samples on an RX stream should return immediately (i.e., without respect to the timeout and regardless of the availability of samples), surfacing any stream error conditions via metadata. This convention was broken in a2f10ee9, causing a recv() of zero samples to wait for the entire timeout period and then return ERROR_CODE_TIMEOUT if no samples are available. This commit restores the desired semantics.
* rfnoc: Add option to disable flow control on rx streamingmattprost2021-04-291-1/+2
| | | | | | | | | | | Disabling this feature will allow the USRP to send a continuous stream of Rx data to a host machine without throttling due to lack of flow control credits. This is unnecessary overhead on lossless transports such as pcie or aurora. Usage: add 'enable_fc=false' to stream_args.args Signed-off-by: mattprost <matt.prost@ni.com>
* lib: Fix unresolved cleanup conflict (sorry!)Aaron Rossetto2021-03-041-5/+1
|
* lib: Add some virtual dtorsMartin Braun2021-03-043-1/+14
| | | | | | | Classes where we call delete (implicitly or explicitly) with a virtual inheritance structure need to declare dtors as virtual. This reduces compiler warnings with clang. There are no known bugs (yet) due to this.
* rfnoc: tx_streamer: Remove EOV size attributeMartin Braun2021-03-041-2/+0
| | | | It is unused, and causes clang warnings.
* host: Update code base using clang-tidyMartin Braun2021-03-047-34/+34
| | | | | | | | | The checks from the new clang-tidy file are applied to the source tree using: $ find . -name "*.cpp" | sort -u | xargs \ --max-procs 8 --max-args 1 clang-tidy --format-style=file \ --fix -p /path/to/compile_commands.json
* uhd: Check for overflow after timeout buff readSteven Koo2021-01-212-18/+15
| | | | | | | | | | | | Error processing has been moved to another thread, so it's possible that consecutive recv calls may miss the signalling that an overflow occurred. We have no guarantee that the flag had been set by the time the second recv call to find the errors occurs. This adds another check for an overflow after calling _get_aligned_buffs with a min 1ms timeout. Hopefully this is long enough for the error to propogate, but it's not guaranteed. Signed-off-by: Steven Koo <steven.koo@ni.com>
* fixup! RFNoC: Handle receive of 0 samplesmichael-west2020-12-211-3/+3
|
* RFNoC: Handle receive of 0 samplesmichael-west2020-12-101-0/+5
| | | | | | | Returns immediately if requested number of samples is zero. Prevents timeout error from being thrown if user requests no samples. Signed-off-by: michael-west <michael.west@ettus.com>
* dpdk: Improve link status detectionAaron Rossetto2020-09-031-0/+1
| | | | | | | | | | | | | | | | This change improves the DPDK link status detection algorithm in the following ways: - The status of the links are checked at an interval of 250 ms. If all links report as being up, the driver proceeds. - If any of the DPDK links has not reported as being up by the end of the link status detection timeout (1000 ms by default), the algorithm throws a runtime error rather than proceeds with one or more down links. - Users may override the default link status detection timeout by passing dpdk_link_timeout=N, where N is the desired timeout in milliseconds, either via device arguments or in the UHD configuration file.
* X300: fix for incorrect PCIe buffer size valuesMatthew Crymble2020-08-041-3/+5
| | | | added extra args to get PCIe buffer sizes from factory method
* RFNoC: Add xport disconnect callbacksmichael-west2020-08-041-0/+7
| | | | | | | | | | | | Transports were not disconnecting their links from the I/O service upon destruction, leaving behind inaccessible send and recv links used by nothing. This led to I/O errors after creating several transports. Added callbacks to transports to automatically disconnect their links from the I/O service when the transport is destroyed. Updated all callers to supply a disconnect callback. Signed-off-by: michael-west <michael.west@ettus.com>
* uhd: remove liberioRobertWalstab2020-07-201-178/+0
|
* uhd: Replace include guards with pragma onceLane Kolbly2020-04-0827-107/+28
| | | | | Pragma once is the more modern version of include guards, eliminating any potential problems with mistyping include guards. Let's use those.
* uhd: Apply clang-format against all .cpp and .hpp files in host/Martin Braun2020-03-037-37/+40
| | | | | Note: template_lvbitx.{cpp,hpp} need to be excluded from the list of files that clang-format gets applied against.
* lib: move OSX specific code to where the values are initially setMichael Dickens2020-02-211-0/+11
|
* uhd: fixing MSVC warningsBrent Stapleton2020-01-091-2/+4
| | | | | | | | | | | Small changes to remove various compiler warnings found in MSVC - Adding uhd::narrow_cast to verious spots - wavetable.hpp: all floats literals in the wavetable. - paths_test: unnecessary character escape - replay example: remove unreferenced noc_id - adfXXXX: Fixing qualifiers to match between parent and derived classes - rpc, block_id: Removing unused name in try...catch
* dpdk: Change client to wait on member variableCiro Nishiguchi2019-12-201-8/+12
| | | | | | Waiting on queue status seems to not always work, the queue state seems to not be updated immediately after pushing an item onto it when queried from a different thread.
* transport,usrp: Make available packet-based flow controlAlex Williams2019-12-201-2/+2
| | | | | | | | DPDK provides a fixed number of fixed-size buffers for the receive window, so it needs packet-based flow control to avoid dropping packets. This change enables counting by packets. Co-authored-by: Ciro Nishiguchi <ciro.nishiguchi@ni.com>
* transport: Fix comment on recv_io_if::fc_callback_tAlex Williams2019-12-201-1/+1
| | | | It referenced the wrong function for releasing recv buffers.
* lib: Remove dpdk_zero_copy filesMartin Braun2019-12-203-406/+0
| | | | These were left here as a reference.
* dpdk: Add new DPDK stack to integrate with I/O servicesAlex Williams2019-12-209-90/+1121
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | 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>
* 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>
* 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
|
* transport: Add new base for DPDK links, based on 18.11Alex Williams2019-11-261-0/+319
| | | | | | | | | | | | 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.
* rfnoc: Make polling I/O service not block on flow controlCiro Nishiguchi2019-11-264-12/+43
| | | | | | 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 Nishiguchi2019-11-262-16/+41
|
* rfnoc: Restrict to inline I/O service based on link restrictionsCiro Nishiguchi2019-11-262-0/+36
| | | | | 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: Split up offload I/O service into multiple filesCiro Nishiguchi2019-11-262-0/+268
|
* uhd: Introduce I/O service managerAaron Rossetto2019-11-265-17/+94
| | | | | | | | | | | | | | | | | | | | | | | | | | - 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>
* transport: Implement eov indications for Rx and Tx streamsAaron Rossetto2019-11-264-45/+270
|
* uhd: Replace usage of boost smart pointers with C++11 counterpartsMartin Braun2019-11-261-2/+2
| | | | | | | | | | | | | | | | | | | This removes the following Boost constructs: - boost::shared_ptr, boost::weak_ptr - boost::enable_shared_from_this - boost::static_pointer_cast, boost::dynamic_pointer_cast The appropriate includes were also removed. All C++11 versions of these require #include <memory>. Note that the stdlib and Boost versions have the exact same syntax, they only differ in the namespace (boost vs. std). The modifications were all done using sed, with the exception of boost::scoped_ptr, which was replaced by std::unique_ptr. References to boost::smart_ptr were also removed. boost::intrusive_ptr is not removed in this commit, since it does not have a 1:1 mapping to a C++11 construct.
* transport: Implement an I/O service that uses an offload threadCiro Nishiguchi2019-11-261-0/+66
| | | | | | | The offload_io_service executes another I/O service instance within an offload thread, and provides synchronization mechanisms to communicate with clients. Frame buffers are passed from the offload thread to the client and back via single-producer, single-consumer queues.
* transport: Port liberio to link_ifAlex Williams2019-11-261-0/+180
|
* rfnoc: Fix transport buffer reservationsCiro Nishiguchi2019-11-261-9/+4
| | | | | | | | | Change transports to reserve the number of frame buffers they actually need from the I/O service. Previously some I/O service clients reserved 0 buffers since they shared frame buffers with other clients, as we know the two clients do not use the links simultaneously. This is possible with the inline_io_service but not with a multithreaded I/O service which queues buffer for clients before they are requested.
* streamer: Add option to ignore sequence errorsCiro Nishiguchi2019-11-263-8/+9
| | | | Add template parameter to ignore sequence errors, used for testing.
* rfnoc: radio/streamer: Handle late commands and burst ACKsMartin Braun2019-11-262-0/+22
| | | | | | | - Burst ACKs are already handled by the TX streamer, but the radio now also sends an action upstream on reception of a burst ACK - Late commands were only acquitted by an 'L', now an action gets sent downstream and is handled in the rx streamer
* rfnoc: Add MTU property to RX streamerCiro Nishiguchi2019-11-262-12/+34
|
* rfnoc: Implement flushing on overrunCiro Nishiguchi2019-11-262-25/+125
| | | | | | | | | | | | | | | | | | | | | | | | | | | | This modifies the overrun handling such that the RX streamer does not restart the radios until the packets that were buffered prior to the overrun are read by the user. When an RX streamer receives an overrun, it will run the following algorithm: 1. Stop all upstream producers. 2. Set an internal flag in the streamer that indicates that the producers have stopped due to an overrun. 3. Continue servicing calls to recv until it runs out of packets in the host buffer (packets that can be read from the transport using a 0 timeout). 4. Once the packets are exhausted, return an overrun error from recv. The radio, if it was in continuous streaming mode before the overrun, includes a flag in its initial action whether or not to restart streaming. 5. If the radio requested a restart, the streamer submits a restart request action upstream. This action will be received by the radio. The radio will then check the current time, and send a stream command action back downstream. 6. The RX streamer receives the stream command action, and uses it to send another stream command to all upstream producers. This way, all upstream producers receive a start command for the same time.
* rfnoc: Add nirio_link link objectMartin Braun2019-11-261-0/+210
|
* transport: Add modeling of physical adaptersAlex Williams2019-11-264-0/+116
| | | | | | | Now link instances must have the ability to report the corresponding physical adapter that is used for the local side of the link. This information can be used to help identify when multiple links share the same adapter.
* rfnoc: Add ability to select transport for streamers to user APIsAlex Williams2019-11-261-0/+1
| | | | Now the user can choose which transport is used in connect() calls.
* rfnoc: tx_streamer: add support for async messagesCiro Nishiguchi2019-11-262-7/+10
| | | | | | Add an async message queue that aggregates errors from multiple sources. Errors can come from the strs packets originating from the stream endpoint or from the radio block through control packets to the host.
* rfnoc: Add MTU property to TX streamerCiro Nishiguchi2019-11-261-5/+24
|
* rfnoc: add rx and tx transports, and amend rfnoc_graphCiro Nishiguchi2019-11-265-0/+1177
| | | | | | | | | | | | | | | | | | | | | | | transports: Transports build on I/O service and implements flow control and sequence number checking. The rx streamer subclass extends the streamer implementation to connect it to the rfnoc graph. It receives configuration values from property propagation and configures the streamer accordingly. It also implements the issue_stream_cmd rx_streamer API method. Add implementation of rx streamer creation and method to connect it to an rfnoc block. rfnoc_graph: Cache more connection info, clarify contract Summary of changes: - rfnoc_graph stores more information about static connections at the beginning. Some search algorithms are replaced by simpler lookups. - The contract for connect() was clarified. It is required to call connect, even for static connections.
* uhd: add udp boost asio implementation of transport interfaceCiro Nishiguchi2019-11-262-0/+315
|
* uhdlib: transport: Add links.hppMartin Braun2019-11-261-0/+43
| | | | This contains both_links_t
generated by cgit v1.2.3 (git 2.39.1) at 2025-01-08 10:37:34 +0000