aboutsummaryrefslogtreecommitdiffstats
path: root/host/lib
Commit message (Collapse)AuthorAgeFilesLines
* rfnoc: Fix back-edge consistency checkMartin Braun2021-12-091-0/+7
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | On back-edges, no properties are forwarded, but properties must be consistent after property resolution. This breaks when the source edge on a back-edge has an edge property which the destination block does not. Consider the following graph: DDC -> Replay -> DDC where both instances of 'DDC' refer to the same block. Now, assume the first edge is declared a back edge (in principle, it shouldn't matter). The DDC block has an edge property `samp_rate` which the Replay block does not. Therefore, it can't forward this edge property to the Replay block's input edge property list. In the consistency check code, we don't check for the existence of edge nodes, because it is assumed edge properties where either forwarded, or aligned through some other manner. This leads to a property lookup failure. With this fix, we skip the consistency check for edge properties which don't exist on the destination node. This is safe because the destination block can not have a property resolver defined for undefined properties. This means the destination block can either: - Drop the property. In this case, there is no value in checking consistency. Even if we could forward edge properties on back-edges, they would always have the same value. - Forward the property. In that case, the consistency check would happen elsewhere in the graph where there's no back-edge.
* host: Fix typos and small thingsLane Kolbly2021-12-092-1/+2
|
* e3xx: Remove unused constantMartin Braun2021-12-091-2/+0
|
* rfnoc: Add ops pending to management opWade Fife2021-12-081-1/+2
| | | | | | The ops pending for each operation was stored implicitly in the data structure. This adds it explicitly, which is useful for debugging and packet dissection.
* dpdk: disable warnings for using an experimental featureAndrew Lynch2021-12-061-0/+6
|
* dpdk: Upgrade to DPDK 19.11 APIAndrew Lynch2021-12-0610-126/+126
| | | | Support DPDK versions 19.11 and 20.11
* rfnoc: Enable drop counter on chdr_ctrl_endpointMartin Braun2021-12-032-4/+23
| | | | | | | | | | | | This class has a member _num_drops, which can be read out using the get_num_drops() API call. However, when dropping packets, this counter was not incremented, which is fixed now. This also includes a very minor optimization from 2 map<> lookups to 1 lookup (they are in O(log N)). Since there are usually a small two-digit number of endpoints connected to the async message receiver, this change is not expected to yield major improvements, but the lookup *is* in a hot loop.
* max287x: Fix key in table of freq rangesMartin Anderseck2021-12-031-1/+1
| | | | | | The keys for the table of frequency ranges for each VCO value counts up consecutively but key "1" was there twice while "2" was missing. This is fixed here.
* uhd: Remove superfluous boost/bind.hpp includesMartin Braun2021-12-031-1/+0
| | | | | On newer versions of Boost, they show deprecation notes. However, they're not actually used any more so they can go.
* rfnoc: Change default block behaviourMartin Braun2021-12-021-5/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | | | The default block controller is used whenever no other block controller is used. It currently defaults to dropping both property propagation and actions. When a custom block is injected into a graph like this for example: Radio -> DDC -> Custom Block -> Rx Streamer This default behaviour causes the Rx Streamer to not be able to send actions (like stream commands) nor does it allow MTU propagation (or any other property's propagation). The default block behaviour is ONE_TO_ONE, meaning that actions and properties on input channel N will get forwarded to output channel N. In absence of an actual block controller, this is more useful default than setting the propagation to DROP for both actions and properties. Most blocks that pass through data, or do some simple processing, will now work in the absence of a block controller. The new disadvantage is that blocks which would modify properties such as sampling rate, scaling, or MTU will no longer work properly in the absence of a block controller. However, the recommended behaviour is anyway not to operate without a block controller. For the cases where no block controller is present, ONE_TO_ONE is considered the generally more useful default.
* mpmd: Add MTU plausibility checkMartin Braun2021-12-021-0/+42
| | | | | | | | | | | | | | | | | | We have noticed that on 1 GbE connections, MTU discovery can become unreliable. Since we now use the MTU directly for deriving spp and other values, a correct MTU is important. Because we don't have a way of knowing if MTU discovery worked or not, we add some heuristics in form of a plausibility check. For now, the only rule in this check is if that the detected MTU is a bit larger than 1472 bytes, we coerce down to 1472, because this is such a standard value (most 1 GbE interfaces default to an IPv4 MTU of 1500 bytes). For the cases where the interface MTU is set to be between 1500 and 1528 bytes, this would cause a very minor performance loss. We accept this performance loss as it is small, and those cases are very rare. MTUs are usually 1500 bytes, or >= 8000 bytes for high-speed links using jumbo frames.
* x300: Remove usage of CHDR_MAX_LEN_HDRMartin Braun2021-12-021-2/+6
| | | | | | | This constant was generally harmful, since it was only correct under certain circumstances (64 bit CHDR with timestamps). The X3x0 code was the last place it was being used, and we remove it without substitute because it was not doing anything useful here.
* rfnoc: Clarify usage of MTU vs. max payload size, remove DEFAULT_SPPMartin Braun2021-12-0213-74/+206
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | 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.
* rfnoc: replay block: Disable prop and action propagationMartin Braun2021-12-011-1/+10
| | | | | | | | | | | | | | The replay block is more like the radio block than like a FIFO. In particular, consider this flow graph: Replay -> DDC -> Replay Imagine you're using the replay block to test the DDC block with prerecorded data. If we treated the Replay Block like a FIFO, then we'd have a loop in the graph (which is already wrong). If we used the DDC to resample, then the input- and output sample rate of the Replay mismatch, which is a legal way to use the Replay block, but not possible if we treat the graph like a loop.
* rfnoc: Fix issue in uhd::rfnoc::connect_through_blocks()Martin Braun2021-12-011-6/+9
| | | | | | | | | | | | | | | | | | | | | | | When connect_through_blocks() was called on blocks within a single chain, there was a bug where the chain was incorrectly cropped. In a standard FPGA image, say one was to use this API call to connect the radio to the DDC. It would generate a chain of blocks hanging off the radio as such: Radio -> DDC -> SEP What the code should do, and what this fix provides, is that the chain gets cropped after the DDC, to look like this: Radio -> DDC With the current bug, it would assume the chain has a dangling edge, and incorrectly throw an exception. Note that this bug would not appear when source and destination block are on separate chains (i.e., both have an SEP in their chain). This patch includes minor logging and comment improvements around the offending lines of code.
* rfnoc: radio: Fix async message handling channel checksMartin Braun2021-11-301-8/+8
| | | | | | | | | | | | | | | | | | | | | The async message handler and the async message validator would erroneously compare channel numbers for RX async messages with the number of valid TX channels. On TwinRX, where there are zero TX channels, this would always fail. Elsewhere in the code, the comparisons for TX and RX channels mixed up input and output ports. The second issue is that the comparison made was a "greater than" rather than "greater or equal". The effect of these two bugs was that potentially, we could have accepted async messages for an invalid port N, where N is the number of valid ports of this block, and that for TwinRX/X300 users, async messages on channel 1 would not get accepted (they would, however, get accepted for channel 0 because of the second issue). This includes overrun handling, which was broken for channel 1 and 3 on an X300. Another effect of the bug was that EPIDs for async messages weren't always programmed correctly.
* multi_usrp_rfnoc: Reduce latency of get_time_now()michael-west2021-11-172-10/+10
| | | | | | | Getting the time from the mb_controller is slow, so try to get the time from the Radio on the fast path first. Signed-off-by: michael-west <michael.west@ettus.com>
* host: Add ability to get time from Radio blockmichael-west2021-11-172-1/+29
| | | | | | Add API calls to Radio control to get ticks and time. Signed-off-by: michael-west <michael.west@ettus.com>
* rfnoc: mgmt_portal: Fix order of validity checksMartin Braun2021-11-151-2/+6
| | | | | | | | | | | The order must: - Check transaction has the right number of hops, then read hop - Check hop has the right number of operations (at least 2), then read those ops - Check the ops have the correct opcodes The code was doing checks in the wrong order. Thanks to Github user johnwstanford for pointing this out.
* rfnoc: Add CHDR width to make argsMartin Braun2021-11-123-0/+4
| | | | | | | This provides every block controller with a copy of its CHDR width. Note: mock blocks always get configured with a 64-bit CHDR width, to retain API compatibility.
* B200: Re-sync timesmichael-west2021-11-102-10/+31
| | | | | | | | | | The times on the device can glitch if either the tick rate changes or the number of active chains changes. This throws off the time if the user gets streamers, changes the sample rate, or changes the tick rate after synchronizing the time. This change re-synchronizes the times automatically in those cases. Signed-off-by: michael-west <michael.west@ettus.com>
* python: multi_usrp: Add set_rx_spp()Wade Fife2021-11-091-0/+1
|
* mpmd: Increase UHD-side MTU cap for 10 GbE and 1 GbEMartin Braun2021-11-081-4/+14
| | | | | | | | | This gets closer to what our hardware can actually support. See the comments for further explanations. This has the side-effect of patching an issue on X410 (using 200 MHz images) where garbage samples would get injected (one per packet). It is not, however, the final fix for that problem.
* host: python: Add gpio_voltage python APILane Kolbly2021-11-051-1/+11
|
* host: Add gpio_voltage discoverable featureLane Kolbly2021-11-052-0/+62
|
* host: Add RPC calls for GPIO voltageLane Kolbly2021-11-051-0/+9
|
* host: python: Return mb_controller with reference_internalLane Kolbly2021-11-041-1/+1
|
* fixup! host: x4xx: Implement GPIO APILane Kolbly2021-11-031-2/+14
|
* host: x4xx: Implement GPIO APILane Kolbly2021-11-035-0/+255
| | | | | | | | | | | | This implements the GPIO API for X410 through get_gpio_attr and set_gpio_attr. In ATR mode, which channel's ATR state is chosen by the set_gpio_src call, setting e.g. DB0_RF0 for channel 0 or DB0_RF1 for channel 1. In manual mode, all 24 bits (for both ports) are set in a single register write. Although the front panel of the device has two ports, labelled GPIO0 and GPIO1, this API exposes them as though they were a single 24-bit GPIO port.
* host: Add GPIO functions to MPM RPC shimLane Kolbly2021-11-031-0/+5
|
* rfnoc: Remove cruft from UHD 3 (constants)Martin Braun2021-11-022-1/+6
| | | | | | | | | | | | | This removes some constants from UHD that were left over from RFNoC/UHD 3.x. They are unused. rfnoc_rx_to_file had a commented-out section that was also UHD-3 only. Note that rfnoc/constants.hpp is pretty bare now, and could be removed. However, it is in the public header section, so we shall leave the used constants where they are. This requires fixing includes in mgmt_portal.cpp.
* uhd: update num_recv_frames calculation for ctrl linksAndrew Lynch2021-11-021-1/+4
|
* host: gpio: Create gpio_atr_offsets to store GPIO registersLane Kolbly2021-10-277-94/+136
| | | | | | | Refactors register addresses into a gpio_atr_offsets structure which contains the various register addresses. This allows creating other devices with different GPIO register layouts with greater ease, and eliminates the use of macros (yay!)
* siggen: Fix direction of rotationWade Fife2021-10-271-9/+9
| | | | | | | | | The I and Q were swapped in sine_tone, which caused confusion and made the rotation of REG_CARTESIAN clockwise by default. This effectively made the resulting frequency negative. This PR makes the I and Q order consistent with RFNoC and fixes the direction of rotation so that a positive value for REG_PHASE_INC (phase increment) results in a counter-clockwise rotation, which yields a positive frequency.
* uhd: Fix spelling errorsA. Maitland Bottoms2021-10-222-2/+2
| | | | Thanks to Mait for pointing these out!
* uhd: math: Replace wrap-frequency math with a single functionMartin Braun2021-10-193-21/+15
| | | | | | | | | In multiple places in the UHD code, we were doing the same calculation for a wrapped frequency (wrap it into the first Nyquist zone). This math was using boost::math, too. Instead of editing every instance, we create a new function, uhd::math::wrap_frequency(), and replace all of its separate implementations with this function. The new function also no longer relies on boost::math::sign.
* lib: Remove all remaining usage of boost::numeric::bounds<>Martin Braun2021-10-191-2/+3
| | | | Replaced by std::numeric_limits<>.
* uhd: Replace Boost mutexes and locks with standard optionsMartin Braun2021-10-1933-231/+230
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | This is a very mechanical task that could almost have been done with sed. Boost versions of mutexes and locks were removed, and replaced with std:: versions. The replacement tables are as follows: == Mutexes == - boost::mutex -> std::mutex - boost::recursive_mutex -> std::recursive_mutex Mutexes behave identically between Boost and std:: and have the same API. == Locks == C++11 has only two types of lock that we use/need in UHD: - std::lock_guard: Identical to boost::lock_guard - std::unique_lock: Identical to boost::unique_lock Boost also has boost::mutex::scoped_lock, which is a typedef for boost::unique_lock<>. However, we often have used scoped_lock where we meant to use lock_guard<>. The name is a bit misleading, "scoped lock" sounding a bit like an RAII mechanism. Therefore, some previous boost::mutex::scoped_lock are now std::lock_guard<>. std::unique_lock is required when doing more than RAII locking (i.e., unlocking, relocking, usage with condition variables, etc.). == Condition Variables == Condition variables were out of the scope of this lock/mutex change, but in UHD, we inconsistently use boost::condition vs. boost::condition_variable. The former is a templated version of the latter, and thus works fine with std::mutex'es. Therefore, some boost::condition_variable where changed to boost::condition. All locks and mutexes use `#include <mutex>`. The corresponding Boost includes were removed. In some cases, this exposed issues with implicit Boost includes elsewhere. The missing explicit includes were added.
* rfnoc: mgmt_portal: Remove two unused variablesMartin Braun2021-10-111-4/+6
| | | | Thanks for github user johnwstanford for pointing those out.
* libusb: Remove unused context variableMartin Braun2021-09-301-2/+5
| | | | | | | | | | | The USB managed buffer implementation created a context every time one was generated. The additional load is not very high, because the global session is a singleton, and simply returns the same context again with only a few branches. Also, managed buffers persist for the entire session. However, the context is never used in the managed buffer. This code is thus confusing for the reader of this code, and we remove the extraneous, unused context variable.
* ad9361: Add comment re overclockingMartin Braun2021-09-301-3/+23
| | | | | | | | | | As GitHub user marcosino points out, we're running the AD9361 in overclocked mode. This is because the driver was written with no longer valid recommendations. We add a comment and some debug messages to clarify this. Should there be RF impairments (signal integrity or other) because of overclocking, users would be able to check DEBUG log statements to correlate with overclocked configurations.
* chdr: Rename var max_size_bytes to avoid confusionMartin Anderseck2021-09-281-9/+9
| | | | | | | | | The variable max_size_bytes has a different name in the source than in the header and is not self-explanatory in both. Therefore when comparing against it in the assertion in line 142 one could assume that a number of bytes needs to be compared with a byte value. Change variable to `buff_size` in source and header file to avoid confusion and add documentation.
* ad9361: Modify set-tx-gain procedure to update gain in one goMartin Braun2021-09-171-7/+13
| | | | | | | | | | | | | | | | | | | | | | | | | The previous behaviour of UHD for setting gain was: 1. Set "Mask Clr Atten Update". This will avoid "Immediately Update TPC Atten" to be cleared. 2. Then, assert "Immediately Update TPC Atten". 3. Poke the LSBs of the attenuation value. 4. Poke the MSB of the attenuation value. This order of operations has the downside of causing large Tx power spikes when setting the attenuation, because you need both registers to properly set the attenuation, but we are updating the gain immediately, even between the two attenuation register's update. Moreover, the upstream Linux driver for AD9361 by ADI also does not do this. We therefore change the procedure to match the kernel driver behaviour, which is: 0. [During initialization: Clear "Mask Clr Atten Update" 1. Poke the attenuation registers 2. Then, assert "Immediately Update TPC Atten". This avoids Tx power spikes. It also reduces the Tx-gain procedure to 3 pokes instead of 4.
* uhd: zbx: Prevent TX antenna config from disrupting RXLane Kolbly2021-09-161-2/+8
| | | | | | | | | | | | | | So, both the set_tx_antenna_switches and set_rx_antenna_switches functions configure the TX0_ANT_11 register (which controls the final switch before the TX/RX port, switching it between the three TX paths and the RX path). The RX antenna configuration code will, if the RX antenna is set to TX/RX, configure that switch to the TX/RX->RX path when the ATR is set to RX. However, the TX antenna config code will always configure that switch to the "bypass" path, for both the 0X and RX ATR modes, regardless of whether the RX side actually needs that path. Ergo, this change makes set_tx_antenna_switches only configure that switch when it is configuring the XX or TX modes.
* cmake: Fix issues with static builds and CMRCMartin Braun2021-09-161-2/+4
| | | | | | | | | | | | | | Adds uhd_rc as a link target to static builds of libuhd. This fixes build errors like this: ``` uhd/lib/cal/database.cpp:12:10: fatal error: cmrc/cmrc.hpp: No such file or directory #include <cmrc/cmrc.hpp> ``` This also adds uhd_rc as to $libuhd_libs instead of just listing it separately, and target objects from uhd_rc to $libuhd_sources.
* X300: Fix error message for wrong reference frequencyMartin Anderseck2021-09-161-1/+2
| | | | | Error message was not adapted when support for 11.52 MHz and 23.04 MHz references was added. Fixing this.
* x300: Remove unused variables in x300_eth_mgr.cppMartin Braun2021-09-161-8/+0
|
* cmake: Replace CMAKE_{SOURCE,BINARY}_DIR with UHD_*_DIRMartin Braun2021-09-103-4/+4
| | | | | | | | | | | | | | | | | | | | | | | | | | See the CMake 3.8 documentation on these two variables: https://cmake.org/cmake/help/v3.8/variable/PROJECT-NAME_SOURCE_DIR.html https://cmake.org/cmake/help/v3.8/variable/CMAKE_SOURCE_DIR.html Under normal circumstances, these two are identical. For sub-projects (i.e., when building UHD as part of something else that is also a CMake project), only the former is useful. There is no discernible downside of using UHD_SOURCE_DIR over CMAKE_SOURCE_DIR. This was changed using sed: $ sed -i "s/CMAKE_SOURCE_DIR/UHD_SOURCE_DIR/g" \ `ag -l CMAKE_SOURCE_DIR **/{CMakeLists.txt,*.cmake}` $ sed -i "s/CMAKE_BINARY_DIR/UHD_BINARY_DIR/g" \ `ag -l CMAKE_BINARY_DIR **/{CMakeLists.txt,*.cmake}` At the same time, we also replace the CMake variable UHD_HOST_ROOT (used in MPM) with UHD_SOURCE_DIR. There's no reason to have two variables with the same meaning and different names, but more importantly, this means that UHD_SOURCE_DIR is defined even in those cases where MPM calls into CMake files from UHD without any additional patches. Shoutout to GitHub user marcobergamin for bringing this up.
* uhd: mpm: Expose filesystem version information on MPM treeLane Kolbly2021-09-081-0/+6
| | | | | This allows UHD clients to determine, for example, whether the currently loaded filesystem is up-to-date.
* python: multi_usrp: Fix overloaded function definitionMartin Anderseck2021-09-081-1/+1
| | | | | Fix function definition set_rx_iq_balance so that Python can reach the overloaded C++ function. There was a copy & paste error in there.
generated by cgit v1.2.3 (git 2.39.1) at 2024-12-30 23:10:29 +0000