Commit graph

1,486 commits

Author SHA1 Message Date
Robert Ghilduta
712d49e28f libbladeRF: add RX reordering and prime worker before RX enable
- Add RX buffer sequence tracking with reorder queue to deliver in-order buffers
  despite out-of-order USB completions. Pause submissions when the reorder
  queue holds data to avoid overwriting, and reset tracking on stream start.
- Introduce sync_prime_stream() to start the sync worker and wait until it is
  running. This is used to avoid start-up gaps.
- Call sync_prime_stream() on RX enable paths for bladerf1/bladerf2 boards
  before backend enable, reducing early stream discontinuities (e.g., with
  RX_MUX_32BIT_COUNTER).
- This does not at affect TX code paths.

This improves RX robustness and eliminates potential early out-of-order batches
on some platforms under high sample rates and low compute throughput.
2025-09-28 20:28:22 -07:00
Ryan Thompson
7d7d87f686 libbladeRF: improve RFIC BIST loopback error message
Change from log_debug to log_error and provide clearer message when
RFIC BIST loopback is attempted in FPGA tuning mode. The new message
explicitly states the requirement (HOST mode) and current state (FPGA
mode), making it actionable for users.
2025-07-16 17:52:11 -05:00
Ryan Thompson
55b5a0d07c libbladeRF: fix OC register from 0x54 to 0xD4 enabling both RX1 and RX2
This occurs because the OC register (0x003) was set to 0x54, which only
enables RX1 in oversample mode. Now both channels are enabled by default
and appropriately gated at the top level adc_assignment_proc
in hosted.vhd.
2025-07-03 21:38:50 -05:00
Ben Morse
918535854c bladeRF-fsk: host: various errors and compiler warning fixes on multiple platforms 2025-06-07 21:58:57 -07:00
Ryan Thompson
8f490c0a67 libbladeRF: fix max sample throughput warning for oversample feature 2025-05-20 01:47:21 -07:00
Robert Ghilduta
09c8208778 libbladeRF: bump version to 2.6.0
v2.6.0 (2025-05-06)
--------------------------------

This version of libbladeRF is intended for use with:

  FX3 Firmware v2.6.0
  FPGA         v0.16.0

New Features:
 * Added SC16_Q11 packed sample format support, improving sample efficiency
 * Increased FX3 GPIF buffer sizes, significantly boosting USB throughput (~356MB/s)
 * Implemented dynamic Nios base address assignment based on RAM size
 * Added parallel build support for hosted FPGA variants (40, 115, A4, A5, A9)
 * Enabled elf-gdb and gdb-server support for improved FPGA debugging capability
 * Added git revision information to FPGA build outputs
 * bladeRF 2.0: Added support for gain calibration loading, enabling, and management
 * bladeRF 2.0: Introduced bladeRF-power CLI tool for managing gain calibration
 * bladeRF 2.0: Added bladerf_get_gain_target function for gain management
 * bladeRF 2.0: Added bladerf_format_to_string API call
 * bladeRF 2.0: Added bladerf_image_print_metadata API to display image metadata
 * bladeRF 2.0: Improved compatibility checking between FPGA and FX3 firmware GPIF buffers

Improvements:
 * libbladeRF now compiles and runs under MinGW and MSVC on Windows platforms
 * Replaced libpthreadswin32 dependency with native libbladeRF threading macros
 * Improved error handling and reporting for out-of-range frequency and sample rate settings
 * Added detailed logging for USB transfer and Nios access timeouts
 * Updated Doxygen to v1.9.8, adding support for light/dark modes and modernized configuration
 * Improved GPIF buffer alignment handling for larger FX3 buffer sizes
 * Added CMake find_package() support for easier integration into external projects
 * Streamlined internal error checking macros, simplifying codebase
 * Enhanced sync_rx_meta example with additional argument parsing and MIMO support
 * Improved FPGA simulations for bladeRF 1.0 and bladeRF 2.0 micro

Bug Fixes:
 * Fixed delayed sample and off-by-one timestamp issues on bladeRF 1.0
 * Corrected TX meta sample gap issue when timestamp set to META_NOW
 * Resolved corrupted UART buffer in command_uart component
 * Fixed TX deadlock scenarios by refactoring RX/TX DMA prioritization
 * Addressed SC16_Q11 packed RX discontinuity issue on overflow condition
 * Fixed false timing paths and resolved multiple timing violations
 * Fixed calibration load check in bladerf_get_gain_calibration
 * Corrected RX meta packet timestamp and flags assignment
 * Fixed sync TX initialization bug

Miscellaneous:
 * Updated API documentation to reflect larger GPIF buffer sizes
 * Improved FX3 SDK setup instructions in README
 * Minor typo and documentation corrections throughout the codebase
2025-05-11 04:22:46 -07:00
Robert Ghilduta
902fc43305 firmware: bump version to 2.6.0
v2.6.0 (2025-05-06)
--------------------------------
 * Fixed missing break in BLADE_USB_CMD_QUERY_FPGA_SOURCE command handling
 * Improved FX3 SDK setup instructions in README
2025-05-11 04:21:32 -07:00
Robert Ghilduta
f746c5bd13 fpga: bump version to 0.16.0
--------------------------------
v0.16.0 (2025-05-06)
--------------------------------

This version brings significant FPGA improvements including sc16q11 packed mode
support and optimized USB throughput, along with important bug fixes for bladeRF
1.0 and 2.0 micro devices.

 Features:
 * hdl: Added sc16q11 packed mode for improved sample efficiency
 * hdl: Increased FX3 GPIF buffer sizes, boosting USB throughput (~356MB/s)
 * hdl: Implemented dynamic Nios base address assignment based on RAM size
 * hdl: Added parallel build support for hosted FPGA variants (40, 115, A4, A5, A9)
 * nios: Enabled elf-gdb and gdb-server support for improved debugging capability
 * hdl: Added git revision information to FPGA build outputs

 Fixes:
 * hdl: Fixed delayed sample and off-by-one timestamp bugs on bladeRF 1.0
 * hdl: Corrected TX meta sample gap issue when timestamp set to META NOW
 * hdl: Fixed corrupted UART buffer in command_uart component
 * hdl: Resolved TX deadlock scenarios by refactoring RX/TX DMA prioritization
 * hdl: Improved FPGA simulation accuracy for bladeRF 1 and 2 micro models
 * hdl: Fixed false timing paths and resolved multiple timing violations
2025-05-11 04:21:30 -07:00
Robert Ghilduta
20b9810eb0 misc: add missing CHANGELOGs
All CHANGELOGs should be up to date now.
2025-05-11 04:17:01 -07:00
Robert Ghilduta
619e058115 matlab: add missing bladerf_image_type_to_string
Add missing bladerf_image_type_to_string API_EXPORT declaration and also
add it to matlab
2025-04-24 04:36:36 -07:00
Robert Ghilduta
6905c37642 host: finish off cross-platform support for Windows
libbladeRF now compiles under MinGW as well as MSVC under Windows
2025-04-24 04:36:36 -07:00
Robert Ghilduta
d673366f4c host: final pthread migration to make linux compile 2025-04-24 04:36:36 -07:00
Robert Ghilduta
5863f983a2 host: remove remaining libpthreadswin32 and add fixups to cmake 2025-04-24 04:36:36 -07:00
Robert Ghilduta
1062b40d46 host: begin removing libpthreadswin32 from cmakes 2025-04-24 04:36:36 -07:00
Robert Ghilduta
5cdcd28e16 host: convert async/sync to libbladeRF thread macros 2025-04-24 04:36:36 -07:00
Robert Ghilduta
1e498900e8 libbladeRF: use libbladeRF thread macros 2025-04-24 04:36:36 -07:00
Ryan Thompson
af0045347f host: update CMake minimum required version format
Update cmake_minimum_required calls to use the 'min...max' version range syntax
 to fix CMake deprecation warnings. Also modernized CMP0115 policy to use the
 NEW behavior instead of OLD.
2025-04-24 04:36:36 -07:00
ifrasch17
b0aa8dd99c libbladeRF: Update API documentation for new GPIF buffer sizes
-fixed the buffer sizes listed in the descriptions of the bladerf META formats
2025-04-24 04:36:36 -07:00
Ryan Thompson
ecbfb99b8a libbladeRF: log sample rate ranges on failed sample rate assignment 2025-04-09 23:17:40 -07:00
Ryan Thompson
50eaaa6e0c libbladeRF: added direction to transfer timeout errors 2025-04-09 23:17:40 -07:00
Ryan Thompson
90bd8bb9cc doc: update sync config buffer size description 2025-04-09 23:17:40 -07:00
Ryan Thompson
234fcfeebf doc: added Doxygen support for light and dark modes 2025-04-09 23:17:40 -07:00
Ryan Thompson
0e2a978554 doc: Doxygen config update v1.8.8 > v1.9.8
└> doxygen -u Doxyfile.in
warning: Tag 'TCL_SUBST' at line 247 of file 'Doxyfile.in' has become obsolete.
         This tag has been removed.
warning: Tag 'COLS_IN_ALPHA_INDEX' at line 1059 of file 'Doxyfile.in' has become obsolete.
         This tag has been removed.
warning: Tag 'HTML_TIMESTAMP' at line 1194 of file 'Doxyfile.in' has become obsolete.
         This tag has been removed.
warning: Tag 'FORMULA_TRANSPARENT' at line 1475 of file 'Doxyfile.in' has become obsolete.
         This tag has been removed.
warning: Tag 'LATEX_SOURCE_CODE' at line 1749 of file 'Doxyfile.in' has become obsolete.
         This tag has been removed.
warning: Tag 'DOCBOOK_PROGRAMLISTING' at line 1911 of file 'Doxyfile.in' has become obsolete.
         This tag has been removed.
warning: Tag 'PERL_PATH' at line 2090 of file 'Doxyfile.in' has become obsolete.
         This tag has been removed.
warning: Tag 'CLASS_DIAGRAMS' at line 2103 of file 'Doxyfile.in' has become obsolete.
         This tag has been removed.
warning: Tag 'MSCGEN_PATH' at line 2112 of file 'Doxyfile.in' has become obsolete.
         This tag has been removed.
warning: Tag 'DOT_FONTNAME' at line 2154 of file 'Doxyfile.in' has become obsolete.
         This tag has been removed.
warning: Tag 'DOT_FONTSIZE' at line 2161 of file 'Doxyfile.in' has become obsolete.
         This tag has been removed.
warning: Tag 'DOT_TRANSPARENT' at line 2372 of file 'Doxyfile.in' has become obsolete.
         This tag has been removed.
warning: Changing CLASS_GRAPH option to TEXT because obsolete option CLASS_DIAGRAM was found and set to NO.
2025-04-09 23:17:40 -07:00
Ryan Thompson
72db17fa2c libbladeRF: fix packet format async stream to bytes
This fixes the sync interface packet format misalignment when sync
config assumes the sync config buffer is in bytes.
2025-04-09 23:17:40 -07:00
Ryan Thompson
2d2904e880 libbladeRF: added bladerf_format_to_string function 2025-04-09 23:17:40 -07:00
Ryan Thompson
0917d0780b libbladeRF: support for smaller GPIF buffer FPGA and FX3 versions 2025-04-09 23:17:40 -07:00
Ryan Thompson
0b00acac09 libbladeRF: error on GPIF buffer size mismatch between FX3 and HDL
While retaining the original non-fail condition for mismatches, we
throw an error for FPGA and FX3 images that have incompatible buffer
sizes. This prevents operational issues when mixing firmware versions
that support larger USB transfer buffers with FPGA versions that don't,
and vice versa.
2025-04-09 23:17:40 -07:00
Ryan Thompson
d586de33c1 libbladeRF: added legacy GPIF buffer size support for FW < v2.5.0 2025-04-09 23:17:40 -07:00
Ryan Thompson
b086beee7e libbladeRF: added error print for set frequencies out of range 2025-04-09 23:17:40 -07:00
Ryan Thompson
bfadb91e06 libbladeRF: fixed rx meta packet timestamp and flags assignment 2025-04-09 23:17:40 -07:00
Ryan Thompson
899cd2274f libbladeRF: refactored bladerf_enable_feature
These changes make it much easier to add features to libbladeRF.
The nested ifs have been restructured for a switch and errors leverage
the goto error. The device handle now is also null checked. The mutex
lock has also been removed.

Regression test passed.
2025-04-09 23:17:40 -07:00
Ryan Thompson
0b54c576e3 host: NULL_CHECK made variadic 2025-04-09 23:17:40 -07:00
Ryan Thompson
5ff41748aa libbladeRF: added sc16q11 packed error when initializing async stream
BLADERF_FORMAT_SC16_Q11_PACKED is only supported with the synchronous
interface.
2025-04-09 23:17:40 -07:00
Ryan Thompson
ff1b3d4e97 libbladeRF: added sc16q11 packed tx support 2025-04-09 23:17:40 -07:00
Ryan Thompson
79cae1e87f libbladeRF: added sc16q11 packed sample format
libbladeRF ingests SC12 Q11 samples and expands them to SC16 Q11 before
filling the end buffer. The trade-off is CPU cycles for the 12b to 16b
conversion.
2025-04-09 23:17:40 -07:00
Ryan Thompson
bdd4f39a3d fx3_firmware: rev libbladeRF, hdl, and fx3 for 2048 sample gpif buffers
FX3 GPIF buffers set to 2048 samples and number of buffers reduced to
11 per dma channel. This substantially increases the theoretical USB
SuperSpeed throughput from roughly 228MB/s to 356MB/s.

$ make test_libbladeRF
[       OK ] libbladeRF.scheduled_retune (19755 ms)
[----------] 29 tests from libbladeRF (308802 ms total)

[----------] Global test environment tear-down
[==========] 29 tests from 1 test suite ran. (308802 ms total)
[  PASSED  ] 29 tests.
2025-04-09 23:17:40 -07:00
Ryan Thompson
a0937cc50c doc: updated buf size limit in examples and tests 2025-04-09 23:17:40 -07:00
Ryan Thompson
a62c1d4903 libbladeRF: support and clamp byte alignment to larger GPIF buffer size
This change also includes a refactor of the redundant format to sample
size conversions found within the synchronous and asynchronous
interfaces.
2025-04-09 23:17:40 -07:00
Ryan Thompson
05ede8f541 fx3_firmware: increase DMA buffer sizes for USB throughput bottleneck
$ make test_hdl
[----------] Global test environment tear-down
[==========] 63 tests from 2 test suites ran. (381209 ms total)
[  PASSED  ] 63 tests.
[100%] Built target test_hdl
2025-04-09 23:17:40 -07:00
Ryan Thompson
a8481c2af7 doc: added packet length warning for BLADERF_FORMAT_PACKET_META
A packet length of (n*256-4) DWORDs will cause the synchronous
buffer it's filled into to indefinitely persist.
2025-04-09 23:17:40 -07:00
Ryan Thompson
73e78639aa doc: fix API sections menu item with Doxygen 1.9.8
The API sections aren't stored as modules anymore. They're topics.
Error likely introduced at doxygen commit: 6d80fc7e5
2025-04-09 23:17:40 -07:00
Ryan Thompson
179fe7c600 hdl: corrected bladeRF 1.0 FPGA version number 2025-01-06 13:36:48 -06:00
Asa Zernik
cbc3d30a27 libbladeRF: Return error code instead of asserting in version_check
Specifically, when there's a missing compat table entry. All callers
specify that they want to be forgiving of version mismatches, so this
seems in the spirit of the call tree.

Ran into this error when playing with a custom-loaded FPGA; if I had
flashed the FPGA instead of simply loading it, this could have bricked
the blade (or at least forced me into a very involved recovery process).
2024-08-13 11:43:21 +03:00
Ryan Thompson
e1a5c9260d libbladeRF: fix calibration load check for get_gain_calibration
Fixes #981
2024-08-10 18:16:51 -07:00
Ryan Thompson
f81b82aa8e libbladeRF: bump to version 2.5.1
Introduces bladeRF-power cli tool and gain calibration functionality.
2024-08-08 00:15:01 -05:00
Ryan Thompson
5e041d0ada libbladeRF: gain_calibration: added serial number to gain cal table
The gain calibration table will compare the serial number in the CSV
with the serial number of the opened device. A warning will be issued
if the serial numbers do not match.
2024-08-07 15:35:55 -05:00
Ryan Thompson
b185d4c60a libbladeRF: gain_calibration: added cal table valid entry check 2024-08-07 15:35:45 -05:00
Ryan Thompson
125d96f4b1 libbladeRF: gain_calibration: restore gain mode after cal load 2024-08-07 15:32:31 -05:00
Ryan Thompson
2ab7adc676 libbladeRF: gain_calibration: added header check for CSV conversions
This blocks RX gain calibrations from being applied to TX and
vice versa.
2024-08-07 15:31:54 -05:00
Ryan Thompson
3c25f57893 libbladeRF: gain_calibration: added bladerf_get_gain_target
The gain target is either set or calculated based on the gain mode. If
using manual gain control, the gain target from the gain table is
returned. If the AGC is enabled, the gain target is calculated based
on the current gain.
2024-08-07 15:31:54 -05:00