Commit message (Collapse) | Author | Age | Files | Lines | |
---|---|---|---|---|---|
* | utils: uhd_find_devices display one device for each unique serial found | Andrej Rode | 2016-11-22 | 1 | -6/+44 |
| | | | | | | | Note: This also is the first precedent for the usage of the 'auto' keyword in UHD. Commits past this one will also be able to use 'auto'. Reviewed-By: Martin Braun <martin.braun@ettus.com> | ||||
* | cmake: set c++11 flags in gcc and clang | Andrej Rode | 2016-11-22 | 1 | -2/+18 |
| | | | | | | | This will enable all C++11 features that are available on all currently supported compilers. Reviewed-By: Martin Braun <martin.braun@ettus.com> | ||||
* | Merge branch 'maint' | Martin Braun | 2016-11-15 | 2 | -33/+33 |
|\ | |||||
| * | fixup! Remove all boost:: namespace prefix for uint32_t, int32_t etc. ↵ | Martin Braun | 2016-11-15 | 5 | -123/+123 |
| | | | | | | | | | | | | (fixed-width types) Now also removes the namespaces in the utils/ directory. | ||||
* | | Merge branch 'maint' | Martin Braun | 2016-11-15 | 3 | -19/+19 |
|\| | |||||
| * | fixup! Remove all boost:: namespace prefix for uint32_t, int32_t etc. ↵ | Martin Braun | 2016-11-15 | 3 | -19/+19 |
| | | | | | | | | | | | | (fixed-width types) Now also removes the namespaces in examples/ | ||||
* | | Merge branch 'maint' | Martin Braun | 2016-11-15 | 2 | -2/+3 |
|\| | |||||
| * | n230: Make second dual-eth addr consistent with standard | Ashish Chaudhari | 2016-11-14 | 1 | -2/+2 |
| | | | | | | | | - The second address argument was changed from "secondary-addr" to "second_addr" | ||||
| * | devtest: test_base add wait time between uhd_usrp_probe calls | Andrej Rode | 2016-11-09 | 1 | -0/+1 |
| | | |||||
* | | Merge branch 'maint' | Martin Braun | 2016-11-09 | 1 | -7/+7 |
|\| | |||||
| * | cmake: Added the ability to specify package names | Paul David | 2016-11-09 | 1 | -7/+7 |
| | | |||||
* | | rfnoc: Minor cleanup in ctrl_iface | Martin Braun | 2016-11-09 | 2 | -100/+12 |
| | | | | | | | | | | | | | | | | - Removed dump queue concept (was geared towards b200 anyway) - Removed some superfluous lines - Requires bidirectional transports Reviewed-By: Michael West <michael.west@ettus.com> | ||||
* | | rfnoc: Removed transport endianness as make arg | Martin Braun | 2016-11-09 | 4 | -8/+0 |
| | | | | | | | | | | | | | | | | | | This is only confusing, since the blocks do not have to care about endianness. The control interface needs to know (and takes care of it internally), and the streaming transport does the same (and does not directly connect to blocks). Reviewed-By: Michael West <michael.west@ettus.com> | ||||
* | | rfnoc: Moved transport endianness as property into both_xports_t | Martin Braun | 2016-11-09 | 6 | -17/+13 |
| | | | | | | | | Reviewed-By: Michael West <michael.west@ettus.com> | ||||
* | | Merge branch 'maint' | Martin Braun | 2016-11-08 | 311 | -3153/+3285 |
|\| | |||||
| * | Remove all boost:: namespace prefix for uint32_t, int32_t etc. (fixed-width ↵ | Martin Braun | 2016-11-08 | 298 | -3179/+3182 |
| | | | | | | | | | | | | | | | | types) - Also removes all references to boost/cstdint.hpp and replaces it with stdint.h (The 'correct' replacement would be <cstdint>, but not all of our compilers support that). | ||||
| * | X300: Fix for recv timeout and wait_for_ack errors over PCIe. | michael-west | 2016-11-08 | 1 | -9/+59 |
| | | | | | | | | | | | | - Reduced allocation of buffers for each control stream. - Copied contol data into new managed buffer so managed buffer of base transport could be released. | ||||
| * | X300: Fix for channel 1 failure to ACK a TX burst. | michael-west | 2016-11-07 | 2 | -24/+63 |
| | | | | | | | | | | | | - Added mapping from radio block and port to channel in legacy compat layer. - Improved parsing of stream args and added parsing of radio ID and port parameters for each channel. - Added proper programming of the response SID based on the new radio ID and port parameters. | ||||
| * | device3: Removed some superfluous includes | Martin Braun | 2016-11-04 | 2 | -12/+1 |
| | | |||||
| * | X300: Check dboard time property before subscribing | Derek Kozel | 2016-11-04 | 2 | -6/+7 |
| | | | | | | | | | | If a daughterboard does not have a time/cmd property then no subscriber should be added to the motherboard time/cmd property. | ||||
| * | Debian: Fix changelog format | Derek Kozel | 2016-11-02 | 1 | -3/+0 |
| | | |||||
| * | Preparing branch for 3.10.1.0 release | Derek Kozel | 2016-11-02 | 2 | -2/+33 |
| | | | | | | | | | | - Updated CHANGELOG - Updated images package | ||||
| * | Windows: Updated and signed USB INF files | Derek Kozel | 2016-11-02 | 12 | -26/+39 |
| | | |||||
| * | TwinRX: Add rev B daughterboard ID | Derek Kozel | 2016-11-01 | 1 | -0/+9 |
| | | |||||
| * | Docs: Fixed dead external link | Derek Kozel | 2016-10-31 | 1 | -1/+1 |
| | | |||||
* | | cmake: Change version numbering to remove leading zeros | Martin Braun | 2016-10-31 | 8 | -48/+25 |
| | | |||||
* | | Merge branch 'maint' | Martin Braun | 2016-10-28 | 12 | -26/+400 |
|\| | |||||
| * | docs: Added section on UHD semantic versioning | Martin Braun | 2016-10-28 | 2 | -0/+144 |
| | | | | | | | | Originally authored by: Ben Hilburn <ben.hilburn@ettus.com> | ||||
| * | Preparing branch for 3.10.1.0 release | Derek Kozel | 2016-10-27 | 1 | -2/+2 |
| | | | | | | | | | | | | - Updated CHANGELOG - Updated fpga-src submodule pointer - Updated images package | ||||
| * | rfnoc legacy compat: Call update_*_streamer on all samp rate updates | Martin Braun | 2016-10-27 | 1 | -5/+44 |
| | | | | | | | | Reviewed-By: Derek Kozel <derek.kozel@ettus.com> | ||||
| * | xcvr: Removed some compiler warnings | Martin Braun | 2016-10-27 | 1 | -1/+1 |
| | | |||||
| * | docs: Added R&D testing procedures | Martin Braun | 2016-10-27 | 2 | -0/+96 |
| | | |||||
| * | TwinRX: Remove unhelpful warning | Derek Kozel | 2016-10-26 | 1 | -8/+0 |
| | | |||||
| * | TwinRX: Use command time to schedule synchronous events | Derek Kozel | 2016-10-26 | 5 | -6/+104 |
| | | |||||
| * | Experts: Always resolve all nodes in the graph | Derek Kozel | 2016-10-26 | 1 | -4/+9 |
| | | |||||
* | | Merge branch 'maint' | Martin Braun | 2016-10-26 | 22 | -61/+75 |
|\| | |||||
| * | CMake: fix the lvbitx path on windows | Paul David | 2016-10-26 | 1 | -6/+6 |
| | | |||||
| * | docs: Fixed notes on transport parameters | Martin Braun | 2016-10-26 | 2 | -15/+3 |
| | | |||||
| * | Merge pull request #1335 from EttusResearch/expert_tweaks | Martin Braun | 2016-10-25 | 1 | -1/+8 |
| |\ | | | | | | | Expert tweaks | ||||
| | * | Experts: Fix typo | Derek Kozel | 2016-10-25 | 1 | -1/+1 |
| | | | |||||
| | * | Experts: Add a printer for time_spec_t | Derek Kozel | 2016-10-25 | 1 | -0/+7 |
| | | | |||||
| * | | x300: Throttle MTU to 3000 for PCIe to avoid underruns | Martin Braun | 2016-10-25 | 1 | -1/+4 |
| |/ | | | | | | | This is a temporary workaround to make PCIe available on lower rates. | ||||
| * | Docs: Minor housekeeping | Derek Kozel | 2016-10-21 | 12 | -31/+31 |
| | | |||||
| * | rfnoc: Fixed case where stream terminators would not notify node about ↵ | Martin Braun | 2016-10-19 | 2 | -0/+3 |
| | | | | | | | | active status | ||||
| * | rfnoc: Fixed graph search algorithm for active-ports only | Martin Braun | 2016-10-19 | 1 | -1/+1 |
| | | |||||
| * | tests: Updated RFNoC tests after some API changes | Martin Braun | 2016-10-19 | 3 | -6/+19 |
| | | |||||
* | | Merge branch 'maint' | Martin Braun | 2016-10-19 | 8 | -31/+39 |
|\| | |||||
| * | b200: Make sure no GPS settings are applied when no gpsdo present. | Martin Braun | 2016-10-18 | 1 | -1/+4 |
| | | | | | | | | | | | | | | This changes the exception message thrown when gpsdo is selected as a clock source, but does not change the exception. Note that before, during its first run, the B2x0 would happily accept gpsdo as a clock source even when none was present. | ||||
| * | Preparing branch for 3.10.1.0 release | Martin Braun | 2016-10-18 | 2 | -3/+3 |
| | | | | | | | | | | | | | | - Updated CHANGELOG - Updated fpga-src submodule pointer - Updated images package - Updated version string | ||||
| * | Fix warning in e300_remote_codec_ctrl.cpp for gcc. | Philip Balister | 2016-10-18 | 1 | -2/+2 |
| | | | | | | | | | | | | | | | | The bigger question is why uhd forces the e300 code to have an implementation of an empty function. But, suppress the warning and let people fight about design later. Signed-off-by: Philip Balister <philip@opensdr.com> |