| 14 | | '''Important:''' The v1.5.2 hardware design uses updated hardware support cores. You must update your [wiki:edk_user_repository edk_user_repository] to build the v1.5.2 hardware or software projects. |
| | 15 | '''Important:''' The v1.5.3 hardware design uses updated hardware support cores. You must update your [wiki:edk_user_repository edk_user_repository] to build the v1.5.3 hardware or software projects. |
| | 16 | |
| | 17 | Release v1.5.3 contains both hardware and software improvements. |
| | 18 | |
| | 19 | '''Changes in v1.5.3:''' |
| | 20 | |
| | 21 | '''PHY / MAC Hardware Changes:''' |
| | 22 | * None - same PHY and MAC cores as v1.5.2. |
| | 23 | |
| | 24 | '''MAC Software Changes:''' |
| | 25 | * Added high framework hooks to enable/disable bridging ETH A with the wireless interface, used by wlan_exp (see below) |
| | 26 | * Fixed time units bug in implementation of Random Schedule LTG ([changeset:5564 changeset 5564]). |
| | 27 | * Fixed bug when Python called {{{enable_DTIM_multicast_buffering()}}} when beacons are already disabled. This sequence caused substantial heap churn in previous releases, eventually crashing the node when the heap couldn't accommodate new allocations elsewhere in the code ([changeset:5563 changeset 5563]). |
| | 28 | * Fixed uninitialized pointer bug in creation of association responses - thanks to [http://warpproject.org/forums/viewtopic.php?pid=15375#p15375 kirchhof] for the bug report ([changeset:5566 changeset 5566]). |
| | 29 | * Fixed null pointer bugs in upper MAC framework/apps - thanks to kirchhof for the bug reports [http://warpproject.org/forums/viewtopic.php?id=3206 1], [http://warpproject.org/forums/viewtopic.php?pid=15404#p15404 2] ([changeset:5570 changeset 5570]). |
| | 30 | |
| | 31 | '''Experiments Framework Changes:''' |
| | 32 | * Added {{{enable_ethernet_portal(bool)}}} node method that controls whether the high-level MAC implements the wired-wireless bridge between the ETH A interface and the wireless interface. It is good practice to disable the Ethernet portal when running LTG-based experiments. This helps avoid issues with unaccounted for Tx/Rx events due to stray Ethernet packets on the wired network. See the updated '''FIXME''' log_capture_two_node_two_flow.py script for an example. |
| | 33 | |
| | 34 | '''Misc Hardware Changes:''' |
| | 35 | * Reworked the debug header IO using the updated w3_userio core. Now both CPUs can read and write debug pins. Individual pins can be configured as hardware-controlled (for MAC/PHY status signals) or software-contollled (for C code benchmarking). See the updated [wiki:../Usage/UserIO#DebugHeaderuser Debug Header user guide page] for details. |
| | 36 | |
| | 37 | ---- |
