DISCO_F769NI EMAC driver may return ethernet packet with illegal
length when driver is under heavy load. In one case, the received
bytes indicate frame length of 53 bytes but advertised data length
was 65518 bytes. In another case EMAC driver variable
`EthHandle.RxFrameInfos.length` contained value 0xFFFF FFFC.
As a work-around accept only 1-1500 bytes long ethernet packets.
ARCH_MAX board uses 0x01 as PHY address, different from other NUCLEO STM32 boards. To faciliate change of PHY address of customized boards, a configuration entry eth-phyaddr was added and can be overriden as needed in mbed_app.json by adding "stm32-emac.eth-phyaddr": X. Macro ETH_ARCH_PHY_ADDRESS was also renamed to ETH_PHY_ADDRESS, because it is not only used by ARCH_MAX board but also other boards.
Issue: udp_sendto() Fails for multicast IPV6 packet when interface is switched from SoftAP to STA mode.
When SoftAP start is called, add_ethernet_interface() will be called internally to add interface details into netif_list.
When switching from SoftAP to STA mode, add_ethernet_interface() will be called again to append the interace details into netif_list.
When udp_sendto() is called, ip6_route() will return interface as NULL since it consider device as single interface.
Fix: SoftAP mode Stop, call remove_ethernet_interface() to remove the interface from the netif_list.
Cypress Target update to support WiFi Manufacturing Test
Adding IOCTLS in WhdSTAInterface for MFG Test.
Fix WHD EMAC interface to release buffer when emac is not powered up or callback
is not registered.
- A shared mutex is added for synchronization
- ScopedMutexLock is used to to protect
- SoftAP: start, stop
- STA: scan, join, disconnect
- Fix switching issue between SoftAP and STA mode for primary interface
- Avoid reinit primary interface by getting mapping the current interface to the other one which is already on
- In concurrent mode, STA is the default if it is up, otherwise SoftAP is default.
- For non-concurrent mode, the most recent started interface is set as default.
- Add WhdAccessPoint to include additional WHD scan info
- To save memory, only move assignment is supported for WhdAccessPoint
- Add scan_whd to scan for WhdAccessPoint
- Set set_blocking(false) to unsupported by return NSAPI_ERROR_UNSUPPORTED
This allows to specify which hal version to use for each family.
It can also be used to modify the thread stack size.
Signed-off-by: Vincent Veron <vincent.veron@st.com>
This port is based on :
* CurryGuy ethernet branch :
https://github.com/CurryGuy/mbed-os/tree/feature-stm32h7-emac
* STM32 Cube example :
Applications/LwIP/LwIP_HTTP_Server_Netconn_RTOS example
Signed-off-by: Vincent Veron <vincent.veron@st.com>
Deprecate wait() in favour of acquire(), try_acquire(),
try_acquire_for() and try_acquire_until().
Brings Semaphore more into line with CMSIS-RTOS 2 (which uses "acquire"),
itself (as it has "release"), and other classes having "try", "try for"
and "try until".
Also steps away from vague "wait" term - the primary operation here is
to acquire the semaphore, and this will of course sleep.
Since commit 12c6b1bd8, the i.MX RT1050 has effectively had its data
cache disabled, as the SDRAM was marked Shareable; for the Cortex-M7,
shareable memory is not cached.
This was done to make the Ethernet driver work without any cache
maintenance code. This commit adds cache maintenance and memory barriers
to the Ethernet driver, and removes the Shareable attribute from the
SDRAM, so the data cache is used again.
Cache code in the base fsl_enet.c driver has not been activated - the
bulk of it is in higher-level Read and Write calls that we're not using,
and there is one flawed invalidate in its initialisation. Instead
imx_emac.cpp takes full cache responsibility.
This commit also marks the SDRAM as read/write-allocate. As the
Cortex-M7 has its "Dynamic read allocate mode" to automatically switch
back to read-allocate in cases where write allocate is working poorly
(eg large memset), this should result in a performance boost with no
downside.
Activating write-allocate is also an attempt to provoke any flaws in
cache maintenance - the Ethernet transmit buffers for example will be
more likely to have a little data in the cache that needs cleaning.
Kintis EMAC is consuming 16 rinbuffers for input, and 8 buffers for output.
This is over-use because input packets are immediately allocated from
heap when passed to LwIP. Therefore the number can be creatly reduced.
Praveen babu chandran
Martin Kojtal
Arto Kinnunen
jeromecoutant
Zhiyonh Li
midd
Michal Paszta
Arun S
Caleb Szalacinski
Mahesh Mahadevan
cyliangtw
Hamza Rizwan
Balaji Subramanyam
fred.li
Bence Kaposzta
Ryan Morse
Lei Zhang
Rohan Fletcher
Tymoteusz Bloch
aqib-ublox
Kyle Kearney
Vincent Veron
Teja Narahari
Volodymyr Medvid
Kevin Bracey
蔡雪 (Xue Cai)
Kari Haapalehto
d-kato
Seppo Takalo
Marcin Tomczyk
c_jin
Jimmy Brisson
chao_king
Qinghao Shi