I started poking around to see if I can get my build files packaged up so we can use them during manufacturing. It looks like the normal build command doesn't create build/$project/APP.bin, but we use that to jlink flash controllers during manufacturing. I started poking around to see how it gets pulled in and it looked like adding either "download" or "package" should build it. I confirmed it happens during download (but that tries to actually use jlink and download to a device). I noticed just adding "package" doesn't do anything. After digging in more I noticed it was setup like this:
$(QUIET)$(ECHO) Created package successfully
$(RELEASE_PACKAGE): create_package_descriptor \
$(STRIPPED_LINK_OUTPUT_FILE) display_map_summary \
package_bootloader $(if $(findstring no_dct,$(MAKECMDGOALS)),,package_dct) package_app package_apps
in tools/makefiles/standard_platform_targets.mk, but RELEASE_PACKAGE isn't defined anywhere. If I pass in something defining it on the make line, I get the next issue:
make: *** No rule to make target `create_package_descriptor',
I can't find any info on create_package_descriptor. I was wondering if this was meant to be removed and/or if I'm even going about this the right way. I'm currently using WICED-Studio-184.108.40.206 but I also tried on WICED-Studio-6.6.1 and saw the same issue. Any help or guidance would be appreciated.
Is there a clear centralized exhaustive list (and history) of know vulnerabilities in the CYW4343W (or all chips) firmware ?
Browsing https://github.com/Infineon/wifi-host-driver commits to RELEASE.md (like that Upload wifi-host-driver 220.127.116.1131 · Infineon/wifi-host-driver@19968e1 (github.com)) I can see that there is a few changelogs related to the CYW4343W firmware.
--- 18.104.22.168 ---
Fix pmk caching
--- 22.214.171.124 ---
Memory usage reduction by disabling debug features
--- 126.96.36.199 ---
Fixed zero stall on UDP
Fixed Tx traffic too less then RX
--- 188.8.131.52 ---
Fixed zero stall on UDP
--- 184.108.40.206 ---
Security fix (KRACK all-zero-key)
--- 220.127.116.11 ---
Security fix(Dragonblood WPA3 attack)
TCP Keepalive Implementation
Security fix(CVE-2019-9501 / CVE-2019-9502)
--- 18.104.22.168 ---
This list is not easy to build and browse, the known vulnerabilities should be centralized.
Is this list exhaustive ?
How can we know what version exactly fixes a vulnerability ? This only show ranges...
Between 22.214.171.124 and 126.96.36.199, it is only mentioned "Security fixes"... Where can we get more details on this/these vulnerability(ies) ?
Looking at this blog post (Potential Fragmentation Vulnerabilities for Wi-Fi ... - Infineon Developer Community), it looks like the CYW4343W could by affected. How can we make sure whether it is or not ?
Any more information about firmware vulnerabilities is welcome.
Thanks and best regardsShow Less
I have a TCP socket opened against our server. We usually transmit up to 40KB/s (although faster would be better). When transmitting, often times, I will either run into a scenario where the tcp send reports that the connection is closed or just an error. I'm not sure what is going on. We are using NetX as the network stack.
Also, when looking at wireshark, I do not see a RST sent by the server. The last thing I see is always the server sending an ACK. I've tried disabling low power, throttling down the transfer rate, changing the MTU size, changing the device's ACK retry timeout.Show Less
I have a STM32F412+CYW43438 EVB to develop a ioT device.
I try to follow all debug configuration settings on forum but I get a problem here.
I set breakpoint at my application_start in Startup option but actually start at WICED/platform/ARM_CM4/crt0_GCC.c _start
Have someone to help this issue?
I am able to write to a dev kit using a segger jlink, however when trying to program our custom hardware (which is very similar to the dev kit) it reads successfully the id 0x5BA00477 but quickly fails afterwards.
SEGGER J-Link Commander V7.63b (Compiled Mar 1 2022 16:42:47) DLL version V7.63b, compiled Mar 1 2022 16:41:28 Connecting to J-Link via USB...O.K. Firmware: J-Link V11 compiled Dec 9 2021 14:14:49 Hardware version: V11.00 S/N: 821008088 License(s): GDB USB speed mode: High speed (480 MBit/s) VTref=3.180V Device "CYW43907" selected. Connecting to target via JTAG TotalIRLen = 4, IRPrint = 0x01 JTAG chain detection found 1 devices: #0 Id: 0x5BA00477, IRLen: 04, CoreSight JTAG-DP Identified core does not match configuration. (Found: None, Configured: Cortex-R4) TotalIRLen = 4, IRPrint = 0x01 JTAG chain detection found 1 devices: #0 Id: 0x5BA00477, IRLen: 04, CoreSight JTAG-DP ****** Error: Could not power up debug port: Control/Status register reads 00000F02 TotalIRLen = 4, IRPrint = 0x01 JTAG chain detection found 1 devices: #0 Id: 0x5BA00477, IRLen: 04, CoreSight JTAG-DP TotalIRLen = 4, IRPrint = 0x01 JTAG chain detection found 1 devices: #0 Id: 0x5BA00477, IRLen: 04, CoreSight JTAG-DP Cannot connect to target.
When looking at JTAG signals, we can see that the voltages are unexpected:
According to segger:
The J-Link has 100 Ohm serial resistors at TDI/TMS and 68 Ohm at TCK.
The voltage drop at TDI/TMS is 0.9V with 9mA load current
The voltage drop at TCK is 0.7V with 9mA load current
According to CYW43907 doc:
GPIO have programmable 2 mA to 16 mA drive strength. Default is 10 mA
Are the GPIOs set as output to GND with an empty flash?Show Less
I want to test this module WM-BAN-BM-09_S, but I can not find any information about it.
I can't contact with the USI global, they do not reply my e-mail.
Does anyone have the datasheet and firmware/nvram ?Show Less
I'm using a wifi module AP6330, the chipset is BCM4330. I want create a 80211an AP with SHORT-GI (72.2Mbps).
My system is run on Linux 4.9 mainline bcmdhd, BCM4330 firmware version is fw_bcm40183b2_ag_apsta.bin
FW Version string:
4330b2-roml/sdio-ag-pool-wme-apsta-idauth Version: 188.8.131.52 CRC: 4b89dc94 Date: Mon 2013-07-22 15:40:35 CST FWID 01-b77f69d6
hostapd-2.9 with this configuration:
then I check the AP on my laptap, the max connection speed is 65Mbps
and I grap the beacon packet, the bit (SHORT-GI-20) is not set int the HTcap field
Does this chipset(BCM4330) support SHORT-GI-20 ?Show Less
I have developed a project using CYW954907AEVAL1F.
I need the possibility to let user to upgrade my APP without using an external programmer.
I browsed "snip/ota_fr" and I notice that it starts an OTA server (see "wiced_ota_server_start" in "ota_fr.c" row 205).
I browsed the source of OTA server on "libraries\daemons\ota2_server" and I observed that in "wiced_ota2_server.c" there are two functions that are called. The first is ...
wiced_ota2_image_write_data( BUFFER, OFFSET, SIZE)
that I understand it programs FLASH with the new incoming APP version.
I modify my project makefile by adding CYW954907AEVAL1F
$(NAME)_COMPONENTS := filesystems/ota2
and I add a call to "wiced_ota2_image_write_data" in my source.
When I try to make my project the linker says to me
"undefined reference to `wiced_ota2_image_write_data'"
What I'm missing ?Show Less
We use Wiced Studio 6.6 for development of Bluetooth applications based on the Laird_Sterling Bluetooth Module (CYW4343W). We now have the need to use the Apple MFI Coprocessor to implement iAP2 over Bluetooth. How can we get sample code, examples, documentation or a Wiced Studio version including Apple MFI code or libraries ? We do have a MFI number.Show Less
The TCP client cannot use the Keepalive function.
Using this function doesn't work:
Wiced_tcp_enable_keepalive (&tcp_client_socket,TCP_KEEP_ALIVE_INTERVAL, TCP_KEEP_ALIVE_PROBES,TCP_KEEP_ALIVE_TIME);
The following methods will not work:
Changing the value of NX_TCP_KEEPALIVE_INITIAL, NX_TCP_KEEPALIVE_RETRIES, and NX_TCP_KEEPALIVE_RETRY in nx_tcp. h and nx_user.h does not work. The macro NX_ENABLE_TCP_KEEPALIVE and NX_ENABLE_TCP_KEEPALIVE are defined.Show Less