Pico-SDK changed the output extension of their compilation from "x.c.obj"
to "x.c.o" meaning the removed Newlib and STDIO SDK wrappers were still
being linked in certain situations. Update make-libpico Cmakefile to
remove the new names.
Fixes debug `printf` output.
Leave that to the main app instead, so we don't reset peripherals twice.
* ota: fix copy error for riscv headers
* ota: initialize bootrom bit ops and add explanation
Using pico-sdk develop branch, add in support for CYW43-based
WiFi/BT/BLE boards on the RP2350 such as the
SparkFun Thing Plus RP2350 or the Pimoroni Pico Plus 2W.
Fixes#2608
Rolls in dynamic SPI divider #2600
* Support LED digitalWrite on RP2350+CYW
Also move "special GPIO" to 64 since the Pimoroni Pico 2W uses the
RP2350B with 48 GPIOs.
* Enable CYW43_PIN_WL_DYNAMIC in IDE and P.IO
Allows calling `cyw43_set_pins_wl(cyw43_pin_array);` to redefine the
CYW43 hookup in the variant initialization.
* Enable LWIP IGMP, MDNS internal server
* Enable MDNS lookup from LWIP DNS
* Add SimpleMDNS responder, small code and no malloc
* Ensure we copy out lwipopts in make-libpico
Adds a small wrapper around the LWIP-provided MDNS responder application.
Drop-in replacement in many basic cases for LEAmDNS.
For FreeRTOS it is important to not allocate memory on an LWIP callback.
LEAmDNS needs to do this to create response objects, leading to crashes.
Increase LWIP timers by bumping the LWIP_ARP number (as done before).
Replace ArduinoOTA LEAmDNS with SimpleMDNS and update a
HTTPUpdateServer example.
33% faster for 4K memcpy using DMAMemcyp example
With this assembly:
CPU: 4835 clock cycles for 4K
DMA: 2169 clock cycles for 4K
Using stock Newlib memcpy:
CPU: 7314 clock cycles for 4K
DMA: 2175 clock cycles for 4K
The flash ROM routines seem to overwrite the QMI configuration we set for
PSRAM, rendering it unreadable after any erase or write or ID command.
Wrap the 4 flash control functions to preserve the QMI state on the
RP2350. On the RP2040, simply pass through the call.
Fixes#2537
* Adds RISC-V compilation option to the IDE and Platform.IO.
* Build RP2350-RISCV libpico, libbearssl
* Fix RP2350 BearSSL library (was copied from RP2040, now built for M33)
* New GCC 14.2 toolchain is required (12.4 RISC-V support is borked)
* Newlib locking fixed prototypes
* Manually force all runtime init code into RP2350 binaries
* Add RISC-V to CI
* Remove RP2350 BOOT2.S files, binaries (not used)
* Clean up minor GCC 14.x warnings
* Add RP2350-RISCV OTA build, link
* Add RISC-V FreeRTOS files (configuration still not running, but builds)
* Add basic documentation
Newlib built using `-Os` causes things like `memcpy` to be very slow on the
RP2350 because it uses byte-wise operations. On the RP2040 this doesn't
matter because there is a ROM routine we use instead of the library, but on
the Pico 2 it's almost 10x slower than the optimal method.
Update GCC to 12.4
Update Newlib to 4.4.0
Move to -O2 library compilation
New toolchain looks to add ~10K to RP2350 flash usage (less on the RP2040).
Split the 1300+ char define-setting line into a multi-line value.
Makes it more maintainable and easier to see when changes needed.
No settings should have been modified.
The RP2350 has a different blob header requirement to identify a working
image. Ensure that header is present in the OTA loader.
Update the PicoOTA examples for the 2350
* Add support for the extra 16 GPIO pins in the menus and core.
* Clean up Generic RP2350 PSRAM ("none" is valid) and flash (other than 16MB) options.
* Add extra GPIO<->peripheral connections
* Add Pimoroni PGA2350 RP2350B-based board
* Pins 32-47 can be used for PIOPrograms
* Avoid hang when PSRAM fails to initialize
* Move libpico to an RP2350B board for SDK (otherwise the SDK drops all GPIOHI support)
* Migrate RP2040-specific bits to separate dirs
* Add chip to boards.txt, isolate RP2040-specifics
* Add RP2350 boot2, bearssl, and libraries
* Platform.IO adjust to new paths
* Add RPIPICO2 JSON for P.IO
* Add RP2350 to Platform.io
* Update Picotool and OpenOCD for all hosts
* Use picotool to generate UF2s
* Build separate libpico blobs serially
Thanks for the review, @aarturo182 !
* Add RP2350 to CI
* Allow Ethernet/WiFi building for RP2350
* Update Adafruit TinyUSB to latest
* Test skip fix
* Make RP2350 Picotool work. update USB ID
* Fix EEPROM/FS flash locations
RP2350 adds a 4K header sector to the UF2, meaning we have 4K less total
flash to work with. Adjust all constants appropriately on the RP2350.
* Adds ilabs board and PSRAM support. (#2342)
* Adds iLabs boards and basic PSRAM support.
* Make PSRAM come up as part of chip init
Uses SparkFun psram.cpp to set timings on clocks which are defined in the
variant file. Prefix things with RP2350_PSRAM_xxx for sanity.
Users don't need to call anything, PSRAM "just appears". Still need to
add in malloc-type allocation.
* Add board SparkFun ProMicro RP2350
Same pinout as the SparkFun ProMicro RP2040 with 8MB PSRAM and RP2350
* Add TLSF library for use w/PSRAM
Fork of upstream to include add'l C++ warning fixes.
* Add pmalloc/pcalloc to use PSRAM memory
free() and realloc() all look at the pointer passed in and jump to the
appropriate handler. Also takes care of stopping IRQs and taking the
malloc mutex to support multicore and FreeRTOS (when that workd)
* Fix BOOTSEL for RP2350
* Add simple rp2040.idleOtherCore test
* Add Generic RP2350 and clean up PSRAM menus
Commercial boards now only have 1 size PSRAM, no need to have menu for them.
* Add Solder Party RP2350 Stamp boards (#2352)
* Add PSRAM heap info helpers, mutex lock mallinfo
* Add RP2350 docs
* FreeRTOS and OTA unsupported warnings for RP2350
* Update to 2.0.0 SDK
* Board type needs to be set before earliest SDK setup
* Platform includes update
* Boot2 files
* Simple compilation issues
* Build and link
* PIO rebuild with version
* Newlib wrapper update
* Force inclusion of runtime_init_* fcns
The linker was dropping all references to the library's included
runtime_init_xxx functions and hence things like the IRQ vector
table and mutexes and alarms weren't properly set up leading to
instant crashes on start up..
Explicitly call out one function from the object file stored in
the .A to force the inclusion of all the functions. May be a better
way, heppy to hear any ideas.
* Fix SPI GPIO calls
* Fix Ethernet GPIO
* Remove SDK warnings
Remove the skipped error messages once the following PR merged:
https://github.com/raspberrypi/pico-sdk/pull/1786
* BTStack moved SBC encode/decode paths
* Platform.IO fixes
* BT No longer has special absolute mouse
* Rebuild and update OTA
* Rebuild BearSSL, too
* Update liker file to latest SDK
* Clean up libpicocmake
* Clean up LWIP/BT library names
The LittleFS folks changed the on-flash format in 2.6.0, making
it unmountable with earlier precompiled code.
Rebuild the OTA bootloader using the 2.9.3 LittleFS release,
matching the core version.
Fixes#2198
Add a "-32K" option to all the IP stack options that doubles the PCB and memory
pools from default. For most use cases this is not necessary, but it could be
helpful in cases where large numbers of TCP clients are connected or high
bandwidth applications.
Fixes#2050
Adds a library to run classic Bluetooth A2DP source (output) audio from
the PicoW. Simple example showing operation and callbacks.
Factor out multiple BT lock/unlock and place in the PicoW variant files.
Use -O2 only on the LWIP checksum routine, resulting in a speedup of
around 13% (checksumming only, not entire LWIP stack) for 72 add'l bytes
of flash.
As seen in debug of #2149, if the LwipIntfDev is already _started,
return false for a ::begin() call.
Also, protect netif_add/_remove on the very small possibilty of being
called by LwipIntfDev devices while the CYW43 driver is doing work.
Avoid issues with interrupts and priority inversions and other deadlocks
and use a SW based random generator for LWIP when under FreeRTOS.
This means removing any overrides for sleep_until and the two
get_rand_xx calls from the SDK, making things much saner.
Related to #1883, #1872, and other random FreeRTOS lockups.
Enable use of wired Ethernet modules as first-class LWIP citizens. All
networking classes like MDNS, WebServer, HTTPClient, WiFiClient, and OTA
can use a wired Ethernet adapter just like built-in WiFi.
Two examples updated to show proper use.
Uses the Async Context support built into the Pico SDK. When running on the
Pico it will use the CYW43 async instance.
Uses modified wired Ethernet drivers, thanks Nicholas Humfrey!
Note, the classic, non-LWIP integrated `Ethernet` and related libraries
should still work fine (but not be able to use WebServer/HTTPS/etc.)
Fixes#775
Fixes#1394
The Pico_Rand SDK calls gather bits from the HW ROSC at precise intervals.
If there is jitter in the sleep_until() call then the ROSC bit collection
will always think it's failed to acquire the right bit and retry infintitely.
Avoid by wrapping the HW random number calls and the sleep_until() routine.
Only when in FreeRTOS set a flag to silently make sleep_until() into a
busy wait loop while in a random number generation step. When not in the
random code, do the normal sleep_until call.
* Update to Pico-SDK v1.5
* Hook in pico_rand, use ioctl to set ipv6 allmulti
* Move into PicoSDK LWIP mutex, hack timer sizes
* Utilize much of the PicoSDK infrastructure for WiFi
* Add WiFi::begin(ssid, pass, bssid)
* WiFiMulti to use BSSID, make more robust
WiFiMulti will now be more aggressive and try all matching SSIDs, in order
of RSSI, using the BSSID to identify individual APs in a mesh.
Before, if the highest RSSI AP didn't connect, it would fail immediately.
Now, it will go down the list, ordered by RSSI, to attempt to get a link.
* Add Bluetooth support from Pico-SDK
Able to build and run the HID Keyboard Demo from the Arduino IDE, almost
as-is.
Will probably need to make BT configurable. Enabling BT on a plain WiFi
sketch uses 50KB of flash and 16KB of RAM even if no BT is used.
* Separate picow libs, BT through menus, example
Build normal Pico.a and 4 different options for PicoW IP/BT configuration.
Use IP=>IP/Bluetooth menu to select between options.
* CMakefile rationalization
* Move BT TLV(pairing) out of last 2 flash sectors
The pairing keys for BT are stored at the end of flash by default, but
we use the last sector of flash for EPROM and the penultimate one for
the filesystem. Overwriting those in BT could cause some real exciting
crashes down the line.
Move the store to an app-build specific address using a dummy const
array to allocate space in the application image itself.
* PicoBluetoothHID with BT Mouse, Joystick, Keyboard
Add simple Bluetooth Classic HID helper function and port the existing
USB HID devices to it. Port their examples.
* Protect BT key storage from multicore
* Add short-n-sweet Bluetooth documents
* Add Bluetooth Serial port library
* Turn off BT when the BT libraries exit
Still need it on the command line in platform.txt for upload/etc., but make the
debug script in lib/picoprobe_cmsis_dap.tcl include the adapter speed setting.
Use GNU LD MAX() to ensure the uninitialized RAM portions are after the
OTA region. For most apps this already happens, so there will be no
overhead added.
The uninitted RAM macro would fail because the OTA code would clear out the
first 50KB or so of RAM to copy its code and global values.
Move all global values to the end of RAM in OTA, and then align any uninitted
vars to a 16KB unit to ensure it won't appear in the 1st part of RAM that is
still needed to copy the OTA executable.
In the normal case, without any uninit_ram vars, no additional space is used.
When uninit_ram is used, up to 16KB of space may be lost to get that alignment,
but it will work properly.
Fixes#1188
Instead of listing each board three times (normal upload, picoprobe,
and pico-debug uploads), list each board once and use a menu to select
the actual upload method.
Also add in picotool as an upload method for those folks who have trouble
with automounting.
Fix#1111
SingleFileDisk allows for exporting a file from the onboard LittleFS
filesystem to a PC through an emulated FAT drive when connected. The
PC can open and copy the file, as well as delete it, but the PC has no
access to the main onboard LittleFS and no actual on-flash FAT
structures are used.
This is handy for things like data loggers. They can run connected to
USB power for some time, and then connected to a PC to dowmload the CSV
log recorded.
It's almost 2023, allow LFN (long file names) on the emulated USB disk.
Reduce the disk buffer size to 64 bytes. The buffer is statically
allocated so it's always present, even in non-USB disk mode, meaning
all apps will pay the RAM price for it. 64 bytes is slower to read
but works and saves ~1/2KB of heap for all apps.
Random crashes, infinite loops, and other lockups were happening to the PicoW
while under high load from multiple clients.
This seems to have been due to two issues:
* The periodic sys_check_timeouts() call from an alarm/IRQ was happening while
the core was in LWIP code. LWIP is not re-entrant or multi-core/thread safe
so this is a bad thing. Some calls may not have been locked with a manual
addition of the LWIPMutex object to hit this.
* The WiFi driver supplies packet data during an interrupt. PBUF work is
legal in an interrupt, but actually calling netif->input() from an IRQ to
queue up the Ethernet packet for processing is illegal if LWIP is already
in progress.
Rearchitect the LWIP interface to fix these problems:
* Disable interrupts during malloc/etc. to avoid the possibility of the
periodic LWIP timeout check interrupting and potentially calling user
code which did a memory operation
* Wrap all used LWIP calls to note LWIP code will be executing, instead of
manually eyeballing and adding in protection in user code.
* Remove all user code LWIPMutex blocking, the wrapping takes care of it.
* When an Ethernet packet is received by interrupt and we're in LWIP code,
just throw the packet away for now. The upper layers can handle retransmit.
(A possible optimization would be to set the packet aside and add a
housekeeping portion to the LWIP wrappers to netif->input() them when safe).
* Ignore callbacks during TCP connection teardown when the ClientContext
passed from LWIP == nullptr
