* Add a FAT filesystem for SD cards to Arduino FS
Arduino forked a copy of SD lib several years ago, put their own wrapper
around it, and it's been languishing in our ESP8266 libraries ever since
as SD. It doesn't support long file names, has class names which
conflict with the ESP8266 internal names, and hasn't been updated in
ages.
The original author of the SD library has continued work in the
meantime, and produced a very feature rich implementation of SdFat. It
unfortunately also conflicts with the class names we use in ESP8266
Arduino and has a different API than the internal SPIFFS or proposed
LittleFS filesystem objects.
This PR puts a wrapper around the latest and greatest SdFat library,
by forking it and wrapping its classes in a private namespace "sdfat,"
and making as thin a wrapper as possible around it to conform to
the ESP8266 FS, File, and Dir classes.
This PR also removes the Arduino SD.h class library and rewrites it
using the new SDFS filesystem to make everything in the ESP8266
Arduino core compatible with each other.
By doing so it lets us use a single interface for anything needing a
file instead of multiple ones (see SDWebServer and how a different
object is needed vs. one serving from SPIFFS even though the logic is
all the same). Same for BearSSL's CertStores and probably a few others
I've missed, cleaning up our code base significantly.
Like LittleFS, silently create directories when a file is created with
a subdirectory specifier ("/path/to/file.txt") if they do not yet exist.
Adds a blacklist of sketches to skip in the CI process (because SdFat
has many examples which do not build properly on the ESP8266).
Now that LittleFS and SDFS have directory support, the FS needs to be
able to communicate whether a name is one or the other. Add a simple
bool FS::isDirectory() and bool FS::isFile() method. SPIFFS doesn't
have directories, so if it's valid it's a file and reported as such.
Add ::mkdir/::rmdir to the FS class to allow users to make and destroy
subdirectories. SPIFFS directory operations will, of course, fail
and return false.
Emulate a 16MB SD card and allow test runner to exercise it by using
a custom SdFat HOST_MOCK-enabled object.
Throw out the original Arduino SD.h class and rewrite from scratch using
only the ESP8266 native SDFS calls. This makes "SD" based applications
compatible with normal ESP8266 "File" and "FS" and "SPIFFS" operations.
The only major visible change for users is that long filenames now are
fully supported and work without any code changes. If there are static
arrays of 11 bytes for old 8.3 names in code, they will need to be
adjusted.
While it is recommended to use the more powerful SDFS class to access SD
cards, this SD.h wrapper allows for use of existing Arduino libraries
which are built to only with with that SD class.
Additional helper functions added to ESP8266 native Filesystem:: classes
to help support this portability.
The rewrite is good enough to run the original SDWebServer and SD
example code without any changes.
* Add a FSConfig and SDFSConfig param to FS.begin()
Allows for configuration values to be passed into a filesystem via the
begin method. By default, a FS will receive a nullptr and should so
whatever is appropriate.
The base FSConfig class has one parameter, _autoFormat, set by the
default constructor to true.
For SPIFFS, you can now disable auto formatting on mount failure by
passing in a FSConfig(false) object.
For SDFS a SDFSConfig parameter can be passed into config specifying the
chip select and SPI configuration. If nothing is passed in, the begin
will fail since there are no safe default values here.
* Add FS::setConfig to set FS-specific options
Add a new call, FS::setConfig(const {SDFS,SPIFFS}Config *cfg), which
takes a FS-specific configuration object and copies any special settings
on a per-FS basis. The call is only valid on unmounted filesystems, and
checks the type of object passed in matches the FS being configured.
Updates the docs and tests to utilize this new configuration method.
* Add ::truncate to File interface
Fixes#3846
* Use polledTimeout for formatting yields, cleanup
Use the new polledTimeout class to ensure a yield every 5ms while
formatting.
Add in default case handling and some debug messages when invalid inputs
specified.
* Make setConfig take const& ref, cleaner code
setConfig now can take a parameter defined directly in the call by using
a const &ref to it, leading to one less line of code to write and
cleaner reading of the code.
Also clean up SDFS implementation pointer definition.
emulation with socket: fix nasty delay
properly check all return values (read() returning 0 indicates closed peer)
ClientContext::state() triggers read() to check for closed peer
uart emulation on tty: add cmdline -b debug option to disable raw tty
hide annoying MOCK messages, add cmdline -v option to show them
uart emulation on tty: check more return values
emulation on host: tcp/udp port shifting: new cmdline -s(port Shift) option
GCC reports the following warning about ignoring the return value of
write() (to the console/UART in this instance).
common/MockUART.cpp: In function ‘void uart_do_write_char(int, char)’:
common/MockUART.cpp:67:8: warning: ignoring return value of ‘ssize_t write(int, const void*, size_t)’, declared with attribute warn_unused_result [-Wunused-result]
write(uart_nr + 1, &c, 1);
~~~~~^~~~~~~~~~~~~~~~~~~~
Add a simple warning printout to STDERR which may not be the same as the
UART stream and could succeed.
* emulation on host: Add full UART driver emulation.
This PR replaces the high level Serial mock with a more complete
UART driver. This way the HardwareSerial works without any
modifications. Additionally the driver supports UART0 and UART1
at the same time (UART0 is directed to stdout and UART1 writes
to stderr). RX is implemented by switching the terminal into raw
non-blocking mode and injecting each key-press directly into the
FIFO of UART0. A new command line switch -c was added to ignore
CTRL-C and send it via serial as well. The decumentation was
updated accordingly.
Reading and setting of GPIOs does only write to stderr, when compiled
with D=1. But this is subject to be replaced with a proper GPIO emu
anyway. Reading from GPIO0 now returns 1 instead of 0 because this is
most likely a low active input.
* Fixed unused variable.
* Remove unused functions, as long as there are no debug macros using them.
* Move user_interface.cc from MOCK_CPP_FILES_EMU to MOCK_CPP_FILES.
* Move optimistic_yield() from user_interface.cpp to Arduino.cpp
* Remove atexit() weak declaration (fixes segfault when calling atexit).
* Improve resetting of terminal after program exit, convert \r to \r\n.
* Show error, when STDOUT is not a TTY, minor code cleanung.
This commit allows switching SDK firmware:
nonos-sdk-pre-v3 shipped with release 2.5.0 has issues:
* Some boards show erratic behavior (radio connection is quickly lost), with an unknown cause.
These boards work well with previous nonos-sdk-2.2.1 firmware (#5736)
* Overall performances seem to have decreased (#5513)
This PR restores sdk2.2.1 (as in core-2.4.2).
SDK-pre-3.0 - which has brought long awaited fixes (WiFi sleep modes) - is still available through a menu option available only with generic board.
BREAKING
* new define `-DNONOSDK221=1` or `-DNONOSDK3V0=1`
* for external build systems: new library directory: `tools/sdk/lib/<version>/lib`
* PIO: variable `PIO_FRAMEWORK_ARDUINO_ESPRESSIF_SDK3` is needed for sdk-pre-v3.
Fix#5736
This patch splits ARDUINO_LIBS into two variables. The second variable is called OPT_ARDUINO_LIBS and can be overridden by the caller as it uses the ?= assignment operator. Additionally it unifies and simplifies collecting the files in common/ by using the addprefix macro. All changes should be 100% backwards compatible.
Reduce String memory overhead from 24 bytes to 16 bytes by limiting the
maximum string length to <64Kbytes (which is larger than heap so no
effective problem).
Add Small String Optimization, SSO, which instead of allocating pointers
to small strings on the heap will store the string in place of the
pointer in the class. This should reduce memory fragmentation as
Save up to 12 chars (11 + \0) in String itself by using the terminating
\0 in the inline string as a flag to identify if this is a SSO or a heap
string.
Add a host test that verifies that no memory is allocated until a
full 11 characters are assigned to a string, as well as checking all
intermediate values.
No user code changes should be required to work with this optimization.
* Remove broken ltoa/ultoa, call itoa/utoa
Use the newlib integer-to-ASCII non-POSIX calls instead of rolling
our own. Should be safe as sizeof(long) == sizeof(int).
The custom functions behaved differently from itoa when passed in
negative values in non-base-10.
Add host tests for negative non-base-10 int/longs
* ci: fix checking for presence of CI_GITHUB_API_KEY
CI_GITHUB_API_KEY is always set when evaluating job include/exclude conditions, and it only gets removed when the job starts. Move the check into the script.
* ci: split build jobs into separate shell scripts
* uart fixes and BW improvements
* uart: read_char straightly use hw buffer
* +attributes for functions called by ISR
* uart: BW improvements
read_char straightly use hw buffer (+ ~10%bw)
read by block (+ ~190%bw) (instead of generic Stream::readBytes)
attributes for functions called by ISR
remove overrun message
remove some ISR flags which were not honoured
* fix merge
* fix buffer overflow
* serial stress test sketch
* astyle
* serial stress example: interactive keyboard, stop reading, overrun
* serial device test: bandwidth & overrun
* update + HardwareSerial::hasError()
* interactive overrun in example
* astyle
* Test using @plerup's SoftwareSerial as submodule (tag 3.4.1)
* update upstream ref (fix warning)
* host mock uart/read(buf,size)
* reset style changes in submodules before style diff
* update build_boards_manager_package.sh for submodules
* trigger CI (removing space)
* cannot reproduce locally the CI issue, setting bash -x option to get live trace
* remove previously added (in this PR) 'set -e' in package builder (passes local tests, not real CI)
script-comment new recipe.hooks.core.prebuild.3 (along with already commented .1 and .2)
moved CI package test to be first on the test list
remove 'set -x', wish me luck
* lwip2: better handling of ipv4_addr/t type + 3 sntp servers
* bump lwip2 version
* Only with FEATURES=1: 3 sntp servers and AutoIP enabled (169.254 when dhcp server fails)
* Only with FEATURES=1: 3 sntp servers and AutoIP enabled (169.254 when dhcp server fails)
* local CI runner: select build type
* new ipv4_addr/t definition makes things easier for IPAddress
* update local CI runner
* lwip2 changes
* lwip2: port esp-ping and espconn
* Move to PROGMEM aware libc, allow PSTR in printf()
A Newlib (libc) patch is in progress to move the _P functions from inside
Arduino into first-class citizens in libc. This Arduino patch cleans up
code that's been migrated there. Binaries for the new libs are included
because it seems they're part of the Arduino git tree, and should be
replaced with @igrr built ones when/if the Newlib changes are accepted.
Notable changes/additions for Arduino:
Allow for use of PROGMEM based format and parameter strings in all
*printf functions. No need for copying PSTR()s into RAM before printing
them out (transparently saves heap space when using _P functions) and
makes it easier to print out constant strings for applications.
Add "%S" (capital-S) format that I've been told, but cannot verify,
is used in Arduino to specify a PROGMEM string parameter in printfs,
as an alias for "%s" since plain "%s" can now handle PROGMEM.
Optimized the memcpy_P, strnlen_P, and strncpy_P functions to use 32-bit
direct reads whenver possible (source and dest alignment mediated), but
there is still room for improvement in others.
Finally, move several constant arrays from RODATA into PROGMEM and
update their accessors. Among these are the ctype array, ~260 bytes,
mprec* arrays, ~300 bytes, and strings/daycounts in the time
formatting functions, ~200 bytes. All told, sketches will see from
300 to 800 additional RAM heap free on startup (depending on their
use of these routines).
* Fix merge error in #ifdef/#endif
* Fix host test using the newlib generic pgmspace.h
Host tests now use the sys/pgmspace.h for compiles instead of the
ESP8266-specific version.
* Update with rebuilt libraries using latest newlib
* Include binaries built directly from @igrr repo
Rebuild the binaries using a git clone of
https://github.com/igrr/newlib-xtensa
Build commands for posterity:
````
rm -rf ./xtensa-lx106-elf/
./configure --prefix=<DIR>/esp8266/tools/sdk/libc --with-newlib \
--enable-multilib --disable-newlib-io-c99-formats \
--disable-newlib-supplied-syscalls \
--enable-newlib-nano-formatted-io --enable-newlib-reent-small \
--enable-target-optspace \
--program-transform-name="s&^&xtensa-lx106-elf-&" \
--disable-option-checking --with-target-subdir=xtensa-lx106-elf \
--target=xtensa-lx106-elf
rm -f etc/config.cache
CROSS_CFLAGS="-fno-omit-frame-pointer -DSIGNAL_PROVIDED -DABORT_PROVIDED"\
" -DMALLOC_PROVIDED" \
PATH=<DIR>/esp8266/tools/xtensa-lx106-elf/bin/:$PATH \
make all install
````
* Fix merge define conflict in c_types.h
* Fix strlen_P misaligned source error
Include fix from newlib-xtensa/fix-strlen branch cleaning up misaligned
access on a non-aligned source string.
* Fix strlen_P and strcpy_P edge cases
Ran the included test suite on ESP8266 tstring.c with the following defines:
#define MAX_1 50
#define memcmp memcmp_P
#define memcpy memcpy_P
#define memmem memmem_P
#define memchr memchr_P
#define strcat strcat_P
#define strncat strncat_P
#define strcpy strcpy_P
#define strlen strlen_P
#define strnlen strnlen_P
#define strcmp strcmp_P
#define strncmp strncmp_P
Uncovered edge case and return value problems in the optimized versions of
the strnlen_P and strncpy_P functions. Corrected.
* Fix memcpy_P return value
memcpy-1.c test suite showed error in return value of memcpy_P. Correct it.
* Fix strnlen_P/strlen_P off-by-4 error
Random crashes, often on String constructors using a PSTR, would occur due
to the accelerated strnlen_P going past the end of the string. Would make
debug builds fail, too (ESP.getVersionString() failure).
Fix to fall through to normal copy on a word that's got a 0 byte anywhere
in it.
* Add device tests for libc functional verification
Add test suite used to debug libc optimized _P functions to the device
tests.
* Rebuild from igrr's repo (same source as prior)
Rebuild .a from igrr's repo at 347260af117b4177389e69fd4d04169b11d87a97
* WIP - add exceptions
* Fix exception to have 0-terminator
* Move some exception constants to TEXT from RODATA
* Remove throw stubs
* Move more exception stuff to ROM
* Enable exceptions in platform.io
* Remove atexit, is duplicated in rebuilt lib
Need to look at the quick-toolchain options, there seems to be a definition
for atexit defined there (libgcc?) that needs to be excised. For now,
remove our local do-nothing copy.
* Update libgcc to remove soft-fp functions
The esp-quick-toolchain generated libgcc.a needed to have the soft-FP routines
that are in ROM removed from it. Remove them in the new esp-quick-toolchain
and update.
* Fix merge typos in Makefile
* Add unhandled exception handler to postmortem
* Return our atexit() handler
* Latest stdc++, minimize exception emercengy area
* Remove atexit from newlib
atexit was defined in newlib strongly, but we also define a noop atexit in core.
Since we never exit, use the core's noop and delete the atexit from libc.a
Updated in esp-quick-toolchain as well.
* Move __FUNCTION__ static strings to PROGMEM
__FUNCTION__ is unlikely to be a timing sensitive variable, so move it to
PROGMEM and not RODATA (RAM) using linker magic.
asserts() now should take no RAM for any strings.
* Clean up linker file, update to latest stdc++
* Update to latest stdc++ which doesn't call strerror
* Update to GCC5.1 exception emergency allocator
Using GCC 5.1's emergency memory allocator for exceptions, much less
space is required in programs which do not use exceptions and when
space is allocated it is managed more efficiently.
* Initial try with new compiler toolchain
* Include newlib built from esp-quick-toolchain
* Update JSON with all new esp-quick-toolchain builds
* Use 64bit Windows compiler on 64bit Windows
* Dump std::exception.what() when possible
When doing the panic on unhandled exceptions, try and grab the
.what() pointer and dump it as part of the termination info.
Makes it easy to see mem errors (std::bad_alloc) or std::runtime_error
strings.
* Use scripted install from esp-quick-toolchain
Makes sure proper libraries and includes are present by using a
scripted installation from esp-quick-install instead of a manual
one.
* Update eqk to remove atexit, fix packaging diff
PIO is taking 27+ minutes now, very close to the runtime limit.
Split into even and odd, just like standard Arduino. Each PIO job
takes ~14-15 minutes and can go in parallel.
* PolledTimeout Class for wrapping millis() loops
* Add yield policies, improve reset, add host tests
* Fix copyright, comments
* adjust host tests for better time precision
* add fuzzyness to timing tests for CI jitter
* add blink example with polledTimeout
* improve namespace and type naming, add copyright, comments
* fix astyle
* WIP compile examples on host with 'make examples'
* WIP bufferize tcp input
* WIP Makefile
* WIP network to rework, tcp/udp to factorize, udp addresses broken
* minor changes to the core
* WIP basic udp working
* WIP mdns
* WIP mcast receiving, not sending
* WIP mdns OK
* beta version
* SSL + doc
* update travis host test command
* licenses
* typo
* doc: arduino builder is not around: declare functions before calling them
* fix with latest SSL PR, compile in 32 bits mode
* fix make clean
* make -m32 optional
* 32bits compiler ability tester
* WIP
* WIP (fix 1 vtable error, still another one to hunt with using spiffs)
* example astyle
* fix os_printf_plus
* load / save mock spiffs
* fix style
* fix using spiffs/mock
* don't mess ram
* update doc
* remove leftover
* optimization -Os except for CI, rename ARCH32 to FORCE32
* revert useless cast (not even compiled)
* remove unused function
* use proper type for pointer arithmetics
* makefile: sketch object and cpp file moved to bin/ directories
easier to clean, and IDE don't like them
* changes for review
* make use of %zd
* less verbose makefile by default (option)
* update readme
* Update to BearSSL 0.6+ release, add AES_CCM modes
Pull in latest BearSSL head (0.6 + minor additions) release and add AES_CCM
modes to the encryption options.
* Enable the aes_ccm initialization in client/server
* Initial attempt
* Working code with second stack thunking
* Remove #ifdefs in .S file, not needed.
* Clean up thunks and remove separate stack flag
* Fix PIO assembler errors
* Remove #ifdef code changes, ensure same code as PC
Remove "#ifdef ESP8266;...;#else;...;#endif" brackets in BearSSL to
ensure the host-tested code is the same as the ESP8266-run code.
* Move to latest BearSSL w/EC progmem savings
* Merge with master
* Add br_thunk_* calls to do ref counting, painting
Add reference counting br_thunk_add/del_ref() to replace stack handling code
in the class.
Add in stack painting and max usage calculation.
* Add in postmortem stack dump hooks
When a crash occurs while in the second stack, dump the BSSL stack and
then also the stack that it was called from (either cont or sys).
* Update stack dump to match decoder expectations
* Move thunk to code core for linkiage
The thunk code needs to be visible to the core routines, so move it to the
cores/esp8266 directory. Probably need to refactor the stack setup and the
bearssl portion to avoid dependency on bearssl libs in cores/esp8266
* Add 2nd stack dump utility routine
* Refactor once more, update stack size, add stress
Make stack_thunks generic, remove bearssl include inside of cores/esp8266.
Allocate the stack on a WiFiServerSecure object creation to avoid
fragmentation since we will need to allocate the stack to do any
connected work, anyway.
A stress test is now included which checks the total BearSSL second
stack usage for a variety of TLS handshake and certificate options
from badssl.org.
* Update to latest to-thunks branch
* Add BearSSL device test using stack stress
Run a series of SSL connection and transmission tests that stress
BearSSL and its stack usage to the device tests.
Modify device tests to include a possible SPIFFS generation and
upload when a make_spiffs.py file is present in a test directory.
* Use bearssl/master branch, not /to-thunks branch
Update to use the merged master branch of bearssl. Should have no code
changes.
* update HTTPClient API usage
skip the second POST as end() has different semantics and nulls the client pointer
use bearssl in ssl tests
add delay in python side when shutting down http web server so MacOS does not complain about address already in use
* fix crash if GET/POST was called after end() without a new begin()
update double POST test to insure no crash if POST called after end()
test now are for both AxTLS and BearSSL
* fix small comment typo
* Add full debug builds for all tests
In order to ensure the -DEBUG builds don't fail with checked-in code, add
two jobs which build even/odd with full debug flags to Travis.
Fixes#5143
* Add in dump of platform.txt to verify build options
* Fix BASH comparison of strings
* Fix debug-mode compile error in ESP8266AVRISP
* Add -Wextra to CI (matches IDE "All" warnings)
boards generator updates:
* simplified mapping description: only flash and spiffs sizes are needed
* some ldscripts are renamed due to autogenerated names
* +2M/0, +2M/512K spiffs, +4M0
* reduce length of hidden strings in boards.txt (#5100, https://github.com/arduino/arduino-builder/issues/284)
* give more details in ldscripts (address, size, +rf-cal, +sdk-wifi-settings)
* add stubs for more exception throw calls
Fixes https://github.com/esp8266/Arduino/issues/3358
* libc: make putc_r implementation weak
newlib provides its own implementation of _putc_r, which will call
_write_r (possibly after buffering). Make our implementation weak to
allow using the one from newlib.
Fixes https://github.com/esp8266/Arduino/issues/4630
* libc: fix incorrect return value of _write_r call
Should return number of bytes written, actually returned zero. This
resulted in std::cout going into failed state after the first write.
* tests: add test for output to std::cout
