This was used in the past because sdld was
very verbose when linking banked hex files. New
sdld versions do not exhibit this level of
verbosity and therefore the redirect can be
stopped
The CC2531 USB stick now identifies itself as a
'Texas Instruments CC2531 USB Dongle' and uses a
TI-assigmed VID:PID. The VID:PID is now configurable
in contiki- or project-conf.h
The sensinode platform does not support .upload and .serialdump
Their presence in the makefile has confused in the past confused
some users. This commit removes them
The commit also removes the $(OBJECTDIR)/%.rel: %.cS recipe which
is not used by either 8051 platform and is probably broken anyway,
since it has been unmaintained for years
Historically $(OBJECTDIR) was created when Makefile.include is read. A
consequence is that combining "clean" with "all" (or any other build
target) results in an error because the clean removes the object
directory that is required to exist when building dependencies.
Creating $(OBJECTDIR) on-demand ensures it is present when needed.
Removed creation of $(OBJECTDIR) on initial read, and added an order-only
dependency forcing its creation all Makefile* rules where the target is
explicitly or implicitly in $(OBJECTDIR).
The boot loader now knows when to go into bootstrap mode by
looking for a specific EEPROM value. Also updated code style
to match Contiki code style guidelines.
This patch removes a defunct EEPROM implementation from the native
platform and provides a new EEPROM implementation for the native cpu.
The previous implementation appears to be vestigal.
This is useful for testing code which uses the EEPROM without running
the code on the actual hardware.
By default the code will create a new temporary file as the EEPROM
backing, reinitializing each time. If you would like to preserve the
EEPROM contents or specify a specific EEPROM file to use, you can set the
`CONTIKI_EEPROM` environment variable to the name of the EEPROM file you
wish to use instead. If it already exists, its contents will be used.
If it does not already exist, it will be created and initialized by
filling it with `0xFF`---just like a real EEPROM.
A new example is also included, which was used to verify the correctness
of the implementation. It can easily be used to verify the EEPROM
implementations of other targets.
- For the CC2538, simplify handling of USB_CDC_ACM_LINE_STATE
events. Ignore the Carrier Control (RTS) bit when receiving
a SET_CONTROL_LINE _STATE request, we are a full duplex device.
- Improve behaviour of the CC2531 USB stick when there is no
process on the host to read IN data. Basically, we adopt the
CC2538 approach and we only send data when a DTE is present
This commit moves the Settings Manager from the AVR codebase
into the Contiki core library. Any platform that implements
the Contiki EEPROM API can now use the Settings Manager's
key-value store for storing their persistent configuration info.
The Settings Manager is a EEPROM-based key-value store. Keys
are 16-bit integers and values may be up to 16,383 bytes long.
It is intended to be used to store configuration-related information,
like network settings, radio channels, etc.
* Robust data format which requires no initialization.
* Supports multiple values with the same key.
* Data can be appended without erasing EEPROM.
* Max size of settings data can be easily increased in the future,
as long as it doesn't overlap with application data.
The format was inspired by the [OLPC manufacturing data format][].
Since the beginning of EEPROM often contains application-specific
information, the best place to store settings is at the end of EEPROM
(the "top"). Because we are starting at the end of EEPROM, it makes
sense to grow the list of key-value pairs downward, toward the start of
EEPROM.
Each key-value pair is stored in memory in the following format:
Order | Size | Name | Description
--------:|---------:|--------------|-------------------------------
0 | 2 | `key` | 16-bit key
-2 | 1 | `size_check` | One's-complement of next byte
-3 | 1 or 2 | `size` | The size of `value`, in bytes
-4 or -5 | variable | `value` | Value associated with `key`
The end of the key-value pairs is denoted by the first invalid entry.
An invalid entry has any of the following attributes:
* The `size_check` byte doesn't match the one's compliment of the
`size` byte (or `size_low` byte).
* The key has a value of 0x0000.
[OLPC manufacturing data format]: http://wiki.laptop.org/go/Manufacturing_data
This is a general cleanup of things like code style issues and code structure of the STM32w port to make it more like the rest of Contiki is structured.
This reverts commit 029bc0ee27, reversing
changes made to a7b3e99644.
This uses LGPL libopencm3. While the patch doesn't include the code,
the resulting binary would force the release of all code as LGPL.
This magic comes from the `--gc-sections` linker flag, which turns on garbage collection for unused input sections. The compiler flags `-ffunction-sections` and `-fdata-sections` make sure that each function and each static data definition have their own section. The result is that GCC can prune away all unused symbols, reducing the size of the resulting executable.
These optimizations may be disabled by setting the Makefile variable
`SMALL` to zero.
Relevant cc65 changes...
General:
- The compiler generates "extended" dependency info (like gcc) so there's no need for postprocessing whatsoever :-)
- The linker is very pernickety regarding the ordering of cmdline options so a custom linker rule is necessary :-(
Apple2:
- The various memory usage scenarios aren't specified anymore via separate linker configs but via defines overriding default values in the builtin linker config.
Atari:
- The builtin linker config allows to override the start addr so there no more need for a custom linker config.
- The C library comes with POSIX directory access. So there's no more need for for a custom coding.
CBM:
- The C library comes with POSIX directory access. So there's no more need for for a custom coding.
This currently supports the stm32f107, but support will be added for
the F105 shortly (missing a linker script and I can't properly test
without a devkit).
This is a major change to how the main tick interrupt is handled on
the mc1322x platforms. Instead of using two timer resources, TMR0 and
RTC, this patch unifies all the timers to use the RTC. This is enabled by
implementing etimers as scheduled rtimers. The main advantage (aside
from freeing TMR0 for general use) is have the Contiki timebase come
from the same source that will be used for sleeping and wakeup.
This patch enables automatic route setup and cleanup when
starting and stopping the minimal-net target on OS X.
Both IPv4 and IPv6 are supported.
Using the minimal-net target on OS X was absolute hell
before I came up with this patch. Now it is painless.
The minimal-net target, as currently written, wakes up the
CPU every millisecond to check for packets, and will only
react in real-time to input from stdin. If you are running
this on a laptop battery, your battery will quickly drain.
This change allows the CPU to idle when there is literally
nothing to do while still being responsive to input from
stein and/or incoming packets. This fix should significantly
improve performance while significantly improving power
usage. Win-win.
Also added `_xassert()` implementation so that the contiki-
provided `assert()` macro will work properly when used
on this platform.
by now it would have worked if you set CC via
command line (e.g. `make CC=clang`, though it
wouldn't work when CC is set in the platform
makefile which includes `Makefile.native`.
configuration system.
(also deprecate TARGET=redbee-econotag)
- mc13224v now automatically probes hardware config for buck converter
and 32kHz crystal as well as automatically monitors battery voltage
and manages the buck accordingly.
- new flashed based config system for mc13224v parameters such has
radio modes (demod, autoack), nvmtype, mac address, channel and
power.
- considerably cleaned up econotag platform code (suffered from severe
case of bit-rot)
This is mainly a naming convention thing, we want to have 'isr'
as part of the name, instead of 'int'. We also want port_2 instead
of p2 because we already had port_1
See Pull Request #18
The P2 Interrupt is shared across many periferal (I2C, USB, GPIO).
This adds a generic interrupt handler on which the differents drivers
can register a handler.
See Pull Request #18
Change to have a real usb VID/PID and better fit the capabilities
of the CC2531 hardware (enpoint size, location).
Compile only if the cdc-acm class is requested.
See Pull Request #18
- Moved to their own file
(so we can later copy the entire thing over to cc2430)
- Renamed the functions
(for naming convention reasons)
- The entire thing can be enabled/disabled
- Added a couple more macros
- Hooked into main()
__bit variables must be located after the 0x20 address
This force the stack to start after 0x20, thus decrease
the stack size by a considerable amount.
This add -fomit-frame-pointer by default, it decrease sliglty the
stack usage when several function call are nested.
On some specific case 30 bytes could be saved on the stack.
This allows us to reduce CODE footprint of SDCC projects
built with --model-huge. Use carefully!
* Added a facility which allows us to enable/disable the
feature from the CPU dir (CC_CONF_NON_BANKED_OPTIMIZATION)
* Added the CC_NON_BANKED keyword to some platform files
(expands to __nonbanked)
* Started using this for some examples
We still use the primary location by default (Info Page) but
this is now configurable. This change is useful for users who
wish to specify their own MAC address. Since the Info Page is
read-only, they need to be able to use the secondary location
This will help us slowly add better support for smaller SoCs
instead of just assuming that all cc253x SoCs are F256
- We build for F256 by default but the project Makefile can override this
- We currently ignore this when passing --code-size. This is a ToDo
- The bank allocator always assumes 7 banks. Once we fix --code-size above this will be irrelevant
This new approach (CLOCK_CONF_ACCURATE=0) was not as inaccurate
as originally thought. In fact, it has pretty much the same
accuracy as the old, stack-hungry version.
* Renamed the define from CLOCK_CONF_ACCURATE (1: old) to
CLOCK_CONF_STACK_FRIENDLY (1: new) to stop implying
that one is more accurate than the other.
* Using CLOCK_CONF_STACK_FRIENDLY by default.
* Bit-Addressable SFRs are now accessed as such,
instead of (N)OR-ing the byte
* A routine was declared as CCIF but not defined as such. Fixed
* Deleted a leftover duplicate define
* Formatting
* Comment updates and clarifications
Since the introduction of a purpose-written routine, the old
generic one (which was written in asm) was never being used.
There was a configuration directive which allowed us to switch
between the two. All references to this configuration directive
have also been removed.
