TIMER2_7_IRQ is shared among several timer sources, including
us_ticker. Raising TIMER2_7_IRQ pending bit will trigger the timer
interrupt, but the timer interrupt handler will not know which timer
source this interrupt is for. This patch sets timer load value to one
tick and force us_ticker to fire almost "immediately".
TIMER2_7_IRQ is handled through a common interrupt handler, and
is automatically cleared. Therefore, there is no need to clear IRQ.
The underlying timer HAL treats load value as micro-seconds and does
conversion internally. Therefore, simply pass micro-seconds to timer
HAL without converting to tick first.
Signed-off-by: Tony Wu <tung7970@gmail.com>
Remove redundant memory regions, and merge multiple RAM regions into one
to solve ielftool zero padding issue.
The side effect is less control over object files placement. It's all up
to linker's best effort.
Signed-off-by: Tony Wu <tonywu@realtek.com>
Fix ARMC6 guard typo introduced in commit 4f3f0cc9cc (Make Realtek link)
While at it, remove redundant ISR_STACK_SIZE assignment for ARMCC.
Signed-off-by: Tony Wu <tonywu@realtek.com>
commit f7bc126 (Re-work Rtl8195AM ticker) tried to address ticker drifting
issue, but in fact broke ticker functionality. This patch adjusts us_ticker
porting to unbreak it, and to fix the following tests:
mbed-os-tests-mbed_drivers-ticker
mbed-os-tests-mbed_drivers-timeout
mbed-os-tests-mbedmicro-rtos-mbed-isr
mbed-os-features-frameworks-utest-tests-unit_tests-minimal_async_scheduler
mbed-os-features-frameworks-utest-tests-unit_tests-case_control_async
mbed-os-features-frameworks-utest-tests-unit_tests-case_async_validate
Changes are:
1. Dont disable timer after every us_ticker interrupt. That basically
breaks ticker functionality.
2. Fine-tune us to tick conversion macro. Previous conversion method
yields 1 tick drift every 2 ms.
3. Remove special placement of ticker objects. No longer necessary.
Signed-off-by: Tony Wu <tonywu@realtek.com>
There was some interesting stuff here.
* ARMC6 treats assembly the same as GNU, but has a different startup
function, so we branch the asm there
* Incorrect discriminant, probably copied from other ports
IAR 8.x compiler throws error for emac_interface_t constructor
rtw_emac.cpp@220,0: [Pe1790]: the default constructor of "emac_interface"
cannot be referenced -- it is a deleted function
Error is because new object is created of structure with constant member.
Resolved it by using malloc instead new.
1. Rearrange SDRAM and SRAM layout. Move timing critical code to SRAM,
and the rest to SDRAM.
2. Add bootloader that's capable of FOTA over mbed cloud.
Signed-off-by: Tony Wu <tonywu@realtek.com>
The FlashIAP is supposed to have access to the full application
memory, so the address scope has been changed to encompass
the entire external flash.
The address translation has also been changed to reflect the memory
layout in the documentation:
Realtek Ameba1 Memory Layout: 5.1 Flash Memory
fire_interrupt function should be used for events in the past. As we have now
64bit timestamp, we can figure out what is in the past, and ask a target to invoke
an interrupt immediately. The previous attemps in the target HAL tickers were not ideal, as it can wrap around easily (16 or 32 bit counters). This new
functionality should solve this problem.
set_interrupt for tickers in HAL code should not handle anything but the next match interrupt. If it was in the past is handled by the upper layer.
It is possible that we are setting next event to the close future, so once it is set it is already in the past. Therefore we add a check after set interrupt to verify it is in future.
If it is not, we fire interrupt immediately. This results in
two events - first one immediate, correct one. The second one might be scheduled in far future (almost entire ticker range),
that should be discarded.
The specification for the fire_interrupts are:
- should set pending bit for the ticker interrupt (as soon as possible),
the event we are scheduling is already in the past, and we do not want to skip
any events
- no arguments are provided, neither return value, not needed
- ticker should be initialized prior calling this function (no need to check if it is already initialized)
All our targets provide this new functionality, removing old misleading if (timestamp is in the past) checks.
The Ameba RTC driver converts time_t to
second/minute/hour/day/month/year in rtc_write and back to time_t in
rtc_read. Replace this with an implementation which uses time_t
directly.
There is an easy default implementation of spi_master_block_write that
just calls spi_master_write in a loop, so the default implementation
of spi_master_block_write has been added to all targets.