Final chunk of register definitions to be pulled up.
Now the "target" files are _only_ defining the list of u(s)arts
available, and any _specific_ functions for that target.
Just small pieces at a time to make it easy to see what's happening.
Taking definitions currently implemented in both f0/f3 headers and
making combined, documented versions in the -v2 header.
Use the usart-common base plus the usart-v2 code, instead of private
implementations. Less code, more common apis across targets.
Of note is the trick to make F0 look like it has an APB2 bus. It's the
only stm32 that doesn't have a documented APB2 bus, but still has
peripherals enabled via an "APB2" register, and they match how other
targets have an APB2. Simply make APB2 an alias of APB1, as it's only
used for clock speed detection.
Instead of declaring that _every_ device has USART1,2,3 and UART4,5, let
the targets themselves define what periphs they do, along with their
USARTx_BASE defines, and let the common headers just have the common
abstractions.
This function was using apb1 for quite a few families instead of apb2.
This only mattered for L1 and F3, and for USART1/USART6, and only if
apb1 speed != apb2 speed.
Instead of using families explicitly, just check for the peripherals
themselves. On F0,F1,F2,F3,F4,F7,H7,L0,L1,L4, usart1/6 are _always_ in
the rcc_apb2 register and the other uarts are all on apb1.
(F0 doesn't actually _have_ apb2, but it's still called the apb2
register)
Similar to how we have abstract defines for the stop bits, parity and
flow control common mode namees, provide abstract flag names for the
"standard" flags. This allows us to start using common API code for v1
and v2 uarts
For stm32f3, drop the "compatibility" defines that simply pollute the
namespace, making it confusingly appear as if f3 has both SR and ISR
registers.
Only pulling out the _common_ stuff. This is a single step in a long
process of eliminating all the duplication and "same, but different"
implementations that are stalling adding nice clean easy support for
l0/f7/l4.
This _ignores_ all currently conflict register definitions, even if they
"do the same thing" it just pulls up the common stuff. A subsequent
commit can look at resolving the implementatations to a single version.
Even if they haven't been referenced from a .c file. Some peripherals
start their life as register definitions only, and they should still
have the documentation generated.
There _will_ be overlap in the generated lists, but doxygen doesn't seem
to mind this.
The ADVANCED_TIMERS define/check was added in 523943a as part of adding L1
support. The runtime checks against TIM1/TIM8 already existed. Since L1
doesn't have TIM1/TIM8, those names are undefined, resulting in a compilation
error until ifdeffed out.
Since I throw out all TIM1/TIM8 checks, there's no references to those names
left, thus no need to keep the ifdef either.
As for the registers themselves, l1/timer.h pulls in common/timer_common_all.h
which defines macros for the superset of all timers, so e.g. TIM_BDTR() is
still available regardless of whether or not the particular chip we're building
for has any timers with a BDTR register.
Instead of expecting a posix-ish "find" in the shell, simply leverage
our knowledge of our source structure to make a single list using make's
builtin wildcard() functionality.
Fixes https://github.com/libopencm3/libopencm3/issues/828
Instead of the fragile and error prone attempts to specifically
include/exclude files from doxygen by name and pattern, simply use the
already generated .d files to provide accurate and up to date lists of
all source files used.
Pros:
* Nothing left to worry about
* Much more encouraging to actually _work_ on the documentation now that
you can be sure the right docs will be generated instead of a confusing
mix.
Downsides/Upsides:
* Automatically includes all CM3/USB in each device's page _as well_ now
Downsides:
* lpc43xx still manually listed. However, completely contained in it's
own dir, so no problems
* No attempt to carry this in latex. easy, but more tempted to drop
latex support outright. (I don't think the generation there has even
worked for a while now)
* Due to the mismatch between lib directories and document roots, the
sourcelist can't be magically created per directory. There has to be
some sort of mapping between the two, so as this is doc generation only,
a static list seems sane for maintennance. (Especially compared to the
old method)
* Source list generation probably doesn't work on windows.
Very new gcc versions add a warning on switch cases that fall through.
While there's an option that accepts any comment as explaining it, it's
easier in our case to just use one of the "blessed" comments. We can't
use the [[]] attributes for standards code, and the gcc specific
attributes are worse than the comments. This has no functional change
whatsoever.
D+ is PA12 not PA11. The reason this worked before, is because the line
before made PA12 output, and without setting the GPIO_ODR register
_before_ hand, this meant as soon as it was switched to output, it
received the reset value of GPIO_ODR for PA12, ie, 0. (Effectively
doing a "free" gpio_clear(GPIOA, GPIO12)
Because GPIO11 wasn't configured to be an output, the confusing
gpio_clear(GPIOA, GPIO11) was simply configuring the pullup/down value
of the input, which was still ignored, as it was (out of reset) in input
floating mode.
Reviewed-by: Karl Palsson <karlp@tweak.net.au>
Clearly staging branch testers weren't testing.
Also, clearly the separation of RST_ bits and RCC_ bits is a complicated
annoyance.
Fixes: d9615a2eb7 update to modern include apis
Added the CAN1 compatibility aliases as has been done for adc and dac to
make code reuse easier. Only for the magic enums, the raw bit
definitions remain as per the ref mans
Originally suggested as https://github.com/libopencm3/libopencm3/pull/802
Karl Palsson