boards/nucleo-l152: configure LSI by default #8545
Conversation
kYc0o
added
Type: bug
kYc0o
added
the
State: waiting for other PR
aabadie
removed
the
State: waiting for other PR
|
I think your problem is not related to the LSE in itself, but maybe something else. This PR makes LSI the default Low Speed clock source for the nucleo-l152 under any revision. If for some reason LSE breaks the board it needs to be investigated. The real problem in #8024 was what it was fixed in #8518, this is only complementary to make work the nucleo-l152 by default (and not related to the stm32l152 CPU, which was affected by #7687). I can't provide a fix for LSE (I don't even know what's the problem) since I don't have a board with it. |
| * 1: external crystal available (always 32.768kHz) | ||
| * | ||
| * LSE might not be available by default in early (C-01) Nucleo boards. | ||
| * If you're sure it is present, define CLOCK_LSE=1 in your project |
There was a problem hiding this comment.
If someone with an C-03 board revision follows what is described here, it will result in a failing board (e.g. no stdio for example).
I suggest removing this line until we have a solution with LSE.
There was a problem hiding this comment.
Hmm... I'm still not completely convinced by this. I think that's obvious that if your board is not working after you set CLOCK_LSE=1 the reason is obviously that LSE is not working properly, so this message still make sense.
Besides, the fact that this board is broken with CLOCK_LSE is a bug, which should be documented somewhere e.g. an issue. Afterwards, of course it needs to be fixed, and it's out of the scope of this PR.
There was a problem hiding this comment.
ok, but this is not what this comment is saying: 'If you're sure it is present' doesn't mean LSE might be broken.
There was a problem hiding this comment.
Ok I can add such information.
There was a problem hiding this comment.
Can you open an issue for this BTW?
|
BTW, with the current change nucleo-l152 works like a charm. |
kYc0o
force-pushed
the
nucleo-l152_lse
branch
from
f35edca to
7eafff1
Compare
|
Changed issue description and directly amended. |
| * | ||
| * LSE might not be available by default in early (C-01) Nucleo boards. | ||
| * For newer revisions, LSE crystal is present, but currently is not working. | ||
| * (issue #8545). |
There was a problem hiding this comment.
@kYc0o, use the full url, e.g. https://github.com/RIOT-OS/RIOT/pull/8545
There was a problem hiding this comment.
It's evident that the comment it's in the context of a github issue. Though I've put the complete address and amended directly the change.
Contribution description
Some nucleo-l152 boards, namely revision C-01, don't have a LSE to rely on for low speed clocking, which is mostly used for RTC.
This PR makes it optional, and define LSI by default.
Issues/PRs references
Fixes #8024 and fixes #8240, depends on
#8518to fix everything.