Add ps_min_ modem feature
#103
Conversation
bjoernQ
commented
Dec 21, 2022
bjoernQ
commented
- on ESP32-C3 the example sometimes runs into a timeout - seems like it's more an issue in the socket util since at least there are data-packets received/sent
- not yet mentioning it in the README since it needs a bit more testing
- definitely the ping gets worse when using the power-safe mode (which is pretty much expected)
| @@ -474,6 +474,7 @@ impl<'s, 'n: 's> Socket<'s, 'n> { | |||
| .network_interface() | |||
| .get_socket_and_context::<TcpSocket>(self.socket_handle); | |||
| let remote_endpoint = (addr, port); | |||
| sock.set_ack_delay(Some(smoltcp::time::Duration::from_millis(100))); | |||
There was a problem hiding this comment.
Is this something that should only be used with the ps_min_modem feature, or just a general improvement?
There was a problem hiding this comment.
I think it shouldn't hurt in any case - the default 10ms are also somewhat optimistic but not really sure t.b.h
| * | ||
| ****************************************************************************/ | ||
| #[no_mangle] | ||
| pub unsafe extern "C" fn esp_timer_get_time() -> i64 { | ||
| trace!("esp_timer_get_time"); | ||
| (crate::timer::get_systimer_count() / crate::timer::TICKS_PER_SECOND / 1_000) as i64 |
There was a problem hiding this comment.
Is this correct by the way? Can't the "time since boot" actually get lower when the systimer wraps around?
Maybe we need a proper rolling timer instead (don't worry about it for this PR though)
There was a problem hiding this comment.
it's probably not really good this way - I think we have a few places where the wrapping isn't correctly handled - I should create an issue for that 🤔
There was a problem hiding this comment.
created issue #104 for that
I haven't measured anything t.b.h. |
