From 6f4910bf81f571790c87da7c55669c9a2b78de49 Mon Sep 17 00:00:00 2001 From: Brian Bulkowski Date: Sun, 28 Jun 2020 13:05:47 -0700 Subject: [PATCH] Update to talk about instability of RMT config --- README.md | 8 ++++++++ 1 file changed, 8 insertions(+) diff --git a/README.md b/README.md index 4f86beb..9e470cb 100644 --- a/README.md +++ b/README.md @@ -499,6 +499,14 @@ function, directly. If you have a 4-wire system, I left a breadcrumb in the fastpin_esp32 directory. If the alternate code works, then you can take the gpio.c out of the hal compile. +## interesting instability in RMT initialization + +The code in ESP-IDF seems to really like using the pattern of having a macro for a +default constructor. In the case of initializing the RMT structure, if I did it "by hand" +and used the -O2 compile option, I got instability. When I switched to using their +constructure, which seems the same to me generally, I get a stable build. I +now wonder if this was some of the issue around not using the IRQ.... + ## message about no hardware SPI pins defined It's true! There are no hardware SPI pins defined. SPI is used for 4-wire LEDs, where