[Core] Use polled waiting on ChibiOS platforms that support it (#17607)

* Use polled waiting on platforms that support it

Due to context switching overhead waiting a very short amount of time on
a sleeping thread is often not accurate and in fact not usable for timing
critical usage i.e. in a driver. Thus we use polled waiting for ranges
in the us range on platforms that support it instead. The fallback is
the thread sleeping mechanism.

This includes:

* ARM platforms with CYCCNT register (ARMv7, ARMv8) this is
  incremented at CPU clock frequency
* GD32VF103 RISC-V port with CSR_MCYCLE register this is incremented at
  CPU clock frequency
* RP2040 ARMv6 port which uses the integrated timer peripheral which is
  incremented with a fixed 1MHz frequency

* Use wait_us() instead of chSysPolledDelayX

...as it is powered by busy waiting now.

* Add chibios waiting methods test bench

2 files changed, 59 insertions, 0 deletions

diff --git a/keyboards/handwired/onekey/keymaps/chibios_waiting_test/config.h b/keyboards/handwired/onekey/keymaps/chibios_waiting_test/config.h
new file mode 100644
index 0000000000..0c75fa3044
--- /dev/null
+++ b/keyboards/handwired/onekey/keymaps/chibios_waiting_test/config.h
@@ -0,0 +1,12 @@
+// Copyright 2022 Stefan Kerkmann
+// SPDX-License-Identifier: GPL-2.0-or-later
+
+#pragma once
+
+#if !defined(QMK_WAITING_TEST_BUSY_PIN)
+#    define QMK_WAITING_TEST_BUSY_PIN A8
+#endif
+
+#if !defined(QMK_WAITING_TEST_YIELD_PIN)
+#    define QMK_WAITING_TEST_YIELD_PIN A9
+#endif
diff --git a/keyboards/handwired/onekey/keymaps/chibios_waiting_test/keymap.c b/keyboards/handwired/onekey/keymaps/chibios_waiting_test/keymap.c
new file mode 100644
index 0000000000..ecf67d3b3c
--- /dev/null
+++ b/keyboards/handwired/onekey/keymaps/chibios_waiting_test/keymap.c
@@ -0,0 +1,47 @@
+// Copyright 2022 Stefan Kerkmann
+// SPDX-License-Identifier: GPL-2.0-or-later
+
+#include QMK_KEYBOARD_H
+
+const uint16_t PROGMEM keymaps[][MATRIX_ROWS][MATRIX_COLS] = {LAYOUT_ortho_1x1(KC_A)};
+
+#if defined(__AVR__)
+#    pragma message "AVR uses polled waiting by default, running theses tests will not show any difference"
+static inline void chThdSleepMicroseconds(uint32_t us) {
+    wait_us(us);
+}
+#endif
+
+void keyboard_post_init_user(void) {
+    setPinOutput(QMK_WAITING_TEST_BUSY_PIN);
+    setPinOutput(QMK_WAITING_TEST_YIELD_PIN);
+}
+
+static inline void wait_us_polling_with_strobe(uint32_t us) {
+    writePinHigh(QMK_WAITING_TEST_BUSY_PIN);
+    wait_us(us);
+    writePinLow(QMK_WAITING_TEST_BUSY_PIN);
+}
+
+static inline void wait_us_yield_with_strobe(uint32_t us) {
+    writePinHigh(QMK_WAITING_TEST_YIELD_PIN);
+    chThdSleepMicroseconds(us);
+    writePinLow(QMK_WAITING_TEST_YIELD_PIN);
+}
+
+static const uint32_t waiting_values[] = {0, 1, 5, 10, 25, 50, 100, 150, 200, 500, 1000};
+
+void housekeeping_task_user(void) {
+    static uint32_t last_bench = 0;
+    if (timer_elapsed32(last_bench) > 500) {
+        for (int i = 0; i < (sizeof(waiting_values) / sizeof(waiting_values[0])); i++) {
+            wait_us_polling_with_strobe(waiting_values[i]);
+            wait_us(10);
+        }
+        for (int i = 0; i < (sizeof(waiting_values) / sizeof(waiting_values[0])); i++) {
+            wait_us_yield_with_strobe(waiting_values[i]);
+            wait_us(10);
+        }
+        last_bench = timer_read32();
+    }
+}