litex: move more critical code to RAM

The XIP SPI flash on Fomu is slow, which results in certain operations
taking a long time. This becomes a problem for time-critical operations
such as USB.

Move various calls into RAM to improve performance.

This includes the call to __modsi3 and __udivsi3 which are used by the
supervisor handler to determine if periodic callbacks need to be run.

This finishes fixing adafruit#3841

Signed-off-by: Sean Cross <[email protected]>

Author: xobs

ports/litex/boards/fomu/fomu-spi.ld

@@ -51,6 +51,11 @@ SECTIONS
         *(.text.tu_edpt_dir)
         *(.text.tu_fifo_empty)
         *(.text.usbd_edpt_busy)
+        *(.text.usb_irq_handler)
+        *(.text.supervisor_tick)
+        *(.text.port_get_raw_ticks)
+        *(.text.__modsi3)
+        *(.text.__udivsi3)
         *(.text.irq_getmask)
         *(.text.irq_setmask)
         *(.text.irq_pending)

ports/litex/common-hal/microcontroller/__init__.c

@@ -59,12 +59,15 @@ void common_hal_mcu_delay_us(uint32_t delay) {
 
 volatile uint32_t nesting_count = 0;
 
+__attribute__((section(".ramtext")))
 void common_hal_mcu_disable_interrupts(void) {
-    irq_setie(0);
-    // __DMB();
+    if (nesting_count == 0) {
+        irq_setie(0);
+    }
     nesting_count++;
 }
 
+__attribute__((section(".ramtext")))
 void common_hal_mcu_enable_interrupts(void) {
     if (nesting_count == 0) {
         // This is very very bad because it means there was mismatched disable/enables so we