FL_ALIGN_PROGMEM

#define FL_ALIGN_PROGMEM

(

N

)

Value:

__attribute__ ((aligned (N)))

Force N-byte alignment for platforms with unaligned access or cache-line optimization.

On some platforms, most notably ARM M0, unaligned access to 'PROGMEM' for multibyte values (e.g. read dword) is not allowed and causes a crash. This macro can help force alignment as needed. The FastLED gradient palette code uses 'read dword', and now uses this macro to make sure that gradient palettes are 4-byte aligned.

Usage: FL_ALIGN_PROGMEM(N) where N is the alignment in bytes

• FL_ALIGN_PROGMEM(4): 4-byte alignment (safe default for all platforms)
• FL_ALIGN_PROGMEM(64): Cache-line optimization on x86/ARM/ESP

Platform-specific behavior:

• x86/WASM: Uses attribute((aligned(N))) for cache-line optimization
• AVR: Caps at 4 bytes (flash PROGMEM doesn't benefit from larger alignment)
• ESP32: Supports full alignment (cache-line optimization)
• ESP8266: Caps at 4 bytes (no memalign support)
• ARM: Supports full alignment

Definition at line 139 of file fastled_progmem.h.