Speed LTO builds by using multiple threads

jepler · jepler · commit cae02f1cdfdf · 2023-06-20T11:01:46.000-05:00
On my i5-1235U laptop this speeds LTO "partition=balanced" builds
substantially, because each "partition" can be run on a separate
CPU thread. I used "pygamer" as my test build with a parallelism of
`-j4`, and took the best elapsed time reported over 4 builds.

The improvement was from 34.6s to 24.0s (-30%).

A link-only build (rm build-pygamer/firmware.elf; make -j...) improved
from1 17.4s to 5.1s (-70%)

The size of the resulting firmware is unchanged.

Boards that are nearly full use "-flto-partition=one" to improve code
size optimization. When LTO partition is "one", this feature doesn't help
but it doesn't seem to negatively affect anything either (tested
building trinket_m0)
diff --git a/py/circuitpy_defns.mk b/py/circuitpy_defns.mk
@@ -70,7 +70,7 @@ endif
 CIRCUITPY_LTO ?= 0
 CIRCUITPY_LTO_PARTITION ?= balanced
 ifeq ($(CIRCUITPY_LTO),1)
-CFLAGS += -flto -flto-partition=$(CIRCUITPY_LTO_PARTITION) -DCIRCUITPY_LTO=1
+CFLAGS += -flto=jobserver -flto-partition=$(CIRCUITPY_LTO_PARTITION) -DCIRCUITPY_LTO=1
 else
 CFLAGS += -DCIRCUITPY_LTO=0
 endif
