arm: Use builtins for ffs/fls
Since ARMv5, the clz instruction allows for efficient implementation of
ffs/fls with builtins. Until ARMv7 (with Thumb-2), this instruction is
only available in ARM mode. LTO makes it difficult to force specific
functions to be in ARM mode, as it is effectively a form of very
aggressive inlining. To work around this, fls/ffs are implemented in
assembly for ARMv5 and ARMv6 when compiling U-Boot in Thumb mode.
Overall, this saves around 75 bytes per call.
This code is synced with v5.15 of the Linux kernel.
Signed-off-by: Sean Anderson <sean.anderson@seco.com>
Reviewed-by: Tom Rini <trini@konsulko.com>
diff --git a/arch/arm/lib/Makefile b/arch/arm/lib/Makefile
index 62cf80f..b1bcd37 100644
--- a/arch/arm/lib/Makefile
+++ b/arch/arm/lib/Makefile
@@ -113,6 +113,11 @@
AFLAGS_REMOVE_memcpy.o := -mthumb -mthumb-interwork
AFLAGS_memset.o := -DMEMSET_NO_THUMB_BUILD
AFLAGS_memcpy.o := -DMEMCPY_NO_THUMB_BUILD
+
+# This is only necessary to force ARM mode on THUMB1 targets.
+ifneq ($(CONFIG_SYS_ARM_ARCH),4)
+obj-y += bitops.o
+endif
endif
endif
diff --git a/arch/arm/lib/bitops.S b/arch/arm/lib/bitops.S
new file mode 100644
index 0000000..29d1524
--- /dev/null
+++ b/arch/arm/lib/bitops.S
@@ -0,0 +1,45 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/*
+ * Copyright (C) 2023 Sean Anderson <sean.anderson@seco.com>
+ *
+ * ARM bitops to call when using THUMB1, which doesn't have these instructions.
+ */
+#include <linux/linkage.h>
+#include <asm/assembler.h>
+
+.pushsection .text.__fls
+ENTRY(__fls)
+ clz r0, r0
+ rsb r0, r0, #31
+ ret lr
+ENDPROC(__fls)
+.popsection
+
+.pushsection .text.__ffs
+ENTRY(__ffs)
+ rsb r3, r0, #0
+ and r0, r0, r3
+ clz r0, r0
+ rsb r0, r0, #31
+ ret lr
+ENDPROC(__ffs)
+.popsection
+
+.pushsection .text.fls
+ENTRY(fls)
+ cmp r0, #0
+ clzne r0, r0
+ rsbne r0, r0, #32
+ ret lr
+ENDPROC(fls)
+.popsection
+
+.pushsection .text.ffs
+ENTRY(ffs)
+ rsb r3, r0, #0
+ and r0, r0, r3
+ clz r0, r0
+ rsb r0, r0, #32
+ ret lr
+ENDPROC(ffs)
+.popsection