Move DSP instruction stuff to its own header file

Audio.cpp

@@ -1,6 +1,6 @@
 #include "Audio.h"
 #include "arm_math.h"
-
+#include "utility/dspinst.h"
 
 
 
@@ -48,15 +48,6 @@ static void copy_to_fft_buffer(void *destination, const void *source)
 	}
 }
 
-// computes limit((val >> rshift), 2**bits)
-static inline int32_t signed_saturate_rshift(int32_t val, int bits, int rshift) __attribute__((always_inline));
-static inline int32_t signed_saturate_rshift(int32_t val, int bits, int rshift)
-{
-	int32_t out;
-	asm volatile("ssat %0, %1, %2, asr %3" : "=r" (out) : "I" (bits), "r" (val), "I" (rshift));
-	return out;
-}
-
 static void apply_window_to_fft_buffer(void *buffer, const void *window)
 {
 	int16_t *buf = (int16_t *)buffer;
@@ -1149,116 +1140,9 @@ void AudioPlayMemory::update(void)
 
 
 
-
-
-
-
-
-
-
 /******************************************************************/
 
 
-
-
-// computes ((a[31:0] * b[15:0]) >> 16)
-static inline int32_t signed_multiply_32x16b(int32_t a, uint32_t b) __attribute__((always_inline));
-static inline int32_t signed_multiply_32x16b(int32_t a, uint32_t b)
-{
-	int32_t out;
-	asm volatile("smulwb %0, %1, %2" : "=r" (out) : "r" (a), "r" (b));
-	return out;
-}
-
-// computes ((a[31:0] * b[31:16]) >> 16)
-static inline int32_t signed_multiply_32x16t(int32_t a, uint32_t b) __attribute__((always_inline));
-static inline int32_t signed_multiply_32x16t(int32_t a, uint32_t b)
-{
-	int32_t out;
-	asm volatile("smulwt %0, %1, %2" : "=r" (out) : "r" (a), "r" (b));
-	return out;
-}
-
-// computes (((int64_t)a[31:0] * (int64_t)b[31:0]) >> 32)
-static inline int32_t multiply_32x32_rshift32(int32_t a, int32_t b) __attribute__((always_inline));
-static inline int32_t multiply_32x32_rshift32(int32_t a, int32_t b)
-{
-	int32_t out;
-	asm volatile("smmul %0, %1, %2" : "=r" (out) : "r" (a), "r" (b));
-	return out;
-}
-
-// computes (((int64_t)a[31:0] * (int64_t)b[31:0] + 0x8000000) >> 32)
-static inline int32_t multiply_32x32_rshift32_rounded(int32_t a, int32_t b) __attribute__((always_inline));
-static inline int32_t multiply_32x32_rshift32_rounded(int32_t a, int32_t b)
-{
-	int32_t out;
-	asm volatile("smmulr %0, %1, %2" : "=r" (out) : "r" (a), "r" (b));
-	return out;
-}
-
-// computes sum + (((int64_t)a[31:0] * (int64_t)b[31:0] + 0x8000000) >> 32)
-static inline int32_t multiply_accumulate_32x32_rshift32_rounded(int32_t sum, int32_t a, int32_t b) __attribute__((always_inline));
-static inline int32_t multiply_accumulate_32x32_rshift32_rounded(int32_t sum, int32_t a, int32_t b)
-{
-	int32_t out;
-	asm volatile("smmlar %0, %2, %3, %1" : "=r" (out) : "r" (sum), "r" (a), "r" (b));
-	return out;
-}
-
-// computes sum - (((int64_t)a[31:0] * (int64_t)b[31:0] + 0x8000000) >> 32)
-static inline int32_t multiply_subtract_32x32_rshift32_rounded(int32_t sum, int32_t a, int32_t b) __attribute__((always_inline));
-static inline int32_t multiply_subtract_32x32_rshift32_rounded(int32_t sum, int32_t a, int32_t b)
-{
-	int32_t out;
-	asm volatile("smmlsr %0, %2, %3, %1" : "=r" (out) : "r" (sum), "r" (a), "r" (b));
-	return out;
-}
-
-// computes ((a[15:0] << 16) | b[15:0])
-static inline uint32_t pack_16x16(int32_t a, int32_t b) __attribute__((always_inline));
-static inline uint32_t pack_16x16(int32_t a, int32_t b)
-{
-	int32_t out;
-	asm volatile("pkhbt %0, %1, %2, lsl #16" : "=r" (out) : "r" (b), "r" (a));
-	return out;
-}
-
-// computes (((a[31:16] + b[31:16]) << 16) | (a[15:0 + b[15:0]))
-static inline uint32_t signed_add_16_and_16(uint32_t a, uint32_t b) __attribute__((always_inline));
-static inline uint32_t signed_add_16_and_16(uint32_t a, uint32_t b)
-{
-	int32_t out;
-	asm volatile("qadd16 %0, %1, %2" : "=r" (out) : "r" (a), "r" (b));
-	return out;
-}
-
-// computes (sum + ((a[31:0] * b[15:0]) >> 16))
-static inline int32_t signed_multiply_accumulate_32x16b(int32_t sum, int32_t a, uint32_t b)
-{
-	int32_t out;
-	asm volatile("smlawb %0, %2, %3, %1" : "=r" (out) : "r" (sum), "r" (a), "r" (b));
-	return out;
-}
-
-// computes (sum + ((a[31:0] * b[31:16]) >> 16))
-static inline int32_t signed_multiply_accumulate_32x16t(int32_t sum, int32_t a, uint32_t b)
-{
-	int32_t out;
-	asm volatile("smlawt %0, %2, %3, %1" : "=r" (out) : "r" (sum), "r" (a), "r" (b));
-	return out;
-}
-
-// computes logical and, forces compiler to allocate register and use single cycle instruction
-static inline uint32_t logical_and(uint32_t a, uint32_t b)
-{
-	asm volatile("and %0, %1" : "+r" (a) : "r" (b));
-	return a;
-}
-
-
-
-
 void applyGain(int16_t *data, int32_t mult)
 {
 	uint32_t *p = (uint32_t *)data;

misc/mksine.pl

@@ -1,17 +0,0 @@
-#! /usr/bin/perl
-
-use Math::Trig ':pi';
-$len = 256;
-print "#define SINE_TABLE_LEN $len\n";
-print "static const int16_t sine_table[] = {\n";
-for ($i=0; $i <= $len; $i++) {
-	$f = sin($i / $len * 2 * pi);
-	$d = sprintf "%.0f", $f * 32767.0;
-	#print $d;
-	printf "%6d", $d + 0;
-	print "," if ($i < $len);
-	print "\n" if ($i % 10) == 9;
-}
-print "\n" unless ($len % 10) == 9;
-print "};\n";
-

utility/dspinst.h

@@ -0,0 +1,111 @@
+#include <stdint.h>
+
+// computes limit((val >> rshift), 2**bits)
+static inline int32_t signed_saturate_rshift(int32_t val, int bits, int rshift) __attribute__((always_inline, unused));
+static inline int32_t signed_saturate_rshift(int32_t val, int bits, int rshift)
+{
+	int32_t out;
+	asm volatile("ssat %0, %1, %2, asr %3" : "=r" (out) : "I" (bits), "r" (val), "I" (rshift));
+	return out;
+}
+
+// computes ((a[31:0] * b[15:0]) >> 16)
+static inline int32_t signed_multiply_32x16b(int32_t a, uint32_t b) __attribute__((always_inline, unused));
+static inline int32_t signed_multiply_32x16b(int32_t a, uint32_t b)
+{
+	int32_t out;
+	asm volatile("smulwb %0, %1, %2" : "=r" (out) : "r" (a), "r" (b));
+	return out;
+}
+
+// computes ((a[31:0] * b[31:16]) >> 16)
+static inline int32_t signed_multiply_32x16t(int32_t a, uint32_t b) __attribute__((always_inline, unused));
+static inline int32_t signed_multiply_32x16t(int32_t a, uint32_t b)
+{
+	int32_t out;
+	asm volatile("smulwt %0, %1, %2" : "=r" (out) : "r" (a), "r" (b));
+	return out;
+}
+
+// computes (((int64_t)a[31:0] * (int64_t)b[31:0]) >> 32)
+static inline int32_t multiply_32x32_rshift32(int32_t a, int32_t b) __attribute__((always_inline, unused));
+static inline int32_t multiply_32x32_rshift32(int32_t a, int32_t b)
+{
+	int32_t out;
+	asm volatile("smmul %0, %1, %2" : "=r" (out) : "r" (a), "r" (b));
+	return out;
+}
+
+// computes (((int64_t)a[31:0] * (int64_t)b[31:0] + 0x8000000) >> 32)
+static inline int32_t multiply_32x32_rshift32_rounded(int32_t a, int32_t b) __attribute__((always_inline, unused));
+static inline int32_t multiply_32x32_rshift32_rounded(int32_t a, int32_t b)
+{
+	int32_t out;
+	asm volatile("smmulr %0, %1, %2" : "=r" (out) : "r" (a), "r" (b));
+	return out;
+}
+
+// computes sum + (((int64_t)a[31:0] * (int64_t)b[31:0] + 0x8000000) >> 32)
+static inline int32_t multiply_accumulate_32x32_rshift32_rounded(int32_t sum, int32_t a, int32_t b) __attribute__((always_inline, unused));
+static inline int32_t multiply_accumulate_32x32_rshift32_rounded(int32_t sum, int32_t a, int32_t b)
+{
+	int32_t out;
+	asm volatile("smmlar %0, %2, %3, %1" : "=r" (out) : "r" (sum), "r" (a), "r" (b));
+	return out;
+}
+
+// computes sum - (((int64_t)a[31:0] * (int64_t)b[31:0] + 0x8000000) >> 32)
+static inline int32_t multiply_subtract_32x32_rshift32_rounded(int32_t sum, int32_t a, int32_t b) __attribute__((always_inline, unused));
+static inline int32_t multiply_subtract_32x32_rshift32_rounded(int32_t sum, int32_t a, int32_t b)
+{
+	int32_t out;
+	asm volatile("smmlsr %0, %2, %3, %1" : "=r" (out) : "r" (sum), "r" (a), "r" (b));
+	return out;
+}
+
+// computes ((a[15:0] << 16) | b[15:0])
+static inline uint32_t pack_16x16(int32_t a, int32_t b) __attribute__((always_inline, unused));
+static inline uint32_t pack_16x16(int32_t a, int32_t b)
+{
+	int32_t out;
+	asm volatile("pkhbt %0, %1, %2, lsl #16" : "=r" (out) : "r" (b), "r" (a));
+	return out;
+}
+
+// computes (((a[31:16] + b[31:16]) << 16) | (a[15:0 + b[15:0]))
+static inline uint32_t signed_add_16_and_16(uint32_t a, uint32_t b) __attribute__((always_inline, unused));
+static inline uint32_t signed_add_16_and_16(uint32_t a, uint32_t b)
+{
+	int32_t out;
+	asm volatile("qadd16 %0, %1, %2" : "=r" (out) : "r" (a), "r" (b));
+	return out;
+}
+
+// computes (sum + ((a[31:0] * b[15:0]) >> 16))
+static inline int32_t signed_multiply_accumulate_32x16b(int32_t sum, int32_t a, uint32_t b) __attribute__((always_inline, unused));
+static inline int32_t signed_multiply_accumulate_32x16b(int32_t sum, int32_t a, uint32_t b)
+{
+	int32_t out;
+	asm volatile("smlawb %0, %2, %3, %1" : "=r" (out) : "r" (sum), "r" (a), "r" (b));
+	return out;
+}
+
+// computes (sum + ((a[31:0] * b[31:16]) >> 16))
+static inline int32_t signed_multiply_accumulate_32x16t(int32_t sum, int32_t a, uint32_t b) __attribute__((always_inline, unused));
+static inline int32_t signed_multiply_accumulate_32x16t(int32_t sum, int32_t a, uint32_t b)
+{
+	int32_t out;
+	asm volatile("smlawt %0, %2, %3, %1" : "=r" (out) : "r" (sum), "r" (a), "r" (b));
+	return out;
+}
+
+// computes logical and, forces compiler to allocate register and use single cycle instruction
+static inline uint32_t logical_and(uint32_t a, uint32_t b) __attribute__((always_inline, unused));
+static inline uint32_t logical_and(uint32_t a, uint32_t b)
+{
+	asm volatile("and %0, %1" : "+r" (a) : "r" (b));
+	return a;
+}
+
+
+