Move very commonly accessed array into IRAM. Should give a nice speedup on Coldfire.

git-svn-id: svn://svn.rockbox.org/rockbox/trunk@13827 a1c6a512-1295-4272-9138-f99709370657

2 files changed, 13 insertions, 9 deletions

diff --git a/apps/codecs/libwma/wmadec.h b/apps/codecs/libwma/wmadec.h
index 395b29f6e8..56f935bbb4 100644
--- a/apps/codecs/libwma/wmadec.h
+++ b/apps/codecs/libwma/wmadec.h
@@ -25,7 +25,7 @@
 #define M_PI    3.14159265358979323846
 
 #define M_PI_F  0x3243f // in fixed 32 format
-#define TWO_M_PI_F  0x6487f	//in fixed 32
+#define TWO_M_PI_F  0x6487f   //in fixed 32
 
 #define MAX_CHANNELS 2
 
@@ -128,7 +128,7 @@ typedef struct WMADecodeContext
     fixed32 exponents[MAX_CHANNELS][BLOCK_MAX_SIZE];
     fixed32 max_exponent[MAX_CHANNELS];
     int16_t coefs1[MAX_CHANNELS][BLOCK_MAX_SIZE];
-    fixed32 coefs[MAX_CHANNELS][BLOCK_MAX_SIZE];
+    fixed32 (*coefs)[MAX_CHANNELS][BLOCK_MAX_SIZE];
     MDCTContext mdct_ctx[BLOCK_NB_SIZES];
     fixed32 *windows[BLOCK_NB_SIZES];
     FFTComplex *mdct_tmp; /* temporary storage for imdct */
diff --git a/apps/codecs/libwma/wmadeci.c b/apps/codecs/libwma/wmadeci.c
index 7eab26863c..a1326abbfe 100644
--- a/apps/codecs/libwma/wmadeci.c
+++ b/apps/codecs/libwma/wmadeci.c
@@ -106,6 +106,8 @@ static void wma_lsp_to_curve_init(WMADecodeContext *s, int frame_len);
 int fft_calc(FFTContext *s, FFTComplex *z);
 
 
+fixed32 coefsarray[MAX_CHANNELS][BLOCK_MAX_SIZE] IBSS_ATTR;
+
 //static variables that replace malloced stuff
 fixed32 stat0[2048], stat1[1024], stat2[512], stat3[256], stat4[128];    //these are the MDCT reconstruction windows
 
@@ -755,6 +757,8 @@ int wma_decode_init(WMADecodeContext* s, asf_waveformatex_t *wfx)
     s->bit_rate = wfx->bitrate;
     s->block_align = wfx->blockalign;
 
+    s->coefs = &coefsarray;
+
     if (wfx->codec_id == ASF_CODEC_ID_WMAV1){
         s->version = 1;
     }else{
@@ -1612,7 +1616,7 @@ static int wma_decode_block(WMADecodeContext *s)
          //   mul = fixtof64(pow_table[total_gain])/(s->block_len/2)/fixtof64(s->max_exponent[ch]);
 
             mult = fixmul64byfixed(mult, mdct_norm);        //what the hell?  This is actually fixed64*2^16!
-            coefs = s->coefs[ch];                                            //VLC exponenents are used to get MDCT coef here!
+            coefs = (*(s->coefs))[ch];                                            //VLC exponenents are used to get MDCT coef here!
 
         n=0;
 
@@ -1750,16 +1754,16 @@ static int wma_decode_block(WMADecodeContext *s)
            never happen */
         if (!s->channel_coded[0])
         {
-            memset(s->coefs[0], 0, sizeof(fixed32) * s->block_len);
+            memset((*(s->coefs))[0], 0, sizeof(fixed32) * s->block_len);
             s->channel_coded[0] = 1;
         }
 
         for(i = 0; i < s->block_len; ++i)
         {
-            a = s->coefs[0][i];
-            b = s->coefs[1][i];
-            s->coefs[0][i] = a + b;
-            s->coefs[1][i] = a - b;
+            a = (*s->coefs)[0][i];
+            b = (*s->coefs)[1][i];
+            (*s->coefs)[0][i] = a + b;
+            (*s->coefs)[1][i] = a - b;
         }
     }
 
@@ -1776,7 +1780,7 @@ static int wma_decode_block(WMADecodeContext *s)
 
             ff_imdct_calc(&s->mdct_ctx[bsize],
                           output,
-                          s->coefs[ch],
+                          (*(s->coefs))[ch],
                           s->mdct_tmp);