getting ready for v 1_2_2

CHANGELOG

@@ -1,9 +1,10 @@
 1.2.2 (May 6, 2005)   The Matthew release
-    Replaced fixed point division with multiply&shift. Thanks to Jean-Marc Valin for the prodding.
+    Replaced fixed point division with multiply&shift.  Thanks to Jean-Marc Valin for 
-
+    discussions regarding.  Considerable speedup.
-    Added C_FIXDIV calls to real fft routines to prevent possible overflow when using fixed point.
-    Credit to Robert Oschler of robodance for pointing me towards the bug.
 
+    Corrected overflow protection in real fft routines  when using fixed point.
+    Finder's Credit goes to Robert Oschler of robodance for pointing me at the bug.
+    This also led to the CHECK_OVERFLOW_OP macro.
 
 1.2.1 (April 4, 2004) 
     compiles cleanly with just about every -W warning flag under the sun

_kiss_fft_guts.h

@@ -43,6 +43,13 @@ struct kiss_fft_state{
  * */
 #ifdef FIXED_POINT
 
+#if defined(CHECK_OVERFLOW)
+#  define CHECK_OVERFLOW_OP(a,op,b)  \
+	if ( (long)(a) op (long)(b) > 32767 || (long)(a) op (long)(b) < -32768 ) { \
+		fprintf(stderr,"WARNING:overflow @ " __FILE__ "(%d): (%d " #op" %d) = %ld\n",__LINE__,(a),(b),(long)(a) op (long)(b) );  }
+#endif
+
+
 #   define smul(a,b) ( (long)(a)*(b) )
 #   define sround( x )  (short)( ( (x) + (1<<14) ) >>15 )
 
@@ -53,7 +60,7 @@ struct kiss_fft_state{
           (m).i = sround( smul((a).r,(b).i) + smul((a).i,(b).r) ); }while(0)
 
 #   define DIVSCALAR(x,k) \
-	(x) = sround( smul(  x, 32768/k ) )
+	(x) = sround( smul(  x, 32767/k ) )
 
 #   define C_FIXDIV(c,div) \
 	do {    DIVSCALAR( (c).r , div);  \
@@ -75,14 +82,38 @@ struct kiss_fft_state{
         (c).i *= (s); }while(0)
 #endif
 
+#ifndef CHECK_OVERFLOW_OP
+#  define CHECK_OVERFLOW_OP(a,op,b) /* noop */
+#endif
+
 #define  C_ADD( res, a,b)\
-    do {    (res).r=(a).r+(b).r;  (res).i=(a).i+(b).i;  }while(0)
+    do { \
+	    CHECK_OVERFLOW_OP((a).r,+,(b).r)\
+	    CHECK_OVERFLOW_OP((a).i,+,(b).i)\
+	    (res).r=(a).r+(b).r;  (res).i=(a).i+(b).i; \
+    }while(0)
 #define  C_SUB( res, a,b)\
-    do {    (res).r=(a).r-(b).r;  (res).i=(a).i-(b).i;  }while(0)
+    do { \
+	    CHECK_OVERFLOW_OP((a).r,-,(b).r)\
+	    CHECK_OVERFLOW_OP((a).i,-,(b).i)\
+	    (res).r=(a).r-(b).r;  (res).i=(a).i-(b).i; \
+    }while(0)
 #define C_ADDTO( res , a)\
-    do {    (res).r += (a).r;  (res).i += (a).i;  }while(0)
+    do { \
+	    CHECK_OVERFLOW_OP((res).r,+,(a).r)\
+	    CHECK_OVERFLOW_OP((res).i,+,(a).i)\
+	    (res).r += (a).r;  (res).i += (a).i;\
+    }while(0)
+
 #define C_SUBFROM( res , a)\
-    do {    (res).r -= (a).r;  (res).i -= (a).i;  }while(0)
+    do {\
+	    CHECK_OVERFLOW_OP((res).r,-,(a).r)\
+	    CHECK_OVERFLOW_OP((res).i,-,(a).i)\
+	    (res).r -= (a).r;  (res).i -= (a).i; \
+    }while(0)
+
+
+
 
 static 
 void kf_cexp(kiss_fft_cpx * x,double phase) /* returns e ** (j*phase)   */

tools/kiss_fftr.c

@@ -65,6 +65,7 @@ void kiss_fftr(kiss_fftr_cfg st,const kiss_fft_scalar *timedata,kiss_fft_cpx *fr
 {
     /* input buffer timedata is stored row-wise */
     int k,N;
+    kiss_fft_cpx fpnk,fpk,f1k,f2k,tw,tdc;
 
     if ( st->substate->inverse) {
         fprintf(stderr,"kiss fft usage error: improper alloc\n");
@@ -76,12 +77,15 @@ void kiss_fftr(kiss_fftr_cfg st,const kiss_fft_scalar *timedata,kiss_fft_cpx *fr
     /*perform the parallel fft of two real signals packed in real,imag*/
     kiss_fft( st->substate , (const kiss_fft_cpx*)timedata, st->tmpbuf );
  
-    freqdata[0].r = st->tmpbuf[0].r + st->tmpbuf[0].i;
+    tdc.r = st->tmpbuf[0].r;
+    tdc.i = st->tmpbuf[0].i;
+    C_FIXDIV(tdc,2);
+
+    CHECK_OVERFLOW_OP(tdc.r ,+, tdc.i);
+    freqdata[0].r = tdc.r + tdc.i;
     freqdata[0].i = 0;
-    C_FIXDIV(freqdata[0],2);
 
     for (k=1;k <= N/2 ; ++k ) {
-        kiss_fft_cpx fpnk,fpk,f1k,f2k,tw;
 
         fpk = st->tmpbuf[k]; 
         fpnk.r =  st->tmpbuf[N-k].r;
@@ -100,9 +104,9 @@ void kiss_fftr(kiss_fftr_cfg st,const kiss_fft_scalar *timedata,kiss_fft_cpx *fr
         freqdata[N-k].r =   (f1k.r - tw.r)/2;
         freqdata[N-k].i = - (f1k.i - tw.i)/2;
     }
-    freqdata[N].r = st->tmpbuf[0].r - st->tmpbuf[0].i;
+    CHECK_OVERFLOW_OP(tdc.r ,-, tdc.i);
+    freqdata[N].r = tdc.r - tdc.i;
     freqdata[N].i = 0;
-    C_FIXDIV(freqdata[N],2);
 }
 
 void kiss_fftri(kiss_fftr_cfg st,const kiss_fft_cpx *freqdata,kiss_fft_scalar *timedata)