Dont inline g726_iterate() the function is big so its inlining will

not help speedwise IMHO.
.o size changes from 70k -> 49k


git-svn-id: file:///var/local/repositories/ffmpeg/trunk@13793 9553f0bf-9b14-0410-a0b8-cfaf0461ba5b

libavcodec/g726.c

@@ -183,7 +183,7 @@ static inline int16_t inverse_quant(G726Context* c, int i)
     return (dql < 0) ? 0 : ((dqt<<7) >> (14-dex));
 }
 
-static inline int16_t g726_iterate(G726Context* c, int16_t I)
+static int16_t g726_iterate(G726Context* c, int16_t I)
 {
     int dq, re_signal, pk0, fa1, i, tr, ylint, ylfrac, thr2, al, dq0;
     Float11 f;