- for (i=b->top&(~7); i>0; i-=8)
- {
- A[0]=B[0]; A[1]=B[1]; A[2]=B[2]; A[3]=B[3];
- A[4]=B[4]; A[5]=B[5]; A[6]=B[6]; A[7]=B[7];
- A+=8;
- B+=8;
- }
- switch (b->top&7)
- {
- case 7:
- A[6]=B[6];
- case 6:
- A[5]=B[5];
- case 5:
- A[4]=B[4];
- case 4:
- A[3]=B[3];
- case 3:
- A[2]=B[2];
- case 2:
- A[1]=B[1];
- case 1:
- A[0]=B[0];
- case 0:
- /* I need the 'case 0' entry for utrix cc.
- * If the optimizer is turned on, it does the
- * switch table by doing
- * a=top&7
- * a--;
- * goto jump_table[a];
- * If top is 0, this makes us jump to 0xffffffc
- * which is rather bad :-(.
- * eric 23-Apr-1998
- */
- ;
- }
+ for (i=b->top&(~7); i>0; i-=8)
+ {
+ A[0]=B[0]; A[1]=B[1]; A[2]=B[2]; A[3]=B[3];
+ A[4]=B[4]; A[5]=B[5]; A[6]=B[6]; A[7]=B[7];
+ A+=8;
+ B+=8;
+ }
+ switch (b->top&7)
+ {
+ case 7:
+ A[6]=B[6];
+ case 6:
+ A[5]=B[5];
+ case 5:
+ A[4]=B[4];
+ case 4:
+ A[3]=B[3];
+ case 3:
+ A[2]=B[2];
+ case 2:
+ A[1]=B[1];
+ case 1:
+ A[0]=B[0];
+ case 0:
+ /* I need the 'case 0' entry for utrix cc.
+ * If the optimizer is turned on, it does the
+ * switch table by doing
+ * a=top&7
+ * a--;
+ * goto jump_table[a];
+ * If top is 0, this makes us jump to 0xffffffc
+ * which is rather bad :-(.
+ * eric 23-Apr-1998
+ */
+ ;
+ }
+#else
+ for (i=b->top>>2; i>0; i--,A+=4,B+=4)
+ {
+ /*
+ * The fact that the loop is unrolled
+ * 4-wise is a tribute to Intel. It's
+ * the one that doesn't have enough
+ * registers to accomodate more data.
+ * I'd unroll it 8-wise otherwise:-)
+ *
+ * <appro@fy.chalmers.se>
+ */
+ BN_ULONG a0,a1,a2,a3;
+ a0=B[0]; a1=B[1]; a2=B[2]; a3=B[3];
+ A[0]=a0; A[1]=a1; A[2]=a2; A[3]=a3;
+ }
+ switch (b->top&3)
+ {
+ case 3: A[2]=B[2];
+ case 2: A[1]=B[1];
+ case 1: A[0]=B[0];
+ case 0: ; /* ultrix cc workaround, see above */
+ }
+#endif
+ }
+
+ /* Now need to zero any data between b->top and b->max */
+
+ A= &(a[b->top]);
+ for (i=(words - b->top)>>3; i>0; i--,A+=8)
+ {
+ A[0]=0; A[1]=0; A[2]=0; A[3]=0;
+ A[4]=0; A[5]=0; A[6]=0; A[7]=0;
+ }
+ for (i=(words - b->top)&7; i>0; i--,A++)
+ A[0]=0;