Don't use vector arithmetic in generic arch_32

src/p448/arch_32/f_impl.h

@@ -10,37 +10,22 @@
 #define LIMB_PLACE_VALUE(i) 28
 
 void gf_add_RAW (gf out, const gf a, const gf b) {
-    for (unsigned int i=0; i<sizeof(*out)/sizeof(uint32xn_t); i++) {
-        ((uint32xn_t*)out)[i] = ((const uint32xn_t*)a)[i] + ((const uint32xn_t*)b)[i];
-    }
-    /*
-    unsigned int i;
-    for (i=0; i<sizeof(*out)/sizeof(out->limb[0]); i++) {
+    for (unsigned int i=0; i<sizeof(*out)/sizeof(out->limb[0]); i++) {
         out->limb[i] = a->limb[i] + b->limb[i];
     }
-    */
 }
 
 void gf_sub_RAW (gf out, const gf a, const gf b) {
-    for (unsigned int i=0; i<sizeof(*out)/sizeof(uint32xn_t); i++) {
-        ((uint32xn_t*)out)[i] = ((const uint32xn_t*)a)[i] - ((const uint32xn_t*)b)[i];
-    }
-    /*
-    unsigned int i;
-    for (i=0; i<sizeof(*out)/sizeof(out->limb[0]); i++) {
+    for (unsigned int i=0; i<sizeof(*out)/sizeof(out->limb[0]); i++) {
         out->limb[i] = a->limb[i] - b->limb[i];
     }
-    */
 }
 
-void gf_bias (gf a, int amt) {
+void gf_bias (gf a, int amt) {    
     uint32_t co1 = ((1ull<<28)-1)*amt, co2 = co1-amt;
-    uint32x4_t lo = {co1,co1,co1,co1}, hi = {co2,co1,co1,co1};
-    uint32x4_t *aa = (uint32x4_t*) a;
-    aa[0] += lo;
-    aa[1] += lo;
-    aa[2] += hi;
-    aa[3] += lo;
+    for (unsigned int i=0; i<sizeof(*a)/sizeof(a->limb[0]); i++) {
+        a->limb[i] += (i==sizeof(*a)/sizeof(a->limb[0])/2) ? co2 : co1;
+    }
 }
 
 void gf_weak_reduce (gf a) {