diff --git a/.github/workflows/clanker_cross.yml b/.github/workflows/clanker_cross.yml
index f35c1b4..98fdee4 100644
--- a/.github/workflows/clanker_cross.yml
+++ b/.github/workflows/clanker_cross.yml
@@ -30,9 +30,7 @@ jobs:
           - arch: i386
             distro: debian:bookworm
             install: apt-get update && apt-get install -y build-essential clang
-          - arch: aarch64
-            distro: ubuntu:22.04
-            install: apt-get update && apt-get install -y build-essential clang
+          # more archs here
     steps:
       - uses: actions/checkout@v4
       - uses: docker/setup-qemu-action@v3
@@ -49,3 +47,15 @@ jobs:
               make test &&
               make example 
             "
+
+  aarch64:
+    name: Linux aarch64 (native)
+    runs-on: ubuntu-22.04-arm
+    steps:
+      - uses: actions/checkout@v4
+      - name: Install dependencies
+        run: sudo apt-get update && sudo apt-get install -y build-essential clang
+      - name: make test
+        run: touch ffc.h && make test
+      - name: make example
+        run: make example
diff --git a/ffc.h b/ffc.h
index 3a0b10b..0a682dd 100644
--- a/ffc.h
+++ b/ffc.h
@@ -322,17 +322,13 @@ bool ffc_int_kind_is_signed(ffc_int_kind ik) {
 
 ffc_internal ffc_inline
 ffc_value_bits ffc_get_value_bits(ffc_value value, ffc_value_kind vk) {
+  ffc_value_bits bits;
   if (vk == FFC_VALUE_KIND_DOUBLE) {
-#if _MSC_VER && !defined(__clang__)
-    ffc_value_bits bits;
     bits.di = ffc_get_double_bits(value.d);
-    return bits;
-#else
-    return (ffc_value_bits){.di=ffc_get_double_bits(value.d)};
-#endif
   } else {
-    return (ffc_value_bits){.fi=ffc_get_float_bits(value.f)};
+    bits.fi = ffc_get_float_bits(value.f);
   }
+  return bits;
 }
 
 ffc_internal ffc_inline
@@ -1422,14 +1418,10 @@ ffc_result ffc_parse_int_string(
   }
 
   if (p == pend || base < 2 || base > 36) {
-#if _MSC_VER && !defined(__clang__)
     ffc_result invalid_input_result;
     invalid_input_result.ptr = (char*)p;
     invalid_input_result.outcome = FFC_OUTCOME_INVALID_INPUT;
     return invalid_input_result;
-#else
-    return (ffc_result){ .ptr = (char*)p, .outcome = FFC_OUTCOME_INVALID_INPUT };
-#endif
   }
 
   ffc_result answer;
@@ -1474,11 +1466,7 @@ ffc_result ffc_parse_int_string(
 
   if (digit_count == 0) {
     if (has_leading_zeros) {
-#if _MSC_VER && !defined(__clang__)
-      value->u64 = 0;
-#else
-      *value = (ffc_int_value){0}; // Largest variants are defined first so this will clear the entire union
-#endif
+      value->u64 = 0; // Must zero the largest variant!
       answer.outcome = FFC_OUTCOME_OK;
       answer.ptr = p;
     } else {
@@ -1887,24 +1875,16 @@ bool ffc_sv_large_add_from_zero(ffc_sv* x, ffc_bigint_limb_span y) {
 // grade-school multiplication algorithm
 ffc_internal
 bool ffc_bigint_long_mul(ffc_sv* x, ffc_bigint_limb_span y) {
-#if defined(_MSC_VER) && !defined(__clang__)
   ffc_bigint_limb_span xs;
   xs.ptr = x->data;
   xs.len = x->len;
-#else
-  ffc_bigint_limb_span xs = (ffc_bigint_limb_span){ .ptr = x->data, .len = x->len };
-#endif
 
   // full copy of x into z
   ffc_sv z = ffc_sv_create(xs);
 
-#if defined(_MSC_VER) && !defined(__clang__)
   ffc_bigint_limb_span zs;
   zs.ptr = z.data;
   zs.len = z.len;
-#else
-  ffc_bigint_limb_span zs = (ffc_bigint_limb_span){ .ptr = z.data, .len = z.len };
-#endif
 
   if (y.len != 0) {
     ffc_bigint_limb y0 = ffc_span_index(y, 0);
@@ -1918,13 +1898,9 @@ bool ffc_bigint_long_mul(ffc_sv* x, ffc_bigint_limb_span y) {
         zi.len = 0;
         FFC_TRY(ffc_sv_try_extend(&zi, zs));
         FFC_TRY(ffc_bigint_small_mul(&zi, yi));
-#if defined(_MSC_VER) && !defined(__clang__)
         ffc_bigint_limb_span zis;
         zis.ptr = zi.data;
         zis.len = zi.len;
-#else
-        ffc_bigint_limb_span zis = (ffc_bigint_limb_span){zi.data, zi.len};
-#endif
         FFC_TRY(ffc_bigint_large_add_from(x, zis, index));
       }
     }
@@ -1999,13 +1975,10 @@ ffc_inline ffc_internal
 ffc_bigint ffc_bigint_empty(void) {
   ffc_sv sv;
   sv.len = 0;
-#if defined(_MSC_VER) && !defined(__clang__)
+
   ffc_bigint sv_bigint;
   sv_bigint.vec = sv;
   return sv_bigint;
-#else
-  return (ffc_bigint){sv};
-#endif
 }
 
 ffc_inline ffc_internal
@@ -2019,13 +1992,10 @@ ffc_bigint ffc_bigint_make(uint64_t value) {
   ffc_sv_push_unchecked(&sv, (uint32_t)(value >> 32));
 #endif
   ffc_sv_normalize(&sv);
-#if defined(_MSC_VER) && !defined(__clang__)
+
   ffc_bigint sv_bigint;
   sv_bigint.vec = sv;
   return sv_bigint;
-#else
-  return (ffc_bigint){sv};
-#endif
 }
 
 // get the high 64 bits from the vector, and if bits were truncated.
@@ -2197,13 +2167,11 @@ ffc_internal ffc_inline
 bool ffc_bigint_pow5(ffc_bigint* me, uint32_t exp) {
   // multiply by a power of 5
   size_t large_length = sizeof(ffc_large_power_of_5) / sizeof(ffc_bigint_limb);
-#if _MSC_VER && !defined(__clang__)
+
   ffc_bigint_limb_span large;
   large.ptr = (ffc_bigint_limb*)ffc_large_power_of_5;
   large.len = large_length;
-#else
-  ffc_bigint_limb_span large = (ffc_bigint_limb_span){ .ptr = (ffc_bigint_limb*)ffc_large_power_of_5, .len = large_length};
-#endif
+
   while (exp >= pow5_tables_large_step) {
     FFC_TRY(ffc_bigint_large_mul(&me->vec, large));
     exp -= pow5_tables_large_step;
diff --git a/src/bigint.h b/src/bigint.h
index 28ef42c..613a986 100644
--- a/src/bigint.h
+++ b/src/bigint.h
@@ -322,24 +322,16 @@ bool ffc_sv_large_add_from_zero(ffc_sv* x, ffc_bigint_limb_span y) {
 // grade-school multiplication algorithm
 ffc_internal
 bool ffc_bigint_long_mul(ffc_sv* x, ffc_bigint_limb_span y) {
-#if defined(_MSC_VER) && !defined(__clang__)
   ffc_bigint_limb_span xs;
   xs.ptr = x->data;
   xs.len = x->len;
-#else
-  ffc_bigint_limb_span xs = (ffc_bigint_limb_span){ .ptr = x->data, .len = x->len };
-#endif
 
   // full copy of x into z
   ffc_sv z = ffc_sv_create(xs);
 
-#if defined(_MSC_VER) && !defined(__clang__)
   ffc_bigint_limb_span zs;
   zs.ptr = z.data;
   zs.len = z.len;
-#else
-  ffc_bigint_limb_span zs = (ffc_bigint_limb_span){ .ptr = z.data, .len = z.len };
-#endif
 
   if (y.len != 0) {
     ffc_bigint_limb y0 = ffc_span_index(y, 0);
@@ -353,13 +345,9 @@ bool ffc_bigint_long_mul(ffc_sv* x, ffc_bigint_limb_span y) {
         zi.len = 0;
         FFC_TRY(ffc_sv_try_extend(&zi, zs));
         FFC_TRY(ffc_bigint_small_mul(&zi, yi));
-#if defined(_MSC_VER) && !defined(__clang__)
         ffc_bigint_limb_span zis;
         zis.ptr = zi.data;
         zis.len = zi.len;
-#else
-        ffc_bigint_limb_span zis = (ffc_bigint_limb_span){zi.data, zi.len};
-#endif
         FFC_TRY(ffc_bigint_large_add_from(x, zis, index));
       }
     }
@@ -434,13 +422,10 @@ ffc_inline ffc_internal
 ffc_bigint ffc_bigint_empty(void) {
   ffc_sv sv;
   sv.len = 0;
-#if defined(_MSC_VER) && !defined(__clang__)
+
   ffc_bigint sv_bigint;
   sv_bigint.vec = sv;
   return sv_bigint;
-#else
-  return (ffc_bigint){sv};
-#endif
 }
 
 ffc_inline ffc_internal
@@ -454,13 +439,10 @@ ffc_bigint ffc_bigint_make(uint64_t value) {
   ffc_sv_push_unchecked(&sv, (uint32_t)(value >> 32));
 #endif
   ffc_sv_normalize(&sv);
-#if defined(_MSC_VER) && !defined(__clang__)
+
   ffc_bigint sv_bigint;
   sv_bigint.vec = sv;
   return sv_bigint;
-#else
-  return (ffc_bigint){sv};
-#endif
 }
 
 // get the high 64 bits from the vector, and if bits were truncated.
@@ -632,13 +614,11 @@ ffc_internal ffc_inline
 bool ffc_bigint_pow5(ffc_bigint* me, uint32_t exp) {
   // multiply by a power of 5
   size_t large_length = sizeof(ffc_large_power_of_5) / sizeof(ffc_bigint_limb);
-#if _MSC_VER && !defined(__clang__)
+
   ffc_bigint_limb_span large;
   large.ptr = (ffc_bigint_limb*)ffc_large_power_of_5;
   large.len = large_length;
-#else
-  ffc_bigint_limb_span large = (ffc_bigint_limb_span){ .ptr = (ffc_bigint_limb*)ffc_large_power_of_5, .len = large_length};
-#endif
+
   while (exp >= pow5_tables_large_step) {
     FFC_TRY(ffc_bigint_large_mul(&me->vec, large));
     exp -= pow5_tables_large_step;
diff --git a/src/common.h b/src/common.h
index e540da9..b7e8c67 100644
--- a/src/common.h
+++ b/src/common.h
@@ -80,17 +80,13 @@ bool ffc_int_kind_is_signed(ffc_int_kind ik) {
 
 ffc_internal ffc_inline
 ffc_value_bits ffc_get_value_bits(ffc_value value, ffc_value_kind vk) {
+  ffc_value_bits bits;
   if (vk == FFC_VALUE_KIND_DOUBLE) {
-#if _MSC_VER && !defined(__clang__)
-    ffc_value_bits bits;
     bits.di = ffc_get_double_bits(value.d);
-    return bits;
-#else
-    return (ffc_value_bits){.di=ffc_get_double_bits(value.d)};
-#endif
   } else {
-    return (ffc_value_bits){.fi=ffc_get_float_bits(value.f)};
+    bits.fi = ffc_get_float_bits(value.f);
   }
+  return bits;
 }
 
 ffc_internal ffc_inline
diff --git a/src/parse.h b/src/parse.h
index ba3532a..51fa7a3 100644
--- a/src/parse.h
+++ b/src/parse.h
@@ -509,14 +509,10 @@ ffc_result ffc_parse_int_string(
   }
 
   if (p == pend || base < 2 || base > 36) {
-#if _MSC_VER && !defined(__clang__)
     ffc_result invalid_input_result;
     invalid_input_result.ptr = (char*)p;
     invalid_input_result.outcome = FFC_OUTCOME_INVALID_INPUT;
     return invalid_input_result;
-#else
-    return (ffc_result){ .ptr = (char*)p, .outcome = FFC_OUTCOME_INVALID_INPUT };
-#endif
   }
 
   ffc_result answer;
@@ -561,11 +557,7 @@ ffc_result ffc_parse_int_string(
 
   if (digit_count == 0) {
     if (has_leading_zeros) {
-#if _MSC_VER && !defined(__clang__)
-      value->u64 = 0;
-#else
-      *value = (ffc_int_value){0}; // Largest variants are defined first so this will clear the entire union
-#endif
+      value->u64 = 0; // Must zero the largest variant!
       answer.outcome = FFC_OUTCOME_OK;
       answer.ptr = p;
     } else {