nac3core: list slice irrt use one function to handle var size

nac3core/src/codegen/irrt/irrt.c

@@ -57,79 +57,6 @@ int32_t __nac3_range_slice_len(const int32_t start, const int32_t end, const int
 // - The end index is *inclusive*,
 // - The length of src and dest slice size should already
 // be checked: if dest.step == 1 then len(src) <= len(dest) else len(src) == len(dest)
-# define DEF_SLICE_ASSIGN(T) int32_t __nac3_list_slice_assign_##T( \
-    int32_t dest_start, \
-    int32_t dest_end, \
-    int32_t dest_step, \
-    T *dest_arr, \
-    int32_t dest_arr_len, \
-    int32_t src_start, \
-    int32_t src_end, \
-    int32_t src_step, \
-    T *src_arr, \
-    int32_t src_arr_len \
-) { \
-    /* if dest_arr_len == 0, do nothing since we do not support extending list */ \
-    if (dest_arr_len == 0) return dest_arr_len; \
-    /* if both step is 1, memmove directly, handle the dropping of the list, and shrink size */ \
-    if (src_step == dest_step && dest_step == 1) { \
-        const int32_t src_len = (src_end >= src_start) ? (src_end - src_start + 1) : 0; \
-        const int32_t dest_len = (dest_end >= dest_start) ? (dest_end - dest_start + 1) : 0; \
-        if (src_len > 0) { \
-            __builtin_memmove( \
-                dest_arr + dest_start, \
-                src_arr + src_start, \
-                src_len * sizeof(T) \
-            ); \
-        } \
-        if (dest_len > 0) { \
-            /* dropping */ \
-            __builtin_memmove( \
-                dest_arr + dest_start + src_len, \
-                dest_arr + dest_end + 1, \
-                (dest_arr_len - dest_end - 1) * sizeof(T) \
-            ); \
-        } \
-        /* shrink size */ \
-        return dest_arr_len - (dest_len - src_len); \
-    } \
-    /* if two range overlaps, need alloca */ \
-    uint8_t need_alloca = \
-        (dest_arr == src_arr) \
-        && !( \
-            MAX(dest_start, dest_end) < MIN(src_start, src_end) \
-            || MAX(src_start, src_end) < MIN(dest_start, dest_end) \
-        ); \
-    if (need_alloca) { \
-        T *tmp = alloca(src_arr_len * sizeof(T)); \
-        __builtin_memcpy(tmp, src_arr, src_arr_len * sizeof(T)); \
-        src_arr = tmp; \
-    } \
-    int32_t src_ind = src_start; \
-    int32_t dest_ind = dest_start; \
-    for (; \
-        (src_step > 0) ? (src_ind <= src_end) : (src_ind >= src_end); \
-        src_ind += src_step, dest_ind += dest_step \
-    ) { \
-        dest_arr[dest_ind] = src_arr[src_ind]; \
-    } \
-    /* only dest_step == 1 can we shrink the dest list. */ \
-    /* size should be ensured prior to calling this function */ \
-    if (dest_step == 1 && dest_end >= dest_start) { \
-        __builtin_memmove( \
-            dest_arr + dest_ind, \
-            dest_arr + dest_end + 1, \
-            (dest_arr_len - dest_end - 1) * sizeof(T) \
-        ); \
-        return dest_arr_len - (dest_end - dest_ind) - 1; \
-    } \
-    return dest_arr_len; \
-} \
-
-DEF_SLICE_ASSIGN(uint8_t)
-DEF_SLICE_ASSIGN(uint32_t)
-DEF_SLICE_ASSIGN(uint64_t)
-
 int32_t __nac3_list_slice_assign_var_size(
     int32_t dest_start,
     int32_t dest_end,
@@ -185,9 +112,18 @@ int32_t __nac3_list_slice_assign_var_size(
         (src_step > 0) ? (src_ind <= src_end) : (src_ind >= src_end);
         src_ind += src_step, dest_ind += dest_step
     ) {
+        /* for constant optimization */
+        if (size == 1) {
+            __builtin_memcpy(dest_arr + dest_ind, src_arr + src_ind, 1);
+        } else if (size == 4) {
+            __builtin_memcpy(dest_arr + dest_ind * 4, src_arr + src_ind * 4, 4);
+        } else if (size == 8) {
+            __builtin_memcpy(dest_arr + dest_ind * 8, src_arr + src_ind * 8, 8);
+        } else {
             /* memcpy for var size, cannot overlap after previous alloca */
             __builtin_memcpy(dest_arr + dest_ind * size, src_arr + src_ind * size, size);
         }
+    }
     /* only dest_step == 1 can we shrink the dest list. */
     /* size should be ensured prior to calling this function */
     if (dest_step == 1 && dest_end >= dest_start) {

nac3core/src/codegen/irrt/mod.rs

@@ -237,30 +237,9 @@ pub fn list_slice_assignment<'ctx, 'a>(
 ) {
     let int8_ptr = ctx.ctx.i8_type().ptr_type(AddressSpace::Generic);
     let int32 = ctx.ctx.i32_type();
-    let int32_ptr = int32.ptr_type(AddressSpace::Generic);
+    let (fun_symbol, elem_ptr_type) = ("__nac3_list_slice_assign_var_size", int8_ptr);
-    let int64_ptr = ctx.ctx.i64_type().ptr_type(AddressSpace::Generic);
-    let elem_size = match ty {
-        BasicTypeEnum::IntType(ty) => match ty.get_bit_width() {
-            w if w < 32 => Some(8),
-            w if w == 32 || w == 64 => Some(w),
-            _ => unreachable!(),
-        },
-        BasicTypeEnum::FloatType(_) => Some(64),
-        BasicTypeEnum::PointerType(_) => match size_ty.get_bit_width() {
-            w if w == 32 || w == 64 => Some(w),
-            _ => unreachable!(),
-        },
-        BasicTypeEnum::StructType(_) => None,
-        _ => unreachable!(),
-    };
-    let (fun_symbol, elem_ptr_type) = match elem_size {
-        Some(8) => ("__nac3_list_slice_assign_uint8_t", int8_ptr),
-        Some(32) => ("__nac3_list_slice_assign_uint32_t", int32_ptr),
-        Some(64) => ("__nac3_list_slice_assign_uint64_t", int64_ptr),
-        _ => ("__nac3_list_slice_assign_var_size", int8_ptr),
-    };
     let slice_assign_fun = {
-        let mut ty_vec = vec![
+        let ty_vec = vec![
             int32.into(),         // dest start idx
             int32.into(),         // dest end idx
             int32.into(),         // dest step
@@ -271,17 +250,10 @@ pub fn list_slice_assignment<'ctx, 'a>(
             int32.into(),         // src step
             elem_ptr_type.into(), // src arr ptr
             int32.into(),         // src arr len
+            int32.into(),         // size
         ];
         ctx.module.get_function(fun_symbol).unwrap_or_else(|| {
-            let fn_t = int32.fn_type(
+            let fn_t = int32.fn_type(ty_vec.as_slice(), false);
-                {
-                    if fun_symbol == "__nac3_list_slice_assign_var_size" {
-                        ty_vec.push(int32.into());
-                    }
-                    ty_vec.as_slice()
-                },
-                false,
-            );
             ctx.module.add_function(fun_symbol, fn_t, None)
         })
     };
@@ -309,7 +281,7 @@ pub fn list_slice_assignment<'ctx, 'a>(
     // TODO: assert if dest.step == 1 then len(src) <= len(dest) else len(src) == len(dest), and
     // throw exception if not satisfied
     let new_len = {
-        let mut args = vec![
+        let args = vec![
             dest_idx.0.into(),   // dest start idx
             dest_idx.1.into(),   // dest end idx
             dest_idx.2.into(),   // dest step
@@ -320,20 +292,20 @@ pub fn list_slice_assignment<'ctx, 'a>(
             src_idx.2.into(),    // src step
             src_arr_ptr.into(),  // src arr ptr
             src_len.into(),      // src arr len
+            {
+                let s = match ty {
+                    BasicTypeEnum::FloatType(t) => t.size_of(),
+                    BasicTypeEnum::IntType(t) => t.size_of(),
+                    BasicTypeEnum::PointerType(t) => t.size_of(),
+                    BasicTypeEnum::StructType(t) => t.size_of().unwrap(),
+                    _ => unreachable!(),
+                };
+                ctx.builder.build_int_truncate_or_bit_cast(s, int32, "size")
+            }
+            .into(),
         ];
         ctx.builder
-            .build_call(
+            .build_call(slice_assign_fun, args.as_slice(), "slice_assign")
-                slice_assign_fun,
-                {
-                    if fun_symbol == "__nac3_list_slice_assign_var_size" {
-                        let s = ty.into_struct_type().size_of().unwrap();
-                        let s = ctx.builder.build_int_truncate_or_bit_cast(s, int32, "size");
-                        args.push(s.into());
-                    }
-                    args.as_slice()
-                },
-                "slice_assign",
-            )
             .try_as_basic_value()
             .unwrap_left()
             .into_int_value()