motiejus/zig
1
Fork 0
Code Releases Activity

more intuitive left shift and right shift operators

Browse Source 
Before:
 * << is left shift, not allowed to shift 1 bits out
 * <<% is left shift, allowed to shift 1 bits out
 * >> is right shift, allowed to shift 1 bits out

After:
 * << is left shift, allowed to shift 1 bits out
 * >> is right shift, allowed to shift 1 bits out
 * @shlExact is left shift, not allowed to shift 1 bits out
 * @shrExact is right shift, not allowed to shift 1 bits out

Closes #413
This commit is contained in:
Andrew Kelley
2017-08-09 10:09:38 -04:00
parent 54675b060a
commit 35d3444e27
28 changed files with 274 additions and 128 deletions
Download Patch File Download Diff File Expand all files Collapse all files

67
src/codegen.cpp
View File

@@ -694,8 +694,10 @@ static Buf *panic_msg_buf(PanicMsgId msg_id) {
return buf_create_from_str("integer cast truncated bits");
case PanicMsgIdIntegerOverflow:
return buf_create_from_str("integer overflow");
case PanicMsgIdShiftOverflowedBits:
case PanicMsgIdShlOverflowedBits:
return buf_create_from_str("left shift overflowed bits");
case PanicMsgIdShrOverflowedBits:
return buf_create_from_str("right shift overflowed bits");
case PanicMsgIdDivisionByZero:
return buf_create_from_str("division by zero");
case PanicMsgIdRemainderDivisionByZero:
@@ -1153,7 +1155,7 @@ static LLVMValueRef ir_render_return(CodeGen *g, IrExecutable *executable, IrIns
static LLVMValueRef gen_overflow_shl_op(CodeGen *g, TypeTableEntry *type_entry,
LLVMValueRef val1, LLVMValueRef val2)
{
// for unsigned left shifting, we do the wrapping shift, then logically shift
// for unsigned left shifting, we do the lossy shift, then logically shift
// right the same number of bits
// if the values don't match, we have an overflow
// for signed left shifting we do the same except arithmetic shift right
@@ -1174,7 +1176,32 @@ static LLVMValueRef gen_overflow_shl_op(CodeGen *g, TypeTableEntry *type_entry,
LLVMBuildCondBr(g->builder, ok_bit, ok_block, fail_block);
LLVMPositionBuilderAtEnd(g->builder, fail_block);
gen_debug_safety_crash(g, PanicMsgIdShiftOverflowedBits);
gen_debug_safety_crash(g, PanicMsgIdShlOverflowedBits);
LLVMPositionBuilderAtEnd(g->builder, ok_block);
return result;
}
static LLVMValueRef gen_overflow_shr_op(CodeGen *g, TypeTableEntry *type_entry,
LLVMValueRef val1, LLVMValueRef val2)
{
assert(type_entry->id == TypeTableEntryIdInt);
LLVMValueRef result;
if (type_entry->data.integral.is_signed) {
result = LLVMBuildAShr(g->builder, val1, val2, "");
} else {
result = LLVMBuildLShr(g->builder, val1, val2, "");
}
LLVMValueRef orig_val = LLVMBuildShl(g->builder, result, val2, "");
LLVMValueRef ok_bit = LLVMBuildICmp(g->builder, LLVMIntEQ, val1, orig_val, "");
LLVMBasicBlockRef ok_block = LLVMAppendBasicBlock(g->cur_fn_val, "OverflowOk");
LLVMBasicBlockRef fail_block = LLVMAppendBasicBlock(g->cur_fn_val, "OverflowFail");
LLVMBuildCondBr(g->builder, ok_bit, ok_block, fail_block);
LLVMPositionBuilderAtEnd(g->builder, fail_block);
gen_debug_safety_crash(g, PanicMsgIdShrOverflowedBits);
LLVMPositionBuilderAtEnd(g->builder, ok_block);
return result;
@@ -1496,12 +1523,12 @@ static LLVMValueRef ir_render_bin_op(CodeGen *g, IrExecutable *executable,
return LLVMBuildXor(g->builder, op1_value, op2_value, "");
case IrBinOpBinAnd:
return LLVMBuildAnd(g->builder, op1_value, op2_value, "");
case IrBinOpBitShiftLeft:
case IrBinOpBitShiftLeftWrap:
case IrBinOpBitShiftLeftLossy:
case IrBinOpBitShiftLeftExact:
{
assert(type_entry->id == TypeTableEntryIdInt);
bool is_wrapping = (op_id == IrBinOpBitShiftLeftWrap);
if (is_wrapping) {
bool is_sloppy = (op_id == IrBinOpBitShiftLeftLossy);
if (is_sloppy) {
return LLVMBuildShl(g->builder, op1_value, op2_value, "");
} else if (want_debug_safety) {
return gen_overflow_shl_op(g, type_entry, op1_value, op2_value);
@@ -1511,12 +1538,24 @@ static LLVMValueRef ir_render_bin_op(CodeGen *g, IrExecutable *executable,
return ZigLLVMBuildNUWShl(g->builder, op1_value, op2_value, "");
}
}
case IrBinOpBitShiftRight:
assert(type_entry->id == TypeTableEntryIdInt);
if (type_entry->data.integral.is_signed) {
return LLVMBuildAShr(g->builder, op1_value, op2_value, "");
} else {
return LLVMBuildLShr(g->builder, op1_value, op2_value, "");
case IrBinOpBitShiftRightLossy:
case IrBinOpBitShiftRightExact:
{
assert(type_entry->id == TypeTableEntryIdInt);
bool is_sloppy = (op_id == IrBinOpBitShiftRightLossy);
if (is_sloppy) {
if (type_entry->data.integral.is_signed) {
return LLVMBuildAShr(g->builder, op1_value, op2_value, "");
} else {
return LLVMBuildLShr(g->builder, op1_value, op2_value, "");
}
} else if (want_debug_safety) {
return gen_overflow_shr_op(g, type_entry, op1_value, op2_value);
} else if (type_entry->data.integral.is_signed) {
return ZigLLVMBuildAShrExact(g->builder, op1_value, op2_value, "");
} else {
return ZigLLVMBuildLShrExact(g->builder, op1_value, op2_value, "");
}
}
case IrBinOpSub:
case IrBinOpSubWrap:
@@ -4556,6 +4595,8 @@ static void define_builtin_fns(CodeGen *g) {
create_builtin_fn(g, BuiltinFnIdMod, "mod", 2);
create_builtin_fn(g, BuiltinFnIdInlineCall, "inlineCall", SIZE_MAX);
create_builtin_fn(g, BuiltinFnIdTypeId, "typeId", 1);
create_builtin_fn(g, BuiltinFnIdShlExact, "shlExact", 2);
create_builtin_fn(g, BuiltinFnIdShrExact, "shrExact", 2);
}
static const char *bool_to_str(bool b) {