Commit ebea56d279
Changed files (4)
lib/std/math/big/int.zig
@@ -2544,8 +2544,7 @@ pub const Const = struct {
const bits_per_limb = @bitSizeOf(Limb);
while (i != 0) {
i -= 1;
- const limb = a.limbs[i];
- const this_limb_lz = @clz(limb);
+ const this_limb_lz = @clz(a.limbs[i]);
total_limb_lz += this_limb_lz;
if (this_limb_lz != bits_per_limb) break;
}
@@ -2557,6 +2556,7 @@ pub const Const = struct {
pub fn ctz(a: Const, bits: Limb) Limb {
// Limbs are stored in little-endian order. Converting a negative number to twos-complement
// flips all bits above the lowest set bit, which does not affect the trailing zero count.
+ if (a.eqlZero()) return bits;
var result: Limb = 0;
for (a.limbs) |limb| {
const limb_tz = @ctz(limb);
lib/std/math/big/int_test.zig
@@ -3332,3 +3332,227 @@ test "(BigInt) negative" {
try testing.expect(mem.eql(u8, a_fmt, "(BigInt)"));
try testing.expect(!mem.eql(u8, b_fmt, "(BigInt)"));
}
+
+test "clz" {
+ const neg_limb_max_squared: std.math.big.int.Const = .{
+ .limbs = &.{ 1, maxInt(Limb) - 1 },
+ .positive = false,
+ };
+ try testing.expect(neg_limb_max_squared.clz(@bitSizeOf(Limb) * 2 + 1) == 0);
+
+ const neg_limb_max_squared_plus_one: std.math.big.int.Const = .{
+ .limbs = &.{ 0, maxInt(Limb) - 1 },
+ .positive = false,
+ };
+ try testing.expect(neg_limb_max_squared_plus_one.clz(@bitSizeOf(Limb) * 2 + 1) == 0);
+
+ const neg_limb_msb_squared: std.math.big.int.Const = .{
+ .limbs = &.{ 0, 1 << @bitSizeOf(Limb) - 2 },
+ .positive = false,
+ };
+ try testing.expect(neg_limb_msb_squared.clz(@bitSizeOf(Limb) * 2) == 0);
+ try testing.expect(neg_limb_msb_squared.clz(@bitSizeOf(Limb) * 2 + 1) == 0);
+
+ const neg_limb_max: std.math.big.int.Const = .{
+ .limbs = &.{maxInt(Limb)},
+ .positive = false,
+ };
+ try testing.expect(neg_limb_max.clz(@bitSizeOf(Limb) + 1) == 0);
+ try testing.expect(neg_limb_max.clz(@bitSizeOf(Limb) * 2 - 1) == 0);
+ try testing.expect(neg_limb_max.clz(@bitSizeOf(Limb) * 2) == 0);
+ try testing.expect(neg_limb_max.clz(@bitSizeOf(Limb) * 2 + 1) == 0);
+
+ const neg_limb_msb: std.math.big.int.Const = .{
+ .limbs = &.{1 << @bitSizeOf(Limb) - 1},
+ .positive = false,
+ };
+ try testing.expect(neg_limb_msb.clz(@bitSizeOf(Limb)) == 0);
+ try testing.expect(neg_limb_msb.clz(@bitSizeOf(Limb) + 1) == 0);
+ try testing.expect(neg_limb_msb.clz(@bitSizeOf(Limb) * 2 - 1) == 0);
+ try testing.expect(neg_limb_msb.clz(@bitSizeOf(Limb) * 2) == 0);
+ try testing.expect(neg_limb_msb.clz(@bitSizeOf(Limb) * 2 + 1) == 0);
+
+ const neg_one: std.math.big.int.Const = .{
+ .limbs = &.{1},
+ .positive = false,
+ };
+ try testing.expect(neg_one.clz(@bitSizeOf(Limb)) == 0);
+ try testing.expect(neg_one.clz(@bitSizeOf(Limb) + 1) == 0);
+ try testing.expect(neg_one.clz(@bitSizeOf(Limb) * 2 - 1) == 0);
+ try testing.expect(neg_one.clz(@bitSizeOf(Limb) * 2) == 0);
+ try testing.expect(neg_one.clz(@bitSizeOf(Limb) * 2 + 1) == 0);
+
+ const zero: std.math.big.int.Const = .{
+ .limbs = &.{0},
+ .positive = true,
+ };
+ try testing.expect(zero.clz(@bitSizeOf(Limb)) == @bitSizeOf(Limb));
+ try testing.expect(zero.clz(@bitSizeOf(Limb) + 1) == @bitSizeOf(Limb) + 1);
+ try testing.expect(zero.clz(@bitSizeOf(Limb) * 2 - 1) == @bitSizeOf(Limb) * 2 - 1);
+ try testing.expect(zero.clz(@bitSizeOf(Limb) * 2) == @bitSizeOf(Limb) * 2);
+ try testing.expect(zero.clz(@bitSizeOf(Limb) * 2 + 1) == @bitSizeOf(Limb) * 2 + 1);
+
+ const one: std.math.big.int.Const = .{
+ .limbs = &.{1},
+ .positive = true,
+ };
+ try testing.expect(one.clz(@bitSizeOf(Limb)) == @bitSizeOf(Limb) - 1);
+ try testing.expect(one.clz(@bitSizeOf(Limb) + 1) == @bitSizeOf(Limb));
+ try testing.expect(one.clz(@bitSizeOf(Limb) * 2 - 1) == @bitSizeOf(Limb) * 2 - 2);
+ try testing.expect(one.clz(@bitSizeOf(Limb) * 2) == @bitSizeOf(Limb) * 2 - 1);
+ try testing.expect(one.clz(@bitSizeOf(Limb) * 2 + 1) == @bitSizeOf(Limb) * 2);
+
+ const limb_msb: std.math.big.int.Const = .{
+ .limbs = &.{1 << @bitSizeOf(Limb) - 1},
+ .positive = true,
+ };
+ try testing.expect(limb_msb.clz(@bitSizeOf(Limb)) == 0);
+ try testing.expect(limb_msb.clz(@bitSizeOf(Limb) + 1) == 1);
+ try testing.expect(limb_msb.clz(@bitSizeOf(Limb) * 2 - 1) == @bitSizeOf(Limb) - 1);
+ try testing.expect(limb_msb.clz(@bitSizeOf(Limb) * 2) == @bitSizeOf(Limb));
+ try testing.expect(limb_msb.clz(@bitSizeOf(Limb) * 2 + 1) == @bitSizeOf(Limb) + 1);
+
+ const limb_max: std.math.big.int.Const = .{
+ .limbs = &.{maxInt(Limb)},
+ .positive = true,
+ };
+ try testing.expect(limb_max.clz(@bitSizeOf(Limb)) == 0);
+ try testing.expect(limb_max.clz(@bitSizeOf(Limb) + 1) == 1);
+ try testing.expect(limb_max.clz(@bitSizeOf(Limb) * 2 - 1) == @bitSizeOf(Limb) - 1);
+ try testing.expect(limb_max.clz(@bitSizeOf(Limb) * 2) == @bitSizeOf(Limb));
+ try testing.expect(limb_max.clz(@bitSizeOf(Limb) * 2 + 1) == @bitSizeOf(Limb) + 1);
+
+ const limb_msb_squared: std.math.big.int.Const = .{
+ .limbs = &.{ 0, 1 << @bitSizeOf(Limb) - 2 },
+ .positive = true,
+ };
+ try testing.expect(limb_msb_squared.clz(@bitSizeOf(Limb) * 2 - 1) == 0);
+ try testing.expect(limb_msb_squared.clz(@bitSizeOf(Limb) * 2) == 1);
+ try testing.expect(limb_msb_squared.clz(@bitSizeOf(Limb) * 2 + 1) == 2);
+
+ const limb_max_squared_minus_one: std.math.big.int.Const = .{
+ .limbs = &.{ 0, maxInt(Limb) - 1 },
+ .positive = true,
+ };
+ try testing.expect(limb_max_squared_minus_one.clz(@bitSizeOf(Limb) * 2) == 0);
+ try testing.expect(limb_max_squared_minus_one.clz(@bitSizeOf(Limb) * 2 + 1) == 1);
+
+ const limb_max_squared: std.math.big.int.Const = .{
+ .limbs = &.{ 1, maxInt(Limb) - 1 },
+ .positive = true,
+ };
+ try testing.expect(limb_max_squared.clz(@bitSizeOf(Limb) * 2) == 0);
+ try testing.expect(limb_max_squared.clz(@bitSizeOf(Limb) * 2 + 1) == 1);
+}
+
+test "ctz" {
+ const neg_limb_max_squared: std.math.big.int.Const = .{
+ .limbs = &.{ 1, maxInt(Limb) - 1 },
+ .positive = false,
+ };
+ try testing.expect(neg_limb_max_squared.ctz(@bitSizeOf(Limb) * 2 + 1) == 0);
+
+ const neg_limb_max_squared_plus_one: std.math.big.int.Const = .{
+ .limbs = &.{ 0, maxInt(Limb) - 1 },
+ .positive = false,
+ };
+ try testing.expect(neg_limb_max_squared_plus_one.ctz(@bitSizeOf(Limb) * 2 + 1) == @bitSizeOf(Limb) + 1);
+
+ const neg_limb_msb_squared: std.math.big.int.Const = .{
+ .limbs = &.{ 0, 1 << @bitSizeOf(Limb) - 2 },
+ .positive = false,
+ };
+ try testing.expect(neg_limb_msb_squared.ctz(@bitSizeOf(Limb) * 2) == @bitSizeOf(Limb) * 2 - 2);
+ try testing.expect(neg_limb_msb_squared.ctz(@bitSizeOf(Limb) * 2 + 1) == @bitSizeOf(Limb) * 2 - 2);
+
+ const neg_limb_max: std.math.big.int.Const = .{
+ .limbs = &.{maxInt(Limb)},
+ .positive = false,
+ };
+ try testing.expect(neg_limb_max.ctz(@bitSizeOf(Limb) + 1) == 0);
+ try testing.expect(neg_limb_max.ctz(@bitSizeOf(Limb) * 2 - 1) == 0);
+ try testing.expect(neg_limb_max.ctz(@bitSizeOf(Limb) * 2) == 0);
+ try testing.expect(neg_limb_max.ctz(@bitSizeOf(Limb) * 2 + 1) == 0);
+
+ const neg_limb_msb: std.math.big.int.Const = .{
+ .limbs = &.{1 << @bitSizeOf(Limb) - 1},
+ .positive = false,
+ };
+ try testing.expect(neg_limb_msb.ctz(@bitSizeOf(Limb)) == @bitSizeOf(Limb) - 1);
+ try testing.expect(neg_limb_msb.ctz(@bitSizeOf(Limb) + 1) == @bitSizeOf(Limb) - 1);
+ try testing.expect(neg_limb_msb.ctz(@bitSizeOf(Limb) * 2 - 1) == @bitSizeOf(Limb) - 1);
+ try testing.expect(neg_limb_msb.ctz(@bitSizeOf(Limb) * 2) == @bitSizeOf(Limb) - 1);
+ try testing.expect(neg_limb_msb.ctz(@bitSizeOf(Limb) * 2 + 1) == @bitSizeOf(Limb) - 1);
+
+ const neg_one: std.math.big.int.Const = .{
+ .limbs = &.{1},
+ .positive = false,
+ };
+ try testing.expect(neg_one.ctz(@bitSizeOf(Limb)) == 0);
+ try testing.expect(neg_one.ctz(@bitSizeOf(Limb) + 1) == 0);
+ try testing.expect(neg_one.ctz(@bitSizeOf(Limb) * 2 - 1) == 0);
+ try testing.expect(neg_one.ctz(@bitSizeOf(Limb) * 2) == 0);
+ try testing.expect(neg_one.ctz(@bitSizeOf(Limb) * 2 + 1) == 0);
+
+ const zero: std.math.big.int.Const = .{
+ .limbs = &.{0},
+ .positive = true,
+ };
+ try testing.expect(zero.ctz(@bitSizeOf(Limb)) == @bitSizeOf(Limb));
+ try testing.expect(zero.ctz(@bitSizeOf(Limb) + 1) == @bitSizeOf(Limb) + 1);
+ try testing.expect(zero.ctz(@bitSizeOf(Limb) * 2 - 1) == @bitSizeOf(Limb) * 2 - 1);
+ try testing.expect(zero.ctz(@bitSizeOf(Limb) * 2) == @bitSizeOf(Limb) * 2);
+ try testing.expect(zero.ctz(@bitSizeOf(Limb) * 2 + 1) == @bitSizeOf(Limb) * 2 + 1);
+
+ const one: std.math.big.int.Const = .{
+ .limbs = &.{1},
+ .positive = true,
+ };
+ try testing.expect(one.ctz(@bitSizeOf(Limb)) == 0);
+ try testing.expect(one.ctz(@bitSizeOf(Limb) + 1) == 0);
+ try testing.expect(one.ctz(@bitSizeOf(Limb) * 2 - 1) == 0);
+ try testing.expect(one.ctz(@bitSizeOf(Limb) * 2) == 0);
+ try testing.expect(one.ctz(@bitSizeOf(Limb) * 2 + 1) == 0);
+
+ const limb_msb: std.math.big.int.Const = .{
+ .limbs = &.{1 << @bitSizeOf(Limb) - 1},
+ .positive = true,
+ };
+ try testing.expect(limb_msb.ctz(@bitSizeOf(Limb)) == @bitSizeOf(Limb) - 1);
+ try testing.expect(limb_msb.ctz(@bitSizeOf(Limb) + 1) == @bitSizeOf(Limb) - 1);
+ try testing.expect(limb_msb.ctz(@bitSizeOf(Limb) * 2 - 1) == @bitSizeOf(Limb) - 1);
+ try testing.expect(limb_msb.ctz(@bitSizeOf(Limb) * 2) == @bitSizeOf(Limb) - 1);
+ try testing.expect(limb_msb.ctz(@bitSizeOf(Limb) * 2 + 1) == @bitSizeOf(Limb) - 1);
+
+ const limb_max: std.math.big.int.Const = .{
+ .limbs = &.{maxInt(Limb)},
+ .positive = true,
+ };
+ try testing.expect(limb_max.ctz(@bitSizeOf(Limb)) == 0);
+ try testing.expect(limb_max.ctz(@bitSizeOf(Limb) + 1) == 0);
+ try testing.expect(limb_max.ctz(@bitSizeOf(Limb) * 2 - 1) == 0);
+ try testing.expect(limb_max.ctz(@bitSizeOf(Limb) * 2) == 0);
+ try testing.expect(limb_max.ctz(@bitSizeOf(Limb) * 2 + 1) == 0);
+
+ const limb_msb_squared: std.math.big.int.Const = .{
+ .limbs = &.{ 0, 1 << @bitSizeOf(Limb) - 2 },
+ .positive = true,
+ };
+ try testing.expect(limb_msb_squared.ctz(@bitSizeOf(Limb) * 2 - 1) == @bitSizeOf(Limb) * 2 - 2);
+ try testing.expect(limb_msb_squared.ctz(@bitSizeOf(Limb) * 2) == @bitSizeOf(Limb) * 2 - 2);
+ try testing.expect(limb_msb_squared.ctz(@bitSizeOf(Limb) * 2 + 1) == @bitSizeOf(Limb) * 2 - 2);
+
+ const limb_max_squared_minus_one: std.math.big.int.Const = .{
+ .limbs = &.{ 0, maxInt(Limb) - 1 },
+ .positive = true,
+ };
+ try testing.expect(limb_max_squared_minus_one.ctz(@bitSizeOf(Limb) * 2) == @bitSizeOf(Limb) + 1);
+ try testing.expect(limb_max_squared_minus_one.ctz(@bitSizeOf(Limb) * 2 + 1) == @bitSizeOf(Limb) + 1);
+
+ const limb_max_squared: std.math.big.int.Const = .{
+ .limbs = &.{ 1, maxInt(Limb) - 1 },
+ .positive = true,
+ };
+ try testing.expect(limb_max_squared.ctz(@bitSizeOf(Limb) * 2) == 0);
+ try testing.expect(limb_max_squared.ctz(@bitSizeOf(Limb) * 2 + 1) == 0);
+}
src/arch/x86_64/CodeGen.zig
@@ -2418,7 +2418,7 @@ fn genBodyBlock(self: *CodeGen, body: []const Air.Inst.Index) InnerError!void {
}
fn genBody(cg: *CodeGen, body: []const Air.Inst.Index) InnerError!void {
- @setEvalBranchQuota(12_600);
+ @setEvalBranchQuota(12_700);
const pt = cg.pt;
const zcu = pt.zcu;
const ip = &zcu.intern_pool;
@@ -2475,7 +2475,6 @@ fn genBody(cg: *CodeGen, body: []const Air.Inst.Index) InnerError!void {
.bitcast => try cg.airBitCast(inst),
- .ctz => try cg.airCtz(inst),
.popcount => try cg.airPopCount(inst),
.bit_reverse => try cg.airBitReverse(inst),
.splat => try cg.airSplat(inst),
@@ -26978,7 +26977,6 @@ fn genBody(cg: *CodeGen, body: []const Air.Inst.Index) InnerError!void {
.call_always_tail => try cg.airCall(inst, .always_tail, .{ .safety = true }),
.call_never_tail => try cg.airCall(inst, .never_tail, .{ .safety = true }),
.call_never_inline => try cg.airCall(inst, .never_inline, .{ .safety = true }),
-
.clz => |air_tag| if (use_old) try cg.airClz(inst) else {
const ty_op = air_datas[@intFromEnum(inst)].ty_op;
var ops = try cg.tempsFromOperands(inst, .{ty_op.operand});
@@ -30084,6 +30082,404 @@ fn genBody(cg: *CodeGen, body: []const Air.Inst.Index) InnerError!void {
};
try res[0].finish(inst, &.{ty_op.operand}, &ops, cg);
},
+ .ctz => |air_tag| if (use_old) try cg.airCtz(inst) else fallback: {
+ const ty_op = air_datas[@intFromEnum(inst)].ty_op;
+ if (ty_op.ty.toType().isVector(zcu)) break :fallback try cg.airCtz(inst);
+ var ops = try cg.tempsFromOperands(inst, .{ty_op.operand});
+ var res: [1]Temp = undefined;
+ cg.select(&res, &.{ty_op.ty.toType()}, &ops, comptime &.{ .{
+ .required_features = .{ .slow_incdec, null, null, null },
+ .src_constraints = .{ .{ .exact_signed_int = 1 }, .any, .any },
+ .patterns = &.{
+ .{ .src = .{ .mut_mem, .none, .none } },
+ .{ .src = .{ .to_mut_gpr, .none, .none } },
+ },
+ .dst_temps = .{ .{ .ref = .src0 }, .unused },
+ .clobbers = .{ .eflags = true },
+ .each = .{ .once = &.{
+ .{ ._, ._, .add, .dst0b, .si(1), ._, ._ },
+ } },
+ }, .{
+ .src_constraints = .{ .{ .exact_signed_int = 1 }, .any, .any },
+ .patterns = &.{
+ .{ .src = .{ .mut_mem, .none, .none } },
+ .{ .src = .{ .to_mut_gpr, .none, .none } },
+ },
+ .dst_temps = .{ .{ .ref = .src0 }, .unused },
+ .clobbers = .{ .eflags = true },
+ .each = .{ .once = &.{
+ .{ ._, ._c, .in, .dst0b, ._, ._, ._ },
+ } },
+ }, .{
+ .src_constraints = .{ .{ .exact_unsigned_int = 1 }, .any, .any },
+ .patterns = &.{
+ .{ .src = .{ .mut_mem, .none, .none } },
+ .{ .src = .{ .to_mut_gpr, .none, .none } },
+ },
+ .dst_temps = .{ .{ .ref = .src0 }, .unused },
+ .clobbers = .{ .eflags = true },
+ .each = .{ .once = &.{
+ .{ ._, ._, .xor, .dst0b, .si(1), ._, ._ },
+ } },
+ }, .{
+ .required_features = .{ .bmi, null, null, null },
+ .src_constraints = .{ .{ .exact_int = 16 }, .any, .any },
+ .patterns = &.{
+ .{ .src = .{ .mem, .none, .none } },
+ .{ .src = .{ .to_gpr, .none, .none } },
+ },
+ .dst_temps = .{ .{ .rc = .general_purpose }, .unused },
+ .clobbers = .{ .eflags = true },
+ .each = .{ .once = &.{
+ .{ ._, ._, .xor, .dst0d, .dst0d, ._, ._ },
+ .{ ._, ._, .tzcnt, .dst0w, .src0w, ._, ._ },
+ } },
+ }, .{
+ .required_features = .{ .bmi, .false_deps_lzcnt_tzcnt, null, null },
+ .src_constraints = .{ .{ .exact_int = 32 }, .any, .any },
+ .patterns = &.{
+ .{ .src = .{ .mem, .none, .none } },
+ .{ .src = .{ .to_gpr, .none, .none } },
+ },
+ .dst_temps = .{ .{ .rc = .general_purpose }, .unused },
+ .clobbers = .{ .eflags = true },
+ .each = .{ .once = &.{
+ .{ ._, ._, .xor, .dst0d, .dst0d, ._, ._ },
+ .{ ._, ._, .tzcnt, .dst0d, .src0d, ._, ._ },
+ } },
+ }, .{
+ .required_features = .{ .bmi, null, null, null },
+ .src_constraints = .{ .{ .exact_int = 32 }, .any, .any },
+ .patterns = &.{
+ .{ .src = .{ .mem, .none, .none } },
+ .{ .src = .{ .to_gpr, .none, .none } },
+ },
+ .dst_temps = .{ .{ .mut_rc = .{ .ref = .src0, .rc = .general_purpose } }, .unused },
+ .clobbers = .{ .eflags = true },
+ .each = .{ .once = &.{
+ .{ ._, ._, .tzcnt, .dst0d, .src0d, ._, ._ },
+ } },
+ }, .{
+ .required_features = .{ .bmi, null, null, null },
+ .src_constraints = .{ .{ .int = .dword }, .any, .any },
+ .patterns = &.{
+ .{ .src = .{ .to_mut_gpr, .none, .none } },
+ },
+ .dst_temps = .{ .{ .ref = .src0 }, .unused },
+ .clobbers = .{ .eflags = true },
+ .each = .{ .once = &.{
+ .{ ._, ._, .@"or", .src0d, .uia(1, .src0, .add_umax), ._, ._ },
+ .{ ._, ._, .tzcnt, .dst0d, .src0d, ._, ._ },
+ } },
+ }, .{
+ .required_features = .{ .@"64bit", .bmi, .false_deps_lzcnt_tzcnt, null },
+ .src_constraints = .{ .{ .exact_int = 64 }, .any, .any },
+ .patterns = &.{
+ .{ .src = .{ .mem, .none, .none } },
+ .{ .src = .{ .to_gpr, .none, .none } },
+ },
+ .dst_temps = .{ .{ .rc = .general_purpose }, .unused },
+ .clobbers = .{ .eflags = true },
+ .each = .{ .once = &.{
+ .{ ._, ._, .xor, .dst0d, .dst0d, ._, ._ },
+ .{ ._, ._, .tzcnt, .dst0q, .src0q, ._, ._ },
+ } },
+ }, .{
+ .required_features = .{ .@"64bit", .bmi, null, null },
+ .src_constraints = .{ .{ .exact_int = 64 }, .any, .any },
+ .patterns = &.{
+ .{ .src = .{ .mem, .none, .none } },
+ .{ .src = .{ .to_gpr, .none, .none } },
+ },
+ .dst_temps = .{ .{ .mut_rc = .{ .ref = .src0, .rc = .general_purpose } }, .unused },
+ .clobbers = .{ .eflags = true },
+ .each = .{ .once = &.{
+ .{ ._, ._, .tzcnt, .dst0q, .src0q, ._, ._ },
+ } },
+ }, .{
+ .required_features = .{ .@"64bit", .bmi, null, null },
+ .src_constraints = .{ .{ .int = .qword }, .any, .any },
+ .patterns = &.{
+ .{ .src = .{ .to_mut_gpr, .none, .none } },
+ },
+ .dst_temps = .{ .{ .ref = .src0 }, .unused },
+ .clobbers = .{ .eflags = true },
+ .each = .{ .once = &.{
+ .{ ._, ._s, .bt, .src0q, .ua(.src0, .add_bit_size), ._, ._ },
+ .{ ._, ._, .tzcnt, .dst0q, .src0q, ._, ._ },
+ } },
+ }, .{
+ .required_features = .{ .cmov, .bsf_bsr_0_clobbers_result, null, null },
+ .src_constraints = .{ .{ .exact_int = 32 }, .any, .any },
+ .patterns = &.{
+ .{ .src = .{ .mem, .none, .none } },
+ .{ .src = .{ .to_gpr, .none, .none } },
+ },
+ .extra_temps = .{
+ .{ .type = .u32, .kind = .{ .rc = .general_purpose } },
+ .unused,
+ .unused,
+ .unused,
+ .unused,
+ .unused,
+ .unused,
+ .unused,
+ .unused,
+ .unused,
+ .unused,
+ },
+ .dst_temps = .{ .{ .mut_rc = .{ .ref = .src0, .rc = .general_purpose } }, .unused },
+ .clobbers = .{ .eflags = true },
+ .each = .{ .once = &.{
+ .{ ._, ._, .mov, .tmp0d, .si(32), ._, ._ },
+ .{ ._, ._f, .bs, .dst0d, .src0d, ._, ._ },
+ .{ ._, ._z, .cmov, .dst0d, .tmp0d, ._, ._ },
+ } },
+ }, .{
+ .required_features = .{ .bsf_bsr_0_clobbers_result, null, null, null },
+ .src_constraints = .{ .{ .exact_int = 32 }, .any, .any },
+ .patterns = &.{
+ .{ .src = .{ .mem, .none, .none } },
+ .{ .src = .{ .to_gpr, .none, .none } },
+ },
+ .dst_temps = .{ .{ .mut_rc = .{ .ref = .src0, .rc = .general_purpose } }, .unused },
+ .clobbers = .{ .eflags = true },
+ .each = .{ .once = &.{
+ .{ ._, ._f, .bs, .dst0d, .src0d, ._, ._ },
+ .{ ._, ._nz, .j, .@"0f", ._, ._, ._ },
+ .{ ._, ._, .mov, .dst0d, .si(32), ._, ._ },
+ } },
+ }, .{
+ .src_constraints = .{ .{ .exact_int = 16 }, .any, .any },
+ .patterns = &.{
+ .{ .src = .{ .mem, .none, .none } },
+ .{ .src = .{ .to_gpr, .none, .none } },
+ },
+ .dst_temps = .{ .{ .rc = .general_purpose }, .unused },
+ .clobbers = .{ .eflags = true },
+ .each = .{ .once = &.{
+ .{ ._, ._, .mov, .dst0d, .si(16), ._, ._ },
+ .{ ._, ._f, .bs, .dst0w, .src0w, ._, ._ },
+ } },
+ }, .{
+ .src_constraints = .{ .{ .exact_int = 32 }, .any, .any },
+ .patterns = &.{
+ .{ .src = .{ .mem, .none, .none } },
+ .{ .src = .{ .to_gpr, .none, .none } },
+ },
+ .dst_temps = .{ .{ .rc = .general_purpose }, .unused },
+ .clobbers = .{ .eflags = true },
+ .each = .{ .once = &.{
+ .{ ._, ._, .mov, .dst0d, .si(32), ._, ._ },
+ .{ ._, ._f, .bs, .dst0d, .src0d, ._, ._ },
+ } },
+ }, .{
+ .src_constraints = .{ .{ .int = .dword }, .any, .any },
+ .patterns = &.{
+ .{ .src = .{ .to_mut_gpr, .none, .none } },
+ },
+ .dst_temps = .{ .{ .ref = .src0 }, .unused },
+ .clobbers = .{ .eflags = true },
+ .each = .{ .once = &.{
+ .{ ._, ._, .@"or", .src0d, .uia(1, .src0, .add_umax), ._, ._ },
+ .{ ._, ._f, .bs, .dst0d, .src0d, ._, ._ },
+ } },
+ }, .{
+ .required_features = .{ .@"64bit", .cmov, .bsf_bsr_0_clobbers_result, null },
+ .src_constraints = .{ .{ .exact_int = 64 }, .any, .any },
+ .patterns = &.{
+ .{ .src = .{ .mem, .none, .none } },
+ .{ .src = .{ .to_gpr, .none, .none } },
+ },
+ .extra_temps = .{
+ .{ .type = .u32, .kind = .{ .rc = .general_purpose } },
+ .unused,
+ .unused,
+ .unused,
+ .unused,
+ .unused,
+ .unused,
+ .unused,
+ .unused,
+ .unused,
+ .unused,
+ },
+ .dst_temps = .{ .{ .mut_rc = .{ .ref = .src0, .rc = .general_purpose } }, .unused },
+ .clobbers = .{ .eflags = true },
+ .each = .{ .once = &.{
+ .{ ._, ._, .mov, .tmp0d, .si(64), ._, ._ },
+ .{ ._, ._f, .bs, .dst0q, .src0q, ._, ._ },
+ .{ ._, ._z, .cmov, .dst0d, .tmp0d, ._, ._ },
+ } },
+ }, .{
+ .required_features = .{ .@"64bit", .bsf_bsr_0_clobbers_result, null, null },
+ .src_constraints = .{ .{ .exact_int = 64 }, .any, .any },
+ .patterns = &.{
+ .{ .src = .{ .mem, .none, .none } },
+ .{ .src = .{ .to_gpr, .none, .none } },
+ },
+ .dst_temps = .{ .{ .mut_rc = .{ .ref = .src0, .rc = .general_purpose } }, .unused },
+ .clobbers = .{ .eflags = true },
+ .each = .{ .once = &.{
+ .{ ._, ._f, .bs, .dst0q, .src0q, ._, ._ },
+ .{ ._, ._nz, .j, .@"0f", ._, ._, ._ },
+ .{ ._, ._, .mov, .dst0d, .si(64), ._, ._ },
+ } },
+ }, .{
+ .src_constraints = .{ .{ .exact_int = 64 }, .any, .any },
+ .patterns = &.{
+ .{ .src = .{ .mem, .none, .none } },
+ .{ .src = .{ .to_gpr, .none, .none } },
+ },
+ .dst_temps = .{ .{ .rc = .general_purpose }, .unused },
+ .clobbers = .{ .eflags = true },
+ .each = .{ .once = &.{
+ .{ ._, ._, .mov, .dst0d, .si(64), ._, ._ },
+ .{ ._, ._f, .bs, .dst0q, .src0q, ._, ._ },
+ } },
+ }, .{
+ .src_constraints = .{ .{ .int = .qword }, .any, .any },
+ .patterns = &.{
+ .{ .src = .{ .to_mut_gpr, .none, .none } },
+ },
+ .dst_temps = .{ .{ .ref = .src0 }, .unused },
+ .clobbers = .{ .eflags = true },
+ .each = .{ .once = &.{
+ .{ ._, ._s, .bt, .src0q, .ua(.src0, .add_bit_size), ._, ._ },
+ .{ ._, ._f, .bs, .dst0q, .src0q, ._, ._ },
+ } },
+ }, .{
+ .required_features = .{ .@"64bit", .bmi, .false_deps_lzcnt_tzcnt, null },
+ .src_constraints = .{ .{ .remainder_int = .{ .of = .qword, .is = .qword } }, .any, .any },
+ .patterns = &.{
+ .{ .src = .{ .to_mem, .none, .none } },
+ },
+ .extra_temps = .{
+ .{ .type = .isize, .kind = .{ .rc = .general_purpose } },
+ .unused,
+ .unused,
+ .unused,
+ .unused,
+ .unused,
+ .unused,
+ .unused,
+ .unused,
+ .unused,
+ .unused,
+ },
+ .dst_temps = .{ .{ .rc = .general_purpose }, .unused },
+ .clobbers = .{ .eflags = true },
+ .each = .{ .once = &.{
+ .{ ._, ._, .mov, .tmp0p, .sa(.src0, .sub_size), ._, ._ },
+ .{ .@"0:", ._, .xor, .dst0d, .dst0d, ._, ._ },
+ .{ ._, ._, .tzcnt, .dst0q, .memia(.src0q, .tmp0, .add_size), ._, ._ },
+ .{ ._, ._nc, .j, .@"0f", ._, ._, ._ },
+ .{ ._, ._, .add, .tmp0p, .si(8), ._, ._ },
+ .{ ._, ._nc, .j, .@"0b", ._, ._, ._ },
+ .{ ._, ._, .mov, .dst0d, .sa(.src0, .add_bit_size_sub_8_size), ._, ._ },
+ .{ .@"0:", ._, .lea, .dst0d, .leasia(.dst0, .@"8", .tmp0, .add_8_src0_size), ._, ._ },
+ } },
+ }, .{
+ .required_features = .{ .@"64bit", .bmi, null, null },
+ .src_constraints = .{ .{ .remainder_int = .{ .of = .qword, .is = .qword } }, .any, .any },
+ .patterns = &.{
+ .{ .src = .{ .to_mem, .none, .none } },
+ },
+ .extra_temps = .{
+ .{ .type = .isize, .kind = .{ .rc = .general_purpose } },
+ .unused,
+ .unused,
+ .unused,
+ .unused,
+ .unused,
+ .unused,
+ .unused,
+ .unused,
+ .unused,
+ .unused,
+ },
+ .dst_temps = .{ .{ .rc = .general_purpose }, .unused },
+ .clobbers = .{ .eflags = true },
+ .each = .{ .once = &.{
+ .{ ._, ._, .mov, .tmp0p, .sa(.src0, .sub_size), ._, ._ },
+ .{ .@"0:", ._, .tzcnt, .dst0q, .memia(.src0q, .tmp0, .add_size), ._, ._ },
+ .{ ._, ._nc, .j, .@"0f", ._, ._, ._ },
+ .{ ._, ._, .add, .tmp0p, .si(8), ._, ._ },
+ .{ ._, ._nc, .j, .@"0b", ._, ._, ._ },
+ .{ ._, ._, .mov, .dst0d, .sa(.src0, .add_bit_size_sub_8_size), ._, ._ },
+ .{ .@"0:", ._, .lea, .dst0d, .leasia(.dst0, .@"8", .tmp0, .add_8_src0_size), ._, ._ },
+ } },
+ }, .{
+ .required_features = .{ .@"64bit", .bsf_bsr_0_clobbers_result, null, null },
+ .src_constraints = .{ .{ .remainder_int = .{ .of = .qword, .is = .qword } }, .any, .any },
+ .patterns = &.{
+ .{ .src = .{ .to_mem, .none, .none } },
+ },
+ .extra_temps = .{
+ .{ .type = .isize, .kind = .{ .rc = .general_purpose } },
+ .unused,
+ .unused,
+ .unused,
+ .unused,
+ .unused,
+ .unused,
+ .unused,
+ .unused,
+ .unused,
+ .unused,
+ },
+ .dst_temps = .{ .{ .rc = .general_purpose }, .unused },
+ .clobbers = .{ .eflags = true },
+ .each = .{ .once = &.{
+ .{ ._, ._, .mov, .tmp0p, .sa(.src0, .sub_size), ._, ._ },
+ .{ .@"0:", ._f, .bs, .dst0q, .memia(.src0q, .tmp0, .add_size), ._, ._ },
+ .{ ._, ._nz, .j, .@"0f", ._, ._, ._ },
+ .{ ._, ._, .add, .tmp0p, .si(8), ._, ._ },
+ .{ ._, ._nc, .j, .@"0b", ._, ._, ._ },
+ .{ ._, ._, .mov, .dst0d, .sa(.src0, .add_bit_size_sub_8_size), ._, ._ },
+ .{ .@"0:", ._, .lea, .dst0d, .leasia(.dst0, .@"8", .tmp0, .add_8_src0_size), ._, ._ },
+ } },
+ }, .{
+ .required_features = .{ .@"64bit", null, null, null },
+ .src_constraints = .{ .{ .remainder_int = .{ .of = .qword, .is = .qword } }, .any, .any },
+ .patterns = &.{
+ .{ .src = .{ .to_mem, .none, .none } },
+ },
+ .extra_temps = .{
+ .{ .type = .isize, .kind = .{ .rc = .general_purpose } },
+ .unused,
+ .unused,
+ .unused,
+ .unused,
+ .unused,
+ .unused,
+ .unused,
+ .unused,
+ .unused,
+ .unused,
+ },
+ .dst_temps = .{ .{ .rc = .general_purpose }, .unused },
+ .clobbers = .{ .eflags = true },
+ .each = .{ .once = &.{
+ .{ ._, ._, .mov, .tmp0p, .sa(.src0, .sub_size), ._, ._ },
+ .{ .@"0:", ._, .xor, .dst0d, .dst0d, ._, ._ },
+ .{ ._, ._f, .bs, .dst0q, .memia(.src0q, .tmp0, .add_size), ._, ._ },
+ .{ ._, ._nz, .j, .@"0f", ._, ._, ._ },
+ .{ ._, ._, .add, .tmp0p, .si(8), ._, ._ },
+ .{ ._, ._nc, .j, .@"0b", ._, ._, ._ },
+ .{ ._, ._, .mov, .dst0d, .sa(.src0, .add_bit_size_sub_8_size), ._, ._ },
+ .{ .@"0:", ._, .lea, .dst0d, .leasia(.dst0, .@"8", .tmp0, .add_8_src0_size), ._, ._ },
+ } },
+ } }) catch |err| switch (err) {
+ error.SelectFailed => return cg.fail("failed to select {s} {} {}", .{
+ @tagName(air_tag),
+ cg.typeOf(ty_op.operand).fmt(pt),
+ ops[0].tracking(cg),
+ }),
+ else => |e| return e,
+ };
+ try res[0].finish(inst, &.{ty_op.operand}, &ops, cg);
+ },
.byte_swap => |air_tag| if (use_old) try cg.airByteSwap(inst) else fallback: {
const ty_op = air_datas[@intFromEnum(inst)].ty_op;
if (ty_op.ty.toType().isVector(zcu)) break :fallback try cg.airByteSwap(inst);
@@ -99931,6 +100327,7 @@ const Select = struct {
unaligned_size,
bit_size,
src0_bit_size,
+ @"8_size_sub_bit_size",
len,
elem_limbs,
elem_size,
@@ -99958,6 +100355,7 @@ const Select = struct {
const sub_size: Adjust = .{ .sign = .neg, .lhs = .size, .op = .mul, .rhs = .@"1" };
const sub_src0_size_div_8: Adjust = .{ .sign = .neg, .lhs = .src0_size, .op = .div, .rhs = .@"8" };
const sub_src0_size: Adjust = .{ .sign = .neg, .lhs = .src0_size, .op = .mul, .rhs = .@"1" };
+ const add_8_src0_size: Adjust = .{ .sign = .pos, .lhs = .src0_size, .op = .mul, .rhs = .@"8" };
const add_delta_size_div_8: Adjust = .{ .sign = .pos, .lhs = .delta_size, .op = .div, .rhs = .@"8" };
const add_delta_elem_size: Adjust = .{ .sign = .pos, .lhs = .delta_elem_size, .op = .mul, .rhs = .@"1" };
const add_delta_elem_size_div_8: Adjust = .{ .sign = .pos, .lhs = .delta_elem_size, .op = .div, .rhs = .@"8" };
@@ -99972,6 +100370,7 @@ const Select = struct {
const sub_bit_size: Adjust = .{ .sign = .neg, .lhs = .bit_size, .op = .mul, .rhs = .@"1" };
const add_src0_bit_size: Adjust = .{ .sign = .pos, .lhs = .src0_bit_size, .op = .mul, .rhs = .@"1" };
const sub_src0_bit_size: Adjust = .{ .sign = .neg, .lhs = .src0_bit_size, .op = .mul, .rhs = .@"1" };
+ const add_bit_size_sub_8_size: Adjust = .{ .sign = .neg, .lhs = .@"8_size_sub_bit_size", .op = .mul, .rhs = .@"1" };
const add_8_len: Adjust = .{ .sign = .pos, .lhs = .len, .op = .mul, .rhs = .@"8" };
const add_4_len: Adjust = .{ .sign = .pos, .lhs = .len, .op = .mul, .rhs = .@"4" };
const add_3_len: Adjust = .{ .sign = .pos, .lhs = .len, .op = .mul, .rhs = .@"3" };
@@ -100683,6 +101082,10 @@ const Select = struct {
.unaligned_size => @intCast(s.cg.unalignedSize(op.flags.base.ref.typeOf(s))),
.bit_size => @intCast(op.flags.base.ref.typeOf(s).scalarType(s.cg.pt.zcu).bitSize(s.cg.pt.zcu)),
.src0_bit_size => @intCast(Select.Operand.Ref.src0.typeOf(s).scalarType(s.cg.pt.zcu).bitSize(s.cg.pt.zcu)),
+ .@"8_size_sub_bit_size" => {
+ const ty = op.flags.base.ref.typeOf(s);
+ break :lhs @intCast(8 * ty.abiSize(s.cg.pt.zcu) - ty.bitSize(s.cg.pt.zcu));
+ },
.len => @intCast(op.flags.base.ref.typeOf(s).vectorLen(s.cg.pt.zcu)),
.elem_limbs => @intCast(@divExact(
op.flags.base.ref.typeOf(s).scalarType(s.cg.pt.zcu).abiSize(s.cg.pt.zcu),
test/behavior/x86_64/math.zig
@@ -19273,6 +19273,14 @@ test clz {
try test_clz.testIntVectors();
}
+inline fn ctz(comptime Type: type, rhs: Type) @TypeOf(@ctz(rhs)) {
+ return @ctz(rhs);
+}
+test ctz {
+ const test_ctz = unary(ctz, .{});
+ try test_ctz.testInts();
+}
+
inline fn byteSwap(comptime Type: type, rhs: Type) RoundBitsUp(Type, 8) {
return @byteSwap(@as(RoundBitsUp(Type, 8), rhs));
}