Fix conversions trapping and fix some weirdness around conditional jumps

This commit is contained in:
Jef
2019-03-21 11:47:50 +01:00
parent 14bff9229e
commit 4eb546c6a5
2 changed files with 184 additions and 115 deletions

View File

@@ -2162,12 +2162,15 @@ impl<'this, M: ModuleContext> Context<'this, M> {
.map(|c| *c)
.unwrap_or_else(|| self.ret_label());
pass_args(self);
let predicate = self.into_reg(I32, val);
dynasm!(self.asm
; test Rd(predicate.rq().unwrap()), Rd(predicate.rq().unwrap())
);
pass_args(self);
dynasm!(self.asm
; jz =>label.0
);
@@ -2188,12 +2191,15 @@ impl<'this, M: ModuleContext> Context<'this, M> {
.map(|c| *c)
.unwrap_or_else(|| self.ret_label());
pass_args(self);
let predicate = self.into_reg(I32, val);
dynasm!(self.asm
; test Rd(predicate.rq().unwrap()), Rd(predicate.rq().unwrap())
);
pass_args(self);
dynasm!(self.asm
; jnz =>label.0
);
@@ -2344,7 +2350,7 @@ impl<'this, M: ModuleContext> Context<'this, M> {
self.pop_into(dst.into());
}
self.set_stack_depth(cc.stack_depth);
self.set_stack_depth_preserve_flags(cc.stack_depth);
}
/// Puts all stack values into "real" locations so that they can i.e. be set to different
@@ -2452,31 +2458,27 @@ impl<'this, M: ModuleContext> Context<'this, M> {
}
fn immediate_to_reg(&mut self, reg: GPR, val: Value) {
if val.as_bytes() == 0 {
self.zero_reg(reg);
} else {
match reg {
GPR::Rq(r) => {
let val = val.as_bytes();
if (val as u64) <= u32::max_value() as u64 {
dynasm!(self.asm
; mov Rd(r), val as i32
);
} else {
dynasm!(self.asm
; mov Rq(r), QWORD val
);
}
}
GPR::Rx(r) => {
let temp = self.block_state.regs.take(I64);
self.immediate_to_reg(temp, val);
match reg {
GPR::Rq(r) => {
let val = val.as_bytes();
if (val as u64) <= u32::max_value() as u64 {
dynasm!(self.asm
; movq Rx(r), Rq(temp.rq().unwrap())
; mov Rd(r), val as i32
);
} else {
dynasm!(self.asm
; mov Rq(r), QWORD val
);
self.block_state.regs.release(temp);
}
}
GPR::Rx(r) => {
let temp = self.block_state.regs.take(I64);
self.immediate_to_reg(temp, val);
dynasm!(self.asm
; movq Rx(r), Rq(temp.rq().unwrap())
);
self.block_state.regs.release(temp);
}
}
}
@@ -3068,18 +3070,18 @@ impl<'this, M: ModuleContext> Context<'this, M> {
let sign_mask = self.aligned_label(4, LabelValue::I32(SIGN_MASK_F32 as i32));
let float_cmp_mask = self.aligned_label(16, LabelValue::I32(0xcf000000u32 as i32));
let zero = self.aligned_label(16, LabelValue::I32(0));
let trap_mask = self.trap_label();
let trap_label = self.trap_label();
dynasm!(self.asm
; cvttss2si Rd(temp.rq().unwrap()), Rx(reg.rx().unwrap())
; cmp Rd(temp.rq().unwrap()), [=>sign_mask.0]
; jne >ret
; ucomiss Rx(reg.rx().unwrap()), Rx(reg.rx().unwrap())
; jp =>trap_mask.0
; jp =>trap_label.0
; ucomiss Rx(reg.rx().unwrap()), [=>float_cmp_mask.0]
; jnae =>trap_mask.0
; jnae =>trap_label.0
; ucomiss Rx(reg.rx().unwrap()), [=>zero.0]
; jnb =>trap_mask.0
; jnb =>trap_label.0
; ret:
);
@@ -3106,21 +3108,21 @@ impl<'this, M: ModuleContext> Context<'this, M> {
let sign_mask = self.aligned_label(4, LabelValue::I32(SIGN_MASK_F32 as i32));
let float_cmp_mask = self.aligned_label(16, LabelValue::I32(0x4f000000u32 as i32));
let zero = self.aligned_label(16, LabelValue::I32(0));
let trap_mask = self.trap_label();
let trap_label = self.trap_label();
dynasm!(self.asm
; ucomiss Rx(reg.rx().unwrap()), [=>float_cmp_mask.0]
; jae >else_
; jp =>trap_mask.0
; jp =>trap_label.0
; cvttss2si Rd(temp.rq().unwrap()), Rx(reg.rx().unwrap())
; cmp Rd(temp.rq().unwrap()), 0
; jnge =>trap_mask.0
; test Rd(temp.rq().unwrap()), Rd(temp.rq().unwrap())
; js =>trap_label.0
; jmp >ret
; else_:
; subss Rx(reg.rx().unwrap()), [=>float_cmp_mask.0]
; cvttss2si Rd(temp.rq().unwrap()), Rx(reg.rx().unwrap())
; cmp Rd(temp.rq().unwrap()), 0
; jnge =>trap_mask.0
; test Rd(temp.rq().unwrap()), Rd(temp.rq().unwrap())
; js =>trap_label.0
; add Rq(temp.rq().unwrap()), [=>sign_mask.0]
; ret:
);
@@ -3147,20 +3149,21 @@ impl<'this, M: ModuleContext> Context<'this, M> {
val = ValueLocation::Reg(reg);
let sign_mask = self.aligned_label(4, LabelValue::I32(SIGN_MASK_F32 as i32));
let float_cmp_mask = self.aligned_label(16, LabelValue::I64(0xc1e0000000200000u64 as i64));
let float_cmp_mask =
self.aligned_label(16, LabelValue::I64(0xc1e0000000200000u64 as i64));
let zero = self.aligned_label(16, LabelValue::I64(0));
let trap_mask = self.trap_label();
let trap_label = self.trap_label();
dynasm!(self.asm
; cvttsd2si Rd(temp.rq().unwrap()), Rx(reg.rx().unwrap())
; cmp Rd(temp.rq().unwrap()), [=>sign_mask.0]
; jne >ret
; ucomisd Rx(reg.rx().unwrap()), Rx(reg.rx().unwrap())
; jp =>trap_mask.0
; jp =>trap_label.0
; ucomisd Rx(reg.rx().unwrap()), [=>float_cmp_mask.0]
; jna =>trap_mask.0
; jna =>trap_label.0
; ucomisd Rx(reg.rx().unwrap()), [=>zero.0]
; jnb =>trap_mask.0
; jnb =>trap_label.0
; ret:
);
@@ -3186,23 +3189,24 @@ impl<'this, M: ModuleContext> Context<'this, M> {
let temp = self.block_state.regs.take(I32);
let sign_mask = self.aligned_label(4, LabelValue::I32(SIGN_MASK_F32 as i32));
let float_cmp_mask = self.aligned_label(16, LabelValue::I64(0x41e0000000000000u64 as i64));
let float_cmp_mask =
self.aligned_label(16, LabelValue::I64(0x41e0000000000000u64 as i64));
let zero = self.aligned_label(16, LabelValue::I64(0));
let trap_mask = self.trap_label();
let trap_label = self.trap_label();
dynasm!(self.asm
; ucomisd Rx(reg.rx().unwrap()), [=>float_cmp_mask.0]
; jae >else_
; jp =>trap_mask.0
; jp =>trap_label.0
; cvttsd2si Rd(temp.rq().unwrap()), Rx(reg.rx().unwrap())
; cmp Rd(temp.rq().unwrap()), 0
; jnge =>trap_mask.0
; test Rd(temp.rq().unwrap()), Rd(temp.rq().unwrap())
; js =>trap_label.0
; jmp >ret
; else_:
; subsd Rx(reg.rx().unwrap()), [=>float_cmp_mask.0]
; cvttsd2si Rd(temp.rq().unwrap()), Rx(reg.rx().unwrap())
; cmp Rd(temp.rq().unwrap()), 0
; jnge =>trap_mask.0
; test Rd(temp.rq().unwrap()), Rd(temp.rq().unwrap())
; js =>trap_label.0
; add Rq(temp.rq().unwrap()), [=>sign_mask.0]
; ret:
);
@@ -3265,30 +3269,84 @@ impl<'this, M: ModuleContext> Context<'this, M> {
|a| wasmparser::Ieee64((a as f64).to_bits())
);
conversion!(
i64_truncate_f32_s,
cvttss2si,
Rx,
rx,
Rq,
rq,
f32,
i64,
as_f32,
|a: wasmparser::Ieee32| f32::from_bits(a.bits()) as i64
);
conversion!(
i64_truncate_f64_s,
cvttsd2si,
Rx,
rx,
Rq,
rq,
f64,
i64,
as_f64,
|a: wasmparser::Ieee64| f64::from_bits(a.bits()) as i64
);
pub fn i64_truncate_f32_s(&mut self) {
let mut val = self.pop();
let out_val = match val {
ValueLocation::Immediate(imm) => ValueLocation::Immediate(
(f32::from_bits(imm.as_f32().unwrap().bits()) as i32).into(),
),
other => {
let reg = self.into_temp_reg(F32, other);
val = ValueLocation::Reg(reg);
let temp = self.block_state.regs.take(I32);
let sign_mask = self.aligned_label(16, LabelValue::I64(SIGN_MASK_F64 as i64));
let float_cmp_mask = self.aligned_label(16, LabelValue::I32(0xdf000000u32 as i32));
let zero = self.aligned_label(16, LabelValue::I64(0));
let trap_label = self.trap_label();
dynasm!(self.asm
; cvttss2si Rq(temp.rq().unwrap()), Rx(reg.rx().unwrap())
; cmp Rq(temp.rq().unwrap()), [=>sign_mask.0]
; jne >ret
; ucomiss Rx(reg.rx().unwrap()), Rx(reg.rx().unwrap())
; jp =>trap_label.0
; ucomiss Rx(reg.rx().unwrap()), [=>float_cmp_mask.0]
; jnae =>trap_label.0
; ucomiss Rx(reg.rx().unwrap()), [=>zero.0]
; jnb =>trap_label.0
; ret:
);
ValueLocation::Reg(temp)
}
};
self.free_value(val);
self.push(out_val);
}
pub fn i64_truncate_f64_s(&mut self) {
let mut val = self.pop();
let out_val = match val {
ValueLocation::Immediate(imm) => ValueLocation::Immediate(
(f64::from_bits(imm.as_f64().unwrap().bits()) as i32).into(),
),
other => {
let reg = self.into_reg(F32, other);
let temp = self.block_state.regs.take(I32);
val = ValueLocation::Reg(reg);
let sign_mask = self.aligned_label(8, LabelValue::I64(SIGN_MASK_F64 as i64));
let float_cmp_mask =
self.aligned_label(16, LabelValue::I64(0xc3e0000000000000u64 as i64));
let zero = self.aligned_label(16, LabelValue::I64(0));
let trap_label = self.trap_label();
dynasm!(self.asm
; cvttsd2si Rq(temp.rq().unwrap()), Rx(reg.rx().unwrap())
; cmp Rq(temp.rq().unwrap()), [=>sign_mask.0]
; jne >ret
; ucomisd Rx(reg.rx().unwrap()), Rx(reg.rx().unwrap())
; jp =>trap_label.0
; ucomisd Rx(reg.rx().unwrap()), [=>float_cmp_mask.0]
; jnae =>trap_label.0
; ucomisd Rx(reg.rx().unwrap()), [=>zero.0]
; jnb =>trap_label.0
; ret:
);
ValueLocation::Reg(temp)
}
};
self.free_value(val);
self.push(out_val);
}
pub fn i64_truncate_f32_u(&mut self) {
struct Trunc;
@@ -3304,17 +3362,24 @@ impl<'this, M: ModuleContext> Context<'this, M> {
val = ValueLocation::Reg(reg);
let temp = self.block_state.regs.take(I64);
let sign_mask = self.aligned_label(16, LabelValue::I64(SIGN_MASK_F64 as i64));
let u64_trunc_f32_const = self.aligned_label(16, LabelValue::I32(0x5F000000));
let trap_label = self.trap_label();
dynasm!(self.asm
; comiss Rx(reg.rx().unwrap()), [=>u64_trunc_f32_const.0]
; jnb >large
; jae >large
; jp =>trap_label.0
; cvttss2si Rq(temp.rq().unwrap()), Rx(reg.rx().unwrap())
; test Rq(temp.rq().unwrap()), Rq(temp.rq().unwrap())
; js =>trap_label.0
; jmp >cont
; large:
; subss Rx(reg.rx().unwrap()), [=>u64_trunc_f32_const.0]
; cvttss2si Rq(temp.rq().unwrap()), Rx(reg.rx().unwrap())
; btc Rq(temp.rq().unwrap()), 0b00111111
; test Rq(temp.rq().unwrap()), Rq(temp.rq().unwrap())
; js =>trap_label.0
; add Rq(temp.rq().unwrap()), [=>sign_mask.0]
; cont:
);
@@ -3339,18 +3404,25 @@ impl<'this, M: ModuleContext> Context<'this, M> {
val = ValueLocation::Reg(reg);
let temp = self.block_state.regs.take(I64);
let sign_mask = self.aligned_label(16, LabelValue::I64(SIGN_MASK_F64 as i64));
let u64_trunc_f64_const =
self.aligned_label(16, LabelValue::I64(0x43E0000000000000));
self.aligned_label(16, LabelValue::I64(0x43e0000000000000));
let trap_label = self.trap_label();
dynasm!(self.asm
; comisd Rx(reg.rx().unwrap()), [=>u64_trunc_f64_const.0]
; jnb >large
; jp =>trap_label.0
; cvttsd2si Rq(temp.rq().unwrap()), Rx(reg.rx().unwrap())
; jmp >cont
; cmp Rq(temp.rq().unwrap()), 0
; jge >cont
; jmp =>trap_label.0
; large:
; subsd Rx(reg.rx().unwrap()), [=>u64_trunc_f64_const.0]
; cvttsd2si Rq(temp.rq().unwrap()), Rx(reg.rx().unwrap())
; btc Rq(temp.rq().unwrap()), 0b00111111
; cmp Rq(temp.rq().unwrap()), 0
; jnge =>trap_label.0
; add Rq(temp.rq().unwrap()), [=>sign_mask.0]
; cont:
);
@@ -3435,6 +3507,7 @@ impl<'this, M: ModuleContext> Context<'this, M> {
; test Rq(reg.rq().unwrap()), Rq(reg.rq().unwrap())
; js >negative
; cvtsi2ss Rx(out.rx().unwrap()), Rq(reg.rq().unwrap())
; jmp >ret
; negative:
; mov Rq(temp.rq().unwrap()), Rq(reg.rq().unwrap())
; shr Rq(temp.rq().unwrap()), 1
@@ -3442,6 +3515,7 @@ impl<'this, M: ModuleContext> Context<'this, M> {
; or Rq(reg.rq().unwrap()), Rq(temp.rq().unwrap())
; cvtsi2ss Rx(out.rx().unwrap()), Rq(reg.rq().unwrap())
; addss Rx(out.rx().unwrap()), Rx(out.rx().unwrap())
; ret:
);
self.free_value(ValueLocation::Reg(temp));
@@ -3466,27 +3540,22 @@ impl<'this, M: ModuleContext> Context<'this, M> {
let reg = self.into_reg(I64, val);
val = ValueLocation::Reg(reg);
let out = self.block_state.regs.take(F64);
let temp = self.block_state.regs.take(F64);
let conv_const_0 = self.aligned_label(
16,
(LabelValue::I32(0x43300000), LabelValue::I32(0x43300000)),
);
let conv_const_1 = self.aligned_label(
16,
(
LabelValue::I64(0x4330000000000000),
LabelValue::I64(0x4530000000000000),
),
);
let out = self.block_state.regs.take(F32);
let temp = self.block_state.regs.take(I64);
dynasm!(self.asm
; movq Rx(temp.rx().unwrap()), rdi
; punpckldq Rx(temp.rx().unwrap()), [=>conv_const_0.0]
; subpd Rx(temp.rx().unwrap()), [=>conv_const_1.0]
; pshufd Rx(out.rx().unwrap()), Rx(temp.rx().unwrap()), 78
; addpd Rx(out.rx().unwrap()), Rx(temp.rx().unwrap())
; test Rq(reg.rq().unwrap()), Rq(reg.rq().unwrap())
; js >negative
; cvtsi2sd Rx(out.rx().unwrap()), Rq(reg.rq().unwrap())
; jmp >ret
; negative:
; mov Rq(temp.rq().unwrap()), Rq(reg.rq().unwrap())
; shr Rq(temp.rq().unwrap()), 1
; and Rq(reg.rq().unwrap()), 1
; or Rq(reg.rq().unwrap()), Rq(temp.rq().unwrap())
; cvtsi2sd Rx(out.rx().unwrap()), Rq(reg.rq().unwrap())
; addsd Rx(out.rx().unwrap()), Rx(out.rx().unwrap())
; ret:
);
self.free_value(ValueLocation::Reg(temp));