cranelift_codegen/isa/pulley_shared/abi.rs
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661 662 663 664 665 666 667 668 669 670 671 672 673 674 675 676 677 678 679 680 681 682 683 684 685 686 687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708 709 710 711 712 713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 728 729 730 731 732 733 734 735 736 737 738 739 740 741 742 743 744 745 746 747 748 749 750 751 752 753 754 755 756 757 758 759 760 761 762 763 764 765 766 767 768 769 770 771 772 773 774 775 776 777 778 779 780 781 782 783 784 785 786 787 788 789 790 791 792 793 794 795 796 797 798 799 800 801 802 803 804 805 806 807 808 809 810 811 812 813 814 815 816 817 818 819 820 821 822 823 824 825 826 827 828 829 830 831 832 833 834 835 836 837 838 839 840 841 842 843 844 845 846 847 848 849 850 851 852 853 854 855 856 857 858 859 860 861 862 863 864 865 866 867 868 869 870 871 872 873 874 875 876 877 878 879 880 881 882 883 884 885 886 887 888 889 890 891 892 893 894 895 896 897 898 899 900 901 902 903 904 905 906 907 908 909 910 911 912 913 914 915 916 917 918 919 920 921 922 923 924 925 926 927 928 929 930 931 932 933 934 935 936 937 938 939 940 941 942 943 944 945 946 947 948 949 950 951 952 953 954 955 956
//! Implementation of a standard Pulley ABI.
use super::{inst::*, PulleyFlags, PulleyTargetKind};
use crate::isa::pulley_shared::{PointerWidth, PulleyBackend};
use crate::{
ir::{self, types::*, MemFlags, Signature},
isa,
machinst::*,
settings, CodegenResult,
};
use alloc::{boxed::Box, vec::Vec};
use core::marker::PhantomData;
use cranelift_bitset::ScalarBitSet;
use regalloc2::{MachineEnv, PReg, PRegSet};
use smallvec::{smallvec, SmallVec};
use std::borrow::ToOwned;
use std::sync::OnceLock;
/// Support for the Pulley ABI from the callee side (within a function body).
pub(crate) type PulleyCallee<P> = Callee<PulleyMachineDeps<P>>;
/// Support for the Pulley ABI from the caller side (at a callsite).
pub(crate) type PulleyABICallSite<P> = CallSite<PulleyMachineDeps<P>>;
/// Pulley-specific ABI behavior. This struct just serves as an implementation
/// point for the trait; it is never actually instantiated.
pub struct PulleyMachineDeps<P>
where
P: PulleyTargetKind,
{
_phantom: PhantomData<P>,
}
impl<P> ABIMachineSpec for PulleyMachineDeps<P>
where
P: PulleyTargetKind,
{
type I = InstAndKind<P>;
type F = PulleyFlags;
/// This is the limit for the size of argument and return-value areas on the
/// stack. We place a reasonable limit here to avoid integer overflow issues
/// with 32-bit arithmetic: for now, 128 MB.
const STACK_ARG_RET_SIZE_LIMIT: u32 = 128 * 1024 * 1024;
fn word_bits() -> u32 {
P::pointer_width().bits().into()
}
/// Return required stack alignment in bytes.
fn stack_align(_call_conv: isa::CallConv) -> u32 {
16
}
fn compute_arg_locs(
call_conv: isa::CallConv,
flags: &settings::Flags,
params: &[ir::AbiParam],
args_or_rets: ArgsOrRets,
add_ret_area_ptr: bool,
mut args: ArgsAccumulator,
) -> CodegenResult<(u32, Option<usize>)> {
// NB: make sure this method stays in sync with
// `cranelift_pulley::interp::Vm::call`.
//
// In general we use the first half of all register banks as argument
// passing registers because, well, why not for now. Currently the only
// exception is x15 which is reserved as a single caller-saved register
// not used for arguments. This is used in `ReturnCallIndirect` to hold
// the location of where we're jumping to.
let x_end = 14;
let f_end = 15;
let v_end = 15;
let mut next_x_reg = 0;
let mut next_f_reg = 0;
let mut next_v_reg = 0;
let mut next_stack: u32 = 0;
let ret_area_ptr = if add_ret_area_ptr {
debug_assert_eq!(args_or_rets, ArgsOrRets::Args);
next_x_reg += 1;
Some(ABIArg::reg(
x_reg(next_x_reg - 1).to_real_reg().unwrap(),
I64,
ir::ArgumentExtension::None,
ir::ArgumentPurpose::Normal,
))
} else {
None
};
for param in params {
// Find the regclass(es) of the register(s) used to store a value of
// this type.
let (rcs, reg_tys) = Self::I::rc_for_type(param.value_type)?;
let mut slots = ABIArgSlotVec::new();
for (rc, reg_ty) in rcs.iter().zip(reg_tys.iter()) {
let next_reg = if (next_x_reg <= x_end) && *rc == RegClass::Int {
let x = Some(x_reg(next_x_reg));
next_x_reg += 1;
x
} else if (next_f_reg <= f_end) && *rc == RegClass::Float {
let f = Some(f_reg(next_f_reg));
next_f_reg += 1;
f
} else if (next_v_reg <= v_end) && *rc == RegClass::Vector {
let v = Some(v_reg(next_v_reg));
next_v_reg += 1;
v
} else {
None
};
if let Some(reg) = next_reg {
slots.push(ABIArgSlot::Reg {
reg: reg.to_real_reg().unwrap(),
ty: *reg_ty,
extension: param.extension,
});
} else {
if args_or_rets == ArgsOrRets::Rets && !flags.enable_multi_ret_implicit_sret() {
return Err(crate::CodegenError::Unsupported(
"Too many return values to fit in registers. \
Use a StructReturn argument instead. (#9510)"
.to_owned(),
));
}
// Compute size and 16-byte stack alignment happens
// separately after all args.
let size = reg_ty.bits() / 8;
let size = std::cmp::max(size, 8);
// Align.
debug_assert!(size.is_power_of_two());
next_stack = align_to(next_stack, size);
slots.push(ABIArgSlot::Stack {
offset: i64::from(next_stack),
ty: *reg_ty,
extension: param.extension,
});
next_stack += size;
}
}
args.push(ABIArg::Slots {
slots,
purpose: param.purpose,
});
}
let pos = if let Some(ret_area_ptr) = ret_area_ptr {
args.push_non_formal(ret_area_ptr);
Some(args.args().len() - 1)
} else {
None
};
next_stack = align_to(next_stack, Self::stack_align(call_conv));
Ok((next_stack, pos))
}
fn gen_load_stack(mem: StackAMode, into_reg: Writable<Reg>, ty: Type) -> Self::I {
let mut flags = MemFlags::trusted();
// Stack loads/stores of vectors always use little-endianess to avoid
// implementing a byte-swap of vectors on big-endian platforms.
if ty.is_vector() {
flags.set_endianness(ir::Endianness::Little);
}
Inst::gen_load(into_reg, mem.into(), ty, flags).into()
}
fn gen_store_stack(mem: StackAMode, from_reg: Reg, ty: Type) -> Self::I {
let mut flags = MemFlags::trusted();
// Stack loads/stores of vectors always use little-endianess to avoid
// implementing a byte-swap of vectors on big-endian platforms.
if ty.is_vector() {
flags.set_endianness(ir::Endianness::Little);
}
Inst::gen_store(mem.into(), from_reg, ty, flags).into()
}
fn gen_move(to_reg: Writable<Reg>, from_reg: Reg, ty: Type) -> Self::I {
Self::I::gen_move(to_reg, from_reg, ty)
}
fn gen_extend(
dst: Writable<Reg>,
src: Reg,
signed: bool,
from_bits: u8,
to_bits: u8,
) -> Self::I {
assert!(from_bits < to_bits);
let src = XReg::new(src).unwrap();
let dst = dst.try_into().unwrap();
match (signed, from_bits) {
(true, 8) => RawInst::Sext8 { dst, src }.into(),
(true, 16) => RawInst::Sext16 { dst, src }.into(),
(true, 32) => RawInst::Sext32 { dst, src }.into(),
(false, 8) => RawInst::Zext8 { dst, src }.into(),
(false, 16) => RawInst::Zext16 { dst, src }.into(),
(false, 32) => RawInst::Zext32 { dst, src }.into(),
_ => unimplemented!("extend {from_bits} to {to_bits} as signed? {signed}"),
}
}
fn get_ext_mode(
_call_conv: isa::CallConv,
specified: ir::ArgumentExtension,
) -> ir::ArgumentExtension {
specified
}
fn gen_args(args: Vec<ArgPair>) -> Self::I {
Inst::Args { args }.into()
}
fn gen_rets(rets: Vec<RetPair>) -> Self::I {
Inst::Rets { rets }.into()
}
fn get_stacklimit_reg(_call_conv: isa::CallConv) -> Reg {
spilltmp_reg()
}
fn gen_add_imm(
_call_conv: isa::CallConv,
into_reg: Writable<Reg>,
from_reg: Reg,
imm: u32,
) -> SmallInstVec<Self::I> {
let dst = into_reg.try_into().unwrap();
let imm = imm as i32;
smallvec![
RawInst::Xconst32 { dst, imm }.into(),
RawInst::Xadd32 {
dst,
src1: from_reg.try_into().unwrap(),
src2: dst.to_reg(),
}
.into()
]
}
fn gen_stack_lower_bound_trap(_limit_reg: Reg) -> SmallInstVec<Self::I> {
unimplemented!("pulley shouldn't need stack bound checks")
}
fn gen_get_stack_addr(mem: StackAMode, dst: Writable<Reg>) -> Self::I {
let dst = dst.to_reg();
let dst = XReg::new(dst).unwrap();
let dst = WritableXReg::from_reg(dst);
let mem = mem.into();
Inst::LoadAddr { dst, mem }.into()
}
fn gen_load_base_offset(into_reg: Writable<Reg>, base: Reg, offset: i32, ty: Type) -> Self::I {
let base = XReg::try_from(base).unwrap();
let mem = Amode::RegOffset { base, offset };
Inst::gen_load(into_reg, mem, ty, MemFlags::trusted()).into()
}
fn gen_store_base_offset(base: Reg, offset: i32, from_reg: Reg, ty: Type) -> Self::I {
let base = XReg::try_from(base).unwrap();
let mem = Amode::RegOffset { base, offset };
Inst::gen_store(mem, from_reg, ty, MemFlags::trusted()).into()
}
fn gen_sp_reg_adjust(amount: i32) -> SmallInstVec<Self::I> {
if amount == 0 {
return smallvec![];
}
let inst = if amount < 0 {
let amount = amount.checked_neg().unwrap();
if let Ok(amt) = u32::try_from(amount) {
RawInst::StackAlloc32 { amt }
} else {
unreachable!()
}
} else {
if let Ok(amt) = u32::try_from(amount) {
RawInst::StackFree32 { amt }
} else {
unreachable!()
}
};
smallvec![inst.into()]
}
/// Generates the entire prologue for the function.
///
/// Note that this is different from other backends where it's not spread
/// out among a few individual functions. That's because the goal here is to
/// generate a single macro-instruction for the entire prologue in the most
/// common cases and we don't want to spread the logic over multiple
/// functions.
///
/// The general machinst methods are split to accommodate stack checks and
/// things like stack probes, all of which are empty on Pulley because
/// Pulley has its own stack check mechanism.
fn gen_prologue_frame_setup(
_call_conv: isa::CallConv,
_flags: &settings::Flags,
_isa_flags: &PulleyFlags,
frame_layout: &FrameLayout,
) -> SmallInstVec<Self::I> {
let mut insts = SmallVec::new();
let incoming_args_diff = frame_layout.tail_args_size - frame_layout.incoming_args_size;
if incoming_args_diff > 0 {
// Decrement SP by the amount of additional incoming argument space
// we need
insts.extend(Self::gen_sp_reg_adjust(-(incoming_args_diff as i32)));
}
let style = frame_layout.pulley_frame_style();
match &style {
FrameStyle::None => {}
FrameStyle::PulleyBasicSetup { frame_size } => {
insts.push(RawInst::PushFrame.into());
insts.extend(Self::gen_sp_reg_adjust(
-i32::try_from(*frame_size).unwrap(),
));
}
FrameStyle::PulleySetupAndSaveClobbers {
frame_size,
saved_by_pulley,
} => insts.push(
RawInst::PushFrameSave {
amt: *frame_size,
regs: pulley_interpreter::UpperRegSet::from_bitset(*saved_by_pulley),
}
.into(),
),
FrameStyle::Manual { frame_size } => insts.extend(Self::gen_sp_reg_adjust(
-i32::try_from(*frame_size).unwrap(),
)),
}
for (offset, ty, reg) in frame_layout.manually_managed_clobbers(&style) {
insts.push(
Inst::gen_store(Amode::SpOffset { offset }, reg, ty, MemFlags::trusted()).into(),
);
}
insts
}
/// Reverse of `gen_prologue_frame_setup`.
fn gen_epilogue_frame_restore(
_call_conv: isa::CallConv,
_flags: &settings::Flags,
_isa_flags: &PulleyFlags,
frame_layout: &FrameLayout,
) -> SmallInstVec<Self::I> {
let mut insts = SmallVec::new();
let style = frame_layout.pulley_frame_style();
// Restore clobbered registers that are manually managed in Cranelift.
for (offset, ty, reg) in frame_layout.manually_managed_clobbers(&style) {
insts.push(
Inst::gen_load(
Writable::from_reg(reg),
Amode::SpOffset { offset },
ty,
MemFlags::trusted(),
)
.into(),
);
}
// Perform the inverse of `gen_prologue_frame_setup`.
match &style {
FrameStyle::None => {}
FrameStyle::PulleyBasicSetup { frame_size } => {
insts.extend(Self::gen_sp_reg_adjust(i32::try_from(*frame_size).unwrap()));
insts.push(RawInst::PopFrame.into());
}
FrameStyle::PulleySetupAndSaveClobbers {
frame_size,
saved_by_pulley,
} => insts.push(
RawInst::PopFrameRestore {
amt: *frame_size,
regs: pulley_interpreter::UpperRegSet::from_bitset(*saved_by_pulley),
}
.into(),
),
FrameStyle::Manual { frame_size } => {
insts.extend(Self::gen_sp_reg_adjust(i32::try_from(*frame_size).unwrap()))
}
}
insts
}
fn gen_return(
_call_conv: isa::CallConv,
_isa_flags: &PulleyFlags,
frame_layout: &FrameLayout,
) -> SmallInstVec<Self::I> {
let mut insts = SmallVec::new();
// Handle final stack adjustments for the tail-call ABI.
if frame_layout.tail_args_size > 0 {
insts.extend(Self::gen_sp_reg_adjust(
frame_layout.tail_args_size.try_into().unwrap(),
));
}
insts.push(RawInst::Ret {}.into());
insts
}
fn gen_probestack(_insts: &mut SmallInstVec<Self::I>, _frame_size: u32) {
// Pulley doesn't implement stack probes since all stack pointer
// decrements are checked already.
}
fn gen_clobber_save(
_call_conv: isa::CallConv,
_flags: &settings::Flags,
_frame_layout: &FrameLayout,
) -> SmallVec<[Self::I; 16]> {
// Note that this is intentionally empty because everything necessary
// was already done in `gen_prologue_frame_setup`.
SmallVec::new()
}
fn gen_clobber_restore(
_call_conv: isa::CallConv,
_flags: &settings::Flags,
_frame_layout: &FrameLayout,
) -> SmallVec<[Self::I; 16]> {
// Intentionally empty as restores happen for Pulley in `gen_return`.
SmallVec::new()
}
fn gen_call(
dest: &CallDest,
_tmp: Writable<Reg>,
mut info: CallInfo<()>,
) -> SmallVec<[Self::I; 2]> {
match dest {
// "near" calls are pulley->pulley calls so they use a normal "call"
// opcode
CallDest::ExtName(name, RelocDistance::Near) => {
// The first four integer arguments to a call can be handled via
// special pulley call instructions. Assert here that
// `info.uses` is sorted in order and then take out x0-x3 if
// they're present and move them from `info.uses` to
// `info.dest.args` to be handled differently during register
// allocation.
let mut args = SmallVec::new();
info.uses.sort_by_key(|arg| arg.preg);
info.uses.retain(|arg| {
if arg.preg != x0() && arg.preg != x1() && arg.preg != x2() && arg.preg != x3()
{
return true;
}
args.push(XReg::new(arg.vreg).unwrap());
false
});
smallvec![Inst::Call {
info: Box::new(info.map(|()| PulleyCall {
name: name.clone(),
args,
}))
}
.into()]
}
// "far" calls are pulley->host calls so they use a different opcode
// which is lowered with a special relocation in the backend.
CallDest::ExtName(name, RelocDistance::Far) => {
smallvec![Inst::IndirectCallHost {
info: Box::new(info.map(|()| name.clone()))
}
.into()]
}
// Indirect calls are all assumed to be pulley->pulley calls
CallDest::Reg(reg) => {
smallvec![Inst::IndirectCall {
info: Box::new(info.map(|()| XReg::new(*reg).unwrap()))
}
.into()]
}
}
}
fn gen_memcpy<F: FnMut(Type) -> Writable<Reg>>(
_call_conv: isa::CallConv,
_dst: Reg,
_src: Reg,
_size: usize,
_alloc_tmp: F,
) -> SmallVec<[Self::I; 8]> {
todo!()
}
fn get_number_of_spillslots_for_value(
rc: RegClass,
_target_vector_bytes: u32,
_isa_flags: &PulleyFlags,
) -> u32 {
// Spill slots are the size of a "word" or a pointer, but Pulley
// registers are 8-byte for integers/floats regardless of pointer size.
// Calculate the number of slots necessary to store 8 bytes.
let slots_for_8bytes = match P::pointer_width() {
PointerWidth::PointerWidth32 => 2,
PointerWidth::PointerWidth64 => 1,
};
match rc {
// Int/float registers are 8-bytes
RegClass::Int | RegClass::Float => slots_for_8bytes,
// Vector registers are 16 bytes
RegClass::Vector => 2 * slots_for_8bytes,
}
}
fn get_machine_env(_flags: &settings::Flags, _call_conv: isa::CallConv) -> &MachineEnv {
static MACHINE_ENV: OnceLock<MachineEnv> = OnceLock::new();
MACHINE_ENV.get_or_init(create_reg_environment)
}
fn get_regs_clobbered_by_call(_call_conv_of_callee: isa::CallConv) -> PRegSet {
DEFAULT_CLOBBERS
}
fn compute_frame_layout(
_call_conv: isa::CallConv,
flags: &settings::Flags,
_sig: &Signature,
regs: &[Writable<RealReg>],
is_leaf: bool,
incoming_args_size: u32,
tail_args_size: u32,
fixed_frame_storage_size: u32,
outgoing_args_size: u32,
) -> FrameLayout {
let mut regs: Vec<Writable<RealReg>> = regs
.iter()
.cloned()
.filter(|r| DEFAULT_CALLEE_SAVES.contains(r.to_reg().into()))
.collect();
regs.sort_unstable();
// Compute clobber size.
let clobber_size = compute_clobber_size(®s);
// Compute linkage frame size.
let setup_area_size = if flags.preserve_frame_pointers()
|| !is_leaf
// The function arguments that are passed on the stack are addressed
// relative to the Frame Pointer.
|| incoming_args_size > 0
|| clobber_size > 0
|| fixed_frame_storage_size > 0
{
P::pointer_width().bytes() * 2 // FP, LR
} else {
0
};
FrameLayout {
incoming_args_size,
tail_args_size,
setup_area_size: setup_area_size.into(),
clobber_size,
fixed_frame_storage_size,
outgoing_args_size,
clobbered_callee_saves: regs,
}
}
fn gen_inline_probestack(
_insts: &mut SmallInstVec<Self::I>,
_call_conv: isa::CallConv,
_frame_size: u32,
_guard_size: u32,
) {
// Pulley doesn't need inline probestacks because it always checks stack
// decrements.
}
}
/// Different styles of management of fp/lr and clobbered registers.
///
/// This helps decide, depending on Cranelift settings and frame layout, what
/// macro instruction is used to setup the pulley frame.
enum FrameStyle {
/// No management is happening, fp/lr aren't saved by Pulley or Cranelift.
/// No stack is being allocated either.
None,
/// Pulley saves the fp/lr combo and then stack adjustments/clobbers are
/// handled manually.
PulleyBasicSetup { frame_size: u32 },
/// Pulley is managing the fp/lr combo, the stack size, and clobbered
/// X-class registers.
///
/// Note that `saved_by_pulley` is not the exhaustive set of clobbered
/// registers. It's only those that are part of the `PushFrameSave`
/// instruction.
PulleySetupAndSaveClobbers {
/// The size of the frame, including clobbers, that's being allocated.
frame_size: u16,
/// Registers that pulley is saving/restoring.
saved_by_pulley: ScalarBitSet<u16>,
},
/// Cranelift is manually managing everything, both clobbers and stack
/// increments/decrements.
///
/// Note that fp/lr are not saved in this mode.
Manual {
/// The size of the stack being allocated.
frame_size: u32,
},
}
/// Pulley-specific helpers when dealing with ABI code.
impl FrameLayout {
/// Whether or not this frame saves fp/lr.
fn setup_frame(&self) -> bool {
self.setup_area_size > 0
}
/// Returns the stack size allocated by this function, excluding incoming
/// tail args or the optional "setup area" of fp/lr.
fn stack_size(&self) -> u32 {
self.clobber_size + self.fixed_frame_storage_size + self.outgoing_args_size
}
/// Returns the style of frame being used for this function.
///
/// See `FrameStyle` for more information.
fn pulley_frame_style(&self) -> FrameStyle {
let saved_by_pulley = self.clobbered_xregs_saved_by_pulley();
match (
self.stack_size(),
self.setup_frame(),
saved_by_pulley.is_empty(),
) {
// No stack allocated, not saving fp/lr, no clobbers, nothing to do
(0, false, true) => FrameStyle::None,
// No stack allocated, saving fp/lr, no clobbers, so this is
// pulley-managed via push/pop_frame.
(0, true, true) => FrameStyle::PulleyBasicSetup { frame_size: 0 },
// Some stack is being allocated and pulley is managing fp/lr. Let
// pulley manage clobbered registers as well, regardless if they're
// present or not.
//
// If the stack is too large then `PulleyBasicSetup` is used
// otherwise we'll be pushing `PushFrameSave` and `PopFrameRestore`.
(frame_size, true, _) => match frame_size.try_into() {
Ok(frame_size) => FrameStyle::PulleySetupAndSaveClobbers {
frame_size,
saved_by_pulley,
},
Err(_) => FrameStyle::PulleyBasicSetup { frame_size },
},
// Some stack is being allocated, but pulley isn't managing fp/lr,
// so we're manually doing everything.
(frame_size, false, true) => FrameStyle::Manual { frame_size },
// If there's no frame setup and there's clobbered registers this
// technically should have already hit a case above, so panic here.
(_, false, false) => unreachable!(),
}
}
/// Returns the set of clobbered registers that Pulley is managing via its
/// macro instructions rather than the generated code.
fn clobbered_xregs_saved_by_pulley(&self) -> ScalarBitSet<u16> {
let mut clobbered: ScalarBitSet<u16> = ScalarBitSet::new();
// Pulley only manages clobbers if it's also managing fp/lr.
if !self.setup_frame() {
return clobbered;
}
let mut found_manual_clobber = false;
for reg in self.clobbered_callee_saves.iter() {
let r_reg = reg.to_reg();
// Pulley can only manage clobbers of integer registers at this
// time, float registers are managed manually.
//
// Also assert that all pulley-managed clobbers come first,
// otherwise the loop below in `manually_managed_clobbers` is
// incorrect.
if r_reg.class() == RegClass::Int {
assert!(!found_manual_clobber);
if let Some(offset) = r_reg.hw_enc().checked_sub(16) {
clobbered.insert(offset);
}
} else {
found_manual_clobber = true;
}
}
clobbered
}
/// Returns an iterator over the clobbers that Cranelift is managing, not
/// Pulley.
///
/// If this frame has clobbers then they're either saved by Pulley with
/// `FrameStyle::PulleySetupAndSaveClobbers`. Cranelift might need to manage
/// these registers depending on Cranelift settings. Cranelift also always
/// manages floating-point registers.
fn manually_managed_clobbers<'a>(
&'a self,
style: &'a FrameStyle,
) -> impl Iterator<Item = (i32, Type, Reg)> + 'a {
let mut offset = self.stack_size();
self.clobbered_callee_saves.iter().filter_map(move |reg| {
// Allocate space for this clobber no matter what. If pulley is
// managing this then we're just accounting for the pulley-saved
// registers as well. Note that all pulley-managed registers come
// first in the list here.
offset -= 8;
let r_reg = reg.to_reg();
let ty = match r_reg.class() {
RegClass::Int => {
// If this register is saved by pulley, skip this clobber.
if let FrameStyle::PulleySetupAndSaveClobbers {
saved_by_pulley, ..
} = style
{
if let Some(reg) = r_reg.hw_enc().checked_sub(16) {
if saved_by_pulley.contains(reg) {
return None;
}
}
}
I64
}
RegClass::Float => F64,
RegClass::Vector => unreachable!("no vector registers are callee-save"),
};
let offset = i32::try_from(offset).unwrap();
Some((offset, ty, Reg::from(reg.to_reg())))
})
}
}
impl<P> PulleyABICallSite<P>
where
P: PulleyTargetKind,
{
pub fn emit_return_call(
mut self,
ctx: &mut Lower<InstAndKind<P>>,
args: isle::ValueSlice,
_backend: &PulleyBackend<P>,
) {
let new_stack_arg_size =
u32::try_from(self.sig(ctx.sigs()).sized_stack_arg_space()).unwrap();
ctx.abi_mut().accumulate_tail_args_size(new_stack_arg_size);
// Put all arguments in registers and stack slots (within that newly
// allocated stack space).
self.emit_args(ctx, args);
self.emit_stack_ret_arg_for_tail_call(ctx);
let dest = self.dest().clone();
let uses = self.take_uses();
match dest {
CallDest::ExtName(name, RelocDistance::Near) => {
let info = Box::new(ReturnCallInfo {
dest: name,
uses,
new_stack_arg_size,
});
ctx.emit(Inst::ReturnCall { info }.into());
}
CallDest::ExtName(_name, RelocDistance::Far) => {
unimplemented!("return-call of a host function")
}
CallDest::Reg(callee) => {
let info = Box::new(ReturnCallInfo {
dest: XReg::new(callee).unwrap(),
uses,
new_stack_arg_size,
});
ctx.emit(Inst::ReturnIndirectCall { info }.into());
}
}
}
}
const DEFAULT_CALLEE_SAVES: PRegSet = PRegSet::empty()
// Integer registers.
.with(px_reg(16))
.with(px_reg(17))
.with(px_reg(18))
.with(px_reg(19))
.with(px_reg(20))
.with(px_reg(21))
.with(px_reg(22))
.with(px_reg(23))
.with(px_reg(24))
.with(px_reg(25))
.with(px_reg(26))
.with(px_reg(27))
.with(px_reg(28))
.with(px_reg(29))
.with(px_reg(30))
.with(px_reg(31))
// Float registers.
.with(pf_reg(16))
.with(pf_reg(17))
.with(pf_reg(18))
.with(pf_reg(19))
.with(pf_reg(20))
.with(pf_reg(21))
.with(pf_reg(22))
.with(pf_reg(23))
.with(pf_reg(24))
.with(pf_reg(25))
.with(pf_reg(26))
.with(pf_reg(27))
.with(pf_reg(28))
.with(pf_reg(29))
.with(pf_reg(30))
.with(pf_reg(31))
// Note: no vector registers are callee-saved.
;
fn compute_clobber_size(clobbers: &[Writable<RealReg>]) -> u32 {
let mut clobbered_size = 0;
for reg in clobbers {
match reg.to_reg().class() {
RegClass::Int => {
clobbered_size += 8;
}
RegClass::Float => {
clobbered_size += 8;
}
RegClass::Vector => unimplemented!("Vector Size Clobbered"),
}
}
align_to(clobbered_size, 16)
}
const DEFAULT_CLOBBERS: PRegSet = PRegSet::empty()
// Integer registers: the first 16 get clobbered.
.with(px_reg(0))
.with(px_reg(1))
.with(px_reg(2))
.with(px_reg(3))
.with(px_reg(4))
.with(px_reg(5))
.with(px_reg(6))
.with(px_reg(7))
.with(px_reg(8))
.with(px_reg(9))
.with(px_reg(10))
.with(px_reg(11))
.with(px_reg(12))
.with(px_reg(13))
.with(px_reg(14))
.with(px_reg(15))
// Float registers: the first 16 get clobbered.
.with(pf_reg(0))
.with(pf_reg(1))
.with(pf_reg(2))
.with(pf_reg(3))
.with(pf_reg(4))
.with(pf_reg(5))
.with(pf_reg(6))
.with(pf_reg(7))
.with(pf_reg(8))
.with(pf_reg(9))
.with(pf_reg(10))
.with(pf_reg(11))
.with(pf_reg(12))
.with(pf_reg(13))
.with(pf_reg(14))
.with(pf_reg(15))
// All vector registers get clobbered.
.with(pv_reg(0))
.with(pv_reg(1))
.with(pv_reg(2))
.with(pv_reg(3))
.with(pv_reg(4))
.with(pv_reg(5))
.with(pv_reg(6))
.with(pv_reg(7))
.with(pv_reg(8))
.with(pv_reg(9))
.with(pv_reg(10))
.with(pv_reg(11))
.with(pv_reg(12))
.with(pv_reg(13))
.with(pv_reg(14))
.with(pv_reg(15))
.with(pv_reg(16))
.with(pv_reg(17))
.with(pv_reg(18))
.with(pv_reg(19))
.with(pv_reg(20))
.with(pv_reg(21))
.with(pv_reg(22))
.with(pv_reg(23))
.with(pv_reg(24))
.with(pv_reg(25))
.with(pv_reg(26))
.with(pv_reg(27))
.with(pv_reg(28))
.with(pv_reg(29))
.with(pv_reg(30))
.with(pv_reg(31));
fn create_reg_environment() -> MachineEnv {
// Prefer caller-saved registers over callee-saved registers, because that
// way we don't need to emit code to save and restore them if we don't
// mutate them.
let preferred_regs_by_class: [Vec<PReg>; 3] = {
let x_registers: Vec<PReg> = (0..16).map(|x| px_reg(x)).collect();
let f_registers: Vec<PReg> = (0..16).map(|x| pf_reg(x)).collect();
let v_registers: Vec<PReg> = (0..32).map(|x| pv_reg(x)).collect();
[x_registers, f_registers, v_registers]
};
let non_preferred_regs_by_class: [Vec<PReg>; 3] = {
let x_registers: Vec<PReg> = (16..XReg::SPECIAL_START)
.map(|x| px_reg(x.into()))
.collect();
let f_registers: Vec<PReg> = (16..32).map(|x| pf_reg(x)).collect();
let v_registers: Vec<PReg> = vec![];
[x_registers, f_registers, v_registers]
};
MachineEnv {
preferred_regs_by_class,
non_preferred_regs_by_class,
fixed_stack_slots: vec![],
scratch_by_class: [None, None, None],
}
}