cmd/compile: switch to a register-based calling convention for Go functions

[aclements]

I propose that we switch the Go internal ABI (used between Go functions) from stack-based to register-based argument and result passing for Go ~1.16~ 1.17.

I lay out the details of our proposal and the work required in this document.

The ABI specification can be found here.

This was previously proposed in #18597, where @dr2chase did some excellent prototyping work that our new proposal builds on. With multiple ABI support, we’re now in a much better position to execute on this without breaking compatibility with the existing body of Go assembly code. I’ve opened this new issue to focus on discussion of our new proposal.

/cc @dr2chase @danscales @thanm @cherrymui @mknyszek @prattmic @randall77

An incomplete and evolving list of tasks:

• [X] Define ABI specification (CL, @aclements)
• [X] Add GOEXPERIMENT=regabi and asm macro (CL)
• [x] cmd/compile: late call lowering (CLs)
• [x] cmd/compile: implement register arguments (@dr2chase)
  • [x] spill around morestack call (CL)
  • [x] add temporary ABI pragma for testing (CL)
  • [x] add abstract argument registers (WIP CL)
  • [x] implement ABI rules (CL)
  • [x] use abstract argument registers
• [x] cmd/compile: implement register returns
  • [x] late return lowering (CL, @dr2chase)
  • [x] implement ABI rules (CL, @thanm)
• [x] cmd/compile: ABI wrappers (CL, old CL)
  • [x] cmd/link: mangle wrapper names for ELF (CL)
  • [x] cmd/link: mangle wrapper names for PE (CL)
  • [x] implement zeroing of X15 in ABI0->ABIInternal wrappers (CL)
• [X] cmd/asm: add syntax for runtime packages to reference ABIInternal symbols (CL)
  • [x] Mark runtime.{goexit,asyncPreempt}, reflect.{makeFuncStub,methodValueCall} definitions ABIInternal (CL)
  • [x] Mark reference to runtime.callbackWrap in sys_windows_amd64.s as ABIInternal (CL)
• [x] runtime: fix assembly closure calls (mcall) (@aclements, CL)
• [x] reflect: support calling ABIInternal (CL)
• [x] reflect: support MakeFunc and method value calls (CL)
• [x] go and defer statements
  • [x] cmd/compile: desugar go/defer to zero-argument (@thanm, CL)
  • [x] cmd/compile: fix defer recover() (@cherrymui, CL)
  • [x] cmd/compile: desugar go/defer with results (@cherrymui, CL)
• [X] cmd/cgo: decouple cgo callbacks from ABI (CL)
• [x] runtime: finalizer support (CL)
• [x] runtime: Windows callback support (CL)
• [x] cmd/internal/obj: introduce and use RegEntryTmp registers in prologue (@mknyszek, CL)
• [x] cmd/compile: generate correct ABI0 argument maps for body-less functions (see WriteFuncMap) (@dr2chase, CL)
• [x] cmd/compile: make cgo_unsafe_args generate an ABI0 function (or an ABIInternal-with-no-registers function) (@cherrymui, CL)
• [x] runtime: update debug call protocol (CL)
• [x] Fix ABI0/ABIInternal confusion in cgo (CL)
• [x] Run signature fuzzer and fix bugs (everyone)
• [x] Get all.bash to pass on Linux, Windows, and Darwin (everyone)

High-priority non-critical path

• [x] Export ABI information across package boundaries to eliminate more wrappers (@aclements, CL)
• [x] Make funcPC always return the "native" PC of a function (maybe also introduce ABIOther) (@cherrymui, CL)
• [x] Strip unnecessary ABIInternal annotations once funcPC is in
• [x] Tracebacks
  • [x] Define extended argument metadata format (in progress, @cherrymui)
  • [x] cmd/compile: produce extended argument metadata (@cherrymui, CL)
  • [x] runtime: consume extended argument metadata (@cherrymui, CL)
  • [x] cmd/compile,runtime: traceback metadata for spill slot liveness
• [x] DWARF
  • [x] cmd/compile: ensure arguments have appropriate DWARF locations
  • [x] Ensure GDB and Delve can find register arguments
  • ~[ ] cmd/compile: add a DWARF vendor attribute with function argument frame size (context)~
• [x] Wrapper mangling for non-critical arches
  • [x] cmd/link: mangle wrapper names for Mach-O (CL)
  • [x] cmd/link: mangle wrapper names for xcoff
  • [x] cmd/link: mangle wrapper names for Plan 9
• [x] runtime: simplify go/defer call paths
  • [x] assert that go/defer frames are zero sized (CL)
  • [x] replace defer reflectcall with a direct closure call (CL)
• [x] cmd/internal/obj: reject splittable ABIInternal functions without morestack spill info (e.g., asm functions) because we can't generate a correct morestack path (CL)
• [x] cmd/asm: don't reference args_stackmap in ABIInternal functions (because it's for ABI0) (CL)
• [x] Port performance-critical assembly to ABIInternal (see CL 296372 for valuable functions) (CL 308931, CL 310184)
• [x] math: fix overhead of assembly->Go calls CL 310331
• [x] Release notes
  • Mention undocumented behavior we've observed applications depending on that may have changed.
  • Using reflect.ValueOf(fn).Pointer() to get the PC of an assembly function will now return the PC of the ABI wrapper
  • Performance surprise: calling from assembly to Go is slightly more expensive.
  • Performance surprise: calling an assembly function via a closure is slightly more expensive.
  • Traceback format is much improved

Enabling steps

• [x] Enable regabiwrappers by default on amd64 (Linux, Windows, and Darwin)
• [x] Enable regabidefer
• [x] Enable regabireflect
• [x] Enable regabig
• [x] Enable regabiargs

Testing

• [ ] Function signature fuzzer (ongoing)
  • [ ] Static call, closure call, method call, interface call, reflect.{ValueOf(target),MakeFunc(x, target),Method(x)}.{Call,Interface}, called from defer, called from go, called as a finalizer
  • [x] {Big,Small} {argument,result} in {memory,registers} and {not addressed,addressed and not leaked to heap,addressed and leaked to heap}
  • [x] Use defer in a test function (check arguments, modify results)
  • [x] Pass pointers-to-stack and pointers-to-heap
  • [x] Cause result to move to heap
  • [x] runtime: add MoveStackOnNextCall, assertions for pointer-to-stack/pointer-to-heap, assertions for live/dead (@aclements, CL)
  • [x] Integrate runtime testing hook
  • [ ] Clean up fuzzer, and get into tools repo.
• [x] cmd/compile: revive clobberdead (CL)
• [x] cmd/compile: add clobberdeadreg (CL)

Post-MVP

• [x] Move g from TLS to register (CL)
  • [x] Use g register on non-Linux OSes (see CL)
• [x] Reserve and use X15 as zero register (CL)
  • ~[ ] Should this be X0 for more compact instruction encoding?~
• [ ] Port to other platforms
  • [x] amd64
  • [x] arm64
  • [ ] mips
  • [x] ppc
  • [ ] s390x
  • [x] risc-v
  • ~[ ] porting guide~
• [x] Eliminate legacy ABI paths for all platforms, even if they're using ABIInternal with 0 registers
• [ ] Eliminate spill slots?
• [ ] Pass pointer-shaped method receiver in context register (#44827)
• [x] runtime: simplify go/defer call paths
  • [x] simplify go/defer paths to assume zero args/results
  • [x] simplify special _defer allocator and remove special _defer cases from mallocgc (because _defer records will be fixed size).
• [ ] cmd/vet: add syntax and checks for register arguments and results in assembly like x+0(FP)
• [ ] cmd/vet: check argument size for ABIInternal functions and fix runtime asm declarations

Cleanup (can be done later)

• [ ] runtime: port all assembly -> Go calls to ABIInternal?
• [x] Eliminate ABIAlias support from cmd/compile, cmd/link, and object format
• [x] remove now-redundant fields from ssa.AuxCall
• [x] rationalize use of LocalsOffset/FrameOffset (for registers that use a link register, e.g. arm64, powerppc).
• [ ] cmd/compile: revisit abiutils.go and clean up the API (also reduce ABIConfig copying)
• [ ] cmd/internal/obj: generate stack maps in obj so we can better compact the morestack/body maps and reduce subtlety?
• [x] cmd/compile,runtime: re-enable passing arguments to go in runtime (context)
• [ ] cmd/compile: always attach ir.Func to function ir.Names (context)
• [x] cmd/compile: support ABIInternal for tail calls (context)
• [x] cmd/compile: once wrapping of defers is on for all arch/OS combinations, it should be possible to simplify or remove some of the openDefer processing in ssagen (notablly we could get rid of openDeferSave()
• ~[ ] runtime: think about an ABI-insensitive debugCall~ Delve uses DWARF information
• [ ] Fix up named slots & value association
• [ ] fix firstpos in ssa.go – sometimes it's late.

[aclements]

I've submitted the ABI spec. It may continue to evolve in small ways, but reflects what all of the code is converging around.

[zx2c4]

I've submitted the ABI spec. It may continue to evolve in small ways, but reflects what all of the code is converging around.

@aclements - In case you didn't see this, https://groups.google.com/g/golang-dev/c/2oHi8t1s0Z8 might be worth considering.

[gopherbot]

Change https://golang.org/cl/292709 mentions this issue: all: merge branch dev.regabi (d3cd4830ad) into master

[gopherbot]

Change https://golang.org/cl/260539 mentions this issue: cmd/compile: add AMD64 parameter register defs, Arg ops, plumb to ssa.Config

[gopherbot]

Change https://golang.org/cl/295789 mentions this issue: cmd/compile/internal-abi: use x87 mode, not MMX mode

[gopherbot]

Change https://golang.org/cl/295792 mentions this issue: cmd/compile/internal-abi: fix ABI0-equivalence for zero-sized values

[gopherbot]

Change https://golang.org/cl/295791 mentions this issue: cmd/compile/internal-abi: fix ABI0-equivalence for zero-sized values

[gopherbot]

Change https://golang.org/cl/295790 mentions this issue: cmd/compile/internal-abi: update internal ABI spec for g register

[aclements]

@aclements - In case you didn't see this, https://groups.google.com/g/golang-dev/c/2oHi8t1s0Z8 might be worth considering.

Thanks @zx2c4 , I just sent a fix. (I'm still catching up from being away for a few weeks.)

[gopherbot]

Change https://golang.org/cl/293390 mentions this issue: cmd/compile: plumb abi info into expandCalls

[gopherbot]

Change https://golang.org/cl/293391 mentions this issue: cmd/compile: change StaticCall to return a "Results"

[gopherbot]

Change https://golang.org/cl/293389 mentions this issue: cmd/compile: plumb abi info into ssagen/ssa

[gopherbot]

Change https://golang.org/cl/293396 mentions this issue: cmd/compile: handle aggregate OpArg in registers

[gopherbot]

Change https://golang.org/cl/293393 mentions this issue: cmd/compile: use abiutils for all rcvr/in/out frame offsets.

[gopherbot]

Change https://golang.org/cl/293394 mentions this issue: cmd/compile: retrieve Args from registers

[gopherbot]

Change https://golang.org/cl/293397 mentions this issue: cmd/compile: refactor out an almost-superfluous arg

[gopherbot]

Change https://golang.org/cl/293392 mentions this issue: cmd/compile: check frame offsets against abi

[gopherbot]

Change https://golang.org/cl/294429 mentions this issue: cmd/compile: pointer maps seem to be right for morestack

[gopherbot]

Change https://golang.org/cl/295309 mentions this issue: cmd/compile: register results seem to work

[gopherbot]

Change https://golang.org/cl/294410 mentions this issue: cmd/compile: register abi, morestack work and mole whacking

[gopherbot]

Change https://golang.org/cl/293889 mentions this issue: cmd/compile: pass register parameters to called function

[gopherbot]

Change https://golang.org/cl/293398 mentions this issue: cmd/compile: pass arguments as register parameters to StaticCall.

[gopherbot]

Change https://golang.org/cl/297637 mentions this issue: cmd/compile: better version of check frame offsets against abi

[gopherbot]

Change https://golang.org/cl/297911 mentions this issue: cmd/compile: make modified Aux type for OpArgXXXX pass ssa/check

[gopherbot]

Change https://golang.org/cl/299269 mentions this issue: reflect: add tests for reflect.Value.Call for the new ABI

[gopherbot]

Change https://golang.org/cl/300113 mentions this issue: runtime: support register ABI Go functions from Windows callbacks

[gopherbot]

Change https://golang.org/cl/298670 mentions this issue: reflect: add register ABI support for makeFuncStub and methodValueCall

[gopherbot]

Change https://golang.org/cl/300112 mentions this issue: runtime: support register ABI for finalizers

[gopherbot]

Change https://golang.org/cl/302049 mentions this issue: cmd/internal/objabi: centralize GOEXPERIMENT parsing

[gopherbot]

Change https://golang.org/cl/302050 mentions this issue: cmd/dist: build bootstrap without GOEXPERIMENT

[gopherbot]

Change https://golang.org/cl/302070 mentions this issue: all: explode GOEXPIREMENT=regabi into 5 sub-experiments

[gopherbot]

Change https://golang.org/cl/302109 mentions this issue: runtime: mark Windows' address-taken asm routines as ABIInternal

[gopherbot]

Change https://golang.org/cl/302071 mentions this issue: cmd/compile: (fixed) spill output parameters passed in registers as autos

[gopherbot]

Change https://golang.org/cl/302249 mentions this issue: cmd/compile: fix open defer of method call

[gopherbot]

Change https://golang.org/cl/302772 mentions this issue: test: make nosplit test invariant to ABI wrappers

[gopherbot]

Change https://golang.org/cl/303069 mentions this issue: runtime: call nanotimeQPC from nanotime1 without a wrapper

[gopherbot]

Change https://golang.org/cl/303070 mentions this issue: runtime: bypass ABI wrapper when calling needm on Windows

[gopherbot]

Change https://golang.org/cl/303314 mentions this issue: cmd/internal/obj/x86: use ABI scratch registers for WRAPPER prologue

[gopherbot]

Change https://golang.org/cl/303432 mentions this issue: cmd/compile: remove now-redundant AuxCall.args

[gopherbot]

Change https://golang.org/cl/303433 mentions this issue: cmd/compile: remove AuxCall.results, cleanup ssagen/ssa.go

[gopherbot]

Change https://golang.org/cl/304189 mentions this issue: cmd/compile: remove more dead code and data structures

[gopherbot]

Change https://golang.org/cl/304233 mentions this issue: cmd/compile: fix array case in types-for-register parameter

[gopherbot]

Change https://golang.org/cl/304549 mentions this issue: cmd/compile: update default ABI choices for calls and bodyless fn stack maps

[emarj]

@alexbrainman

@aclements I reckon there are very few instances of syscall.NewCallback usage. But they are used a lot. ... These two pretty much cover syscall.NewCallback usage. ... I reckon, that is about it. For Microsoft APIs (maybe google code for syscall.NewCallback or windows.NewCallback).

Actually there are other instances where NewCallback is used. One example is to interface with the Core Audio APIs. In this specific case, the support for callbacks with floating point argument is needed. See for example this callback. I think the situation may be similar in graphics APIs, as mentioned in this issue about syscalls https://github.com/golang/go/issues/37273

Would be great if we could get support for floats args (also) in callbacks.

PS: This is beyond the scope of this issue, but a full support of Windows syscall/callbacks could be useful since Microsoft is now trying to make APIs more accessible to other languages by describing them in metadata (announcement).

[alexbrainman]

@emarj

@alexbrainman

@aclements I reckon there are very few instances of syscall.NewCallback usage. But they are used a lot. ... These two pretty much cover syscall.NewCallback usage. ... I reckon, that is about it. For Microsoft APIs (maybe google code for syscall.NewCallback or windows.NewCallback).

Actually there are other instances where NewCallback is used.

No argument.

One example is to interface with the Core Audio APIs.

I did not know about that one. And I would argue that this API is rarely used comparing to building a Windows service or a Windows GUI.

Alex

[elagergren-spideroak]

These two pretty much cover syscall.NewCallback usage.

WinHTTP uses them too.

[gopherbot]

Change https://golang.org/cl/305271 mentions this issue: cmd/compile: set ir.Name.Func in more cases

[gopherbot]

Change https://golang.org/cl/305273 mentions this issue: cmd/compile: eliminate -abiwraplimit

[gopherbot]

Change https://golang.org/cl/305274 mentions this issue: cmd/compile: restructure ABI wrapper generation, export ABI

[gopherbot]

Change https://golang.org/cl/305272 mentions this issue: cmd/compile: assert that function values reference ABIInternal