Merge branch 'master' into no_std

.travis.yml

@@ -5,6 +5,7 @@ rust:
     - nightly
 matrix:
     allow_failures:
+        - rust: beta
         - rust: nightly
 dist: trusty
 sudo: false

README.rst

@@ -6,23 +6,52 @@ Cretonne is a low-level retargetable code generator. It translates a `target-ind
 intermediate representation <https://cretonne.readthedocs.io/en/latest/langref.html>`_ into executable
 machine code.
 
-*This is a work in progress that is not yet functional.*
-
 .. image:: https://readthedocs.org/projects/cretonne/badge/?version=latest
     :target: https://cretonne.readthedocs.io/en/latest/?badge=latest
     :alt: Documentation Status
 
-.. image:: https://travis-ci.org/Cretonne/cretonne.svg?branch=master
-    :target: https://travis-ci.org/Cretonne/cretonne
+.. image:: https://travis-ci.org/cretonne/cretonne.svg?branch=master
+    :target: https://travis-ci.org/cretonne/cretonne
     :alt: Build Status
 
-.. image:: https://badges.gitter.im/Cretonne/cretonne.png
-    :target: https://gitter.im/Cretonne/Lobby/~chat
+.. image:: https://badges.gitter.im/cretonne/cretonne.png
+    :target: https://gitter.im/cretonne/Lobby/~chat
     :alt: Gitter chat
 
 For more information, see `the documentation
 <https://cretonne.readthedocs.io/en/latest/?badge=latest>`_.
 
+Status
+------
+
+Cretonne currently supports enough functionality to run a wide variety of
+programs, including all the functionality needed to execute WebAssembly MVP
+functions, although it needs to be used within an external WebAssembly
+embedding to be part of a complete WebAssembly implementation.
+
+The x86-64 backend is currently the most complete and stable; other
+architectures are in various stages of development. Cretonne currently supports
+the System V AMD64 ABI calling convention used on many platforms, but does not
+yet support the Windows x64 calling convention. The performance of code
+produced by Cretonne is not yet impressive, though we have plans to fix that.
+
+The core codegen crates have minimal dependencies, and do not require any host
+floating-point support. Support for `no_std` mode in the core codegen crates is
+`in development <https://github.com/cretonne/cretonne/tree/no_std>`_.
+
+Cretonne does not yet perform mitigations for Spectre or related security
+issues, though it may do so in the future. It does not currently make any
+security-relevant instruction timing guarantees. It has seen a fair amount
+of testing and fuzzing, although more work is needed before it would be
+ready for a production use case.
+
+Cretonne's APIs are not yet stable.
+
+Cretonne currently supports Rust 1.22.1 and later. We intend to always support
+the latest *stable* Rust. And, we currently support the version of Rust in the
+latest Ubuntu LTS, although whether we will always do so is not yet determined.
+Cretonne requires Python 2.7 or Python 3 to build.
+
 Planned uses
 ------------
 

cranelift/.gitignore

@@ -10,3 +10,4 @@ Cargo.lock
 cretonne.dbg*
 .mypy_cache
 rusty-tags.*
+docs/_build

cranelift/Cargo.toml

@@ -1,11 +1,11 @@
 [package]
 name = "cretonne-tools"
 authors = ["The Cretonne Project Developers"]
-version = "0.4.1"
-description = "Binaries for testing the Cretonne library"
+version = "0.5.0"
+description = "Binaries for testing the Cretonne libraries"
 license = "Apache-2.0"
 documentation = "https://cretonne.readthedocs.io/"
-repository = "https://github.com/Cretonne/cretonne"
+repository = "https://github.com/cretonne/cretonne"
 publish = false
 
 [[bin]]
@@ -13,12 +13,16 @@ name = "cton-util"
 path = "src/cton-util.rs"
 
 [dependencies]
-cretonne = { path = "lib/cretonne", version = "0.4.1" }
-cretonne-reader = { path = "lib/reader", version = "0.4.1" }
-cretonne-frontend = { path = "lib/frontend", version = "0.4.1" }
-cretonne-wasm = { path = "lib/wasm", version = "0.4.1" }
-cretonne-native = { path = "lib/native", version = "0.4.1" }
-cretonne-filetests = { path = "lib/filetests", version = "0.4.1" }
+cretonne-codegen = { path = "lib/codegen", version = "0.5.0" }
+cretonne-reader = { path = "lib/reader", version = "0.5.0" }
+cretonne-frontend = { path = "lib/frontend", version = "0.5.0" }
+cretonne-wasm = { path = "lib/wasm", version = "0.5.0" }
+cretonne-native = { path = "lib/native", version = "0.5.0" }
+cretonne-filetests = { path = "lib/filetests", version = "0.5.0" }
+cretonne-module = { path = "lib/module", version = "0.5.0" }
+cretonne-faerie = { path = "lib/faerie", version = "0.5.0" }
+cretonne-simplejit = { path = "lib/simplejit", version = "0.5.0" }
+cretonne = { path = "lib/umbrella", version = "0.5.0" }
 filecheck = "0.2.1"
 docopt = "0.8.0"
 serde = "1.0.8"
@@ -30,7 +34,7 @@ term = "0.5.1"
 
 # Enable debug assertions and parallel compilation when building cretonne-tools
 # since they are for testing and development mostly. This doesn't affect the
-# flags used to build the Cretonne crate when used as a dependency.
+# flags used to build the cretonne-* crates when used as a dependency.
 [profile.release]
 opt-level = 2
 debug-assertions = true

cranelift/docs/Makefile

@@ -14,7 +14,7 @@ help:
 	@$(SPHINXBUILD) -M help "$(SOURCEDIR)" "$(BUILDDIR)" $(SPHINXOPTS) $(O)
 
 autohtml: html
-	$(SPHINXABUILD) -z ../lib/cretonne/meta --ignore '.*' -b html -E $(ALLSPHINXOPTS) $(BUILDDIR)/html
+	$(SPHINXABUILD) -z ../lib/codegen/meta --ignore '.*' -b html -E $(ALLSPHINXOPTS) $(BUILDDIR)/html
 
 .PHONY: help Makefile
 

cranelift/docs/callex.cton

@@ -1,7 +1,7 @@
 test verifier
 
 function %gcd(i32 uext, i32 uext) -> i32 uext system_v {
-    fn1 = function %divmod(i32 uext, i32 uext) -> i32 uext, i32 uext
+    fn1 = %divmod(i32 uext, i32 uext) -> i32 uext, i32 uext
 
 ebb1(v1: i32, v2: i32):
     brz v2, ebb2

cranelift/docs/compare-llvm.rst

@@ -6,8 +6,8 @@ Cretonne compared to LLVM
 a set of C++ libraries. It can be used to build both JIT compilers and static
 compilers like `Clang <https://clang.llvm.org>`_, and it is deservedly very
 popular. `Chris Lattner's chapter about LLVM
-<http://www.aosabook.org/en/llvm.html>`_ in the `Architecture of Open Source
-Applications <http://aosabook.org/en/index.html>`_ book gives an excellent
+<https://www.aosabook.org/en/llvm.html>`_ in the `Architecture of Open Source
+Applications <https://aosabook.org/en/index.html>`_ book gives an excellent
 overview of the architecture and design of LLVM.
 
 Cretonne and LLVM are superficially similar projects, so it is worth
@@ -174,7 +174,7 @@ is emitted, there are opcodes for every native instruction that can be
 generated. There is a lot of overlap between different ISAs, so for example the
 :cton:inst:`iadd_imm` instruction is used by every ISA that can add an
 immediate integer to a register. A simple RISC ISA like RISC-V can be defined
-with only shared instructions, while an Intel ISA needs a number of specific
+with only shared instructions, while x86 needs a number of specific
 instructions to model addressing modes.
 
 Undefined behavior

cranelift/docs/conf.py

@@ -23,7 +23,7 @@ sys.path.insert(0, os.path.abspath('.'))
 
 # Also add the meta directory to sys.path so autodoc can find the Cretonne meta
 # language definitions.
-sys.path.insert(0, os.path.abspath('../lib/cretonne/meta'))
+sys.path.insert(0, os.path.abspath('../lib/codegen/meta'))
 
 # -- General configuration ------------------------------------------------
 

cranelift/docs/langref.rst

@@ -48,8 +48,7 @@ A ``.cton`` file consists of a sequence of independent function definitions:
 
 .. productionlist::
     function_list : { function }
-    function      : function_spec "{" preamble function_body "}"
-    function_spec : "function" function_name signature
+    function      : "function" function_name signature "{" preamble function_body "}"
     preamble      : { preamble_decl }
     function_body : { extended_basic_block }
 
@@ -75,7 +74,7 @@ SSA values: In the entry block, ``v4`` is the initial value. In the loop block
 variable during each iteration. Finally, ``v12`` is computed as the induction
 variable value for the next iteration.
 
-The `cton_frontend` crate contains utilities for translating from programs
+The `cretonne_frontend` crate contains utilities for translating from programs
 containing multiple assignments to the same variables into SSA form for
 Cretonne :term:`IR`.
 
@@ -409,10 +408,14 @@ compilers.
 Functions that are called directly must be declared in the :term:`function
 preamble`:
 
-.. inst:: FN = function NAME signature
+.. inst:: FN = [colocated] NAME signature
 
     Declare a function so it can be called directly.
 
+    If the colocated keyword is present, the symbol's definition will be
+    defined along with the current function, such that it can use more
+    efficient addressing.
+
     :arg NAME: Name of the function, passed to the linker for resolution.
     :arg signature: Function signature. See below.
     :result FN: A function identifier that can be used with :inst:`call`.
@@ -570,13 +573,17 @@ runtime data structures.
                  variable.
     :result GV: Global variable.
 
-.. inst:: GV = globalsym name
+.. inst:: GV = [colocated] globalsym name
 
     Declare a global variable at a symbolic address.
 
     The address of GV is symbolic and will be assigned a relocation, so that
     it can be resolved by a later linking phase.
 
+    If the colocated keyword is present, the symbol's definition will be
+    defined along with the current function, such that it can use more
+    efficient addressing.
+
     :arg name: External name.
     :result GV: Global variable.
 
@@ -998,20 +1005,20 @@ ISA-specific instructions
 Target ISAs can define supplemental instructions that do not make sense to
 support generally.
 
-Intel
+x86
 -----
 
-Instructions that can only be used by the Intel target ISA.
+Instructions that can only be used by the x86 target ISA.
 
-.. autoinst:: isa.intel.instructions.sdivmodx
-.. autoinst:: isa.intel.instructions.udivmodx
-.. autoinst:: isa.intel.instructions.cvtt2si
-.. autoinst:: isa.intel.instructions.fmin
-.. autoinst:: isa.intel.instructions.fmax
-.. autoinst:: isa.intel.instructions.bsf
-.. autoinst:: isa.intel.instructions.bsr
-.. autoinst:: isa.intel.instructions.push
-.. autoinst:: isa.intel.instructions.pop
+.. autoinst:: isa.x86.instructions.sdivmodx
+.. autoinst:: isa.x86.instructions.udivmodx
+.. autoinst:: isa.x86.instructions.cvtt2si
+.. autoinst:: isa.x86.instructions.fmin
+.. autoinst:: isa.x86.instructions.fmax
+.. autoinst:: isa.x86.instructions.bsf
+.. autoinst:: isa.x86.instructions.bsr
+.. autoinst:: isa.x86.instructions.push
+.. autoinst:: isa.x86.instructions.pop
 
 Instruction groups
 ==================
@@ -1023,7 +1030,7 @@ group.
 
 Target ISAs may define further instructions in their own instruction groups:
 
-.. autoinstgroup:: isa.intel.instructions.GROUP
+.. autoinstgroup:: isa.x86.instructions.GROUP
 
 Implementation limits
 =====================

cranelift/docs/metaref.rst

@@ -11,7 +11,7 @@ domain specific language embedded in Python. This document describes the Python
 modules that form the embedded DSL.
 
 The meta language descriptions are Python modules under the
-:file:`lib/cretonne/meta` directory. The descriptions are processed in two
+:file:`lib/codegen/meta` directory. The descriptions are processed in two
 steps:
 
 1. The Python modules are imported. This has the effect of building static data
@@ -23,8 +23,8 @@ steps:
    constant tables.
 
 The main driver for this source code generation process is the
-:file:`lib/cretonne/meta/build.py` script which is invoked as part of the build
-process if anything in the :file:`lib/cretonne/meta` directory has changed
+:file:`lib/codegen/meta/build.py` script which is invoked as part of the build
+process if anything in the :file:`lib/codegen/meta` directory has changed
 since the last build.
 
 
@@ -38,7 +38,7 @@ of code generation. Each setting is defined in the meta language so a compact
 and consistent Rust representation can be generated. Shared settings are defined
 in the :mod:`base.settings` module. Some settings are specific to a target ISA,
 and defined in a :file:`settings.py` module under the appropriate
-:file:`lib/cretonne/meta/isa/*` directory.
+:file:`lib/codegen/meta/isa/*` directory.
 
 Settings can take boolean on/off values, small numbers, or explicitly enumerated
 symbolic values. Each type is represented by a sub-class of :class:`Setting`:
@@ -400,7 +400,7 @@ Fixed register operands
 -----------------------
 
 Some instructions use hard-coded input and output registers for some value
-operands. An example is the ``pblendvb`` Intel SSE instruction which takes one
+operands. An example is the ``pblendvb`` x86 SSE instruction which takes one
 of its three value operands in the hard-coded ``%xmm0`` register::
 
     XMM0 = FPR[0]
@@ -433,13 +433,13 @@ architectures. Each ISA is represented by a :py:class:`cdsl.isa.TargetISA` insta
 .. autoclass:: TargetISA
 
 The definitions for each supported target live in a package under
-:file:`lib/cretonne/meta/isa`.
+:file:`lib/codegen/meta/isa`.
 
 .. automodule:: isa
     :members:
 
 .. automodule:: isa.riscv
-.. automodule:: isa.intel
+.. automodule:: isa.x86
 .. automodule:: isa.arm32
 .. automodule:: isa.arm64
 

cranelift/docs/regalloc.rst

@@ -79,7 +79,7 @@ Different register banks
 
 Instructions with fixed operands
     Some instructions use a fixed register for an operand. This happens on the
-    Intel ISAs:
+    x86 ISAs:
 
     - Dynamic shift and rotate instructions take the shift amount in CL.
     - Division instructions use RAX and RDX for both input and output operands.
@@ -109,7 +109,7 @@ ABI boundaries
 Aliasing registers
     Different registers sometimes share the same bits in the register bank.
     This can make it difficult to measure register pressure. For example, the
-    Intel registers RAX, EAX, AX, AL, and AH overlap.
+    x86 registers RAX, EAX, AX, AL, and AH overlap.
 
     If only one of the aliasing registers can be used at a time, the aliasing
     doesn't cause problems since the registers can simply be counted as one

cranelift/docs/testing.rst

@@ -51,7 +51,7 @@ tested::
     //!
     //! # Example
     //! ```
-    //! use cretonne::settings::{self, Configurable};
+    //! use cretonne_codegen::settings::{self, Configurable};
     //!
     //! let mut b = settings::builder();
     //! b.set("opt_level", "fastest");
@@ -73,9 +73,9 @@ test.
 
 These tests are usually found in the :file:`tests` top-level directory where
 they have access to all the crates in the Cretonne repository. The
-:file:`lib/cretonne` and :file:`lib/reader` crates have no external
+:file:`lib/codegen` and :file:`lib/reader` crates have no external
 dependencies, which can make testing tedious. Integration tests that don't need
-to depend on other crates can be placed in :file:`lib/cretonne/tests` and
+to depend on other crates can be placed in :file:`lib/codegen/tests` and
 :file:`lib/reader/tests`.
 
 File tests
@@ -109,7 +109,7 @@ header:
     isa_spec      : "isa" isa_name { `option` } "\n"
 
 The options given on the ``isa`` line modify the ISA-specific settings defined in
-:file:`lib/cretonne/meta/isa/*/settings.py`.
+:file:`lib/codegen/meta/isa/*/settings.py`.
 
 All types of tests allow shared Cretonne settings to be modified:
 
@@ -119,7 +119,7 @@ All types of tests allow shared Cretonne settings to be modified:
     option        : flag | setting "=" value
 
 The shared settings available for all target ISAs are defined in
-:file:`lib/cretonne/meta/base/settings.py`.
+:file:`lib/codegen/meta/base/settings.py`.
 
 The ``set`` lines apply settings cumulatively::
 

cranelift/filetests/isa/intel/binary64-pic.cton

@@ -1,54 +0,0 @@
-; binary emission of 64-bit code.
-test binemit
-set is_64bit
-set is_compressed
-set is_pic
-isa intel haswell
-
-; The binary encodings can be verified with the command:
-;
-;   sed -ne 's/^ *; asm: *//p' filetests/isa/intel/binary64-pic.cton | llvm-mc -show-encoding -triple=x86_64
-;
-
-; Tests for i64 instructions.
-function %I64() {
-    sig0 = ()
-    fn0 = function %foo()
-
-    gv0 = globalsym %some_gv
-
-    ; Use incoming_arg stack slots because they won't be relocated by the frame
-    ; layout.
-    ss0 = incoming_arg 8, offset 0
-    ss1 = incoming_arg 1024, offset -1024
-    ss2 = incoming_arg 1024, offset -2048
-    ss3 = incoming_arg 8, offset -2056
-
-ebb0:
-
-    ; asm: call foo@PLT
-    call fn0()                                  ; bin: e8 PLTRel4(%foo-4) 00000000
-
-    ; asm: mov 0x0(%rip), %rax
-    [-,%rax]            v0 = func_addr.i64 fn0        ; bin: 48 8b 05 GOTPCRel4(%foo-4) 00000000
-    ; asm: mov 0x0(%rip), %rsi
-    [-,%rsi]            v1 = func_addr.i64 fn0        ; bin: 48 8b 35 GOTPCRel4(%foo-4) 00000000
-    ; asm: mov 0x0(%rip), %r10
-    [-,%r10]            v2 = func_addr.i64 fn0        ; bin: 4c 8b 15 GOTPCRel4(%foo-4) 00000000
-
-    ; asm: call *%rax
-    call_indirect sig0, v0()                  ; bin: ff d0
-    ; asm: call *%rsi
-    call_indirect sig0, v1()                  ; bin: ff d6
-    ; asm: call *%r10
-    call_indirect sig0, v2()                  ; bin: 41 ff d2
-
-    ; asm: mov 0x0(%rip), %rcx
-    [-,%rcx]            v3 = globalsym_addr.i64 gv0    ; bin: 48 8b 0d GOTPCRel4(%some_gv-4) 00000000
-    ; asm: mov 0x0(%rip), %rsi
-    [-,%rsi]            v4 = globalsym_addr.i64 gv0    ; bin: 48 8b 35 GOTPCRel4(%some_gv-4) 00000000
-    ; asm: mov 0x0(%rip), %r10
-    [-,%r10]            v5 = globalsym_addr.i64 gv0    ; bin: 4c 8b 15 GOTPCRel4(%some_gv-4) 00000000
-
-    return
-}

cranelift/filetests/isa/intel/prologue-epilogue.cton

@@ -1,32 +0,0 @@
-test compile
-set is_64bit
-set is_compressed
-isa intel haswell
-
-function %foo() {
-    ss0 = explicit_slot 168
-ebb0:
-    return
-}
-
-; check: function %foo(i64 fp [%rbp], i64 csr [%rbx], i64 csr [%r12], i64 csr [%r13], i64 csr [%r14], i64 csr [%r15]) -> i64 fp [%rbp], i64 csr [%rbx], i64 csr [%r12], i64 csr [%r13], i64 csr [%r14], i64 csr [%r15] system_v {
-; nextln:     ss0 = explicit_slot 168, offset -224
-; nextln:     ss1 = incoming_arg 56, offset -56
-; check: ebb0(v0: i64 [%rbp], v1: i64 [%rbx], v2: i64 [%r12], v3: i64 [%r13], v4: i64 [%r14], v5: i64 [%r15]):
-; nextln:     x86_push v0
-; nextln:     copy_special %rsp -> %rbp
-; nextln:     x86_push v1
-; nextln:     x86_push v2
-; nextln:     x86_push v3
-; nextln:     x86_push v4
-; nextln:     x86_push v5
-; nextln:     adjust_sp_imm -168
-; nextln:     adjust_sp_imm 168
-; nextln:     v11 = x86_pop.i64
-; nextln:     v10 = x86_pop.i64
-; nextln:     v9 = x86_pop.i64
-; nextln:     v8 = x86_pop.i64
-; nextln:     v7 = x86_pop.i64
-; nextln:     v6 = x86_pop.i64
-; nextln:     return v6, v7, v8, v9, v10, v11
-; nextln: }

cranelift/filetests/isa/riscv/binary32.cton

@@ -4,7 +4,7 @@ isa riscv
 
 function %RV32I(i32 link [%x1]) -> i32 link [%x1] {
     sig0 = ()
-    fn0 = function %foo()
+    fn0 = %foo()
 
 ebb0(v9999: i32):
     [-,%x10]            v1 = iconst.i32 1

cranelift/filetests/isa/riscv/legalize-abi.cton

@@ -17,8 +17,8 @@ ebb0(v0: i64):
 }
 
 function %split_call_arg(i32) {
-    fn1 = function %foo(i64)
-    fn2 = function %foo(i32, i64)
+    fn1 = %foo(i64)
+    fn2 = %foo(i32, i64)
 ebb0(v0: i32):
     v1 = uextend.i64 v0
     call fn1(v1)
@@ -30,7 +30,7 @@ ebb0(v0: i32):
 }
 
 function %split_ret_val() {
-    fn1 = function %foo() -> i64
+    fn1 = %foo() -> i64
 ebb0:
     v1 = call fn1()
     ; check: ebb0($(link=$V): i32):
@@ -45,7 +45,7 @@ ebb1(v10: i64):
 
 ; First return value is fine, second one is expanded.
 function %split_ret_val2() {
-    fn1 = function %foo() -> i32, i64
+    fn1 = %foo() -> i32, i64
 ebb0:
     v1, v2 = call fn1()
     ; check: ebb0($(link=$V): i32):
@@ -70,7 +70,7 @@ ebb0(v1: i8, v2: i8, v3: i8):
 
 ; Function produces single return value, still need to copy.
 function %ext_ret_val() {
-    fn1 = function %foo() -> i8 sext
+    fn1 = %foo() -> i8 sext
 ebb0:
     v1 = call fn1()
     ; check: ebb0($V: i32):
@@ -124,7 +124,7 @@ ebb0(v0: i32, v1: f32x2):
 ; Call a function that takes arguments on the stack.
 function %stack_args(i32) {
     ; check: $(ss0=$SS) = outgoing_arg 4
-    fn1 = function %foo(i64, i64, i64, i64, i32)
+    fn1 = %foo(i64, i64, i64, i64, i32)
 ebb0(v0: i32):
     v1 = iconst.i64 1
     call fn1(v1, v1, v1, v1, v0)

cranelift/filetests/isa/riscv/parse-encoding.cton

@@ -27,9 +27,9 @@ function %parse_encoding(i32 [%x5]) -> i32 [%x10] {
     ; check: sig5 = () -> f32 [0] system_v
 
     ; function + signature
-    fn0 = function %bar(i32 [%x10]) -> b1 [%x10] system_v
+    fn0 = %bar(i32 [%x10]) -> b1 [%x10] system_v
     ; check: sig6 = (i32 [%x10]) -> b1 [%x10] system_v
-    ; nextln: fn0 = sig6 %bar
+    ; nextln: fn0 = %bar sig6
 
 ebb0(v0: i32):
     return v0

cranelift/filetests/isa/riscv/verify-encoding.cton

@@ -2,7 +2,7 @@ test verifier
 isa riscv
 
 function %RV32I(i32 link [%x1]) -> i32 link [%x1] {
-    fn0 = function %foo()
+    fn0 = %foo()
 
 ebb0(v9999: i32):
     ; iconst.i32 needs legalizing, so it should throw a
@@ -11,7 +11,7 @@ ebb0(v9999: i32):
 }
 
 function %RV32I(i32 link [%x1]) -> i32 link [%x1] {
-    fn0 = function %foo()
+    fn0 = %foo()
 
 ebb0(v9999: i32):
     v1 = iconst.i32 1

cranelift/filetests/isa/x86/abi-bool.cton

@@ -1,6 +1,6 @@
 test compile
 set is_64bit=1
-isa intel haswell
+isa x86 haswell
 
 function %foo(i64, i64, i64, i32) -> b1 system_v {
 ebb3(v0: i64, v1: i64, v2: i64, v3: i32):

cranelift/filetests/isa/x86/abi32.cton

@@ -1,6 +1,6 @@
 ; Test the legalization of function signatures.
 test legalizer
-isa intel
+isa x86
 
 ; regex: V=v\d+
 

cranelift/filetests/isa/x86/abi64.cton

@@ -1,7 +1,7 @@
 ; Test the legalization of function signatures.
 test legalizer
 set is_64bit
-isa intel
+isa x86
 
 ; regex: V=v\d+
 
@@ -21,7 +21,7 @@ ebb0:
 
 function %pass_stack_int64(i64, i64, i64, i64, i64, i64, i64, i64, i64, i64, i64, i64, i64, i64, i64, i64, i64, i64, i64, i64, i64 vmctx) spiderwasm {
     sig0 = (i64, i64, i64, i64, i64, i64, i64, i64, i64, i64, i64, i64, i64, i64, i64, i64, i64, i64, i64, i64, i64 vmctx) spiderwasm
-    fn0 = sig0 u0:0
+    fn0 = u0:0 sig0
 
 ebb0(v0: i64, v1: i64, v2: i64, v3: i64, v4: i64, v5: i64, v6: i64, v7: i64, v8: i64, v9: i64, v10: i64, v11: i64, v12: i64, v13: i64, v14: i64, v15: i64, v16: i64, v17: i64, v18: i64, v19: i64, v20: i64):
     call fn0(v0, v1, v2, v3, v4, v5, v6, v7, v8, v9, v10, v11, v12, v13, v14, v15, v16, v17, v18, v19, v20)

cranelift/filetests/isa/x86/allones_funcaddrs32.cton

@@ -2,17 +2,17 @@
 test binemit
 set is_compressed
 set allones_funcaddrs
-isa intel haswell
+isa x86 haswell
 
 ; The binary encodings can be verified with the command:
 ;
-;   sed -ne 's/^ *; asm: *//p' filetests/isa/intel/allones_funcaddrs32.cton | llvm-mc -show-encoding -triple=i386
+;   sed -ne 's/^ *; asm: *//p' filetests/isa/x86/allones_funcaddrs32.cton | llvm-mc -show-encoding -triple=i386
 ;
 
 ; Tests from binary32.cton affected by allones_funcaddrs.
 function %I32() {
     sig0 = ()
-    fn0 = function %foo()
+    fn0 = %foo()
 
 ebb0:
 

cranelift/filetests/isa/x86/allones_funcaddrs64.cton

@@ -3,17 +3,17 @@ test binemit
 set is_64bit
 set is_compressed
 set allones_funcaddrs
-isa intel haswell
+isa x86 haswell
 
 ; The binary encodings can be verified with the command:
 ;
-;   sed -ne 's/^ *; asm: *//p' filetests/isa/intel/allones_funcaddrs64.cton | llvm-mc -show-encoding -triple=x86_64
+;   sed -ne 's/^ *; asm: *//p' filetests/isa/x86/allones_funcaddrs64.cton | llvm-mc -show-encoding -triple=x86_64
 ;
 
 ; Tests from binary64.cton affected by allones_funcaddrs.
 function %I64() {
     sig0 = ()
-    fn0 = function %foo()
+    fn0 = %foo()
 
 ebb0:
 

cranelift/filetests/isa/x86/baseline_clz_ctz_popcount.cton

@@ -1,7 +1,7 @@
 
 test compile
 set is_64bit
-isa intel baseline
+isa x86 baseline
 
 
 ; clz/ctz on 64 bit operands

cranelift/filetests/isa/x86/baseline_clz_ctz_popcount_encoding.cton

@@ -2,11 +2,11 @@
 test binemit
 set is_64bit
 set is_compressed
-isa intel baseline
+isa x86 baseline
 
 ; The binary encodings can be verified with the command:
 ;
-;   sed -ne 's/^ *; asm: *//p' filetests/isa/intel/baseline_clz_ctz_popcount_encoding.cton | llvm-mc -show-encoding -triple=x86_64
+;   sed -ne 's/^ *; asm: *//p' filetests/isa/x86/baseline_clz_ctz_popcount_encoding.cton | llvm-mc -show-encoding -triple=x86_64
 ;
 
 function %Foo() {

cranelift/filetests/isa/x86/binary32-float.cton

@@ -1,10 +1,10 @@
 ; Binary emission of 32-bit floating point code.
 test binemit
-isa intel haswell
+isa x86 haswell
 
 ; The binary encodings can be verified with the command:
 ;
-;   sed -ne 's/^ *; asm: *//p' filetests/isa/intel/binary32-float.cton | llvm-mc -show-encoding -triple=i386
+;   sed -ne 's/^ *; asm: *//p' filetests/isa/x86/binary32-float.cton | llvm-mc -show-encoding -triple=i386
 ;
 
 function %F32() {

cranelift/filetests/isa/x86/binary32.cton

@@ -1,16 +1,16 @@
 ; binary emission of x86-32 code.
 test binemit
 set is_compressed
-isa intel haswell
+isa x86 haswell
 
 ; The binary encodings can be verified with the command:
 ;
-;   sed -ne 's/^ *; asm: *//p' filetests/isa/intel/binary32.cton | llvm-mc -show-encoding -triple=i386
+;   sed -ne 's/^ *; asm: *//p' filetests/isa/x86/binary32.cton | llvm-mc -show-encoding -triple=i386
 ;
 
 function %I32() {
     sig0 = ()
-    fn0 = function %foo()
+    fn0 = %foo()
 
     gv0 = globalsym %some_gv
 
@@ -352,7 +352,7 @@ ebb0:
     [-,%rsi]             v351 = bint.i32 v301   ; bin: 0f b6 f2
 
     ; asm: call foo
-    call fn0()                                  ; bin: e8 PCRel4(%foo) 00000000
+    call fn0()                                  ; bin: e8 PCRel4(%foo-4) 00000000
 
     ; asm: movl $0, %ecx
     [-,%rcx]            v400 = func_addr.i32 fn0        ; bin: b9 Abs4(%foo) 00000000

cranelift/filetests/isa/x86/binary64-float.cton

@@ -2,11 +2,11 @@
 test binemit
 set is_64bit
 set is_compressed
-isa intel haswell
+isa x86 haswell
 
 ; The binary encodings can be verified with the command:
 ;
-;   sed -ne 's/^ *; asm: *//p' filetests/isa/intel/binary64-float.cton | llvm-mc -show-encoding -triple=x86_64
+;   sed -ne 's/^ *; asm: *//p' filetests/isa/x86/binary64-float.cton | llvm-mc -show-encoding -triple=x86_64
 ;
 
 function %F32() {

cranelift/filetests/isa/x86/binary64-pic.cton

@@ -0,0 +1,84 @@
+; binary emission of 64-bit code.
+test binemit
+set is_64bit
+set is_compressed
+set is_pic
+isa x86 haswell
+
+; The binary encodings can be verified with the command:
+;
+;   sed -ne 's/^ *; asm: *//p' filetests/isa/x86/binary64-pic.cton | llvm-mc -show-encoding -triple=x86_64
+;
+
+; Tests for i64 instructions.
+function %I64() {
+    sig0 = ()
+    fn0 = %foo()
+    fn1 = colocated %bar()
+
+    gv0 = globalsym %some_gv
+    gv1 = globalsym colocated %some_gv
+
+    ; Use incoming_arg stack slots because they won't be relocated by the frame
+    ; layout.
+    ss0 = incoming_arg 8, offset 0
+    ss1 = incoming_arg 1024, offset -1024
+    ss2 = incoming_arg 1024, offset -2048
+    ss3 = incoming_arg 8, offset -2056
+
+ebb0:
+
+    ; Colocated functions.
+
+    ; asm: call foo
+    call fn1()                                  ; bin: e8 PCRel4(%bar-4) 00000000
+
+    ; asm: lea 0x0(%rip), %rax
+    [-,%rax]            v0 = func_addr.i64 fn1        ; bin: 48 8d 05 PCRel4(%bar-4) 00000000
+    ; asm: lea 0x0(%rip), %rsi
+    [-,%rsi]            v1 = func_addr.i64 fn1        ; bin: 48 8d 35 PCRel4(%bar-4) 00000000
+    ; asm: lea 0x0(%rip), %r10
+    [-,%r10]            v2 = func_addr.i64 fn1        ; bin: 4c 8d 15 PCRel4(%bar-4) 00000000
+
+    ; asm: call *%rax
+    call_indirect sig0, v0()                  ; bin: ff d0
+    ; asm: call *%rsi
+    call_indirect sig0, v1()                  ; bin: ff d6
+    ; asm: call *%r10
+    call_indirect sig0, v2()                  ; bin: 41 ff d2
+
+    ; Non-colocated functions.
+
+    ; asm: call foo@PLT
+    call fn0()                                  ; bin: e8 PLTRel4(%foo-4) 00000000
+
+    ; asm: mov 0x0(%rip), %rax
+    [-,%rax]            v100 = func_addr.i64 fn0        ; bin: 48 8b 05 GOTPCRel4(%foo-4) 00000000
+    ; asm: mov 0x0(%rip), %rsi
+    [-,%rsi]            v101 = func_addr.i64 fn0        ; bin: 48 8b 35 GOTPCRel4(%foo-4) 00000000
+    ; asm: mov 0x0(%rip), %r10
+    [-,%r10]            v102 = func_addr.i64 fn0        ; bin: 4c 8b 15 GOTPCRel4(%foo-4) 00000000
+
+    ; asm: call *%rax
+    call_indirect sig0, v100()                  ; bin: ff d0
+    ; asm: call *%rsi
+    call_indirect sig0, v101()                  ; bin: ff d6
+    ; asm: call *%r10
+    call_indirect sig0, v102()                  ; bin: 41 ff d2
+
+    ; asm: mov 0x0(%rip), %rcx
+    [-,%rcx]            v3 = globalsym_addr.i64 gv0    ; bin: 48 8b 0d GOTPCRel4(%some_gv-4) 00000000
+    ; asm: mov 0x0(%rip), %rsi
+    [-,%rsi]            v4 = globalsym_addr.i64 gv0    ; bin: 48 8b 35 GOTPCRel4(%some_gv-4) 00000000
+    ; asm: mov 0x0(%rip), %r10
+    [-,%r10]            v5 = globalsym_addr.i64 gv0    ; bin: 4c 8b 15 GOTPCRel4(%some_gv-4) 00000000
+
+    ; asm: lea 0x0(%rip), %rcx
+    [-,%rcx]            v6 = globalsym_addr.i64 gv1    ; bin: 48 8d 0d PCRel4(%some_gv-4) 00000000
+    ; asm: lea 0x0(%rip), %rsi
+    [-,%rsi]            v7 = globalsym_addr.i64 gv1    ; bin: 48 8d 35 PCRel4(%some_gv-4) 00000000
+    ; asm: lea 0x0(%rip), %r10
+    [-,%r10]            v8 = globalsym_addr.i64 gv1    ; bin: 4c 8d 15 PCRel4(%some_gv-4) 00000000
+
+    return
+}

cranelift/filetests/isa/x86/binary64.cton

@@ -2,17 +2,18 @@
 test binemit
 set is_64bit
 set is_compressed
-isa intel haswell
+isa x86 haswell
 
 ; The binary encodings can be verified with the command:
 ;
-;   sed -ne 's/^ *; asm: *//p' filetests/isa/intel/binary64.cton | llvm-mc -show-encoding -triple=x86_64
+;   sed -ne 's/^ *; asm: *//p' filetests/isa/x86/binary64.cton | llvm-mc -show-encoding -triple=x86_64
 ;
 
 ; Tests for i64 instructions.
 function %I64() {
     sig0 = ()
-    fn0 = function %foo()
+    fn0 = %foo()
+    fn1 = colocated %bar()
 
     gv0 = globalsym %some_gv
 
@@ -473,15 +474,17 @@ ebb0:
     ; asm: movzbq %dl, %rsi
     [-,%rsi]             v351 = bint.i64 v301   ; bin: 0f b6 f2
 
-    ; asm: call foo
-    call fn0()                                  ; bin: e8 PCRel4(%foo) 00000000
+    ; Colocated functions.
 
-    ; asm: movabsq $0, %rcx
-    [-,%rcx]            v400 = func_addr.i64 fn0        ; bin: 48 b9 Abs8(%foo) 0000000000000000
-    ; asm: movabsq $0, %rsi
-    [-,%rsi]            v401 = func_addr.i64 fn0        ; bin: 48 be Abs8(%foo) 0000000000000000
-    ; asm: movabsq $0, %r10
-    [-,%r10]            v402 = func_addr.i64 fn0        ; bin: 49 ba Abs8(%foo) 0000000000000000
+    ; asm: call bar
+    ; call fn1()                                ; bin: e8 PCRel4(%bar-4) 00000000
+
+    ; asm: lea 0x0(%rip), %rcx
+    [-,%rcx]            v400 = func_addr.i64 fn1        ; bin: 48 8d 0d PCRel4(%bar-4) 00000000
+    ; asm: lea 0x0(%rip), %rsi
+    [-,%rsi]            v401 = func_addr.i64 fn1        ; bin: 48 8d 35 PCRel4(%bar-4) 00000000
+    ; asm: lea 0x0(%rip), %r10
+    [-,%r10]            v402 = func_addr.i64 fn1        ; bin: 4c 8d 15 PCRel4(%bar-4) 00000000
 
     ; asm: call *%rcx
     call_indirect sig0, v400()                  ; bin: ff d1
@@ -490,6 +493,22 @@ ebb0:
     ; asm: call *%r10
     call_indirect sig0, v402()                  ; bin: 41 ff d2
 
+    ; Non-colocated functions. Note that there is no non-colocated non-PIC call.
+
+    ; asm: movabsq $0, %rcx
+    [-,%rcx]            v410 = func_addr.i64 fn0        ; bin: 48 b9 Abs8(%foo) 0000000000000000
+    ; asm: movabsq $0, %rsi
+    [-,%rsi]            v411 = func_addr.i64 fn0        ; bin: 48 be Abs8(%foo) 0000000000000000
+    ; asm: movabsq $0, %r10
+    [-,%r10]            v412 = func_addr.i64 fn0        ; bin: 49 ba Abs8(%foo) 0000000000000000
+
+    ; asm: call *%rcx
+    call_indirect sig0, v410()                  ; bin: ff d1
+    ; asm: call *%rsi
+    call_indirect sig0, v411()                  ; bin: ff d6
+    ; asm: call *%r10
+    call_indirect sig0, v412()                  ; bin: 41 ff d2
+
     ; asm: movabsq $-1, %rcx
     [-,%rcx]            v450 = globalsym_addr.i64 gv0    ; bin: 48 b9 Abs8(%some_gv) 0000000000000000
     ; asm: movabsq $-1, %rsi
@@ -551,9 +570,9 @@ ebb0:
     [-,%rsi]             v517 = sshr_imm v2, 32   ; bin: 48 c1 fe 20
     ; asm: sarq $33, %r8
     [-,%r8]              v518 = sshr_imm v4, 33   ; bin: 49 c1 f8 21
-    ; asm: shrl $62, %rsi
+    ; asm: shrq $62, %rsi
     [-,%rsi]             v519 = ushr_imm v2, 62   ; bin: 48 c1 ee 3e
-    ; asm: shrl $63, %r8
+    ; asm: shrq $63, %r8
     [-,%r8]              v520 = ushr_imm v4, 63   ; bin: 49 c1 e8 3f
 
 
@@ -710,7 +729,7 @@ ebb0:
 ; be done by an instruction shrinking pass.
 function %I32() {
     sig0 = ()
-    fn0 = function %foo()
+    fn0 = %foo()
 
     ss0 = incoming_arg 8, offset 0
     ss1 = incoming_arg 1024, offset -1024
@@ -1043,7 +1062,7 @@ ebb0:
     ; asm: setl %bl
     [-,%rbx]            v320 = icmp_imm slt v1, 37  ; bin: 83 f9 25 0f 9c c3
 
-    ; asm: cmpq $100000, %ecx
+    ; asm: cmpl $100000, %ecx
     ; asm: setl %bl
     [-,%rbx]            v321 = icmp_imm slt v1, 100000 ; bin: 81 f9 000186a0 0f 9c c3
 

cranelift/filetests/isa/x86/legalize-custom.cton

@@ -1,8 +1,8 @@
 ; Test the custom legalizations.
 test legalizer
-isa intel
+isa x86
 set is_64bit
-isa intel
+isa x86
 
 ; regex: V=v\d+
 ; regex: EBB=ebb\d+

cranelift/filetests/isa/x86/legalize-div-traps.cton

@@ -3,7 +3,7 @@ test legalizer
 set is_64bit
 ; See also legalize-div.cton.
 set avoid_div_traps=1
-isa intel
+isa x86
 
 ; regex: V=v\d+
 ; regex: EBB=ebb\d+

cranelift/filetests/isa/x86/legalize-div.cton

@@ -3,7 +3,7 @@ test legalizer
 set is_64bit
 ; See also legalize-div-traps.cton.
 set avoid_div_traps=0
-isa intel
+isa x86
 
 ; regex: V=v\d+
 ; regex: EBB=ebb\d+

cranelift/filetests/isa/x86/legalize-libcall.cton

@@ -2,7 +2,8 @@ test legalizer
 
 ; Pre-SSE 4.1, we need to use runtime library calls for floating point rounding operations.
 set is_64bit
-isa intel
+set is_pic
+isa x86
 
 function %floor(f32) -> f32 {
 ebb0(v0: f32):
@@ -11,5 +12,5 @@ ebb0(v0: f32):
 }
 ; check: function %floor(f32 [%xmm0]) -> f32 [%xmm0] system_v {
 ; check: sig0 = (f32) -> f32 system_v
-; check: fn0 = sig0 %FloorF32
+; check: fn0 = %FloorF32 sig0
 ; check: v1 = call fn0(v0)

cranelift/filetests/isa/x86/legalize-memory.cton

@@ -1,7 +1,7 @@
 ; Test the legalization of memory objects.
 test legalizer
 set is_64bit
-isa intel
+isa x86
 
 ; regex: V=v\d+
 ; regex: EBB=ebb\d+

cranelift/filetests/isa/x86/legalize-mulhi.cton

@@ -1,7 +1,7 @@
 
 test compile
 set is_64bit
-isa intel baseline
+isa x86 baseline
 
 ; umulhi/smulhi on 64 bit operands
 

cranelift/filetests/isa/x86/prologue-epilogue.cton

@@ -0,0 +1,231 @@
+test compile
+set is_64bit
+set is_compressed
+set is_pic
+isa x86 haswell
+
+; An empty function.
+
+function %empty() {
+ebb0:
+    return
+}
+
+; check: function %empty(i64 fp [%rbp]) -> i64 fp [%rbp] system_v {
+; nextln:     ss0 = incoming_arg 16, offset -16
+; nextln: 
+; nextln: ebb0(v0: i64 [%rbp]):
+; nextln:     x86_push v0
+; nextln:     copy_special %rsp -> %rbp
+; nextln:     v1 = x86_pop.i64
+; nextln:     return v1
+; nextln: }
+
+; A function with a single stack slot.
+
+function %one_stack_slot() {
+    ss0 = explicit_slot 168
+ebb0:
+    return
+}
+
+; check: function %one_stack_slot(i64 fp [%rbp]) -> i64 fp [%rbp] system_v {
+; nextln:     ss0 = explicit_slot 168, offset -184
+; nextln:     ss1 = incoming_arg 16, offset -16
+; nextln: 
+; nextln: ebb0(v0: i64 [%rbp]):
+; nextln:     x86_push v0
+; nextln:     copy_special %rsp -> %rbp
+; nextln:     adjust_sp_imm -176
+; nextln:     adjust_sp_imm 176
+; nextln:     v1 = x86_pop.i64
+; nextln:     return v1
+; nextln: }
+
+; A function performing a call.
+
+function %call() {
+    fn0 = %foo()
+
+ebb0:
+    call fn0()
+    return
+}
+
+; check: function %call(i64 fp [%rbp]) -> i64 fp [%rbp] system_v {
+; nextln:     ss0 = incoming_arg 16, offset -16
+; nextln:     sig0 = () system_v
+; nextln:     fn0 = %foo sig0
+; nextln: 
+; nextln: ebb0(v0: i64 [%rbp]):
+; nextln:     x86_push v0
+; nextln:     copy_special %rsp -> %rbp
+; nextln:     call fn0()
+; nextln:     v1 = x86_pop.i64
+; nextln:     return v1
+; nextln: }
+
+; A function that uses a lot of registers but doesn't quite need to spill.
+
+function %no_spill(i64, i64) {
+ebb0(v0: i64, v1: i64):
+    v2 = load.i32 v0+0
+    v3 = load.i32 v0+8
+    v4 = load.i32 v0+16
+    v5 = load.i32 v0+24
+    v6 = load.i32 v0+32
+    v7 = load.i32 v0+40
+    v8 = load.i32 v0+48
+    v9 = load.i32 v0+56
+    v10 = load.i32 v0+64
+    v11 = load.i32 v0+72
+    v12 = load.i32 v0+80
+    v13 = load.i32 v0+88
+    v14 = load.i32 v0+96
+    store.i32 v2, v1+0
+    store.i32 v3, v1+8
+    store.i32 v4, v1+16
+    store.i32 v5, v1+24
+    store.i32 v6, v1+32
+    store.i32 v7, v1+40
+    store.i32 v8, v1+48
+    store.i32 v9, v1+56
+    store.i32 v10, v1+64
+    store.i32 v11, v1+72
+    store.i32 v12, v1+80
+    store.i32 v13, v1+88
+    store.i32 v14, v1+96
+    return
+}
+
+; check: function %no_spill(i64 [%rdi], i64 [%rsi], i64 fp [%rbp], i64 csr [%rbx], i64 csr [%r12], i64 csr [%r13], i64 csr [%r14], i64 csr [%r15]) -> i64 fp [%rbp], i64 csr [%rbx], i64 csr [%r12], i64 csr [%r13], i64 csr [%r14], i64 csr [%r15] system_v {
+; nextln:     ss0 = incoming_arg 56, offset -56
+; nextln: 
+; nextln: ebb0(v0: i64 [%rdi], v1: i64 [%rsi], v15: i64 [%rbp], v16: i64 [%rbx], v17: i64 [%r12], v18: i64 [%r13], v19: i64 [%r14], v20: i64 [%r15]):
+; nextln:     x86_push v15
+; nextln:     copy_special %rsp -> %rbp
+; nextln:     x86_push v16
+; nextln:     x86_push v17
+; nextln:     x86_push v18
+; nextln:     x86_push v19
+; nextln:     x86_push v20
+; nextln:     adjust_sp_imm -8
+; nextln:     v2 = load.i32 v0
+; nextln:     v3 = load.i32 v0+8
+; nextln:     v4 = load.i32 v0+16
+; nextln:     v5 = load.i32 v0+24
+; nextln:     v6 = load.i32 v0+32
+; nextln:     v7 = load.i32 v0+40
+; nextln:     v8 = load.i32 v0+48
+; nextln:     v9 = load.i32 v0+56
+; nextln:     v10 = load.i32 v0+64
+; nextln:     v11 = load.i32 v0+72
+; nextln:     v12 = load.i32 v0+80
+; nextln:     v13 = load.i32 v0+88
+; nextln:     v14 = load.i32 v0+96
+; nextln:     store v2, v1
+; nextln:     store v3, v1+8
+; nextln:     store v4, v1+16
+; nextln:     store v5, v1+24
+; nextln:     store v6, v1+32
+; nextln:     store v7, v1+40
+; nextln:     store v8, v1+48
+; nextln:     store v9, v1+56
+; nextln:     store v10, v1+64
+; nextln:     store v11, v1+72
+; nextln:     store v12, v1+80
+; nextln:     store v13, v1+88
+; nextln:     store v14, v1+96
+; nextln:     adjust_sp_imm 8
+; nextln:     v26 = x86_pop.i64
+; nextln:     v25 = x86_pop.i64
+; nextln:     v24 = x86_pop.i64
+; nextln:     v23 = x86_pop.i64
+; nextln:     v22 = x86_pop.i64
+; nextln:     v21 = x86_pop.i64
+; nextln:     return v21, v22, v23, v24, v25, v26
+; nextln: }
+
+; This function requires too many registers and must spill.
+
+function %yes_spill(i64, i64) {
+ebb0(v0: i64, v1: i64):
+    v2 = load.i32 v0+0
+    v3 = load.i32 v0+8
+    v4 = load.i32 v0+16
+    v5 = load.i32 v0+24
+    v6 = load.i32 v0+32
+    v7 = load.i32 v0+40
+    v8 = load.i32 v0+48
+    v9 = load.i32 v0+56
+    v10 = load.i32 v0+64
+    v11 = load.i32 v0+72
+    v12 = load.i32 v0+80
+    v13 = load.i32 v0+88
+    v14 = load.i32 v0+96
+    v15 = load.i32 v0+104
+    store.i32 v2, v1+0
+    store.i32 v3, v1+8
+    store.i32 v4, v1+16
+    store.i32 v5, v1+24
+    store.i32 v6, v1+32
+    store.i32 v7, v1+40
+    store.i32 v8, v1+48
+    store.i32 v9, v1+56
+    store.i32 v10, v1+64
+    store.i32 v11, v1+72
+    store.i32 v12, v1+80
+    store.i32 v13, v1+88
+    store.i32 v14, v1+96
+    store.i32 v15, v1+104
+    return
+}
+
+; check: function %yes_spill(i64 [%rdi], i64 [%rsi], i64 fp [%rbp], i64 csr [%rbx], i64 csr [%r12], i64 csr [%r13], i64 csr [%r14], i64 csr [%r15]) -> i64 fp [%rbp], i64 csr [%rbx], i64 csr [%r12], i64 csr [%r13], i64 csr [%r14], i64 csr [%r15] system_v {
+; check:     ss0 = spill_slot
+
+; check: ebb0(v16: i64 [%rdi], v17: i64 [%rsi], v48: i64 [%rbp], v49: i64 [%rbx], v50: i64 [%r12], v51: i64 [%r13], v52: i64 [%r14], v53: i64 [%r15]):
+; nextln:     x86_push v48
+; nextln:     copy_special %rsp -> %rbp
+; nextln:     x86_push v49
+; nextln:     x86_push v50
+; nextln:     x86_push v51
+; nextln:     x86_push v52
+; nextln:     x86_push v53
+; nextln:     adjust_sp_imm
+
+; check:      spill
+
+; check:      fill
+
+; check:     adjust_sp_imm
+; nextln:     v59 = x86_pop.i64
+; nextln:     v58 = x86_pop.i64
+; nextln:     v57 = x86_pop.i64
+; nextln:     v56 = x86_pop.i64
+; nextln:     v55 = x86_pop.i64
+; nextln:     v54 = x86_pop.i64
+; nextln:     return v54, v55, v56, v57, v58, v59
+; nextln: }
+
+; A function which uses diverted registers.
+
+function %divert(i32) -> i32 system_v {
+ebb0(v0: i32):
+    v2 = iconst.i32 0
+    v3 = iconst.i32 1
+    jump ebb3(v0, v3, v2)
+
+ebb3(v4: i32, v5: i32, v6: i32):
+    brz v4, ebb4
+    v7 = iadd v5, v6
+    v8 = iadd_imm v4, -1
+    jump ebb3(v8, v7, v5)
+
+ebb4:
+    return v5
+}
+
+; check: function %divert
+; check: regmove v5, %rcx -> %rbx
+; check: [RexOp1popq#58,%rbx]                v15 = x86_pop.i64

cranelift/filetests/parser/call.cton

@@ -26,22 +26,22 @@ ebb1:
 function %signatures() {
     sig10 = ()
     sig11 = (i32, f64) -> i32, b1 spiderwasm
-    fn5 = sig11 %foo
-    fn8 = function %bar(i32) -> b1
+    fn5 = %foo sig11
+    fn8 = %bar(i32) -> b1
 }
 ; sameln: function %signatures() system_v {
 ; check:      sig10 = () system_v
 ; check:      sig11 = (i32, f64) -> i32, b1 spiderwasm
 ; check:      sig12 = (i32) -> b1 system_v
 ; not:        fn0
-; check:      fn5 = sig11 %foo
-; check:      fn8 = sig12 %bar
+; check:      fn5 = %foo sig11
+; check:      fn8 = %bar sig12
 ; check:  }
 
 function %direct() {
-    fn0 = function %none()
-    fn1 = function %one() -> i32
-    fn2 = function %two() -> i32, f32
+    fn0 = %none()
+    fn1 = %one() -> i32
+    fn2 = %two() -> i32, f32
 
 ebb0:
     call fn0()
@@ -72,7 +72,7 @@ ebb0(v0: i64):
 
 function %long_call() {
     sig0 = ()
-    fn0 = sig0 %none
+    fn0 = %none sig0
 
 ebb0:
     v0 = func_addr.i32 fn0

cranelift/filetests/postopt/basic.cton

@@ -1,5 +1,5 @@
 test postopt
-isa intel
+isa x86
 
 ; Test that compare+branch sequences are folded effectively on x86.
 

cranelift/filetests/preopt/div_by_const_indirect.cton

@@ -1,6 +1,6 @@
 
 test preopt
-isa intel baseline
+isa x86 baseline
 
 ; Cases where the denominator is created by an iconst
 

cranelift/filetests/preopt/div_by_const_non_power_of_2.cton

@@ -1,6 +1,6 @@
 
 test preopt
-isa intel baseline
+isa x86 baseline
 
 ; -------- U32 --------
 

cranelift/filetests/preopt/div_by_const_power_of_2.cton

@@ -1,6 +1,6 @@
 
 test preopt
-isa intel baseline
+isa x86 baseline
 
 ; -------- U32 --------
 

cranelift/filetests/preopt/rem_by_const_non_power_of_2.cton

@@ -1,6 +1,6 @@
 
 test preopt
-isa intel baseline
+isa x86 baseline
 
 ; -------- U32 --------
 

cranelift/filetests/preopt/rem_by_const_power_of_2.cton

@@ -1,6 +1,6 @@
 
 test preopt
-isa intel baseline
+isa x86 baseline
 
 ; -------- U32 --------
 

cranelift/filetests/preopt/simplify.cton

@@ -1,5 +1,5 @@
 test preopt
-isa intel
+isa x86
 
 function %iadd_imm(i32) -> i32 {
 ebb0(v0: i32):
@@ -75,6 +75,6 @@ ebb0(v0: i32):
 ; sameln: function %irsub_imm
 ; nextln: ebb0(v0: i32):
 ; nextln:     v1 = iconst.i32 2
-; nextln:     v2 = irsub_imm v1, 2
+; nextln:     v2 = irsub_imm v0, 2
 ; nextln:     return v2
 ; nextln: }

cranelift/filetests/regalloc/aliases.cton

@@ -1,6 +1,6 @@
 test regalloc
 set is_64bit
-isa intel haswell
+isa x86 haswell
 
 function %value_aliases(i32, f32, i64 vmctx) spiderwasm {
     gv0 = vmctx

cranelift/filetests/regalloc/basic.cton

@@ -53,7 +53,7 @@ ebb1(v10: i32):
 
 ; Pass an EBB argument as a function argument.
 function %callebb(i32, i32) -> i32 {
-    fn0 = function %foo(i32) -> i32
+    fn0 = %foo(i32) -> i32
 
 ebb0(v1: i32, v2: i32):
     brnz v1, ebb1(v1)
@@ -66,7 +66,7 @@ ebb1(v10: i32):
 
 ; Pass an EBB argument as a jump argument.
 function %jumpebb(i32, i32) -> i32 {
-    fn0 = function %foo(i32) -> i32
+    fn0 = %foo(i32) -> i32
 
 ebb0(v1: i32, v2: i32):
     brnz v1, ebb1(v1, v2)

cranelift/filetests/regalloc/coalescing-207.cton

@@ -1,8 +1,8 @@
 test regalloc
 set is_64bit
-isa intel haswell
+isa x86 haswell
 
-; Reported as https://github.com/Cretonne/cretonne/issues/207
+; Reported as https://github.com/cretonne/cretonne/issues/207
 ;
 ; The coalescer creates a virtual register with two interfering values.
 function %pr207(i64 vmctx, i32, i32) -> i32 system_v {
@@ -11,9 +11,9 @@ function %pr207(i64 vmctx, i32, i32) -> i32 system_v {
     sig0 = (i64 vmctx, i32, i32) -> i32 system_v
     sig1 = (i64 vmctx, i32, i32, i32) -> i32 system_v
     sig2 = (i64 vmctx, i32, i32, i32) -> i32 system_v
-    fn0 = sig0 u0:2
-    fn1 = sig1 u0:0
-    fn2 = sig2 u0:1
+    fn0 = u0:2 sig0
+    fn1 = u0:0 sig1
+    fn2 = u0:1 sig2
 
 ebb0(v0: i64, v1: i32, v2: i32):
     v3 = iconst.i32 0
@@ -1038,7 +1038,7 @@ function %musl(f64 [%xmm0], i64 vmctx [%rdi]) -> f64 [%xmm0] system_v {
     gv0 = vmctx
     heap0 = static gv0, min 0, bound 0x0001_0000_0000, guard 0x8000_0000
     sig0 = (f64 [%xmm0], i32 [%rdi], i64 vmctx [%rsi]) -> f64 [%xmm0] system_v
-    fn0 = sig0 u0:517
+    fn0 = u0:517 sig0
 
 ebb0(v0: f64, v1: i64):
     v3 = iconst.i64 0

cranelift/filetests/regalloc/coalescing-216.cton

@@ -1,8 +1,8 @@
 test regalloc
 set is_64bit
-isa intel haswell
+isa x86 haswell
 
-; Reported as https://github.com/Cretonne/cretonne/issues/216 from the Binaryen fuzzer.
+; Reported as https://github.com/cretonne/cretonne/issues/216 from the Binaryen fuzzer.
 ;
 ; The (old) coalescer creates a virtual register with two identical values.
 function %pr216(i32 [%rdi], i64 vmctx [%rsi]) -> i64 [%rax] system_v {

cranelift/filetests/regalloc/coloring-227.cton

@@ -1,6 +1,6 @@
 test regalloc
 set is_64bit
-isa intel haswell
+isa x86 haswell
 
 function %pr227(i32 [%rdi], i32 [%rsi], i32 [%rdx], i32 [%rcx], i64 vmctx [%r8]) system_v {
     gv0 = vmctx

cranelift/filetests/regalloc/constraints.cton

@@ -1,5 +1,5 @@
 test regalloc
-isa intel
+isa x86
 
 ; regex: V=v\d+
 ; regex: REG=%r([abcd]x|[sd]i)

cranelift/filetests/regalloc/ghost-param.cton

@@ -1,6 +1,6 @@
 test regalloc
 set is_64bit
-isa intel haswell
+isa x86 haswell
 
 ; This test case would create an EBB parameter that was a ghost value.
 ; The coalescer would insert a copy of the ghost value, leading to verifier errors.

cranelift/filetests/regalloc/global-constraints.cton

@@ -1,11 +1,11 @@
 test regalloc
-isa intel
+isa x86
 
 ; This test covers the troubles when values with global live ranges are defined
 ; by instructions with constrained register classes.
 ;
 ; The icmp_imm instrutions write their b1 result to the ABCD register class on
-; 32-bit Intel. So if we define 5 live values, they can't all fit.
+; 32-bit x86. So if we define 5 live values, they can't all fit.
 function %global_constraints(i32) {
 ebb0(v0: i32):
     v1 = icmp_imm eq v0, 1

cranelift/filetests/regalloc/global-fixed.cton

@@ -1,6 +1,6 @@
 test regalloc
 set is_64bit=1
-isa intel haswell
+isa x86 haswell
 
 function %foo() system_v {
 ebb4:

cranelift/filetests/regalloc/iterate.cton

@@ -1,6 +1,6 @@
 test regalloc
 set is_64bit
-isa intel haswell
+isa x86 haswell
 
 function u0:9(i64 [%rdi], f32 [%xmm0], f64 [%xmm1], i32 [%rsi], i32 [%rdx], i64 vmctx [%r14]) -> i64 [%rax] spiderwasm {
 ebb0(v0: i64, v1: f32, v2: f64, v3: i32, v4: i32, v5: i64):

cranelift/filetests/regalloc/multi-constraints.cton

@@ -1,10 +1,10 @@
 test regalloc
 set is_64bit
-isa intel haswell
+isa x86 haswell
 
 ; Test combinations of constraints.
 ;
-; The Intel ushr instruction requires its second operand to be passed in %rcx and its output is
+; The x86 ushr instruction requires its second operand to be passed in %rcx and its output is
 ; tied to the first input operand.
 ;
 ; If we pass the same value to both operands, both constraints must be satisfied.

cranelift/filetests/regalloc/output-interference.cton

@@ -1,11 +1,11 @@
 test regalloc
 set is_64bit=1
-isa intel haswell
+isa x86 haswell
 
 function %test(i64) -> i64 system_v {
 ebb0(v0: i64):
     v2 = iconst.i64 12
-    ; This division clobbers two of its fixed input registers on Intel.
+    ; This division clobbers two of its fixed input registers on x86.
     ; These are FixedTied constraints that the spiller needs to resolve.
     v5 = udiv v0, v2
     v6 = iconst.i64 13

cranelift/filetests/regalloc/reload-208.cton

@@ -1,10 +1,10 @@
 test regalloc
 set is_64bit
-isa intel haswell
+isa x86 haswell
 
 ; regex: V=v\d+
 
-; Filed as https://github.com/Cretonne/cretonne/issues/208
+; Filed as https://github.com/cretonne/cretonne/issues/208
 ;
 ; The verifier complains about a branch argument that is not in the same virtual register as the
 ; corresponding EBB argument.
@@ -16,8 +16,8 @@ function %pr208(i64 vmctx [%rdi]) system_v {
     heap0 = static gv0, min 0, bound 0x5000, guard 0x0040_0000
     sig0 = (i64 vmctx [%rdi]) -> i32 [%rax] system_v
     sig1 = (i64 vmctx [%rdi], i32 [%rsi]) system_v
-    fn0 = sig0 u0:1
-    fn1 = sig1 u0:3
+    fn0 = u0:1 sig0
+    fn1 = u0:3 sig1
 
 ebb0(v0: i64):
     v1 = iconst.i32 0

cranelift/filetests/regalloc/reload.cton

@@ -5,7 +5,7 @@ isa riscv enable_e
 
 ; Check that we can handle a function return value that got spilled.
 function %spill_return() -> i32 {
-    fn0 = function %foo() -> i32 system_v
+    fn0 = %foo() -> i32 system_v
 
 ebb0:
     v0 = call fn0()

cranelift/filetests/regalloc/schedule-moves.cton

@@ -1,5 +1,5 @@
 test regalloc
-isa intel haswell
+isa x86 haswell
 
 function %pr165() system_v {
 ebb0:

cranelift/filetests/regalloc/spill-noregs.cton

@@ -1,12 +1,12 @@
 test regalloc
 set is_64bit
-isa intel
+isa x86
 
 ; Test case found by the Binaryen fuzzer.
 ;
 ; The spiller panics with a
 ; 'Ran out of GPR registers when inserting copy before v68 = icmp.i32 eq v66, v67',
-; lib/cretonne/src/regalloc/spilling.rs:425:28 message.
+; lib/codegen/src/regalloc/spilling.rs:425:28 message.
 ;
 ; The process_reg_uses() function is trying to insert a copy before the icmp instruction in ebb4
 ; and runs out of registers to spill. Note that ebb7 has a lot of dead parameter values.

cranelift/filetests/regalloc/spill.cton

@@ -70,7 +70,7 @@ ebb0(v1: i32):
 
 ; All values live across a call must be spilled
 function %across_call(i32) {
-    fn0 = function %foo(i32)
+    fn0 = %foo(i32)
 ebb0(v1: i32):
     ; check: v1 = spill
     call fn0(v1)
@@ -83,7 +83,7 @@ ebb0(v1: i32):
 
 ; The same value used for two function arguments.
 function %doubleuse(i32) {
-    fn0 = function %xx(i32, i32)
+    fn0 = %xx(i32, i32)
 ebb0(v0: i32):
     ; check: $(c=$V) = copy v0
     call fn0(v0, v0)

cranelift/filetests/regalloc/unreachable_code.cton

@@ -2,7 +2,7 @@
 test compile
 
 set is_64bit
-isa intel haswell
+isa x86 haswell
 
 ; This function contains unreachable blocks which trip up the register
 ; allocator if they don't get cleared out.

cranelift/filetests/regalloc/x86-regres.cton

@@ -1,6 +1,6 @@
 test regalloc
 
-isa intel
+isa x86
 
 ; regex: V=v\d+
 

cranelift/filetests/verifier/flags.cton

@@ -1,5 +1,5 @@
 test verifier
-isa intel
+isa x86
 
 ; Simple, correct use of CPU flags.
 function %simple(i32) -> i32 {

cranelift/filetests/verifier/type_check.cton

@@ -42,14 +42,14 @@ function %type_mismatch_controlling_variable() {
 }
 
 function %fn_call_too_few_args() {
-    fn2 = function %great_fn(i32, f32)
+    fn2 = %great_fn(i32, f32)
     ebb0:
         call fn2() ; error: mismatched argument count, got 0, expected 2
         return
 }
 
 function %fn_call_too_many_args() {
-    fn5 = function %best_fn()
+    fn5 = %best_fn()
     ebb0:
         v0 = iconst.i64 56
         v1 = f32const 0.0

cranelift/filetests/wasm/control.cton

@@ -2,10 +2,10 @@
 test compile
 
 set is_64bit=0
-isa intel haswell
+isa x86 haswell
 
 set is_64bit=1
-isa intel haswell
+isa x86 haswell
 
 function %br_if(i32) -> i32 {
 ebb0(v0: i32):

cranelift/filetests/wasm/conversions.cton

@@ -2,7 +2,7 @@
 test compile
 
 set is_64bit=1
-isa intel haswell
+isa x86 haswell
 
 function %i32_wrap_i64(i64) -> i32 {
 ebb0(v0: i64):

cranelift/filetests/wasm/f32-arith.cton

@@ -2,10 +2,10 @@
 test compile
 
 set is_64bit=0
-isa intel haswell
+isa x86 haswell
 
 set is_64bit=1
-isa intel haswell
+isa x86 haswell
 
 ; Constants.
 

cranelift/filetests/wasm/f32-compares.cton

@@ -2,10 +2,10 @@
 test compile
 
 set is_64bit=0
-isa intel haswell
+isa x86 haswell
 
 set is_64bit=1
-isa intel haswell
+isa x86 haswell
 
 function %f32_eq(f32, f32) -> i32 {
 ebb0(v0: f32, v1: f32):

cranelift/filetests/wasm/f32-memory64.cton

@@ -4,7 +4,7 @@ test compile
 ; We only test on 64-bit since the heap_addr instructions and vmctx parameters
 ; explicitly mention the pointer width.
 set is_64bit=1
-isa intel haswell
+isa x86 haswell
 
 function %f32_load(i32, i64 vmctx) -> f32 {
     gv0 = vmctx

cranelift/filetests/wasm/f64-arith.cton

@@ -2,7 +2,7 @@
 test compile
 
 set is_64bit=1
-isa intel haswell
+isa x86 haswell
 
 ; Constants.
 

cranelift/filetests/wasm/f64-compares.cton

@@ -2,10 +2,10 @@
 test compile
 
 set is_64bit=0
-isa intel haswell
+isa x86 haswell
 
 set is_64bit=1
-isa intel haswell
+isa x86 haswell
 
 function %f64_eq(f64, f64) -> i32 {
 ebb0(v0: f64, v1: f64):

cranelift/filetests/wasm/f64-memory64.cton

@@ -4,7 +4,7 @@ test compile
 ; We only test on 64-bit since the heap_addr instructions and vmctx parameters
 ; explicitly mention the pointer width.
 set is_64bit=1
-isa intel haswell
+isa x86 haswell
 
 function %f64_load(i32, i64 vmctx) -> f64 {
     gv0 = vmctx

cranelift/filetests/wasm/i32-arith.cton

@@ -2,10 +2,10 @@
 test compile
 
 set is_64bit=0
-isa intel haswell
+isa x86 haswell
 
 set is_64bit=1
-isa intel haswell
+isa x86 haswell
 
 ; Constants.
 

cranelift/filetests/wasm/i32-compares.cton

@@ -2,10 +2,10 @@
 test compile
 
 set is_64bit=0
-isa intel haswell
+isa x86 haswell
 
 set is_64bit=1
-isa intel haswell
+isa x86 haswell
 
 function %i32_eqz(i32) -> i32 {
 ebb0(v0: i32):

cranelift/filetests/wasm/i32-memory64.cton

@@ -4,7 +4,7 @@ test compile
 ; We only test on 64-bit since the heap_addr instructions and vmctx parameters
 ; explicitly mention the pointer width.
 set is_64bit=1
-isa intel haswell
+isa x86 haswell
 
 function %i32_load(i32, i64 vmctx) -> i32 {
     gv0 = vmctx

cranelift/filetests/wasm/i64-arith.cton

@@ -2,7 +2,7 @@
 test compile
 
 set is_64bit=1
-isa intel haswell
+isa x86 haswell
 
 ; Constants.
 

cranelift/filetests/wasm/i64-compares.cton

@@ -2,7 +2,7 @@
 test compile
 
 set is_64bit=1
-isa intel haswell
+isa x86 haswell
 
 function %i64_eqz(i64) -> i32 {
 ebb0(v0: i64):

cranelift/filetests/wasm/i64-memory64.cton

@@ -4,7 +4,7 @@ test compile
 ; We only test on 64-bit since the heap_addr instructions and vmctx parameters
 ; explicitly mention the pointer width.
 set is_64bit=1
-isa intel haswell
+isa x86 haswell
 
 function %i64_load(i32, i64 vmctx) -> i64 {
     gv0 = vmctx

cranelift/filetests/wasm/select.cton

@@ -2,10 +2,10 @@
 test compile
 
 set is_64bit=0
-isa intel haswell
+isa x86 haswell
 
 set is_64bit=1
-isa intel haswell
+isa x86 haswell
 
 function %select_i32(i32, i32, i32) -> i32 {
 ebb0(v0: i32, v1: i32, v2: i32):

cranelift/publish-all.sh

@@ -4,7 +4,7 @@ cd $(dirname "$0")
 topdir="$(pwd)"
 
 # All the cretonne-* crates have the same version number
-version="0.4.1"
+version="0.5.0"
 
 # Update all of the Cargo.toml files.
 #
@@ -27,8 +27,14 @@ cargo update
 
 echo git commit -a -m "\"Bump version to $version"\"
 echo git push
-for crate in cretonne frontend native reader wasm; do
+for crate in entity codegen frontend native reader wasm module simplejit faerie umbrella ; do
+    if [ "$crate" == "umbrella" ]; then
+        dir="cretonne"
+    else
+        dir="$crate"
+    fi
+
     echo cargo publish --manifest-path "lib/$crate/Cargo.toml"
 done
 echo
-echo Then, go to https://github.com/Cretonne/cretonne/releases/ and define a new release.
+echo Then, go to https://github.com/cretonne/cretonne/releases/ and define a new release.

cranelift/src/cat.rs

@@ -4,7 +4,7 @@
 //! normalizing formatting and removing comments.
 
 use CommandResult;
-use cton_reader::parse_functions;
+use cretonne_reader::parse_functions;
 use utils::read_to_string;
 
 pub fn run(files: &[String]) -> CommandResult {

cranelift/src/compile.rs

@@ -1,10 +1,10 @@
 //! CLI tool to read Cretonne IR files and compile them into native code.
 
-use cretonne::Context;
-use cretonne::print_errors::pretty_error;
-use cretonne::settings::FlagsOrIsa;
-use cretonne::{binemit, ir};
-use cton_reader::parse_test;
+use cretonne_codegen::Context;
+use cretonne_codegen::print_errors::pretty_error;
+use cretonne_codegen::settings::FlagsOrIsa;
+use cretonne_codegen::{binemit, ir};
+use cretonne_reader::parse_test;
 use std::path::Path;
 use std::path::PathBuf;
 use utils::{parse_sets_and_isa, read_to_string};

cranelift/src/cton-util.rs

@@ -1,7 +1,7 @@
-extern crate cretonne;
-extern crate cton_filetests;
-extern crate cton_reader;
-extern crate cton_wasm;
+extern crate cretonne_codegen;
+extern crate cretonne_filetests;
+extern crate cretonne_reader;
+extern crate cretonne_wasm;
 extern crate docopt;
 extern crate filecheck;
 #[macro_use]
@@ -9,7 +9,7 @@ extern crate serde_derive;
 extern crate tempdir;
 extern crate term;
 
-use cretonne::{timing, VERSION};
+use cretonne_codegen::{timing, VERSION};
 use docopt::Docopt;
 use std::io::{self, Write};
 use std::process;
@@ -86,7 +86,7 @@ fn cton_util() -> CommandResult {
 
     // Find the sub-command to execute.
     let result = if args.cmd_test {
-        cton_filetests::run(args.flag_verbose, &args.arg_file).map(|_time| ())
+        cretonne_filetests::run(args.flag_verbose, &args.arg_file).map(|_time| ())
     } else if args.cmd_cat {
         cat::run(&args.arg_file)
     } else if args.cmd_filecheck {

cranelift/src/print_cfg.rs

@@ -4,8 +4,8 @@
 //! in graphviz format.
 
 use CommandResult;
-use cretonne::cfg_printer::CFGPrinter;
-use cton_reader::parse_functions;
+use cretonne_codegen::cfg_printer::CFGPrinter;
+use cretonne_reader::parse_functions;
 use utils::read_to_string;
 
 pub fn run(files: &[String]) -> CommandResult {

cranelift/src/utils.rs

@@ -1,9 +1,9 @@
 //! Utility functions.
 
-use cretonne::isa;
-use cretonne::isa::TargetIsa;
-use cretonne::settings::{self, FlagsOrIsa};
-use cton_reader::{parse_options, Location};
+use cretonne_codegen::isa;
+use cretonne_codegen::isa::TargetIsa;
+use cretonne_codegen::settings::{self, FlagsOrIsa};
+use cretonne_reader::{parse_options, Location};
 use std::fs::File;
 use std::io::{self, Read};
 use std::path::Path;

cranelift/src/wasm.rs

@@ -1,12 +1,13 @@
-//! CLI tool to use the functions provided by the [cretonne-wasm](../cton_wasm/index.html) crate.
+//! CLI tool to use the functions provided by the [cretonne-wasm](../cretonne_wasm/index.html)
+//! crate.
 //!
 //! Reads Wasm binary files, translates the functions' code to Cretonne IR.
 #![cfg_attr(feature = "cargo-clippy", allow(too_many_arguments, cyclomatic_complexity))]
 
-use cretonne::Context;
-use cretonne::print_errors::{pretty_error, pretty_verifier_error};
-use cretonne::settings::FlagsOrIsa;
-use cton_wasm::{translate_module, DummyEnvironment, ModuleEnvironment};
+use cretonne_codegen::Context;
+use cretonne_codegen::print_errors::{pretty_error, pretty_verifier_error};
+use cretonne_codegen::settings::FlagsOrIsa;
+use cretonne_wasm::{translate_module, DummyEnvironment, ModuleEnvironment};
 use std::error::Error;
 use std::fs::File;
 use std::io;

cranelift/test-all.sh

@@ -31,13 +31,13 @@ fi
 # Check if any Python files have changed since we last checked them.
 tsfile=$topdir/target/meta-checked
 if [ -f $tsfile ]; then
-    needcheck=$(find $topdir/lib/cretonne/meta -name '*.py' -newer $tsfile)
+    needcheck=$(find $topdir/lib/codegen/meta -name '*.py' -newer $tsfile)
 else
     needcheck=yes
 fi
 if [ -n "$needcheck" ]; then
     banner "$(python --version 2>&1), $(python3 --version 2>&1)"
-    $topdir/lib/cretonne/meta/check.sh
+    $topdir/lib/codegen/meta/check.sh
     touch $tsfile || echo no target directory
 fi
 

cranelift/test-no_std.sh

@@ -15,7 +15,7 @@ function banner() {
 }
 
 # Test those packages which have no_std support.
-LIBS="cretonne frontend wasm native"
+LIBS="codegen frontend wasm native"
 cd "$topdir"
 for LIB in $LIBS
 do

cranelift/tests/filetests.rs

@@ -1,7 +1,7 @@
-extern crate cton_filetests;
+extern crate cretonne_filetests;
 
 #[test]
 fn filetests() {
     // Run all the filetests in the following directories.
-    cton_filetests::run(false, &["filetests".into(), "docs".into()]).expect("test harness");
+    cretonne_filetests::run(false, &["filetests".into(), "docs".into()]).expect("test harness");
 }

lib/codegen/Cargo.toml

@@ -1,20 +1,18 @@
 [package]
 authors = ["The Cretonne Project Developers"]
-name = "cretonne"
-version = "0.4.1"
+name = "cretonne-codegen"
+version = "0.5.0"
 description = "Low-level code generator library"
 license = "Apache-2.0"
 documentation = "https://cretonne.readthedocs.io/"
-repository = "https://github.com/Cretonne/cretonne"
+repository = "https://github.com/cretonne/cretonne"
 readme = "README.md"
 keywords = ["compile", "compiler", "jit"]
 build = "build.rs"
 
-[lib]
-name = "cretonne"
-
 [dependencies]
-# It is a goal of the cretonne crate to have minimal external dependencies.
+cretonne-entity = { path = "../entity", version = "0.5.0" }
+# It is a goal of the cretonne-codegen crate to have minimal external dependencies.
 # Please don't add any unless they are essential to the task of creating binary
 # machine code. Integration tests that need external dependencies can be
 # accomodated in `tests`.
@@ -35,4 +33,4 @@ core = ["hashmap_core"]
 
 [badges]
 maintenance = { status = "experimental" }
-travis-ci = { repository = "Cretonne/cretonne" }
+travis-ci = { repository = "cretonne/cretonne" }

lib/codegen/build.rs

@@ -1,7 +1,7 @@
 // Build script.
 //
-// This program is run by Cargo when building lib/cretonne. It is used to generate Rust code from
-// the language definitions in the lib/cretonne/meta directory.
+// This program is run by Cargo when building lib/codegen. It is used to generate Rust code from
+// the language definitions in the lib/codegen/meta directory.
 //
 // Environment:
 //
@@ -77,7 +77,7 @@ fn main() {
 #[derive(Copy, Clone)]
 enum Isa {
     Riscv,
-    Intel,
+    X86,
     Arm32,
     Arm64,
 }
@@ -103,14 +103,14 @@ impl Isa {
 
     /// Returns all supported isa targets.
     fn all() -> [Isa; 4] {
-        [Isa::Riscv, Isa::Intel, Isa::Arm32, Isa::Arm64]
+        [Isa::Riscv, Isa::X86, Isa::Arm32, Isa::Arm64]
     }
 
     /// Returns name of the isa target.
     fn name(&self) -> &'static str {
         match *self {
             Isa::Riscv => "riscv",
-            Isa::Intel => "intel",
+            Isa::X86 => "x86",
             Isa::Arm32 => "arm32",
             Isa::Arm64 => "arm64",
         }
@@ -120,7 +120,7 @@ impl Isa {
     fn is_arch_applicable(&self, arch: &str) -> bool {
         match *self {
             Isa::Riscv => arch == "riscv",
-            Isa::Intel => ["x86_64", "i386", "i586", "i686"].contains(&arch),
+            Isa::X86 => ["x86_64", "i386", "i586", "i686"].contains(&arch),
             Isa::Arm32 => arch.starts_with("arm") || arch.starts_with("thumb"),
             Isa::Arm64 => arch == "aarch64",
         }

lib/codegen/meta/base/entities.py

@@ -20,7 +20,7 @@ stack_slot = EntityRefKind('stack_slot', 'A stack slot.')
 global_var = EntityRefKind('global_var', 'A global variable.')
 
 #: A reference to a function sugnature declared in the function preamble.
-#: Tbis is used to provide the call signature in an indirect call instruction.
+#: This is used to provide the call signature in a call_indirect instruction.
 sig_ref = EntityRefKind('sig_ref', 'A function signature.')
 
 #: A reference to an external function declared in the function preamble.

lib/codegen/meta/base/formats.py

@@ -53,7 +53,7 @@ BranchIcmp = InstructionFormat(intcc, VALUE, VALUE, ebb, VARIABLE_ARGS)
 BranchTable = InstructionFormat(VALUE, entities.jump_table)
 
 Call = InstructionFormat(func_ref, VARIABLE_ARGS)
-IndirectCall = InstructionFormat(sig_ref, VALUE, VARIABLE_ARGS)
+CallIndirect = InstructionFormat(sig_ref, VALUE, VARIABLE_ARGS)
 FuncAddr = InstructionFormat(func_ref)
 
 Load = InstructionFormat(memflags, VALUE, offset32)

lib/codegen/meta/base/legalize.py

@@ -65,6 +65,9 @@ expand_flags = XFormGroup('expand_flags', """
 expand.custom_legalize(insts.global_addr, 'expand_global_addr')
 expand.custom_legalize(insts.heap_addr, 'expand_heap_addr')
 
+# Custom expansions for calls.
+expand.custom_legalize(insts.call, 'expand_call')
+
 # Custom expansions that need to change the CFG.
 # TODO: Add sufficient XForm syntax that we don't need to hand-code these.
 expand.custom_legalize(insts.trapz, 'expand_cond_trap')