llvm-project/llvm/lib/Target/WebAssembly/WebAssemblyTargetMachine.cpp

Ignoring revisions in .git-blame-ignore-revs. Click here to bypass and see the normal blame view.

519 lines
19 KiB
C++
Raw Normal View History

//===- WebAssemblyTargetMachine.cpp - Define TargetMachine for WebAssembly -==//
//
// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
// See https://llvm.org/LICENSE.txt for license information.
// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
//
//===----------------------------------------------------------------------===//
///
/// \file
/// This file defines the WebAssembly-specific subclass of TargetMachine.
///
//===----------------------------------------------------------------------===//
#include "WebAssemblyTargetMachine.h"
#include "MCTargetDesc/WebAssemblyMCTargetDesc.h"
#include "TargetInfo/WebAssemblyTargetInfo.h"
#include "WebAssembly.h"
#include "WebAssemblyMachineFunctionInfo.h"
#include "WebAssemblyTargetObjectFile.h"
#include "WebAssemblyTargetTransformInfo.h"
#include "llvm/CodeGen/MIRParser/MIParser.h"
#include "llvm/CodeGen/MachineFunctionPass.h"
#include "llvm/CodeGen/Passes.h"
#include "llvm/CodeGen/RegAllocRegistry.h"
#include "llvm/CodeGen/TargetPassConfig.h"
#include "llvm/IR/Function.h"
#include "llvm/Support/TargetRegistry.h"
#include "llvm/Target/TargetOptions.h"
#include "llvm/Transforms/Scalar.h"
#include "llvm/Transforms/Scalar/LowerAtomic.h"
#include "llvm/Transforms/Utils.h"
using namespace llvm;
#define DEBUG_TYPE "wasm"
// Emscripten's asm.js-style exception handling
static cl::opt<bool> EnableEmException(
"enable-emscripten-cxx-exceptions",
cl::desc("WebAssembly Emscripten-style exception handling"),
cl::init(false));
// Emscripten's asm.js-style setjmp/longjmp handling
static cl::opt<bool> EnableEmSjLj(
"enable-emscripten-sjlj",
cl::desc("WebAssembly Emscripten-style setjmp/longjmp handling"),
cl::init(false));
// A command-line option to keep implicit locals
// for the purpose of testing with lit/llc ONLY.
// This produces output which is not valid WebAssembly, and is not supported
// by assemblers/disassemblers and other MC based tools.
static cl::opt<bool> WasmDisableExplicitLocals(
"wasm-disable-explicit-locals", cl::Hidden,
cl::desc("WebAssembly: output implicit locals in"
" instruction output for test purposes only."),
cl::init(false));
CMake: Make most target symbols hidden by default Summary: For builds with LLVM_BUILD_LLVM_DYLIB=ON and BUILD_SHARED_LIBS=OFF this change makes all symbols in the target specific libraries hidden by default. A new macro called LLVM_EXTERNAL_VISIBILITY has been added to mark symbols in these libraries public, which is mainly needed for the definitions of the LLVMInitialize* functions. This patch reduces the number of public symbols in libLLVM.so by about 25%. This should improve load times for the dynamic library and also make abi checker tools, like abidiff require less memory when analyzing libLLVM.so One side-effect of this change is that for builds with LLVM_BUILD_LLVM_DYLIB=ON and LLVM_LINK_LLVM_DYLIB=ON some unittests that access symbols that are no longer public will need to be statically linked. Before and after public symbol counts (using gcc 8.2.1, ld.bfd 2.31.1): nm before/libLLVM-9svn.so | grep ' [A-Zuvw] ' | wc -l 36221 nm after/libLLVM-9svn.so | grep ' [A-Zuvw] ' | wc -l 26278 Reviewers: chandlerc, beanz, mgorny, rnk, hans Reviewed By: rnk, hans Subscribers: merge_guards_bot, luismarques, smeenai, ldionne, lenary, s.egerton, pzheng, sameer.abuasal, MaskRay, wuzish, echristo, Jim, hiraditya, michaelplatings, chapuni, jholewinski, arsenm, dschuff, jyknight, dylanmckay, sdardis, nemanjai, jvesely, javed.absar, sbc100, jgravelle-google, aheejin, kbarton, fedor.sergeev, asb, rbar, johnrusso, simoncook, apazos, sabuasal, niosHD, jrtc27, zzheng, edward-jones, mgrang, atanasyan, rogfer01, MartinMosbeck, brucehoult, the_o, PkmX, jocewei, kristina, jsji, llvm-commits Tags: #llvm Differential Revision: https://reviews.llvm.org/D54439
2020-01-15 11:15:07 +08:00
extern "C" LLVM_EXTERNAL_VISIBILITY void LLVMInitializeWebAssemblyTarget() {
// Register the target.
RegisterTargetMachine<WebAssemblyTargetMachine> X(
getTheWebAssemblyTarget32());
RegisterTargetMachine<WebAssemblyTargetMachine> Y(
getTheWebAssemblyTarget64());
// Register backend passes
auto &PR = *PassRegistry::getPassRegistry();
initializeWebAssemblyAddMissingPrototypesPass(PR);
initializeWebAssemblyLowerEmscriptenEHSjLjPass(PR);
initializeLowerGlobalDtorsPass(PR);
initializeFixFunctionBitcastsPass(PR);
initializeOptimizeReturnedPass(PR);
initializeWebAssemblyArgumentMovePass(PR);
initializeWebAssemblySetP2AlignOperandsPass(PR);
initializeWebAssemblyReplacePhysRegsPass(PR);
initializeWebAssemblyPrepareForLiveIntervalsPass(PR);
initializeWebAssemblyOptimizeLiveIntervalsPass(PR);
initializeWebAssemblyMemIntrinsicResultsPass(PR);
initializeWebAssemblyRegStackifyPass(PR);
initializeWebAssemblyRegColoringPass(PR);
initializeWebAssemblyFixIrreducibleControlFlowPass(PR);
initializeWebAssemblyLateEHPreparePass(PR);
initializeWebAssemblyExceptionInfoPass(PR);
initializeWebAssemblyCFGSortPass(PR);
initializeWebAssemblyCFGStackifyPass(PR);
initializeWebAssemblyExplicitLocalsPass(PR);
initializeWebAssemblyLowerBrUnlessPass(PR);
initializeWebAssemblyRegNumberingPass(PR);
initializeWebAssemblyDebugFixupPass(PR);
initializeWebAssemblyPeepholePass(PR);
}
//===----------------------------------------------------------------------===//
// WebAssembly Lowering public interface.
//===----------------------------------------------------------------------===//
static Reloc::Model getEffectiveRelocModel(Optional<Reloc::Model> RM,
const Triple &TT) {
if (!RM.hasValue()) {
// Default to static relocation model. This should always be more optimial
// than PIC since the static linker can determine all global addresses and
// assume direct function calls.
return Reloc::Static;
}
if (!TT.isOSEmscripten()) {
// Relocation modes other than static are currently implemented in a way
// that only works for Emscripten, so disable them if we aren't targeting
// Emscripten.
return Reloc::Static;
}
return *RM;
}
/// Create an WebAssembly architecture model.
///
WebAssemblyTargetMachine::WebAssemblyTargetMachine(
const Target &T, const Triple &TT, StringRef CPU, StringRef FS,
const TargetOptions &Options, Optional<Reloc::Model> RM,
Optional<CodeModel::Model> CM, CodeGenOpt::Level OL, bool JIT)
: LLVMTargetMachine(T,
TT.isArch64Bit() ? "e-m:e-p:64:64-i64:64-n32:64-S128"
: "e-m:e-p:32:32-i64:64-n32:64-S128",
TT, CPU, FS, Options, getEffectiveRelocModel(RM, TT),
getEffectiveCodeModel(CM, CodeModel::Large), OL),
TLOF(new WebAssemblyTargetObjectFile()) {
// WebAssembly type-checks instructions, but a noreturn function with a return
// type that doesn't match the context will cause a check failure. So we lower
// LLVM 'unreachable' to ISD::TRAP and then lower that to WebAssembly's
// 'unreachable' instructions which is meant for that case.
this->Options.TrapUnreachable = true;
// WebAssembly treats each function as an independent unit. Force
// -ffunction-sections, effectively, so that we can emit them independently.
this->Options.FunctionSections = true;
this->Options.DataSections = true;
this->Options.UniqueSectionNames = true;
initAsmInfo();
// Note that we don't use setRequiresStructuredCFG(true). It disables
// optimizations than we're ok with, and want, such as critical edge
// splitting and tail merging.
}
WebAssemblyTargetMachine::~WebAssemblyTargetMachine() = default; // anchor.
const WebAssemblySubtarget *
WebAssemblyTargetMachine::getSubtargetImpl(std::string CPU,
std::string FS) const {
auto &I = SubtargetMap[CPU + FS];
if (!I) {
I = std::make_unique<WebAssemblySubtarget>(TargetTriple, CPU, FS, *this);
}
return I.get();
}
const WebAssemblySubtarget *
WebAssemblyTargetMachine::getSubtargetImpl(const Function &F) const {
Attribute CPUAttr = F.getFnAttribute("target-cpu");
Attribute FSAttr = F.getFnAttribute("target-features");
std::string CPU = !CPUAttr.hasAttribute(Attribute::None)
? CPUAttr.getValueAsString().str()
: TargetCPU;
std::string FS = !FSAttr.hasAttribute(Attribute::None)
? FSAttr.getValueAsString().str()
: TargetFS;
// This needs to be done before we create a new subtarget since any
// creation will depend on the TM and the code generation flags on the
// function that reside in TargetOptions.
resetTargetOptions(F);
return getSubtargetImpl(CPU, FS);
}
namespace {
class CoalesceFeaturesAndStripAtomics final : public ModulePass {
// Take the union of all features used in the module and use it for each
// function individually, since having multiple feature sets in one module
// currently does not make sense for WebAssembly. If atomics are not enabled,
// also strip atomic operations and thread local storage.
static char ID;
WebAssemblyTargetMachine *WasmTM;
public:
CoalesceFeaturesAndStripAtomics(WebAssemblyTargetMachine *WasmTM)
: ModulePass(ID), WasmTM(WasmTM) {}
bool runOnModule(Module &M) override {
FeatureBitset Features = coalesceFeatures(M);
std::string FeatureStr = getFeatureString(Features);
for (auto &F : M)
replaceFeatures(F, FeatureStr);
[WebAssembly] Implement thread-local storage (local-exec model) Summary: Thread local variables are placed inside a `.tdata` segment. Their symbols are offsets from the start of the segment. The address of a thread local variable is computed as `__tls_base` + the offset from the start of the segment. `.tdata` segment is a passive segment and `memory.init` is used once per thread to initialize the thread local storage. `__tls_base` is a wasm global. Since each thread has its own wasm instance, it is effectively thread local. Currently, `__tls_base` must be initialized at thread startup, and so cannot be used with dynamic libraries. `__tls_base` is to be initialized with a new linker-synthesized function, `__wasm_init_tls`, which takes as an argument a block of memory to use as the storage for thread locals. It then initializes the block of memory and sets `__tls_base`. As `__wasm_init_tls` will handle the memory initialization, the memory does not have to be zeroed. To help allocating memory for thread-local storage, a new compiler intrinsic is introduced: `__builtin_wasm_tls_size()`. This instrinsic function returns the size of the thread-local storage for the current function. The expected usage is to run something like the following upon thread startup: __wasm_init_tls(malloc(__builtin_wasm_tls_size())); Reviewers: tlively, aheejin, kripken, sbc100 Subscribers: dschuff, jgravelle-google, hiraditya, sunfish, jfb, cfe-commits, llvm-commits Tags: #clang, #llvm Differential Revision: https://reviews.llvm.org/D64537 llvm-svn: 366272
2019-07-17 06:00:45 +08:00
bool StrippedAtomics = false;
bool StrippedTLS = false;
if (!Features[WebAssembly::FeatureAtomics])
StrippedAtomics = stripAtomics(M);
if (!Features[WebAssembly::FeatureBulkMemory])
StrippedTLS = stripThreadLocals(M);
if (StrippedAtomics && !StrippedTLS)
stripThreadLocals(M);
else if (StrippedTLS && !StrippedAtomics)
stripAtomics(M);
[WebAssembly] Implement thread-local storage (local-exec model) Summary: Thread local variables are placed inside a `.tdata` segment. Their symbols are offsets from the start of the segment. The address of a thread local variable is computed as `__tls_base` + the offset from the start of the segment. `.tdata` segment is a passive segment and `memory.init` is used once per thread to initialize the thread local storage. `__tls_base` is a wasm global. Since each thread has its own wasm instance, it is effectively thread local. Currently, `__tls_base` must be initialized at thread startup, and so cannot be used with dynamic libraries. `__tls_base` is to be initialized with a new linker-synthesized function, `__wasm_init_tls`, which takes as an argument a block of memory to use as the storage for thread locals. It then initializes the block of memory and sets `__tls_base`. As `__wasm_init_tls` will handle the memory initialization, the memory does not have to be zeroed. To help allocating memory for thread-local storage, a new compiler intrinsic is introduced: `__builtin_wasm_tls_size()`. This instrinsic function returns the size of the thread-local storage for the current function. The expected usage is to run something like the following upon thread startup: __wasm_init_tls(malloc(__builtin_wasm_tls_size())); Reviewers: tlively, aheejin, kripken, sbc100 Subscribers: dschuff, jgravelle-google, hiraditya, sunfish, jfb, cfe-commits, llvm-commits Tags: #clang, #llvm Differential Revision: https://reviews.llvm.org/D64537 llvm-svn: 366272
2019-07-17 06:00:45 +08:00
recordFeatures(M, Features, StrippedAtomics || StrippedTLS);
// Conservatively assume we have made some change
return true;
}
private:
FeatureBitset coalesceFeatures(const Module &M) {
FeatureBitset Features =
WasmTM
->getSubtargetImpl(std::string(WasmTM->getTargetCPU()),
std::string(WasmTM->getTargetFeatureString()))
->getFeatureBits();
for (auto &F : M)
Features |= WasmTM->getSubtargetImpl(F)->getFeatureBits();
return Features;
}
std::string getFeatureString(const FeatureBitset &Features) {
std::string Ret;
for (const SubtargetFeatureKV &KV : WebAssemblyFeatureKV) {
if (Features[KV.Value])
Ret += (StringRef("+") + KV.Key + ",").str();
}
return Ret;
}
void replaceFeatures(Function &F, const std::string &Features) {
F.removeFnAttr("target-features");
F.removeFnAttr("target-cpu");
F.addFnAttr("target-features", Features);
}
bool stripAtomics(Module &M) {
// Detect whether any atomics will be lowered, since there is no way to tell
// whether the LowerAtomic pass lowers e.g. stores.
bool Stripped = false;
for (auto &F : M) {
for (auto &B : F) {
for (auto &I : B) {
if (I.isAtomic()) {
Stripped = true;
goto done;
}
}
}
}
done:
if (!Stripped)
return false;
LowerAtomicPass Lowerer;
FunctionAnalysisManager FAM;
for (auto &F : M)
Lowerer.run(F, FAM);
return true;
}
bool stripThreadLocals(Module &M) {
bool Stripped = false;
for (auto &GV : M.globals()) {
if (GV.getThreadLocalMode() !=
GlobalValue::ThreadLocalMode::NotThreadLocal) {
Stripped = true;
GV.setThreadLocalMode(GlobalValue::ThreadLocalMode::NotThreadLocal);
}
}
return Stripped;
}
void recordFeatures(Module &M, const FeatureBitset &Features, bool Stripped) {
for (const SubtargetFeatureKV &KV : WebAssemblyFeatureKV) {
[WebAssembly] Disallow 'shared-mem' rather than 'atomics' Summary: The WebAssembly backend automatically lowers atomic operations and TLS to nonatomic operations and non-TLS data when either are present and the atomics or bulk-memory features are not present, respectively. The resulting object is no longer thread-safe, so the linker has to be told not to allow it to be linked into a module with shared memory. This was previously done by disallowing the 'atomics' feature, which prevented any objct with its atomic operations or TLS removed from being linked with any object containing atomics or TLS, and therefore preventing it from being linked into a module with shared memory since shared memory requires atomics. However, as of https://github.com/WebAssembly/threads/issues/144, the validation rules are relaxed to allow atomic operations to validate with unshared memories, which makes it perfectly safe to link an object with stripped atomics and TLS with another object that still contains TLS and atomics as long as the resulting module has an unshared memory. To allow this kind of link, this patch disallows a pseudo-feature 'shared-mem' rather than 'atomics' to communicate to the linker that the object is not thread-safe. This means that the 'atomics' feature is available to accurately reflect whether or not an object has atomics enabled. As a drive-by tweak, this change also requires that bulk-memory be enabled in addition to atomics in order to use shared memory. This is because initializing shared memories requires bulk-memory operations. Reviewers: aheejin, sbc100 Subscribers: dschuff, jgravelle-google, hiraditya, sunfish, jfb, llvm-commits Tags: #llvm Differential Revision: https://reviews.llvm.org/D79542
2020-05-07 10:33:24 +08:00
if (Features[KV.Value]) {
// Mark features as used
std::string MDKey = (StringRef("wasm-feature-") + KV.Key).str();
M.addModuleFlag(Module::ModFlagBehavior::Error, MDKey,
wasm::WASM_FEATURE_PREFIX_USED);
}
}
[WebAssembly] Disallow 'shared-mem' rather than 'atomics' Summary: The WebAssembly backend automatically lowers atomic operations and TLS to nonatomic operations and non-TLS data when either are present and the atomics or bulk-memory features are not present, respectively. The resulting object is no longer thread-safe, so the linker has to be told not to allow it to be linked into a module with shared memory. This was previously done by disallowing the 'atomics' feature, which prevented any objct with its atomic operations or TLS removed from being linked with any object containing atomics or TLS, and therefore preventing it from being linked into a module with shared memory since shared memory requires atomics. However, as of https://github.com/WebAssembly/threads/issues/144, the validation rules are relaxed to allow atomic operations to validate with unshared memories, which makes it perfectly safe to link an object with stripped atomics and TLS with another object that still contains TLS and atomics as long as the resulting module has an unshared memory. To allow this kind of link, this patch disallows a pseudo-feature 'shared-mem' rather than 'atomics' to communicate to the linker that the object is not thread-safe. This means that the 'atomics' feature is available to accurately reflect whether or not an object has atomics enabled. As a drive-by tweak, this change also requires that bulk-memory be enabled in addition to atomics in order to use shared memory. This is because initializing shared memories requires bulk-memory operations. Reviewers: aheejin, sbc100 Subscribers: dschuff, jgravelle-google, hiraditya, sunfish, jfb, llvm-commits Tags: #llvm Differential Revision: https://reviews.llvm.org/D79542
2020-05-07 10:33:24 +08:00
// Code compiled without atomics or bulk-memory may have had its atomics or
// thread-local data lowered to nonatomic operations or non-thread-local
// data. In that case, we mark the pseudo-feature "shared-mem" as disallowed
// to tell the linker that it would be unsafe to allow this code ot be used
// in a module with shared memory.
if (Stripped) {
M.addModuleFlag(Module::ModFlagBehavior::Error, "wasm-feature-shared-mem",
wasm::WASM_FEATURE_PREFIX_DISALLOWED);
}
}
};
char CoalesceFeaturesAndStripAtomics::ID = 0;
/// WebAssembly Code Generator Pass Configuration Options.
class WebAssemblyPassConfig final : public TargetPassConfig {
public:
WebAssemblyPassConfig(WebAssemblyTargetMachine &TM, PassManagerBase &PM)
: TargetPassConfig(TM, PM) {}
WebAssemblyTargetMachine &getWebAssemblyTargetMachine() const {
return getTM<WebAssemblyTargetMachine>();
}
FunctionPass *createTargetRegisterAllocator(bool) override;
void addIRPasses() override;
bool addInstSelector() override;
void addPostRegAlloc() override;
bool addGCPasses() override { return false; }
void addPreEmitPass() override;
// No reg alloc
bool addRegAssignmentFast() override { return false; }
// No reg alloc
bool addRegAssignmentOptimized() override { return false; }
};
} // end anonymous namespace
TargetTransformInfo
WebAssemblyTargetMachine::getTargetTransformInfo(const Function &F) {
return TargetTransformInfo(WebAssemblyTTIImpl(this, F));
}
TargetPassConfig *
WebAssemblyTargetMachine::createPassConfig(PassManagerBase &PM) {
return new WebAssemblyPassConfig(*this, PM);
}
FunctionPass *WebAssemblyPassConfig::createTargetRegisterAllocator(bool) {
return nullptr; // No reg alloc
}
//===----------------------------------------------------------------------===//
// The following functions are called from lib/CodeGen/Passes.cpp to modify
// the CodeGen pass sequence.
//===----------------------------------------------------------------------===//
void WebAssemblyPassConfig::addIRPasses() {
// Runs LowerAtomicPass if necessary
addPass(new CoalesceFeaturesAndStripAtomics(&getWebAssemblyTargetMachine()));
// This is a no-op if atomics are not used in the module
addPass(createAtomicExpandPass());
// Add signatures to prototype-less function declarations
addPass(createWebAssemblyAddMissingPrototypes());
// Lower .llvm.global_dtors into .llvm_global_ctors with __cxa_atexit calls.
addPass(createWebAssemblyLowerGlobalDtors());
// Fix function bitcasts, as WebAssembly requires caller and callee signatures
// to match.
addPass(createWebAssemblyFixFunctionBitcasts());
// Optimize "returned" function attributes.
if (getOptLevel() != CodeGenOpt::None)
addPass(createWebAssemblyOptimizeReturned());
// If exception handling is not enabled and setjmp/longjmp handling is
// enabled, we lower invokes into calls and delete unreachable landingpad
// blocks. Lowering invokes when there is no EH support is done in
// TargetPassConfig::addPassesToHandleExceptions, but this runs after this
// function and SjLj handling expects all invokes to be lowered before.
if (!EnableEmException &&
TM->Options.ExceptionModel == ExceptionHandling::None) {
addPass(createLowerInvokePass());
// The lower invoke pass may create unreachable code. Remove it in order not
// to process dead blocks in setjmp/longjmp handling.
addPass(createUnreachableBlockEliminationPass());
}
// Handle exceptions and setjmp/longjmp if enabled.
if (EnableEmException || EnableEmSjLj)
addPass(createWebAssemblyLowerEmscriptenEHSjLj(EnableEmException,
EnableEmSjLj));
// Expand indirectbr instructions to switches.
addPass(createIndirectBrExpandPass());
TargetPassConfig::addIRPasses();
}
bool WebAssemblyPassConfig::addInstSelector() {
(void)TargetPassConfig::addInstSelector();
addPass(
createWebAssemblyISelDag(getWebAssemblyTargetMachine(), getOptLevel()));
// Run the argument-move pass immediately after the ScheduleDAG scheduler
// so that we can fix up the ARGUMENT instructions before anything else
// sees them in the wrong place.
addPass(createWebAssemblyArgumentMove());
// Set the p2align operands. This information is present during ISel, however
// it's inconvenient to collect. Collect it now, and update the immediate
// operands.
addPass(createWebAssemblySetP2AlignOperands());
return false;
}
void WebAssemblyPassConfig::addPostRegAlloc() {
// TODO: The following CodeGen passes don't currently support code containing
// virtual registers. Consider removing their restrictions and re-enabling
// them.
// These functions all require the NoVRegs property.
disablePass(&MachineCopyPropagationID);
disablePass(&PostRAMachineSinkingID);
disablePass(&PostRASchedulerID);
disablePass(&FuncletLayoutID);
disablePass(&StackMapLivenessID);
disablePass(&LiveDebugValuesID);
disablePass(&PatchableFunctionID);
disablePass(&ShrinkWrapID);
// This pass hurts code size for wasm because it can generate irreducible
// control flow.
disablePass(&MachineBlockPlacementID);
TargetPassConfig::addPostRegAlloc();
}
void WebAssemblyPassConfig::addPreEmitPass() {
TargetPassConfig::addPreEmitPass();
// Eliminate multiple-entry loops.
addPass(createWebAssemblyFixIrreducibleControlFlow());
// Do various transformations for exception handling.
[WebAssembly] Exception handling: Switch to the new proposal Summary: This switches the EH implementation to the new proposal: https://github.com/WebAssembly/exception-handling/blob/master/proposals/Exceptions.md (The previous proposal was https://github.com/WebAssembly/exception-handling/blob/master/proposals/old/Exceptions.md) - Instruction changes - Now we have one single `catch` instruction that returns a except_ref value - `throw` now can take variable number of operations - `rethrow` does not have 'depth' argument anymore - `br_on_exn` queries an except_ref to see if it matches the tag and branches to the given label if true. - `extract_exception` is a pseudo instruction that simulates popping values from wasm stack. This is to make `br_on_exn`, a very special instruction, work: `br_on_exn` puts values onto the stack only if it is taken, and the # of values can vay depending on the tag. - Now there's only one `catch` per `try`, this patch removes all special handling for terminate pad with a call to `__clang_call_terminate`. Before it was the only case there are two catch clauses (a normal `catch` and `catch_all` per `try`). - Make `rethrow` act as a terminator like `throw`. This splits BB after `rethrow` in WasmEHPrepare, and deletes an unnecessary `unreachable` after `rethrow` in LateEHPrepare. - Now we stop at all catchpads (because we add wasm `catch` instruction that catches all exceptions), this creates new `findWasmUnwindDestinations` function in SelectionDAGBuilder. - Now we use `br_on_exn` instrution to figure out if an except_ref matches the current tag or not, LateEHPrepare generates this sequence for catch pads: ``` catch block i32 br_on_exn $__cpp_exception end_block extract_exception ``` - Branch analysis for `br_on_exn` in WebAssemblyInstrInfo - Other various misc. changes to switch to the new proposal. Reviewers: dschuff Subscribers: sbc100, jgravelle-google, sunfish, llvm-commits Differential Revision: https://reviews.llvm.org/D57134 llvm-svn: 352598
2019-01-30 11:21:57 +08:00
// Every CFG-changing optimizations should come before this.
addPass(createWebAssemblyLateEHPrepare());
// Now that we have a prologue and epilogue and all frame indices are
// rewritten, eliminate SP and FP. This allows them to be stackified,
// colored, and numbered with the rest of the registers.
addPass(createWebAssemblyReplacePhysRegs());
[WebAssembly] Exception handling: Switch to the new proposal Summary: This switches the EH implementation to the new proposal: https://github.com/WebAssembly/exception-handling/blob/master/proposals/Exceptions.md (The previous proposal was https://github.com/WebAssembly/exception-handling/blob/master/proposals/old/Exceptions.md) - Instruction changes - Now we have one single `catch` instruction that returns a except_ref value - `throw` now can take variable number of operations - `rethrow` does not have 'depth' argument anymore - `br_on_exn` queries an except_ref to see if it matches the tag and branches to the given label if true. - `extract_exception` is a pseudo instruction that simulates popping values from wasm stack. This is to make `br_on_exn`, a very special instruction, work: `br_on_exn` puts values onto the stack only if it is taken, and the # of values can vay depending on the tag. - Now there's only one `catch` per `try`, this patch removes all special handling for terminate pad with a call to `__clang_call_terminate`. Before it was the only case there are two catch clauses (a normal `catch` and `catch_all` per `try`). - Make `rethrow` act as a terminator like `throw`. This splits BB after `rethrow` in WasmEHPrepare, and deletes an unnecessary `unreachable` after `rethrow` in LateEHPrepare. - Now we stop at all catchpads (because we add wasm `catch` instruction that catches all exceptions), this creates new `findWasmUnwindDestinations` function in SelectionDAGBuilder. - Now we use `br_on_exn` instrution to figure out if an except_ref matches the current tag or not, LateEHPrepare generates this sequence for catch pads: ``` catch block i32 br_on_exn $__cpp_exception end_block extract_exception ``` - Branch analysis for `br_on_exn` in WebAssemblyInstrInfo - Other various misc. changes to switch to the new proposal. Reviewers: dschuff Subscribers: sbc100, jgravelle-google, sunfish, llvm-commits Differential Revision: https://reviews.llvm.org/D57134 llvm-svn: 352598
2019-01-30 11:21:57 +08:00
// Preparations and optimizations related to register stackification.
if (getOptLevel() != CodeGenOpt::None) {
// LiveIntervals isn't commonly run this late. Re-establish preconditions.
addPass(createWebAssemblyPrepareForLiveIntervals());
// Depend on LiveIntervals and perform some optimizations on it.
addPass(createWebAssemblyOptimizeLiveIntervals());
// Prepare memory intrinsic calls for register stackifying.
addPass(createWebAssemblyMemIntrinsicResults());
// Mark registers as representing wasm's value stack. This is a key
// code-compression technique in WebAssembly. We run this pass (and
// MemIntrinsicResults above) very late, so that it sees as much code as
// possible, including code emitted by PEI and expanded by late tail
// duplication.
addPass(createWebAssemblyRegStackify());
// Run the register coloring pass to reduce the total number of registers.
// This runs after stackification so that it doesn't consider registers
// that become stackified.
addPass(createWebAssemblyRegColoring());
}
// Sort the blocks of the CFG into topological order, a prerequisite for
// BLOCK and LOOP markers.
addPass(createWebAssemblyCFGSort());
// Insert BLOCK and LOOP markers.
addPass(createWebAssemblyCFGStackify());
// Insert explicit local.get and local.set operators.
if (!WasmDisableExplicitLocals)
addPass(createWebAssemblyExplicitLocals());
// Lower br_unless into br_if.
addPass(createWebAssemblyLowerBrUnless());
// Perform the very last peephole optimizations on the code.
if (getOptLevel() != CodeGenOpt::None)
addPass(createWebAssemblyPeephole());
// Create a mapping from LLVM CodeGen virtual registers to wasm registers.
addPass(createWebAssemblyRegNumbering());
// Fix debug_values whose defs have been stackified.
if (!WasmDisableExplicitLocals)
addPass(createWebAssemblyDebugFixup());
}
yaml::MachineFunctionInfo *
WebAssemblyTargetMachine::createDefaultFuncInfoYAML() const {
return new yaml::WebAssemblyFunctionInfo();
}
yaml::MachineFunctionInfo *WebAssemblyTargetMachine::convertFuncInfoToYAML(
const MachineFunction &MF) const {
const auto *MFI = MF.getInfo<WebAssemblyFunctionInfo>();
return new yaml::WebAssemblyFunctionInfo(*MFI);
}
bool WebAssemblyTargetMachine::parseMachineFunctionInfo(
const yaml::MachineFunctionInfo &MFI, PerFunctionMIParsingState &PFS,
SMDiagnostic &Error, SMRange &SourceRange) const {
const auto &YamlMFI =
reinterpret_cast<const yaml::WebAssemblyFunctionInfo &>(MFI);
MachineFunction &MF = PFS.MF;
MF.getInfo<WebAssemblyFunctionInfo>()->initializeBaseYamlFields(YamlMFI);
return false;
}