v8/test/common/wasm/test-signatures.h

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

196 lines
6.2 KiB
C
Raw Normal View History

// Copyright 2015 the V8 project authors. All rights reserved.
// Use of this source code is governed by a BSD-style license that can be
// found in the LICENSE file.
#ifndef TEST_SIGNATURES_H
#define TEST_SIGNATURES_H
#include "src/codegen/signature.h"
#include "src/wasm/value-type.h"
#include "src/wasm/wasm-opcodes.h"
namespace v8 {
namespace internal {
namespace wasm {
// A helper class with many useful signatures in order to simplify tests.
class TestSignatures {
public:
TestSignatures()
: sig_i_v(1, 0, kIntTypes4),
sig_i_i(1, 1, kIntTypes4),
sig_i_ii(1, 2, kIntTypes4),
sig_i_iii(1, 3, kIntTypes4),
sig_i_f(1, 1, kIntFloatTypes4),
sig_i_ff(1, 2, kIntFloatTypes4),
sig_i_d(1, 1, kIntDoubleTypes4),
sig_i_dd(1, 2, kIntDoubleTypes4),
sig_i_e(1, 1, kIntExternRefTypes4),
sig_i_ee(1, 2, kIntExternRefTypes4),
sig_i_c(1, 1, kIntFuncRefTypes4),
sig_i_s(1, 1, kIntSimd128Types4),
sig_l_v(1, 0, kLongTypes4),
sig_l_l(1, 1, kLongTypes4),
sig_l_ll(1, 2, kLongTypes4),
sig_i_ll(1, 2, kIntLongTypes4),
sig_f_f(1, 1, kFloatTypes4),
sig_f_ff(1, 2, kFloatTypes4),
sig_d_d(1, 1, kDoubleTypes4),
sig_d_dd(1, 2, kDoubleTypes4),
sig_e_v(1, 0, kExternRefTypes4),
sig_c_v(1, 0, kFuncTypes4),
sig_e_e(1, 1, kExternRefTypes4),
sig_c_c(1, 1, kFuncTypes4),
sig_v_v(0, 0, kIntTypes4),
sig_v_i(0, 1, kIntTypes4),
sig_v_ii(0, 2, kIntTypes4),
sig_v_iii(0, 3, kIntTypes4),
sig_v_e(0, 1, kExternRefTypes4),
sig_v_c(0, 1, kFuncTypes4),
[compiler] Re-reland "Faster JS-to-Wasm calls" This is a reland of 6ada6a90ee387d8de183208b0ef8b786f1768665 - Fixed a GC issue https://bugs.chromium.org/p/v8/issues/detail?id=11335: GC expected all arguments on the stack from code with CodeKind::TURBOFAN to be tagged objects. This is not the case now with inlined Wasm calls, and this information can be passed in SafepointEntry for each call site. - Disabled JS-to-Wasm inlining for calls inside try/catch. For more details, see updated doc: https://docs.google.com/document/d/1mXxYnYN77tK-R1JOVo6tFG3jNpMzfueQN1Zp5h3r9aM/edit# Bug: v8:11092 Original change's description: > Reland "Faster JS-to-Wasm calls" > > This is a reland of 860fcb1bd2bd6447e08f3636874ac7abcd77b781 > > - Disabled the tests for this feature in V8-lite mode (the original > change broke V8-lite tests). > - Also modified test console-profile-wasm.js that was brittle with this > change because it assumed that there was always a JS-to-Wasm wrapper > but this is not the case when the TurboFan compilation completes before > the Liftoff-compiled code starts to run. > > More changes in Patchset 8: > > - Moved inlining of the "JSToWasm Wrapper" away from simplified-lowering, > into a new phase, wasm-inlining that reuses the JSInliner reducer. > The doc > https://docs.google.com/document/d/1mXxYnYN77tK-R1JOVo6tFG3jNpMzfueQN1Zp5h3r9aM/edit# > describes the new logic. > > - Fixed a couple of small issues in wasm_compiler.cc to make sure that > the graph "JSToWasm Wrapper" subgraph has a valid Control chain; > this should solve the problem we had inlining the calls in functions > that can throw exception. Original change's description: > Faster JS-to-Wasm calls > > This replaces https://chromium-review.googlesource.com/c/v8/v8/+/2376165/. > > Currently JS-to-Wasm calls go through a wrapper/trampoline, built on > the basis of the signature of a Wasm function to call, and whose task > is to: > - set "thread_in_wasm_flag" to true > - convert the arguments from tagged types into Wasm native types > - calculate the address of the Wasm function to call and call it > - convert back the result from Wasm native types into tagged types > - reset "thread_in_wasm_flag" to false. > > This CL tries to improve the performance of JS-to-Wasm calls by > inlining the code of the JS-to-Wasm wrappers in the call site. > > It introduces a new IR operand, JSWasmCall, which replaces JSCall for > this kind of calls. A 'JSWasmCall' node is associated to > WasmCallParameters, which contain information about the signature of > the Wasm function to call. > > WasmWrapperGraphBuilder::BuildJSToWasmWrapper is modified to avoid > generating code to convert the types for the arguments > of the Wasm function, when the conversion is not necessary. > The actual inlining of the graph generated for this wrapper happens in > the simplified-lowering phase. > > A new builtin, JSToWasmLazyDeoptContinuation, is introduced to manage > lazy deoptimizations that can happen if the Wasm function callee calls > back some JS code that invalidates the compiled JS caller function. > Bug: v8:11092 Cq-Include-Trybots: luci.v8.try:v8_linux_arm_lite_rel_ng Change-Id: Ie052634598754feab4ff36d10fd04e008b5227a5 Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2649777 Commit-Queue: Paolo Severini <paolosev@microsoft.com> Reviewed-by: Jakob Gruber <jgruber@chromium.org> Reviewed-by: Andreas Haas <ahaas@chromium.org> Reviewed-by: Georg Neis <neis@chromium.org> Cr-Commit-Position: refs/heads/master@{#72541}
2021-02-04 15:46:10 +00:00
sig_v_d(0, 1, kDoubleTypes4),
sig_s_i(1, 1, kSimd128IntTypes4),
sig_s_s(1, 1, kSimd128Types4),
sig_s_ss(1, 2, kSimd128Types4),
sig_ii_v(2, 0, kIntTypes4),
sig_iii_v(3, 0, kIntTypes4) {
// I used C++ and you won't believe what happened next....
for (int i = 0; i < 4; i++) kIntTypes4[i] = kWasmI32;
for (int i = 0; i < 4; i++) kLongTypes4[i] = kWasmI64;
for (int i = 0; i < 4; i++) kFloatTypes4[i] = kWasmF32;
for (int i = 0; i < 4; i++) kDoubleTypes4[i] = kWasmF64;
for (int i = 0; i < 4; i++) kExternRefTypes4[i] = kWasmExternRef;
for (int i = 0; i < 4; i++) kFuncTypes4[i] = kWasmFuncRef;
for (int i = 1; i < 4; i++) kIntLongTypes4[i] = kWasmI64;
for (int i = 1; i < 4; i++) kIntFloatTypes4[i] = kWasmF32;
for (int i = 1; i < 4; i++) kIntDoubleTypes4[i] = kWasmF64;
for (int i = 1; i < 4; i++) kIntExternRefTypes4[i] = kWasmExternRef;
for (int i = 1; i < 4; i++) kIntFuncRefTypes4[i] = kWasmFuncRef;
for (int i = 0; i < 4; i++) kSimd128Types4[i] = kWasmS128;
for (int i = 1; i < 4; i++) kIntSimd128Types4[i] = kWasmS128;
for (int i = 0; i < 4; i++) kSimd128IntTypes4[i] = kWasmS128;
kIntLongTypes4[0] = kWasmI32;
kIntFloatTypes4[0] = kWasmI32;
kIntDoubleTypes4[0] = kWasmI32;
kIntExternRefTypes4[0] = kWasmI32;
kIntFuncRefTypes4[0] = kWasmI32;
kIntSimd128Types4[0] = kWasmI32;
kSimd128IntTypes4[1] = kWasmI32;
}
FunctionSig* i_v() { return &sig_i_v; }
FunctionSig* i_i() { return &sig_i_i; }
FunctionSig* i_ii() { return &sig_i_ii; }
FunctionSig* i_iii() { return &sig_i_iii; }
FunctionSig* i_f() { return &sig_i_f; }
FunctionSig* i_ff() { return &sig_i_ff; }
FunctionSig* i_d() { return &sig_i_d; }
FunctionSig* i_dd() { return &sig_i_dd; }
FunctionSig* l_v() { return &sig_l_v; }
FunctionSig* l_l() { return &sig_l_l; }
FunctionSig* l_ll() { return &sig_l_ll; }
FunctionSig* i_ll() { return &sig_i_ll; }
FunctionSig* i_e() { return &sig_i_e; }
FunctionSig* i_ee() { return &sig_i_ee; }
FunctionSig* i_c() { return &sig_i_c; }
FunctionSig* i_s() { return &sig_i_s; }
FunctionSig* f_f() { return &sig_f_f; }
FunctionSig* f_ff() { return &sig_f_ff; }
FunctionSig* d_d() { return &sig_d_d; }
FunctionSig* d_dd() { return &sig_d_dd; }
FunctionSig* e_v() { return &sig_e_v; }
FunctionSig* a_v() { return &sig_c_v; }
FunctionSig* e_e() { return &sig_e_e; }
FunctionSig* c_c() { return &sig_c_c; }
FunctionSig* v_v() { return &sig_v_v; }
FunctionSig* v_i() { return &sig_v_i; }
FunctionSig* v_ii() { return &sig_v_ii; }
FunctionSig* v_iii() { return &sig_v_iii; }
FunctionSig* v_e() { return &sig_v_e; }
FunctionSig* v_c() { return &sig_v_c; }
[compiler] Re-reland "Faster JS-to-Wasm calls" This is a reland of 6ada6a90ee387d8de183208b0ef8b786f1768665 - Fixed a GC issue https://bugs.chromium.org/p/v8/issues/detail?id=11335: GC expected all arguments on the stack from code with CodeKind::TURBOFAN to be tagged objects. This is not the case now with inlined Wasm calls, and this information can be passed in SafepointEntry for each call site. - Disabled JS-to-Wasm inlining for calls inside try/catch. For more details, see updated doc: https://docs.google.com/document/d/1mXxYnYN77tK-R1JOVo6tFG3jNpMzfueQN1Zp5h3r9aM/edit# Bug: v8:11092 Original change's description: > Reland "Faster JS-to-Wasm calls" > > This is a reland of 860fcb1bd2bd6447e08f3636874ac7abcd77b781 > > - Disabled the tests for this feature in V8-lite mode (the original > change broke V8-lite tests). > - Also modified test console-profile-wasm.js that was brittle with this > change because it assumed that there was always a JS-to-Wasm wrapper > but this is not the case when the TurboFan compilation completes before > the Liftoff-compiled code starts to run. > > More changes in Patchset 8: > > - Moved inlining of the "JSToWasm Wrapper" away from simplified-lowering, > into a new phase, wasm-inlining that reuses the JSInliner reducer. > The doc > https://docs.google.com/document/d/1mXxYnYN77tK-R1JOVo6tFG3jNpMzfueQN1Zp5h3r9aM/edit# > describes the new logic. > > - Fixed a couple of small issues in wasm_compiler.cc to make sure that > the graph "JSToWasm Wrapper" subgraph has a valid Control chain; > this should solve the problem we had inlining the calls in functions > that can throw exception. Original change's description: > Faster JS-to-Wasm calls > > This replaces https://chromium-review.googlesource.com/c/v8/v8/+/2376165/. > > Currently JS-to-Wasm calls go through a wrapper/trampoline, built on > the basis of the signature of a Wasm function to call, and whose task > is to: > - set "thread_in_wasm_flag" to true > - convert the arguments from tagged types into Wasm native types > - calculate the address of the Wasm function to call and call it > - convert back the result from Wasm native types into tagged types > - reset "thread_in_wasm_flag" to false. > > This CL tries to improve the performance of JS-to-Wasm calls by > inlining the code of the JS-to-Wasm wrappers in the call site. > > It introduces a new IR operand, JSWasmCall, which replaces JSCall for > this kind of calls. A 'JSWasmCall' node is associated to > WasmCallParameters, which contain information about the signature of > the Wasm function to call. > > WasmWrapperGraphBuilder::BuildJSToWasmWrapper is modified to avoid > generating code to convert the types for the arguments > of the Wasm function, when the conversion is not necessary. > The actual inlining of the graph generated for this wrapper happens in > the simplified-lowering phase. > > A new builtin, JSToWasmLazyDeoptContinuation, is introduced to manage > lazy deoptimizations that can happen if the Wasm function callee calls > back some JS code that invalidates the compiled JS caller function. > Bug: v8:11092 Cq-Include-Trybots: luci.v8.try:v8_linux_arm_lite_rel_ng Change-Id: Ie052634598754feab4ff36d10fd04e008b5227a5 Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2649777 Commit-Queue: Paolo Severini <paolosev@microsoft.com> Reviewed-by: Jakob Gruber <jgruber@chromium.org> Reviewed-by: Andreas Haas <ahaas@chromium.org> Reviewed-by: Georg Neis <neis@chromium.org> Cr-Commit-Position: refs/heads/master@{#72541}
2021-02-04 15:46:10 +00:00
FunctionSig* v_d() { return &sig_v_d; }
FunctionSig* s_i() { return &sig_s_i; }
FunctionSig* s_s() { return &sig_s_s; }
FunctionSig* s_ss() { return &sig_s_ss; }
FunctionSig* ii_v() { return &sig_ii_v; }
FunctionSig* iii_v() { return &sig_iii_v; }
FunctionSig* many(Zone* zone, ValueType ret, ValueType param, int count) {
FunctionSig::Builder builder(zone, ret == kWasmVoid ? 0 : 1, count);
if (ret != kWasmVoid) builder.AddReturn(ret);
for (int i = 0; i < count; i++) {
builder.AddParam(param);
}
return builder.Build();
}
private:
ValueType kIntTypes4[4];
ValueType kLongTypes4[4];
ValueType kFloatTypes4[4];
ValueType kDoubleTypes4[4];
ValueType kExternRefTypes4[4];
ValueType kFuncTypes4[4];
ValueType kIntLongTypes4[4];
ValueType kIntFloatTypes4[4];
ValueType kIntDoubleTypes4[4];
ValueType kIntExternRefTypes4[4];
ValueType kIntFuncRefTypes4[4];
ValueType kSimd128Types4[4];
ValueType kIntSimd128Types4[4];
ValueType kSimd128IntTypes4[4];
FunctionSig sig_i_v;
FunctionSig sig_i_i;
FunctionSig sig_i_ii;
FunctionSig sig_i_iii;
FunctionSig sig_i_f;
FunctionSig sig_i_ff;
FunctionSig sig_i_d;
FunctionSig sig_i_dd;
FunctionSig sig_i_e;
FunctionSig sig_i_ee;
FunctionSig sig_i_c;
FunctionSig sig_i_s;
FunctionSig sig_l_v;
FunctionSig sig_l_l;
FunctionSig sig_l_ll;
FunctionSig sig_i_ll;
FunctionSig sig_f_f;
FunctionSig sig_f_ff;
FunctionSig sig_d_d;
FunctionSig sig_d_dd;
FunctionSig sig_e_v;
FunctionSig sig_c_v;
FunctionSig sig_e_e;
FunctionSig sig_c_c;
FunctionSig sig_v_v;
FunctionSig sig_v_i;
FunctionSig sig_v_ii;
FunctionSig sig_v_iii;
FunctionSig sig_v_e;
FunctionSig sig_v_c;
[compiler] Re-reland "Faster JS-to-Wasm calls" This is a reland of 6ada6a90ee387d8de183208b0ef8b786f1768665 - Fixed a GC issue https://bugs.chromium.org/p/v8/issues/detail?id=11335: GC expected all arguments on the stack from code with CodeKind::TURBOFAN to be tagged objects. This is not the case now with inlined Wasm calls, and this information can be passed in SafepointEntry for each call site. - Disabled JS-to-Wasm inlining for calls inside try/catch. For more details, see updated doc: https://docs.google.com/document/d/1mXxYnYN77tK-R1JOVo6tFG3jNpMzfueQN1Zp5h3r9aM/edit# Bug: v8:11092 Original change's description: > Reland "Faster JS-to-Wasm calls" > > This is a reland of 860fcb1bd2bd6447e08f3636874ac7abcd77b781 > > - Disabled the tests for this feature in V8-lite mode (the original > change broke V8-lite tests). > - Also modified test console-profile-wasm.js that was brittle with this > change because it assumed that there was always a JS-to-Wasm wrapper > but this is not the case when the TurboFan compilation completes before > the Liftoff-compiled code starts to run. > > More changes in Patchset 8: > > - Moved inlining of the "JSToWasm Wrapper" away from simplified-lowering, > into a new phase, wasm-inlining that reuses the JSInliner reducer. > The doc > https://docs.google.com/document/d/1mXxYnYN77tK-R1JOVo6tFG3jNpMzfueQN1Zp5h3r9aM/edit# > describes the new logic. > > - Fixed a couple of small issues in wasm_compiler.cc to make sure that > the graph "JSToWasm Wrapper" subgraph has a valid Control chain; > this should solve the problem we had inlining the calls in functions > that can throw exception. Original change's description: > Faster JS-to-Wasm calls > > This replaces https://chromium-review.googlesource.com/c/v8/v8/+/2376165/. > > Currently JS-to-Wasm calls go through a wrapper/trampoline, built on > the basis of the signature of a Wasm function to call, and whose task > is to: > - set "thread_in_wasm_flag" to true > - convert the arguments from tagged types into Wasm native types > - calculate the address of the Wasm function to call and call it > - convert back the result from Wasm native types into tagged types > - reset "thread_in_wasm_flag" to false. > > This CL tries to improve the performance of JS-to-Wasm calls by > inlining the code of the JS-to-Wasm wrappers in the call site. > > It introduces a new IR operand, JSWasmCall, which replaces JSCall for > this kind of calls. A 'JSWasmCall' node is associated to > WasmCallParameters, which contain information about the signature of > the Wasm function to call. > > WasmWrapperGraphBuilder::BuildJSToWasmWrapper is modified to avoid > generating code to convert the types for the arguments > of the Wasm function, when the conversion is not necessary. > The actual inlining of the graph generated for this wrapper happens in > the simplified-lowering phase. > > A new builtin, JSToWasmLazyDeoptContinuation, is introduced to manage > lazy deoptimizations that can happen if the Wasm function callee calls > back some JS code that invalidates the compiled JS caller function. > Bug: v8:11092 Cq-Include-Trybots: luci.v8.try:v8_linux_arm_lite_rel_ng Change-Id: Ie052634598754feab4ff36d10fd04e008b5227a5 Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2649777 Commit-Queue: Paolo Severini <paolosev@microsoft.com> Reviewed-by: Jakob Gruber <jgruber@chromium.org> Reviewed-by: Andreas Haas <ahaas@chromium.org> Reviewed-by: Georg Neis <neis@chromium.org> Cr-Commit-Position: refs/heads/master@{#72541}
2021-02-04 15:46:10 +00:00
FunctionSig sig_v_d;
FunctionSig sig_s_i;
FunctionSig sig_s_s;
FunctionSig sig_s_ss;
FunctionSig sig_ii_v;
FunctionSig sig_iii_v;
};
} // namespace wasm
} // namespace internal
} // namespace v8
#endif // TEST_SIGNATURES_H