AuroraMiddleware/v8
1
0
Fork 0
Code Issues 2 Pull Requests Releases Wiki Activity
44c24bf827
v8/test/cctest/interpreter/bytecode_expectations/StandardForLoop.golden

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

574 lines
17 KiB
Plaintext
Raw Normal View History

[parser] avoid complex for-loop desugaring when possible let/const declarations in "standard" C-style for-loops have some complex desugaring to accommodate the case where loop loop variables may be captured. This slows down the baseline performance of for-loops with let variables. This change attempts to avoid this desugaring if it's known that the loop variable is not captured at any point. A side effect of this change is that let/const loop variables, when not captured within the loop body, are not necessarily shown in the debugger, similar to other stack-allocated vars. BUG=v8:4762, v8:5460 R=marja@chromium.org, adamk@chromium.org, yangguo@chromium.org Change-Id: I8dbe545a12c086f675972bdba60c94998268311a Reviewed-on: https://chromium-review.googlesource.com/472247 Commit-Queue: Caitlin Potter <caitp@igalia.com> Reviewed-by: Yang Guo <yangguo@chromium.org> Reviewed-by: Adam Klein <adamk@chromium.org> Reviewed-by: Marja Hölttä <marja@chromium.org> Cr-Commit-Position: refs/heads/master@{#44731}
2017-04-18 14:56:06 +00:00
#
# Autogenerated by generate-bytecode-expectations.
#
---
wrap: no
test function name: f
---
snippet: "
function f() {
for (let x = 0; x < 10; ++x) { let y = x; }
}
f();
"
frame size: 2
parameter count: 1
Skip hole initialization of lexical variables when possible This patch expands scope analysis to skip hole initialization when it can be determined statically that no hole checks will be generated at runtime. Two conditions must be met to safely eliminate hole initialization: - There must not exist a VariableProxy referencing this Variable whose HoleCheckMode is kRequired - The Variable must be stack allocated; any other allocation implies that it may be accessed from not-yet-analyzed scopes (other modules, inner functions, or eval code) and that code may require hole checks. The new logic required removing debug code in full-codegen which is now incorrect in some cases. Also fixed Variable's bitfield helpers to take no more space than needed. Bug: chromium:651637 Change-Id: Ie5ac326af4e05b7a5c3c37cd4d0afba6a51a504d Reviewed-on: https://chromium-review.googlesource.com/494006 Commit-Queue: Adam Klein <adamk@chromium.org> Reviewed-by: Georg Neis <neis@chromium.org> Reviewed-by: Leszek Swirski <leszeks@chromium.org> Reviewed-by: Michael Starzinger <mstarzinger@chromium.org> Cr-Commit-Position: refs/heads/master@{#45170}
2017-05-08 18:55:54 +00:00
bytecode array length: 26
[parser] avoid complex for-loop desugaring when possible let/const declarations in "standard" C-style for-loops have some complex desugaring to accommodate the case where loop loop variables may be captured. This slows down the baseline performance of for-loops with let variables. This change attempts to avoid this desugaring if it's known that the loop variable is not captured at any point. A side effect of this change is that let/const loop variables, when not captured within the loop body, are not necessarily shown in the debugger, similar to other stack-allocated vars. BUG=v8:4762, v8:5460 R=marja@chromium.org, adamk@chromium.org, yangguo@chromium.org Change-Id: I8dbe545a12c086f675972bdba60c94998268311a Reviewed-on: https://chromium-review.googlesource.com/472247 Commit-Queue: Caitlin Potter <caitp@igalia.com> Reviewed-by: Yang Guo <yangguo@chromium.org> Reviewed-by: Adam Klein <adamk@chromium.org> Reviewed-by: Marja Hölttä <marja@chromium.org> Cr-Commit-Position: refs/heads/master@{#44731}
2017-04-18 14:56:06 +00:00
bytecodes: [
/* 10 E> */ B(StackCheck),
/* 30 S> */ B(LdaZero),
B(Star), R(1),
/* 35 S> */ B(LdaSmi), I8(10),
[objects] Make feedback vector a first-class object Instead of having feedback vector as a subtype of FixedArray with reserved slots, make it a first-class variable-sized object with a fixed-size header. This allows us to compress counters to ints in the header, rather than forcing them to be Smis. Cq-Include-Trybots: master.tryserver.chromium.linux:linux_chromium_rel_ng Change-Id: Icc5f088ffbc2e2651b845bc71ea42060639e3e48 Reviewed-on: https://chromium-review.googlesource.com/585129 Commit-Queue: Leszek Swirski <leszeks@chromium.org> Reviewed-by: Michael Stanton <mvstanton@chromium.org> Reviewed-by: Ross McIlroy <rmcilroy@chromium.org> Reviewed-by: Michael Lippautz <mlippautz@chromium.org> Cr-Commit-Position: refs/heads/master@{#46935}
2017-07-27 12:45:00 +00:00
/* 35 E> */ B(TestLessThan), R(1), U8(0),
Skip hole initialization of lexical variables when possible This patch expands scope analysis to skip hole initialization when it can be determined statically that no hole checks will be generated at runtime. Two conditions must be met to safely eliminate hole initialization: - There must not exist a VariableProxy referencing this Variable whose HoleCheckMode is kRequired - The Variable must be stack allocated; any other allocation implies that it may be accessed from not-yet-analyzed scopes (other modules, inner functions, or eval code) and that code may require hole checks. The new logic required removing debug code in full-codegen which is now incorrect in some cases. Also fixed Variable's bitfield helpers to take no more space than needed. Bug: chromium:651637 Change-Id: Ie5ac326af4e05b7a5c3c37cd4d0afba6a51a504d Reviewed-on: https://chromium-review.googlesource.com/494006 Commit-Queue: Adam Klein <adamk@chromium.org> Reviewed-by: Georg Neis <neis@chromium.org> Reviewed-by: Leszek Swirski <leszeks@chromium.org> Reviewed-by: Michael Starzinger <mstarzinger@chromium.org> Cr-Commit-Position: refs/heads/master@{#45170}
2017-05-08 18:55:54 +00:00
B(JumpIfFalse), U8(15),
[parser] avoid complex for-loop desugaring when possible let/const declarations in "standard" C-style for-loops have some complex desugaring to accommodate the case where loop loop variables may be captured. This slows down the baseline performance of for-loops with let variables. This change attempts to avoid this desugaring if it's known that the loop variable is not captured at any point. A side effect of this change is that let/const loop variables, when not captured within the loop body, are not necessarily shown in the debugger, similar to other stack-allocated vars. BUG=v8:4762, v8:5460 R=marja@chromium.org, adamk@chromium.org, yangguo@chromium.org Change-Id: I8dbe545a12c086f675972bdba60c94998268311a Reviewed-on: https://chromium-review.googlesource.com/472247 Commit-Queue: Caitlin Potter <caitp@igalia.com> Reviewed-by: Yang Guo <yangguo@chromium.org> Reviewed-by: Adam Klein <adamk@chromium.org> Reviewed-by: Marja Hölttä <marja@chromium.org> Cr-Commit-Position: refs/heads/master@{#44731}
2017-04-18 14:56:06 +00:00
/* 17 E> */ B(StackCheck),
/* 56 S> */ B(Mov), R(1), R(0),
/* 43 S> */ B(Ldar), R(0),
[objects] Make feedback vector a first-class object Instead of having feedback vector as a subtype of FixedArray with reserved slots, make it a first-class variable-sized object with a fixed-size header. This allows us to compress counters to ints in the header, rather than forcing them to be Smis. Cq-Include-Trybots: master.tryserver.chromium.linux:linux_chromium_rel_ng Change-Id: Icc5f088ffbc2e2651b845bc71ea42060639e3e48 Reviewed-on: https://chromium-review.googlesource.com/585129 Commit-Queue: Leszek Swirski <leszeks@chromium.org> Reviewed-by: Michael Stanton <mvstanton@chromium.org> Reviewed-by: Ross McIlroy <rmcilroy@chromium.org> Reviewed-by: Michael Lippautz <mlippautz@chromium.org> Cr-Commit-Position: refs/heads/master@{#46935}
2017-07-27 12:45:00 +00:00
B(Inc), U8(1),
[parser] avoid complex for-loop desugaring when possible let/const declarations in "standard" C-style for-loops have some complex desugaring to accommodate the case where loop loop variables may be captured. This slows down the baseline performance of for-loops with let variables. This change attempts to avoid this desugaring if it's known that the loop variable is not captured at any point. A side effect of this change is that let/const loop variables, when not captured within the loop body, are not necessarily shown in the debugger, similar to other stack-allocated vars. BUG=v8:4762, v8:5460 R=marja@chromium.org, adamk@chromium.org, yangguo@chromium.org Change-Id: I8dbe545a12c086f675972bdba60c94998268311a Reviewed-on: https://chromium-review.googlesource.com/472247 Commit-Queue: Caitlin Potter <caitp@igalia.com> Reviewed-by: Yang Guo <yangguo@chromium.org> Reviewed-by: Adam Klein <adamk@chromium.org> Reviewed-by: Marja Hölttä <marja@chromium.org> Cr-Commit-Position: refs/heads/master@{#44731}
2017-04-18 14:56:06 +00:00
B(Star), R(1),
Skip hole initialization of lexical variables when possible This patch expands scope analysis to skip hole initialization when it can be determined statically that no hole checks will be generated at runtime. Two conditions must be met to safely eliminate hole initialization: - There must not exist a VariableProxy referencing this Variable whose HoleCheckMode is kRequired - The Variable must be stack allocated; any other allocation implies that it may be accessed from not-yet-analyzed scopes (other modules, inner functions, or eval code) and that code may require hole checks. The new logic required removing debug code in full-codegen which is now incorrect in some cases. Also fixed Variable's bitfield helpers to take no more space than needed. Bug: chromium:651637 Change-Id: Ie5ac326af4e05b7a5c3c37cd4d0afba6a51a504d Reviewed-on: https://chromium-review.googlesource.com/494006 Commit-Queue: Adam Klein <adamk@chromium.org> Reviewed-by: Georg Neis <neis@chromium.org> Reviewed-by: Leszek Swirski <leszeks@chromium.org> Reviewed-by: Michael Starzinger <mstarzinger@chromium.org> Cr-Commit-Position: refs/heads/master@{#45170}
2017-05-08 18:55:54 +00:00
B(JumpLoop), U8(17), I8(0),
[parser] avoid complex for-loop desugaring when possible let/const declarations in "standard" C-style for-loops have some complex desugaring to accommodate the case where loop loop variables may be captured. This slows down the baseline performance of for-loops with let variables. This change attempts to avoid this desugaring if it's known that the loop variable is not captured at any point. A side effect of this change is that let/const loop variables, when not captured within the loop body, are not necessarily shown in the debugger, similar to other stack-allocated vars. BUG=v8:4762, v8:5460 R=marja@chromium.org, adamk@chromium.org, yangguo@chromium.org Change-Id: I8dbe545a12c086f675972bdba60c94998268311a Reviewed-on: https://chromium-review.googlesource.com/472247 Commit-Queue: Caitlin Potter <caitp@igalia.com> Reviewed-by: Yang Guo <yangguo@chromium.org> Reviewed-by: Adam Klein <adamk@chromium.org> Reviewed-by: Marja Hölttä <marja@chromium.org> Cr-Commit-Position: refs/heads/master@{#44731}
2017-04-18 14:56:06 +00:00
B(LdaUndefined),
/* 61 S> */ B(Return),
]
constant pool: [
]
handlers: [
]
---
snippet: "
function f() {
for (let x = 0; x < 10; ++x) { eval('1'); }
}
f();
"
[Interpreter] Remove new.target from fixed frame slot. Removes the new.target slot from the interpreter's fixed frame. Instead adds a field to BytecodeArray to get the bytecode's incoming new.target or generator object register. The InterpreterEntryTrampoline then sets this register with the incoming new.target (or generator object) when the function is called. This register can be directly the new.target or generator object variable if they are LOCAL location, otherwise it is a temporary register which is then moved to the variable's location during the function prologue. This fixes a hack in the deoptimizer where we would set the new.target fixed slot to undefined in order to avoid extending it's lifetime through the optimized code - now it's just a standard register and can be optimized away as normal. Bug=v8:6644 Change-Id: Ieb8cc34cccefd9fb6634a90cbc77c6002a54f2ae Reviewed-on: https://chromium-review.googlesource.com/608966 Commit-Queue: Ross McIlroy <rmcilroy@chromium.org> Reviewed-by: Michael Starzinger <mstarzinger@chromium.org> Cr-Commit-Position: refs/heads/master@{#47320}
2017-08-11 15:16:31 +00:00
frame size: 15
[parser] avoid complex for-loop desugaring when possible let/const declarations in "standard" C-style for-loops have some complex desugaring to accommodate the case where loop loop variables may be captured. This slows down the baseline performance of for-loops with let variables. This change attempts to avoid this desugaring if it's known that the loop variable is not captured at any point. A side effect of this change is that let/const loop variables, when not captured within the loop body, are not necessarily shown in the debugger, similar to other stack-allocated vars. BUG=v8:4762, v8:5460 R=marja@chromium.org, adamk@chromium.org, yangguo@chromium.org Change-Id: I8dbe545a12c086f675972bdba60c94998268311a Reviewed-on: https://chromium-review.googlesource.com/472247 Commit-Queue: Caitlin Potter <caitp@igalia.com> Reviewed-by: Yang Guo <yangguo@chromium.org> Reviewed-by: Adam Klein <adamk@chromium.org> Reviewed-by: Marja Hölttä <marja@chromium.org> Cr-Commit-Position: refs/heads/master@{#44731}
2017-04-18 14:56:06 +00:00
parameter count: 1
Replace Context::closure with Context::scope_info, allowing closure to die. There are likely cleanups that can be done after this CL: - context-related functions in the interpreter and compiler take ScopeInfo as well as ScopeType and slot-count as input. The latter 2 should be directly derived from the former. We should be able to drop FunctionContextParameters. - ContextExtension is probably not needed anymore, since we now always have the correct scope_info directly in the SCOPE_INFO_INDEX slot. Bug: v8:7066 Cq-Include-Trybots: luci.chromium.try:linux_chromium_rel_ng;master.tryserver.blink:linux_trusty_blink_rel Change-Id: Ie1f6134c686a9f2183e54730d9cdd598a9e5ab67 Reviewed-on: https://chromium-review.googlesource.com/785151 Commit-Queue: Toon Verwaest <verwaest@chromium.org> Reviewed-by: Hannes Payer <hpayer@chromium.org> Reviewed-by: Adam Klein <adamk@chromium.org> Reviewed-by: Ross McIlroy <rmcilroy@chromium.org> Reviewed-by: Michael Starzinger <mstarzinger@chromium.org> Cr-Commit-Position: refs/heads/master@{#52952}
2018-05-02 13:57:26 +00:00
bytecode array length: 165
[parser] avoid complex for-loop desugaring when possible let/const declarations in "standard" C-style for-loops have some complex desugaring to accommodate the case where loop loop variables may be captured. This slows down the baseline performance of for-loops with let variables. This change attempts to avoid this desugaring if it's known that the loop variable is not captured at any point. A side effect of this change is that let/const loop variables, when not captured within the loop body, are not necessarily shown in the debugger, similar to other stack-allocated vars. BUG=v8:4762, v8:5460 R=marja@chromium.org, adamk@chromium.org, yangguo@chromium.org Change-Id: I8dbe545a12c086f675972bdba60c94998268311a Reviewed-on: https://chromium-review.googlesource.com/472247 Commit-Queue: Caitlin Potter <caitp@igalia.com> Reviewed-by: Yang Guo <yangguo@chromium.org> Reviewed-by: Adam Klein <adamk@chromium.org> Reviewed-by: Marja Hölttä <marja@chromium.org> Cr-Commit-Position: refs/heads/master@{#44731}
2017-04-18 14:56:06 +00:00
bytecodes: [
Replace Context::closure with Context::scope_info, allowing closure to die. There are likely cleanups that can be done after this CL: - context-related functions in the interpreter and compiler take ScopeInfo as well as ScopeType and slot-count as input. The latter 2 should be directly derived from the former. We should be able to drop FunctionContextParameters. - ContextExtension is probably not needed anymore, since we now always have the correct scope_info directly in the SCOPE_INFO_INDEX slot. Bug: v8:7066 Cq-Include-Trybots: luci.chromium.try:linux_chromium_rel_ng;master.tryserver.blink:linux_trusty_blink_rel Change-Id: Ie1f6134c686a9f2183e54730d9cdd598a9e5ab67 Reviewed-on: https://chromium-review.googlesource.com/785151 Commit-Queue: Toon Verwaest <verwaest@chromium.org> Reviewed-by: Hannes Payer <hpayer@chromium.org> Reviewed-by: Adam Klein <adamk@chromium.org> Reviewed-by: Ross McIlroy <rmcilroy@chromium.org> Reviewed-by: Michael Starzinger <mstarzinger@chromium.org> Cr-Commit-Position: refs/heads/master@{#52952}
2018-05-02 13:57:26 +00:00
B(CreateFunctionContext), U8(0), U8(3),
[Interpreter] Remove new.target from fixed frame slot. Removes the new.target slot from the interpreter's fixed frame. Instead adds a field to BytecodeArray to get the bytecode's incoming new.target or generator object register. The InterpreterEntryTrampoline then sets this register with the incoming new.target (or generator object) when the function is called. This register can be directly the new.target or generator object variable if they are LOCAL location, otherwise it is a temporary register which is then moved to the variable's location during the function prologue. This fixes a hack in the deoptimizer where we would set the new.target fixed slot to undefined in order to avoid extending it's lifetime through the optimized code - now it's just a standard register and can be optimized away as normal. Bug=v8:6644 Change-Id: Ieb8cc34cccefd9fb6634a90cbc77c6002a54f2ae Reviewed-on: https://chromium-review.googlesource.com/608966 Commit-Queue: Ross McIlroy <rmcilroy@chromium.org> Reviewed-by: Michael Starzinger <mstarzinger@chromium.org> Cr-Commit-Position: refs/heads/master@{#47320}
2017-08-11 15:16:31 +00:00
B(PushContext), R(4),
[parser] avoid complex for-loop desugaring when possible let/const declarations in "standard" C-style for-loops have some complex desugaring to accommodate the case where loop loop variables may be captured. This slows down the baseline performance of for-loops with let variables. This change attempts to avoid this desugaring if it's known that the loop variable is not captured at any point. A side effect of this change is that let/const loop variables, when not captured within the loop body, are not necessarily shown in the debugger, similar to other stack-allocated vars. BUG=v8:4762, v8:5460 R=marja@chromium.org, adamk@chromium.org, yangguo@chromium.org Change-Id: I8dbe545a12c086f675972bdba60c94998268311a Reviewed-on: https://chromium-review.googlesource.com/472247 Commit-Queue: Caitlin Potter <caitp@igalia.com> Reviewed-by: Yang Guo <yangguo@chromium.org> Reviewed-by: Adam Klein <adamk@chromium.org> Reviewed-by: Marja Hölttä <marja@chromium.org> Cr-Commit-Position: refs/heads/master@{#44731}
2017-04-18 14:56:06 +00:00
B(Ldar), R(this),
B(StaCurrentContextSlot), U8(4),
B(CreateMappedArguments),
B(StaCurrentContextSlot), U8(6),
[Interpreter] Remove new.target from fixed frame slot. Removes the new.target slot from the interpreter's fixed frame. Instead adds a field to BytecodeArray to get the bytecode's incoming new.target or generator object register. The InterpreterEntryTrampoline then sets this register with the incoming new.target (or generator object) when the function is called. This register can be directly the new.target or generator object variable if they are LOCAL location, otherwise it is a temporary register which is then moved to the variable's location during the function prologue. This fixes a hack in the deoptimizer where we would set the new.target fixed slot to undefined in order to avoid extending it's lifetime through the optimized code - now it's just a standard register and can be optimized away as normal. Bug=v8:6644 Change-Id: Ieb8cc34cccefd9fb6634a90cbc77c6002a54f2ae Reviewed-on: https://chromium-review.googlesource.com/608966 Commit-Queue: Ross McIlroy <rmcilroy@chromium.org> Reviewed-by: Michael Starzinger <mstarzinger@chromium.org> Cr-Commit-Position: refs/heads/master@{#47320}
2017-08-11 15:16:31 +00:00
B(Ldar), R(3),
[parser] avoid complex for-loop desugaring when possible let/const declarations in "standard" C-style for-loops have some complex desugaring to accommodate the case where loop loop variables may be captured. This slows down the baseline performance of for-loops with let variables. This change attempts to avoid this desugaring if it's known that the loop variable is not captured at any point. A side effect of this change is that let/const loop variables, when not captured within the loop body, are not necessarily shown in the debugger, similar to other stack-allocated vars. BUG=v8:4762, v8:5460 R=marja@chromium.org, adamk@chromium.org, yangguo@chromium.org Change-Id: I8dbe545a12c086f675972bdba60c94998268311a Reviewed-on: https://chromium-review.googlesource.com/472247 Commit-Queue: Caitlin Potter <caitp@igalia.com> Reviewed-by: Yang Guo <yangguo@chromium.org> Reviewed-by: Adam Klein <adamk@chromium.org> Reviewed-by: Marja Hölttä <marja@chromium.org> Cr-Commit-Position: refs/heads/master@{#44731}
2017-04-18 14:56:06 +00:00
B(StaCurrentContextSlot), U8(5),
/* 10 E> */ B(StackCheck),
Replace Context::closure with Context::scope_info, allowing closure to die. There are likely cleanups that can be done after this CL: - context-related functions in the interpreter and compiler take ScopeInfo as well as ScopeType and slot-count as input. The latter 2 should be directly derived from the former. We should be able to drop FunctionContextParameters. - ContextExtension is probably not needed anymore, since we now always have the correct scope_info directly in the SCOPE_INFO_INDEX slot. Bug: v8:7066 Cq-Include-Trybots: luci.chromium.try:linux_chromium_rel_ng;master.tryserver.blink:linux_trusty_blink_rel Change-Id: Ie1f6134c686a9f2183e54730d9cdd598a9e5ab67 Reviewed-on: https://chromium-review.googlesource.com/785151 Commit-Queue: Toon Verwaest <verwaest@chromium.org> Reviewed-by: Hannes Payer <hpayer@chromium.org> Reviewed-by: Adam Klein <adamk@chromium.org> Reviewed-by: Ross McIlroy <rmcilroy@chromium.org> Reviewed-by: Michael Starzinger <mstarzinger@chromium.org> Cr-Commit-Position: refs/heads/master@{#52952}
2018-05-02 13:57:26 +00:00
B(CreateBlockContext), U8(1),
[Interpreter] Remove new.target from fixed frame slot. Removes the new.target slot from the interpreter's fixed frame. Instead adds a field to BytecodeArray to get the bytecode's incoming new.target or generator object register. The InterpreterEntryTrampoline then sets this register with the incoming new.target (or generator object) when the function is called. This register can be directly the new.target or generator object variable if they are LOCAL location, otherwise it is a temporary register which is then moved to the variable's location during the function prologue. This fixes a hack in the deoptimizer where we would set the new.target fixed slot to undefined in order to avoid extending it's lifetime through the optimized code - now it's just a standard register and can be optimized away as normal. Bug=v8:6644 Change-Id: Ieb8cc34cccefd9fb6634a90cbc77c6002a54f2ae Reviewed-on: https://chromium-review.googlesource.com/608966 Commit-Queue: Ross McIlroy <rmcilroy@chromium.org> Reviewed-by: Michael Starzinger <mstarzinger@chromium.org> Cr-Commit-Position: refs/heads/master@{#47320}
2017-08-11 15:16:31 +00:00
B(PushContext), R(5),
[parser] avoid complex for-loop desugaring when possible let/const declarations in "standard" C-style for-loops have some complex desugaring to accommodate the case where loop loop variables may be captured. This slows down the baseline performance of for-loops with let variables. This change attempts to avoid this desugaring if it's known that the loop variable is not captured at any point. A side effect of this change is that let/const loop variables, when not captured within the loop body, are not necessarily shown in the debugger, similar to other stack-allocated vars. BUG=v8:4762, v8:5460 R=marja@chromium.org, adamk@chromium.org, yangguo@chromium.org Change-Id: I8dbe545a12c086f675972bdba60c94998268311a Reviewed-on: https://chromium-review.googlesource.com/472247 Commit-Queue: Caitlin Potter <caitp@igalia.com> Reviewed-by: Yang Guo <yangguo@chromium.org> Reviewed-by: Adam Klein <adamk@chromium.org> Reviewed-by: Marja Hölttä <marja@chromium.org> Cr-Commit-Position: refs/heads/master@{#44731}
2017-04-18 14:56:06 +00:00
B(LdaTheHole),
B(StaCurrentContextSlot), U8(4),
/* 30 S> */ B(LdaZero),
/* 30 E> */ B(StaCurrentContextSlot), U8(4),
B(LdaCurrentContextSlot), U8(4),
B(Star), R(0),
B(LdaSmi), I8(1),
B(Star), R(1),
/* 59 E> */ B(StackCheck),
Replace Context::closure with Context::scope_info, allowing closure to die. There are likely cleanups that can be done after this CL: - context-related functions in the interpreter and compiler take ScopeInfo as well as ScopeType and slot-count as input. The latter 2 should be directly derived from the former. We should be able to drop FunctionContextParameters. - ContextExtension is probably not needed anymore, since we now always have the correct scope_info directly in the SCOPE_INFO_INDEX slot. Bug: v8:7066 Cq-Include-Trybots: luci.chromium.try:linux_chromium_rel_ng;master.tryserver.blink:linux_trusty_blink_rel Change-Id: Ie1f6134c686a9f2183e54730d9cdd598a9e5ab67 Reviewed-on: https://chromium-review.googlesource.com/785151 Commit-Queue: Toon Verwaest <verwaest@chromium.org> Reviewed-by: Hannes Payer <hpayer@chromium.org> Reviewed-by: Adam Klein <adamk@chromium.org> Reviewed-by: Ross McIlroy <rmcilroy@chromium.org> Reviewed-by: Michael Starzinger <mstarzinger@chromium.org> Cr-Commit-Position: refs/heads/master@{#52952}
2018-05-02 13:57:26 +00:00
B(CreateBlockContext), U8(2),
[Interpreter] Remove new.target from fixed frame slot. Removes the new.target slot from the interpreter's fixed frame. Instead adds a field to BytecodeArray to get the bytecode's incoming new.target or generator object register. The InterpreterEntryTrampoline then sets this register with the incoming new.target (or generator object) when the function is called. This register can be directly the new.target or generator object variable if they are LOCAL location, otherwise it is a temporary register which is then moved to the variable's location during the function prologue. This fixes a hack in the deoptimizer where we would set the new.target fixed slot to undefined in order to avoid extending it's lifetime through the optimized code - now it's just a standard register and can be optimized away as normal. Bug=v8:6644 Change-Id: Ieb8cc34cccefd9fb6634a90cbc77c6002a54f2ae Reviewed-on: https://chromium-review.googlesource.com/608966 Commit-Queue: Ross McIlroy <rmcilroy@chromium.org> Reviewed-by: Michael Starzinger <mstarzinger@chromium.org> Cr-Commit-Position: refs/heads/master@{#47320}
2017-08-11 15:16:31 +00:00
B(PushContext), R(6),
[parser] avoid complex for-loop desugaring when possible let/const declarations in "standard" C-style for-loops have some complex desugaring to accommodate the case where loop loop variables may be captured. This slows down the baseline performance of for-loops with let variables. This change attempts to avoid this desugaring if it's known that the loop variable is not captured at any point. A side effect of this change is that let/const loop variables, when not captured within the loop body, are not necessarily shown in the debugger, similar to other stack-allocated vars. BUG=v8:4762, v8:5460 R=marja@chromium.org, adamk@chromium.org, yangguo@chromium.org Change-Id: I8dbe545a12c086f675972bdba60c94998268311a Reviewed-on: https://chromium-review.googlesource.com/472247 Commit-Queue: Caitlin Potter <caitp@igalia.com> Reviewed-by: Yang Guo <yangguo@chromium.org> Reviewed-by: Adam Klein <adamk@chromium.org> Reviewed-by: Marja Hölttä <marja@chromium.org> Cr-Commit-Position: refs/heads/master@{#44731}
2017-04-18 14:56:06 +00:00
B(LdaTheHole),
B(StaCurrentContextSlot), U8(4),
B(Ldar), R(0),
B(StaCurrentContextSlot), U8(4),
B(LdaSmi), I8(1),
[objects] Make feedback vector a first-class object Instead of having feedback vector as a subtype of FixedArray with reserved slots, make it a first-class variable-sized object with a fixed-size header. This allows us to compress counters to ints in the header, rather than forcing them to be Smis. Cq-Include-Trybots: master.tryserver.chromium.linux:linux_chromium_rel_ng Change-Id: Icc5f088ffbc2e2651b845bc71ea42060639e3e48 Reviewed-on: https://chromium-review.googlesource.com/585129 Commit-Queue: Leszek Swirski <leszeks@chromium.org> Reviewed-by: Michael Stanton <mvstanton@chromium.org> Reviewed-by: Ross McIlroy <rmcilroy@chromium.org> Reviewed-by: Michael Lippautz <mlippautz@chromium.org> Cr-Commit-Position: refs/heads/master@{#46935}
2017-07-27 12:45:00 +00:00
B(TestEqual), R(1), U8(0),
[parser] avoid complex for-loop desugaring when possible let/const declarations in "standard" C-style for-loops have some complex desugaring to accommodate the case where loop loop variables may be captured. This slows down the baseline performance of for-loops with let variables. This change attempts to avoid this desugaring if it's known that the loop variable is not captured at any point. A side effect of this change is that let/const loop variables, when not captured within the loop body, are not necessarily shown in the debugger, similar to other stack-allocated vars. BUG=v8:4762, v8:5460 R=marja@chromium.org, adamk@chromium.org, yangguo@chromium.org Change-Id: I8dbe545a12c086f675972bdba60c94998268311a Reviewed-on: https://chromium-review.googlesource.com/472247 Commit-Queue: Caitlin Potter <caitp@igalia.com> Reviewed-by: Yang Guo <yangguo@chromium.org> Reviewed-by: Adam Klein <adamk@chromium.org> Reviewed-by: Marja Hölttä <marja@chromium.org> Cr-Commit-Position: refs/heads/master@{#44731}
2017-04-18 14:56:06 +00:00
B(JumpIfFalse), U8(7),
B(LdaZero),
B(Star), R(1),
B(Jump), U8(8),
/* 43 S> */ B(LdaCurrentContextSlot), U8(4),
[objects] Make feedback vector a first-class object Instead of having feedback vector as a subtype of FixedArray with reserved slots, make it a first-class variable-sized object with a fixed-size header. This allows us to compress counters to ints in the header, rather than forcing them to be Smis. Cq-Include-Trybots: master.tryserver.chromium.linux:linux_chromium_rel_ng Change-Id: Icc5f088ffbc2e2651b845bc71ea42060639e3e48 Reviewed-on: https://chromium-review.googlesource.com/585129 Commit-Queue: Leszek Swirski <leszeks@chromium.org> Reviewed-by: Michael Stanton <mvstanton@chromium.org> Reviewed-by: Ross McIlroy <rmcilroy@chromium.org> Reviewed-by: Michael Lippautz <mlippautz@chromium.org> Cr-Commit-Position: refs/heads/master@{#46935}
2017-07-27 12:45:00 +00:00
B(Inc), U8(1),
[parser] avoid complex for-loop desugaring when possible let/const declarations in "standard" C-style for-loops have some complex desugaring to accommodate the case where loop loop variables may be captured. This slows down the baseline performance of for-loops with let variables. This change attempts to avoid this desugaring if it's known that the loop variable is not captured at any point. A side effect of this change is that let/const loop variables, when not captured within the loop body, are not necessarily shown in the debugger, similar to other stack-allocated vars. BUG=v8:4762, v8:5460 R=marja@chromium.org, adamk@chromium.org, yangguo@chromium.org Change-Id: I8dbe545a12c086f675972bdba60c94998268311a Reviewed-on: https://chromium-review.googlesource.com/472247 Commit-Queue: Caitlin Potter <caitp@igalia.com> Reviewed-by: Yang Guo <yangguo@chromium.org> Reviewed-by: Adam Klein <adamk@chromium.org> Reviewed-by: Marja Hölttä <marja@chromium.org> Cr-Commit-Position: refs/heads/master@{#44731}
2017-04-18 14:56:06 +00:00
/* 43 E> */ B(StaCurrentContextSlot), U8(4),
B(LdaSmi), I8(1),
B(Star), R(2),
/* 35 S> */ B(LdaCurrentContextSlot), U8(4),
[Interpreter] Remove new.target from fixed frame slot. Removes the new.target slot from the interpreter's fixed frame. Instead adds a field to BytecodeArray to get the bytecode's incoming new.target or generator object register. The InterpreterEntryTrampoline then sets this register with the incoming new.target (or generator object) when the function is called. This register can be directly the new.target or generator object variable if they are LOCAL location, otherwise it is a temporary register which is then moved to the variable's location during the function prologue. This fixes a hack in the deoptimizer where we would set the new.target fixed slot to undefined in order to avoid extending it's lifetime through the optimized code - now it's just a standard register and can be optimized away as normal. Bug=v8:6644 Change-Id: Ieb8cc34cccefd9fb6634a90cbc77c6002a54f2ae Reviewed-on: https://chromium-review.googlesource.com/608966 Commit-Queue: Ross McIlroy <rmcilroy@chromium.org> Reviewed-by: Michael Starzinger <mstarzinger@chromium.org> Cr-Commit-Position: refs/heads/master@{#47320}
2017-08-11 15:16:31 +00:00
B(Star), R(7),
[parser] avoid complex for-loop desugaring when possible let/const declarations in "standard" C-style for-loops have some complex desugaring to accommodate the case where loop loop variables may be captured. This slows down the baseline performance of for-loops with let variables. This change attempts to avoid this desugaring if it's known that the loop variable is not captured at any point. A side effect of this change is that let/const loop variables, when not captured within the loop body, are not necessarily shown in the debugger, similar to other stack-allocated vars. BUG=v8:4762, v8:5460 R=marja@chromium.org, adamk@chromium.org, yangguo@chromium.org Change-Id: I8dbe545a12c086f675972bdba60c94998268311a Reviewed-on: https://chromium-review.googlesource.com/472247 Commit-Queue: Caitlin Potter <caitp@igalia.com> Reviewed-by: Yang Guo <yangguo@chromium.org> Reviewed-by: Adam Klein <adamk@chromium.org> Reviewed-by: Marja Hölttä <marja@chromium.org> Cr-Commit-Position: refs/heads/master@{#44731}
2017-04-18 14:56:06 +00:00
B(LdaSmi), I8(10),
[Interpreter] Remove new.target from fixed frame slot. Removes the new.target slot from the interpreter's fixed frame. Instead adds a field to BytecodeArray to get the bytecode's incoming new.target or generator object register. The InterpreterEntryTrampoline then sets this register with the incoming new.target (or generator object) when the function is called. This register can be directly the new.target or generator object variable if they are LOCAL location, otherwise it is a temporary register which is then moved to the variable's location during the function prologue. This fixes a hack in the deoptimizer where we would set the new.target fixed slot to undefined in order to avoid extending it's lifetime through the optimized code - now it's just a standard register and can be optimized away as normal. Bug=v8:6644 Change-Id: Ieb8cc34cccefd9fb6634a90cbc77c6002a54f2ae Reviewed-on: https://chromium-review.googlesource.com/608966 Commit-Queue: Ross McIlroy <rmcilroy@chromium.org> Reviewed-by: Michael Starzinger <mstarzinger@chromium.org> Cr-Commit-Position: refs/heads/master@{#47320}
2017-08-11 15:16:31 +00:00
/* 35 E> */ B(TestLessThan), R(7), U8(2),
[parser] avoid complex for-loop desugaring when possible let/const declarations in "standard" C-style for-loops have some complex desugaring to accommodate the case where loop loop variables may be captured. This slows down the baseline performance of for-loops with let variables. This change attempts to avoid this desugaring if it's known that the loop variable is not captured at any point. A side effect of this change is that let/const loop variables, when not captured within the loop body, are not necessarily shown in the debugger, similar to other stack-allocated vars. BUG=v8:4762, v8:5460 R=marja@chromium.org, adamk@chromium.org, yangguo@chromium.org Change-Id: I8dbe545a12c086f675972bdba60c94998268311a Reviewed-on: https://chromium-review.googlesource.com/472247 Commit-Queue: Caitlin Potter <caitp@igalia.com> Reviewed-by: Yang Guo <yangguo@chromium.org> Reviewed-by: Adam Klein <adamk@chromium.org> Reviewed-by: Marja Hölttä <marja@chromium.org> Cr-Commit-Position: refs/heads/master@{#44731}
2017-04-18 14:56:06 +00:00
B(JumpIfFalse), U8(4),
B(Jump), U8(6),
[Interpreter] Remove new.target from fixed frame slot. Removes the new.target slot from the interpreter's fixed frame. Instead adds a field to BytecodeArray to get the bytecode's incoming new.target or generator object register. The InterpreterEntryTrampoline then sets this register with the incoming new.target (or generator object) when the function is called. This register can be directly the new.target or generator object variable if they are LOCAL location, otherwise it is a temporary register which is then moved to the variable's location during the function prologue. This fixes a hack in the deoptimizer where we would set the new.target fixed slot to undefined in order to avoid extending it's lifetime through the optimized code - now it's just a standard register and can be optimized away as normal. Bug=v8:6644 Change-Id: Ieb8cc34cccefd9fb6634a90cbc77c6002a54f2ae Reviewed-on: https://chromium-review.googlesource.com/608966 Commit-Queue: Ross McIlroy <rmcilroy@chromium.org> Reviewed-by: Michael Starzinger <mstarzinger@chromium.org> Cr-Commit-Position: refs/heads/master@{#47320}
2017-08-11 15:16:31 +00:00
B(PopContext), R(6),
[parser] avoid complex for-loop desugaring when possible let/const declarations in "standard" C-style for-loops have some complex desugaring to accommodate the case where loop loop variables may be captured. This slows down the baseline performance of for-loops with let variables. This change attempts to avoid this desugaring if it's known that the loop variable is not captured at any point. A side effect of this change is that let/const loop variables, when not captured within the loop body, are not necessarily shown in the debugger, similar to other stack-allocated vars. BUG=v8:4762, v8:5460 R=marja@chromium.org, adamk@chromium.org, yangguo@chromium.org Change-Id: I8dbe545a12c086f675972bdba60c94998268311a Reviewed-on: https://chromium-review.googlesource.com/472247 Commit-Queue: Caitlin Potter <caitp@igalia.com> Reviewed-by: Yang Guo <yangguo@chromium.org> Reviewed-by: Adam Klein <adamk@chromium.org> Reviewed-by: Marja Hölttä <marja@chromium.org> Cr-Commit-Position: refs/heads/master@{#44731}
2017-04-18 14:56:06 +00:00
B(Jump), U8(77),
B(LdaSmi), I8(1),
[objects] Make feedback vector a first-class object Instead of having feedback vector as a subtype of FixedArray with reserved slots, make it a first-class variable-sized object with a fixed-size header. This allows us to compress counters to ints in the header, rather than forcing them to be Smis. Cq-Include-Trybots: master.tryserver.chromium.linux:linux_chromium_rel_ng Change-Id: Icc5f088ffbc2e2651b845bc71ea42060639e3e48 Reviewed-on: https://chromium-review.googlesource.com/585129 Commit-Queue: Leszek Swirski <leszeks@chromium.org> Reviewed-by: Michael Stanton <mvstanton@chromium.org> Reviewed-by: Ross McIlroy <rmcilroy@chromium.org> Reviewed-by: Michael Lippautz <mlippautz@chromium.org> Cr-Commit-Position: refs/heads/master@{#46935}
2017-07-27 12:45:00 +00:00
B(TestEqual), R(2), U8(3),
[parser] avoid complex for-loop desugaring when possible let/const declarations in "standard" C-style for-loops have some complex desugaring to accommodate the case where loop loop variables may be captured. This slows down the baseline performance of for-loops with let variables. This change attempts to avoid this desugaring if it's known that the loop variable is not captured at any point. A side effect of this change is that let/const loop variables, when not captured within the loop body, are not necessarily shown in the debugger, similar to other stack-allocated vars. BUG=v8:4762, v8:5460 R=marja@chromium.org, adamk@chromium.org, yangguo@chromium.org Change-Id: I8dbe545a12c086f675972bdba60c94998268311a Reviewed-on: https://chromium-review.googlesource.com/472247 Commit-Queue: Caitlin Potter <caitp@igalia.com> Reviewed-by: Yang Guo <yangguo@chromium.org> Reviewed-by: Adam Klein <adamk@chromium.org> Reviewed-by: Marja Hölttä <marja@chromium.org> Cr-Commit-Position: refs/heads/master@{#44731}
2017-04-18 14:56:06 +00:00
B(JumpIfFalse), U8(54),
/* 17 E> */ B(StackCheck),
Replace Context::closure with Context::scope_info, allowing closure to die. There are likely cleanups that can be done after this CL: - context-related functions in the interpreter and compiler take ScopeInfo as well as ScopeType and slot-count as input. The latter 2 should be directly derived from the former. We should be able to drop FunctionContextParameters. - ContextExtension is probably not needed anymore, since we now always have the correct scope_info directly in the SCOPE_INFO_INDEX slot. Bug: v8:7066 Cq-Include-Trybots: luci.chromium.try:linux_chromium_rel_ng;master.tryserver.blink:linux_trusty_blink_rel Change-Id: Ie1f6134c686a9f2183e54730d9cdd598a9e5ab67 Reviewed-on: https://chromium-review.googlesource.com/785151 Commit-Queue: Toon Verwaest <verwaest@chromium.org> Reviewed-by: Hannes Payer <hpayer@chromium.org> Reviewed-by: Adam Klein <adamk@chromium.org> Reviewed-by: Ross McIlroy <rmcilroy@chromium.org> Reviewed-by: Michael Starzinger <mstarzinger@chromium.org> Cr-Commit-Position: refs/heads/master@{#52952}
2018-05-02 13:57:26 +00:00
/* 48 S> */ B(LdaLookupGlobalSlot), U8(3), U8(4), U8(3),
[parser] avoid complex for-loop desugaring when possible let/const declarations in "standard" C-style for-loops have some complex desugaring to accommodate the case where loop loop variables may be captured. This slows down the baseline performance of for-loops with let variables. This change attempts to avoid this desugaring if it's known that the loop variable is not captured at any point. A side effect of this change is that let/const loop variables, when not captured within the loop body, are not necessarily shown in the debugger, similar to other stack-allocated vars. BUG=v8:4762, v8:5460 R=marja@chromium.org, adamk@chromium.org, yangguo@chromium.org Change-Id: I8dbe545a12c086f675972bdba60c94998268311a Reviewed-on: https://chromium-review.googlesource.com/472247 Commit-Queue: Caitlin Potter <caitp@igalia.com> Reviewed-by: Yang Guo <yangguo@chromium.org> Reviewed-by: Adam Klein <adamk@chromium.org> Reviewed-by: Marja Hölttä <marja@chromium.org> Cr-Commit-Position: refs/heads/master@{#44731}
2017-04-18 14:56:06 +00:00
B(Star), R(7),
Replace Context::closure with Context::scope_info, allowing closure to die. There are likely cleanups that can be done after this CL: - context-related functions in the interpreter and compiler take ScopeInfo as well as ScopeType and slot-count as input. The latter 2 should be directly derived from the former. We should be able to drop FunctionContextParameters. - ContextExtension is probably not needed anymore, since we now always have the correct scope_info directly in the SCOPE_INFO_INDEX slot. Bug: v8:7066 Cq-Include-Trybots: luci.chromium.try:linux_chromium_rel_ng;master.tryserver.blink:linux_trusty_blink_rel Change-Id: Ie1f6134c686a9f2183e54730d9cdd598a9e5ab67 Reviewed-on: https://chromium-review.googlesource.com/785151 Commit-Queue: Toon Verwaest <verwaest@chromium.org> Reviewed-by: Hannes Payer <hpayer@chromium.org> Reviewed-by: Adam Klein <adamk@chromium.org> Reviewed-by: Ross McIlroy <rmcilroy@chromium.org> Reviewed-by: Michael Starzinger <mstarzinger@chromium.org> Cr-Commit-Position: refs/heads/master@{#52952}
2018-05-02 13:57:26 +00:00
B(LdaConstant), U8(4),
[Interpreter] Remove new.target from fixed frame slot. Removes the new.target slot from the interpreter's fixed frame. Instead adds a field to BytecodeArray to get the bytecode's incoming new.target or generator object register. The InterpreterEntryTrampoline then sets this register with the incoming new.target (or generator object) when the function is called. This register can be directly the new.target or generator object variable if they are LOCAL location, otherwise it is a temporary register which is then moved to the variable's location during the function prologue. This fixes a hack in the deoptimizer where we would set the new.target fixed slot to undefined in order to avoid extending it's lifetime through the optimized code - now it's just a standard register and can be optimized away as normal. Bug=v8:6644 Change-Id: Ieb8cc34cccefd9fb6634a90cbc77c6002a54f2ae Reviewed-on: https://chromium-review.googlesource.com/608966 Commit-Queue: Ross McIlroy <rmcilroy@chromium.org> Reviewed-by: Michael Starzinger <mstarzinger@chromium.org> Cr-Commit-Position: refs/heads/master@{#47320}
2017-08-11 15:16:31 +00:00
B(Star), R(8),
[parser] avoid complex for-loop desugaring when possible let/const declarations in "standard" C-style for-loops have some complex desugaring to accommodate the case where loop loop variables may be captured. This slows down the baseline performance of for-loops with let variables. This change attempts to avoid this desugaring if it's known that the loop variable is not captured at any point. A side effect of this change is that let/const loop variables, when not captured within the loop body, are not necessarily shown in the debugger, similar to other stack-allocated vars. BUG=v8:4762, v8:5460 R=marja@chromium.org, adamk@chromium.org, yangguo@chromium.org Change-Id: I8dbe545a12c086f675972bdba60c94998268311a Reviewed-on: https://chromium-review.googlesource.com/472247 Commit-Queue: Caitlin Potter <caitp@igalia.com> Reviewed-by: Yang Guo <yangguo@chromium.org> Reviewed-by: Adam Klein <adamk@chromium.org> Reviewed-by: Marja Hölttä <marja@chromium.org> Cr-Commit-Position: refs/heads/master@{#44731}
2017-04-18 14:56:06 +00:00
B(LdaZero),
B(Star), R(12),
[Interpreter] Remove new.target from fixed frame slot. Removes the new.target slot from the interpreter's fixed frame. Instead adds a field to BytecodeArray to get the bytecode's incoming new.target or generator object register. The InterpreterEntryTrampoline then sets this register with the incoming new.target (or generator object) when the function is called. This register can be directly the new.target or generator object variable if they are LOCAL location, otherwise it is a temporary register which is then moved to the variable's location during the function prologue. This fixes a hack in the deoptimizer where we would set the new.target fixed slot to undefined in order to avoid extending it's lifetime through the optimized code - now it's just a standard register and can be optimized away as normal. Bug=v8:6644 Change-Id: Ieb8cc34cccefd9fb6634a90cbc77c6002a54f2ae Reviewed-on: https://chromium-review.googlesource.com/608966 Commit-Queue: Ross McIlroy <rmcilroy@chromium.org> Reviewed-by: Michael Starzinger <mstarzinger@chromium.org> Cr-Commit-Position: refs/heads/master@{#47320}
2017-08-11 15:16:31 +00:00
B(LdaSmi), I8(31),
[parser] avoid complex for-loop desugaring when possible let/const declarations in "standard" C-style for-loops have some complex desugaring to accommodate the case where loop loop variables may be captured. This slows down the baseline performance of for-loops with let variables. This change attempts to avoid this desugaring if it's known that the loop variable is not captured at any point. A side effect of this change is that let/const loop variables, when not captured within the loop body, are not necessarily shown in the debugger, similar to other stack-allocated vars. BUG=v8:4762, v8:5460 R=marja@chromium.org, adamk@chromium.org, yangguo@chromium.org Change-Id: I8dbe545a12c086f675972bdba60c94998268311a Reviewed-on: https://chromium-review.googlesource.com/472247 Commit-Queue: Caitlin Potter <caitp@igalia.com> Reviewed-by: Yang Guo <yangguo@chromium.org> Reviewed-by: Adam Klein <adamk@chromium.org> Reviewed-by: Marja Hölttä <marja@chromium.org> Cr-Commit-Position: refs/heads/master@{#44731}
2017-04-18 14:56:06 +00:00
B(Star), R(13),
[Interpreter] Remove new.target from fixed frame slot. Removes the new.target slot from the interpreter's fixed frame. Instead adds a field to BytecodeArray to get the bytecode's incoming new.target or generator object register. The InterpreterEntryTrampoline then sets this register with the incoming new.target (or generator object) when the function is called. This register can be directly the new.target or generator object variable if they are LOCAL location, otherwise it is a temporary register which is then moved to the variable's location during the function prologue. This fixes a hack in the deoptimizer where we would set the new.target fixed slot to undefined in order to avoid extending it's lifetime through the optimized code - now it's just a standard register and can be optimized away as normal. Bug=v8:6644 Change-Id: Ieb8cc34cccefd9fb6634a90cbc77c6002a54f2ae Reviewed-on: https://chromium-review.googlesource.com/608966 Commit-Queue: Ross McIlroy <rmcilroy@chromium.org> Reviewed-by: Michael Starzinger <mstarzinger@chromium.org> Cr-Commit-Position: refs/heads/master@{#47320}
2017-08-11 15:16:31 +00:00
B(LdaSmi), I8(48),
B(Star), R(14),
[parser] avoid complex for-loop desugaring when possible let/const declarations in "standard" C-style for-loops have some complex desugaring to accommodate the case where loop loop variables may be captured. This slows down the baseline performance of for-loops with let variables. This change attempts to avoid this desugaring if it's known that the loop variable is not captured at any point. A side effect of this change is that let/const loop variables, when not captured within the loop body, are not necessarily shown in the debugger, similar to other stack-allocated vars. BUG=v8:4762, v8:5460 R=marja@chromium.org, adamk@chromium.org, yangguo@chromium.org Change-Id: I8dbe545a12c086f675972bdba60c94998268311a Reviewed-on: https://chromium-review.googlesource.com/472247 Commit-Queue: Caitlin Potter <caitp@igalia.com> Reviewed-by: Yang Guo <yangguo@chromium.org> Reviewed-by: Adam Klein <adamk@chromium.org> Reviewed-by: Marja Hölttä <marja@chromium.org> Cr-Commit-Position: refs/heads/master@{#44731}
2017-04-18 14:56:06 +00:00
B(Mov), R(7), R(9),
[Interpreter] Remove new.target from fixed frame slot. Removes the new.target slot from the interpreter's fixed frame. Instead adds a field to BytecodeArray to get the bytecode's incoming new.target or generator object register. The InterpreterEntryTrampoline then sets this register with the incoming new.target (or generator object) when the function is called. This register can be directly the new.target or generator object variable if they are LOCAL location, otherwise it is a temporary register which is then moved to the variable's location during the function prologue. This fixes a hack in the deoptimizer where we would set the new.target fixed slot to undefined in order to avoid extending it's lifetime through the optimized code - now it's just a standard register and can be optimized away as normal. Bug=v8:6644 Change-Id: Ieb8cc34cccefd9fb6634a90cbc77c6002a54f2ae Reviewed-on: https://chromium-review.googlesource.com/608966 Commit-Queue: Ross McIlroy <rmcilroy@chromium.org> Reviewed-by: Michael Starzinger <mstarzinger@chromium.org> Cr-Commit-Position: refs/heads/master@{#47320}
2017-08-11 15:16:31 +00:00
B(Mov), R(8), R(10),
B(Mov), R(closure), R(11),
B(CallRuntime), U16(Runtime::kResolvePossiblyDirectEval), R(9), U8(6),
B(Star), R(7),
[Cleanup][Interpreter] Move feedback slot allocation to bytecode generator Moves the feedback vector slot allocation out of ast-numbering and into bytecode generation directly. This has a couple of benifits, including reduced AST size, avoid code duplication and reduced feedback vector sizes in many cases due to only allocating slots when needed. Also removes AstProperties since this is no longer needed. AstNumbering is now only used to allocate suspend ids for generators. BUG=v8:6921 Change-Id: I103e8593c94ef5b2e56c34ef4f77bd6e7d64796f Reviewed-on: https://chromium-review.googlesource.com/722959 Commit-Queue: Ross McIlroy <rmcilroy@chromium.org> Reviewed-by: Michael Stanton <mvstanton@chromium.org> Reviewed-by: Michael Starzinger <mstarzinger@chromium.org> Cr-Commit-Position: refs/heads/master@{#48757}
2017-10-19 15:12:42 +00:00
/* 48 E> */ B(CallUndefinedReceiver1), R(7), R(8), U8(6),
[parser] avoid complex for-loop desugaring when possible let/const declarations in "standard" C-style for-loops have some complex desugaring to accommodate the case where loop loop variables may be captured. This slows down the baseline performance of for-loops with let variables. This change attempts to avoid this desugaring if it's known that the loop variable is not captured at any point. A side effect of this change is that let/const loop variables, when not captured within the loop body, are not necessarily shown in the debugger, similar to other stack-allocated vars. BUG=v8:4762, v8:5460 R=marja@chromium.org, adamk@chromium.org, yangguo@chromium.org Change-Id: I8dbe545a12c086f675972bdba60c94998268311a Reviewed-on: https://chromium-review.googlesource.com/472247 Commit-Queue: Caitlin Potter <caitp@igalia.com> Reviewed-by: Yang Guo <yangguo@chromium.org> Reviewed-by: Adam Klein <adamk@chromium.org> Reviewed-by: Marja Hölttä <marja@chromium.org> Cr-Commit-Position: refs/heads/master@{#44731}
2017-04-18 14:56:06 +00:00
B(LdaZero),
B(Star), R(2),
B(LdaCurrentContextSlot), U8(4),
B(Star), R(0),
B(JumpLoop), U8(56), I8(1),
B(LdaSmi), I8(1),
[objects] Make feedback vector a first-class object Instead of having feedback vector as a subtype of FixedArray with reserved slots, make it a first-class variable-sized object with a fixed-size header. This allows us to compress counters to ints in the header, rather than forcing them to be Smis. Cq-Include-Trybots: master.tryserver.chromium.linux:linux_chromium_rel_ng Change-Id: Icc5f088ffbc2e2651b845bc71ea42060639e3e48 Reviewed-on: https://chromium-review.googlesource.com/585129 Commit-Queue: Leszek Swirski <leszeks@chromium.org> Reviewed-by: Michael Stanton <mvstanton@chromium.org> Reviewed-by: Ross McIlroy <rmcilroy@chromium.org> Reviewed-by: Michael Lippautz <mlippautz@chromium.org> Cr-Commit-Position: refs/heads/master@{#46935}
2017-07-27 12:45:00 +00:00
/* 59 E> */ B(TestEqual), R(2), U8(8),
[parser] avoid complex for-loop desugaring when possible let/const declarations in "standard" C-style for-loops have some complex desugaring to accommodate the case where loop loop variables may be captured. This slows down the baseline performance of for-loops with let variables. This change attempts to avoid this desugaring if it's known that the loop variable is not captured at any point. A side effect of this change is that let/const loop variables, when not captured within the loop body, are not necessarily shown in the debugger, similar to other stack-allocated vars. BUG=v8:4762, v8:5460 R=marja@chromium.org, adamk@chromium.org, yangguo@chromium.org Change-Id: I8dbe545a12c086f675972bdba60c94998268311a Reviewed-on: https://chromium-review.googlesource.com/472247 Commit-Queue: Caitlin Potter <caitp@igalia.com> Reviewed-by: Yang Guo <yangguo@chromium.org> Reviewed-by: Adam Klein <adamk@chromium.org> Reviewed-by: Marja Hölttä <marja@chromium.org> Cr-Commit-Position: refs/heads/master@{#44731}
2017-04-18 14:56:06 +00:00
B(JumpIfFalse), U8(6),
[Interpreter] Remove new.target from fixed frame slot. Removes the new.target slot from the interpreter's fixed frame. Instead adds a field to BytecodeArray to get the bytecode's incoming new.target or generator object register. The InterpreterEntryTrampoline then sets this register with the incoming new.target (or generator object) when the function is called. This register can be directly the new.target or generator object variable if they are LOCAL location, otherwise it is a temporary register which is then moved to the variable's location during the function prologue. This fixes a hack in the deoptimizer where we would set the new.target fixed slot to undefined in order to avoid extending it's lifetime through the optimized code - now it's just a standard register and can be optimized away as normal. Bug=v8:6644 Change-Id: Ieb8cc34cccefd9fb6634a90cbc77c6002a54f2ae Reviewed-on: https://chromium-review.googlesource.com/608966 Commit-Queue: Ross McIlroy <rmcilroy@chromium.org> Reviewed-by: Michael Starzinger <mstarzinger@chromium.org> Cr-Commit-Position: refs/heads/master@{#47320}
2017-08-11 15:16:31 +00:00
B(PopContext), R(6),
[parser] avoid complex for-loop desugaring when possible let/const declarations in "standard" C-style for-loops have some complex desugaring to accommodate the case where loop loop variables may be captured. This slows down the baseline performance of for-loops with let variables. This change attempts to avoid this desugaring if it's known that the loop variable is not captured at any point. A side effect of this change is that let/const loop variables, when not captured within the loop body, are not necessarily shown in the debugger, similar to other stack-allocated vars. BUG=v8:4762, v8:5460 R=marja@chromium.org, adamk@chromium.org, yangguo@chromium.org Change-Id: I8dbe545a12c086f675972bdba60c94998268311a Reviewed-on: https://chromium-review.googlesource.com/472247 Commit-Queue: Caitlin Potter <caitp@igalia.com> Reviewed-by: Yang Guo <yangguo@chromium.org> Reviewed-by: Adam Klein <adamk@chromium.org> Reviewed-by: Marja Hölttä <marja@chromium.org> Cr-Commit-Position: refs/heads/master@{#44731}
2017-04-18 14:56:06 +00:00
B(Jump), U8(7),
[Interpreter] Remove new.target from fixed frame slot. Removes the new.target slot from the interpreter's fixed frame. Instead adds a field to BytecodeArray to get the bytecode's incoming new.target or generator object register. The InterpreterEntryTrampoline then sets this register with the incoming new.target (or generator object) when the function is called. This register can be directly the new.target or generator object variable if they are LOCAL location, otherwise it is a temporary register which is then moved to the variable's location during the function prologue. This fixes a hack in the deoptimizer where we would set the new.target fixed slot to undefined in order to avoid extending it's lifetime through the optimized code - now it's just a standard register and can be optimized away as normal. Bug=v8:6644 Change-Id: Ieb8cc34cccefd9fb6634a90cbc77c6002a54f2ae Reviewed-on: https://chromium-review.googlesource.com/608966 Commit-Queue: Ross McIlroy <rmcilroy@chromium.org> Reviewed-by: Michael Starzinger <mstarzinger@chromium.org> Cr-Commit-Position: refs/heads/master@{#47320}
2017-08-11 15:16:31 +00:00
B(PopContext), R(6),
Replace Context::closure with Context::scope_info, allowing closure to die. There are likely cleanups that can be done after this CL: - context-related functions in the interpreter and compiler take ScopeInfo as well as ScopeType and slot-count as input. The latter 2 should be directly derived from the former. We should be able to drop FunctionContextParameters. - ContextExtension is probably not needed anymore, since we now always have the correct scope_info directly in the SCOPE_INFO_INDEX slot. Bug: v8:7066 Cq-Include-Trybots: luci.chromium.try:linux_chromium_rel_ng;master.tryserver.blink:linux_trusty_blink_rel Change-Id: Ie1f6134c686a9f2183e54730d9cdd598a9e5ab67 Reviewed-on: https://chromium-review.googlesource.com/785151 Commit-Queue: Toon Verwaest <verwaest@chromium.org> Reviewed-by: Hannes Payer <hpayer@chromium.org> Reviewed-by: Adam Klein <adamk@chromium.org> Reviewed-by: Ross McIlroy <rmcilroy@chromium.org> Reviewed-by: Michael Starzinger <mstarzinger@chromium.org> Cr-Commit-Position: refs/heads/master@{#52952}
2018-05-02 13:57:26 +00:00
B(JumpLoop), U8(123), I8(0),
[Interpreter] Remove new.target from fixed frame slot. Removes the new.target slot from the interpreter's fixed frame. Instead adds a field to BytecodeArray to get the bytecode's incoming new.target or generator object register. The InterpreterEntryTrampoline then sets this register with the incoming new.target (or generator object) when the function is called. This register can be directly the new.target or generator object variable if they are LOCAL location, otherwise it is a temporary register which is then moved to the variable's location during the function prologue. This fixes a hack in the deoptimizer where we would set the new.target fixed slot to undefined in order to avoid extending it's lifetime through the optimized code - now it's just a standard register and can be optimized away as normal. Bug=v8:6644 Change-Id: Ieb8cc34cccefd9fb6634a90cbc77c6002a54f2ae Reviewed-on: https://chromium-review.googlesource.com/608966 Commit-Queue: Ross McIlroy <rmcilroy@chromium.org> Reviewed-by: Michael Starzinger <mstarzinger@chromium.org> Cr-Commit-Position: refs/heads/master@{#47320}
2017-08-11 15:16:31 +00:00
B(PopContext), R(5),
[parser] avoid complex for-loop desugaring when possible let/const declarations in "standard" C-style for-loops have some complex desugaring to accommodate the case where loop loop variables may be captured. This slows down the baseline performance of for-loops with let variables. This change attempts to avoid this desugaring if it's known that the loop variable is not captured at any point. A side effect of this change is that let/const loop variables, when not captured within the loop body, are not necessarily shown in the debugger, similar to other stack-allocated vars. BUG=v8:4762, v8:5460 R=marja@chromium.org, adamk@chromium.org, yangguo@chromium.org Change-Id: I8dbe545a12c086f675972bdba60c94998268311a Reviewed-on: https://chromium-review.googlesource.com/472247 Commit-Queue: Caitlin Potter <caitp@igalia.com> Reviewed-by: Yang Guo <yangguo@chromium.org> Reviewed-by: Adam Klein <adamk@chromium.org> Reviewed-by: Marja Hölttä <marja@chromium.org> Cr-Commit-Position: refs/heads/master@{#44731}
2017-04-18 14:56:06 +00:00
B(LdaUndefined),
/* 61 S> */ B(Return),
]
constant pool: [
Replace Context::closure with Context::scope_info, allowing closure to die. There are likely cleanups that can be done after this CL: - context-related functions in the interpreter and compiler take ScopeInfo as well as ScopeType and slot-count as input. The latter 2 should be directly derived from the former. We should be able to drop FunctionContextParameters. - ContextExtension is probably not needed anymore, since we now always have the correct scope_info directly in the SCOPE_INFO_INDEX slot. Bug: v8:7066 Cq-Include-Trybots: luci.chromium.try:linux_chromium_rel_ng;master.tryserver.blink:linux_trusty_blink_rel Change-Id: Ie1f6134c686a9f2183e54730d9cdd598a9e5ab67 Reviewed-on: https://chromium-review.googlesource.com/785151 Commit-Queue: Toon Verwaest <verwaest@chromium.org> Reviewed-by: Hannes Payer <hpayer@chromium.org> Reviewed-by: Adam Klein <adamk@chromium.org> Reviewed-by: Ross McIlroy <rmcilroy@chromium.org> Reviewed-by: Michael Starzinger <mstarzinger@chromium.org> Cr-Commit-Position: refs/heads/master@{#52952}
2018-05-02 13:57:26 +00:00
SCOPE_INFO_TYPE,
[runtime] Add SCOPE_INFO_TYPE InstanceType Bug: v8:7310 Change-Id: I82e7ada4c0f7e415887a859719eb01bb45fd3012 Reviewed-on: https://chromium-review.googlesource.com/921742 Reviewed-by: Michael Lippautz <mlippautz@chromium.org> Reviewed-by: Peter Marshall <petermarshall@chromium.org> Commit-Queue: Camillo Bruni <cbruni@chromium.org> Cr-Commit-Position: refs/heads/master@{#51324}
2018-02-16 12:36:47 +00:00
SCOPE_INFO_TYPE,
SCOPE_INFO_TYPE,
[parser] avoid complex for-loop desugaring when possible let/const declarations in "standard" C-style for-loops have some complex desugaring to accommodate the case where loop loop variables may be captured. This slows down the baseline performance of for-loops with let variables. This change attempts to avoid this desugaring if it's known that the loop variable is not captured at any point. A side effect of this change is that let/const loop variables, when not captured within the loop body, are not necessarily shown in the debugger, similar to other stack-allocated vars. BUG=v8:4762, v8:5460 R=marja@chromium.org, adamk@chromium.org, yangguo@chromium.org Change-Id: I8dbe545a12c086f675972bdba60c94998268311a Reviewed-on: https://chromium-review.googlesource.com/472247 Commit-Queue: Caitlin Potter <caitp@igalia.com> Reviewed-by: Yang Guo <yangguo@chromium.org> Reviewed-by: Adam Klein <adamk@chromium.org> Reviewed-by: Marja Hölttä <marja@chromium.org> Cr-Commit-Position: refs/heads/master@{#44731}
2017-04-18 14:56:06 +00:00
ONE_BYTE_INTERNALIZED_STRING_TYPE ["eval"],
ONE_BYTE_INTERNALIZED_STRING_TYPE ["1"],
]
handlers: [
]
---
snippet: "
function f() {
for (let x = 0; x < 10; ++x) { (function() { return x; })(); }
}
f();
"
frame size: 6
parameter count: 1
Replace Context::closure with Context::scope_info, allowing closure to die. There are likely cleanups that can be done after this CL: - context-related functions in the interpreter and compiler take ScopeInfo as well as ScopeType and slot-count as input. The latter 2 should be directly derived from the former. We should be able to drop FunctionContextParameters. - ContextExtension is probably not needed anymore, since we now always have the correct scope_info directly in the SCOPE_INFO_INDEX slot. Bug: v8:7066 Cq-Include-Trybots: luci.chromium.try:linux_chromium_rel_ng;master.tryserver.blink:linux_trusty_blink_rel Change-Id: Ie1f6134c686a9f2183e54730d9cdd598a9e5ab67 Reviewed-on: https://chromium-review.googlesource.com/785151 Commit-Queue: Toon Verwaest <verwaest@chromium.org> Reviewed-by: Hannes Payer <hpayer@chromium.org> Reviewed-by: Adam Klein <adamk@chromium.org> Reviewed-by: Ross McIlroy <rmcilroy@chromium.org> Reviewed-by: Michael Starzinger <mstarzinger@chromium.org> Cr-Commit-Position: refs/heads/master@{#52952}
2018-05-02 13:57:26 +00:00
bytecode array length: 106
[parser] avoid complex for-loop desugaring when possible let/const declarations in "standard" C-style for-loops have some complex desugaring to accommodate the case where loop loop variables may be captured. This slows down the baseline performance of for-loops with let variables. This change attempts to avoid this desugaring if it's known that the loop variable is not captured at any point. A side effect of this change is that let/const loop variables, when not captured within the loop body, are not necessarily shown in the debugger, similar to other stack-allocated vars. BUG=v8:4762, v8:5460 R=marja@chromium.org, adamk@chromium.org, yangguo@chromium.org Change-Id: I8dbe545a12c086f675972bdba60c94998268311a Reviewed-on: https://chromium-review.googlesource.com/472247 Commit-Queue: Caitlin Potter <caitp@igalia.com> Reviewed-by: Yang Guo <yangguo@chromium.org> Reviewed-by: Adam Klein <adamk@chromium.org> Reviewed-by: Marja Hölttä <marja@chromium.org> Cr-Commit-Position: refs/heads/master@{#44731}
2017-04-18 14:56:06 +00:00
bytecodes: [
/* 10 E> */ B(StackCheck),
/* 30 S> */ B(LdaZero),
B(Star), R(0),
B(Star), R(1),
B(LdaSmi), I8(1),
B(Star), R(2),
/* 78 E> */ B(StackCheck),
B(CreateBlockContext), U8(0),
B(PushContext), R(4),
B(LdaTheHole),
B(StaCurrentContextSlot), U8(4),
B(Ldar), R(1),
B(StaCurrentContextSlot), U8(4),
B(LdaSmi), I8(1),
[objects] Make feedback vector a first-class object Instead of having feedback vector as a subtype of FixedArray with reserved slots, make it a first-class variable-sized object with a fixed-size header. This allows us to compress counters to ints in the header, rather than forcing them to be Smis. Cq-Include-Trybots: master.tryserver.chromium.linux:linux_chromium_rel_ng Change-Id: Icc5f088ffbc2e2651b845bc71ea42060639e3e48 Reviewed-on: https://chromium-review.googlesource.com/585129 Commit-Queue: Leszek Swirski <leszeks@chromium.org> Reviewed-by: Michael Stanton <mvstanton@chromium.org> Reviewed-by: Ross McIlroy <rmcilroy@chromium.org> Reviewed-by: Michael Lippautz <mlippautz@chromium.org> Cr-Commit-Position: refs/heads/master@{#46935}
2017-07-27 12:45:00 +00:00
B(TestEqual), R(2), U8(0),
[parser] avoid complex for-loop desugaring when possible let/const declarations in "standard" C-style for-loops have some complex desugaring to accommodate the case where loop loop variables may be captured. This slows down the baseline performance of for-loops with let variables. This change attempts to avoid this desugaring if it's known that the loop variable is not captured at any point. A side effect of this change is that let/const loop variables, when not captured within the loop body, are not necessarily shown in the debugger, similar to other stack-allocated vars. BUG=v8:4762, v8:5460 R=marja@chromium.org, adamk@chromium.org, yangguo@chromium.org Change-Id: I8dbe545a12c086f675972bdba60c94998268311a Reviewed-on: https://chromium-review.googlesource.com/472247 Commit-Queue: Caitlin Potter <caitp@igalia.com> Reviewed-by: Yang Guo <yangguo@chromium.org> Reviewed-by: Adam Klein <adamk@chromium.org> Reviewed-by: Marja Hölttä <marja@chromium.org> Cr-Commit-Position: refs/heads/master@{#44731}
2017-04-18 14:56:06 +00:00
B(JumpIfFalse), U8(7),
B(LdaZero),
B(Star), R(2),
B(Jump), U8(8),
/* 43 S> */ B(LdaCurrentContextSlot), U8(4),
[objects] Make feedback vector a first-class object Instead of having feedback vector as a subtype of FixedArray with reserved slots, make it a first-class variable-sized object with a fixed-size header. This allows us to compress counters to ints in the header, rather than forcing them to be Smis. Cq-Include-Trybots: master.tryserver.chromium.linux:linux_chromium_rel_ng Change-Id: Icc5f088ffbc2e2651b845bc71ea42060639e3e48 Reviewed-on: https://chromium-review.googlesource.com/585129 Commit-Queue: Leszek Swirski <leszeks@chromium.org> Reviewed-by: Michael Stanton <mvstanton@chromium.org> Reviewed-by: Ross McIlroy <rmcilroy@chromium.org> Reviewed-by: Michael Lippautz <mlippautz@chromium.org> Cr-Commit-Position: refs/heads/master@{#46935}
2017-07-27 12:45:00 +00:00
B(Inc), U8(1),
[parser] avoid complex for-loop desugaring when possible let/const declarations in "standard" C-style for-loops have some complex desugaring to accommodate the case where loop loop variables may be captured. This slows down the baseline performance of for-loops with let variables. This change attempts to avoid this desugaring if it's known that the loop variable is not captured at any point. A side effect of this change is that let/const loop variables, when not captured within the loop body, are not necessarily shown in the debugger, similar to other stack-allocated vars. BUG=v8:4762, v8:5460 R=marja@chromium.org, adamk@chromium.org, yangguo@chromium.org Change-Id: I8dbe545a12c086f675972bdba60c94998268311a Reviewed-on: https://chromium-review.googlesource.com/472247 Commit-Queue: Caitlin Potter <caitp@igalia.com> Reviewed-by: Yang Guo <yangguo@chromium.org> Reviewed-by: Adam Klein <adamk@chromium.org> Reviewed-by: Marja Hölttä <marja@chromium.org> Cr-Commit-Position: refs/heads/master@{#44731}
2017-04-18 14:56:06 +00:00
/* 43 E> */ B(StaCurrentContextSlot), U8(4),
B(LdaSmi), I8(1),
B(Star), R(3),
/* 35 S> */ B(LdaCurrentContextSlot), U8(4),
B(Star), R(5),
B(LdaSmi), I8(10),
[objects] Make feedback vector a first-class object Instead of having feedback vector as a subtype of FixedArray with reserved slots, make it a first-class variable-sized object with a fixed-size header. This allows us to compress counters to ints in the header, rather than forcing them to be Smis. Cq-Include-Trybots: master.tryserver.chromium.linux:linux_chromium_rel_ng Change-Id: Icc5f088ffbc2e2651b845bc71ea42060639e3e48 Reviewed-on: https://chromium-review.googlesource.com/585129 Commit-Queue: Leszek Swirski <leszeks@chromium.org> Reviewed-by: Michael Stanton <mvstanton@chromium.org> Reviewed-by: Ross McIlroy <rmcilroy@chromium.org> Reviewed-by: Michael Lippautz <mlippautz@chromium.org> Cr-Commit-Position: refs/heads/master@{#46935}
2017-07-27 12:45:00 +00:00
/* 35 E> */ B(TestLessThan), R(5), U8(2),
[parser] avoid complex for-loop desugaring when possible let/const declarations in "standard" C-style for-loops have some complex desugaring to accommodate the case where loop loop variables may be captured. This slows down the baseline performance of for-loops with let variables. This change attempts to avoid this desugaring if it's known that the loop variable is not captured at any point. A side effect of this change is that let/const loop variables, when not captured within the loop body, are not necessarily shown in the debugger, similar to other stack-allocated vars. BUG=v8:4762, v8:5460 R=marja@chromium.org, adamk@chromium.org, yangguo@chromium.org Change-Id: I8dbe545a12c086f675972bdba60c94998268311a Reviewed-on: https://chromium-review.googlesource.com/472247 Commit-Queue: Caitlin Potter <caitp@igalia.com> Reviewed-by: Yang Guo <yangguo@chromium.org> Reviewed-by: Adam Klein <adamk@chromium.org> Reviewed-by: Marja Hölttä <marja@chromium.org> Cr-Commit-Position: refs/heads/master@{#44731}
2017-04-18 14:56:06 +00:00
B(JumpIfFalse), U8(4),
B(Jump), U8(6),
B(PopContext), R(4),
B(Jump), U8(45),
B(LdaSmi), I8(1),
[objects] Make feedback vector a first-class object Instead of having feedback vector as a subtype of FixedArray with reserved slots, make it a first-class variable-sized object with a fixed-size header. This allows us to compress counters to ints in the header, rather than forcing them to be Smis. Cq-Include-Trybots: master.tryserver.chromium.linux:linux_chromium_rel_ng Change-Id: Icc5f088ffbc2e2651b845bc71ea42060639e3e48 Reviewed-on: https://chromium-review.googlesource.com/585129 Commit-Queue: Leszek Swirski <leszeks@chromium.org> Reviewed-by: Michael Stanton <mvstanton@chromium.org> Reviewed-by: Ross McIlroy <rmcilroy@chromium.org> Reviewed-by: Michael Lippautz <mlippautz@chromium.org> Cr-Commit-Position: refs/heads/master@{#46935}
2017-07-27 12:45:00 +00:00
B(TestEqual), R(3), U8(3),
[parser] avoid complex for-loop desugaring when possible let/const declarations in "standard" C-style for-loops have some complex desugaring to accommodate the case where loop loop variables may be captured. This slows down the baseline performance of for-loops with let variables. This change attempts to avoid this desugaring if it's known that the loop variable is not captured at any point. A side effect of this change is that let/const loop variables, when not captured within the loop body, are not necessarily shown in the debugger, similar to other stack-allocated vars. BUG=v8:4762, v8:5460 R=marja@chromium.org, adamk@chromium.org, yangguo@chromium.org Change-Id: I8dbe545a12c086f675972bdba60c94998268311a Reviewed-on: https://chromium-review.googlesource.com/472247 Commit-Queue: Caitlin Potter <caitp@igalia.com> Reviewed-by: Yang Guo <yangguo@chromium.org> Reviewed-by: Adam Klein <adamk@chromium.org> Reviewed-by: Marja Hölttä <marja@chromium.org> Cr-Commit-Position: refs/heads/master@{#44731}
2017-04-18 14:56:06 +00:00
B(JumpIfFalse), U8(22),
/* 17 E> */ B(StackCheck),
[Cleanup][Interpreter] Move feedback slot allocation to bytecode generator Moves the feedback vector slot allocation out of ast-numbering and into bytecode generation directly. This has a couple of benifits, including reduced AST size, avoid code duplication and reduced feedback vector sizes in many cases due to only allocating slots when needed. Also removes AstProperties since this is no longer needed. AstNumbering is now only used to allocate suspend ids for generators. BUG=v8:6921 Change-Id: I103e8593c94ef5b2e56c34ef4f77bd6e7d64796f Reviewed-on: https://chromium-review.googlesource.com/722959 Commit-Queue: Ross McIlroy <rmcilroy@chromium.org> Reviewed-by: Michael Stanton <mvstanton@chromium.org> Reviewed-by: Michael Starzinger <mstarzinger@chromium.org> Cr-Commit-Position: refs/heads/master@{#48757}
2017-10-19 15:12:42 +00:00
/* 48 S> */ B(CreateClosure), U8(1), U8(4), U8(2),
[parser] avoid complex for-loop desugaring when possible let/const declarations in "standard" C-style for-loops have some complex desugaring to accommodate the case where loop loop variables may be captured. This slows down the baseline performance of for-loops with let variables. This change attempts to avoid this desugaring if it's known that the loop variable is not captured at any point. A side effect of this change is that let/const loop variables, when not captured within the loop body, are not necessarily shown in the debugger, similar to other stack-allocated vars. BUG=v8:4762, v8:5460 R=marja@chromium.org, adamk@chromium.org, yangguo@chromium.org Change-Id: I8dbe545a12c086f675972bdba60c94998268311a Reviewed-on: https://chromium-review.googlesource.com/472247 Commit-Queue: Caitlin Potter <caitp@igalia.com> Reviewed-by: Yang Guo <yangguo@chromium.org> Reviewed-by: Adam Klein <adamk@chromium.org> Reviewed-by: Marja Hölttä <marja@chromium.org> Cr-Commit-Position: refs/heads/master@{#44731}
2017-04-18 14:56:06 +00:00
B(Star), R(5),
[Cleanup][Interpreter] Move feedback slot allocation to bytecode generator Moves the feedback vector slot allocation out of ast-numbering and into bytecode generation directly. This has a couple of benifits, including reduced AST size, avoid code duplication and reduced feedback vector sizes in many cases due to only allocating slots when needed. Also removes AstProperties since this is no longer needed. AstNumbering is now only used to allocate suspend ids for generators. BUG=v8:6921 Change-Id: I103e8593c94ef5b2e56c34ef4f77bd6e7d64796f Reviewed-on: https://chromium-review.googlesource.com/722959 Commit-Queue: Ross McIlroy <rmcilroy@chromium.org> Reviewed-by: Michael Stanton <mvstanton@chromium.org> Reviewed-by: Michael Starzinger <mstarzinger@chromium.org> Cr-Commit-Position: refs/heads/master@{#48757}
2017-10-19 15:12:42 +00:00
/* 74 E> */ B(CallUndefinedReceiver0), R(5), U8(5),
[parser] avoid complex for-loop desugaring when possible let/const declarations in "standard" C-style for-loops have some complex desugaring to accommodate the case where loop loop variables may be captured. This slows down the baseline performance of for-loops with let variables. This change attempts to avoid this desugaring if it's known that the loop variable is not captured at any point. A side effect of this change is that let/const loop variables, when not captured within the loop body, are not necessarily shown in the debugger, similar to other stack-allocated vars. BUG=v8:4762, v8:5460 R=marja@chromium.org, adamk@chromium.org, yangguo@chromium.org Change-Id: I8dbe545a12c086f675972bdba60c94998268311a Reviewed-on: https://chromium-review.googlesource.com/472247 Commit-Queue: Caitlin Potter <caitp@igalia.com> Reviewed-by: Yang Guo <yangguo@chromium.org> Reviewed-by: Adam Klein <adamk@chromium.org> Reviewed-by: Marja Hölttä <marja@chromium.org> Cr-Commit-Position: refs/heads/master@{#44731}
2017-04-18 14:56:06 +00:00
B(LdaZero),
B(Star), R(3),
B(LdaCurrentContextSlot), U8(4),
B(Star), R(1),
B(JumpLoop), U8(24), I8(1),
B(LdaSmi), I8(1),
[objects] Make feedback vector a first-class object Instead of having feedback vector as a subtype of FixedArray with reserved slots, make it a first-class variable-sized object with a fixed-size header. This allows us to compress counters to ints in the header, rather than forcing them to be Smis. Cq-Include-Trybots: master.tryserver.chromium.linux:linux_chromium_rel_ng Change-Id: Icc5f088ffbc2e2651b845bc71ea42060639e3e48 Reviewed-on: https://chromium-review.googlesource.com/585129 Commit-Queue: Leszek Swirski <leszeks@chromium.org> Reviewed-by: Michael Stanton <mvstanton@chromium.org> Reviewed-by: Ross McIlroy <rmcilroy@chromium.org> Reviewed-by: Michael Lippautz <mlippautz@chromium.org> Cr-Commit-Position: refs/heads/master@{#46935}
2017-07-27 12:45:00 +00:00
/* 78 E> */ B(TestEqual), R(3), U8(7),
[parser] avoid complex for-loop desugaring when possible let/const declarations in "standard" C-style for-loops have some complex desugaring to accommodate the case where loop loop variables may be captured. This slows down the baseline performance of for-loops with let variables. This change attempts to avoid this desugaring if it's known that the loop variable is not captured at any point. A side effect of this change is that let/const loop variables, when not captured within the loop body, are not necessarily shown in the debugger, similar to other stack-allocated vars. BUG=v8:4762, v8:5460 R=marja@chromium.org, adamk@chromium.org, yangguo@chromium.org Change-Id: I8dbe545a12c086f675972bdba60c94998268311a Reviewed-on: https://chromium-review.googlesource.com/472247 Commit-Queue: Caitlin Potter <caitp@igalia.com> Reviewed-by: Yang Guo <yangguo@chromium.org> Reviewed-by: Adam Klein <adamk@chromium.org> Reviewed-by: Marja Hölttä <marja@chromium.org> Cr-Commit-Position: refs/heads/master@{#44731}
2017-04-18 14:56:06 +00:00
B(JumpIfFalse), U8(6),
B(PopContext), R(4),
B(Jump), U8(7),
B(PopContext), R(4),
Replace Context::closure with Context::scope_info, allowing closure to die. There are likely cleanups that can be done after this CL: - context-related functions in the interpreter and compiler take ScopeInfo as well as ScopeType and slot-count as input. The latter 2 should be directly derived from the former. We should be able to drop FunctionContextParameters. - ContextExtension is probably not needed anymore, since we now always have the correct scope_info directly in the SCOPE_INFO_INDEX slot. Bug: v8:7066 Cq-Include-Trybots: luci.chromium.try:linux_chromium_rel_ng;master.tryserver.blink:linux_trusty_blink_rel Change-Id: Ie1f6134c686a9f2183e54730d9cdd598a9e5ab67 Reviewed-on: https://chromium-review.googlesource.com/785151 Commit-Queue: Toon Verwaest <verwaest@chromium.org> Reviewed-by: Hannes Payer <hpayer@chromium.org> Reviewed-by: Adam Klein <adamk@chromium.org> Reviewed-by: Ross McIlroy <rmcilroy@chromium.org> Reviewed-by: Michael Starzinger <mstarzinger@chromium.org> Cr-Commit-Position: refs/heads/master@{#52952}
2018-05-02 13:57:26 +00:00
B(JumpLoop), U8(91), I8(0),
[parser] avoid complex for-loop desugaring when possible let/const declarations in "standard" C-style for-loops have some complex desugaring to accommodate the case where loop loop variables may be captured. This slows down the baseline performance of for-loops with let variables. This change attempts to avoid this desugaring if it's known that the loop variable is not captured at any point. A side effect of this change is that let/const loop variables, when not captured within the loop body, are not necessarily shown in the debugger, similar to other stack-allocated vars. BUG=v8:4762, v8:5460 R=marja@chromium.org, adamk@chromium.org, yangguo@chromium.org Change-Id: I8dbe545a12c086f675972bdba60c94998268311a Reviewed-on: https://chromium-review.googlesource.com/472247 Commit-Queue: Caitlin Potter <caitp@igalia.com> Reviewed-by: Yang Guo <yangguo@chromium.org> Reviewed-by: Adam Klein <adamk@chromium.org> Reviewed-by: Marja Hölttä <marja@chromium.org> Cr-Commit-Position: refs/heads/master@{#44731}
2017-04-18 14:56:06 +00:00
B(LdaUndefined),
/* 80 S> */ B(Return),
]
constant pool: [
[runtime] Add SCOPE_INFO_TYPE InstanceType Bug: v8:7310 Change-Id: I82e7ada4c0f7e415887a859719eb01bb45fd3012 Reviewed-on: https://chromium-review.googlesource.com/921742 Reviewed-by: Michael Lippautz <mlippautz@chromium.org> Reviewed-by: Peter Marshall <petermarshall@chromium.org> Commit-Queue: Camillo Bruni <cbruni@chromium.org> Cr-Commit-Position: refs/heads/master@{#51324}
2018-02-16 12:36:47 +00:00
SCOPE_INFO_TYPE,
[parser] avoid complex for-loop desugaring when possible let/const declarations in "standard" C-style for-loops have some complex desugaring to accommodate the case where loop loop variables may be captured. This slows down the baseline performance of for-loops with let variables. This change attempts to avoid this desugaring if it's known that the loop variable is not captured at any point. A side effect of this change is that let/const loop variables, when not captured within the loop body, are not necessarily shown in the debugger, similar to other stack-allocated vars. BUG=v8:4762, v8:5460 R=marja@chromium.org, adamk@chromium.org, yangguo@chromium.org Change-Id: I8dbe545a12c086f675972bdba60c94998268311a Reviewed-on: https://chromium-review.googlesource.com/472247 Commit-Queue: Caitlin Potter <caitp@igalia.com> Reviewed-by: Yang Guo <yangguo@chromium.org> Reviewed-by: Adam Klein <adamk@chromium.org> Reviewed-by: Marja Hölttä <marja@chromium.org> Cr-Commit-Position: refs/heads/master@{#44731}
2017-04-18 14:56:06 +00:00
SHARED_FUNCTION_INFO_TYPE,
]
handlers: [
]
---
snippet: "
function f() {
for (let { x, y } = { x: 0, y: 3 }; y > 0; --y) { let z = x + y; }
}
f();
"
frame size: 6
parameter count: 1
Skip hole initialization of lexical variables when possible This patch expands scope analysis to skip hole initialization when it can be determined statically that no hole checks will be generated at runtime. Two conditions must be met to safely eliminate hole initialization: - There must not exist a VariableProxy referencing this Variable whose HoleCheckMode is kRequired - The Variable must be stack allocated; any other allocation implies that it may be accessed from not-yet-analyzed scopes (other modules, inner functions, or eval code) and that code may require hole checks. The new logic required removing debug code in full-codegen which is now incorrect in some cases. Also fixed Variable's bitfield helpers to take no more space than needed. Bug: chromium:651637 Change-Id: Ie5ac326af4e05b7a5c3c37cd4d0afba6a51a504d Reviewed-on: https://chromium-review.googlesource.com/494006 Commit-Queue: Adam Klein <adamk@chromium.org> Reviewed-by: Georg Neis <neis@chromium.org> Reviewed-by: Leszek Swirski <leszeks@chromium.org> Reviewed-by: Michael Starzinger <mstarzinger@chromium.org> Cr-Commit-Position: refs/heads/master@{#45170}
2017-05-08 18:55:54 +00:00
bytecode array length: 68
[parser] avoid complex for-loop desugaring when possible let/const declarations in "standard" C-style for-loops have some complex desugaring to accommodate the case where loop loop variables may be captured. This slows down the baseline performance of for-loops with let variables. This change attempts to avoid this desugaring if it's known that the loop variable is not captured at any point. A side effect of this change is that let/const loop variables, when not captured within the loop body, are not necessarily shown in the debugger, similar to other stack-allocated vars. BUG=v8:4762, v8:5460 R=marja@chromium.org, adamk@chromium.org, yangguo@chromium.org Change-Id: I8dbe545a12c086f675972bdba60c94998268311a Reviewed-on: https://chromium-review.googlesource.com/472247 Commit-Queue: Caitlin Potter <caitp@igalia.com> Reviewed-by: Yang Guo <yangguo@chromium.org> Reviewed-by: Adam Klein <adamk@chromium.org> Reviewed-by: Marja Hölttä <marja@chromium.org> Cr-Commit-Position: refs/heads/master@{#44731}
2017-04-18 14:56:06 +00:00
bytecodes: [
/* 10 E> */ B(StackCheck),
[objects] Make feedback vector a first-class object Instead of having feedback vector as a subtype of FixedArray with reserved slots, make it a first-class variable-sized object with a fixed-size header. This allows us to compress counters to ints in the header, rather than forcing them to be Smis. Cq-Include-Trybots: master.tryserver.chromium.linux:linux_chromium_rel_ng Change-Id: Icc5f088ffbc2e2651b845bc71ea42060639e3e48 Reviewed-on: https://chromium-review.googlesource.com/585129 Commit-Queue: Leszek Swirski <leszeks@chromium.org> Reviewed-by: Michael Stanton <mvstanton@chromium.org> Reviewed-by: Ross McIlroy <rmcilroy@chromium.org> Reviewed-by: Michael Lippautz <mlippautz@chromium.org> Cr-Commit-Position: refs/heads/master@{#46935}
2017-07-27 12:45:00 +00:00
B(CreateObjectLiteral), U8(0), U8(0), U8(41), R(4),
[parser] avoid complex for-loop desugaring when possible let/const declarations in "standard" C-style for-loops have some complex desugaring to accommodate the case where loop loop variables may be captured. This slows down the baseline performance of for-loops with let variables. This change attempts to avoid this desugaring if it's known that the loop variable is not captured at any point. A side effect of this change is that let/const loop variables, when not captured within the loop body, are not necessarily shown in the debugger, similar to other stack-allocated vars. BUG=v8:4762, v8:5460 R=marja@chromium.org, adamk@chromium.org, yangguo@chromium.org Change-Id: I8dbe545a12c086f675972bdba60c94998268311a Reviewed-on: https://chromium-review.googlesource.com/472247 Commit-Queue: Caitlin Potter <caitp@igalia.com> Reviewed-by: Yang Guo <yangguo@chromium.org> Reviewed-by: Adam Klein <adamk@chromium.org> Reviewed-by: Marja Hölttä <marja@chromium.org> Cr-Commit-Position: refs/heads/master@{#44731}
2017-04-18 14:56:06 +00:00
B(Mov), R(4), R(3),
B(Ldar), R(3),
B(JumpIfUndefined), U8(6),
B(Ldar), R(3),
B(JumpIfNotNull), U8(16),
[Intl] Implement Intl.ListFromat format() and formatToParts(). Spec: http://tc39.github.io/proposal-intl-list-format/ Design Doc: go/add-intl.listformat-to-v8 Test: intl/list-format/* R=gsathya@chromium.org, mvstanton@chromium.org Bug: v8:7871 Cq-Include-Trybots: luci.v8.try:v8_linux_noi18n_rel_ng Change-Id: I820c205ca842c228ffe37f7e1648667f30f80bd8 Reviewed-on: https://chromium-review.googlesource.com/1126683 Commit-Queue: Frank Tang <ftang@chromium.org> Reviewed-by: Adam Klein <adamk@chromium.org> Cr-Commit-Position: refs/heads/master@{#54991}
2018-08-08 23:04:11 +00:00
B(LdaSmi), I8(81),
[parser] avoid complex for-loop desugaring when possible let/const declarations in "standard" C-style for-loops have some complex desugaring to accommodate the case where loop loop variables may be captured. This slows down the baseline performance of for-loops with let variables. This change attempts to avoid this desugaring if it's known that the loop variable is not captured at any point. A side effect of this change is that let/const loop variables, when not captured within the loop body, are not necessarily shown in the debugger, similar to other stack-allocated vars. BUG=v8:4762, v8:5460 R=marja@chromium.org, adamk@chromium.org, yangguo@chromium.org Change-Id: I8dbe545a12c086f675972bdba60c94998268311a Reviewed-on: https://chromium-review.googlesource.com/472247 Commit-Queue: Caitlin Potter <caitp@igalia.com> Reviewed-by: Yang Guo <yangguo@chromium.org> Reviewed-by: Adam Klein <adamk@chromium.org> Reviewed-by: Marja Hölttä <marja@chromium.org> Cr-Commit-Position: refs/heads/master@{#44731}
2017-04-18 14:56:06 +00:00
B(Star), R(4),
B(LdaConstant), U8(1),
B(Star), R(5),
B(CallRuntime), U16(Runtime::kNewTypeError), R(4), U8(2),
[parser] Better error message when destructuring against undefined/null Previously, when destructuring against null or undefined we would print: d8> var { x } = null (d8):1: TypeError: Cannot match against 'undefined' or 'null'. var { x } = null ^ TypeError: Cannot match against 'undefined' or 'null'. at (d8):1:1 The above message uses the term "match" which isn't a common term in JavaScript to describe destructuring. This message also doesn't provide the name of the property that fails destructuring. This patch changes the error message to be: d8> var { x } = null; (d8):1: TypeError: Cannot destructure property `x` of 'undefined' or 'null'. var { x } = null; ^ TypeError: Cannot destructure property `x` of 'undefined' or 'null'. at (d8):1:1 This patch changes the message to say "destructure" instead of "match". This patch adds support for printing property names that are string literals. We iterate through every property and pick the first string literal property name if it exists. This provides at least some feedback to the developer. This patch also makes the pointer point to the position of the property name that fails destructuring. For computed and numeric property names, we print a generic error: d8> var { 1: x } = null (d8):1: TypeError: Cannot destructure against 'undefined' or 'null'. var { 1: x } = null ^ TypeError: Cannot destructure against 'undefined' or 'null'. at (d8):1:1 Bug: v8:6499 Change-Id: I35b1ac749489828686f042975294b9926e2dfc53 Reviewed-on: https://chromium-review.googlesource.com/537341 Reviewed-by: Adam Klein <adamk@chromium.org> Commit-Queue: Sathya Gunasekaran <gsathya@chromium.org> Cr-Commit-Position: refs/heads/master@{#45965}
2017-06-15 21:24:37 +00:00
/* 28 E> */ B(Throw),
[Cleanup][Interpreter] Move feedback slot allocation to bytecode generator Moves the feedback vector slot allocation out of ast-numbering and into bytecode generation directly. This has a couple of benifits, including reduced AST size, avoid code duplication and reduced feedback vector sizes in many cases due to only allocating slots when needed. Also removes AstProperties since this is no longer needed. AstNumbering is now only used to allocate suspend ids for generators. BUG=v8:6921 Change-Id: I103e8593c94ef5b2e56c34ef4f77bd6e7d64796f Reviewed-on: https://chromium-review.googlesource.com/722959 Commit-Queue: Ross McIlroy <rmcilroy@chromium.org> Reviewed-by: Michael Stanton <mvstanton@chromium.org> Reviewed-by: Michael Starzinger <mstarzinger@chromium.org> Cr-Commit-Position: refs/heads/master@{#48757}
2017-10-19 15:12:42 +00:00
/* 37 S> */ B(LdaNamedProperty), R(3), U8(1), U8(1),
[parser] avoid complex for-loop desugaring when possible let/const declarations in "standard" C-style for-loops have some complex desugaring to accommodate the case where loop loop variables may be captured. This slows down the baseline performance of for-loops with let variables. This change attempts to avoid this desugaring if it's known that the loop variable is not captured at any point. A side effect of this change is that let/const loop variables, when not captured within the loop body, are not necessarily shown in the debugger, similar to other stack-allocated vars. BUG=v8:4762, v8:5460 R=marja@chromium.org, adamk@chromium.org, yangguo@chromium.org Change-Id: I8dbe545a12c086f675972bdba60c94998268311a Reviewed-on: https://chromium-review.googlesource.com/472247 Commit-Queue: Caitlin Potter <caitp@igalia.com> Reviewed-by: Yang Guo <yangguo@chromium.org> Reviewed-by: Adam Klein <adamk@chromium.org> Reviewed-by: Marja Hölttä <marja@chromium.org> Cr-Commit-Position: refs/heads/master@{#44731}
2017-04-18 14:56:06 +00:00
B(Star), R(1),
[Cleanup][Interpreter] Move feedback slot allocation to bytecode generator Moves the feedback vector slot allocation out of ast-numbering and into bytecode generation directly. This has a couple of benifits, including reduced AST size, avoid code duplication and reduced feedback vector sizes in many cases due to only allocating slots when needed. Also removes AstProperties since this is no longer needed. AstNumbering is now only used to allocate suspend ids for generators. BUG=v8:6921 Change-Id: I103e8593c94ef5b2e56c34ef4f77bd6e7d64796f Reviewed-on: https://chromium-review.googlesource.com/722959 Commit-Queue: Ross McIlroy <rmcilroy@chromium.org> Reviewed-by: Michael Stanton <mvstanton@chromium.org> Reviewed-by: Michael Starzinger <mstarzinger@chromium.org> Cr-Commit-Position: refs/heads/master@{#48757}
2017-10-19 15:12:42 +00:00
/* 37 S> */ B(LdaNamedProperty), R(3), U8(2), U8(3),
[parser] avoid complex for-loop desugaring when possible let/const declarations in "standard" C-style for-loops have some complex desugaring to accommodate the case where loop loop variables may be captured. This slows down the baseline performance of for-loops with let variables. This change attempts to avoid this desugaring if it's known that the loop variable is not captured at any point. A side effect of this change is that let/const loop variables, when not captured within the loop body, are not necessarily shown in the debugger, similar to other stack-allocated vars. BUG=v8:4762, v8:5460 R=marja@chromium.org, adamk@chromium.org, yangguo@chromium.org Change-Id: I8dbe545a12c086f675972bdba60c94998268311a Reviewed-on: https://chromium-review.googlesource.com/472247 Commit-Queue: Caitlin Potter <caitp@igalia.com> Reviewed-by: Yang Guo <yangguo@chromium.org> Reviewed-by: Adam Klein <adamk@chromium.org> Reviewed-by: Marja Hölttä <marja@chromium.org> Cr-Commit-Position: refs/heads/master@{#44731}
2017-04-18 14:56:06 +00:00
B(Star), R(2),
/* 55 S> */ B(LdaZero),
[Cleanup][Interpreter] Move feedback slot allocation to bytecode generator Moves the feedback vector slot allocation out of ast-numbering and into bytecode generation directly. This has a couple of benifits, including reduced AST size, avoid code duplication and reduced feedback vector sizes in many cases due to only allocating slots when needed. Also removes AstProperties since this is no longer needed. AstNumbering is now only used to allocate suspend ids for generators. BUG=v8:6921 Change-Id: I103e8593c94ef5b2e56c34ef4f77bd6e7d64796f Reviewed-on: https://chromium-review.googlesource.com/722959 Commit-Queue: Ross McIlroy <rmcilroy@chromium.org> Reviewed-by: Michael Stanton <mvstanton@chromium.org> Reviewed-by: Michael Starzinger <mstarzinger@chromium.org> Cr-Commit-Position: refs/heads/master@{#48757}
2017-10-19 15:12:42 +00:00
/* 55 E> */ B(TestGreaterThan), R(2), U8(5),
Skip hole initialization of lexical variables when possible This patch expands scope analysis to skip hole initialization when it can be determined statically that no hole checks will be generated at runtime. Two conditions must be met to safely eliminate hole initialization: - There must not exist a VariableProxy referencing this Variable whose HoleCheckMode is kRequired - The Variable must be stack allocated; any other allocation implies that it may be accessed from not-yet-analyzed scopes (other modules, inner functions, or eval code) and that code may require hole checks. The new logic required removing debug code in full-codegen which is now incorrect in some cases. Also fixed Variable's bitfield helpers to take no more space than needed. Bug: chromium:651637 Change-Id: Ie5ac326af4e05b7a5c3c37cd4d0afba6a51a504d Reviewed-on: https://chromium-review.googlesource.com/494006 Commit-Queue: Adam Klein <adamk@chromium.org> Reviewed-by: Georg Neis <neis@chromium.org> Reviewed-by: Leszek Swirski <leszeks@chromium.org> Reviewed-by: Michael Starzinger <mstarzinger@chromium.org> Cr-Commit-Position: refs/heads/master@{#45170}
2017-05-08 18:55:54 +00:00
B(JumpIfFalse), U8(19),
[parser] avoid complex for-loop desugaring when possible let/const declarations in "standard" C-style for-loops have some complex desugaring to accommodate the case where loop loop variables may be captured. This slows down the baseline performance of for-loops with let variables. This change attempts to avoid this desugaring if it's known that the loop variable is not captured at any point. A side effect of this change is that let/const loop variables, when not captured within the loop body, are not necessarily shown in the debugger, similar to other stack-allocated vars. BUG=v8:4762, v8:5460 R=marja@chromium.org, adamk@chromium.org, yangguo@chromium.org Change-Id: I8dbe545a12c086f675972bdba60c94998268311a Reviewed-on: https://chromium-review.googlesource.com/472247 Commit-Queue: Caitlin Potter <caitp@igalia.com> Reviewed-by: Yang Guo <yangguo@chromium.org> Reviewed-by: Adam Klein <adamk@chromium.org> Reviewed-by: Marja Hölttä <marja@chromium.org> Cr-Commit-Position: refs/heads/master@{#44731}
2017-04-18 14:56:06 +00:00
/* 17 E> */ B(StackCheck),
[inspector] moved var initialization break location before init expression (reland) This CL improves break locations for expressions like 'var a = <expr>'. Without CL we use <expr> position as break location for initialization statement, with this CL we use position of first character after '=' as position. Benefits (see test for details): - only one break in expressions which includes mix of property lookup and calls, e.g. var p = Promise.resolve().then(x => x * 2), - removed redundant break location for expressions like: let { x, y } = { x: 1, y: 2}. TBR=dgozman@chromium.org,rmcilroy@chromium.org,machenbach@chromium.org,marja@chromium.org,kozyatinskiy@chromium.org,devtools-reviews@chromium.org,v8-reviews@googlegroups.com # Not skipping CQ checks because original CL landed > 1 day ago. Bug: v8:5909 Change-Id: Ie84fa79afeed09e28cf8478ba610a0cfbfdfc294 Reviewed-on: https://chromium-review.googlesource.com/518116 Commit-Queue: Aleksey Kozyatinskiy <kozyatinskiy@chromium.org> Reviewed-by: Aleksey Kozyatinskiy <kozyatinskiy@chromium.org> Cr-Commit-Position: refs/heads/master@{#45598}
2017-05-30 12:25:55 +00:00
/* 75 S> */ B(Ldar), R(2),
[Cleanup][Interpreter] Move feedback slot allocation to bytecode generator Moves the feedback vector slot allocation out of ast-numbering and into bytecode generation directly. This has a couple of benifits, including reduced AST size, avoid code duplication and reduced feedback vector sizes in many cases due to only allocating slots when needed. Also removes AstProperties since this is no longer needed. AstNumbering is now only used to allocate suspend ids for generators. BUG=v8:6921 Change-Id: I103e8593c94ef5b2e56c34ef4f77bd6e7d64796f Reviewed-on: https://chromium-review.googlesource.com/722959 Commit-Queue: Ross McIlroy <rmcilroy@chromium.org> Reviewed-by: Michael Stanton <mvstanton@chromium.org> Reviewed-by: Michael Starzinger <mstarzinger@chromium.org> Cr-Commit-Position: refs/heads/master@{#48757}
2017-10-19 15:12:42 +00:00
/* 77 E> */ B(Add), R(1), U8(6),
[parser] avoid complex for-loop desugaring when possible let/const declarations in "standard" C-style for-loops have some complex desugaring to accommodate the case where loop loop variables may be captured. This slows down the baseline performance of for-loops with let variables. This change attempts to avoid this desugaring if it's known that the loop variable is not captured at any point. A side effect of this change is that let/const loop variables, when not captured within the loop body, are not necessarily shown in the debugger, similar to other stack-allocated vars. BUG=v8:4762, v8:5460 R=marja@chromium.org, adamk@chromium.org, yangguo@chromium.org Change-Id: I8dbe545a12c086f675972bdba60c94998268311a Reviewed-on: https://chromium-review.googlesource.com/472247 Commit-Queue: Caitlin Potter <caitp@igalia.com> Reviewed-by: Yang Guo <yangguo@chromium.org> Reviewed-by: Adam Klein <adamk@chromium.org> Reviewed-by: Marja Hölttä <marja@chromium.org> Cr-Commit-Position: refs/heads/master@{#44731}
2017-04-18 14:56:06 +00:00
B(Star), R(0),
/* 62 S> */ B(Ldar), R(2),
[Cleanup][Interpreter] Move feedback slot allocation to bytecode generator Moves the feedback vector slot allocation out of ast-numbering and into bytecode generation directly. This has a couple of benifits, including reduced AST size, avoid code duplication and reduced feedback vector sizes in many cases due to only allocating slots when needed. Also removes AstProperties since this is no longer needed. AstNumbering is now only used to allocate suspend ids for generators. BUG=v8:6921 Change-Id: I103e8593c94ef5b2e56c34ef4f77bd6e7d64796f Reviewed-on: https://chromium-review.googlesource.com/722959 Commit-Queue: Ross McIlroy <rmcilroy@chromium.org> Reviewed-by: Michael Stanton <mvstanton@chromium.org> Reviewed-by: Michael Starzinger <mstarzinger@chromium.org> Cr-Commit-Position: refs/heads/master@{#48757}
2017-10-19 15:12:42 +00:00
B(Dec), U8(7),
[parser] avoid complex for-loop desugaring when possible let/const declarations in "standard" C-style for-loops have some complex desugaring to accommodate the case where loop loop variables may be captured. This slows down the baseline performance of for-loops with let variables. This change attempts to avoid this desugaring if it's known that the loop variable is not captured at any point. A side effect of this change is that let/const loop variables, when not captured within the loop body, are not necessarily shown in the debugger, similar to other stack-allocated vars. BUG=v8:4762, v8:5460 R=marja@chromium.org, adamk@chromium.org, yangguo@chromium.org Change-Id: I8dbe545a12c086f675972bdba60c94998268311a Reviewed-on: https://chromium-review.googlesource.com/472247 Commit-Queue: Caitlin Potter <caitp@igalia.com> Reviewed-by: Yang Guo <yangguo@chromium.org> Reviewed-by: Adam Klein <adamk@chromium.org> Reviewed-by: Marja Hölttä <marja@chromium.org> Cr-Commit-Position: refs/heads/master@{#44731}
2017-04-18 14:56:06 +00:00
B(Star), R(2),
Skip hole initialization of lexical variables when possible This patch expands scope analysis to skip hole initialization when it can be determined statically that no hole checks will be generated at runtime. Two conditions must be met to safely eliminate hole initialization: - There must not exist a VariableProxy referencing this Variable whose HoleCheckMode is kRequired - The Variable must be stack allocated; any other allocation implies that it may be accessed from not-yet-analyzed scopes (other modules, inner functions, or eval code) and that code may require hole checks. The new logic required removing debug code in full-codegen which is now incorrect in some cases. Also fixed Variable's bitfield helpers to take no more space than needed. Bug: chromium:651637 Change-Id: Ie5ac326af4e05b7a5c3c37cd4d0afba6a51a504d Reviewed-on: https://chromium-review.googlesource.com/494006 Commit-Queue: Adam Klein <adamk@chromium.org> Reviewed-by: Georg Neis <neis@chromium.org> Reviewed-by: Leszek Swirski <leszeks@chromium.org> Reviewed-by: Michael Starzinger <mstarzinger@chromium.org> Cr-Commit-Position: refs/heads/master@{#45170}
2017-05-08 18:55:54 +00:00
B(JumpLoop), U8(20), I8(0),
[parser] avoid complex for-loop desugaring when possible let/const declarations in "standard" C-style for-loops have some complex desugaring to accommodate the case where loop loop variables may be captured. This slows down the baseline performance of for-loops with let variables. This change attempts to avoid this desugaring if it's known that the loop variable is not captured at any point. A side effect of this change is that let/const loop variables, when not captured within the loop body, are not necessarily shown in the debugger, similar to other stack-allocated vars. BUG=v8:4762, v8:5460 R=marja@chromium.org, adamk@chromium.org, yangguo@chromium.org Change-Id: I8dbe545a12c086f675972bdba60c94998268311a Reviewed-on: https://chromium-review.googlesource.com/472247 Commit-Queue: Caitlin Potter <caitp@igalia.com> Reviewed-by: Yang Guo <yangguo@chromium.org> Reviewed-by: Adam Klein <adamk@chromium.org> Reviewed-by: Marja Hölttä <marja@chromium.org> Cr-Commit-Position: refs/heads/master@{#44731}
2017-04-18 14:56:06 +00:00
B(LdaUndefined),
/* 84 S> */ B(Return),
]
constant pool: [
[runtime][parser] Use ArrayBoilerplateDescription all the way * Rename BoilerplateDescription to ObjectBoilerplateDescription * Add literal_type flag to ObjectBoilerplateDescription, which is stored as zeroth element of Fixed array * Create ArrayBoilerplateDescription with elements_kind and constant_elements field * Replace CompileTimeValue and ConstantElementPair with ArrayBoilerplateDescription * Kill ConstantElementPair and CompileTimeValue Change-Id: Icb42dcfd575a27e2b64ffd5e2e61f9d703d5e986 Bug: v8:7787, chromium:818642 Reviewed-on: https://chromium-review.googlesource.com/1122411 Commit-Queue: Chandan Reddy <chandanreddy@google.com> Reviewed-by: Michael Lippautz <mlippautz@chromium.org> Reviewed-by: Camillo Bruni <cbruni@chromium.org> Reviewed-by: Jaroslav Sevcik <jarin@chromium.org> Reviewed-by: Ross McIlroy <rmcilroy@chromium.org> Reviewed-by: Marja Hölttä <marja@chromium.org> Cr-Commit-Position: refs/heads/master@{#54272}
2018-07-05 18:34:09 +00:00
OBJECT_BOILERPLATE_DESCRIPTION_TYPE,
[parser] avoid complex for-loop desugaring when possible let/const declarations in "standard" C-style for-loops have some complex desugaring to accommodate the case where loop loop variables may be captured. This slows down the baseline performance of for-loops with let variables. This change attempts to avoid this desugaring if it's known that the loop variable is not captured at any point. A side effect of this change is that let/const loop variables, when not captured within the loop body, are not necessarily shown in the debugger, similar to other stack-allocated vars. BUG=v8:4762, v8:5460 R=marja@chromium.org, adamk@chromium.org, yangguo@chromium.org Change-Id: I8dbe545a12c086f675972bdba60c94998268311a Reviewed-on: https://chromium-review.googlesource.com/472247 Commit-Queue: Caitlin Potter <caitp@igalia.com> Reviewed-by: Yang Guo <yangguo@chromium.org> Reviewed-by: Adam Klein <adamk@chromium.org> Reviewed-by: Marja Hölttä <marja@chromium.org> Cr-Commit-Position: refs/heads/master@{#44731}
2017-04-18 14:56:06 +00:00
ONE_BYTE_INTERNALIZED_STRING_TYPE ["x"],
ONE_BYTE_INTERNALIZED_STRING_TYPE ["y"],
]
handlers: [
]
---
snippet: "
function* f() {
for (let x = 0; x < 10; ++x) { let y = x; }
}
f();
"
[ignition] Single-switch generator bytecode Currently, yields and awaits inside loops compile to bytecode which switches to the top of the loop header, and switch again once inside the loop. This is to make loops reducible. This replaces this switching logic with a single switch bytecode that directly jumps to the bytecode being resumed. Among other things, this allows us to no longer maintain the generator state after the switch at the top of the function, and avoid having to track loop suspend counts. TurboFan still needs to have reducible loops, so we now insert loop header switches during bytecode graph building, for suspends that are discovered to be inside loops during bytecode analysis. We do, however, do some environment magic across loop headers since we know that we will continue switching if and only if we reached that loop header via a generator resume. This allows us to generate fewer phis and tighten liveness. Change-Id: Id2720ce1d6955be9a48178322cc209b3a4b8d385 Reviewed-on: https://chromium-review.googlesource.com/866734 Commit-Queue: Leszek Swirski <leszeks@chromium.org> Reviewed-by: Jaroslav Sevcik <jarin@chromium.org> Cr-Commit-Position: refs/heads/master@{#50804}
2018-01-23 13:51:38 +00:00
frame size: 5
[parser] avoid complex for-loop desugaring when possible let/const declarations in "standard" C-style for-loops have some complex desugaring to accommodate the case where loop loop variables may be captured. This slows down the baseline performance of for-loops with let variables. This change attempts to avoid this desugaring if it's known that the loop variable is not captured at any point. A side effect of this change is that let/const loop variables, when not captured within the loop body, are not necessarily shown in the debugger, similar to other stack-allocated vars. BUG=v8:4762, v8:5460 R=marja@chromium.org, adamk@chromium.org, yangguo@chromium.org Change-Id: I8dbe545a12c086f675972bdba60c94998268311a Reviewed-on: https://chromium-review.googlesource.com/472247 Commit-Queue: Caitlin Potter <caitp@igalia.com> Reviewed-by: Yang Guo <yangguo@chromium.org> Reviewed-by: Adam Klein <adamk@chromium.org> Reviewed-by: Marja Hölttä <marja@chromium.org> Cr-Commit-Position: refs/heads/master@{#44731}
2017-04-18 14:56:06 +00:00
parameter count: 1
[ignition] Reduce generator switch boilerplate The SwitchOnGeneratorState bytecode now also falls through if the generator object is undefined (so that we don't need that jump) and restores generator context (so that we don't need that PushContext). This saves 10 bytes per generator. Change-Id: Ie0872c827119b9f1d1e9244d3be6496a30cd9620 Reviewed-on: https://chromium-review.googlesource.com/867051 Commit-Queue: Leszek Swirski <leszeks@chromium.org> Reviewed-by: Ross McIlroy <rmcilroy@chromium.org> Reviewed-by: Georg Neis <neis@chromium.org> Cr-Commit-Position: refs/heads/master@{#50845}
2018-01-24 11:19:51 +00:00
bytecode array length: 67
[parser] avoid complex for-loop desugaring when possible let/const declarations in "standard" C-style for-loops have some complex desugaring to accommodate the case where loop loop variables may be captured. This slows down the baseline performance of for-loops with let variables. This change attempts to avoid this desugaring if it's known that the loop variable is not captured at any point. A side effect of this change is that let/const loop variables, when not captured within the loop body, are not necessarily shown in the debugger, similar to other stack-allocated vars. BUG=v8:4762, v8:5460 R=marja@chromium.org, adamk@chromium.org, yangguo@chromium.org Change-Id: I8dbe545a12c086f675972bdba60c94998268311a Reviewed-on: https://chromium-review.googlesource.com/472247 Commit-Queue: Caitlin Potter <caitp@igalia.com> Reviewed-by: Yang Guo <yangguo@chromium.org> Reviewed-by: Adam Klein <adamk@chromium.org> Reviewed-by: Marja Hölttä <marja@chromium.org> Cr-Commit-Position: refs/heads/master@{#44731}
2017-04-18 14:56:06 +00:00
bytecodes: [
[ignition] Single-switch generator bytecode Currently, yields and awaits inside loops compile to bytecode which switches to the top of the loop header, and switch again once inside the loop. This is to make loops reducible. This replaces this switching logic with a single switch bytecode that directly jumps to the bytecode being resumed. Among other things, this allows us to no longer maintain the generator state after the switch at the top of the function, and avoid having to track loop suspend counts. TurboFan still needs to have reducible loops, so we now insert loop header switches during bytecode graph building, for suspends that are discovered to be inside loops during bytecode analysis. We do, however, do some environment magic across loop headers since we know that we will continue switching if and only if we reached that loop header via a generator resume. This allows us to generate fewer phis and tighten liveness. Change-Id: Id2720ce1d6955be9a48178322cc209b3a4b8d385 Reviewed-on: https://chromium-review.googlesource.com/866734 Commit-Queue: Leszek Swirski <leszeks@chromium.org> Reviewed-by: Jaroslav Sevcik <jarin@chromium.org> Cr-Commit-Position: refs/heads/master@{#50804}
2018-01-23 13:51:38 +00:00
B(SwitchOnGeneratorState), R(2), U8(0), U8(1),
B(Mov), R(closure), R(3),
B(Mov), R(this), R(4),
B(InvokeIntrinsic), U8(Runtime::k_CreateJSGeneratorObject), R(3), U8(2),
[parser] avoid for-loop desugaring unless loop var may be captured In https://chromium-review.googlesource.com/c/472247/, I avoided running DesugarLexicalBindingsInForStatement() if there were no lexical loop variables, the function was not resumable, and the variables are not captured by eval or a function declaration. I think it's now possible to limit this further, and only do the more extensive desugaring if there's a function declaration / eval() call in the loop body. `yield` and `await` are not an issue as those loop variables are written to the register file and not lost. This change just removes the `is_resumable()` condition. If it passes tests, I think it's safe. BUG=v8:4762, v8:5460, v8:6579 Change-Id: I92d0308ad9401c1338411bc9ae9021f978803d3a Reviewed-on: https://chromium-review.googlesource.com/563587 Commit-Queue: Caitlin Potter <caitp@igalia.com> Reviewed-by: Georg Neis <neis@chromium.org> Reviewed-by: Adam Klein <adamk@chromium.org> Cr-Commit-Position: refs/heads/master@{#46536}
2017-07-07 20:48:16 +00:00
B(Star), R(2),
[parser] avoid complex for-loop desugaring when possible let/const declarations in "standard" C-style for-loops have some complex desugaring to accommodate the case where loop loop variables may be captured. This slows down the baseline performance of for-loops with let variables. This change attempts to avoid this desugaring if it's known that the loop variable is not captured at any point. A side effect of this change is that let/const loop variables, when not captured within the loop body, are not necessarily shown in the debugger, similar to other stack-allocated vars. BUG=v8:4762, v8:5460 R=marja@chromium.org, adamk@chromium.org, yangguo@chromium.org Change-Id: I8dbe545a12c086f675972bdba60c94998268311a Reviewed-on: https://chromium-review.googlesource.com/472247 Commit-Queue: Caitlin Potter <caitp@igalia.com> Reviewed-by: Yang Guo <yangguo@chromium.org> Reviewed-by: Adam Klein <adamk@chromium.org> Reviewed-by: Marja Hölttä <marja@chromium.org> Cr-Commit-Position: refs/heads/master@{#44731}
2017-04-18 14:56:06 +00:00
/* 11 E> */ B(StackCheck),
[ignition] Single-switch generator bytecode Currently, yields and awaits inside loops compile to bytecode which switches to the top of the loop header, and switch again once inside the loop. This is to make loops reducible. This replaces this switching logic with a single switch bytecode that directly jumps to the bytecode being resumed. Among other things, this allows us to no longer maintain the generator state after the switch at the top of the function, and avoid having to track loop suspend counts. TurboFan still needs to have reducible loops, so we now insert loop header switches during bytecode graph building, for suspends that are discovered to be inside loops during bytecode analysis. We do, however, do some environment magic across loop headers since we know that we will continue switching if and only if we reached that loop header via a generator resume. This allows us to generate fewer phis and tighten liveness. Change-Id: Id2720ce1d6955be9a48178322cc209b3a4b8d385 Reviewed-on: https://chromium-review.googlesource.com/866734 Commit-Queue: Leszek Swirski <leszeks@chromium.org> Reviewed-by: Jaroslav Sevcik <jarin@chromium.org> Cr-Commit-Position: refs/heads/master@{#50804}
2018-01-23 13:51:38 +00:00
/* 11 E> */ B(SuspendGenerator), R(2), R(0), U8(3), U8(0),
B(ResumeGenerator), R(2), R(0), U8(3),
B(Star), R(3),
[Interpreter] Remove new.target from fixed frame slot. Removes the new.target slot from the interpreter's fixed frame. Instead adds a field to BytecodeArray to get the bytecode's incoming new.target or generator object register. The InterpreterEntryTrampoline then sets this register with the incoming new.target (or generator object) when the function is called. This register can be directly the new.target or generator object variable if they are LOCAL location, otherwise it is a temporary register which is then moved to the variable's location during the function prologue. This fixes a hack in the deoptimizer where we would set the new.target fixed slot to undefined in order to avoid extending it's lifetime through the optimized code - now it's just a standard register and can be optimized away as normal. Bug=v8:6644 Change-Id: Ieb8cc34cccefd9fb6634a90cbc77c6002a54f2ae Reviewed-on: https://chromium-review.googlesource.com/608966 Commit-Queue: Ross McIlroy <rmcilroy@chromium.org> Reviewed-by: Michael Starzinger <mstarzinger@chromium.org> Cr-Commit-Position: refs/heads/master@{#47320}
2017-08-11 15:16:31 +00:00
B(InvokeIntrinsic), U8(Runtime::k_GeneratorGetResumeMode), R(2), U8(1),
[ignition] Use jump table for generator resume mode Bug: v8:6351 Change-Id: Iad82fcfb4172d7a9cbadffa1e40fd1cd26866895 Reviewed-on: https://chromium-review.googlesource.com/521107 Commit-Queue: Leszek Swirski <leszeks@chromium.org> Reviewed-by: Ross McIlroy <rmcilroy@chromium.org> Cr-Commit-Position: refs/heads/master@{#45705}
2017-06-05 13:23:20 +00:00
B(SwitchOnSmiNoFeedback), U8(1), U8(2), I8(0),
[ignition] Single-switch generator bytecode Currently, yields and awaits inside loops compile to bytecode which switches to the top of the loop header, and switch again once inside the loop. This is to make loops reducible. This replaces this switching logic with a single switch bytecode that directly jumps to the bytecode being resumed. Among other things, this allows us to no longer maintain the generator state after the switch at the top of the function, and avoid having to track loop suspend counts. TurboFan still needs to have reducible loops, so we now insert loop header switches during bytecode graph building, for suspends that are discovered to be inside loops during bytecode analysis. We do, however, do some environment magic across loop headers since we know that we will continue switching if and only if we reached that loop header via a generator resume. This allows us to generate fewer phis and tighten liveness. Change-Id: Id2720ce1d6955be9a48178322cc209b3a4b8d385 Reviewed-on: https://chromium-review.googlesource.com/866734 Commit-Queue: Leszek Swirski <leszeks@chromium.org> Reviewed-by: Jaroslav Sevcik <jarin@chromium.org> Cr-Commit-Position: refs/heads/master@{#50804}
2018-01-23 13:51:38 +00:00
B(Ldar), R(3),
[ignition] Use jump table for generator resume mode Bug: v8:6351 Change-Id: Iad82fcfb4172d7a9cbadffa1e40fd1cd26866895 Reviewed-on: https://chromium-review.googlesource.com/521107 Commit-Queue: Leszek Swirski <leszeks@chromium.org> Reviewed-by: Ross McIlroy <rmcilroy@chromium.org> Cr-Commit-Position: refs/heads/master@{#45705}
2017-06-05 13:23:20 +00:00
/* 11 E> */ B(Throw),
[ignition] Single-switch generator bytecode Currently, yields and awaits inside loops compile to bytecode which switches to the top of the loop header, and switch again once inside the loop. This is to make loops reducible. This replaces this switching logic with a single switch bytecode that directly jumps to the bytecode being resumed. Among other things, this allows us to no longer maintain the generator state after the switch at the top of the function, and avoid having to track loop suspend counts. TurboFan still needs to have reducible loops, so we now insert loop header switches during bytecode graph building, for suspends that are discovered to be inside loops during bytecode analysis. We do, however, do some environment magic across loop headers since we know that we will continue switching if and only if we reached that loop header via a generator resume. This allows us to generate fewer phis and tighten liveness. Change-Id: Id2720ce1d6955be9a48178322cc209b3a4b8d385 Reviewed-on: https://chromium-review.googlesource.com/866734 Commit-Queue: Leszek Swirski <leszeks@chromium.org> Reviewed-by: Jaroslav Sevcik <jarin@chromium.org> Cr-Commit-Position: refs/heads/master@{#50804}
2018-01-23 13:51:38 +00:00
B(Ldar), R(3),
Move generator-close on exception from the generator function to the GeneratorResume builtin. The change also moves creation of the iterator result from the parser to the bytecode generator. Unfortunately, async generators will stay on the old scheme (try-finally around generator body) because I am not exactly sure how they work. Review-Url: https://codereview.chromium.org/2917263002 Cr-Commit-Position: refs/heads/master@{#45713}
2017-06-05 19:54:14 +00:00
/* 62 S> */ B(Return),
[parser] avoid complex for-loop desugaring when possible let/const declarations in "standard" C-style for-loops have some complex desugaring to accommodate the case where loop loop variables may be captured. This slows down the baseline performance of for-loops with let variables. This change attempts to avoid this desugaring if it's known that the loop variable is not captured at any point. A side effect of this change is that let/const loop variables, when not captured within the loop body, are not necessarily shown in the debugger, similar to other stack-allocated vars. BUG=v8:4762, v8:5460 R=marja@chromium.org, adamk@chromium.org, yangguo@chromium.org Change-Id: I8dbe545a12c086f675972bdba60c94998268311a Reviewed-on: https://chromium-review.googlesource.com/472247 Commit-Queue: Caitlin Potter <caitp@igalia.com> Reviewed-by: Yang Guo <yangguo@chromium.org> Reviewed-by: Adam Klein <adamk@chromium.org> Reviewed-by: Marja Hölttä <marja@chromium.org> Cr-Commit-Position: refs/heads/master@{#44731}
2017-04-18 14:56:06 +00:00
/* 31 S> */ B(LdaZero),
Make non-Module generators only context allocate parameters. In particular, local variables should be allocated on stack (in bytecode register), and stored/loaded to the generator object on generator suspend/resume. The CL is based on @adamk's change to scoping/parsers (https://chromium-review.googlesource.com/c/498538/), I only made the debugger cope with this change. I should note that the CL changes the scope type of suspended generators from ScopeType.Closure to ScopeType.Local. In the future we might want to introduce ScopeType.SuspendedGenerator to make the distinction explicit. Some of the changes in the tests have been made because the debugger functions do not return scopes of closed generators anymore. Generators should be allowed to throw away their internal state when they finish. BUG=v8:6368 Review-Url: https://codereview.chromium.org/2898163002 Cr-Commit-Position: refs/heads/master@{#45515}
2017-05-24 13:54:57 +00:00
B(Star), R(1),
/* 36 S> */ B(LdaSmi), I8(10),
[objects] Make feedback vector a first-class object Instead of having feedback vector as a subtype of FixedArray with reserved slots, make it a first-class variable-sized object with a fixed-size header. This allows us to compress counters to ints in the header, rather than forcing them to be Smis. Cq-Include-Trybots: master.tryserver.chromium.linux:linux_chromium_rel_ng Change-Id: Icc5f088ffbc2e2651b845bc71ea42060639e3e48 Reviewed-on: https://chromium-review.googlesource.com/585129 Commit-Queue: Leszek Swirski <leszeks@chromium.org> Reviewed-by: Michael Stanton <mvstanton@chromium.org> Reviewed-by: Ross McIlroy <rmcilroy@chromium.org> Reviewed-by: Michael Lippautz <mlippautz@chromium.org> Cr-Commit-Position: refs/heads/master@{#46935}
2017-07-27 12:45:00 +00:00
/* 36 E> */ B(TestLessThan), R(1), U8(0),
[parser] avoid for-loop desugaring unless loop var may be captured In https://chromium-review.googlesource.com/c/472247/, I avoided running DesugarLexicalBindingsInForStatement() if there were no lexical loop variables, the function was not resumable, and the variables are not captured by eval or a function declaration. I think it's now possible to limit this further, and only do the more extensive desugaring if there's a function declaration / eval() call in the loop body. `yield` and `await` are not an issue as those loop variables are written to the register file and not lost. This change just removes the `is_resumable()` condition. If it passes tests, I think it's safe. BUG=v8:4762, v8:5460, v8:6579 Change-Id: I92d0308ad9401c1338411bc9ae9021f978803d3a Reviewed-on: https://chromium-review.googlesource.com/563587 Commit-Queue: Caitlin Potter <caitp@igalia.com> Reviewed-by: Georg Neis <neis@chromium.org> Reviewed-by: Adam Klein <adamk@chromium.org> Cr-Commit-Position: refs/heads/master@{#46536}
2017-07-07 20:48:16 +00:00
B(JumpIfFalse), U8(15),
[parser] avoid complex for-loop desugaring when possible let/const declarations in "standard" C-style for-loops have some complex desugaring to accommodate the case where loop loop variables may be captured. This slows down the baseline performance of for-loops with let variables. This change attempts to avoid this desugaring if it's known that the loop variable is not captured at any point. A side effect of this change is that let/const loop variables, when not captured within the loop body, are not necessarily shown in the debugger, similar to other stack-allocated vars. BUG=v8:4762, v8:5460 R=marja@chromium.org, adamk@chromium.org, yangguo@chromium.org Change-Id: I8dbe545a12c086f675972bdba60c94998268311a Reviewed-on: https://chromium-review.googlesource.com/472247 Commit-Queue: Caitlin Potter <caitp@igalia.com> Reviewed-by: Yang Guo <yangguo@chromium.org> Reviewed-by: Adam Klein <adamk@chromium.org> Reviewed-by: Marja Hölttä <marja@chromium.org> Cr-Commit-Position: refs/heads/master@{#44731}
2017-04-18 14:56:06 +00:00
/* 18 E> */ B(StackCheck),
Make non-Module generators only context allocate parameters. In particular, local variables should be allocated on stack (in bytecode register), and stored/loaded to the generator object on generator suspend/resume. The CL is based on @adamk's change to scoping/parsers (https://chromium-review.googlesource.com/c/498538/), I only made the debugger cope with this change. I should note that the CL changes the scope type of suspended generators from ScopeType.Closure to ScopeType.Local. In the future we might want to introduce ScopeType.SuspendedGenerator to make the distinction explicit. Some of the changes in the tests have been made because the debugger functions do not return scopes of closed generators anymore. Generators should be allowed to throw away their internal state when they finish. BUG=v8:6368 Review-Url: https://codereview.chromium.org/2898163002 Cr-Commit-Position: refs/heads/master@{#45515}
2017-05-24 13:54:57 +00:00
/* 57 S> */ B(Mov), R(1), R(0),
[parser] avoid for-loop desugaring unless loop var may be captured In https://chromium-review.googlesource.com/c/472247/, I avoided running DesugarLexicalBindingsInForStatement() if there were no lexical loop variables, the function was not resumable, and the variables are not captured by eval or a function declaration. I think it's now possible to limit this further, and only do the more extensive desugaring if there's a function declaration / eval() call in the loop body. `yield` and `await` are not an issue as those loop variables are written to the register file and not lost. This change just removes the `is_resumable()` condition. If it passes tests, I think it's safe. BUG=v8:4762, v8:5460, v8:6579 Change-Id: I92d0308ad9401c1338411bc9ae9021f978803d3a Reviewed-on: https://chromium-review.googlesource.com/563587 Commit-Queue: Caitlin Potter <caitp@igalia.com> Reviewed-by: Georg Neis <neis@chromium.org> Reviewed-by: Adam Klein <adamk@chromium.org> Cr-Commit-Position: refs/heads/master@{#46536}
2017-07-07 20:48:16 +00:00
/* 44 S> */ B(Ldar), R(0),
[objects] Make feedback vector a first-class object Instead of having feedback vector as a subtype of FixedArray with reserved slots, make it a first-class variable-sized object with a fixed-size header. This allows us to compress counters to ints in the header, rather than forcing them to be Smis. Cq-Include-Trybots: master.tryserver.chromium.linux:linux_chromium_rel_ng Change-Id: Icc5f088ffbc2e2651b845bc71ea42060639e3e48 Reviewed-on: https://chromium-review.googlesource.com/585129 Commit-Queue: Leszek Swirski <leszeks@chromium.org> Reviewed-by: Michael Stanton <mvstanton@chromium.org> Reviewed-by: Ross McIlroy <rmcilroy@chromium.org> Reviewed-by: Michael Lippautz <mlippautz@chromium.org> Cr-Commit-Position: refs/heads/master@{#46935}
2017-07-27 12:45:00 +00:00
B(Inc), U8(1),
[parser] avoid for-loop desugaring unless loop var may be captured In https://chromium-review.googlesource.com/c/472247/, I avoided running DesugarLexicalBindingsInForStatement() if there were no lexical loop variables, the function was not resumable, and the variables are not captured by eval or a function declaration. I think it's now possible to limit this further, and only do the more extensive desugaring if there's a function declaration / eval() call in the loop body. `yield` and `await` are not an issue as those loop variables are written to the register file and not lost. This change just removes the `is_resumable()` condition. If it passes tests, I think it's safe. BUG=v8:4762, v8:5460, v8:6579 Change-Id: I92d0308ad9401c1338411bc9ae9021f978803d3a Reviewed-on: https://chromium-review.googlesource.com/563587 Commit-Queue: Caitlin Potter <caitp@igalia.com> Reviewed-by: Georg Neis <neis@chromium.org> Reviewed-by: Adam Klein <adamk@chromium.org> Cr-Commit-Position: refs/heads/master@{#46536}
2017-07-07 20:48:16 +00:00
B(Star), R(1),
B(JumpLoop), U8(17), I8(0),
[parser] avoid complex for-loop desugaring when possible let/const declarations in "standard" C-style for-loops have some complex desugaring to accommodate the case where loop loop variables may be captured. This slows down the baseline performance of for-loops with let variables. This change attempts to avoid this desugaring if it's known that the loop variable is not captured at any point. A side effect of this change is that let/const loop variables, when not captured within the loop body, are not necessarily shown in the debugger, similar to other stack-allocated vars. BUG=v8:4762, v8:5460 R=marja@chromium.org, adamk@chromium.org, yangguo@chromium.org Change-Id: I8dbe545a12c086f675972bdba60c94998268311a Reviewed-on: https://chromium-review.googlesource.com/472247 Commit-Queue: Caitlin Potter <caitp@igalia.com> Reviewed-by: Yang Guo <yangguo@chromium.org> Reviewed-by: Adam Klein <adamk@chromium.org> Reviewed-by: Marja Hölttä <marja@chromium.org> Cr-Commit-Position: refs/heads/master@{#44731}
2017-04-18 14:56:06 +00:00
B(LdaUndefined),
/* 62 S> */ B(Return),
]
constant pool: [
[ignition] Single-switch generator bytecode Currently, yields and awaits inside loops compile to bytecode which switches to the top of the loop header, and switch again once inside the loop. This is to make loops reducible. This replaces this switching logic with a single switch bytecode that directly jumps to the bytecode being resumed. Among other things, this allows us to no longer maintain the generator state after the switch at the top of the function, and avoid having to track loop suspend counts. TurboFan still needs to have reducible loops, so we now insert loop header switches during bytecode graph building, for suspends that are discovered to be inside loops during bytecode analysis. We do, however, do some environment magic across loop headers since we know that we will continue switching if and only if we reached that loop header via a generator resume. This allows us to generate fewer phis and tighten liveness. Change-Id: Id2720ce1d6955be9a48178322cc209b3a4b8d385 Reviewed-on: https://chromium-review.googlesource.com/866734 Commit-Queue: Leszek Swirski <leszeks@chromium.org> Reviewed-by: Jaroslav Sevcik <jarin@chromium.org> Cr-Commit-Position: refs/heads/master@{#50804}
2018-01-23 13:51:38 +00:00
Smi [22],
Move closing of generators upon final return to the generator-resume builtin. Review-Url: https://codereview.chromium.org/2936813002 Cr-Commit-Position: refs/heads/master@{#45906}
2017-06-13 13:49:45 +00:00
Smi [10],
[ignition] Use jump table for generator resume mode Bug: v8:6351 Change-Id: Iad82fcfb4172d7a9cbadffa1e40fd1cd26866895 Reviewed-on: https://chromium-review.googlesource.com/521107 Commit-Queue: Leszek Swirski <leszeks@chromium.org> Reviewed-by: Ross McIlroy <rmcilroy@chromium.org> Cr-Commit-Position: refs/heads/master@{#45705}
2017-06-05 13:23:20 +00:00
Smi [7],
[parser] avoid complex for-loop desugaring when possible let/const declarations in "standard" C-style for-loops have some complex desugaring to accommodate the case where loop loop variables may be captured. This slows down the baseline performance of for-loops with let variables. This change attempts to avoid this desugaring if it's known that the loop variable is not captured at any point. A side effect of this change is that let/const loop variables, when not captured within the loop body, are not necessarily shown in the debugger, similar to other stack-allocated vars. BUG=v8:4762, v8:5460 R=marja@chromium.org, adamk@chromium.org, yangguo@chromium.org Change-Id: I8dbe545a12c086f675972bdba60c94998268311a Reviewed-on: https://chromium-review.googlesource.com/472247 Commit-Queue: Caitlin Potter <caitp@igalia.com> Reviewed-by: Yang Guo <yangguo@chromium.org> Reviewed-by: Adam Klein <adamk@chromium.org> Reviewed-by: Marja Hölttä <marja@chromium.org> Cr-Commit-Position: refs/heads/master@{#44731}
2017-04-18 14:56:06 +00:00
]
handlers: [
]
---
snippet: "
function* f() {
for (let x = 0; x < 10; ++x) yield x;
}
f();
"
[ignition] Single-switch generator bytecode Currently, yields and awaits inside loops compile to bytecode which switches to the top of the loop header, and switch again once inside the loop. This is to make loops reducible. This replaces this switching logic with a single switch bytecode that directly jumps to the bytecode being resumed. Among other things, this allows us to no longer maintain the generator state after the switch at the top of the function, and avoid having to track loop suspend counts. TurboFan still needs to have reducible loops, so we now insert loop header switches during bytecode graph building, for suspends that are discovered to be inside loops during bytecode analysis. We do, however, do some environment magic across loop headers since we know that we will continue switching if and only if we reached that loop header via a generator resume. This allows us to generate fewer phis and tighten liveness. Change-Id: Id2720ce1d6955be9a48178322cc209b3a4b8d385 Reviewed-on: https://chromium-review.googlesource.com/866734 Commit-Queue: Leszek Swirski <leszeks@chromium.org> Reviewed-by: Jaroslav Sevcik <jarin@chromium.org> Cr-Commit-Position: refs/heads/master@{#50804}
2018-01-23 13:51:38 +00:00
frame size: 4
[parser] avoid complex for-loop desugaring when possible let/const declarations in "standard" C-style for-loops have some complex desugaring to accommodate the case where loop loop variables may be captured. This slows down the baseline performance of for-loops with let variables. This change attempts to avoid this desugaring if it's known that the loop variable is not captured at any point. A side effect of this change is that let/const loop variables, when not captured within the loop body, are not necessarily shown in the debugger, similar to other stack-allocated vars. BUG=v8:4762, v8:5460 R=marja@chromium.org, adamk@chromium.org, yangguo@chromium.org Change-Id: I8dbe545a12c086f675972bdba60c94998268311a Reviewed-on: https://chromium-review.googlesource.com/472247 Commit-Queue: Caitlin Potter <caitp@igalia.com> Reviewed-by: Yang Guo <yangguo@chromium.org> Reviewed-by: Adam Klein <adamk@chromium.org> Reviewed-by: Marja Hölttä <marja@chromium.org> Cr-Commit-Position: refs/heads/master@{#44731}
2017-04-18 14:56:06 +00:00
parameter count: 1
[ignition] Reduce generator switch boilerplate The SwitchOnGeneratorState bytecode now also falls through if the generator object is undefined (so that we don't need that jump) and restores generator context (so that we don't need that PushContext). This saves 10 bytes per generator. Change-Id: Ie0872c827119b9f1d1e9244d3be6496a30cd9620 Reviewed-on: https://chromium-review.googlesource.com/867051 Commit-Queue: Leszek Swirski <leszeks@chromium.org> Reviewed-by: Ross McIlroy <rmcilroy@chromium.org> Reviewed-by: Georg Neis <neis@chromium.org> Cr-Commit-Position: refs/heads/master@{#50845}
2018-01-24 11:19:51 +00:00
bytecode array length: 99
[parser] avoid complex for-loop desugaring when possible let/const declarations in "standard" C-style for-loops have some complex desugaring to accommodate the case where loop loop variables may be captured. This slows down the baseline performance of for-loops with let variables. This change attempts to avoid this desugaring if it's known that the loop variable is not captured at any point. A side effect of this change is that let/const loop variables, when not captured within the loop body, are not necessarily shown in the debugger, similar to other stack-allocated vars. BUG=v8:4762, v8:5460 R=marja@chromium.org, adamk@chromium.org, yangguo@chromium.org Change-Id: I8dbe545a12c086f675972bdba60c94998268311a Reviewed-on: https://chromium-review.googlesource.com/472247 Commit-Queue: Caitlin Potter <caitp@igalia.com> Reviewed-by: Yang Guo <yangguo@chromium.org> Reviewed-by: Adam Klein <adamk@chromium.org> Reviewed-by: Marja Hölttä <marja@chromium.org> Cr-Commit-Position: refs/heads/master@{#44731}
2017-04-18 14:56:06 +00:00
bytecodes: [
[ignition] Single-switch generator bytecode Currently, yields and awaits inside loops compile to bytecode which switches to the top of the loop header, and switch again once inside the loop. This is to make loops reducible. This replaces this switching logic with a single switch bytecode that directly jumps to the bytecode being resumed. Among other things, this allows us to no longer maintain the generator state after the switch at the top of the function, and avoid having to track loop suspend counts. TurboFan still needs to have reducible loops, so we now insert loop header switches during bytecode graph building, for suspends that are discovered to be inside loops during bytecode analysis. We do, however, do some environment magic across loop headers since we know that we will continue switching if and only if we reached that loop header via a generator resume. This allows us to generate fewer phis and tighten liveness. Change-Id: Id2720ce1d6955be9a48178322cc209b3a4b8d385 Reviewed-on: https://chromium-review.googlesource.com/866734 Commit-Queue: Leszek Swirski <leszeks@chromium.org> Reviewed-by: Jaroslav Sevcik <jarin@chromium.org> Cr-Commit-Position: refs/heads/master@{#50804}
2018-01-23 13:51:38 +00:00
B(SwitchOnGeneratorState), R(1), U8(0), U8(2),
B(Mov), R(closure), R(2),
B(Mov), R(this), R(3),
B(InvokeIntrinsic), U8(Runtime::k_CreateJSGeneratorObject), R(2), U8(2),
[parser] avoid for-loop desugaring unless loop var may be captured In https://chromium-review.googlesource.com/c/472247/, I avoided running DesugarLexicalBindingsInForStatement() if there were no lexical loop variables, the function was not resumable, and the variables are not captured by eval or a function declaration. I think it's now possible to limit this further, and only do the more extensive desugaring if there's a function declaration / eval() call in the loop body. `yield` and `await` are not an issue as those loop variables are written to the register file and not lost. This change just removes the `is_resumable()` condition. If it passes tests, I think it's safe. BUG=v8:4762, v8:5460, v8:6579 Change-Id: I92d0308ad9401c1338411bc9ae9021f978803d3a Reviewed-on: https://chromium-review.googlesource.com/563587 Commit-Queue: Caitlin Potter <caitp@igalia.com> Reviewed-by: Georg Neis <neis@chromium.org> Reviewed-by: Adam Klein <adamk@chromium.org> Cr-Commit-Position: refs/heads/master@{#46536}
2017-07-07 20:48:16 +00:00
B(Star), R(1),
[parser] avoid complex for-loop desugaring when possible let/const declarations in "standard" C-style for-loops have some complex desugaring to accommodate the case where loop loop variables may be captured. This slows down the baseline performance of for-loops with let variables. This change attempts to avoid this desugaring if it's known that the loop variable is not captured at any point. A side effect of this change is that let/const loop variables, when not captured within the loop body, are not necessarily shown in the debugger, similar to other stack-allocated vars. BUG=v8:4762, v8:5460 R=marja@chromium.org, adamk@chromium.org, yangguo@chromium.org Change-Id: I8dbe545a12c086f675972bdba60c94998268311a Reviewed-on: https://chromium-review.googlesource.com/472247 Commit-Queue: Caitlin Potter <caitp@igalia.com> Reviewed-by: Yang Guo <yangguo@chromium.org> Reviewed-by: Adam Klein <adamk@chromium.org> Reviewed-by: Marja Hölttä <marja@chromium.org> Cr-Commit-Position: refs/heads/master@{#44731}
2017-04-18 14:56:06 +00:00
/* 11 E> */ B(StackCheck),
[ignition] Single-switch generator bytecode Currently, yields and awaits inside loops compile to bytecode which switches to the top of the loop header, and switch again once inside the loop. This is to make loops reducible. This replaces this switching logic with a single switch bytecode that directly jumps to the bytecode being resumed. Among other things, this allows us to no longer maintain the generator state after the switch at the top of the function, and avoid having to track loop suspend counts. TurboFan still needs to have reducible loops, so we now insert loop header switches during bytecode graph building, for suspends that are discovered to be inside loops during bytecode analysis. We do, however, do some environment magic across loop headers since we know that we will continue switching if and only if we reached that loop header via a generator resume. This allows us to generate fewer phis and tighten liveness. Change-Id: Id2720ce1d6955be9a48178322cc209b3a4b8d385 Reviewed-on: https://chromium-review.googlesource.com/866734 Commit-Queue: Leszek Swirski <leszeks@chromium.org> Reviewed-by: Jaroslav Sevcik <jarin@chromium.org> Cr-Commit-Position: refs/heads/master@{#50804}
2018-01-23 13:51:38 +00:00
/* 11 E> */ B(SuspendGenerator), R(1), R(0), U8(2), U8(0),
B(ResumeGenerator), R(1), R(0), U8(2),
B(Star), R(2),
[Interpreter] Remove new.target from fixed frame slot. Removes the new.target slot from the interpreter's fixed frame. Instead adds a field to BytecodeArray to get the bytecode's incoming new.target or generator object register. The InterpreterEntryTrampoline then sets this register with the incoming new.target (or generator object) when the function is called. This register can be directly the new.target or generator object variable if they are LOCAL location, otherwise it is a temporary register which is then moved to the variable's location during the function prologue. This fixes a hack in the deoptimizer where we would set the new.target fixed slot to undefined in order to avoid extending it's lifetime through the optimized code - now it's just a standard register and can be optimized away as normal. Bug=v8:6644 Change-Id: Ieb8cc34cccefd9fb6634a90cbc77c6002a54f2ae Reviewed-on: https://chromium-review.googlesource.com/608966 Commit-Queue: Ross McIlroy <rmcilroy@chromium.org> Reviewed-by: Michael Starzinger <mstarzinger@chromium.org> Cr-Commit-Position: refs/heads/master@{#47320}
2017-08-11 15:16:31 +00:00
B(InvokeIntrinsic), U8(Runtime::k_GeneratorGetResumeMode), R(1), U8(1),
[ignition] Use jump table for generator resume mode Bug: v8:6351 Change-Id: Iad82fcfb4172d7a9cbadffa1e40fd1cd26866895 Reviewed-on: https://chromium-review.googlesource.com/521107 Commit-Queue: Leszek Swirski <leszeks@chromium.org> Reviewed-by: Ross McIlroy <rmcilroy@chromium.org> Cr-Commit-Position: refs/heads/master@{#45705}
2017-06-05 13:23:20 +00:00
B(SwitchOnSmiNoFeedback), U8(2), U8(2), I8(0),
[ignition] Single-switch generator bytecode Currently, yields and awaits inside loops compile to bytecode which switches to the top of the loop header, and switch again once inside the loop. This is to make loops reducible. This replaces this switching logic with a single switch bytecode that directly jumps to the bytecode being resumed. Among other things, this allows us to no longer maintain the generator state after the switch at the top of the function, and avoid having to track loop suspend counts. TurboFan still needs to have reducible loops, so we now insert loop header switches during bytecode graph building, for suspends that are discovered to be inside loops during bytecode analysis. We do, however, do some environment magic across loop headers since we know that we will continue switching if and only if we reached that loop header via a generator resume. This allows us to generate fewer phis and tighten liveness. Change-Id: Id2720ce1d6955be9a48178322cc209b3a4b8d385 Reviewed-on: https://chromium-review.googlesource.com/866734 Commit-Queue: Leszek Swirski <leszeks@chromium.org> Reviewed-by: Jaroslav Sevcik <jarin@chromium.org> Cr-Commit-Position: refs/heads/master@{#50804}
2018-01-23 13:51:38 +00:00
B(Ldar), R(2),
[ignition] Use jump table for generator resume mode Bug: v8:6351 Change-Id: Iad82fcfb4172d7a9cbadffa1e40fd1cd26866895 Reviewed-on: https://chromium-review.googlesource.com/521107 Commit-Queue: Leszek Swirski <leszeks@chromium.org> Reviewed-by: Ross McIlroy <rmcilroy@chromium.org> Cr-Commit-Position: refs/heads/master@{#45705}
2017-06-05 13:23:20 +00:00
/* 11 E> */ B(Throw),
[ignition] Single-switch generator bytecode Currently, yields and awaits inside loops compile to bytecode which switches to the top of the loop header, and switch again once inside the loop. This is to make loops reducible. This replaces this switching logic with a single switch bytecode that directly jumps to the bytecode being resumed. Among other things, this allows us to no longer maintain the generator state after the switch at the top of the function, and avoid having to track loop suspend counts. TurboFan still needs to have reducible loops, so we now insert loop header switches during bytecode graph building, for suspends that are discovered to be inside loops during bytecode analysis. We do, however, do some environment magic across loop headers since we know that we will continue switching if and only if we reached that loop header via a generator resume. This allows us to generate fewer phis and tighten liveness. Change-Id: Id2720ce1d6955be9a48178322cc209b3a4b8d385 Reviewed-on: https://chromium-review.googlesource.com/866734 Commit-Queue: Leszek Swirski <leszeks@chromium.org> Reviewed-by: Jaroslav Sevcik <jarin@chromium.org> Cr-Commit-Position: refs/heads/master@{#50804}
2018-01-23 13:51:38 +00:00
B(Ldar), R(2),
Move generator-close on exception from the generator function to the GeneratorResume builtin. The change also moves creation of the iterator result from the parser to the bytecode generator. Unfortunately, async generators will stay on the old scheme (try-finally around generator body) because I am not exactly sure how they work. Review-Url: https://codereview.chromium.org/2917263002 Cr-Commit-Position: refs/heads/master@{#45713}
2017-06-05 19:54:14 +00:00
/* 56 S> */ B(Return),
[parser] avoid complex for-loop desugaring when possible let/const declarations in "standard" C-style for-loops have some complex desugaring to accommodate the case where loop loop variables may be captured. This slows down the baseline performance of for-loops with let variables. This change attempts to avoid this desugaring if it's known that the loop variable is not captured at any point. A side effect of this change is that let/const loop variables, when not captured within the loop body, are not necessarily shown in the debugger, similar to other stack-allocated vars. BUG=v8:4762, v8:5460 R=marja@chromium.org, adamk@chromium.org, yangguo@chromium.org Change-Id: I8dbe545a12c086f675972bdba60c94998268311a Reviewed-on: https://chromium-review.googlesource.com/472247 Commit-Queue: Caitlin Potter <caitp@igalia.com> Reviewed-by: Yang Guo <yangguo@chromium.org> Reviewed-by: Adam Klein <adamk@chromium.org> Reviewed-by: Marja Hölttä <marja@chromium.org> Cr-Commit-Position: refs/heads/master@{#44731}
2017-04-18 14:56:06 +00:00
/* 31 S> */ B(LdaZero),
[parser] avoid for-loop desugaring unless loop var may be captured In https://chromium-review.googlesource.com/c/472247/, I avoided running DesugarLexicalBindingsInForStatement() if there were no lexical loop variables, the function was not resumable, and the variables are not captured by eval or a function declaration. I think it's now possible to limit this further, and only do the more extensive desugaring if there's a function declaration / eval() call in the loop body. `yield` and `await` are not an issue as those loop variables are written to the register file and not lost. This change just removes the `is_resumable()` condition. If it passes tests, I think it's safe. BUG=v8:4762, v8:5460, v8:6579 Change-Id: I92d0308ad9401c1338411bc9ae9021f978803d3a Reviewed-on: https://chromium-review.googlesource.com/563587 Commit-Queue: Caitlin Potter <caitp@igalia.com> Reviewed-by: Georg Neis <neis@chromium.org> Reviewed-by: Adam Klein <adamk@chromium.org> Cr-Commit-Position: refs/heads/master@{#46536}
2017-07-07 20:48:16 +00:00
B(Star), R(0),
Make non-Module generators only context allocate parameters. In particular, local variables should be allocated on stack (in bytecode register), and stored/loaded to the generator object on generator suspend/resume. The CL is based on @adamk's change to scoping/parsers (https://chromium-review.googlesource.com/c/498538/), I only made the debugger cope with this change. I should note that the CL changes the scope type of suspended generators from ScopeType.Closure to ScopeType.Local. In the future we might want to introduce ScopeType.SuspendedGenerator to make the distinction explicit. Some of the changes in the tests have been made because the debugger functions do not return scopes of closed generators anymore. Generators should be allowed to throw away their internal state when they finish. BUG=v8:6368 Review-Url: https://codereview.chromium.org/2898163002 Cr-Commit-Position: refs/heads/master@{#45515}
2017-05-24 13:54:57 +00:00
/* 36 S> */ B(LdaSmi), I8(10),
[objects] Make feedback vector a first-class object Instead of having feedback vector as a subtype of FixedArray with reserved slots, make it a first-class variable-sized object with a fixed-size header. This allows us to compress counters to ints in the header, rather than forcing them to be Smis. Cq-Include-Trybots: master.tryserver.chromium.linux:linux_chromium_rel_ng Change-Id: Icc5f088ffbc2e2651b845bc71ea42060639e3e48 Reviewed-on: https://chromium-review.googlesource.com/585129 Commit-Queue: Leszek Swirski <leszeks@chromium.org> Reviewed-by: Michael Stanton <mvstanton@chromium.org> Reviewed-by: Ross McIlroy <rmcilroy@chromium.org> Reviewed-by: Michael Lippautz <mlippautz@chromium.org> Cr-Commit-Position: refs/heads/master@{#46935}
2017-07-27 12:45:00 +00:00
/* 36 E> */ B(TestLessThan), R(0), U8(0),
[ignition] Single-switch generator bytecode Currently, yields and awaits inside loops compile to bytecode which switches to the top of the loop header, and switch again once inside the loop. This is to make loops reducible. This replaces this switching logic with a single switch bytecode that directly jumps to the bytecode being resumed. Among other things, this allows us to no longer maintain the generator state after the switch at the top of the function, and avoid having to track loop suspend counts. TurboFan still needs to have reducible loops, so we now insert loop header switches during bytecode graph building, for suspends that are discovered to be inside loops during bytecode analysis. We do, however, do some environment magic across loop headers since we know that we will continue switching if and only if we reached that loop header via a generator resume. This allows us to generate fewer phis and tighten liveness. Change-Id: Id2720ce1d6955be9a48178322cc209b3a4b8d385 Reviewed-on: https://chromium-review.googlesource.com/866734 Commit-Queue: Leszek Swirski <leszeks@chromium.org> Reviewed-by: Jaroslav Sevcik <jarin@chromium.org> Cr-Commit-Position: refs/heads/master@{#50804}
2018-01-23 13:51:38 +00:00
B(JumpIfFalse), U8(47),
[parser] avoid complex for-loop desugaring when possible let/const declarations in "standard" C-style for-loops have some complex desugaring to accommodate the case where loop loop variables may be captured. This slows down the baseline performance of for-loops with let variables. This change attempts to avoid this desugaring if it's known that the loop variable is not captured at any point. A side effect of this change is that let/const loop variables, when not captured within the loop body, are not necessarily shown in the debugger, similar to other stack-allocated vars. BUG=v8:4762, v8:5460 R=marja@chromium.org, adamk@chromium.org, yangguo@chromium.org Change-Id: I8dbe545a12c086f675972bdba60c94998268311a Reviewed-on: https://chromium-review.googlesource.com/472247 Commit-Queue: Caitlin Potter <caitp@igalia.com> Reviewed-by: Yang Guo <yangguo@chromium.org> Reviewed-by: Adam Klein <adamk@chromium.org> Reviewed-by: Marja Hölttä <marja@chromium.org> Cr-Commit-Position: refs/heads/master@{#44731}
2017-04-18 14:56:06 +00:00
/* 18 E> */ B(StackCheck),
[interpreter] refactor BuildGeneratorSuspend/Resume into BuildSuspendPoint Simplify the model for generating Awaits, because the resume point is always immediately following the suspend point, and registers used are always the same for both operations. Includes a minor refactoring of BytecodeGenerator::VisitYield() to perform iterator result creation before the SuspendGenerator bytecode, rather than between SuspendGenerator and Return. This adds a small number of bytecodes for each yield. BUG=v8:2355, v8:5855 Change-Id: I4868b89a6bc1b251f887d2a45890c8fa19f7b089 Reviewed-on: https://chromium-review.googlesource.com/576286 Reviewed-by: Ross McIlroy <rmcilroy@chromium.org> Reviewed-by: Georg Neis <neis@chromium.org> Commit-Queue: Caitlin Potter <caitp@igalia.com> Cr-Commit-Position: refs/heads/master@{#46820}
2017-07-20 12:35:28 +00:00
/* 47 S> */ B(LdaFalse),
[parser] avoid for-loop desugaring unless loop var may be captured In https://chromium-review.googlesource.com/c/472247/, I avoided running DesugarLexicalBindingsInForStatement() if there were no lexical loop variables, the function was not resumable, and the variables are not captured by eval or a function declaration. I think it's now possible to limit this further, and only do the more extensive desugaring if there's a function declaration / eval() call in the loop body. `yield` and `await` are not an issue as those loop variables are written to the register file and not lost. This change just removes the `is_resumable()` condition. If it passes tests, I think it's safe. BUG=v8:4762, v8:5460, v8:6579 Change-Id: I92d0308ad9401c1338411bc9ae9021f978803d3a Reviewed-on: https://chromium-review.googlesource.com/563587 Commit-Queue: Caitlin Potter <caitp@igalia.com> Reviewed-by: Georg Neis <neis@chromium.org> Reviewed-by: Adam Klein <adamk@chromium.org> Cr-Commit-Position: refs/heads/master@{#46536}
2017-07-07 20:48:16 +00:00
B(Star), R(3),
[ignition] Single-switch generator bytecode Currently, yields and awaits inside loops compile to bytecode which switches to the top of the loop header, and switch again once inside the loop. This is to make loops reducible. This replaces this switching logic with a single switch bytecode that directly jumps to the bytecode being resumed. Among other things, this allows us to no longer maintain the generator state after the switch at the top of the function, and avoid having to track loop suspend counts. TurboFan still needs to have reducible loops, so we now insert loop header switches during bytecode graph building, for suspends that are discovered to be inside loops during bytecode analysis. We do, however, do some environment magic across loop headers since we know that we will continue switching if and only if we reached that loop header via a generator resume. This allows us to generate fewer phis and tighten liveness. Change-Id: Id2720ce1d6955be9a48178322cc209b3a4b8d385 Reviewed-on: https://chromium-review.googlesource.com/866734 Commit-Queue: Leszek Swirski <leszeks@chromium.org> Reviewed-by: Jaroslav Sevcik <jarin@chromium.org> Cr-Commit-Position: refs/heads/master@{#50804}
2018-01-23 13:51:38 +00:00
B(Mov), R(0), R(2),
B(InvokeIntrinsic), U8(Runtime::k_CreateIterResultObject), R(2), U8(2),
/* 47 E> */ B(SuspendGenerator), R(1), R(0), U8(2), U8(1),
B(ResumeGenerator), R(1), R(0), U8(2),
B(Star), R(2),
[Interpreter] Remove new.target from fixed frame slot. Removes the new.target slot from the interpreter's fixed frame. Instead adds a field to BytecodeArray to get the bytecode's incoming new.target or generator object register. The InterpreterEntryTrampoline then sets this register with the incoming new.target (or generator object) when the function is called. This register can be directly the new.target or generator object variable if they are LOCAL location, otherwise it is a temporary register which is then moved to the variable's location during the function prologue. This fixes a hack in the deoptimizer where we would set the new.target fixed slot to undefined in order to avoid extending it's lifetime through the optimized code - now it's just a standard register and can be optimized away as normal. Bug=v8:6644 Change-Id: Ieb8cc34cccefd9fb6634a90cbc77c6002a54f2ae Reviewed-on: https://chromium-review.googlesource.com/608966 Commit-Queue: Ross McIlroy <rmcilroy@chromium.org> Reviewed-by: Michael Starzinger <mstarzinger@chromium.org> Cr-Commit-Position: refs/heads/master@{#47320}
2017-08-11 15:16:31 +00:00
B(InvokeIntrinsic), U8(Runtime::k_GeneratorGetResumeMode), R(1), U8(1),
[ignition] Single-switch generator bytecode Currently, yields and awaits inside loops compile to bytecode which switches to the top of the loop header, and switch again once inside the loop. This is to make loops reducible. This replaces this switching logic with a single switch bytecode that directly jumps to the bytecode being resumed. Among other things, this allows us to no longer maintain the generator state after the switch at the top of the function, and avoid having to track loop suspend counts. TurboFan still needs to have reducible loops, so we now insert loop header switches during bytecode graph building, for suspends that are discovered to be inside loops during bytecode analysis. We do, however, do some environment magic across loop headers since we know that we will continue switching if and only if we reached that loop header via a generator resume. This allows us to generate fewer phis and tighten liveness. Change-Id: Id2720ce1d6955be9a48178322cc209b3a4b8d385 Reviewed-on: https://chromium-review.googlesource.com/866734 Commit-Queue: Leszek Swirski <leszeks@chromium.org> Reviewed-by: Jaroslav Sevcik <jarin@chromium.org> Cr-Commit-Position: refs/heads/master@{#50804}
2018-01-23 13:51:38 +00:00
B(SwitchOnSmiNoFeedback), U8(4), U8(2), I8(0),
B(Ldar), R(2),
[ignition] Use jump table for generator resume mode Bug: v8:6351 Change-Id: Iad82fcfb4172d7a9cbadffa1e40fd1cd26866895 Reviewed-on: https://chromium-review.googlesource.com/521107 Commit-Queue: Leszek Swirski <leszeks@chromium.org> Reviewed-by: Ross McIlroy <rmcilroy@chromium.org> Cr-Commit-Position: refs/heads/master@{#45705}
2017-06-05 13:23:20 +00:00
/* 47 E> */ B(Throw),
[ignition] Single-switch generator bytecode Currently, yields and awaits inside loops compile to bytecode which switches to the top of the loop header, and switch again once inside the loop. This is to make loops reducible. This replaces this switching logic with a single switch bytecode that directly jumps to the bytecode being resumed. Among other things, this allows us to no longer maintain the generator state after the switch at the top of the function, and avoid having to track loop suspend counts. TurboFan still needs to have reducible loops, so we now insert loop header switches during bytecode graph building, for suspends that are discovered to be inside loops during bytecode analysis. We do, however, do some environment magic across loop headers since we know that we will continue switching if and only if we reached that loop header via a generator resume. This allows us to generate fewer phis and tighten liveness. Change-Id: Id2720ce1d6955be9a48178322cc209b3a4b8d385 Reviewed-on: https://chromium-review.googlesource.com/866734 Commit-Queue: Leszek Swirski <leszeks@chromium.org> Reviewed-by: Jaroslav Sevcik <jarin@chromium.org> Cr-Commit-Position: refs/heads/master@{#50804}
2018-01-23 13:51:38 +00:00
B(Ldar), R(2),
Move generator-close on exception from the generator function to the GeneratorResume builtin. The change also moves creation of the iterator result from the parser to the bytecode generator. Unfortunately, async generators will stay on the old scheme (try-finally around generator body) because I am not exactly sure how they work. Review-Url: https://codereview.chromium.org/2917263002 Cr-Commit-Position: refs/heads/master@{#45713}
2017-06-05 19:54:14 +00:00
/* 56 S> */ B(Return),
[parser] avoid for-loop desugaring unless loop var may be captured In https://chromium-review.googlesource.com/c/472247/, I avoided running DesugarLexicalBindingsInForStatement() if there were no lexical loop variables, the function was not resumable, and the variables are not captured by eval or a function declaration. I think it's now possible to limit this further, and only do the more extensive desugaring if there's a function declaration / eval() call in the loop body. `yield` and `await` are not an issue as those loop variables are written to the register file and not lost. This change just removes the `is_resumable()` condition. If it passes tests, I think it's safe. BUG=v8:4762, v8:5460, v8:6579 Change-Id: I92d0308ad9401c1338411bc9ae9021f978803d3a Reviewed-on: https://chromium-review.googlesource.com/563587 Commit-Queue: Caitlin Potter <caitp@igalia.com> Reviewed-by: Georg Neis <neis@chromium.org> Reviewed-by: Adam Klein <adamk@chromium.org> Cr-Commit-Position: refs/heads/master@{#46536}
2017-07-07 20:48:16 +00:00
/* 44 S> */ B(Ldar), R(0),
[objects] Make feedback vector a first-class object Instead of having feedback vector as a subtype of FixedArray with reserved slots, make it a first-class variable-sized object with a fixed-size header. This allows us to compress counters to ints in the header, rather than forcing them to be Smis. Cq-Include-Trybots: master.tryserver.chromium.linux:linux_chromium_rel_ng Change-Id: Icc5f088ffbc2e2651b845bc71ea42060639e3e48 Reviewed-on: https://chromium-review.googlesource.com/585129 Commit-Queue: Leszek Swirski <leszeks@chromium.org> Reviewed-by: Michael Stanton <mvstanton@chromium.org> Reviewed-by: Ross McIlroy <rmcilroy@chromium.org> Reviewed-by: Michael Lippautz <mlippautz@chromium.org> Cr-Commit-Position: refs/heads/master@{#46935}
2017-07-27 12:45:00 +00:00
B(Inc), U8(1),
[parser] avoid for-loop desugaring unless loop var may be captured In https://chromium-review.googlesource.com/c/472247/, I avoided running DesugarLexicalBindingsInForStatement() if there were no lexical loop variables, the function was not resumable, and the variables are not captured by eval or a function declaration. I think it's now possible to limit this further, and only do the more extensive desugaring if there's a function declaration / eval() call in the loop body. `yield` and `await` are not an issue as those loop variables are written to the register file and not lost. This change just removes the `is_resumable()` condition. If it passes tests, I think it's safe. BUG=v8:4762, v8:5460, v8:6579 Change-Id: I92d0308ad9401c1338411bc9ae9021f978803d3a Reviewed-on: https://chromium-review.googlesource.com/563587 Commit-Queue: Caitlin Potter <caitp@igalia.com> Reviewed-by: Georg Neis <neis@chromium.org> Reviewed-by: Adam Klein <adamk@chromium.org> Cr-Commit-Position: refs/heads/master@{#46536}
2017-07-07 20:48:16 +00:00
B(Star), R(0),
[ignition] Single-switch generator bytecode Currently, yields and awaits inside loops compile to bytecode which switches to the top of the loop header, and switch again once inside the loop. This is to make loops reducible. This replaces this switching logic with a single switch bytecode that directly jumps to the bytecode being resumed. Among other things, this allows us to no longer maintain the generator state after the switch at the top of the function, and avoid having to track loop suspend counts. TurboFan still needs to have reducible loops, so we now insert loop header switches during bytecode graph building, for suspends that are discovered to be inside loops during bytecode analysis. We do, however, do some environment magic across loop headers since we know that we will continue switching if and only if we reached that loop header via a generator resume. This allows us to generate fewer phis and tighten liveness. Change-Id: Id2720ce1d6955be9a48178322cc209b3a4b8d385 Reviewed-on: https://chromium-review.googlesource.com/866734 Commit-Queue: Leszek Swirski <leszeks@chromium.org> Reviewed-by: Jaroslav Sevcik <jarin@chromium.org> Cr-Commit-Position: refs/heads/master@{#50804}
2018-01-23 13:51:38 +00:00
B(JumpLoop), U8(49), I8(0),
[parser] avoid complex for-loop desugaring when possible let/const declarations in "standard" C-style for-loops have some complex desugaring to accommodate the case where loop loop variables may be captured. This slows down the baseline performance of for-loops with let variables. This change attempts to avoid this desugaring if it's known that the loop variable is not captured at any point. A side effect of this change is that let/const loop variables, when not captured within the loop body, are not necessarily shown in the debugger, similar to other stack-allocated vars. BUG=v8:4762, v8:5460 R=marja@chromium.org, adamk@chromium.org, yangguo@chromium.org Change-Id: I8dbe545a12c086f675972bdba60c94998268311a Reviewed-on: https://chromium-review.googlesource.com/472247 Commit-Queue: Caitlin Potter <caitp@igalia.com> Reviewed-by: Yang Guo <yangguo@chromium.org> Reviewed-by: Adam Klein <adamk@chromium.org> Reviewed-by: Marja Hölttä <marja@chromium.org> Cr-Commit-Position: refs/heads/master@{#44731}
2017-04-18 14:56:06 +00:00
B(LdaUndefined),
/* 56 S> */ B(Return),
]
constant pool: [
[ignition] Single-switch generator bytecode Currently, yields and awaits inside loops compile to bytecode which switches to the top of the loop header, and switch again once inside the loop. This is to make loops reducible. This replaces this switching logic with a single switch bytecode that directly jumps to the bytecode being resumed. Among other things, this allows us to no longer maintain the generator state after the switch at the top of the function, and avoid having to track loop suspend counts. TurboFan still needs to have reducible loops, so we now insert loop header switches during bytecode graph building, for suspends that are discovered to be inside loops during bytecode analysis. We do, however, do some environment magic across loop headers since we know that we will continue switching if and only if we reached that loop header via a generator resume. This allows us to generate fewer phis and tighten liveness. Change-Id: Id2720ce1d6955be9a48178322cc209b3a4b8d385 Reviewed-on: https://chromium-review.googlesource.com/866734 Commit-Queue: Leszek Swirski <leszeks@chromium.org> Reviewed-by: Jaroslav Sevcik <jarin@chromium.org> Cr-Commit-Position: refs/heads/master@{#50804}
2018-01-23 13:51:38 +00:00
Smi [22],
Smi [68],
Move closing of generators upon final return to the generator-resume builtin. Review-Url: https://codereview.chromium.org/2936813002 Cr-Commit-Position: refs/heads/master@{#45906}
2017-06-13 13:49:45 +00:00
Smi [10],
[ignition] Use jump table for generator resume mode Bug: v8:6351 Change-Id: Iad82fcfb4172d7a9cbadffa1e40fd1cd26866895 Reviewed-on: https://chromium-review.googlesource.com/521107 Commit-Queue: Leszek Swirski <leszeks@chromium.org> Reviewed-by: Ross McIlroy <rmcilroy@chromium.org> Cr-Commit-Position: refs/heads/master@{#45705}
2017-06-05 13:23:20 +00:00
Smi [7],
Move closing of generators upon final return to the generator-resume builtin. Review-Url: https://codereview.chromium.org/2936813002 Cr-Commit-Position: refs/heads/master@{#45906}
2017-06-13 13:49:45 +00:00
Smi [10],
[ignition] Use jump table for generator resume mode Bug: v8:6351 Change-Id: Iad82fcfb4172d7a9cbadffa1e40fd1cd26866895 Reviewed-on: https://chromium-review.googlesource.com/521107 Commit-Queue: Leszek Swirski <leszeks@chromium.org> Reviewed-by: Ross McIlroy <rmcilroy@chromium.org> Cr-Commit-Position: refs/heads/master@{#45705}
2017-06-05 13:23:20 +00:00
Smi [7],
[parser] avoid complex for-loop desugaring when possible let/const declarations in "standard" C-style for-loops have some complex desugaring to accommodate the case where loop loop variables may be captured. This slows down the baseline performance of for-loops with let variables. This change attempts to avoid this desugaring if it's known that the loop variable is not captured at any point. A side effect of this change is that let/const loop variables, when not captured within the loop body, are not necessarily shown in the debugger, similar to other stack-allocated vars. BUG=v8:4762, v8:5460 R=marja@chromium.org, adamk@chromium.org, yangguo@chromium.org Change-Id: I8dbe545a12c086f675972bdba60c94998268311a Reviewed-on: https://chromium-review.googlesource.com/472247 Commit-Queue: Caitlin Potter <caitp@igalia.com> Reviewed-by: Yang Guo <yangguo@chromium.org> Reviewed-by: Adam Klein <adamk@chromium.org> Reviewed-by: Marja Hölttä <marja@chromium.org> Cr-Commit-Position: refs/heads/master@{#44731}
2017-04-18 14:56:06 +00:00
]
handlers: [
]
---
snippet: "
async function f() {
for (let x = 0; x < 10; ++x) { let y = x; }
}
f();
"
[Interpreter] Make CallJSRuntime implicitly use undefined reciever. JS runtime calls are always created with undefined recievers, so make the bytecode behave similarly to CallUndefinedReciever such that we don't need to push an explicit undefined register for the receiver for such calls. Modifies the Async[Generator/Function]Await[Caught/Uncaught] runtime calls to pass the generator in the first argument rather than the reciever since these runtime calls were desugered in the bytecode generator and explicitly passed the generator in the receiver. Change-Id: I36c8087bb3b663dccd805bfdb1eea04eb6a73269 Reviewed-on: https://chromium-review.googlesource.com/654257 Reviewed-by: Leszek Swirski <leszeks@chromium.org> Reviewed-by: Jaroslav Sevcik <jarin@chromium.org> Commit-Queue: Ross McIlroy <rmcilroy@chromium.org> Cr-Commit-Position: refs/heads/master@{#47870}
2017-09-06 22:48:36 +00:00
frame size: 12
[parser] avoid complex for-loop desugaring when possible let/const declarations in "standard" C-style for-loops have some complex desugaring to accommodate the case where loop loop variables may be captured. This slows down the baseline performance of for-loops with let variables. This change attempts to avoid this desugaring if it's known that the loop variable is not captured at any point. A side effect of this change is that let/const loop variables, when not captured within the loop body, are not necessarily shown in the debugger, similar to other stack-allocated vars. BUG=v8:4762, v8:5460 R=marja@chromium.org, adamk@chromium.org, yangguo@chromium.org Change-Id: I8dbe545a12c086f675972bdba60c94998268311a Reviewed-on: https://chromium-review.googlesource.com/472247 Commit-Queue: Caitlin Potter <caitp@igalia.com> Reviewed-by: Yang Guo <yangguo@chromium.org> Reviewed-by: Adam Klein <adamk@chromium.org> Reviewed-by: Marja Hölttä <marja@chromium.org> Cr-Commit-Position: refs/heads/master@{#44731}
2017-04-18 14:56:06 +00:00
parameter count: 1
[inspector] reworked async stack instrumentation for async functions New intstrumentation consists of: - kAsyncFunctionSuspended when async function is suspended on await (called on each await), - kAsyncFunctionFinished when async function is finished. Old instrumentation was based on reusing async function promise. Using this promise produces couple side effects: - for any promise instrumentation we first need to check if it is special case for async function promise or not - it requires expensive reading from promise object. - we capture stack for async functions even if it does not contain awaits. - we do not properly cancel async task created for async function. New intsrumntation resolved all these problems as well as provide clear mapping between async task and generator which we can use later to fetch scope information for async functions on pause. R=dgozman@chromium.org,yangguo@chromium.org Bug: v8:7078 Cq-Include-Trybots: master.tryserver.blink:linux_trusty_blink_rel Change-Id: Ifdcec947d91e6e3d4d5f9029bc080a19b8e23d41 Reviewed-on: https://chromium-review.googlesource.com/1043096 Reviewed-by: Sathya Gunasekaran <gsathya@chromium.org> Reviewed-by: Benedikt Meurer <bmeurer@chromium.org> Reviewed-by: Dmitry Gozman <dgozman@chromium.org> Commit-Queue: Aleksey Kozyatinskiy <kozyatinskiy@chromium.org> Cr-Commit-Position: refs/heads/master@{#53445}
2018-05-30 14:21:39 +00:00
bytecode array length: 140
[parser] avoid complex for-loop desugaring when possible let/const declarations in "standard" C-style for-loops have some complex desugaring to accommodate the case where loop loop variables may be captured. This slows down the baseline performance of for-loops with let variables. This change attempts to avoid this desugaring if it's known that the loop variable is not captured at any point. A side effect of this change is that let/const loop variables, when not captured within the loop body, are not necessarily shown in the debugger, similar to other stack-allocated vars. BUG=v8:4762, v8:5460 R=marja@chromium.org, adamk@chromium.org, yangguo@chromium.org Change-Id: I8dbe545a12c086f675972bdba60c94998268311a Reviewed-on: https://chromium-review.googlesource.com/472247 Commit-Queue: Caitlin Potter <caitp@igalia.com> Reviewed-by: Yang Guo <yangguo@chromium.org> Reviewed-by: Adam Klein <adamk@chromium.org> Reviewed-by: Marja Hölttä <marja@chromium.org> Cr-Commit-Position: refs/heads/master@{#44731}
2017-04-18 14:56:06 +00:00
bytecodes: [
[ignition] Move generator object creation to BytecodeGenerator This lets us avoid allocating the "this" variable for every generator, since the BytecodeGenerator can directly read the receiver via BytecodeArrayBuilder::Receive() when passing it into %_CreateJSGeneratorObject. Bug: v8:6351 Change-Id: Ib5e1f3303b6b5d5fc051ce76ea62129fd6afac65 Reviewed-on: https://chromium-review.googlesource.com/500507 Commit-Queue: Adam Klein <adamk@chromium.org> Reviewed-by: Leszek Swirski <leszeks@chromium.org> Reviewed-by: Georg Neis <neis@chromium.org> Reviewed-by: Caitlin Potter <caitp@igalia.com> Cr-Commit-Position: refs/heads/master@{#45281}
2017-05-12 17:29:53 +00:00
/* 16 E> */ B(StackCheck),
[Interpreter] Make CallJSRuntime implicitly use undefined reciever. JS runtime calls are always created with undefined recievers, so make the bytecode behave similarly to CallUndefinedReciever such that we don't need to push an explicit undefined register for the receiver for such calls. Modifies the Async[Generator/Function]Await[Caught/Uncaught] runtime calls to pass the generator in the first argument rather than the reciever since these runtime calls were desugered in the bytecode generator and explicitly passed the generator in the receiver. Change-Id: I36c8087bb3b663dccd805bfdb1eea04eb6a73269 Reviewed-on: https://chromium-review.googlesource.com/654257 Reviewed-by: Leszek Swirski <leszeks@chromium.org> Reviewed-by: Jaroslav Sevcik <jarin@chromium.org> Commit-Queue: Ross McIlroy <rmcilroy@chromium.org> Cr-Commit-Position: refs/heads/master@{#47870}
2017-09-06 22:48:36 +00:00
B(CallJSRuntime), U8(%async_function_promise_create), R(0), U8(0),
[parser] avoid for-loop desugaring unless loop var may be captured In https://chromium-review.googlesource.com/c/472247/, I avoided running DesugarLexicalBindingsInForStatement() if there were no lexical loop variables, the function was not resumable, and the variables are not captured by eval or a function declaration. I think it's now possible to limit this further, and only do the more extensive desugaring if there's a function declaration / eval() call in the loop body. `yield` and `await` are not an issue as those loop variables are written to the register file and not lost. This change just removes the `is_resumable()` condition. If it passes tests, I think it's safe. BUG=v8:4762, v8:5460, v8:6579 Change-Id: I92d0308ad9401c1338411bc9ae9021f978803d3a Reviewed-on: https://chromium-review.googlesource.com/563587 Commit-Queue: Caitlin Potter <caitp@igalia.com> Reviewed-by: Georg Neis <neis@chromium.org> Reviewed-by: Adam Klein <adamk@chromium.org> Cr-Commit-Position: refs/heads/master@{#46536}
2017-07-07 20:48:16 +00:00
B(Star), R(3),
B(Mov), R(context), R(6),
B(Mov), R(context), R(7),
/* 36 S> */ B(LdaZero),
Make non-Module generators only context allocate parameters. In particular, local variables should be allocated on stack (in bytecode register), and stored/loaded to the generator object on generator suspend/resume. The CL is based on @adamk's change to scoping/parsers (https://chromium-review.googlesource.com/c/498538/), I only made the debugger cope with this change. I should note that the CL changes the scope type of suspended generators from ScopeType.Closure to ScopeType.Local. In the future we might want to introduce ScopeType.SuspendedGenerator to make the distinction explicit. Some of the changes in the tests have been made because the debugger functions do not return scopes of closed generators anymore. Generators should be allowed to throw away their internal state when they finish. BUG=v8:6368 Review-Url: https://codereview.chromium.org/2898163002 Cr-Commit-Position: refs/heads/master@{#45515}
2017-05-24 13:54:57 +00:00
B(Star), R(1),
/* 41 S> */ B(LdaSmi), I8(10),
[objects] Make feedback vector a first-class object Instead of having feedback vector as a subtype of FixedArray with reserved slots, make it a first-class variable-sized object with a fixed-size header. This allows us to compress counters to ints in the header, rather than forcing them to be Smis. Cq-Include-Trybots: master.tryserver.chromium.linux:linux_chromium_rel_ng Change-Id: Icc5f088ffbc2e2651b845bc71ea42060639e3e48 Reviewed-on: https://chromium-review.googlesource.com/585129 Commit-Queue: Leszek Swirski <leszeks@chromium.org> Reviewed-by: Michael Stanton <mvstanton@chromium.org> Reviewed-by: Ross McIlroy <rmcilroy@chromium.org> Reviewed-by: Michael Lippautz <mlippautz@chromium.org> Cr-Commit-Position: refs/heads/master@{#46935}
2017-07-27 12:45:00 +00:00
/* 41 E> */ B(TestLessThan), R(1), U8(0),
[parser] avoid for-loop desugaring unless loop var may be captured In https://chromium-review.googlesource.com/c/472247/, I avoided running DesugarLexicalBindingsInForStatement() if there were no lexical loop variables, the function was not resumable, and the variables are not captured by eval or a function declaration. I think it's now possible to limit this further, and only do the more extensive desugaring if there's a function declaration / eval() call in the loop body. `yield` and `await` are not an issue as those loop variables are written to the register file and not lost. This change just removes the `is_resumable()` condition. If it passes tests, I think it's safe. BUG=v8:4762, v8:5460, v8:6579 Change-Id: I92d0308ad9401c1338411bc9ae9021f978803d3a Reviewed-on: https://chromium-review.googlesource.com/563587 Commit-Queue: Caitlin Potter <caitp@igalia.com> Reviewed-by: Georg Neis <neis@chromium.org> Reviewed-by: Adam Klein <adamk@chromium.org> Cr-Commit-Position: refs/heads/master@{#46536}
2017-07-07 20:48:16 +00:00
B(JumpIfFalse), U8(15),
[parser] avoid complex for-loop desugaring when possible let/const declarations in "standard" C-style for-loops have some complex desugaring to accommodate the case where loop loop variables may be captured. This slows down the baseline performance of for-loops with let variables. This change attempts to avoid this desugaring if it's known that the loop variable is not captured at any point. A side effect of this change is that let/const loop variables, when not captured within the loop body, are not necessarily shown in the debugger, similar to other stack-allocated vars. BUG=v8:4762, v8:5460 R=marja@chromium.org, adamk@chromium.org, yangguo@chromium.org Change-Id: I8dbe545a12c086f675972bdba60c94998268311a Reviewed-on: https://chromium-review.googlesource.com/472247 Commit-Queue: Caitlin Potter <caitp@igalia.com> Reviewed-by: Yang Guo <yangguo@chromium.org> Reviewed-by: Adam Klein <adamk@chromium.org> Reviewed-by: Marja Hölttä <marja@chromium.org> Cr-Commit-Position: refs/heads/master@{#44731}
2017-04-18 14:56:06 +00:00
/* 23 E> */ B(StackCheck),
Make non-Module generators only context allocate parameters. In particular, local variables should be allocated on stack (in bytecode register), and stored/loaded to the generator object on generator suspend/resume. The CL is based on @adamk's change to scoping/parsers (https://chromium-review.googlesource.com/c/498538/), I only made the debugger cope with this change. I should note that the CL changes the scope type of suspended generators from ScopeType.Closure to ScopeType.Local. In the future we might want to introduce ScopeType.SuspendedGenerator to make the distinction explicit. Some of the changes in the tests have been made because the debugger functions do not return scopes of closed generators anymore. Generators should be allowed to throw away their internal state when they finish. BUG=v8:6368 Review-Url: https://codereview.chromium.org/2898163002 Cr-Commit-Position: refs/heads/master@{#45515}
2017-05-24 13:54:57 +00:00
/* 62 S> */ B(Mov), R(1), R(0),
[parser] avoid for-loop desugaring unless loop var may be captured In https://chromium-review.googlesource.com/c/472247/, I avoided running DesugarLexicalBindingsInForStatement() if there were no lexical loop variables, the function was not resumable, and the variables are not captured by eval or a function declaration. I think it's now possible to limit this further, and only do the more extensive desugaring if there's a function declaration / eval() call in the loop body. `yield` and `await` are not an issue as those loop variables are written to the register file and not lost. This change just removes the `is_resumable()` condition. If it passes tests, I think it's safe. BUG=v8:4762, v8:5460, v8:6579 Change-Id: I92d0308ad9401c1338411bc9ae9021f978803d3a Reviewed-on: https://chromium-review.googlesource.com/563587 Commit-Queue: Caitlin Potter <caitp@igalia.com> Reviewed-by: Georg Neis <neis@chromium.org> Reviewed-by: Adam Klein <adamk@chromium.org> Cr-Commit-Position: refs/heads/master@{#46536}
2017-07-07 20:48:16 +00:00
/* 49 S> */ B(Ldar), R(0),
[objects] Make feedback vector a first-class object Instead of having feedback vector as a subtype of FixedArray with reserved slots, make it a first-class variable-sized object with a fixed-size header. This allows us to compress counters to ints in the header, rather than forcing them to be Smis. Cq-Include-Trybots: master.tryserver.chromium.linux:linux_chromium_rel_ng Change-Id: Icc5f088ffbc2e2651b845bc71ea42060639e3e48 Reviewed-on: https://chromium-review.googlesource.com/585129 Commit-Queue: Leszek Swirski <leszeks@chromium.org> Reviewed-by: Michael Stanton <mvstanton@chromium.org> Reviewed-by: Ross McIlroy <rmcilroy@chromium.org> Reviewed-by: Michael Lippautz <mlippautz@chromium.org> Cr-Commit-Position: refs/heads/master@{#46935}
2017-07-27 12:45:00 +00:00
B(Inc), U8(1),
[parser] avoid for-loop desugaring unless loop var may be captured In https://chromium-review.googlesource.com/c/472247/, I avoided running DesugarLexicalBindingsInForStatement() if there were no lexical loop variables, the function was not resumable, and the variables are not captured by eval or a function declaration. I think it's now possible to limit this further, and only do the more extensive desugaring if there's a function declaration / eval() call in the loop body. `yield` and `await` are not an issue as those loop variables are written to the register file and not lost. This change just removes the `is_resumable()` condition. If it passes tests, I think it's safe. BUG=v8:4762, v8:5460, v8:6579 Change-Id: I92d0308ad9401c1338411bc9ae9021f978803d3a Reviewed-on: https://chromium-review.googlesource.com/563587 Commit-Queue: Caitlin Potter <caitp@igalia.com> Reviewed-by: Georg Neis <neis@chromium.org> Reviewed-by: Adam Klein <adamk@chromium.org> Cr-Commit-Position: refs/heads/master@{#46536}
2017-07-07 20:48:16 +00:00
B(Star), R(1),
B(JumpLoop), U8(17), I8(0),
[parser] avoid complex for-loop desugaring when possible let/const declarations in "standard" C-style for-loops have some complex desugaring to accommodate the case where loop loop variables may be captured. This slows down the baseline performance of for-loops with let variables. This change attempts to avoid this desugaring if it's known that the loop variable is not captured at any point. A side effect of this change is that let/const loop variables, when not captured within the loop body, are not necessarily shown in the debugger, similar to other stack-allocated vars. BUG=v8:4762, v8:5460 R=marja@chromium.org, adamk@chromium.org, yangguo@chromium.org Change-Id: I8dbe545a12c086f675972bdba60c94998268311a Reviewed-on: https://chromium-review.googlesource.com/472247 Commit-Queue: Caitlin Potter <caitp@igalia.com> Reviewed-by: Yang Guo <yangguo@chromium.org> Reviewed-by: Adam Klein <adamk@chromium.org> Reviewed-by: Marja Hölttä <marja@chromium.org> Cr-Commit-Position: refs/heads/master@{#44731}
2017-04-18 14:56:06 +00:00
B(LdaUndefined),
[Interpreter] Make CallJSRuntime implicitly use undefined reciever. JS runtime calls are always created with undefined recievers, so make the bytecode behave similarly to CallUndefinedReciever such that we don't need to push an explicit undefined register for the receiver for such calls. Modifies the Async[Generator/Function]Await[Caught/Uncaught] runtime calls to pass the generator in the first argument rather than the reciever since these runtime calls were desugered in the bytecode generator and explicitly passed the generator in the receiver. Change-Id: I36c8087bb3b663dccd805bfdb1eea04eb6a73269 Reviewed-on: https://chromium-review.googlesource.com/654257 Reviewed-by: Leszek Swirski <leszeks@chromium.org> Reviewed-by: Jaroslav Sevcik <jarin@chromium.org> Commit-Queue: Ross McIlroy <rmcilroy@chromium.org> Cr-Commit-Position: refs/heads/master@{#47870}
2017-09-06 22:48:36 +00:00
B(Star), R(9),
B(Mov), R(3), R(8),
[builtins] Refactor the promise resolution and rejection logic. This introduces dedicated builtins - FulfillPromise, - RejectPromise, and - ResolvePromise, which perform the corresponding operations from the language specification, and removes the redundant entry points and the excessive inlining of these operations into other builtins. We also add the same logic on the C++ side, so that we don't need to go into JavaScript land when resolving/rejecting from the API. The C++ side has a complete implementation, including full support for the debugger and the current PromiseHook machinery. This is to avoid constantly crossing the boundary for those cases, and to also simplify the CSA side (and soon the TurboFan side), where we only do the fast-path and bail out to the runtime for the general handling. On top of this we introduce %_RejectPromise and %_ResolvePromise, which are entry points used by the bytecode and parser desugarings for async functions, and also used by the V8 Extras API. Thanks to this we can uniformly optimize these in TurboFan, where we have corresponding operators JSRejectPromise and JSResolvePromise, which currently just call into the builtins, but middle-term can be further optimized, i.e. to skip the "then" lookup for JSResolvePromise when we know something about the resolution. In TurboFan we can also already inline the default PromiseCapability [[Reject]] and [[Resolve]] functions, although this is not as effective as it can be right now, until we have inlining support for the Promise constructor (being worked on by petermarshall@ right now) and/or SFI based CALL_IC feedback. Overall this change is meant as a refactoring without significant performance impact anywhere; it seems to improve performance of simple async functions a bit, but otherwise is neutral. Bug: v8:7253 Change-Id: Id0b979f9b2843560e38cd8df4b02627dad4b6d8c Reviewed-on: https://chromium-review.googlesource.com/911632 Reviewed-by: Sathya Gunasekaran <gsathya@chromium.org> Reviewed-by: Benedikt Meurer <bmeurer@chromium.org> Reviewed-by: Georg Neis <neis@chromium.org> Commit-Queue: Benedikt Meurer <bmeurer@chromium.org> Cr-Commit-Position: refs/heads/master@{#51260}
2018-02-12 19:10:29 +00:00
/* 49 E> */ B(InvokeIntrinsic), U8(Runtime::k_ResolvePromise), R(8), U8(2),
[parser] avoid complex for-loop desugaring when possible let/const declarations in "standard" C-style for-loops have some complex desugaring to accommodate the case where loop loop variables may be captured. This slows down the baseline performance of for-loops with let variables. This change attempts to avoid this desugaring if it's known that the loop variable is not captured at any point. A side effect of this change is that let/const loop variables, when not captured within the loop body, are not necessarily shown in the debugger, similar to other stack-allocated vars. BUG=v8:4762, v8:5460 R=marja@chromium.org, adamk@chromium.org, yangguo@chromium.org Change-Id: I8dbe545a12c086f675972bdba60c94998268311a Reviewed-on: https://chromium-review.googlesource.com/472247 Commit-Queue: Caitlin Potter <caitp@igalia.com> Reviewed-by: Yang Guo <yangguo@chromium.org> Reviewed-by: Adam Klein <adamk@chromium.org> Reviewed-by: Marja Hölttä <marja@chromium.org> Cr-Commit-Position: refs/heads/master@{#44731}
2017-04-18 14:56:06 +00:00
B(LdaZero),
[parser] avoid for-loop desugaring unless loop var may be captured In https://chromium-review.googlesource.com/c/472247/, I avoided running DesugarLexicalBindingsInForStatement() if there were no lexical loop variables, the function was not resumable, and the variables are not captured by eval or a function declaration. I think it's now possible to limit this further, and only do the more extensive desugaring if there's a function declaration / eval() call in the loop body. `yield` and `await` are not an issue as those loop variables are written to the register file and not lost. This change just removes the `is_resumable()` condition. If it passes tests, I think it's safe. BUG=v8:4762, v8:5460, v8:6579 Change-Id: I92d0308ad9401c1338411bc9ae9021f978803d3a Reviewed-on: https://chromium-review.googlesource.com/563587 Commit-Queue: Caitlin Potter <caitp@igalia.com> Reviewed-by: Georg Neis <neis@chromium.org> Reviewed-by: Adam Klein <adamk@chromium.org> Cr-Commit-Position: refs/heads/master@{#46536}
2017-07-07 20:48:16 +00:00
B(Star), R(4),
B(Mov), R(3), R(5),
Remove the catch variable name from the extension field of catch contexts Instead rely on the scope info containing the name as well. Change-Id: Ie1f96ea023a793b11209510566f6831b1dfd40ab Reviewed-on: https://chromium-review.googlesource.com/1042567 Commit-Queue: Toon Verwaest <verwaest@chromium.org> Reviewed-by: Michael Starzinger <mstarzinger@chromium.org> Cr-Commit-Position: refs/heads/master@{#52983}
2018-05-04 09:29:20 +00:00
B(Jump), U8(55),
B(Jump), U8(39),
[parser] avoid for-loop desugaring unless loop var may be captured In https://chromium-review.googlesource.com/c/472247/, I avoided running DesugarLexicalBindingsInForStatement() if there were no lexical loop variables, the function was not resumable, and the variables are not captured by eval or a function declaration. I think it's now possible to limit this further, and only do the more extensive desugaring if there's a function declaration / eval() call in the loop body. `yield` and `await` are not an issue as those loop variables are written to the register file and not lost. This change just removes the `is_resumable()` condition. If it passes tests, I think it's safe. BUG=v8:4762, v8:5460, v8:6579 Change-Id: I92d0308ad9401c1338411bc9ae9021f978803d3a Reviewed-on: https://chromium-review.googlesource.com/563587 Commit-Queue: Caitlin Potter <caitp@igalia.com> Reviewed-by: Georg Neis <neis@chromium.org> Reviewed-by: Adam Klein <adamk@chromium.org> Cr-Commit-Position: refs/heads/master@{#46536}
2017-07-07 20:48:16 +00:00
B(Star), R(8),
Remove the catch variable name from the extension field of catch contexts Instead rely on the scope info containing the name as well. Change-Id: Ie1f96ea023a793b11209510566f6831b1dfd40ab Reviewed-on: https://chromium-review.googlesource.com/1042567 Commit-Queue: Toon Verwaest <verwaest@chromium.org> Reviewed-by: Michael Starzinger <mstarzinger@chromium.org> Cr-Commit-Position: refs/heads/master@{#52983}
2018-05-04 09:29:20 +00:00
B(CreateCatchContext), R(8), U8(0),
[parser] avoid for-loop desugaring unless loop var may be captured In https://chromium-review.googlesource.com/c/472247/, I avoided running DesugarLexicalBindingsInForStatement() if there were no lexical loop variables, the function was not resumable, and the variables are not captured by eval or a function declaration. I think it's now possible to limit this further, and only do the more extensive desugaring if there's a function declaration / eval() call in the loop body. `yield` and `await` are not an issue as those loop variables are written to the register file and not lost. This change just removes the `is_resumable()` condition. If it passes tests, I think it's safe. BUG=v8:4762, v8:5460, v8:6579 Change-Id: I92d0308ad9401c1338411bc9ae9021f978803d3a Reviewed-on: https://chromium-review.googlesource.com/563587 Commit-Queue: Caitlin Potter <caitp@igalia.com> Reviewed-by: Georg Neis <neis@chromium.org> Reviewed-by: Adam Klein <adamk@chromium.org> Cr-Commit-Position: refs/heads/master@{#46536}
2017-07-07 20:48:16 +00:00
B(Star), R(7),
[parser] avoid complex for-loop desugaring when possible let/const declarations in "standard" C-style for-loops have some complex desugaring to accommodate the case where loop loop variables may be captured. This slows down the baseline performance of for-loops with let variables. This change attempts to avoid this desugaring if it's known that the loop variable is not captured at any point. A side effect of this change is that let/const loop variables, when not captured within the loop body, are not necessarily shown in the debugger, similar to other stack-allocated vars. BUG=v8:4762, v8:5460 R=marja@chromium.org, adamk@chromium.org, yangguo@chromium.org Change-Id: I8dbe545a12c086f675972bdba60c94998268311a Reviewed-on: https://chromium-review.googlesource.com/472247 Commit-Queue: Caitlin Potter <caitp@igalia.com> Reviewed-by: Yang Guo <yangguo@chromium.org> Reviewed-by: Adam Klein <adamk@chromium.org> Reviewed-by: Marja Hölttä <marja@chromium.org> Cr-Commit-Position: refs/heads/master@{#44731}
2017-04-18 14:56:06 +00:00
B(LdaTheHole),
B(SetPendingMessage),
[parser] avoid for-loop desugaring unless loop var may be captured In https://chromium-review.googlesource.com/c/472247/, I avoided running DesugarLexicalBindingsInForStatement() if there were no lexical loop variables, the function was not resumable, and the variables are not captured by eval or a function declaration. I think it's now possible to limit this further, and only do the more extensive desugaring if there's a function declaration / eval() call in the loop body. `yield` and `await` are not an issue as those loop variables are written to the register file and not lost. This change just removes the `is_resumable()` condition. If it passes tests, I think it's safe. BUG=v8:4762, v8:5460, v8:6579 Change-Id: I92d0308ad9401c1338411bc9ae9021f978803d3a Reviewed-on: https://chromium-review.googlesource.com/563587 Commit-Queue: Caitlin Potter <caitp@igalia.com> Reviewed-by: Georg Neis <neis@chromium.org> Reviewed-by: Adam Klein <adamk@chromium.org> Cr-Commit-Position: refs/heads/master@{#46536}
2017-07-07 20:48:16 +00:00
B(Ldar), R(7),
B(PushContext), R(8),
[parser] avoid complex for-loop desugaring when possible let/const declarations in "standard" C-style for-loops have some complex desugaring to accommodate the case where loop loop variables may be captured. This slows down the baseline performance of for-loops with let variables. This change attempts to avoid this desugaring if it's known that the loop variable is not captured at any point. A side effect of this change is that let/const loop variables, when not captured within the loop body, are not necessarily shown in the debugger, similar to other stack-allocated vars. BUG=v8:4762, v8:5460 R=marja@chromium.org, adamk@chromium.org, yangguo@chromium.org Change-Id: I8dbe545a12c086f675972bdba60c94998268311a Reviewed-on: https://chromium-review.googlesource.com/472247 Commit-Queue: Caitlin Potter <caitp@igalia.com> Reviewed-by: Yang Guo <yangguo@chromium.org> Reviewed-by: Adam Klein <adamk@chromium.org> Reviewed-by: Marja Hölttä <marja@chromium.org> Cr-Commit-Position: refs/heads/master@{#44731}
2017-04-18 14:56:06 +00:00
B(LdaImmutableCurrentContextSlot), U8(4),
[Interpreter] Make CallJSRuntime implicitly use undefined reciever. JS runtime calls are always created with undefined recievers, so make the bytecode behave similarly to CallUndefinedReciever such that we don't need to push an explicit undefined register for the receiver for such calls. Modifies the Async[Generator/Function]Await[Caught/Uncaught] runtime calls to pass the generator in the first argument rather than the reciever since these runtime calls were desugered in the bytecode generator and explicitly passed the generator in the receiver. Change-Id: I36c8087bb3b663dccd805bfdb1eea04eb6a73269 Reviewed-on: https://chromium-review.googlesource.com/654257 Reviewed-by: Leszek Swirski <leszeks@chromium.org> Reviewed-by: Jaroslav Sevcik <jarin@chromium.org> Commit-Queue: Ross McIlroy <rmcilroy@chromium.org> Cr-Commit-Position: refs/heads/master@{#47870}
2017-09-06 22:48:36 +00:00
B(Star), R(10),
[parser] avoid complex for-loop desugaring when possible let/const declarations in "standard" C-style for-loops have some complex desugaring to accommodate the case where loop loop variables may be captured. This slows down the baseline performance of for-loops with let variables. This change attempts to avoid this desugaring if it's known that the loop variable is not captured at any point. A side effect of this change is that let/const loop variables, when not captured within the loop body, are not necessarily shown in the debugger, similar to other stack-allocated vars. BUG=v8:4762, v8:5460 R=marja@chromium.org, adamk@chromium.org, yangguo@chromium.org Change-Id: I8dbe545a12c086f675972bdba60c94998268311a Reviewed-on: https://chromium-review.googlesource.com/472247 Commit-Queue: Caitlin Potter <caitp@igalia.com> Reviewed-by: Yang Guo <yangguo@chromium.org> Reviewed-by: Adam Klein <adamk@chromium.org> Reviewed-by: Marja Hölttä <marja@chromium.org> Cr-Commit-Position: refs/heads/master@{#44731}
2017-04-18 14:56:06 +00:00
B(LdaFalse),
[Interpreter] Make CallJSRuntime implicitly use undefined reciever. JS runtime calls are always created with undefined recievers, so make the bytecode behave similarly to CallUndefinedReciever such that we don't need to push an explicit undefined register for the receiver for such calls. Modifies the Async[Generator/Function]Await[Caught/Uncaught] runtime calls to pass the generator in the first argument rather than the reciever since these runtime calls were desugered in the bytecode generator and explicitly passed the generator in the receiver. Change-Id: I36c8087bb3b663dccd805bfdb1eea04eb6a73269 Reviewed-on: https://chromium-review.googlesource.com/654257 Reviewed-by: Leszek Swirski <leszeks@chromium.org> Reviewed-by: Jaroslav Sevcik <jarin@chromium.org> Commit-Queue: Ross McIlroy <rmcilroy@chromium.org> Cr-Commit-Position: refs/heads/master@{#47870}
2017-09-06 22:48:36 +00:00
B(Star), R(11),
B(Mov), R(3), R(9),
[builtins] Refactor the promise resolution and rejection logic. This introduces dedicated builtins - FulfillPromise, - RejectPromise, and - ResolvePromise, which perform the corresponding operations from the language specification, and removes the redundant entry points and the excessive inlining of these operations into other builtins. We also add the same logic on the C++ side, so that we don't need to go into JavaScript land when resolving/rejecting from the API. The C++ side has a complete implementation, including full support for the debugger and the current PromiseHook machinery. This is to avoid constantly crossing the boundary for those cases, and to also simplify the CSA side (and soon the TurboFan side), where we only do the fast-path and bail out to the runtime for the general handling. On top of this we introduce %_RejectPromise and %_ResolvePromise, which are entry points used by the bytecode and parser desugarings for async functions, and also used by the V8 Extras API. Thanks to this we can uniformly optimize these in TurboFan, where we have corresponding operators JSRejectPromise and JSResolvePromise, which currently just call into the builtins, but middle-term can be further optimized, i.e. to skip the "then" lookup for JSResolvePromise when we know something about the resolution. In TurboFan we can also already inline the default PromiseCapability [[Reject]] and [[Resolve]] functions, although this is not as effective as it can be right now, until we have inlining support for the Promise constructor (being worked on by petermarshall@ right now) and/or SFI based CALL_IC feedback. Overall this change is meant as a refactoring without significant performance impact anywhere; it seems to improve performance of simple async functions a bit, but otherwise is neutral. Bug: v8:7253 Change-Id: Id0b979f9b2843560e38cd8df4b02627dad4b6d8c Reviewed-on: https://chromium-review.googlesource.com/911632 Reviewed-by: Sathya Gunasekaran <gsathya@chromium.org> Reviewed-by: Benedikt Meurer <bmeurer@chromium.org> Reviewed-by: Georg Neis <neis@chromium.org> Commit-Queue: Benedikt Meurer <bmeurer@chromium.org> Cr-Commit-Position: refs/heads/master@{#51260}
2018-02-12 19:10:29 +00:00
B(InvokeIntrinsic), U8(Runtime::k_RejectPromise), R(9), U8(3),
[parser] avoid for-loop desugaring unless loop var may be captured In https://chromium-review.googlesource.com/c/472247/, I avoided running DesugarLexicalBindingsInForStatement() if there were no lexical loop variables, the function was not resumable, and the variables are not captured by eval or a function declaration. I think it's now possible to limit this further, and only do the more extensive desugaring if there's a function declaration / eval() call in the loop body. `yield` and `await` are not an issue as those loop variables are written to the register file and not lost. This change just removes the `is_resumable()` condition. If it passes tests, I think it's safe. BUG=v8:4762, v8:5460, v8:6579 Change-Id: I92d0308ad9401c1338411bc9ae9021f978803d3a Reviewed-on: https://chromium-review.googlesource.com/563587 Commit-Queue: Caitlin Potter <caitp@igalia.com> Reviewed-by: Georg Neis <neis@chromium.org> Reviewed-by: Adam Klein <adamk@chromium.org> Cr-Commit-Position: refs/heads/master@{#46536}
2017-07-07 20:48:16 +00:00
B(PopContext), R(8),
[ignition] Re-use deferred return/rethrow commands Since deferred return/async return/rethrow commands don't have an associated statement, we can reuse their entries in the deferred command list. Bug: v8:4280 Bug: v8:6218 Change-Id: Ic51fb735d4ea13ebea1d5e456ec5689c0d98a4ac Reviewed-on: https://chromium-review.googlesource.com/506077 Commit-Queue: Leszek Swirski <leszeks@chromium.org> Reviewed-by: Ross McIlroy <rmcilroy@chromium.org> Cr-Commit-Position: refs/heads/master@{#45344}
2017-05-16 15:20:09 +00:00
B(LdaZero),
[parser] avoid for-loop desugaring unless loop var may be captured In https://chromium-review.googlesource.com/c/472247/, I avoided running DesugarLexicalBindingsInForStatement() if there were no lexical loop variables, the function was not resumable, and the variables are not captured by eval or a function declaration. I think it's now possible to limit this further, and only do the more extensive desugaring if there's a function declaration / eval() call in the loop body. `yield` and `await` are not an issue as those loop variables are written to the register file and not lost. This change just removes the `is_resumable()` condition. If it passes tests, I think it's safe. BUG=v8:4762, v8:5460, v8:6579 Change-Id: I92d0308ad9401c1338411bc9ae9021f978803d3a Reviewed-on: https://chromium-review.googlesource.com/563587 Commit-Queue: Caitlin Potter <caitp@igalia.com> Reviewed-by: Georg Neis <neis@chromium.org> Reviewed-by: Adam Klein <adamk@chromium.org> Cr-Commit-Position: refs/heads/master@{#46536}
2017-07-07 20:48:16 +00:00
B(Star), R(4),
B(Mov), R(3), R(5),
[ignition] Always write the deferred command result register For deferred commands (such as in try-finally), some deferred commands save and restore the accumulator using a result register (e.g. return, throw, rethrow), while others don't (e.g. break, continue, fall-through). However, conditionally reading this result register that may not ever be written caused it to be considered live from the start of the function, as far as the liveness analysis could statically tell. Now, we write the result register for all deferred commands, including the fall-through. As a micro-optimization, we re-use the Smi command tokeen to clobber the result, rather than emitting an LdaUndefined. Bug: chromium:758472 Change-Id: I2ea65e2249b40ee6403216e654a8bb88d50bec3b Reviewed-on: https://chromium-review.googlesource.com/635592 Commit-Queue: Leszek Swirski <leszeks@chromium.org> Reviewed-by: Ross McIlroy <rmcilroy@chromium.org> Cr-Commit-Position: refs/heads/master@{#47612}
2017-08-25 16:04:42 +00:00
B(Jump), U8(16),
[parser] avoid complex for-loop desugaring when possible let/const declarations in "standard" C-style for-loops have some complex desugaring to accommodate the case where loop loop variables may be captured. This slows down the baseline performance of for-loops with let variables. This change attempts to avoid this desugaring if it's known that the loop variable is not captured at any point. A side effect of this change is that let/const loop variables, when not captured within the loop body, are not necessarily shown in the debugger, similar to other stack-allocated vars. BUG=v8:4762, v8:5460 R=marja@chromium.org, adamk@chromium.org, yangguo@chromium.org Change-Id: I8dbe545a12c086f675972bdba60c94998268311a Reviewed-on: https://chromium-review.googlesource.com/472247 Commit-Queue: Caitlin Potter <caitp@igalia.com> Reviewed-by: Yang Guo <yangguo@chromium.org> Reviewed-by: Adam Klein <adamk@chromium.org> Reviewed-by: Marja Hölttä <marja@chromium.org> Cr-Commit-Position: refs/heads/master@{#44731}
2017-04-18 14:56:06 +00:00
B(LdaSmi), I8(-1),
[ignition] Always write the deferred command result register For deferred commands (such as in try-finally), some deferred commands save and restore the accumulator using a result register (e.g. return, throw, rethrow), while others don't (e.g. break, continue, fall-through). However, conditionally reading this result register that may not ever be written caused it to be considered live from the start of the function, as far as the liveness analysis could statically tell. Now, we write the result register for all deferred commands, including the fall-through. As a micro-optimization, we re-use the Smi command tokeen to clobber the result, rather than emitting an LdaUndefined. Bug: chromium:758472 Change-Id: I2ea65e2249b40ee6403216e654a8bb88d50bec3b Reviewed-on: https://chromium-review.googlesource.com/635592 Commit-Queue: Leszek Swirski <leszeks@chromium.org> Reviewed-by: Ross McIlroy <rmcilroy@chromium.org> Cr-Commit-Position: refs/heads/master@{#47612}
2017-08-25 16:04:42 +00:00
B(Star), R(5),
[parser] avoid for-loop desugaring unless loop var may be captured In https://chromium-review.googlesource.com/c/472247/, I avoided running DesugarLexicalBindingsInForStatement() if there were no lexical loop variables, the function was not resumable, and the variables are not captured by eval or a function declaration. I think it's now possible to limit this further, and only do the more extensive desugaring if there's a function declaration / eval() call in the loop body. `yield` and `await` are not an issue as those loop variables are written to the register file and not lost. This change just removes the `is_resumable()` condition. If it passes tests, I think it's safe. BUG=v8:4762, v8:5460, v8:6579 Change-Id: I92d0308ad9401c1338411bc9ae9021f978803d3a Reviewed-on: https://chromium-review.googlesource.com/563587 Commit-Queue: Caitlin Potter <caitp@igalia.com> Reviewed-by: Georg Neis <neis@chromium.org> Reviewed-by: Adam Klein <adamk@chromium.org> Cr-Commit-Position: refs/heads/master@{#46536}
2017-07-07 20:48:16 +00:00
B(Star), R(4),
[parser] avoid complex for-loop desugaring when possible let/const declarations in "standard" C-style for-loops have some complex desugaring to accommodate the case where loop loop variables may be captured. This slows down the baseline performance of for-loops with let variables. This change attempts to avoid this desugaring if it's known that the loop variable is not captured at any point. A side effect of this change is that let/const loop variables, when not captured within the loop body, are not necessarily shown in the debugger, similar to other stack-allocated vars. BUG=v8:4762, v8:5460 R=marja@chromium.org, adamk@chromium.org, yangguo@chromium.org Change-Id: I8dbe545a12c086f675972bdba60c94998268311a Reviewed-on: https://chromium-review.googlesource.com/472247 Commit-Queue: Caitlin Potter <caitp@igalia.com> Reviewed-by: Yang Guo <yangguo@chromium.org> Reviewed-by: Adam Klein <adamk@chromium.org> Reviewed-by: Marja Hölttä <marja@chromium.org> Cr-Commit-Position: refs/heads/master@{#44731}
2017-04-18 14:56:06 +00:00
B(Jump), U8(8),
[parser] avoid for-loop desugaring unless loop var may be captured In https://chromium-review.googlesource.com/c/472247/, I avoided running DesugarLexicalBindingsInForStatement() if there were no lexical loop variables, the function was not resumable, and the variables are not captured by eval or a function declaration. I think it's now possible to limit this further, and only do the more extensive desugaring if there's a function declaration / eval() call in the loop body. `yield` and `await` are not an issue as those loop variables are written to the register file and not lost. This change just removes the `is_resumable()` condition. If it passes tests, I think it's safe. BUG=v8:4762, v8:5460, v8:6579 Change-Id: I92d0308ad9401c1338411bc9ae9021f978803d3a Reviewed-on: https://chromium-review.googlesource.com/563587 Commit-Queue: Caitlin Potter <caitp@igalia.com> Reviewed-by: Georg Neis <neis@chromium.org> Reviewed-by: Adam Klein <adamk@chromium.org> Cr-Commit-Position: refs/heads/master@{#46536}
2017-07-07 20:48:16 +00:00
B(Star), R(5),
[Interpreter] Allocate context registers as temporary registers. Rather than trying to pre-calculate the number of contexts required during scope analysis, instead just allocate context registers in the register allocator. This reduces frame size a bit due to reusing of registers when the context isn't pushed. BUG=v8:6322, chromium:716265 Change-Id: I145e38fcb3797a3b86c91e90ea9326a6e55b9b89 Reviewed-on: https://chromium-review.googlesource.com/514087 Reviewed-by: Marja Hölttä <marja@chromium.org> Reviewed-by: Mythri Alle <mythria@chromium.org> Commit-Queue: Ross McIlroy <rmcilroy@chromium.org> Cr-Commit-Position: refs/heads/master@{#45522}
2017-05-24 15:48:27 +00:00
B(LdaSmi), I8(1),
[parser] avoid for-loop desugaring unless loop var may be captured In https://chromium-review.googlesource.com/c/472247/, I avoided running DesugarLexicalBindingsInForStatement() if there were no lexical loop variables, the function was not resumable, and the variables are not captured by eval or a function declaration. I think it's now possible to limit this further, and only do the more extensive desugaring if there's a function declaration / eval() call in the loop body. `yield` and `await` are not an issue as those loop variables are written to the register file and not lost. This change just removes the `is_resumable()` condition. If it passes tests, I think it's safe. BUG=v8:4762, v8:5460, v8:6579 Change-Id: I92d0308ad9401c1338411bc9ae9021f978803d3a Reviewed-on: https://chromium-review.googlesource.com/563587 Commit-Queue: Caitlin Potter <caitp@igalia.com> Reviewed-by: Georg Neis <neis@chromium.org> Reviewed-by: Adam Klein <adamk@chromium.org> Cr-Commit-Position: refs/heads/master@{#46536}
2017-07-07 20:48:16 +00:00
B(Star), R(4),
[parser] avoid complex for-loop desugaring when possible let/const declarations in "standard" C-style for-loops have some complex desugaring to accommodate the case where loop loop variables may be captured. This slows down the baseline performance of for-loops with let variables. This change attempts to avoid this desugaring if it's known that the loop variable is not captured at any point. A side effect of this change is that let/const loop variables, when not captured within the loop body, are not necessarily shown in the debugger, similar to other stack-allocated vars. BUG=v8:4762, v8:5460 R=marja@chromium.org, adamk@chromium.org, yangguo@chromium.org Change-Id: I8dbe545a12c086f675972bdba60c94998268311a Reviewed-on: https://chromium-review.googlesource.com/472247 Commit-Queue: Caitlin Potter <caitp@igalia.com> Reviewed-by: Yang Guo <yangguo@chromium.org> Reviewed-by: Adam Klein <adamk@chromium.org> Reviewed-by: Marja Hölttä <marja@chromium.org> Cr-Commit-Position: refs/heads/master@{#44731}
2017-04-18 14:56:06 +00:00
B(LdaTheHole),
B(SetPendingMessage),
[parser] avoid for-loop desugaring unless loop var may be captured In https://chromium-review.googlesource.com/c/472247/, I avoided running DesugarLexicalBindingsInForStatement() if there were no lexical loop variables, the function was not resumable, and the variables are not captured by eval or a function declaration. I think it's now possible to limit this further, and only do the more extensive desugaring if there's a function declaration / eval() call in the loop body. `yield` and `await` are not an issue as those loop variables are written to the register file and not lost. This change just removes the `is_resumable()` condition. If it passes tests, I think it's safe. BUG=v8:4762, v8:5460, v8:6579 Change-Id: I92d0308ad9401c1338411bc9ae9021f978803d3a Reviewed-on: https://chromium-review.googlesource.com/563587 Commit-Queue: Caitlin Potter <caitp@igalia.com> Reviewed-by: Georg Neis <neis@chromium.org> Reviewed-by: Adam Klein <adamk@chromium.org> Cr-Commit-Position: refs/heads/master@{#46536}
2017-07-07 20:48:16 +00:00
B(Star), R(6),
[inspector] reworked async stack instrumentation for async functions New intstrumentation consists of: - kAsyncFunctionSuspended when async function is suspended on await (called on each await), - kAsyncFunctionFinished when async function is finished. Old instrumentation was based on reusing async function promise. Using this promise produces couple side effects: - for any promise instrumentation we first need to check if it is special case for async function promise or not - it requires expensive reading from promise object. - we capture stack for async functions even if it does not contain awaits. - we do not properly cancel async task created for async function. New intsrumntation resolved all these problems as well as provide clear mapping between async task and generator which we can use later to fetch scope information for async functions on pause. R=dgozman@chromium.org,yangguo@chromium.org Bug: v8:7078 Cq-Include-Trybots: master.tryserver.blink:linux_trusty_blink_rel Change-Id: Ifdcec947d91e6e3d4d5f9029bc080a19b8e23d41 Reviewed-on: https://chromium-review.googlesource.com/1043096 Reviewed-by: Sathya Gunasekaran <gsathya@chromium.org> Reviewed-by: Benedikt Meurer <bmeurer@chromium.org> Reviewed-by: Dmitry Gozman <dgozman@chromium.org> Commit-Queue: Aleksey Kozyatinskiy <kozyatinskiy@chromium.org> Cr-Commit-Position: refs/heads/master@{#53445}
2018-05-30 14:21:39 +00:00
B(LdaFalse),
B(Star), R(8),
B(Mov), R(3), R(7),
B(CallJSRuntime), U8(%async_function_promise_release), R(7), U8(2),
[parser] avoid for-loop desugaring unless loop var may be captured In https://chromium-review.googlesource.com/c/472247/, I avoided running DesugarLexicalBindingsInForStatement() if there were no lexical loop variables, the function was not resumable, and the variables are not captured by eval or a function declaration. I think it's now possible to limit this further, and only do the more extensive desugaring if there's a function declaration / eval() call in the loop body. `yield` and `await` are not an issue as those loop variables are written to the register file and not lost. This change just removes the `is_resumable()` condition. If it passes tests, I think it's safe. BUG=v8:4762, v8:5460, v8:6579 Change-Id: I92d0308ad9401c1338411bc9ae9021f978803d3a Reviewed-on: https://chromium-review.googlesource.com/563587 Commit-Queue: Caitlin Potter <caitp@igalia.com> Reviewed-by: Georg Neis <neis@chromium.org> Reviewed-by: Adam Klein <adamk@chromium.org> Cr-Commit-Position: refs/heads/master@{#46536}
2017-07-07 20:48:16 +00:00
B(Ldar), R(6),
[parser] avoid complex for-loop desugaring when possible let/const declarations in "standard" C-style for-loops have some complex desugaring to accommodate the case where loop loop variables may be captured. This slows down the baseline performance of for-loops with let variables. This change attempts to avoid this desugaring if it's known that the loop variable is not captured at any point. A side effect of this change is that let/const loop variables, when not captured within the loop body, are not necessarily shown in the debugger, similar to other stack-allocated vars. BUG=v8:4762, v8:5460 R=marja@chromium.org, adamk@chromium.org, yangguo@chromium.org Change-Id: I8dbe545a12c086f675972bdba60c94998268311a Reviewed-on: https://chromium-review.googlesource.com/472247 Commit-Queue: Caitlin Potter <caitp@igalia.com> Reviewed-by: Yang Guo <yangguo@chromium.org> Reviewed-by: Adam Klein <adamk@chromium.org> Reviewed-by: Marja Hölttä <marja@chromium.org> Cr-Commit-Position: refs/heads/master@{#44731}
2017-04-18 14:56:06 +00:00
B(SetPendingMessage),
[parser] avoid for-loop desugaring unless loop var may be captured In https://chromium-review.googlesource.com/c/472247/, I avoided running DesugarLexicalBindingsInForStatement() if there were no lexical loop variables, the function was not resumable, and the variables are not captured by eval or a function declaration. I think it's now possible to limit this further, and only do the more extensive desugaring if there's a function declaration / eval() call in the loop body. `yield` and `await` are not an issue as those loop variables are written to the register file and not lost. This change just removes the `is_resumable()` condition. If it passes tests, I think it's safe. BUG=v8:4762, v8:5460, v8:6579 Change-Id: I92d0308ad9401c1338411bc9ae9021f978803d3a Reviewed-on: https://chromium-review.googlesource.com/563587 Commit-Queue: Caitlin Potter <caitp@igalia.com> Reviewed-by: Georg Neis <neis@chromium.org> Reviewed-by: Adam Klein <adamk@chromium.org> Cr-Commit-Position: refs/heads/master@{#46536}
2017-07-07 20:48:16 +00:00
B(Ldar), R(4),
Remove the catch variable name from the extension field of catch contexts Instead rely on the scope info containing the name as well. Change-Id: Ie1f96ea023a793b11209510566f6831b1dfd40ab Reviewed-on: https://chromium-review.googlesource.com/1042567 Commit-Queue: Toon Verwaest <verwaest@chromium.org> Reviewed-by: Michael Starzinger <mstarzinger@chromium.org> Cr-Commit-Position: refs/heads/master@{#52983}
2018-05-04 09:29:20 +00:00
B(SwitchOnSmiNoFeedback), U8(1), U8(2), I8(0),
[ignition] Re-use deferred return/rethrow commands Since deferred return/async return/rethrow commands don't have an associated statement, we can reuse their entries in the deferred command list. Bug: v8:4280 Bug: v8:6218 Change-Id: Ic51fb735d4ea13ebea1d5e456ec5689c0d98a4ac Reviewed-on: https://chromium-review.googlesource.com/506077 Commit-Queue: Leszek Swirski <leszeks@chromium.org> Reviewed-by: Ross McIlroy <rmcilroy@chromium.org> Cr-Commit-Position: refs/heads/master@{#45344}
2017-05-16 15:20:09 +00:00
B(Jump), U8(8),
[parser] avoid for-loop desugaring unless loop var may be captured In https://chromium-review.googlesource.com/c/472247/, I avoided running DesugarLexicalBindingsInForStatement() if there were no lexical loop variables, the function was not resumable, and the variables are not captured by eval or a function declaration. I think it's now possible to limit this further, and only do the more extensive desugaring if there's a function declaration / eval() call in the loop body. `yield` and `await` are not an issue as those loop variables are written to the register file and not lost. This change just removes the `is_resumable()` condition. If it passes tests, I think it's safe. BUG=v8:4762, v8:5460, v8:6579 Change-Id: I92d0308ad9401c1338411bc9ae9021f978803d3a Reviewed-on: https://chromium-review.googlesource.com/563587 Commit-Queue: Caitlin Potter <caitp@igalia.com> Reviewed-by: Georg Neis <neis@chromium.org> Reviewed-by: Adam Klein <adamk@chromium.org> Cr-Commit-Position: refs/heads/master@{#46536}
2017-07-07 20:48:16 +00:00
B(Ldar), R(5),
[parser] avoid complex for-loop desugaring when possible let/const declarations in "standard" C-style for-loops have some complex desugaring to accommodate the case where loop loop variables may be captured. This slows down the baseline performance of for-loops with let variables. This change attempts to avoid this desugaring if it's known that the loop variable is not captured at any point. A side effect of this change is that let/const loop variables, when not captured within the loop body, are not necessarily shown in the debugger, similar to other stack-allocated vars. BUG=v8:4762, v8:5460 R=marja@chromium.org, adamk@chromium.org, yangguo@chromium.org Change-Id: I8dbe545a12c086f675972bdba60c94998268311a Reviewed-on: https://chromium-review.googlesource.com/472247 Commit-Queue: Caitlin Potter <caitp@igalia.com> Reviewed-by: Yang Guo <yangguo@chromium.org> Reviewed-by: Adam Klein <adamk@chromium.org> Reviewed-by: Marja Hölttä <marja@chromium.org> Cr-Commit-Position: refs/heads/master@{#44731}
2017-04-18 14:56:06 +00:00
/* 67 S> */ B(Return),
[parser] avoid for-loop desugaring unless loop var may be captured In https://chromium-review.googlesource.com/c/472247/, I avoided running DesugarLexicalBindingsInForStatement() if there were no lexical loop variables, the function was not resumable, and the variables are not captured by eval or a function declaration. I think it's now possible to limit this further, and only do the more extensive desugaring if there's a function declaration / eval() call in the loop body. `yield` and `await` are not an issue as those loop variables are written to the register file and not lost. This change just removes the `is_resumable()` condition. If it passes tests, I think it's safe. BUG=v8:4762, v8:5460, v8:6579 Change-Id: I92d0308ad9401c1338411bc9ae9021f978803d3a Reviewed-on: https://chromium-review.googlesource.com/563587 Commit-Queue: Caitlin Potter <caitp@igalia.com> Reviewed-by: Georg Neis <neis@chromium.org> Reviewed-by: Adam Klein <adamk@chromium.org> Cr-Commit-Position: refs/heads/master@{#46536}
2017-07-07 20:48:16 +00:00
B(Ldar), R(5),
[parser] avoid complex for-loop desugaring when possible let/const declarations in "standard" C-style for-loops have some complex desugaring to accommodate the case where loop loop variables may be captured. This slows down the baseline performance of for-loops with let variables. This change attempts to avoid this desugaring if it's known that the loop variable is not captured at any point. A side effect of this change is that let/const loop variables, when not captured within the loop body, are not necessarily shown in the debugger, similar to other stack-allocated vars. BUG=v8:4762, v8:5460 R=marja@chromium.org, adamk@chromium.org, yangguo@chromium.org Change-Id: I8dbe545a12c086f675972bdba60c94998268311a Reviewed-on: https://chromium-review.googlesource.com/472247 Commit-Queue: Caitlin Potter <caitp@igalia.com> Reviewed-by: Yang Guo <yangguo@chromium.org> Reviewed-by: Adam Klein <adamk@chromium.org> Reviewed-by: Marja Hölttä <marja@chromium.org> Cr-Commit-Position: refs/heads/master@{#44731}
2017-04-18 14:56:06 +00:00
B(ReThrow),
B(LdaUndefined),
/* 67 S> */ B(Return),
]
constant pool: [
[runtime] Add SCOPE_INFO_TYPE InstanceType Bug: v8:7310 Change-Id: I82e7ada4c0f7e415887a859719eb01bb45fd3012 Reviewed-on: https://chromium-review.googlesource.com/921742 Reviewed-by: Michael Lippautz <mlippautz@chromium.org> Reviewed-by: Peter Marshall <petermarshall@chromium.org> Commit-Queue: Camillo Bruni <cbruni@chromium.org> Cr-Commit-Position: refs/heads/master@{#51324}
2018-02-16 12:36:47 +00:00
SCOPE_INFO_TYPE,
[ignition] Use jump table for "finally" deferred commands Bug: v8:6218 Change-Id: I5d245ea485c93fcc03fcd95c82ae0474f31ffa5e Reviewed-on: https://chromium-review.googlesource.com/505491 Commit-Queue: Leszek Swirski <leszeks@chromium.org> Reviewed-by: Ross McIlroy <rmcilroy@chromium.org> Cr-Commit-Position: refs/heads/master@{#45341}
2017-05-16 11:36:04 +00:00
Smi [6],
Smi [9],
[parser] avoid complex for-loop desugaring when possible let/const declarations in "standard" C-style for-loops have some complex desugaring to accommodate the case where loop loop variables may be captured. This slows down the baseline performance of for-loops with let variables. This change attempts to avoid this desugaring if it's known that the loop variable is not captured at any point. A side effect of this change is that let/const loop variables, when not captured within the loop body, are not necessarily shown in the debugger, similar to other stack-allocated vars. BUG=v8:4762, v8:5460 R=marja@chromium.org, adamk@chromium.org, yangguo@chromium.org Change-Id: I8dbe545a12c086f675972bdba60c94998268311a Reviewed-on: https://chromium-review.googlesource.com/472247 Commit-Queue: Caitlin Potter <caitp@igalia.com> Reviewed-by: Yang Guo <yangguo@chromium.org> Reviewed-by: Adam Klein <adamk@chromium.org> Reviewed-by: Marja Hölttä <marja@chromium.org> Cr-Commit-Position: refs/heads/master@{#44731}
2017-04-18 14:56:06 +00:00
]
handlers: [
Remove the catch variable name from the extension field of catch contexts Instead rely on the scope info containing the name as well. Change-Id: Ie1f96ea023a793b11209510566f6831b1dfd40ab Reviewed-on: https://chromium-review.googlesource.com/1042567 Commit-Queue: Toon Verwaest <verwaest@chromium.org> Reviewed-by: Michael Starzinger <mstarzinger@chromium.org> Cr-Commit-Position: refs/heads/master@{#52983}
2018-05-04 09:29:20 +00:00
[10, 93, 101],
[Interpreter] Make CallJSRuntime implicitly use undefined reciever. JS runtime calls are always created with undefined recievers, so make the bytecode behave similarly to CallUndefinedReciever such that we don't need to push an explicit undefined register for the receiver for such calls. Modifies the Async[Generator/Function]Await[Caught/Uncaught] runtime calls to pass the generator in the first argument rather than the reciever since these runtime calls were desugered in the bytecode generator and explicitly passed the generator in the receiver. Change-Id: I36c8087bb3b663dccd805bfdb1eea04eb6a73269 Reviewed-on: https://chromium-review.googlesource.com/654257 Reviewed-by: Leszek Swirski <leszeks@chromium.org> Reviewed-by: Jaroslav Sevcik <jarin@chromium.org> Commit-Queue: Ross McIlroy <rmcilroy@chromium.org> Cr-Commit-Position: refs/heads/master@{#47870}
2017-09-06 22:48:36 +00:00
[13, 54, 56],
[parser] avoid complex for-loop desugaring when possible let/const declarations in "standard" C-style for-loops have some complex desugaring to accommodate the case where loop loop variables may be captured. This slows down the baseline performance of for-loops with let variables. This change attempts to avoid this desugaring if it's known that the loop variable is not captured at any point. A side effect of this change is that let/const loop variables, when not captured within the loop body, are not necessarily shown in the debugger, similar to other stack-allocated vars. BUG=v8:4762, v8:5460 R=marja@chromium.org, adamk@chromium.org, yangguo@chromium.org Change-Id: I8dbe545a12c086f675972bdba60c94998268311a Reviewed-on: https://chromium-review.googlesource.com/472247 Commit-Queue: Caitlin Potter <caitp@igalia.com> Reviewed-by: Yang Guo <yangguo@chromium.org> Reviewed-by: Adam Klein <adamk@chromium.org> Reviewed-by: Marja Hölttä <marja@chromium.org> Cr-Commit-Position: refs/heads/master@{#44731}
2017-04-18 14:56:06 +00:00
]
---
snippet: "
async function f() {
for (let x = 0; x < 10; ++x) await x;
}
f();
"
[ignition] Single-switch generator bytecode Currently, yields and awaits inside loops compile to bytecode which switches to the top of the loop header, and switch again once inside the loop. This is to make loops reducible. This replaces this switching logic with a single switch bytecode that directly jumps to the bytecode being resumed. Among other things, this allows us to no longer maintain the generator state after the switch at the top of the function, and avoid having to track loop suspend counts. TurboFan still needs to have reducible loops, so we now insert loop header switches during bytecode graph building, for suspends that are discovered to be inside loops during bytecode analysis. We do, however, do some environment magic across loop headers since we know that we will continue switching if and only if we reached that loop header via a generator resume. This allows us to generate fewer phis and tighten liveness. Change-Id: Id2720ce1d6955be9a48178322cc209b3a4b8d385 Reviewed-on: https://chromium-review.googlesource.com/866734 Commit-Queue: Leszek Swirski <leszeks@chromium.org> Reviewed-by: Jaroslav Sevcik <jarin@chromium.org> Cr-Commit-Position: refs/heads/master@{#50804}
2018-01-23 13:51:38 +00:00
frame size: 11
[parser] avoid complex for-loop desugaring when possible let/const declarations in "standard" C-style for-loops have some complex desugaring to accommodate the case where loop loop variables may be captured. This slows down the baseline performance of for-loops with let variables. This change attempts to avoid this desugaring if it's known that the loop variable is not captured at any point. A side effect of this change is that let/const loop variables, when not captured within the loop body, are not necessarily shown in the debugger, similar to other stack-allocated vars. BUG=v8:4762, v8:5460 R=marja@chromium.org, adamk@chromium.org, yangguo@chromium.org Change-Id: I8dbe545a12c086f675972bdba60c94998268311a Reviewed-on: https://chromium-review.googlesource.com/472247 Commit-Queue: Caitlin Potter <caitp@igalia.com> Reviewed-by: Yang Guo <yangguo@chromium.org> Reviewed-by: Adam Klein <adamk@chromium.org> Reviewed-by: Marja Hölttä <marja@chromium.org> Cr-Commit-Position: refs/heads/master@{#44731}
2017-04-18 14:56:06 +00:00
parameter count: 1
[inspector] reworked async stack instrumentation for async functions New intstrumentation consists of: - kAsyncFunctionSuspended when async function is suspended on await (called on each await), - kAsyncFunctionFinished when async function is finished. Old instrumentation was based on reusing async function promise. Using this promise produces couple side effects: - for any promise instrumentation we first need to check if it is special case for async function promise or not - it requires expensive reading from promise object. - we capture stack for async functions even if it does not contain awaits. - we do not properly cancel async task created for async function. New intsrumntation resolved all these problems as well as provide clear mapping between async task and generator which we can use later to fetch scope information for async functions on pause. R=dgozman@chromium.org,yangguo@chromium.org Bug: v8:7078 Cq-Include-Trybots: master.tryserver.blink:linux_trusty_blink_rel Change-Id: Ifdcec947d91e6e3d4d5f9029bc080a19b8e23d41 Reviewed-on: https://chromium-review.googlesource.com/1043096 Reviewed-by: Sathya Gunasekaran <gsathya@chromium.org> Reviewed-by: Benedikt Meurer <bmeurer@chromium.org> Reviewed-by: Dmitry Gozman <dgozman@chromium.org> Commit-Queue: Aleksey Kozyatinskiy <kozyatinskiy@chromium.org> Cr-Commit-Position: refs/heads/master@{#53445}
2018-05-30 14:21:39 +00:00
bytecode array length: 191
[parser] avoid complex for-loop desugaring when possible let/const declarations in "standard" C-style for-loops have some complex desugaring to accommodate the case where loop loop variables may be captured. This slows down the baseline performance of for-loops with let variables. This change attempts to avoid this desugaring if it's known that the loop variable is not captured at any point. A side effect of this change is that let/const loop variables, when not captured within the loop body, are not necessarily shown in the debugger, similar to other stack-allocated vars. BUG=v8:4762, v8:5460 R=marja@chromium.org, adamk@chromium.org, yangguo@chromium.org Change-Id: I8dbe545a12c086f675972bdba60c94998268311a Reviewed-on: https://chromium-review.googlesource.com/472247 Commit-Queue: Caitlin Potter <caitp@igalia.com> Reviewed-by: Yang Guo <yangguo@chromium.org> Reviewed-by: Adam Klein <adamk@chromium.org> Reviewed-by: Marja Hölttä <marja@chromium.org> Cr-Commit-Position: refs/heads/master@{#44731}
2017-04-18 14:56:06 +00:00
bytecodes: [
[ignition] Single-switch generator bytecode Currently, yields and awaits inside loops compile to bytecode which switches to the top of the loop header, and switch again once inside the loop. This is to make loops reducible. This replaces this switching logic with a single switch bytecode that directly jumps to the bytecode being resumed. Among other things, this allows us to no longer maintain the generator state after the switch at the top of the function, and avoid having to track loop suspend counts. TurboFan still needs to have reducible loops, so we now insert loop header switches during bytecode graph building, for suspends that are discovered to be inside loops during bytecode analysis. We do, however, do some environment magic across loop headers since we know that we will continue switching if and only if we reached that loop header via a generator resume. This allows us to generate fewer phis and tighten liveness. Change-Id: Id2720ce1d6955be9a48178322cc209b3a4b8d385 Reviewed-on: https://chromium-review.googlesource.com/866734 Commit-Queue: Leszek Swirski <leszeks@chromium.org> Reviewed-by: Jaroslav Sevcik <jarin@chromium.org> Cr-Commit-Position: refs/heads/master@{#50804}
2018-01-23 13:51:38 +00:00
B(SwitchOnGeneratorState), R(1), U8(0), U8(1),
B(Mov), R(closure), R(3),
B(Mov), R(this), R(4),
B(InvokeIntrinsic), U8(Runtime::k_CreateJSGeneratorObject), R(3), U8(2),
[parser] avoid for-loop desugaring unless loop var may be captured In https://chromium-review.googlesource.com/c/472247/, I avoided running DesugarLexicalBindingsInForStatement() if there were no lexical loop variables, the function was not resumable, and the variables are not captured by eval or a function declaration. I think it's now possible to limit this further, and only do the more extensive desugaring if there's a function declaration / eval() call in the loop body. `yield` and `await` are not an issue as those loop variables are written to the register file and not lost. This change just removes the `is_resumable()` condition. If it passes tests, I think it's safe. BUG=v8:4762, v8:5460, v8:6579 Change-Id: I92d0308ad9401c1338411bc9ae9021f978803d3a Reviewed-on: https://chromium-review.googlesource.com/563587 Commit-Queue: Caitlin Potter <caitp@igalia.com> Reviewed-by: Georg Neis <neis@chromium.org> Reviewed-by: Adam Klein <adamk@chromium.org> Cr-Commit-Position: refs/heads/master@{#46536}
2017-07-07 20:48:16 +00:00
B(Star), R(1),
[ignition] Move generator object creation to BytecodeGenerator This lets us avoid allocating the "this" variable for every generator, since the BytecodeGenerator can directly read the receiver via BytecodeArrayBuilder::Receive() when passing it into %_CreateJSGeneratorObject. Bug: v8:6351 Change-Id: Ib5e1f3303b6b5d5fc051ce76ea62129fd6afac65 Reviewed-on: https://chromium-review.googlesource.com/500507 Commit-Queue: Adam Klein <adamk@chromium.org> Reviewed-by: Leszek Swirski <leszeks@chromium.org> Reviewed-by: Georg Neis <neis@chromium.org> Reviewed-by: Caitlin Potter <caitp@igalia.com> Cr-Commit-Position: refs/heads/master@{#45281}
2017-05-12 17:29:53 +00:00
/* 16 E> */ B(StackCheck),
[Interpreter] Make CallJSRuntime implicitly use undefined reciever. JS runtime calls are always created with undefined recievers, so make the bytecode behave similarly to CallUndefinedReciever such that we don't need to push an explicit undefined register for the receiver for such calls. Modifies the Async[Generator/Function]Await[Caught/Uncaught] runtime calls to pass the generator in the first argument rather than the reciever since these runtime calls were desugered in the bytecode generator and explicitly passed the generator in the receiver. Change-Id: I36c8087bb3b663dccd805bfdb1eea04eb6a73269 Reviewed-on: https://chromium-review.googlesource.com/654257 Reviewed-by: Leszek Swirski <leszeks@chromium.org> Reviewed-by: Jaroslav Sevcik <jarin@chromium.org> Commit-Queue: Ross McIlroy <rmcilroy@chromium.org> Cr-Commit-Position: refs/heads/master@{#47870}
2017-09-06 22:48:36 +00:00
B(CallJSRuntime), U8(%async_function_promise_create), R(0), U8(0),
[parser] avoid for-loop desugaring unless loop var may be captured In https://chromium-review.googlesource.com/c/472247/, I avoided running DesugarLexicalBindingsInForStatement() if there were no lexical loop variables, the function was not resumable, and the variables are not captured by eval or a function declaration. I think it's now possible to limit this further, and only do the more extensive desugaring if there's a function declaration / eval() call in the loop body. `yield` and `await` are not an issue as those loop variables are written to the register file and not lost. This change just removes the `is_resumable()` condition. If it passes tests, I think it's safe. BUG=v8:4762, v8:5460, v8:6579 Change-Id: I92d0308ad9401c1338411bc9ae9021f978803d3a Reviewed-on: https://chromium-review.googlesource.com/563587 Commit-Queue: Caitlin Potter <caitp@igalia.com> Reviewed-by: Georg Neis <neis@chromium.org> Reviewed-by: Adam Klein <adamk@chromium.org> Cr-Commit-Position: refs/heads/master@{#46536}
2017-07-07 20:48:16 +00:00
B(Star), R(2),
[ignition] Single-switch generator bytecode Currently, yields and awaits inside loops compile to bytecode which switches to the top of the loop header, and switch again once inside the loop. This is to make loops reducible. This replaces this switching logic with a single switch bytecode that directly jumps to the bytecode being resumed. Among other things, this allows us to no longer maintain the generator state after the switch at the top of the function, and avoid having to track loop suspend counts. TurboFan still needs to have reducible loops, so we now insert loop header switches during bytecode graph building, for suspends that are discovered to be inside loops during bytecode analysis. We do, however, do some environment magic across loop headers since we know that we will continue switching if and only if we reached that loop header via a generator resume. This allows us to generate fewer phis and tighten liveness. Change-Id: Id2720ce1d6955be9a48178322cc209b3a4b8d385 Reviewed-on: https://chromium-review.googlesource.com/866734 Commit-Queue: Leszek Swirski <leszeks@chromium.org> Reviewed-by: Jaroslav Sevcik <jarin@chromium.org> Cr-Commit-Position: refs/heads/master@{#50804}
2018-01-23 13:51:38 +00:00
B(Mov), R(context), R(5),
[Interpreter] Remove new.target from fixed frame slot. Removes the new.target slot from the interpreter's fixed frame. Instead adds a field to BytecodeArray to get the bytecode's incoming new.target or generator object register. The InterpreterEntryTrampoline then sets this register with the incoming new.target (or generator object) when the function is called. This register can be directly the new.target or generator object variable if they are LOCAL location, otherwise it is a temporary register which is then moved to the variable's location during the function prologue. This fixes a hack in the deoptimizer where we would set the new.target fixed slot to undefined in order to avoid extending it's lifetime through the optimized code - now it's just a standard register and can be optimized away as normal. Bug=v8:6644 Change-Id: Ieb8cc34cccefd9fb6634a90cbc77c6002a54f2ae Reviewed-on: https://chromium-review.googlesource.com/608966 Commit-Queue: Ross McIlroy <rmcilroy@chromium.org> Reviewed-by: Michael Starzinger <mstarzinger@chromium.org> Cr-Commit-Position: refs/heads/master@{#47320}
2017-08-11 15:16:31 +00:00
B(Mov), R(context), R(6),
[parser] avoid for-loop desugaring unless loop var may be captured In https://chromium-review.googlesource.com/c/472247/, I avoided running DesugarLexicalBindingsInForStatement() if there were no lexical loop variables, the function was not resumable, and the variables are not captured by eval or a function declaration. I think it's now possible to limit this further, and only do the more extensive desugaring if there's a function declaration / eval() call in the loop body. `yield` and `await` are not an issue as those loop variables are written to the register file and not lost. This change just removes the `is_resumable()` condition. If it passes tests, I think it's safe. BUG=v8:4762, v8:5460, v8:6579 Change-Id: I92d0308ad9401c1338411bc9ae9021f978803d3a Reviewed-on: https://chromium-review.googlesource.com/563587 Commit-Queue: Caitlin Potter <caitp@igalia.com> Reviewed-by: Georg Neis <neis@chromium.org> Reviewed-by: Adam Klein <adamk@chromium.org> Cr-Commit-Position: refs/heads/master@{#46536}
2017-07-07 20:48:16 +00:00
/* 36 S> */ B(LdaZero),
B(Star), R(0),
Make non-Module generators only context allocate parameters. In particular, local variables should be allocated on stack (in bytecode register), and stored/loaded to the generator object on generator suspend/resume. The CL is based on @adamk's change to scoping/parsers (https://chromium-review.googlesource.com/c/498538/), I only made the debugger cope with this change. I should note that the CL changes the scope type of suspended generators from ScopeType.Closure to ScopeType.Local. In the future we might want to introduce ScopeType.SuspendedGenerator to make the distinction explicit. Some of the changes in the tests have been made because the debugger functions do not return scopes of closed generators anymore. Generators should be allowed to throw away their internal state when they finish. BUG=v8:6368 Review-Url: https://codereview.chromium.org/2898163002 Cr-Commit-Position: refs/heads/master@{#45515}
2017-05-24 13:54:57 +00:00
/* 41 S> */ B(LdaSmi), I8(10),
[objects] Make feedback vector a first-class object Instead of having feedback vector as a subtype of FixedArray with reserved slots, make it a first-class variable-sized object with a fixed-size header. This allows us to compress counters to ints in the header, rather than forcing them to be Smis. Cq-Include-Trybots: master.tryserver.chromium.linux:linux_chromium_rel_ng Change-Id: Icc5f088ffbc2e2651b845bc71ea42060639e3e48 Reviewed-on: https://chromium-review.googlesource.com/585129 Commit-Queue: Leszek Swirski <leszeks@chromium.org> Reviewed-by: Michael Stanton <mvstanton@chromium.org> Reviewed-by: Ross McIlroy <rmcilroy@chromium.org> Reviewed-by: Michael Lippautz <mlippautz@chromium.org> Cr-Commit-Position: refs/heads/master@{#46935}
2017-07-27 12:45:00 +00:00
/* 41 E> */ B(TestLessThan), R(0), U8(0),
[ignition] Single-switch generator bytecode Currently, yields and awaits inside loops compile to bytecode which switches to the top of the loop header, and switch again once inside the loop. This is to make loops reducible. This replaces this switching logic with a single switch bytecode that directly jumps to the bytecode being resumed. Among other things, this allows us to no longer maintain the generator state after the switch at the top of the function, and avoid having to track loop suspend counts. TurboFan still needs to have reducible loops, so we now insert loop header switches during bytecode graph building, for suspends that are discovered to be inside loops during bytecode analysis. We do, however, do some environment magic across loop headers since we know that we will continue switching if and only if we reached that loop header via a generator resume. This allows us to generate fewer phis and tighten liveness. Change-Id: Id2720ce1d6955be9a48178322cc209b3a4b8d385 Reviewed-on: https://chromium-review.googlesource.com/866734 Commit-Queue: Leszek Swirski <leszeks@chromium.org> Reviewed-by: Jaroslav Sevcik <jarin@chromium.org> Cr-Commit-Position: refs/heads/master@{#50804}
2018-01-23 13:51:38 +00:00
B(JumpIfFalse), U8(50),
[parser] avoid complex for-loop desugaring when possible let/const declarations in "standard" C-style for-loops have some complex desugaring to accommodate the case where loop loop variables may be captured. This slows down the baseline performance of for-loops with let variables. This change attempts to avoid this desugaring if it's known that the loop variable is not captured at any point. A side effect of this change is that let/const loop variables, when not captured within the loop body, are not necessarily shown in the debugger, similar to other stack-allocated vars. BUG=v8:4762, v8:5460 R=marja@chromium.org, adamk@chromium.org, yangguo@chromium.org Change-Id: I8dbe545a12c086f675972bdba60c94998268311a Reviewed-on: https://chromium-review.googlesource.com/472247 Commit-Queue: Caitlin Potter <caitp@igalia.com> Reviewed-by: Yang Guo <yangguo@chromium.org> Reviewed-by: Adam Klein <adamk@chromium.org> Reviewed-by: Marja Hölttä <marja@chromium.org> Cr-Commit-Position: refs/heads/master@{#44731}
2017-04-18 14:56:06 +00:00
/* 23 E> */ B(StackCheck),
[ignition] Single-switch generator bytecode Currently, yields and awaits inside loops compile to bytecode which switches to the top of the loop header, and switch again once inside the loop. This is to make loops reducible. This replaces this switching logic with a single switch bytecode that directly jumps to the bytecode being resumed. Among other things, this allows us to no longer maintain the generator state after the switch at the top of the function, and avoid having to track loop suspend counts. TurboFan still needs to have reducible loops, so we now insert loop header switches during bytecode graph building, for suspends that are discovered to be inside loops during bytecode analysis. We do, however, do some environment magic across loop headers since we know that we will continue switching if and only if we reached that loop header via a generator resume. This allows us to generate fewer phis and tighten liveness. Change-Id: Id2720ce1d6955be9a48178322cc209b3a4b8d385 Reviewed-on: https://chromium-review.googlesource.com/866734 Commit-Queue: Leszek Swirski <leszeks@chromium.org> Reviewed-by: Jaroslav Sevcik <jarin@chromium.org> Cr-Commit-Position: refs/heads/master@{#50804}
2018-01-23 13:51:38 +00:00
/* 52 S> */ B(Mov), R(1), R(7),
B(Mov), R(0), R(8),
B(Mov), R(2), R(9),
Revert promises optimizations due to regressions in async hooks Revert "[async-await] Eliminate throwaway promise in async functions." This reverts commit a840f1f8f7e6c9ab32dec456a954612619855093. Revert "[async-generators] Also avoid throwaway promise here." This reverts commit feb545ceba5cb477ba883b46b60142f441ad5a3d. Revert "[async-await] Turn await closures into intrinsics." This reverts commit d97bb317386506a6798386b74647cbe0502fce35. Revert "[async-generators] Add fast-path for primitives in AsyncGeneratorYield." This reverts commit e57b500eb26dc595609613963468910eb4bcae2e. Revert "[async-generators] Add fast-path to skip "then" lookup in AsyncGeneratorResolve." This reverts commit c15802e11e30976528d1b711a9b9e6ff6b490117. Revert "[promises] Correctly run before/after hooks for await." This reverts commit ca7639239fc13e992e857710cbf7dbaac87163bf. Bug: v8:7253, v8:7745 Change-Id: I25ad0d2df3cfbc84dbb431aa25b268bce8a39e89 Reviewed-on: https://chromium-review.googlesource.com/1049975 Commit-Queue: Maya Lekova <mslekova@chromium.org> Reviewed-by: Benedikt Meurer <bmeurer@chromium.org> Cr-Commit-Position: refs/heads/master@{#53139}
2018-05-14 08:55:07 +00:00
B(CallJSRuntime), U8(%async_function_await_uncaught), R(7), U8(3),
[ignition] Single-switch generator bytecode Currently, yields and awaits inside loops compile to bytecode which switches to the top of the loop header, and switch again once inside the loop. This is to make loops reducible. This replaces this switching logic with a single switch bytecode that directly jumps to the bytecode being resumed. Among other things, this allows us to no longer maintain the generator state after the switch at the top of the function, and avoid having to track loop suspend counts. TurboFan still needs to have reducible loops, so we now insert loop header switches during bytecode graph building, for suspends that are discovered to be inside loops during bytecode analysis. We do, however, do some environment magic across loop headers since we know that we will continue switching if and only if we reached that loop header via a generator resume. This allows us to generate fewer phis and tighten liveness. Change-Id: Id2720ce1d6955be9a48178322cc209b3a4b8d385 Reviewed-on: https://chromium-review.googlesource.com/866734 Commit-Queue: Leszek Swirski <leszeks@chromium.org> Reviewed-by: Jaroslav Sevcik <jarin@chromium.org> Cr-Commit-Position: refs/heads/master@{#50804}
2018-01-23 13:51:38 +00:00
/* 52 E> */ B(SuspendGenerator), R(1), R(0), U8(7), U8(0),
B(ResumeGenerator), R(1), R(0), U8(7),
B(Star), R(7),
[Interpreter] Remove new.target from fixed frame slot. Removes the new.target slot from the interpreter's fixed frame. Instead adds a field to BytecodeArray to get the bytecode's incoming new.target or generator object register. The InterpreterEntryTrampoline then sets this register with the incoming new.target (or generator object) when the function is called. This register can be directly the new.target or generator object variable if they are LOCAL location, otherwise it is a temporary register which is then moved to the variable's location during the function prologue. This fixes a hack in the deoptimizer where we would set the new.target fixed slot to undefined in order to avoid extending it's lifetime through the optimized code - now it's just a standard register and can be optimized away as normal. Bug=v8:6644 Change-Id: Ieb8cc34cccefd9fb6634a90cbc77c6002a54f2ae Reviewed-on: https://chromium-review.googlesource.com/608966 Commit-Queue: Ross McIlroy <rmcilroy@chromium.org> Reviewed-by: Michael Starzinger <mstarzinger@chromium.org> Cr-Commit-Position: refs/heads/master@{#47320}
2017-08-11 15:16:31 +00:00
B(InvokeIntrinsic), U8(Runtime::k_GeneratorGetResumeMode), R(1), U8(1),
[ignition] Single-switch generator bytecode Currently, yields and awaits inside loops compile to bytecode which switches to the top of the loop header, and switch again once inside the loop. This is to make loops reducible. This replaces this switching logic with a single switch bytecode that directly jumps to the bytecode being resumed. Among other things, this allows us to no longer maintain the generator state after the switch at the top of the function, and avoid having to track loop suspend counts. TurboFan still needs to have reducible loops, so we now insert loop header switches during bytecode graph building, for suspends that are discovered to be inside loops during bytecode analysis. We do, however, do some environment magic across loop headers since we know that we will continue switching if and only if we reached that loop header via a generator resume. This allows us to generate fewer phis and tighten liveness. Change-Id: Id2720ce1d6955be9a48178322cc209b3a4b8d385 Reviewed-on: https://chromium-review.googlesource.com/866734 Commit-Queue: Leszek Swirski <leszeks@chromium.org> Reviewed-by: Jaroslav Sevcik <jarin@chromium.org> Cr-Commit-Position: refs/heads/master@{#50804}
2018-01-23 13:51:38 +00:00
B(Star), R(8),
[parser] avoid complex for-loop desugaring when possible let/const declarations in "standard" C-style for-loops have some complex desugaring to accommodate the case where loop loop variables may be captured. This slows down the baseline performance of for-loops with let variables. This change attempts to avoid this desugaring if it's known that the loop variable is not captured at any point. A side effect of this change is that let/const loop variables, when not captured within the loop body, are not necessarily shown in the debugger, similar to other stack-allocated vars. BUG=v8:4762, v8:5460 R=marja@chromium.org, adamk@chromium.org, yangguo@chromium.org Change-Id: I8dbe545a12c086f675972bdba60c94998268311a Reviewed-on: https://chromium-review.googlesource.com/472247 Commit-Queue: Caitlin Potter <caitp@igalia.com> Reviewed-by: Yang Guo <yangguo@chromium.org> Reviewed-by: Adam Klein <adamk@chromium.org> Reviewed-by: Marja Hölttä <marja@chromium.org> Cr-Commit-Position: refs/heads/master@{#44731}
2017-04-18 14:56:06 +00:00
B(LdaZero),
[ignition] Optimizing Smi only comparisons There are various situations where we explicitly compare a SMI against another SMI (e.g., BuildIndexedJump). This is also a common pattern for generated code (e.g., comparing a loop variable with an integer). Instead of using the generic equality/strict-equality stub for this, which is expensive, this CL offers a simple comparison stub, repurposing the TestEqualStrictNoFeedback bytecode to TestReferenceEqual Bug: v8:5310 Change-Id: Ib2b47cd24d5386cf0d20d3bd794776dc6e3a02a5 Reviewed-on: https://chromium-review.googlesource.com/1007542 Reviewed-by: Leszek Swirski <leszeks@chromium.org> Commit-Queue: Christian O. Andersson <cricke@chromium.org> Cr-Commit-Position: refs/heads/master@{#52655}
2018-04-18 00:04:34 +00:00
B(TestReferenceEqual), R(8),
[async-await] desugar Await in BytecodeGenerator This includes several changes. From most to least interesting: - No longer implement AwaitExpressions using a do-expression. - Reduces frame-size of async generators by not allocating temporary variables to hold results of Await epxressions. - Streamline and reduce generated bytecodes for Await. - Debugger no longer emits a debug::kCallBreakLocation breakpoint for the JS-builtin call performed for Await, and instead only emits such a breakpoint if the operand of Await is actually a call. - Push fewer parameters to Await* builtins, using the receiver for the first parameter (possible now that the CallRuntime invocation not part of the AST). - Adds a new Await AST node. No new members or anything, but it seemed palatable to avoid having `if (is_await())` in a number of VisitSuspend functions. BUG=v8:5855, v8:5099, v8:4483 R=rmcilroy@chromium.org, kozyatinskiy@chromium.org, yangguo@chromium.org TBR=bmeurer@chromium.org Change-Id: I9cd3fda99cd40295c04fdf1aea01b5d83fac6caf Reviewed-on: https://chromium-review.googlesource.com/558806 Commit-Queue: Georg Neis <neis@chromium.org> Reviewed-by: Georg Neis <neis@chromium.org> Reviewed-by: Ross McIlroy <rmcilroy@chromium.org> Reviewed-by: Tobias Tebbi <tebbi@chromium.org> Reviewed-by: Aleksey Kozyatinskiy <kozyatinskiy@chromium.org> Cr-Commit-Position: refs/heads/master@{#46666}
2017-07-13 13:53:13 +00:00
B(JumpIfTrue), U8(5),
[ignition] Single-switch generator bytecode Currently, yields and awaits inside loops compile to bytecode which switches to the top of the loop header, and switch again once inside the loop. This is to make loops reducible. This replaces this switching logic with a single switch bytecode that directly jumps to the bytecode being resumed. Among other things, this allows us to no longer maintain the generator state after the switch at the top of the function, and avoid having to track loop suspend counts. TurboFan still needs to have reducible loops, so we now insert loop header switches during bytecode graph building, for suspends that are discovered to be inside loops during bytecode analysis. We do, however, do some environment magic across loop headers since we know that we will continue switching if and only if we reached that loop header via a generator resume. This allows us to generate fewer phis and tighten liveness. Change-Id: Id2720ce1d6955be9a48178322cc209b3a4b8d385 Reviewed-on: https://chromium-review.googlesource.com/866734 Commit-Queue: Leszek Swirski <leszeks@chromium.org> Reviewed-by: Jaroslav Sevcik <jarin@chromium.org> Cr-Commit-Position: refs/heads/master@{#50804}
2018-01-23 13:51:38 +00:00
B(Ldar), R(7),
Reland "[async-iteration] implement spec-change to `yield` in async generators" Per https://github.com/tc39/proposal-async-iteration/pull/102/files: AsyncGeneratorResolve no longer unwraps a value component. Instead, the value is unwrapped before the builtin call via Await, allowing Promise rejections to affect the generator control flow. Thus, all `yield <expr>` implicitly become `yield await <expr>`. Additionally, `return <expr>` becomes `return await <expr>`. Finally, when the generator is resumed with `.return()`, the parameter passed to .return() is awaited before generator execution properly continues). BUG=v8:6187, v8:5855 R=littledan@chromium.org, neis@chromium.org, adamk@chromium.org TBR=rmcilroy@chromium.org, neis@chromium.org Cq-Include-Trybots: master.tryserver.v8:v8_linux_noi18n_rel_ng Change-Id: Id7718028fd555481f9f4ca0dbecfa788e3057c48 Reviewed-on: https://chromium-review.googlesource.com/594500 Reviewed-by: Caitlin Potter <caitp@igalia.com> Reviewed-by: Ross McIlroy <rmcilroy@chromium.org> Reviewed-by: Jakob Gruber <jgruber@chromium.org> Commit-Queue: Caitlin Potter <caitp@igalia.com> Cr-Commit-Position: refs/heads/master@{#47058}
2017-08-01 14:45:03 +00:00
B(ReThrow),
[parser] avoid for-loop desugaring unless loop var may be captured In https://chromium-review.googlesource.com/c/472247/, I avoided running DesugarLexicalBindingsInForStatement() if there were no lexical loop variables, the function was not resumable, and the variables are not captured by eval or a function declaration. I think it's now possible to limit this further, and only do the more extensive desugaring if there's a function declaration / eval() call in the loop body. `yield` and `await` are not an issue as those loop variables are written to the register file and not lost. This change just removes the `is_resumable()` condition. If it passes tests, I think it's safe. BUG=v8:4762, v8:5460, v8:6579 Change-Id: I92d0308ad9401c1338411bc9ae9021f978803d3a Reviewed-on: https://chromium-review.googlesource.com/563587 Commit-Queue: Caitlin Potter <caitp@igalia.com> Reviewed-by: Georg Neis <neis@chromium.org> Reviewed-by: Adam Klein <adamk@chromium.org> Cr-Commit-Position: refs/heads/master@{#46536}
2017-07-07 20:48:16 +00:00
/* 49 S> */ B(Ldar), R(0),
[objects] Make feedback vector a first-class object Instead of having feedback vector as a subtype of FixedArray with reserved slots, make it a first-class variable-sized object with a fixed-size header. This allows us to compress counters to ints in the header, rather than forcing them to be Smis. Cq-Include-Trybots: master.tryserver.chromium.linux:linux_chromium_rel_ng Change-Id: Icc5f088ffbc2e2651b845bc71ea42060639e3e48 Reviewed-on: https://chromium-review.googlesource.com/585129 Commit-Queue: Leszek Swirski <leszeks@chromium.org> Reviewed-by: Michael Stanton <mvstanton@chromium.org> Reviewed-by: Ross McIlroy <rmcilroy@chromium.org> Reviewed-by: Michael Lippautz <mlippautz@chromium.org> Cr-Commit-Position: refs/heads/master@{#46935}
2017-07-27 12:45:00 +00:00
B(Inc), U8(1),
[parser] avoid for-loop desugaring unless loop var may be captured In https://chromium-review.googlesource.com/c/472247/, I avoided running DesugarLexicalBindingsInForStatement() if there were no lexical loop variables, the function was not resumable, and the variables are not captured by eval or a function declaration. I think it's now possible to limit this further, and only do the more extensive desugaring if there's a function declaration / eval() call in the loop body. `yield` and `await` are not an issue as those loop variables are written to the register file and not lost. This change just removes the `is_resumable()` condition. If it passes tests, I think it's safe. BUG=v8:4762, v8:5460, v8:6579 Change-Id: I92d0308ad9401c1338411bc9ae9021f978803d3a Reviewed-on: https://chromium-review.googlesource.com/563587 Commit-Queue: Caitlin Potter <caitp@igalia.com> Reviewed-by: Georg Neis <neis@chromium.org> Reviewed-by: Adam Klein <adamk@chromium.org> Cr-Commit-Position: refs/heads/master@{#46536}
2017-07-07 20:48:16 +00:00
B(Star), R(0),
[ignition] Single-switch generator bytecode Currently, yields and awaits inside loops compile to bytecode which switches to the top of the loop header, and switch again once inside the loop. This is to make loops reducible. This replaces this switching logic with a single switch bytecode that directly jumps to the bytecode being resumed. Among other things, this allows us to no longer maintain the generator state after the switch at the top of the function, and avoid having to track loop suspend counts. TurboFan still needs to have reducible loops, so we now insert loop header switches during bytecode graph building, for suspends that are discovered to be inside loops during bytecode analysis. We do, however, do some environment magic across loop headers since we know that we will continue switching if and only if we reached that loop header via a generator resume. This allows us to generate fewer phis and tighten liveness. Change-Id: Id2720ce1d6955be9a48178322cc209b3a4b8d385 Reviewed-on: https://chromium-review.googlesource.com/866734 Commit-Queue: Leszek Swirski <leszeks@chromium.org> Reviewed-by: Jaroslav Sevcik <jarin@chromium.org> Cr-Commit-Position: refs/heads/master@{#50804}
2018-01-23 13:51:38 +00:00
B(JumpLoop), U8(52), I8(0),
[parser] avoid complex for-loop desugaring when possible let/const declarations in "standard" C-style for-loops have some complex desugaring to accommodate the case where loop loop variables may be captured. This slows down the baseline performance of for-loops with let variables. This change attempts to avoid this desugaring if it's known that the loop variable is not captured at any point. A side effect of this change is that let/const loop variables, when not captured within the loop body, are not necessarily shown in the debugger, similar to other stack-allocated vars. BUG=v8:4762, v8:5460 R=marja@chromium.org, adamk@chromium.org, yangguo@chromium.org Change-Id: I8dbe545a12c086f675972bdba60c94998268311a Reviewed-on: https://chromium-review.googlesource.com/472247 Commit-Queue: Caitlin Potter <caitp@igalia.com> Reviewed-by: Yang Guo <yangguo@chromium.org> Reviewed-by: Adam Klein <adamk@chromium.org> Reviewed-by: Marja Hölttä <marja@chromium.org> Cr-Commit-Position: refs/heads/master@{#44731}
2017-04-18 14:56:06 +00:00
B(LdaUndefined),
[ignition] Single-switch generator bytecode Currently, yields and awaits inside loops compile to bytecode which switches to the top of the loop header, and switch again once inside the loop. This is to make loops reducible. This replaces this switching logic with a single switch bytecode that directly jumps to the bytecode being resumed. Among other things, this allows us to no longer maintain the generator state after the switch at the top of the function, and avoid having to track loop suspend counts. TurboFan still needs to have reducible loops, so we now insert loop header switches during bytecode graph building, for suspends that are discovered to be inside loops during bytecode analysis. We do, however, do some environment magic across loop headers since we know that we will continue switching if and only if we reached that loop header via a generator resume. This allows us to generate fewer phis and tighten liveness. Change-Id: Id2720ce1d6955be9a48178322cc209b3a4b8d385 Reviewed-on: https://chromium-review.googlesource.com/866734 Commit-Queue: Leszek Swirski <leszeks@chromium.org> Reviewed-by: Jaroslav Sevcik <jarin@chromium.org> Cr-Commit-Position: refs/heads/master@{#50804}
2018-01-23 13:51:38 +00:00
B(Star), R(8),
B(Mov), R(2), R(7),
[builtins] Refactor the promise resolution and rejection logic. This introduces dedicated builtins - FulfillPromise, - RejectPromise, and - ResolvePromise, which perform the corresponding operations from the language specification, and removes the redundant entry points and the excessive inlining of these operations into other builtins. We also add the same logic on the C++ side, so that we don't need to go into JavaScript land when resolving/rejecting from the API. The C++ side has a complete implementation, including full support for the debugger and the current PromiseHook machinery. This is to avoid constantly crossing the boundary for those cases, and to also simplify the CSA side (and soon the TurboFan side), where we only do the fast-path and bail out to the runtime for the general handling. On top of this we introduce %_RejectPromise and %_ResolvePromise, which are entry points used by the bytecode and parser desugarings for async functions, and also used by the V8 Extras API. Thanks to this we can uniformly optimize these in TurboFan, where we have corresponding operators JSRejectPromise and JSResolvePromise, which currently just call into the builtins, but middle-term can be further optimized, i.e. to skip the "then" lookup for JSResolvePromise when we know something about the resolution. In TurboFan we can also already inline the default PromiseCapability [[Reject]] and [[Resolve]] functions, although this is not as effective as it can be right now, until we have inlining support for the Promise constructor (being worked on by petermarshall@ right now) and/or SFI based CALL_IC feedback. Overall this change is meant as a refactoring without significant performance impact anywhere; it seems to improve performance of simple async functions a bit, but otherwise is neutral. Bug: v8:7253 Change-Id: Id0b979f9b2843560e38cd8df4b02627dad4b6d8c Reviewed-on: https://chromium-review.googlesource.com/911632 Reviewed-by: Sathya Gunasekaran <gsathya@chromium.org> Reviewed-by: Benedikt Meurer <bmeurer@chromium.org> Reviewed-by: Georg Neis <neis@chromium.org> Commit-Queue: Benedikt Meurer <bmeurer@chromium.org> Cr-Commit-Position: refs/heads/master@{#51260}
2018-02-12 19:10:29 +00:00
/* 49 E> */ B(InvokeIntrinsic), U8(Runtime::k_ResolvePromise), R(7), U8(2),
[ignition] Re-use deferred return/rethrow commands Since deferred return/async return/rethrow commands don't have an associated statement, we can reuse their entries in the deferred command list. Bug: v8:4280 Bug: v8:6218 Change-Id: Ic51fb735d4ea13ebea1d5e456ec5689c0d98a4ac Reviewed-on: https://chromium-review.googlesource.com/506077 Commit-Queue: Leszek Swirski <leszeks@chromium.org> Reviewed-by: Ross McIlroy <rmcilroy@chromium.org> Cr-Commit-Position: refs/heads/master@{#45344}
2017-05-16 15:20:09 +00:00
B(LdaZero),
[ignition] Single-switch generator bytecode Currently, yields and awaits inside loops compile to bytecode which switches to the top of the loop header, and switch again once inside the loop. This is to make loops reducible. This replaces this switching logic with a single switch bytecode that directly jumps to the bytecode being resumed. Among other things, this allows us to no longer maintain the generator state after the switch at the top of the function, and avoid having to track loop suspend counts. TurboFan still needs to have reducible loops, so we now insert loop header switches during bytecode graph building, for suspends that are discovered to be inside loops during bytecode analysis. We do, however, do some environment magic across loop headers since we know that we will continue switching if and only if we reached that loop header via a generator resume. This allows us to generate fewer phis and tighten liveness. Change-Id: Id2720ce1d6955be9a48178322cc209b3a4b8d385 Reviewed-on: https://chromium-review.googlesource.com/866734 Commit-Queue: Leszek Swirski <leszeks@chromium.org> Reviewed-by: Jaroslav Sevcik <jarin@chromium.org> Cr-Commit-Position: refs/heads/master@{#50804}
2018-01-23 13:51:38 +00:00
B(Star), R(3),
B(Mov), R(2), R(4),
Remove the catch variable name from the extension field of catch contexts Instead rely on the scope info containing the name as well. Change-Id: Ie1f96ea023a793b11209510566f6831b1dfd40ab Reviewed-on: https://chromium-review.googlesource.com/1042567 Commit-Queue: Toon Verwaest <verwaest@chromium.org> Reviewed-by: Michael Starzinger <mstarzinger@chromium.org> Cr-Commit-Position: refs/heads/master@{#52983}
2018-05-04 09:29:20 +00:00
B(Jump), U8(55),
B(Jump), U8(39),
[Interpreter] Remove new.target from fixed frame slot. Removes the new.target slot from the interpreter's fixed frame. Instead adds a field to BytecodeArray to get the bytecode's incoming new.target or generator object register. The InterpreterEntryTrampoline then sets this register with the incoming new.target (or generator object) when the function is called. This register can be directly the new.target or generator object variable if they are LOCAL location, otherwise it is a temporary register which is then moved to the variable's location during the function prologue. This fixes a hack in the deoptimizer where we would set the new.target fixed slot to undefined in order to avoid extending it's lifetime through the optimized code - now it's just a standard register and can be optimized away as normal. Bug=v8:6644 Change-Id: Ieb8cc34cccefd9fb6634a90cbc77c6002a54f2ae Reviewed-on: https://chromium-review.googlesource.com/608966 Commit-Queue: Ross McIlroy <rmcilroy@chromium.org> Reviewed-by: Michael Starzinger <mstarzinger@chromium.org> Cr-Commit-Position: refs/heads/master@{#47320}
2017-08-11 15:16:31 +00:00
B(Star), R(7),
Remove the catch variable name from the extension field of catch contexts Instead rely on the scope info containing the name as well. Change-Id: Ie1f96ea023a793b11209510566f6831b1dfd40ab Reviewed-on: https://chromium-review.googlesource.com/1042567 Commit-Queue: Toon Verwaest <verwaest@chromium.org> Reviewed-by: Michael Starzinger <mstarzinger@chromium.org> Cr-Commit-Position: refs/heads/master@{#52983}
2018-05-04 09:29:20 +00:00
B(CreateCatchContext), R(7), U8(1),
[ignition] Single-switch generator bytecode Currently, yields and awaits inside loops compile to bytecode which switches to the top of the loop header, and switch again once inside the loop. This is to make loops reducible. This replaces this switching logic with a single switch bytecode that directly jumps to the bytecode being resumed. Among other things, this allows us to no longer maintain the generator state after the switch at the top of the function, and avoid having to track loop suspend counts. TurboFan still needs to have reducible loops, so we now insert loop header switches during bytecode graph building, for suspends that are discovered to be inside loops during bytecode analysis. We do, however, do some environment magic across loop headers since we know that we will continue switching if and only if we reached that loop header via a generator resume. This allows us to generate fewer phis and tighten liveness. Change-Id: Id2720ce1d6955be9a48178322cc209b3a4b8d385 Reviewed-on: https://chromium-review.googlesource.com/866734 Commit-Queue: Leszek Swirski <leszeks@chromium.org> Reviewed-by: Jaroslav Sevcik <jarin@chromium.org> Cr-Commit-Position: refs/heads/master@{#50804}
2018-01-23 13:51:38 +00:00
B(Star), R(6),
[parser] avoid complex for-loop desugaring when possible let/const declarations in "standard" C-style for-loops have some complex desugaring to accommodate the case where loop loop variables may be captured. This slows down the baseline performance of for-loops with let variables. This change attempts to avoid this desugaring if it's known that the loop variable is not captured at any point. A side effect of this change is that let/const loop variables, when not captured within the loop body, are not necessarily shown in the debugger, similar to other stack-allocated vars. BUG=v8:4762, v8:5460 R=marja@chromium.org, adamk@chromium.org, yangguo@chromium.org Change-Id: I8dbe545a12c086f675972bdba60c94998268311a Reviewed-on: https://chromium-review.googlesource.com/472247 Commit-Queue: Caitlin Potter <caitp@igalia.com> Reviewed-by: Yang Guo <yangguo@chromium.org> Reviewed-by: Adam Klein <adamk@chromium.org> Reviewed-by: Marja Hölttä <marja@chromium.org> Cr-Commit-Position: refs/heads/master@{#44731}
2017-04-18 14:56:06 +00:00
B(LdaTheHole),
B(SetPendingMessage),
[ignition] Single-switch generator bytecode Currently, yields and awaits inside loops compile to bytecode which switches to the top of the loop header, and switch again once inside the loop. This is to make loops reducible. This replaces this switching logic with a single switch bytecode that directly jumps to the bytecode being resumed. Among other things, this allows us to no longer maintain the generator state after the switch at the top of the function, and avoid having to track loop suspend counts. TurboFan still needs to have reducible loops, so we now insert loop header switches during bytecode graph building, for suspends that are discovered to be inside loops during bytecode analysis. We do, however, do some environment magic across loop headers since we know that we will continue switching if and only if we reached that loop header via a generator resume. This allows us to generate fewer phis and tighten liveness. Change-Id: Id2720ce1d6955be9a48178322cc209b3a4b8d385 Reviewed-on: https://chromium-review.googlesource.com/866734 Commit-Queue: Leszek Swirski <leszeks@chromium.org> Reviewed-by: Jaroslav Sevcik <jarin@chromium.org> Cr-Commit-Position: refs/heads/master@{#50804}
2018-01-23 13:51:38 +00:00
B(Ldar), R(6),
B(PushContext), R(7),
[parser] avoid complex for-loop desugaring when possible let/const declarations in "standard" C-style for-loops have some complex desugaring to accommodate the case where loop loop variables may be captured. This slows down the baseline performance of for-loops with let variables. This change attempts to avoid this desugaring if it's known that the loop variable is not captured at any point. A side effect of this change is that let/const loop variables, when not captured within the loop body, are not necessarily shown in the debugger, similar to other stack-allocated vars. BUG=v8:4762, v8:5460 R=marja@chromium.org, adamk@chromium.org, yangguo@chromium.org Change-Id: I8dbe545a12c086f675972bdba60c94998268311a Reviewed-on: https://chromium-review.googlesource.com/472247 Commit-Queue: Caitlin Potter <caitp@igalia.com> Reviewed-by: Yang Guo <yangguo@chromium.org> Reviewed-by: Adam Klein <adamk@chromium.org> Reviewed-by: Marja Hölttä <marja@chromium.org> Cr-Commit-Position: refs/heads/master@{#44731}
2017-04-18 14:56:06 +00:00
B(LdaImmutableCurrentContextSlot), U8(4),
[ignition] Single-switch generator bytecode Currently, yields and awaits inside loops compile to bytecode which switches to the top of the loop header, and switch again once inside the loop. This is to make loops reducible. This replaces this switching logic with a single switch bytecode that directly jumps to the bytecode being resumed. Among other things, this allows us to no longer maintain the generator state after the switch at the top of the function, and avoid having to track loop suspend counts. TurboFan still needs to have reducible loops, so we now insert loop header switches during bytecode graph building, for suspends that are discovered to be inside loops during bytecode analysis. We do, however, do some environment magic across loop headers since we know that we will continue switching if and only if we reached that loop header via a generator resume. This allows us to generate fewer phis and tighten liveness. Change-Id: Id2720ce1d6955be9a48178322cc209b3a4b8d385 Reviewed-on: https://chromium-review.googlesource.com/866734 Commit-Queue: Leszek Swirski <leszeks@chromium.org> Reviewed-by: Jaroslav Sevcik <jarin@chromium.org> Cr-Commit-Position: refs/heads/master@{#50804}
2018-01-23 13:51:38 +00:00
B(Star), R(9),
This is a first step towards reducing the number of stores/loads when suspending/resuming a generator. Unfortunately, even for an empty generator, we still use 8 register for various things (try-finally, copies of generator object, parser-introduced temporaries). I will try to get rid of these in separate CLs. Changes: - SuspendGenerator bytecode now takes register list to save. - ResumeGenerator was split into two bytecodes: * Resume generator reads the state out and marks the generator as 'executing'. * RestoreGeneratorRegisters reloads the registers from the generator. + this required adding support for output register list. - Introduced generator_object_ register in the bytecode generator. * in subsequent CLs, I will make better use of it, the goal is to get rid if the .generator_object local variable. - Taught register optimizer to flush unassigned registers. BUG=v8:6379 Review-Url: https://codereview.chromium.org/2894293003 Cr-Commit-Position: refs/heads/master@{#45675}
2017-06-02 11:55:48 +00:00
B(LdaFalse),
[ignition] Single-switch generator bytecode Currently, yields and awaits inside loops compile to bytecode which switches to the top of the loop header, and switch again once inside the loop. This is to make loops reducible. This replaces this switching logic with a single switch bytecode that directly jumps to the bytecode being resumed. Among other things, this allows us to no longer maintain the generator state after the switch at the top of the function, and avoid having to track loop suspend counts. TurboFan still needs to have reducible loops, so we now insert loop header switches during bytecode graph building, for suspends that are discovered to be inside loops during bytecode analysis. We do, however, do some environment magic across loop headers since we know that we will continue switching if and only if we reached that loop header via a generator resume. This allows us to generate fewer phis and tighten liveness. Change-Id: Id2720ce1d6955be9a48178322cc209b3a4b8d385 Reviewed-on: https://chromium-review.googlesource.com/866734 Commit-Queue: Leszek Swirski <leszeks@chromium.org> Reviewed-by: Jaroslav Sevcik <jarin@chromium.org> Cr-Commit-Position: refs/heads/master@{#50804}
2018-01-23 13:51:38 +00:00
B(Star), R(10),
B(Mov), R(2), R(8),
[builtins] Refactor the promise resolution and rejection logic. This introduces dedicated builtins - FulfillPromise, - RejectPromise, and - ResolvePromise, which perform the corresponding operations from the language specification, and removes the redundant entry points and the excessive inlining of these operations into other builtins. We also add the same logic on the C++ side, so that we don't need to go into JavaScript land when resolving/rejecting from the API. The C++ side has a complete implementation, including full support for the debugger and the current PromiseHook machinery. This is to avoid constantly crossing the boundary for those cases, and to also simplify the CSA side (and soon the TurboFan side), where we only do the fast-path and bail out to the runtime for the general handling. On top of this we introduce %_RejectPromise and %_ResolvePromise, which are entry points used by the bytecode and parser desugarings for async functions, and also used by the V8 Extras API. Thanks to this we can uniformly optimize these in TurboFan, where we have corresponding operators JSRejectPromise and JSResolvePromise, which currently just call into the builtins, but middle-term can be further optimized, i.e. to skip the "then" lookup for JSResolvePromise when we know something about the resolution. In TurboFan we can also already inline the default PromiseCapability [[Reject]] and [[Resolve]] functions, although this is not as effective as it can be right now, until we have inlining support for the Promise constructor (being worked on by petermarshall@ right now) and/or SFI based CALL_IC feedback. Overall this change is meant as a refactoring without significant performance impact anywhere; it seems to improve performance of simple async functions a bit, but otherwise is neutral. Bug: v8:7253 Change-Id: Id0b979f9b2843560e38cd8df4b02627dad4b6d8c Reviewed-on: https://chromium-review.googlesource.com/911632 Reviewed-by: Sathya Gunasekaran <gsathya@chromium.org> Reviewed-by: Benedikt Meurer <bmeurer@chromium.org> Reviewed-by: Georg Neis <neis@chromium.org> Commit-Queue: Benedikt Meurer <bmeurer@chromium.org> Cr-Commit-Position: refs/heads/master@{#51260}
2018-02-12 19:10:29 +00:00
B(InvokeIntrinsic), U8(Runtime::k_RejectPromise), R(8), U8(3),
[ignition] Single-switch generator bytecode Currently, yields and awaits inside loops compile to bytecode which switches to the top of the loop header, and switch again once inside the loop. This is to make loops reducible. This replaces this switching logic with a single switch bytecode that directly jumps to the bytecode being resumed. Among other things, this allows us to no longer maintain the generator state after the switch at the top of the function, and avoid having to track loop suspend counts. TurboFan still needs to have reducible loops, so we now insert loop header switches during bytecode graph building, for suspends that are discovered to be inside loops during bytecode analysis. We do, however, do some environment magic across loop headers since we know that we will continue switching if and only if we reached that loop header via a generator resume. This allows us to generate fewer phis and tighten liveness. Change-Id: Id2720ce1d6955be9a48178322cc209b3a4b8d385 Reviewed-on: https://chromium-review.googlesource.com/866734 Commit-Queue: Leszek Swirski <leszeks@chromium.org> Reviewed-by: Jaroslav Sevcik <jarin@chromium.org> Cr-Commit-Position: refs/heads/master@{#50804}
2018-01-23 13:51:38 +00:00
B(PopContext), R(7),
[ignition] Re-use deferred return/rethrow commands Since deferred return/async return/rethrow commands don't have an associated statement, we can reuse their entries in the deferred command list. Bug: v8:4280 Bug: v8:6218 Change-Id: Ic51fb735d4ea13ebea1d5e456ec5689c0d98a4ac Reviewed-on: https://chromium-review.googlesource.com/506077 Commit-Queue: Leszek Swirski <leszeks@chromium.org> Reviewed-by: Ross McIlroy <rmcilroy@chromium.org> Cr-Commit-Position: refs/heads/master@{#45344}
2017-05-16 15:20:09 +00:00
B(LdaZero),
[ignition] Single-switch generator bytecode Currently, yields and awaits inside loops compile to bytecode which switches to the top of the loop header, and switch again once inside the loop. This is to make loops reducible. This replaces this switching logic with a single switch bytecode that directly jumps to the bytecode being resumed. Among other things, this allows us to no longer maintain the generator state after the switch at the top of the function, and avoid having to track loop suspend counts. TurboFan still needs to have reducible loops, so we now insert loop header switches during bytecode graph building, for suspends that are discovered to be inside loops during bytecode analysis. We do, however, do some environment magic across loop headers since we know that we will continue switching if and only if we reached that loop header via a generator resume. This allows us to generate fewer phis and tighten liveness. Change-Id: Id2720ce1d6955be9a48178322cc209b3a4b8d385 Reviewed-on: https://chromium-review.googlesource.com/866734 Commit-Queue: Leszek Swirski <leszeks@chromium.org> Reviewed-by: Jaroslav Sevcik <jarin@chromium.org> Cr-Commit-Position: refs/heads/master@{#50804}
2018-01-23 13:51:38 +00:00
B(Star), R(3),
B(Mov), R(2), R(4),
[ignition] Always write the deferred command result register For deferred commands (such as in try-finally), some deferred commands save and restore the accumulator using a result register (e.g. return, throw, rethrow), while others don't (e.g. break, continue, fall-through). However, conditionally reading this result register that may not ever be written caused it to be considered live from the start of the function, as far as the liveness analysis could statically tell. Now, we write the result register for all deferred commands, including the fall-through. As a micro-optimization, we re-use the Smi command tokeen to clobber the result, rather than emitting an LdaUndefined. Bug: chromium:758472 Change-Id: I2ea65e2249b40ee6403216e654a8bb88d50bec3b Reviewed-on: https://chromium-review.googlesource.com/635592 Commit-Queue: Leszek Swirski <leszeks@chromium.org> Reviewed-by: Ross McIlroy <rmcilroy@chromium.org> Cr-Commit-Position: refs/heads/master@{#47612}
2017-08-25 16:04:42 +00:00
B(Jump), U8(16),
[parser] avoid complex for-loop desugaring when possible let/const declarations in "standard" C-style for-loops have some complex desugaring to accommodate the case where loop loop variables may be captured. This slows down the baseline performance of for-loops with let variables. This change attempts to avoid this desugaring if it's known that the loop variable is not captured at any point. A side effect of this change is that let/const loop variables, when not captured within the loop body, are not necessarily shown in the debugger, similar to other stack-allocated vars. BUG=v8:4762, v8:5460 R=marja@chromium.org, adamk@chromium.org, yangguo@chromium.org Change-Id: I8dbe545a12c086f675972bdba60c94998268311a Reviewed-on: https://chromium-review.googlesource.com/472247 Commit-Queue: Caitlin Potter <caitp@igalia.com> Reviewed-by: Yang Guo <yangguo@chromium.org> Reviewed-by: Adam Klein <adamk@chromium.org> Reviewed-by: Marja Hölttä <marja@chromium.org> Cr-Commit-Position: refs/heads/master@{#44731}
2017-04-18 14:56:06 +00:00
B(LdaSmi), I8(-1),
[Interpreter] Remove new.target from fixed frame slot. Removes the new.target slot from the interpreter's fixed frame. Instead adds a field to BytecodeArray to get the bytecode's incoming new.target or generator object register. The InterpreterEntryTrampoline then sets this register with the incoming new.target (or generator object) when the function is called. This register can be directly the new.target or generator object variable if they are LOCAL location, otherwise it is a temporary register which is then moved to the variable's location during the function prologue. This fixes a hack in the deoptimizer where we would set the new.target fixed slot to undefined in order to avoid extending it's lifetime through the optimized code - now it's just a standard register and can be optimized away as normal. Bug=v8:6644 Change-Id: Ieb8cc34cccefd9fb6634a90cbc77c6002a54f2ae Reviewed-on: https://chromium-review.googlesource.com/608966 Commit-Queue: Ross McIlroy <rmcilroy@chromium.org> Reviewed-by: Michael Starzinger <mstarzinger@chromium.org> Cr-Commit-Position: refs/heads/master@{#47320}
2017-08-11 15:16:31 +00:00
B(Star), R(4),
[ignition] Single-switch generator bytecode Currently, yields and awaits inside loops compile to bytecode which switches to the top of the loop header, and switch again once inside the loop. This is to make loops reducible. This replaces this switching logic with a single switch bytecode that directly jumps to the bytecode being resumed. Among other things, this allows us to no longer maintain the generator state after the switch at the top of the function, and avoid having to track loop suspend counts. TurboFan still needs to have reducible loops, so we now insert loop header switches during bytecode graph building, for suspends that are discovered to be inside loops during bytecode analysis. We do, however, do some environment magic across loop headers since we know that we will continue switching if and only if we reached that loop header via a generator resume. This allows us to generate fewer phis and tighten liveness. Change-Id: Id2720ce1d6955be9a48178322cc209b3a4b8d385 Reviewed-on: https://chromium-review.googlesource.com/866734 Commit-Queue: Leszek Swirski <leszeks@chromium.org> Reviewed-by: Jaroslav Sevcik <jarin@chromium.org> Cr-Commit-Position: refs/heads/master@{#50804}
2018-01-23 13:51:38 +00:00
B(Star), R(3),
This is a first step towards reducing the number of stores/loads when suspending/resuming a generator. Unfortunately, even for an empty generator, we still use 8 register for various things (try-finally, copies of generator object, parser-introduced temporaries). I will try to get rid of these in separate CLs. Changes: - SuspendGenerator bytecode now takes register list to save. - ResumeGenerator was split into two bytecodes: * Resume generator reads the state out and marks the generator as 'executing'. * RestoreGeneratorRegisters reloads the registers from the generator. + this required adding support for output register list. - Introduced generator_object_ register in the bytecode generator. * in subsequent CLs, I will make better use of it, the goal is to get rid if the .generator_object local variable. - Taught register optimizer to flush unassigned registers. BUG=v8:6379 Review-Url: https://codereview.chromium.org/2894293003 Cr-Commit-Position: refs/heads/master@{#45675}
2017-06-02 11:55:48 +00:00
B(Jump), U8(8),
[Interpreter] Remove new.target from fixed frame slot. Removes the new.target slot from the interpreter's fixed frame. Instead adds a field to BytecodeArray to get the bytecode's incoming new.target or generator object register. The InterpreterEntryTrampoline then sets this register with the incoming new.target (or generator object) when the function is called. This register can be directly the new.target or generator object variable if they are LOCAL location, otherwise it is a temporary register which is then moved to the variable's location during the function prologue. This fixes a hack in the deoptimizer where we would set the new.target fixed slot to undefined in order to avoid extending it's lifetime through the optimized code - now it's just a standard register and can be optimized away as normal. Bug=v8:6644 Change-Id: Ieb8cc34cccefd9fb6634a90cbc77c6002a54f2ae Reviewed-on: https://chromium-review.googlesource.com/608966 Commit-Queue: Ross McIlroy <rmcilroy@chromium.org> Reviewed-by: Michael Starzinger <mstarzinger@chromium.org> Cr-Commit-Position: refs/heads/master@{#47320}
2017-08-11 15:16:31 +00:00
B(Star), R(4),
[ignition] Single-switch generator bytecode Currently, yields and awaits inside loops compile to bytecode which switches to the top of the loop header, and switch again once inside the loop. This is to make loops reducible. This replaces this switching logic with a single switch bytecode that directly jumps to the bytecode being resumed. Among other things, this allows us to no longer maintain the generator state after the switch at the top of the function, and avoid having to track loop suspend counts. TurboFan still needs to have reducible loops, so we now insert loop header switches during bytecode graph building, for suspends that are discovered to be inside loops during bytecode analysis. We do, however, do some environment magic across loop headers since we know that we will continue switching if and only if we reached that loop header via a generator resume. This allows us to generate fewer phis and tighten liveness. Change-Id: Id2720ce1d6955be9a48178322cc209b3a4b8d385 Reviewed-on: https://chromium-review.googlesource.com/866734 Commit-Queue: Leszek Swirski <leszeks@chromium.org> Reviewed-by: Jaroslav Sevcik <jarin@chromium.org> Cr-Commit-Position: refs/heads/master@{#50804}
2018-01-23 13:51:38 +00:00
B(LdaSmi), I8(1),
B(Star), R(3),
[parser] avoid complex for-loop desugaring when possible let/const declarations in "standard" C-style for-loops have some complex desugaring to accommodate the case where loop loop variables may be captured. This slows down the baseline performance of for-loops with let variables. This change attempts to avoid this desugaring if it's known that the loop variable is not captured at any point. A side effect of this change is that let/const loop variables, when not captured within the loop body, are not necessarily shown in the debugger, similar to other stack-allocated vars. BUG=v8:4762, v8:5460 R=marja@chromium.org, adamk@chromium.org, yangguo@chromium.org Change-Id: I8dbe545a12c086f675972bdba60c94998268311a Reviewed-on: https://chromium-review.googlesource.com/472247 Commit-Queue: Caitlin Potter <caitp@igalia.com> Reviewed-by: Yang Guo <yangguo@chromium.org> Reviewed-by: Adam Klein <adamk@chromium.org> Reviewed-by: Marja Hölttä <marja@chromium.org> Cr-Commit-Position: refs/heads/master@{#44731}
2017-04-18 14:56:06 +00:00
B(LdaTheHole),
B(SetPendingMessage),
[ignition] Single-switch generator bytecode Currently, yields and awaits inside loops compile to bytecode which switches to the top of the loop header, and switch again once inside the loop. This is to make loops reducible. This replaces this switching logic with a single switch bytecode that directly jumps to the bytecode being resumed. Among other things, this allows us to no longer maintain the generator state after the switch at the top of the function, and avoid having to track loop suspend counts. TurboFan still needs to have reducible loops, so we now insert loop header switches during bytecode graph building, for suspends that are discovered to be inside loops during bytecode analysis. We do, however, do some environment magic across loop headers since we know that we will continue switching if and only if we reached that loop header via a generator resume. This allows us to generate fewer phis and tighten liveness. Change-Id: Id2720ce1d6955be9a48178322cc209b3a4b8d385 Reviewed-on: https://chromium-review.googlesource.com/866734 Commit-Queue: Leszek Swirski <leszeks@chromium.org> Reviewed-by: Jaroslav Sevcik <jarin@chromium.org> Cr-Commit-Position: refs/heads/master@{#50804}
2018-01-23 13:51:38 +00:00
B(Star), R(5),
[inspector] reworked async stack instrumentation for async functions New intstrumentation consists of: - kAsyncFunctionSuspended when async function is suspended on await (called on each await), - kAsyncFunctionFinished when async function is finished. Old instrumentation was based on reusing async function promise. Using this promise produces couple side effects: - for any promise instrumentation we first need to check if it is special case for async function promise or not - it requires expensive reading from promise object. - we capture stack for async functions even if it does not contain awaits. - we do not properly cancel async task created for async function. New intsrumntation resolved all these problems as well as provide clear mapping between async task and generator which we can use later to fetch scope information for async functions on pause. R=dgozman@chromium.org,yangguo@chromium.org Bug: v8:7078 Cq-Include-Trybots: master.tryserver.blink:linux_trusty_blink_rel Change-Id: Ifdcec947d91e6e3d4d5f9029bc080a19b8e23d41 Reviewed-on: https://chromium-review.googlesource.com/1043096 Reviewed-by: Sathya Gunasekaran <gsathya@chromium.org> Reviewed-by: Benedikt Meurer <bmeurer@chromium.org> Reviewed-by: Dmitry Gozman <dgozman@chromium.org> Commit-Queue: Aleksey Kozyatinskiy <kozyatinskiy@chromium.org> Cr-Commit-Position: refs/heads/master@{#53445}
2018-05-30 14:21:39 +00:00
B(LdaTrue),
B(Star), R(7),
B(Mov), R(2), R(6),
B(CallJSRuntime), U8(%async_function_promise_release), R(6), U8(2),
[ignition] Single-switch generator bytecode Currently, yields and awaits inside loops compile to bytecode which switches to the top of the loop header, and switch again once inside the loop. This is to make loops reducible. This replaces this switching logic with a single switch bytecode that directly jumps to the bytecode being resumed. Among other things, this allows us to no longer maintain the generator state after the switch at the top of the function, and avoid having to track loop suspend counts. TurboFan still needs to have reducible loops, so we now insert loop header switches during bytecode graph building, for suspends that are discovered to be inside loops during bytecode analysis. We do, however, do some environment magic across loop headers since we know that we will continue switching if and only if we reached that loop header via a generator resume. This allows us to generate fewer phis and tighten liveness. Change-Id: Id2720ce1d6955be9a48178322cc209b3a4b8d385 Reviewed-on: https://chromium-review.googlesource.com/866734 Commit-Queue: Leszek Swirski <leszeks@chromium.org> Reviewed-by: Jaroslav Sevcik <jarin@chromium.org> Cr-Commit-Position: refs/heads/master@{#50804}
2018-01-23 13:51:38 +00:00
B(Ldar), R(5),
[parser] avoid complex for-loop desugaring when possible let/const declarations in "standard" C-style for-loops have some complex desugaring to accommodate the case where loop loop variables may be captured. This slows down the baseline performance of for-loops with let variables. This change attempts to avoid this desugaring if it's known that the loop variable is not captured at any point. A side effect of this change is that let/const loop variables, when not captured within the loop body, are not necessarily shown in the debugger, similar to other stack-allocated vars. BUG=v8:4762, v8:5460 R=marja@chromium.org, adamk@chromium.org, yangguo@chromium.org Change-Id: I8dbe545a12c086f675972bdba60c94998268311a Reviewed-on: https://chromium-review.googlesource.com/472247 Commit-Queue: Caitlin Potter <caitp@igalia.com> Reviewed-by: Yang Guo <yangguo@chromium.org> Reviewed-by: Adam Klein <adamk@chromium.org> Reviewed-by: Marja Hölttä <marja@chromium.org> Cr-Commit-Position: refs/heads/master@{#44731}
2017-04-18 14:56:06 +00:00
B(SetPendingMessage),
[ignition] Single-switch generator bytecode Currently, yields and awaits inside loops compile to bytecode which switches to the top of the loop header, and switch again once inside the loop. This is to make loops reducible. This replaces this switching logic with a single switch bytecode that directly jumps to the bytecode being resumed. Among other things, this allows us to no longer maintain the generator state after the switch at the top of the function, and avoid having to track loop suspend counts. TurboFan still needs to have reducible loops, so we now insert loop header switches during bytecode graph building, for suspends that are discovered to be inside loops during bytecode analysis. We do, however, do some environment magic across loop headers since we know that we will continue switching if and only if we reached that loop header via a generator resume. This allows us to generate fewer phis and tighten liveness. Change-Id: Id2720ce1d6955be9a48178322cc209b3a4b8d385 Reviewed-on: https://chromium-review.googlesource.com/866734 Commit-Queue: Leszek Swirski <leszeks@chromium.org> Reviewed-by: Jaroslav Sevcik <jarin@chromium.org> Cr-Commit-Position: refs/heads/master@{#50804}
2018-01-23 13:51:38 +00:00
B(Ldar), R(3),
Remove the catch variable name from the extension field of catch contexts Instead rely on the scope info containing the name as well. Change-Id: Ie1f96ea023a793b11209510566f6831b1dfd40ab Reviewed-on: https://chromium-review.googlesource.com/1042567 Commit-Queue: Toon Verwaest <verwaest@chromium.org> Reviewed-by: Michael Starzinger <mstarzinger@chromium.org> Cr-Commit-Position: refs/heads/master@{#52983}
2018-05-04 09:29:20 +00:00
B(SwitchOnSmiNoFeedback), U8(2), U8(2), I8(0),
[ignition] Re-use deferred return/rethrow commands Since deferred return/async return/rethrow commands don't have an associated statement, we can reuse their entries in the deferred command list. Bug: v8:4280 Bug: v8:6218 Change-Id: Ic51fb735d4ea13ebea1d5e456ec5689c0d98a4ac Reviewed-on: https://chromium-review.googlesource.com/506077 Commit-Queue: Leszek Swirski <leszeks@chromium.org> Reviewed-by: Ross McIlroy <rmcilroy@chromium.org> Cr-Commit-Position: refs/heads/master@{#45344}
2017-05-16 15:20:09 +00:00
B(Jump), U8(8),
[ignition] Single-switch generator bytecode Currently, yields and awaits inside loops compile to bytecode which switches to the top of the loop header, and switch again once inside the loop. This is to make loops reducible. This replaces this switching logic with a single switch bytecode that directly jumps to the bytecode being resumed. Among other things, this allows us to no longer maintain the generator state after the switch at the top of the function, and avoid having to track loop suspend counts. TurboFan still needs to have reducible loops, so we now insert loop header switches during bytecode graph building, for suspends that are discovered to be inside loops during bytecode analysis. We do, however, do some environment magic across loop headers since we know that we will continue switching if and only if we reached that loop header via a generator resume. This allows us to generate fewer phis and tighten liveness. Change-Id: Id2720ce1d6955be9a48178322cc209b3a4b8d385 Reviewed-on: https://chromium-review.googlesource.com/866734 Commit-Queue: Leszek Swirski <leszeks@chromium.org> Reviewed-by: Jaroslav Sevcik <jarin@chromium.org> Cr-Commit-Position: refs/heads/master@{#50804}
2018-01-23 13:51:38 +00:00
B(Ldar), R(4),
[parser] avoid complex for-loop desugaring when possible let/const declarations in "standard" C-style for-loops have some complex desugaring to accommodate the case where loop loop variables may be captured. This slows down the baseline performance of for-loops with let variables. This change attempts to avoid this desugaring if it's known that the loop variable is not captured at any point. A side effect of this change is that let/const loop variables, when not captured within the loop body, are not necessarily shown in the debugger, similar to other stack-allocated vars. BUG=v8:4762, v8:5460 R=marja@chromium.org, adamk@chromium.org, yangguo@chromium.org Change-Id: I8dbe545a12c086f675972bdba60c94998268311a Reviewed-on: https://chromium-review.googlesource.com/472247 Commit-Queue: Caitlin Potter <caitp@igalia.com> Reviewed-by: Yang Guo <yangguo@chromium.org> Reviewed-by: Adam Klein <adamk@chromium.org> Reviewed-by: Marja Hölttä <marja@chromium.org> Cr-Commit-Position: refs/heads/master@{#44731}
2017-04-18 14:56:06 +00:00
/* 61 S> */ B(Return),
[ignition] Single-switch generator bytecode Currently, yields and awaits inside loops compile to bytecode which switches to the top of the loop header, and switch again once inside the loop. This is to make loops reducible. This replaces this switching logic with a single switch bytecode that directly jumps to the bytecode being resumed. Among other things, this allows us to no longer maintain the generator state after the switch at the top of the function, and avoid having to track loop suspend counts. TurboFan still needs to have reducible loops, so we now insert loop header switches during bytecode graph building, for suspends that are discovered to be inside loops during bytecode analysis. We do, however, do some environment magic across loop headers since we know that we will continue switching if and only if we reached that loop header via a generator resume. This allows us to generate fewer phis and tighten liveness. Change-Id: Id2720ce1d6955be9a48178322cc209b3a4b8d385 Reviewed-on: https://chromium-review.googlesource.com/866734 Commit-Queue: Leszek Swirski <leszeks@chromium.org> Reviewed-by: Jaroslav Sevcik <jarin@chromium.org> Cr-Commit-Position: refs/heads/master@{#50804}
2018-01-23 13:51:38 +00:00
B(Ldar), R(4),
[parser] avoid complex for-loop desugaring when possible let/const declarations in "standard" C-style for-loops have some complex desugaring to accommodate the case where loop loop variables may be captured. This slows down the baseline performance of for-loops with let variables. This change attempts to avoid this desugaring if it's known that the loop variable is not captured at any point. A side effect of this change is that let/const loop variables, when not captured within the loop body, are not necessarily shown in the debugger, similar to other stack-allocated vars. BUG=v8:4762, v8:5460 R=marja@chromium.org, adamk@chromium.org, yangguo@chromium.org Change-Id: I8dbe545a12c086f675972bdba60c94998268311a Reviewed-on: https://chromium-review.googlesource.com/472247 Commit-Queue: Caitlin Potter <caitp@igalia.com> Reviewed-by: Yang Guo <yangguo@chromium.org> Reviewed-by: Adam Klein <adamk@chromium.org> Reviewed-by: Marja Hölttä <marja@chromium.org> Cr-Commit-Position: refs/heads/master@{#44731}
2017-04-18 14:56:06 +00:00
B(ReThrow),
B(LdaUndefined),
/* 61 S> */ B(Return),
]
constant pool: [
[ignition] Single-switch generator bytecode Currently, yields and awaits inside loops compile to bytecode which switches to the top of the loop header, and switch again once inside the loop. This is to make loops reducible. This replaces this switching logic with a single switch bytecode that directly jumps to the bytecode being resumed. Among other things, this allows us to no longer maintain the generator state after the switch at the top of the function, and avoid having to track loop suspend counts. TurboFan still needs to have reducible loops, so we now insert loop header switches during bytecode graph building, for suspends that are discovered to be inside loops during bytecode analysis. We do, however, do some environment magic across loop headers since we know that we will continue switching if and only if we reached that loop header via a generator resume. This allows us to generate fewer phis and tighten liveness. Change-Id: Id2720ce1d6955be9a48178322cc209b3a4b8d385 Reviewed-on: https://chromium-review.googlesource.com/866734 Commit-Queue: Leszek Swirski <leszeks@chromium.org> Reviewed-by: Jaroslav Sevcik <jarin@chromium.org> Cr-Commit-Position: refs/heads/master@{#50804}
2018-01-23 13:51:38 +00:00
Smi [58],
[runtime] Add SCOPE_INFO_TYPE InstanceType Bug: v8:7310 Change-Id: I82e7ada4c0f7e415887a859719eb01bb45fd3012 Reviewed-on: https://chromium-review.googlesource.com/921742 Reviewed-by: Michael Lippautz <mlippautz@chromium.org> Reviewed-by: Peter Marshall <petermarshall@chromium.org> Commit-Queue: Camillo Bruni <cbruni@chromium.org> Cr-Commit-Position: refs/heads/master@{#51324}
2018-02-16 12:36:47 +00:00
SCOPE_INFO_TYPE,
[ignition] Use jump table for "finally" deferred commands Bug: v8:6218 Change-Id: I5d245ea485c93fcc03fcd95c82ae0474f31ffa5e Reviewed-on: https://chromium-review.googlesource.com/505491 Commit-Queue: Leszek Swirski <leszeks@chromium.org> Reviewed-by: Ross McIlroy <rmcilroy@chromium.org> Cr-Commit-Position: refs/heads/master@{#45341}
2017-05-16 11:36:04 +00:00
Smi [6],
Smi [9],
[parser] avoid complex for-loop desugaring when possible let/const declarations in "standard" C-style for-loops have some complex desugaring to accommodate the case where loop loop variables may be captured. This slows down the baseline performance of for-loops with let variables. This change attempts to avoid this desugaring if it's known that the loop variable is not captured at any point. A side effect of this change is that let/const loop variables, when not captured within the loop body, are not necessarily shown in the debugger, similar to other stack-allocated vars. BUG=v8:4762, v8:5460 R=marja@chromium.org, adamk@chromium.org, yangguo@chromium.org Change-Id: I8dbe545a12c086f675972bdba60c94998268311a Reviewed-on: https://chromium-review.googlesource.com/472247 Commit-Queue: Caitlin Potter <caitp@igalia.com> Reviewed-by: Yang Guo <yangguo@chromium.org> Reviewed-by: Adam Klein <adamk@chromium.org> Reviewed-by: Marja Hölttä <marja@chromium.org> Cr-Commit-Position: refs/heads/master@{#44731}
2017-04-18 14:56:06 +00:00
]
handlers: [
Remove the catch variable name from the extension field of catch contexts Instead rely on the scope info containing the name as well. Change-Id: Ie1f96ea023a793b11209510566f6831b1dfd40ab Reviewed-on: https://chromium-review.googlesource.com/1042567 Commit-Queue: Toon Verwaest <verwaest@chromium.org> Reviewed-by: Michael Starzinger <mstarzinger@chromium.org> Cr-Commit-Position: refs/heads/master@{#52983}
2018-05-04 09:29:20 +00:00
[26, 144, 152],
[ignition] Reduce generator switch boilerplate The SwitchOnGeneratorState bytecode now also falls through if the generator object is undefined (so that we don't need that jump) and restores generator context (so that we don't need that PushContext). This saves 10 bytes per generator. Change-Id: Ie0872c827119b9f1d1e9244d3be6496a30cd9620 Reviewed-on: https://chromium-review.googlesource.com/867051 Commit-Queue: Leszek Swirski <leszeks@chromium.org> Reviewed-by: Ross McIlroy <rmcilroy@chromium.org> Reviewed-by: Georg Neis <neis@chromium.org> Cr-Commit-Position: refs/heads/master@{#50845}
2018-01-24 11:19:51 +00:00
[29, 105, 107],
[parser] avoid complex for-loop desugaring when possible let/const declarations in "standard" C-style for-loops have some complex desugaring to accommodate the case where loop loop variables may be captured. This slows down the baseline performance of for-loops with let variables. This change attempts to avoid this desugaring if it's known that the loop variable is not captured at any point. A side effect of this change is that let/const loop variables, when not captured within the loop body, are not necessarily shown in the debugger, similar to other stack-allocated vars. BUG=v8:4762, v8:5460 R=marja@chromium.org, adamk@chromium.org, yangguo@chromium.org Change-Id: I8dbe545a12c086f675972bdba60c94998268311a Reviewed-on: https://chromium-review.googlesource.com/472247 Commit-Queue: Caitlin Potter <caitp@igalia.com> Reviewed-by: Yang Guo <yangguo@chromium.org> Reviewed-by: Adam Klein <adamk@chromium.org> Reviewed-by: Marja Hölttä <marja@chromium.org> Cr-Commit-Position: refs/heads/master@{#44731}
2017-04-18 14:56:06 +00:00
]
Reference in New Issue Copy Permalink