- 15 Jul, 2020 18 commits
-
-
Dominik Inführ authored
This is a reland of 273f4e42 Original change's description: > [heap] Avoid ParkedMutexGuard during allocation > > Since main thread allocation does not start incremental marking anymore > while holding allocation_mutex_, background allocation does not need > ParkedMutexGuard anymore to avoid deadlocks. > > This also means background thread allocation isn't paused anymore to > perform a GC, which already resulted in subtle bugs (e.g. in > ExpandBackground with incremental marking). We also do not > stop-the-world anymore while holding allocation_mutex_. > > Bug: v8:10315 > Change-Id: Iadf00bc26434c765722b82a10497ab06151f15cc > Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2289771 > Reviewed-by: Ulan Degenbaev <ulan@chromium.org> > Commit-Queue: Dominik Inführ <dinfuehr@chromium.org> > Cr-Commit-Position: refs/heads/master@{#68754} Bug: v8:10315 Change-Id: If5aec78370685369ad0f1d7a76002d45f149ddfb Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2297468Reviewed-by:
Ulan Degenbaev <ulan@chromium.org> Commit-Queue: Dominik Inführ <dinfuehr@chromium.org> Cr-Commit-Position: refs/heads/master@{#68873}
-
Zeynep Cankara authored
Change names of global functions from globalXYZEvent to handleXYZ format to increase readability. Bug: v8:10667 Change-Id: Ie5e7b6b7ab1c535f5c6beb65361d0a78cac96e46 Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2299362 Commit-Queue: Zeynep Cankara <zcankara@google.com> Reviewed-by:
Camillo Bruni <cbruni@chromium.org> Cr-Commit-Position: refs/heads/master@{#68872}
-
Maya Lekova authored
This change adds a has_error parameter on the stack which allows the fast callback to report an error. In case this parameter is set to non-zero, the generated code calls the slow (default) callback, which can throw the exception. Bug: chromium:1052746 Change-Id: Ib11f6b0bef37d5eb1d04cd6d0a3ef59028dcc448 Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2183929Reviewed-by:
Georg Neis <neis@chromium.org> Reviewed-by:
Toon Verwaest <verwaest@chromium.org> Reviewed-by:
Jakob Gruber <jgruber@chromium.org> Commit-Queue: Maya Lekova <mslekova@chromium.org> Cr-Commit-Position: refs/heads/master@{#68871}
-
Milad Farazmand authored
Change-Id: I96b8b6735b4a1d8bb42040ecd9e46c5e97675749 Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2298141Reviewed-by:
Junliang Yan <jyan@ca.ibm.com> Commit-Queue: Milad Farazmand <miladfar@ca.ibm.com> Cr-Commit-Position: refs/heads/master@{#68870}
-
Zeynep Cankara authored
This CL establishes a naming convention to increase readability of the code. Bug: v8:10667, v8:10644 Change-Id: Id9fc9c558ba3957132b5a09aa975f9e8873c9dfd Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2297475 Commit-Queue: Zeynep Cankara <zcankara@google.com> Reviewed-by:
-
Michael Achenbach authored
Bug: v8:10680 Change-Id: I3a6055372b757fac4c5e28840536d1389e857437 Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2297381Reviewed-by:
Tamer Tas <tmrts@chromium.org> Commit-Queue: Michael Achenbach <machenbach@chromium.org> Cr-Commit-Position: refs/heads/master@{#68868}
-
Zeynep Cankara authored
This CL attaches global functions of the panels as instance variables of an app class. Bug: v8:10667, v8:10644 Change-Id: Ib76730652f977ac81e3558ddb18165e938859512 Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2297476 Commit-Queue: Zeynep Cankara <zcankara@google.com> Reviewed-by:
-
Jakob Kummerow authored
In addition to decoding them, we also have to evaluate the initializer instructions when instantiating a module. Drive-by fix: use "big-endian" encoding (prefix comes first) when emitting initializers in the module builder. Bug: v8:7748 Change-Id: Idfa0f5db298a8f6c6100fc09e1984e4a2e170e4a Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2298004 Commit-Queue: Jakob Kummerow <jkummerow@chromium.org> Reviewed-by:
Andreas Haas <ahaas@chromium.org> Cr-Commit-Position: refs/heads/master@{#68866}
-
Zhao Jiazhong authored
mjsunit/regress/regress-896326.js failed on mips simulator, because mips simulator has larger stack size and won't throw the expected RangeError exception. This CL set sim-stack-size to 100K in regress-896326 just like setting the native machine's stack-size. Change-Id: I51328b10a7b54addab2adb90401680c0581d7ee2 Bug: v8:10709 Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2299880 Commit-Queue: Zhao Jiazhong <zhaojiazhong-hf@loongson.cn> Reviewed-by:
-
Zeynep Cankara authored
This CL maps the colors being used in the web app to variables to make it easier to change color palette without having to track color changes across files. Bug: v8:10673 Change-Id: Icf1c53396b8a831367c4ed420931e6233d780a07 Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2298005 Commit-Queue: Zeynep Cankara <zcankara@google.com> Reviewed-by:
-
Jakob Gruber authored
When creating a new JSFunction (either through Factory::NewFunction or the FastNewClosure builtin), install the cached Code object if one exists. In the former, this happens explicitly; in the former implicitly through %CompileLazy. Drive-by: Clean up nci tracing methods. Drive-by: Rename maybe_has_... to may_have_cached_code. Bug: v8:8888 Change-Id: I98d87df10df496c45749b3fd76c072c36af45b68 Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2294662Reviewed-by:
-
Anton Bikineev authored
With CagedHeapLocalData, size of allocatable area in the caged heap has reduced, but CL that introduced it didn't change the size passed to BoundedPageAllocator. Change-Id: I3720820589c88c3467af68cd7da1b305dc5a77b4 Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2297474Reviewed-by:
Omer Katz <omerkatz@chromium.org> Commit-Queue: Anton Bikineev <bikineev@chromium.org> Cr-Commit-Position: refs/heads/master@{#68862}
-
Daniel Bevenius authored
Change-Id: Icccd7d8d0701ddbf9fb5088f5a47caa992f1b1e2 Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2120296Reviewed-by:
Marja Hölttä <marja@chromium.org> Commit-Queue: Marja Hölttä <marja@chromium.org> Cr-Commit-Position: refs/heads/master@{#68861}
-
Igor Sheludko authored
This is a follow-up fix for https://chromium-review.googlesource.com/c/v8/v8/+/2292230 In this CL fixes the case when the property cell is added to the dictionary but the value is not actually stored which leaves PropertyCell with the hole in the dictionary. Now the logic for GlobalDictionary matches the logic for NameDictionary - the property cell is added to the dictionary in LookupIterator::ApplyTransitionToDataProperty(). Bug: chromium:1104711, chromium:1105383 Change-Id: I56da16d85d13288fbc41fd60dbce556fec5e7d18 Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2297472Reviewed-by:
-
Zeynep Cankara authored
This CL fixes the Map Panel colors. The getColor function of Edge class was always falling to the default color unable to show correct colors in the timeline panel. Change-Id: Ide13b35703a656251222f512b2b9282f9f34cc04 Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2297473Reviewed-by:
-
Clemens Backes authored
Instead of storing a weak pointer per isolate, store exactly one weak pointer to the native module per engine. This is a small preparation for switching to the jobs API. R=ahaas@chromium.org Bug: chromium:1101340 Change-Id: I5f6590421c890998aa95a0a3b34596f59f2f2690 Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2297471Reviewed-by:
-
v8-ci-autoroll-builder authored
Rolling v8/build: https://chromium.googlesource.com/chromium/src/build/+log/b8f14c0..7a85719 Rolling v8/third_party/aemu-linux-x64: sruBLlVcbmt9F1gAt88JIykKImWZUcWy2HLp9bT1IBYC..7UW9Qc56aQ1K23_T0oZ9qoVs9Hbuka0eN2NRy-z2UV0C Rolling v8/third_party/catapult: https://chromium.googlesource.com/catapult/+log/09e01d1..49d9f03 Rolling v8/third_party/depot_tools: https://chromium.googlesource.com/chromium/tools/depot_tools/+log/6897807..35dec5f Rolling v8/tools/clang: https://chromium.googlesource.com/chromium/src/tools/clang/+log/d027d75..58914a5 TBR=machenbach@chromium.org,tmrts@chromium.org,v8-waterfall-sheriff@grotations.appspotmail.com Change-Id: I2e501280a1420e18f8c0ed2f807244786b49209b Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2298363Reviewed-by:
v8-ci-autoroll-builder <v8-ci-autoroll-builder@chops-service-accounts.iam.gserviceaccount.com> Commit-Queue: v8-ci-autoroll-builder <v8-ci-autoroll-builder@chops-service-accounts.iam.gserviceaccount.com> Cr-Commit-Position: refs/heads/master@{#68857}
-
Zhao Jiazhong authored
Port d0e6ff15 https://crrev.com/c/2285149 Port 34871edd https://crrev.com/c/2284212 Port dc82799d https://crrev.com/c/2290623 Change-Id: I8cceface23368dafc6a029edaa7c6a125a0760ed Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2291306 Commit-Queue: Zhao Jiazhong <zhaojiazhong-hf@loongson.cn> Reviewed-by:
Zhi An Ng <zhin@chromium.org> Cr-Commit-Position: refs/heads/master@{#68856}
-
- 14 Jul, 2020 22 commits
-
-
Ng Zhi An authored
Change-Id: I946c9f0db1dcb91ab9414be2de8285444741ca3e Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2293499 Commit-Queue: Zhi An Ng <zhin@chromium.org> Reviewed-by:
Bill Budge <bbudge@chromium.org> Cr-Commit-Position: refs/heads/master@{#68855}
-
Frank Tang authored
Make locales and options required and no default for type in options. Bug: v8:10623 Change-Id: I5df065a95e82ecb3b8b036d1b4738f296aa7243f Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2291617Reviewed-by:
Jakob Kummerow <jkummerow@chromium.org> Commit-Queue: Frank Tang <ftang@chromium.org> Cr-Commit-Position: refs/heads/master@{#68854}
-
Dominik Inführ authored
Fix data race between concurrent threads allocating (accessing gc_state_ that way) and the main thread starting tear down. Bug: v8:10315 Change-Id: Icc24811e43268512c8d7fdaf92ecd3fc7b3ecd57 Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2297390Reviewed-by:
-
Ross McIlroy authored
Adds basic framework to pipeline.cc to enable a seperate fast register allocator for the TurboProp mid-tier. As part of this, common logic as well as a base class for RegisterAllocationData is moved to a seperate register-allocation.h header file. The current register allocator's RegisterAllocationData is renamed to TopTierRegisterAllocationData, and the former name is the new base class held in PipelineData. BUG=v8:9684 Change-Id: I28285b7d6112505bf90e88ea3cda66d03dfabc74 Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2295359 Commit-Queue: Ross McIlroy <rmcilroy@chromium.org> Reviewed-by:
-
Daniel Bevenius authored
Currently, when specifying '--help' with mksnapshot it will only print the v8/d8 help message and options and then exit the process. This means that the usage message from mksnapshot will never be displayed. This commit suggests adding an option to SetFlagsFromCommandLine that can disable this printing and exiting. This allows mksnapshot to display the usage and print the options after that. While this works, it does seems a little strange that SetFlagsFromCommandLine prints the help message and exits the process but I'm probably missing some background details around this. Change-Id: I28932adf3478b88b05eed4db70bf74946f8abf2d Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2290852Reviewed-by:
Dan Elphick <delphick@chromium.org> Reviewed-by:
-
Ng Zhi An authored
There is a sign-extension bug happening when packing 2 32-bit ints into a 64-bit int. We are OR-ing int32_t with a uint64_t, so an integral conversion converts int32_t to uint64_t, which is a sign extension, and this gives unexpected results for a negative value: 0x80000000 | uint64_t{0} -> 0xffffffff80000000 What we want is 0x0000000080000000. Created a helper function to do this work of combining two uint32_t into one uint64_t. The use of this function will also ensure that if callers passed a int32_t, it would first be converted to a uint32_t, and will not have this sign extension bug. Sneaked a small regression test into the existing v128.const cctest, and also cleanup the loop to reset `expected` array to 0. Bug: chromium:1104033 Change-Id: Icaca4c5ba42077dd4463697b9220cdbca9974b5e Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2293044 Commit-Queue: Zhi An Ng <zhin@chromium.org> Reviewed-by:Clemens Backes <clemensb@chromium.org> Reviewed-by:
-
Zeynep Cankara authored
This CL modifies the logging pipeline of V8 to track timestamps of the IC events across the log file. Modifies the current IC-explorer's code to make it compatible with the IC event time processing. Change-Id: I2a0f652e2657bdebe8cecd7862a7545f7b050cdb Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2274613 Commit-Queue: Zeynep Cankara <zcankara@google.com> Reviewed-by:
Sathya Gunasekaran <gsathya@chromium.org> Cr-Commit-Position: refs/heads/master@{#68849}
-
Ulan Degenbaev authored
The host object may have an impossible markbit pattern if it is a one-word filler followed by an already marked object. Bug: v8:10698 Change-Id: I498e6f0768fbdb181fc893f98f224dd3cd0e37e6 Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2295600Reviewed-by:
Dominik Inführ <dinfuehr@chromium.org> Commit-Queue: Ulan Degenbaev <ulan@chromium.org> Cr-Commit-Position: refs/heads/master@{#68848}
-
Zeynep Cankara authored
This CL identifies dependencies between modules and convert existing javascript files to ES6 standard modules. It cleans the unused code and remove duplicate code throughout the app. Bug: v8:10670 Change-Id: I787de8ca0d76c56aec5aeb3faa94a9e158a94c72 Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2292237 Commit-Queue: Zeynep Cankara <zcankara@google.com> Reviewed-by:
-
Maya Lekova authored
This reverts commit 2a1abac5. Reason for revert: Breaking Arm CFI bot - https://ci.chromium.org/p/v8/builders/ci/V8%20Linux%20-%20arm64%20-%20sim%20-%20CFI/1354? Original change's description: > [Atomics.waitAsync] Implement Atomics.waitAsync > > Original design doc: > https://docs.google.com/document/d/1dthXsVHMc1Sd_oYf9a-KZSFOd_a8dUgnt4REAG8YIXA > > Design changes: > https://docs.google.com/document/d/1aeEGDm1XSqoJkQQKz9F75WqnuAa2caktxGy_O_KpO9Y > > > Bug: v8:10239 > Change-Id: Iab94ccab85d7b4ff23cff1955774b42edf5be541 > Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2202981 > Commit-Queue: Marja Hölttä <marja@chromium.org> > Reviewed-by: Igor Sheludko <ishell@chromium.org> > Reviewed-by: Andreas Haas <ahaas@chromium.org> > Reviewed-by: Ulan Degenbaev <ulan@chromium.org> > Reviewed-by: Shu-yu Guo <syg@chromium.org> > Cr-Commit-Position: refs/heads/master@{#68844} TBR=ulan@chromium.org,marja@chromium.org,ahaas@chromium.org,ishell@chromium.org,syg@chromium.org Change-Id: I1a1164ab29112bd0113b8b1823c78a3895cfd6cc No-Presubmit: true No-Tree-Checks: true No-Try: true Bug: v8:10239 Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2297469Reviewed-by:
Maya Lekova <mslekova@chromium.org> Commit-Queue: Maya Lekova <mslekova@chromium.org> Cr-Commit-Position: refs/heads/master@{#68846}
-
Zeynep Cankara authored
This CL adds input fields to the IC Panel to filter IC events based on the event creation time. Filtered events across time reflected back to the IC-panel statistics which helps to examine statistics about the events in the selected time range. Change-Id: Ib2d66caab25140b09daa4d6249758254f8c75ce8 Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2295601Reviewed-by:
-
Marja Hölttä authored
Original design doc: https://docs.google.com/document/d/1dthXsVHMc1Sd_oYf9a-KZSFOd_a8dUgnt4REAG8YIXA Design changes: https://docs.google.com/document/d/1aeEGDm1XSqoJkQQKz9F75WqnuAa2caktxGy_O_KpO9Y Bug: v8:10239 Change-Id: Iab94ccab85d7b4ff23cff1955774b42edf5be541 Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2202981 Commit-Queue: Marja Hölttä <marja@chromium.org> Reviewed-by:
Igor Sheludko <ishell@chromium.org> Reviewed-by:
Shu-yu Guo <syg@chromium.org> Cr-Commit-Position: refs/heads/master@{#68844}
-
Dominik Inführ authored
Before actually failing to allocate, let the background thread help to sweep all pages of that space. As a drive-by also rename allocation functions to make background and main thread allocation more similar. Bug: v8:10315 Change-Id: I26d4b622de949d4943e35071cee1df8b3d2889c2 Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2297383Reviewed-by:
-
Dominik Inführ authored
Help sweeper tasks complete sweeping sooner but do not refill free lists on shutdown. This races with allocating background threads. Background threads will refill free lists themselves if more memory is required. Bug: v8:10315 Change-Id: Ie615983229701e8c9434b4352bd055e9dbbb8671 Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2297466Reviewed-by:
-
Santiago Aboy Solanes authored
Bug: v8:7703 Change-Id: I685ad610e29f277381103c4be2c374bf0b7336ef Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2294979 Auto-Submit: Santiago Aboy Solanes <solanes@chromium.org> Reviewed-by:
-
Jakob Kummerow authored
The named LoadIC code was missing a check for "names" that convert to TypedArray indices. This was flushed out by the recent bump of the max TypedArray size from 2^32-1 to 2^32. Named StoreICs had the same bug; fixed here as well. Bug: v8:4153 Fixed: chromium:1104608 Change-Id: I6bd2552d6ccc238104f92e7b95d19970d4a75dae Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2295606Reviewed-by:
-
Dominik Inführ authored
This is a reland of fc48a228 Original change's description: > Reland "[heap] Move start of incremental marking in allocation" > > This is a reland of d6a14abe > > Test wasn't written with incremental/concurrent marking in mind, so > simply disabling it for this particular unittest. > > Original change's description: > > [heap] Move start of incremental marking in allocation > > > > Move start of incremental marking out of > > RefillLinearAllocationAreaFromFreeList. This avoids a potential > > safepoint while holding allocation_mutex_. > > > > Bug: v8:10315 > > Change-Id: Ieb60ac68f26199eea7b6b7ad6d874851382f3d69 > > Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2287496 > > Commit-Queue: Dominik Inführ <dinfuehr@chromium.org> > > Reviewed-by: Ulan Degenbaev <ulan@chromium.org> > > Cr-Commit-Position: refs/heads/master@{#68751} > > Bug: v8:10315 > Change-Id: I2a665400d9a784b1557474a051839d5c8b45e9e2 > Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2292241 > Reviewed-by: Ulan Degenbaev <ulan@chromium.org> > Commit-Queue: Dominik Inführ <dinfuehr@chromium.org> > Cr-Commit-Position: refs/heads/master@{#68818} Bug: v8:10315 Change-Id: I7873c6c20e39d6636bd95a26d0c1cfc8f89366bd Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2295363Reviewed-by:
-
Santiago Aboy Solanes authored
For the first test, we just test that we can search on two threads at the same time. This CL sets the base for the future tests for more complicated cases. Bug: v8:7790 Change-Id: I1becf4493897b55e7ee0a7f37ab5bf1203bf14eb Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2241530 Commit-Queue: Santiago Aboy Solanes <solanes@chromium.org> Reviewed-by:
Michael Stanton <mvstanton@chromium.org> Cr-Commit-Position: refs/heads/master@{#68838}
-
Jakob Gruber authored
... on Code objects. Refactors: create a dedicated WasmCode constructor, hide the internal constructor, constify members, and let SafepointTable handle out-of-line tables. Expose a new Code::SafepointTableAddress() helper as the source of truth. Some safepoint tables may move out-of-line in the near future. Bug: v8:7777,v8:10707 Change-Id: I4e2d954ed2d157235e9dfa3e7a5ca08800896683 Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2297459Reviewed-by:
-
Santiago Aboy Solanes authored
This CL adds functionality to read the source positions directly from the JS heap rather than from serialized data. In order to do this, we create a PersistentHandles container in the OptimizedCompilationInfo which gets passed onto the JSHeapBroker. This allows us to create the handles in the main thread and pass them safely to the background thread. In order to read safely from the background thread, we need a LocalHeap which blocks the GC from running and potentially moving the handles. This LocalHeap is created only when the JSHeapBroker has finalized serializing and destroyed when retiring it. Bug: v8:7790 Change-Id: I19f8b08d12e5be0a3df34d6af2043310c0c7b6fe Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2277802Reviewed-by:
-
Jakob Gruber authored
Handler tables (and other inlined Code metadata) will have to move outside the .text section. This CL creates Code::HandlerTableAddress() as a single chokepoint for accessing the handler table of a Code object. Drive-by: Create a dedicated constructor for WasmCode handler tables. Bug: v8:7777 Change-Id: I01c5157b732ba509b2c76f2744fde271c2ba1411 Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2295605 Commit-Queue: Jakob Gruber <jgruber@chromium.org> Commit-Queue: Clemens Backes <clemensb@chromium.org> Auto-Submit: Jakob Gruber <jgruber@chromium.org> Reviewed-by:
-
Zhao Jiazhong authored
The Sw/Sd in FillStackSlotsWithZero should't use kSystemPointerSize as address offset, because the start address should be inclusive, and the end address should be exclusive. The skip-stack-guard-page test case failed due to this bug, and besides, it also needs larger stack size on mips simulator. Change-Id: Ieff55fe2c5a13e6dad1c5d073e1c0d22fe789d41 Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2282663 Commit-Queue: Zhao Jiazhong <zhaojiazhong-hf@loongson.cn> Reviewed-by:
-
