summaryrefslogtreecommitdiffstats
path: root/js
Commit message (Collapse)AuthorAgeLines
* Use C++11 thread-safe statics with MSVC. (js)wolfbeast2019-02-14-4/+0
|
* Fix possible data race while updating scope object during compacting GC.wolfbeast2019-02-07-3/+8
|
* Align `instanceof` with the final ES6 spec.wolfbeast2019-02-07-18/+79
|
* Merge branch 'cycle_collector-work'wolfbeast2019-02-03-3/+0
|\
| * Merge branch 'master' into cycle_collector-workwolfbeast2019-01-29-6/+21
| |\
| * | Remove NS_IMPL_CYCLE_COLLECTION_TRAVERSE_SCRIPT_OBJECTSwolfbeast2019-01-18-3/+0
| | |
* | | Fix build error with JS_HAS_TOSOURCE undefined.wolfbeast2019-02-02-1/+4
| | |
* | | Remove telemetry leftovers from JS engine.adeshkp2019-01-30-461/+63
| |/ |/|
* | Reduce number of allocations in AutoStopwatchDavid Teller2019-01-29-6/+21
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | This patch fixes two related issues. 1. The AutoStopwatch uses a stack-allocated `mozilla::Vector` to communicate with its callback during each compartment switch. This vector was designed to allow its contents to be stack-allocated but they turned out to be accidentally heap-allocated. 2. During each tick, the stopwatch fills a vector `recentGroups_`. This vector always started with minimal capacity and had to grow repeatedly as groups were added, causing repeated reallocations. This patch preallocates `recentGroups_` to have the same capacity as the previous tick. We expect that this should eventually reach a stable size that closely matches the actual needs of the process.
* | Revert "Reduce number of allocations in AutoStopwatch"wolfbeast2019-01-29-18/+8
| | | | | | | | This reverts commit 3476c1d60ec29c5497123194acd7a9310b1023d2.
* | Reduce number of allocations in AutoStopwatchDavid Teller2019-01-28-8/+18
|/ | | | | | | | | | | | | | | This patch fixes two related issues. 1. The AutoStopwatch uses a stack-allocated `mozilla::Vector` to communicate with its callback during each compartment switch. This vector was designed to allow its contents to be stack-allocated but they turned out to be accidentally heap-allocated. 2. During each tick, the stopwatch fills a vector `recentGroups_`. This vector always started with minimal capacity and had to grow repeatedly as groups were added, causing repeated reallocations. This patch preallocates `recentGroups_` to have the same capacity as the previous tick. We expect that this should eventually reach a stable size that closely matches the actual needs of the process.
* Telemetry: Remove stubs and related codeadeshkp2019-01-12-49/+0
|
* Restrict ExtractLinearSum to find non-wrapping results.wolfbeast2018-12-13-1/+1
|
* Use canonical function in TypeNewScript::rollbackPartiallyInitializedObjects.wolfbeast2018-12-13-7/+26
|
* Fix mistakingly flagged instructions.wolfbeast2018-12-12-1/+1
| | | | This enables optimizations which were wrongly inhibited before by this typo.
* Fix a longstanding IndexedDB correctness issue.wolfbeast2018-11-02-4/+20
| | | | Standards Compliance fix, port of Bug 1492737
* Simplify SIMD conversion from Uint64 to Double.wolfbeast2018-11-01-31/+21
|
* Don't mark MGetFirstDollarIndex as movable.wolfbeast2018-10-31-1/+4
|
* Code style improvements (no bug).wolfbeast2018-10-23-9/+18
|
* Improve graph edge resolution code.wolfbeast2018-10-23-17/+48
| | | | This is a follow-up to ca7ecd37c94e268972697a37eec4e46771c6e6f2 further improving the DiD resolution for CVE-2018-12386.
* Bug 1493900.wolfbeast2018-10-04-3/+12
| | | | DiD fix.
* Rename TypeSet::clone to TypeSet::cloneIntoUninitialized to indicate that it ↵trav902018-09-19-9/+11
| | | | | | freshly initializes the TemporaryTypeSet* provided to it. Also removes existing code that, quite unnecessarily, partly initialized that argument.
* Don't use PodCopy/PodMove to implement typed-array element-to-element copyingtrav902018-09-19-4/+17
| | | | Standard std::copy and std::copy_n are readily optimized to the same thing, and they don't have a non-obvious requirement that the type being copied be trivial.
* Give uint8_clamped a defaulted (and also trivial) default constructor, copy ↵trav902018-09-19-6/+3
| | | | | | constructor, and copy-assignment operator. This also allows uint8_clamped to be permissibly memmove'd and memcpy'd.
* Don't memset-zero the BacktrackingAllocator::vregs array of non-trivial ↵trav902018-09-12-15/+11
| | | | VirtualRegister instances
* Call memset on a void*, not a T*, in js_delete_poison to avoid ↵trav902018-09-12-1/+1
| | | | memset-on-nontrivial warnings with gcc that don't matter for an object whose lifetime is about to end
* Initialize some asm.js structures using in-class initializers instead of PodZerotrav902018-09-12-9/+9
|
* Stop using PodZero in several places to initialize values of non-trivial typetrav902018-09-12-73/+56
|
* Merge pull request #754 from trav90/class-memaccess-errorsMoonchild2018-09-10-95/+180
|\ | | | | Fix more -Wclass-memaccess warnings (GCC8)
| * Fix build bustagetrav902018-09-02-1/+1
| |
| * Call the relevant scope-data constructor when allocating it, and poison/mark ↵trav902018-09-02-31/+72
| | | | | | | | as undefined the memory for the trailing array of BindingNames, ratther than impermissibly PodZero-ing non-trivial classes.
| * Convert the trailing array of BindingNames at the end of the various kinds ↵trav902018-09-02-64/+108
| | | | | | | | of scope data into raw unsigned chars into which those BindingNames are placement-new'd, rather than memcpy-ing non-trivial classes around and failing to comply with the C++ object model
* | Merge branch 'master' of https://github.com/MoonchildProductions/UXPwolfbeast2018-09-04-101/+1
|\ \
| * | Remove all C++ Telemetry Accumulation calls.wolfbeast2018-09-03-101/+1
| |/ | | | | | | | | This creates a number of stubs and leaves some surrounding code that may be irrelevant (eg. recorded time stamps, status variables). Stub resolution/removal should be a follow-up to this.
* / Add extra check for assembler buffer space.wolfbeast2018-09-04-1/+2
|/
* Revert "Bug 1444668 - Avoid allocating large AssemblerBuffers. r=luke, ↵wolfbeast2018-08-30-49/+9
| | | | | | r=bbouvier, a=RyanVM" This reverts commit 9472136272f01b858412f2d9d7854d2daa82496f.
* Avoid using memset on a not-trivial type like ServoSizestrav902018-08-18-1/+9
|
* Avoid using memset on a not-trivial type like TabSizestrav902018-08-18-1/+7
|
* Avoid doing a memset on a non-POD structuretrav902018-08-18-13/+33
| | | | | | |entryCount| tracks -- in fast-to-check manner -- the number of entries in the hashtable. But to actually enumerate entries, we have to loop through all of |table|, checking for entries that are actually live. A live entry is indicated by a zero |hash| in the entry. The |memset| would properly zero that; removing the memset will not. It's not entirely clear whether a memset that overwrites a lot of stuff but is maybe simpler, is faster than compiler-generated likely-SIMD code that zeroes out *just* |hash| fields in all the entries. But I am going to guess that SIMD is good enough. For now, we should just do the simple and thing: don't distinguish POD and non-POD, and know that the compiler is going to recognize that |mem.addr()->~T()| is a no-op when T is trivial. So with POD, the loop should degenerate to just zeroing |hash| at consistent offset, and SIMD will eat that up, and it can't be *that* different from the memset in performance (if it is at all).
* Merge pull request #707 from trav90/class-memaccess-errorsMoonchild2018-08-13-28/+10
|\ | | | | Simplify HeapSlot to make it trivially copyable
| * Simplify HeapSlot to make it trivially copyabletrav902018-08-12-29/+8
| | | | | | | | This removes the constructors, which were never called since we allocate arrays of HeapSlot with pod_malloc. The destructor is only ever called explicitly since we free this memory with js_free so it has been renamed to destroy(). Also removed is an unused manual barrier.
| * Avoid using memcpy on HeapSlot that is not trivially copyable.trav902018-08-12-4/+7
| |
* | js::atomics_wait: Remove unnecessary parentheses in declaration of 'addr'trav902018-08-10-1/+1
|/ | | | Silences a warning with GCC 8.
* Merge pull request #628 from janekptacijarabaci/build_bustage_debug_1Moonchild2018-07-13-0/+3
|\ | | | | Fix Build Bustage - with "--enable-debug"
| * Fix Build Bustage - with "--enable-debug"janekptacijarabaci2018-07-12-0/+3
| |
* | Remove recover instruction results after bailouts.wolfbeast2018-07-13-38/+20
|/
* Issue #578: Applications cannot start without /proc (chroot).wolfbeast2018-07-02-3/+65
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | UXP uses the current stack frame address and the stack size as a sort of heuristic for various things in the JavaScript engine. The js::GetNativeStackBaseImpl() function is used to get the base stack address (i.e. the address from which the stack grows, so this can be either the first or last memory address of the stack memory space depending on the CPU architecture). On Linux, this function is implemented using the pthreads APIs. For non-main threads, the queried thread info is stored in memory. The main thread does not have this information on hand, so it gets the stack memory range via the /proc/self/maps file (see glibc's pthread_get_attr_np.c). Fortunately (per discussions with the firefox devs in #jsapi) the base address only needs to be approximate. In reality, environment variables, args, and other things are stored in stack space between the end/beginning of the mapped stack memory and the 'top' of the stack space used by stack frames. When using glibc, we can get the top of this usable stack from __libc_stack_end, which is a void* set by glibc during program initialization, avoiding the need to access /proc. Non-main threads still get their stack-base through the usual pthreads APIs. Other libc implementations like musl will fall back to the standard UNIX-like implementation which calls pthread's pthread_attr_getstack() also from the main thread, which may imply /proc access and not work in restricted environments.
* Bug 1464829 - Ensure the recover instruction vector has the expected size.Nicolas B. Pierron2018-06-07-5/+1
|
* Bug 1465108 - Use function pointers rather than virtual run method for GC ↵Jon Coppeard2018-06-07-37/+72
| | | | parallel tasks r=sfink a=abillings a=RyanVM
* Fix count of compacting update tasks started.wolfbeast2018-06-07-1/+1
|
generated by cgit v1.2.3 (git 2.25.1) at 2024-11-19 15:23:31 +0000