issue_owner_repo
listlengths 2
2
| issue_body
stringlengths 0
261k
⌀ | issue_title
stringlengths 1
925
| issue_comments_url
stringlengths 56
81
| issue_comments_count
int64 0
2.5k
| issue_created_at
stringlengths 20
20
| issue_updated_at
stringlengths 20
20
| issue_html_url
stringlengths 37
62
| issue_github_id
int64 387k
2.46B
| issue_number
int64 1
127k
|
|---|---|---|---|---|---|---|---|---|---|
[
"chakra-core",
"ChakraCore"
] |
Hi,
One of our deployed website is slow because of several reasons. We are constantly trying to fix the latency issues. Our end users do not have much concerns about latency but missing controls like buttons etc while using IE11 and Edge. They have to refresh the page again to get the controls to be visible. This is not the case with Chrome or Firefox.
Looks like the way IE or Edge interprets javascript code and looking for dependent files is not efficient in slow connections. This is our observation so far.
Please give us any hints/suggestions to resolve this during slow connections if there are settings to tuneup at browser level.There are around 160 CSS & JS files downloaded while accessing our site and it takes 20 to 40 secs on average.
you can try accessing our website at https://connex.ngsmedicare.com
|
IE11 does not render controls when there is a slow connection but Chrome and Firefox do
|
https://api.github.com/repos/chakra-core/ChakraCore/issues/4142/comments
| 2 |
2017-11-05T11:26:31Z
|
2017-11-06T22:32:40Z
|
https://github.com/chakra-core/ChakraCore/issues/4142
| 271,269,245 | 4,142 |
[
"chakra-core",
"ChakraCore"
] |
Example:
```js
Object.getOwnPropertyDescriptor(function (a, b) { return 0; }, "length").configurable // true
Object.getOwnPropertyDescriptor(function (a, b) { return 0; }.bind({}), "length").configurable // false
```
This is blocking tests 10.3.2_1_a_L15.js, 12.3.2_1_a_L15, 11.3.2_1_a_L15, 10.2.2_L15.js, 12.2.2_L15.js, and 11.2.2_L15.js in test262/test/intl402
/cc @bterlson
|
Functions returned by Function.prototype.bind() have non-configurable lengths
|
https://api.github.com/repos/chakra-core/ChakraCore/issues/4122/comments
| 7 |
2017-11-02T18:54:54Z
|
2018-08-20T00:59:32Z
|
https://github.com/chakra-core/ChakraCore/issues/4122
| 270,767,656 | 4,122 |
[
"chakra-core",
"ChakraCore"
] |
See: https://github.com/Microsoft/ChakraCore/pull/4071
Not doing this in the PR itself because getting templates to compile across MSVC and clang is nitpicky and this is a minor point in refactoring that doesn't affect runtime performance or memory usage.
|
RegExp Bytecode: (follow-up #4071) Refactor Switch##n macros to templates.
|
https://api.github.com/repos/chakra-core/ChakraCore/issues/4117/comments
| 0 |
2017-11-01T22:50:02Z
|
2019-06-07T19:05:52Z
|
https://github.com/chakra-core/ChakraCore/issues/4117
| 270,478,214 | 4,117 |
[
"chakra-core",
"ChakraCore"
] |
Hi, we recently started fuzzing Chakra using [OSS-Fuzz](https://github.com/google/oss-fuzz) (google/oss-fuzz#925), and have discovered a number of crashes. Some of them are potentially security vulnerabilities (buffer overflows, bad address dereferences etc).
We'd like to report these bugs by asking Chakra developers/maintainers to send a PR to add their emails (needs to be associated with a [Google account](https://github.com/google/oss-fuzz/blob/master/docs/faq.md#why-do-you-require-a-google-account-for-authentication)) to [this file](https://github.com/google/oss-fuzz/blob/master/projects/chakra/project.yaml) ([example](https://github.com/google/oss-fuzz/blob/master/projects/freetype2/project.yaml)) so that you are CC'ed on current and future bug reports in our issue tracker ([disclosure policy](https://github.com/google/oss-fuzz#bug-disclosure-guidelines)).
Our infrastructure comes with regression range testing, fix verification, de-duplication, testcase minimization, and more.
Thanks!
|
Reportings bugs found by OSS-Fuzz
|
https://api.github.com/repos/chakra-core/ChakraCore/issues/4112/comments
| 10 |
2017-11-01T11:41:22Z
|
2017-11-30T23:35:01Z
|
https://github.com/chakra-core/ChakraCore/issues/4112
| 270,282,154 | 4,112 |
[
"chakra-core",
"ChakraCore"
] |
- ~~`BuiltInPropertyRecords::XXXXXXXX.Equals` calls use 1:1 string comparison. Worst case use `PropertyRecord` from specialized strings and compare the propertyId~~ DONE #4123
- ~~`String::Equals` may share the content once `String1 == String2` Further comparison of those strings would be instant and we could GC the unnecessary copy. This needs more work.~~ DONE #4163
- `CharacterBuffer` internal class usage in TypeHandler Get/Set etc. calls require a quick investigation. Overall, some usage of `CharacterBuffer` can be replaced with propertyId from specialized string.
- We do convert from CHAR16 to CHAR for almost everything (JSRT). Experiment replacing CHAR16 to CHAR and see the gain. In other words, see what would happen if ChakraCore internally use CHAR type.
- CompoundString alternative for concatenating two strings ?
|
perf: Improve string usage
|
https://api.github.com/repos/chakra-core/ChakraCore/issues/4110/comments
| 2 |
2017-11-01T01:46:54Z
|
2018-01-25T23:55:00Z
|
https://github.com/chakra-core/ChakraCore/issues/4110
| 270,169,008 | 4,110 |
[
"chakra-core",
"ChakraCore"
] |
Open with MS Edge:
http://esenthel.com/site/Esenthel%20Fantasy%20Demo/Esenthel%20Fantasy%20Demo.html
And get crash:
```
xception thrown: RuntimeError: Memory index is out of range,RuntimeError: Memory index is out of range
at Module[dynCall_v] (http://esenthel.com/site/Esenthel%20Fantasy%20Demo/Esenthel%20Fantasy%20Demo.js:1:374378)
at browserIterationFunc (http://esenthel.com/site/Esenthel%20Fantasy%20Demo/Esenthel%20Fantasy%20Demo.js:1:174851)
at Browser.mainLoop.runIter (http://esenthel.com/site/Esenthel%20Fantasy%20Demo/Esenthel%20Fantasy%20Demo.js:1:177943)
at Browser_mainLoop_runner (http://esenthel.com/site/Esenthel%20Fantasy%20Demo/Esenthel%20Fantasy%20Demo.js:1:176388)
SCRIPT7017: SCRIPT7017: Memory index is out of range
```
Works fine on Chrome, FireFox and Opera.
|
Emscripten WebGL JS WASM crash
|
https://api.github.com/repos/chakra-core/ChakraCore/issues/4099/comments
| 12 |
2017-10-31T08:16:26Z
|
2018-11-30T00:31:05Z
|
https://github.com/chakra-core/ChakraCore/issues/4099
| 269,870,903 | 4,099 |
[
"chakra-core",
"ChakraCore"
] |
1 - How we allocate the memory
ChakraCore's internal memory managers reserve (i.e. Arena Allocator reserves 128KB memory) good amount of memory at once. However, the problem is, the managers commit the memory per consumer.
This corresponds to lots of calls to mmap on *nix. Neither `mmap` or PAL's VirtualAlloc are good candidates for often calls.
2 - We use VirtualProtect unnecessarily often
We do allocate with write/read rights, and then set no access, and then set read/write again. Similarly, `mprotect` and PAL's VirtualProtect calls are not something we should call per each allocation.
|
Memory management improvements neeeded
|
https://api.github.com/repos/chakra-core/ChakraCore/issues/4093/comments
| 0 |
2017-10-30T18:19:06Z
|
2018-05-07T23:53:21Z
|
https://github.com/chakra-core/ChakraCore/issues/4093
| 269,703,377 | 4,093 |
[
"chakra-core",
"ChakraCore"
] |
Tried to make a 32-bit ChakraCore build for Linux (Ubuntu), but the build failed. I was able to build for x64 without a hitch, but x86 proved futile.
Command line:
```
fatcerberus@pigcult-vm:~/src/ChakraCore$ ./build.sh --embed-icu -j=4 --lto --arch=x86
```
Gave me the following output:
```
lto: ranlib disabled
Custom CXX /home/fatcerberus/src/ChakraCore/cc-toolchain/build/bin/clang++
Custom CC /home/fatcerberus/src/ChakraCore/cc-toolchain/build/bin/clang
Build path: /home/fatcerberus/src/ChakraCore/out/Release
-- /home/fatcerberus/src/ChakraCore/bin/ch/DbgController.js.h is up to date. skipping.
Compile Target : x86
Generating Release makefiles
-- found ICU libs: deps/icu/source/output/include//../lib/
-- Configuring done
-- Generating done
-- Build files have been written to: /home/fatcerberus/src/ChakraCore/out/Release
[ 0%] Building CXX object lib/Jsrt/Core/CMakeFiles/Chakra.Jsrt.Core.dir/JsrtContextCore.cpp.o
[ 1%] Building CXX object pal/src/CMakeFiles/Chakra.Pal.dir/cruntime/file.cpp.o
clang-4.0: error: argument unused during compilation: '-arch i386' [-Werror,-Wunused-command-line-argument]
lib/Jsrt/Core/CMakeFiles/Chakra.Jsrt.Core.dir/build.make:62: recipe for target 'lib/Jsrt/Core/CMakeFiles/Chakra.Jsrt.Core.dir/JsrtContextCore.cpp.o' failed
make[2]: *** [lib/Jsrt/Core/CMakeFiles/Chakra.Jsrt.Core.dir/JsrtContextCore.cpp.o] Error 1
CMakeFiles/Makefile2:1448: recipe for target 'lib/Jsrt/Core/CMakeFiles/Chakra.Jsrt.Core.dir/all' failed
make[1]: *** [lib/Jsrt/Core/CMakeFiles/Chakra.Jsrt.Core.dir/all] Error 2
make[1]: *** Waiting for unfinished jobs....
clang-4.0: error: argument unused during compilation: '-arch i386' [-Werror,-Wunused-command-line-argument]
pal/src/CMakeFiles/Chakra.Pal.dir/build.make:62: recipe for target 'pal/src/CMakeFiles/Chakra.Pal.dir/cruntime/file.cpp.o' failed
make[2]: *** [pal/src/CMakeFiles/Chakra.Pal.dir/cruntime/file.cpp.o] Error 1
make[2]: *** Waiting for unfinished jobs....
[ 1%] Building CXX object pal/src/CMakeFiles/Chakra.Pal.dir/cruntime/filecrt.cpp.o
[ 1%] Building CXX object pal/src/CMakeFiles/Chakra.Pal.Singular.dir/cruntime/filecrt.cpp.o
[ 1%] Building CXX object pal/src/CMakeFiles/Chakra.Pal.Singular.dir/cruntime/file.cpp.o
clang-4.0: error: argument unused during compilation: '-arch i386' [-Werror,-Wunused-command-line-argument]
pal/src/CMakeFiles/Chakra.Pal.dir/build.make:86: recipe for target 'pal/src/CMakeFiles/Chakra.Pal.dir/cruntime/filecrt.cpp.o' failed
make[2]: *** [pal/src/CMakeFiles/Chakra.Pal.dir/cruntime/filecrt.cpp.o] Error 1
CMakeFiles/Makefile2:142: recipe for target 'pal/src/CMakeFiles/Chakra.Pal.dir/all' failed
make[1]: *** [pal/src/CMakeFiles/Chakra.Pal.dir/all] Error 2
clang-4.0: error: argument unused during compilation: '-arch i386' [-Werror,-Wunused-command-line-argument]
pal/src/CMakeFiles/Chakra.Pal.Singular.dir/build.make:62: recipe for target 'pal/src/CMakeFiles/Chakra.Pal.Singular.dir/cruntime/file.cpp.o' failed
make[2]: *** [pal/src/CMakeFiles/Chakra.Pal.Singular.dir/cruntime/file.cpp.o] Error 1
make[2]: *** Waiting for unfinished jobs....
clang-4.0: error: argument unused during compilation: '-arch i386' [-Werror,-Wunused-command-line-argument]
pal/src/CMakeFiles/Chakra.Pal.Singular.dir/build.make:86: recipe for target 'pal/src/CMakeFiles/Chakra.Pal.Singular.dir/cruntime/filecrt.cpp.o' failed
make[2]: *** [pal/src/CMakeFiles/Chakra.Pal.Singular.dir/cruntime/filecrt.cpp.o] Error 1
[ 1%] Building CXX object pal/src/CMakeFiles/Chakra.Pal.Singular.dir/cruntime/finite.cpp.o
clang-4.0: error: argument unused during compilation: '-arch i386' [-Werror,-Wunused-command-line-argument]
pal/src/CMakeFiles/Chakra.Pal.Singular.dir/build.make:110: recipe for target 'pal/src/CMakeFiles/Chakra.Pal.Singular.dir/cruntime/finite.cpp.o' failed
make[2]: *** [pal/src/CMakeFiles/Chakra.Pal.Singular.dir/cruntime/finite.cpp.o] Error 1
CMakeFiles/Makefile2:105: recipe for target 'pal/src/CMakeFiles/Chakra.Pal.Singular.dir/all' failed
make[1]: *** [pal/src/CMakeFiles/Chakra.Pal.Singular.dir/all] Error 2
[ 1%] Building CXX object lib/Backend/CMakeFiles/Chakra.Backend.dir/AgenPeeps.cpp.o
[ 1%] Building CXX object lib/Backend/CMakeFiles/Chakra.Backend.dir/AsmJsJITInfo.cpp.o
[ 1%] Building CXX object lib/Backend/CMakeFiles/Chakra.Backend.dir/Backend.cpp.o
clang-4.0clang-4.0: : errorerror: : argument unused during compilation: '-arch i386' [-Werror,-Wunused-command-line-argument]argument unused during compilation: '-arch i386' [-Werror,-Wunused-command-line-argument]
lib/Backend/CMakeFiles/Chakra.Backend.dir/build.make:62: recipe for target 'lib/Backend/CMakeFiles/Chakra.Backend.dir/AgenPeeps.cpp.o' failed
make[2]: *** [lib/Backend/CMakeFiles/Chakra.Backend.dir/AgenPeeps.cpp.o] Error 1
make[2]: *** Waiting for unfinished jobs....
lib/Backend/CMakeFiles/Chakra.Backend.dir/build.make:86: recipe for target 'lib/Backend/CMakeFiles/Chakra.Backend.dir/AsmJsJITInfo.cpp.o' failed
make[2]: *** [lib/Backend/CMakeFiles/Chakra.Backend.dir/AsmJsJITInfo.cpp.o] Error 1
[ 1%] Building CXX object lib/Backend/CMakeFiles/Chakra.Backend.dir/BackendApi.cpp.o
clang-4.0: error: argument unused during compilation: '-arch i386' [-Werror,-Wunused-command-line-argument]
clang-4.0: error: argument unused during compilation: '-arch i386' [-Werror,-Wunused-command-line-argument]
lib/Backend/CMakeFiles/Chakra.Backend.dir/build.make:110: recipe for target 'lib/Backend/CMakeFiles/Chakra.Backend.dir/Backend.cpp.o' failed
make[2]: *** [lib/Backend/CMakeFiles/Chakra.Backend.dir/Backend.cpp.o] Error 1
lib/Backend/CMakeFiles/Chakra.Backend.dir/build.make:134: recipe for target 'lib/Backend/CMakeFiles/Chakra.Backend.dir/BackendApi.cpp.o' failed
make[2]: *** [lib/Backend/CMakeFiles/Chakra.Backend.dir/BackendApi.cpp.o] Error 1
CMakeFiles/Makefile2:285: recipe for target 'lib/Backend/CMakeFiles/Chakra.Backend.dir/all' failed
make[1]: *** [lib/Backend/CMakeFiles/Chakra.Backend.dir/all] Error 2
Makefile:83: recipe for target 'all' failed
make: *** [all] Error 2
See error details above. Exit code was 2
```
|
Disable x86 (32-bit) Linux builds or issue a warning
|
https://api.github.com/repos/chakra-core/ChakraCore/issues/4091/comments
| 9 |
2017-10-30T16:52:56Z
|
2024-04-25T04:53:59Z
|
https://github.com/chakra-core/ChakraCore/issues/4091
| 269,674,167 | 4,091 |
[
"chakra-core",
"ChakraCore"
] |
Xplat CI CCRobot / Linux / ChakraCore Master
```
out/Debug/ch -WERExceptionSupport -ExtendedErrorStackForTestHost -BaselineMode -maxInterpretCount:1 -maxSimpleJitRunCount:1 -bgjit- -dynamicprofilecache:profile.dpl.137 -RecyclerVerifyMark -RecyclerConcurrentStress -VerifyBarrierBit -KeepRecyclerTrackData -maic:0 -maxInterpretCount:0 test/AsmJs/argTest.js
Output:
----------------------------
ASSERTION 19183: (lib/Runtime/Base/FunctionBody.cpp, line 1852) Cannot shrink register usage
Failure: (m_inParamCount <= newInParamCount)
----------------------------
exit code: -4
```
|
Assert crash while running AsmJs/argTest.js
|
https://api.github.com/repos/chakra-core/ChakraCore/issues/4090/comments
| 3 |
2017-10-29T11:19:39Z
|
2018-01-13T03:37:47Z
|
https://github.com/chakra-core/ChakraCore/issues/4090
| 269,390,778 | 4,090 |
[
"chakra-core",
"ChakraCore"
] |
UserAgent: Mozilla/5.0 (Windows NT 10.0; Win64; x64; ServiceUI 9) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/52.0.2743.116 Safari/537.36 Edge/15.15063
Steps to reproduce the problem:
1. Configure Windows to use 150% scaling.
2. Configure Edge to use 200% scaling.
3. window.open( 'test.html', new Date().getTime(), 'width=600, height=300' );
What is the expected behavior?
The resulting window window should have
window.innerWidth == 600 && window.innerHeight == 300
https://drafts.csswg.org/cssom-view/#the-features-argument-to-the-open()-method
What went wrong?
window.open measures width and height in "windows pixels" instead of CSS pixels.
If the window contains a image with 600px x 300px it will not fit.
First test:
window.open( 'test.html', new Date().getTime(), 'width=600, height=300' );
Windows Edge Result (innerWidth x innerHeight)
100% 100% 600 x 300 (correct)
100% 200% 400 x 200 (incorrect)
150% 100% 602 x 302 (almost correct)
150% 200% 301 x 151 (incorrect)
The problem can be shifted:
window.open( 'test.html', new Date().getTime(), 'width=' + ( 600 * devicePixelRatio ) + ', height=' + ( 300 * devicePixelRatio ) );
Windows Edge Result (innerWidth x innerHeight)
100% 100% 600 x 300 (correct)
100% 200% 600 x 300 (correct)
150% 100% 902 x 452 (incorrect)
150% 200% 901 x 451 (incorrect)
So the solution can either handle windows scaling or chrome scaling but not both.
If there was a way to detect the scale factor of edge or windows independently, the problem would be easy to solve.
However I only found devicePixelRatio which gives the product of windows and edge scaling (i.e. 3.0 if windows is at 150% and edge is at 200%).
Firefox and Internet Explorer work correctly on Windows 10.
Chrome and Edge behave equally incorrect.
Is there a workaround? devicePixelRatio only helps for browser scaling but makes windows scaling even worse... I did not find a way to detect windows pixel scaling from within the browser. It would also help to detect browser scaling alone (devicePixelRatio / browser scaling = windows scaling)...
|
window.open scales incorrect
|
https://api.github.com/repos/chakra-core/ChakraCore/issues/4089/comments
| 2 |
2017-10-28T14:35:12Z
|
2017-10-31T01:56:44Z
|
https://github.com/chakra-core/ChakraCore/issues/4089
| 269,322,157 | 4,089 |
[
"chakra-core",
"ChakraCore"
] |
ChakraCore does not currently propagate relevant information in error messages from within ES Modules; it would be good to see this fixed
**EDIT1:** see 2nd comment for proposed solution to these issues.
**EDIT2:** see https://github.com/Microsoft/ChakraCore/pull/4080 for PR that addresses points 1 and 3.
**I think I could make a PR to address all of the below points - but a couple of them involve decisions - particularly the fact that currently there is no interface to set a URL for a module see related issue** #3618
Please let me know what the view of the team is on the below points and if a Pull Request for one or all of them would be welcome.
**Examples - make any error in a module in ch, standard cases:**
1. Syntax error - an error message is given along with the name of the top level parent module but not the file the error is actually in - and no line number (line number is a ch issue - CC provides it)
2. import/export name not found as above, name of top level module and - SyntaxError: Module import insert_name_here cannot be resolved
3. Run time error - stack trace with no filename or URL or specifier
**Looking into this in ChakraCore - key information is not provided when errors are set within modules:**
**1. As far as I cans see syntax errors can be thrown from these two places:**
https://github.com/Microsoft/ChakraCore/blob/9b255666b90e8594054ad851cb88d27675353582/lib/Runtime/Language/SourceTextModuleRecord.cpp#L174
https://github.com/Microsoft/ChakraCore/blob/9b255666b90e8594054ad851cb88d27675353582/lib/Runtime/Language/SourceTextModuleRecord.cpp#L834
In both cases the "optional" 3rd parameter of a URL for ```JavascriptError::CreateFromCompileScriptException``` is not provided. Meaning that the host receives no specifier/url/filename for the error.
I note that as discussed in issue https://github.com/Microsoft/ChakraCore/issues/3618 modules don't have URLs set at the moment (as there is actually no API to provide the URL) BUT the specifier is available ```this->GetSpecifierSz()``` and for most cases could be usable for error handling.
**2. import/export names not found**
https://github.com/Microsoft/ChakraCore/blob/9b255666b90e8594054ad851cb88d27675353582/lib/Runtime/Language/SourceTextModuleRecord.cpp#L630
https://github.com/Microsoft/ChakraCore/blob/9b255666b90e8594054ad851cb88d27675353582/lib/Runtime/Language/SourceTextModuleRecord.cpp#L979
https://github.com/Microsoft/ChakraCore/blob/9b255666b90e8594054ad851cb88d27675353582/lib/Runtime/Language/SourceTextModuleRecord.cpp#L1111
The property name that can't be found is given but the specifier is not provided - for debugging any large module based codebase this will be painful to work with, particularly as default exports will always come up with "default" as their name. In these cases the "fix" looks to be to change the sahpe of the error message slightly, possibly concatenating module specifier with the name of the import/export that can't be resolved.
**3. Run time error - no filename/URL in stack trace**
This comes back to the no URL issue (#3618) I pointed at above and the choice in https://github.com/Microsoft/ChakraCore/blob/09ab25c9a4e28dba22d2130ed323ab791676b11d/lib/Jsrt/Core/JsrtCore.cpp#L80 to give a nullptr as the URL for CreateSourceContextInfo as opposed to setting the specifier in the absence of a URL.
|
Error handling for ES Modules
|
https://api.github.com/repos/chakra-core/ChakraCore/issues/4068/comments
| 6 |
2017-10-26T00:21:48Z
|
2018-05-31T00:31:58Z
|
https://github.com/chakra-core/ChakraCore/issues/4068
| 268,591,298 | 4,068 |
[
"chakra-core",
"ChakraCore"
] |
The compareLocale method seems to ignore minus signs, even when the numeric option is set:
````javascript
'-2'.localeCompare('1', undefined, {numeric: true});
````
results in
````javascript
1
````
on Edge. This seems incorrect to me; other browsers also seem to handle this differently. The behavior is especially frustrating when trying to give a natural sorting to numbers. Example:
```javascript
var strings = ['3', '1', '-2'];
var ascCompare = function(elem1, elem2) {
return elem1.localeCompare(elem2, undefined, {numeric: true, sensitivity: 'variant'});
};
strings.sort(ascCompare);
```
results in
````javascript
[1, -2, 3]
````
For a working example, see: https://jsfiddle.net/9rff9xLt/3/
|
localeCompare doesn't take into account minus signs
|
https://api.github.com/repos/chakra-core/ChakraCore/issues/4060/comments
| 6 |
2017-10-25T12:37:28Z
|
2017-10-30T20:03:55Z
|
https://github.com/chakra-core/ChakraCore/issues/4060
| 268,380,967 | 4,060 |
[
"chakra-core",
"ChakraCore"
] |
Building Chakra with MSVC + /permissive- failed with error C4596, the issue is in https://github.com/Microsoft/ChakraCore/blob/master/lib/Backend/Opnd.h#L1502, and involved by commit https://github.com/Microsoft/ChakraCore/commit/fa9151ead415e84ec0c07f9acf87db2f36826a9b
Using full-qualified name to declare members inside class is ill-formed and incorrectly allowed by MSVC. MSVC now gives error when /permissive- is used. The fix is remove ‘ListOpnd::’
Failure like:
d:\chakra\src\lib\backend\Opnd.h(1502): error C4596: 'New': illegal qualified name in member declaration
|
Issues found when building Chakra with MSVC + /permissive-
|
https://api.github.com/repos/chakra-core/ChakraCore/issues/4052/comments
| 0 |
2017-10-24T03:04:54Z
|
2017-10-25T00:52:47Z
|
https://github.com/chakra-core/ChakraCore/issues/4052
| 267,882,068 | 4,052 |
[
"chakra-core",
"ChakraCore"
] |
https://github.com/Microsoft/ChakraCore/blob/669ed3c20f3e0ad96cc105d716b7392ae6a139de/lib/Runtime/Library/InJavascript/Intl.js#L640-L680
We can produce this table by running this function on our end and placing the lookup table itself in the source code. There isn't a good reason that every scriptContext that initializes Intl.js has to pay the cost of generating this table, other than ensuring that it is kept in sync, however the table is updated so infrequently (never?) that a comment indicating that it should be kept in sync with a JS snippet to run (approx this code) that actually produces the reverse lookup table is probably sufficient.
For context on this table in the source, see final table of https://msdn.microsoft.com/en-us/library/cc233968.aspx
/cc @jackhorton
|
Intl.js: pre-compute RFC 4646 <-> RFC 5646 reverse lookup table -- can be statically computed.
|
https://api.github.com/repos/chakra-core/ChakraCore/issues/4044/comments
| 1 |
2017-10-23T20:42:56Z
|
2017-11-22T02:07:44Z
|
https://github.com/chakra-core/ChakraCore/issues/4044
| 267,810,697 | 4,044 |
[
"chakra-core",
"ChakraCore"
] |
Will avoid crashes due to returning null or something else not useful from the underlying Intl implementation.
|
Intl-ICU: Don't call platform.registerBuiltInFunction for (String|Date|Number).prototype.* functions until they are properly implemented
|
https://api.github.com/repos/chakra-core/ChakraCore/issues/4043/comments
| 1 |
2017-10-23T19:26:34Z
|
2018-01-04T00:08:56Z
|
https://github.com/chakra-core/ChakraCore/issues/4043
| 267,787,963 | 4,043 |
[
"chakra-core",
"ChakraCore"
] |
Follow up to #4006
/cc @jackhorton
|
Intl: not enough Date.prototype.toLocale(Date|Time)?String -Intl and -Intl- tests
|
https://api.github.com/repos/chakra-core/ChakraCore/issues/4042/comments
| 1 |
2017-10-23T19:21:16Z
|
2018-04-09T22:51:19Z
|
https://github.com/chakra-core/ChakraCore/issues/4042
| 267,786,424 | 4,042 |
[
"chakra-core",
"ChakraCore"
] |
Hi.
I've got a minimal class which looks as follows:
```javascript
class DB {
constructor () {
}
save(key, val) {
localStorage.setItem(key, JSON.stringify(val));
window[key] = val;
}
load(key, def='{}') {
var x = localStorage.getItem(key) || def;
window[key] = JSON.parse(x);
return window[key];
}
}
```
Apparently, Edge is complaining about the use of `def='{}'` in the parameter list. The error is `Expected ')' on Line 11` which tells me that Edge doesn't like the [default parameter](https://developer.mozilla.org/en-US/docs/Web/JavaScript/Reference/Functions/Default_parameters)
The code works in almost every other browser, which got me thinking. Maybe this isn't actual native JavaScript syntax? Or maybe Edge is just being Edge. But as far as I can tell, it is standard syntax.
Edge is however having a `?` in the [Browser Compatibility list](https://developer.mozilla.org/en-US/docs/Web/JavaScript/Reference/Functions/Default_parameters#Browser_compatibility). Either way, I would strongly recommend that this be implemented in Chakra - seeing as it won't break anything, just add functionality.
Sources:
* https://developer.mozilla.org/en-US/docs/Web/JavaScript/Reference/Functions/Default_parameters
|
Default value for function parameters
|
https://api.github.com/repos/chakra-core/ChakraCore/issues/4039/comments
| 4 |
2017-10-23T13:39:36Z
|
2017-10-24T06:08:37Z
|
https://github.com/chakra-core/ChakraCore/issues/4039
| 267,671,835 | 4,039 |
[
"chakra-core",
"ChakraCore"
] |
Hi,
I've found a case where ES6 modules are not working for me in current MS Edge (Creators Update 41.16299.15.0), that is when there is a try/catch in an imported module. It's also reproducable in ChakraCore.
Here is a hosted reduced test case:
https://rawgit.com/herrernst/es6-module-try-catch/master/index.html
It works in current Chrome, Safari, and Firefox (with dom.moduleScripts.enabled), but not in MS Edge (nothing happens). If you [remove the try/catch](https://github.com/herrernst/es6-module-try-catch/blob/master/utils.js#L4) it works.
When running in Chakra (tested with 1.7.3 OSX binary release from GitHub) via `ch ./chakra-wrap.js`, this is printed:
```
thrown at main.js:
^
SyntaxError: Module import or export statement unexpected here
```
|
ES6 modules not working with try/catch in module
|
https://api.github.com/repos/chakra-core/ChakraCore/issues/4038/comments
| 4 |
2017-10-23T12:47:47Z
|
2017-10-24T17:24:44Z
|
https://github.com/chakra-core/ChakraCore/issues/4038
| 267,654,507 | 4,038 |
[
"chakra-core",
"ChakraCore"
] |
When miniSphere is run with the `--version` switch, I print out the version number of the engine and all its major dependencies, like this:
```
C:\src\spectacles-i>spherun --version
miniSphere 5.0a5 JS game engine (x64)
a lightweight JavaScript-powered game engine
(c) 2015-2017 Fat Cerberus
Allegro: v5.2.3.0 libmng: v1.0.10
Dyad.c: v0.2.0 zlib: v1.2.11
```
I haven't found a way to detect the version number of the ChakraCore DLL in use. Is this possible?
|
API to get ChakraCore DLL version?
|
https://api.github.com/repos/chakra-core/ChakraCore/issues/4037/comments
| 2 |
2017-10-22T18:07:05Z
|
2017-10-24T23:12:16Z
|
https://github.com/chakra-core/ChakraCore/issues/4037
| 267,485,719 | 4,037 |
[
"chakra-core",
"ChakraCore"
] |
I've a file (reconnecting websoket js, see: https://github.com/joewalnes/reconnecting-websocket/blob/master/reconnecting-websocket.js)
<img width="1002" alt="bildschirmfoto 2017-10-21 um 11 48 49" src="https://user-images.githubusercontent.com/364896/31850681-bc13536e-b656-11e7-9a6f-b3e2d7c3a115.png">
|
Edge complains about syntax error, don't know why
|
https://api.github.com/repos/chakra-core/ChakraCore/issues/4034/comments
| 2 |
2017-10-21T09:55:41Z
|
2017-10-21T10:02:59Z
|
https://github.com/chakra-core/ChakraCore/issues/4034
| 267,367,757 | 4,034 |
[
"chakra-core",
"ChakraCore"
] |
```js
new Intl.Collator("de-DE-u-co-phonebk").resolvedOptions()
```
Expected result (as in Chrome and Firefox)
```js
{
locale: "de-u-co-phonebk",
collation: "phonebk"
...
}
```
Actual result (Edge and IE11)
```js
{
locale: "de-DE",
collation: "phonebk"
...
}
```
This is caused by some very special/intricate/poor handling of locales that we do for collation options, since the system API that we use on Windows for comparing strings based on locale expects the language tag in a different format than the Intl spec works in. @bterlson is the locale value that we report here actually incorrect?
|
Intl: resolvedOptions().locale does not correctly represent the collation value
|
https://api.github.com/repos/chakra-core/ChakraCore/issues/4031/comments
| 6 |
2017-10-21T01:30:34Z
|
2018-01-04T00:08:56Z
|
https://github.com/chakra-core/ChakraCore/issues/4031
| 267,340,954 | 4,031 |
[
"chakra-core",
"ChakraCore"
] |
In particular I think we might want to check GetProcessInformation for AvailableCommit.
|
Chakra should treat xbox as a low memory device
|
https://api.github.com/repos/chakra-core/ChakraCore/issues/4025/comments
| 2 |
2017-10-20T21:47:45Z
|
2019-06-07T18:26:40Z
|
https://github.com/chakra-core/ChakraCore/issues/4025
| 267,313,886 | 4,025 |
[
"chakra-core",
"ChakraCore"
] |
When trying to implement authentication in an application, I've come across the following - when using `window.fetch` to make an API call that results in 401 error, the response is not returned and an empty string is returned instead. For example:
```
window.fetch('/api/method', {
method: 'POST', body: '{"data":"data"}'
}).then(function(r) {return r.json();})
.then(function(ok) {console.log('OK', ok);},
function(e) {console.log('ERROR',err);});
```
If `/api/method` returns any other HTTP code (i.e. 200, 201, 404), then the result is returned and OK logs the json data. However, if the returned code is 401, `.json()` throws an error that it can't parse the body.
All other browsers I've tested (Chrome 61, Firefox 56) do return the response properly and the issue only happens with Edge.
Ideally the authentication model for the application I am working on will depend on HTTP code being 401 and the response providing enough information to authenticate the user by the frontend and perform auth again.
This does not seem to be possible with Edge.
As 403 code is already used for other things, it can't be switched to that code. Also 401, not 403 should be used to indicate the application has to provide Authorization HTTP header.
|
window.fetch() requests that return 401 code do not get their data returned
|
https://api.github.com/repos/chakra-core/ChakraCore/issues/4019/comments
| 3 |
2017-10-20T13:26:35Z
|
2017-10-20T20:35:28Z
|
https://github.com/chakra-core/ChakraCore/issues/4019
| 267,175,400 | 4,019 |
[
"chakra-core",
"ChakraCore"
] |
In Edge 15:
```js
new Intl.Collator(undefined, { caseFirst: 'lower' }).compare("a", "A") // returns -1
new Intl.Collator(undefined, { caseFirst: 'upper' }).compare("a", "A") // returns -1
new Intl.Collator(undefined, { caseFirst: 'lower' }).resolvedOptions() // returns object with caseFirst: 'lower'
```
My understanding of the spec is that we are allowed to not support this option as long as any given values for caseFirst (or `-u-kf`) are validated _and_ we do not report caseFirst as a resolved option. @bterlson is this a correct interpretation? If so, we should have an intermediate fix of removing the caseFirst key from being reported by resolvedOptions. Regardless, we should look into implementing this in the Windows scenario.
|
Intl: Chakra does not correctly implement caseFirst support
|
https://api.github.com/repos/chakra-core/ChakraCore/issues/4015/comments
| 8 |
2017-10-19T23:32:27Z
|
2017-12-19T22:51:11Z
|
https://github.com/chakra-core/ChakraCore/issues/4015
| 267,021,021 | 4,015 |
[
"chakra-core",
"ChakraCore"
] |
/cc @jackhorton
|
Update EntryIntl_RegisterBuiltInFunction and Intl.js to use an enum instead of magic constants for entry point IDs
|
https://api.github.com/repos/chakra-core/ChakraCore/issues/4012/comments
| 0 |
2017-10-19T21:47:15Z
|
2017-10-23T19:33:14Z
|
https://github.com/chakra-core/ChakraCore/issues/4012
| 266,999,904 | 4,012 |
[
"chakra-core",
"ChakraCore"
] |
These enums were introduced to eliminate magic constants, but at the time they were only used in ICU-based Intl code. The WinGlob code uses these same constants for the same meaning.
Move these enums to a header that ICU and WinGlob have in common and update the WinGlob code.
|
Use enum classes NumberFormatStyle and NumberFormatCurrencyDisplay in WinGlob code
|
https://api.github.com/repos/chakra-core/ChakraCore/issues/4010/comments
| 2 |
2017-10-19T21:22:05Z
|
2018-02-17T01:21:02Z
|
https://github.com/chakra-core/ChakraCore/issues/4010
| 266,993,637 | 4,010 |
[
"chakra-core",
"ChakraCore"
] |
```
## Source
print(new Intl.DateTimeFormat().format.length)
┌──────────────────┬───┐
│ d8 │ 1 │
│ jsc │ │
│ node │ │
│ sm │ │
├──────────────────┼───┤
│ ch-1.7.1 │ 0 │
│ ch-dev-che3 │ │
│ ch-master-latest │ │
│ node-ch │ │
└──────────────────┴───┘
```
`Intl.NumberFormat().format.length` is correct.
```
## Source
print(new Intl.NumberFormat().format.length)
┌──────────────────┬───┐
│ ch-1.7.1 │ 1 │
│ ch-dev-che3 │ │
│ ch-master-latest │ │
│ d8 │ │
│ jsc │ │
│ node │ │
│ node-ch │ │
│ sm │ │
└──────────────────┴───┘
```
|
Intl.DateTimeFormat().format.length is incorrect (0, should be 1)
|
https://api.github.com/repos/chakra-core/ChakraCore/issues/4009/comments
| 1 |
2017-10-19T19:57:59Z
|
2017-10-23T19:33:13Z
|
https://github.com/chakra-core/ChakraCore/issues/4009
| 266,968,793 | 4,009 |
[
"chakra-core",
"ChakraCore"
] |
`new Date(STRING)` is a hot call on acme-air.
Micro-bench
```
var arr_dates = [];
var days = ['Sun', 'Mon', 'Tue', 'Wed', 'Sat'];
var mons = ['Oct', 'Nov', 'Sep', 'Jun', 'Jan'];
for (var i = 1980; i < 2050; i++)
{
for(var mday = 1; mday < 29; mday++)
{
for(var j = 0; j < days.length; j++)
{
for(var k = 0; k < mons.length; k++)
{
arr_dates.push(days[j] + " " + mons[k] + " " + mday + " 00:00:00 GMT " + i);
}
}
}
}
// 50K+ dates.. we won't be measuring the cached results
var start = Date.now();
var len = arr_dates.length;
var tt = 0;
for(var i = 0; i < len; i++)
{
tt += new Date(arr_dates[i]).getMonth();
tt = tt & 63; // MOD 64
}
var end = Date.now() - start;
console.log("Total: ", end, tt)
```
CC is 6x slower than v8 on above micro bench.
|
perf: new Date(STRING) is slow
|
https://api.github.com/repos/chakra-core/ChakraCore/issues/4008/comments
| 5 |
2017-10-19T13:32:33Z
|
2017-11-06T19:41:18Z
|
https://github.com/chakra-core/ChakraCore/issues/4008
| 266,843,750 | 4,008 |
[
"chakra-core",
"ChakraCore"
] |
Follow up #4006
/cc @jackhorton
|
Follow up #4006: Add caching to other Intl-based toLocaleString functions
|
https://api.github.com/repos/chakra-core/ChakraCore/issues/4007/comments
| 2 |
2017-10-19T01:10:09Z
|
2018-09-06T20:47:33Z
|
https://github.com/chakra-core/ChakraCore/issues/4007
| 266,690,048 | 4,007 |
[
"chakra-core",
"ChakraCore"
] |
See https://github.com/Microsoft/ChakraCore/blob/9427f0feefc6b7fd4bb38f700768e395986d2802/lib/Runtime/Library/JavascriptString.cpp#L3896-L3899
Do we have to do this ? It will unnecessarily flatten the string.
Not sure the perf gain on Acme-Air but #3752 case (1) suffers here with 25% perf drop. I would like to investigate it.
|
perf: String Equal is slow
|
https://api.github.com/repos/chakra-core/ChakraCore/issues/4002/comments
| 9 |
2017-10-18T23:03:28Z
|
2017-11-01T01:19:46Z
|
https://github.com/chakra-core/ChakraCore/issues/4002
| 266,669,971 | 4,002 |
[
"chakra-core",
"ChakraCore"
] |
Hi,
I am porting my UWP app to Creators update, and I noticed that when I passed big amount of JS code into Function.apply(null, jsCodeAsArgs) debugger crashes. When I start the same app without debugger attached (ie. from menu Start), it is working fine. It seems to be debugger issue. Details below: Exception, Native Stack Trace, Modules loaded (with newest VS updates issue is the same).
Issue redirected from VS community tracker, as not related to debugger.
Exception
Unhandled exception at 0x00007FFCA0FEE492 (Chakra.dll) in WWAHost.exe: 0xC0000005: Access violation reading location 0x0000000000000008.
Stack trace
Index Function
--------------------------------------------------------------------------------
*1 Chakra.dll!00007ffca0fee492()
2 Chakra.dll!00007ffca10b9e20()
3 Chakra.dll!00007ffca0f597db()
4 Chakra.dll!00007ffca0f592ec()
5 Chakra.dll!00007ffca0f5b5b9()
6 Chakra.dll!00007ffca0f517ca()
7 Chakra.dll!00007ffca0f56126()
8 Chakra.dll!00007ffca0f54f47()
9 Chakra.dll!00007ffca0f57200()
10 Chakra.dll!00007ffca0f54e98()
11 Chakra.dll!00007ffca0f57200()
12 Chakra.dll!00007ffca0f55da5()
13 Chakra.dll!00007ffca0f58370()
14 Chakra.dll!00007ffca0f5929d()
15 Chakra.dll!00007ffca0e48c8c()
16 Chakra.dll!00007ffca0e4900d()
17 Chakra.dll!00007ffca1059d84()
18 Chakra.dll!00007ffca0e2fe06()
19 Chakra.dll!00007ffca121a758()
20 Chakra.dll!00007ffca105e093()
21 Chakra.dll!00007ffca0fe2f13()
22 Chakra.dll!00007ffca0fe2da4()
23 Chakra.dll!00007ffca0fe2c01()
24 Chakra.dll!00007ffca105e093()
25 Chakra.dll!00007ffca0fe2f13()
26 Chakra.dll!00007ffca130a610()
27 Chakra.dll!00007ffca130cfd9()
28 Chakra.dll!00007ffca130bb14()
29 Chakra.dll!00007ffca105e093()
30 Chakra.dll!00007ffca0f2ecac()
31 Chakra.dll!00007ffca0f31c03()
32 Chakra.dll!00007ffca11d32e7()
33 Chakra.dll!00007ffca10ad7d9()
34 Chakra.dll!00007ffca11d30f2()
35 Chakra.dll!00007ffca11c00f4()
36 Chakra.dll!00007ffca10613fb()
37 Chakra.dll!00007ffca102bc65()
38 00000242117e084a()
39 Chakra.dll!00007ffca105e093()
40 Chakra.dll!00007ffca0fe2f13()
41 Chakra.dll!00007ffca130bb6d()
42 Chakra.dll!00007ffca105e093()
43 Chakra.dll!00007ffca0f2ecac()
44 Chakra.dll!00007ffca0f31e89()
45 Chakra.dll!00007ffca11d3200()
46 Chakra.dll!00007ffca11c00f4()
47 Chakra.dll!00007ffca10613fb()
48 Chakra.dll!00007ffca102bc65()
49 00000242117e0852()
50 Chakra.dll!00007ffca105e093()
51 Chakra.dll!00007ffca0fe2f13()
52 Chakra.dll!00007ffca130bb6d()
53 Chakra.dll!00007ffca105e093()
54 Chakra.dll!00007ffca0fcd82f()
55 Chakra.dll!00007ffca0f8d42f()
56 Chakra.dll!00007ffca105e093()
57 Chakra.dll!00007ffca0fe2f13()
58 Chakra.dll!00007ffca130a610()
59 Chakra.dll!00007ffca130cfd9()
60 Chakra.dll!00007ffca130bb14()
61 Chakra.dll!00007ffca105e093()
62 Chakra.dll!00007ffca0fe2f13()
63 Chakra.dll!00007ffca0fc9f10()
64 Chakra.dll!00007ffca0fc9017()
65 Chakra.dll!00007ffca103fa06()
66 Chakra.dll!00007ffca0fc8e54()
67 Chakra.dll!00007ffca0fc5656()
68 edgehtml.dll!00007ffc9f354ed1()
69 edgehtml.dll!00007ffc9f2030fb()
70 edgehtml.dll!00007ffc9f3530d6()
71 edgehtml.dll!00007ffc9f353785()
72 edgehtml.dll!00007ffc9f354e85()
73 edgehtml.dll!00007ffc9f389f25()
74 edgehtml.dll!00007ffc9f389c7b()
75 edgehtml.dll!00007ffc9f3581aa()
76 edgehtml.dll!00007ffc9f330c46()
77 edgehtml.dll!00007ffc9f37d5ad()
78 edgehtml.dll!00007ffc9f37d33b()
79 edgehtml.dll!00007ffc9f37cf16()
80 edgehtml.dll!00007ffc9f1f92b5()
81 edgehtml.dll!00007ffc9f4c491b()
82 edgehtml.dll!00007ffc9f3530d6()
83 edgehtml.dll!00007ffc9f49f760()
84 edgehtml.dll!00007ffc9f49f583()
85 edgehtml.dll!00007ffc9f353c83()
86 edgehtml.dll!00007ffc9f352175()
87 user32.dll!00007ffcc74dbc50()
88 user32.dll!00007ffcc74db5cf()
89 combase.dll!00007ffcc5171358()
90 combase.dll!00007ffcc5172048()
91 combase.dll!00007ffcc520970e()
92 combase.dll!00007ffcc5209605()
93 combase.dll!00007ffcc51618db()
94 edgehtml.dll!00007ffc9fd08a6f()
95 edgehtml.dll!00007ffc9fd089f8()
96 edgehtml.dll!00007ffc9fd08c30()
97 edgehtml.dll!00007ffc9fd08b41()
98 edgehtml.dll!00007ffc9fd08d2e()
99 kernel32.dll!00007ffcc7932774()
100 ntdll.dll!00007ffcc7be0d51()
Symbols loaded
Debug.ListModules /SymbolStatus:no /Version:yes
Name Address Path Version AppDomain
-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
WWAHost.exe 00007FF6826A0000-00007FF68276B000 C:\Windows\System32\WWAHost.exe 10.0.15063.608 (WinBuild.160101.0800)
ntdll.dll 00007FFCC7B70000-00007FFCC7D4B000 C:\Windows\System32\ntdll.dll 10.0.15063.608 (WinBuild.160101.0800)
kernel32.dll 00007FFCC7920000-00007FFCC79CE000 C:\Windows\System32\kernel32.dll 10.0.15063.296 (WinBuild.160101.0800)
KernelBase.dll 00007FFCC4270000-00007FFCC44B9000 C:\Windows\System32\KernelBase.dll 10.0.15063.608 (WinBuild.160101.0800)
apphelp.dll 00007FFCC27C0000-00007FFCC283E000 C:\Windows\System32\apphelp.dll 10.0.15063.0 (WinBuild.160101.0800)
msvcrt.dll 00007FFCC5D10000-00007FFCC5DAD000 C:\Windows\System32\msvcrt.dll 7.0.15063.0 (WinBuild.160101.0800)
kernel.appcore.dll 00007FFCC4070000-00007FFCC4081000 C:\Windows\System32\kernel.appcore.dll 10.0.15063.0 (WinBuild.160101.0800)
rpcrt4.dll 00007FFCC7620000-00007FFCC7745000 C:\Windows\System32\rpcrt4.dll 10.0.15063.0 (WinBuild.160101.0800)
combase.dll 00007FFCC5130000-00007FFCC5429000 C:\Windows\System32\combase.dll 10.0.15063.608 (WinBuild.160101.0800)
ucrtbase.dll 00007FFCC4580000-00007FFCC4676000 C:\Windows\System32\ucrtbase.dll 10.0.15063.413 (WinBuild.160101.0800)
bcryptprimitives.dll 00007FFCC4680000-00007FFCC46EA000 C:\Windows\System32\bcryptprimitives.dll 10.0.15063.0 (WinBuild.160101.0800)
profapi.dll 00007FFCC3FE0000-00007FFCC3FF5000 C:\Windows\System32\profapi.dll 10.0.15063.0 (WinBuild.160101.0800)
SHCore.dll 00007FFCC5EC0000-00007FFCC5F6A000 C:\Windows\System32\SHCore.dll 10.0.15063.608 (WinBuild.160101.0800)
sechost.dll 00007FFCC5C10000-00007FFCC5C69000 C:\Windows\System32\sechost.dll 10.0.15063.0 (WinBuild.160101.0800)
iertutil.dll 00007FFCB73B0000-00007FFCB763A000 C:\Windows\System32\iertutil.dll 11.00.15063.608 (WinBuild.160101.0800)
advapi32.dll 00007FFCC5DB0000-00007FFCC5E51000 C:\Windows\System32\advapi32.dll 10.0.15063.0 (WinBuild.160101.0800)
WinTypes.dll 00007FFCC0CD0000-00007FFCC0E09000 C:\Windows\System32\WinTypes.dll 10.0.15063.0 (WinBuild.160101.0800)
bcrypt.dll 00007FFCC3F20000-00007FFCC3F45000 C:\Windows\System32\bcrypt.dll 10.0.15063.0 (WinBuild.160101.0800)
WwaExt.dll 00007FFCC0E60000-00007FFCC0E6E000 C:\Windows\System32\WwaExt.dll 10.0.15063.0 (WinBuild.160101.0800)
edgehtml.dll 00007FFC9F020000-00007FFCA06D2000 C:\Windows\System32\edgehtml.dll 11.00.15063.608 (WinBuild.160101.0800)
ole32.dll 00007FFCC5A50000-00007FFCC5B95000 C:\Windows\System32\ole32.dll 10.0.15063.502 (WinBuild.160101.0800)
gdi32.dll 00007FFCC5BA0000-00007FFCC5BC7000 C:\Windows\System32\gdi32.dll 10.0.15063.0 (WinBuild.160101.0800)
Chakra.dll 00007FFCA0DF0000-00007FFCA15D2000 C:\Windows\System32\Chakra.dll 11.00.15063.608 (WinBuild.160101.0800)
gdi32full.dll 00007FFCC4090000-00007FFCC4218000 C:\Windows\System32\gdi32full.dll 10.0.15063.608 (WinBuild.160101.0800)
msvcp_win.dll 00007FFCC44E0000-00007FFCC457A000 C:\Windows\System32\msvcp_win.dll 10.0.15063.0 (WinBuild.160101.0800)
cryptsp.dll 00007FFCC3A30000-00007FFCC3A47000 C:\Windows\System32\cryptsp.dll 10.0.15063.0 (WinBuild.160101.0800)
user32.dll 00007FFCC74D0000-00007FFCC761A000 C:\Windows\System32\user32.dll 10.0.15063.608 (WinBuild.160101.0800)
win32u.dll 00007FFCC44C0000-00007FFCC44DE000 C:\Windows\System32\win32u.dll 10.0.15063.608 (WinBuild.160101.0800)
mlang.dll 00007FFCB56A0000-00007FFCB56DF000 C:\Windows\System32\mlang.dll 10.0.15063.0 (WinBuild.160101.0800)
imm32.dll 00007FFCC5BD0000-00007FFCC5BFD000 C:\Windows\System32\imm32.dll 10.0.15063.0 (WinBuild.160101.0800)
twinapi.appcore.dll 00007FFCC29B0000-00007FFCC2B20000 C:\Windows\System32\twinapi.appcore.dll 10.0.15063.332 (WinBuild.160101.0800)
windows.storage.dll 00007FFCC4970000-00007FFCC5063000 C:\Windows\System32\windows.storage.dll 10.0.15063.608 (WinBuild.160101.0800)
shlwapi.dll 00007FFCC5E60000-00007FFCC5EB1000 C:\Windows\System32\shlwapi.dll 10.0.15063.502 (WinBuild.160101.0800)
powrprof.dll 00007FFCC4020000-00007FFCC406C000 C:\Windows\System32\powrprof.dll 10.0.15063.0 (WinBuild.160101.0800)
urlmon.dll 00007FFCB86B0000-00007FFCB8878000 C:\Windows\System32\urlmon.dll 11.00.15063.608 (WinBuild.160101.0800)
cryptbase.dll 00007FFCC3A50000-00007FFCC3A5B000 C:\Windows\System32\cryptbase.dll 10.0.15063.0 (WinBuild.160101.0800)
MrmCoreR.dll 00007FFCB6930000-00007FFCB6A32000 C:\Windows\System32\MrmCoreR.dll 10.0.15063.608 (WinBuild.160101.0800)
oleaut32.dll 00007FFCC7860000-00007FFCC7920000 C:\Windows\System32\oleaut32.dll 10.0.15063.332 (WinBuild.160101.0800)
Windows.UI.dll 00007FFCB6370000-00007FFCB6476000 C:\Windows\System32\Windows.UI.dll 10.0.15063.502 (WinBuild.160101.0800)
TextInputFramework.dll 00007FFCB6830000-00007FFCB68B2000 C:\Windows\System32\TextInputFramework.dll 10.0.15063.502 (WinBuild.160101.0800)
CoreUIComponents.dll 00007FFCB6090000-00007FFCB6362000 C:\Windows\System32\CoreUIComponents.dll 10.00.15063.608
CoreMessaging.dll 00007FFCC1890000-00007FFCC1973000 C:\Windows\System32\CoreMessaging.dll 10.00.15063.608
usermgrcli.dll 00007FFCC00A0000-00007FFCC00B5000 C:\Windows\System32\usermgrcli.dll 10.0.15063.0 (WinBuild.160101.0800)
ntmarta.dll 00007FFCC31C0000-00007FFCC31F1000 C:\Windows\System32\ntmarta.dll 10.0.15063.0 (WinBuild.160101.0800)
rmclient.dll 00007FFCC1410000-00007FFCC142D000 C:\Windows\System32\rmclient.dll 10.0.15063.0 (WinBuild.160101.0800)
BCP47Langs.dll 00007FFCB68C0000-00007FFCB6929000 C:\Windows\System32\BCP47Langs.dll 10.0.15063.0 (WinBuild.160101.0800)
d2d1.dll 00007FFCB24F0000-00007FFCB2A94000 C:\Windows\System32\d2d1.dll 10.0.15063.483 (WinBuild.160101.0800)
crypt32.dll 00007FFCC46F0000-00007FFCC48B9000 C:\Windows\System32\crypt32.dll 10.0.15063.296 (WinBuild.160101.0800)
msasn1.dll 00007FFCC4000000-00007FFCC4011000 C:\Windows\System32\msasn1.dll 10.0.15063.0 (WinBuild.160101.0800)
uxtheme.dll 00007FFCC28E0000-00007FFCC2975000 C:\Windows\System32\uxtheme.dll 10.0.15063.0 (WinBuild.160101.0800)
OneCoreUAPCommonProxyStub.dll 00007FFCB6A40000-00007FFCB6F71000 C:\Windows\System32\OneCoreUAPCommonProxyStub.dll 10.0.15063.608 (WinBuild.160101.0800)
DWrite.dll 00007FFCB1E30000-00007FFCB20EF000 C:\Windows\System32\DWrite.dll 10.0.15063.483 (WinBuild.160101.0800)
dwmapi.dll 00007FFCC2640000-00007FFCC266A000 C:\Windows\System32\dwmapi.dll 10.0.15063.0 (WinBuild.160101.0800)
dxgi.dll 00007FFCC2F30000-00007FFCC2FD4000 C:\Windows\System32\dxgi.dll 10.0.15063.502 (WinBuild.160101.0800)
msctf.dll 00007FFCC58E0000-00007FFCC5A46000 C:\Windows\System32\msctf.dll 10.0.15063.332 (WinBuild.160101.0800)
UiaManager.dll 00007FFCB9430000-00007FFCB9448000 C:\Windows\System32\UiaManager.dll 10.0.15063.0 (WinBuild.160101.0800)
Windows.Internal.SecurityMitigationsBroker.dll 00007FFCBCD30000-00007FFCBCD3F000 C:\Windows\System32\Windows.Internal.SecurityMitigationsBroker.dll 10.0.15063.0 (WinBuild.160101.0800)
userenv.dll 00007FFCC3EE0000-00007FFCC3F09000 C:\Windows\System32\userenv.dll 10.0.15063.0 (WinBuild.160101.0800)
d3d11.dll 00007FFCC1E20000-00007FFCC20FF000 C:\Windows\System32\d3d11.dll 10.0.15063.0 (WinBuild.160101.0800)
igd10iumd64.dll 00007FFCB0160000-00007FFCB0D02000 C:\Windows\System32\igd10iumd64.dll 10.18.10.4425
ncrypt.dll 00007FFCC3B10000-00007FFCC3B35000 C:\Windows\System32\ncrypt.dll 10.0.15063.0 (WinBuild.160101.0800)
ntasn1.dll 00007FFCC3AD0000-00007FFCC3B06000 C:\Windows\System32\ntasn1.dll 10.0.15063.0 (WinBuild.160101.0800)
igdusc64.dll 00007FFCB17B0000-00007FFCB1C15000 C:\Windows\System32\igdusc64.dll 10.18.10.4425
Windows.UI.Core.TextInput.dll 00007FFCABAE0000-00007FFCABB5A000 C:\Windows\System32\Windows.UI.Core.TextInput.dll 10.0.15063.332 (WinBuild.160101.0800)
Windows.Graphics.dll 00007FFCABA80000-00007FFCABADA000 C:\Windows\System32\Windows.Graphics.dll 10.0.15063.0 (WinBuild.160101.0800)
srpapi.dll 00007FFCABCF0000-00007FFCABD18000 C:\Windows\System32\srpapi.dll 10.0.15063.608 (WinBuild.160101.0800)
sspicli.dll 00007FFCC3EB0000-00007FFCC3EE0000 C:\Windows\System32\sspicli.dll 10.0.15063.0 (WinBuild.160101.0800)
msIso.dll 00007FFCB6760000-00007FFCB67D4000 C:\Windows\System32\msIso.dll 11.00.15063.608 (WinBuild.160101.0800)
ninput.dll 00007FFCB3F10000-00007FFCB3F6D000 C:\Windows\System32\ninput.dll 10.0.15063.0 (WinBuild.160101.0800)
msimtf.dll 00007FFCBBFE0000-00007FFCBBFF0000 C:\Windows\System32\msimtf.dll 10.0.15063.0 (WinBuild.160101.0800)
directmanipulation.dll 00007FFCBD6D0000-00007FFCBD758000 C:\Windows\System32\directmanipulation.dll 10.0.15063.0 (WinBuild.160101.0800)
dcomp.dll 00007FFCC2400000-00007FFCC2522000 C:\Windows\System32\dcomp.dll 10.0.15063.483 (WinBuild.160101.0800)
twinapi.dll 00007FFCAF210000-00007FFCAF292000 C:\Windows\System32\twinapi.dll 10.0.15063.0 (WinBuild.160101.0800)
WwaApi.dll 00007FFCB4A00000-00007FFCB4A45000 C:\Windows\System32\WwaApi.dll 10.0.15063.0 (WinBuild.160101.0800)
pdm.dll 00007FFCA2040000-00007FFCA20D3000 C:\Program Files\Common Files\microsoft shared\VS7Debug\pdm.dll 16.00.71.0
msdbg2.dll 00007FFCAEBC0000-00007FFCAEC22000 C:\Program Files\Common Files\microsoft shared\VS7Debug\msdbg2.dll 16.00.71.0
pdmproxy100.dll 00007FFCB8BA0000-00007FFCB8BBB000 C:\Program Files\Common Files\microsoft shared\VS7Debug\pdmproxy100.dll 16.00.71.0
VSDebugScriptAgent140.dll 00007FFCB8A70000-00007FFCB8A9F000 C:\Program Files\Common Files\microsoft shared\VS7Debug\VSDebugScriptAgent140.dll 14.0.23107.0 built by: D14REL
version.dll 00007FFCB9C00000-00007FFCB9C0A000 C:\Windows\System32\version.dll 10.0.15063.0 (WinBuild.160101.0800)
shell32.dll 00007FFCC5F80000-00007FFCC73B6000 C:\Windows\System32\shell32.dll 10.0.15063.608 (WinBuild.160101.0800)
cfgmgr32.dll 00007FFCC4220000-00007FFCC4269000 C:\Windows\System32\cfgmgr32.dll 10.0.15063.0 (WinBuild.160101.0800)
wininet.dll 00007FFCB6F80000-00007FFCB72AE000 C:\Windows\System32\wininet.dll 11.00.15063.608 (WinBuild.160101.0800)
ieapfltr.dll 00007FFCA2620000-00007FFCA27AA000 C:\Windows\System32\ieapfltr.dll 11.00.15063.608
smartscreenps.dll 00007FFCB64B0000-00007FFCB64E8000 C:\Windows\System32\smartscreenps.dll 10.0.15063.332 (WinBuild.160101.0800)
DataExchange.dll 00007FFCAE260000-00007FFCAE2A7000 C:\Windows\System32\DataExchange.dll 10.0.15063.0 (WinBuild.160101.0800)
DiagnosticsTap.dll 00007FFCA1FC0000-00007FFCA2036000 C:\Program Files (x86)\Microsoft Visual Studio 14.0\Common7\IDE\Script Diagnostics\amd64\DiagnosticsTap.dll 14.0.23107.0 built by: D14REL
atlthunk.dll 00007FFCB9A30000-00007FFCB9A3B000 C:\Windows\System32\atlthunk.dll 10.0.15063.0 (WinBuild.160101.0800)
Windows.Globalization.dll 00007FFCBD2A0000-00007FFCBD428000 C:\Windows\System32\Windows.Globalization.dll 10.0.15063.0 (WinBuild.160101.0800)
profext.dll 00007FFCC3230000-00007FFCC3251000 C:\Windows\System32\profext.dll 10.0.15063.0 (WinBuild.160101.0800)
rometadata.dll 00007FFCB9370000-00007FFCB93AB000 C:\Windows\System32\rometadata.dll 4.7.2046.0 built by: NET47REL1
Windows.UI.Immersive.dll 00007FFCB65B0000-00007FFCB6756000 C:\Windows\System32\Windows.UI.Immersive.dll 10.0.15063.502 (WinBuild.160101.0800)
sxs.dll 00007FFCC3DB0000-00007FFCC3E4A000 C:\Windows\System32\sxs.dll 10.0.15063.0 (WinBuild.160101.0800)
WindowsCodecs.dll 00007FFCB3250000-00007FFCB33FC000 C:\Windows\System32\WindowsCodecs.dll 10.0.15063.483 (WinBuild.160101.0800)
xmllite.dll 00007FFCBCA90000-00007FFCBCAC5000 C:\Windows\System32\xmllite.dll 10.0.15063.0 (WinBuild.160101.0800)
cabinet.dll 00007FFCB8A40000-00007FFCB8A67000 C:\Windows\System32\cabinet.dll 5.00 (WinBuild.160101.0800)
UIAnimation.dll 00007FFCC1CE0000-00007FFCC1D26000 C:\Windows\System32\UIAnimation.dll 10.0.15063.0 (WinBuild.160101.0800)
wuceffects.dll 00007FFCB1360000-00007FFCB139C000 C:\Windows\System32\wuceffects.dll 10.0.15063.0 (WinBuild.160101.0800)
|
Access violation while debugging a JS UWP App
|
https://api.github.com/repos/chakra-core/ChakraCore/issues/3987/comments
| 7 |
2017-10-17T19:27:30Z
|
2017-12-01T00:11:12Z
|
https://github.com/chakra-core/ChakraCore/issues/3987
| 266,249,287 | 3,987 |
[
"chakra-core",
"ChakraCore"
] |
Calling `Date.parse()` with the format `YYYY-MM-DD HH:mm:ss.sss` returns `NaN`:
`Date.parse('2017-10-17 08:00:00.000')`
I'm not sure whether the specification requires this format, but it works in Chrome and Firefox.
Tested in Edge 40.15063.674.0
|
Date.parse() not working with milliseconds
|
https://api.github.com/repos/chakra-core/ChakraCore/issues/3980/comments
| 0 |
2017-10-17T14:06:06Z
|
2017-11-21T22:59:32Z
|
https://github.com/chakra-core/ChakraCore/issues/3980
| 266,138,492 | 3,980 |
[
"chakra-core",
"ChakraCore"
] |
I was investigating the viability of porting the current source to Android (and possibly iOS) for a project I'm working on. I tested the performance and usability on desktop and the performance seems really good (on par with our current V8/JavaScriptCore solution), but we need to support Android+consoles as well.
Looking at the [roadmap](https://github.com/Microsoft/ChakraCore/wiki/Roadmap) I don't see these platforms on the horizon, but seeing that Edge made it to a preview on Android/iOS maybe there are some plans to make it fully cross-platform and move away from WebKit/Blink.
There are some \_\_ANDROID\_\_ defines scattered here and there in the code, and also an ARM configuration (that I don't know if it's functional) so it's not entirely unsupported.
I wanted to share my ~2 day experience and see if there's anyone else who tried doing this port with any success, or has any pointers what is going wrong or if it's even feasible to do in the current state.
_Preparation:_
- I checked the current issues and found #3270 and #3179 relevant, although there wasn't much information there. PR #2292 suggested that at some point the Android port was working.
_Compilation:_
- First, I tried to compile the Android port on Windows, but that had some unexplainable troubles with the PATH env var and others, so I switched on a Linux machine and the compilation was fairly straightforward.
- I was unable to get the project compile with any other Android NDK other than R15c (too many other errors with previous NDKs).
- There was a redefinition error for `int8_t` in [pal_mstypes.h](https://github.com/Microsoft/ChakraCore/blob/master/pal/inc/pal_mstypes.h#L179) which was fixed by adding `&& !defined(__ANDROID__)` to the define
- There were some missing includes for the JIT build, so I focused on the NoJIT one, which compiled easily.
```
tools/android_toolchain.sh <path_to_ndk_r15c>
./build.sh --target=android --arch=arm --no-icu --no-jit --static --debug -j
```
After that I had `libChakraCoreStatic.a`.
We're building our Android project with VS on Windows, so I wanted to be sure that there's no NDK incompatibility of some sort, but VS supports NDKs up to R13. R14+ NDKs cause _clang_ to [emit warnings](https://developercommunity.visualstudio.com/content/problem/38376/vs2017-android-projects-dont-work-with-ndk-14b-wit.html) so I had to write a small tool that replaced the _clang_ binary with one that transformed the command line args' backslashes to forward slashes and forwarded that to the real _clang_ executable.
With everything built with the same compiler and NDK it was time to run it.
_Running:_
- I tried to run it in a rather complex application that I had, but got runtime errors almost immediately. Initialization went fine, but script execution was broken.
- The *ARGUMENTS* macro used [DECLARE_ARGS_VARARRAY](https://github.com/Microsoft/ChakraCore/blob/master/lib/Runtime/Language/Arguments.h#24) but the stack layout doesn't look like what it's expected to be and the app crashes. I spent a couple of hours in the ARM docs and the disassembly trying to figure out where the var args are supposed to be, without much success. The WIN32 version of the macro couldn't be used, because on gcc/clang `va_list` is a built-in and cannot be cast, so I simply made a stack of 32 `Js::Var`s and filled it with `va_start`/`va_arg`/`va_end` for testing (which meant that I can have more than 32 JS args, but I wanted to see where this will get me).
- After the `va_list` "fix" I got another crash which didn't make any sense and this is where I stopped. The call stack went from `Js::InterpreterStackFrame::OP_CallCommon<Js::OpLayoutT_CallI<Js::LayoutSizePolicy<(Js::LayoutSize)0> > >(Js::InterpreterStackFrame * this, const Js::OpLayoutT_CallI<Js::LayoutSizePolicy<SmallLayout> > * playout, Js::RecyclableObject * function, unsigned int flags, const Js::AuxArray<unsigned int> * spreadIndices)` into `Js::UriHelper::ValidateURIProps` and the loop there from 0-127 had a value way out of range and triggered an assert. I tried removing the debug check there, but the crash just moved to `UriHelper::Decode`, which leads me to believe some jump address is incorrect, maybe the execution relied on the stack satisfying some certain layout.
That's as far as I got, if there is/will be any development or advice how to get the Android (and also iOS) port running, I'd love to hear it.
|
Running ChakraCore on Android?
|
https://api.github.com/repos/chakra-core/ChakraCore/issues/3966/comments
| 9 |
2017-10-16T11:59:08Z
|
2018-02-13T23:49:00Z
|
https://github.com/chakra-core/ChakraCore/issues/3966
| 265,744,093 | 3,966 |
[
"chakra-core",
"ChakraCore"
] |
When using `JsDiagGetProperties()` to inspect an object in the debugger, is there a way to get the property attributes (writable, configurable, enumerable) for each property? I thought that was what `propertyAttributes` was for, but that's not the case:
```
/// propertyAttributes is a bit mask of
/// NONE = 0x1,
/// HAVE_CHILDRENS = 0x2,
/// READ_ONLY_VALUE = 0x4,
```
|
How to get property attributes in the debugger
|
https://api.github.com/repos/chakra-core/ChakraCore/issues/3965/comments
| 5 |
2017-10-14T17:18:09Z
|
2017-12-07T18:20:14Z
|
https://github.com/chakra-core/ChakraCore/issues/3965
| 265,507,994 | 3,965 |
[
"chakra-core",
"ChakraCore"
] |
Slightly convoluted issue I discovered whilst fixing https://github.com/Microsoft/ChakraCore/issues/3902
Note I believe that this is an issue in ch not ChakraCore itself. BUT this issue was causing a CI test failure when I was fixing 3902; and could cause other test errors in the future.
Issue is slightly convoluted and is best illustrated by 5 test cases:
**Case 1**
```js
//testStart.js
WScript.LoadScriptFile("folder/script.js");
```
```js
//folder/script.js
import("../moduleOne.js");
```
```js
//moduleOne.js
console.log("Completed");
```
Output:
```
Couldn't load file.
thrown at moduleOne.js:
^
URIError: [snip]/ChakraCore-master/out/Test/moduleOne.js
```
**The path for the import() from the Script is expected to be relative to the CWD by ch and so this case fails.**
***Case 2***
```js
//testStart.js
WScript.LoadScriptFile("folder/script.js");
```
```js
//folder/script.js
import("moduleOne.js");
```
```js
//moduleOne.js
console.log("Completed");
```
Output:
```
Completed
```
**The path for the import() from the Script is expected to be relative to the CWD by ch and so this case succeeds.**
***Case 3***
```js
//testStart.js
WScript.LoadScriptFile("folder/module.js","module");
```
```js
//folder/module.js
import("../moduleOne.js");
```
```js
//moduleOne.js
console.log("Completed");
```
Output:
```
Completed
```
**The path for the import() from the Module is expected to be relative to the Module by ch and so this case succeeds.**
***Case 4***
```js
//testStart.js
WScript.LoadScriptFile("folder/module.js","module");
WScript.LoadScriptFile("folder/script.js");
```
```js
//folder/module.js
import("../moduleOne.js");
```
```js
//folder/script.js
import("../moduleOne.js");
```
```js
//moduleOne.js
console.log("Completed");
```
Output:
```
Completed
```
**Running both cases 1 and 3 (with 3 first) expect the failure from case 1 BUT ch substitutes the version of moduleOne.js loaded for module.js when moduleOne.js is requested by script.js and so this case succeeds - this behaviour is clearly inconsistent - either the path used in script.js is wrong or it isn't, its accuracy should not be determined by whether the module has already been imported**
***Case 5***
```js
//testStart.js
WScript.LoadScriptFile("folder/script.js");
WScript.LoadScriptFile("folder/module.js","module");
```
```js
//folder/module.js
import("../moduleOne.js");
```
```js
//folder/script.js
import("../moduleOne.js");
```
```js
//moduleOne.js
console.log("Completed");
```
Output:
```
Couldn't load file.
thrown at moduleOne.js:
^
URIError: [snip]/ChakraCore-master/out/Test/moduleOne.js
```
**Running both cases 1 and 3 again but this time with 1 first and the failure is restored**
In the slightly more involved case of the CI test for bug_issue_3257 this behaviour was leading to a hard crash - I had to change the paths in the equivalent script file to suppress it; see also https://github.com/Microsoft/ChakraCore/pull/3934
|
BUG: Inconsistent module path handling in ch
|
https://api.github.com/repos/chakra-core/ChakraCore/issues/3964/comments
| 4 |
2017-10-14T13:22:48Z
|
2018-05-31T00:31:58Z
|
https://github.com/chakra-core/ChakraCore/issues/3964
| 265,490,969 | 3,964 |
[
"chakra-core",
"ChakraCore"
] |
Might be related to this [other issue](https://github.com/Microsoft/ChakraCore/issues/2997).
**Problem**: when subclassing `CustomEvent`, the `instanceof` operator fails in returning the correct boolean on an instance of that subclass.
Example:
```js
class MySubCustomEvent extends CustomEvent{
constructor(){
super('subcustom');
}
}
const instance = new MySubCustomEvent();
instance instanceof MySubCustomEvent; // returns false (expected true);
instance instanceof CustomEvent; // return true
```
Here's a [jsfiddle](https://jsfiddle.net/8m0fjfh1/1/) that works fine for Chrome, Opera, Firefox, ...but not for Edge.
|
CustomEvent ES2015 SubClass vs instanceof operator
|
https://api.github.com/repos/chakra-core/ChakraCore/issues/3952/comments
| 3 |
2017-10-13T10:24:38Z
|
2017-10-16T19:27:25Z
|
https://github.com/chakra-core/ChakraCore/issues/3952
| 265,241,397 | 3,952 |
[
"chakra-core",
"ChakraCore"
] |
```js
new Intl.NumberFormat('en-US', { useGrouping: false }).format(123456)
```
Expected return: `"123456"`
Actual return: `"123,456"`
|
Intl-ICU: useGrouping option for NumberFormat is not respected
|
https://api.github.com/repos/chakra-core/ChakraCore/issues/3948/comments
| 0 |
2017-10-12T23:50:48Z
|
2017-10-13T20:47:57Z
|
https://github.com/chakra-core/ChakraCore/issues/3948
| 265,126,388 | 3,948 |
[
"chakra-core",
"ChakraCore"
] |
ch.exe poc.js
0:000> g
ModLoad: 000007fe`fe940000 000007fe`fe96e000 C:\Windows\system32\IMM32.DLL
ModLoad: 000007fe`fe170000 000007fe`fe279000 C:\Windows\system32\MSCTF.dll
ModLoad: 000007fe`dcfa0000 000007fe`dd60b000 D:\cc_windows_1_7_3\x64_release\chakracore.dll
ModLoad: 000007fe`fe5e0000 000007fe`fe6bb000 C:\Windows\system32\ADVAPI32.dll
ModLoad: 000007fe`fe080000 000007fe`fe09f000 C:\Windows\SYSTEM32\sechost.dll
ModLoad: 000007fe`fd970000 000007fe`fd97f000 C:\Windows\system32\cryptbase.dll
(1a30.1120): Access violation - code c0000005 (first chance)
First chance exceptions are reported before any exception handling.
This exception may be expected and handled.
00000000`01dd1997 498b4708 mov rax,qword ptr [r15+8] ds:00040000`00000008=????????????????
0:001> r
rax=0000000000000000 rbx=8000000280000002 rcx=000007fedd12c601
rdx=0000000000000001 rsi=000007fedd421328 rdi=0000000000000001
rip=0000000001dd1997 rsp=0000000001f5ebf0 rbp=0000000001f5ec80
r8=7ff0000000000000 r9=000fffffffffffff r10=000007fedd5795f8
r11=0000000001f5ebe0 r12=0000000000000001 r13=000007fedd5848e0
r14=0000000001c2c020 r15=0004000000000000
iopl=0 nv up ei pl nz na pe nc
cs=0033 ss=002b ds=002b es=002b fs=0053 gs=002b efl=00010202
00000000`01dd1997 498b4708 mov rax,qword ptr [r15+8] ds:00040000`00000008=????????????????
[poc.zip](https://github.com/Microsoft/ChakraCore/files/1377872/poc.zip)
=====================
qflb.wu () dbappsecurity com cn
|
A bug
|
https://api.github.com/repos/chakra-core/ChakraCore/issues/3942/comments
| 2 |
2017-10-12T02:56:13Z
|
2018-01-11T02:07:27Z
|
https://github.com/chakra-core/ChakraCore/issues/3942
| 264,798,445 | 3,942 |
[
"chakra-core",
"ChakraCore"
] |
Currently we get lines like `TryGetProperty<true, true, true, true, true, true, !PolymorphicInlineCache::IsPolymorphic, PolymorphicInlineCache::IsPolymorphic, false>;` this is less readable than it could be. Changing the boolean template parameters to enum classes would allow us to have clearer code around this and make it both easier to review and less error-prone in the first place.
|
Replace InlineCache template parameters with enum classes
|
https://api.github.com/repos/chakra-core/ChakraCore/issues/3939/comments
| 0 |
2017-10-12T01:08:17Z
|
2018-05-07T22:21:44Z
|
https://github.com/chakra-core/ChakraCore/issues/3939
| 264,783,137 | 3,939 |
[
"chakra-core",
"ChakraCore"
] |
New feature from GitHub: GitHub will now directly surface “help wanted” PRs.
We are currently using “Accepting PRs” and have that label set up with up-for-grabs: https://github.com/Microsoft/ChakraCore/wiki/Label-Glossary#accepting-prs
We can update our entry for up-for-grabs.net here: https://github.com/up-for-grabs/up-for-grabs.net/blob/gh-pages/_data/projects/ChakraCore.yml
Typescript has already made this change in their issue tracker:
https://github.com/Microsoft/TypeScript/pull/19105
This is a quick change:
* [x] https://github.com/Microsoft/ChakraCore/labels - rename “Accepting PRs” to “help wanted”
* [x] Update https://github.com/Microsoft/ChakraCore/wiki/Label-Glossary
* [x] Update references to this tag in ChakraCore-wiki
* [x] Update references to this tag in README etc in ChakraCore
* [x] Submit a PR for https://github.com/up-for-grabs/up-for-grabs.net/blob/gh-pages/_data/projects/ChakraCore.yml
* [x] Confirm up-for-grabs PR is merged and the resulting [query](http://up-for-grabs.net/#/tags/chakra) looks right.
Any objections?
|
Proposal: Rename tag "Accepting PRs" to "help wanted"
|
https://api.github.com/repos/chakra-core/ChakraCore/issues/3938/comments
| 6 |
2017-10-12T00:49:39Z
|
2017-10-17T22:52:43Z
|
https://github.com/chakra-core/ChakraCore/issues/3938
| 264,780,515 | 3,938 |
[
"chakra-core",
"ChakraCore"
] |
1st part ->
`13a0220fb7af048af1c137e1ef4f9d6a3df806cb` -> `965098b204069d131177c3c0bd9ca847ccff950d`
2nd part ->
`965098b204069d131177c3c0bd9ca847ccff950d` -> `bd0e524ff7d31acb2bb8eca5303c9c2984fdea39`
pre-after this commit ranges above, Acme-AIR LTO perf loss is ~6.5%
UPDATE:
Looked once again the builds around the commit IDs given above. Unfortunately, hard to blame a single place for the regression. Might have multiple changes involved.
|
perf: acme-air perf regression in a particular commit range
|
https://api.github.com/repos/chakra-core/ChakraCore/issues/3933/comments
| 1 |
2017-10-11T21:48:41Z
|
2017-10-12T16:10:17Z
|
https://github.com/chakra-core/ChakraCore/issues/3933
| 264,747,197 | 3,933 |
[
"chakra-core",
"ChakraCore"
] |
Made a quick experiment with locking algorithms and other cross thread sync implementations under PAL, saw 2% AcmeAIR LTO gain after basically updating from Sleep/Wait to PThread/Mutex
This tells me that sometimes we lock longer than Sleep timer hence we end up Vol. context switching twice or more!
This might be causing us a bottle-neck and will need some attention.
|
linux: less kernel calls
|
https://api.github.com/repos/chakra-core/ChakraCore/issues/3930/comments
| 5 |
2017-10-11T15:30:29Z
|
2017-11-09T17:29:57Z
|
https://github.com/chakra-core/ChakraCore/issues/3930
| 264,633,363 | 3,930 |
[
"chakra-core",
"ChakraCore"
] |
My apologies if this is the wrong place to submit this.
I have found what I suspect is a bug in Edge. In JavaScript, the lines:
```
a = 0.0003;
b = a.toPrecision(3);
```
result in b containing "0.000200". (IE, Chrome and Firefox all give the "0.000300" that I was expecting.)
See [http://jsfiddle.net/dajp/2b47qLg0/](http://jsfiddle.net/dajp/2b47qLg0/)
I'm using Edge 40.15063.0.0.
This is causing me some difficulties. Many thanks.
|
toPrecision() in Edge gives incorrect output in rare cases
|
https://api.github.com/repos/chakra-core/ChakraCore/issues/3929/comments
| 8 |
2017-10-11T12:33:40Z
|
2018-01-17T16:14:34Z
|
https://github.com/chakra-core/ChakraCore/issues/3929
| 264,568,427 | 3,929 |
[
"chakra-core",
"ChakraCore"
] |
https://github.com/tc39/proposal-regexp-named-groups
Tests: https://github.com/tc39/test262/tree/master/test/built-ins/RegExp/named-groups
|
Implement RegExp named captures
|
https://api.github.com/repos/chakra-core/ChakraCore/issues/3928/comments
| 0 |
2017-10-11T08:12:54Z
|
2018-05-07T22:21:50Z
|
https://github.com/chakra-core/ChakraCore/issues/3928
| 264,494,042 | 3,928 |
[
"chakra-core",
"ChakraCore"
] |
I am trying to port ChakraCore to [Buck Build](https://buckbuild.com), but the include paths used in ChakraCore are making this difficult.
For instance:
- `RuntimeBasePch.h` contains `#include "Runtime.h"`;
- `Runtime.h` contains `#include "../JITIDL/JITTypes.h"`.
This causes the compiler to walk outside of the top-level of the build, which breaks reproducibility.
This scenario is avoidable if a path without `../` is used instead. For example:
```
#include "JITIDL/JITTypes.h"
```
Or better yet, something like this:
```
#include <ChakraCore/JITIDL/JITTypes.h>
```
Ideally, all headers would be included from the same include-path. This is common practice is many C++ libraries, such as [Boost](https://github.com/boostorg/graph/blob/develop/include/boost/graph/subgraph.hpp#L19).
What are the ChakraCore team's thoughts on this?
|
Include paths prevent hermetically sealed builds
|
https://api.github.com/repos/chakra-core/ChakraCore/issues/3916/comments
| 2 |
2017-10-10T12:09:14Z
|
2018-05-07T23:10:11Z
|
https://github.com/chakra-core/ChakraCore/issues/3916
| 264,203,487 | 3,916 |
[
"chakra-core",
"ChakraCore"
] |
sessionStorage in Javascript doesn't work when I run a page in local without a server directly on a disk. Example:
file:///C:/UwAmp/ww....HTML/Example.html
but it works if you read the page via the network at the adress:
file://pc_tower/ww....HTML/Example.html
Regards
André
|
Edge: sessionStorage in Javascript doesn't work in local
|
https://api.github.com/repos/chakra-core/ChakraCore/issues/3906/comments
| 1 |
2017-10-09T15:14:01Z
|
2017-10-09T16:42:44Z
|
https://github.com/chakra-core/ChakraCore/issues/3906
| 263,931,009 | 3,906 |
[
"chakra-core",
"ChakraCore"
] |
I am unable to build 1.7.2 (`ee46fc449c5e7aa6753622720177b56dc4cc3abd`) on macOS using the provided build script.
Build command:
```
./build.sh --static --icu=/usr/local/opt/icu4c/include --test-build
```
Error message:
```
error: definition of builtin function '_rotl'
unsigned int __cdecl _rotl(unsigned int value, int shift)
```
Environment:
```
$ clang --version
clang version 4.0.1 (tags/RELEASE_401/final)
Target: x86_64-apple-darwin16.7.0
Thread model: posix
InstalledDir: /usr/local/opt/llvm/bin
$ c++ --version
Apple LLVM version 9.0.0 (clang-900.0.37)
Target: x86_64-apple-darwin16.7.0
Thread model: posix
InstalledDir: /Applications/Xcode.app/Contents/Developer/Toolchains/XcodeDefault.xctoolchain/usr/bin
$ cmake --version
cmake version 3.7.1
CMake suite maintained and supported by Kitware (kitware.com/cmake).
$ brew install icu4c
Warning: icu4c 59.1 is already installed
$ sw_vers
ProductName: Mac OS X
ProductVersion: 10.12.6
BuildVersion: 16G29
$ xcodebuild -version
Xcode 9.0
Build version 9A235
```
|
Build script does not work on my macOS setup
|
https://api.github.com/repos/chakra-core/ChakraCore/issues/3905/comments
| 1 |
2017-10-09T10:30:43Z
|
2017-10-09T11:54:26Z
|
https://github.com/chakra-core/ChakraCore/issues/3905
| 263,848,793 | 3,905 |
[
"chakra-core",
"ChakraCore"
] |
A dynamic `import()` of a module that wasn't already imported statically will never resolve. This only affects `import()` done in an ES module--dynamic imports loaded using the FetchImportedModuleFromScript strategy resolve just fine.
Repro:
```js
// test.js
console.log("running test.js");
WScript.LoadScriptFile('./pig.js', 'module');
```
```js
// pig.js
console.log("running pig.js");
console.log("starting to load cow.js");
import("./cow.js").then(module => {
console.log("successfully loaded cow.mjs");
console.log(`the cow says: ${module.Cow}`);
}).catch(e => {
console.log(`error: ${e}`);
});
```
```js
// cow.js
console.log("running cow.js");
export const Cow = "cowz eat kittiez";
```
Output:
```
D:\src\ChakraCore\Build\VcBuild\bin\x64_debug>ch test.js
running test.js
running pig.js
starting to load cow.js
```
By contrast, `import()`ing from global code:
```
D:\src\ChakraCore\Build\VcBuild\bin\x64_debug>ch pig.js
running pig.js
starting to load cow.js
running cow.js
successfully loaded cow.mjs
the cow says: cowz eat kittiez
```
|
Dynamic import from within an ES module never resolves
|
https://api.github.com/repos/chakra-core/ChakraCore/issues/3902/comments
| 9 |
2017-10-08T03:15:42Z
|
2017-10-19T19:25:59Z
|
https://github.com/chakra-core/ChakraCore/issues/3902
| 263,689,066 | 3,902 |
[
"chakra-core",
"ChakraCore"
] |
If the reviver changes the result of JSON.parse, it can make infinite recursive calls, which trigger the stackoverflow.
This is a PoC code.
```
var o = {0:1,1:1};
JSON.parse('[0,0]', function () {
this[1] = o;
})
```
|
Stackoverflow in JSON.parse
|
https://api.github.com/repos/chakra-core/ChakraCore/issues/3900/comments
| 1 |
2017-10-07T08:40:08Z
|
2017-10-18T04:22:54Z
|
https://github.com/chakra-core/ChakraCore/issues/3900
| 263,626,731 | 3,900 |
[
"chakra-core",
"ChakraCore"
] |
`fast-json-stringify` benchmark results as of #3894
```
Chakra: JSON.stringify array x 1,685 ops/sec ±4.13% (77 runs sampled)
v8: JSON.stringify array x 3,421 ops/sec ±0.64% (92 runs sampled)
Chakra: JSON.stringify long string x 30,920 ops/sec ±0.46% (92 runs sampled)
v8: JSON.stringify long string x 14,839 ops/sec ±0.20% (94 runs sampled)
Chakra: JSON.stringify short string x 5,585,836 ops/sec ±3.49% (77 runs sampled)
v8 : JSON.stringify short string x 4,586,806 ops/sec ±0.36% (90 runs sampled)
Chakra: JSON.stringify obj x 1,518,252 ops/sec ±3.22% (81 runs sampled)
v8: JSON.stringify obj x 1,604,359 ops/sec ±0.75% (94 runs sampled)
```
**JSON.stringify of Array needs some attention.**
**P.S.** when I run the benchmarks under v8 powered node.js, `fast-json-stringify` was outperforming v8 significantly. However, when I run the benchmarks under chakracore powered node.js, `fast-json-stringify` wasn't performing good. This is possibly due to engine specific optimizations made there. In order not to compare apples to oranges, I kept the numbers for `fast-jason-stringify` aside.
|
perf: JSON.stringify Array performance
|
https://api.github.com/repos/chakra-core/ChakraCore/issues/3895/comments
| 3 |
2017-10-06T13:49:26Z
|
2018-03-17T00:33:21Z
|
https://github.com/chakra-core/ChakraCore/issues/3895
| 263,452,477 | 3,895 |
[
"chakra-core",
"ChakraCore"
] |
I'm trying to propagate exceptions caught during static module loading back to my event loop where they can be dealt with cleanly (by reporting them and then terminating). In order to do that, I have the following logic in my module loader:
```c
JsModuleEvaluation(module_record, &result);
JsHasException(&have_error);
if (have_error) {
JsGetAndClearException(&exception); // don't want to throw here
JsSetModuleHostInfo(module_record, JsModuleHostInfo_Exception, exception);
}
```
And later:
```c
JsGetModuleHostInfo(module, JsModuleHostInfo_Exception, &exception);
if (exception != JS_INVALID_REFERENCE)
throw_value(exception);
```
However, `exception` always comes out as NULL in the latter code and nothing is thrown. So as a followup I tried the following test:
```c
JsSetModuleHostInfo(module_record, JsModuleHostInfo_Exception, exception);
JsGetModuleHostInfo(module_record, JsModuleHostInfo_Exception, &exception);
```
`exception` goes into JsSetModuleHostInfo as a non-null reference and comes out of JsGetModuleHostInfo as NULL. The exception is apparently getting eaten once I set it on the module record.
|
Exception stored on module record gets eaten
|
https://api.github.com/repos/chakra-core/ChakraCore/issues/3871/comments
| 9 |
2017-10-04T05:23:45Z
|
2018-01-30T22:31:39Z
|
https://github.com/chakra-core/ChakraCore/issues/3871
| 262,669,376 | 3,871 |
[
"chakra-core",
"ChakraCore"
] |
We have seen in some benchmarks that class constructor calls can be extremely slow compared to object literal assignment.
For example, replacing calls to `new Point()` with `{x:0, y:0}` improves perf by 50% in the benchmark below (brought up by #3828):
````js
class Point {
constructor(x = 0, y = 0) {
this.x = x;
this.y = y;
};
}
class Param {
constructor(x1 = 0, x2 = 0, a = 0, b = 0, c = 0, d = 0) {
this.x1 = x1;
this.x2 = x2;
this.a = a;
this.b = b;
this.c = c;
this.d = d;
};
}
class Spline {
constructor(input) {
const clen = input.length - 1;
this.last = 0;
this.params = [];
let ds = new Float64Array(clen);
let ms = new Float64Array(clen);
for (let i = 0; i < clen; i++) {
let c = input[i];
let n = input[i + 1];
const dx = n.x - c.x;
const dy = n.y - c.y;
ds[i] = dx;
ms[i] = dy / dx;
const p = new Param();
if (i) {
c = ms[i - 1];
n = ms[i];
if ((c * n) <= 0) {
p.c = 0.0;
} else {
const pdx = ds[i - 1];
const com = pdx + dx;
p.c = 3.0 * com / ((com + dx) / c + (com + pdx) / n);
}
} else {
p.c = ms[0];
}
this.params[i] = p;
}
this.params[clen] = new Param();
this.params[clen].c = ms[clen - 1];
for (let i = 0; i < clen; i++) {
const p = this.params[i];
const c = input[i];
const n = input[i + 1];
const m = ms[i];
p.x1 = c.x;
p.x2 = n.x;
p.d = c.y;
const inv = 1.0 / ds[i];
const com = p.c + this.params[i + 1].c - m - m;
p.b = (m - p.c - com) * inv;
p.a = com * inv * inv;
}
this.params[clen - 1].x2 = Number.MAX_VALUE;
};
evaluate(x) {
let c = this.params[this.last];
if (x < c.x1 || x > c.x2) {
for (let i = 0, len = this.params.length; i < len; i++) {
if (x <= this.params[i].x2) {
this.last = i;
c = this.params[i];
break;
}
}
}
const dif = x - c.x1;
const out = new Point();
out.x = ((c.a * dif + c.b) * dif + c.c) * dif + c.d;
out.y = (3.0 * c.a * dif + 2.0 * c.b) * dif + c.c;
return out;
};
}
class Dac {
constructor(bits = 8) {
this.dac = new Float32Array(bits);
};
kinked(model) {
const len = this.dac.length;
const inf = 1e6;
let r2 = 2.20;
let term = false;
let vsum = 0;
if (model == 2) {
r2 = 2.00;
term = true;
}
for (let i = 0; i < len; i++) {
let vn = 1;
let rn = (term) ? r2 : inf;
let b, ti;
for (b = 0; b < i; b++) {
rn = (rn == inf) ? (1 + r2) : (1 + r2 * rn / (r2 + rn));
}
if (rn == inf) {
rn = r2;
} else {
rn = r2 * rn / (r2 + rn);
vn = vn * rn / r2;
}
for (++b; b < len; b++) {
rn++;
ti = vn / rn;
rn = r2 * rn / (r2 + rn);
vn = rn * ti;
}
vsum += vn;
this.dac[i] = vn;
}
vsum /= (1 << len);
for (let i = 0; i < len; i++) {
this.dac[i] /= vsum;
}
};
output(input) {
let value = 0;
for (let i = 0, len = this.dac.length; i < len; i++) {
if (input & (1 << i)) {
value += this.dac[i];
}
}
return value;
};
}
class OpAmp {
constructor(opamp, kvddt) {
this.opamp = new Spline(opamp);
this.kvddt = kvddt;
this.vmin = opamp[0].x;
this.vmax = opamp[opamp.length - 1].x;
this.x = 0;
};
reset() {
this.x = this.vmin;
};
solve(n, vi) {
let ak = this.vmin;
let bk = this.vmax;
const a = n + 1;
const b = this.kvddt;
let bvi = b - vi;
if (bvi < 0) { bvi = 0; }
const c = n * (bvi * bvi);
do {
const xk = this.x;
let out = this.opamp.evaluate(xk);
let bvx = b - xk;
let bvo = b - out.x;
if (bvx < 0) { bvx = 0; }
if (bvo < 0) { bvo = 0; }
const f = a * (bvx * bvx) - c - (bvo * bvo);
this.x -= (f / (2 * (bvo * out.y - a * bvx)));
if (Math.abs(this.x - xk) < 1e-8) {
out = this.opamp.evaluate(this.x);
return out.x;
}
if (f < 0) {
bk = xk;
} else {
ak = xk;
}
if (this.x <= ak || this.x >= bk) {
this.x = (ak + bk) * 0.5;
}
} while (true);
};
}
class Filter {
constructor() {
this.gain = null;
this.mixer = null;
this.resonance = null;
this.summer = null;
this.enabled = true;
this.filter = 0;
this.volume = 0;
this.vhp = 0;
this.vbp = 0;
this.vlp = 0;
this.ve = 0;
this.fc = 0;
this.hp = false;
this.bp = false;
this.lp = false;
this.filt1 = false;
this.filt2 = false;
this.filt3 = false;
this.filtE = false;
this.voice3off = false;
this.setEnable = (value) => {
this.enabled = value;
if (value) {
this.resFilt = this.filter;
} else {
this.filt1 = false;
this.filt2 = false;
this.filt3 = false;
this.filtE = false;
}
}
};
set fcLo(value) {
this.fc = (this.fc & 0x7f8) | (value & 7);
this.updateCenterFreq();
};
set fcHi(value) {
this.fc = ((value << 3) & 0x7f8) | (this.fc & 7);
this.updateCenterFreq();
};
set modeVol(value) {
this.volume = value & 0x0f;
this.lp = (value & 0x10) != 0;
this.bp = (value & 0x20) != 0;
this.hp = (value & 0x40) != 0;
this.voice3off = (value & 0x80) != 0;
this.updateMixing();
};
set resFilt(value) {
this.filter = value;
this.updateResonance((value >> 4) & 0x0f);
if (this.enabled) {
this.filt1 = (value & 1) != 0;
this.filt2 = (value & 2) != 0;
this.filt3 = (value & 4) != 0;
this.filtE = (value & 8) != 0;
}
this.updateMixing();
};
reset() {
this.fc = 0;
this.updateCenterFreq();
this.modeVol = 0;
this.resFilt = 0;
};
}
class Model6581 {
constructor() {
const voltage = [
new Point( 0.81, 10.31),
new Point( 2.40, 10.31),
new Point( 2.60, 10.30),
new Point( 2.70, 10.29),
new Point( 2.80, 10.26),
new Point( 2.90, 10.17),
new Point( 3.00, 10.04),
new Point( 3.10, 9.83),
new Point( 3.20, 9.58),
new Point( 3.30, 9.32),
new Point( 3.50, 8.69),
new Point( 3.70, 8.00),
new Point( 4.00, 6.89),
new Point( 4.40, 5.21),
new Point( 4.54, 4.54),
new Point( 4.60, 4.19),
new Point( 4.80, 3.00),
new Point( 4.90, 2.30),
new Point( 4.95, 2.03),
new Point( 5.00, 1.88),
new Point( 5.05, 1.77),
new Point( 5.10, 1.69),
new Point( 5.20, 1.58),
new Point( 5.40, 1.44),
new Point( 5.60, 1.33),
new Point( 5.80, 1.26),
new Point( 6.00, 1.21),
new Point( 6.40, 1.12),
new Point( 7.00, 1.02),
new Point( 7.50, 0.97),
new Point( 8.50, 0.89),
new Point(10.00, 0.81),
new Point(10.31, 0.81)
];
const ut = 26.0e-3;
const vdd = 12.18;
const vth = 1.31;
const wlvcr = 9;
this.voltrange = 1.5;
this.dcvoltage = 5;
this.c = 470e-12;
this.k = 1;
this.kvddt = this.k * (vdd - vth);
this.ucox = 20e-6;
this.wlsnake = 1 / 115;
this.vmin = voltage[0].x;
this.vmax = voltage[0].y;
if (this.kvddt > this.vmax) {
this.vmax = this.kvddt;
}
this.denorm = this.vmax - this.vmin;
this.norm = 1 / this.denorm;
this.n16 = this.norm * 65535;
this.voiceDc = (this.n16 * (this.dcvoltage - this.vmin)) >> 0;
this.scale14 = ((this.norm * 16383) * this.voltrange) >> 0;
this.gain = [];
this.mixer = [];
this.summer = [];
this.dacbits = 11;
this.daczero = 6.65;
this.dacscale = 2.63;
this.dac = new Dac(this.dacbits);
this.dac.kinked(1);
this.opampRev = new Uint16Array(65536);
this.vcrkvg = new Uint16Array(65536);
this.vcrnTerm = new Uint16Array(65536);
const scaled = [];
for (let i = 0, len = voltage.length; i < len; i++) {
let p = new Point();
let o = voltage[i];
p.x = (this.n16 * (o.x - o.y) + 65536) / 2;
p.y = this.n16 * (o.x - this.vmin);
scaled[i] = p;
}
const s = new Spline(scaled);
for (let i = 0; i < 65536; i++) {
const o = s.evaluate(i);
if (o.x < 0) { o.x = 0; }
this.opampRev[i] = o.x + 0.5;
}
const model = new OpAmp(voltage, this.kvddt);
for (let i = 0; i < 5; i++) {
const idiv = 2 + i;
const size = idiv << 16;
const narr = new Uint16Array(size);
model.reset();
for (let vi = 0; vi < size; vi++) {
const vin = this.vmin + vi / this.n16 / idiv;
narr[vi] = ((model.solve(idiv, vin) - this.vmin) * this.n16) + 0.5;
}
this.summer[i] = narr;
}
for (let i = 0; i < 8; i++) {
let idiv = 1;
let size = 1;
if (i) {
idiv = i;
size = i << 16;
}
const narr = new Uint16Array(size);
const n = i * 8 / 6;
model.reset();
for (let vi = 0; vi < size; vi++) {
const vin = this.vmin + vi / this.n16 / idiv;
narr[vi] = ((model.solve(n, vin) - this.vmin) * this.n16) + 0.5;
}
this.mixer[i] = narr;
}
for (let i = 0; i < 16; i++) {
const narr = new Uint16Array(65536);
const n = i / 8;
model.reset();
for (let vi = 0; vi < 65536; vi++) {
const vin = this.vmin + vi / this.n16;
narr[vi] = ((model.solve(n, vin) - this.vmin) * this.n16) + 0.5;
}
this.gain[i] = narr;
}
const nkvddt = this.n16 * this.kvddt;
const nvmin = this.n16 * this.vmin;
const kvt = this.k * vth;
const is = 2 * this.ucox * ut * ut / this.k * wlvcr;
const n15 = this.norm * 32767;
const nis = n15 * 1.0e-6 / this.c * is;
const ut2 = 2 * ut;
for (let i = 0; i < 65536; i++) {
const log = Math.log1p(Math.exp((i / this.n16 - kvt) / ut2));
this.vcrnTerm[i] = (nis * log * log) + 0.5;
this.vcrkvg[i] = (this.k * (nkvddt - Math.sqrt(i << 16)) - nvmin) + 0.5;
}
};
}
console.time("worker");
const sid6581 = new Model6581();
console.timeEnd("worker");
````
|
Optimize class constructor calls
|
https://api.github.com/repos/chakra-core/ChakraCore/issues/3869/comments
| 3 |
2017-10-04T00:06:37Z
|
2018-05-07T23:52:22Z
|
https://github.com/chakra-core/ChakraCore/issues/3869
| 262,628,667 | 3,869 |
[
"chakra-core",
"ChakraCore"
] |
Motivated by perf issues in this microbenchmark brought up in #3828. We should be able to figure out that result is only used as an int, and therefore use int division.
````js
const buffer = new Uint16Array(5000000);
start = new Date();
function bar() {
for (let i = 0, len = buffer.length; i < len; i++) {
let idiv = 1;
let size = 1;
if (i) {
idiv = i;
size = i << 16;
}
const n = i * 8 / 6;
buffer[i] = n;
}
}
for (let i = 0; i < 100; ++i) {
bar()
}
console.log(new Date() - start)
````
|
Optimize divide to use int divide when floating point result isn't needed
|
https://api.github.com/repos/chakra-core/ChakraCore/issues/3867/comments
| 2 |
2017-10-03T22:49:47Z
|
2018-05-07T23:52:27Z
|
https://github.com/chakra-core/ChakraCore/issues/3867
| 262,615,362 | 3,867 |
[
"chakra-core",
"ChakraCore"
] |
On MS Edge 15, `Element.addEventListener('click', null)` or `Element.addEventListener('click', undefined)` will throw the error `Unable to get property 'handleEvent' of undefined or null reference` when the F12 developer tools is opened.
If the script is run before opening the dev tools, the errors are not logged in the console.
To reproduce, execute this code on Edge 15 ([jsbin here](http://output.jsbin.com/bidusoloku/1/)):
```html
<body>
<script>
document.body.addEventListener('click', undefined);
document.body.addEventListener('click', null);
</script>
</body>
```
Not sure if this is a ChakraCore bug or Dev Tools bug, ended up filing it both in this repo and [here](https://developer.microsoft.com/en-us/microsoft-edge/platform/issues/14063806/), sorry in advance and feel free to close the one in the wrong place :)
|
Element.addEventListener('click', null) throws error only when dev tools opened
|
https://api.github.com/repos/chakra-core/ChakraCore/issues/3863/comments
| 3 |
2017-10-03T19:52:47Z
|
2017-10-07T00:59:48Z
|
https://github.com/chakra-core/ChakraCore/issues/3863
| 262,563,527 | 3,863 |
[
"chakra-core",
"ChakraCore"
] |
Memory::Recycler::AddMark has a mismatched exception specification between the declaration and definition. Starting with version 19.12 of MSVC this will be diagnosed under /permissive- or /std:c++17.
I've submitted PR #3853 to correct this, but I note that none of the other declarations in this class have exception specifications, so it may be preferable to remove throw() from the definition instead of adding it to the declaration. I can make that change if you prefer.
|
Exception specification mismatches will cause build breaks in new versions of VS 2017
|
https://api.github.com/repos/chakra-core/ChakraCore/issues/3862/comments
| 0 |
2017-10-03T19:24:26Z
|
2017-10-18T00:04:53Z
|
https://github.com/chakra-core/ChakraCore/issues/3862
| 262,554,902 | 3,862 |
[
"chakra-core",
"ChakraCore"
] |
Not sure what's going on here but we need to take a deeper look.
Repro:
```
utf8::NarrowToWide wname("Length", 7);
fprintf(stdout, "%d \n", wname.Length() == 6); // --> prints false!!
```
|
utf8::NarrowToWide fails to return correct length
|
https://api.github.com/repos/chakra-core/ChakraCore/issues/3852/comments
| 7 |
2017-10-02T20:48:33Z
|
2017-10-03T11:09:00Z
|
https://github.com/chakra-core/ChakraCore/issues/3852
| 262,223,656 | 3,852 |
[
"chakra-core",
"ChakraCore"
] |
See @MSLaguana's review of https://github.com/Microsoft/ChakraCore/pull/3840. We may not need to use the StringBufferAutoPtr class at all, and we can get rid of our own math here.
/cc @dilijev
|
Use Utf8Codex helper classes in intl-icu
|
https://api.github.com/repos/chakra-core/ChakraCore/issues/3851/comments
| 2 |
2017-10-02T18:32:04Z
|
2018-02-15T16:44:25Z
|
https://github.com/chakra-core/ChakraCore/issues/3851
| 262,181,767 | 3,851 |
[
"chakra-core",
"ChakraCore"
] |
Currently `null` and `undefined` are created per `JavascriptLibrary`.
- We can't compare a `JsVar` to `null` or `undefined` because we don't know its' origin.
- `JsVar` -> ToRecyc..Object -> Compare... is unnecessarily more complicated
- There are some places we know `JsVar` is actually a `Recyc..Object` hence things are easier (this improvement is already merged to master). Yet, those places are very limited
Any objections to making `null`, `undefined` values are allocated per framework only once?
Is there anything else that we could use the same approach?
|
perf: Context independent, shared null && undefined
|
https://api.github.com/repos/chakra-core/ChakraCore/issues/3847/comments
| 5 |
2017-10-02T09:34:55Z
|
2018-05-07T23:53:11Z
|
https://github.com/chakra-core/ChakraCore/issues/3847
| 262,029,497 | 3,847 |
[
"chakra-core",
"ChakraCore"
] |
I have the following pattern of code in JS
```
return new Promise((resolve, reject) => {
try {
csharpFunction(text, function (bodyContent) {
if (somthing) {
resolve(true);
}
else {
resolve(false);
}
});
} catch (err) {
reject(err);
}
```
In the managed code I am doing AddRef on the callback.
```
private JavaScriptValue CSharpFunction(JavaScriptValue callee, bool isConstructCall, JavaScriptValue[] arguments, ushort argumentCount, IntPtr callbackData)
{
var queue = (JsQueue)((GCHandle)callbackData).Target;
queue.Lock.Increment();
var text = arguments[1].ToString();
var refCount = arguments[2].AddRef();
var callback = arguments[2];
var context = JavaScriptContext.Current;
var t = Task.Run(async () =>
{
var response = await something();
queue.Schedule(new JsFunc(() =>
{
callback.CallFunction(JavaScriptValue.GlobalObject, luisResponse);
}
});
return JavaScriptValue.Undefined;
}
}
```
While calling CallFunction I am getting the following exception. Please note that **AddRef** is done on 'callback'
Managed Debugging Assistant 'CallbackOnCollectedDelegate' occurred
Message=Managed Debugging Assistant 'CallbackOnCollectedDelegate' : 'A callback was made on a garbage collected delegate of type 'Microsoft.Bots.Runtime!Microsoft.Bots.Hosting.JavaScriptNativeFunction::Invoke'. This may cause application crashes, corruption and data loss. When passing delegates to unmanaged code, they must be kept alive by the managed application until it is guaranteed that they will never be called.'
|
Collected exception even after AddRef was done
|
https://api.github.com/repos/chakra-core/ChakraCore/issues/3831/comments
| 3 |
2017-09-28T19:15:17Z
|
2017-09-28T22:42:35Z
|
https://github.com/chakra-core/ChakraCore/issues/3831
| 261,430,136 | 3,831 |
[
"chakra-core",
"ChakraCore"
] |
This test program takes about 300ms in both Chrome (canary 63) and Firefox (Nightly 58.0a1) and slightly less using a worker.
In Edge (EdgeHTML 15.15063) it takes about 13 seconds and the worker version never prints anything at all.
You can find the test program here:
http://www.neoartcr.com/edge_test/index_worker.htm
http://www.neoartcr.com/edge_test/index_noworker.htm
Keep the console open to see the time it takes to run the test.
Thanks
Christian
|
Edge javascript slowdown
|
https://api.github.com/repos/chakra-core/ChakraCore/issues/3828/comments
| 23 |
2017-09-28T02:43:17Z
|
2020-03-25T17:16:08Z
|
https://github.com/chakra-core/ChakraCore/issues/3828
| 261,176,382 | 3,828 |
[
"chakra-core",
"ChakraCore"
] |
The following code behaves differently between Chakra and v8:
```
async function bar() {
throw new Error("Whoops");
}
async function foo() {
try {
await bar();
} catch (e){
console.log("Caught ");
}
}
var oldThen = Promise.prototype.then;
Promise.prototype.then = function(thenx, catchx) {
console.log(`then: ${thenx}, catch: ${catchx}`)
oldThen.apply(this, [ thenx, catchx ]);
}
foo()
```
In ChakraCore's case, we see the following output:
```
then: function() {
[native code]
}, catch: undefined
then: undefined, catch: function() {
[native code]
}
Caught
```
In v8's case, we see the following output:
```
then: function () { [native code] }, catch: function () { [native code] }
Caught
```
Basically, v8 calls then with resolve and reject in one step, whereas ChakraCore calls it in two (then followed by catch which calls then). It appears that this behavior also trips up zone.js (if you use zone.js with the above snippet, you'll see an unhandled promise rejection in ChakraCore but not in v8).
It's unclear from reading the [spec](https://tc39.github.io/ecma262/#sec-async-functions-abstract-operations-async-function-await) whether there is one mandated approach, but it's worth understanding to see if this is a bug in ChakraCore or Zone.js is just being pedantic in it's expectation of awaited promise resolution.
|
Async-await difference between ChakraCore and v8
|
https://api.github.com/repos/chakra-core/ChakraCore/issues/3827/comments
| 1 |
2017-09-28T01:20:58Z
|
2017-12-05T16:40:49Z
|
https://github.com/chakra-core/ChakraCore/issues/3827
| 261,164,808 | 3,827 |
[
"chakra-core",
"ChakraCore"
] |
Opening this item to track a performance issue.
Issue:
Calling a method defined on C land is slow.
i.e. an example C method.
```
JsValueRef DummyFunction(JsValueRef callee, bool isConstructCall, JsValueRef *arguments, unsigned short argumentCount, void *callbackState)
{
JsValueRef undefinedValue;
if (JsGetUndefinedValue(&undefinedValue) == JsNoError)
{
return undefinedValue;
}
else
{
return nullptr;
}
}
```
No surprise. Expectedly, `call to native land` process takes time.
Can we make it better?
|
perf: calls to native land is slow [investigate]
|
https://api.github.com/repos/chakra-core/ChakraCore/issues/3821/comments
| 1 |
2017-09-27T18:56:46Z
|
2017-11-17T22:12:21Z
|
https://github.com/chakra-core/ChakraCore/issues/3821
| 261,081,781 | 3,821 |
[
"chakra-core",
"ChakraCore"
] |
Currently it seems to not be possible to name a regular (non static) class method `static`. Edge errors out with `Expected identifier, string or number` so I assume that it gets interpreted as the static keyword instead of the method name.
I've encountered this problem in Edge 15 and the Edge 16 preview.
Here's a small reproduction:
```js
class Foo {
static() {
console.log('this does not work in Edge');
}
}
const foo = new Foo();
foo.static();
```
I am not 100% sure if the name is allowed, but it seems to work in other JavaScript engines
|
Class methods cannot be called static
|
https://api.github.com/repos/chakra-core/ChakraCore/issues/3816/comments
| 6 |
2017-09-26T23:03:07Z
|
2017-10-13T00:28:46Z
|
https://github.com/chakra-core/ChakraCore/issues/3816
| 260,796,464 | 3,816 |
[
"chakra-core",
"ChakraCore"
] |
(1d,P3) Enable Intl tests in ChakraCore Linux --with-intl builds
|
https://api.github.com/repos/chakra-core/ChakraCore/issues/3811/comments
| 1 |
2017-09-26T19:45:50Z
|
2017-12-19T02:38:14Z
|
https://github.com/chakra-core/ChakraCore/issues/3811
| 260,746,541 | 3,811 |
|
[
"chakra-core",
"ChakraCore"
] |
@jackhorton Some Intl.NumberFormat tests (etc.) should pass at this point, but not all tests. Depending on a variety of factors, it may not yet make sense to do this. If so, we can delay this work.
/cc @digitalinfinity
|
(4h,P2) Enable (a subset of) Intl tests in Node-ChakraCore
|
https://api.github.com/repos/chakra-core/ChakraCore/issues/3810/comments
| 4 |
2017-09-26T19:45:46Z
|
2018-08-17T19:01:51Z
|
https://github.com/chakra-core/ChakraCore/issues/3810
| 260,746,524 | 3,810 |
[
"chakra-core",
"ChakraCore"
] |
(2h,P1) Intl.js fallback: platform.getDefaultTimeZone
|
https://api.github.com/repos/chakra-core/ChakraCore/issues/3801/comments
| 1 |
2017-09-25T20:31:01Z
|
2017-12-19T02:40:37Z
|
https://github.com/chakra-core/ChakraCore/issues/3801
| 260,405,117 | 3,801 |
|
[
"chakra-core",
"ChakraCore"
] |
Minimal sub-task of #3755
See: http://userguide.icu-project.org/icudata#TOC-ICU-Data-Directory
The easiest thing for us to do is to ship icudt59l.dat (etc) with Node-ChakraCore and have ChakraCore dynamically calculate the directory on disk where it is stored from the directory that the executable lives in + a relative path from there to the data file.
|
Determine how to compile ICU data alongside ChakraCore
|
https://api.github.com/repos/chakra-core/ChakraCore/issues/3800/comments
| 1 |
2017-09-25T20:27:34Z
|
2019-06-07T18:37:34Z
|
https://github.com/chakra-core/ChakraCore/issues/3800
| 260,404,174 | 3,800 |
[
"chakra-core",
"ChakraCore"
] |
According to documentation, JsInitializeModuleRecord() will return the already existing module record if the same normalized specifier is passed in twice. Parsing source for an already-loaded module will then return an "already parsed" error, which should be enough to ensure only one instance of a module ever gets created. There were a few cases, however, where I noticed I was loading the same module more than once, so I investigated.
As an experiment, I tried the following:
```c
JsInitializeModuleRecord(NULL, specifier_ref, &module1);
JsInitializeModuleRecord(NULL, specifier_ref, &module2);
```
Both calls succeed, but `module2 != module1`. Is the documentation incorrect, or is this a bug?
|
Calling JsInitializeModuleRecord a second time creates a new record
|
https://api.github.com/repos/chakra-core/ChakraCore/issues/3798/comments
| 8 |
2017-09-25T16:45:56Z
|
2018-05-04T23:50:09Z
|
https://github.com/chakra-core/ChakraCore/issues/3798
| 260,343,190 | 3,798 |
[
"chakra-core",
"ChakraCore"
] |
The equivalent of the following pseudocode works in Windows but fails on *nix (at least Linux and macOS) with `JsErrorWrongThread`:
```c
// assuming a runtime and context were created a while ago...
JsSetCurrentContext(JS_INVALID_REFERENCE);
JsDisposeRuntime(runtime);
JsCreateRuntime(..., &runtime); // this succeeds, but...
JsCreateContext(runtime, &context); // this returns JsErrorWrongThread!
```
This happens for me even though my engine is single-threaded and the runtime/context is therefore never called from a different thread. miniSphere has a function to restart the engine and I want to create a fresh runtime when that happens so that globals from the previous run don't leak into the new session. This issue (is it a bug?) prevents that from working.
|
Destroying and then recreating a runtime/context fails with JsErrorWrongThread
|
https://api.github.com/repos/chakra-core/ChakraCore/issues/3797/comments
| 5 |
2017-09-25T14:41:48Z
|
2018-01-24T22:33:00Z
|
https://github.com/chakra-core/ChakraCore/issues/3797
| 260,302,169 | 3,797 |
[
"chakra-core",
"ChakraCore"
] |
In miniSphere at present calling `import(...)` makes the engine segfault while static `import` works just fine, which makes me think I’m missing a callback. What do I need to do for dynamic imports to work?
|
What’s needed to make dynamic import work?
|
https://api.github.com/repos/chakra-core/ChakraCore/issues/3793/comments
| 13 |
2017-09-22T19:48:33Z
|
2018-02-12T19:19:54Z
|
https://github.com/chakra-core/ChakraCore/issues/3793
| 259,918,495 | 3,793 |
[
"chakra-core",
"ChakraCore"
] |
Using Jsrt, there is no direct way to use a `JavascriptString` object for property identifier.
`expensive loop` : Get the string from JsValueRef -> Convert to Utf16 -> Lookup / convert to JsPropertyRef... -> etc.. and finally Get OR Set the property!
So.. If we had an interface that accepts JsValueRef, we wouldn't have to do all these stuff.
This alone should change the things for node-chakracore. (AcmeAir perf wise.)
However, I would like to take it one step further and implement a string type that is very similar to Js::PropertyString with some differences.
`Js::JsrtString` (or however we are going to call)
Structure-wise, apart from being a regular `JavascriptString`... it will be holding a pointer to `PropertyRecord` + another pointer for extension / Jsrt call caching purposes. (Memory wise, it will cost +16 bytes on top of standard JavascriptString)
`PropertyRecord` ptr by default will be `nullptr`... It will be filled, once this`JsrtString` object is used for any property related purposes. Additional usages will be saving from `string TO propertyId` hashing / lookup... (name it!) And... from GC point.. Yes, as long as this string object is alive, PropertyRecord will be alive too.
`JsrtCreateString...` will create this new `JsrtString` by default. (In case the string length is bigger than 128 char, it will create the good old `JavascriptString` instead.`
** Q/A **
- Why not using `Js::PropertyString`?
That would cost +32 bytes. It also expects every string variable to be property string... and will ask for PropertyRecord from the beginning... etc. ( <- some cache related stuff )
- Expected gain?
Somewhere between 2% to 10% AcmeAIR LTO gain.
- 2% or 10% ? How did you calculate this?
Average AcmeAIR benchmark request forces the above mentioned "expensive loop" aprox. 50+ times.. In other words, ~80 times each milliseconds. (considering 1730 requests/second rate on the machine I do the benchmark)...
1 ms ==> 2.7Ghz => 2.7 e6 ticks (aprox CPU hertz) / 80times == > So... per each `33750` cpu ticks, we do one of these heavy loops. One loop costs somewhere between 2% to 10% of the available cpu ticks... Hence, this work may end-up giving 2% gain or something around 10%... or less...
- How much change needed on Chakrashim end?
Not much.
- Chakracore?
Not much.
I will start hacking soon on this.
Opinions? Objections?
|
perf: JsrtGet/Set property from JsValueRef and more.
|
https://api.github.com/repos/chakra-core/ChakraCore/issues/3790/comments
| 6 |
2017-09-22T12:52:42Z
|
2017-11-17T22:06:41Z
|
https://github.com/chakra-core/ChakraCore/issues/3790
| 259,805,124 | 3,790 |
[
"chakra-core",
"ChakraCore"
] |
Offline with @Cellule and learned`libwabt` shouldn't be part of xplat release builds.
Opening this issue to track. Assigning myself && @Cellule . One of us should take care of this.
|
libwabt: Remove from xplat release builds
|
https://api.github.com/repos/chakra-core/ChakraCore/issues/3789/comments
| 3 |
2017-09-22T09:49:01Z
|
2017-10-02T09:36:14Z
|
https://github.com/chakra-core/ChakraCore/issues/3789
| 259,761,479 | 3,789 |
[
"chakra-core",
"ChakraCore"
] |
The wiki entry: https://github.com/Microsoft/ChakraCore/wiki/JsCopyString states that JSCopyString has 5 parameters.
However this commit which was merged 17 days ago: https://github.com/Microsoft/ChakraCore/commit/f2078fed54fdeca4c3842fbd8d007de2502b06d5
Changed it to 4 parameters.
I'm not sure if this change was intentional, noting that the parameter change breaks things that depend on JSCopyString, but either way one of the two is now wrong (either the wiki or the codebase) - which one should be updated?
|
JSCopyString number of parameters changed Wiki is wrong
|
https://api.github.com/repos/chakra-core/ChakraCore/issues/3787/comments
| 2 |
2017-09-22T07:15:39Z
|
2017-09-22T16:13:40Z
|
https://github.com/chakra-core/ChakraCore/issues/3787
| 259,724,064 | 3,787 |
[
"chakra-core",
"ChakraCore"
] |
Current NuGet release is only for MSVC 2015 `v140` toolchain. miniSphere builds against `v141_xp` for maximum portability, but this prevents me from using the CC NuGet package and forces me to commit the ChakraCore binaries, which are quite large. This is unwieldy and makes upgrading to new ChakraCore releases painful.
Are there any plans to support MSVC toolchains other than `v140` in NuGet?
|
NuGet packages for MSVC 2017, XP toolchain
|
https://api.github.com/repos/chakra-core/ChakraCore/issues/3786/comments
| 7 |
2017-09-22T06:48:54Z
|
2018-01-23T05:07:19Z
|
https://github.com/chakra-core/ChakraCore/issues/3786
| 259,718,971 | 3,786 |
[
"chakra-core",
"ChakraCore"
] |
OS: macOS 10.12.6
Xcode: 9.0 (just updated)
Command Line Tools (macOS Sierra version 10.12) for Xcode: 9.0
```
/Applications/Xcode.app/Contents/Developer/Toolchains/XcodeDefault.xctoolchain/
usr/include/c++/v1/__undef_min_max:29:2:
warning: : macro max is incompatible with C++. #undefing max [-W#warnings]
```
many of these warning is happen on building
|
build: warnning issue after updated xcode9
|
https://api.github.com/repos/chakra-core/ChakraCore/issues/3779/comments
| 4 |
2017-09-21T15:51:29Z
|
2018-05-31T00:31:58Z
|
https://github.com/chakra-core/ChakraCore/issues/3779
| 259,544,970 | 3,779 |
[
"chakra-core",
"ChakraCore"
] |
Saw in CI test build. It doesn't repro.. every time.
```
... -WERExceptionSupport -ExtendedErrorStackForTestHost -BaselineMode -maxInterpretCount:1 -maxSimpleJitRunCount:1 -bgjit- -dynamicprofilecache:profile.dpl.2160 -ES6Module -Es6ToStringTag -args summary -endargs test/es6/module-namespace.js
...
... Output:
... ----------------------------
... *** Async result of test #8 (7): Issue3246: namespace property names are sorted
... Test threw exception: assert.areEqual failed: expected: $ actual: undefined: ModuleNamespace
...
... FAILED
... Summary of tests: total executed: 8; passed: 7; failed: 1
...
... ----------------------------
... exit code: 0
```
Not sure to whom this should be assigned?
|
test/es6/module-namespace.js: Random CI failure (release/1.7)
|
https://api.github.com/repos/chakra-core/ChakraCore/issues/3778/comments
| 8 |
2017-09-21T01:37:11Z
|
2018-01-04T00:08:56Z
|
https://github.com/chakra-core/ChakraCore/issues/3778
| 259,350,015 | 3,778 |
[
"chakra-core",
"ChakraCore"
] |
These tags will be added in an upcoming PR.
|
intl-icu: Fix tests tagged require_winglob so that they run under ICU as well.
|
https://api.github.com/repos/chakra-core/ChakraCore/issues/3777/comments
| 0 |
2017-09-21T01:30:47Z
|
2018-02-15T01:12:27Z
|
https://github.com/chakra-core/ChakraCore/issues/3777
| 259,349,182 | 3,777 |
[
"chakra-core",
"ChakraCore"
] |
#3750 changed the bytecode version by introducing a new PropertyId via adding to JnDirectFields.h
Since 1.7 (the parent branch) is in RELEASE mode (see lib/Common/ChakraCoreVersion.h), this means we need to update the GUID.
See lib/Runtime/ByteCode/ByteCodeCacheReleaseFileVersion.h
@jackhorton we spoke offline about how to do this.
|
intl-icu branch needs an update to bytecode version GUID
|
https://api.github.com/repos/chakra-core/ChakraCore/issues/3773/comments
| 0 |
2017-09-20T21:58:41Z
|
2017-10-03T22:35:04Z
|
https://github.com/chakra-core/ChakraCore/issues/3773
| 259,314,848 | 3,773 |
[
"chakra-core",
"ChakraCore"
] |
See: https://github.com/Microsoft/ChakraCore/pull/3769#discussion_r140083352
|
Intl.js: General refactoring of `var` to `let` or `const`
|
https://api.github.com/repos/chakra-core/ChakraCore/issues/3771/comments
| 1 |
2017-09-20T21:02:19Z
|
2018-02-15T16:51:23Z
|
https://github.com/chakra-core/ChakraCore/issues/3771
| 259,300,947 | 3,771 |
[
"chakra-core",
"ChakraCore"
] |
See: https://github.com/Microsoft/ChakraCore/pull/3769#discussion_r140082740
|
Intl.js: refactor fallback logic to be more general to any platform implementation behavior
|
https://api.github.com/repos/chakra-core/ChakraCore/issues/3770/comments
| 1 |
2017-09-20T21:01:18Z
|
2017-12-19T22:45:14Z
|
https://github.com/chakra-core/ChakraCore/issues/3770
| 259,300,670 | 3,770 |
[
"chakra-core",
"ChakraCore"
] |
Configurations for example and as-applies:
WinGlob / ICU
Windows / Linux / (optionally OSX if there's a reason to believe it's different)
ChakraCore / Node-ChakraCore
See https://github.com/bterlson/test262-harness
/cc @jackhorton @digitalinfinity
Note: waiting on completion of the majority of the [milestone 1.7 Intl-ICU tasks](https://github.com/Microsoft/ChakraCore/issues?q=is%3Aopen+is%3Aissue+label%3AIntl-ICU+milestone%3A1.7)
|
(1d,P3) Intl-ICU: Collect test262 conformance numbers in various configs
|
https://api.github.com/repos/chakra-core/ChakraCore/issues/3764/comments
| 5 |
2017-09-19T23:05:21Z
|
2018-08-17T17:34:56Z
|
https://github.com/chakra-core/ChakraCore/issues/3764
| 258,991,002 | 3,764 |
[
"chakra-core",
"ChakraCore"
] |
It's not currently possible, as far as I can tell, to make a constructor that creates an external object without returning the object directly (i.e. you can't use the default `this`), this leaves me to create the object and then manually set its prototype to `new.target.prototype` in order to support subclassing. However, that doesn't seem to be possible to do in a JsNativeFunction.
Is there an API to get the value of `new.target`? If not, I would like to propose such an API.
|
API to get value of `new.target` in native function
|
https://api.github.com/repos/chakra-core/ChakraCore/issues/3762/comments
| 12 |
2017-09-19T14:29:03Z
|
2018-01-11T23:06:06Z
|
https://github.com/chakra-core/ChakraCore/issues/3762
| 258,842,053 | 3,762 |
[
"chakra-core",
"ChakraCore"
] |
How do I put a function on a constructed object and tie my internal object to the instance? The code
```
var client = new client();
client.get("value");
```
I am able to get the first line to call a native function as a ctor call with the code..
```
JsPropertyIdRef clientId;
JsGetPropertyIdFromName(L"client", &clientId);
JsValueRef clientCtor;
JsCreateFunction(Client, nullptr, &clientctor);
JsSetProperty(globalObject, clientId, clientctor, true);
```
I want to have a function on the object (get) that can call to native, but I can't figure how. I suppose I need to associate a pointer to the instance and define a call back that is assigned to client. I've tried a number of ways but cannot figure out.
|
Construct with new keyword
|
https://api.github.com/repos/chakra-core/ChakraCore/issues/3761/comments
| 2 |
2017-09-19T14:07:41Z
|
2017-11-30T23:39:42Z
|
https://github.com/chakra-core/ChakraCore/issues/3761
| 258,834,163 | 3,761 |
[
"chakra-core",
"ChakraCore"
] |
`new Date()` or `new Date( some args here! )` ... are slow.
Regardless from the source, we convert the `Date` to local. -> https://github.com/Microsoft/ChakraCore/blob/release/1.7/lib/Runtime/Library/JavascriptDate.cpp#L231
...and then UTC etc..
Besides, each time we create a string, we go an find out the DST string for the data. We keep DST related value without much reason.
This Branch : https://github.com/obastemur/ChakraCore/tree/fast_date makes `Date` object new instance creation 4 times faster and some memory improvements. Usage of prototype methods are similarly fast (i.e. calling `getDay` etc)
However, it's a quick hack and needs `Windows` side implementation for PlatformAgnostic and some stabilization work.
I did this, in order to measure the effect of `Date` object on AcmeAIR. Well, although it is 4 times faster with this hack.. no gain on AcmeAIR end. Hence, I will left this as a TODO item with the guide branch sitting on my repo.
If `Date` shows up any heat signature on any place, kindly consider this issue and my branch as a kick start. I may also turn back to this though.
|
Date object is slow
|
https://api.github.com/repos/chakra-core/ChakraCore/issues/3760/comments
| 5 |
2017-09-19T06:09:53Z
|
2019-06-07T19:01:34Z
|
https://github.com/chakra-core/ChakraCore/issues/3760
| 258,706,554 | 3,760 |
[
"chakra-core",
"ChakraCore"
] |
Example:
let arr = [...document.querySelectorAll("*")]
This throws the error, "Function expected"
|
"Function expected" error when using spread syntax on a NodeList to convert it to array
|
https://api.github.com/repos/chakra-core/ChakraCore/issues/3759/comments
| 5 |
2017-09-19T05:31:31Z
|
2017-09-21T21:12:52Z
|
https://github.com/chakra-core/ChakraCore/issues/3759
| 258,701,041 | 3,759 |
[
"chakra-core",
"ChakraCore"
] |
We aren't turning this on any time soon, so we shouldn't unnecessarily bloat the binary with it.
|
Remove asm.js templatized jit from release builds
|
https://api.github.com/repos/chakra-core/ChakraCore/issues/3757/comments
| 0 |
2017-09-18T21:55:37Z
|
2017-09-21T02:24:20Z
|
https://github.com/chakra-core/ChakraCore/issues/3757
| 258,633,620 | 3,757 |
[
"chakra-core",
"ChakraCore"
] |
See http://userguide.icu-project.org/icudata
For now the workaround is to set `ICU_DATA` in the environment to the directory containing the ICU data .dat file (e.g. `icudt59l.dat`) to let ICU manage it's own data loading from the default data file.
We need to consider how we will package up the data and have the shipping versions of ChakraCore and Node-ChakraCore deal with this problem without having to have the user's environment set in a particular way.
/cc @digitalinfinity @jackhorton
|
(1d,P2) Intl-ICU: Custom management of ICU data lifetime (as resource, etc.)
|
https://api.github.com/repos/chakra-core/ChakraCore/issues/3755/comments
| 2 |
2017-09-18T19:08:20Z
|
2018-02-17T01:12:30Z
|
https://github.com/chakra-core/ChakraCore/issues/3755
| 258,587,976 | 3,755 |
[
"chakra-core",
"ChakraCore"
] |
If the code call APIs often such as JSON.stringify, window.postMessage, or getBoundingClientRect in the background of the Xbox, the fragmentation quickly increased such that it reached the Xbox memory limit (128M) and the app crash.
|
Large fragmentation by using JSON.stringify, window.postMessage.
|
https://api.github.com/repos/chakra-core/ChakraCore/issues/3752/comments
| 15 |
2017-09-18T14:50:10Z
|
2018-07-17T17:13:55Z
|
https://github.com/chakra-core/ChakraCore/issues/3752
| 258,508,204 | 3,752 |
[
"chakra-core",
"ChakraCore"
] |
Is there any way to deal with asynchronous javascript? Can you return promise and then await in C#?
Or is there any other way to return to c# value that come from js async code, maybe can you call c# callback from js?
|
Embedding ChakraCore in C# - async
|
https://api.github.com/repos/chakra-core/ChakraCore/issues/3748/comments
| 1 |
2017-09-15T20:34:00Z
|
2018-02-13T23:48:32Z
|
https://github.com/chakra-core/ChakraCore/issues/3748
| 258,156,839 | 3,748 |
[
"chakra-core",
"ChakraCore"
] |
Follow-up to #2751 (PR #3635)
See comments on PR: https://github.com/Microsoft/ChakraCore/pull/3635
/cc @xiaoyinl
|
Follow-up #2751: Number#toString: add tests on the edge cases between written-out and exponential notation
|
https://api.github.com/repos/chakra-core/ChakraCore/issues/3740/comments
| 0 |
2017-09-15T01:32:08Z
|
2017-10-21T01:52:54Z
|
https://github.com/chakra-core/ChakraCore/issues/3740
| 257,901,881 | 3,740 |
[
"chakra-core",
"ChakraCore"
] |
Follow-up to #2751 (PR #3635)
Compatibility: align with other engines' implementation here.
See @littledan's comment https://github.com/Microsoft/ChakraCore/issues/2751#issuecomment-326902301
/cc @xiaoyinl @littledan @bterlson
|
Follow-up #2751: For Number#toString with a radix other than 10, never switch to exponential display
|
https://api.github.com/repos/chakra-core/ChakraCore/issues/3739/comments
| 2 |
2017-09-15T01:28:24Z
|
2017-10-21T01:52:53Z
|
https://github.com/chakra-core/ChakraCore/issues/3739
| 257,901,393 | 3,739 |
[
"chakra-core",
"ChakraCore"
] |
This type doesn't disable interrupts.
|
AutoDisableInterrupt is a confusing name
|
https://api.github.com/repos/chakra-core/ChakraCore/issues/3735/comments
| 2 |
2017-09-14T22:37:19Z
|
2019-06-07T19:08:44Z
|
https://github.com/chakra-core/ChakraCore/issues/3735
| 257,876,676 | 3,735 |
[
"chakra-core",
"ChakraCore"
] |
Edge 14 & 15 on Windows 10.
`var a = new Date().toLocaleTimeString('en-us', { timeZoneName: 'short' }).split(' ')[2].split('')[0];`
Inspecting `a` reports an empty string. But `a === ''` or `a == ''` reports `false`. What is going on here?
In my timezone, `var tz = new Date().toLocaleTimeString('en-us', { timeZoneName: 'short' }).split(' ')[2]` reports `CEST`, but `tz.length` reports `5`. This should obviously be `4`. I can get around this problem like so, but would rather not use this workaround at all:
```
var tzArray = tz.split('');
tz = tzArray.slice(1, tzArray.length).join('');
tz.length // reports 4
```
These problems don't happen in Chrome / Firefox / Safari / Opera.
|
Unknown empty character in string (BiDi markers inserted by WinGlob in Intl-defined functions)
|
https://api.github.com/repos/chakra-core/ChakraCore/issues/3728/comments
| 5 |
2017-09-14T09:20:32Z
|
2017-11-08T00:45:51Z
|
https://github.com/chakra-core/ChakraCore/issues/3728
| 257,650,573 | 3,728 |
[
"chakra-core",
"ChakraCore"
] |
1 - Virtual memory backend we use on xplat is pretty expensive.
- memory region is stored on a linked list
- Each VirtualProtect / Commit etc. requests end-up searching through the linked list.
- we acquire a lock each time since this linked list is process wide.
#3725 reduces the number of times we loop the whole list.
2 - Backend is unaware the intentions.
Consumer (CC) already has start boundary information (and others) on many places. However, current common interface we use for both Windows and xplat doesn't allow us to use the data without looking up for memory region store.
3 - Complicated mprotect usage.
mprotect is not that cheap to call on small parts of the mapped area.
|
TODO: Platform agnostic virtual memory management
|
https://api.github.com/repos/chakra-core/ChakraCore/issues/3726/comments
| 3 |
2017-09-14T04:58:22Z
|
2017-11-21T08:19:43Z
|
https://github.com/chakra-core/ChakraCore/issues/3726
| 257,595,344 | 3,726 |
[
"chakra-core",
"ChakraCore"
] |
(1d) Intl-ICU: platform.cacheNumberFormat
|
https://api.github.com/repos/chakra-core/ChakraCore/issues/3716/comments
| 0 |
2017-09-13T00:11:57Z
|
2017-10-03T22:35:05Z
|
https://github.com/chakra-core/ChakraCore/issues/3716
| 257,220,535 | 3,716 |
|
[
"chakra-core",
"ChakraCore"
] |
Getting segmentation fault when running `npm prune` on `body-parser` module.
Ref: https://github.com/expressjs/body-parser/pull/233#issuecomment-328359075
Debugged this a bit and noticed something wrong inside `OP_GetElementI()`.
Repro steps:
1. clone `body-parser`
2. `npm install`
3. `npm prune`
|
Segmentation fault inside OP_GetElementI
|
https://api.github.com/repos/chakra-core/ChakraCore/issues/3715/comments
| 2 |
2017-09-12T22:20:49Z
|
2018-01-22T22:58:06Z
|
https://github.com/chakra-core/ChakraCore/issues/3715
| 257,201,380 | 3,715 |
[
"chakra-core",
"ChakraCore"
] |
See: https://tc39.github.io/ecma402/#sup-string.prototype.tolocaleuppercase
|
Intl (ECMA 402 Draft) Implement Intl.js version of String.prototype.toLocaleUpperCase
|
https://api.github.com/repos/chakra-core/ChakraCore/issues/3711/comments
| 1 |
2017-09-12T03:39:08Z
|
2019-06-07T18:37:38Z
|
https://github.com/chakra-core/ChakraCore/issues/3711
| 256,908,184 | 3,711 |
[
"chakra-core",
"ChakraCore"
] |
See: https://tc39.github.io/ecma402/#sup-string.prototype.tolocalelowercase
|
Intl (ECMA 402 Draft) Implement Intl.js version of String.prototype.toLocaleLowerCase
|
https://api.github.com/repos/chakra-core/ChakraCore/issues/3710/comments
| 4 |
2017-09-12T03:39:02Z
|
2019-06-07T18:37:43Z
|
https://github.com/chakra-core/ChakraCore/issues/3710
| 256,908,173 | 3,710 |
[
"chakra-core",
"ChakraCore"
] |
If they are indeed different. If they are NOT effectively different in our implementation (I believe this is the case in our implementation), add a note in a comment Intl.js to make this clear for future maintainers.
See: https://github.com/tc39/ecma402/issues/173
See: https://tc39.github.io/ecma402/#sup-array.prototype.tolocalestring
See: https://tc39.github.io/ecma402/#sup-array.prototype.tolocalestring
See also https://tc39.github.io/ecma262/#sec-%typedarray%.prototype.tolocalestring which is dependent on the `Array.prototype.toLocaleString` implementation, and which is much simpler to reason about if `Array.prototype.toLocaleString` is actually implemented in native code instead of JS, the same regardless of Intl (because the only difference is in passing the reserved parameters around, which has the correct effect whether Intl is present or not).
|
Intl: Implement Intl.js version of Array.prototype.toLocaleString (if different)
|
https://api.github.com/repos/chakra-core/ChakraCore/issues/3709/comments
| 0 |
2017-09-12T03:38:35Z
|
2019-06-07T18:37:47Z
|
https://github.com/chakra-core/ChakraCore/issues/3709
| 256,908,118 | 3,709 |
[
"chakra-core",
"ChakraCore"
] |
To illustrate just one of the glitches I've encountered (there have been others):
```
SSj 5.0a1 Sphere JavaScript debugger (x64)
the powerful symbolic JS debugger for Sphere
(c) 2015-2017 Fat Cerberus
starting 'C:/src/kh2bar/dist/'... OK.
connecting to 127.0.0.1:1208... OK.
establishing communication... OK.
downloading game information... OK.
platform: miniSphere 5.0a1
title: kh2Bar Showcase
author: Fat Cerberus
-> # 0: module(), at #/game_modules/sphere-runtime.mjs:33
33 export const assert = require('assert'),
#/game_modules/sphere-runtime.mjs:33 module()
(ssj) bp src/main.mjs:32
breakpoint # 0 set at src/main.mjs:32.
32 this.colorID = 0;
#/game_modules/sphere-runtime.mjs:33 module()
(ssj) c
log: kh2Bar Showcase Debug Console
log: Sphere v2 API lv. 0 (miniSphere 5.0a1)
log:
-> # 0: Showcase(), at src/main.mjs:32
32 this.colorID = 0;
src/main.mjs:32 Showcase()
(ssj) l
27 {
28 constructor()
29 {
30 super({ priority: 0 });
31
-> 32 this.colorID = 0;
33 this.comboTimer = 0;
34 this.hp = 812;
35 this.isHidden = false;
36
src/main.mjs:32 Showcase()
(ssj) e this
eval() = class Showcase extends Thread
{
constructor()
{
super({ priority: 0 });
this.colorID = 0;
this.comboTimer = 0;
this.hp = 812;
this.isHidden = false;
// construct a new HP gauge and make it visible
this.lifeBar = new HPGauge(16
src/main.mjs:32 Showcase()
(ssj) _
```
Why did `this` evaluate to the class itself rather than an instance?
|
Scoping seems screwy for JsDiagEvaluate()
|
https://api.github.com/repos/chakra-core/ChakraCore/issues/3708/comments
| 6 |
2017-09-12T03:12:42Z
|
2018-06-07T00:08:45Z
|
https://github.com/chakra-core/ChakraCore/issues/3708
| 256,904,338 | 3,708 |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.