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"
] | https://github.com/Microsoft/ChakraCore/wiki/Build-Status-%28linux%29#legacy-builds
```
00:15:57 d:\j\workspace\daily_dev12_x---b34c9c6b\lib\common\Codex\Utf8Codex.cpp(486): error C2220: warning treated as error - no 'object' file generated [d:\j\workspace\daily_dev12_x---b34c9c6b\lib\common\Codex\Chakra.Common.Codex.vcxproj]
00:15:57 d:\j\workspace\daily_dev12_x---b34c9c6b\lib\common\Codex\Utf8Codex.cpp(513) : see reference to function template instantiation 'size_t utf8::EncodeIntoImpl<true>(LPUTF8,const char16 *,charcount_t)' being compiled
00:15:57 d:\j\workspace\daily_dev12_x---b34c9c6b\lib\common\Codex\Utf8Codex.cpp(486): warning C4127: conditional expression is constant [d:\j\workspace\daily_dev12_x---b34c9c6b\lib\common\Codex\Chakra.Common.Codex.vcxproj]
```
Repro locally: https://github.com/Microsoft/ChakraCore/wiki/Jenkins-Repro-Steps#legacy-builds-windows-only
| [Legacy] [Linux Branch] {all flavors} Build Error | https://api.github.com/repos/chakra-core/ChakraCore/issues/1398/comments | 3 | 2016-08-06T00:47:06Z | 2016-08-25T21:33:34Z | https://github.com/chakra-core/ChakraCore/issues/1398 | 169,720,179 | 1,398 |
[
"chakra-core",
"ChakraCore"
] | ERROR: type should be string, got "https://github.com/Microsoft/ChakraCore/wiki/Build-Status-%28linux%29#builds-with-slow-tests\n\n```\n16:19:06 1>ERROR: Test failed to run correctly: pass not found in output file ():\n16:19:06 1> ch.exe -DumpOnCrash -WERExceptionSupport -ExtendedErrorStackForTestHost -BaselineMode -maxInterpretCount:1 -maxSimpleJitRunCount:1 -bgjit- -dynamicprofilecache:profile.dpl.UnnamedTest1675 -es6tostringtag -ES6ObjectLiterals -ES6Species -args summary -endargs D:\\j\\workspace\\daily_slow_x8---e1770805\\test\\es6\\ES6TypedArrayExtensions.js >testout1615 2>&1\n16:19:06 1>ERROR: name of output file: D:\\j\\workspace\\daily_slow_x8---e1770805\\test\\es6\\testout1615; size: 350; creation: Fri Aug 5 16:14:58 2016, last access: Fri Aug 5 16:14:58 2016, now: Fri Aug 5 16:14:59 2016\n16:19:06 1>ERROR: bad output file follows ============\n16:19:06 1>*** Running test #12 (11): %TypedArrayPrototype%[@@toStringTag] behavior\n16:19:06 1>Test threw exception: assert.throws failed: expected: TypeError 'this' is not a typed array object, actual: <no exception>: %TypedArrayPrototype%[@@toStringTag] throws TypeError if called with no parameter\n16:19:06 1>FAILED\n16:19:06 1>Summary of tests: total executed: 67; passed: 66; failed: 1\n16:19:06 1>ERROR: end of bad output file ============\n```\n\nHow to repro locally: https://github.com/Microsoft/ChakraCore/wiki/Jenkins-Repro-Steps#windows-daily\n" | [Slow] [Linux Branch] {x64,x86} x {debug,test} Test failure: TypedArrayPrototype | https://api.github.com/repos/chakra-core/ChakraCore/issues/1397/comments | 3 | 2016-08-06T00:44:08Z | 2016-08-21T15:33:53Z | https://github.com/chakra-core/ChakraCore/issues/1397 | 169,719,979 | 1,397 |
[
"chakra-core",
"ChakraCore"
] | [ECMAScript unified the handling](https://github.com/tc39/ecma262/pull/410) of indexes across all the TypedArray and related data structures. A number of minor semantics have changed.
| Implement TypedArray updates to index handling | https://api.github.com/repos/chakra-core/ChakraCore/issues/1391/comments | 5 | 2016-08-05T17:43:19Z | 2017-01-11T22:12:34Z | https://github.com/chakra-core/ChakraCore/issues/1391 | 169,659,563 | 1,391 |
[
"chakra-core",
"ChakraCore"
] | Server Error in '/' Application.
A potentially dangerous Request.Form value was detected from the client (ctl00$ctl00$ctl00$ctl00$cphRoot$cphSite$cphLC$cphC$txtCandidate_Notes_Web="...Maghsoudi <dralireza.maghsoudi...").
Description: Request Validation has detected a potentially dangerous client input value, and processing of the request has been aborted. This value may indicate an attempt to compromise the security of your application, such as a cross-site scripting attack. You can disable request validation by setting validateRequest=false in the Page directive or in the configuration section. However, it is strongly recommended that your application explicitly check all inputs in this case.
Exception Details: System.Web.HttpRequestValidationException: A potentially dangerous Request.Form value was detected from the client (ctl00$ctl00$ctl00$ctl00$cphRoot$cphSite$cphLC$cphC$txtCandidate_Notes_Web="...Maghsoudi <dralireza.maghsoudi...").
Source Error:
[No relevant source lines]
Source File: c:\Windows\Microsoft.NET\Framework64\v2.0.50727\Temporary ASP.NET Files\root\a24ecd5a\8d0ac35\App_Web_registerdetails.aspx.2587f01f.5ikejyg9.0.cs Line: 0
Stack Trace:
[HttpRequestValidationException (0x80004005): A potentially dangerous Request.Form value was detected from the client (ctl00$ctl00$ctl00$ctl00$cphRoot$cphSite$cphLC$cphC$txtCandidate_Notes_Web="...Maghsoudi <dralireza.maghsoudi...").]
System.Web.HttpRequest.ValidateString(String s, String valueName, String collectionName) +11188987
System.Web.HttpRequest.ValidateNameValueCollection(NameValueCollection nvc, String collectionName) +71
System.Web.HttpRequest.get_Form() +178
System.Web.HttpRequest.get_HasForm() +11189223
System.Web.UI.Page.GetCollectionBasedOnMethod(Boolean dontReturnNull) +124
System.Web.UI.Page.DeterminePostBackMode() +83
System.Web.UI.Page.ProcessRequestMain(Boolean includeStagesBeforeAsyncPoint, Boolean includeStagesAfterAsyncPoint) +11155303
System.Web.UI.Page.ProcessRequest(Boolean includeStagesBeforeAsyncPoint, Boolean includeStagesAfterAsyncPoint) +11154842
System.Web.UI.Page.ProcessRequest() +91
System.Web.UI.Page.ProcessRequest(HttpContext context) +240
ASP.candidate_register_registerdetails_aspx.ProcessRequest(HttpContext context) in c:\Windows\Microsoft.NET\Framework64\v2.0.50727\Temporary ASP.NET Files\root\a24ecd5a\8d0ac35\App_Web_registerdetails.aspx.2587f01f.5ikejyg9.0.cs:0
System.Web.CallHandlerExecutionStep.System.Web.HttpApplication.IExecutionStep.Execute() +599
System.Web.HttpApplication.ExecuteStep(IExecutionStep step, Boolean& completedSynchronously) +171
Version Information: Microsoft .NET Framework Version:2.0.50727.5485; ASP.NET Version:2.0.50727.5491
| Commissioner for Justice | https://api.github.com/repos/chakra-core/ChakraCore/issues/1364/comments | 0 | 2016-07-31T21:51:27Z | 2016-07-31T21:58:26Z | https://github.com/chakra-core/ChakraCore/issues/1364 | 168,540,273 | 1,364 |
[
"chakra-core",
"ChakraCore"
] | Somehow the error message is not correctly propagated when not using -on:WasmLazyTrap or -on:WasmDeferred
```
Error: Compiling wasm failed: JavaScript compilation error
at Global code (c:\projects\chakra\core\test\wasm\misc.js:8:1)
```
instead of
```
Error: Compiling wasm failed: function trunc[2]: Operator F32Trunc NYI
at Global code (c:\projects\chakra\core\test\wasm\misc.js:18:1)
```
| WebAssembly wrong error message without LazyTrap | https://api.github.com/repos/chakra-core/ChakraCore/issues/1359/comments | 0 | 2016-07-29T22:29:40Z | 2016-12-20T00:33:25Z | https://github.com/chakra-core/ChakraCore/issues/1359 | 168,426,972 | 1,359 |
[
"chakra-core",
"ChakraCore"
] | Under Ubuntu 16.04 I built the Hello World sample from here: https://github.com/Microsoft/Chakra-Samples/blob/master/ChakraCore%20Samples/Hello%20World/Linux_OSX/sample.cpp (I took a copy of the Linux_OSX directory and put it in my ChakraCore directory) . That worked fine, but when I attempted to instead dynamically link to libChakraCore.so I am getting a failed assert in ThreadContextTLSEntry::InitializeThread
I built ChakraCore from the linux branch (commit 23098cf0998a6cafaf09569667d9abc73d2e0183) using:
```
./build.sh --debug
```
I built the sample using the following:
```
root@ljw-desktop:~/ChakraCore/Linux_OSX# clang++ sample.cpp -g -lstdc++ -std=c++0x -I../lib/Jsrt -L../BuildLinux/Debug/ -pthread -lm -ldl -licuuc -lunwind-x86_64 -lChakraCore -Wno-c++11-compat-deprecated-writable-strings -Wno-deprecated-declarations -Wno-unknown-warning-option -o sample_dynamic
```
Note that is using the following clang++:
```
root@ljw-desktop:~/ChakraCore# clang++ --version
clang version 3.8.0-2ubuntu4 (tags/RELEASE_380/final)
Target: x86_64-pc-linux-gnu
Thread model: posix
InstalledDir: /usr/bin
```
I then set LD_LIBRARY_PATH to pick up the ChakraCore shared library:
```
export LD_LIBRARY_PATH=/root/ChakraCore/BuildLinux/Debug
```
Running under gdb I get the following backtrace
```
root@ljw-desktop:~/ChakraCore/Linux_OSX# gdb ./sample_dynamic
GNU gdb (Ubuntu 7.11.1-0ubuntu1~16.04) 7.11.1
Copyright (C) 2016 Free Software Foundation, Inc.
License GPLv3+: GNU GPL version 3 or later <http://gnu.org/licenses/gpl.html>
This is free software: you are free to change and redistribute it.
There is NO WARRANTY, to the extent permitted by law. Type "show copying"
and "show warranty" for details.
This GDB was configured as "x86_64-linux-gnu".
Type "show configuration" for configuration details.
For bug reporting instructions, please see:
<http://www.gnu.org/software/gdb/bugs/>.
Find the GDB manual and other documentation resources online at:
<http://www.gnu.org/software/gdb/documentation/>.
For help, type "help".
Type "apropos word" to search for commands related to "word"...
Reading symbols from ./sample_dynamic...done.
(gdb) run
Starting program: /root/ChakraCore/Linux_OSX/sample_dynamic
[Thread debugging using libthread_db enabled]
Using host libthread_db library "/lib/x86_64-linux-gnu/libthread_db.so.1".
Program received signal SIGILL, Illegal instruction.
0x00007ffff54017e4 in ThreadContextTLSEntry::InitializeThread () at /root/ChakraCore/lib/Runtime/Base/ThreadContextTlsEntry.cpp:31
31 Assert(s_tlsSlot != TLS_OUT_OF_INDEXES);
(gdb) backtrace
#0 0x00007ffff54017e4 in ThreadContextTLSEntry::InitializeThread () at /root/ChakraCore/lib/Runtime/Base/ThreadContextTlsEntry.cpp:31
#1 0x00007ffff527635a in VALIDATE_ENTER_CURRENT_THREAD () at /root/ChakraCore/lib/Jsrt/JsrtHelper.cpp:190
#2 0x00007ffff5215746 in CreateRuntimeCore (attributes=JsRuntimeAttributeNone, optRecordUri=0x0, optRecordUriCount=0, optDebugUri=0x0, optDebugUriCount=0, snapInterval=4294967295, snapHistoryLength=4294967295,
threadService=0x0, runtimeHandle=0x7fffffffe0f0) at /root/ChakraCore/lib/Jsrt/Jsrt.cpp:106
#3 0x00007ffff521686c in JsCreateRuntime (attributes=JsRuntimeAttributeNone, threadService=0x0, runtimeHandle=0x7fffffffe0f0) at /root/ChakraCore/lib/Jsrt/Jsrt.cpp:410
#4 0x0000000000400b63 in main () at sample.cpp:23
```
| ThreadContextTLSEntry::InitializeThread Assert hit when dynamically linked to libChakraCore.so | https://api.github.com/repos/chakra-core/ChakraCore/issues/1353/comments | 9 | 2016-07-28T22:39:02Z | 2016-08-10T14:24:51Z | https://github.com/chakra-core/ChakraCore/issues/1353 | 168,213,227 | 1,353 |
[
"chakra-core",
"ChakraCore"
] | ArchLinux usually builds out of the box whatever other distro can build.
In this case there is a good old problem about undeclared identifier via mbstring
``` sh
src/cruntime/mbstring.cpp:183:20: error: use of undeclared identifier 'min'
ret += min(count, strlen((const char*)string));
```
It seems to be something related to the `clang` version which is currently `version 3.8.0 (tags/RELEASE_380/final)`.
Following the `uname`:
Linux localhost 4.6.4-1-ARCH #1 SMP PREEMPT Mon Jul 11 19:12:32 CEST 2016 x86_64 GNU/Linux
All dependencies have been installed. Any way we can solve this?
AFAIK ArchLinux is the second most popular distro after Ubuntu, it would be great.
Thanks for any sort of hint or fix.
| ArchLinux undeclared identifier min / max | https://api.github.com/repos/chakra-core/ChakraCore/issues/1352/comments | 2 | 2016-07-28T06:46:41Z | 2016-07-28T14:25:09Z | https://github.com/chakra-core/ChakraCore/issues/1352 | 168,024,950 | 1,352 |
[
"chakra-core",
"ChakraCore"
] | ```
// IE11 / Edge
new Date().toLocaleString( 'Ja-JP',{ month: 'long', day: 'numeric' } );
"7月月27日"
// Chrome / Firefox
new Date().toLocaleString( 'Ja-JP',{ month: 'long', day: 'numeric' } );
"7月27日"
```
The second 月 is redundant.
This affects both IE11 and Edge. There are various permutations of this bug across Japanese and Chinese.
| date.toLocaleString shows duplicate 月 character when 'month' is specified as 'long' | https://api.github.com/repos/chakra-core/ChakraCore/issues/1351/comments | 2 | 2016-07-28T00:55:25Z | 2016-07-28T18:14:11Z | https://github.com/chakra-core/ChakraCore/issues/1351 | 167,990,201 | 1,351 |
[
"chakra-core",
"ChakraCore"
] | ``` js
eval('new.target');
```
Actual: ReferenceError: '<_lexicalNewTargetSymbol>' is undefined
Expected: ReferenceError: new.target is not allowed here (or something)
| new.target inside eval has bad error message | https://api.github.com/repos/chakra-core/ChakraCore/issues/1348/comments | 1 | 2016-07-27T21:32:21Z | 2016-08-08T16:12:44Z | https://github.com/chakra-core/ChakraCore/issues/1348 | 167,961,777 | 1,348 |
[
"chakra-core",
"ChakraCore"
] | Per https://github.com/tc39/ecma262/pull/593 - if the result of `[[GetOwnPropertyDescriptor]]` is undefined, it should _not_ be stored on the returned descriptors object.
| Update `Object.getOwnPropertyDescriptors` with spec changes | https://api.github.com/repos/chakra-core/ChakraCore/issues/1342/comments | 4 | 2016-07-26T21:43:34Z | 2016-07-29T20:31:04Z | https://github.com/chakra-core/ChakraCore/issues/1342 | 167,719,396 | 1,342 |
[
"chakra-core",
"ChakraCore"
] | ``` js
'use strict';
for(var x = 0 in {}) { }
```
Expected: syntax error
Actual: x is initialized to 0
(This should remain working in sloppy mode. See also: https://github.com/tc39/ecma262/pull/614).
| For-in initializer should be a syntax error in strict mode | https://api.github.com/repos/chakra-core/ChakraCore/issues/1340/comments | 0 | 2016-07-26T19:06:37Z | 2016-07-27T01:09:55Z | https://github.com/chakra-core/ChakraCore/issues/1340 | 167,685,996 | 1,340 |
[
"chakra-core",
"ChakraCore"
] | Test does pass on test builds. On debug builds, we hit the following assert:
ASSERT [THREAD ] at ../../pal/src/thread/pal_thread.cpp.2858: Expression: status == 0, Description: pthread_getattr_np call failed
| Regex/BoiHardFail.js fails on Fedora Debug Builds | https://api.github.com/repos/chakra-core/ChakraCore/issues/1334/comments | 3 | 2016-07-26T01:40:36Z | 2016-08-29T22:06:58Z | https://github.com/chakra-core/ChakraCore/issues/1334 | 167,503,070 | 1,334 |
[
"chakra-core",
"ChakraCore"
] | Currently, the spec says this behavior is implementation defined. We have decided that means to default these parameters to zero. All other implementations simply return NaN. I think we should align if possible and update the spec to mandate this behavior. Are there compelling reasons to keep our current behavior? It'd be a compat break for UWP/etc. but this behavior doesn't seem particularly useful so hopefully it's not used...
See also https://github.com/whatwg/javascript/issues/31
| Date.UTC should return NaN with < 2 arguments | https://api.github.com/repos/chakra-core/ChakraCore/issues/1318/comments | 7 | 2016-07-22T21:04:39Z | 2016-08-17T06:33:25Z | https://github.com/chakra-core/ChakraCore/issues/1318 | 167,132,397 | 1,318 |
[
"chakra-core",
"ChakraCore"
] | https://github.com/Microsoft/ChakraCore/wiki/Build-Status#legacy-builds
http://dotnet-ci.cloudapp.net/job/Microsoft_ChakraCore/job/master/job/daily_dev12_x64_debug/109/consoleFull#85178576a82fefab-f698-416f-8fca-58544c94cd4e
```
10:58:08
d:\j\workspace\daily_dev12_x---55cdd258\lib\common\Codex\Utf8Codex.cpp(486): error C2220: warning treated as error - no 'object' file generated [d:\j\workspace\daily_dev12_x---55cdd258\lib\common\Codex\Chakra.Common.Codex.vcxproj]
10:58:08 d:\j\workspace\daily_dev12_x---55cdd258\lib\common\Codex\Utf8Codex.cpp(513) : see reference to function template instantiation 'size_t utf8::EncodeIntoImpl<true>(LPUTF8,const char16 *,charcount_t)' being compiled
10:58:08 d:\j\workspace\daily_dev12_x---55cdd258\lib\common\Codex\Utf8Codex.cpp(486): warning C4127: conditional expression is constant [d:\j\workspace\daily_dev12_x---55cdd258\lib\common\Codex\Chakra.Common.Codex.vcxproj]
```
**Edit:** Failing in all flavors.
| [Legacy] Build failures in master | https://api.github.com/repos/chakra-core/ChakraCore/issues/1308/comments | 1 | 2016-07-22T00:04:56Z | 2016-08-09T02:03:37Z | https://github.com/chakra-core/ChakraCore/issues/1308 | 166,944,550 | 1,308 |
[
"chakra-core",
"ChakraCore"
] | Related to _https://github.com/angular/angular/issues/10080_
Seems like for formatting any component from `Intl.DateTimeFormat` it add and empty unicode character before result `‎`.
Reproduced in plnkr _http://plnkr.co/edit/57fHqSmRBsESgtJw79pq?p=preview_
In chrome - length of 19 is 2.
In edge - length of 19 is 3.
If you try to conert edge displayed result you may see `‎19`
| Intl.DateTimeFormat adds extra unicode empty characters in Edge | https://api.github.com/repos/chakra-core/ChakraCore/issues/1282/comments | 2 | 2016-07-15T15:33:02Z | 2016-08-06T01:00:09Z | https://github.com/chakra-core/ChakraCore/issues/1282 | 165,815,207 | 1,282 |
[
"chakra-core",
"ChakraCore"
] | I'm on Windows 7 and VS2012 and tried to build the ChakraCore but no luck.
I prefer to have just ChakraCore.dll , plug that into my C# app and start using it ...
Can you guys provide the X86 and X64 ChakraCore.dll files...
Please don't make stupid like Linux that we have to build it all the time from source.
| ChakraCore.dll should be available directly for use | https://api.github.com/repos/chakra-core/ChakraCore/issues/1281/comments | 6 | 2016-07-15T15:10:54Z | 2016-07-18T18:42:19Z | https://github.com/chakra-core/ChakraCore/issues/1281 | 165,809,808 | 1,281 |
[
"chakra-core",
"ChakraCore"
] | Hi there,
I'm a researcher studying software evolution. As part of my current research, I'm studying the implications of open-sourcing a proprietary software, for instance, if the project succeed in attracting newcomers. However, I observed that some projects, like _ChakraCore_, deleted their software history when transitioning to open-source.
https://github.com/Microsoft/ChakraCore/commit/5d8406741f8c60e7bf0a06e4fb71a5cf7a6458dc
Knowing that software history is indispensable for developers (e.g., developers need to refer to history several times a day), I would like to ask ChakraCore developers the following four brief questions:
1. Why did you decide to not keep the software history?
2. Do the core developers faced any kind of problems, when trying to refer to the old history? If so, how did they solve these problems?
3. Do the newcomers faced any kind of problems, when trying to refer to the old history? If so, how did they solve these problems?
4. How does the lack of history impacted on software evolution? Does it placed any burden in understanding and evolving the software?
Thanks in advance for your collaboration,
Gustavo Pinto, PhD
http://www.gustavopinto.org
| Why the software history was not kept? | https://api.github.com/repos/chakra-core/ChakraCore/issues/1280/comments | 3 | 2016-07-15T02:16:29Z | 2018-03-14T21:12:44Z | https://github.com/chakra-core/ChakraCore/issues/1280 | 165,697,643 | 1,280 |
[
"chakra-core",
"ChakraCore"
] | Hi!
We are developing an application for Windows 10 where we are using Chakra engine. In this application the end-user can write JavaScript code which we run inside our app.
We like to deliver the users some kind of possibility to debug their code. But I couldn't find any way yet as to how can I debug the code which I run on Chakra.
Now, we have a sample app about this, where I simply create a Js Runtime and Context, then I set up CurrentContext, then I simply have the JS Code in a string which I running on Chakra with JSRunScript.
I tried to call JsStartDebugging which should enable us to debug the code with Visual studio, but it didn't helped altough the call returns with NoError.
Could you give me some info about this? Is this even possible somehow?
| How can I debug JS code running in an app which embeds ChakraCore? | https://api.github.com/repos/chakra-core/ChakraCore/issues/1274/comments | 10 | 2016-07-13T09:31:45Z | 2016-12-21T09:34:03Z | https://github.com/chakra-core/ChakraCore/issues/1274 | 165,277,328 | 1,274 |
[
"chakra-core",
"ChakraCore"
] | Today we don't generate fast path for below case.
``` js
var x = new Array(5); // argument should be constant between 0 and 8 inclusive.
```
That is because in [lowerer](https://github.com/Microsoft/ChakraCore/blob/master/lib/Backend/Lower.cpp#L4935) , we fetch the wrong source of instruction. I tried fixing it by extracting the right argument and based on that emit fast path. For below test case I saw 10% win.
``` js
var start = new Date().getTime();
function test() {
var arr = [];
for (var j = 0; j < 100; j++) {
arr.push(new Array(5));
}
return arr;
}
for (var i = 0; i < 1000000; i++) {
test();
}
var end = new Date().getTime();
var time = end - start;
print(time);
```
However, I later realized that `Array` ctor can take multiple parameters that are array elements and we need to emit code to initialize those elements in fast path. If we should/do not want to emit fast path if parameters are array elements (although less than or equal to 8), then my solution should work and I can submit the PR for this.
This was never caught because the only other code path that creates this fast path is `NewScObjArrayNoArg`.
I was also curious why we decided to emit fast path only for array length up to 8 and not beyond. On scanning through benchmarks, I noticed that this would improve Sunspider but other benchmarks like Octane, Jetstream, etc. won't be improved because they create `Array` from ctor of length around 1024 or greater.
| No FastPath for new Array(constant) | https://api.github.com/repos/chakra-core/ChakraCore/issues/1271/comments | 3 | 2016-07-13T02:38:39Z | 2017-06-01T07:32:06Z | https://github.com/chakra-core/ChakraCore/issues/1271 | 165,224,089 | 1,271 |
[
"chakra-core",
"ChakraCore"
] | https://github.com/Microsoft/ChakraCore/pull/1254 adds new JSRT APIs. Should add native tests for the following as per #1224
```
JsInitializeModuleRecord
JsParseModuleSource
JsModuleEvaluation
JsSetModuleHostInfo
JsGetModuleHostInfo
```
@Yongqu @akroshg
| Add native tests for new JSRT APIs | https://api.github.com/repos/chakra-core/ChakraCore/issues/1269/comments | 1 | 2016-07-12T20:59:47Z | 2018-01-08T19:25:32Z | https://github.com/chakra-core/ChakraCore/issues/1269 | 165,179,499 | 1,269 |
[
"chakra-core",
"ChakraCore"
] | f64.promote/f32 with SSE 1 on x86 currently asserts.
| Wasm f64.promote/f32 /w SSE1 | https://api.github.com/repos/chakra-core/ChakraCore/issues/1268/comments | 2 | 2016-07-12T00:54:00Z | 2016-12-20T00:33:25Z | https://github.com/chakra-core/ChakraCore/issues/1268 | 164,971,546 | 1,268 |
[
"chakra-core",
"ChakraCore"
] | See failures in DisableJIT builds here:
https://github.com/Microsoft/ChakraCore/wiki/Build-Status#disablejit-builds
Beginning with this commit:
https://github.com/Microsoft/ChakraCore/commit/19ad691f60fb2de1781ae657182505d9787174eb (@akroshg)
Builds since then have failed. See:
http://dotnet-ci.cloudapp.net/job/Microsoft_ChakraCore/job/master/job/daily_disablejit_x86_test/100/
Flaky failure text:
```
01:01:23 Summary: D:\j\workspace\daily_disable---91bab980\test had 1212 tests; 0 failures
01:01:23 -- jenkins.testone.cmd >>
01:01:23
01:01:23 -- jenkins.testone.cmd >> Tests failed!
```
As you can see there were no failures but the script detected a non-zero exit code, reported "Tests failed!" and exited with a non-zero exit code, causing Jenkins builds to fail.
Repro with these commands:
```
jenkins\buildone.cmd x86 debug "/p:BuildJIT=false"
jenkins\testone.cmd x86 debug -disablejit
```
When testing a fix please request:
```
@dotnet-bot test nojit tests please
```
| [DisableJIT] Test flakiness since Native Tests change | https://api.github.com/repos/chakra-core/ChakraCore/issues/1266/comments | 0 | 2016-07-11T23:00:37Z | 2016-07-12T00:29:16Z | https://github.com/chakra-core/ChakraCore/issues/1266 | 164,958,730 | 1,266 |
[
"chakra-core",
"ChakraCore"
] | With a project I was working on I noticed that Edge is more than 15 times slower than my Firefox.
Project: http://javascript.nilspace.nl/Cuberite/GrownVisualizer/index.html
The pixel rendering (gets activated when the size of the map gets larger than the canvas) is taking ~130ms while it's done in ~8ms on my Firefox, and ~20ms on Chrome. On my mobile Edge is taking over one second.
| Chakra is slow on a project I'm working on. | https://api.github.com/repos/chakra-core/ChakraCore/issues/1262/comments | 6 | 2016-07-11T07:30:24Z | 2018-06-06T23:36:46Z | https://github.com/chakra-core/ChakraCore/issues/1262 | 164,775,904 | 1,262 |
[
"chakra-core",
"ChakraCore"
] | Tagging @kunalspathak and @akroshg as they are listed in the changes since the last good build: http://dotnet-ci.cloudapp.net/job/Microsoft_ChakraCore/job/master/job/daily_slow_x64_test/99/changes
See e.g. https://github.com/Microsoft/ChakraCore/wiki/Build-Status#builds-with-slow-tests
Output below:
```
20:18:21 2>ERROR: Test failed to run correctly: pass not found in output file ():
20:18:21 2> ch.exe -WERExceptionSupport -ExtendedErrorStackForTestHost -BaselineMode -maxInterpretCount:1 -maxSimpleJitRunCount:1 -bgjit- -dynamicprofilecache:profile.dpl.UnnamedTest1669 -ES6 -ES6TypedArrayExtensions -es6tostringtag -ES6ObjectLiterals -es7Builtins -ES6Species -args summary -endargs D:\j\workspace\daily_slow_x6---8997bd32\test\es6\ES6TypedArrayExtensions.js >testout1536 2>&1
20:18:21 2>ERROR: name of output file: D:\j\workspace\daily_slow_x6---8997bd32\test\es6\testout1536; size: 107; creation: Thu Jul 7 20:17:19 2016, last access: Thu Jul 7 20:17:19 2016, now: Thu Jul 7 20:17:19 2016
20:18:21 2>ERROR: bad output file follows ============
20:18:21 2>Invalid Flag : -es7Builtins
20:18:21 2>
20:18:21 2>Usage: ch.exe <source file>
20:18:21 2>[flaglist] is not supported for Release mode
20:18:21 2>ERROR: end of bad output file ============
```
You can repro locally with:
```
jenkins\buildone.cmd x64 debug
jenkins\testone.cmd x64 debug -includeSlow -dirs es6
```
(or `x86 debug` or `x64 test`)
---
Also seeing failures for Legacy Builds: https://github.com/Microsoft/ChakraCore/wiki/Build-Status#legacy-builds
```
10:50:32 CodexAssert.cpp(12): error C2220: warning treated as error - no 'object' file generated [d:\j\workspace\daily_dev12_x---06a2e6be\bin\NativeTests\NativeTests.vcxproj]
10:50:32 rlrun.cpp
10:50:32 CodexAssert.cpp(12): warning C4127: conditional expression is constant [d:\j\workspace\daily_dev12_x---06a2e6be\bin\NativeTests\NativeTests.vcxproj]
10:50:32 NativeTests.cpp
10:50:32 rlregr.cpp
10:50:32 FileLoadHelpers.cpp(46): error C2220: warning treated as error - no 'object' file generated [d:\j\workspace\daily_dev12_x---06a2e6be\bin\NativeTests\NativeTests.vcxproj]
10:50:32 FileLoadHelpers.cpp(46): warning C4127: conditional expression is constant [d:\j\workspace\daily_dev12_x---06a2e6be\bin\NativeTests\NativeTests.vcxproj]
10:50:32 FileLoadHelpers.cpp(64): warning C4127: conditional expression is constant [d:\j\workspace\daily_dev12_x---06a2e6be\bin\NativeTests\NativeTests.vcxproj]
10:50:32 FileLoadHelpers.cpp(90): warning C4127: conditional expression is constant [d:\j\workspace\daily_dev12_x---06a2e6be\bin\NativeTests\NativeTests.vcxproj]
10:50:32 FileLoadHelpers.cpp(113): warning C4127: conditional expression is constant [d:\j\workspace\daily_dev12_x---06a2e6be\bin\NativeTests\NativeTests.vcxproj]
10:50:32 FileLoadHelpers.cpp(122): warning C4127: conditional expression is constant [d:\j\workspace\daily_dev12_x---06a2e6be\bin\NativeTests\NativeTests.vcxproj]
10:50:32 FileLoadHelpers.cpp(123): warning C4127: conditional expression is constant [d:\j\workspace\daily_dev12_x---06a2e6be\bin\NativeTests\NativeTests.vcxproj]
10:50:32 ThreadServiceTest.cpp
10:50:32 GCStress.cpp
10:50:33 xmlreader.cpp
10:50:33 RecyclerTestObject.cpp
10:50:33 JsRTApiTest.cpp(1426): error C2220: warning treated as error - no 'object' file generated [d:\j\workspace\daily_dev12_x---06a2e6be\bin\NativeTests\NativeTests.vcxproj]
10:50:33 JsRTApiTest.cpp(1426): warning C4127: conditional expression is constant [d:\j\workspace\daily_dev12_x---06a2e6be\bin\NativeTests\NativeTests.vcxproj]
10:50:33 Generating Code...
10:50:33 rl.vcxproj -> d:\j\workspace\daily_dev12_x---06a2e6be\Build\VcBuild\bin\x86_debug\rl.exe
```
Assuming you have Dev12 installed, repro locally with:
```
jenkins\buildone.cmd x64 debug msbuild12
```
Or any other flavor.
---
When submitting fixes please request tests for slow and dev12 tests respectively:
`@dotnet-bot test slow tests please`
`@dotnet-bot test dev12 tests please`
---
/cc @digitalinfinity - It might be a good idea to wait to merge from master into linux until these issues are resolved.
| [Tests] [Slow] [Legacy] Slow and Legacy Test Failures in master branch | https://api.github.com/repos/chakra-core/ChakraCore/issues/1260/comments | 1 | 2016-07-09T00:13:26Z | 2016-07-11T19:03:34Z | https://github.com/chakra-core/ChakraCore/issues/1260 | 164,638,714 | 1,260 |
[
"chakra-core",
"ChakraCore"
] | Upon reviewing #1202 (see https://github.com/Microsoft/ChakraCore/pull/1202#discussion_r69828994) @Yongqu @tcare and I decided we should strongly type the return types of APIs used exclusively internal to the runtime rather than using `Var`.
Motivation comes from `JavascriptOperators::GetOwnPropertyKeys`. Code review raised concern about assuming the return value is always a `JavascriptArray*` and that callers should handle the case this isn't true. It is always true however and so far most of the callers have made this assumption and merely Asserted it, rather than handling the alternative in release builds.
Clearly there is unnecessary cognitive overhead when working with `Var` return types on APIs we know to have guarantees on the type of their result values. We should use specific return types for these APIs and only generalize to Var when necessary e.g. when the API is used as a JIT helper.
This issue recommends changing the following APIs and their dependencies to use `JavascriptArray*` for their return type rather than `Var`:
- `JavascriptOperators::GetOwnPropertyNames`
- `JavascriptOperators::GetOwnPropertySymbols`
- `JavascriptOperators::GetOwnPropertyKeys`
Also #1251 `JavascriptConversion::ToPropertyKey` should return `void` instead of `BOOL`.
| Use specific strong return types for JavascriptOperators::GetOwnPropertyKeys() et al | https://api.github.com/repos/chakra-core/ChakraCore/issues/1252/comments | 3 | 2016-07-06T23:36:20Z | 2016-07-20T03:06:02Z | https://github.com/chakra-core/ChakraCore/issues/1252 | 164,199,007 | 1,252 |
[
"chakra-core",
"ChakraCore"
] | See https://github.com/Microsoft/ChakraCore/pull/1202#discussion_r69828994
`ToPropertyKey` always returns `TRUE` making the `BOOL` return type confusing for callers about whether they need to check the return value in release builds. If it always returns true, the return type should be void to remove that cognitive overhead when using this API.
| JavascriptConversion::ToPropertyKey should return void instead of BOOL | https://api.github.com/repos/chakra-core/ChakraCore/issues/1251/comments | 0 | 2016-07-06T23:34:27Z | 2016-07-18T07:47:08Z | https://github.com/chakra-core/ChakraCore/issues/1251 | 164,198,731 | 1,251 |
[
"chakra-core",
"ChakraCore"
] | Currently ignoring these tests by using the tag `require_backend` (see #1242).
| Investigate whether TTBasic and TTExecuteBasic tests could be run in DisableJIT mode. | https://api.github.com/repos/chakra-core/ChakraCore/issues/1244/comments | 0 | 2016-07-06T00:21:00Z | 2016-08-12T22:39:41Z | https://github.com/chakra-core/ChakraCore/issues/1244 | 163,974,250 | 1,244 |
[
"chakra-core",
"ChakraCore"
] | `AsmJsLink.obj : warning LNK4221: This object file does not define any previously undefined public symbols, so it will not be used by any link operation that consumes this library [D:\j\workspace\daily_disable---1fca0551\lib\Runtime\Language\Chakra.Runtime.Language.vcxproj]`
See e.g. http://dotnet-ci.cloudapp.net/job/Microsoft_ChakraCore/job/master/job/daily_disablejit_x86_debug/98/consoleText (find on page: LNK)
See also: https://github.com/Microsoft/ChakraCore/wiki/Build-Status#disablejit-builds, https://github.com/Microsoft/ChakraCore/wiki/Build-Status-%28linux%29#disablejit-builds
FYI @digitalinfinity @MikeHolman @Cellule
cc @aneeshdk
| DisableJIT linker warning (unused obj file) | https://api.github.com/repos/chakra-core/ChakraCore/issues/1240/comments | 0 | 2016-07-05T21:42:35Z | 2016-07-06T03:08:09Z | https://github.com/chakra-core/ChakraCore/issues/1240 | 163,953,576 | 1,240 |
[
"chakra-core",
"ChakraCore"
] | Hi!
I have an application in which I have to run JavaScript code and I intend to use ChakraCore to do so. My problem is that this application is a cross-platform Xamarin app, thus I should build ChakraCore into a Shared Object (.so) and into a static library (.a)
Is it possibly to build ChakraCore like this? If it is, could somebody help me in this, just a little direction would be much appreciated!
| Is it possible to build ChakraCore into .so and .a to use it on Android and iOS? | https://api.github.com/repos/chakra-core/ChakraCore/issues/1237/comments | 7 | 2016-07-05T07:32:04Z | 2016-08-29T22:00:42Z | https://github.com/chakra-core/ChakraCore/issues/1237 | 163,788,042 | 1,237 |
[
"chakra-core",
"ChakraCore"
] | I was looking into the [code](https://github.com/Microsoft/ChakraCore/blob/master/lib/Runtime/Language/JavascriptOperators.cpp#L1405) of OP_LdCustomSpreadIteratorList and noticed that we optimize it by not creating SpreadArgument if parameter passed to OP_LdCustomSpreadIteratorList does not have user defined iterator. We do this check only for JavascriptArray and TypedArrayBase and we can expand it for ArgumentsObject as well. As per spec, both [unmapped](https://tc39.github.io/ecma262/#sec-createunmappedargumentsobject) and [mapped](https://tc39.github.io/ecma262/#sec-createmappedargumentsobject) argument objects has default `@@iterator` set to `Array.proto.values`.
| Optimize OP_LdCustomSpreadIteratorList for arguments object | https://api.github.com/repos/chakra-core/ChakraCore/issues/1230/comments | 10 | 2016-07-01T16:20:00Z | 2018-08-13T19:03:32Z | https://github.com/chakra-core/ChakraCore/issues/1230 | 163,429,127 | 1,230 |
[
"chakra-core",
"ChakraCore"
] | As per [spec](https://tc39.github.io/ecma262/#sec-runtime-semantics-classdefinitionevaluation), if derived class doesn't have constructor, it should create a constructor something like this -
``` js
constructor(... args){ super (...args);}
```
Instead we do something like this -
``` js
constructor(){ super (...arguments);}
```
`...arguments` part loads argumentsObject on heap which is not performance efficient.
| default constructor creation for derived class is not spec compliant | https://api.github.com/repos/chakra-core/ChakraCore/issues/1228/comments | 3 | 2016-07-01T00:25:53Z | 2016-07-08T19:46:40Z | https://github.com/chakra-core/ChakraCore/issues/1228 | 163,295,584 | 1,228 |
[
"chakra-core",
"ChakraCore"
] | eg.
`JsSetCurrentContext`
takes `nullptr`as input but the API annotation has only`__In__,`so this raises the preFAST error. There are other APIs which have the same behavior. We need to look at other APIs and fix that.
| SAL annotation for some JsRT APIs should be changed. | https://api.github.com/repos/chakra-core/ChakraCore/issues/1225/comments | 4 | 2016-06-30T22:00:56Z | 2019-11-21T08:36:18Z | https://github.com/chakra-core/ChakraCore/issues/1225 | 163,276,609 | 1,225 |
[
"chakra-core",
"ChakraCore"
] | Related to _https://github.com/angular/angular/issues/9524_
Seems like there is issue with Intl.DateTimeFormat.
When you try to format date with only minutes and/or seconds it ignore it and format as default.
It seems like there is only hour + minutes + seconds pattern support properly.
Here plnkr to describe issue: _http://plnkr.co/edit/OyjNyyb1fJhB1thtuEnA?p=preview_
Version:
IE 11.0.9600.18350
Edge 25.10586.00
| Intl.DateTimeFormat doesn't format minutes and seconds properly in IE 11 and Edge | https://api.github.com/repos/chakra-core/ChakraCore/issues/1223/comments | 9 | 2016-06-30T15:17:25Z | 2017-12-08T19:58:35Z | https://github.com/chakra-core/ChakraCore/issues/1223 | 163,191,797 | 1,223 |
[
"chakra-core",
"ChakraCore"
] | Character wise string matching performs bad on EDGE as compared to chrome or firefox. Even worst on IE. @LouisLaf already has context for the test case.
eg. string1[i] === string2[j] performs 3 times better on chrome as compared to EDGE. Firefox performs best overall.
| String accesses and comparisons using indexes not optimized in Chakra | https://api.github.com/repos/chakra-core/ChakraCore/issues/1222/comments | 0 | 2016-06-30T12:21:34Z | 2018-06-27T23:27:36Z | https://github.com/chakra-core/ChakraCore/issues/1222 | 163,152,093 | 1,222 |
[
"chakra-core",
"ChakraCore"
] | ``` JS
function f3(a) {
b = () => eval("arguments[0]");
print(b()); // Should print 1
}
f3(1);
```
This throws an error saying arguments is undefined
| Unable to access arguments in eval inside an arrow method | https://api.github.com/repos/chakra-core/ChakraCore/issues/1209/comments | 4 | 2016-06-28T17:07:20Z | 2019-05-31T18:28:41Z | https://github.com/chakra-core/ChakraCore/issues/1209 | 162,737,309 | 1,209 |
[
"chakra-core",
"ChakraCore"
] | - Unicode 8 has 109,830 `ID_Start` symbols; Unicode 9 has 117,007, i.e. 7,177 more (no removals).
- Unicode 8 has 112,352 `ID_Continue` symbols; Unicode 9 has 119,691, i.e. 7,339 more (no removals).
E.g. these should not throw per Unicode 9:
``` js
Function('var \u{1E943}'); // new ID_Start
Function('var _\u{1E959}'); // new ID_Continue
```
I’ve attached a tarball containing `results.js` which contains the full list of new `ID_Start` and `ID_Continue` symbols in Unicode 9, and the Node.js script used to generate it.
[`unicode-9-identifiers.tar.gz`](https://github.com/Microsoft/ChakraCore/files/336677/unicode-9-identifiers.tar.gz)
---
See also:
- JavaScriptCore: https://bugs.webkit.org/show_bug.cgi?id=159203
- SpiderMonkey: https://bugzilla.mozilla.org/show_bug.cgi?id=1282724
- V8: https://bugs.chromium.org/p/v8/issues/detail?id=5155
| [Unicode] Use Unicode 9’s `ID_Start` & `ID_Continue` for identifiers | https://api.github.com/repos/chakra-core/ChakraCore/issues/1208/comments | 13 | 2016-06-28T09:37:28Z | 2017-09-05T18:00:55Z | https://github.com/chakra-core/ChakraCore/issues/1208 | 162,643,161 | 1,208 |
[
"chakra-core",
"ChakraCore"
] | The code `new URL("http://username:password@127.0.0.1:80/jsonrpc")` throws `A security problem occurred.` on Edge (14.14372).
The standard never mentions about disallowing credential in URLs, and the code above runs successfully in Firefox (48) and Chrome (52) and returns `username` and `password` without problem.
I'm porting an extension from Chrome, it relies on this API to parse URL and make requests.
| URL constructor doesn't allow credential | https://api.github.com/repos/chakra-core/ChakraCore/issues/1196/comments | 4 | 2016-06-24T16:27:02Z | 2016-06-27T16:45:57Z | https://github.com/chakra-core/ChakraCore/issues/1196 | 162,183,127 | 1,196 |
[
"chakra-core",
"ChakraCore"
] | Now that https://github.com/tc39/ecma262/pull/525 has landed, `\u017F` (LATIN SMALL LETTER LONG S) and `\u212A` (KELVIN SIGN) are word characters for `/iu` patterns.
Expected vs. actual behavior:
``` js
/\w/iu.test('\u017F') // true
/\w/iu.test('\u212A') // true
/\W/iu.test('\u017F') // false, actual: true
/\W/iu.test('\u212A') // false, actual: true
/\W/iu.test('s') // false
/\W/iu.test('S') // false
/\W/iu.test('K') // false
/\W/iu.test('k') // false
```
#### See also
- V8: https://bugs.chromium.org/p/v8/issues/detail?id=5145
- JavaScriptCore: https://bugs.webkit.org/show_bug.cgi?id=159055
- SpiderMonkey: https://bugzilla.mozilla.org/show_bug.cgi?id=1281739
- regexpu-core: https://github.com/mathiasbynens/regexpu-core/issues/8
| [RegExp] Match updated spec for `/\w/iu` and `/\W/iu` | https://api.github.com/repos/chakra-core/ChakraCore/issues/1181/comments | 4 | 2016-06-23T09:24:12Z | 2017-01-13T22:06:55Z | https://github.com/chakra-core/ChakraCore/issues/1181 | 161,877,269 | 1,181 |
[
"chakra-core",
"ChakraCore"
] | It would be great, if Chakra will provide nightly builds (with ch), same as do Firefox SpiderMonkey (https://ftp.mozilla.org/pub/firefox/nightly/latest-mozilla-central/jsshell-win64.zip and others).
It will provide simple way to test big libraries with nightly builds and react early on potential Chakra, library or compatibility issues.
| Are public nightly builds of ChakraCore available somewhere? | https://api.github.com/repos/chakra-core/ChakraCore/issues/1180/comments | 6 | 2016-06-23T08:06:04Z | 2016-06-28T00:37:17Z | https://github.com/chakra-core/ChakraCore/issues/1180 | 161,863,218 | 1,180 |
[
"chakra-core",
"ChakraCore"
] | As title. I build ChakraCore with /p:BuildJIT=False. I am wondering how many threads exist simultaneously if I execute ".\Build\VcBuild.NoJIT\bin\x64_release\ch.exe test.js"?. I noticed there are _beginthreadex and WaitForSingleObject calls in wmain. Does it mean that the max number of threads is 2 during execution?
Thanks
| [Question] How many threads exist simultaneously if ChakraCore executes interpreter only (build with /p:BuildJIT=False)? | https://api.github.com/repos/chakra-core/ChakraCore/issues/1179/comments | 3 | 2016-06-23T06:29:31Z | 2016-09-02T12:43:15Z | https://github.com/chakra-core/ChakraCore/issues/1179 | 161,849,372 | 1,179 |
[
"chakra-core",
"ChakraCore"
] | Originally discovered in https://bugs.chromium.org/p/v8/issues/detail?id=5142#c3.
I misread the spec in one case with HTML comments, where a multiline comment has no line break:
/*
*/ --> This is a comment
/\* */ --> This is not a comment
| Single line delimited comment incorrectly parses an HTML comment | https://api.github.com/repos/chakra-core/ChakraCore/issues/1176/comments | 1 | 2016-06-22T22:32:09Z | 2017-05-25T21:35:12Z | https://github.com/chakra-core/ChakraCore/issues/1176 | 161,802,033 | 1,176 |
[
"chakra-core",
"ChakraCore"
] | Opened this issue to track discussion of adding things like this outside of the PR #1168 where `console.log` is added.
@obastemur suggested it might be nice to add other common `console` methods.
Most likely candidate is `console.error` but I don't know that this team has any real use for that since `ch.exe` is mostly used for testing and we look for any output being different.
- [x] Map `console.log` to `WScript.Echo` to unblock common use case (#1168)
- [ ] (If/when needed:) Provide a complete implementation of `console.log` including formatting and full specification of the API.
| Implement selected console.* methods in ch.exe (since it is a console host after all) | https://api.github.com/repos/chakra-core/ChakraCore/issues/1175/comments | 8 | 2016-06-22T20:41:08Z | 2018-02-24T01:54:10Z | https://github.com/chakra-core/ChakraCore/issues/1175 | 161,780,658 | 1,175 |
[
"chakra-core",
"ChakraCore"
] | I am contributing to a project where we have to calculate CRC32 values for files to upload to a server:
https://github.com/blackcity/backload
The CRC32 value is also calculated on the server to ensure file integrity.
When we run the code on the client we have disappointing results for Edge (not to name IE). Chrome is very fast. Try this code:
<pre>
// Lookup table for the polynomial 3988292384 (0xEDB88320)
var crc32Lookup = [
0, 0x77073096, 0xEE0E612C, 0x990951BA,
0x076DC419, 0x706AF48F, 0xE963A535, 0x9E6495A3,
0x0EDB8832, 0x79DCB8A4, 0xE0D5E91E, 0x97D2D988,
0x09B64C2B, 0x7EB17CBD, 0xE7B82D07, 0x90BF1D91,
0x1DB71064, 0x6AB020F2, 0xF3B97148, 0x84BE41DE,
0x1ADAD47D, 0x6DDDE4EB, 0xF4D4B551, 0x83D385C7,
0x136C9856, 0x646BA8C0, 0xFD62F97A, 0x8A65C9EC,
0x14015C4F, 0x63066CD9, 0xFA0F3D63, 0x8D080DF5,
0x3B6E20C8, 0x4C69105E, 0xD56041E4, 0xA2677172,
0x3C03E4D1, 0x4B04D447, 0xD20D85FD, 0xA50AB56B,
0x35B5A8FA, 0x42B2986C, 0xDBBBC9D6, 0xACBCF940,
0x32D86CE3, 0x45DF5C75, 0xDCD60DCF, 0xABD13D59,
0x26D930AC, 0x51DE003A, 0xC8D75180, 0xBFD06116,
0x21B4F4B5, 0x56B3C423, 0xCFBA9599, 0xB8BDA50F,
0x2802B89E, 0x5F058808, 0xC60CD9B2, 0xB10BE924,
0x2F6F7C87, 0x58684C11, 0xC1611DAB, 0xB6662D3D,
0x76DC4190, 0x01DB7106, 0x98D220BC, 0xEFD5102A,
0x71B18589, 0x06B6B51F, 0x9FBFE4A5, 0xE8B8D433,
0x7807C9A2, 0x0F00F934, 0x9609A88E, 0xE10E9818,
0x7F6A0DBB, 0x086D3D2D, 0x91646C97, 0xE6635C01,
0x6B6B51F4, 0x1C6C6162, 0x856530D8, 0xF262004E,
0x6C0695ED, 0x1B01A57B, 0x8208F4C1, 0xF50FC457,
0x65B0D9C6, 0x12B7E950, 0x8BBEB8EA, 0xFCB9887C,
0x62DD1DDF, 0x15DA2D49, 0x8CD37CF3, 0xFBD44C65,
0x4DB26158, 0x3AB551CE, 0xA3BC0074, 0xD4BB30E2,
0x4ADFA541, 0x3DD895D7, 0xA4D1C46D, 0xD3D6F4FB,
0x4369E96A, 0x346ED9FC, 0xAD678846, 0xDA60B8D0,
0x44042D73, 0x33031DE5, 0xAA0A4C5F, 0xDD0D7CC9,
0x5005713C, 0x270241AA, 0xBE0B1010, 0xC90C2086,
0x5768B525, 0x206F85B3, 0xB966D409, 0xCE61E49F,
0x5EDEF90E, 0x29D9C998, 0xB0D09822, 0xC7D7A8B4,
0x59B33D17, 0x2EB40D81, 0xB7BD5C3B, 0xC0BA6CAD,
0xEDB88320, 0x9ABFB3B6, 0x03B6E20C, 0x74B1D29A,
0xEAD54739, 0x9DD277AF, 0x04DB2615, 0x73DC1683,
0xE3630B12, 0x94643B84, 0x0D6D6A3E, 0x7A6A5AA8,
0xE40ECF0B, 0x9309FF9D, 0x0A00AE27, 0x7D079EB1,
0xF00F9344, 0x8708A3D2, 0x1E01F268, 0x6906C2FE,
0xF762575D, 0x806567CB, 0x196C3671, 0x6E6B06E7,
0xFED41B76, 0x89D32BE0, 0x10DA7A5A, 0x67DD4ACC,
0xF9B9DF6F, 0x8EBEEFF9, 0x17B7BE43, 0x60B08ED5,
0xD6D6A3E8, 0xA1D1937E, 0x38D8C2C4, 0x4FDFF252,
0xD1BB67F1, 0xA6BC5767, 0x3FB506DD, 0x48B2364B,
0xD80D2BDA, 0xAF0A1B4C, 0x36034AF6, 0x41047A60,
0xDF60EFC3, 0xA867DF55, 0x316E8EEF, 0x4669BE79,
0xCB61B38C, 0xBC66831A, 0x256FD2A0, 0x5268E236,
0xCC0C7795, 0xBB0B4703, 0x220216B9, 0x5505262F,
0xC5BA3BBE, 0xB2BD0B28, 0x2BB45A92, 0x5CB36A04,
0xC2D7FFA7, 0xB5D0CF31, 0x2CD99E8B, 0x5BDEAE1D,
0x9B64C2B0, 0xEC63F226, 0x756AA39C, 0x026D930A,
0x9C0906A9, 0xEB0E363F, 0x72076785, 0x05005713,
0x95BF4A82, 0xE2B87A14, 0x7BB12BAE, 0x0CB61B38,
0x92D28E9B, 0xE5D5BE0D, 0x7CDCEFB7, 0x0BDBDF21,
0x86D3D2D4, 0xF1D4E242, 0x68DDB3F8, 0x1FDA836E,
0x81BE16CD, 0xF6B9265B, 0x6FB077E1, 0x18B74777,
0x88085AE6, 0xFF0F6A70, 0x66063BCA, 0x11010B5C,
0x8F659EFF, 0xF862AE69, 0x616BFFD3, 0x166CCF45,
0xA00AE278, 0xD70DD2EE, 0x4E048354, 0x3903B3C2,
0xA7672661, 0xD06016F7, 0x4969474D, 0x3E6E77DB,
0xAED16A4A, 0xD9D65ADC, 0x40DF0B66, 0x37D83BF0,
0xA9BCAE53, 0xDEBB9EC5, 0x47B2CF7F, 0x30B5FFE9,
0xBDBDF21C, 0xCABAC28A, 0x53B39330, 0x24B4A3A6,
0xBAD03605, 0xCDD70693, 0x54DE5729, 0x23D967BF,
0xB3667A2E, 0xC4614AB8, 0x5D681B02, 0x2A6F2B94,
0xB40BBE37, 0xC30C8EA1, 0x5A05DF1B, 0x2D02EF8D
];
// Calculate CRC32
function crc32(data) {
var crc = 0xFFFFFFFF
for (var i = 0; i < data.length; i++)
crc = (crc >>> 8) ^ crc32Lookup[data[i] ^ (crc & 0x000000FF)];
crc = crc = ~crc;
return crc < 0 ? (0xFFFFFFFF + crc + 1) : crc;
}
// Main test method
function runTest()
{
var len = 50000000, // 50MB
data = new Uint8Array(len),
data2,
blob,
crc;
// Fill array with random values from 0-255
// IE: 45sec, Edge: 30sec, Chrome 4sec
for (var i = 0; i < len; i++) {
data[i] = Math.floor(Math.random() * 256);
}
// Simulate file data (blob)
blob = new Blob([data], { type: 'application/octet-binary' });
// Using a FileReader to read the blob into an ArrayBuffer
var reader = new FileReader();
reader.onload = function (e) {
var result = e.target.result;
var data = new Uint8Array(result, 0, result.byteLength);
// Calculate CRC (IE: 25sec, Edge: 25sec, Chrome 4sec)
var crc = crc32(data);
};
reader.readAsArrayBuffer(blob);
}
</pre>
| Perf issues with CRC32 Calculation | https://api.github.com/repos/chakra-core/ChakraCore/issues/1171/comments | 7 | 2016-06-22T14:36:34Z | 2016-06-29T17:33:50Z | https://github.com/chakra-core/ChakraCore/issues/1171 | 161,700,301 | 1,171 |
[
"chakra-core",
"ChakraCore"
] | Hi,
is node-mssql already supported on ChakraCore? I'm unable to connect to Azure DB (simple select example). It will just crash an app - no part of it will work anymore - I need to restart / redeploy to run/debug.
| node-mssql unable to connect to azure db | https://api.github.com/repos/chakra-core/ChakraCore/issues/1157/comments | 2 | 2016-06-20T21:48:53Z | 2016-06-21T23:26:09Z | https://github.com/chakra-core/ChakraCore/issues/1157 | 161,304,249 | 1,157 |
[
"chakra-core",
"ChakraCore"
] | I'm working on sq/JSIL#993 and try to find a way for fast invocation of method stored in object by run-time calculated key. I've write some tests to found a way for finding proper JS form, and looks like Edge works much slower than other browsers. Please see my results in table https://github.com/sq/JSIL/pull/993#issuecomment-226718472.
What is really bad, it works very slow (30-50 times) for last too options (Stable/Generated), which are the best for other browsers, and really only way for implementing dynamic method invocation.
Here are my test [cases](http://plnkr.co/edit/hVsmbE?p=preview).
| Function invocation slow when stored by run-time calculated key | https://api.github.com/repos/chakra-core/ChakraCore/issues/1153/comments | 10 | 2016-06-17T22:53:51Z | 2016-11-02T23:23:05Z | https://github.com/chakra-core/ChakraCore/issues/1153 | 160,989,226 | 1,153 |
[
"chakra-core",
"ChakraCore"
] | Chakra returns `false` for each of these tests, while V8 and JavaScriptCore return `true` for all of them:
``` js
/\u01F1/i.test('\u01F3')
/\u0345/i.test('\u03B9')
/\u037F/i.test('\u03F3')
/\u0528/i.test('\u0529')
/\u052A/i.test('\u052B')
/\u052C/i.test('\u052D')
/\u052E/i.test('\u052F')
/\uA698/i.test('\uA699')
/\uA69A/i.test('\uA69B')
/\uA796/i.test('\uA797')
/\uA798/i.test('\uA799')
/\uA79A/i.test('\uA79B')
/\uA79C/i.test('\uA79D')
/\uA79E/i.test('\uA79F')
/\uA7AB/i.test('\u025C')
/\uA7AC/i.test('\u0261')
/\uA7AD/i.test('\u026C')
/\uA7B0/i.test('\u029E')
/\uA7B1/i.test('\u0287')
```
https://mathiasbynens.be/demo/regex-i
(See also: [similar SpiderMonkey bug](https://bugzilla.mozilla.org/show_bug.cgi?id=1280046).)
| RegExp: case-insensitive matching not compatible with V8/JSC | https://api.github.com/repos/chakra-core/ChakraCore/issues/1136/comments | 5 | 2016-06-14T14:09:00Z | 2016-06-21T23:43:57Z | https://github.com/chakra-core/ChakraCore/issues/1136 | 160,190,876 | 1,136 |
[
"chakra-core",
"ChakraCore"
] | After fading in a modal div, the document looses focus causing keydown and keyup events to no longer fire. This behavior is specific to Edge. IE, Chrome, and Firefox do not have this problem.
They key events begin firing again if you click somewhere on the modal div or page, or if you use $(document).focus(); to reset focus on the page.
## Some example code from my project:
$('#modal-AjaxCommunicating').fadeIn();
---
```
// Close modal dialog by pressing enter or escape
$(document).keyup(function (e) {
if (e.keyCode == 13 || e.keyCode == 27) {
if ($('#modal-AjaxCommunicating').is(':visible')) {
$('#modal-AjaxCommunicating').fadeOut();
}
}
});
```
| Javascript keydown events not firing after .fadein() | https://api.github.com/repos/chakra-core/ChakraCore/issues/1129/comments | 8 | 2016-06-12T18:57:24Z | 2016-06-13T22:24:06Z | https://github.com/chakra-core/ChakraCore/issues/1129 | 159,841,992 | 1,129 |
[
"chakra-core",
"ChakraCore"
] | ``` js
var obj = {};
obj.foo = function () {
print(obj.foo.name);
}
obj.foo();
```
Above test case prints function name only if function is defer parsed.
``` cmd
ch.exe –Es6Experimental –forceDeferParse test.js
Prints “foo”
ch.exe –Es6Experimental test.js
Prints nothing.
```
| function.name doesn't return name if function is not defer parsed | https://api.github.com/repos/chakra-core/ChakraCore/issues/1128/comments | 3 | 2016-06-12T06:45:18Z | 2016-06-21T23:28:13Z | https://github.com/chakra-core/ChakraCore/issues/1128 | 159,811,730 | 1,128 |
[
"chakra-core",
"ChakraCore"
] | This is a list of thing we need to implement for Wasm MVP
- [x] Br yielding values #1111
- [x] I32Popcnt
- [x] F32Min/Max #659
- [x] Trunc JIT
- [x] Nearest JIT
- [x] Int64 (@Cellule #1806)
- [x] Unreachable operator
- [x] Unreachable state
- [ ] Missing OpCodes
- [x] current_memory (0x3b)
- [ ] grow_memory (0x39) (@MikeHolman)
- [x] f64.copysign (0x91)
- [ ] i32.trunc_u/f32 (0x9f) (@Krovatkin #1812)
- [ ] i32.trunc_u/f64 (0xa0) (@Krovatkin #1812)
- [ ] f32.convert_u/i32 (0xa9) (@Krovatkin)
- [x] f32.reinterpret/i32 (0xad) (@Krovatkin #1813)
- [x] get_global (0xbb)
- [x] set_global (0xbc)
- [ ] Int64 Conversions
- [ ] i64.trunc_s/f32
- [ ] i64.trunc_s/f64
- [ ] i64.trunc_u/f32
- [ ] i64.trunc_u/f64
- [ ] i64.extend_s/i32 (@meg-gupta)
- [ ] i64.extend_u/i32 (@meg-gupta)
- [ ] f32.convert_s/i64
- [ ] f32.convert_u/i64
- [ ] f64.convert_s/i64
- [ ] f64.convert_u/i64
- [ ] f64.reinterpret/i64 (@arunetm)
- [ ] i64.reinterpret/f64 (@arunetm)
- [x] [Globals](https://github.com/WebAssembly/design/blob/binary_0xc/BinaryEncoding.md#global-section)
- [ ] [Init Expressions](https://github.com/WebAssembly/design/blob/binary_0xc/BinaryEncoding.md#init_expr)
- [ ] For element segments' [offset](https://github.com/WebAssembly/design/blob/binary_0xc/BinaryEncoding.md#element-section) (@arunetm)
- [x] For data segments' [offset](https://github.com/WebAssembly/design/blob/binary_0xc/BinaryEncoding.md#data-section)
- [x] For globals' [initial value](https://github.com/WebAssembly/design/blob/binary_0xc/BinaryEncoding.md#global-entry)
- [ ] Correctly call imported functions from the function table (intermediary function used for the conversion) (@Cellule #1807)
- [ ] [Import types](https://github.com/WebAssembly/design/blob/binary_0xc/BinaryEncoding.md#import-entry)
- [ ] Kind: Table
- [ ] Kind: Memory (@MikeHolman)
- [x] Kind: Global
- [ ] [Export types](https://github.com/WebAssembly/design/blob/binary_0xc/BinaryEncoding.md#export-entry)
- [ ] Kind: Table
- [x] Kind: Global
- [ ] Runtime Traps
- [ ] Array access out of bound (@meg-gupta)
- [ ] Verify [math trapping](https://github.com/WebAssembly/design/blob/binary_0xc/AstSemantics.md#32-bit-integer-operators)
- [ ] Trap on invalid indirect call (out of bounds and invalid signature)
- [ ] Spec compliant
- [ ] Ignore the names section until we have use of it (@cellule #1855)
- [ ] Copy the binary buffer to prevent modification after call to `wasm.instantiateModule()` or `new WebAssembly.Module()` (@Cellule #1808)
- [x] 0xD format (@Cellule #1822)
- [x] [OpCode Reordering](https://github.com/WebAssembly/design/blob/binary_0xd/BinaryEncoding.md#function-bodies)
- [x] [Negative type constructors](https://github.com/WebAssembly/design/blob/binary_0xd/BinaryEncoding.md#language-types)
- [x] [Block signature has byte -0x40 for no return value](https://github.com/WebAssembly/design/blob/binary_0xd/BinaryEncoding.md#block_type)
- [x] Enable deferred parsing by default
- [ ] Investigate Debugging/Profiling scenario on Edge
- [ ] Implement final Javascript API (@MikeHolman)
- [ ] Remove `Wasm` Object from the global scope #1834
- [x] Retire WebAssembly branch completely (all dev work done in master)
- [ ] Security analysis (@Penguinwizzard)
- [ ] Known Bug Fixes
- [ ] Assert on `print(Wasm.instantiateModule(buffer, ffi).exports.a);`
- [ ] Call import with int64 and float32 in arguments
- [ ] Performance
- [ ] Use 4G array to detect out of bounds access and trap
- [ ] Fold const math operations (#1852) | Wasm todo | https://api.github.com/repos/chakra-core/ChakraCore/issues/1127/comments | 13 | 2016-06-11T01:31:30Z | 2016-12-20T00:33:19Z | https://github.com/chakra-core/ChakraCore/issues/1127 | 159,749,190 | 1,127 |
[
"chakra-core",
"ChakraCore"
] | Create a new type Unreachable for EmitInfo.
Unreachable is valid for typecheck.
If 1 statement is unreachable in a block all the following statement has to be unreachable (not really important at first, shouldn't happen in real code)
Also, this will allow to correctly invalidate the following
``` js
function () : i32 {
if (cond) {
return 0;
} else {
call something // Right now, the IfElse block will yield void and we will assume unreachable when checking for the function return value
}
}
```
| Wasm unreachable EmitInfo | https://api.github.com/repos/chakra-core/ChakraCore/issues/1112/comments | 1 | 2016-06-09T20:54:22Z | 2016-12-20T00:33:19Z | https://github.com/chakra-core/ChakraCore/issues/1112 | 159,505,703 | 1,112 |
[
"chakra-core",
"ChakraCore"
] | Depends on #1110
Capture the next tmp locations when entering a block in a stack
When doing a BR of that block, move the yielding value to the location and give that block the corresponding type.
This way a block will have a type and you can type check with every br yields and also the last block statement
| Wasm br yield | https://api.github.com/repos/chakra-core/ChakraCore/issues/1111/comments | 0 | 2016-06-09T20:50:37Z | 2016-12-20T00:33:19Z | https://github.com/chakra-core/ChakraCore/issues/1111 | 159,504,879 | 1,111 |
[
"chakra-core",
"ChakraCore"
] | Capture what is the next tmp register for all types when entering the block (do not acquire)
At the end of the block, if we have to yield something, move it to the corresponding location, release everything then yield that location.
| Wasm block yielding | https://api.github.com/repos/chakra-core/ChakraCore/issues/1110/comments | 0 | 2016-06-09T20:48:27Z | 2016-12-20T00:33:19Z | https://github.com/chakra-core/ChakraCore/issues/1110 | 159,504,423 | 1,110 |
[
"chakra-core",
"ChakraCore"
] | Instead of having a stack of stack for the eval stack, change it to a single stack with delimiter between nested blocks
| Wasm single eval stack | https://api.github.com/repos/chakra-core/ChakraCore/issues/1109/comments | 0 | 2016-06-09T20:46:27Z | 2016-12-20T00:33:19Z | https://github.com/chakra-core/ChakraCore/issues/1109 | 159,504,012 | 1,109 |
[
"chakra-core",
"ChakraCore"
] | In Chrome, using V8, using a **let** rather than **var** to declare a variable avoids a JavaScript closure behavior that often surprises developers. In browsers based on Chakra, this distinction between **let** and **var** does not seem to apply.
The following fiddle demonstrates: http://jsfiddle.net/rmXcF/4/. In Chrome, clicking on the red "3" div alerts with the number 3, but in IE and edge, the displayed number is 6.
I'm really not sure which behavior is correct per the specification, but Chrome's behavior is certainly less surprising.
| let vs var and closures | https://api.github.com/repos/chakra-core/ChakraCore/issues/1106/comments | 9 | 2016-06-09T15:57:07Z | 2016-06-21T19:10:02Z | https://github.com/chakra-core/ChakraCore/issues/1106 | 159,444,545 | 1,106 |
[
"chakra-core",
"ChakraCore"
] | I integrated master last night to the WebAssembly branch.
I had to comment out the copysign implementation (WasmMath.inl) because it doesn't compile in ChakraFull.
Need to investigate how to make it build in both solution.
| WASM copysign fix build | https://api.github.com/repos/chakra-core/ChakraCore/issues/1103/comments | 2 | 2016-06-09T06:36:39Z | 2016-12-20T00:33:18Z | https://github.com/chakra-core/ChakraCore/issues/1103 | 159,337,648 | 1,103 |
[
"chakra-core",
"ChakraCore"
] | At https://github.com/Microsoft/ChakraCore/blob/master/bin/ch/WScriptJsrt.cpp#L271
We preallocate `fullPathNarrow` to a buffer with `_MAX_PATH`, however the method `NarrowStringToWideDynamic` does a malloc and replaces that buffer.
We need to free the buffer returned by the helper call.
Also, we should check if the helper failed ie: `IfFailGo(Helpers::NarrowStringToWideDynamic(fullPathNarrow, &fullPath))`
| WScript LoadScript memory leak | https://api.github.com/repos/chakra-core/ChakraCore/issues/1102/comments | 2 | 2016-06-09T05:35:20Z | 2016-06-09T18:05:37Z | https://github.com/chakra-core/ChakraCore/issues/1102 | 159,331,599 | 1,102 |
[
"chakra-core",
"ChakraCore"
] | Hi when I have an web app which calls js scripts on different domains when i run the app in doesn't load any of those scripts when I use Microsoft Edge debug tools, it shows "SCRIPT70: Permission denied" for all unloaded scripts. FYI this only happens on Microsoft Edge 25.10586.0.0
Thanks.
| SCRIPT70: Permission denied | https://api.github.com/repos/chakra-core/ChakraCore/issues/1100/comments | 11 | 2016-06-08T06:53:31Z | 2016-09-09T01:27:34Z | https://github.com/chakra-core/ChakraCore/issues/1100 | 159,091,867 | 1,100 |
[
"chakra-core",
"ChakraCore"
] | Source buffer may have a broken/unsupported unicode entry that we could normalize / convert into UTF-8. At the moment, this is considered as a consumer problem rather than a feature that ChakraCore provides cross platform.
| Handling UTF-8 conversion/normalization internally | https://api.github.com/repos/chakra-core/ChakraCore/issues/1098/comments | 4 | 2016-06-07T17:44:29Z | 2016-09-07T10:29:28Z | https://github.com/chakra-core/ChakraCore/issues/1098 | 158,982,944 | 1,098 |
[
"chakra-core",
"ChakraCore"
] | Checking arguments.length slow with Chakra. Here is sample page:
```
<html>
<body>
<script>
var start = Date.now();
function one(a){
}
function two(a,b){
}
function route(a,b){
switch (arguments.length) {
case 1:
return one(a);
case 2:
return two(a,b);
}
}
for (var i = 0; i<1000000; i++) {
route(1,2);
}
document.write(Date.now() - start);
</script>
</body>
</html>
```
My results:
- Edge: ~380ms
- Chrome: ~20ms
- Firefox: ~3ms
| Checking arguments.length slow | https://api.github.com/repos/chakra-core/ChakraCore/issues/1083/comments | 14 | 2016-06-03T07:47:50Z | 2016-09-08T21:18:55Z | https://github.com/chakra-core/ChakraCore/issues/1083 | 158,310,132 | 1,083 |
[
"chakra-core",
"ChakraCore"
] | This is the story of IE 11 on Windows 10(build 14332), IE 11 does not show any prompt when I use window.clipboardData.setData even though the settings in
“Allow Programmatic clipboard access” is "prompt".
The test page is:
http://www.broadband-xp.com/test/ie11/clipboard.html
If I change the settings in “Allow Programmatic clipboard access” from the default value of "prompt" to "enabled", it works. But if I change the settings back to "prompt", IE does not show a prompt but clipboard copying fails silently.
---
Repro code:
``` html
<html>
<head>
<meta http-equiv="Content-Type" content="text/html; charset=Shift_JIS">
<title>test</title>
<script>
function a(){
var b=window.clipboardData.setData("text","This is a test. Sucess!");
if(b){
alert("success");
}else{
alert("fail");
}
}
</script>
</head>
<body bgcolor=#FFFFFF>
<input type=button onClick="a();" value="Copy to Clipboard">
</body>
</html>
```
| In Windows 10 (build 14332) window.clipboardData.setData does not work as expected. | https://api.github.com/repos/chakra-core/ChakraCore/issues/1071/comments | 11 | 2016-06-01T09:36:57Z | 2016-06-13T18:51:49Z | https://github.com/chakra-core/ChakraCore/issues/1071 | 157,863,477 | 1,071 |
[
"chakra-core",
"ChakraCore"
] | Thanks for Ian's help on getting me unblocked, but for outsiders a how to run tests page would be very helpful.
| Wiki needs a help page on how to run tests. | https://api.github.com/repos/chakra-core/ChakraCore/issues/1068/comments | 3 | 2016-05-31T22:19:29Z | 2016-11-15T23:35:34Z | https://github.com/chakra-core/ChakraCore/issues/1068 | 157,783,631 | 1,068 |
[
"chakra-core",
"ChakraCore"
] | Currently, [JsRuntimeAttributeEnableExperimentalFeatures](https://github.com/Microsoft/ChakraCore/blob/release/1.2/lib/Jsrt/ChakraCommon.h#L311) flag is the only way to turn ON all the ES6 experimental features. Instead have a JSRT API / enum to expose more flags that user can use to control which experimental features should be turned ON/OFF (similar to `--harmony-*` that v8 has).
| Expose JSRT Apis to control ES6 experimental features | https://api.github.com/repos/chakra-core/ChakraCore/issues/1064/comments | 8 | 2016-05-31T07:59:13Z | 2016-06-22T21:53:17Z | https://github.com/chakra-core/ChakraCore/issues/1064 | 157,611,362 | 1,064 |
[
"chakra-core",
"ChakraCore"
] | In https://github.com/Shiney/ChakraCore/tree/AsmJsChangeHeapAssertionFailure
I have added a new version of the asmjs test relink.js with a small change to one of the constants to make it not match the expected change heap function.
https://github.com/Shiney/ChakraCore/commit/3d7ee3e40a5645b174cd902448e7cf62f258dfa3
https://github.com/Shiney/ChakraCore/blob/AsmJsChangeHeapAssertionFailure/test/AsmJs/relinkFail.js
https://github.com/Shiney/ChakraCore/blob/AsmJsChangeHeapAssertionFailure/test/AsmJs/relinkFail.baseline
https://github.com/Shiney/ChakraCore/blob/AsmJsChangeHeapAssertionFailure/test/AsmJs/rlexe.xml
Running the test with e.g.
runtests -x64debug -dirs asmjs
Causes an assertion failure as shown below.
2>ASSERTION 40948: (D:\dev\ChakraCore\ChakraCore\lib\Runtime\Base\FunctionBody.c
pp, line 1485) this->GetFunctionProxy() == this
2> Failure: (this->GetFunctionProxy() == this)
Also getting the same issue when trying to implement #65
| Assertion failure in tests when changeHeap is not correctly implemented | https://api.github.com/repos/chakra-core/ChakraCore/issues/1062/comments | 1 | 2016-05-28T13:21:07Z | 2017-05-24T04:17:20Z | https://github.com/chakra-core/ChakraCore/issues/1062 | 157,338,572 | 1,062 |
[
"chakra-core",
"ChakraCore"
] | The max address space for a process is hardcoded on Linux. We should find a better mechanism of obtaining this information so that the GC can function correctly.
| Fix hardcoding of max process address space on Linux | https://api.github.com/repos/chakra-core/ChakraCore/issues/1059/comments | 3 | 2016-05-27T20:44:57Z | 2016-09-07T09:59:14Z | https://github.com/chakra-core/ChakraCore/issues/1059 | 157,287,512 | 1,059 |
[
"chakra-core",
"ChakraCore"
] | IsInAssert is supposed to be a thread static variable but on Linux it's marked as a global static variable. It's use needs to be examined and fixed up accordingly.
| Fix use of IsInAssert on Linux | https://api.github.com/repos/chakra-core/ChakraCore/issues/1058/comments | 0 | 2016-05-27T20:43:52Z | 2016-09-07T07:47:24Z | https://github.com/chakra-core/ChakraCore/issues/1058 | 157,287,349 | 1,058 |
[
"chakra-core",
"ChakraCore"
] | Only tested in a stable Edge release. Please confirm.
According to [ES6 6.1.7.3](http://www.ecma-international.org/ecma-262/6.0/#sec-invariants-of-the-essential-internal-methods) Invariants of the Essential Internal Methods:
<blockquote>
If a property P is described as a data property with Desc.[[Value]] equal to v and Desc.[[Writable]] and Desc.[[Configurable]] are both false, then the SameValue must be returned for the Desc.[[Value]] attribute of the property on all future calls to [[GetOwnProperty]] ( P ).
</blockquote>
Testcase against `Function#arguments`
``` js
function f() { return Object.getOwnPropertyDescriptor(f, 'arguments'); }
Object.getOwnPropertyDescriptor(f, 'arguments');
// value: null, writable: false, configurable: false
f();
// value: Arguments[], writable: false, configurable: false
```
Testcase against `Function#caller`
``` js
function g() { return Object.getOwnPropertyDescriptor(g, 'caller'); }
function h() { return g(); }
Object.getOwnPropertyDescriptor(g, 'caller');
// value: null, writable: false, configurable: false
h();
// value: function h(), writable: false, configurable: false
```
Suggested fix: Use getters.
| Own "arguments" and "caller" properties of non-strict functions don't observe the invariability of nonconfigurable, nonwritable properties | https://api.github.com/repos/chakra-core/ChakraCore/issues/1048/comments | 5 | 2016-05-26T12:36:26Z | 2018-07-23T22:00:17Z | https://github.com/chakra-core/ChakraCore/issues/1048 | 156,973,654 | 1,048 |
[
"chakra-core",
"ChakraCore"
] | In the official test suite address.wast the unaligned memory load give bad results.
We didn't want to regress asm.js which doesn't allow unaligned load so we will have to create a different path for wasm that needs to check the alignment.
| Wasm interpreter for unaligned memory load | https://api.github.com/repos/chakra-core/ChakraCore/issues/1042/comments | 5 | 2016-05-25T20:53:22Z | 2016-12-20T00:33:18Z | https://github.com/chakra-core/ChakraCore/issues/1042 | 156,847,776 | 1,042 |
[
"chakra-core",
"ChakraCore"
] | Hi!
When I try recreate Chakra Runtime, create Chakra Context and acquire Context Scope several times I got the following exception when trying to call a function. Here is additional message:
> "Additional information: Attempted to read or write protected memory. This is often an indication that other memory is corrupt."
I am using the latest Chakra version with release build (x64).
| AccessViolationException | https://api.github.com/repos/chakra-core/ChakraCore/issues/1033/comments | 13 | 2016-05-24T06:23:54Z | 2018-12-19T00:36:57Z | https://github.com/chakra-core/ChakraCore/issues/1033 | 156,428,500 | 1,033 |
[
"chakra-core",
"ChakraCore"
] | It looks like [bn.js](https://github.com/indutny/bn.js) is running 6x slower on a Chakra than it does on V8, or IonMonkey (or whatever it is called now). My suspicion is that it may be related to the integer/double representations and conversions between them.
See: https://github.com/indutny/bn.js/issues/146
| bn.js performance issues | https://api.github.com/repos/chakra-core/ChakraCore/issues/1020/comments | 11 | 2016-05-21T20:29:26Z | 2016-05-27T00:59:43Z | https://github.com/chakra-core/ChakraCore/issues/1020 | 156,120,127 | 1,020 |
[
"chakra-core",
"ChakraCore"
] | var o = {get m() {}};
var f = Object.getOwnPropertyDescriptor(o, 'm').get;
console.log(f.hasOwnProperty('prototype')); // expected: false, actual: true
| Method defined in object literal should not have 'prototype' property | https://api.github.com/repos/chakra-core/ChakraCore/issues/1014/comments | 1 | 2016-05-20T17:13:31Z | 2016-05-23T21:45:31Z | https://github.com/chakra-core/ChakraCore/issues/1014 | 156,010,036 | 1,014 |
[
"chakra-core",
"ChakraCore"
] | The resulting string from Intl.DateTimeFormat doesn't equal the string literal that I expect. This has some resulting different behavior from other browsers if I try to do any validation against the string. I know it's a new specification and I can polyfill it to get the behavior I want, but thought I should report it in case it's not on anyone's radar.
| Intl.DateTimeFormat string comparison | https://api.github.com/repos/chakra-core/ChakraCore/issues/1013/comments | 2 | 2016-05-20T16:10:07Z | 2016-05-23T17:11:44Z | https://github.com/chakra-core/ChakraCore/issues/1013 | 155,998,223 | 1,013 |
[
"chakra-core",
"ChakraCore"
] | the following code is throwing object doesn't support this action on the this.Key.
it runs fine for about 30 seconds and then starts erring. this error only happens in Edge.
class Value {
constructor(key, data) {
this.Key = key;
this.Data = data;
}
}
| ES6 constructor erring with object doesn't support this action | https://api.github.com/repos/chakra-core/ChakraCore/issues/1011/comments | 7 | 2016-05-19T17:10:38Z | 2016-09-07T21:33:09Z | https://github.com/chakra-core/ChakraCore/issues/1011 | 155,788,416 | 1,011 |
[
"chakra-core",
"ChakraCore"
] | I am running the Node interop with ES modules in https://github.com/nodejs/node-eps/blob/master/002-es6-modules.md . I'm trying to find a good way to contact Yong Qu (or maybe someone else?) about some gotchas, but could not find social media or email location.
In particular getters and non-object values:
``` js
// commonjs
module.exports = {
get y() {/*...*/}
}
```
``` js
import x,{y} from 'commonjs';
console.log(x)
console.log(y) // does perform getter
```
and
``` js
// commonjs
module.exports = ()=>{}
```
``` js
import fn,{call} from 'commonjs';
fn();
call(); // throws, not a function
```
Have had a some calls w/ TC39/Node/WHATWG, now contacting engines.
| Contact point for Node ES module interop | https://api.github.com/repos/chakra-core/ChakraCore/issues/1002/comments | 1 | 2016-05-18T14:28:04Z | 2016-05-18T15:00:01Z | https://github.com/chakra-core/ChakraCore/issues/1002 | 155,515,698 | 1,002 |
[
"chakra-core",
"ChakraCore"
] | I tried to embed Chakra into UWP by following the steps from https://github.com/Microsoft/ChakraCore/wiki/Embedding-ChakraCore guide. Everything works fine the appx fails WAC certification for Release_x64 configuration (supported apis). Isn't chakra supposed to run in UWP store application? Below you can find all not supported api reported by WAC tool:
API _CxxThrowException in vcruntime140.dll is not supported for this application type. ChakraCore.dll calls this API.
API __C_specific_handler in vcruntime140.dll is not supported for this application type. ChakraCore.dll calls this API.
API __CxxFrameHandler3 in vcruntime140.dll is not supported for this application type. ChakraCore.dll calls this API.
API __std_terminate in vcruntime140.dll is not supported for this application type. ChakraCore.dll calls this API.
API __std_type_info_destroy_list in vcruntime140.dll is not supported for this application type. ChakraCore.dll calls this API.
API __telemetry_main_invoke_trigger in vcruntime140.dll is not supported for this application type. ChakraCore.dll calls this API.
API __telemetry_main_return_trigger in vcruntime140.dll is not supported for this application type. ChakraCore.dll calls this API.
API _purecall in vcruntime140.dll is not supported for this application type. ChakraCore.dll calls this API.
API memcmp in vcruntime140.dll is not supported for this application type. ChakraCore.dll calls this API.
API memcpy in vcruntime140.dll is not supported for this application type. ChakraCore.dll calls this API.
API memmove in vcruntime140.dll is not supported for this application type. ChakraCore.dll calls this API.
API memset in vcruntime140.dll is not supported for this application type. ChakraCore.dll calls this API.
API wcschr in vcruntime140.dll is not supported for this application type. ChakraCore.dll calls this API.
API wcsrchr in vcruntime140.dll is not supported for this application type. ChakraCore.dll calls this API.
API wcsstr in vcruntime140.dll is not supported for this application type. ChakraCore.dll calls this API.
API RegCloseKey in advapi32.dll is not supported for this application type. ChakraCore.dll calls this API.
API RegGetValueW in advapi32.dll is not supported for this application type. ChakraCore.dll calls this API.
API RegOpenKeyExW in advapi32.dll is not supported for this application type. ChakraCore.dll calls this API.
API AddAtomW in kernel32.dll is not supported for this application type. ChakraCore.dll calls this API.
API CompareStringW in kernel32.dll is not supported for this application type. ChakraCore.dll calls this API.
API DebugBreak in kernel32.dll is not supported for this application type. ChakraCore.dll calls this API.
API DeleteAtom in kernel32.dll is not supported for this application type. ChakraCore.dll calls this API.
API FindAtomW in kernel32.dll is not supported for this application type. ChakraCore.dll calls this API.
API FindResourceExW in kernel32.dll is not supported for this application type. ChakraCore.dll calls this API.
API FlushInstructionCache in kernel32.dll is not supported for this application type. ChakraCore.dll calls this API.
API GetConsoleScreenBufferInfo in kernel32.dll is not supported for this application type. ChakraCore.dll calls this API.
API GetDateFormatW in kernel32.dll is not supported for this application type. ChakraCore.dll calls this API.
API GetEnvironmentVariableW in kernel32.dll is not supported for this application type. ChakraCore.dll calls this API.
API GetLocaleInfoW in kernel32.dll is not supported for this application type. ChakraCore.dll calls this API.
API GetLogicalProcessorInformation in kernel32.dll is not supported for this application type. ChakraCore.dll calls this API.
API GetModuleHandleW in kernel32.dll is not supported for this application type. ChakraCore.dll calls this API.
API GetProcessIoCounters in kernel32.dll is not supported for this application type. ChakraCore.dll calls this API.
API GetStdHandle in kernel32.dll is not supported for this application type. ChakraCore.dll calls this API.
API GetSystemTimeAdjustment in kernel32.dll is not supported for this application type. ChakraCore.dll calls this API.
API GetTickCount in kernel32.dll is not supported for this application type. ChakraCore.dll calls this API.
API GetTimeFormatW in kernel32.dll is not supported for this application type. ChakraCore.dll calls this API.
API GetUserDefaultLCID in kernel32.dll is not supported for this application type. ChakraCore.dll calls this API.
API GlobalMemoryStatusEx in kernel32.dll is not supported for this application type. ChakraCore.dll calls this API.
API LCMapStringW in kernel32.dll is not supported for this application type. ChakraCore.dll calls this API.
API LoadLibraryExA in kernel32.dll is not supported for this application type. ChakraCore.dll calls this API.
API LoadLibraryExW in kernel32.dll is not supported for this application type. ChakraCore.dll calls this API.
API LoadResource in kernel32.dll is not supported for this application type. ChakraCore.dll calls this API.
API LockResource in kernel32.dll is not supported for this application type. ChakraCore.dll calls this API.
API QueryThreadCycleTime in kernel32.dll is not supported for this application type. ChakraCore.dll calls this API.
API RtlAddFunctionTable in kernel32.dll is not supported for this application type. ChakraCore.dll calls this API.
API RtlCaptureContext in kernel32.dll is not supported for this application type. ChakraCore.dll calls this API.
API RtlDeleteFunctionTable in kernel32.dll is not supported for this application type. ChakraCore.dll calls this API.
API RtlVirtualUnwind in kernel32.dll is not supported for this application type. ChakraCore.dll calls this API.
API SetConsoleTextAttribute in kernel32.dll is not supported for this application type. ChakraCore.dll calls this API.
API SetThreadStackGuarantee in kernel32.dll is not supported for this application type. ChakraCore.dll calls this API.
API SizeofResource in kernel32.dll is not supported for this application type. ChakraCore.dll calls this API.
API TlsAlloc in kernel32.dll is not supported for this application type. ChakraCore.dll calls this API.
API TlsFree in kernel32.dll is not supported for this application type. ChakraCore.dll calls this API.
API TlsGetValue in kernel32.dll is not supported for this application type. ChakraCore.dll calls this API.
API TlsSetValue in kernel32.dll is not supported for this application type. ChakraCore.dll calls this API.
API UnhandledExceptionFilter in kernel32.dll is not supported for this application type. ChakraCore.dll calls this API.
API VerSetConditionMask in kernel32.dll is not supported for this application type. ChakraCore.dll calls this API.
API VerifyVersionInfoW in kernel32.dll is not supported for this application type. ChakraCore.dll calls this API.
API VirtualAlloc in kernel32.dll is not supported for this application type. ChakraCore.dll calls this API.
API VirtualProtect in kernel32.dll is not supported for this application type. ChakraCore.dll calls this API.
API CharLowerBuffW in user32.dll is not supported for this application type. ChakraCore.dll calls this API.
API CharUpperBuffW in user32.dll is not supported for this application type. ChakraCore.dll calls this API.
API GetFileVersionInfoExW in version.dll is not supported for this application type. ChakraCore.dll calls this API.
API GetFileVersionInfoSizeExW in version.dll is not supported for this application type. ChakraCore.dll calls this API.
API VerQueryValueW in version.dll is not supported for this application type. ChakraCore.dll calls this API.
| Chakra embedded in UWP application fails WAC tool | https://api.github.com/repos/chakra-core/ChakraCore/issues/989/comments | 3 | 2016-05-14T23:51:40Z | 2016-06-14T14:44:08Z | https://github.com/chakra-core/ChakraCore/issues/989 | 154,881,686 | 989 |
[
"chakra-core",
"ChakraCore"
] | Steps to reproduce in Edge 13:
Point browser to a page serving the following JS file:
``` js
eval('(function(){let a; class A { x() { return a; } } })()');
```
This will cause the page to fail to load, immediately reloading the HTML and JS from the server a second time, and then giving up with a message of "This page is having a problem loading. We tried to load this page for you a few times ...".
The crash can be averted in any of several different ways:
1. Remove the `eval` - the code itself works fine outside an eval
2. Change `let` to `var` - no problems there [note that changing to `const` (and assigning something, as required) does _not_ prevent the crash]
3. Change the method to a function (i.e. remove the class) - ordinary functions have no problems
4. Remove the IIFE - top-level `let`s seem to be fine, too
5. After a crash, open devtools (F12), highlight the URL, and hit enter - may take a few tries before it finally succeeds, but once it does, it continues to succeed
The crash appears to be independent of strict mode - adding `"use strict";` either at the beginning of the eval, or immediately inside the function - neither makes a difference.
| Crash when a class closes over lexically-bound variables inside an eval | https://api.github.com/repos/chakra-core/ChakraCore/issues/974/comments | 1 | 2016-05-12T01:27:42Z | 2016-05-12T02:33:11Z | https://github.com/chakra-core/ChakraCore/issues/974 | 154,377,835 | 974 |
[
"chakra-core",
"ChakraCore"
] | ``` js
Function('a//', 'return a;')(1);
```
Expected: returns 1
Actual: SyntaxError
Per 11.4, comments are treated like whitespace and are ignored.
In order to construct something that can toString properly, we can align with the stage 3 [F.p.toString proposal](https://tc39.github.io/Function-prototype-toString-revision/#sec-createdynamicfunction). To wit:
``` js
Function('a//', 'return a;').toString();
/*
function anonymous(a//
) {return a;}
*/
```
| Various Function constructors should allow single-line comments | https://api.github.com/repos/chakra-core/ChakraCore/issues/971/comments | 2 | 2016-05-11T23:12:41Z | 2016-05-14T04:51:59Z | https://github.com/chakra-core/ChakraCore/issues/971 | 154,363,582 | 971 |
[
"chakra-core",
"ChakraCore"
] | See daily build statuses for all branches here:
https://github.com/Microsoft/ChakraCore/wiki/Build-Status#disablejit-builds
https://github.com/Microsoft/ChakraCore/wiki/Build-Status-%28release-1.2%29#disablejit-builds
https://github.com/Microsoft/ChakraCore/wiki/Build-Status-%28linux%29#disablejit-builds
See logs for example:
http://dotnet-ci.cloudapp.net/job/Microsoft_ChakraCore/job/master/job/daily_disablejit_x64_debug/58/console
http://dotnet-ci.cloudapp.net/job/Microsoft_ChakraCore/job/release_1.2/job/daily_disablejit_x64_debug/40/console
http://dotnet-ci.cloudapp.net/job/Microsoft_ChakraCore/job/linux/job/daily_disablejit_x64_debug/9/console
@Krovatkin I believe this is your bug because you're the only person who has touched these files. Please fix with high priority to unblock merging the linux branch into master.
FYI @digitalinfinity
| [Test failure] [Daily] [DisableJIT] [All Branches] testSimdManyVars.js b95.js b108.js failing in Daily DisableJIT builds. | https://api.github.com/repos/chakra-core/ChakraCore/issues/964/comments | 2 | 2016-05-10T23:21:47Z | 2016-05-11T19:33:33Z | https://github.com/chakra-core/ChakraCore/issues/964 | 154,128,072 | 964 |
[
"chakra-core",
"ChakraCore"
] | e.g. CallInfo.h, is following "unused" patch applicable to xplat x64?
```
CallInfo(CallFlags flags, ushort count)
: Flags(flags)
, Count(count)
#ifdef _WIN64
, unused(0)
#endif
{
}
// Assumes big-endian layout
// If the size of the count is changed, change should happen at following places also
// - scriptdirect.idl
// - LowererMDArch::LoadInputParamCount
//
unsigned Count : 24;
CallFlags Flags : 8;
#ifdef _WIN64
unsigned unused : 32;
#endif
```
| xplat: examine _WIN64, should it be _M_X64? | https://api.github.com/repos/chakra-core/ChakraCore/issues/963/comments | 5 | 2016-05-10T19:53:07Z | 2018-05-16T22:31:50Z | https://github.com/chakra-core/ChakraCore/issues/963 | 154,092,034 | 963 |
[
"chakra-core",
"ChakraCore"
] | It is still correct in the sense that it will get at least all of the files edited by a pull request, but it is checking a much larger set of files than necessary. Would be nice to clean this up.
| Copyright Check takes a diff of the current branch versus master instead of the target branch. | https://api.github.com/repos/chakra-core/ChakraCore/issues/962/comments | 9 | 2016-05-10T18:58:30Z | 2017-05-10T20:15:03Z | https://github.com/chakra-core/ChakraCore/issues/962 | 154,080,981 | 962 |
[
"chakra-core",
"ChakraCore"
] | At the moment, ChakraCore linux support requires a certain minimum clang version. We have a couple of command line arguments to feed build script with the clang++ and clang.
The output is not clear;
- If the user only provides the path for cc.
- If the clang or clang++ from path are not matching the minimum required versions.
| cross-platform: Improve clang error reporting with build.sh | https://api.github.com/repos/chakra-core/ChakraCore/issues/954/comments | 0 | 2016-05-09T19:26:39Z | 2016-07-06T20:45:29Z | https://github.com/chakra-core/ChakraCore/issues/954 | 153,850,397 | 954 |
[
"chakra-core",
"ChakraCore"
] | Currently `target` and `handler` as internal fields of `JavascriptProxy.cpp` and there is no way to retrieve them natively. Recently node [started taking dependency](https://github.com/nodejs/node/commit/ba6196f84369f1c9c6ab87635de61f6162682cc3#diff-2551d2db08a9c000abf562d51e3de404R42) on these fields natively. This breaks node on chakracore as chakracore doesn't have these fields exposed via JSRT. Additionally as per ES6 spec, there is no way to check if an object is a proxy or not.
This issue is to track adding below JSRT APIs
- `GetTargetOfProxy`
- `GetHandlerOfProxy`
- `IsProxy`
| JSRT APIs for Proxy | https://api.github.com/repos/chakra-core/ChakraCore/issues/950/comments | 12 | 2016-05-09T00:03:43Z | 2018-04-25T16:37:40Z | https://github.com/chakra-core/ChakraCore/issues/950 | 153,684,851 | 950 |
[
"chakra-core",
"ChakraCore"
] | Currently we expect shamt to be Unsigned. This was inspired by the SIMD.js spec to cast the shamt to Uint32 ([link](http://tc39.github.io/ecmascript_simd/#simd-shift-left-by-scalar))
However, Emscripten passes Ints. So we have to relax the signatures of SIMD shift ops.
We will add a fix for this in https://github.com/Microsoft/ChakraCore/pull/941
Repro case:
``` javascript
function asmModule(stdlib, imports,buffer) {
"use asm";
var i4 = stdlib.SIMD.Int32x4;
var i4check = i4.check;
var i4shiftLeftByScalar = i4.shiftLeftByScalar;
var i4shiftRightByScalar = i4.shiftRightByScalar;
var u4 = stdlib.SIMD.Uint32x4;
var u4check = u4.check;
var u4shiftLeftByScalar = u4.shiftLeftByScalar ;
var u4shiftRightByScalar= u4.shiftRightByScalar ;
var loopCOUNT = 3;
function func1(a)
{
a = i4check(a);
var x = u4(0, 0, 0, 0);
var y = u4(0, 0, 0, 0);
var loopIndex = 0;
var loopCOUNT = 3;
while ( (loopIndex|0) < (loopCOUNT|0)) {
x = u4shiftLeftByScalar(x, loopIndex);
y = u4shiftRightByScalar(y, loopIndex);
loopIndex = (loopIndex + 1) | 0;
}
return i4check(a);
}
return {func1:func1};
}
var buffer = new ArrayBuffer(0x10000);
var m = asmModule(this, {}, buffer);
var v = SIMD.Int32x4(1, 2, 3, 4 );
var ret1 = m.func1(v);
print(ret1);
```
Actual:
```
ch.exe -simdjs -asmjs -testtrace:asmjs -asmjsstoponerror shiftBug.js
shiftBug.js(26, 18)
Asm.js Compilation Error function : asmModule::func1
SIMD builtin function doesn't support arguments
Asm.js compilation failed.
FATAL ERROR: ch.exe failed due to exception code c0000005
```
Expected:
```
ch.exe -simdjs -asmjs -testtrace:asmjs -asmjsstoponerror shiftBug.js
Successfully compiled asm.js code
SIMD.Int32x4(1, 2, 3, 4)
```
| Simd.js in Asm.js: Relax validation of shift amount type for SIMD shift ops | https://api.github.com/repos/chakra-core/ChakraCore/issues/944/comments | 3 | 2016-05-06T19:26:04Z | 2016-05-11T17:52:35Z | https://github.com/chakra-core/ChakraCore/issues/944 | 153,521,398 | 944 |
[
"chakra-core",
"ChakraCore"
] | Lowerer sequence for Simd.RightShift clobbers register's value even though it might still be live downstream.
Testcase:
[shr_bug.js.txt](https://github.com/Microsoft/ChakraCore/files/252821/shr_bug.js.txt)
``` javascript
function asmModule(stdlib, imports) {
"use asm";
var i16 = stdlib.SIMD.Int8x16;
var i16c = i16.check;
var i4 = stdlib.SIMD.Int32x4;
var i16shiftRightByScalar = i16.shiftRightByScalar;
var i16add = i16.add;
var i4extractLane = i4.extractLane;
function f2(a)
{
a = i16c(a);
return a;
}
function testShiftLeftScalarLocal()
{
var a = i16(255, 40, -40, 40, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16);
var b = i4(255,0,0,0);
var i = 0;
i = i4extractLane (b, 0);
a = i16shiftRightByScalar(a, i >>> 0);
a = i16c(f2(a));
return i|0;
}
return {
testShiftLeftScalarLocal: testShiftLeftScalarLocal };
}
var m = asmModule(this, {});
print (m.testShiftLeftScalarLocal());
print (m.testShiftLeftScalarLocal());
```
Actual:
```
ch.exe -simdjs -asmjs -maic:0 shamt.js
7
7
```
Expected
```
ch.exe -simdjs -asmjs- shamt.js
255
255
```
| Asm.js Simd.js : Wrong result due to Simd.RightShift clobbers register's value | https://api.github.com/repos/chakra-core/ChakraCore/issues/940/comments | 1 | 2016-05-06T18:27:18Z | 2016-05-12T20:46:50Z | https://github.com/chakra-core/ChakraCore/issues/940 | 153,510,645 | 940 |
[
"chakra-core",
"ChakraCore"
] | For the test/CI related work, IMO we better have a separate branch (can be forked from linux) and we do the work there. Once it's up and running we `squash the commits` and `cherry pick the one` back to linux or whatever the origin branch was.
Sounds good ?
| CI/Test related commits | https://api.github.com/repos/chakra-core/ChakraCore/issues/937/comments | 4 | 2016-05-06T16:11:25Z | 2016-05-06T19:28:43Z | https://github.com/chakra-core/ChakraCore/issues/937 | 153,485,636 | 937 |
[
"chakra-core",
"ChakraCore"
] | Simd builtin return values should be allowed to be coerced to all the types where a valid coercion exists as allowed by the asm.js spec.
http://asmjs.org/spec/latest/#unary-operators
**Repro Case**
``` javascript
function asmModule(stdlib, imports, buffer) {
"use asm";
var fround = stdlib.Math.fround;
var f4 = stdlib.SIMD.Float32x4;
var f4ext = f4.extractLane;
function func1()
{
//var retn = fround(0);
var retn = 0.0;
var st = f4(1.0, 2.0, 3.0, 4.0);
retn = +f4ext(st,0);
}
return {func1:func1};
}
```
**Expected Results**
Asm.js Compilation completed successfully
**Actual Result**
Asm.js Compilation Error function : asmModule::func1
| Asmjs compilation error on assigning float/signed/unsigned return values of simd builtin to a double var | https://api.github.com/repos/chakra-core/ChakraCore/issues/933/comments | 3 | 2016-05-06T00:29:11Z | 2016-05-10T00:04:19Z | https://github.com/chakra-core/ChakraCore/issues/933 | 153,354,961 | 933 |
[
"chakra-core",
"ChakraCore"
] | See: https://github.com/Microsoft/ChakraCore/wiki/Build-Status-%28linux%29#legacy-builds
Logs:
http://dotnet-ci.cloudapp.net/job/Microsoft_ChakraCore/job/linux/job/daily_dev12_x86_debug/34/
http://dotnet-ci.cloudapp.net/job/Microsoft_ChakraCore/job/linux/job/daily_dev12_x64_debug/34/
Sample error:
`Chakra.Runtime.Base.lib(Constants.obj) : error LNK2005: "public: static unsigned int const Js::Constants::InvalidSourceIndex" (?InvalidSourceIndex@Constants@Js@@2IB) already defined in Chakra.Runtime.Base.lib(FunctionBody.obj) [d:\j\workspace\daily_dev12_x---e0d643ad\bin\ChakraCore\ChakraCore.vcxproj]`
| Link errors in legacy debug builds in linux branch | https://api.github.com/repos/chakra-core/ChakraCore/issues/928/comments | 0 | 2016-05-05T21:39:36Z | 2016-08-06T00:52:16Z | https://github.com/chakra-core/ChakraCore/issues/928 | 153,332,392 | 928 |
[
"chakra-core",
"ChakraCore"
] | These are the items on our asm.js/simd.js spring cleaning list:
- [x] Disable **Float64x2** and **FromFloat64x2** opcodes
- [x] Remove **"and, or, xor, not"** for float32x4 type
- [x] Remove **"min/max"** for integral types
- [x] Remove/clean-up unit tests
- [ ] Get a go from MSFT
- [ ] lunch?
Actual Output:
```
Not Available
```
Expected Output:
```
Not Available
```
| Simd.js Asm.js : Disable Float64x2 and remove SIMD ops not in the spec | https://api.github.com/repos/chakra-core/ChakraCore/issues/907/comments | 6 | 2016-05-02T18:27:48Z | 2016-05-10T00:49:10Z | https://github.com/chakra-core/ChakraCore/issues/907 | 152,612,988 | 907 |
[
"chakra-core",
"ChakraCore"
] | Followed the information at http://ariya.ofilabs.com/2016/01/playing-with-chakracore-on-linux.html
except that I use virtualbox as the VM and install the ubuntu15.10 from http://releases.ubuntu.com/15.10/
the error message I am seeing is:
c++: error: unrecognized command line option ‘-fdelayed-template-parsing’
pal/src/CMakeFiles/Chakra.Pal.dir/build.make:54: 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
CMakeFiles/Makefile2:90: recipe for target 'pal/src/CMakeFiles/Chakra.Pal.dir/all' failed
make[1]: **\* [pal/src/CMakeFiles/Chakra.Pal.dir/all] Error 2
Makefile:116: recipe for target 'all' failed
make: **\* [all] Error 2
| ChakraCore cannot build on Ubuntu 15.10 (x64) | https://api.github.com/repos/chakra-core/ChakraCore/issues/906/comments | 12 | 2016-05-02T17:27:29Z | 2016-05-11T18:28:44Z | https://github.com/chakra-core/ChakraCore/issues/906 | 152,600,922 | 906 |
[
"chakra-core",
"ChakraCore"
] | ``` HTML
<!DOCTYPE html>
<html>
<head>
<title></title>
<meta charset="utf-8" />
<script src="https://ajax.googleapis.com/ajax/libs/jquery/2.2.2/jquery.min.js"></script>
<script>
$(document).ready(function () {
$('#doc').prop('data', 'anotherDocument.pdf');
});
</script>
</head>
<body>
<object id="doc" type="application/pdf" data="document.pdf" width="500px" height="500px"></object>
</body>
</html>
```
| modify data attribute dynamically of object tag not working | https://api.github.com/repos/chakra-core/ChakraCore/issues/903/comments | 9 | 2016-05-01T02:58:26Z | 2016-11-15T22:07:27Z | https://github.com/chakra-core/ChakraCore/issues/903 | 152,065,729 | 903 |
[
"chakra-core",
"ChakraCore"
] | Asm.js functions with simd args/return values are not allowed to use unsigned simd types for args/returns.
tc39/ecmascript_simd#307 (comment)
"Arguments passing : No unsigned Integer SIMD arguments are allowed. No SIMD arguments are allowed to imported functions."
**Repro Case**
``` javascript
function asmModule(stdlib) {
"use asm";
var u4 = stdlib.SIMD.Uint32x4;
var u4check = u4.check;
var u4ff4 = u4.fromFloat32x4;
var f4 = stdlib.SIMD.Float32x4;
function func1()
{
var a = f4(1.9, 0.0, -0.9, -0.0);
return u4check(u4ff4(a));
}
return {func1:func1};
}
var m = asmModule(this);
var ret2 = m.func1();
WScript.Echo(ret2);
```
**Expected Results**
> stDstBug.js(7, 5)
> Asm.js Compilation Error function : asmModule::func1
> Expression for return must be subtype of Signed, Double, or Float
>
> Asm.js compilation failed.
> FATAL ERROR: ch.exe failed due to exception code c0000005
**Actual Results**
> Successfully compiled asm.js code
> SIMD.Uint32x4(1, 0, 0, 0);
| Simd.js: Restrict Asm.js functions to use unsigned simd values as arguments or return values. | https://api.github.com/repos/chakra-core/ChakraCore/issues/897/comments | 0 | 2016-04-29T19:38:54Z | 2016-05-03T00:22:41Z | https://github.com/chakra-core/ChakraCore/issues/897 | 151,952,313 | 897 |
[
"chakra-core",
"ChakraCore"
] | While converting from float32 to Uint32, the values needs to be truncated before performing bound check to be able to convert float32 values between 0 and -1.0. Only values <= -1.0 should throw.
http://tc39.github.io/ecmascript_simd/#simd-to-timd-logical
**Repro Case**
``` javascript
function asmModule(stdlib) {
"use asm";
var u4 = stdlib.SIMD.Uint32x4;
var u4check = u4.check;
var u4ff4 = u4.fromFloat32x4;
var f4 = stdlib.SIMD.Float32x4;
var f4check = f4.check;
function func1()
{
var a = f4(1.9, 0.0, -0.9, -0.0);
return f4check(f4);
}
return {func1:func1};
}
var m = asmModule(this);
var ret2 = m.func1();
WScript.Echo(ret2);
```
**Expected Results**
> Successfully compiled asm.js code
> SIMD.Uint32x4(1, 0, 0, 0)
**Actual Results**
> Successfully compiled asm.js code
> RangeError: argument out of range
> at Global code (g:\bugs\stdstbug.js:15:1)
| Simd.js Bug: F32 truncation to be done before bound check for Uint32x4.fromFloat32x4 conversion builtin | https://api.github.com/repos/chakra-core/ChakraCore/issues/890/comments | 1 | 2016-04-29T16:43:13Z | 2016-05-09T23:34:09Z | https://github.com/chakra-core/ChakraCore/issues/890 | 151,916,495 | 890 |
[
"chakra-core",
"ChakraCore"
] | Handling of extended opcodes with Medium or large layouts is disabled in InterpreterLoop.inl. This will cause a crash when interpreter SIMD opcodes (extended ones) with functions with over 256 bytecode registers.
https://github.com/Microsoft/ChakraCore/pull/831/files/5dfff42fdd6fe01a8b5680dc0b1e89b827594370#r61509222 (see for more information)
| Asm.js Simd.js : Handling of extended opcodes with Medium or large layouts is disabled in InterpreterLoop.inl for Asm.js | https://api.github.com/repos/chakra-core/ChakraCore/issues/888/comments | 8 | 2016-04-29T16:36:16Z | 2016-05-06T23:29:04Z | https://github.com/chakra-core/ChakraCore/issues/888 | 151,914,685 | 888 |
[
"chakra-core",
"ChakraCore"
] | Repro test case:
``` js
var h = { has : function() { WScript.Echo("has"); return true; }};
var p = new Proxy({}, h);
var o = {};
Object.defineProperty(o, "x", p);
```
Output:
```
has
ASSERTION 3060: (D:\src\Chakra\core\lib\Runtime\Language\JavascriptOperators.cpp, line 8661) Proxy : HasProperty and GetProperty's result don't match for 'enumerable'. Failure: (0)
FATAL ERROR: jshost.exe failed due to exception code c0000420
```
Expected output:
```
has
has
has
has
has
has
TypeError: Property cannot have both accessors and a value
```
Each `has` is for the Proxy properties: configurable, enumerable, value, writable, get, and set that are checked when there is a proxy around the PropertyDescriptor being used with Object.defineProperty.
See: https://developer.mozilla.org/en-US/docs/Web/JavaScript/Reference/Global_Objects/Proxy
See: https://tc39.github.io/ecma262/2016/#sec-object.defineproperty
See: https://tc39.github.io/ecma262/2016/#sec-topropertydescriptor
The reason there should be a `TypeError` is because according to spec semantics, when the proxy says `has()` is true for a property, we should try to look it up, and if we discover that it is not present, then we should set it to the default value (falsy: undefined for functions and false for non-functions). Since certain combinations of these PropertyDescriptor properties are disallowed, we see a `TypeError` in this case.
Converted from [7019635](https://microsoft.visualstudio.com/web/wi.aspx?pcguid=cb55739e-4afe-46a3-970f-1b49d8ee7564&id=7019635)
/cc @bterlson to double check details.
| Proxy: ToPropertyDescriptor assert; when Proxy has() always lies and returns true, should result in TypeError | https://api.github.com/repos/chakra-core/ChakraCore/issues/884/comments | 2 | 2016-04-28T21:09:35Z | 2016-04-30T00:09:02Z | https://github.com/chakra-core/ChakraCore/issues/884 | 151,744,021 | 884 |
[
"chakra-core",
"ChakraCore"
] | // Issue in Edge but not in chrome
var t = dt.toLocaleTimeString(); // Say it returns 13:04:26
var tArr = t.split(':');
console.log(parseInt(tArr[0], 10); // returns NaN in Edge, other browsers return 13
console.log(+tArr[0]); // returns NaN in Edge, other browsers return 13
console.log(1 \* tArr[0]); // returns NaN in Edge, other browsers return 13
// Solution to the problem for Edge and possibly IE11
function removeNonAscii(str) {
var regx = new RegExp("[^\u0000-\u007F]", "g");
return str.replace(regx, '');
}
var t = removeNonAscii(dt.toLocaleTimeString()); // Say it returns 13:04:26
var tArr = t.split(':');
console.log(parseInt(tArr[0], 10); // returns 13
console.log(+tArr[0]); // returns 13
console.log(1 \* tArr[0]); // returns 13
| toLocaleTimeString returns non-ascii characters | https://api.github.com/repos/chakra-core/ChakraCore/issues/883/comments | 2 | 2016-04-28T19:18:03Z | 2016-04-30T00:12:02Z | https://github.com/chakra-core/ChakraCore/issues/883 | 151,721,708 | 883 |
[
"chakra-core",
"ChakraCore"
] | modify data attribute dynamically of object tag not working | https://api.github.com/repos/chakra-core/ChakraCore/issues/881/comments | 3 | 2016-04-28T11:26:30Z | 2016-11-15T22:07:03Z | https://github.com/chakra-core/ChakraCore/issues/881 | 151,616,653 | 881 | |
[
"chakra-core",
"ChakraCore"
] | I hit some less-than-pointer (like 3 bytes on x86) allocation with leaf bit off in both chakra and memGC
we should fix the caller(runtime code) to make such allocation is leaf, the benefits are:
1. reduce addresses to scan
2. reduce false positive in marking
we can add an assert in recycler to find out such allocations
| Assert less-than-pointer allocation to be leaf | https://api.github.com/repos/chakra-core/ChakraCore/issues/880/comments | 2 | 2016-04-28T05:59:34Z | 2018-02-22T00:06:37Z | https://github.com/chakra-core/ChakraCore/issues/880 | 151,555,869 | 880 |
[
"chakra-core",
"ChakraCore"
] | When allocating pages sometimes we only need a part of the pages to be commited. if page allocator support this we can:
1. simplify pageheap, we do not need to decommit guard pages and maintain the page status in heap block code
2. some medium block has useless pages in the end, we can only commit necessary part of the pages
| make pageAllocator able to commit part of requested pages | https://api.github.com/repos/chakra-core/ChakraCore/issues/879/comments | 4 | 2016-04-28T05:55:22Z | 2018-03-22T00:12:18Z | https://github.com/chakra-core/ChakraCore/issues/879 | 151,555,408 | 879 |
[
"chakra-core",
"ChakraCore"
] | This seems like a register allocation bug when nested calls are involved. The issue happens when creating/finding register for dstOpnd during the irbuilding phase of the SIMD128ReturnU4 opcode. There seem to be a mismatch in the return register type (Expecting U4 and finding I4)
**Repro Case**
> ``` javascript
> function asmModule(stdlib, imports) {
> "use asm";
> var i4 = stdlib.SIMD.Int32x4;
> var i4check = i4.check;
> var u4 = stdlib.SIMD.Uint32x4;
> var u4check = u4.check;
> function func1(a)
> {
> a = i4check(a);
> return a;
> }
> function func2()
> {
> var x = u4(-1, -2, -3, -4);
> var k = i4(-1, -2, -3, -4);;
> k = i4check(func1(k));
> return u4check(x); //Deliberate use of x here
> }
> return {func1:func1, func2:func2};
> }
> var m = asmModule(this, {g1:SIMD.Float32x4(90934.2,123.9,419.39,449.0), g2:SIMD.Uint32x4(-1065353216, -1073741824,-1077936128, -1082130432)});
> var ret2 = m.func2();
> ```
**Flags**
> x64_debug
> -DumpOnCrash -bgjit- -on:asmjs -testtrace:asmjs -simdjs -AsmJsStopOnError -maic:0 -lic:0
**Actual result**
> Successfully compiled asm.js code
> ASSERTION 30176: (G:\arun\git\chakra-310\core\lib\Backend\Sym.cpp, line 382) stackSym->GetType() == type
> Failure: (stackSym->GetType() == type)
> dump filename C:\Users\b-arpuru\AppData\Local\Temp\CH_30176_18664.dmp
> FATAL ERROR: ch.exe failed due to exception code c0000420
| Simd.js Bug: Destination register type mismatch assertion while returning simd types from asm functions. | https://api.github.com/repos/chakra-core/ChakraCore/issues/867/comments | 2 | 2016-04-26T21:13:16Z | 2016-04-29T21:34:14Z | https://github.com/chakra-core/ChakraCore/issues/867 | 151,240,406 | 867 |
[
"chakra-core",
"ChakraCore"
] | The helper method used to build IR for simd type conversion operations sets the destination register type to 'TySimd128I4' always.
**Repro Case**
``` javascript
function asmModule(stdlib) {
"use asm";
var i4 = stdlib.SIMD.Int32x4;
var i4check = i4.check;
var u4 = stdlib.SIMD.Uint32x4;
var u4swizzle = u4.swizzle;
var u4shuffle = u4.shuffle;
var u4fi4 = u4.fromInt32x4Bits;
var i4fu4 = i4.fromUint32x4Bits;
function func1(a)
{
a = i4check(a);
var x = u4(0, 0, 0, 0);
x = u4swizzle(x, 3, 2, 1, 0);
x = u4shuffle(u4fi4(a), x, 0, 1, 7, 6);
return i4check(i4fu4(x));
}
return {func1:func1};
}
var m = asmModule(this);
var ret2 = m.func1(SIMD.Int32x4(1,2,3,4));
```
**Expected Results**
> Successfully compiled asm.js code
> PASS
**Actual Results**
> Successfully compiled asm.js code
> ASSERTION 16232: (G:\arun\git\chakra-411\core\lib\Backend\Sym.cpp, line 382) stackSym->GetType() == type
> Failure: (stackSym->GetType() == type)
> FATAL ERROR: ch.exe failed due to exception code c0000420
| Simd.js Bug: Simd type conversions in asmjs sets the dst type incorrectly | https://api.github.com/repos/chakra-core/ChakraCore/issues/866/comments | 2 | 2016-04-26T21:10:26Z | 2016-04-29T21:34:24Z | https://github.com/chakra-core/ChakraCore/issues/866 | 151,239,853 | 866 |
[
"chakra-core",
"ChakraCore"
] | Incorrect register reuse for unused simd temp registers. If the simd function call is not part of an assignment a temp dst register is created and since its unused its marked for reuse. However the stack symbol associated to this tmp register is typed and will assert when reused for a simd value with a different type.
**Repro Case**
``` javascript
function asmModule(stdlib) {
"use asm";
var i4 = stdlib.SIMD.Int32x4;
var i4check = i4.check;
var u4 = stdlib.SIMD.Uint32x4;
var u4add = u4.add;
function func1(a, b)
{
a = i4check(a);
b = i4check(b);
var x = u4(-1, -2, -3, -4);
u4add(x,x);
return i4check(a);
}
return {func1:func1};
}
var m = asmModule(this);
var ret2 = m.func1(SIMD.Int32x4(-1, -2, -3, -4),SIMD.Int32x4(-1, -2, -3, -4));
```
**Expected Results**
> Successfully compiled asm.js code
**Actual Results**
> Successfully compiled asm.js code
> ASSERTION 15960: (G:\arun\git\chakra-411\core\lib\Backend\Sym.cpp, line 382) stackSym->GetType() == type
> Failure: (stackSym->GetType() == type)
> FATAL ERROR: ch.exe failed due to exception code c0000420
| Simd.js Bug: Incorrect reuse of unused simd temp registers in IRBuilder register allocation | https://api.github.com/repos/chakra-core/ChakraCore/issues/865/comments | 0 | 2016-04-26T21:09:39Z | 2016-04-29T00:21:57Z | https://github.com/chakra-core/ChakraCore/issues/865 | 151,239,689 | 865 |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.