Re: Semantic of WebAssembly "select" instruction?

So, It means the (cond ? then : else) kind of semantic has to be implemented using "if" right ? Le mercredi 13 novembre 2019 20:10:20 UTC+1, Thomas Lively a écrit : > > No, it unconditionally evaluates all arguments. > > On Wed, Nov 13, 2019 at 11:09 AM Stéphane Le

Semantic of WebAssembly "select" instruction?

Does WebAssembly "select" instruction behaves like C/C++ (cond ? then : else), that is compute *only* one of the two branches depending of the "cond" value? Thanks. -- You received this message because you are subscribed to the Google Groups "emscripten-discuss" group. To unsubscribe from

Re: uncaught_exceptionv erri

cussion_r284040606 > > which was pulled off into > > https://github.com/emscripten-core/emscripten/issues/8678 > > As a workaround until that is fixed, removing -s ASSERTIONS=1 should avoid > the problem. > > - Alon > > > On Thu, Jul 4, 2019 at 7:23 AM Stéphane

Strange error in EMCC compiled code

We get this error when using C++ code compiled with EMCC 1.38.31 ReferenceError: tempDouble is not defined at Object.doStat (http://127.0.0.1:8000/libwasm-worklet-glue.js:4848:46) at ___syscall197 (http://127.0.0.1:8000/libwasm-worklet-glue.js:6494:23) at _fstat (wasm-function[1849]:0x

uncaught_exceptionv erri

I get this error: libwasm-worklet-glue.js:8 Uncaught TypeError: Cannot assign to read only property '__ZSt18uncaught_exceptionv' of object '[object Object]' at libwasm-worklet-glue.js:8 at clarinetMIDI-processor1.js:7 when loading and starting a WASM compiled library: import FaustModu

Re: Accessing WebAssembly memory ?

re it there (and export it, > if you want). > > - Alon > > > On Thu, May 16, 2019 at 10:22 AM Stéphane Letz > wrote: > >> In a previous version of Emscripten, I was able to access the >> underlying WebAssembly memory by just using the Module['wasmMemory&

Accessing WebAssembly memory ?

In a previous version of Emscripten, I was able to access the underlying WebAssembly memory by just using the Module['wasmMemory'] field. This does not work anymore with code compiled using Emscripten with 1.38.30. Is there an official and reliable way to access the WebAssembly memory? I need

Can EM_JS/EM_ASM be used with await/asyn kind of code?

Can the EM_JS/EM_ASM syntax be used with await/asyn(= asynchronous) kind of code? Thanks. -- You received this message because you are subscribed to the Google Groups "emscripten-discuss" group. To unsubscribe from this group and stop receiving emails from it, send an email to emscripten-disc

How to convert long lasting C/C++ code in async/await JS model ?

We are using a C/C++ library compiled with emscripten. We need to call some long lasting functions and would like to see them as async/await model on the JS side, to avoid blocking the main thread. I read some stuff around emscripten_async_call but cannot see if this is the path do go. Are them

Does WebAssembly.Memory has to be explicitely deleted?

Hu, We are using "new WebAssembly.Memory" kind of API to allocate wasm memory. Does this memory segment has to be explicitly deleted with some API (that I could not find) ? Or is the memory segment just GC by JS as soon as no reference on it exists ? Thanks. -- You received this message bec

Re: Error using emscripten fetch API

', '_cleanupAfterException', '_getErrorAfterException']" -s EXTRA_EXPORTED_RUNTIME_METHODS="['cwrap','Pointer_stringify','stringToUTF8']" gives: clang++: error: unknown argument: '-–proxy-to-worker' So as soon as -s FETCH=1 is

Error using emscripten fetch API

I'm trying to replace old code with the emscripten fetch API, using the example code given here: https://kripken.github.io/emscripten-site/docs/api_reference/fetch.html following the "Synchronous Fetches" section. Our library is then linked with --proxy-to-worker flag. At run time, I get an ex

Re: Symbol access issue with emcc 1.37.32

t have to be explicitly exported. So the code faust_module._malloc, faust_module._free continue to work. Le jeudi 1 février 2018 14:23:36 UTC+1, Stéphane Letz a écrit : > > Compiled with emcc 1.37.32 and -s WASM=1 -s EXPORT_NAME="'FaustModule'" -s > MODULARIZE

Symbol access issue with emcc 1.37.32

Compiled with emcc 1.37.32 and -s WASM=1 -s EXPORT_NAME="'FaustModule'" -s MODULARIZE=1 options, our code was doing: var faust_module = FaustModule(); // Emscripten generated module then we could write this kind of code: faust_module.stringToUTF8(name, name_ptr, faust_module.lengthBytesUTF8(na

Getting "TypeError: Underlying ArrayBuffer has been detached from the view" error with WebAssembly code

Hi, We are getting a "TypeError: Underlying ArrayBuffer has been detached from the view" error with WebAssembly code compiled with Emscripten 1.37.21 and running under Webkit. A more complete bug report has been done on Webkit Bugzilla here : https://bugs.webkit.org/show_bug.cgi?id=182327. The

Re: Linking wasm modules with wasm-merge ?

The problem was in my code. I succeeded running the final wasm-merged code in the browser. But beside very simple cases where I see a gain, using this emcc compiled fast-code code merged with the Faust generated wasm modules shows big regressions (like 2 to 4 times slower...), I don't understand

Re: Linking wasm modules with wasm-merge ?

Thanks, that works well. Now the 2 modules can be properly linked with wasm-merge and the resulting one compiled and instantiated in JS context. Next problem is this kind of error when running it : "RuntimeError: index out of bounds" when calling (for instance...) the fast_log10f function of

Re: Linking wasm modules with wasm-merge ?

OK. I'll have to checkout for memory segments. And I had to change the import naming convention off our generated module to use "env". Now "call merge" are correctly generated. One issue still. The "fastmath" C++ module has 2 set of functions, one using "float" type and one using "double". The

Re: Linking wasm modules with wasm-merge ?

Both modules have memory segments. By asking the lines (import "env" "memoryBase" (global $memoryBase i32)) and (import "env" "tableBase" (global $tableBase i32)) in our generated oddly, the wasm-merge runs without errors. But I'm still not clear in how functions exported by one module (the

Linking wasm modules with wasm-merge ?

Hi, Our Faust compiler (faust.grame.fr) can directly generate wasm modules from the Faust DSP source code. We typically generate modules that need mathematical functions (log, sin, pow...) which are imported from the JS context (by generating the appropriate module "import" section). We would

Debugging a wasm module not generated with Emscripten ?

Hi, When compiling a simple hello_world.c win emcc (./emcc tests/hello_world.c - o hello.html), I see that the resulting HTML can be debugged in FirefoxNightly (see added screenshot). How is that working ? Does emscripten follows code generating convention sso that browsers can access the .wasm

Strange EMCC optimization on some C code

Hi, Our Faust DSP audio language (faust.grame.fr) can generate C,C++ ... or wast/wasm code. On a simple example (signal integrator), we are comparing the wasm code directly generated from our wast/wasm backend, to the C backed later on compiled/optimized in wasm with EMCC. Here are the results

Re: Local stack allocated arrays, and stack management in WebAssembly code

Memory out from a Wasm function local, so >>>>>>> that it can be referenced by a pointer in other functions. >>>>>>> WebAssembly >>>>>>> itself does not care about Emscripten's STACKTOP etc., but all it >>>>>>&g

Re: Local stack allocated arrays, and stack management in WebAssembly code

sm modules). >>>>> > >>>>> > In our wasm code, it is not clear yet how we could possibly define >>>>> stack >>>>> > allocated arrays. Compiling a C++ code example, I see that >>>>> Emscripten >>

Re: Optimization in binaryen.js library

I see. Then I guess we can use the "wasp-opt" tool when we prepare wasm modules "statically" (that is in advance before deploying them in the Web). But I guess that when doing "dynamic compilation" (that is producing the wasm binary module on the fly with the embedded Faust compiler, as it is

Re: Optimization in binaryen.js library

I did optimization tests on a bunch of Faust compiled DSP sources files, using wasp-opt (-03 or 0z), testing on nodejs (version 8.4.0) on OS X El capitan: - something like 10 to 25% maximum save in binary code size - using Date.getTime API to measure the actual audio DSP loop, I could not find

Re: Optimization in binaryen.js library

- yes autovectorization usually helps a lot with this kind of code, as we can see when we produce C/C++ code from the Faust compiler. - code size is usually not an issue, but speed is - for speed measurement, should I simply use all Date.getTime kind of API ? Or are they any more sophisticated

Re: Optimization in binaryen.js library

Our compiler (for the Faust DSP audio language http://faust.grame.fr/ ) generates numerical processing code. Precompiled pages containing Faust generated WebAudio nodes can be found here: http://faust.grame.fr/modules/, using the RUN button to

Optimization in binaryen.js library

Hi, Our compiler directly generates (probably sub-optimal) wasm code. I see that then binaryen "optimize" in binaryen.js library could possibly be used to later optimize the code. What kind of optimisation and so speed gain can be expect? Will binaryen possibly to SIMD auto-vectorization opt

Re: Local stack allocated arrays, and stack management in WebAssembly code

ample, I see that Emscripten >>> > runtime code has some stack operations (like >>> > stackSave/stackAlloc/stackRestore...) and generates stack related code >>> in >>> > the function header. So I understand that Emscripten runtime allocat

Re: Local stack allocated arrays, and stack management in WebAssembly code

s stack related code > in > > the function header. So I understand that Emscripten runtime allocates a > > block of memory to be used for the stack yes?, but what to do in if we > don't > > interact with the Emscripten runtime? Do we need to manage our own

Local stack allocated arrays, and stack management in WebAssembly code

in a similar way as Emscripten runtime does? Or are they any simpler recommended way? Thanks. Stéphane Letz -- You received this message because you are subscribed to the Google Groups "emscripten-discuss" group. To unsubscribe from this group and stop receiving emails from it, se

WebAssembly code generation question

I have this expression : (select v1 v2 (i64.and t1 t2)), when I need to convert the i64 (i64.and t1 t2) back to i32 for the select to correctly work. What should I use to convert (i64.and t1 t2) subexpression to i32 ? Is i32.wrap/i64 adapted for that, like : (select v1 v2 (i32.wrap/i64 (i64.

Speed issues with WebAssembly using JavaScript math functions

Hi, We are testing our wasm backend for Faust DSP audio language (http://faust.grame.fr/news/2017/01/13/faust-webassembly.html). For some examples that mainly use math functions imported from the JavaScript Math context (like code, pow...etc..), we see poor speed performances compared to the n

Re: Equivalent of C remainder function in asm.js?

But : var x = 9.2; var y = 2; console.log(x % y); gives : 1.2 Le jeudi 16 mars 2017 18:38:46 UTC+1, le...@grame.fr a écrit : > > Is there an equivalent of C remainder function in asm.js ? Or if not any > code to emulate it? > > Thanks. > -- You received this message because you are subscribe

Re: Equivalent of C remainder function in asm.js?

Are you sure ? int main() { printf("res %f\n", remainderf(9.2f, 2.0f)); } gives : res -0.80 when : var x = 9.2; var y = 2; console.log((x >>> 0) % (y >>> 0)); console.log((x | 0) % (y | 0)); gives (twice the same result) 1 1 Le jeudi 16 mars 2017 18:38:46 UTC+1, le...@

Equivalent of C remainder function in asm.js?

Is there an equivalent of C remainder function in asm.js ? Or if not any code to emulate it? Thanks. -- You received this message because you are subscribed to the Google Groups "emscripten-discuss" group. To unsubscribe from this group and stop receiving emails from it, send an email to emsc

Re: Memory allocation issue using emcc 1.37.1

Finally I've found the problem in our code, sorry for the noise ! Le mardi 17 janvier 2017 19:46:06 UTC+1, Alon Zakai a écrit : > > The first thing I would do to debug it is to bisect on emcc versions, that > is, find on which version the problem begins. The git log between the two > should be

Memory allocation issue using emcc 1.37.1

Hi, We had a C++ library successfully compiled and running with emcc 1.35. Recompiling and using the code with recent emcc 1.37.1 fais with memory issues. We tried raising the TOTAL_MEMORY value or even using the ALLOW_MEMORY_GROWTH=1 flag without success : an exception is finally raised. We a

Generating WebAssembly binary from our compiler

Hi, We recently added a WebAssembly textual format backend in our audio DSP language compiler (Faust : faust.grame.fr). Now we would like to directly generate binary, but this seems more difficult. Since a lot of code has already been written to deal with WebAssembly (AST, reader/writer..etc..

Re: Assertion failed: !"vector length_error" with emcripten 1.35

Sorry but how to upgrade to SDK after 1.35 (OS X here) ? (./emsdk update ./emsdk install latest...etc..) now stay on 1.35, why that? Le lundi 5 septembre 2016 20:04:15 UTC+2, Alon Zakai a écrit : > > A debug build might help, the stack trace would be more useful. Also see > the settings mention

Assertion failed: !"vector length_error" with emcripten 1.35

Hi, We get the following error with emcripten 1.35 under OS X. We tried activating ALLOW_MEMORY_GROWTH and so on, but no change. What could be the reason? Thanks. == Assertion failed: !"vector length_error", at: /Documents/emsdk_portable/emscripten/1.35.0/system/in

Re: Playing with WebAssembly in Faust

Found the solution: the memory block has to be taken from inside the WebAssembly module... Le mercredi 13 juillet 2016 16:11:14 UTC+2, le...@grame.fr a écrit : > > Hi, > > I am trying to convert our Faust (faust.grame.fr) to asm.js compilation > chain to WebAssembly. Basically what I am doing is

Playing with WebAssembly in Faust

Hi, I am trying to convert our Faust (faust.grame.fr) to asm.js compilation chain to WebAssembly. Basically what I am doing is : 1) Faust DSP ==> asm.js code using the Faust asm.js backend; 2) binaryen tool asm2wasm to convert the asm.js module to a wasm textual file 3) binaryen tool wasm-

Re: asm.js running on iOS?

n a > WKWebView, not UIWebView. (The older UIWebView disables the JIT compiler.) > With WKWebView you should get about the same performance as in Safari, > which is pretty decent on 64-bit (A7 and later CPUs). > > -- brion > > On Wed, Sep 16, 2015 at 6:43 AM, Stéphane Letz > wr

Re: asm.js running on iOS?

an > run on iOS browsers. > > There exists a WebKit blog entry about their FTL JIT, which optimizes for > asm.js style of code: > https://www.webkit.org/blog/3362/introducing-the-webkit-ftl-jit/ . Safari > 8 should have that enabled (as well as having WebGL enabled) > >

asm.js running on iOS?

Hi, Can asm.js code be run on iOS ? (Safari ? Chrome? anything else?) Thanks. Stéphane Letz -- You received this message because you are subscribed to the Google Groups "emscripten-discuss" group. To unsubscribe from this group and stop receiving emails from it, send an email to

asm.js code in worker thread?

Is is possible to use asm.js code in worker threads? If yes, what should be the "stdlib" parameter that is supposed to be given to the module? (in function MyAsmModule(stdlib, foreign, heap); ) Thanks. Stéphane Letz -- You received this message because you are subscribed to

How to transfer char* using EM_ASM ?

Hi, Can string (= char* buffers) be transferred to/from C code calling JS using the EM_ASM construct? Thanks Stéphane Letz -- You received this message because you are subscribed to the Google Groups "emscripten-discuss" group. To unsubscribe from this group and stop receiving e

Re: powf function issue?

> However, an argument could be made that we should turn PRECISE_F32 on in > debug builds or -O0 or something like that, so -O0 is guaranteed to have > the right semantics. But I'm not sure that's the right thing either. > Thoughts? > > - Alon > > > On Fri, J

Re: powf function issue?

m in the first place? Thanks. Stéphane Letz Le jeudi 29 janvier 2015 21:03:00 UTC+1, Alon Zakai a écrit : > > Ok, that is very surprising, so I did more checking. It looks like your > original results were optimized? I tried without any optimization flags, > which leads to good results

Re: powf function issue?

that I > get almost identical results, but still some tiny rounding errors. But even > without it, the errors are much smaller than yours. > > - Alon > > > > On Wed, Jan 28, 2015 at 1:43 PM, Stéphane Letz > wrote: > >> The following code : >

powf function issue?

51 155.563492 pitch 52 164.813778 pitch 52 162.450876 pitch 53 174.614116 pitch 53 177.153937 pitch 54 184.997211 When could be the reason for this strange behavior of the powf function ? Thanks. Stéphane Letz -- You received this message because you are subscribed to the Google Groups &qu

Asm.js module head size

that somewhere? Thanks. Stéphane Letz -- You received this message because you are subscribed to the Google Groups "emscripten-discuss" group. To unsubscribe from this group and stop receiving emails from it, send an email to emscripten-discuss+unsubscr...@googlegroups.com. For mo

Asm.js and stack management

if we don't interact with the Emscripten runtime? Do we need to manage our own stack in a similar way as Emscripten runtime does? Or are they any simpler recommended way? Thanks. Stéphane Letz -- You received this message because you are subscribed to the Google Groups "ems

Re: Error using "chdir"

Well it actually happens (at least in those two cases…) if the "dir" argument of "chdir" is either "" (empty string) or "/". Stéphane Letz Le jeudi 15 janvier 2015 22:09:28 UTC+1, Alon Zakai a écrit : > > You can build with assertions, safe heap,

Re: Error using "chdir"

Compiling i debug mode, I get a bit more explicit error message, does it help? libfaust.js:1 Uncaught TypeError: Cannot read property 'mode' of null : Error at jsStackTrace (file:///Users/letz/Sites/libfaust.js:1:18691) at stackTrace (file:///Users/letz/Sites/libfaust.js:1:188

Re: Error using "chdir"

Not really… use of a big C++ library compiled in JS with emscripten… Is there anything special I should check at the original C++ side ? Stéphane Letz Le jeudi 15 janvier 2015 10:54:00 UTC+1, Stéphane Letz a écrit : > > Hi, > > When compiling C++ code that used "chdir"

Error using "chdir"

Hi, When compiling C++ code that used "chdir", we get the following error : Uncaught TypeError: Cannot read property 'mode' of null : Error at jsStackTrace (libfaust.js:1:18691 <http://localhost/~letz/libfaust.js>) at stackTrace (libfaust.js:1:18874 <http://loc

Compiling a single file without adding the Emscripten runtime?

Hi, I would like to compile a single C++ file in JS without having the complete Emscripten runtime be added in the .js file output. Its that possible and how? Thanks. Stéphane Letz -- You received this message because you are subscribed to the Google Groups "emscripten-discuss"

Re: Emscripten SDK 1.21.0 is out, now with Linux support!

; > > > On Fri, Jul 18, 2014 at 4:52 AM, Stéphane Letz > wrote: > >> After updating to latest SDK 1.21.0, we now get this error when compiling >> our C++ code: >> >> >> /Documents/emsdk-portable/emscripten/1.21.0/tools/eliminator/node_modules/uglif

Re: Emscripten SDK 1.21.0 is out, now with Linux support!

/eliminator/node_modules/uglify-js/lib/parse-js.js:1209:27) Any idea? Thanks. Stéphane Letz Le lundi 7 juillet 2014 15:12:16 UTC+2, jj a écrit : > > Hi everyone, > > it's due time for a new emsdk update, and this one finally brings in > support for Linux as well! > > Down

Re: Memory management issue

see SAFE_HEAP etc. in > https://github.com/kripken/emscripten/wiki/Debugging > > - Alon > > > > On Wed, Jun 25, 2014 at 2:05 AM, Stéphane Letz > wrote: > >> Hi, >> >> We have successfully ported a C++ library in JS using emscripten. Our >>

Http access / web socket

understanding correct? Does that mean that it will be mandatory to have all servers be WS compatible so that our library can access them? Or is there any alternative? Thanks. Stéphane Letz -- You received this message because you are subscribed to the Google Groups "emscripten-discuss"

Memory management issue

5 ) at _abort ( file:///Documents/faust-sf/compiler/libfaust.js:7605 ..n. How could we better trace what happens? Thanks. Stéphane Letz -- You received this message because you are subscribed to the Google Groups "emscripten-discuss" group. To unsubscribe from

Denormalized float numbers in asm.js

he problem in another way ? Thanks. Stéphane Letz -- You received this message because you are subscribed to the Google Groups "emscripten-discuss" group. To unsubscribe from this group and stop receiving emails from it, send an email to emscripten-discuss+unsubscr...@google

Re: Dynamically compiling and evaluating asm.js code?

e firefox console, saying it > succeeded or did not succeed? > > - Alon > > > > On Mon, Jun 9, 2014 at 9:34 AM, Stéphane Letz > > wrote: > >> Hi, >> >> We have successfully ported our Faust (audio DSP) compiler to JS using >> emscripten. We

Dynamically compiling and evaluating asm.js code?

ng and evaluating asm.js code actually possible? Thanks. Stéphane Letz -- You received this message because you are subscribed to the Google Groups "emscripten-discuss" group. To unsubscribe from this group and stop receiving emails from it, send an email to emscripten-discuss+unsu

Example of asm.js code to start with?

Hi, We are interested to directly generate asm.js code from our audio DSP language Faust (http://faust.grame.fr). Is there any simple example of asm.js code/module we could look at to start with? Thanks in advance. Stephane Letz -- You received this message because you are subscribed to the

What about compiler for SIMD.js ?

See : https://software.intel.com/sites/billboard/article/simd-javascript-faster-html5-apps?utm_source=yesmail&utm_medium=article&utm_content=Ealerts&mmid=1098992&utm_campaign=06052014-Parallelizing%20the%20Web-EMEA&uid=155575 Stéphane Letz -- You received this mes

Re: Emscripten fastcomp backend interesting for Faust

d support almost anything that clang will > emit into LLVM IR, minus things like threads. > > - Alon > > > > On Tue, Mar 11, 2014 at 8:50 AM, Stéphane Letz wrote: > I've recently read the thread on LLVM list about Emscripten fastcomp backend > and I am very inte

Emscripten fastcomp backend interesting for Faust

VM IR (… some IR==> IR optimisation passes…) ==> Emscripten fast comp ==> asm.js Would this work? Thanks. Stéphane Letz -- You received this message because you are subscribed to the Google Groups "emscripten-discuss" group. To unsubscribe from this group and stop receiv