Well, that's a temporary bug, but true...

- Alon



On Wed, Mar 5, 2014 at 2:01 PM, Chad Austin <[email protected]> wrote:

> For temporary debugging purposes, it's probably better if it _does_ break
> validation, because asm.js code doesn't show in the call stack, making
> tracing the failure hard. :)
>
> Sent from my iPhone
>
> On Mar 5, 2014, at 1:31 PM, Alon Zakai <[email protected]> wrote:
>
> I implemented emscripten_debugger() with one function call, so it is
> closer. Emitting it directly in the code will take some checking to make
> sure it does not break asm validation, I'll try to do that later.
>
> - Alon
>
>
>
> On Wed, Mar 5, 2014 at 11:49 AM, Chad Austin <[email protected]> wrote:
>
>> On Wed, Mar 5, 2014 at 11:22 AM, Alon Zakai <[email protected]> wrote:
>>
>>> asm() has some odd behavior of how escaping and so forth works - it's
>>> designed for typical inline assembly, not JS. EM_ASM is in practice easier
>>> to use for that reason.
>>>
>>
>> Ah, interesting.
>>
>> Also, expanding the code inline can break asm.js validation - it's
>>> arbitrary JS - so we expand it outline.
>>>
>>
>>> I guess there might be a use case for actually expanding inline though,
>>> for things like `debugger;`. But, perhaps we should just add an intrinsic
>>> emscripten_debugger() that you would call. Is there anything aside from
>>> `debugger;` like this?
>>>
>>
>> debugger; is the only use case I've had for asm() so far!  I like the
>> idea of emscripten_debugger();
>>
>>
>>>
>>> - Alon
>>>
>>>
>>>
>>> On Tue, Mar 4, 2014 at 11:55 PM, Chad Austin <[email protected]> wrote:
>>>
>>>> I'm playing with fastcomp to see what it would take to get a few embind
>>>> demos running in asm.js.
>>>>
>>>> It was pretty convenient to put asm("debugger"); in Emscripten output,
>>>> but now asm() doesn't work.  You have to use EM_ASM, which breaks into the
>>>> debugger a couple entries down in the stack, which is just a bit less
>>>> convenient.
>>>>
>>>> Is there a particular reason why?
>>>>
>>>> --
>>>> Chad Austin
>>>> Technical Director, IMVU
>>>> http://engineering.imvu.com <http://www.imvu.com/members/Chad/>
>>>> http://chadaustin.me
>>>>
>>>>
>>>>  --
>>>> 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 [email protected].
>>>> For more options, visit https://groups.google.com/groups/opt_out.
>>>>
>>>
>>>  --
>>> 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 [email protected].
>>> For more options, visit https://groups.google.com/groups/opt_out.
>>>
>>
>>
>>
>> --
>> Chad Austin
>> Technical Director, IMVU
>> http://engineering.imvu.com <http://www.imvu.com/members/Chad/>
>> http://chadaustin.me
>>
>>
>>  --
>> 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 [email protected].
>> For more options, visit https://groups.google.com/groups/opt_out.
>>
>
>  --
> 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 [email protected].
> For more options, visit https://groups.google.com/groups/opt_out.
>
>  --
> 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 [email protected].
> For more options, visit https://groups.google.com/groups/opt_out.
>

-- 
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 [email protected].
For more options, visit https://groups.google.com/groups/opt_out.

Reply via email to