Thanks for the reply JJ. So, it seems like the link's not implemented yet
then.
I tried hacking my own in ealier and it failed but I gave it another go
just now,
hacking in this line in my generated 'anthraciteWASM.js' file and lo and
behold it kinda worked. The buffer's being copied alright: it looks a
little off but I think that might be down to other small discrepancies
between my desktop build and WASM.
Progress though: I've got some full screen effects now. Is there any chance
that somebody can implement the function in the library as below?
function _glBlitFramebuffer(a,b,c,d,e,f,g,h,i,j) {
GLctx["blitFramebuffer"](a,b,c,d,e,f,g,h,i,j);
}
On Tuesday, 11 February 2020 18:26:16 UTC, jj wrote:
>
> I think Alon was looking too quickly: glBlitFramebuffer is
> implemented, but it is one of those trivial "pass-through" functions
> (functions that have only primitive numbers as args) so you'll find it
> by searching for 'blitframebuffer' rather than 'glblitframebuffer'.
> Its implementation is done on this line:
>
>
> https://github.com/emscripten-core/emscripten/blob/7b2d3d0a3fc0d5238d5dee0371de7cc24d3542a7/src/library_webgl2.js#L1015
>
>
> There is one test in the test suite that does touch glBlitframebuffer,
> see
> https://github.com/emscripten-core/emscripten/blob/7b2d3d0a3fc0d5238d5dee0371de7cc24d3542a7/tests/webgl_with_closure.cpp#L176
>
> .
>
> See
> https://github.com/emscripten-core/emscripten/blob/7b2d3d0a3fc0d5238d5dee0371de7cc24d3542a7/tests/test_browser.py#L2659
>
> for the test. It can be run from command line with "python
> tests/runner.py browser.test_webgl_with_closure"
>
> The flag -s USE_WEBGL2=1 is an alias/synonym to -s
> MAX_WEBGL_VERSION=2. (passing either one has identical behavior)
>
> -s MAX_WEBGL_VERSION=2 means that the generated build supports
> creating WebGL 2 contexts (in addition to supporting WebGL 1
> contexts). Some of the emulated libraries (SDL,GLUT,GLFW,GLEW,...
> emulation stubs in src/library_x.js) will default to creating WebGL 2
> contexts if -s MAX_WEBGL_VERSION=2 is passed. The html5.h API WebGL
> context creation function has a version field that is set to specify
> the desired context versoin.
>
> Passing -s MIN_WEBGL_VERSION=2 drops support for WebGL 1 altogether,
> to get a size micro-optimization for the build output. This is
> intended for pages that know they cannot operate with WebGL 1, and
> wish not to have to pay code size for generated WebGL 1 support
> functions.
>
> Try passing -lwebgl2.js explicitly on the link line if there is any
> oddness. Running with environment variable EMCC_DEBUG=1 and linker
> flag -s VERBOSE=1 may help troubleshooting some linking decisions that
> are being made.
>
> ti 11. helmik. 2020 klo 19.42 Alon Zakai ([email protected] <javascript:>)
> kirjoitti:
> >
> > Looking in src/*, it seems like that function is not implemented yet -
> it needs to be added there. A PR would be very welcome if someone is
> interested! It should be in src/library_webgl2.js.
> >
> > Linking with -s MAX_WEBGL_VERSION=2 should have been enough for this to
> work, if the function were there (that says "the max version that will be
> used is 2, so anything between the min version - default 1 - and the max is
> supported; the default max is 1, so raising the max is necessary).
> >
> > On Mon, Feb 10, 2020 at 4:07 PM Tim Luther Lewis <[email protected]
> <javascript:>> wrote:
> >>
> >> I tried that, although I hadn't thought of linking libraries directly
> like that (emscripten seems to do it's own thing with '-s USE_WEBGL2=1',
> which probably resolves to some library using internal magic that I haven't
> deep dived on yet)
> >>
> >> In my generated WASM.js file, I still see this:
> >>
> >> function _glBlitFramebuffer(
> >> ) {
> >> err('missing function: glBlitFramebuffer'); abort(-1);
> >> }
> >>
> >> Which is clearly why it's not working: it's not there! Why it's not
> there though, is still a mystery to me :/
> >>
> >>
> >> On Monday, 10 February 2020 20:23:17 UTC, Charles Vaughn wrote:
> >>>
> >>> Looks like you're missing a linker input, do you have -lGLESv2 on your
> link command?
> >>>
> >>> On Monday, February 10, 2020 at 1:01:11 AM UTC-8, Tim Luther Lewis
> wrote:
> >>>>
> >>>> Thanks for the responses guys but unfortunately, I'm still getting
> 'anthraciteWASM.js:8995 missing function: glBlitFramebuffer'
> >>>>
> >>>> I've added both '-s MIN_WEBGL_VERSION=2' and '-s MAX_WEBGL_VERSION=2'
> and I'm creating my context with 'const context =
> canvas.getContext("webgl2", {antialias: false, depth: true});'
> >>>>
> >>>> I do get a bunch of linker warnings when I build, however:
> >>>>
> >>>> ---8<-------------------------------
> >>>> warning: undefined symbol: emscripten_glBeginQuery
> >>>> warning: undefined symbol: emscripten_glBeginTransformFeedback
> >>>> warning: undefined symbol: emscripten_glBindBufferBase
> >>>> warning: undefined symbol: emscripten_glBindBufferRange
> >>>> warning: undefined symbol: emscripten_glBindSampler
> >>>> warning: undefined symbol: emscripten_glBindTransformFeedback
> >>>> warning: undefined symbol: emscripten_glBlitFramebuffer
> >>>> warning: undefined symbol: emscripten_glClearBufferfi
> >>>> warning: undefined symbol: emscripten_glClearBufferfv
> >>>> warning: undefined symbol: emscripten_glClearBufferiv
> >>>> warning: undefined symbol: emscripten_glClearBufferuiv
> >>>> warning: undefined symbol: emscripten_glClientWaitSync
> >>>> warning: undefined symbol: emscripten_glCompressedTexImage3D
> >>>> ...
> >>>> --8<--------------------------------------------
> >>>> but seeing as everything ran fine, other than the blit, I assumed
> this was just some sort of late binding issue for external functions.
> >>>>
> >>>> Is it possible that there's a problem with glBlitFramebuffer that
> hasn't been picked up?
> >>>>
> >>>>
> >>>>
> >>>> On Sunday, 9 February 2020 20:31:49 UTC, Tim Luther Lewis wrote:
> >>>>>
> >>>>> Hi,
> >>>>> I'm porting my engine to WASM and it's all going well so far but
> I've hit a snag with this. I'm linking with -s FULL_ES2=1 -s FULL_ES3=1
> but it still gives me 'missing function: glBlitFramebuffer' whenever I call
> it.
> >>>>> I noticed there was a discussion about this function on github back
> in February 2019 (https://github.com/emscripten-core/emscripten/pull/7588).
> Does anybody know if it's broken or perhaps I've missed something vital?
> >>>>>
> >>>>> Thanks in advance,
> >>>>> Tim.
> >>>>>
> >> --
> >> 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] <javascript:>.
>
> >> To view this discussion on the web visit
> https://groups.google.com/d/msgid/emscripten-discuss/9c7adf13-64e9-4993-8f87-f6f669a5b787%40googlegroups.com.
>
>
> >
> > --
> > 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] <javascript:>.
>
> > To view this discussion on the web visit
> https://groups.google.com/d/msgid/emscripten-discuss/CAEX4NpT9mrc%3DczjFbhVBaG1fdChUn7WKnLnHEa9EBOg_LH6nbA%40mail.gmail.com.
>
>
>
--
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].
To view this discussion on the web visit
https://groups.google.com/d/msgid/emscripten-discuss/0db7c526-95c7-4619-b366-8912ae71f1b1%40googlegroups.com.
