Thanks! I think I see what's wrong. This PR should fix it, please test if
you can,

https://github.com/emscripten-core/emscripten/pull/9880

- Alon

On Wed, Nov 20, 2019 at 7:29 AM Eric Mandel <[email protected]> wrote:

> I (laboriously) added print statements to find the library routine that
> eventually fails and it looks like the non-asm Emscripten might indeed be
> trying to call a function where there is no function. Here is a snippet of
> the code that is being run, indicating where print statements are located:
>
> fprintf(stdout, "2\n"); fflush(stdout);
>
>     if (*status == NOT_IMAGE)
>         *status = tstatus;    /* ignore 'unknown extension type' error */
>     else if (*status > 0)
>         return(*status);
>
> fprintf(stdout, "3\n"); fflush(stdout);
>
>     /*
>        the logical end of the header is 80 bytes before the current
> position,
>        minus any trailing blank keywords just before the END keyword.
>     */
>     (fptr->Fptr)->headend = (fptr->Fptr)->nextkey - (80 * (nspace + 1));
>
> fprintf(stdout, "4 [%d %d]\n", (int)(fptr->Fptr)->headend,
> (int)(fptr->Fptr)->nextkey); fflush(stdout);
>
>     /* the data unit begins at the beginning of the next logical block */
>     (fptr->Fptr)->datastart = (((fptr->Fptr)->nextkey - 80) / 2880 + 1)
>                               * 2880;
>
> fprintf(stdout, "5\n"); fflush(stdout);
>
> and here is the output in the Chrome debugger for the non-wasm case:
>
> [image: Screen Shot 2019-11-20 at 10.13.49 AM.png]
>
> The values of 400 and 480 displayed in print statement #4 are correct for
> this astronomical image, so things look right at that point. But the next
> line of code should simply set datastart to the first 2880 byte block after
> the image header, and it looks like it might be calling a non-existent
> routine instead. So we never reach print statement #5.
>
> Note that headend, nextkey, and datastart are all declared as long long,
> if that is any hint.
>
> For comparison, here is the Chrome debugger output from the wasm build:
>
> [image: Screen Shot 2019-11-20 at 10.24.35 AM.png]
>
>
> Let me know how I can help.
>
> --
> 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/39322779-c9c5-4fbc-9231-203200ffe086%40googlegroups.com
> <https://groups.google.com/d/msgid/emscripten-discuss/39322779-c9c5-4fbc-9231-203200ffe086%40googlegroups.com?utm_medium=email&utm_source=footer>
> .
>

-- 
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/CAEX4NpR0RO_dZr%3Dpb7uE6JHzZX_81iFtjXMU9aKRCCv%3DxNgG_g%40mail.gmail.com.

Reply via email to