Oh, good call! Took me a long time to get these right, and once I
figured it out, I was a bit panicked about touching them. :) In addition
to the stray \, I also found a missing ,. I also moved _main first, and
flattened everything onto a single line. One or both of those seems to
have resolved it.
Thanks!
On 08/27/2018 08:45 AM, Alon Zakai wrote:
My guess is that there may be a parsing bug here - we did refactor our
argument parser for things like that recently. In particular, the slash on
'_sqlite3_finalize',\
and the fact that _main is the last item in the list, are the things
that make me think it could be parsing-related.
Specifically, it could be related to
https://github.com/kripken/emscripten/issues/6581#issuecomment-412333543
which is the only known issue I am aware of there.
Also helpful to debug this might be to build with EMCC_DEBUG=1 in the
env, which will print out the arguments passed, and also the flags
passed to LLVM DCE (search for "main", or
"-internalize-public-api-list=").
On Mon, Aug 27, 2018 at 6:10 AM Nolan Darilek <[email protected]
<mailto:[email protected]>> wrote:
I alluded to this issue earlier. Upgrading *only emsdk-portable* from
1.37.36 to 1.38.11 and keeping everything else the same results in a
wasm file that won't start. Further, its size is ~2.4M whereas the
size
under 1.37.36 is just over 10M. I'm using Rust nightly, cargo-web and
the wasm32-unknown-emscripten target. All needed dependencies are
rebuilt under 1.38.11 before rebuilding the Rust binary. _main is
definitely specified in the list of exported functions. Here are my
linker arguments as passed to rustc:
link-args = [
"-s", "ASSERTIONS=1",
"-L", "vendor/prefix/lib",
"-l", "proj",
"-l", "geos_c",
"-l", "geos",
"-l", "z",
"-s", "EXTRA_EXPORTED_RUNTIME_METHODS=['ccall','cwrap']",
"-s", """EXPORTED_FUNCTIONS=[
'_spatialite_version',
'_spatialite_alloc_connection'
'_spatialite_init_ex',
'_spatialite_cleanup_ex',
'_load_shapefile_ex',
'_malloc',
'_free',
'_sqlite3_open',
'_sqlite3_exec',
'_sqlite3_free',
'_sqlite3_errmsg',
'_sqlite3_prepare_v2',
'_sqlite3_bind_text',
'_sqlite3_bind_blob',
'_sqlite3_bind_double',
'_sqlite3_bind_int',
'_sqlite3_bind_parameter_index',
'_sqlite3_step',
'_sqlite3_data_count',
'_sqlite3_column_double',
'_sqlite3_column_text',
'_sqlite3_column_blob',
'_sqlite3_column_bytes',
'_sqlite3_column_type',
'_sqlite3_column_name',
'_sqlite3_reset',
'_sqlite3_clear_bindings',
'_sqlite3_finalize',\
'_sqlite3_close_v2',
'_main'
]"""
]
Is there any known issue between 1.37.36 and 1.38.11 that might
have an
effect here? How do I even go about debugging this? I have an
older copy
of the intermediate Rust artifacts, including the final .wasm file.
Thanks for any 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]
<mailto:emscripten-discuss%[email protected]>.
For more options, visit https://groups.google.com/d/optout.
--
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]
<mailto:[email protected]>.
For more options, visit https://groups.google.com/d/optout.
--
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/d/optout.
