Sam Clegg

I'm not aware of any users, but I have maintained it over time. If nobody is using it, it would be great to remove it. If folks are using it...

``` +/report_result?exception: Maximum call stack size exceeded / RangeError: Maximum call stack size exceeded at test.wasm.operator_new_impl(unsigned long) (http://localhost:8888/test.wasm:wasm-function[142]:0x68bf) at test.wasm.operator new(unsigned long) (http://localhost:8888/test.wasm:wasm-function[141]:0x68b0) at test.wasm.(anonymous namespace)::OPFSBackend::createDirectory(unsigned int) (http://localhost:8888/test.wasm:wasm-function[297]:0x8848) at test.wasm.doMkdir(wasmfs::path::P[..]...

Hmm.. I as able to repro the other day, and I've seen in in the CI.. but I'm not sure if it consistent. I'll try again now..

I'm seeing this issue on some CI runs. e.g.: https://app.circleci.com/pipelines/github/emscripten-core/emscripten/34717/workflows/01f23013-b724-4ec3-81b3-33a49727d0ab/jobs/775610 ``` ====================================================================== FAIL [3.019s]: test_wasmfs_opfs_jspi (test_browser.browser_2gb) ---------------------------------------------------------------------- Traceback (most recent call last): File "/root/project/test/common.py", line 2080, in run_browser self.assertContained(expected, output)...

Currently we have 41 uses of `withStackSave` in the codebase. It looks like about 1/2 of them are trivially replaceable with a pair of `stackSave`/`stackRestore` calls.

I'm not a cmake expert either I'm afraid but it looks like this should be tested as part of `test_cmake_find_modules`.

I'm in favor of this change too.. just waiting on a test.

It looks like the `./boost/config/requires_threads.hpp` header in boost requires threading support. In fact that whole point of that header seems to be to error out when threads are not supported,...

Regarding that compile issue it looks like `platform.hpp` would need to be configured somehow to not include `requires_threads.hpp`. I assume that is possible, otherwise `BOOST_DISABLE_THREADS` wouldn't be a thing? Regarding...

Sorry yes, I believe that warning is no longer valid and we should allow `-sEXPORTED_FUNCTIONS` in this case.