[03:54] *** GLaDOS has quit IRC (Quit: Oh crap, I died.) [05:12] *** balrog has quit IRC (Ping timeout: 260 seconds) [06:15] *** balrog has joined #jsmess [20:23] WebAssembly.compile(): Result = expected version 0b 00 00 00, found 0c 00 00 00 @+4 [20:23] derp [20:24] Should we yank azakai into this channel? Would it help? [20:24] We're obviously the biggest worst thing they're going to shove through it for a while [20:24] let me poke a bit first [20:24] It's like a brand new day, like we had with the first days of JSMESS [20:24] "What do you mean you have an entire mixing system in aaaaaaaaaaaaaaaaaaa" [20:25] hah, then in firefox... [20:25] wasm validation error: at offset 8: binary version 0xc does not match expected version 0xd [20:26] so chrome wants 0xb, the compiler outputs 0xc, and firefox wants 0xd [20:52] It might be good to bring him in if only that as he finds things are fucko, he might be able to fix emscripten [20:53] yeah, I'm gonna drop into #emscripten when I'm off this call [21:04] Also, it really seems like MAME development has heated up [21:05] Might be reflective of some changes of the code structure, though [21:29] graphs don't seem too dramatic https://github.com/mamedev/mame/graphs/contributors [21:34] *** azakai has joined #jsmess [22:16] *** db48x has joined #jsmess [22:20] hey azakai [22:21] been looking into the wasm builds of mame. I got it to build by disabling -Werror, but now I'm having problems loading in both ff and chrome [22:21] chrome says: WebAssembly.compile(): Result = expected version 0b 00 00 00, found 0c 00 00 00 @+4 [22:21] ff nightly says: wasm validation error: at offset 8: binary version 0xc does not match expected version 0xd [22:22] sounds like chrome and ff are using different versions of the binary format, but neither of them match what emscripten incoming branch is incoming? [22:22] is generating* [22:38] *** SketchCow sets mode: +o azakai [22:45] *** azakai_ has joined #jsmess [22:45] *** azakai_ has left [23:28] bai: that looks like an older browser version for chrome, it should expect 0xd. as for firefox, that it sees 0xc suggests it's using an older binaryen - current emscripten incoming by default should use 0xd, unless you built older binaryen yourself? [23:33] bai: Is there a way you're doing this so it's on a browsable link? [23:33] Although you're all in the same geographical time zone\ [23:51] azakai: yeah this is with the current release version of chrome, I guess I should be using...canary? [23:51] but yeah, http://baicoianu.com/~bai/webassembly/pacman.html is the url I'm testing with right now [23:52] azakai: I built a fresh version of emscripten from incoming two days ago. I had to manually build binaryen but I did so using the version that the build process itself checked out [23:53] it was trying to build it with "emmake make" when it should have just been running "make" [23:57] Uncaught ReferenceError: WebAssembly is not defined at XMLHttpRequest. (pacman.html:10) [23:57] yeah, you need to enable it in your browser, and be running a dev build [23:57] bai: yeah, need canary i think [23:58] bai: it's possible the binaryen update to 0xd was in the last 2 days. it was definitely in the last 3 [23:58] ah ok, I'll rebuild [23:58] trying again, with enabled [23:59] VM29 pacman.html:10 Uncaught (in promise) CompileError: WebAssembly.compile(): Wasm decoding failedResult = expected version 0d 00 00 00, found 0c 00 00 00 @+4 [23:59] at XMLHttpRequest. (http://baicoianu.com/~bai/webassembly/pacman.html:10:30) [23:59] (As expected) [23:59] OK, I'm prepped here on the same general setup for comparison. [23:59] cool, yeah in chrome dev branch, I get the compiled output