cljsrn 2018-07-18 | Slack Archive

j1mr10rd4n06:07:02

I’ve been banging my head on this one for a day or so now and wondered if anyone would have a moment to try and reproduce please… using re-natal 0.9.0 I generate an om-next application with: `re-natal init FutureApp -i om-next then

cd future-app
lein clean && re-natal use-android-device avd && re-natal use-figwheel && rlwrap lein figwheel android

wait for figwheel to start, start up an emulator, then in another terminal:

react-native start

the metro bundler starts and the FutureApp loads into the emulator figwheel bridge begins downloading the compiled cljs into FutureApp then FutureApp crashes and the emulator displays the home screen 💣💥:white_frowning_face: after doing some digging around in figwheel-bridge, this happens when it tries to eval() the compiled version of https://github.com/omcljs/om/blob/1.0.0-beta3/src/main/om/next.cljc other notes: * the compiled om/next.js file is 344Kb * the production-build of the app runs successfully in the android emulator * the ios figwheel build runs and hot-reloads successfully in the ios simulator * apps generated with the other frameworks (reagent and rum) run and hot-reload successfully in android and ios * the emulator logcat contains the following segfault trace after the app crashes

--------- beginning of crash
07-18 15:29:49.463  4639  4684 F libc    : Fatal signal 11 (SIGSEGV), code 1 (SEGV_MAPERR), fault addr 0x0 in tid 4684 (mqt_js), pid 4639 (futureappfulcro)
07-18 15:29:49.564  4799  4799 I crash_dump32: obtaining output fd from tombstoned, type: kDebuggerdTombstone
07-18 15:29:49.566  1712  1712 I /system/bin/tombstoned: received crash request for pid 4684
07-18 15:29:49.573  4799  4799 I crash_dump32: performing dump of process 4639 (target tid = 4684)
07-18 15:29:49.580  4799  4799 F DEBUG   : *** *** *** *** *** *** *** *** *** *** *** *** *** *** *** ***
07-18 15:29:49.580  4799  4799 F DEBUG   : Build fingerprint: 'google/sdk_gphone_x86/generic_x86:9/PPP4.180612.007/4860066:userdebug/dev-keys'
07-18 15:29:49.580  4799  4799 F DEBUG   : Revision: '0'
07-18 15:29:49.580  4799  4799 F DEBUG   : ABI: 'x86'
07-18 15:29:49.580  4799  4799 F DEBUG   : pid: 4639, tid: 4684, name: mqt_js  >>> com.futureappfulcro <<<
07-18 15:29:49.580  4799  4799 F DEBUG   : signal 11 (SIGSEGV), code 1 (SEGV_MAPERR), fault addr 0x0
07-18 15:29:49.580  4799  4799 F DEBUG   : Cause: null pointer dereference
07-18 15:29:49.580  4799  4799 F DEBUG   :     eax 00000000  ebx 7d666d9c  ecx ffffffff  edx 00000000
07-18 15:29:49.580  4799  4799 F DEBUG   :     edi 0001788d  esi 00000000
07-18 15:29:49.580  4799  4799 F DEBUG   :     ebp 7cd796a8  esp 7cd7961c  eip 00000000
07-18 15:29:49.628  1693  1693 I Zygote  : Process 4582 exited due to signal (9)
07-18 15:29:49.628  1831  3880 I ActivityManager: Process com.android.defcontainer (pid 4582) has died: cch  CEM

• the same behaviour happened on two different android emulators with API levels 26 and 28 • the android emulator instances were configured with 2GB RAM and 1GB VM heap • similar behaviour happens with an application configured to use fulcro-2.5.12 instead of om-next, this time when it gets to the compiled version of https://github.com/fulcrologic/fulcro/blob/2.5.12/src/main/fulcro/client/primitives.cljc (fulcro is derived from om-next and this file compiles to a hefty 449Kb)

j1mr10rd4n06:07:49

some versions from my platform: re-natal: 0.9.0; react-native-cli: 2.0.1; node 10.5.0; npm 6.1.0; lein 2.8.1 on Java 1.8.0_172 OpenJDK 64-Bit Server V; Android emulator version 27.3.8.0

2018-07-18

Channels