In bug 912121, I am moving all DevTools files up to a top-level /devtools directory. This has led to the following situation for XPCShell on B2G Emulator ICS debug: # Overview * Emulator debug XPCShell tests are split into 4 chunks * Tests are sorted alphabetically by source tree path before dividing into chunks # Current State (before bug 912121 changes) * Example run: https://treeherder.mozilla.org/#/jobs?repo=fx-team&revision=30fffdc35b41&filter-searchStr=ics%20xpcshell * Chunk 1 starts the RIL tests up to test_ril_worker_cw.js, then the chunk ends * Chunk 2 runs the remaining RIL tests # Future State (after bug 912121 changes) * Example run: https://treeherder.mozilla.org/#/jobs?repo=try&revision=052b34bc7034&filter-searchStr=ics%20xpcshell * All DevTools tests are now before RIL in alphabetic sort * Chunk 1 starts the RIL tests up to test_ril_worker_icc_ICCPDUHelper.js, then the chunk ends * Chunk 2 tries to run remaining RIL tests, but test_ril_worker_cellbroadcast.js now permafails * Bug 912121 caused 16 RIL tests to move from chunk 1 to chunk 2 # Solutions I tried disabling the RIL tests, but other tests continue to permafail in chunk 2. In order to land bug 912121 for now, I have disabled the following directories on Gonk: devtools/shared/apps devtools/shared/qrcode devtools/shared/security devtools/shared/shared devtools/shared/sourcemap We should investigate a better solution so we can re-enable the DevTools tests here.

Keep tracking

Hi Edgar, According to comment 0, there was an unknown issue causing "devtools" tests failures on emu-ics (debug), and this bug is aiming at fixing the issue to make the devtools tests green and re-enabled. Per our chat, I learned that you set "see also bug 1234972" was because you were suspecting that if bug 1234972 crashed due to the same (unknown) issue reported on comment 0, and if we would consider to disable the test on 1234972 first according to the workaround mentioned on comment 0. However, I still don't get the strong enough connection that why this bug blocks the x86-kk meta bug 1192135. Could you please check again or explain more? Thank you!

(In reply to Hsin-Yi Tsai [:hsinyi] from comment #2) > Hi Edgar, > > According to comment 0, there was an unknown issue causing "devtools" tests > failures on emu-ics (debug), and this bug is aiming at fixing the issue to > make the devtools tests green and re-enabled. > > Per our chat, I learned that you set "see also bug 1234972" was because you > were suspecting that if bug 1234972 crashed due to the same (unknown) issue > reported on comment 0, and if we would consider to disable the test on > 1234972 first according to the workaround mentioned on comment 0. > > However, I still don't get the strong enough connection that why this bug > blocks the x86-kk meta bug 1192135. Could you please check again or explain > more? Thank you! In addition to bug 1234972, I saw some other failures happening in devtools tests as well. My intention was using this bug to track all devtools tests. But you are right, it is not clear enough as this bug is filed for "B2G Emulator Debug" (tests for x86 KK are currently running on OPT build only). I will check status again, then file separated bug for each of the failures or link a proper existing bug instead. Thank you.

Hi Ken, May I understand why this is P1? After checking with Edgar, according to the clues we have, there's no strong enough reason to show this blocks emu-x86-kk.

I thought this bug blocked bug 1192135. I overlook this bug was removed in comment 3.