By checking the code in [1], gecko did not handle wild char properly while reading EF_OPL. It will cause wrong operator name display if there is any wild char pre-allocated in SIM EF_OPL. // 3GPP TS 31.102, clause 4.2.59 EFOPL (Operator PLMN List) ... Coding: PLMN : according to TS 24.008 [9]/TS 24.301 [51] A BCD value of 'D' in any of the MCC and/or MNC digits shall be used to indicate a "wild" value for that corresponding MCC/MNC digit [1] - http://dxr.mozilla.org/mozilla-central/source/dom/system/gonk/ril_worker.js?from=ril_worker.js&case=true#14190

Bug 1046649 was filed for 1.3t/1.4 only due to time-to-market. Bug 1051664 is filed to re-consider the solution/test cases on m-c.

(In reply to shawn ku [:sku] from comment #1) > Bug 1046649 was filed for 1.3t/1.4 only due to time-to-market. Bug 1051664 > is filed to re-consider the solution/test cases on m-c. According to the landing rule [1], we should land patches to master first then uplift to 1.4 branch, the landing process will be done in bug 1046649. This bug now focus on adding marionette test for EF_OPL wild char. [1] https://wiki.mozilla.org/Release_Management/B2G_Landing

Comment on attachment 8475062 [details] Bug 1051664 Part1: qemu patch - B2G RIL: Add marionette tests for EF_OPL wild char. Hi Vicamo: Can I have your time to review this PR? Thanks!! Shawn

Comment on attachment 8475062 [details] Bug 1051664 Part1: qemu patch - B2G RIL: Add marionette tests for EF_OPL wild char. Hi, we have all mcc set to 001 because it's explicitly listed in Test Networks in wikipedia. Since Gaia have its own operator variants db, I'd like to avoid possible collisions if that doesn't cost much effort. And surely it doesn't. You may simply set OPL item 5 to be effective only with mcc=001, mnc=2dd, and OPL item 6 with mcc=001, mnc=20d. Thank you. [1]: http://en.wikipedia.org/wiki/Mobile_country_code#Test_networks

(In reply to Vicamo Yang [:vicamo][:vyang] from comment #6) > Comment on attachment 8475062 [details] > Bug 1051664 Part1: qemu patch - B2G RIL: Add marionette tests for EF_OPL > wild char. > > Hi, we have all mcc set to 001 because it's explicitly listed in Test > Networks in wikipedia. Since Gaia have its own operator variants db, I'd > like to avoid possible collisions if that doesn't cost much effort. And > surely it doesn't. You may simply set OPL item 5 to be effective only with > mcc=001, mnc=2dd, and OPL item 6 with mcc=001, mnc=20d. Thank you. > > [1]: http://en.wikipedia.org/wiki/Mobile_country_code#Test_networks Thanks for your time!! Can I simply change it to: // MCC = 001, MNC = 2D, START=0000, END=FFFE, PNN = 05, // MCC = 001, MNC = 3DD, START=0000, END=FFFE, PNN = 06, because 20d is a subset of 2dd, 001-20d will always be treated as the same as 001-2dd.

(In reply to shawn ku [:sku] from comment #7) > Thanks for your time!! > > Can I simply change it to: > // MCC = 001, MNC = 2D, START=0000, END=FFFE, PNN = 05, > // MCC = 001, MNC = 3DD, START=0000, END=FFFE, PNN = 06, > because 20d is a subset of 2dd, 001-20d will always be treated as the same > as 001-2dd. Sure. That was one of questions in my mind when I first saw your pull request. I thought that was intended so that you may test some kind of fallback algorithm.

Comment on attachment 8475776 [details] [diff] [review] Bug 1051664 Part2 : test patch - B2G RIL: Add marionette tests for EF_OPL wild char. Review of attachment 8475776 [details] [diff] [review]: ----------------------------------------------------------------- Hi Edgar: May I have your time to review? Thanks!! Shawn

Comment on attachment 8475776 [details] [diff] [review] Bug 1051664 Part2 : test patch - B2G RIL: Add marionette tests for EF_OPL wild char. Review of attachment 8475776 [details] [diff] [review]: ----------------------------------------------------------------- Looks good, thank you.

Hi Vicamo: Could you please help push the PR for me? I would like to run the full marionette test before submit gecko patch. Thanks!! Local result: ------- 2:45.85 LOG: MainThread INFO passed: 32 2:45.85 LOG: MainThread INFO failed: 0 2:45.85 LOG: MainThread INFO todo: 0 2:46.97 SUITE_END: MainThread Summary ======= Ran 32 tests Expected results: 32 Unexpected results: 0 OK

(In reply to shawn ku [:sku] from comment #17) > Hi Vicamo: > Could you please help push the PR for me? > I would like to run the full marionette test before submit gecko patch. You don't need this next time. You may edit '${gecko}/b2g/configs/emulator-ics/sources.xml', push it to tryserver as other ordinary patches do.

(In reply to Vicamo Yang [:vicamo][:vyang] from comment #19) > (In reply to shawn ku [:sku] from comment #17) > > Hi Vicamo: > > Could you please help push the PR for me? > > I would like to run the full marionette test before submit gecko patch. > > You don't need this next time. You may edit > '${gecko}/b2g/configs/emulator-ics/sources.xml', push it to tryserver as > other ordinary patches do. Thanks for your useful information. :-)

https://tbpl.mozilla.org/?tree=Try&rev=1e9c8245d6e4 try green