Bug 793111 introduced PLMN entries reading, taking as argument the number of entries to read. According to the current codebase, I can see the argument is tlvLen/6. As far as I remember, I had the same tlvLen value, but did divide it by 3. I see no justification but the /6, and running tests on my system (Free Mobile SIM card, Nexus S), I get: with /6: $ adb logcat | egrep -i "readPLMN" I/Gecko ( 77): RIL Worker: readPLMNEntries: PLMN entries length = 1 I/Gecko ( 77): RIL Worker: readPLMNEntries: Reading PLMN entry: [0]: '2,248,16' I/Gecko ( 77): RIL Worker: readPLMNEntries: PLMN = 208, 1 with /3: $ adb logcat | egrep -i "readPLMN" I/Gecko ( 77): RIL Worker: readPLMNEntries: PLMN entries length = 2 I/Gecko ( 77): RIL Worker: readPLMNEntries: Reading PLMN entry: [0]: '2,248,16' I/Gecko ( 77): RIL Worker: readPLMNEntries: PLMN = 208, 1 I/Gecko ( 77): RIL Worker: readPLMNEntries: Reading PLMN entry: [1]: '2,248,32' I/Gecko ( 77): RIL Worker: readPLMNEntries: PLMN = 208, 2 with /2: $ adb logcat | egrep -i "readPLMN" I/Gecko ( 77): RIL Worker: readPLMNEntries: PLMN entries length = 2 I/Gecko ( 77): RIL Worker: readPLMNEntries: Reading PLMN entry: [0]: '2,248,16' I/Gecko ( 77): RIL Worker: readPLMNEntries: PLMN = 208, 1 I/Gecko ( 77): RIL Worker: readPLMNEntries: Reading PLMN entry: [1]: '2,248,32' I/Gecko ( 77): RIL Worker: readPLMNEntries: PLMN = 208, 2 I/Gecko ( 77): RIL Worker: readPLMNEntries: Reading PLMN entry: [2]: '255,255,255' Do you have spec/link suggesting/confirming the /6 value ? All I can find is ETSI spec, TS 131.102, section 4.2.66 which states: "Note: the length is 3*n bytes, where n denotes the number of PLMN entries. The length can be coded on one or more bytes."

And for your information, 20801 and 20802 are other valid assigned PLMN for Orange network in France, which Free is roaming on.

Works well on Free Mobile USIM card, and should comply to legacy SIM specifications (but no way for me to test).

20801 is main Orange network identifier, and 20802 is a special one for zones with low coverage and specific multi-operator network to cover at lower cost.

Comment on attachment 690208 [details] [diff] [review] Trivial fix Review of attachment 690208 [details] [diff] [review]: ----------------------------------------------------------------- I think you're right, and we'll still need more test case for this. Please don't check-in now.

Adding EF_SPDI file, which contains an SPDI TLV (tag: A3) and an Service Provider PLMN List TLV that contains 2 entries: 234136 and 46692.

Comment on attachment 691754 [details] Pull request: change in emulator Please changes comments of EF_SPDI as well.

Merged fix from bug 819786 and bug 819784.

Comment on attachment 692191 [details] [diff] [review] Patch Review of attachment 692191 [details] [diff] [review]: ----------------------------------------------------------------- Great!

BB? for bug 819786 and bug 819784 are both bb+.

(In reply to Vicamo Yang [:vicamo][:vyang] from comment #12) > BB? for bug 819786 and bug 819784 are both bb+. Yep, this is a blocker for that reason.

https://hg.mozilla.org/releases/mozilla-aurora/rev/a3f1771e71c9 https://hg.mozilla.org/releases/mozilla-b2g18/rev/0d0c852fc724