726233 - [meta] B2G RIL: akami support

Status: RESOLVED FIXED

(Firefox OS Graveyard :: RIL, defect)

Description

[[:fabrice] Fabrice Desré]

Here's the startup log when turning DEBUG to true in the ril worker.

Nothing more from the RIL afterward.

I/Gecko   (  275): RIL Worker: Received 16 bytes.
I/Gecko   (  275): RIL Worker: Already read 0
I/Gecko   (  275): RIL Worker: New incoming parcel of size 12
I/Gecko   (  275): RIL Worker: Parcel (size 12): 1,0,0,0,232,3,0,0,1,0,0,0
I/Gecko   (  275): RIL Worker: We have at least one complete parcel.
I/Gecko   (  275): RIL Worker: Unsolicited response for request type 1000
I/Gecko   (  275): RIL Worker: Handling parcel as UNSOLICITED_RESPONSE_RADIO_STATE_CHANGED
I/Gecko   (  275): RIL Worker: Radio state changed from 1 to 1
I/Gecko   (  275): RIL Worker: Next parcel size unknown, going to sleep.

Comment 1

[Kyle Machulis [:qdot] [:kmachulis] (INACTIVE)]

I can't seem to repro this. I've got a SIM card in and am seeing status updates about radio strength just fine (as in it's actually showing strength updates not just 99's.)

Comment 2

[Kyle Machulis [:qdot] [:kmachulis] (INACTIVE)]

Ok something is definitely up at least. I can't get SMS messages thru to it.

Comment 3

[Kyle Machulis [:qdot] [:kmachulis] (INACTIVE)]

It doesn't look like we're registering on the network correctly.

I/Gecko   (   82): RIL Worker: New incoming parcel of size 28
I/Gecko   (   82): RIL Worker: Parcel (size 28): 0,0,0,0,67,1,0,0,0,0,0,0,3,0,0,0,255,255,255,255,255,255,255,255,255,255,255,255
I/Gecko   (   82): RIL Worker: We have at least one complete parcel.
I/Gecko   (   82): RIL Worker: Solicited response for request type 22, token 323
I/Gecko   (   82): RIL Worker: Handling parcel as REQUEST_OPERATOR
I/Gecko   (   82): RIL Worker: Next parcel size unknown, going to sleep.

That operator message should be showing us AT&T (for the sim card I have in). However that 0x3 15 bytes down is signifying an error. So this is two bugs: We don't properly pick up operator errors, and the akami doesn't work.

Comment 4

[Philipp von Weitershausen [:philikon]]

I don't think that's right, Kyle. Deconstructing your parcel from above:

0,0,0,0,    // RESPONSE_TYPE_SOLICITED
67,1,0,0,   // token
0,0,0,0,    // error (none)
3,0,0,0,    // string list with 3 elements
255,255,255,255,255,255,255,255,255,255,255,255 // WTF?!?

Something's definitely fishy here, and the problem seems to be with akami's RIL. Any other clues from the logcat?

Comment 5

[Kyle Machulis [:qdot] [:kmachulis] (INACTIVE)]

Looks like there may be firmware issues out of our control. Leaving open until we have a resolution, but not expecting patches landing via this.

Comment 6

[Michael Vines [:m1] [:evilmachines]]

This is almost certainly due to the modem being configured for DSDS and Gecko is not initializing the DSDS ril correctly.  Next week I can tweak the modem back to single SIM mode and this issue should be resolved.  I've done this for my Akami and am able to make calls.

Comment 7

[Philipp von Weitershausen [:philikon]]

Seems like the big issues are resolved with the modem tweaks Michael mentioned. Morphing this bug to track any outstanding RIL issues on the akami.

Comment 8

[JP Rosevear [:jpr]]

Blocks a blocker and depends on a blocker.

Comment 9

[JP Rosevear [:jpr]]

Nothing Akami blocks basecamp now.

Comment 10

[Dietrich Ayala (:dietrich)]

Removing product phone metabug dependency.

Comment 11

[Hsin-Yi Tsai (she/her) [:hsinyi]]

Close as all dependencies fixed.