946165 - [B2G] [DSDS] [Fugu] Miss STK menu proactive on fugu

Status: RESOLVED FIXED

(Firefox OS Graveyard :: RIL, defect)

Description

[Edgar Chen [:edgar]]

Attachment: fugu_stk_ril.txt

+++ This bug was initially created as a clone of Bug #942096 +++

In bug 942096, we add iccId into 'icc-stkcommand' to support DSDS, and there is a timing between reading iccId and receiving stk menu proactive command from sim card, so we ignore the command if iccId is not ready yet [1]. It works well in unagi device, because unagi will keep generating stk menu proactive command if did not receive response. However, in Fugu the modem will not generate new command, but wait for response if did not receive it.

It seem if iccId is not ready yet, we should cache proactive commands and send then after iccId becomes ready.

[1] Please see bug 942096 comment #2.
[2] Please see bug 935802 comment #1.

Comment 1

[Edgar Chen [:edgar]]

Attachment: Patch, v1

1). Cache stk command if iccId is not ready.
2). When iccId becomes ready, send pending stk command if any.

I have tested this patch in unagi and fugu, both of them are worked well. ie, I can see STK entry in setting and also can browse the menu.

Comment 2

[Yoshi Cheng-Hao Huang [:allstars.chh][:allstarschh][:yoshi]]

Comment on attachment 8343500 [details] [diff] [review]
Patch, v1

Review of attachment 8343500 [details] [diff] [review]:
-----------------------------------------------------------------

Can we do it in ril_worker?

And modify the timing to call reportStkServiceIsRunning?

Comment 3

[Edgar Chen [:edgar]]

(In reply to Yoshi Huang[:allstars.chh][:yoshi] from comment #2)
> Comment on attachment 8343500 [details] [diff] [review]
> Patch, v1
> 
> Review of attachment 8343500 [details] [diff] [review]:
> -----------------------------------------------------------------
> 
> Can we do it in ril_worker?
> 
> And modify the timing to call reportStkServiceIsRunning?

Hmm, sounds better.
I will try this, thanks your suggestion.

Comment 4

[Edgar Chen [:edgar]]

Comment on attachment 8343500 [details] [diff] [review]
Patch, v1

Per comment #3, cancel review first.

Comment 5

[Edgar Chen [:edgar]]

(In reply to Edgar Chen [:edgar][:echen] from comment #3)
> (In reply to Yoshi Huang[:allstars.chh][:yoshi] from comment #2)
> > Comment on attachment 8343500 [details] [diff] [review]
> > Patch, v1
> > 
> > Review of attachment 8343500 [details] [diff] [review]:
> > -----------------------------------------------------------------
> > 
> > Can we do it in ril_worker?
> > 
> > And modify the timing to call reportStkServiceIsRunning?
> 
> Hmm, sounds better.
> I will try this, thanks your suggestion.

If we try to fix this by modifying the timing to call reportStkServiceIsRunning(), we need to consider below two cases,
1). cardState is ready, but iccId is not ready.
2). iccId is ready, but cardState is not ready. (ie, pin locked enabled case)

Comment 6

[Edgar Chen [:edgar]]

Attachment: Send stk service ready after cardState and iccid are both ready, v1

Another possible fix which is suggested in comment #2. 
I have tested this patch, it works well in {unagi|fugu} {with|without} sim pin lock enabled.

Comment 7

[Edgar Chen [:edgar]]

Attachment: Patch, v3

After discussing with Yoshi, we don't need to wait cardState becomes to 'ready'.
And I have tested unagi and fugu, both of them works well.

Comment 8

[Edgar Chen [:edgar]]

Comment on attachment 8343640 [details] [diff] [review]
Patch, v3

Oops, upload wrong patch.

Comment 9

[Edgar Chen [:edgar]]

Attachment: Patch, v3

Upload correct patch.
Please see comment #7 for details.

Comment 10

[Edgar Chen [:edgar]]

Attachment: Patch, v4, r=allstars.chh

Rebase only.

Comment 11

[Edgar Chen [:edgar]]

https://hg.mozilla.org/integration/b2g-inbound/rev/fe1ac343ac94

Comment 12

[Ryan VanderMeulen [:RyanVM]]

https://hg.mozilla.org/mozilla-central/rev/fe1ac343ac94