1117619 - [FFOS7715 v2.1][STK]USAT case 27.22.4.7.1 REFRESH SEQ 1.1 fail，TR返0x12（No response from user）

Status: RESOLVED WONTFIX

(Firefox OS Graveyard :: Gaia::Settings, defect)

Description

[jinghua.xing]

[Testing Steps     ]:USAT case 27.22.4.7.1 REFRESH SEQ 1.1	REFRESH, USIM Initialization

[Expected Result  ]:case pass
BER-TLV: 81 03 01 01 03 82 02 82 81 83 01 00或
BER-TLV: 81 03 01 01 03 82 02 82 81 83 01 03

[Test Result      ]:返0x12（No response from user）

Comment 1

[jinghua.xing]

When we test the case 27.22.4.7.1 REFRESH SEQ 1.1, it failed and the TR returned 0x12. I found the message of REFRESH is 
{"iccId":"89860112811023010685","command":{"commandNumber":1,"typeOfCommand":1,"commandQualifier":3,"rilMessageType":"stkcommand","options":null,"rilMessageClientId":0}}

We discard the message because the options property is null, and in handleSTKCommand() in the file system icc.js, as the code:
if (!message || !message.iccId || !message.command ||
        !message.command.typeOfCommand || !message.command.options) {
      return DUMP('STK Proactive Command bad formed: ', message);
    } 
the message is discarded.So there isn't any response for this case and it failed.

Is the !message.command.options really needed as in this case the options is null?

Comment 2

[jinghua.xing]

See also https://bugzilla.mozilla.org/show_bug.cgi?id=975299

Comment 3

[jinghua.xing]

Attachment: stk_refresh.patch

Can we just remove the " !message.command.options"

Comment 4

[siiaceon]

pls help ckeck 975299 effect, thanks

Comment 5

[Fernando R. Sela (no CC, needinfo please) [:frsela]]

Comment on attachment 8544551 [details] [diff] [review]
stk_refresh.patch

Review of attachment 8544551 [details] [diff] [review]:
-----------------------------------------------------------------

LGTM but please, if no options are provided, assign an empty object to avoid crashes:

if (!message.command.options) {
  message.command.options = {}
}

Comment 6

[shawn ku [:sku]]

Hi Jinghua:
 As my understanding, there should be *no* TR for SIM REFRESH PR.
We remove TR at master (see [1]), 2.1 is not up-lifted.

This REFRESH TR should be responded by modem.
After applying your patch, gaia will reply TR to SIM card which is not a valid design to me.


Please go check pure AOSP behavior as well.
 

[1] - https://github.com/mozilla-b2g/gaia/blob/master/apps/system/js/icc_worker.js#L29

Comment 7

[shawn ku [:sku]]

Make statement more precise:
 As my understanding, there should be *no* Gaia/Gecko TR for SIM REFRESH PR.

Comment 8

[siiaceon]

yes shawn, I agree with you. but modem need face all platform such as android,FeaturePhone,FFOS. so it is hard to make them modify except SPEC directly point out that they should do.
In the current sprd modem design , app level should let SIM reset, let PS off then on..., trigger order and trigger time all base on MMI recieved ACK from the "worker" .

So some case is a touble for us, but we still will keep talking with modem.

if we need TR from Gaia/Gecko, this patch which Fernado reviewed is OK, right?

Thanks

Comment 9

[jinghua.xing]

Our modem can not return the response for SIM REFRESH PR, so it need us to send a TR to the sim

Comment 10

[shawn ku [:sku]]

Hi Jinghua:
 All right, you can do this things on your own branch.

But, what I am trying to said is, there should be *no* TR for REFRESH/SEND_SS/SEND_DTMF/SEND_SMS/PLAY_TONE etc... from telephony framework or UI on pure AOSP.

I think it should be better for SPRD to follow what AOSP has been done.

Comment 11

[jinghua.xing]

Shawn:

I have fixed this on our branch, Thank you for your help

Comment 12

[BMO Automation]

Firefox OS is not being worked on