1117614 - [FFOS7715 v2.1][STK]USAT case 27.22.4.7.1 REFRESH SEQ 1.5 fail，STATUS [P1='02'] after FETCH not performed

Status: RESOLVED WONTFIX

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

Description

[jinghua.xing]

[Testing Steps     ]:USAT case 27.22.4.7.1 REFRESH SEQ 1.5 fail，STATUS [P1='02'] after FETCH not performed

[Expected Result  ]:After FETCH the Terminal shall send STATUS[P1='02'], reset the UICC and perform USIM initialisation including sending STATUS[P1='01'].The Terminal shall not send a TERMINAL RESPONSE

[Test Result      ]:STATUS [P1='02'] after FETCH not performed

Comment 1

[jinghua.xing]

12-30 01:15:59.480   130   130 I Gecko   : -*- RadioInterface[0]: Received message from worker: {"commandNumber":1,"typeOfCommand":1,"commandQualifier":4,"rilMessageType":"stkcommand","options":null,"rilMessageClientId":0}

The commandQualifier is 4 and it means now is STK_REFRESH_UICC_RESET

Comment 2

[jinghua.xing]

How do our FFOS support STK_REFRESH_UICC_RESET command?

Comment 3

[shawn ku [:sku]]

Hi Jinghua:
 Does this issue mean no TR (terminal response) for UICC_RESET? or other behavior is wrong?

Per comment 1, it only means "SIM reset" 0x04 was received by ril_worker.js.
Do you expect radio_power off -> on?

Thanks!!
Shawn

Comment 4

[jinghua.xing]

As the protocol says: 
>This mode causes the ME to run the GSM session termination procedure and to deactivate the SIM in accordance with >TS 11.11 [20]. Subsequently, the ME activates the SIM again and starts a new card session. 

>In the case The ME shall not send the TERMINAL RESPONSE;  The SIM Application shall interpret a new activation of >the contacts of the SIM as an implicit TERMINAL RESPONSE. 

For this, I think we should make the radio_power off -> on.

Do you have more advices about this? Thank you.

Comment 5

[jinghua.xing]

The AP should send AT commands to modem :Close protocol stack、Close SIM、Open SIM、 Open protocol stack

Comment 6

[jinghua.xing]

sku,

Did there any way to achieve the four steps in comment5 in our ril_worker or in gaia side?

Thank you.

Comment 7

[shawn ku [:sku]]

Hi Jinghua:
 There are two ways to finish this task.

1. Modem to finish the SIM reset by itself.
2. you can turn radio off at [1] if receiving refreshType == 4, gaia will turn radio on after receiving radiostate off.
(ex: this.context.RIL.setRadioEnabled({enabled: false});)

[1] - http://dxr.mozilla.org/mozilla-central/source/dom/system/gonk/ril_worker.js?from=ril_worker.js&case=true#10705

Comment 8

[jinghua.xing]

shawn:

Using "this.context.RIL.setRadioEnabled({enabled: false})" will only process the 'Close protocol stack' and 'Open protocol stack' two actions. Between the two actions, we should also send two commands to the modem to close SIM and to open SIM.

Thank you.

Comment 9

[shawn ku [:sku]]

Hi Jinghua:
 I don't think there is an interface/API to toggle SIM per standard AOSP design.
Please indicate which message between rild/gecko can do this by checking [1].

Besides, please also share the design that you have in rild to us, we can have a further discussion then.


[1] - https://github.com/android/platform_hardware_ril/blob/master/include/telephony/ril.h

Comment 10

[jinghua.xing]

Attachment: stk_refresh_1.5.patch

This is my patch for this issue in gecko. Can you help me to see is this way to reset the UICC ok?

Thank you.

Comment 11

[jinghua.xing]

Attachment: stk_refresh_1.5.txt

This is the log for this case after I merge my patch above.

Comment 12

[shawn ku [:sku]]

Comment on attachment 8545112 [details] [diff] [review]
stk_refresh_1.5.patch

Review of attachment 8545112 [details] [diff] [review]:
-----------------------------------------------------------------

::: dom/system/gonk/ril_worker.js
@@ +10847,5 @@
> +            this.context.debug("STK_Refresh, STK_REFRESH_UICC_RESET");
> +        }
> +        this.context.RIL.setRadioEnabled({enabled: false});
> +        this.context.RIL.setSimEnabled({enabled: false});
> +        this.context.RIL.setSimEnabled({enabled: true});

How can we know if setSimEnabled({enabled: false}); or setSimEnabled({enabled: true}); success or not?

Can we wait sim status after setSimEnabled({enabled: false});, and request "setSimEnabled({enabled: true});" again?

Comment 13

[shawn ku [:sku]]

Please check my reply in comment 12.

Comment 14

[jinghua.xing]

shawn：

Thank you for your feedback

Are reset the radio power and reset sim two separate process？ And we should close the sim after close the radio_power, if we do this in the response of REQUEST_RADIO_POWER， how should we determine if we were in the process of STK_REFRESH_UICC_RESET or normal radio reset process? As gaia will turn radio on after receiving radiostate off, is it possible we turn radio on before we complete the process of reset the sim?

Thank you.

Comment 15

[shawn ku [:sku]]

Hi Jinghua:
 Per comment 11, my two cents is:
1. turn SIM power to off after receiving REFREST_REST.
2. turn off RADIO power too, and flag this (example: turnSimOn = true) for later use.
3. once gaia trun radio power on, check if turnSimOn = true, trigger SIM power on command and RADIO POWER on then.

Comment 16

[jinghua.xing]

Shawn：

Thank you for your help. I have made a solution for this issue according your kindly advices

Comment 17

[BMO Automation]

Firefox OS is not being worked on