[ 
https://issues.apache.org/jira/browse/MYNEWT-64?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
 ]

Christopher Collins resolved MYNEWT-64.
---------------------------------------
    Resolution: Fixed

This is implemented via two changes:
* Add a ble_hci_sched mutex that gets locked whenever the list of HCI tx slots 
is read or modified.
* Add a handle per HCI tx slot.  This allows client code to cancel an HCI 
transmission before the tx callback is rendered unusable.


> BLE Host - HCI callback thread-safety
> -------------------------------------
>
>                 Key: MYNEWT-64
>                 URL: https://issues.apache.org/jira/browse/MYNEWT-64
>             Project: Mynewt
>          Issue Type: Bug
>          Components: Nimble
>            Reporter: Christopher Collins
>            Assignee: Christopher Collins
>             Fix For: v0_8_0_beta2
>
>
> When the host schedules an outgoing HCI command, it configures a callback 
> which does the actual command formation.  Some events can occur before the 
> callback is executed which render the callback invalid (e.g., connection 
> terminated).
> The fix is to ensure the appropriate resources are protected from deletion 
> while an HCI command is scheduled.



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)

Reply via email to