On Thu, Nov 12, 2009 at 03:00:48PM +0000, Daniel P. Berrange wrote:
> There is a race condition in HAL driver startup where the callback
> can get triggered before we have finished startup. This then causes
> a deadlock in the driver.
>
> * src/node_device/node_device_hal.c: RElease driver lock before
> registering DBus callbacks
> ---
> src/node_device/node_device_hal.c | 14 ++++++++++----
> 1 files changed, 10 insertions(+), 4 deletions(-)
>
> diff --git a/src/node_device/node_device_hal.c b/src/node_device/node_device_hal.c
> index 818c7d6..918a3a9 100644
> --- a/src/node_device/node_device_hal.c
> +++ b/src/node_device/node_device_hal.c
> @@ -742,6 +742,16 @@ static int halDeviceMonitorStartup(int privileged ATTRIBUTE_UNUSED)
> goto failure;
> }
>
> + /* Populate with known devices */
> + driverState->privateData = hal_ctx;
> +
> + /* We need to unlock state now, since setting these callbacks cause
> + * a dbus RPC call, and while this call is waiting for the reply,
> + * a signal may already arrive, triggering the callback and thus
> + * requiring the lock !
> + */
> + nodeDeviceUnlock(driverState);
> +
> /* Register HAL event callbacks */
> if (!libhal_ctx_set_device_added(hal_ctx, device_added) ||
> !libhal_ctx_set_device_removed(hal_ctx, device_removed) ||
> @@ -753,10 +763,6 @@ static int halDeviceMonitorStartup(int privileged ATTRIBUTE_UNUSED)
> goto failure;
> }
>
> - /* Populate with known devices */
> - driverState->privateData = hal_ctx;
> -
> - nodeDeviceUnlock(driverState);
> udi = libhal_get_all_devices(hal_ctx, &num_devs, &err);
> if (udi == NULL) {
> VIR_ERROR0("libhal_get_all_devices failed\n");
Tricky, ACK,
Daniel
--
Daniel Veillard | libxml Gnome XML XSLT toolkit http://xmlsoft.org/
daniel@xxxxxxxxxxxx | Rpmfind RPM search engine http://rpmfind.net/
http://veillard.com/ | virtualization library http://libvirt.org/
--
Libvir-list mailing list
Libvir-list@xxxxxxxxxx
https://www.redhat.com/mailman/listinfo/libvir-list