[libvirt] [PATCH] Doc: Explicitly declaring that nodedev-destroy only works for vHBA

Osier Yang jyang at redhat.com
Thu Nov 28 08:00:53 UTC 2013


On 22/11/13 20:55, Osier Yang wrote:

Can someone give an obvious ACK?

> Though trying to destroy a physical HBA doesn't make sense at all,
> it's still a bit misleading with saying "only works for HBA".
>
> Signed-off-by: Osier Yang <jyang at redhat.com>
> ---
>   src/libvirt.c   | 5 +++--
>   tools/virsh.pod | 6 +++---
>   2 files changed, 6 insertions(+), 5 deletions(-)
>
> diff --git a/src/libvirt.c b/src/libvirt.c
> index ae05300..4ebd13f 100644
> --- a/src/libvirt.c
> +++ b/src/libvirt.c
> @@ -16065,8 +16065,9 @@ error:
>    * virNodeDeviceDestroy:
>    * @dev: a device object
>    *
> - * Destroy the device object. The virtual device is removed from the host operating system.
> - * This function may require privileged access
> + * Destroy the device object. The virtual device (only works for vHBA
> + * currently) is removed from the host operating system.  This function
> + * may require privileged access.
>    *
>    * Returns 0 in case of success and -1 in case of failure.
>    */
> diff --git a/tools/virsh.pod b/tools/virsh.pod
> index dac9a08..557df9c 100644
> --- a/tools/virsh.pod
> +++ b/tools/virsh.pod
> @@ -2158,9 +2158,9 @@ contains xml for a top-level <device> description of a node device.
>   =item B<nodedev-destroy> I<device>
>   
>   Destroy (stop) a device on the host. I<device> can be either device
> -name or wwn pair in "wwnn,wwpn" format (only works for HBA). Note
> -that this makes libvirt quit managing a host device, and may even make
> -that device unusable by the rest of the physical host until a reboot.
> +name or wwn pair in "wwnn,wwpn" format (only works for vHBA currently).
> +Note that this makes libvirt quit managing a host device, and may even
> +make that device unusable by the rest of the physical host until a reboot.
>   
>   =item B<nodedev-detach> I<nodedev> [I<--driver backend_driver>]
>   




More information about the libvir-list mailing list