On 05/04/2012 02:25 AM, Guannan Ren wrote:
> refactor qemuPrepareHostdevUSBDevices function, make it focus on
> adding usb device to activeUsbHostdevs after check. After that,
> the usb hotplug function qemuDomainAttachHostDevice also could use
> it.
>
> expand qemuPrepareHostUSBDevices to perform the usb search,
> rollback on failure.
> ---
> src/qemu/qemu_hostdev.c | 117 +++++++++++++++++++++++++++-------------------
> src/qemu/qemu_hostdev.h | 3 +-
> 2 files changed, 70 insertions(+), 50 deletions(-)
>
> diff --git a/src/qemu/qemu_hostdev.c b/src/qemu/qemu_hostdev.c
> index b98fd8f..0cb89a6 100644
> --- a/src/qemu/qemu_hostdev.c
> +++ b/src/qemu/qemu_hostdev.c
> @@ -565,13 +565,50 @@ qemuPrepareHostPCIDevices(struct qemud_driver *driver,
> int
> qemuPrepareHostdevUSBDevices(struct qemud_driver *driver,
> const char *name,
> - virDomainHostdevDefPtr *hostdevs,
> - int nhostdevs)
> + usbDeviceList *list)
> {
> - int ret = -1;
> int i;
> + usbDevice *tmp;
> +
> + for (i = 0; i < usbDeviceListCount(list); i++) {
Thankfully, usbDeviceListCount() is O(1); if it were O(n) then you would
have a quadratic loop. Still, it is slightly faster to factor the
function call out of the for loop. On the other hand, right now we
aren't using gcc's attribute((pure/const)) because the warnings in gcc
4.6 were buggy; if we were to use that, then usbDeviceListCount() would
already be marked and there would be nothing to factor out. So, this is
fine as is, and we can save the optimization for a future day when we
rely on gcc 4.7 to help us mark the attributes where needed.
Unfortunately, this failed to compile on its own:
qemu/qemu_hotplug.c: In function 'qemuDomainAttachHostDevice':
qemu/qemu_hotplug.c:1137:9: error: passing argument 3 of
'qemuPrepareHostdevUSBDevices' from incompatible pointer type [-Werror]
qemu/qemu_hostdev.h:39:5: note: expected 'struct usbDeviceList *' but
argument is of type 'struct virDomainHostdevDef **'
qemu/qemu_hotplug.c:1137:9: error: too many arguments to function
'qemuPrepareHostdevUSBDevices'
qemu/qemu_hostdev.h:39:5: note: declared here
But since applying patch 3/3 fixes the failure:
ACK if you squash 2/3 and 3/3 into a single patch, so that the total
series is only 2 patches instead of 3.
--
Eric Blake eblake@xxxxxxxxxx +1-919-301-3266
Libvirt virtualization library http://libvirt.org
Attachment:
signature.asc
Description: OpenPGP digital signature
-- libvir-list mailing list libvir-list@xxxxxxxxxx https://www.redhat.com/mailman/listinfo/libvir-list