>>> On 25.08.11 at 08:34, Li Dongyang <lidongyang@xxxxxxxxxx> wrote:
> On Wed, Aug 24, 2011 at 6:26 PM, Jan Beulich <JBeulich@xxxxxxxxxx> wrote:
>>>>> On 24.08.11 at 11:23, Li Dongyang <lidongyang@xxxxxxxxxx> wrote:
>>> This adds the BLKIF_OP_TRIM for blkfront and blkback, also 2 enums telling
>>> us the type of the backend, used in blkback to determine what to do when we
>>> see a trim request.
>>> The BLKIF_OP_TRIM part is just taken from Owen Smith, Thanks
>>>
>>> Signed-off-by: Owen Smith <owen.smith@xxxxxxxxxx>
>>> Signed-off-by: Li Dongyang <lidongyang@xxxxxxxxxx>
>>> ---
>>> drivers/block/xen-blkback/common.h | 10 +++++++++-
>>> include/xen/interface/io/blkif.h | 19 +++++++++++++++++++
>>> 2 files changed, 28 insertions(+), 1 deletions(-)
>>>
>>> diff --git a/drivers/block/xen-blkback/common.h
>>> b/drivers/block/xen-blkback/common.h
>>> index 9e40b28..146d325 100644
>>> --- a/drivers/block/xen-blkback/common.h
>>> +++ b/drivers/block/xen-blkback/common.h
>>> @@ -113,6 +113,11 @@ enum blkif_protocol {
>>> BLKIF_PROTOCOL_X86_64 = 3,
>>> };
>>>
>>> +enum blkif_backend_type {
>>> + BLKIF_BACKEND_PHY = 1,
>>> + BLKIF_BACKEND_FILE = 2,
>>> +};
>>> +
>>> struct xen_vbd {
>>> /* What the domain refers to this vbd as. */
>>> blkif_vdev_t handle;
>>> @@ -138,6 +143,7 @@ struct xen_blkif {
>>> unsigned int irq;
>>> /* Comms information. */
>>> enum blkif_protocol blk_protocol;
>>> + enum blkif_backend_type blk_backend_type;
>>> union blkif_back_rings blk_rings;
>>> struct vm_struct *blk_ring_area;
>>> /* The VBD attached to this interface. */
>>> @@ -159,8 +165,10 @@ struct xen_blkif {
>>> int st_wr_req;
>>> int st_oo_req;
>>> int st_f_req;
>>> + int st_tr_req;
>>> int st_rd_sect;
>>> int st_wr_sect;
>>> + int st_tr_sect;
>>
>> Just to repeat - I don't think this piece of statistic is very useful, the
>> more that you use "int" here while ...
>>
>>>
>>> wait_queue_head_t waiting_to_free;
>>>
>>> @@ -182,7 +190,7 @@ struct xen_blkif {
>>>
>>> struct phys_req {
>>> unsigned short dev;
>>> - unsigned short nr_sects;
>>> + blkif_sector_t nr_sects;
>>
>> ... you specifically widen the field to 64 bits here.
>>
> sounds reasonable, gonna kill the stat stuff
I didn't mean to say kill it all - the st_tr_req is quite reasonable to
have (in line with st_f_req).
Jan
_______________________________________________
Xen-devel mailing list
Xen-devel@xxxxxxxxxxxxxxxxxxx
http://lists.xensource.com/xen-devel
|
Home