Displaying 3 results from an estimated 3 matches for "nbd_cmd_flag".
Did you mean:
nbd_cmd_flag_
2020 Sep 11
3
[libnbd PATCH] api: Add LIBNBD_SHUTDOWN_IMMEDIATE flag
..._shutdown was to let us specify
different behaviors at the libnbd level, rather than NBD protocol
flags (for that, the user has nbd_aio_disconnect). But when we later
parameterized OFlags to accept various bitmasks (commit f891340b), we
failed to mark nbd_shutdown as using a different bitmask than
NBD_CMD_FLAG.
I've finally found a use for such a flag. By itself,
nbd_aio_disconnect merely queues itself at the back of all pending
commands. But if the server is processing responses slowly, it can be
desirable to elevate a disconnect request to the front of the queue,
intentionally abandoning all sub...
2020 Sep 17
1
Re: [libnbd PATCH] api: Add LIBNBD_SHUTDOWN_IMMEDIATE flag
...erent behaviors at the libnbd level, rather than NBD protocol
>> flags (for that, the user has nbd_aio_disconnect). But when we later
>> parameterized OFlags to accept various bitmasks (commit f891340b), we
>> failed to mark nbd_shutdown as using a different bitmask than
>> NBD_CMD_FLAG.
>>
>> I've finally found a use for such a flag. By itself,
>> nbd_aio_disconnect merely queues itself at the back of all pending
>> commands. But if the server is processing responses slowly, it can be
>> desirable to elevate a disconnect request to the front of...
2020 Sep 17
0
Re: [libnbd PATCH] api: Add LIBNBD_SHUTDOWN_IMMEDIATE flag
...pecify
> different behaviors at the libnbd level, rather than NBD protocol
> flags (for that, the user has nbd_aio_disconnect). But when we later
> parameterized OFlags to accept various bitmasks (commit f891340b), we
> failed to mark nbd_shutdown as using a different bitmask than
> NBD_CMD_FLAG.
>
> I've finally found a use for such a flag. By itself,
> nbd_aio_disconnect merely queues itself at the back of all pending
> commands. But if the server is processing responses slowly, it can be
> desirable to elevate a disconnect request to the front of the queue,
> int...