+ * Send I/O requests to the server.
+ *
+ * This function sends requests to the server, links the requests to
+ * the outstanding_list in BDRVNBDState, and exits without waiting for
+ * the response. The responses are received in the `aio_read_response'
+ * function which is called
On Tue, Feb 22, 2011 at 11:48 AM, Nicholas Thomas n...@bytemark.co.uk wrote:
+ * Send I/O requests to the server.
+ *
+ * This function sends requests to the server, links the requests to
+ * the outstanding_list in BDRVNBDState, and exits without waiting for
+ * the response. The
At Mon, 21 Feb 2011 17:48:49 +0100,
Kevin Wolf wrote:
Am 21.02.2011 17:31, schrieb Nicholas Thomas:
Hi again,
Thanks for looking through the patches. I'm just going through and
making the suggested changes now. I've also got qemu-nbd and block/nbd.c
working over IPv6 :) - hopefully
Am 18.02.2011 13:55, schrieb Nick Thomas:
This preserves the previous behaviour where the NBD server is
unavailable or goes away during guest execution, but switches the
NBD backend to present the AIO interface instead of the sync IO
interface.
We also split write requests into 1 MiB blocks
Hi again,
Thanks for looking through the patches. I'm just going through and
making the suggested changes now. I've also got qemu-nbd and block/nbd.c
working over IPv6 :) - hopefully I'll be able to provide patches in a
couple of days. Just a few questions about some of the changes...
Canceled
Am 21.02.2011 17:31, schrieb Nicholas Thomas:
Hi again,
Thanks for looking through the patches. I'm just going through and
making the suggested changes now. I've also got qemu-nbd and block/nbd.c
working over IPv6 :) - hopefully I'll be able to provide patches in a
couple of days. Just a
On Fri, Feb 18, 2011 at 12:55:29PM +, Nick Thomas wrote:
+static inline AIOReq *alloc_aio_req(BDRVNBDState *s, NBDAIOCB *acb,
+size_t data_len,
+off_t offset,
+off_t iov_offset)
+{