Add support for BUSY status coide from the target and just pass this
back to the application as is (instead of converting it to _ERROR).
This allows the application to trap task->status==SCSI_STATUS_BUSY and
decide what/how to proceed.
Dont call READCAPACITY from every single test.
Remove two arguments from the test signatures and turn them into globals.
remove the ret++ silliness. Tests return 0 : all ok, -2, skipped or -1 : test failed
Ordering by itt adds the special case of handling
itt == 0xffffffff. Order by CmdSN instead as described
by RFC3720.
Signed-off-by: Peter Lieven <pl@kamp.de>
Queing packets with itt = 0xffffffff (e.g. NOP-Out replies) ahead
of queue because they might have a higher CmdSN than following
packets. This again could lead to a deadlock AND its a protocol
violotion:
RFC3720 Section 3.2.2.1 second last paragraph on Page 21:
"On any connection, the iSCSI initiator MUST send the commands in
increasing order of CmdSN, except for commands that are retransmitted
due to digest error recovery and connection recovery."
Signed-off-by: Peter Lieven <pl@kamp.de>
This patch adds support for read/writev to directly read and write
from/to iovectors. Before this patch on read and write from/to socket
the operation was limited by the iovec boundaries. If there is enough
data in the buffer or enough buffer space available its now possible
to transfer the whole data in one atomic operaion.
Signed-off-by: Peter Lieven <pl@kamp.de>
This will set TCP_NODELAY on the socket connection
to the target. This is the first step to improve latency.
For systems supporting TCP_CORK we plan to add a cork
around certain PDUs e.g. DATA-Out, but this needs further
testing.
Signed-off-by: Peter Lieven <pl@kamp.de>
We first merge waitpdu and outqueue queues and then requeue
the elements. iscsi_add_to_outqueue() will sort them accordingly
so this is no problem.
Signed-off-by: Peter Lieven <pl@kamp.de>
We where modifying out_offset and out_len in iscsi_write_to_socket().
If the packet that was being sent before reconnect was a write command
the was a significant change that out_offset and out_len where already
touched. When requeing the packet after reconnect we where
sending garbage!
Signed-off-by: Peter Lieven <pl@kamp.de>
This patch adds the abilitiy to libiscsi to count the number
of consecutive outstanding NOPs.
With this ability its fairly easy to implement a keepalive
check with NOPs in your application.
Periodically (e.g. every 5 secs) create a NOP-Out with:
iscsi_nop_out_async(iscsi, NULL, NULL, 0, NULL);
At that time check the number of consecutive missing NOP-Ins
with
iscsi_get_nops_in_flight(iscsi) > N.
Where N is the number of missing NOP-Ins you will allow.
Please note that it is legitime for the Target to ignore
a NOP if the load is very high as those packet are mark
as IMMEDIATE.
Signed-off-by: Peter Lieven <pl@kamp.de>
Just like DATA_OUT we should just discard NOPs instead of requeueing them
on session reconnect.
Add new flag that to indicate this behaviour on reconnect and set it for
both data out and nops
In specific situation it might be useful to give up if a reconnect
is not successful or after a given number reconnect retries.
This patch adds the ability to change that. The default remains
the same: retry forever.
Signed-off-by: Peter Lieven <pl@kamp.de>
The preallocation logic was messing with size and offset of pdu->indata.
This was broken when we removed iscsi_data->alloc_size. Removing
entirely since reading without iovectors is deprecated.
Signed-off-by: Peter Lieven <pl@kamp.de>
A storage might sent an R2T response for a WRITE command while
we still sending out the WRITE command PDU. This is especially
the case when the command PDU carries immediata data.
Without this patch the R2T response will get lost as
the cmdpdu for the R2T cannot be found in iscsi_process_pdu()
leading to a deadlock.
Signed-off-by: Peter Lieven <pl@kamp.de>
