linux-toradex.git/include/target/target_core_base.h, branch v3.10.2

target: Re-instate sess_wait_list for target_wait_for_sess_cmds

2013-05-30T04:30:33+00:00

Switch back to pre commit 1c7b13fe652 list splicing logic for active I/O
shutdown with tcm_qla2xxx + ib_srpt fabrics.

The original commit was done under the incorrect assumption that it's safe to
walk se_sess->sess_cmd_list unprotected in target_wait_for_sess_cmds() after
sess->sess_tearing_down = 1 has been set by target_sess_cmd_list_set_waiting()
during session shutdown.

So instead of adding sess->sess_cmd_lock protection around sess->sess_cmd_list
during target_wait_for_sess_cmds(), switch back to sess->sess_wait_list to
allow wait_for_completion() + TFO->release_cmd() to occur without having to
walk ->sess_cmd_list after the list_splice.

Also add a check to exit if target_sess_cmd_list_set_waiting() has already
been called, and add a WARN_ON to check for any fabric bug where new se_cmds
are added to sess->sess_cmd_list after sess->sess_tearing_down = 1 has already
been set.

Cc: Joern Engel 
Cc: Roland Dreier 
Cc: stable@vger.kernel.org
Signed-off-by: Nicholas Bellinger

target: removed unused transport_state flag

2013-05-15T08:46:53+00:00

Signed-off-by: Joern Engel 
Signed-off-by: Nicholas Bellinger

target: Remove unused struct members in se_dev_entry

2013-05-03T23:42:18+00:00

Some were incremented, but never used anywhere from what I could tell.

Signed-off-by: Andy Grover 
Signed-off-by: Nicholas Bellinger

target: Fix error checking for UNMAP commands

2013-02-13T20:16:05+00:00

SBC-3 (revision 35) says:

    The PARAMETER LIST LENGTH field specifies the length in bytes of the
    UNMAP parameter list that is available to be transferred from the
    Data-Out Buffer. If the parameter list length is greater than zero
    and less than 0008h (i.e., eight), then the device server shall
    terminate the command with CHECK CONDITION status with the sense key
    set to ILLEGAL REQUEST and the additional sense code set to
    PARAMETER LIST LENGTH ERROR. A PARAMETER LIST LENGTH set to zero
    specifies that no data shall be sent.

so our sense code for too-short descriptors was wrong, and we were
incorrectly failing commands that didn't transfer any descriptors.

While we're at it, also handle the UNMAP check:

    If the ANCHOR bit is set to one, and the ANC_SUP bit in the Logical
    Block Provisioning VPD page (see 6.6.4) is set to zero, then the
    device server shall terminate the command with CHECK CONDITION
    status with the sense key set to ILLEGAL REQUEST and the additional
    sense code set to INVALID FIELD IN CDB.

(chris boot: Fix wrong cut+paste comment in transport_send_check_condition_and_sense)

Signed-off-by: Roland Dreier 
Signed-off-by: Nicholas Bellinger

target: Add device attribute to expose config_item_name for INQUIRY model

2013-02-13T19:27:58+00:00

This patch changes LIO to use the configfs backend device name as the
model if you echo '1' to an individual device's emulate_model_alias attribute.
This is a valid operation only on devices with an export count of 0.

Signed-off-by: Tregaron Bayly 
Signed-off-by: Nicholas Bellinger

target: change sprintf to snprintf in transport_dump_vpd_ident

2013-02-13T19:27:31+00:00

"buf" is 128 characters and "vpd->device_identifier" is 256.  It makes
the static checkers complain.

Also bump VPD_TMP_BUF_SIZE to match INQUIRY_VPD_DEVICE_IDENTIFIER_LEN.

Signed-off-by: Dan Carpenter 
Signed-off-by: Nicholas Bellinger

target: Remove never-used TMR_FABRIC_TMR enum value

2013-02-13T19:27:22+00:00

Signed-off-by: Roland Dreier 
Signed-off-by: Nicholas Bellinger

target: Introduce TCM_NO_SENSE

2013-01-11T04:06:08+00:00

Introduce TCM_NO_SENSE, mapping to sense code
'Not ready, no additional sense information'.

Signed-off-by: Hannes Reinecke 
Cc: Nicholas Bellinger 
Signed-off-by: Nicholas Bellinger

target/iscsi_target: Add NodeACL tags for initiator group support

2012-12-13T22:18:09+00:00

Thanks for reviews, looking a lot better.

---- 8< ----

Initiator access config could be easier. The way other storage vendors
have addressed this is to support initiator groups: the admin adds
initiator WWNs to the group, and then LUN permissions can be granted for
the entire group at once.

Instead of changing ktarget's configfs interface, this patch keeps
the configfs interface per-initiator-wwn and just adds a 'tag' field
for each. This should be enough for user tools like targetcli to group
initiator ACLs and sync their configurations.

acl_tag is not used internally, but needs to be kept in configfs so that
all user tools can avoid dependencies on each other.

Code tested to work, although userspace pieces still to be implemented.

Signed-off-by: Andy Grover 
Signed-off-by: Nicholas Bellinger

target: Add link_magic for fabric allow_link destination target_items

2012-12-05T08:11:36+00:00

This patch adds [dev,lun]_link_magic value assignment + checks within generic
target_fabric_port_link() and target_fabric_mappedlun_link() code to ensure
destination config_item *target_item sent from configfs_symlink() ->
config_item_operations->allow_link() is the underlying se_device->dev_group
and se_lun->lun_group that we expect to symlink.

Reported-by: Sebastian Andrzej Siewior 
Cc: Sebastian Andrzej Siewior 
Cc: stable@vger.kernel.org
Signed-off-by: Nicholas Bellinger