On 5 Mar 2016, at 07:33, Nicholas A. Bellinger <[email protected]> wrote: > > Hi Dan + BootC, > > On Wed, 2016-03-02 at 13:09 +0300, Dan Carpenter wrote: >> We changed this from kzalloc to sbp_mgt_get_req() so we need to change >> from checking for NULL to check for error pointers. >> >> Fixes: c064b2a78989 ('sbp-target: Conversion to percpu_ida tag >> pre-allocation') >> Signed-off-by: Dan Carpenter <[email protected]> >> >> diff --git a/drivers/target/sbp/sbp_target.c >> b/drivers/target/sbp/sbp_target.c >> index 251d532..a04b0605f 100644 >> --- a/drivers/target/sbp/sbp_target.c >> +++ b/drivers/target/sbp/sbp_target.c >> @@ -951,7 +951,7 @@ static void tgt_agent_fetch_work(struct work_struct >> *work) >> >> while (next_orb && tgt_agent_check_active(agent)) { >> req = sbp_mgt_get_req(sess, sess->card, next_orb); >> - if (!req) { >> + if (IS_ERR(req)) { >> spin_lock_bh(&agent->lock); >> agent->state = AGENT_STATE_DEAD; >> spin_unlock_bh(&agent->lock); > > Fixed + folded into the original patch. > > Thanks Dan. > > Chris, would you be so kind to review the original changes here: > > sbp-target: Conversion to percpu_ida tag pre-allocation > http://www.spinics.net/lists/target-devel/msg11778.html > > sbp-target: Convert to TARGET_SCF_ACK_KREF I/O krefs > http://www.spinics.net/lists/target-devel/msg11780.html > > and verify on your local IEEE1394 target setup..?
Hi Nic, Dan, I’m away this weekend so I can’t test these for a few days at least, unfortunately. I must admit I only vaguely follow the changes here as I haven’t been keeping up with the pace of change in target-devel lately, but it generally looks OK I think. Are these in linux-next or another branch somewhere I can easily clone them from? How soon do you need my ACK/NAK on these? Cheers, Chris -- Chris Boot [email protected]
