From: Roland Dreier Date: Tue, 12 Jun 2012 01:23:16 +0000 (-0700) Subject: qla2xxx: Don't leak commands we give up on in qlt_do_work() X-Git-Tag: v3.5-rc6~5^2~5 X-Git-Url: http://review.tizen.org/git/?a=commitdiff_plain;h=fae9eaf81323eda64f28b9f4f4aeffb807e5b5f4;p=profile%2Fivi%2Fkernel-x86-ivi.git qla2xxx: Don't leak commands we give up on in qlt_do_work() If we go to the "out_term:" exit path in qlt_do_work(), we call qlt_send_term_exchange() with a NULL cmd, which means that it can't possibly free the cmd for us. Add an explicit call to free the command memory, so we don't leak the allocation. This will also fix warnings about "BUG qla_tgt_cmd_cachep: Objects remaining on kmem_cache_close" from slub when unloading the qla2xxx target module. Signed-off-by: Roland Dreier Signed-off-by: Nicholas Bellinger --- diff --git a/drivers/scsi/qla2xxx/qla_target.c b/drivers/scsi/qla2xxx/qla_target.c index c263f90..e912920 100644 --- a/drivers/scsi/qla2xxx/qla_target.c +++ b/drivers/scsi/qla2xxx/qla_target.c @@ -2725,10 +2725,12 @@ static void qlt_do_work(struct work_struct *work) out_term: ql_dbg(ql_dbg_tgt_mgt, vha, 0xf020, "Terminating work cmd %p", cmd); /* - * cmd has not sent to target yet, so pass NULL as the second argument + * cmd has not sent to target yet, so pass NULL as the second + * argument to qlt_send_term_exchange() and free the memory here. */ spin_lock_irqsave(&ha->hardware_lock, flags); qlt_send_term_exchange(vha, NULL, &cmd->atio, 1); + kmem_cache_free(qla_tgt_cmd_cachep, cmd); spin_unlock_irqrestore(&ha->hardware_lock, flags); if (sess) ha->tgt.tgt_ops->put_sess(sess);