clear-pkgs-linux-iot-lts2018/0543-vhm-remove-re-schedule-for-ioreq-tasklet.patch

From 96ddd327f08322129e37fdd185f92cbc4ad9dd30 Mon Sep 17 00:00:00 2001
From: Shuo Liu <shuo.a.liu@intel.com>
Date: Fri, 31 Aug 2018 10:59:02 +0800
Subject: [PATCH 075/214] vhm: remove re-schedule for ioreq tasklet

io_req_tasklet can process all existing or even incoming ioreqs in the
ioreq shared page once. So when the ioreq IPI rising, we needn't rearm
the tasklet if the previous one havn't got running. tasklet_schedule
can gurantee to execute once after schedule, and all pending ioreqs
can be processed once. tasklet_schedule also can rearm the running
tasklet so no ioreq interrupt lost.

Signed-off-by: Shuo Liu <shuo.a.liu@intel.com>
Reviewed-by: Jason Chen CJ <jason.cj.chen@intel.com>
---
 drivers/char/vhm/vhm_dev.c | 11 +----------
 1 file changed, 1 insertion(+), 10 deletions(-)

diff --git a/drivers/char/vhm/vhm_dev.c b/drivers/char/vhm/vhm_dev.c
index e454b9efcda2..39204b5d7898 100644
--- a/drivers/char/vhm/vhm_dev.c
+++ b/drivers/char/vhm/vhm_dev.c
@@ -96,7 +96,6 @@ static int    major;
 static struct class *vhm_class;
 static struct device *vhm_device;
 static struct tasklet_struct vhm_io_req_tasklet;
-static atomic_t ioreq_retry = ATOMIC_INIT(0);
 
 struct table_iomems {
 	/* list node for this table_iomems */
@@ -622,19 +621,11 @@ static void io_req_tasklet(unsigned long data)
 
 		acrn_ioreq_distribute_request(vm);
 	}
-
-	if (atomic_read(&ioreq_retry) > 0) {
-		atomic_dec(&ioreq_retry);
-		tasklet_schedule(&vhm_io_req_tasklet);
-	}
 }
 
 static void vhm_intr_handler(void)
 {
-	if (test_bit(TASKLET_STATE_SCHED, &(vhm_io_req_tasklet.state)))
-		atomic_inc(&ioreq_retry);
-	else
-		tasklet_schedule(&vhm_io_req_tasklet);
+	tasklet_schedule(&vhm_io_req_tasklet);
 }
 
 static int vhm_dev_release(struct inode *inodep, struct file *filep)
-- 
2.19.1