Thread (13 messages) 13 messages, 2 authors, 2023-11-02
STALE975d
Revisions (34)
  1. v1 [diff vs current]
  2. v1 [diff vs current]
  3. v2 [diff vs current]
  4. v1 [diff vs current]
  5. v1 [diff vs current]
  6. v2 [diff vs current]
  7. v1 current
  8. v2 [diff vs current]
  9. v1 [diff vs current]
  10. v1 [diff vs current]
  11. v2 [diff vs current]
  12. v3 [diff vs current]
  13. v4 [diff vs current]
  14. v1 [diff vs current]
  15. v2 [diff vs current]
  16. v3 [diff vs current]
  17. v1 [diff vs current]
  18. v1 [diff vs current]
  19. v1 [diff vs current]
  20. v2 [diff vs current]
  21. v3 [diff vs current]
  22. resend [diff vs current]
  23. v2 [diff vs current]
  24. v2 [diff vs current]
  25. v3 [diff vs current]
  26. v1 [diff vs current]
  27. v1 [diff vs current]
  28. v1 [diff vs current]
  29. v1 [diff vs current]
  30. v2 [diff vs current]
  31. v1 [diff vs current]
  32. v1 [diff vs current]
  33. v1 [diff vs current]
  34. v2 [diff vs current]

[PATCH net 2/7] net: hns3: add barrier in vf mailbox reply process

From: Jijie Shao <shaojijie@huawei.com>
Date: 2023-10-28 03:02:49
Also in: lkml
Subsystem: hisilicon network subsystem 3 driver (hns3), hisilicon network subsystem driver, networking drivers, the rest · Maintainers: Jian Shen, Jijie Shao, Andrew Lunn, "David S. Miller", Eric Dumazet, Jakub Kicinski, Paolo Abeni, Linus Torvalds

From: Yonglong Liu <redacted>

In hclgevf_mbx_handler() and hclgevf_get_mbx_resp() functions,
there is a typical store-store and load-load scenario between
received_resp and additional_info. This patch adds barrier
to fix the problem.

Fixes: 4671042f1ef0 ("net: hns3: add match_id to check mailbox response from PF to VF")
Signed-off-by: Yonglong Liu <redacted>
Signed-off-by: Jijie Shao <shaojijie@huawei.com>
---
 drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_mbx.c | 7 +++++++
 1 file changed, 7 insertions(+)
diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_mbx.c b/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_mbx.c
index bbf7b14079de..85c2a634c8f9 100644
--- a/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_mbx.c
+++ b/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_mbx.c
@@ -63,6 +63,9 @@ static int hclgevf_get_mbx_resp(struct hclgevf_dev *hdev, u16 code0, u16 code1,
 		i++;
 	}
 
+	/* ensure additional_info will be seen after received_resp */
+	smp_rmb();
+
 	if (i >= HCLGEVF_MAX_TRY_TIMES) {
 		dev_err(&hdev->pdev->dev,
 			"VF could not get mbx(%u,%u) resp(=%d) from PF in %d tries\n",
@@ -178,6 +181,10 @@ static void hclgevf_handle_mbx_response(struct hclgevf_dev *hdev,
 	resp->resp_status = hclgevf_resp_to_errno(resp_status);
 	memcpy(resp->additional_info, req->msg.resp_data,
 	       HCLGE_MBX_MAX_RESP_DATA_SIZE * sizeof(u8));
+
+	/* ensure additional_info will be seen before setting received_resp */
+	smp_wmb();
+
 	if (match_id) {
 		/* If match_id is not zero, it means PF support match_id.
 		 * if the match_id is right, VF get the right response, or
-- 
2.30.0
Keyboard shortcuts
hback out one level
jnext message in thread
kprevious message in thread
ldrill in
Escclose help / fold thread tree
?toggle this help