linux-kernel - [PATCH net-next 07/12] net: hns3: always assume no drop TC for performance reason

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Date:   Thu, 11 Apr 2019 20:25:26 +0800
From:   Huazhong Tan <tanhuazhong@...wei.com>
To:     <davem@...emloft.net>
CC:     <netdev@...r.kernel.org>, <linux-kernel@...r.kernel.org>,
        <salil.mehta@...wei.com>, <yisen.zhuang@...wei.com>,
        <linuxarm@...wei.com>, Yunsheng Lin <linyunsheng@...wei.com>,
        Peng Li <lipeng321@...wei.com>,
        Huazhong Tan <tanhuazhong@...wei.com>
Subject: [PATCH net-next 07/12] net: hns3: always assume no drop TC for performance reason

From: Yunsheng Lin <linyunsheng@...wei.com>

Currently rx shared buffer' threshold size for speific TC is
set to smaller value when the TC's PFC is not enabled, which may
cause performance problem because hardware may not have enough
hardware buffer when PFC is not enabled.

This patch sets the same threshold size for all TC no matter if
the specific TC's PFC is enabled.

Signed-off-by: Yunsheng Lin <linyunsheng@...wei.com>
Signed-off-by: Peng Li <lipeng321@...wei.com>
Signed-off-by: Huazhong Tan <tanhuazhong@...wei.com>
---
 .../ethernet/hisilicon/hns3/hns3pf/hclge_main.c    | 29 +++-------------------
 1 file changed, 4 insertions(+), 25 deletions(-)

diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c
index 69d1113..5926016 100644
--- a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c
+++ b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c
@@ -1432,17 +1432,6 @@ static int hclge_get_tc_num(struct hclge_dev *hdev)
 	return cnt;
 }
 
-static int hclge_get_pfc_enalbe_num(struct hclge_dev *hdev)
-{
-	int i, cnt = 0;
-
-	for (i = 0; i < HCLGE_MAX_TC_NUM; i++)
-		if (hdev->hw_tc_map & BIT(i) &&
-		    hdev->tm_info.hw_pfc_map & BIT(i))
-			cnt++;
-	return cnt;
-}
-
 /* Get the number of pfc enabled TCs, which have private buffer */
 static int hclge_get_pfc_priv_num(struct hclge_dev *hdev,
 				  struct hclge_pkt_buf_alloc *buf_alloc)
@@ -1507,13 +1496,11 @@ static bool  hclge_is_rx_buf_ok(struct hclge_dev *hdev,
 				u32 rx_all)
 {
 	u32 shared_buf_min, shared_buf_tc, shared_std;
-	int tc_num, pfc_enable_num;
+	int tc_num = hclge_get_tc_num(hdev);
 	u32 shared_buf, aligned_mps;
 	u32 rx_priv;
 	int i;
 
-	tc_num = hclge_get_tc_num(hdev);
-	pfc_enable_num = hclge_get_pfc_enalbe_num(hdev);
 	aligned_mps = roundup(hdev->mps, HCLGE_BUF_SIZE_UNIT);
 
 	if (hnae3_dev_dcb_supported(hdev))
@@ -1522,9 +1509,7 @@ static bool  hclge_is_rx_buf_ok(struct hclge_dev *hdev,
 		shared_buf_min = aligned_mps + HCLGE_NON_DCB_ADDITIONAL_BUF
 					+ hdev->dv_buf_size;
 
-	shared_buf_tc = pfc_enable_num * aligned_mps +
-			(tc_num - pfc_enable_num) * aligned_mps / 2 +
-			aligned_mps;
+	shared_buf_tc = tc_num * aligned_mps + aligned_mps;
 	shared_std = roundup(max_t(u32, shared_buf_min, shared_buf_tc),
 			     HCLGE_BUF_SIZE_UNIT);
 
@@ -1546,14 +1531,8 @@ static bool  hclge_is_rx_buf_ok(struct hclge_dev *hdev,
 	}
 
 	for (i = 0; i < HCLGE_MAX_TC_NUM; i++) {
-		if ((hdev->hw_tc_map & BIT(i)) &&
-		    (hdev->tm_info.hw_pfc_map & BIT(i))) {
-			buf_alloc->s_buf.tc_thrd[i].low = aligned_mps;
-			buf_alloc->s_buf.tc_thrd[i].high = 2 * aligned_mps;
-		} else {
-			buf_alloc->s_buf.tc_thrd[i].low = 0;
-			buf_alloc->s_buf.tc_thrd[i].high = aligned_mps;
-		}
+		buf_alloc->s_buf.tc_thrd[i].low = aligned_mps;
+		buf_alloc->s_buf.tc_thrd[i].high = 2 * aligned_mps;
 	}
 
 	return true;
-- 
2.7.4