Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]


Groups > linux.kernel > #1738693

[PATCH review for 4.9 10/50] net/mlx4_en: fix overflow in mlx4_en_init_timestamp()

Path csiph.com!news.mixmin.net!news.unit0.net!news.panservice.it!bofh.it!news.nic.it!robomod
From "Levin, Alexander (Sasha Levin)" <alexander.levin@verizon.com>
Newsgroups linux.kernel
Subject [PATCH review for 4.9 10/50] net/mlx4_en: fix overflow in mlx4_en_init_timestamp()
Date Mon, 25 Sep 2017 03:40:02 +0200
Message-ID <utqxk-7O2-31@gated-at.bofh.it> (permalink)
References <utqdY-7Fr-3@gated-at.bofh.it>
X-Original-To "linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>, "stable@vger.kernel.org" <stable@vger.kernel.org>
Dkim-Signature v=1; a=rsa-sha256; c=relaxed/simple; d=verizon.com; i=@verizon.com; q=dns/txt; s=corp; t=1506302050; x=1537838050; h=from:cc:to:subject:date:message-id:references: in-reply-to:content-transfer-encoding:mime-version; bh=sBS+SGdChURZ6eRC76Fi9iJo770m9t/APIPJ26tOv9E=; b=YYHZJezVd/uSyy87JpmnBGuFsQICcjfMXz2UDTd1XP/jLRZhLIKJVjY0 8OdsuW54wLoXyPfmAOEeewSK+4YqQiQIwgPw2lIJgUpXfTXf0y1N0ffM6 ubT/cfKf//5WEc4C+xbuq2bavQSRNdKJLkh1shi/Bs6SEeLw50h9D+J6/ Y=;
Dkim-Signature v=1; a=rsa-sha256; c=relaxed/simple; d=verizon.com; i=@verizon.com; q=dns/txt; s=corp; t=1506301968; x=1537837968; h=from:to:cc:subject:date:message-id:references: in-reply-to:content-transfer-encoding:mime-version; bh=sBS+SGdChURZ6eRC76Fi9iJo770m9t/APIPJ26tOv9E=; b=WCNoznF+Fon6eaKzns05CEveIdz7jak0Z5OSgMuIFVdCfMLFrokTf+JH jrS4gmKCl7dpbkYFyb+O6gpnOjkaunUQ5cxYgl7pLlXcab8qnNTnBIEcq UiJ4cypcID1hI+Re/KrGKGEMQLgoSQe62bn/k4FMCvZXRNLClUA0lzfXB 4=;
Dkim-Signature v=1; a=rsa-sha256; c=relaxed/simple; d=verizon.com; i=@verizon.com; q=dns/txt; s=corp; t=1506301967; x=1537837967; h=from:to:cc:subject:date:message-id:references: in-reply-to:content-transfer-encoding:mime-version; bh=sBS+SGdChURZ6eRC76Fi9iJo770m9t/APIPJ26tOv9E=; b=QuXGCSN/ZRnXjhGZYACZiJCVQRxGDTrzuuCAc/ieobbErk26tbYbo1aI tUUsUkso0dZ7egZTC4JlfWpgTfjEVV2rm5JL3uqMELwkV6I01NERZJA7l wTgEyzAJPx4dPFsJGavMzgnHmYkwAlHUQAZ+5UsP29B7cqBECPkB4R3WN 0=;
X-Host viking.odc.vzwcorp.com
Thread-Topic [PATCH review for 4.9 10/50] net/mlx4_en: fix overflow in mlx4_en_init_timestamp()
Thread-Index AQHTNZtjjybBPXFZCEi+lqtgzboRyg==
Accept-Language en-US
Content-Language en-US
X-Ms-Exchange-Messagesentrepresentingtype 1
X-Ms-Exchange-Transport-Fromentityheader Hosted
X-Originating-IP [10.144.60.250]
Content-Type text/plain; charset="iso-8859-1"
Content-Transfer-Encoding quoted-printable
MIME-Version 1.0
Sender robomod@news.nic.it
List-ID <linux-kernel.vger.kernel.org>
X-Mailing-List linux-kernel@vger.kernel.org
Approved robomod@news.nic.it
Lines 102
Organization linux.* mail to news gateway
X-Original-Cc Eric Dumazet <edumazet@google.com>, Tariq Toukan <tariqt@mellanox.com>, Eugenia Emantayev <eugenia@mellanox.com>, "David S . Miller" <davem@davemloft.net>, "Levin, Alexander (Sasha Levin)" <alexander.levin@verizon.com>
X-Original-Date Mon, 25 Sep 2017 01:12:43 +0000
X-Original-Message-ID <20170925011225.10029-10-alexander.levin@verizon.com>
X-Original-References <20170925011225.10029-1-alexander.levin@verizon.com>
X-Original-Sender linux-kernel-owner@vger.kernel.org
Xref csiph.com linux.kernel:1738693

Show key headers only | View raw


From: Eric Dumazet <edumazet@google.com>

[ Upstream commit 47d3a07528ecbbccf53bc4390d70b4e3d1c04fcf ]

The cited commit makes a great job of finding optimal shift/multiplier
values assuming a 10 seconds wrap around, but forgot to change the
overflow_period computation.

It overflows in cyclecounter_cyc2ns(), and the final result is 804 ms,
which is silly.

Lets simply use 5 seconds, no need to recompute this, given how it is
supposed to work.

Later, we will use a timer instead of a work queue, since the new RX
allocation schem will no longer need mlx4_en_recover_from_oom() and the
service_task firing every 250 ms.

Fixes: 31c128b66e5b ("net/mlx4_en: Choose time-stamping shift value according to HW frequency")
Signed-off-by: Eric Dumazet <edumazet@google.com>
Cc: Tariq Toukan <tariqt@mellanox.com>
Cc: Eugenia Emantayev <eugenia@mellanox.com>
Reviewed-by: Tariq Toukan <tariqt@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Sasha Levin <alexander.levin@verizon.com>
---
 drivers/net/ethernet/mellanox/mlx4/en_clock.c | 18 ++++++++----------
 drivers/net/ethernet/mellanox/mlx4/mlx4_en.h  |  1 -
 2 files changed, 8 insertions(+), 11 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx4/en_clock.c b/drivers/net/ethernet/mellanox/mlx4/en_clock.c
index a5fc46bbcbe2..d4d97ca12e83 100644
--- a/drivers/net/ethernet/mellanox/mlx4/en_clock.c
+++ b/drivers/net/ethernet/mellanox/mlx4/en_clock.c
@@ -88,10 +88,17 @@ void mlx4_en_remove_timestamp(struct mlx4_en_dev *mdev)
 	}
 }
 
+#define MLX4_EN_WRAP_AROUND_SEC	10UL
+/* By scheduling the overflow check every 5 seconds, we have a reasonably
+ * good chance we wont miss a wrap around.
+ * TOTO: Use a timer instead of a work queue to increase the guarantee.
+ */
+#define MLX4_EN_OVERFLOW_PERIOD (MLX4_EN_WRAP_AROUND_SEC * HZ / 2)
+
 void mlx4_en_ptp_overflow_check(struct mlx4_en_dev *mdev)
 {
 	bool timeout = time_is_before_jiffies(mdev->last_overflow_check +
-					      mdev->overflow_period);
+					      MLX4_EN_OVERFLOW_PERIOD);
 	unsigned long flags;
 
 	if (timeout) {
@@ -236,7 +243,6 @@ static const struct ptp_clock_info mlx4_en_ptp_clock_info = {
 	.enable		= mlx4_en_phc_enable,
 };
 
-#define MLX4_EN_WRAP_AROUND_SEC	10ULL
 
 /* This function calculates the max shift that enables the user range
  * of MLX4_EN_WRAP_AROUND_SEC values in the cycles register.
@@ -261,7 +267,6 @@ void mlx4_en_init_timestamp(struct mlx4_en_dev *mdev)
 {
 	struct mlx4_dev *dev = mdev->dev;
 	unsigned long flags;
-	u64 ns, zero = 0;
 
 	/* mlx4_en_init_timestamp is called for each netdev.
 	 * mdev->ptp_clock is common for all ports, skip initialization if
@@ -285,13 +290,6 @@ void mlx4_en_init_timestamp(struct mlx4_en_dev *mdev)
 			 ktime_to_ns(ktime_get_real()));
 	write_unlock_irqrestore(&mdev->clock_lock, flags);
 
-	/* Calculate period in seconds to call the overflow watchdog - to make
-	 * sure counter is checked at least once every wrap around.
-	 */
-	ns = cyclecounter_cyc2ns(&mdev->cycles, mdev->cycles.mask, zero, &zero);
-	do_div(ns, NSEC_PER_SEC / 2 / HZ);
-	mdev->overflow_period = ns;
-
 	/* Configure the PHC */
 	mdev->ptp_clock_info = mlx4_en_ptp_clock_info;
 	snprintf(mdev->ptp_clock_info.name, 16, "mlx4 ptp");
diff --git a/drivers/net/ethernet/mellanox/mlx4/mlx4_en.h b/drivers/net/ethernet/mellanox/mlx4/mlx4_en.h
index a3528dd1e72e..df0f39611c5e 100644
--- a/drivers/net/ethernet/mellanox/mlx4/mlx4_en.h
+++ b/drivers/net/ethernet/mellanox/mlx4/mlx4_en.h
@@ -419,7 +419,6 @@ struct mlx4_en_dev {
 	struct cyclecounter	cycles;
 	struct timecounter	clock;
 	unsigned long		last_overflow_check;
-	unsigned long		overflow_period;
 	struct ptp_clock	*ptp_clock;
 	struct ptp_clock_info	ptp_clock_info;
 	struct notifier_block	nb;
-- 
2.11.0

Back to linux.kernel | Previous | Next | Find similar | Unroll thread


Thread

[PATCH review for 4.9 10/50] net/mlx4_en: fix overflow in  mlx4_en_init_timestamp() "Levin, Alexander (Sasha Levin)" <alexander.levin@verizon.com> - 2017-09-25 03:40 +0200

csiph-web