netdev - RE: [PATCH net-next v2 4/6] ice: sleep, don't busy-wait, for ICE_CTL_Q_SQ_CMD

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 for Android: free password hash cracker in your pocket

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

Date:   Fri, 14 Apr 2023 17:29:12 +0000
From:   "Kubalewski, Arkadiusz" <arkadiusz.kubalewski@...el.com>
To:     mschmidt <mschmidt@...hat.com>,
        "intel-wired-lan@...ts.osuosl.org" <intel-wired-lan@...ts.osuosl.org>
CC:     "netdev@...r.kernel.org" <netdev@...r.kernel.org>,
        "Brandeburg, Jesse" <jesse.brandeburg@...el.com>,
        "Kolacinski, Karol" <karol.kolacinski@...el.com>,
        "Nguyen, Anthony L" <anthony.l.nguyen@...el.com>,
        Simon Horman <simon.horman@...igine.com>,
        "Michalik, Michal" <michal.michalik@...el.com>,
        poros <poros@...hat.com>, Andrew Lunn <andrew@...n.ch>,
        Brent Rowsell <browsell@...hat.com>
Subject: RE: [PATCH net-next v2 4/6] ice: sleep, don't busy-wait, for
 ICE_CTL_Q_SQ_CMD_TIMEOUT

>From: Michal Schmidt <mschmidt@...hat.com>
>Sent: Wednesday, April 12, 2023 10:19 AM
>
>The driver polls for ice_sq_done() with a 100 µs period for up to 1 s
>and it uses udelay to do that.
>
>Let's use usleep_range instead. We know sleeping is allowed here,
>because we're holding a mutex (cq->sq_lock). To preserve the total
>max waiting time, measure the timeout in jiffies.
>
>ICE_CTL_Q_SQ_CMD_TIMEOUT is used also in ice_release_res(), but there
>the polling period is 1 ms (i.e. 10 times longer). Since the timeout was
>expressed in terms of the number of loops, the total timeout in this
>function is 10 s. I do not know if this is intentional. This patch keeps
>it.
>
>The patch lowers the CPU usage of the ice-gnss-<dev_name> kernel thread
>on my system from ~8 % to less than 1 %.
>
>I received a report of high CPU usage with ptp4l where the busy-waiting
>in ice_sq_send_cmd dominated the profile. This patch has been tested in
>that usecase too and it made a huge improvement there.
>
>Tested-by: Brent Rowsell <browsell@...hat.com>
>Signed-off-by: Michal Schmidt <mschmidt@...hat.com>
>---
> drivers/net/ethernet/intel/ice/ice_common.c   | 14 +++++++-------
> drivers/net/ethernet/intel/ice/ice_controlq.c |  9 +++++----
> drivers/net/ethernet/intel/ice/ice_controlq.h |  2 +-
> 3 files changed, 13 insertions(+), 12 deletions(-)
>
>diff --git a/drivers/net/ethernet/intel/ice/ice_common.c
>b/drivers/net/ethernet/intel/ice/ice_common.c
>index f4c256563248..3638598d732b 100644
>--- a/drivers/net/ethernet/intel/ice/ice_common.c
>+++ b/drivers/net/ethernet/intel/ice/ice_common.c
>@@ -1992,19 +1992,19 @@ ice_acquire_res(struct ice_hw *hw, enum
>ice_aq_res_ids res,
>  */
> void ice_release_res(struct ice_hw *hw, enum ice_aq_res_ids res)
> {
>-	u32 total_delay = 0;
>+	unsigned long timeout;
> 	int status;
>
>-	status = ice_aq_release_res(hw, res, 0, NULL);
>-
> 	/* there are some rare cases when trying to release the resource
> 	 * results in an admin queue timeout, so handle them correctly
> 	 */
>-	while ((status == -EIO) && (total_delay < ICE_CTL_Q_SQ_CMD_TIMEOUT))
>{
>-		mdelay(1);
>+	timeout = jiffies + 10 * ICE_CTL_Q_SQ_CMD_TIMEOUT;
>+	do {
> 		status = ice_aq_release_res(hw, res, 0, NULL);
>-		total_delay++;
>-	}
>+		if (status != -EIO)
>+			break;
>+		usleep_range(1000, 2000);
>+	} while (time_before(jiffies, timeout));
> }
>
> /**
>diff --git a/drivers/net/ethernet/intel/ice/ice_controlq.c
>b/drivers/net/ethernet/intel/ice/ice_controlq.c
>index c8fb10106ec3..d2faf1baad2f 100644
>--- a/drivers/net/ethernet/intel/ice/ice_controlq.c
>+++ b/drivers/net/ethernet/intel/ice/ice_controlq.c
>@@ -964,7 +964,7 @@ ice_sq_send_cmd(struct ice_hw *hw, struct
>ice_ctl_q_info *cq,
> 	struct ice_aq_desc *desc_on_ring;
> 	bool cmd_completed = false;
> 	struct ice_sq_cd *details;
>-	u32 total_delay = 0;
>+	unsigned long timeout;
> 	int status = 0;
> 	u16 retval = 0;
> 	u32 val = 0;
>@@ -1057,13 +1057,14 @@ ice_sq_send_cmd(struct ice_hw *hw, struct
>ice_ctl_q_info *cq,
> 		cq->sq.next_to_use = 0;
> 	wr32(hw, cq->sq.tail, cq->sq.next_to_use);
>
>+	timeout = jiffies + ICE_CTL_Q_SQ_CMD_TIMEOUT;
> 	do {
> 		if (ice_sq_done(hw, cq))
> 			break;
>
>-		udelay(ICE_CTL_Q_SQ_CMD_USEC);
>-		total_delay++;
>-	} while (total_delay < ICE_CTL_Q_SQ_CMD_TIMEOUT);
>+		usleep_range(ICE_CTL_Q_SQ_CMD_USEC,
>+			     ICE_CTL_Q_SQ_CMD_USEC * 3 / 2);
>+	} while (time_before(jiffies, timeout));
>
> 	/* if ready, copy the desc back to temp */
> 	if (ice_sq_done(hw, cq)) {
>diff --git a/drivers/net/ethernet/intel/ice/ice_controlq.h
>b/drivers/net/ethernet/intel/ice/ice_controlq.h
>index e790b2f4e437..950b7f4a7a05 100644
>--- a/drivers/net/ethernet/intel/ice/ice_controlq.h
>+++ b/drivers/net/ethernet/intel/ice/ice_controlq.h
>@@ -34,7 +34,7 @@ enum ice_ctl_q {
> };
>
> /* Control Queue timeout settings - max delay 1s */
>-#define ICE_CTL_Q_SQ_CMD_TIMEOUT	10000 /* Count 10000 times */
>+#define ICE_CTL_Q_SQ_CMD_TIMEOUT	HZ    /* Wait max 1s */
> #define ICE_CTL_Q_SQ_CMD_USEC		100   /* Check every 100usec */
> #define ICE_CTL_Q_ADMIN_INIT_TIMEOUT	10    /* Count 10 times */
> #define ICE_CTL_Q_ADMIN_INIT_MSEC	100   /* Check every 100msec */
>--
>2.39.2

Looks good, thank you Michal!

Reviewed-by: Arkadiusz Kubalewski <arkadiusz.kubalewski@...el.com>