[PATCH iwl-net 0/3] ice: restore timestamp config after reset

[PATCH iwl-net 0/3] ice: restore timestamp config after reset

[Jacob Keller]

We recently discovered during internal validation that the ice driver has
not been properly restoring Tx timestamp configuration after a device reset,
which resulted in application failures after a device reset.

After some digging, it turned out this problem is two-fold. Since the
introduction of the PTP support the driver has been clobbering the storage
of the current timestamp configuration during reset. Thus after a reset, the
driver will no longer perform Tx or Rx timestamps, and will report
timestamp configuration as disabled if SIOCGHWTSTAMP ioctl is issued.

In addition, the recently merged auxiliary bus support code missed that
PFINT_TSYN_MSK must be reprogrammed on the clock owner for E822 devices.
Failure to restore this register configuration results in the driver no
longer responding to interrupts from other ports. Depending on the traffic
pattern, this can either result in increased latency responding to
timestamps on the non-owner ports, or it can result in the driver never
reporting any timestamps. The configuration of PFINT_TSYN_MSK was only done
during initialization. Due to this, the Tx timestamp issue persists even if
userspace reconfigures timestamping.

This series fixes both issues, as well as removes a redundant Tx ring field
since we can rely on the skb flag as the primary detector for a Tx timestamp
request.

Note that I don't think this series will directly apply to older stable
releases (even v6.6) as we recently refactored a lot of the PTP code to
support auxiliary bus. Patch 2/3 only matters for the post-auxiliary bus
implementation. The principle of patch 1/3 and 3/3 could apply as far back
as the initial PTP support, but I don't think it will apply cleanly as-is.

This was originally posted to net-next before 6.6 release, and later posted
as a v2 to net. I'm re-posting this to Tony's Intel Wired LAN dev queue so
we can queue it up since its driver only changes.

Changes since v2
* rebased on to Tony's net dev-queue to send through IWL

net V2 was posted here:
https://lore.kernel.org/netdev/20231031222725.2819172-1-jacob.e.keller@intel.com/

Changes since v1
* Update target to net and rebase onto current net tree.
* Add appropriate fixes tag to 1/3.
* Slightly update the cover message.
* picked up Jesse's Reviewed-by tag.

net-next V1 was posted here:
https://lore.kernel.org/netdev/20231028002322.2224777-1-jacob.e.keller@intel.com/

Jacob Keller (3):
  ice: remove ptp_tx ring parameter flag
  ice: unify logic for programming PFINT_TSYN_MSK
  ice: restore timestamp configuration after device reset

 drivers/net/ethernet/intel/ice/ice_main.c |  12 +-
 drivers/net/ethernet/intel/ice/ice_ptp.c  | 146 ++++++++++++----------
 drivers/net/ethernet/intel/ice/ice_ptp.h  |   5 +-
 drivers/net/ethernet/intel/ice/ice_txrx.c |   3 -
 drivers/net/ethernet/intel/ice/ice_txrx.h |   1 -
 5 files changed, 84 insertions(+), 83 deletions(-)


base-commit: 1405b6c08fc9d3ba6c01de477556d127534ce52f
-- 
2.41.0

[PATCH iwl-net 1/3] ice: remove ptp_tx ring parameter flag

[Jacob Keller]

Before performing a Tx timestamp in ice_stamp(), the driver checks a ptp_tx
ring variable to see if timestamping is enabled on that ring. This value is
set for all rings whenever userspace configures Tx timestamping.

Ostensibly this was done to avoid wasting cycles checking other fields when
timestamping has not been enabled. However, for Tx timestamps we already
get an individual per-SKB flag indicating whether userspace wants to
request a timestamp on that packet. We do not gain much by also having
a separate flag to check for whether timestamping was enabled.

In fact, the driver currently fails to restore the field after a PF reset.
Because of this, if a PF reset occurs, timestamps will be disabled.

Since this flag doesn't add value in the hotpath, remove it and always
provide a timestamp if the SKB flag has been set.

A following change will fix the reset path to properly restore user
timestamping configuration completely.

This went unnoticed for some time because one of the most common
applications using Tx timestamps, ptp4l, will reconfigure the socket as
part of its fault recovery logic.

Fixes: ea9b847cda64 ("ice: enable transmit timestamps for E810 devices")
Signed-off-by: Jacob Keller <jacob.e.keller@intel.com>
Reviewed-by: Jesse Brandeburg <jesse.brandeburg@intel.com>
---
 drivers/net/ethernet/intel/ice/ice_ptp.c  | 14 --------------
 drivers/net/ethernet/intel/ice/ice_txrx.c |  3 ---
 drivers/net/ethernet/intel/ice/ice_txrx.h |  1 -
 3 files changed, 18 deletions(-)

diff --git a/drivers/net/ethernet/intel/ice/ice_ptp.c b/drivers/net/ethernet/intel/ice/ice_ptp.c
index 1eddcbe89b0c..affd90622a68 100644
--- a/drivers/net/ethernet/intel/ice/ice_ptp.c
+++ b/drivers/net/ethernet/intel/ice/ice_ptp.c
@@ -280,20 +280,6 @@ static void ice_ptp_configure_tx_tstamp(struct ice_pf *pf, bool on)
  */
 static void ice_set_tx_tstamp(struct ice_pf *pf, bool on)
 {
-	struct ice_vsi *vsi;
-	u16 i;
-
-	vsi = ice_get_main_vsi(pf);
-	if (!vsi)
-		return;
-
-	/* Set the timestamp enable flag for all the Tx rings */
-	ice_for_each_txq(vsi, i) {
-		if (!vsi->tx_rings[i])
-			continue;
-		vsi->tx_rings[i]->ptp_tx = on;
-	}
-
 	if (pf->ptp.tx_interrupt_mode == ICE_PTP_TX_INTERRUPT_SELF)
 		ice_ptp_configure_tx_tstamp(pf, on);
 
diff --git a/drivers/net/ethernet/intel/ice/ice_txrx.c b/drivers/net/ethernet/intel/ice/ice_txrx.c
index 52d0a126eb61..9e97ea863068 100644
--- a/drivers/net/ethernet/intel/ice/ice_txrx.c
+++ b/drivers/net/ethernet/intel/ice/ice_txrx.c
@@ -2306,9 +2306,6 @@ ice_tstamp(struct ice_tx_ring *tx_ring, struct sk_buff *skb,
 	if (likely(!(skb_shinfo(skb)->tx_flags & SKBTX_HW_TSTAMP)))
 		return;
 
-	if (!tx_ring->ptp_tx)
-		return;
-
 	/* Tx timestamps cannot be sampled when doing TSO */
 	if (first->tx_flags & ICE_TX_FLAGS_TSO)
 		return;
diff --git a/drivers/net/ethernet/intel/ice/ice_txrx.h b/drivers/net/ethernet/intel/ice/ice_txrx.h
index 166413fc33f4..daf7b9dbb143 100644
--- a/drivers/net/ethernet/intel/ice/ice_txrx.h
+++ b/drivers/net/ethernet/intel/ice/ice_txrx.h
@@ -380,7 +380,6 @@ struct ice_tx_ring {
 #define ICE_TX_FLAGS_RING_VLAN_L2TAG2	BIT(2)
 	u8 flags;
 	u8 dcb_tc;			/* Traffic class of ring */
-	u8 ptp_tx;
 } ____cacheline_internodealigned_in_smp;
 
 static inline bool ice_ring_uses_build_skb(struct ice_rx_ring *ring)
-- 
2.41.0

[PATCH iwl-net 2/3] ice: unify logic for programming PFINT_TSYN_MSK

[Jacob Keller]

Commit d938a8cca88a ("ice: Auxbus devices & driver for E822 TS") modified
how Tx timestamps are handled for E822 devices. On these devices, only the
clock owner handles reading the Tx timestamp data from firmware. To do
this, the PFINT_TSYN_MSK register is modified from the default value to one
which enables reacting to a Tx timestamp on all PHY ports.

The driver currently programs PFINT_TSYN_MSK in different places depending
on whether the port is the clock owner or not. For the clock owner, the
PFINT_TSYN_MSK value is programmed during ice_ptp_init_owner just before
calling ice_ptp_tx_ena_intr to program the PHY ports.

For the non-clock owner ports, the PFINT_TSYN_MSK is programmed during
ice_ptp_init_port.

If a large enough device reset occurs, the PFINT_TSYN_MSK register will be
reset to the default value in which only the PHY associated directly with
the PF will cause the Tx timestamp interrupt to trigger.

The driver lacks logic to reprogram the PFINT_TSYN_MSK register after a
device reset. For the E822 device, this results in the PF no longer
responding to interrupts for other ports. This results in failure to
deliver Tx timestamps to user space applications.

Rename ice_ptp_configure_tx_tstamp to ice_ptp_cfg_tx_interrupt, and unify
the logic for programming PFINT_TSYN_MSK and PFINT_OICR_ENA into one place.
This function will program both registers according to the combination of
user configuration and device requirements.

This ensures that PFINT_TSYN_MSK is always restored when we configure the
Tx timestamp interrupt.

Fixes: d938a8cca88a ("ice: Auxbus devices & driver for E822 TS")
Signed-off-by: Jacob Keller <jacob.e.keller@intel.com>
Reviewed-by: Jesse Brandeburg <jesse.brandeburg@intel.com>
---
 drivers/net/ethernet/intel/ice/ice_ptp.c | 60 ++++++++++++++----------
 1 file changed, 34 insertions(+), 26 deletions(-)

diff --git a/drivers/net/ethernet/intel/ice/ice_ptp.c b/drivers/net/ethernet/intel/ice/ice_ptp.c
index affd90622a68..624d05b4bbd9 100644
--- a/drivers/net/ethernet/intel/ice/ice_ptp.c
+++ b/drivers/net/ethernet/intel/ice/ice_ptp.c
@@ -256,21 +256,42 @@ ice_verify_pin_e810t(struct ptp_clock_info *info, unsigned int pin,
 }
 
 /**
- * ice_ptp_configure_tx_tstamp - Enable or disable Tx timestamp interrupt
- * @pf: The PF pointer to search in
- * @on: bool value for whether timestamp interrupt is enabled or disabled
+ * ice_ptp_cfg_tx_interrupt - Configure Tx timestamp interrupt for the device
+ * @pf: Board private structure
+ *
+ * Program the device to respond appropriately to the Tx timestamp interrupt
+ * cause.
  */
-static void ice_ptp_configure_tx_tstamp(struct ice_pf *pf, bool on)
+static void ice_ptp_cfg_tx_interrupt(struct ice_pf *pf)
 {
+	struct ice_hw *hw = &pf->hw;
+	bool enable;
 	u32 val;
 
+	switch (pf->ptp.tx_interrupt_mode) {
+	case ICE_PTP_TX_INTERRUPT_ALL:
+		/* React to interrupts across all quads. */
+		wr32(hw, PFINT_TSYN_MSK + (0x4 * hw->pf_id), (u32)0x1f);
+		enable = true;
+		break;
+	case ICE_PTP_TX_INTERRUPT_NONE:
+		/* Do not react to interrupts on any quad. */
+		wr32(hw, PFINT_TSYN_MSK + (0x4 * hw->pf_id), (u32)0x0);
+		enable = false;
+		break;
+	case ICE_PTP_TX_INTERRUPT_SELF:
+	default:
+		enable = pf->ptp.tstamp_config.tx_type == HWTSTAMP_TX_ON;
+		break;
+	}
+
 	/* Configure the Tx timestamp interrupt */
-	val = rd32(&pf->hw, PFINT_OICR_ENA);
-	if (on)
+	val = rd32(hw, PFINT_OICR_ENA);
+	if (enable)
 		val |= PFINT_OICR_TSYN_TX_M;
 	else
 		val &= ~PFINT_OICR_TSYN_TX_M;
-	wr32(&pf->hw, PFINT_OICR_ENA, val);
+	wr32(hw, PFINT_OICR_ENA, val);
 }
 
 /**
@@ -280,10 +301,9 @@ static void ice_ptp_configure_tx_tstamp(struct ice_pf *pf, bool on)
  */
 static void ice_set_tx_tstamp(struct ice_pf *pf, bool on)
 {
-	if (pf->ptp.tx_interrupt_mode == ICE_PTP_TX_INTERRUPT_SELF)
-		ice_ptp_configure_tx_tstamp(pf, on);
-
 	pf->ptp.tstamp_config.tx_type = on ? HWTSTAMP_TX_ON : HWTSTAMP_TX_OFF;
+
+	ice_ptp_cfg_tx_interrupt(pf);
 }
 
 /**
@@ -2789,15 +2809,7 @@ static int ice_ptp_init_owner(struct ice_pf *pf)
 	/* Release the global hardware lock */
 	ice_ptp_unlock(hw);
 
-	if (pf->ptp.tx_interrupt_mode == ICE_PTP_TX_INTERRUPT_ALL) {
-		/* The clock owner for this device type handles the timestamp
-		 * interrupt for all ports.
-		 */
-		ice_ptp_configure_tx_tstamp(pf, true);
-
-		/* React on all quads interrupts for E82x */
-		wr32(hw, PFINT_TSYN_MSK + (0x4 * hw->pf_id), (u32)0x1f);
-
+	if (!ice_is_e810(hw)) {
 		/* Enable quad interrupts */
 		err = ice_ptp_tx_ena_intr(pf, true, itr);
 		if (err)
@@ -2867,13 +2879,6 @@ static int ice_ptp_init_port(struct ice_pf *pf, struct ice_ptp_port *ptp_port)
 	case ICE_PHY_E810:
 		return ice_ptp_init_tx_e810(pf, &ptp_port->tx);
 	case ICE_PHY_E822:
-		/* Non-owner PFs don't react to any interrupts on E82x,
-		 * neither on own quad nor on others
-		 */
-		if (!ice_ptp_pf_handles_tx_interrupt(pf)) {
-			ice_ptp_configure_tx_tstamp(pf, false);
-			wr32(hw, PFINT_TSYN_MSK + (0x4 * hw->pf_id), (u32)0x0);
-		}
 		kthread_init_delayed_work(&ptp_port->ov_work,
 					  ice_ptp_wait_for_offsets);
 
@@ -3018,6 +3023,9 @@ void ice_ptp_init(struct ice_pf *pf)
 	/* Start the PHY timestamping block */
 	ice_ptp_reset_phy_timestamping(pf);
 
+	/* Configure initial Tx interrupt settings */
+	ice_ptp_cfg_tx_interrupt(pf);
+
 	set_bit(ICE_FLAG_PTP, pf->flags);
 	err = ice_ptp_init_work(pf, ptp);
 	if (err)
-- 
2.41.0

[PATCH iwl-net 3/3] ice: restore timestamp configuration after device reset

[Jacob Keller]

The driver calls ice_ptp_cfg_timestamp() during ice_ptp_prepare_for_reset()
to disable timestamping while the device is resetting. This operation
destroys the user requested configuration. While the driver does call
ice_ptp_cfg_timestamp in ice_rebuild() to restore some hardware settings
after a reset, it unconditionally passes true or false, resulting in
failure to restore previous user space configuration.

This results in a device reset forcibly disabling timestamp configuration
regardless of current user settings.

This was not detected previously due to a quirk of the LinuxPTP ptp4l
application. If ptp4l detects a missing timestamp, it enters a fault state
and performs recovery logic which includes executing SIOCSHWTSTAMP again,
restoring the now accidentally cleared configuration.

Not every application does this, and for these applications, timestamps
will mysteriously stop after a PF reset, without being restored until an
application restart.

Fix this by replacing ice_ptp_cfg_timestamp() with two new functions:

1) ice_ptp_disable_timestamp_mode() which unconditionally disables the
   timestamping logic in ice_ptp_prepare_for_reset() and ice_ptp_release()

2) ice_ptp_restore_timestamp_mode() which calls
   ice_ptp_restore_tx_interrupt() to restore Tx timestamping configuration,
   calls ice_set_rx_tstamp() to restore Rx timestamping configuration, and
   issues an immediate TSYN_TX interrupt to ensure that timestamps which
   may have occurred during the device reset get processed.

Modify the ice_ptp_set_timestamp_mode to directly save the user
configuration and then call ice_ptp_restore_timestamp_mode. This way, reset
no longer destroys the saved user configuration.

This obsoletes the ice_set_tx_tstamp() function which can now be safely
removed.

With this change, all devices should now restore Tx and Rx timestamping
functionality correctly after a PF reset without application intervention.

Fixes: 77a781155a65 ("ice: enable receive hardware timestamping")
Fixes: ea9b847cda64 ("ice: enable transmit timestamps for E810 devices")
Signed-off-by: Jacob Keller <jacob.e.keller@intel.com>
Reviewed-by: Jesse Brandeburg <jesse.brandeburg@intel.com>
---
 drivers/net/ethernet/intel/ice/ice_main.c | 12 +---
 drivers/net/ethernet/intel/ice/ice_ptp.c  | 76 ++++++++++++++---------
 drivers/net/ethernet/intel/ice/ice_ptp.h  |  5 +-
 3 files changed, 52 insertions(+), 41 deletions(-)

diff --git a/drivers/net/ethernet/intel/ice/ice_main.c b/drivers/net/ethernet/intel/ice/ice_main.c
index 6607fa6fe556..fb9c93f37e84 100644
--- a/drivers/net/ethernet/intel/ice/ice_main.c
+++ b/drivers/net/ethernet/intel/ice/ice_main.c
@@ -7401,15 +7401,6 @@ static void ice_rebuild(struct ice_pf *pf, enum ice_reset_req reset_type)
 		goto err_vsi_rebuild;
 	}
 
-	/* configure PTP timestamping after VSI rebuild */
-	if (test_bit(ICE_FLAG_PTP_SUPPORTED, pf->flags)) {
-		if (pf->ptp.tx_interrupt_mode == ICE_PTP_TX_INTERRUPT_SELF)
-			ice_ptp_cfg_timestamp(pf, false);
-		else if (pf->ptp.tx_interrupt_mode == ICE_PTP_TX_INTERRUPT_ALL)
-			/* for E82x PHC owner always need to have interrupts */
-			ice_ptp_cfg_timestamp(pf, true);
-	}
-
 	err = ice_vsi_rebuild_by_type(pf, ICE_VSI_SWITCHDEV_CTRL);
 	if (err) {
 		dev_err(dev, "Switchdev CTRL VSI rebuild failed: %d\n", err);
@@ -7461,6 +7452,9 @@ static void ice_rebuild(struct ice_pf *pf, enum ice_reset_req reset_type)
 	ice_plug_aux_dev(pf);
 	if (ice_is_feature_supported(pf, ICE_F_SRIOV_LAG))
 		ice_lag_rebuild(pf);
+
+	/* Restore timestamp mode settings after VSI rebuild */
+	ice_ptp_restore_timestamp_mode(pf);
 	return;
 
 err_vsi_rebuild:
diff --git a/drivers/net/ethernet/intel/ice/ice_ptp.c b/drivers/net/ethernet/intel/ice/ice_ptp.c
index 624d05b4bbd9..71f405f8a6fe 100644
--- a/drivers/net/ethernet/intel/ice/ice_ptp.c
+++ b/drivers/net/ethernet/intel/ice/ice_ptp.c
@@ -294,18 +294,6 @@ static void ice_ptp_cfg_tx_interrupt(struct ice_pf *pf)
 	wr32(hw, PFINT_OICR_ENA, val);
 }
 
-/**
- * ice_set_tx_tstamp - Enable or disable Tx timestamping
- * @pf: The PF pointer to search in
- * @on: bool value for whether timestamps are enabled or disabled
- */
-static void ice_set_tx_tstamp(struct ice_pf *pf, bool on)
-{
-	pf->ptp.tstamp_config.tx_type = on ? HWTSTAMP_TX_ON : HWTSTAMP_TX_OFF;
-
-	ice_ptp_cfg_tx_interrupt(pf);
-}
-
 /**
  * ice_set_rx_tstamp - Enable or disable Rx timestamping
  * @pf: The PF pointer to search in
@@ -317,7 +305,7 @@ static void ice_set_rx_tstamp(struct ice_pf *pf, bool on)
 	u16 i;
 
 	vsi = ice_get_main_vsi(pf);
-	if (!vsi)
+	if (!vsi || !vsi->rx_rings)
 		return;
 
 	/* Set the timestamp flag for all the Rx rings */
@@ -326,23 +314,50 @@ static void ice_set_rx_tstamp(struct ice_pf *pf, bool on)
 			continue;
 		vsi->rx_rings[i]->ptp_rx = on;
 	}
-
-	pf->ptp.tstamp_config.rx_filter = on ? HWTSTAMP_FILTER_ALL :
-					       HWTSTAMP_FILTER_NONE;
 }
 
 /**
- * ice_ptp_cfg_timestamp - Configure timestamp for init/deinit
+ * ice_ptp_disable_timestamp_mode - Disable current timestamp mode
  * @pf: Board private structure
- * @ena: bool value to enable or disable time stamp
  *
- * This function will configure timestamping during PTP initialization
- * and deinitialization
+ * Called during preparation for reset to temporarily disable timestamping on
+ * the device. Called during remove to disable timestamping while cleaning up
+ * driver resources.
  */
-void ice_ptp_cfg_timestamp(struct ice_pf *pf, bool ena)
+static void ice_ptp_disable_timestamp_mode(struct ice_pf *pf)
 {
-	ice_set_tx_tstamp(pf, ena);
-	ice_set_rx_tstamp(pf, ena);
+	struct ice_hw *hw = &pf->hw;
+	u32 val;
+
+	val = rd32(hw, PFINT_OICR_ENA);
+	val &= ~PFINT_OICR_TSYN_TX_M;
+	wr32(hw, PFINT_OICR_ENA, val);
+
+	ice_set_rx_tstamp(pf, false);
+}
+
+/**
+ * ice_ptp_restore_timestamp_mode - Restore timestamp configuration
+ * @pf: Board private structure
+ *
+ * Called at the end of rebuild to restore timestamp configuration after
+ * a device reset.
+ */
+void ice_ptp_restore_timestamp_mode(struct ice_pf *pf)
+{
+	struct ice_hw *hw = &pf->hw;
+	bool enable_rx;
+
+	ice_ptp_cfg_tx_interrupt(pf);
+
+	enable_rx = pf->ptp.tstamp_config.rx_filter == HWTSTAMP_FILTER_ALL;
+	ice_set_rx_tstamp(pf, enable_rx);
+
+	/* Trigger an immediate software interrupt to ensure that timestamps
+	 * which occurred during reset are handled now.
+	 */
+	wr32(hw, PFINT_OICR, PFINT_OICR_TSYN_TX_M);
+	ice_flush(hw);
 }
 
 /**
@@ -2043,10 +2058,10 @@ ice_ptp_set_timestamp_mode(struct ice_pf *pf, struct hwtstamp_config *config)
 {
 	switch (config->tx_type) {
 	case HWTSTAMP_TX_OFF:
-		ice_set_tx_tstamp(pf, false);
+		pf->ptp.tstamp_config.tx_type = HWTSTAMP_TX_OFF;
 		break;
 	case HWTSTAMP_TX_ON:
-		ice_set_tx_tstamp(pf, true);
+		pf->ptp.tstamp_config.tx_type = HWTSTAMP_TX_ON;
 		break;
 	default:
 		return -ERANGE;
@@ -2054,7 +2069,7 @@ ice_ptp_set_timestamp_mode(struct ice_pf *pf, struct hwtstamp_config *config)
 
 	switch (config->rx_filter) {
 	case HWTSTAMP_FILTER_NONE:
-		ice_set_rx_tstamp(pf, false);
+		pf->ptp.tstamp_config.rx_filter = HWTSTAMP_FILTER_NONE;
 		break;
 	case HWTSTAMP_FILTER_PTP_V1_L4_EVENT:
 	case HWTSTAMP_FILTER_PTP_V1_L4_SYNC:
@@ -2070,12 +2085,15 @@ ice_ptp_set_timestamp_mode(struct ice_pf *pf, struct hwtstamp_config *config)
 	case HWTSTAMP_FILTER_PTP_V2_L4_DELAY_REQ:
 	case HWTSTAMP_FILTER_NTP_ALL:
 	case HWTSTAMP_FILTER_ALL:
-		ice_set_rx_tstamp(pf, true);
+		pf->ptp.tstamp_config.rx_filter = HWTSTAMP_FILTER_ALL;
 		break;
 	default:
 		return -ERANGE;
 	}
 
+	/* Immediately update the device timestamping mode */
+	ice_ptp_restore_timestamp_mode(pf);
+
 	return 0;
 }
 
@@ -2743,7 +2761,7 @@ void ice_ptp_prepare_for_reset(struct ice_pf *pf)
 	clear_bit(ICE_FLAG_PTP, pf->flags);
 
 	/* Disable timestamping for both Tx and Rx */
-	ice_ptp_cfg_timestamp(pf, false);
+	ice_ptp_disable_timestamp_mode(pf);
 
 	kthread_cancel_delayed_work_sync(&ptp->work);
 
@@ -3061,7 +3079,7 @@ void ice_ptp_release(struct ice_pf *pf)
 		return;
 
 	/* Disable timestamping for both Tx and Rx */
-	ice_ptp_cfg_timestamp(pf, false);
+	ice_ptp_disable_timestamp_mode(pf);
 
 	ice_ptp_remove_auxbus_device(pf);
 
diff --git a/drivers/net/ethernet/intel/ice/ice_ptp.h b/drivers/net/ethernet/intel/ice/ice_ptp.h
index 8f6f94392756..06a330867fc9 100644
--- a/drivers/net/ethernet/intel/ice/ice_ptp.h
+++ b/drivers/net/ethernet/intel/ice/ice_ptp.h
@@ -292,7 +292,7 @@ int ice_ptp_clock_index(struct ice_pf *pf);
 struct ice_pf;
 int ice_ptp_set_ts_config(struct ice_pf *pf, struct ifreq *ifr);
 int ice_ptp_get_ts_config(struct ice_pf *pf, struct ifreq *ifr);
-void ice_ptp_cfg_timestamp(struct ice_pf *pf, bool ena);
+void ice_ptp_restore_timestamp_mode(struct ice_pf *pf);
 
 void ice_ptp_extts_event(struct ice_pf *pf);
 s8 ice_ptp_request_ts(struct ice_ptp_tx *tx, struct sk_buff *skb);
@@ -317,8 +317,7 @@ static inline int ice_ptp_get_ts_config(struct ice_pf *pf, struct ifreq *ifr)
 	return -EOPNOTSUPP;
 }
 
-static inline void ice_ptp_cfg_timestamp(struct ice_pf *pf, bool ena) { }
-
+static inline void ice_ptp_restore_timestamp_mode(struct ice_pf *pf) { }
 static inline void ice_ptp_extts_event(struct ice_pf *pf) { }
 static inline s8
 ice_ptp_request_ts(struct ice_ptp_tx *tx, struct sk_buff *skb)
-- 
2.41.0

Re: [PATCH iwl-net 1/3] ice: remove ptp_tx ring parameter flag

[Simon Horman]

On Fri, Nov 03, 2023 at 04:46:56PM -0700, Jacob Keller wrote:
> Before performing a Tx timestamp in ice_stamp(), the driver checks a ptp_tx
> ring variable to see if timestamping is enabled on that ring. This value is
> set for all rings whenever userspace configures Tx timestamping.
> 
> Ostensibly this was done to avoid wasting cycles checking other fields when
> timestamping has not been enabled. However, for Tx timestamps we already
> get an individual per-SKB flag indicating whether userspace wants to
> request a timestamp on that packet. We do not gain much by also having
> a separate flag to check for whether timestamping was enabled.
> 
> In fact, the driver currently fails to restore the field after a PF reset.
> Because of this, if a PF reset occurs, timestamps will be disabled.
> 
> Since this flag doesn't add value in the hotpath, remove it and always
> provide a timestamp if the SKB flag has been set.
> 
> A following change will fix the reset path to properly restore user
> timestamping configuration completely.
> 
> This went unnoticed for some time because one of the most common
> applications using Tx timestamps, ptp4l, will reconfigure the socket as
> part of its fault recovery logic.
> 
> Fixes: ea9b847cda64 ("ice: enable transmit timestamps for E810 devices")
> Signed-off-by: Jacob Keller <jacob.e.keller@intel.com>
> Reviewed-by: Jesse Brandeburg <jesse.brandeburg@intel.com>

Reviewed-by: Simon Horman <horms@kernel.org>

Re: [PATCH iwl-net 2/3] ice: unify logic for programming PFINT_TSYN_MSK

[Simon Horman]

On Fri, Nov 03, 2023 at 04:46:57PM -0700, Jacob Keller wrote:
> Commit d938a8cca88a ("ice: Auxbus devices & driver for E822 TS") modified
> how Tx timestamps are handled for E822 devices. On these devices, only the
> clock owner handles reading the Tx timestamp data from firmware. To do
> this, the PFINT_TSYN_MSK register is modified from the default value to one
> which enables reacting to a Tx timestamp on all PHY ports.
> 
> The driver currently programs PFINT_TSYN_MSK in different places depending
> on whether the port is the clock owner or not. For the clock owner, the
> PFINT_TSYN_MSK value is programmed during ice_ptp_init_owner just before
> calling ice_ptp_tx_ena_intr to program the PHY ports.
> 
> For the non-clock owner ports, the PFINT_TSYN_MSK is programmed during
> ice_ptp_init_port.
> 
> If a large enough device reset occurs, the PFINT_TSYN_MSK register will be
> reset to the default value in which only the PHY associated directly with
> the PF will cause the Tx timestamp interrupt to trigger.
> 
> The driver lacks logic to reprogram the PFINT_TSYN_MSK register after a
> device reset. For the E822 device, this results in the PF no longer
> responding to interrupts for other ports. This results in failure to
> deliver Tx timestamps to user space applications.
> 
> Rename ice_ptp_configure_tx_tstamp to ice_ptp_cfg_tx_interrupt, and unify
> the logic for programming PFINT_TSYN_MSK and PFINT_OICR_ENA into one place.
> This function will program both registers according to the combination of
> user configuration and device requirements.
> 
> This ensures that PFINT_TSYN_MSK is always restored when we configure the
> Tx timestamp interrupt.
> 
> Fixes: d938a8cca88a ("ice: Auxbus devices & driver for E822 TS")
> Signed-off-by: Jacob Keller <jacob.e.keller@intel.com>
> Reviewed-by: Jesse Brandeburg <jesse.brandeburg@intel.com>

Reviewed-by: Simon Horman <horms@kernel.org>

Re: [PATCH iwl-net 3/3] ice: restore timestamp configuration after device reset

[Simon Horman]

On Fri, Nov 03, 2023 at 04:46:58PM -0700, Jacob Keller wrote:
> The driver calls ice_ptp_cfg_timestamp() during ice_ptp_prepare_for_reset()
> to disable timestamping while the device is resetting. This operation
> destroys the user requested configuration. While the driver does call
> ice_ptp_cfg_timestamp in ice_rebuild() to restore some hardware settings
> after a reset, it unconditionally passes true or false, resulting in
> failure to restore previous user space configuration.
> 
> This results in a device reset forcibly disabling timestamp configuration
> regardless of current user settings.
> 
> This was not detected previously due to a quirk of the LinuxPTP ptp4l
> application. If ptp4l detects a missing timestamp, it enters a fault state
> and performs recovery logic which includes executing SIOCSHWTSTAMP again,
> restoring the now accidentally cleared configuration.
> 
> Not every application does this, and for these applications, timestamps
> will mysteriously stop after a PF reset, without being restored until an
> application restart.
> 
> Fix this by replacing ice_ptp_cfg_timestamp() with two new functions:
> 
> 1) ice_ptp_disable_timestamp_mode() which unconditionally disables the
>    timestamping logic in ice_ptp_prepare_for_reset() and ice_ptp_release()
> 
> 2) ice_ptp_restore_timestamp_mode() which calls
>    ice_ptp_restore_tx_interrupt() to restore Tx timestamping configuration,
>    calls ice_set_rx_tstamp() to restore Rx timestamping configuration, and
>    issues an immediate TSYN_TX interrupt to ensure that timestamps which
>    may have occurred during the device reset get processed.
> 
> Modify the ice_ptp_set_timestamp_mode to directly save the user
> configuration and then call ice_ptp_restore_timestamp_mode. This way, reset
> no longer destroys the saved user configuration.
> 
> This obsoletes the ice_set_tx_tstamp() function which can now be safely
> removed.
> 
> With this change, all devices should now restore Tx and Rx timestamping
> functionality correctly after a PF reset without application intervention.
> 
> Fixes: 77a781155a65 ("ice: enable receive hardware timestamping")
> Fixes: ea9b847cda64 ("ice: enable transmit timestamps for E810 devices")
> Signed-off-by: Jacob Keller <jacob.e.keller@intel.com>
> Reviewed-by: Jesse Brandeburg <jesse.brandeburg@intel.com>

Reviewed-by: Simon Horman <horms@kernel.org>

RE: [Intel-wired-lan] [PATCH iwl-net 3/3] ice: restore timestamp configuration after device reset

[Pucha, HimasekharX Reddy]

> -----Original Message-----
> From: Intel-wired-lan <intel-wired-lan-bounces@osuosl.org> On Behalf Of Jacob Keller
> Sent: Saturday, November 4, 2023 5:17 AM
> To: Nguyen, Anthony L <anthony.l.nguyen@intel.com>
> Cc: Keller, Jacob E <jacob.e.keller@intel.com>; netdev@vger.kernel.org; Intel Wired LAN <intel-wired-lan@lists.osuosl.org>; Brandeburg, Jesse <jesse.brandeburg@intel.com>
> Subject: [Intel-wired-lan] [PATCH iwl-net 3/3] ice: restore timestamp configuration after device reset
>
> The driver calls ice_ptp_cfg_timestamp() during ice_ptp_prepare_for_reset()
> to disable timestamping while the device is resetting. This operation
> destroys the user requested configuration. While the driver does call
> ice_ptp_cfg_timestamp in ice_rebuild() to restore some hardware settings
> after a reset, it unconditionally passes true or false, resulting in
> failure to restore previous user space configuration.
>
> This results in a device reset forcibly disabling timestamp configuration
> regardless of current user settings.
>
> This was not detected previously due to a quirk of the LinuxPTP ptp4l
> application. If ptp4l detects a missing timestamp, it enters a fault state
> and performs recovery logic which includes executing SIOCSHWTSTAMP again,
> restoring the now accidentally cleared configuration.
>
> Not every application does this, and for these applications, timestamps
> will mysteriously stop after a PF reset, without being restored until an
> application restart.
>
> Fix this by replacing ice_ptp_cfg_timestamp() with two new functions:
>
> 1) ice_ptp_disable_timestamp_mode() which unconditionally disables the
>    timestamping logic in ice_ptp_prepare_for_reset() and ice_ptp_release()
>
> 2) ice_ptp_restore_timestamp_mode() which calls
>    ice_ptp_restore_tx_interrupt() to restore Tx timestamping configuration,
>    calls ice_set_rx_tstamp() to restore Rx timestamping configuration, and
>    issues an immediate TSYN_TX interrupt to ensure that timestamps which
>    may have occurred during the device reset get processed.
>
> Modify the ice_ptp_set_timestamp_mode to directly save the user
> configuration and then call ice_ptp_restore_timestamp_mode. This way, reset
> no longer destroys the saved user configuration.
>
> This obsoletes the ice_set_tx_tstamp() function which can now be safely
> removed.
>
> With this change, all devices should now restore Tx and Rx timestamping
> functionality correctly after a PF reset without application intervention.
>
> Fixes: 77a781155a65 ("ice: enable receive hardware timestamping")
> Fixes: ea9b847cda64 ("ice: enable transmit timestamps for E810 devices")
> Signed-off-by: Jacob Keller <jacob.e.keller@intel.com>
> Reviewed-by: Jesse Brandeburg <jesse.brandeburg@intel.com>
>  ---
>  drivers/net/ethernet/intel/ice/ice_main.c | 12 +---
>  drivers/net/ethernet/intel/ice/ice_ptp.c  | 76 ++++++++++++++---------
>  drivers/net/ethernet/intel/ice/ice_ptp.h  |  5 +-
>  3 files changed, 52 insertions(+), 41 deletions(-)
>

Tested-by: Pucha Himasekhar Reddy <himasekharx.reddy.pucha@intel.com> (A Contingent worker at Intel)

RE: [Intel-wired-lan] [PATCH iwl-net 2/3] ice: unify logic for programming PFINT_TSYN_MSK

[Pucha, HimasekharX Reddy]

> -----Original Message-----
> From: Intel-wired-lan <intel-wired-lan-bounces@osuosl.org> On Behalf Of Jacob Keller
> Sent: Saturday, November 4, 2023 5:17 AM
> To: Nguyen, Anthony L <anthony.l.nguyen@intel.com>
> Cc: Keller, Jacob E <jacob.e.keller@intel.com>; netdev@vger.kernel.org; Intel Wired LAN <intel-wired-lan@lists.osuosl.org>; Brandeburg, Jesse <jesse.brandeburg@intel.com>
> Subject: [Intel-wired-lan] [PATCH iwl-net 2/3] ice: unify logic for programming PFINT_TSYN_MSK
>
> Commit d938a8cca88a ("ice: Auxbus devices & driver for E822 TS") modified
> how Tx timestamps are handled for E822 devices. On these devices, only the
> clock owner handles reading the Tx timestamp data from firmware. To do
> this, the PFINT_TSYN_MSK register is modified from the default value to one
> which enables reacting to a Tx timestamp on all PHY ports.
>
> The driver currently programs PFINT_TSYN_MSK in different places depending
> on whether the port is the clock owner or not. For the clock owner, the
> PFINT_TSYN_MSK value is programmed during ice_ptp_init_owner just before
> calling ice_ptp_tx_ena_intr to program the PHY ports.
>
> For the non-clock owner ports, the PFINT_TSYN_MSK is programmed during
> ice_ptp_init_port.
>
> If a large enough device reset occurs, the PFINT_TSYN_MSK register will be
> reset to the default value in which only the PHY associated directly with
> the PF will cause the Tx timestamp interrupt to trigger.
>
> The driver lacks logic to reprogram the PFINT_TSYN_MSK register after a
> device reset. For the E822 device, this results in the PF no longer
> responding to interrupts for other ports. This results in failure to
> deliver Tx timestamps to user space applications.
>
> Rename ice_ptp_configure_tx_tstamp to ice_ptp_cfg_tx_interrupt, and unify
> the logic for programming PFINT_TSYN_MSK and PFINT_OICR_ENA into one place.
> This function will program both registers according to the combination of
> user configuration and device requirements.
>
> This ensures that PFINT_TSYN_MSK is always restored when we configure the
> Tx timestamp interrupt.
>
> Fixes: d938a8cca88a ("ice: Auxbus devices & driver for E822 TS")
> Signed-off-by: Jacob Keller <jacob.e.keller@intel.com>
> Reviewed-by: Jesse Brandeburg <jesse.brandeburg@intel.com>
> ---
>  drivers/net/ethernet/intel/ice/ice_ptp.c | 60 ++++++++++++++----------
>  1 file changed, 34 insertions(+), 26 deletions(-)
>

Tested-by: Pucha Himasekhar Reddy <himasekharx.reddy.pucha@intel.com> (A Contingent worker at Intel)

RE: [Intel-wired-lan] [PATCH iwl-net 1/3] ice: remove ptp_tx ring parameter flag

[Pucha, HimasekharX Reddy]

> -----Original Message-----
> From: Intel-wired-lan <intel-wired-lan-bounces@osuosl.org> On Behalf Of Jacob Keller
> Sent: Saturday, November 4, 2023 5:17 AM
> To: Nguyen, Anthony L <anthony.l.nguyen@intel.com>
> Cc: Keller, Jacob E <jacob.e.keller@intel.com>; netdev@vger.kernel.org; Intel Wired LAN <intel-wired-lan@lists.osuosl.org>; Brandeburg, Jesse <jesse.brandeburg@intel.com>
> Subject: [Intel-wired-lan] [PATCH iwl-net 1/3] ice: remove ptp_tx ring parameter flag
>
> Before performing a Tx timestamp in ice_stamp(), the driver checks a ptp_tx
> ring variable to see if timestamping is enabled on that ring. This value is
> set for all rings whenever userspace configures Tx timestamping.
>
> Ostensibly this was done to avoid wasting cycles checking other fields when
> timestamping has not been enabled. However, for Tx timestamps we already
> get an individual per-SKB flag indicating whether userspace wants to
> request a timestamp on that packet. We do not gain much by also having
> a separate flag to check for whether timestamping was enabled.
>
> In fact, the driver currently fails to restore the field after a PF reset.
> Because of this, if a PF reset occurs, timestamps will be disabled.
>
> Since this flag doesn't add value in the hotpath, remove it and always
> provide a timestamp if the SKB flag has been set.
>
> A following change will fix the reset path to properly restore user
> timestamping configuration completely.
>
> This went unnoticed for some time because one of the most common
> applications using Tx timestamps, ptp4l, will reconfigure the socket as
> part of its fault recovery logic.
>
> Fixes: ea9b847cda64 ("ice: enable transmit timestamps for E810 devices")
> Signed-off-by: Jacob Keller <jacob.e.keller@intel.com>
> Reviewed-by: Jesse Brandeburg <jesse.brandeburg@intel.com>
> ---
>  drivers/net/ethernet/intel/ice/ice_ptp.c  | 14 --------------
>  drivers/net/ethernet/intel/ice/ice_txrx.c |  3 ---
>  drivers/net/ethernet/intel/ice/ice_txrx.h |  1 -
>  3 files changed, 18 deletions(-)
>

Tested-by: Pucha Himasekhar Reddy <himasekharx.reddy.pucha@intel.com> (A Contingent worker at Intel)