[PATCH net-next 00/15][pull request] Intel Wired LAN Driver Updates 2026-06-09 (idpf, ice, i40e, iavf, ixgbe, igc, igb, e1000e, e1000)

[PATCH net-next 00/15][pull request] Intel Wired LAN Driver Updates 2026-06-09 (idpf, ice, i40e, iavf, ixgbe, igc, igb, e1000e, e1000)

[Tony Nguyen]

Marco Crivellari replaces obsolete use of system_unbound_wq to
system_dfl_wq for idpf.

Natalia removes redundant PTP checks on ice.

Corinna Vinschen removes redundant MAC address check on iavf.

Jakub Raczynski replaces open-coded array size calculation to use
ARRAY_SIZE for i40e and iavf drivers.

Piotr removes a couple redundant assignments on ixgbe.

Kurt Kanzenbach moves processing of Tx timestamps from system workqueue
to BH workqueue for igb.

Alex utilizes ktime_get_* helpers for igb and e1000e.

Daiki Harada replaces napi_schedule() to, more appropriate,
napi_schedule_irqoff() call in igb and igc.

Matt Vollrath does the same on e1000e.

Agalakov Daniil skips unnecessary endian conversions on e1000e and
e1000.

Maximilian Pezzullo fixes some typos on igb and igc.

The following are changes since commit 9415471e01c1aaac43daa6af3a261dc0c6c3a47c:
  selftests/net/openvswitch: guard command substitutions against empty output
and are available in the git repository at:
  git://git.kernel.org/pub/scm/linux/kernel/git/tnguy/next-queue 1GbE

Agalakov Daniil (2):
  e1000: limit endianness conversion to boundary words
  e1000e: limit endianness conversion to boundary words

Aleksandr Loktionov (2):
  igb: use ktime_get_real helpers in igb_ptp_reset()
  e1000e: use ktime_get_real_ns() in e1000e_systim_reset()

Corinna Vinschen (1):
  iavf: iavf_virtchnl_completion: drop duplicate ether_addr_equal() test

Daiki Harada (2):
  igb: use napi_schedule_irqoff() instead of napi_schedule()
  igc: use napi_schedule_irqoff() instead of napi_schedule()

Jakub Raczynski (1):
  net/intel: Replace manual array size calculation with ARRAY_SIZE

Kurt Kanzenbach (1):
  igb: Retrieve Tx timestamp from BH workqueue

Marco Crivellari (1):
  idpf: Replace use of system_unbound_wq with system_dfl_wq

Matt Vollrath (1):
  e1000e: Use __napi_schedule_irqoff()

Maximilian Pezzullo (2):
  igb: fix typos in comments
  igc: fix typos in comments

Natalia Wochtman (1):
  ice: remove redundant checks from PTP init

Piotr Kwapulinski (1):
  ixgbe: e610: remove redundant assignment

 .../net/ethernet/intel/e1000/e1000_ethtool.c  | 10 ++++++----
 drivers/net/ethernet/intel/e1000e/ethtool.c   | 19 ++++++++++++-------
 drivers/net/ethernet/intel/e1000e/netdev.c    |  8 ++++----
 drivers/net/ethernet/intel/i40e/i40e_adminq.h |  2 +-
 drivers/net/ethernet/intel/iavf/iavf_adminq.h |  2 +-
 .../net/ethernet/intel/iavf/iavf_virtchnl.c   | 12 +++++-------
 drivers/net/ethernet/intel/ice/ice_ptp.c      | 14 ++++----------
 drivers/net/ethernet/intel/idpf/idpf_txrx.c   |  2 +-
 .../net/ethernet/intel/igb/e1000_defines.h    |  2 +-
 drivers/net/ethernet/intel/igb/e1000_mac.c    |  2 +-
 drivers/net/ethernet/intel/igb/e1000_mbx.h    |  2 +-
 drivers/net/ethernet/intel/igb/e1000_nvm.c    |  2 +-
 drivers/net/ethernet/intel/igb/igb_main.c     | 10 +++++-----
 drivers/net/ethernet/intel/igb/igb_ptp.c      |  7 ++++---
 drivers/net/ethernet/intel/igc/igc_diag.c     |  2 +-
 drivers/net/ethernet/intel/igc/igc_main.c     | 10 +++++-----
 drivers/net/ethernet/intel/ixgbe/ixgbe_e610.c | 14 ++++----------
 17 files changed, 57 insertions(+), 63 deletions(-)

-- 
2.47.1

[PATCH net-next 01/15] idpf: Replace use of system_unbound_wq with system_dfl_wq

[Tony Nguyen]

From: Marco Crivellari <marco.crivellari@suse.com>

This patch continues the effort to refactor workqueue APIs, which has begun
with the changes introducing new workqueues and a new alloc_workqueue flag:

   commit 128ea9f6ccfb ("workqueue: Add system_percpu_wq and system_dfl_wq")
   commit 930c2ea566af ("workqueue: Add new WQ_PERCPU flag")

The point of the refactoring is to eventually alter the default behavior of
workqueues to become unbound by default so that their workload placement is
optimized by the scheduler.

Before that to happen, workqueue users must be converted to the better named
new workqueues with no intended behaviour changes:

   system_wq -> system_percpu_wq
   system_unbound_wq -> system_dfl_wq

This way the old obsolete workqueues (system_wq, system_unbound_wq) can be
removed in the future.

Link: https://lore.kernel.org/all/20250221112003.1dSuoGyc@linutronix.de/
Suggested-by: Tejun Heo <tj@kernel.org>
Signed-off-by: Marco Crivellari <marco.crivellari@suse.com>
Reviewed-by: Przemek Kitszel <przemyslaw.kitszel@intel.com>
Tested-by: Samuel Salin <Samuel.salin@intel.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
---
 drivers/net/ethernet/intel/idpf/idpf_txrx.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/ethernet/intel/idpf/idpf_txrx.c b/drivers/net/ethernet/intel/idpf/idpf_txrx.c
index f6b3b15364ff..7f9056404f64 100644
--- a/drivers/net/ethernet/intel/idpf/idpf_txrx.c
+++ b/drivers/net/ethernet/intel/idpf/idpf_txrx.c
@@ -2022,7 +2022,7 @@ static void idpf_tx_read_tstamp(struct idpf_tx_queue *txq, struct sk_buff *skb)
 		/* Fetch timestamp from completion descriptor through
 		 * virtchnl msg to report to stack.
 		 */
-		queue_work(system_unbound_wq, txq->tstamp_task);
+		queue_work(system_dfl_wq, txq->tstamp_task);
 		break;
 	}
 
-- 
2.47.1

[PATCH net-next 02/15] ice: remove redundant checks from PTP init

[Tony Nguyen]

From: Natalia Wochtman <natalia.wochtman@intel.com>

Remove unnecessary condition checks in ice_ptp_setup_adapter() and
ice_ptp_init(). They are duplicated in ice_pf_src_tmr_owned().

Change ice_ptp_setup_adapter() to return void.

Reviewed-by: Przemek Kitszel <przemyslaw.kitszel@intel.com>
Reviewed-by: Aleksandr Loktionov <aleksandr.loktionov@intel.com>
Signed-off-by: Natalia Wochtman <natalia.wochtman@intel.com>
Tested-by: Rinitha S <sx.rinitha@intel.com> (A Contingent worker at Intel)
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
---
 drivers/net/ethernet/intel/ice/ice_ptp.c | 14 ++++----------
 1 file changed, 4 insertions(+), 10 deletions(-)

diff --git a/drivers/net/ethernet/intel/ice/ice_ptp.c b/drivers/net/ethernet/intel/ice/ice_ptp.c
index 36df742c326c..cca004265662 100644
--- a/drivers/net/ethernet/intel/ice/ice_ptp.c
+++ b/drivers/net/ethernet/intel/ice/ice_ptp.c
@@ -3059,14 +3059,9 @@ void ice_ptp_rebuild(struct ice_pf *pf, enum ice_reset_req reset_type)
 	dev_err(ice_pf_to_dev(pf), "PTP reset failed %d\n", err);
 }
 
-static int ice_ptp_setup_adapter(struct ice_pf *pf)
+static void ice_ptp_setup_adapter(struct ice_pf *pf)
 {
-	if (!ice_pf_src_tmr_owned(pf) || !ice_is_primary(&pf->hw))
-		return -EPERM;
-
 	pf->adapter->ctrl_pf = pf;
-
-	return 0;
 }
 
 static int ice_ptp_setup_pf(struct ice_pf *pf)
@@ -3309,10 +3304,9 @@ void ice_ptp_init(struct ice_pf *pf)
 	/* If this function owns the clock hardware, it must allocate and
 	 * configure the PTP clock device to represent it.
 	 */
-	if (ice_pf_src_tmr_owned(pf) && ice_is_primary(hw)) {
-		err = ice_ptp_setup_adapter(pf);
-		if (err)
-			goto err_exit;
+	if (ice_pf_src_tmr_owned(pf)) {
+		ice_ptp_setup_adapter(pf);
+
 		err = ice_ptp_init_owner(pf);
 		if (err)
 			goto err_exit;
-- 
2.47.1

[PATCH net-next 03/15] iavf: iavf_virtchnl_completion: drop duplicate ether_addr_equal() test

[Tony Nguyen]

From: Corinna Vinschen <vinschen@redhat.com>

This is just a simple cleanup fix.  Commit 35a2443d0910f ("iavf: Add
waiting for response from PF in set mac") introduced a duplicate
ether_addr_equal() check, so the current code tests the new MAC twice
against the former MAC.

Remove the outer ether_addr_equal() test, remnant of commit c5c922b3e09b
("iavf: fix MAC address setting for VFs when filter is rejected")

Signed-off-by: Corinna Vinschen <vinschen@redhat.com>
Reviewed-by: Aleksandr Loktionov <aleksandr.loktionov@intel.com>
Tested-by: Rafal Romanowski <rafal.romanowski@intel.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
---
 drivers/net/ethernet/intel/iavf/iavf_virtchnl.c | 12 +++++-------
 1 file changed, 5 insertions(+), 7 deletions(-)

diff --git a/drivers/net/ethernet/intel/iavf/iavf_virtchnl.c b/drivers/net/ethernet/intel/iavf/iavf_virtchnl.c
index 4f2defd2331b..ec234cc8bd9d 100644
--- a/drivers/net/ethernet/intel/iavf/iavf_virtchnl.c
+++ b/drivers/net/ethernet/intel/iavf/iavf_virtchnl.c
@@ -2550,13 +2550,11 @@ void iavf_virtchnl_completion(struct iavf_adapter *adapter,
 	case VIRTCHNL_OP_ADD_ETH_ADDR:
 		if (!v_retval)
 			iavf_mac_add_ok(adapter);
-		if (!ether_addr_equal(netdev->dev_addr, adapter->hw.mac.addr))
-			if (!ether_addr_equal(netdev->dev_addr,
-					      adapter->hw.mac.addr)) {
-				netif_addr_lock_bh(netdev);
-				eth_hw_addr_set(netdev, adapter->hw.mac.addr);
-				netif_addr_unlock_bh(netdev);
-			}
+		if (!ether_addr_equal(netdev->dev_addr, adapter->hw.mac.addr)) {
+			netif_addr_lock_bh(netdev);
+			eth_hw_addr_set(netdev, adapter->hw.mac.addr);
+			netif_addr_unlock_bh(netdev);
+		}
 		wake_up(&adapter->vc_waitqueue);
 		break;
 	case VIRTCHNL_OP_GET_STATS: {
-- 
2.47.1

[PATCH net-next 04/15] net/intel: Replace manual array size calculation with ARRAY_SIZE

[Tony Nguyen]

From: Jakub Raczynski <j.raczynski@samsung.com>

There are still places in the code where manual calculation of array size
exist, but it is good to enforce usage of single macro through the whole
code as it makes code bit more readable.
While at it, beautify condition surrounding it by reversing check and remove
unnecessary casting.

Signed-off-by: Jakub Raczynski <j.raczynski@samsung.com>
Reviewed-by: Dan Carpenter <error27@gmail.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
---
 drivers/net/ethernet/intel/i40e/i40e_adminq.h | 2 +-
 drivers/net/ethernet/intel/iavf/iavf_adminq.h | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/net/ethernet/intel/i40e/i40e_adminq.h b/drivers/net/ethernet/intel/i40e/i40e_adminq.h
index 1be97a3a86ce..dcf3baec7b73 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_adminq.h
+++ b/drivers/net/ethernet/intel/i40e/i40e_adminq.h
@@ -109,7 +109,7 @@ static inline int i40e_aq_rc_to_posix(int aq_ret, int aq_rc)
 		-EFBIG,      /* I40E_AQ_RC_EFBIG */
 	};
 
-	if (!((u32)aq_rc < (sizeof(aq_to_posix) / sizeof((aq_to_posix)[0]))))
+	if (aq_rc >= ARRAY_SIZE(aq_to_posix))
 		return -ERANGE;
 
 	return aq_to_posix[aq_rc];
diff --git a/drivers/net/ethernet/intel/iavf/iavf_adminq.h b/drivers/net/ethernet/intel/iavf/iavf_adminq.h
index bbf5c4b3a2ae..dd2f61172157 100644
--- a/drivers/net/ethernet/intel/iavf/iavf_adminq.h
+++ b/drivers/net/ethernet/intel/iavf/iavf_adminq.h
@@ -113,7 +113,7 @@ static inline int iavf_aq_rc_to_posix(int aq_ret, int aq_rc)
 	if (aq_ret == IAVF_ERR_ADMIN_QUEUE_TIMEOUT)
 		return -EAGAIN;
 
-	if (!((u32)aq_rc < (sizeof(aq_to_posix) / sizeof((aq_to_posix)[0]))))
+	if (aq_rc >= ARRAY_SIZE(aq_to_posix))
 		return -ERANGE;
 
 	return aq_to_posix[aq_rc];
-- 
2.47.1

[PATCH net-next 05/15] ixgbe: e610: remove redundant assignment

[Tony Nguyen]

From: Piotr Kwapulinski <piotr.kwapulinski@intel.com>

Remove unnecessary code. No functional impact.

Signed-off-by: Piotr Kwapulinski <piotr.kwapulinski@intel.com>
Reviewed-by: Aleksandr Loktionov <aleksandr.loktionov@intel.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
---
 drivers/net/ethernet/intel/ixgbe/ixgbe_e610.c | 14 ++++----------
 1 file changed, 4 insertions(+), 10 deletions(-)

diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_e610.c b/drivers/net/ethernet/intel/ixgbe/ixgbe_e610.c
index 665a9813e251..4d8ae5b56145 100644
--- a/drivers/net/ethernet/intel/ixgbe/ixgbe_e610.c
+++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_e610.c
@@ -142,20 +142,14 @@ static int ixgbe_aci_send_cmd_execute(struct ixgbe_hw *hw,
 			  IXGBE_PF_HICR);
 
 	/* Read sync Admin Command response */
-	if ((hicr & IXGBE_PF_HICR_SV)) {
-		for (i = 0; i < IXGBE_ACI_DESC_SIZE_IN_DWORDS; i++) {
+	if ((hicr & IXGBE_PF_HICR_SV))
+		for (i = 0; i < IXGBE_ACI_DESC_SIZE_IN_DWORDS; i++)
 			raw_desc[i] = IXGBE_READ_REG(hw, IXGBE_PF_HIDA(i));
-			raw_desc[i] = raw_desc[i];
-		}
-	}
 
 	/* Read async Admin Command response */
-	if ((hicr & IXGBE_PF_HICR_EV) && !(hicr & IXGBE_PF_HICR_C)) {
-		for (i = 0; i < IXGBE_ACI_DESC_SIZE_IN_DWORDS; i++) {
+	if ((hicr & IXGBE_PF_HICR_EV) && !(hicr & IXGBE_PF_HICR_C))
+		for (i = 0; i < IXGBE_ACI_DESC_SIZE_IN_DWORDS; i++)
 			raw_desc[i] = IXGBE_READ_REG(hw, IXGBE_PF_HIDA_2(i));
-			raw_desc[i] = raw_desc[i];
-		}
-	}
 
 	/* Handle timeout and invalid state of HICR register */
 	if (hicr & IXGBE_PF_HICR_C)
-- 
2.47.1

[PATCH net-next 06/15] igb: Retrieve Tx timestamp from BH workqueue

[Tony Nguyen]

From: Kurt Kanzenbach <kurt@linutronix.de>

Retrieve Tx timestamp from system BH instead of regular system workqueue.

The current implementation uses schedule_work() which is executed by the
system work queue and kworkers to retrieve Tx timestamps. This increases
latency and can lead to timeouts in case of heavy system load. i210 is
often used in industrial systems, where timestamp timeouts can be fatal.

Therefore, switch to the system BH workqueues which are executed in softirq
context shortly after the IRQ handler returns.

Tested between Intel i210 and i350 with ptp4l gPTP profile:

|ptp4l[30.405]: rms    4 max    7 freq +12825 +/-   3 delay   247 +/-   0
|ptp4l[31.406]: rms    2 max    3 freq +12829 +/-   3 delay   248 +/-   0
|ptp4l[32.406]: rms    3 max    3 freq +12827 +/-   3 delay   248 +/-   0
|ptp4l[33.406]: rms    2 max    3 freq +12827 +/-   3 delay   248 +/-   0
|ptp4l[34.407]: rms    3 max    6 freq +12825 +/-   4 delay   248 +/-   0
|ptp4l[35.407]: rms    3 max    6 freq +12822 +/-   4 delay   246 +/-   0
|ptp4l[36.407]: rms    7 max   10 freq +12812 +/-   5 delay   248 +/-   0
|ptp4l[37.408]: rms    5 max    8 freq +12808 +/-   3 delay   248 +/-   0

Furthermore, Miroslav Lichvar tested with ntpperf and chrony on Intel i350:

Without the patch:

|               |          responses            |        response time (ns)
|rate   clients |  lost invalid   basic  xleave |    min    mean     max stddev
|150000   15000   0.00%   0.00%   0.00% 100.00%    +4188  +36475 +193328  16179
|157500   15750   0.02%   0.00%   0.02%  99.96%    +6373  +42969 +683894  22682
|165375   16384   0.03%   0.00%   0.00%  99.97%    +7911  +43960 +692471  24454
|173643   16384   0.06%   0.00%   0.00%  99.94%    +8323  +45627 +707240  28452
|182325   16384   0.06%   0.00%   0.00%  99.94%    +8404  +47292 +722524  26936
|191441   16384   0.00%   0.00%   0.00% 100.00%    +8930  +51738 +223727  14272
|201013   16384   0.05%   0.00%   0.00%  99.95%    +9634  +53696 +776445  23783
|211063   16384   0.00%   0.00%   0.00% 100.00%   +14393  +54558 +329546  20473
|221616   16384   2.59%   0.00%   0.05%  97.36%   +23924 +321205 +518192  21838
|232696   16384   7.00%   0.00%   0.10%  92.90%   +33396 +337709 +575661  21017
|244330   16384  10.82%   0.00%   0.15%  89.03%   +34188 +340248 +556237  20880
|
|With the patch:
|150000   15000   5.11%   0.00%   0.00%  94.88%    +4426 +460642 +640884  83746
|157500   15750  11.54%   0.00%   0.26%  88.20%   +14434 +543656 +738355  30349
|165375   16384  15.61%   0.00%   0.31%  84.08%   +35822 +515304 +833859  25596
|173643   16384  19.58%   0.00%   0.37%  80.05%   +20762 +568962 +900100  28118
|182325   16384  23.46%   0.00%   0.42%  76.13%   +41829 +547974 +804170  27890
|191441   16384  27.23%   0.00%   0.46%  72.31%   +15182 +557920 +798212  28868
|201013   16384  30.51%   0.00%   0.49%  69.00%   +15980 +560764 +805576  29979
|211063   16384   0.06%   0.00%   0.00%  99.94%   +12668  +80487 +410555  62182
|221616   16384   2.94%   0.00%   0.05%  97.00%   +21587 +342769 +517566  23359
|232696   16384   6.94%   0.00%   0.10%  92.96%   +16581 +336068 +484574  18453
|244330   16384  11.45%   0.00%   0.14%  88.41%   +23608 +345023 +564130  19177

There are some minor differences at lower rates, but no performance
regressions at higher ones.

Reviewed-by: Paul Menzel <pmenzel@molgen.mpg.de>
Reviewed-by: Aleksandr Loktionov <aleksandr.loktionov@intel.com>
Signed-off-by: Kurt Kanzenbach <kurt@linutronix.de>
Tested-by: Rinitha S <sx.rinitha@intel.com> (A Contingent worker at Intel)
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
---
 drivers/net/ethernet/intel/igb/igb_main.c | 4 ++--
 drivers/net/ethernet/intel/igb/igb_ptp.c  | 2 +-
 2 files changed, 3 insertions(+), 3 deletions(-)

diff --git a/drivers/net/ethernet/intel/igb/igb_main.c b/drivers/net/ethernet/intel/igb/igb_main.c
index 49d46670695f..17eff90af9e3 100644
--- a/drivers/net/ethernet/intel/igb/igb_main.c
+++ b/drivers/net/ethernet/intel/igb/igb_main.c
@@ -6571,7 +6571,7 @@ netdev_tx_t igb_xmit_frame_ring(struct sk_buff *skb,
 			adapter->ptp_tx_skb = skb_get(skb);
 			adapter->ptp_tx_start = jiffies;
 			if (adapter->hw.mac.type == e1000_82576)
-				schedule_work(&adapter->ptp_tx_work);
+				queue_work(system_bh_wq, &adapter->ptp_tx_work);
 		} else {
 			adapter->tx_hwtstamp_skipped++;
 		}
@@ -7075,7 +7075,7 @@ static void igb_tsync_interrupt(struct igb_adapter *adapter)
 
 	if (tsicr & E1000_TSICR_TXTS) {
 		/* retrieve hardware timestamp */
-		schedule_work(&adapter->ptp_tx_work);
+		queue_work(system_bh_wq, &adapter->ptp_tx_work);
 	}
 
 	if (tsicr & TSINTR_TT0)
diff --git a/drivers/net/ethernet/intel/igb/igb_ptp.c b/drivers/net/ethernet/intel/igb/igb_ptp.c
index bd85d02ecadd..7b44f9090631 100644
--- a/drivers/net/ethernet/intel/igb/igb_ptp.c
+++ b/drivers/net/ethernet/intel/igb/igb_ptp.c
@@ -832,7 +832,7 @@ static void igb_ptp_tx_work(struct work_struct *work)
 		igb_ptp_tx_hwtstamp(adapter);
 	else
 		/* reschedule to check later */
-		schedule_work(&adapter->ptp_tx_work);
+		queue_work(system_bh_wq, &adapter->ptp_tx_work);
 }
 
 static void igb_ptp_overflow_check(struct work_struct *work)
-- 
2.47.1

[PATCH net-next 07/15] igb: use ktime_get_real helpers in igb_ptp_reset()

[Tony Nguyen]

From: Aleksandr Loktionov <aleksandr.loktionov@intel.com>

Replace ktime_to_ns(ktime_get_real()) with the direct equivalent
ktime_get_real_ns() and ktime_to_timespec64(ktime_get_real()) with
ktime_get_real_ts64() in igb_ptp_reset().  Using the combined helpers
makes the intent clearer.

Suggested-by: Jacob Keller <jacob.e.keller@intel.com>
Suggested-by: Simon Horman <horms@kernel.org>
Reviewed-by: Paul Menzel <pmenzel@molgen.mpg.de>
Signed-off-by: Aleksandr Loktionov <aleksandr.loktionov@intel.com>
Reviewed-by: Simon Horman <horms@kernel.org>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
---
 drivers/net/ethernet/intel/igb/igb_ptp.c | 5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)

diff --git a/drivers/net/ethernet/intel/igb/igb_ptp.c b/drivers/net/ethernet/intel/igb/igb_ptp.c
index 7b44f9090631..f63765e9249b 100644
--- a/drivers/net/ethernet/intel/igb/igb_ptp.c
+++ b/drivers/net/ethernet/intel/igb/igb_ptp.c
@@ -1500,12 +1500,13 @@ void igb_ptp_reset(struct igb_adapter *adapter)
 
 	/* Re-initialize the timer. */
 	if ((hw->mac.type == e1000_i210) || (hw->mac.type == e1000_i211)) {
-		struct timespec64 ts = ktime_to_timespec64(ktime_get_real());
+		struct timespec64 ts;
 
+		ktime_get_real_ts64(&ts);
 		igb_ptp_write_i210(adapter, &ts);
 	} else {
 		timecounter_init(&adapter->tc, &adapter->cc,
-				 ktime_to_ns(ktime_get_real()));
+				 ktime_get_real_ns());
 	}
 out:
 	spin_unlock_irqrestore(&adapter->tmreg_lock, flags);
-- 
2.47.1

[PATCH net-next 08/15] e1000e: use ktime_get_real_ns() in e1000e_systim_reset()

[Tony Nguyen]

From: Aleksandr Loktionov <aleksandr.loktionov@intel.com>

Replace ktime_to_ns(ktime_get_real()) with the direct equivalent
ktime_get_real_ns() in e1000e_systim_reset().  Using the combined helper
avoids the unnecessary intermediate ktime_t variable and makes the
intent clearer.

Suggested-by: Jacob Keller <jacob.e.keller@intel.com>
Suggested-by: Simon Horman <horms@kernel.org>
Signed-off-by: Aleksandr Loktionov <aleksandr.loktionov@intel.com>
Tested-by: Avigail Dahan <avigailx.dahan@intel.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
---
 drivers/net/ethernet/intel/e1000e/netdev.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/ethernet/intel/e1000e/netdev.c b/drivers/net/ethernet/intel/e1000e/netdev.c
index 5b7ba1594f0d..3bdeb5f0dbea 100644
--- a/drivers/net/ethernet/intel/e1000e/netdev.c
+++ b/drivers/net/ethernet/intel/e1000e/netdev.c
@@ -3943,7 +3943,7 @@ static void e1000e_systim_reset(struct e1000_adapter *adapter)
 	/* reset the systim ns time counter */
 	spin_lock_irqsave(&adapter->systim_lock, flags);
 	timecounter_init(&adapter->tc, &adapter->cc,
-			 ktime_to_ns(ktime_get_real()));
+			 ktime_get_real_ns());
 	spin_unlock_irqrestore(&adapter->systim_lock, flags);
 
 	/* restore the previous hwtstamp configuration settings */
-- 
2.47.1

[PATCH net-next 09/15] igb: use napi_schedule_irqoff() instead of napi_schedule()

[Tony Nguyen]

From: Daiki Harada <daiky0325@gmail.com>

Replace napi_schedule() with napi_schedule_irqoff()
in the interrupt handler path in igb driver

Tested on QEMU with igb NIC emulation (-nic user,model=igb)

Suggested-by: Kohei Enju <kohei@enjuk.jp>
Signed-off-by: Daiki Harada <daiky0325@gmail.com>
Reviewed-by: Aleksandr Loktionov <aleksandr.loktionov@intel.com>
Tested-by: Rinitha S <sx.rinitha@intel.com> (A Contingent worker at Intel)
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
---
 drivers/net/ethernet/intel/igb/igb_main.c | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/drivers/net/ethernet/intel/igb/igb_main.c b/drivers/net/ethernet/intel/igb/igb_main.c
index 17eff90af9e3..056a5546956e 100644
--- a/drivers/net/ethernet/intel/igb/igb_main.c
+++ b/drivers/net/ethernet/intel/igb/igb_main.c
@@ -7157,7 +7157,7 @@ static irqreturn_t igb_msix_ring(int irq, void *data)
 	/* Write the ITR value calculated from the previous interrupt. */
 	igb_write_itr(q_vector);
 
-	napi_schedule(&q_vector->napi);
+	napi_schedule_irqoff(&q_vector->napi);
 
 	return IRQ_HANDLED;
 }
@@ -8198,7 +8198,7 @@ static irqreturn_t igb_intr_msi(int irq, void *data)
 	if (icr & E1000_ICR_TS)
 		igb_tsync_interrupt(adapter);
 
-	napi_schedule(&q_vector->napi);
+	napi_schedule_irqoff(&q_vector->napi);
 
 	return IRQ_HANDLED;
 }
@@ -8244,7 +8244,7 @@ static irqreturn_t igb_intr(int irq, void *data)
 	if (icr & E1000_ICR_TS)
 		igb_tsync_interrupt(adapter);
 
-	napi_schedule(&q_vector->napi);
+	napi_schedule_irqoff(&q_vector->napi);
 
 	return IRQ_HANDLED;
 }
-- 
2.47.1

[PATCH net-next 10/15] igc: use napi_schedule_irqoff() instead of napi_schedule()

[Tony Nguyen]

From: Daiki Harada <daiky0325@gmail.com>

Replace napi_schedule() with napi_schedule_irqoff()
in the interrupt handler path in igc driver
Tested on Intel Corporation Ethernet Controller I226-V.

Suggested-by: Kohei Enju <kohei@enjuk.jp>
Signed-off-by: Daiki Harada <daiky0325@gmail.com>
Reviewed-by: Aleksandr Loktionov <aleksandr.loktionov@intel.com>
Reviewed-by: Dima Ruinskiy <dima.ruinskiy@intel.com>
Tested-by: Moriya Kadosh <moriyax.kadosh@intel.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
---
 drivers/net/ethernet/intel/igc/igc_main.c | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/drivers/net/ethernet/intel/igc/igc_main.c b/drivers/net/ethernet/intel/igc/igc_main.c
index 89b9e1d00fc7..9143d591083f 100644
--- a/drivers/net/ethernet/intel/igc/igc_main.c
+++ b/drivers/net/ethernet/intel/igc/igc_main.c
@@ -5688,7 +5688,7 @@ static irqreturn_t igc_msix_ring(int irq, void *data)
 	/* Write the ITR value calculated from the previous interrupt. */
 	igc_write_itr(q_vector);
 
-	napi_schedule(&q_vector->napi);
+	napi_schedule_irqoff(&q_vector->napi);
 
 	return IRQ_HANDLED;
 }
@@ -6059,7 +6059,7 @@ static irqreturn_t igc_intr_msi(int irq, void *data)
 	if (icr & IGC_ICR_TS)
 		igc_tsync_interrupt(adapter);
 
-	napi_schedule(&q_vector->napi);
+	napi_schedule_irqoff(&q_vector->napi);
 
 	return IRQ_HANDLED;
 }
@@ -6105,7 +6105,7 @@ static irqreturn_t igc_intr(int irq, void *data)
 	if (icr & IGC_ICR_TS)
 		igc_tsync_interrupt(adapter);
 
-	napi_schedule(&q_vector->napi);
+	napi_schedule_irqoff(&q_vector->napi);
 
 	return IRQ_HANDLED;
 }
-- 
2.47.1

[PATCH net-next 11/15] e1000e: Use __napi_schedule_irqoff()

[Tony Nguyen]

From: Matt Vollrath <tactii@gmail.com>

The __napi_schedule_irqoff() macro is intended to bypass saving and
restoring IRQ state when scheduling is requested from an IRQ handler,
where hard interrupts are already disabled. Use this macro in all three
interrupt handlers.

This was tested on a system with an I218-V and MSI interrupts. Because
this is an optimization, I was interested in measuring the impact, so I
added ktime_get() time measurement to e1000_intr_msi and a print of the
last sample in the watchdog task. For each test case I ran a
bi-directional iperf3 to saturate the line. With some help from awk,
here are the statistics.

49 samples each, all units ns
previous: min 678 max 1265 mean 879.429 median 806 stddev 137.188
noirq:    min 707 max 1165 mean 811.857 median 790 stddev  89.486

According to this informal comparison, the mean time to handle an
interrupt from start to finish is improved by about 8% under load.

Signed-off-by: Matt Vollrath <tactii@gmail.com>
Reviewed-by: Aleksandr Loktionov <aleksandr.loktionov@intel.com>
Tested-by: Michal Cohen <michalx.cohen@intel.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
---
 drivers/net/ethernet/intel/e1000e/netdev.c | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/drivers/net/ethernet/intel/e1000e/netdev.c b/drivers/net/ethernet/intel/e1000e/netdev.c
index 3bdeb5f0dbea..808e5cddd6a9 100644
--- a/drivers/net/ethernet/intel/e1000e/netdev.c
+++ b/drivers/net/ethernet/intel/e1000e/netdev.c
@@ -1803,7 +1803,7 @@ static irqreturn_t e1000_intr_msi(int __always_unused irq, void *data)
 		adapter->total_tx_packets = 0;
 		adapter->total_rx_bytes = 0;
 		adapter->total_rx_packets = 0;
-		__napi_schedule(&adapter->napi);
+		__napi_schedule_irqoff(&adapter->napi);
 	}
 
 	return IRQ_HANDLED;
@@ -1882,7 +1882,7 @@ static irqreturn_t e1000_intr(int __always_unused irq, void *data)
 		adapter->total_tx_packets = 0;
 		adapter->total_rx_bytes = 0;
 		adapter->total_rx_packets = 0;
-		__napi_schedule(&adapter->napi);
+		__napi_schedule_irqoff(&adapter->napi);
 	}
 
 	return IRQ_HANDLED;
@@ -1951,7 +1951,7 @@ static irqreturn_t e1000_intr_msix_rx(int __always_unused irq, void *data)
 	if (napi_schedule_prep(&adapter->napi)) {
 		adapter->total_rx_bytes = 0;
 		adapter->total_rx_packets = 0;
-		__napi_schedule(&adapter->napi);
+		__napi_schedule_irqoff(&adapter->napi);
 	}
 	return IRQ_HANDLED;
 }
-- 
2.47.1

[PATCH net-next 12/15] e1000: limit endianness conversion to boundary words

[Tony Nguyen]

From: Agalakov Daniil <ade@amicon.ru>

[Why]
In e1000_set_eeprom(), the eeprom_buff is allocated to hold a range of
words. However, only the boundary words (the first and the last) are
populated from the EEPROM if the write request is not word-aligned.
The words in the middle of the buffer remain uninitialized because they
are intended to be completely overwritten by the new data via memcpy().

The previous implementation had a loop that performed le16_to_cpus()
on the entire buffer. This resulted in endianness conversion being
performed on uninitialized memory for all interior words.

Fix this by converting the endianness only for the boundary words
immediately after they are successfully read from the EEPROM.

Found by Linux Verification Center (linuxtesting.org) with SVACE.

Co-developed-by: Iskhakov Daniil <dish@amicon.ru>
Signed-off-by: Iskhakov Daniil <dish@amicon.ru>
Signed-off-by: Agalakov Daniil <ade@amicon.ru>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
---
 drivers/net/ethernet/intel/e1000/e1000_ethtool.c | 10 ++++++----
 1 file changed, 6 insertions(+), 4 deletions(-)

diff --git a/drivers/net/ethernet/intel/e1000/e1000_ethtool.c b/drivers/net/ethernet/intel/e1000/e1000_ethtool.c
index 4dcbeabb3ad2..c15ad95c63c1 100644
--- a/drivers/net/ethernet/intel/e1000/e1000_ethtool.c
+++ b/drivers/net/ethernet/intel/e1000/e1000_ethtool.c
@@ -499,6 +499,9 @@ static int e1000_set_eeprom(struct net_device *netdev,
 		if (ret_val)
 			goto out;
 
+		/* Device's eeprom is always little-endian, word addressable */
+		le16_to_cpus(&eeprom_buff[0]);
+
 		ptr++;
 	}
 	if ((eeprom->offset + eeprom->len) & 1) {
@@ -509,11 +512,10 @@ static int e1000_set_eeprom(struct net_device *netdev,
 					    &eeprom_buff[last_word - first_word]);
 		if (ret_val)
 			goto out;
-	}
 
-	/* Device's eeprom is always little-endian, word addressable */
-	for (i = 0; i < last_word - first_word + 1; i++)
-		le16_to_cpus(&eeprom_buff[i]);
+		/* Device's eeprom is always little-endian, word addressable */
+		le16_to_cpus(&eeprom_buff[last_word - first_word]);
+	}
 
 	memcpy(ptr, bytes, eeprom->len);
 
-- 
2.47.1

[PATCH net-next 13/15] e1000e: limit endianness conversion to boundary words

[Tony Nguyen]

From: Agalakov Daniil <ade@amicon.ru>

[Why]
In e1000_set_eeprom(), the eeprom_buff is allocated to hold a range of
words. However, only the boundary words (the first and the last) are
populated from the EEPROM if the write request is not word-aligned.
The words in the middle of the buffer remain uninitialized because they
are intended to be completely overwritten by the new data via memcpy().

The previous implementation had a loop that performed le16_to_cpus()
on the entire buffer. This resulted in endianness conversion being
performed on uninitialized memory for all interior words.

Fix this by converting the endianness only for the boundary words
immediately after they are successfully read from the EEPROM.

Found by Linux Verification Center (linuxtesting.org) with SVACE.

Co-developed-by: Iskhakov Daniil <dish@amicon.ru>
Signed-off-by: Iskhakov Daniil <dish@amicon.ru>
Signed-off-by: Agalakov Daniil <ade@amicon.ru>
Tested-by: Avigail Dahan <avigailx.dahan@intel.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
---
 drivers/net/ethernet/intel/e1000e/ethtool.c | 19 ++++++++++++-------
 1 file changed, 12 insertions(+), 7 deletions(-)

diff --git a/drivers/net/ethernet/intel/e1000e/ethtool.c b/drivers/net/ethernet/intel/e1000e/ethtool.c
index dbed30943ef4..a8b35ae41141 100644
--- a/drivers/net/ethernet/intel/e1000e/ethtool.c
+++ b/drivers/net/ethernet/intel/e1000e/ethtool.c
@@ -583,20 +583,25 @@ static int e1000_set_eeprom(struct net_device *netdev,
 		/* need read/modify/write of first changed EEPROM word */
 		/* only the second byte of the word is being modified */
 		ret_val = e1000_read_nvm(hw, first_word, 1, &eeprom_buff[0]);
+		if (ret_val)
+			goto out;
+
+		/* Device's eeprom is always little-endian, word addressable */
+		le16_to_cpus(&eeprom_buff[0]);
+
 		ptr++;
 	}
-	if (((eeprom->offset + eeprom->len) & 1) && (!ret_val))
+	if ((eeprom->offset + eeprom->len) & 1) {
 		/* need read/modify/write of last changed EEPROM word */
 		/* only the first byte of the word is being modified */
 		ret_val = e1000_read_nvm(hw, last_word, 1,
 					 &eeprom_buff[last_word - first_word]);
+		if (ret_val)
+			goto out;
 
-	if (ret_val)
-		goto out;
-
-	/* Device's eeprom is always little-endian, word addressable */
-	for (i = 0; i < last_word - first_word + 1; i++)
-		le16_to_cpus(&eeprom_buff[i]);
+		/* Device's eeprom is always little-endian, word addressable */
+		le16_to_cpus(&eeprom_buff[last_word - first_word]);
+	}
 
 	memcpy(ptr, bytes, eeprom->len);
 
-- 
2.47.1

[PATCH net-next 14/15] igb: fix typos in comments

[Tony Nguyen]

From: Maximilian Pezzullo <maximilianpezzullo@gmail.com>

Fix spelling errors in code comments:
 - e1000_nvm.c: 'likley' -> 'likely'
 - e1000_mac.c: 'auto-negotitation' -> 'auto-negotiation'
 - e1000_mbx.h: 'exra' -> 'extra'
 - e1000_defines.h: 'Aserted' -> 'Asserted'

Signed-off-by: Maximilian Pezzullo <maximilianpezzullo@gmail.com>
Reviewed-by: Aleksandr Loktionov <aleksandr.loktionov@intel.com>
Reviewed-by: Joe Damato <joe@dama.to>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
---
 drivers/net/ethernet/intel/igb/e1000_defines.h | 2 +-
 drivers/net/ethernet/intel/igb/e1000_mac.c     | 2 +-
 drivers/net/ethernet/intel/igb/e1000_mbx.h     | 2 +-
 drivers/net/ethernet/intel/igb/e1000_nvm.c     | 2 +-
 4 files changed, 4 insertions(+), 4 deletions(-)

diff --git a/drivers/net/ethernet/intel/igb/e1000_defines.h b/drivers/net/ethernet/intel/igb/e1000_defines.h
index fa028928482f..7e6f9aa2d57b 100644
--- a/drivers/net/ethernet/intel/igb/e1000_defines.h
+++ b/drivers/net/ethernet/intel/igb/e1000_defines.h
@@ -442,7 +442,7 @@
 /* Interrupt Cause Set */
 #define E1000_ICS_LSC       E1000_ICR_LSC       /* Link Status Change */
 #define E1000_ICS_RXDMT0    E1000_ICR_RXDMT0    /* rx desc min. threshold */
-#define E1000_ICS_DRSTA     E1000_ICR_DRSTA     /* Device Reset Aserted */
+#define E1000_ICS_DRSTA     E1000_ICR_DRSTA     /* Device Reset Asserted */
 
 /* Extended Interrupt Cause Set */
 /* E1000_EITR_CNT_IGNR is only for 82576 and newer */
diff --git a/drivers/net/ethernet/intel/igb/e1000_mac.c b/drivers/net/ethernet/intel/igb/e1000_mac.c
index fa3dfafd2bb1..2bcce6eef0c7 100644
--- a/drivers/net/ethernet/intel/igb/e1000_mac.c
+++ b/drivers/net/ethernet/intel/igb/e1000_mac.c
@@ -1581,7 +1581,7 @@ s32 igb_disable_pcie_master(struct e1000_hw *hw)
  *  igb_validate_mdi_setting - Verify MDI/MDIx settings
  *  @hw: pointer to the HW structure
  *
- *  Verify that when not using auto-negotitation that MDI/MDIx is correctly
+ *  Verify that when not using auto-negotiation that MDI/MDIx is correctly
  *  set, which is forced to MDI mode only.
  **/
 s32 igb_validate_mdi_setting(struct e1000_hw *hw)
diff --git a/drivers/net/ethernet/intel/igb/e1000_mbx.h b/drivers/net/ethernet/intel/igb/e1000_mbx.h
index 178e60ec71d4..9e44527f5eea 100644
--- a/drivers/net/ethernet/intel/igb/e1000_mbx.h
+++ b/drivers/net/ethernet/intel/igb/e1000_mbx.h
@@ -30,7 +30,7 @@
 /* Indicates that VF is still clear to send requests */
 #define E1000_VT_MSGTYPE_CTS	0x20000000
 #define E1000_VT_MSGINFO_SHIFT	16
-/* bits 23:16 are used for exra info for certain messages */
+/* bits 23:16 are used for extra info for certain messages */
 #define E1000_VT_MSGINFO_MASK	(0xFF << E1000_VT_MSGINFO_SHIFT)
 
 #define E1000_VF_RESET		0x01 /* VF requests reset */
diff --git a/drivers/net/ethernet/intel/igb/e1000_nvm.c b/drivers/net/ethernet/intel/igb/e1000_nvm.c
index c8638502c2be..cf4e5d0e9417 100644
--- a/drivers/net/ethernet/intel/igb/e1000_nvm.c
+++ b/drivers/net/ethernet/intel/igb/e1000_nvm.c
@@ -405,7 +405,7 @@ s32 igb_read_nvm_eerd(struct e1000_hw *hw, u16 offset, u16 words, u16 *data)
  *  Writes data to EEPROM at offset using SPI interface.
  *
  *  If e1000_update_nvm_checksum is not called after this function , the
- *  EEPROM will most likley contain an invalid checksum.
+ *  EEPROM will most likely contain an invalid checksum.
  **/
 s32 igb_write_nvm_spi(struct e1000_hw *hw, u16 offset, u16 words, u16 *data)
 {
-- 
2.47.1

[PATCH net-next 15/15] igc: fix typos in comments

[Tony Nguyen]

From: Maximilian Pezzullo <maximilianpezzullo@gmail.com>

Fix spelling errors in code comments:
 - igc_diag.c: 'autonegotioation' -> 'autonegotiation'
 - igc_main.c: 'revisons' -> 'revisions' (two occurrences)

Signed-off-by: Maximilian Pezzullo <maximilianpezzullo@gmail.com>
Reviewed-by: Aleksandr Loktionov <aleksandr.loktionov@intel.com>
Reviewed-by: Joe Damato <joe@dama.to>
Tested-by: Avigail Dahan <avigailx.dahan@intel.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
---
 drivers/net/ethernet/intel/igc/igc_diag.c | 2 +-
 drivers/net/ethernet/intel/igc/igc_main.c | 4 ++--
 2 files changed, 3 insertions(+), 3 deletions(-)

diff --git a/drivers/net/ethernet/intel/igc/igc_diag.c b/drivers/net/ethernet/intel/igc/igc_diag.c
index a43d7244ee70..031561fdce49 100644
--- a/drivers/net/ethernet/intel/igc/igc_diag.c
+++ b/drivers/net/ethernet/intel/igc/igc_diag.c
@@ -172,7 +172,7 @@ bool igc_link_test(struct igc_adapter *adapter, u64 *data)
 
 	*data = 0;
 
-	/* add delay to give enough time for autonegotioation to finish */
+	/* add delay to give enough time for autonegotiation to finish */
 	ssleep(5);
 
 	link_up = igc_has_link(adapter);
diff --git a/drivers/net/ethernet/intel/igc/igc_main.c b/drivers/net/ethernet/intel/igc/igc_main.c
index 9143d591083f..e0d719c11105 100644
--- a/drivers/net/ethernet/intel/igc/igc_main.c
+++ b/drivers/net/ethernet/intel/igc/igc_main.c
@@ -1797,7 +1797,7 @@ static const enum pkt_hash_types igc_rss_type_table[IGC_RSS_TYPE_MAX_TABLE] = {
 	[IGC_RSS_TYPE_HASH_UDP_IPV6_EX] = PKT_HASH_TYPE_L4,
 	[10] = PKT_HASH_TYPE_NONE, /* RSS Type above 9 "Reserved" by HW  */
 	[11] = PKT_HASH_TYPE_NONE, /* keep array sized for SW bit-mask   */
-	[12] = PKT_HASH_TYPE_NONE, /* to handle future HW revisons       */
+	[12] = PKT_HASH_TYPE_NONE, /* to handle future HW revisions       */
 	[13] = PKT_HASH_TYPE_NONE,
 	[14] = PKT_HASH_TYPE_NONE,
 	[15] = PKT_HASH_TYPE_NONE,
@@ -7039,7 +7039,7 @@ static enum xdp_rss_hash_type igc_xdp_rss_type[IGC_RSS_TYPE_MAX_TABLE] = {
 	[IGC_RSS_TYPE_HASH_UDP_IPV6_EX] = XDP_RSS_TYPE_L4_IPV6_UDP_EX,
 	[10] = XDP_RSS_TYPE_NONE, /* RSS Type above 9 "Reserved" by HW  */
 	[11] = XDP_RSS_TYPE_NONE, /* keep array sized for SW bit-mask   */
-	[12] = XDP_RSS_TYPE_NONE, /* to handle future HW revisons       */
+	[12] = XDP_RSS_TYPE_NONE, /* to handle future HW revisions       */
 	[13] = XDP_RSS_TYPE_NONE,
 	[14] = XDP_RSS_TYPE_NONE,
 	[15] = XDP_RSS_TYPE_NONE,
-- 
2.47.1

Re: [PATCH net-next 06/15] igb: Retrieve Tx timestamp from BH workqueue

[Simon Horman]

This is an AI-generated review of your patch. The human sending this
email has considered the AI review valid, or at least plausible.
Full review at: https://sashiko.dev
---
igb: Retrieve Tx timestamp from BH workqueue

Retrieve Tx timestamp from system BH instead of regular system workqueue.

The current implementation uses schedule_work() which is executed by the
system work queue and kworkers to retrieve Tx timestamps. This increases
latency and can lead to timeouts in case of heavy system load. i210 is
often used in industrial systems, where timestamp timeouts can be fatal.

Simon says:

I think that first issue below, flagged as pre-existing, can be considered
in the context of possible follow-up.

However, the issues flagged below that, regarding scheduling work
and possible softirq storms, do seem to be something that warrants
investigation in the context of this patch-set.

> diff --git a/drivers/net/ethernet/intel/igb/igb_main.c b/drivers/net/ethernet/intel/igb/igb_main.c
> index 49d46670695f3..17eff90af9e3d 100644
> --- a/drivers/net/ethernet/intel/igb/igb_main.c
> +++ b/drivers/net/ethernet/intel/igb/igb_main.c
> @@ -6571,7 +6571,7 @@ netdev_tx_t igb_xmit_frame_ring(struct sk_buff *skb,

[Severity: High]
This is a pre-existing issue, but does this function modify the shared SKB
data area without verifying exclusive ownership?

In igb_xmit_frame_ring(), just before the work is queued:

	if (adapter->tstamp_config.tx_type == HWTSTAMP_TX_ON &&
	    !test_and_set_bit_lock(__IGB_PTP_TX_IN_PROGRESS,
				   &adapter->state)) {
		skb_shinfo(skb)->tx_flags |= SKBTX_IN_PROGRESS;

If the SKB is cloned (e.g. via tc mirred or packet sniffers), won't modifying
the shared shinfo region create a data race with other network paths?

[Severity: High]
This is a pre-existing issue, but can the error path sleep in atomic context?

igb_xmit_frame_ring() runs in atomic context as the ndo_start_xmit callback,
but the cleanup path calls a blocking function:

cleanup_tx_tstamp:
	if (unlikely(tx_flags & IGB_TX_FLAGS_TSTAMP)) {
		...
		if (adapter->hw.mac.type == e1000_82576)
			cancel_work_sync(&adapter->ptp_tx_work);

Could this trigger scheduling while atomic warnings or potential deadlocks?

>  			adapter->ptp_tx_skb = skb_get(skb);
>  			adapter->ptp_tx_start = jiffies;
>  			if (adapter->hw.mac.type == e1000_82576)
> -				schedule_work(&adapter->ptp_tx_work);
> +				queue_work(system_bh_wq, &adapter->ptp_tx_work);

[Severity: High]
Does this change introduce a softirq polling storm for e1000_82576 MACs?

For this specific MAC type, the work is scheduled immediately upon packet
transmission. When queueing to system_bh_wq instead of a regular workqueue,
the polling loop in igb_ptp_tx_work() will now run in softirq context.

>  		} else {
>  			adapter->tx_hwtstamp_skipped++;
>  		}

[ ... ]

> diff --git a/drivers/net/ethernet/intel/igb/igb_ptp.c b/drivers/net/ethernet/intel/igb/igb_ptp.c
> index bd85d02ecadd8..7b44f9090631a 100644
> --- a/drivers/net/ethernet/intel/igb/igb_ptp.c
> +++ b/drivers/net/ethernet/intel/igb/igb_ptp.c
> @@ -832,7 +832,7 @@ static void igb_ptp_tx_work(struct work_struct *work)
>  		igb_ptp_tx_hwtstamp(adapter);
>  	else
>  		/* reschedule to check later */
> -		schedule_work(&adapter->ptp_tx_work);
> +		queue_work(system_bh_wq, &adapter->ptp_tx_work);

[Severity: High]
If the hardware timestamp is not yet valid, won't this work item
unconditionally reschedule itself to system_bh_wq without delay? 

Because the work requeues itself immediately, this could create a tight
polling loop in softirq context, maximizing the CPU's softirq time limit
and degrading network processing and overall system latency.

[Severity: High]
This is a pre-existing issue, but is there a use-after-free race condition
between igb_ptp_tx_work() and the watchdog task igb_ptp_tx_hang()?

Both functions independently check if the timeout has elapsed using
adapter->ptp_tx_start. If igb_ptp_tx_work() detects a timeout, it frees the
SKB and clears the state bit, allowing a new PTP packet to be queued:

	if (time_is_before_jiffies(adapter->ptp_tx_start +
				   IGB_PTP_TX_TIMEOUT)) {
		dev_kfree_skb_any(adapter->ptp_tx_skb);
		...
		clear_bit_unlock(__IGB_PTP_TX_IN_PROGRESS, &adapter->state);

If igb_ptp_tx_hang() concurrently evaluates the timeout as true based on the
old ptp_tx_start but is preempted, could it resume and mistakenly cancel
the work for the newly queued packet?

igb_ptp_tx_hang() {
	bool timeout = time_is_before_jiffies(adapter->ptp_tx_start +
					      IGB_PTP_TX_TIMEOUT);
	...
	if (timeout) {
		cancel_work_sync(&adapter->ptp_tx_work);
		dev_kfree_skb_any(adapter->ptp_tx_skb);

Would this cause a use-after-free by freeing the new SKB and corrupting the
driver's PTP state?

>  }