On Tue, Aug 20, 2024 at 11:13:54PM +0200, Petr Valenta wrote:
> Dne 20. 08. 24 v 20:09 Bjorn Helgaas napsal(a):
> > On Mon, Aug 19, 2024 at 07:23:42AM +0200, Jiri Slaby wrote:
> > > On 19. 08. 24, 6:50, Jiri Slaby wrote:
> > > > CC e1000e guys + Jesse (due to 75a3f93b5383) + Bjorn (due to 
> > > > b2c289415b2b)
> > > 
> > > Bjorn,
> > > 
> > > I am confused by these changes:
> > > ==========================================
> > > @@ -291,16 +288,13 @@ static int e1000_set_link_ksettings(struct 
> > > net_device
> > > *net
> > > dev,
> > >           * duplex is forced.
> > >           */
> > >          if (cmd->base.eth_tp_mdix_ctrl) {
> > > -               if (hw->phy.media_type != e1000_media_type_copper) {
> > > -                       ret_val = -EOPNOTSUPP;
> > > -                       goto out;
> > > -               }
> > > +               if (hw->phy.media_type != e1000_media_type_copper)
> > > +                       return -EOPNOTSUPP;
> > > 
> > >                  if ((cmd->base.eth_tp_mdix_ctrl != ETH_TP_MDI_AUTO) &&
> > >                      (cmd->base.autoneg != AUTONEG_ENABLE)) {
> > >                          e_err("forcing MDI/MDI-X state is not supported 
> > > when
> > > lin
> > > k speed and/or duplex are forced\n");
> > > -                       ret_val = -EINVAL;
> > > -                       goto out;
> > > +                       return -EINVAL;
> > >                  }
> > >          }
> > > 
> > > @@ -347,7 +341,6 @@ static int e1000_set_link_ksettings(struct net_device
> > > *netde
> > > v,
> > >          }
> > > 
> > >   out:
> > > -       pm_runtime_put_sync(netdev->dev.parent);
> > >          clear_bit(__E1000_RESETTING, &adapter->state);
> > >          return ret_val;
> > >   }
> > > ==========================================
> > > 
> > > So no more clear_bit(__E1000_RESETTING in the above fail paths. Is that
> > > intentional?
> > 
> > Not intentional.  Petr, do you have the ability to test the patch
> > below?  I'm not sure it's the correct fix, but it reverts the pieces
> > of b2c289415b2b that Jiri pointed out.
> 
> I tested the patch below but it didn't help. After the first boot with new
> kernel it looked promising as the irq storm only appeared for a few seconds,
> but with subsequent reboots it was the same as without the patch.

Thank you very much for testing that!

> To be sure, I also send the md5sum of ethtool.c after applying the patch:
> 
> a25c003257538f16994b4d7c4260e894 ethtool.c

Thanks, that matches what I get when applying the patch on v6.10.

I'm at a loss.  You could try reverting the entire b2c289415b2b commit
(patch for that is below).

If that doesn't help, I guess you could try reverting the other
commits Jiri mentioned:

  76a0a3f9cc2f e1000e: fix force smbus during suspend flow
  c93a6f62cb1b e1000e: Fix S0ix residency on corporate systems
  bfd546a552e1 e1000e: move force SMBUS near the end of enable_ulp function
  6918107e2540 net: e1000e & ixgbe: Remove PCI_HEADER_TYPE_MFD duplicates
  1eb2cded45b3 net: annotate writes on dev->mtu from ndo_change_mtu()
  b2c289415b2b e1000e: Remove redundant runtime resume for ethtool_ops
  75a3f93b5383 net: intel: implement modern PM ops declarations

If you do this, I would revert 76a0a3f9cc2f, test, then revert
c93a6f62cb1b in addition, test, then revert bfd546a552e1 in addition,
etc.

commit 5e92945ffe5c ("Revert "e1000e: Remove redundant runtime resume for 
ethtool_ops"")
Author: Bjorn Helgaas <bhelg...@google.com>
Date:   Tue Aug 20 16:18:32 2024 -0500

    Revert "e1000e: Remove redundant runtime resume for ethtool_ops"
    
    This reverts commit b2c289415b2b2ef112b78d5e73b4acecf5db409e.


diff --git a/drivers/net/ethernet/intel/e1000e/ethtool.c 
b/drivers/net/ethernet/intel/e1000e/ethtool.c
index 9364bc2b4eb1..61fa2f6b3708 100644
--- a/drivers/net/ethernet/intel/e1000e/ethtool.c
+++ b/drivers/net/ethernet/intel/e1000e/ethtool.c
@@ -156,7 +156,7 @@ static int e1000_get_link_ksettings(struct net_device 
*netdev,
                        speed = adapter->link_speed;
                        cmd->base.duplex = adapter->link_duplex - 1;
                }
-       } else {
+       } else if (!pm_runtime_suspended(netdev->dev.parent)) {
                u32 status = er32(STATUS);
 
                if (status & E1000_STATUS_LU) {
@@ -274,13 +274,16 @@ static int e1000_set_link_ksettings(struct net_device 
*netdev,
        ethtool_convert_link_mode_to_legacy_u32(&advertising,
                                                cmd->link_modes.advertising);
 
+       pm_runtime_get_sync(netdev->dev.parent);
+
        /* When SoL/IDER sessions are active, autoneg/speed/duplex
         * cannot be changed
         */
        if (hw->phy.ops.check_reset_block &&
            hw->phy.ops.check_reset_block(hw)) {
                e_err("Cannot change link characteristics when SoL/IDER is 
active.\n");
-               return -EINVAL;
+               ret_val = -EINVAL;
+               goto out;
        }
 
        /* MDI setting is only allowed when autoneg enabled because
@@ -288,13 +291,16 @@ static int e1000_set_link_ksettings(struct net_device 
*netdev,
         * duplex is forced.
         */
        if (cmd->base.eth_tp_mdix_ctrl) {
-               if (hw->phy.media_type != e1000_media_type_copper)
-                       return -EOPNOTSUPP;
+               if (hw->phy.media_type != e1000_media_type_copper) {
+                       ret_val = -EOPNOTSUPP;
+                       goto out;
+               }
 
                if ((cmd->base.eth_tp_mdix_ctrl != ETH_TP_MDI_AUTO) &&
                    (cmd->base.autoneg != AUTONEG_ENABLE)) {
                        e_err("forcing MDI/MDI-X state is not supported when 
link speed and/or duplex are forced\n");
-                       return -EINVAL;
+                       ret_val = -EINVAL;
+                       goto out;
                }
        }
 
@@ -341,6 +347,7 @@ static int e1000_set_link_ksettings(struct net_device 
*netdev,
        }
 
 out:
+       pm_runtime_put_sync(netdev->dev.parent);
        clear_bit(__E1000_RESETTING, &adapter->state);
        return ret_val;
 }
@@ -376,6 +383,8 @@ static int e1000_set_pauseparam(struct net_device *netdev,
        while (test_and_set_bit(__E1000_RESETTING, &adapter->state))
                usleep_range(1000, 2000);
 
+       pm_runtime_get_sync(netdev->dev.parent);
+
        if (adapter->fc_autoneg == AUTONEG_ENABLE) {
                hw->fc.requested_mode = e1000_fc_default;
                if (netif_running(adapter->netdev)) {
@@ -408,6 +417,7 @@ static int e1000_set_pauseparam(struct net_device *netdev,
        }
 
 out:
+       pm_runtime_put_sync(netdev->dev.parent);
        clear_bit(__E1000_RESETTING, &adapter->state);
        return retval;
 }
@@ -438,6 +448,8 @@ static void e1000_get_regs(struct net_device *netdev,
        u32 *regs_buff = p;
        u16 phy_data;
 
+       pm_runtime_get_sync(netdev->dev.parent);
+
        memset(p, 0, E1000_REGS_LEN * sizeof(u32));
 
        regs->version = (1u << 24) |
@@ -483,6 +495,8 @@ static void e1000_get_regs(struct net_device *netdev,
        e1e_rphy(hw, MII_STAT1000, &phy_data);
        regs_buff[24] = (u32)phy_data;  /* phy local receiver status */
        regs_buff[25] = regs_buff[24];  /* phy remote receiver status */
+
+       pm_runtime_put_sync(netdev->dev.parent);
 }
 
 static int e1000_get_eeprom_len(struct net_device *netdev)
@@ -515,6 +529,8 @@ static int e1000_get_eeprom(struct net_device *netdev,
        if (!eeprom_buff)
                return -ENOMEM;
 
+       pm_runtime_get_sync(netdev->dev.parent);
+
        if (hw->nvm.type == e1000_nvm_eeprom_spi) {
                ret_val = e1000_read_nvm(hw, first_word,
                                         last_word - first_word + 1,
@@ -528,6 +544,8 @@ static int e1000_get_eeprom(struct net_device *netdev,
                }
        }
 
+       pm_runtime_put_sync(netdev->dev.parent);
+
        if (ret_val) {
                /* a read error occurred, throw away the result */
                memset(eeprom_buff, 0xff, sizeof(u16) *
@@ -577,6 +595,8 @@ static int e1000_set_eeprom(struct net_device *netdev,
 
        ptr = (void *)eeprom_buff;
 
+       pm_runtime_get_sync(netdev->dev.parent);
+
        if (eeprom->offset & 1) {
                /* need read/modify/write of first changed EEPROM word */
                /* only the second byte of the word is being modified */
@@ -617,6 +637,7 @@ static int e1000_set_eeprom(struct net_device *netdev,
                ret_val = e1000e_update_nvm_checksum(hw);
 
 out:
+       pm_runtime_put_sync(netdev->dev.parent);
        kfree(eeprom_buff);
        return ret_val;
 }
@@ -712,6 +733,8 @@ static int e1000_set_ringparam(struct net_device *netdev,
                }
        }
 
+       pm_runtime_get_sync(netdev->dev.parent);
+
        e1000e_down(adapter, true);
 
        /* We can't just free everything and then setup again, because the
@@ -750,6 +773,7 @@ static int e1000_set_ringparam(struct net_device *netdev,
                e1000e_free_tx_resources(temp_tx);
 err_setup:
        e1000e_up(adapter);
+       pm_runtime_put_sync(netdev->dev.parent);
 free_temp:
        vfree(temp_tx);
        vfree(temp_rx);
@@ -1792,6 +1816,8 @@ static void e1000_diag_test(struct net_device *netdev,
        u8 autoneg;
        bool if_running = netif_running(netdev);
 
+       pm_runtime_get_sync(netdev->dev.parent);
+
        set_bit(__E1000_TESTING, &adapter->state);
 
        if (!if_running) {
@@ -1877,6 +1903,8 @@ static void e1000_diag_test(struct net_device *netdev,
        }
 
        msleep_interruptible(4 * 1000);
+
+       pm_runtime_put_sync(netdev->dev.parent);
 }
 
 static void e1000_get_wol(struct net_device *netdev,
@@ -2018,11 +2046,15 @@ static int e1000_set_coalesce(struct net_device *netdev,
                adapter->itr_setting = adapter->itr & ~3;
        }
 
+       pm_runtime_get_sync(netdev->dev.parent);
+
        if (adapter->itr_setting != 0)
                e1000e_write_itr(adapter, adapter->itr);
        else
                e1000e_write_itr(adapter, 0);
 
+       pm_runtime_put_sync(netdev->dev.parent);
+
        return 0;
 }
 
@@ -2036,7 +2068,9 @@ static int e1000_nway_reset(struct net_device *netdev)
        if (!adapter->hw.mac.autoneg)
                return -EINVAL;
 
+       pm_runtime_get_sync(netdev->dev.parent);
        e1000e_reinit_locked(adapter);
+       pm_runtime_put_sync(netdev->dev.parent);
 
        return 0;
 }
@@ -2050,8 +2084,12 @@ static void e1000_get_ethtool_stats(struct net_device 
*netdev,
        int i;
        char *p = NULL;
 
+       pm_runtime_get_sync(netdev->dev.parent);
+
        dev_get_stats(netdev, &net_stats);
 
+       pm_runtime_put_sync(netdev->dev.parent);
+
        for (i = 0; i < E1000_GLOBAL_STATS_LEN; i++) {
                switch (e1000_gstrings_stats[i].type) {
                case NETDEV_STATS:
@@ -2108,7 +2146,9 @@ static int e1000_get_rxnfc(struct net_device *netdev,
                struct e1000_hw *hw = &adapter->hw;
                u32 mrqc;
 
+               pm_runtime_get_sync(netdev->dev.parent);
                mrqc = er32(MRQC);
+               pm_runtime_put_sync(netdev->dev.parent);
 
                if (!(mrqc & E1000_MRQC_RSS_FIELD_MASK))
                        return 0;
@@ -2171,9 +2211,13 @@ static int e1000e_get_eee(struct net_device *netdev, 
struct ethtool_keee *edata)
                return -EOPNOTSUPP;
        }
 
+       pm_runtime_get_sync(netdev->dev.parent);
+
        ret_val = hw->phy.ops.acquire(hw);
-       if (ret_val)
+       if (ret_val) {
+               pm_runtime_put_sync(netdev->dev.parent);
                return -EBUSY;
+       }
 
        /* EEE Capability */
        ret_val = e1000_read_emi_reg_locked(hw, cap_addr, &phy_data);
@@ -2213,6 +2257,8 @@ static int e1000e_get_eee(struct net_device *netdev, 
struct ethtool_keee *edata)
        if (ret_val)
                ret_val = -ENODATA;
 
+       pm_runtime_put_sync(netdev->dev.parent);
+
        return ret_val;
 }
 
@@ -2253,12 +2299,16 @@ static int e1000e_set_eee(struct net_device *netdev, 
struct ethtool_keee *edata)
 
        hw->dev_spec.ich8lan.eee_disable = !edata->eee_enabled;
 
+       pm_runtime_get_sync(netdev->dev.parent);
+
        /* reset the link */
        if (netif_running(netdev))
                e1000e_reinit_locked(adapter);
        else
                e1000e_reset(adapter);
 
+       pm_runtime_put_sync(netdev->dev.parent);
+
        return 0;
 }
 

Reply via email to