r8169: fix issue with forced threading in combination with shared interrupts

[ Upstream commit 2734a24 ]

As reported by Serge flag IRQF_NO_THREAD causes an error if the
interrupt is actually shared and the other driver(s) don't have this
flag set. This situation can occur if a PCI(e) legacy interrupt is
used in combination with forced threading.
There's no good way to deal with this properly, therefore we have to
remove flag IRQF_NO_THREAD. For fixing the original forced threading
issue switch to napi_schedule().

Fixes: 424a646 ("r8169: fix operation under forced interrupt threading")
Link: https://www.spinics.net/lists/netdev/msg694960.html
Reported-by: Serge Belyshev <belyshev@depni.sinp.msu.ru>
Signed-off-by: Heiner Kallweit <hkallweit1@gmail.com>
Tested-by: Serge Belyshev <belyshev@depni.sinp.msu.ru>
Link: https://lore.kernel.org/r/b5b53bfe-35ac-3768-85bf-74d1290cf394@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

Author: hkallweit

drivers/net/ethernet/realtek/r8169_main.c

@@ -6264,7 +6264,7 @@ static irqreturn_t rtl8169_interrupt(int irq, void *dev_instance)
 	}
 
 	rtl_irq_disable(tp);
-	napi_schedule_irqoff(&tp->napi);
+	napi_schedule(&tp->napi);
 out:
 	rtl_ack_events(tp, status);
 
@@ -6470,7 +6470,7 @@ static int rtl_open(struct net_device *dev)
 	rtl_request_firmware(tp);
 
 	retval = request_irq(pci_irq_vector(pdev, 0), rtl8169_interrupt,
-			     IRQF_NO_THREAD | IRQF_SHARED, dev->name, tp);
+			     IRQF_SHARED, dev->name, tp);
 	if (retval < 0)
 		goto err_release_fw_2;