linux-kernel - Re: PROBLEM: Intel I210AT NIC resets while using PCI passthrough on ESXi (regression)

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: Windows password security audit tool. GUI, reports in PDF.

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

Message-ID: <alpine.DEB.2.11.1607261342310.19896@nanos>
Date:	Tue, 26 Jul 2016 13:46:11 +0200 (CEST)
From:	Thomas Gleixner <tglx@...utronix.de>
To:	Foster Snowhill <forst@...stwoof.ru>
cc:	LKML <linux-kernel@...r.kernel.org>,
	Matthias Prager <linux@...thiasprager.de>,
	Jason Taylor <jason.taylor@...plivity.com>,
	Marc Zyngier <marc.zyngier@....com>
Subject: Re: PROBLEM: Intel I210AT NIC resets while using PCI passthrough on
 ESXi (regression)

On Mon, 25 Jul 2016, Foster Snowhill wrote:
> On 25.07.16 13:56, Thomas Gleixner wrote:
i> > Could you please give the patch below a try? It might be related, but I'm
> > not sure whether it will cure that particular vmware oddity.
>
> Patch fixed the issue for me. Did 4-5 reboots just to be sure, all working
> properly, no adapter resets. Just for the sake of experiment built a clean
> 4.7 kernel without the patch, issue still present there as expected.

Can you please try whether the replacement patch below fixes your issue as
well?

Thanks,

	tglx

8<------------
diff --git a/drivers/pci/msi.c b/drivers/pci/msi.c
index a080f4496fe2..142341f8331b 100644
--- a/drivers/pci/msi.c
+++ b/drivers/pci/msi.c
@@ -645,6 +645,15 @@ static int msi_capability_init(struct pci_dev *dev, int nvec)
 		return ret;
 	}
 
+	/*
+	 * The mask can be ignored and PCI 2.3 does not specify mask bits for
+	 * each MSI interrupt. So in case of hierarchical irqdomains we need
+	 * to make sure that if masking is not available that the msi message
+	 * is written prior to setting the MSI enable bit in the device.
+	 */
+	if (pci_msi_ignore_mask || !entry->msi_attrib.maskbit)
+		irq_domain_activate_irq(irq_get_irq_data(entry->irq));
+
 	/* Set MSI enabled bits	 */
 	pci_intx_for_msi(dev, 0);
 	pci_msi_set_enable(dev, 1);