[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

[Xen-changelog] [linux-2.6.18-xen] pciback: only check PF if actually dealing with a VF


  • To: xen-changelog@xxxxxxxxxxxxxxxxxxx
  • From: Xen patchbot-linux-2.6.18-xen <patchbot@xxxxxxx>
  • Date: Mon, 09 Jan 2017 11:33:01 +0000
  • Delivery-date: Mon, 09 Jan 2017 11:33:08 +0000
  • List-id: "Change log for Mercurial \(receive only\)" <xen-changelog.lists.xen.org>

# HG changeset patch
# User Jan Beulich <jbeulich@xxxxxxxx>
# Date 1483959360 -3600
#      Mon Jan 09 11:56:00 2017 +0100
# Node ID 85a408a07bccb3419e9bf8a46db3566bdf11af3e
# Parent  de7a5d5195f8c07f66ae6bdcdc16e9ed3e47ae6e
pciback: only check PF if actually dealing with a VF

c/s c00c8be428cd ("pciback: check PF instead of VF for
PCI_COMMAND_MEMORY") has a blatant backporting mistake: ->is_physfn
gets set only for SR-IOV devices once they've got enabled; we really
need to check ->is_virtfn here.

Signed-off-by: Jan Beulich <jbeulich@xxxxxxxx>
---


diff -r de7a5d5195f8 -r 85a408a07bcc 
drivers/xen/pciback/conf_space_capability_msi.c
--- a/drivers/xen/pciback/conf_space_capability_msi.c   Mon Jan 09 11:53:58 
2017 +0100
+++ b/drivers/xen/pciback/conf_space_capability_msi.c   Mon Jan 09 11:56:00 
2017 +0100
@@ -62,7 +62,7 @@ int pciback_enable_msix(struct pciback_d
         * But VF devices are unique in which the PF needs to be checked.
         */
 #ifdef CONFIG_PCI_IOV
-       phys_dev = dev->is_physfn ? dev : dev->physfn;
+       phys_dev = dev->is_virtfn ? dev->physfn : dev;
 #endif
        pci_read_config_word(phys_dev, PCI_COMMAND, &cmd);
        if (dev->msi_enabled || !(cmd & PCI_COMMAND_MEMORY))

_______________________________________________
Xen-changelog mailing list
Xen-changelog@xxxxxxxxxxxxx
https://lists.xenproject.org/xen-changelog

 


Rackspace

Lists.xenproject.org is hosted with RackSpace, monitoring our
servers 24x7x365 and backed by RackSpace's Fanatical Support®.