Re: [Xen-devel] Issues with PCI-Passtrough (VT-d) in HVM with Xen 4.6

[Andrey Grodzovsky <andrey2805@xxxxxxxxx>]

In reply to - http://lists.xen.org/archives/html/xen-devel/2016-06/msg00622.html

HI, I am working with Jurgen on the issue, as per Jan's request I tried to write explicitly only latency timer to be written - 

bool force_write = false;

if ((dev_data->permissive || xen_pcibk_permissive) &&

              offset == PCI_CACHE_LINE_SIZE && size == 4)

             force_write = true;

...

if ((force_write || !handled) && !err) {...}

But then it exposed another issue, the command register field seems not to be restored also

because I think the bits which are to be restored are not in PCI_COMMAND_GUEST mask.

The only hack which always works for me is force write even if the confpsace fields filter doens't allow it - 

+       int remainder = size;

+       

        if (unlikely(verbose_request))

                printk(KERN_DEBUG

                       DRV_NAME ": %s: write request %d bytes at 0x%x = %x\n",

@@ -252,10 +254,11 @@ int xen_pcibk_config_write(struct pci_dev *dev, int offset, int size, u32 value)

                         * special helpers to work correctly.

                         */

                        handled = 1;

+                       remainder -= field->size

                }

        }

 

-       if (!handled && !err) {

+       if ((remainder || !handled) && !err) {

Thanks.

Andrey Grodzovsky.

_______________________________________________
Xen-devel mailing list
Xen-devel@xxxxxxxxxxxxx
http://lists.xen.org/xen-devel