Re: [Qemu-devel] [PATCH 1/2] pci-assign: Fix a bug when map MSI-X table memory failed

2014-04-09 Thread Michael S. Tsirkin
On Thu, Apr 03, 2014 at 01:18:23PM +0800, arei.gong...@huawei.com wrote: > From: Gonglei > > when map MSI-X table memory failed, the dev->msix_table not be > set to NULL, the assigned_dev_unregister_msix_mmio() will case > a segfault when munmap the failed dev->msix_table. > > Signed-off-by: Gon

Re: [Qemu-devel] [PATCH 1/2] pci-assign: Fix a bug when map MSI-X table memory failed

2014-04-08 Thread Michael S. Tsirkin
On Thu, Apr 03, 2014 at 01:18:23PM +0800, arei.gong...@huawei.com wrote: > From: Gonglei > > when map MSI-X table memory failed, the dev->msix_table not be > set to NULL, the assigned_dev_unregister_msix_mmio() will case > a segfault when munmap the failed dev->msix_table. > > Signed-off-by: Gon

Re: [Qemu-devel] [PATCH 1/2] pci-assign: Fix a bug when map MSI-X table memory failed

2014-04-08 Thread Gonglei (Arei)
Hi, mst and alex. Ping... These two bug fix can be accepted for KVM pci-assign ? Thanks. BTW, I have finished the testing work of the Emulex Corporation OneConnect NIC (Lancer) Nic by vfio-pci, and the pass-troughed VF works well. My environment of testing as follows: Host: 3.12.16-0.6.6-defa

[Qemu-devel] [PATCH 1/2] pci-assign: Fix a bug when map MSI-X table memory failed

2014-04-02 Thread arei.gonglei
From: Gonglei when map MSI-X table memory failed, the dev->msix_table not be set to NULL, the assigned_dev_unregister_msix_mmio() will case a segfault when munmap the failed dev->msix_table. Signed-off-by: Gonglei --- hw/i386/kvm/pci-assign.c | 1 + 1 file changed, 1 insertion(+) diff --git a