qemu-devel.nongnu.org archive mirror
 help / color / mirror / Atom feed
From: Yongji Xie <xyjxie@linux.vnet.ibm.com>
To: qemu-devel@nongnu.org, alex.williamson@redhat.com
Cc: nikunj@linux.vnet.ibm.com, zhong@linux.vnet.ibm.com,
	aik@ozlabs.ru, paulus@samba.org, mpe@ellerman.id.au,
	warrier@linux.vnet.ibm.com,
	Yongji Xie <xyjxie@linux.vnet.ibm.com>
Subject: [Qemu-devel] [RFC PATCH 3/3] vfio/pci: Add support for mmapping MSI-X table
Date: Fri, 11 Dec 2015 17:05:18 +0800	[thread overview]
Message-ID: <1449824719-3407-4-git-send-email-xyjxie@linux.vnet.ibm.com> (raw)
In-Reply-To: <1449824719-3407-1-git-send-email-xyjxie@linux.vnet.ibm.com>

The VFIO-PCI ioctl VFIO_DEVICE_FLAGS_PCI_MSIX_MMAP flag indicate that
platform support mmapping MSI-X table.

With this flag set, we skip some MSI-X table check so that QEMU can
mmap MSI-X table successfully. And we also raise the priority of mmap
memory region in case of overlap with MSI-X/PBA memory region which
is added in msix_init().

Signed-off-by: Yongji Xie <xyjxie@linux.vnet.ibm.com>
---
 hw/vfio/pci.c |   19 ++++++++++++++-----
 1 file changed, 14 insertions(+), 5 deletions(-)

diff --git a/hw/vfio/pci.c b/hw/vfio/pci.c
index a7c6171..bec301c 100644
--- a/hw/vfio/pci.c
+++ b/hw/vfio/pci.c
@@ -1339,7 +1339,8 @@ static void vfio_mmap_set_enabled(VFIOPCIDevice *vdev, bool enabled)
         }
 
         memory_region_set_enabled(&bar->region.mmap_mem, enabled);
-        if (vdev->msix && vdev->msix->table_bar == i) {
+        if (vdev->msix && vdev->msix->table_bar == i &&
+            !(vdev->vbasedev.flags & VFIO_DEVICE_FLAGS_PCI_MSIX_MMAP)) {
             memory_region_set_enabled(&vdev->msix->mmap_mem, enabled);
         }
     }
@@ -1357,7 +1358,8 @@ static void vfio_unregister_bar(VFIOPCIDevice *vdev, int nr)
 
     memory_region_del_subregion(&bar->region.mem, &bar->region.mmap_mem);
 
-    if (vdev->msix && vdev->msix->table_bar == nr) {
+    if (vdev->msix && vdev->msix->table_bar == nr &&
+        !(vdev->vbasedev.flags & VFIO_DEVICE_FLAGS_PCI_MSIX_MMAP)) {
         memory_region_del_subregion(&bar->region.mem, &vdev->msix->mmap_mem);
     }
 }
@@ -1374,7 +1376,8 @@ static void vfio_unmap_bar(VFIOPCIDevice *vdev, int nr)
 
     munmap(bar->region.mmap, memory_region_size(&bar->region.mmap_mem));
 
-    if (vdev->msix && vdev->msix->table_bar == nr) {
+    if (vdev->msix && vdev->msix->table_bar == nr &&
+        !(vdev->vbasedev.flags & VFIO_DEVICE_FLAGS_PCI_MSIX_MMAP)) {
         munmap(vdev->msix->mmap, memory_region_size(&vdev->msix->mmap_mem));
     }
 }
@@ -1420,7 +1423,8 @@ static void vfio_map_bar(VFIOPCIDevice *vdev, int nr)
      * We can't mmap areas overlapping the MSIX vector table, so we
      * potentially insert a direct-mapped subregion before and after it.
      */
-    if (vdev->msix && vdev->msix->table_bar == nr) {
+    if (vdev->msix && vdev->msix->table_bar == nr &&
+        !(vdev->vbasedev.flags & VFIO_DEVICE_FLAGS_PCI_MSIX_MMAP)) {
         size = vdev->msix->table_offset & qemu_real_host_page_mask;
     }
 
@@ -1429,9 +1433,14 @@ static void vfio_map_bar(VFIOPCIDevice *vdev, int nr)
                       &bar->region.mmap_mem, &bar->region.mmap,
                       size, 0, name)) {
         error_report("%s unsupported. Performance may be slow", name);
+    } else if (vdev->vbasedev.flags & VFIO_DEVICE_FLAGS_PCI_MSIX_MMAP) {
+        memory_region_del_subregion(&bar->region.mem, &bar->region.mmap_mem);
+        memory_region_add_subregion_overlap(&bar->region.mem, 0,
+                      &bar->region.mmap_mem, 1);
     }
 
-    if (vdev->msix && vdev->msix->table_bar == nr) {
+    if (vdev->msix && vdev->msix->table_bar == nr &&
+        !(vdev->vbasedev.flags & VFIO_DEVICE_FLAGS_PCI_MSIX_MMAP)) {
         uint64_t start;
 
         start = REAL_HOST_PAGE_ALIGN((uint64_t)vdev->msix->table_offset +
-- 
1.7.9.5

      parent reply	other threads:[~2015-12-11  9:06 UTC|newest]

Thread overview: 4+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2015-12-11  9:05 [Qemu-devel] [RFC PATCH 0/3] vfio/pci: Add support for mmapping sub-page MMIO BARs and MSI-X table Yongji Xie
2015-12-11  9:05 ` [Qemu-devel] [RFC PATCH 1/3] linux-headers: Update VFIO headers from linux-next tag ToBeFilled Yongji Xie
2015-12-11  9:05 ` [Qemu-devel] [RFC PATCH 2/3] vfio/pci: Add support for mmapped sub-page MMIO BARs Yongji Xie
2015-12-11  9:05 ` Yongji Xie [this message]

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=1449824719-3407-4-git-send-email-xyjxie@linux.vnet.ibm.com \
    --to=xyjxie@linux.vnet.ibm.com \
    --cc=aik@ozlabs.ru \
    --cc=alex.williamson@redhat.com \
    --cc=mpe@ellerman.id.au \
    --cc=nikunj@linux.vnet.ibm.com \
    --cc=paulus@samba.org \
    --cc=qemu-devel@nongnu.org \
    --cc=warrier@linux.vnet.ibm.com \
    --cc=zhong@linux.vnet.ibm.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).