Re: [Qemu-devel] [PATCH qemu v7 3/4] vfio-pci: Allow mmap of MSIX BAR

2018-02-12 Thread David Gibson
On Fri, Feb 09, 2018 at 06:55:02PM +1100, Alexey Kardashevskiy wrote:
> At the moment we unconditionally avoid mapping MSIX data of a BAR and
> emulate MSIX table in QEMU. However it is 1) not always necessary as
> a platform may prodive a paravirt interface for MSIX configuration;
> 2) can affect the speed of MMIO access by emulating them in QEMU when
> frequently accessed registers share same system page with MSIX data,
> this is particularly a problem for systems with the page size bigger
> than 4KB.
> 
> A new capability - VFIO_REGION_INFO_CAP_MSIX_MAPPABLE - has been added
> to the kernel [1] which tells the userspace that mapping of the MSIX data
> is possible now. This makes use of it so from now on QEMU tries mapping
> the entire BAR as a whole and emulate MSIX on top of that.
> 
> [1] 
> https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?id=a32295c612c57990d17fb0f41e7134394b2f35f6
> 
> Signed-off-by: Alexey Kardashevskiy 

Reviewed-by: David Gibson 

> ---
> Changes:
> v7:
> * test iova/llsize against pgmask in vfio_listener_region_add/del
> * s/vfio_is_cap_present/vfio_has_region_cap/
> * added comments here and there
> * s/vdev->msix->table_bar/region-nr/
> ---
>  include/hw/vfio/vfio-common.h |  1 +
>  hw/vfio/common.c  | 15 +++
>  hw/vfio/pci.c |  9 +
>  3 files changed, 25 insertions(+)
> 
> diff --git a/include/hw/vfio/vfio-common.h b/include/hw/vfio/vfio-common.h
> index f3a2ac9..42dd2b0 100644
> --- a/include/hw/vfio/vfio-common.h
> +++ b/include/hw/vfio/vfio-common.h
> @@ -171,6 +171,7 @@ int vfio_get_region_info(VFIODevice *vbasedev, int index,
>   struct vfio_region_info **info);
>  int vfio_get_dev_region_info(VFIODevice *vbasedev, uint32_t type,
>   uint32_t subtype, struct vfio_region_info 
> **info);
> +bool vfio_has_region_cap(VFIODevice *vbasedev, int region, uint16_t 
> cap_type);
>  #endif
>  extern const MemoryListener vfio_prereg_listener;
>  
> diff --git a/hw/vfio/common.c b/hw/vfio/common.c
> index 736f271..b99ae77 100644
> --- a/hw/vfio/common.c
> +++ b/hw/vfio/common.c
> @@ -1464,6 +1464,21 @@ int vfio_get_dev_region_info(VFIODevice *vbasedev, 
> uint32_t type,
>  return -ENODEV;
>  }
>  
> +bool vfio_has_region_cap(VFIODevice *vbasedev, int region, uint16_t cap_type)
> +{
> +struct vfio_region_info *info = NULL;
> +bool ret = false;
> +
> +if (!vfio_get_region_info(vbasedev, region, )) {
> +if (vfio_get_region_info_cap(info, cap_type)) {
> +ret = true;
> +}
> +g_free(info);
> +}
> +
> +return ret;
> +}
> +
>  /*
>   * Interfaces for IBM EEH (Enhanced Error Handling)
>   */
> diff --git a/hw/vfio/pci.c b/hw/vfio/pci.c
> index 879510c..ae9098d 100644
> --- a/hw/vfio/pci.c
> +++ b/hw/vfio/pci.c
> @@ -1294,6 +1294,15 @@ static void vfio_pci_fixup_msix_region(VFIOPCIDevice 
> *vdev)
>  VFIORegion *region = >bars[vdev->msix->table_bar].region;
>  
>  /*
> + * If the host driver allows mapping of a MSIX data, we are going to
> + * do map the entire BAR and emulate MSIX table on top of that.
> + */
> +if (vfio_has_region_cap(>vbasedev, region->nr,
> +VFIO_REGION_INFO_CAP_MSIX_MAPPABLE)) {
> +return;
> +}
> +
> +/*
>   * We expect to find a single mmap covering the whole BAR, anything else
>   * means it's either unsupported or already setup.
>   */

-- 
David Gibson| I'll have my music baroque, and my code
david AT gibson.dropbear.id.au  | minimalist, thank you.  NOT _the_ _other_
| _way_ _around_!
http://www.ozlabs.org/~dgibson


signature.asc
Description: PGP signature


[Qemu-devel] [PATCH qemu v7 3/4] vfio-pci: Allow mmap of MSIX BAR

2018-02-08 Thread Alexey Kardashevskiy
At the moment we unconditionally avoid mapping MSIX data of a BAR and
emulate MSIX table in QEMU. However it is 1) not always necessary as
a platform may prodive a paravirt interface for MSIX configuration;
2) can affect the speed of MMIO access by emulating them in QEMU when
frequently accessed registers share same system page with MSIX data,
this is particularly a problem for systems with the page size bigger
than 4KB.

A new capability - VFIO_REGION_INFO_CAP_MSIX_MAPPABLE - has been added
to the kernel [1] which tells the userspace that mapping of the MSIX data
is possible now. This makes use of it so from now on QEMU tries mapping
the entire BAR as a whole and emulate MSIX on top of that.

[1] 
https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?id=a32295c612c57990d17fb0f41e7134394b2f35f6

Signed-off-by: Alexey Kardashevskiy 
---
Changes:
v7:
* test iova/llsize against pgmask in vfio_listener_region_add/del
* s/vfio_is_cap_present/vfio_has_region_cap/
* added comments here and there
* s/vdev->msix->table_bar/region-nr/
---
 include/hw/vfio/vfio-common.h |  1 +
 hw/vfio/common.c  | 15 +++
 hw/vfio/pci.c |  9 +
 3 files changed, 25 insertions(+)

diff --git a/include/hw/vfio/vfio-common.h b/include/hw/vfio/vfio-common.h
index f3a2ac9..42dd2b0 100644
--- a/include/hw/vfio/vfio-common.h
+++ b/include/hw/vfio/vfio-common.h
@@ -171,6 +171,7 @@ int vfio_get_region_info(VFIODevice *vbasedev, int index,
  struct vfio_region_info **info);
 int vfio_get_dev_region_info(VFIODevice *vbasedev, uint32_t type,
  uint32_t subtype, struct vfio_region_info **info);
+bool vfio_has_region_cap(VFIODevice *vbasedev, int region, uint16_t cap_type);
 #endif
 extern const MemoryListener vfio_prereg_listener;
 
diff --git a/hw/vfio/common.c b/hw/vfio/common.c
index 736f271..b99ae77 100644
--- a/hw/vfio/common.c
+++ b/hw/vfio/common.c
@@ -1464,6 +1464,21 @@ int vfio_get_dev_region_info(VFIODevice *vbasedev, 
uint32_t type,
 return -ENODEV;
 }
 
+bool vfio_has_region_cap(VFIODevice *vbasedev, int region, uint16_t cap_type)
+{
+struct vfio_region_info *info = NULL;
+bool ret = false;
+
+if (!vfio_get_region_info(vbasedev, region, )) {
+if (vfio_get_region_info_cap(info, cap_type)) {
+ret = true;
+}
+g_free(info);
+}
+
+return ret;
+}
+
 /*
  * Interfaces for IBM EEH (Enhanced Error Handling)
  */
diff --git a/hw/vfio/pci.c b/hw/vfio/pci.c
index 879510c..ae9098d 100644
--- a/hw/vfio/pci.c
+++ b/hw/vfio/pci.c
@@ -1294,6 +1294,15 @@ static void vfio_pci_fixup_msix_region(VFIOPCIDevice 
*vdev)
 VFIORegion *region = >bars[vdev->msix->table_bar].region;
 
 /*
+ * If the host driver allows mapping of a MSIX data, we are going to
+ * do map the entire BAR and emulate MSIX table on top of that.
+ */
+if (vfio_has_region_cap(>vbasedev, region->nr,
+VFIO_REGION_INFO_CAP_MSIX_MAPPABLE)) {
+return;
+}
+
+/*
  * We expect to find a single mmap covering the whole BAR, anything else
  * means it's either unsupported or already setup.
  */
-- 
2.11.0