diff mbox series

[2/2] net/virtio: do not require IO permissions

Message ID 1571313388-32142-3-git-send-email-david.marchand@redhat.com (mailing list archive)
State Superseded, archived
Headers
Series Using virtio ethdev ports as non-root |

Checks

Context Check Description
ci/checkpatch success coding style OK
ci/Intel-compilation success Compilation OK
ci/travis-robot success Travis build: passed

Commit Message

David Marchand Oct. 17, 2019, 11:56 a.m. UTC 
  On x86, iopl permissions are only needed when the virtio devices are
bound to a uio kernel module.

When running a dpdk application as non root, the virtio driver was
refusing to register because of this check while it could work when
binding the device to vfio (requires to have a vIOMMU configured).

We still need to call rte_eal_iopl_init() in the constructor so that
the interrupt thread would inherit this permission in the case it could
be used with UIO later.
Log a warning message for the user to understand what is wrong.

Signed-off-by: David Marchand <david.marchand@redhat.com>
---
 drivers/net/virtio/virtio_ethdev.c | 8 ++------
 1 file changed, 2 insertions(+), 6 deletions(-)

Comments

Tiwei Bie Oct. 18, 2019, 8:16 a.m. UTC | #1 
On Thu, Oct 17, 2019 at 01:56:28PM +0200, David Marchand wrote:
> On x86, iopl permissions are only needed when the virtio devices are
> bound to a uio kernel module.
> 
> When running a dpdk application as non root, the virtio driver was
> refusing to register because of this check while it could work when
> binding the device to vfio (requires to have a vIOMMU configured).
> 
> We still need to call rte_eal_iopl_init() in the constructor so that
> the interrupt thread would inherit this permission in the case it could
> be used with UIO later.
> Log a warning message for the user to understand what is wrong.
> 
> Signed-off-by: David Marchand <david.marchand@redhat.com>
> ---
>  drivers/net/virtio/virtio_ethdev.c | 8 ++------
>  1 file changed, 2 insertions(+), 6 deletions(-)
> 
> diff --git a/drivers/net/virtio/virtio_ethdev.c b/drivers/net/virtio/virtio_ethdev.c
> index 7261109..3506ca0 100644
> --- a/drivers/net/virtio/virtio_ethdev.c
> +++ b/drivers/net/virtio/virtio_ethdev.c
> @@ -1995,11 +1995,6 @@ exit:
>  static int eth_virtio_pci_probe(struct rte_pci_driver *pci_drv __rte_unused,
>  	struct rte_pci_device *pci_dev)
>  {
> -	if (rte_eal_iopl_init() != 0) {
> -		PMD_INIT_LOG(ERR, "IOPL call failed - cannot use virtio PMD");
> -		return 1;
> -	}
> -
>  	/* virtio pmd skips probe if device needs to work in vdpa mode */
>  	if (vdpa_mode_selected(pci_dev->device.devargs))
>  		return 1;
> @@ -2031,7 +2026,8 @@ static struct rte_pci_driver rte_virtio_pmd = {
>  
>  RTE_INIT(rte_virtio_pmd_init)
>  {
> -	rte_eal_iopl_init();
> +	if (rte_eal_iopl_init() != 0)
> +		PMD_INIT_LOG(WARNING, "IOPL call failed - virtio devices won't be functional if bound to UIO drivers");

Basically this will undo what below commit did, i.e.
annoying log will show again.

https://github.com/DPDK/dpdk/commit/705dced4a72a1053368c84c4b68f04f028a78b30

Maybe it's better to print this error when we really
need the port IO (legacy device):

https://github.com/DPDK/dpdk/blob/31b798a6f08e9b333b94b8bb26910209aa810b73/drivers/net/virtio/virtio_pci.c#L679-L680
https://github.com/DPDK/dpdk/blob/31b798a6f08e9b333b94b8bb26910209aa810b73/drivers/net/virtio/virtio_ethdev.c#L1834-L1835

Thanks,
Tiwei

>  	rte_pci_register(&rte_virtio_pmd);
>  }
>  
> -- 
> 1.8.3.1
>
David Marchand Oct. 18, 2019, 8:33 a.m. UTC | #2 
On Fri, Oct 18, 2019 at 10:19 AM Tiwei Bie <tiwei.bie@intel.com> wrote:
>
> On Thu, Oct 17, 2019 at 01:56:28PM +0200, David Marchand wrote:
> > On x86, iopl permissions are only needed when the virtio devices are
> > bound to a uio kernel module.
> >
> > When running a dpdk application as non root, the virtio driver was
> > refusing to register because of this check while it could work when
> > binding the device to vfio (requires to have a vIOMMU configured).
> >
> > We still need to call rte_eal_iopl_init() in the constructor so that
> > the interrupt thread would inherit this permission in the case it could
> > be used with UIO later.
> > Log a warning message for the user to understand what is wrong.
> >
> > Signed-off-by: David Marchand <david.marchand@redhat.com>
> > ---
> >  drivers/net/virtio/virtio_ethdev.c | 8 ++------
> >  1 file changed, 2 insertions(+), 6 deletions(-)
> >
> > diff --git a/drivers/net/virtio/virtio_ethdev.c b/drivers/net/virtio/virtio_ethdev.c
> > index 7261109..3506ca0 100644
> > --- a/drivers/net/virtio/virtio_ethdev.c
> > +++ b/drivers/net/virtio/virtio_ethdev.c
> > @@ -1995,11 +1995,6 @@ exit:
> >  static int eth_virtio_pci_probe(struct rte_pci_driver *pci_drv __rte_unused,
> >       struct rte_pci_device *pci_dev)
> >  {
> > -     if (rte_eal_iopl_init() != 0) {
> > -             PMD_INIT_LOG(ERR, "IOPL call failed - cannot use virtio PMD");
> > -             return 1;
> > -     }
> > -
> >       /* virtio pmd skips probe if device needs to work in vdpa mode */
> >       if (vdpa_mode_selected(pci_dev->device.devargs))
> >               return 1;
> > @@ -2031,7 +2026,8 @@ static struct rte_pci_driver rte_virtio_pmd = {
> >
> >  RTE_INIT(rte_virtio_pmd_init)
> >  {
> > -     rte_eal_iopl_init();
> > +     if (rte_eal_iopl_init() != 0)
> > +             PMD_INIT_LOG(WARNING, "IOPL call failed - virtio devices won't be functional if bound to UIO drivers");
>
> Basically this will undo what below commit did, i.e.
> annoying log will show again.
>
> https://github.com/DPDK/dpdk/commit/705dced4a72a1053368c84c4b68f04f028a78b30

Yes.. true.
I wanted to keep a trace for debug, so maybe lower this to debug
level, or just drop this message.


>
> Maybe it's better to print this error when we really
> need the port IO (legacy device):

virtio calls rte_pci_ioport_map, so I suppose we can change the log
level to ERR in the first patch of this series:
http://patchwork.dpdk.org/patch/61370/

And I suppose I could add some context in this log, like the device name.

>
> https://github.com/DPDK/dpdk/blob/31b798a6f08e9b333b94b8bb26910209aa810b73/drivers/net/virtio/virtio_pci.c#L679-L680
> https://github.com/DPDK/dpdk/blob/31b798a6f08e9b333b94b8bb26910209aa810b73/drivers/net/virtio/virtio_ethdev.c#L1834-L1835
Tiwei Bie Oct. 18, 2019, 10:05 a.m. UTC | #3 
On Fri, Oct 18, 2019 at 10:33:47AM +0200, David Marchand wrote:
> On Fri, Oct 18, 2019 at 10:19 AM Tiwei Bie <tiwei.bie@intel.com> wrote:
> > On Thu, Oct 17, 2019 at 01:56:28PM +0200, David Marchand wrote:
> > > On x86, iopl permissions are only needed when the virtio devices are
> > > bound to a uio kernel module.
> > >
> > > When running a dpdk application as non root, the virtio driver was
> > > refusing to register because of this check while it could work when
> > > binding the device to vfio (requires to have a vIOMMU configured).
> > >
> > > We still need to call rte_eal_iopl_init() in the constructor so that
> > > the interrupt thread would inherit this permission in the case it could
> > > be used with UIO later.
> > > Log a warning message for the user to understand what is wrong.
> > >
> > > Signed-off-by: David Marchand <david.marchand@redhat.com>
> > > ---
> > >  drivers/net/virtio/virtio_ethdev.c | 8 ++------
> > >  1 file changed, 2 insertions(+), 6 deletions(-)
> > >
> > > diff --git a/drivers/net/virtio/virtio_ethdev.c b/drivers/net/virtio/virtio_ethdev.c
> > > index 7261109..3506ca0 100644
> > > --- a/drivers/net/virtio/virtio_ethdev.c
> > > +++ b/drivers/net/virtio/virtio_ethdev.c
> > > @@ -1995,11 +1995,6 @@ exit:
> > >  static int eth_virtio_pci_probe(struct rte_pci_driver *pci_drv __rte_unused,
> > >       struct rte_pci_device *pci_dev)
> > >  {
> > > -     if (rte_eal_iopl_init() != 0) {
> > > -             PMD_INIT_LOG(ERR, "IOPL call failed - cannot use virtio PMD");
> > > -             return 1;
> > > -     }
> > > -
> > >       /* virtio pmd skips probe if device needs to work in vdpa mode */
> > >       if (vdpa_mode_selected(pci_dev->device.devargs))
> > >               return 1;
> > > @@ -2031,7 +2026,8 @@ static struct rte_pci_driver rte_virtio_pmd = {
> > >
> > >  RTE_INIT(rte_virtio_pmd_init)
> > >  {
> > > -     rte_eal_iopl_init();
> > > +     if (rte_eal_iopl_init() != 0)
> > > +             PMD_INIT_LOG(WARNING, "IOPL call failed - virtio devices won't be functional if bound to UIO drivers");
> >
> > Basically this will undo what below commit did, i.e.
> > annoying log will show again.
> >
> > https://github.com/DPDK/dpdk/commit/705dced4a72a1053368c84c4b68f04f028a78b30
> 
> Yes.. true.
> I wanted to keep a trace for debug, so maybe lower this to debug
> level, or just drop this message.
> 
> 
> >
> > Maybe it's better to print this error when we really
> > need the port IO (legacy device):
> 
> virtio calls rte_pci_ioport_map, so I suppose we can change the log
> level to ERR in the first patch of this series:
> http://patchwork.dpdk.org/patch/61370/
> 
> And I suppose I could add some context in this log, like the device name.

Sounds like a good idea to me.

Thanks,
Tiwei


> 
> >
> > https://github.com/DPDK/dpdk/blob/31b798a6f08e9b333b94b8bb26910209aa810b73/drivers/net/virtio/virtio_pci.c#L679-L680
> > https://github.com/DPDK/dpdk/blob/31b798a6f08e9b333b94b8bb26910209aa810b73/drivers/net/virtio/virtio_ethdev.c#L1834-L1835
> 
> 
> -- 
> David Marchand
diff mbox series

Patch

diff --git a/drivers/net/virtio/virtio_ethdev.c b/drivers/net/virtio/virtio_ethdev.c
index 7261109..3506ca0 100644
--- a/drivers/net/virtio/virtio_ethdev.c
+++ b/drivers/net/virtio/virtio_ethdev.c
@@ -1995,11 +1995,6 @@  exit:
 static int eth_virtio_pci_probe(struct rte_pci_driver *pci_drv __rte_unused,
 	struct rte_pci_device *pci_dev)
 {
-	if (rte_eal_iopl_init() != 0) {
-		PMD_INIT_LOG(ERR, "IOPL call failed - cannot use virtio PMD");
-		return 1;
-	}
-
 	/* virtio pmd skips probe if device needs to work in vdpa mode */
 	if (vdpa_mode_selected(pci_dev->device.devargs))
 		return 1;
@@ -2031,7 +2026,8 @@  static struct rte_pci_driver rte_virtio_pmd = {
 
 RTE_INIT(rte_virtio_pmd_init)
 {
-	rte_eal_iopl_init();
+	if (rte_eal_iopl_init() != 0)
+		PMD_INIT_LOG(WARNING, "IOPL call failed - virtio devices won't be functional if bound to UIO drivers");
 	rte_pci_register(&rte_virtio_pmd);
 }