[0/4] make vhost PMD available for secondary processes

  From: Itsuro Oda <oda@valinux.co.jp>

vhost PMD has not been available for secondary processes since
DPDK v18.11.  (https://bugs.dpdk.org/show_bug.cgi?id=194)
(for a long term !)
This series of patches intend to make vhost PMD available for
secondary processes.
Because now setting vhost driver to communicate with a vhost-user
master (ex. Qemu) is accomplished by the probe function of the
primary process, only the primary process can be a vhost-user
slave.
With this patch, setting vhost driver is delayed at eth_dev
configuration in order to be able to set it from a secondary
process. Because (in the first place,) setting vhost driver is not
necessary to be done at probe (it is enough to be done up to eth_dev
start), this fix is no problem for the primary process.
There is a precondition that the same process has to operate
a vhost interface from begining to end (eth_dev configuration to
eth_dev close). (This patch leaves it to user's responsibility.)
This precondition will not be a problem in most use cases
(including SPP).

Itsuro Oda (4):
  net/vhost: remove an unused member
  net/vhost: allocate iface_name from heap
  net/vhost: delay vhost driver setup
  net/vhost: make secondary probe complete

 drivers/net/vhost/rte_eth_vhost.c | 152 +++++++++++++++++-------------
 1 file changed, 88 insertions(+), 64 deletions(-)
  

Hi,

I will fix to add signed-off-by and correct spelling error
according to test-report@dpdk.
Please check the content until then.

Thanks.

On Wed,  8 Jan 2020 15:25:06 +0900
oda@valinux.co.jp wrote:

> From: Itsuro Oda <oda@valinux.co.jp>
> 
> vhost PMD has not been available for secondary processes since
> DPDK v18.11.  (https://bugs.dpdk.org/show_bug.cgi?id=194)
> (for a long term !)
> This series of patches intend to make vhost PMD available for
> secondary processes.
> Because now setting vhost driver to communicate with a vhost-user
> master (ex. Qemu) is accomplished by the probe function of the
> primary process, only the primary process can be a vhost-user
> slave.
> With this patch, setting vhost driver is delayed at eth_dev
> configuration in order to be able to set it from a secondary
> process. Because (in the first place,) setting vhost driver is not
> necessary to be done at probe (it is enough to be done up to eth_dev
> start), this fix is no problem for the primary process.
> There is a precondition that the same process has to operate
> a vhost interface from begining to end (eth_dev configuration to
> eth_dev close). (This patch leaves it to user's responsibility.)
> This precondition will not be a problem in most use cases
> (including SPP).
> 
> Itsuro Oda (4):
>   net/vhost: remove an unused member
>   net/vhost: allocate iface_name from heap
>   net/vhost: delay vhost driver setup
>   net/vhost: make secondary probe complete
> 
>  drivers/net/vhost/rte_eth_vhost.c | 152 +++++++++++++++++-------------
>  1 file changed, 88 insertions(+), 64 deletions(-)
> 
> -- 
> 2.17.0
  

vhost PMD has not been available for secondary processes since
DPDK v18.11.  (https://bugs.dpdk.org/show_bug.cgi?id=194)
(for a long term !)
This series of patches intend to make vhost PMD available for
secondary processes.
Because now setting vhost driver to communicate with a vhost-user
master (ex. Qemu) is accomplished by the probe function of the
primary process, only the primary process can be a vhost-user
slave.
With this patch, setting vhost driver is delayed at eth_dev
configuration in order to be able to set it from a secondary
process. Because (in the first place,) setting vhost driver is not
necessary to be done at probe (it is enough to be done up to eth_dev
start), this fix is no problem for the primary process.
There is a precondition that the same process has to operate
a vhost interface from beginning to end (from eth_dev configuration
to eth_dev close). (This patch leaves it to user's responsibility.)
This precondition will not be a problem in most use cases
(including SPP).

v2:
- add signed-off-by
- fix spelling error

Itsuro Oda (4):
  net/vhost: remove an unused member
  net/vhost: allocate iface_name from heap
  net/vhost: delay vhost driver setup
  net/vhost: make secondary probe complete

 drivers/net/vhost/rte_eth_vhost.c | 152 +++++++++++++++++-------------
 1 file changed, 88 insertions(+), 64 deletions(-)
  

Hi vhost PMD maitainers,

I have not got any feedback yet. 
Since this is the first time I submit a patch, something
may be wrong, would you tell me what should I do ?

Thanks.

On Thu,  9 Jan 2020 08:22:05 +0900
Itsuro Oda <oda@valinux.co.jp> wrote:

> vhost PMD has not been available for secondary processes since
> DPDK v18.11.  (https://bugs.dpdk.org/show_bug.cgi?id=194)
> (for a long term !)
> This series of patches intend to make vhost PMD available for
> secondary processes.
> Because now setting vhost driver to communicate with a vhost-user
> master (ex. Qemu) is accomplished by the probe function of the
> primary process, only the primary process can be a vhost-user
> slave.
> With this patch, setting vhost driver is delayed at eth_dev
> configuration in order to be able to set it from a secondary
> process. Because (in the first place,) setting vhost driver is not
> necessary to be done at probe (it is enough to be done up to eth_dev
> start), this fix is no problem for the primary process.
> There is a precondition that the same process has to operate
> a vhost interface from beginning to end (from eth_dev configuration
> to eth_dev close). (This patch leaves it to user's responsibility.)
> This precondition will not be a problem in most use cases
> (including SPP).
> 
> v2:
> - add signed-off-by
> - fix spelling error
> 
> Itsuro Oda (4):
>   net/vhost: remove an unused member
>   net/vhost: allocate iface_name from heap
>   net/vhost: delay vhost driver setup
>   net/vhost: make secondary probe complete
> 
>  drivers/net/vhost/rte_eth_vhost.c | 152 +++++++++++++++++-------------
>  1 file changed, 88 insertions(+), 64 deletions(-)
> 
> -- 
> 2.17.0
  

Hi Itsuro,

On 1/20/20 3:17 AM, Itsuro ODA wrote:
> Hi vhost PMD maitainers,
> 
> I have not got any feedback yet. 
> Since this is the first time I submit a patch, something
> may be wrong, would you tell me what should I do ?

Sorry for the delay, and thanks for the contribution.

You series does not apply properly on dpdk-next-virtio master branch:
https://git.dpdk.org/next/dpdk-next-virtio

I will review it, so when doing v3, please rebase it.

More generally, you series comprises fixes (patch 2 to 4), and
cleanup (patch 1).

Cleanup patch should be the last, in order to ease the backporting
of the fixes to LTSes (we avoid backporting cleanup patches).

Regarding fixes patches, it should tag the faulty commit in
master branch, and stable@dpdk.org should be Cc'ed.

Example of commit message with fixes tag:
http://patches.dpdk.org/patch/63305/

Finally, when you'll post the v3, please prefix the patches subject
with the revision number:

git format-patch --subject-prefix"PATCH v3" ...

Thanks,
Maxime

> Thanks.
> 
> On Thu,  9 Jan 2020 08:22:05 +0900
> Itsuro Oda <oda@valinux.co.jp> wrote:
> 
>> vhost PMD has not been available for secondary processes since
>> DPDK v18.11.  (https://bugs.dpdk.org/show_bug.cgi?id=194)
>> (for a long term !)
>> This series of patches intend to make vhost PMD available for
>> secondary processes.
>> Because now setting vhost driver to communicate with a vhost-user
>> master (ex. Qemu) is accomplished by the probe function of the
>> primary process, only the primary process can be a vhost-user
>> slave.
>> With this patch, setting vhost driver is delayed at eth_dev
>> configuration in order to be able to set it from a secondary
>> process. Because (in the first place,) setting vhost driver is not
>> necessary to be done at probe (it is enough to be done up to eth_dev
>> start), this fix is no problem for the primary process.
>> There is a precondition that the same process has to operate
>> a vhost interface from beginning to end (from eth_dev configuration
>> to eth_dev close). (This patch leaves it to user's responsibility.)
>> This precondition will not be a problem in most use cases
>> (including SPP).
>>
>> v2:
>> - add signed-off-by
>> - fix spelling error
>>
>> Itsuro Oda (4):
>>   net/vhost: remove an unused member
>>   net/vhost: allocate iface_name from heap
>>   net/vhost: delay vhost driver setup
>>   net/vhost: make secondary probe complete
>>
>>  drivers/net/vhost/rte_eth_vhost.c | 152 +++++++++++++++++-------------
>>  1 file changed, 88 insertions(+), 64 deletions(-)
>>
>> -- 
>> 2.17.0
>
  

Hi Maxime,

Thank you for your reply and review.
I will make fixes of the patches according to your
indication and post them.

Thanks.
Itsuro Oda

On Tue, 4 Feb 2020 18:54:23 +0100
Maxime Coquelin <maxime.coquelin@redhat.com> wrote:

> Hi Itsuro,
> 
> On 1/20/20 3:17 AM, Itsuro ODA wrote:
> > Hi vhost PMD maitainers,
> > 
> > I have not got any feedback yet. 
> > Since this is the first time I submit a patch, something
> > may be wrong, would you tell me what should I do ?
> 
> Sorry for the delay, and thanks for the contribution.
> 
> You series does not apply properly on dpdk-next-virtio master branch:
> https://git.dpdk.org/next/dpdk-next-virtio
> 
> I will review it, so when doing v3, please rebase it.
> 
> More generally, you series comprises fixes (patch 2 to 4), and
> cleanup (patch 1).
> 
> Cleanup patch should be the last, in order to ease the backporting
> of the fixes to LTSes (we avoid backporting cleanup patches).
> 
> Regarding fixes patches, it should tag the faulty commit in
> master branch, and stable@dpdk.org should be Cc'ed.
> 
> Example of commit message with fixes tag:
> http://patches.dpdk.org/patch/63305/
> 
> Finally, when you'll post the v3, please prefix the patches subject
> with the revision number:
> 
> git format-patch --subject-prefix"PATCH v3" ...
> 
> Thanks,
> Maxime
> 
> > Thanks.
> > 
> > On Thu,  9 Jan 2020 08:22:05 +0900
> > Itsuro Oda <oda@valinux.co.jp> wrote:
> > 
> >> vhost PMD has not been available for secondary processes since
> >> DPDK v18.11.  (https://bugs.dpdk.org/show_bug.cgi?id=194)
> >> (for a long term !)
> >> This series of patches intend to make vhost PMD available for
> >> secondary processes.
> >> Because now setting vhost driver to communicate with a vhost-user
> >> master (ex. Qemu) is accomplished by the probe function of the
> >> primary process, only the primary process can be a vhost-user
> >> slave.
> >> With this patch, setting vhost driver is delayed at eth_dev
> >> configuration in order to be able to set it from a secondary
> >> process. Because (in the first place,) setting vhost driver is not
> >> necessary to be done at probe (it is enough to be done up to eth_dev
> >> start), this fix is no problem for the primary process.
> >> There is a precondition that the same process has to operate
> >> a vhost interface from beginning to end (from eth_dev configuration
> >> to eth_dev close). (This patch leaves it to user's responsibility.)
> >> This precondition will not be a problem in most use cases
> >> (including SPP).
> >>
> >> v2:
> >> - add signed-off-by
> >> - fix spelling error
> >>
> >> Itsuro Oda (4):
> >>   net/vhost: remove an unused member
> >>   net/vhost: allocate iface_name from heap
> >>   net/vhost: delay vhost driver setup
> >>   net/vhost: make secondary probe complete
> >>
> >>  drivers/net/vhost/rte_eth_vhost.c | 152 +++++++++++++++++-------------
> >>  1 file changed, 88 insertions(+), 64 deletions(-)
> >>
> >> -- 
> >> 2.17.0
> >