Add a new virtual device named eth_cvio, it can be used just like
eth_ring, eth_null, etc.
Configured parameters include:
- rx (optional, 1 by default): number of rx, only allowed to be
1 for now.
- tx (optional, 1 by default): number of tx, only allowed
Depends on the type of vhost file: vhost-user is used if the given
path points to a unix socket; vhost-net is used if the given path
points to a char device.
NOTE: we now keep CONFIG_RTE_VIRTIO_VDEV undefined by default, need
to be uncommented when in use.
Signed-off-by: Huawei Xie
Signed-off-by
A new API named rte_eal_get_backfile_info() and a new data
struct back_file is added to obstain information of memory-
backed file info.
Signed-off-by: Huawei Xie
Signed-off-by: Jianfeng Tan
---
lib/librte_eal/common/include/rte_memory.h | 16 +
lib/librte_eal/linuxapp/eal/eal_memor
Originally, there're two cons in using hugepage: a. needs root
privilege to touch /proc/self/pagemap, which is a premise to
alllocate physically contiguous memseg; b. possibly too many
hugepage file are created, especially used with 2M hugepage.
For virtual devices, they don't care about physical-
This patchset is to provide high performance networking interface (virtio)
for container-based DPDK applications. The way of starting DPDK apps in
containers with ownership of NIC devices exclusively is beyond the scope.
The basic idea here is to present a new virtual device (named eth_cvio),
which
Hello David
Thanks for your huge contribution!
May you help to describe more details of why you made these huge changes?
What benefit we can have with your changes. I guess there must have, while you
did not tell that here.
Regards,
Helin
> -Original Message-
> From: dev [mailto:dev-bou
Really ...
Last mail for today ...
On Sun, Jan 10, 2016 at 2:26 PM, David Marchand
wrote:
>
>
> On Sun, Jan 10, 2016 at 2:24 PM, Thomas Monjalon <
> thomas.monjalon at 6wind.com> wrote:
>
>> 2016-01-10 13:58, David Marchand:
>> > Hello Thomas,
>> >
>> > Sorry, don't know how I ended up with this
On Sun, Jan 10, 2016 at 2:24 PM, Thomas Monjalon
wrote:
> 2016-01-10 13:58, David Marchand:
> > Hello Thomas,
> >
> > Sorry, don't know how I ended up with this, I wrote an incorrect mail
> > address for you (I should not send patches after sunday meal ;-)).
>
> No worries :)
> I was not expectin
2016-01-10 13:58, David Marchand:
> Hello Thomas,
>
> Sorry, don't know how I ended up with this, I wrote an incorrect mail
> address for you (I should not send patches after sunday meal ;-)).
No worries :)
I was not expecting a patch so quickly after introducing you in this thread.
Glad to see t
Hello Thomas,
Sorry, don't know how I ended up with this, I wrote an incorrect mail
address for you (I should not send patches after sunday meal ;-)).
--
David Marchand
We could do something ? la modinfo, but let's keep it simple for now.
With this, you can extract the devices that need to be bound to uio / vfio
with tools like objdump :
$ objdump -j rte_pci_id_uio -s build/lib/librte_pmd_fm10k.so
Contents of section rte_pci_id_uio:
15760 8680a415 868
Now that all pci device ids are in their respective drivers, we can remove
this header.
Signed-off-by: David Marchand
---
doc/api/doxy-api-index.md | 1 -
doc/guides/prog_guide/dev_kit_build_system.rst | 75 ++--
lib/librte_eal/common/Makefile
Since we are going to remove a header in next commit, let's first refresh
documentation.
Signed-off-by: David Marchand
---
doc/guides/prog_guide/dev_kit_build_system.rst | 57 +-
1 file changed, 38 insertions(+), 19 deletions(-)
diff --git a/doc/guides/prog_guide/dev_kit
Reused defines from the driver and moved broadcom vendor id macro.
Signed-off-by: David Marchand
---
drivers/net/bnx2x/bnx2x.c | 3 +-
drivers/net/bnx2x/bnx2x_ethdev.c| 25 ---
lib/librte_eal/common/include/rte_pci_dev_ids.h | 60 ---
Moved cisco vendor id since the driver had no such information.
Signed-off-by: David Marchand
---
drivers/net/enic/enic_ethdev.c | 12
lib/librte_eal/common/include/rte_pci_dev_ids.h | 17 -
2 files changed, 4 insertions(+), 25 deletions(-)
diff --g
Moved vmware device ids macro since the driver had no such information.
Signed-off-by: David Marchand
---
drivers/net/vmxnet3/vmxnet3_ethdev.c| 7 +++
lib/librte_eal/common/include/rte_pci_dev_ids.h | 16
2 files changed, 3 insertions(+), 20 deletions(-)
diff -
Reused defines from virtio_pci.h.
Signed-off-by: David Marchand
---
drivers/net/virtio/virtio_ethdev.c | 5 +
lib/librte_eal/common/include/rte_pci_dev_ids.h | 17 -
2 files changed, 1 insertion(+), 21 deletions(-)
diff --git a/drivers/net/virtio/virtio_ethdev.
Since the base driver already defines all pci device ids, no need to
redefine them, let's just drop the previous RTE_PCI_DEV_ID_DECL* stuff.
Signed-off-by: David Marchand
---
drivers/net/fm10k/fm10k_ethdev.c| 8 +++
lib/librte_eal/common/include/rte_pci_dev_ids.h | 29 --
Since the base driver already defines all pci device ids, no need to
redefine them, let's just drop the previous RTE_PCI_DEV_ID_DECL* stuff.
Signed-off-by: David Marchand
---
drivers/net/i40e/i40e_ethdev.c | 18 +++-
drivers/net/i40e/i40e_ethdev_vf.c | 6 ++-
test application and kni still want to know ixgbe pci devices.
So let's create a header in the driver that will be used by them.
Signed-off-by: David Marchand
---
app/test-pmd/Makefile | 2 +
app/test-pmd/cmdline.c | 2 +-
app/test/Makefile
test application and kni still want to know e1000 pci devices.
So let's create headers in the driver that will be used by them.
Signed-off-by: David Marchand
---
app/test/Makefile | 3 +
app/test/test_pci.c | 3 +-
drivers/net/e1000/e
This patchset moves all pci device ids from eal to the pmds that need them
(patches 1 to 8).
Global pci device id list is then removed (patches 9, 10).
In last patch, all those device ids are put in a dedicated section for
retrieval by external tools.
--
David Marchand
David Marchand (11):
e1
On Sun, 10 Jan 2016 13:50:46 +0100
David Marchand wrote:
> +{ RTE_PCI_DEVICE(I40E_INTEL_VENDOR_ID, I40E_DEV_ID_SFP_XL710) },
> +{ RTE_PCI_DEVICE(I40E_INTEL_VENDOR_ID, I40E_DEV_ID_QEMU) },
> +{ RTE_PCI_DEVICE(I40E_INTEL_VENDOR_ID, I40E_DEV_ID_KX_A) },
> +{ RTE_PCI_DEVICE(I40E_INTEL_VENDOR_ID, I40E
23 matches
Mail list logo
