On 02/23/2015 11:16 PM, Matthew Hall wrote:
> On Mon, Feb 23, 2015 at 08:48:57AM -0600, Matt Laswell wrote:
>> Apologies in advance for likely being a bit long-winded.
> Long winded is great, helps me get context.
>
>> First, you really need to take cache performance into account when you're
>> cho
> > While VFIO doesn't allow us to map complete BARs with MSI-X tables,
> > it does allow us to map around them in PAGE_SIZE granularity. There
> > might be adapters that provide their registers in the same BAR
> > but on a different page. For example, Intel's NVME adapter, though
> > not a network
The patch introduces following commands.
- port attach [ident]
- port detach [port_id]
- attach: attaching a port
- detach: detaching a port
- ident: pci address of physical device.
Or device name and parameters of virtual device.
(ex. :02:00.0, eth_pcap0,iface=eth0)
- po
This patch adds finalization code to free resources allocated by the
PMD.
v6:
- Fix a paramter of rte_eth_dev_free().
v4:
- Change function name.
Signed-off-by: Tetsuya Mukawa
---
lib/librte_pmd_pcap/rte_eth_pcap.c | 40 ++
1 file changed, 40 insertions(+)
This patch adds a new section for describing port hotplug framework.
Signed-off-by: Tetsuya Mukawa
---
doc/guides/prog_guide/index.rst | 1 +
doc/guides/prog_guide/port_hotplug_framework.rst | 110 +++
2 files changed, 111 insertions(+)
create mode 100644
These functions are used for attaching or detaching a port.
When rte_eal_dev_attach() is called, the function tries to realize the
device name as pci address. If this is done successfully,
rte_eal_dev_attach() will attach physical device port. If not, attaches
virtual devive port.
When rte_eal_dev_
This new parameter is needed to keep device type like PCI or virtual.
Port detaching processes are different between PCI device and virtual
device.
RTE_ETH_DEV_PCI indicates device type is PCI. RTE_ETH_DEV_VIRTUAL
indicates device is virtual.
v12:
- Add missing symbol in version map.
(Thanks to
- Add pci_close_all_drivers()
The function tries to find a driver for the specified device, and
then close the driver.
- Add rte_eal_pci_probe_one() and rte_eal_pci_close_one()
The functions are used for probe and close a device.
First the function tries to find a device that has the specif
The patch adds functions for unmapping igb_uio resources. The patch is only
for Linux and igb_uio environment. VFIO and BSD are not supported.
v9:
- Remove "rte_dev_hotplug.h".
- Remove needless "#ifdef".
(Thanks to Thomas Monjalon and Neil Horman)
- Remove pci_unmap_device(). It will be impleme
The patch adds following functions.
- rte_eth_dev_save()
The function is used for saving current rte_eth_dev structures.
- rte_eth_dev_get_changed_port()
The function receives the rte_eth_dev structures, then compare
these with current values to know which port is actually
attached or deta
The patch adds function pointer to rte_pci_driver and eth_driver
structure. These function pointers are used when ports are detached.
Also, the patch adds rte_eth_dev_uninit(). So far, it's not called
by anywhere, but it will be called when port hotplug function is
implemented.
v10:
- Add size par
This patch adds rte_eth_dev_release_port(). The function is used for
changing an attached status of the device that has specified name.
v9:
- rte_eth_dev_free() is replaced by rte_eth_dev_release_port().
(Thanks to Thomas Monjalon)
v6:
- Use rte_eth_dev structure as the paramter of rte_eth_dev_f
This patch replaces pci_addr_comparison() and memcmp() of pci addresses by
rte_eal_compare_pci_addr().
To compare PCI addresses, rte_eal_compare_pci_addr() doesn't use memcmp().
This is because sizeof(struct rte_pci_addr) returns 6, but actually
this structure is like below.
struct rte_pci_addr {
To remove assumption, do like followings.
This patch adds "RTE_PCI_DRV_DETACHABLE" to drv_flags of rte_pci_driver
structure. The flags indicate the driver can detach devices at runtime.
Also, remove assumption that port will not be detached.
To remove the assumption.
- Add 'attached' member to rt
From: Michael Qiu
With the driver type flag in struct rte_pci_dev, we do not need
to always map uio devices with vfio related function when
vfio enabled.
Signed-off-by: Michael Qiu
Signed-off-by: Tetsuya Mukawa
---
lib/librte_eal/linuxapp/eal/eal_pci.c | 30 +-
1
From: Michael Qiu
Currently, dpdk has no ability to know which type of driver(
vfio-pci/igb_uio/uio_pci_generic) the device used. It only can
check whether vfio is enabled or not staticly.
It really useful to have the flag, becasue different type need to
handle differently in runtime. For exampl
The patch adds CONFIG_RTE_LIBRTE_EAL_HOTPLUG in Linux and BSD
configuration. So far, Hotplug functions only support linux.
v9:
- Move this patch at the top of this patch series.
(Thanks to Thomas Monjalon)
Signed-off-by: Tetsuya Mukawa
---
config/common_bsdapp | 6 ++
config/common_linu
This patch series adds a dynamic port hotplug framework to DPDK.
With the patches, DPDK apps can attach or detach ports at runtime.
The basic concept of the port hotplug is like followings.
- DPDK apps must have responsibility to manage ports.
DPDK apps only know which ports are attached or deta
On 2015/02/23 20:39, Iremonger, Bernard wrote:
the pointer diff --git a/lib/librte_ether/rte_ether_version.map
b/lib/librte_ether/rte_ether_version.map
index f66fd2d..099c769 100644
--- a/lib/librte_ether/rte_ether_version.map
+++ b/lib/librte_ether/rte_ether_version.map
>>
On 2015/02/23 20:12, Bruce Richardson wrote:
> On Mon, Feb 23, 2015 at 11:33:45AM +0900, Tetsuya Mukawa wrote:
>> On 2015/02/23 5:46, Bruce Richardson wrote:
>>> On Sun, Feb 22, 2015 at 02:30:02PM +0900, Tetsuya Mukawa wrote:
Hi,
In my environment, testpmd in latest master branch ret
On 2015/02/23 20:01, Iremonger, Bernard wrote:
>> -Original Message-
>> From: Tetsuya Mukawa [mailto:mukawa at igel.co.jp]
>> Sent: Monday, February 23, 2015 5:09 AM
>> To: dev at dpdk.org
>> Cc: Qiu, Michael; Iremonger, Bernard; maxime.leroy at 6wind.com; Tetsuya
>> Mukawa
>> Subject: [PA
> > This patch tries to remove the RTE_MBUF_REFCNT config options and
> > dependencies
> > by introducing a new mbuf flag IND_ATTACHED_MBUF that would indicate when
> > the mbuf
> > is an indirect attached mbuf, to differentiate between indirect mbufs and
> > mbufs
> > with external memory buffe
2015-02-17 01:21, Zhang, Helin:
> From: dev [mailto:dev-bounces at dpdk.org] On Behalf Of Marc Sune
> > This patch introduces CONFIG_RTE_KNI_PREEMPT_DEFAULT flag. When set to
> > 'no', KNI kernel thread(s) do not call schedule_timeout_interruptible(),
> > which
> > improves overall KNI performance
2015-02-19 15:21, Bruce Richardson:
> Confirmed, this worked for me too.
> Looking at the patches, they look good. However, one thing I think we are
> missing
> is a unit test to verify that all our CRC implementations give the same
> result.
> That would be useful as a sanity check of the softwa
From: Richardson, Bruce
Example showing how callbacks can be used to insert a timestamp
into each packet on RX. On TX the timestamp is used to calculate
the packet latency through the app, in cycles.
Signed-off-by: Bruce Richardson
Signed-off-by: John McNamara
---
MAINTAINERS
From: Richardson, Bruce
Add optional support for inline processing of packets inside the RX
or TX call. For an RX callback, what happens is that we get a set of
packets from the NIC and then pass them to a callback function, if
configured, to allow additional processing to be done on them, e.g.
f
From: Richardson, Bruce
The 'callbacks' member of the rte_eth_dev structure has been renamed
to 'link_intr_cbs' to make it clear that it refers to callbacks from
NIC interrupts. This allows us to add other types of callbacks to
the structure without ambiguity.
Signed-off-by: Bruce Richardson
Si
This patchset is for a small optional addition to the ethdev library,
to add support for callbacks at the RX and TX stages. This allows
packet processing to be done on packets before they get returned
to applications using rte_eth_rx_burst call.
See the RFC cover letter for the use cases:
htt
On Sat, Feb 14, 2015 at 5:28 PM, Neil Horman wrote:
> On Sat, Feb 14, 2015 at 10:32:58AM -0500, Stephen Hemminger wrote:
> > Device driver should log via DPDK log, not to printf which is
> > sends to /dev/null in a daemon application.
> >
> > Signed-off-by: Stephen Hemminger
> > ---
> > lib/lib
From: "Xie, Huawei"
* support calling rte_vhost_driver_register after rte_vhost_driver_session_start
* add mutext to protect fdset from concurrent access
* add busy flag in fdentry. this flag is set before cb and cleared after cb is
finished.
mutex lock scenario in vhost:
* event_dispatch(in r
From: "Xie, Huawei"
for vhost-cuse, ifname is the name of the tap device
for vhost-user, ifname is the name of the unix domain socket path
Signed-off-by: Huawei Xie
Signed-off-by: Przemyslaw Czesnowicz
---
lib/librte_vhost/Makefile | 2 +-
lib/librte_vhost/rte_virtio_net.
From: "Xie, Huawei"
In rte_vhost_driver_register(), vhost unix domain socket listener fd is created
and added to polled(based on select) fdset.
In rte_vhost_driver_session_start(), fds in the fdset are checked for
processing. If there is new connection from qemu, connection fd accepted is
added
From: "Xie, Huawei"
for more generic event driven processing, refer to:
http://libevent.org/
Signed-off-by: Huawei Xie
Acked-by: Konstantin Ananyev
---
lib/librte_vhost/vhost_user/fd_man.c | 207 +++
lib/librte_vhost/vhost_user/fd_man.h | 64 ++
From: "Xie, Huawei"
remove set_memory_table ops
vhost-cuse or vhost-user will both implement their own set_memory_region
handler.
In current vhost-cuse implementation, guest numa memory isn't supported.
Assume that guest memory is backed by only one file.
Signed-off-by: Huawei Xie
Signed-off
From: "Xie, Huawei"
This functions accepts a virtual address and pid(qemu), and maps it into
current process(vhost)'s address space.
The memory behind the virtual address should be backed by a file,
and virtual address should be the starting address.
Signed-off-by: Huawei Xie
---
lib/librte_v
From: "Xie, Huawei"
Signed-off-by: Huawei Xie
---
lib/librte_vhost/vhost_cuse/virtio-net-cdev.c | 257 ++
1 file changed, 257 insertions(+)
create mode 100644 lib/librte_vhost/vhost_cuse/virtio-net-cdev.c
diff --git a/lib/librte_vhost/vhost_cuse/virtio-net-cdev.c
b/li
From: "Xie, Huawei"
vhost-user doesn't need eventfd kernel module to copy fds between processes.
Signed-off-by: Huawei Xie
Signed-off-by: Przemyslaw Czesnowicz
---
lib/librte_vhost/Makefile| 2 +-
lib/librte_vhost/vhost_cuse/eventfd_copy.c | 88 +
From: "Xie, Huawei"
This file defines common operations provided by virtio-net(.c).
Signed-off-by: Huawei Xie
---
lib/librte_vhost/vhost-net-cdev.h| 113 ---
lib/librte_vhost/vhost-net.h | 113 +++
lib/librte_vhost/vho
From: "Xie, Huawei"
vhost-cuse driver will be divided into two parts: cuse driver specific message
handling(in cuse directory) and common message handling(in virtio-net.c).
vhost ioctl message is pre-processed in cuse and then sent to virtio-net
if is not terminated.
virtio-net.c provides commo
From: "Xie, Huawei"
In virtnet_send_command:
/* Caller should know better */
BUG_ON(!virtio_has_feature(vi->vdev, VIRTIO_NET_F_CTRL_VQ) ||
(out + in > VIRTNET_SEND_COMMAND_SG_MAX));
Signed-off-by: Huawei Xie
---
lib/librte_vhost/virtio-net.c | 3 ++-
1 file cha
v3 changes:
* move things around to make all patches compile
Xie, Huawei (11):
lib/librte_vhost: enable VIRTIO_NET_F_CTRL_RX VIRTIO_NET_F_CTRL_RX is
dependant on VIRTIO_NET_F_CTRL_VQ. Observed that virtio-net driver
in guest would crash with only CTRL_RX enabled.
lib/librte_vhost: c
> -Original Message-
> From: Thomas Monjalon [mailto:thomas.monjalon at 6wind.com]
> Sent: Monday, February 23, 2015 3:12 PM
> To: Mcnamara, John
> Cc: dev at dpdk.org; Richardson, Bruce; nhorman at tuxdriver.com;
> stephen at networkplumber.org; Doherty, Declan
> Subject: Re: [PATCH v5 2/3
> -Original Message-
> From: Stephen Hemminger [mailto:stephen at networkplumber.org]
> Sent: Tuesday, February 24, 2015 12:59 AM
> To: Zhou, Danny
> Cc: dev at dpdk.org
> Subject: Re: [dpdk-dev] [PATCH v5 1/6] ethdev: add rx interrupt
> enable/disable functions
>
> On Tue, 24 Feb 2015 0
2015-02-23 15:55, Jastrzebski, MichalX K:
> From: Thomas Monjalon [mailto:thomas.monjalon at 6wind.com]
> > 2015-02-23 14:36, Jastrzebski, MichalX K:
> > > From: Thomas Monjalon [mailto:thomas.monjalon at 6wind.com]
> > > > 2015-02-20 15:46, Jastrzebski, MichalX K:
> > > > > From: dev [mailto:dev-b
Rather than scanning the resource file in sysfs a second time, we
can pull the information on physical addresses of BARs from the
pci resource information already present in the dev structure.
Signed-off-by: Bruce Richardson
---
lib/librte_eal/linuxapp/eal/eal_pci_uio.c | 163 +++
Instead of distinguishing the BAR mappings via offset within a single
file, originally /dev/uioX, switch to mapping each individual bar via
the appropriately numbered resourceX file.
Signed-off-by: Bruce Richardson
---
lib/librte_eal/common/include/rte_pci.h| 2 +-
lib/librte_eal/linuxapp/e
This patch does some cleanup of the uio mapping code to
a) fix issue with mmap of PCI bars reported by Tetsuya and confirmed
by others.
b) eliminate redundant code and reduce scans of /sys
Bruce Richardson (2):
eal: mmap uio resources using resourceX files
eal: populate uio_maps from pci mem
Separately comparing major and minor versions becomes seriously clumsy
when with major version changes, convert the entire version string into
a numeric value (ie 4.6.0 becomes 460 and 5.0.0 becomes 500) and use
that for comparisons, eliminate unnecessary negations while at it.
This makes the compa
When I tried to launch test-pmd on KVM guest of Fedora21, I got following error:
Configuring Port 0 (socket 0)
Fail to configure port 0 tx queues
EAL: Error - exiting with code: 1
Cause: Start ports failed
I found that the error caused here, and actual error message was "TX checksum
offload no
2015-02-20 17:01, Declan Doherty:
> The patch set supports NVGRE on i40e.
>
> It includes:
> - Support RX filters for NVGRE packet. It uses MAC and VLAN to point
>to a queue. The filter types supported are listed below:
>
>1. Inner MAC and Inner VLAN ID
>
>2. Inner MAC address, inne
Hello,
Ok this is coming too late, but anyway, my comments.
On Fri, Feb 20, 2015 at 5:59 PM, Bruce Richardson <
bruce.richardson at intel.com> wrote:
[snip]
diff --git a/lib/librte_eal/linuxapp/eal/eal_pci_uio.c
> b/lib/librte_eal/linuxapp/eal/eal_pci_uio.c
> index 54cce08..2b16fcb 100644
> --
On 02/23/2015 03:55 PM, Neil Horman wrote:
> On Mon, Feb 23, 2015 at 10:19:23AM +0200, Panu Matilainen wrote:
>> On 02/21/2015 09:33 PM, Neil Horman wrote:
>>> On Fri, Feb 20, 2015 at 05:55:21PM -0800, Stephen Hemminger wrote:
On Thu, 12 Feb 2015 16:54:44 +0200
Panu Matilainen wrote:
>>>
2015-02-23 15:02, Zhou, Danny:
> From: Thomas Monjalon [mailto:thomas.monjalon at 6wind.com]
> > 2015-02-23 11:47, Zhou, Danny:
> > > From: Thomas Monjalon [mailto:thomas.monjalon at 6wind.com]
> > > > 2015-02-19 21:48, Zhou Danny:
> > > > > --- a/lib/librte_eal/linuxapp/eal/Makefile
> > > > > +++
Hi John,
2015-02-20 17:03, John McNamara:
> From: Richardson, Bruce
>
> Add optional support for inline processing of packets inside the RX
> or TX call. For an RX callback, what happens is that we get a set of
> packets from the NIC and then pass them to a callback function, if
> configured, to
> -Original Message-
> From: Thomas Monjalon [mailto:thomas.monjalon at 6wind.com]
> Sent: Monday, February 23, 2015 3:47 PM
> To: Jastrzebski, MichalX K
> Cc: Wodkowski, PawelX; dev at dpdk.org; Neil Horman
> Subject: Re: [dpdk-dev] [PATCH v5 0/3] new headroom stats library and
> example a
2015-02-23 14:36, Jastrzebski, MichalX K:
> From: Thomas Monjalon [mailto:thomas.monjalon at 6wind.com]
> > 2015-02-20 15:46, Jastrzebski, MichalX K:
> > > From: dev [mailto:dev-bounces at dpdk.org] On Behalf Of Neil Horman
> > > > On Thu, Feb 19, 2015 at 01:18:41PM +0100, Pawel Wodkowski wrote:
>
> -Original Message-
> From: dev [mailto:dev-bounces at dpdk.org] On Behalf Of Bruce Richardson
> Sent: Monday, February 23, 2015 3:00 PM
> To: dev at dpdk.org
> Subject: Re: [dpdk-dev] [PATCH] eal: mmap uio resources using resourceX files
>
> On Mon, Feb 23, 2015 at 02:57:24PM +, Br
> -Original Message-
> From: Thomas Monjalon [mailto:thomas.monjalon at 6wind.com]
> Sent: Monday, February 23, 2015 11:19 PM
> To: Zhou, Danny
> Cc: Gonzalez Monroy, Sergio; dev at dpdk.org
> Subject: Re: [dpdk-dev] [PATCH v4 4/5] eal: add per rx queue interrupt
> handling based on VFIO
> -Original Message-
> From: Thomas Monjalon [mailto:thomas.monjalon at 6wind.com]
> Sent: Monday, February 23, 2015 9:20 PM
> To: Zhou, Danny
> Cc: dev at dpdk.org
> Subject: Re: [dpdk-dev] [PATCH v4 2/5] ixgbe: enable rx queue interrupts for
> both PF and VF
>
> 2015-02-23 11:23, Zhou,
Declaration of fgets() is
char *fgets(char *str, int size, FILE *stream);
Klocwork complain about passing "sizeof()" as size parameter since
implicit casting size_t to int might cause loss of precision.
Signed-off-by: Pawel Wodkowski
---
lib/librte_cfgfile/rte_cfgfile.c| 2 +-
Fix warning reported by klocwork about size_t to int cast when passing
parameters to parse_set_list().
This patch fix code formating errors that give checkpatch.pl errors
after generating patch.
Signed-off-by: Pawel Wodkowski
---
lib/librte_cmdline/cmdline_parse_portlist.c | 4 ++--
1 file chan
Free kvlist on function exit to avoid memory leak.
Signed-off-by: Pawel Wodkowski
---
lib/librte_pmd_ring/rte_eth_ring.c | 6 +++---
1 file changed, 3 insertions(+), 3 deletions(-)
diff --git a/lib/librte_pmd_ring/rte_eth_ring.c
b/lib/librte_pmd_ring/rte_eth_ring.c
index a23e933..582a621 10064
It is desired that all type of *_free() functions mimic behaviour of
libc free() function. This function does nothing if given parameter is
NULL. This patch add this behaviour for rte_kvargs_free().
Signed-off-by: Pawel Wodkowski
---
lib/librte_kvargs/rte_kvargs.c | 4
lib/librte_kvargs/rte
Declaration for function pointer should be
typedef ret_type (*type_name)(args...)
not
typedef ret_type (type_name)(args...)
although compiler treat both of them the same, the static analysis tool
like klocwork complain about that.
Signed-off-by: Pawel Wodkowski
---
lib/librte_timer/rte_timer.h
Klockwork report some issues against current DPDK version. Most of them need
only cosmetic code changes (changing type of variable or adding explicit cast).
One issue related with ring pmd fix real memory leak problem.
Pawel Wodkowski (5):
rte_timer: fix invalid declaration of rte_timer_cb_t
2015-02-23 13:53, Czesnowicz, Przemyslaw:
> > I tried to locally applied the patches, waiting comments are closed.
> > But I stopped after patch 04/11 which makes compilation failing.
> > I'm so sorry that we still don't have a vhost-user support integrated in
> > DPDK.
> > I feel it won't be read
> -Original Message-
> From: Thomas Monjalon [mailto:thomas.monjalon at 6wind.com]
> Sent: Monday, February 23, 2015 9:28 PM
> To: Zhou, Danny
> Cc: Gonzalez Monroy, Sergio; dev at dpdk.org
> Subject: Re: [dpdk-dev] [PATCH v4 4/5] eal: add per rx queue interrupt
> handling based on VFIO
>
On Mon, Feb 23, 2015 at 02:57:24PM +, Bruce Richardson wrote:
> Instead of distinguishing the BAR mappings via offset within a single
> file, originally /dev/uioX, switch to mapping each individual bar via
> the appropriately numbered resourceX file.
>
> Signed-off-by: Bruce Richardson
> ---
On 23/02/2015 13:52, Neil Horman wrote:
> On Mon, Feb 23, 2015 at 10:25:01AM +, Gonzalez Monroy, Sergio wrote:
>> On 22/02/2015 23:37, Neil Horman wrote:
>>> On Fri, Feb 20, 2015 at 02:31:36PM +, Gonzalez Monroy, Sergio wrote:
On 13/02/2015 12:51, Neil Horman wrote:
> On Fri, Feb 1
Instead of distinguishing the BAR mappings via offset within a single
file, originally /dev/uioX, switch to mapping each individual bar via
the appropriately numbered resourceX file.
Signed-off-by: Bruce Richardson
---
lib/librte_eal/common/include/rte_pci.h| 2 +-
lib/librte_eal/linuxapp/e
> -Original Message-
> From: Thomas Monjalon [mailto:thomas.monjalon at 6wind.com]
> Sent: Monday, February 23, 2015 12:46 PM
> To: Wodkowski, PawelX
> Cc: dev at dpdk.org; Jastrzebski, MichalX K; Neil Horman
> Subject: Re: [dpdk-dev] [PATCH v5 0/3] new headroom stats library and
> example
Hi Tetsuya,
On Mon, Feb 23, 2015 at 6:09 AM, Tetsuya Mukawa wrote:
> These functions are used for attaching or detaching a port.
[...]
>
> +static int
> +rte_eal_vdev_init(const char *name, const char *args)
> +{
> + struct rte_driver *driver;
> +
> + if (name == NULL)
> +
2015-02-23 11:47, Zhou, Danny:
> From: Thomas Monjalon [mailto:thomas.monjalon at 6wind.com]
> > 2015-02-19 21:48, Zhou Danny:
> > > --- a/lib/librte_eal/linuxapp/eal/Makefile
> > > +++ b/lib/librte_eal/linuxapp/eal/Makefile
> > > @@ -43,6 +43,7 @@ CFLAGS += -I$(SRCDIR)/include
> > > CFLAGS += -I$
2015-02-23 11:23, Zhou, Danny:
> I noticed the V4 patch conflicts with the latest code on the main branch due
> to lots of code merged, and it
> is mentioned by Jun Xu in previous email, and I will have to rebase the patch
> and send out V5 version.
Maybe you misunderstood my comment.
I'm quoti
> 2015-02-23 13:53, Czesnowicz, Przemyslaw:
> > > I tried to locally applied the patches, waiting comments are closed.
> > > But I stopped after patch 04/11 which makes compilation failing.
> > > I'm so sorry that we still don't have a vhost-user support integrated in
> > > DPDK.
> > > I feel it w
This patch adds port hotplug support to Null PMD.
v9:
- Use rte_eth_dev_release_port() instead of rte_eth_dev_free().
v7:
- Add parameter checkings.
(Thanks to Iremonger, Bernard)
v6:
- Fix a parameter of rte_eth_dev_free().
v4:
- Fix commit title.
Signed-off-by: Tetsuya Mukawa
---
lib/l
Null PMD is a driver of the virtual device particularly designed to measure
performance of DPDK PMDs. When an application call rx, Null PMD just allocates
mbufs and returns those. Also tx, the PMD just frees mbufs.
The PMD has following options.
- size: specify packe size allocated by RX. Default
The patch introduces following commands.
- port attach [ident]
- port detach [port_id]
- attach: attaching a port
- detach: detaching a port
- ident: pci address of physical device.
Or device name and parameters of virtual device.
(ex. :02:00.0, eth_pcap0,iface=eth0)
- po
This patch adds finalization code to free resources allocated by the
PMD.
v6:
- Fix a paramter of rte_eth_dev_free().
v4:
- Change function name.
Signed-off-by: Tetsuya Mukawa
---
lib/librte_pmd_pcap/rte_eth_pcap.c | 40 ++
1 file changed, 40 insertions(+)
This patch adds a new section for describing port hotplug framework.
Signed-off-by: Tetsuya Mukawa
---
doc/guides/prog_guide/index.rst | 1 +
doc/guides/prog_guide/port_hotplug_framework.rst | 110 +++
2 files changed, 111 insertions(+)
create mode 100644
These functions are used for attaching or detaching a port.
When rte_eal_dev_attach() is called, the function tries to realize the
device name as pci address. If this is done successfully,
rte_eal_dev_attach() will attach physical device port. If not, attaches
virtual devive port.
When rte_eal_dev_
This new parameter is needed to keep device type like PCI or virtual.
Port detaching processes are different between PCI device and virtual
device.
RTE_ETH_DEV_PCI indicates device type is PCI. RTE_ETH_DEV_VIRTUAL
indicates device is virtual.
v10:
- Change order of version.map.
(Thanks to Thomas
- Add pci_close_all_drivers()
The function tries to find a driver for the specified device, and
then close the driver.
- Add rte_eal_pci_probe_one() and rte_eal_pci_close_one()
The functions are used for probe and close a device.
First the function tries to find a device that has the specif
The patch adds functions for unmapping igb_uio resources. The patch is only
for Linux and igb_uio environment. VFIO and BSD are not supported.
v9:
- Remove "rte_dev_hotplug.h".
- Remove needless "#ifdef".
(Thanks to Thomas Monjalon and Neil Horman)
- Remove pci_unmap_device(). It will be impleme
The patch adds following functions.
- rte_eth_dev_save()
The function is used for saving current rte_eth_dev structures.
- rte_eth_dev_get_changed_port()
The function receives the rte_eth_dev structures, then compare
these with current values to know which port is actually
attached or deta
The patch adds function pointer to rte_pci_driver and eth_driver
structure. These function pointers are used when ports are detached.
Also, the patch adds rte_eth_dev_uninit(). So far, it's not called
by anywhere, but it will be called when port hotplug function is
implemented.
v10:
- Add size par
This patch adds rte_eth_dev_release_port(). The function is used for
changing an attached status of the device that has specified name.
v9:
- rte_eth_dev_free() is replaced by rte_eth_dev_release_port().
(Thanks to Thomas Monjalon)
v6:
- Use rte_eth_dev structure as the paramter of rte_eth_dev_f
This patch replaces pci_addr_comparison() and memcmp() of pci addresses by
rte_eal_compare_pci_addr().
To compare PCI addresses, rte_eal_compare_pci_addr() doesn't use memcmp().
This is because sizeof(struct rte_pci_addr) returns 6, but actually
this structure is like below.
struct rte_pci_addr {
To remove assumption, do like followings.
This patch adds "RTE_PCI_DRV_DETACHABLE" to drv_flags of rte_pci_driver
structure. The flags indicate the driver can detach devices at runtime.
Also, remove assumption that port will not be detached.
To remove the assumption.
- Add 'attached' member to rt
From: Michael Qiu
With the driver type flag in struct rte_pci_dev, we do not need
to always map uio devices with vfio related function when
vfio enabled.
Signed-off-by: Michael Qiu
Signed-off-by: Tetsuya Mukawa
---
lib/librte_eal/linuxapp/eal/eal_pci.c | 30 +-
1
From: Michael Qiu
Currently, dpdk has no ability to know which type of driver(
vfio-pci/igb_uio/uio_pci_generic) the device used. It only can
check whether vfio is enabled or not staticly.
It really useful to have the flag, becasue different type need to
handle differently in runtime. For exampl
The patch adds CONFIG_RTE_LIBRTE_EAL_HOTPLUG in Linux and BSD
configuration. So far, Hotplug functions only support linux.
v9:
- Move this patch at the top of this patch series.
(Thanks to Thomas Monjalon)
Signed-off-by: Tetsuya Mukawa
---
config/common_bsdapp | 6 ++
config/common_linu
This patch series adds a dynamic port hotplug framework to DPDK.
With the patches, DPDK apps can attach or detach ports at runtime.
The basic concept of the port hotplug is like followings.
- DPDK apps must have responsibility to manage ports.
DPDK apps only know which ports are attached or deta
> I tried to locally applied the patches, waiting comments are closed.
> But I stopped after patch 04/11 which makes compilation failing.
> I'm so sorry that we still don't have a vhost-user support integrated in DPDK.
> I feel it won't be ready in next days to be able to enter in 2.0 version.
Hi
Fix for Klockwork identified issue.
Signed-off-by: John McNamara
---
lib/librte_pmd_af_packet/rte_eth_af_packet.c | 11 ---
1 files changed, 8 insertions(+), 3 deletions(-)
diff --git a/lib/librte_pmd_af_packet/rte_eth_af_packet.c
b/lib/librte_pmd_af_packet/rte_eth_af_packet.c
index
> -Original Message-
> From: dev [mailto:dev-bounces at dpdk.org] On Behalf Of John McNamara
> Sent: Monday, February 23, 2015 2:17 PM
> To: dev at dpdk.org
> Subject: [dpdk-dev] [PATCH] ring: fix minor memory leak of kvlist in dev init
>
> Fix for Klockwork identified issue.
>
> Signed
On Mon, Feb 23, 2015 at 02:58:30PM +, Gonzalez Monroy, Sergio wrote:
> On 23/02/2015 13:52, Neil Horman wrote:
> >On Mon, Feb 23, 2015 at 10:25:01AM +, Gonzalez Monroy, Sergio wrote:
> >>On 22/02/2015 23:37, Neil Horman wrote:
> >>>On Fri, Feb 20, 2015 at 02:31:36PM +, Gonzalez Monroy,
Fix for Klockwork identified issue.
Signed-off-by: John McNamara
---
lib/librte_pmd_ring/rte_eth_ring.c |3 ++-
1 files changed, 2 insertions(+), 1 deletions(-)
diff --git a/lib/librte_pmd_ring/rte_eth_ring.c
b/lib/librte_pmd_ring/rte_eth_ring.c
index a23e933..88a1382 100644
--- a/lib/libr
On Mon, Feb 23, 2015 at 08:48:57AM -0600, Matt Laswell wrote:
> Apologies in advance for likely being a bit long-winded.
Long winded is great, helps me get context.
> First, you really need to take cache performance into account when you're
> choosing a data structure. Something like a balanced
2015-02-20 15:46, Jastrzebski, MichalX K:
> From: dev [mailto:dev-bounces at dpdk.org] On Behalf Of Neil Horman
> > On Thu, Feb 19, 2015 at 01:18:41PM +0100, Pawel Wodkowski wrote:
> > > Hi community,
> > > I would like to introduce library for measuring load of some arbitrary
> > > jobs.
> > > It
1 - 100 of 119 matches
Mail list logo
