[dpdk-dev] [PATCH v5 4/4] DO NOT MERGE: Tests for new eventfd_link module

To use: 1. Compile and load the new eventfd_link module (as root): # (cd lib/librte_vhost/eventfd_link; make; insmod ./eventfd_link.ko) 2. Compile the test program: $ make -C test_eventfd_copy 3. Run it as root: # sudo ./test_eventfd_copy/test_eventfd_copy --check Stealing FD OK --- test_event

[dpdk-dev] [PATCH v5 3/4] vhost: using EVENTFD_COPY2

Signed-off-by: Pavel Boldin --- lib/librte_vhost/vhost_cuse/eventfd_copy.c | 54 ++-- lib/librte_vhost/vhost_cuse/eventfd_copy.h | 6 lib/librte_vhost/vhost_cuse/vhost-net-cdev.c | 3 ++ 3 files changed, 44 insertions(+), 19 deletions(-) diff --git a/lib/librte

[dpdk-dev] [PATCH v5 2/4] vhost: add EVENTFD_COPY2 ioctl

Signed-off-by: Pavel Boldin --- lib/librte_vhost/eventfd_link/eventfd_link.c | 61 lib/librte_vhost/eventfd_link/eventfd_link.h | 28 ++--- 2 files changed, 84 insertions(+), 5 deletions(-) diff --git a/lib/librte_vhost/eventfd_link/eventfd_link.c b/lib/libr

[dpdk-dev] [PATCH v5 1/4] vhost: eventfd_link: refactoring EVENTFD_COPY handler

* Move ioctl `EVENTFD_COPY' code to a separate function * Remove extra #includes * Introduce function fget_from_files * Fix ioctl return values Signed-off-by: Pavel Boldin --- lib/librte_vhost/eventfd_link/eventfd_link.c | 188 +++ 1 file changed, 103 insertions(+), 85 de

[dpdk-dev] Build instructions

On 8/28/15, 2:49 PM, "dev on behalf of Ratliff, Stanley" wrote: >Hello everyone, > >I'm a DPDK newbie, so my apologies for the na?ve question. I'm trying to >use the instructions at http://dpdk.org/doc/quick-start. > >I've cloned the DPDK repo, & started with the make command. I got: > >stanr at

[dpdk-dev] Build instructions

Hello again, My mistake. Typo. Sorry for the noise. Regards, Stan > -Original Message- > From: dev [mailto:dev-bounces at dpdk.org] On Behalf Of Ratliff, Stanley > Sent: Friday, August 28, 2015 3:49 PM > To: dev at dpdk.org > Subject: [dpdk-dev] Build instructions > > Hello everyone,

[dpdk-dev] Build instructions

Hello everyone, I'm a DPDK newbie, so my apologies for the na?ve question. I'm trying to use the instructions at http://dpdk.org/doc/quick-start. I've cloned the DPDK repo, & started with the make command. I got: stanr at stansubu:~$ git clone http://dpdk.org/git/dpdk Cloning into 'dpdk'...

[dpdk-dev] [RFC PATCH 0/6] remove pci driver from vdevs

On 8/28/15, 5:32 AM, "dev on behalf of Neil Horman" wrote: >On Fri, Aug 28, 2015 at 08:15:47AM +, Iremonger, Bernard wrote: >> Hi John, >> >> > -Original Message- >> > From: John W. Linville [mailto:linville at tuxdriver.com] >> > Sent: Thursday, August 27, 2015 6:44 PM >> > To: Irem

[dpdk-dev] [PATCH] kni: set kni mac on ioctl_create

Hi, Probably I missed something in understanding why the mac is not set on kni creation. Any comments would be highly appreciated. Thanks, Sergey On Friday 28 August 2015 16:06:27 Sergey Balabanov wrote: > There is a situation when ioctl returns zero mac address (00:00:00:00:00:00) > for just c

[dpdk-dev] [PATCH] kni: set kni mac on ioctl_create

There is a situation when ioctl returns zero mac address (00:00:00:00:00:00) for just created kni. The situation happens because kni mac is set on 'ipconfig up' event (kni_net_open callback) not on kni creation (kni_ioctl_create). Signed-off-by: Sergey Balabanov --- lib/librte_eal/linuxapp/kni/

[dpdk-dev] flow_director_filter error!!

cc-ing Bruce -- 'cos found an earlier thread on this... http://dpdk.info/ml/archives/dev/2014-February/001381.html Any pointers Bruce -- on how to setup flow-director-filter using x540? Thanks -Navneet -Original Message- From: Navneet Rao Sent: Friday, August 28, 2015 3:11 PM To: De L

[dpdk-dev] flow_director_filter error!!

Hi John et al: I am using the X540, and setting it up to receive UDP traffic. Want to use the flow-director filter, to steer packets to different receive queues. Using the testpmd app, I am running into the following error -- testpmd> flow_director_filter 0 add flow ipv4-udp src 192.168.10.100

[dpdk-dev] [RFC PATCH 0/6] remove pci driver from vdevs

On Fri, Aug 28, 2015 at 08:15:47AM +, Iremonger, Bernard wrote: > Hi John, > > > -Original Message- > > From: John W. Linville [mailto:linville at tuxdriver.com] > > Sent: Thursday, August 27, 2015 6:44 PM > > To: Iremonger, Bernard > > Cc: dev at dpdk.org > > Subject: Re: [dpdk-dev] [

[dpdk-dev] BUG - KNI broken in 4.2 kernel

On Thu, Aug 27, 2015 at 10:45:43AM -0700, Stephen Hemminger wrote: > On Thu, 27 Aug 2015 15:56:16 + > "Zhang, Helin" wrote: > > > Based on my experience, only one or two users asked for ethtool support, > > then we have it. Before that time, we don?t have KNI ethtool support. > > I did not m

[dpdk-dev] Hugetlbfs mounted with size option leads to mmap failure

Hi Monroy, As hugetlbfs mounted with option "size=xxx", and xxx is smaller than that from /sys/kernel/mm/hugepages/hugepages-1048576kB/nr_hugepages, error is raised like, EAL: map_all_hugepages(): mmap failed: Cannot allocate memory. Although this can be avoided by the parameter of "--sock-mem=",

[dpdk-dev] [RFC PATCH] vhost: Add VHOST PMD

The patch introduces a new PMD. This PMD is implemented as thin wrapper of librte_vhost. It means librte_vhost is also needed to compile the PMD. The PMD can have 'iface' parameter like below to specify a path to connect to a virtio-net device. $ ./testpmd -c f -n 4 --vdev 'eth_vhost0,iface=/tmp/s

[dpdk-dev] [RFC PATCH] Add VHOST PMD

The patch introduces a new PMD. This PMD is implemented as thin wrapper of librte_vhost. To work the PMD correctly, below patches are needed. - [PATCH 1/3] vhost: Fix return value of GET_VRING_BASE message - [PATCH 2/3] vhost: Fix RESET_OWNER handling not to close callfd - [PATCH 3/3] vhost: Fi

[dpdk-dev] rte_eth_rx_burst only returns up to 32 packets

Hi, The documentation of this method says: Applications implementing a "retrieve as much received packets as possible" policy can check this specific case and keep invoking the rte_eth_rx_burst() function until a v

[dpdk-dev] about new timesync feature in 2.1.0

Hi, I'm very interested in getting accurate timestamps for received packets. What is the best way to do it? I found here: http://www.wand.net.nz/trac/libtrace/browser/Intel%20DPDK%20Patches/ some patch to enable timestamping but only on e1000 driver (and honestly I don't know if that patch wor

[dpdk-dev] [RFC PATCH] vhost: Add VHOST PMD

On Fri, 28 Aug 2015 12:21:41 +0900 Tetsuya Mukawa wrote: > The patch introduces a new PMD. This PMD is implemented as thin wrapper > of librte_vhost. It means librte_vhost is also needed to compile the PMD. > The PMD can have 'iface' parameter like below to specify a path to connect > to a virtio

[dpdk-dev] [PATCH 2/2] rte_sched: remove useless bitmap_free

Coverity reports that rte_bitmap_free() does nothing and caller does not check return value. Just remove it. Also since rte_free(NULL) is a nop, remove useless check here. Signed-off-by: Stephen Hemminger --- lib/librte_sched/rte_bitmap.h | 19 --- lib/librte_sched/rte_sched.c

[dpdk-dev] [PATCH 1/2] rte_sched: whitespace cleanup

Break long lines, remove {} on single line conditionals, fix indentation, and break needlessly long lines. Signed-off-by: Stephen Hemminger --- lib/librte_sched/rte_sched.c | 363 ++- 1 file changed, 182 insertions(+), 181 deletions(-) diff --git a/lib/li

[dpdk-dev] [PATCH 0/2] rte_sched: cleanups

Minor stuff found by Coverity and checkpatch Stephen Hemminger (2): rte_sched: whitespace cleanup rte_sched: remove useless bitmap_free lib/librte_sched/rte_bitmap.h | 19 --- lib/librte_sched/rte_sched.c | 366 +- 2 files changed, 181 insertions(+),

[dpdk-dev] [PATCH 3/3] virtio: fix possible NULL dereference

Found by Coverity. In virtio_dev_queue_release if the queue pointer is NULL, then driver is dereferencing it to get hw pointer. Also, don't do useless assignment Signed-off-by: Stephen Hemminger --- drivers/net/virtio/virtio_ethdev.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) dif

[dpdk-dev] [PATCH 2/3] virtio: fix Coverity unsigned warnings

There are some places in virtio driver where uint16_t or int are used where it would be safer to use unsigned. Signed-off-by: Stephen Hemminger --- drivers/net/virtio/virtio_ethdev.c |7 +- drivers/net/virtio/virtio_ethdev.c.orig | 1577 +++ drivers/net/virti

[dpdk-dev] [PATCH 1/3] virtio: don't report link state feature unless available

From: Stephen Hemminger If host does not support virtio link state (like current DPDK vhost) then don't set the flag. This keeps applications from incorrectly assuming that link state is available when it is not. It also avoids useless "guess what works in the config". Signed-off-by: Stephen Hem

[dpdk-dev] [PATCH 0/3] virtio: patches for 2.1+

These are some recently found bug fixes for virtio driver. Stephen Hemminger (3): virtio: don't report link state feature unless available virtio: fix Coverity unsigned warnings virtio: fix possible NULL dereference drivers/net/virtio/virtio_ethdev.c | 20 +- drivers/net/virtio/virt

[dpdk-dev] "cannot use T= with gcov target" when doing "makefile clean" with DPDK-2.1.0

Hi all, I found that after unzipping dpdk-2.1.0.tar.gz if I run: # make T=x86_64-native-linuxapp-gcc O=x86_64-native-linuxapp-gcc config And then # make V=1 T=x86_64-native-linuxapp-gcc O=x86_64-native-linuxapp-gcc clean I get: [...] == Clean app/proc_info make -f /home

[dpdk-dev] [RFC PATCH 5/6] bonding: remove pci device driver

For DPDK 2.1, there are these:

[dpdk-dev] BUG - KNI broken in 4.2 kernel

On Fri, 28 Aug 2015 13:44:28 +0100 Bruce Richardson wrote: > On Thu, Aug 27, 2015 at 10:45:43AM -0700, Stephen Hemminger wrote: > > On Thu, 27 Aug 2015 15:56:16 + > > "Zhang, Helin" wrote: > > > > > Based on my experience, only one or two users asked for ethtool support, > > > then we have

[dpdk-dev] [RFC PATCH 5/6] bonding: remove pci device driver

Hi Stephen, > -Original Message- > From: Stephen Hemminger [mailto:stephen at networkplumber.org] > Sent: Thursday, August 27, 2015 6:48 PM > To: Iremonger, Bernard > Cc: dev at dpdk.org > Subject: Re: [dpdk-dev] [RFC PATCH 5/6] bonding: remove pci device driver > > On Thu, 27 Aug 2015 16

[dpdk-dev] [RFC PATCH 0/6] remove pci driver from vdevs

Hi John, > -Original Message- > From: John W. Linville [mailto:linville at tuxdriver.com] > Sent: Thursday, August 27, 2015 6:44 PM > To: Iremonger, Bernard > Cc: dev at dpdk.org > Subject: Re: [dpdk-dev] [RFC PATCH 0/6] remove pci driver from vdevs > > On Thu, Aug 27, 2015 at 04:40:35PM

[dpdk-dev] [RFC PATCH 0/6] remove pci driver from vdevs

On Fri, Aug 28, 2015 at 08:15:47AM +, Iremonger, Bernard wrote: > Hi John, > > > -Original Message- > > From: John W. Linville [mailto:linville at tuxdriver.com] > > Sent: Thursday, August 27, 2015 6:44 PM > > To: Iremonger, Bernard > > Cc: dev at dpdk.org > > Subject: Re: [dpdk-dev] [

[dpdk-dev] [PATCH] rte_eth_af_packet: refactor error handling to avoid NULL pointer dereference

On Thu, Aug 27, 2015 at 01:17:13PM -0400, John W. Linville wrote: > Coverity CID # 13321 > > Checking *internals != NULL before accessing req is not good enough, > because **internals is a function argument and the function doesn't > really know what is passed-in. We can close our eyes and ignore