[dpdk-dev] [PATCH 1/3] rte_ring: remove deprecated functions

2015-06-12 15:00, Bruce Richardson: > On Fri, Jun 12, 2015 at 08:46:55AM +0300, Panu Matilainen wrote: > > On 06/12/2015 08:18 AM, Stephen Hemminger wrote: > Also, patch title should be prefixed with "ring pmd" or "drivers/net/ring" > rather > than rte_ring, since this is a patch for the ring

[dpdk-dev] [PATCH] vhost: enable live migration

2015-05-27 11:01, Huawei Xie: > When we migrate VM, without this feature, qemu will report error: > "migrate: Migration disabled: vhost lacks VHOST_F_LOG_ALL feature". > > Signed-off-by: Krishna Murthy Assumed the author is Krishna Murthy (not explicit in the patch). Applied, thanks.

[dpdk-dev] [PATCH] mlx4: fix compilation warnings for 32 bit

> > Fix warning messages "cast to pointer from integer of different size" when > > compiling DPDK in 32 bit with Mellanox PMD. > > > > SGE addresses are 64 bit integers, converting them to pointers must be done > > through uintptr_t to avoid compilation warnings when those have a different > >

[dpdk-dev] [PATCH 3/6] ethdev: extend struct to support flow director in VFs

2015-05-11 11:46, Jingjing Wu: > This patch extends struct rte_eth_fdir_flow_ext to support flow > director in VFs. > > Signed-off-by: Jingjing Wu > --- a/lib/librte_ether/rte_eth_ctrl.h > +++ b/lib/librte_ether/rte_eth_ctrl.h > @@ -394,6 +394,8 @@ struct rte_eth_fdir_flow_ext { >

[dpdk-dev] [PATCH 2/3] librte_compat: fix macro deffinition

Hi, More fixes are needed in rte_compat.h 2015-05-29 17:26, Maciej Gajdzica: > --- a/lib/librte_compat/rte_compat.h > +++ b/lib/librte_compat/rte_compat.h > @@ -103,7 +103,7 @@ > #define VERSION_SYMBOL(b, e, v) Should be b,e,n > #define __vsym > #define BASE_SYMBOL(b, n) Should be b,e >

[dpdk-dev] [PATCH] qos_sched: example modification to use librte_cfgfile

> > This is a modification of qos_sched example to use > > librte_cfgfile for parsing configuration file. > > > > Signed-off-by: Michal Jastrzebski > > Acked-by: Cristian Dumitrescu Applied, thanks

[dpdk-dev] [PATCH 3/3] librte_cfgfile: rework

2015-05-29 17:26, Maciej Gajdzica: > From: Pawel Wodkowski > > This patch add following features: > 1. Make entry variable length - remove 64 byte limit. > 2. Add line continue character '\' to make multiline values possible. > > This patch also fixes following issues: > 1. BUG: section were

[dpdk-dev] [PATCH] app/test: fix default memory assignment

> Each test requires a certain minimal amount of memory. > Spreading memory on all sockets means that the test will get less memory than > what it wanted on multi sockets system. > So replace all_sockets() with per_sockets(). > > Also doubled memory on group_5 as current requirement is not

[dpdk-dev] [PATCH] app/test: return error code on failed tests

2015-06-11 15:48, John McNamara: > This change returns a system error code if tests fail when > running any of the "make test" targets. > > This allows the tests to report failures while running in > continuous integration environments. > > Previously "make test" returned $? == 0 for all

[dpdk-dev] [PATCH] mem: allow mem size to be specified when no hugetblfs

2015-05-29 15:23, Paul Atkins: > The config option to turn off huge table support does not work with > the existing -m option to specify the amount of memory to use. Add > a new option --no-huge-mem-size that takes a paramater > to use as the heap size instead of the value specified by >

[dpdk-dev] [PATCH v2] eal: Allow combining -m and --no-huge

> > Needed to run as non-root but with higher memory allocations, and > > removes a constraint on no-huge mode being limited to 64M. A usage > > example is if running with file input with the pcap PMD, which can be > > done as non-root after this patch via e.g., > > > > ./test-dpdk --no-huge

[dpdk-dev] [PATCH v3 1/6] ethdev: add an field for querying hash key size

2015-06-12 15:33, Helin Zhang: > v3 changes: > * Moved the newly added element right after 'uint16_t reta_size', where it > was a padding. So it will not break any ABI compatibility, and no need to > disable it by default. [...] > @@ -918,6 +918,7 @@ struct rte_eth_dev_info { > uint32_t

[dpdk-dev] [PATCH] e1000: enable allmulticast support for VF

We still have no maintainer for e1000. Anyone to double-check this short patch? 2015-05-28 17:46, Yury Kylulin: > Add support to enable and disable reception of all multicast packets by the > VF using standard API > rte_eth_allmulticast_enable()/rte_eth_allmulticast_disable(). > >

[dpdk-dev] [PATCH 00/26] update ixgbe base driver

> > Wenzhuo Lu (26): > > ixgbe/base: update copyright and readme > > ixgbe/base: fix code comment, double from > > ixgbe/base: fix typo error in code comment > > ixgbe/base: check return value after calling > > ixgbe/base: allow tunneled UDP and TCP frames to reach their > >

[dpdk-dev] [PATCH] ixgbe: fix x550 flow director issue

> > On x550, flow director doesn't support other IP packets directly. > > If we want to monitor IP other packets, the L4 protocol and ports must be > > masked. It means, on x550, if we want to add a flow director filter for > > other > > IP packets, a flow director mask must have been configed to

[dpdk-dev] [PATCH v3 00/10] Add a VXLAN sample

> Acked-by: Helin Zhang Sorry, there are too many checkpatch errors to peacefully apply this series.

[dpdk-dev] [PATCH v2] Add toeplitz hash algorithm used by RSS

2015-06-03 16:07, Thomas Monjalon: > 2015-05-08 10:58, Vladimir Medvedkin: > > Software implementation of the Toeplitz hash function used by RSS. > > Can be used either for packet distribution on single queue NIC > > or for simulating of RSS computation on specific NIC (for

[dpdk-dev] [PATCH v6 08/10] hash: remove duplicated code

The following patch is needed and will be squashed to fix some doxygen issues: --- a/lib/librte_hash/rte_jhash.h +++ b/lib/librte_hash/rte_jhash.h @@ -306,13 +306,13 @@ rte_jhash2(const uint32_t *k, uint32_t length, uint32_t initval) * with seeds. If you pass in (*pb)=0, the output (*pc) will

[dpdk-dev] [PATCH v6 00/10] update jhash function

2015-06-12 11:37, Bruce Richardson: > On Wed, Jun 10, 2015 at 04:25:17PM +0100, Pablo de Lara wrote: > > Jenkins hash function was developed originally in 1996, > > and was integrated in first versions of DPDK. > > The function has been improved in 2006, > > achieving up to 35% better performance,

[dpdk-dev] [PATCH v6 08/10] hash: remove duplicated code

2015-06-16 10:31, De Lara Guarch, Pablo: > From: Thomas Monjalon [mailto:thomas.monjalon at 6wind.com] > > The following patch is needed and will be squashed to fix some doxygen > > issues: |...] |> Thanks for spotting this! > Are you going to do it yourself or you want me t

[dpdk-dev] [PATCH v3 00/10] Add a VXLAN sample

2015-06-16 12:49, Liu, Jijiang: > From: Thomas Monjalon [mailto:thomas.monjalon at 6wind.com] > > Sorry, there are too many checkpatch errors to peacefully apply this series. > > Dou you mean there are too many warnings or errors when using checkpatch.pl? > > If yo

[dpdk-dev] [PATCH 0/6] Cuckoo hash

Hi Pablo, 2015-06-05 15:33, Pablo de Lara: > Things left for v2: > - Improve unit tests to show clearer performance numbers > - Documentation changes When do you plan to submit a v2?

[dpdk-dev] [PATCH 1/3] kni: minor opto

2015-06-16 01:15, Zhang, Helin: > Yes, your patches are OK for me. I have acked it. Series applied, thanks

[dpdk-dev] Debugging SR-IOV related packet flow problem with DPDK

Hi, That's third email in 8 hours for this question. Do you plan to send an email every 4 hours? It's really hard to help you without knowing which application you run? in which VM? which VLAN configuration? etc 2015-06-16 20:22, Anjali Kulkarni: > Hi, > > Can someone please help with this?

[dpdk-dev] [PATCH] kni: Use utsrelease.h to determine Ubuntu kernel version

2015-05-27 13:45, Simon Kagstrom: > /proc/version_signature is the version for the host machine, but in > e.g., chroots, this does not need to match that DPDK is built for. Use > utsrelease.h from the kernel sources instead and fake the upload > version. Sorry, I don't really understand the

[dpdk-dev] [dpdk-announce] important design choices - statistics - ABI

Hi all, Sometimes there are some important discussions about architecture or design which require opinions from several developers. Unfortunately, we cannot read every threads. Maybe that using the announce mailing list will help to bring more audience to these discussions. Please note that

[dpdk-dev] [PATCH 1/3] pmd_ring: remove deprecated functions

2015-06-16 16:05, Stephen Hemminger: > On Tue, 16 Jun 2015 14:52:16 +0100 > Bruce Richardson wrote: > > > On Mon, Jun 15, 2015 at 09:51:11AM -0700, Stephen Hemminger wrote: > > > From: Stephen Hemminger > > > > > > These were deprecated in 2.0 so remove them from 2.1 > > > > > >

[dpdk-dev] [dpdk-announce] important design choices - statistics - ABI

2015-06-16 22:28, Stephen Hemminger: > On Tue, Jun 16, 2015 at 9:36 PM, Matthew Hall > wrote: > > On Wed, Jun 17, 2015 at 01:29:47AM +0200, Thomas Monjalon wrote: > > > There were some debates about software statistics disabling. > > > Should they be always on or p

[dpdk-dev] [PATCH] vfio: Fix overflow while assigning vfio BAR region offset and size

Hi Rahul, 2015-06-16 19:46, Rahul Lakkireddy: > After the commit-id 90a1633b2 (eal/linux: allow to map BARs with MSI-X > tables), Please show this information before the Signed-off-by lines: Fixes: 90a1633b2347 ("eal/linux: allow to map BARs with MSI-X tables") generated with this git alias:

[dpdk-dev] [PATCH v2 02/11] cxgbe: add cxgbe poll mode driver.

It doesn't build as a shared library. Please add EXPORT_MAP and LIBABIVER to the Makefile. It would be easier to test if build config was merged in this patch, with the Makefile. Please merge also the maintainer infos. 2015-06-01 23:00, Rahul Lakkireddy: > +#define ARRAY_SIZE(arr) (sizeof(arr) /

[dpdk-dev] [PATCH 1/4] ixgbe: expose extended error statistics

2015-06-09 17:51, Stephen Hemminger: > On Fri, 5 Jun 2015 18:35:02 +0100 > Maryam Tahhan wrote: > > > Implement xstats_get() and xstats_reset() in dev_ops for ixgbe to expose > > detailed error statistics to DPDK applications. > > > > Signed-off-by: Maryam Tahhan > > Also, the bug where CRC

[dpdk-dev] [PATCH] kni: Passing task_pid_vnr(current) to get_net_ns_by_pid() in kni_ioctl_create()

> > In containers like docker, current->pid returns current process's global PID > > instead of its own PID under containers's PID namespace, and > > get_net_ns_by_pid() suppose to accept a virtual PID under its own namespace, > > so we should use task_pid_vnr(current) to get current process's

[dpdk-dev] [PATCH v2 0/3] add support for HTM lock elision for x86

2015-06-17 14:05, Bruce Richardson: > On Tue, Jun 16, 2015 at 10:16:43AM -0700, Roman Dementiev wrote: > > This series of patches adds methods that use hardware memory transactions > > (HTM) > > on fast-path for DPDK locks (a.k.a. lock elision). Here the methods are > > implemented for x86 using

[dpdk-dev] [PATCH 2/4] ethdev: expose extended error stats

2015-06-05 18:35, Maryam Tahhan: > Extend rte_eth_xstats_get to retrieve additional stats from the device > driver as well the top level extended stats. Add additional drop > counters to the extended stats. > > Signed-off-by: Maryam Tahhan [..] Patch 1/4 doesn't compile without patch 2/4. > ---

[dpdk-dev] [PATCH v2 1/7] ethdev: add additional error stats

2015-06-09 17:29, Stephen Hemminger: > On Tue, 9 Jun 2015 16:10:40 +0100 > Maryam Tahhan wrote: > > > Add MAC error and drop statistics to struct rte_eth_stats and the > > extended stats. > > Signed-off-by: Maryam Tahhan > > --- > > lib/librte_ether/rte_ethdev.c | 4 > >

[dpdk-dev] rte_mbuf.next in 2nd cacheline

2015-06-17 13:55, Damjan Marion: > > > On 15 Jun 2015, at 16:12, Bruce Richardson > > wrote: > > > > The next pointers always start out as NULL when the mbuf pool is created. > > The > > only time it is set to non-NULL is when we have chained mbufs. If we never > > have > > any chained

[dpdk-dev] [PATCH v7 0/4] Fix vhost enqueue/dequeue issue

2015-06-10 06:49, Xie, Huawei: > On 6/9/2015 9:03 AM, Ouyang, Changchun wrote: > > Fix enqueue/dequeue can't handle chained vring descriptors; > > Remove unnecessary vring descriptor length updating; > > Add support copying scattered mbuf to vring; > > > > Changchun Ouyang (4): > > lib_vhost:

[dpdk-dev] rte_mbuf.next in 2nd cacheline

2015-06-17 14:23, Damjan Marion: > > > On 17 Jun 2015, at 16:06, Bruce Richardson > > wrote: > > > > On Wed, Jun 17, 2015 at 01:55:57PM +, Damjan Marion (damarion) wrote: > >> > >>> On 15 Jun 2015, at 16:12, Bruce Richardson >>> intel.com> wrote: > >>> > >>> The next pointers always

[dpdk-dev] [PATCH 2/2] vhost: realloc virtio_net and virtqueue to the same node of vring desc table

2015-06-05 11:13, Huawei Xie: > --- a/mk/rte.app.mk > +++ b/mk/rte.app.mk > @@ -92,6 +92,9 @@ endif # ! CONFIG_RTE_BUILD_COMBINE_LIBS > > _LDLIBS-$(CONFIG_RTE_LIBRTE_PMD_PCAP) += -lpcap > > +ifeq ($(CONFIG_RTE_LIBRTE_VHOST_NUMA),y) > +_LDLIBS-$(CONFIG_RTE_LIBRTE_VHOST) +=

[dpdk-dev] [PATCH 2/2] vhost: realloc virtio_net and virtqueue to the same node of vring desc table

2015-06-17 18:47, Thomas Monjalon: > 2015-06-05 11:13, Huawei Xie: > > --- a/mk/rte.app.mk > > +++ b/mk/rte.app.mk > > @@ -92,6 +92,9 @@ endif # ! CONFIG_RTE_BUILD_COMBINE_LIBS > > > > _LDLIBS-$(CONFIG_RTE_LIBRTE_PMD_PCAP) += -lpcap > > >

[dpdk-dev] [PATCH 1/2] vhost: malloc -> rte_malloc for virtio_net and virt queue allocation

2015-06-05 11:13, Huawei Xie: > use rte_malloc/free for virtio_net and virt queue allocation/free > > Signed-off-by: Huawei Xie I suggest this title: vhost: use rte_malloc to allocate device and queues

[dpdk-dev] [PATCH v2] vhost: provide vhost API to unregister vhost unix domain socket

2015-06-05 11:26, Huawei Xie: > rte_vhost_driver_unregister will remove the listenfd from event list, and > then close it. > > Signed-off-by: Huawei Xie > Signed-off-by: Peng Sun > --- > lib/librte_vhost/rte_virtio_net.h| 3 ++ > lib/librte_vhost/vhost_cuse/vhost-net-cdev.c | 9

[dpdk-dev] [PATCH v2 1/3] spinlock: add support for HTM lock elision for x86

2015-06-16 10:16, Roman Dementiev: > --- a/lib/librte_eal/common/include/arch/x86/rte_spinlock.h > +++ b/lib/librte_eal/common/include/arch/x86/rte_spinlock.h > @@ -39,6 +39,13 @@ extern "C" { > #endif > > #include "generic/rte_spinlock.h" > +#include "rte_rtm.h" > +#include "rte_cpuflags.h" >

[dpdk-dev] [PATCH v2 5/6] cfgfile: fixed calling free for each section in rte_cfgfile_close

2015-06-17 16:48, Maciej Gajdzica: > From: Pawel Wodkowski > > Signed-off-by: Pawel Wodkowski What is fixed exactly? What was the problem? > @@ -60,6 +60,15 @@ struct rte_cfgfile { > * for new entries do we add in */ > #define CFG_ALLOC_ENTRY_BATCH 16 > > +/* Helpers */ > + > +#define

[dpdk-dev] [PATCH v2 6/6] cfgfile: added new implementation of section parsing

2015-06-17 16:48, Maciej Gajdzica: > + char *start = _skip_spaceses(str); Why not use a function with "char *" parameter instead of a macro?

[dpdk-dev] [PATCH v2 0/6] cfgfile: config file parsing extension

2015-06-17 16:48, Maciej Gajdzica: > Added new implementation of section parsing in config file. Refactored > existing code by spliting it to smaller functions. Changed section > allocation scheme and added new features - variable length entry value > and line continue character '\'. > > Pawel

[dpdk-dev] [PATCH] doc: fix doxygen warnings for QoS API

> > This patch fix doxygen warnings when generating documentation > > for qos_meter and qos_sched > > > > Signed-off-by: Michal Jastrzebski > > Acked-by: Cristian Dumitrescu Applied, thanks

[dpdk-dev] [PATCH v3 0/4] port: add ethdev_writer and ring_writer nodrop ports

> > When nodrop writer port fails to send data, it retries until reach maximum > > number of retries. Also added new tx_bulk implementation for ring writer > > port. > > > > Changes in v2: > > - added missing signoffs > > > > Changes in v3: > > - removed second implementation of tx_bulk

[dpdk-dev] [PATCH v4] doc: guidelines for library statistics

We need a conclusion here. It seems the policy proposed by Cristian is the best consensus. If nobody nack in this thread with a counter-proposal, it will be adopted on Monday. 2015-06-18 16:38, Morten Br?rup: > The suggested solution with only one single flag per library prevents > implementing

[dpdk-dev] [PATCH v2 1/2] Added ETH_SPEED_CAP bitmap in rte_eth_dev_info

2015-06-18 17:06, Marc Sune: > On 18/06/15 16:43, Morten Br?rup wrote: > > Regarding the PHY speed ABI: > > > > 1. The Ethernet PHY ABI for speed, duplex, etc. should be common throughout > > the entire DPDK. It might be confusing if some structures/functions use a > > bitmask to indicate PHY

[dpdk-dev] [PATCHv2 0/8] acl: various fixes and cleanups

2015-06-08 11:41, Konstantin Ananyev: > Several fixes and code cleanups for the librte_acl. > New test-cases for acl UT. > > Konstantin Ananyev (8): > acl: fix invalid rule wildness calculation for bitmask field type > acl: code cleanup - use global EAL macro, instead of creating a local >

[dpdk-dev] [PATCH] doc: fix doxygen warnings

Signed-off-by: Thomas Monjalon --- .../common/include/generic/rte_cpuflags.h | 2 +- lib/librte_eal/common/include/rte_common.h | 6 ++--- lib/librte_eal/common/include/rte_debug.h | 6 ++--- lib/librte_eal/common/include/rte_devargs.h| 2 +- lib/librte_eal

[dpdk-dev] vhost doxygen

Please Changchun, Could you check and complete the vhost API comments? It can be checked with this command: make doc-api-html 2>&1 | grep '/librte_vhost/.*warning: Thanks

[dpdk-dev] [PATCH] doc: fix doxygen warnings

2015-06-19 10:20, Ouyang Changchun: > Fix doxygen warnings in vhost > > Signed-off-by: Changchun Ouyang Applied, thanks

[dpdk-dev] [PATCH] doc: fix doxygen warnings

> > Signed-off-by: Thomas Monjalon > > Acked-by: Bruce Richardson Applied

[dpdk-dev] [dpdk-announce] important design choices - statistics - ABI

2015-06-19 06:26, Neil Horman: > On Thu, Jun 18, 2015 at 04:55:45PM +, O'Driscoll, Tim wrote: > > For the 2.1 release, I think we should agree to make patches that change > > the ABI controllable via a compile-time option. I like Olivier's proposal > > on using a single option

[dpdk-dev] [dpdk-announce] important design choices - statistics - ABI

2015-06-19 09:02, Neil Horman: > On Fri, Jun 19, 2015 at 02:32:33PM +0200, Thomas Monjalon wrote: > > 2015-06-19 06:26, Neil Horman: > > > On Thu, Jun 18, 2015 at 04:55:45PM +, O'Driscoll, Tim wrote: > > > > For the 2.1 release, I think we should agr

[dpdk-dev] [PATCH v2 1/3] spinlock: add support for HTM lock elision for x86

2015-06-18 11:00, Bruce Richardson: > On Wed, Jun 17, 2015 at 11:29:49PM +0200, Thomas Monjalon wrote: > > Introducing rte_cpuflags.h in this header breaks the compilation of > > the mlx4 pmd with CONFIG_RTE_LIBRTE_MLX4_DEBUG=y. > > Indeed, it triggers the -pedantic flag w

[dpdk-dev] [PATCH v3 0/3] add support for HTM lock elision for x86

2015-06-19 13:08, Roman Dementiev: > This series of patches adds methods that use hardware memory transactions > (HTM) > on fast-path for DPDK locks (a.k.a. lock elision). Here the methods are > implemented for x86 using Restricted Transactional Memory instructions > (Intel(r) > Transactional

[dpdk-dev] Issue with rte_compat versioning macros

2015-06-19 14:38, Gajdzica, MaciejX T: > There is an issue with macros in rte_compat.h. For shared library case, macro > BIND_DEFAULT_SYMBOL > takes three arguments and for other case it takes only two arguments. Also > letters for macro variable names > are not consistent in these two cases.

[dpdk-dev] [dpdk-announce] important design choices - statistics - ABI

2015-06-19 11:27, Neil Horman: > On Fri, Jun 19, 2015 at 03:16:53PM +0200, Thomas Monjalon wrote: > > 2015-06-19 09:02, Neil Horman: > > > On Fri, Jun 19, 2015 at 02:32:33PM +0200, Thomas Monjalon wrote: > > > > 2015-06-19 06:26, Neil Horman: > > > >

[dpdk-dev] [dpdk-announce] important design choices - statistics - ABI

2015-06-19 12:13, Thomas F Herbert: > > On 6/19/15 9:16 AM, Thomas Monjalon wrote: > > 2015-06-19 09:02, Neil Horman: > >> On Fri, Jun 19, 2015 at 02:32:33PM +0200, Thomas Monjalon wrote: > >>> 2015-06-19 06:26, Neil Horman: > >>>> On Thu, Jun 18,

[dpdk-dev] [PATCH v2 0/3] mempool: clean-up

2015-06-19 18:16, Olivier Matz: > These 3 patches try to make mempool code easier to read, they > have no functional impact. > > Changes in v2: > - fix compilation between patch 1 and 2 > > Olivier Matz (3): > mempool: introduce objhdr structure for object headers > mempool: introduce objtlr

[dpdk-dev] [PATCH v4 01/10] examples/tep_termination:initialize the VXLAN sample

2015-06-18 09:03, Jijiang Liu: > This sample uses the basic virtio devices management function from the vHost > example, which includes virtio device creation, destroying and maintenance. Please wrap commit message to about 72 characters.

[dpdk-dev] [PATCH v2 05/10] eal: add and use unaligned integer types

2015-06-19 10:34, Cyril Chemparathy: > On machines that are strict on pointer alignment, current code breaks > on GCC's -Wcast-align checks on casts from narrower to wider types. > This patch introduces new unaligned_uint(16|32|64)_t types, which > correctly retain alignment in such cases. [...] >

[dpdk-dev] [PATCH 09/11] tile: initial TILE-Gx support.

2015-06-19 12:15, Cyril Chemparathy: > +CONFIG_RTE_LIBNAME="tile_dpdk" Not sure it makes sense to have different library names. Why not renaming all to "dpdk"?

[dpdk-dev] [PATCH] ixgbe: fix the flow director flexbytes offset issues

> > > The flexbytes offset can not be set, because the value is over written > > > when fdir is enabled. > > > This patch fixes this issue, and also removes some reduplicate lines. > > > > > > Reported-by: David Marchand > > > Signed-off-by: Jingjing Wu > > > > Tested-by: Gaetan Rivet > >

[dpdk-dev] [PATCH] ixgbevf: fix incorrect RX function selection

> > The logic to select ixgbe VF RX function is different than PF side. > > > > There are a few issues with its current state: > > - it does not allow to select ixgbe_recv_pkts_vec among other options. > > - it can cause memory corruption for scatter mode as it does not allocate > >enough

[dpdk-dev] [PATCH] ixgbevf: fix link status for PF up/down events

2015-06-16 15:38, Sergio Gonzalez Monroy: > Current ixgbe VF base driver only really read the status register when: > - get_link_status is true > - link reset > - mailbox timeout. > > We only set get_link_status to true when we start the PF/VF, so > following calls to ixgbe_dev_link_update

[dpdk-dev] [PATCH v3 00/10] vmxnet3: multisegment and bugfixes

2015-03-10 14:42, Thomas Monjalon: > 2015-03-05 16:10, Stephen Hemminger: > > From: Stephen Hemminger > > > > Revised version of earlier patches. > > Incorporate the small packet optimization > > Add more cleanups > > > > Stephen Hemminger

[dpdk-dev] [PATCH v2 0/3] fm10k: Add promiscuous mode support

2015-06-12 15:06, Chen Jing D: > From: "Chen Jing D(Mark)" > > v2: > Add debug messages in allmulticast enable/disable functions. > > The patch set add promiscuous mode configuration and 2 bug fixes. > > Chen Jing D(Mark) (3): > fm10k: Add promiscuous mode support > fm10k: remove mbuf

[dpdk-dev] [PATCH 0/6] fm10k: A series of bug fixes

> > This patch set include a few bug fixes and enhancements on fm10k driver. > > > > Chen Jing D(Mark) (6): > > fm10k: Fix improper RX buffer size assignment > > fm10k: Fix jumbo frame issue > > fm10k: Fix data integrity issue with multi-segment frame > > fm10k: Fix issue that MAC addr

[dpdk-dev] [PATCH v3 0/3] fm10k: update MAC/VLAN filter and VLAN offload features

> > v3 changes: > > - add more result check for MAC table mailbox operation for VLAN filter > > - remove two extra indentation tabs > > > > v2 changes: > > - add result check when updating MAC address table for VLAN filter > > - move VLAN Table updating to the end of VLAN filter process > >

[dpdk-dev] [PATCH v2] fm10k: fix PF/VF MAC address register and clean up bug

> > v2 changes: > > - rebase this patch onto the HEAD > > - add more description for this patch > > > > This patch includes 3 changes related to MAC/VLAN address table > > when the system(e.g. testpmd) is started and closed: > > - remove default MAC address with fixed VLAN 0 which was for the

[dpdk-dev] [PATCH] ixgbe: fix x550 shutdown API issue

> > For there're only laser ports on x550 before, we only considered laser > > ports for the testpmd CLIs "port start/stop ...". Now we have new x550 > > devices which have copper ports. Use the API for copper to enable/disable > > these ports. > > > > And also let the testpmd CLI "set

[dpdk-dev] [PATCH] pmd_ring: return new port id on ethdev creation

> The rte_eth_from_rings API allowed the creation of an ethdev port at > runtime using rte_rings as the underlying storage. However, the return > value from this function was either 0 or -1, and these values were never > actually documented in the API documentation. Unfortunately, the programmers

[dpdk-dev] [PATCH v2] lib_vhost:reset secure_len when rte_atomic16_cmpset failed

2015-06-01 06:14, Ouyang, Changchun: > From: Wei li, June 1, 2015 2:12 PM: > > when rte_atomic16_cmpset return 0 in first loop, secure_len should be > > reset to 0 in second loop, otherwise (pkt_len > secure_len) always be > > false, > > the num of desc maybe not enough > > > > Signed-off-by:

[dpdk-dev] [PATCH v2] kni: ignore double calls to rte_kni_init()

> > Prevent double initialization of the KNI subsytem. > > > > v2: added warning trace > > > > Signed-off-by: Marc Sune > Acked-by: Helin Zhang Applied, thanks

[dpdk-dev] [PATCH] ivshmem: fix segmentation fault in corner case

2015-06-18 13:27, Sergio Gonzalez Monroy: > Depending on the configured segments it is possible to hit a > segmentation fault as a result of decrementing an unsigned index with > value 0. > > To avoid it, exit the loop if the index has value 0. > > Signed-off-by: Sergio Gonzalez Monroy

[dpdk-dev] [PATCH v2] log: Properly reset log_history_size in rte_log_dump_history()

2015-06-01 11:46, Olivier MATZ: > On 06/01/2015 11:30 AM, Jan Blunck wrote: > > In rte_log_dump_history() the log_history list is reinitialized without > > resetting the log_history_size. In the next call to rte_log_add_in_history() > > the log_history_size > RTE_LOG_HISTORY and the code

[dpdk-dev] [PATCH v2 05/10] eal: add and use unaligned integer types

2015-06-22 09:47, Cyril Chemparathy: > On Mon, 22 Jun 2015 11:36:20 +0200 > Thomas Monjalon wrote: > > > 2015-06-19 10:34, Cyril Chemparathy: > > > On machines that are strict on pointer alignment, current code > > > breaks on GCC's -Wcast-align checks on casts f

[dpdk-dev] [PATCH 09/11] tile: initial TILE-Gx support.

2015-06-22 09:54, Cyril Chemparathy: > On Mon, 22 Jun 2015 11:39:07 +0200 > Thomas Monjalon wrote: > > > > +CONFIG_RTE_LIBNAME="tile_dpdk" > > > > Not sure it makes sense to have different library names. > > Why not renaming all to "dpdk

[dpdk-dev] [PATCH 09/11] tile: initial TILE-Gx support.

2015-06-22 10:03, Cyril Chemparathy: > On Mon, 22 Jun 2015 19:00:44 +0200 > Thomas Monjalon wrote: > > > > > Not sure it makes sense to have different library names. > > > > Why not renaming all to "dpdk"? > > > > > > This config do

[dpdk-dev] [PATCH v2] eal:Fix log messages always being printed from rte_eal_cpu_init

2015-06-19 11:54, David Marchand: > On Mon, Jun 8, 2015 at 11:55 PM, Keith Wiles wrote: > > The RTE_LOG(DEBUG, ...) messages in rte_eal_cpu_init() are printed > > even when the log level on the command line was set to INFO or lower. > > > > The problem is the rte_eal_cpu_init() routine was called

[dpdk-dev] [PATCH] lib: fix RTE_MBUF_METADATA macros

2015-06-05 15:31, Dumitrescu, Cristian: > > Fix RTE_MBUF_METADATA macros to allow for unaligned accesses to > > meta-data fields. > > Forcing aligned accesses is not really required, so this is removing an > > unneeded constraint. > > This issue was met during testing of the new version of the

[dpdk-dev] [PATCH] rte_reorder: Allow sequence numbers > 0 as starting point

2015-05-29 17:37, Gonzalez Monroy, Sergio: > On 20/05/2015 12:02, Simon Kagstrom wrote: > > We use sequence numbers from a generator which has potentially started > > long before the receiver. Therefore, the first number will typically > > be > 0. The rte_reorder code will not work in this case,

[dpdk-dev] [PATCH] examples/distributor: fix missing "; " in debug macro

2015-06-05 17:01, Bruce Richardson: > The macro to turn on additional debug output when the app was compiled > with "-DDEBUG" was missing a ";". > > Fixes: 07db4a975094 ("examples/distributor: new sample app") > > Signed-off-by: Anbarasan Murugesan > Signed-off-by: Bruce Richardson Applied,

[dpdk-dev] [PATCH] lib: fix RTE_MBUF_METADATA macros

2015-06-22 13:23, Cyril Chemparathy: > On Mon, 22 Jun 2015 22:16:59 +0200 > Thomas Monjalon wrote: > > Cyril, feel free to fix it if it breaks with Tile arch. > > Why define these locally within rte_port.h? Shouldn't these macros > really be in rte_mbuf.h? No, an applicati

[dpdk-dev] [PATCH] examples/distributor: fix missing "; " in debug macro

2015-06-08 11:58, Bruce Richardson: > On Fri, Jun 05, 2015 at 10:45:04PM +0200, Thomas Monjalon wrote: > > It shows that such dead code is almost never tested. > > It would be saner if this command would return no result: > > git grep 'ifdef.*DEBUG' examples >

[dpdk-dev] [PATCH v1] app/test: fix pmd_perf issue in no NUMA case

2015-06-08 14:33, Cunming Liang: > Reported-by: Jayakumar, Muthurajan > Signed-off-by: Cunming Liang Please explain exactly what you try to fix. Is it still needed since this patch? http://dpdk.org/browse/dpdk/commit/?id=94ef2964148a4540 > + socket_id =

[dpdk-dev] [PATCH] app/test: Fix potential null pointer accesses in test_mbuf

2015-06-12 11:54, Tetsuya Mukawa: > The patch fixes potential null pointer accesses in test_mbuf. > if 'm[i]' is null, stop accessing it. > > Signed-off-by: Tetsuya Mukawa Applied, thanks

[dpdk-dev] [PATCH v5 01/10] examples/tep_termination:initialize the VXLAN sample

2015-06-23 00:40, Jijiang Liu: > examples/Makefile |1 + > examples/tep_termination/Makefile | 55 ++ > examples/tep_termination/main.c | 1130 > + > examples/tep_termination/main.h | 123 > 4 files changed, 1309 insertions(+),

[dpdk-dev] [PATCH v5 00/10] Add a VXLAN sample

2015-06-23 00:40, Jijiang Liu: > v2 changes: > Fix an issue about the 'nb_ports' duplication in check_ports_num(). > Remove the inaccurate comment in main.c > Fix an issue about TSO offload. > > v3 changes: > Change some viriable name that don't follow coding rules. > Remove

[dpdk-dev] [PATCH 0/4] rte_ethdev: cleanups

2015-04-09 14:29, Stephen Hemminger: > A bunch of small (almost trivial) patches to fix style and other > issues in the base Ethernet driver interface code. > > Stephen Hemminger (4): > rte_ethdev: remove extra inline > rte_ethdev: whitespace cleanup > rte_ethdev: make tables const >

[dpdk-dev] [PATCH] net: remove unneeded include in rte_ip.h

2015-04-23 13:49, Bruce Richardson: > There are no memcpy functions in rte_ip.h so there is no need to include > rte_memcpy.h in that file. > > Signed-off-by: Bruce Richardson Applied, thanks

[dpdk-dev] DPDK Community Call - Software QoS

2015-04-24 09:14, Dumitrescu, Cristian: > Hi Tim, > > Thank you for your help in recording this readout! > The slides that you attached to the email are dropped by the dpdk.org email > server. No, they are not dropped: http://dpdk.org/ml/archives/dev/2015-April/016842.html

[dpdk-dev] [PATCH 00/36] Update IXGBE base codes

> > This patch set update IXGBE base codes to the version of cid-10g-shared- > > code.2015.02.03. > > and this patch set add 2 new device id as following: > > #define IXGBE_DEV_ID_X550EM_X_10G_T0x15AD > > #define IXGBE_DEV_ID_X550EM_X_1G_T 0x15AE > > > > Changchun Ouyang

[dpdk-dev] [RFC PATCH] ethdev: remove old flow director API

2015-04-20 09:45, Venky Venkatesan: > On 04/20/2015 09:33 AM, Neil Horman wrote: > > On Mon, Apr 20, 2015 at 04:11:43PM +0200, Thomas Monjalon wrote: > >> It's time to remove this old API. > >> It seems some work is still needed to rely only on eth_ctrl API. > >&

[dpdk-dev] [PATCH 03/18] i40e: adjustment of register definitions and relevant

2015-04-20 16:22, Helin Zhang: > More register definitions and their relevant masks are added > publically. Also few useless macros are deleted. > > Signed-off-by: Helin Zhang > --- > lib/librte_pmd_i40e/i40e/i40e_register.h | 1981 > +- All these values are not

[dpdk-dev] [PATCH 11/18] i40e: support of building both PF and VF driver together

2015-04-20 16:22, Helin Zhang: > Macros of PF_DRIVER, VF_DRIVER and INTEGRATED_VF were defined to > support building both PF and VF driver together. PF_DRIVER needs > to be defined if a build is for PF only, while VF_DRIVER for VF > only. PF_DRIVER, VF_DRIVER and INTEGRATED_VF are all needed for >

<    2   3   4   5   6   7   8   9   10   11   >