Re: [dpdk-dev] [PATCH 4/4] examples/vhost: fix uninitialized desc indexes

On 06/02/2017 01:20 PM, Jerin Jacob wrote: Fixing the below error by returning from the function early when count == 0. Issue flagged by GCC 7.1.1 examples/vhost/virtio_net.c:370:38: error: ‘desc_indexes[0]’ may be used uninitialized in this function [-Werror=maybe-uninitialized] rte_prefe

Re: [dpdk-dev] [RFC] proposal of allowing personal/project repos on DPDK.org

Dear All, Hope github is best place as many people do. On Fri, 6/2/17, Dumitrescu, Cristian wrote: Subject: Re: [dpdk-dev] [RFC] proposal of allowing personal/project repos on DPDK.org To: "Thomas Monjalon" , "Bie, Tiwei" Cc: "dev@dpdk.

Re: [dpdk-dev] [PATCH v2] ring: use aligned memzone allocation

> > The PROD/CONS_ALIGN values on x86-64 are set to 2 cache lines, so members of > struct rte_ring are 128 byte aligned, >and therefore the whole struct needs 128-byte alignment according to the ABI >so that the 128-byte alignment of the fields can be guaranteed. Ah ok, missed the fact that

Re: [dpdk-dev] [RFCv2] service core concept

Hi Harry, > > > > Thanks everybody for the input on the RFC - appreciated! From an > > application point-of-view, I > > see merit in Konstantin's suggestions for the API, over the RFC I sent > > previously. I will re- > > work the API taking inspiration from both APIs and send an RFCv2, you'll

Re: [dpdk-dev] Getting meta data from different pipelines in ip_pipeline application

Hi Nidhia, You should go ahead and test it. Did you test it and it did not work as you expect? Personally, I cannot see any reasons why it should not work, but you need to test it yourself. I am assuming that pipelines 4, 5,6 below have the passthrough type and pipeline 7 has the flow classifi

Re: [dpdk-dev] [dpdk-stable] [PATCH] vhost: fix crash on NUMA

On Fri, Jun 02, 2017 at 10:20:38AM +0200, Jens Freimann wrote: > On Fri, Jun 02, 2017 at 08:14:46AM +0800, Yuanhan Liu wrote: > > The queue allocation was changed, from allocating one queue-pair at a > > time to one queue at a time. Most of the changes have been done, but > > just with one being mi

[dpdk-dev] [PATCH v3]net/mlx5: implement drop action in hardware classifier

remove/add blank line according to convensions Shachar Beiser (1): net/mlx5: implement drop action in hardware classifier drivers/net/mlx5/Makefile| 5 + drivers/net/mlx5/mlx5_flow.c | 17 + 2 files changed, 22 insertions(+) -- 1.8.3.1

[dpdk-dev] [PATCH v3] net/mlx5: implement drop action in hardware classifier

The current drop action is implemented as a queue tail drop, requiring to instantiate multiple WQs to maintain high drop rate. This commit, implements the drop action in hardware classifier. This enables to reduce the amount of contexts needed for the drop, without affecting the drop rate. Signed-

[dpdk-dev] [PATCH v3 0/2] Various fixes for contigmem

The vm_page_replace_checked() was introduced in FreeBSD 11, so the build is broken on FreeBSD 10. The fix is to use vm_page_replace() directly and do the check in caller. --- v2: destroy the initialized mtx before failing or unloading v3: fix build on FreeBSD 10, refine comments Tiwei Bie (2):

[dpdk-dev] [PATCH v3 2/2] contigmem: don't zero the pages during each mmap

Don't zero the pages during each mmap. Instead, only zero the pages when they are not already mmapped. Otherwise, the multi-process support will be broken, as the pages will be zeroed when secondary processes map the memory. Besides, track the open and mmap operations on the cdev, and prevent the m

[dpdk-dev] [PATCH v3 1/2] contigmem: free the allocated memory when error occurs

Fixes: 764bf26873b9 ("add FreeBSD support") Cc: sta...@dpdk.org Signed-off-by: Tiwei Bie Acked-by: Bruce Richardson --- lib/librte_eal/bsdapp/contigmem/contigmem.c | 19 +++ 1 file changed, 15 insertions(+), 4 deletions(-) diff --git a/lib/librte_eal/bsdapp/contigmem/contigmem.