[dpdk-dev] [PATCH v5] vhost: support virtqueue interrupt/notification suppression

2017-12-23 Thread Junjie Chen
The driver can suppress interrupt when VIRTIO_F_EVENT_IDX feature bit is negotiated. The driver set vring flags to 0, and MAY use used_event in available ring to advise device interrupt util reach an index specified by used_event. The device ignore the lower bit of vring flags, and send an interrup

Re: [dpdk-dev] [PATCH 3/6] test: fix memory leak in ring autotest

2017-12-23 Thread Burakov, Anatoly
On 22-Dec-17 4:26 PM, Olivier MATZ wrote: Hi, On Fri, Dec 22, 2017 at 10:12:07AM +, Anatoly Burakov wrote: Fixes: af75078fece3 ("first public release") Not sure about this commit id: freeing rings is only possible since commit 4e32101f9b01 ("ring: support freeing"). You're correct, i'll

Re: [dpdk-dev] [PATCH] member: fix memory leak on error

2017-12-23 Thread Burakov, Anatoly
On 22-Dec-17 6:33 PM, Wang, Yipeng1 wrote: -Original Message- From: Burakov, Anatoly Yep, i can see that now. Didn't think to look inside rte_member_free() :/ However, you're creating a race condition there - you're unlocking a tailq, and then locking (and unlocking) it again inside rte_m

Re: [dpdk-dev] [PATCH v2 0/5] Introduce virtual PMD for Hyper-V/Azure platforms

2017-12-23 Thread Thomas Monjalon
23/12/2017 03:06, Stephen Hemminger: > Why does this need to be a PMD? It needs to be a driver on top of buses. > Maybe we need some platform infrastructure? What would be such infrastructure? A new driver type? Something like drivers/platform/? I am not sure it is required for this driver give

Re: [dpdk-dev] [PATCH v4 6/8] maintainers: add the opdl pmd maintainer information

2017-12-23 Thread Thomas Monjalon
The maintainers infos can be added while adding the new files. No need for a separate patch. Thanks

Re: [dpdk-dev] [RFC v4 PATCH 0/8] event: eventdev OPDL PMD

2017-12-23 Thread Thomas Monjalon
22/12/2017 12:23, Liang Ma: > ChangeLoga > [v4] > 1. fix 2 coding style issue > [v3] > 1. add dynamic log support. > 2. update maintainer, release notes. > 3. fix issues with review comments. > > [v2] > 1. merge the opdl eventdev unit test code into opdl pmd. > 2. propose three new cap

Re: [dpdk-dev] [PATCH] bus/dpaa/base: fix build when assert enabled

2017-12-23 Thread Thomas Monjalon
22/12/2017 08:05, Shreyansh Jain: > On Friday 22 December 2017 07:43 AM, Ferruh Yigit wrote: > > Enabling CONFIG_RTE_ENABLE_ASSERT cause build error because some fields > > protected by RTE_LIBRTE_DPAA_HWDEBUG macro accessed within DPAA_ASSERT. > > > > All DPAA_ASSERT macro usage causing the build

Re: [dpdk-dev] [PATCH 2/5] ethdev: add port ownership

2017-12-23 Thread Matan Azrad
Hi > -Original Message- > From: Neil Horman [mailto:nhor...@tuxdriver.com] > Sent: Friday, December 22, 2017 4:27 PM > To: Matan Azrad > Cc: Thomas Monjalon ; dev@dpdk.org; Bruce > Richardson ; Ananyev, Konstantin > ; Gaëtan Rivet ; > Wu, Jingjing > Subject: Re: [dpdk-dev] [PATCH 2/5] et

Re: [dpdk-dev] [PATCH 1/2] mempool: notify mempool area after mempool alloc

2017-12-23 Thread Pavan Nikhilesh
On Fri, Dec 22, 2017 at 04:11:21PM +0100, Olivier MATZ wrote: > On Fri, Dec 15, 2017 at 09:30:30PM +0530, Pavan Nikhilesh wrote: > > Mempool creation needs to be completed first before notifying mempool to > > register the mempool area. > > > > Signed-off-by: Pavan Nikhilesh > > Looks good to me.