Re: [dpdk-dev] CONFIG_RTE_EAL_NUMA_AWARE_HUGEPAGES: no difference in memory pool allocations, when enabling/disabling this configuration

2018-12-09 Thread Asaf Sinai
Hi all, Thanks for the detailed explanations! So, what we understood from that, is the following (please correct, if it is wrong): Before 18.05 version: - Dividing huge pages between NUMAs was based, by default, on Linux good will. - Enforcing Linux to divide huge pages between NUMAs, required e

Re: [dpdk-dev] [PATCH v2 1/1] eventdev: add new software event timer adapter

2018-12-09 Thread Mattias Rönnblom
On 2018-12-07 21:34, Erik Gabriel Carrillo wrote: This patch introduces a new version of the event timer adapter software PMD. In the original design, timer event producer lcores in the primary and secondary processes enqueued event timers into a ring, and a service core in the primary process de

Re: [dpdk-dev] [PATCH] eal: fix rte_zalloc_socket to zero memory

2018-12-09 Thread Mattias Rönnblom
On 2018-12-07 23:24, David Harton wrote: The zalloc and calloc functions do not actually zero the memory. Added memset to rte_zmalloc_socket() so allocated memory is cleared. Signed-off-by: David Harton --- lib/librte_eal/common/rte_malloc.c | 4 +++- 1 file changed, 3 insertions(+), 1 delet

[dpdk-dev] [PATCH v2] eal: fix rte_zalloc_socket to zero memory

2018-12-09 Thread David Harton
The zalloc and calloc functions do not actually zero the memory. Added memset to rte_zmalloc_socket() so allocated memory is cleared. Signed-off-by: David Harton --- v2 Indented if clause lib/librte_eal/common/rte_malloc.c | 5 - 1 file changed, 4 insertions(+), 1 deletion(-) diff --git a

[dpdk-dev] [PATCH v1] net/i40e: fix memory leak for dev reset

2018-12-09 Thread Zhirun Yan
Before this patch port reset will call i40evf_dev_uninit() and i40evf_dev_init(), i40evf_dev_uninit() will call i40e_dev_free_queues to free queues. If set dev->data->nb_tx_queues = 0 in i40evf_dev_close, the next time config queue will never free these queues. Fixex: 6b4537128394 ("i40e: free que

Re: [dpdk-dev] [PATCH] net/virtio: move bytes accounting to common function

2018-12-09 Thread Tiwei Bie
On Mon, Dec 03, 2018 at 07:20:13PM +0300, Ilya Maximets wrote: > There is no need to count 'bytes' separately. > > Signed-off-by: Ilya Maximets > --- > drivers/net/virtio/virtio_rxtx.c | 7 ++- > 1 file changed, 2 insertions(+), 5 deletions(-) Reviewed-by: Tiwei Bie

Re: [dpdk-dev] [PATCH] ethdev: support double precision RED queue weight

2018-12-09 Thread Rao, Nikhil
> -Original Message- > From: Stephen Hemminger [mailto:step...@networkplumber.org] > Sent: Thursday, November 29, 2018 11:43 AM > To: Rao, Nikhil > Cc: Dumitrescu, Cristian ; Singh, Jasvinder > ; dev@dpdk.org > Subject: Re: [dpdk-dev] [PATCH] ethdev: support double precision RED queue > we

[dpdk-dev] [PATCH v1] net/i40e: remove redundant code in free queues.

2018-12-09 Thread Zhirun Yan
Before this patch, there are two functions that will clear RX/TX queues number: rte_eth_dev_close() and i40e_dev_free_queues(). This patch remove redundant code in i40e_dev_free_queues(). Signed-off-by: Zhirun Yan Signed-off-by: Haiyue Wang --- drivers/net/i40e/i40e_rxtx.c | 2 -- 1 file chang

[dpdk-dev] [PATCH v3] net/i40e: fix VF/PF port close

2018-12-09 Thread Zhirun Yan
Port reset will call i40evf_uninit_vf() to release resource. It wants to call i40evf_dev_close() to release resources. Befort this patch, hw->adapter_stopped was used to mark the status about start/stop and close. So it will never call i40evf_dev_close() after stoped the port. This patch added hw-

[dpdk-dev] [PATCH v3] net/i40e: fix VF/PF port close

2018-12-09 Thread Zhirun Yan
Port reset will call i40evf_uninit_vf() to release resource. It wants to call i40evf_dev_close() to release resources. Befort this patch, hw->adapter_stopped was used to mark the status about start/stop and close. So it will never call i40evf_dev_close() after stoped the port. This patch added hw-

[dpdk-dev] [PATCH v4] net/i40e: fix VF/PF port close

2018-12-09 Thread Zhirun Yan
Port reset will call i40evf_uninit_vf() to release resource. It wants to call i40evf_dev_close() to release resources. Befort this patch, hw->adapter_stopped was used to mark the status about start/stop and close. So it will never call i40evf_dev_close() after stopped the port. This patch added hw

[dpdk-dev] Incorrectly applied ethdev: fix queue start to 17.11

2018-12-09 Thread Andrew Rybchenko
Hi, Do I understand correctly that the following patch was incorrectly applied on 17.11 since changes are done in queue stop functions instead of queue start functions? If so, could you fix it, please. Also I think it makes sense to check other stable branches as well. We observe regressions b

Re: [dpdk-dev] Incorrectly applied ethdev: fix queue start to 17.11

2018-12-09 Thread Zhang, Qi Z
Your understand is correct, I saw the issue happens on both 17.11 and 16.11 both From: Andrew Rybchenko [mailto:arybche...@solarflare.com] Sent: Monday, December 10, 2018 2:57 PM To: Zhang, Qi Z ; Yongseok Koh Cc: dev@dpdk.org; Ivan Malov Subject: Incorrectly applied ethdev: fix queue start to

[dpdk-dev] rte_eal_hotplug_remove() generates error message

2018-12-09 Thread Hideyuki Yamashita
Subject: Dear Thomas and all, I hope you all get safely back home after DPDK summit. (When I get back Japan, it is chilling. (start of winter)) On DPDK 18.11.0, we tried to remove vhost device by using rte_eal_hotplug_remove(). However, following syslog message is printed. “EAL: Error: Invalid