Re: [SPDK] [SPDK-NVMeF]: Issues with Namespace Reservation

* Re: [SPDK] [SPDK-NVMeF]: Issues with Namespace Reservation
@ 2017-06-14  9:29 Ankur Srivastava
  0 siblings, 0 replies; 3+ messages in thread
From: Ankur Srivastava @ 2017-06-14  9:29 UTC (permalink / raw)
  To: spdk

[-- Attachment #1: Type: text/plain, Size: 3829 bytes --]

Hi Liu,

Thanks for the update.

Till when are you planning to push that fix in the SPDK ?? Since we
are stuck in the SPDK reservation scenarios.

Thanks in Advance.

Regards
Ankur

On Mon, Jun 12, 2017 at 1:01 PM, Liu, Changpeng <changpeng.liu(a)intel.com> wrote:
> Thanks Ankur,
>
> It's an issue of SPDK, because the example used a stack buffer instead of spdk_dma_malloc's contiguous physical memory.
> I'll fix it soon.
>
>
>> -----Original Message-----
>> From: SPDK [mailto:spdk-bounces(a)lists.01.org] On Behalf Of Ankur Srivastava
>> Sent: Monday, June 12, 2017 3:04 PM
>> To: Storage Performance Development Kit <spdk(a)lists.01.org>
>> Subject: [SPDK] [SPDK-NVMeF]: Issues with Namespace Reservation
>>
>> Hi All,
>>
>> I was trying some reservation related stuff through the examples given
>> in the SPDK folder,
>>
>> Through nvme-manage, I was able to create a namespace and also
>> attached it with the controller.
>>
>> While executing the reserve application, I am getting the following error:
>>
>> Setup: PCIe attached Intel NVMe SSD
>>
>> [root(a)localhost reserve]# ./reserve
>> EAL: Detected 12 lcore(s)
>> EAL: Auto-detected process type: PRIMARY
>> EAL: Probing VFIO support...
>> EAL: PCI device 0000:06:00.0 on NUMA socket 0
>> EAL:   probe driver: 8086:984 spdk_nvme
>> EAL: PCI device 0000:06:00.0 on NUMA socket 0
>> EAL:   probe driver: 8086:984 spdk_nvme
>> =====================================================
>> NVMe Controller at PCI bus 6, device 0, function 0
>> =====================================================
>> Reservations:                Supported
>> Set Feature: Host Identifier 0xababababcdcdcdcd
>> could not find 2MB vfn 0x3ffe4e9 in DPDK mem config
>> SET FEATURES (09) sqid:0 cid:55 nsid:0 cdw10:00000081 cdw11:00000000
>> INVALID FIELD (00/02) sqid:0 cid:55 cdw0:0 sqhd:0000 p:0 m:0 dnr:1
>> Set Feature: Failed
>> could not find 2MB vfn 0x3ffe4e9 in DPDK mem config
>> GET FEATURES (0a) sqid:0 cid:55 nsid:0 cdw10:00000081 cdw11:00000000
>> INVALID FIELD (00/02) sqid:0 cid:55 cdw0:0 sqhd:0000 p:0 m:0 dnr:1
>> Get Feature: Failed
>> RESERVATION REGISTER (0d) sqid:1 cid:127 nsid:1
>> INVALID FIELD (00/02) sqid:1 cid:127 cdw0:0 sqhd:0001 p:1 m:1 dnr:0
>> Reservation Register Failed
>> RESERVATION ACQUIRE (11) sqid:1 cid:127 nsid:1
>> INVALID FIELD (00/02) sqid:1 cid:127 cdw0:0 sqhd:0002 p:1 m:1 dnr:0
>> Reservation Acquire Failed
>> Reservation Generation Counter                  0
>> Reservation type                                0
>> Reservation Number of Registered Controllers    0
>> Reservation Persist Through Power Loss State    0
>> RESERVATION RELEASE (15) sqid:1 cid:127 nsid:1
>> INVALID FIELD (00/02) sqid:1 cid:127 cdw0:0 sqhd:0004 p:1 m:1 dnr:0
>> Reservation Release Failed
>> Cleaning up...
>>
>>
>> Although, I have reserved and also checked the hugepages:
>>
>> mount -t hugetlbfs nodev /mnt/huge
>> echo 2048 > /sys/devices/system/node/node0/hugepages/hugepages-
>> 1048576kB/nr_hugepages
>> echo 2048 > /sys/devices/system/node/node0/hugepages/hugepages-
>> 2048kB/nr_hugepages
>> ./setup.sh
>>
>> [root(a)localhost reserve]# cat
>> /sys/devices/system/node/node0/hugepages/hugepages-2048kB/nr_hugepages
>> 1195
>> [root(a)localhost reserve]# cat
>> /sys/devices/system/node/node0/hugepages/hugepages-
>> 1048576kB/nr_hugepages
>> 12
>>
>>
>>
>>
>> Am I doing something wrong or missing something ?? Thanks in Advance.
>>
>>
>>
>> Regards
>> Ankur
>> _______________________________________________
>> SPDK mailing list
>> SPDK(a)lists.01.org
>> https://lists.01.org/mailman/listinfo/spdk
> _______________________________________________
> SPDK mailing list
> SPDK(a)lists.01.org
> https://lists.01.org/mailman/listinfo/spdk

^ permalink raw reply	[flat|nested] 3+ messages in thread