OWC Aura Pro X SSD

OWC Aura Pro X SSD

[Winston Smith]

hi all. i recently bought an NVME SSD for my aging macbook air and could
use a little help with it.

i dual-boot arch linux and macos on my laptop, but spend 99.99% of my
time in linux. i only keep macos around for firmware updates...useless os.

anyhow, since installing the new nvme ssd i am getting some pretty bad
battery life. i run tlp configure my laptop's power savings, and use
powertop to analyze power usage.

lspci shows that the Aura SSD uses a controller by Silicon Motion, the
'Silicon Motion, Inc. Device 2260'

powertop shows my pci and usb devices usage at 100%, but using 0 mW, is
that indicative of an issue?

i decided to try and look at my ssd directly, and here i get to the meat
of my question for this mailing list. i read about the power saving
kernel patch on the arch wiki:

https://wiki.archlinux.org/index.php/Solid_state_drive/NVMe#NVME_Power_Saving_Patch

when i run "nvme get-feature -f 0x0c -H /dev/nvme0" i get an error:

"NVMe Status:INVALID_FIELD"

i ran "nvme id-ctrl /dev/nvme0", here is the output:

> NVME Identify Controller:
> vid       : 0x126f
> ssvid     : 0x126f
> sn        : OW18112010047103E   
> mn        : OWC Aura Pro X SSD                      
> fr        : C2.3.23A
> rab       : 6
> ieee      : 000000
> cmic      : 0
> mdts      : 5
> cntlid    : 1
> ver       : 10200
> rtd3r     : 186a0
> rtd3e     : 13880
> oaes      : 0
> ctratt    : 0
> rrls      : 0
> oacs      : 0x16
> acl       : 4
> aerl      : 4
> frmw      : 0x12
> lpa       : 0x3
> elpe      : 63
> npss      : 4
> avscc     : 0x1
> apsta     : 0
> wctemp    : 335
> cctemp    : 345
> mtfa      : 50
> hmpre     : 0
> hmmin     : 0
> tnvmcap   : 0
> unvmcap   : 0
> rpmbs     : 0
> edstt     : 5
> dsto      : 1
> fwug      : 0
> kas       : 0
> hctma     : 0
> mntmt     : 0
> mxtmt     : 0
> sanicap   : 0
> hmminds   : 0
> hmmaxd    : 0
> nsetidmax : 0
> sqes      : 0x66
> cqes      : 0x44
> maxcmd    : 0
> nn        : 1
> oncs      : 0x1c
> fuses     : 0
> fna       : 0x1
> vwc       : 0x1
> awun      : 4
> awupf     : 0
> nvscc     : 1
> acwu      : 0
> sgls      : 0
> subnqn    : 
> ioccsz    : 0
> iorcsz    : 0
> icdoff    : 0
> ctrattr   : 0
> msdbd     : 0
> ps    0 : mp:9.00W operational enlat:5 exlat:5 rrt:0 rrl:0
>           rwt:0 rwl:0 idle_power:- active_power:-
> ps    1 : mp:4.60W operational enlat:30 exlat:30 rrt:1 rrl:1
>           rwt:1 rwl:1 idle_power:- active_power:-
> ps    2 : mp:3.80W operational enlat:30 exlat:30 rrt:2 rrl:2
>           rwt:2 rwl:2 idle_power:- active_power:-
> ps    3 : mp:0.0700W non-operational enlat:30000 exlat:10000 rrt:3 rrl:3
>           rwt:3 rwl:3 idle_power:- active_power:-
> ps    4 : mp:0.0050W non-operational enlat:50000 exlat:10000 rrt:4 rrl:4
>           rwt:4 rwl:4 idle_power:- active_power:-

is nvme power saving available with this (admittedly exotic) ssd? i see
the 'ps' fields in the id_ctrl output, but don't know what to do about
them. how can i find the correct feature number to use to check the
power saving status of my nvme drive?

thank you all in advance for any help you can provide.

winston smith

-------------------------------------------------

ONLY AT VFEmail! - Use our Metadata Mitigator to keep your email out of the NSA's hands!
$24.95 ONETIME Lifetime accounts with Privacy Features!  
15GB disk! No bandwidth quotas!
Commercial and Bulk Mail Options!  

OWC Aura Pro X SSD

[Keith Busch]

On Wed, Jan 02, 2019@10:03:00PM +0000, Winston Smith wrote:
> i decided to try and look at my ssd directly, and here i get to the meat
> of my question for this mailing list. i read about the power saving
> kernel patch on the arch wiki:
> 
> https://wiki.archlinux.org/index.php/Solid_state_drive/NVMe#NVME_Power_Saving_Patch
> 
> when i run "nvme get-feature -f 0x0c -H /dev/nvme0" i get an error:
> 
> "NVMe Status:INVALID_FIELD"

This is saying your controller doesn't support autonomous power state
transitions. For such devices, you can reach your their low power states
only through direct host managed transitions, but the driver doesn't
support that mode.
 
> i ran "nvme id-ctrl /dev/nvme0", here is the output:
> 
> > NVME Identify Controller:
> > ps    0 : mp:9.00W operational enlat:5 exlat:5 rrt:0 rrl:0
> >           rwt:0 rwl:0 idle_power:- active_power:-
> > ps    1 : mp:4.60W operational enlat:30 exlat:30 rrt:1 rrl:1
> >           rwt:1 rwl:1 idle_power:- active_power:-
> > ps    2 : mp:3.80W operational enlat:30 exlat:30 rrt:2 rrl:2
> >           rwt:2 rwl:2 idle_power:- active_power:-
> > ps    3 : mp:0.0700W non-operational enlat:30000 exlat:10000 rrt:3 rrl:3
> >           rwt:3 rwl:3 idle_power:- active_power:-
> > ps    4 : mp:0.0050W non-operational enlat:50000 exlat:10000 rrt:4 rrl:4
> >           rwt:4 rwl:4 idle_power:- active_power:-
> 
> is nvme power saving available with this (admittedly exotic) ssd? i see
> the 'ps' fields in the id_ctrl output, but don't know what to do about
> them. how can i find the correct feature number to use to check the
> power saving status of my nvme drive?

You can mess with those with nvme-cli using 'set-feature -f 2' with a
'-v' between 0 and 3, but you'll likely just get IO timeouts if you
select a non-operational power state since this SSD won't automatically
transition to operation on a new read/write command.

OWC Aura Pro X SSD

[Winston Smith]

thanks for the response keith!

how important is it to get some sort or power savings out of an nvme
drive? my total power usage reported by powertop is around 7.5W at
typical back-light settings and application usage, with the wifi card on.

my pci bus shows only negligible power usage. the 100% usage values only
indicated the device is available 100% of the time, so that is good
actually.

am i imagining the power drain if powertop says everything is kosher?

lastly, should i try to find a better supported ssd and return this one
while i can? OWC has great support and return policy.

thanks again!

Keith Busch:
> On Wed, Jan 02, 2019@10:03:00PM +0000, Winston Smith wrote:
>> i decided to try and look at my ssd directly, and here i get to the meat
>> of my question for this mailing list. i read about the power saving
>> kernel patch on the arch wiki:
>>
>> https://wiki.archlinux.org/index.php/Solid_state_drive/NVMe#NVME_Power_Saving_Patch
>>
>> when i run "nvme get-feature -f 0x0c -H /dev/nvme0" i get an error:
>>
>> "NVMe Status:INVALID_FIELD"
> 
> This is saying your controller doesn't support autonomous power state
> transitions. For such devices, you can reach your their low power states
> only through direct host managed transitions, but the driver doesn't
> support that mode.
>  
>> i ran "nvme id-ctrl /dev/nvme0", here is the output:
>>
>>> NVME Identify Controller:
>>> ps    0 : mp:9.00W operational enlat:5 exlat:5 rrt:0 rrl:0
>>>           rwt:0 rwl:0 idle_power:- active_power:-
>>> ps    1 : mp:4.60W operational enlat:30 exlat:30 rrt:1 rrl:1
>>>           rwt:1 rwl:1 idle_power:- active_power:-
>>> ps    2 : mp:3.80W operational enlat:30 exlat:30 rrt:2 rrl:2
>>>           rwt:2 rwl:2 idle_power:- active_power:-
>>> ps    3 : mp:0.0700W non-operational enlat:30000 exlat:10000 rrt:3 rrl:3
>>>           rwt:3 rwl:3 idle_power:- active_power:-
>>> ps    4 : mp:0.0050W non-operational enlat:50000 exlat:10000 rrt:4 rrl:4
>>>           rwt:4 rwl:4 idle_power:- active_power:-
>>
>> is nvme power saving available with this (admittedly exotic) ssd? i see
>> the 'ps' fields in the id_ctrl output, but don't know what to do about
>> them. how can i find the correct feature number to use to check the
>> power saving status of my nvme drive?
> 
> You can mess with those with nvme-cli using 'set-feature -f 2' with a
> '-v' between 0 and 3, but you'll likely just get IO timeouts if you
> select a non-operational power state since this SSD won't automatically
> transition to operation on a new read/write command.
> 

-------------------------------------------------

ONLY AT VFEmail! - Use our Metadata Mitigator to keep your email out of the NSA's hands!
$24.95 ONETIME Lifetime accounts with Privacy Features!  
15GB disk! No bandwidth quotas!
Commercial and Bulk Mail Options!