Skip site navigation (1)Skip section navigation (2)
Date:      Mon, 23 Sep 2024 10:47:16 +0100
From:      Warner Losh <imp@bsdimp.com>
To:        Alexey Sukhoguzov <mail@eseipi.net>
Cc:        FreeBSD Hackers <freebsd-hackers@freebsd.org>
Subject:   Re: nvme(4): some non-operational power states are broken
Message-ID:  <CANCZdfpv6EbB_HSnZG4Sy4crOJBbX16PGo3iwjCdkBzvTE7SVA@mail.gmail.com>
In-Reply-To: <ZvE3f_gTC4D6TqBe@eseipi.net>
References:  <ZvE3f_gTC4D6TqBe@eseipi.net>

next in thread | previous in thread | raw e-mail | index | archive | help
--000000000000dce7450622c6464f
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable

On Mon, Sep 23, 2024, 10:41=E2=80=AFAM Alexey Sukhoguzov <mail@eseipi.net> =
wrote:

> Hi,
>
> My NVMe controller is Toshiba XG5, and it has 6 power states: the
> first three (0-2) are normal and the last three (3-5) are NOPS.
> Here is 'nvmecontrol power -l nvme0' output:
>
>  #   Max pwr  Enter Lat  Exit Lat RT RL WT WL Idle Pwr  Act Pwr Workloadd
> --  --------  --------- --------- -- -- -- -- -------- -------- --
>  0:  8.0000W    0.000ms   0.000ms  0  0  0  0  0.0000W  0.0000W 0
>  1:  3.9000W    0.000ms   0.000ms  1  1  1  1  0.0000W  0.0000W 0
>  2:  2.0000W    0.000ms   0.000ms  2  2  2  2  0.0000W  0.0000W 0
>  3:  0.0500W*   1.500ms   1.500ms  3  3  3  3  0.0000W  0.0000W 0
>  4:  0.0050W*   6.000ms  14.000ms  4  4  4  4  0.0000W  0.0000W 0
>  5:  0.0030W*  50.000ms  80.000ms  5  5  5  5  0.0000W  0.0000W 0
>
> The problem is that only one of the NOPS is working as expected
> (state 3). Another two (states 4-5) skyrocket the controller's power
> consumption far beyond normal (0-2) power states do, and far beyond
> reasonable. For example, when the controller is in state 3, my
> system consumes about 3-3.5 W at idle (according to acpiconf with
> laptop power cable unplugged), in states 0-2 - about 4 W, and in
> states 4-5 consumption is approaching 6 W. Thus, the NVMe becomes
> the hottest part of the system (>50C, still idle), and it eats up
> almost half of the battery alone.
>
> Linux doesn't have this issue, so it seems to be nvme(4) related.
> All the above data is collected on 14.1-RELEASE Live USB with no
> filesystem mounted. 15-CURRENT has the same problem.
>
> Any ideas what it might be?
>

Does Linux have active power state management?

Warner


> Regards,
> Alexey
>
>

--000000000000dce7450622c6464f
Content-Type: text/html; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable

<div dir=3D"auto"><div><br><br><div class=3D"gmail_quote"><div dir=3D"ltr" =
class=3D"gmail_attr">On Mon, Sep 23, 2024, 10:41=E2=80=AFAM Alexey Sukhoguz=
ov &lt;<a href=3D"mailto:mail@eseipi.net">mail@eseipi.net</a>&gt; wrote:<br=
></div><blockquote class=3D"gmail_quote" style=3D"margin:0 0 0 .8ex;border-=
left:1px #ccc solid;padding-left:1ex">Hi,<br>
<br>
My NVMe controller is Toshiba XG5, and it has 6 power states: the<br>
first three (0-2) are normal and the last three (3-5) are NOPS.<br>
Here is &#39;nvmecontrol power -l nvme0&#39; output:<br>
<br>
=C2=A0#=C2=A0 =C2=A0Max pwr=C2=A0 Enter Lat=C2=A0 Exit Lat RT RL WT WL Idle=
 Pwr=C2=A0 Act Pwr Workloadd<br>
--=C2=A0 --------=C2=A0 --------- --------- -- -- -- -- -------- -------- -=
-<br>
=C2=A00:=C2=A0 8.0000W=C2=A0 =C2=A0 0.000ms=C2=A0 =C2=A00.000ms=C2=A0 0=C2=
=A0 0=C2=A0 0=C2=A0 0=C2=A0 0.0000W=C2=A0 0.0000W 0<br>
=C2=A01:=C2=A0 3.9000W=C2=A0 =C2=A0 0.000ms=C2=A0 =C2=A00.000ms=C2=A0 1=C2=
=A0 1=C2=A0 1=C2=A0 1=C2=A0 0.0000W=C2=A0 0.0000W 0<br>
=C2=A02:=C2=A0 2.0000W=C2=A0 =C2=A0 0.000ms=C2=A0 =C2=A00.000ms=C2=A0 2=C2=
=A0 2=C2=A0 2=C2=A0 2=C2=A0 0.0000W=C2=A0 0.0000W 0<br>
=C2=A03:=C2=A0 0.0500W*=C2=A0 =C2=A01.500ms=C2=A0 =C2=A01.500ms=C2=A0 3=C2=
=A0 3=C2=A0 3=C2=A0 3=C2=A0 0.0000W=C2=A0 0.0000W 0<br>
=C2=A04:=C2=A0 0.0050W*=C2=A0 =C2=A06.000ms=C2=A0 14.000ms=C2=A0 4=C2=A0 4=
=C2=A0 4=C2=A0 4=C2=A0 0.0000W=C2=A0 0.0000W 0<br>
=C2=A05:=C2=A0 0.0030W*=C2=A0 50.000ms=C2=A0 80.000ms=C2=A0 5=C2=A0 5=C2=A0=
 5=C2=A0 5=C2=A0 0.0000W=C2=A0 0.0000W 0<br>
<br>
The problem is that only one of the NOPS is working as expected<br>
(state 3). Another two (states 4-5) skyrocket the controller&#39;s power<br=
>
consumption far beyond normal (0-2) power states do, and far beyond<br>
reasonable. For example, when the controller is in state 3, my<br>
system consumes about 3-3.5 W at idle (according to acpiconf with<br>
laptop power cable unplugged), in states 0-2 - about 4 W, and in<br>
states 4-5 consumption is approaching 6 W. Thus, the NVMe becomes<br>
the hottest part of the system (&gt;50C, still idle), and it eats up<br>
almost half of the battery alone.<br>
<br>
Linux doesn&#39;t have this issue, so it seems to be nvme(4) related.<br>
All the above data is collected on 14.1-RELEASE Live USB with no<br>
filesystem mounted. 15-CURRENT has the same problem.<br>
<br>
Any ideas what it might be?<br></blockquote></div></div><div dir=3D"auto"><=
br></div><div dir=3D"auto">Does Linux have active power state management?</=
div><div dir=3D"auto"><br></div><div dir=3D"auto">Warner</div><div dir=3D"a=
uto"><br></div><div dir=3D"auto"><div class=3D"gmail_quote"><blockquote cla=
ss=3D"gmail_quote" style=3D"margin:0 0 0 .8ex;border-left:1px #ccc solid;pa=
dding-left:1ex">
<br>
Regards,<br>
Alexey<br>
<br>
</blockquote></div></div></div>

--000000000000dce7450622c6464f--



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?CANCZdfpv6EbB_HSnZG4Sy4crOJBbX16PGo3iwjCdkBzvTE7SVA>