From: kai.heng.feng@canonical.com (Kai-Heng Feng)
Subject: [Regression] Commit "nvme/pci: Use host managed power state for suspend" has problems
Date: Fri, 26 Jul 2019 02:20:19 +0800 [thread overview]
Message-ID: <46857EDA-081E-4622-A8D7-D337127A38DE@canonical.com> (raw)
In-Reply-To: <CAJZ5v0iDQ4=kTUgW94tKGt7oJzA_3uVU_M6HAMbNCRXwp_do8A@mail.gmail.com>
at 01:03, Rafael J. Wysocki <rafael@kernel.org> wrote:
> On Thu, Jul 25, 2019@6:24 PM <Mario.Limonciello@dell.com> wrote:
>> +Rajat
>>
>>> -----Original Message-----
>>> From: Kai-Heng Feng <kai.heng.feng at canonical.com>
>>> Sent: Thursday, July 25, 2019 9:03 AM
>>> To: Rafael J. Wysocki
>>> Cc: Keith Busch; Christoph Hellwig; Sagi Grimberg; linux-
>>> nvme at lists.infradead.org; Limonciello, Mario; Linux PM; LKML
>>> Subject: Re: [Regression] Commit "nvme/pci: Use host managed power
>>> state for
>>> suspend" has problems
>>>
>>>
>>> [EXTERNAL EMAIL]
>>>
>>> Hi Rafael,
>>>
>>>@17:51, Rafael J. Wysocki <rjw@rjwysocki.net> wrote:
>>>
>>>> Hi Keith,
>>>>
>>>> Unfortunately,
>>>>
>>>> commit d916b1be94b6dc8d293abed2451f3062f6af7551
>>>> Author: Keith Busch <keith.busch at intel.com>
>>>> Date: Thu May 23 09:27:35 2019 -0600
>>>>
>>>> nvme-pci: use host managed power state for suspend
>>>>
>>>> doesn't universally improve things. In fact, in some cases it makes
>>>> things worse.
>>>>
>>>> For example, on the Dell XPS13 9380 I have here it prevents the
>>>> processor
>>>> package
>>>> from reaching idle states deeper than PC2 in suspend-to-idle (which, of
>>>> course, also
>>>> prevents the SoC from reaching any kind of S0ix).
>>>>
>>>> That can be readily explained too. Namely, with the commit above the
>>>> NVMe device
>>>> stays in D0 over suspend/resume, so the root port it is connected to
>>>> also
>>>> has to stay in
>>>> D0 and that "blocks" package C-states deeper than PC2.
>>>>
>>>> In order for the root port to be able to go to D3, the device connected
>>>> to it also needs
>>>> to go into D3, so it looks like (at least on this particular machine,
>>>> but
>>>> maybe in
>>>> general), both D3 and the NVMe-specific PM are needed.
>>
>> Well this is really unfortunate to hear. I recall that with some disks
>> we were
>> seeing problems where NVME specific PM wasn't working when the disk was
>> in D3.
>>
>> On your specific disk, it would be good to know if just removing the
>> pci_save_state(pdev)
>> call helps.
>
> Yes, it does help.
>
>> If so, :
>> * that might be a better option to add as a parameter.
>> * maybe we should double check all the disks one more time with that
>> tweak.
>
> At this point it seems so.
>
>>>> I'm not sure what to do here, because evidently there are systems where
>>>> that commit
>>>> helps. I was thinking about adding a module option allowing the user to
>>>> override the
>>>> default behavior which in turn should be compatible with 5.2 and earlier
>>>> kernels.
>>>
>>> I just briefly tested s2i on XPS 9370, and the power meter shows a
>>> 0.8~0.9W
>>> power consumption so at least I don?t see the issue on XPS 9370.
>>
>> To me that confirms NVME is down, but it still seems higher than I would
>> have
>> expected. We should be more typically in the order of ~0.3W I think.
From what I?ve observed, ~0.8W s2idle is already better than Windows (~1W).
0.3W is what I see during S5.
>
> It may go to PC10, but not reach S0ix.
>
> Anyway, I run the s2idle tests under turbostat which then tells me
> what has happened more precisely.
The XPS 9370 at my hand does reach to s0ix during s2idle:
# cat /sys/kernel/debug/pmc_core/slp_s0_residency_usec
15998400
So I think keep the root port in D0 is not the culprit here.
Maybe something is wrong on the ASPM settings?
Kai-Heng
next prev parent reply other threads:[~2019-07-25 18:20 UTC|newest]
Thread overview: 75+ messages / expand[flat|nested] mbox.gz Atom feed top
2019-07-25 9:51 [Regression] Commit "nvme/pci: Use host managed power state for suspend" has problems Rafael J. Wysocki
2019-07-25 14:02 ` Kai-Heng Feng
2019-07-25 16:23 ` Mario.Limonciello
2019-07-25 17:03 ` Rafael J. Wysocki
2019-07-25 17:23 ` Mario.Limonciello
2019-07-25 18:20 ` Kai-Heng Feng [this message]
2019-07-25 19:09 ` Mario.Limonciello
2019-07-30 10:45 ` Rafael J. Wysocki
2019-07-30 14:41 ` Keith Busch
2019-07-30 17:14 ` Mario.Limonciello
2019-07-30 18:50 ` Kai-Heng Feng
2019-07-30 19:19 ` Keith Busch
2019-07-30 21:05 ` Mario.Limonciello
2019-07-30 21:31 ` Keith Busch
2019-07-31 21:25 ` Rafael J. Wysocki
2019-07-31 22:19 ` Keith Busch
2019-07-31 22:33 ` Rafael J. Wysocki
2019-08-01 9:05 ` Kai-Heng Feng
2019-08-01 17:29 ` Rafael J. Wysocki
2019-08-01 19:05 ` Mario.Limonciello
2019-08-01 22:26 ` Rafael J. Wysocki
2019-08-02 10:55 ` Kai-Heng Feng
2019-08-02 11:04 ` Rafael J. Wysocki
2019-08-05 19:13 ` Kai-Heng Feng
2019-08-05 21:28 ` Rafael J. Wysocki
2019-08-06 14:02 ` Mario.Limonciello
2019-08-06 15:00 ` Rafael J. Wysocki
2019-08-07 10:29 ` Rafael J. Wysocki
2019-08-01 20:22 ` Keith Busch
2019-08-07 9:48 ` Rafael J. Wysocki
2019-08-07 10:45 ` Christoph Hellwig
2019-08-07 10:54 ` Rafael J. Wysocki
2019-08-07 9:53 ` [PATCH] nvme-pci: Do not prevent PCI bus-level PM from being used Rafael J. Wysocki
2019-08-07 10:14 ` Rafael J. Wysocki
2019-08-07 10:43 ` Christoph Hellwig
2019-08-07 14:37 ` Keith Busch
2019-08-08 8:36 ` [PATCH] nvme-pci: Allow PCI bus-level PM to be used if ASPM is disabled Rafael J. Wysocki
2019-08-08 8:48 ` Christoph Hellwig
2019-08-08 9:06 ` Rafael J. Wysocki
2019-08-08 10:03 ` [PATCH v2 0/2] " Rafael J. Wysocki
2019-08-08 10:06 ` [PATCH v2 1/2] PCI: PCIe: ASPM: Introduce pcie_aspm_enabled_mask() Rafael J. Wysocki
2019-08-08 13:15 ` Bjorn Helgaas
2019-08-08 14:48 ` Rafael J. Wysocki
2019-08-08 10:10 ` [PATCH v2 2/2] nvme-pci: Allow PCI bus-level PM to be used if ASPM is disabled Rafael J. Wysocki
2019-08-08 13:43 ` Bjorn Helgaas
2019-08-08 14:47 ` Rafael J. Wysocki
2019-08-08 17:06 ` Rafael J. Wysocki
2019-08-08 18:39 ` Bjorn Helgaas
2019-08-08 20:01 ` Keith Busch
2019-08-08 20:05 ` Mario.Limonciello
2019-08-08 20:41 ` Rafael J. Wysocki
2019-08-09 4:47 ` Bjorn Helgaas
2019-08-09 8:04 ` Rafael J. Wysocki
2019-08-08 21:51 ` [PATCH v3 0/2] " Rafael J. Wysocki
2019-08-08 21:55 ` [PATCH v3 1/2] PCI: PCIe: ASPM: Introduce pcie_aspm_enabled() Rafael J. Wysocki
2019-08-09 4:50 ` Bjorn Helgaas
2019-08-09 8:00 ` Rafael J. Wysocki
2019-10-07 22:34 ` Bjorn Helgaas
2019-10-08 9:27 ` Rafael J. Wysocki
2019-10-08 21:16 ` Bjorn Helgaas
2019-10-08 22:54 ` Rafael J. Wysocki
2019-10-09 12:49 ` Bjorn Helgaas
2019-08-08 21:58 ` [PATCH v3 2/2] nvme-pci: Allow PCI bus-level PM to be used if ASPM is disabled Rafael J. Wysocki
2019-08-08 22:13 ` [PATCH v3 0/2] " Keith Busch
2019-08-09 8:05 ` Rafael J. Wysocki
2019-08-09 14:52 ` Keith Busch
2019-07-25 16:59 ` [Regression] Commit "nvme/pci: Use host managed power state for suspend" has problems Rafael J. Wysocki
2019-07-25 14:52 ` Keith Busch
2019-07-25 19:48 ` Rafael J. Wysocki
2019-07-25 19:52 ` Keith Busch
2019-07-25 20:02 ` Rafael J. Wysocki
2019-07-26 14:02 ` Kai-Heng Feng
2019-07-27 12:55 ` Rafael J. Wysocki
2019-07-29 15:51 ` Mario.Limonciello
2019-07-29 22:05 ` Rafael J. Wysocki
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=46857EDA-081E-4622-A8D7-D337127A38DE@canonical.com \
--to=kai.heng.feng@canonical.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).