From: Heiner Kallweit <hkallweit1@gmail.com>
To: AceLan Kao <acelan.kao@canonical.com>,
Realtek linux nic maintainers <nic_swsd@realtek.com>,
"David S. Miller" <davem@davemloft.net>,
netdev@vger.kernel.org, linux-kernel@vger.kernel.org
Subject: Re: [PATCH] r8169: only disable ASPM L1.1 support, instead of disabling them all
Date: Wed, 18 Mar 2020 12:09:14 +0100 [thread overview]
Message-ID: <78aa4db2-6dec-e23c-03f4-f76577de756f@gmail.com> (raw)
In-Reply-To: <20200318014548.14547-1-acelan.kao@canonical.com>
On 18.03.2020 02:45, AceLan Kao wrote:
> The issues which have been seen by enabling ASPM support are from the
> BIOS that enables the ASPM L1.1 support on the device. It leads to some
> strange behaviors when the device enter L1.1 state.
> So, we don't have to disable ASPM support entriely, just disable L1.1
> state, that fixes the issues and also has good power management.
>
Meanwhile you can use sysfs to re-enable selected ASPM states, see
entries in "link" directory under the PCI device (provided that BIOS
allows OS to control ASPM). This allows users with mobile devices
w/o the ASPM issue to benefit from the ASPM power savings.
There are ~ 50 RTL8168 chip versions, used on different platforms and
dozens of consumer mainboards (with more or less buggy BIOS versions).
This leaves a good chance that some users may face issues with L0s/L1
enabled. And unfortunately the symptoms of ASPM issues haven't always
been obvious, sometimes just the performance was reduced.
Having said that I'd prefer to keep ASPM an opt-in feature.
Heiner
> Signed-off-by: AceLan Kao <acelan.kao@canonical.com>
> ---
> drivers/net/ethernet/realtek/r8169_main.c | 7 +++----
> 1 file changed, 3 insertions(+), 4 deletions(-)
>
> diff --git a/drivers/net/ethernet/realtek/r8169_main.c b/drivers/net/ethernet/realtek/r8169_main.c
> index a2168a14794c..b52680e7323b 100644
> --- a/drivers/net/ethernet/realtek/r8169_main.c
> +++ b/drivers/net/ethernet/realtek/r8169_main.c
> @@ -5473,11 +5473,10 @@ static int rtl_init_one(struct pci_dev *pdev, const struct pci_device_id *ent)
> if (rc)
> return rc;
>
> - /* Disable ASPM completely as that cause random device stop working
> - * problems as well as full system hangs for some PCIe devices users.
> + /* r8169 suppots ASPM L0 and L1 well, and doesn't support L1.1,
> + * so disable ASPM L1.1 only.
> */
> - rc = pci_disable_link_state(pdev, PCIE_LINK_STATE_L0S |
> - PCIE_LINK_STATE_L1);
> + rc = pci_disable_link_state(pdev, PCIE_LINK_STATE_L1_1);
> tp->aspm_manageable = !rc;
>
> /* enable device (incl. PCI PM wakeup and hotplug setup) */
>
next prev parent reply other threads:[~2020-03-18 11:09 UTC|newest]
Thread overview: 3+ messages / expand[flat|nested] mbox.gz Atom feed top
2020-03-18 1:45 AceLan Kao
2020-03-18 11:09 ` Heiner Kallweit [this message]
2020-03-22 3:20 ` David Miller
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=78aa4db2-6dec-e23c-03f4-f76577de756f@gmail.com \
--to=hkallweit1@gmail.com \
--cc=acelan.kao@canonical.com \
--cc=davem@davemloft.net \
--cc=linux-kernel@vger.kernel.org \
--cc=netdev@vger.kernel.org \
--cc=nic_swsd@realtek.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
all inboxes | Powered by JetHome®