From: Bjorn Helgaas <bhelgaas-hpIqsD4AKlfQT0dZR+AlfA@public.gmane.org>
To: Emmanuel Grumbach <egrumbach-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org>
Cc: Matthew Garrett
<matthew.garrett-05XSO3Yj/JvQT0dZR+AlfA@public.gmane.org>,
Stanislaw Gruszka
<sgruszka-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org>,
"linux-pci-u79uwXL29TY76Z2rM5mHXA@public.gmane.org"
<linux-pci-u79uwXL29TY76Z2rM5mHXA@public.gmane.org>,
linux-wireless
<linux-wireless-u79uwXL29TY76Z2rM5mHXA@public.gmane.org>,
John Linville <linville-2XuSBdqkA4R54TAoqtyWWQ@public.gmane.org>,
Roman Yepishev
<roman.yepishev-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org>,
"Guy,
Wey-Yi" <wey-yi.w.guy-ral2JQCrhuEAvxtiuMwx3w@public.gmane.org>,
Mike Miller <mike.miller-VXdhtT5mjnY@public.gmane.org>,
iss_storagedev-VXdhtT5mjnY@public.gmane.org,
Guo-Fu Tseng <cooldavid-eHdHF5TYP0iWVfeAwA7xHQ@public.gmane.org>,
netdev-u79uwXL29TY76Z2rM5mHXA@public.gmane.org,
Francois Romieu <romieu-W8zweXLXuWQS+FvcfC7Uqw@public.gmane.org>,
nic_swsd-Rasf1IRRPZFBDgjK7y7TUQ@public.gmane.org,
aacraid-ugMvIWC9OiRBDgjK7y7TUQ@public.gmane.org,
"Rafael J. Wysocki" <rjw-KKrjLPT3xs0@public.gmane.org>,
linux-kernel-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
Subject: Re: is L1 really disabled in iwlwifi
Date: Fri, 10 May 2013 16:52:57 -0600 [thread overview]
Message-ID: <20130510225257.GA10847@google.com> (raw)
In-Reply-To: <CAErSpo6OCyTy19u6Xaf=xr0TSDriwCD3n-oMc7eJyLzuJ9d60g-JsoAwUIsXosN+BqQ9rBEUg@public.gmane.org>
[+cc Rafael, other pci_disable_link_state() users]
On Wed, May 01, 2013 at 11:13:15AM -0600, Bjorn Helgaas wrote:
> On Wed, May 1, 2013 at 2:31 AM, Emmanuel Grumbach <egrumbach-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org> wrote:
> > [from Bjorn's mail]
> >> In Emmanuel's case, we don't get _OSC control, so
> >> pci_disable_link_state() does nothing.
> >
> > Right, but this is true with the specific log I sent to you. Is it
> > possible that another platform / BIOS, we *will* get _OSC control and
> > that pci_disable_link_state() will actually do something? In that case
> > I would prefer not to remove the call to pcie_disable_link_state().
>
> Yes, absolutely, on many platforms we will get _OSC control, and
> pci_disable_link_state() will work as expected. The problem is that
> the driver doesn't have a good way to know whether pci_disable_link()
> did anything or not.
>
> Today I think we have:
>
> 1) If the BIOS grants the OS permission to control PCIe services via
> _OSC, pci_disable_link_state() works and L1 will be disabled.
>
> 2) If the BIOS does not grant permission, pci_disable_link_state()
> does nothing and L1 may be enabled or not depending on what
> configuration the BIOS did.
>
> If the device really doesn't work reliably when L1 is enabled, we're
> currently at the mercy of the BIOS -- if the BIOS enables L1 but
> doesn't grant us permission via _OSC, L1 will remain enabled (as it is
> on your system).
I propose the following patch. Any comments?
commit cd11e3f87c4d2777cf8921c0454500c9baa54b46
Author: Bjorn Helgaas <bhelgaas-hpIqsD4AKlfQT0dZR+AlfA@public.gmane.org>
Date: Fri May 10 15:54:35 2013 -0600
PCI/ASPM: Allow drivers to disable ASPM unconditionally
Some devices have hardware problems related to using ASPM. Drivers for
these devices use pci_disable_link_state() to prevent their device from
entering L0s or L1. But on platforms where the OS doesn't have permission
to manage ASPM, pci_disable_link_state() does nothing, and the driver has
no way to know this.
Therefore, if the BIOS enables ASPM but declines (either via the FADT
ACPI_FADT_NO_ASPM bit or the _OSC method) to allow the OS to manage it,
the device can still use ASPM and trip over the hardware issue.
This patch makes pci_disable_link_state() disable ASPM unconditionally,
regardless of whether the OS has permission to manage ASPM in general.
Reported-by: Emmanuel Grumbach <egrumbach-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org>
Reference: https://lkml.kernel.org/r/CANUX_P3F5YhbZX3WGU-j1AGpbXb_T9Bis2ErhvKkFMtDvzatVQ-JsoAwUIsXosN+BqQ9rBEUg@public.gmane.org
Reference: https://bugzilla.kernel.org/show_bug.cgi?id=57331
Signed-off-by: Bjorn Helgaas <bhelgaas-hpIqsD4AKlfQT0dZR+AlfA@public.gmane.org>
diff --git a/drivers/pci/pcie/aspm.c b/drivers/pci/pcie/aspm.c
index d320df6..9ef4ab8 100644
--- a/drivers/pci/pcie/aspm.c
+++ b/drivers/pci/pcie/aspm.c
@@ -718,15 +718,11 @@ void pcie_aspm_powersave_config_link(struct pci_dev *pdev)
* pci_disable_link_state - disable pci device's link state, so the link will
* never enter specific states
*/
-static void __pci_disable_link_state(struct pci_dev *pdev, int state, bool sem,
- bool force)
+static void __pci_disable_link_state(struct pci_dev *pdev, int state, bool sem)
{
struct pci_dev *parent = pdev->bus->self;
struct pcie_link_state *link;
- if (aspm_disabled && !force)
- return;
-
if (!pci_is_pcie(pdev))
return;
@@ -757,13 +753,13 @@ static void __pci_disable_link_state(struct pci_dev *pdev, int state, bool sem,
void pci_disable_link_state_locked(struct pci_dev *pdev, int state)
{
- __pci_disable_link_state(pdev, state, false, false);
+ __pci_disable_link_state(pdev, state, false);
}
EXPORT_SYMBOL(pci_disable_link_state_locked);
void pci_disable_link_state(struct pci_dev *pdev, int state)
{
- __pci_disable_link_state(pdev, state, true, false);
+ __pci_disable_link_state(pdev, state, true);
}
EXPORT_SYMBOL(pci_disable_link_state);
@@ -781,7 +777,7 @@ void pcie_clear_aspm(struct pci_bus *bus)
__pci_disable_link_state(child, PCIE_LINK_STATE_L0S |
PCIE_LINK_STATE_L1 |
PCIE_LINK_STATE_CLKPM,
- false, true);
+ false);
}
}
--
To unsubscribe from this list: send the line "unsubscribe linux-wireless" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
next parent reply other threads:[~2013-05-10 22:52 UTC|newest]
Thread overview: 5+ messages / expand[flat|nested] mbox.gz Atom feed top
[not found] <CANUX_P2Hy02SnyYS24dyUGLv3wB3L5xkXt8Y1s+8_RG9d5ReAw@mail.gmail.com>
[not found] ` <CANUX_P3znXgMN5aBZbEVx_zKpTQ98my1C_cBG5+mERFWrrrLEw@mail.gmail.com>
[not found] ` <CANUX_P3Mt3X8JXq5siSijrWGkjeWbTCcdmJ-Fuv6U8jUqrFouQ@mail.gmail.com>
[not found] ` <CAErSpo66=UngrHNcEDeh4gXnt5c0qJHwbxum0mBz-rSWGduLhg@mail.gmail.com>
[not found] ` <CANUX_P0hpx8NNvX6cJXfOZMNYN8hrEF-gzf9hBN2Uz=k0WiwgA@mail.gmail.com>
[not found] ` <CANUX_P1JnFk7yovC7kXBMDWiErGhvzDQrNCm8te39m41SonCNg@mail.gmail.com>
[not found] ` <CAErSpo5WbdCx0iyt14bJxCiNBtUWjT2rMeg5YwNQaRt=M+_v8Q@mail.gmail.com>
[not found] ` <1367362536.9976.9.camel@x230>
[not found] ` <CANUX_P1sS8+3s-8-TyovHEwpa5qgQDj4W8P_hvfQjE2MKcTNEA@mail.gmail.com>
[not found] ` <CAErSpo6OCyTy19u6Xaf=xr0TSDriwCD3n-oMc7eJyLzuJ9d60g@mail.gmail.com>
[not found] ` <CAErSpo6OCyTy19u6Xaf=xr0TSDriwCD3n-oMc7eJyLzuJ9d60g-JsoAwUIsXosN+BqQ9rBEUg@public.gmane.org>
2013-05-10 22:52 ` Bjorn Helgaas [this message]
[not found] ` <20130510225257.GA10847-hpIqsD4AKlfQT0dZR+AlfA@public.gmane.org>
2013-05-11 20:26 ` is L1 really disabled in iwlwifi Rafael J. Wysocki
2013-05-11 20:22 ` Matthew Garrett
2013-05-16 22:55 ` Bjorn Helgaas
2013-05-17 5:49 ` Emmanuel Grumbach
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=20130510225257.GA10847@google.com \
--to=bhelgaas-hpiqsd4aklfqt0dzr+alfa@public.gmane.org \
--cc=aacraid-ugMvIWC9OiRBDgjK7y7TUQ@public.gmane.org \
--cc=cooldavid-eHdHF5TYP0iWVfeAwA7xHQ@public.gmane.org \
--cc=egrumbach-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org \
--cc=iss_storagedev-VXdhtT5mjnY@public.gmane.org \
--cc=linux-kernel-u79uwXL29TY76Z2rM5mHXA@public.gmane.org \
--cc=linux-pci-u79uwXL29TY76Z2rM5mHXA@public.gmane.org \
--cc=linux-wireless-u79uwXL29TY76Z2rM5mHXA@public.gmane.org \
--cc=linville-2XuSBdqkA4R54TAoqtyWWQ@public.gmane.org \
--cc=matthew.garrett-05XSO3Yj/JvQT0dZR+AlfA@public.gmane.org \
--cc=mike.miller-VXdhtT5mjnY@public.gmane.org \
--cc=netdev-u79uwXL29TY76Z2rM5mHXA@public.gmane.org \
--cc=nic_swsd-Rasf1IRRPZFBDgjK7y7TUQ@public.gmane.org \
--cc=rjw-KKrjLPT3xs0@public.gmane.org \
--cc=roman.yepishev-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org \
--cc=romieu-W8zweXLXuWQS+FvcfC7Uqw@public.gmane.org \
--cc=sgruszka-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org \
--cc=wey-yi.w.guy-ral2JQCrhuEAvxtiuMwx3w@public.gmane.org \
/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).