linux-pci.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Bjorn Helgaas <helgaas@kernel.org>
To: Ville Syrjala <ville.syrjala@linux.intel.com>
Cc: intel-gfx@lists.freedesktop.org,
	Bjorn Helgaas <bhelgaas@google.com>,
	"Rafael J. Wysocki" <rafael@kernel.org>,
	Rodrigo Vivi <rodrigo.vivi@intel.com>,
	linux-pci@vger.kernel.org
Subject: Re: [PATCH 6/6] drm/i915/pm: Use pci_dev->skip_bus_pm for hibernate vs. D3 workaround
Date: Wed, 25 Sep 2024 14:28:56 -0500	[thread overview]
Message-ID: <20240925192856.GA10650@bhelgaas> (raw)
In-Reply-To: <20240925144526.2482-7-ville.syrjala@linux.intel.com>

On Wed, Sep 25, 2024 at 05:45:26PM +0300, Ville Syrjala wrote:
> From: Ville Syrjälä <ville.syrjala@linux.intel.com>
> 
> On some older laptops we have to leave the device in D0
> during hibernation, or else the BIOS just hangs and never
> finishes the hibernation.
> 
> Currently we are achieving that by skipping the
> pci_set_power_state(D3). However we also need to call
> pci_save_state() ahead of time, or else
> pci_pm_suspend_noirq() will do the pci_set_power_state(D3)
> anyway.
> 
> This is all rather ugly, and might cause us to deviate from
> standard pci pm behaviour in unknown ways since we always
> call pci_save_state() for any kind of suspend operation.
> 
> Stop calling pci_save_state()+pci_set_power_state() entirely
> (apart from the switcheroo paths) and instead set
> pci_dev->skip_bus_pm=true to prevent the D3 during hibernation
> on old machines. Apart from that we'll just let the normal
> pci pm code take care of everything for us.
> 
> Cc: Bjorn Helgaas <bhelgaas@google.com>
> Cc: "Rafael J. Wysocki" <rafael@kernel.org>
> Cc: Rodrigo Vivi <rodrigo.vivi@intel.com>
> Cc: linux-pci@vger.kernel.org
> Cc: intel-gfx@lists.freedesktop.org
> Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
> ---
>  drivers/gpu/drm/i915/i915_driver.c | 12 ++++++------
>  1 file changed, 6 insertions(+), 6 deletions(-)
> 
> diff --git a/drivers/gpu/drm/i915/i915_driver.c b/drivers/gpu/drm/i915/i915_driver.c
> index c3e7225ea1ba..05948d00a874 100644
> --- a/drivers/gpu/drm/i915/i915_driver.c
> +++ b/drivers/gpu/drm/i915/i915_driver.c
> @@ -1123,13 +1123,9 @@ static int i915_drm_suspend_noirq(struct drm_device *dev, bool hibernation)
>  	 * Lenovo Thinkpad X301, X61s, X60, T60, X41
>  	 * Fujitsu FSC S7110
>  	 * Acer Aspire 1830T
> -	 *
> -	 * pci_save_state() is needed to prevent driver/pci from
> -	 * automagically putting the device into D3.
>  	 */
> -	pci_save_state(pdev);
> -	if (!(hibernation && GRAPHICS_VER(dev_priv) < 6))
> -		pci_set_power_state(pdev, PCI_D3hot);
> +	if (hibernation && GRAPHICS_VER(dev_priv) < 6)
> +		pdev->skip_bus_pm = true;

.skip_bus_pm was previously strictly internal to
drivers/pci/pci-driver.c.  Not sure about using it outside
drivers/pci/; would want Rafael to chime in.

>  	return 0;
>  }
> @@ -1137,6 +1133,7 @@ static int i915_drm_suspend_noirq(struct drm_device *dev, bool hibernation)
>  int i915_driver_suspend_switcheroo(struct drm_i915_private *i915,
>  				   pm_message_t state)
>  {
> +	struct pci_dev *pdev = to_pci_dev(i915->drm.dev);
>  	int error;
>  
>  	if (drm_WARN_ON_ONCE(&i915->drm, state.event != PM_EVENT_SUSPEND &&
> @@ -1158,6 +1155,9 @@ int i915_driver_suspend_switcheroo(struct drm_i915_private *i915,
>  	if (error)
>  		return error;
>  
> +	pci_save_state(pdev);
> +	pci_set_power_state(pdev, PCI_D3hot);
> +
>  	return 0;
>  }
>  
> -- 
> 2.44.2
> 

      reply	other threads:[~2024-09-25 19:28 UTC|newest]

Thread overview: 23+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2024-09-25 14:45 [PATCH 0/6] drm/i915/pm: Clean up the hibernate vs. PCI D3 quirk Ville Syrjala
2024-09-25 14:45 ` [PATCH 1/6] PCI/PM: Respect pci_dev->skip_bus_pm in the .poweroff() path Ville Syrjala
2024-09-25 19:28   ` Bjorn Helgaas
2024-09-26 16:03     ` Ville Syrjälä
2024-09-30 19:50       ` Bjorn Helgaas
2024-10-01 13:12         ` Ville Syrjälä
2024-10-23 15:31     ` Ville Syrjälä
2024-09-25 14:45 ` [PATCH 2/6] drm/i915/pm: Hoist pci_save_state()+pci_set_power_state() to the end of pm _late() hook Ville Syrjala
2024-09-26 14:43   ` Rodrigo Vivi
2024-09-26 15:41     ` Ville Syrjälä
2024-09-26 16:40       ` Rodrigo Vivi
2024-09-25 14:45 ` [PATCH 3/6] drm/i915/pm: Simplify pm hook documentation Ville Syrjala
2024-09-26 14:45   ` Rodrigo Vivi
2024-09-26 15:38     ` Ville Syrjälä
2024-09-26 16:10       ` Rodrigo Vivi
2024-09-25 14:45 ` [PATCH 4/6] drm/i915/pm: Move the hibernate+D3 quirk stuff into noirq() pm hooks Ville Syrjala
2024-09-25 19:28   ` Bjorn Helgaas
2024-09-25 14:45 ` [PATCH 5/6] drm/i915/pm: Do pci_restore_state() in switcheroo resume hook Ville Syrjala
2024-09-26 14:48   ` Rodrigo Vivi
2024-09-26 15:36     ` Ville Syrjälä
2024-09-26 16:27       ` Rodrigo Vivi
2024-09-25 14:45 ` [PATCH 6/6] drm/i915/pm: Use pci_dev->skip_bus_pm for hibernate vs. D3 workaround Ville Syrjala
2024-09-25 19:28   ` Bjorn Helgaas [this message]

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=20240925192856.GA10650@bhelgaas \
    --to=helgaas@kernel.org \
    --cc=bhelgaas@google.com \
    --cc=intel-gfx@lists.freedesktop.org \
    --cc=linux-pci@vger.kernel.org \
    --cc=rafael@kernel.org \
    --cc=rodrigo.vivi@intel.com \
    --cc=ville.syrjala@linux.intel.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).