linux-kernel - Re: [PATCH] platform/x86/intel/pmc/mtl: Put GNA/IPU/VPU devices in D3

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [day] [month] [year] [list]

Message-ID: <55d2d2d2b20df6e086748cf7404e17e2f4307994.camel@linux.intel.com>
Date:   Sun, 09 Apr 2023 12:14:00 -0700
From:   "David E. Box" <david.e.box@...ux.intel.com>
To:     Hans de Goede <hdegoede@...hat.com>,
        srinivas.pandruvada@...ux.intel.com, irenic.rajneesh@...il.com,
        markgross@...nel.org, rjw@...ysocki.net
Cc:     platform-driver-x86@...r.kernel.org, linux-kernel@...r.kernel.org
Subject: Re: [PATCH] platform/x86/intel/pmc/mtl: Put GNA/IPU/VPU devices in
 D3

On Sat, 2023-04-08 at 10:57 +0200, Hans de Goede wrote:
> Hi David,
> 
> On 4/8/23 04:26, David E. Box wrote:
> > On Meteor Lake, the GNA, IPU, and VPU devices are booted in D0 power state
> > and will block the SoC from going into the deepest Package C-state if a
> > driver is not present. Put each device in D3hot if no driver is found.
> > 
> > Signed-off-by: David E. Box <david.e.box@...ux.intel.com>
> > ---
> >  drivers/platform/x86/intel/pmc/mtl.c | 31 ++++++++++++++++++++++++++++
> >  1 file changed, 31 insertions(+)
> > 
> > diff --git a/drivers/platform/x86/intel/pmc/mtl.c
> > b/drivers/platform/x86/intel/pmc/mtl.c
> > index eeb3bd8c2502..33aa98b54049 100644
> > --- a/drivers/platform/x86/intel/pmc/mtl.c
> > +++ b/drivers/platform/x86/intel/pmc/mtl.c
> > @@ -8,6 +8,7 @@
> >   *
> >   */
> >  
> > +#include <linux/pci.h>
> >  #include "core.h"
> >  
> >  const struct pmc_reg_map mtl_reg_map = {
> > @@ -45,8 +46,38 @@ void mtl_core_configure(struct pmc_dev *pmcdev)
> >         pmc_core_send_ltr_ignore(pmcdev, 3);
> >  }
> >  
> > +#define MTL_GNA_PCI_DEV        0x7e4c
> > +#define MTL_IPU_PCI_DEV        0x7d19
> > +#define MTL_VPU_PCI_DEV        0x7d1d
> > +void mtl_set_device_d3(unsigned int device)
> 
> As pointed out by "kernel test robot <lkp@...el.com>" this needs to
> be static. Otherwise this LGTM. With this fixed please add my:
> 
> Reviewed-by: Hans de Goede <hdegoede@...hat.com>
> 
> to version 2 of the patch.
> 
> Regards,
> 
> Hans

Thanks Hans

> 
> 
> 
> 
> > +{
> > +       struct pci_dev *pcidev;
> > +
> > +       pcidev = pci_get_device(PCI_VENDOR_ID_INTEL, device, NULL);
> > +       if (pcidev) {
> > +               if (!device_trylock(&pcidev->dev)) {
> > +                       pci_dev_put(pcidev);
> > +                       return;
> > +               }
> > +               if (!pcidev->dev.driver) {
> > +                       dev_info(&pcidev->dev, "Setting to D3hot\n");
> > +                       pci_set_power_state(pcidev, PCI_D3hot);
> > +               }
> > +               device_unlock(&pcidev->dev);
> > +               pci_dev_put(pcidev);
> > +       }
> > +}
> > +
> >  void mtl_core_init(struct pmc_dev *pmcdev)
> >  {
> >         pmcdev->map = &mtl_reg_map;
> >         pmcdev->core_configure = mtl_core_configure;
> > +
> > +       /*
> > +        * Set power state of select devices that do not have drivers to D3
> > +        * so that they do not block Package C entry.
> > +        */
> > +       mtl_set_device_d3(MTL_GNA_PCI_DEV);
> > +       mtl_set_device_d3(MTL_IPU_PCI_DEV);
> > +       mtl_set_device_d3(MTL_VPU_PCI_DEV);
> >  }
> > 
> > base-commit: 4f59630a5ed0a4e7d275bd7e5d253a8f5a425c5a
>