linux-kernel - Re: [PATCH 4/5] PCI/ACPI: Centralize pcie_ports

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

Message-ID: <20201128214552.GA907628@bjorn-Precision-5520>
Date:   Sat, 28 Nov 2020 15:45:52 -0600
From:   Bjorn Helgaas <helgaas@...nel.org>
To:     "Kuppuswamy, Sathyanarayanan" 
        <sathyanarayanan.kuppuswamy@...ux.intel.com>
Cc:     ashok.raj@...el.com, knsathya@...nel.org,
        linux-pci@...r.kernel.org, linux-kernel@...r.kernel.org,
        Bjorn Helgaas <bhelgaas@...gle.com>
Subject: Re: [PATCH 4/5] PCI/ACPI: Centralize pcie_ports_native checking

On Wed, Nov 25, 2020 at 07:20:53PM -0800, Kuppuswamy, Sathyanarayanan wrote:
> On 11/25/20 5:18 PM, Bjorn Helgaas wrote:
> > From: Kuppuswamy Sathyanarayanan <sathyanarayanan.kuppuswamy@...ux.intel.com>
> > 
> > If the user booted with "pcie_ports=native", we take control of the PCIe
> > features unconditionally, regardless of what _OSC says.
> > 
> > Centralize the testing of pcie_ports_native in acpi_pci_root_create(),
> > where we interpret the _OSC results, so other places only have to check
> > host_bridge->native_X and we don't have to sprinkle tests of
> > pcie_ports_native everywhere.
> > 
> > [bhelgaas: commit log, rework OSC_PCI_EXPRESS_CONTROL_MASKS, logging]
> > Link: https://lore.kernel.org/r/bc87c9e675118960949043a832bed86bc22becbd.1603766889.git.sathyanarayanan.kuppuswamy@linux.intel.com
> > Signed-off-by: Kuppuswamy Sathyanarayanan <sathyanarayanan.kuppuswamy@...ux.intel.com>
> > Signed-off-by: Bjorn Helgaas <bhelgaas@...gle.com>
> > ---
> >   drivers/acpi/pci_root.c           | 19 +++++++++++++++++++
> >   drivers/pci/hotplug/pciehp_core.c |  2 +-
> >   drivers/pci/pci-acpi.c            |  3 ---
> >   drivers/pci/pcie/aer.c            |  2 +-
> >   drivers/pci/pcie/portdrv_core.c   |  9 +++------
> >   5 files changed, 24 insertions(+), 11 deletions(-)
> > 
> > diff --git a/drivers/acpi/pci_root.c b/drivers/acpi/pci_root.c
> > index 6db071038fd5..36142ed7b8f8 100644
> > --- a/drivers/acpi/pci_root.c
> > +++ b/drivers/acpi/pci_root.c
> > @@ -882,6 +882,8 @@ static void acpi_pci_root_release_info(struct pci_host_bridge *bridge)
> >   			flag = 0;	\
> >   	} while (0)
> > +#define FLAG(x)		((x) ? '+' : '-')
> > +
> >   struct pci_bus *acpi_pci_root_create(struct acpi_pci_root *root,
> >   				     struct acpi_pci_root_ops *ops,
> >   				     struct acpi_pci_root_info *info,
> > @@ -930,6 +932,23 @@ struct pci_bus *acpi_pci_root_create(struct acpi_pci_root *root,
> >   	OSC_OWNER(ctrl, OSC_PCI_EXPRESS_LTR_CONTROL, host_bridge->native_ltr);
> >   	OSC_OWNER(ctrl, OSC_PCI_EXPRESS_DPC_CONTROL, host_bridge->native_dpc);
> > +	if (pcie_ports_native) {
> > +		dev_info(&root->device->dev, "Taking control of PCIe-related features because \"pcie_ports=native\" specified; may conflict with firmware\n");
> > +		host_bridge->native_pcie_hotplug = 1;
> > +		host_bridge->native_aer = 1;
> > +		host_bridge->native_pme = 1;
> > +		host_bridge->native_ltr = 1;
> > +		host_bridge->native_dpc = 1;
> > +	}
> Won't it be better if its merged with above OSC_OWNER() calls? If
> pcie_ports_native is set, then above OSC_OWNER() calls for PCIe
> related features are redundant. Let me know.

Yes, I think you're right.  I think it would also be nice to know
exactly which services we're overriding _OSC for, i.e., when we turn
native_aer from 0 to 1 and so on.  If we merge this back in, we'll
have a way to keep track of those.

> > @@ -208,8 +208,7 @@ static int get_port_device_capability(struct pci_dev *dev)
> >   	struct pci_host_bridge *host = pci_find_host_bridge(dev->bus);
> >   	int services = 0;
> > -	if (dev->is_hotplug_bridge &&
> > -	    (pcie_ports_native || host->native_pcie_hotplug)) {
> > +	if (host->native_pcie_hotplug && dev->is_hotplug_bridge) {
> May be this reordering can be split into a different patch ?

Good point, that makes the diff a little easier to read, thanks!