Received-SPF: pass (google.com: best guess record for domain of linux-kernel-owner@vger.kernel.org designates 209.132.180.67 as permitted sender) client-ip=209.132.180.67;
DMARC-Filter: OpenDMARC Filter v1.3.2 mail.kernel.org E7D7A217A8
Date:   Fri, 30 Mar 2018 18:10:54 -0500
From:   Bjorn Helgaas <helgaas@kernel.org>
To:     KarimAllah Ahmed <karahmed@amazon.de>
Cc:     linux-kernel@vger.kernel.org, Bjorn Helgaas <bhelgaas@google.com>,
        linux-pci@vger.kernel.org
Subject: Re: [PATCH v4 1/2] PCI/IOV: Use VF0 cached config registers for
 other VFs
Message-ID: <20180330231054.GI9322@bhelgaas-glaptop.roam.corp.google.com>
References: <1521489961-18291-1-git-send-email-karahmed@amazon.de>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <1521489961-18291-1-git-send-email-karahmed@amazon.de>
User-Agent: Mutt/1.9.2 (2017-12-15)
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk

On Mon, Mar 19, 2018 at 09:06:00PM +0100, KarimAllah Ahmed wrote:
> Cache some config data from VF0 and use it for all other VFs instead of
> reading it from the config space of each VF.  We assume these items are the
> same across all associated VFs:
> 
>    Revision ID
>    Class Code
>    Subsystem Vendor ID
>    Subsystem ID
> 
> This is an optimization when enabling SR-IOV on a device with many VFs.
> 
> Cc: Bjorn Helgaas <bhelgaas@google.com>
> Cc: linux-pci@vger.kernel.org
> Cc: linux-kernel@vger.kernel.org
> Signed-off-by: KarimAllah Ahmed <karahmed@amazon.de>
> [bhelgaas: changelog, simplify comments, remove unused "device"]
> Signed-off-by: Bjorn Helgaas <helgaas@kernel.org>

I applied this one to pci/virtualization for v4.17, thanks!

I changed the ifdefs from CONFIG_PCI_ATS to CONFIG_PCI_IOV.  I know we
use CONFIG_PCI_ATS in linux/pci.h, but I think that's a mistake.

> ---
> v3->v4:
> - Restructure the code to handle CONFIG_PCI_ATS
> 
>  drivers/pci/iov.c   | 42 +++++++++++++++++++++++++++++++++++-------
>  drivers/pci/pci.h   |  4 ++++
>  drivers/pci/probe.c | 47 ++++++++++++++++++++++++++++++++++++++++++-----
>  3 files changed, 81 insertions(+), 12 deletions(-)
> 
> diff --git a/drivers/pci/iov.c b/drivers/pci/iov.c
> index 677924a..30bf8f7 100644
> --- a/drivers/pci/iov.c
> +++ b/drivers/pci/iov.c
> @@ -114,6 +114,29 @@ resource_size_t pci_iov_resource_size(struct pci_dev *dev, int resno)
>  	return dev->sriov->barsz[resno - PCI_IOV_RESOURCES];
>  }
>  
> +static void pci_read_vf_config_common(struct pci_dev *virtfn)
> +{
> +	struct pci_dev *physfn = virtfn->physfn;
> +
> +	/*
> +	 * Some config registers are the same across all associated VFs.
> +	 * Read them once from VF0 so we can skip reading them from the
> +	 * other VFs.
> +	 *
> +	 * PCIe r4.0, sec 9.3.4.1, technically doesn't require all VFs to
> +	 * have the same Revision ID and Subsystem ID, but we assume they
> +	 * do.
> +	 */
> +	pci_read_config_dword(virtfn, PCI_CLASS_REVISION,
> +			      &physfn->sriov->class);
> +	pci_read_config_byte(virtfn, PCI_HEADER_TYPE,
> +			     &physfn->sriov->hdr_type);
> +	pci_read_config_word(virtfn, PCI_SUBSYSTEM_VENDOR_ID,
> +			     &physfn->sriov->subsystem_vendor);
> +	pci_read_config_word(virtfn, PCI_SUBSYSTEM_ID,
> +			     &physfn->sriov->subsystem_device);
> +}
> +
>  int pci_iov_add_virtfn(struct pci_dev *dev, int id)
>  {
>  	int i;
> @@ -136,13 +159,17 @@ int pci_iov_add_virtfn(struct pci_dev *dev, int id)
>  	virtfn->devfn = pci_iov_virtfn_devfn(dev, id);
>  	virtfn->vendor = dev->vendor;
>  	virtfn->device = iov->vf_device;
> +	virtfn->is_virtfn = 1;
> +	virtfn->physfn = pci_dev_get(dev);
> +
> +	if (id == 0)
> +		pci_read_vf_config_common(virtfn);
> +
>  	rc = pci_setup_device(virtfn);
>  	if (rc)
> -		goto failed0;
> +		goto failed1;
>  
>  	virtfn->dev.parent = dev->dev.parent;
> -	virtfn->physfn = pci_dev_get(dev);
> -	virtfn->is_virtfn = 1;
>  	virtfn->multifunction = 0;
>  
>  	for (i = 0; i < PCI_SRIOV_NUM_BARS; i++) {
> @@ -163,10 +190,10 @@ int pci_iov_add_virtfn(struct pci_dev *dev, int id)
>  	sprintf(buf, "virtfn%u", id);
>  	rc = sysfs_create_link(&dev->dev.kobj, &virtfn->dev.kobj, buf);
>  	if (rc)
> -		goto failed1;
> +		goto failed2;
>  	rc = sysfs_create_link(&virtfn->dev.kobj, &dev->dev.kobj, "physfn");
>  	if (rc)
> -		goto failed2;
> +		goto failed3;
>  
>  	kobject_uevent(&virtfn->dev.kobj, KOBJ_CHANGE);
>  
> @@ -174,11 +201,12 @@ int pci_iov_add_virtfn(struct pci_dev *dev, int id)
>  
>  	return 0;
>  
> -failed2:
> +failed3:
>  	sysfs_remove_link(&dev->dev.kobj, buf);
> +failed2:
> +	pci_stop_and_remove_bus_device(virtfn);
>  failed1:
>  	pci_dev_put(dev);
> -	pci_stop_and_remove_bus_device(virtfn);
>  failed0:
>  	virtfn_remove_bus(dev->bus, bus);
>  failed:
> diff --git a/drivers/pci/pci.h b/drivers/pci/pci.h
> index fcd8191..bdb4ba2 100644
> --- a/drivers/pci/pci.h
> +++ b/drivers/pci/pci.h
> @@ -271,6 +271,10 @@ struct pci_sriov {
>  	u16		driver_max_VFs;	/* Max num VFs driver supports */
>  	struct pci_dev	*dev;		/* Lowest numbered PF */
>  	struct pci_dev	*self;		/* This PF */
> +	u32		class;		/* VF device */
> +	u8		hdr_type;	/* VF header type */
> +	u16		subsystem_vendor; /* VF subsystem vendor */
> +	u16		subsystem_device; /* VF subsystem device */
>  	resource_size_t	barsz[PCI_SRIOV_NUM_BARS];	/* VF BAR size */
>  	bool		drivers_autoprobe; /* Auto probing of VFs by driver */
>  };
> diff --git a/drivers/pci/probe.c b/drivers/pci/probe.c
> index ef53774..21ee1c3 100644
> --- a/drivers/pci/probe.c
> +++ b/drivers/pci/probe.c
> @@ -1389,6 +1389,43 @@ int pci_cfg_space_size(struct pci_dev *dev)
>  	return PCI_CFG_SPACE_SIZE;
>  }
>  
> +static int pci_cfg_space_class(struct pci_dev *dev)
> +{
> +	int class;
> +
> +#ifdef CONFIG_PCI_ATS
> +	if (dev->is_virtfn)
> +		return dev->physfn->sriov->class;
> +#endif
> +	pci_read_config_dword(dev, PCI_CLASS_REVISION, &class);
> +	return class;
> +}
> +
> +static void pci_cfg_space_subsystem(struct pci_dev *dev, u16 *vendor, u16 *device)
> +{
> +#ifdef CONFIG_PCI_ATS
> +	if (dev->is_virtfn) {
> +		*vendor = dev->physfn->sriov->subsystem_vendor;
> +		*device = dev->physfn->sriov->subsystem_device;
> +		return;
> +	}
> +#endif
> +	pci_read_config_word(dev, PCI_SUBSYSTEM_VENDOR_ID, vendor);
> +	pci_read_config_word(dev, PCI_SUBSYSTEM_ID, device);
> +}
> +
> +static u8 pci_cfg_space_hdr_type(struct pci_dev *dev)
> +{
> +	u8 hdr_type;
> +
> +#ifdef CONFIG_PCI_ATS
> +	if (dev->is_virtfn)
> +		return dev->physfn->sriov->hdr_type;
> +#endif
> +	pci_read_config_byte(dev, PCI_HEADER_TYPE, &hdr_type);
> +	return hdr_type;
> +}
> +
>  #define LEGACY_IO_RESOURCE	(IORESOURCE_IO | IORESOURCE_PCI_FIXED)
>  
>  static void pci_msi_setup_pci_dev(struct pci_dev *dev)
> @@ -1454,8 +1491,7 @@ int pci_setup_device(struct pci_dev *dev)
>  	struct pci_bus_region region;
>  	struct resource *res;
>  
> -	if (pci_read_config_byte(dev, PCI_HEADER_TYPE, &hdr_type))
> -		return -EIO;
> +	hdr_type = pci_cfg_space_hdr_type(dev);
>  
>  	dev->sysdata = dev->bus->sysdata;
>  	dev->dev.parent = dev->bus->bridge;
> @@ -1477,7 +1513,8 @@ int pci_setup_device(struct pci_dev *dev)
>  		     dev->bus->number, PCI_SLOT(dev->devfn),
>  		     PCI_FUNC(dev->devfn));
>  
> -	pci_read_config_dword(dev, PCI_CLASS_REVISION, &class);
> +	class = pci_cfg_space_class(dev);
> +
>  	dev->revision = class & 0xff;
>  	dev->class = class >> 8;		    /* upper 3 bytes */
>  
> @@ -1517,8 +1554,8 @@ int pci_setup_device(struct pci_dev *dev)
>  			goto bad;
>  		pci_read_irq(dev);
>  		pci_read_bases(dev, 6, PCI_ROM_ADDRESS);
> -		pci_read_config_word(dev, PCI_SUBSYSTEM_VENDOR_ID, &dev->subsystem_vendor);
> -		pci_read_config_word(dev, PCI_SUBSYSTEM_ID, &dev->subsystem_device);
> +
> +		pci_cfg_space_subsystem(dev, &dev->subsystem_vendor, &dev->subsystem_device);
>  
>  		/*
>  		 * Do the ugly legacy mode stuff here rather than broken chip
> -- 
> 2.7.4
>