linux-pci.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Ajay Kaher <akaher@vmware.com>
To: "bhelgaas@google.com" <bhelgaas@google.com>,
	"helgaas@kernel.org" <helgaas@kernel.org>,
	"gregkh@linuxfoundation.org" <gregkh@linuxfoundation.org>
Cc: "ymcavikas@gmail.com" <ymcavikas@gmail.com>,
	Alexey Makhalov <amakhalov@vmware.com>,
	Anish Swaminathan <anishs@vmware.com>,
	"linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>,
	"linux-pci@vger.kernel.org" <linux-pci@vger.kernel.org>,
	Srinidhi Rao <srinidhir@vmware.com>,
	"srivatsa@csail.mit.edu" <srivatsa@csail.mit.edu>,
	Srivatsa Bhat <srivatsab@vmware.com>,
	Vasavi Sirnapalli <vsirnapalli@vmware.com>
Subject: Re: [PATCH v3] PCI: Speed up device init by parsing capabilities all at once
Date: Thu, 5 May 2022 11:27:06 +0000	[thread overview]
Message-ID: <8773546D-E43A-4951-9713-44A09025AF5A@vmware.com> (raw)
In-Reply-To: <1645616049-19712-1-git-send-email-akaher@vmware.com>


On 23/02/22, 5:20 PM, "Ajay Kaher" <akaher@vmware.com> wrote:

> From: Vikash Bansal <bvikas@vmware.com>

Bjorn, as per your guidance we sent v3 (hope it's as per your expectation). 

Please let us know if you are looking for more improvement or 
wanted us to re-sent this patch.

-Ajay

> In the current implementation, the PCI capability list is parsed from
> the beginning to find each capability, which results in a large number
> of redundant PCI reads.
> 
> Instead, we can parse the complete list just once, store it in the
> pci_dev structure, and get the offset of each capability directly from
> the pci_dev structure.
> 
> This implementation improves pci devices initialization time  by ~2-3%
> (from 270ms to 261ms) in case of bare metal and 7-8% (From 201ms to 184ms)
> in case of VM running on ESXi.
> 
> It also adds a memory overhead of 20bytes (value of PCI_CAP_ID_MAX) per
> PCI device.
> 
> Ran pahole for pci_dev structure. This patch is not adding any padding
> bytes.
> 
> Signed-off-by: Vikash Bansal <bvikas@vmware.com>
> Signed-off-by: Ajay Kaher <akaher@vmware.com>
> 
> ---
> 
> Changes in v3:
>   - Added check before access cap_off from pci_find_capability().
>   - Moved declaration of pci_find_all_capabilities() to drivers/pci/pci.h.
>
> Changes in v2:
>  - Ran pahole tool.
>  - Modified comments to add "clock time".
>  - Removed comments before call to pci_find_all_capabilities.
>
> ---
> drivers/pci/pci.c   | 44 +++++++++++++++++++++++++++++++++++++-------
> drivers/pci/pci.h   |  1 +
> drivers/pci/probe.c |  1 +
> include/linux/pci.h |  1 +
> 4 files changed, 40 insertions(+), 7 deletions(-)
>
> diff --git a/drivers/pci/pci.c b/drivers/pci/pci.c
> index 3d2fb39..cf54811 100644
> --- a/drivers/pci/pci.c
> +++ b/drivers/pci/pci.c
> @@ -469,6 +469,40 @@ static u8 __pci_bus_find_cap_start(struct pci_bus *bus,
> }
>
> /**
> + * pci_find_all_capabilities - Read all capabilities
> + * @dev: the PCI device
> + *
> + * Read all capabilities and store offsets in cap_off
> + * array in pci_dev structure.
> + */
> +void pci_find_all_capabilities(struct pci_dev *dev)
> +{
> +	int ttl = PCI_FIND_CAP_TTL;
> +	u16 ent;
> +	u8 pos;
> +	u8 id;
> +
> +	pos = __pci_bus_find_cap_start(dev->bus, dev->devfn, dev->hdr_type);
> +	if (!pos)
> +		return;
> +	pci_bus_read_config_byte(dev->bus, dev->devfn, pos, &pos);
> +	while (ttl--) {
> +		if (pos < 0x40)
> +			break;
> +		pos &= ~3;
> +		pci_bus_read_config_word(dev->bus, dev->devfn, pos, &ent);
> +		id = ent & 0xff;
> +		if (id == 0xff)
> +			break;
> +
> +		/* Read first instance of capability */
> +		if (!(dev->cap_off[id]))
> +			dev->cap_off[id] = pos;
> +		pos = (ent >> 8);
> +	}
> +}
> +
> +/**
>   * pci_find_capability - query for devices' capabilities
>   * @dev: PCI device to query
>   * @cap: capability code
> @@ -489,13 +523,9 @@ static u8 __pci_bus_find_cap_start(struct pci_bus *bus,
>   */
> u8 pci_find_capability(struct pci_dev *dev, int cap)
> {
> -	u8 pos;
> -
> -	pos = __pci_bus_find_cap_start(dev->bus, dev->devfn, dev->hdr_type);
> -	if (pos)
> -		pos = __pci_find_next_cap(dev->bus, dev->devfn, pos, cap);
> -
> -	return pos;
> +	if(cap >= PCI_CAP_ID_MAX)
> +		return 0;
> +	return dev->cap_off[cap];
> }
> EXPORT_SYMBOL(pci_find_capability);
> 
> diff --git a/drivers/pci/pci.h b/drivers/pci/pci.h
> index 3d60cab..3cb70d5 100644
> --- a/drivers/pci/pci.h
> +++ b/drivers/pci/pci.h
> @@ -87,6 +87,7 @@ bool pci_bridge_d3_possible(struct pci_dev *dev);
> void pci_bridge_d3_update(struct pci_dev *dev);
> void pci_bridge_wait_for_secondary_bus(struct pci_dev *dev);
> void pci_bridge_reconfigure_ltr(struct pci_dev *dev);
> +void pci_find_all_capabilities(struct pci_dev *dev);
>
> static inline void pci_wakeup_event(struct pci_dev *dev)
> {
> diff --git a/drivers/pci/probe.c b/drivers/pci/probe.c
> index 087d365..d75e1fa 100644
> --- a/drivers/pci/probe.c
> +++ b/drivers/pci/probe.c
> @@ -1839,6 +1839,7 @@ int pci_setup_device(struct pci_dev *dev)
> 	dev->hdr_type = hdr_type & 0x7f;
> 	dev->multifunction = !!(hdr_type & 0x80);
> 	dev->error_state = pci_channel_io_normal;
> +	pci_find_all_capabilities(dev);
> 	set_pcie_port_type(dev);
>
> 	pci_set_of_node(dev);
> diff --git a/include/linux/pci.h b/include/linux/pci.h
> index 18a75c8..bd61cee 100644
> --- a/include/linux/pci.h
> +++ b/include/linux/pci.h
> @@ -326,6 +326,7 @@ struct pci_dev {
> 	unsigned int	class;		/* 3 bytes: (base,sub,prog-if) */
> 	u8		revision;	/* PCI revision, low byte of class word */
> 	u8		hdr_type;	/* PCI header type (`multi' flag masked out) */
> +	u8		cap_off[PCI_CAP_ID_MAX]; /* Offsets of all pci capabilities */
> #ifdef CONFIG_PCIEAER
> 	u16		aer_cap;	/* AER capability offset */
> 	struct aer_stats *aer_stats;	/* AER stats for this device */
> --
> 2.7.4



      reply	other threads:[~2022-05-05 11:27 UTC|newest]

Thread overview: 5+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2022-01-28  4:51 [PATCH v2] PCI: Speed up device init by parsing capabilities all at once Vikash Bansal
2022-02-02 21:35 ` Bjorn Helgaas
2022-02-03  5:52   ` Vikash Bansal
2022-02-23 11:34     ` [PATCH v3] " Ajay Kaher
2022-05-05 11:27       ` Ajay Kaher [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=8773546D-E43A-4951-9713-44A09025AF5A@vmware.com \
    --to=akaher@vmware.com \
    --cc=amakhalov@vmware.com \
    --cc=anishs@vmware.com \
    --cc=bhelgaas@google.com \
    --cc=gregkh@linuxfoundation.org \
    --cc=helgaas@kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-pci@vger.kernel.org \
    --cc=srinidhir@vmware.com \
    --cc=srivatsa@csail.mit.edu \
    --cc=srivatsab@vmware.com \
    --cc=vsirnapalli@vmware.com \
    --cc=ymcavikas@gmail.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).