From: Don Dutile <ddutile@redhat.com>
To: Jiang Liu <liuj97@gmail.com>
Cc: Bjorn Helgaas <bhelgaas@google.com>,
Jiang Liu <jiang.liu@huawei.com>, Yinghai Lu <yinghai@kernel.org>,
Taku Izumi <izumi.taku@jp.fujitsu.com>,
"Rafael J . Wysocki" <rjw@sisk.pl>,
Kenji Kaneshige <kaneshige.kenji@jp.fujitsu.com>,
Yijing Wang <wangyijing@huawei.com>,
linux-kernel@vger.kernel.org, linux-pci@vger.kernel.org
Subject: Re: [RFC PATCH v2 04/32] PCI: add PCIe capabilities access functions to hide differences among PCIe specs
Date: Tue, 24 Jul 2012 17:12:31 -0400 [thread overview]
Message-ID: <500F0FBF.9080903@redhat.com> (raw)
In-Reply-To: <1343147504-25891-5-git-send-email-jiang.liu@huawei.com>
On 07/24/2012 12:31 PM, Jiang Liu wrote:
> From: Jiang Liu<jiang.liu@huawei.com>
>
> Introduce five configuration access functions for PCIe capabilities registers
> to hide differences among PCIe Base Spec versions.
>
> Function pci_pcie_capability_read_word/dword() stores the PCIe Capabilities
> register value by the passed parameter val. If related PCIe Capabilities
> register is not implemented on the PCIe device, the passed parameter val
> will be set 0.
>
> Function pci_pcie_capability_write_word/dowrd() writes the value to PCIe
> Capability register.
>
> Function pci_pcie_capability_reg_implemeneted() checks whether a Capabilities
> register is implemented by the PCIe device.
>
> Signed-off-by: Jiang Liu<liuj97@gmail.com>
> Signed-off-by: Yijing Wang<wangyijing@huawei.com>
> ---
> drivers/pci/access.c | 157 ++++++++++++++++++++++++++++++++++++++++++++++
> include/linux/pci.h | 6 ++
> include/linux/pci_regs.h | 2 +
> 3 files changed, 165 insertions(+)
>
> diff --git a/drivers/pci/access.c b/drivers/pci/access.c
> index ba91a7e..59409e8 100644
> --- a/drivers/pci/access.c
> +++ b/drivers/pci/access.c
> @@ -469,3 +469,160 @@ void pci_cfg_access_unlock(struct pci_dev *dev)
> raw_spin_unlock_irqrestore(&pci_lock, flags);
> }
> EXPORT_SYMBOL_GPL(pci_cfg_access_unlock);
> +
> +static inline int pci_pcie_cap_version(const struct pci_dev *dev)
> +{
> + return dev->pcie_flags_reg& PCI_EXP_FLAGS_VERS;
> +}
> +
> +static inline bool pci_pcie_cap_has_devctl(const struct pci_dev *dev)
> +{
> + return true;
> +}
> +
> +static inline bool pci_pcie_cap_has_lnkctl(const struct pci_dev *dev)
> +{
> + int type = pci_pcie_type(dev);
> +
> + return pci_pcie_cap_version(dev)> 1 ||
> + type == PCI_EXP_TYPE_ROOT_PORT ||
> + type == PCI_EXP_TYPE_ENDPOINT ||
> + type == PCI_EXP_TYPE_LEG_END;
> +}
> +
> +static inline bool pci_pcie_cap_has_sltctl(const struct pci_dev *dev)
> +{
> + int type = pci_pcie_type(dev);
> +
> + return pci_pcie_cap_version(dev)> 1 ||
> + type == PCI_EXP_TYPE_ROOT_PORT ||
> + (type == PCI_EXP_TYPE_DOWNSTREAM&&
> + dev->pcie_flags_reg& PCI_EXP_FLAGS_SLOT);
> +}
> +
> +static inline bool pci_pcie_cap_has_rtctl(const struct pci_dev *dev)
> +{
> + int type = pci_pcie_type(dev);
> +
> + return pci_pcie_cap_version(dev)> 1 ||
> + type == PCI_EXP_TYPE_ROOT_PORT ||
> + type == PCI_EXP_TYPE_RC_EC;
> +}
> +
> +bool pci_pcie_capability_reg_implemented(struct pci_dev *dev, int pos)
> +{
> + if (!pci_is_pcie(dev))
> + return false;
> +
> + switch (pos) {
> + case PCI_EXP_FLAGS_TYPE:
> + return true;
> + case PCI_EXP_DEVCAP:
> + case PCI_EXP_DEVCTL:
> + case PCI_EXP_DEVSTA:
> + return pci_pcie_cap_has_devctl(dev);
> + case PCI_EXP_LNKCAP:
> + case PCI_EXP_LNKCTL:
> + case PCI_EXP_LNKSTA:
> + return pci_pcie_cap_has_lnkctl(dev);
> + case PCI_EXP_SLTCAP:
> + case PCI_EXP_SLTCTL:
> + case PCI_EXP_SLTSTA:
> + return pci_pcie_cap_has_sltctl(dev);
> + case PCI_EXP_RTCTL:
> + case PCI_EXP_RTCAP:
> + case PCI_EXP_RTSTA:
> + return pci_pcie_cap_has_rtctl(dev);
> + case PCI_EXP_DEVCAP2:
> + case PCI_EXP_DEVCTL2:
> + case PCI_EXP_LNKCAP2:
> + case PCI_EXP_LNKCTL2:
> + case PCI_EXP_LNKSTA2:
> + return pci_pcie_cap_version(dev)> 1;
> + default:
> + return false;
> + }
> +}
> +EXPORT_SYMBOL(pci_pcie_capability_reg_implemented);
> +
> +/*
> + * Quotation from PCIe Base Spec 3.0:
> + * For Functions that do not implement the Slot Capabilities,
> + * Slot Status, and Slot Control registers, these spaces must
> + * be hardwired to 0b, with the exception of the Presence Detect
> + * State bit in the Slot Status register of Downstream Ports,
> + * which must be hardwired to 1b.
> + */
> +int pci_pcie_capability_read_word(struct pci_dev *dev, int pos, u16 *val)
> +{
> + int ret = 0;
> +
> + *val = 0;
> + if (pos& 1)
> + return -EINVAL;
> +
> + if (pci_pcie_capability_reg_implemented(dev, pos)) {
> + ret = pci_read_config_word(dev, pci_pcie_cap(dev) + pos, val);
> + /*
> + * Reset *val to 0 if pci_read_config_word() fails, it may
> + * have been written as 0xFFFF if hardware error happens
> + * during pci_read_config_word().
> + */
> + if (ret)
> + *val = 0;
> + } else if (pos == PCI_EXP_SLTSTA&&
> + pci_pcie_type(dev) == PCI_EXP_TYPE_DOWNSTREAM) {
> + *val = PCI_EXP_SLTSTA_PDS;
> + }
Don't you want the above if check done 1st, and not the
pci_pcie_capability_reg_implemented(dev, pos) check ?
Isn't PCI_EXP_SLTCTL an implemented register, looking at this snippet?:
> + switch (pos) {
<snip>
> + case PCI_EXP_SLTCAP:
> + case PCI_EXP_SLTCTL:
> + case PCI_EXP_SLTSTA:
> + return pci_pcie_cap_has_sltctl(dev);
and the above function is:
> +static inline bool pci_pcie_cap_has_sltctl(const struct pci_dev *dev)
> +{
> + int type = pci_pcie_type(dev);
> +
> + return pci_pcie_cap_version(dev)> 1 ||
> + type == PCI_EXP_TYPE_ROOT_PORT ||
> + (type == PCI_EXP_TYPE_DOWNSTREAM&&
> + dev->pcie_flags_reg& PCI_EXP_FLAGS_SLOT);
> +}
or is PCI_EXP_FLAGS_SLOT not set when it should be, then the first condition
fails, and this function forces the val to 1b ?
> +
> + return ret;
> +}
> +EXPORT_SYMBOL(pci_pcie_capability_read_word);
> +
> +int pci_pcie_capability_read_dword(struct pci_dev *dev, int pos, u32 *val)
> +{
> + int ret = 0;
> +
> + *val = 0;
> + if (pos& 3)
> + return -EINVAL;
> +
> + if (pci_pcie_capability_reg_implemented(dev, pos)) {
> + ret = pci_read_config_dword(dev, pci_pcie_cap(dev) + pos, val);
> + /*
> + * Reset *val to 0 if pci_read_config_dword() fails, it may
> + * have been written as 0xFFFFFFFF if hardware error happens
> + * during pci_read_config_dword().
> + */
> + if (ret)
> + *val = 0;
> + } else if (pos == PCI_EXP_SLTCTL&&
> + pci_pcie_type(dev) == PCI_EXP_TYPE_DOWNSTREAM) {
> + *val = PCI_EXP_SLTSTA_PDS;
> + }
ditto above...
> +
> + return ret;
> +}
> +EXPORT_SYMBOL(pci_pcie_capability_read_dword);
> +
> +int pci_pcie_capability_write_word(struct pci_dev *dev, int pos, u16 val)
> +{
> + if (pos& 1)
> + return -EINVAL;
> + else if (!pci_pcie_capability_reg_implemented(dev, pos))
> + return 0;
> +
> + return pci_write_config_word(dev, pci_pcie_cap(dev) + pos, val);
> +}
> +EXPORT_SYMBOL(pci_pcie_capability_write_word);
> +
> +int pci_pcie_capability_write_dword(struct pci_dev *dev, int pos, u32 val)
> +{
> + if (pos& 3)
> + return -EINVAL;
> + else if (!pci_pcie_capability_reg_implemented(dev, pos))
> + return 0;
> +
> + return pci_write_config_dword(dev, pci_pcie_cap(dev) + pos, val);
> +}
> +EXPORT_SYMBOL(pci_pcie_capability_write_dword);
> diff --git a/include/linux/pci.h b/include/linux/pci.h
> index 9807da5..a9b7605 100644
> --- a/include/linux/pci.h
> +++ b/include/linux/pci.h
> @@ -816,6 +816,12 @@ static inline int pci_write_config_dword(const struct pci_dev *dev, int where,
> return pci_bus_write_config_dword(dev->bus, dev->devfn, where, val);
> }
>
> +int pci_pcie_capability_read_word(struct pci_dev *dev, int where, u16 *val);
> +int pci_pcie_capability_read_dword(struct pci_dev *dev, int where, u32 *val);
> +int pci_pcie_capability_write_word(struct pci_dev *dev, int where, u16 val);
> +int pci_pcie_capability_write_dword(struct pci_dev *dev, int where, u32 val);
> +bool pci_pcie_capability_reg_implemented(struct pci_dev *dev, int where);
> +
> /* user-space driven config access */
> int pci_user_read_config_byte(struct pci_dev *dev, int where, u8 *val);
> int pci_user_read_config_word(struct pci_dev *dev, int where, u16 *val);
> diff --git a/include/linux/pci_regs.h b/include/linux/pci_regs.h
> index 53274bf..5300fdf 100644
> --- a/include/linux/pci_regs.h
> +++ b/include/linux/pci_regs.h
> @@ -543,7 +543,9 @@
> #define PCI_EXP_OBFF_MSGB_EN 0x4000 /* OBFF enable with Message type B */
> #define PCI_EXP_OBFF_WAKE_EN 0x6000 /* OBFF using WAKE# signaling */
> #define PCI_CAP_EXP_ENDPOINT_SIZEOF_V2 44 /* v2 endpoints end here */
> +#define PCI_EXP_LNKCAP2 44 /* Link Capabilities 2 */
> #define PCI_EXP_LNKCTL2 48 /* Link Control 2 */
> +#define PCI_EXP_LNKSTA2 50 /* Link Status 2 */
> #define PCI_EXP_SLTCTL2 56 /* Slot Control 2 */
>
> /* Extended Capabilities (PCI-X 2.0 and Express) */
next prev parent reply other threads:[~2012-07-24 21:12 UTC|newest]
Thread overview: 34+ messages / expand[flat|nested] mbox.gz Atom feed top
2012-07-24 16:31 [RFC PATCH v2 00/32] provide interfaces to access PCIe capabilities registers Jiang Liu
2012-07-24 16:31 ` [RFC PATCH v2 01/32] PCI: add pcie_flags_reg into struct pci_dev to cache PCIe capabilities register Jiang Liu
2012-07-25 15:12 ` Don Dutile
2012-07-26 13:47 ` Yijing Wang
2012-07-24 16:31 ` [RFC PATCH v2 02/32] PCI: introduce pci_pcie_type(dev) to replace pci_dev->pcie_type Jiang Liu
2012-07-24 16:31 ` [RFC PATCH v2 03/32] PCI: remove unused field pcie_type from struct pci_dev Jiang Liu
2012-07-24 16:31 ` [RFC PATCH v2 04/32] PCI: add PCIe capabilities access functions to hide differences among PCIe specs Jiang Liu
2012-07-24 21:12 ` Don Dutile [this message]
2012-07-29 16:22 ` Jiang Liu
2012-07-24 16:31 ` [RFC PATCH v2 05/32] PCI/core: use PCIe capabilities access functions to simplify implementation Jiang Liu
2012-07-25 21:12 ` Don Dutile
2012-07-29 2:12 ` Jiang Liu
2012-07-24 16:31 ` [RFC PATCH v2 06/32] PCI/hotplug: " Jiang Liu
2012-07-24 16:31 ` [RFC PATCH v2 07/32] PCI/portdrv: " Jiang Liu
2012-07-25 5:51 ` Kaneshige, Kenji
2012-07-25 9:44 ` Jiang Liu
2012-07-24 16:31 ` [RFC PATCH v2 08/32] PCI/pciehp: " Jiang Liu
2012-07-24 16:31 ` [RFC PATCH v2 09/32] PCI/PME: " Jiang Liu
2012-07-24 16:31 ` [RFC PATCH v2 10/32] PCI/AER: " Jiang Liu
2012-07-24 16:31 ` [RFC PATCH v2 11/32] PCI/ASPM: " Jiang Liu
2012-07-24 16:31 ` [RFC PATCH v2 12/32] PCI/ARM: " Jiang Liu
2012-07-24 16:31 ` [RFC PATCH v2 13/32] PCI/MIPS: " Jiang Liu
2012-07-24 16:31 ` [RFC PATCH v2 14/32] PCI/tile: " Jiang Liu
2012-07-24 16:31 ` [RFC PATCH v2 15/32] PCI/r8169: " Jiang Liu
2012-07-24 16:31 ` [RFC PATCH v2 16/32] PCI/broadcom: " Jiang Liu
2012-07-24 16:31 ` [RFC PATCH v2 17/32] PCI/igb: " Jiang Liu
2012-07-24 16:31 ` [RFC PATCH v2 18/32] PCI/vxge: " Jiang Liu
2012-07-24 16:31 ` [RFC PATCH v2 19/32] PCI/mlx4: " Jiang Liu
2012-07-24 16:31 ` [RFC PATCH v2 20/32] PCI/niu: " Jiang Liu
2012-07-24 16:31 ` [RFC PATCH v2 21/32] PCI/myri10ge: " Jiang Liu
2012-07-24 16:31 ` [RFC PATCH v2 22/32] PCI/chelsio: " Jiang Liu
2012-07-24 16:31 ` [RFC PATCH v2 23/32] PCI/atl1c: " Jiang Liu
2012-07-24 21:09 ` [RFC PATCH v2 00/32] provide interfaces to access PCIe capabilities registers Don Dutile
2012-07-29 2:26 ` Jiang Liu
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=500F0FBF.9080903@redhat.com \
--to=ddutile@redhat.com \
--cc=bhelgaas@google.com \
--cc=izumi.taku@jp.fujitsu.com \
--cc=jiang.liu@huawei.com \
--cc=kaneshige.kenji@jp.fujitsu.com \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-pci@vger.kernel.org \
--cc=liuj97@gmail.com \
--cc=rjw@sisk.pl \
--cc=wangyijing@huawei.com \
--cc=yinghai@kernel.org \
/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).