linux-pci.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Shanker R Donthineni <sdonthineni@nvidia.com>
To: Bjorn Helgaas <helgaas@kernel.org>,
	Amey Narkhede <ameynarkhede03@gmail.com>
Cc: Bjorn Helgaas <bhelgaas@google.com>, <alex.williamson@redhat.com>,
	Raphael Norwitz <raphael.norwitz@nutanix.com>,
	<linux-pci@vger.kernel.org>, <linux-kernel@vger.kernel.org>,
	<kw@linux.com>, Sinan Kaya <okaya@kernel.org>,
	"Rafael J. Wysocki" <rjw@rjwysocki.net>
Subject: Re: [PATCH v5 5/7] PCI: Add support for a function level reset based on _RST method
Date: Sun, 6 Jun 2021 00:00:40 -0500	[thread overview]
Message-ID: <b24be059-4f90-f1b5-b49c-ba7076ace9ba@nvidia.com> (raw)
In-Reply-To: <20210605205317.GA2254430@bjorn-Precision-5520>

Hi Bjorn,

On 6/5/21 3:53 PM, Bjorn Helgaas wrote:
> This is a little sketchy.  We shouldn't be doing device config stuff
> after device_add() because that's when it becomes available for
> drivers to bind to the device.  If we do anything with the device
> after that point, we may interfere with a driver.
>
> I think the problem is that we don't call acpi_bind_one() until
> device_add().  There's some hackery in pci-acpi.c to deal with a
> similar problem for something else -- see acpi_pci_bridge_d3().
>
> I don't know how to fix this yet.  Here's the call graph that I think
> is relevant:
I've refactored pci_dev_acpi_reset() to avoid dependency on acpi_bind_one().
It can be called any time after creating pci_dev object. The code logic is
not exactly same as acpi_pci_bridge_d3() but similar flow. No need to set
ACPI_COMPANION since it would be updated eventually after probing the
reset methods.

Please review the below code and provide suggestions for the next step.

Updated patch:

[PATCH v5 5/7] PCI: Add support for ACPI _RST reset method

The _RST is a standard method specified in the ACPI specification. It
provides a function level reset when it is described in the acpi_device
context associated with PCI-device.

Implement a new reset function pci_dev_acpi_reset() for probing RST
method and execute if it is defined in the firmware. The ACPI binding
information is available only after calling device_add(). Since the
ACPI_COMPANION was not done before calling pci_init_reset_methods(),
use acpi_pci_find_companion() to know the ACPI binding.

The default priority of the ACPI reset is set to below device-specific
and above hardware resets.

Signed-off-by: Shanker Donthineni <sdonthineni@nvidia.com>
Suggested-by: Alex Williamson <alex.williamson@redhat.com>
Reviewed-by: Sinan Kaya <okaya@kernel.org>
---
 drivers/pci/pci-acpi.c | 30 ++++++++++++++++++++++++++++++
 drivers/pci/pci.c      |  1 +
 drivers/pci/pci.h      |  6 ++++++
 include/linux/pci.h    |  2 +-
 4 files changed, 38 insertions(+), 1 deletion(-)

diff --git a/drivers/pci/pci-acpi.c b/drivers/pci/pci-acpi.c
index 36bc23e217592..c344c33f5c910 100644
--- a/drivers/pci/pci-acpi.c
+++ b/drivers/pci/pci-acpi.c
@@ -934,6 +934,36 @@ static pci_power_t acpi_pci_choose_state(struct pci_dev *pdev)

 static struct acpi_device *acpi_pci_find_companion(struct device *dev);

+/**
+ * pci_dev_acpi_reset - do a function level reset using _RST method
+ * @dev: device to reset
+ * @probe: check if _RST method is included in the acpi_device context.
+ */
+int pci_dev_acpi_reset(struct pci_dev *dev, int probe)
+{
+       acpi_handle handle = ACPI_HANDLE(&dev->dev);
+
+       /* Find out ACPI_HANDLE if not available in the device context */
+       if (!handle) {
+               handle = acpi_device_handle(acpi_pci_find_companion(&dev->dev));
+               if (!handle)
+                       return -ENOTTY;
+       }
+
+       if (!acpi_has_method(handle, "_RST"))
+               return -ENOTTY;
+
+       if (probe)
+               return 0;
+
+       if (ACPI_FAILURE(acpi_evaluate_object(handle, "_RST", NULL, NULL))) {
+               pci_warn(dev, "ACPI _RST failed\n");
+               return -EINVAL;
+       }
+
+       return 0;
+}
+
 static bool acpi_pci_bridge_d3(struct pci_dev *dev)
 {
        const struct fwnode_handle *fwnode;
diff --git a/drivers/pci/pci.c b/drivers/pci/pci.c
index bbed852d977f1..5726d120b70a2 100644
--- a/drivers/pci/pci.c
+++ b/drivers/pci/pci.c
@@ -5122,6 +5122,7 @@ static void pci_dev_restore(struct pci_dev *dev)
  */
 const struct pci_reset_fn_method pci_reset_fn_methods[] = {
        { &pci_dev_specific_reset, .name = "device_specific" },
+       { &pci_dev_acpi_reset, .name = "acpi" },
        { &pcie_reset_flr, .name = "flr" },
        { &pci_af_flr, .name = "af_flr" },
        { &pci_pm_reset, .name = "pm" },
diff --git a/drivers/pci/pci.h b/drivers/pci/pci.h
index 13ec6bd6f4f76..f3974ed1a99c2 100644
--- a/drivers/pci/pci.h
+++ b/drivers/pci/pci.h
@@ -703,7 +703,13 @@ static inline int pci_aer_raw_clear_status(struct pci_dev *dev) { return -EINVAL
 #ifdef CONFIG_ACPI
 int pci_acpi_program_hp_params(struct pci_dev *dev);
 extern const struct attribute_group pci_dev_acpi_attr_group;
+int pci_dev_acpi_reset(struct pci_dev *dev, int probe);
 #else
+static inline int pci_dev_acpi_reset(struct pci_dev *dev, int probe)
+{
+       return -ENOTTY;
+}
+
 static inline int pci_acpi_program_hp_params(struct pci_dev *dev)
 {
        return -ENODEV;
diff --git a/include/linux/pci.h b/include/linux/pci.h
index 6e9bc4f9cdab4..a7f063da2fe5f 100644
--- a/include/linux/pci.h
+++ b/include/linux/pci.h
@@ -49,7 +49,7 @@
                               PCI_STATUS_SIG_TARGET_ABORT | \
                               PCI_STATUS_PARITY)

-#define PCI_RESET_METHODS_NUM 5
+#define PCI_RESET_METHODS_NUM 6




  parent reply	other threads:[~2021-06-06  5:00 UTC|newest]

Thread overview: 20+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-05-29 19:25 [PATCH v5 0/7] Expose and manage PCI device reset Amey Narkhede
2021-05-29 19:25 ` [PATCH v5 1/7] PCI: Add pcie_reset_flr to follow calling convention of other reset methods Amey Narkhede
2021-05-29 19:25 ` [PATCH v5 2/7] PCI: Add new array for keeping track of ordering of " Amey Narkhede
2021-06-02 21:37   ` Shanker R Donthineni
2021-06-05 20:55   ` Bjorn Helgaas
2021-05-29 19:25 ` [PATCH v5 3/7] PCI: Remove reset_fn field from pci_dev Amey Narkhede
2021-06-02 22:16   ` Shanker R Donthineni
2021-06-05 20:57   ` Bjorn Helgaas
2021-05-29 19:25 ` [PATCH v5 4/7] PCI/sysfs: Allow userspace to query and set device reset mechanism Amey Narkhede
2021-06-02 22:17   ` Shanker R Donthineni
2021-06-06 12:58   ` Krzysztof Wilczyński
2021-06-06 14:27     ` Shanker R Donthineni
2021-05-29 19:25 ` [PATCH v5 5/7] PCI: Add support for a function level reset based on _RST method Amey Narkhede
2021-06-05 20:53   ` Bjorn Helgaas
2021-06-06  0:04     ` Shanker R Donthineni
2021-06-06  5:00     ` Shanker R Donthineni [this message]
2021-06-09 12:51     ` Shanker R Donthineni
2021-05-29 19:25 ` [PATCH v5 6/7] PCI: Enable NO_BUS_RESET quirk for Nvidia GPUs Amey Narkhede
2021-05-29 19:25 ` [PATCH v5 7/7] PCI: Change the type of probe argument in reset functions Amey Narkhede
2021-06-02 22:15   ` Shanker R Donthineni

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=b24be059-4f90-f1b5-b49c-ba7076ace9ba@nvidia.com \
    --to=sdonthineni@nvidia.com \
    --cc=alex.williamson@redhat.com \
    --cc=ameynarkhede03@gmail.com \
    --cc=bhelgaas@google.com \
    --cc=helgaas@kernel.org \
    --cc=kw@linux.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-pci@vger.kernel.org \
    --cc=okaya@kernel.org \
    --cc=raphael.norwitz@nutanix.com \
    --cc=rjw@rjwysocki.net \
    --subject='Re: [PATCH v5 5/7] PCI: Add support for a function level reset based on _RST method' \
    /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

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).