From: "Rafael J. Wysocki" <rjw@rjwysocki.net>
To: Linux PCI <linux-pci@vger.kernel.org>
Cc: Linux PM <linux-pm@vger.kernel.org>,
LKML <linux-kernel@vger.kernel.org>,
Bjorn Helgaas <helgaas@kernel.org>,
Mika Westerberg <mika.westerberg@linux.intel.com>
Subject: [PATCH v3 8/9] PCI/PM: Avoid redundant current_state update
Date: Thu, 14 Apr 2022 15:18:15 +0200 [thread overview]
Message-ID: <1844408.CQOukoFCf9@kreacher> (raw)
In-Reply-To: <5838942.lOV4Wx5bFT@kreacher>
From: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
Notice that if pci_power_up() returns success early, without updating
the given device's PCI_PM_CTRL register, because it has verified that
the power state of the device is D0 already at that point, the
pci_update_current_state() invocation in pci_pm_default_resume_early()
is redundant.
Avoid that redundant call by making pci_power_up() return 1 when it
has updated the device's PCI_PM_CTRL register and checking its return
value in pci_pm_default_resume_early().
Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
Reviewed-by: Mika Westerberg <mika.westerberg@linux.intel.com>
---
v2 -> v3:
* Added R-by from Mika.
---
drivers/pci/pci-driver.c | 5 +++--
drivers/pci/pci.c | 10 ++++++++--
2 files changed, 11 insertions(+), 4 deletions(-)
Index: linux-pm/drivers/pci/pci-driver.c
===================================================================
--- linux-pm.orig/drivers/pci/pci-driver.c
+++ linux-pm/drivers/pci/pci-driver.c
@@ -553,8 +553,9 @@ static void pci_pm_default_resume(struct
static void pci_pm_default_resume_early(struct pci_dev *pci_dev)
{
- pci_power_up(pci_dev);
- pci_update_current_state(pci_dev, PCI_D0);
+ if (pci_power_up(pci_dev))
+ pci_update_current_state(pci_dev, PCI_D0);
+
pci_restore_state(pci_dev);
pci_pme_restore(pci_dev);
}
Index: linux-pm/drivers/pci/pci.c
===================================================================
--- linux-pm.orig/drivers/pci/pci.c
+++ linux-pm/drivers/pci/pci.c
@@ -1189,6 +1189,12 @@ static int pci_dev_wait(struct pci_dev *
/**
* pci_power_up - Put the given device into D0
* @dev: PCI device to power up
+ *
+ * Use the platform firmware and the PCI_PM_CTRL register to put @dev into D0.
+ *
+ * Return 0 if invoking the platform firmware was sufficient to put @dev into D0
+ * (and so the PCI_PM_CTRL register was not updated), or 1 if it was necessary to
+ * update the PCI_PM_CTRL register, or -ENODEV if the device was not accessible.
*/
int pci_power_up(struct pci_dev *dev)
{
@@ -1244,7 +1250,7 @@ int pci_power_up(struct pci_dev *dev)
udelay(PCI_PM_D2_DELAY);
dev->current_state = PCI_D0;
- return 0;
+ return 1;
fail:
pci_err(dev, "Unable to change power state to D0, device inaccessible\n");
@@ -1294,7 +1300,7 @@ static int pci_set_full_power_state(stru
int ret;
ret = pci_power_up(dev);
- if (ret)
+ if (ret < 0)
return ret;
if (!dev->pm_cap)
next prev parent reply other threads:[~2022-04-14 13:24 UTC|newest]
Thread overview: 47+ messages / expand[flat|nested] mbox.gz Atom feed top
2022-04-09 13:03 [PATCH v1 0/7] PCI/PM: Improvements related to device transitions into D0 Rafael J. Wysocki
2022-04-09 13:05 ` [PATCH v1 1/7] PCI/PM: Resume subordinate bus in bus type callbacks Rafael J. Wysocki
2022-04-09 13:06 ` [PATCH v1 2/7] PCI/PM: Drop the runtime_d3cold PCI device flag Rafael J. Wysocki
2022-04-09 13:08 ` [PATCH v1 3/7] PCI/PM: Rearrange pci_update_current_state() Rafael J. Wysocki
2022-04-09 13:22 ` [PATCH v1 4/7] PCI/PM: Rework changing power states of PCI devices Rafael J. Wysocki
2022-04-09 13:23 ` [PATCH v1 5/7] PCI/PM: Move pci_set_low_power_state() next to its caller Rafael J. Wysocki
2022-04-09 13:26 ` [PATCH v1 6/7] PCI/PM: Clean up pci_set_low_power_state() Rafael J. Wysocki
2022-04-09 13:28 ` [PATCH v1 7/7] PCI/PM: Rearrange pci_set_power_state() Rafael J. Wysocki
2022-04-11 14:17 ` [PATCH v2 0/9] PCI/PM: Improvements related to device transitions into D0 Rafael J. Wysocki
2022-04-11 14:19 ` [PATCH v2 1/9] PCI/PM: Resume subordinate bus in bus type callbacks Rafael J. Wysocki
2022-04-11 14:20 ` [PATCH v2 2/9] PCI/PM: Drop the runtime_d3cold device flag Rafael J. Wysocki
2022-04-11 14:21 ` [PATCH v2 3/9] PCI/PM: Rearrange pci_update_current_state() Rafael J. Wysocki
2022-04-12 9:42 ` Mika Westerberg
2022-04-12 10:56 ` Rafael J. Wysocki
2022-04-11 14:25 ` [PATCH v2 4/9] PCI/PM: Rework changing power states of PCI devices Rafael J. Wysocki
2022-04-12 10:00 ` Mika Westerberg
2022-04-12 11:31 ` Rafael J. Wysocki
2022-04-12 11:38 ` Mika Westerberg
2022-04-11 14:27 ` [PATCH v2 5/9] PCI/PM: Move pci_set_low_power_state() next to its caller Rafael J. Wysocki
2022-04-11 14:28 ` [PATCH v2 6/9] PCI/PM: Clean up pci_set_low_power_state() Rafael J. Wysocki
2022-04-11 14:30 ` [PATCH v2 7/9] PCI/PM: Rearrange pci_set_power_state() Rafael J. Wysocki
2022-04-11 14:31 ` [PATCH v2 8/9] PCI/PM: Avoid redundant current_state update Rafael J. Wysocki
2022-04-11 14:33 ` [PATCH v2 9/9] PCI/PM: Replace pci_set_power_state() in pci_pm_thaw_noirq() Rafael J. Wysocki
2022-04-11 14:33 ` [PATCH v2 0/9] PCI/PM: Improvements related to device transitions into D0 Rafael J. Wysocki
2022-04-12 10:08 ` Mika Westerberg
2022-04-12 11:34 ` Rafael J. Wysocki
2022-04-14 13:00 ` [PATCH v3 " Rafael J. Wysocki
2022-04-14 13:04 ` [PATCH v3 1/9] PCI/PM: Resume subordinate bus in bus type callbacks Rafael J. Wysocki
2022-04-14 13:04 ` [PATCH v3 2/9] PCI/PM: Drop the runtime_d3cold device flag Rafael J. Wysocki
2022-04-14 13:07 ` [PATCH v3 3/9] PCI/PM: Rearrange pci_update_current_state() Rafael J. Wysocki
2022-04-14 13:11 ` [PATCH v3 4/9] PCI/PM: Rework changing power states of PCI devices Rafael J. Wysocki
2022-05-03 17:59 ` Nathan Chancellor
2022-05-04 12:59 ` Rafael J. Wysocki
2022-05-04 15:54 ` Anders Roxell
2022-05-04 16:36 ` Nathan Chancellor
2022-05-04 18:00 ` Rafael J. Wysocki
2022-05-04 19:35 ` Nathan Chancellor
2022-05-05 11:58 ` Rafael J. Wysocki
2022-05-04 16:54 ` Bjorn Helgaas
2022-05-04 17:15 ` Rafael J. Wysocki
2022-04-14 13:14 ` [PATCH v3 5/9] PCI/PM: Move pci_set_low_power_state() next to its caller Rafael J. Wysocki
2022-04-14 13:15 ` [PATCH v3 6/9] PCI/PM: Clean up pci_set_low_power_state() Rafael J. Wysocki
2022-04-14 13:17 ` [PATCH v3 7/9] PCI/PM: Rearrange pci_set_power_state() Rafael J. Wysocki
2022-04-14 13:18 ` Rafael J. Wysocki [this message]
2022-04-14 13:19 ` [PATCH v3 9/9] PCI/PM: Replace pci_set_power_state() in pci_pm_thaw_noirq() Rafael J. Wysocki
2022-04-20 16:25 ` [PATCH v3 0/9] PCI/PM: Improvements related to device transitions into D0 Bjorn Helgaas
2022-04-20 16:28 ` Rafael J. Wysocki
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=1844408.CQOukoFCf9@kreacher \
--to=rjw@rjwysocki.net \
--cc=helgaas@kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-pci@vger.kernel.org \
--cc=linux-pm@vger.kernel.org \
--cc=mika.westerberg@linux.intel.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).