public inbox for linux-pm@vger.kernel.org
 help / color / mirror / Atom feed
From: "Rafael J. Wysocki" <rjw@rjwysocki.net>
To: Linux ACPI <linux-acpi@vger.kernel.org>
Cc: LKML <linux-kernel@vger.kernel.org>,
	Linux PM <linux-pm@vger.kernel.org>,
	Andy Shevchenko <andriy.shevchenko@linux.intel.com>,
	Mika Westerberg <mika.westerberg@linux.intel.com>,
	Hans de Goede <hdegoede@redhat.com>,
	Sakari Ailus <sakari.ailus@linux.intel.com>
Subject: [PATCH v2 01/16] ACPI: glue: Use acpi_dev_for_each_child()
Date: Mon, 13 Jun 2022 20:05:17 +0200	[thread overview]
Message-ID: <5595342.DvuYhMxLoT@kreacher> (raw)
In-Reply-To: <2653857.mvXUDI8C0e@kreacher>

From: Rafael J. Wysocki <rafael.j.wysocki@intel.com>

Instead of walking the list of children of an ACPI device directly,
use acpi_dev_for_each_child() to carry out an action for all of
the given ACPI device's children.

This will help to eliminate the children list head from struct
acpi_device as it is redundant and it is used in questionable ways
in some places (in particular, locking is needed for walking the
list pointed to it safely, but it is often missing).

Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
Reviewed-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
---

v1 -> v2:
   * Add R-by from Andy.

---
 drivers/acpi/glue.c |  103 +++++++++++++++++++++++++++++++---------------------
 1 file changed, 63 insertions(+), 40 deletions(-)

Index: linux-pm/drivers/acpi/glue.c
===================================================================
--- linux-pm.orig/drivers/acpi/glue.c
+++ linux-pm/drivers/acpi/glue.c
@@ -105,51 +105,74 @@ static int find_child_checks(struct acpi
 	return FIND_CHILD_MAX_SCORE;
 }
 
+struct find_child_walk_data {
+	struct acpi_device *adev;
+	u64 address;
+	int score;
+	bool check_children;
+};
+
+static int check_one_child(struct acpi_device *adev, void *data)
+{
+	struct find_child_walk_data *wd = data;
+	int score;
+
+	if (!adev->pnp.type.bus_address || acpi_device_adr(adev) != wd->address)
+		return 0;
+
+	if (!wd->adev) {
+		/* This is the first matching object.  Save it and continue. */
+		wd->adev = adev;
+		return 0;
+	}
+
+	/*
+	 * There is more than one matching device object with the same _ADR
+	 * value.  That really is unexpected, so we are kind of beyond the scope
+	 * of the spec here.  We have to choose which one to return, though.
+	 *
+	 * First, get the score for the previously found object and terminate
+	 * the walk if it is maximum.
+	*/
+	if (!wd->score) {
+		score = find_child_checks(wd->adev, wd->check_children);
+		if (score == FIND_CHILD_MAX_SCORE)
+			return 1;
+
+		wd->score = score;
+	}
+	/*
+	 * Second, if the object that has just been found has a better score,
+	 * replace the previously found one with it and terminate the walk if
+	 * the new score is maximum.
+	 */
+	score = find_child_checks(adev, wd->check_children);
+	if (score > wd->score) {
+		wd->adev = adev;
+		if (score == FIND_CHILD_MAX_SCORE)
+			return 1;
+
+		wd->score = score;
+	}
+
+	/* Continue, because there may be better matches. */
+	return 0;
+}
+
 struct acpi_device *acpi_find_child_device(struct acpi_device *parent,
 					   u64 address, bool check_children)
 {
-	struct acpi_device *adev, *ret = NULL;
-	int ret_score = 0;
+	struct find_child_walk_data wd = {
+		.address = address,
+		.check_children = check_children,
+		.adev = NULL,
+		.score = 0,
+	};
 
-	if (!parent)
-		return NULL;
+	if (parent)
+		acpi_dev_for_each_child(parent, check_one_child, &wd);
 
-	list_for_each_entry(adev, &parent->children, node) {
-		acpi_bus_address addr = acpi_device_adr(adev);
-		int score;
-
-		if (!adev->pnp.type.bus_address || addr != address)
-			continue;
-
-		if (!ret) {
-			/* This is the first matching object.  Save it. */
-			ret = adev;
-			continue;
-		}
-		/*
-		 * There is more than one matching device object with the same
-		 * _ADR value.  That really is unexpected, so we are kind of
-		 * beyond the scope of the spec here.  We have to choose which
-		 * one to return, though.
-		 *
-		 * First, check if the previously found object is good enough
-		 * and return it if so.  Second, do the same for the object that
-		 * we've just found.
-		 */
-		if (!ret_score) {
-			ret_score = find_child_checks(ret, check_children);
-			if (ret_score == FIND_CHILD_MAX_SCORE)
-				return ret;
-		}
-		score = find_child_checks(adev, check_children);
-		if (score == FIND_CHILD_MAX_SCORE) {
-			return adev;
-		} else if (score > ret_score) {
-			ret = adev;
-			ret_score = score;
-		}
-	}
-	return ret;
+	return wd.adev;
 }
 EXPORT_SYMBOL_GPL(acpi_find_child_device);
 




  reply	other threads:[~2022-06-13 20:07 UTC|newest]

Thread overview: 77+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2022-06-09 13:44 [PATCH v1 00/16] ACPI: Get rid of the list of children in struct acpi_device Rafael J. Wysocki
2022-06-09 13:47 ` [PATCH v1 01/16] ACPI: glue: Use acpi_dev_for_each_child() Rafael J. Wysocki
2022-06-09 13:49 ` [PATCH v1 02/16] ACPI: glue: Introduce acpi_dev_has_children() Rafael J. Wysocki
2022-06-09 13:54 ` [PATCH v1 03/16] ACPI: glue: Introduce acpi_find_child_by_adr() Rafael J. Wysocki
2022-06-09 13:54 ` [PATCH v1 04/16] thunderbolt: ACPI: Use acpi_find_child_by_adr() Rafael J. Wysocki
2022-06-09 15:25   ` Andy Shevchenko
2022-06-09 15:36     ` Rafael J. Wysocki
2022-06-10  6:46   ` Heikki Krogerus
2022-06-10 13:12     ` Rafael J. Wysocki
2022-06-09 13:56 ` [PATCH v1 05/16] USB: " Rafael J. Wysocki
2022-06-09 15:27   ` Andy Shevchenko
2022-06-09 15:37     ` Rafael J. Wysocki
2022-06-10  6:47   ` Heikki Krogerus
2022-06-10 13:14     ` Rafael J. Wysocki
2022-06-09 13:58 ` [PATCH v1 06/16] ACPI: container: Use acpi_dev_for_each_child() Rafael J. Wysocki
2022-06-09 15:29   ` Andy Shevchenko
2022-06-09 15:58     ` Rafael J. Wysocki
2022-06-09 13:59 ` [PATCH v1 07/16] ACPI: property: Use acpi_dev_for_each_child() for child lookup Rafael J. Wysocki
2022-06-09 14:02 ` [PATCH v1 08/16] ACPI: bus: Export acpi_dev_for_each_child() to modules Rafael J. Wysocki
2022-06-09 14:03 ` [PATCH v1 09/16] ACPI: video: Use acpi_dev_for_each_child() Rafael J. Wysocki
2022-06-09 15:40   ` Andy Shevchenko
2022-06-09 14:06 ` [PATCH v1 10/16] ACPI: bus: Introduce acpi_dev_for_each_child_reverse() Rafael J. Wysocki
2022-06-09 15:40   ` Andy Shevchenko
2022-06-09 14:07 ` [PATCH v1 11/16] ACPI: scan: Walk ACPI device's children using driver core Rafael J. Wysocki
2022-06-09 14:09 ` [PATCH v1 12/16] platform/x86/thinkpad_acpi: Use acpi_dev_for_each_child() Rafael J. Wysocki
2022-06-09 15:48   ` Andy Shevchenko
2022-06-09 15:56     ` Rafael J. Wysocki
2022-06-09 14:12 ` [PATCH v1 13/16] mfd: core: " Rafael J. Wysocki
2022-06-09 14:16 ` [PATCH v1 14/16] soundwire: " Rafael J. Wysocki
2022-06-09 15:22   ` Pierre-Louis Bossart
2022-06-09 16:13     ` Rafael J. Wysocki
2022-06-09 16:21       ` Pierre-Louis Bossart
2022-06-09 17:35         ` Rafael J. Wysocki
2022-06-09 19:08           ` Pierre-Louis Bossart
2022-06-09 14:18 ` [PATCH v1 15/16] ACPI / MMC: PM: Unify fixing up device power Rafael J. Wysocki
2022-06-09 15:33   ` Adrian Hunter
2022-06-10 12:16   ` Ulf Hansson
2022-06-09 14:19 ` [PATCH v1 16/16] ACPI: bus: Drop unused list heads from struct acpi_device Rafael J. Wysocki
2022-06-09 15:12 ` [PATCH v1 00/16] ACPI: Get rid of the list of children in " Andy Shevchenko
2022-06-09 20:24   ` Frank Rowand
2022-06-09 15:56 ` Andy Shevchenko
2022-06-09 15:59   ` Rafael J. Wysocki
2022-06-13 18:03 ` [PATCH v2 " Rafael J. Wysocki
2022-06-13 18:05   ` Rafael J. Wysocki [this message]
2022-06-13 18:06   ` [PATCH v2 02/16] ACPI: glue: Introduce acpi_dev_has_children() Rafael J. Wysocki
2022-06-13 18:10   ` [PATCH v2 03/16] ACPI: glue: Introduce acpi_find_child_by_adr() Rafael J. Wysocki
2022-06-13 18:11   ` [PATCH v2 04/16] thunderbolt: ACPI: Replace tb_acpi_find_port() with acpi_find_child_by_adr() Rafael J. Wysocki
2022-06-13 18:55     ` Andy Shevchenko
2022-06-14  6:07     ` Mika Westerberg
2022-06-14 18:25       ` Rafael J. Wysocki
2022-06-15  6:27         ` Mika Westerberg
2022-06-15 19:52           ` Rafael J. Wysocki
2022-06-14  7:36     ` Heikki Krogerus
2022-06-13 18:15   ` [PATCH v2 06/16] ACPI: container: Use acpi_dev_for_each_child() Rafael J. Wysocki
2022-06-13 18:16   ` [PATCH v2 07/16] ACPI: property: Use acpi_dev_for_each_child() for child lookup Rafael J. Wysocki
2022-06-13 18:26   ` [PATCH v2 08/16] ACPI: bus: Export acpi_dev_for_each_child() to modules Rafael J. Wysocki
2022-06-13 18:26   ` [PATCH v2 09/16] ACPI: video: Use acpi_dev_for_each_child() Rafael J. Wysocki
2022-06-13 18:26   ` [PATCH v2 10/16] ACPI: bus: Introduce acpi_dev_for_each_child_reverse() Rafael J. Wysocki
2022-06-13 18:27   ` [PATCH v2 11/16] ACPI: scan: Walk ACPI device's children using driver core Rafael J. Wysocki
2022-06-13 18:30   ` [PATCH v2 12/16] platform/x86/thinkpad_acpi: Use acpi_dev_for_each_child() Rafael J. Wysocki
2022-06-13 18:54     ` Andy Shevchenko
2022-06-13 20:50     ` Hans de Goede
2022-06-13 18:31   ` [PATCH v2 13/16] mfd: core: " Rafael J. Wysocki
2022-06-15 22:39     ` Lee Jones
2022-06-16 17:31       ` Rafael J. Wysocki
2022-06-27 11:38     ` [GIT PULL] Immutable branch between MFD and ACPI due for the v5.20 merge window Lee Jones
2022-06-27 12:15       ` Rafael J. Wysocki
2022-06-13 18:35   ` [PATCH v2 14/16] soundwire: Use acpi_dev_for_each_child() Rafael J. Wysocki
2022-06-23  8:10     ` Vinod Koul
2022-06-23 12:29       ` Rafael J. Wysocki
2022-06-23 12:41         ` Vinod Koul
2022-06-23 13:26           ` Rafael J. Wysocki
2022-06-13 18:36   ` [PATCH v2 15/16] ACPI / MMC: PM: Unify fixing up device power Rafael J. Wysocki
2022-06-13 18:38   ` [PATCH v2 16/16] ACPI: bus: Drop unused list heads from struct acpi_device Rafael J. Wysocki
2022-06-13 18:39   ` [PATCH v2 05/16] USB: ACPI: Replace usb_acpi_find_port() with acpi_find_child_by_adr() Rafael J. Wysocki
2022-06-13 18:53     ` Andy Shevchenko
2022-06-14  7:37     ` Heikki Krogerus

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=5595342.DvuYhMxLoT@kreacher \
    --to=rjw@rjwysocki.net \
    --cc=andriy.shevchenko@linux.intel.com \
    --cc=hdegoede@redhat.com \
    --cc=linux-acpi@vger.kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-pm@vger.kernel.org \
    --cc=mika.westerberg@linux.intel.com \
    --cc=sakari.ailus@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