linux-acpi.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Hans de Goede <hdegoede@redhat.com>
To: "Rafael J . Wysocki" <rjw@rjwysocki.net>, Len Brown <lenb@kernel.org>
Cc: Hans de Goede <hdegoede@redhat.com>, linux-acpi@vger.kernel.org
Subject: [PATCH 6/7] ACPI: scan: Fix battery devices not working with acpi.defer_scan_based_on_dep=1
Date: Sat, 21 Nov 2020 21:30:39 +0100	[thread overview]
Message-ID: <20201121203040.146252-7-hdegoede@redhat.com> (raw)
In-Reply-To: <20201121203040.146252-1-hdegoede@redhat.com>

When battery devices get added during the second step of the now 2 step
scan-process, then acpi_walk_dep_device_list() may have already been
called for some deps during the first step.

In this case acpi_device_dep_initialize() should not add these deps to
the acpi_dep_list; and it should not increase adev->dep_unmet.

Add a check for already registered (and bound to a driver) devices to
acpi_device_dep_initialize(). This fixes battery devices (which honor the
dep_unmet value) not working with acpi.defer_scan_based_on_dep=1.

Signed-off-by: Hans de Goede <hdegoede@redhat.com>
---
 drivers/acpi/scan.c | 29 +++++++++++++++++++++++++++++
 1 file changed, 29 insertions(+)

diff --git a/drivers/acpi/scan.c b/drivers/acpi/scan.c
index 9927036bfe77..44001610f6a4 100644
--- a/drivers/acpi/scan.c
+++ b/drivers/acpi/scan.c
@@ -17,6 +17,8 @@
 #include <linux/platform_data/x86/apple.h>
 #include <linux/pgtable.h>
 
+#include "acpica/accommon.h"
+#include "acpica/acnamesp.h"
 #include "internal.h"
 
 #define _COMPONENT		ACPI_BUS_COMPONENT
@@ -1935,7 +1937,10 @@ static void acpi_device_dep_initialize(struct acpi_device *adev)
 	}
 
 	for (i = 0; i < dep_devices.count; i++) {
+		struct acpi_device *dep_adev = NULL;
+		struct acpi_namespace_node *node;
 		struct acpi_device_info *info;
+		struct device *dep_dev;
 		int skip;
 
 		status = acpi_get_object_info(dep_devices.handles[i], &info);
@@ -1950,6 +1955,30 @@ static void acpi_device_dep_initialize(struct acpi_device *adev)
 		if (skip)
 			continue;
 
+		/*
+		 * Skip devices which already have a driver bound
+		 * and thus are already available for use.
+		 * We only need to do this during the second scan step,
+		 * when acpi_check_defer_add == false.
+		 */
+		if (!acpi_check_defer_add) {
+			/*
+			 * Cannot call acpi_bus_get_device here as we are called
+			 * with ACPI_MTX_NAMESPACE held.
+			 */
+			node = acpi_ns_validate_handle(dep_devices.handles[i]);
+			if (!node)
+				continue; /* Should never happen */
+
+			status = acpi_ns_get_attached_data(node, acpi_scan_drop_device,
+							   (void **)&dep_adev);
+			if (ACPI_SUCCESS(status) && dep_adev) {
+				dep_dev = acpi_get_first_physical_node(dep_adev);
+				if (dep_dev && dep_dev->driver)
+					continue;
+			}
+		}
+
 		dep = kzalloc(sizeof(struct acpi_dep_data), GFP_KERNEL);
 		if (!dep)
 			return;
-- 
2.28.0


  parent reply	other threads:[~2020-11-21 20:31 UTC|newest]

Thread overview: 25+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2020-11-21 20:30 [PATCH 0/7] ACPI: scan: Split root scanning into 2 steps Hans de Goede
2020-11-21 20:30 ` [PATCH 1/7] ACPI: scan: Add an acpi_info_matches_hids() helper Hans de Goede
2020-11-21 20:30 ` [PATCH 2/7] ACPI: scan: Call acpi_get_object_info() from acpi_add_single_object() Hans de Goede
2020-11-21 20:30 ` [PATCH 3/7] ACPI: scan: Add a separate cleanup exit-path to acpi_scan_init() Hans de Goede
2020-11-21 20:30 ` [PATCH 4/7] ACPI: scan: Split root scanning into 2 steps Hans de Goede
2020-11-21 20:30 ` [PATCH 5/7] ACPI: scan: Add support for deferring adding devices to the second scan phase based on the _DEP list Hans de Goede
2020-11-23 12:17   ` Rafael J. Wysocki
2020-11-23 13:30     ` Hans de Goede
2020-11-23 12:41       ` Rafael J. Wysocki
2020-11-23 13:49         ` Hans de Goede
2020-11-21 20:30 ` Hans de Goede [this message]
2020-12-02 13:52   ` [PATCH 6/7] ACPI: scan: Fix battery devices not working with acpi.defer_scan_based_on_dep=1 Rafael J. Wysocki
2020-11-21 20:30 ` [PATCH 7/7] ACPI: scan: Add some HIDs which are never bound on Cherry Trail devices to acpi_ignore_dep_hids Hans de Goede
2020-12-02 13:49 ` [PATCH 0/7] ACPI: scan: Split root scanning into 2 steps Rafael J. Wysocki
2020-12-02 15:51   ` Rafael J. Wysocki
2020-12-02 19:46     ` Rafael J. Wysocki
2020-12-02 19:39   ` Hans de Goede
2020-12-02 19:57     ` Rafael J. Wysocki
2020-12-03  9:53       ` Hans de Goede
2020-12-03 14:27         ` Rafael J. Wysocki
2020-12-05 15:44           ` Rafael J. Wysocki
2020-12-05 17:02             ` Hans de Goede
2020-12-07 17:27               ` Rafael J. Wysocki
2020-12-07 18:15                 ` Hans de Goede
2021-04-29  3:43 ` [PATCH] ACPI: scan: Defer enumeration of devices with _DEP lists youling257

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=20201121203040.146252-7-hdegoede@redhat.com \
    --to=hdegoede@redhat.com \
    --cc=lenb@kernel.org \
    --cc=linux-acpi@vger.kernel.org \
    --cc=rjw@rjwysocki.net \
    /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).