linux-kernel - [PATCH RESEND v2 2/2] driver core: bus: Correct API bus_rescan

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-Id: <20241022-bus_match_unlikely-v2-2-1a6f8e6839a0@quicinc.com>
Date: Tue, 22 Oct 2024 19:18:02 +0800
From: Zijun Hu <zijun_hu@...oud.com>
To: Greg Kroah-Hartman <gregkh@...uxfoundation.org>, 
 "Rafael J. Wysocki" <rafael@...nel.org>
Cc: Zijun Hu <zijun_hu@...oud.com>, linux-kernel@...r.kernel.org, 
 Zijun Hu <quic_zijuhu@...cinc.com>
Subject: [PATCH RESEND v2 2/2] driver core: bus: Correct API
 bus_rescan_devices() behavior

From: Zijun Hu <quic_zijuhu@...cinc.com>

API bus_rescan_devices() should ideally scan drivers for a bus's
devices as many as possible, but it will really stop scanning for
remaining devices even if a device encounters inconsequential errors
such as -EPROBE_DEFER, -ENODEV and -EBUSY, fixed by ignoring such
inconsequential errors and continue to scan drivers for next device.

By the way, in order to eliminate risk:
 - the API's return value is not changed by recording the first
   error code during scanning which is returned.
 - API device_reprobe()'s existing logic is not changed as well.

Signed-off-by: Zijun Hu <quic_zijuhu@...cinc.com>
---
 drivers/base/bus.c | 38 +++++++++++++++++++++++++-------------
 1 file changed, 25 insertions(+), 13 deletions(-)

diff --git a/drivers/base/bus.c b/drivers/base/bus.c
index 4b5958c5ee7d..fa68acd55bf8 100644
--- a/drivers/base/bus.c
+++ b/drivers/base/bus.c
@@ -58,9 +58,6 @@ static int __must_check bus_rescan_single_device(struct device *dev)
 	return ret;
 }
 
-static int __must_check bus_rescan_devices_helper(struct device *dev,
-						void *data);
-
 /**
  * bus_to_subsys - Turn a struct bus_type into a struct subsys_private
  *
@@ -797,15 +794,23 @@ static int __must_check bus_rescan_devices_helper(struct device *dev,
 						  void *data)
 {
 	int ret = 0;
+	int *first_error = data;
 
-	if (!dev->driver) {
-		if (dev->parent && dev->bus->need_parent_lock)
-			device_lock(dev->parent);
-		ret = device_attach(dev);
-		if (dev->parent && dev->bus->need_parent_lock)
-			device_unlock(dev->parent);
-	}
-	return ret < 0 ? ret : 0;
+	ret = bus_rescan_single_device(dev);
+
+	if (ret >= 0)
+		return 0;
+
+	if (!*first_error)
+		*first_error = ret;
+	/*
+	 * Ignore these inconsequential errors and continue to
+	 * scan drivers for next device.
+	 */
+	if (ret == -EPROBE_DEFER || ret == -ENODEV || ret == -EBUSY)
+		return 0;
+
+	return ret;
 }
 
 /**
@@ -818,7 +823,10 @@ static int __must_check bus_rescan_devices_helper(struct device *dev,
  */
 int bus_rescan_devices(const struct bus_type *bus)
 {
-	return bus_for_each_dev(bus, NULL, NULL, bus_rescan_devices_helper);
+	int err = 0;
+
+	bus_for_each_dev(bus, NULL, &err, bus_rescan_devices_helper);
+	return err;
 }
 EXPORT_SYMBOL_GPL(bus_rescan_devices);
 
@@ -833,9 +841,13 @@ EXPORT_SYMBOL_GPL(bus_rescan_devices);
  */
 int device_reprobe(struct device *dev)
 {
+	int ret;
+
 	if (dev->driver)
 		device_driver_detach(dev);
-	return bus_rescan_devices_helper(dev, NULL);
+
+	ret = bus_rescan_single_device(dev);
+	return ret < 0 ? ret : 0;
 }
 EXPORT_SYMBOL_GPL(device_reprobe);
 

-- 
2.34.1