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-next>] [day] [month] [year] [list]
Message-ID: <20220516045058.GA7933@linuxonhyperv3.guj3yctzbm1etfxqx2vob5hsef.xx.internal.cloudapp.net>
Date:   Sun, 15 May 2022 21:50:58 -0700
From:   Shradha Gupta <shradhagupta@...ux.microsoft.com>
To:     linux-hyperv@...r.kernel.org, linux-kernel@...r.kernel.org,
        "K. Y. Srinivasan" <kys@...rosoft.com>,
        Haiyang Zhang <haiyangz@...rosoft.com>,
        Stephen Hemminger <sthemmin@...rosoft.com>,
        Wei Liu <wei.liu@...nel.org>, Dexuan Cui <decui@...rosoft.com>
Subject: [PATCH v2] hv_balloon: Fix balloon_probe() and balloon_remove()
 error handling

Add missing cleanup in balloon_probe() if the call to
balloon_connect_vsp() fails.  Also correctly handle cleanup in
balloon_remove() when dm_state is DM_INIT_ERROR because
balloon_resume() failed.

Signed-off-by: Shradha Gupta <shradhagupta@...rosoft.com>

---

Changes in v2:
* Use a goto instead of inline code to handle the cleanup
   in balloon_probe()
* Add a comment in balloon_remove() explaining the
   cleanup scenario
* Add missing disable of page reporting when resume
   from hibernation fails

---
 drivers/hv/hv_balloon.c | 21 ++++++++++++++++-----
 1 file changed, 16 insertions(+), 5 deletions(-)

diff --git a/drivers/hv/hv_balloon.c b/drivers/hv/hv_balloon.c
index eee7402cfc02..98fcfb516bbc 100644
--- a/drivers/hv/hv_balloon.c
+++ b/drivers/hv/hv_balloon.c
@@ -1842,7 +1842,7 @@ static int balloon_probe(struct hv_device *dev,
 
 	ret = balloon_connect_vsp(dev);
 	if (ret != 0)
-		return ret;
+		goto connect_error;
 
 	enable_page_reporting();
 	dm_device.state = DM_INITIALIZED;
@@ -1861,6 +1861,7 @@ static int balloon_probe(struct hv_device *dev,
 	dm_device.thread  = NULL;
 	disable_page_reporting();
 	vmbus_close(dev->channel);
+connect_error:
 #ifdef CONFIG_MEMORY_HOTPLUG
 	unregister_memory_notifier(&hv_memory_nb);
 	restore_online_page_callback(&hv_online_page);
@@ -1882,12 +1883,21 @@ static int balloon_remove(struct hv_device *dev)
 	cancel_work_sync(&dm->ha_wrk.wrk);
 
 	kthread_stop(dm->thread);
-	disable_page_reporting();
-	vmbus_close(dev->channel);
+
+	/*
+	 * This is to handle the case when balloon_resume()
+	 * call has failed and some cleanup has been done as
+	 * a part of the error handling.
+	 */
+	if (dm_device.state != DM_INIT_ERROR) {
+		disable_page_reporting();
+		vmbus_close(dev->channel);
 #ifdef CONFIG_MEMORY_HOTPLUG
-	unregister_memory_notifier(&hv_memory_nb);
-	restore_online_page_callback(&hv_online_page);
+		unregister_memory_notifier(&hv_memory_nb);
+		restore_online_page_callback(&hv_online_page);
 #endif
+	}
+
 	spin_lock_irqsave(&dm_device.ha_lock, flags);
 	list_for_each_entry_safe(has, tmp, &dm->ha_region_list, list) {
 		list_for_each_entry_safe(gap, tmp_gap, &has->gap_list, list) {
@@ -1948,6 +1958,7 @@ static int balloon_resume(struct hv_device *dev)
 	vmbus_close(dev->channel);
 out:
 	dm_device.state = DM_INIT_ERROR;
+	disable_page_reporting();
 #ifdef CONFIG_MEMORY_HOTPLUG
 	unregister_memory_notifier(&hv_memory_nb);
 	restore_online_page_callback(&hv_online_page);
-- 
2.17.1

Powered by blists - more mailing lists

Powered by Openwall GNU/*/Linux Powered by OpenVZ