linux-kernel - [char-misc-next 2/4] mei: pxp: recover from recv fail under memory pressure

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 for Android: free password hash cracker in your pocket

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

Message-ID: <20231011110157.247552-3-tomas.winkler@intel.com>
Date:   Wed, 11 Oct 2023 14:01:55 +0300
From:   Tomas Winkler <tomas.winkler@...el.com>
To:     Greg Kroah-Hartman <gregkh@...uxfoundation.org>
Cc:     Alexander Usyskin <alexander.usyskin@...el.com>,
        Vitaly Lubart <vitaly.lubart@...el.com>,
        linux-kernel@...r.kernel.org,
        Alan Previn <alan.previn.teres.alexis@...el.com>,
        Tomas Winkler <tomas.winkler@...el.com>
Subject: [char-misc-next 2/4] mei: pxp: recover from recv fail under memory pressure

From: Alexander Usyskin <alexander.usyskin@...el.com>

Under memory pressure recv fails due to kmalloc failure,
and if drivers(pxp) retry send/receive, send blocks
indefinitely.
Send without recv leaves the channel in a bad state.

Retry send attempt after small timeout and reset the channel if
the retry failed on kmalloc failure too.

Signed-off-by: Alexander Usyskin <alexander.usyskin@...el.com>
Signed-off-by: Alan Previn <alan.previn.teres.alexis@...el.com>
Signed-off-by: Tomas Winkler <tomas.winkler@...el.com>
---
 drivers/misc/mei/pxp/mei_pxp.c | 25 ++++++++++++++++++++++++-
 1 file changed, 24 insertions(+), 1 deletion(-)

diff --git a/drivers/misc/mei/pxp/mei_pxp.c b/drivers/misc/mei/pxp/mei_pxp.c
index 2dcb9169e404c3bcbbccc7b4..c6cdd6a47308ebcc72f34c38 100644
--- a/drivers/misc/mei/pxp/mei_pxp.c
+++ b/drivers/misc/mei/pxp/mei_pxp.c
@@ -11,6 +11,7 @@
  * negotiation messages to ME FW command payloads and vice versa.
  */
 
+#include <linux/delay.h>
 #include <linux/module.h>
 #include <linux/slab.h>
 #include <linux/mei.h>
@@ -61,16 +62,38 @@ mei_pxp_receive_message(struct device *dev, void *buffer, size_t size)
 {
 	struct mei_cl_device *cldev;
 	ssize_t byte;
+	bool retry = false;
 
 	if (!dev || !buffer)
 		return -EINVAL;
 
 	cldev = to_mei_cl_device(dev);
 
+retry:
 	byte = mei_cldev_recv(cldev, buffer, size);
 	if (byte < 0) {
 		dev_dbg(dev, "mei_cldev_recv failed. %zd\n", byte);
-		return byte;
+		if (byte != -ENOMEM)
+			return byte;
+
+		/* Retry the read when pages are reclaimed */
+		msleep(20);
+		if (!retry) {
+			retry = true;
+			goto retry;
+		} else {
+			dev_warn(dev, "No memory on data receive after retry, trying to reset the channel...\n");
+			byte = mei_cldev_disable(cldev);
+			if (byte < 0)
+				dev_warn(dev, "mei_cldev_disable failed. %zd\n", byte);
+			/*
+			 * Explicitly ignoring disable failure,
+			 * enable may fix the states and succeed
+			 */
+			byte = mei_cldev_enable(cldev);
+			if (byte < 0)
+				dev_err(dev, "mei_cldev_enable failed. %zd\n", byte);
+		}
 	}
 
 	return byte;
-- 
2.41.0