The quilt patch titled
Subject: ipc/msg: avoid negative value by overflow in msginfo
has been removed from the -mm tree. Its filename was
ipc-msg-mitigate-the-lock-contention-with-percpu-counter-fix.patch
This patch was dropped because it was folded into ipc-msg-mitigate-the-lock-contention-with-percpu-counter.patch
------------------------------------------------------
From: Jiebin Sun <jiebin.sun@xxxxxxxxx>
Subject: ipc/msg: avoid negative value by overflow in msginfo
Date: Tue, 20 Sep 2022 23:08:09 +0800
The 32-bit value in msginfo struct could be negative if we get it from
signed 64-bit. Clamping it to INT_MAX helps to avoid the negative value
by overflow.
Link: https://lkml.kernel.org/r/20220920150809.4014944-1-jiebin.sun@xxxxxxxxx
Signed-off-by: Jiebin Sun <jiebin.sun@xxxxxxxxx>
Reviewed-by: Manfred Spraul <manfred@xxxxxxxxxxxxxxx>
Reviewed-by: Tim Chen <tim.c.chen@xxxxxxxxxxxxxxx>
Signed-off-by: Andrew Morton <akpm@xxxxxxxxxxxxxxxxxxxx>
---
ipc/msg.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
--- a/ipc/msg.c~ipc-msg-mitigate-the-lock-contention-with-percpu-counter-fix
+++ a/ipc/msg.c
@@ -501,8 +501,8 @@ static int msgctl_info(struct ipc_namesp
max_idx = ipc_get_maxidx(&msg_ids(ns));
up_read(&msg_ids(ns).rwsem);
if (cmd == MSG_INFO) {
- msginfo->msgmap = percpu_counter_sum(&ns->percpu_msg_hdrs);
- msginfo->msgtql = percpu_counter_sum(&ns->percpu_msg_bytes);
+ msginfo->msgmap = min(percpu_counter_sum(&ns->percpu_msg_hdrs), INT_MAX);
+ msginfo->msgtql = min(percpu_counter_sum(&ns->percpu_msg_bytes), INT_MAX);
} else {
msginfo->msgmap = MSGMAP;
msginfo->msgpool = MSGPOOL;
_
Patches currently in -mm which might be from jiebin.sun@xxxxxxxxx are
percpu-add-percpu_counter_add_local-and-percpu_counter_sub_local.patch
ipc-msg-mitigate-the-lock-contention-with-percpu-counter.patch
