[PATCH net v4] ipv6: fix a BUG in rt6_get_pcpu_route() under PREEMPT_RT

[Jiayuan Chen <jiayuan.chen@linux.dev>]

On PREEMPT_RT kernels, after rt6_get_pcpu_route() returns NULL, the
current task can be preempted. Another task running on the same CPU
may then execute rt6_make_pcpu_route() and successfully install a
pcpu_rt entry. When the first task resumes execution, its cmpxchg()
in rt6_make_pcpu_route() will fail because rt6i_pcpu is no longer
NULL, triggering the BUG_ON(prev). It's easy to reproduce it by adding
mdelay() after rt6_get_pcpu_route().

Using preempt_disable/enable is not appropriate here because
ip6_rt_pcpu_alloc() may sleep.

Fix this by handling the cmpxchg() failure gracefully on PREEMPT_RT:
free our allocation and return the existing pcpu_rt installed by
another task. The BUG_ON is replaced by WARN_ON_ONCE for non-PREEMPT_RT
kernels where such races should not occur.

Link: https://syzkaller.appspot.com/bug?extid=9b35e9bc0951140d13e6
Fixes: d2d6422f8bd1 ("x86: Allow to enable PREEMPT_RT.")
Reported-by: syzbot+9b35e9bc0951140d13e6@syzkaller.appspotmail.com
Closes: https://lore.kernel.org/all/6918cd88.050a0220.1c914e.0045.GAE@google.com/T/
Signed-off-by: Jiayuan Chen <jiayuan.chen@linux.dev>

---
v2 -> v4: Use correct Fixes tag; Replace BUG_ON by WARN_ON_ONCE.
          https://lore.kernel.org/netdev/20251219025140.77695-1-jiayuan.chen@linux.dev/
v1 -> v2: Drop migrate_{disable, enabled} suggested by Steven and Paolo.
          https://lore.kernel.org/all/20251209124805.379112-1-jiayuan.chen@linux.dev/T/
---
 net/ipv6/route.c | 13 ++++++++++++-
 1 file changed, 12 insertions(+), 1 deletion(-)

diff --git a/net/ipv6/route.c b/net/ipv6/route.c
index aee6a10b112a..a3e051dc66ee 100644
--- a/net/ipv6/route.c
+++ b/net/ipv6/route.c
@@ -1470,7 +1470,18 @@ static struct rt6_info *rt6_make_pcpu_route(struct net *net,
 
 	p = this_cpu_ptr(res->nh->rt6i_pcpu);
 	prev = cmpxchg(p, NULL, pcpu_rt);
-	BUG_ON(prev);
+	if (unlikely(prev)) {
+		/*
+		 * Another task on this CPU already installed a pcpu_rt.
+		 * This can happen on PREEMPT_RT where preemption is possible.
+		 * Free our allocation and return the existing one.
+		 */
+		WARN_ON_ONCE(!IS_ENABLED(CONFIG_PREEMPT_RT));
+
+		dst_dev_put(&pcpu_rt->dst);
+		dst_release(&pcpu_rt->dst);
+		return prev;
+	}
 
 	if (res->f6i->fib6_destroying) {
 		struct fib6_info *from;
-- 
2.43.0