pstore/zone: Use GFP_ATOMIC to allocate zone buffer

commit 99b3b83 upstream.

There is a case found when triggering a panic_on_oom, pstore fails to dump
kmsg. Because psz_kmsg_write_record can't get the new buffer.

Handle this by using GFP_ATOMIC to allocate a buffer at lower watermark.

Signed-off-by: Qiujun Huang <hqjagain@gmail.com>
Fixes: 335426c ("pstore/zone: Provide way to skip "broken" zone for MTD devices")
Cc: WeiXiong Liao <gmpy.liaowx@gmail.com>
Cc: stable@vger.kernel.org
Signed-off-by: Kees Cook <keescook@chromium.org>
Link: https://lore.kernel.org/r/CAJRQjofRCF7wjrYmw3D7zd5QZnwHQq+F8U-mJDJ6NZ4bddYdLA@mail.gmail.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

fs/pstore/zone.c

@@ -761,7 +761,7 @@ static inline int notrace psz_kmsg_write_record(struct psz_context *cxt,
 		/* avoid destroying old data, allocate a new one */
 		len = zone->buffer_size + sizeof(*zone->buffer);
 		zone->oldbuf = zone->buffer;
-		zone->buffer = kzalloc(len, GFP_KERNEL);
+		zone->buffer = kzalloc(len, GFP_ATOMIC);
 		if (!zone->buffer) {
 			zone->buffer = zone->oldbuf;
 			return -ENOMEM;