Date:   Mon,  6 Mar 2023 23:00:04 +0000
In-Reply-To: <20230306230004.1387007-1-jthoughton@google.com>
Mime-Version: 1.0
References: <20230306230004.1387007-1-jthoughton@google.com>
Message-ID: <20230306230004.1387007-3-jthoughton@google.com>
Subject: [PATCH 2/2] mm: rmap: increase COMPOUND_MAPPED to support 512G
 HugeTLB pages
From:   James Houghton <jthoughton@google.com>
To:     Mike Kravetz <mike.kravetz@oracle.com>,
        Hugh Dickins <hughd@google.com>,
        Muchun Song <songmuchun@bytedance.com>,
        Peter Xu <peterx@redhat.com>,
        "Matthew Wilcox (Oracle)" <willy@infradead.org>,
        Andrew Morton <akpm@linux-foundation.org>
Cc:     "Kirill A . Shutemov" <kirill.shutemov@linux.intel.com>,
        David Hildenbrand <david@redhat.com>,
        David Rientjes <rientjes@google.com>,
        Axel Rasmussen <axelrasmussen@google.com>,
        Jiaqi Yan <jiaqiyan@google.com>, linux-mm@kvack.org,
        linux-kernel@vger.kernel.org,
        James Houghton <jthoughton@google.com>
Content-Type: text/plain; charset="UTF-8"
Precedence: bulk

Some architectures may want to support 512G HugeTLB pages in the future.
Adjust COMPOUND_MAPPED to support this page size.

Signed-off-by: James Houghton <jthoughton@google.com>

diff --git a/mm/internal.h b/mm/internal.h
index 400451a4dd49..994381fd5155 100644
--- a/mm/internal.h
+++ b/mm/internal.h
@@ -53,11 +53,11 @@ struct folio_batch;
 void page_writeback_init(void);
 
 /*
- * If a 16GB hugetlb folio were mapped by PTEs of all of its 4kB pages,
- * its nr_pages_mapped would be 0x400000: choose the COMPOUND_MAPPED bit
- * above that range, instead of 2*(PMD_SIZE/PAGE_SIZE).
+ * If a 512G hugetlb folio were mapped by PTEs of all of its 4kB pages,
+ * its nr_pages_mapped would be 2^27 (0x8000000): choose the COMPOUND_MAPPED bit
+ * above that range.
  */
-#define COMPOUND_MAPPED		0x800000
+#define COMPOUND_MAPPED		(1 << 28)
 #define FOLIO_PAGES_MAPPED	(COMPOUND_MAPPED - 1)
 
 /*
-- 
2.40.0.rc0.216.gc4246ad0f0-goog