Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 23.128.96.35 as permitted sender) client-ip=23.128.96.35;
From:   Alexandre Ghiti <alexghiti@rivosinc.com>
To:     Paul Walmsley <paul.walmsley@sifive.com>,
        Palmer Dabbelt <palmer@dabbelt.com>,
        Albert Ou <aou@eecs.berkeley.edu>,
        linux-riscv@lists.infradead.org, linux-kernel@vger.kernel.org
Cc:     Alexandre Ghiti <alexghiti@rivosinc.com>
Subject: [PATCH] riscv: Use hugepage mappings for vmemmap
Date:   Thu, 14 Dec 2023 14:29:35 +0100
Message-Id: <20231214132935.212864-1-alexghiti@rivosinc.com>
MIME-Version: 1.0
Content-Transfer-Encoding: 8bit
Precedence: bulk

This will allow better TLB utilization and then should be more performant.

Before:

---[ vmemmap start ]---
0xffff8d8002000000-0xffff8d8012000000    0x000000046ec00000       256M PTE .   ..     ..   D A G . . W R V
---[ vmemmap end ]---

After:

---[ vmemmap start ]---
0xffff8d8002000000-0xffff8d8012000000    0x000000046ec00000       256M PMD .   ..     ..   D A G . . W R V
---[ vmemmap end ]---

Signed-off-by: Alexandre Ghiti <alexghiti@rivosinc.com>
---
 arch/riscv/mm/init.c | 21 ++++++++++++++++++++-
 1 file changed, 20 insertions(+), 1 deletion(-)

diff --git a/arch/riscv/mm/init.c b/arch/riscv/mm/init.c
index 2e011cbddf3a..f636336e6401 100644
--- a/arch/riscv/mm/init.c
+++ b/arch/riscv/mm/init.c
@@ -1383,10 +1383,29 @@ void __init misc_mem_init(void)
 }
 
 #ifdef CONFIG_SPARSEMEM_VMEMMAP
+void __meminit vmemmap_set_pmd(pmd_t *pmd, void *p, int node,
+			       unsigned long addr, unsigned long next)
+{
+	pmd_set_huge(pmd, virt_to_phys(p), PAGE_KERNEL);
+}
+
+int __meminit vmemmap_check_pmd(pmd_t *pmdp, int node,
+				unsigned long addr, unsigned long next)
+{
+	vmemmap_verify((pte_t *)pmdp, node, addr, next);
+	return 1;
+}
+
 int __meminit vmemmap_populate(unsigned long start, unsigned long end, int node,
 			       struct vmem_altmap *altmap)
 {
-	return vmemmap_populate_basepages(start, end, node, NULL);
+	/*
+	 * Note that SPARSEMEM_VMEMMAP is only selected for rv64 and that we
+	 * can't use hugepage mappings for 2-level page table because in case of
+	 * memory hotplug, we are not able to update all the page tables with
+	 * the new PMDs.
+	 */
+	return vmemmap_populate_hugepages(start, end, node, NULL);
 }
 #endif
 
-- 
2.39.2