riscv: Enable HAVE_ARCH_HUGE_VMAP for 64BIT

This sets the HAVE_ARCH_HUGE_VMAP option, and defines the required page
table functions. With this feature, ioremap area will be mapped with
huge page granularity according to its actual size. This feature can be
disabled by kernel parameter "nohugeiomap".

Signed-off-by: Liu Shixin <liushixin2@huawei.com>
Reviewed-by: Björn Töpel <bjorn@kernel.org>
Tested-by: Björn Töpel <bjorn@kernel.org>
Link: https://lore.kernel.org/r/20221012120038.1034354-2-liushixin2@huawei.com
[Palmer: minor formatting]
Signed-off-by: Palmer Dabbelt <palmer@rivosinc.com>

5 files changed, 104 insertions, 1 deletions

diff --git a/Documentation/features/vm/huge-vmap/arch-support.txt b/Documentation/features/vm/huge-vmap/arch-support.txt
index 13b4940e0c3a..7274a4b15bcc 100644
--- a/Documentation/features/vm/huge-vmap/arch-support.txt
+++ b/Documentation/features/vm/huge-vmap/arch-support.txt
@@ -21,7 +21,7 @@
     |    openrisc: | TODO |
     |      parisc: | TODO |
     |     powerpc: |  ok  |
-    |       riscv: | TODO |
+    |       riscv: |  ok  |
     |        s390: | TODO |
     |          sh: | TODO |
     |       sparc: | TODO |
diff --git a/arch/riscv/Kconfig b/arch/riscv/Kconfig
index 6b48a3ae9843..db2082dd456d 100644
--- a/arch/riscv/Kconfig
+++ b/arch/riscv/Kconfig
@@ -72,6 +72,7 @@ config RISCV
 	select GENERIC_VDSO_TIME_NS if HAVE_GENERIC_VDSO
 	select HARDIRQS_SW_RESEND
 	select HAVE_ARCH_AUDITSYSCALL
+	select HAVE_ARCH_HUGE_VMAP if MMU && 64BIT && !XIP_KERNEL
 	select HAVE_ARCH_JUMP_LABEL if !XIP_KERNEL
 	select HAVE_ARCH_JUMP_LABEL_RELATIVE if !XIP_KERNEL
 	select HAVE_ARCH_KASAN if MMU && 64BIT
diff --git a/arch/riscv/include/asm/vmalloc.h b/arch/riscv/include/asm/vmalloc.h
index ff9abc00d139..48da5371f1e9 100644
--- a/arch/riscv/include/asm/vmalloc.h
+++ b/arch/riscv/include/asm/vmalloc.h
@@ -1,4 +1,22 @@
 #ifndef _ASM_RISCV_VMALLOC_H
 #define _ASM_RISCV_VMALLOC_H
 
+#ifdef CONFIG_HAVE_ARCH_HUGE_VMAP
+
+#define IOREMAP_MAX_ORDER (PUD_SHIFT)
+
+#define arch_vmap_pud_supported arch_vmap_pud_supported
+static inline bool arch_vmap_pud_supported(pgprot_t prot)
+{
+	return true;
+}
+
+#define arch_vmap_pmd_supported arch_vmap_pmd_supported
+static inline bool arch_vmap_pmd_supported(pgprot_t prot)
+{
+	return true;
+}
+
+#endif
+
 #endif /* _ASM_RISCV_VMALLOC_H */
diff --git a/arch/riscv/mm/Makefile b/arch/riscv/mm/Makefile
index d76aabf4b94d..ce7f121ad2dc 100644
--- a/arch/riscv/mm/Makefile
+++ b/arch/riscv/mm/Makefile
@@ -13,6 +13,7 @@ obj-y += extable.o
 obj-$(CONFIG_MMU) += fault.o pageattr.o
 obj-y += cacheflush.o
 obj-y += context.o
+obj-y += pgtable.o
 
 ifeq ($(CONFIG_MMU),y)
 obj-$(CONFIG_SMP) += tlbflush.o
diff --git a/arch/riscv/mm/pgtable.c b/arch/riscv/mm/pgtable.c
new file mode 100644
index 000000000000..6645ead1a7c1
--- /dev/null
+++ b/arch/riscv/mm/pgtable.c
@@ -0,0 +1,83 @@
+// SPDX-License-Identifier: GPL-2.0
+
+#include <asm/pgalloc.h>
+#include <linux/gfp.h>
+#include <linux/kernel.h>
+#include <linux/pgtable.h>
+
+#ifdef CONFIG_HAVE_ARCH_HUGE_VMAP
+int p4d_set_huge(p4d_t *p4d, phys_addr_t addr, pgprot_t prot)
+{
+	return 0;
+}
+
+void p4d_clear_huge(p4d_t *p4d)
+{
+}
+
+int pud_set_huge(pud_t *pud, phys_addr_t phys, pgprot_t prot)
+{
+	pud_t new_pud = pfn_pud(__phys_to_pfn(phys), prot);
+
+	set_pud(pud, new_pud);
+	return 1;
+}
+
+int pud_clear_huge(pud_t *pud)
+{
+	if (!pud_leaf(READ_ONCE(*pud)))
+		return 0;
+	pud_clear(pud);
+	return 1;
+}
+
+int pud_free_pmd_page(pud_t *pud, unsigned long addr)
+{
+	pmd_t *pmd = pud_pgtable(*pud);
+	int i;
+
+	pud_clear(pud);
+
+	flush_tlb_kernel_range(addr, addr + PUD_SIZE);
+
+	for (i = 0; i < PTRS_PER_PMD; i++) {
+		if (!pmd_none(pmd[i])) {
+			pte_t *pte = (pte_t *)pmd_page_vaddr(pmd[i]);
+
+			pte_free_kernel(NULL, pte);
+		}
+	}
+
+	pmd_free(NULL, pmd);
+
+	return 1;
+}
+
+int pmd_set_huge(pmd_t *pmd, phys_addr_t phys, pgprot_t prot)
+{
+	pmd_t new_pmd = pfn_pmd(__phys_to_pfn(phys), prot);
+
+	set_pmd(pmd, new_pmd);
+	return 1;
+}
+
+int pmd_clear_huge(pmd_t *pmd)
+{
+	if (!pmd_leaf(READ_ONCE(*pmd)))
+		return 0;
+	pmd_clear(pmd);
+	return 1;
+}
+
+int pmd_free_pte_page(pmd_t *pmd, unsigned long addr)
+{
+	pte_t *pte = (pte_t *)pmd_page_vaddr(*pmd);
+
+	pmd_clear(pmd);
+
+	flush_tlb_kernel_range(addr, addr + PMD_SIZE);
+	pte_free_kernel(NULL, pte);
+	return 1;
+}
+
+#endif /* CONFIG_HAVE_ARCH_HUGE_VMAP */