Merge patch series "riscv, mm: detect svnapot cpu support at runtime"

Qinglin Pan <panqinglin00@gmail.com> says:

Svnapot is a RISC-V extension for marking contiguous 4K pages as a non-4K
page. This patch set is for using Svnapot in hugetlb fs and huge vmap.

This patchset adds a Kconfig item for using Svnapot in
"Platform type"->"SVNAPOT extension support". Its default value is on,
and people can set it off if they don't allow kernel to detect Svnapot
hardware support and leverage it.

Tested on:
  - qemu rv64 with "Svnapot support" off and svnapot=true.
  - qemu rv64 with "Svnapot support" on and svnapot=true.
  - qemu rv64 with "Svnapot support" off and svnapot=false.
  - qemu rv64 with "Svnapot support" on and svnapot=false.

* b4-shazam-merge:
  riscv: mm: support Svnapot in huge vmap
  riscv: mm: support Svnapot in hugetlb page
  riscv: mm: modify pte format for Svnapot

Link: https://lore.kernel.org/r/20230209131647.17245-1-panqinglin00@gmail.com
[Palmer: fix up the feature ordering in the merge]
Signed-off-by: Palmer Dabbelt <palmer@rivosinc.com>

arch/riscv/Kconfig

@@ -44,7 +44,7 @@ config RISCV
 	select ARCH_USE_QUEUED_RWLOCKS
 	select ARCH_WANT_DEFAULT_TOPDOWN_MMAP_LAYOUT if MMU
 	select ARCH_WANT_FRAME_POINTERS
-	select ARCH_WANT_GENERAL_HUGETLB
+	select ARCH_WANT_GENERAL_HUGETLB if !RISCV_ISA_SVNAPOT
 	select ARCH_WANT_HUGETLB_PAGE_OPTIMIZE_VMEMMAP
 	select ARCH_WANT_HUGE_PMD_SHARE if 64BIT
 	select ARCH_WANT_LD_ORPHAN_WARN if !XIP_KERNEL
@@ -398,6 +398,25 @@ config RISCV_ISA_C
 
 	  If you don't know what to do here, say Y.
 
+config RISCV_ISA_SVNAPOT
+	bool "SVNAPOT extension support"
+	depends on 64BIT && MMU
+	default y
+	select RISCV_ALTERNATIVE
+	help
+	  Allow kernel to detect the SVNAPOT ISA-extension dynamically at boot
+	  time and enable its usage.
+
+	  The SVNAPOT extension is used to mark contiguous PTEs as a range
+	  of contiguous virtual-to-physical translations for a naturally
+	  aligned power-of-2 (NAPOT) granularity larger than the base 4KB page
+	  size. When HUGETLBFS is also selected this option unconditionally
+	  allocates some memory for each NAPOT page size supported by the kernel.
+	  When optimizing for low memory consumption and for platforms without
+	  the SVNAPOT extension, it may be better to say N here.
+
+	  If you don't know what to do here, say Y.
+
 config RISCV_ISA_SVPBMT
 	bool "SVPBMT extension support"
 	depends on 64BIT && MMU

arch/riscv/include/asm/hugetlb.h

@@ -2,7 +2,6 @@
 #ifndef _ASM_RISCV_HUGETLB_H
 #define _ASM_RISCV_HUGETLB_H
 
-#include <asm-generic/hugetlb.h>
 #include <asm/page.h>
 
 static inline void arch_clear_hugepage_flags(struct page *page)
@@ -11,4 +10,37 @@ static inline void arch_clear_hugepage_flags(struct page *page)
 }
 #define arch_clear_hugepage_flags arch_clear_hugepage_flags
 
+#ifdef CONFIG_RISCV_ISA_SVNAPOT
+#define __HAVE_ARCH_HUGE_PTE_CLEAR
+void huge_pte_clear(struct mm_struct *mm, unsigned long addr,
+		    pte_t *ptep, unsigned long sz);
+
+#define __HAVE_ARCH_HUGE_SET_HUGE_PTE_AT
+void set_huge_pte_at(struct mm_struct *mm,
+		     unsigned long addr, pte_t *ptep, pte_t pte);
+
+#define __HAVE_ARCH_HUGE_PTEP_GET_AND_CLEAR
+pte_t huge_ptep_get_and_clear(struct mm_struct *mm,
+			      unsigned long addr, pte_t *ptep);
+
+#define __HAVE_ARCH_HUGE_PTEP_CLEAR_FLUSH
+pte_t huge_ptep_clear_flush(struct vm_area_struct *vma,
+			    unsigned long addr, pte_t *ptep);
+
+#define __HAVE_ARCH_HUGE_PTEP_SET_WRPROTECT
+void huge_ptep_set_wrprotect(struct mm_struct *mm,
+			     unsigned long addr, pte_t *ptep);
+
+#define __HAVE_ARCH_HUGE_PTEP_SET_ACCESS_FLAGS
+int huge_ptep_set_access_flags(struct vm_area_struct *vma,
+			       unsigned long addr, pte_t *ptep,
+			       pte_t pte, int dirty);
+
+pte_t arch_make_huge_pte(pte_t entry, unsigned int shift, vm_flags_t flags);
+#define arch_make_huge_pte arch_make_huge_pte
+
+#endif /*CONFIG_RISCV_ISA_SVNAPOT*/
+
+#include <asm-generic/hugetlb.h>
+
 #endif /* _ASM_RISCV_HUGETLB_H */

arch/riscv/include/asm/hwcap.h

@@ -42,6 +42,7 @@
 #define RISCV_ISA_EXT_ZBB		30
 #define RISCV_ISA_EXT_ZICBOM		31
 #define RISCV_ISA_EXT_ZIHINTPAUSE	32
+#define RISCV_ISA_EXT_SVNAPOT		33
 
 #define RISCV_ISA_EXT_MAX		64
 #define RISCV_ISA_EXT_NAME_LEN_MAX	32

arch/riscv/include/asm/page.h

@@ -16,11 +16,6 @@
 #define PAGE_SIZE	(_AC(1, UL) << PAGE_SHIFT)
 #define PAGE_MASK	(~(PAGE_SIZE - 1))
 
-#ifdef CONFIG_64BIT
-#define HUGE_MAX_HSTATE		2
-#else
-#define HUGE_MAX_HSTATE		1
-#endif
 #define HPAGE_SHIFT		PMD_SHIFT
 #define HPAGE_SIZE		(_AC(1, UL) << HPAGE_SHIFT)
 #define HPAGE_MASK              (~(HPAGE_SIZE - 1))

arch/riscv/include/asm/pgtable-64.h

@@ -78,6 +78,40 @@ typedef struct {
  */
 #define _PAGE_PFN_MASK  GENMASK(53, 10)
 
+/*
+ * [63] Svnapot definitions:
+ * 0 Svnapot disabled
+ * 1 Svnapot enabled
+ */
+#define _PAGE_NAPOT_SHIFT	63
+#define _PAGE_NAPOT		BIT(_PAGE_NAPOT_SHIFT)
+/*
+ * Only 64KB (order 4) napot ptes supported.
+ */
+#define NAPOT_CONT_ORDER_BASE 4
+enum napot_cont_order {
+	NAPOT_CONT64KB_ORDER = NAPOT_CONT_ORDER_BASE,
+	NAPOT_ORDER_MAX,
+};
+
+#define for_each_napot_order(order)						\
+	for (order = NAPOT_CONT_ORDER_BASE; order < NAPOT_ORDER_MAX; order++)
+#define for_each_napot_order_rev(order)						\
+	for (order = NAPOT_ORDER_MAX - 1;					\
+	     order >= NAPOT_CONT_ORDER_BASE; order--)
+#define napot_cont_order(val)	(__builtin_ctzl((val.pte >> _PAGE_PFN_SHIFT) << 1))
+
+#define napot_cont_shift(order)	((order) + PAGE_SHIFT)
+#define napot_cont_size(order)	BIT(napot_cont_shift(order))
+#define napot_cont_mask(order)	(~(napot_cont_size(order) - 1UL))
+#define napot_pte_num(order)	BIT(order)
+
+#ifdef CONFIG_RISCV_ISA_SVNAPOT
+#define HUGE_MAX_HSTATE		(2 + (NAPOT_ORDER_MAX - NAPOT_CONT_ORDER_BASE))
+#else
+#define HUGE_MAX_HSTATE		2
+#endif
+
 /*
  * [62:61] Svpbmt Memory Type definitions:
  *

arch/riscv/include/asm/pgtable.h

@@ -264,10 +264,47 @@ static inline pte_t pud_pte(pud_t pud)
 	return __pte(pud_val(pud));
 }
 
+#ifdef CONFIG_RISCV_ISA_SVNAPOT
+
+static __always_inline bool has_svnapot(void)
+{
+	return riscv_has_extension_likely(RISCV_ISA_EXT_SVNAPOT);
+}
+
+static inline unsigned long pte_napot(pte_t pte)
+{
+	return pte_val(pte) & _PAGE_NAPOT;
+}
+
+static inline pte_t pte_mknapot(pte_t pte, unsigned int order)
+{
+	int pos = order - 1 + _PAGE_PFN_SHIFT;
+	unsigned long napot_bit = BIT(pos);
+	unsigned long napot_mask = ~GENMASK(pos, _PAGE_PFN_SHIFT);
+
+	return __pte((pte_val(pte) & napot_mask) | napot_bit | _PAGE_NAPOT);
+}
+
+#else
+
+static __always_inline bool has_svnapot(void) { return false; }
+
+static inline unsigned long pte_napot(pte_t pte)
+{
+	return 0;
+}
+
+#endif /* CONFIG_RISCV_ISA_SVNAPOT */
+
 /* Yields the page frame number (PFN) of a page table entry */
 static inline unsigned long pte_pfn(pte_t pte)
 {
-	return __page_val_to_pfn(pte_val(pte));
+	unsigned long res  = __page_val_to_pfn(pte_val(pte));
+
+	if (has_svnapot() && pte_napot(pte))
+		res = res & (res - 1UL);
+
+	return res;
 }
 
 #define pte_page(x)     pfn_to_page(pte_pfn(x))

arch/riscv/include/asm/vmalloc.h

@@ -17,6 +17,65 @@ static inline bool arch_vmap_pmd_supported(pgprot_t prot)
 	return true;
 }
 
-#endif
+#ifdef CONFIG_RISCV_ISA_SVNAPOT
+#include <linux/pgtable.h>
 
+#define arch_vmap_pte_range_map_size arch_vmap_pte_range_map_size
+static inline unsigned long arch_vmap_pte_range_map_size(unsigned long addr, unsigned long end,
+							 u64 pfn, unsigned int max_page_shift)
+{
+	unsigned long map_size = PAGE_SIZE;
+	unsigned long size, order;
+
+	if (!has_svnapot())
+		return map_size;
+
+	for_each_napot_order_rev(order) {
+		if (napot_cont_shift(order) > max_page_shift)
+			continue;
+
+		size = napot_cont_size(order);
+		if (end - addr < size)
+			continue;
+
+		if (!IS_ALIGNED(addr, size))
+			continue;
+
+		if (!IS_ALIGNED(PFN_PHYS(pfn), size))
+			continue;
+
+		map_size = size;
+		break;
+	}
+
+	return map_size;
+}
+
+#define arch_vmap_pte_supported_shift arch_vmap_pte_supported_shift
+static inline int arch_vmap_pte_supported_shift(unsigned long size)
+{
+	int shift = PAGE_SHIFT;
+	unsigned long order;
+
+	if (!has_svnapot())
+		return shift;
+
+	WARN_ON_ONCE(size >= PMD_SIZE);
+
+	for_each_napot_order_rev(order) {
+		if (napot_cont_size(order) > size)
+			continue;
+
+		if (!IS_ALIGNED(size, napot_cont_size(order)))
+			continue;
+
+		shift = napot_cont_shift(order);
+		break;
+	}
+
+	return shift;
+}
+
+#endif /* CONFIG_RISCV_ISA_SVNAPOT */
+#endif /* CONFIG_HAVE_ARCH_HUGE_VMAP */
 #endif /* _ASM_RISCV_VMALLOC_H */

arch/riscv/kernel/cpu.c

@@ -191,6 +191,7 @@ static struct riscv_isa_ext_data isa_ext_arr[] = {
 	__RISCV_ISA_EXT_DATA(sscofpmf, RISCV_ISA_EXT_SSCOFPMF),
 	__RISCV_ISA_EXT_DATA(sstc, RISCV_ISA_EXT_SSTC),
 	__RISCV_ISA_EXT_DATA(svinval, RISCV_ISA_EXT_SVINVAL),
+	__RISCV_ISA_EXT_DATA(svnapot, RISCV_ISA_EXT_SVNAPOT),
 	__RISCV_ISA_EXT_DATA(svpbmt, RISCV_ISA_EXT_SVPBMT),
 	__RISCV_ISA_EXT_DATA("", RISCV_ISA_EXT_MAX),
 };

arch/riscv/kernel/cpufeature.c

@@ -224,6 +224,7 @@ void __init riscv_fill_hwcap(void)
 				SET_ISA_EXT_MAP("sscofpmf", RISCV_ISA_EXT_SSCOFPMF);
 				SET_ISA_EXT_MAP("sstc", RISCV_ISA_EXT_SSTC);
 				SET_ISA_EXT_MAP("svinval", RISCV_ISA_EXT_SVINVAL);
+				SET_ISA_EXT_MAP("svnapot", RISCV_ISA_EXT_SVNAPOT);
 				SET_ISA_EXT_MAP("svpbmt", RISCV_ISA_EXT_SVPBMT);
 				SET_ISA_EXT_MAP("zbb", RISCV_ISA_EXT_ZBB);
 				SET_ISA_EXT_MAP("zicbom", RISCV_ISA_EXT_ZICBOM);

arch/riscv/mm/hugetlbpage.c

@@ -2,6 +2,305 @@
 #include <linux/hugetlb.h>
 #include <linux/err.h>
 
+#ifdef CONFIG_RISCV_ISA_SVNAPOT
+pte_t *huge_pte_alloc(struct mm_struct *mm,
+		      struct vm_area_struct *vma,
+		      unsigned long addr,
+		      unsigned long sz)
+{
+	unsigned long order;
+	pte_t *pte = NULL;
+	pgd_t *pgd;
+	p4d_t *p4d;
+	pud_t *pud;
+	pmd_t *pmd;
+
+	pgd = pgd_offset(mm, addr);
+	p4d = p4d_alloc(mm, pgd, addr);
+	if (!p4d)
+		return NULL;
+
+	pud = pud_alloc(mm, p4d, addr);
+	if (!pud)
+		return NULL;
+
+	if (sz == PUD_SIZE) {
+		pte = (pte_t *)pud;
+		goto out;
+	}
+
+	if (sz == PMD_SIZE) {
+		if (want_pmd_share(vma, addr) && pud_none(*pud))
+			pte = huge_pmd_share(mm, vma, addr, pud);
+		else
+			pte = (pte_t *)pmd_alloc(mm, pud, addr);
+		goto out;
+	}
+
+	pmd = pmd_alloc(mm, pud, addr);
+	if (!pmd)
+		return NULL;
+
+	for_each_napot_order(order) {
+		if (napot_cont_size(order) == sz) {
+			pte = pte_alloc_map(mm, pmd, addr & napot_cont_mask(order));
+			break;
+		}
+	}
+
+out:
+	WARN_ON_ONCE(pte && pte_present(*pte) && !pte_huge(*pte));
+	return pte;
+}
+
+pte_t *huge_pte_offset(struct mm_struct *mm,
+		       unsigned long addr,
+		       unsigned long sz)
+{
+	unsigned long order;
+	pte_t *pte = NULL;
+	pgd_t *pgd;
+	p4d_t *p4d;
+	pud_t *pud;
+	pmd_t *pmd;
+
+	pgd = pgd_offset(mm, addr);
+	if (!pgd_present(*pgd))
+		return NULL;
+
+	p4d = p4d_offset(pgd, addr);
+	if (!p4d_present(*p4d))
+		return NULL;
+
+	pud = pud_offset(p4d, addr);
+	if (sz == PUD_SIZE)
+		/* must be pud huge, non-present or none */
+		return (pte_t *)pud;
+
+	if (!pud_present(*pud))
+		return NULL;
+
+	pmd = pmd_offset(pud, addr);
+	if (sz == PMD_SIZE)
+		/* must be pmd huge, non-present or none */
+		return (pte_t *)pmd;
+
+	if (!pmd_present(*pmd))
+		return NULL;
+
+	for_each_napot_order(order) {
+		if (napot_cont_size(order) == sz) {
+			pte = pte_offset_kernel(pmd, addr & napot_cont_mask(order));
+			break;
+		}
+	}
+	return pte;
+}
+
+static pte_t get_clear_contig(struct mm_struct *mm,
+			      unsigned long addr,
+			      pte_t *ptep,
+			      unsigned long pte_num)
+{
+	pte_t orig_pte = ptep_get(ptep);
+	unsigned long i;
+
+	for (i = 0; i < pte_num; i++, addr += PAGE_SIZE, ptep++) {
+		pte_t pte = ptep_get_and_clear(mm, addr, ptep);
+
+		if (pte_dirty(pte))
+			orig_pte = pte_mkdirty(orig_pte);
+
+		if (pte_young(pte))
+			orig_pte = pte_mkyoung(orig_pte);
+	}
+
+	return orig_pte;
+}
+
+static pte_t get_clear_contig_flush(struct mm_struct *mm,
+				    unsigned long addr,
+				    pte_t *ptep,
+				    unsigned long pte_num)
+{
+	pte_t orig_pte = get_clear_contig(mm, addr, ptep, pte_num);
+	struct vm_area_struct vma = TLB_FLUSH_VMA(mm, 0);
+	bool valid = !pte_none(orig_pte);
+
+	if (valid)
+		flush_tlb_range(&vma, addr, addr + (PAGE_SIZE * pte_num));
+
+	return orig_pte;
+}
+
+pte_t arch_make_huge_pte(pte_t entry, unsigned int shift, vm_flags_t flags)
+{
+	unsigned long order;
+
+	for_each_napot_order(order) {
+		if (shift == napot_cont_shift(order)) {
+			entry = pte_mknapot(entry, order);
+			break;
+		}
+	}
+	if (order == NAPOT_ORDER_MAX)
+		entry = pte_mkhuge(entry);
+
+	return entry;
+}
+
+void set_huge_pte_at(struct mm_struct *mm,
+		     unsigned long addr,
+		     pte_t *ptep,
+		     pte_t pte)
+{
+	int i, pte_num;
+
+	if (!pte_napot(pte)) {
+		set_pte_at(mm, addr, ptep, pte);
+		return;
+	}
+
+	pte_num = napot_pte_num(napot_cont_order(pte));
+	for (i = 0; i < pte_num; i++, ptep++, addr += PAGE_SIZE)
+		set_pte_at(mm, addr, ptep, pte);
+}
+
+int huge_ptep_set_access_flags(struct vm_area_struct *vma,
+			       unsigned long addr,
+			       pte_t *ptep,
+			       pte_t pte,
+			       int dirty)
+{
+	struct mm_struct *mm = vma->vm_mm;
+	unsigned long order;
+	pte_t orig_pte;
+	int i, pte_num;
+
+	if (!pte_napot(pte))
+		return ptep_set_access_flags(vma, addr, ptep, pte, dirty);
+
+	order = napot_cont_order(pte);
+	pte_num = napot_pte_num(order);
+	ptep = huge_pte_offset(mm, addr, napot_cont_size(order));
+	orig_pte = get_clear_contig_flush(mm, addr, ptep, pte_num);
+
+	if (pte_dirty(orig_pte))
+		pte = pte_mkdirty(pte);
+
+	if (pte_young(orig_pte))
+		pte = pte_mkyoung(pte);
+
+	for (i = 0; i < pte_num; i++, addr += PAGE_SIZE, ptep++)
+		set_pte_at(mm, addr, ptep, pte);
+
+	return true;
+}
+
+pte_t huge_ptep_get_and_clear(struct mm_struct *mm,
+			      unsigned long addr,
+			      pte_t *ptep)
+{
+	pte_t orig_pte = ptep_get(ptep);
+	int pte_num;
+
+	if (!pte_napot(orig_pte))
+		return ptep_get_and_clear(mm, addr, ptep);
+
+	pte_num = napot_pte_num(napot_cont_order(orig_pte));
+
+	return get_clear_contig(mm, addr, ptep, pte_num);
+}
+
+void huge_ptep_set_wrprotect(struct mm_struct *mm,
+			     unsigned long addr,
+			     pte_t *ptep)
+{
+	pte_t pte = ptep_get(ptep);
+	unsigned long order;
+	int i, pte_num;
+
+	if (!pte_napot(pte)) {
+		ptep_set_wrprotect(mm, addr, ptep);
+		return;
+	}
+
+	order = napot_cont_order(pte);
+	pte_num = napot_pte_num(order);
+	ptep = huge_pte_offset(mm, addr, napot_cont_size(order));
+
+	for (i = 0; i < pte_num; i++, addr += PAGE_SIZE, ptep++)
+		ptep_set_wrprotect(mm, addr, ptep);
+}
+
+pte_t huge_ptep_clear_flush(struct vm_area_struct *vma,
+			    unsigned long addr,
+			    pte_t *ptep)
+{
+	pte_t pte = ptep_get(ptep);
+	int pte_num;
+
+	if (!pte_napot(pte))
+		return ptep_clear_flush(vma, addr, ptep);
+
+	pte_num = napot_pte_num(napot_cont_order(pte));
+
+	return get_clear_contig_flush(vma->vm_mm, addr, ptep, pte_num);
+}
+
+void huge_pte_clear(struct mm_struct *mm,
+		    unsigned long addr,
+		    pte_t *ptep,
+		    unsigned long sz)
+{
+	pte_t pte = READ_ONCE(*ptep);
+	int i, pte_num;
+
+	if (!pte_napot(pte)) {
+		pte_clear(mm, addr, ptep);
+		return;
+	}
+
+	pte_num = napot_pte_num(napot_cont_order(pte));
+	for (i = 0; i < pte_num; i++, addr += PAGE_SIZE, ptep++)
+		pte_clear(mm, addr, ptep);
+}
+
+static __init bool is_napot_size(unsigned long size)
+{
+	unsigned long order;
+
+	if (!has_svnapot())
+		return false;
+
+	for_each_napot_order(order) {
+		if (size == napot_cont_size(order))
+			return true;
+	}
+	return false;
+}
+
+static __init int napot_hugetlbpages_init(void)
+{
+	if (has_svnapot()) {
+		unsigned long order;
+
+		for_each_napot_order(order)
+			hugetlb_add_hstate(order);
+	}
+	return 0;
+}
+arch_initcall(napot_hugetlbpages_init);
+
+#else
+
+static __init bool is_napot_size(unsigned long size)
+{
+	return false;
+}
+
+#endif /*CONFIG_RISCV_ISA_SVNAPOT*/
+
 int pud_huge(pud_t pud)
 {
 	return pud_leaf(pud);
@@ -18,6 +317,8 @@ bool __init arch_hugetlb_valid_size(unsigned long size)
 		return true;
 	else if (IS_ENABLED(CONFIG_64BIT) && size == PUD_SIZE)
 		return true;
+	else if (is_napot_size(size))
+		return true;
 	else
 		return false;
 }