diff mbox series

[v8,1/3] riscv: mm: modify pte format for Svnapot

Message ID 20221128022719.328770-2-panqinglin2020@iscas.ac.cn (mailing list archive)
State Superseded
Headers show
Series riscv, mm: detect svnapot cpu support at runtime | expand

Checks

Context Check Description
conchuod/patch_count success Link
conchuod/cover_letter success Series has a cover letter
conchuod/tree_selection success Guessed tree name to be for-next
conchuod/fixes_present success Fixes tag not required for -next series
conchuod/verify_signedoff success Signed-off-by tag matches author and committer
conchuod/kdoc success Errors and warnings before: 0 this patch: 0
conchuod/module_param success Was 0 now: 0
conchuod/build_rv32_defconfig success Build OK
conchuod/build_warn_rv64 success Errors and warnings before: 0 this patch: 0
conchuod/dtb_warn_rv64 success Errors and warnings before: 0 this patch: 0
conchuod/header_inline success No static functions without inline keyword in header files
conchuod/checkpatch success total: 0 errors, 0 warnings, 0 checks, 165 lines checked
conchuod/source_inline success Was 0 now: 0
conchuod/build_rv64_nommu_k210_defconfig success Build OK
conchuod/verify_fixes success No Fixes tag
conchuod/build_rv64_nommu_virt_defconfig success Build OK

Commit Message

Qinglin Pan Nov. 28, 2022, 2:27 a.m. UTC 
From: Qinglin Pan <panqinglin2020@iscas.ac.cn>

Add one static key to enable/disable svnapot support, enable this static
key when "svnapot" is in the "riscv,isa" field of fdt and SVNAPOT compile
option is set. It will influence the behavior of has_svnapot. All code
dependent on svnapot should make sure that has_svnapot return true firstly.

Modify PTE definition for Svnapot, and creates some functions in pgtable.h
to mark a PTE as napot and check if it is a Svnapot PTE. Until now, only
64KB napot size is supported in spec, so some macros has only 64KB version.

Signed-off-by: Qinglin Pan <panqinglin2020@iscas.ac.cn>

Comments

Andrew Jones Nov. 29, 2022, 9:49 a.m. UTC | #1 
On Mon, Nov 28, 2022 at 10:27:17AM +0800, panqinglin2020@iscas.ac.cn wrote:
> From: Qinglin Pan <panqinglin2020@iscas.ac.cn>
> 
> Add one static key to enable/disable svnapot support, enable this static
> key when "svnapot" is in the "riscv,isa" field of fdt and SVNAPOT compile
> option is set. It will influence the behavior of has_svnapot. All code
> dependent on svnapot should make sure that has_svnapot return true firstly.
> 
> Modify PTE definition for Svnapot, and creates some functions in pgtable.h
> to mark a PTE as napot and check if it is a Svnapot PTE. Until now, only
> 64KB napot size is supported in spec, so some macros has only 64KB version.
> 
> Signed-off-by: Qinglin Pan <panqinglin2020@iscas.ac.cn>
> 
> diff --git a/arch/riscv/Kconfig b/arch/riscv/Kconfig
> index 3b41165a8b10..1671938f2f81 100644
> --- a/arch/riscv/Kconfig
> +++ b/arch/riscv/Kconfig
> @@ -394,6 +394,20 @@ 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
> +	select RISCV_ALTERNATIVE
> +	default y
> +	help
> +	  Allow kernel to detect SVNAPOT ISA-extension dynamically in boot time
> +	  and enable its usage.
> +
> +	  SVNAPOT extension helps to mark contiguous PTEs as a range
> +	  of contiguous virtual-to-physical translations, with a naturally
> +	  aligned power-of-2 (NAPOT) granularity larger than the base 4KB page
> +	  size.
> +
>  config RISCV_ISA_SVPBMT
>  	bool "SVPBMT extension support"
>  	depends on 64BIT && MMU
> diff --git a/arch/riscv/include/asm/hwcap.h b/arch/riscv/include/asm/hwcap.h
> index b22525290073..15cda8f131aa 100644
> --- a/arch/riscv/include/asm/hwcap.h
> +++ b/arch/riscv/include/asm/hwcap.h
> @@ -54,6 +54,7 @@ extern unsigned long elf_hwcap;
>   */
>  enum riscv_isa_ext_id {
>  	RISCV_ISA_EXT_SSCOFPMF = RISCV_ISA_EXT_BASE,
> +	RISCV_ISA_EXT_SVNAPOT,
>  	RISCV_ISA_EXT_SVPBMT,
>  	RISCV_ISA_EXT_ZICBOM,
>  	RISCV_ISA_EXT_ZIHINTPAUSE,
> @@ -69,6 +70,7 @@ enum riscv_isa_ext_id {
>   */
>  enum riscv_isa_ext_key {
>  	RISCV_ISA_EXT_KEY_FPU,		/* For 'F' and 'D' */
> +	RISCV_ISA_EXT_KEY_SVNAPOT,
>  	RISCV_ISA_EXT_KEY_ZIHINTPAUSE,
>  	RISCV_ISA_EXT_KEY_SVINVAL,
>  	RISCV_ISA_EXT_KEY_MAX,
> @@ -90,6 +92,8 @@ static __always_inline int riscv_isa_ext2key(int num)
>  		return RISCV_ISA_EXT_KEY_FPU;
>  	case RISCV_ISA_EXT_d:
>  		return RISCV_ISA_EXT_KEY_FPU;
> +	case RISCV_ISA_EXT_SVNAPOT:
> +		return RISCV_ISA_EXT_KEY_SVNAPOT;
>  	case RISCV_ISA_EXT_ZIHINTPAUSE:
>  		return RISCV_ISA_EXT_KEY_ZIHINTPAUSE;
>  	case RISCV_ISA_EXT_SVINVAL:
> diff --git a/arch/riscv/include/asm/page.h b/arch/riscv/include/asm/page.h
> index ac70b0fd9a9a..349fad5e35de 100644
> --- a/arch/riscv/include/asm/page.h
> +++ b/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))
> diff --git a/arch/riscv/include/asm/pgtable-64.h b/arch/riscv/include/asm/pgtable-64.h
> index dc42375c2357..598958cbda50 100644
> --- a/arch/riscv/include/asm/pgtable-64.h
> +++ b/arch/riscv/include/asm/pgtable-64.h
> @@ -74,6 +74,40 @@ 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:
>   *
> diff --git a/arch/riscv/include/asm/pgtable.h b/arch/riscv/include/asm/pgtable.h
> index c61ae83aadee..af6174e3fd97 100644
> --- a/arch/riscv/include/asm/pgtable.h
> +++ b/arch/riscv/include/asm/pgtable.h
> @@ -6,10 +6,12 @@
>  #ifndef _ASM_RISCV_PGTABLE_H
>  #define _ASM_RISCV_PGTABLE_H
>  
> +#include <linux/jump_label.h>
>  #include <linux/mmzone.h>
>  #include <linux/sizes.h>
>  
>  #include <asm/pgtable-bits.h>
> +#include <asm/hwcap.h>
>  
>  #ifndef CONFIG_MMU
>  #define KERNEL_LINK_ADDR	PAGE_OFFSET
> @@ -264,10 +266,45 @@ static inline pte_t pud_pte(pud_t pud)
>  	return __pte(pud_val(pud));
>  }
>  
> +static __always_inline bool has_svnapot(void)
> +{
> +	return static_branch_likely(&riscv_isa_ext_keys[RISCV_ISA_EXT_KEY_SVNAPOT]);

I'm not sure if this should be likely or unlikely.

> +}
> +
> +#ifdef CONFIG_RISCV_ISA_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 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))

We've been burned with static branches inside heavily used inline
functions before (see [1]). There's a series[2] that was meant to
help with this. I haven't seen a refresh of that though.

[1] https://lore.kernel.org/linux-riscv/20220922060958.44203-1-samuel@sholland.org/
[2] https://lore.kernel.org/all/20221006070818.3616-1-jszhang@kernel.org/

> +		res = res & (res - 1UL);
> +
> +	return res;
>  }
>  
>  #define pte_page(x)     pfn_to_page(pte_pfn(x))
> diff --git a/arch/riscv/kernel/cpu.c b/arch/riscv/kernel/cpu.c
> index bf9dd6764bad..88495f5fcafd 100644
> --- a/arch/riscv/kernel/cpu.c
> +++ b/arch/riscv/kernel/cpu.c
> @@ -165,6 +165,7 @@ 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(zicbom, RISCV_ISA_EXT_ZICBOM),
>  	__RISCV_ISA_EXT_DATA(zihintpause, RISCV_ISA_EXT_ZIHINTPAUSE),
> diff --git a/arch/riscv/kernel/cpufeature.c b/arch/riscv/kernel/cpufeature.c
> index 694267d1fe81..ad12fb5363c3 100644
> --- a/arch/riscv/kernel/cpufeature.c
> +++ b/arch/riscv/kernel/cpufeature.c
> @@ -205,6 +205,7 @@ void __init riscv_fill_hwcap(void)
>  				SET_ISA_EXT_MAP("zihintpause", RISCV_ISA_EXT_ZIHINTPAUSE);
>  				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);
>  			}
>  #undef SET_ISA_EXT_MAP
>  		}
> -- 
> 2.37.4
>

Besides the static branch questions, this looks good to me

Reviewed-by: Andrew Jones <ajones@ventanamicro.com>

Thanks,
drew
Andrew Jones Nov. 29, 2022, 2:37 p.m. UTC | #2 
On Mon, Nov 28, 2022 at 10:27:17AM +0800, panqinglin2020@iscas.ac.cn wrote:
> From: Qinglin Pan <panqinglin2020@iscas.ac.cn>
> 
> Add one static key to enable/disable svnapot support, enable this static
> key when "svnapot" is in the "riscv,isa" field of fdt and SVNAPOT compile
> option is set. It will influence the behavior of has_svnapot. All code
> dependent on svnapot should make sure that has_svnapot return true firstly.
> 
> Modify PTE definition for Svnapot, and creates some functions in pgtable.h
> to mark a PTE as napot and check if it is a Svnapot PTE. Until now, only
> 64KB napot size is supported in spec, so some macros has only 64KB version.
> 
> Signed-off-by: Qinglin Pan <panqinglin2020@iscas.ac.cn>
> 
> diff --git a/arch/riscv/Kconfig b/arch/riscv/Kconfig
> index 3b41165a8b10..1671938f2f81 100644
> --- a/arch/riscv/Kconfig
> +++ b/arch/riscv/Kconfig
> @@ -394,6 +394,20 @@ 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
> +	select RISCV_ALTERNATIVE
> +	default y
> +	help
> +	  Allow kernel to detect SVNAPOT ISA-extension dynamically in boot time
> +	  and enable its usage.
> +
> +	  SVNAPOT extension helps to mark contiguous PTEs as a range
> +	  of contiguous virtual-to-physical translations, with a naturally
> +	  aligned power-of-2 (NAPOT) granularity larger than the base 4KB page
> +	  size.
> +
>  config RISCV_ISA_SVPBMT
>  	bool "SVPBMT extension support"
>  	depends on 64BIT && MMU
> diff --git a/arch/riscv/include/asm/hwcap.h b/arch/riscv/include/asm/hwcap.h
> index b22525290073..15cda8f131aa 100644
> --- a/arch/riscv/include/asm/hwcap.h
> +++ b/arch/riscv/include/asm/hwcap.h
> @@ -54,6 +54,7 @@ extern unsigned long elf_hwcap;
>   */
>  enum riscv_isa_ext_id {
>  	RISCV_ISA_EXT_SSCOFPMF = RISCV_ISA_EXT_BASE,
> +	RISCV_ISA_EXT_SVNAPOT,
>  	RISCV_ISA_EXT_SVPBMT,
>  	RISCV_ISA_EXT_ZICBOM,
>  	RISCV_ISA_EXT_ZIHINTPAUSE,
> @@ -69,6 +70,7 @@ enum riscv_isa_ext_id {
>   */
>  enum riscv_isa_ext_key {
>  	RISCV_ISA_EXT_KEY_FPU,		/* For 'F' and 'D' */
> +	RISCV_ISA_EXT_KEY_SVNAPOT,
>  	RISCV_ISA_EXT_KEY_ZIHINTPAUSE,
>  	RISCV_ISA_EXT_KEY_SVINVAL,
>  	RISCV_ISA_EXT_KEY_MAX,
> @@ -90,6 +92,8 @@ static __always_inline int riscv_isa_ext2key(int num)
>  		return RISCV_ISA_EXT_KEY_FPU;
>  	case RISCV_ISA_EXT_d:
>  		return RISCV_ISA_EXT_KEY_FPU;
> +	case RISCV_ISA_EXT_SVNAPOT:
> +		return RISCV_ISA_EXT_KEY_SVNAPOT;

BTW, while you're touching this, can you make this opportunistic cleanup?

@@ -87,7 +87,6 @@ static __always_inline int riscv_isa_ext2key(int num)
 {
        switch (num) {
        case RISCV_ISA_EXT_f:
-               return RISCV_ISA_EXT_KEY_FPU;
        case RISCV_ISA_EXT_d:
                return RISCV_ISA_EXT_KEY_FPU;
        case RISCV_ISA_EXT_ZIHINTPAUSE:

Thanks,
drew


>  	case RISCV_ISA_EXT_ZIHINTPAUSE:
>  		return RISCV_ISA_EXT_KEY_ZIHINTPAUSE;
>  	case RISCV_ISA_EXT_SVINVAL:
> diff --git a/arch/riscv/include/asm/page.h b/arch/riscv/include/asm/page.h
> index ac70b0fd9a9a..349fad5e35de 100644
> --- a/arch/riscv/include/asm/page.h
> +++ b/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))
> diff --git a/arch/riscv/include/asm/pgtable-64.h b/arch/riscv/include/asm/pgtable-64.h
> index dc42375c2357..598958cbda50 100644
> --- a/arch/riscv/include/asm/pgtable-64.h
> +++ b/arch/riscv/include/asm/pgtable-64.h
> @@ -74,6 +74,40 @@ 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:
>   *
> diff --git a/arch/riscv/include/asm/pgtable.h b/arch/riscv/include/asm/pgtable.h
> index c61ae83aadee..af6174e3fd97 100644
> --- a/arch/riscv/include/asm/pgtable.h
> +++ b/arch/riscv/include/asm/pgtable.h
> @@ -6,10 +6,12 @@
>  #ifndef _ASM_RISCV_PGTABLE_H
>  #define _ASM_RISCV_PGTABLE_H
>  
> +#include <linux/jump_label.h>
>  #include <linux/mmzone.h>
>  #include <linux/sizes.h>
>  
>  #include <asm/pgtable-bits.h>
> +#include <asm/hwcap.h>
>  
>  #ifndef CONFIG_MMU
>  #define KERNEL_LINK_ADDR	PAGE_OFFSET
> @@ -264,10 +266,45 @@ static inline pte_t pud_pte(pud_t pud)
>  	return __pte(pud_val(pud));
>  }
>  
> +static __always_inline bool has_svnapot(void)
> +{
> +	return static_branch_likely(&riscv_isa_ext_keys[RISCV_ISA_EXT_KEY_SVNAPOT]);
> +}
> +
> +#ifdef CONFIG_RISCV_ISA_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 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))
> diff --git a/arch/riscv/kernel/cpu.c b/arch/riscv/kernel/cpu.c
> index bf9dd6764bad..88495f5fcafd 100644
> --- a/arch/riscv/kernel/cpu.c
> +++ b/arch/riscv/kernel/cpu.c
> @@ -165,6 +165,7 @@ 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(zicbom, RISCV_ISA_EXT_ZICBOM),
>  	__RISCV_ISA_EXT_DATA(zihintpause, RISCV_ISA_EXT_ZIHINTPAUSE),
> diff --git a/arch/riscv/kernel/cpufeature.c b/arch/riscv/kernel/cpufeature.c
> index 694267d1fe81..ad12fb5363c3 100644
> --- a/arch/riscv/kernel/cpufeature.c
> +++ b/arch/riscv/kernel/cpufeature.c
> @@ -205,6 +205,7 @@ void __init riscv_fill_hwcap(void)
>  				SET_ISA_EXT_MAP("zihintpause", RISCV_ISA_EXT_ZIHINTPAUSE);
>  				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);
>  			}
>  #undef SET_ISA_EXT_MAP
>  		}
> -- 
> 2.37.4
>
Jisheng Zhang Nov. 29, 2022, 3:42 p.m. UTC | #3 
On Tue, Nov 29, 2022 at 10:49:46AM +0100, Andrew Jones wrote:
> On Mon, Nov 28, 2022 at 10:27:17AM +0800, panqinglin2020@iscas.ac.cn wrote:
> > From: Qinglin Pan <panqinglin2020@iscas.ac.cn>
> > 
> > Add one static key to enable/disable svnapot support, enable this static
> > key when "svnapot" is in the "riscv,isa" field of fdt and SVNAPOT compile
> > option is set. It will influence the behavior of has_svnapot. All code
> > dependent on svnapot should make sure that has_svnapot return true firstly.
> > 
> > Modify PTE definition for Svnapot, and creates some functions in pgtable.h
> > to mark a PTE as napot and check if it is a Svnapot PTE. Until now, only
> > 64KB napot size is supported in spec, so some macros has only 64KB version.
> > 
> > Signed-off-by: Qinglin Pan <panqinglin2020@iscas.ac.cn>
> > 
> > diff --git a/arch/riscv/Kconfig b/arch/riscv/Kconfig
> > index 3b41165a8b10..1671938f2f81 100644
> > --- a/arch/riscv/Kconfig
> > +++ b/arch/riscv/Kconfig
> > @@ -394,6 +394,20 @@ 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
> > +	select RISCV_ALTERNATIVE
> > +	default y
> > +	help
> > +	  Allow kernel to detect SVNAPOT ISA-extension dynamically in boot time
> > +	  and enable its usage.
> > +
> > +	  SVNAPOT extension helps to mark contiguous PTEs as a range
> > +	  of contiguous virtual-to-physical translations, with a naturally
> > +	  aligned power-of-2 (NAPOT) granularity larger than the base 4KB page
> > +	  size.
> > +
> >  config RISCV_ISA_SVPBMT
> >  	bool "SVPBMT extension support"
> >  	depends on 64BIT && MMU
> > diff --git a/arch/riscv/include/asm/hwcap.h b/arch/riscv/include/asm/hwcap.h
> > index b22525290073..15cda8f131aa 100644
> > --- a/arch/riscv/include/asm/hwcap.h
> > +++ b/arch/riscv/include/asm/hwcap.h
> > @@ -54,6 +54,7 @@ extern unsigned long elf_hwcap;
> >   */
> >  enum riscv_isa_ext_id {
> >  	RISCV_ISA_EXT_SSCOFPMF = RISCV_ISA_EXT_BASE,
> > +	RISCV_ISA_EXT_SVNAPOT,
> >  	RISCV_ISA_EXT_SVPBMT,
> >  	RISCV_ISA_EXT_ZICBOM,
> >  	RISCV_ISA_EXT_ZIHINTPAUSE,
> > @@ -69,6 +70,7 @@ enum riscv_isa_ext_id {
> >   */
> >  enum riscv_isa_ext_key {
> >  	RISCV_ISA_EXT_KEY_FPU,		/* For 'F' and 'D' */
> > +	RISCV_ISA_EXT_KEY_SVNAPOT,
> >  	RISCV_ISA_EXT_KEY_ZIHINTPAUSE,
> >  	RISCV_ISA_EXT_KEY_SVINVAL,
> >  	RISCV_ISA_EXT_KEY_MAX,
> > @@ -90,6 +92,8 @@ static __always_inline int riscv_isa_ext2key(int num)
> >  		return RISCV_ISA_EXT_KEY_FPU;
> >  	case RISCV_ISA_EXT_d:
> >  		return RISCV_ISA_EXT_KEY_FPU;
> > +	case RISCV_ISA_EXT_SVNAPOT:
> > +		return RISCV_ISA_EXT_KEY_SVNAPOT;
> >  	case RISCV_ISA_EXT_ZIHINTPAUSE:
> >  		return RISCV_ISA_EXT_KEY_ZIHINTPAUSE;
> >  	case RISCV_ISA_EXT_SVINVAL:
> > diff --git a/arch/riscv/include/asm/page.h b/arch/riscv/include/asm/page.h
> > index ac70b0fd9a9a..349fad5e35de 100644
> > --- a/arch/riscv/include/asm/page.h
> > +++ b/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))
> > diff --git a/arch/riscv/include/asm/pgtable-64.h b/arch/riscv/include/asm/pgtable-64.h
> > index dc42375c2357..598958cbda50 100644
> > --- a/arch/riscv/include/asm/pgtable-64.h
> > +++ b/arch/riscv/include/asm/pgtable-64.h
> > @@ -74,6 +74,40 @@ 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:
> >   *
> > diff --git a/arch/riscv/include/asm/pgtable.h b/arch/riscv/include/asm/pgtable.h
> > index c61ae83aadee..af6174e3fd97 100644
> > --- a/arch/riscv/include/asm/pgtable.h
> > +++ b/arch/riscv/include/asm/pgtable.h
> > @@ -6,10 +6,12 @@
> >  #ifndef _ASM_RISCV_PGTABLE_H
> >  #define _ASM_RISCV_PGTABLE_H
> >  
> > +#include <linux/jump_label.h>
> >  #include <linux/mmzone.h>
> >  #include <linux/sizes.h>
> >  
> >  #include <asm/pgtable-bits.h>
> > +#include <asm/hwcap.h>
> >  
> >  #ifndef CONFIG_MMU
> >  #define KERNEL_LINK_ADDR	PAGE_OFFSET
> > @@ -264,10 +266,45 @@ static inline pte_t pud_pte(pud_t pud)
> >  	return __pte(pud_val(pud));
> >  }
> >  
> > +static __always_inline bool has_svnapot(void)
> > +{
> > +	return static_branch_likely(&riscv_isa_ext_keys[RISCV_ISA_EXT_KEY_SVNAPOT]);
> 
> I'm not sure if this should be likely or unlikely.
> 
> > +}
> > +
> > +#ifdef CONFIG_RISCV_ISA_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 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))
> 
> We've been burned with static branches inside heavily used inline
> functions before (see [1]). There's a series[2] that was meant to
> help with this. I haven't seen a refresh of that though.
> 
> [1] https://lore.kernel.org/linux-riscv/20220922060958.44203-1-samuel@sholland.org/
> [2] https://lore.kernel.org/all/20221006070818.3616-1-jszhang@kernel.org/

I will send a refresh of [2] tomorrow.

Thanks
Qinglin Pan Dec. 4, 2022, 9:21 a.m. UTC | #4 
On 2022/11/29 17:49, Andrew Jones wrote:
> On Mon, Nov 28, 2022 at 10:27:17AM +0800, panqinglin2020@iscas.ac.cn wrote:
>> From: Qinglin Pan <panqinglin2020@iscas.ac.cn>
>>
>> Add one static key to enable/disable svnapot support, enable this static
>> key when "svnapot" is in the "riscv,isa" field of fdt and SVNAPOT compile
>> option is set. It will influence the behavior of has_svnapot. All code
>> dependent on svnapot should make sure that has_svnapot return true firstly.
>>
>> Modify PTE definition for Svnapot, and creates some functions in pgtable.h
>> to mark a PTE as napot and check if it is a Svnapot PTE. Until now, only
>> 64KB napot size is supported in spec, so some macros has only 64KB version.
>>
>> Signed-off-by: Qinglin Pan <panqinglin2020@iscas.ac.cn>
>>
>> diff --git a/arch/riscv/Kconfig b/arch/riscv/Kconfig
>> index 3b41165a8b10..1671938f2f81 100644
>> --- a/arch/riscv/Kconfig
>> +++ b/arch/riscv/Kconfig
>> @@ -394,6 +394,20 @@ 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
>> +	select RISCV_ALTERNATIVE
>> +	default y
>> +	help
>> +	  Allow kernel to detect SVNAPOT ISA-extension dynamically in boot time
>> +	  and enable its usage.
>> +
>> +	  SVNAPOT extension helps to mark contiguous PTEs as a range
>> +	  of contiguous virtual-to-physical translations, with a naturally
>> +	  aligned power-of-2 (NAPOT) granularity larger than the base 4KB page
>> +	  size.
>> +
>>   config RISCV_ISA_SVPBMT
>>   	bool "SVPBMT extension support"
>>   	depends on 64BIT && MMU
>> diff --git a/arch/riscv/include/asm/hwcap.h b/arch/riscv/include/asm/hwcap.h
>> index b22525290073..15cda8f131aa 100644
>> --- a/arch/riscv/include/asm/hwcap.h
>> +++ b/arch/riscv/include/asm/hwcap.h
>> @@ -54,6 +54,7 @@ extern unsigned long elf_hwcap;
>>    */
>>   enum riscv_isa_ext_id {
>>   	RISCV_ISA_EXT_SSCOFPMF = RISCV_ISA_EXT_BASE,
>> +	RISCV_ISA_EXT_SVNAPOT,
>>   	RISCV_ISA_EXT_SVPBMT,
>>   	RISCV_ISA_EXT_ZICBOM,
>>   	RISCV_ISA_EXT_ZIHINTPAUSE,
>> @@ -69,6 +70,7 @@ enum riscv_isa_ext_id {
>>    */
>>   enum riscv_isa_ext_key {
>>   	RISCV_ISA_EXT_KEY_FPU,		/* For 'F' and 'D' */
>> +	RISCV_ISA_EXT_KEY_SVNAPOT,
>>   	RISCV_ISA_EXT_KEY_ZIHINTPAUSE,
>>   	RISCV_ISA_EXT_KEY_SVINVAL,
>>   	RISCV_ISA_EXT_KEY_MAX,
>> @@ -90,6 +92,8 @@ static __always_inline int riscv_isa_ext2key(int num)
>>   		return RISCV_ISA_EXT_KEY_FPU;
>>   	case RISCV_ISA_EXT_d:
>>   		return RISCV_ISA_EXT_KEY_FPU;
>> +	case RISCV_ISA_EXT_SVNAPOT:
>> +		return RISCV_ISA_EXT_KEY_SVNAPOT;
>>   	case RISCV_ISA_EXT_ZIHINTPAUSE:
>>   		return RISCV_ISA_EXT_KEY_ZIHINTPAUSE;
>>   	case RISCV_ISA_EXT_SVINVAL:
>> diff --git a/arch/riscv/include/asm/page.h b/arch/riscv/include/asm/page.h
>> index ac70b0fd9a9a..349fad5e35de 100644
>> --- a/arch/riscv/include/asm/page.h
>> +++ b/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))
>> diff --git a/arch/riscv/include/asm/pgtable-64.h b/arch/riscv/include/asm/pgtable-64.h
>> index dc42375c2357..598958cbda50 100644
>> --- a/arch/riscv/include/asm/pgtable-64.h
>> +++ b/arch/riscv/include/asm/pgtable-64.h
>> @@ -74,6 +74,40 @@ 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:
>>    *
>> diff --git a/arch/riscv/include/asm/pgtable.h b/arch/riscv/include/asm/pgtable.h
>> index c61ae83aadee..af6174e3fd97 100644
>> --- a/arch/riscv/include/asm/pgtable.h
>> +++ b/arch/riscv/include/asm/pgtable.h
>> @@ -6,10 +6,12 @@
>>   #ifndef _ASM_RISCV_PGTABLE_H
>>   #define _ASM_RISCV_PGTABLE_H
>>   
>> +#include <linux/jump_label.h>
>>   #include <linux/mmzone.h>
>>   #include <linux/sizes.h>
>>   
>>   #include <asm/pgtable-bits.h>
>> +#include <asm/hwcap.h>
>>   
>>   #ifndef CONFIG_MMU
>>   #define KERNEL_LINK_ADDR	PAGE_OFFSET
>> @@ -264,10 +266,45 @@ static inline pte_t pud_pte(pud_t pud)
>>   	return __pte(pud_val(pud));
>>   }
>>   
>> +static __always_inline bool has_svnapot(void)
>> +{
>> +	return static_branch_likely(&riscv_isa_ext_keys[RISCV_ISA_EXT_KEY_SVNAPOT]);
> 
> I'm not sure if this should be likely or unlikely.
> 
>> +}
>> +
>> +#ifdef CONFIG_RISCV_ISA_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 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))
> 
> We've been burned with static branches inside heavily used inline
> functions before (see [1]). There's a series[2] that was meant to
> help with this. I haven't seen a refresh of that though.
> 
> [1] https://lore.kernel.org/linux-riscv/20220922060958.44203-1-samuel@sholland.org/
> [2] https://lore.kernel.org/all/20221006070818.3616-1-jszhang@kernel.org/

Hi Andrew,

I will reimplement has_svnapot with alternative in next version
patchset (as I did in previous versions) :-(
And we can replace has_svnapot with
riscv_has_extension_{likely, unlikely} from [2] in the future.

Thanks,
Qinglin

> 
>> +		res = res & (res - 1UL);
>> +
>> +	return res;
>>   }
>>   
>>   #define pte_page(x)     pfn_to_page(pte_pfn(x))
>> diff --git a/arch/riscv/kernel/cpu.c b/arch/riscv/kernel/cpu.c
>> index bf9dd6764bad..88495f5fcafd 100644
>> --- a/arch/riscv/kernel/cpu.c
>> +++ b/arch/riscv/kernel/cpu.c
>> @@ -165,6 +165,7 @@ 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(zicbom, RISCV_ISA_EXT_ZICBOM),
>>   	__RISCV_ISA_EXT_DATA(zihintpause, RISCV_ISA_EXT_ZIHINTPAUSE),
>> diff --git a/arch/riscv/kernel/cpufeature.c b/arch/riscv/kernel/cpufeature.c
>> index 694267d1fe81..ad12fb5363c3 100644
>> --- a/arch/riscv/kernel/cpufeature.c
>> +++ b/arch/riscv/kernel/cpufeature.c
>> @@ -205,6 +205,7 @@ void __init riscv_fill_hwcap(void)
>>   				SET_ISA_EXT_MAP("zihintpause", RISCV_ISA_EXT_ZIHINTPAUSE);
>>   				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);
>>   			}
>>   #undef SET_ISA_EXT_MAP
>>   		}
>> -- 
>> 2.37.4
>>
> 
> Besides the static branch questions, this looks good to me
> 
> Reviewed-by: Andrew Jones <ajones@ventanamicro.com>
> 
> Thanks,
> drew
diff mbox series

Patch

diff --git a/arch/riscv/Kconfig b/arch/riscv/Kconfig
index 3b41165a8b10..1671938f2f81 100644
--- a/arch/riscv/Kconfig
+++ b/arch/riscv/Kconfig
@@ -394,6 +394,20 @@  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
+	select RISCV_ALTERNATIVE
+	default y
+	help
+	  Allow kernel to detect SVNAPOT ISA-extension dynamically in boot time
+	  and enable its usage.
+
+	  SVNAPOT extension helps to mark contiguous PTEs as a range
+	  of contiguous virtual-to-physical translations, with a naturally
+	  aligned power-of-2 (NAPOT) granularity larger than the base 4KB page
+	  size.
+
 config RISCV_ISA_SVPBMT
 	bool "SVPBMT extension support"
 	depends on 64BIT && MMU
diff --git a/arch/riscv/include/asm/hwcap.h b/arch/riscv/include/asm/hwcap.h
index b22525290073..15cda8f131aa 100644
--- a/arch/riscv/include/asm/hwcap.h
+++ b/arch/riscv/include/asm/hwcap.h
@@ -54,6 +54,7 @@  extern unsigned long elf_hwcap;
  */
 enum riscv_isa_ext_id {
 	RISCV_ISA_EXT_SSCOFPMF = RISCV_ISA_EXT_BASE,
+	RISCV_ISA_EXT_SVNAPOT,
 	RISCV_ISA_EXT_SVPBMT,
 	RISCV_ISA_EXT_ZICBOM,
 	RISCV_ISA_EXT_ZIHINTPAUSE,
@@ -69,6 +70,7 @@  enum riscv_isa_ext_id {
  */
 enum riscv_isa_ext_key {
 	RISCV_ISA_EXT_KEY_FPU,		/* For 'F' and 'D' */
+	RISCV_ISA_EXT_KEY_SVNAPOT,
 	RISCV_ISA_EXT_KEY_ZIHINTPAUSE,
 	RISCV_ISA_EXT_KEY_SVINVAL,
 	RISCV_ISA_EXT_KEY_MAX,
@@ -90,6 +92,8 @@  static __always_inline int riscv_isa_ext2key(int num)
 		return RISCV_ISA_EXT_KEY_FPU;
 	case RISCV_ISA_EXT_d:
 		return RISCV_ISA_EXT_KEY_FPU;
+	case RISCV_ISA_EXT_SVNAPOT:
+		return RISCV_ISA_EXT_KEY_SVNAPOT;
 	case RISCV_ISA_EXT_ZIHINTPAUSE:
 		return RISCV_ISA_EXT_KEY_ZIHINTPAUSE;
 	case RISCV_ISA_EXT_SVINVAL:
diff --git a/arch/riscv/include/asm/page.h b/arch/riscv/include/asm/page.h
index ac70b0fd9a9a..349fad5e35de 100644
--- a/arch/riscv/include/asm/page.h
+++ b/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))
diff --git a/arch/riscv/include/asm/pgtable-64.h b/arch/riscv/include/asm/pgtable-64.h
index dc42375c2357..598958cbda50 100644
--- a/arch/riscv/include/asm/pgtable-64.h
+++ b/arch/riscv/include/asm/pgtable-64.h
@@ -74,6 +74,40 @@  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:
  *
diff --git a/arch/riscv/include/asm/pgtable.h b/arch/riscv/include/asm/pgtable.h
index c61ae83aadee..af6174e3fd97 100644
--- a/arch/riscv/include/asm/pgtable.h
+++ b/arch/riscv/include/asm/pgtable.h
@@ -6,10 +6,12 @@ 
 #ifndef _ASM_RISCV_PGTABLE_H
 #define _ASM_RISCV_PGTABLE_H
 
+#include <linux/jump_label.h>
 #include <linux/mmzone.h>
 #include <linux/sizes.h>
 
 #include <asm/pgtable-bits.h>
+#include <asm/hwcap.h>
 
 #ifndef CONFIG_MMU
 #define KERNEL_LINK_ADDR	PAGE_OFFSET
@@ -264,10 +266,45 @@  static inline pte_t pud_pte(pud_t pud)
 	return __pte(pud_val(pud));
 }
 
+static __always_inline bool has_svnapot(void)
+{
+	return static_branch_likely(&riscv_isa_ext_keys[RISCV_ISA_EXT_KEY_SVNAPOT]);
+}
+
+#ifdef CONFIG_RISCV_ISA_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 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))
diff --git a/arch/riscv/kernel/cpu.c b/arch/riscv/kernel/cpu.c
index bf9dd6764bad..88495f5fcafd 100644
--- a/arch/riscv/kernel/cpu.c
+++ b/arch/riscv/kernel/cpu.c
@@ -165,6 +165,7 @@  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(zicbom, RISCV_ISA_EXT_ZICBOM),
 	__RISCV_ISA_EXT_DATA(zihintpause, RISCV_ISA_EXT_ZIHINTPAUSE),
diff --git a/arch/riscv/kernel/cpufeature.c b/arch/riscv/kernel/cpufeature.c
index 694267d1fe81..ad12fb5363c3 100644
--- a/arch/riscv/kernel/cpufeature.c
+++ b/arch/riscv/kernel/cpufeature.c
@@ -205,6 +205,7 @@  void __init riscv_fill_hwcap(void)
 				SET_ISA_EXT_MAP("zihintpause", RISCV_ISA_EXT_ZIHINTPAUSE);
 				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);
 			}
 #undef SET_ISA_EXT_MAP
 		}