Add update_mmu_tlb_range() to simplify code
Subsystems (6)
v2 cover Add update_mmu_tlb_range() to simplify code
This series of commits mainly adds the update_mmu_tlb_range() to batch update tlb in an address range.
After the commit 19eaf44954df ("mm: thp: support allocation of anonymous multi-size THP"), We may need to batch update tlb of a certain address range by calling update_mmu_tlb() in a loop. Using the update_mmu_tlb_range(), we can simplify the code and possibly reduce the execution of some unnecessary code in some architectures.
Thanks,
Bang
Changes since v1 [1]:
- Add __HAVE_ARCH_UPDATE_MMU_TLB_RANGE macro (per Lance Yang)
[1] https://lore.kernel.org/linux-mm/[email protected]/
Bang Li (5):
LoongArch: Add update_mmu_tlb_range()
mips: Add update_mmu_tlb_range()
riscv: Add update_mmu_tlb_range()
xtensa: Add update_mmu_tlb_range()
mm: Add update_mmu_tlb_range()
arch/loongarch/include/asm/pgtable.h | 4 ++++ arch/mips/include/asm/pgtable.h | 4 ++++ arch/riscv/include/asm/pgtable.h | 4 ++++ arch/xtensa/include/asm/pgtable.h | 4 ++++ arch/xtensa/mm/tlb.c | 6 ++++++ include/linux/pgtable.h | 8 ++++++++ mm/memory.c | 4 +--- 7 files changed, 31 insertions(+), 3 deletions(-)
Signature · 1 line
2.19.1.6.gb485710b
PATCH v2 1/5 LoongArch: Add update_mmu_tlb_range()
Added update_mmu_tlb_range function, we can batch update tlb of an address range.
--- arch/loongarch/include/asm/pgtable.h | 4 ++++ 1 file changed, 4 insertions(+)
arch/loongarch/include/asm/pgtable.h +4 -0
@@ -470,6 +470,10 @@ static inline void update_mmu_cache_range(struct vm_fault *vmf,#define __HAVE_ARCH_UPDATE_MMU_TLB#define update_mmu_tlb update_mmu_cache +#define __HAVE_ARCH_UPDATE_MMU_TLB_RANGE +#define update_mmu_tlb_range (vma, addr, ptep, nr) \ + update_mmu_cache_range(NULL, vma, addr, ptep, nr) +static inline void update_mmu_cache_pmd (struct vm_area_struct *vma,unsigned long address,pmd_t *pmdp) {
Signature · 1 line
2.19.1.6.gb485710b
PATCH v2 2/5 mips: Add update_mmu_tlb_range()
In reply to Bang LiAdded update_mmu_tlb_range function, we can batch update tlb of an address range.
--- arch/mips/include/asm/pgtable.h | 4 ++++ 1 file changed, 4 insertions(+)
arch/mips/include/asm/pgtable.h +4 -0
@@ -597,6 +597,10 @@ static inline void update_mmu_cache_range(struct vm_fault *vmf,#define __HAVE_ARCH_UPDATE_MMU_TLB#define update_mmu_tlb update_mmu_cache +#define __HAVE_ARCH_UPDATE_MMU_TLB_RANGE +#define update_mmu_tlb_range (vma, address, ptep, nr) \ + update_mmu_cache_range(NULL, vma, address, ptep, nr) +static inline void update_mmu_cache_pmd (struct vm_area_struct *vma,unsigned long address,pmd_t *pmdp) {
Signature · 1 line
2.19.1.6.gb485710b
PATCH v2 3/5 riscv: Add update_mmu_tlb_range()
In reply to Bang LiAdded update_mmu_tlb_range function, we can batch update tlb of an address range.
--- arch/riscv/include/asm/pgtable.h | 4 ++++ 1 file changed, 4 insertions(+)
arch/riscv/include/asm/pgtable.h +4 -0
@@ -489,6 +489,10 @@ static inline void update_mmu_cache_range(struct vm_fault *vmf,#define __HAVE_ARCH_UPDATE_MMU_TLB#define update_mmu_tlb update_mmu_cache +#define __HAVE_ARCH_UPDATE_MMU_TLB_RANGE +#define update_mmu_tlb_range (vma, addr, ptep, nr) \ + update_mmu_cache_range(NULL, vma, addr, ptep, nr) +static inline void update_mmu_cache_pmd (struct vm_area_struct *vma,unsigned long address,pmd_t *pmdp) {
Signature · 1 line
2.19.1.6.gb485710b
Re: [PATCH v2 3/5] riscv: Add update_mmu_tlb_range()
Hi Bang,
Quoted 13 lines · Bang Li
Added
update_mmu_tlb_rangefunction, we can batch update tlb of an address range.Signed-off-by Bang Li <[email protected]>--- arch/riscv/include/asm/pgtable.h | 4 ++++ 1 file changed, 4 insertions(+)arch/riscv/include/asm/pgtable.h +0
-0
arch/riscv/include/asm/pgtable.h static inline void update_mmu_cache_range(struct vm_fault *vmf, · line 498
Quoted 10 lines
arch/riscv/include/asm/pgtable.h +4
-0@@ -489,6 +489,10 @@ static inline void update_mmu_cache_range(struct vm_fault *vmf,#define __HAVE_ARCH_UPDATE_MMU_TLB#define update_mmu_tlb update_mmu_cache +#define __HAVE_ARCH_UPDATE_MMU_TLB_RANGE +#define update_mmu_tlb_range (vma, addr, ptep, nr) \ + update_mmu_cache_range(NULL, vma, addr, ptep, nr) +static inline void update_mmu_cache_pmd (struct vm_area_struct *vma,unsigned long address,pmd_t *pmdp) {
You can add:
Thanks,
Alex
Re: [PATCH v2 3/5] riscv: Add update_mmu_tlb_range()
Thanks, Alex!
Quoted 35 lines · Alexandre Ghiti
Hi Bang,
Quoted 13 lines · Bang Li
Added
update_mmu_tlb_rangefunction, we can batch update tlb of an address range.Signed-off-by Bang Li <[email protected]>---
arch/riscv/include/asm/pgtable.h| 4 ++++1 file changed, 4 insertions(+)arch/riscv/include/asm/pgtable.h +0
-0arch/riscv/include/asm/pgtable.h static inline void update_mmu_cache_range(struct · line 498
Quoted 10 lines
arch/riscv/include/asm/pgtable.h +4
-0@@ -489,6 +489,10 @@ static inline void update_mmu_cache_range(struct vm_fault *vmf, #define __HAVE_ARCH_UPDATE_MMU_TLB #define update_mmu_tlb update_mmu_cache +#define __HAVE_ARCH_UPDATE_MMU_TLB_RANGE +#define update_mmu_tlb_range (vma, addr, ptep, nr) \ + update_mmu_cache_range(NULL, vma, addr, ptep, nr) + static inline void update_mmu_cache_pmd(struct vm_area_struct *vma, unsigned long address, pmd_t *pmdp) {You can add:
Reviewed-by Alexandre Ghiti <[email protected]>Thanks,
Alex
PATCH v2 4/5 xtensa: Add update_mmu_tlb_range()
In reply to Bang LiAdded update_mmu_tlb_range function, we can batch update tlb of an address range.
--- arch/xtensa/include/asm/pgtable.h | 4 ++++ arch/xtensa/mm/tlb.c | 6 ++++++ 2 files changed, 10 insertions(+)
arch/xtensa/include/asm/pgtable.h +4 -0
@@ -414,6 +414,10 @@ void update_mmu_tlb(struct vm_area_struct *vma, unsigned long address, pte_t *ptep);#define __HAVE_ARCH_UPDATE_MMU_TLB +void update_mmu_tlb_range (struct vm_area_struct *vma, +unsigned long address,pte_t *ptep,unsigned int nr); +#define __HAVE_ARCH_UPDATE_MMU_TLB_RANGE +#endif /* !defined (__ASSEMBLY__) */ #define __HAVE_ARCH_PTEP_TEST_AND_CLEAR_YOUNG
arch/xtensa/mm/tlb.c +6 -0
@@ -169,6 +169,12 @@ void update_mmu_tlb(struct vm_area_struct *vma,local_flush_tlb_page (vma, address); } +void update_mmu_tlb_range (struct vm_area_struct *vma, +unsigned long address,pte_t *ptep,unsigned int nr) +{ +local_flush_tlb_range (vma, address, address +PAGE_SIZE * nr); +} +#ifdef CONFIG_DEBUG_TLB_SANITY static unsigned get_pte_for_vaddr (unsigned vaddr)
Signature · 1 line
2.19.1.6.gb485710b
PATCH v2 5/5 mm: Add update_mmu_tlb_range()
In reply to Bang LiAfter the commit 19eaf44954df ("mm: thp: support allocation of anonymous multi-size THP"), it may need to batch update tlb of an address range through the update_mmu_tlb function. We can simplify this operation by adding the update_mmu_tlb_range function, which may also reduce the execution of some unnecessary code in some architectures.
--- include/linux/pgtable.h | 8 ++++++++ mm/memory.c | 4 +--- 2 files changed, 9 insertions(+), 3 deletions(-)
include/linux/pgtable.h +8 -0
@@ -737,6 +737,14 @@ static inline void update_mmu_tlb(struct vm_area_struct *vma,#define __HAVE_ARCH_UPDATE_MMU_TLB#endif +#ifndef __HAVE_ARCH_UPDATE_MMU_TLB_RANGE +static inline void update_mmu_tlb_range (struct vm_area_struct *vma, +unsigned long address,pte_t *ptep,unsigned int nr) +{ +} +#define __HAVE_ARCH_UPDATE_MMU_TLB_RANGE +#endif +/* * Some architectures may be able to avoid expensive synchronization * primitives when modifications are made to PTE's which are already
mm/memory.c +1 -3
@@ -4421,7 +4421,6 @@ static vm_fault_t do_anonymous_page(struct vm_fault *vmf)vm_fault_t ret =0 ;int nr_pages =1 ;pte_t entry; -int i;/* File mapping without ->vm_ops ? */ if (vma->vm_flags &VM_SHARED )
@@ -4491,8 +4490,7 @@ static vm_fault_t do_anonymous_page(struct vm_fault *vmf)update_mmu_tlb (vma, addr, vmf->pte); goto release; }else if (nr_pages >1 && !pte_range_none (vmf->pte, nr_pages)) { -for (i =0 ; i < nr_pages; i++) -update_mmu_tlb (vma, addr +PAGE_SIZE * i, vmf->pte + i); +update_mmu_tlb_range (vma, addr, vmf->pte, nr_pages); goto release; }
Signature · 1 line
2.19.1.6.gb485710b
Re: [PATCH v2 5/5] mm: Add update_mmu_tlb_range()
Quoted 18 lines · Bang Li
After the commit
19eaf44954df("mm: thp: support allocation of anonymous multi-size THP"), it may need to batch update tlb of an address range through theupdate_mmu_tlbfunction. We can simplify this operation by adding theupdate_mmu_tlb_rangefunction, which may also reduce the execution of some unnecessary code in some architectures.Signed-off-by Bang Li <[email protected]>--- include/linux/pgtable.h | 8 ++++++++ mm/memory.c | 4 +--- 2 files changed, 9 insertions(+), 3 deletions(-)include/linux/pgtable.h +0
-0
include/linux/pgtable.h static inline void update_mmu_tlb(struct vm_area_struct *vma, · line 740
Quoted 4 lines
include/linux/pgtable.h +1
-0@@ -737,6 +737,14 @@ static inline void update_mmu_tlb(struct vm_area_struct *vma,#define __HAVE_ARCH_UPDATE_MMU_TLB#endif +#ifndef __HAVE_ARCH_UPDATE_MMU_TLB_RANGE
IIRC, the contemporary practice is to define a macro with the same name as the function if it is being overridden.
Thanks,
Lance
Quoted 35 lines
include/linux/pgtable.h +7
-0+static inline void update_mmu_tlb_range (struct vm_area_struct *vma, +unsigned long address,pte_t *ptep,unsigned int nr) +{ +} +#define __HAVE_ARCH_UPDATE_MMU_TLB_RANGE +#endif +/* * Some architectures may be able to avoid expensive synchronization * primitives when modifications are made to PTE's which are already mm/memory.c +1
-4@@ -4421,7 +4421,6 @@ static vm_fault_t do_anonymous_page(struct vm_fault *vmf)vm_fault_t ret =0 ;int nr_pages =1 ;pte_t entry; -int i;/* File mapping without ->vm_ops ? */ if (vma->vm_flags &VM_SHARED )@@ -4491,8 +4490,7 @@ static vm_fault_t do_anonymous_page(struct vm_fault *vmf)update_mmu_tlb (vma, addr, vmf->pte); goto release; }else if (nr_pages >1 && !pte_range_none (vmf->pte, nr_pages)) { -for (i =0 ; i < nr_pages; i++) -update_mmu_tlb (vma, addr +PAGE_SIZE * i, vmf->pte + i); +update_mmu_tlb_range (vma, addr, vmf->pte, nr_pages); goto release; } --2.19.1.6.gb485710b
Re: [PATCH v2 5/5] mm: Add update_mmu_tlb_range()
Hey Lance,
Thanks for taking time to review!
Quoted 22 lines · Lance Yang
Quoted 21 lines · Bang Li
After the commit
19eaf44954df("mm: thp: support allocation of anonymous multi-size THP"), it may need to batch update tlb of an address range through theupdate_mmu_tlbfunction. We can simplify this operation by adding theupdate_mmu_tlb_rangefunction, which may also reduce the execution of some unnecessary code in some architectures.Signed-off-by Bang Li <[email protected]>--- include/linux/pgtable.h | 8 ++++++++ mm/memory.c | 4 +--- 2 files changed, 9 insertions(+), 3 deletions(-)include/linux/pgtable.h +0
-0@@ -737,6 +737,14 @@ static inline void update_mmu_tlb(struct vm_area_struct *vma,#define __HAVE_ARCH_UPDATE_MMU_TLB#endif
Quoted 4 lines
include/linux/pgtable.h static inline void update_mmu_tlb(struct vm_area_struct *vma, · line 740
Quoted 1 line
include/linux/pgtable.h +1
-0+#ifndef __HAVE_ARCH_UPDATE_MMU_TLB_RANGEIIRC, the contemporary practice is to define a macro with the same name as the function if it is being overridden.
The macro __HAVE_ARCH_UPDATE_MMU_TLB_RANGE defined here is aligned with the macro __HAVE_ARCH_UPDATE_MMU_TLB corresponding to the update_mmu_tlb function. IMO, it should be better to use my method in this case.
Thanks,
Bang
Quoted 39 lines
Thanks,
LanceQuoted 35 lines
include/linux/pgtable.h +7
-0+static inline void update_mmu_tlb_range (struct vm_area_struct *vma, +unsigned long address,pte_t *ptep,unsigned int nr) +{ +} +#define __HAVE_ARCH_UPDATE_MMU_TLB_RANGE +#endif +/* * Some architectures may be able to avoid expensive synchronization * primitives when modifications are made to PTE's which are already mm/memory.c +1
-4@@ -4421,7 +4421,6 @@ static vm_fault_t do_anonymous_page(struct vm_fault *vmf)vm_fault_t ret =0 ;int nr_pages =1 ;pte_t entry; -int i;/* File mapping without ->vm_ops ? */ if (vma->vm_flags &VM_SHARED )@@ -4491,8 +4490,7 @@ static vm_fault_t do_anonymous_page(struct vm_fault *vmf)update_mmu_tlb (vma, addr, vmf->pte); goto release; }else if (nr_pages >1 && !pte_range_none (vmf->pte, nr_pages)) { -for (i =0 ; i < nr_pages; i++) -update_mmu_tlb (vma, addr +PAGE_SIZE * i, vmf->pte + i); +update_mmu_tlb_range (vma, addr, vmf->pte, nr_pages); goto release; } --2.19.1.6.gb485710b
Re: [PATCH v2 5/5] mm: Add update_mmu_tlb_range()
In reply to Bang LiQuoted 17 lines · Bang Li
After the commit
19eaf44954df("mm: thp: support allocation of anonymous multi-size THP"), it may need to batch update tlb of an address range through theupdate_mmu_tlbfunction. We can simplify this operation by adding theupdate_mmu_tlb_rangefunction, which may also reduce the execution of some unnecessary code in some architectures.Signed-off-by Bang Li <[email protected]>--- include/linux/pgtable.h | 8 ++++++++ mm/memory.c | 4 +--- 2 files changed, 9 insertions(+), 3 deletions(-)include/linux/pgtable.h +0
-0
include/linux/pgtable.h static inline void update_mmu_tlb(struct vm_area_struct *vma, · line 738
Quoted 2 lines
include/linux/pgtable.h +0
-0@@ -737,6 +737,14 @@ static inline void update_mmu_tlb(struct vm_area_struct *vma,#define __HAVE_ARCH_UPDATE_MMU_TLB#endif
Given you are implementing update_mmu_tlb_range() in all the arches that currently override update_mmu_tlb() I wonder if it would be cleaner to remove update_mmu_tlb() from all those arches, and define generically, removing the ability for arches to override it:
static inline void update_mmu_tlb(struct vm_area_struct *vma,
unsigned long address, pte_t *ptep)
{
update_mmu_tlb_range(vma, address, ptep, 1);
}include/linux/pgtable.h static inline void update_mmu_tlb(struct vm_area_struct *vma, · line 746
Quoted 8 lines
include/linux/pgtable.h +7
-0+#ifndef __HAVE_ARCH_UPDATE_MMU_TLB_RANGE +static inline void update_mmu_tlb_range (struct vm_area_struct *vma, +unsigned long address,pte_t *ptep,unsigned int nr) +{ +} +#define __HAVE_ARCH_UPDATE_MMU_TLB_RANGE +#endif
Then you could use the modern override scheme as Lance suggested and you won't have any confusion with __HAVE_ARCH_UPDATE_MMU_TLB because it won't exist anymore.
Quoted 17 lines
include/linux/pgtable.h +1
-0+/* * Some architectures may be able to avoid expensive synchronization * primitives when modifications are made to PTE's which are already mm/memory.c +0
-1@@ -4421,7 +4421,6 @@ static vm_fault_t do_anonymous_page(struct vm_fault *vmf)vm_fault_t ret =0 ;int nr_pages =1 ;pte_t entry; -int i;/* File mapping without ->vm_ops ? */ if (vma->vm_flags &VM_SHARED )
mm/memory.c static vm_fault_t do_anonymous_page(struct vm_fault *vmf) · line 4493
Quoted 6 lines
mm/memory.c +1
-2@@ -4491,8 +4490,7 @@ static vm_fault_t do_anonymous_page(struct vm_fault *vmf)update_mmu_tlb (vma, addr, vmf->pte); goto release; }else if (nr_pages >1 && !pte_range_none (vmf->pte, nr_pages)) { -for (i =0 ; i < nr_pages; i++) -update_mmu_tlb (vma, addr +PAGE_SIZE * i, vmf->pte + i); +update_mmu_tlb_range (vma, addr, vmf->pte, nr_pages);
I certainly agree that this will be a useful helper to have. I expect there will be more users in future.
Quoted 3 lines
mm/memory.c +0
-0goto release; }
Re: [PATCH v2 5/5] mm: Add update_mmu_tlb_range()
Quoted 14 lines · Ryan Roberts
Quoted 12 lines · Bang Li
After the commit
19eaf44954df("mm: thp: support allocation of anonymous multi-size THP"), it may need to batch update tlb of an address range through theupdate_mmu_tlbfunction. We can simplify this operation by adding theupdate_mmu_tlb_rangefunction, which may also reduce the execution of some unnecessary code in some architectures.Signed-off-by Bang Li <[email protected]>--- include/linux/pgtable.h | 8 ++++++++ mm/memory.c | 4 +--- 2 files changed, 9 insertions(+), 3 deletions(-)
Quoted 12 lines
include/linux/pgtable.h static inline void update_mmu_tlb(struct vm_area_struct *vma, · line 738
Quoted 7 lines
include/linux/pgtable.h +0
-0@@ -737,6 +737,14 @@ static inline void update_mmu_tlb(struct vm_area_struct *vma,#define __HAVE_ARCH_UPDATE_MMU_TLB#endif Given you are implementing
update_mmu_tlb_range()in all the arches that currently overrideupdate_mmu_tlb()I wonder if it would be cleaner to removeupdate_mmu_tlb()from all those arches, and define generically, removing the ability for arches to override it:
Sounds great! Let's get it done.
Quoted 8 lines
static inline void update_mmu_tlb(struct
vm_area_struct*vma,
unsigned long address,pte_t*ptep){ update_mmu_tlb_range(vma, address, ptep, 1); }Quoted 1 line
Quoted 10 lines
include/linux/pgtable.h static inline void update_mmu_tlb(struct vm_area_struct *vma, · line 746
Quoted 7 lines
include/linux/pgtable.h +7
-0+#ifndef __HAVE_ARCH_UPDATE_MMU_TLB_RANGE +static inline void update_mmu_tlb_range (struct vm_area_struct *vma, +unsigned long address,pte_t *ptep,unsigned int nr) +{ +} +#define __HAVE_ARCH_UPDATE_MMU_TLB_RANGE +#endif Then you could use the modern override scheme as Lance suggested and you won't have any confusion with
__HAVE_ARCH_UPDATE_MMU_TLBbecause it won't exist anymore.
+1. It might be better to use the modern override scheme :)
Thanks,
Lance
Quoted 32 lines
Quoted 18 lines
include/linux/pgtable.h +1
-0+/* * Some architectures may be able to avoid expensive synchronization * primitives when modifications are made to PTE's which are already mm/memory.c +0
-1@@ -4421,7 +4421,6 @@ static vm_fault_t do_anonymous_page(struct vm_fault *vmf)vm_fault_t ret =0 ;int nr_pages =1 ;pte_t entry; -int i;/* File mapping without ->vm_ops ? */ if (vma->vm_flags &VM_SHARED )mm/memory.c static vm_fault_t do_anonymous_page(struct vm_fault *vmf) · line 4493
Quoted 6 lines
mm/memory.c +1
-2@@ -4491,8 +4490,7 @@ static vm_fault_t do_anonymous_page(struct vm_fault *vmf)update_mmu_tlb (vma, addr, vmf->pte); goto release; }else if (nr_pages >1 && !pte_range_none (vmf->pte, nr_pages)) { -for (i =0 ; i < nr_pages; i++) -update_mmu_tlb (vma, addr +PAGE_SIZE * i, vmf->pte + i); +update_mmu_tlb_range (vma, addr, vmf->pte, nr_pages);I certainly agree that this will be a useful helper to have. I expect there will be more users in future.
Quoted 3 lines
mm/memory.c +0
-0goto release; }
Re: [PATCH v2 5/5] mm: Add update_mmu_tlb_range()
In reply to Ryan RobertsHi Ryan,
Thanks for you review!
Quoted 26 lines · Ryan Roberts
Quoted 17 lines · Bang Li
After the commit
19eaf44954df("mm: thp: support allocation of anonymous multi-size THP"), it may need to batch update tlb of an address range through theupdate_mmu_tlbfunction. We can simplify this operation by adding theupdate_mmu_tlb_rangefunction, which may also reduce the execution of some unnecessary code in some architectures.Signed-off-by Bang Li <[email protected]>--- include/linux/pgtable.h | 8 ++++++++ mm/memory.c | 4 +--- 2 files changed, 9 insertions(+), 3 deletions(-)include/linux/pgtable.h +0
-0include/linux/pgtable.h static inline void update_mmu_tlb(struct vm_area_struct *vma, · line 738
Quoted 2 lines
include/linux/pgtable.h +0
-0@@ -737,6 +737,14 @@ static inline void update_mmu_tlb(struct vm_area_struct *vma,#define __HAVE_ARCH_UPDATE_MMU_TLB#endif Given you are implementing
update_mmu_tlb_range()in all the arches that currently overrideupdate_mmu_tlb()I wonder if it would be cleaner to removeupdate_mmu_tlb()from all those arches, and define generically, removing the ability for arches to override it:
Quoted 5 lines
static inline void update_mmu_tlb(struct
vm_area_struct*vma,
unsigned long address,pte_t*ptep){ update_mmu_tlb_range(vma, address, ptep, 1); }
Agreed! Thank you for your suggestion, I will modify it in the next version.
Quoted 12 lines
include/linux/pgtable.h static inline void update_mmu_tlb(struct vm_area_struct *vma, · line 746
Quoted 9 lines
include/linux/pgtable.h +7
-0+#ifndef __HAVE_ARCH_UPDATE_MMU_TLB_RANGE +static inline void update_mmu_tlb_range (struct vm_area_struct *vma, +unsigned long address,pte_t *ptep,unsigned int nr) +{ +} +#define __HAVE_ARCH_UPDATE_MMU_TLB_RANGE +#endif Then you could use the modern override scheme as Lance suggested and you won't have any confusion with
__HAVE_ARCH_UPDATE_MMU_TLBbecause it won't exist anymore.
Yes, use update_mmu_tlb_range to implement update_mmu_tlb, we only need to define the update_mmu_tlb_range macro.
Quoted 15 lines
Quoted 15 lines
include/linux/pgtable.h +1
-0+/* * Some architectures may be able to avoid expensive synchronization * primitives when modifications are made to PTE's which are already mm/memory.c +0
-1@@ -4421,7 +4421,6 @@ static vm_fault_t do_anonymous_page(struct vm_fault *vmf)vm_fault_t ret =0 ;int nr_pages =1 ;pte_t entry; -int i;
Quoted 12 lines
mm/memory.c static vm_fault_t do_anonymous_page(struct vm_fault *vmf) · line 4493
Quoted 9 lines
mm/memory.c +1
-2/* File mapping without ->vm_ops ? */ if (vma->vm_flags &VM_SHARED )@@ -4491,8 +4490,7 @@ static vm_fault_t do_anonymous_page(struct vm_fault *vmf)update_mmu_tlb (vma, addr, vmf->pte); goto release; }else if (nr_pages >1 && !pte_range_none (vmf->pte, nr_pages)) { -for (i =0 ; i < nr_pages; i++) -update_mmu_tlb (vma, addr +PAGE_SIZE * i, vmf->pte + i); +update_mmu_tlb_range (vma, addr, vmf->pte, nr_pages);I certainly agree that this will be a useful helper to have. I expect there will be more users in future.
Thank you for your affirmation. Baolin’s "add mTHP support for anonymous shmem" series[1] can also use this function to simplify the code.
[1] https://lore.kernel.org/linux-mm/[email protected]/
Thanks,
Bang
Quoted 4 lines
Quoted 4 lines
mm/memory.c +0
-0goto release; }