2023-02-28 21:39:16

by Matthew Wilcox

[permalink] [raw]
Subject: [PATCH v3 21/34] s390: Implement the new page table range API

Add set_ptes() and update_mmu_cache_range().

Signed-off-by: Matthew Wilcox (Oracle) <[email protected]>
Cc: Heiko Carstens <[email protected]>
Cc: Vasily Gorbik <[email protected]>
Cc: Alexander Gordeev <[email protected]>
Cc: Christian Borntraeger <[email protected]>
Cc: Sven Schnelle <[email protected]>
Cc: [email protected]
---
arch/s390/include/asm/pgtable.h | 34 ++++++++++++++++++++++++---------
1 file changed, 25 insertions(+), 9 deletions(-)

diff --git a/arch/s390/include/asm/pgtable.h b/arch/s390/include/asm/pgtable.h
index 2c70b4d1263d..46bf475116f1 100644
--- a/arch/s390/include/asm/pgtable.h
+++ b/arch/s390/include/asm/pgtable.h
@@ -50,6 +50,7 @@ void arch_report_meminfo(struct seq_file *m);
* tables contain all the necessary information.
*/
#define update_mmu_cache(vma, address, ptep) do { } while (0)
+#define update_mmu_cache_range(vma, addr, ptep, nr) do { } while (0)
#define update_mmu_cache_pmd(vma, address, ptep) do { } while (0)

/*
@@ -1317,21 +1318,36 @@ pgprot_t pgprot_writecombine(pgprot_t prot);
pgprot_t pgprot_writethrough(pgprot_t prot);

/*
- * Certain architectures need to do special things when PTEs
- * within a page table are directly modified. Thus, the following
- * hook is made available.
+ * Set multiple PTEs to consecutive pages with a single call. All PTEs
+ * are within the same folio, PMD and VMA.
*/
-static inline void set_pte_at(struct mm_struct *mm, unsigned long addr,
- pte_t *ptep, pte_t entry)
+static inline void set_ptes(struct mm_struct *mm, unsigned long addr,
+ pte_t *ptep, pte_t entry, unsigned int nr)
{
if (pte_present(entry))
entry = clear_pte_bit(entry, __pgprot(_PAGE_UNUSED));
- if (mm_has_pgste(mm))
- ptep_set_pte_at(mm, addr, ptep, entry);
- else
- set_pte(ptep, entry);
+ if (mm_has_pgste(mm)) {
+ for (;;) {
+ ptep_set_pte_at(mm, addr, ptep, entry);
+ if (--nr == 0)
+ break;
+ ptep++;
+ entry = __pte(pte_val(entry) + PAGE_SIZE);
+ addr += PAGE_SIZE;
+ }
+ } else {
+ for (;;) {
+ set_pte(ptep, entry);
+ if (--nr == 0)
+ break;
+ ptep++;
+ entry = __pte(pte_val(entry) + PAGE_SIZE);
+ }
+ }
}

+#define set_pte_at(mm, addr, ptep, pte) set_ptes(mm, addr, ptep, pte, 1)
+
/*
* Conversion functions: convert a page and protection to a page entry,
* and a page entry and page directory to the page they refer to.
--
2.39.1



2023-03-02 13:32:23

by Gerald Schaefer

[permalink] [raw]
Subject: Re: [PATCH v3 21/34] s390: Implement the new page table range API

On Tue, 28 Feb 2023 21:37:24 +0000
"Matthew Wilcox (Oracle)" <[email protected]> wrote:

> Add set_ptes() and update_mmu_cache_range().
>
> Signed-off-by: Matthew Wilcox (Oracle) <[email protected]>
> Cc: Heiko Carstens <[email protected]>
> Cc: Vasily Gorbik <[email protected]>
> Cc: Alexander Gordeev <[email protected]>
> Cc: Christian Borntraeger <[email protected]>
> Cc: Sven Schnelle <[email protected]>
> Cc: [email protected]
> ---
> arch/s390/include/asm/pgtable.h | 34 ++++++++++++++++++++++++---------
> 1 file changed, 25 insertions(+), 9 deletions(-)
>
> diff --git a/arch/s390/include/asm/pgtable.h b/arch/s390/include/asm/pgtable.h
> index 2c70b4d1263d..46bf475116f1 100644
> --- a/arch/s390/include/asm/pgtable.h
> +++ b/arch/s390/include/asm/pgtable.h
> @@ -50,6 +50,7 @@ void arch_report_meminfo(struct seq_file *m);
> * tables contain all the necessary information.
> */
> #define update_mmu_cache(vma, address, ptep) do { } while (0)
> +#define update_mmu_cache_range(vma, addr, ptep, nr) do { } while (0)
> #define update_mmu_cache_pmd(vma, address, ptep) do { } while (0)
>
> /*
> @@ -1317,21 +1318,36 @@ pgprot_t pgprot_writecombine(pgprot_t prot);
> pgprot_t pgprot_writethrough(pgprot_t prot);
>
> /*
> - * Certain architectures need to do special things when PTEs
> - * within a page table are directly modified. Thus, the following
> - * hook is made available.
> + * Set multiple PTEs to consecutive pages with a single call. All PTEs
> + * are within the same folio, PMD and VMA.
> */
> -static inline void set_pte_at(struct mm_struct *mm, unsigned long addr,
> - pte_t *ptep, pte_t entry)
> +static inline void set_ptes(struct mm_struct *mm, unsigned long addr,
> + pte_t *ptep, pte_t entry, unsigned int nr)
> {
> if (pte_present(entry))
> entry = clear_pte_bit(entry, __pgprot(_PAGE_UNUSED));
> - if (mm_has_pgste(mm))
> - ptep_set_pte_at(mm, addr, ptep, entry);
> - else
> - set_pte(ptep, entry);
> + if (mm_has_pgste(mm)) {
> + for (;;) {
> + ptep_set_pte_at(mm, addr, ptep, entry);

There might be room for additional optimization here, regarding the
preempt_disable/enable() in ptep_set_pte_at(), i.e. move it out of
ptep_set_pte_at() and do it only once in this loop.

We could add that later with an add-on patch, but for this series it
all looks good.

Reviewed-by: Gerald Schaefer <[email protected]>