powerpc/mm/book3s/radix: Add mapping statistics

Add statistics that show how memory is mapped within the kernel linear mapping.
This is similar to commit 37cd944c8d ("s390/pgtable: add mapping statistics")

We don't do this with Hash translation mode. Hash uses one size (mmu_linear_psize)
to map the kernel linear mapping and we print the linear psize during boot as
below.

"Page orders: linear mapping = 24, virtual = 16, io = 16, vmemmap = 24"

A sample output looks like:

DirectMap4k:           0 kB
DirectMap64k:       18432 kB
DirectMap2M:     1030144 kB
DirectMap1G:    11534336 kB

Signed-off-by: Aneesh Kumar K.V <aneesh.kumar@linux.ibm.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
This commit is contained in:
Aneesh Kumar K.V
2018-08-13 11:14:57 +05:30
committed by Michael Ellerman
parent 241b5f7ffc
commit a2dc009afa
4 changed files with 46 additions and 5 deletions

View File

@@ -227,4 +227,11 @@ static inline void __pte_free_tlb(struct mmu_gather *tlb, pgtable_t table,
#define check_pgt_cache() do { } while (0)
extern atomic_long_t direct_pages_count[MMU_PAGE_COUNT];
static inline void update_page_count(int psize, long count)
{
if (IS_ENABLED(CONFIG_PROC_FS))
atomic_long_add(count, &direct_pages_count[psize]);
}
#endif /* _ASM_POWERPC_BOOK3S_64_PGALLOC_H */

View File

@@ -32,6 +32,9 @@
#define RADIX_PUD_BAD_BITS 0x60000000000000e0UL
#define RADIX_PGD_BAD_BITS 0x60000000000000e0UL
#define RADIX_PMD_SHIFT (PAGE_SHIFT + RADIX_PTE_INDEX_SIZE)
#define RADIX_PUD_SHIFT (RADIX_PMD_SHIFT + RADIX_PMD_INDEX_SIZE)
#define RADIX_PGD_SHIFT (RADIX_PUD_SHIFT + RADIX_PUD_INDEX_SIZE)
/*
* Size of EA range mapped by our pagetables.
*/