sparc64 mm: Fix base TSB sizing when hugetlb pages are used
do_sparc64_fault() calculates both the base and huge page RSS sizes and
uses this information in calls to tsb_grow(). The calculation for base
page TSB size is not correct if the task uses hugetlb pages. hugetlb
pages are not accounted for in RSS, therefore the call to get_mm_rss(mm)
does not include hugetlb pages. However, the number of pages based on
huge_pte_count (which does include hugetlb pages) is subtracted from
this value. This will result in an artificially small and often negative
RSS calculation. The base TSB size is then often set to max_tsb_size
as the passed RSS is unsigned, so a negative value looks really big.
THP pages are also accounted for in huge_pte_count, and THP pages are
accounted for in RSS so the calculation in do_sparc64_fault() is correct
if a task only uses THP pages.
A single huge_pte_count is not sufficient for TSB sizing if both hugetlb
and THP pages can be used. Instead of a single counter, use two: one
for hugetlb and one for THP.
Signed-off-by: Mike Kravetz <mike.kravetz@oracle.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
diff --git a/arch/sparc/include/asm/mmu_64.h b/arch/sparc/include/asm/mmu_64.h
index 70067ce..f7de0db 100644
--- a/arch/sparc/include/asm/mmu_64.h
+++ b/arch/sparc/include/asm/mmu_64.h
@@ -92,7 +92,8 @@
typedef struct {
spinlock_t lock;
unsigned long sparc64_ctx_val;
- unsigned long huge_pte_count;
+ unsigned long hugetlb_pte_count;
+ unsigned long thp_pte_count;
struct tsb_config tsb_block[MM_NUM_TSBS];
struct hv_tsb_descr tsb_descr[MM_NUM_TSBS];
} mm_context_t;
diff --git a/arch/sparc/mm/fault_64.c b/arch/sparc/mm/fault_64.c
index 6c43b92..3a16ba0 100644
--- a/arch/sparc/mm/fault_64.c
+++ b/arch/sparc/mm/fault_64.c
@@ -476,14 +476,14 @@
up_read(&mm->mmap_sem);
mm_rss = get_mm_rss(mm);
-#if defined(CONFIG_HUGETLB_PAGE) || defined(CONFIG_TRANSPARENT_HUGEPAGE)
- mm_rss -= (mm->context.huge_pte_count * (HPAGE_SIZE / PAGE_SIZE));
+#if defined(CONFIG_TRANSPARENT_HUGEPAGE)
+ mm_rss -= (mm->context.thp_pte_count * (HPAGE_SIZE / PAGE_SIZE));
#endif
if (unlikely(mm_rss >
mm->context.tsb_block[MM_TSB_BASE].tsb_rss_limit))
tsb_grow(mm, MM_TSB_BASE, mm_rss);
#if defined(CONFIG_HUGETLB_PAGE) || defined(CONFIG_TRANSPARENT_HUGEPAGE)
- mm_rss = mm->context.huge_pte_count;
+ mm_rss = mm->context.hugetlb_pte_count + mm->context.thp_pte_count;
if (unlikely(mm_rss >
mm->context.tsb_block[MM_TSB_HUGE].tsb_rss_limit)) {
if (mm->context.tsb_block[MM_TSB_HUGE].tsb)
diff --git a/arch/sparc/mm/hugetlbpage.c b/arch/sparc/mm/hugetlbpage.c
index ba52e64..d69b66e 100644
--- a/arch/sparc/mm/hugetlbpage.c
+++ b/arch/sparc/mm/hugetlbpage.c
@@ -180,7 +180,7 @@
unsigned long nptes;
if (!pte_present(*ptep) && pte_present(entry))
- mm->context.huge_pte_count++;
+ mm->context.hugetlb_pte_count++;
addr &= HPAGE_MASK;
@@ -212,7 +212,7 @@
entry = *ptep;
if (pte_present(entry))
- mm->context.huge_pte_count--;
+ mm->context.hugetlb_pte_count--;
addr &= HPAGE_MASK;
nptes = 1 << HUGETLB_PAGE_ORDER;
diff --git a/arch/sparc/mm/init_64.c b/arch/sparc/mm/init_64.c
index aec508e..aef153f 100644
--- a/arch/sparc/mm/init_64.c
+++ b/arch/sparc/mm/init_64.c
@@ -346,7 +346,8 @@
spin_lock_irqsave(&mm->context.lock, flags);
#if defined(CONFIG_HUGETLB_PAGE) || defined(CONFIG_TRANSPARENT_HUGEPAGE)
- if (mm->context.huge_pte_count && is_hugetlb_pte(pte))
+ if ((mm->context.hugetlb_pte_count || mm->context.thp_pte_count) &&
+ is_hugetlb_pte(pte))
__update_mmu_tsb_insert(mm, MM_TSB_HUGE, REAL_HPAGE_SHIFT,
address, pte_val(pte));
else
diff --git a/arch/sparc/mm/tlb.c b/arch/sparc/mm/tlb.c
index f81cd97..3659d37 100644
--- a/arch/sparc/mm/tlb.c
+++ b/arch/sparc/mm/tlb.c
@@ -175,9 +175,9 @@
if ((pmd_val(pmd) ^ pmd_val(orig)) & _PAGE_PMD_HUGE) {
if (pmd_val(pmd) & _PAGE_PMD_HUGE)
- mm->context.huge_pte_count++;
+ mm->context.thp_pte_count++;
else
- mm->context.huge_pte_count--;
+ mm->context.thp_pte_count--;
/* Do not try to allocate the TSB hash table if we
* don't have one already. We have various locks held
diff --git a/arch/sparc/mm/tsb.c b/arch/sparc/mm/tsb.c
index a0604a4..6725ed4 100644
--- a/arch/sparc/mm/tsb.c
+++ b/arch/sparc/mm/tsb.c
@@ -470,7 +470,7 @@
int init_new_context(struct task_struct *tsk, struct mm_struct *mm)
{
#if defined(CONFIG_HUGETLB_PAGE) || defined(CONFIG_TRANSPARENT_HUGEPAGE)
- unsigned long huge_pte_count;
+ unsigned long total_huge_pte_count;
#endif
unsigned int i;
@@ -479,12 +479,14 @@
mm->context.sparc64_ctx_val = 0UL;
#if defined(CONFIG_HUGETLB_PAGE) || defined(CONFIG_TRANSPARENT_HUGEPAGE)
- /* We reset it to zero because the fork() page copying
+ /* We reset them to zero because the fork() page copying
* will re-increment the counters as the parent PTEs are
* copied into the child address space.
*/
- huge_pte_count = mm->context.huge_pte_count;
- mm->context.huge_pte_count = 0;
+ total_huge_pte_count = mm->context.hugetlb_pte_count +
+ mm->context.thp_pte_count;
+ mm->context.hugetlb_pte_count = 0;
+ mm->context.thp_pte_count = 0;
#endif
/* copy_mm() copies over the parent's mm_struct before calling
@@ -500,8 +502,8 @@
tsb_grow(mm, MM_TSB_BASE, get_mm_rss(mm));
#if defined(CONFIG_HUGETLB_PAGE) || defined(CONFIG_TRANSPARENT_HUGEPAGE)
- if (unlikely(huge_pte_count))
- tsb_grow(mm, MM_TSB_HUGE, huge_pte_count);
+ if (unlikely(total_huge_pte_count))
+ tsb_grow(mm, MM_TSB_HUGE, total_huge_pte_count);
#endif
if (unlikely(!mm->context.tsb_block[MM_TSB_BASE].tsb))