[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-Id: <8ecffbf990afd1c8ccc195a2ec321d55f0923908.1733305182.git.zhengqi.arch@bytedance.com>
Date: Wed, 4 Dec 2024 19:09:45 +0800
From: Qi Zheng <zhengqi.arch@...edance.com>
To: david@...hat.com,
jannh@...gle.com,
hughd@...gle.com,
willy@...radead.org,
muchun.song@...ux.dev,
vbabka@...nel.org,
peterx@...hat.com,
akpm@...ux-foundation.org
Cc: mgorman@...e.de,
catalin.marinas@....com,
will@...nel.org,
dave.hansen@...ux.intel.com,
luto@...nel.org,
peterz@...radead.org,
x86@...nel.org,
lorenzo.stoakes@...cle.com,
zokeefe@...gle.com,
rientjes@...gle.com,
linux-mm@...ck.org,
linux-kernel@...r.kernel.org,
Qi Zheng <zhengqi.arch@...edance.com>
Subject: [PATCH v4 05/11] mm: skip over all consecutive none ptes in do_zap_pte_range()
Skip over all consecutive none ptes in do_zap_pte_range(), which helps
optimize away need_resched() + force_break + incremental pte/addr
increments etc.
Suggested-by: David Hildenbrand <david@...hat.com>
Signed-off-by: Qi Zheng <zhengqi.arch@...edance.com>
---
mm/memory.c | 27 ++++++++++++++++++++-------
1 file changed, 20 insertions(+), 7 deletions(-)
diff --git a/mm/memory.c b/mm/memory.c
index abe07e6bdd1bb..7f8869a22b57c 100644
--- a/mm/memory.c
+++ b/mm/memory.c
@@ -1665,17 +1665,30 @@ static inline int do_zap_pte_range(struct mmu_gather *tlb,
{
pte_t ptent = ptep_get(pte);
int max_nr = (end - addr) / PAGE_SIZE;
+ int nr = 0;
- if (pte_none(ptent))
- return 1;
+ /* Skip all consecutive none ptes */
+ if (pte_none(ptent)) {
+ for (nr = 1; nr < max_nr; nr++) {
+ ptent = ptep_get(pte + nr);
+ if (!pte_none(ptent))
+ break;
+ }
+ max_nr -= nr;
+ if (!max_nr)
+ return nr;
+ pte += nr;
+ addr += nr * PAGE_SIZE;
+ }
if (pte_present(ptent))
- return zap_present_ptes(tlb, vma, pte, ptent, max_nr,
- addr, details, rss, force_flush,
- force_break);
+ nr += zap_present_ptes(tlb, vma, pte, ptent, max_nr, addr,
+ details, rss, force_flush, force_break);
+ else
+ nr += zap_nonpresent_ptes(tlb, vma, pte, ptent, max_nr, addr,
+ details, rss);
- return zap_nonpresent_ptes(tlb, vma, pte, ptent, max_nr, addr,
- details, rss);
+ return nr;
}
static unsigned long zap_pte_range(struct mmu_gather *tlb,
--
2.20.1
Powered by blists - more mailing lists