mirror of https://lore.kernel.org/lkml/
 help / color / mirror / Atom feed
From: Hugh Dickins <hugh@veritas.com>
To: Andrew Morton <akpm@osdl.org>
Cc: linux-kernel@vger.kernel.org
Subject: [PATCH 05/21] mm: zap_pte out of line
Date: Thu, 13 Oct 2005 01:48:30 +0100 (BST)	[thread overview]
Message-ID: <Pine.LNX.4.61.0510130147560.4060@goblin.wat.veritas.com> (raw)
In-Reply-To: <Pine.LNX.4.61.0510130143240.4060@goblin.wat.veritas.com>

There used to be just one call to zap_pte, but it shouldn't be inline
now there are two.  Check for the common case pte_none before calling,
and move its rss accounting up into install_page or install_file_pte -
which helps the next patch.

Signed-off-by: Hugh Dickins <hugh@veritas.com>
---

 mm/fremap.c |   19 +++++++++----------
 1 files changed, 9 insertions(+), 10 deletions(-)

--- mm04/mm/fremap.c	2005-10-11 12:16:50.000000000 +0100
+++ mm05/mm/fremap.c	2005-10-11 23:54:15.000000000 +0100
@@ -20,34 +20,32 @@
 #include <asm/cacheflush.h>
 #include <asm/tlbflush.h>
 
-static inline void zap_pte(struct mm_struct *mm, struct vm_area_struct *vma,
+static int zap_pte(struct mm_struct *mm, struct vm_area_struct *vma,
 			unsigned long addr, pte_t *ptep)
 {
 	pte_t pte = *ptep;
+	struct page *page = NULL;
 
-	if (pte_none(pte))
-		return;
 	if (pte_present(pte)) {
 		unsigned long pfn = pte_pfn(pte);
-		struct page *page;
-
 		flush_cache_page(vma, addr, pfn);
 		pte = ptep_clear_flush(vma, addr, ptep);
 		if (unlikely(!pfn_valid(pfn))) {
 			print_bad_pte(vma, pte, addr);
-			return;
+			goto out;
 		}
 		page = pfn_to_page(pfn);
 		if (pte_dirty(pte))
 			set_page_dirty(page);
 		page_remove_rmap(page);
 		page_cache_release(page);
-		dec_mm_counter(mm, file_rss);
 	} else {
 		if (!pte_file(pte))
 			free_swap_and_cache(pte_to_swp_entry(pte));
 		pte_clear(mm, addr, ptep);
 	}
+out:
+	return !!page;
 }
 
 /*
@@ -93,9 +91,9 @@ int install_page(struct mm_struct *mm, s
 	if (!page->mapping || page->index >= size)
 		goto err_unlock;
 
-	zap_pte(mm, vma, addr, pte);
+	if (pte_none(*pte) || !zap_pte(mm, vma, addr, pte))
+		inc_mm_counter(mm, file_rss);
 
-	inc_mm_counter(mm, file_rss);
 	flush_icache_page(vma, page);
 	set_pte_at(mm, addr, pte, mk_pte(page, prot));
 	page_add_file_rmap(page);
@@ -142,7 +140,8 @@ int install_file_pte(struct mm_struct *m
 	if (!pte)
 		goto err_unlock;
 
-	zap_pte(mm, vma, addr, pte);
+	if (!pte_none(*pte) && zap_pte(mm, vma, addr, pte))
+		dec_mm_counter(mm, file_rss);
 
 	set_pte_at(mm, addr, pte, pgoff_to_pte(pgoff));
 	pte_val = *pte;

  parent reply	other threads:[~2005-10-13  0:49 UTC|newest]

Thread overview: 25+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2005-10-13  0:44 [PATCH 00/21] mm: page fault scalability nearer Hugh Dickins
2005-10-13  0:45 ` [PATCH 01/21] mm: copy_one_pte inc rss Hugh Dickins
2005-10-13  0:46 ` [PATCH 02/21] mm: zap_pte_range dec rss Hugh Dickins
2005-10-13  0:47 ` [PATCH 03/21] mm: do_swap_page race major Hugh Dickins
2005-10-13  0:47 ` [PATCH 04/21] mm: do_mremap current mm Hugh Dickins
2005-10-13  0:48 ` Hugh Dickins [this message]
2005-10-13  0:50 ` [PATCH 06/21] mm: update_hiwaters just in time Hugh Dickins
2005-10-13  1:17   ` Christoph Lameter
2005-10-13 19:18     ` Jay Lan
2005-10-13  0:52 ` [PATCH 07/21] mm: mm_struct hiwaters moved Hugh Dickins
2005-10-13  0:53 ` [PATCH 08/21] mm: ia64 use expand_upwards Hugh Dickins
2005-10-13  0:55 ` [PATCH 09/21] mm: init_mm without ptlock Hugh Dickins
2005-10-13  0:55 ` [PATCH 10/21] mm: ptd_alloc inline and out Hugh Dickins
2005-10-13  0:56 ` [PATCH 11/21] mm: ptd_alloc take ptlock Hugh Dickins
2005-10-13  0:59 ` [PATCH 12/21] mm: arches skip ptlock Hugh Dickins
2005-10-13  1:13 ` [PATCH 13/21] mm: page fault handler locking Hugh Dickins
2005-10-13  1:14 ` [PATCH 14/21] mm: pte_offset_map_lock loops Hugh Dickins
2005-10-13  1:16 ` [PATCH 15/21] mm: flush_tlb_range outside ptlock Hugh Dickins
2005-10-13  1:17 ` [PATCH 16/21] mm: unlink vma before pagetables Hugh Dickins
2005-10-13  1:18 ` [PATCH 17/21] mm: unmap_vmas with inner ptlock Hugh Dickins
2005-10-13  1:20 ` [PATCH 18/21] mm: xip_unmap ZERO_PAGE fix Hugh Dickins
2005-10-13 12:39   ` Carsten Otte
2005-10-13  1:21 ` [PATCH 19/21] mm: rmap with inner ptlock Hugh Dickins
2005-10-13  1:22 ` [PATCH 20/21] mm: kill check_user_page_readable Hugh Dickins
2005-10-13  1:23 ` [PATCH 21/21] mm: follow_page with inner ptlock Hugh Dickins

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=Pine.LNX.4.61.0510130147560.4060@goblin.wat.veritas.com \
    --to=hugh@veritas.com \
    --cc=akpm@osdl.org \
    --cc=linux-kernel@vger.kernel.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox

all inboxes | Powered by JetHome®