linux-kernel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Pavel Tatashin <pasha.tatashin@oracle.com>
To: steven.sistare@oracle.com, daniel.m.jordan@oracle.com,
	akpm@linux-foundation.org, mgorman@techsingularity.net,
	mhocko@suse.com, linux-mm@kvack.org,
	linux-kernel@vger.kernel.org, gregkh@linuxfoundation.org,
	vbabka@suse.cz, bharata@linux.vnet.ibm.com, tglx@linutronix.de,
	mingo@redhat.com, hpa@zytor.com, x86@kernel.org,
	dan.j.williams@intel.com, kirill.shutemov@linux.intel.com,
	bhe@redhat.com, alexander.levin@microsoft.com
Subject: [v6 4/6] mm/memory_hotplug: optimize probe routine
Date: Tue,  3 Apr 2018 14:16:41 -0400	[thread overview]
Message-ID: <20180403181643.28127-5-pasha.tatashin@oracle.com> (raw)
In-Reply-To: <20180403181643.28127-1-pasha.tatashin@oracle.com>

When memory is hotplugged pages_correctly_reserved() is called to verify
that the added memory is present, this routine traverses through every
struct page and verifies that PageReserved() is set. This is a slow
operation especially if a large amount of memory is added.

Instead of checking every page, it is enough to simply check that the
section is present, has mapping (struct page array is allocated), and the
mapping is online.

In addition, we should not excpect that probe routine sets flags in struct
page, as the struct pages have not yet been initialized. The initialization
should be done in __init_single_page(), the same as during boot.

Signed-off-by: Pavel Tatashin <pasha.tatashin@oracle.com>
Reviewed-by: Ingo Molnar <mingo@kernel.org>
Acked-by: Michal Hocko <mhocko@suse.com>
---
 drivers/base/memory.c | 36 ++++++++++++++++++++----------------
 1 file changed, 20 insertions(+), 16 deletions(-)

diff --git a/drivers/base/memory.c b/drivers/base/memory.c
index fe4b24f05f6a..deb3f029b451 100644
--- a/drivers/base/memory.c
+++ b/drivers/base/memory.c
@@ -187,13 +187,14 @@ int memory_isolate_notify(unsigned long val, void *v)
 }
 
 /*
- * The probe routines leave the pages reserved, just as the bootmem code does.
- * Make sure they're still that way.
+ * The probe routines leave the pages uninitialized, just as the bootmem code
+ * does. Make sure we do not access them, but instead use only information from
+ * within sections.
  */
-static bool pages_correctly_reserved(unsigned long start_pfn)
+static bool pages_correctly_probed(unsigned long start_pfn)
 {
-	int i, j;
-	struct page *page;
+	unsigned long section_nr = pfn_to_section_nr(start_pfn);
+	unsigned long section_nr_end = section_nr + sections_per_block;
 	unsigned long pfn = start_pfn;
 
 	/*
@@ -201,21 +202,24 @@ static bool pages_correctly_reserved(unsigned long start_pfn)
 	 * SPARSEMEM_VMEMMAP. We lookup the page once per section
 	 * and assume memmap is contiguous within each section
 	 */
-	for (i = 0; i < sections_per_block; i++, pfn += PAGES_PER_SECTION) {
+	for (; section_nr < section_nr_end; section_nr++) {
 		if (WARN_ON_ONCE(!pfn_valid(pfn)))
 			return false;
-		page = pfn_to_page(pfn);
-
-		for (j = 0; j < PAGES_PER_SECTION; j++) {
-			if (PageReserved(page + j))
-				continue;
-
-			printk(KERN_WARNING "section number %ld page number %d "
-				"not reserved, was it already online?\n",
-				pfn_to_section_nr(pfn), j);
 
+		if (!present_section_nr(section_nr)) {
+			pr_warn("section %ld pfn[%lx, %lx) not present",
+				section_nr, pfn, pfn + PAGES_PER_SECTION);
+			return false;
+		} else if (!valid_section_nr(section_nr)) {
+			pr_warn("section %ld pfn[%lx, %lx) no valid memmap",
+				section_nr, pfn, pfn + PAGES_PER_SECTION);
+			return false;
+		} else if (online_section_nr(section_nr)) {
+			pr_warn("section %ld pfn[%lx, %lx) is already online",
+				section_nr, pfn, pfn + PAGES_PER_SECTION);
 			return false;
 		}
+		pfn += PAGES_PER_SECTION;
 	}
 
 	return true;
@@ -237,7 +241,7 @@ memory_block_action(unsigned long phys_index, unsigned long action, int online_t
 
 	switch (action) {
 	case MEM_ONLINE:
-		if (!pages_correctly_reserved(start_pfn))
+		if (!pages_correctly_probed(start_pfn))
 			return -EBUSY;
 
 		ret = online_pages(start_pfn, nr_pages, online_type);
-- 
2.16.3

  parent reply	other threads:[~2018-04-03 18:16 UTC|newest]

Thread overview: 7+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2018-04-03 18:16 [v6 0/6] optimize memory hotplug Pavel Tatashin
2018-04-03 18:16 ` [v6 1/6] mm/memory_hotplug: enforce block size aligned range check Pavel Tatashin
2018-04-03 18:16 ` [v6 2/6] x86/mm/memory_hotplug: determine block size based on the end of boot memory Pavel Tatashin
2018-04-03 18:16 ` [v6 3/6] mm: add uninitialized struct page poisoning sanity checking Pavel Tatashin
2018-04-03 18:16 ` Pavel Tatashin [this message]
2018-04-03 18:16 ` [v6 5/6] mm/memory_hotplug: don't read nid from struct page during hotplug Pavel Tatashin
2018-04-03 18:16 ` [v6 6/6] mm/memory_hotplug: optimize memory hotplug Pavel Tatashin

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=20180403181643.28127-5-pasha.tatashin@oracle.com \
    --to=pasha.tatashin@oracle.com \
    --cc=akpm@linux-foundation.org \
    --cc=alexander.levin@microsoft.com \
    --cc=bharata@linux.vnet.ibm.com \
    --cc=bhe@redhat.com \
    --cc=dan.j.williams@intel.com \
    --cc=daniel.m.jordan@oracle.com \
    --cc=gregkh@linuxfoundation.org \
    --cc=hpa@zytor.com \
    --cc=kirill.shutemov@linux.intel.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-mm@kvack.org \
    --cc=mgorman@techsingularity.net \
    --cc=mhocko@suse.com \
    --cc=mingo@redhat.com \
    --cc=steven.sistare@oracle.com \
    --cc=tglx@linutronix.de \
    --cc=vbabka@suse.cz \
    --cc=x86@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;
as well as URLs for NNTP newsgroup(s).