linux-kernel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Nadav Amit <namit@vmware.com>
To: Arnd Bergmann <arnd@arndb.de>, <gregkh@linuxfoundation.org>
Cc: Xavier Deguillard <xdeguillard@vmware.com>,
	<linux-kernel@vger.kernel.org>, Nadav Amit <namit@vmware.com>
Subject: [PATCH v3 07/20] vmw_balloon: treat all refused pages equally
Date: Wed, 26 Sep 2018 12:13:23 -0700	[thread overview]
Message-ID: <20180926191336.101885-8-namit@vmware.com> (raw)
In-Reply-To: <20180926191336.101885-1-namit@vmware.com>

Currently, when the hypervisor rejects a page during lock operation, the
VM treats pages differently according to the error-code: in certain
cases the page is immediately freed, and in others it is put on a
rejection list and only freed later.

The behavior does not make too much sense. If the page is freed
immediately it is very likely to be used again in the next batch of
allocations, and be rejected again.

In addition, for support of compaction and OOM notifiers, we wish to
separate the logic that communicates with the hypervisor (as well as
analyzes the status of each page) from the logic that allocates or free
pages.

Treat all errors the same way, queuing the pages on the refuse list.
Move to the next allocation size (4k) when too many pages are refused.
Free the refused pages when moving to the next size to avoid situations
in which too much memory is waiting to be freed on the refused list.

Reviewed-by: Xavier Deguillard <xdeguillard@vmware.com>
Signed-off-by: Nadav Amit <namit@vmware.com>
---
 drivers/misc/vmw_balloon.c | 52 +++++++++++++++++++++-----------------
 1 file changed, 29 insertions(+), 23 deletions(-)

diff --git a/drivers/misc/vmw_balloon.c b/drivers/misc/vmw_balloon.c
index 96dde120bbd5..4e067d269706 100644
--- a/drivers/misc/vmw_balloon.c
+++ b/drivers/misc/vmw_balloon.c
@@ -543,29 +543,13 @@ static int vmballoon_lock(struct vmballoon *b, unsigned int num_pages,
 		/* Error occurred */
 		STATS_INC(b->stats.refused_alloc[is_2m_pages]);
 
-		switch (status) {
-		case VMW_BALLOON_ERROR_PPN_PINNED:
-		case VMW_BALLOON_ERROR_PPN_INVALID:
-			/*
-			 * Place page on the list of non-balloonable pages
-			 * and retry allocation, unless we already accumulated
-			 * too many of them, in which case take a breather.
-			 */
-			if (page_size->n_refused_pages
-					< VMW_BALLOON_MAX_REFUSED) {
-				list_add(&p->lru, &page_size->refused_pages);
-				page_size->n_refused_pages++;
-				break;
-			}
-			/* Fallthrough */
-		case VMW_BALLOON_ERROR_RESET:
-		case VMW_BALLOON_ERROR_PPN_NOTNEEDED:
-			vmballoon_free_page(p, is_2m_pages);
-			break;
-		default:
-			/* This should never happen */
-			WARN_ON_ONCE(true);
-		}
+		/*
+		 * Place page on the list of non-balloonable pages
+		 * and retry allocation, unless we already accumulated
+		 * too many of them, in which case take a breather.
+		 */
+		list_add(&p->lru, &page_size->refused_pages);
+		page_size->n_refused_pages++;
 	}
 
 	return batch_status == VMW_BALLOON_SUCCESS ? 0 : -EIO;
@@ -712,9 +696,31 @@ static void vmballoon_inflate(struct vmballoon *b)
 
 		vmballoon_add_page(b, num_pages++, page);
 		if (num_pages == b->batch_max_pages) {
+			struct vmballoon_page_size *page_size =
+					&b->page_sizes[is_2m_pages];
+
 			error = vmballoon_lock(b, num_pages, is_2m_pages);
 
 			num_pages = 0;
+
+			/*
+			 * Stop allocating this page size if we already
+			 * accumulated too many pages that the hypervisor
+			 * refused.
+			 */
+			if (page_size->n_refused_pages >=
+			    VMW_BALLOON_MAX_REFUSED) {
+				if (!is_2m_pages)
+					break;
+
+				/*
+				 * Release the refused pages as we move to 4k
+				 * pages.
+				 */
+				vmballoon_release_refused_pages(b, true);
+				is_2m_pages = true;
+			}
+
 			if (error)
 				break;
 		}
-- 
2.17.1


  parent reply	other threads:[~2018-09-26 19:16 UTC|newest]

Thread overview: 27+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2018-09-26 19:13 [PATCH v3 00/20] vmw_balloon: compaction, shrinker, 64-bit, etc Nadav Amit
2018-09-26 19:13 ` [PATCH v3 01/20] vmw_balloon: handle commands in a single function Nadav Amit
2018-09-26 19:13 ` [PATCH v3 02/20] vmw_balloon: unify commands tracing and stats Nadav Amit
2018-09-26 19:13 ` [PATCH v3 03/20] vmw_balloon: merge send_lock and send_unlock path Nadav Amit
2018-09-26 19:13 ` [PATCH v3 04/20] vmw_balloon: simplifying batch access Nadav Amit
2018-09-26 19:13 ` [PATCH v3 05/20] vmw_balloon: remove sleeping allocations Nadav Amit
2018-09-26 19:13 ` [PATCH v3 06/20] vmw_balloon: change batch/single lock abstractions Nadav Amit
2018-09-26 19:13 ` Nadav Amit [this message]
2018-09-26 19:13 ` [PATCH v3 08/20] vmw_balloon: rename VMW_BALLOON_2M_SHIFT to VMW_BALLOON_2M_ORDER Nadav Amit
2018-09-26 19:13 ` [PATCH v3 09/20] vmw_balloon: refactor change size from vmballoon_work Nadav Amit
2018-09-26 19:13 ` [PATCH v3 10/20] vmw_balloon: simplify vmballoon_send_get_target() Nadav Amit
2018-09-26 19:13 ` [PATCH v3 11/20] vmw_balloon: stats rework Nadav Amit
2018-09-26 19:13 ` [PATCH v3 12/20] vmw_balloon: rework the inflate and deflate loops Nadav Amit
2018-09-26 19:13 ` [PATCH v3 13/20] vmw_balloon: general style cleanup Nadav Amit
2018-09-26 19:13 ` [PATCH v3 14/20] vmw_balloon: add reset stat Nadav Amit
2018-09-26 19:13 ` [PATCH v3 15/20] mm/balloon_compaction: suppress allocation warnings Nadav Amit
2018-09-26 19:13 ` [PATCH v3 16/20] mm/balloon_compaction: list interfaces Nadav Amit
2018-09-28 19:48   ` Nadav Amit
2018-10-17  2:42     ` Nadav Amit
2018-09-26 19:13 ` [PATCH v3 17/20] vmw_balloon: compaction support Nadav Amit
2018-09-26 19:13 ` [PATCH v3 18/20] vmw_balloon: support 64-bit memory limit Nadav Amit
2018-09-26 19:13 ` [PATCH v3 19/20] vmw_balloon: memory shrinker Nadav Amit
2018-09-26 19:13 ` [PATCH v3 20/20] vmw_balloon: split refused pages Nadav Amit
2018-10-30 16:32 ` [PATCH v3 00/20] vmw_balloon: compaction, shrinker, 64-bit, etc Nadav Amit
2018-10-30 16:51   ` gregkh
2018-10-30 16:52     ` Nadav Amit
2018-10-30 17:05       ` gregkh

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=20180926191336.101885-8-namit@vmware.com \
    --to=namit@vmware.com \
    --cc=arnd@arndb.de \
    --cc=gregkh@linuxfoundation.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=xdeguillard@vmware.com \
    /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).