qemu-devel.nongnu.org archive mirror
 help / color / mirror / Atom feed
From: Kevin Wolf <kwolf@redhat.com>
To: qemu-devel@nongnu.org
Cc: kwolf@redhat.com
Subject: [Qemu-devel] [PULL 07/29] qcow2: Constant cache size in bytes
Date: Fri, 22 Aug 2014 16:51:31 +0200	[thread overview]
Message-ID: <1408719113-5316-8-git-send-email-kwolf@redhat.com> (raw)
In-Reply-To: <1408719113-5316-1-git-send-email-kwolf@redhat.com>

From: Max Reitz <mreitz@redhat.com>

Specifying the metadata cache sizes in clusters results in less clusters
(and much less bytes) covered for small cluster sizes and vice versa.
Using a constant byte size reduces this difference, and makes it
possible to manually specify the cache size in an easily comprehensible
unit.

Signed-off-by: Max Reitz <mreitz@redhat.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
---
 block/qcow2.c | 15 +++++++++++++--
 block/qcow2.h | 10 ++++++++--
 2 files changed, 21 insertions(+), 4 deletions(-)

diff --git a/block/qcow2.c b/block/qcow2.c
index 435e0e1..0f1be2e 100644
--- a/block/qcow2.c
+++ b/block/qcow2.c
@@ -470,6 +470,7 @@ static int qcow2_open(BlockDriverState *bs, QDict *options, int flags,
     uint64_t l1_vm_state_index;
     const char *opt_overlap_check;
     int overlap_check_template = 0;
+    uint64_t l2_cache_size, refcount_cache_size;
 
     ret = bdrv_pread(bs->file, 0, &header, sizeof(header));
     if (ret < 0) {
@@ -707,8 +708,18 @@ static int qcow2_open(BlockDriverState *bs, QDict *options, int flags,
     }
 
     /* alloc L2 table/refcount block cache */
-    s->l2_table_cache = qcow2_cache_create(bs, L2_CACHE_SIZE);
-    s->refcount_block_cache = qcow2_cache_create(bs, REFCOUNT_CACHE_SIZE);
+    l2_cache_size = DEFAULT_L2_CACHE_BYTE_SIZE / s->cluster_size;
+    if (l2_cache_size < MIN_L2_CACHE_SIZE) {
+        l2_cache_size = MIN_L2_CACHE_SIZE;
+    }
+
+    refcount_cache_size = l2_cache_size / DEFAULT_L2_REFCOUNT_SIZE_RATIO;
+    if (refcount_cache_size < MIN_REFCOUNT_CACHE_SIZE) {
+        refcount_cache_size = MIN_REFCOUNT_CACHE_SIZE;
+    }
+
+    s->l2_table_cache = qcow2_cache_create(bs, l2_cache_size);
+    s->refcount_block_cache = qcow2_cache_create(bs, refcount_cache_size);
     if (s->l2_table_cache == NULL || s->refcount_block_cache == NULL) {
         error_setg(errp, "Could not allocate metadata caches");
         ret = -ENOMEM;
diff --git a/block/qcow2.h b/block/qcow2.h
index b49424b..671783d 100644
--- a/block/qcow2.h
+++ b/block/qcow2.h
@@ -64,10 +64,16 @@
 #define MIN_CLUSTER_BITS 9
 #define MAX_CLUSTER_BITS 21
 
-#define L2_CACHE_SIZE 16
+#define MIN_L2_CACHE_SIZE 1 /* cluster */
 
 /* Must be at least 4 to cover all cases of refcount table growth */
-#define REFCOUNT_CACHE_SIZE 4
+#define MIN_REFCOUNT_CACHE_SIZE 4 /* clusters */
+
+#define DEFAULT_L2_CACHE_BYTE_SIZE 1048576 /* bytes */
+
+/* The refblock cache needs only a fourth of the L2 cache size to cover as many
+ * clusters */
+#define DEFAULT_L2_REFCOUNT_SIZE_RATIO 4
 
 #define DEFAULT_CLUSTER_SIZE 65536
 
-- 
1.8.3.1

  parent reply	other threads:[~2014-08-22 14:52 UTC|newest]

Thread overview: 32+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2014-08-22 14:51 [Qemu-devel] [PULL 00/29] Block patches Kevin Wolf
2014-08-22 14:51 ` [Qemu-devel] [PULL 01/29] block: Use g_new() & friends where that makes obvious sense Kevin Wolf
2014-08-22 14:51 ` [Qemu-devel] [PULL 02/29] block: Use g_new() & friends to avoid multiplying sizes Kevin Wolf
2014-08-22 14:51 ` [Qemu-devel] [PULL 03/29] qemu-io-cmds: g_renew() can't fail, bury dead error handling Kevin Wolf
2014-08-22 14:51 ` [Qemu-devel] [PULL 04/29] block: Drop some superfluous casts from void * Kevin Wolf
2014-08-22 14:51 ` [Qemu-devel] [PULL 05/29] runner: Add an argument for test duration Kevin Wolf
2014-08-22 14:51 ` [Qemu-devel] [PULL 06/29] runner: Kill a program under test by time-out Kevin Wolf
2014-08-22 14:51 ` Kevin Wolf [this message]
2014-08-22 14:51 ` [Qemu-devel] [PULL 08/29] qcow2: Use g_try_new0() for cache array Kevin Wolf
2014-08-22 14:51 ` [Qemu-devel] [PULL 09/29] qcow2: Add runtime options for cache sizes Kevin Wolf
2014-08-22 14:51 ` [Qemu-devel] [PULL 10/29] iotests: Add test for qcow2's cache options Kevin Wolf
2014-08-22 14:51 ` [Qemu-devel] [PULL 11/29] test-coroutine: test cost introduced by coroutine Kevin Wolf
2014-08-22 14:51 ` [Qemu-devel] [PULL 12/29] qemu-iotests: Fix 028 reference output for qed Kevin Wolf
2014-08-22 14:51 ` [Qemu-devel] [PULL 13/29] block: acquire AioContext in qmp_block_resize() Kevin Wolf
2014-08-22 14:51 ` [Qemu-devel] [PULL 14/29] virtio-blk: allow block_resize with dataplane Kevin Wolf
2014-08-22 14:51 ` [Qemu-devel] [PULL 15/29] virtio-blk: fix reference a pointer which might be freed Kevin Wolf
2014-08-22 14:51 ` [Qemu-devel] [PULL 16/29] block: Add bdrv_refresh_filename() Kevin Wolf
2014-08-22 14:51 ` [Qemu-devel] [PULL 17/29] blkdebug: Implement bdrv_refresh_filename() Kevin Wolf
2014-08-22 14:51 ` [Qemu-devel] [PULL 18/29] blkverify: " Kevin Wolf
2014-08-22 14:51 ` [Qemu-devel] [PULL 19/29] nbd: " Kevin Wolf
2014-08-22 14:51 ` [Qemu-devel] [PULL 20/29] quorum: " Kevin Wolf
2014-08-22 14:51 ` [Qemu-devel] [PULL 21/29] iotests: Add test for image filename construction Kevin Wolf
2014-08-22 14:51 ` [Qemu-devel] [PULL 22/29] block/vvfat.c: remove debugging code to reinit stderr if NULL Kevin Wolf
2014-08-22 14:51 ` [Qemu-devel] [PULL 23/29] block/iscsi: fix memory corruption on iscsi resize Kevin Wolf
2014-08-22 14:51 ` [Qemu-devel] [PULL 24/29] raw-posix: fix O_DIRECT short reads Kevin Wolf
2014-08-22 14:51 ` [Qemu-devel] [PULL 25/29] qemu-iotests: add test case 101 for short file I/O Kevin Wolf
2014-08-22 14:51 ` [Qemu-devel] [PULL 26/29] blkdebug: Delete BH in bdrv_aio_cancel Kevin Wolf
2014-08-22 14:51 ` [Qemu-devel] [PULL 27/29] vmdk: Use bdrv_nb_sectors() where sectors, not bytes are wanted Kevin Wolf
2014-08-22 14:51 ` [Qemu-devel] [PULL 28/29] qemu-img: Allow source cache mode specification Kevin Wolf
2014-08-22 14:51 ` [Qemu-devel] [PULL 29/29] qemu-img: Allow cache mode specification for amend Kevin Wolf
2014-08-22 15:09 ` [Qemu-devel] [PULL 00/29] Block patches Daniel H Barboza
2014-08-22 16:23 ` Peter Maydell

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=1408719113-5316-8-git-send-email-kwolf@redhat.com \
    --to=kwolf@redhat.com \
    --cc=qemu-devel@nongnu.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).