openkylin
/
qemu
mirror of https://gitee.com/openkylin/qemu.git
9
0
Fork 0
Code Issues Projects Releases Wiki Activity
qemu/block
History
Alberto Garcia 72e80b8901 qcow2: use one single memory block for the L2/refcount cache tables 
The qcow2 L2/refcount cache contains one separate table for each cache
entry. Doing one allocation per table adds unnecessary overhead and it
also requires us to store the address of each table separately.

Since the size of the cache is constant during its lifetime, it's
better to have an array that contains all the tables using one single
allocation.

In my tests measuring freshly created caches with sizes 128MB (L2) and
32MB (refcount) this uses around 10MB of RAM less.

Signed-off-by: Alberto Garcia <berto@igalia.com>
Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com>
Reviewed-by: Max Reitz <mreitz@redhat.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
 		2015-05-22 17:08:01 +02:00
..
Makefile.objs block: move I/O request processing to block/io.c 2015-04-28 15:36:17 +02:00
accounting.c block: add accounting for merged requests 2015-02-06 17:24:21 +01:00
archipelago.c block: remove superfluous '\n' around error_report/error_setg 2015-03-10 08:15:33 +03:00
backup.c block: Ensure consistent bitmap function prototypes 2015-04-28 15:36:10 +02:00
blkdebug.c blkdebug: Add bdrv_truncate() 2015-04-28 15:36:09 +02:00
blkverify.c block: Rename BlockDriverCompletionFunc to BlockCompletionFunc 2014-10-20 13:41:27 +02:00
block-backend.c block-backend: Expose bdrv_write_zeroes() 2015-04-28 15:36:08 +02:00
bochs.c block: Use g_new() & friends to avoid multiplying sizes 2014-08-20 11:51:28 +02:00
cloop.c cloop: Handle failure for potentially large allocations 2014-08-15 15:07:15 +02:00
commit.c block: let commit blockjob run in BDS AioContext 2014-11-03 11:41:49 +00:00
curl.c block/curl: Improve type safety of s->timeout. 2014-11-03 11:41:47 +00:00
dmg.c block/dmg: improve zeroes handling 2015-02-06 17:24:21 +01:00
gluster.c block: don't convert file size to sector size 2014-09-12 15:43:06 +02:00
io.c block: get_block_status: use "else" when testing the opposite condition 2015-05-22 09:37:33 +01:00
iscsi.c block/iscsi: use the allocationmap also if cache.direct=on 2015-04-28 15:36:10 +02:00
linux-aio.c linux-aio: simplify removal of completed iocbs from the list 2014-12-12 16:57:55 +00:00
mirror.c block/mirror: Always call block_job_sleep_ns() 2015-04-28 15:36:11 +02:00
nbd-client.c nbd: Set block size to BDRV_SECTOR_SIZE 2015-03-18 12:07:01 +01:00
nbd-client.h nbd: Set block size to BDRV_SECTOR_SIZE 2015-03-18 12:07:01 +01:00
nbd.c nbd: Fix nbd_establish_connection()'s return value 2015-03-18 12:05:38 +01:00
nfs.c block/nfs: Add create_opts 2014-12-10 10:31:19 +01:00
null.c block/null: Support reopen 2015-04-28 15:36:09 +02:00
parallels.c block/parallels: improve image writing performance further 2015-05-22 09:37:32 +01:00
qapi.c qobject: Clean up around qtype_code 2015-05-11 08:59:07 -04:00
qcow.c block: use bdrv_get_device_or_node_name() in error messages 2015-04-28 15:36:09 +02:00
qcow2-cache.c qcow2: use one single memory block for the L2/refcount cache tables 2015-05-22 17:08:01 +02:00
qcow2-cluster.c qcow2: use one single memory block for the L2/refcount cache tables 2015-05-22 17:08:01 +02:00
qcow2-refcount.c qcow2: use one single memory block for the L2/refcount cache tables 2015-05-22 17:08:01 +02:00
qcow2-snapshot.c savevm: create snapshot failed when id_str already exists 2015-04-28 15:36:08 +02:00
qcow2.c block: add 'node-name' field to BLOCK_IMAGE_CORRUPTED 2015-04-28 15:36:09 +02:00
qcow2.h qcow2: use one single memory block for the L2/refcount cache tables 2015-05-22 17:08:01 +02:00
qed-check.c block: Use g_new() & friends to avoid multiplying sizes 2014-08-20 11:51:28 +02:00
qed-cluster.c Use glib memory allocation and free functions 2011-08-20 23:01:08 -05:00
qed-gencb.c block: Rename BlockDriverCompletionFunc to BlockCompletionFunc 2014-10-20 13:41:27 +02:00
qed-l2-cache.c qed: do not evict in-use L2 table cache entries 2012-03-12 15:14:06 +01:00
qed-table.c block: Rename BlockDriverCompletionFunc to BlockCompletionFunc 2014-10-20 13:41:27 +02:00
qed.c block: use bdrv_get_device_or_node_name() in error messages 2015-04-28 15:36:09 +02:00
qed.h qed: Really remove unused field QEDAIOCB.finished 2015-02-06 17:24:21 +01:00
quorum.c block: add bdrv_get_device_or_node_name() 2015-04-28 15:36:09 +02:00
raw-aio.h linux-aio: drop return code from laio_io_unplug and ioq_submit 2014-12-12 16:57:55 +00:00
raw-posix.c block: align bounce buffers to page 2015-05-22 09:37:33 +01:00
raw-win32.c block: Remove "growable" from BDS 2015-02-16 15:07:19 +00:00
raw_bsd.c block: Add driver methods to probe blocksizes and geometry 2015-03-10 14:02:22 +01:00
rbd.c Convert (ffs(val) - 1) to ctz32(val) 2015-04-28 15:36:08 +02:00
sheepdog.c sheepdog: fix resource leak with sd_snapshot_create 2015-05-08 14:11:10 +03:00
snapshot.c block: use bdrv_get_device_or_node_name() in error messages 2015-04-28 15:36:09 +02:00
ssh.c ssh: Don't crash if either host or path is not specified. 2014-10-03 10:30:33 +01:00
stream.c block: let stream blockjob run in BDS AioContext 2014-11-03 11:41:49 +00:00
vdi.c block: use bdrv_get_device_or_node_name() in error messages 2015-04-28 15:36:09 +02:00
vhdx-endian.c block: VHDX endian fixes 2014-08-15 15:07:14 +02:00
vhdx-log.c block: Drop some superfluous casts from void * 2014-08-20 11:51:28 +02:00
vhdx.c block: use bdrv_get_device_or_node_name() in error messages 2015-04-28 15:36:09 +02:00
vhdx.h block: vhdx - update PAYLOAD_BLOCK_UNMAPPED value to match 1.00 spec 2014-12-12 15:42:22 +00:00
vmdk.c vmdk: Fix overflow if l1_size is 0x20000000 2015-05-22 17:08:01 +02:00
vpc.c block: use bdrv_get_device_or_node_name() in error messages 2015-04-28 15:36:09 +02:00
vvfat.c block: use bdrv_get_device_or_node_name() in error messages 2015-04-28 15:36:09 +02:00
win32-aio.c block: Rename BlockDriverCompletionFunc to BlockCompletionFunc 2014-10-20 13:41:27 +02:00
write-threshold.c block: Fix block-set-write-threshold not to use funky error class 2015-03-16 17:07:25 +01:00