[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
[PULL 13/17] qcow2: Use bs->bl.request_alignment when updating an L1 ent
From: |
Max Reitz |
Subject: |
[PULL 13/17] qcow2: Use bs->bl.request_alignment when updating an L1 entry |
Date: |
Thu, 6 Feb 2020 13:51:28 +0100 |
From: Alberto Garcia <address@hidden>
When updating an L1 entry the qcow2 driver writes a (512-byte) sector
worth of data to avoid a read-modify-write cycle. Instead of always
writing 512 bytes we should follow the alignment requirements of the
storage backend.
(the only exception is when the alignment is larger than the cluster
size because then we could be overwriting data after the L1 table)
Signed-off-by: Alberto Garcia <address@hidden>
Message-id: address@hidden
Reviewed-by: Max Reitz <address@hidden>
Signed-off-by: Max Reitz <address@hidden>
---
block/qcow2-cluster.c | 25 +++++++++++++++----------
1 file changed, 15 insertions(+), 10 deletions(-)
diff --git a/block/qcow2-cluster.c b/block/qcow2-cluster.c
index 0384fb2339..1947f13a2d 100644
--- a/block/qcow2-cluster.c
+++ b/block/qcow2-cluster.c
@@ -216,26 +216,31 @@ static int l2_load(BlockDriverState *bs, uint64_t offset,
}
/*
- * Writes one sector of the L1 table to the disk (can't update single entries
- * and we really don't want bdrv_pread to perform a read-modify-write)
+ * Writes an L1 entry to disk (note that depending on the alignment
+ * requirements this function may write more that just one entry in
+ * order to prevent bdrv_pwrite from performing a read-modify-write)
*/
-#define L1_ENTRIES_PER_SECTOR (512 / 8)
int qcow2_write_l1_entry(BlockDriverState *bs, int l1_index)
{
BDRVQcow2State *s = bs->opaque;
- uint64_t buf[L1_ENTRIES_PER_SECTOR] = { 0 };
int l1_start_index;
int i, ret;
+ int bufsize = MAX(sizeof(uint64_t),
+ MIN(bs->file->bs->bl.request_alignment,
s->cluster_size));
+ int nentries = bufsize / sizeof(uint64_t);
+ g_autofree uint64_t *buf = g_try_new0(uint64_t, nentries);
- l1_start_index = l1_index & ~(L1_ENTRIES_PER_SECTOR - 1);
- for (i = 0; i < L1_ENTRIES_PER_SECTOR && l1_start_index + i < s->l1_size;
- i++)
- {
+ if (buf == NULL) {
+ return -ENOMEM;
+ }
+
+ l1_start_index = QEMU_ALIGN_DOWN(l1_index, nentries);
+ for (i = 0; i < MIN(nentries, s->l1_size - l1_start_index); i++) {
buf[i] = cpu_to_be64(s->l1_table[l1_start_index + i]);
}
ret = qcow2_pre_write_overlap_check(bs, QCOW2_OL_ACTIVE_L1,
- s->l1_table_offset + 8 * l1_start_index, sizeof(buf), false);
+ s->l1_table_offset + 8 * l1_start_index, bufsize, false);
if (ret < 0) {
return ret;
}
@@ -243,7 +248,7 @@ int qcow2_write_l1_entry(BlockDriverState *bs, int l1_index)
BLKDBG_EVENT(bs->file, BLKDBG_L1_UPDATE);
ret = bdrv_pwrite_sync(bs->file,
s->l1_table_offset + 8 * l1_start_index,
- buf, sizeof(buf));
+ buf, bufsize);
if (ret < 0) {
return ret;
}
--
2.24.1
- [PULL 03/17] block: fix memleaks in bdrv_refresh_filename, (continued)
- [PULL 03/17] block: fix memleaks in bdrv_refresh_filename, Max Reitz, 2020/02/06
- [PULL 04/17] qcow2: Use a GString in report_unsupported_feature(), Max Reitz, 2020/02/06
- [PULL 05/17] iotests: remove 'linux' from default supported platforms, Max Reitz, 2020/02/06
- [PULL 06/17] iotests: Test 041 only works on certain systems, Max Reitz, 2020/02/06
- [PULL 07/17] iotests: Test 183 does not work on macOS and OpenBSD, Max Reitz, 2020/02/06
- [PULL 09/17] iotests: Skip Python-based tests if QEMU does not support virtio-blk, Max Reitz, 2020/02/06
- [PULL 08/17] iotests: Check for the availability of the required devices in 267 and 127, Max Reitz, 2020/02/06
- [PULL 10/17] iotests: Enable more tests in the 'auto' group to improve test coverage, Max Reitz, 2020/02/06
- [PULL 12/17] qcow2: Tighten cluster_offset alignment assertions, Max Reitz, 2020/02/06
- [PULL 11/17] qcow2: Don't round the L1 table allocation up to the sector size, Max Reitz, 2020/02/06
- [PULL 13/17] qcow2: Use bs->bl.request_alignment when updating an L1 entry,
Max Reitz <=
- [PULL 14/17] qcow2: Don't require aligned offsets in qcow2_co_copy_range_from(), Max Reitz, 2020/02/06
- [PULL 15/17] qcow2: Use BDRV_SECTOR_SIZE instead of the hardcoded value, Max Reitz, 2020/02/06
- [PULL 16/17] block/backup-top: fix failure path, Max Reitz, 2020/02/06
- [PULL 17/17] iotests: add test for backup-top failure on permission activation, Max Reitz, 2020/02/06
- Re: [PULL 00/17] Block patches, Peter Maydell, 2020/02/06