From: Kevin Wolf <kwolf@redhat.com>
To: qemu-block@nongnu.org
Cc: kwolf@redhat.com, peter.maydell@linaro.org, qemu-devel@nongnu.org
Subject: [PULL 22/51] block: Add generic bdrv_inherited_options()
Date: Fri, 15 May 2020 14:44:52 +0200 [thread overview]
Message-ID: <20200515124521.335403-23-kwolf@redhat.com> (raw)
In-Reply-To: <20200515124521.335403-1-kwolf@redhat.com>
From: Max Reitz <mreitz@redhat.com>
After the series this patch belongs to, we want to have a common
BdrvChildClass that encompasses all of child_file, child_format, and
child_backing. Such a single class needs a single .inherit_options()
implementation, and this patch introduces it.
The next patch will show how the existing implementations can fall back
to it just by passing appropriate BdrvChildRole and parent_is_format
values.
Signed-off-by: Max Reitz <mreitz@redhat.com>
Message-Id: <20200513110544.176672-11-mreitz@redhat.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
---
block.c | 81 +++++++++++++++++++++++++++++++++++++++++++++++++++++++++
1 file changed, 81 insertions(+)
diff --git a/block.c b/block.c
index a17c29f54b..9b0e13d537 100644
--- a/block.c
+++ b/block.c
@@ -1356,6 +1356,87 @@ const BdrvChildClass child_backing = {
.set_aio_ctx = bdrv_child_cb_set_aio_ctx,
};
+/*
+ * Returns the options and flags that a generic child of a BDS should
+ * get, based on the given options and flags for the parent BDS.
+ */
+static void __attribute__((unused))
+ bdrv_inherited_options(BdrvChildRole role, bool parent_is_format,
+ int *child_flags, QDict *child_options,
+ int parent_flags, QDict *parent_options)
+{
+ int flags = parent_flags;
+
+ /*
+ * First, decide whether to set, clear, or leave BDRV_O_PROTOCOL.
+ * Generally, the question to answer is: Should this child be
+ * format-probed by default?
+ */
+
+ /*
+ * Pure and non-filtered data children of non-format nodes should
+ * be probed by default (even when the node itself has BDRV_O_PROTOCOL
+ * set). This only affects a very limited set of drivers (namely
+ * quorum and blkverify when this comment was written).
+ * Force-clear BDRV_O_PROTOCOL then.
+ */
+ if (!parent_is_format &&
+ (role & BDRV_CHILD_DATA) &&
+ !(role & (BDRV_CHILD_METADATA | BDRV_CHILD_FILTERED)))
+ {
+ flags &= ~BDRV_O_PROTOCOL;
+ }
+
+ /*
+ * All children of format nodes (except for COW children) and all
+ * metadata children in general should never be format-probed.
+ * Force-set BDRV_O_PROTOCOL then.
+ */
+ if ((parent_is_format && !(role & BDRV_CHILD_COW)) ||
+ (role & BDRV_CHILD_METADATA))
+ {
+ flags |= BDRV_O_PROTOCOL;
+ }
+
+ /*
+ * If the cache mode isn't explicitly set, inherit direct and no-flush from
+ * the parent.
+ */
+ qdict_copy_default(child_options, parent_options, BDRV_OPT_CACHE_DIRECT);
+ qdict_copy_default(child_options, parent_options, BDRV_OPT_CACHE_NO_FLUSH);
+ qdict_copy_default(child_options, parent_options, BDRV_OPT_FORCE_SHARE);
+
+ if (role & BDRV_CHILD_COW) {
+ /* backing files are opened read-only by default */
+ qdict_set_default_str(child_options, BDRV_OPT_READ_ONLY, "on");
+ qdict_set_default_str(child_options, BDRV_OPT_AUTO_READ_ONLY, "off");
+ } else {
+ /* Inherit the read-only option from the parent if it's not set */
+ qdict_copy_default(child_options, parent_options, BDRV_OPT_READ_ONLY);
+ qdict_copy_default(child_options, parent_options,
+ BDRV_OPT_AUTO_READ_ONLY);
+ }
+
+ /*
+ * bdrv_co_pdiscard() respects unmap policy for the parent, so we
+ * can default to enable it on lower layers regardless of the
+ * parent option.
+ */
+ qdict_set_default_str(child_options, BDRV_OPT_DISCARD, "unmap");
+
+ /* Clear flags that only apply to the top layer */
+ flags &= ~(BDRV_O_SNAPSHOT | BDRV_O_NO_BACKING | BDRV_O_COPY_ON_READ);
+
+ if (role & BDRV_CHILD_METADATA) {
+ flags &= ~BDRV_O_NO_IO;
+ }
+ if (role & BDRV_CHILD_COW) {
+ flags &= ~BDRV_O_TEMPORARY;
+ }
+
+ *child_flags = flags;
+}
+
static int bdrv_open_flags(BlockDriverState *bs, int flags)
{
int open_flags = flags;
--
2.25.4
next prev parent reply other threads:[~2020-05-15 13:04 UTC|newest]
Thread overview: 54+ messages / expand[flat|nested] mbox.gz Atom feed top
2020-05-15 12:44 [PULL 00/51] Block layer patches Kevin Wolf
2020-05-15 12:44 ` [PULL 01/51] iotests/109: Don't mirror with mismatched size Kevin Wolf
2020-05-15 12:44 ` [PULL 02/51] iotests/229: Use blkdebug to inject an error Kevin Wolf
2020-05-15 12:44 ` [PULL 03/51] mirror: Make sure that source and target size match Kevin Wolf
2020-05-15 12:44 ` [PULL 04/51] iotests: Mirror with different source/target size Kevin Wolf
2020-05-15 12:44 ` [PULL 05/51] block/replication.c: Avoid cancelling the job twice Kevin Wolf
2020-05-15 12:44 ` [PULL 06/51] iotests: Fix incomplete type declarations Kevin Wolf
2020-05-15 12:44 ` [PULL 07/51] iotests: Run pylint and mypy in a testcase Kevin Wolf
2020-05-15 12:44 ` [PULL 08/51] block: Add bdrv_make_empty() Kevin Wolf
2020-05-15 12:44 ` [PULL 09/51] block: Add blk_make_empty() Kevin Wolf
2020-05-15 12:44 ` [PULL 10/51] block: Use blk_make_empty() after commits Kevin Wolf
2020-05-15 12:44 ` [PULL 11/51] replication: Avoid blk_make_empty() on read-only child Kevin Wolf
2020-05-15 12:44 ` [PULL 12/51] block: Use bdrv_make_empty() where possible Kevin Wolf
2020-05-15 12:44 ` [PULL 13/51] block: Mark commit, mirror, blkreplay as filters Kevin Wolf
2020-05-15 12:44 ` [PULL 14/51] block: Add BlockDriver.is_format Kevin Wolf
2020-05-15 12:44 ` [PULL 15/51] block: Rename BdrvChildRole to BdrvChildClass Kevin Wolf
2020-05-15 12:44 ` [PULL 16/51] block: Add BdrvChildRole and BdrvChildRoleBits Kevin Wolf
2020-05-15 12:44 ` [PULL 17/51] block: Add BdrvChildRole to BdrvChild Kevin Wolf
2020-05-15 12:44 ` [PULL 18/51] block: Pass BdrvChildRole to bdrv_child_perm() Kevin Wolf
2020-05-15 12:44 ` [PULL 19/51] block: Pass BdrvChildRole to .inherit_options() Kevin Wolf
2020-05-15 12:44 ` [PULL 20/51] block: Pass parent_is_format " Kevin Wolf
2020-05-15 12:44 ` [PULL 21/51] block: Rename bdrv_inherited_options() Kevin Wolf
2020-05-15 12:44 ` Kevin Wolf [this message]
2020-05-15 12:44 ` [PULL 23/51] block: Use bdrv_inherited_options() Kevin Wolf
2020-05-15 12:44 ` [PULL 24/51] block: Unify bdrv_child_cb_attach() Kevin Wolf
2020-05-15 12:44 ` [PULL 25/51] block: Unify bdrv_child_cb_detach() Kevin Wolf
2020-05-15 12:44 ` [PULL 26/51] block: Add child_of_bds Kevin Wolf
2020-05-15 12:44 ` [PULL 27/51] block: Distinguish paths in *_format_default_perms Kevin Wolf
2020-05-15 12:44 ` [PULL 28/51] block: Pull out bdrv_default_perms_for_cow() Kevin Wolf
2020-05-15 12:44 ` [PULL 29/51] block: Pull out bdrv_default_perms_for_storage() Kevin Wolf
2020-05-15 12:45 ` [PULL 30/51] block: Relax *perms_for_storage for data children Kevin Wolf
2020-05-15 12:45 ` [PULL 31/51] block: Add bdrv_default_perms() Kevin Wolf
2020-05-15 12:45 ` [PULL 32/51] raw-format: Split raw_read_options() Kevin Wolf
2020-05-15 12:45 ` [PULL 33/51] block: Switch child_format users to child_of_bds Kevin Wolf
2020-05-15 12:45 ` [PULL 34/51] block: Drop child_format Kevin Wolf
2020-05-15 12:45 ` [PULL 35/51] block: Make backing files child_of_bds children Kevin Wolf
2020-05-15 12:45 ` [PULL 36/51] block: Drop child_backing Kevin Wolf
2020-05-15 12:45 ` [PULL 37/51] block: Make format drivers use child_of_bds Kevin Wolf
2020-05-15 12:45 ` [PULL 38/51] block: Make filter " Kevin Wolf
2020-05-15 12:45 ` [PULL 39/51] block: Use child_of_bds in remaining places Kevin Wolf
2020-05-15 12:45 ` [PULL 40/51] tests: Use child_of_bds instead of child_file Kevin Wolf
2020-05-15 12:45 ` [PULL 41/51] block: Use bdrv_default_perms() Kevin Wolf
2020-05-15 12:45 ` [PULL 42/51] block: Make bdrv_filter_default_perms() static Kevin Wolf
2020-05-15 12:45 ` [PULL 43/51] block: Drop bdrv_format_default_perms() Kevin Wolf
2020-05-15 12:45 ` [PULL 44/51] block: Drop child_file Kevin Wolf
2020-05-15 12:45 ` [PULL 45/51] block: Pass BdrvChildRole in remaining cases Kevin Wolf
2020-05-15 12:45 ` [PULL 46/51] block: Drop @child_class from bdrv_child_perm() Kevin Wolf
2020-05-15 12:45 ` [PULL 47/51] block/block-copy: Fix uninitialized variable in block_copy_task_entry Kevin Wolf
2020-05-15 12:45 ` [PULL 48/51] block/block-copy: Simplify block_copy_do_copy() Kevin Wolf
2020-05-15 12:45 ` [PULL 49/51] iotests: log messages from notrun() Kevin Wolf
2020-05-15 12:45 ` [PULL 50/51] hw/ide/ahci: Log lost IRQs Kevin Wolf
2020-05-15 12:45 ` [PULL 51/51] iotests/030: Reduce run time by unthrottling job earlier Kevin Wolf
2020-05-15 14:28 ` [PULL 00/51] Block layer patches Peter Maydell
2020-05-15 20:30 ` no-reply
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=20200515124521.335403-23-kwolf@redhat.com \
--to=kwolf@redhat.com \
--cc=peter.maydell@linaro.org \
--cc=qemu-block@nongnu.org \
--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).