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] [RFC PATCH 07/16] qcow2: Factor out handle_dependencies()
Date: Tue, 18 Sep 2012 13:40:33 +0200	[thread overview]
Message-ID: <1347968442-8860-8-git-send-email-kwolf@redhat.com> (raw)
In-Reply-To: <1347968442-8860-1-git-send-email-kwolf@redhat.com>

Signed-off-by: Kevin Wolf <kwolf@redhat.com>
---
 block/qcow2-cluster.c |   70 +++++++++++++++++++++++++++++-------------------
 1 files changed, 42 insertions(+), 28 deletions(-)

diff --git a/block/qcow2-cluster.c b/block/qcow2-cluster.c
index 7a038ac..468ef1b 100644
--- a/block/qcow2-cluster.c
+++ b/block/qcow2-cluster.c
@@ -753,38 +753,16 @@ out:
 }
 
 /*
- * Allocates new clusters for the given guest_offset.
- *
- * At most *nb_clusters are allocated, and on return *nb_clusters is updated to
- * contain the number of clusters that have been allocated and are contiguous
- * in the image file.
- *
- * If *host_offset is non-zero, it specifies the offset in the image file at
- * which the new clusters must start. *nb_clusters can be 0 on return in this
- * case if the cluster at host_offset is already in use. If *host_offset is
- * zero, the clusters can be allocated anywhere in the image file.
- *
- * *host_offset is updated to contain the offset into the image file at which
- * the first allocated cluster starts.
- *
- * Return 0 on success and -errno in error cases. -EAGAIN means that the
- * function has been waiting for another request and the allocation must be
- * restarted, but the whole request should not be failed.
+ * Check if there already is an AIO write request in flight which allocates
+ * the same cluster. In this case we need to wait until the previous
+ * request has completed and updated the L2 table accordingly.
  */
-static int do_alloc_cluster_offset(BlockDriverState *bs, uint64_t guest_offset,
-    uint64_t *host_offset, unsigned int *nb_clusters)
+static int handle_dependencies(BlockDriverState *bs, uint64_t guest_offset,
+    unsigned int *nb_clusters)
 {
     BDRVQcowState *s = bs->opaque;
     QCowL2Meta *old_alloc;
 
-    trace_qcow2_do_alloc_clusters_offset(qemu_coroutine_self(), guest_offset,
-                                         *host_offset, *nb_clusters);
-
-    /*
-     * Check if there already is an AIO write request in flight which allocates
-     * the same cluster. In this case we need to wait until the previous
-     * request has completed and updated the L2 table accordingly.
-     */
     QLIST_FOREACH(old_alloc, &s->cluster_allocs, next_in_flight) {
 
         uint64_t start = guest_offset >> s->cluster_bits;
@@ -817,6 +795,42 @@ static int do_alloc_cluster_offset(BlockDriverState *bs, uint64_t guest_offset,
         abort();
     }
 
+    return 0;
+}
+
+/*
+ * Allocates new clusters for the given guest_offset.
+ *
+ * At most *nb_clusters are allocated, and on return *nb_clusters is updated to
+ * contain the number of clusters that have been allocated and are contiguous
+ * in the image file.
+ *
+ * If *host_offset is non-zero, it specifies the offset in the image file at
+ * which the new clusters must start. *nb_clusters can be 0 on return in this
+ * case if the cluster at host_offset is already in use. If *host_offset is
+ * zero, the clusters can be allocated anywhere in the image file.
+ *
+ * *host_offset is updated to contain the offset into the image file at which
+ * the first allocated cluster starts.
+ *
+ * Return 0 on success and -errno in error cases. -EAGAIN means that the
+ * function has been waiting for another request and the allocation must be
+ * restarted, but the whole request should not be failed.
+ */
+static int do_alloc_cluster_offset(BlockDriverState *bs, uint64_t guest_offset,
+    uint64_t *host_offset, unsigned int *nb_clusters)
+{
+    BDRVQcowState *s = bs->opaque;
+    int ret;
+
+    trace_qcow2_do_alloc_clusters_offset(qemu_coroutine_self(), guest_offset,
+                                         *host_offset, *nb_clusters);
+
+    ret = handle_dependencies(bs, guest_offset, nb_clusters);
+    if (ret < 0) {
+        return ret;
+    }
+
     /* Allocate new clusters */
     trace_qcow2_cluster_alloc_phys(qemu_coroutine_self());
     if (*host_offset == 0) {
@@ -828,7 +842,7 @@ static int do_alloc_cluster_offset(BlockDriverState *bs, uint64_t guest_offset,
         *host_offset = cluster_offset;
         return 0;
     } else {
-        int ret = qcow2_alloc_clusters_at(bs, *host_offset, *nb_clusters);
+        ret = qcow2_alloc_clusters_at(bs, *host_offset, *nb_clusters);
         if (ret < 0) {
             return ret;
         }
-- 
1.7.6.5

  parent reply	other threads:[~2012-09-18 11:41 UTC|newest]

Thread overview: 31+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2012-09-18 11:40 [Qemu-devel] [RFC PATCH 00/16] qcow2: Delayed COW Kevin Wolf
2012-09-18 11:40 ` [Qemu-devel] [RFC PATCH 01/16] qcow2: Round QCowL2Meta.offset down to cluster boundary Kevin Wolf
2012-09-18 11:40 ` [Qemu-devel] [RFC PATCH 02/16] qcow2: Introduce Qcow2COWRegion Kevin Wolf
2012-09-18 11:40 ` [Qemu-devel] [RFC PATCH 03/16] qcow2: Allocate l2meta dynamically Kevin Wolf
2012-09-18 11:40 ` [Qemu-devel] [RFC PATCH 04/16] qcow2: Drop l2meta.cluster_offset Kevin Wolf
2012-09-18 11:40 ` [Qemu-devel] [RFC PATCH 05/16] qcow2: Allocate l2meta only for cluster allocations Kevin Wolf
2012-09-18 11:40 ` [Qemu-devel] [RFC PATCH 06/16] qcow2: Enable dirty flag in qcow2_alloc_cluster_link_l2 Kevin Wolf
2012-09-18 11:40 ` Kevin Wolf [this message]
2012-09-18 11:40 ` [Qemu-devel] [RFC PATCH 08/16] qcow2: Reading from areas not in L2 tables yet Kevin Wolf
2012-09-18 11:40 ` [Qemu-devel] [RFC PATCH 09/16] qcow2: Move COW and L2 update into own coroutine Kevin Wolf
2012-09-18 14:24   ` Paolo Bonzini
2012-09-18 14:44     ` Kevin Wolf
2012-09-18 14:59       ` Paolo Bonzini
2012-09-18 11:40 ` [Qemu-devel] [RFC PATCH 10/16] qcow2: Delay the COW Kevin Wolf
2012-09-18 14:27   ` Paolo Bonzini
2012-09-18 14:49     ` Kevin Wolf
2012-09-19 18:47   ` Blue Swirl
2012-09-20  6:58     ` Kevin Wolf
2012-09-18 11:40 ` [Qemu-devel] [RFC PATCH 11/16] qcow2: Add error handling to the l2meta coroutine Kevin Wolf
2012-09-18 14:29   ` Paolo Bonzini
2012-09-18 11:40 ` [Qemu-devel] [RFC PATCH 12/16] qcow2: Handle dependencies earlier Kevin Wolf
2012-09-18 11:40 ` [Qemu-devel] [RFC PATCH 13/16] qcow2: Change handle_dependency to byte granularity Kevin Wolf
2012-09-18 11:40 ` [Qemu-devel] [RFC PATCH 14/16] qcow2: Execute run_dependent_requests() without lock Kevin Wolf
2012-09-18 14:33   ` Paolo Bonzini
2012-09-18 14:54     ` Kevin Wolf
2012-09-18 11:40 ` [Qemu-devel] [RFC PATCH 15/16] qcow2: Cancel COW when overwritten Kevin Wolf
2012-09-18 14:44   ` Paolo Bonzini
2012-09-18 15:02     ` Kevin Wolf
2012-09-18 15:05       ` Paolo Bonzini
2012-09-18 15:08         ` Paolo Bonzini
2012-09-18 11:40 ` [Qemu-devel] [RFC PATCH 16/16] [BROKEN] qcow2: Overwrite COW and allocate new cluster at the same time Kevin Wolf

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=1347968442-8860-8-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).