qemu-devel.nongnu.org archive mirror
 help / color / mirror / Atom feed
From: Kevin Wolf <kwolf@redhat.com>
To: anthony@codemonkey.ws
Cc: kwolf@redhat.com, qemu-devel@nongnu.org
Subject: [Qemu-devel] [PULL 13/30] sheepdog: pass copy_policy in the request
Date: Thu, 31 Oct 2013 16:48:27 +0100	[thread overview]
Message-ID: <1383234524-372-14-git-send-email-kwolf@redhat.com> (raw)
In-Reply-To: <1383234524-372-1-git-send-email-kwolf@redhat.com>

From: Liu Yuan <namei.unix@gmail.com>

Currently copy_policy isn't used. Recent sheepdog supports erasure coding, which
make use of copy_policy internally, but require client explicitly passing
copy_policy from base inode to newly creately inode for snapshot related
operations.

If connected sheep daemon doesn't utilize copy_policy, passing it to sheep
daemon is just one extra null effect operation. So no compatibility problem.

With this patch, sheepdog can provide erasure coded volume for QEMU VM.

Cc: Kevin Wolf <kwolf@redhat.com>
Cc: Stefan Hajnoczi <stefanha@redhat.com>
Signed-off-by: Liu Yuan <namei.unix@gmail.com>
Acked-by: MORITA Kazutaka <morita.kazutaka@lab.ntt.co.jp>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
---
 block/sheepdog.c | 20 +++++++++++++-------
 1 file changed, 13 insertions(+), 7 deletions(-)

diff --git a/block/sheepdog.c b/block/sheepdog.c
index b8a2985..9f0757b 100644
--- a/block/sheepdog.c
+++ b/block/sheepdog.c
@@ -126,7 +126,8 @@ typedef struct SheepdogObjReq {
     uint64_t oid;
     uint64_t cow_oid;
     uint8_t copies;
-    uint8_t reserved[7];
+    uint8_t copy_policy;
+    uint8_t reserved[6];
     uint64_t offset;
 } SheepdogObjReq;
 
@@ -139,7 +140,8 @@ typedef struct SheepdogObjRsp {
     uint32_t data_length;
     uint32_t result;
     uint8_t copies;
-    uint8_t reserved[3];
+    uint8_t copy_policy;
+    uint8_t reserved[2];
     uint32_t pad[6];
 } SheepdogObjRsp;
 
@@ -153,7 +155,8 @@ typedef struct SheepdogVdiReq {
     uint64_t vdi_size;
     uint32_t vdi_id;
     uint8_t copies;
-    uint8_t reserved[3];
+    uint8_t copy_policy;
+    uint8_t reserved[2];
     uint32_t snapid;
     uint32_t pad[3];
 } SheepdogVdiReq;
@@ -1346,7 +1349,8 @@ out:
 }
 
 static int do_sd_create(BDRVSheepdogState *s, char *filename, int64_t vdi_size,
-                        uint32_t base_vid, uint32_t *vdi_id, int snapshot)
+                        uint32_t base_vid, uint32_t *vdi_id, int snapshot,
+                        uint8_t copy_policy)
 {
     SheepdogVdiReq hdr;
     SheepdogVdiRsp *rsp = (SheepdogVdiRsp *)&hdr;
@@ -1376,6 +1380,7 @@ static int do_sd_create(BDRVSheepdogState *s, char *filename, int64_t vdi_size,
 
     hdr.data_length = wlen;
     hdr.vdi_size = vdi_size;
+    hdr.copy_policy = copy_policy;
 
     ret = do_req(fd, (SheepdogReq *)&hdr, buf, &wlen, &rlen);
 
@@ -1528,7 +1533,8 @@ static int sd_create(const char *filename, QEMUOptionParameter *options,
         bdrv_unref(bs);
     }
 
-    ret = do_sd_create(s, vdi, vdi_size, base_vid, &vid, 0);
+    /* TODO: allow users to specify copy number */
+    ret = do_sd_create(s, vdi, vdi_size, base_vid, &vid, 0, 0);
     if (!prealloc || ret) {
         goto out;
     }
@@ -1718,7 +1724,7 @@ static int sd_create_branch(BDRVSheepdogState *s)
      */
     deleted = sd_delete(s);
     ret = do_sd_create(s, s->name, s->inode.vdi_size, s->inode.vdi_id, &vid,
-                       !deleted);
+                       !deleted, s->inode.copy_policy);
     if (ret) {
         goto out;
     }
@@ -2008,7 +2014,7 @@ static int sd_snapshot_create(BlockDriverState *bs, QEMUSnapshotInfo *sn_info)
     }
 
     ret = do_sd_create(s, s->name, s->inode.vdi_size, s->inode.vdi_id, &new_vid,
-                       1);
+                       1, s->inode.copy_policy);
     if (ret < 0) {
         error_report("failed to create inode for snapshot. %s",
                      strerror(errno));
-- 
1.8.1.4

  parent reply	other threads:[~2013-10-31 15:49 UTC|newest]

Thread overview: 38+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2013-10-31 15:48 [Qemu-devel] [PULL v2 00/30] Block patches Kevin Wolf
2013-10-31 15:48 ` [Qemu-devel] [PULL 01/30] qapi: fix documentation example Kevin Wolf
2013-10-31 15:48 ` [Qemu-devel] [PULL 02/30] qcow2: Restore total_sectors value in save_vmstate Kevin Wolf
2013-10-31 15:48 ` [Qemu-devel] [PULL 03/30] qcow2: Unset zero_beyond_eof " Kevin Wolf
2013-10-31 15:48 ` [Qemu-devel] [PULL 04/30] qemu-img: add special exit code if bdrv_check is not supported Kevin Wolf
2013-10-31 15:48 ` [Qemu-devel] [PULL 05/30] block/vpc: check that the image has not been truncated Kevin Wolf
2013-10-31 15:48 ` [Qemu-devel] [PULL 06/30] qemu-iotests: Test for loading VM state from qcow2 Kevin Wolf
2013-10-31 15:48 ` [Qemu-devel] [PULL 07/30] qcow2: Flush image after creation Kevin Wolf
2013-10-31 15:48 ` [Qemu-devel] [PULL 08/30] exec: Fix bounce buffer allocation in address_space_map() Kevin Wolf
2013-10-31 15:48 ` [Qemu-devel] [PULL 09/30] ide-test: Check what happens with bus mastering disabled Kevin Wolf
2013-10-31 15:48 ` [Qemu-devel] [PULL 10/30] tests: Multiboot mmap test case Kevin Wolf
2013-10-31 15:48 ` [Qemu-devel] [PULL 11/30] block: Don't copy backing file name on error Kevin Wolf
2013-10-31 15:48 ` [Qemu-devel] [PULL 12/30] sheepdog: explicitly set copies as type uint8_t Kevin Wolf
2013-10-31 15:48 ` Kevin Wolf [this message]
2013-10-31 15:48 ` [Qemu-devel] [PULL 14/30] ahci: fix win7 hang on boot Kevin Wolf
2013-10-31 15:48 ` [Qemu-devel] [PULL 15/30] block: Disable BDRV_O_COPY_ON_READ for the backing file Kevin Wolf
2013-10-31 15:48 ` [Qemu-devel] [PULL 16/30] block: Avoid unecessary drv->bdrv_getlength() calls Kevin Wolf
2013-10-31 15:48 ` [Qemu-devel] [PULL 17/30] qemu-iotests: Fix 051 reference output Kevin Wolf
2013-10-31 15:48 ` [Qemu-devel] [PULL 18/30] qemu-iotests: drop duplicated "create_image" Kevin Wolf
2013-10-31 15:48 ` [Qemu-devel] [PULL 19/30] qemu-iotests: Test case for backing file deletion Kevin Wolf
2013-10-31 15:48 ` [Qemu-devel] [PULL 20/30] sheepdog: check return values of qemu_co_recv/send correctly Kevin Wolf
2013-10-31 15:48 ` [Qemu-devel] [PULL 21/30] sheepdog: handle vdi objects in resend_aio_req Kevin Wolf
2013-10-31 15:48 ` [Qemu-devel] [PULL 22/30] sheepdog: reload inode outside of resend_aioreq Kevin Wolf
2013-10-31 15:48 ` [Qemu-devel] [PULL 23/30] coroutine: add co_aio_sleep_ns() to allow sleep in block drivers Kevin Wolf
2013-10-31 15:48 ` [Qemu-devel] [PULL 24/30] sheepdog: try to reconnect to sheepdog after network error Kevin Wolf
2013-10-31 15:48 ` [Qemu-devel] [PULL 25/30] sheepdog: make add_aio_request and send_aioreq void functions Kevin Wolf
2013-10-31 15:48 ` [Qemu-devel] [PULL 26/30] sheepdog: cancel aio requests if possible Kevin Wolf
2013-10-31 15:48 ` [Qemu-devel] [PULL 27/30] sheepdog: check simultaneous create in resend_aioreq Kevin Wolf
2013-10-31 15:48 ` [Qemu-devel] [PULL 28/30] qemu-iotests: prefill some data to test image Kevin Wolf
2013-10-31 15:48 ` [Qemu-devel] [PULL 29/30] qapi: Add optional field 'compressed' to ImageInfo Kevin Wolf
2013-10-31 15:48 ` [Qemu-devel] [PULL 30/30] vmdk: Implment bdrv_get_specific_info Kevin Wolf
2013-10-31 16:13   ` Eric Blake
2013-10-31 16:20     ` Kevin Wolf
2013-11-01  9:32       ` Fam Zheng
2013-10-31 20:50 ` [Qemu-devel] [PULL v2 00/30] Block patches Anthony Liguori
2013-10-31 21:52   ` Paolo Bonzini
2013-11-01  7:22     ` Anthony Liguori
2013-11-04 11:00   ` 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=1383234524-372-14-git-send-email-kwolf@redhat.com \
    --to=kwolf@redhat.com \
    --cc=anthony@codemonkey.ws \
    --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).