Skip to contentSkip to conversation

Quorum disk image corruption resiliency

PATCH Superseded Benoît Canet 2 participants 14 messages Oct 23, 2012 Block layer

v1 of 9 Superseded Latest is v9 →

lore
Share & tools
lore
#0 lore raw

cover Quorum disk image corruption resiliency

This patchset is posted for inclusion into 1.3.

It create a block driver implementing a quorum using total qemu disk images. Writes are mirrored on the $total files.
For the reading part the $total files are read at the same time and a vote is done to determine if a qiov version is present $threshold or more times. It then return this majority version to the upper layers.
When i < $threshold versions of the data are returned by the lower layer the quorum is broken and the read return -EIO.

The goal of this patchset is to be turned in a QEMU block filter living just above raw-*.c and below qcow2/qed when the required infrastructure will be done.

Main use of this feature will be people using NFS appliances which can be subjected to bitflip errors.

This patchset can be used to replace blkverify and the out of tree blkmirror.

usage: -drive file=quorum:threshold/total:image_1.raw:...:image_total.raw,if=virtio,cache=none

since latest RFC:

Eric Blake: revert back separator to ":"
rewrite quorum_getlength

Benoît Canet: use memcmp to compare iovec excepted for the blkverify case
use strstart to parse argument in open

Benoît Canet (11):
quorum: Create quorum.c, add QuorumSingleAIOCB and QuorumAIOCB.
quorum: Create BDRVQuorumState and BlkDriver and do init.
quorum: Add quorum_open() and quorum_close().
quorum: Add quorum_aio_writev and its dependencies.
blkverify: Extract qemu_iovec_clone() and qemu_iovec_compare() from
blkverify.
quorum: Add quorum_co_flush().
quorum: Add quorum_aio_readv.
quorum: Add quorum mechanism.
quorum: Add quorum_getlength().
quorum: Add quorum_invalidate_cache().
quorum: Add quorum_co_is_allocated.

 block/Makefile.objs |    1 +
 block/blkverify.c   |  108 +-------
 block/quorum.c      |  689 +++++++++++++++++++++++++++++++++++++++++++++++++++
 cutils.c            |  103 ++++++++
 qemu-common.h       |    2 +
 5 files changed, 797 insertions(+), 106 deletions(-)
 create mode 100644 block/quorum.c
Signature · 1 line
1.7.10.4
#1 lore raw

PATCH 1/11 quorum: Create quorum.c, add QuorumSingleAIOCB and QuorumAIOCB.

---
 block/Makefile.objs |    1 +
 block/quorum.c      |   45 +++++++++++++++++++++++++++++++++++++++++++++
 2 files changed, 46 insertions(+)
 create mode 100644 block/quorum.c
block/Makefile.objs +1 -0
index 554f429..f94e06f 100644
@@ -2,6 +2,7 @@ block-obj-y += raw.o cow.o qcow.o vdi.o vmdk.o cloop.o dmg.o bochs.o vpc.o vvfat
 block-obj-y += qcow2.o qcow2-refcount.o qcow2-cluster.o qcow2-snapshot.o qcow2-cache.o
 block-obj-y += qed.o qed-gencb.o qed-l2-cache.o qed-table.o qed-cluster.o
 block-obj-y += qed-check.o
+block-obj-y += quorum.o
 block-obj-y += parallels.o nbd.o blkdebug.o sheepdog.o blkverify.o
 block-obj-$(CONFIG_WIN32) += raw-win32.o
 block-obj-$(CONFIG_POSIX) += raw-posix.o
block/quorum.c new file +45 -0

45 lines not on this page · raw message

Signature · 1 line
1.7.10.4
#2 lore raw

PATCH 2/11 quorum: Create BDRVQuorumState and BlkDriver and do init.

In reply to Benoît Canet
---
 block/quorum.c |   22 ++++++++++++++++++++++
 1 file changed, 22 insertions(+)
block/quorum.c +22 -0
index 65a6b55..19a9a44 100644
@@ -15,6 +15,13 @@
 
 #include "block_int.h"
 
+typedef struct {
+    BlockDriverState **bs;
+    int threshold;
+    int total;
+    char **filenames;
+} BDRVQuorumState;
+
 typedef struct QuorumAIOCB QuorumAIOCB;
 
 typedef struct QuorumSingleAIOCB {
@@ -26,6 +33,7 @@ typedef struct QuorumSingleAIOCB {
 
 struct QuorumAIOCB {
     BlockDriverAIOCB common;
+    BDRVQuorumState *bqs;
     QEMUBH *bh;
 
     /* Request metadata */
@@ -43,3 +51,17 @@ struct QuorumAIOCB {
     void (*vote)(QuorumAIOCB *acb);
     int vote_ret;
 };
+
+static BlockDriver bdrv_quorum = {
+    .format_name        = "quorum",
+    .protocol_name      = "quorum",
+
+    .instance_size      = sizeof(BDRVQuorumState),
+};
+
+static void bdrv_quorum_init(void)
+{
+    bdrv_register(&bdrv_quorum);
+}
+
+block_init(bdrv_quorum_init);
Signature · 1 line
1.7.10.4
#3 lore raw

PATCH 3/11 quorum: Add quorum_open() and quorum_close().

In reply to Benoît Canet

Valid quorum resources look like quorum:threshold/total:path/to/image_1: ... :path/to/image_total

':' is used as a separator to allow to use networked path '\' is the escaping character for filename containing ':' '\' escape itself

On the command line for quorum files "img,test.raw", "img2.raw" and "img3.raw" invocation look like:

-drive file=quorum:2/3:img\\:test.raw:img2.raw:img3.raw (note the double \\)

---
 block/quorum.c |  122 ++++++++++++++++++++++++++++++++++++++++++++++++++++++++
 1 file changed, 122 insertions(+)
block/quorum.c +122 -0
index 19a9a44..7b20f08 100644
@@ -52,11 +52,133 @@ struct QuorumAIOCB {
     int vote_ret;
 };
 
+/* Valid quorum resources look like
+ * quorum:threshold/total:path/to/image_1: ... :path/to/image_total
+ *
+ * ':' is used as a separator
+ * '\' is the escaping character for filename containing ':'
+ */
+static int quorum_open(BlockDriverState *bs, const char *filename, int flags)
+{
+    BDRVQuorumState *s = bs->opaque;
+    int i, j, k, len, ret = 0;
+    char *a, *b, *names;
+    const char *start;
+    bool escape;
+
+    /* Parse the quorum: prefix */
+    if (!strstart(filename, "quorum:", &start)) {
+        return -EINVAL;
+    }
+
+    /* Get threshold */
+    errno = 0;
+    s->threshold = strtoul(start, &a, 10);
+    if (*a != '/' || errno) {
+        return -EINVAL;
+    }
+    a++;
+
+    /* Get total */
+    errno = 0;
+    s->total = strtoul(a, &b, 10);
+    if (*b != ':' || errno) {
+        return -EINVAL;
+    }
+    b++;
+
+    if (s->threshold < 1 || s->total < 2) {
+        return -EINVAL;
+    }
+
+    if (s->threshold > s->total) {
+        return -EINVAL;
+    }
+
+    s->bs = g_malloc0(sizeof(BlockDriverState *) * s->total);
+    /* Two allocations for all filenames: simpler to free */
+    s->filenames = g_malloc0(sizeof(char *) * s->total);
+    names = g_strdup(b);
+
+    /* Get the filenames pointers */
+    escape = false;
+    s->filenames[0] = names;
+    len = strlen(names);
+    for (i = j = k = 0; i < len && j < s->total; i++) {
+        /* separation between two files */
+        if (!escape && names[i] == ':') {
+            char *prev = s->filenames[j];
+            prev[k] = '\0';
+            s->filenames[++j] = prev + k + 1;
+            k = 0;
+            continue;
+        }
+
+        escape = !escape && names[i] == '\\';
+
+        /* if we are not escaping copy */
+        if (!escape) {
+            s->filenames[j][k++] = names[i];
+        }
+    }
+    /* terminate last string */
+    s->filenames[j][k] = '\0';
+
+    if ((j + 1) != s->total) {
+        ret = -EINVAL;
+        goto free_exit;
+    }
+
+    /* Open files */
+    for (i = 0; i < s->total; i++) {
+        s->bs[i] = bdrv_new("");
+        ret = bdrv_open(s->bs[i], s->filenames[i], flags, NULL);
+        if (ret < 0) {
+            goto error_exit;
+        }
+    }
+
+    goto exit;
+
+error_exit:
+    for (; i >= 0; i--) {
+        bdrv_delete(s->bs[i]);
+        s->bs[i] = NULL;
+    }
+free_exit:
+    g_free(s->filenames[0]);
+    g_free(s->filenames);
+    s->filenames = NULL;
+    g_free(s->bs);
+exit:
+    return ret;
+}
+
+static void quorum_close(BlockDriverState *bs)
+{
+    BDRVQuorumState *s = bs->opaque;
+    int i;
+
+    for (i = 0; i < s->total; i++) {
+        /* Ensure writes reach stable storage */
+        bdrv_flush(s->bs[i]);
+        bdrv_delete(s->bs[i]);
+    }
+
+    g_free(s->filenames[0]);
+    g_free(s->filenames);
+    s->filenames = NULL;
+    g_free(s->bs);
+}
+
 static BlockDriver bdrv_quorum = {
     .format_name        = "quorum",
     .protocol_name      = "quorum",
 
     .instance_size      = sizeof(BDRVQuorumState),
+
+    .bdrv_file_open     = quorum_open,
+    .bdrv_close         = quorum_close,
 };
 
 static void bdrv_quorum_init(void)
Signature · 1 line
1.7.10.4
#4 lore raw

Re: [Qemu-devel] [PATCH 03/11] quorum: Add quorum_open() and quorum_close().

Quoted 8 lines · Benoît Canet

Valid quorum resources look like quorum:threshold/total:path/to/image_1: ... :path/to/image_total

':' is used as a separator to allow to use networked path '\' is the escaping character for filename containing ':' '\' escape itself

On the command line for quorum files "img,test.raw", "img2.raw"

Here, you used a comma,

Quoted 3 lines

and "img3.raw" invocation look like:

-drive file=quorum:2/3:img\\:test.raw:img2.raw:img3.raw

but here, you are using a colon for the underlying name 'img:test.raw'.

Quoted 1 line

(note the double \\)

You also didn't give an example of comma escaping using typical command line comma escaping; but at least your example looks okay once you fix the typo in the lead-in text. The code appears to match the commit message.

Signature · 2 lines
Eric Blake   [email protected]    +1-919-301-3266
Libvirt virtualization library http://libvirt.org
#5 lore raw

PATCH 4/11 quorum: Add quorum_aio_writev and its dependencies.

In reply to Benoît Canet
---
 block/quorum.c |  112 ++++++++++++++++++++++++++++++++++++++++++++++++++++++++
 1 file changed, 112 insertions(+)
block/quorum.c +112 -0
index 7b20f08..878d930 100644
@@ -171,6 +171,116 @@ static void quorum_close(BlockDriverState *bs)
     g_free(s->bs);
 }
 
+static void quorum_aio_cancel(BlockDriverAIOCB *blockacb)
+{
+    QuorumAIOCB *acb = container_of(blockacb, QuorumAIOCB, common);
+    bool finished = false;
+
+    /* Wait for the request to finish */
+    acb->finished = &finished;
+    while (!finished) {
+        qemu_aio_wait();
+    }
+}
+
+static AIOPool quorum_aio_pool = {
+    .aiocb_size         = sizeof(QuorumAIOCB),
+    .cancel             = quorum_aio_cancel,
+};
+
+static void quorum_aio_bh(void *opaque)
+{
+    QuorumAIOCB *acb = opaque;
+    BDRVQuorumState *s = acb->bqs;
+    int ret;
+
+    ret = s->threshold <= acb->success_count ? 0 : -EIO;
+
+    qemu_bh_delete(acb->bh);
+    acb->common.cb(acb->common.opaque, ret);
+    if (acb->finished) {
+        *acb->finished = true;
+    }
+    g_free(acb->aios);
+    g_free(acb->qiovs);
+    qemu_aio_release(acb);
+}
+
+static QuorumAIOCB *quorum_aio_get(BDRVQuorumState *s,
+                                   BlockDriverState *bs,
+                                   QEMUIOVector *qiov,
+                                   int64_t sector_num,
+                                   int nb_sectors,
+                                   BlockDriverCompletionFunc *cb,
+                                   void *opaque)
+{
+    QuorumAIOCB *acb = qemu_aio_get(&quorum_aio_pool, bs, cb, opaque);
+    int i;
+
+    acb->aios = g_new0(QuorumSingleAIOCB, s->total);
+    acb->qiovs = g_new0(QEMUIOVector, s->total);
+
+    acb->bqs = s;
+    acb->qiov = qiov;
+    acb->bh = NULL;
+    acb->count = 0;
+    acb->success_count = 0;
+    acb->sector_num = sector_num;
+    acb->nb_sectors = nb_sectors;
+    acb->vote = NULL;
+    acb->vote_ret = 0;
+
+    for (i = 0; i < s->total; i++) {
+        acb->aios[i].buf = NULL;
+        acb->aios[i].ret = 0;
+        acb->aios[i].parent = acb;
+    }
+
+    return acb;
+}
+
+static void quorum_aio_cb(void *opaque, int ret)
+{
+    QuorumSingleAIOCB *sacb = opaque;
+    QuorumAIOCB *acb = sacb->parent;
+    BDRVQuorumState *s = acb->bqs;
+
+    sacb->ret = ret;
+    acb->count++;
+    if (ret == 0) {
+        acb->success_count++;
+    }
+    assert(acb->count <= s->total);
+    assert(acb->success_count <= s->total);
+    if (acb->count < s->total) {
+        return;
+    }
+
+    acb->bh = qemu_bh_new(quorum_aio_bh, acb);
+    qemu_bh_schedule(acb->bh);
+}
+
+static BlockDriverAIOCB *quorum_aio_writev(BlockDriverState *bs,
+                                          int64_t sector_num,
+                                          QEMUIOVector *qiov,
+                                          int nb_sectors,
+                                          BlockDriverCompletionFunc *cb,
+                                          void *opaque)
+{
+    BDRVQuorumState *s = bs->opaque;
+    QuorumAIOCB *acb = quorum_aio_get(s, bs, qiov, sector_num, nb_sectors,
+                                      cb, opaque);
+    int i;
+
+    for (i = 0; i < s->total; i++) {
+        acb->aios[i].aiocb = bdrv_aio_writev(s->bs[i], sector_num, qiov,
+                                             nb_sectors, &quorum_aio_cb,
+                                             &acb->aios[i]);
+    }
+
+    return &acb->common;
+}
+
 static BlockDriver bdrv_quorum = {
     .format_name        = "quorum",
     .protocol_name      = "quorum",
@@ -179,6 +289,8 @@ static BlockDriver bdrv_quorum = {
 
     .bdrv_file_open     = quorum_open,
     .bdrv_close         = quorum_close,
+
+    .bdrv_aio_writev    = quorum_aio_writev,
 };
 
 static void bdrv_quorum_init(void)
Signature · 1 line
1.7.10.4
#6 lore raw

PATCH 5/11 blkverify: Extract qemu_iovec_clone() and qemu_iovec_compare() from blkverify.

In reply to Benoît Canet
---
 block/blkverify.c |  108 +----------------------------------------------------
 cutils.c          |  103 ++++++++++++++++++++++++++++++++++++++++++++++++++
 qemu-common.h     |    2 +
 3 files changed, 107 insertions(+), 106 deletions(-)
block/blkverify.c +2 -106
index 9d5f1ec..79d36d5 100644
@@ -123,110 +123,6 @@ static int64_t blkverify_getlength(BlockDriverState *bs)
     return bdrv_getlength(s->test_file);
 }
 
-/**
- * Check that I/O vector contents are identical
- *
- * @a:          I/O vector
- * @b:          I/O vector
- * @ret:        Offset to first mismatching byte or -1 if match
- */
-static ssize_t blkverify_iovec_compare(QEMUIOVector *a, QEMUIOVector *b)
-{
-    int i;
-    ssize_t offset = 0;
-
-    assert(a->niov == b->niov);
-    for (i = 0; i < a->niov; i++) {
-        size_t len = 0;
-        uint8_t *p = (uint8_t *)a->iov[i].iov_base;
-        uint8_t *q = (uint8_t *)b->iov[i].iov_base;
-
-        assert(a->iov[i].iov_len == b->iov[i].iov_len);
-        while (len < a->iov[i].iov_len && *p++ == *q++) {
-            len++;
-        }
-
-        offset += len;
-
-        if (len != a->iov[i].iov_len) {
-            return offset;
-        }
-    }
-    return -1;
-}
-
-typedef struct {
-    int src_index;
-    struct iovec *src_iov;
-    void *dest_base;
-} IOVectorSortElem;
-
-static int sortelem_cmp_src_base(const void *a, const void *b)
-{
-    const IOVectorSortElem *elem_a = a;
-    const IOVectorSortElem *elem_b = b;
-
-    /* Don't overflow */
-    if (elem_a->src_iov->iov_base < elem_b->src_iov->iov_base) {
-        return -1;
-    } else if (elem_a->src_iov->iov_base > elem_b->src_iov->iov_base) {
-        return 1;
-    } else {
-        return 0;
-    }
-}
-
-static int sortelem_cmp_src_index(const void *a, const void *b)
-{
-    const IOVectorSortElem *elem_a = a;
-    const IOVectorSortElem *elem_b = b;
-
-    return elem_a->src_index - elem_b->src_index;
-}
-
-/**
- * Copy contents of I/O vector
- *
- * The relative relationships of overlapping iovecs are preserved.  This is
- * necessary to ensure identical semantics in the cloned I/O vector.
- */
-static void blkverify_iovec_clone(QEMUIOVector *dest, const QEMUIOVector *src,
-                                  void *buf)
-{
-    IOVectorSortElem sortelems[src->niov];
-    void *last_end;
-    int i;
-
-    /* Sort by source iovecs by base address */
-    for (i = 0; i < src->niov; i++) {
-        sortelems[i].src_index = i;
-        sortelems[i].src_iov = &src->iov[i];
-    }
-    qsort(sortelems, src->niov, sizeof(sortelems[0]), sortelem_cmp_src_base);
-
-    /* Allocate buffer space taking into account overlapping iovecs */
-    last_end = NULL;
-    for (i = 0; i < src->niov; i++) {
-        struct iovec *cur = sortelems[i].src_iov;
-        ptrdiff_t rewind = 0;
-
-        /* Detect overlap */
-        if (last_end && last_end > cur->iov_base) {
-            rewind = last_end - cur->iov_base;
-        }
-
-        sortelems[i].dest_base = buf - rewind;
-        buf += cur->iov_len - MIN(rewind, cur->iov_len);
-        last_end = MAX(cur->iov_base + cur->iov_len, last_end);
-    }
-
-    /* Sort by source iovec index and build destination iovec */
-    qsort(sortelems, src->niov, sizeof(sortelems[0]), sortelem_cmp_src_index);
-    for (i = 0; i < src->niov; i++) {
-        qemu_iovec_add(dest, sortelems[i].dest_base, src->iov[i].iov_len);
-    }
-}
-
 static BlkverifyAIOCB *blkverify_aio_get(BlockDriverState *bs, bool is_write,
                                          int64_t sector_num, QEMUIOVector *qiov,
                                          int nb_sectors,
@@ -290,7 +186,7 @@ static void blkverify_aio_cb(void *opaque, int ret)
 
 static void blkverify_verify_readv(BlkverifyAIOCB *acb)
 {
-    ssize_t offset = blkverify_iovec_compare(acb->qiov, &acb->raw_qiov);
+    ssize_t offset = qemu_iovec_compare(acb->qiov, &acb->raw_qiov);
     if (offset != -1) {
         blkverify_err(acb, "contents mismatch in sector %" PRId64,
                       acb->sector_num + (int64_t)(offset / BDRV_SECTOR_SIZE));
@@ -308,7 +204,7 @@ static BlockDriverAIOCB *blkverify_aio_readv(BlockDriverState *bs,
     acb->verify = blkverify_verify_readv;
     acb->buf = qemu_blockalign(bs->file, qiov->size);
     qemu_iovec_init(&acb->raw_qiov, acb->qiov->niov);
-    blkverify_iovec_clone(&acb->raw_qiov, qiov, acb->buf);
+    qemu_iovec_clone(&acb->raw_qiov, qiov, acb->buf);
 
     bdrv_aio_readv(s->test_file, sector_num, qiov, nb_sectors,
                    blkverify_aio_cb, acb);
cutils.c +103 -0

109 lines not on this page · raw message

qemu-common.h +2 -0
index b54612b..78467d4 100644
@@ -386,6 +386,8 @@ size_t qemu_iovec_from_buf(QEMUIOVector *qiov, size_t offset,
                            const void *buf, size_t bytes);
 size_t qemu_iovec_memset(QEMUIOVector *qiov, size_t offset,
                          int fillc, size_t bytes);
+ssize_t qemu_iovec_compare(QEMUIOVector *a, QEMUIOVector *b);
+void qemu_iovec_clone(QEMUIOVector *dest, const QEMUIOVector *src, void *buf);
 
 bool buffer_is_zero(const void *buf, size_t len);
 
Signature · 1 line
1.7.10.4
#7 lore raw

PATCH 6/11 quorum: Add quorum_co_flush().

In reply to Benoît Canet
---
 block/quorum.c |   13 +++++++++++++
 1 file changed, 13 insertions(+)
block/quorum.c +13 -0
index 878d930..1b40081 100644
@@ -281,6 +281,18 @@ static BlockDriverAIOCB *quorum_aio_writev(BlockDriverState *bs,
     return &acb->common;
 }
 
+static coroutine_fn int quorum_co_flush(BlockDriverState *bs)
+{
+    BDRVQuorumState *s = bs->opaque;
+    int i;
+
+    for (i = 0; i < s->total; i++) {
+        bdrv_co_flush(s->bs[i]);
+    }
+
+    return 0;
+}
+
 static BlockDriver bdrv_quorum = {
     .format_name        = "quorum",
     .protocol_name      = "quorum",
@@ -289,6 +301,7 @@ static BlockDriver bdrv_quorum = {
 
     .bdrv_file_open     = quorum_open,
     .bdrv_close         = quorum_close,
+    .bdrv_co_flush_to_disk = quorum_co_flush,
 
     .bdrv_aio_writev    = quorum_aio_writev,
 };
Signature · 1 line
1.7.10.4
#8 lore raw

Re: [Qemu-devel] [PATCH 06/11] quorum: Add quorum_co_flush().

Quoted 10 lines · Benoît Canet
---
 block/quorum.c |   13 +++++++++++++
 1 file changed, 13 insertions(+)
block/quorum.c +0 -0
index 878d930..1b40081 100644

block/quorum.c static BlockDriverAIOCB *quorum_aio_writev(BlockDriverState *bs, · line 290

Quoted 10 lines
block/quorum.c +7 -0
@@ -281,6 +281,18 @@ static BlockDriverAIOCB *quorum_aio_writev(BlockDriverState *bs,
     return &acb->common;
 }
 
+static coroutine_fn int quorum_co_flush(BlockDriverState *bs)
+{
+    BDRVQuorumState *s = bs->opaque;
+    int i;
+
+    for (i = 0; i < s->total; i++) {
+        bdrv_co_flush(s->bs[i]);

What if any of the individual flush calls return non-zero?

block/quorum.c static BlockDriverAIOCB *quorum_aio_writev(BlockDriverState *bs, · line 293

Quoted 3 lines
block/quorum.c +3 -0
+    }
+
+    return 0;

Shouldn't you then do likewise?

Signature · 2 lines
Eric Blake   [email protected]    +1-919-301-3266
Libvirt virtualization library http://libvirt.org
#9 lore raw

PATCH 7/11 quorum: Add quorum_aio_readv.

In reply to Benoît Canet
---
 block/quorum.c |   38 +++++++++++++++++++++++++++++++++++++-
 1 file changed, 37 insertions(+), 1 deletion(-)
block/quorum.c +37 -1
index 1b40081..51301fb 100644
@@ -192,15 +192,24 @@ static void quorum_aio_bh(void *opaque)
 {
     QuorumAIOCB *acb = opaque;
     BDRVQuorumState *s = acb->bqs;
-    int ret;
+    int i, ret;
 
     ret = s->threshold <= acb->success_count ? 0 : -EIO;
 
+    for (i = 0; i < s->total; i++) {
+        qemu_vfree(acb->aios[i].buf);
+        acb->aios[i].buf = NULL;
+        acb->aios[i].ret = 0;
+    }
+
     qemu_bh_delete(acb->bh);
     acb->common.cb(acb->common.opaque, ret);
     if (acb->finished) {
         *acb->finished = true;
     }
+    for (i = 0; i < s->total; i++) {
+        qemu_iovec_destroy(&acb->qiovs[i]);
+    }
     g_free(acb->aios);
     g_free(acb->qiovs);
     qemu_aio_release(acb);
@@ -260,6 +269,32 @@ static void quorum_aio_cb(void *opaque, int ret)
     qemu_bh_schedule(acb->bh);
 }
 
+static BlockDriverAIOCB *quorum_aio_readv(BlockDriverState *bs,
+                                         int64_t sector_num,
+                                         QEMUIOVector *qiov,
+                                         int nb_sectors,
+                                         BlockDriverCompletionFunc *cb,
+                                         void *opaque)
+{
+    BDRVQuorumState *s = bs->opaque;
+    QuorumAIOCB *acb = quorum_aio_get(s, bs, qiov, sector_num,
+                                      nb_sectors, cb, opaque);
+    int i;
+
+    for (i = 0; i < s->total; i++) {
+        acb->aios[i].buf = qemu_blockalign(bs->file, qiov->size);
+        qemu_iovec_init(&acb->qiovs[i], qiov->niov);
+        qemu_iovec_clone(&acb->qiovs[i], qiov, acb->aios[i].buf);
+    }
+
+    for (i = 0; i < s->total; i++) {
+        bdrv_aio_readv(s->bs[i], sector_num, qiov, nb_sectors,
+                       quorum_aio_cb, &acb->aios[i]);
+    }
+
+    return &acb->common;
+}
+
 static BlockDriverAIOCB *quorum_aio_writev(BlockDriverState *bs,
                                           int64_t sector_num,
                                           QEMUIOVector *qiov,
@@ -303,6 +338,7 @@ static BlockDriver bdrv_quorum = {
     .bdrv_close         = quorum_close,
     .bdrv_co_flush_to_disk = quorum_co_flush,
 
+    .bdrv_aio_readv     = quorum_aio_readv,
     .bdrv_aio_writev    = quorum_aio_writev,
 };
 
Signature · 1 line
1.7.10.4
#10 lore raw

PATCH 8/11 quorum: Add quorum mechanism.

In reply to Benoît Canet
---
 block/quorum.c |  278 +++++++++++++++++++++++++++++++++++++++++++++++++++++++-
 1 file changed, 277 insertions(+), 1 deletion(-)
block/quorum.c +277 -1
index 51301fb..6d76c49 100644
@@ -14,6 +14,20 @@
  */
 
 #include "block_int.h"
+#include "zlib.h"
+
+typedef struct QuorumVoteItem {
+    int index;
+    QLIST_ENTRY(QuorumVoteItem) next;
+} QuorumVoteItem;
+
+typedef struct QuorumVoteVersion {
+    unsigned long value;
+    int index;
+    int vote_count;
+    QLIST_HEAD(, QuorumVoteItem) items;
+    QLIST_ENTRY(QuorumVoteVersion) next;
+} QuorumVoteVersion;
 
 typedef struct {
     BlockDriverState **bs;
@@ -31,6 +45,10 @@ typedef struct QuorumSingleAIOCB {
     QuorumAIOCB *parent;
 } QuorumSingleAIOCB;
 
+typedef struct QuorumVotes {
+    QLIST_HEAD(, QuorumVoteVersion) vote_list;
+} QuorumVotes;
+
 struct QuorumAIOCB {
     BlockDriverAIOCB common;
     BDRVQuorumState *bqs;
@@ -48,6 +66,8 @@ struct QuorumAIOCB {
     int success_count;          /* number of successfully completed AIOCB */
     bool *finished;             /* completion signal for cancel */
 
+    QuorumVotes votes;
+
     void (*vote)(QuorumAIOCB *acb);
     int vote_ret;
 };
@@ -203,6 +223,11 @@ static void quorum_aio_bh(void *opaque)
     }
 
     qemu_bh_delete(acb->bh);
+
+    if (acb->vote_ret) {
+        ret = acb->vote_ret;
+    }
+
     acb->common.cb(acb->common.opaque, ret);
     if (acb->finished) {
         *acb->finished = true;
@@ -238,6 +263,7 @@ static QuorumAIOCB *quorum_aio_get(BDRVQuorumState *s,
     acb->nb_sectors = nb_sectors;
     acb->vote = NULL;
     acb->vote_ret = 0;
+    QLIST_INIT(&acb->votes.vote_list);
 
     for (i = 0; i < s->total; i++) {
         acb->aios[i].buf = NULL;
@@ -265,10 +291,258 @@ static void quorum_aio_cb(void *opaque, int ret)
         return;
     }
 
+    /* Do the vote */
+    if (acb->vote) {
+        acb->vote(acb);
+    }
+
     acb->bh = qemu_bh_new(quorum_aio_bh, acb);
     qemu_bh_schedule(acb->bh);
 }
 
+static void quorum_print_bad(QuorumAIOCB *acb, const char *filename)
+{
+    fprintf(stderr, "quorum: corrected error in quorum file %s: sector_num=%"
+            PRId64 " nb_sectors=%i\n", filename, acb->sector_num,
+            acb->nb_sectors);
+}
+
+static void quorum_print_failure(QuorumAIOCB *acb)
+{
+    fprintf(stderr, "quorum: failure sector_num=%" PRId64 " nb_sectors=%i\n",
+            acb->sector_num, acb->nb_sectors);
+}
+
+static void quorum_print_bad_versions(QuorumAIOCB *acb,
+                                      unsigned long checksum)
+{
+    QuorumVoteVersion *version;
+    QuorumVoteItem *item;
+    BDRVQuorumState *s = acb->bqs;
+
+    QLIST_FOREACH(version, &acb->votes.vote_list, next) {
+        if (version->value == checksum) {
+            continue;
+        }
+        QLIST_FOREACH(item, &version->items, next) {
+            quorum_print_bad(acb, s->filenames[item->index]);
+        }
+    }
+}
+
+static void quorum_copy_qiov(QEMUIOVector *dest, QEMUIOVector *source)
+{
+    int i;
+    assert(dest->niov == source->niov);
+    assert(dest->size == source->size);
+    for (i = 0; i < source->niov; i++) {
+        assert(dest->iov[i].iov_len == source->iov[i].iov_len);
+        memcpy(dest->iov[i].iov_base,
+               source->iov[i].iov_base,
+               source->iov[i].iov_len);
+    }
+}
+
+static void quorum_count_vote(QuorumVotes *votes,
+                              unsigned long checksum,
+                              int index)
+{
+    QuorumVoteVersion *v = NULL, *version = NULL;
+    QuorumVoteItem *item;
+
+    /* look if we have something with this checksum */
+    QLIST_FOREACH(v, &votes->vote_list, next) {
+        if (v->value == checksum) {
+            version = v;
+            break;
+        }
+    }
+
+    /* It's a version not yet in the list add it */
+    if (!version) {
+        version = g_new0(QuorumVoteVersion, 1);
+        QLIST_INIT(&version->items);
+        version->value = checksum;
+        version->index = index;
+        version->vote_count = 0;
+        QLIST_INSERT_HEAD(&votes->vote_list, version, next);
+    }
+
+    version->vote_count++;
+
+    item = g_new0(QuorumVoteItem, 1);
+    item->index = index;
+    QLIST_INSERT_HEAD(&version->items, item, next);
+}
+
+static void quorum_free_vote_list(QuorumVotes *votes)
+{
+    QuorumVoteVersion *version, *next_version;
+    QuorumVoteItem *item, *next_item;
+
+    QLIST_FOREACH_SAFE(version, &votes->vote_list, next, next_version) {
+        QLIST_REMOVE(version, next);
+        QLIST_FOREACH_SAFE(item, &version->items, next, next_item) {
+            QLIST_REMOVE(item, next);
+            g_free(item);
+        }
+        g_free(version);
+    }
+}
+
+static unsigned long quorum_compute_checksum(QuorumAIOCB *acb, int i)
+{
+    int j;
+    unsigned long adler = adler32(0L, Z_NULL, 0);
+    QEMUIOVector *qiov = &acb->qiovs[i];
+
+    for (j = 0; j < qiov->niov; j++) {
+        adler = adler32(adler,
+                        qiov->iov[j].iov_base,
+                        qiov->iov[j].iov_len);
+    }
+
+    return adler;
+}
+
+static QuorumVoteVersion *quorum_get_vote_winner(QuorumVotes *votes)
+{
+    int i = 0;
+    QuorumVoteVersion *candidate, *winner = NULL;
+
+    QLIST_FOREACH(candidate, &votes->vote_list, next) {
+        if (candidate->vote_count > i) {
+            i = candidate->vote_count;
+            winner = candidate;
+        }
+    }
+
+    return winner;
+}
+
+static bool quorum_iovec_compare(QEMUIOVector *a, QEMUIOVector *b)
+{
+    int i;
+    int result;
+
+    assert(a->niov == b->niov);
+    for (i = 0; i < a->niov; i++) {
+        assert(a->iov[i].iov_len == b->iov[i].iov_len);
+        result = memcmp(a->iov[i].iov_base,
+                        b->iov[i].iov_base,
+                        a->iov[i].iov_len);
+        if (result) {
+            return false;
+        }
+    }
+
+    return true;
+}
+
+static void GCC_FMT_ATTR(2, 3) quorum_err(QuorumAIOCB *acb,
+                                          const char *fmt, ...)
+{
+    va_list ap;
+
+    va_start(ap, fmt);
+    fprintf(stderr, "quorum: sector_num=%" PRId64 " nb_sectors=%d ",
+            acb->sector_num, acb->nb_sectors);
+    vfprintf(stderr, fmt, ap);
+    fprintf(stderr, "\n");
+    va_end(ap);
+    exit(1);
+}
+
+static bool quorum_compare(QuorumAIOCB *acb,
+                           QEMUIOVector *a,
+                           QEMUIOVector *b)
+{
+    BDRVQuorumState *s = acb->bqs;
+    bool blkverify = false;
+    ssize_t offset;
+
+    if (s->total == 2 && s->threshold == 2) {
+        blkverify = true;
+    }
+
+    if (blkverify) {
+        offset = qemu_iovec_compare(a, b);
+        if (offset != -1) {
+            quorum_err(acb, "contents mismatch in sector %" PRId64,
+                       acb->sector_num +
+                       (int64_t)(offset / BDRV_SECTOR_SIZE));
+        }
+        return true;
+    }
+
+    return quorum_iovec_compare(a, b);
+}
+
+static void quorum_vote(QuorumAIOCB *acb)
+{
+    bool quorum = true;
+    int i, j;
+    unsigned long checksum = 0;
+    BDRVQuorumState *s = acb->bqs;
+    QuorumVoteVersion *winner;
+
+    /* get the index of the first successfull read */
+    for (i = 0; i < s->total; i++) {
+        if (!acb->aios[i].ret) {
+            break;
+        }
+    }
+
+    /* compare this read with all other successfull read looking for quorum */
+    for (j = i + 1; j < s->total; j++) {
+        if (acb->aios[j].ret) {
+            continue;
+        }
+
+        quorum = quorum_compare(acb, &acb->qiovs[i], &acb->qiovs[j]);
+        if (!quorum) {
+            break;
+        }
+    }
+
+    /* Every successfull read agrees -> Quorum */
+    if (quorum) {
+        quorum_copy_qiov(acb->qiov, &acb->qiovs[i]);
+        return;
+    }
+
+    /* compute checksums for each successfull read, also store indexes */
+    for (i = 0; i < s->total; i++) {
+        if (acb->aios[i].ret) {
+            continue;
+        }
+        checksum = quorum_compute_checksum(acb, i);
+        quorum_count_vote(&acb->votes, checksum, i);
+    }
+
+    /* vote to select the most represented version */
+    winner = quorum_get_vote_winner(&acb->votes);
+    assert(winner != NULL);
+
+    /* if the winner count is smaller than threshold read fail */
+    if (winner->vote_count < s->threshold) {
+        quorum_print_failure(acb);
+        acb->vote_ret = -EIO;
+        fprintf(stderr, "quorum: vote result inferior to threshold\n");
+        goto free_exit;
+    }
+
+    /* we have a winner: copy it */
+    quorum_copy_qiov(acb->qiov, &acb->qiovs[winner->index]);
+
+    /* if some versions are bad print them */
+    quorum_print_bad_versions(acb, winner->value);
+
+free_exit:
+    /* free lists */
+    quorum_free_vote_list(&acb->votes);
+}
+
 static BlockDriverAIOCB *quorum_aio_readv(BlockDriverState *bs,
                                          int64_t sector_num,
                                          QEMUIOVector *qiov,
@@ -281,6 +555,8 @@ static BlockDriverAIOCB *quorum_aio_readv(BlockDriverState *bs,
                                       nb_sectors, cb, opaque);
     int i;
 
+    acb->vote = quorum_vote;
+
     for (i = 0; i < s->total; i++) {
         acb->aios[i].buf = qemu_blockalign(bs->file, qiov->size);
         qemu_iovec_init(&acb->qiovs[i], qiov->niov);
@@ -288,7 +564,7 @@ static BlockDriverAIOCB *quorum_aio_readv(BlockDriverState *bs,
     }
 
     for (i = 0; i < s->total; i++) {
-        bdrv_aio_readv(s->bs[i], sector_num, qiov, nb_sectors,
+        bdrv_aio_readv(s->bs[i], sector_num, &acb->qiovs[i], nb_sectors,
                        quorum_aio_cb, &acb->aios[i]);
     }
 
Signature · 1 line
1.7.10.4
#11 lore raw

PATCH 9/11 quorum: Add quorum_getlength().

In reply to Benoît Canet

Check that every bs file return the same length.
If not return -EIO to disable the quorum and
avoid length discrepancy.

---
 block/quorum.c |   20 ++++++++++++++++++++
 1 file changed, 20 insertions(+)
block/quorum.c +20 -0
index 6d76c49..a01f712 100644
@@ -604,12 +604,32 @@ static coroutine_fn int quorum_co_flush(BlockDriverState *bs)
     return 0;
 }
 
+static int64_t quorum_getlength(BlockDriverState *bs)
+{
+    BDRVQuorumState *s = bs->opaque;
+    int64_t result;
+    int i;
+
+    /* check that every file have the same length */
+    result = bdrv_getlength(s->bs[0]);
+    for (i = 1; i < s->total; i++) {
+        int64_t value = bdrv_getlength(s->bs[i]);
+        if (value != result) {
+            return -EIO;
+        }
+    }
+
+    return result;
+}
+
 static BlockDriver bdrv_quorum = {
     .format_name        = "quorum",
     .protocol_name      = "quorum",
 
     .instance_size      = sizeof(BDRVQuorumState),
 
+    .bdrv_getlength     = quorum_getlength,
+
     .bdrv_file_open     = quorum_open,
     .bdrv_close         = quorum_close,
     .bdrv_co_flush_to_disk = quorum_co_flush,
Signature · 1 line
1.7.10.4
#12 lore raw

PATCH 10/11 quorum: Add quorum_invalidate_cache().

In reply to Benoît Canet
---
 block/quorum.c |   11 +++++++++++
 1 file changed, 11 insertions(+)
block/quorum.c +11 -0
index a01f712..9550216 100644
@@ -622,6 +622,16 @@ static int64_t quorum_getlength(BlockDriverState *bs)
     return result;
 }
 
+static void quorum_invalidate_cache(BlockDriverState *bs)
+{
+    BDRVQuorumState *s = bs->opaque;
+    int i;
+
+    for (i = 0; i < s->total; i++) {
+        bdrv_invalidate_cache(s->bs[i]);
+    }
+}
+
 static BlockDriver bdrv_quorum = {
     .format_name        = "quorum",
     .protocol_name      = "quorum",
@@ -636,6 +646,7 @@ static BlockDriver bdrv_quorum = {
 
     .bdrv_aio_readv     = quorum_aio_readv,
     .bdrv_aio_writev    = quorum_aio_writev,
+    .bdrv_invalidate_cache = quorum_invalidate_cache,
 };
 
 static void bdrv_quorum_init(void)
Signature · 1 line
1.7.10.4
#13 lore raw

PATCH 11/11 quorum: Add quorum_co_is_allocated.

In reply to Benoît Canet
---
 block/quorum.c |   32 ++++++++++++++++++++++++++++++++
 1 file changed, 32 insertions(+)
block/quorum.c +32 -0
index 9550216..587796d 100644
@@ -632,6 +632,37 @@ static void quorum_invalidate_cache(BlockDriverState *bs)
     }
 }
 
+static int coroutine_fn quorum_co_is_allocated(BlockDriverState *bs,
+                                               int64_t sector_num,
+                                               int nb_sectors,
+                                               int *pnum)
+{
+    BDRVQuorumState *s = bs->opaque;
+    QuorumVoteVersion *winner = NULL;
+    QuorumVotes result_votes, num_votes;
+    int i, result, num;
+
+    QLIST_INIT(&result_votes.vote_list);
+    QLIST_INIT(&num_votes.vote_list);
+
+    for (i = 0; i < s->total; i++) {
+        result = bdrv_co_is_allocated(s->bs[i], sector_num, nb_sectors, &num);
+        quorum_count_vote(&result_votes, result, i);
+        quorum_count_vote(&num_votes, num, i);
+    }
+
+    winner = quorum_get_vote_winner(&result_votes);
+    result = winner->value;
+
+    winner = quorum_get_vote_winner(&num_votes);
+    *pnum = winner->value;
+
+    quorum_free_vote_list(&result_votes);
+    quorum_free_vote_list(&num_votes);
+
+    return result;
+}
+
 static BlockDriver bdrv_quorum = {
     .format_name        = "quorum",
     .protocol_name      = "quorum",
@@ -647,6 +678,7 @@ static BlockDriver bdrv_quorum = {
     .bdrv_aio_readv     = quorum_aio_readv,
     .bdrv_aio_writev    = quorum_aio_writev,
     .bdrv_invalidate_cache = quorum_invalidate_cache,
+    .bdrv_co_is_allocated  = quorum_co_is_allocated,
 };
 
 static void bdrv_quorum_init(void)
Signature · 1 line
1.7.10.4