[Qemu-devel] [PULL 00/11] Block patches

[Qemu-devel] [PULL 00/11] Block patches

[Max Reitz]

The following changes since commit 19eb2d4e736dc895f31fbd6b520e514f10cc08e0:

  Merge remote-tracking branch 'remotes/thibault/tags/samuel-thibault' into staging (2019-05-07 10:43:32 +0100)

are available in the Git repository at:

  https://git.xanclic.moe/XanClic/qemu.git tags/pull-block-2019-05-07

for you to fetch changes up to 1278dce7927301bf3d004a40061dbd2c1e0846a8:

  iotests: Fix iotests 110 and 126 (2019-05-07 17:14:21 +0200)

----------------------------------------------------------------
Block patches:
- Fixes to qcow2's implementation of qemu-img check
- Our SSH driver now supports bdrv_refresh_filename()
- Miscellaneous fixes

----------------------------------------------------------------
Alberto Garcia (2):
  block: Assert that drv->bdrv_child_perm is set in bdrv_child_perm()
  commit: Use bdrv_append() in commit_start()

Andrey Shinkevich (1):
  qcow2: discard bitmap when removed

Max Reitz (3):
  block/ssh: Implement .bdrv_refresh_filename()
  block/ssh: Implement .bdrv_dirname()
  iotests: Fix iotests 110 and 126

Vladimir Sementsov-Ogievskiy (5):
  qcow2-refcount: fix check_oflag_copied
  qcow2-refcount: avoid eating RAM
  qcow2-refcount: check_refcounts_l2: reduce ignored overlaps
  qcow2-refcount: check_refcounts_l2: don't count fixed cluster as
    allocated
  qcow2-refcount: don't mask corruptions under internal errors

 block.c                       |  9 ++--
 block/commit.c                | 11 +----
 block/qcow2-bitmap.c          |  2 +-
 block/qcow2-refcount.c        | 80 ++++++++++++++++++++++-------------
 block/ssh.c                   | 73 +++++++++++++++++++++++++++++---
 tests/qemu-iotests/110        | 10 +++--
 tests/qemu-iotests/126        | 10 +++--
 tests/qemu-iotests/138        | 12 +++---
 tests/qemu-iotests/138.out    |  5 ++-
 tests/qemu-iotests/207        | 10 ++---
 tests/qemu-iotests/207.out    | 10 ++---
 tests/qemu-iotests/common.rc  |  2 +-
 tests/qemu-iotests/iotests.py |  2 +-
 13 files changed, 159 insertions(+), 77 deletions(-)

-- 
2.20.1

[Qemu-devel] [PULL 01/11] qcow2-refcount: fix check_oflag_copied

[Max Reitz]

From: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com>

Increase corruptions_fixed only after successful fix.

Signed-off-by: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com>
Reviewed-by: Max Reitz <mreitz@redhat.com>
Message-id: 20190227131433.197063-2-vsementsov@virtuozzo.com
Signed-off-by: Max Reitz <mreitz@redhat.com>
---
 block/qcow2-refcount.c | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

diff --git a/block/qcow2-refcount.c b/block/qcow2-refcount.c
index e0fe322500..ed3d8ebd57 100644
--- a/block/qcow2-refcount.c
+++ b/block/qcow2-refcount.c
@@ -1846,7 +1846,7 @@ static int check_oflag_copied(BlockDriverState *bs, BdrvCheckResult *res,
     for (i = 0; i < s->l1_size; i++) {
         uint64_t l1_entry = s->l1_table[i];
         uint64_t l2_offset = l1_entry & L1E_OFFSET_MASK;
-        bool l2_dirty = false;
+        int l2_dirty = 0;
 
         if (!l2_offset) {
             continue;
@@ -1912,8 +1912,7 @@ static int check_oflag_copied(BlockDriverState *bs, BdrvCheckResult *res,
                         l2_table[j] = cpu_to_be64(refcount == 1
                                     ? l2_entry |  QCOW_OFLAG_COPIED
                                     : l2_entry & ~QCOW_OFLAG_COPIED);
-                        l2_dirty = true;
-                        res->corruptions_fixed++;
+                        l2_dirty++;
                     } else {
                         res->corruptions++;
                     }
@@ -1921,7 +1920,7 @@ static int check_oflag_copied(BlockDriverState *bs, BdrvCheckResult *res,
             }
         }
 
-        if (l2_dirty) {
+        if (l2_dirty > 0) {
             ret = qcow2_pre_write_overlap_check(bs, QCOW2_OL_ACTIVE_L2,
                                                 l2_offset, s->cluster_size,
                                                 false);
@@ -1940,6 +1939,7 @@ static int check_oflag_copied(BlockDriverState *bs, BdrvCheckResult *res,
                 res->check_errors++;
                 goto fail;
             }
+            res->corruptions_fixed += l2_dirty;
         }
     }
 
-- 
2.20.1

[Qemu-devel] [PULL 02/11] qcow2-refcount: avoid eating RAM

[Max Reitz]

From: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com>

qcow2_inc_refcounts_imrt() (through realloc_refcount_array()) can eat
an unpredictable amount of memory on corrupted table entries, which are
referencing regions far beyond the end of file.

Prevent this, by skipping such regions from further processing.

Interesting that iotest 138 checks exactly the behavior which we fix
here. So, change the test appropriately.

Signed-off-by: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com>
Reviewed-by: Max Reitz <mreitz@redhat.com>
Message-id: 20190227131433.197063-3-vsementsov@virtuozzo.com
Signed-off-by: Max Reitz <mreitz@redhat.com>
---
 block/qcow2-refcount.c     | 19 +++++++++++++++++++
 tests/qemu-iotests/138     | 12 +++++-------
 tests/qemu-iotests/138.out |  5 ++++-
 3 files changed, 28 insertions(+), 8 deletions(-)

diff --git a/block/qcow2-refcount.c b/block/qcow2-refcount.c
index ed3d8ebd57..21e50dacf4 100644
--- a/block/qcow2-refcount.c
+++ b/block/qcow2-refcount.c
@@ -1520,12 +1520,31 @@ int qcow2_inc_refcounts_imrt(BlockDriverState *bs, BdrvCheckResult *res,
 {
     BDRVQcow2State *s = bs->opaque;
     uint64_t start, last, cluster_offset, k, refcount;
+    int64_t file_len;
     int ret;
 
     if (size <= 0) {
         return 0;
     }
 
+    file_len = bdrv_getlength(bs->file->bs);
+    if (file_len < 0) {
+        return file_len;
+    }
+
+    /*
+     * Last cluster of qcow2 image may be semi-allocated, so it may be OK to
+     * reference some space after file end but it should be less than one
+     * cluster.
+     */
+    if (offset + size - file_len >= s->cluster_size) {
+        fprintf(stderr, "ERROR: counting reference for region exceeding the "
+                "end of the file by one cluster or more: offset 0x%" PRIx64
+                " size 0x%" PRIx64 "\n", offset, size);
+        res->corruptions++;
+        return 0;
+    }
+
     start = start_of_cluster(s, offset);
     last = start_of_cluster(s, offset + size - 1);
     for(cluster_offset = start; cluster_offset <= last;
diff --git a/tests/qemu-iotests/138 b/tests/qemu-iotests/138
index f353ac8219..6a731370db 100755
--- a/tests/qemu-iotests/138
+++ b/tests/qemu-iotests/138
@@ -54,15 +54,13 @@ $QEMU_IO -c 'write 0 512' "$TEST_IMG" | _filter_qemu_io
 # Put the data cluster at a multiple of 2 TB, resulting in the image apparently
 # having a multiple of 2^32 clusters
 # (To be more specific: It is at 32 PB)
-poke_file "$TEST_IMG" 2048 "\x80\x80\x00\x00\x00\x00\x00\x00"
+poke_file "$TEST_IMG" $((2048 + 8)) "\x00\x80\x00\x00\x00\x00\x00\x00"
 
 # An offset of 32 PB results in qemu-img check having to allocate an in-memory
-# refcount table of 128 TB (16 bit refcounts, 512 byte clusters).
-# This should be generally too much for any system and thus fail.
-# What this test is checking is that the qcow2 driver actually tries to allocate
-# such a large amount of memory (and is consequently aborting) instead of having
-# truncated the cluster count somewhere (which would result in much less memory
-# being allocated and then a segfault occurring).
+# refcount table of 128 TB (16 bit refcounts, 512 byte clusters), if qemu-img
+# don't check that referenced data cluster is far beyond the end of file.
+# But starting from 4.0, qemu-img does this check, and instead of "Cannot
+# allocate memory", we have an error showing that l2 entry is invalid.
 _check_test_img
 
 # success, all done
diff --git a/tests/qemu-iotests/138.out b/tests/qemu-iotests/138.out
index 3fe911f85a..aca7d47a80 100644
--- a/tests/qemu-iotests/138.out
+++ b/tests/qemu-iotests/138.out
@@ -5,5 +5,8 @@ QA output created by 138
 Formatting 'TEST_DIR/t.IMGFMT', fmt=IMGFMT size=512
 wrote 512/512 bytes at offset 0
 512 bytes, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
-qemu-img: Check failed: Cannot allocate memory
+ERROR: counting reference for region exceeding the end of the file by one cluster or more: offset 0x80000000000000 size 0x200
+
+1 errors were found on the image.
+Data may be corrupted, or further writes to the image may corrupt it.
 *** done
-- 
2.20.1

[Qemu-devel] [PULL 03/11] qcow2-refcount: check_refcounts_l2: reduce ignored overlaps

[Max Reitz]

From: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com>

Reduce number of structures ignored in overlap check: when checking
active table ignore active tables, when checking inactive table ignore
inactive ones.

Signed-off-by: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com>
Reviewed-by: Max Reitz <mreitz@redhat.com>
Message-id: 20190227131433.197063-4-vsementsov@virtuozzo.com
Signed-off-by: Max Reitz <mreitz@redhat.com>
---
 block/qcow2-refcount.c | 16 +++++++++-------
 1 file changed, 9 insertions(+), 7 deletions(-)

diff --git a/block/qcow2-refcount.c b/block/qcow2-refcount.c
index 21e50dacf4..1ac2b0490d 100644
--- a/block/qcow2-refcount.c
+++ b/block/qcow2-refcount.c
@@ -1591,7 +1591,7 @@ enum {
 static int check_refcounts_l2(BlockDriverState *bs, BdrvCheckResult *res,
                               void **refcount_table,
                               int64_t *refcount_table_size, int64_t l2_offset,
-                              int flags, BdrvCheckMode fix)
+                              int flags, BdrvCheckMode fix, bool active)
 {
     BDRVQcow2State *s = bs->opaque;
     uint64_t *l2_table, l2_entry;
@@ -1682,11 +1682,12 @@ static int check_refcounts_l2(BlockDriverState *bs, BdrvCheckResult *res,
                     if (fix & BDRV_FIX_ERRORS) {
                         uint64_t l2e_offset =
                             l2_offset + (uint64_t)i * sizeof(uint64_t);
+                        int ign = active ? QCOW2_OL_ACTIVE_L2 :
+                                           QCOW2_OL_INACTIVE_L2;
 
                         l2_entry = QCOW_OFLAG_ZERO;
                         l2_table[i] = cpu_to_be64(l2_entry);
-                        ret = qcow2_pre_write_overlap_check(bs,
-                                QCOW2_OL_ACTIVE_L2 | QCOW2_OL_INACTIVE_L2,
+                        ret = qcow2_pre_write_overlap_check(bs, ign,
                                 l2e_offset, sizeof(uint64_t), false);
                         if (ret < 0) {
                             fprintf(stderr, "ERROR: Overlap check failed\n");
@@ -1762,7 +1763,7 @@ static int check_refcounts_l1(BlockDriverState *bs,
                               void **refcount_table,
                               int64_t *refcount_table_size,
                               int64_t l1_table_offset, int l1_size,
-                              int flags, BdrvCheckMode fix)
+                              int flags, BdrvCheckMode fix, bool active)
 {
     BDRVQcow2State *s = bs->opaque;
     uint64_t *l1_table = NULL, l2_offset, l1_size2;
@@ -1818,7 +1819,7 @@ static int check_refcounts_l1(BlockDriverState *bs,
             /* Process and check L2 entries */
             ret = check_refcounts_l2(bs, res, refcount_table,
                                      refcount_table_size, l2_offset, flags,
-                                     fix);
+                                     fix, active);
             if (ret < 0) {
                 goto fail;
             }
@@ -2109,7 +2110,7 @@ static int calculate_refcounts(BlockDriverState *bs, BdrvCheckResult *res,
     /* current L1 table */
     ret = check_refcounts_l1(bs, res, refcount_table, nb_clusters,
                              s->l1_table_offset, s->l1_size, CHECK_FRAG_INFO,
-                             fix);
+                             fix, true);
     if (ret < 0) {
         return ret;
     }
@@ -2138,7 +2139,8 @@ static int calculate_refcounts(BlockDriverState *bs, BdrvCheckResult *res,
             continue;
         }
         ret = check_refcounts_l1(bs, res, refcount_table, nb_clusters,
-                                 sn->l1_table_offset, sn->l1_size, 0, fix);
+                                 sn->l1_table_offset, sn->l1_size, 0, fix,
+                                 false);
         if (ret < 0) {
             return ret;
         }
-- 
2.20.1

[Qemu-devel] [PULL 04/11] qcow2-refcount: check_refcounts_l2: don't count fixed cluster as allocated

[Max Reitz]

From: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com>

Do not count a cluster which is fixed to be ZERO as allocated.

Signed-off-by: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com>
Reviewed-by: Max Reitz <mreitz@redhat.com>
Message-id: 20190227131433.197063-5-vsementsov@virtuozzo.com
Signed-off-by: Max Reitz <mreitz@redhat.com>
---
 block/qcow2-refcount.c | 18 +++++++++---------
 1 file changed, 9 insertions(+), 9 deletions(-)

diff --git a/block/qcow2-refcount.c b/block/qcow2-refcount.c
index 1ac2b0490d..6851807ffc 100644
--- a/block/qcow2-refcount.c
+++ b/block/qcow2-refcount.c
@@ -1660,15 +1660,6 @@ static int check_refcounts_l2(BlockDriverState *bs, BdrvCheckResult *res,
         {
             uint64_t offset = l2_entry & L2E_OFFSET_MASK;
 
-            if (flags & CHECK_FRAG_INFO) {
-                res->bfi.allocated_clusters++;
-                if (next_contiguous_offset &&
-                    offset != next_contiguous_offset) {
-                    res->bfi.fragmented_clusters++;
-                }
-                next_contiguous_offset = offset + s->cluster_size;
-            }
-
             /* Correct offsets are cluster aligned */
             if (offset_into_cluster(s, offset)) {
                 if (qcow2_get_cluster_type(bs, l2_entry) ==
@@ -1721,6 +1712,15 @@ static int check_refcounts_l2(BlockDriverState *bs, BdrvCheckResult *res,
                 }
             }
 
+            if (flags & CHECK_FRAG_INFO) {
+                res->bfi.allocated_clusters++;
+                if (next_contiguous_offset &&
+                    offset != next_contiguous_offset) {
+                    res->bfi.fragmented_clusters++;
+                }
+                next_contiguous_offset = offset + s->cluster_size;
+            }
+
             /* Mark cluster as used */
             if (!has_data_file(bs)) {
                 ret = qcow2_inc_refcounts_imrt(bs, res, refcount_table,
-- 
2.20.1

[Qemu-devel] [PULL 05/11] qcow2-refcount: don't mask corruptions under internal errors

[Max Reitz]

From: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com>

No reasons for not reporting found corruptions as corruptions in case
of some internal errors, especially in case of just failed to fix l2
entry (and in this case, missed corruptions may influence comparing
logic, when we calculate difference between corruptions fields of two
results)

Signed-off-by: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com>
Message-id: 20190227131433.197063-6-vsementsov@virtuozzo.com
Reviewed-by: Max Reitz <mreitz@redhat.com>
Signed-off-by: Max Reitz <mreitz@redhat.com>
---
 block/qcow2-refcount.c | 19 +++++++++----------
 1 file changed, 9 insertions(+), 10 deletions(-)

diff --git a/block/qcow2-refcount.c b/block/qcow2-refcount.c
index 6851807ffc..fa7ac1f7cb 100644
--- a/block/qcow2-refcount.c
+++ b/block/qcow2-refcount.c
@@ -1662,6 +1662,8 @@ static int check_refcounts_l2(BlockDriverState *bs, BdrvCheckResult *res,
 
             /* Correct offsets are cluster aligned */
             if (offset_into_cluster(s, offset)) {
+                res->corruptions++;
+
                 if (qcow2_get_cluster_type(bs, l2_entry) ==
                     QCOW2_CLUSTER_ZERO_ALLOC)
                 {
@@ -1697,18 +1699,16 @@ static int check_refcounts_l2(BlockDriverState *bs, BdrvCheckResult *res,
                             /* Do not abort, continue checking the rest of this
                              * L2 table's entries */
                         } else {
+                            res->corruptions--;
                             res->corruptions_fixed++;
                             /* Skip marking the cluster as used
                              * (it is unused now) */
                             continue;
                         }
-                    } else {
-                        res->corruptions++;
                     }
                 } else {
                     fprintf(stderr, "ERROR offset=%" PRIx64 ": Data cluster is "
                         "not properly aligned; L2 entry corrupted.\n", offset);
-                    res->corruptions++;
                 }
             }
 
@@ -1879,6 +1879,7 @@ static int check_oflag_copied(BlockDriverState *bs, BdrvCheckResult *res,
             continue;
         }
         if ((refcount == 1) != ((l1_entry & QCOW_OFLAG_COPIED) != 0)) {
+            res->corruptions++;
             fprintf(stderr, "%s OFLAG_COPIED L2 cluster: l1_index=%d "
                     "l1_entry=%" PRIx64 " refcount=%" PRIu64 "\n",
                     repair ? "Repairing" : "ERROR", i, l1_entry, refcount);
@@ -1891,9 +1892,8 @@ static int check_oflag_copied(BlockDriverState *bs, BdrvCheckResult *res,
                     res->check_errors++;
                     goto fail;
                 }
+                res->corruptions--;
                 res->corruptions_fixed++;
-            } else {
-                res->corruptions++;
             }
         }
 
@@ -1925,6 +1925,7 @@ static int check_oflag_copied(BlockDriverState *bs, BdrvCheckResult *res,
                     }
                 }
                 if ((refcount == 1) != ((l2_entry & QCOW_OFLAG_COPIED) != 0)) {
+                    res->corruptions++;
                     fprintf(stderr, "%s OFLAG_COPIED data cluster: "
                             "l2_entry=%" PRIx64 " refcount=%" PRIu64 "\n",
                             repair ? "Repairing" : "ERROR", l2_entry, refcount);
@@ -1933,8 +1934,6 @@ static int check_oflag_copied(BlockDriverState *bs, BdrvCheckResult *res,
                                     ? l2_entry |  QCOW_OFLAG_COPIED
                                     : l2_entry & ~QCOW_OFLAG_COPIED);
                         l2_dirty++;
-                    } else {
-                        res->corruptions++;
                     }
                 }
             }
@@ -1959,6 +1958,7 @@ static int check_oflag_copied(BlockDriverState *bs, BdrvCheckResult *res,
                 res->check_errors++;
                 goto fail;
             }
+            res->corruptions -= l2_dirty;
             res->corruptions_fixed += l2_dirty;
         }
     }
@@ -1997,6 +1997,7 @@ static int check_refblocks(BlockDriverState *bs, BdrvCheckResult *res,
         }
 
         if (cluster >= *nb_clusters) {
+            res->corruptions++;
             fprintf(stderr, "%s refcount block %" PRId64 " is outside image\n",
                     fix & BDRV_FIX_ERRORS ? "Repairing" : "ERROR", i);
 
@@ -2036,6 +2037,7 @@ static int check_refblocks(BlockDriverState *bs, BdrvCheckResult *res,
                     goto resize_fail;
                 }
 
+                res->corruptions--;
                 res->corruptions_fixed++;
                 ret = qcow2_inc_refcounts_imrt(bs, res,
                                                refcount_table, nb_clusters,
@@ -2049,12 +2051,9 @@ static int check_refblocks(BlockDriverState *bs, BdrvCheckResult *res,
                 continue;
 
 resize_fail:
-                res->corruptions++;
                 *rebuild = true;
                 fprintf(stderr, "ERROR could not resize image: %s\n",
                         strerror(-ret));
-            } else {
-                res->corruptions++;
             }
             continue;
         }
-- 
2.20.1

[Qemu-devel] [PULL 06/11] qcow2: discard bitmap when removed

[Max Reitz]

From: Andrey Shinkevich <andrey.shinkevich@virtuozzo.com>

Bitmap data may take a lot of disk space, so it's better to discard it
always.

Signed-off-by: Andrey Shinkevich <andrey.shinkevich@virtuozzo.com>
Message-id: 1551346019-293202-1-git-send-email-andrey.shinkevich@virtuozzo.com
Reviewed-by: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com>
[mreitz: Use the commit message proposed by Vladimir]
Signed-off-by: Max Reitz <mreitz@redhat.com>
---
 block/qcow2-bitmap.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/block/qcow2-bitmap.c b/block/qcow2-bitmap.c
index e53a1609d7..8a75366c92 100644
--- a/block/qcow2-bitmap.c
+++ b/block/qcow2-bitmap.c
@@ -202,7 +202,7 @@ static void clear_bitmap_table(BlockDriverState *bs, uint64_t *bitmap_table,
             continue;
         }
 
-        qcow2_free_clusters(bs, addr, s->cluster_size, QCOW2_DISCARD_OTHER);
+        qcow2_free_clusters(bs, addr, s->cluster_size, QCOW2_DISCARD_ALWAYS);
         bitmap_table[i] = 0;
     }
 }
-- 
2.20.1

[Qemu-devel] [PULL 07/11] block/ssh: Implement .bdrv_refresh_filename()

[Max Reitz]

This requires some changes to keep iotests 104 and 207 working.

qemu-img info in 104 will now return a filename including the user name
and the port, which need to be filtered by adjusting REMOTE_TEST_DIR in
common.rc.  This additional information has to be marked optional,
however (which is simple as REMOTE_TEST_DIR is a regex), because
otherwise 197 and 215 would fail: They use it (indirectly) to filter
qemu-img create output which contains a backing filename they have
passed to it -- which probably does not contain a user name or port
number.

The problem in 207 is a nice one to have: qemu-img info used to return
json:{} filenames, but with this patch it returns nice plain ones.  We
now need to adjust the filtering to hide the user name (and port number
while we are at it).  The simplest way to do this is to include both in
iotests.remote_filename() so that bdrv_refresh_filename() will not
change it, and then iotests.img_info_log() will filter it correctly
automatically.

Signed-off-by: Max Reitz <mreitz@redhat.com>
Tested-by: Richard W.M. Jones <rjones@redhat.com>
Message-id: 20190225190828.17726-2-mreitz@redhat.com
Signed-off-by: Max Reitz <mreitz@redhat.com>
---
 block/ssh.c                   | 52 +++++++++++++++++++++++++++++++----
 tests/qemu-iotests/207        | 10 +++----
 tests/qemu-iotests/207.out    | 10 +++----
 tests/qemu-iotests/common.rc  |  2 +-
 tests/qemu-iotests/iotests.py |  2 +-
 5 files changed, 59 insertions(+), 17 deletions(-)

diff --git a/block/ssh.c b/block/ssh.c
index 859249113d..2eaeab84d5 100644
--- a/block/ssh.c
+++ b/block/ssh.c
@@ -75,6 +75,14 @@ typedef struct BDRVSSHState {
 
     /* Used to warn if 'flush' is not supported. */
     bool unsafe_flush_warning;
+
+    /*
+     * Store the user name for ssh_refresh_filename() because the
+     * default depends on the system you are on -- therefore, when we
+     * generate a filename, it should always contain the user name we
+     * are actually using.
+     */
+    char *user;
 } BDRVSSHState;
 
 static void ssh_state_init(BDRVSSHState *s)
@@ -87,6 +95,8 @@ static void ssh_state_init(BDRVSSHState *s)
 
 static void ssh_state_free(BDRVSSHState *s)
 {
+    g_free(s->user);
+
     if (s->sftp_handle) {
         libssh2_sftp_close(s->sftp_handle);
     }
@@ -628,14 +638,13 @@ static int connect_to_ssh(BDRVSSHState *s, BlockdevOptionsSsh *opts,
                           int ssh_flags, int creat_mode, Error **errp)
 {
     int r, ret;
-    const char *user;
     long port = 0;
 
     if (opts->has_user) {
-        user = opts->user;
+        s->user = g_strdup(opts->user);
     } else {
-        user = g_get_user_name();
-        if (!user) {
+        s->user = g_strdup(g_get_user_name());
+        if (!s->user) {
             error_setg_errno(errp, errno, "Can't get user name");
             ret = -errno;
             goto err;
@@ -685,7 +694,7 @@ static int connect_to_ssh(BDRVSSHState *s, BlockdevOptionsSsh *opts,
     }
 
     /* Authenticate. */
-    ret = authenticate(s, user, errp);
+    ret = authenticate(s, s->user, errp);
     if (ret < 0) {
         goto err;
     }
@@ -1242,6 +1251,38 @@ static int coroutine_fn ssh_co_truncate(BlockDriverState *bs, int64_t offset,
     return ssh_grow_file(s, offset, errp);
 }
 
+static void ssh_refresh_filename(BlockDriverState *bs)
+{
+    BDRVSSHState *s = bs->opaque;
+    const char *path, *host_key_check;
+    int ret;
+
+    /*
+     * None of these options can be represented in a plain "host:port"
+     * format, so if any was given, we have to abort.
+     */
+    if (s->inet->has_ipv4 || s->inet->has_ipv6 || s->inet->has_to ||
+        s->inet->has_numeric)
+    {
+        return;
+    }
+
+    path = qdict_get_try_str(bs->full_open_options, "path");
+    assert(path); /* mandatory option */
+
+    host_key_check = qdict_get_try_str(bs->full_open_options, "host_key_check");
+
+    ret = snprintf(bs->exact_filename, sizeof(bs->exact_filename),
+                   "ssh://%s@%s:%s%s%s%s",
+                   s->user, s->inet->host, s->inet->port, path,
+                   host_key_check ? "?host_key_check=" : "",
+                   host_key_check ?: "");
+    if (ret >= sizeof(bs->exact_filename)) {
+        /* An overflow makes the filename unusable, so do not report any */
+        bs->exact_filename[0] = '\0';
+    }
+}
+
 static const char *const ssh_strong_runtime_opts[] = {
     "host",
     "port",
@@ -1268,6 +1309,7 @@ static BlockDriver bdrv_ssh = {
     .bdrv_getlength               = ssh_getlength,
     .bdrv_co_truncate             = ssh_co_truncate,
     .bdrv_co_flush_to_disk        = ssh_co_flush,
+    .bdrv_refresh_filename        = ssh_refresh_filename,
     .create_opts                  = &ssh_create_opts,
     .strong_runtime_opts          = ssh_strong_runtime_opts,
 };
diff --git a/tests/qemu-iotests/207 b/tests/qemu-iotests/207
index dfd3c51bd1..b3816136f7 100755
--- a/tests/qemu-iotests/207
+++ b/tests/qemu-iotests/207
@@ -66,7 +66,7 @@ with iotests.FilePath('t.img') as disk_path, \
                           'size': 4194304 })
     vm.shutdown()
 
-    iotests.img_info_log(remote_path, filter_path=disk_path)
+    iotests.img_info_log(remote_path)
     iotests.log("")
     iotests.img_info_log(disk_path)
 
@@ -91,7 +91,7 @@ with iotests.FilePath('t.img') as disk_path, \
                           'size': 8388608 })
     vm.shutdown()
 
-    iotests.img_info_log(remote_path, filter_path=disk_path)
+    iotests.img_info_log(remote_path)
 
     vm.launch()
     blockdev_create(vm, { 'driver': 'ssh',
@@ -108,7 +108,7 @@ with iotests.FilePath('t.img') as disk_path, \
                           'size': 4194304 })
     vm.shutdown()
 
-    iotests.img_info_log(remote_path, filter_path=disk_path)
+    iotests.img_info_log(remote_path)
 
     md5_key = subprocess.check_output(
         'ssh-keyscan -t rsa 127.0.0.1 2>/dev/null | grep -v "\\^#" | ' +
@@ -146,7 +146,7 @@ with iotests.FilePath('t.img') as disk_path, \
                           'size': 8388608 })
     vm.shutdown()
 
-    iotests.img_info_log(remote_path, filter_path=disk_path)
+    iotests.img_info_log(remote_path)
 
     sha1_key = subprocess.check_output(
         'ssh-keyscan -t rsa 127.0.0.1 2>/dev/null | grep -v "\\^#" | ' +
@@ -184,7 +184,7 @@ with iotests.FilePath('t.img') as disk_path, \
                           'size': 4194304 })
     vm.shutdown()
 
-    iotests.img_info_log(remote_path, filter_path=disk_path)
+    iotests.img_info_log(remote_path)
 
     #
     # Invalid path and user
diff --git a/tests/qemu-iotests/207.out b/tests/qemu-iotests/207.out
index 979d5cf745..ec9823793a 100644
--- a/tests/qemu-iotests/207.out
+++ b/tests/qemu-iotests/207.out
@@ -5,7 +5,7 @@
 {"execute": "job-dismiss", "arguments": {"id": "job0"}}
 {"return": {}}
 
-image: json:{"driver": "IMGFMT", "file": {"server.host": "127.0.0.1", "server.port": "22", "driver": "ssh", "path": "TEST_IMG"}}
+image: TEST_IMG
 file format: IMGFMT
 virtual size: 4 MiB (4194304 bytes)
 
@@ -21,7 +21,7 @@ virtual size: 4 MiB (4194304 bytes)
 {"execute": "job-dismiss", "arguments": {"id": "job0"}}
 {"return": {}}
 
-image: json:{"driver": "IMGFMT", "file": {"server.host": "127.0.0.1", "server.port": "22", "driver": "ssh", "path": "TEST_IMG"}}
+image: TEST_IMG
 file format: IMGFMT
 virtual size: 8 MiB (8388608 bytes)
 
@@ -30,7 +30,7 @@ virtual size: 8 MiB (8388608 bytes)
 {"execute": "job-dismiss", "arguments": {"id": "job0"}}
 {"return": {}}
 
-image: json:{"driver": "IMGFMT", "file": {"server.host": "127.0.0.1", "server.port": "22", "driver": "ssh", "path": "TEST_IMG"}}
+image: TEST_IMG
 file format: IMGFMT
 virtual size: 4 MiB (4194304 bytes)
 
@@ -45,7 +45,7 @@ Job failed: remote host key does not match host_key_check 'wrong'
 {"execute": "job-dismiss", "arguments": {"id": "job0"}}
 {"return": {}}
 
-image: json:{"driver": "IMGFMT", "file": {"server.host": "127.0.0.1", "server.port": "22", "driver": "ssh", "path": "TEST_IMG"}}
+image: TEST_IMG
 file format: IMGFMT
 virtual size: 8 MiB (8388608 bytes)
 
@@ -60,7 +60,7 @@ Job failed: remote host key does not match host_key_check 'wrong'
 {"execute": "job-dismiss", "arguments": {"id": "job0"}}
 {"return": {}}
 
-image: json:{"driver": "IMGFMT", "file": {"server.host": "127.0.0.1", "server.port": "22", "driver": "ssh", "path": "TEST_IMG"}}
+image: TEST_IMG
 file format: IMGFMT
 virtual size: 4 MiB (4194304 bytes)
 
diff --git a/tests/qemu-iotests/common.rc b/tests/qemu-iotests/common.rc
index a543e546c2..93f87389b6 100644
--- a/tests/qemu-iotests/common.rc
+++ b/tests/qemu-iotests/common.rc
@@ -158,7 +158,7 @@ else
         TEST_IMG="nbd:127.0.0.1:10810"
     elif [ "$IMGPROTO" = "ssh" ]; then
         TEST_IMG_FILE=$TEST_DIR/t.$IMGFMT
-        REMOTE_TEST_DIR="ssh://127.0.0.1$TEST_DIR"
+        REMOTE_TEST_DIR="ssh://\\($USER@\\)\\?127.0.0.1\\(:[0-9]\\+\\)\\?$TEST_DIR"
         TEST_IMG="ssh://127.0.0.1$TEST_IMG_FILE"
     elif [ "$IMGPROTO" = "nfs" ]; then
         TEST_IMG_FILE=$TEST_DIR/t.$IMGFMT
diff --git a/tests/qemu-iotests/iotests.py b/tests/qemu-iotests/iotests.py
index 997dc910cb..f811f69135 100644
--- a/tests/qemu-iotests/iotests.py
+++ b/tests/qemu-iotests/iotests.py
@@ -411,7 +411,7 @@ def remote_filename(path):
     if imgproto == 'file':
         return path
     elif imgproto == 'ssh':
-        return "ssh://127.0.0.1%s" % (path)
+        return "ssh://%s@127.0.0.1:22%s" % (os.environ.get('USER'), path)
     else:
         raise Exception("Protocol %s not supported" % (imgproto))
 
-- 
2.20.1

[Qemu-devel] [PULL 08/11] block/ssh: Implement .bdrv_dirname()

[Max Reitz]

ssh_bdrv_dirname() is basically the generic bdrv_dirname(), except it
takes care not to silently chop off any query string (i.e.,
host_key_check).

Signed-off-by: Max Reitz <mreitz@redhat.com>
Tested-by: Richard W.M. Jones <rjones@redhat.com>
Message-id: 20190225190828.17726-3-mreitz@redhat.com
Signed-off-by: Max Reitz <mreitz@redhat.com>
---
 block/ssh.c | 21 +++++++++++++++++++++
 1 file changed, 21 insertions(+)

diff --git a/block/ssh.c b/block/ssh.c
index 2eaeab84d5..12fd4f39e8 100644
--- a/block/ssh.c
+++ b/block/ssh.c
@@ -1283,6 +1283,26 @@ static void ssh_refresh_filename(BlockDriverState *bs)
     }
 }
 
+static char *ssh_bdrv_dirname(BlockDriverState *bs, Error **errp)
+{
+    if (qdict_haskey(bs->full_open_options, "host_key_check")) {
+        /*
+         * We cannot generate a simple prefix if we would have to
+         * append a query string.
+         */
+        error_setg(errp,
+                   "Cannot generate a base directory with host_key_check set");
+        return NULL;
+    }
+
+    if (bs->exact_filename[0] == '\0') {
+        error_setg(errp, "Cannot generate a base directory for this ssh node");
+        return NULL;
+    }
+
+    return path_combine(bs->exact_filename, "");
+}
+
 static const char *const ssh_strong_runtime_opts[] = {
     "host",
     "port",
@@ -1310,6 +1330,7 @@ static BlockDriver bdrv_ssh = {
     .bdrv_co_truncate             = ssh_co_truncate,
     .bdrv_co_flush_to_disk        = ssh_co_flush,
     .bdrv_refresh_filename        = ssh_refresh_filename,
+    .bdrv_dirname                 = ssh_bdrv_dirname,
     .create_opts                  = &ssh_create_opts,
     .strong_runtime_opts          = ssh_strong_runtime_opts,
 };
-- 
2.20.1

[Qemu-devel] [PULL 09/11] block: Assert that drv->bdrv_child_perm is set in bdrv_child_perm()

[Max Reitz]

From: Alberto Garcia <berto@igalia.com>

There is no need to check for this because all block drivers that have
children implement bdrv_child_perm and all callers already ensure that
bs->drv is set.

Furthermore, if this check would fail then the callers would end up
with uninitialized values for nperm and nshared.

This patch replaces the check with an assertion.

Signed-off-by: Alberto Garcia <berto@igalia.com>
Message-id: 20190404112953.4058-1-berto@igalia.com
Signed-off-by: Max Reitz <mreitz@redhat.com>
---
 block.c | 9 ++++-----
 1 file changed, 4 insertions(+), 5 deletions(-)

diff --git a/block.c b/block.c
index 9ae5c0ed2f..7dc8fe289a 100644
--- a/block.c
+++ b/block.c
@@ -1743,11 +1743,10 @@ static void bdrv_child_perm(BlockDriverState *bs, BlockDriverState *child_bs,
                             uint64_t parent_perm, uint64_t parent_shared,
                             uint64_t *nperm, uint64_t *nshared)
 {
-    if (bs->drv && bs->drv->bdrv_child_perm) {
-        bs->drv->bdrv_child_perm(bs, c, role, reopen_queue,
-                                 parent_perm, parent_shared,
-                                 nperm, nshared);
-    }
+    assert(bs->drv && bs->drv->bdrv_child_perm);
+    bs->drv->bdrv_child_perm(bs, c, role, reopen_queue,
+                             parent_perm, parent_shared,
+                             nperm, nshared);
     /* TODO Take force_share from reopen_queue */
     if (child_bs && child_bs->force_share) {
         *nshared = BLK_PERM_ALL;
-- 
2.20.1

[Qemu-devel] [PULL 10/11] commit: Use bdrv_append() in commit_start()

[Max Reitz]

From: Alberto Garcia <berto@igalia.com>

This function combines bdrv_set_backing_hd() and bdrv_replace_node()
so we can use it to simplify the code a bit in commit_start().

Signed-off-by: Alberto Garcia <berto@igalia.com>
Message-id: 20190403143748.9790-1-berto@igalia.com
Signed-off-by: Max Reitz <mreitz@redhat.com>
---
 block/commit.c | 11 +----------
 1 file changed, 1 insertion(+), 10 deletions(-)

diff --git a/block/commit.c b/block/commit.c
index 27537d995b..14e5bb394c 100644
--- a/block/commit.c
+++ b/block/commit.c
@@ -303,23 +303,14 @@ void commit_start(const char *job_id, BlockDriverState *bs,
     commit_top_bs->total_sectors = top->total_sectors;
     bdrv_set_aio_context(commit_top_bs, bdrv_get_aio_context(top));
 
-    bdrv_set_backing_hd(commit_top_bs, top, &local_err);
+    bdrv_append(commit_top_bs, top, &local_err);
     if (local_err) {
-        bdrv_unref(commit_top_bs);
-        commit_top_bs = NULL;
-        error_propagate(errp, local_err);
-        goto fail;
-    }
-    bdrv_replace_node(top, commit_top_bs, &local_err);
-    if (local_err) {
-        bdrv_unref(commit_top_bs);
         commit_top_bs = NULL;
         error_propagate(errp, local_err);
         goto fail;
     }
 
     s->commit_top_bs = commit_top_bs;
-    bdrv_unref(commit_top_bs);
 
     /* Block all nodes between top and base, because they will
      * disappear from the chain after this operation. */
-- 
2.20.1

[Qemu-devel] [PULL 11/11] iotests: Fix iotests 110 and 126

[Max Reitz]

A recent patch results in qemu-img reporting the backing file format of
vmdk images as vmdk.  This broke iotests 110 and 126.

Fixes: 7502be838e2fb62cc00f9e55f632e0b88ead5e6e
Signed-off-by: Max Reitz <mreitz@redhat.com>
Message-id: 20190415154129.31021-1-mreitz@redhat.com
Signed-off-by: Max Reitz <mreitz@redhat.com>
---
 tests/qemu-iotests/110 | 10 +++++++---
 tests/qemu-iotests/126 | 10 +++++++---
 2 files changed, 14 insertions(+), 6 deletions(-)

diff --git a/tests/qemu-iotests/110 b/tests/qemu-iotests/110
index fad672c1ae..33b169ffd4 100755
--- a/tests/qemu-iotests/110
+++ b/tests/qemu-iotests/110
@@ -53,8 +53,12 @@ TEST_IMG="$TEST_IMG.base" _make_test_img 64M
 _make_test_img -b "$TEST_IMG_REL.base" 64M
 # qemu should be able to reconstruct the filename, so relative backing names
 # should work
+# (We have to filter the backing file format because vmdk always
+# reports it (as vmdk), whereas other image formats would do so only
+# with the backing_fmt creation option, which neither vmdk nor qcow
+# support)
 TEST_IMG="json:{'driver':'$IMGFMT','file':{'driver':'file','filename':'$TEST_IMG'}}" \
-    _img_info | _filter_img_info
+    _img_info | _filter_img_info | grep -v 'backing file format'
 
 echo
 echo '=== Non-reconstructable filename ==='
@@ -78,7 +82,7 @@ TEST_IMG="json:{
             }
         ]
     }
-}" _img_info | _filter_img_info
+}" _img_info | _filter_img_info | grep -v 'backing file format'
 
 echo
 echo '=== Backing name is always relative to the backed image ==='
@@ -110,7 +114,7 @@ TEST_IMG="json:{
             }
         ]
     }
-}" _img_info | _filter_img_info
+}" _img_info | _filter_img_info | grep -v 'backing file format'
 
 
 # success, all done
diff --git a/tests/qemu-iotests/126 b/tests/qemu-iotests/126
index 96dc048d59..e3ee65c606 100755
--- a/tests/qemu-iotests/126
+++ b/tests/qemu-iotests/126
@@ -62,8 +62,12 @@ TOP_IMG="$TEST_DIR/image:top.$IMGFMT"
 TEST_IMG=$BASE_IMG _make_test_img 64M
 TEST_IMG=$TOP_IMG _make_test_img -b ./image:base.$IMGFMT
 
-# The default cluster size depends on the image format
-TEST_IMG=$TOP_IMG _img_info | grep -v 'cluster_size'
+# (1) The default cluster size depends on the image format
+# (2) vmdk only supports vmdk backing files, so it always reports the
+#     format of its backing file as such (but neither it nor qcow
+#     support the backing_fmt creation option, so we cannot use that to
+#     harmonize the output across all image formats this test supports)
+TEST_IMG=$TOP_IMG _img_info | grep -ve 'cluster_size' -e 'backing file format'
 
 _rm_test_img "$BASE_IMG"
 _rm_test_img "$TOP_IMG"
@@ -79,7 +83,7 @@ TOP_IMG="file:image:top.$IMGFMT"
 TEST_IMG=$BASE_IMG _make_test_img 64M
 TEST_IMG=$TOP_IMG _make_test_img -b "$BASE_IMG"
 
-TEST_IMG=$TOP_IMG _img_info | grep -v 'cluster_size'
+TEST_IMG=$TOP_IMG _img_info | grep -ve 'cluster_size' -e 'backing file format'
 
 _rm_test_img "$BASE_IMG"
 _rm_test_img "image:top.$IMGFMT"
-- 
2.20.1

Re: [Qemu-devel] [PULL 00/11] Block patches

[Peter Maydell]

On Tue, 7 May 2019 at 16:18, Max Reitz <mreitz@redhat.com> wrote:
>
> The following changes since commit 19eb2d4e736dc895f31fbd6b520e514f10cc08e0:
>
>   Merge remote-tracking branch 'remotes/thibault/tags/samuel-thibault' into staging (2019-05-07 10:43:32 +0100)
>
> are available in the Git repository at:
>
>   https://git.xanclic.moe/XanClic/qemu.git tags/pull-block-2019-05-07
>
> for you to fetch changes up to 1278dce7927301bf3d004a40061dbd2c1e0846a8:
>
>   iotests: Fix iotests 110 and 126 (2019-05-07 17:14:21 +0200)

Attempting to fetch from this remote hangs:

$ git fetch -v xanclic
POST git-upload-pack (gzip 1798 to 966 bytes)
POST git-upload-pack (gzip 1798 to 965 bytes)
POST git-upload-pack (gzip 2648 to 1393 bytes)
POST git-upload-pack (gzip 4248 to 2201 bytes)
POST git-upload-pack (gzip 7498 to 3833 bytes)
POST git-upload-pack (gzip 13998 to 7092 bytes)
POST git-upload-pack (gzip 27648 to 13930 bytes)
POST git-upload-pack (gzip 55148 to 27782 bytes)
POST git-upload-pack (gzip 108948 to 54371 bytes)
POST git-upload-pack (gzip 215798 to 107233 bytes)
[no further output]

thanks
-- PMM

Re: [Qemu-devel] [PULL 00/11] Block patches

[Max Reitz]

[-- Attachment #1: Type: text/plain, Size: 1441 bytes --]

On 09.05.19 10:49, Peter Maydell wrote:
> On Tue, 7 May 2019 at 16:18, Max Reitz <mreitz@redhat.com> wrote:
>>
>> The following changes since commit 19eb2d4e736dc895f31fbd6b520e514f10cc08e0:
>>
>>   Merge remote-tracking branch 'remotes/thibault/tags/samuel-thibault' into staging (2019-05-07 10:43:32 +0100)
>>
>> are available in the Git repository at:
>>
>>   https://git.xanclic.moe/XanClic/qemu.git tags/pull-block-2019-05-07
>>
>> for you to fetch changes up to 1278dce7927301bf3d004a40061dbd2c1e0846a8:
>>
>>   iotests: Fix iotests 110 and 126 (2019-05-07 17:14:21 +0200)
> 
> Attempting to fetch from this remote hangs:
> 
> $ git fetch -v xanclic
> POST git-upload-pack (gzip 1798 to 966 bytes)
> POST git-upload-pack (gzip 1798 to 965 bytes)
> POST git-upload-pack (gzip 2648 to 1393 bytes)
> POST git-upload-pack (gzip 4248 to 2201 bytes)
> POST git-upload-pack (gzip 7498 to 3833 bytes)
> POST git-upload-pack (gzip 13998 to 7092 bytes)
> POST git-upload-pack (gzip 27648 to 13930 bytes)
> POST git-upload-pack (gzip 55148 to 27782 bytes)
> POST git-upload-pack (gzip 108948 to 54371 bytes)
> POST git-upload-pack (gzip 215798 to 107233 bytes)
> [no further output]

Hm, that's unfortunate.  It works for me.  (At least now.)

I've pushed the tag to Github:

https://github.com/XanClic/qemu.git tags/pull-block-2019-05-07

I hope that works better.


Sorry for the inconvenience,

Max


[-- Attachment #2: OpenPGP digital signature --]
[-- Type: application/pgp-signature, Size: 488 bytes --]

Re: [Qemu-devel] [PULL 00/11] Block patches

[Peter Maydell]

On Thu, 9 May 2019 at 14:27, Max Reitz <mreitz@redhat.com> wrote:
>
> On 09.05.19 10:49, Peter Maydell wrote:
> > On Tue, 7 May 2019 at 16:18, Max Reitz <mreitz@redhat.com> wrote:
> >>
> >> The following changes since commit 19eb2d4e736dc895f31fbd6b520e514f10cc08e0:
> >>
> >>   Merge remote-tracking branch 'remotes/thibault/tags/samuel-thibault' into staging (2019-05-07 10:43:32 +0100)
> >>
> >> are available in the Git repository at:
> >>
> >>   https://git.xanclic.moe/XanClic/qemu.git tags/pull-block-2019-05-07
> >>
> >> for you to fetch changes up to 1278dce7927301bf3d004a40061dbd2c1e0846a8:
> >>
> >>   iotests: Fix iotests 110 and 126 (2019-05-07 17:14:21 +0200)
> >
> > Attempting to fetch from this remote hangs:
> >
> > $ git fetch -v xanclic
> > POST git-upload-pack (gzip 1798 to 966 bytes)
> > POST git-upload-pack (gzip 1798 to 965 bytes)
> > POST git-upload-pack (gzip 2648 to 1393 bytes)
> > POST git-upload-pack (gzip 4248 to 2201 bytes)
> > POST git-upload-pack (gzip 7498 to 3833 bytes)
> > POST git-upload-pack (gzip 13998 to 7092 bytes)
> > POST git-upload-pack (gzip 27648 to 13930 bytes)
> > POST git-upload-pack (gzip 55148 to 27782 bytes)
> > POST git-upload-pack (gzip 108948 to 54371 bytes)
> > POST git-upload-pack (gzip 215798 to 107233 bytes)
> > [no further output]
>
> Hm, that's unfortunate.  It works for me.  (At least now.)
>
> I've pushed the tag to Github:
>
> https://github.com/XanClic/qemu.git tags/pull-block-2019-05-07
>
> I hope that works better.

Yep, that worked fine: I've applied that.

Please update the changelog at https://wiki.qemu.org/ChangeLog/4.1
for any user-visible changes.

(Strace says that we seem to successfully connect to
the git.xanclic.moe server and do some talking to it,
but the first time the client issues a "fetch" command
the server never replies and the client is just blocked
in read().)

thanks
-- PMM