Merge remote-tracking branch 'remotes/maxreitz/tags/pull-block-2019-09-03' into staging

Block patches:
- qemu-io now accepts a file to read a write pattern from
- Ensure that raw files have their first block allocated so we can probe
  the O_DIRECT alignment if necessary
- Various fixes

# gpg: Signature made Tue 03 Sep 2019 13:58:57 BST
# gpg:                using RSA key 91BEB60A30DB3E8857D11829F407DB0061D5CF40
# gpg:                issuer "mreitz@redhat.com"
# gpg: Good signature from "Max Reitz <mreitz@redhat.com>" [full]
# Primary key fingerprint: 91BE B60A 30DB 3E88 57D1  1829 F407 DB00 61D5 CF40

* remotes/maxreitz/tags/pull-block-2019-09-03:
  iotests: Unify cache mode quoting
  tests/check-block: Skip iotests when sanitizers are enabled
  iotests: Check for enabled drivers before testing them
  iotests: Add -display none to the qemu options
  file-posix: fix request_alignment typo
  iotests: Disable 126 for flat vmdk subformats
  iotests: Disable 110 for vmdk.twoGbMaxExtentSparse
  iotests: Disable broken streamOptimized tests
  vmdk: Reject invalid compressed writes
  iotests: Keep testing broken relative extent paths
  vmdk: Use bdrv_dirname() for relative extent paths
  iotests: Fix _filter_img_create()
  iotests: Test allocate_first_block() with O_DIRECT
  block: posix: Always allocate the first block
  block: fix permission update in bdrv_replace_node
  qemu-io: add pattern file for write command

Signed-off-by: Peter Maydell <peter.maydell@linaro.org>

block.c

@@ -4165,7 +4165,6 @@ void bdrv_replace_node(BlockDriverState *from, BlockDriverState *to,
 {
     BdrvChild *c, *next;
     GSList *list = NULL, *p;
-    uint64_t old_perm, old_shared;
     uint64_t perm = 0, shared = BLK_PERM_ALL;
     int ret;
 
@@ -4211,8 +4210,8 @@ void bdrv_replace_node(BlockDriverState *from, BlockDriverState *to,
         bdrv_unref(from);
     }
 
-    bdrv_get_cumulative_perm(to, &old_perm, &old_shared);
-    bdrv_set_perm(to, old_perm | perm, old_shared | shared);
+    bdrv_get_cumulative_perm(to, &perm, &shared);
+    bdrv_set_perm(to, perm, shared);
 
 out:
     g_slist_free(list);

block/file-posix.c

@@ -380,7 +380,7 @@ static void raw_probe_alignment(BlockDriverState *bs, int fd, Error **errp)
         for (i = 0; i < ARRAY_SIZE(alignments); i++) {
             align = alignments[i];
             if (raw_is_io_aligned(fd, buf + align, max_align)) {
-                /* Fallback to request_aligment. */
+                /* Fallback to request_alignment. */
                 s->buf_align = (align != 1) ? align : bs->bl.request_alignment;
                 break;
             }
@@ -1749,6 +1749,43 @@ static int handle_aiocb_discard(void *opaque)
     return ret;
 }
 
+/*
+ * Help alignment probing by allocating the first block.
+ *
+ * When reading with direct I/O from unallocated area on Gluster backed by XFS,
+ * reading succeeds regardless of request length. In this case we fallback to
+ * safe alignment which is not optimal. Allocating the first block avoids this
+ * fallback.
+ *
+ * fd may be opened with O_DIRECT, but we don't know the buffer alignment or
+ * request alignment, so we use safe values.
+ *
+ * Returns: 0 on success, -errno on failure. Since this is an optimization,
+ * caller may ignore failures.
+ */
+static int allocate_first_block(int fd, size_t max_size)
+{
+    size_t write_size = (max_size < MAX_BLOCKSIZE)
+        ? BDRV_SECTOR_SIZE
+        : MAX_BLOCKSIZE;
+    size_t max_align = MAX(MAX_BLOCKSIZE, getpagesize());
+    void *buf;
+    ssize_t n;
+    int ret;
+
+    buf = qemu_memalign(max_align, write_size);
+    memset(buf, 0, write_size);
+
+    do {
+        n = pwrite(fd, buf, write_size, 0);
+    } while (n == -1 && errno == EINTR);
+
+    ret = (n == -1) ? -errno : 0;
+
+    qemu_vfree(buf);
+    return ret;
+}
+
 static int handle_aiocb_truncate(void *opaque)
 {
     RawPosixAIOData *aiocb = opaque;
@@ -1788,6 +1825,17 @@ static int handle_aiocb_truncate(void *opaque)
                 /* posix_fallocate() doesn't set errno. */
                 error_setg_errno(errp, -result,
                                  "Could not preallocate new data");
+            } else if (current_length == 0) {
+                /*
+                 * posix_fallocate() uses fallocate() if the filesystem
+                 * supports it, or fallback to manually writing zeroes. If
+                 * fallocate() was used, unaligned reads from the fallocated
+                 * area in raw_probe_alignment() will succeed, hence we need to
+                 * allocate the first block.
+                 *
+                 * Optimize future alignment probing; ignore failures.
+                 */
+                allocate_first_block(fd, offset);
             }
         } else {
             result = 0;
@@ -1849,6 +1897,9 @@ static int handle_aiocb_truncate(void *opaque)
         if (ftruncate(fd, offset) != 0) {
             result = -errno;
             error_setg_errno(errp, -result, "Could not resize file");
+        } else if (current_length == 0 && offset > current_length) {
+            /* Optimize future alignment probing; ignore failures. */
+            allocate_first_block(fd, offset);
         }
         return result;
     default:

block/vmdk.c

@@ -1076,8 +1076,7 @@ static const char *next_line(const char *s)
 }
 
 static int vmdk_parse_extents(const char *desc, BlockDriverState *bs,
-                              const char *desc_file_path, QDict *options,
-                              Error **errp)
+                              QDict *options, Error **errp)
 {
     int ret;
     int matches;
@@ -1087,6 +1086,7 @@ static int vmdk_parse_extents(const char *desc, BlockDriverState *bs,
     const char *p, *np;
     int64_t sectors = 0;
     int64_t flat_offset;
+    char *desc_file_dir = NULL;
     char *extent_path;
     BdrvChild *extent_file;
     BDRVVmdkState *s = bs->opaque;
@@ -1130,16 +1130,23 @@ static int vmdk_parse_extents(const char *desc, BlockDriverState *bs,
             continue;
         }
 
-        if (!path_is_absolute(fname) && !path_has_protocol(fname) &&
-            !desc_file_path[0])
-        {
-            bdrv_refresh_filename(bs->file->bs);
-            error_setg(errp, "Cannot use relative extent paths with VMDK "
-                       "descriptor file '%s'", bs->file->bs->filename);
-            return -EINVAL;
-        }
+        if (path_is_absolute(fname)) {
+            extent_path = g_strdup(fname);
+        } else {
+            if (!desc_file_dir) {
+                desc_file_dir = bdrv_dirname(bs->file->bs, errp);
+                if (!desc_file_dir) {
+                    bdrv_refresh_filename(bs->file->bs);
+                    error_prepend(errp, "Cannot use relative paths with VMDK "
+                                  "descriptor file '%s': ",
+                                  bs->file->bs->filename);
+                    ret = -EINVAL;
+                    goto out;
+                }
+            }
 
-        extent_path = path_combine(desc_file_path, fname);
+            extent_path = g_strconcat(desc_file_dir, fname, NULL);
+        }
 
         ret = snprintf(extent_opt_prefix, 32, "extents.%d", s->num_extents);
         assert(ret < 32);
@@ -1149,7 +1156,8 @@ static int vmdk_parse_extents(const char *desc, BlockDriverState *bs,
         g_free(extent_path);
         if (local_err) {
             error_propagate(errp, local_err);
-            return -EINVAL;
+            ret = -EINVAL;
+            goto out;
         }
 
         /* save to extents array */
@@ -1160,7 +1168,7 @@ static int vmdk_parse_extents(const char *desc, BlockDriverState *bs,
                             0, 0, 0, 0, 0, &extent, errp);
             if (ret < 0) {
                 bdrv_unref_child(bs, extent_file);
-                return ret;
+                goto out;
             }
             extent->flat_start_offset = flat_offset << 9;
         } else if (!strcmp(type, "SPARSE") || !strcmp(type, "VMFSSPARSE")) {
@@ -1175,24 +1183,27 @@ static int vmdk_parse_extents(const char *desc, BlockDriverState *bs,
             g_free(buf);
             if (ret) {
                 bdrv_unref_child(bs, extent_file);
-                return ret;
+                goto out;
             }
             extent = &s->extents[s->num_extents - 1];
         } else if (!strcmp(type, "SESPARSE")) {
             ret = vmdk_open_se_sparse(bs, extent_file, bs->open_flags, errp);
             if (ret) {
                 bdrv_unref_child(bs, extent_file);
-                return ret;
+                goto out;
             }
             extent = &s->extents[s->num_extents - 1];
         } else {
             error_setg(errp, "Unsupported extent type '%s'", type);
             bdrv_unref_child(bs, extent_file);
-            return -ENOTSUP;
+            ret = -ENOTSUP;
+            goto out;
         }
         extent->type = g_strdup(type);
     }
-    return 0;
+
+    ret = 0;
+    goto out;
 
 invalid:
     np = next_line(p);
@@ -1201,7 +1212,11 @@ invalid:
         np--;
     }
     error_setg(errp, "Invalid extent line: %.*s", (int)(np - p), p);
-    return -EINVAL;
+    ret = -EINVAL;
+
+out:
+    g_free(desc_file_dir);
+    return ret;
 }
 
 static int vmdk_open_desc_file(BlockDriverState *bs, int flags, char *buf,
@@ -1228,8 +1243,7 @@ static int vmdk_open_desc_file(BlockDriverState *bs, int flags, char *buf,
     }
     s->create_type = g_strdup(ct);
     s->desc_offset = 0;
-    ret = vmdk_parse_extents(buf, bs, bs->file->bs->exact_filename, options,
-                             errp);
+    ret = vmdk_parse_extents(buf, bs, options, errp);
 exit:
     return ret;
 }
@@ -1720,6 +1734,16 @@ static int vmdk_write_extent(VmdkExtent *extent, int64_t cluster_offset,
     if (extent->compressed) {
         void *compressed_data;
 
+        /* Only whole clusters */
+        if (offset_in_cluster ||
+            n_bytes > (extent->cluster_sectors * SECTOR_SIZE) ||
+            (n_bytes < (extent->cluster_sectors * SECTOR_SIZE) &&
+             offset + n_bytes != extent->end_sector * SECTOR_SIZE))
+        {
+            ret = -EINVAL;
+            goto out;
+        }
+
         if (!extent->has_marker) {
             ret = -EINVAL;
             goto out;

qemu-io-cmds.c

@@ -350,6 +350,79 @@ static void qemu_io_free(void *p)
     qemu_vfree(p);
 }
 
+/*
+ * qemu_io_alloc_from_file()
+ *
+ * Allocates the buffer and populates it with the content of the given file
+ * up to @len bytes. If the file length is less than @len, then the buffer
+ * is populated with the file content cyclically.
+ *
+ * @blk - the block backend where the buffer content is going to be written to
+ * @len - the buffer length
+ * @file_name - the file to read the content from
+ *
+ * Returns: the buffer pointer on success
+ *          NULL on error
+ */
+static void *qemu_io_alloc_from_file(BlockBackend *blk, size_t len,
+                                     const char *file_name)
+{
+    char *buf, *buf_origin;
+    FILE *f = fopen(file_name, "r");
+    int pattern_len;
+
+    if (!f) {
+        perror(file_name);
+        return NULL;
+    }
+
+    if (qemuio_misalign) {
+        len += MISALIGN_OFFSET;
+    }
+
+    buf_origin = buf = blk_blockalign(blk, len);
+
+    if (qemuio_misalign) {
+        buf_origin += MISALIGN_OFFSET;
+        buf += MISALIGN_OFFSET;
+        len -= MISALIGN_OFFSET;
+    }
+
+    pattern_len = fread(buf_origin, 1, len, f);
+
+    if (ferror(f)) {
+        perror(file_name);
+        goto error;
+    }
+
+    if (pattern_len == 0) {
+        fprintf(stderr, "%s: file is empty\n", file_name);
+        goto error;
+    }
+
+    fclose(f);
+
+    if (len > pattern_len) {
+        len -= pattern_len;
+        buf += pattern_len;
+
+        while (len > 0) {
+            size_t len_to_copy = MIN(pattern_len, len);
+
+            memcpy(buf, buf_origin, len_to_copy);
+
+            len -= len_to_copy;
+            buf += len_to_copy;
+        }
+    }
+
+    return buf_origin;
+
+error:
+    qemu_io_free(buf_origin);
+    return NULL;
+}
+
 static void dump_buffer(const void *buffer, int64_t offset, int64_t len)
 {
     uint64_t i;
@@ -948,6 +1021,7 @@ static void write_help(void)
 " -n, -- with -z, don't allow slow fallback\n"
 " -p, -- ignored for backwards compatibility\n"
 " -P, -- use different pattern to fill file\n"
+" -s, -- use a pattern file to fill the write buffer\n"
 " -C, -- report statistics in a machine parsable format\n"
 " -q, -- quiet mode, do not show I/O statistics\n"
 " -u, -- with -z, allow unmapping\n"
@@ -964,7 +1038,7 @@ static const cmdinfo_t write_cmd = {
     .perm       = BLK_PERM_WRITE,
     .argmin     = 2,
     .argmax     = -1,
-    .args       = "[-bcCfnquz] [-P pattern] off len",
+    .args       = "[-bcCfnquz] [-P pattern | -s source_file] off len",
     .oneline    = "writes a number of bytes at a specified offset",
     .help       = write_help,
 };
@@ -973,7 +1047,7 @@ static int write_f(BlockBackend *blk, int argc, char **argv)
 {
     struct timespec t1, t2;
     bool Cflag = false, qflag = false, bflag = false;
-    bool Pflag = false, zflag = false, cflag = false;
+    bool Pflag = false, zflag = false, cflag = false, sflag = false;
     int flags = 0;
     int c, cnt, ret;
     char *buf = NULL;
@@ -982,8 +1056,9 @@ static int write_f(BlockBackend *blk, int argc, char **argv)
     /* Some compilers get confused and warn if this is not initialized.  */
     int64_t total = 0;
     int pattern = 0xcd;
+    const char *file_name = NULL;
 
-    while ((c = getopt(argc, argv, "bcCfnpP:quz")) != -1) {
+    while ((c = getopt(argc, argv, "bcCfnpP:qs:uz")) != -1) {
         switch (c) {
         case 'b':
             bflag = true;
@@ -1013,6 +1088,10 @@ static int write_f(BlockBackend *blk, int argc, char **argv)
         case 'q':
             qflag = true;
             break;
+        case 's':
+            sflag = true;
+            file_name = optarg;
+            break;
         case 'u':
             flags |= BDRV_REQ_MAY_UNMAP;
             break;
@@ -1050,8 +1129,9 @@ static int write_f(BlockBackend *blk, int argc, char **argv)
         return -EINVAL;
     }
 
-    if (zflag && Pflag) {
-        printf("-z and -P cannot be specified at the same time\n");
+    if (zflag + Pflag + sflag > 1) {
+        printf("Only one of -z, -P, and -s "
+               "can be specified at the same time\n");
         return -EINVAL;
     }
 
@@ -1087,7 +1167,14 @@ static int write_f(BlockBackend *blk, int argc, char **argv)
     }
 
     if (!zflag) {
-        buf = qemu_io_alloc(blk, count, pattern);
+        if (sflag) {
+            buf = qemu_io_alloc_from_file(blk, count, file_name);
+            if (!buf) {
+                return -EINVAL;
+            }
+        } else {
+            buf = qemu_io_alloc(blk, count, pattern);
+        }
     }
 
     clock_gettime(CLOCK_MONOTONIC, &t1);

tests/check-block.sh

@@ -21,6 +21,11 @@ if grep -q "TARGET_GPROF=y" *-softmmu/config-target.mak 2>/dev/null ; then
     exit 0
 fi
 
+if grep -q "CFLAGS.*-fsanitize" config-host.mak 2>/dev/null ; then
+    echo "Sanitizers are enabled ==> Not running the qemu-iotests."
+    exit 0
+fi
+
 if [ -z "$(find . -name 'qemu-system-*' -print)" ]; then
     echo "No qemu-system binary available ==> Not running the qemu-iotests."
     exit 0

tests/qemu-iotests/002

@@ -38,6 +38,7 @@ trap "_cleanup; exit \$status" 0 1 2 3 15
 
 _supported_fmt generic
 _supported_proto generic
+_unsupported_imgopts "subformat=streamOptimized"
 
 
 size=128M

tests/qemu-iotests/003

@@ -38,6 +38,7 @@ trap "_cleanup; exit \$status" 0 1 2 3 15
 
 _supported_fmt generic
 _supported_proto generic
+_unsupported_imgopts "subformat=streamOptimized"
 
 size=128M
 offset=67M

tests/qemu-iotests/005

@@ -43,7 +43,8 @@ _supported_fmt generic
 _supported_proto generic
 _supported_os Linux
 _unsupported_imgopts "subformat=twoGbMaxExtentFlat" \
-                     "subformat=twoGbMaxExtentSparse"
+                     "subformat=twoGbMaxExtentSparse" \
+                     "subformat=streamOptimized"
 
 # vpc is limited to 127GB, so we can't test it here
 if [ "$IMGFMT" = "vpc" ]; then

tests/qemu-iotests/009

@@ -38,6 +38,7 @@ trap "_cleanup; exit \$status" 0 1 2 3 15
 
 _supported_fmt generic
 _supported_proto generic
+_unsupported_imgopts "subformat=streamOptimized"
 
 
 size=6G

tests/qemu-iotests/010

@@ -38,6 +38,7 @@ trap "_cleanup; exit \$status" 0 1 2 3 15
 
 _supported_fmt generic
 _supported_proto generic
+_unsupported_imgopts "subformat=streamOptimized"
 
 
 size=6G

tests/qemu-iotests/011

@@ -38,6 +38,7 @@ trap "_cleanup; exit \$status" 0 1 2 3 15
 
 _supported_fmt generic
 _supported_proto generic
+_unsupported_imgopts "subformat=streamOptimized"
 
 
 size=6G

tests/qemu-iotests/017

@@ -41,7 +41,8 @@ trap "_cleanup; exit \$status" 0 1 2 3 15
 _supported_fmt qcow qcow2 vmdk qed
 _supported_proto generic
 _unsupported_proto vxhs
-_unsupported_imgopts "subformat=monolithicFlat" "subformat=twoGbMaxExtentFlat"
+_unsupported_imgopts "subformat=monolithicFlat" "subformat=twoGbMaxExtentFlat" \
+                     "subformat=streamOptimized"
 
 TEST_OFFSETS="0 4294967296"
 

tests/qemu-iotests/018

@@ -41,7 +41,8 @@ trap "_cleanup; exit \$status" 0 1 2 3 15
 _supported_fmt qcow qcow2 vmdk qed
 _supported_proto file
 _supported_os Linux
-_unsupported_imgopts "subformat=monolithicFlat" "subformat=twoGbMaxExtentFlat"
+_unsupported_imgopts "subformat=monolithicFlat" "subformat=twoGbMaxExtentFlat" \
+                     "streamOptimized"
 
 TEST_OFFSETS="0 4294967296"
 

tests/qemu-iotests/019

@@ -47,7 +47,8 @@ _supported_proto file
 _supported_os Linux
 _unsupported_imgopts "subformat=monolithicFlat" \
                      "subformat=twoGbMaxExtentFlat" \
-                     "subformat=twoGbMaxExtentSparse"
+                     "subformat=twoGbMaxExtentSparse" \
+                     "subformat=streamOptimized"
 
 TEST_OFFSETS="0 4294967296"
 CLUSTER_SIZE=65536

tests/qemu-iotests/020

@@ -44,7 +44,8 @@ _supported_fmt qcow qcow2 vmdk qed
 _supported_proto file
 _unsupported_imgopts "subformat=monolithicFlat" \
                      "subformat=twoGbMaxExtentFlat" \
-                     "subformat=twoGbMaxExtentSparse"
+                     "subformat=twoGbMaxExtentSparse" \
+                     "subformat=streamOptimized"
 
 TEST_OFFSETS="0 4294967296"
 

tests/qemu-iotests/026

@@ -41,8 +41,8 @@ trap "_cleanup; exit \$status" 0 1 2 3 15
 # Currently only qcow2 supports rebasing
 _supported_fmt qcow2
 _supported_proto file
-_default_cache_mode "writethrough"
-_supported_cache_modes "writethrough" "none"
+_default_cache_mode writethrough
+_supported_cache_modes writethrough none
 # The refcount table tests expect a certain minimum width for refcount entries
 # (so that the refcount table actually needs to grow); that minimum is 16 bits,
 # being the default refcount entry width.

tests/qemu-iotests/027

@@ -38,6 +38,7 @@ trap "_cleanup; exit \$status" 0 1 2 3 15
 
 _supported_fmt vmdk qcow qcow2 qed
 _supported_proto generic
+_unsupported_imgopts "subformat=streamOptimized"
 
 
 size=128M

tests/qemu-iotests/032

@@ -42,6 +42,7 @@ trap "_cleanup; exit \$status" 0 1 2 3 15
 # This works for any image format (though unlikely to segfault for raw)
 _supported_fmt generic
 _supported_proto generic
+_unsupported_imgopts "subformat=streamOptimized"
 
 echo
 echo === Prepare image ===

tests/qemu-iotests/033

@@ -38,6 +38,7 @@ trap "_cleanup; exit \$status" 0 1 2 3 15
 
 _supported_fmt generic
 _supported_proto generic
+_unsupported_imgopts "subformat=streamOptimized"
 
 
 size=128M

tests/qemu-iotests/034

@@ -41,7 +41,8 @@ _supported_proto file
 _supported_os Linux
 _unsupported_imgopts "subformat=monolithicFlat" \
                      "subformat=twoGbMaxExtentFlat" \
-                     "subformat=twoGbMaxExtentSparse"
+                     "subformat=twoGbMaxExtentSparse" \
+                     "subformat=streamOptimized"
 
 CLUSTER_SIZE=4k
 size=128M