1 files changed, 79 insertions, 43 deletions

diff --git a/block/file-posix.c b/block/file-posix.c
index 8067e238cb..8cc39a1ef6 100644
--- a/block/file-posix.c
+++ b/block/file-posix.c
@@ -62,10 +62,12 @@
 #include <sys/ioctl.h>
 #include <sys/param.h>
 #include <sys/syscall.h>
+#include <sys/vfs.h>
 #include <linux/cdrom.h>
 #include <linux/fd.h>
 #include <linux/fs.h>
 #include <linux/hdreg.h>
+#include <linux/magic.h>
 #include <scsi/sg.h>
 #ifdef __s390__
 #include <asm/dasd.h>
@@ -300,6 +302,28 @@ static int probe_physical_blocksize(int fd, unsigned int *blk_size)
 #endif
 }
 
+/*
+ * Returns true if no alignment restrictions are necessary even for files
+ * opened with O_DIRECT.
+ *
+ * raw_probe_alignment() probes the required alignment and assume that 1 means
+ * the probing failed, so it falls back to a safe default of 4k. This can be
+ * avoided if we know that byte alignment is okay for the file.
+ */
+static bool dio_byte_aligned(int fd)
+{
+#ifdef __linux__
+    struct statfs buf;
+    int ret;
+
+    ret = fstatfs(fd, &buf);
+    if (ret == 0 && buf.f_type == NFS_SUPER_MAGIC) {
+        return true;
+    }
+#endif
+    return false;
+}
+
 /* Check if read is allowed with given memory buffer and length.
  *
  * This function is used to check O_DIRECT memory buffer and request alignment.
@@ -401,6 +425,39 @@ static void raw_probe_alignment(BlockDriverState *bs, int fd, Error **errp)
     }
 }
 
+static int check_hdev_writable(int fd)
+{
+#if defined(BLKROGET)
+    /* Linux block devices can be configured "read-only" using blockdev(8).
+     * This is independent of device node permissions and therefore open(2)
+     * with O_RDWR succeeds.  Actual writes fail with EPERM.
+     *
+     * bdrv_open() is supposed to fail if the disk is read-only.  Explicitly
+     * check for read-only block devices so that Linux block devices behave
+     * properly.
+     */
+    struct stat st;
+    int readonly = 0;
+
+    if (fstat(fd, &st)) {
+        return -errno;
+    }
+
+    if (!S_ISBLK(st.st_mode)) {
+        return 0;
+    }
+
+    if (ioctl(fd, BLKROGET, &readonly) < 0) {
+        return -errno;
+    }
+
+    if (readonly) {
+        return -EACCES;
+    }
+#endif /* defined(BLKROGET) */
+    return 0;
+}
+
 static void raw_parse_flags(int bdrv_flags, int *open_flags, bool has_writers)
 {
     bool read_write = false;
@@ -585,6 +642,15 @@ static int raw_open_common(BlockDriverState *bs, QDict *options,
     }
     s->fd = fd;
 
+    /* Check s->open_flags rather than bdrv_flags due to auto-read-only */
+    if (s->open_flags & O_RDWR) {
+        ret = check_hdev_writable(s->fd);
+        if (ret < 0) {
+            error_setg_errno(errp, -ret, "The device is not writable");
+            goto fail;
+        }
+    }
+
     s->perm = 0;
     s->shared_perm = BLK_PERM_ALL;
 
@@ -629,7 +695,7 @@ static int raw_open_common(BlockDriverState *bs, QDict *options,
 
     s->has_discard = true;
     s->has_write_zeroes = true;
-    if ((bs->open_flags & BDRV_O_NOCACHE) != 0) {
+    if ((bs->open_flags & BDRV_O_NOCACHE) != 0 && !dio_byte_aligned(s->fd)) {
         s->needs_alignment = true;
     }
 
@@ -707,6 +773,9 @@ static int raw_open_common(BlockDriverState *bs, QDict *options,
     }
     ret = 0;
 fail:
+    if (ret < 0 && s->fd != -1) {
+        qemu_close(s->fd);
+    }
     if (filename && (bdrv_flags & BDRV_O_TEMPORARY)) {
         unlink(filename);
     }
@@ -977,6 +1046,15 @@ static int raw_reconfigure_getfd(BlockDriverState *bs, int flags,
         }
     }
 
+    if (fd != -1 && (*open_flags & O_RDWR)) {
+        ret = check_hdev_writable(fd);
+        if (ret < 0) {
+            qemu_close(fd);
+            error_setg_errno(errp, -ret, "The device is not writable");
+            return -1;
+        }
+    }
+
     return fd;
 }
 
@@ -3299,39 +3377,6 @@ static int hdev_probe_device(const char *filename)
     return 0;
 }
 
-static int check_hdev_writable(BDRVRawState *s)
-{
-#if defined(BLKROGET)
-    /* Linux block devices can be configured "read-only" using blockdev(8).
-     * This is independent of device node permissions and therefore open(2)
-     * with O_RDWR succeeds.  Actual writes fail with EPERM.
-     *
-     * bdrv_open() is supposed to fail if the disk is read-only.  Explicitly
-     * check for read-only block devices so that Linux block devices behave
-     * properly.
-     */
-    struct stat st;
-    int readonly = 0;
-
-    if (fstat(s->fd, &st)) {
-        return -errno;
-    }
-
-    if (!S_ISBLK(st.st_mode)) {
-        return 0;
-    }
-
-    if (ioctl(s->fd, BLKROGET, &readonly) < 0) {
-        return -errno;
-    }
-
-    if (readonly) {
-        return -EACCES;
-    }
-#endif /* defined(BLKROGET) */
-    return 0;
-}
-
 static void hdev_parse_filename(const char *filename, QDict *options,
                                 Error **errp)
 {
@@ -3454,15 +3499,6 @@ hdev_open_Mac_error:
     /* Since this does ioctl the device must be already opened */
     bs->sg = hdev_is_sg(bs);
 
-    if (flags & BDRV_O_RDWR) {
-        ret = check_hdev_writable(s);
-        if (ret < 0) {
-            raw_close(bs);
-            error_setg_errno(errp, -ret, "The device is not writable");
-            return ret;
-        }
-    }
-
     return ret;
 }