[13/31] file-posix: Switch to .bdrv_co_block_status()

Commit Message

Eric Blake April 18, 2017, 1:33 a.m. UTC

We are gradually moving away from sector-based interfaces, towards
byte-based.  Update the file protocol driver accordingly.

Signed-off-by: Eric Blake <[e-mail]>
---
 block/file-posix.c | 47 +++++++++++++++++++++++------------------------
 1 file changed, 23 insertions(+), 24 deletions(-)

Comments

Eric Blake April 18, 2017, 8:56 p.m. UTC | #1

On 04/17/2017 08:33 PM, Eric Blake wrote:
> We are gradually moving away from sector-based interfaces, towards
> byte-based.  Update the file protocol driver accordingly.
> 
> Signed-off-by: Eric Blake <[e-mail]>
> ---
>  block/file-posix.c | 47 +++++++++++++++++++++++------------------------
>  1 file changed, 23 insertions(+), 24 deletions(-)

This patch makes qemu-iotests 109 fail, due to an assertion in 6/31 that
the nested call to bdrv_block_status() [thanks to BDRV_BLOCK_RAW] is
still sector-aligned.  The culprit:

> +static int64_t coroutine_fn raw_co_block_status(BlockDriverState *bs,
> +                                                int64_t offset,
> +                                                int64_t bytes, int64_t *pnum,
> +                                                BlockDriverState **file)
>  {
> -    off_t start, data = 0, hole = 0;
> +    off_t data = 0, hole = 0;
>      int64_t total_size;
>      int ret;
> 
> @@ -1856,39 +1856,38 @@ static int64_t coroutine_fn raw_co_get_block_status(BlockDriverState *bs,
>          return ret;
>      }
> 
> -    start = sector_num * BDRV_SECTOR_SIZE;
>      total_size = bdrv_getlength(bs);

total_size is always sector-aligned (bdrv_getlength() purposefully
widens the input file size, 560 bytes in the case of an empty qcow image
in test 109, out to a sector boundary)...

>      if (total_size < 0) {
>          return total_size;
> -    } else if (start >= total_size) {
> +    } else if (offset >= total_size) {
>          *pnum = 0;
>          return 0;
> -    } else if (start + nb_sectors * BDRV_SECTOR_SIZE > total_size) {
> -        nb_sectors = DIV_ROUND_UP(total_size - start, BDRV_SECTOR_SIZE);
> +    } else if (offset + bytes > total_size) {
> +        bytes = total_size - offset;
>      }
> 
> -    ret = find_allocation(bs, start, &data, &hole);
> +    ret = find_allocation(bs, offset, &data, &hole);

...while find_allocation still obeys byte-granularity limits of the
underlying file system, and therefore reports a hole starting at offset
560 (the only time a hole starts at a non-sector boundary); pre-patch
that didn't matter (because we rounded before returning sectors through
*pnum), but post-patch, returning 560 instead of 1024 triggers problems.

I'm still debating whether the best fix is to tweak the file-posix
limits to just round up to sector boundaries at EOF, or to make the
block layer itself special-case EOF in case other protocols also have
byte-granularity support.  Any advice you have while reviewing is
appreciated, but it does mean I'll need a v2 of the series one way or
another.

And I didn't spot it during my earlier testing because I was skipping
109 as failing prior to my patches even started testing it - but now
that Jeff and Fam have both posted patches for two separate problems in
test 109 (the need to munge "len", and failure to do proper cleanup), I
no longer have that excuse.

Patch

diff --git a/block/file-posix.c b/block/file-posix.c
index 1941fb6..690bd45 100644
--- a/block/file-posix.c
+++ b/block/file-posix.c
@@ -1832,22 +1832,22 @@  static int find_allocation(BlockDriverState *bs, off_t start,
 /*
  * Returns the allocation status of the specified sectors.
  *
- * If 'sector_num' is beyond the end of the disk image the return value is 0
+ * If 'offset' is beyond the end of the disk image the return value is 0
  * and 'pnum' is set to 0.
  *
- * 'pnum' is set to the number of sectors (including and immediately following
- * the specified sector) that are known to be in the same
+ * 'pnum' is set to the number of bytes (including and immediately following
+ * the specified offset) that are known to be in the same
  * allocated/unallocated state.
  *
- * 'nb_sectors' is the max value 'pnum' should be set to.  If nb_sectors goes
+ * 'bytes' is the max value 'pnum' should be set to.  If bytes goes
  * beyond the end of the disk image it will be clamped.
  */
-static int64_t coroutine_fn raw_co_get_block_status(BlockDriverState *bs,
-                                                    int64_t sector_num,
-                                                    int nb_sectors, int *pnum,
-                                                    BlockDriverState **file)
+static int64_t coroutine_fn raw_co_block_status(BlockDriverState *bs,
+                                                int64_t offset,
+                                                int64_t bytes, int64_t *pnum,
+                                                BlockDriverState **file)
 {
-    off_t start, data = 0, hole = 0;
+    off_t data = 0, hole = 0;
     int64_t total_size;
     int ret;

@@ -1856,39 +1856,38 @@  static int64_t coroutine_fn raw_co_get_block_status(BlockDriverState *bs,
         return ret;
     }

-    start = sector_num * BDRV_SECTOR_SIZE;
     total_size = bdrv_getlength(bs);
     if (total_size < 0) {
         return total_size;
-    } else if (start >= total_size) {
+    } else if (offset >= total_size) {
         *pnum = 0;
         return 0;
-    } else if (start + nb_sectors * BDRV_SECTOR_SIZE > total_size) {
-        nb_sectors = DIV_ROUND_UP(total_size - start, BDRV_SECTOR_SIZE);
+    } else if (offset + bytes > total_size) {
+        bytes = total_size - offset;
     }

-    ret = find_allocation(bs, start, &data, &hole);
+    ret = find_allocation(bs, offset, &data, &hole);
     if (ret == -ENXIO) {
         /* Trailing hole */
-        *pnum = nb_sectors;
+        *pnum = bytes;
         ret = BDRV_BLOCK_ZERO;
     } else if (ret < 0) {
         /* No info available, so pretend there are no holes */
-        *pnum = nb_sectors;
+        *pnum = bytes;
         ret = BDRV_BLOCK_DATA;
-    } else if (data == start) {
-        /* On a data extent, compute sectors to the end of the extent,
+    } else if (data == offset) {
+        /* On a data extent, compute bytes to the end of the extent,
          * possibly including a partial sector at EOF. */
-        *pnum = MIN(nb_sectors, DIV_ROUND_UP(hole - start, BDRV_SECTOR_SIZE));
+        *pnum = MIN(bytes, hole - offset);
         ret = BDRV_BLOCK_DATA;
     } else {
-        /* On a hole, compute sectors to the beginning of the next extent.  */
-        assert(hole == start);
-        *pnum = MIN(nb_sectors, (data - start) / BDRV_SECTOR_SIZE);
+        /* On a hole, compute bytes to the beginning of the next extent.  */
+        assert(hole == offset);
+        *pnum = MIN(bytes, data - offset);
         ret = BDRV_BLOCK_ZERO;
     }
     *file = bs;
-    return ret | BDRV_BLOCK_OFFSET_VALID | start;
+    return ret | BDRV_BLOCK_OFFSET_VALID | (offset & BDRV_BLOCK_OFFSET_MASK);
 }

 static coroutine_fn BlockAIOCB *raw_aio_pdiscard(BlockDriverState *bs,
@@ -1963,7 +1962,7 @@  BlockDriver bdrv_file = {
     .bdrv_close = raw_close,
     .bdrv_create = raw_create,
     .bdrv_has_zero_init = bdrv_has_zero_init_1,
-    .bdrv_co_get_block_status = raw_co_get_block_status,
+    .bdrv_co_block_status = raw_co_block_status,
     .bdrv_co_pwrite_zeroes = raw_co_pwrite_zeroes,

     .bdrv_co_preadv         = raw_co_preadv,