All of lore.kernel.org
 help / color / mirror / Atom feed
From: Eric Blake <eblake@redhat.com>
To: qemu-devel@nongnu.org
Cc: kwolf@redhat.com, jsnow@redhat.com, qemu-block@nongnu.org,
	el13635@mail.ntua.gr, Max Reitz <mreitz@redhat.com>,
	Stefan Hajnoczi <stefanha@redhat.com>,
	Fam Zheng <famz@redhat.com>
Subject: [Qemu-devel] [PATCH v2 14/15] block: Align block status requests
Date: Mon,  3 Jul 2017 17:14:55 -0500	[thread overview]
Message-ID: <20170703221456.30817-15-eblake@redhat.com> (raw)
In-Reply-To: <20170703221456.30817-1-eblake@redhat.com>

Any device that has request_alignment greater than 512 should be
unable to report status at a finer granularity; it may also be
simpler for such devices to be guaranteed that the block layer
has rounded things out to the granularity boundary (the way the
block layer already rounds all other I/O out).  Besides, getting
the code correct for super-sector alignment also benefits us
for the fact that our public interface now has byte granularity,
even though none of our drivers have byte-level callbacks.

Add an assertion in blkdebug that proves that the block layer
never requests status of unaligned sections, similar to what it
does on other requests (while still keeping the generic helper
in place for when future patches add a throttle driver).  Note
note that iotest 177 already covers this (it would fail if you
use just the blkdebug.c hunk without the io.c changes).
Meanwhile, we can drop assertions in callers that no longer have
to pass in sector-aligned addresses.

Signed-off-by: Eric Blake <eblake@redhat.com>

---
v2: new patch
---
 include/block/block_int.h |  3 ++-
 block/blkdebug.c          | 13 +++++++++++-
 block/io.c                | 53 ++++++++++++++++++++++++++++++++---------------
 3 files changed, 50 insertions(+), 19 deletions(-)

diff --git a/include/block/block_int.h b/include/block/block_int.h
index ffa22c7..5f6ba5d 100644
--- a/include/block/block_int.h
+++ b/include/block/block_int.h
@@ -173,7 +173,8 @@ struct BlockDriver {
      * according to the current layer, and should not set
      * BDRV_BLOCK_ALLOCATED, but may set BDRV_BLOCK_RAW.  See block.h
      * for the meaning of _DATA, _ZERO, and _OFFSET_VALID.  The block
-     * layer guarantees non-NULL pnum and file.
+     * layer guarantees input aligned to request_alignment, as well as
+     * non-NULL pnum and file.
      */
     int64_t coroutine_fn (*bdrv_co_get_block_status)(BlockDriverState *bs,
         int64_t sector_num, int nb_sectors, int *pnum,
diff --git a/block/blkdebug.c b/block/blkdebug.c
index f1539db..67736b4 100644
--- a/block/blkdebug.c
+++ b/block/blkdebug.c
@@ -641,6 +641,17 @@ static int coroutine_fn blkdebug_co_pdiscard(BlockDriverState *bs,
     return bdrv_co_pdiscard(bs->file->bs, offset, bytes);
 }

+static int64_t coroutine_fn blkdebug_co_get_block_status(
+    BlockDriverState *bs, int64_t sector_num, int nb_sectors, int *pnum,
+    BlockDriverState **file)
+{
+    assert(QEMU_IS_ALIGNED(sector_num | nb_sectors,
+                           DIV_ROUND_UP(bs->bl.request_alignment,
+                                        BDRV_SECTOR_SIZE)));
+    return bdrv_co_get_block_status_from_file(bs, sector_num, nb_sectors,
+                                              pnum, file);
+}
+
 static void blkdebug_close(BlockDriverState *bs)
 {
     BDRVBlkdebugState *s = bs->opaque;
@@ -915,7 +926,7 @@ static BlockDriver bdrv_blkdebug = {
     .bdrv_co_flush_to_disk  = blkdebug_co_flush,
     .bdrv_co_pwrite_zeroes  = blkdebug_co_pwrite_zeroes,
     .bdrv_co_pdiscard       = blkdebug_co_pdiscard,
-    .bdrv_co_get_block_status = bdrv_co_get_block_status_from_file,
+    .bdrv_co_get_block_status = blkdebug_co_get_block_status,

     .bdrv_debug_event           = blkdebug_debug_event,
     .bdrv_debug_breakpoint      = blkdebug_debug_breakpoint,
diff --git a/block/io.c b/block/io.c
index 91d3e99..5ed1ac7 100644
--- a/block/io.c
+++ b/block/io.c
@@ -1746,7 +1746,8 @@ static int64_t coroutine_fn bdrv_co_block_status(BlockDriverState *bs,
     int64_t n; /* bytes */
     int64_t ret, ret2;
     BlockDriverState *local_file = NULL;
-    int count; /* sectors */
+    int64_t aligned_offset, aligned_bytes;
+    uint32_t align;

     assert(pnum);
     total_size = bdrv_getlength(bs);
@@ -1788,27 +1789,44 @@ static int64_t coroutine_fn bdrv_co_block_status(BlockDriverState *bs,
     }

     bdrv_inc_in_flight(bs);
-    /*
-     * TODO: Rather than require aligned offsets, we could instead
-     * round to the driver's request_alignment here, then touch up
-     * count afterwards back to the caller's expectations.
-     */
-    assert(QEMU_IS_ALIGNED(offset | bytes, BDRV_SECTOR_SIZE));
-    ret = bs->drv->bdrv_co_get_block_status(bs, offset >> BDRV_SECTOR_BITS,
-                                            bytes >> BDRV_SECTOR_BITS, &count,
-                                            &local_file);
-    if (ret < 0) {
-        *pnum = 0;
-        goto out;
+
+    /* Round out to request_alignment boundaries */
+    align = MAX(bs->bl.request_alignment, BDRV_SECTOR_SIZE);
+    aligned_offset = QEMU_ALIGN_DOWN(offset, align);
+    aligned_bytes = ROUND_UP(offset + bytes, align) - aligned_offset;
+
+    {
+        int count; /* sectors */
+
+        assert(QEMU_IS_ALIGNED(aligned_offset | aligned_bytes,
+                               BDRV_SECTOR_SIZE));
+        ret = bs->drv->bdrv_co_get_block_status(
+            bs, aligned_offset >> BDRV_SECTOR_BITS,
+            aligned_bytes >> BDRV_SECTOR_BITS, &count, &local_file);
+        if (ret < 0) {
+            *pnum = 0;
+            goto out;
+        }
+        *pnum = count * BDRV_SECTOR_SIZE;
+    }
+
+    /* Clamp pnum and ret to original request */
+    assert(QEMU_IS_ALIGNED(*pnum, align));
+    *pnum -= offset - aligned_offset;
+    if (aligned_offset >> BDRV_SECTOR_BITS != offset >> BDRV_SECTOR_BITS &&
+        ret & BDRV_BLOCK_OFFSET_VALID) {
+        ret += QEMU_ALIGN_DOWN(offset - aligned_offset, BDRV_SECTOR_SIZE);
+    }
+    if (*pnum > bytes) {
+        *pnum = bytes;
     }
-    *pnum = count * BDRV_SECTOR_SIZE;

     if (ret & BDRV_BLOCK_RAW) {
         assert(ret & BDRV_BLOCK_OFFSET_VALID && local_file);
         ret = bdrv_co_block_status(local_file, allocation,
-                                   ret & BDRV_BLOCK_OFFSET_MASK,
+                                   (ret & BDRV_BLOCK_OFFSET_MASK) |
+                                   (offset & ~BDRV_BLOCK_OFFSET_MASK),
                                    *pnum, pnum, &local_file);
-        assert(QEMU_IS_ALIGNED(*pnum, BDRV_SECTOR_SIZE));
         goto out;
     }

@@ -1832,7 +1850,8 @@ static int64_t coroutine_fn bdrv_co_block_status(BlockDriverState *bs,
         int64_t file_pnum;

         ret2 = bdrv_co_block_status(local_file, true,
-                                    ret & BDRV_BLOCK_OFFSET_MASK,
+                                    (ret & BDRV_BLOCK_OFFSET_MASK) |
+                                    (offset & ~BDRV_BLOCK_OFFSET_MASK),
                                     *pnum, &file_pnum, NULL);
         if (ret2 >= 0) {
             /* Ignore errors.  This is just providing extra information, it
-- 
2.9.4

  parent reply	other threads:[~2017-07-03 22:15 UTC|newest]

Thread overview: 41+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2017-07-03 22:14 [Qemu-devel] [PATCH v2 00/15] make bdrv_get_block_status byte-based Eric Blake
2017-07-03 22:14 ` [Qemu-devel] [PATCH v2 01/15] block: add default implementations for bdrv_co_get_block_status() Eric Blake
2017-07-04  7:33   ` Fam Zheng
2017-07-05 22:00   ` Eric Blake
2017-07-03 22:14 ` [Qemu-devel] [PATCH v2 02/15] block: Allow NULL file for bdrv_get_block_status() Eric Blake
2017-07-04  7:33   ` Fam Zheng
2017-07-03 22:14 ` [Qemu-devel] [PATCH v2 03/15] block: Add flag to avoid wasted work in bdrv_is_allocated() Eric Blake
2017-07-04  7:06   ` Fam Zheng
2017-07-05 11:56     ` Eric Blake
2017-07-05 12:07       ` Fam Zheng
2017-07-05 14:01         ` Eric Blake
2017-07-05 14:21           ` Fam Zheng
2017-07-06  2:35   ` Eric Blake
2017-07-03 22:14 ` [Qemu-devel] [PATCH v2 04/15] block: Make bdrv_round_to_clusters() signature more useful Eric Blake
2017-07-04  7:34   ` Fam Zheng
2017-07-03 22:14 ` [Qemu-devel] [PATCH v2 05/15] qcow2: Switch is_zero_sectors() to byte-based Eric Blake
2017-07-04  7:34   ` Fam Zheng
2017-07-03 22:14 ` [Qemu-devel] [PATCH v2 06/15] block: Switch bdrv_make_zero() " Eric Blake
2017-07-04  7:35   ` Fam Zheng
2017-07-03 22:14 ` [Qemu-devel] [PATCH v2 07/15] qemu-img: Switch get_block_status() " Eric Blake
2017-07-04  7:40   ` Fam Zheng
2017-07-03 22:14 ` [Qemu-devel] [PATCH v2 08/15] block: Convert bdrv_get_block_status() to bytes Eric Blake
2017-07-04  8:40   ` Fam Zheng
2017-07-03 22:14 ` [Qemu-devel] [PATCH v2 09/15] block: Switch bdrv_co_get_block_status() to byte-based Eric Blake
2017-07-04  8:59   ` Fam Zheng
2017-07-03 22:14 ` [Qemu-devel] [PATCH v2 10/15] block: Switch BdrvCoGetBlockStatusData " Eric Blake
2017-07-04  9:00   ` Fam Zheng
2017-07-03 22:14 ` [Qemu-devel] [PATCH v2 11/15] block: Switch bdrv_common_block_status_above() " Eric Blake
2017-07-04  9:02   ` Fam Zheng
2017-07-03 22:14 ` [Qemu-devel] [PATCH v2 12/15] block: Switch bdrv_co_get_block_status_above() " Eric Blake
2017-07-04  9:28   ` Fam Zheng
2017-07-03 22:14 ` [Qemu-devel] [PATCH v2 13/15] block: Convert bdrv_get_block_status_above() to bytes Eric Blake
2017-07-04  9:32   ` Fam Zheng
2017-07-03 22:14 ` Eric Blake [this message]
2017-07-04  9:44   ` [Qemu-devel] [PATCH v2 14/15] block: Align block status requests Fam Zheng
2017-07-05 12:01     ` Eric Blake
2017-07-05 12:12       ` Fam Zheng
2017-07-03 22:14 ` [Qemu-devel] [PATCH v2 15/15] qemu-io: Relax 'alloc' now that block-status doesn't assert Eric Blake
2017-07-04  9:50   ` Fam Zheng
2017-07-03 22:20 ` [Qemu-devel] [RFC PATCH v2 16/15] block: Add .bdrv_co_block_status() callback Eric Blake
2017-07-04 11:30   ` Fam Zheng

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20170703221456.30817-15-eblake@redhat.com \
    --to=eblake@redhat.com \
    --cc=el13635@mail.ntua.gr \
    --cc=famz@redhat.com \
    --cc=jsnow@redhat.com \
    --cc=kwolf@redhat.com \
    --cc=mreitz@redhat.com \
    --cc=qemu-block@nongnu.org \
    --cc=qemu-devel@nongnu.org \
    --cc=stefanha@redhat.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.