diff mbox series

[for-6.2?,2/2] nbd/server: Simplify zero and trim

Message ID 20211117170230.1128262-3-eblake@redhat.com (mailing list archive)
State New, archived
Headers show
Series NBD 6.2-rc fixes | expand

Commit Message

Eric Blake Nov. 17, 2021, 5:02 p.m. UTC
Now that the block layer supports 64-bit operations, we no longer have
to self-fragment requests larger than 2G, reverting the workaround
added in 890cbccb08 (nbd: Fix large trim/zero requests).

Signed-off-by: Eric Blake <eblake@redhat.com>
---
 nbd/server.c | 23 +++--------------------
 1 file changed, 3 insertions(+), 20 deletions(-)

Comments

Vladimir Sementsov-Ogievskiy Nov. 17, 2021, 6:04 p.m. UTC | #1
17.11.2021 20:02, Eric Blake wrote:
> Now that the block layer supports 64-bit operations, we no longer have
> to self-fragment requests larger than 2G, reverting the workaround
> added in 890cbccb08 (nbd: Fix large trim/zero requests).
> 
> Signed-off-by: Eric Blake<eblake@redhat.com>

Reviewed-by: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com>
Eric Blake Nov. 17, 2021, 8:49 p.m. UTC | #2
On Wed, Nov 17, 2021 at 09:04:34PM +0300, Vladimir Sementsov-Ogievskiy wrote:
> 17.11.2021 20:02, Eric Blake wrote:
> > Now that the block layer supports 64-bit operations, we no longer have
> > to self-fragment requests larger than 2G, reverting the workaround
> > added in 890cbccb08 (nbd: Fix large trim/zero requests).
> > 
> > Signed-off-by: Eric Blake<eblake@redhat.com>
> 
> Reviewed-by: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com>

Any preferences on whether this should be in 6.2, or deferred to 7.0?
Vladimir Sementsov-Ogievskiy Nov. 17, 2021, 8:52 p.m. UTC | #3
17.11.2021 23:49, Eric Blake wrote:
> On Wed, Nov 17, 2021 at 09:04:34PM +0300, Vladimir Sementsov-Ogievskiy wrote:
>> 17.11.2021 20:02, Eric Blake wrote:
>>> Now that the block layer supports 64-bit operations, we no longer have
>>> to self-fragment requests larger than 2G, reverting the workaround
>>> added in 890cbccb08 (nbd: Fix large trim/zero requests).
>>>
>>> Signed-off-by: Eric Blake<eblake@redhat.com>
>>
>> Reviewed-by: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com>
> 
> Any preferences on whether this should be in 6.2, or deferred to 7.0?
> 

In my opinion it's good for 6.2, if we are not very strict on "only real bug fixes in hard freeze".

The commit is obviously safe: if it break something, it means that feature we already included into 6.2 is broken anyway :)
diff mbox series

Patch

diff --git a/nbd/server.c b/nbd/server.c
index 85877f630533..1b3945220bd3 100644
--- a/nbd/server.c
+++ b/nbd/server.c
@@ -2509,16 +2509,8 @@  static coroutine_fn int nbd_handle_request(NBDClient *client,
         if (request->flags & NBD_CMD_FLAG_FAST_ZERO) {
             flags |= BDRV_REQ_NO_FALLBACK;
         }
-        ret = 0;
-        /* FIXME simplify this when blk_pwrite_zeroes switches to 64-bit */
-        while (ret >= 0 && request->len) {
-            int align = client->check_align ?: 1;
-            int len = MIN(request->len, QEMU_ALIGN_DOWN(BDRV_REQUEST_MAX_BYTES,
-                                                        align));
-            ret = blk_pwrite_zeroes(exp->common.blk, request->from, len, flags);
-            request->len -= len;
-            request->from += len;
-        }
+        ret = blk_pwrite_zeroes(exp->common.blk, request->from, request->len,
+                                flags);
         return nbd_send_generic_reply(client, request->handle, ret,
                                       "writing to file failed", errp);

@@ -2532,16 +2524,7 @@  static coroutine_fn int nbd_handle_request(NBDClient *client,
                                       "flush failed", errp);

     case NBD_CMD_TRIM:
-        ret = 0;
-        /* FIXME simplify this when blk_co_pdiscard switches to 64-bit */
-        while (ret >= 0 && request->len) {
-            int align = client->check_align ?: 1;
-            int len = MIN(request->len, QEMU_ALIGN_DOWN(BDRV_REQUEST_MAX_BYTES,
-                                                        align));
-            ret = blk_co_pdiscard(exp->common.blk, request->from, len);
-            request->len -= len;
-            request->from += len;
-        }
+        ret = blk_co_pdiscard(exp->common.blk, request->from, request->len);
         if (ret >= 0 && request->flags & NBD_CMD_FLAG_FUA) {
             ret = blk_co_flush(exp->common.blk);
         }