When there is a format driver between the backend, it's not guaranteed that exposing the opt_transfer_length for the format driver results in the optimal requests (because of fragmentation etc.), but it can't make things worse, so let's just do it.
Signed-off-by: Kevin Wolf <kw...@redhat.com> Reviewed-by: Wenchao Xia <xiaw...@linux.vnet.ibm.com> --- block.c | 20 +++++++++++++++++++- 1 file changed, 19 insertions(+), 1 deletion(-) diff --git a/block.c b/block.c index 3bbcad4..79a325f 100644 --- a/block.c +++ b/block.c @@ -485,7 +485,25 @@ static int bdrv_refresh_limits(BlockDriverState *bs) memset(&bs->bl, 0, sizeof(bs->bl)); - if (drv && drv->bdrv_refresh_limits) { + if (!drv) { + return 0; + } + + /* Take some limits from the children as a default */ + if (bs->file) { + bdrv_refresh_limits(bs->file); + bs->bl.opt_transfer_length = bs->file->bl.opt_transfer_length; + } + + if (bs->backing_hd) { + bdrv_refresh_limits(bs->backing_hd); + bs->bl.opt_transfer_length = + MAX(bs->bl.opt_transfer_length, + bs->backing_hd->bl.opt_transfer_length); + } + + /* Then let the driver override it */ + if (drv->bdrv_refresh_limits) { return drv->bdrv_refresh_limits(bs); } -- 1.8.1.4