On 05/01/15 14:29, Peter Lieven wrote:
If bs->bl.max_write_zeroes is large and we end up in the unsupported path we might allocate a lot of memory for the iovector and/or even generate an oversized requests.Fix this by limiting the request by the minimum of the reported maximum transfer size or 16MB (32768 sectors). Reported-by: Denis V. Lunev <d...@openvz.org> Signed-off-by: Peter Lieven <p...@kamp.de> --- block.c | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) diff --git a/block.c b/block.c index a612594..8009478 100644 --- a/block.c +++ b/block.c @@ -3203,6 +3203,9 @@ static int coroutine_fn bdrv_co_do_write_zeroes(BlockDriverState *bs, if (ret == -ENOTSUP) { /* Fall back to bounce buffer if write zeroes is unsupported */ + int max_xfer_len = MIN_NON_ZERO(bs->bl.max_transfer_length, + MAX_WRITE_ZEROES_DEFAULT); + num = MIN(num, max_xfer_len); iov.iov_len = num * BDRV_SECTOR_SIZE; if (iov.iov_base == NULL) { iov.iov_base = qemu_try_blockalign(bs, num * BDRV_SECTOR_SIZE); @@ -3219,7 +3222,7 @@ static int coroutine_fn bdrv_co_do_write_zeroes(BlockDriverState *bs, /* Keep bounce buffer around if it is big enough for all * all future requests. */ - if (num < max_write_zeroes) { + if (num < max_xfer_len) { qemu_vfree(iov.iov_base); iov.iov_base = NULL; }
Reviewed-by: Denis V. Lunev <d...@openvz.org> Though pls consider my patch v3, it avoids allocation of 16 Mb here and uses only 1 Mb of memory.
