On 02/02/15 23:40, Peter Lieven wrote:
Am 02.02.2015 um 21:09 schrieb Denis V. Lunev:qemu_gluster_co_discard calculates size to discard as follows size_t size = nb_sectors * BDRV_SECTOR_SIZE; ret = glfs_discard_async(s->fd, offset, size, &gluster_finish_aiocb, acb);glfs_discard_async is declared as follows: int glfs_discard_async (glfs_fd_t *fd, off_t length, size_t lent, glfs_io_cbk fn, void *data) __THROW This is problematic on i686 as sizeof(size_t) == 4. Set bl_max_discard to SIZE_MAX >> BDRV_SECTOR_BITS to avoid overflow on i386. Signed-off-by: Denis V. Lunev <d...@openvz.org> CC: Kevin Wolf <kw...@redhat.com> CC: Peter Lieven <p...@kamp.de> --- block/gluster.c | 9 +++++++++ 1 file changed, 9 insertions(+) diff --git a/block/gluster.c b/block/gluster.c index 1eb3a8c..8a8c153 100644 --- a/block/gluster.c +++ b/block/gluster.c @@ -622,6 +622,11 @@ out: return ret; }+static void qemu_gluster_refresh_limits(BlockDriverState *bs, Error **errp)+{ + bs->bl.max_discard = MIN(SIZE_MAX >> BDRV_SECTOR_BITS, INT_MAX); +} +Looking at the gluster code bl.max_transfer_length should have the same limit, but thats a different patch.
ha, the same applies to nbd code too. I'll do this stuff tomorrow and also I think that some audit in other drivers could reveal something interesting. Den
