On 31.05.23 20:40, Eric Blake wrote:
On Wed, May 31, 2023 at 08:00:43PM +0300, Vladimir Sementsov-Ogievskiy wrote:
On 15.05.23 22:53, Eric Blake wrote:
Because we use NBD_CMD_FLAG_REQ_ONE with NBD_CMD_BLOCK_STATUS, a
client in narrow mode should not be able to provoke a server into
sending a bl
On Wed, May 31, 2023 at 08:00:43PM +0300, Vladimir Sementsov-Ogievskiy wrote:
> On 15.05.23 22:53, Eric Blake wrote:
> > Because we use NBD_CMD_FLAG_REQ_ONE with NBD_CMD_BLOCK_STATUS, a
> > client in narrow mode should not be able to provoke a server into
> > sending a block status result larger th
On 15.05.23 22:53, Eric Blake wrote:
Because we use NBD_CMD_FLAG_REQ_ONE with NBD_CMD_BLOCK_STATUS, a
client in narrow mode should not be able to provoke a server into
sending a block status result larger than the client's 32-bit request.
But in extended mode, a 64-bit status request must be able
Because we use NBD_CMD_FLAG_REQ_ONE with NBD_CMD_BLOCK_STATUS, a
client in narrow mode should not be able to provoke a server into
sending a block status result larger than the client's 32-bit request.
But in extended mode, a 64-bit status request must be able to handle a
64-bit status result, once
