On Tue, 21 Jan 2025 14:20:28 +1030
David Gibson
On Mon, Jan 20, 2025 at 02:15:22PM +0100, Laurent Vivier wrote:
Replace ASSERT() on the number of iovec in the element and on the first entry length by a debug() message.
Signed-off-by: Laurent Vivier
Removing the ASSERT() makes sense, but is it safe to carry on to the tap_add_packet() if the packet is not in the layout we expect?
Safe for us and for TCP yes, but it's bad to mangle UDP packets like that.
Should we be bailing out of the function (effectively dropping the packet) instead?
I even proposed that in: https://archives.passt.top/passt-dev/20250115233302.23b24862@elisabeth/ and forgot about it during review. :( This is applied now. Laurent, a follow-up patch would be appreciated. -- Stefano