Date: Wed, 22 Jan 2020 23:07:04 +0000 From: bugzilla-noreply@freebsd.org To: virtualization@FreeBSD.org Subject: [Bug 215737] [bhyve] utilizing virtio-net truncates jumbo frames at 4084 bytes length Message-ID: <bug-215737-27103-uqz69FDGda@https.bugs.freebsd.org/bugzilla/> In-Reply-To: <bug-215737-27103@https.bugs.freebsd.org/bugzilla/> References: <bug-215737-27103@https.bugs.freebsd.org/bugzilla/>
next in thread | previous in thread | raw e-mail | index | archive | help
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=215737 Vincenzo Maffione <vmaffione@FreeBSD.org> changed: What |Removed |Added ---------------------------------------------------------------------------- CC| |vmaffione@FreeBSD.org --- Comment #19 from Vincenzo Maffione <vmaffione@FreeBSD.org> --- If I'm not mistaken if_vtnet with jumbo frames should work even if the host does not support rx mergeable buffers... if that doesn't work I would be inclined to think that this particular situation is not supported by the driver (IOW 64KB packets are not handled). Can somebody test it again with the stable/11 code? And yes, currently rx mergeable buffers are advertised by the host only when the vale net backend is used (for both if_vtnet and e1000). However, now that I think about that, rx mergeable buffers is not something that actually depends on the net backend, so if we advertise them in any case (e.g., also with the tap backend), things should magically work. I'll test this theory in the next days. -- You are receiving this mail because: You are the assignee for the bug.
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?bug-215737-27103-uqz69FDGda>