Date: Tue, 28 Jan 2020 10:15:11 +0000 From: bugzilla-noreply@freebsd.org To: net@FreeBSD.org Subject: [Bug 243392] if_vmx(4): Input buffer corruption Message-ID: <bug-243392-7501-IKlpskbcDh@https.bugs.freebsd.org/bugzilla/> In-Reply-To: <bug-243392-7501@https.bugs.freebsd.org/bugzilla/> References: <bug-243392-7501@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=3D243392 --- Comment #6 from Andriy Gapon <avg@FreeBSD.org> --- I probably should open a new bug for my side of the things as my problem ca= n be different from the original problem reported here. But for the moment here is a new bit of information. I tried to gather a little bit more of information with tcpdump and I got a very surprising result. If I run tcpdump on the vmx side, then the copy speed never degrades. If I start copying before starting tcpdump then the speed can degrade, but = as soon as I start tcpdump (on the vmx side) the speed immediately recovers. I have tested that many times with the same result. It seems that enabling bpf changes the packet processing flow and works aro= und whatever problem there is. Does anyone have an idea what that change could be and where to dig further? Thanks! --=20 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-243392-7501-IKlpskbcDh>