Date: Thu, 11 Sep 2025 18:00:22 +0000 From: bugzilla-noreply@freebsd.org To: net@FreeBSD.org Subject: [Bug 289333] [Feature Request] HFSC overhead calculation Message-ID: <bug-289333-7501-STfHqwu7O6@https.bugs.freebsd.org/bugzilla/> In-Reply-To: <bug-289333-7501@https.bugs.freebsd.org/bugzilla/> References: <bug-289333-7501@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=3D289333 --- Comment #4 from Daniel Engel <freebsd@danielengel.com> --- (In reply to Oleksandr Kryvulia from comment #3) I don't think this configuration would address the problem at all.=20=20 I'll leave aside the potential 'burst' issue with bufferbloat for now, sinc= e I don't actually know how big of a buffer my modem has internally.=20=20 The main problem is that not all 700 kbps traffic flows are created equal.= =20 Consider the range of TCP/IP packet sizes:=20 * Large packet case (e.g. file upload): 58 pkt/s * 1492 bytes/pkt * 8 =3D 692288 kbps (dummynet output) 58 pkt/s * (1492+38) bytes/pkt * (53/48) * 8 =3D 783870 kbps (modem outpu= t) * Small packet case (e.g. ACK):=20 2187 pkt/s * 40 bytes/pkt =3D 699840 kbps (dummynet output) 2187 pkt/s * (40+38) bytes/pkt * (53/48) * 8 =3D 1506843 kbps (modem outp= ut) In the latter case, the modem would have to drop almost 50% of packets even though dummynet is shaping the link to the same throughput in both cases.=20 NOTE 1: I'm not 100% sure on the specific framing differences between ADSL = and ADSL2, or exactly what layers are present on my ADSL2 connection. I'm basi= ng my calculations on the summary linked below. It seems consistent with my real-world experience:=20=20 https://blog.ipspace.net/2009/03/adsl-overhead/ NOTE 2: Throughput for ACKs is even worse than the calculation above. ATM = will add padding if necessary to fill out the last cell. So, a 78 byte ACK(20 T= CP + 20 IP + 38 ADSL) actually wastes an additional 18 bytes to fill out 2 ATM frames before encoding (106 bytes on the wire).=20=20 Regarding "stateless rules", I use PFsense floating rules and haven't seen = any issues with queuing inbound or outbound traffic correctly. --=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-289333-7501-STfHqwu7O6>