From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mail-qk0-x22b.google.com (mail-qk0-x22b.google.com [IPv6:2607:f8b0:400d:c09::22b]) (using TLSv1 with cipher RC4-SHA (128/128 bits)) (Client CN "smtp.gmail.com", Issuer "Google Internet Authority G2" (verified OK)) by huchra.bufferbloat.net (Postfix) with ESMTPS id EEC8521F309; Sat, 23 May 2015 23:51:10 -0700 (PDT) Received: by qkx62 with SMTP id 62so43492821qkx.3; Sat, 23 May 2015 23:51:08 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113; h=mime-version:in-reply-to:references:date:message-id:subject:from:to :cc:content-type; bh=RWtTapEUIpDxrYIyJodVkSLREa3Cxoq1rGeurr/UMk4=; b=at0Ypaz1YrgHMOX4z7BjaAMuk9iGvc4zLDh8yx/tleXkdm6XmhveqrqblBk1zj/SiM BKFJL6TL8c6Z0GkHi8AuKXRVYGqOIoOklSM6Nxn/LYiK/CoTmWImglQkcHzBEa6l1xxi 2bsmuF/ouQkQCmpXKDGlqU+4Z8uuym9R3ltoVvJ7K7QecAsgeKRIhYcdVwFBD+Yy9Rlp aR74zLPb5WADxe7nS0z/6ACVuC0szJehk9ZFbC2qcAPr5t6Q0xdnNQh54jfhV2G0mH6s EAqvXjxr6RUQ0qoBd6IFonD7/R1LvnDm6rwf1s+QDTGSZ4P7mO/1eVBWOoYET8x5oI8f sM7w== MIME-Version: 1.0 X-Received: by 10.140.134.148 with SMTP id 142mr21835765qhg.100.1432450268405; Sat, 23 May 2015 23:51:08 -0700 (PDT) Received: by 10.96.187.71 with HTTP; Sat, 23 May 2015 23:51:08 -0700 (PDT) In-Reply-To: References: Date: Sat, 23 May 2015 23:51:08 -0700 Message-ID: From: Aaron Wood To: Dave Taht Content-Type: multipart/alternative; boundary=001a113534146826730516ce5198 Cc: cerowrt-devel , bloat Subject: Re: [Cerowrt-devel] [Bloat] sqm-scripts on WRT1900AC X-BeenThere: cerowrt-devel@lists.bufferbloat.net X-Mailman-Version: 2.1.13 Precedence: list List-Id: Development issues regarding the cerowrt test router project List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Sun, 24 May 2015 06:51:40 -0000 --001a113534146826730516ce5198 Content-Type: text/plain; charset=UTF-8 On Sat, May 23, 2015 at 11:44 PM, Dave Taht wrote: > > And it has a fan. Hate fans. Amusingly (I guess), I had this same > chipset to fiddle with in the "mirabox" and it ran waaaay too hot. > I haven't hit the fan, yet.... > It is not clear why you are getting an inaccurate rate out of it, either. > I feel like the rate has never been really accurate, or perhaps it's just that we're jut not properly accounting in our measurement tools (netperf), for the ACK bandwidth. But my current "limited" rate is set higher than anything I've ever measured, and it's clearly still working. Part of my next rounds of testing are to keep pushing that up until it's clear that I've tripped over the "real" rate. It is a leading candidate, but I would prefer to find a hardware > partner that cared about our issues enough to work with us, rather > than ignore us as netgear did. > I fear that's going to be a continual issue while we ask for things that most people don't understand that they want/need. But I think that we'll have an easier time if we're talking with a chipset vendor (like Marvell) vs an OEM (like Netgear), given the relative distance each has from the kernel drivers... -Aaron --001a113534146826730516ce5198 Content-Type: text/html; charset=UTF-8 Content-Transfer-Encoding: quoted-printable
On Sat, May 23, 2015 at 11:44 PM, Dave Taht <dave.taht@= gmail.com> wrote:

And it has a fan. Hate fans. Amusingly (I guess), I had this same
chipset to fiddle with in the "mirabox" and it ran waaaay too hot= .

I haven't hit the fan, yet....
=C2=A0
It is not clear why you = are getting an inaccurate rate out of it, either.

=
I feel like the rate has never been really accurate, or perhaps = it's just that we're jut not properly accounting in our measurement= tools (netperf), for the ACK bandwidth.=C2=A0 But my current "limited= " rate is set higher than anything I've ever measured, and it'= s clearly still working.=C2=A0 Part of my next rounds of testing are to kee= p pushing that up until it's clear that I've tripped over the "= ;real" rate.
=C2=A0

It is a leading candidate, but I would prefer to find a hardware=
partner that cared about our issues enough to work with us, rather
than ignore us as netgear did.

I fear that's g= oing to be a continual issue while we ask for things that most people don&#= 39;t understand that they want/need.=C2=A0 But I think that we'll have = an easier time if we're talking with a chipset vendor (like Marvell) vs= an OEM (like Netgear), given the relative distance each has from the kerne= l drivers...

-Aaron
--001a113534146826730516ce5198--