From owner-freebsd-chromium@freebsd.org Mon Aug 22 06:25:59 2016 Return-Path: Delivered-To: freebsd-chromium@mailman.ysv.freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:1900:2254:206a::19:1]) by mailman.ysv.freebsd.org (Postfix) with ESMTP id 9804FBC1F2B for ; Mon, 22 Aug 2016 06:25:59 +0000 (UTC) (envelope-from ws@au.dyndns.ws) Received: from ipmail04.adl6.internode.on.net (ipmail04.adl6.internode.on.net [150.101.137.141]) by mx1.freebsd.org (Postfix) with ESMTP id 32DCB1CBF for ; Mon, 22 Aug 2016 06:25:58 +0000 (UTC) (envelope-from ws@au.dyndns.ws) Received: from ppp103-111.static.internode.on.net (HELO lillith-iv.ovirt.dyndns.ws) ([150.101.103.111]) by ipmail04.adl6.internode.on.net with ESMTP; 22 Aug 2016 15:50:41 +0930 X-Envelope-From: ws@au.dyndns.ws X-Envelope-To: freebsd-chromium@freebsd.org Received: from predator-ii.buffyverse (predator-ii.buffyverse [172.17.17.136]) by lillith-iv.ovirt.dyndns.ws (8.14.9/8.14.9) with ESMTP id u7M6KTP0038316; Mon, 22 Aug 2016 15:50:30 +0930 (ACST) (envelope-from ws@au.dyndns.ws) Message-ID: <1471846829.4089.12.camel@au.dyndns.ws> Subject: Re: 52.0.2743.82 (64-bit) to go, Aw snap fixed as well. From: Wayne Sierke To: Arto Pekkanen , clutton Cc: freebsd-chromium@freebsd.org Date: Mon, 22 Aug 2016 15:50:29 +0930 In-Reply-To: <3cdd3ac94cc67da02f16c7899e405201@kapsi.fi> References: <1471486169.7533.8.camel@zoho.com> <1471487392.7533.22.camel@zoho.com> <3cdd3ac94cc67da02f16c7899e405201@kapsi.fi> Content-Type: text/plain; charset="UTF-8" X-Mailer: Evolution 3.18.5.1 FreeBSD GNOME Team Port Mime-Version: 1.0 Content-Transfer-Encoding: 8bit X-Greylist: Sender IP whitelisted, not delayed by milter-greylist-4.4.3 (lillith-iv.ovirt.dyndns.ws [172.17.17.142]); Mon, 22 Aug 2016 15:50:30 +0930 (ACST) X-Scanned-By: MIMEDefang 2.75 on 172.17.17.142 X-Scanned-By: SpamAssassin 3.004000(2014-02-07) X-Scanned-By: ClamAV X-Spam-Score: -1 () ALL_TRUSTED X-BeenThere: freebsd-chromium@freebsd.org X-Mailman-Version: 2.1.22 Precedence: list List-Id: FreeBSD-specific Chromium issues List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Mon, 22 Aug 2016 06:25:59 -0000 On Sat, 2016-08-20 at 21:58 +0300, Arto Pekkanen wrote: > You are so AWESOME!!! :) Thank you very much for your efforts! > > There was a paid bounty posted for solving this problem, did anybody  > claim it yet? > > Also, I want to ask a dumb question (since I am not a dev): was the  > reason for "Aw snap!" tab crashes because of the difference in > behavior  > of mmap() between Linux and FreeBSD? If not, what was the actual > reason  > for this problem? I mean these problems did not manifest in Linux, > so  > this whole issue was really arcane :S > > clutton kirjoitti 18.08.2016 05:29: > > > > On Thu, 2016-08-18 at 05:09 +0300, clutton wrote: > > > > > > I've just fixed the Aw, snap. I believe so. > > > > > > The patch is going to be huge, not because of this simple bug. > > > New > > > version was ported as well with some extra things. I need few > > > more > > > days > > > to polish everything and create patches for some things that are > > > to > > > ugly now. And probably few days more to be sure that it works > > > everywhere 9, 10, HEAD. The current porting was done on HEAD. > > I see. 52 was already ported. I actually asked on this maillist, If > > anyone working on this? It's bad when the same work done twice. > > > > From my liaskos/freebsd-chromium reply::: > > Oh, I did my own porting as well. My first working version were > > done a > > week ago but real fixing 'Aw, snap' were done just now. Such a pity > > that we both did the work. I spend endless time with lldb. I also > > ported tools/gn/bootstrap and a lot of other stuff were polished. Unfortunately while the current pkg:  chromium-52.0.2743.116         www/chromium appears to have eliminated the "Aw, snap!" behaviour, pages now just stall quietly, becoming unresponsive. Stopping the page loading and refreshing the page doesn't recover and the stalled tab has to be closed and re-opened. >From a UE perspective the "Aw, snap!" behaviour was actually better since it at least provided a definite indicator of failure and (in my case at least) failed "Aw, snap!" pages were recoverable with just a refresh - albeit of late the "Aw, snap!" was tending to recur more frequently on refreshes than I had noticed with earlier versions of chromium.