From mboxrd@z Thu Jan 1 00:00:00 1970 MIME-Version: 1.0 In-Reply-To: References: Date: Sat, 1 Mar 2014 23:14:52 +0000 Message-ID: From: Charles Forsyth To: Fans of the OS Plan 9 from Bell Labs <9fans@9fans.net> Content-Type: multipart/alternative; boundary=001a11c34908c47d3e04f393b844 Subject: Re: [9fans] imagereclaim() Topicbox-Message-UUID: c06fc380-ead8-11e9-9d60-3106f5b1d025 --001a11c34908c47d3e04f393b844 Content-Type: text/plain; charset=UTF-8 On 1 March 2014 23:05, wrote: > how is this possible? the old page is locked during the call > to duppage(). so duppage() will not run concurrently on the > same page. > there's a race to decide whether to do it, and what happens to the page once it has. it reaches back into the page load code. you can easily end up with many copies of the same page. worse, duppage calls a putmmu/mmuput, which in some implementations calls newpage, which can lead to deadlock, since duppage has the conch. the whole thing is meant well, but misconceived. i can't even remember the details since it's so long since i cut out all that stuff. --001a11c34908c47d3e04f393b844 Content-Type: text/html; charset=UTF-8 Content-Transfer-Encoding: quoted-printable

= On 1 March 2014 23:05, <cinap_lenrek@felloff.net> wr= ote:
how is this possibl= e? the old page is locked during the call
to duppage(). so duppage() will not run concurrently on the
same page.

there's a race to decide whether= to do it, and what happens to the page once it has. it reaches back into t= he page load code.
you can easily end up wi= th many copies of the same page. worse, duppage calls a putmmu/mmuput,
which in some implementations calls newpage, whi= ch can lead to deadlock, since duppage has the conch.
the whole thing is meant well, but misconceived. i can't eve= n remember the details since it's so long since i
cut out all that stuff.
--001a11c34908c47d3e04f393b844--