Skip site navigation (1)Skip section navigation (2)
Date:      Tue, 13 Dec 2016 08:12:25 -0600
From:      Kyle Evans <kevans91@ksu.edu>
To:        <koobs@freebsd.org>
Cc:        Marcus von Appen <mva@freebsd.org>, <python@freebsd.org>
Subject:   Re: lang/python3* ports, __pycache__ included
Message-ID:  <CACNAnaFMe=%2B%2BZW2vgRnWNMLLU_RG21QeaFhyi-cgeBLHYyrWzg@mail.gmail.com>
In-Reply-To: <fef69799-bee5-d836-70cf-d928707ccfdb@FreeBSD.org>
References:  <CACNAnaEueRdkEuuf9MmZwqqaz8HB6hSW14a_VmqZ9%2B8ub3235g@mail.gmail.com> <ema74d97cd-d111-4740-909d-419a4d12c8a6@hora> <CACNAnaF1=D-P0ZTbZP7Wt=BBoWwmDnwe=-mRmz%2Bga_H__zk4-g@mail.gmail.com> <fef69799-bee5-d836-70cf-d928707ccfdb@FreeBSD.org>

next in thread | previous in thread | raw e-mail | index | archive | help
On Tue, Dec 13, 2016 at 4:59 AM, Kubilay Kocak <koobs@freebsd.org> wrote:
>
> Note that they will be re-created on import unless one overrides the
> default for the interpreter to produce these optimization files.
>
> I'm not sure (or cant remember) if there is a good way to disable this
> on a global or system-wide basis.

Yeah, I was aware of this -- my response to mva@ mentioned that
runtime creation is ok. =) What I'm really trying to go for (and still
working out the details -- still planning, decent amount of time
before implementation), is going to have a small-disk constraint
(Think < 1GB), but a decent amount of memory. What I *really* want to
do that I haven't even started to work out the logistics on is
effectively overlay this disk into memory with no writing back to
disk. Something similar to a nullfs with copy-on-write semantics is
what I was hoping for, but that doesn't seem to be easy to accomplish.



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?CACNAnaFMe=%2B%2BZW2vgRnWNMLLU_RG21QeaFhyi-cgeBLHYyrWzg>