Skip site navigation (1)Skip section navigation (2)
Date:      Mon, 15 Jan 2018 23:47:28 +0000
From:      bugzilla-noreply@freebsd.org
To:        freebsd-bugs@FreeBSD.org
Subject:   [Bug 225197] `make buildkernel' fails on a machine with 1GB RAM
Message-ID:  <bug-225197-8-WhLAfp9HCN@https.bugs.freebsd.org/bugzilla/>
In-Reply-To: <bug-225197-8@https.bugs.freebsd.org/bugzilla/>
References:  <bug-225197-8@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=3D225197

Mark Millard <markmi@dsl-only.net> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |markmi@dsl-only.net

--- Comment #4 from Mark Millard <markmi@dsl-only.net> ---
(In reply to Wolfram Schneider from comment #3)

Also, as I understand it, with only 1 GiByte of
RAM assigned FreeBSD will self limit the swap space,
reporting somethings like (not necessarily with
matching figures):

warning: total configured swap (524288 pages) exceeds maximum recommended
amount (405460 pages).

[Note the above is from a RPi2B V1.1 with 1 GiByte
of RAM (armv7). An RPi3B (aarch64) allows far more
but still limits it.]

It will also say something like:

warning: increase kern.maxswzone or reduce amount of swap.

but looking at the code that "increase kern.maxswzone" only
applies if one has already set kern.maxswzone to a figure
that limited the amount of swap. The default value of
zero makes kern.maxswzone be otherwise ignored. (But
i386 seemed to have a non-zero default, if I remember
correctly.)

Quoting "man 8 loader" :

    kern.maxswzone
                  . . .

                  Note that swap metadata can be fragmented, which means th=
at
                  the system can run out of space before it reaches the
                  theoretical limit.  Therefore, care should be taken to not
                  configure more swap than approximately half of the
                  theoretical maximum.

                 . . .

It turns out that the "maximum recommended amount" earlier is that
"half". But the theoretical maximum is not always the "8 times the
amount of physical memory" that man page indicates (or anywhere
near even 6 times). (The RPi2B V1.1 is an example of that.)

--=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-225197-8-WhLAfp9HCN>