Re: WITHOUT_BZIP2?

Warner Losh <[email protected]> Sat, 4 Apr 2026 20:59:28 -0600
Newsgroups gmane.os.freebsd.architechture
Message-ID <CANCZdfo74f9cywv8DG2TVfv7r28D-YLd0U6FeTUBxxEu4V3w7A__33512.6774780051$1775358010$gmane$org@mail.gmail.com>
--000000000000be11a0064eadbdb6
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable

On Fri, Apr 3, 2026 at 10:43=E2=80=AFAM Enji Cooper (yaneurabeya) <
[email protected]> wrote:

>
> > On Apr 3, 2026, at 9:05=E2=80=AFAM, Lexi Winter <[email protected]> wrote=
:
> >
> > Enji Cooper (yaneurabeya) wrote in <
> [email protected]>:
> >>> On Mar 29, 2026, at 9:34=E2=80=AFAM, Dag-Erling Sm=C3=B8rgrav <des@Fr=
eeBSD.org>
> wrote:
> >>> We should probably drop both options outright, since we've never
> >>> provided corresponding options for gzip, xz / lzma, or zstd.
> >>
> >> If there was a way to build bootloaders without bzip2 support, that
> >> could save boot loader blocks for other things. That=E2=80=99s one cas=
e I can
> >> think of where it makes sense making the support work as desired.
> >
> > but this would need to be a different option, since you wouldn't want t=
o
> > build the entire system without bzip2 support just for a smaller loader=
.
> >
> > similar to how WITHOUT_ZFS and WITHOUT_LOADER_ZFS work today.
>
> Making it a dependent option though would be a good idea, e.g., if
> `${MK_BZIP2} =3D=3D no`, then `${MK_LOADER_BZIP2} =3D=3D no`. That=E2=80=
=99s what I was had
> in mind.
>

I'm not so sure. They've never been coupled before, and the loader BZIP2 is
completely independent of userland or kernel builds.

Warner

--000000000000be11a0064eadbdb6
Content-Type: text/html; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable

<div dir=3D"ltr"><div dir=3D"ltr"><br></div><br><div class=3D"gmail_quote g=
mail_quote_container"><div dir=3D"ltr" class=3D"gmail_attr">On Fri, Apr 3, =
2026 at 10:43=E2=80=AFAM Enji Cooper (yaneurabeya) &lt;<a href=3D"mailto:ya=
[email protected]">[email protected]</a>&gt; wrote:<br></div><blockqu=
ote class=3D"gmail_quote" style=3D"margin:0px 0px 0px 0.8ex;border-left:1px=
 solid rgb(204,204,204);padding-left:1ex"><br>
&gt; On Apr 3, 2026, at 9:05=E2=80=AFAM, Lexi Winter &lt;<a href=3D"mailto:=
[email protected]" target=3D"_blank">[email protected]</a>&gt; wrote:<br>
&gt; <br>
&gt; Enji Cooper (yaneurabeya) wrote in &lt;<a href=3D"mailto:9CB15D18-A91C=
[email protected]" target=3D"_blank">9CB15D18-A91C-4318-883=
[email protected]</a>&gt;:<br>
&gt;&gt;&gt; On Mar 29, 2026, at 9:34=E2=80=AFAM, Dag-Erling Sm=C3=B8rgrav =
&lt;[email protected]&gt; wrote:<br>
&gt;&gt;&gt; We should probably drop both options outright, since we&#39;ve=
 never<br>
&gt;&gt;&gt; provided corresponding options for gzip, xz / lzma, or zstd.<b=
r>
&gt;&gt; <br>
&gt;&gt; If there was a way to build bootloaders without bzip2 support, tha=
t<br>
&gt;&gt; could save boot loader blocks for other things. That=E2=80=99s one=
 case I can<br>
&gt;&gt; think of where it makes sense making the support work as desired.<=
br>
&gt; <br>
&gt; but this would need to be a different option, since you wouldn&#39;t w=
ant to<br>
&gt; build the entire system without bzip2 support just for a smaller loade=
r.<br>
&gt; <br>
&gt; similar to how WITHOUT_ZFS and WITHOUT_LOADER_ZFS work today.<br>
<br>
Making it a dependent option though would be a good idea, e.g., if `${MK_BZ=
IP2} =3D=3D no`, then `${MK_LOADER_BZIP2} =3D=3D no`. That=E2=80=99s what I=
 was had in mind.<br></blockquote><div><br></div><div>I&#39;m not so sure. =
They&#39;ve never been coupled before, and the loader BZIP2 is completely i=
ndependent of userland or kernel builds.</div><div><br></div><div>Warner=C2=
=A0</div></div></div>

--000000000000be11a0064eadbdb6--