Re: Experimental musl Port for FreeBSD
Pedro Giffuni <[email protected]> Sun, 26 Jul 2026 22:39:22 +0000 (UTC)
| Newsgroups | gmane.os.freebsd.devel.hackers |
|---|---|
| Message-ID | <[email protected]> |
------=_Part_2436785_764867873.1785105562853
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: quoted-printable
JFYI;
Someone started an implementation for OpenBSD , actually hyperbolaBSD, as w=
ell:
https://git.hyperbola.info:50100/hyperbolabsd/musl.git/
Cheers,
Pedro.
On Sunday, July 26, 2026 at 10:42:48 AM GMT-5, ShengYi Hung <aokblast@f=
reebsd.org> wrote: =20
=20
=20
For me, having another libc implementation would make it easier to perform =
A/B testing when debugging LLVM components (such as compiler-rt, LLDB, and =
other tools) on FreeBSD. I have also seen people test libm precision across=
multiple platforms with different libc implementations. In this case, havi=
ng another libc implementation can help determine whether an issue is speci=
fic to FreeBSD or not.
=20
Additionally, some users may want to build their own userspace with a diff=
erent libc. Since the libc interface is relatively stable, I think it is wo=
rthwhile to spend some time porting another libc implementation to FreeBSD.=
If this project gains enough interest, I am willing to rewrite the AI-gene=
rated patch myself and work toward getting it merged upstream.
=20
Regarding the musl libc AI policy, my understanding is that they do not ac=
cept AI-generated patches. I also do not feel comfortable submitting AI-gen=
erated patches to a project that I am not familiar with, so I will not do t=
hat.
=20
=20
=20
Best
=20
ShengYi Hung.
=20
=20
On 7/26/26 04:43, Pedro Giffuni wrote:
=20
=20
Kudos!=20
On Sat, Jul 25, 2026 at 8:31=E2=80=AFPM ShengYi Hung <[email protected]=
g> wrote: > > Hi all FreeBSD hackers, > > Since Claude reset my token quota=
last week and I couldn't use all of it, I decided > to spend some time on =
an interesting side project: porting musl to FreeBSD.=20
FWIW, I was proposing that as a GSoC some years ago. It would not ever r=
eplace our native libc but there are always people wanting to run experimen=
ts like creating distributions like Chimera BSD or Alpine BSD. Pretty much =
the reason there is/was a FreeBSD glibc port.=20
I wonder if musl libc has some policy regarding AI contibutions :).=20
Pedro. =20
------=_Part_2436785_764867873.1785105562853
Content-Type: text/html; charset=UTF-8
Content-Transfer-Encoding: quoted-printable
<html><head></head><body><div class=3D"ydp5594d611yahoo-style-wrap" style=
=3D"font-family:Helvetica Neue, Helvetica, Arial, sans-serif;font-size:16px=
;"><div></div>
<div dir=3D"ltr" data-setdir=3D"false">JFYI;</div><div dir=3D"ltr" =
data-setdir=3D"false"><br></div><div dir=3D"ltr" data-setdir=3D"false">Some=
one started an implementation for OpenBSD , actually hyperbolaBSD, as well:=
</div><div dir=3D"ltr" data-setdir=3D"false"><br></div><div dir=3D"ltr" dat=
a-setdir=3D"false"><a href=3D"https://git.hyperbola.info:50100/hyperbolabsd=
/musl.git/" rel=3D"nofollow" target=3D"_blank">https://git.hyperbola.info:5=
0100/hyperbolabsd/musl.git/</a><br></div><div dir=3D"ltr" data-setdir=3D"fa=
lse"><br></div><div dir=3D"ltr" data-setdir=3D"false">Cheers,</div><div dir=
=3D"ltr" data-setdir=3D"false"><br></div><div dir=3D"ltr" data-setdir=3D"fa=
lse">Pedro.</div><div><br></div>
=20
</div><div id=3D"ydp3dbc356ayahoo_quoted_5510516866" class=3D"ydp3d=
bc356ayahoo_quoted">
<div style=3D"font-family:'Helvetica Neue', Helvetica, Arial, s=
ans-serif;font-size:13px;">
=20
<div>
On Sunday, July 26, 2026 at 10:42:48 AM GMT-5, ShengYi =
Hung <[email protected]> wrote:
</div>
<div><br></div>
<div><br></div>
<div><div id=3D"ydp3dbc356ayiv6265170617"><div>
<p>For me, having another libc implementation would make it easier
to perform A/B testing when debugging LLVM components (such as
compiler-rt, LLDB, and other tools) on FreeBSD. I have also seen
people test libm precision across multiple platforms with
different libc implementations. In this case, having another libc
implementation can help determine whether an issue is specific to
FreeBSD or not.<br clear=3D"none">
<br clear=3D"none">
Additionally, some users may want to build their own userspace
with a different libc. Since the libc interface is relatively
stable, I think it is worthwhile to spend some time porting
another libc implementation to FreeBSD. If this project gains
enough interest, I am willing to rewrite the AI-generated patch
myself and work toward getting it merged upstream.<br clear=3D"none">
<br clear=3D"none">
Regarding the musl libc AI policy, my understanding is that they
do not accept AI-generated patches. I also do not feel comfortable
submitting AI-generated patches to a project that I am not
familiar with, so I will not do that.<br clear=3D"none">
<br clear=3D"none">
</p>
<p>Best</p>
<p>ShengYi Hung.</p>
<p><br clear=3D"none">
</p>
<div id=3D"ydp3dbc356ayiv6265170617yqt04734" class=3D"ydp3dbc356ayiv626=
5170617yqt5941283284"><div class=3D"ydp3dbc356ayiv6265170617moz-cite-prefix=
">On 7/26/26 04:43, Pedro Giffuni wrote:<br clear=3D"none">
</div>
<blockquote type=3D"cite">
</blockquote></div></div><div id=3D"ydp3dbc356ayiv6265170617yqt84207"=
class=3D"ydp3dbc356ayiv6265170617yqt5941283284"><div><div style=3D"font-fa=
mily:Helvetica Neue, Helvetica, Arial, sans-serif;font-size:16px;" class=3D=
"ydp3dbc356ayiv6265170617yahoo-style-wrap">
<div dir=3D"ltr">
<div>
<div dir=3D"ltr">Kudos!</div>
<div dir=3D"ltr"><br clear=3D"none">
</div>
<div>On Sat, Jul 25, 2026 at 8:31=E2=80=AFPM ShengYi Hung
<a shape=3D"rect" href=3D"mailto:[email protected]" class=
=3D"ydp3dbc356ayiv6265170617moz-txt-link-rfc2396E" rel=3D"nofollow" target=
=3D"_blank"><[email protected]></a> wrote:</div>
<div>></div>
<div>> Hi all FreeBSD hackers,</div>
<div>></div>
<div>> Since Claude reset my token quota last week and I
couldn't use all of it, I decided</div>
<div>> to spend some time on an interesting side project:
porting musl to FreeBSD.</div>
<div dir=3D"ltr"><br clear=3D"none">
</div>
</div>
FWIW, I was proposing that as a GSoC some years ago. It would
not ever replace</div>
<div dir=3D"ltr">our native libc but there are
always people wanting to run experiments like creating</div>
<div dir=3D"ltr">distributions like Chimera
BSD or Alpine BSD. Pretty much the reason there is/was</div>
<div dir=3D"ltr">a FreeBSD glibc port.</div>
<div dir=3D"ltr"><br clear=3D"none">
</div>
<div dir=3D"ltr">I wonder if musl libc has
some policy regarding AI contibutions :).</div>
<div dir=3D"ltr"><br clear=3D"none">
</div>
<div dir=3D"ltr">Pedro.</div>
</div>
=20
</div></div></div></div>
</div>
</div></body></html>
------=_Part_2436785_764867873.1785105562853--