Re: PS/2 deprecation situation

Warner Losh <[email protected]> Sat, 18 Jul 2026 22:10:51 -0600
Newsgroups gmane.os.freebsd.devel.hackers
Message-ID <CANCZdfo93ij+xpNbAzGO3u71ofTm_SpW4AmKmeULKk0AVOWDvQ@mail.gmail.com>
--000000000000572de30656eefaff
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable

On Sat, Jul 18, 2026 at 5:24=E2=80=AFPM Tomoaki AOKI <[email protected]=
e.jp>
wrote:

> On Sun, 19 Jul 2026 00:16:48 +0300
> Peter Pentchev <[email protected]> wrote:
>
> > On Sun, Jul 19, 2026 at 12:09:39AM +0300, Peter Pentchev wrote:
> > > On Sat, Jul 18, 2026 at 10:15:22PM +0200, vermaden wrote:
> > > >
> > > > > What is the reason to remove it?
> > > > > Isn't it reasonable to keep it and,
> > > > > if needed, adapt it to current needs?
> > > >
> > > > If You would ask me - then I would just 'keep' it - I do not see an=
y
> reason to remove it.
> > >
> > > Note that I am ot...
> > [snip]
> > > [...] comromise [...]
> > [snip]
> > > their continued maintenance - actual users actually, wel, use them :)
> > > Stil, [...]
> > [snip even more dropped letters every here and there]
> >
> > ...and it is a bit ironic that I would reply to a thread about PS/2
> > keyboard support and write my reply on a (different type of) keyboard
> > that is having more and more problems with keypresses not registering.
> >
> > Apologies if anyone found my message difficult to read because of that =
:)
> >
> > G'luck,
> > Peter
> >
> > --
> > Peter Pentchev  [email protected] [email protected] [email protected]
> > PGP key:        https://www.ringlet.net/roam/roam.key.asc
> > Key fingerprint 2EE7 A7A5 17FC 124C F115  C354 651E EFB0 2527 DF13
>
> The point of the deprecation notice is to make FreeBSD kernel
> "Giant Lock Free", and almost all others are fixed (IIRC, mostly
> devices on non-ISA buses) or removed (IIRC, mostly ISA bus only
> and old, unused devices). Actual deletions would mean "OK,
> this is too hard to fix and no one states using it on polls.
> Let's remove!".
>
> IIRC, Warner and some others are still trying to fix PS/2
> device drivers but in relatively lower priorities compared
> to his / their other works.
>
> And as vermaden already noted, these are too important
> and still widely and actually used, the schedule is slipped
> (delayed) several major releases, rather than giving up and
> removing.
>

We're removing Giant. That's non-negotiable. It's going to happen. When...
that's a more complicated issue.

We're getting close to having Giant removed. We're not going to remove
psm or atkbd or atkbdc. They'll get converted. They are one of the few
old-school devices that are still relevant and if Intel can't get rid of
them entirely, I doubt we can any time soon. I'm loathe to remove
the warning, though, since it keeps the pressure on.

Warner

--000000000000572de30656eefaff
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 Sat, Jul 18,=
 2026 at 5:24=E2=80=AFPM Tomoaki AOKI &lt;<a href=3D"mailto:[email protected]=
kura.ne.jp">[email protected]</a>&gt; wrote:<br></div><blockquote c=
lass=3D"gmail_quote" style=3D"margin:0px 0px 0px 0.8ex;border-left:1px soli=
d rgb(204,204,204);padding-left:1ex">On Sun, 19 Jul 2026 00:16:48 +0300<br>
Peter Pentchev &lt;<a href=3D"mailto:[email protected]" target=3D"_blank">ro=
[email protected]</a>&gt; wrote:<br>
<br>
&gt; On Sun, Jul 19, 2026 at 12:09:39AM +0300, Peter Pentchev wrote:<br>
&gt; &gt; On Sat, Jul 18, 2026 at 10:15:22PM +0200, vermaden wrote:<br>
&gt; &gt; &gt; <br>
&gt; &gt; &gt; &gt; What is the reason to remove it?<br>
&gt; &gt; &gt; &gt; Isn&#39;t it reasonable to keep it and,<br>
&gt; &gt; &gt; &gt; if needed, adapt it to current needs?<br>
&gt; &gt; &gt; <br>
&gt; &gt; &gt; If You would ask me - then I would just &#39;keep&#39; it - =
I do not see any reason to remove it.<br>
&gt; &gt; <br>
&gt; &gt; Note that I am ot...<br>
&gt; [snip]<br>
&gt; &gt; [...] comromise [...]<br>
&gt; [snip]<br>
&gt; &gt; their continued maintenance - actual users actually, wel, use the=
m :)<br>
&gt; &gt; Stil, [...]<br>
&gt; [snip even more dropped letters every here and there]<br>
&gt; <br>
&gt; ...and it is a bit ironic that I would reply to a thread about PS/2<br=
>
&gt; keyboard support and write my reply on a (different type of) keyboard<=
br>
&gt; that is having more and more problems with keypresses not registering.=
<br>
&gt; <br>
&gt; Apologies if anyone found my message difficult to read because of that=
 :)<br>
&gt; <br>
&gt; G&#39;luck,<br>
&gt; Peter<br>
&gt; <br>
&gt; -- <br>
&gt; Peter Pentchev=C2=A0 <a href=3D"mailto:[email protected]" target=3D"_bl=
ank">[email protected]</a> <a href=3D"mailto:[email protected]" target=3D"_bla=
nk">[email protected]</a> <a href=3D"mailto:[email protected]" target=3D"_=
blank">[email protected]</a><br>
&gt; PGP key:=C2=A0 =C2=A0 =C2=A0 =C2=A0 <a href=3D"https://www.ringlet.net=
/roam/roam.key.asc" rel=3D"noreferrer" target=3D"_blank">https://www.ringle=
t.net/roam/roam.key.asc</a><br>
&gt; Key fingerprint 2EE7 A7A5 17FC 124C F115=C2=A0 C354 651E EFB0 2527 DF1=
3<br>
<br>
The point of the deprecation notice is to make FreeBSD kernel<br>
&quot;Giant Lock Free&quot;, and almost all others are fixed (IIRC, mostly<=
br>
devices on non-ISA buses) or removed (IIRC, mostly ISA bus only<br>
and old, unused devices). Actual deletions would mean &quot;OK,<br>
this is too hard to fix and no one states using it on polls.<br>
Let&#39;s remove!&quot;.<br>
<br>
IIRC, Warner and some others are still trying to fix PS/2<br>
device drivers but in relatively lower priorities compared<br>
to his / their other works.<br>
<br>
And as vermaden already noted, these are too important<br>
and still widely and actually used, the schedule is slipped<br>
(delayed) several major releases, rather than giving up and<br>
removing.<br></blockquote><div><br></div><div>We&#39;re removing Giant. Tha=
t&#39;s non-negotiable. It&#39;s going to happen. When...</div><div>that&#3=
9;s a more complicated issue.</div><div><br></div><div>We&#39;re getting cl=
ose to having Giant removed. We&#39;re not going to remove</div><div>psm or=
 atkbd or atkbdc. They&#39;ll get converted. They are one of the few</div><=
div>old-school devices that are still relevant and if Intel can&#39;t get r=
id of</div><div>them entirely, I doubt we can any time soon. I&#39;m loathe=
 to remove</div><div>the warning, though, since it keeps the pressure on.</=
div><div><br></div><div><span style=3D"background-color:transparent">Warner=
</span></div></div></div>

--000000000000572de30656eefaff--