Re: Call for tests: DRM fixes for upcoming NetBSD 10.1
Vitaly Shevtsov <[email protected]> Wed, 9 Oct 2024 02:57:43 +0500
| Newsgroups | gmane.os.netbsd.devel.x11,gmane.os.netbsd.current,gmane.os.netbsd.ports.x86-64,gmane.os.netbsd.ports.arm,gmane.os.netbsd.ports.i386 |
|---|---|
| Message-ID | <CAKydt3YBnjb38Y=JA=F5BMBH=S8Q1=OLBNeAKOmPV=Ng7dGq8A@mail.gmail.com> |
--000000000000caaed40623fe3a76 Content-Type: text/plain; charset="UTF-8" Content-Transfer-Encoding: quoted-printable I am so sorry for that noise, I didn't notice that it doesn't work with current userland. Okay then. =D1=81=D1=80, 9 =D0=BE=D0=BA=D1=82. 2024 =D0=B3., 2:27 David Brownlee <abs@= absd.org>: > Tested the Fri Oct 4 build from nyftp. No apparent regressions and > working well enough to show https://get.webgl.org in Firefox on amd64 > on the following: > - ThinkPad W520 > - ThinkPad T430 > - ThinkPad T530 > - ThinkPad T480 > - Thinkpad T495 (software rendering, wrong resolution, awful performance) > - Ryzen 5 5500/Radeon RX 5700 desktop (software rendering, correct > resolution, performance fine) > > There is something odd with booting and running with root on USB on > netbsd-10 - all but the desktop hung after "poweroff" after syncing > disks, but again - not a regression. > > David > > > On Sat, 5 Oct 2024 at 06:38, Rin Okuyama <[email protected]> wrote: > > > > Hi, > > > > We will release NetBSD 10.1 soon. Usually, non-trivial pullups > > should be avoided just before release, but we've exceptionally > > pulled up DRM changes recently: > > > > https://releng.netbsd.org/cgi-bin/req-10.cgi?show=3D928 > > > > We expect that some DRM-related problems for 10.0 have been fixed. > > On the other hand, we concern potential regressions caused by the > > pullup. It would really be appreciated if you can test 10.0_STABLE > > during **this weekend**. If it doesn't work well (compared with > > 10.0 or -current), something goes wrong. > > > > If you have GPU supported by our DRM-based drivers, and working > > 10.0 (or 10.0_STABLE) userland, please try new kernel > > (**don't forget to backup your original kernel**): > > > > - amd64: > > > https://nycdn.netbsd.org/pub/NetBSD-daily/netbsd-10/202410041820Z/amd64/b= inary/sets/kern-GENERIC.tar.xz > > > > - evbarm64: > > > https://nycdn.netbsd.org/pub/NetBSD-daily/netbsd-10/202410041820Z/evbarm-= aarch64/binary/sets/kern-GENERIC64.tar.xz > > > > - i386: > > > https://nycdn.netbsd.org/pub/NetBSD-daily/netbsd-10/202410041820Z/i386/bi= nary/sets/kern-GENERIC.tgz > > > > If kernel boots successfully, please startx(1) to see log for > > GPU-accelerated driver in /var/run/Xorg.0.log. > > > > Note that -current userland **do not** work on netbsd-10 kernel. > > > > Thanks! > > rin > --000000000000caaed40623fe3a76 Content-Type: text/html; charset="UTF-8" Content-Transfer-Encoding: quoted-printable <p dir=3D"ltr">I am so sorry for that noise, I didn't notice that it do= esn't work with current userland. Okay then. </p> <br><div class=3D"gmail_quote"><div dir=3D"ltr" class=3D"gmail_attr">=D1=81= =D1=80, 9 =D0=BE=D0=BA=D1=82. 2024 =D0=B3., 2:27 David Brownlee <<a href= =3D"mailto:[email protected]">[email protected]</a>>:<br></div><blockquote class= =3D"gmail_quote" style=3D"margin:0 0 0 .8ex;border-left:1px #ccc solid;padd= ing-left:1ex">Tested the Fri Oct=C2=A0 4 build from nyftp. No apparent regr= essions and<br> working well enough to show <a href=3D"https://get.webgl.org" rel=3D"norefe= rrer noreferrer" target=3D"_blank">https://get.webgl.org</a> in Firefox on = amd64<br> on the following:<br> - ThinkPad W520<br> - ThinkPad T430<br> - ThinkPad T530<br> - ThinkPad T480<br> - Thinkpad T495 (software rendering, wrong resolution, awful performance)<b= r> - Ryzen 5 5500/Radeon RX 5700 desktop (software rendering, correct<br> resolution, performance fine)<br> <br> There is something odd with booting and running with root on USB on<br> netbsd-10 - all but the desktop hung after "poweroff" after synci= ng<br> disks, but again - not a regression.<br> <br> David<br> <br> <br> On Sat, 5 Oct 2024 at 06:38, Rin Okuyama <<a href=3D"mailto:rokuyama.rk@= gmail.com" target=3D"_blank" rel=3D"noreferrer">[email protected]</a>&g= t; wrote:<br> ><br> > Hi,<br> ><br> > We will release NetBSD 10.1 soon. Usually, non-trivial pullups<br> > should be avoided just before release, but we've exceptionally<br> > pulled up DRM changes recently:<br> ><br> > <a href=3D"https://releng.netbsd.org/cgi-bin/req-10.cgi?show=3D928" re= l=3D"noreferrer noreferrer" target=3D"_blank">https://releng.netbsd.org/cgi= -bin/req-10.cgi?show=3D928</a><br> ><br> > We expect that some DRM-related problems for 10.0 have been fixed.<br> > On the other hand, we concern potential regressions caused by the<br> > pullup. It would really be appreciated if you can test 10.0_STABLE<br> > during **this weekend**. If it doesn't work well (compared with<br= > > 10.0 or -current), something goes wrong.<br> ><br> > If you have GPU supported by our DRM-based drivers, and working<br> > 10.0 (or 10.0_STABLE) userland, please try new kernel<br> > (**don't forget to backup your original kernel**):<br> ><br> > - amd64:<br> > <a href=3D"https://nycdn.netbsd.org/pub/NetBSD-daily/netbsd-10/2024100= 41820Z/amd64/binary/sets/kern-GENERIC.tar.xz" rel=3D"noreferrer noreferrer"= target=3D"_blank">https://nycdn.netbsd.org/pub/NetBSD-daily/netbsd-10/2024= 10041820Z/amd64/binary/sets/kern-GENERIC.tar.xz</a><br> ><br> > - evbarm64:<br> > <a href=3D"https://nycdn.netbsd.org/pub/NetBSD-daily/netbsd-10/2024100= 41820Z/evbarm-aarch64/binary/sets/kern-GENERIC64.tar.xz" rel=3D"noreferrer = noreferrer" target=3D"_blank">https://nycdn.netbsd.org/pub/NetBSD-daily/net= bsd-10/202410041820Z/evbarm-aarch64/binary/sets/kern-GENERIC64.tar.xz</a><b= r> ><br> > - i386:<br> > <a href=3D"https://nycdn.netbsd.org/pub/NetBSD-daily/netbsd-10/2024100= 41820Z/i386/binary/sets/kern-GENERIC.tgz" rel=3D"noreferrer noreferrer" tar= get=3D"_blank">https://nycdn.netbsd.org/pub/NetBSD-daily/netbsd-10/20241004= 1820Z/i386/binary/sets/kern-GENERIC.tgz</a><br> ><br> > If kernel boots successfully, please startx(1) to see log for<br> > GPU-accelerated driver in /var/run/Xorg.0.log.<br> ><br> > Note that -current userland **do not** work on netbsd-10 kernel.<br> ><br> > Thanks!<br> > rin<br> </blockquote></div> --000000000000caaed40623fe3a76--