NetBSD Problem Report #55198
From www@netbsd.org Thu Apr 23 01:10:52 2020
Return-Path: <www@netbsd.org>
Received: from mail.netbsd.org (mail.netbsd.org [199.233.217.200])
(using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits))
(Client CN "mail.NetBSD.org", Issuer "mail.NetBSD.org CA" (not verified))
by mollari.NetBSD.org (Postfix) with ESMTPS id 880D11A9218
for <gnats-bugs@gnats.NetBSD.org>; Thu, 23 Apr 2020 01:10:52 +0000 (UTC)
Message-Id: <20200423011051.5850D1A924D@mollari.NetBSD.org>
Date: Thu, 23 Apr 2020 01:10:51 +0000 (UTC)
From: joaopaulo1511@hotmail.com
Reply-To: joaopaulo1511@hotmail.com
To: gnats-bugs@NetBSD.org
Subject: When the video card is Intel Mobile 965GM Express, i915drmkms has error CPU pipe A FIFO underrun, also X freezes/hangs
X-Send-Pr-Version: www-1.0
>Number: 55198
>Category: kern
>Synopsis: When the video card is Intel Mobile 965GM Express, i915drmkms has error CPU pipe A FIFO underrun, also X freezes/hangs
>Confidential: no
>Severity: serious
>Priority: medium
>Responsible: kern-bug-people
>State: closed
>Class: sw-bug
>Submitter-Id: net
>Arrival-Date: Thu Apr 23 01:15:00 +0000 2020
>Closed-Date: Sat Jul 04 08:26:04 +0000 2020
>Last-Modified: Sat Jul 04 08:35:01 +0000 2020
>Originator: Joćo Paulo Vinha Bittar
>Release: 9.0
>Organization:
home user
>Environment:
NetBSD localhost 9.0 NetBSD 9.0 (GENERIC) #0: Fri Feb 14 00:06:28 UTC 2020 mkrepro@mkrepro.NetBSD.org:/usr/src/sys/arch/i386/compile/GENERIC i386
>Description:
This problem report is similar to the one reported at "http://gnats.netbsd.org/cgi-bin/query-pr-single.pl?number=55133", but I think it is sufficiently different to deserve its own PR#.
On a laptop with Intel Mobile 965GM Express Chipset (i915drmkms driver on kernel), booting NetBSD 9.0 gives the following "dmesg | grep intel" lines:
[ 5.467170] kern error: [drm:(/usr/src/sys/external/bsd/drm2/dist/drm/i915/intel_fifo_underrun.c:363)intel_cpu_fifo_underrun_irq_handler] *ERROR* CPU pipe A FIFO underrun
[ 5.747346] intelfb0 at i915drmkms0
[ 5.747346] intelfb0: framebuffer at 0xdb891000, size 1280x800, depth 32, stride 5120
[ 6.697919] wsdisplay0 at intelfb0 (kbdmux ignored): console (default, vt100 emulation), using wskbd0
[ 15.893517] kern error: [drm:(/usr/src/sys/external/bsd/drm2/dist/drm/i915/intel_fifo_underrun.c:363)intel_cpu_fifo_underrun_irq_handler] *ERROR* CPU pipe A FIFO underrun
[ 5.398721] kern error: [drm:(/usr/src/sys/external/bsd/drm2/dist/drm/i915/intel_fifo_underrun.c:141)i9xx_set_fifo_underrun_reporting] *ERROR* pipe A underrun
[ 5.398721] kern error: [drm:(/usr/src/sys/external/bsd/drm2/dist/drm/i915/intel_fifo_underrun.c:363)intel_cpu_fifo_underrun_irq_handler] *ERROR* CPU pipe A FIFO underrun
[ 5.678906] intelfb0 at i915drmkms0
[ 5.678906] intelfb0: framebuffer at 0xdb88b000, size 1280x800, depth 32, stride 5120
[ 6.409366] wsdisplay0 at intelfb0 (kbdmux ignored): console (default, vt100 emulation), using wskbd0
[ 15.735309] kern error: [drm:(/usr/src/sys/external/bsd/drm2/dist/drm/i915/intel_fifo_underrun.c:363)intel_cpu_fifo_underrun_irq_handler] *ERROR* CPU pipe A FIFO underrun
[ 6.368271] kern error: [drm:(/usr/src/sys/external/bsd/drm2/dist/drm/i915/intel_fifo_underrun.c:363)intel_cpu_fifo_underrun_irq_handler] *ERROR* CPU pipe A FIFO underrun
[ 6.648435] intelfb0 at i915drmkms0
[ 6.648435] intelfb0: framebuffer at 0xdb8b0000, size 1280x800, depth 32, stride 5120
[ 7.618971] wsdisplay0 at intelfb0 (kbdmux ignored): console (default, vt100 emulation), using wskbd0
[ 16.784096] kern error: [drm:(/usr/src/sys/external/bsd/drm2/dist/drm/i915/intel_fifo_underrun.c:363)intel_cpu_fifo_underrun_irq_handler] *ERROR* CPU pipe A FIFO underrun
Also, running startx doesn't open the default twm environment with three xterm and xclock opened, as X freezes/hangs after showing the first xterm. It is needed to change back to the virtual terminal (CTRL+ALT+F#) and kill the X server, as even the only xterm that shows up is frozen/hung.
>How-To-Repeat:
Install NetBSD 9.0 on a laptop with Intel Mobile 965GM Express Chipset, boot it and run startx. You will see the dmsg error messages in green and X server hanging up/freezing.
I did that on an Acer Aspire 5720Z which came with this specific graphics card (some models may come with Intel Mobile 960GM or 965PM).
>Fix:
I don't know how to fix it, but I know that FreeBSD 12.1 (i386) boots without errors and X works just by running startx.
>Release-Note:
>Audit-Trail:
From: Izumi Tsutsui <tsutsui@ceres.dti.ne.jp>
To: gnats-bugs@netbsd.org
Cc: tsutsui@ceres.dti.ne.jp
Subject: Re: kern/55198: When the video card is Intel Mobile 965GM Express,
i915drmkms has error CPU pipe A FIFO underrun, also X freezes/hangs
Date: Fri, 24 Apr 2020 04:58:13 +0900
Maybe something similar as PR/54995 on Atom N455?
https://gnats.netbsd.org/54995
How about adding
> option "AccelMethod" "uxa"
in /etc/X11/xorg.conf ?
If SNA accelaration is problematic on old Intel chips,
maybe we should fallback to UXA on such pre-2010 chipset?
---
Izumi Tsutsui
From: =?Windows-1252?Q?Jo=E3o_Paulo_Vinha_Bittar?= <joaopaulo1511@hotmail.com>
To: "kern-bug-people@netbsd.org" <kern-bug-people@netbsd.org>,
"gnats-admin@netbsd.org" <gnats-admin@netbsd.org>, "netbsd-bugs@netbsd.org"
<netbsd-bugs@netbsd.org>, "gnats-bugs@netbsd.org" <gnats-bugs@netbsd.org>
Cc:
Subject: Re: kern/55198: When the video card is Intel Mobile 965GM Express,
i915drmkms has error CPU pipe A FIFO underrun, also X freezes/hangs
Date: Sat, 25 Apr 2020 20:43:17 +0000
--_000_DM6PR04MB4906B4CC95635AE7F430632CA7D00DM6PR04MB4906namp_
Content-Type: text/plain; charset="Windows-1252"
Content-Transfer-Encoding: quoted-printable
Hi Izumi Tsutsui,
I forgot to mention, but I did run X -configure, edited the xorg.conf.new t=
o have "AccelMethod" "uxa", moved it to /etc/X11/xorg.conf, but it didn't w=
ork on NetBSD 9.0.
I also tried with "Accel" "false", but X still hanged.
I tried the oldest NetBSD version still available at ftp.NetBSD.org, which =
is 7.1.2 (this time on amd64), and X did work out of the box. It also didn=
't have any =93dmesg | grep intel=94 errors.
When I upgraded to NetBSD 7.2 (also amd64), X didn't work out of the box, n=
either worked after modifying xorg.conf. It also didn't have any errors sh=
ow when running =93dmesg | grep intel=94.
So I think something happened between 7.1.2 and 7.2.
________________________________
From: Izumi Tsutsui <tsutsui@ceres.dti.ne.jp>
Sent: Thursday, April 23, 2020 4:00:03 PM
To: kern-bug-people@netbsd.org <kern-bug-people@netbsd.org>; gnats-admin@ne=
tbsd.org <gnats-admin@netbsd.org>; netbsd-bugs@netbsd.org <netbsd-bugs@netb=
sd.org>; joaopaulo1511@hotmail.com <joaopaulo1511@hotmail.com>
Subject: Re: kern/55198: When the video card is Intel Mobile 965GM Express,=
i915drmkms has error CPU pipe A FIFO underrun, also X freezes/hangs
The following reply was made to PR kern/55198; it has been noted by GNATS.
From: Izumi Tsutsui <tsutsui@ceres.dti.ne.jp>
To: gnats-bugs@netbsd.org
Cc: tsutsui@ceres.dti.ne.jp
Subject: Re: kern/55198: When the video card is Intel Mobile 965GM Express,
i915drmkms has error CPU pipe A FIFO underrun, also X freezes/hang=
s
Date: Fri, 24 Apr 2020 04:58:13 +0900
Maybe something similar as PR/54995 on Atom N455?
https://gnats.netbsd.org/54995
How about adding
> option "AccelMethod" "uxa"
in /etc/X11/xorg.conf ?
If SNA accelaration is problematic on old Intel chips,
maybe we should fallback to UXA on such pre-2010 chipset?
---
Izumi Tsutsui
--_000_DM6PR04MB4906B4CC95635AE7F430632CA7D00DM6PR04MB4906namp_
Content-Type: text/html; charset="Windows-1252"
Content-Transfer-Encoding: quoted-printable
<html>
<head>
<meta http-equiv=3D"Content-Type" content=3D"text/html; charset=3DWindows-1=
252">
</head>
<body>
<div data-ogsc=3D"" style=3D"">
<meta content=3D"text/html; charset=3Diso-8859-1" data-ogsc=3D"" style=3D""=
>
</div>
<div dir=3D"auto" style=3D"direction: ltr; margin: 0px; padding: 0px; font-=
family: sans-serif; font-size: 11pt; color: black;">
Hi Izumi Tsutsui,<br>
<br>
</div>
<div dir=3D"auto" style=3D"direction: ltr; margin: 0px; padding: 0px; font-=
family: sans-serif; font-size: 11pt; color: black;">
I forgot to mention, but I did run X -configure, edited the xorg.conf.new t=
o have "AccelMethod" "uxa", moved it to /etc/X11/xorg.c=
onf, but it didn't work on NetBSD 9.0.<br>
<br>
</div>
<div dir=3D"auto" style=3D"direction: ltr; margin: 0px; padding: 0px; font-=
family: sans-serif; font-size: 11pt; color: black;">
I also tried with "Accel" "false", but X still hanged.<=
br>
</div>
<div dir=3D"auto" style=3D"direction: ltr; margin: 0px; padding: 0px; font-=
family: sans-serif; font-size: 11pt; color: black;">
<br>
</div>
<div dir=3D"auto" style=3D"direction: ltr; margin: 0px; padding: 0px; font-=
family: sans-serif; font-size: 11pt; color: black; text-align: left;">
I tried the oldest NetBSD version still available at ftp.NetBSD.org, which =
is 7.1.2 (this time on amd64), and X did work out of the box. It also=
didn't have any =93dmesg | grep intel=94 errors.</div>
<div dir=3D"auto" style=3D"direction: ltr; margin: 0px; padding: 0px; font-=
family: sans-serif; font-size: 11pt; color: black; text-align: left;">
<br>
</div>
<div dir=3D"auto" style=3D"direction: ltr; margin: 0px; padding: 0px; font-=
family: sans-serif; font-size: 11pt; color: black; text-align: left;">
When I upgraded to NetBSD 7.2 (also amd64), X didn't work out of the box, n=
either worked after modifying xorg.conf. It also didn't have any erro=
rs show when running =93dmesg | grep intel=94.</div>
<div dir=3D"auto" style=3D"direction: ltr; margin: 0px; padding: 0px; font-=
family: sans-serif; font-size: 11pt; color: black; text-align: left;">
<br>
</div>
<div dir=3D"auto" style=3D"direction: ltr; margin: 0px; padding: 0px; font-=
family: sans-serif; font-size: 11pt; color: black; text-align: left;">
So I think something happened between 7.1.2 and 7.2.</div>
<hr tabindex=3D"-1" style=3D"display: inline-block; width: 98%;" data-ogsc=
=3D"">
<div id=3D"divRplyFwdMsg" dir=3D"ltr" data-ogsc=3D"" style=3D""><font face=
=3D"Calibri, sans-serif" color=3D"#000000" style=3D"font-size: 11pt;" data-=
ogsc=3D"">From: Izumi Tsutsui <tsutsui@ceres.dti.ne.jp><br>
Sent: Thursday, April 23, 2020 4:00:03 PM<br>
To: kern-bug-people@netbsd.org <kern-bug-people@netbsd.org>; g=
nats-admin@netbsd.org <gnats-admin@netbsd.org>; netbsd-bugs@netbsd.or=
g <netbsd-bugs@netbsd.org>; joaopaulo1511@hotmail.com <joaopaulo15=
11@hotmail.com><br>
Subject: Re: kern/55198: When the video card is Intel Mobile 965GM E=
xpress, i915drmkms has error CPU pipe A FIFO underrun, also X freezes/hangs=
</font>
<div> </div>
</div>
<div class=3D"BodyFragment" data-ogsc=3D"" style=3D""><font size=3D"2"><spa=
n style=3D"font-size:11pt">
<div class=3D"PlainText">The following reply was made to PR kern/55198; it =
has been noted by GNATS.<br>
<br>
From: Izumi Tsutsui <tsutsui@ceres.dti.ne.jp><br>
To: gnats-bugs@netbsd.org<br>
Cc: tsutsui@ceres.dti.ne.jp<br>
Subject: Re: kern/55198: When the video card is Intel Mobile 965GM Express,=
<br>
i915drmkms has error CPU p=
ipe A FIFO underrun, also X freezes/hangs<br>
Date: Fri, 24 Apr 2020 04:58:13 +0900<br>
<br>
Maybe something similar as PR/54995 on Atom N455?<br>
<a href=3D"https://gnats.netbsd.org/54995" data-ogsc=3D"" style=3D""=
>https://gnats.netbsd.org/54995</a><br>
<br>
How about adding<br>
> option "AccelMethod" "uxa"<br>
in /etc/X11/xorg.conf ?<br>
<br>
If SNA accelaration is problematic on old Intel chips,<br>
maybe we should fallback to UXA on such pre-2010 chipset?<br>
<br>
---<br>
Izumi Tsutsui<br>
<br>
</div>
</span></font></div>
</body>
</html>
--_000_DM6PR04MB4906B4CC95635AE7F430632CA7D00DM6PR04MB4906namp_--
From: =?iso-8859-1?Q?Jo=E3o_Paulo_Vinha_Bittar?= <joaopaulo1511@hotmail.com>
To: "kern-bug-people@netbsd.org" <kern-bug-people@netbsd.org>,
"gnats-admin@netbsd.org" <gnats-admin@netbsd.org>, "netbsd-bugs@netbsd.org"
<netbsd-bugs@netbsd.org>, "gnats-bugs@netbsd.org" <gnats-bugs@netbsd.org>
Cc:
Subject: RE: kern/55198: When the video card is Intel Mobile 965GM Express,
i915drmkms has error CPU pipe A FIFO underrun, also X freezes/hangs
Date: Tue, 28 Apr 2020 21:01:45 +0000
Hi Izumi Tsutsui,=0A=
=0A=
I could make X work after I added the "AccelMethod" "uxa" line to both of t=
he two sections created on xorg.conf.new by X -configure (It didn't occurre=
d me earlier to do that, sorry).=0A=
=0A=
So I think the easiest fix would make "uxa" the default acceleration method=
on old chips as Intel 965GM.=0A=
=0A=
Could you please in the meantime add a suggestion to fix this error on "htt=
p://netbsd.org/docs/guide/en/chap-x.html#chap-x-configuration" page? I thi=
nk the following text could be helpful to add right after the line "If the =
above test was successful, move the file into place as /etc/X11/xorg.conf a=
nd you are ready to go":=0A=
=0A=
"If the test was not successful and your graphics card is an old Intel mode=
l, edit /etc/X11/xorg.conf, find all lines where "AccelMethod" is commented=
with a starting "#", remove the "#" on the beginning and add "uxa" in the =
end, so the lines will look like: "AccelMethod" "uxa". It is needed to fin=
d all lines because some computers have two video outputs."=0A=
=0A=
But I have to point out this "uxa fix" only helps with X, it doesn't make t=
he error boot messages go away.=0A=
=0A=
Thank you very much,=0A=
=0A=
Jo=E3o Paulo Vinha Bittar=
From: Izumi Tsutsui <tsutsui@ceres.dti.ne.jp>
To: joaopaulo1511@hotmail.com
Cc: gnats-bugs@netbsd.org, tsutsui@ceres.dti.ne.jp
Subject: Re: kern/55198: When the video card is Intel Mobile 965GM Express,i915drmkms
has error CPU pipe A FIFO underrun, also X freezes/hangs
Date: Thu, 30 Apr 2020 03:19:37 +0900
> I could make X work after I added the "AccelMethod" "uxa" line to both of the two sections created on xorg.conf.new by X -configure (It didn't occurred me earlier to do that, sorry).
>
> So I think the easiest fix would make "uxa" the default acceleration method on old chips as Intel 965GM.
Mine is N455 GMA 3150. Maybe all GMA Intel GPUs are affected?
https://en.wikipedia.org/wiki/Intel_GMA#Generation_three_based
GMA 900 (910G, 915G, 915Gx)
GMA 950 (945G)
GMA 3100 (Q33, G35, G31, G33)
GMA 3150 (Atom D4xx, D5xx, N4xx, N5xx)
etc.
("SNA" seems to stand for "Sandybridge's New Acceleration").
https://en.wikipedia.org/wiki/SNA_(computer_graphics)
> Could you please in the meantime add a suggestion to fix this error on "http://netbsd.org/docs/guide/en/chap-x.html#chap-x-configuration" page?
I've surveyed how we can change xorg.conf simpler, and then
it's turned out it's enough to have the following simple
/etc/X11/xorg.conf just to change only the "AccelMethod":
---
Section "Device"
Identifier "Intel Gen3 GPU workaround"
Driver "intel"
Option "AccelMethod" "uxa"
EndSection
---
Maybe we should an independent section about the trouble shooting?
---
Izumi Tsutsui
From: coypu@sdf.org
To: gnats-bugs@netbsd.org
Cc:
Subject: Re: kern/55198: When the video card is Intel Mobile 965GM
Express,i915drmkms has error CPU pipe A FIFO underrun, also X freezes/hangs
Date: Wed, 29 Apr 2020 21:54:22 +0000
in the long run we need to switch to the xf86-video-modesetting driver
(once GLAMOR works), and it also eliminates this issue I think.
From: "John D. Baker" <jdbaker@consolidated.net>
To: gnats-bugs@NetBSD.org
Cc:
Subject: Re: kern/55198: When the video card is Intel Mobile 965GM
Express,i915drmkms has error CPU pipe A FIFO underrun, also X freezes/hangs
Date: Wed, 29 Apr 2020 18:49:34 -0500 (CDT)
There was a thread about this issue in current-users some time back. I
don't recall if another PR was filed about it. In that thread, X didn't
freeze/hang but the display was corrupt/sluggish.
The gist of it is, if your intel video hardware is old enough,
Accelmethod "SNA"
(default) works. If it is new enough, it works, albeit "TearFree" "True"
is required and was made default.
Then there are some in the middle where Accelmethod "SNA" used to work
(INTEL_2014 driver), but no-longer works (INTEL_2019 driver).
Old enough: i82810e (no DRMKMS), i82915, others? (Set "TearFree" "False"
to save performance penalty.)
Used to work but doesn't anymore: i82Q45, i82G41, i82945, i82946, others.
Must force Accelmethod "UXA" (set "TearFree" "False" to save performance
penalty) or use the "modesetting" driver.
New enough: (See other people's messages wherein it works for them with
"TearFree" "True". I don't have any of these, so can't say myself.)
--
|/"\ John D. Baker, KN5UKS NetBSD Darwin/MacOS X
|\ / jdbaker[snail]consolidated[flyspeck]net OpenBSD FreeBSD
| X No HTML/proprietary data in email. BSD just sits there and works!
|/ \ GPGkeyID: D703 4A7E 479F 63F8 D3F4 BD99 9572 8F23 E4AD 1645
From: "John D. Baker" <jdbaker@consolidated.net>
To: gnats-bugs@NetBSD.org
Cc:
Subject: Re: kern/55198: When the video card is Intel Mobile 965GM
Express,i915drmkms has error CPU pipe A FIFO underrun, also X freezes/hangs
Date: Fri, 1 May 2020 00:26:08 -0500 (CDT)
On Wed, 29 Apr 2020, John D. Baker wrote:
> Old enough: i82810e (no DRMKMS), i82915, others? (Set "TearFree" "False"
> to save performance penalty.)
Add the i82845 to this list. It now works very well with SNA and without
"TearFree". (See kern/49707 for previous issues with DRMKMS and intel
Xorg driver on this chip.)
--
|/"\ John D. Baker, KN5UKS NetBSD Darwin/MacOS X
|\ / jdbaker[snail]consolidated[flyspeck]net OpenBSD FreeBSD
| X No HTML/proprietary data in email. BSD just sits there and works!
|/ \ GPGkeyID: D703 4A7E 479F 63F8 D3F4 BD99 9572 8F23 E4AD 1645
From: Izumi Tsutsui <tsutsui@ceres.dti.ne.jp>
To: gnats-bugs@netbsd.org
Cc: netbsd-bugs@netbsd.org, tsutsui@ceres.dti.ne.jp
Subject: Re: kern/55198: When the video card is Intel Mobile 965GM Express,i915drmkmshas
error CPU pipe A FIFO underrun, also X freezes/hangs
Date: Sat, 9 May 2020 00:14:31 +0900
> > So I think the easiest fix would make "uxa" the default acceleration method on old chips as Intel 965GM.
>
> Mine is N455 GMA 3150. Maybe all GMA Intel GPUs are affected?
> https://en.wikipedia.org/wiki/Intel_GMA#Generation_three_based
The attachend diff makes old (Gen4 and prior) chipsets use UXA
by default. Is it worth to apply netbsd-9 for workaround?
---
Index: external/mit/xf86-video-intel/dist/src/intel_module.c
===================================================================
RCS file: /cvsroot/xsrc/external/mit/xf86-video-intel/dist/src/intel_module.c,v
retrieving revision 1.1.1.4
diff -u -p -d -r1.1.1.4 intel_module.c
--- external/mit/xf86-video-intel/dist/src/intel_module.c 20 Mar 2019 06:34:02 -0000 1.1.1.4
+++ external/mit/xf86-video-intel/dist/src/intel_module.c 8 May 2020 15:05:46 -0000
@@ -567,9 +567,8 @@ _xf86findDriver(const char *ident, XF86C
return NULL;
}
-static enum accel_method { NOACCEL, SNA, UXA } get_accel_method(void)
+static enum accel_method { NOACCEL, SNA, UXA } get_accel_method(enum accel_method accel_method)
{
- enum accel_method accel_method = DEFAULT_ACCEL_METHOD;
XF86ConfDevicePtr dev;
if (hosted())
@@ -641,7 +640,12 @@ intel_scrn_create(DriverPtr driver,
#endif
#if KMS
- switch (get_accel_method()) {
+ enum accel_method default_accel_method = DEFAULT_ACCEL_METHOD;
+ /* Force UXA by default for pre-HD chipset to avoid screen corruption */
+ if ((unsigned)((struct intel_device_info *)match_data)->gen < 050)
+ default_accel_method = UXA;
+
+ switch (get_accel_method(default_accel_method)) {
#if USE_SNA
case NOACCEL:
case SNA:
---
Izumi Tsutsui
From: "John D. Baker" <jdbaker@consolidated.net>
To: gnats-bugs@NetBSD.org
Cc: Izumi Tsutsui <tsutsui@ceres.dti.ne.jp>
Subject: Re: kern/55198: When the video card is Intel Mobile 965GM
Express,i915drmkms has error CPU pipe A FIFO underrun, also X freezes/hangs
Date: Fri, 8 May 2020 11:14:44 -0500 (CDT)
Please consider a lower bound on chips defaulting to UXA. As I have posted
in this PR, chips that are "old enough" work with SNA just fine. There
are only some chips "in the middle" that fail with SNA.
Old, works w/SNA: i82810e, i82845, i82915
Not as old, fails w/SNA: i82945, i82946, i82G41, i82Q45
New, works w/SNA, needs TearFree: ???
These are the ones I know about and have tested. I don't know what the
"generation" values are for the working chips. If a lower bound is
implemented, perhaps it can be refined as more data becomes available.
--
|/"\ John D. Baker, KN5UKS NetBSD Darwin/MacOS X
|\ / jdbaker[snail]consolidated[flyspeck]net OpenBSD FreeBSD
| X No HTML/proprietary data in email. BSD just sits there and works!
|/ \ GPGkeyID: D703 4A7E 479F 63F8 D3F4 BD99 9572 8F23 E4AD 1645
From: Izumi Tsutsui <tsutsui@ceres.dti.ne.jp>
To: jdbaker@consolidated.net
Cc: gnats-bugs@NetBSD.org, tsutsui@ceres.dti.ne.jp
Subject: Re: kern/55198: When the video card is Intel Mobile 965GMExpress,i915drmkms
has error CPU pipe A FIFO underrun, also X freezes/hangs
Date: Sat, 9 May 2020 02:47:29 +0900
> Old, works w/SNA: i82810e, i82845, i82915
I don't think i810 uses SNA (even KMS):
---
static const struct intel_device_info intel_i81x_info = {
.gen = 010,
};
static const struct intel_device_info intel_i830_info = {
.gen = 020,
};
static const struct intel_device_info intel_i845_info = {
.gen = 020,
};
:
#if UMS
if ((unsigned)((struct intel_device_info *)match_data)->gen < 020)
return lg_i810_init(scrn);
#endif
---
Is there any visible advantage with SNA on such old chips?
(i.e. is it worth to maintain a version conditinal table?)
> These are the ones I know about and have tested. I don't know what the
> "generation" values are for the working chips.
https://github.com/NetBSD/xsrc/blob/trunk/external/mit/xf86-video-intel/dist/src/intel_module.c
https://en.wikipedia.org/wiki/List_of_Intel_graphics_processing_units
---
Izumi Tsutsui
From: Izumi Tsutsui <tsutsui@ceres.dti.ne.jp>
To: gnats-bugs@NetBSD.org
Cc: jdbaker@consolidated.net, tsutsui@ceres.dti.ne.jp
Subject: Re: kern/55198: When the video card is Intel Mobile 965GMExpress,i915drmkmshas
error CPU pipe A FIFO underrun, also X freezes/hangs
Date: Mon, 22 Jun 2020 03:56:09 +0900
> > These are the ones I know about and have tested. I don't know what the
> > "generation" values are for the working chips.
I've confirmed both Gen2 830M and 855GM works with the default SNA
(at least no screen corruption) on NetBSD/i386 9.0.
Xorg.0.log files are here:
https://gist.github.com/tsutsui/1e17959dc3f9896be47b91a70bdb2c12
https://gist.github.com/tsutsui/3f31b17c09e6c3a71728b7fd7b689030
https://gist.github.com/tsutsui/609f04aab6d36047b13e68722f7ac7f1
> https://github.com/NetBSD/xsrc/blob/trunk/external/mit/xf86-video-intel/dist/src/intel_module.c
> https://en.wikipedia.org/wiki/List_of_Intel_graphics_processing_units
Here is a rough summary SNA status per chipset and generation:
---
Gen1
810 OK (SNA is not used)
Gen2
830M OK
845M OK (has the same .gen value as intel_i830_info)
855GM OK
865G ??
Gen3
915G (GMA 900) OK
945G (GMA 950) NG
G33 (GMA 3100) ??
N455 (GMA 3150) NG (maybe shares the same struct intel_device_info as G33?)
Gen4
Q965 (GMA 3000) ?? (defined as PCI_CHIP_I965_Q in intel_driver.h) (*)
G965 (GMA X3000) ?? (defined as PCI_CHIP_I965_G in intel_driver.h)
G41 (GMA X4500) OK
G45 (GMA X4500HD) OK (shares the same struct intel_device_info as G41?)
Gen5
Ironlake ??
Gen6
SandyBridge ??
(*) Wikipedia says:
https://en.wikipedia.org/wiki/Intel_GMA#Generation_four_based
``The GMA 3000 3D core is very different from the X3000,''
``It (GMA 3000)_is based more directly on the previous generation
GMA 900 and GMA 950 graphics, and belonging to the same "i915"
family with them.''
But no independent struct intel_device_info is defined in
intel_module.c for G965 and Q965??
---
Anyway I'll commit the attached updated change and send a pullup request.
(we can update the list if there is a new report)
---
Index: intel_module.c
===================================================================
RCS file: /cvsroot/xsrc/external/mit/xf86-video-intel/dist/src/intel_module.c,v
retrieving revision 1.1.1.4
diff -u -p -d -r1.1.1.4 intel_module.c
--- intel_module.c 20 Mar 2019 06:34:02 -0000 1.1.1.4
+++ intel_module.c 21 Jun 2020 16:10:25 -0000
@@ -567,9 +567,8 @@ _xf86findDriver(const char *ident, XF86C
return NULL;
}
-static enum accel_method { NOACCEL, SNA, UXA } get_accel_method(void)
+static enum accel_method { NOACCEL, SNA, UXA } get_accel_method(enum accel_method accel_method)
{
- enum accel_method accel_method = DEFAULT_ACCEL_METHOD;
XF86ConfDevicePtr dev;
if (hosted())
@@ -641,7 +640,17 @@ intel_scrn_create(DriverPtr driver,
#endif
#if KMS
- switch (get_accel_method()) {
+ enum accel_method default_accel_method = DEFAULT_ACCEL_METHOD;
+ /*
+ * XXX
+ * Force UXA by default for Gen3 chipsets (except 915) to avoid
+ * screen corruption. (PR/54995, PR/55198)
+ */
+ if ((unsigned)((struct intel_device_info *)match_data)->gen > 030 &&
+ (unsigned)((struct intel_device_info *)match_data)->gen < 050)
+ default_accel_method = UXA;
+
+ switch (get_accel_method(default_accel_method)) {
#if USE_SNA
case NOACCEL:
case SNA:
---
Izumi Tsutsui
From: Izumi Tsutsui <tsutsui@ceres.dti.ne.jp>
To: gnats-bugs@NetBSD.org
Cc: tsutsui@ceres.dti.ne.jp
Subject: Re: kern/55198: When the video card is Intel Mobile 965GMExpress,i915drmkmshaserror
CPU pipe A FIFO underrun, also X freezes/hangs
Date: Tue, 23 Jun 2020 22:13:19 +0900
> Gen4
> Q965 (GMA 3000) ?? (defined as PCI_CHIP_I965_Q in intel_driver.h) (*)
> G965 (GMA X3000) ?? (defined as PCI_CHIP_I965_G in intel_driver.h)
I missed the subject of this PR, 965GM doesn't work on SNA.
(not sure about Q965, but it shares the same struct)
> --- intel_module.c 20 Mar 2019 06:34:02 -0000 1.1.1.4
> +++ intel_module.c 21 Jun 2020 16:10:25 -0000
:
> - switch (get_accel_method()) {
> + enum accel_method default_accel_method = DEFAULT_ACCEL_METHOD;
> + /*
> + * XXX
> + * Force UXA by default for Gen3 chipsets (except 915) to avoid
> + * screen corruption. (PR/54995, PR/55198)
> + */
> + if ((unsigned)((struct intel_device_info *)match_data)->gen > 030 &&
> + (unsigned)((struct intel_device_info *)match_data)->gen < 050)
> + default_accel_method = UXA;
On the other hand G41/G45 work so the upper limit should be "gen < 045".
---
Izumi Tsutsui
From: "John D. Baker" <jdbaker@consolidated.net>
To: gnats-bugs@NetBSD.org
Cc:
Subject: Re: kern/55198: When the video card is Intel Mobile 965GMExpress,i915drmkmshas
error CPU pipe A FIFO underrun, also X freezes/hangs
Date: Tue, 23 Jun 2020 09:56:06 -0500 (CDT)
On Mon, 22 Jun 2020, Izumi Tsutsui wrote:
> Here is a rough summary SNA status per chipset and generation:
> [snip]
> Gen3
> 915G (GMA 900) OK
> 945G (GMA 950) NG
> G33 (GMA 3100) ??
> N455 (GMA 3150) NG (maybe shares the same struct intel_device_info as G33?)
I have a system with:
946GZ (GMA 3000) NG
As noted above, it does not work with SNA.
> Gen4
> Q965 (GMA 3000) ?? (defined as PCI_CHIP_I965_Q in intel_driver.h) (*)
> G965 (GMA X3000) ?? (defined as PCI_CHIP_I965_G in intel_driver.h)
> G41 (GMA X4500) OK
I have a G41-based machine (82G41TY) and it definitely does NOT work
with SNA.
> G45 (GMA X4500HD) OK (shares the same struct intel_device_info as G41?)
>
> Gen5
> Ironlake ??
I think 82Q45 falls into this category and the machine I have with this
video device also does not work with SNA.
> + if ((unsigned)((struct intel_device_info *)match_data)->gen > 030 &&
> + (unsigned)((struct intel_device_info *)match_data)->gen < 050)
> + default_accel_method = UXA;
I've kludged up a similar patch, but set the upper limit at 060.
--
|/"\ John D. Baker, KN5UKS NetBSD Darwin/MacOS X
|\ / jdbaker[snail]consolidated[flyspeck]net OpenBSD FreeBSD
| X No HTML/proprietary data in email. BSD just sits there and works!
|/ \ GPGkeyID: D703 4A7E 479F 63F8 D3F4 BD99 9572 8F23 E4AD 1645
From: Izumi Tsutsui <tsutsui@ceres.dti.ne.jp>
To: gnats-bugs@netbsd.org
Cc: jdbaker@consolidated.net, tsutsui@ceres.dti.ne.jp
Subject: Re: kern/55198: When the video card is Intel Mobile 965GMExpress,i915drmkmshaserror
CPU pipe A FIFO underrun, also X freezes/hangs
Date: Wed, 24 Jun 2020 00:24:26 +0900
> > G41 (GMA X4500) OK
>
> I have a G41-based machine (82G41TY) and it definitely does NOT work
> with SNA.
>
> > G45 (GMA X4500HD) OK (shares the same struct intel_device_info as G41?)
> >
> > Gen5
> > Ironlake ??
>
> I think 82Q45 falls into this category and the machine I have with this
> video device also does not work with SNA.
Ah, I misread your previous message..
> > + if ((unsigned)((struct intel_device_info *)match_data)->gen > 030 &&
> > + (unsigned)((struct intel_device_info *)match_data)->gen < 050)
> > + default_accel_method = UXA;
>
> I've kludged up a similar patch, but set the upper limit at 060.
060 is sandybridge and 050 is ironlake (i7-6xx etc.):
---
static const struct intel_device_info intel_g4x_info = {
.gen = 045,
};
static const struct intel_device_info intel_ironlake_info = {
.gen = 050,
};
static const struct intel_device_info intel_sandybridge_info = {
.gen = 060,
};
---
So maybe the upperlimit "gen < 050" is okay?
---
Izumi Tsutsui
From: matthew green <mrg@eterna.com.au>
To: gnats-bugs@netbsd.org
Cc: kern-bug-people@netbsd.org, gnats-admin@netbsd.org,
netbsd-bugs@netbsd.org, joaopaulo1511@hotmail.com
Subject: re: kern/55198: When the video card is Intel Mobile 965GMExpress,i915drmkmshas error CPU pipe A FIFO underrun, also X freezes/hangs
Date: Wed, 24 Jun 2020 16:30:35 +1000
> Anyway I'll commit the attached updated change and send a pullup request.
> (we can update the list if there is a new report)
sounds great. thanks for researching this thoroughly.
.mrg.
From: "John D. Baker" <jdbaker@consolidated.net>
To: gnats-bugs@netbsd.org
Cc:
Subject: Re: kern/55198: When the video card is Intel Mobile
965GMExpress,i915drmkmshaserror CPU pipe A FIFO underrun, also X freezes/hangs
Date: Wed, 24 Jun 2020 14:32:57 -0500 (CDT)
On Wed, 24 Jun 2020, Izumi Tsutsui wrote:
> So maybe the upperlimit "gen < 050" is okay?
I just re-checked my Q45 system and when using SNA it reports using the
"EagleLake (gen 4.5)" backend, which doesn't work. So, if the octal
encoding accounts for fractional generation values, "gen < 050" should
cover all the cases I can test.
It should then be possible to get a working display on more intel-video
machines w/o needing an "xorg.conf" file (important for live-CD/USB
installs).
Also, the "TearFree" option was defaulted to "True" some time back.
Apparently some devices need that to produce a usable display.
I find the older devices on which SNA works and those later devices
which must revert to UXA work just fine with "TearFree" set to "False"
(and thus save a bit of performance hit). It would be nice to adjust
that default in a similar fashion as the current patch does for
"AccelMethod".
--
|/"\ John D. Baker, KN5UKS NetBSD Darwin/MacOS X
|\ / jdbaker[snail]consolidated[flyspeck]net OpenBSD FreeBSD
| X No HTML/proprietary data in email. BSD just sits there and works!
|/ \ GPGkeyID: D703 4A7E 479F 63F8 D3F4 BD99 9572 8F23 E4AD 1645
From: Izumi Tsutsui <tsutsui@ceres.dti.ne.jp>
To: gnats-bugs@netbsd.org
Cc: tsutsui@ceres.dti.ne.jp
Subject: Re: kern/55198: When the video card is Intel Mobile965GMExpress,i915drmkmshaserror
CPU pipe A FIFO underrun, also X freezes/hangs
Date: Sat, 27 Jun 2020 21:41:55 +0900
There is a report (from isaki@) that SNA has some problem also
on ThinkPad X201s (i7-L620), i.e. Gen5 Ironlake.
I've also confirmed Gen7 (IvyBridge i7-3770) works.
Here is an updated summary:
https://github.com/NetBSD/xsrc/blob/trunk/external/mit/xf86-video-intel/dist/src/intel_module.c
https://en.wikipedia.org/wiki/List_of_Intel_graphics_processing_units
---
Gen1
810 OK (SNA is not used)
Gen2
830M OK (tsutsui@, Panasonic CF-R1)
845M OK (jdbaker@)
855GM OK (tsutsui@, CF-R3; xdm has problems, but it also happens on UXA)
865G ??
Gen3
915G (GMA 900) OK (jdbaker@)
945G (GMA 950) NG (jdbaker@)
G33 (GMA 3100) NG? (there is a blog that says SNA failed on FreeBSD 12.1)
N455 (GMA 3150) NG (tsutsui@, MSI U135DX)
Gen4
Q965 (GMA 3000) ??
G965 (GMA X3000) NG (joaopaulo1511@, Acer Aspire 5720Z in this PR)
G41 (GMA X4500) NG (jdbaker@)
Q45 (GMA X4500HD) NG (jdbaker@)
Gen5
Ironlake NG (isaki@, ThinkPad X201s; xdm corruption, UXA works)
Gen6
SandyBridge ??
Gen7
IvyBridge OK (tsutsui@, DELL Optiplex 9010)
---
I'll commit the attached version RSN and send a pullup request.
---
Index: intel_module.c
===================================================================
RCS file: /cvsroot/xsrc/external/mit/xf86-video-intel/dist/src/intel_module.c,v
retrieving revision 1.1.1.4
diff -u -p -d -r1.1.1.4 intel_module.c
--- intel_module.c 20 Mar 2019 06:34:02 -0000 1.1.1.4
+++ intel_module.c 27 Jun 2020 12:15:05 -0000
@@ -567,9 +567,8 @@ _xf86findDriver(const char *ident, XF86C
return NULL;
}
-static enum accel_method { NOACCEL, SNA, UXA } get_accel_method(void)
+static enum accel_method { NOACCEL, SNA, UXA } get_accel_method(enum accel_method accel_method)
{
- enum accel_method accel_method = DEFAULT_ACCEL_METHOD;
XF86ConfDevicePtr dev;
if (hosted())
@@ -641,7 +640,17 @@ intel_scrn_create(DriverPtr driver,
#endif
#if KMS
- switch (get_accel_method()) {
+ enum accel_method default_accel_method = DEFAULT_ACCEL_METHOD;
+ /*
+ * XXX
+ * Use UXA by default for Gen3/4/5 chipsets (except 915) to avoid
+ * screen corruption etc. (PR/54995, PR/55198)
+ */
+ if ((unsigned)((struct intel_device_info *)match_data)->gen > 030 &&
+ (unsigned)((struct intel_device_info *)match_data)->gen < 060)
+ default_accel_method = UXA;
+
+ switch (get_accel_method(default_accel_method)) {
#if USE_SNA
case NOACCEL:
case SNA:
---
Note I think the options "TearFree" should be checked independently.
---
Izumi Tsutsui
From: "John D. Baker" <jdbaker@consolidated.net>
To: gnats-bugs@netbsd.org
Cc:
Subject: Re: kern/55198: When the video card is Intel Mobile
965GMExpress,i915drmkmshaserror CPU pipe A FIFO underrun, also X freezes/hangs
Date: Sat, 27 Jun 2020 11:22:48 -0500 (CDT)
On Sat, 27 Jun 2020 12:45:02 +0000 (UTC), Izumi Tsutsui
<tsutsui@ceres.dti.ne.jp> wrote:
> Gen3
> 915G (GMA 900) OK (jdbaker@)
> 945G (GMA 950) NG (jdbaker@)
| ^
While I do have a 945GCL system, X is not installed on it so I have
never tested it.
What I did report is:
946GZIS (GMA 3000) NG
| ^
(nine-four-six, not nine-four-five)
--
|/"\ John D. Baker, KN5UKS NetBSD Darwin/MacOS X
|\ / jdbaker[snail]consolidated[flyspeck]net OpenBSD FreeBSD
| X No HTML/proprietary data in email. BSD just sits there and works!
|/ \ GPGkeyID: D703 4A7E 479F 63F8 D3F4 BD99 9572 8F23 E4AD 1645
From: Izumi Tsutsui <tsutsui@ceres.dti.ne.jp>
To: gnats-bugs@netbsd.org
Cc: tsutsui@ceres.dti.ne.jp
Subject: Re: kern/55198: When the video card is Intel Mobile965GMExpress,i915drmkmshaserror
CPU pipe A FIFO underrun, also X freezes/hangs
Date: Sun, 28 Jun 2020 02:05:08 +0900
> > Gen3
> > 915G (GMA 900) OK (jdbaker@)
> > 945G (GMA 950) NG (jdbaker@)
> | ^
> While I do have a 945GCL system, X is not installed on it so I have
> never tested it.
Well i just referred https://mail-index.netbsd.org/netbsd-bugs/2020/05/08/msg067539.html
> What I did report is:
>
> 946GZIS (GMA 3000) NG
It turns out it's part of Gen4 "I965GM"
https://github.com/NetBSD/xsrc/blob/fe8aea9e5e26bc2302ca376e8528700245935fde/external/mit/xf86-video-intel/dist/src/i915_pciids.h#L85
---
Izumi Tsutsui
From: "Izumi Tsutsui" <tsutsui@netbsd.org>
To: gnats-bugs@gnats.NetBSD.org
Cc:
Subject: PR/55198 CVS commit: xsrc/external/mit/xf86-video-intel/dist/src
Date: Sat, 27 Jun 2020 17:09:59 +0000
Module Name: xsrc
Committed By: tsutsui
Date: Sat Jun 27 17:09:59 UTC 2020
Modified Files:
xsrc/external/mit/xf86-video-intel/dist/src: intel_module.c
Log Message:
Use UXA (not SNA) by default on Intel Gen3/4/5 chipsets (except 915).
This works around PR/54995 (screen corruption on N455)
and PR/55198 (X freezes/hangs on 965GM). Ok'ed by mrg@.
Should be pulled up to netbsd-9.
To generate a diff of this commit:
cvs rdiff -u -r1.1.1.4 -r1.2 \
xsrc/external/mit/xf86-video-intel/dist/src/intel_module.c
Please note that diffs are not public domain; they are subject to the
copyright notices on the relevant files.
From: "David H. Gutteridge" <david@gutteridge.ca>
To: gnats-bugs@netbsd.org
Cc:
Subject: RE: kern/55198: When the video card is Intel Mobile 965GM Express,
i915drmkms has error CPU pipe A FIFO underrun, also X freezes/hangs
Date: Sat, 27 Jun 2020 14:18:45 -0400
On Sat, 27 Jun 2020 at 11:22:48 -0500 (CDT), John D. Baker wrote:
> On Sat, 27 Jun 2020 12:45:02 +0000 (UTC), Izumi Tsutsui
> <tsutsui@ceres.dti.ne.jp> wrote:
>
> > Gen3
> > 915G (GMA 900) OK (jdbaker@)
> > 945G (GMA 950) NG (jdbaker@)
> | ^
> While I do have a 945GCL system, X is not installed on it so I have
> never tested it.
>
> What I did report is:
>
> 946GZIS (GMA 3000) NG
> | ^
> (nine-four-six, not nine-four-five)
I have a laptop with a 945GME running i386/9.0_STABLE that I've tested
UXA vs. SNA with. UXA is indeed an improvement for it, e.g., use of UXA
prevents some crashes with Xfce4, fixes window decoration rendering
issues in Fluxbox, etc.
Thanks,
Dave
State-Changed-From-To: open->pending-pullups
State-Changed-By: tsutsui@NetBSD.org
State-Changed-When: Mon, 29 Jun 2020 15:14:04 +0000
State-Changed-Why:
[pullup-9 #975]
From: Izumi Tsutsui <tsutsui@ceres.dti.ne.jp>
To: gnats-bugs@netbsd.org
Cc: david@gutteridge.ca, tsutsui@ceres.dti.ne.jp
Subject: Re: kern/55198: When the video card is Intel Mobile 965GM Express,i915drmkms
has error CPU pipe A FIFO underrun, also X freezes/hangs
Date: Tue, 30 Jun 2020 00:12:04 +0900
> I have a laptop with a 945GME running i386/9.0_STABLE that I've tested
> UXA vs. SNA with. UXA is indeed an improvement for it, e.g., use of UXA
> prevents some crashes with Xfce4, fixes window decoration rendering
> issues in Fluxbox, etc.
Thank you for your confirmation.
I've committed a fix and sent a pullup request for netbsd-9.
---
Izumi Tsutsui
From: "Martin Husemann" <martin@netbsd.org>
To: gnats-bugs@gnats.NetBSD.org
Cc:
Subject: PR/55198 CVS commit: [netbsd-9] xsrc/external/mit/xf86-video-intel/dist/src
Date: Tue, 30 Jun 2020 18:32:48 +0000
Module Name: xsrc
Committed By: martin
Date: Tue Jun 30 18:32:47 UTC 2020
Modified Files:
xsrc/external/mit/xf86-video-intel/dist/src [netbsd-9]: intel_module.c
Log Message:
Pull up following revision(s) (requested by tsutsui in ticket #975):
external/mit/xf86-video-intel/dist/src/intel_module.c: revision 1.2
Use UXA (not SNA) by default on Intel Gen3/4/5 chipsets (except 915).
This works around PR/54995 (screen corruption on N455)
and PR/55198 (X freezes/hangs on 965GM). Ok'ed by mrg@.
Should be pulled up to netbsd-9.
To generate a diff of this commit:
cvs rdiff -u -r1.1.1.4 -r1.1.1.4.2.1 \
xsrc/external/mit/xf86-video-intel/dist/src/intel_module.c
Please note that diffs are not public domain; they are subject to the
copyright notices on the relevant files.
From: David Brownlee <abs@absd.org>
To: gnats-bugs@netbsd.org, tsutsui@netbsd.org
Cc: kern-bug-people@netbsd.org, gnats-admin@netbsd.org, netbsd-bugs@netbsd.org,
joaopaulo1511@hotmail.com
Subject: Re: kern/55198: When the video card is Intel Mobile965GMExpress,i915drmkmshaserror
CPU pipe A FIFO underrun, also X freezes/hangs
Date: Fri, 3 Jul 2020 21:14:24 +0100
Hi, update on this - I'm seeing the same issues (FIFO underrun, X
freezes with SNA, OK with UXA) on a Thinkpad T480, latest netbsd
current
- dmesg shows: Intel UHD Graphics 620 (GT2)
- Xorg.0.log shows: intel(0): SNA initialized with Kabylake (gen9) backend
Thanks
David
State-Changed-From-To: pending-pullups->closed
State-Changed-By: tsutsui@NetBSD.org
State-Changed-When: Sat, 04 Jul 2020 08:26:04 +0000
State-Changed-Why:
Pullup complete.
From: Izumi Tsutsui <tsutsui@ceres.dti.ne.jp>
To: abs@absd.org
Cc: gnats-bugs@netbsd.org, tsutsui@ceres.dti.ne.jp
Subject: Re: kern/55198: When the video card is Intel Mobile965GMExpress,i915drmkmshaserrorCPU
pipe A FIFO underrun, also X freezes/hangs
Date: Sat, 4 Jul 2020 17:23:54 +0900
> Hi, update on this - I'm seeing the same issues (FIFO underrun, X
> freezes with SNA, OK with UXA) on a Thinkpad T480, latest netbsd
> current
> - dmesg shows: Intel UHD Graphics 620 (GT2)
> - Xorg.0.log shows: intel(0): SNA initialized with Kabylake (gen9) backend
For recent chipsets, it would be better to file a new PR
to fix actual problems in SNA, rather than workaround by UXA.
Note FIFO error messages are independent problems, it seems.
(seen on various chipsets)
---
Izumi Tsutsui
>Unformatted:
(Contact us)
$NetBSD: query-full-pr,v 1.46 2020/01/03 16:35:01 leot Exp $
$NetBSD: gnats_config.sh,v 1.9 2014/08/02 14:16:04 spz Exp $
Copyright © 1994-2020
The NetBSD Foundation, Inc. ALL RIGHTS RESERVED.
Home