SlackwareThis Forum is for the discussion of Slackware Linux.
Notices
Welcome to LinuxQuestions.org, a friendly and active Linux Community.
You are currently viewing LQ as a guest. By joining our community you will have the ability to post topics, receive our newsletter, use the advanced search, subscribe to threads and access many other special features. Registration is quick, simple and absolutely free. Join our community today!
Note that registered members see fewer ads, and ContentLink is completely disabled once you log in.
If you have any problems with the registration process or your account login, please contact us. If you need to reset your password, click here.
Having a problem logging in? Please visit this page to clear all LQ-related cookies.
Get a virtual cloud desktop with the Linux distro that you want in less than five minutes with Shells! With over 10 pre-installed distros to choose from, the worry-free installation life is here! Whether you are a digital nomad or just looking for flexibility, Shells can put your Linux machine on the device that you want to use.
Exclusive for LQ members, get up to 45% off per month. Click here for more info.
Mon Apr 18 22:21:58 UTC 2016
x/xf86-video-intel-git_20160417_81029be-x86_64-1.txz: Upgraded.
Removed --with-default-dri=2 --with-default-accel=uxa options
Added patch for modesetting + glamor fallback on gen9+.
Hopefully the latest git, these option changes, and fallback patch will solve
the issues reported on LQ. Please test!
I think "Removed --with-default-dri=2 --with-default-accel=uxa options" is a nice touch.
Not sure about "Added patch for modesetting + glamor fallback on gen9+."
The patch dates back to Mon, 29 Feb 2016 14:43:35 +0100
And this commit, dating back to 2016-04-06 16:24:01 (GMT) adds SNA on gen9.
The commit title is somehow worrisome "sna/gen9: Quick and dirty implementation" and I don't poses gen9 to rise my voice too loud.
4.6.0-rc4 base w xf86-video-intel-git_20160417_81029be and xf86-video-intel-git_20160416_f2a4645 as the only variables.
Hardware is MSI GS60 6QE (MS-16H7), A.M. bios E16H7IMS.111 (intel 74 microcode), i7-6700HQ, HD530 + nvidia NV124/Maxwell. nvidia was switched off in /etc/rc.local on a 15s delay, default runlevel is 4 (xfce).
Of the dozen+ A-B tests, all 81029be runs failed to complete init 4:
Code:
[ 28.117] (EE) No devices detected.
[ 28.117] (EE)
Fatal server error:
[ 28.117] (EE) no screens found(EE)
[ 28.117] (EE)
Code:
(@51sec)
kernel panic not syncing timeout not all cpus entered broadcast exception handler
shutting down cpus w NMI
offset 0x2c000000 from 0xfffffff81000000
Most were hard lockups 5-15 seconds after the console login prompt showed, plus other occasional artifacts such as the early Atheros ethernet probe failing.
Update: After a hard lock where holding the power button for 4+ seconds is needed to recover, if the Atheros probe fails on subsequent boot a 2nd lockup will happen within a minute or two. This does not happen with normal power on or reboot. MSI 'extra' perhaps?
All of the f2a4645 boots completed, including full X initialization. If the kernel was viable thru init 3, logs for dmesg, boot params, and Xorg were saved for each run, so ping if these are useful.
@lazardo Try the latest commit without Pat's provided patch.
Because of "Removed --with-default-dri=2 --with-default-accel=uxa options" everything is probably going to default to SNA now.
Was git_20160416_f2a4645 with SNA working fine for you?
My guess is it was, and now Pat's provided patch disables SNA for you and things break.
EDIT:
Probably what I want to say is that we're better of with SNA than modesetting + glamor.
I don't have sufficient knowledge and might be bullshitting, though.
--
Best regards,
Andrzej Telszewski
Last edited by atelszewski.versades; 04-19-2016 at 06:49 AM.
OK, "Removed --with-default-dri=2 --with-default-accel=uxa options" has the desired effect and now X picks SNA as default acceleration method in my case, DRI stays at 2, though.
Since I was already running f2a4645 and it was working fine, this update hasn't changed much other than the defaulting to SNA.
I'm on gen8, so cannot speak much for you Skylake guys
As of this writing, I'm running and testing git_20160418_562ae1f.
X does not start with the -current 20160417_81029be:
Code:
[ 55.829] (II) Loading sub module "fb"
[ 55.829] (II) LoadModule: "fb"
[ 55.829] (II) Loading /usr/lib64/xorg/modules/libfb.so
[ 55.831] (II) Module fb: vendor="X.Org Foundation"
[ 55.831] compiled for 1.18.3, module version = 1.0.0
[ 55.831] ABI class: X.Org ANSI C Emulation, version 0.4
[ 55.831] (II) UnloadModule: "vesa"
[ 55.831] (II) Unloading vesa
[ 55.831] (==) Depth 24 pixmap format is 32 bpp
[ 55.831] (EE) modeset(0): drmSetMaster failed: Invalid argument
[ 55.831] (EE)
Fatal server error:
[ 55.831] (EE) AddScreen/ScreenInit failed for driver 0
[ 55.831] (EE)
[ 55.831] (EE)
Please consult the The X.Org Foundation support
at http://wiki.x.org
for help.
[ 55.831] (EE) Please also check the log file at "/var/log/Xorg.0.log" for additional information.
[ 55.832] (EE)
[ 55.846] (EE) Server terminated with error (1). Closing log file.
Now I'm running git_20160418_562ae1f with "Removed --with-default-dri=2 --with-default-accel=uxa options". It seems to work fine.
I would help with more tests if needed.
Skylake 6700K HD 530 here.
Edit: the text in the edit box was drawn with glitches when I deleted some letters. I think someone above mentioned such an issue.
Retested w 81029be and stubbed out 20-intel: no X, hard lock.
Slackware 14.2 must succeed for the 99% else we end up with platform-specific branches for server, desktop, laptop and problem children named Skylake. At some point Pat changes his name to Vishnu (eg, 8 arms required to keep up
Retested w 81029be and stubbed out 20-intel: no X, hard lock.
Slackware 14.2 must succeed for the 99% else we end up with platform-specific branches for server, desktop, laptop and problem children named Skylake. At some point Pat changes his name to Vishnu (eg, 8 arms required to keep up
The Intel guys should just get their act together. Skylake has been out for well over 8 months now.
LinuxQuestions.org is looking for people interested in writing
Editorials, Articles, Reviews, and more. If you'd like to contribute
content, let us know.
