View Issue Details

IDProjectCategoryView StatusLast Update
0000710channel: elrepo/el7kmod-fglrxpublic2017-04-26 11:13
Reporterjcools Assigned Towolfy  
PrioritynormalSeveritymajorReproducibilityalways
Status assignedResolutionopen 
Summary0000710: fglrx-x11-drv fails to install on kernel version 3.10.0-514
DescriptionHello,

I Have a AMD Directe CUII HD7970 graphics card. I tried installing the drivers on CentOS Linux release 7.3.1611 (Core) using the yum package manager and get kernel dependencies which are incorrect.

Could this package be rebuilt for this kernel version?
3.10.0-514.2.2.el7.x86_64

Thank you!


Additional InformationLoaded plugins: fastestmirror, langpacks
base | 3.6 kB 00:00
elrepo | 2.9 kB 00:00
epel/x86_64/metalink | 24 kB 00:00
epel | 4.3 kB 00:00
extras | 3.4 kB 00:00
updates | 3.4 kB 00:00
(1/4): epel/x86_64/updateinfo | 706 kB 00:00
(2/4): epel/x86_64/primary_db | 4.5 MB 00:00
(3/4): updates/7/x86_64/primary_db | 1.3 MB 00:00
(4/4): elrepo/primary_db | 393 kB 00:00:07
Loading mirror speeds from cached hostfile
 * base: centos.cu.be
 * elrepo: ftp.nluug.nl
 * epel: epel.mirror.wearetriple.com
 * extras: centos.cu.be
 * updates: centos.cu.be
Resolving Dependencies
--> Running transaction check
---> Package fglrx-x11-drv.x86_64 0:15.12-3.el7.elrepo will be installed
--> Processing Dependency: fglrx-kmod = 15.12-3.el7.elrepo for package: fglrx-x11-drv-15.12-3.el7.elrepo.x86_64
--> Processing Dependency: fglrx-kmod = 15.12-3.el7.elrepo for package: fglrx-x11-drv-15.12-3.el7.elrepo.x86_64
--> Processing Dependency: acpid for package: fglrx-x11-drv-15.12-3.el7.elrepo.x86_64
--> Running transaction check
---> Package acpid.x86_64 0:2.0.19-6.el7 will be installed
---> Package kmod-fglrx.x86_64 0:15.12-3.el7.elrepo will be installed
--> Processing Dependency: kernel(kill_fasync) = 0xdd9d7a7d for package: kmod-fglrx-15.12-3.el7.elrepo.x86_64
--> Processing Dependency: kernel(fasync_helper) = 0x124106ba for package: kmod-fglrx-15.12-3.el7.elrepo.x86_64
--> Processing Dependency: kernel(amd_iommu_unbind_pasid) = 0xc3022564 for package: kmod-fglrx-15.12-3.el7.elrepo.x86_64
--> Processing Dependency: kernel(amd_iommu_set_invalidate_ctx_cb) = 0x904f86f0 for package: kmod-fglrx-15.12-3.el7.elrepo.x86_64
--> Processing Dependency: kernel(amd_iommu_set_invalid_ppr_cb) = 0xe693e19b for package: kmod-fglrx-15.12-3.el7.elrepo.x86_64
--> Processing Dependency: kernel(amd_iommu_init_device) = 0xbf4e6777 for package: kmod-fglrx-15.12-3.el7.elrepo.x86_64
--> Processing Dependency: kernel(amd_iommu_free_device) = 0x90b26027 for package: kmod-fglrx-15.12-3.el7.elrepo.x86_64
--> Processing Dependency: kernel(amd_iommu_enable_device_erratum) = 0x4ea602e8 for package: kmod-fglrx-15.12-3.el7.elrepo.x86_64
--> Processing Dependency: kernel(amd_iommu_device_info) = 0x0eab57ba for package: kmod-fglrx-15.12-3.el7.elrepo.x86_64
--> Processing Dependency: kernel(amd_iommu_bind_pasid) = 0x8bd98934 for package: kmod-fglrx-15.12-3.el7.elrepo.x86_64
--> Finished Dependency Resolution
Error: Package: kmod-fglrx-15.12-3.el7.elrepo.x86_64 (elrepo)
           Requires: kernel(amd_iommu_enable_device_erratum) = 0x4ea602e8
           Installed: kernel-3.10.0-514.el7.x86_64 (@anaconda)
               kernel(amd_iommu_enable_device_erratum) = 0x9f948ff4
           Installed: kernel-3.10.0-514.2.2.el7.x86_64 (@updates)
               kernel(amd_iommu_enable_device_erratum) = 0x9f948ff4
           Available: kernel-debug-3.10.0-514.el7.x86_64 (base)
               kernel(amd_iommu_enable_device_erratum) = 0xb46b9634
           Available: kernel-debug-3.10.0-514.2.2.el7.x86_64 (updates)
               kernel(amd_iommu_enable_device_erratum) = 0xb46b9634
Error: Package: kmod-fglrx-15.12-3.el7.elrepo.x86_64 (elrepo)
           Requires: kernel(amd_iommu_init_device) = 0xbf4e6777
           Installed: kernel-3.10.0-514.el7.x86_64 (@anaconda)
               kernel(amd_iommu_init_device) = 0x57cbdc50
           Installed: kernel-3.10.0-514.2.2.el7.x86_64 (@updates)
               kernel(amd_iommu_init_device) = 0x57cbdc50
           Available: kernel-debug-3.10.0-514.el7.x86_64 (base)
               kernel(amd_iommu_init_device) = 0x9c5faa83
           Available: kernel-debug-3.10.0-514.2.2.el7.x86_64 (updates)
               kernel(amd_iommu_init_device) = 0x9c5faa83
Error: Package: kmod-fglrx-15.12-3.el7.elrepo.x86_64 (elrepo)
           Requires: kernel(amd_iommu_set_invalid_ppr_cb) = 0xe693e19b
           Installed: kernel-3.10.0-514.el7.x86_64 (@anaconda)
               kernel(amd_iommu_set_invalid_ppr_cb) = 0x6d4aeaee
           Installed: kernel-3.10.0-514.2.2.el7.x86_64 (@updates)
               kernel(amd_iommu_set_invalid_ppr_cb) = 0x6d4aeaee
           Available: kernel-debug-3.10.0-514.el7.x86_64 (base)
               kernel(amd_iommu_set_invalid_ppr_cb) = 0xcfa4eac3
           Available: kernel-debug-3.10.0-514.2.2.el7.x86_64 (updates)
               kernel(amd_iommu_set_invalid_ppr_cb) = 0xcfa4eac3
Error: Package: kmod-fglrx-15.12-3.el7.elrepo.x86_64 (elrepo)
           Requires: kernel(amd_iommu_device_info) = 0x0eab57ba
           Installed: kernel-3.10.0-514.el7.x86_64 (@anaconda)
               kernel(amd_iommu_device_info) = 0xc41feff7
           Installed: kernel-3.10.0-514.2.2.el7.x86_64 (@updates)
               kernel(amd_iommu_device_info) = 0xc41feff7
           Available: kernel-debug-3.10.0-514.el7.x86_64 (base)
               kernel(amd_iommu_device_info) = 0x45cae3f2
           Available: kernel-debug-3.10.0-514.2.2.el7.x86_64 (updates)
               kernel(amd_iommu_device_info) = 0x45cae3f2
Error: Package: kmod-fglrx-15.12-3.el7.elrepo.x86_64 (elrepo)
           Requires: kernel(amd_iommu_unbind_pasid) = 0xc3022564
           Installed: kernel-3.10.0-514.el7.x86_64 (@anaconda)
               kernel(amd_iommu_unbind_pasid) = 0x621fab5b
           Installed: kernel-3.10.0-514.2.2.el7.x86_64 (@updates)
               kernel(amd_iommu_unbind_pasid) = 0x621fab5b
           Available: kernel-debug-3.10.0-514.el7.x86_64 (base)
               kernel(amd_iommu_unbind_pasid) = 0x55f4368b
           Available: kernel-debug-3.10.0-514.2.2.el7.x86_64 (updates)
               kernel(amd_iommu_unbind_pasid) = 0x55f4368b
Error: Package: kmod-fglrx-15.12-3.el7.elrepo.x86_64 (elrepo)
           Requires: kernel(amd_iommu_bind_pasid) = 0x8bd98934
           Installed: kernel-3.10.0-514.el7.x86_64 (@anaconda)
               kernel(amd_iommu_bind_pasid) = 0xe93fbffa
           Installed: kernel-3.10.0-514.2.2.el7.x86_64 (@updates)
               kernel(amd_iommu_bind_pasid) = 0xe93fbffa
           Available: kernel-debug-3.10.0-514.el7.x86_64 (base)
               kernel(amd_iommu_bind_pasid) = 0x99e05e3c
           Available: kernel-debug-3.10.0-514.2.2.el7.x86_64 (updates)
               kernel(amd_iommu_bind_pasid) = 0x99e05e3c
Error: Package: kmod-fglrx-15.12-3.el7.elrepo.x86_64 (elrepo)
           Requires: kernel(fasync_helper) = 0x124106ba
           Installed: kernel-3.10.0-514.el7.x86_64 (@anaconda)
               kernel(fasync_helper) = 0xc682ddc7
           Installed: kernel-3.10.0-514.2.2.el7.x86_64 (@updates)
               kernel(fasync_helper) = 0xc682ddc7
           Available: kernel-debug-3.10.0-514.el7.x86_64 (base)
               kernel(fasync_helper) = 0x906e1140
           Available: kernel-debug-3.10.0-514.2.2.el7.x86_64 (updates)
               kernel(fasync_helper) = 0x906e1140
Error: Package: kmod-fglrx-15.12-3.el7.elrepo.x86_64 (elrepo)
           Requires: kernel(amd_iommu_free_device) = 0x90b26027
           Installed: kernel-3.10.0-514.el7.x86_64 (@anaconda)
               kernel(amd_iommu_free_device) = 0xfea37b21
           Installed: kernel-3.10.0-514.2.2.el7.x86_64 (@updates)
               kernel(amd_iommu_free_device) = 0xfea37b21
           Available: kernel-debug-3.10.0-514.el7.x86_64 (base)
               kernel(amd_iommu_free_device) = 0xa7d33d6a
           Available: kernel-debug-3.10.0-514.2.2.el7.x86_64 (updates)
               kernel(amd_iommu_free_device) = 0xa7d33d6a
Error: Package: kmod-fglrx-15.12-3.el7.elrepo.x86_64 (elrepo)
           Requires: kernel(kill_fasync) = 0xdd9d7a7d
           Installed: kernel-3.10.0-514.el7.x86_64 (@anaconda)
               kernel(kill_fasync) = 0xfbb6c6eb
           Installed: kernel-3.10.0-514.2.2.el7.x86_64 (@updates)
               kernel(kill_fasync) = 0xfbb6c6eb
           Available: kernel-debug-3.10.0-514.el7.x86_64 (base)
               kernel(kill_fasync) = 0x9636806e
           Available: kernel-debug-3.10.0-514.2.2.el7.x86_64 (updates)
               kernel(kill_fasync) = 0x9636806e
Error: Package: kmod-fglrx-15.12-3.el7.elrepo.x86_64 (elrepo)
           Requires: kernel(amd_iommu_set_invalidate_ctx_cb) = 0x904f86f0
           Installed: kernel-3.10.0-514.el7.x86_64 (@anaconda)
               kernel(amd_iommu_set_invalidate_ctx_cb) = 0x2080cb57
           Installed: kernel-3.10.0-514.2.2.el7.x86_64 (@updates)
               kernel(amd_iommu_set_invalidate_ctx_cb) = 0x2080cb57
           Available: kernel-debug-3.10.0-514.el7.x86_64 (base)
               kernel(amd_iommu_set_invalidate_ctx_cb) = 0x87d12216
           Available: kernel-debug-3.10.0-514.2.2.el7.x86_64 (updates)
               kernel(amd_iommu_set_invalidate_ctx_cb) = 0x87d12216
 You could try using --skip-broken to work around the problem
** Found 3 pre-existing rpmdb problem(s), 'yum check' output follows:
ipa-client-4.4.0-14.el7.centos.1.1.x86_64 has installed conflicts freeipa-client: ipa-client-4.4.0-14.el7.centos.1.1.x86_64
ipa-client-common-4.4.0-14.el7.centos.1.1.noarch has installed conflicts freeipa-client-common: ipa-client-common-4.4.0-14.el7.centos.1.1.noarch
ipa-common-4.4.0-14.el7.centos.1.1.noarch has installed conflicts freeipa-common: ipa-common-4.4.0-14.el7.centos.1.1.noarch
TagsNo tags attached.
Attached Files
yum_output.txt (613 bytes)   
Scriptlet output:
   1 Working. This may take some time ...
   2 depmod: ERROR: fstatat(4, amdgpu.ko): No such file or directory
   3 depmod: ERROR: fstatat(4, amdttm.ko): No such file or directory
   4 depmod: ERROR: fstatat(4, amdkcl.ko): No such file or directory
   5 depmod: ERROR: fstatat(4, amdgpu.ko): No such file or directory
   6 depmod: ERROR: fstatat(4, amdttm.ko): No such file or directory
   7 depmod: ERROR: fstatat(4, amdkcl.ko): No such file or directory
   8 Done.
   9 Uninitialised file found, configuring.
  10 Using /etc/X11/xorg.conf
  11 Saving back-up to /etc/X11/xorg.conf.original-1

yum_output.txt (613 bytes)   
Reported upstream

Activities

wolfy

2017-01-11 07:09

developer   ~0005034

It's a bit more complicated, the package I built for EL7 still use an application ( ati-config-display) which relies on a library that was removed from the distro. I must replace that application probably relying on aticonfig instead but $NewYearHoliday and $WORK kind of delayed that

Meanwhile you can use http://wdl.lug.ro/fglrx/ which were rebuilt for the current EL7 kernel but were not pushed to the official repos because of the above mentioned issue.

jcools

2017-01-11 14:21

reporter   ~0005041

Thank you very much! That helped me out!

nugai

2017-01-19 13:35

reporter   ~0005043

FYI, as reported in the Centos support forum at https://www.centos.org/forums/viewtopic.php?f=49&t=60735&p=257194#p257194 and https://www.centos.org/forums/viewtopic.php?f=49&t=59089&p=257495#p257495, the 15.12-5 drivers seem to be working well for me under kernel-3.10.0-514.2.2 (installed last week) and kernel-3.10.0-514.6.1 (installed today).

I suggest pushing that driver forward into elrepo-testing as soon as possible.

Thank you wolfy!!!

wolfy

2017-03-01 21:16

developer   ~0005080

Just as a short update
- the packages which were tested and reported functional in EL 7.3 are now available from my personal site, at http://wdl.lug.ro/fglrx/stable
- there is a newer set of packages at http://wdl.lug.ro/fglrx/testing which no longer rely on ati-config-display ( a Python script that was added by me and which does not work in EL 7.3 due to the fact that RHEL no longer ships a library this script depended on ). They do work but with a caveat: for a reason that I was not yet able to identify, X can be started as startx but not automatic as graphical login from multi-user.target. I am still debugging this issue which puzzles me, as X is clearly happy, only the DE is not.

nugai

2017-03-01 23:15

reporter   ~0005081

Thanks, wolfy!

FYI, using a manual rpm -Uvh install, your 15.12-6 (x86_64) testing drivers ran without problems and all screens started correctly on the first try on my dual HD 6450 4-monitor system (using latest 3.10.0-514.6.2.el7.x86_64 kernel).

As a rule of thumb, I have my SElinux environment set to permissive. Given the multi-user.target problem you describe, have you checked your SElinux settings and/or policies (and/or disabled SElinux for testing)?

nugai

2017-03-01 23:20

reporter   ~0005082

for clarification, I meant: "all screens auto-started correctly"

wolfy

2017-03-02 06:40

developer   ~0005083

I just switched selinux to permissive and disabled and tested .. without luck. I have a strong feeling I might be missing some package but OTOH the system is happy with the default drivers...

nugai

2017-03-03 15:25

reporter   ~0005088

...and another clarification: For about 2 years now, I have been using Xfce as Desktop because I ran into problems with Gnome (child windows 3-5 levels inside Eclipse being blank). So, your notion that the DE could be the culprit may well be correct.

yaekee

2017-03-13 01:56

reporter   ~0005092

I use AMD radeon HD 7570.
I install a new CentOS 7.2 and update to 7.3 with "yum update -y"
Then I install http://wdl.lug.ro/fglrx/testing/kmod-fglrx-15.12-6.el7.elrepo.x86_64.rpm and http://wdl.lug.ro/fglrx/testing/fglrx-x11-drv-15.12-6.el7.elrepo.x86_64.rpm.

The XWindows fail to start,I find the following in /var/log/Xorg.0.log
(EE) AIGLX error: failed to open /usr/X11R6/lib64/modules/dri/fglrx_dri.so, error[/usr/X11R6/lib64/modules/dri/fglrx_dri.so: cannot open shared object file: No such file or directory]

I check /etc/default/grub, I find the following
GRUB_CMDLINE_LINUX="rd.lvm.lv=rootvg/root nomodeset rhgb quiet radeon.modeset=0 rd.driver.blacklist=radeon"

I know there's a similar bug before.

wolfy

2017-03-13 02:13

developer   ~0005093

@yaekee: The problem you see is exactly the reason why I have not pushed the packages to the elrepo repository. You can start the graphical environment by using startx from a text console ( at least this works on my test system ). However I was not yet able to allocate enough time to investigate the reason for the regular GUI not being able to start.
For the time being please use the set of packages from the /stable directory ( use force to install them if needed )

Hint: I welcome any help !

yaekee

2017-03-13 04:26

reporter   ~0005094

I also find another warning when I remove the testing rpms:
warning: file /etc/ld.so.conf.d/ati.conf.disable: remove failed: No such file or directory

wolfy

2017-03-13 07:51

developer   ~0005095

That's a benign error perpetuated ever since the package was created.

hunter86_bg

2017-04-18 12:50

reporter   ~0005122

I have a setup of HD7770 and R7 250X. If you need some testing, do not hesitate to ask.

wolfy

2017-04-19 00:51

developer   ~0005124

@hunter86_bg: your help would be awesome because I am once again out of hardware suitable for testing purposes ( we use only CentOS 6 here ). Kindly please do the following two tests
a)
- install a fresh RHEL or CentOS 7.3 system with either KDE ( this is what I normally use ) or Gnome ( RH's default DM ) and make sure it works as it should in GUI mode
- put aside the Xorg log, it might be needed later for debugging purposes
- go to http://wdl.lug.ro/fglrx/testing/ and install the kmod and fglrx-x11-drv-15.12-6.el7.elrepo.x86_64.rpm. They depend on each other so you need to install them in a single transaction.
- reboot
- the system will start "almost" normally but will probably refuse to go into graphical mode. put aside the Xorg log if there is a new one and ( from console 2 as probably as #1 will be tied by the failed X server ) try to use startx ( as "startx -- :1" ) -- on my system this worked as it should.
- attach here the 3 Xorg.log files ( initial i.e. prior to installing my packages, after install and failed and the one created after running startx )
- put aside the file /etc/X11/xorg.conf that was created

b) repeat the above but using the packages from http://wdl.lug.ro/fglrx/stable. These should work by default but will trigger an error regarding a missing python module.
Please attach the file /etc/X11/xorg.conf that was created when you installed the drivers as well as the one preserved since installing release 6 of the packages.


As for explanations: I am shooting in the dark trying to find out why the systems used to work when xorg.conf was generated by a script which is not really usable any more but same systems fail if the config is generated by ATI's own tool.

hunter86_bg

2017-04-19 01:08

reporter   ~0005125

I've already done that (without the fresh install) using Gnome , IOMMU enabled from bios and latest kernel and I can tell you that neither the testing nor the stable one work for me.
I've noticed that during the "yum install" it complains alot.Stable - looking for some kernels , while the testing is provided in the attachment.
 The testing packages lead to a strange situation where I have a mouse and small colourful dots on the screen. It's obvious that I somehow managed to login , but nothing is working.
At the end of the week, when I will be able to clean it up and start over as I can't find the Xorg log with the kernel parameters passed with the packages(something modeset=0 ,etc).

wolfy

2017-04-19 01:25

developer   ~0005126

In terms of kernel requirements, there is absolutely no difference between the two sets of packages. The only difference between release 5 and 6 is the way the xorg.conf file is generated - release 5 uses a script I inherited and reused since the RHEL 6 era while with release 6 I attempted to use ATI's own mechanism ( which was introduced well after the usage of previously mentioned script was needed )

The yum output you provided looks like the one generated by ATI's tool.

Normally you can find X logs as /var/log/Xorg.*

hunter86_bg

2017-04-19 01:51

reporter   ~0005127

I will make a clean install this weekend (as I tried once to install AMD's drivers from their website) and then I will report back what happened.

hunter86_bg

2017-04-24 10:48

reporter   ~0005144

@Wolfy,

it seems that there is some issue with the HD7770 and R7 250X in RHEL7.3 (fresh install)
I have cleaned up the Xorg logs, rebooted , installed the testing , rebooted , tried "startx -- :1" , reboot ,removed testing, reboot , install stable , reboot and remove stable too.

Both testing and stable cause a strange behaviour where only a mouse pointer is available.I managed to login but no change - pure black screen. I know I have managed to login as I could do:
1. Alt +F2
2. enter "gnome-terminal"
3. enter "su -"
4. reboot

wolfy

2017-04-25 02:14

developer   ~0005148

All I can suggest is to try the genuine AMD installer. If the issue persists ( which should happen since the difference between my packages and the AMD bundle is just in packaging ) I suspect that either these drivers are not suitable for your system or you need to tweak either the xorg.conf file or /etc/ati/amdpcsdb

Anyway, before that I am curious to see the content of Xorg.log after installing any of my packages, even if you get just a black screen

hunter86_bg

2017-04-25 02:23

reporter   ~0005149

I've already attached them in "testing_hd7770_r7250x.tgz" with the output of the yum during install/removal.

wolfy

2017-04-25 03:41

developer   ~0005150

Sorry, I did not notice the attachment.

Something is a bit odd in your logs because both for the stable and testing packages, Xorg.0.log.old and Xorg.1.* show that the binary drivers are/were installed and used but Xorg.0.log claims that the currently running drivers are the ones from the distribution. Should I assume that you actually retrieved the logs after removing my packages ?

Anyway, Xorg.0.log.old and Xorg.1.* indicate that the proprietary drivers were correctly installed and used so I am afraid there is nothing more that I can so short of suggesting to look in ATI's forums and/or asking them for help. I very strongly suspect one of the their private settings stored in /etc/ati/amdpcsdb would solve the problem you encounter.

nugai

2017-04-25 04:00

reporter   ~0005151

Ah, I didn't notice the uploaded Xorg.log files either.

Taking a brief look at some of the files, I see 2 major differences to my working system, related to ModulePath and whether or not fglrx is loaded:

hunter86_bg's system shows:
[ 20.892] (==) ModulePath set to "/usr/lib64/xorg/modules"

...while my system shows:
[ 14.328] (**) ModulePath set to "/usr/lib64/xorg/modules/extensions/fglrx,/usr/lib64/xorg/modules"

...consequently, hunter86_bg's system shows:
[ 20.896] (II) Loading /usr/lib64/xorg/modules/extensions/libglx.so
[ 20.901] (II) Module glx: vendor="X.Org Foundation"
[ 20.901] compiled for 1.17.2, module version = 1.0.0
[ 20.901] ABI class: X.Org Server Extension, version 9.0

...while my system shows:
[ 14.336] (II) Loading /usr/lib64/xorg/modules/extensions/fglrx/libglx.so
[ 14.337] (II) Module glx: vendor="Advanced Micro Devices, Inc."
[ 14.337] compiled for 6.9.0, module version = 1.0.0
[ 14.337] (II) LoadModule: "fglrx"
[ 14.340] (II) Loading /usr/lib64/xorg/modules/drivers/fglrx_drv.so
[ 14.508] (II) Module fglrx: vendor="FireGL - AMD Technologies Inc."
[ 14.508] compiled for 1.4.99.906, module version = 15.30.3
[ 14.508] Module class: X.Org Video Driver
[ 14.512] (II) Loading sub module "fglrxdrm"
[ 14.512] (II) LoadModule: "fglrxdrm"
[ 14.512] (II) Loading /usr/lib64/xorg/modules/linux/libfglrxdrm.so
[ 14.513] (II) Module fglrxdrm: vendor="FireGL - AMD Technologies Inc."
[ 14.513] compiled for 1.4.99.906, module version = 15.30.3
[ 14.513] (II) AMD Proprietary Linux Driver Version Identifier:15.30.3
[ 14.513] (II) AMD Proprietary Linux Driver Release Identifier: UNSUPPORTED-15.302
[ 14.513] (II) AMD Proprietary Linux Driver Build Date: Dec 17 2015 02:43:16

I hope this helps somewhat.

hunter86_bg

2017-04-25 04:06

reporter   ~0005153

"Should I assume that you actually retrieved the logs after removing my packages ?" - yes.I have just separated them , for convinience - but they do duplicate.

@wolfy, Is there a way to enable higher level of debugging ?

I will check what is the situation with the "/etc/ati/amdpcsdb" will try to find the reason for this behaviour.

wolfy

2017-04-25 04:26

developer   ~0005154

@nugai: compare your logs with either hunter's Xorg.0.log.old or with Xorg.1.log*. Xorg.0.log was created after removing the binary drivers therefore they show the behaviour after using the stock radeon driver.


@hunter86_bg: As far as I know, no, but I might be wrong.

nugai

2017-04-25 12:16

reporter   ~0005156

@wolfy: Thanks for the hint. Maybe I shouldn't try looking at log files at 3:00 AM. ;-)

Here are my observations of hunter86_bg's Xorg.1.log files for "stable" and "testing" drivers:

1. Both stable and testing seem to be failing in exactly the same way. The Xorg.1.log files show the following entries (line numbers correspond to after_testing log file) that are not present in my Xorg.0.log file (running 2x HD 6450 with 4x 1920x1080 monitors).

Line 723: [ 110.156] (WW) fglrx(0): os shared memory open failed
...
Line 812: [ 162.670] (WW) xf86CloseConsole: KDSETMODE failed: Input/output error
Line 813: [ 162.670] (WW) xf86CloseConsole: VT_GETMODE failed: Input/output error
Line 814: [ 162.670] (EE)
Line 815: Fatal server error:
Line 816: [ 162.670] (EE) xf86CloseConsole: VT_ACTIVATE failed: Input/output error
Line 817: [ 162.670] (EE)
Line 818: [ 162.670] (EE)

2. It appears the difference between stable and testing is that testing adds the following 3 lines to the log file:

Line 796: [ 133.576] (II) AIGLX: Suspending AIGLX clients for VT switch
Line 797: [ 133.576] (II) fglrx(0): Backup framebuffer data.
Line 798: [ 133.727] (II) fglrx(0): Backup complete.

None of these lines show up in my Xorg.0.log file.

wolfy

2017-04-26 01:41

developer   ~0005157

From my point of view, the only difference between stable and testing ( -- which you can verify by running rpm -q --scripts over the two ) is who generates the xorg.conf file at install time ( /usr/sbin/ati-config-display for stable, AMD's own aticonfig for the testing )

nugai

2017-04-26 11:13

reporter   ~0005158

OK, so maybe we should see xorg.conf files as well?
...and the output of "dmesg | grep fglrx" when X server fails?

Issue History

Date Modified Username Field Change
2017-01-11 03:40 jcools New Issue
2017-01-11 03:40 jcools Status new => assigned
2017-01-11 03:40 jcools Assigned To => stindall
2017-01-11 07:02 toracat Assigned To stindall => wolfy
2017-01-11 07:09 wolfy Note Added: 0005034
2017-01-11 14:21 jcools Note Added: 0005041
2017-01-19 13:35 nugai Note Added: 0005043
2017-03-01 21:16 wolfy Note Added: 0005080
2017-03-01 23:15 nugai Note Added: 0005081
2017-03-01 23:20 nugai Note Added: 0005082
2017-03-02 06:40 wolfy Note Added: 0005083
2017-03-03 15:25 nugai Note Added: 0005088
2017-03-13 01:56 yaekee Note Added: 0005092
2017-03-13 02:13 wolfy Note Added: 0005093
2017-03-13 04:26 yaekee Note Added: 0005094
2017-03-13 07:51 wolfy Note Added: 0005095
2017-03-13 15:08 burakkucat Project elrepo bug tracker => channel: elrepo/el7
2017-03-13 15:09 burakkucat Category --bug-tracker--request-for-enhancement-- => kmod-fglrx
2017-04-18 12:50 hunter86_bg Note Added: 0005122
2017-04-19 00:51 wolfy Note Added: 0005124
2017-04-19 01:08 hunter86_bg Note Added: 0005125
2017-04-19 01:08 hunter86_bg File Added: yum_output.txt
2017-04-19 01:25 wolfy Note Added: 0005126
2017-04-19 01:51 hunter86_bg Note Added: 0005127
2017-04-24 10:48 hunter86_bg Note Added: 0005144
2017-04-24 10:48 hunter86_bg File Added: testing_hd7770_r7250x.tgz
2017-04-25 02:14 wolfy Note Added: 0005148
2017-04-25 02:23 hunter86_bg Note Added: 0005149
2017-04-25 03:41 wolfy Note Added: 0005150
2017-04-25 04:00 nugai Note Added: 0005151
2017-04-25 04:06 hunter86_bg Note Added: 0005153
2017-04-25 04:26 wolfy Note Added: 0005154
2017-04-25 12:16 nugai Note Added: 0005156
2017-04-26 01:41 wolfy Note Added: 0005157
2017-04-26 11:13 nugai Note Added: 0005158