-
Posts: 88
- Joined: 25 Aug 2012
-
Posts: 2,238
- Joined: 16 Dec 2007
#32
oh yeah... is there a new amd driver to install in stretch?
kmathern wrote:fglrx was removed from Stretch a year or so ago because it doesn't support Xorg versions 1.18 and higher (Stretch is using Xorg 1.19.2), so there's nothing to install.anticapitalista wrote:see a couple of posts above yourseugen-b wrote:But clicking AMD/ATI flgrx installer in CC doesn't open any window.
oh yeah... is there a new amd driver to install in stretch?
-
Posts: 88
- Joined: 25 Aug 2012
#33
There's the amdgpu driver, but that only works with newer AMD/ATI gpus.dolphin_oracle wrote:oh yeah... is there a new amd driver to install in stretch?
-
Posts: 2,238
- Joined: 16 Dec 2007
#34
I think there is an xorg package for that already.
========= SCRAPER REMOVED AN EMBEDDED LINK HERE ===========
url was:"https://packages.debian.org/stretch/xserver-xorg-video-amdgpu"
linktext was:"https://packages.debian.org/stretch/xse ... deo-amdgpu"
====================================
so I suppose nothing to do for amd proprietary drivers in stretch, right. then we can drop that function of ddm-mx.
kmathern wrote:There's the amdgpu driver, but that only works with newer AMD/ATI gpus.dolphin_oracle wrote:oh yeah... is there a new amd driver to install in stretch?
I think there is an xorg package for that already.
========= SCRAPER REMOVED AN EMBEDDED LINK HERE ===========
url was:"https://packages.debian.org/stretch/xserver-xorg-video-amdgpu"
linktext was:"https://packages.debian.org/stretch/xse ... deo-amdgpu"
====================================
so I suppose nothing to do for amd proprietary drivers in stretch, right. then we can drop that function of ddm-mx.
-
anticapitalista
Posts: 5,955
- Site Admin
- Joined: 11 Sep 2007
#35
Glad we sorted that out in alpha that's why we do this to check - good work kent.
-
Posts: 307
- Joined: 23 Aug 2015
#36
========= SCRAPER REMOVED AN EMBEDDED LINK HERE ===========
url was:"https://github.com/Tids/mhwd-addon-amdgfx-hwe"
linktext was:"https://github.com/Tids/mhwd-addon-amdgfx-hwe"
====================================
The driver video-amdgpu-experimental+ works on my GPU.
Here is a pretty good overview about AMD drivers for"older/newer" hardware (GCN generations)kmathern wrote:There's the amdgpu driver, but that only works with newer AMD/ATI gpus.dolphin_oracle wrote:oh yeah... is there a new amd driver to install in stretch?
========= SCRAPER REMOVED AN EMBEDDED LINK HERE ===========
url was:"https://github.com/Tids/mhwd-addon-amdgfx-hwe"
linktext was:"https://github.com/Tids/mhwd-addon-amdgfx-hwe"
====================================
The driver video-amdgpu-experimental+ works on my GPU.
-
Posts: 88
- Joined: 25 Aug 2012
#37
I haven't really tried antiX-17-a2 yet, on my MX-16 install I can get it to use the amdgpu driver if I install xserver-xorg-video-amdgpu and add the following to the kernel optionseugen-b wrote:Here is a pretty good overview about AMD drivers for"older/newer" hardware (GCN generations)
========= SCRAPER REMOVED AN EMBEDDED LINK HERE ===========
url was:"https://github.com/Tids/mhwd-addon-amdgfx-hwe"
linktext was:"https://github.com/Tids/mhwd-addon-amdgfx-hwe"
====================================
The driver video-amdgpu-experimental+ works on my GPU.
$ cat /proc/cmdline
BOOT_IMAGE=/boot/vmlinuz-4.10.1-antix.1-amd64-smp root=UUID=ab940fe4-7c83-45ad-852c-8addf5ebee5a ro zswap.zpool=zsmalloc BOOT_IMAGE=/antiX/vmlinuz quiet radeon.modeset=0
Code: Select all
$ inxi -Gxxx
Graphics: Card: Advanced Micro Devices [AMD/ATI] Kabini [Radeon HD 8240 / R3 Series]
bus-ID: 00:01.0 chip-ID: 1002:9838
Display Server: X.Org 1.16.4 drivers: ati,amdgpu (unloaded: modesetting,fbdev,vesa,radeon)
Resolution: 1600x900@60.00hz
GLX Renderer: Gallium 0.4 on AMD KABINI (DRM 3.9.0 / 4.10.1-antix.1-amd64-smp, LLVM 3.9.1)
GLX Version: 3.0 Mesa 13.0.5 Direct Rendering: Yes
-
Posts: 307
- Joined: 23 Aug 2015
#38
And we happen to have very similar hardware.
Maybe Tids' configuration does s.th. similar. He writes that those are not different amdgpu drivers but differen install scripts.kmathern wrote:I haven't really tried antiX-17-a2 yet, on my MX-16 install I can get it to use the amdgpu driver if I install xserver-xorg-video-amdgpu and add the following to the kernel optionseugen-b wrote:Here is a pretty good overview about AMD drivers for"older/newer" hardware (GCN generations)
========= SCRAPER REMOVED AN EMBEDDED LINK HERE ===========
url was:"https://github.com/Tids/mhwd-addon-amdgfx-hwe"
linktext was:"https://github.com/Tids/mhwd-addon-amdgfx-hwe"
====================================
The driver video-amdgpu-experimental+ works on my GPU.$ cat /proc/cmdline
BOOT_IMAGE=/boot/vmlinuz-4.10.1-antix.1-amd64-smp root=UUID=ab940fe4-7c83-45ad-852c-8addf5ebee5a ro zswap.zpool=zsmalloc BOOT_IMAGE=/antiX/vmlinuz quiet radeon.modeset=0Code: Select all
$ inxi -Gxxx Graphics: Card: Advanced Micro Devices [AMD/ATI] Kabini [Radeon HD 8240 / R3 Series] bus-ID: 00:01.0 chip-ID: 1002:9838 Display Server: X.Org 1.16.4 drivers: ati,amdgpu (unloaded: modesetting,fbdev,vesa,radeon) Resolution: 1600x900@60.00hz GLX Renderer: Gallium 0.4 on AMD KABINI (DRM 3.9.0 / 4.10.1-antix.1-amd64-smp, LLVM 3.9.1) GLX Version: 3.0 Mesa 13.0.5 Direct Rendering: Yes
And we happen to have very similar hardware.
Code: Select all
$ inxi -G
Graphics: Card: Advanced Micro Devices [AMD/ATI] Kabini [Radeon HD 8280 / R3 Series]
Display Server: X.Org 1.19.2 drivers: ati,radeon (unloaded: modesetting,fbdev,vesa)
Resolution: 1920x1080@60.00hz
GLX Renderer: Gallium 0.4 on AMD KABINI (DRM 2.49.0 / 4.10.1-antix.1-486-smp, LLVM 3.9.1)
GLX Version: 3.0 Mesa 13.0.5
Last edited by eugen-b on 22 Mar 2017, 23:30, edited 1 time in total.
-
Posts: 307
- Joined: 23 Aug 2015
#39
I created a new live USB stick, this time with GPT, wrote the ISO with dd and booted in UEFI mode. The UEFI menu is very attractive, IMO! I booted with toram option.
I tried to install again on a btrfs partition and got the same error as described before, step 3. stops at 3% if I say yes to format partiotion, it stops immediately if I select no, then it goes back to step 1.
Installing on ext4 works.
The option to install Grub to ESP is greyed out, even though I booted in UEFImode and there is a 256MB FAT32 partition.
I tried to install again on a btrfs partition and got the same error as described before, step 3. stops at 3% if I say yes to format partiotion, it stops immediately if I select no, then it goes back to step 1.
Installing on ext4 works.
The option to install Grub to ESP is greyed out, even though I booted in UEFImode and there is a 256MB FAT32 partition.
-
Posts: 2,238
- Joined: 16 Dec 2007
#40
the 32bit cannot install in UEFI mode. we should probably have a warning pop up about that.eugen-b wrote:I created a new live USB stick, this time with GPT, wrote the ISO with dd and booted in UEFI mode. The UEFI menu is very attractive, IMO! I booted with toram option.
I tried to install again on a btrfs partition and got the same error as described before, step 3. stops at 3% if I say yes to format partiotion, it stops immediately if I select no, then it goes back to step 1.
Installing on ext4 works.
The option to install Grub to ESP is greyed out, even though I booted in UEFImode and there is a 256MB FAT32 partition.
- Posts: 307 eugen-b
- Joined: 23 Aug 2015
#41
I heard about that and it makes sense, but you often realize such facts when they hit you in your face. __{{emoticon}}__dolphin_oracle wrote:the 32bit cannot install in UEFI mode. we should probably have a warning pop up about that.
-
Posts: 1,445
- Joined: 09 Feb 2012
#42
eugen, for btrfs, what mount options should it specify?
? Something like"defaults,noatime,compress=lzo,commit=120 0 0"
AFAICT, here's how the installer currently attempts to handle btrfs
? Something like"defaults,noatime,compress=lzo,commit=120 0 0"
AFAICT, here's how the installer currently attempts to handle btrfs
Code: Select all
https://github.com/antiX-Linux/antiX-installer/blob/master/minstall.cpp#L2459
FILE *fp = fopen("/mnt/antiX/etc/fstab","w");
. . .
} else if (isFormatBtrfs) {
sprintf(line,"%s / auto defaults,noatime 1 1\n", rootdev);
========= SCRAPER REMOVED AN EMBEDDED LINK HERE ===========
url was:"https://wiki.archlinux.org/index.php/fstab"
linktext was:"https://wiki.archlinux.org/index.php/fstab"
====================================
Note: If the root file system is btrfs, the fsck order should be set to 0 instead of 1
-
Posts: 1,445
- Joined: 09 Feb 2012
#43
the `killall conky` during changeover isn't happening. The new WM session winds up with multiple, overlapped, conkies running.1. when leaving icewm to another wm, then coming back, format of workspace area changes. (sometimes 1 2, sometime alpha, beta)
2. changing theme in icewm makes conky get a little weird
-
Posts: 98
- Joined: 29 Jan 2016
#44
hello anticapitalista and everyone
This Nvidia NV34M always works fine upon installs. Here has never need experience of other nvidia avenues. How would you wish this set-up tested? In a repository entry of deb multimedia Distribution shows"stecth." (is that normal?) The ram usage throughout upgrade rest below 166 M. Same usb that ran live the T42 for fifty hours installed here the D800. live used gparted to format target. install to entire drive. upgraded showing repositories ; is that right?
This Nvidia NV34M always works fine upon installs. Here has never need experience of other nvidia avenues. How would you wish this set-up tested? In a repository entry of deb multimedia Distribution shows"stecth." (is that normal?) The ram usage throughout upgrade rest below 166 M. Same usb that ran live the T42 for fifty hours installed here the D800. live used gparted to format target. install to entire drive. upgraded showing repositories ; is that right?
Code: Select all
System: Host: phoema2x17 Kernel: 4.10.1-antix.1-486-smp i686 (32 bit gcc: 6.3.0)
Desktop: Fluxbox 1.3.5 dm: slim Distro: antiX-17.a2_386-full keerfa 19 March 2017
Machine: Device: portable System: Dell product: Latitude D800 Chassis: type: 8
Mobo: Dell model: 0T1957 BIOS: Dell v: A11 date: 09/03/2004
CPU: Single core Intel Pentium M (-UP-) cache: 2048 KB bmips: 3388 speed/min/max: 1700/600/1700 MHz
CPU Flags: acpi bts clflush cmov cx8 de dts est fpu fxsr mca mce mmx msr mtrr pbe pge pse sep ss sse
sse2 tm tm2 tsc vme
Graphics: Card: NVIDIA NV34M [GeForce FX Go5200 64M] bus-ID: 01:00.0 chip-ID: 10de:0324
Display Server: X.Org 1.19.2 drivers: nouveau (unloaded: modesetting,fbdev,vesa)
Resolution: 1680x1050@60.01hz
GLX Renderer: Gallium 0.4 on NV34 GLX Version: 1.5 Mesa 13.0.5 Direct Rendering: Yes
Network: Card-1: Broadcom Limited NetXtreme BCM5705M Gigabit Ethernet
driver: tg3 v: 3.137 bus-ID: 02:00.0 chip-ID: 14e4:165d
Card-2: Intel PRO/Wireless 2200BG [Calexico2] Network Connection
driver: ipw2200 v: 1.2.2kmprq bus-ID: 02:03.0 chip-ID: 8086:4220
Drives: HDD Total Size: 16.0GB (32.8% used)
Repos: Active apt sources in file: /etc/apt/sources.list.d/antix.list
deb http://repo.antixlinux.com/stretch/ stretch main nosystemd
Active apt sources in file: /etc/apt/sources.list.d/debian.list
deb http://ftp.gr.debian.org/debian/ stretch non-free contrib main
deb http://security.debian.org/ stretch/updates non-free contrib main
Weather: Conditions: 32 F (0 C) - Clear Wind: From the NNE at 8 MPH Humidity: 34%
Pressure: 30.55 in (1034 mb) Wind Chill: 25 F (-4 C)
Location: Washington DC (US) Altitude: 12 ft
Time: March 23, 4:33 AM EDT Observation Time: March 23, 3:52 AM EDT
Info: Processes: 127 Uptime: 1:36 Memory: 258.2/1002.7MB
Init: SysVinit v: 2.88 runlevel: 5 default: 5 Gcc sys: 6.3.0
Client: Shell (bash 4.4.111 running in lxterminal) inxi: 2.3.8
ihere@phoema2x17:~
-
Posts: 307
- Joined: 23 Aug 2015
#45
I used to apply the command in a separate terminal window
which saves me about 45% disk space.
For fstab I apply noatime,compress=lzo,ssd(or autodefrag when on HDD),commit=120 0 0. The zero at the end is not necessary, the fsck would be skipped with an error message anyway at boot.
But this might not be the preferred way for everybody. For example s.o. uses mutt and needs relatime. Or doesn't trust transparent compression. commit=120 delays disk writes to every two minutes, default is 60 seconds.
So, I would be careful to change mount default options and notify the user about them.
Best thing would be if the user could get the chance to enter his or her preferred mount options.skidoo wrote:eugen, for btrfs, what mount options should it specify?
? Something like"defaults,noatime,compress=lzo,commit=120 0 0"
Note: If the root file system is btrfs, the fsck order should be set to 0 instead of 1
I used to apply the command in a separate terminal window
Code: Select all
sudo mount -o remount,compress-force=lzo,ssd(or autodefrag),noatime -t btrfs /dev/sda1 /mnt/antix
For fstab I apply noatime,compress=lzo,ssd(or autodefrag when on HDD),commit=120 0 0. The zero at the end is not necessary, the fsck would be skipped with an error message anyway at boot.
But this might not be the preferred way for everybody. For example s.o. uses mutt and needs relatime. Or doesn't trust transparent compression. commit=120 delays disk writes to every two minutes, default is 60 seconds.
So, I would be careful to change mount default options and notify the user about them.