Tiny Core 14.0 Alpha 1 Testing

[curaga]

md5 files are not checked on boot, only on download. This has always been so, specifically because it would make boot very slow on old cpus.

[CNK]

PS: I wonder how many programs (as % from total) still run under Xvesa /Xfbdev in today tc14. Because most of them ask for gtk3, which in turn ask for Xlibs -> Xorg. Any proper (modern) browser  asks for Xorg.

I think you're mixing up the X server and Xlib. Xvesa/Xfbdev were always supposed to work with Xorg's Xlib, they're separate things (Xlib allows programs to talk to the X server).

You might be interested to look at Nano Linux, as it's a Tiny Core based distro that uses a "nano" version of both the X server and Xlib.

[nick65go]

I love 32 bits software. But unfortunately (for me) my UEFI-64bits booting in 32 bits will show only 2GB RAM, out of 8 GB RAM. So no go for Xvesa and 32 bits kernel, only for chroot. Neither mixture kernel 64 bits and Xfbdev, because insane radeon luminosity without xf86-video-radeon. [I hate Dell /HP firmware for (L)users]

So, to summarize:
1. OK, md5 route for onboot.lst was already implemented in TC, sorry that I overlook it (-1). But the merge for 100 tczs into one Xorg-full.tcz for fast speed was not yet challenged (+1).

2. Sorting/indexing files inside squashfs-ed tcz for speed was not challenged (+1). Neither "parallel" processing /piping (even with naive "interval halving method" like sub-call in sub-call) (+1)

3. from distrowatch -> Nano Linux has status discounted from 2015-12-26 (-1). My mistake to say (talk about) Xlib.(-1). Not interested.

4. All online-banking ask for "modern"/bloated browser WITH java-script. So dillo / links / netsurf etc are dead for this task. The main point was about GTK3! and no up-to-date (security concerns) browser runs without gtk3 and Xorg.(+1).

5. May I repeat the question again to (long experience) users: How many programs (as % from total) still run under Xvesa /Xfbdev in today tc14?

[nick65go]

Testing TC14 alpha 64 bits, I found a small bug in filetool for backup/restore (png attached). The ESC-characters (maybe for colors?) are not interpreted, instead they are shown ad-literal, like " ^ [ [ ".

[GNUser]

Hi, nick65go. That's a screenshot of /usr/local/bin/filetool (GUI frontend to filetool.sh) (included in Xprogs.tcz). It seems that it does not support colors.

filetool.sh was colorized recently. I'm actually finding the colors a bit jarring since it's the only command I use over SSH that does colors. I could get used to it, but if the colors are creating problems for the GUI frontend then maybe the colors should be scrapped?

[curaga]

Oh, if it causes trouble for the GUI, indeed should be reverted. Will revert.

[Rich]

Hi nick65go
fltk-1.3 has been patched to recognized ANSI color escape sequences.

[curaga]

Un-reverting the script change.

[nick65go]

Hi Tinycore Team! Thank you very much, I confirm that in TC14 alpha1 64 bits:

- the FLWM windows-manager now has the corner arrows (thanks Rich)
- the filetool.sh runs with colors, same as is its GUI filetool. (thanks curaga)
- Xfe file-manager was updated to version 1.45 (thanks GNUnuser)

-Xorg-server (1.21.1.6), kernel (6.1.2), busybox (1.36) are up-to-date as in other Linux-distributions.
Maybe there are other good news (not discovered yet by me versus tc13).
PS: I am not picky, just say that filetools, shell versus gui, has different output text colors (see attach).

[nick65go]

I have a problem with Xorg-3D. My laptop has an APU (=CPU+GPU). The kernel sees an AMDGPU (so core GCN 2.x?), but UEFI console and Xorg want RADEON.

I exit from Xorg and I instaled from condole the graphics-6.1.2-tinycore64.tcz, without instaling xf86-video-ati.tcz, neither xf86-video-amdgpu.tcz.
Then /var/log/Xorg.0.log is:

Code: [Select]

[   660.876] (II) Platform probe for /sys/devices/pci0000:00/0000:00:01.0/drm/card0
[   660.961] (WW) Warning, couldn't open module ati
[   660.961] (EE) Failed to load module "ati" (module does not exist, 0)[   661.171] (II) modeset(0): glamor X acceleration enabled on KABINI (, LLVM 13.0.0, DRM 2.50, 6.1.2-tinycore64)
[   661.289] (II) modeset(0): [DRI2] Setup complete
[   661.289] (II) modeset(0): [DRI2]   DRI driver: radeonsi
[   661.289] (II) modeset(0): [DRI2]   VDPAU driver: radeonsi
[   664.064] (II) modeset(0): Modeline "1600x900"x0.0  128.78  1600 1648 1680 2298  900 902 907 934 -hsync -vsync (56.0 kHz eP)
[   664.162] (EE)
[   664.162] (EE) Backtrace:
[   664.162] (EE) 0: /usr/local/lib/xorg/Xorg (xorg_backtrace+0x56) [0x5609e30bfd3f]
...
[   664.162] (EE) 2: /lib/libc.so.6 (0x7f6a3f66a000+0x36870) [0x7f6a3f6a0870][   660.876] (II) Platform probe for /sys/devices/pci0000:00/0000:00:01.0/drm/card0
[   660.961] (WW) Warning, couldn't open module ati
[   660.961] (EE) Failed to load module "ati" (module does not exist, 0)[   661.171] (II) modeset(0): glamor X acceleration enabled on KABINI (, LLVM 13.0.0, DRM 2.50, 6.1.2-tinycore64)
[   661.289] (II) modeset(0): [DRI2] Setup complete
[   661.289] (II) modeset(0): [DRI2]   DRI driver: radeonsi
[   661.289] (II) modeset(0): [DRI2]   VDPAU driver: radeonsi
[   664.064] (II) modeset(0): Modeline "1600x900"x0.0  128.78  1600 1648 1680 2298  900 902 907 934 -hsync -vsync (56.0 kHz eP)
[   664.162] (EE)
[   664.162] (EE) Backtrace:
[   664.162] (EE) 0: /usr/local/lib/xorg/Xorg (xorg_backtrace+0x56) [0x5609e30bfd3f]
...
[   664.162] (EE) 2: /lib/libc.so.6 (0x7f6a3f66a000+0x36870) [0x7f6a3f6a0870]
...
[   664.164] (EE) 10: /usr/local/lib/libLLVM-13.so (LLVMBuildInBoundsGEP2+0x51) [0x7f6a3a6cbc7d]
[   664.164] (EE) 11: /usr/local/lib/dri/radeonsi_dri.so (0x7f6a3d971000+0x924823) [0x7f6a3e295823]
[   664.165] (EE)
Fatal server error:
[   664.165] (EE) Caught signal 6 (Aborted). Server aborting
...
[   664.164] (EE) 10: /usr/local/lib/libLLVM-13.so (LLVMBuildInBoundsGEP2+0x51) [0x7f6a3a6cbc7d]
[   664.164] (EE) 11: /usr/local/lib/dri/radeonsi_dri.so (0x7f6a3d971000+0x924823) [0x7f6a3e295823]
[   664.165] (EE)
Fatal server error:
[   664.165] (EE) Caught signal 6 (Aborted). Server aborting
So the Xorg is focused on radeon driver, even if in dmesg I have

Code: [Select]

cat 1.txt | grep -i amd
smpboot: CPU0: AMD A6-6310 APU with AMD Radeon R4 Graphics (family: 0x16, model: 0x30, stepping: 0x1)
Performance Events: AMD PMU driver.
perf: AMD IBS detected (0x000000ff)
amd_uncore: 4  amd_nb counters detected
amd_uncore: 4  amd_l2 counters detected
AMD-Vi: AMD IOMMUv2 functionality not available on this system - This is not a bug.
ehci-pci 0000:00:12.0: applying AMD SB700/SB800/Hudson-2/3 EHCI dummy qh workaround
ehci-pci 0000:00:13.0: applying AMD SB700/SB800/Hudson-2/3 EHCI dummy qh workaround
[drm] amdgpu kernel modesetting enabled.
amdgpu: Virtual CRAT table created for CPU
amdgpu: Topology: Add CPU node
amdgpu 0000:00:01.0: vgaarb: deactivate vga console
amdgpu 0000:00:01.0: amdgpu: get invalid ip discovery binary signature from vram
amdgpu 0000:00:01.0: amdgpu: amdgpu_discovery is not set properly
amdgpu 0000:00:01.0: amdgpu: failed to read ip discovery binary from file
[drm:0xffffffffa0295938] *ERROR* amdgpu_discovery_init failed
amdgpu 0000:00:01.0: amdgpu: Fatal error during GPU init
amdgpu 0000:00:01.0: amdgpu: amdgpu: finishing device.
amdgpu: probe of 0000:00:01.0 failed with error -22
ATOM BIOS: AMD

Code: [Select]

cat 1.txt | grep -i radeon
smpboot: CPU0: AMD A6-6310 APU with AMD Radeon R4 Graphics (family: 0x16, model: 0x30, stepping: 0x1)
[drm] radeon kernel modesetting enabled.
radeon 0000:00:01.0: vgaarb: deactivate vga console
radeon 0000:00:01.0: VRAM: 1024M 0x0000000000000000 - 0x000000003FFFFFFF (1024M used)
radeon 0000:00:01.0: GTT: 2048M 0x0000000040000000 - 0x00000000BFFFFFFF
[drm] radeon: 1024M of VRAM memory ready
[drm] radeon: 2048M of GTT memory ready.
[drm] radeon: dpm initialized
radeon 0000:00:01.0: WB enabled
radeon 0000:00:01.0: fence driver on ring 0 use gpu addr 0x0000000040000c00
radeon 0000:00:01.0: fence driver on ring 7 use gpu addr 0x0000000040000c1c
radeon 0000:00:01.0: radeon: using MSI.
[drm] radeon: irq initialized.
[drm] radeon atom DIG backlight initialized
[drm] Radeon Display Connectors
fbcon: radeondrmfb (fb0) is primary device
radeon 0000:00:01.0: [drm] fb0: radeondrmfb frame buffer device
[drm] Initialized radeon 2.50.0 20080528 for 0000:00:01.0 on minor 0
PS: To avoid any old firmaware bug, I tried both with firmware-radeon and firmware-amdgpu from TC original then from Alpine.
But I have the same problem, for Xorg-3D.
EDIT: I does not help the fact that in http://tinycorelinux.net/14.x/x86_64/release/src/kernel/config-6.1.2-tinycore64 we have

Code: [Select]

CONFIG_DRM_AMDGPU=m
# CONFIG_DRM_AMDGPU_SI is not set
# CONFIG_DRM_AMDGPU_CIK is not set

[nick65go]

To avoid any doubts about mixing firmware + drivers etc, in a console I did:

Code: [Select]

ls /tmp/tcloop/ | grep xf86-video
xf86-video-ati/
xf86-video-fbdev/
xf86-video-vesa/
and

Code: [Select]

lsmod | grep amdoutput nothing, only kernel-radeon drv is loaded
and

Code: [Select]

cat onboot.lst
kmaps.tcz
regdb.tcz
HP-radeon.tcz
graphics-6.1.2-tinycore64.tcz
Xorg-7.7-3d.tcz
flwm.tcz
aterm.tcz
wbar.tcz
wifi.tcz
acpid.tczSo, at least for RADEON case: I suspect /usr/local/lib/dri/radeonsi_dri.so has a linking problem.

[Juanito]

If you don't load xf86-video-amdgpu or xf86-video-ati, but do load graphics-KERNEL, does the modesetting driver work?

[curaga]

There is no problem with ati/radeon vs amdgpu. Your card is supported by ati/radeon, and the optional/experimental amdgpu support for it is correctly disabled. X failed to start because llvm called abort(), meaning you are hitting a llvm or mesa bug. You can disable 3d in your xorg.conf or try older versions.

[nick65go]

There is no problem with ati/radeon vs amdgpu. Your card is supported by ati/radeon, and the optional/experimental amdgpu support for it is correctly disabled. X failed to start because llvm called abort(), meaning you are hitting a llvm or mesa bug. You can disable 3d in your xorg.conf or try older versions.

Right I hit a bug in LLVM  or mesa.
So for clarity, lets forget ALL future discussion about amdgpu, and concentrate only on radeon.

Xorg-3D without graphics-KERNEL works OK, using xorg-FBDEV (so is not accelerated).

If you don't load xf86-video-amdgpu or xf86-video-ati, but do load graphics-KERNEL, does the modesetting driver work?

1. After exit xorg and loading (in console) the graphics-KERNEL, then the console/prompter is frozen and I must do cold reset the laptop.

2.Fresh boot.After exit xorg and loading (in console) FIRST the firmware-radeon.tcz and THEN the graphics-KERNEL, now I stay console/prompter. When I start Xorg now, it shows briefly (1 second) the Xorg-window + mouse cursor, then it quickly quits back in console.
The less /var/log/Xorg.0.log shows that Xorg successfully loaded modeset-drv +llinput etc and then Xorg spits (EE) about libLLVM-13.so + radeonsi_dri.so.

3. Fresh boot.After exit xorg and loading (in console), first firmare-radeon, then Xf86-video-ati then graphics-kernel. And now I have the same briefly Xorg shown and then it quickly quits back in console. The less /var/log/Xorg.0.log shows that Xorg successfully loaded RADEON-drv +llinput etc, and then it spits (EE) about libLLVM-13.so + radeonsi_dri.so, right afler "RADEON (0): Modeline "1600x900" etc.

[nick65go]

For the case loading firmware-radeon + xf86-video-ati + graphics-kernel, all there tcz loaded from console before starting xorg, here is attached the full /var/log/Xorg.0.log. You can see the (EE) from libLLVM-13.so + radeonsi_dri.so.
FYI: I know is in alpha1 stage so work in progress. but to use Xorg-3D without acceleration is a NOT-go for anyone with a laptop [so I think].