Carolina: Vanguard Edition Release 2 December 2014

[Marv]

@battleshooter

As long as its vanguarding you are, I think it would be fine to put the newer ipw2200 version 3.1.5 firmware from http://puppylinuxstuff.meownplanet.net/ ... rmware.pet in in place of the 3.1.0 firmware currently in Carolina and vlina. I've used it on many ipw2200 machines and it really may be somewhat better wrt the dropout issue with no downsides. Could also just be put in the repo so it's more visible. Don't think it's there but haven't checked lately.

[battleshooter]

If it is going in as a fix, after flogging /root/Startup/9acpid more mercilessly, I would move the test for a running acpid to line 10 and leave line 9 as is. Modified version attached for your perusal.

Makes sense to me, adding a test instead of replacing the existing one. I'll be adding the file you attached.

Edit: Pulled the CPU frequency governor from the Bay Trail box vlina (unneeded there and left over from the savefile swap while working on the Xorg pets) and the glxgears FPS is up to 2150 FPS

LxPupTahr 14.10.1 non-PAE with std 3.14.20 kernel
Xorg version: 1.15.1
Intel driver version: 2.99.910 (compiled for 1.15.1)
Mesa version: 10.1.3
maximum driver: 'intel'
glxgears (vblank_mode=0) 2915 FPS

Hmm, how is Tahr able to still pull 800 more frames out of thin air? Anyway, don't mind me, just nit-picking I actually am very happy we've got the box working

Hey battleshooter, what if you just use the date as a sub-title? Then you wouldn't have to change the logo and everyone would be able to tell which is the latest version.

It might be for my own benefit rather than anyone else's, I tend to get by builds mixed up like nobody's business Having a build number on the splash screen helps me immeasurably I still will take your suggestion and put it on the subtitle on top of maybe putting a small "b2" on the bottom corner of the splash for my sake

As long as its vanguarding you are, I think it would be fine to put the newer ipw2200 version 3.1.5 firmware from http://puppylinuxstuff.meownplanet.net/ ... rmware.pet in in place of the 3.1.0 firmware currently in Carolina and vlina. I've used it on many ipw2200 machines and it really may be somewhat better wrt the dropout issue with no downsides.

I'll put it in, don't see any reason not to.

[battleshooter]

This currently the contents of my TODO list:

Code: Select all

Hardinfo pre patch
ipw2200 firmware
Marv's 9acpi update
Edit Skype
Edit MasterPDF
delete  /root/.config/xfce4/desktop/backdrop.list

For Xorg:
Add Xorg update
Add 
	setxkbmap -option terminate:ctrl_alt_bksp 
to .xinitrc

Remove clashing sven keycodes , 
	236 default email
	178 browser
	214 grandr

Sigh. Recompile with CONFIG_HIGHMEM64G=y

Recompile gcc with intel support

Has anyone noticed any program bugs, hardware incompatibility that hasn't been covered yet? (I can't tell you how scared I am to ask that )

Any suggestions for improvement? I'm recompiling the kernel so if there's desired options, now would be a good time to suggest them.

I'm fairly sure CONFIG_HIGHMEM64G=y is the option for PAE, but I'm not sure as they removed the old CONFIG_X86_PAE option that was pretty obvious. can anyone confirm this?

[Marv]

Hmm, how is Tahr able to still pull 800 more frames out of thin air? Anyway, don't mind me, just nit-picking I actually am very happy we've got the box working

Don't know at this point. When the dust settles and my head clears, I'll run them head to head again. If it is to be a PAE kernel only, it would be great to have the forcepae patch in so it would run on the Pentium Ms. It's a Carolina thing, but at first glance I didn't see a default image viewer. I added Viewnior, small no dependencies. I've been using vlina with the Xorg pet and have been through multiple reboots and restarts chasing the intel microcode grail with no problems at all. Swapped the WM theme to tyrex. It likes the default wallpaper, as do I.

[gcmartin]

Kernel - add KVM feature if not already there. @StemSee has this in all of his EmSee distros.

[battleshooter]

Kernel - add KVM feature if not already there. @StemSee has this in all of his EmSee distros.

Yep, KVM's already in since the k3.17 build.

Works too with AQemu and a couple of frontends I compiled for testing.

Off topic, I had a look at compiling Samba 4 instead of asking 01micko for his pet. While it compiled cleanly I had some issues turning it into a pet. Weird glitches. Probably look at it again later. Just thought I'd mention that.

If it is to be a PAE kernel only, it would be great to have the forcepae patch in so it would run on the Pentium Ms.

Ok. I remember coming across that in some of my online travels. I'll look in to that.

[Marv]

The following would enable microcode updating by the kernel. Of interest mostly with older CPUs so it's not a vital but may help with the 'pipe underrun' stuff.

Code: Select all

Enabling Intel early microcode loading in custom kernels

In order for early loading to work in custom kernels, "CPU microcode loading support" needs to be compiled into the kernel, NOT compiled as a module. This will enable the "Early load microcode" prompt which should be set to "Y".

CONFIG_MICROCODE=y
CONFIG_MICROCODE_INTEL=y
CONFIG_MICROCODE_INTEL_EARLY=y
CONFIG_MICROCODE_EARLY=y

[666philb]

If it is to be a PAE kernel only, it would be great to have the forcepae patch in so it would run on the Pentium Ms.

Ok. I remember coming across that in some of my online travels. I'll look in to that.

here's a link for the one i used in tahrpup ... although there may be a newer one now https://lkml.org/lkml/2014/3/7/180

[Marv]

@battleshooter

The forcepae patch in Tahr works fine for me. Another compile option that would help in the microcode quest is:

Code: Select all

CONFIG_X86_CPUID=y 

[gcmartin]

This is a clear discussion occuring on kernel options.

It is appearing almost as a clinic for building a 32bit kernel with the features that are important to most traditional distro for PCs used by community members.

A clinic as such probably would be useful as well for 64bit kernel builds where a developer appeals to community for recommendations to a solid kernel build.

In each/either case, important parms are brought to the forefront allowing other developers to benefit from and contribute to the combined wisdom.

Thanks @BattleShooter for exposing this as an approach.

[Chili Dog]

This is a clear discussion occuring on kernel options.

It is appearing almost as a clinic for building a 32bit kernel with the features that are important to most traditional distro for PCs used by community members.

A clinic as such probably would be useful as well for 64bit kernel builds where a developer appeals to community for recommendations to a solid kernel build.

In each/either case, important parms are brought to the forefront allowing other developers to benefit from and contribute to the combined wisdom.

Thanks @BattleShooter for exposing this as an approach.

BUT DOES IT HAVE SAMBA?

[anikin]

This is a clear discussion occuring on kernel options.

It is appearing almost as a clinic for building a 32bit kernel with the features that are important to most traditional distro for PCs used by community members.

A clinic as such probably would be useful as well for 64bit kernel builds where a developer appeals to community for recommendations to a solid kernel build.

In each/either case, important parms are brought to the forefront allowing other developers to benefit from and contribute to the combined wisdom.

Thanks @BattleShooter for exposing this as an approach.

gcmartin,
Your wisdom is a source of inspiration for our developers. This community owes you a huge debt of gratitude for all your selfless work and guidance. Please, be our Supreme Spiritual Leader. Amen.

[Geoffrey]

BUT DOES IT HAVE SAMBA?

I don't understand the reasoning for that response to gcmartin, but the word flippant comes to mind.

Samba-TNG is used and along with gvfs works well enough for me.

[musher0]

Hello, battleshooter.

You did get me "sold" on your product .

However, your new Carolina girl leaves me out in the cold after (giving
me the) boot-up, exactly at

Code: Select all

root login on "tty1"

I mean: the boot process goes well and then the computer freezes badly
at that point.

BTW, if it's any help, my computer's an AMD Sempron 3100+ from 2007.
Or could it be the 3.18 kernel?

Best regards.

musher0

Second try with fresh download, as battleshooter recommended: same thing.

[Marv]

Hello, battleshooter.

You did get me "sold" on your product .

However, your new Carolina girl leaves me out in the cold after (giving
me the) boot-up, exactly at

Code: Select all

root login on "tty1"

I mean: the boot process goes well and then the computer freezes badly
at that point.

BTW, if it's any help, my computer's an AMD Sempron 3100+ from 2007.
Or could it be the 3.18 kernel?

Best regards.

musher0

Second try with fresh download, as battleshooter recommended: same thing.

Newer kernels do require that modesetting be used and done by the kernel. Is the video radeon? That point is pretty much when xorgwizard is autorun and locked me out pretty solidly until the driver got sorted. What broke the logjam for me was using xxx.modeset=0 as a kernel parameter where xxx is intel, radeon, etc and/or turning the auto xorgwizard off so it was forced to use an xorg.conf I wrote. The place to toggle that is in the last line of DISTRO_SPECS in the initrd.gz. Once I got it limpingly going, I could troubleshoot.

[battleshooter]

The following would enable microcode updating by the kernel. Of interest mostly with older CPUs so it's not a vital but may help with the 'pipe underrun' stuff.

Yeah I read about that too, but thanks for digging out the actual kernel options.

here's a link for the one i used in tahrpup ... although there may be a newer one now https://lkml.org/lkml/2014/3/7/180

Many thanks Phil, I appreciate this rather than me finding it on my own as I now know it's been pretested

Code: Select all

CONFIG_MICROCODE=y
CONFIG_MICROCODE_INTEL=y
CONFIG_MICROCODE_INTEL_EARLY=y
CONFIG_MICROCODE_EARLY=y 
	
CONFIG_X86_CPUID=y 

All on the to do list.

This is a clear discussion occuring on kernel options.

It is appearing almost as a clinic for building a 32bit kernel with the features that are important to most traditional distro for PCs used by community members.

A clinic as such probably would be useful as well for 64bit kernel builds where a developer appeals to community for recommendations to a solid kernel build.

In each/either case, important parms are brought to the forefront allowing other developers to benefit from and contribute to the combined wisdom.

Thanks @BattleShooter for exposing this as an approach.

I like how you make it sound like I know what I'm doing

I just ask because there are people on the forum that know more or different things from me. But thanks

BUT DOES IT HAVE SAMBA?

Yeah, as Geoff said, a small file sharing version is included by default, but the full Samba package can be downloaded from the repos.

Please, be our Supreme Spiritual Leader. Amen.

Yeah, he's an all round nice guy isn't he?

Newer kernels do require that modesetting be used and done by the kernel.

Second try with fresh download, as battleshooter recommended: same thing.

What graphics card does the machine use Musher ?

[battleshooter]

I upgraded Xorg here for Build 2, working nicely on a Radeon, booting up on a fresh save with correct resolution and acceleration with no additional kernel parameters. However on a nVidia I have to manually modprobe nouveau. I'm not sure how to make that automatic. Maybe I'll try Tahr's XorgWizard again for automatic detection.

The main issue I'm having at the moment is working out why GCC/make isn't compiling on Intel processors. Most leads seem to be dead ends. The annoying thing is I don't have an Intel processor computer so it's a bit tricky to test.

Started a thread on the issue here if anyone wants to chip in.

All other fixes seem to be working fine, image list isn't crashing, MasterPDF and Skype boot up fine. Xorg is stable besides the non detection issue. Probably won't get on to the kernel recompile till I can sort the GCC issue.

[musher0]

Newer kernels do require that modesetting be used and done by the kernel. Is the video radeon? That point is pretty much when xorgwizard is autorun and locked me out pretty solidly until the driver got sorted. What broke the logjam for me was using xxx.modeset=0 as a kernel parameter where xxx is intel, radeon, etc and/or turning the auto xorgwizard off so it was forced to use an xorg.conf I wrote. The place to toggle that is in the last line of DISTRO_SPECS in the initrd.gz. Once I got it limpingly going, I could troubleshoot.

Thanks Marv.

I use the "default" graphics chip rather than a video card on this AMD computer. I'll
check which brand it is, and try your command on the kernel line in grub4dos.

BFN.

musher0

[yerc1]

I think the only issues left are:

• GCC Intel incompatibility
  Musher's root tty issue
  Yerc's dear touchpad

Hah having a VIP feeling atm LOL

But seriously though thanks battleshooter... and don't let our issues ruin the holiday spirit.

[Billtoo]

I used the pets that I made in Racy 5.5 and then compiled GCC 492 in
my Carolina Vanguard 1.3 install on a pc with an Athlon processor, then moved the flash drive to my Acer
desktop which has an Intel processor.
-------------------------------------------------------------------------------

I compiled lxterminal and grun on this pc with an Intel i3 processor so
it seems that compiling works now.

video-info-glx 1.5.3 Fri 19 Dec 2014 on Vanguard-Carolina 1.3 Linux 3.18.0 i686
2.0 VGA compatible controller: Intel Corporation Core Processor Integrated Graphics Controller (rev 18)
oem: Intel(R)Ironlake Desktop Graphics Chipset Accelerated VGA BIOS
product: Intel(R)Ironlake Desktop Graphics Controller Hardware Version 0.0

X Server: Xorg Driver: intel
X.Org version: 1.11.0
dimensions: 1920x1080 pixels (507x285 millimeters)
depth of root window: 24 planes

direct rendering: Yes
server glx vendor string: SGI
server glx version string: 1.4
OpenGL vendor string: Tungsten Graphics, Inc
OpenGL renderer string: Mesa DRI Intel(R) Ironlake Desktop x86/MMX/SSE2
OpenGL version string: 2.1 Mesa 7.10.3

Intel(R) Core(TM) i3 CPU 540 @ 3.07GHz
Core 0: @1200 1: @1200 2: @1200 3: @1200 MHz

EDIT: I compiled Gcc 492 on my Ivybridge install:

video-info-glx 1.5.3 Sat 20 Dec 2014 on Vanguard-Carolina 1.3 Linux 3.18.0 i686
2.0 VGA compatible controller: Intel Corporation Ivy Bridge Graphics Controller (rev 09)
oem: Intel(R) Sandybridge/Ivybridge Graphics Chipset Accelerated VGA BIOS
product: Intel(R) Sandybridge/Ivybridge Graphics Controller Hardware Version 0.0

X Server: Xorg Driver: intel
X.Org version: 1.16.1
dimensions: 1920x1080 pixels (507x285 millimeters)
depth of root window: 24 planes

direct rendering: Yes
server glx vendor string: SGI
server glx version string: 1.4
OpenGL vendor string: Intel Open Source Technology Center
OpenGL renderer string: Mesa DRI Intel(R) Ivybridge Desktop x86/MMX/SSE2
OpenGL version string: 3.0 Mesa 10.3.2

It's working great