Gentoo Forums
Gentoo Forums
Gentoo Forums
Quick Search: in
Use old amd64 gentoo image on new amd64 hardware, possible?
View unanswered posts
View posts from last 24 hours

Goto page Previous  1, 2  
Reply to topic    Gentoo Forums Forum Index Kernel & Hardware
View previous topic :: View next topic  
Author Message
wcg
Guru
Guru


Joined: 06 Jan 2009
Posts: 588

PostPosted: Sun Nov 04, 2012 12:33 pm    Post subject: Reply with quote

AMD 900 series chipsets:
http://en.wikipedia.org/wiki/AMD_900_chipset_series

(See the note at the bottom of the document above
about HPET/MSI and SATA/MSI issues.
http://en.wikipedia.org/wiki/Message_Signaled_Interrupts
)

I do not know if the kernel has workarounds for the SATA/MSI
issues on these chipsets that would prevent them from
misbehaving with 4+ sata drive raid arrays.

I use this in /etc/portage/make.profile (my /etc/make.profile is now
a symbolic link to that file):
Code:

CHOST=x86_64-pc-linux-gnu
CFLAGS="-march=native -O2 -fno-strict-aliasing -pipe -ggdb -Wall"
CXXFLAGS="$CFLAGS -fpermissive"
LDFLAGS="$LDFLAGS -Wl,--as-needed"
MAKEOPTS="-j5"
FEATURES="splitdebug"


(The debugging flags are my own choice. The system is fast enough
with them enabled, and I do not have to rebuild anything to get
a backtrace of a reproducible bug from gdb. The one thing that -ggdb
and splitdebug seem to miss is debugging versions of gcc builtins,
like libgcc_s.so.1. So one sometimes gets a "no debug symbols"
in places in a gdb backtrace when it is looking for some symbol
in the binary being debugged that happens to be a reference to
a function in libgcc_s.so.1. In those cases the debug information
is simply not there for gdb to display with splitdebug enabled
at emerge time. Hopefully what went wrong happened before or
after that position in the code. The -O2 may seem a little radical
for debugging, but one can usually see in the gdb output where
a symbol is missing that is actually in the source code simply
because gcc optimized it away, and that way gdb is tracing
the exact same instruction sequences that produced the bug
in the first place.)

Note that mplayer stable was updated in the last couple of weeks
from mplayer-1.0_rc4_[dated version] to mplayer-1.1-r1, so your
old and newly merged versions of it probably do not match.
I do not know what all was changed. It looks like one thing it
did was split out mpg123 support from an embedded-into-mplayer
version to using a shared system mpg123 library. The newer
mplayer required mpg123 as a dependency, and the newer
mplayer binary is considerably smaller than the previous
version (around 275k vs around 850k on my system). An ffmpeg
update (stable version) was merged at the same time.
_________________
TIA
Back to top
View user's profile Send private message
wcg
Guru
Guru


Joined: 06 Jan 2009
Posts: 588

PostPosted: Mon Nov 05, 2012 4:27 am    Post subject: Reply with quote

So you have a Realtek ethernet chip on an Extreme4? All of the Asrock
motherboards that I looked at descriptions of had the Broadcom chip.
I guess I didn't go back far enough in the models to see any models
with an onboard Realtek ethernet chip.

I actually do have a Realtek codec on board as part of the onboard
audio, though. I did not page down far enough in the hda-intel
options to see that the driver for it was still enabled the last time
I ran make menuconfig. I tried the onboard audio using the generic
codec driver, and alsamixer only showed the master and pcm volume
controls. With the Realtek codec driver (and quirks) enabled, alsamixer
shows controls for master, headphones, pcm, front, mikes, surround,
etc. I looked for the codec chip when installing the motherboard,
but IIRC it was under one of the chipset heatsinks somewhere,
and I could not see what codec the on-board sound was using.

I also tried to disable hpet support in the kernel, thinking about
the comments referencing hpet/MSI quirks in the Wikipedia description
of the AMD 9xx chipsets, but disabling the kernel option for the
hpet device only disables the userspace interface to hpet devices,
/dev/hpet. dmesg showed that the kernel was still seeing and
enabling an hpet device, probably for the use of kernel drivers
(who would be expected to know about quirks relative to the
use of hpet devices on particular chipsets).

Also, I happened to go into BIOS setup again on the 990FX Extreme4
to see if the iommu was really enabled. I noticed that there are not
one but two separate "cpu overclock" menu selections, on the
same page of the BIOS setup display but in different sections,
more or less.

The first one has the "manual mode" or "overclock" choice
(the one where I would expect to see "disable" as an option,
which unfortunately is not there). The second cpu overclock
menu item has "auto" and "manual mode" options. Selecting
"manual mode" on that second overclock menu item causes
three more menu items to appear for adjusting various
clock speeds on peripheral buses, etc.

(I don't care about all that "virtual hot-rodding". I don't play
games on this motherboard, I don't count fps, etc. I simply
want it to run until I don't need it anymore. If I don't fall asleep
at the keyboard waiting for a kernel to compile, it is fast enough.)
_________________
TIA
Back to top
View user's profile Send private message
Tony0945
Apprentice
Apprentice


Joined: 25 Jul 2006
Posts: 276

PostPosted: Mon Nov 05, 2012 10:52 am    Post subject: Reply with quote

Quote:
If I don't fall asleep
at the keyboard waiting for a kernel to compile, it is fast enough.


Amen
Back to top
View user's profile Send private message
miroR
Guru
Guru


Joined: 05 Mar 2008
Posts: 344

PostPosted: Wed Nov 07, 2012 9:04 pm    Post subject: Reply with quote

First a quick note (I intended it to be quick, but I see it's not such quick note, indeed) why I'm late, for which I'm sorry!
An issue I had upon hotswapping/hotplugging ...
4 systems each with 3 HDD 1TB raid5 on it, 3 of which 4 systems is now with 3xHDD Seagate SATA st1000vm002 (Pipeline Series, 5900 rpm, truly silent), and one is not... It is not because this is, as I try to point out, a province of very whimsical Corporate and Soviet EU, and that is all for short, long explanation would be too long....
So, one is not, since it will yet in this story end up differently, because having waited too long, last attempt I made while still waiting, was, I said, what do I do, I don't feel like rsync-cloning my systems when dd cloning is cleaner and much simpler... With dd'ing which is straigtforward and actually quick in comparison to rsync'ing, I only need changing hostnames and network address stuff, and all is just fine...
And missing 3 HDD to have 4 times 3HDD raids in my 4 systems, and having only one spare HDD (we're talking all st1000vm002 so far) I embarked on removing one HDD from raid5 on one system, and installing it (970 Extreme4 supports hotplugging/hotswapping) on a system that had rsync-cloned old system (some two weeks old) on it, and likewise on yet another.
Now all the systems had the ability to be simply cloned (anyone looking for info about cloning, there I gave a link in a previous post of this thread to how I do it)...
And I cloned them.
So they got the ability to be cloned and were cloned, and I got paranoia and anxiety... once I cloned them
Because I didn't anymore feel safe with my 4 raid5 systems of which 3 were degraded!
:oops:
So I had to make another search for the desired Pipeline Series Seagate 1T, but in this province of the Corp. & Sov. EU. no such thing there for me...
So I went for the standard st1000dm003 Seagate 1T, the exact missing number of 3 of them.
But, knowing that it's best to have exactly same models in raids, I decided to do the hotswapping a few more times (just go and figure how many is necessary and how many times I needed to wait for the raids to resync!...) and have all the Pipiline Seagates in the three and these standard Seagates in one single box...
Which would be fine, if human imperfection and design insufficiency weren't to interfere...
And upon doing it on one of the boxes, I made a journey in such way as to find myself in the position to examine and, hesitatingly and with some amount of horror and a touch of despair, investigate whether the "--run" and "--force" options of mdadm do what the "man mdadm" says they do...
8O
Get my drift?
Know what I'm talking about...?
Dear Lord, that would have been a loss!
What happened is, the Extreme4 has a peculiar positioned SATA connectors, on the side, which is good for some, but not good for all MBO's, and unable to stuck one more necessary SATA male connector in the female on the MBO, I unstuck one of only two SATA connectors of the two HDD's of an array of three HDD's (actually partitions are in array, but I'm simplifying here)... of which array of three disks one disk was already missing!
A raid left to run from what it had copied over into RAM, actually a dead RAID.
You can imagine how I felt...
The system was still running, you know?
And... The RAM being already plentiful, so shutdown was nearly entirely software, the stopping of services and all...
With raid6 I had already had similar experience:
http://forum.cgsecurity.org/phpBB3/post438.html
There's not much there to explain about this dreadful adventure.
Just, this time around, I can simply tell you that, because I don't anymore even remember the exact command that I used, because I had issues with testing Corsair DDR3 memory modules later (protracted issue), so I forgot all the circumstances, but the manual says it right...
So I can just tell you that the options --run and/or --force do work, and that system of mine is still fine.. as well as the others...
In short, I got the raid started on those two HDD's, because the --run option made mdadm decide to consider the unstuck HDD as clean, as it seems to have been, because I still haven't found anything broken/missing/whatnot in the raid5 partitions of it.
I intend to relate the issue with the memory modules but this adventure with the raid5 unstuck I felt I needed to relate to you for completeness.
I do admire that great mdadm GNU Linux program, Neil Brown and company do it just fine, really!
Now I need to go over Tony0945's and especially wcg's replies and see where I got to delve deeper yet, and then I have the Corsair issue to recount.
Cheers!
Back to top
View user's profile Send private message
miroR
Guru
Guru


Joined: 05 Mar 2008
Posts: 344

PostPosted: Wed Nov 07, 2012 9:38 pm    Post subject: Reply with quote

Tony0945 wrote:
Quote:
I don't get it.
I see:
Code:
# man -ea world

What would that do? I doesn't fit your explanation.


Should have been "emerge -ea world". Sorry for the typo.

I recall changing form k8 to k10. That was from an Athlon II X2 (Newcastle?) to Phenom II X6 (Deneb). I used the emerge -ea world. Don't recall any problems, but it took a long time with nearly a thousand packages installed.

I get that (similar to my upgrade, except I am upgradiing from Athlon I series.
Tony0945 wrote:
Another time I changed from i686 to K6. In that case I had a lot of crashes because the K6 doesn't run all i686 opcodes. Booting an old system rescue CD (i386 or i486, I don't remember which) and using distcc helped.

I get that too, approximately, just what is distcc . Oh, Isee:
https://en.wikipedia.org/wiki/Distcc
Great!
Back to top
View user's profile Send private message
miroR
Guru
Guru


Joined: 05 Mar 2008
Posts: 344

PostPosted: Wed Nov 07, 2012 10:22 pm    Post subject: Reply with quote

wcg wrote:
AMD 900 series chipsets:
http://en.wikipedia.org/wiki/AMD_900_chipset_series

(See the note at the bottom of the document above
about HPET/MSI and SATA/MSI issues.
http://en.wikipedia.org/wiki/Message_Signaled_Interrupts
)

I do not know if the kernel has workarounds for the SATA/MSI
issues on these chipsets that would prevent them from
misbehaving with 4+ sata drive raid arrays.

I use this in /etc/portage/make.profile (my /etc/make.profile is now
a symbolic link to that file):
Code:

CHOST=x86_64-pc-linux-gnu
CFLAGS="-march=native -O2 -fno-strict-aliasing -pipe -ggdb -Wall"
CXXFLAGS="$CFLAGS -fpermissive"
LDFLAGS="$LDFLAGS -Wl,--as-needed"
MAKEOPTS="-j5"
FEATURES="splitdebug"

You gave me extra homework again!
But I'll have to cut corners.
I've been missing in recent about two weeks with what I regularly posted and people in Croatia are waiting for on my channel:
https://www.youtube.com/user/miroR2
I'll have to leave the debugging out. I like computing, I'd like to program, but my contemporary history in video collecting (that's how I view news and documentaries) and planning on making my own stuff is stronger desire in me.
wcg wrote:

(The debugging flags are my own choice. The system is fast enough
with them enabled, and I do not have to rebuild anything to get
a backtrace of a reproducible bug from gdb. The one thing that -ggdb
and splitdebug seem to miss is debugging versions of gcc builtins,
like libgcc_s.so.1. So one sometimes gets a "no debug symbols"
in places in a gdb backtrace when it is looking for some symbol
in the binary being debugged that happens to be a reference to
a function in libgcc_s.so.1. In those cases the debug information
is simply not there for gdb to display with splitdebug enabled
at emerge time. Hopefully what went wrong happened before or
after that position in the code. The -O2 may seem a little radical
for debugging, but one can usually see in the gdb output where
a symbol is missing that is actually in the source code simply
because gcc optimized it away, and that way gdb is tracing
the exact same instruction sequences that produced the bug
in the first place.)

Great, but too costly for me, timewise, to learn that much.
wcg wrote:

Note that mplayer stable was updated in the last couple of weeks
from mplayer-1.0_rc4_[dated version] to mplayer-1.1-r1, so your
old and newly merged versions of it probably do not match.
I do not know what all was changed. It looks like one thing it
did was split out mpg123 support from an embedded-into-mplayer
version to using a shared system mpg123 library. The newer
mplayer required mpg123 as a dependency, and the newer
mplayer binary is considerably smaller than the previous
version (around 275k vs around 850k on my system). An ffmpeg
update (stable version) was merged at the same time.

That is great. I always liked mplayer.
And mencoder is still the best option, I believe I came to know, for capturing on old Hauppauge composite input, which is what I do (also use satellite and DVB-T input --plain Kaffeine is my best friend there--, I distribute these on three Hauppauge cards, in three of the boxes, currently.
Code:
# mplayer --version
Unknown option on the command line: --version
Error parsing option on the command line: --version
MPlayer 1.1-4.6.3 (C) 2000-2012 MPlayer Team
#

and I am next to do emerge...
I'm thinking about emerge -ea as Tony0945 advised.
True, that would be 1290 packages to install. On the other hand, I guess the flags above is first, than only emerge -ea. Would really taka a longish time... :x
But, what do I clone my systems for if not to use ones that are not in maintenance mode? Right?
Oh, I not done with your teachiing me yet... I'll go and do my homework now. Actually after I reply to the other post of yours that is left.
Cheers!
Actually, one thing here first, both for the Tony0945 and your attention.
Sine, IMO the flags above is first, than only emerge -ea. and since I have to leave debugging out, and have two peers/senior to use advice from...
Since I have the occasion, what would your flags above without debug translate into?
Alse, if there is a safer option to use, and you don't mind my asking, let me know.
I want to be on the relatively safe side.
Let me explain, I am on the testing ~amd64 branch, so I don't mean completely safe, but I'd call it balanced safe side...
I am writing this before I do my homework, so I may myself know more than now later, but...
Currently, as I posted my flags here:
http://forums.gentoo.org/viewtopic-t-940916.html#7176772, without comments they are still:
Code:
CFLAGS="-march=k8 -O2 -pipe"
CXXFLAGS="${CFLAGS}"
CHOST="x86_64-pc-linux-gnu"

You have as in the beginning of this post, above, and I copy only differences from my flags (only one line is same, actually):
Code:

CFLAGS="-march=native -O2 -fno-strict-aliasing -pipe -ggdb -Wall"
CXXFLAGS="$CFLAGS -fpermissive"
LDFLAGS="$LDFLAGS -Wl,--as-needed"
MAKEOPTS="-j5"
FEATURES="splitdebug"

So if I just change your gdb containing line into:
Code:
CFLAGS="-march=native -O2 -fno-strict-aliasing -pipe -Wall"

and leave out:
Code:
FEATURES="splitdebug"
, would I be fine to use your setup.
I won't just go on and do it.
I am now soon to read where you pointed me to, and other places possibly...
(P.S. I looked up your other post, and can only declare I need another look to reply well. I don't figure out all the talk, yet. Haven't even beein into manual mode in my 970 Extreme4 in any detail, yet, and iommu and hpet are still unknown personalities to me.
One is, as already stated, to be introduced to me on one of the wikipedia pages.
Cheers!)
Back to top
View user's profile Send private message
miroR
Guru
Guru


Joined: 05 Mar 2008
Posts: 344

PostPosted: Wed Nov 07, 2012 11:40 pm    Post subject: Reply with quote

wcg wrote:
I also tried to disable hpet support in the kernel, thinking about
the comments referencing hpet/MSI quirks in the Wikipedia description
of the AMD 9xx chipsets, but disabling the kernel option for the
hpet device only disables the userspace interface to hpet devices,
/dev/hpet. dmesg showed that the kernel was still seeing and
enabling an hpet device, probably for the use of kernel drivers
(who would be expected to know about quirks relative to the
use of hpet devices on particular chipsets).

Interesting.
My MBO ships with Advanced ACPI HPET (IIRC, can't go in again now) set to disabled.
Anyway, I remember downloading Flash BIOS and trying to flash it with Rufus disk (IIRC), and it said:
Version (or similar word) mismatch. ...Not recommended...
And the version of my BIOS that it reported I did made myself to remember:
2.39
while available from Asrock is:
2.20
Oh, well
wcg wrote:
Also, I happened to go into BIOS setup again on the 990FX Extreme4
to see if the iommu was really enabled.

I found it under Northbridge, was disabled. and I enabled it.
wcg wrote:
I noticed that there are not
one but two separate "cpu overclock" menu selections, on the
same page of the BIOS setup display but in different sections,
more or less.
...[snip]...

I really have to leave out even modest overclocking for now, no time. Sticking with the defaults in the MBO as far as OC goes.
wcg wrote:
(I don't care about all that "virtual hot-rodding". I don't play
games on this motherboard, I don't count fps, etc. I simply
want it to run until I don't need it anymore. If I don't fall asleep
at the keyboard waiting for a kernel to compile, it is fast enough.)

Sure enough!
Anyway, I went through and have now a basic, or should I admit vague to rudimentary only, understanding of IOMMU, HPET, MSI and friends.
Back to top
View user's profile Send private message
miroR
Guru
Guru


Joined: 05 Mar 2008
Posts: 344

PostPosted: Thu Nov 08, 2012 12:34 am    Post subject: Reply with quote

miroR wrote:
wcg wrote:

I use this in /etc/portage/make.profile (my /etc/make.profile is now
a symbolic link to that file):
Code:

CHOST=x86_64-pc-linux-gnu
CFLAGS="-march=native -O2 -fno-strict-aliasing -pipe -ggdb -Wall"
CXXFLAGS="$CFLAGS -fpermissive"
LDFLAGS="$LDFLAGS -Wl,--as-needed"
MAKEOPTS="-j5"
FEATURES="splitdebug"

...
Currently, as I posted my flags here:
http://forums.gentoo.org/viewtopic-t-940916.html#7176772, without comments they are still:
Code:
CFLAGS="-march=k8 -O2 -pipe"
CXXFLAGS="${CFLAGS}"
CHOST="x86_64-pc-linux-gnu"

You have as in the beginning of this post, above, and I copy only differences from my flags (only one line is same, actually):
Code:

CFLAGS="-march=native -O2 -fno-strict-aliasing -pipe -ggdb -Wall"
CXXFLAGS="$CFLAGS -fpermissive"
LDFLAGS="$LDFLAGS -Wl,--as-needed"
MAKEOPTS="-j5"
FEATURES="splitdebug"

So if I just change your gdb containing line into:
Code:
CFLAGS="-march=native -O2 -fno-strict-aliasing -pipe -Wall"

and leave out:
Code:
FEATURES="splitdebug"
, would I be fine to use your setup.
I won't just go on and do it.
...

From:
http://forums.gentoo.org/viewtopic-p-7179816.html#7179816
I went on:
http://en.gentoo-wiki.com/wiki/Safe_Cflags
and then found rather simplistic suggestion:
http://en.gentoo-wiki.com/wiki/Safe_Cflags/AMD#Athlon_X2_7x50.2C_Phenom_X3.2FX4.2C_Phenom_II.2C_Athlon_II_X2.2FX3.2FX4.2C_Turion_II
Code:
CHOST="x86_64-pc-linux-gnu"
CFLAGS="-march=amdfam10 -O2 -pipe"
CXXFLAGS="${CFLAGS}"

I might go for it, if I don't learn better.
EDIT: Yeah. path of least resistance, the easy way out. Unless I get a good reasno, I'll go with the flow.
Waiting a few hours for replies, then adopting these flags and doing emerge -ea!
Back to top
View user's profile Send private message
wcg
Guru
Guru


Joined: 06 Jan 2009
Posts: 588

PostPosted: Thu Nov 08, 2012 5:38 pm    Post subject: Reply with quote

[-march=]
"-march=amdfam10" is what you should get from "-march=native"
with a Phemon II. If you do not do debugging with gdb, you can
leave out -ggdb and Features="splitdebug". (Did I post the gentoo
wiki link explaining these? Note that with splitdebug, you are still
running stripped binaries, so they should not be any slower to
load and run than binaries compiled without -ggdb.)

[mplayer]
You can use this command
Code:

emerge --info mplayer

to see what mplayer package version is installed, what USE
flags it was installed with, etc. All of that is in a line near
the end of the output.

mplayer-1.1-r1 is working fine here, but I don't do much
video conversion, etc, with it. Mostly I only use it to play
audio. I just opened an .mp4 video in smplayer, and it
displayed just fine. So I expect that mplayer-1.1-r1 should
work as well as the previous mplayer-1.0_rc4_p20110322-r1.
Let us hope that menconder in the new version works as well,
too.

[hpet]
I can disable hpet in the acpi options in BIOS setup? Other peoples'
comments lead to me to believe that hpet is no longer necessary
with chipsets and cpus this new (there are other high
precision timers available to the kernel). I will try disabling it
and see if anything breaks. From the description of MSI (PCIe hardware
interrupt protocol), PCIe MSI would be more valuable for performance
than hpet.

[other CFLAGS]
In CFLAGS, "-fno-strict-aliasing" is to keep gcc from using an optimization
that is unsafe with certain kinds of code. "-fpermissive" in CXXFLAGS
allows some older C++ code that would be an error with modern
g++ to only be a warning instead. (In the original C++ "standard",
certain things were defined to require the compiler to issue a warning.
Later standards for the language defined some of those same code forms
that earlier compilers issued warnings for to be errors instead, and
the compiler would stop compiling. "-fpermissive" in CXXFLAGS allows
C++ code using "the old way" to still compile with newer g++.)

[tomato]
Amuse your neighbors: their tomatoes have been introduced to the US.
http://t.tatianastomatobase.com/wiki/Brandywine_from_Croatia
_________________
TIA
Back to top
View user's profile Send private message
miroR
Guru
Guru


Joined: 05 Mar 2008
Posts: 344

PostPosted: Thu Nov 08, 2012 10:16 pm    Post subject: Reply with quote

wcg wrote:
[-march=]
"-march=amdfam10" is what you should get from "-march=native"
with a Phemon II. If you do not do debugging with gdb, you can
leave out -ggdb and Features="splitdebug". (Did I post the gentoo
wiki link explaining these? Note that with splitdebug, you are still
running stripped binaries, so they should not be any slower to
load and run than binaries compiled without -ggdb.)

Too late for me to use the advice.
I've only about 220 of 1289 packages left to install, since
Code:
# emerge -ea --keep-going --with-bdeps @world

(IIRC, I forgot "q", so what I issued can't see anymore, it's too verbose)
started a reinstall of 1289 packages.
And I went for the easy advice from the en.gentoo-wiki.com link I gave (I don't remember you giving one, but I have to move on to the issues still unsolved).
Taking me, all the 1289 packages, about, what, probably about 15 hours. Not much for that number of packages.
And with this new hardware setup the recoding with ffmpeg, as an example (and ffmpeg does wonders, and is probably fastest there is), takes about 5 times less than the old setup used to take me...

wcg wrote:
[mplayer]
You can use this command
Code:

emerge --info mplayer

to see what mplayer package version is installed, what USE
flags it was installed with, etc. All of that is in a line near
the end of the output.

mplayer-1.1-r1 is working fine here, but I don't do much
video conversion, etc, with it. Mostly I only use it to play
audio. I just opened an .mp4 video in smplayer, and it
displayed just fine. So I expect that mplayer-1.1-r1 should
work as well as the previous mplayer-1.0_rc4_p20110322-r1.
Let us hope that menconder in the new version works as well,
too.

I have precise issue with mencoder capturing on old Composite input the video (that is ok), while not capturing any sound from Line-in, as I described here:
http://forums.gentoo.org/viewtopic-t-940916.html#7175518
Suspicion was cast onto pulseaudio, but pulseaudio is working with ffmpeg...
Which (the ffmpeg) doesn't capture (or grab) video well...
I believe this is a more complex issue and that I might need to engage in different mailing lists, next best effort might be the pulseaudio list and get some developer to maybe tell us how to apply pulseaudio in mencoder capture...
In the next to that post (in this very thread that the kind reader is reading), in this one:
http://forums.gentoo.org/viewtopic-t-940916.html#7175644
I pointed to where I already asked for help on the issue.
The latest are these:
http://ffmpeg.org/pipermail/ffmpeg-user/2012-November/011092.html
(pls. read there the warning on clumsiness of the 15 minute video:
www.youtube.com/watch?v=M7AZ6el8AK4 )
I have just given myself extra work capturing video with mencoder and separately audio with ffmpeg, which is soo much more work, but did the work (that couldn't be posponed).

Now back to your advice... Risking to make all the thread a little harder to manage, I think it might be a good idea to ask you to look up, and bearing in mind the precise issue with mencoder not getiing audio through Line-in as it did for me for a few years...
With that in mind, is there anything there that could be the culprit to bring to justice fot my no audio issue in the following output?:

Code:
# emerge --info mplayer
Portage 2.1.11.31 (hardened/linux/amd64, gcc-4.6.3, glibc-2.16.0, 3.6.5-hardened-121103_1900 x86_64)
=================================================================
                        System Settings
=================================================================
System uname: Linux-3.6.5-hardened-121103_1900-x86_64-AMD_Phenom-tm-_II_X4_965_Processor-with-gentoo-2.2
Timestamp of tree: Thu, 08 Nov 2012 00:15:01 +0000
ld GNU ld (GNU Binutils) 2.23
app-shells/bash:          4.2_p39
dev-java/java-config:     2.1.12
dev-lang/python:          2.7.3-r2, 3.2.3-r1
dev-util/cmake:           2.8.9-r1
dev-util/pkgconfig:       0.27.1
sys-apps/baselayout:      2.2
sys-apps/openrc:          0.11.4
sys-apps/sandbox:         2.6
sys-devel/autoconf:       2.13, 2.69
sys-devel/automake:       1.9.6-r3, 1.11.6, 1.12.4
sys-devel/binutils:       2.23
sys-devel/gcc:            4.6.3
sys-devel/gcc-config:     1.7.3
sys-devel/libtool:        2.4.2
sys-devel/make:           3.82-r4
sys-kernel/linux-headers: 3.6 (virtual/os-headers)
sys-libs/glibc:           2.16.0
Repositories: gentoo multimedia sunrise zugaina vdr-devel vdr-testing seden udev x-portage x-my_ebuilds
ACCEPT_KEYWORDS="amd64 ~amd64"
ACCEPT_LICENSE="* -@EULA Oracle-BCLA-JavaSE"
CBUILD="x86_64-pc-linux-gnu"
CFLAGS="-march=amdfam10 -O2 -pipe"
CHOST="x86_64-pc-linux-gnu"
CONFIG_PROTECT="/etc /usr/share/config /usr/share/gnupg/qualified.txt /var/lib/hsqldb"
CONFIG_PROTECT_MASK="${EPREFIX}/etc/gconf /etc/ca-certificates.conf /etc/env.d /etc/fonts/fonts.conf /etc/gconf /etc/gentoo-release /etc/php/apache2-php5.4/ext-active/ /etc/php/cgi-php5.4/ext-active/ /etc/php/cli-php5.4/ext-active/ /etc/revdep-rebuild /etc/sandbox.d /etc/terminfo"
CXXFLAGS="-march=amdfam10 -O2 -pipe"
DISTDIR="/usr/portage/distfiles"
EMERGE_DEFAULT_OPTS="--autounmask-write=y"
FCFLAGS="-O2 -pipe"
FEATURES="assume-digests binpkg-logs clean-logs config-protect-if-modified distlocks ebuild-locks fixlafiles merge-sync news parallel-fetch protect-owned sandbox sfperms strict unknown-features-warn unmerge-logs unmerge-orphans userfetch"
FFLAGS="-O2 -pipe"
GENTOO_MIRRORS="http://ftp.gentoo.bg/ ftp://gentoo.mirror.web4u.cz/ http://mirror.opteamax.de/gentoo/ rsync://mirror.opteamax.de/gentoo/ ftp://91.121.124.139/gentoo-distfiles/"
LANG="en_GB.UTF-8"
LC_ALL="en_GB.UTF-8"
LDFLAGS="-Wl,-O1 -Wl,--as-needed"
LINGUAS="en_GB"
PKGDIR="/usr/portage/packages"
PORTAGE_CONFIGROOT="/"
PORTAGE_RSYNC_OPTS="--recursive --links --safe-links --perms --times --compress --force --whole-file --delete --stats --human-readable --timeout=180 --exclude=/distfiles --exclude=/local --exclude=/packages"
PORTAGE_TMPDIR="/dev/shm"
PORTDIR="/usr/portage"
PORTDIR_OVERLAY="/var/lib/layman/multimedia /var/lib/layman/sunrise /var/lib/layman/zugaina /var/lib/layman/vdr-devel /var/lib/layman/vdr-testing /var/lib/layman/seden /var/lib/layman/udev /var/lib/portage /var/lib/layman/my_ebuilds"
SYNC="rsync://rsync.europe.gentoo.org/gentoo-portage"
USE="acl amd64 berkdb bzip2 cli cracklib crypt cups cxx dri gdbm gpm hardened iconv ipv6 justify mmx modules mudflap multilib ncurses nls nptl openmp pam pax_kernel pcre pppd readline rule_generator session sse sse2 ssl tcpd unicode urandom zlib" ALSA_CARDS="ali5451 als4000 atiixp atiixp-modem bt87x ca0106 cmipci emu10k1x ens1370 ens1371 es1938 es1968 fm801 hda-intel intel8x0 intel8x0m maestro3 trident usb-audio via82xx via82xx-modem ymfpci" ALSA_PCM_PLUGINS="adpcm alaw asym copy dmix dshare dsnoop empty extplug file hooks iec958 ioplug ladspa lfloat linear meter mmap_emul mulaw multi null plug rate route share shm softvol" APACHE2_MODULES="authn_core authz_core socache_shmcb unixd actions alias auth_basic authn_alias authn_anon authn_dbm authn_default authn_file authz_dbm authz_default authz_groupfile authz_host authz_owner authz_user autoindex cache cgi cgid dav dav_fs dav_lock deflate dir disk_cache env expires ext_filter file_cache filter headers include info log_config logio mem_cache mime mime_magic negotiation rewrite setenvif speling status unique_id userdir usertrack vhost_alias" CALLIGRA_FEATURES="kexi words flow plan sheets stage tables krita karbon braindump" CAMERAS="ptp2" COLLECTD_PLUGINS="df interface irq load memory rrdtool swap syslog" ELIBC="glibc" GPSD_PROTOCOLS="ashtech aivdm earthmate evermore fv18 garmin garmintxt gpsclock itrax mtk3301 nmea ntrip navcom oceanserver oldstyle oncore rtcm104v2 rtcm104v3 sirf superstar2 timing tsip tripmate tnt ubx" GRUB_PLATFORMS="pc multiboot" INPUT_DEVICES="evdev" KERNEL="linux" LCD_DEVICES="bayrad cfontz cfontz633 glk hd44780 lb216 lcdm001 mtxorb ncurses text" LIBREOFFICE_EXTENSIONS="presenter-console presenter-minimizer" LINGUAS="en_GB" LIRC_DEVICES="devinput" PHP_TARGETS="php5-3" PYTHON_TARGETS="python3_2 python2_7" RUBY_TARGETS="ruby18 ruby19" USERLAND="GNU" VIDEO_CARDS="radeon" XTABLES_ADDONS="quota2 psd pknock lscan length2 ipv4options ipset ipp2p iface geoip fuzzy condition tee tarpit sysrq steal rawnat logmark ipmark dhcpmac delude chaos account"
Unset:  CPPFLAGS, CTARGET, INSTALL_MASK, MAKEOPTS, PORTAGE_BUNZIP2_COMMAND, PORTAGE_COMPRESS, PORTAGE_COMPRESS_FLAGS, PORTAGE_RSYNC_EXTRA_OPTS, USE_PYTHON

=================================================================
                        Package Settings
=================================================================

media-video/mplayer-1.1-r1 was built with the following:
USE="3dnow 3dnowext X a52 aalib alsa cdio directfb doc dts dv dvb dvd dvdnav enca encode faac faad fbcon iconv ipv6 jpeg jpeg2k libass libcaca libmpeg2 live mad mmx mmxext mp3 (multilib) network opengl osdmenu png pnm pulseaudio quicktime rar real rtc sdl shm speex sse sse2 theora toolame tremor truetype twolame unicode v4l vorbis x264 xanim xinerama xscreensaver xv xvid (-altivec) (-aqua) -bidi -bindist -bl -bluray -bs2b -cddb -cdparanoia -cpudetection -debug -dga -dxr3 -ftp -ggi -gif -gsm -jack -joystick -ladspa -lirc -lzo -md5sum -mng -nas -nut -openal -oss -pvr -radio -rtmp -samba -ssse3 -tga -vdpau (-vidix) (-win32codecs) -xvmc -zoran" VIDEO_CARDS="-mga -s3virge -tdfx"
CFLAGS=""
CXXFLAGS="-march=k8 -O2 -pipe"

I find both alsa and pulseadio there under Package Settings...
OTOH, mplayer is one of the packages not yet updated (CXXFLAGS="-march=k8 -O2 -pipe" there)... The problem might not go away just like that with the update... I'm afraid.
wcg wrote:
[hpet]
I can disable hpet in the acpi options in BIOS setup? Other peoples'
comments lead to me to believe that hpet is no longer necessary
with chipsets and cpus this new (there are other high
precision timers available to the kernel). I will try disabling it
and see if anything breaks. From the description of MSI (PCIe hardware
interrupt protocol), PCIe MSI would be more valuable for performance
than hpet.

[other CFLAGS]
In CFLAGS, "-fno-strict-aliasing" is to keep gcc from using an optimization
that is unsafe with certain kinds of code. "-fpermissive" in CXXFLAGS
allows some older C++ code that would be an error with modern
g++ to only be a warning instead. (In the original C++ "standard",
certain things were defined to require the compiler to issue a warning.
Later standards for the language defined some of those same code forms
that earlier compilers issued warnings for to be errors instead, and
the compiler would stop compiling. "-fpermissive" in CXXFLAGS allows
C++ code using "the old way" to still compile with newer g++.)

Fine! Will bear these in mind.

And this below, it is cute, sure, is that somewhere near where you live?
wcg wrote:
[tomato]
Amuse your neighbors: their tomatoes have been introduced to the US.
http://t.tatianastomatobase.com/wiki/Brandywine_from_Croatia


Last edited by miroR on Thu Nov 08, 2012 10:22 pm; edited 1 time in total
Back to top
View user's profile Send private message
Tony0945
Apprentice
Apprentice


Joined: 25 Jul 2006
Posts: 276

PostPosted: Thu Nov 08, 2012 10:21 pm    Post subject: Reply with quote

My abbreviated make.conf
Code:
CFLAGS="-march=native -O2 -pipe"
CHOST="x86_64-pc-linux-gnu"
CXXFLAGS="${CFLAGS}"
MAKEOPTS="-j7"

PORTAGE_ELOG_CLASSES="info error log"
PORTAGE_ELOG_SYSTEM="save"

PORTDIR_OVERLAY="/usr/local/portage"



#ACCEPT_KEYWORDS="amd64"

LINGUAS="en en_US"

# NOTE:  use flag eds below is evolution-data-server not esd (esound)
USE=" alsa apm applet audio audiofile automount apache2 dvb cairo consolekit \
     cdr cdrom cdsound corba cpudetection dbus dirac dv dvb dvd dvdr \
     elf esd fontconfig fortran fortran95 ftp jpeg gcc64  gnome guile gnuplot g$
     gstreamer010 gtkhtml gtkspell gtk gtkmm -hal hpcups imagemagick lame mad m$
     mpm-prefork mysql\
     nautilus nfs no-old-linux no-seamonkey nptl nptlonly numeric nvtv -sandbox\
     opengl openntpd policykit png ppds perl python recode reiserfs samba\
     scanner schroedinger silvercity sndfile \
     -sqlite \
     sockets suid sysfs syslog svg thunderbird tk v4l2 vdr vfat X wxwindows \
     win32codecs \
     xine -kde -nls -pppd -quicktime -qt3 -qt4 bitmap-fonts \
     -bluetooth -eds -epiphany -evolution -eds -ipv6 -krb5  -ldap -swig\
     -gnome-keyring -xulrunner -ruby -static-libs \
     truetype truetype-fonts type1-fonts sse sse2 sse3 3dnow 3dnowext mmx mmxex$
     udev"

CONFIG_PROTECT_MASK="/etc/services /etc/make.conf.example"
FEATURES="-ccache -distcc -sandbox parallel-fetch"


The rest is hardware and locale specific.
I probably have way too many use flags. I wouldn't advise anyone to blindly copy them.
You won't be using "-j7" because I have an X6 an yours is X3(?)
You will have different LINGUAS too I'm sure.
The CFLAGS is simple but I've used it for 6 years. Of course with a specific -march before -march=native was added to compiler.
And I used -Os instead of -O2 in the old days of small memory. My understanding is that -Os is not longer recommended.
Back to top
View user's profile Send private message
miroR
Guru
Guru


Joined: 05 Mar 2008
Posts: 344

PostPosted: Thu Nov 08, 2012 10:34 pm    Post subject: Reply with quote

Tony0945 wrote:
My abbreviated make.conf
Code:
CFLAGS="-march=native -O2 -pipe"
CHOST="x86_64-pc-linux-gnu"
CXXFLAGS="${CFLAGS}"
MAKEOPTS="-j7"

PORTAGE_ELOG_CLASSES="info error log"
PORTAGE_ELOG_SYSTEM="save"

PORTDIR_OVERLAY="/usr/local/portage"



#ACCEPT_KEYWORDS="amd64"

LINGUAS="en en_US"

# NOTE:  use flag eds below is evolution-data-server not esd (esound)
USE=" alsa apm applet audio audiofile automount apache2 dvb cairo consolekit \
     cdr cdrom cdsound corba cpudetection dbus dirac dv dvb dvd dvdr \
     elf esd fontconfig fortran fortran95 ftp jpeg gcc64  gnome guile gnuplot g$
     gstreamer010 gtkhtml gtkspell gtk gtkmm -hal hpcups imagemagick lame mad m$
     mpm-prefork mysql\
     nautilus nfs no-old-linux no-seamonkey nptl nptlonly numeric nvtv -sandbox\
     opengl openntpd policykit png ppds perl python recode reiserfs samba\
     scanner schroedinger silvercity sndfile \
     -sqlite \
     sockets suid sysfs syslog svg thunderbird tk v4l2 vdr vfat X wxwindows \
     win32codecs \
     xine -kde -nls -pppd -quicktime -qt3 -qt4 bitmap-fonts \
     -bluetooth -eds -epiphany -evolution -eds -ipv6 -krb5  -ldap -swig\
     -gnome-keyring -xulrunner -ruby -static-libs \
     truetype truetype-fonts type1-fonts sse sse2 sse3 3dnow 3dnowext mmx mmxex$
     udev"

CONFIG_PROTECT_MASK="/etc/services /etc/make.conf.example"
FEATURES="-ccache -distcc -sandbox parallel-fetch"


The rest is hardware and locale specific.
I probably have way too many use flags. I wouldn't advise anyone to blindly copy them.
You won't be using "-j7" because I have an X6 an yours is X3(?)
You will have different LINGUAS too I'm sure.
The CFLAGS is simple but I've used it for 6 years. Of course with a specific -march before -march=native was added to compiler.
And I used -Os instead of -O2 in the old days of small memory. My understanding is that -Os is not longer recommended.

I see.
You say: I have an X6 an yours is X3(?)
I don't know. I got it all mixed up a little. I think it's X4. It does have four cores, the processors are four on one die... It's Phenom 965 II Deneb(?)
I'll familiarize with my hardware with time. :)
And don't forget I get so clumsy to unstuck cables on a running system...
It may be rare to happen (it doesn't happen to me often), but only mdadm perfection saved me! Like I describe in the first post in the last 24 hours or so, a little way back from here.
So my thanks go to Neil Brown and his crew!
On I go now to resolve the mencoder issue.
Because as I was able to see, I think mencoder is the way to go to capture on old S-video inputs from either cable provider channels or VHS old home video equipment.
Only it don't do the pulseadio well.
Back to top
View user's profile Send private message
Tony0945
Apprentice
Apprentice


Joined: 25 Jul 2006
Posts: 276

PostPosted: Fri Nov 09, 2012 1:22 am    Post subject: Reply with quote

I took pulseaudio out of my system. It seems to work better that way.

Sorry for the Xn mix-up. I thought you had decided to buy an X3. Whatever n is in Xn, use -j=n+1. After doing the arithmetic, of course.

Don't worry if emerge -ea world takes about four hours. You and I have a lot of packages. You might just want to emerge your performance critical apps and let the others wait on updates to be optimized. It's a 50-50 choice.
Back to top
View user's profile Send private message
miroR
Guru
Guru


Joined: 05 Mar 2008
Posts: 344

PostPosted: Fri Nov 09, 2012 1:43 am    Post subject: Reply with quote

Tony0945 wrote:

My abbreviated make.conf
Code:
...[snip]...
#ACCEPT_KEYWORDS="amd64"
...[snip]...




If you didn't leave out the tilde, the curly character "~", like so:

Code:
ACCEPT_KEYWORDS="~amd64"


then you needn't comment it, because it would give you the testing branch, just like I am on.

Poor eyesight. I notice I make typoes for the same reasons too...

OTOH, with emerge -ea , just another hour or two, it's not a problem.

See, there I now truly pursue the mencoder issue with a lot of hope.

The thread is by subject:

[MEncoder-users] With time honoured always working alsa settings no more audio with mencoder...

and the address of the post I wrote a while ago is:

http://lists.mplayerhq.hu/pipermail/mencoder-users/2012-November/013229.html

And now I think I'll search the web a little, and, depending on what I find do something else or go and ask for more help on pulseaudio mailing list... I think.
EDIT: Yes, I just did that:
http://lists.freedesktop.org/archives/pulseaudio-discuss/2012-November/015238.html
Back to top
View user's profile Send private message
Tony0945
Apprentice
Apprentice


Joined: 25 Jul 2006
Posts: 276

PostPosted: Fri Nov 09, 2012 10:45 am    Post subject: Reply with quote

ACCEPT_KEYWORDS is commented out because it is unnecessary. I run stable branch as much as possible. Some applications, however, do not have a stable branch, so specific versions are listed in /usr/portage/keywords.
Goo luck on your project. Be sure to post your eventual success.
Back to top
View user's profile Send private message
miroR
Guru
Guru


Joined: 05 Mar 2008
Posts: 344

PostPosted: Fri Nov 09, 2012 11:57 pm    Post subject: Reply with quote

Tony0945 wrote:
ACCEPT_KEYWORDS is commented out because it is unnecessary. I run stable branch as much as possible. Some applications, however, do not have a stable branch, so specific versions are listed in /usr/portage/keywords.
Goo luck on your project. Be sure to post your eventual success.

Thanx, Tony0945!
But there is a missing "~" 8)
Never mind.
I hit another snag.
I was also successful in my quest to fix my mencoder profiles.
The snag is being solved (could be HPET or IOMMU related, really can't tell yet).
And the solution is with the quest on mencoder Composite input (on my TV-card, for video) and Line-in (for audio) capture...
My problem was addressed with patience by Tanu Kaskinen, one of the four creators of pulseaudio, and I feel honored https://en.wikipedia.org/wiki/PulseAudio.
Go read, dear brother, the Finnish can understand Croats:
http://lists.freedesktop.org/archives/pulseaudio-discuss/2012-November/015243.html
Go read, dear brother, the Finnish can understand Croats, because they had similar issues in the historical tempest that made the current world order in which an American President, and a, I dare say, treacherous German Premier, Obama and Merkel, trample over mass graves of innocent civilians of Dresden.
I remember having watched Obama (the lier on any changes whatsoever), in visit to Germany, with Merkel, the truly-Soviet-in-the-core neocommunist, swaggering in and around Dresden, and visiting some minor concentration camp in the vicinity...
Honoring the Allies, and the few victims of the Nazi terror, but on a site that saw about at least two hundred thousand non-military victims in 1945.
And not a word, not a memory, not a memento for those two hundred thousand.
Who couldn't have in any grotesque possibility been all nazi.
Who were just civilians running for their lives.
May God reckon with both of you, stupid leaders!
That was an event that happened in the Corporate and Soviet European Union maybe some three to almost four years ago, I think at the beginning of the lier's (on change, not because is black, pls. take notice), at the beginning of this killer of children (yea, support for abortion is getting absolute in the USA, never yet was!) Obama (a bad black man, not a good among the colored man, he really is no Martin Luther King; he is actually candid as some Black people from USA like to call him, he's candid they say, he's not really black, so my despise of him is not racial, pls. take into consideration).
It was a swaggering Dresden walk about three to four years ago. And I watched in dismay on Al Jazeera English TV back then when it was a model News station in in its last years or so, was already going sour actually), and it stuck with me.
Croatian people, as well as the Finnish, were on the "wrong" side in WWII, because we were only fighting for our own freedom, and I hope to God that the Finnish have a better life than we in Croatia have, even though it is not yet that bad, and I can see in truthful and persecuted Intl TV's like Iran's International PressTV, where so many from among the best of the USA's dissenters often feature as guests, how bad things are going in the USA also...
Back to top
View user's profile Send private message
miroR
Guru
Guru


Joined: 05 Mar 2008
Posts: 344

PostPosted: Sat Nov 10, 2012 2:38 am    Post subject: Reply with quote

No, doesn't have to be IOMMU related.
On two MBO's (reminder: both MBO same model, on both same hardware, very few differences), one working, onw not. both have IOMMU disabled.
It may not be HPET related either.
On two MBO's, one working, onw not. both have IOMMU enabled.

The one working, all is well.
The one not working, one of the first failures that I see is:

This is making rounds on a KVM switch (four input computers' keybord-video-mouse on one controlling keybord and one controlling mouse and one controlling monitor)... and copying (is there a better way?)... I mean if I touch it the messages go on and, sure enough, away...

Code:
Setting system clock using the hardware clock [UTC]
hwclock: Cannot access the hardware clock with any known method
hwclock: Use the --debug option to see the details of our search for an access method
Failed to set the system clock           [!!!]
...[snip]...

Starting up RAID               [!!!]
Setting up the Logical Volume Manager
Failed to find the sysfs mount point
No volume groups found.
(((...enumerates the volumes here... none found )))
Checking local filesystems
Unable to resolve ext4 [UUID="here good UUID from /etc/mdadm.conf")

Remounting root filesystem read/write

Root filesystem could not be mounted read/write
((( and on from here you can faily well imagine that nothing really works on a root that can not be really mounted )))


I sure could have gone for the "hwclock: Use the --debug" (where is that to be set anyway), but I think that is just the symptom, not the cause.
Actually, I believe that my keyboard has succumbed to my heavy blows and given up bearing up with me.
If only Gentoo Forums bear on with me, that will be great, but I'm fearing all kinds from dear administrators, since I seem to be already (tell me what else it is?), partially censored.
Namely, as I wrote in reply to Tanu Kaskinen's mail who gave such patient and kind advice to me for my issue with mencoder and pulseaudio (of which he is one of the principal developer), on the pulseaudio mailing list:
http://lists.freedesktop.org/archives/pulseaudio-discuss/2012-November/015250.html
actually:
http://lists.freedesktop.org/archives/pulseaudio-discuss/2012-November/015249.html
and I wrote this:
Quote:
and Gentoo forums
- -- where I seem to be starting to be censored as well:

https://www.youtube.com/watch?v=JExAwgsUuLo
, "Screen 2012_11_03, my posts on Gentoo forums",

mostly for the following post, someone get that thread archived
somewhere!:

http://forums.gentoo.org/viewtopic-t-905472.html ,
"System attacked, Konqueror went on window-popping spree!",

which is about NSA, USA's secret services' institution,

getting a hold in GNU Linux,

get me serious folks, please!

Because GNU Linux is sacred and if the Gentoo Linux, which is still
just about the best there is in free software, goes to the "ruling
elite", because it's not about the USA's people, but about the shadow
rulers, we, the good of the World, have lost a lot!)


Dear administrators, you get an apology if that is not purposefully done on your behalf!

Sure, I can only believe it was not done on purpose if I can from now on, again see all my posts, as I always was able to...

In which case: Thank you!

If only Gentoo Forums bear on with me, not for my being wrong, no! But the time is tough and it is not easy to be truthful.

The keyboard has not born any longer with me. It died.
And it might have provoked a shock through a short circuit on the MBO's it was attached to, of which one got a more severe blow than the other three.
Because all the keyboads are attached to the MBO in the same fashion.

Skip or search at around some 30 or more lines back where I explain my KVM switch.
Like in other occasions when I get a minor or major shock or stress, let me see an example (so you can also get a glimpse of what read you might be missing if I get banned or hidden from ordinary access on Gentoo Forums)... I'm going to search with Tor now and see how much Google is willing to tell about me on a particular issue...
The search is "Rovis sysresccd gentoo".
And Tor sure searches Google through https://startpage.com/do/search.
It does return at least some good results:
https://www.youtube.com/watch?v=CdEOKGP-Ftw
http://vimeo.com/48282458
http://forums.gentoo.org/viewtopic-p-7157330.html?sid=0b8603798b5325fe475573fc4e062e1f
I'm not rummaging these now.
I only wanted to say that just like I was confused back then when François Dupoux banned me from his site, so I was this time with my keyboard that died and made a mess on my clones.
I really wasn't able to boot with none of my system at first!
Then they slowly recovered, and one of the four hasn't yet.
It is being recovered with --force :-)
I mean, it's actually the one that two of the others were cloned from! Completely they have its content. Just the modifications for network and some other little thingies.
Now it is being restored, because it is the same dd'd files that are used to clone and to restore.
I'm able to clone a system in less than an hour, with this hardware.
dd cloning explained earlier in this topic, actually link given to explanation.
In an old topic by others, in reply to 4est, the topic is something "...libselinux... system horked" ...
Trying to recollect what happened...
At a particular time, and I don't recall what the immediate cause was, I haven't delivered any blows onto the table or keyboard in recent days actually, last time I got really frustrated and angry was when I unstuck the SATA cable on a degraded SATA HDD, which I also explained earlier... That was about four or five days ago.
At a particular time the keyboard wouldn't respond to switching control from one onto another of the four computers connected to the KVM switch.
I kept trying, but the light on the KVM just blinked, and none of the system I could select to control...
Since I couldn't control any of the systems, instead of stopping and thinking more slowly (even though it wouldn't help anyway had I decided to do differently, actually I did the right move, since I wasn't, and am still not certain completely that the keyboard is dead, only the likelyhood appear very very probable), I went to unstuck the keyboard female connector from the KVM switch (that's the one keyboard which is in front of me all the time, and that through KVM switch controls the selected conputer of the four connected to KVM switch)...
And I connected the female connector of the probably short circuited keyboard onto one of the systems...
Hey wait a minute... I thought it wasn't related... I thought it was...
But wait an intermeriary wait first.
My assumption was right.
The one system that the other two (it's the one system that I did, as my friend Tony0945 advised my, emerge -ea world on, and reinstalled 1289 packages in about 15 to 20, less than 20 certainly hours, with the safe CFLAGS (this time I didn't go for the more interesting flags that the kind friend wcg gave me advice on, no time)...
That one system is now up and running.
(P.S. Just a note. Surely I did attach and am writing all this with a spare and working other keyboard attached to KVM switch)
That exact one system is which I gave you the "... Failed to set the system clock..." and the rest, manually copied while making rounds on my KVM switch...
So I must have been mostly right on the keyboard's death.
Now you waited the intermediary wait.
The previous wait now.
The system that got the second shock from the dying short circuited keyboard (like if I get a stroke, it won't necessarily be from the immediate blows, just like Mohammed Ali, a fine Muslim and a fine American -- I am Catholic to the core, but respect all religions, and you good Americans, stop your country bashing Muslims for once! -- didn't get his Alzheimers but only later on... But I was saying, if I get a stroke, it will be from the years that I received bashing and blows, when I trained some martial arts a little, and also lived in underworld for a few years, no longer the case anymore since about 15 or 20 years now, completely clean I am now; wait, I don't want to say I was good at martial arts, only was foolishly stubborn, get me out of comparison with Ali other than the blows received, and that only as completely frail constitution mine being in comparison with his)...
But I was saying, the system that got the second shock from the dying short circuited keyboard long after the blows, just like me or Mohammed Ali...
The keyboard did give the first strong shock on the system that got some certain impact, and that is now restored. I am yet to know if my data (I keep data comppletely saparate from /root ) is damaged, but it doesn't look like it was.
But it also gave a second minor shock onto the other system that is not yet fully restored, and on which I suspected other reasons for not fully restoring.
On that other system I suspected the kernel 3.6.6-hardened) the reason for X windows not starting (because the cloned files were on Turks Sapphire 6570 system while that system has about 5 year old Radeon 1650 or somesuch... Unimportant, because...
I mean, above I said that only one system was affected, but I may not have been right.
I'll be able to tell you if I restore that system if I was right or not.

Enough talking now.

My thanks go to God for His help!

I hope I made sufficiently comprehensible reading. Remember that English is not my mothertongue. I do listen to English daily, because I follow world events, but still...
Allow later corrections, but in follow-ups if need be.
Namely: get this for absolute certainty:
I don't intend in my right mind and in the absense of duress, to delete any whatsoever of my posts that I posted anywhere on Gentooo Forums!
So if any will be missing, I didn't delete them!
For similar reason, I won't be editing what I wrote too much later either. I'll rather be writing follow-ups.
Back to top
View user's profile Send private message
miroR
Guru
Guru


Joined: 05 Mar 2008
Posts: 344

PostPosted: Sat Nov 10, 2012 5:30 am    Post subject: Reply with quote

No, there was just one short circuit by the keyboard.
And the previous assumption was correct (on the MBO with the old Radeon).
This got me back into X:
http://forums.gentoo.org/viewtopic-t-924726.html#7047332
And I'm tired.
Good night!
Back to top
View user's profile Send private message
wcg
Guru
Guru


Joined: 06 Jan 2009
Posts: 588

PostPosted: Mon Nov 12, 2012 2:38 am    Post subject: Reply with quote

[mplayer, menconder]
I think your best resource would be reading mplayer-specifc user
forums and mailing lists.

[NSA]
I doubt that the NSA is your personal enemy. The NSA is really
part of the military defense establishment in the US. Their responsibility
is "signals intelligence". They listen to all sorts of electronic signals:
radio, television, internet backbone traffic, etc. But only things related
to national defense and the military are really interesting to them.
Everything else is noise, from their point of view. They are not going
to spend any of their expensive resources on issues unrelated
to military security.

Are you stealing data from a US company designing or building
hardware or software for the US military and sending that data
to some foreign country? The NSA is definitely interested. Are
you collaborating in some way with "international terrorist" groups
(a vague term there, "international terrorist")? The NSA may be
interested. Are you making bombs in your basement for heroin
smugglers? The NSA is *not* interested in electronic signals traffic
to or from you, because investigating heroin smuggling is someone
else's job.

If you are only objecting to US foreign policy, US economic policy,
and other issues that are basically political, that does not make you
interesting to the NSA.

If your systems get cracked, it is likely some criminal looking for
credit card numbers, passwords to bank accounts and investment
accounts, and assets of that nature rather than agents of a foreign
government that disagree with your political views.

edit:
As far as Gentoo security, the NSA would not be interested in
Gentoo any more than any other linux distribution. They would be far
more interested in breaking into routers, gateway firewalls, etc.,
than into backend servers and workstations. Finding out what data
is passing through routers fits their job description. (Spending agency
resources on something not in your job description is a fast way
to become unemployed in the US.)

The NSA's job is not political, it is operational. If the prime minister
of some government sends an email to his neighbor that says,
"What the US is doing is bad, we should oppose it," the NSA does not
care about that. If that same prime minister calls a general and tells
him to move an army closer to a border, the NSA does care about that.
That is the sort of information that their operations are designed to
to listen for and report.
_________________
TIA
Back to top
View user's profile Send private message
wcg
Guru
Guru


Joined: 06 Jan 2009
Posts: 588

PostPosted: Tue Nov 13, 2012 10:27 pm    Post subject: Reply with quote

[keyboard, kvm switch]

So you have a keyboard, mouse, and monitor plugged into a kvm
switch. You have outlets to 4 computers. 3 of them work correctly
and one does not.

The solution is to unplug the keyboard, mouse, and video outputs
from the computer that does not work correctly. Then unplug
the keyboard, mouse, and video outputs from one of the computers
that works correctly and plug them into the computer that has
the problem.

If that computer now works correctly, then what is broken is the kvm
switch (one or more of it's outputs is non-functional). If that computer
still does not work, then what is broken is that computer.

(It might only be partly broken. If the keyboard does not work, but
the video and mouse work, try plugging the keyboard outlet from
the kvm switch into the mouse port on the computer and leave
the mouse unplugged. If you have USB instead of PS/2
peripherals, you can use a different USB port. etc.)
_________________
TIA
Back to top
View user's profile Send private message
miroR
Guru
Guru


Joined: 05 Mar 2008
Posts: 344

PostPosted: Wed Nov 14, 2012 11:54 pm    Post subject: Reply with quote

Hi, wcg!

I admire your love of your country for reasons of which you disagree with me. But it's deeper than you see it.

https://www.eff.org/nsa-spying

...I was just right in my assumption way back, and I now need to find and link...

http://forums.gentoo.org/viewtopic-t-905472.html#7125428
and stuff on Surveillance-ready Linux?...

I can't now analyze more in detail...

I have just survived more shocks...:
http://lists.claws-mail.org/pipermail/users/2012-November/004212.html
(by the title "[Users] Claws not downloading mail (how do I check and prove the matter?"), last one:
http://lists.claws-mail.org/pipermail/users/2012-November/004254.html
where:
Iskon 5, why Claws-mail appeared as it did?
http://www.youtube.com/watch?v=cPcFj95FovM
is the newest.

I don't blame NSA for that last bit above!

I blame'em for having sold onto non-GNU espousing devs or manager or you-name-'em, who are well positioned in some of the GNU Linux distro's (or is it buying, and not selling, to place it?, who can tell? are we free to have our doubts?)...

I blame'em for there now being something that almost broke (find it in the grsecurity author's, Brad Spender Spengler's pages...) the GNU licence. The SELinux is the one. Out, I say! GNU Linux is sacred! Boycott it, people!
Also, go and investigate... I feel saddened for not having any time left any more, since this four-times hardware update took all my strength, and GNU Linux always gives you new challenges...

Also, go and investigate... SELinux was introduced to the public as RedHat's program, Red Hat devs introduced it as if they were the creators of it... They really did try to offer it to the public as somebody else's... Why?

These things you won't find on:
"System attacked, Konqueror went on window-popping spree!"
http://forums.gentoo.org/viewtopic-t-905472.html

So I brought them up again (I wrote them when I discovered them, somewhere else in these forums).
It would be wrong to repeat myself here. No other new notions about that topic in my awareness.

The system (the one with the bad boot full of errors) I fixed, I believe, on my right assumption, and I reported back. It was so broken that it wouldn't start most of the services, and it is, just as the other three systems, now essentially equally of impeccable performance. after cloning it back, or in this case, I should call it restoring from backup. (OTOH, I still fear using that keyboard!)

You didn't tell me if those Croatian Serbs are near your place (I only have issues with the Greater-Serbs --the expansionists against my territory and nation, and these neighbors of yours seem fine).

I now have to go back to solving other issues.

I couldn't report back prior to this time, but people, Gentoo GNU Linux is a great tool, a paramount OS, and I'm glad if this topic that is probably nearing it's closure has been of use to especially newbies, and wish so much I could, as user of intermediate level, still be of some benefit to the community.

God bless!

Miroslav Rovis
m.rovis@inet.hr
miro.rovis@gmail.com
Zagreb, Croatia
#=== was sentenced to jail by my neocommunist regime: ===#
http://www.croatiafidelis.hr/Miroslav_Rovis_politicki_progon/
#=== Sorry! no English there at this time ===#
http://www.exDeo.com #some English
http://groups.google.com/group/croatian-news/
http://www.youtube.com/user/miroR2 #some English
http://vimeo.com/user9621785 #some English
Back to top
View user's profile Send private message
wcg
Guru
Guru


Joined: 06 Jan 2009
Posts: 588

PostPosted: Fri Nov 16, 2012 5:06 pm    Post subject: Reply with quote

[NSA]
I simply think one should not look for outside interference as the source
of problems that are probably only technical problems (misconfigured
hardware, software). I am familiar with the reports of illegal spying on
civilian communications. The NSA itself, however, is only an electronic
ear of the US military. The information that military is most interested
in is what *other country's militaries* are doing.

Imagine a tank sitting on a hill somwhere. The tank has a radar
antenna on top of it. The person in the tank is on guard duty,
watching the radar display inside the tank. The NSA is like that
radar antenna, only it is not watching a single potential battlefield,
it is watching the whole planet.

Few users fit that category, "other country's militaries", so there is little
incentive for the NSA to pay the costs in programming, debugging, and
so on to subvert Gentoo's security. The payoff in terms of military security
is not likely worth the cost in man-hours to do it (and to maintain it
as the kernel and userspace programs evolve).

(There was a lot of hysterical behavior in the government security
agencies after the September 11, 2001 attack on the Twin Towers.
That was doubtless how most of these programs of illegal spying
got started: people in the government imagining Islamic terrorists
hiding behind every bush and tree on the street outside their
offices.)

I would be more worried about "government spyware" programs leaking
to the private sector, to *corporate* security contractors, than I would
be worried about what the NSA itself is doing with the information.

[Selinux]
There have been conflicts with the licensing on some parts of the
selinux code and FSF licensing from the beginning of the selinux
project. Some programmers pointed out years ago that where that
licensing is a problem for someone, they are free to rip out the parts
of the selinux code with restrictive licensing and roll their own
replacement code for it.

[Balkans]
I am elsewhere. FYI, though, the average person's opinion in the US
is that lining up whole villages of civilians, anywhere, and mowing
them down with machine-guns is a crime.
_________________
TIA
Back to top
View user's profile Send private message
miroR
Guru
Guru


Joined: 05 Mar 2008
Posts: 344

PostPosted: Wed Nov 21, 2012 2:36 pm    Post subject: Reply with quote

wcg wrote:
[NSA]
I simply think ...

The NSA-spying is the link by:
https://www.eff.org/nsa-spying

wcg wrote:
[Selinux]
There have been conflicts with the licensing on some parts of the
selinux code and FSF licensing from the beginning of the selinux
project. Some programmers pointed out years ago that where that
licensing is a problem for someone, they are free to rip out the parts
of the selinux code with restrictive licensing and roll their own
replacement code for it.


While other devs...? What did the others say?
But the more important of my points you just don't seem to like to take into acount.

wcg wrote:
[Balkans]
I am elsewhere. FYI, though, the average person's opinion in the US
is that lining up whole villages of civilians, anywhere, and mowing
them down with machine-guns is a crime.


Did the Haag International Court Tribunal for former Yugoslavia hold us guilty as we stood accused, wrongly, for about one decade before the international community, in the final verdict delivered just back when you wrote this, on the 16th of November 2012?
No, but the State of Croatia, accused through it's top generals, was completely acquitted of state organized and aided crimes, by the Haag International Court Tribunal.
These what you wrote were absolutely not the usual ways Croatian Army dealt with civilians (and no Nation is comprised of solely Saints, less than others some extremely powerful nations of the day, and crimes simply are virtually unavoidable, not entirely, no!, in times of war, I guess everyone will agree... You can prevent some, but never all crimes in a theater of war)...
But unfortunately there was genocide, the worst, the most numerous, the most gory and heinous, there was genocide perpetrated on the Muslim population (Muslims of Bosnia are of the same provenance as the Catholics, same blood, same, truly, same nation, although now they partly declare themselves a separate nation, Bosniac nation, and that was the true brotherly conflict, btwn us Catholics and them Muslims...
the Serbs were the agressors in the war in the Balkans in 1990s, just like they were the hegemon with their dictatorial regimes aided by traitors from non-Serb nations of Yugoslavia, plundering wealth from those non-Serb nations, the most badly hurt by their hated and unjust domination being the Albanians and the Croats, in the most of the 20th century.
The Srebrenica genocide, and the Vukovar massacre are what kids should be learning about in history books for centuries to come.
I am always sad to hear of crimes commited by my connationals and there is here and there some largely unwarranted hatred of Serbs in Croatia, unwarranted because neither man can be guilty for simply being Rohingya (those that Obama couldn't give an honest word for in Burma/Myanmar these days, did he?) or Basque or German (say in WWII, when some of the Germans were found guilty even when they truly were not) or American or Serb or Croat, or Black or White or Redskin...
But the crimes commited by my connationals fade away in comparison to weeks of killing in Srebrenica and about a week of killing in Vukovar.
Every single crime is gruesome and no crime is ever acceptable, but it is the Serbs who perpetrated the majority of crimes in Croatia (historically) or Bosnia and Herzegovina and Croatia as they have their Yugoslav frontiers nowadays, which gave them their Tito, the Traitor and Slaughterer of Croats after, attention, after the conflicts of WWII were over, Tito the Slaughterer of Croats...
Again, it is the Serbs who perpetrated the majority of crimes in historical Croatian territories in the Balkan war of 1990s. The absolute majority of crimes in historical Croatian territories in 1990s were done by the agressors, the Serbs.
I want to emphasize here that there is in among the Serbian nation, a strain of those who I do, and always will hail as honorable and noble, who detested the crimes, tried to prevent those crimes, but, alas, those are far too difficult to prevent in times of war, so their efforts were mostly unsuccessful.
Similarly on our side, where much smaller was the extent of the atrocities.
Croatia holds acquitted of war crimes at the International Tribunal for the former Yugoslavia, having general Ante Gotovina and general Mladen Markač had their unjust first degree sentences overturned, which is nearly certain to be final at the Haague around Operation Storm and the Croatia's presumed guilt.
My thanks go to God and to the thankfully sufficient number of honest judges for the justice to be delivered at the Hague. The Jew Theodore Meron who was the President of the Court bestows to his Israeli Nation some justice, sadly too little in comparison to the heinous and inhumane killing of civilians in Gaza these daye (I cried many times seeing those kids slain in Gaza, of course, I hardly could have seen that true face of Israeli genocidal agression on BBC or CNN, but I luckily am able to watch PressTV Inernational TV, Iran based, but so honest in comparison to the others mentioned a few words back in this paragraph).
Here you can see our Generals coming back to Croatia. English resume will be there too, God permitting:
Nova, 2012-11-16, (uvod) Oslobođeni!, dolazak Hrvatskih Generala iz Haaga!
https://www.youtube.com/watch?v=J7Wle5isH7M
(I'll just translate the title I gave to the video: "Nova, 2012-11-16, (intro) Freed!, the arrival of Croatian Generals from the Haague!")

But, I still have only one issue unsolved. And I can only be back in some more time. And that is the one bad DDR2 memory module. It's way back in some previous post in this thread...
Back to top
View user's profile Send private message
Display posts from previous:   
Reply to topic    Gentoo Forums Forum Index Kernel & Hardware All times are GMT
Goto page Previous  1, 2
Page 2 of 2

 
Jump to:  
You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot vote in polls in this forum