Re: [gentoo-user] Re: [gentoo-dev] Current unavoidable use of xz utils in Gentoo

[Håkon Alstadheim]

Den 31.03.2024 14:33, skrev Rich Freeman:

(moving this to gentoo-user as this is really getting off-topic for -dev)


It might also happen with commercial software, but the challenge there
is HR as you can't just pay 1 person to masquerade as 10 when they all
need to deal with payroll taxes.


For commercial entities, the government could just contact the company 
and apply pressure, no need to sneak the backdoor in. Cf. RSA .

Re: [gentoo-user] 6.1.53-gentoo-r1 kernel not booting

[Håkon Alstadheim]

Den 01.10.2023 21:31, skrev Frank Steinmetzger:

Am Sat, Sep 30, 2023 at 11:25:46PM +0200 schrieb Håkon Alstadheim:

Den 30.09.2023 22:57, skrev Valmor F. de Almeida:

Hello,

For a while now (3 weeks or so) I have been upgrading the linux kernel
on a Dell XPS laptop starting from 6.1.41-gentoo (which is my current
working kernel) to 6.1.53-gentoo-r1. No kernel I have built since is
able to boot. I have been following the same method for many years: make
oldconfig, etc...

The booting error starts at:

[snip]

* INIT: Entering runlevel: 3
[snip]
* Starting cronie ...
* Starting DHCP Client Daemon ...
* Starting laptop_mode ...
* Mounting network filesystems ...
/etc/init.d/netmount: line 45 /lib/rc/bin/ewend: Input/output error
/lib/rc/sh/rc-cgroup.sh: line 184: rmdir: command not found
INIT:
INIT: cannot execute "/sbin/agetty"
INIT: cannot execute "/sbin/agetty"
INIT: cannot execute "/sbin/agetty"



Can you show /etc/fstab and the console-log for the entire boot? Seems /sbin
is not readable. You sure you have the kernel modules loaded? Are you using
an initramfs? If so, does that build without errors ?

The input/output error – to me – indicates a hardware problem. When you
mounted the FS by hand, can you read ewend? For instance with md5sum.

except it boots ok with older kernels. When you've eliminated the 
impossible, whatever remains, however improbable has to be a kernel 
config change (missing or erroneous and unintended) , or 
initramfs failing to build/install correctly. Check error output from 
your kernel build.

Re: [gentoo-user] 6.1.53-gentoo-r1 kernel not booting

[Håkon Alstadheim]

Den 01.10.2023 06:56, skrev Valmor F. de Almeida:

On 9/30/23 17:25, Håkon Alstadheim wrote:


Den 30.09.2023 22:57, skrev Valmor F. de Almeida:


Hello,

For a while now (3 weeks or so) I have been upgrading the linux 
kernel on a Dell XPS laptop starting from 6.1.41-gentoo (which is my 
current working kernel) to 6.1.53-gentoo-r1. No kernel I have built 
since is able to boot. I have been following the same method for 
many years: make oldconfig, etc...


The booting error starts at:

[snip]

* INIT: Entering runlevel: 3
[snip]
* Starting cronie ...
* Starting DHCP Client Daemon ...
* Starting laptop_mode ...
* Mounting network filesystems ...
/etc/init.d/netmount: line 45 /lib/rc/bin/ewend: Input/output error
/lib/rc/sh/rc-cgroup.sh: line 184: rmdir: command not found
INIT:
INIT: cannot execute "/sbin/agetty"
INIT: cannot execute "/sbin/agetty"
INIT: cannot execute "/sbin/agetty"


Can you show /etc/fstab and the console-log for the entire boot? 
Seems /sbin is not readable. You sure you have the kernel modules 
loaded? Are you using an initramfs? If so, does that build without 
errors ?



Here is fstab:

/dev/nvme0n1p2  /boot   ext2 defaults    0 2
/dev/nvme0n1p3  none    swap sw  0 0
/dev/nvme0n1p4  /   ext4 noatime,discard    0 1
/dev/cdrom  /mnt/cdrom  auto noauto,user 0 0

I have not changed anything from 6.1.41-gentoo (which compiles and 
boots) except updating the config file for compiling the new kernel. 
Then I do: make && modules_install. Which runs without errors. After 
that: grub-mkconfig -o /boot/grub/grub.cfg.

I am not using initramfs.

I can't see anything fishy in fstab, no network-mounts that might have 
failed. From the original message it looks like your network interface 
is not being set up correctly, but you should still be able to boot. 
Also "rmdir: command not found" probably means your PATH does not get 
set, or else you don't have execute permissions on rmdir. Odd  that this 
should be dependent on kernel version. You should go over and verify 
that the transfer of your linux .config went OK. Then verify that the 
new config does in fact enable modules for your network card, and enable 
the necessary settings for your init system (looks like you are using 
openrc?)

Re: [gentoo-user] 6.1.53-gentoo-r1 kernel not booting

[Håkon Alstadheim]

Den 30.09.2023 22:57, skrev Valmor F. de Almeida:


Hello,

For a while now (3 weeks or so) I have been upgrading the linux kernel 
on a Dell XPS laptop starting from 6.1.41-gentoo (which is my current 
working kernel) to 6.1.53-gentoo-r1. No kernel I have built since is 
able to boot. I have been following the same method for many years: 
make oldconfig, etc...


The booting error starts at:

[snip]

* INIT: Entering runlevel: 3
[snip]
* Starting cronie ...
* Starting DHCP Client Daemon ...
* Starting laptop_mode ...
* Mounting network filesystems ...
/etc/init.d/netmount: line 45 /lib/rc/bin/ewend: Input/output error
/lib/rc/sh/rc-cgroup.sh: line 184: rmdir: command not found
INIT:
INIT: cannot execute "/sbin/agetty"
INIT: cannot execute "/sbin/agetty"
INIT: cannot execute "/sbin/agetty"


Can you show /etc/fstab and the console-log for the entire boot? Seems 
/sbin is not readable. You sure you have the kernel modules loaded? Are 
you using an initramfs? If so, does that build without errors ?

Re: [gentoo-user] Password questions, looking for opinions. cryptsetup question too.

[Håkon Alstadheim]

Den 23.09.2023 15:42, skrev Dale:

Wols Lists wrote:

On 19/09/2023 10:13, Dale wrote:

That's a interesting way to come up with passwords tho.  I've seen that
is a few whodunit type shows.  Way back in the old days, they had some
interesting ways of coding messages.  Passwords are sort of similar.

Back when we were busy conquering India ...

The story goes of a General trying to send a message back of his
latest conquest, but he didn't want to use codes because he had a
suspicion the Indians could read them if his messenger was captured.

It appears the story is apocryphal, but the message he sent read
"peccavi".

https://www.ft.com/content/49036e66-ac48-11e8-94bd-cba20d67390c

Cheers,
Wol




It seems that requires a subscription.  Oh well.
Try 
https://www.euronews.com/culture/2023/02/17/culture-re-view-peccavi-a-misattributed-quote-and-the-british-raj

Probably ripped off from FT, but I was  curious :-) .

Re: [gentoo-user] Re: How to move ext4 partition

[Håkon Alstadheim]

Den 22.09.2023 08:48, skrev Wols Lists:

On 20/09/2023 23:39, Grant Edwards wrote:

Assuming GParted is smart enough to do overlapping moves, is it smart
enough to only copy filesystem data and not copy "empty" sectors?
According to various forum posts, it is not: moving a partion copies
every sector. [That's certainly the obvious, safe thing to do.]


Seeing as it knows nothing about filesystems, and everything about 
partitions, it will treat the partition as an opaque blob and move it 
as a single object ...


The partition in question is 200GB, but only 7GB is used, so I think
backup/restore is the way to go...


You would think so :-)

I use ext4, and make heavy use of hard links. Last time I tried a 
straight copy (not backup/restore) I think the copied partition would 
have been three times the size of the original - that is if it hadn't 
run out of space first :-)


Just for completeness, you should check out the e2image command, though 
NOT for direct in-place overlapping move. Something like "e2image -ra 
$sourcevol $targetvol" has worked very well for me. That is obviously 
NOT for overlapping move, but for a backup it is quite fast. If you are 
moving to a bigger volume, I'd do resize as a separate step after the move.

[gentoo-user] Downgrading zfs-kmod 9999 -> 2.1.12 -- Heads up

[Håkon Alstadheim]

Heads up for anyone trying this. Filesystem layout for kernel modules 
has changed, I ended up with left-over modules from  when 
downgrading, needs manual clean-up. Would maybe be a good idea to update 
the ebuilds for the non-git ebuilds to either the new layout or to clean 
up the version- files in postinst.


Symptom will obviously be undefined symbols when doing depmod. Took me 
quite a while before the cause dawned on me :-D

Re: [gentoo-user] Re: [SOLVED] Pipewire not a dependency?

[Håkon Alstadheim]

Den 02.10.2022 11:47, skrev Michael:

On Saturday, 1 October 2022 19:32:11 BST Daniel Sonck wrote:

On zaterdag 1 oktober 2022 19:11:19 CEST Wol wrote:

On 01/10/2022 17:56, Michael wrote:

Anyway, I ventured into pipewire because I wanted to see if Skype would
work without pulseaudio and in this system it won't.  After I manually
installed pipewire Skype won't access the microphone.  

I've got some vague feeling that pipewire is designed to happily sit
under pulseaudio. The design aim was to replace both Jack and pulseaudio
but it basically just presents a sound device to the layers above, so
just like you can stack block devices for disk access, you can stack
jack, pulseaudio and pipewire for sound.

Well, it is actually designed as a drop-in replacement and won't present
audio devices in the sense pulseaudio wants to receive it. I guess it would
theoretically be possible to use pulseaudio's jack sink to talk to
pipewire, but pipewire has the full pulseaudio interface for pulseaudio
applications.

At the moment only some applications support PipeWire's native API, but most
support PulseAudio's API.  When you come across an application like Skype
which expects PulseAudio, the solution is to enable USE="sound-server
pipewire-alsa" for PipeWire and in addition to PipeWire also install media-
libs/libpulse.  No other PulseAudio packages are needed.
To get that, I seem to need media-sound/pulseaudio (meta package) with  
USE="-daemon"


Thereafter an application requiring PulseAudio uses PipeWire, the latter
emulating PulseAudio's server by using PulseAudio's API via libpulse.

I applied the above and now the microphone in Skype works again.  I assume the
same applies to other PulseAudio friendly applications, which won't play
nicely with PipeWire only.  I suppose at some point PulseAudio will be
completely replaced by PipeWire and applications will update their code
accordingly.

Re: [gentoo-user] power button to shutdown for openrc? [SOLVED]

[Håkon Alstadheim]

Den 17.11.2019 12:22, skrev Dale:

Neil Bothwick wrote:

On Sun, 17 Nov 2019 10:50:08 +, Mick wrote:


Magic SysReq would probably have helped in those situations. ext3/4
only journal metadata by default, you can specify a mount option to
also journal data but it impacts performance.

When X hangs and I lose the keyboard to the extent where neither
Ctrl+Alt+F1, or Ctrl+Alt+Del would work, I will use ssh to connect
remotely and stop the hanging process or restat the X server.  If ssh
is also not working, I use the magic SysReq sequence to stop processes,
sync the disks and reboot, or shutdown.  I don't recall losing data in
such cases, although when I have time I run fsck with Live media just
in case.

I do the same, but configuring the power button to do a clean shutdown
may like life simpler.




I don't have a second system to ssh in with but if I did, that's what
I'd do as well, or try at least.


You can get ssh clients for Android or Apple iOS. Just remember to set 
up with public keys before you need them, and also give them a static 
address if the box in question is router/dhcp-server.

Re: [gentoo-user] Globally disabling colour

[Håkon Alstadheim]

Den 02.10.2019 11:01, skrev Dr Rainer Woitok:

Hakon,

On Tuesday, 2019-10-01 16:07:55 +0200, you wrote:


...
I've got this in make.conf, which gets me part-way there:

CMAKE_COLOR_MAKEFILE=OFF
NOCOLOR="true"
GCC_COLORS=""

I already had the  "NOCOLOR" environment variable set  (there was a typo
in my mail, not in my "make.conf" file :-),  but the Cmake and Gcc spec-
ific environment variables were not yet there.  Thankyou for this tip.


I've also got this in /usr/local/bin/rustc and symlinked to
/usr/local/bin/cargo:

This assumes "/usr/local/bin" to precede "/usr/bin" in environment vari-
able "PATH".  Can I set this for Portage only in "make.conf"?

Correct.

/usr/local/bin/rustc:#!/usr/bin/perlmy @newargs=();my $skipnext=0;my 
$me=$0;$me =~
s(.*\/)();push @newargs, "--color";push @newargs, "never";foreach $arg
(@ARGV) {    if($arg eq "--color"){    $skipnext =1;    }
elsif($skipnext == 1){    $skipnext = 0;    } else { # any arg EXCEPT
color gets passed along verbatim     push @newargs, $arg  Â
}}exec("/usr/bin/${me}",@newargs);
---

Alas,  this code snippet did not survive  mail transmission :-(.   Could
you please run

perl -MMIME::Base64 -ne 'print encode_base64($_)' < /usr/local/bin/rustc

and post (or send me privately) the base64 encoded output?  Thankyou.

Sincerely,
   Rainer


Just doing an attachment, so you'll get the charset as well. Don't know 
if that gets through to the list.



#!/usr/bin/perl
my @newargs=();
my $skipnext=0;
my $me=$0;
$me =~ s(.*\/)();
#print STDERR "$me\n";
push @newargs, "--color";
push @newargs, "never";
foreach $arg (@ARGV) {
if($arg eq "--color"){
$skipnext =1;
} elsif($skipnext == 1){
$skipnext = 0;
} else {
push @newargs, $arg
}
#print STDERR "$arg\n";
}
exec("/usr/bin/${me}",@newargs);
#foreach $arg (@newargs) {
#print STDERR "$arg\n";
#}
#exit 0;
#exec /usr/bin/rustc "$@"

Re: [gentoo-user] Globally disabling colour

[Håkon Alstadheim]

Den 01.10.2019 13:28, skrev Dr Rainer Woitok:

Greetings,

having freshly erm ... converted from Ubuntu to Gentoo and thus being an
absolute Gentoo newbie  I'm desparately looking  for a way to get rid of
all the colour in the output  produced by "emerge"  but also by "qlist",
"eix", "e-file" and so on.

Is there a way to globally configure this for all these commands?   I've
already added  'NOCLOR="false"'  to my "make.conf" file but this doesn't
seem to do the trick.  Do I really have to remember for each call to one
of these commands to disable colour individually and in addition have to
remember what's the correct "nocolor" option for the command at hand?

Any pointers welcome ... :-)


I've got this in make.conf, which gets me part-way there:

CMAKE_COLOR_MAKEFILE=OFF
NOCOLOR="true"
GCC_COLORS=""

I've also got this in /usr/local/bin/rustc and symlinked to 
/usr/local/bin/cargo:


/usr/local/bin/rustc:#!/usr/bin/perlmy @newargs=();my $skipnext=0;my $me=$0;$me =~ 
s(.*\/)();push @newargs, "--color";push @newargs, "never";foreach $arg 
(@ARGV) {    if($arg eq "--color"){    $skipnext =1;    } 
elsif($skipnext == 1){    $skipnext = 0;    } else { # any arg EXCEPT 
color gets passed along verbatim     push @newargs, $arg    
}}exec("/usr/bin/${me}",@newargs);

---

This last trick could probably be used for other commands as well 
(modulo specific argument)

Re: [gentoo-user] Keeping 17-year-old Kylix software alive

[Håkon Alstadheim]

Den 09.06.2019 11:24, skrev Matthias Hanft:


The old Kylix libraries just do some comprehensive calculations.
Not too difficult to manually translate that into PHP (or any
other script language, or even plain C), but just hundreds of
code lines to type...

Find a Pascal-to-C++ translator program, and/or hire a college student 
over the summer to do it.

Re: [gentoo-user] Fw: updating /etc/package.accept_keywords

[Håkon Alstadheim]

Den 29.05.2019 23:58, skrev n952...@web.de:

And, what are the consequences that I'm suffering, that I haven't done that 
before, for over a year?


Gesendet: Mittwoch, 29. Mai 2019 um 23:55 Uhr
Von: n952...@web.de
An: gentoo-user@lists.gentoo.org
Betreff: updating /etc/package.accept_keywords

I have many files like ._cfg_package.accept_keywords.
Is the right way to handle this to do something like:

sort -u ._cfg_package.accept_keywords >| package.accept_keywords


For package.accept_keywords specifically, it most likely means you have 
said yes to question to allowing unstable packages, and never actually 
allowed those. So I'd just remove them, and pay more attention on the 
next "emerge -u --autounmask-write @world". See emerge(1)


Apart from that, what the others said.

Re: [gentoo-user] portage sandbox path-depth limit ?

[Håkon Alstadheim]

Den 30. okt. 2018 10:01, skrev Mick:
> On Tuesday, 30 October 2018 06:30:23 GMT Håkon Alstadheim wrote:
>> I'm having fun enabling "test" in FEATURES on my gentoo-desktop. One
>> interesting failure, that brings to mind build failures I have had in
>> the past:
>>
>> Building sys-apps/mlocate-0.26-r2, I get
>>
>>  43: updatedb: Very deep hierarchy   FAILED
>> (updatedb.at:261)
>>
>> Trying to reproduce, as root I do "make check" in the work/mlocate-0.26/
>> , and the test passes.
>>
>>  43: updatedb: Very deep hierarchy   ok
>>
>> I'd really like to get to the bottom of this, as I believe it must have
>> the same root-cause as issues I have had compiling large packages such
>> as firefox.
>>
>> Re-running both the emerge and the make check, I get the same results.
>> emerge fails, make check succeeds. I made a local copy of the ebuild and
>> inserted a "ulimit -a" in pre_src_test.
>>
>> ulimit from root-shell:
>>
>> # ulimit -a
>> core file size  (blocks, -c) unlimited
>> data seg size   (kbytes, -d) unlimited
>> scheduling priority (-e) 0
>> file size   (blocks, -f) unlimited
>> pending signals (-i) 59958
>> max locked memory   (kbytes, -l) 16384
>> max memory size (kbytes, -m) unlimited
>> open files  (-n) 1024
>> pipe size(512 bytes, -p) 8
>> POSIX message queues (bytes, -q) 819200
>> real-time priority  (-r) 0
>> stack size  (kbytes, -s) 8192
>> cpu time   (seconds, -t) unlimited
>> max user processes  (-u) 1
>> virtual memory  (kbytes, -v) unlimited
>> file locks  (-x) unlimited
>>
>> ulimit from emerge:
>>>>> Source compiled.
>> core file size  (blocks, -c) unlimited
>> data seg size   (kbytes, -d) unlimited
>> scheduling priority (-e) 0
>> file size   (blocks, -f) unlimited
>> pending signals (-i) 59958
>> max locked memory   (kbytes, -l) 16384
>> max memory size (kbytes, -m) unlimited
>> open files  (-n) 1024
>> pipe size(512 bytes, -p) 8
>> POSIX message queues (bytes, -q) 819200
>> real-time priority  (-r) 0
>> stack size  (kbytes, -s) 9788
>> cpu time   (seconds, -t) unlimited
>> max user processes  (-u) 1
>> virtual memory  (kbytes, -v) unlimited
>> file locks  (-x) unlimited
>>
>>>>> Test phase: sys-apps/mlocate-0.26-r2
>> I have plenty of space in my portage temp directory (/pt):
>>
>>  # df -hT ./
>> Filsystem  Type Størrelse Brukt Tilgj. Bruk% Montert på
>> /dev/xvdc  ext4  163G  8,0G   147G6% /pt
>>
>> Portage temp is at /pt due to the earlier mentioned issues with firefox.
>>
>> At my wits end here. Anyone ?
> I have not looked or used the test FEATURES of portage, but have also noticed 
> over time certain packages fail to build on machines with low RAM.  As low 
> here I consider anything less than 4G.  It seems each thread is now consuming 
> so much memory they cumulatively use up all RAM available and then start 
> swapping endlessly until the compilation invariably fails.  Increasingly more 
> and more packages have been suffering from this, the last two I noticed are 
> qtwebkit and qtwebengine.
>
> My solution has been to create a package.env file in which I specify MAKEOPTS 
> limiting the number of jobs and average load for any of these packages which 
> chew up all the RAM.
Memory should not be a problem here. Fails with only that one emerge
running,
succeeds if run directly as root, or with FEATURES="-sandbox -usersandbox".

Memory is >14GB:
# vmstat
procs ---memory-- ---swap-- -io -system--
--cpu-
 r  b   swpd   free   buff  cache   si   so    bi    bo   in   cs us sy
id wa st
 3  4  28416 6904608 174112 4616144    0    0    65   266   13    4 10 
2 84  4  0

[gentoo-user] portage sandbox path-depth limit ?

[Håkon Alstadheim]

I'm having fun enabling "test" in FEATURES on my gentoo-desktop. One
interesting failure, that brings to mind build failures I have had in
the past:

Building sys-apps/mlocate-0.26-r2, I get

 43: updatedb: Very deep hierarchy   FAILED
(updatedb.at:261)

Trying to reproduce, as root I do "make check" in the work/mlocate-0.26/
, and the test passes.

 43: updatedb: Very deep hierarchy   ok

I'd really like to get to the bottom of this, as I believe it must have
the same root-cause as issues I have had compiling large packages such
as firefox.

Re-running both the emerge and the make check, I get the same results.
emerge fails, make check succeeds. I made a local copy of the ebuild and
inserted a "ulimit -a" in pre_src_test.

ulimit from root-shell:

# ulimit -a
core file size  (blocks, -c) unlimited
data seg size   (kbytes, -d) unlimited
scheduling priority (-e) 0
file size   (blocks, -f) unlimited
pending signals (-i) 59958
max locked memory   (kbytes, -l) 16384
max memory size (kbytes, -m) unlimited
open files  (-n) 1024
pipe size    (512 bytes, -p) 8
POSIX message queues (bytes, -q) 819200
real-time priority  (-r) 0
stack size  (kbytes, -s) 8192
cpu time   (seconds, -t) unlimited
max user processes  (-u) 1
virtual memory  (kbytes, -v) unlimited
file locks  (-x) unlimited

ulimit from emerge:

>>> Source compiled.
core file size  (blocks, -c) unlimited
data seg size   (kbytes, -d) unlimited
scheduling priority (-e) 0
file size   (blocks, -f) unlimited
pending signals (-i) 59958
max locked memory   (kbytes, -l) 16384
max memory size (kbytes, -m) unlimited
open files  (-n) 1024
pipe size    (512 bytes, -p) 8
POSIX message queues (bytes, -q) 819200
real-time priority  (-r) 0
stack size  (kbytes, -s) 9788
cpu time   (seconds, -t) unlimited
max user processes  (-u) 1
virtual memory  (kbytes, -v) unlimited
file locks  (-x) unlimited
>>> Test phase: sys-apps/mlocate-0.26-r2

I have plenty of space in my portage temp directory (/pt):

 # df -hT ./
Filsystem  Type Størrelse Brukt Tilgj. Bruk% Montert på
/dev/xvdc  ext4  163G  8,0G   147G    6% /pt

Portage temp is at /pt due to the earlier mentioned issues with firefox.

At my wits end here. Anyone ?

Re: [gentoo-user] Per-process-tree memory quotas?

[Håkon Alstadheim]

Den 20. okt. 2018 11:08, skrev Mick:
> On Saturday, 20 October 2018 05:13:13 BST Andrew Udvare wrote:
>>> On 2018-10-19, at 23:24, Alan Grimes  wrote:
>>>
>>> How do I do this?
>> Cgroups were sort of invented for this reason. Yes it requires Systemd.
> No, I don't think systemd is obligatory, although it may be for archlinux?
>
>
>> https://wiki.archlinux.org/index.php/cgroups
>>
>> I too have 32 GiB of RAM and I'm curious how Chromium acts under a
>> constrained environment, or what limits can be placed especially on the
>> CPU. In my experience using Chrome inside VMs, it acts very poorly.
> Have a look here instead for the Gentoo alternative:
>
> https://wiki.gentoo.org/wiki/OpenRC/CGroups
>
"support for userlevel cgroups" is on the list of possible improvements
there. This means some reading and manual setup to limit your browser
processes. Definitely doable though.
Start with turning on CGroups in /etc/rc.conf, and take it from there.

Re: [gentoo-user] [OT] AER: Corrected error received: 0000:00:03.0

[Håkon Alstadheim]

Den 28. sep. 2018 00:02, skrev Mick:
> On Thursday, 27 September 2018 21:51:42 BST Håkon Alstadheim wrote:
>> (Sorry for the OT, don't know where to go for generic hardware questions)
>>
>> I'm wondering if my main board dying? On sept. 1 I started getting error
>> messages like the ones at the  end of this mail. I just noticed them. I
>> am including logs from as far back as I have, just for completeness.
>>
>> This mainboard:
>>
>> # smbios-sys-info
>> Libsmbios version:  2.3.2
>> Product Name:   Z10PE-D8 WS
>> Vendor: ASUSTeK COMPUTER INC.
>> BIOS Version:   3703
>>
>> ... has given me a lot of grief, so I've been mucking about with it
>> quite a lot. I don't know if I did anything on september 1. Could it be
>> some bios-setting I've changed, or som firmware corruption ? Other
>> weirdness happening recently is sensors have started having episodes of
>> all values totally bogus, and one pwm fan-sensor seems to be permanently
>> slightly loony. 
>>
>> OS running is kept up-to-date ~amd64 gentoo-sources running as Dom0
>> under Xen-4.11 latest.
>>
>> --
>>
>>
>> 0:root@gentoo log # zgrep 'pcieport :00:03.0' kern.log* | sed -e
>> 's/^[^:]*://' | sort -M
>> Aug 19 19:38:55 gentoo kernel: [1.181466] pcieport :00:03.0: AER
>> enabled with IRQ 134
>> Aug 19 19:38:55 gentoo kernel: [1.181700] pcieport :00:03.0:
>> Signaling PME with IRQ 134
> Did you enable CONFIG_PCIEAER_INJECT in your kernel?
Yes, enabled already.
>
> What do you get when you boot with the kernel option:
>
> pcie_ports=auto
>
Still happens I'm afraid.

# cat /proc/cmdline
placeholder root=LABEL=SAS-GENTOO ro
xen-pciback.hide=(02:00.*)(04:00.*)(05:00.*)(08:00.*)(81:00.*)(82:00.*)
usbcore.autosuspend=-1 scsi_mod.use_blk_mq=1 xen-netback.max-queues=8
xen-pciback.permissive=1 xen-pciback.verbose_request=1
xen-blkback.max_persistent_grants=1024 intel_iommu=on domodules domdadm 
console=tty1 pcie_ports=auto

 # dmesg -c
[13701.614600] pcieport :00:03.0: AER: Corrected error received:
:00:03.0
[13701.614607] pcieport :00:03.0: PCIe Bus Error:
severity=Corrected, type=Data Link Layer, (Transmitter ID)
[13701.614609] pcieport :00:03.0:   device [8086:2f08] error
status/mask=1000/2000
[13701.614610] pcieport :00:03.0:    [12] Replay Timer Timeout 
[24039.831939] pcieport :00:03.0: AER: Corrected error received:
:00:03.0
[24039.831952] pcieport :00:03.0: PCIe Bus Error:
severity=Corrected, type=Data Link Layer, (Transmitter ID)
[24039.831955] pcieport :00:03.0:   device [8086:2f08] error
status/mask=1000/2000
[24039.831958] pcieport :00:03.0:    [12] Replay Timer Timeout 
0:root@gentoo ~ #

[gentoo-user] [OT] AER: Corrected error received: 0000:00:03.0

[Håkon Alstadheim]

(Sorry for the OT, don't know where to go for generic hardware questions)

I'm wondering if my main board dying? On sept. 1 I started getting error
messages like the ones at the  end of this mail. I just noticed them. I
am including logs from as far back as I have, just for completeness.

This mainboard:

# smbios-sys-info
Libsmbios version:  2.3.2
Product Name:   Z10PE-D8 WS
Vendor: ASUSTeK COMPUTER INC.
BIOS Version:   3703

... has given me a lot of grief, so I've been mucking about with it
quite a lot. I don't know if I did anything on september 1. Could it be
some bios-setting I've changed, or som firmware corruption ? Other
weirdness happening recently is sensors have started having episodes of
all values totally bogus, and one pwm fan-sensor seems to be permanently
slightly loony. 

OS running is kept up-to-date ~amd64 gentoo-sources running as Dom0
under Xen-4.11 latest.

--


0:root@gentoo log # zgrep 'pcieport :00:03.0' kern.log* | sed -e
's/^[^:]*://' | sort -M
Aug 19 19:38:55 gentoo kernel: [    1.181466] pcieport :00:03.0: AER
enabled with IRQ 134
Aug 19 19:38:55 gentoo kernel: [    1.181700] pcieport :00:03.0:
Signaling PME with IRQ 134
Aug 20 11:13:02 gentoo kernel: [    1.184663] pcieport :00:03.0: AER
enabled with IRQ 134
Aug 20 11:13:02 gentoo kernel: [    1.184896] pcieport :00:03.0:
Signaling PME with IRQ 134
Aug 20 11:40:44 gentoo kernel: [    1.182309] pcieport :00:03.0: AER
enabled with IRQ 134
Aug 20 11:40:44 gentoo kernel: [    1.182543] pcieport :00:03.0:
Signaling PME with IRQ 134
Aug 20 16:45:31 gentoo kernel: [    1.183570] pcieport :00:03.0: AER
enabled with IRQ 134
Aug 20 16:45:31 gentoo kernel: [    1.183807] pcieport :00:03.0:
Signaling PME with IRQ 134
Aug 21 19:37:47 gentoo kernel: [    6.380332] pcieport :00:03.0: AER
enabled with IRQ 134
Aug 21 19:37:47 gentoo kernel: [    6.460584] pcieport :00:03.0:
Signaling PME with IRQ 134
Aug 22 03:09:06 gentoo kernel: [    3.301579] pcieport :00:03.0: AER
enabled with IRQ 134
Aug 22 03:09:06 gentoo kernel: [    3.301814] pcieport :00:03.0:
Signaling PME with IRQ 134
Aug 22 03:31:11 gentoo kernel: [    1.186599] pcieport :00:03.0: AER
enabled with IRQ 134
Aug 22 03:31:11 gentoo kernel: [    1.186838] pcieport :00:03.0:
Signaling PME with IRQ 134
Aug 22 09:52:46 gentoo kernel: [    1.185736] pcieport :00:03.0: AER
enabled with IRQ 134
Aug 22 09:52:46 gentoo kernel: [    1.185974] pcieport :00:03.0:
Signaling PME with IRQ 134
Aug 22 10:27:41 gentoo kernel: [    1.186577] pcieport :00:03.0: AER
enabled with IRQ 134
Aug 22 10:27:41 gentoo kernel: [    1.186849] pcieport :00:03.0:
Signaling PME with IRQ 134
Aug 23 16:00:11 gentoo kernel: [    1.183100] pcieport :00:03.0: AER
enabled with IRQ 134
Aug 23 16:00:11 gentoo kernel: [    1.183336] pcieport :00:03.0:
Signaling PME with IRQ 134
Aug 25 15:08:56 gentoo kernel: [    1.182612] pcieport :00:03.0: AER
enabled with IRQ 134
Aug 25 15:08:56 gentoo kernel: [    1.182845] pcieport :00:03.0:
Signaling PME with IRQ 134
Aug 26 13:57:30 gentoo kernel: [    1.213208] pcieport :00:03.0: AER
enabled with IRQ 120
Aug 26 13:57:30 gentoo kernel: [    1.213445] pcieport :00:03.0:
Signaling PME with IRQ 120
Aug 26 19:34:56 gentoo kernel: [    1.180446] pcieport :00:03.0: AER
enabled with IRQ 134
Aug 26 19:34:56 gentoo kernel: [    1.180681] pcieport :00:03.0:
Signaling PME with IRQ 134
Aug 26 19:54:39 gentoo kernel: [    1.208132] pcieport :00:03.0: AER
enabled with IRQ 120
Aug 26 19:54:39 gentoo kernel: [    1.208372] pcieport :00:03.0:
Signaling PME with IRQ 120
Aug 27 00:24:11 gentoo kernel: [    1.201904] pcieport :00:03.0: AER
enabled with IRQ 120
Aug 27 00:24:11 gentoo kernel: [    1.202151] pcieport :00:03.0:
Signaling PME with IRQ 120
Aug 30 00:32:38 gentoo kernel: [    1.202498] pcieport :00:03.0: AER
enabled with IRQ 120
Aug 30 00:32:38 gentoo kernel: [    1.202732] pcieport :00:03.0:
Signaling PME with IRQ 120
Aug 31 17:32:55 gentoo kernel: [    1.201787] pcieport :00:03.0: AER
enabled with IRQ 119
Aug 31 17:32:55 gentoo kernel: [    1.202054] pcieport :00:03.0:
Signaling PME with IRQ 119
Aug 31 17:41:37 gentoo kernel: [    5.364084] pcieport :00:03.0: AER
enabled with IRQ 133
Aug 31 17:41:37 gentoo kernel: [    5.364349] pcieport :00:03.0:
Signaling PME with IRQ 133
Aug 31 18:15:15 gentoo kernel: [    3.299535] pcieport :00:03.0: AER
enabled with IRQ 119
Aug 31 18:15:15 gentoo kernel: [    3.299789] pcieport :00:03.0:
Signaling PME with IRQ 119
Aug 31 18:27:13 gentoo kernel: [    1.205078] pcieport :00:03.0: AER
enabled with IRQ 119
Aug 31 18:27:13 gentoo kernel: [    1.205372] pcieport :00:03.0:
Signaling PME with IRQ 119
Sep  1 05:42:01 gentoo kernel: [40519.193721] pcieport :00:03.0:
AER: Corrected error received: :00:03.0
Sep  1 05:42:01 gentoo kernel: 

Re: [gentoo-user] Any utility to forcibly freeze or swap out a specific pid?

[Håkon Alstadheim]

Den 02. juli 2018 11:34, skrev Vadim A. Misbakh-Soloviov:
>> kill -s SIGSTOP 
>> kill -s SIGCONT 
> Although, such a "freezing" doesn't free any RAM :-/
>
>
>
It will allow the process to be swapped out without provoking thrashing.
Should work, plugins might give you some grief though.

Re: [gentoo-user] Re: Console serial terminal/console with command history?

[Håkon Alstadheim]

Den 22. mai 2018 20:28, skrev Ian Zimmerman:
> On 2018-05-22 12:00, Grant Taylor wrote:
>
>> You might also want to check out using vim or emacs as they have
>> terminal emulators built in.  They might be able to apply some command
>> line history / editing (in a round about way).
> Indeed, if there isn't a prepackaged way the next easiest is probably
> customizing emacs "comint" mode (which is the base mode behind shell
> mode and various other specialized interpreter modes).
>
Second that, use emacs. Basically you just need to teach emacs to
recognize the prompt. Most likely it will already work.

There are at least three ways to use emacs for this. The best would be
to run emacs on your local machine and then M-x serial-term or M-x
shell. ('M' stands for meta, usually means you can push the 'alt' key
together with a letter, but there are other ways if you have no alt key,
like on a serial connection) . Third way would be to run emacs in text
mode on the remote machine. If you have ssh running on the remote
machine, check out the "tramp" package in emacs.

With no gui, you would need to learn the old tty-commands in emacs, like
push  followed by 'x' instead of using ALT-x, but it should all be
there.

Pro-tip: if running emacs on the remote machine, make sure your
terminal-connection does not interpret Control-S as a STOP signal, i.e.
anything to do with XON/XOFF you do NOT want enabled in your
shell-connection.

Re: [gentoo-user] Is sys-firmware/intel-microcode-20180108 complete?

[Håkon Alstadheim]

I've got this in my boot directory:
-rwxr-xr-x 1 root root   67584 Jan 10 15:12 /boot/microcode.bin
-rwxr-xr-x 1 root root   33792 Jan  7 23:56 /boot/microcode.bin.prev

These are the split ucode file for my cpu, specifically a copy of
/lib/firmware/intel-ucode/06-3f-02.

I'm running xen, an xl dmesg says (from loading microcode at boot):
(XEN) microcode: CPU0 updated from revision 0x39 to 0x3b, date = 2017-11-17
(XEN) xstate: size: 0x340 and states: 0x7

0x3b is the same as with the previous microcode, but the size of the
file has changed. What is up with that?

Den 10. jan. 2018 16:47, skrev Wolfgang Mueller:
>> It looks as though my CPU hasn't been fixed yet. Is that right?
> 
> It seems that patches are being pushed out as they are being received,
> so back when that one was released, no other updates were available.
> 
> See https://bugs.gentoo.org/643430#c10
> 
> We should get the full range of updates in the next few days.
> 

[gentoo-user] Warnings on shutting down bcache: WARNING: CPU: 1 PID: 4378 at lib/idr.c:383

[Håkon Alstadheim]

I'm getting some noise in my logs when shutting down bcache.
My system:
Linux gentoo 4.12.7-gentoo #1 SMP Sun Aug 13 22:56:20 CEST 2017 x86_64
Intel(R) Xeon(R) CPU E5-2620 v3 @ 2.40GHz GenuineIntel GNU/Linux
Running as dom0 under Xen-4.9.0

My shutdown script loops over the bcache devices like so:
---
for f in /sys/fs/bcache/854fc80e-5499-413f-b90b-87a803eaca7e/bdev*; do
bdev=$(basename $f )
if [ -e $f/stop ]; then
logger -t "init.d/probe-home-partition" -p daemon.warning\
"Stopping $bdev"
einfo "Stopping $bdev"
if ! echo 1 > $f/stop ; then
  ewarn "Could not stop $bdev"
  logger -t "init.d/probe-home-partition" -p daemon.warning\
 "Could not stop $bdev"
fi
usleep 300
sleep 1
else
einfo "$bdev not cached?"
logger -t "init.d/probe-home-partition" -p daemon.warning \
"$bdev not cached ?"
fi
done

-
This gives me in daemon.log:
-
Aug 14 09:31:21 gentoo init.d/probe-home-partition: Stopping bdev15
Aug 14 09:31:22 gentoo init.d/probe-home-partition: Stopping bdev16
Aug 14 09:31:23 gentoo init.d/probe-home-partition: Stopping bdev17
Aug 14 09:31:24 gentoo init.d/probe-home-partition: Stopping bdev18
Aug 14 09:31:25 gentoo init.d/probe-home-partition: Stopping bdev19
Aug 14 09:31:26 gentoo init.d/probe-home-partition: Stopping cache set
-
But it seems bcache is doing something weird, stopping a device called
bcache0, which I do not have, before stopping the first of my bcache
devices. ALL runs of "echo 1 > $device/stop" provoke a warning in
lib/idr.c on that same line. It may be that my stopping of the bcache
devices is wrong, un-necessary and misguided, but it should still work
the way I am doing it, should it not?


---kern.log:---
...
[140395.421994] nfsd: last server has exited, flushing export cache
[140397.011932] XFS (bcache16): Unmounting Filesystem
[140397.248911] bcache: bcache_device_free() bcache0 stopped
[140398.300187] bcache: bcache_device_free() bcache16 stopped
[140398.381951] ida_remove called for id=16 which is not allocated.
[140398.381975] [ cut here ]
[140398.381985] WARNING: CPU: 1 PID: 4378 at lib/idr.c:383
ida_remove+0xde/0xef
[140398.381987] Modules linked in: xt_physdev iptable_filter ip_tables
x_tables nfsd auth_rpcgss oid_registry nfsv4 dns_resolver nfsv3 nfs_acl
iTCO_wdt iTCO_vendor_support binfmt_misc tun usbip_host usbip_core
pktcdvd xen_wdt xen_blkback intel_rapl amdgpu intel_powerclamp
crc32c_intel serio_raw drm_kms_helper pcspkr syscopyarea sysfillrect
sysimgblt fb_sys_fops ttm lpc_ich i2c_i801 snd_hda_codec_realtek drm
snd_hda_codec_generic bcache snd_usb_audio snd_usbmidi_lib snd_hda_intel
snd_rawmidi snd_hda_codec snd_seq_device snd_hwdep snd_hda_core snd_pcm
snd_timer snd cp210x usbserial nct6775 input_leds hwmon_vid shpchp
ipmi_ssif wmi acpi_power_meter dm_zero dm_thin_pool dm_persistent_data
dm_bio_prison dm_service_time dm_round_robin dm_queue_length
dm_multipath dm_log_userspace cn virtio_pci virtio_scsi
[140398.382056]  virtio_blk virtio_console virtio_balloon xts aes_x86_64
cbc sha512_generic sha1_generic libiscsi scsi_transport_iscsi macvlan
virtio_net virtio_ring virtio e1000 fuse overlay nfs lockd grace sunrpc
fscache jfs multipath linear raid10 raid1 raid0 dm_raid raid456
async_raid6_recov async_memcpy async_pq async_xor xor async_tx raid6_pq
dm_snapshot dm_bufio dm_crypt dm_mirror dm_region_hash dm_log dm_mod dax
hid_sunplus hid_sony hid_samsung hid_pl hid_petalynx hid_monterey
hid_microsoft hid_logitech ff_memless hid_gyration hid_ezkey hid_cypress
hid_chicony hid_cherry hid_a4tech sl811_hcd xhci_plat_hcd ohci_pci
ohci_hcd uhci_hcd aic94xx lpfc qla2xxx aacraid sx8 DAC960 hpsa cciss
3w_9xxx 3w_ mptsas mptfc scsi_transport_fc mptspi mptscsih mptbase
atp870u dc395x qla1280 imm parport dmx3191d
[140398.382139]  sym53c8xx gdth initio BusLogic arcmsr aic7xxx aic79xx
sg pdc_adma sata_inic162x sata_mv sata_qstor sata_vsc sata_uli sata_sis
sata_sx4 sata_nv sata_via sata_svw sata_sil24 sata_sil sata_promise
pata_sis usbhid led_class igb ptp dca i2c_algo_bit ehci_pci ehci_hcd
xhci_pci megaraid_sas xhci_hcd
[140398.382175] CPU: 1 PID: 4378 Comm: kworker/1:2 Not tainted
4.12.6-gentoo #1
[140398.382177] Hardware name: ASUSTeK COMPUTER INC. Z10PE-D8
WS/Z10PE-D8 WS, BIOS 3407 03/10/2017
[140398.382190] Workqueue: events cached_dev_free [bcache]
[140398.382193] task: 880216b24240 task.stack: c900431e4000
[140398.382197] RIP: e030:ida_remove+0xde/0xef
[140398.382199] RSP: e02b:c900431e7db0 EFLAGS: 00010082
[140398.382202] RAX: 0033 RBX: 0012 RCX:

[140398.382204] RDX: 880249654450 RSI: 88024964dc08 RDI:
88024964dc08
[140398.382205] RBP: c900431e7e10 R08:  R09:
000448da
[140398.382207] R10: 1668 R11: 81f1dad0 R12:
c08114c8
[140398.382209] R13: c08114c0 R14: 00fa R15:
0010
[140398.382223] FS:  

Re: [gentoo-user] I have a patch for kde-plasma/plasma-desktop-5.10.1 where do I send it?

[Håkon Alstadheim]

Den 08. juni 2017 22:46, skrev Dale:
> Håkon Alstadheim wrote:
>> Hi all I'd like to run the latest kde desktop , but stuff keeps
>> crashing. Regular X just drops me right back at login, wayland lets me
>> log in but has some issues.
>>
>> Among other things I am unable tune my keyboard settings. SIGSEGV on
>> XInternAtom in kcms/keyboard/x11_helper.cpp .
>>
>> I found a report of a similar bug in digiKam on the web, and just for
>> kicks I tried the attached patch. I am now able to change keyboard
>> settings in plasma (wayland).
>>
>> So, how do I find out if this is fixed upstream, and if not where do I
>> send a bug-report?
> 
> 
> It doesn't seem to be Gentoo's bug so I would start here:
> 
> https://bugs.kde.org/
> 
> If it is something Gentoo, file it on Gentoo's bug system. 
> 
> As to already reported, this is the only bug I found.
> 
> https://bugs.kde.org/show_bug.cgi?id=367080 
> 
> I don't think that is the same problem.  The other results were years
> old.  You may have found a bug and a solution as well, which they would
> likely like to hear about.  ;-)
> 
> Hope that helps.
> 
> Dale
> 
> :-)  :-) 
> 

I think it IS the same problem. Posted my patch to that bug, thanks for
the help. My head was already swimming a bit after finding the problem.
Turns out I already had a log-in for that page so the rest was easy :-) .

[gentoo-user] I have a patch for kde-plasma/plasma-desktop-5.10.1 where do I send it?

[Håkon Alstadheim]

Hi all I'd like to run the latest kde desktop , but stuff keeps
crashing. Regular X just drops me right back at login, wayland lets me
log in but has some issues.

Among other things I am unable tune my keyboard settings. SIGSEGV on
XInternAtom in kcms/keyboard/x11_helper.cpp .

I found a report of a similar bug in digiKam on the web, and just for
kicks I tried the attached patch. I am now able to change keyboard
settings in plasma (wayland).

So, how do I find out if this is fixed upstream, and if not where do I
send a bug-report?
--- a/kcms/keyboard/x11_helper.cpp	2017-06-08 21:04:52.65000 +0200
+++ b/kcms/keyboard/x11_helper.cpp	2017-06-08 21:02:23.88000 +0200
@@ -239,6 +239,13 @@
 	char *prop_data = NULL;
 	Status ret;
 
+	if (!QX11Info::isPlatformX11())
+	{
+	qCWarning(KCM_KEYBOARD) << "Desktop platform is not X11";
+	return false;
+	}
+
+
 	Atom rules_atom = XInternAtom(display, _XKB_RF_NAMES_PROP_ATOM, False);
 
 	/* no such atom! */

[gentoo-user] xen-tools-4.8.1.ebuild pulling in old ovmf.bin

[Håkon Alstadheim]

I have a problem where xen hvm guests using ovmf (efi) as "bios" will
not read efi variables stored in /NvVars. I believe that may
be a known bug in old versions of tianocore.

Building xen-tools-4.8.1 now, and I notice that the ebuild still has
OVMF_PV=20151110 . Tried building a newer one myself for 4.8.0 by
pulling the sources that are referred to in the upstream makefile.
However, I can't seem to get it to boot :-/, never mind getting the
ebuild to make it.

Is there any hope that the ebuild might be updated with newer ovmf source ?

Re: [gentoo-user] HD 630?

[Håkon Alstadheim]

Den 28. mars 2017 00:03, skrev Jorge Almeida:

This may be a stupid question, for one of two possible reasons, but
here it goes:

I'm thinking of buying a recent Intel CPU (7th generation, in
saleslang), say an i5-7400, and it came to mind, not too late yet,
that the integrated GPU may not be supported in linux. I'm talking
about the latest kernels, not necessarily the gentoo-packaged one.
Anyone knows something about it? And if not supported, is it likely
that it will be sometime soon? I can use a spare Radeon card
meanwhile...
Google "linux intel graphics", and you'll find lots of hits from Intel 
itself. Most likely also from linux kernel maintainers. My impression 
(though I do not have one of these cpus) is that this is very well 
supported, and Intel is eager to get their latest offering out into the 
linux community, in a fairly stable form.




(And, for someone who is not a gamer, is a 7th generation CPU worth it
at all, as opposed to a 6th generation one?)
Read some benchmarks for your work-load. My guess: yes. Calculate 
percent increase performance, compare prices. Prices change all the 
time, so you need to do this the same week as you place your order.


Point to keep in mind: not just raw power but also heat/performance is 
relevant. Newer cpus waste less energy/need fewer fans/make less noise 
per work-unit done.


Running gentoo you might also be well positioned to take advantage of 
latest optimizations, but this can be a huge sink for your personal 
time. For reasonable effort, expect to wait at least a couple of years 
before stable gcc and kernel modules will push your rig to the max.

Re: [gentoo-user] Weird warning message when emerging gcc

[Håkon Alstadheim]

Den 30. des. 2016 14:44, skrev lee:
> Nikos Chantziaras  writes:
> 
>> A world update emerged gcc-5.4.0-r2 (update from 5.4.0). At the end of
>> the build, I got this:
>>
>>  * Python seems to be broken, attempting to locate CHOST ourselves ...
>>  * Switching native-compiler to x86_64-pc-linux-gnu-5.4.0
>> ...PORTAGE_BZIP2_COMMAND setting is invalid: 'bzip2'
>> PORTAGE_BZIP2_COMMAND setting from make.globals is invalid: 'bzip2'
>>
>> I'm not seeing how python is broken here (works fine), and why
>> PORTAGE_BZIP2_COMMAND is invalid. Can someone explain what's going on
>> here?
> 
> Since there is such a command, is it possible (and worthwhile) to use
> lbzip2 instead of bzip2 with portage?  (lbzip2 is ridiculously fast when
> you have the cores and the RAM ...)
> 
> 

What is the difference between theese two? :

* app-arch/lbzip2
 Available versions:  2.3-r1 ~2.5 {debug symlink}
 Homepage:https://github.com/kjn/lbzip2/
 Description: Parallel bzip2 utility

[I] app-arch/pbzip2
 Available versions:  1.1.12 ~1.1.12-r1 ~1.1.13 {static symlink}
 Installed versions:  1.1.12(kl. 16.51 +0100 09. nov. 2016)(symlink
-static)
 Homepage:http://compression.ca/pbzip2/
https://launchpad.net/pbzip2
 Description: Parallel bzip2 (de)compressor using libbz2


I just picked one at random.

Re: [gentoo-user] Gentoo-sources-4.9.0

[Håkon Alstadheim]

Kernel now has two possible kernel modules for a lot of the radeon
cards, "radeon" and "amdgpu" . Don't remember the differences, google is
your friend. "amdgpu" in 4.9 supports more cards than it used to. Rather
prominent in the linux-4.9 announcements. Are you now (automatically? )
loading amdgpu whereas you maybe used radeon before? .

Both modules, and the drm module, have options for verbosity. Turn them
on and compare boot-logs from 4.8.14 and 4.9. Most likely you'll spot
where it goes bad.

Den 13. des. 2016 18:09, skrev Peter Humphrey:
> Hello list,
> 
> Has anyone stumbled over a display problem with this kernel? Copying the 
> .config from 4.8.14, tweaking with oldconfig and compiling left me with a 
> blank display and no booting activity.
> 
> The display card is an AMD/ATI Tonga Radeon R9 380X and I load the latest 
> microcode during kernel compilation. I have VIDEO_CARDS="amdgpu radeonsi" in 
> make.conf. Fine with earlier kernel versions but not with 4.9.0.
> 

Re: [gentoo-user] emerge --sync

[Håkon Alstadheim]

Den 24. okt. 2016 17:21, skrev Jorge Almeida:
> I want to do emerge --sync on computer A and then update computer B by
> copying /usr/portage. Is this safe? The point is: does emerge --sync
> just updates the contents of /usr/portage or does it also change
> something else ?
I have one box that I run emerge --sync on, and then export /usr/portage
with nfs to a few other gentoo boxes. Works well. For eix, you can do
eix-update on the nfs-clients. I have been experimenting with building
once and using the nfs-server as binhost. Small variations between the
boxes have so far made me abandon that, but a single version of
/usr/portage has given me no headaches.

Re: [gentoo-user] Re: Strive for zero swap usage?

[Håkon Alstadheim]

Den 08. okt. 2016 21:27, skrev Kai Krakow:
>
> You may want to try setting your io scheduler to deadline (or even noop
> if you are using a RAID controller with bbu and write cache). Since you
> seem to prefer response times over throughput you should be using
> deadline io scheduler anyways. Actually, don't use the default CFQ if
> your server is virtualized. At least in my tests, CFQ seems to work a
> lot against what virtualized IO seems to achieve.
>
> I also suggest using maybe XFS as a filesystem. Which one are you using?
>
I second XFS and deadline, at least on RAID or on a VM. I got bitten by
XFS recently though, had a crash with the linux-4.8.0 bug which left my
root fs in a bad state. I had no xfs_repair in my initramfs,  (no help
from genkernel) and the systemrescueCD images from portage were stored
on the non-mountable fs :-). My GRML CD had a version of xfs_repair that
was too old to deal. Had to find a lap-top and plug it straight to the
outside line and get a systemrescueCD from some shady download-provider.
Believe it was sourceforge. Make sure you have an up-to date
systemrescueCD with recent-ish xfsprogs residing in you boot device if
you go XFS for your root-device. fsck.xfs is basically a no-op, so if
bad things happen, you mount to play back the journal, then umount, then
xfs_repair.

Packages:
app-admin/systemrescuecd-x86 (doctored to store latest .iso on your
boot-device, which you back up to memory-stick)
sys-boot/systemrescuecd-x86-grub
sys-fs/xfsprogs (up to date !)

Re: [gentoo-user] Digest verification failed: Filesize does not match recorded size

[Håkon Alstadheim]

Den 08. okt. 2016 11:24, skrev Bill Kenworthy:
> On 08/10/16 16:49, Mick wrote:
>> On Saturday 08 Oct 2016 11:45:05 Alexander Kapshuk wrote:
>>> On Sat, Oct 8, 2016 at 11:33 AM, Andrew Lowe  wrote:
 On 08/10/16 15:59, Alexander Kapshuk wrote:
> I've been getting this output since last night.
>
> Calculating dependencies  ... ... done!
> [ebuild U  ] app-shells/bash-4.3_p48 [4.3_p46-r1]
 [snip]
 ...
 ...
 [snip]

> Or is there some other way to fix this, like running 'ebuild
> /path/to/ebuild manifest', as suggested here:
> https://forums.gentoo.org/viewtopic-p-7414010.html
>
> Thanks.
>
 Unless Alexander is in Perth, Australia, I can confirm the same

 thing, exactly as Alexander described, is happening here - Perth that is.

 Andrew
>>> Ha, ha. I'm not that Down Under. Pun intended.
>>>
>>> I used a European mirror.
>> I'm getting the same on a UK mirror this morning.
>>
> Worked for me - iinet mirror, Perth Australia.
>
> Tried that one?
>
> BillK
>
>
>
  Same here, from Norway. I did not want to force accept of potentially
bad .emerge by running "emerge .emerge digest". Rather I
deleted the Manifest in the affected directories, and ran emerge --sync
again. Fixed.

[gentoo-user] What is current "worst practice" for maximum bells?

[Håkon Alstadheim]

What is a good strategy for setting use-flags to get the most out of a
fairly recent graphics card these days?

It used to be (for me at least) that I just made sure I had the latest
opengl and associated libs, and the latest mesa if my card was
supported. These days things get more complicated, especially if you
want to try out KDE and plasma alongside Gnome.

Do I just turn everything on globally (wayland egl gles gles1 gles2
opengl mesa gtk* ) , and then go through all conflicts one by one ? If
not, what flags do I set globally, and which packages do I add specific
use-flags to?

My primary desktop is just fluxbox, but occasionally I like to run both
KDE and Gnome stuff with everything turned on to the max.

I have plenty of CPU, so the stuff with "embedded" in the descriptions
sound like I should leave them alone, but when I get down to resolving
conflicts, it feels like I am going against the grain, with cascading
dependencies leading into a murky mess of details.

Some road signs, anyone?

Re: [gentoo-user] Gummiboot -> efibootmgr

[Håkon Alstadheim]

Den 22. aug. 2016 17:05, skrev Peter Humphrey:
> Hello list,
>
> Following today's marking of gummiboot as to be deleted in a month, I had a 
> look at efibootmgr in the wiki pages. It looks as though I'll be able to use 
> it instead, but one thing puzzles me: is it possible to create a set of 
> configs for several kernels, the way gummiboot does in 
> /boot/loader/entries/*.conf? Actually, I'd also like to specify each of two 
> kernel versions with three different command lines to start different run 
> levels.
>
> My brain isn't working very well today, so would anyone like to offer me 
> some advice? Please?  :)
>
As far as I can tell, you need a separate executable for each
menu-entry. The "menu" would then be the modern descendant of the "bios"
boot-list. Works well for two or three standard boot-configurations.

Booting straight into linux on an EFI system without a boot-loader means
you have no way to provide command-line or initramfs as far as I can
tell, all modules must be compiled in, and default command-line needs to
be set in the kernel config.

EFI executables can also be created for example by dracut, which will
allow kernel, command-line and initramfs in one efi-file. I have never
tried this, I'm using xen.efi (a hypervisor kernel)  with a config file,
which allows loading of separate kernel and initramfs. Config is just a
text file with the same name as the executable, with a .cfg extension.
Linux has no such magic available I believe.

Copying of executables gets unwieldy fast, plus I have found no way to
edit the command-line.

Therefore I keep a grub2 install as well, which has editable
commmand-lines and such, and also  sys-boot/systemrescuecd-x86-grub. I
put GRUB_PLATFORMS="efi-64" in /etc/portage/make.conf. I let
/boot/EFI/Boot/BOOTX64.EFI be a copy of the Grub EFI binary, so that
whenever I update the bios and wipe all efi variables, I can boot into
Grub, given that /boot/EFI is where I mount my EFI boot partition. Works
as advertised. Grub is not able to load xen.efi, so I need to keep
xen.gz around as well for that.

Re: [gentoo-user] 2000 emails - printing, sorting by date

[Håkon Alstadheim]

Den 18. aug. 2016 08:39, skrev Stroller:
>
>> On 17 Aug 2016, at 15:12, Daniel Quinn > > wrote:
>>
>> I’m a Python guy, so my answer to this would be "use Python" :-)
>>
>> [The ReportLab
>> library](https://www.reportlab.com/docs/reportlab-userguide.pdf) is
>> extremely powerful and can be used to generate a PDF for every email
>> or a pdf for all emails.  I've not used it myself, but I hear it's
>> very good.
>>
>> … 
>>
>> At that point you have a sorted list of email objects which you can
>> then use ReportLab to generate a PDF.
>
>
> That's a little more complicated than I hoped for.
>
Have you looked at app-misc/muttprint ? Never tried it, but looks to fit
the bill. You will need various command-line tools that mutt-print will
use to parse mails and generate graphics.

[gentoo-user] gcc5 and xen-tools

[Håkon Alstadheim]

Apologies for ipad mail, I'll not make a habit of using it. Away from desktop 
four weeks atm.

I'm getting compilation-errors on xen-tools-4.6.3-r1 using gcc-5.4.0

Can't find xen mentioned here : 
https://bugs.gentoo.org/buglist.cgi?quicksearch=gcc-5.3_id=3005290

Adding " append-flags -Wno-error=trampolines" to src_compile in the ebuild 
for xen-tools-4.6.3 (no -r1) allowed that to compile and is running fine.

 Build log excerpt: -
xenalyze.c: In function 'interval_cr3_schedule_ordered_output': 

 xenalyze.c:2649:9: error: trampoline generated for 
nested function 'cr3_time_compare' [-Werror=trampolines]
   int 
cr3_time_compare(const void *_a, const void *_b) {  

 ^  

 xenalyze.c: In function 
'dump_eip': 

 xenalyze.c:2969:9: error: trampoline generated for nested function 
'eip_compare' [-Werror=trampolines] 
   int eip_compare(const void *_a, 
const void *_b) {   

 ^  

 xenalyze.c: In function 'cr3_dump_list':   

  xenalyze.c:4141:9: error: 
trampoline generated for nested function 'cr3_compare_start' 
[-Werror=trampolines]   
   int cr3_compare_start(const void *_a, const void 
*_b) {  
^   


cc1: all warnings being treated as errors   

 
/var/tmp/portage/app-emulation/xen-tools-4.6.3-r1/work/xen-4.6.3/tools/xentrace/../../tools/Rules.mk:107:
 recipe for target 'xenalyze.o' failed  
make[2]: *** [xenalyze.o] Error 1   

 make[2]: Leaving directory 
'/var/tmp/portage/app-emulation/xen-tools-4.6.3-r1/work/xen-4.6.3/tools/xentrace'

 
/var/tmp/portage/app-emulation/xen-tools-4.6.3-r1/work/xen-4.6.3/tools/../tools/Rules.mk:123:
 recipe for target 'subdir-all-xentrace' failed 
make[1]: *** [subdir-all-xentrace] Error 2  

 make[1]: Leaving directory 
'/var/tmp/portage/app-emulation/xen-tools-4.6.3-r1/work/xen-4.6.3/tools'

  
/var/tmp/portage/app-emulation/xen-tools-4.6.3-r1/work/xen-4.6.3/tools/../tools/Rules.mk:118:
 recipe for target 'subdirs-all' failed 
make: *** [subdirs-all] Error 2 

 make: Leaving directory 
'/var/tmp/portage/app-emulation/xen-tools-4.6.3-r1/work/xen-4.6.3/tools'

  * ERROR: app-emulation/xen-tools-4.6.3-r1::gentoo failed (compile pha




Håkon Alstadheim 
mob:+47 47353938 epost:ha...@alstadheim.priv.no

Re: [gentoo-user] Re: [OT] How to be a penguin.

[Håkon Alstadheim]

Den 29. mai 2016 03:17, skrev Gregory Woodbury:
>
> WOW!
>
> Alan just wants to start it and walk away, as if Gentoo was a binary
> distribution
> that handles it all upstream.  He doesn't want to take the time to
> review what
> emerge is proposing and see if changes are needed first.
>
It IS actually possible to do that, at least for non-critical systems.
Just make sure to mail yourself with the emerge output, so you can fix
things before the next automatic run.

The point being that you don't have to sit and watch while emerge works.
You can have the output of any blocks or failures waiting for you in
your inbox at your convenience.

For this to work in a timely fashion you need to stay on stable packages
as much as possible, and also keep other customizations to a minimum,
e.g. don't use --autounmask-write.

Re: [gentoo-user] [Solved?] Getting a valid /etc/adjtime while using ntpd ?

[Håkon Alstadheim]

On 03/17/2016 11:31 PM, Mick wrote:

On Friday 18 Mar 2016 06:01:17 Bill Kenworthy wrote:

On 18/03/16 05:59, Bill Kenworthy wrote:

On 18/03/16 05:14, Alan McKinnon wrote:

On 17/03/2016 22:02, Håkon Alstadheim wrote:

On 03/17/2016 02:03 PM, Bill Kenworthy wrote:

On 17/03/16 20:26, Alan McKinnon wrote:

On 17/03/2016 08:50, Håkon Alstadheim wrote:

I have a server SUPPOSED to be running 24/7, but every once in a
while
during a prolonged absence the box will go down. The Real Time Clock
will drift, and in the rush to get the box up again I let everything
boot up automatically and get both wrong time on the main systems,
and
different times on the various systems.

My setup has a main server which does NTP, but with no direct link to
the outside. Router /have/ to be booted booted later (dumb
setup, don't ask), after which I can finally get correct time from
NTP.

NTP initiates "11 minute mode", which makes /etc/adjtime useless as
far
as I understand. Anybody have a /correct/ way to account for RTC
drift
on a box running ntpd? Right now I have a ---file in
/etc/cron.d/time-bad like so:
* * * * * root adjtimex -S 5 >/dev/null 2>&1 
Have you looked at adjtimex ... its in portage


 From the man page ...
"For a standalone or intermittently connected machine, where it’s not
ossible to run ntpd, you may use adjtimex instead to correct the sys-tem
clock for systematic drift.

There are several ways to estimate the drift rate.  If your

computer can be connected to the net, you might run ntpd for at least
several hours and run "adjtimex --print" to learn what values of tick
and freq it settled on.  Alternately, you could estimate values using as
a reference the CMOS clock (see the --compare and --adjust switches),
another host (see --host and --review), or some other source of time
(see --watch and --review).  You could then add a line to rc.local
invoking adjtimex, or configure /etc/init.d/adjtimex or
/etc/default/adjtimex, to set those parameters each time you reboot."

Used it at one time for dialup which approximates your condition.

BillK

forget it ... I forgot that's where you started from ... must be getting
old :(

Nobody mentioned net-misc/chrony.  Would it be more appropriate for this use
case?

This is looking really good. I never considered chrony since I scrapped 
my modems ~15 years ago, but chrony has these things going for it:


- Good documentation in info format
- It acknowleges the existence of the rtc
- Allows turning OFF system -> rtc updates
- Keeps its own drift files, which I believe watches both rtc and system 
clock drift.


I'll run chrony with "-r -s" , with "rtcfile" and without "rtcsync".
My hopes are high. :-D .

Re: [gentoo-user] Getting a valid /etc/adjtime while using ntpd ?

[Håkon Alstadheim]

On 03/17/2016 10:59 PM, Bill Kenworthy wrote:

On 18/03/16 05:14, Alan McKinnon wrote:

On 17/03/2016 22:02, Håkon Alstadheim wrote:

On 03/17/2016 02:03 PM, Bill Kenworthy wrote:

On 17/03/16 20:26, Alan McKinnon wrote:

On 17/03/2016 08:50, Håkon Alstadheim wrote:

I have a server SUPPOSED to be running 24/7, but every once in a while
during a prolonged absence the box will go down. The Real Time Clock
will drift, and in the rush to get the box up again I let everything
boot up automatically and get both wrong time on the main systems, and
different times on the various systems.

My setup has a main server which does NTP, but with no direct link to
the outside. Router /have/ to be booted booted later (dumb
setup, don't ask), after which I can finally get correct time from NTP.

NTP initiates "11 minute mode", which makes /etc/adjtime useless as far
as I understand. Anybody have a /correct/ way to account for RTC drift
on a box running ntpd? Right now I have a ---file in
/etc/cron.d/time-bad like so:
* * * * * root adjtimex -S 5 >/dev/null 2>&1 
Have you looked at adjtimex ... its in portage


 From the man page ...
"For a standalone or intermittently connected machine, where it’s not
ossible to run ntpd,

I /can/ and do run ntpd once everything is up and running.

  you may use adjtimex instead to correct the sys-tem
clock
System clock is fine, what I'm after is drift of the RTC clock ("bios 
clock"). Ntp does nothing for that, as far as I understand. Now, I'd be 
very happy if someone could tell me I've misunderstood.

  for systematic drift.



There are several ways to estimate the drift rate.  If your
computer can be connected to the net, you might run ntpd for at least
several hours and run "adjtimex --print" to learn what values of tick
and freq it settled on.  Alternately, you could estimate values using as
a reference the CMOS clock (see the --compare and --adjust switches),
another host (see --host and --review), or some other source of time
(see --watch and --review).  You could then add a line to rc.local
invoking adjtimex, or configure /etc/init.d/adjtimex or
/etc/default/adjtimex, to set those parameters each time you reboot."

Used it at one time for dialup which approximates your condition.

[gentoo-user] Getting a valid /etc/adjtime while using ntpd ?

[Håkon Alstadheim]

I have a server SUPPOSED to be running 24/7, but every once in a while
during a prolonged absence the box will go down. The Real Time Clock
will drift, and in the rush to get the box up again I let everything
boot up automatically and get both wrong time on the main systems, and
different times on the various systems.

My setup has a main server which does NTP, but with no direct link to
the outside. Router /have/ to be booted booted later (dumb
setup, don't ask), after which I can finally get correct time from NTP.

NTP initiates "11 minute mode", which makes /etc/adjtime useless as far
as I understand. Anybody have a /correct/ way to account for RTC drift
on a box running ntpd? Right now I have a ---file in
/etc/cron.d/time-bad like so:
* * * * * root adjtimex -S 5 >/dev/null 2>&1 

Re: [gentoo-user] Getting a valid /etc/adjtime while using ntpd ?

[Håkon Alstadheim]

On 03/17/2016 11:31 PM, Mick wrote:

On Friday 18 Mar 2016 06:01:17 Bill Kenworthy wrote:

On 18/03/16 05:59, Bill Kenworthy wrote:

On 18/03/16 05:14, Alan McKinnon wrote:

On 17/03/2016 22:02, Håkon Alstadheim wrote:

On 03/17/2016 02:03 PM, Bill Kenworthy wrote:

On 17/03/16 20:26, Alan McKinnon wrote:

On 17/03/2016 08:50, Håkon Alstadheim wrote:

I have a server SUPPOSED to be running 24/7, but every once in a
while
during a prolonged absence the box will go down. The Real Time Clock
will drift, and in the rush to get the box up again I let everything
boot up automatically and get both wrong time on the main systems,
and
different times on the various systems.

My setup has a main server which does NTP, but with no direct link to
the outside. Router /have/ to be booted booted later (dumb
setup, don't ask), after which I can finally get correct time from
NTP.

NTP initiates "11 minute mode", which makes /etc/adjtime useless as
far
as I understand. Anybody have a /correct/ way to account for RTC
drift
on a box running ntpd? Right now I have a ---file in
/etc/cron.d/time-bad like so:
* * * * * root adjtimex -S 5 >/dev/null 2>&1 
Have you looked at adjtimex ... its in portage


 From the man page ...
"For a standalone or intermittently connected machine, where it’s not
ossible to run ntpd, you may use adjtimex instead to correct the sys-tem
clock for systematic drift.

There are several ways to estimate the drift rate.  If your

computer can be connected to the net, you might run ntpd for at least
several hours and run "adjtimex --print" to learn what values of tick
and freq it settled on.  Alternately, you could estimate values using as
a reference the CMOS clock (see the --compare and --adjust switches),
another host (see --host and --review), or some other source of time
(see --watch and --review).  You could then add a line to rc.local
invoking adjtimex, or configure /etc/init.d/adjtimex or
/etc/default/adjtimex, to set those parameters each time you reboot."

Used it at one time for dialup which approximates your condition.

BillK

forget it ... I forgot that's where you started from ... must be getting
old :(

Nobody mentioned net-misc/chrony.  Would it be more appropriate for this use
case?


I see it also claims to contain an ntp server. I'll check it out.

Re: [gentoo-user] Getting a valid /etc/adjtime while using ntpd ?

[Håkon Alstadheim]

On 03/17/2016 02:03 PM, Bill Kenworthy wrote:

On 17/03/16 20:26, Alan McKinnon wrote:

On 17/03/2016 08:50, Håkon Alstadheim wrote:

I have a server SUPPOSED to be running 24/7, but every once in a while
during a prolonged absence the box will go down. The Real Time Clock
will drift, and in the rush to get the box up again I let everything
boot up automatically and get both wrong time on the main systems, and
different times on the various systems.

My setup has a main server which does NTP, but with no direct link to
the outside. Router /have/ to be booted booted later (dumb
setup, don't ask), after which I can finally get correct time from NTP.

NTP initiates "11 minute mode", which makes /etc/adjtime useless as far
as I understand. Anybody have a /correct/ way to account for RTC drift
on a box running ntpd? Right now I have a ---file in
/etc/cron.d/time-bad like so:
* * * * * root adjtimex -S 5 >/dev/null 2>&1 

When the box was off, all questions of accurate ntp tracking are moot.
ntp is designed around the idea that every second happens but from your
machine's point of view they didn't happen since it was powered down.

I would go the really simple route and force ntpdate to run once during
boot up before ntpd is started, thereby avoiding the entire issue.
Why can't I have proper drift information for my RTC ("bios clock") ? 
The old way ( where  "system-time as set by ntp"  minus "RTC time"  
gives "drift-value" written to /etc/adjtime )   used to work perfectly 
for me for several years. Is there no canonical way of getting that 
these days?


My problem is that my WAN connection can not be brought up until well 
after the main server is up (stupid I know, but rearranging things 
entails a major overhaul). Thus a bios clock without drift information 
gives me a choice between ntpdate (which messes up my logs) and ntp with 
incremental adjustments (which might leave clocks wrong for several days).


I really need the logs to be on the same clock for all systems. Don't 
ask, just assume I know why it's called bleeding edge . I also really 
need sub-minute accuracy on all clocks. I suppose I should try running 
ntpdate on everything once the WAN connection is up, just to see how bad 
the mess is.




Sometimes correctness really doesn't matter, this looks like one of those.


alan


add a cheap gps setup as the reference clock to the server,
That sounds like a real option. I have an old Nokia N900 lying around 
with a broken usb-port, so I'd need to solder in a power lead. Any 
pointers for how to read time signal from the gps on a maemo system?

or even
better is a dedicated time server (either a real one or a raspberry
pi/gps) on the network if you have internal connectivity.  Going super
cheap, but not quite as accurate for me was an arduino and rtc on a
bluetooth pan for when the network was down but I needed a reference (to
power up the real server :).


google "arduino time server" for plenty of options :)

This led me to finding some serial-port connected gps modules. 
Serial-ports I have, so this is sounding good. 35 to 40 euro for a gps 
device when hwclock and /etc/adjtime should give ballpark-correct time 
on boot makes me hate this though.


This reminds me one reason I need a valid time is my DVB-T TV-receiver 
card. It should be possible to find a clock source in the broadcast 
stream. I'll research that first, while I leave my "adjtime -S 5 " hack 
running, even though I still don't know if that makes any sense at all. 
At least now there is something different from "0." in the 
driftvalue, which gives me some hope I'm on to something.

Re: [gentoo-user] Re: openssl upgrade may miss some needed rebuilds

[Håkon Alstadheim]

On 03. mars 2016 12:26, Rich Freeman wrote:
> On Thu, Mar 3, 2016 at 3:15 AM, Håkon Alstadheim
> <ha...@alstadheim.priv.no> wrote:
>> Would "revdep-rebuild.sh -i -L "libssl\.so.*" -- -f" before emerging, be
>> sufficient ? I.e. that should obviate the need for compiling wget with
>> gnutls ?
>>
> No, and no.  The problem is the ABI is silently changing.  Rebuilding
> everything 
Not rebuilding, "-f" is supposed to mean "fetchonly" , i.e. all wget
would get run first, and then a second run of emerge would do the actual
compilation with the distfiles already on local disk, That is if I
understadnd "-f" correctly, and given enough disk-space and no sneaky
purging distfiles between runs.

> before updating openssl will just recompile everything with
> the old ABI, and then when it silently changes everything will still
> break.  The purpose of building wget with gnutls is to make it no
> longer use openssl, so then it doesn't break at all.
>

Re: [gentoo-user] Re: openssl upgrade may miss some needed rebuilds

[Håkon Alstadheim]

Den 02. mars 2016 19:19, skrev »Q«:
> On Wed, 2 Mar 2016 10:49:59 -0500
> Rich Freeman  wrote:
>
>> https://forums.gentoo.org/viewtopic-p-7886940.html
>> https://bugs.gentoo.org/show_bug.cgi?id=576128
> I had wget with USE="gnutls" already, so I took the plunge yesterday
> and followed PolynomialC's instructions at the first link above.
>
> When I used
>
> revdep-rebuild.sh -i -L "libssl\.so.*" -- --exclude=openssl --keep-going
>
> the only package that failed to rebuild was
> www-client/w3mmee-0.3.2_p24-r7, and that failure is due to
> , nothing to do with
> openssl.
>
>
>
Would "revdep-rebuild.sh -i -L "libssl\.so.*" -- -f" before emerging, be
sufficient ? I.e. that should obviate the need for compiling wget with
gnutls ?

Re: [gentoo-user] Nouveau blank screen

[Håkon Alstadheim]

On 01/12/2016 10:17 PM, lee wrote:

Thanks for helping, I' try to answer in-line.

Håkon Alstadheim <ha...@alstadheim.priv.no> writes:


I have an old but good graphics card, "NVIDIA Corporation GT200GL
[Quadro FX 3800]". The proprietary driver is EOL, not supported after
kernel 3.14.*, so I'd like to switch to nouveau. I'm having trouble
getting nouveau to work at all, it is giving me a blank screen and
apparently not grabbing my keyboard (ctrl:swapcaps has no effect).

Nothing stands out as errors in Xorg.0.log, same errors are both under
nvidia and nouveau, but nvidia gives me a useable desktop. Both seem
to detect my monitor (benq) correctly.

---
$ grep '(EE)' Xorg.0.log.nvidia Xorg.0.log.nouveau | grep -v '(WW)'
Xorg.0.log.nvidia:[39.193] (EE) systemd-logind: failed to get
session: PID 2112 does not belong to any known session
Xorg.0.log.nouveau:[35.428] (EE) systemd-logind: failed to get
session: PID 2167 does not belong to any known session
Xorg.0.log.nouveau:[37.322] (EE) NOUVEAU(0): [COPY] failed to
allocate class.
---
The PID belongs to /usr/bin/X, see below.
---
I'm running gentoo-sources-4.3.3 kernel with experimental feature to
select Haswell architecture. The host is a virtual machine running
under app-emulation/xen-4.6.0-r6. Driver is
x11-drivers/xf86-video-nouveau-1.0.11, use-flag glamor enabled.

Have you passed the graphics card through to the VM?
Yes, it is passed through. No audio on the card, so there is only one 
function, and X manages to get EDID data from the monitor, so there is 
obviously some kind of connection in place.

Is the user trying to run the X server in the video group?

(pardon the LOCALE setting, should be intelligible none the less :-/ .
-
$ ls -l /dev/dri/*
crw-rw+ 1 root video 226,   0 jan.  13 10:05 /dev/dri/card0
crw-rw  1 root video 226,  64 jan.  13 10:05 /dev/dri/controlD64
crw-rw+ 1 root video 226, 128 jan.  13 10:05 /dev/dri/renderD128
# id hakon
uid=1001(hakon) gid=1001(hakon) 
grupper=1001(hakon),0(root),4(adm),7(lp),10(uucp),27(sudo),29(audio),37(operator),44(video),50(staff),60(games),100(users),1011(realtime),131(lpadmin),123(fuse),139(quaggavty),143(wireshark),162(davfs2),992(wheel),85(usb)


---
Just for completeness, some more data:
-
gt ~ # dmesg | grep -i nou
[0.00] Build-time adjustment of leaf fanout to 64.
[0.00] RCU: Adjusting geometry for rcu_fanout_leaf=64, nr_cpu_ids=15
[   14.036936] nouveau :00:05.0: NVIDIA GT200 (0a0f80b1)
[   14.146653] nouveau :00:05.0: bios: version 62.00.74.00.0b
[   14.312151] nouveau :00:05.0: fb: 1024 MiB GDDR3
[   14.483500] nouveau :00:05.0: DRM: VRAM: 1024 MiB
[   14.485210] nouveau :00:05.0: DRM: GART: 1048576 MiB
[   14.487219] nouveau :00:05.0: DRM: TMDS table version 2.0
[   14.489185] nouveau :00:05.0: DRM: DCB version 4.0
[   14.491000] nouveau :00:05.0: DRM: DCB outp 00: 04000320 0028
[   14.493539] nouveau :00:05.0: DRM: DCB outp 01: 01000322 00020030
[   14.495835] nouveau :00:05.0: DRM: DCB outp 02: 01121336 0f220e00
[   14.498363] nouveau :00:05.0: DRM: DCB outp 03: 01121332 00020e00
[   14.501005] nouveau :00:05.0: DRM: DCB outp 04: 02132336 0f220d00
[   14.503523] nouveau :00:05.0: DRM: DCB outp 05: 02132332 00020d00
[   14.505994] nouveau :00:05.0: DRM: DCB conn 00: 1030
[   14.507959] nouveau :00:05.0: DRM: DCB conn 01: a146
[   14.510084] nouveau :00:05.0: DRM: DCB conn 02: 00050246
[   14.512625] nouveau :00:05.0: DRM: DCB conn 03: 0360
[   14.727073] nouveau :00:05.0: DRM: MM: using CRYPT for buffer copies
[   15.344825] nouveau :00:05.0: DRM: allocated 1920x1080 fb: 
0x5, bo 8803ae4fb800

[   15.349163] fbcon: nouveaufb (fb0) is primary device
[   16.026158] nouveau :00:05.0: fb0: nouveaufb frame buffer device
[   16.140035] [drm] Initialized nouveau 1.3.0 20120801 for :00:05.0 
on minor 0


Grepping for "nv" just gets me a sata_nv


Systemd appears to complicate things greatly.  Have you tried to use
startx?
Tried it now. Nvidia drivers gives me a useable twm , nouveau gives med 
no signal to the monitor (goes into sleep mode immediately).


With nvidia-drivers:
-
0:hakon@gt:~$ startx
xauth:  file /home/hakon/.serverauth.16044 does not exist


X.Org X Server 1.17.4
Release Date: 2015-10-28
X Protocol Version 11, Revision 0
Build Operating System: Linux 3.18.22-gentoo x86_64 Gentoo
Current Operating System: Linux gt 4.3.3-gentoo #1 SMP Mon Jan 11 
15:45:22 CET 2016 x86_64
Kernel command line: 
BOOT_IMAGE=/boot/kernel-genkernel-x86_64-4.3.3-gentoo 
root=UUID=a202adf2-9b91-4d10-97ff-ab12aeb6009f ro intel_iommu=on 
init=/usr/lib/systemd/systemd tmem net.ifnames=0 console=tty0 
console=ttyS0,115200n8

Build Date: 02 November 2015  01:48:49AM

Current version of pixman: 0.32.8
Before reporting problems, check http://wiki.x.org
to make sure that you have the latest ver

Re: [gentoo-user] Nouveau blank screen

[Håkon Alstadheim]

Just for the record: Tried switcing off gfx_passthru, and over to 
device-model qemu-xen. This gave me another (virtual) graphics-card on 
this vm. After adding BusID  "00:05:0" to the device section, I'm right 
back where I started :-(




On 01/13/2016 10:19 AM, Håkon Alstadheim wrote:


On 01/12/2016 10:17 PM, lee wrote:

Thanks for helping, I' try to answer in-line.

Håkon Alstadheim <ha...@alstadheim.priv.no> writes:


I have an old but good graphics card, "NVIDIA Corporation GT200GL
[Quadro FX 3800]". The proprietary driver is EOL, not supported after
kernel 3.14.*, so I'd like to switch to nouveau. I'm having trouble
getting nouveau to work at all, it is giving me a blank screen and
apparently not grabbing my keyboard (ctrl:swapcaps has no effect).

Nothing stands out as errors in Xorg.0.log, same errors are both under
nvidia and nouveau, but nvidia gives me a useable desktop. Both seem
to detect my monitor (benq) correctly.

---
$ grep '(EE)' Xorg.0.log.nvidia Xorg.0.log.nouveau | grep -v '(WW)'
Xorg.0.log.nvidia:[39.193] (EE) systemd-logind: failed to get
session: PID 2112 does not belong to any known session
Xorg.0.log.nouveau:[35.428] (EE) systemd-logind: failed to get
session: PID 2167 does not belong to any known session
Xorg.0.log.nouveau:[37.322] (EE) NOUVEAU(0): [COPY] failed to
allocate class.
---
The PID belongs to /usr/bin/X, see below.
---
I'm running gentoo-sources-4.3.3 kernel with experimental feature to
select Haswell architecture. The host is a virtual machine running
under app-emulation/xen-4.6.0-r6. Driver is
x11-drivers/xf86-video-nouveau-1.0.11, use-flag glamor enabled.

Have you passed the graphics card through to the VM?
Yes, it is passed through. No audio on the card, so there is only one 
function, and X manages to get EDID data from the monitor, so there is 
obviously some kind of connection in place.

Is the user trying to run the X server in the video group?

(pardon the LOCALE setting, should be intelligible none the less :-/ .
-
$ ls -l /dev/dri/*
crw-rw+ 1 root video 226,   0 jan.  13 10:05 /dev/dri/card0
crw-rw  1 root video 226,  64 jan.  13 10:05 /dev/dri/controlD64
crw-rw+ 1 root video 226, 128 jan.  13 10:05 /dev/dri/renderD128
# id hakon
uid=1001(hakon) gid=1001(hakon) 
grupper=1001(hakon),0(root),4(adm),7(lp),10(uucp),27(sudo),29(audio),37(operator),44(video),50(staff),60(games),100(users),1011(realtime),131(lpadmin),123(fuse),139(quaggavty),143(wireshark),162(davfs2),992(wheel),85(usb)


---
Just for completeness, some more data:
-
gt ~ # dmesg | grep -i nou
[0.00] Build-time adjustment of leaf fanout to 64.
[0.00] RCU: Adjusting geometry for rcu_fanout_leaf=64, 
nr_cpu_ids=15

[   14.036936] nouveau :00:05.0: NVIDIA GT200 (0a0f80b1)
[   14.146653] nouveau :00:05.0: bios: version 62.00.74.00.0b
[   14.312151] nouveau :00:05.0: fb: 1024 MiB GDDR3
[   14.483500] nouveau :00:05.0: DRM: VRAM: 1024 MiB
[   14.485210] nouveau :00:05.0: DRM: GART: 1048576 MiB
[   14.487219] nouveau :00:05.0: DRM: TMDS table version 2.0
[   14.489185] nouveau :00:05.0: DRM: DCB version 4.0
[   14.491000] nouveau :00:05.0: DRM: DCB outp 00: 04000320 0028
[   14.493539] nouveau :00:05.0: DRM: DCB outp 01: 01000322 00020030
[   14.495835] nouveau :00:05.0: DRM: DCB outp 02: 01121336 0f220e00
[   14.498363] nouveau :00:05.0: DRM: DCB outp 03: 01121332 00020e00
[   14.501005] nouveau :00:05.0: DRM: DCB outp 04: 02132336 0f220d00
[   14.503523] nouveau :00:05.0: DRM: DCB outp 05: 02132332 00020d00
[   14.505994] nouveau :00:05.0: DRM: DCB conn 00: 1030
[   14.507959] nouveau :00:05.0: DRM: DCB conn 01: a146
[   14.510084] nouveau :00:05.0: DRM: DCB conn 02: 00050246
[   14.512625] nouveau :00:05.0: DRM: DCB conn 03: 0360
[   14.727073] nouveau :00:05.0: DRM: MM: using CRYPT for buffer 
copies
[   15.344825] nouveau :00:05.0: DRM: allocated 1920x1080 fb: 
0x5, bo 8803ae4fb800

[   15.349163] fbcon: nouveaufb (fb0) is primary device
[   16.026158] nouveau :00:05.0: fb0: nouveaufb frame buffer device
[   16.140035] [drm] Initialized nouveau 1.3.0 20120801 for 
:00:05.0 on minor 0


Grepping for "nv" just gets me a sata_nv


Systemd appears to complicate things greatly.  Have you tried to use
startx?
Tried it now. Nvidia drivers gives me a useable twm , nouveau gives 
med no signal to the monitor (goes into sleep mode immediately).


With nvidia-drivers:
-
0:hakon@gt:~$ startx
xauth:  file /home/hakon/.serverauth.16044 does not exist


X.Org X Server 1.17.4
Release Date: 2015-10-28
X Protocol Version 11, Revision 0
Build Operating System: Linux 3.18.22-gentoo x86_64 Gentoo
Current Operating System: Linux gt 4.3.3-gentoo #1 SMP Mon Jan 11 
15:45:22 CET 2016 x86_64
Kernel command line: 
BOOT_IMAGE=/boot/kernel-genkernel-x86_64-4.3.3-gentoo 
root=UUID=a202adf2-9b91-4d10-97ff-

Re: [gentoo-user] Nouveau blank screen

[Håkon Alstadheim]

Den 11. jan. 2016 18:22, skrev Alexander Kapshuk:
[...]
> Section "Device"
>
>  Identifier "noveau device"
>
>
> Typo here. This should read 'nouveau'.
>
Thanks for checking, but I believe the Identifier is just for human
consumption, i.e. it does not matter. The Driver line has it spelled
correctly :/ .
>From Xorg.0.log:

[35.341] (==) No device specified for screen "Default Screen Section".
Using the first device section listed.
[35.341] (**) |   |-->Device "noveau device"


 I've corrected the typo however.
> What is the output of 'grep NOUVEAU /path/to/your/kernel/sources/.config'?
>
zgrep -i nouveau /proc/config.gz
CONFIG_DRM_NOUVEAU=m
CONFIG_NOUVEAU_DEBUG=5
CONFIG_NOUVEAU_DEBUG_DEFAULT=3
# CONFIG_DRM_NOUVEAU_BACKLIGHT is not set

 
>  
>
>  Driver "nouveau"
> EndSection
> XORGX11
> ;;
>

Still stumped :-/ . I've since upgraded to gentoo-sources-4.4.0. Still 
no go.

[gentoo-user] Nouveau blank screen

[Håkon Alstadheim]

I have an old but good graphics card, "NVIDIA Corporation GT200GL 
[Quadro FX 3800]". The proprietary driver is EOL, not supported after 
kernel 3.14.*, so I'd like to switch to nouveau. I'm having trouble 
getting nouveau to work at all, it is giving me a blank screen and 
apparently not grabbing my keyboard (ctrl:swapcaps has no effect).


Nothing stands out as errors in Xorg.0.log, same errors are both under 
nvidia and nouveau, but nvidia gives me a useable desktop. Both seem to 
detect my monitor (benq) correctly.


---
$ grep '(EE)' Xorg.0.log.nvidia Xorg.0.log.nouveau | grep -v '(WW)'
Xorg.0.log.nvidia:[39.193] (EE) systemd-logind: failed to get 
session: PID 2112 does not belong to any known session
Xorg.0.log.nouveau:[35.428] (EE) systemd-logind: failed to get 
session: PID 2167 does not belong to any known session
Xorg.0.log.nouveau:[37.322] (EE) NOUVEAU(0): [COPY] failed to 
allocate class.

---
The PID belongs to /usr/bin/X, see below.
---
I'm running gentoo-sources-4.3.3 kernel with experimental feature to 
select Haswell architecture. The host is a virtual machine running under 
app-emulation/xen-4.6.0-r6. Driver is 
x11-drivers/xf86-video-nouveau-1.0.11, use-flag glamor enabled.


I've run emerge -e @world after installing the kernel, just in case some 
part of the system was incompatible with the kernel. Specifically pam 
has been recompiled several times. I have jiggled countless other knobs, 
to no effect.


-- systemctl log for kdm (disabled means it is loaded manually, 
autologin into fluxbox works) : ---

# systemctl status -l kdm
● kdm.service - KDM Display Manager
   Loaded: loaded (/etc/systemd/system/kdm.service; disabled; vendor 
preset: enabled)

   Active: active (running) since ma. 2016-01-11 16:56:11 CET; 28min ago
 Main PID: 2157 (kdm)
   CGroup: /system.slice/kdm.service
   ├─2157 /usr/bin/kdm -debug 0x108 -nodaemon
   └─2167 /usr/bin/X -br -novtswitch -quiet :0 vt7 -nolisten 
tcp -auth /var/run/xauth/A:0-y7Kmbb


jan. 11 16:56:18 gt kdm[2157]: receiving command from sub-daemon for 
display :0 ...

jan. 11 16:56:18 gt kdm[2157]:  -> 1
jan. 11 16:56:18 gt kdm[2157]: receiving int from sub-daemon for display 
:0 ...

jan. 11 16:56:18 gt kdm[2157]:  -> 1001 (0x3e9)
jan. 11 16:56:18 gt kdm[2157]: receiving string from sub-daemon for 
display :0 ...

jan. 11 16:56:18 gt kdm[2157]:  -> 6 bytes
jan. 11 16:56:18 gt kdm[2157]:  -> "hakon"
jan. 11 16:56:18 gt kdm[2157]: receiving string from sub-daemon for 
display :0 ...

jan. 11 16:56:18 gt kdm[2157]:  -> 8 bytes
jan. 11 16:56:18 gt kdm[2157]:  -> "fluxbox"
--
--
The [COPY] message is supposed to be harmless according to a discussion 
with the developers I found on some website, since my card does not 
support that feature.


I'm using this script to switch graphics-driver, before rebooting and 
launching kdm with autologin:


cat `which graphics-nvidia `
#!/bin/bash
NEW=`basename $0`
case $NEW in
graphics-nouveau)
eselect opengl set xorg-x11
eselect opencl set mesa
rm -f /etc/modprobe.d/blacklist.conf.gml
mv /etc/modprobe.d/blacklist.conf /etc/modprobe.d/blacklist.conf.gml
perl -pe 's(^ *blacklist nouveau)(#$&);s(^ *# *blacklist 
nvidia$)(blacklist nvidia)' /etc/modprobe.d/blacklist.conf.gml > 
/etc/modprobe.d/blacklist.conf

# dummy empty nvidia.rules
touch /etc/udev/rules.d/99-nvidia.rules
cat < /etc/X11/xorg.conf.d/10device.conf
Section "Device"
 Identifier "noveau device"
 Driver "nouveau"
EndSection
XORGX11
;;
graphics-nvidia)
rm -f /etc/modprobe.d/blacklist.conf.gml
mv /etc/modprobe.d/blacklist.conf /etc/modprobe.d/blacklist.conf.gml
perl -pe 's(^ *blacklist nvidia$)(#$&);s(^ *# *blacklist 
nouveau)(blacklist nouveau)' /etc/modprobe.d/blacklist.conf.gml > 
/etc/modprobe.d/blacklist.conf

eselect opengl set nvidia
eselect opencl set nvidia
# remove dummy empty nvidia.rules
rm /etc/udev/rules.d/99-nvidia.rules
cat < /etc/X11/xorg.conf.d/10device.conf
Section "Device"
 Identifier "nvidia device"
 Driver "nvidia"
EndSection
XORGNV

;;
esac
---

The script is not pretty, I know :-/

So, it seems everything is working with nouveau, except there is no 
picture and no keyboard.


[35.202] 
X.Org X Server 1.17.4
Release Date: 2015-10-28
[35.202] X Protocol Version 11, Revision 0
[35.202] Build Operating System: Linux 3.18.22-gentoo x86_64 Gentoo
[35.202] Current Operating System: Linux gt 4.3.3-gentoo #1 SMP Mon Jan 11 
15:45:22 CET 2016 x86_64
[35.202] Kernel command line: 
BOOT_IMAGE=/boot/kernel-genkernel-x86_64-4.3.3-gentoo 
root=UUID=a202adf2-9b91-4d10-97ff-ab12aeb6009f ro intel_iommu=on 
init=/usr/lib/systemd/systemd tmem net.ifnames=0 console=tty0 
console=ttyS0,115200n8
[35.202] Build Date: 02 November 2015  01:48:49AM
[35.202]  
[35.202] Current version of pixman: 0.32.8
[35.202]Before reporting problems, check http://wiki.x.org
to make sure that you have the 

Re: [gentoo-user] Re: OT - Midnight Commander and hiding terminal output

[Håkon Alstadheim]

Den 03. jan. 2016 16:56, skrev Skippy:
>
> On 01/02/2016 12:38 AM, Håkon Alstadheim wrote:
>> Den 01. jan. 2016 00:49, skrev Linux:
>>> On 12/30/2015 10:32 AM, Roman Dobosz wrote:
>>>
>>> snip
>>   Should be as easy as hitting Ctrl-L when your screen is messed up,
>> should it not? Don't use MC, so haven't tried it. If it does not work,
>> look in manual for key-binding for "redraw" .
>>
> I didn't know about ctrl-L. Sure enough that works as well.
> Thank you.  Good work around until I fix it otherwise.
> Skippy
>
>

ctrl-L (C-l for short)  is old standard terminal key-binding, sending an
actual control-character. Control-characters are like their regular
counterparts but with a numeric value 0x40 less. Others that you
probably know are C-c (0x03, break) , C-d, C-/ . More obscure are C-s
(aka XOFF) to suspend terminal output, and C-q (aka XON) to continue
output. All these are control-characters used forever on unix terminals.

Also fairly standard C-h (backspace) C-p (previous) C-n (next). C-j is
synonym for carriage return aka \r, C-m is line-feed aka \n. Good to
know when keyboard mappings get screwed up.

In this context C-l is actually sending page-break control-character,
which will usually redraw the terminal window, unless the app decides to
use it for something else.

These control-codes can be seen in an acii table, they are the first ten
code-points.

Re: [gentoo-user] Re: OT - Midnight Commander and hiding terminal output

[Håkon Alstadheim]

Den 01. jan. 2016 00:49, skrev Linux:
> On 12/30/2015 10:32 AM, Roman Dobosz wrote:
>
> snip
>
>> Just redirect the standard end error output to the void, like:
>>
>> --- 8< ~/.config/mc/mc.ext ---
>> include/video2
>>  Open=(mpv -vf-clr %f >/dev/null 2>&1 &)
>>  View=%view{ascii} midentify %f
>>  Edit=if [ -n "$DISPLAY" ]; then (avidemux3_qt4 %f 2>&1
>> >/dev/null &); fi
>> --- >8 
>>
>> so opening the file (pressing enter on the file), or editing it (via
>> pressing F4) I don;t see any output from neither mpv nor avidemux.
>> Note, that both of the processes are instantly put to the background
>> (amperand at the end of the command) so that mc is still operatable.
>>
>> Hope, that helps :)
>
> This does help.  Thank you very much.  I'll be trying it out soon.
> -Skippy
>
>
>
  Should be as easy as hitting Ctrl-L when your screen is messed up,
should it not? Don't use MC, so haven't tried it. If it does not work,
look in manual for key-binding for "redraw" .

Re: [gentoo-user] Hard drive noise

[Håkon Alstadheim]

Den 19. des. 2015 01:01, skrev Dale:
...
> It sounds like the heads are doing random
> reads/writes and the heads are moving but doing so noisily.  Thing is,
> there is no drive activity according to gkrellm or iotop.  All the
> drives should be basically idle. 
...
>  Model=ST3000DM001
What kind of file system ? How full ? Raid ? lvm ? What kind of SATA
controller? There is lots of stuff about these things I don't know, but
I /do/ know all those cause different quirks and telling more specifics
would maybe prod someone with experience with your particular type of
set-up to chime in.

Just as an example, ext4 will delay some of the work on a newly
formatted drive. Lots of other system activity can cause background
jobs. If this coincides with a bad sector on a drive, you get noise,
access latency and system wait. This system wait will not necessarily be
counted against any specific process in "top(1)" (depending on how your
kernel was compiled) , so you can even get 0%idle and still have no
specific process sticking out in "top".

Watch your logs and kernel messages for access errors. Difficulties in
reading might cause retries without actually timing out. If this is your
problem, it will persist until that sector is marked as failing AND
SOMETHING IS WRITTEN TO IT. The sector will not be mapped out until
something gets written to it.

Also, if you are running raid, google around for permutations of
"smartctl -T permissive -l scterc,70,70", seeing as this drive

is not a server type drive. But, remember, NEVER blindly run something
on your system that some random guy sent you by email without
understanding what it does. :-D .

[gentoo-user] Bug-reporting best practice? Xen/Gentoo -- Preparing to log interrupt info sometime after feb. 2016.

[Håkon Alstadheim]

Hi all, I'm not a developer but I would like to set up my Xen system to
get bug reports as useful as possible. At present I am logging the xen
console via serial port, and I am running Xen compiled from gentoo
ebuild with "debug" use flag. Is there anything more I need to set up to
be able to collect as much useful info as possible?

At present, this is my xen command line:
xen_commandline: ssd-xen-debug-marker console_timestamps=date
loglvl=all guest_loglvl=all sync_console iommu=1,verbose,debug
iommu_inclusive_mapping=1 com1=115200,8n1 console=com1 dom0_max_vcpus=4
dom0_vcpus_pin=1 dom0_mem=8G,max:8G cpufreq=xen:performance,verbose
tmem=1 sched_smt_power_savings=1 apic_verbosity=debug e820-verbose=1
core_parking=power

(yes, I know, 8G for dom0, well, it is also acting as file server, I
want to keep as much of the file-system in cache as I can )

Any changes I should make for maximally useful bug-reports?

This is in preparation for the 4.7 release (april 2016 I gather) . I
have some PCIe cards that are acting up, something to do with
interrupts. I see some activity on that front right now, so I've decided
to be an early adopter of 4.7, to be able to help as much as possible,
maybe get any tweaks I might need for my system done in the source.

I'm running gentoo, and have tried installing xen-*- from the
virtualization overlay, but it fails, seems it is trying to fetch the
qemu sources twice, and failing.  Is it me doing something wrong? Use
flag "system-qemu" does not help.

---
Additional info:
The PCIe cards that are acting up are:
00:1b.0 Audio device: Intel Corporation C610/X99 series chipset HD Audio
Controller (rev 05)
02:00.0 USB controller: ASMedia Technology Inc. ASM1042A USB 3.0 Host
Controller
81:00.0 USB controller: ASMedia Technology Inc. Device 1242

Trying to pass them through to any domain is highly hit-or-miss. Seems
like interrupt set-up fails quite often, but I have not been able to
find any actual error-messages in any logs. The two first there are
built-in to the motherboard, the last one is an add-in card. I have
tried several other add-in cards for USB, all acting decidedly weird.

Re: [gentoo-user] installing Gentoo as a xen guest

[Håkon Alstadheim]

Den 26. sep. 2015 14:00, skrev J. Roeleveld:
>
>> Depending on your hardware you will want to use hvm or pvm for
>> efficiency. (VT-x means hvm is more efficient).
> What do you base this on?
> Without VT-x, HVM doesn't even work, which means PV is only option.
I stand corrected :/ .

I'l refrain from confusing the issue further, I believe I was thinking
VT-d. This
 is
the right place to learn about guest types :). Some of the alphabet-soup
is explained here 

The rule of thumb I live by for my private, ad-hoc systems is: since I
have hardware support for virtualization and also VT-d, I use it, and I
keep references handy while i configure things :-D.


> With VT-x, PV still has higher performance as the drivers inside the guest 
> talk directly to the host.
>

PV on HVM will be best of both worlds, and drivers are available at
least for windows and linux, so I was taking pv drivers as a given.
Also, where I said VT-x read VT-d.

>> If running hvm on
>> quemu-xen-traditional, you HAVE to use a bootloader inside the VM, or
>> some kind of netboot/pvgrub thing. If running upstream quemu for a hvm,
>> you can choose. I find it less of a hassle to use bootloader inside the
>> VM.
> It's simple, if you don't have full access to the host.
> If you have full access, it's actually simpler as you don't have to worry 
> about boot-order, partitioning and a bootloader.
I'm sure you are right, it is just pvgrub is an extra piece of kit I
haven't bothered learning. I'll go back to lurking now :-~

Re: [gentoo-user] installing Gentoo as a xen guest

[Håkon Alstadheim]

Den 24. sep. 2015 14:53, skrev Rich Freeman:
> On Thu, Sep 24, 2015 at 6:05 AM, hw  wrote:
>> Hi,
>>
>> I'm installing Gentoo as a xen PV guest.  Do I need to install a bootloader
>> like grub, or should I rather just specify the kernel to boot in the
>> definition file of the guest?  If I do the latter, what about the kernel
>> command line?
>>
>> Is there anything I should consider?  The host is also running Gentoo.
>>
> You can do either.  I suspect it would be easier to just use grub.  I
> don't know if xen provides a way to provide a command-line, if not you
> could build a default command-line into your kernel.
>
Depending on your hardware you will want to use hvm or pvm for
efficiency. (VT-x means hvm is more efficient). If running hvm on
quemu-xen-traditional, you HAVE to use a bootloader inside the VM, or
some kind of netboot/pvgrub thing. If running upstream quemu for a hvm,
you can choose. I find it less of a hassle to use bootloader inside the
VM. If running pv, I BELIEVE you have to specify kernel and boot options
outside the VM-image, possibly through pvgrub. Pvgrub is supposed to be
able to fish out kernel and initrd from the VM disk. I never bothered to
get that working. Here is an example excerpt from a PVM I use (for a
debian vm):
--print.pvm---
builder = "generic"
kernel = "/etc/xen/wheezy/vmlinuz-3.16.0-0.bpo.4-amd64"
ramdisk = "/etc/xen/wheezy/initrd.img-3.16.0-0.bpo.4-amd64"
extra = "root=/dev/xvda console=hvc0 tmem"
--
For explanation check "man xl.cfg". When you specify
kernel in the vm config, you do NOT need a /boot/
directory on your vm.  Obviously when you build a kernel inside the vm,
it will end up in /boot, so you have to provide a way for the
dom0/hypervisor to load your kernel, either (when using full
virtualization) a regular boot-loader inside the vm, or (for a PV guest)
through some other means to let the dom0/hypervisor load your image.

Re: [gentoo-user] Re: Flaky USB 3.0

[Håkon Alstadheim]

Den 08. sep. 2015 05:32, skrev James:
> Håkon Alstadheim  alstadheim.priv.no> writes:
>
>
>> In a nutshell:
>> My PCIE USB 3.0 card keeps quitting, eg. no light from my laser mouse.
> I see no advantage to using usb3 for a mouse. Try to plug into another
> usb port. Look at your mobo manual and find one that is usb-2 and see
> if the mouse does not work reliable on that usb-2 port. This should
> at least get your mouse working correctly as you experiment/debug
> the usb-3 with other devices that need that sort of bw.
>
I use the mouse primarily for testing. It is a dead simple device that
always works, it is an old HP mouse. It gives off a nice red light when
the USB bus is operating, so it is easy to spot when there is any sign
of life.

I need a separate PCI device that I can pass through to a Xen domain.
Xen usb is in a bit of a flux at the moment. I've looked online for PCIE
USB 2.x cards, but there are none to be found. For my linux guests I can
use usbip, but I have not found a usbip driver for windows. Anyhow usbip
gets unwieldy when I have several domains that need keyboard

The PCIE usb card is still dead, I've found out that ASUS released a new
bios, barely half a year after the last update. Installed the latst BIOS
last night, but the only thing that bought me was another couple of
hours re-entering (from memory) the settings that I needed to get things
working. Back to square one.

Re: [gentoo-user] Flaky USB 3.0 -- typo-fix -- more logs

[Håkon Alstadheim]

Den 07. sep. 2015 21:23, skrev Håkon Alstadheim:
> Den 07. sep. 2015 20:26, skrev Håkon Alstadheim:
>> In a nutshell:
>> My PCIE USB 3.0 card keeps quitting, eg. no light from my laser mouse.
>> No life in anything I plug in.
>> Lately I've had three cold starts (that usually will revive the thing) ,
>> but still no go. Will try unplugging for a couple of minutes next time I
>> am able to reboot.
Did a reboot and tried wiggling some knobs in the bios. Noticed the
mouse lit up while I was in there, but it goes out at once when boot
starts. :/ .
Got a dmesg output from gentoo on the bare metal:

0:gentoo ~ # echo 1 > /sys/bus/pci/devices/:04:00.0/remove
0:gentoo ~ # dmesg -c
0:gentoo ~ # echo 1 > /sys/bus/pci/rescan
0:gentoo ~ # dmesg -c
[  863.007306] pci :04:00.0: [1106:3483] type 00 class 0x0c0330
[  863.007331] pci :04:00.0: reg 0x10: [mem 0xc650-0xc6500fff 64bit]
[  863.007450] pci :04:00.0: PME# supported from D0 D1 D2 D3hot D3cold
[  863.007739] dmar: [Firmware Bug]: RMRR entry for device 04:00.0 is
broken - applying workaround
[  863.018667] pci :04:00.0: BAR 0: assigned [mem
0xc650-0xc6500fff 64bit]
[  863.018682] ast :0c:00.0: BAR 6: [??? 0x flags 0x2] has
bogus alignment
[  863.034924] pci :04:00.0: xHCI HW did not halt within 16000 usec
status = 0x1004
[  863.035127] xhci_hcd :04:00.0: xHCI Host Controller
[  863.035138] xhci_hcd :04:00.0: new USB bus registered, assigned
bus number 7
[  863.059990] xhci_hcd :04:00.0: Host not halted after 16000
microseconds.
[  863.059994] xhci_hcd :04:00.0: can't setup: -110
[  863.059998] xhci_hcd :04:00.0: USB bus 7 deregistered
[  863.060057] xhci_hcd :04:00.0: init :04:00.0 fail, -110
[  863.060063] xhci_hcd: probe of :04:00.0 failed with error -110

---
Now I have got to stop fiddling with this box, need to keep it up 24/7
for at least a week now before I try anything drastic again : - ~
---

>> General note about the commands and switches below: Most have been found
>> by trial and error, and the setup used may be utterly WRONG. If you spot
>> any mistakes, please point them out to me.
>>
>> Googling around led me to run then following, with the output seen
>> below. Similar output can be seen right after boot.
>>
>> 
>> root@steam:~# echo 1 > /sys/bus/pci/devices/:00:05.0/remove
>> root@steam:~# dmesg -c
>> root@steam:~# echo "1" > /sys/bus/pci/rescan
>> root@steam:~# dmesg -c
>> [ 3164.999480] pci :00:05.0: [1106:3483] type 00 class 0x0c0330
>> [ 3165.008397] pci :00:05.0: reg 0x10: [mem 0xf42a4000-0xf42a4fff 64bit]
>> [ 3165.051792] pci :00:05.0: BAR 0: assigned [mem
>> 0xf400-0xf4000fff 64bit]
>> [ 3165.059245] pci :00:05.0: can't find IRQ for PCI INT A; probably
>> buggy MP table
>> [ 3165.075349] pci :00:05.0: xHCI HW did not halt within 16000 usec
>> status = 0x1004
>> [ 3165.075961] xhci_hcd :00:05.0: can't find IRQ for PCI INT A;
>> probably buggy MP table
>> [ 3165.076150] xhci_hcd :00:05.0: xHCI Host Controller
>> [ 3165.076157] xhci_hcd :00:05.0: new USB bus registered, assigned
>> bus number 1
>> [ 3165.100651] xhci_hcd :00:05.0: Host not halted after 16000
>> microseconds.
>> [ 3165.100653] xhci_hcd :00:05.0: can't setup: -110
>> [ 3165.100655] xhci_hcd :00:05.0: USB bus 1 deregistered
>> [ 3165.100774] xhci_hcd :00:05.0: init :00:05.0 fail, -110
>> [ 3165.100776] xhci_hcd: probe of :00:05.0 failed with error -110
>> ---
>>
>> This happens on Asus hardware viz:
>>
>> -
>> # smbios-sys-info
>> Libsmbios version:  2.2.28
>> Product Name:   Z10PE-D8 WS
>> Vendor: ASUSTeK COMPUTER INC.
>> BIOS Version:   1001
>> -
>> On top of this I'm running Xen, viz:
>> 0:gentoo ~ # xl info
>> host   : gentoo
>> release: 4.0.9-gentoo
>> version: #1 SMP Tue Sep 1 01:10:52 CEST 2015
>> machine: x86_64
>> ...
>> cpu_mhz: 2394
>> hw_caps:
>> bfebfbff:2c100800::7f00:77fefbff::0021:37ab
>> virt_caps  : hvm hvm_directio
>> ...
>> xen_version: 4.5.1
>> xen_caps   : xen-3.0-x86_64 xen-3.0-x86_32p hvm-3.0-x86_32
>> hvm-3.0-x86_32p hvm-3.0-x86_64
>> ...
>> xen_commandline: ssd-xen-g-4-00-marker  loglvl=warn
>> guest_loglvl=warn noreboot iommu=1,verbose iommu_inclusive_mapping=1
>> com1=115200,8n1 console=com1 dom0_max_vcpus=4 dom0_vcpus_pin=1
>> dom0_

Re: [gentoo-user] Re: Flaky USB 3.0

[Håkon Alstadheim]

> Den 8. sep. 2015 kl. 17.07 skrev James <wirel...@tampabay.rr.com>:
> 
> Håkon Alstadheim  alstadheim.priv.no> writes:
> 
> 
>>>> My PCIE USB 3.0 card keeps quitting, eg. no light from my laser mouse.
> 
>>> I see no advantage to using usb3 for a mouse. Try to plug into another
>>> usb port. Look at your mobo manual and find one that is usb-2 and see
>>> if the mouse does not work reliable on that usb-2 port. This should
>>> at least get your mouse working correctly as you experiment/debug
>>> the usb-3 with other devices that need that sort of bw.


I need the separate PCIE device so I can pass it through to a virtual machine. 
There are no USB 2.x PCIE cards that I know of. 
> 
> USB negotiates with devices as to what speed/standard to use. If your 
> device only needs a lesser speed (usb 1.1) it's not going to negoiate
> with the host for a usb-3 speed. If you want to debug usb3 speeds
> you'll need a truly usb 3.0 device. A hi res usb 3.0 camera device
> would be keen to test your usb 3.0 buss/chip/negotiations.
> 
I posted a log saying something about interrupt not working, so I would think 
it fails before there is any traffic on the USB bus. If there is anything to 
see, wireshark is supposed to be able to capture it. I'll see if I can get it 
set up.
> 
>> I use the mouse primarily for testing. It is a dead simple device that
>> always works, it is an old HP mouse. It gives off a nice red light when
>> the USB bus is operating, so it is easy to spot when there is any sign
>> of life.
> 
> 
> You need to find some usb sniffer software and see what the negotiations
> are doing when the device is first hooked up and then running a while.
> There are sniffers for windows aplenty, so you might have to do this
> diagnostic work under windows. [1] I do not think this list is
> complete so look around. Also look in the sources for the usb 3.0
> kernel as often the comments are most excellent for device and driver
> debugging. Some vendors use several different chipsets for the same
> product name, so start looking for the usb chipset for that card/board/device.
> 
> Also go through your kernel configs and verify what you need is set
> in the kernel properly (and loading if as a module).
> 
> 
Yes, I'm sure my settings are worth a look, (kernel and BIOS). Problem is I 
don't know what I'm looking for. Currently I'm on alert for anything to do with 
MSI, irq, DMA, RMRR and a whole slew of other thing I barely understand.
> 
> 
> [1] http://www.linux-usb.org/tools.html
Thanks for the link!

Re: [gentoo-user] Re: Flaky USB 3.0

[Håkon Alstadheim]

>>> 
>>> I need the separate PCIE device so I can pass it through to a virtual 
> machine. There are no USB 2.x PCIE cards that I know of. 
>> 
>> How about this one?
>> http://www.startech.com/Cards-Adapters/USB-2/Card/5-Port-USB-20-PCI-Express-Card~PEX400USB2
>> 
>> Startech model # PEX400USB2
>> 
>> It is a 5 port USB2 PCI-e adapter. 4 external, one internal.
>> 

Looks discontinued  "may be available at resellers"

>> That was a bugger to find.
>> 
>> Dan
> 
> There's a few on ebay too for less than half that price.
> 
Yes, ebay may turn out to be my only option. Feels wrong to buy outdated tech 
though.

Trying a 3.1 card first. Different chipset. Might allow full "docking" 
functionality, which would be just perfect.

[gentoo-user] Flaky USB 3.0

[Håkon Alstadheim]

In a nutshell:
My PCIE USB 3.0 card keeps quitting, eg. no light from my laser mouse.
No life in anything I plug in.
Lately I've had three cold starts (that usually will revive the thing) ,
but still no go. Will try unplugging for a couple of minutes next time I
am able to reboot.

General note about the commands and switches below: Most have been found
by trial and error, and the setup used may be utterly WRONG. If you spot
any mistakes, please point them out to me.

Googling around led me to run then following, with the output seen
below. Similar output can be seen right after boot.


root@steam:~# echo 1 > /sys/bus/pci/devices/:00:05.0/remove
root@steam:~# dmesg -c
root@steam:~# echo "1" > /sys/bus/pci/rescan
root@steam:~# dmesg -c
[ 3164.999480] pci :00:05.0: [1106:3483] type 00 class 0x0c0330
[ 3165.008397] pci :00:05.0: reg 0x10: [mem 0xf42a4000-0xf42a4fff 64bit]
[ 3165.051792] pci :00:05.0: BAR 0: assigned [mem
0xf400-0xf4000fff 64bit]
[ 3165.059245] pci :00:05.0: can't find IRQ for PCI INT A; probably
buggy MP table
[ 3165.075349] pci :00:05.0: xHCI HW did not halt within 16000 usec
status = 0x1004
[ 3165.075961] xhci_hcd :00:05.0: can't find IRQ for PCI INT A;
probably buggy MP table
[ 3165.076150] xhci_hcd :00:05.0: xHCI Host Controller
[ 3165.076157] xhci_hcd :00:05.0: new USB bus registered, assigned
bus number 1
[ 3165.100651] xhci_hcd :00:05.0: Host not halted after 16000
microseconds.
[ 3165.100653] xhci_hcd :00:05.0: can't setup: -110
[ 3165.100655] xhci_hcd :00:05.0: USB bus 1 deregistered
[ 3165.100774] xhci_hcd :00:05.0: init :00:05.0 fail, -110
[ 3165.100776] xhci_hcd: probe of :00:05.0 failed with error -110
---

This happens on Asus hardware viz:

-
# smbios-sys-info
Libsmbios version:  2.2.28
Product Name:   Z10PE-D8 WS
Vendor: ASUSTeK COMPUTER INC.
BIOS Version:   1001
-
On top of this I'm running Xen, viz:
0:gentoo ~ # xl info
host   : gentoo
release: 4.0.9-gentoo
version: #1 SMP Tue Sep 1 01:10:52 CEST 2015
machine: x86_64
...
cpu_mhz: 2394
hw_caps:
bfebfbff:2c100800::7f00:77fefbff::0021:37ab
virt_caps  : hvm hvm_directio
...
xen_version: 4.5.1
xen_caps   : xen-3.0-x86_64 xen-3.0-x86_32p hvm-3.0-x86_32
hvm-3.0-x86_32p hvm-3.0-x86_64
...
xen_commandline: ssd-xen-g-4-00-marker  loglvl=warn
guest_loglvl=warn noreboot iommu=1,verbose iommu_inclusive_mapping=1
com1=115200,8n1 console=com1 dom0_max_vcpus=4 dom0_vcpus_pin=1
dom0_mem=8G,max:8G cpufreq=xen,performance,verbose tmem=1 
dom0_nodes=0,relaxed sched_smt_power_savings=1
cc_compiler: x86_64-pc-linux-gnu-gcc (Gentoo 4.9.3 p1.2,
pie-0.6.3) 4.9.3
cc_compile_by  :
cc_compile_domain  : alstadheim.priv.no
cc_compile_date: Mon Aug 31 05:54:06 CEST 2015
xend_config_format : 4
--
Note: 4.0.9 is the gcc compiler version.
--
Dom 0 is:
# uname -a
Linux gentoo 4.0.9-gentoo #1 SMP Tue Sep 1 01:10:52 CEST 2015 x86_64
Intel(R) Xeon(R) CPU E5-2620 v3 @ 2.40GHz GenuineIntel GNU/Linux
--
# cat /proc/cmdline
placeholder root=LABEL=ssdroot ro
xen-pciback.hide=(00:1b.0)(02:00.0)(04:00.0)(81:00.0)(82:00.0)(09:00.0)
console=hvc0 console=vga domodules domdadm dolvm intel_iommu=on
earlyprintk=xen usbip_core.usbip_debug_flag=0 usbcore.autosuspend=-1
--
The USB card in question is:
root@steam:~# lspci -vvv -s :00:05.0
00:05.0 USB controller: VIA Technologies, Inc. Device 3483 (rev 01)
(prog-if 30 [XHCI])
Subsystem: VIA Technologies, Inc. Device 3483
Physical Slot: 5
Control: I/O- Mem+ BusMaster- SpecCycle- MemWINV- VGASnoop- ParErr-
Stepping- SERR- FastB2B- DisINTx-
Status: Cap+ 66MHz- UDF- FastB2B- ParErr- DEVSEL=fast >TAbort-
SERR- 

Re: [gentoo-user] Flaky USB 3.0 -- typo-fix

[Håkon Alstadheim]

Den 07. sep. 2015 20:26, skrev Håkon Alstadheim:
> In a nutshell:
> My PCIE USB 3.0 card keeps quitting, eg. no light from my laser mouse.
> No life in anything I plug in.
> Lately I've had three cold starts (that usually will revive the thing) ,
> but still no go. Will try unplugging for a couple of minutes next time I
> am able to reboot.
>
> General note about the commands and switches below: Most have been found
> by trial and error, and the setup used may be utterly WRONG. If you spot
> any mistakes, please point them out to me.
>
> Googling around led me to run then following, with the output seen
> below. Similar output can be seen right after boot.
>
> 
> root@steam:~# echo 1 > /sys/bus/pci/devices/:00:05.0/remove
> root@steam:~# dmesg -c
> root@steam:~# echo "1" > /sys/bus/pci/rescan
> root@steam:~# dmesg -c
> [ 3164.999480] pci :00:05.0: [1106:3483] type 00 class 0x0c0330
> [ 3165.008397] pci :00:05.0: reg 0x10: [mem 0xf42a4000-0xf42a4fff 64bit]
> [ 3165.051792] pci :00:05.0: BAR 0: assigned [mem
> 0xf400-0xf4000fff 64bit]
> [ 3165.059245] pci :00:05.0: can't find IRQ for PCI INT A; probably
> buggy MP table
> [ 3165.075349] pci :00:05.0: xHCI HW did not halt within 16000 usec
> status = 0x1004
> [ 3165.075961] xhci_hcd :00:05.0: can't find IRQ for PCI INT A;
> probably buggy MP table
> [ 3165.076150] xhci_hcd :00:05.0: xHCI Host Controller
> [ 3165.076157] xhci_hcd :00:05.0: new USB bus registered, assigned
> bus number 1
> [ 3165.100651] xhci_hcd :00:05.0: Host not halted after 16000
> microseconds.
> [ 3165.100653] xhci_hcd :00:05.0: can't setup: -110
> [ 3165.100655] xhci_hcd :00:05.0: USB bus 1 deregistered
> [ 3165.100774] xhci_hcd :00:05.0: init :00:05.0 fail, -110
> [ 3165.100776] xhci_hcd: probe of :00:05.0 failed with error -110
> ---
>
> This happens on Asus hardware viz:
>
> -
> # smbios-sys-info
> Libsmbios version:  2.2.28
> Product Name:   Z10PE-D8 WS
> Vendor: ASUSTeK COMPUTER INC.
> BIOS Version:   1001
> -
> On top of this I'm running Xen, viz:
> 0:gentoo ~ # xl info
> host   : gentoo
> release: 4.0.9-gentoo
> version: #1 SMP Tue Sep 1 01:10:52 CEST 2015
> machine: x86_64
> ...
> cpu_mhz: 2394
> hw_caps:
> bfebfbff:2c100800::7f00:77fefbff::0021:37ab
> virt_caps  : hvm hvm_directio
> ...
> xen_version: 4.5.1
> xen_caps   : xen-3.0-x86_64 xen-3.0-x86_32p hvm-3.0-x86_32
> hvm-3.0-x86_32p hvm-3.0-x86_64
> ...
> xen_commandline: ssd-xen-g-4-00-marker  loglvl=warn
> guest_loglvl=warn noreboot iommu=1,verbose iommu_inclusive_mapping=1
> com1=115200,8n1 console=com1 dom0_max_vcpus=4 dom0_vcpus_pin=1
> dom0_mem=8G,max:8G cpufreq=xen,performance,verbose tmem=1 
> dom0_nodes=0,relaxed sched_smt_power_savings=1
> cc_compiler: x86_64-pc-linux-gnu-gcc (Gentoo 4.9.3 p1.2,
> pie-0.6.3) 4.9.3
> cc_compile_by  :
> cc_compile_domain  : alstadheim.priv.no
> cc_compile_date: Mon Aug 31 05:54:06 CEST 2015
> xend_config_format : 4
> --
Corrected: My gcc-version is x86_64-pc-linux-gnu-4.9.3
> --
> Dom 0 is:
> # uname -a
> Linux gentoo 4.0.9-gentoo #1 SMP Tue Sep 1 01:10:52 CEST 2015 x86_64
> Intel(R) Xeon(R) CPU E5-2620 v3 @ 2.40GHz GenuineIntel GNU/Linux
> --
> # cat /proc/cmdline
> placeholder root=LABEL=ssdroot ro
> xen-pciback.hide=(00:1b.0)(02:00.0)(04:00.0)(81:00.0)(82:00.0)(09:00.0)
> console=hvc0 console=vga domodules domdadm dolvm intel_iommu=on
> earlyprintk=xen usbip_core.usbip_debug_flag=0 usbcore.autosuspend=-1
> --
> The USB card in question is:
> root@steam:~# lspci -vvv -s :00:05.0
> 00:05.0 USB controller: VIA Technologies, Inc. Device 3483 (rev 01)
> (prog-if 30 [XHCI])
> Subsystem: VIA Technologies, Inc. Device 3483
> Physical Slot: 5
> Control: I/O- Mem+ BusMaster- SpecCycle- MemWINV- VGASnoop- ParErr-
> Stepping- SERR- FastB2B- DisINTx-
> Status: Cap+ 66MHz- UDF- FastB2B- ParErr- DEVSEL=fast >TAbort-
> SERR-  Interrupt: pin A routed to IRQ 10
> Region 0: Memory at f400 (64-bit, non-prefetchable) [size=4K]
> Capabilities: [80] Power Management version 3
> Flags: PMEClk- DSI- D1- D2- AuxCurrent=0mA
> PME(D0-,D1-,D2-,D3hot-,D3cold-)
> Status: D0 NoSoftRst+ PME-Enable- DSel=0 DScale=0 PME-
> Capabilities: [90] MSI: Enable- Count=1/1 Maskable- 64bit+
> Address:   Data: 
>

Re: [gentoo-user] Firefox 38.1.0 :-(

[Håkon Alstadheim]

 Den 2. aug. 2015 kl. 02.24 skrev Frank Steinmetzger war...@gmx.de:
 
 On Sat, Aug 01, 2015 at 05:31:45PM +, Alan Mackenzie wrote:
 
 The 'default' profile selection does not stick.  Deleting the new
 'dev-edition-default' profile causes it to be recreated afresh at the
 next start up.
 
 Yes.  This is the sort of developer attitude that is making me want to
 use a proper browser.  What the heck does a browser need prefiles for,
 anyway?  It's supposed to be a web browser, for goodness sake.
 
 I don’t use it often, but having it is nice. Actually many contemporary
 browsers do. I have my main profile that I usually use. But if I want to
 visit some site that shall not have any way of obtaining information I don’t
 want it to have (or because it just would not work with my restrictive
 security setup), I quickly create a throwaway profile.
 

My online bank used to require java. I have a separate profile I use for that 
bank and nothing else.

Re: [gentoo-user] Picking dom0 OS for new XEN server

[Håkon Alstadheim]

On 22. jan. 2015 08:49, Tomas Mozes wrote:

On 2015-01-22 08:32, Håkon Alstadheim wrote:

Pondering dom0 os for new Xen server (Xeon e5 2620 v3 cpu) . Server is
for home use. (routing/firewall; dns/dhcp, mail servers, etc. etc;
linux gui desktop; linux media-server; windows as separate domains).

I have basic skills in debian, and gentoo. A bit rusty on the
rpm-based distros. Totally unfamiliar with *BSD. Total newbie in XEN

So, given this list is bound to be a bit biased, : why would I pick
gentoo for dom0, and what version of XEN would I use ?


Picking gentoo for dom0 is probably the same question as why would I 
use gentoo instead of other linux distribution. Me personally have 
very good experience in running xen dom0 machines on gentoo. If you 
separate your dom0 machine and the services are in domUs, then the 
dom0 is very small, clean and easy to maintain (for years). Ask on the 
debian list and I'm sure they will answer the same way so it's your 
choice ;)


If it's a new server you may try the new 4.5 release. I don't know how 
stable it is, I just started my first VM on 4.5 today, but I'm running 
4.3 in production and 4.4 in pre-production. Since you don't need to 
migrate in production, it's not a problem I believe.





Arye you using the 4.5 ebuilds from portage ?
-
Debian Jessie tells me (as an example):
# apt-cache search xen-hyp
xen-hypervisor-4.4-amd64 - Xen Hypervisor on AMD64
xen-hypervisor-4.0-amd64 - The Xen Hypervisor on AMD64
xen-hypervisor-4.0-i386 - The Xen Hypervisor on i386
xen-hypervisor-4.1-amd64 - Xen Hypervisor on AMD64
xen-hypervisor-4.1-i386 - Xen Hypervisor on i386
# apt-cache policy xen-hypervisor-4.4-amd64
xen-hypervisor-4.4-amd64:
  Installed: (none)
  Candidate: 4.4.1-6
  Version table:
 4.4.1-6 0
990 http://ftp.se.debian.org/debian/ jessie/main amd64 Packages
500 http://ftp.se.debian.org/debian/ unstable/main amd64 Packages

--
My gentoo box tells me:
# eix -e xen
* app-emulation/xen
 Available versions:  4.2.5-r3^t ~4.2.5-r4^t *4.3.3-r3^t 
~*4.3.3-r4^t ~*4.4.1-r4^t ~*4.4.1-r5^t ~*4.5.0^t {custom-cflags debug 
efi flask pae xsm}

 Homepage:http://xen.org/
 Description: The Xen virtual machine monitor
--
So it seems that if I decide xen 4.5, gentoo might be less hassle ?

[gentoo-user] Picking dom0 OS for new XEN server

[Håkon Alstadheim]

Pondering dom0 os for new Xen server (Xeon e5 2620 v3 cpu) . Server is 
for home use. (routing/firewall; dns/dhcp, mail servers, etc. etc; linux 
gui desktop; linux media-server; windows as separate domains).


I have basic skills in debian, and gentoo. A bit rusty on the rpm-based 
distros. Totally unfamiliar with *BSD. Total newbie in XEN


So, given this list is bound to be a bit biased, : why would I pick 
gentoo for dom0, and what version of XEN would I use ?

Re: [gentoo-user] new installation - partitions

[Håkon Alstadheim]

On 05. sep. 2014 04:44, Daniel Frey wrote:
It is possible to boot in EFI mode off of a USB, as I used a Mint ISO 
to boot from in EFI mode. I would presume the USB needs to have the 
FAT partition that EFI requires. Dan 
Sounds good. Having /boot on a stick makes it easy to have whatever I 
might need available when my fancy-schmanzy root-fs fails to show up at 
boot :-) .
Always have a known good kernel and initramfs to fall back on, and tuck 
away some extra tools on the stick. Put some (statically linked) 
*parted,lvm,md and formatting binaries on there and you can easily 
rearrange things before mounting the root fs.

Re: [gentoo-user] new installation - partitions

[Håkon Alstadheim]

On 04. sep. 2014 16:52, Tom H wrote:

On Thu, Sep 4, 2014 at 10:00 AM, Rich Freeman ri...@gentoo.org wrote:

When booting from EFI you need a GPT boot partition (FAT - ugh) that
actually contains the image that gets booted, so it needs to have room
for at least a couple of kernels/initramfs - so that will be larger.


I'm working on getting a new motherboard, Will I still be able to have 
my boot filesystem on a flash-stick? Currently I have everything except 
/boot on LVM on top of Physical Volumes on unpartitioned raid volumes. 
Having a single drive with an odd size makes swapping drives around when 
they fail and drop out of the raid a hassle, and I do not want to waste 
2G on every drive just to have a 2G boot partition. A flash stick (and 
another one for backup) is very pleasant to work with. Especially when i 
bork my initramfs or need to run maintenance without mounting my root 
filesystem. Will this work on an EFI board ?

[gentoo-user] ebuild for canon ufr ii lt driver . Anyone got it ?

[Håkon Alstadheim]

Canon just came out with a driver for my printer i-Sensys lbp 7100cn 
(yay!) . It was promised september 2013 (grumble). It is downloadable 
from here : 
http://www.canon.no/Support/Consumer_Products/products/printers/Laser/i-SENSYS_LBP7100Cn.aspx 
. I could not for the life of me create a scraper for that page (grumble).


Anyhow: Does anyone have an ebuild to install from the 
Linux_UFRIILT_PrinterDriver_V100_uk_EN.tar.gz into gentoo ?


The common files seem to be the same as the ones in 
net-print/cndrvcups-common-lb (except for a version change from 2.70 to 
2.80) . The other file however is different, 
cndrvcups-ncap-1.00-1.tar.gz inside the above mentioned tar-file. Just 
running the commands in the README does not work, due to differences in 
the gentoo directory-structure I guess.


--
Håkon Alstadheim / N-7510 Skatval / email: ha...@alstadheim.priv.no
tlf: 74 82 60 27 mob: 47 35 39 38
http://alstadheim.priv.no/hakon/

Re: [gentoo-user] Re: How to stem the flood of new packages?

[Håkon Alstadheim]

On 12. feb. 2013 02:14, Stroller wrote:

On 12 February 2013, at 00:30, Peter Humphrey wrote:

On Tuesday 12 February 2013 00:01:00 Dale wrote:


This makes me think, if a Government suddenly decided for people to
switch which side of the road they are supposed to drive on.

This happened in my lifetime in, I think, Sweden. I can't remember when
though.

They had a special day for it, with no driving permitted between certain hours 
whilst they changed road signs over:
http://en.wikipedia.org/wiki/Dagen_H

This image probably looks worse than it was in Stockholm, but it illustrates 
Dale's point:
http://en.wikipedia.org/wiki/File:Kungsgatan_1967.jpg

Stroller.
The story in Norway is that the Swedes did the switchover gradually, to 
ease into it. Lorries changed over on monday, the rest of the vehicles 
on tuesday :-).

Re: [gentoo-user] Re: can one tell me: gentoo vs opensuse

[Håkon Alstadheim]

Den 07. des. 2011 01:15, skrev Indi:

On Wed, Dec 07, 2011 at 12:40:01AM +0100, Grant Edwards wrote:

On 2011-12-06, ny6...@gmail.comny6...@gmail.com  wrote:

On Tue, Dec 06, 2011 at 11:15:31AM +, Neil Bothwick wrote:

On Tue, 6 Dec 2011 05:54:22 -0500, LinuxIsOne wrote:


But I don't know why the creators of Gentoo made it so difficult for
beginners!

Because Gentoo is not for beginners, there are already plenty of distros,
like Mandriva and Ubuntu, catering for first time users. Gentoo is a
power users distro.

And yet the documentation is clear enough for anyone to follow along.
Which leads me to my next point: the Gentoo documentation is far and
away the best of any distro I have tried.

Definitely.

The Ubuntu documentation seems to be mainly user-forum threads full of
wrong answers posted by people who didn't understand the question.


Whoever writes these docs deserves a heap of accolades for his
efforts.

The Gentoo docs are indeed brilliant.


They really are. When you have great documentaion, a shell, and a keyboard
that *is* user-friendly! ;)!



It tells you something that quite often when I google for 
some-problem-description and debian/ubuntu/suse/windows/whatever, 
quite often the only hits to come up are gentoo-related. On debian the 
gentoo answer is usually useful, on ubuntu ... well, 'nuff sed.

Re: [gentoo-user] Tcl in your system...

[Håkon Alstadheim]

Den 29. nov. 2011 06:03, skrev Michael Mol:

On Mon, Nov 28, 2011 at 11:33 PM, Pandu Poluanpa...@poluan.info  wrote:

Just wondering if anyone here ever use Tcl for scripting (i.e., automating
repetitive procedures) or even *gasp* serious programming.

Not me, but Tcl is one of the best-represented langauges on the site I run.



I use it for small stuff (with expect) regarding keeping my cheap 
consumer-grade routers/switches running. Those things usually have 
either a telnet or ssh interface that allows me to 
monitor/configure/reboot, but they are usually too ficle to be 
programmed in any predictable way. Muddle through a problem manually, 
and record the solution in an expect script to keep until next time. 
Over the years I have accumulated several that are stable enough to run 
unattended. Last couple of times that my most troublesome wifi-box got 
alzheimers, it got rebooted automatically. Dynamic DNS (10+ hostnames) 
and ADSL rebooting happens within 5 minutes of my ISP deciding to do 
something weird. expectk brings a semblance of sanity.


No serious programming though. Did use exmh as my main mailclient until 
~9 years ago. It is still being developed, look into that if you want an 
example of Tcl in a largeish project.

Re: [gentoo-user] New HD monitor stretches everything. How to teach Xorg?

[Håkon Alstadheim]

Den 24. aug. 2010 04:27, skrev Kevin O'Gorman:
I had to replace an 4:3 Westinghouse monitor this weekend.  I got a 
new ASUS VH242H, which is very wide.  But Xorg is still running 
1280x1024, instead of the monitor's normal 1920x1080, according to 
xorg logs because of lack of video memory (using the ATI on the 
motherboard).  I can make the screen use a 4:3 aspect ratio, so I'm up 
and running, much better than I started, but I'd like to do better.


I guess I've gotta look for a video card, but all I have is PCIX 
slots, so I don't want to put a lot of money into it (I'll be 
upgrading the mobo when finances permit -- which is not right now.)


Just did a cursory read of the entire thread here. I notice the card is 
on the mobo, did you try to see if there is a BIOS setting to increase 
the amount of video RAM? I.e enter BIOS setup during boot, and look 
around in the chip setup.

Re: [gentoo-user] Constraining X display resolutions

[Håkon Alstadheim]

Frank Steinmetzger skrev:

Am Mittwoch, 28. April 2010 schrieb Mick:

  

However, Linux GUIs are very good at geometric upscaling, so I suggest
increasing font and icon sizes.


I'll try that anyway; it may give me a better compromise. Thanks.
  

I've had the same problem on a high resolution (1920x1080), small size
screen (15.6).  The characters are tiny and anything else but native
resolution makes images and characters blurred.  The solution was to
increase the font size on the terminals and KDE apps.  However, I don't
know how to make the characters in the Firefox menus and body larger.  Am
I supposed to run gconftool-2 with some esoteric options?



There's a package that lets GTK apps look like KDE apps, including font, 
called kcm_gtk. It adds a page to System settings under

Appearance-Appearance called GTK styles and fonts.
  


Running fluxbox myself, but the idea should work across desktops: use 
xrandr and lie to X about the physical  size of your screen. On my TV I 
run xrandr first once without arguments to get the actual size, dive the 
sizes by two and run xrandr like so: xrandr --fbmm 443x247. This is a 
32 16:9 TV. Stick this last bit somewhere early in your login-sequence. 
Works beautifylly.

Re: [gentoo-user] Opera 10.10 says I've got Java?

[Håkon Alstadheim]

Mick skrev:

Version information

Version 10.10
Build 4742
Platform Linux
System i686, 2.6.30-gentoo-r8
Qt library 4.5.3
Java Java Runtime Environment installed

However, for all I know I do not have anything Java installed and 
about:plugins does not show Java either.  Is this worth reporting as bug?
  



Opera does not run java as a plugin, you point opera directly at the 
jre. Find the setting grouped somewhere under web contents.

Re: [gentoo-user] my xorg-server 1.6 seems a bit unstable - what am I doing wrong?

[Håkon Alstadheim]

I've been having some issues myself (mythtv manages to mess up my screen 
royally, I need to switch to a text virtual console and back to X to 
make the screen legible after viewing a TV-recording in myth). I just 
noticed that the version of the headers for xcb do not match the version 
on libxcb itself. Could this be the cause of my (and Daves) problems? I 
have libxcb 1.4, but headers version 1.5, no 1.4 available for the 
headers. (running a new sync of my portage tree now, just in case I've 
missed something)

Viz:

1:medisin ~ # eix xcb
[I] x11-libs/libxcb
Available versions:  1.0 1.1 ~1.1.90.1 1.4-r1 {debug doc selinux}
Installed versions:  1.4-r1(22:28:37 10/11/09)(-debug -doc -selinux)
Homepage:http://xcb.freedesktop.org/
Description: X C-language Bindings library

[I] x11-libs/xcb-util
Available versions:  0.3.3 ~0.3.4 ~0.3.5 0.3.6 {debug test}
Installed versions:  0.3.6(18:11:17 10/08/09)(-debug -test)
Homepage:http://xcb.freedesktop.org/
Description: X C-language Bindings sample implementations

* x11-misc/xcb
Available versions:  2.4 {motif}
Homepage:http://www.goof.com/pcg/marc/xcb.html
Description: Marc Lehmann's improved X Cut Buffers

[I] x11-proto/xcb-proto
Available versions:  1.0 1.1 ~1.2 1.5
Installed versions:  1.5(05:16:09 10/06/09)
Homepage:http://xcb.freedesktop.org/
Description: X C-language Bindings protocol headers

Found 4 matches.

Re: [gentoo-user] Question about xorg and a kill process

[Håkon Alstadheim]

Daniel da Veiga wrote:

On Wed, May 27, 2009 at 10:03, Dale rdalek1...@gmail.com wrote:
  

Hung Dang wrote:


I often use Ctrl+F1 to F6 to back to the command line and if your X is
fine you can back the X screen using Ctrl+F7. Or if you want to kill
your X then Ctrl+Alt+Backspace may be helpful.

Hung


Dale wrote:
  

I know the subject is a bit lacking but here goes.  I'm thinking about
trying this xorg-server upgrade again.  I been thinking about a way to
do this and not have to pull the plug on my rig if it fails, which I bet
it does.  This is the command I am thinking about trying.
/etc/init.d/xdm start  sleep 5m  /etc/init.d/xdm stop

I'm thinking this way.  Start X first.  If it fails, it will stop in 5
minutes and come back to a console.  Think this will work?  If xorg
works, I can switch back to a console and ctrl C the command and
carry on.

Thoughts?  Better ideas?

Dale

:-)  :-)



That won't work because if xorg-server fails, my keyboard doesn't work
when I switch to X.  If the keyboard doesn't work, I can't switch back
to anything or type anything.

I done been through this one time.  I'm trying to figure out how to get
back to console with a keyboard that doesn't work at all.




Best way to do this is using a remote shell (SSH for example) in
another machine. If that's not an option or X driver fails in
conflicts with the kernel (mine did before I found a suitable config)
then you're pretty much lost, cause your video is gone for good.

You can try SYSREQ combinations to kill the server and if that fails
even cleanly reboot the rig, but as I said before, depending on the
problem your video is gone.

  

stick a chvt 1  after the xdm stop.

Re: [gentoo-user] GPRS connection through usb

[Håkon Alstadheim]

Simon wrote:

Hi there,
  i've recently acquired a pda (htc tytn) and a good data plan.  Right
now, i can access the internet from my computer by setting up the
phone as a wifi ad-hoc station, but this takes so much battery, it's
not the optimal setup.

  What i would need is to use the usb cable to connect to the cell
modem.  But I'm a bit lost as to what to do...  the documents i found
on this don't seem to talk about the same things (there seems to be
many ways of doing this...?) and many are out of date...

  I was wondering if any of you could give me a general idea of what i
need (specially if you could clarify whats the best vocabulary to use
in google) or confirm what I found so far...

  From what i understand, I believe i will need some kind of kernel
module (device driver) which will turn the usb device into a
networking interface.  Then I think i would setup some kind of
connection normally (is it PPP? or something else?).  The cellphone
would also need to be setup to share its data connection to usb (but
this seems quite simple... the linux part is the problem).

  
You do NOT need PPP, but you do need a fairly recent kernel (= 2.6.27 
at least, not sure of the details). You need at least 
CONFIG_USB_RNDIS_WLAN as m or y. With udev and 'm' things work mostly 
automatically on the linux side. On the phone you enable advanced 
networking (or else you will see the phone as ttyUSB? and you will need 
ppp). Then you plug in and run dmesg -c until things quiet down. You 
should see a new eth? interface. Enable internet sharing on the phone, 
and you should see in dmesg -c that your new ethernet interface 
disappears and reappears. If it does not reappear your kernel is too 
old. On my gentoo box i can now start the interface like a regular 
ethernet interface. I have a debian box with a kernel from unstable 
where dhcp fails on the phone 'ehternet' device, so I have to do ip 
addr add and ip route add manually to get the network going (somehow 
the ifup command on debian prints the assigned ip and the gateway ip, 
gut does not configure the device properly). Do some googling to fill in 
the details, I'm not using this regularly.

Re: [gentoo-user] ssmtp att woes

[Håkon Alstadheim]

Since this thread has been going on for so long without a resolution, I 
thought I'd mention that I recently switched to nullmailer from ssmtp. 
Im using port 587 with STARTTLS, and I find nullmailer way easier to set 
up. Just put --user and --pass parameters in /etc/nullmailer/remotes.

Re: [gentoo-user] Sending RSS feeds via email

[Håkon Alstadheim]

Alan McKinnon wrote:

On Thursday 18 September 2008 22:24:32 [EMAIL PROTECTED] wrote:
  

Does anyone know a tool that fetches RSS feeds and sends new articles to a
given email address? 
Momesso Andrea




What's wrong with cron, wget, diff and mail?


  
See http://rss2email.infogami.com/. It is packaged in debian, but it 
looks like it is not in portage. Caveat: I have not tried it.

[gentoo-user] Top values don't add up

[Håkon Alstadheim]

If anybody knows a better arena to field this question, please let me know.

My system is a single-core old fashioned intel system. uname -a reports:
Linux medisin 2.6.25-gentoo-r7 #3 PREEMPT Sun Aug 3 11:40:41 CEST 2008 
i686 Intel(R) Pentium(R) 4 CPU 2.00GHz GenuineIntel GNU/Linux


# qfile `which top`
sys-process/procps (/usr/bin/top)
# eix -e procps
[I] sys-process/procps
Available versions:  3.2.4-r3 3.2.5-r1 3.2.6 3.2.7 {n32}
Installed versions:  3.2.7(03:27:10 05/13/08)(-n32)
Homepage:http://procps.sourceforge.net/
Description: Standard informational utilities and 
process-handling tools


Top reports ~70% idle, while at the same time the topmost couple of 
processes are reported as using 70%CPU. Is there anything I could use 
that reports more sensible values ?


I'm running the machine for multi-media-use, and I would like to make 
sure that I tune the media-programs to leave sufficcient cpu to handle 
the odd house-keeping task, while at the same time doing as much 
post-processing for image and sound quality as possible. Is there a way 
to get top to make sense, or are there other tools you good people would 
recommend ?

Re: [gentoo-user] Top values don't add up

[Håkon Alstadheim]

Alan McKinnon wrote:

On Tuesday 19 August 2008 15:30:07 Dale wrote:
  

Håkon Alstadheim wrote:


 SNIP 

I'm running the machine for multi-media-use, and I would like to make
sure that I tune the media-programs to leave sufficcient cpu to handle
the odd house-keeping task, while at the same time doing as much
post-processing for image and sound quality as possible. Is there a
way to get top to make sense, or are there other tools you good people
would recommend ?
  

The only thing I can think of that may help is to use the nice command
to give some processes more or less priority.  That may or may not help
you any but it is a thought.

Otherwise, I tend to agree with Alan.  Leave it to the kernel.



I thought of one thing that might help - perhaps the media apps in question 
are single threaded and might benefit from a different preemption model 
algorithm. Give the old-fashioned server model a try instead of desktop or 
low-latency desktop. It's worth a try, YMMV 
  


When I say tune I mean things like picking a resolution and a 
deinterlace method for video that is as good as possible, while still 
leaving enough headroom to avoid uneven playback. Given that a schedule 
update or a backup run might kick in at a bad time despite all efforts 
to write a good crontab, knowing when to stop is not always easy. I run 
all system tasks under nice ionice -c3, and they will still cause 
hiccups if the system is maxed out. Looks like the good old trial and 
error is the only method that will work. Being on the bleeding edge, 
that means erring on the side of caution, or else you never know whether 
you have hit a system bottleneck or a bug in the software  :-) .


mplayer kept saying my system was too slow, while the cpu was idling at 
20%. Turns out top was correct in that instance, mplayer was 
misinterpreting input data, trying to play back at half the intended 
frame-rate. Now I'm past that hurdle and adding deinterlacing and other 
filters. I'll just have to hold back on the temptation to go all out on 
the filter options :-)


Thanks for keeping me from wasting any more time with top though.

Re: [gentoo-user] Re: Horribly off-topic linux distro question...

[Håkon Alstadheim]

Mick wrote:

On Thursday 07 February 2008, Neil Bothwick wrote:
  

On Thu, 07 Feb 2008 15:27:51 +0100, Michael Schmarck wrote:


In the context of online banking, where Windows of some flavour is the
desktop OS, I see a substantial risk arising through spyware and/or
viruses.  I suspect that a neat way to mitigate this would be to run
an OS from a CD which offers nothing more fancy than a basic
web-browser.

Is there anything like this already available?


DSL should come fairly close.
  

Dillo doesn't work with the online banking sites, and many others, that I
tried.



Basic web browsers do not have the javascript, Java (and soon enough flash?) 
functionality that the majority of banking sites require.  Wouldn't Knoppix 
with its Firefox and equivalents do the job for you, after you set root and 
knoppix passwds?  BTW, Konqueror will also work with many banking sites, but 
you may need to change the browser agent identification, treatment of cookies 
and so on.  YMMV.
  
I've had some success (one of two sites) with the opera browser. Free as 
in beer.

--
gentoo-user@lists.gentoo.org mailing list

Re: [gentoo-user] Postfix problem

[Håkon Alstadheim]

James Homuth wrote:

At 06:58 AM 29/01/2008, you wrote:

Hello,

I have problem with postfix. I just want to send email to other mail 
server from my new postfix/courier server. Every time I write address 
different then local server (postfix) returns info 5.7.1 (some 
address): Relay access denied. What should I do? I don't know is it 
forwarding or relaying. What is wrong?
Sounds like there's a problem in the way you have postfix configured. 
Namely, you didn't configure $mydomain, $mydestination, and 
$mynetworks. Either that, or one or all of them are misconfigured. 
http://www.postfix.org/BASIC_CONFIGURATION_README.html
... and once those are ok, you need to have a permit_mynetworks BEFORE 
the reject_unauth_destination (you definitely want that last one in 
there).


The postconf command is a good way to get the actual values that are 
in effect. Run postconf | less and search for reject_unauth_destination.


--
gentoo-user@lists.gentoo.org mailing list

Re: [gentoo-user] Re: [konqueror] Can't invoke and editor when `view source'

[Håkon Alstadheim]

[EMAIL PROTECTED] wrote:

Mick [EMAIL PROTECTED] writes:

  

Don't know the first thing about emacs, but it may need bringing up
a terminal first which in turn runs emacs.  You may want to try
selecting the Run in terminal or invoke it like so:

xterm -e /usr/bin/emacs



This should not be a factor with X enabled emacs.  And in fact calling
emacs at a cmd prompt just brigs up emacs in it own window, not
another xterm.

However, and surprisingly it does work... Inserting the xterm -e
command  at: 
   right click/ open with/ other


Brings first an xterm which immediately spawns a new emacs window (not
in the xterm but on its own)

I'm pretty sure this is not what SHOULD happen though.  I SHOULD be
able to just insert /usr/bin/emacs  since it does not run in an
xterm. But... thanks .. at least I can edit a page with emacs now.

  
I have this same problem on some machines. Notably the ones where I've 
put the most cruft in .emacs. I suspect that there is some bug that 
stops garbage-collection from happening during startup so emacs runs out 
of memory. Somehow having a tty attached works around that. A way to 
test is to launch emacs in the background from a terminal (with  at the 
end). If emacs hangs, I do fg in the shell  and hit enter a couple of 
times and emacs will (sometimes) continue its initialization process.


A better work-around than firing off emacs in the foreground is to make 
more of your .emacs do AUTOLOAD rather than REQUIRE or LOAD.


Better yet is to have emacs-server or gnuserv running, and putting 
emacsclient/gnuclient in the browser editor-config.

Re: [gentoo-user] Re: [konqueror] Can't invoke and editor when `view source'

[Håkon Alstadheim]

[EMAIL PROTECTED] wrote:

Håkon Alstadheim [EMAIL PROTECTED] writes:

  

[EMAIL PROTECTED] wrote:


Mick [EMAIL PROTECTED] writes:

  
  

Don't know the first thing about emacs, but it may need bringing up
a terminal first which in turn runs emacs.  You may want to try
selecting the Run in terminal or invoke it like so:

xterm -e /usr/bin/emacs



This should not be a factor with X enabled emacs.  And in fact calling
emacs at a cmd prompt just brigs up emacs in it own window, not
another xterm.

However, and surprisingly it does work... Inserting the xterm -e
command  at:right click/ open with/ other

Brings first an xterm which immediately spawns a new emacs window (not
in the xterm but on its own)

I'm pretty sure this is not what SHOULD happen though.  I SHOULD be
able to just insert /usr/bin/emacs  since it does not run in an
xterm. But... thanks .. at least I can edit a page with emacs now.

  
  

I have this same problem on some machines. Notably the ones where I've
put the most cruft in .emacs. I suspect that there is some bug that
stops garbage-collection from happening during startup so emacs runs
out of memory. Somehow having a tty attached works around that. 



A little more on this.  Do you mean it works just fine on some
machines as well?

Something to test your theory... (I tried it here with no better
results) is from Konq, right click/open with/other  and put /usr/bin/emacs -Q

Which will start emacs with no site-file or ~/.emacs being loaded.
I tried that here but still just got the bouncing emacs icon/cursor.

Can you start emacs without problems from konqueror at right
click/open with/  on any machine? I mean without `xterm -e emacs'


  
Not running konqueror here, but yes, emacs would fire up quite reliably 
when I had it as my source editor in mozilla, way back when. You most 
definitely SHOULD be able to run it without a terminal window. If you 
are really keen on getting this to work, you could write a little script 
like so to use as your editor: 

#!/bin/bash
date ~/emacs.log
emacs $@ ~/emacs.log 21
---
Come to think of it (while looking up the $@ semantics), could it be 
that the argument fed from konqueror has spaces in it? Maybe it needs 
quoting?


Experiment with adding /dev/null onto there (that never did it for me). 
Have a look at what the log says.



--
gentoo-user@lists.gentoo.org mailing list

Re: [gentoo-user] Be careful when using dhcp with a LiveCD

[Håkon Alstadheim]

Billy Holmes wrote:


I guess eventually all dhcp implementations will catch up with this 
change,
although for now it is bound to create some problems with particular 
DHCP


You could try disconnecting your cable modem for about 10 minutes, 
ensuring the ISP recognizes that it's offline, and thus remove the 
entries in it's DHCP table for your connection.


then when you boot your system normally, it might work like it used to.

Yes, that sound reasonable. The livecd was probably shut down without 
releasing the lease on the ip-address. Check the manual page on the 
dhcp-client on the live cd. There is probably some signal you can send 
it to make it release its lease.


When the regular os comes back up in this situation, it wants /its/ old 
ip back. This might hang on a server restriction on only one ip per 
subscriber/mac address, or it might just trigger a bug somewhere.


--
[EMAIL PROTECTED] mailing list

Re: [gentoo-user] {OT} web/mail server as nameserver

[Håkon Alstadheim]

Crayon Shin Chan wrote:
 On Friday 11 May 2007 18:48, [EMAIL PROTECTED] wrote:

   
 Poor security of bind is imho similar superstition as it is
 for sendmail: once in the past this software had some problem,
 so now a lot of people think they should forever avoid using it...
 

 If the OP doesn't need any bind-specific feature then why not use djbdns 
 which has a better security track record. djb software are built from the 
 ground up to be secure (as is possible), he also splits the program 
 into smaller executables, each having a specific job thus making each of 
 them secure a simpler task. Whilst bind and sendmail have made 
 substantial efforts to be more secure, they are still built on legacy and 
 bloated monolithic code.

   
Just to fill in the picture a bit, the djb* software also has a long
flip-the-bird-at-any-rfc-you-don't-like track-record.

-- 
Håkon Alstadheim 
spamtrap: [EMAIL PROTECTED] -- 1 hit  you are out

-- 
[EMAIL PROTECTED] mailing list

Re: [gentoo-user] [OT] umask and sticky bit

[Håkon Alstadheim]

Jorge Almeida wrote:
 I would like the default permissions for directories created by a
 particular user to be 1775. Is there some way to achieve this? I think
 umask doesn't deal with the sticky bit.

umask is a MASK. The application suggests permissions on new files,
umask turns off the bits you do not want. I can not think of a fully
automatic solution to what you want to do.

-- 
Håkon Alstadheim / N-7510 Skatval / email:[EMAIL PROTECTED]
tlf: 74 82 60 27 mob: 47 35 39 38
http://alstadheim.priv.no/hakon/ 
spamtrap: [EMAIL PROTECTED] -- 1 hit  you are out

-- 
gentoo-user@gentoo.org mailing list

Re: [gentoo-user] The next step in AV

[Håkon Alstadheim]

Grant wrote:
 I currently have an HDTV hooked up to a desktop computer running
 Gentoo and xfce4, all controlled by a wireless keyboard/mouse from the
 couch.  It's awesome.  However, I think the next step is to control
 everything from a laptop on the couch.  There would be a normal xfce4
 desktop on the laptop, but you could issue certain commands to have
 video or audio played through the desktop connected to the TV across
 the room.

 How would you set this up?  Maybe the actual xfce4 session is being
 run on the desktop and the laptop uses VNC or NX to control that
 session?  Is there a way to have certain audio or video launched on
 the laptop via VNC or NX, but outputted by the desktop?

 - Grant
The easy way is to use a big cursor theme and just use your laptop as a
remote keyboard/mouse for your TV. (in addition to stuff like mythweb
and the like)

Do like so: (assuming tv runs on display :0, which it most likely does )
Get mythtv or whatever running by auto-login and startup script. Assume
that your tv is running as user tv, hostname tv. From your laptop:
ssh -Y [EMAIL PROTECTED], then x2x -to :0 -north. This makes the tv-screen 
look
like an extension of your laptop screen, situated above (north) of your
laptop. Now whenever you move your mouse up, out of the laptop screen,
your laptop keyboard and mouse controls the tv.

Some details:
You might need to run emerge x2x on your tv box first. Also look for
packages named something with cursor. Kde allows you to select cursor
theme easily once you have the themes installed.

The crusial bit to get your head around:
An X session runs on a display, named like :0 or localhost:10.0. The
default for a standard X session starts at :0, so your tv will most
likely be :0. Ssh forwarded displays, xvnc displays etc. will have other
values. Check out the DISPLAY environment variable. To be allowed to
write to a display you use the xauth system, usually by way of a file
name ~/.Xauthority, check man xauth. Just log in as the user owning
the display (as per the .Xauthority file) and issue export DISPLAY=:0
to make whatever apps you start show themselves on the tv rather than on
your laptop.

-- 
Håkon Alstadheim
spamtrap: [EMAIL PROTECTED] -- 1 hit  you are out


-- 
gentoo-user@gentoo.org mailing list

[gentoo-user] emerge sys-apps/hwinfo-13.0 fails, missing function dbus_connection_disconnect in sys-apps/dbus-1.0.2

[Håkon Alstadheim]

Emerge sys-apps/hwinfo-13.0 fails, missing function
dbus_connection_disconnect ( should be in sys-apps/dbus-1.0.2 ?)

I'm not really a C programmer but I tried replacing disconnect with
close in that one call from hal.c in hwinfo. That makes it compile ok,
but running it makes dbus complain that I'm not allowed to close a
shared connection.

The most logical explanation AFAICT is that this version of hwinfo
expects some other version of dbus. Anybody have any hints ?

Coming from SuSE I'm sort-of lost when I don't have hwinfo, I go there
all the time to double-check what drivers I should use for disks, X
(drm) and all manner of other things.

The obvious replacement for hwinfo --cpu is cat /proc/cpuinfo, but
for the other stuff I feel like I'm stumbling in the dark...

-- 
Håkon Alstadheimtlf: 74 82 60 27 mob: 47 35 39 38
7510 Skatval
http://alstadheim.priv.no/hakon/

-- 
gentoo-user@gentoo.org mailing list