Re: [gentoo-user] from Firefox52: NO pure ALSA?, WAS: Firefox 49.0 & Youtube... Audio: No

[Neil Bothwick]

On Tue, 27 Dec 2016 19:53:47 +0100, lee wrote:

> > I would imagine because it cannot be used without some initial
> > configuration. The default provides the greatest reliability out of
> > the box, at the expense of less readable (which is not the same as
> > unrecognisable, a value judgement you are imposing on the names)
> > names.  
> 
> I call them unrecognisable because they are hard to recognise, as in
> hard to read and impossible to remember.  I find that annoying.  I can
> call them "annoying names" if you prefer that :)

I do, or "difficult to remember" or "cryptic", but they are not
unrecognisable - except to those that wish them to be.
 
> > There is nothing wrong with wanting things to work as you do, but it
> > requires input to do so. It you have to start editing files to make it
> > work properly, there is little point in making it the default.  
> 
> Right, and it could work without editing files manually.  A
> configuration file assigning editable names to the annoying names could
> be created automatically and filled by assigning the name an interface
> already has to it (because when it has a name, the name is known, which
> is easier than trying to make up all possible names in advance).  Then
> only if you wanted you would edit the configuration file to assign the
> name(s) of your choosing, and if you don't want to do that, you simply
> get the names you get now.  There would be no change to how the names
> are now, only an additional option.
> 
> That would also have the advantage that when the annoying name of an
> interface changes, you can choose to either adjust all configuration
> files in which you have specified a particular interface or simply
> adjust the one configuration file that assigns the names.
> 
> I actually wonder why they didn't virtualise the names.  It makes too
> much sense for not to do it, and you could do likewise with other
> devices (especially disks).

That's a reasonable approach, and you could have the ebuild set it up
with a USE flag. All it takes is for someone that cares enough about it
to do something.


-- 
Neil Bothwick

I have seen the truth, and it makes no sense.


pgpi_cPQaXuTe.pgp
Description: OpenPGP digital signature

Re: [gentoo-user] At last! A Qt5 version of KMail-2 - but here be dragons!

[Neil Bothwick]

On Tue, 27 Dec 2016 19:54:57 +, J. Roeleveld wrote:

> Speaking of English. I always wonder which would be a better match for
> someone who uses English mostly to deal with international friends and
> computers. I am Dutch myself, which means it's officially a second
> language for me. (Always wondered about that, as I speak several
> languages and fluency is simply a matter of which I speak regularly). I
> usually throw a dice to decide between US and UK whenever I get asked.

As a native English speaking pedant, I would throw "a die" or "some dice"
to decide, dice is the plural :P


-- 
Neil Bothwick

WORM: (n.) acronym for Write Once, Read Mangled. Used to describe a
  normally-functioning computer disk of the very latest design.


pgp84eUdsm5cU.pgp
Description: OpenPGP digital signature

Re: [gentoo-user] from Firefox52: NO pure ALSA?, WAS: Firefox 49.0 & Youtube... Audio: No

[Neil Bothwick]

On Tue, 27 Dec 2016 20:21:19 +0100, lee wrote:

> >  Even more reasonable:
> >
> >  eselect news read new
> >
> > will only come up with the latest as yet unread news, rather than a
> > long list which could have accumulated over the years.  
> 
> It seems to be clearing out the list automatically.
> 
> [1] says the mailer module of eselect was removed.  Is there a better
> way to read them than with eselect?

Put this script in /etc/portage/postsync.d and make it executable

#!/bin/sh

if [ $( eselect news count new ) != "0" ]; then
   eselect news list | mail y...@wherever.you.are
   fi


-- 
Neil Bothwick

To most people solutions mean finding the answers.  But to chemists
solutions are things that are still all mixed up.


pgp_SE676b14D.pgp
Description: OpenPGP digital signature

Re: [gentoo-user] Re: should everything compile?

[Dale]

J. Roeleveld wrote:
> On December 27, 2016 6:55:31 PM GMT+01:00, Dale  wrote:
>> Alan Grimes wrote:
>>> Holger Hoffstätte wrote:
 ..indicates a mismatch in C++11 ABI which changed in gcc5. What
>> happens is that one the
 dependencies of openimageio was built against the old C++11
>> std::string ABI (hence the
 link errors), and needs to be rebuilt. It looks to be "Imf" aka
>> libIlmImf,
 whatever that is. Try to rebuild it with --oneshot and it should
>> work.
 If a similar error pops up for a different dependency, repeat. :)

 -h
>>> Yeah, I emptytree world my system after each Y in X.Y.Z compiler
>> version
>>> bump. Since I sad it, everyone will tell you it's bad advice but
>> really
>>> not. The binary distros will compile everything with the same
>> compiler
>>> so crap doesn't happen. Now it's not super important but then you
>> have
>>> no idea how many other abi link errors are hiding out there.
>>>
>>>
>> I do the same here.  When I switch to a new version of gcc, I do a
>> emerge -e world.  If I've read that it really changes some things, like
>> this one appears to do, I do it twice.  The second time may be overkill
>> but I'd rather have overkill than some weird problem that is difficult
>> to figure out the solution.  I don't think anyone would say doing that
>> is bad.  A ounce of prevention is always better than a pound of cure. 
>> ;-) 
>>
>> There's another upgrade that I do that after too.  I can't recall the
>> name right now but maybe it is glibc or something 
>>
>> Dale
>>
>> :-)  :-)
> I usually do (if encountering weird issues):
> # emerge -1 gcc
> # emerge -1 glibc
> # emerge -e @system
> # emerge -e @world
>
> If there is a better method requiring less time, please let me know.
>
> A full rebuild like this into binary packages using a chroot is a good way to 
> prepare for a toolchain update. That way all the packages are already 
> prepared and the downtime will be minimized.
>
> --
> Joost
>
>

Giving it some thought, I would think your way would be the fastest. 
When you emerge -e system, that should rebuild everything toolchain
wise.  Then when you go back and do world, that should rebuild
everything with the new toolchain.  I have ran emerge -e world twice in
the past but that requires more time than your way.  Your way should
guarantee success and be the quickest.

I might add, in the past when I run into something weird, even with no
gcc or glibc changes, I would do a emerge -e world.  Sometimes there can
be a change that doesn't get picked up on by emerge or even the devs. 
Even the logs from a build failure may not give any real clue.  That
said, it has been a while since I've had that sort of problem.  I run a
mix of unstable and stable and still have a pretty sane system.  I think
that says a lot about how portage/emerge/etc does its job now.  The devs
have really worked out some serious kinks. 

Now that I said that, something will come along pretty soon and just
bork everything up.  LOL 

Dale

:-)  :-) 

Re: [gentoo-user] from Firefox52: NO pure ALSA?, WAS: Firefox 49.0 & Youtube... Audio: No

[Mick]

On Tuesday 27 Dec 2016 20:21:19 lee wrote:
> Mick  writes:
> > On Tuesday 27 Dec 2016 08:21:53 lee wrote:
> >> Rich Freeman  writes:
> >> > On Mon, Dec 26, 2016 at 5:47 PM, lee  wrote:
> >> >> Yes, and that doesn't show me news before I sync, or does it?
> >> > 
> >> > Correct.
> >> > 
> >> > The order to do this in is:
> >> > 
> >> > Sync
> >> > Read news.
> >> > Apply updates.
> >> 
> >> sounds reasonable
> > 
> > Even more reasonable:
> >  eselect news read new
> > 
> > will only come up with the latest as yet unread news, rather than a long
> > list which could have accumulated over the years.
> 
> It seems to be clearing out the list automatically.
> 
> [1] says the mailer module of eselect was removed.  Is there a better
> way to read them than with eselect?
> 
> 
> [1]: https://gitweb.gentoo.org/proj/eselect.git/plain/NEWS

Unless you're running some script to sync portage and emerge, when you 
synchronise portage manually with emerge --sync, it will let you know if there 
are any new news items before you start emerging any packages.  You could get 
portage to email you all elogs, but I am not sure if this will also email you 
any news items - I've never used this feature.  It may require your own script 
to do it.
-- 
Regards,
Mick

signature.asc
Description: This is a digitally signed message part.

Re: [gentoo-user] At last! A Qt5 version of KMail-2 - but here be dragons!

[J. Roeleveld]

On December 27, 2016 5:39:09 PM GMT+01:00, Peter Humphrey 
 wrote:
>On Tuesday, 27 December 2016 15:19:29 GMT J. Roeleveld wrote:
>> On December 27, 2016 3:38:28 PM GMT+01:00, Peter Humphrey 
> wrote:
>> >On Sunday, 18 December 2016 19:07:23 GMT J. Roeleveld wrote:
>> >> More important, how is the latest kmail behaving?
>> >
>> >My first impression is one of horror. It's ghastly!
>> >
>> >I've never seen such profligate waste of screen space. I've attached
>a
>> >couple of screen shots to show you what I mean.
>> 
>> Thanks :)
>> 
>> >Take the folder list, for example. I used to be able to show all
>those
>> >folders in one panel with no scroll-bars, with no difficulty reading
>> >them; now eight folders spill over. I may be able to find a more
>compact
>> >arrangement but this is the best I've managed so far. At least with
>> >kmail:4 I could tweak Qt settings to condense it; now nothing I do
>makes
>> >any improvement.
>> 
>> I would expect font settings. In kmail:4 there is a display part in
>the
>> settings menu. I never did anything directly with Qt to chamge
>anything.
>> I have too many folders anyway for it to fit (unless I would have a
>BIG
>> 4K screen...)
>
>You see how spaced out the folder list is? There's no setting for that
>in 
>KMail as far as I can see. The Qt Configuration tool has an Interface
>tab. 
>Under Global Strut you can set minimum width and height. The default is
>
>zero, giving Qt free rein to do what it likes, but I could set those to
>1 
>and kmail:4 would tuck all those folder entries up so that they fitted 
>without scrolling. It has no effect on kmail:5 though.

Is the qt config for qt4 or qt5?
You might need to find the settings/tool to apply to qt5. Kmail:4 is qt4 based.
(I never played with that tool, which package is it? And which command do I 
run?)

>> >Then the message view. Message.png shows what your message looks
>like
>> >in this version of KMail (the message I'm replying to now). This is
>with
>> >all the bells and whistles I can find switched off.
>> 
>> That actually looks nice to me. Do the minus (-) blocks actually
>allow the
>> quotes to be folded up?
>
>To each his own of course, but for myself, I just want the text with
>some 
>quoting. Yes the minus blocks do fold up, but all of them or none.
>There's 
>no individual control.

That sounds like it's only halfway implemented. Hope they fix that.
The quoting is coloured like that based on a 'simple' algorithm, I would guess.
In kmail4, I think that was configurable?

>I found a new header selection. After switching it back from 5.2 style
>to 
>Fancy, it's back the way I like it.

Is that similar to the 'fancy' style from kmail4? I find that one the more 
natural option myself.

>> >Next, after I'd emerged kde-apps/kmai-16.12.0, it was incomplete. I
>had
>> >to
>> >install several other packages to complete it, including the import
>> >wizard.
>> >Rather than messing about, I just emerged kdepim-meta and had done
>with
>> >it.
>> >
>> >Even after doing that, I get "No backend available for spell
>checking,"
>> >even
>> >though I've set everything up that I can see. Myspell and hunspell
>are
>> >both
>> >installed.
>> 
>> Hmm... that's not nice. Then again, I only use spell check for
>documents.
>> Not emails. Never really trusted those ever since I saw dictionaries
>get
>> automatically ruined by 'do you want to add this word to the
>dictionary'.
>
>Kmail:4 used to default to Australian English, no matter how many times
>I 
>reminded it I'm in the UK. This one has no back-end at all.
>
>--->8

Speaking of English. I always wonder which would be a better match for someone 
who uses English mostly to deal with international friends and computers. I am 
Dutch myself, which means it's officially a second language for me. (Always 
wondered about that, as I speak several languages and fluency is simply a 
matter of which I speak regularly).
I usually throw a dice to decide between US and UK whenever I get asked.

>> >Nothing to do with KMail, but the display of gkrellm has changed
>> >dramatically. I use its Invisible theme, which hasn't actually been
>> >invisible since the switch from KDE-3 to 4, but it had a plain,
>> >unobtrusive
>> >grey scheme and showed what I wanted to see, clearly and with no
>drama.
>> >Now,
>> >the chart backgrounds have changed from charcoal-grey to a dark red,
>> >and
>> >what was grey is now a dreadful salmon-pink. Of course I can't see
>the
>> >red
>> >traces any longer. Perhaps I'm missing a KDE or Qt component.
>> 
>> Always possible. I tend to just install the meta stuff and be done
>with
>> it.
>
>I've found a lot more KDE-5 meta-packages and installed those, but they
>
>haven't helped with this.

What about plasma-meta?
I think that pulls in the whole shebang?

>> >Oh, and when I start a reboot in KDE, akonadi crashes with a
>> >segmentation fault.
>> 
>> That is a bug. Hope it gets ironed out soon. Maybe still busy
>indexing
>> 

Re: [gentoo-user] kernel 4.9.0 + nvidia problem

[J. Roeleveld]

On December 27, 2016 4:23:06 PM GMT+01:00, Philip Webb  
wrote:
>I successfully configured, compiled & installed Kernel 4.9.0 (testing)
>& compiled Nvidia 375.26 (testing) to match ;
>there was a problem trying to use 4.9.0 with 361.28 (below).
>After reboot, X started, but with a primitive display (overlarge chars
>etc).
>I tried it with KDE 5 & with Fluxbox, but no difference :
>KDE couldn't start Plasma, but showed a few apps.
>There was a warning msg re Opengl, but nothing helpful.
>
>Previously & now again, everything is ok with Kernel 4.2.0-r1 + Nvidia
>361.28.
>
>There are many kernel versions available & many Nvidia versions ;
>there are also many USE flags & many kernel options.
>Before I try permutating them all (smile), has anyone else had trouble
>here ?

I actually use the latest 4.8.x with the currently most recent stable nvidia 
drivers and not seen anything weird.

Will test with a few 3D heavy stuff tomorrow to be certain.

--
Joost
-- 
Sent from my Android device with K-9 Mail. Please excuse my brevity.

Re: [gentoo-user] Re: should everything compile?

[J. Roeleveld]

On December 27, 2016 6:55:31 PM GMT+01:00, Dale  wrote:
>Alan Grimes wrote:
>> Holger Hoffstätte wrote:
>>> ..indicates a mismatch in C++11 ABI which changed in gcc5. What
>happens is that one the
>>> dependencies of openimageio was built against the old C++11
>std::string ABI (hence the
>>> link errors), and needs to be rebuilt. It looks to be "Imf" aka
>libIlmImf,
>>> whatever that is. Try to rebuild it with --oneshot and it should
>work.
>>> If a similar error pops up for a different dependency, repeat. :)
>>>
>>> -h
>> Yeah, I emptytree world my system after each Y in X.Y.Z compiler
>version
>> bump. Since I sad it, everyone will tell you it's bad advice but
>really
>> not. The binary distros will compile everything with the same
>compiler
>> so crap doesn't happen. Now it's not super important but then you
>have
>> no idea how many other abi link errors are hiding out there.
>>
>>
>
>I do the same here.  When I switch to a new version of gcc, I do a
>emerge -e world.  If I've read that it really changes some things, like
>this one appears to do, I do it twice.  The second time may be overkill
>but I'd rather have overkill than some weird problem that is difficult
>to figure out the solution.  I don't think anyone would say doing that
>is bad.  A ounce of prevention is always better than a pound of cure. 
>;-) 
>
>There's another upgrade that I do that after too.  I can't recall the
>name right now but maybe it is glibc or something 
>
>Dale
>
>:-)  :-)

I usually do (if encountering weird issues):
# emerge -1 gcc
# emerge -1 glibc
# emerge -e @system
# emerge -e @world

If there is a better method requiring less time, please let me know.

A full rebuild like this into binary packages using a chroot is a good way to 
prepare for a toolchain update. That way all the packages are already prepared 
and the downtime will be minimized.

--
Joost


-- 
Sent from my Android device with K-9 Mail. Please excuse my brevity.

Re: [gentoo-user] from Firefox52: NO pure ALSA?, WAS: Firefox 49.0 & Youtube... Audio: No

[lee]

Alan McKinnon  writes:

> On 27/12/2016 01:02, lee wrote:
>> Alan McKinnon  writes:
>> 
>>> On 26/12/2016 21:42, lee wrote:
 Well, I guess you haven't realised yet that reality doesn't exist.
 Bubbles are a self-imposed limit for those who believe in reality.
 You probably hit that wall and now try hard to remain confined.

 Unfortunately, this won't make sense to you until you come to realise.
>>>
>>>
>>> I *strongly* recommend you cease insulting people's intelligence on this
>>> list.
>> 
>> It was not my intention to do that, and if I did, I apologize.
>> 
>> Different people realise different things.  There are many things I will
>> never realise and others I have.  There are things I won't understand
>> before I have realised what is necessary to realise to understand them.
>> Finding that someone won't understand something before they have
>> realised something doesn't insult anyones intelligence.
>
>
> OK.
>
> I think you need to step back a little and apply the above to this
> situation. By that I mean how you are interacting with others, not the
> various questions about systemd, how many NICs a board has in general
> and so on.
>
> The results you are getting are far from optimum - you may eventually
> get an answer that satisfies you but in general it is involving long
> winding threads that frustrate others.
>
> So I suggest you apply reason and investigation to determine why that
> might be so.
>
> One highly workable method is when you find yourself taking a contrary
> position and about to explain why you think what you think, then reverse
> it. Instead, state that you disagree, that you think something else and
> invite the other to explain why they are saying what they are saying.
> This method has high success in revealing to you what it is you have to
> realise first, as you mention above)
>
>
>> 
>> 
>>> This list (gentoo-user) contains the brightest minds, widest range of
>>> experience (both CS-related and just generally in life), most articulate
>>> and surprisingly, most tolerant, bunch of people I have ever come across
>>> online; and I've been here for 10 years and doing online for 20+ years
>>> and the above is not meant idly.
>>>
>>> I can't watch you type and I can't get in your headspace but based just
>>> on what I read from you, your communications say something that is
>>> frankly, very insulting to those individuals. There is no need to
>>> disparage someone else just because their frame of reference differs
>>> from yours.
>>>
>>> Alan
>>>
>>> p.s. You have a very long way to go still before you begin to match
>>> Dale's contributions here.
>> 
>> So this is supposed to be a competition?
>
> No, it's about people and how people communicate concepts and ideas.
> It's about how Dale is a long term contributor and people generally
> think well of him and how statistically he is right more often than he
> is wrong. He's worth paying attention to.

I see what you mean.  Dale must be pretty annoyed by me.

Sorry, Dale.

Re: [gentoo-user] from Firefox52: NO pure ALSA?, WAS: Firefox 49.0 & Youtube... Audio: No

[lee]

Neil Bothwick  writes:

> On Mon, 26 Dec 2016 21:01:22 +0100, lee wrote:
>
>> > AFAIK, you have three possibilities.
>> >
>> > 1) If you're renaming a NIC via its MAC address, you have to edit the
>> > config file thatlinks the NIC's names and its MAC address.
>> >
>> > 2) If you're using udev's predictable names, the NIC'll have the same
>> > (more or less complex) name if you use the same slot.
>> >
>> > 3) If you're using the kernel names, you have no guarantee that ethX
>> > will be assigned to the same NIC at every bot.  
>> 
>> So there's no good option because names may change unless you make and
>> maintain an assignment.  I wonder why that isn't the default ...
>
> I would imagine because it cannot be used without some initial
> configuration. The default provides the greatest reliability out of the
> box, at the expense of less readable (which is not the same as
> unrecognisable, a value judgement you are imposing on the names) names.

I call them unrecognisable because they are hard to recognise, as in
hard to read and impossible to remember.  I find that annoying.  I can
call them "annoying names" if you prefer that :)

> There is nothing wrong with wanting things to work as you do, but it
> requires input to do so. It you have to start editing files to make it
> work properly, there is little point in making it the default.

Right, and it could work without editing files manually.  A
configuration file assigning editable names to the annoying names could
be created automatically and filled by assigning the name an interface
already has to it (because when it has a name, the name is known, which
is easier than trying to make up all possible names in advance).  Then
only if you wanted you would edit the configuration file to assign the
name(s) of your choosing, and if you don't want to do that, you simply
get the names you get now.  There would be no change to how the names
are now, only an additional option.

That would also have the advantage that when the annoying name of an
interface changes, you can choose to either adjust all configuration
files in which you have specified a particular interface or simply
adjust the one configuration file that assigns the names.

I actually wonder why they didn't virtualise the names.  It makes too
much sense for not to do it, and you could do likewise with other
devices (especially disks).

Re: [gentoo-user] from Firefox52: NO pure ALSA?, WAS: Firefox 49.0 & Youtube... Audio: No

[lee]

Mick  writes:

> On Tuesday 27 Dec 2016 08:21:53 lee wrote:
>> Rich Freeman  writes:
>> > On Mon, Dec 26, 2016 at 5:47 PM, lee  wrote:
>> >> Yes, and that doesn't show me news before I sync, or does it?
>> > 
>> > Correct.
>> > 
>> > The order to do this in is:
>> > 
>> > Sync
>> > Read news.
>> > Apply updates.
>> 
>> sounds reasonable
>
> Even more reasonable:
>
>  eselect news read new
>
> will only come up with the latest as yet unread news, rather than a long list 
> which could have accumulated over the years.

It seems to be clearing out the list automatically.

[1] says the mailer module of eselect was removed.  Is there a better
way to read them than with eselect?


[1]: https://gitweb.gentoo.org/proj/eselect.git/plain/NEWS

Re: [gentoo-user] kernel 4.9.0 + nvidia problem

[lee]

Philip Webb  writes:

> I successfully configured, compiled & installed Kernel 4.9.0 (testing)
> & compiled Nvidia 375.26 (testing) to match ;
> there was a problem trying to use 4.9.0 with 361.28 (below).
> After reboot, X started, but with a primitive display (overlarge chars etc).
> I tried it with KDE 5 & with Fluxbox, but no difference :
> KDE couldn't start Plasma, but showed a few apps.
> There was a warning msg re Opengl, but nothing helpful.
>
> Previously & now again, everything is ok with Kernel 4.2.0-r1 + Nvidia 361.28.
>
> There are many kernel versions available & many Nvidia versions ;
> there are also many USE flags & many kernel options.
> Before I try permutating them all (smile), has anyone else had trouble here ?

4.4.6 and 375.26 seem to work fine here.


 * Found these USE flags for x11-drivers/nvidia-drivers-375.26:
 U I
 + + X   : Install the X.org driver, OpenGL libraries, XvMC libraries, 
and VDPAU libraries
 - - acpi: Add support for Advanced Configuration and Power Interface
 - - compat  : Install non-GLVND libGL for backwards compatibility
 + + driver  : Install the kernel driver module
 + + gtk3: Install nvidia-settings with support for GTK+ 3
 + + kms : Enable support for kernel mode setting (KMS)
 - - pax_kernel  : PaX patches from the PaX project
 - - static-libs : Build static versions of dynamic libraries as well
 + + tools   : Install additional tools such as nvidia-settings
 - - uvm : Install the Unified Memory kernel module (nvidia-uvm) for 
sharing memory between CPU and GPU in CUDA programs
 - - wayland : Enable dev-libs/wayland backend


For kernel options, I followed the wiki.

Re: [gentoo-user] Re: should everything compile?

[Neil Bothwick]

On Tue, 27 Dec 2016 10:25:14 -0500, Alan Grimes wrote:

> Yeah, I emptytree world my system after each Y in X.Y.Z compiler version
> bump. Since I sad it, everyone will tell you it's bad advice but really
> not.

It's not bad advice unless CPU cycles, heat or energy consumption are an
issue. It's generally unnecessary but harmless IMO.


-- 
Neil Bothwick

Who messed with my anti-paranoia shot?


pgpxb1MbG4X9s.pgp
Description: OpenPGP digital signature

Re: [gentoo-user] Re: should everything compile?

[Dale]

Alan Grimes wrote:
> Holger Hoffstätte wrote:
>> ..indicates a mismatch in C++11 ABI which changed in gcc5. What happens is 
>> that one the
>> dependencies of openimageio was built against the old C++11 std::string ABI 
>> (hence the
>> link errors), and needs to be rebuilt. It looks to be "Imf" aka libIlmImf,
>> whatever that is. Try to rebuild it with --oneshot and it should work.
>> If a similar error pops up for a different dependency, repeat. :)
>>
>> -h
> Yeah, I emptytree world my system after each Y in X.Y.Z compiler version
> bump. Since I sad it, everyone will tell you it's bad advice but really
> not. The binary distros will compile everything with the same compiler
> so crap doesn't happen. Now it's not super important but then you have
> no idea how many other abi link errors are hiding out there.
>
>

I do the same here.  When I switch to a new version of gcc, I do a
emerge -e world.  If I've read that it really changes some things, like
this one appears to do, I do it twice.  The second time may be overkill
but I'd rather have overkill than some weird problem that is difficult
to figure out the solution.  I don't think anyone would say doing that
is bad.  A ounce of prevention is always better than a pound of cure.  ;-) 

There's another upgrade that I do that after too.  I can't recall the
name right now but maybe it is glibc or something 

Dale

:-)  :-)

Re: [gentoo-user] kernel 4.9.0 + nvidia problem

[Alan Grimes]

Philip Webb wrote:
> I successfully configured, compiled & installed Kernel 4.9.0 (testing)
> & compiled Nvidia 375.26 (testing) to match ;
> there was a problem trying to use 4.9.0 with 361.28 (below).
> After reboot, X started, but with a primitive display (overlarge chars etc).
> I tried it with KDE 5 & with Fluxbox, but no difference :
> KDE couldn't start Plasma, but showed a few apps.
> There was a warning msg re Opengl, but nothing helpful.
>
> Previously & now again, everything is ok with Kernel 4.2.0-r1 + Nvidia 361.28.

I'm having very good luck with 4.6.7, I'm version frozen on this version
until potential hardware upgrade to Ryzen CPU, pending more information
about it's open systems acceptability.


-- 
Strange Game.
The only winning move is not to play. 

Powers are not rights.

Re: [gentoo-user] At last! A Qt5 version of KMail-2 - but here be dragons!

[Peter Humphrey]

On Tuesday, 27 December 2016 15:19:29 GMT J. Roeleveld wrote:
> On December 27, 2016 3:38:28 PM GMT+01:00, Peter Humphrey 
 wrote:
> >On Sunday, 18 December 2016 19:07:23 GMT J. Roeleveld wrote:
> >> More important, how is the latest kmail behaving?
> >
> >My first impression is one of horror. It's ghastly!
> >
> >I've never seen such profligate waste of screen space. I've attached a
> >couple of screen shots to show you what I mean.
> 
> Thanks :)
> 
> >Take the folder list, for example. I used to be able to show all those
> >folders in one panel with no scroll-bars, with no difficulty reading
> >them; now eight folders spill over. I may be able to find a more compact
> >arrangement but this is the best I've managed so far. At least with
> >kmail:4 I could tweak Qt settings to condense it; now nothing I do makes
> >any improvement.
> 
> I would expect font settings. In kmail:4 there is a display part in the
> settings menu. I never did anything directly with Qt to chamge anything.
> I have too many folders anyway for it to fit (unless I would have a BIG
> 4K screen...)

You see how spaced out the folder list is? There's no setting for that in 
KMail as far as I can see. The Qt Configuration tool has an Interface tab. 
Under Global Strut you can set minimum width and height. The default is 
zero, giving Qt free rein to do what it likes, but I could set those to 1 
and kmail:4 would tuck all those folder entries up so that they fitted 
without scrolling. It has no effect on kmail:5 though.

> >Then the message view. Message.png shows what your message looks like
> >in this version of KMail (the message I'm replying to now). This is with
> >all the bells and whistles I can find switched off.
> 
> That actually looks nice to me. Do the minus (-) blocks actually allow the
> quotes to be folded up?

To each his own of course, but for myself, I just want the text with some 
quoting. Yes the minus blocks do fold up, but all of them or none. There's 
no individual control.

I found a new header selection. After switching it back from 5.2 style to 
Fancy, it's back the way I like it.

> >Next, after I'd emerged kde-apps/kmai-16.12.0, it was incomplete. I had
> >to
> >install several other packages to complete it, including the import
> >wizard.
> >Rather than messing about, I just emerged kdepim-meta and had done with
> >it.
> >
> >Even after doing that, I get "No backend available for spell checking,"
> >even
> >though I've set everything up that I can see. Myspell and hunspell are
> >both
> >installed.
> 
> Hmm... that's not nice. Then again, I only use spell check for documents.
> Not emails. Never really trusted those ever since I saw dictionaries get
> automatically ruined by 'do you want to add this word to the dictionary'.

Kmail:4 used to default to Australian English, no matter how many times I 
reminded it I'm in the UK. This one has no back-end at all.

--->8

> >Nothing to do with KMail, but the display of gkrellm has changed
> >dramatically. I use its Invisible theme, which hasn't actually been
> >invisible since the switch from KDE-3 to 4, but it had a plain,
> >unobtrusive
> >grey scheme and showed what I wanted to see, clearly and with no drama.
> >Now,
> >the chart backgrounds have changed from charcoal-grey to a dark red,
> >and
> >what was grey is now a dreadful salmon-pink. Of course I can't see the
> >red
> >traces any longer. Perhaps I'm missing a KDE or Qt component.
> 
> Always possible. I tend to just install the meta stuff and be done with
> it.

I've found a lot more KDE-5 meta-packages and installed those, but they 
haven't helped with this.

> >Oh, and when I start a reboot in KDE, akonadi crashes with a
> >segmentation fault.
> 
> That is a bug. Hope it gets ironed out soon. Maybe still busy indexing
> your emails?

Could be. I'll try leaving it for an hour or two.

Since I wrote, I've noticed that the messages I imported from kmail archive 
are not all in the right folders - about 2000 have magically jumped from the 
Purchases folder to the inbox. I'm also seeing a reversion to that old 
problem of silly numbers of duplicates appearing in random places at random 
times. I had hoped that would be among the first things to be ironed out in 
the :5 version.

> >I dare say version 16.12.0 of KMail-2 will make a decent platform for
> >development, now that it's finally here, but a very great deal of work
> >lies ahead. I can see that I'll be doing my fair share of shouting too,
> >at it and at the devs.
> 
> I'll try to get my desktop converted this week. And will join in the
> bugreporting party.

It would be good to have someone to compare notes with.

--->8

> >I think I'll have to go down the pub to drown my sorrows.
> 
> Just don't climb behind the wheel of a car afterwards

No, it's just a wee stroll. Takes longer to get my shoes on.   :-)

-- 
Regards
Peter

Re: [gentoo-user] Re: should everything compile?

[Alan Grimes]

Holger Hoffstätte wrote:
> ..indicates a mismatch in C++11 ABI which changed in gcc5. What happens is 
> that one the
> dependencies of openimageio was built against the old C++11 std::string ABI 
> (hence the
> link errors), and needs to be rebuilt. It looks to be "Imf" aka libIlmImf,
> whatever that is. Try to rebuild it with --oneshot and it should work.
> If a similar error pops up for a different dependency, repeat. :)
>
> -h

Yeah, I emptytree world my system after each Y in X.Y.Z compiler version
bump. Since I sad it, everyone will tell you it's bad advice but really
not. The binary distros will compile everything with the same compiler
so crap doesn't happen. Now it's not super important but then you have
no idea how many other abi link errors are hiding out there.


-- 
Strange Game.
The only winning move is not to play. 

Powers are not rights.

[gentoo-user] kernel 4.9.0 + nvidia problem

[Philip Webb]

I successfully configured, compiled & installed Kernel 4.9.0 (testing)
& compiled Nvidia 375.26 (testing) to match ;
there was a problem trying to use 4.9.0 with 361.28 (below).
After reboot, X started, but with a primitive display (overlarge chars etc).
I tried it with KDE 5 & with Fluxbox, but no difference :
KDE couldn't start Plasma, but showed a few apps.
There was a warning msg re Opengl, but nothing helpful.

Previously & now again, everything is ok with Kernel 4.2.0-r1 + Nvidia 361.28.

There are many kernel versions available & many Nvidia versions ;
there are also many USE flags & many kernel options.
Before I try permutating them all (smile), has anyone else had trouble here ?

-- 
,,
SUPPORT ___//___,   Philip Webb
ELECTRIC   /] [] [] [] [] []|   Cities Centre, University of Toronto
TRANSIT`-O--O---'   purslowatchassdotutorontodotca

Re: [gentoo-user] At last! A Qt5 version of KMail-2 - but here be dragons!

[J. Roeleveld]

On December 27, 2016 3:38:28 PM GMT+01:00, Peter Humphrey 
 wrote:
>On Sunday, 18 December 2016 19:07:23 GMT J. Roeleveld wrote:
>
>> More important, how is the latest kmail behaving?
>
>My first impression is one of horror. It's ghastly!
>
>I've never seen such profligate waste of screen space. I've attached a 
>couple of screen shots to show you what I mean.

Thanks :)

>Take the folder list, for example. I used to be able to show all those 
>folders in one panel with no scroll-bars, with no difficulty reading
>them; 
>now eight folders spill over. I may be able to find a more compact 
>arrangement but this is the best I've managed so far. At least with
>kmail:4 
>I could tweak Qt settings to condense it; now nothing I do makes any 
>improvement.

I would expect font settings. In kmail:4 there is a display part in the 
settings menu. I never did anything directly with Qt to chamge anything.
I have too many folders anyway for it to fit (unless I would have a BIG 4K 
screen...)

>Then the message view. Message.png shows what your message looks like
>in 
>this version of KMail (the message I'm replying to now). This is with
>all 
>the bells and whistles I can find switched off.

That actually looks nice to me. Do the minus (-) blocks actually allow the 
quotes to be folded up?

>Next, after I'd emerged kde-apps/kmai-16.12.0, it was incomplete. I had
>to 
>install several other packages to complete it, including the import
>wizard. 
>Rather than messing about, I just emerged kdepim-meta and had done with
>it.
>
>Even after doing that, I get "No backend available for spell checking,"
>even 
>though I've set everything up that I can see. Myspell and hunspell are
>both 
>installed.

Hmm... that's not nice. Then again, I only use spell check for documents. Not 
emails. Never really trusted those ever since I saw dictionaries get 
automatically ruined by 'do you want to add this word to the dictionary'.

>In the message list I have next-to-no control over the font. I can set
>the 
>basic one, but not those for unread, important or action items. They're
>now 
>displayed in a reduced-density form of the basic font (while pretending
>
>they're going to use the same font as the message itself). The
>designers 
>evidently know what I want better than I do (anyone might think this
>was 
>Gnome).

I hope that can be fixed somehow.

>Nothing to do with KMail, but the display of gkrellm has changed 
>dramatically. I use its Invisible theme, which hasn't actually been 
>invisible since the switch from KDE-3 to 4, but it had a plain,
>unobtrusive 
>grey scheme and showed what I wanted to see, clearly and with no drama.
>Now, 
>the chart backgrounds have changed from charcoal-grey to a dark red,
>and 
>what was grey is now a dreadful salmon-pink. Of course I can't see the
>red 
>traces any longer. Perhaps I'm missing a KDE or Qt component.

Always possible. I tend to just install the meta stuff and be done with it.

>Oh, and when I start a reboot in KDE, akonadi crashes with a
>segmentation 
>fault.

That is a bug. Hope it gets ironed out soon. Maybe still busy indexing your 
emails?

>I dare say version 16.12.0 of KMail-2 will make a decent platform for 
>development, now that it's finally here, but a very great deal of work
>lies 
>ahead. I can see that I'll be doing my fair share of shouting too, at
>it and 
>at the devs.

I'll try to get my desktop converted this week. And will join in the 
bugreporting party.

>It's taken me about 30 hours to get this far. I ditched the old system 
>altogether and built a new one on the kde-plasma profile. I didn't ask
>for 
>anything in a slot 4, just slot 5 versions. I also ditched my old user
>and 
>set up a new one from scratch. Headache? What headache?

Great

>I think I'll have to go down the pub to drown my sorrows.

Just don't climb behind the wheel of a car afterwards

--
Joost


-- 
Sent from my Android device with K-9 Mail. Please excuse my brevity.

Re: [gentoo-user] Firefox 49.0 & Youtube....Video: Yes - Audio: No...

[Alarig Le Lay]

On Sat Oct 15 20:27:43 2016, meino.cra...@gmx.de wrote:
> Hi,
> 
> this evening I updated GENTOO and a new firefox was installed.
> This one seem completly to disable flash video finally...
> since I got no video/audio at all.
> 
> I disabled all flash-related addons of my firefox and
> restarted  it.
> 
> Now I got a video ... but without any audio.
> (I am running jackd by the way).
> I check with qjackctl whether there were any
> ports which I missed to connect...nothing.
> 
> Hmmm...
> 
> Is there any fix for that?

Hi, I had the same problem today (well, the problem was really old, but
it was the first time I tried to play audio in firefox since 45).

If I play the file from mpv, it works, from firefox, no sound.
I added the ffmpeg useflag and it works again now. I still use pure ALSA
as sound stack.

$ equery uses www-client/firefox | grep +
+dbus
+ffmpeg
+gmp-autoupdate
+gstreamer
+hwaccel
+jemalloc3
+jit
+l10n_fr
+startup-notification

-- 
alarig


signature.asc
Description: PGP signature

Re: [gentoo-user] from Firefox52: NO pure ALSA?, WAS: Firefox 49.0 & Youtube... Audio: No

[Neil Bothwick]

On Mon, 26 Dec 2016 21:01:22 +0100, lee wrote:

> > AFAIK, you have three possibilities.
> >
> > 1) If you're renaming a NIC via its MAC address, you have to edit the
> > config file thatlinks the NIC's names and its MAC address.
> >
> > 2) If you're using udev's predictable names, the NIC'll have the same
> > (more or less complex) name if you use the same slot.
> >
> > 3) If you're using the kernel names, you have no guarantee that ethX
> > will be assigned to the same NIC at every bot.  
> 
> So there's no good option because names may change unless you make and
> maintain an assignment.  I wonder why that isn't the default ...

I would imagine because it cannot be used without some initial
configuration. The default provides the greatest reliability out of the
box, at the expense of less readable (which is not the same as
unrecognisable, a value judgement you are imposing on the names) names.

There is nothing wrong with wanting things to work as you do, but it
requires input to do so. It you have to start editing files to make it
work properly, there is little point in making it the default.


-- 
Neil Bothwick

GOTO: (n.) an efficient and general way of controlling a program, much
despised by academics and others whose brains have been ruined by
overexposure to Pascal.


pgp5fSFWYrnWV.pgp
Description: OpenPGP digital signature

Re: [gentoo-user] from Firefox52: NO pure ALSA?, WAS: Firefox 49.0 & Youtube... Audio: No

[Mick]

On Tuesday 27 Dec 2016 08:21:53 lee wrote:
> Rich Freeman  writes:
> > On Mon, Dec 26, 2016 at 5:47 PM, lee  wrote:
> >> Yes, and that doesn't show me news before I sync, or does it?
> > 
> > Correct.
> > 
> > The order to do this in is:
> > 
> > Sync
> > Read news.
> > Apply updates.
> 
> sounds reasonable

Even more reasonable:

 eselect news read new

will only come up with the latest as yet unread news, rather than a long list 
which could have accumulated over the years.

-- 
Regards,
Mick

signature.asc
Description: This is a digitally signed message part.

Re: [gentoo-user] Installing Gentoo on a VPS with little RAM

[Mick]

On Tuesday 27 Dec 2016 07:13:36 J. Roeleveld wrote:
> On Monday, December 26, 2016 09:45:47 PM Francesco Turco wrote:
> > Hello.
> > 
> > I have a Vultr VPS instance with Arch Linux but I'd like to replace it
> > with Gentoo Linux. The last time I tried that I couldn't build some
> > packages because the kernel killed gcc after a while. Please notice this
> > VPS instance has only 768 MiB of RAM. What can I try besides removing
> > -pipe from C(XX)FLAGS and setting MAKEOPTS to -j1? Should I add a swap
> > partition? Currently there's only a single root btrfs filesystem with @,
> > @boot and @home subvolumes. Btrfs doesn't support a swap file as far as
> > I know.
> > 
> > My VPS is currently used for the following things:
> > - Static personal website
> > - Shaarli (PHP application with no database)
> > - Tiny Tiny RSS (PHP application with database)
> > - ZNC server
> > 
> > Thanks.
> 
> You could try building all packages on your own system and then installing
> using "-k".
> That should avoid any out-of-memory issues.
> 
> --
> Joost

Many packages require a lot of memory these days to compile, gcc being one of 
them.  With only 768 MiB of RAM you will *definitely* need a swap 
file/partition.   In addition, with so little memory you won't be able to 
compile packages in tmpfs.  Thrashing of your disks will go off scale as the 
machine will try to swap memory to disk and back again.

For these reasons what Joost suggests is the best way.  You could try to share 
/usr/portage via NFS, but since this is a remote server I am not sure the 
security model will be adequate for most use cases.  It would be easier to 
copy your VPS filesystem over to your local machine which is not resouce 
constrained, chroot into it and then sync portage and emerge everything you 
need with --buildpckg.  Then copy over the binary packages to the VPS and 
emerge them there with --usepkg or --usepkgonly.  If you are disk space 
constrained on the server, you don't have to copy over the binary packages 
manually each time.  Instead you can set up your home PC as a binary host and 
serve the binary packages over ssh or over http/s to the VPS client.

Essentially, you will be creating your own bespoke binary distribution on your 
home PC, then downloading it and installing it at your VPS.  Have a look here 
for some pointers:

https://wiki.gentoo.org/wiki/Binary_package_guide

-- 
Regards,
Mick

signature.asc
Description: This is a digitally signed message part.

Re: [gentoo-user] from Firefox52: NO pure ALSA?, WAS: Firefox 49.0 & Youtube... Audio: No

[Alan McKinnon]

On 27/12/2016 01:02, lee wrote:
> Alan McKinnon  writes:
> 
>> On 26/12/2016 21:42, lee wrote:
>>> Well, I guess you haven't realised yet that reality doesn't exist.
>>> Bubbles are a self-imposed limit for those who believe in reality.
>>> You probably hit that wall and now try hard to remain confined.
>>>
>>> Unfortunately, this won't make sense to you until you come to realise.
>>
>>
>> I *strongly* recommend you cease insulting people's intelligence on this
>> list.
> 
> It was not my intention to do that, and if I did, I apologize.
> 
> Different people realise different things.  There are many things I will
> never realise and others I have.  There are things I won't understand
> before I have realised what is necessary to realise to understand them.
> Finding that someone won't understand something before they have
> realised something doesn't insult anyones intelligence.


OK.

I think you need to step back a little and apply the above to this
situation. By that I mean how you are interacting with others, not the
various questions about systemd, how many NICs a board has in general
and so on.

The results you are getting are far from optimum - you may eventually
get an answer that satisfies you but in general it is involving long
winding threads that frustrate others.

So I suggest you apply reason and investigation to determine why that
might be so.

One highly workable method is when you find yourself taking a contrary
position and about to explain why you think what you think, then reverse
it. Instead, state that you disagree, that you think something else and
invite the other to explain why they are saying what they are saying.
This method has high success in revealing to you what it is you have to
realise first, as you mention above)


> 
> 
>> This list (gentoo-user) contains the brightest minds, widest range of
>> experience (both CS-related and just generally in life), most articulate
>> and surprisingly, most tolerant, bunch of people I have ever come across
>> online; and I've been here for 10 years and doing online for 20+ years
>> and the above is not meant idly.
>>
>> I can't watch you type and I can't get in your headspace but based just
>> on what I read from you, your communications say something that is
>> frankly, very insulting to those individuals. There is no need to
>> disparage someone else just because their frame of reference differs
>> from yours.
>>
>> Alan
>>
>> p.s. You have a very long way to go still before you begin to match
>> Dale's contributions here.
> 
> So this is supposed to be a competition?

No, it's about people and how people communicate concepts and ideas.
It's about how Dale is a long term contributor and people generally
think well of him and how statistically he is right more often than he
is wrong. He's worth paying attention to.

-- 
Alan McKinnon
alan.mckin...@gmail.com