LinuxQuestions.org - Requests for -current (14.2-->15.0)

Page 10 of 197

Show 50 post(s) from this thread on one page

- Slackware (https://www.linuxquestions.org/questions/slackware-14/)

- - Requests for -current (14.2-->15.0) (https://www.linuxquestions.org/questions/slackware-14/requests-for-current-14-2-15-0-a-4175620463/)

saxa	02-18-2018 02:07 PM

Quote:

Originally Posted by Didier Spaier (Post 5821306)

If you don't want to make your own directory, use /floppy ;)

But at the end and being realist, probably would be already time to make an good name for a free available dir in the installer boot disk to allow
mounting of the tarballs for install.

I also use /floppy but naming it with a more moden name would be really appreciated.

Didier Spaier

02-18-2018 02:17 PM

Quote:

Originally Posted by saxa (Post 5821324)

I also use /floppy but naming it with a more modern name would be really appreciated.

Who cares how directories are named in RAM? They won't survive a reboot anyway. I am happy with /floppy which reminds me the 8-inch disks I used nearly forty years ago. These were really floppy.

Darth Vader

02-18-2018 03:02 PM

Quote:

Originally Posted by Didier Spaier (Post 5821326)

Who cares how directories are named in RAM? They won't survive a reboot anyway. I am happy with /floppy which reminds me the 8-inch disks I used nearly forty years ago. These were really floppy.

Well, I believe that /floppy thing is still useful, considering that there are Iomega JAZ drives, which have 2GB. ;)

Plenty of space for a full install "floppy" disk if you avoid the KDE...

And if you go for the Iomega REV 120GB variant, go figure! :D

saxa	02-18-2018 05:15 PM

Surely the name is really not important for the machine, but for a human who have to think about what to mount, where to mount surely it is. I doubt you would mount your drive under /bin
as you know what bin means, but /floppy surely can be more appropriated than /bin. In that sense I am talking about making it a more user friendly, to understand which point in the fs tree should be used or not. As if you think the installer sets up its newly created dirs tree under /mnt/ which is IMHO the most logical point where something should be mounted.

Didier Spaier

02-18-2018 05:39 PM

As pointed out by Alien Bob you should not mount the source of packages and the target on the same directory for the installation to succeed, that's all. Name the directory you use for the source of packages as you see fit, it's just a matter of typing "mkdir <mypreferednameforthesourcedirectory>", but please don't request to fix something that's not broken.

Thom1b

02-19-2018 04:40 AM

Hi,

It could be nice to have sftp support with proftpd.
Adding "mod_sftp" in PROFTPD_MODS (in the SlackBuild) is the only thing to do.

saxa	02-19-2018 04:52 AM

Dear Didier, my post was not a request, it was just an side note or an heads up that would be much easier for the user to have an intuitive name for something that is not intuitive. I have did that many times and it worked always, but I have to admit that
for somebody who never installed slackware this is not very intuitive name for a mount point. Just simply that. It was a mere comment.
And the reality is that /floppy as a directory for mounting something other than a floppy disk is not very intuitive name.
Of course anybody who have experience with the install process will not care about this at all.

As for the Alien post I know that it is written and very clearly in the installer to not use /mnt.

gmgf	02-19-2018 07:35 AM

Quote:

Originally Posted by willysr (Post 5821162)

This is a new request FYI since this is GTK+3 version of glade. The one shipped on Slackware was the GTK+2 version.
Good news is that it's co-installable just like GTK+X
See the information on the website

Today, I doubt that many people, develop for or with gtk2 ;)

Didier Spaier

02-19-2018 07:55 AM

Quote:

Originally Posted by gmgf (Post 5821548)

Today, I doubt that many people, develop for or with gtk2 ;)

LXDE is still alive and supported even if less heavily developed than LXQT. By the way the devs chose to switch to Qt instead of gtk3. Which has the inconvenience that accessibility level of Qt apps is generally still way behind those of GTK apps, but I digress.

Pixxt

02-19-2018 04:47 PM

mesa 17.3.5

https://lists.freedesktop.org/archiv...ry/000401.html

https://mesa.freedesktop.org/archive/mesa-17.3.5.tar.xz

bifferos

02-19-2018 05:54 PM

Openvpn is included in Slackware but lacks a bit of polish.

Can we have:

1) rc.openvpn example copied from here: https://docs.slackware.com/howtos:ne...rvices:openvpn and added to the openvpn package.

2) Corresponding entry in rc.inet2 for above.

3) Example server file copied to e.g. /etc/openvpn/openvpn.conf-server.sample (have to get it from the source)

4) Example client file copied to e.g. /etc/openvpn/openvpn.conf-client.sample (have to get it from the source)

At the moment we have just /etc/openvpn/openvpn.conf.sample (which is presumably a server example?), and the initially promising looking directory /usr/doc/openvpn-2.4.4/sample-config-files/ but which according to MANIFEST.bz2 for -current lacks any content.

Adding tunctl to help with adding tap devices to bridges would be nice, but not essential.

bassmadrigal

02-19-2018 10:40 PM

In conjunction with openvpn requests, could we consider adding openresolv? This allows you to have your resolv.conf updated with different nameservers after openvpn connects. It is developed by the same guy who develops dhcpcd.

I have a working SlackBuild for it that you're free to use.

franzen

02-20-2018 06:59 AM

Hi,

in /etc/rc.d/rc.M, smartd is disabled, request to enable it:

Code:

# Start smartd, which monitors the status of S.M.A.R.T. compatible

# hard drives and reports any problems.  Note some devices (which aren't

# smart, I guess ;) will hang if probed by smartd, so it's commented out

# by default.

-#if [ -x /usr/sbin/smartd ]; then

-#  /usr/sbin/smartd

-#fi

+if [ -x /usr/sbin/smartd ]; then

+  /usr/sbin/smartd

+fi

According to https://www.smartmontools.org/wiki/Warnings the hangs may not be a problem any more. But still, to be on the the safe side, DEVICESCAN may be commented in /etc/smartd.conf. If commented, and therefor smartd.conf ist empty, smartd will not start with returncode 17.

Quote:

# The word DEVICESCAN will cause any remaining lines in this
# configuration file to be ignored: it tells smartd to scan for all
# ATA and SCSI devices. DEVICESCAN may be followed by any of the
# Directives listed below, which will be applied to all devices that
# are found. Most users should comment out DEVICESCAN and explicitly
# list the devices that they wish to monitor.
-DEVICESCAN
+#DEVICESCAN

TracyTiger

02-20-2018 08:22 AM

Quote:

Originally Posted by Thom1b (Post 5821502)

It could be nice to have sftp support with proftpd.
Adding "mod_sftp" in PROFTPD_MODS (in the SlackBuild) is the only thing to do.

IF you are reviewing the modules to include in proftpd, it would be convenient to have mod_facl.c included by default. Some clients require more complicated permissions configurations that can only be handled with ACLs.

This is not a strong request as it is easy for me to recomplile with the modules I need.

( I'm not running -current so it may already be included now. :-) )

ttk	02-20-2018 11:21 AM

A host running the ssh service accepts sftp connections already. It never occurred to me to use proftpd for sftp, but I guess the advantage is that proftp offers more sophisticated access control options?

kjhambrick

02-20-2018 12:43 PM

Heads up ...

Not sure if it is intentional or if it is an artifact or ... ???

When I invoked Alien Bob's slackware mirror rsync script today ( Tue Feb 20 ), I saw a '?' file scroll by.

Code:

# pwd

/home/dld/slackware/slackware64-current



# ls -lad '?'

-rw-r--r-- 1 4015 4015 0 Feb 19 16:55 ?

Repeated the download from rsync://ftp.slackware.com/slackware/ and the '?' file existed there too ...

Edit: I see it here: ftp://ftp.slackware.com/pub/slackwar...are64-current/

HTH.

-- kjh

volkerdi

02-20-2018 01:04 PM

Quote:

Originally Posted by kjhambrick (Post 5822101)

Heads up ...

Not sure if it is intentional or if it is an artifact or ... ???

When I invoked Alien Bob's slackware mirror rsync script today ( Tue Feb 20 ), I saw a '?' file scroll by.

Code:

# pwd

/home/dld/slackware/slackware64-current



# ls -lad '?'

-rw-r--r-- 1 4015 4015 0 Feb 19 16:55 ?

That was supposed to be saved for April 1st. Sorry.

kjhambrick

02-20-2018 01:22 PM

Quote:

Originally Posted by volkerdi (Post 5822110)

That was supposed to be saved for April 1st. Sorry.

NP at all !

Isn't is always April 1st somewhere ?

-- kjh( :) getting old ... maybe it was: it's always 5 O'Clock somewhere :) )

bifferos

02-20-2018 03:45 PM

Quote:

Originally Posted by ttk (Post 5822059)

I wondered the same thing. In combination with sshfs, ssh provides pretty much everything I need for personal use. Then I realised that proftpd may scale better for a heavily loaded public facing site. <shrug>

bifferos

02-20-2018 04:10 PM

Quote:

Originally Posted by saxa (Post 5821376)

but /floppy surely can be more appropriated than /bin. In that sense I am talking about making it a more user friendly, to understand which point in the fs tree should be used or not.

Did I miss your suggestion for an alternate name? :-). That's a serious question, because I've no idea what I'd call it.

I always create a new directory so I can call it what I want. I'm frequently creating directories like /sda1 /sda2 and so on because I want the names to match what they actually are. I don't think anyone else would want crap like that appearing all over their otherwise clean RAMFS, and I'd create the directories as I mounted them, so I knew what was actually there, to keep track.

NB: I was also initially perplexed with the /mnt directory and the message in it, until I had time to think about it. Then, as with most things Slackware related I realised it's pretty sensible. Its somewhere to mount stuff. Just not somewhere for *me* to mount stuff :-).

franzen

02-21-2018 03:01 AM

Quote:

Originally Posted by ttk (Post 5822059)

Looking at spectre/meltdown, it is an disadvantage to have lots of systemaccounts on a server, but that's unavoidable with sftp per openssh.
With proftpd there's no additional systemaccount needed, aka virtual sftp-users can be configured.

TurboBlaze

02-21-2018 10:52 AM

glibc-i18n

Why I need to generate ru_RU.UTF-8 locale /usr/lib64/locale/locale-archive for the glibc-i18n-2.27-x86_64-1.txz via

Code:

localedef -i ru_RU -f UTF-8 ru_RU.UTF-8

? Otherwise I get

Code:

locale not supported by C library" error

and broken localization with unstable OS.

BTW, with glibc-i18n-2.26 all was ok.

Thanks.

Didier Spaier

02-21-2018 02:31 PM

Same issue for fr_FR.utf8, it seems only the C locale be active. Caveat: I installed in on a Slint system based on Slackware64-14.2 as I don't have Slackware-current installed. Could this be linked to changes noted in the NEWS in the source archive? For instance:

Code:

* Support for two grammatical forms of month names has been added.

  In a call to strftime, the "%B" and "%b" format specifiers will now

  produce the grammatical form required when the month is used as part

  of a complete date.  New "%OB" and "%Ob" specifiers produce the form

  required when the month is named by itself.  For instance, in Greek

  and in many Slavic and Baltic languages, "%B" will produce the month

  in genitive case, and "%OB" will produce the month in nominative case.



  In a call to strptime, "%B", "%b", "%h", "%OB", "%Ob", and "%Oh"

  are all valid and will all accept any known form of month

  name---standalone or complete, abbreviated or full.  In a call to

  nl_langinfo, the query constants MON_1..12 and ABMON_1..12 return

  the strings used by "%B" and "%b", respectively.  New query

  constants ALTMON_1..12 and _NL_ABALTMON_1..12 return the strings

  used by "%OB" and "%Ob", respectively.



  In a locale definition file, use "alt_mon" and "ab_alt_mon" to

  define the strings for %OB and %Ob, respectively; these have the

  same syntax as "mon" and "abmon".  These arrays are optional; if they

  are not provided then they have the same content as "mon" and "abmon",

  respectively.



  These features are provided for locales which define "alt_mon" and/or

  "ab_alt_mon" in their locale source data.  This release includes such

  alternative month name data for the following languages: Belarusian,

  Croatian, Greek, Lithuanian, Polish, Russian, and Ukrainian.



  This feature is currently a GNU extension, but it is expected to

  be added to the next revision of POSIX, and it is also already

  available on some BSD-derived operating systems.



  This feature will cause existing statically compiled applications

  to fail to load locales and fall back to the builtin C/POSIX locales.

  See notes below for other changes affecting compatibility.



<snip>



* Statically compiled applications attempting to load locales compiled for the

  GNU C Library version 2.27 will fail and fall back to the builtin C/POSIX

  locale.  The reason for this is that the addition of the new "%OB" and "%Ob",

  support for two grammatical forms of the month names, also extends the locale

  data binary format.  Static applications needing locale support must be

  recompiled to match the runtime and data they are deployed with. In some

  distributions there is an upgrade window where dynamically linked applications

  may use a new library but the old locale data and also fall back to the

  builtin C/POSIX locales; restarting the application process is sufficient to

  fix this.

Anyway I see other probable effects like pulseaudio not working properly or a python app like orca failing or this message when running indirectly upgradepkg (localized version):

Code:

/sbin/installpkg: line 64: warning: setlocale: LC_TIME: cannot change locale (fr_FR.utf8)

Again I am not running -current so if all that doesn't happen on -current I apologize for the noise.

All this with LANG set to fr_FR.utf8

qunying

02-21-2018 05:22 PM

Latest update less-530-x86_64-1.txz (Feb-20) install /usr/bin/lesspipe.sh without execution permission, causing less on a tar.gz file shown the raw content in stead of uncompressed directory listing of the tar.gz file.

volkerdi

02-21-2018 07:15 PM

Quote:

Originally Posted by TurboBlaze (Post 5822452)

Why I need to generate ru_RU.UTF-8 locale /usr/lib64/locale/locale-archive for the glibc-i18n-2.27-x86_64-1.txz via

Code:

localedef -i ru_RU -f UTF-8 ru_RU.UTF-8

? Otherwise I get

Code:

locale not supported by C library" error

and broken localization with unstable OS.

BTW, with glibc-i18n-2.26 all was ok.

Thanks.

What's a command that will cause that error message?

USUARIONUEVO

02-21-2018 07:36 PM

Hi , xorg protocol packages are now all-in-one package

xorgproto-2018.3
https://www.x.org/releases/individua...2018.3.tar.bz2

gmgf	02-22-2018 12:18 AM

meson-0.44.1:

https://github.com/mesonbuild/meson/...-0.44.1.tar.gz

Thom1b

02-22-2018 02:43 AM

postfix-3.3.0 is released:

https://de.postfix.org/ftpmirror/off...x-3.3.0.tar.gz

TurboBlaze

02-22-2018 08:06 AM

Quote:

Originally Posted by volkerdi (Post 5822630)

What's a command that will cause that error message?

Code:

perl -v

this is output of the 2>

Code:

perl: warning: Setting locale failed.

perl: warning: Please check that your locale settings:

        LANGUAGE = (unset),

        LC_ALL = (unset),

        LC_COLLATE = "C",

        LANG = "ru_RU.utf8"

    are supported and installed on your system.

perl: warning: Falling back to the standard locale ("C").

Code:

locale

this is output of the 2>

Code:

locale: Cannot set LC_ALL to default locale: No such file or directory exists

Code:

localedef --no-archive -c -i ru_RU -f UTF-8 ru

Code:

[error] LC_IDENTIFICATION: unknown standard `ru_RU:2000' for category `LC_CTYPE'

[error] LC_IDENTIFICATION: unknown standard `ru_RU:2000' for category `LC_NUMERIC'

[error] LC_IDENTIFICATION: unknown standard `ru_RU:2000' for category `LC_TIME'

[error] LC_IDENTIFICATION: unknown standard `ru_RU:2000' for category `LC_COLLATE'

[error] LC_IDENTIFICATION: unknown standard `ru_RU:2000' for category `LC_MONETARY'

[error] LC_IDENTIFICATION: unknown standard `ru_RU:2000' for category `LC_MESSAGES'

[error] LC_IDENTIFICATION: unknown standard `ru_RU:2000' for category `LC_PAPER'

[error] LC_IDENTIFICATION: unknown standard `ru_RU:2000' for category `LC_NAME'

[error] LC_IDENTIFICATION: unknown standard `ru_RU:2000' for category `LC_ADDRESS'

[error] LC_IDENTIFICATION: unknown standard `ru_RU:2000' for category `LC_TELEPHONE'

[error] LC_IDENTIFICATION: unknown standard `ru_RU:2000' for category `LC_MEASUREMENT'

[error] LC_IDENTIFICATION: unknown standard `ru_RU:2000' for category `LC_IDENTIFICATION'

Working gnome-terminal stopped working with the "locale not supported by C library" error
I can not to use gnome-terminal (only xterm works for me) http://s7.uploads.ru/x5GEk.png
I can't to give Russian name for the file or directory

Didier Spaier

02-22-2018 08:19 AM

I am downloading Eric's Slackware64-current ISO dated Thu Feb 22 04:34:23 UTC 2018. Will see if get the same weirdness as you and as on 14.2 and try to investigate further.

@TurboBlaze: are your findings on Porteus or on genuine Slackware?

TurboBlaze

02-22-2018 08:40 AM

Didier Spaier, if it matters, I tested this in Porteus based on Slackware-current. In advance I apologize if I was mistaken with a topic.

BTW, system-config-printer must be run as root user or it gives freezes.

Code:

sed -i 's|system-config-printer|gksu system-config-printer|g' /usr/share/applications/system-config-printer.desktop

Cheers

gmgf	02-22-2018 08:49 AM

Package, 'system-config-printer' , on a fresh current install, the root auth authentication dialog box is broken, it need this patch (work here):

https://git.archlinux.org/svntogit/p...config-printer

already posted, but probably omitted ;)

patch is also available on system-config-printer git now:

https://github.com/zdohnal/system-co...commits/master

ponce

02-22-2018 09:31 AM

Quote:

Originally Posted by TurboBlaze (Post 5822833)

Didier Spaier, if it matters, I tested this in Porteus based on Slackware-current. In advance I apologize if I was mistaken with a topic.

TurboBlaze: porteus <> slackware-current, don't report porteus issues here...

Didier Spaier

02-22-2018 10:04 AM

1 Attachment(s)

To make sure I installed Slakware64-current in a VM and ran this script:

Code:

#/bin/sh/

for LANG in $(locale -a); do

    strip=${LANG%.*}

    mylocale=${strip%@*}

    mycharmap=$(locale -k charmap|sed 's/[^"]*"//;s/"//')

    echo "localedef -f $mycharmap -i $mylocale $LANG"

    localedef -f $mycharmap -i $mylocale $LANG

done

Only 4 locales are missing in /usr/lib64/locale/locale-archive (according to localedef --list-archive) out of 815 reported by locale -a, namely:
et_EE
et_EE.iso885915
et_EE.utf8
hy_AM.armscii8
due to formatting errors in the input files from /usr/share/i18n/locales, no other issues but a few warnings (full log attached), and no other weirdness detected.

So, false alarm as far I can tell.

Maybe the files shipped in the glibc-i18n package can't be used with another glibc version, in other words are not so independent that I thought?

codeguy

02-22-2018 10:30 AM

Can I add another request for adding CONFIG_CIFS_SMB2 to the kernel? I see it mentioned but not sure if it made it in yet.

Pretty please?

codeguy

02-22-2018 05:47 PM

Oh! Its in the changelog!
k/kernel-source-4.9.28-noarch-1.txz: Upgraded.
CIFS_SMB2 n -> y

Looks like its set, sorry for the noise.

gmgf	02-23-2018 01:09 AM

gnupg-2.2.5:

https://lists.gnupg.org/pipermail/gn...q1/000420.html
https://gnupg.org/ftp/gcrypt/gnupg/gnupg-2.2.5.tar.bz2

kjhambrick

02-23-2018 08:02 AM

new smartd ...

Very nice Pat and Team !

The new /etc/rc.d/rc.{M,smartd} scripts work as expected on my Slackware64 -current Laptop.

Thank you !

-- kjh

Code:

# mv /etc/rc.M.new /etc/rc.M

# mv /etc/rc.d/rc.smartd.new /etc/rc.d/rc.smartd    # oops ... left this out in OP

# chmod 755 /etc/rc.d/rc.smartd

# /etc/rc.d/rc.smartd restart

# grep smartd /var/log/massages



<<snip>>--------------------------------------------------------------------

Feb 23 07:27:42 samsung smartd[32163]: smartd received signal 15: Terminated 

Feb 23 07:27:42 samsung smartd[32163]: smartd is exiting (exit status 0) 

Feb 23 07:27:43 samsung smartd[32316]: smartd 6.6 2017-11-05 r4594 [x86_64-linux-4.14.20] (local build) 

Feb 23 07:27:43 samsung smartd[32316]: Copyright (C) 2002-17, Bruce Allen, Christian Franke, www.smartmontools.org 

Feb 23 07:27:43 samsung smartd[32316]: Opened configuration file /etc/smartd.conf 

Feb 23 07:27:43 samsung smartd[32316]: Configuration file /etc/smartd.conf parsed. 

Feb 23 07:27:43 samsung smartd[32316]: Device: /dev/sda [SAT], opened 

Feb 23 07:27:43 samsung smartd[32316]: Device: /dev/sda [SAT], HGST HTE721010A9E630, S/N:JR10034M2KTNZK, WWN:5-000cca-8a8e445f3, FW:JB0OA3M0, 1.00 TB 

Feb 23 07:27:43 samsung smartd[32316]: Device: /dev/sda [SAT], found in smartd database: HGST Travelstar 7K1000 

Feb 23 07:27:44 samsung smartd[32316]: Device: /dev/sda [SAT], is SMART capable. Adding to "monitor" list. 

Feb 23 07:27:44 samsung smartd[32316]: Monitoring 1 ATA/SATA, 0 SCSI/SAS and 0 NVMe devices 

Feb 23 07:27:45 samsung smartd[32318]: smartd has fork()ed into background mode. New PID=32318. 

Feb 23 07:27:45 samsung smartd[32318]: file /run/smartd.pid written containing PID 32318 



# cat /run/smartd.pid

32318



# ps -ef |grep [s]martd

root    32318    1  0 07:27 ?        00:00:00 /usr/sbin/smartd -p /run/smartd.pid

dugan

02-23-2018 11:04 AM

Quote:

Originally Posted by Didier Spaier (Post 5821553)

Eh? Wasn't the GTK 3 port over 90 percent done?

Didier Spaier

02-23-2018 12:53 PM

Quote:

Originally Posted by dugan (Post 5823250)

Eh? Wasn't the GTK 3 port over 90 percent done?

No, as far as I know.
As long as 洪任諭 (Hong Jen Yee, aka PCMan) prefers Qt5 over GTK3 this won't change. Unless Andriy Grytsenko forks it, but I wouldn't bet on that either.
Links:
https://github.com/lxde/lxqt/wiki/History
https://blog.lxde.org/2014/05/09/som...o-are-curious/

So, if accessibility matters I recommend MATE.

USUARIONUEVO

02-23-2018 05:13 PM

what arround wayland support ?

Pixxt

02-23-2018 06:17 PM

mesa-demos 8.4.0

https://lists.freedesktop.org/archiv...ry/186542.html

https://mesa.freedesktop.org/archive...-8.4.0.tar.bz2

qunying

02-23-2018 06:41 PM

xfig 3.2.6a
https://sourceforge.net/projects/mcj...-3.2.6a.tar.xz

Darth Vader

02-23-2018 07:13 PM

I have a proposal about enhancing the LAMP stack:

Major PHP frameworks like Laravel, CakePHP or Symfony use a fast database named Redis. There are tons of sites made this way.

So, there are already builds for:

https://slackbuilds.org/repository/14.2/system/redis/
https://slackbuilds.org/repository/1...ies/php-redis/

And the generated packages are very small.

BUT, I think that adding Redis and its PHP bindings to Slackware is a great improvement. There is an entire world beyond Drupal, Joomla! and Wordpress. ;)

Only on Laravel is estimated to had be made hundred of thousands of (custom) sites. Same on CakePHP and Symfony is generally used by pros.

Which pros are pros in PHP, not in Slackware and busy people, then giving them Redis, they will not dismiss from start our glorious operating system.

PS. I hope that there are no political issues with libvdpau-va-gl: https://github.com/i-rinat/libvdpau-va-gl

Without it, the MPlayer and Flash Player cannot use hardware decoding for videos in ANY Intel (graphics) system, and on laptops that results in eating fast the batteries.

USUARIONUEVO

02-23-2018 08:13 PM

@Darth Vader

In your first attemp to request libvdpau-va-gl , i think is good , cause in my slackware 14.2 box im not have video acceleration (i7 7500u with enbedded graphics) , but in current YES , video acceleration works great. (not needed i think)

Testing this libvdpau-va-gl in my 14.2 box i see "videos" , are now with "foggy" , and i uninstalled , i prefer in media player preferences set "x11" acceleration.

Thats my experience after test this package , i think now say NO , cause video quality down extremely...as i can say in "current" , video acceleration in my intel works great.

For test this a easy way is , install libvdpau-va-gl , start a video and make frame capture , uninstall ... make frame capture , and open the two images , and see how , with va-gl , the image is "too foggy" , especially in dark scenes.

Thats my little experience testing that package.

Darth Vader

02-23-2018 08:38 PM

@USUARIONUEVO

Yet, you have ability to configure the video driver in MPlayer and to disable the hardware acceleration in Flash, IF it does not do well.

BUT, just because in some cases does not work properly, is not a must to exclude it. AND the Intel thingies have strong personalities. ;)

BTW, NO hardware decoding for you in MPlayer, if this library is not installed. AND, "X11" output driver from MPlayer has no acceleration at all. It is the bare one.

PS. In fact, your issue is in the VAAPI, this library is just a VDPAU bridge to it, do no processing.

USUARIONUEVO

02-23-2018 09:21 PM

I only share my experience with this package in system , no good for all.

And i see now , i have accelerated video in 14.2

Quote:

glxinfo|grep -i accel
Accelerated: yes

But i need choose x11 render in media player for fullscreen.

And the positive thing is in "current" , no need nothing , no extra packages , probably other intel machines do the same , cause intel video driver is more new in current.

Darth Vader

02-23-2018 10:03 PM

Logical that your OpenGL is hardware accelerated, Mr. Watson! :D

Unless you chose to use the software LLVM thingie, OpenGL work well in a modern Intel chipset.

BUT, the fact that your OpenGL is hardware accelerated is not the same that your MPlayer can use the hardware decoding in a H264 video. ;)

For doing hardware decoding on a video, for MPlayer the flow is:

MPlayer -> VDPAU -> libvdpau-va-gl -> VAAPI -> Video Card <- there is decoded the video.

Because the system lacks the "libvdpau-va-gl", Mplayer will use always the CPU for video decoding.

gmgf	02-24-2018 12:04 AM

libwacom-0.28:

https://sourceforge.net/projects/lin...iles/libwacom/

bassmadrigal

02-24-2018 12:11 AM

Quote:

Originally Posted by USUARIONUEVO (Post 5823466)

I only share my experience with this package in system , no good for all.

And i see now , i have accelerated video in 14.2

But i need choose x11 render in media player for fullscreen.

And the positive thing is in "current" , no need nothing , no extra packages , probably other intel machines do the same , cause intel video driver is more new in current.

As far as I know, glxinfo doesn't provide any information on hardware video decoding. You'd want to use vdpauinfo to find out what is supported on your system with your current software/drivers/libraries. Check out the "Decoder capabilities" output to see what is supported (I made it red to see it easily).

Code:

jbhansen@craven-moorhead:~$ DISPLAY=:0 vdpauinfo

display: :0  screen: 0

API version: 1

Information string: G3DVL VDPAU Driver Shared Library version 1.0



Video surface:



name  width height types

-------------------------------------------

420    16384 16384  NV12 YV12

422    16384 16384  UYVY YUYV

444    16384 16384  Y8U8V8A8 V8U8Y8A8



Decoder capabilities:



name                        level macbs width height

----------------------------------------------------

MPEG1                          --- not supported ---

MPEG2_SIMPLE                    3  9216  2048  1152

MPEG2_MAIN                      3  9216  2048  1152

H264_BASELINE                  41  9216  2048  1152

H264_MAIN                      41  9216  2048  1152

H264_HIGH                      41  9216  2048  1152

VC1_SIMPLE                      1  9216  2048  1152

VC1_MAIN                        2  9216  2048  1152

VC1_ADVANCED                    4  9216  2048  1152

MPEG4_PART2_SP                  3  9216  2048  1152

MPEG4_PART2_ASP                5  9216  2048  1152

DIVX4_QMOBILE                  --- not supported ---

DIVX4_MOBILE                  --- not supported ---

DIVX4_HOME_THEATER            --- not supported ---

DIVX4_HD_1080P                --- not supported ---

DIVX5_QMOBILE                  --- not supported ---

DIVX5_MOBILE                  --- not supported ---

DIVX5_HOME_THEATER            --- not supported ---

DIVX5_HD_1080P                --- not supported ---

H264_CONSTRAINED_BASELINE      --- not supported ---

H264_EXTENDED                  --- not supported ---

H264_PROGRESSIVE_HIGH          --- not supported ---

H264_CONSTRAINED_HIGH          --- not supported ---

H264_HIGH_444_PREDICTIVE      --- not supported ---

HEVC_MAIN                      --- not supported ---

HEVC_MAIN_10                  --- not supported ---

HEVC_MAIN_STILL                --- not supported ---

HEVC_MAIN_12                  --- not supported ---

HEVC_MAIN_444                  --- not supported ---



Output surface:



name              width height nat types

----------------------------------------------------

B8G8R8A8        16384 16384    y  NV12 YV12 UYVY YUYV Y8U8V8A8 V8U8Y8A8 A8I8 I8A8

R8G8B8A8        16384 16384    y  NV12 YV12 UYVY YUYV Y8U8V8A8 V8U8Y8A8 A8I8 I8A8

R10G10B10A2      16384 16384    y  NV12 YV12 UYVY YUYV Y8U8V8A8 V8U8Y8A8 A8I8 I8A8

B10G10R10A2      16384 16384    y  NV12 YV12 UYVY YUYV Y8U8V8A8 V8U8Y8A8 A8I8 I8A8



Bitmap surface:



name              width height

------------------------------

B8G8R8A8        16384 16384

R8G8B8A8        16384 16384

R10G10B10A2      16384 16384

B10G10R10A2      16384 16384

A8              16384 16384



Video mixer:



feature name                    sup

------------------------------------

DEINTERLACE_TEMPORAL            y

DEINTERLACE_TEMPORAL_SPATIAL    -

INVERSE_TELECINE                -

NOISE_REDUCTION                  y

SHARPNESS                        y

LUMA_KEY                        -

HIGH QUALITY SCALING - L1        -

HIGH QUALITY SCALING - L2        -

HIGH QUALITY SCALING - L3        -

HIGH QUALITY SCALING - L4        -

HIGH QUALITY SCALING - L5        -

HIGH QUALITY SCALING - L6        -

HIGH QUALITY SCALING - L7        -

HIGH QUALITY SCALING - L8        -

HIGH QUALITY SCALING - L9        -



parameter name                  sup      min      max

-----------------------------------------------------

VIDEO_SURFACE_WIDTH              y        48    2048

VIDEO_SURFACE_HEIGHT            y        48    1152

CHROMA_TYPE                      y

LAYERS                          y        0        4



attribute name                  sup      min      max

-----------------------------------------------------

BACKGROUND_COLOR                y

CSC_MATRIX                      y

NOISE_REDUCTION_LEVEL            y      0.00    1.00

SHARPNESS_LEVEL                  y    -1.00    1.00

LUMA_KEY_MIN_LUMA                y

LUMA_KEY_MAX_LUMA                y

All times are GMT -5. The time now is 03:24 AM.

Page 10 of 197

« First

110

Last »

Show 50 post(s) from this thread on one page