| View previous topic :: View next topic |
| Author |
Message |
GDH-gentoo
Veteran
Joined: 20 Jul 2019
Posts: 1806
Location: South America
|
 Posted: Sun Jun 18, 2023 3:32 pm Post subject: |
 |
|
| mv wrote: | | Without dbus, gtk+ can be used on non-linux systems, for instance FreeBSD. |
Could you elaborate? I see the reference D-Bus implementation in FreeBSD's Ports and Packages Collection.
| stefan11111 wrote: | | Makes me wonder though, do people really don't mind dbus? |
I don't mind. It provides an IPC mechanism, and, unlike systemd, its scope has stayed at that. Is there a better way to implement an IPC mechanism? Maybe. Am I going to develop a new, better one? Certainly not. Are software packages that currently use D-Bus going to switch to a different one? If there is a strong motivation to make the necessary code adjustments, maybe. Otherwise, likely no.
That said, IMO, using D-Bus only for passing file descriptors to Xorg is overkill, and the combination of using D-Bus for communication with the init system's service manager, and embedding the service manager in the PID 1 program —therefore requiring the PID 1 program to implement D-Bus client functionality—, as systemd does, is a horrible implementation choice.
| stefan11111 wrote: | | Afaik dbus is actually useful on wayland, but it's just bloat on xorg. |
It is used for communication between desktop environment components, regardless of whether it's an X11 or Wayland DE... Without IPC, you don't have a DE. You just have a bunch of unrelated processes running at the same time.
| szatox wrote: | It's just that instead of message flow like: A -> socket -> B
we get something like: A -> socket -> dbus -> socket -> B. |
The network equivalent of that implementation choice is hub-and-spoke topology vs. a full mesh. One socket bound to a pathname vs N sockets bound to pathnames.
| szatox wrote: | Talking to multiple clients at the same time? Yeah, multiple clients talk to dbus over a single socket, just like they would to the server directly.
Multiple servers responding to a single client call? I can vaguely imagine this being useful in a deeply integrated desktop environment, but in this case, you probably also need some application-specific logic at the event forwarder, making generic solutions useless anyway... |
The network equivalent of what D-Bus provides is a multiaccess network with support for both unicast and broadcast communications. Yes, that needs some sort of addressing. D-Bus uses bus names for that.
_________________
| NeddySeagoon wrote: | I'm not a witch, I'm a retired electronics engineer  |
| Ionen wrote: | | As a packager I just don't want things to get messier with weird build systems and multiple toolchains requirements though |
Last edited by GDH-gentoo on Sun Jun 18, 2023 4:34 pm; edited 1 time in total |
|
| Back to top |
|
 |
mv
Watchman
Joined: 20 Apr 2005
Posts: 6780
|
| Posted: Sun Jun 18, 2023 4:08 pm Post subject: |
|
|
It seems that my memory was wrong: I picked it from some NetBSD patch (not FreeBSD) which in turn I found somewhere under the name gnome-next. I cannot say anything about the history or further development of these patches under NetBSD as I never had found any further reference. |
|
| Back to top |
|
|
stefan11111
l33t
Joined: 29 Jan 2023
Posts: 951
Location: Romania
|
| Posted: Sun Jun 18, 2023 5:02 pm Post subject: |
|
|
| Anon-E-moose wrote: | | I was one of the early people messing with gtk+:3 and the at bridge |
Thank you for your contributions.
| Anon-E-moose wrote: |
At the time, I was using minimal X (openbox, tint) and didn't need dbus (which at-spi* bridge pulls in) thus the patches, etc. |
I run X with dwm, so this is exactly my use case.
| Anon-E-moose wrote: |
I've since walked away from X, in favor of wayland and as mentioned it's pretty much impossible to use without dbus, thus I haven't used a patched gtk+:3 in quite a while. I've long walked away from gtk2 as it's unmaintained, and won't work with most newer software, especially wayland stuff. |
I don't use wayland for 2 reasons:
1. Dwm and slstatus don't work on wayland, though there is dwl.
2. Wayland seems like a bunch of hacks on top of one another, evan having a full X server embedded in it.
| Anon-E-moose wrote: |
but I'm not against progress as a general rule. |
Progress is good, bloat is not. Look at math development to see the difference.
| Anon-E-moose wrote: |
I use an ancient eudev (because it works) |
I believe you know about this already, but:
https://wiki.gentoo.org/wiki/Old_Fashioned_Gentoo_Install
_________________
My overlay: https://github.com/stefan11111/stefan_overlay
INSTALL_MASK="/etc/systemd /lib/systemd /usr/lib/systemd /usr/lib/modules-load.d *udev* /usr/lib/tmpfiles.d *tmpfiles* /var/lib/dbus /usr/bin/gdbus /lib/udev" |
|
| Back to top |
|
|
stefan11111
l33t
Joined: 29 Jan 2023
Posts: 951
Location: Romania
|
| Posted: Sun Jun 18, 2023 5:05 pm Post subject: |
|
|
| GDH-gentoo wrote: |
| stefan11111 wrote: | | Afaik dbus is actually useful on wayland, but it's just bloat on xorg. |
It is used for communication between desktop environment components, regardless of whether it's an X11 or Wayland DE... Without IPC, you don't have a DE. You just have a bunch of unrelated processes running at the same time.
|
I get the need for inter-process communication, but if things work fine without dbus, why should I add it to my system? I use dwm, which is as minimal as it gets and doesn't need dbus.
_________________
My overlay: https://github.com/stefan11111/stefan_overlay
INSTALL_MASK="/etc/systemd /lib/systemd /usr/lib/systemd /usr/lib/modules-load.d *udev* /usr/lib/tmpfiles.d *tmpfiles* /var/lib/dbus /usr/bin/gdbus /lib/udev" |
|
| Back to top |
|
|
GDH-gentoo
Veteran
Joined: 20 Jul 2019
Posts: 1806
Location: South America
|
| Posted: Sun Jun 18, 2023 7:03 pm Post subject: |
|
|
| stefan11111 wrote: | | I get the need for inter-process communication, but if things work fine without dbus, why should I add it to my system? |
Gentoo won't add it if nothing that you have installed needs it.
_________________
| NeddySeagoon wrote: | | I'm not a witch, I'm a retired electronics engineer |
| Ionen wrote: | | As a packager I just don't want things to get messier with weird build systems and multiple toolchains requirements though |
|
|
| Back to top |
|
|
stefan11111
l33t
Joined: 29 Jan 2023
Posts: 951
Location: Romania
|
| Posted: Sun Jun 18, 2023 7:52 pm Post subject: |
|
|
| GDH-gentoo wrote: | | stefan11111 wrote: | | I get the need for inter-process communication, but if things work fine without dbus, why should I add it to my system? |
Gentoo won't add it if nothing that you have installed needs it. |
kek
If only. Look at my package.provided. I already posted it.
_________________
My overlay: https://github.com/stefan11111/stefan_overlay
INSTALL_MASK="/etc/systemd /lib/systemd /usr/lib/systemd /usr/lib/modules-load.d *udev* /usr/lib/tmpfiles.d *tmpfiles* /var/lib/dbus /usr/bin/gdbus /lib/udev" |
|
| Back to top |
|
|
szatox
Advocate
Joined: 27 Aug 2013
Posts: 3503
|
| Posted: Sun Jun 18, 2023 9:13 pm Post subject: |
|
|
| GDH-gentoo wrote: |
| szatox wrote: | It's just that instead of message flow like: A -> socket -> B
we get something like: A -> socket -> dbus -> socket -> B. |
The network equivalent of that implementation choice is hub-and-spoke topology vs. a full mesh. One socket bound to a pathname vs N sockets bound to pathnames.
|
Servers won't talk to other servers and clients won't talk to other clients though, will they?
I imagine a handful of hubs with a few spokes each instead of 1 hub with all spokes. Even if it does indeed use more connections (which translates to file descriptors), it won't be anywhere near exponential.
E.g. a bunch of applications I have use dbus to show notifications. Even if delegating this is a good choice (press X to doubt), those applications have no reason to talk to each other.
How many servers does one client program talk to? Notifications is 1.
Does browser understand acpi status? No.
Does mail client understand file download completed? No.
Does irc understand "you've got mail"? No.
Does torrent care about network status changes? Yes, but iproute does not send updates and network is inherently unreliable anyway, so do it yourself, torrent. And browser. And mail. And irc. And... And...
On my laptop with lxde (and dbus) I have like 250 connections over 20 unix sockets.
On a headless server it's 90 connections over 30 sockets, without dbus. Mostly postfix talking to itself. I'm not worried. |
|
| Back to top |
|
|
Hu
Administrator
Joined: 06 Mar 2007
Posts: 23100
|
| Posted: Sun Jun 18, 2023 10:44 pm Post subject: |
|
|
| stefan11111 wrote: | | GDH-gentoo wrote: | | stefan11111 wrote: | | I get the need for inter-process communication, but if things work fine without dbus, why should I add it to my system? |
Gentoo won't add it if nothing that you have installed needs it. |
kek
If only. Look at my package.provided. I already posted it. |
As GDH-gentoo wrote, if no ebuild specifies a dependency on dbus, then Portage will not install dbus unless specifically requested by the user. If an ebuild does specify a dependency, then Portage will install it. If that dependency is incorrect, and you have test results to prove it to be incorrect, please advise that the ebuild should be modified. Note that proof in this context means that:- The package builds correctly without dbus installed, possibly with configure options to tell it not to use dbus. The build does not display any scary messages warning about the inability to find dbus.
- The package runs correctly without dbus installed. Again, there should be no scary messages about its absence, if it was built not to use dbus.
- The package either does not use dbus, or:
- The functionality lost by the lack of dbus does not render the package unusable and
- The Gentoo maintainer can be convinced to place that usage behind a USE flag.
Per recent comments on the forum from Gentoo maintainers, I would not be surprised to see that last point to be a difficult argument to make. As I understand those comments, packages which use dbus at all tend to use it in ways that the maintainers believe the average desktop user will want and will be surprised to find missing. Therefore, I would expect that if you convince the maintainer at all, the dependency will move behind a default-enabled USE flag, so that users continue to get dbus by default. |
|
| Back to top |
|
|
GDH-gentoo
Veteran
Joined: 20 Jul 2019
Posts: 1806
Location: South America
|
| Posted: Sun Jun 18, 2023 11:18 pm Post subject: |
|
|
| szatox wrote: | | Servers won't talk to other servers and clients won't talk to other clients though, will they? |
In the desktop environment case, it's several clients talking to several servers using a common transport: the bus abstraction implemented by dbus-daemon. And interested 'nodes' receiving broadcasted notifications (signals) on said transport. In all cases, using a common protocol.
_________________
| NeddySeagoon wrote: | | I'm not a witch, I'm a retired electronics engineer |
| Ionen wrote: | | As a packager I just don't want things to get messier with weird build systems and multiple toolchains requirements though |
|
|
| Back to top |
|
|
stefan11111
l33t
Joined: 29 Jan 2023
Posts: 951
Location: Romania
|
| Posted: Mon Jun 19, 2023 12:22 am Post subject: |
|
|
| Hu wrote: | | stefan11111 wrote: | | GDH-gentoo wrote: | | stefan11111 wrote: | | I get the need for inter-process communication, but if things work fine without dbus, why should I add it to my system? |
Gentoo won't add it if nothing that you have installed needs it. |
kek
If only. Look at my package.provided. I already posted it. |
As GDH-gentoo wrote, if no ebuild specifies a dependency on dbus, then Portage will not install dbus unless specifically requested by the user. If an ebuild does specify a dependency, then Portage will install it. If that dependency is incorrect, and you have test results to prove it to be incorrect, please advise that the ebuild should be modified. Note that proof in this context means that:- The package builds correctly without dbus installed, possibly with configure options to tell it not to use dbus. The build does not display any scary messages warning about the inability to find dbus.
- The package runs correctly without dbus installed. Again, there should be no scary messages about its absence, if it was built not to use dbus.
- The package either does not use dbus, or:
- The functionality lost by the lack of dbus does not render the package unusable and
- The Gentoo maintainer can be convinced to place that usage behind a USE flag.
Per recent comments on the forum from Gentoo maintainers, I would not be surprised to see that last point to be a difficult argument to make. As I understand those comments, packages which use dbus at all tend to use it in ways that the maintainers believe the average desktop user will want and will be surprised to find missing. Therefore, I would expect that if you convince the maintainer at all, the dependency will move behind a default-enabled USE flag, so that users continue to get dbus by default. |
Not dbus, but:
| Code: | $ equery d rgb
* These packages depend on rgb:
x11-base/xorg-server-21.1.8 (>=x11-apps/rgb-1.0.3) |
Builds correctly.
Runs correctly.
Don't know if it needs rgb, but I am running xorg right now without rgb.
Anything else?
One more:
| Code: | $ equery d freedesktop-icon-theme
* These packages depend on freedesktop-icon-theme:
www-client/librewolf-114.0.1_p2 (virtual/freedesktop-icon-theme)
x11-libs/gtk+-3.24.38 (!adwaita-icon-theme ? virtual/freedesktop-icon-theme) |
| Code: | $ equery d gtk-update-icon-cache
* These packages depend on gtk-update-icon-cache:
dev-qt/qtgui-5.15.10 (dev-util/gtk-update-icon-cache)
x11-libs/gtk+-3.24.38 (>=dev-util/gtk-update-icon-cache-3) |
Builds correctly.
Runs correctly, although gtk apps through a warning:
| Code: | (librewolf:11784): Gtk-WARNING **: 20:41:45.906: Could not find the icon 'user-home-symbolic-ltr'. The 'hicolor' theme
was not found either, perhaps you need to install it.
You can get a copy from:
http://icon-theme.freedesktop.org/releases |
Is needed for very few things and does not affect most functionality.
_________________
My overlay: https://github.com/stefan11111/stefan_overlay
INSTALL_MASK="/etc/systemd /lib/systemd /usr/lib/systemd /usr/lib/modules-load.d *udev* /usr/lib/tmpfiles.d *tmpfiles* /var/lib/dbus /usr/bin/gdbus /lib/udev" |
|
| Back to top |
|
|
Zucca
Moderator
Joined: 14 Jun 2007
Posts: 3939
Location: Rasi, Finland
|
| Posted: Mon Jun 19, 2023 11:09 am Post subject: |
|
|
Out of curiosity (and off-topic): Why don't you ditch xorg if you're going for the minimalist possible setup? xorg is pretty much hacks layered atop of another. The earliest in the pile from the 1980s.
There is https://github.com/ghaerr/microwindows this for example.
I think there's at least one another minimal X implementation.
I have wondered why didn't these caught up much. Not that it does matter to me any more. I've moved to various wayland based compisitors long time ago.
Then there's also DirectFB2...
_________________
..: Zucca :..
| My gentoo installs: | init=/sbin/openrc-init
-systemd -logind -elogind seatd |
| Quote: | | I am NaN! I am a man! |
|
|
| Back to top |
|
|
stefan11111
l33t
Joined: 29 Jan 2023
Posts: 951
Location: Romania
|
| Posted: Mon Jun 19, 2023 11:32 am Post subject: |
|
|
| Zucca wrote: | Out of curiosity (and off-topic): Why don't you ditch xorg if you're going for the minimalist possible setup? xorg is pretty much hacks layered atop of another. The earliest in the pile from the 1980s.
There is https://github.com/ghaerr/microwindows this for example.
I think there's at least one another minimal X implementation.
I have wondered why didn't these caught up much. Not that it does matter to me any more. I've moved to various wayland based compisitors long time ago.
Then there's also DirectFB2... |
I will look into these, thanks.
| Code: | /usr/lib/gcc/x86_64-pc-linux-gnu/13/../../../../x86_64-pc-linux-gnu/bin/ld: /home/stefan/Downloads/microwindows/src/obj/drivers/mou_devmice.o:(.data.rel.local+0x0): multiple definition of `mousedev'; /home/stefan/Downloads/microwindows/src/obj/drivers/mou_x11.o:(.data.rel.local+0x0): first defined here
collect2: error: ld returned 1 exit status
make[1]: *** [/home/stefan/Downloads/microwindows/src/Makefile.rules:516: /home/stefan/Downloads/microwindows/src/lib/libmwin.so] Error 1
make[1]: Leaving directory '/home/stefan/Downloads/microwindows/src/mwin'
make: *** [/home/stefan/Downloads/microwindows/src/Makefile.rules:481: subdir-/home/stefan/Downloads/microwindows/src/mwin] Error 2 |
Still, doesn't wayland have a full X server embedded in it? Doesn't that make it at least as bloated and hacky as Xorg by default? Not to mention all the things that wayland needs that xorg doesn't.
_________________
My overlay: https://github.com/stefan11111/stefan_overlay
INSTALL_MASK="/etc/systemd /lib/systemd /usr/lib/systemd /usr/lib/modules-load.d *udev* /usr/lib/tmpfiles.d *tmpfiles* /var/lib/dbus /usr/bin/gdbus /lib/udev" |
|
| Back to top |
|
|
Zucca
Moderator
Joined: 14 Jun 2007
Posts: 3939
Location: Rasi, Finland
|
| Posted: Mon Jun 19, 2023 11:46 am Post subject: |
|
|
| stefan11111 wrote: | | Still, doesn't wayland have a full X server embedded in it? |
No. Wayland is the protocol. It's up to the wayland compositor how to implement it.
You may confuse the "X server inside wayland" with xwayland. That's a program which allows X11-only programs display their graphic side on a wayland compositor. Usually xwayland support is either toggleable via configuration or a compile-time option.
EDIT: Fix typo: xwindows --> xwayland
_________________
..: Zucca :..
| My gentoo installs: | init=/sbin/openrc-init
-systemd -logind -elogind seatd |
| Quote: | | I am NaN! I am a man! |
Last edited by Zucca on Mon Jun 19, 2023 2:19 pm; edited 1 time in total |
|
| Back to top |
|
|
NeddySeagoon
Administrator
Joined: 05 Jul 2003
Posts: 54848
Location: 56N 3W
|
| Posted: Mon Jun 19, 2023 12:02 pm Post subject: |
|
|
stefan11111,
There is/was tinyx too.
It was originally for embedded system and has died and been reborn several times,
I'll leave it to you to find a fork you like.
_________________
Regards,
NeddySeagoon
Computer users fall into two groups:-
those that do backups
those that have never had a hard drive fail. |
|
| Back to top |
|
|
stefan11111
l33t
Joined: 29 Jan 2023
Posts: 951
Location: Romania
|
| Posted: Mon Jun 19, 2023 1:21 pm Post subject: |
|
|
| NeddySeagoon wrote: | stefan11111,
There is/was tinyx too.
It was originally for embedded system and has died and been reborn several times,
I'll leave it to you to find a fork you like. |
It's not in the repos, so I have to hunt down dependencies manually.
tinyx depends on libxfont.
libxfont depends on fontcacheproto and xorg-macros
If I install xorg-macros and copy xorg-macros.pc to /usr/lib*/pkgconfig, ./configure still doesn't detect it.
| Code: | stefan@GentooPC ~/Downloads/libxfont $ ./autogen.sh
autoreconf-2.71: export WARNINGS=
autoreconf-2.71: Entering directory '.'
autoreconf-2.71: configure.ac: not using Gettext
autoreconf-2.71: running: aclocal
configure.ac:41: error: must install xorg-macros 1.10 or later before running autoconf/autogen
configure.ac:41: the top level
autom4te-2.71: error: /usr/bin/m4 failed with exit status: 1
aclocal-1.16: error: /usr/bin/autom4te-2.71 failed with exit status: 1
autoreconf-2.71: error: aclocal failed with exit status: 1 |
I found a version of libXfont that doesn't look for xorg-macros, but it fails to build:
| Code: | ftfuncs.c:57:10: error: #include expects "FILENAME" or <FILENAME>
57 | #include FT_INTERNAL_TRUETYPE_TYPES_H
| ^~~~~~~~~~~~~~~~~~~~~~~~~~~~
ftfuncs.c:59:10: error: #include expects "FILENAME" or <FILENAME>
59 | #include FT_INTERNAL_SFNT_H
| ^~~~~~~~~~~~~~~~~~
ftfuncs.c:60:10: error: #include expects "FILENAME" or <FILENAME>
60 | #include FT_INTERNAL_STREAM_H
| ^~~~~~~~~~~~~~~~~~~~
ftfuncs.c: In function 'FreeTypeOpenInstance':
ftfuncs.c:467:9: error: unknown type name 'TT_Face'; did you mean 'FT_Face'?
467 | TT_Face tt_face;
| ^~~~~~~
| FT_Face
ftfuncs.c:471:9: error: unknown type name 'SFNT_Service'
471 | SFNT_Service sfnt;
| ^~~~~~~~~~~~
ftfuncs.c:472:18: error: 'TT_Face' undeclared (first use in this function); did you mean 'FT_Face'?
472 | tt_face=(TT_Face)face->face;
| ^~~~~~~
| FT_Face
ftfuncs.c:472:18: note: each undeclared identifier is reported only once for each function it appears in
ftfuncs.c:472:26: error: expected ';' before 'face'
472 | tt_face=(TT_Face)face->face;
| ^~~~
| ;
ftfuncs.c:475:23: error: invalid type argument of '->' (have 'int')
475 | if ( ( tt_face->header.Flags & 8 ) != 0 ) {
| ^~
ftfuncs.c:486:19: error: 'SFNT_Service' undeclared (first use in this function)
486 | sfnt = (SFNT_Service)tt_face->sfnt;
| ^~~~~~~~~~~~
ftfuncs.c:486:32: error: expected ';' before 'tt_face'
486 | sfnt = (SFNT_Service)tt_face->sfnt;
| ^~~~~~~
| ;
ftfuncs.c:487:19: error: invalid type argument of '->' (have 'int')
487 | err = sfnt->set_sbit_strike(tt_face,tt_x_ppem,tt_y_ppem,&instance->strike_index);
| ^~
ftfuncs.c: In function 'tt_get_metrics':
ftfuncs.c:815:5: error: unknown type name 'TT_LongMetrics'
815 | TT_LongMetrics longs_m;
| ^~~~~~~~~~~~~~
ftfuncs.c:824:21: error: 'TT_LongMetrics' undeclared (first use in this function)
824 | longs_m = (TT_LongMetrics )header->long_metrics + idx;
| ^~~~~~~~~~~~~~
ftfuncs.c:824:37: error: expected ';' before 'header'
824 | longs_m = (TT_LongMetrics )header->long_metrics + idx;
| ^~~~~~
| ;
ftfuncs.c:825:27: error: invalid type argument of '->' (have 'int')
825 | *bearing = longs_m->bearing;
| ^~
ftfuncs.c:826:27: error: invalid type argument of '->' (have 'int')
826 | *advance = longs_m->advance;
| ^~
ftfuncs.c:829:22: error: 'TT_ShortMetrics' undeclared (first use in this function); did you mean 'FT_Size_Metrics'?
829 | *bearing = ((TT_ShortMetrics*)header->short_metrics)[idx - k];
| ^~~~~~~~~~~~~~~
| FT_Size_Metrics
ftfuncs.c:829:38: error: expected expression before ')' token
829 | *bearing = ((TT_ShortMetrics*)header->short_metrics)[idx - k];
| ^
ftfuncs.c:830:38: error: expected ')' before 'header'
830 | *advance = ((TT_LongMetrics )header->long_metrics)[k - 1].advance;
| ~ ^~~~~~
| )
ftfuncs.c: In function 'ft_get_very_lazy_bbox':
ftfuncs.c:845:9: error: unknown type name 'TT_Face'; did you mean 'FT_Face'?
845 | TT_Face ttface = (TT_Face)face;
| ^~~~~~~
| FT_Face
ftfuncs.c:845:29: error: 'TT_Face' undeclared (first use in this function); did you mean 'FT_Face'?
845 | TT_Face ttface = (TT_Face)face;
| ^~~~~~~
| FT_Face
ftfuncs.c:845:37: error: expected ',' or ';' before 'face'
845 | TT_Face ttface = (TT_Face)face;
| ^~~~
ftfuncs.c:852:31: error: invalid type argument of '->' (have 'int')
852 | tt_get_metrics(&ttface->horizontal, index,
| ^~
ftfuncs.c: In function 'FT_Do_SBit_Metrics':
ftfuncs.c:914:5: error: unknown type name 'SFNT_Service'
914 | SFNT_Service sfnt;
| ^~~~~~~~~~~~
ftfuncs.c:915:5: error: unknown type name 'TT_Face'; did you mean 'FT_Face'?
915 | TT_Face face;
| ^~~~~~~
| FT_Face
ftfuncs.c:918:5: error: unknown type name 'TT_SBit_Strike'
918 | TT_SBit_Strike strike;
| ^~~~~~~~~~~~~~
ftfuncs.c:919:5: error: unknown type name 'TT_SBit_Range'
919 | TT_SBit_Range range;
| ^~~~~~~~~~~~~
ftfuncs.c:920:5: error: unknown type name 'TT_SBit_MetricsRec'
920 | TT_SBit_MetricsRec elem_metrics;
| ^~~~~~~~~~~~~~~~~~
ftfuncs.c:931:13: error: 'TT_Face' undeclared (first use in this function); did you mean 'FT_Face'?
931 | face = (TT_Face)ft_face;
| ^~~~~~~
| FT_Face
ftfuncs.c:931:21: error: expected ';' before 'ft_face'
931 | face = (TT_Face)ft_face;
| ^~~~~~~
| ;
ftfuncs.c:932:15: error: 'SFNT_Service' undeclared (first use in this function)
932 | sfnt = (SFNT_Service)face->sfnt;
| ^~~~~~~~~~~~
ftfuncs.c:932:28: error: expected ';' before 'face'
932 | sfnt = (SFNT_Service)face->sfnt;
| ^~~~
| ;
ftfuncs.c:934:48: error: invalid type argument of '->' (have 'int')
934 | if (strike_index != 0xFFFFU && sfnt && sfnt->find_sbit_image &&
| ^~
ftfuncs.c:935:17: error: invalid type argument of '->' (have 'int')
935 | sfnt->load_sbits) {
| ^~
ftfuncs.c:937:21: error: invalid type argument of '->' (have 'int')
937 | error = sfnt->find_sbit_image( face, glyph_index, strike_index,
| ^~
ftfuncs.c:945:18: error: invalid type argument of '->' (have 'int')
945 | stream = face->root.stream;
| ^~
ftfuncs.c:949:17: error: invalid type argument of '->' (have 'int')
949 | error = face->goto_table( face, TTAG_EBDT, stream, 0 );
| ^~
ftfuncs.c:951:19: error: invalid type argument of '->' (have 'int')
951 | error = face->goto_table( face, TTAG_bdat, stream, 0 );
| ^~
ftfuncs.c:955:16: warning: implicit declaration of function 'FT_STREAM_POS' [-Wimplicit-function-declaration]
955 | ebdt_pos = FT_STREAM_POS();
| ^~~~~~~~~~~~~
ftfuncs.c:955:16: warning: nested extern declaration of 'FT_STREAM_POS' [-Wnested-externs]
ftfuncs.c:958:10: warning: implicit declaration of function 'FT_STREAM_SEEK' [-Wimplicit-function-declaration]
958 | if ( FT_STREAM_SEEK( ebdt_pos + glyph_offset ) )
| ^~~~~~~~~~~~~~
ftfuncs.c:958:10: warning: nested extern declaration of 'FT_STREAM_SEEK' [-Wnested-externs]
ftfuncs.c:961:17: error: invalid type argument of '->' (have 'int')
961 | error = sfnt->load_sbit_metrics( stream, range, &elem_metrics );
| ^~
ftfuncs.c:965:50: error: request for member 'width' in something not a structure or union
965 | metrics_return->width = (FT_Pos)elem_metrics.width << 6;
| ^
ftfuncs.c:966:50: error: request for member 'height' in something not a structure or union
966 | metrics_return->height = (FT_Pos)elem_metrics.height << 6;
| ^
ftfuncs.c:968:56: error: request for member 'horiBearingX' in something not a structure or union
968 | metrics_return->horiBearingX = (FT_Pos)elem_metrics.horiBearingX << 6;
| ^
ftfuncs.c:969:56: error: request for member 'horiBearingY' in something not a structure or union
969 | metrics_return->horiBearingY = (FT_Pos)elem_metrics.horiBearingY << 6;
| ^
ftfuncs.c:970:56: error: request for member 'horiAdvance' in something not a structure or union
970 | metrics_return->horiAdvance = (FT_Pos)elem_metrics.horiAdvance << 6;
| ^
ftfuncs.c:972:56: error: request for member 'vertBearingX' in something not a structure or union
972 | metrics_return->vertBearingX = (FT_Pos)elem_metrics.vertBearingX << 6;
| ^
ftfuncs.c:973:56: error: request for member 'vertBearingY' in something not a structure or union
973 | metrics_return->vertBearingY = (FT_Pos)elem_metrics.vertBearingY << 6;
| ^
ftfuncs.c:974:56: error: request for member 'vertAdvance' in something not a structure or union
974 | metrics_return->vertAdvance = (FT_Pos)elem_metrics.vertAdvance << 6;
| ^
ftfuncs.c: In function 'restrict_code_range_by_str':
ftfuncs.c:1963:29: warning: implicit declaration of function 'isspace' [-Wimplicit-function-declaration]
1963 | while (',' == *p || isspace(*p))
| ^~~~~~~
ftfuncs.c:74:1: note: include '<ctype.h>' or provide a declaration of 'isspace'
73 | #include "xttcap.h"
+++ |+#include <ctype.h> |
_________________
My overlay: https://github.com/stefan11111/stefan_overlay
INSTALL_MASK="/etc/systemd /lib/systemd /usr/lib/systemd /usr/lib/modules-load.d *udev* /usr/lib/tmpfiles.d *tmpfiles* /var/lib/dbus /usr/bin/gdbus /lib/udev" |
|
| Back to top |
|
|
stefan11111
l33t
Joined: 29 Jan 2023
Posts: 951
Location: Romania
|
| Posted: Mon Jun 19, 2023 11:56 pm Post subject: |
|
|
| Zucca wrote: | | stefan11111 wrote: | | Still, doesn't wayland have a full X server embedded in it? |
No. Wayland is the protocol. It's up to the wayland compositor how to implement it.
You may confuse the "X server inside wayland" with xwayland. That's a program which allows X11-only programs display their graphic side on a wayland compositor. Usually xwayland support is either toggleable via configuration or a compile-time option.
EDIT: Fix typo: xwindows --> xwayland |
Is wayland useful without the X server? Just having a WM/DE with no apps or crippled apps doesn't seem that useful.
From my understanding, there is no way to write a 4k lines of code WM for wayland, like with xorg.
There is no way to run a WM with a different compositor or with none.
Wayland needs pipewire for screen capture, xorg doesn't.
Wayland needs pam and dbus to work, xorg doesn't.
There are probably other things I don't know about.
_________________
My overlay: https://github.com/stefan11111/stefan_overlay
INSTALL_MASK="/etc/systemd /lib/systemd /usr/lib/systemd /usr/lib/modules-load.d *udev* /usr/lib/tmpfiles.d *tmpfiles* /var/lib/dbus /usr/bin/gdbus /lib/udev" |
|
| Back to top |
|
|
stefan11111
l33t
Joined: 29 Jan 2023
Posts: 951
Location: Romania
|
| Posted: Tue Jun 20, 2023 12:29 am Post subject: |
|
|
For now I added mv's gtk to my overlay, fixed the dependencies and added the patch to make gtk2 build with my atk.
Still, I would like to know how to have local patches for different versions of the same program.
I have that gtk2 patch. How can I get it to apply to gtk2 and not to gtk3 without a modified ebuild?
_________________
My overlay: https://github.com/stefan11111/stefan_overlay
INSTALL_MASK="/etc/systemd /lib/systemd /usr/lib/systemd /usr/lib/modules-load.d *udev* /usr/lib/tmpfiles.d *tmpfiles* /var/lib/dbus /usr/bin/gdbus /lib/udev" |
|
| Back to top |
|
|
mv
Watchman
Joined: 20 Apr 2005
Posts: 6780
|
| Posted: Tue Jun 20, 2023 5:06 am Post subject: |
|
|
| stefan11111 wrote: | | Is wayland useful without the X server? Just having a WM/DE with no apps or crippled apps doesn't seem that useful. |
The latter is what you are running on a minimalistic X server, isn't it? As always, it depends on what programs you use.
Most programs which I use run meanwhile without xwayland, most notably my terminal program kitty and my main web browser google-chrome. I consider this an enormous security gain compared to X where your web browser (even if run as a different user with limited permissions) - once hacked - can read the password you enter in the terminal.
| Quote: | | From my understanding, there is no way to write a 4k lines of code WM for wayland, like with xorg. |
Yes, instead you have that bloat (e.g. hardware drivers for input devices) in the X server while the wayland server is minimalistic (AKA: does not exist).
| Quote: | | There is no way to run a WM with a different compositor or with none. |
The concepts are different. The compositor is responsible for everything instead of the X server. You use libraries instead of communicating to a server. Many compositors use the same libraries (e.g. for the mentioned input devices), the minimalistic compositors usually base on gui-libs/wlroots. This is a big advantage: Instead of being forced to let every application talk to a server (and therefore practically giving it full hardware access which enables such things as password sniffing), the compositor is in full control on how to use the libraries and for what.
| Quote: | | Wayland needs pipewire for screen capture, xorg doesn't. |
So wayland obeys the "one tool one job" tradition instead of trying to be everything at once as X (communication platform + graphic drivers + input drivers + 3d accelleration drivers + ...). Yet another advantage.
| Quote: | | Wayland needs pam and dbus to work, xorg doesn't. |
No, wayland does not need pam (it was so in the beginning, but meanwhile I could remove pam again from my system). I am not sure about dbus.
Edit: I just learnt that emacs-29.5(?) is running meanwhile without xwayland, too. Why don't we have emacs:29 in gentoo? |
|
| Back to top |
|
|
colo-des
Tux's lil' helper
Joined: 20 May 2011
Posts: 97
|
| Posted: Tue Jun 20, 2023 6:35 am Post subject: |
|
|
Hi Stefan.
Reading the thread, I see that it has been complicated with atk, a long time ago I presumed that there would
be problems with atk and masked it the same for gtk.
| Code: | $ cat /etc/portage/package.mask/package.mask |grep atk
>dev-libs/atk-2.38.0
$ cat /etc/portage/package.mask/package.mask |grep atk
>x11-libs/gtk+-3.24.22 |
In blfs I now have these versions, and looking to create packages for the new versions.
| Code: | my-user [ ~ ]$ pkginfo -i |grep gtk
gtk 2.24.33-1
my-user [ ~ ]$ pkginfo -i |grep gtk3
gtk3 3.24.34-1 |
Just cloned your git... thanks for keeping those treasures. )
I have four complete copies of the official portage tree in case you need anything from there.
Before synchronizing I always make copies just in case.
| Code: |
portage-09-05-2021
portage-12-03-2023
portage-28-12-2021
portage-29-08-2021 |
To tell the truth I don't want to advance beyond atk-2.38.0 and gtk-3.24.34, since everything points to gtk4
At this point the updates are part of a gtk harakiri.
I'm looking at a gtk fork called ctk, wanting to create blfs ports and completely separate from gtk.
https://github.com/cafe-desktop/ctk
It would be interesting to be able to contact outside the forum to exchange experiences.
Greetings and once again thank you very much for uploading the mv to your git.
Also eternally grateful to mv for creating and maintaining the ebuilds and patches to disable atk-bridge
without which I could not remove dbus.
-------------------------------------
P.S. here in blfs and gentoo without: pam, dbus, atk-spi2, elogind not even graphical login. |
|
| Back to top |
|
|
stefan11111
l33t
Joined: 29 Jan 2023
Posts: 951
Location: Romania
|
| Posted: Tue Jun 20, 2023 2:54 pm Post subject: |
|
|
| colo-des wrote: | Hi Stefan.
Reading the thread, I see that it has been complicated with atk, a long time ago I presumed that there would
be problems with atk and masked it the same for gtk.
| Code: | $ cat /etc/portage/package.mask/package.mask |grep atk
>dev-libs/atk-2.38.0
$ cat /etc/portage/package.mask/package.mask |grep atk
>x11-libs/gtk+-3.24.22 |
In blfs I now have these versions, and looking to create packages for the new versions.
| Code: | my-user [ ~ ]$ pkginfo -i |grep gtk
gtk 2.24.33-1
my-user [ ~ ]$ pkginfo -i |grep gtk3
gtk3 3.24.34-1 |
Just cloned your git... thanks for keeping those treasures. )
I have four complete copies of the official portage tree in case you need anything from there.
Before synchronizing I always make copies just in case.
| Code: |
portage-09-05-2021
portage-12-03-2023
portage-28-12-2021
portage-29-08-2021 |
To tell the truth I don't want to advance beyond atk-2.38.0 and gtk-3.24.34, since everything points to gtk4
At this point the updates are part of a gtk harakiri.
I'm looking at a gtk fork called ctk, wanting to create blfs ports and completely separate from gtk.
https://github.com/cafe-desktop/ctk
It would be interesting to be able to contact outside the forum to exchange experiences.
Greetings and once again thank you very much for uploading the mv to your git.
Also eternally grateful to mv for creating and maintaining the ebuilds and patches to disable atk-bridge
without which I could not remove dbus.
|
I added gtk3 24.24-r1 and 24.22.
For now, there is also perfect gentleman's overlay, which also has a patched gtk.
There are some patches in mv's overlay that aren't in perfect gentleman's overlay, but I don't think they are really needed for new gtk.
I also updated this to work with these older versions of gtk3.
| colo-des wrote: |
P.S. here in blfs and gentoo without: pam, dbus, atk-spi2, elogind not even graphical login. |
Bloat is a capricious beast. What about *kit, udev and systemd-utils?
Also, can someone tell me when do distfiles with no ebuild in ::gentoo?
To mitigate this, we have needy doing god's work, keeping distfiles even older than me.
_________________
My overlay: https://github.com/stefan11111/stefan_overlay
INSTALL_MASK="/etc/systemd /lib/systemd /usr/lib/systemd /usr/lib/modules-load.d *udev* /usr/lib/tmpfiles.d *tmpfiles* /var/lib/dbus /usr/bin/gdbus /lib/udev" |
|
| Back to top |
|
|
stefan11111
l33t
Joined: 29 Jan 2023
Posts: 951
Location: Romania
|
| Posted: Tue Jun 20, 2023 3:25 pm Post subject: |
|
|
| mv wrote: | | stefan11111 wrote: | | Is wayland useful without the X server? Just having a WM/DE with no apps or crippled apps doesn't seem that useful. |
The latter is what you are running on a minimalistic X server, isn't it? As always, it depends on what programs you use.
|
Most apps work on my minimalist setup. Those that don't work are proprietary games I have no way to debug.
| mv wrote: |
Most programs which I use run meanwhile without xwayland, most notably my terminal program kitty and my main web browser google-chrome. I consider this an enormous security gain compared to X where your web browser (even if run as a different user with limited permissions) - once hacked - can read the password you enter in the terminal.
|
All security goes out the window of you use backdoored proprietary software that steals your data.
| mv wrote: |
| Quote: | | From my understanding, there is no way to write a 4k lines of code WM for wayland, like with xorg. |
Yes, instead you have that bloat (e.g. hardware drivers for input devices) in the X server while the wayland server is minimalistic (AKA: does not exist).
|
| Code: | xf86-input-keyboard-..> 2007-09-28 16:22 230K
xf86-input-mouse-1.9..> 2018-06-19 04:36 373K |
I'll live with this if this is what it takes to not sell my soul to redhat.
| mv wrote: |
| Quote: | | There is no way to run a WM with a different compositor or with none. |
The concepts are different. The compositor is responsible for everything instead of the X server. You use libraries instead of communicating to a server. Many compositors use the same libraries (e.g. for the mentioned input devices), the minimalistic compositors usually base on gui-libs/wlroots. This is a big advantage: Instead of being forced to let every application talk to a server (and therefore practically giving it full hardware access which enables such things as password sniffing), the compositor is in full control on how to use the libraries and for what.
|
So no way to not use a compositor or a very minimal one like xcompmgr to get the best performance possible, say in games.
At the other end, there is no way to use compiz.
| mv wrote: |
| Quote: | | Wayland needs pipewire for screen capture, xorg doesn't. |
So wayland obeys the "one tool one job" tradition instead of trying to be everything at once as X (communication platform + graphic drivers + input drivers + 3d accelleration drivers + ...). Yet another advantage.
|
Yes just a few lines above you say that the compositor/WM/DE implements everything. How is that any better that Xorg implementing everything, except compositing.
Also, some things should be done together(though definitely not int the systemd way). Overly separating tasks is oop.
Also, you have those things called portals on wayland. How is that not completely unnecessary and hacky.
| mv wrote: |
| Quote: | | Wayland needs pam and dbus to work, xorg doesn't. |
No, wayland does not need pam (it was so in the beginning, but meanwhile I could remove pam again from my system). I am not sure about dbus.
Edit: I just learnt that emacs-29.5(?) is running meanwhile without xwayland, too. Why don't we have emacs:29 in gentoo? |
Still, after removing systemd-utils from package.mask:
| Code: | $ doas emerge -1av wlroots
These are the packages that would be merged, in order:
Calculating dependencies... done!
Dependency resolution took 1.33 s.
[ebuild N ] gui-libs/eglexternalplatform-1.1::gentoo 11 KiB
[ebuild N ] app-text/scdoc-1.11.2-r1::gentoo 13 KiB
[ebuild N ] acct-group/audio-0-r1::gentoo 0 KiB
[ebuild N ] acct-group/cdrom-0-r1::gentoo 0 KiB
[ebuild N ] acct-group/dialout-0-r1::gentoo 0 KiB
[ebuild N ] acct-group/disk-0-r1::gentoo 0 KiB
[ebuild N ] acct-group/floppy-0::gentoo 0 KiB
[ebuild N ] acct-group/input-0-r1::gentoo 0 KiB
[ebuild N ] acct-group/kmem-0-r1::gentoo 0 KiB
[ebuild N ] acct-group/kvm-0-r1::gentoo 0 KiB
[ebuild N ] acct-group/lp-0-r1::gentoo 0 KiB
[ebuild N ] acct-group/render-0-r1::gentoo 0 KiB
[ebuild N ] acct-group/sgx-0::gentoo 0 KiB
[ebuild N ] acct-group/tape-0-r1::gentoo 0 KiB
[ebuild N ] acct-group/tty-0-r1::gentoo 0 KiB
[ebuild N ] acct-group/usb-0-r1::gentoo 0 KiB
[ebuild N ] dev-util/wayland-scanner-1.22.0::gentoo 225 KiB
[ebuild N ] sys-auth/seatd-0.7.0-r2:0/1::gentoo USE="-builtin -elogind -server -systemd" 39 KiB
[ebuild N ] dev-libs/wayland-1.22.0::gentoo USE="-doc -test" ABI_X86="(64) -32 (-x32)" 0 KiB
[ebuild N ] dev-libs/wayland-protocols-1.31::gentoo USE="-test" 0 KiB
[ebuild N ] gui-libs/egl-wayland-1.1.12::gentoo 66 KiB
[ebuild N ] x11-base/xwayland-23.1.2::gentoo USE="(-selinux) -unwind -xcsecurity" VIDEO_CARDS="nvidia" 1,258 KiB
[ebuild N ] sys-apps/systemd-utils-253.5::gentoo USE="(split-usr) udev -acl -boot -kmod (-selinux) -sysusers -test -tmpfiles" ABI_X86="(64) -32 (-x32)" 11,735 KiB
[ebuild N ] virtual/udev-217-r6::gentoo USE="-systemd" 0 KiB
[ebuild N ] sys-fs/udev-init-scripts-35::gentoo 0 KiB
[ebuild N ] virtual/libudev-232-r8:0/1::gentoo USE="-systemd" ABI_X86="(64) -32 (-x32)" 0 KiB
[ebuild N ] gui-libs/wlroots-0.16.2-r1:0/16::gentoo USE="X -drm -libinput -tinywl -vulkan -x11-backend" 576 KiB
Total: 27 packages (27 new), Size of downloads: 13,918 KiB
The following USE changes are necessary to proceed:
(see "package.use" in the portage(5) man page for more details)
# required by virtual/udev-217-r6::gentoo[-systemd]
# required by sys-fs/udev-init-scripts-35::gentoo
>=sys-apps/systemd-utils-253.5 udev
Would you like to add these changes to your config files? [Yes/No] n
|
udev and seatd. Hard pass. There is no USE="suid" for wayland.
Also no way to do this.
_________________
My overlay: https://github.com/stefan11111/stefan_overlay
INSTALL_MASK="/etc/systemd /lib/systemd /usr/lib/systemd /usr/lib/modules-load.d *udev* /usr/lib/tmpfiles.d *tmpfiles* /var/lib/dbus /usr/bin/gdbus /lib/udev" |
|
| Back to top |
|
|
stefan11111
l33t
Joined: 29 Jan 2023
Posts: 951
Location: Romania
|
| Posted: Tue Jun 20, 2023 3:37 pm Post subject: |
|
|
I tried to write a git ebuild for gtk3:
https://github.com/stefan11111/pastebin/blob/main/gtk%2B-3.9999.ebuild
However, I couldn' get it to work. It tries to download a .tar.gz:
| Code: | !!! Couldn't download 'gtk+-3.9999.tar.xz'. Aborting.
!!! Fetch failed for gtk+-3.9999.tar.xz, can't update Manifest
|
Can someone help me fix this?
_________________
My overlay: https://github.com/stefan11111/stefan_overlay
INSTALL_MASK="/etc/systemd /lib/systemd /usr/lib/systemd /usr/lib/modules-load.d *udev* /usr/lib/tmpfiles.d *tmpfiles* /var/lib/dbus /usr/bin/gdbus /lib/udev" |
|
| Back to top |
|
|
sdauth
l33t
Joined: 19 Sep 2018
Posts: 679
Location: Ásgarðr
|
| Posted: Tue Jun 20, 2023 5:18 pm Post subject: |
|
|
| Try to move the inherit [...] line below EAPI=8 |
|
| Back to top |
|
|
stefan11111
l33t
Joined: 29 Jan 2023
Posts: 951
Location: Romania
|
| Posted: Tue Jun 20, 2023 5:34 pm Post subject: |
|
|
| sdauth wrote: | | Try to move the inherit [...] line below EAPI=8 |
It is below:
| Code: | EAPI=8
DESCRIPTION="Gimp ToolKit +"
HOMEPAGE="https://www.gtk.org/"
EGIT_REPO_URI="https://gitlab.gnome.org/GNOME/gtk.git"
inherit git-r3 gnome2 meson-multilib multilib virtualx |
Also, it seems like this may not be the proper way to manage a range of dependency versions:
| Code: | !atk-bridge? ( >=dev-libs/atk-2.10.0[introspection(+)?,${MULTILIB_USEDEP}] )
!atk-bridge? ( <dev-libs/atk-2.46.0[introspection(+)?,${MULTILIB_USEDEP}] ) |
Because when running emerge with certain combinations, I get some warnings:
| Code: | $ doas emerge -pv atk =gtk+-3.24.38
These are the packages that would be merged, in order:
Calculating dependencies... done!
Dependency resolution took 2.18 s.
[ebuild R ] dev-libs/atk-2.38.0::stefan_overlay 0 KiB
[ebuild R ] x11-libs/gtk+-3.24.38:3::stefan_overlay USE="X -adwaita-icon-theme (-aqua) -atk-bridge -broadway -cloudproviders -colord -cups -examples -gtk-doc -introspection -sysprof -test -vim-syntax -wayland -xinerama" ABI_X86="(64) -32 (-x32)" 0 KiB
Total: 2 packages (2 reinstalls), Size of downloads: 0 KiB
WARNING: One or more updates/rebuilds have been skipped due to a dependency conflict:
dev-libs/atk:0
(dev-libs/atk-2.46.0:0/0::gentoo, ebuild scheduled for merge) USE="-introspection" ABI_X86="(64) -32 (-x32)" conflicts with
<dev-libs/atk-2.46.0[abi_x86_64(-)] required by (x11-libs/gtk+-3.24.38:3/3::stefan_overlay, ebuild scheduled for merge) USE="X -adwaita-icon-theme (-aqua) -atk-bridge -broadway -cloudproviders -colord -cups -examples -gtk-doc -introspection -sysprof -test -vim-syntax -wayland -xinerama" ABI_X86="(64) -32 (-x32)" |
_________________
My overlay: https://github.com/stefan11111/stefan_overlay
INSTALL_MASK="/etc/systemd /lib/systemd /usr/lib/systemd /usr/lib/modules-load.d *udev* /usr/lib/tmpfiles.d *tmpfiles* /var/lib/dbus /usr/bin/gdbus /lib/udev" |
|
| Back to top |
|
|
colo-des
Tux's lil' helper
Joined: 20 May 2011
Posts: 97
|
| Posted: Tue Jun 20, 2023 5:54 pm Post subject: |
|
|
@stefan11111
It's just that I didn't want to name them all... I don't have either: gvfs, polkit and nothing at all from systemd and its utils
I only use eudev, which I don't update to.
The main problem is the package update system and it doesn't matter what distribution it is, when you update you
they force you to accept the changes, with all that that means, until you reach the point where we are now, where
has completely broken trust.
This was the reason to jumped into blfs and create my own binary packages, one by one with the dependencies I want.
otherwise, no updates.
It is evident... there is an agenda, they want to turn gnu/linux into a business platform like windows or mac.
By dint of buying wills they are twisting the destiny of free software.
We got to this and the loop will close more and more, RH (among others) has sufficient means.
And we already know how powerful is the "gentleman is a gift of money" is, as the genius of Paco Ivañez sang.
(activate CC for your language).
https://www.youtube.com/watch?v=N_ZRK0tGC0U
Greetings. |
|
| Back to top |
|
|
|
|
You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot vote in polls in this forum
|
|
Unsupported Software
