bentahyr | 25 May 00:00 2016

Re: Wine 1.9.10 : compilation trial

Thanks for the link but DirectX installation went fine, and since 2010, winetricks appeared and helped to
automate quite a lot of these tiedous tasks.

What I had no luck with is that with or without DirectX native implementation, I have the same memory error.
I'll keep that under the hood when have some time amd try to see with the Wine guys if we can get it to work.

As it used to work on opensolaris, I though it should work but I tried wine 1.7.7 on OI151a9 and if it doesn't
expose the same issue, it doesn't render (this is a very quick test with a very limited number of software so
someone might have different results).

Just as a ref, wine is quite used to have access to proprietary music software and one may want to use it for
picture/photo manipulation software as well (I known, I kind of lost my sense of humor lately but I'm
working on it).

I personally use wine for Win32 legacy software so I'm glad with the current situation but I'd like to
propose a complete solution. I'm running out of time regarding this so I'll pile it up with the other things
I would like to port/package to OI.

Best regards.
Ben

On Tue, May 24, 2016 at 11:37:41AM +0000, ken mays via oi-dev wrote:
> Ben,
> Read this: http://www.dedoimedo.com/games/wine-directx.html
> 
> 
> 
> Ken 
> 
> 
(Continue reading)

Adam Števko | 24 May 13:03 2016
Picon

Vagrant

Hello guys,

while working on OI infra I came across a need to rapidly provision OI hipster boxes. 

I'd be interested in how many people actually use vagrant and would do on Oi hipster. If there is enough
interest, I might package it and bring it to Oi. I'll also need testers. We will be able to build VirtualBox
base boxes only (I don't use vmware on OS X).

I'll also start producing official OI vagrant boxes.

Let me know the interest in the package and how many people are to help test vagrant on Oi.

Cheers,
Adam
bentahyr | 24 May 02:25 2016

Re: Wine 1.9.10 : compilation trial

Yes, you're right...
I was reading what I sent and realized that I may have redefined gregs being regs.
Then I read your mail.

Ok... here is the state :
[...]
gmake[1]: Leaving directory '/home/ben/tmp/wine-1.9.10/programs/winetest'
Wine build complete.

Cool,

To sum up
$ ./configure CFLAGS="-std=gnu99" CXXFLAGS="-std=gnu99"
Patch dlls/ntdll/signal_i386.c to add
#if defined(__sun) && defined(__SVR4)
#include <sys/regset.h>
#endif
$ gmake -j8
Et voila!!

I'm running gmake test and will test a few software.
But it already made my day.

Thanks for your help
Did I say "Cool!!" ?

Best regards.
Ben

On Tue, May 24, 2016 at 01:45:41AM +0200, Aurélien Larcher wrote:
(Continue reading)

bentahyr | 23 May 04:25 2016

Wine 1.9.10 : compilation trial

Hi
I have a bit of free time today and thought I could give wine compilation a go.

configure stage is ok
I need to add the -std=gnu99  to C(XX)FLAGS to pass first error then unfortunately it fails a lot later with
the following error :

gmake[1]: Entering directory '/home/franck/tmp/wine-1.9.10/dlls/ntdll'
gcc -c -o signal_i386.o signal_i386.c -I. -I../../include -D__WINESRC__ -D_NTSYSTEM_ -D_REENTRANT
-fPIC \
  -Wall -pipe -fno-strict-aliasing -Wdeclaration-after-statement -Wempty-body
-Wignored-qualifiers \
  -Wstrict-prototypes -Wtype-limits -Wunused-but-set-parameter -Wvla -Wwrite-strings
-Wpointer-arith \
  -Wlogical-op -fno-omit-frame-pointer -std=gnu99
In file included from signal_i386.c:60:0:
signal_i386.c: In function 'init_handler':
signal_i386.c:342:60: error: 'FS' undeclared (first use in this function)
 #define FS_sig(context)      ((context)->uc_mcontext.gregs[FS])
                                                            ^
../../include/windef.h:337:52: note: in definition of macro 'LOWORD'
 #define LOWORD(l)              ((WORD)((DWORD_PTR)(l) & 0xFFFF))
                                                    ^
signal_i386.c:965:18: note: in expansion of macro 'FS_sig'
     *fs = LOWORD(FS_sig(sigcontext));
                  ^
signal_i386.c:342:60: note: each undeclared identifier is reported only once for each function it
appears in
 #define FS_sig(context)      ((context)->uc_mcontext.gregs[FS])
                                                            ^
(Continue reading)

Denys Rtveliashvili | 22 May 23:50 2016
Gravatar

Re: first steps at compiling libvirt

Hello Bob,

I am afraid that did not help.

Also tried adding -lc, that did not work either.

On the other hand, when I tried to package libvirt those errors seemed 
to went away.
New ones have added however, so I will have to spend some time figuring 
out what do they mean.

With kind regards,
Denys Rtveliashvili

> On Sat, 21 May 2016, Denys Rtveliashvili wrote:
>>
>> The linking command looks OKish to me. The only thing I managed to 
>> find about _fini and _init is that they are related to initialisation 
>> and shutting down of shared libraries. Also, -nostdlib option seems 
>> to "fix" the problem but I am not sure if it is what really should be 
>> done. And I have never seen this on Linux.
>
> Does the linking problem go away if you remove the '-Wl,now'?
>
> Bob
Denys Rtveliashvili | 22 May 14:26 2016
Gravatar

Re: Making VIM run in a modern mode by default


Hi Volker,

I understand. But so far I have not seen a single case of anything 
non-standard being distracting. While the opposite is very much true.

If you launch "vim -N" (non-compatible mode), what could you possibly 
find distracting?

> Hello all!
>
>
>>      The list of gains is huge but the most obvious to me are:
>>      a) syntax highlighting makes it easier to read code and configuration files
>>      b) navigation is possible with arrow keys, page-up/down, home/end
>>      c) selecting blocks of texts and running regexps over them is trivial
>>
>>      And I see not a single downside in the non-compatible vim mode.
>>
>>
>> Maybe my question was stupid but: are there environment for which using "vim"
>> as "vi" breaks things ? (serial consoles ?)
>> If it is the case then Volker has a point.
> Wow, interesting discussion.
>
> Maybe I should clarify my concern.  When I log in as root via the
> system console on a newly installed system, I do not want to be
> distracted by anything "non-standard".  I have not experienced any
> problems yet simply because I have never used vim in non-compatible
> mode as root.
(Continue reading)

Denys Rtveliashvili | 21 May 19:17 2016
Gravatar

Re: Making VIM run in a modern mode by default

Hello Volker,

The list of gains is huge but the most obvious to me are:
a) syntax highlighting makes it easier to read code and configuration files
b) navigation is possible with arrow keys, page-up/down, home/end
c) selecting blocks of texts and running regexps over them is trivial

And I see not a single downside in the non-compatible vim mode.

I have never seen it not working in a limited environment. As long as 
terminal is correctly specified it acts accordingly.

Yes, OpenIndiana is not Linux. But does it mean that it should be worse 
than Linux?

"vi" is a system utility, no question about it. But it is not "ed" or 
"sed". It is used by real human beings. And that's why I would like it 
to be more human oriented.

The original Solaris "vi" also was not set in stone I suppose. Surely it 
was updated over time as people needed a useful tool.

Well to me the olden "vi" is broken. I cannot use it without much 
shivering. I prefer a tool which was polished by many people to be an 
excellent, if not best, text editor. Why would I want to use a 
fossilised "vi" mode which has not effectively developed with time if 
there is a better tool doing all the same thing plus many more?

Really, can you name a few actual examples why "non compatible" mode is 
worse than a "compatible" one?
(Continue reading)

Denys Rtveliashvili | 21 May 18:13 2016
Gravatar

first steps at compiling libvirt

Hello,

I am trying to compile libvirt on OpenIndiana.
There is some good progress here.

However, I found a strange thing which may be Illumos or Solaris  - specific.
Perhaps someone could comment?

root <at> zone1:~/libvirt-1.3.4/src# gcc -std=gnu99 -shared  -fPIC -DPIC -Wl,-z -Wl,text -Wl,-h -Wl,libvirt-qemu.so.0 -o .libs/libvirt-qemu.so.0.1003.4  .libs/libvirt_qemu_la-libvirt-qemu.o   -R/export/home/rtvd/libvirt-1.3.4/src/.libs -R/usr/local/lib ./.libs/libvirt.so -lthread -lxml2 -lavahi-common -lavahi-client -ldbus-1 -lgnutls -lssh2 -lcurl -lsocket -lnsl -lintl  -g -O2 -Wl,-M -Wl,./libvirt_qemu.syms -Wl,-z -Wl,now
Undefined                       first referenced
 symbol                             in file
_fini                               /usr/lib/crti.o  (symbol has no version assigned)
_init                               /usr/lib/crti.o  (symbol has no version assigned)
_lib_version                        /usr/lib/values-Xa.o  (symbol has no version assigned)
ld: fatal: symbol referencing errors. No output written to .libs/libvirt-qemu.so.0.1003.4
collect2: error: ld returned 1 exit status

The linking command looks OKish to me. The only thing I managed to find about _fini and _init is that they are related to initialisation and shutting down of shared libraries. Also, -nostdlib option seems to "fix" the problem but I am not sure if it is what really should be done. And I have never seen this on Linux.

With kind regards,
Denys Rtveliashvili
_______________________________________________
oi-dev mailing list
oi-dev <at> openindiana.org
http://openindiana.org/mailman/listinfo/oi-dev
Denys Rtveliashvili | 21 May 18:07 2016
Gravatar

Making VIM run in a modern mode by default

Hello,

I would like VIM to be more modern by default.

I have also raised a ticket for that: https://www.illumos.org/issues/6986

VIM runs in non-compatible mode by default on many if not all Linux 
distributions and frankly I see no reasons why would someone want to 
have it in "compatible" mode.
Also, VIM manual assumes that VIM would be used in the modern mode 
rather than in "compatible" one.

The tweak is a matter of adding a system-wide file.

So the question is this: are there any real reasons why it must not be done?
If there are none, I would make the change.

With kind regards,
Denys Rtveliashvili
Aurélien Larcher | 18 May 12:07 2016
Picon
Gravatar

#6981 VLC 2.2.2 misses icons after libpng update

Hello,
Just pointing to a recent report:
Jim, did you test that VLC works OK after rebuild ?

--
---
Praise the Caffeine embeddings
_______________________________________________
oi-dev mailing list
oi-dev <at> openindiana.org
http://openindiana.org/mailman/listinfo/oi-dev
Michael Kruger | 18 May 03:33 2016
Picon

Public Documentation License (terms)

Hello All,

Given the topic of licensing has come up several times in recent 
discussions, I thought this might be a good time to talk a little more 
about the PDL license and the terms it imposes on authored works subject 
to this license.

Below I have cobbled together a few things I found on the Internet, and 
have drawn a few conclusions from it all. Please let me know whether I 
am incorrect in any of my conclusions.

First, the firebird SQL website 
(http://www.firebirdsql.org/manual/licenses-pdl.html), gives the 
following summary of the PDL:

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

To give you a quick idea of what the PDL is all about, here is a short 
overview of the rights granted and requirements imposed by the PDL:

Free use: everyone may use and distribute PDL-licensed works, for free 
or for money, as long as the license notice is kept intact.

Right to modify: everyone may modify and redistribute PDL-licensed 
works, as long as any modified versions are PDL-licensed too, the 
original license notice is kept intact, and the modifications are 
documented.

Larger works: everyone may incorporate PDL-licensed documentation in a 
larger work. The larger work as a whole need not be released under the 
PDL, but the license requirements must be fulfilled for the PDL-licensed 
parts.

Please be advised that this overview is for informational purposes only 
and has no legal status whatsoever. Only the license text itself - 
following hereafter - is legally binding.

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

The site then follows it up with the full text of the license:

http://www.firebirdsql.org/manual/licenses-pdl-text.html

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

For an example of a license notice, there is the dtrace guide from 
dtrace.org (http://dtrace.org/guide/license.html)

This page says:

Public Documentation License Notice

The contents of this Documentation are subject to the Public 
Documentation License Version 1.01 (the "License"); you may only use 
this Documentation if you comply with the terms of this License. A copy 
of the License is available at http://illumos.org/license/PDL

The Original Documentation is the Solaris Dynamic Tracing Guide. The 
Initial Writer of the Original Documentation is Sun Microsystems 
Copyright (C) 2003-2009. All Rights Reserved. (Initial Writer 
contact(s): http://sun.com).

Contributor(s): Joyent, Inc. Portions created by Joyent, Inc. are 
Copyright (C) 2012. All Rights Reserved. (Contributor contact(s): 
http://joyent.com).

This documentation was derived from the source at illumos docbooks. For 
full changes as required by the PDL, please see the above URL.

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

The above example is essentially the PDL appendix with the blank spaces 
completed.

This suggests every PDL licensed document modified by the OpenIndiana 
community (for example each book included in the PDL licensed 
redistributable docs) would need to included such a notice.

Interestingly, not all of the redistributable books appear to include a 
visible PDL notice.

For example: 
http://makruger.github.io/website/pages/books/content/getstart/html/docinfo.html

I don't see a PDL notice listed anywhere on that page.

However, the XML sources for this book all include a license declaration.

https://github.com/makruger/website/tree/master/pages/books/content/getstart

So, what's going on here. Well, to  answer that question, you need to 
look at the HTML source:

https://github.com/makruger/website/blob/master/pages/books/content/getstart/html/docinfo.html

There it is....a statement saying the document is subject to the PDL.

Other books are a bit more explicit about the license. For example on 
this page 
(http://makruger.github.io/website/pages/books/content/SYSADV1/html/docinfo.html), 
the license declaration is clearly listed.

Presumably only documents modified since their original publication 
would require a completed PDL appendix.

So, what about the 'OpenSolaris' book hosted on the Oracle website?
Are they PDL licensed as well?

The answer to that question is apparently NO.

The 'OpenSolaris' documents hosted on the Oracle website do NOT contain 
any such PDL license declaration either visibly or contained in the HTML 
source. I think it's safe to conclude PDF copies of these documents are 
not PDL licensed either.

--------------------------------------------------------------------
Taking the discussion in a slightly different direction, and looking 
once more at the PDL licensed Dtrace guide on Dtrace.org, the following 
question comes to mind:

How exactly does one go about identifying the original portions from the 
Joyent modified portions?

The answer to this question is found in the last sentence of the license 
notice which says "see the above URL". This URL references the 
Illumos-Docbooks Github repository.

This effectively means for the purpose of complying to the terms of the 
PDL, the editors of that document (illumos/Joyent) relied on the git 
logs as the means of identifying the changes made to the document and 
the contributor(s) who made those changes.

One might even say editing a PDL licensed document outside of some kind 
of change tracking system would likely NOT be in compliance with the 
terms of the license UNLESS they also manually (and might I say 
painstakingly) documented exactly what changed, and precisely who made 
the changes.

The natural conclusion here is using a change tracking system is the 
only practical way to ensure full compliance with the PDL.

I am curious whether the change tracking capabilities of a CMS system 
would be also considered 'in compliance'. I suppose as long as it tracks 
all the changes and who made them, then yes, it would be.

---------------------------------------------------------------------
Taking the discussion in yet another direction,

Given neither Illumos or OpenIndiana are legal entities, I presume for 
the purpose of complying to the terms of the PDL, the individual names 
of each contributor would need to be listed in the PDL notice of any 
document licensed under the terms of the PDL.

So, if the content found on openindiana.org or wiki.openindiana.org were 
PDL licensed, then it would also follow that any document created from 
such content would then need to include all the contributor names in 
it's PDL notice.

And if a large document such as a new handbook was created from such 
content, then the list of contributors would need to include all the 
authors from each and every source document.

While both of the websites support change tracking, this concern might 
be completely irrelevant because none of the documents found on 
openindiana.org appear to contain visible PDL license declarations or 
notices.

Just some thoughts on licensing.....

Michael

Gmane