[OpenIndiana-discuss] ANNOUNCE: OpenSXCE-org/FireFox-43-port-for-all-OpenSolaris-distros with gcc4 flash-plugin support released at bin PKG and the src on GitHub
Alexander Pyhalov
alp at rsu.ru
Wed Nov 25 20:07:27 UTC 2015
Hello.
I've built FF with these patches and got flash player working. But
plugin-container 100% dumps core when I switch VK player to fullscreen.
Stack is not very useful:
sudo pstack /var/cores/plugin-container.1401
Password:
core '/var/cores/plugin-container.1401' of
1401: /usr/lib/firefox/plugin-container
/usr/lib/firefox/browser/plugins/lib
----------------- lwp# 1 / thread# 1 --------------------
e5e5e5e5 ???????? ()
----------------- lwp# 2 / thread# 2 --------------------
feeee835 __pollsys (f6a36200, 2, 0, 0, f69fef74, 0) + 15
fee819e6 poll (f6a36200, 2, ffffffff, fac03585, f69fed78, 0) + 66
fac03672 poll_dispatch (1, f6a36200, 0, 0, 0, 0) + f9
00000002 ???????? ()
----------------- lwp# 3 / thread# 3 --------------------
feeea129 __lwp_park (f661c850, f661c860, 0, feee1e64, f6d11240, 0) + 19
feee4168 cond_wait_queue (f661c850, f661c860, 0, feee29b6, f661c86c, 0)
+ 6a
feee47e0 __cond_wait (f661c850, f661c860, 0, f661c860, fef5d000, 0) +
8f
feee4834 cond_wait (f661c850, f661c860, f66ad070, 0, f664f108,
f661c860) + 2e
feee487d pthread_cond_wait (f661c850, f661c860, f6ae51a0, f661c890) +
24
f3b2e732 ???????? (f661c850, ffffffff, 574, 5cc)
f3c8f705 ???????? (f664f100, 0, f351ef58, f3b1a070)
f3b2e5b4 ???????? (f664f108, 62, f351efc8, f3b2e225)
f3b2e250 ???????? (f664f108, 0, 0, 0)
feee9f3b _thrp_setup (f6d11240) + 88
feeea0d0 _lwp_start (f6d11240, 0, 0, 0, 0, 0)
----------------- lwp# 4 / thread# 4 --------------------
feeea129 __lwp_park (f661c8b0, f661c8c0, 0, 0, 0, 0) + 19
feee4168 cond_wait_queue (f661c8b0, f661c8c0, 0, 1, 0, 0) + 6a
feee47e0 __cond_wait (f661c8b0, f661c8c0, fec90b00, f661c8c0, fef5d000,
0) + 8f
feee4834 cond_wait (f661c8b0, f661c8c0, f31feed8, feee996e, 10,
f661c8c0) + 2e
feee487d pthread_cond_wait (f661c8b0, f661c8c0, 0, fef5d000) + 24
f3b2e732 ???????? (f661c8b0, ffffffff, f31fef38, feee1be8)
f3c8f705 ???????? (f664f150, 0, f31fef58, f3b1a070)
f3b2e5b4 ???????? (f664f158, 62, f31fefc8, f3b2e225)
f3b2e250 ???????? (f664f158, 0, 0, 0)
feee9f3b _thrp_setup (f6d11a40) + 88
feeea0d0 _lwp_start (f6d11a40, 0, 0, 0, 0, 0)
----------------- lwp# 5 / thread# 5 --------------------
feeea129 __lwp_park (f329795c, f329796c, f2e93e48, 0, 0, 0) + 19
feee4168 cond_wait_queue (f329795c, f329796c, f2e93e48, feec6c88, 3,
f2e93de8) + 6a
feee443c cond_wait_common (f329795c, f329796c, f2e93e48, 0, 0, 0) + 27b
feee461e __cond_timedwait (f329795c, f329796c, f2e93ec0, f329796c,
fef5d000, 135347e8) + 85
feee46e4 cond_timedwait (f329795c, f329796c, f2e93ec0, fef5d000,
f6d12240, f329796c) + 35
feee4734 pthread_cond_timedwait (f329795c, f329796c, f2e93ec0,
f329795c) + 2b
f3b2e7ba ???????? (f329795c, 3e8, f2e93f18, fef5d000)
f3b5aa56 ???????? (f3297000, 0, f2e93f58, f3b1a070)
f3b2e5b4 ???????? (f3297928, 76, f2e93fc8, f3b2e225)
f3b2e250 ???????? (f3297928, 0, 0, 0)
feee9f3b _thrp_setup (f6d12240) + 88
feeea0d0 _lwp_start (f6d12240, 0, 0, 0, 0, 0)
----------------- lwp# 6 / thread# 6 --------------------
feeea129 __lwp_park (f2c16d00, f2c16d10, 0, fef5d000, f6d12a40, 0) + 19
feee4168 cond_wait_queue (f2c16d00, f2c16d10, 0, feebe30c, fef6083c, 0)
+ 6a
feee47e0 __cond_wait (f2c16d00, f2c16d10, f6d12a40, f2c16d10, fef5d000,
0) + 8f
feee4834 cond_wait (f2c16d00, f2c16d10, 0, fef5d000, f6d12a40,
f2c16d10) + 2e
feee487d pthread_cond_wait (f2c16d00, f2c16d10, 0, f28db4d8) + 24
f3b2e732 ???????? (f2c16d00, ffffffff, 229, f13fec8c)
f40b9b70 ???????? (f28db020, 0, f13fef58, f3b1a070)
f3b2e5b4 ???????? (f28db048, 66, f13fefc8, f3b2e225)
f3b2e250 ???????? (f28db048, 0, 0, 0)
feee9f3b _thrp_setup (f6d12a40) + 88
feeea0d0 _lwp_start (f6d12a40, 0, 0, 0, 0, 0)
----------------- lwp# 7 / thread# 7 --------------------
feeef1c5 __write (12, f214cfb0, 1000, f3fe0603) + 15
f3fe069b ???????? (f66ef000, 0, 0, 0)
feee9f3b _thrp_setup (f6d13240) + 88
feeea0d0 _lwp_start (f6d13240, 0, 0, 0, 0, 0)
----------------- lwp# 8 / thread# 8 --------------------
feeea129 __lwp_park (f0c61034, f0c6101c, 0, feee1a77, f6d52400,
f0c6104c) + 19
feee4168 cond_wait_queue (f0c61034, f0c6101c, 0, fef5d000, f6d13a40,
fec92a40) + 6a
feee47e0 __cond_wait (f0c61034, f0c6101c, f6d52440, f0c6101c, fef5d000,
f0c61014) + 8f
feee4834 cond_wait (f0c61034, f0c6101c, f07fef58, feee36f3, f6d13a40,
f0c6101c) + 2e
feee487d pthread_cond_wait (f0c61034, f0c6101c, 0, 0) + 24
f4463ddc ???????? (f0c61014, ffffffff, 0, 0)
f4438c5c ???????? (f0c61010, 0, 0, 0)
feee9f3b _thrp_setup (f6d13a40) + 88
feeea0d0 _lwp_start (f6d13a40, 0, 0, 0, 0, 0)
----------------- lwp# 9 / thread# 9 --------------------
feeea129 __lwp_park (f0c8b594, f0c8b57c, 0, f4486165, f0c8b5f0, 3) + 19
feee4168 cond_wait_queue (f0c8b594, f0c8b57c, 0, feee3d98, ef2d7fd8, 0)
+ 6a
feee47e0 __cond_wait (f0c8b594, f0c8b57c, f1bfef38, f0c8b57c, fef5d000,
f0c8b574) + 8f
feee4834 cond_wait (f0c8b594, f0c8b57c, f1bfef58, feee36d4, f0c8b5ac,
f0c8b57c) + 2e
feee487d pthread_cond_wait (f0c8b594, f0c8b57c, f1bfef58, 0) + 24
f4463ddc ???????? (f0c8b574, ffffffff, 0, 0)
f4438c5c ???????? (f0c8b570, 0, 0, 0)
feee9f3b _thrp_setup (f6d14240) + 88
feeea0d0 _lwp_start (f6d14240, 0, 0, 0, 0, 0)
~/.xsession-errors contains
preedit has not been enabled yet. (1, 1)
preedit has not been enabled yet. (1, 2)
Client shut down the connection owned by im_id(1).
Window manager warning: Log level 8: Source ID 472 was not found when
attempting to remove it
###!!! [Parent][MessageChannel::Call] Error: Channel error: cannot
send/recv
###!!! [Parent][MessageChannel::Call] Error:
(msgtype=0xA80007,name=PPluginScriptableObject::Msg_HasMethod) Channel
error: cannot send/recv
###!!! [Parent][MessageChannel::Call] Error:
(msgtype=0xA8000D,name=PPluginScriptableObject::Msg_HasProperty) Channel
error: cannot send/recv
###!!! [Parent][MessageChannel::Call] Error:
(msgtype=0xA80007,name=PPluginScriptableObject::Msg_HasMethod) Channel
error: cannot send/recv
###!!! [Parent][MessageChannel::Call] Error:
(msgtype=0xA8000D,name=PPluginScriptableObject::Msg_HasProperty) Channel
error: cannot send/recv
###!!! [Parent][MessageChannel::Call] Error:
(msgtype=0xA80007,name=PPluginScriptableObject::Msg_HasMethod) Channel
error: cannot send/recv
###!!! [Parent][MessageChannel::Call] Error:
(msgtype=0xA8000D,name=PPluginScriptableObject::Msg_HasProperty) Channel
error: cannot send/recv
###!!! [Parent][MessageChannel::Call] Error:
(msgtype=0xA80007,name=PPluginScriptableObject::Msg_HasMethod) Channel
error: cannot send/recv
---
System Administrator of Southern Federal University Computer Center
Martin Bochnig писал 18.11.2015 06:14:
> Demo-Screenshots:
> http://opensxce.org/.FF/43.0b3/i386/demo_screenshots/
>
> SRC:
> https://github.com/OpenSXCE-org/FireFox-43-port-for-all-OpenSolaris-distros/commits?author=MartinBochnig
> https://github.com/OpenSXCE-org/FireFox-43-port-for-all-OpenSolaris-distros/commit/a217b4c934e5ab7dceb2d8253f03ce8687e505c7
>
> The diff is also atatched and can freely be used for whatever you want:
> https://github.com/OpenSXCE-org/FireFox-43-port-for-all-OpenSolaris-distros/blob/master/FF43.0b3_OpenSolaris_x86_x64.patch
> <<FF43.0b3_OpenSolaris_x86_x64.patch>>
>
>
> BINS:
> http://opensxce.org/.FF/43.0b3/i386/
>
>
> https://raw.githubusercontent.com/OpenSXCE-org/FireFox-43-port-for-all-OpenSolaris-distros/master/README_INSTALLATION/HOW_TO_BUILD.txt
> Easy:
>
> A) clone the repo
>
> B) run ./MAKE_WORLDx.sh
> (if you run into headers related errors until you correct some pieces,
> that's normal - but happens at different places from system to system)
>
> Alternatively: For users of Sun/Oracle's SFW/userland derived
> userlands everything is prepared ready to run under
> oracle-userland-integration/.
>
> C) for SVR4 package creation, such as on OpenSXCE, TribbliX, Sun
> SXCE11 or Sun/Oracle Solaris10.x the place to look is under
> oracle-userland-integration/opensxce/.
>
> Good luck and happy coding :)
>
>
>
> A few words which I cannot keep untold, not possible:
>
> https://github.com/OpenSXCE-org/FireFox-43-port-for-all-OpenSolaris-distros/tree/master/README_INSTALLATION/
>
>
>
>
>
> DONATION: FireFox43.0b3 pkg and src in GitHub unleashed (initial
> version x86/x64 only, SPARC next week [if wanted], Illumos-x86/x64
> kernel patch for flash-plugin support also next week if wanted]*, also
> DRM/KMS for Intel can* be released any time, also my openXsun patches
> and a SPARC-OpenIndiana)
>
> * := unless some so far unknown non-contributor that one never saw
> here for months suddenly opens his western bigmouth and urges me away
> once again.
>
>
> Martin B. aka OpenSXCE.org (_HE_!) herewith **donates** 6
> hyper-fulltime weeks of on average 15 unpaid hours per day (no joke,
> btw including weekends) to the
> OpenSXCE/DilOS/Illumos/OpenSolaris/Sun-Solaris/Oracle-Solaris/OpenIndiana/Hipster
> (and all other OpenSolaris based distros') communities.
>
> Originally he wanted to publish this stuff in April (at the time for
> FireFox37 which he then after certain events only published as binary
> package, which can still be obtained from
> http://opensxce.org/.FF/37/i386/). But testing the water faced him
> with egoistic GIMME GIMME MOAR _____demands*_____, rather than any
> noticable support, like last time when he published OpenSXCE, like
> last time when he had created the SPARC-OpenIndiana LiveDVD, like last
> time when he released SPARC-Xorg in 2006, like last time when he
> published the Qemu port to SPARC-Solaris hosts in 2006, like last time
> when [...]
>
> Folks called him a "capitalist", only because he dared to ask for some
> (so called) "donations".
> Look how foolish this is: A capitalist is somebody who does something
> for a profit. Profit would mean getting more - (a lot!) more - than
> having invested, rather than getting back maybe 5% of the actual costs
> (electricity, computer, hdd, food, powerdrinks, room, all bills in
> Germany are extremely high), plus additionally getting nothing for the
> actual time spent nor any other reward for the lack of income during
> all this time spent. Even a toilet-cleaner is rich in contrast,
> because he does get something for his time.
>
> Who doesn't know (or understand) the difference between M.B. (who
> spent more than ten [probably more than twenty] thousand unpaid hours
> on OpenSolaris community related work since 2005 [and at Blastwave
> since 2003) should read:
> https://en.wikipedia.org/wiki/Capitalist_mode_of_production_%28Marxist_theory%29
>
> "Anti-Social Capitalist" ???
> --->> That's not him who loves OpenSolaris so much that he ruined his
> career.
> NOPE: That's most readers of this mailing list, who only show up once
> in 2 years to make some useless one-liner comment.
> And if they do, then only to criticize those few real OpenIndiana
> contributors that are really working on this (Hipster team!).
> Yesterday there were such comments again (like "Hipster is going
> nowhere" or "write up instructions").
> Thos instructions are on the openindiana page for years. But only
> those truly interested in contributing something would come across
> them, rather than some show-offs and time-wasters who can only blame
> and complain (the referred post was only a single example, other
> mailing list members are much worse).
> True Anti-Social beings are those biz owners or well paid biz-admins
> who deploy OI countless times for commercial installations while
> keeping all the 1000's of Dollars for themselves, rather than giving
> anything back to the Hipster team.
> I must say: It is a mistake that those few working on Hipster are
> giving it away for free.
>
> Free software - the meaning has to be interpreted as open-src, but not
> necessarily free of charge all the time.
> Especially not the binary iso's! Just take this hint as a well
> intended recommendation.
> 1000's are exploiting you, Hipster (rgds. to Rostov).
>
> NetBSD Is Paying For DRM With KMS/GEM
> http://www.phoronix.com/scan.php?page=news_item&px=MTMxMzg
>
> But from me you expect that I release everything always for free,
> including my therefore (!) so far unpublished KMS/DRM (now upgraded to
> Oracle's better version)?
> It fits well to the rest of western hypocrisy to always point the
> finger at others.
> Overpaid folks who repeatedly stated that they have no time to
> contribute anything to OpenSolaris (because it's apparently not
> important enough to them, so they rather take the money elsewhere)
> BLAME me for being an "anti-social capitalist", only because I do
> contribute something for one and a half decades, but find it hard to
> survive?
> That's like Obama saying ISIS is an enemy of the West (rather than its
> own creation and useful tool [proof in Part II below, simply listen to
> McCain's and H.Clinton's own statements on video tape]).
>
>
> {
> # # # # # # # # # # # # # # # # # # # # # # # # # # # #
> # # # Part I.) TECHNICAL SCOPE: # # # # # # # # # # #
> # # # # # # # # # # # # # # # # # # # # # # # # # # # #
> Unlike others who 1:1 either cloned&& paste-used Sun/Oracle's
> Studio-diffs or NetBSD's Firefox gcc-port (or rather: tried hard ...),
> I developed this patch from scratch and used the former only as
> reference, where really required (e.g. where the build terminated or -
> felt lightyears later [when the build would finally succeed for the
> first time!!!] - in cases where some functionality was missing or
> broken, and where it was more convenient to look what's already there,
> although in most cases this was the easy part either way), thanks to
> Rich Teer's book "Solaris Systems programming", the famous Solaris
> series including "Solaris Internals" and Joerg Schilling's OpenSolaris
> book (available in German language) and of course the according man
> pages and system headers.
>
> Before that I emulated being under the most-supported (Non-MacOS X)
> Firefox-UNIX platform: LinUX, by simply enabling all LinUX related
> preprocessor #ifdef directives:
>
> by putting:
>
> #elif defined(__sun__) || defined(__sun)
> #define OS_LINUX 1
>
> into ipc/chromium/src/build/build_config.h
>
> and such as by putting:
>
> elif CONFIG['OS_ARCH'] == 'SunOS':
> os_solaris = 1
>
> into ipc/chromium/moz.build.
>
> After a while I found out about and tracked all the problems and to
> get an estimate how much trouble is ahead continued with -k or touch.
> Many open tabs, vi sessions, log files, certain suffixes I added to
> renamed files etc were involved.
> Then I fixed all these invalid function calls. Then Did all the same
> for OpenBSD #ifdefs etc, in the hope this might shed some light into
> the flash-plugin under gcc built FF saga (the flash-plugin support on
> Illumos patched kernels is yet another topic, and affects all also
> Studio-compiled versions, I developed a workaround for OpenSXCE
> already in 2013 but never released it because Illumos thinks that I'm
> not valuable for their code base, so that was their idea, and that
> alone explains why I no longer wanted to publish the src, the rest
> depends on THEIR attitude, because my attitude has always been to play
> fair, but that was probably naive in the first place). I did that
> because I didn't test LinUX nor any BSD for a decade and always
> thought that those would not have flash-plugin problems with gcc (now
> after having isolated the problem, I'm not so sure anymore).
>
> Arg, well, by doing so I used to grep the entire src hierarchy several
> times per 10 minutes, for days and weeks, and from the third */*/* on
> exceeded even g-grep's variable arguments list limits (needed to split
> it down into */*/[a-m]* and */*/[n-z]* etc [into up to five regex
> groups per each of the more populated levels]).
> ... and found out that some #ifdefs were forgotten in the Sun-
> provided and NetBSD patches.
> But fixing them still didn't solve the main goal: What the hell caused
> flash-plugins from 9.x to 10. to crash with error A) and 11.x even
> with Error A) and B)???
> I almost grew into a senior citizen with gray hair from that (but
> fortunately I don't have much hair anymore since I started with
> OpenSolaris). It was like what one would have called a war, before
> history reminded us how REAL WAR outside looks and feels. So this all
> here is rather fun.
> All this was done on my poor Celeron G530 2.4GHz inside snv_130 (for
> optimum inter-distro compatibility) running in vbox, wait a pain -
> waaaaaiiiiiiiting ...
>
> But before indeed reaching the luxury of messing with a non-working
> flash first one needed to actually get FF built!!!!!
>
> But until I finally achieved this March ran through and then it was
> the end of April.
> All daily/nightly/weekly/weekendly fulltime pain and unpaid!
> To give an estimate: The FF git contains so many _file_ objects:
>
> $ find firefox-43.0b3|wc -l
> 138070
>
> If on average each file has 1000 lines, then:
>
> $ bc
> 138070*1000
> 138070000
>
> voila - 138 MILLION lines of code.
> And even if too many files are shorter that 1000 lines and if it
> "only" totals to 50 million lines: Happy reading!
> In the past one said oh - WOW, this or that OS has a few million(s)
> lines of code.
> Not so long ago ...
> Indeed extremely impressive what the mozilla.org team is achieving!!
> (But are they doing this for fun or in the frame of industry-paid
> jobs??)
>
> To shorten things up a bit, here the major difficulties and
> accordingly their references to related posts by others and followed
> by the solution or at least workaround:
>
> /* {0.}.libxul linkage
> {1.} plugin (kernel)
> {2.} plugin (ggc-FF)
> {3.} contextMenu
> {4.} langpacks */
>
> (problem/references/solution)
>
> {0.}.libxul linkage
> ###PROBLEM:
> Success was at first mostly prevented by the libxul linkage problem.
> Which was probably the major stopper for most of all those who tried
> to get newer FireFox 30++ versions at least somehow at all compiled
> and especially linked to the finish line.
> Because even with Studio - only the Sun-/Oracle diffs patched into
> FF24 were still easily buildable on snv_130 or also on Illumos (with
> include files derived from svc_147, which is pretty outdated compared
> to what Oracle meanwhile uses). Sun-Oracle's FF31 diffs patched into
> FF31 - no easy way. Not even with Studio 12.4. And for gcc Sun's
> patches are more a disadvantage than any help.
> Also: Of course only a gcc built FF is what I wanted ;) But one which
> can play the flash-plugin (at least version 10 [as I thought at the
> time]). Meanwhile all this is fixed and solved like a charme.
>
> ###REFERENCES:
> Oh, I see that one of the few others who ever really invested their
> own life and 1000's of hours of their private unpaid time to rescue
> OpenSolaris also ran into the same problem:
>
> [oi-dev] NS_COM_GLUE and upstream FF patches
> http://openindiana.org/pipermail/oi-dev/2013-December/002962.html
>
> This describes the problem you would end up with forever.
> Not exactly this, but variations of it.
> We are not dealing with some usual trivial missing symbols here, but
> with pointers to pointers to pointers to functions dealing with
> object, implemented and various files, but never there where the
> output would suggest, but in different files.
> While only one unresolved symbol would be shown per linker run, if it
> didn't exist, the next one would follow, and so on forever
> (literally).
> It is samething that can keep you busy for a year. Especially
> troublesome are x86 assembly files for the media codecs, such as in
>
> media/libvpx/vpx_config_x86-linux-gcc.asm
>
> It is a complex gcc visibility issue caused by the fact that Solaris
> still needs to default to its proprietary linker and even the very
> latest binutils still don't provide a fully functional gld or bfld or
> gold even though the problem was known for centuries.
>
>
> Googling brings:
> "gcc visibility error hidden firefox libxul.so"
> https://www.google.de/search?q=gcc+visibility+error+hidden+gcc+firefox+li&ie=utf-8&oe=utf-8&gws_rd=cr&ei=kclKVtiIMcH0asTCl9gF#q=gcc+visibility+error+hidden+gcc+firefox+libxul.so
>
>
> and only for "gcc visibility error hidden" (in both cases omit the
> double quotes, obviously) :
> https://www.google.de/search?q=gcc+visibility+error+hidden+gcc+firefox+li&ie=utf-8&oe=utf-8&gws_rd=cr&ei=kclKVtiIMcH0asTCl9gF
>
> You can add the most verbose output to CFLAGS and LDFLAGS, you can
> globally force brute-force visibility on or off in
> configure.in/configure or directly in:
>
> config/gcc_hidden.h
> /* This Source Code Form is subject to the terms of the Mozilla Public
> * License, v. 2.0. If a copy of the MPL was not distributed with this
> * file, You can obtain one at http://mozilla.org/MPL/2.0/. */
>
> /* Begin all files as hidden visibility */
> #pragma GCC visibility push(hidden)
>
> ... you can run a hundred manual compile or link commands: One ends
> staying hopeless for some while.
>
> The point is, that the Solaris ld does not fully understand gcc's
> visibility logic.
> HENCE the obvious solution would be to link FF with gnu-ld or bfld or
> gold, right?!
> YEP, and I see that I'm not the only one who had that idea. But as
> written before: gld does not properly function on Solaris.
> Which I find unbelievable. Maybe I will look myself into more details,
> as soon as time and $$$ial situation permits (if nobody helps, I must
> once again instantly stop all efforts and first job somewhere, so DO
> NOT COMPLAIN about unpublished src or delays, delays, deferrals!).
>
> LD_ALTEXEC=/usr/bin/gld would instruct /usr/bin/ld to forward all
> linking requests by the gcc/g++ frontent to gld, rather than linking
> the objects itself.
> But if you would attempt to set them globally your FF (or other larger
> project) build would not come all that far. The different command
> syntax for calling linking scripts is not the problem, although you
> might have to first deal with it.
>
> Wrong compiler-calling linker and passing args for calling mapfiles:
> -Wl,-M, .. results in:
>
> (here during a configure run)
>
> gld:/opt/csw/lib/gcc/i386-pc-solaris2.10/4.9.2/../../../libgcc-unwind.map:
> file format not recognized; treating as linker script
> gld:/opt/csw/lib/gcc/i386-pc-solaris2.10/4.9.2/../../../libgcc-unwind.map:1:
> syntax error
> configure: failed program was:
>
> #line 3724 "configure"
> #include "confdefs.h"
>
> int main(){return(0);}
> configure: error: installation or configuration problem: C++ compiler
> cannot create executables.
>
> Or you may get: gld: unrecognized option '-Bdirect'
> But all that's only cosmetics (wrong flags) and therefore no deal.
>
> Next possible issue, although it also appears to involve broken
> gcc/gld visibility:
>
> I'm in this case explicitly _not_ referring to:
>
> Subject: [Bug ld/14852] gnu ld can't use a library linked via solaris
> ld, if that library uses a library linked with GNU ld
>
> https://lists.gnu.org/archive/html/bug-binutils/2012-11/msg00049.html
>
> nor
>
> https://lists.gnu.org/archive/html/bug-binutils/2012-11/msg00050.html
>
> But the true problem is much worse and much older,
> a set of age old real bugs in binutils' gld:
>
> Timewarp 15 years back:
> Re: ld broken on Solaris 8
> https://sourceware.org/ml/binutils/2001-05/msg00173.html
> http://www.sourceware.org/ml/binutils/2000-04/msg00333.html
>
> But here we come already a bit closer (and look at the date!) :
>
> https://sourceware.org/bugzilla/show_bug.cgi?id=1021
>
> Matthias Kurz 2005-06-17 09:41:56 UTC
>
> Building mozilla under Solaris 10 fails. When g++ (3.4.4) tries to link
> regxpcom
> it bombs out with the following errors:
> mozilla/dist/bin/libnspr4.so: undefined reference to `fcntl at SUNW_0.9'
> mozilla/dist/bin/libnspr4.so: undefined reference to `dlsym at SUNW_0.7'
> mozilla/dist/bin/libnspr4.so: undefined reference to
> `pthread_join at SUNW_0.9'
> mozilla/dist/bin/libnspr4.so: undefined reference to `select at SUNW_1.2'
> mozilla/dist/bin/libnspr4.so: undefined reference to
> `rw_unlock at SUNW_0.9'
> mozilla/dist/bin/libnspr4.so: undefined reference to
> `pthread_attr_destroy at SUNW_
> 0.9'
> mozilla/dist/bin/libnspr4.so: undefined reference to `dlerror at SUNW_0.7'
> .. and so on
>
> I saw similiar reports from other users (google, see also
> http://sourceware.org/ml/binutils/2005-06/msg00466.html).
>
> I need to build with GNU binutils because i use OpenPKG and it is the
> default there.
>
>
> (mk)
>
>
>
> Whatever, this topic goes under in this long difficult to read ascii
> file.
> And it is worth being discussed in a dedicated discussion thread
> @binutils and/or illumos.
> To keep it short: The key to success was to link everything with
> Sun-ld, _except_ libxul.so
>
> And to implement it properly without user interaction this resulted in:
>
> +++ 001__PORTED/firefox-beta/toolkit/library/autoconflibxul.mk
> 2015-11-11 04:12:33.031527320 +0100
> @@ -0,0 +1,29 @@
> +#*
> +#* CDDL HEADER START
> +#*
> +#* The contents of this file are subject to the terms of the
> +#* Common Development and Distribution License, Version 1.0 only
> +#* (the "License"). You may not use this file except in compliance
> +#* with the License.
> +#*
> +#* You can obtain a copy of the license at usr/src/OPENSOLARIS.LICENSE
> +#* or http://www.opensolaris.org/os/licensing.
> +#* See the License for the specific language governing permissions
> +#* and limitations under the License.
> +#*
> +#* When distributing Covered Code, include this CDDL HEADER in each
> +#* file and include the License file at usr/src/OPENSOLARIS.LICENSE.
> +#* If applicable, add the following below this CDDL HEADER, with the
> +#* fields enclosed by brackets "[]" replaced with your own identifying
> +#* information: Portions Copyright [yyyy] [name of copyright owner]
> +#*
> +#* CDDL HEADER END
> +#*
> +#*
> +#* Copyright 2015 OpenSXCE.org Martin Bochnig<opensxce at mail.ru>
> +#* FireFox 20/30/40++ gcc4.x port with Flash support for OpenSolaris++
> x86/x64
> +#*
> +
> +//MKSHLIB = /usr/bin/gld $(DSO_PIC_CFLAGS) $(DSO_LDOPTS) -h
> $(DSO_SONAME) -o $@
> +//OS_LDFLAGS = -lpthread -z text -z ignore
> -R,'$$ORIGIN:$$ORIGIN/..' -z lazyload -z combreloc -z muldefs
> +MKSHLIB = /usr/bin/env LD_ALTEXEC=/usr/bin/gld
> PATH=/usr/gcc/4.5/bin:/usr/gcc/4.7/bin:/usr/gcc/4.8/bin:$PATH $(CXX)
> $(CXXFLAGS) $(DSO_PIC_CFLAGS) $(DSO_LDOPTS) -Wl,-h,$(DSO_SONAME) -o $@
>
>
> and
>
>
> diff -Nurb 000__VANILLA/firefox-beta/toolkit/library/Makefile.in
> 001__PORTED/firefox-beta/toolkit/library/Makefile.in
> --- 000__VANILLA/firefox-beta/toolkit/library/Makefile.in
> 2015-11-09 23:56:20.000000000 +0100
> +++ 001__PORTED/firefox-beta/toolkit/library/Makefile.in
> 2015-11-11 04:23:35.762542329 +0100
> @@ -1,8 +1,14 @@
> +#*
> +#* Copyright 2015 OpenSXCE.org Martin Bochnig<opensxce at mail.ru>
> +#* FireFox 20/30/40++ gcc4.x port with Flash support for OpenSolaris++
> x86/x64
> +#*
> +
> # This Source Code Form is subject to the terms of the Mozilla Public
> # License, v. 2.0. If a copy of the MPL was not distributed with this
> file,
> # You can obtain one at http://mozilla.org/MPL/2.0/.
>
> include $(topsrcdir)/toolkit/library/libxul.mk
> +include $(topsrcdir)/toolkit/library/autoconflibxul.mk
>
> include $(topsrcdir)/config/config.mk
>
>
>
> This locally overrides values imported by included and inter-included
> config files, although it took several hours before it was actually
> clear what is what and relevant versus not, example:
>
>
>
> $ grep -n MKSHLIB ../config/*
> ../config/config.mk:627:EXPAND_MKSHLIB_ARGS = --uselist
> ../config/config.mk:629:EXPAND_MKSHLIB_ARGS += --symbol-order
> $(SYMBOL_ORDER)
> ../config/config.mk:631:EXPAND_MKSHLIB = $(EXPAND_LIBS_EXEC)
> $(EXPAND_MKSHLIB_ARGS) -- $(MKSHLIB)
> ../config/rules.mk:162:ifdef MKSHLIB
> ../config/rules.mk:165:MKSHLIB = $(MKCSHLIB)
> ../config/rules.mk:170:endif # MKSHLIB
> ../config/rules.mk:445:MKSHLIB += -Wl,+eNSGetModule -Wl,+eerrno
> ../config/rules.mk:448:MKSHLIB += -Wl,+e_shlInit
> ../config/rules.mk:459:MKSHLIB +=
> -bE:$(MOZILLA_DIR)/build/unix/aix.exp -bnoexpall
> ../config/rules.mk:834: $(EXPAND_MKSHLIB) $(SHLIB_LDSTARTFILE) $(OBJS)
> $(SUB_SHLOBJS) $(DTRACE_PROBE_OBJ) $(MOZILLA_PROBE_LIBS) $(RESFILE)
> $(LDFLAGS) $(WRAP_LDFLAGS) $(STATIC_LIBS) $(SHARED_LIBS)
> $(EXTRA_DSO_LDOPTS) $(MOZ_GLUE_LDFLAGS) $(EXTRA_LIBS) $(OS_LIBS)
> $(SHLIB_LDENDFILE)
> ../config/rules.mk:837: $(EXPAND_MKSHLIB) $(SHLIB_LDSTARTFILE) $(OBJS)
> $(SUB_SHLOBJS) $(RESFILE) $(LDFLAGS) $(WRAP_LDFLAGS) $(STATIC_LIBS)
> $(SHARED_LIBS) $(EXTRA_DSO_LDOPTS) $(MOZ_GLUE_LDFLAGS) $(EXTRA_LIBS)
> $(OS_LIBS) $(SHLIB_LDENDFILE)
>
>
> {1.} plugin (kernel)
> Exactly like all the other goodies (initial version x86/x64 only,
> SPARC next week [if wanted], Illumos-x86/x64 kernel patch for
> flash-plugin support also next week if wanted]*, also DRM/KMS (now
> upgraded to Oracle's better version) for Intel can* be released any
> time, also my openXsun patches and a SPARC-OpenIndiana, also my
> studio2gcc userland wrapper enhanced cw [enabling me to build an
> almost vanilla Oracle-userland-Sol12.latest without having to correct
> all their studio-only CFLAGS, CXXFLAGS and LDFLAGS pollution every
> time {I say pollution not in disrespect of Studio 12.4, but because in
> the original design userland was compiler-clean, and only carelessness
> by Oracle employees caused all this}])
> ... this can or could be published anytime.
> BUT: Only if the "community" behaves. Otherwise I regret that I shared
> this FF43 patches and insides with you.
>
> {2.} plugin (ggc-FF)
> ###problem: As you all know (because it was reported several times on
> this list) ...:
>
> Flash plugin 9.x and 10.x would still function, but not 11.x
> (at different levels of modifications by default not even the older
> plugins worked on gcc4.[5-9] compiled firefox, I tested them all)
>
> Error with 9.x and 10.x:
> Not GTK2 toolkit (got 0).
>
>
> Error with 11.x:
> bash-4.0$ [NPAPI 21961] WARNING: pipe error: Broken pipe: file
> /export/home/martin/DEVEL/FF/43/FF43.0b2__RELEASE/TEST/PATCHCREATION/001__PORTED/RELEASE/firefox-43.0b3/ipc/chromium/src/chrome/common/ipc_channel_posix.cc,
> line 760
> [NPAPI 21961] ###!!! ABORT: Aborting on channel error.: file
> /export/home/martin/DEVEL/FF/43/FF43.0b2__RELEASE/TEST/PATCHCREATION/001__PORTED/RELEASE/firefox-43.0b3/ipc/glue/MessageChannel.cpp,
> line 1768
> [NPAPI 21961] ###!!! ABORT: Aborting on channel error.: file
> /export/home/martin/DEVEL/FF/43/FF43.0b2__RELEASE/TEST/PATCHCREATION/001__PORTED/RELEASE/firefox-43.0b3/ipc/glue/MessageChannel.cpp,
> line 1768
>
>
> ###references: After all the weeks it took me to find this bug it is
> now somewhat frustrating so find out, that another one already found
> out the very same and even reported about it at mozilla - at first in
> 2011!!! (while firefox upstream nevertheless still continues to keep
> the unfixed version in its dev git).
> It appears to be the same guy who ported the chrome browser to Solaris
> and got only ass-kicks from certain (NON-CONTRIBUTING) folks on this
> openindiana list.
> Now - as you see he did in fact publish sources. Had those
> flamewar-lovers reacted in a somewhat more civilised way, maybe he
> would have helped you with the FF port.
>
> GCC 3 ABI mask breaks Flash 10.3 compatibility on Solaris for
> Chromium compiled with gcc 4 #3
> Open
>
> https://github.com/mozilla/npapi-sdk/issues/3
>
> ###solution:
> third_party/npapi/bindings/npapi.h
> @@ -334,7 +334,7
> -#if (defined(XP_UNIX)&& defined(__GNUC__)&& (__GNUC__>= 3))
> +#if (defined(XP_UNIX)&& defined(__GNUC__)&& (__GNUC__ == 3))
>
> As I found out myself, a single byte (== instead of>=) was causing all
> this trouble on Solaris kernels.
> What's new after seeing this post: Aha! Only Solaris may be affected,
> not LiunUX nor BSD nor MacOSX (?)
>
> {3.} contextMenu
> ###problem: At all Beta src releases after FF41.0b9, hence from
> FF42.0b1 on, suddenly on the canvas the main context-menu would be
> extremely extremely long (several screens) and show all sorts of ever
> envisioned entries, while no single of them (except Select All) would
> actually invoke any functionality.
> The entries would show all crap - including once planned but never
> implemented functions - and would show always the same long menu
> independently of the context (no matter on which object one would
> place the mourse cursor).
>
> ###references to similar problems:
> Screenshot:
> http://forums.thedailywtf.com/forums/p/27789/324362.aspx
>
> Search:
> https://www.google.de/search?q=context+menu+long+firefox&ie=utf-8&oe=utf-8&gws_rd=cr&ei=XJBKVp-LLsS4UfnZi_AP
>
> What's about the suddenly extremely long right mouse-click menu?
> https://support.mozilla.org/en-US/questions/956675
>
> very long right-click context menu is present
> https://support.mozilla.org/en-US/questions/958458
>
> Right click menu is huge, how do I get the old menu back so I can
> highlight a word and then search for it?
> https://support.mozilla.org/en-US/questions/949976
>
> ###solution: It was one such thing which made me really upset.
> Analysing the long list of git commits I knew it would be related to
> WebGLFormat, whatever I merged in or backed out, all the things that
> appeared to make sense did not work.
> By patching and Solaris-building all 41.0b? and 42.0b? builds in a
> manner of reducing the intervall of the last known working versus
> first known unknown version I could track down, that the break appears
> from 41.0b9 to 42.0b1. But the gdiff -Nurb of those 2 resulted in
> thousands and thousands of lines :(
>
> --->> It was clear that the only solution was building a debug
> version of some post-41.b9 with all syms.
> But even then I had made the mistake to only look after the 2 error
> messages created when instantiating the contextMenu by right-clicking
> onto the canvas.
>
> And those were as you may observe in:
>
>
> $ PATH=dist/lib:dist/bin:$PATH LD_LIBRARY_PATH_32=dist/lib:dist/bin
> /usr/lib/ld.so.1 dist/bin/firefox
> File descriptors set to 512
> ++DOCSHELL f045e800 == 1 [pid = 19010] [id = 1]
> ++DOMWINDOW == 1 (f045f000) [pid = 19010] [serial = 1] [outer = 0]
> [19010] WARNING: Hardware Vsync support not yet implemented. Falling
> back to software timers: file
> /export/home/martin/DEVEL/FF/43/FF43.0b2__RELEASE/TEST/PATCHCREATION/001__PORTED/firefox-43.0b2/gfx/thebes/gfxPlatform.cpp,
> line 2404
> ++DOMWINDOW == 2 (f0463000) [pid = 19010] [serial = 2] [outer =
> f045f000]
> ++DOCSHELL eb276800 == 2 [pid = 19010] [id = 2]
> ++DOMWINDOW == 3 (eb27d800) [pid = 19010] [serial = 3] [outer = 0]
> ++DOMWINDOW == 4 (eb27e000) [pid = 19010] [serial = 4] [outer =
> eb27d800]
> ++DOMWINDOW == 5 (eb2d7400) [pid = 19010] [serial = 5] [outer =
> f045f000]
> JavaScript error: resource://gre/modules/LoginManagerContextMenu.jsm,
> line 20: ReferenceError: Intl is not defined
> ++DOCSHELL ea7f6400 == 3 [pid = 19010] [id = 3]
> ++DOMWINDOW == 6 (ea7f6800) [pid = 19010] [serial = 6] [outer = 0]
> ++DOCSHELL ea7f6c00 == 4 [pid = 19010] [id = 4]
> ++DOMWINDOW == 7 (eb248f60) [pid = 19010] [serial = 7] [outer = 0]
> [19010] WARNING: NS_ENSURE_SUCCESS(rv, rv) failed with result
> 0x80040111: file
> /export/home/martin/DEVEL/FF/43/FF43.0b2__RELEASE/TEST/PATCHCREATION/001__PORTED/firefox-43.0b2/dom/base/nsFrameLoader.cpp,
> line 268
> ++DOCSHELL e9d8ec00 == 5 [pid = 19010] [id = 5]
> ++DOMWINDOW == 8 (eb24a5e0) [pid = 19010] [serial = 8] [outer = 0]
> [19010] WARNING: NS_ENSURE_SUCCESS(rv, rv) failed with result
> 0x80040111: file
> /export/home/martin/DEVEL/FF/43/FF43.0b2__RELEASE/TEST/PATCHCREATION/001__PORTED/firefox-43.0b2/dom/base/nsFrameLoader.cpp,
> line 268
> ++DOMWINDOW == 9 (eb24ad60) [pid = 19010] [serial = 9] [outer =
> eb24a5e0]
> [19010] WARNING: NS_ENSURE_TRUE(globalObject&&
> globalObject->GetGlobalJSObject()) failed: file
> /export/home/martin/DEVEL/FF/43/FF43.0b2__RELEASE/TEST/PATCHCREATION/001__PORTED/firefox-43.0b2/dom/base/nsDocument.cpp,
> line 8326
> [19010] WARNING: NS_ENSURE_TRUE(globalObject&&
> globalObject->GetGlobalJSObject()) failed: file
> /export/home/martin/DEVEL/FF/43/FF43.0b2__RELEASE/TEST/PATCHCREATION/001__PORTED/firefox-43.0b2/dom/base/nsDocument.cpp,
> line 8326
> ++DOMWINDOW == 10 (e9585800) [pid = 19010] [serial = 10] [outer =
> ea7f6800]
> ++DOMWINDOW == 11 (eb24c980) [pid = 19010] [serial = 11] [outer =
> eb248f60]
> ++DOMWINDOW == 12 (eb24cd40) [pid = 19010] [serial = 12] [outer =
> eb24a5e0]
> ++DOMWINDOW == 13 (e830f520) [pid = 19010] [serial = 13] [outer =
> eb24a5e0]
> [19010] WARNING: attempt to modify an immutable nsStandardURL: file
> /export/home/martin/DEVEL/FF/43/FF43.0b2__RELEASE/TEST/PATCHCREATION/001__PORTED/firefox-43.0b2/netwerk/base/nsStandardURL.cpp,
> line 1300
> [19010] WARNING: NS_ENSURE_SUCCESS(rv, rv) failed with result
> 0x80070057: file
> /export/home/martin/DEVEL/FF/43/FF43.0b2__RELEASE/TEST/PATCHCREATION/001__PORTED/firefox-43.0b2/dom/quota/QuotaManager.cpp,
> line 3437
> ++DOCSHELL e5fba400 == 6 [pid = 19010] [id = 6]
> ++DOMWINDOW == 14 (e5fba800) [pid = 19010] [serial = 14] [outer = 0]
> ++DOMWINDOW == 15 (e5fbb000) [pid = 19010] [serial = 15] [outer =
> e5fba800]
> JavaScript error: chrome://browser/content/nsContextMenu.js, line 21:
> TypeError: this.initMenu is not a function
> [19010] ###!!! ASSERTION: unable to use nl_langinfo(CODESET): 'Error',
> file
> /export/home/martin/DEVEL/FF/43/FF43.0b2__RELEASE/TEST/PATCHCREATION/001__PORTED/firefox-43.0b2/intl/locale/unix/nsUNIXCharset.cpp,
> line 142
> [19010] WARNING: GetDefaultCharsetForLocale: need to add multi locale
> support: file
> /export/home/martin/DEVEL/FF/43/FF43.0b2__RELEASE/TEST/PATCHCREATION/001__PORTED/firefox-43.0b2/intl/locale/unix/nsUNIXCharset.cpp,
> line 101
> ++DOCSHELL e3bb6000 == 7 [pid = 19010] [id = 7]
> ++DOMWINDOW == 16 (e3bb6400) [pid = 19010] [serial = 16] [outer = 0]
> ++DOMWINDOW == 17 (e3bb7000) [pid = 19010] [serial = 17] [outer =
> e3bb6400]
> ++DOMWINDOW == 18 (e3bb9400) [pid = 19010] [serial = 18] [outer =
> e3bb6400]
> ++DOCSHELL e3bba000 == 8 [pid = 19010] [id = 8]
> ++DOMWINDOW == 19 (e4fb3f20) [pid = 19010] [serial = 19] [outer = 0]
> ++DOMWINDOW == 20 (e4fb46a0) [pid = 19010] [serial = 20] [outer =
> e4fb3f20]
> [19010] WARNING: Could not get disk status from nsIDiskSpaceWatcher:
> file
> /export/home/martin/DEVEL/FF/43/FF43.0b2__RELEASE/TEST/PATCHCREATION/001__PORTED/firefox-43.0b2/uriloader/prefetch/nsOfflineCacheUpdateService.cpp,
> line 315
> [19010] WARNING: 'err', file
> /export/home/martin/DEVEL/FF/43/FF43.0b2__RELEASE/TEST/PATCHCREATION/001__PORTED/firefox-43.0b2/netwerk/base/nsSocketTransport2.cpp,
> line 3008
> [19010] WARNING: nsSocketTransport Failed setting TCP_KEEPIDLE
> [e3bd6480] OS error[0x63] Option not supported by protocol: file
> /export/home/martin/DEVEL/FF/43/FF43.0b2__RELEASE/TEST/PATCHCREATION/001__PORTED/firefox-43.0b2/netwerk/base/nsSocketTransport2.cpp,
> line 2932
> [19010] WARNING: 'NS_FAILED(rv)', file
> /export/home/martin/DEVEL/FF/43/FF43.0b2__RELEASE/TEST/PATCHCREATION/001__PORTED/firefox-43.0b2/netwerk/base/nsSocketTransport2.cpp,
> line 2784
> [19010] WARNING: 'err', file
> /export/home/martin/DEVEL/FF/43/FF43.0b2__RELEASE/TEST/PATCHCREATION/001__PORTED/firefox-43.0b2/netwerk/base/nsSocketTransport2.cpp,
> line 3008
> [19010] WARNING: nsSocketTransport Failed setting TCP_KEEPIDLE
> [e3bd6600] OS error[0x63] Option not supported by protocol: file
> /export/home/martin/DEVEL/FF/43/FF43.0b2__RELEASE/TEST/PATCHCREATION/001__PORTED/firefox-43.0b2/netwerk/base/nsSocketTransport2.cpp,
> line 2932
> [19010] WARNING: 'NS_FAILED(rv)', file
> /export/home/martin/DEVEL/FF/43/FF43.0b2__RELEASE/TEST/PATCHCREATION/001__PORTED/firefox-43.0b2/netwerk/base/nsSocketTransport2.cpp,
> line 2784
> [19010] WARNING: 'err', file
> /export/home/martin/DEVEL/FF/43/FF43.0b2__RELEASE/TEST/PATCHCREATION/001__PORTED/firefox-43.0b2/netwerk/base/nsSocketTransport2.cpp,
> line 3008
> [19010] WARNING: nsSocketTransport Failed setting TCP_KEEPIDLE
> [e3bd6480] OS error[0x63] Option not supported by protocol: file
> /export/home/martin/DEVEL/FF/43/FF43.0b2__RELEASE/TEST/PATCHCREATION/001__PORTED/firefox-43.0b2/netwerk/base/nsSocketTransport2.cpp,
> line 2932
> [19010] WARNING: 'NS_FAILED(rv)', file
> /export/home/martin/DEVEL/FF/43/FF43.0b2__RELEASE/TEST/PATCHCREATION/001__PORTED/firefox-43.0b2/netwerk/base/nsSocketTransport2.cpp,
> line 2634
> [19010] WARNING: 'NS_FAILED(rv)', file
> /export/home/martin/DEVEL/FF/43/FF43.0b2__RELEASE/TEST/PATCHCREATION/001__PORTED/firefox-43.0b2/netwerk/base/nsSocketTransport2.cpp,
> line 2721
> ++DOMWINDOW == 21 (e4fb4c40) [pid = 19010] [serial = 21] [outer =
> e4fb3f20]
> [19010] WARNING: 'err', file
> /export/home/martin/DEVEL/FF/43/FF43.0b2__RELEASE/TEST/PATCHCREATION/001__PORTED/firefox-43.0b2/netwerk/base/nsSocketTransport2.cpp,
> line 3008
> [19010] WARNING: nsSocketTransport Failed setting TCP_KEEPIDLE
> [e3bd6a80] OS error[0x63] Option not supported by protocol: file
> /export/home/martin/DEVEL/FF/43/FF43.0b2__RELEASE/TEST/PATCHCREATION/001__PORTED/firefox-43.0b2/netwerk/base/nsSocketTransport2.cpp,
> line 2932
> [19010] WARNING: 'NS_FAILED(rv)', file
> /export/home/martin/DEVEL/FF/43/FF43.0b2__RELEASE/TEST/PATCHCREATION/001__PORTED/firefox-43.0b2/netwerk/base/nsSocketTransport2.cpp,
> line 2784
> [19010] WARNING: 'err', file
> /export/home/martin/DEVEL/FF/43/FF43.0b2__RELEASE/TEST/PATCHCREATION/001__PORTED/firefox-43.0b2/netwerk/base/nsSocketTransport2.cpp,
> line 3008
> [19010] WARNING: nsSocketTransport Failed setting TCP_KEEPIDLE
> [e3bd6f00] OS error[0x63] Option not supported by protocol: file
> /export/home/martin/DEVEL/FF/43/FF43.0b2__RELEASE/TEST/PATCHCREATION/001__PORTED/firefox-43.0b2/netwerk/base/nsSocketTransport2.cpp,
> line 2932
> [19010] WARNING: 'NS_FAILED(rv)', file
> /export/home/martin/DEVEL/FF/43/FF43.0b2__RELEASE/TEST/PATCHCREATION/001__PORTED/firefox-43.0b2/netwerk/base/nsSocketTransport2.cpp,
> line 2784
> [19010] WARNING: 'err', file
> /export/home/martin/DEVEL/FF/43/FF43.0b2__RELEASE/TEST/PATCHCREATION/001__PORTED/firefox-43.0b2/netwerk/base/nsSocketTransport2.cpp,
> line 3008
> [19010] WARNING: nsSocketTransport Failed setting TCP_KEEPIDLE
> [e3bd7380] OS error[0x63] Option not supported by protocol: file
> /export/home/martin/DEVEL/FF/43/FF43.0b2__RELEASE/TEST/PATCHCREATION/001__PORTED/firefox-43.0b2/netwerk/base/nsSocketTransport2.cpp,
> line 2932
> [19010] WARNING: 'NS_FAILED(rv)', file
> /export/home/martin/DEVEL/FF/43/FF43.0b2__RELEASE/TEST/PATCHCREATION/001__PORTED/firefox-43.0b2/netwerk/base/nsSocketTransport2.cpp,
> line 2784
> JavaScript error: chrome://browser/content/browser.xul, line 1:
> TypeError: gContextMenu is null
> JavaScript error: chrome://browser/content/browser.xul, line 1:
> TypeError: gContextMenu is null
> --DOCSHELL e5fba400 == 7 [pid = 19010] [id = 6]
> JavaScript error: chrome://browser/content/nsContextMenu.js, line 21:
> TypeError: this.initMenu is not a function
> [19010] WARNING: 'err', file
> /export/home/martin/DEVEL/FF/43/FF43.0b2__RELEASE/TEST/PATCHCREATION/001__PORTED/firefox-43.0b2/netwerk/base/nsSocketTransport2.cpp,
> line 3008
> [19010] WARNING: nsSocketTransport Failed setting TCP_KEEPIDLE
> [e95a3e80] OS error[0x63] Option not supported by protocol: file
> /export/home/martin/DEVEL/FF/43/FF43.0b2__RELEASE/TEST/PATCHCREATION/001__PORTED/firefox-43.0b2/netwerk/base/nsSocketTransport2.cpp,
> line 2932
> [19010] WARNING: 'NS_FAILED(rv)', file
> /export/home/martin/DEVEL/FF/43/FF43.0b2__RELEASE/TEST/PATCHCREATION/001__PORTED/firefox-43.0b2/netwerk/base/nsSocketTransport2.cpp,
> line 2784
> --DOMWINDOW == 20 (e3bb7000) [pid = 19010] [serial = 17] [outer =
> e3bb6400] [url = about:blank]
> --DOMWINDOW == 19 (e4fb46a0) [pid = 19010] [serial = 20] [outer =
> e4fb3f20] [url = about:blank]
> --DOMWINDOW == 18 (eb24ad60) [pid = 19010] [serial = 9] [outer =
> eb24a5e0] [url = about:blank]
> --DOMWINDOW == 17 (f0463000) [pid = 19010] [serial = 2] [outer =
> f045f000] [url = about:blank]
> [19010] WARNING: 'err', file
> /export/home/martin/DEVEL/FF/43/FF43.0b2__RELEASE/TEST/PATCHCREATION/001__PORTED/firefox-43.0b2/netwerk/base/nsSocketTransport2.cpp,
> line 3008
> [19010] WARNING: nsSocketTransport Failed setting TCP_KEEPIDLE
> [e3bd6600] OS error[0x63] Option not supported by protocol: file
> /export/home/martin/DEVEL/FF/43/FF43.0b2__RELEASE/TEST/PATCHCREATION/001__PORTED/firefox-43.0b2/netwerk/base/nsSocketTransport2.cpp,
> line 2932
> [19010] WARNING: 'NS_FAILED(rv)', file
> /export/home/martin/DEVEL/FF/43/FF43.0b2__RELEASE/TEST/PATCHCREATION/001__PORTED/firefox-43.0b2/netwerk/base/nsSocketTransport2.cpp,
> line 2634
> [19010] WARNING: 'NS_FAILED(rv)', file
> /export/home/martin/DEVEL/FF/43/FF43.0b2__RELEASE/TEST/PATCHCREATION/001__PORTED/firefox-43.0b2/netwerk/base/nsSocketTransport2.cpp,
> line 2721
> JavaScript error: chrome://browser/content/browser.xul, line 1:
> TypeError: gContextMenu is null
> JavaScript error: chrome://browser/content/browser.xul, line 1:
> TypeError: gContextMenu is null
> --DOMWINDOW == 16 (eb24cd40) [pid = 19010] [serial = 12] [outer = 0]
> [url = about:blank]
> ++DOMWINDOW == 17 (e9d6e980) [pid = 19010] [serial = 22] [outer =
> eb24a5e0]
> [19010] WARNING: 'err', file
> /export/home/martin/DEVEL/FF/43/FF43.0b2__RELEASE/TEST/PATCHCREATION/001__PORTED/firefox-43.0b2/netwerk/base/nsSocketTransport2.cpp,
> line 3008
> [19010] WARNING: nsSocketTransport Failed setting TCP_KEEPIDLE
> [f0673300] OS error[0x63] Option not supported by protocol: file
> /export/home/martin/DEVEL/FF/43/FF43.0b2__RELEASE/TEST/PATCHCREATION/001__PORTED/firefox-43.0b2/netwerk/base/nsSocketTransport2.cpp,
> line 2932
> [19010] WARNING: 'NS_FAILED(rv)', file
> /export/home/martin/DEVEL/FF/43/FF43.0b2__RELEASE/TEST/PATCHCREATION/001__PORTED/firefox-43.0b2/netwerk/base/nsSocketTransport2.cpp,
> line 2784
> [19010] WARNING: RasterImage::Init failed: file
> /export/home/martin/DEVEL/FF/43/FF43.0b2__RELEASE/TEST/PATCHCREATION/001__PORTED/firefox-43.0b2/image/ImageFactory.cpp,
> line 109
> JavaScript error: chrome://browser/content/nsContextMenu.js, line 21:
> TypeError: this.initMenu is not a function
> --DOMWINDOW == 16 (e5fbb000) [pid = 19010] [serial = 15] [outer =
> e5fba800] [url = about:blank]
> --DOMWINDOW == 15 (e5fba800) [pid = 19010] [serial = 14] [outer = 0]
> [url = about:srcdoc]
> JavaScript error: chrome://browser/content/browser.xul, line 1:
> TypeError: gContextMenu is null
> JavaScript error: chrome://browser/content/browser.xul, line 1:
> TypeError: gContextMenu is null
> JavaScript error: chrome://browser/content/nsContextMenu.js, line 21:
> TypeError: this.initMenu is not a function
> JavaScript error: chrome://browser/content/browser.xul, line 1:
> TypeError: gContextMenu is null
> JavaScript error: chrome://browser/content/browser.xul, line 1:
> TypeError: gContextMenu is null
> --DOCSHELL e3bb6000 == 6 [pid = 19010] [id = 7]
> --DOCSHELL f045e800 == 5 [pid = 19010] [id = 1]
> [19010] WARNING: cannot post event if not initialized: file
> /export/home/martin/DEVEL/FF/43/FF43.0b2__RELEASE/TEST/PATCHCREATION/001__PORTED/firefox-43.0b2/netwerk/protocol/http/nsHttpConnectionMgr.cpp,
> line 194
> [19010] WARNING: cannot post event if not initialized: file
> /export/home/martin/DEVEL/FF/43/FF43.0b2__RELEASE/TEST/PATCHCREATION/001__PORTED/firefox-43.0b2/netwerk/protocol/http/nsHttpConnectionMgr.cpp,
> line 194
> [19010] WARNING: cannot post event if not initialized: file
> /export/home/martin/DEVEL/FF/43/FF43.0b2__RELEASE/TEST/PATCHCREATION/001__PORTED/firefox-43.0b2/netwerk/protocol/http/nsHttpConnectionMgr.cpp,
> line 194
> [19010] WARNING: cannot post event if not initialized: file
> /export/home/martin/DEVEL/FF/43/FF43.0b2__RELEASE/TEST/PATCHCREATION/001__PORTED/firefox-43.0b2/netwerk/protocol/http/nsHttpConnectionMgr.cpp,
> line 194
> --DOCSHELL e3bba000 == 4 [pid = 19010] [id = 8]
> --DOCSHELL eb276800 == 3 [pid = 19010] [id = 2]
> --DOCSHELL e9d8ec00 == 2 [pid = 19010] [id = 5]
> --DOCSHELL ea7f6400 == 1 [pid = 19010] [id = 3]
> --DOCSHELL ea7f6c00 == 0 [pid = 19010] [id = 4]
> [19010] WARNING: NS_ENSURE_TRUE(context) failed: file
> /export/home/martin/DEVEL/FF/43/FF43.0b2__RELEASE/TEST/PATCHCREATION/001__PORTED/firefox-43.0b2/xpcom/threads/nsThread.cpp,
> line 761
> --DOMWINDOW == 14 (eb2d7400) [pid = 19010] [serial = 5] [outer =
> f045f000] [url = resource://gre-resources/hiddenWindow.html]
> --DOMWINDOW == 13 (e9585800) [pid = 19010] [serial = 10] [outer =
> ea7f6800] [url = about:blank]
> --DOMWINDOW == 12 (eb24c980) [pid = 19010] [serial = 11] [outer =
> eb248f60] [url = about:blank]
> --DOMWINDOW == 11 (f045f000) [pid = 19010] [serial = 1] [outer = 0]
> [url = resource://gre-resources/hiddenWindow.html]
> --DOMWINDOW == 10 (ea7f6800) [pid = 19010] [serial = 6] [outer = 0]
> [url = about:blank]
> --DOMWINDOW == 9 (eb248f60) [pid = 19010] [serial = 7] [outer = 0]
> [url = about:blank]
> Hit MOZ_CRASH(Shutdown too long, probably frozen, causing a crash.) at
> /export/home/martin/DEVEL/FF/43/FF43.0b2__RELEASE/TEST/PATCHCREATION/001__PORTED/firefox-43.0b2/toolkit/components/terminator/nsTerminator.cpp:151
> /usr/bin/pstack: 19010: No such file or directory
> --DOMWINDOW == 8 (eb27e000) [pid = 19010] [serial = 4] [outer = 0]
> [url = about:blank]
> --DOMWINDOW == 7 (eb27d800) [pid = 19010] [serial = 3] [outer = 0]
> [url = chrome://browser/content/browser.xul]
> --DOMWINDOW == 6 (eb24a5e0) [pid = 19010] [serial = 8] [outer = 0]
> [url = http://opensxce.org/]
> --DOMWINDOW == 5 (e830f520) [pid = 19010] [serial = 13] [outer = 0]
> [url = about:home]
> --DOMWINDOW == 4 (e4fb3f20) [pid = 19010] [serial = 19] [outer = 0]
> [url = https://self-repair.mozilla.org/en-US/repair/]
> --DOMWINDOW == 3 (e3bb9400) [pid = 19010] [serial = 18] [outer = 0]
> [url =
> data:application/vnd.mozilla.xul+xml;charset=utf-8,<window%20id='win'/>]
> --DOMWINDOW == 2 (e3bb6400) [pid = 19010] [serial = 16] [outer = 0]
> [url =
> data:application/vnd.mozilla.xul+xml;charset=utf-8,<window%20id='win'/>]
> --DOMWINDOW == 1 (e9d6e980) [pid = 19010] [serial = 22] [outer = 0]
> [url = http://opensxce.org/]
> --DOMWINDOW == 0 (e4fb4c40) [pid = 19010] [serial = 21] [outer = 0]
> [url = https://self-repair.mozilla.org/en-US/repair/]
> nsStringStats
> => mAllocCount: 99212
> => mReallocCount: 11728
> => mFreeCount: 99212
> => mShareCount: 180933
> => mAdoptCount: 5780
> => mAdoptFreeCount: 5780
> => Process ID: 19010, Thread ID: 1
>
>
> ... thse were:
>
> JavaScript error: chrome://browser/content/browser.xul, line 1:
> TypeError: gContextMenu is null
>
> and
>
> JavaScript error: chrome://browser/content/nsContextMenu.js, line 21:
> TypeError: this.initMenu is not a function
>
>
> This again lead me to some similar bug scenarios, but not identical
> ones:
>
> https://www.google.de/search?q=%22TypeError:+gContextMenu+is+null%22&ie=utf-8&oe=utf-8&gws_rd=cr&ei=SJVKVo6MEMf_ac26sNAN
>
> TypeError: gContextMenu is null
> http://www.camp-firefox.de/forum/viewtopic.php?f=8&t=75369&start=0
>
> Bug 978082 - Intermittent |
> browser_save_video.js,browser_save_video_frame.js | uncaught exception
> - TypeError: gContextMenu is null at
> chrome://browser/content/browser.xul:1
> https://bugzilla.mozilla.org/show_bug.cgi?id=978082
>
> Bug 913176 - Intermittent browser_default_image_filename.js | uncaught
> exception - TypeError: gContextMenu is null at
> chrome://browser/content/browser.xul:1 | Test timed out
> https://bugzilla.mozilla.org/show_bug.cgi?id=913176
>
> Target Undefined When Right Clicking In Firefox With Context Menu
> http://stackoverflow.com/questions/14651785/target-undefined-when-right-clicking-in-firefox-with-context-menu
>
>
> ... and for the second error message:
>
> No single identical result for this (even shortened) string if in
> double quotes:
>
> "TypeError: this.initMenu is not a function"
>
> No safe node or rm -R ~/.mozilla helped, so no extensions/add-ons
> could have been causing this.
> Then I spent days reading, compiling, grep'ing, checking firefox's git
> log etc etc etc.
> After a week I finally realized that with JS it is a mistake to only
> look for the final error messages. Because if something goes wrong
> during JS initialization some or all functionality might be broken
> until the hosting application gets restarted.
> BANG!!!
>
> So let's have a look at the full stdout and stderr from the console
> window running the debug version:
>
> First interesting error:
> JavaScript error: resource://gre/modules/LoginManagerContextMenu.jsm,
> line 20: ReferenceError: Intl is not defined
>
> And at that moment it became all clear to me:
>
> As silly as life can be, of course I had simply disabled
> internationalization support last spring, because I personally always
> prefer the C locale anyway, and also had enough other problems with
> first getting the port itself done, at the time.
> And until now this has never been a problem and worked fine until
> FF41.0b9
>
> Configuring with ../configure --enable-application=browser
> --prefix=/usr/lib/firefox`/bin/cat /tmp/FFvers`
> --enable-official-branding --disable-updater --enable-debug-symbols=no
> --enable-update-channel=esr --disable-tests --enable-jemalloc
> --disable-dtrace --disable-xinerama --disable-crashreporter
> --disable-pulseaudio --disable-gstreamer --with-intl-api
> --disable-debug --enable-ipc --enable-default-toolkit=cairo-gtk2
>
> soon brought the relieving confirmation, that indeed and in fact only
> this had been to problem from 42.0b1++ on.
> NICE, how stupid. Okay and now that internationalization was on and
> fully building like a charm of course I wanted to give you also the
> language kits in preinstalled ready2run form in the binary package and
> althernative bins-tarball.
>
> ... which instantly brought up the next day of questions, finding odd
> user-unfriendliness in FF's handling of this, namely:
>
>
> {4.} langpacks */
> How to get the langpacks not only somehow to function on by one
> forcing end-users to perform manual installations one at a time
> following bogus outdated and often self-contradicting instructions
> somewhere on the mozilla site, often referring to oldtimers such as
> FF3 or 4.
>
> No, my goal was:
>
> ###problem:
> A) * having all available 90++ langpacks installed
>
> B) * having them installed GLOBALLY, not even Oracle does this in
> Sol11.3 as I had to find out, they use jack's local ~/.mozilla, but I
> offer not a distro this time, but only installable software in a pkg
> and tar ball, cannot mess with anybody's $HOME
>
> C) * I want the entire FF-UI to be in the corresponding language
> (rather than only in the objects dealing with the filesystem [open
> file/save as etc.]), where-ever this is fully imlemented in a language
> pack (not all have that)
>
> D) * All plugins shall not get disabled by default on first run (if no
> ~/.mozilla exists, or if you already had that but run the new FF43 for
> the first time)
>
> E) of course I want full automation:
> It shall instantly come up in the locale that the user is using and
> have its entire UI in the associated language.
> No matter what the user specified during installation system-wide or
> locally for one session from gdm, it shall always work smoothly and
> without interaction. Without first forcing the user to:
> i) install any langpack
> ii) having to enable it
> ii) having to set a single variable in about:config
>
> In in extreme case, from the same env and cmd line one shall be able to
> say:
> LANG=de_DE.UTF-8 /usr/lib/firefox43.0b3/lib/firefox-43.0/firefox
> and it should come up like a genuine german version
>
> while seconds later from the same window without changing the env:
>
> LANG=ru_RU.UTF-8 /usr/lib/firefox43.0b3/lib/firefox-43.0/firefox
> should bring up a fully localized Russian version and all the same for
> all other 90 locales (if and where UI-localization has full support)
> including zh_CN or Japanese.
>
> I was surprised to learn, that in 40++ versions of FF this is no
> longer easy to be accomplished (as global installation that has zero
> dependencies to whatever in ~/.mozilla). At least not unless you found
> out how to do this.
> I fund out by testing that Sun/Oracle Solaris 11.3 can does so only by
> means of jack's $HOME, while Hipster with its FF24 does have the
> langpacks all in place in the global extensions dir, but no single
> langpack appeared to be installed (regsitered) let alone activated not
> to mention activated by intl.locale.matchOS=true or
> general.useragent.locale
>
> ###REFERENCES:
> Integrating Add-ons into Firefox
> https://mike.kaply.com/2012/02/09/integrating-add-ons-into-firefox/
>
> distribution/bundles Directory Gone in Firefox 40
> https://mike.kaply.com/2015/05/19/distributionbundles-directory-gone-in-firefox-40/
>
> Deploying Firefox in an enterprise environment
> https://developer.mozilla.org/en-US/Firefox/Enterprise_deployment
>
> AddOns for all users
> https://support.mozilla.org/en-US/questions/966922
>
> Extension Packaging
> https://developer.mozilla.org/en-US/Add-ons/Extension_Packaging
>
> Works for Seamonkey, but not for FF: Firefox/Seamonkey: How to install
> add-ons system-wide?
> http://www.linuxquestions.org/questions/linux-software-2/firefox-seamonkey-how-to-install-add-ons-system-wide-4175421566/
>
> How do I install a language XPI globally and make it the default
> language?
> http://www.cla5h.com/how-do-i-install-a-language-xpi-globally-and-make-it-the-default-language.html
>
> Well, all of that did not work, unzipping it in place under
> /usr/lib/firefox43.0b3/lib/firefox-43.0/browser/extensions/ also made
> no difference over not unzipping them.
> Let me shorten it and here comes the ...
>
> ###SOLUTION:
> ---
> 000__VANILLA/toolkit/mozapps/extensions/internal/XPIProviderUtils.js 2015-11-12
> 23:49:00.000000000 +0100
> +++
> 001__PORTED/toolkit/mozapps/extensions/internal/XPIProviderUtils.js 2015-11-16
> 01:54:30.570857314 +0100
> @@ -1675,12 +1675,12 @@
>
> if (isDetectedInstall&& aNewAddon.foreignInstall) {
> // If the add-on is a foreign install and is in a scope where
> add-ons
> - // that were dropped in should default to disabled then disable
> it
> + // that were dropped in should default to disabled then enable
> it anyway on OpenSolaris (preinstalled langpacks)
> let disablingScopes =
> Preferences.get(PREF_EM_AUTO_DISABLED_SCOPES, 0);
> if (aInstallLocation.scope& disablingScopes) {
> - logger.warn("Disabling foreign installed add-on " +
> aNewAddon.id + " in "
> + logger.warn("Enabling foreign installed add-on " +
> aNewAddon.id + " in "
> + aInstallLocation.name);
> - aNewAddon.userDisabled = true;
> + aNewAddon.userDisabled = false;
> }
> }
>
> --- 000__VANILLA/modules/libpref/init/all.js 2015-11-12
> 23:48:53.000000000 +0100
> +++ 001__PORTED/modules/libpref/init/all.js 2015-11-16
> 01:15:18.702792127 +0100
> @@ -1836,7 +1836,7 @@
> pref("intl.charset.fallback.override", "");
> pref("intl.charset.fallback.tld", true);
> pref("intl.ellipsis",
> "chrome://global-platform/locale/intl.properties");
> -pref("intl.locale.matchOS", false);
> +pref("intl.locale.matchOS", true);
> // fallback charset list for Unicode conversion (converting from
> Unicode)
> // currently used for mail send only to handle symbol characters (e.g
> Euro, trademark, smartquotes)
> // for ISO-8859-1
>
>
> Simply putting the default untouched langpacks from
> http://ftp.mozilla.org/pub/firefox/releases/43.0b3/linux-i686/xpi/
> into a vanilla
> /usr/lib/firefox43.0b3/lib/firefox-43.0/browser/extensions/ would
> recognize the LANG or LC_ALL exported by env in all file menu
> open/save as dialogues out of the box. BIG BUT: In Firefox's
> Tool>Add-Ons Menu would _not_ show any menu entry "Languages".
>
> In addition to that: For some unknown reasons, the xpi files (must
> always match the corresponding release, which I cared of) only get
> recognized as installed/registered langpacks not by means of manually
> faking the installation tricks the tough way by creating dirs such as
>
> d none
> usr/lib/firefox43.0b3/lib/firefox-43.0/browser/extensions/{972ce4c6-7e08-4474-a285-3208198ce6fd}
> 0755 root bin
> f none
> usr/lib/firefox43.0b3/lib/firefox-43.0/browser/extensions/{972ce4c6-7e08-4474-a285-3208198ce6fd}/icon.png
> 0644 root bin
> f none
> usr/lib/firefox43.0b3/lib/firefox-43.0/browser/extensions/{972ce4c6-7e08-4474-a285-3208198ce6fd}/chrome.manifest
> 0644 root bin
> f none {972ce4c6-7e08-4474-a285-3208198ce6fd}/install.rdf 0644 root bin
>
> manually (even though all ids matched it would always complain: id
> doesn't match ..).
>
> But the solution was (despite tough to find, only testing brought me
> to it) simpler:
> #0.) perform a local installation of all xpi files stored somewhere
> into the local user's ~/.mozilla simply via
> Tool>Add-Ons>Extensions>Install_Add-on_from_file and selecting all at
> once, then clicking ok (which is not visible), but one can blindly hit
> [TAB] 4 times and then [SPACE].
> #1.) Restart and close FF once (and make sure they were installed,
> accepted and enabled)
> #2.) (cd /export/home/martin/.mozilla/firefox/0deafbum.default;tar cvf
> - ./extensions)|(cd
> /usr/lib/firefox43.0b3/lib/firefox-43.0/browser/.;tar xvf -)
>
> Any other variation did not succeed.
> And doing the same from the same ~/.mozilla after a 2nd start of FF it
> would also break it again (fortunately I too a snapshot before that).
> Needs further investigation for the next verion.
> ATTENTION: Subfolders of
> /usr/lib/firefox43.0b3/lib/firefox-43.0/browser/extensions including
> /usr/lib/firefox43.0b3/lib/firefox-43.0/browser/extensions/{972ce4c6-7e08-4474-a285-3208198ce6fd}
> need x-access for the user who wants to start firefox, therefore
> permissions must be 755. Otherwise the langpacks will not work. This
> appears to explain the subtle errors I got at first. Because after
> wget'ing them from mozilla at first they are all 0600.
>
>
> INSTALL bins:
> @@@@@UPDATE@@@@@: Flash Plugin 9/10/11.x
> ... works now inside this gcc compiled Firefox!
> That was a complete Novum when I reached this step in April. This took
> another tough week
> of work night and day and night.
> Nobody has ever achieved this before with a gcc build of FF on Solaris
> (at least nobody in the records and nobody who offered to donate this
> to the broader planet OpenSolaris).
> From the src it appears that this must also affect other UNIX/LinUX OS
> ports, but I never tested this.
>
>
> It was tested and is known to work with:
> (instabilities and freezes related to videos that are in a too new
> format can always happen, but are not related to this port, while
> flash is closed src and I can only take their bins and make them work
> in FF as good as technically possible)
>
> drwxr-xr-x 2 martin other 4 Dec 27 2007
> flash_player_9_solaris_r115_x86
> drwx------ 2 martin other 3 Nov 8 2011
> flash_player_solaris_10_3_183_11_x86
> drwx------ 2 martin other 3 Nov 3 2011
> flash_player_solaris_11_1_102_56_x86
> -rw------- 1 martin other 4752427 Nov 9 2011
> flashplayer10_3r183_11_solaris_x86.tar.bz2
> -rw------- 1 martin other 4755108 Aug 9 2011
> flashplayer10_3r183_5_solaris_x86.tar.bz2
> -rw------- 1 martin other 4752536 Aug 23 2011
> flashplayer10_3r183_7_solaris_x86.tar.bz2
> -rw------- 1 martin other 6084262 Nov 4 2011
> flashplayer11_1r102_55_solaris_x86.tar.bz2
> -rw------- 1 martin other 6212851 Mar 20 2012
> flashplayer11_2r202_228_solaris_x86.tar.bz2
> -rw------- 1 martin other 2859276 Jun 10 2010
> flashplayer9r115_solaris_x86.tar.bz2
>
>
> Attention: On all newer Illumos based Distros _except_ OpenSXCE you
> probably need some loading order
> nonsense to get the Flash-Plugin 11.x initializing itself rather than
> crashing (only works sometimes, not with all codecs).
> In such cases you should remove your local libflashplayer.so from the
> plugins directory and set HTML5 as your default player (e.g. on
> youtube).
> But that drmatically reduces compatibility and rather often you cannot
> watch a news or music video. Because ... : HTML5 nice and good, but
> often they still force use to use flash.
>
>
>
>
> INSTALLATION:
>
> $ cd ~/Downloads
> $ wget
> http://opensxce.org/.FF/43.0b3/i386/firefox-43.0b3___Multilang__gcc4x-built__supports-flash-plugin__opensolaris-snv_130++_all_distros-i386.pkg.bz2
> $ bunzip2
> ./firefox-43.0b3___Multilang__gcc4x-built__supports-flash-plugin__opensolaris-snv_130++_all_distros-i386.pkg.bz2
> $ su
> # pkgrm SUNWfirefox43.0b3 (all other versions can be installed without
> a single namespace conflict)
> # pkgadd -d
> ./firefox-43.0b3___Multilang__gcc4x-built__supports-flash-plugin__opensolaris-snv_130++_all_distros-i386.pkg
>
>
>
> WHERE IS THE PLUGINS INSTALL DIR:
>
> Global location: /usr/lib/firefox43.0b3/plugins (symlink) or
> /usr/lib/firefox43.0b3/lib/firefox-43.0/browser/plugins
> Because by definition, given that I frequently want to test different
> flash plugins for testing, I always use ~/firefox/plugins which works.
> To verify which plugins got recognized open FF and enter the special
> URL: about:plugins
> Don't forget that Flash will crash on Illumos kernels except on
> OpenSXCE (which is the only one with a kernel fix).
>
>
>
> USAGE:
>
> FF43.0b3 can be started in 3 different ways, either of the following
> ...
>
> $ /usr/bin/firefox43.0b3
>
> $ /usr/lib/firefox43.0b3/bin/firefox
>
> or under JDS/Gnome from {Start}->{All
> Applications}->{Internet}->{Firefox 43.0b3 Beta}
>
>
>
> UNINSTALL:
> # pkgrm SUNWfirefox43-0b3
>
>
> best regards, %m
>
> email: opensxce at mail.ru
> (This would in theory also be my PAYPAL address, but you know the joke
> about theory vs. practice)
>
> }
>
>
>
>
>
> {
> # # # # # # # # # # # # # # # # # # # # # # # # # # # #
> # # # Part II.) *non*-TECHNICAL SCOPE: # # # # # # # # #
> # # # # # # # # # # # # # # # # # # # # # # # # # # # #
>
>
> Both OpenSXCE and the FF37 port could and can be downloaded
> anonymously by ANYone.
> http://opensxce.org/.FF/37/i386/README.txt
> http://svr4.opensxce.org/201405/i386/5.11/
>
> No single capitalist ever paid the license price (as I requested in
> http://svr4.opensxce.org/RELEASE_NOTES__OpenSXCE2014.05_x86_x64_INTEL32-64_AMD32-64____sun4u_sun4v_SPARC.txt
> ).
>
> The few donations (BIG THANKS TO THE FEW FRIENDS!!!) got smaller and
> more scarce.
> No single so called "donation" arrived for months. I say "so called"
> because a more fitting term is: pay at least a small fraction of the
> costs that arise through this volunteer work).
>
> But nevertheless all downloads are still available.
> Is that what a "capitalist" would do?
> M.B. lives in empoverished and hopelessly over-indebted conditions and
> of a total bank account balance of -1000 EUR has currently 1 Cent
> available (-999.999 EUR). Screenshot anytime ...
> One person claimed he wanted to "sponsor him" with 450 EUR per month.
> A closer investigation proved that his company looked for some project
> assigned engineer that they could hire, but only had a budget of a
> montly 450 EUR.
> What first appeared to be a helpful citizen who wanted to support
> OpenSXCE going to GitHub at closer investigation turned out to be
> attempted exploitation.
> More: During the endless instructions for the 2 commercial projects I
> should work on for those promised 450 EUR (== no donation!) he
> reiterated several times: Hahaha, no, we here in my company *never*
> pay anything for software, hahaha. The only single exception is xyz
> (some commercial product he mentioned).
> Nice community friend indeed!
>
> Experiences with the so called "community" stretch back more than 10
> years.
> Here another such example from 2007 (at the time I had already
> contributed MartUX [now OpenSXCE] as src and bins, the Xorg port to
> SPARC as src and bins, the port of the aperture driver to amd64 before
> xsvc got opened, the qemu-port to SPARC-Solaris hosts) :
>
> http://echelog.com/logs/browse/opensolaris/1195513200
>
> [06:33:47]<g4lt-U60> YAY m. Bochning is finally removed from the
> lists! happy day indeed!
> [06:40:52]<Tempt> Bochning?
> [06:40:58]<Tempt> Another noisy listtwit?
> [06:42:18]<g4lt-U60> Tempt, he's the one that was doing the "unscreib
> me" crap for the last week
> [06:42:37]<Triskelios> Tempt: martin, as in martux
> [06:42:48]<Tempt> oh
> [06:42:51]<Tempt> loonix wankstain
> [06:43:35]<Triskelios> eh... he's done good work. he was one of the
> main FOX people
> [06:43:43]<Tempt> Oh.
> [06:43:46]<Triskelios> has serious ... personality issues though
> [06:43:57]<jmcp> Tempt: Martin of the "I deserve funding because I've
> spent all my time working on martux rather than trying to get a job
> now i'm in serious debt"
> [06:44:07]<Tempt> Does anyone use Martux?
> [06:44:44] * jmcp shrugs
> [06:45:57]<Tempt> Meh
> [06:46:08]<Triskelios> the SPARC Xorg stuff was written for it
> [06:46:34]<Tempt> People with those sort of personality issues tend
> not to be able to sustain employment anyway.
>
>
>> And here from just this August:
>> http://openindiana.org/pipermail/openindiana-discuss/2015-August/017864.html
>>
>> jay at m5.chicago.il.us
>> Mon Aug 24 01:43:58 UTC 2015
>>
>> And, speaking of OpenSXCE (from which one can apparently obtain a
>> working version of Firefox 37) -- there was a huge amount of
>> discussion on this mailing list a while back about OpenSXCE. I do
>> not
>> want to re-awaken those sleeping threads, I truly do not want to read
>> any more postings about whether the man who gave us OpenSXCE is
>> irredeemably and irretrievably antisocial, or whether it is moral to
>> donate money to someone whose political sympathies lie with people
>> who
>> blow up schoolbuses.
>
>
> --->> Who<<--- is anti-social here?
> This person who probably never donated much - if any - of his labour
> for free?
> Which moral if any right do such "persons" have to talk about me like
> this?
> His only achievement (pls. correct me if this is wrong) I know of
> consisted of having posted a trolling from time to time.
>
> I know, this is a technical list. And I do understand the need to
> avoid political discussions.
> But this jay at m5.chicago.il.us "person" claimed devastating lies
> about me and not a single person had the balls to contradict his
> claims.
> What kind of "community" am I dealing with?
>
> Here is my short response to him:
>
> REMINDER:
>
> General Wesley Clark explains ISIS was created by U.S. Allies
> https://www.youtube.com/watch?v=ojcoKnTGf4s
>
> Wesley Clark: "Our friends and allies funded ISIS to destroy Hezbollah"
> https://www.youtube.com/watch?v=QHLqaSZPe98
>
> Hillary Clinton : We created Al-Qaeda
> https://www.youtube.com/watch?v=Dqn0bm4E9yw
>
> Hillary Clinton - We created ISIS
> https://www.youtube.com/watch?v=h5ff0EOj2WU
>
>
> Exposing the I.S.I.S Lie: How President Obama& Hillary Clinton Created
> I.S.I.S.
> https://www.youtube.com/watch?v=RU8-bgfR83Q
>
>
>
> UNFASSBAR!!! US Hauptziel IST kRIEG in Europa
> (UNBELIEVABLE!!! USA's main goal is WAR IN EUROPE
> https://www.youtube.com/watch?v=xi7M_P2ldsQ
>
>
>
> NATO wants to "fight ISIS" to - in reality - support them even better
> than until now.
> GOAL: To remove Assad and to invade Iran. Then Russia/China are
> finally completely encircled.
>
>
> https://www.youtube.com/results?search_query=mc+cain+isis
>
> US Senator John McCain: "I know ISIS intimately, I've met them and I
> talk to them all the time."
> https://www.youtube.com/watch?v=tgLj5s0Xk8A
>
> ISIS : John McCain admits he met ISIS and says "We know these people
> intimately" (Sept 16, 2014)
> https://www.youtube.com/watch?v=vItuKKuz_7Y
>
> McCain's ISIS Ties Are Too Numerous For America
> https://www.youtube.com/watch?v=C7VHPKlWJUQ
>
>
> ISIS ISIL Leader with John McCain (US GOV)? - BREAKING NEWS ALERT
> https://www.youtube.com/watch?v=Ba0jpRjGFDo
>
>
> Confirmation now in that Paris attacks were false flag.
>
> Because nothing says ‘False Flag’ like a ‘takfiri terror cell’
> planning their exercise on the exact same day as the ‘government
> terror cell’.
> http://www.globalresearch.ca/we-were-prepared-large-military-exercise-took-place-on-same-day-as-paris-terrorist-attacks/5489164
>
> Washington Refines Its False Flag Operations — Paul Craig Roberts
> http://www.paulcraigroberts.org/2015/11/16/washington-refines-its-false-flag-operations-paul-craig-roberts/
>
> Thank you, America: Sinai plane terror victims:
> https://www.google.de/search?q=sinai+plane+victims&source=lnms&tbm=isch&sa=X&ved=0CAgQ_AUoAmoVChMIgNqTpp-WyQIVyNgaCh0HDweA&biw=1280&bih=873
> (mostly young women and children were on board)
>
> I didn't see anybody changing his fb or twitter avatar to the Russian
> tricolore.
> Unlike now in Paris.
>
> This heartbreaking image shows 10 months young Darina, who is among the
> victims
> http://www.zerohedge.com/sites/default/files/images/user92183/imageroot/2015/10/PlanesDarina_0.png
>
> Russian Plane "Broke Apart In The Air," Officials Say As Investigators
> Frantically Search For "Clues"
> http://www.zerohedge.com/news/2015-11-01/russian-plane-broke-apart-air-officials-say-investigators-frantically-search-clues
>
> May I also ask: How was it possible that the USA knew in minutes the
> plane was shot down by ISIS, and that in less than 1 week sattelite
> images and so called "intelligence" could be provided, which are
> mysteriously still not available for MH17 let alone MH370 in 1.5
> years??????????
>
> Move on:
>
> Published: October 2nd: INSANE MCCAIN: ARM SYRIAN REBELS TO SHOOT DOWN
> RUSSIAN PLANES 'Just like we did in Afghanistan'
> https://www.youtube.com/watch?v=qXinE5lSQYI
>
> John McCain and Free Syrian Army of Israeli State Finally Shot Down a
> Russian Plane
> https://www.youtube.com/watch?v=lqwmdpvawog
>
> Leuren Moret: McCain in false flag to down Russian airliner, caught
> just prior to Sinai crash
> https://www.youtube.com/watch?v=ChT4VDm_n8M
>
> On FOX-News: US Senator John McCain says he'd give terrorists means to
> shoot down Russian planes.
> https://www.youtube.com/watch?v=paR_5iqnZsU
>
> Published September 30th: BREAKING: John McCain To Arm Rebels with STA
> Missiles To Shoot Down Russian Jets
> https://www.youtube.com/watch?v=EM-bV19J6w8
>
> Published October 2nd: Insane McCain Wants ISIS To Shoot Down Russian
> Jet Fighters
> https://www.youtube.com/watch?v=NaDPumW5qMQ
>
> Published October 2nd: US Senator John McCain: Arm Syrian Rebels to
> shoot down Russian planes!
> https://www.youtube.com/watch?v=tDFql8eE0TA
>
> Published October 2nd: Senator John McCain Wants Russian Fighter Jets
> Shot Down In Syria
> https://www.youtube.com/watch?v=YQIHCScb77o
>
> Published October 31st: BREAKING: [CENSORED] SHOOTS DOWN RUSSIAN
> AIRLINER THANKS TO JOHN MCCAIN
> https://www.youtube.com/watch?v=t7AUfWlGlmk
>
>
>
> WHO is supporting terrorists, valued Mr. jay at m5.chicago.il.us???
> Answer: IT IS *****YOU*****:
>
> Kiev started shooting their own civilians weeks before the war started
> in Mai 2014.
> YT is full of evidence (search in cyrillic).
> It has nothing to do with the Kremlin.
>
> Захват БМП в Мариуполе (другой ракурс) 11:42 9 мая 2014
> https://www.youtube.com/watch?v=MhfPkHvFBvE
>
>
> Let alone the Odessa massacre, the entire Marioupol massacre, and the
> massacred Bus in Fevruary 2014 with people from Crimea that wanted to
> travel back home from Anti-Maidan in Kiev.
> And believe me: The West is the worst tyrant of all times.
>
>
> https://www.google.de/search?q=odessa+pregnant+woman+massacre&source=lnms&tbm=isch&sa=X&ei=8lQfVaDPM47raMTugPgH&ved=0CAcQ_AUoAQ&biw=1280&bih=853
>
> https://www.youtube.com/watch?v=LvVuFCYPx0Y
>
> http://scgnews.com/the-odessa-massacre-what-really-happened
>
> Very Real Neo-Nazi Threat in Ukraine. A compilation of what mainstream
> media isn't showing.
> https://www.youtube.com/watch?v=pYjsCJwcPWk
>
> How the US Government Conquered Ukraine to Start WW3 With Russia
> https://www.youtube.com/watch?v=V6yL_8lal3c
>
>
> NATO bombing of Yugoslavia 1999. / NATO bombardovanje Jugoslavije 1999.
> https://www.youtube.com/watch?v=gelCe981y0I
>
> Documentary: "Why Zasto" - NATO bombing Yugoslavia IN 1999
> https://www.youtube.com/watch?v=HxDcvhH2Ypc&list=PLRImknr7RoMDl2OvaP14Mx2mBMKgMeLbK
>
> Reason Why I Hate USA - American Bombing of Serbia (based on lies and
> false completely made up pretexts, as every time!)
> https://www.youtube.com/watch?v=Ju-Y73Qn05Q
>
>
>
> I was born as a simple East-German in wonderful communist Soviet
> East-Berlin:
> It was a peaceful country, not invading "region" after "region", year
> after year since 1945, killing tens of millions!!!
> And YES, I do have a right to say all this! Because before 1989 your
> God-darn Western propaganda mouthpieces promised us "Freedom of
> speech" and "democracy", "freedom" and all that noise!
>
> What we in fact got from the golden West was this:
>
> TREUHANDKRIMINALITAET (aks "Privatization") :
> beutezug ost https://www.youtube.com/results?search_query=beutezug+ost
>
> and in Russia and now Ukraine:
> Всё идёт по плану (Everything goes [well] according to plan)
> https://www.youtube.com/watch?v=QS54GkLRtrI
>
> Compare this to """evil evil evil""" communism, my homeland
> East-Germany which got robbed off, de-industrialized and ensalved.
> Here before 1989:
>
> GDR:
> Weiße Maus Verkehrspolizei DDR 1987
> https://www.youtube.com/watch?v=37iy1MB1Wh8
>
> Berlin, Hauptstadt der DDR Teil 1 / Берлин, ГДР
> https://www.youtube.com/watch?v=oZdxuOK3YuE
>
> Soviet Union:
> Наташа Королёва Жёлтые тюльпаны Песня 1990
> https://www.youtube.com/watch?v=E57T8XYa7OM
>
> Moscow Olympic 1980 (BOYCOTTED by the USA, which dound it better to
> create Al-Quaeda during that time! According to Hillary Clinton
> herself)
> https://www.youtube.com/watch?v=qt3xVMt0k4k
>
> Yugoslavia:
> Few Americans know little if anything about: Yugoslavia - the Good Old
> Days of Tito-style Communism
> https://www.youtube.com/watch?v=0RRnCnO1Y2c
>
>
>
> I feel ashamed and DISGUSTED by Germany's EU's and generally
> Western actions. And also by such "community members" who care a damn
> shit for dead Russian women and children in Donbass or Egypt, let
> alone Syrians, Iraqis, Libyans, Afghans, Serbs and all the list
> through their amazing history as an "exceptional nation".
>
> IMAGINE: I got excluded from the Illumos discuss list in 2013 by
> Madame Deidre Straughan (gets $$$ for being a "community architect"
> destroying the community but never submitted a line of code) for
> having said the truth. And the US-sponsored terror attacks in
> Wolgograd DID KILL circa 20 CIVILIANS during the Sochi Olympics, when
> busses where blown up. Nobody had a tear. Now after the next western
> false flag everybody jumps on the MSM hysteria and calls for WAR.
> Although the only french airplane carrier headed from France to the
> Middle East 2 days _before_ the "attacks" took place.
>
> The ones who are spilling blood all over the planet - hint: It's not
> the Russians.
> And it were not the communists in my home country -
> GermanDemocraticRepublic.
> It's the West and the terror groups it create[d|s] and finance[d|s] for
> decades.
> I'm not talking about politics here. Sure - that shall not be done on
> such lists.
> But what we are dealing with here is not "politics", but
> mass-slaughter and endless genocides co-committed by our beloved
> "Nobel peace prize" winner Obomber and the EU of "western values".
> I not only have every right to speak about this, but it is my duty.
> And should be everyone's who still wants to look in the mirror.
> The day comes closer on which we no longer have the possibility to
> attend mailing lists. Who wants to let that happen and wait until his
> death with a paralyzed smile is free to do so! But I personally won't.
> GET THAT.
>
> Let me close this non-technical section forced onto us triggered by
> folks who never contributed technically, such as this jay at
> m5.chicago.il.us
> (he started it and nobody ended it, this forces me to write all this,
> yet another time ...)
> So if anybody wants to stop seeing such non-technical paragraphs,
> break the patttern ..., else it repeats while true.
>
> Listen to murdered US presidents (the only US-presidents who really
> wanted to serve their nation (and the world) all got assassinated.
> That alone would cause any true American patriot to think twice about
> CNN-News sort of "reporting" across ___all___ western MSM, who are all
> repeating the very identical "reports" from Reuters and AFP (which all
> across the alleged spectrum are owned by 7 families!)
>
> JFK - The Speech That Killed Him
> https://www.youtube.com/watch?v=y8HTr-F-FVM
>
> Eisenhower warns us of the military industrial complex.
> https://www.youtube.com/watch?v=8y06NSBBRtY
>
> (From books) Thomas Jefferson warns of the dangers of central banks
> and national debt
> https://www.youtube.com/watch?v=UrxKOO0nKwc
>
> In the old days (not only in communist countries) folks were educated
> and smart.
> Not only "expert" idiots, but had some general wisdom, were interested
> in all truly important aspects of life and defended PEACE AND JUSTICE.
> What a pathetic mass of equalized yes-sayers has this been washed into
> by "your" TV in just 3 decades?
>
> Imagine somebody during WW2 reporting about the horrible crimes
> committed by Nazi-Germany would have come to this list with
> eyewittness reporting, tons of youtube footage and waterproof
> evidence.
> And your only reaction would have been: ^^Sir, you violated the newly
> introduced "no-kveching-rule" which was just created _after_ your
> "hatred post".^^
> THAT's what Illumos-devel has done in 2013.
> And by doing so they are in fact complicit with genocides committed by
> their "government".
> Contradicting the American Costitution by all accounts (as every
> US-administration since JFK).
>
> For years I had not managed to grasp, why the LinUX, *BSD and other
> open-src communities disliked Solaris so much, even though Sun did
> open it in the form of OpenSolaris, and even though Solaris is one of
> the most adavanced operating systems ever created.
> Over time when dealing with some folks the reason can only be: It has
> little to do with the software, but more with such McCain -Palin
> maniac fanboys in its Sun kernel team.
>
> p.s. Back to the "Anti-Social Capitalist" bs: Why is nobody
> complaining about the fact, that certain users who never contribute
> anything substantial pollute the public oi- and other lists by
> spreading SPAM and ADS for their own private COMMERCIAL businesses,
> even in cases where all they "contributed" was a one-liner joke, such
> as here:
> http://openindiana.org/pipermail/openindiana-discuss/2015-November/018225.html
> ???
>
> If we want equal standards for all, then some should not be "more
> equal".
> }
>
>
>
>
>
>
> {
> # # # # # # # # # # # # # # # # # # # # # # # # # # # #
> # # # Part III.) THANKS: # # # # # # # # # # # # # # #
> # # # # # # # # # # # # # # # # # # # # # # # # # # # #
>
> Most of all I would like to thank my girlfriend and my family who have
> limitless amounts of
> understanding for the financial and time-realted problems caused by
> this.
> I LOVE HER
>
> And then: You can thank Al Hopper who pays for all my OpenSolaris
> related hosting, domain names etc since 2005!!!!!
> Thanks also to all who ever sent me money (as listed in
> http://svr4.opensxce.org/RELEASE_NOTES__OpenSXCE2014.05_x86_x64_INTEL32-64_AMD32-64____sun4u_sun4v_SPARC.txt
> and to Ken Mays for his never-ending series of mentioning me on-list
> in some positive way, plus for having designed OpenSXCE.org's old (now
> outdated) website.
>
> If anybody attempted to contact me via martin at martux.org since July:
> Please don't be upset. I no longer have any pw access because google
> claims "I'm running a biz", which would be nice but is not the case.
> They want 5 EUR per month and I cannot by any means pay that
> (especially not backwards from tip).
> My new eMail address is: opensxce [[at]] mail.ru
> If anybody would in fact want to give me some $$$ back for my work
> (after he tested if he likes it), then plese use that new email
> address, tnx.
> If not, I cannot do anything about that, as all the way back to 2005
> and 2003.
> }
>
> Sorry - won't spell-check or correction-read this anomore.
> I dislike writing such long ANNOUNCE files.
> Everything that needed to addressed has been said. By releasing this
> FF port the ball is now in your field.
> If OpenIndiana and or Illumos is interested in any friendship, then
> everybody can see that I once again attempted to reach out and gave
> you my hand.
> This time you have all FF related SRC patches. So it proves that I
> trust you.
>
> You can thank friendly long-term community members from good old osol
> times, because without having seen that they repeatedly had some good
> words for me on the ml list and openindiana.org's FF page and the
> illumos site, this release would not have happened (regards and BIG
> tnx to all my misc. friends, as listed in
> http://svr4.opensxce.org/RELEASE_NOTES__OpenSXCE2014.05_x86_x64_INTEL32-64_AMD32-64____sun4u_sun4v_SPARC.txt
> !).
>
>
>
> FOR THE DEMOCRATIC REPUBLICS, IMAGINE
>
> Martin Bochnig
> opensxce at mail.ru
>
>
>
>
>
>
>
> _______________________________________________
> openindiana-discuss mailing list
> openindiana-discuss at openindiana.org
> http://openindiana.org/mailman/listinfo/openindiana-discuss
More information about the openindiana-discuss
mailing list