Google Groups no longer supports new Usenet posts or subscriptions. Historical content remains viewable.
Dismiss
Firefox and Seamonkey support for non ss2
157 views
Skip to first unread message
Richmond
Dec 24, 2016, 4:37:51 PM12/24/16
to
I read that mozilla is withdrawing support for processors without the
sse2 instruction set (and so it will disappear from seamonkey too). Will
there be a build with slackware that can run without sse2 instructions?
(I am not sure why hardware independent languages like C require
particular instructions, I am just accepting there is a reason.
sse2 instruction set (and so it will disappear from seamonkey too). Will
there be a build with slackware that can run without sse2 instructions?
(I am not sure why hardware independent languages like C require
particular instructions, I am just accepting there is a reason.
Aragorn
Dec 24, 2016, 5:12:11 PM12/24/16
to
On Saturday 24 December 2016 22:37, Richmond conveyed the following to
alt.os.linux.slackware...
> I read that mozilla is withdrawing support for processors without the
> sse2 instruction set (and so it will disappear from seamonkey too).
This will most likely be the case in their readily downloadable and
installable packages then. Distribution vendors may or may not decide
to bundle those versions as part of the distribution's repository, or
they may decide to build the software in question from source code.
The above said, I'm not a programmer, but I strongly suspect that
nothing in the Firefox/Mozilla code itself would be going so low-level
as to directly address the SSE2 registers, and that it would be rather a
compile-time option.
> Will there be a build with slackware that can run without sse2
> instructions?
You'd have to ask Pat that question. :)
> (I am not sure why hardware independent languages like C require
> particular instructions, I am just accepting there is a reason.
I suppose it's got to do with a greater requirement for multimedia
support ─ HTML5, among other things.
--
= Aragorn =
alt.os.linux.slackware...
> I read that mozilla is withdrawing support for processors without the
> sse2 instruction set (and so it will disappear from seamonkey too).
installable packages then. Distribution vendors may or may not decide
to bundle those versions as part of the distribution's repository, or
they may decide to build the software in question from source code.
The above said, I'm not a programmer, but I strongly suspect that
nothing in the Firefox/Mozilla code itself would be going so low-level
as to directly address the SSE2 registers, and that it would be rather a
compile-time option.
> Will there be a build with slackware that can run without sse2
> instructions?
> (I am not sure why hardware independent languages like C require
> particular instructions, I am just accepting there is a reason.
support ─ HTML5, among other things.
--
= Aragorn =
Askfor
Dec 27, 2016, 7:30:03 AM12/27/16
to
Aragorn wrote:
> On Saturday 24 December 2016 22:37, Richmond conveyed the following to
> alt.os.linux.slackware...
>
>> I read that mozilla is withdrawing support for processors without the
>> sse2 instruction set (and so it will disappear from seamonkey too).
>
> This will most likely be the case in their readily downloadable and
> installable packages then. Distribution vendors may or may not decide
> to bundle those versions as part of the distribution's repository, or
> they may decide to build the software in question from source code.
>
> The above said, I'm not a programmer, but I strongly suspect that
> nothing in the Firefox/Mozilla code itself would be going so low-level
> as to directly address the SSE2 registers, and that it would be rather a
> compile-time option.
>
>> Will there be a build with slackware that can run without sse2
>> instructions?
>
> You'd have to ask Pat that question. :)
>
>> (I am not sure why hardware independent languages like C require
>> particular instructions, I am just accepting there is a reason.
>
I don't think that C or C++ require instruction set extension. However,
> On Saturday 24 December 2016 22:37, Richmond conveyed the following to
> alt.os.linux.slackware...
>
>> I read that mozilla is withdrawing support for processors without the
>> sse2 instruction set (and so it will disappear from seamonkey too).
>
> This will most likely be the case in their readily downloadable and
> installable packages then. Distribution vendors may or may not decide
> to bundle those versions as part of the distribution's repository, or
> they may decide to build the software in question from source code.
>
> The above said, I'm not a programmer, but I strongly suspect that
> nothing in the Firefox/Mozilla code itself would be going so low-level
> as to directly address the SSE2 registers, and that it would be rather a
> compile-time option.
>
>> Will there be a build with slackware that can run without sse2
>> instructions?
>
> You'd have to ask Pat that question. :)
>
>> (I am not sure why hardware independent languages like C require
>> particular instructions, I am just accepting there is a reason.
>
Mozilla may choose to build from source with appropriate command line
option which may cause compiler to emit such op codes (instructions).
Computers without such instruction set will be catching some sort of
signal and aborting execution. I think that Foxit PDF Reader is built
that way.
I suppose it won't stop anyone to build different executable from source.
Like it or not 32-bit is going away. There is less and less support for
it. 32-bit is perfectly fine for home users, but hardware vendors are
making 64-bit, and those who write and/or package software often can't
be bothered to maintain two versions.
I guess this move is more to give additional push to 32-bit users than
due to real need. That is progress being pushed down our throats and
there are always idiots and vegetables in human form who say that "one
has to embrace the change" or something.
Slacky
Dec 27, 2016, 1:56:17 PM12/27/16
to
On 2016-12-27, Askfor <ask...@nowhere.net> wrote:
>> The above said, I'm not a programmer, but I strongly suspect that
>> nothing in the Firefox/Mozilla code itself would be going so low-level
>> as to directly address the SSE2 registers, and that it would be rather a
>> compile-time option.
Firefox actually has a bunch of assembly code which is why it is not
>> The above said, I'm not a programmer, but I strongly suspect that
>> nothing in the Firefox/Mozilla code itself would be going so low-level
>> as to directly address the SSE2 registers, and that it would be rather a
>> compile-time option.
available on most platforms.
> Like it or not 32-bit is going away. There is less and less support for
> it. 32-bit is perfectly fine for home users, but hardware vendors are
> making 64-bit, and those who write and/or package software often can't
> be bothered to maintain two versions.
If it weren't for application bloat there would be essentially no reason you
couldn't live with 4GB of RAM for a desktop. Given how much apps like
Firefox and Acroread use even 16GB feels like a small desktop these days.
I have a box for work with a core i7, 32GB of DDR4 and an SSD and it still
can't run Windows 7 acceptably. Life sucks when you have to run bloatware.
Linux is getting there. You could say it's already there.
> I guess this move is more to give additional push to 32-bit users than
> due to real need. That is progress being pushed down our throats and
> there are always idiots and vegetables in human form who say that "one
> has to embrace the change" or something.
there is any hardware but Intel. I happen to know there is because I have a
bunch of it sitting in front of me. But those same people don't realize
there are any OS besides Windows or Linux so...
They don't want to have platform dependent code for other than Intel x86_64
anyway. Many of these guys have day jobs so it is inevitable.
Slacky
>
>
>
Jens Stuckelberger
Dec 27, 2016, 2:53:12 PM12/27/16
to
On Tue, 27 Dec 2016 18:56:14 +0000, Slacky wrote:
> I have a box for work with a core i7, 32GB of DDR4 and an SSD and it
> still can't run Windows 7 acceptably. Life sucks when you have to run
> bloatware. Linux is getting there. You could say it's already there.
It is, when you use distributions like Fedora and Ubuntu, and
> I have a box for work with a core i7, 32GB of DDR4 and an SSD and it
> still can't run Windows 7 acceptably. Life sucks when you have to run
> bloatware. Linux is getting there. You could say it's already there.
derivatives, or a desktop system like Gnome, KDE or Unity. Which is why
I, for one, hope that year of Linux in the desktop will never arrive.
Aragorn
Dec 27, 2016, 3:51:55 PM12/27/16
to
On Tuesday 27 December 2016 19:56, Slacky conveyed the following to
alt.os.linux.slackware...
> Until Acroread comes in a native 64 bit version I'm staying with Slack
> 32.
Why on Earth would you use such a crappy piece of proprietary software?
There are plenty of FLOSS alternatives that work _a lot_ better.
Personally, I like Okular, but other people's mileage may vary.
--
= Aragorn =
alt.os.linux.slackware...
> Until Acroread comes in a native 64 bit version I'm staying with Slack
> 32.
There are plenty of FLOSS alternatives that work _a lot_ better.
Personally, I like Okular, but other people's mileage may vary.
--
= Aragorn =
Askfor
Dec 27, 2016, 5:30:03 PM12/27/16
to
Slacky wrote:
> On 2016-12-27, Askfor <ask...@nowhere.net> wrote:
>
>>> The above said, I'm not a programmer, but I strongly suspect that
>>> nothing in the Firefox/Mozilla code itself would be going so low-level
>>> as to directly address the SSE2 registers, and that it would be rather a
>>> compile-time option.
>
> Firefox actually has a bunch of assembly code which is why it is not
> available on most platforms.
>
>> Like it or not 32-bit is going away. There is less and less support for
>> it. 32-bit is perfectly fine for home users, but hardware vendors are
>> making 64-bit, and those who write and/or package software often can't
>> be bothered to maintain two versions.
>
> Until Acroread comes in a native 64 bit version I'm staying with Slack 32.
> If it weren't for application bloat there would be essentially no reason you
> couldn't live with 4GB of RAM for a desktop. Given how much apps like
> Firefox and Acroread use even 16GB feels like a small desktop these days.
Actually, I had problems with 32-bit Acroread on one computer. It kept
> On 2016-12-27, Askfor <ask...@nowhere.net> wrote:
>
>>> The above said, I'm not a programmer, but I strongly suspect that
>>> nothing in the Firefox/Mozilla code itself would be going so low-level
>>> as to directly address the SSE2 registers, and that it would be rather a
>>> compile-time option.
>
> Firefox actually has a bunch of assembly code which is why it is not
> available on most platforms.
>
>> Like it or not 32-bit is going away. There is less and less support for
>> it. 32-bit is perfectly fine for home users, but hardware vendors are
>> making 64-bit, and those who write and/or package software often can't
>> be bothered to maintain two versions.
>
> Until Acroread comes in a native 64 bit version I'm staying with Slack 32.
> If it weren't for application bloat there would be essentially no reason you
> couldn't live with 4GB of RAM for a desktop. Given how much apps like
> Firefox and Acroread use even 16GB feels like a small desktop these days.
throwing some error mentioning Debian !!??? I don't remember exactly any
more.
>
> I have a box for work with a core i7, 32GB of DDR4 and an SSD and it still
> can't run Windows 7 acceptably. Life sucks when you have to run bloatware.
> Linux is getting there. You could say it's already there.
>
Agree
> I have a box for work with a core i7, 32GB of DDR4 and an SSD and it still
> can't run Windows 7 acceptably. Life sucks when you have to run bloatware.
> Linux is getting there. You could say it's already there.
>
>> I guess this move is more to give additional push to 32-bit users than
>> due to real need. That is progress being pushed down our throats and
>> there are always idiots and vegetables in human form who say that "one
>> has to embrace the change" or something.
>
> If you're talking about Firefox it has to do with most people in denial that
> there is any hardware but Intel. I happen to know there is because I have a
> bunch of it sitting in front of me. But those same people don't realize
> there are any OS besides Windows or Linux so...
>
there are ARM, SPARC and others. I worked on 32 and 64-bit SPARCs for
quite a while. Particularly I am talking about 32-bit Intel. At least
there should have been a decent emulation. When 16-bit was going away,
i386 and i486 did decent job running 16-bit apps without extra
compatibility layer.
> They don't want to have platform dependent code for other than Intel x86_64
> anyway. Many of these guys have day jobs so it is inevitable.
>
notbob
Dec 27, 2016, 6:20:08 PM12/27/16
to
On 2016-12-27, Aragorn <thor...@telenet.be> wrote:
> Personally, I like Okular, but other people's mileage may vary.
Works fer me!
> Personally, I like Okular, but other people's mileage may vary.
I tried Adobe Reader on my Slack 14.1 box. Total junk. I hadda kill the
browser and re-boot it every time AR changed a document. Total
nonsense. I tossed it and use okular, which operates independently of
the browser.
I see zero advantage to anything from Adopey. I think they are one of
the most proprietary thugs on the block. They have a total lock on
our e-books from both our local libraries with their Adobe Digital
Editions (ADE).
Not sure if that's the librarian's fault, but it might be. Both are
M$/Mac bastions. :(
nb
Eef Hartman
Dec 27, 2016, 6:27:46 PM12/27/16
to
Richmond <dnom...@gmx.com> wrote:
> I read that mozilla is withdrawing support for processors without the
> sse2 instruction set (and so it will disappear from seamonkey too).
The sse2 instruction set was first introduced by Intel with the
> I read that mozilla is withdrawing support for processors without the
> sse2 instruction set (and so it will disappear from seamonkey too).
initial version of the Pentium 4 in 2001.
So it would only exclude Pentium III and older CPU's from usage.
And you can keep on using those with an (then) older version of
firefox cq seamonkey.
Askfor
Dec 27, 2016, 7:30:03 PM12/27/16
to
more. Glad I am wrong. Fuck progress.
Mike Spencer
Dec 28, 2016, 12:23:17 AM12/28/16
to
Netscape 4.76 and FF 2.x. Yes, I have a laptop with SLack 14.1, lotsa
RAM and a newer FF, but still no KDE/Gnome.
I get tired of trying to beat back progress but so far I'm ahead on
points.
--
Mike Spencer Nova Scotia, Canada
Richmond
Dec 28, 2016, 1:53:18 AM12/28/16
to
Eef Hartman
Dec 28, 2016, 6:08:22 AM12/28/16
to
Except for the Athlon64 (2800+ and higher), but indeed, my old
AMD Athlon XP 1800+ wouldn't have been able to run firefox with
SSE2 instructions.
That one, though, has given up about a year ago so now I'm using a
real Intel Core 2 Duo (3 GHz) machine (HP xw4600 Workstation I got
cheap 2nd hand).
AMD Athlon XP 1800+ wouldn't have been able to run firefox with
SSE2 instructions.
That one, though, has given up about a year ago so now I'm using a
real Intel Core 2 Duo (3 GHz) machine (HP xw4600 Workstation I got
cheap 2nd hand).
Slacky
Dec 29, 2016, 8:27:16 AM12/29/16
to
Intel-dependent code like we're talking about in Firefox is still platform
dependent. It's just they think Intel is the only platform and that is a
self-fulfilling screwology.
Slacky
Jim Diamond
Dec 30, 2016, 8:35:04 AM12/30/16
to
On 2016-12-27 at 19:20 AST, notbob <not...@nothome.com> wrote:
> On 2016-12-27, Aragorn <thor...@telenet.be> wrote:
>> Personally, I like Okular, but other people's mileage may vary.
> Works fer me!
> I tried Adobe Reader on my Slack 14.1 box. Total junk. I hadda kill the
> browser and re-boot it every time AR changed a document. Total
> nonsense. I tossed it and use okular, which operates independently of
> the browser.
I think you have some major confusion about things. AR does not need
> On 2016-12-27, Aragorn <thor...@telenet.be> wrote:
>> Personally, I like Okular, but other people's mileage may vary.
> Works fer me!
> I tried Adobe Reader on my Slack 14.1 box. Total junk. I hadda kill the
> browser and re-boot it every time AR changed a document. Total
> nonsense. I tossed it and use okular, which operates independently of
> the browser.
a web browser. You can (and maybe should) run it as a stand-alone
program.
> I see zero advantage to anything from Adopey.
superior to Okular/evince/xpdf as well as any other open-source PDF viewer
that I have seen. I wish it wasn't so, since I would be happy getting
rid of Adobe's bloatware, but so far I haven't found anything that
comes close.
Jim
Sylvain Robitaille
Dec 30, 2016, 12:41:14 PM12/30/16
to
On 2016-12-29, Slacky wrote:
> Yeah but I think more than a few people are missing the essential
> point Intel-dependent code like we're talking about in Firefox is
> still platform dependent. ...
> Yeah but I think more than a few people are missing the essential
> point Intel-dependent code like we're talking about in Firefox is
What Intel-dependent code, though? I thought this discussion was about
the planned removal of support for non sse2-capable processors?
: elvira[syl] ~; grep -w vendor_id /proc/cpuinfo |uniq -c
8 vendor_id : AuthenticAMD
: elvira[syl] ~; grep -w ^flags /proc/cpuinfo | uniq -c
8 flags : ... sse sse2 ...
For what it's worth, my Asus EeePC's Celeron processor has the sse2
flag, so even it won't be left behind.
--
----------------------------------------------------------------------
Sylvain Robitaille s...@encs.concordia.ca
Systems analyst / AITS Concordia University
Faculty of Engineering and Computer Science Montreal, Quebec, Canada
----------------------------------------------------------------------
Henrik Carlqvist
Dec 30, 2016, 2:16:28 PM12/30/16
to
On Tue, 27 Dec 2016 21:51:53 +0100, Aragorn wrote:
> Why on Earth would you use such a crappy piece of proprietary software?
> There are plenty of FLOSS alternatives that work _a lot_ better.
One reason to use Adobe Acrobat reader (and also install multilib with
> Why on Earth would you use such a crappy piece of proprietary software?
> There are plenty of FLOSS alternatives that work _a lot_ better.
all its disadvantages to be able to run 32-bit binaries) is to be able to
look at 3D models in pdf files. There is a simple example of such a 3D
model in the two pdf files in pdf_3D.zip from
http://rainnic.altervista.org/en/content/embed-3ds-pdf-latex-u3d
I have not yet seen any other pdf viewer being capable of showing such
interactive 3D models in pdf files where you can rotate, pan and zoom the
model.
regards Henrik
--
The address in the header is only to prevent spam. My real address is:
hc351(at)poolhem.se Examples of addresses which go to spammers:
root@localhost postmaster@localhost
Eef Hartman
Dec 30, 2016, 9:28:14 PM12/30/16
to
Richmond <dnom...@gmx.com> wrote:
> I read that mozilla is withdrawing support for processors without the
> sse2 instruction set (and so it will disappear from seamonkey too). Will
> there be a build with slackware that can run without sse2 instructions?
Try it, with the just released seamonkey 2.46 packages (for Slackware
> I read that mozilla is withdrawing support for processors without the
> sse2 instruction set (and so it will disappear from seamonkey too). Will
> there be a build with slackware that can run without sse2 instructions?
14.1 and later).
Richmond
Jan 8, 2017, 1:12:40 PM1/8/17
to
seamonkey
(seamonkey:32549): Gdk-WARNING **: gdkproperty-x11.c:325 invalid X atom:
453
[32549] ###!!! ABORT: X_ShmDetach: BadShmSeg (invalid shared segment
parameter): file
/tmp/seamonkey-2.46/mozilla/toolkit/xre/nsX11ErrorHandler.cpp, line 157
[32549] ###!!! ABORT: X_ShmDetach: BadShmSeg (invalid shared segment
parameter): file
/tmp/seamonkey-2.46/mozilla/toolkit/xre/nsX11ErrorHandler.cpp, line 157
Segmentation fault
I used ssh -Y to connect from host to guest. The guest does not have X
running, I am using the host X server. When I ran xterm and then ran
seamonkey from inside that, it worked. When I quit xterm, and ran
seamonkey from the original shell, the error no longer occured.
0 new messages