f



Modules with GCC - possible?

Hi,
  I've just got myself a copy of Foundation CD 8 from RISC OS Ltd and I'd
like to have a go at developing a PDumper. Unfortunately, I don't have a copy
of the Castle or Acorn C compilers (if Castle were selling the compiler for
�50 or even �75 I'd buy it, but �200 is a but much). Is there any way to
generate a module using GCC or LCC (ISTR someone ported that to RISC OS)?
Failing that, can anyone suggest a decent ARM assembler tutorial? My ARM ASM
skills are incredibly rusty (not that they were particularly good in the
first place)...

Thanks.
-- 
Phil.
philpem@despammed.com  <<-- This address is valid
http://www.philpem.dsl.pipex.com/
.... A fail-safe circuit will destroy others.
0
philpem (337)
7/2/2003 1:22:04 PM
comp.sys.acorn.programmer 2499 articles. 0 followers. Post Follow

22 Replies
430 Views

Similar Articles

[PageSpeed] 19

In message <slrnbg5p0o.epi.{tony}@realh.co.uk>
          Tony Houghton <{tony}@realh.co.uk> wrote:

> It contains complete working versions of the C++ and C compilers (the
> latter is version 5.06 and can output 32-bit code despite its age)
Well, that could be useful...

> which
> can be run from the command line. Everything else you need is available
> as freeware, all you'll be missing is the GUI front-ends which are of
> dubious use anyway.
I know Justin Fletcher has written an open-source alternative to CMHG called
"Cmunge", so I can probably use that to generate the module headers. As for
the build scripts, David Pilling rather thoughtfully used Obey files instead
of Acorn Make. In his words - 
  "Honestly, I did once study and understand the Acorn Make program, and I
   came to the conclusion that it couldn't do everything I needed and would  
   be more complicated to use!"

Thanks.
-- 
Phil.
philpem@despammed.com  <<-- This address is valid
http://www.philpem.dsl.pipex.com/
"High voltage. Yeah, right. I can hold this 20kV wire with my fingers no prob... YEAARGH!" *thud*
0
philpem (337)
7/2/2003 2:18:36 PM
In article <8d8ab70b4c.philpem@dsl.pipex.com>, Philip Pemberton
<URL:mailto:philpem@despammed.com> wrote:
> Hi,
>   I've just got myself a copy of Foundation CD 8 from RISC OS Ltd and I'd
> like to have a go at developing a PDumper. Unfortunately, I don't have a copy
> of the Castle or Acorn C compilers (if Castle were selling the compiler for
> �50 or even �75 I'd buy it, but �200 is a but much).

We have it on a 30% Offer at 138 GBP inc VAT & delivery.

We also just happen to have it in stock;-)


Chris Evans

-- 
CJE Micro's / NCS / Fourth Dimension     'RISC OS Specialists'
Telephone: (01903) 523222                  Fax: (01903) 523679
chris@cjemicros.co.uk              http://www.cjemicros.co.uk/
78 Brighton Road, Worthing, West Sussex, BN11 2EN, UK.

0
chris8168 (2937)
7/2/2003 2:40:32 PM
Well, it looks as if I've hit quite a major problem. I've got Cmunge to
create an assembler file from the CMHG file, but ASM and AS are refusing to
accept the asm file as valid. Is there an assembler - besides objasm - that
will work with Cmunge?

Thanks.
-- 
Phil.
philpem@despammed.com  <<-- This address is valid
http://www.philpem.dsl.pipex.com/
.... Which sparks some mnemonic circuitry.
0
philpem (337)
7/2/2003 3:54:09 PM
Hmm... I've just managed to get Cmunge working - I'm using the "as" assembler
from the latest stable version of GCC.
Now I've got another problem - the C compiler didn't come with any
documentation and I have no idea where it's looking for its headers. I also
don't have a clue what linker to use - will GCC's Drlink work?

Thanks
-- 
Phil.
philpem@despammed.com  <<-- This address is valid
http://www.philpem.dsl.pipex.com/
Microsoft? Is that some form of toilet paper?
0
philpem (337)
7/2/2003 4:00:26 PM
Hi Philip,

> Failing that, can anyone suggest a decent ARM assembler tutorial?

The book, _ARM Assembly Language Programming_, is now available for
viewing online, or for download.

    http://www.peter-cockerell.net:8080/aalp/

Cheers,

-- 
Ralph Corderoy.      http://inputplus.co.uk/ralph/     http://troff.org/
0
ralph6281 (150)
7/2/2003 5:00:49 PM
On Wed, 2 Jul 2003, Philip Pemberton wrote:

> In message <slrnbg5p0o.epi.{tony}@realh.co.uk>
>           Tony Houghton <{tony}@realh.co.uk> wrote:
>
> > It contains complete working versions of the C++ and C compilers (the
> > latter is version 5.06 and can output 32-bit code despite its age)
> Well, that could be useful...
>
> > which
> > can be run from the command line. Everything else you need is available
> > as freeware, all you'll be missing is the GUI front-ends which are of
> > dubious use anyway.
> I know Justin Fletcher has written an open-source alternative to CMHG called
> "Cmunge", so I can probably use that to generate the module headers. As for

To be fair, Robin Watts wrote it and I just maintain it these days.

> the build scripts, David Pilling rather thoughtfully used Obey files instead
> of Acorn Make. In his words -
>   "Honestly, I did once study and understand the Acorn Make program, and I
>    came to the conclusion that it couldn't do everything I needed and would
>    be more complicated to use!"

I've used amu for most tasks on RISC OS and aside from its one or two
limitations it's been suitable for just about everything. Admittedly the
recent updates for functions have been an amazing boon for more generic
makefiles, but these same features are probably also available in the GCC
development tools.

-- 
Gerph {djf0-.3w6e2w2.226,6q6w2q2,2.3,2m4}
URL: http://www.movspclr.co.uk/
.... In my field of paper flowers and candy clouds of lullaby
    I lie inside myself for hours and watch my purple sky fly over me
0
7/2/2003 5:26:34 PM
In message <slrnbg623p.h0k.{tony}@realh.co.uk>
          Tony Houghton <{tony}@realh.co.uk> wrote:

> Drlink
> always did work with Norcroft, but I don't know whether it can manage
> modules, especially 32-bit ones.
I don't see any point in making it 32-bit capable. I heard a rumour that the
Iyonix didn't have a printer port. If that's true, a HP PPA driver is not
likely to be much good to Iyonix users.

> For libraries and headers you can use ROL's StubsG package, the website
> says it's "free".
I've been using RISC_OSLib with GCC to compile Jacob Rief's autorouter
anyway. I'm just cleaning it up and (with a bit of luck) I should be able to
make a GPL- or BSD-licensed release within a few weeks.

As far as SPrinter goes, I've got it compiling, but I'm getting a rather
worrying warning from the C compiler:

Norcroft RISC OS ARM C vsn 5.06 (Acorn Computers Ltd) [May 25 1995]
"c.main", line 278: Warning: out-of-bound offset 11 in address
c.main: 1 warning, 0 errors, 0 serious errors

.... which lands in the following block of code:

/* this is the main dumper entry point */
_kernel_oserror * dumper_entry(_kernel_swi_regs *r)
{
[ BIG SNIP ]

 switch(r->r[11])      <<--- Error here

Now, I know _kernel_swi_regs only covers R0 to R10 (IIRC), so why the devil
was this code written like this? It seems to work fine, but - like I said -
that warning is a bit disconcerting...
 
Thanks.
-- 
Phil.
philpem@despammed.com  <<-- This address is valid
http://www.philpem.dsl.pipex.com/
.... Techs would rather pee on an electric fence for the light show
0
philpem (337)
7/2/2003 5:29:55 PM
On Wed, 2 Jul 2003, Tony Houghton wrote:

> In <ed76c50b4c.philpem@dsl.pipex.com>,
> Philip Pemberton <philpem@despammed.com> wrote:
>
> > Well, it looks as if I've hit quite a major problem. I've got Cmunge to
> > create an assembler file from the CMHG file, but ASM and AS are refusing to
> > accept the asm file as valid. Is there an assembler - besides objasm - that
> > will work with Cmunge?
>
> You should be able to get cmunge to output an AOF instead.

Not without re-writing the output generator. To make things really easy,
CMunge only generates assembler source which is then compiled. If anyone
finds that it doesn't work with other assemblers, please let me know and
we can see what needs updating. The most recent source isn't up on a site
yet, 'cos I'm busy with other things, but if anyone wants a play they can
email me for the sources.

-- 
Gerph {djf0-.3w6e2w2.226,6q6w2q2,2.3,2m4}
URL: http://www.movspclr.co.uk/
.... I believe in you; I'll give everything just to find you.
0
7/2/2003 5:29:58 PM
Philip Pemberton <philpem@despammed.com> wrote:

> As far as SPrinter goes, I've got it compiling, but I'm getting a rather
> worrying warning from the C compiler:
> 
> Norcroft RISC OS ARM C vsn 5.06 (Acorn Computers Ltd) [May 25 1995]
> "c.main", line 278: Warning: out-of-bound offset 11 in address c.main: 1
> warning, 0 errors, 0 serious errors
> 
> ... which lands in the following block of code:
> 
> /* this is the main dumper entry point */ _kernel_oserror *
> dumper_entry(_kernel_swi_regs *r) { [ BIG SNIP ]
> 
>  switch(r->r[11])      <<--- Error here
> 
> Now, I know _kernel_swi_regs only covers R0 to R10 (IIRC), so why the
> devil was this code written like this? It seems to work fine, but - like I
> said - that warning is a bit disconcerting...

The code is clearly faulty, as the array of registers is only 10 entries
long (R0 - R9, inclusive).

Define "seems to work fine".  It is possible that, by pure luck, the other
registers just happen to be in memory after the _kernel_swi_regs structure. 
This can happen if the assembler veneer just stacks all the registers and
passes the address of the stored R0 as the "_kernel_swi_regs *".  However,
that is relying on an implementation detail of the veneer, so unless you are
manually writing the veneer and putting in comments to indicate that you are
relying on the stacked (R10 and) R11 values being in memory above the R9 of
the _kernel_swi_regs, I'd try and find a way of changing that.

This is the sort of code that exhibits the "seems to work fine" state that
then suddenly doesn't work on new machines for no apparent reason.

-- 
Stewart Brodie
0
7/2/2003 8:19:04 PM
Justin Fletcher <justin.fletcher@ntlworld.com> wrote:

> On Wed, 2 Jul 2003, Philip Pemberton wrote:
> 
> > David Pilling rather thoughtfully used Obey files instead of Acorn Make.
> > In his words -
> >   "Honestly, I did once study and understand the Acorn Make program, and
> >    I came to the conclusion that it couldn't do everything I needed and
> >    would be more complicated to use!"

Look, *nobody* understands the !Make program!  What got me was that it was
necessary to have spaces on the end of all the lines of the Makefile.  And
that some of the vital information wasn't stored in the Makefile but inside
!Make itself.  It makes me shudder to even think about the damned program.

> I've used amu for most tasks on RISC OS and aside from its one or two
> limitations it's been suitable for just about everything. Admittedly the
> recent updates for functions have been an amazing boon for more generic
> makefiles, but these same features are probably also available in the GCC
> development tools.

amu is certainly much closer to GNU make now - which was my aim so that we
could use the same Makefiles in both GNU make and amu on RISC OS and on UNIX
systems.  Any similarity between the functionality of amu and that of GNU
make 3.79.1 and that defined by GNU make's documentation is purely
intentional :)

My one regret was never finishing the implementation of the wildcard
function.  It was very complicated, basically by the need to copy with
directory reversals and non-reversals, possibly even in the same wildcard
match.  (i.e.  $(wildcard *.c)  would have to look for anything/c and all
files in a sub-directory called 'c'.)

Curiously enough, amu was rather well designed.  I was able to rip out the
makefile lexer and replace it completely, which fixed the lexing faults and
allowed the use of include directives and conditionals.  Previously, the
code was horrendously complicated - sort of like the code generated by lex,
but worse as it was clearly somebody'd written this all by hand - and with a
sprinkling of bugs.  The macro implementation was very basic too.  After its
total rewrite, it was possible to *reliably* refer to other variables in the
definitions of variables, use environment variables as make variables,
override makefile definitions with command-line definitions and all the
sorts of critical things that other makes rely on.

Those were fun days.


-- 
Stewart Brodie
0
7/2/2003 8:32:07 PM
In <Pine.LNX.4.55.0307021827360.1685@buttercup>,
Justin Fletcher <justin.fletcher@ntlworld.com> wrote:

> On Wed, 2 Jul 2003, Tony Houghton wrote:
> 
>> You should be able to get cmunge to output an AOF instead.
> 
> Not without re-writing the output generator. To make things really easy,
> CMunge only generates assembler source which is then compiled. If anyone
> finds that it doesn't work with other assemblers, please let me know and
> we can see what needs updating. The most recent source isn't up on a site
> yet, 'cos I'm busy with other things, but if anyone wants a play they can
> email me for the sources.

Odd, I'll have to check what's really going on with SmartOpenDir and
WinSnap, because it all seemed to be working OK. Or does it
automatically invoke objasm if you ask it to output a .o file?

-- 
Use Reply-To and DO NOT remove .nospam when replying
0
Tony
7/2/2003 9:36:32 PM
In <8a3bce0b4c.philpem@dsl.pipex.com>,
Philip Pemberton <philpem@despammed.com> wrote:

> In message <slrnbg623p.h0k.{tony}@realh.co.uk>
>           Tony Houghton <{tony}@realh.co.uk> wrote:
> 
>> Drlink
>> always did work with Norcroft, but I don't know whether it can manage
>> modules, especially 32-bit ones.
> I don't see any point in making it 32-bit capable. I heard a rumour that the
> Iyonix didn't have a printer port. If that's true, a HP PPA driver is not
> likely to be much good to Iyonix users.

OK, but then you won't be able to link with StubsG. There are other free
Stubs alternatives though (I used to use one by Mark Wooding
(mdw)/Straylight because of a bug in the official _swi IIRC, but I don't
know where you'd find that now), and the headers with StubsG should
probably still work with them.

Actually, I've just remembered the compiler has its own internal copies
of the shared library headers anyway, and you use -J rather than -I to
override it.

-- 
Use Reply-To and DO NOT remove .nospam when replying
0
Tony
7/2/2003 9:42:04 PM
In message <gemini.3f033e38000dd8fb%stewart.brodie@ntlworld.com>
          Stewart Brodie <stewart.brodie@ntlworld.com> wrote:

> Philip Pemberton <philpem@despammed.com> wrote:
> 
> > /* this is the main dumper entry point */ _kernel_oserror *
> > dumper_entry(_kernel_swi_regs *r) { [ BIG SNIP ]
> > 
> >  switch(r->r[11])      <<--- Error here
> > 
> > Now, I know _kernel_swi_regs only covers R0 to R10 (IIRC), so why the
> > devil was this code written like this? It seems to work fine, but - like I
> > said - that warning is a bit disconcerting...
> 
> The code is clearly faulty, as the array of registers is only 10 entries
> long (R0 - R9, inclusive).
> 
> Define "seems to work fine".  It is possible that, by pure luck, the other
> registers just happen to be in memory after the _kernel_swi_regs structure. 
> This can happen if the assembler veneer just stacks all the registers and
> passes the address of the stored R0 as the "_kernel_swi_regs *".

This is precisely what the assembler veneer in SPrinter does. So in this
particular case I don't think there's any need to worry about the
warning.

Alan

-- 
RISC OS - you know it makes cents
0
spamhater1 (1060)
7/2/2003 10:00:41 PM
Tony Houghton <{tony}@realh.co.uk> wrote:

> In <gemini.3f0341470019ca8a%stewart.brodie@ntlworld.com>, Stewart Brodie
> <stewart.brodie@ntlworld.com> wrote:
> 
> > override makefile definitions with command-line definitions
> 
> Ah, you mean that didn't work in previous versions of amu? That might
> explain some problems I've had.

Correct.  If you makefile contained the like "CC=cc" and then you actually
ran "amu CC=gcc", then previously (i.e. in versions before we started
releasing 32-bit versions of things), an expansion of CC in a command would
give you 'cc'.  The modern version fixes this and you would get 'gcc'.

The makefile could say "override CC=cc" in order to beat your command-line
definition though.  In fact, the new amu supports all 8(iirc?) different
macro "origins" (and the origin function).  These origins are ranked and the
highest ranking definition is that which is used for macro expansion.

There is a command-line switch to revert the new amu to the old (perceived)
rankings (-E IIRC).  I actually just have a two static arrays of the enum
values and use the new one by default and a slightly different one for
backward compatibility.

-- 
Stewart Brodie
0
7/2/2003 10:02:55 PM
In message <8605e70b4c.spamhater@keepyourfilthyspamtoyourself.co.uk>
          Alan Wrigley <spamhater@keepyourfilthyspamtoyourself.co.uk> wrote:

> This is precisely what the assembler veneer in SPrinter does. So in this
> particular case I don't think there's any need to worry about the
> warning.
It certainly looks that way. It does an "STMFD sp!, {r0-r9, sl, fp, ip, lr}"
just before it jumps to the handler. I suppose that code should be cleaned up
a bit - maybe create a new typedef - something like...
  typedef struct {
    int r[14];   /* registers r0-r9, sl, fp, ir, lr */
  } regs_extended;

Anyone want to make a guess at the chances of that little hack actually
_working_?

Later.
-- 
Phil.
philpem@despammed.com  <<-- This address is valid
http://www.philpem.dsl.pipex.com/
Microsoft? Is that some form of toilet paper?
0
philpem (337)
7/2/2003 10:24:11 PM
In message <gemini.3f0341470019ca8a%stewart.brodie@ntlworld.com>
          Stewart Brodie <stewart.brodie@ntlworld.com> wrote:

> Look, *nobody* understands the !Make program!  What got me was that it was
> necessary to have spaces on the end of all the lines of the Makefile.  And
> that some of the vital information wasn't stored in the Makefile but inside
> !Make itself.  It makes me shudder to even think about the damned program.
I use GNU Make. 'Nuff said :)
I basically have a basic template that I modify to suit the program I'm
writing. Say I want to make a Desklib based WIMP program. OK, copy the
Desklib template, change a few constants and I can start writing code. It
certainly saves a fair bit of time.

Later.
-- 
Phil.
philpem@despammed.com  <<-- This address is valid
http://www.philpem.dsl.pipex.com/
.... Techs would rather pee on an electric fence for the light show
0
philpem (337)
7/2/2003 10:28:09 PM
On 2 Jul 2003 Stewart Brodie <stewart.brodie@ntlworld.com> wrote:

> Alan Wrigley <spamhater@keepyourfilthyspamtoyourself.co.uk> wrote:
> 
> > In message <gemini.3f033e38000dd8fb%stewart.brodie@ntlworld.com>
> >           Stewart Brodie <stewart.brodie@ntlworld.com> wrote:
> > 
> > > Philip Pemberton <philpem@despammed.com> wrote:
> > > 
> > > > /* this is the main dumper entry point */ _kernel_oserror *
> > > > dumper_entry(_kernel_swi_regs *r) { [ BIG SNIP ]
> > > > 
> > > >  switch(r->r[11])      <<--- Error here
> > > > 
> > > > Now, I know _kernel_swi_regs only covers R0 to R10 (IIRC), so why the
> > > > devil was this code written like this? It seems to work fine, but -
> > > > like I said - that warning is a bit disconcerting...

[snip]

> Is it well-commented?  If not, I'd strongly advise it was - in BOTH the C
> and assembler code.
> 
> To get rid of the warning, you should be able to change the expression to:
> 
>   &r->r[0] + 11
> 
> However, it isn't always easy to fool it - it depends if the compiler
> manages to spot the hack ;-)

Eeek! less readable. It would be better to change the type of r to
a more appropriate type. _kernel_registerset* if 16 registers are required.

---druck

-- 
The ARM Club Free Software - http://www.armclub.org.uk/free/
The 32bit Conversions Page - http://www.quantumsoft.co.uk/druck/
0
news5843 (7461)
7/2/2003 11:45:48 PM
Hi Stewart,

> > > >  switch(r->r[11])      <<--- Error here
>
> To get rid of the warning, you should be able to change the expression
> to:
>
>  &r->r[0] + 11

That should be

    *(&r->r[0] + 11)

or

    *(r->r + 11)

if that still manages to fool the compiler.

Personally, I'd cast r->r to be a pointer so the array bounds aren't
checked.  It's a compile-time op. with no run-time cost in this case.
Something like

    int *reg16;

    reg16 = r->r;    /* We've ensured that all regs are present. */
    switch (reg16[11]) {

Cheers,

-- 
Ralph Corderoy.      http://inputplus.co.uk/ralph/     http://troff.org/
0
ralph6281 (150)
7/3/2003 9:21:17 AM
ralph@inputplus.co.uk (Ralph Corderoy) wrote:

> Hi Stewart,
> 
> > > > >  switch(r->r[11])      <<--- Error here
> >
> > To get rid of the warning, you should be able to change the expression
> > to:
> >
> >  &r->r[0] + 11
> 
> That should be
> 
>     *(&r->r[0] + 11)

Yes, I missed the dereference.

> 
> or
> 
>     *(r->r + 11)
> 
> if that still manages to fool the compiler.

Nooooo - it's not *that* easy :-)  IIRC, the compiler should remember the
size of object that r->r points to and thus still be able to check the
bounds.  This isn't as surprising as it might seem since the compiler will
treat "r->r[11]" as identical to "*(r->r+11)".  The standard even describes
that the [] notation is purely syntactic sugar.

It's also leads down the path to one of the truly evil constructs in C which
has extremely few uses unless the pointer is generated by a very complex
expression, and then you should probably do it differently:

a[b] is identical to *(a+b) by definition

*(a+b) is identical to *(b+a) by semantics of addition operator

*(b+a) is identical to b[a] by the inverse of the first definition.

Therefore, a[b] is identical to b[a].

Therefore, you *could* try "11[r->r]" and see if that confuses the compiler
(but I don't think it will ;-)


> Personally, I'd cast r->r to be a pointer so the array bounds aren't
> checked.  It's a compile-time op. with no run-time cost in this case.
> Something like
> 
>     int *reg16;
> 
>     reg16 = r->r;    /* We've ensured that all regs are present. */
>     switch (reg16[11]) {

Yes, unless the information about the real size of the object pointed to is
also copied by the compiler to its list of known properties of 'reg16' when
the assignment is done.  You could try a straight cast to "int *".


-- 
Stewart Brodie
0
7/3/2003 11:33:35 AM
Hi Stewart,

> bounds.  This isn't as surprising as it might seem since the compiler
> will treat "r->r[11]" as identical to "*(r->r+11)".  The standard even
> describes that the [] notation is purely syntactic sugar.

Yes, many people who write C aren't familiar with the equivalence
between arrays and pointers.  That is sometimes the reason why

    &buf[0] + FIXED_OFFSET

is written when

    buf + FIXED_OFFSET

would be more idiomatic and thus clearer for the reader.  It works since

        &buf[0]
     == &(*(buf + 0))
     == &(*(buf))
     == buf

Afterall, those same people don't write `i + 0 - 0' instead of `i'.

> It's also leads down the path to one of the truly evil constructs in C
> which has extremely few uses unless the pointer is generated by a very
> complex expression, and then you should probably do it differently:

Even with a complex expression some terms can normally be swapped to get
a `base' pointer before the `['.

One place where it's sometimes seen is a quick conversion from integer
to character, e.g. instead of writing

    "0123456789abcdef"[i]

some dislike seeing the string before the `[' and write

    i["0123456789abcdef"]

I think they're barking;  there's nothing wrong with the former  :-)

Cheers,

-- 
Ralph Corderoy.      http://inputplus.co.uk/ralph/     http://troff.org/
0
ralph6281 (150)
7/3/2003 6:37:32 PM
I demand that Alan Wrigley may or may not have written...

[snip]
> Incidentally, I can't see how the code is going to work if an error is
> returned. After your C routine has returned, you get:

>   ; Save the returned value in r8
>   MOVS    r8, r0
>   ; Get the stuff off the stack
>   LDMFD   sp!, {r0-r9, sl, fp, ip, lr}
>   ; If returned value indicates an error, then put it back in r0
>   MOVNE   r0, r8

> which is going to result in your error block pointing to $deity knows
> where. [...] My own solution would be to use R9 as your work register
> instead of R8, store R0-R8 instead of R0-R9, [...]

  TEQ     r0,#0
  STRNE   r0,[sp]
  LDMFD   sp!,{r0-r9,sl,fp,ip,lr} ; or ...ip,pc}

-- 
| Darren Salt | d youmustbejoking,demon,co,uk | nr. Ashington,
| RISC OS,    | s zap,tartarus,org            | Northumberland
| Linux       | @                             | Toon Army
|   <URL:http://www.youmustbejoking.demon.co.uk/> (PGP 2.6, GPG keys)

When working a problem, it helps to know the answer.
0
news64 (1253)
7/3/2003 11:43:28 PM
In message <gemini.3f052f1e00105de1%stewart.brodie@ntlworld.com>
          Stewart Brodie <stewart.brodie@ntlworld.com> wrote:

> Philip Pemberton <philpem@dsl.pipex.com> wrote:
> 
> > Stewart,
> >   Just out of interest, are you the same Stewart Brodie that wrote ArcWeb?
> Yes,
Would you mind contacting me off-group? I'm one of the Netsurf developers and
I'd like to discuss the possibility of borrowing some of Arcweb's code for it
(I'm developing the GIF loader for NS).
For my current valid email address, replace "despammed.com" with
"dsl.pipex.com". Despammed is a filter and it does tend to be a bit - well -
over-enthusiastic...

> > Your email address seems to be bouncing...
> 
> Which one:  ntlworld.com? pace.co.uk? metahusky.net?
NTLWorld. Looks like it's just Pipex's server being stupid though - it just
bounced another message and it still managed to get to the recipient's
server...

> I rarely read any e-mail to my ntlworld.com account, so it's possible it
> might have filled up.  It's a sort of sacrificial newsgroup account that
> only ever attracts spam.
A bit like my BTOpenworld one then ;-)

Later.
-- 
Phil.
philpem@despammed.com  <<-- This address is valid
http://www.philpem.dsl.pipex.com/
.... You may be a tech if you're entertained by a 6-pack and sparking HV.
0
philpem (337)
7/4/2003 7:47:52 AM
Reply:

Similar Artilces:

Programmers, Programmers, Programmers, ...
As Steve Balmer correctly stated, while making his monkey dance, it is applications and hence programmers that make a platform. The fact though is that if you want to do professional programming, then Linux is the platform for you. I know that this statement will get the heckels up on a lot of trolls in C.O.L.A, but I have a recent experience that proves this. I am currently working for a Windows only house producing a system that receives and transmits around 1000 telegrams per second in each direction on a UDP socket, translates them into a different format and creates a log entry for each ...

problem with reload(sys) (doing reload on the sys module)
Python 2.4.2 (#67, Sep 28 2005, 12:41:11) [MSC v.1310 32 bit (Intel)] on win32 Type "copyright", "credits" or "license()" for more information. **************************************************************** Personal firewall software may warn about the connection IDLE makes to its subprocess using this computer's internal loopback interface. This connection is not visible on any external interface and no data is sent to or received from the Internet. **************************************************************** IDLE 1.1.2 >&g...

imputils
Hello, I've created an archive import class (with the help of importers.py from the python /demo/imputil dir) which imports packages and modules from a dictionary. So far so good. The problem is that every import statement after the code: imputil.ImportManager().install() sys.path.insert(0, DictImporter(archive)) sys.path.insert(0, imputil.BuiltinImporter()) tries to import from the dictionary and doesn't use the modules that were imported before. For example, I would like to use os in a module that's stored in the dictionary. The import os statement tries to import os from the dictionary and does not 'use' the already imported os module that is in the modules list. Is it possible to get access to both, already loaded modules in the modules list and modules in the dictionary? Regards, Remy Cool Problem solved!!! Switched to python 2.3 and created a class using the brand new sys.path_hooks. See python PEP302 for more info on this subject. Remy > Hello, > > I've created an archive import class (with the help of importers.py from > the python /demo/imputil dir) which imports packages and modules from a > dictionary. So far so good. > > The problem is that every import statement after the code: > > imputil.ImportManager().install() > sys.path.insert(0, DictImporter(archive)) > sys.path.insert(0, imputil.BuiltinImporter()) > > tries to import from the dictionary and doesn't use the modules th...

gcc-3.0 <-> gcc-2.95 usage for kernel/modules
Hi, i noticed that there is some problem in using a mixture of gcc-3.0 and gcc-2.95 in kernel space. My system runs a vanilla kernel compiled with gcc-2.95. A module i'm currently working on needs to be compiled with gcc-3.0. Everything works fine except scheduling using sleep_on and wake_up which crashes utterly. Using a kernel compiled with gcc-3.0 solves this problem. Is there some trick appart from porting my code to 2.95? Mit freundlichen Gruessen / Best regards Dipl.-Phys. Christoph Baumann --- SORCUS Computer GmbH Im Breitspiel 11 c D-69126 Heidelberg Tel.: +...

GCC
unsigned char numbers[8]= {0b10000000, 0b01000000, 0b00100000, 0b00010000, 0b00001000, 0b00000100, 0b00001000, 0b00010000}; My 15 year old C compiler works with this, why doesnt GCC !!! Anyone know how I can do this without pages of hex ?? On Mon, 24 Nov 2003 20:37:28 +0000, root wrote: > > unsigned char numbers[8]= > {0b10000000, > 0b01000000, > 0b00100000, > 0b00010000, > 0b00001000, > 0b00000100, > 0b00001000, > 0b00010000}; > > My 15 year old C compiler works with this, why doesnt GCC !!! Because it is NOT standard C. > Anyone know how I can do this without pages of hex ?? If you can't do this by heart, you should really practice your bin2hex conversion. Any programmer should be able to convert this in seconds. Especially one that has been programming for 15 years. Sybren -- (o_ Q: God, root, what is difference? //\ A: God can change the byte order on the CPU, root can't. V_/_ root wrote: > > > unsigned char numbers[8]= > {0b10000000, > 0b01000000, >...

RISCOS/Acorn Eprom programmer
Hi, Does anyone know where I can get an EPROM programmer for a RISCOS 3.11 A5000? Must blow up to 256Kbit bog standard ST EPROMs. Cheers all, - Chris. In message <3fdd0a07$0$10357$fa0fcedb@lovejoy.zen.co.uk> "Chris Smith" <chris@FARPinfinitemonkeys.org.uk> wrote: > Hi, > > Does anyone know where I can get an EPROM programmer for a RISCOS 3.11 > A5000? Must blow up to 256Kbit bog standard ST EPROMs. Did you try IFEL's web site ? I think they were selling one not so long ago <http://users.argonet.co.uk/ifel> (or there abouts) -- J�r�me Mathevet On Mon, 15 Dec 2003 01:10:31 -0000, "Chris Smith" <chris@FARPinfinitemonkeys.org.uk> wrote: >Hi, > >Does anyone know where I can get an EPROM programmer for a RISCOS 3.11 >A5000? Must blow up to 256Kbit bog standard ST EPROMs. What'd you be willing to pay ? John Kortink -- Email : kortink@inter.nl.net Homepage : http://www.inter.nl.net/users/J.Kortink ViewFinder, the high performance graphics card for RISC PC's : visit http://www.windfall.nl for more details and pricing. ...

sys.modules and __main__ obscureness
Hello all, I am messing with namespaces, so that code I exec thinks it is executing in the __main__ module. I have the following code : import imp import sys # can't call the module '__main__' or 'new_module' returns the real one module = imp.new_module('_') namespace = module.__dict__ namespace['__name__'] = '__main__' # next put things into the names # e.g. : namespace['variable'] = 3 sys.modules['__main__'] = module print module print namespace import __main__ print __main__.__dict__['variable'] This code behaves di...

Why there is no "setdefaultencoding" in sys module?
Hi, everyone I'm a new hand at python. I tried to set system default encoding by using "import sys; sys.setdefaultencoding('utf-f')", but I got error message: >>> sys.setdefaultencoding('utf-8') Traceback (most recent call last): File "<stdin>", line 1, in <module> AttributeError: 'module' object has no attribute 'setdefaultencoding' Then I checked dir(sys), seems there was no function named "setdefaultencoding" in "sys" module. But in python's document, it said I s...

Comments on using Sys::PortIO module
I am working on a home automation project. As part of my project, I need to do the following; - control 4 irrigation zones - control the fan motor in the furnace - monitor the position (locked/unlocked) of the dead bolts on specific doors - monitor garage door position (close if necessary) - monitor rain sensor - monitor HVAC with a home brew system (heating, cooling and/or fan on) - monitor certain doors for open or closed position I have built a I/O card based on the Intel 8255 PPI chip (http://www.boondog.com) and will be using it on a RedHat Fedora 2 system. The card was tested wit...

Writing a Module Using Sourcery And GCC
Hiya apologies in adavnce for any stupidity included within I have the lastest versions of gcc and cmunge with sourcery and are attempting to write a module. My current setup works well with gcc for compiling applications in C. When i try and use it to build a module i get no end of errors and unusual problems occuring. Has anyone got either of the following they can send me; (zip or spark of entire module source would be cool) 1) A module wirtten in C that is a sourcery project 2) A module written in C that compiles with gcc using a makefile or similar. Essentially so i can look at setup and compare it to mine. Thanks in advance Tim Tim Wright wrote: > Hiya apologies in adavnce for any stupidity included within > > I have the lastest versions of gcc and cmunge with sourcery and are > attempting to write a module. > My current setup works well with gcc for compiling applications in C. > > When i try and use it to build a module i get no end of errors and > unusual problems occuring. Without knowing any of the details of the errors, there isn't much we can do. > > Has anyone got either of the following they can send me; (zip or spark > of entire module source would be cool) > > 1) A module wirtten in C that is a sourcery project > > 2) A module written in C that compiles with gcc using a makefile or > similar. GCC comes with a number of examples, including sample module code. Thanks Paul, I am writing a mod...

RISC OS modules with stock gcc?
Does anyone know if it's possible to persuade gcc & ld to produce sensible RISC OS modules? My experimentation is showing that there appear to be two major choices with gcc: firstly, generating binaries with absolute addresses baked into them, which of course doesn't work; and secondly, to generate binaries that look up all data via the GOT, which of course I don't have (and, I suspect, requires a lot of setup). I don't need anything pretty, merely something that works. I notice that the latest gccsdk seems to support doing this; is this custom hackery or a stock gcc feature? (Note that I *don't* want to use gccsdk if I can help it. Mainly because I'm cross-compiling and would very much rather use a stock gcc for simplicity.) -- ┌─── dg@cowlark.com ───── http://www.cowlark.com ───── │ │ ⍎'⎕',∊N⍴⊂S←'←⎕←(3=T)⋎M⋏2=T←⊃+/(V⌽"⊂M),(V⊝"M),(V,⌽V)⌽"(V,V←1⎺1)⊝"⊂M)' │ --- Conway's Game Of Life, in one line of APL On Sun, 12 Oct 2008 22:39:04 +0100 David Given <dg@cowlark.com> wrote: > I notice that the latest gccsdk seems to support doing this; is this > custom hackery or a stock gcc feature? > > (Note that I *don't* want to use gccsdk if I can help it. Mainly > because I'm cross-compiling and would very much rather use a stock > gcc for simplicity.) You can't. Stock GCC doesn't support the features RISC OS requires (such as chunked stacks in the ARM backend.) It's ...

Experimental GCC now with module support
[Posted and mailed] Nick Burrett has now largely completed the somewhat naive attempt I had at putting module support in GCC. So far, the results are encouraging. I've put up a test version: http://gccsdk.riscos.info/downloads/testing/ There are quite a number of caveats, and in time, these will be addressed properly. In the meantime, to try this out (and please do), you will need to: Compile your source code to objects with something like: gcc -mlibscl -mmodule -c mymodule.c (Exactly the same with assembler) Link explictly with Castle C/C++ stubs or ROL StubsG, as GCC's stubs do not have module support: drlink -rmf mymodule.o C:stubs.o -o mymodule Potential missing symbols: __gcc_vsnprintf - put #undef vsnprintf after #include <stdio.h> |LC..xx| - Try recompiling the C file in question without optimisation others - Possibly maths symbols from libgcc.o, which we will need to do a module version of. No C++ modules yet - will need to do module versions of the C++ runtime libraries. If you have any other problems not listed here, we'd be interested to know. -- Peter Naulls - peter@chocky.org | http://www.chocky.org/ ---------------------------------------------------------------------------- Unix Programs on RISC OS | http://www.riscos.info/unix/ ...

GCC support for RISC OS modules
Hope posting this to csa.misc is OK, I'm not after a deeply techy discussion. Is this new feature of GCC 3.4.4 Release 3 something of a final piece in the jigsaw for using GCC with RISC OS, or do I misunderstand? -- Cheers, Guy guy@use.steelpillow.less.com To reply, remove useless bits. On 8 Jul 2005 Guy Inchbald <guy@efjdolkhvgf.com> wrote: > Hope posting this to csa.misc is OK, I'm not after a deeply techy > discussion. csa.programmer is the place for any kind of programming discussion. > Is this new feature of GCC 3.4.4 Release 3 something of a final piec...

gcc compiled perl runtime module
Hi together one of my customers want to have a perl runtime which was compiled with gcc. He isn't able to do so himself, because he got many errors compiling the perl sources. Is there any place where I can download a gcc-precompiled perl.rte. Alternatively can someone post the options to compile the perl sources with gcc ? Thanks in advance Friedhelm In article <g0s36s$l1e$1@online.de>, Friedhelm Neyer <Friedhelm.Neyer@t-online.de> writes: > Hi together > > one of my customers want to have a perl runtime which was compiled with > gcc. He isn't able to...

gcc is including all modules in libraries I link against
gcc 3.4.0 AIX 5.2 My company has a large application we normally build with xlC. I'm charged with porting it to the GNU toolchain in preparation to porting it to other architectures. I built a gcc 3.4.0 bootstrapped from a gcc 3.3.3 I downloaded. I've built binutils 2.15, but I can't get anything linked with it, so I'm still using the native linker. I have a two-fold problem. 1) The table of contents is > 64K, and according to their respective documentations, all options that either make the TOC bigger in ld, or put less in the TOC in gcc, make the program run slower....

Supply Mini-Com Coupler Modules ,Mini-com Coupler Modules,Panduit Mini-com Coupler Module,Panduit Minicom Modules,Panduit Mini-Com Module,Panduit Mini-com Fiber Optic Adapter Modules
Original:China Supply Mini-Com Coupler Modules ,Mini-com Coupler Modules,Panduit Mini-com Coupler Module,Panduit Minicom Modules,Panduit Mini-Com Module,Panduit Mini-com Fiber Optic Adapter Modules Mini-Com TX6 Plus Jack Module TP Series,Mini-Com Coupler Modules ,Mini-com Coupler Modules,Panduit Mini-com Coupler Module,Panduit Minicom Modules,Panduit Mini-Com Module,Panduit Mini-com Fiber Optic Adapter Modules,Panduit Mini-Com ST Fiber Optic Adapter Modules,Panduit Mini-Com SC Fiber Optic Adapter Modules,Panduit Mini-Com FC Fiber Optic Adapter Modules,Panduit Mini-Com Surface Mount Boxes,Fiber Optic Surface Mount Boxes,Fiber Surface Mounted Boxes Surface mount box accepts four Mini-Com modules. See More Details at the Link: http://www.cheerwe-telecom.com http://indonetwork.net/cheerwetelecom Category 5e, RJ45, 8 position, 8 wire universal coupler module. Features: Designed to exceed performance requirements of the TIA/EIA-568-B.2 Category 5e and ISO 11801 2nd Edition Class D channel standards RJ45 8 position, 8 wire female to RJ45 8 position, 8 wire female pass through module No termination tool required Accepts 6 and 8 position modular plugs without damage Can be clearly identified with optional labels and icons Not suitable for surface mount box applications ...

Supply Mini-Com Coupler Modules ,Mini-com Coupler Modules,Panduit Mini-com Coupler Module,Panduit Minicom Modules,Panduit Mini-Com Module,Panduit Mini-com Fiber Optic Adapter Modules
Original:China Supply Mini-Com Coupler Modules ,Mini-com Coupler Modules,Panduit Mini-com Coupler Module,Panduit Minicom Modules,Panduit Mini-Com Module,Panduit Mini-com Fiber Optic Adapter Modules Mini-Com TX6 Plus Jack Module TP Series,Mini-Com Coupler Modules ,Mini-com Coupler Modules,Panduit Mini-com Coupler Module,Panduit Minicom Modules,Panduit Mini-Com Module,Panduit Mini-com Fiber Optic Adapter Modules,Panduit Mini-Com ST Fiber Optic Adapter Modules,Panduit Mini-Com SC Fiber Optic Adapter Modules,Panduit Mini-Com FC Fiber Optic Adapter Modules,Panduit Mini-Com Surface Mount Boxes,Fiber Optic Surface Mount Boxes,Fiber Surface Mounted Boxes Surface mount box accepts four Mini-Com modules. See More Details at the Link: http://www.cheerwe-telecom.com http://indonetwork.net/cheerwetelecom Category 5e, RJ45, 8 position, 8 wire universal coupler module. Features: Designed to exceed performance requirements of the TIA/EIA-568-B.2 Category 5e and ISO 11801 2nd Edition Class D channel standards RJ45 8 position, 8 wire female to RJ45 8 position, 8 wire female pass through module No termination tool required Accepts 6 and 8 position modular plugs without damage Can be clearly identified with optional labels and icons Not suitable for surface mount box applications ...

Supply Mini-Com Coupler Modules ,Mini-com Coupler Modules,Panduit Mini-com Coupler Module,Panduit Minicom Modules,Panduit Mini-Com Module,Panduit Mini-com Fiber Optic Adapter Modules
Original:China Supply Mini-Com Coupler Modules ,Mini-com Coupler Modules,Panduit Mini-com Coupler Module,Panduit Minicom Modules,Panduit Mini-Com Module,Panduit Mini-com Fiber Optic Adapter Modules Mini-Com TX6 Plus Jack Module TP Series,Mini-Com Coupler Modules ,Mini-com Coupler Modules,Panduit Mini-com Coupler Module,Panduit Minicom Modules,Panduit Mini-Com Module,Panduit Mini-com Fiber Optic Adapter Modules,Panduit Mini-Com ST Fiber Optic Adapter Modules,Panduit Mini-Com SC Fiber Optic Adapter Modules,Panduit Mini-Com FC Fiber Optic Adapter Modules,Panduit Mini-Com Surface Mount Boxes,Fiber Optic Surface Mount Boxes,Fiber Surface Mounted Boxes Surface mount box accepts four Mini-Com modules. See More Details at the Link: http://www.cheerwe-telecom.com http://indonetwork.net/cheerwetelecom Category 5e, RJ45, 8 position, 8 wire universal coupler module. Features: Designed to exceed performance requirements of the TIA/EIA-568-B.2 Category 5e and ISO 11801 2nd Edition Class D channel standards RJ45 8 position, 8 wire female to RJ45 8 position, 8 wire female pass through module No termination tool required Accepts 6 and 8 position modular plugs without damage Can be clearly identified with optional labels and icons Not suitable for surface mount box applications ...

Module import via sys.path and py2exe
Hi, i have a program with is built like this: startup.py dir1/__init__.py dir1/file1.py dir2/__init__.py dir2/file2.py dir3/__init__.py dir3/file3.py The program works but now i want to add the functionality into an existing program 2. This is program 2: program2.py program2dir1/__init__.py program2dir1/file4.py When i copy the directories over i get this situation: program2.py program2dir1/__init__.py program2dir1/file4.py program1/__init__.py program1/dir1/__init__.py program1/dir1/file1.py program1/dir2/__init__.py program1/dir2/file2.py program1/dir3/__init__.py program1/dir3/file3.p...

Appending to sys.path during module install with distutils
I'm trying to use distutils to install a collection of modules in /usr/local/lib/python2.7/site-packages. My distribution (Fedora 15) doesn't include any /usr/local paths in sys.path, so the import fails when running the program. The distutils documentation suggests adding a $NAME.pth file to an existing site-packages directory in sys.path. Is there a preferred/accepted method of doing this? I considered just adding some code to my setup.py to generate a braindump.pth file containing something like: PREFIX/lib/pythonMAJOR.MINOR/site-packages and then walking the existing sys.path and picking one of those site-packages directories to install braindump.pth to. I'm not sure how to determine which is the appropriate path. Maybe I'm going about this completely wrong as well - anyone care to help steer me in the right direction? The project is located here: http://braindump.dvhart.com in case anyone wants it for reference. Thanks, -- Darren Hart ...

Problem with using GCC/CMunge for modules with C++
Hi. I've read through the documentation, but still haven't found the answer, so I'm trying my luck here: I've installed GCC, and when testing for example the "Simple" module in the Examples.Module directory, it compiles and runs just fine. I've then modified the makefile slightly to use g++ instead of gcc, and saved gccmodule in a "cc" directory, and added code to call a couple of overloaded functions from gccmodule_init() (to test if it's really compiled as C++), printing some text, which still works. However, if I try to use any of the C++ standard library there, I run into problems... For example, adding this code to the top of gccmodule.cc: #include <vector> std::vector<int> v; gives this error: gccmodule.o: In function '__static_initialization_and_destruction_0(int, int): gccmodule.cc:(.text+0xd8): undefined reference to 'operator delete(void*)' collect2: ld returned 1 exit status Has anyone been able to use the C++ standard library in GCC modules? If so, then I'd appreciate either a working example, or some pointer to how I might fix this problem... Regards, Terje On 27/08/2011 19:03, Terje Sletteb� wrote: > Has anyone been able to use the C++ standard library in GCC modules? Thinking back to ANSILib [*]; if it is a standalone bit of library code that links to the executable (as opposed to SharedCLib), perhaps you need a version compiled as module code? Best wishes, Rick. * -...

Displayed area module, Graphic annotation module and Graphic layer module
Hi, 1. Displayed area module, Graphic annotation module and Graphic layer module are used in Presentation State. Can I use these modules in regular DICOM Images (with no presentation state)? If not, How can I represent text value on a Image (Curves and regular Overlays are not good for text values)? 2. Which tag is used for "inverse" operation? Thanks, Erez. erez bibi wrote: > Displayed area module, Graphic annotation module and Graphic layer > module are used in Presentation State. > Can I use these modules in regular DICOM Images (with no presentation > state)? No. > If not, How can I represent text value on a Image (Curves and regular > Overlays are not good for text values)? By sending a presentation state ! That is what it was designed for. > 2. > Which tag is used for "inverse" operation? There is no such attribute in most images (though one could lie about Photometric Interpretation, but this is not a good idea and won't work for many SOP Classes). Within a presentation state, one can use a Presentation LUT Shape (2050,0020) of INVERSE rather than IDENTITY. david ...

Is there a Perl Module The Does GCC like precompile on C++ files ?
Is there a Perl Module The Does GCC like precompile on C++ files ? I am not seeing anything on CPAN ... but maybe my coffee has not kicked in yet :-) Cheers, Jim www.gomba.ca Random Task <Random@Task.be> wrote: > Is there a Perl Module The Does GCC like precompile on C++ files ? If you mean a preprocessor, then see the -P switch in perldoc perlrun -- Tad McClellan SGML consulting tadmc@augustmail.com Perl programming Fort Worth, Texas Tad McClellan <tadmc@augustmail.com> wrote in comp.lang.perl.misc: > Ra...

programms for engineer Pro Engineer Wildfire wuth moduls
Cristmans sales!!!!!!! more information for order: sw_2003@list.ru Top sales: Solidworks 2004 =3cd-rs= USD 100 NEW - USD 70 !!!!!!! Solidworks 2003 SP4 =3cd-rs= USD 100 - USD 60 !!!!!!! Solid Edge 14 =2 cd-rs= USD 80 - USD 50 !!!!!!! ANSYS 8 USD 100 NEW - USD 50 !!!!!!! Ideas 10 =9cd-rs= USD 200 - USD 100 !!!!!! Pro Engineer 2003 Wildfire =2cd-rs= USD 100 USD 70 !!!!!! Pro Engineer library 2003 USD 75 USD 50 !!!!!! ArchiCAD 8.2 ...

Errant Move of /sys/module/ in Fedora Core 3
Through some misunderstanding, I believed I wanted to move /sys/module/parport to /hold/parport in order to keep the parallel port drivers from being loaded. The move completed but some warnings (msgs anyway) were issued. I do not recall what they were, possibly about links. What functionality did I actually lose? If I now move the parport directory back from /hold/parport, should I expect whatever functionality was lost to be restored properly? As far as I can tell, no harm came from the move, but I really don't use the parallel port for the printer anyway. parport_pc was mo...

Web resources about - Modules with GCC - possible? - comp.sys.acorn.programmer

Brussels attack: Friends get ‘notification’ on Infosys man missing after blasts
External Affairs Minister Sushma Swaraj said on Thursday that Ganeshan’s last phone call had been tracked to the metro rail.

That was quick: New iOS build fixes Apple activation bug
iOS 9.3 has a serious bug , which is not exactly a surprise something like this seems to happen every release. Some owners of older iToys complained ...

Earl Hamner Jr., Creator Of The Iconic 70s TV Show ‘The Waltons,’ Dies At 92
Earl Hamner Jr., writer and creator of the iconic 1970s TV show, The Waltons , has died. He was 92. Hamner took his Depression-era upbringing ...

'Justice League,' 'X-Men' Comics Worth Over $200K Stolen
Rare "Justice League" and "X-Men" comic books valued at more than $200,000 have been stolen from a central Georgia store

It’s Netflix that’s throttling your video, not AT&T or Verizon
It turns out that neither AT&T nor Verizon are throttling Netflix videos, as T-Mobile CEO John Legere suggested last week. If you’re an AT&T ...

Top Twitter users throw a surprise 10th birthday party — Biz Please No. 10
Twitter celebrated its 10th anniversary this week, and we here at Biz Please wanted to get our favorite microblogging social network platform ...

Former Bosnian Serb leader Karadzic convicted of genocide, gets 40-year sentence
The Nation Former Bosnian Serb leader Karadzic convicted of genocide, gets 40-year sentence The Denver Post Bida Smajlovic, 64, a survivor ...

5 useful Gmail and Google Apps add-ons for Firefox
Chrome may not be your primary browser, but you can still get in on the Google goodness. Firefox’s famous add-on support lets you integrate many ...

Ole Miss Student Pleads Guilty To Putting A Noose Around Civil Rights Statue
A former University of Mississippi student could face up to a year in prison after pleading guilty to desecrating a civil rights statue. Austin ...

iPhone SE reviews are glowing as device now available for pre-order
The iPhone SE unveiling may not have been accompanied by the type of fanfare that usually drapes iPhone announcements, but make no mistake about ...

Resources last updated: 3/25/2016 11:41:42 AM