Duty calls. Can someone pick up the summary for this and the next couple of
weeks? Email Simon for details. Thanks,
--
Bryan C. Warnock
[EMAIL PROTECTED]
In the following list, "I" is an op which is implemented but not documented;
"D" is an op which is documented but not implemented. I suggest we clean up
the "I"s first. :)
call_method D
can D
chopn_s_ic I
clear_eh D
clear_i I
clear_n I
clear_p I
clear_s I
dec_n I
dec_n_nc I
end I
eq_i_ic
Simon Cozens sent the following bits through the ether:
> eq_i_ic I
Which reminds me, we currently have:
eq_i_ic4 I I D D
Do we really need the extra D? I'm generating twice as many labels for
little gain. ISTR Dan saying it slipped in somehow. May I suggest it
slips out again ;
Here are the costs in seconds for running 100 million iterations of each
op. This only really makes sense on *my* machine, but it gives you an
idea of what sort of tradeoffs there are. Given that the difference between
most ops is negligible, I'd suggest that a lot of work is still happening
in th
I've commited a change which allows local labels to be used in parrot.
The labels are local relative to the last non-local label defined (i.e.
local labels are forgotten when a non-local is defined).
Here's my test program:
main: print "test 1\n"
branch $ok
$ng:p
On Sun, 16 Sep 2001, Simon Cozens wrote:
> In the following list, "I" is an op which is implemented but not documented;
> "D" is an op which is documented but not implemented. I suggest we clean up
> the "I"s first. :)
A lot of those Is are really documented. This, for example:
> le_i_ic I
wo
Yeah, I ran a test where I substituted a huge switch statement with all of
the operations inlined for the function calls and I went from 27 seconds to
21 seconds on the benchmark.
-Original Message-
From: Simon Cozens
To: [EMAIL PROTECTED]
Sent: 9/16/2001 10:01 AM
Subject: Op cost models
Thanks. Applied.
On Sat, 2001-09-15 at 20:56, Buggs wrote:
> Hoi,
>
> Just not to lie.
>
> Buggs
>
> Index: little_languages/jakoc
> ===
> RCS file: /home/perlcvs/parrot/little_languages/jakoc,v
> retrieving revision 1.1
> diff -
Changes from the last one:
* some tidying in the assembly
( now uses set I4, 4 instead of set_i_ic I4, 4 )
* moved lib/Test/Parrot.pm to Parrot/Test.pm
* now run with perl t/harness
Regards
Mattia
diff -r -b -u -2 -N parrot.cvs/Parrot/Test.pm parrot/Parrot/Test.pm
--- parrot.cvs/Parrot/Tes
Dan --
> > In working to make Jako be able to handle while (i < 0) as well as
> > while (i < n), I added a bunch of integer and numerical comparison
> > ops. I also tinkered with the assembler's op inferencing to get it
> > to pick the ops I intended.
>
> Please back these out. I'd rather not ha
All --
Now that the reg-const comparison ops are gone from Parrot, I've
updated jakoc to use pseudo-ops. It uses a naiive implementation
where registers [INPS]0 are reseved for use as temporaries and it
reloads the register each time before the comparison.
Of course, the optimizer can remove tha
All --
Anyone care to post a subcall.pasm example file that shows the
implementation of a subroutine and a call to it? I was thinking
of starting from euclid.pasm (since it has two args), but I'm
not sure I understand what the calling convention is really
supposed to look like...
Regards,
-- G
Another point:
I added the -O3 on each test:
function dispatch switch statement%speedup
-O0 27s21s 30
-O3 16s12s 25
AS you can see -O3 helped out quite a bit g
I don't think anyone objects to this...I vote go ahead and move the .h files
and modify Makefile.in according to Brent's email.
Tanton
-Original Message-
From: Gregor N. Purdy
To: Gibbs Tanton - tgibbs
Cc: 'Robert Spier '; '[EMAIL PROTECTED]"'
Sent: 9/14/2001 7:05 PM
Subject: [propo
On Sun, Sep 16, 2001 at 12:28:50PM +0200, Mattia Barbon wrote:
Content-Description: Mail message body
> Changes from the last one:
> * some tidying in the assembly
> ( now uses set I4, 4 instead of set_i_ic I4, 4 )
> * moved lib/Test/Parrot.pm to Parrot/Test.pm
> * now run with perl t/harness
On Sun, Sep 16, 2001 at 10:35:10AM -0400, Dan Sugalski wrote:
> This, for example:
>
> > le_i_ic I
>
> would show in the docs as "le".
I know. I've taken account of that.
--
"I'd crawl over an acre of 'Visual This++' and 'Integrated Development
That' to get to gcc, Emacs, and gdb. Thank you
Brent,
This patch allows the Makefile to work for me...if the
$(CC) -o $*$(O) -c $*.c
line is not below each .c file cygwin's make does not do the right thing and
does not make properly. This patch adds those lines. If you have a better
way to do it, I would be glad to implement that instead.
The not op seems to be doing a logical not rather than a bitwise
not. Patch to fix it is as follows:
Index: basic_opcodes.ops
===
RCS file: /home/perlcvs/parrot/basic_opcodes.ops,v
retrieving revision 1.17
diff -u -r1.17 basic_opcode
On Sun, Sep 16, 2001 at 05:29:22PM +0100, Tom Hughes wrote:
> The not op seems to be doing a logical not rather than a bitwise
> not. Patch to fix it is as follows:
Thanks, applied.
--
The best book on programming for the layman is "Alice in Wonderland";
but that's because it's the best book on
On 16 Sep 2001, Gregor N. Purdy wrote:
> All --
>
> Anyone care to post a subcall.pasm example file that shows the
> implementation of a subroutine and a call to it? I was thinking
> of starting from euclid.pasm (since it has two args), but I'm
> not sure I understand what the calling convention
All --
> Anyone care to post a subcall.pasm example file that shows the
> implementation of a subroutine and a call to it? I was thinking
> of starting from euclid.pasm (since it has two args), but I'm
> not sure I understand what the calling convention is really
> supposed to look like...
Here
Ok, here is the rule based makefile...
-Original Message-
From: [EMAIL PROTECTED]
To: Gibbs Tanton - tgibbs
Cc: '[EMAIL PROTECTED] '
Sent: 9/16/2001 2:49 PM
Subject: Re: [PATCH]Makefile.in
This should be done with an implicit rule or a pattern rule.
By putting all the explicit lines i
This should be done with an implicit rule or a pattern rule.
By putting all the explicit lines in, it'll be harder to
change later, and errors can crop up.
The makefile needs a cleanup - we're not making good use of
any of the 'features' of make.
Are we allowed to use gnu makeisms?
-R (holding
Gibbs Tanton - tgibbs writes:
>Ok, here is the rule based makefile...
Thanks
Looks good (and works for me.)
I suggest the following patch on top of Gibbs' patch, to seperate form
and function of CC and CFLAGS. (Also, the "standard" (fwiw) is
CFLAGS, not C_FLAGS)
-R
--- Makefile.in.1
On Sun, Sep 16, 2001 at 12:49:36PM -0700, [EMAIL PROTECTED] wrote:
> This should be done with an implicit rule or a pattern rule.
>
> By putting all the explicit lines in, it'll be harder to
> change later, and errors can crop up.
>
> The makefile needs a cleanup - we're not making good use of
>
Robert Spier:
# Gibbs Tanton - tgibbs writes:
# >Ok, here is the rule based makefile...
#
# Thanks
#
# Looks good (and works for me.)
#
# I suggest the following patch on top of Gibbs' patch, to seperate form
# and function of CC and CFLAGS. (Also, the "standard" (fwiw) is
# CFLAGS, not C_FLAGS)
I think there's an extra / in the cvs add line in the loop.
Gibbs Tanton - tgibbs writes:
>I don't think anyone objects to this...I vote go ahead and move the .h files
>and modify Makefile.in according to Brent's email.
>
>Tanton
>
>$ cvs -q upd -AdP
>$ mkdir -p include/parrot
>$
Hoi,
probably obsolete soon, but still.
Buggs
Index: Configure.pl
===
RCS file: /home/perlcvs/parrot/Configure.pl,v
retrieving revision 1.5
diff -u -3 -p -r1.5 Configure.pl
--- Configure.pl2001/09/15 00:57:42 1.5
+++ Co
Here is a potential patch for all of the non documented opcodes.
It is a patch for parrot_assembly.pod
Thanks!
Tanton
-Original Message-
From: Simon Cozens
To: Dan Sugalski
Cc: [EMAIL PROTECTED]
Sent: 9/16/2001 11:12 AM
Subject: Re: Op documentation versus implementation
On Sun, Sep 16
Small hack to keep process_opfunc.pl from generating
extra return() statements.
I believe I incorrectly sent this to the wrong party before.
My apologies.
===
--- process_opfunc.bak.pl Sun Sep 16 13:23:41 2001
+++ process_opfunc.pl
30 matches
Mail list logo
