Re: [PATCH] Disable GC at startup

On Thursday 11 April 2002 07:03 pm, Michel J Lambert wrote: > Dan Sugalski wrote: > > That's because GC_DEBUG is fundamentally broken by design. It's doing > > things it shouldn't be doing--the system is such that it shouldn't > > ever trigger GC before the memory allocation system is set up. > >

Re: [Applied] Fix up comments in pmc.c

Cool, applied. At 18:21 on 04/11/2002 EDT, Simon Glover <[EMAIL PROTECTED]> wrote: > > We no longer pass a PMC pointer into pmc_new, but the comment hasn't been > changed to reflect that. Patch below corrects, and also adds an > appropriate comment for pmc_new_sized. > > Simon > > --- p

Re: [Applied] String comment fixes

This one too. --Josh At 19:10 on 04/11/2002 EDT, Simon Glover <[EMAIL PROTECTED]> wrote: > > This patch tidies up a few of the comments in string.c, and fixes one > actual documentation bug -- namely, string_chopn removes the last > n _characters_, not the last n _bytes_. > > Simon > >

[DOC PATCH] String comment fixes

This patch tidies up a few of the comments in string.c, and fixes one actual documentation bug -- namely, string_chopn removes the last n _characters_, not the last n _bytes_. Simon --- string.c.oldThu Apr 11 19:01:31 2002 +++ string.cThu Apr 11 19:06:47 2002 @@ -87,7 +87,7 @@

Re: [PATCH] Disable GC at startup

Dan Sugalski wrote: > That's because GC_DEBUG is fundamentally broken by design. It's doing > things it shouldn't be doing--the system is such that it shouldn't > ever trigger GC before the memory allocation system is set up. > > If this patch disables the GC only when GC_DEBUG is in place, I'll p

[DOC PATCH] Fix up comments in pmc.c

We no longer pass a PMC pointer into pmc_new, but the comment hasn't been changed to reflect that. Patch below corrects, and also adds an appropriate comment for pmc_new_sized. Simon --- pmc.c.old Thu Apr 11 18:02:16 2002 +++ pmc.c Thu Apr 11 18:17:30 2002 @@ -16,14 +16,12 @@ #incl

Re: [PATCH] Disable GC at startup

At 5:35 PM -0400 4/11/02, Michel J Lambert wrote: >There are still plenty of problems with GC_DEBUG turned on, but at least >it passes all the non-pmc tests now. Has anyone done anything with >attempting to get a make tinder and tindertest set up? That's because GC_DEBUG is fundamentally broken b

[PATCH] Disable GC at startup

The attached patch disables GC at startup. If you turn on GC_DEBUG, you'll see parrot crash at startup because dod runs and collection runs are being performed before the interpreter is fully initialized, causing all sorts of havoc. This also means that if our default GC allocation values do not

Re: What I'm working on before I go offline for 2 weeks

At 11:12 PM -0400 4/10/02, Melvin Smith wrote: >3) Going to fixup the IO system to be more Parrot like, including stringifying >and making a real PMC out the PIO object. >Dan I did send you an initial PDD for IO which you >acknowledged but then >I think you ate it. Still n

Re: What I'm working on before I go offline for 2 weeks

At 12:02 PM +0100 4/11/02, Simon Cozens wrote: > > I'm also chipping away at a certain unnamed company to donate an AIX box >> for official testbed > >Another unnamed company offered us one, and nobody collected it in >time, so it got scrapped. :( Argh, bummer. On the other hand, a certain unna

Re: TODO additions

im a newcomer, particulalry intrested in the jit part. is there anyone working on the sparc asm? /joh On Thu, 11 Apr 2002, Steve Fink wrote: > These are my idea of new TODO items. Note that I don't really know > anything about some of the things I added, so someone should probably > read this

Re: Minor patch to make the GC count total bytes copied [APPLIED]

At 5:13 PM +0200 4/11/02, Peter Gibbs wrote: >The memory_collected GC statistic does not get updated at present. Patch >below fixes. > >Note that a 5000-generation run of life.pasm allocates 32K, and copies >almost 58MB. Applied, as I make the interpreter crash badly in another directory. :) I'm

Re: internal structure

On Thu, 11 Apr 2002, [latin1] Josef Höök wrote: > > How about the JIT engine then. > When does parrot use it? > how well does it scale? ( has anyone done some benchmarking ) The JIT engine produce machine code for a given bytecode, but there are only a few opcodes that are coded in assembly so

RE: internal structure

Jonathan E. Paton: # > How about the JIT engine then. # # JIT runs on bytecode, producing bytecode. No, JIT runs on bytecode, producing equivalent machine code. # > When does parrot use it? # # Before executing bytecode. # # > how well does it scale? ( has anyone done some benchmarking ) # # Don

Re: internal structure

> > How about the JIT engine then. > > JIT runs on bytecode, producing bytecode. > > > When does parrot use it? > > Before executing bytecode. > > > how well does it scale? ( has anyone done some benchmarking ) > > Don't ask me :P If you haven't already, then you'd > be best looking at the b

Minor patch to make the GC count total bytes copied

The memory_collected GC statistic does not get updated at present. Patch below fixes. Note that a 5000-generation run of life.pasm allocates 32K, and copies almost 58MB. -- Peter Gibbs EmKel Systems Index: resources.c === RCS file:

Re: internal structure

How about the JIT engine then. When does parrot use it? how well does it scale? ( has anyone done some benchmarking ) On Thu, 11 Apr 2002, [iso-8859-1] Jonathan E. Paton wrote: > > Where in the parrot code does the actual translation > > from byte code to binary code occur? > > Parrot eq. a

Re: internal structure

> Where in the parrot code does the actual translation > from byte code to binary code occur? Parrot eq. an interpreter, all the byte codes are like commands to tell it what actions to take... it doesn't directly take byte codes and turn them into binary code. Conversion would be compiling, but

internal structure

Hello. Where in the parrot code does the actual translation from byte code to binary code occur? where does it get executed? Im having a hard time finding things in the code.. /joh

Re: What I'm working on before I go offline for 2 weeks

Melvin Smith: > I'm getting married Saturday and going on honeymoon for 10 days. Congratulations! Enjoy. > I'm also chipping away at a certain unnamed company to donate an AIX box > for official testbed Another unnamed company offered us one, and nobody collected it in time, so it got scrapped.