On Thu, Aug 09, 2007 at 08:41:22AM +0200, felix winkelmann wrote:
libffi support has been dropped. Hand written assembly glue
code is now used and should work with the autotools build
for x86, ppc and x86-64 platforms and MSVC.
Does this mean Chicken will not build anymore on, say, SPARC
On Tue, Aug 14, 2007 at 05:41:15PM +0900, Ivan Raikov wrote:
Famous last words. ;-)
*GLP*
Peter Bex [EMAIL PROTECTED] writes:
It can never be worse than autohell.
___
Chicken-hackers mailing list
Chicken-hackers@nongnu.org
http
On Tue, Aug 14, 2007 at 01:55:17PM +0200, felix winkelmann wrote:
- Generation and installation of the info manual from texi sources
I got this to work by simply changing INSTALL(TARGETS chicken.info ..) to
INSTALL(FILES chicken.info ...) (see attached patch)
If I understand the documentation
Hi all,
For a project I'm currently working on I'm implementing a network
protocol in pure Scheme. This protocol uses numeric constants as
message types. Since Chicken is targeted very much at interoperability
with existing C code, I thought the following macro would make a nice
addition to
Hi there,
When debugging the sendfile egg we discovered that Chicken itself does
not report correct file sizes on really big files ( 1Gb) on 32-bit
architectures. We're not sure if this happens on all systems, but
we've tried on NetBSD and Linux.
We're still not sure what's going on here, but
On Sun, Oct 28, 2007 at 09:49:38PM +0100, Peter Bex wrote:
Applying the attached patch to posixunix.scm fixes the problem - for us.
Terribly sorry, but this patch DOES NOT WORK. I ripped this code from
the code to 'du', which is supposed to work on large files. Turns out
that my system's 'du
On Sun, Oct 28, 2007 at 10:34:36PM -0500, Zbigniew wrote:
OK, I extended define-enum in miscmacros to support resetting the
counter. Would this suffice?
Excellent! Thanks.
Peter
--
http://sjamaan.ath.cx
--
The process of preparing programs for a digital computer
is especially attractive,
On Sat, Nov 24, 2007 at 11:38:09AM -0200, Mario Domenech Goulart wrote:
This doesn't seem that hard from my (posibly naive) perspective.
In every release an svn cp .../eggs .../rel-a.b.c/eggs is done.
chicken-setup merely needs to a) know the version and b) get the eggs
from
On Sun, Jan 13, 2008 at 09:51:23AM -0800, Elf wrote:
at a guess, i imagine that the reason for felix (and chicken) using gnu make
is that a) its available for all platforms supported by chicken and b) its
common on all of these platforms. i suppose if you wanted to port bsd's make
as a
On Sun, Jan 13, 2008 at 10:30:13AM -0800, Elf wrote:
a) who uses pkgsrc besides the bsds?
Solaris people tend to rely on pkgsrc because it's the only sane way to
install 3rd party packages. There are even some Linux distros that use
pkgsrc for their package manager. I've used it on Mac OS X
Hello everyone,
This weekend's hackathon was a great success. Fun was had by all, and lots
of work got done. Most importantly, a number of new people have become
more active in the project.
Summary
---
We have accomplished the following things:
- About twenty eggs got their old-style
On Mon, Feb 25, 2008 at 06:27:31AM -0800, Kon Lovett wrote:
Thanks to everyone for all the work. I just wish test-
infrastructure wasn't translated since it is obsolete. The current
is testbase.
Even obsolete stuff should be moved to the wiki. Everything in one
place means everything in
On Mon, Apr 14, 2008 at 10:21:40AM +0200, felix winkelmann wrote:
On Sun, Apr 13, 2008 at 3:43 PM, Peter Bex [EMAIL PROTECTED] wrote:
Hi all,
According to Elf the main Chicken hackers (Felix and Kon) do not
read the janitors list and few others do.
I just submitted a ticket
On Wed, Oct 15, 2008 at 10:11:53AM -0300, Mario Domenech Goulart wrote:
Hi guys
I've moved salmonella to a new, faster, machine. Besides running
salmonella faster, it alleviates galinha's load.
The new salmonella home is http://salmonella.ucpel.tche.br
Please, update your bookmarks and
On Sun, Nov 09, 2008 at 05:53:08PM +0100, Peter Bex wrote:
Hi all,
Whenever I request a status list from subversion, it keeps telling me
about all the .import.scm files that got created by running
chicken-setup from the egg dir. How about adding '*.import.scm' to
the global-ignores
On Mon, Nov 24, 2008 at 01:33:08PM +0100, felix winkelmann wrote:
Hi, Peter!
I think I found the problem: an incorrectly named makefile
variable. I hope the last commit (rev. 12585) helps.
Thanks, that did the trick!
Cheers,
Peter
--
http://sjamaan.ath.cx
--
The process of preparing
On Mon, Apr 06, 2009 at 09:28:38AM +0200, Peter Bex wrote:
I have created a man/ subdirectory. In fact, it would be cool, if you
could help me and move the old manual into man/3.
Sure, I will do this tonight.
The wiki is merrily crunching away on my changes right now.
I've moved all
On Mon, Apr 06, 2009 at 01:21:19PM -0700, Vincent Manis wrote:
I believe I am the original author of `Getting Started'. I'll take a
look at it and see if there's anything (minor, I hope) that should
change.
Thanks
I make a pest of myself on some free software projects for asking if
Hey all,
I was wondering when we will be seeing a new minor release of Chicken 4.
Several annoying bugs have been fixed, and we now have a nice new
scrutinizer, which would warrant a new release IMHO. Are there some
important open tickets to fix first? I assume ticket #34 should be
fixed, but
On Fri, Jun 26, 2009 at 10:30:36AM +0900, Ivan Raikov wrote:
The Chicken development snapshot script has been adapted to Chicken
4. From now on, Chicken 4 snapshots will be built and linked to on the
development snapshots page:
http://chicken.wiki.br/dev-snapshots/
hurray! Thanks,
On Fri, Jul 31, 2009 at 05:23:34PM +0900, Ivan Raikov wrote:
Ok, I can do some research and see if there is some existing wiki
parser that can be adapted to read the svnwiki dialect and produce
s-expressions. If there are some reasonable solutions out there, then I
will propose an
Hi all,
I've finally made a Chicken 4.1.0 package for pkgsrc. This package uses
www.call-with-current-continuation.org/chicken-4.1.0.tar.gz but also
chicken.wiki.br/releases/4.1.0/chicken-4.1.0.tar.gz as a mirror.
Unfortunately, these files are not the same! For some reason the files
differ,
On Tue, Sep 01, 2009 at 09:57:47AM +0200, felix winkelmann wrote:
Hello!
It's probably just me, but if I have to write (use ports files
data-structures utils extras) one more
time, I'll go crazy.
I think the splitting up of the extras unit was done haphazardly. It
doesn't really give any
On Mon, Oct 05, 2009 at 08:26:03PM -0700, Kon Lovett wrote:
Hi,
Personally I want a core Chicken with a component orientation towards
the major sub-systems. I mean pluggable composable string, number,
gc, concurrency, file components; not to be construed as an exhaustive
list.
I
On Thu, Apr 08, 2010 at 11:09:50AM -0400, John Cowan wrote:
Thomas Chust scripsit:
I consider it extremely bad style to rely on the mutability of
literals, so I wouldn't mind if structurally equal literal constants
were folded, whether there is inlining involved or not.
+1. I would
Hello Chicken hackers,
I've traced an issue with Spiffy in HTTPS mode to a nasty interaction
between OpenSSL and sendfile.
For those who don't know sendfile: it is a Chicken egg that attempts
to push data from an input port to an output port as fast as possible.
It can use the Linux-specific
On Sun, Jul 04, 2010 at 01:17:52AM +0200, Felix wrote:
Hello!
Because I can't figure out how to extend `find-files' without
changing its signature (something I don't want for compatibility
reasons), can someone suggest an alternative name for a new
variant that can later supersede
On Sun, Jul 04, 2010 at 01:03:27AM +0200, Felix wrote:
So... You could use the reserved fields (you actually *should* use
the port-data, but some silly implementor (me) has already broken
this convention), so you could use the reserved fields of the port
(see around line 1630 in library.scm)
On Sun, Jul 04, 2010 at 02:59:14PM +0200, Thomas Chust wrote:
Hello,
this looks alright to me. I have three small remarks:
* I would suggest renaming ssl-port-tcp-fd into ssl-port-tcp-port,
since it doesn't return a file descriptor, but rather another
port.
Whoops, that's
On Sun, Jul 04, 2010 at 04:36:34PM +0200, Thomas Chust wrote:
Hello,
maybe ssl-addresses is important enough to warrant a convenience
routine. Maybe an additional ssl-port-fileno procedure would also
make sense. If minimalism is a goal, exporting ssl-port-tcp-port
would be sufficient,
On Mon, Aug 23, 2010 at 12:43:21PM -0700, Kon Lovett wrote:
Assume a component of package A uses something that is GPL'ed, but no
other component in that package uses the GPL tainted component (it is
just along for the ride). Then all components of package A are
tainted?
If package A is
On Mon, Aug 23, 2010 at 01:44:48PM -0700, Kon Lovett wrote:
In the format-compiler-base used by error-utils in check-errors
situation nothing got linked, only packaged. Unless errorf is
actually used then no linkage occurs. (And, at the time, the only
module using errorf was
On Tue, Aug 24, 2010 at 12:02:49PM -0400, John Cowan wrote:
Also note that the LGPL is incompatible with the GPL in that you cannot
link a LGPL library or program to a GPL library(!)
That's definitely false. The whole point of the LGPL (2.1 or 3) is that
you can use an LGPLed library in a
On Tue, Sep 07, 2010 at 09:16:13AM +0200, Felix wrote:
Any ideas for a better name? Any considerations that
should be taken into account?
Maybe equal?* or equal*? or something? Or equal/=? or equal=? or
equal?/=
The equal?? is a bit confusing because the SSAX tools already use
the
On Tue, Sep 07, 2010 at 08:22:07PM +0200, Felix wrote:
Any ideas for a better name? Any considerations that
should be taken into account?
How about structurally-equal? A little long, but very clear.
But isn't equal? also structurally-equal? ?
What about `same?' or `similar?' ?
On Mon, Oct 18, 2010 at 04:16:52AM -0400, Felix wrote:
Hello!
I have added read-syntax for blob-literals and currently use
#{hex...}
Since this may block the use of #{ ... } for user-defined read-syntax,
I want to ask if perhaps another syntax might be preferrable.
I think the braces
On Tue, Nov 23, 2010 at 03:59:48AM -0500, Felix wrote:
Should it be possible to allow multiple threads to set different file
creation modes? As far as I understand, Scsh allows that (I may be wrong,
though).
I'm strongly against this. I don't see the advantage of doing so
and I find it
On Tue, Nov 23, 2010 at 07:54:08AM -0500, Felix wrote:
Should we still have a separate set! procedure, aside from the
generic setter? It would make sense to me, since most procedures
have both a getter and a setter procedure, regardless of whether
they have a generic setter.
You can
On Wed, Dec 01, 2010 at 01:50:33PM +0100, Thomas Chust wrote:
2010/11/30 Mario Domenech Goulart mario.goul...@gmail.com:
[...]
A poll on the umask support has been set. See
http://bugs.call-cc.org/ticket/424#comment:13
[...]
Hello,
my apologies for the stupid question, but how does
Hello,
This CR's time has also expired, but we haven't really reached a
conclusion on what exactly to vote, so I'll make it simple and
make a yes/no poll on the original patch posted in the ticket.
If people disagree, please give a motivation on this list and
let us know what implementation
On Tue, Mar 15, 2011 at 02:19:20AM -0400, John Cowan wrote:
Felix scripsit:
I like the idea of tgz-submission (did John suggest it? I can't
remember, it's already 60 seconds later). How about a central store
that accepts tgz's and external clients, that keep their own
infrastructure
On Thu, Mar 17, 2011 at 12:16:19PM +, Imran Rafique wrote:
Its been interesting reading through this thread. I'd like to suggest
some amendments to Peter's proposals, addressing some of the concerns
highlighted.
I haven't had time to digest this fully, but here are some initial remarks:
On Fri, Mar 18, 2011 at 08:59:10AM +, Imran Rafique wrote:
Felix said:
When mixing trees that way, how do you ensure the combination of them
is consistent?
The global tree. I can't write to this.
+ networking/
+ zmq/
+ recipe-zmq-1.0
+ recipe-zmq-2.0
On Fri, Mar 18, 2011 at 09:00:20AM +, Imran Rafique wrote:
Dynamite is popular, and gets repeatedly forked on github.
Forking's bad, mmmkay?
How do you handle this in the official egg tree?
We could have 3 seperate eggs, dynamite, dynamite-peter and
dynamite-felix. If there were any
On Fri, Mar 18, 2011 at 09:53:19AM +, Imran Rafique wrote:
I don't see why chicken-install needs to be bloated with all this stuff.
Why not maintain a mercurial patch queue or use quilt. You could also
just have a local clone of a repo with your own changes?
Right. Till you have a
On Fri, Mar 18, 2011 at 10:02:17AM +, Imran Rafique wrote:
On 18 March 2011 09:20, Peter Bex peter@xs4all.nl wrote:
On Fri, Mar 18, 2011 at 09:00:20AM +, Imran Rafique wrote:
Dynamite is popular, and gets repeatedly forked on github.
Forking's bad, mmmkay?
Are you lecturing
On Wed, May 11, 2011 at 05:25:12AM -0400, Felix wrote:
I was wondering what use does (null-pointer?) has.
Historical. I will deprecate.
Thanks, Felix. I noticed the documentation says
Another way to say (address-pointer 0). Should the address-pointer
procedure return #f when given 0?
On Wed, May 11, 2011 at 08:21:32AM -0400, Felix wrote:
On Wed, May 11, 2011 at 05:25:12AM -0400, Felix wrote:
I was wondering what use does (null-pointer?) has.
Historical. I will deprecate.
Thanks, Felix. I noticed the documentation says
Another way to say (address-pointer 0).
On Sun, May 22, 2011 at 04:53:18PM +0200, Felix wrote:
Any pointers to code, or heavy-duty regexen that need long to execute
would be very welcome.
You've probably already seen it, but just in case you missed it, there's
a benchmarks directory in the irregex upstream hg repo.
They mostly test
On Sun, May 22, 2011 at 11:25:44AM -0700, matt welland wrote:
Hi Felix,
It isn't a particularly complex benchmark but my logpro app relies
heavily on regexes and I'm seeing some somewhat slow performance. You
can get it here http://www.kiatoa.com/fossils/logpro.
I tried to clone it, but
On Mon, Jul 04, 2011 at 02:47:01AM -0400, John Cowan wrote:
Felix scripsit:
* Those who want IEEE behaviour can have it using fpexpt.
True. However, I tend to expect that the fp* functions will provide
efficient versions of the regular procedures (because they know their
arguments are
On Mon, Jul 04, 2011 at 10:41:22AM +0200, Felix wrote:
+1 on that. Having the fp (or fx, for that matter) procedures behave
differently from the generic ones on flonums is unintuitive and seems
to me like trouble in the making.
Is this intuition, or are you simply being used to it? I
On Mon, Jul 04, 2011 at 11:20:57AM +0200, Felix wrote:
This deeply disappoints me. You all are a bunch of robots.
A very smart man I deeply respect once said: It is not necessary to say
that
See, another robot.
Bzzt. Does not compute. ERROR ERROR
FAILURE
*Crashes to the floor,
On Mon, Jul 04, 2011 at 12:59:25PM +0200, Moritz Heidkamp wrote:
This deeply disappoints me. You all are a bunch of robots.
Who would have thought that floating point numbers can stir emotions
like that! On second thought ...
In the words of Pennywise the clown:
We all float down here!
On Thu, Jul 14, 2011 at 04:51:48AM -0400, Felix wrote:
Hello!
HEAD in master of the core system is currently broken (make check
fails).
I think I've fixed this in d2fd11237651c2f4f114420705ed99279410b65b
of expander-simplifications.
Cheers,
Peter
--
http://sjamaan.ath.cx
--
The process of
On Wed, Aug 10, 2011 at 01:04:19PM -0700, Derrell Piper wrote:
Anyone else trying to build trunk on Lion? Or is this the problem Felix
alluded to a while back?
make C_COMPILER=gcc-4.2 PLATFORM=macosx PREFIX=/usr/local ARCH=x86-64
It looks like you're trying to build Chicken from git with
On Mon, Aug 22, 2011 at 11:56:55AM +0200, Felix wrote:
Hi!
From now on I'll better ask first, so:
Would anybody be very unhappy if always?, none? and never?
(unit data-structures) get deprecated? They can be trivially defined
using constantly. The situation for any? is different IMHO,
On Mon, Aug 22, 2011 at 03:59:01PM +0200, Felix wrote:
I don't understand how any? is different.
Like this:
(define (any? x) #t)
(define (none? x) #f) ; DEPRECATED
(define (always? . _) #t) ; DEPRECATED
(define (never? . _) #f) ; DEPRECATED
Hi all,
I'm working on fixing several bugs in numbers, and now I stumbled
upon Chicken's behaviour with zero-flonums:
(/ 1.0 0.0) = error: division by zero
(/ -1.0 0.0) = error: division by zero
(/ 0.0 0.0) = error: division by zero
However, an inexact zero is not exactly zero; it could be
On Wed, Sep 07, 2011 at 05:14:45PM +0200, Christian Kellermann wrote:
Thanks! We just had a little chat on IRC about this. For the records
let me summarise:
- the patch does basically what alist-copy does in srfi-1
Yep
- why not move all the alist stuff *into* srfi-1 even our extensions to
On Wed, Aug 24, 2011 at 09:18:44PM +0200, Peter Bex wrote:
All other Schemes I tested with (Guile, Gauche, Scheme48, Racket and Gambit)
return +inf.0 for (/ 1.0 0.0) and -inf.0 for (/ -1.0 0.0).
They return +nan.0 for (/ 0.0 0.0). Extra data points would be welcome.
I'd like to open a CR
On Thu, Sep 08, 2011 at 12:09:11PM +0100, Alaric Snell-Pym wrote:
The egg index is pretty nicely categorised as it is. Its certainly good
enough for someone to search through, and quickly zero in on 1 or 2 (or 3)
eggs of specific interest to the problem at hand. I'm not sure what else
you
Hi all,
As some of you know I've been doing some work on the numbers egg which
culminated in a horrific torture test for numerical syntax that was
sent to the R7RS discussion list in an attempt to get rid of the ugly '#'
padding syntax in numbers which it inherited from R5RS (which seems to
have
On Tue, Sep 13, 2011 at 10:21:58AM +0200, Felix wrote:
Attached is a patch that adds documentation for the pure
declaration, which declares a toplevel identifier to name
a procedure that is free of side effects (and thus calls to it
may be eliminated). This declaration was previously named
On Sun, Sep 11, 2011 at 12:26:45AM +0200, Felix wrote:
Hello!
The attached patch implements a correction of the available sequence
type specifiers. Currently (list T) and (vector T) designate lists
or vectors with an unknown number of elements of the given type. This
is suboptimal, because
On Sun, Sep 11, 2011 at 12:26:45AM +0200, Felix wrote:
The attached patch implements a correction of the available sequence
type specifiers.
This is quite a big patch, but I can see that the changes to types.db
and the tests are mostly about mechanical changes.
I don't quite grok the
On Wed, Sep 14, 2011 at 03:52:58AM -0400, Felix wrote:
On Wed, Sep 14, 2011 at 12:51:55AM +0200, Felix wrote:
I think you overlooked these entries:
- reverse: argument and result
- member, assv, assoc: specialization types
Also, member has a (forall a) in its declaration
On Wed, Sep 07, 2011 at 10:01:52PM +0200, Peter Bex wrote:
On Wed, Aug 24, 2011 at 09:18:44PM +0200, Peter Bex wrote:
All other Schemes I tested with (Guile, Gauche, Scheme48, Racket and Gambit)
return +inf.0 for (/ 1.0 0.0) and -inf.0 for (/ -1.0 0.0).
They return +nan.0 for (/ 0.0 0.0
.
-- Donald Knuth
From 6c39851e1504d01b2de05e014f5542913e3e5c5e Mon Sep 17 00:00:00 2001
From: Peter Bex peter@xs4all.nl
Date: Sat, 17 Sep 2011 17:41:58 +0200
Subject: [PATCH] Do not drop precision in flonums when compiling
---
c-backend.scm|2 ++
tests/compiler-tests.scm
On Sat, Sep 17, 2011 at 05:56:52PM +0200, Peter Bex wrote:
Hi all,
This is a patch and bugreport rolled into one ;)
While testing the numbers egg I found out that programs
containing flonum literals can behave differently when compiled
or interpreted.
I forgot, but here's a simple test
On Mon, Sep 19, 2011 at 01:13:18PM +0200, Felix wrote:
Removes broken makefile target compile-all, since the compile-all
script in the scripts directory does the same.
What is the reason the Makefile rule is replaced by a script?
Make seems like a better way to do this as it won't
On Fri, Sep 23, 2011 at 08:36:14AM +0200, Felix wrote:
Looks fine, please apply it.
I still can't make git apply patches with git am while preserving
commit author and message, and I couldn't find the patch in your
felix-pending branch.
Sorry, I forgot to push this one.
Is your
On Sun, Oct 02, 2011 at 11:20:37AM +0200, Christian Kellermann wrote:
The patch still makes sense for architectures with older gcc versions.
I think we should test for it and wrap it then (this may also apply
to the -fwrap switch).
How do other BSDs deal with this?
NetBSD uses -lpthread in
On Sat, Oct 08, 2011 at 04:30:12PM -0400, John Cowan wrote:
Here's my argument: file-exists? should return #t if the file definitely
exists and #f if it definitely doesn't, and should throw an exception
if the implementation cannot tell. In the case of ENAMETOOLONG, the
file *cannot* exist,
On Wed, Oct 12, 2011 at 04:28:37AM -0400, Felix wrote:
The following patch disables a test in
tests/numbers-string-conversion-tests.scm,
which causes it to fail on Windows (mingw) systems. Apparently the runtime
library prints floating-point numbers differently.
There possibly is a better
On Wed, Oct 12, 2011 at 06:42:56AM -0400, Felix wrote:
From: Felix fe...@call-with-current-continuation.org
Subject: disable failing test on Windows
Date: Wed, 12 Oct 2011 04:28:37 -0400 (EDT)
The following patch disables a test in
tests/numbers-string-conversion-tests.scm,
which
On Wed, Oct 12, 2011 at 04:30:10AM -0400, Felix wrote:
The tests takes very long to compile and only tests library functionality,
so running it in the interpreter may be sufficient.
Pushed, thanks.
--
http://sjamaan.ath.cx
--
The process of preparing programs for a digital computer
is
On Fri, Oct 21, 2011 at 03:33:03AM +0200, Felix wrote:
The attached patch resolves macro-aliases correctly when resolving
identifier names in declare forms. Without this, identifiers that
went through several macro-expansions (and thus aliasing) where not
resolved to their true names,
On Fri, Oct 21, 2011 at 01:35:13PM +0200, Felix wrote:
Here it is. Thanks for pointing out the problem with the old
patch.
Applied and pushed. Thanks!
Peter
--
http://sjamaan.ath.cx
--
The process of preparing programs for a digital computer
is especially attractive, not only because it can
or music.
-- Donald Knuth
From 85930543f14bed93693c7565f3aee3b33e2a852a Mon Sep 17 00:00:00 2001
From: Peter Bex peter@xs4all.nl
Date: Fri, 9 Dec 2011 10:48:49 +0100
Subject: [PATCH] Apply upstream changesets ba70feace1dd and 78ba6b09e021
On Thu, Dec 15, 2011 at 02:50:16AM -0500, Felix wrote:
The attached patch adds special cases for list-ref and list-tail
to the scrutinizer to obtain more precise result-type information
when the index argument is a constant (and the list argument
is of a known fixed-length list type).
Does
On Sun, Dec 18, 2011 at 02:02:55PM +0100, Felix wrote:
(sorry for being obnoxious)
You aren't. Questions like these are important to be cleared up.
And thank you for doing so! This really helped.
I've applied and pushed the change.
Cheers,
Peter
--
http://sjamaan.ath.cx
--
The process
On Sun, Jan 01, 2012 at 01:08:39PM +0100, Jörg F. Wittenberger wrote:
On Dec 31 2011, John Cowan wrote:
Huh? The point is that well-chosen hash collisions can force the
algorithm into its worst case behavior, and if that's linear, it's a
problem. Choosing a linear algorithm to begin with is
On Sun, Jan 01, 2012 at 09:57:15AM -0500, John Cowan wrote:
Peter Bex scripsit:
Agreed. Changing datastructures is not an option.
Why not?
Because not fixing hash tables makes them into another footgun.
Also because fixing them transparently is simple and has almost
zero overhead.
It's
On Sun, Jan 01, 2012 at 10:29:18AM -0500, John Cowan wrote:
Peter Bex scripsit:
Yes, and doing it in *every* *freaking* program. Including
third-party libraries written long ago or by people assuming a sane
srfi-69 implementation (or more likely, not having thought about
On Sun, Jan 01, 2012 at 01:41:26PM +0100, Peter Bex wrote:
That's confirmed. Attached is a preliminary patch to fix this issue.
This patch was indeed *very* preliminary and provided almost zero
protection. That's because:
- When a hash collision is found in the *underlying* algorithm's
Hi all,
The thought just popped into my head that my hash table patch breaks
SRFI-69 compatibility. When the user passes a custom hash procedure,
this procedure is expected to accept an extra argument, which is not
part of the SRFI-69 API.
While looking into this, I also found that the result
On Sun, Jan 08, 2012 at 04:13:49PM +0100, Jörg F. Wittenberger wrote:
On Jan 7 2012, Peter Bex wrote:
The thought just popped into my head that my hash table patch breaks
SRFI-69 compatibility. When the user passes a custom hash procedure,
That would be bad bad. In the end.
Yes.
2
On Thu, Jan 12, 2012 at 01:18:42AM -0600, Jim Ursetto wrote:
C_string_to_symbol was allocating 6 words when it needed 7 (symbols gained an
extra slot at some point, I think). I caught this when trying to get LLVM
gcc working.
Maybe it's also related to the intermittent symbol GC test
On Sat, Jan 21, 2012 at 08:38:39PM +0100, Peter Bex wrote:
Hi all,
Afterwards, I added the option to selectively turn on profiling for
parts in Chicken itself to the Makefiles (see patch 0001 or changeset
57fff73344771739e7723e28df0c0019058a0268 in the sjamaan-pending branch).
This helped
On Sat, Jan 21, 2012 at 08:38:39PM +0100, Peter Bex wrote:
Instead of using sprintf and allocating a new string and copying over
every time, I've changed real-name to use build up a list and use
string-intersperse to join them together at the end (see patch 0002 or
changeset
.
-- Donald Knuth
From 3b3e278df64b5d8ae4dd1dcd8576f9597d32c5ca Mon Sep 17 00:00:00 2001
From: Peter Bex peter@xs4all.nl
Date: Mon, 23 Jan 2012 18:39:34 +0100
Subject: [PATCH] Limit depth for procedure nesting reports to ensure linear
scaling of compilation times on input file size
, but also because it can be an aesthetic
experience much like composing poetry or music.
-- Donald Knuth
From d9de1eb6b0dfef4d119748c99efa2fd0f9f37adb Mon Sep 17 00:00:00 2001
From: Peter Bex peter@xs4all.nl
Date: Thu, 26 Jan 2012 20:34:10
On Mon, Jan 23, 2012 at 06:02:04AM +0100, Felix wrote:
Hello!
The attached patch adds slight improvements to the flow analysis pass
done by the scrutinizer:
1) Variables that are assigned now retain that type-information unless
they are not captured (previously all assigned variables
13343f2c6e9846b1fccd0d8d4cbbb66a64b8284b Mon Sep 17 00:00:00 2001
From: Peter Bex peter@xs4all.nl
Date: Sun, 29 Jan 2012 15:23:15 +0100
Subject: [PATCH] In the analysis phase, keep around a copy of localenv appended
to env. This ensures that deeply nested let forms don't cause exponential
On Mon, Jan 30, 2012 at 08:35:22AM +0100, Felix wrote:
Since the scrutinizer tracks program-flow, assignment can theoretically
be used to update the currently known type-information for a variable
(that's the blist, a list of updated type-information of variables,
kept separate from the env,
On Mon, Jan 30, 2012 at 10:28:18AM +0100, Felix wrote:
Note: there is a certain class of variable manipulation that we can not
track: assignments from other threads and side-effects caused by
finalizers. I see no way to handle this without being excessively
conservative (and thus reducing
On Mon, Jan 30, 2012 at 11:49:15AM +0100, Jörg F. Wittenberger wrote:
I'm asking this because I'm trying for several weeks to track
down a certain segfault. As it happens I can observe that
one only on ARM, never on AMD64. It occurs randomly and rare
enough to make a full call trace a no-go.
poetry or music.
-- Donald Knuth
From 1b6c8f6797ec4a142074c7408aada9d44d2e1674 Mon Sep 17 00:00:00 2001
From: Peter Bex peter@xs4all.nl
Date: Thu, 2 Feb 2012 21:27:28 +0100
Subject: [PATCH 1/2] When preparing for compilations, don't keep re
.
-- Donald Knuth
From 0c909da170e82eaca7e1255b7ba20b24945e84c3 Mon Sep 17 00:00:00 2001
From: Peter Bex peter@xs4all.nl
Date: Sun, 5 Feb 2012 22:54:34 +0100
Subject: [PATCH] While optimizing, don't traverse the same chain of replacable
variables multiple times; update all variables encountered during
1 - 100 of 1541 matches
Mail list logo
