Perl also has threading support (and no GIL) and no, it does not copy the entire interpreter state, and hasn't done so since perl 5.6 at least.
No, that is wrong.
Perl threading (ithreads - as currently implemented in 5.x) DOES really copy the entire interpreter state.
And the implementation is incredibly dumb: think of a fork() that (instead of copying the entire memory at once
or doing COW on pages on VM systems, etc.) will walk painfully through all the pointers, strings and objects and
copy each other in turn, recursively.
You come to wonder if the people who designed it were really morons, or it was some kind of joke.
Hide macro definitions in amongst rubbish comments. The programmer will get bored and not finish reading the comments thus never discover the macro. Ensure that the macro replaces what looks like a perfectly legitimate assignment with some bizarre operation, a simple example:
#define a=b a=0-b
I really don't understand what that guy want to demonstrate with this.
His ignorance about how the preprocessor works ?
He's also not able to make the difference between matters of just style and code that's
invoking undefined behaviour (see the part about "lawyer's code").
'open source' was an attempt to confiscate the Free Software movement, and
build a 'business model' on other people's work.
If the political & religious (sic) implications of "Free as in Freedom" make you
nervous, just stay away from the whole thing.
Don't forget that FSF started by actually paying people salaries for writing
free software; people have to eat in order to write software, free or proprietary.
Those 'open source' lies about people writing working software in their free time
remind me of sport in socialism, where professional athletes were supposed
to work hard in factories, and train only during the weekend.
Nobody is writing a functional driver at home at night after 10 or 12 hours of hard,
stressing day work.
If you have to lie for corporate ideological reasons, at least don't pretend that
people are actually believing you:)
and stop that silly drivel now: if I driver is complete
and isn't written with obfuscation in mind, all the
great 'secrets' about the workings of the device should
be already compromised.
they want people to sign NDAs in order to force them
to write half-functional, obfuscated crap so they could
keep the impression that their proprietary model is still
the rule, and try to shove it again down people's throat
after that 'open-source' fad dies out.
it's ironic that OpenBSD came to be more close to the ideals
of free software (as of FSF) than linux - but linux just
became too important, and unscrupulous self-promoting assholes
and attention whores are already considering it their
exclusive playground.
It's a small, moderately prosperous and
racis^WWestern Democratic Values bla bla bla
country in/Northern/ Europe.
Most Estonian will be happy to explain you how
*all* the crime in their country is carried out by
Russians or more recent immigrants. You know,
stuff as usual.
Now can we just take this quote to mean that exploiting this part of the IPv6 specification has an extremely low barrier to entry as it was intended and move along?
I got it.
It's like saying: "everything's so simple, even
an American can figure it out."
But I still think they had to give proper credits
in their CVS log. Not doing so in the first
place + trying to play down the importance of the
bug was dishonest and despicable.
I do use OpenBSD too, and appreciate the OpenBSD
team, but I think they somehow deserve a bit of
undeserved public humiliation for this:)
*Credits*
This vulnerability was found and researched by Alfredo Ortega from
Core Security Technologies. The proof-of-concept code included in the
advisory was developed by Alfredo Ortega with assistance from
Mario Vilas and Gerardo Richarte.
From the OpenBSD CVS log:
revision 1.27
date: 2007/02/26 20:15:33; author: claudio; state: Exp; lines: +2 -6
m_dup1() copies the packet header and allocates the mbuf cluster in the
wrong order. M_DUP_PKTHDR needs to be called with an empty mbuf.
Allocating an mbuf cluster beforehand is not allowed as the resulting mbuf is
no longer considered empty (part of the header is initialized).
The correct order is to allocate an mbuf via MGETHDR(), copy the packet header
and as last step allocate the cluster.
Issue found by JINMEI Tatuya.
OK canacar@ deraadt@ mglocker@ additional input itojun@
he goes around calling us all "FUCKING IDIOTS" and "INTERFACE NAZI'S"
he's right.
Re:first post
on
Define - /etc?
·
· Score: 2, Informative
Actually, "." and ".." are part of the filesystem
(i.e. they're stored on disk, as directory entries)
in the FAT filesystem.
Assuming that all filesystems are implemented
100 % similar to the one(s) you know about _is_
noob and pretentious: the implementer of the FS is free to do things
the way he sees fit as long as it provides reasonable
semantics.
In fact he doesn't have to do
directories or files at all - he may implement
everything as a big hash with different entries sharing the same blocks.
Our dumbhead president apparently tried to be nice, and managed to botched it.
He should have consulted someone more knoledgable, but even then, it takes a lot of effort & active indoctrination to make an outsider understand all this 'intellectual property' nonsense.
For a rational human being, it's just as hard to find copyright infringement morally repulsive as it is for a non-bigot to consider sex outside marriage a mortal sin.
you must be kidding with your 'widely recognized as superior' linux & freebsd.
until recently, you would rather openly admit that you're gay than being suspected of using linux or any kind of free software in Romania.
there were pockets of resistance in universities, etc - but in general, you would met with rabid reactions from microsoft worshippers (yes they were in fact pirating microsoft, but what difference does this make ?)
Third, it IS in the linux kernel, Linus accepted it just fine. He had to put it in the kernel so he could use it because Ulrich Drepper is too dense to let it into glibc.
I don't get this. Is linux (the kernel) linking against glibc ? Even if strl{cat,cpy} were in glibc, this would not make any difference for kernel programming.
Of course, having the strl* functions available as builtins in gcc (like most other string functions) would be more interesting, but I don't thing Ulrich Drepper has any say in this.
it's the second time this week that I stumble on crap open-source source that tries to run 'sudo' from 'make install' in the Makefile.
if those people even DARE to do this immensely silly thing (assuming that it will work), this means that are thousands and thousands of 'linux' idiots where it is the norm to simply prepend 'sudo' to a command to run it as root.
when searching through google (I was searching for a zip code list in plain text, something that the Post in my country considers its intelectual property:)) I have found a sudoers file with some thirty users that could run any command without password. That was the home directory of a senior administrator of a well-known ISP (1-2 million users).
and neither the kde nor the gnome drones were able to write a decent tetris clone, that does drop & shift, like the original Pazhitnov's. or a terminal emulator (just try kterm, gnome-terminal and xterm on a remote X11 session, through a ssh channel).
kde/gnome are good for presentations - to show that 'linux' is 'like windows'.
they're unix tools in fact - they do one thing well (look like Windows), and JUST that.
The whole network stack is called the Open Systems Interconnection Reference Model, the OSI Reference Model, or even just the OSI Model. It was published in 1984 by both the ISO, as standard ISO 7498, TCP/IP is an implementation of parts of this stack.
How could TCP/IP be just an implementation of the OSI stack ?
TCP/IP predates OSI, and is based on a different model (DoD).
Of course,
you can teach TCP/IP in terms of OSI layers to innocent students,
but good luck _implementing_ it that way:)
You're like saying that UNIX is an implementation of parts of VMS.
I'm sick of the 'don't re-invent the wheel' argument being dragged out and used to justify people not studying properly, or for that matter, not teaching properly.
I've recently met a Python programmer (who has written some very large & complex applications)
who simply wasn't able to come up with an "algorithm" to calculate how many Sundays are
between two random dates in the calendar.
Slashdot Mirror
No, that is wrong. Perl threading (ithreads - as currently implemented in 5.x) DOES really copy the entire interpreter state.
And the implementation is incredibly dumb: think of a fork() that (instead of copying the entire memory at once or doing COW on pages on VM systems, etc.) will walk painfully through all the pointers, strings and objects and copy each other in turn, recursively.
You come to wonder if the people who designed it were really morons, or it was some kind of joke.
I really don't understand what that guy want to demonstrate with this.
His ignorance about how the preprocessor works ?
He's also not able to make the difference between matters of just style and code that's invoking undefined behaviour (see the part about "lawyer's code").
Thanks for having wasted my time with this crap.
His lawyer seems to be a moron, and not even a motivated one.
Looks like Hans Reiser has really screwed up with his defense,
just as he did with his business in the past.
but this time it was about saving his life, not pushing some
piece of shit filesystem/database hybrid that nobody has any
use for.
'open source' was an attempt to confiscate the Free Software movement, and build a 'business model' on other people's work.
If the political & religious (sic) implications of "Free as in Freedom" make you nervous, just stay away from the whole thing.
Don't forget that FSF started by actually paying people salaries for writing free software; people have to eat in order to write software, free or proprietary.
Those 'open source' lies about people writing working software in their free time remind me of sport in socialism, where professional athletes were supposed to work hard in factories, and train only during the weekend.
Nobody is writing a functional driver at home at night after 10 or 12 hours of hard, stressing day work.
If you have to lie for corporate ideological reasons, at least don't pretend that people are actually believing you :)
s/diplomatic/hypocritical/
and stop that silly drivel now: if I driver is complete and isn't written with obfuscation in mind, all the great 'secrets' about the workings of the device should be already compromised.
they want people to sign NDAs in order to force them to write half-functional, obfuscated crap so they could keep the impression that their proprietary model is still the rule, and try to shove it again down people's throat after that 'open-source' fad dies out.
it's ironic that OpenBSD came to be more close to the ideals of free software (as of FSF) than linux - but linux just became too important, and unscrupulous self-promoting assholes and attention whores are already considering it their exclusive playground.
No, it's like saying that dogs exist, cats exist, but God does NOT exist.
I don't believe you.
define "Eastern Europe".
Estonia is not Romania or Bulgaria.
It's a small, moderately prosperous and racis^WWestern Democratic Values bla bla bla country in /Northern/ Europe.
Most Estonian will be happy to explain you how *all* the crime in their country is carried out by Russians or more recent immigrants. You know, stuff as usual.
I got it.
It's like saying: "everything's so simple, even an American can figure it out."
Well, the artistic and litterary taste of computer geeks is ALWAYS embarassing.
With the time you'll learn to get over it.
The OpenBSD guys at least don't push so much Tolkien, Monty Python, Star Trek or other exasperating crap. That alone is quite refreshing.
Congratulations.
But I still think they had to give proper credits in their CVS log. Not doing so in the first place + trying to play down the importance of the bug was dishonest and despicable.
I do use OpenBSD too, and appreciate the OpenBSD team, but I think they somehow deserve a bit of undeserved public humiliation for this :)
From the bugtraq advisory:
From the OpenBSD CVS log:
So, who found the bug in the first place ?
he's right.
Actually, "." and ".." are part of the filesystem (i.e. they're stored on disk, as directory entries) in the FAT filesystem.
Assuming that all filesystems are implemented 100 % similar to the one(s) you know about _is_ noob and pretentious: the implementer of the FS is free to do things the way he sees fit as long as it provides reasonable semantics.
In fact he doesn't have to do directories or files at all - he may implement everything as a big hash with different entries sharing the same blocks.
Our dumbhead president apparently tried to be nice, and managed to botched it.
He should have consulted someone more knoledgable, but even then, it takes a lot of
effort & active indoctrination to make an outsider understand all this
'intellectual property' nonsense.
For a rational human being, it's just as hard to find copyright infringement morally
repulsive as it is for a non-bigot to consider sex outside marriage a mortal sin.
you must be kidding with your 'widely recognized as superior' linux & freebsd.
until recently, you would rather openly admit that you're gay than being suspected
of using linux or any kind of free software in Romania.
there were pockets of resistance in universities, etc - but in general, you
would met with rabid reactions from microsoft worshippers (yes they were in fact
pirating microsoft, but what difference does this make ?)
what kind of notation are you using ?
1^n == 1, no matter how big 'n' is.
your example is syntactically incorrect, and the order of arguments to fgets() is completely wrong.
and that example of using a for(;;) loop gratuitously is crap anyway ...
I doubt that google is archiving the same spam twice.
:)
Just think about the rsync algorithm or the venti backup system.
It turns out they don't HAVE TO archive such massive
volumes of spam. Unless they're incompetents.
I don't think the entropy of spam grows faster than
the size of their disks
I don't get this. Is linux (the kernel) linking against glibc ?
Even if strl{cat,cpy} were in glibc, this would not make any
difference for kernel programming.
Of course, having the strl* functions available as builtins
in gcc (like most other string functions) would be more
interesting, but I don't thing Ulrich Drepper has any say in
this.
it's the second time this week that I stumble on crap open-source source that tries to run 'sudo'
:)) I have found a sudoers file with some thirty users
from 'make install' in the Makefile.
if those people even DARE to do this immensely silly thing (assuming that it will work), this means that
are thousands and thousands of 'linux' idiots where it is the norm to simply prepend 'sudo' to a command
to run it as root.
when searching through google (I was searching for a zip code list in plain text, something that the Post
in my country considers its intelectual property
that could run any command without password. That was the home directory of a senior administrator of a
well-known ISP (1-2 million users).
I said 'remote'.
:)
Compare a xterm/X11-classic application to a kde/gnome-thing on a connection with
latency > 20ms and small bandwidth.
Of course, everything works like a charm with the display on a laptop in my bathroom
and the apps on a machine in my basement
yes, iagno.
and neither the kde nor the gnome drones were able to write a decent tetris clone,
that does drop & shift, like the original Pazhitnov's.
or a terminal emulator (just try kterm, gnome-terminal and xterm on a remote X11
session, through a ssh channel).
kde/gnome are good for presentations - to show that 'linux' is 'like windows'.
they're unix tools in fact - they do one thing well (look like Windows), and JUST that.
How could TCP/IP be just an implementation of the OSI stack ? TCP/IP predates OSI, and is based on a different model (DoD).
Of course, you can teach TCP/IP in terms of OSI layers to innocent students, but good luck _implementing_ it that way :)
You're like saying that UNIX is an implementation of parts of VMS.
I've recently met a Python programmer (who has written some very large & complex applications) who simply wasn't able to come up with an "algorithm" to calculate how many Sundays are between two random dates in the calendar.
I'm not kidding.