> If you remember... the Xbox runs on a right off the shelf Pentium.
> Now, I don't know a lot about linux, but if you can put the kernel libs on a cd that the XBox can read..., it wouldn't require any special kernel hacking
The kind of processor used in the machine is the least significant problem here.
My fears:
If the firmware refuses to boot non signed-DVDs (ie: loads code to memory, then check that there is a signature for it on the disc itself), it can be very difficult to boot non MS approved software on the X-box.
Hacking around this would require rewriting various low-level components (BIOS/firmware), which would be extremely difficult, and would need physically modifying the machine (unless you have a BIOS/firware patcher that would be signed, which would be really unlikely).
Do you wonder why encription is now exportable ? This is probably the answer. Anyway, expect most future hardware to have this kind of protection.
And don't forget that under DMCA, it will be illegal for USIan to defeat such a protection. In that case, it would be a unhackable product. Plain and simple.
Oops. Forgot to answer the question. The thing uses the Spigot Algorithm, by Stanley Rabinowitz and Stan Wagon.
See (for instance) <http://www.physik.tu-muenchen.de/lehrstuehle/T 32/matpack/html/Mathematics/Pi.html> to get a description. Remove slashdot added spaces from the URL.
So, definitively, no. You won't be a contender with an extremly common stolen program.
> Although I am clearly biased on this point, I just dont see any other need for a tech-union, perhaps someone else can enlighten me on this issue.
Just answer those simple/not-so-simple questions:
1/ How old are you ?
2/ How many hours a week do you work ?
3/ How much holidays a year do you have ?
4/ Do you have children or a familly ?
5/ What do you want to do with your life ?
> Then, you install _every available linux package in the known universe_, even those that don't compile, leaving your drive full of useless tar files.
bla-bla
> How much of the HD have we used, at this point? 8, 10 gigs? That's 1/4 of the space of this, by today's standards, fairly conservatively sized hard drive.
It is much more probably in the Tera range. Anyone could make an educated guess ?
> The latest trend in DSS is using emulation software on a PC to intercept the signal and then sending it to your reciever. It truly is an innovative solution!
Mmm. Using computers hooked to the receiver to decrypt TV signal have been done since a long time (often emulating the missing circuit). Or maybe I don't understand what you meant.
They did the right thing. A nice fight, and a regular victory. First round for the hackers, second round for DirectTV.
Waiting for last round, but both side have my support. This is the way the fight should be done.
This recall me the old time of copied adventure games on the Apple ][. There was one (don't remember its name), in which you had to escape from a jail, and the copy protection was smart enough to slighly change the game logic, so you could NOT escape. And there was an (rather hard to find) inscription on the toilet room (IIRC), that said, more or less: "You don't expect beeing able to complete the game with a pirated version, do you ?".
This is the way things should be done. Hardware companies should try to outsmart hackers/crackers which should try to do the same.
> I`d have used a case statement, instead of the if..else if business!
Actaully, I first wrote it with an added if (n<0) {/* Error handling case */ } that cannot be dealt nicely in a switch. Then I removed the error test, because I felt it was pedantic to post it that way (more precisely, in real world, I would expect an "assert( n>=0 )" before the if/else)
Btw, I generally avoid the switches because it is sometimes difficult to add new entries (it is not difficult, but it is error prone in the case where you need 'n<0', 'n==0', 'n==1' and 'n>=2', because you have to turn the whole switch into an if-else. I learnt the hard way to minimize changes to working code). Also, switches fails on non-integral types, so you need to master two idioms (one for the integral types, one for the floating point or function call decision tree, etc, etc)
The less idioms I use, the less bugs I make, so I now only use switches when the tested value is an enumeration (because it never needs to be changed into an if/else), and where the default case is error handling or not used (so if I extend the enumeration I'll get either a warning or a run-time error, not a software that silently do something unexpected).
It is a question of style, and you may feel better with a switch in that case.
This is the original Duff device (in pre-ANSI C). A gem, that the ANSI comitee carefully looked at and declared valid.
Take a look at how the loop unrolling is performed. See how the loop is misplaced between the first and second case label. This is beautifull, because it have been a creative act to even think that the compiler could grok it.
send(to, from, count)
register short *to, *from;
register count;
{
register n=(count+7)/8;
switch(count%8){
case 0: do{ *to = *from++;
case 7: *to = *from++;
case 6: *to = *from++;
case 5: *to = *from++;
case 4: *to = *from++;
case 3: *to = *from++;
case 2: *to = *from++;
case 1: *to = *from++;
} while(--n>0);
}
}
See author comments at:
http://www.lysator.liu.se/c/duffs-device.html
(But clearly, I would not want that in production code)
Btw, what I call beautifull code is crystal clear code. The first ncsa httpd was of that kind (imho). Anyway, beautifull code matters much less than beautifull _interfaces_.
Anyway, this is too a bad idea, as those constructs are not localisable. You'd better aim at something like:
if (n==0)
printf( LOCALIZE( "No filed copied" ) );
else if (n==1)
printf( LOCALIZE( "1 file copied" ) );
else
printf( LOCALIZE( "%d files copied" ), n );
for whatever definition of LOCALIZE you happend to have.
Longer, more boring, but definitely nicer code.
Cheers,
--fred
Re:Do this on other platforms
on
2.2 vs 2.4
·
· Score: 1
I'm going to stop responding here.
> Use cvsup. Look at/usr/share/examples/mumble for a working template supfile and edit out the ports you arent interested in. (PS: man cvsup.)
I probably have mis-expressed myself. I do cvsup the port tree every night. My problem is different:
Say I install/usr/ports/foo/bar
cd/usr/ports/foo/bar
make install
One week later, I update the port tree
cvsup -L 2 ports-supfile
My port tree is now up-to date, which is cool if I want to seek for new tools, but the bar binary is not up to date.
Good.
cd/usr/ports/foo/bar
First problem: sometimes bar have moved away. For instance, the mac utilities moved around in the port tree a few month ago. Not a big deal, but human intervention is needed.
Most problematic, is that by doing this blindly I ended up with multiple versions of libraries installed and binaries that stopped working.
Last but not least, I currently have 153 ports installed on my machine (slowly migrating to a freebsd-only environment. I expect to have much more than that package count at the end). The cvsup/make approach is painfull, needs huge disk space for compilation, and take a lot of time (compiling mozilla is not a fast operation, and keeping the 1Gb objects files laying around is not a pleasant idea).
Grabing the packages in binary is easier for some package, but in those case, the cvsup approach won't work.
There is nothing aggressive in pointing this. The package system is even planned to be revamped on BSD. And, yes, I beleive that apt-get is superior in *some* points of view (in particular, it enable a binary upgrade of installed software in a single command, something like a working "pkg_version -c |/bin/sh" [which sorta works but may screw dependencies])
The is no automated packages updating system in FreeBSD. That is even pointed in the BSD documentation. No need to start a holly war here.
Cheers,
--fred
Re:Do this on other platforms
on
2.2 vs 2.4
·
· Score: 1
> What can I tell you, stick with Linux and when you learn something about unix, give freebsd another try
You are probably the same asshole that responded to the other post.
Why are you anonymous ? What are you afraid of ?
I'd like to measure the length of our respective dicks. I think you need a little public embarassment. I am offensed. Let's do a coding contest.
Please, reply while logged, so I can know who you are (you can easily know who I am if you know how to use the net).
Cheers,
--fred
Re:Do this on other platforms
on
2.2 vs 2.4
·
· Score: 1
Dear anonymous asshole,
> How goddamn hard is that for linux fed weenies to understand.
I suspect that I probably used UNIX before you. In any way, I started using UNIX way before linux or freebsd existed. To be precise, I started using unix in september 1986. Some people on slashdot started before, but not that much.
> make mutt
> make inn
> make this
> make that
bash-2.03$ pkg_version | wc -l
143
bash-2.03$
In the same vein, are you aware that the build tree for mozilla weight 1Gb ? Should I keep that around, dear asshole ?
> We dont download the entire source, just the diffs.
Why do you beleive I don't know how CVSup works ? I can even read the modula-3 source, which is probably not your case (most agressive asshole I met were pretty lame developers. You are probably not an exception)
I don't care about the bandwidth. But, I do care when ports are moved around in the port tree, and that what used to be at one place moves into another, breaking the update scripts. I also do care when two ports on the system requires different version of ORBit. Or when a port suddently stop compiling because of a discovered unfixed security hole.
In one word, I do care when there is a need for *human* intervention.
> Stick to linux ?
Why do you think I am using linux ? I am not.
12:18:42|152 [ladybug:~] fred% hostinfo
Mach kernel version:
Kernel Release 5.5:
Fri Jul 2 13:10:26 PDT 1999; root(rcbuilder):Objects/kernel-154.14.obj~10/RELEA SE_PPC
Copyright (c) 1988-1995,1997-1999 Apple Computer, Inc. All Rights Reserved.
Re:Small niggle with the article
on
2.2 vs 2.4
·
· Score: 1
I used to think that, but I recently hacked a cheap machine for use at home. ATA-66 (VIA 82C686), 40Gb EIDE drive (Maxtor 54098H8).
Under FreeBSD 4-2:
su-2.03# dd if=/dev/ad0s1 of=/dev/null bs=524288
1302+1 records in
1302+1 records out
682665984 bytes transferred in 23.191338 secs (29436248 bytes/sec)
su-2.03#
This almost 30Mb/s sustained. I was shocked. I thought I would get a much lower number.
SCSI is still superior when dealing with mutliple devices, or when doing complex thing, but EIDE is pretty good those days...
Cheers,
--fred
Re:Do this on other platforms
on
2.2 vs 2.4
·
· Score: 1
> Better to edit make.conf once and for all according to your installation, then do a cvsup && make world_ports_whathaveyou.
Interesting idea. I already do something similar with a shell script, but failed on me several times when the port tree is re-organised.
Cheers,
--fred
Re:Do this on other platforms
on
2.2 vs 2.4
·
· Score: 1
I am already aware of the various incantations of pkg_version.
Let's look at the man page for pkg_version:
-c
Enable commands output. Commands output includes the commands
you should type to update your installed packages to the latest
versions in the ports system. This feature does not constitute
an automated packages updating system. The output of this com-
mand must be edited, in order to avoid destroying dependencies
between installed packages.
I don't think this qualify as 'more powerfull than debian apt-get' (but, I do admit it is usefull). The AC I was replying to stated "The ports system is superior to debian packages (in particular, upgrade.)", which is doubtfull.
Btw, 'pkg_version -c' is only good for an install from source, which can be very long for packages like X, gnome or mozilla. I was asking for "upgrade the installed ports/added packages binaries". Afaik, it have to be done by hand, and is (IMHO) really tedious. This contrast sharply with the upgrade of the base OS which is increadibly easy, or the installation of a new port/package.
I do appreciate that you took time to answer, anyway,
> I suspect it was more of a case of "This is what we have now, if you don't like it, help fix it."
"Don't like" is not a good motivation to fix something. If it ain't broken, don't fix it.:-)
More seriously, recall the mindcraft study. Sure, the benchmarks were founded by M$ and obviously biased, but the bottom line was that TCP/IP performance sucked on SMP (or, more precisely, the whole kernel sucked on SMP [*]). This study have been based to death here. People were screaming that the tests were wrong, that mindcraft lied, etc, etc. It tooks weeks to settle down. Beside few minor points, the study was right, and this helped tremendously the kernel development.
Saying "This is what we have now, if you don't like it, help fix it." is okay, but it often turn into "If you are not going to fix it, shut up", which is much less sane.
Unfortunately most posters here have the "Don't say anything bad about it" attitude that I was pointing to. (For instance the AC in another thread that tells me that the port system of FreeBSD is superior to apt, in particular for upgrades).
Cheers,
--fred
[*] Naked truth. I am/was a NeXTstep/OPENSTEP lover. Well, I can tell you that NeXTstep performance really sucked on SMP (ie: it just ran on a single CPU)
Re:Do this on other platforms
on
2.2 vs 2.4
·
· Score: 1
> Incorrect. The ports system is superior to debian packages (in particular, upgrade.)
This is very good news for me, as I have a few FreeBSD boxes out there. Can you tell me how I am supposed to upgrade the installed ports/added packages binaries to the latest versions ? (This is a serious query)
In one word, where is the
apt-get upgrade / apt-get dist-upgrade
of FreeBSD ?
Cheers,
--fred
Re:Do this on other platforms
on
2.2 vs 2.4
·
· Score: 1
> I mean't, the kernel, I don't consider the rest of the distribution to be the O/S.
You said "full version release", so I suposed that it was the whole system.
Now, if you consider that the kernel is the OS, you "OS" is not even self-hosting. Pretty unfortunate, don't you think ?
And upgrading the kernel without upgrading modutils/libc/etc, is really asking for troubles. In particular in the context of the current article (A whole new 2.4. Most people would upgrade from 2.2)
Cheers,
--fred
Re:Do this on other platforms
on
2.2 vs 2.4
·
· Score: 1
> If I'm not mistaken, Debian can do this
This is very possible (IMHO, Debian is the linux distro that sucks less). But generally debian is not at the bleeding edge, hence I suspect that the poster did not use debian to get a 2.4 kernel.
> Debian's about as close as it gets to ports
The few commands I posted are not related to ports. Ports is for the software that is _not_ included in the OS. The few command I posted upgrade the base OS (ie: what is located in/usr/src)
Btw, debian is superior to the port system. The port system sucks by many aspects (in particular upgrade).
Cheers,
--fred
Re:Do this on other platforms
on
2.2 vs 2.4
·
· Score: 2
> On your lunch hour, download the next full version release of your operating system (service packs are for sissys). Configure it, compile it, install it and be back working with no problems by the end of the hour. That's what I did.
The *full* version ? This means everything from the kernel to the applications, including the C library. On what linux distro did you do that ?
Btw, I do it once a week on my operating system (freebsd)
cvsup -L 2 stable-supfile
make buildworld
make buildkernel KERNEL=SIDONIE
make installkernel KERNEL=SIDONIE
make installworld
mergemaster
Did you really do this on your platform, or are you just making noises ?
> As far as using it as a server, 2.4 is FAST. Much faster than 2.2. Our SCSI RAID goes about 3x faster and NFS goes twice as fast (over gigabit ethernet).
Rotfl. So does it means that 2.2 sucked despite all the claims that it was a server-class OS ?
I used 2.0 and 2.2 for a loooong time. A lot of things sucked badly (overall, it sucked less then the NT4 boxes I rhad), but I generally got bashed when pointing those suckage. Now, it is pollitically correct to talk about shortcomings of 2.2 ? (As long as I don't point any 2.4 problem...)
Sometimes, I receive junkmail that is manually put in my mailbox (ie: no address on it, I receive the same as everyone else).
I love those. I fake the name and address of my best friends, and fill the card with bogus information. Quite pleasant. I also do at least one little error in the first/lastname/address, and tell my friends about it. Then, I have this warm feeling because I know that sometimes, somewhere, somebody is thinking of me.
Btw, when I order something by mail, or give my address for whatever purpose, I _always_ make a slight error in the address (for instance, there are no appartments where I live, so I add a random appartment number). This way, I know who sells my name/address to who. Fascinating, sometimes.
Slashdot Mirror
> If you remember... the Xbox runs on a right off the shelf Pentium.
> Now, I don't know a lot about linux, but if you can put the kernel libs on a cd that the XBox can read..., it wouldn't require any special kernel hacking
The kind of processor used in the machine is the least significant problem here.
My fears:
If the firmware refuses to boot non signed-DVDs (ie: loads code to memory, then check that there is a signature for it on the disc itself), it can be very difficult to boot non MS approved software on the X-box.
Hacking around this would require rewriting various low-level components (BIOS/firmware), which would be extremely difficult, and would need physically modifying the machine (unless you have a BIOS/firware patcher that would be signed, which would be really unlikely).
Do you wonder why encription is now exportable ? This is probably the answer. Anyway, expect most future hardware to have this kind of protection.
And don't forget that under DMCA, it will be illegal for USIan to defeat such a protection. In that case, it would be a unhackable product. Plain and simple.
Cheers,
--fred
Oops. Forgot to answer the question. The thing uses the Spigot Algorithm, by Stanley Rabinowitz and Stan Wagon.
See (for instance) <http://www.physik.tu-muenchen.de/lehrstuehle/T 32/matpack/html/Mathematics/Pi.html> to get a description. Remove slashdot added spaces from the URL.
So, definitively, no. You won't be a contender with an extremly common stolen program.
Cheers,
--fred
> Think I could have be a contender?
Nope. Even in obfuscated C, main cannot be declared void.
Btw, you should give credit to the original author of this program, Dick T Winter. (Which btw, did not put the 'void' before main())
Cheers,
--fred
> Although I am clearly biased on this point, I just dont see any other need for a tech-union, perhaps someone else can enlighten me on this issue.
Just answer those simple/not-so-simple questions:
1/ How old are you ?
2/ How many hours a week do you work ?
3/ How much holidays a year do you have ?
4/ Do you have children or a familly ?
5/ What do you want to do with your life ?
Cheers,
--fred
bla-bla
> Then, you install _every available linux package in the known universe_, even those that don't compile, leaving your drive full of useless tar files.
bla-bla
> How much of the HD have we used, at this point? 8, 10 gigs? That's 1/4 of the space of this, by today's standards, fairly conservatively sized hard drive.
It is much more probably in the Tera range. Anyone could make an educated guess ?
Cheers,
--fred
> The latest trend in DSS is using emulation software on a PC to intercept the signal and then sending it to your reciever. It truly is an innovative solution!
Mmm. Using computers hooked to the receiver to decrypt TV signal have been done since a long time (often emulating the missing circuit). Or maybe I don't understand what you meant.
Cheers,
--fred
the congratulation to DirectTV.
They did the right thing. A nice fight, and a regular victory. First round for the hackers, second round for DirectTV.
Waiting for last round, but both side have my support. This is the way the fight should be done.
This recall me the old time of copied adventure games on the Apple ][. There was one (don't remember its name), in which you had to escape from a jail, and the copy protection was smart enough to slighly change the game logic, so you could NOT escape. And there was an (rather hard to find) inscription on the toilet room (IIRC), that said, more or less: "You don't expect beeing able to complete the game with a pirated version, do you ?".
This is the way things should be done. Hardware companies should try to outsmart hackers/crackers which should try to do the same.
Cheers,
--fred
> I`d have used a case statement, instead of the if..else if business!
/* Error handling case */ } that cannot be dealt nicely in a switch. Then I removed the error test, because I felt it was pedantic to post it that way (more precisely, in real world, I would expect an "assert( n>=0 )" before the if/else)
Actaully, I first wrote it with an added if (n<0) {
Btw, I generally avoid the switches because it is sometimes difficult to add new entries (it is not difficult, but it is error prone in the case where you need 'n<0', 'n==0', 'n==1' and 'n>=2', because you have to turn the whole switch into an if-else. I learnt the hard way to minimize changes to working code). Also, switches fails on non-integral types, so you need to master two idioms (one for the integral types, one for the floating point or function call decision tree, etc, etc)
The less idioms I use, the less bugs I make, so I now only use switches when the tested value is an enumeration (because it never needs to be changed into an if/else), and where the default case is error handling or not used (so if I extend the enumeration I'll get either a warning or a run-time error, not a software that silently do something unexpected).
It is a question of style, and you may feel better with a switch in that case.
Cheers,
--fred
> Trust me, I'm an economist...
This is fucking hilarious.
In the same vein, we can have:
"Trust me, I'm a statistician..."
"Trust me, I'm a lawyer..."
"Trust me, I'm a politician..."
...
Cheers,
--fred
This is the original Duff device (in pre-ANSI C). A gem, that the ANSI comitee carefully looked at and declared valid.
Take a look at how the loop unrolling is performed. See how the loop is misplaced between the first and second case label. This is beautifull, because it have been a creative act to even think that the compiler could grok it.
send(to, from, count)
register short *to, *from;
register count;
{
register n=(count+7)/8;
switch(count%8){
case 0: do{ *to = *from++;
case 7: *to = *from++;
case 6: *to = *from++;
case 5: *to = *from++;
case 4: *to = *from++;
case 3: *to = *from++;
case 2: *to = *from++;
case 1: *to = *from++;
} while(--n>0);
}
}
See author comments at:
http://www.lysator.liu.se/c/duffs-device.html
(But clearly, I would not want that in production code)
Btw, what I call beautifull code is crystal clear code. The first ncsa httpd was of that kind (imho). Anyway, beautifull code matters much less than beautifull _interfaces_.
Cheers,
--fred
> http://miranda-icq.sourceforge.net
> elegant c code.
> here is my favourite gem:
> printf("%d file%s copied\n", n, &["s"](n==1));
Elegant ? The one that coded that did not know C.
A less obfuscated way to get the same result is:
printf( "%d file%s copied\n", n, "s"+(n==1) );
Anyway, this is too a bad idea, as those constructs are not localisable. You'd better aim at something like:
if (n==0)
printf( LOCALIZE( "No filed copied" ) );
else if (n==1)
printf( LOCALIZE( "1 file copied" ) );
else
printf( LOCALIZE( "%d files copied" ), n );
for whatever definition of LOCALIZE you happend to have.
Longer, more boring, but definitely nicer code.
Cheers,
--fred
I'm going to stop responding here.
/usr/share/examples/mumble for a working template supfile and edit out the ports you arent interested in. (PS: man cvsup.)
/usr/ports/foo/bar
/usr/ports/foo/bar
/usr/ports/foo/bar
/bin/sh" [which sorta works but may screw dependencies])
> Use cvsup. Look at
I probably have mis-expressed myself. I do cvsup the port tree every night. My problem is different:
Say I install
cd
make install
One week later, I update the port tree
cvsup -L 2 ports-supfile
My port tree is now up-to date, which is cool if I want to seek for new tools, but the bar binary is not up to date.
Good.
cd
First problem: sometimes bar have moved away. For instance, the mac utilities moved around in the port tree a few month ago. Not a big deal, but human intervention is needed.
Most problematic, is that by doing this blindly I ended up with multiple versions of libraries installed and binaries that stopped working.
Last but not least, I currently have 153 ports installed on my machine (slowly migrating to a freebsd-only environment. I expect to have much more than that package count at the end). The cvsup/make approach is painfull, needs huge disk space for compilation, and take a lot of time (compiling mozilla is not a fast operation, and keeping the 1Gb objects files laying around is not a pleasant idea).
Grabing the packages in binary is easier for some package, but in those case, the cvsup approach won't work.
There is nothing aggressive in pointing this. The package system is even planned to be revamped on BSD. And, yes, I beleive that apt-get is superior in *some* points of view (in particular, it enable a binary upgrade of installed software in a single command, something like a working "pkg_version -c |
The is no automated packages updating system in FreeBSD. That is even pointed in the BSD documentation. No need to start a holly war here.
Cheers,
--fred
> What can I tell you, stick with Linux and when you learn something about unix, give freebsd another try
You are probably the same asshole that responded to the other post.
Why are you anonymous ? What are you afraid of ?
I'd like to measure the length of our respective dicks. I think you need a little public embarassment. I am offensed. Let's do a coding contest.
Please, reply while logged, so I can know who you are (you can easily know who I am if you know how to use the net).
Cheers,
--fred
Dear anonymous asshole,
A SE_PPC
> How goddamn hard is that for linux fed weenies to understand.
I suspect that I probably used UNIX before you. In any way, I started using UNIX way before linux or freebsd existed. To be precise, I started using unix in september 1986. Some people on slashdot started before, but not that much.
> make mutt
> make inn
> make this
> make that
bash-2.03$ pkg_version | wc -l
143
bash-2.03$
In the same vein, are you aware that the build tree for mozilla weight 1Gb ? Should I keep that around, dear asshole ?
> We dont download the entire source, just the diffs.
Why do you beleive I don't know how CVSup works ? I can even read the modula-3 source, which is probably not your case (most agressive asshole I met were pretty lame developers. You are probably not an exception)
I don't care about the bandwidth. But, I do care when ports are moved around in the port tree, and that what used to be at one place moves into another, breaking the update scripts. I also do care when two ports on the system requires different version of ORBit. Or when a port suddently stop compiling because of a discovered unfixed security hole.
In one word, I do care when there is a need for *human* intervention.
> Stick to linux ?
Why do you think I am using linux ? I am not.
12:18:42|152 [ladybug:~] fred% hostinfo
Mach kernel version:
Kernel Release 5.5:
Fri Jul 2 13:10:26 PDT 1999; root(rcbuilder):Objects/kernel-154.14.obj~10/RELE
Copyright (c) 1988-1995,1997-1999 Apple Computer, Inc. All Rights Reserved.
Kernel configured for a single processor only.
1 processor is physically available.
Processor type: ppc750 (PowerPC 750)
Processor active: 0
Primary memory available: 512.00 megabytes.
Default processor set: 78 tasks, 214 threads, 1 processors
Load average: 4.74, Mach factor: 0.17
12:18:57|153 [ladybug:~] fred%
> Stick to Linux.
You don't seem very bright, do you ?
Cheers,
--fred
I used to think that, but I recently hacked a cheap machine for use at home. ATA-66 (VIA 82C686), 40Gb EIDE drive (Maxtor 54098H8).
Under FreeBSD 4-2:
su-2.03# dd if=/dev/ad0s1 of=/dev/null bs=524288
1302+1 records in
1302+1 records out
682665984 bytes transferred in 23.191338 secs (29436248 bytes/sec)
su-2.03#
This almost 30Mb/s sustained. I was shocked. I thought I would get a much lower number.
SCSI is still superior when dealing with mutliple devices, or when doing complex thing, but EIDE is pretty good those days...
Cheers,
--fred
> Better to edit make.conf once and for all according to your installation, then do a cvsup && make world_ports_whathaveyou.
Interesting idea. I already do something similar with a shell script, but failed on me several times when the port tree is re-organised.
Cheers,
--fred
I am already aware of the various incantations of pkg_version.
Let's look at the man page for pkg_version:
-c
Enable commands output. Commands output includes the commands
you should type to update your installed packages to the latest
versions in the ports system. This feature does not constitute
an automated packages updating system. The output of this com-
mand must be edited, in order to avoid destroying dependencies
between installed packages.
I don't think this qualify as 'more powerfull than debian apt-get' (but, I do admit it is usefull). The AC I was replying to stated "The ports system is superior to debian packages (in particular, upgrade.)", which is doubtfull.
Btw, 'pkg_version -c' is only good for an install from source, which can be very long for packages like X, gnome or mozilla. I was asking for "upgrade the installed ports/added packages binaries". Afaik, it have to be done by hand, and is (IMHO) really tedious. This contrast sharply with the upgrade of the base OS which is increadibly easy, or the installation of a new port/package.
I do appreciate that you took time to answer, anyway,
Cheers,
--fred
> I suspect it was more of a case of "This is what we have now, if you don't like it, help fix it."
:-)
"Don't like" is not a good motivation to fix something. If it ain't broken, don't fix it.
More seriously, recall the mindcraft study. Sure, the benchmarks were founded by M$ and obviously biased, but the bottom line was that TCP/IP performance sucked on SMP (or, more precisely, the whole kernel sucked on SMP [*]). This study have been based to death here. People were screaming that the tests were wrong, that mindcraft lied, etc, etc. It tooks weeks to settle down. Beside few minor points, the study was right, and this helped tremendously the kernel development.
Saying "This is what we have now, if you don't like it, help fix it." is okay, but it often turn into "If you are not going to fix it, shut up", which is much less sane.
Unfortunately most posters here have the "Don't say anything bad about it" attitude that I was pointing to. (For instance the AC in another thread that tells me that the port system of FreeBSD is superior to apt, in particular for upgrades).
Cheers,
--fred
[*] Naked truth. I am/was a NeXTstep/OPENSTEP lover. Well, I can tell you that NeXTstep performance really sucked on SMP (ie: it just ran on a single CPU)
> Incorrect. The ports system is superior to debian packages (in particular, upgrade.)
This is very good news for me, as I have a few FreeBSD boxes out there. Can you tell me how I am supposed to upgrade the installed ports/added packages binaries to the latest versions ? (This is a serious query)
In one word, where is the
apt-get upgrade / apt-get dist-upgrade
of FreeBSD ?
Cheers,
--fred
> I mean't, the kernel, I don't consider the rest of the distribution to be the O/S.
You said "full version release", so I suposed that it was the whole system.
Now, if you consider that the kernel is the OS, you "OS" is not even self-hosting. Pretty unfortunate, don't you think ?
And upgrading the kernel without upgrading modutils/libc/etc, is really asking for troubles. In particular in the context of the current article (A whole new 2.4. Most people would upgrade from 2.2)
Cheers,
--fred
> If I'm not mistaken, Debian can do this
/usr/src)
This is very possible (IMHO, Debian is the linux distro that sucks less). But generally debian is not at the bleeding edge, hence I suspect that the poster did not use debian to get a 2.4 kernel.
> Debian's about as close as it gets to ports
The few commands I posted are not related to ports. Ports is for the software that is _not_ included in the OS. The few command I posted upgrade the base OS (ie: what is located in
Btw, debian is superior to the port system. The port system sucks by many aspects (in particular upgrade).
Cheers,
--fred
> On your lunch hour, download the next full version release of your operating system (service packs are for sissys). Configure it, compile it, install it and be back working with no problems by the end of the hour. That's what I did.
The *full* version ? This means everything from the kernel to the applications, including the C library. On what linux distro did you do that ?
Btw, I do it once a week on my operating system (freebsd)
cvsup -L 2 stable-supfile
make buildworld
make buildkernel KERNEL=SIDONIE
make installkernel KERNEL=SIDONIE
make installworld
mergemaster
Did you really do this on your platform, or are you just making noises ?
Cheers,
--fred
> As far as using it as a server, 2.4 is FAST. Much faster than 2.2. Our SCSI RAID goes about 3x faster and NFS goes twice as fast (over gigabit ethernet).
Rotfl. So does it means that 2.2 sucked despite all the claims that it was a server-class OS ?
I used 2.0 and 2.2 for a loooong time. A lot of things sucked badly (overall, it sucked less then the NT4 boxes I rhad), but I generally got bashed when pointing those suckage. Now, it is pollitically correct to talk about shortcomings of 2.2 ? (As long as I don't point any 2.4 problem...)
Cheers,
--fred
Sometimes, I receive junkmail that is manually put in my mailbox (ie: no address on it, I receive the same as everyone else).
I love those. I fake the name and address of my best friends, and fill the card with bogus information. Quite pleasant. I also do at least one little error in the first/lastname/address, and tell my friends about it. Then, I have this warm feeling because I know that sometimes, somewhere, somebody is thinking of me.
Btw, when I order something by mail, or give my address for whatever purpose, I _always_ make a slight error in the address (for instance, there are no appartments where I live, so I add a random appartment number). This way, I know who sells my name/address to who. Fascinating, sometimes.
Cheers,
--fred
You can cut yourself a finger and mail it back to them. It would be pretty disgusting, hence it'll probably be very effective.
Use slashback to tell us if it worked.
Thanks,
--fred