The chip used in the broadcom kit is used in some military applications. As such even if Broadcom wanted to open up spec's they wouldnt be allowed to as the powers that be would deem it a threat to national security (probably something to do with the encryption side of it).
Care to substantiate that claim?
When my employer was picking a wireless chip for an embedded system, Broadcom was unwilling to give us anything other than binaries even though we were willing to do whatever NDA stuff was required. I don't know the reason behind it, and always assumed it was because they're like most other wireless hardware vendors. Since source is an absolute requirement (we have a strange Linux system and a few architectures to support), we ended up with an Intersil chip.
AFAIK, ndiswrapper won't work, the windows driver is an x86 binary.
"Airport Extreme" is really just the Apple name for a broadcom chip. Broadcom does not provide documentation for their chip, the OS X driver is binary only.
Even if you do not alter the GPL'd source, you must provide the source when distributing binaries, or alternatively include a written offer to provide it on demand.
See item 3 of the terms and conditions. An exception is given for non-commercial distribution, but that doesn't apply here.
From the anaconda release notes: "The anaconda installer works on a wide variety of Linux-based computing architectures (ia32, Itanium, Alpha, S/390, PowerPC), and is designed to make
it easy to add platforms."
FWIW, there's a screen mailing list at gnu-screen@yahoogroups.com... I've put up a few other screen links at the debian screen package page
remapping "mouse buttons" under Linux
on
OS X Hacks
·
· Score: 1
In Linux, you can map an arbitrary keycode to look like a mouse button. On my ibook, I have the funky enter key to the right of the space bar look like a middle click, and F11 is right click (I rarely right click, but X11 cut and paste is all about middle click).
I rarely use OS X natively, so I don't know how to "fix" this "problem" in it.
The version of screen in Debian stable is 3.9.11-5, and it has no problems displaying the same window to multiple screen clients simultaneously. You're trolling, or you're doing something horribly wrong.
If you think this is wrong, please file a bug and I'll walk you through a simple example of how to do this in my reply to the bug report.
All those are cool, except for the case insensetive nfs server for unix users who _expect_ a case sensetive filesystem.
What happens when I check out a project from cvs that has a directory named 'cvs' in it? That's right, it sucks. The same goes for tarballs with a README and a readme, or configure and CONFIGURE, etc.
No, it decodes mp3s in software. The board has two arm processors, and it's probably safe to bet that one is used for decoding the audio while the other is used for everything else.
You could also print to postscript in windows... I think it involves installing a driver for an HP 4M and setting it up to print to file or something... crazy windows.
I use johncompanies to host a coupla domains now, and they're pretty good. I'd give them a better review, but I have a few experiences that have made me not as happy as I could be:
They'll change stuff on your filesystem without notifying you. I've had this happen 2 times already:
changing the perms on/dev/kmem so that only root could read it (broke top for normal users, security risk according to them)
disabling daily/weekly/monthly crontabs in/etc/crontab. This was to avoid daily reboots of the box when everybody's daily cronjobs would trigger at the same time and take it down.
Both times, I emailed them asking what was going on, and asking them to at least notify me whenever they change files in my jail. I emailed them a couple of times over the course of 2 months about the crontab disabled thing. I got "we're still looking into that" and then no replies. I later re-enabled it in my jail.
Software upgrades suck. I guess this is the nature of the beast with FreeBSD. My box is stuck at FreeBSD 4.5, and since there's limited filesystem space I'm not going to make world to upgrade base (this is a problem when there are security issues in base - yes, I know of the overwrite-base ports). I'm also reluctant to upgrade the base system without upgrading the kernel (johnco controls that part).
Non existent notifications of system changes, security issues, etc. I guess this is semi related to the first issue, but it would be nice to know what's going on when they have to change things in my jail. It would also be nice to know that when there is a security issue with software in the base system (i.e. almost guaranteed to be software in every jail they run), that johnco has a solution to the problem.
Despite those issues, I'm a pretty happy customer, and have even recommended other people to them. If you're really interested in the service, I'd suggest testing it out first to see if it's a good fit for what you want to do.
With this new round of ibooks I'd be more concerned with being able to put it to sleep than with getting hardware accel working.
This new rev of ibooks ships with the ATI Mobility 7500 chipset. The ppc kernel hackers are waiting on help from ATI to get it to play nice when put to sleep. From what they say, ATI is pretty responsive for x86 linux stuff, but they're pretty silent when it comes to "mac" ATI cards.
Actually, you can remap caps lock to control using ucontrol in OS X. There's a linux patch to do the equivalent as noted in the post you linked to. It was based on the stuff from ucontrol (called icontrol back then).
I used that patch in Linux and ucontrol in OS X for almost a year without too many big problems. Occasionally you'd have to hit caps lock (ctrl) when coming back from sleep. Nothing too big.
This is qualitative, not quantitative, unfortunately, but I installed OS X on my powerbook G4 with ufs once. The install alone took over an hour. Booting OS X took forever, things started up slowly, I thought OS X was a dog.
Later I was reinstalling OS X (repartitioning the disk a bit) and decided to go to HFS+ because a friend said OS X on his powerbook with the same hardware wasn't nearly as slow as I was describing. I re-installed onto HFS+ and the install only took 20 minutes this time and things ran noticably faster.
Next time, I'll run bonnie++ on both and find out for sure how much slower ufs is than HFS+. Regardless, my qualitative feeling is that OS X's UFS sucks performance wise.
My biggest nit pick is that I wish Portage had a better way of tracking changes between package versions. Sometimes the only difference between two versions is a few lines of a Gentoo-supplied script or config file. When you upgrade the package it forces you you to recompile the whole thing, even though the changes didn't do anything that would have affected compilation.
I don't know if there's a good Gentoo specific answer to your question, but you could speed up the n+1th c/c++ compile if you used ccache to compile your software. See ccache.samba.org.
ccache and its predecessor, compilercache, save me large quantities of time compiling things for work and play.
Slashdot Mirror
Care to substantiate that claim?
When my employer was picking a wireless chip for an embedded system, Broadcom was unwilling to give us anything other than binaries even though we were willing to do whatever NDA stuff was required. I don't know the reason behind it, and always assumed it was because they're like most other wireless hardware vendors. Since source is an absolute requirement (we have a strange Linux system and a few architectures to support), we ended up with an Intersil chip.
AFAIK, ndiswrapper won't work, the windows driver is an x86 binary.
"Airport Extreme" is really just the Apple name for a broadcom chip. Broadcom does not provide documentation for their chip, the OS X driver is binary only.
Now if only they'd support freaking WebDAV over SSL...
Even if you do not alter the GPL'd source, you must provide the source when distributing binaries, or alternatively include a written offer to provide it on demand.
See item 3 of the terms and conditions. An exception is given for non-commercial distribution, but that doesn't apply here.
From the anaconda release notes: "The anaconda installer works on a wide variety of Linux-based computing architectures (ia32, Itanium, Alpha, S/390, PowerPC), and is designed to make it easy to add platforms."
FWIW, there's a screen mailing list at gnu-screen@yahoogroups.com ... I've put up a few other screen links at the debian screen package page
In Linux, you can map an arbitrary keycode to look like a mouse button. On my ibook, I have the funky enter key to the right of the space bar look like a middle click, and F11 is right click (I rarely right click, but X11 cut and paste is all about middle click).
I rarely use OS X natively, so I don't know how to "fix" this "problem" in it.
See groups.google.com... rms changed the OS detection in emacs to spit out lignux or something...
("regular user" != Anonymous Coward)
The version of screen in Debian stable is 3.9.11-5, and it has no problems displaying the same window to multiple screen clients simultaneously. You're trolling, or you're doing something horribly wrong.
If you think this is wrong, please file a bug and I'll walk you through a simple example of how to do this in my reply to the bug report.
Upgrade your version of screen, trollboy.
Speaking as the Debian maintainer for screen, I can tell you that multiple displays showing the same window works fine in any recent version (3.9.x).
You're doing something wrong, or your ancient screen version doesn't support it.
I use screen -x daily, and it works fine to have a single window displayed on multiple displays.
You should definitely upgrade from 3.7.6 though... 3.9.15 is the latest release.
All those are cool, except for the case insensetive nfs server for unix users who _expect_ a case sensetive filesystem.
What happens when I check out a project from cvs that has a directory named 'cvs' in it? That's right, it sucks. The same goes for tarballs with a README and a readme, or configure and CONFIGURE, etc.
No, it decodes mp3s in software. The board has two arm processors, and it's probably safe to bet that one is used for decoding the audio while the other is used for everything else.
You can print to pdf natively in OS X.
You could also print to postscript in windows... I think it involves installing a driver for an HP 4M and setting it up to print to file or something... crazy windows.
I use johncompanies to host a coupla domains now, and they're pretty good. I'd give them a better review, but I have a few experiences that have made me not as happy as I could be:
I've had this happen 2 times already:
- changing the perms on
/dev/kmem so that only root could read it (broke top for normal users, security risk according to them)
- disabling daily/weekly/monthly crontabs in
/etc/crontab. This was to avoid daily reboots of the box when everybody's daily cronjobs would trigger at the same time and take it down.
Both times, I emailed them asking what was going on, and asking them to at least notify me whenever they change files in my jail. I emailed them a couple of times over the course of 2 months about the crontab disabled thing. I got "we're still looking into that" and then no replies. I later re-enabled it in my jail.Despite those issues, I'm a pretty happy customer, and have even recommended other people to them. If you're really interested in the service, I'd suggest testing it out first to see if it's a good fit for what you want to do.
Looks like they want a binary similar to the one described in A Whirlwind Tutorial on Creating Really Teensy ELF Executables for Linux , except it has to print text and not just return 42.
If only I had some spare time to play along at home...
With this new round of ibooks I'd be more concerned with being able to put it to sleep than with getting hardware accel working.
This new rev of ibooks ships with the ATI Mobility 7500 chipset. The ppc kernel hackers are waiting on help from ATI to get it to play nice when put to sleep. From what they say, ATI is pretty responsive for x86 linux stuff, but they're pretty silent when it comes to "mac" ATI cards.
Actually, you can remap caps lock to control using ucontrol in OS X. There's a linux patch to do the equivalent as noted in the post you linked to. It was based on the stuff from ucontrol (called icontrol back then).
I used that patch in Linux and ucontrol in OS X for almost a year without too many big problems. Occasionally you'd have to hit caps lock (ctrl) when coming back from sleep. Nothing too big.
Now are you going to stop your whining?
It's slow in comparison to Linux on the same hardware. See the lmbench numbers at http://clustermonkey.org/~laz/pbook/rob.lmbench.tx t
http://clustermonkey.org/~laz/pbook/rob.lmbench.tx t
See wondershaper for a semi user friendly script and the Linux Advanced Routing & Shaping HOWTO for docs if you want to tweak it.
I use newsisfree.com to glob together the headlines from my usual news sources. I migrated there shortly after geekboys.org died.
This is qualitative, not quantitative, unfortunately, but I installed OS X on my powerbook G4 with ufs once. The install alone took over an hour. Booting OS X took forever, things started up slowly, I thought OS X was a dog.
Later I was reinstalling OS X (repartitioning the disk a bit) and decided to go to HFS+ because a friend said OS X on his powerbook with the same hardware wasn't nearly as slow as I was describing. I re-installed onto HFS+ and the install only took 20 minutes this time and things ran noticably faster.
Next time, I'll run bonnie++ on both and find out for sure how much slower ufs is than HFS+. Regardless, my qualitative feeling is that OS X's UFS sucks performance wise.
My biggest nit pick is that I wish Portage had a better way of tracking changes between package versions. Sometimes the only difference between two versions is a few lines of a Gentoo-supplied script or config file. When you upgrade the package it forces you you to recompile the whole thing, even though the changes didn't do anything that would have affected compilation.
I don't know if there's a good Gentoo specific answer to your question, but you could speed up the n+1th c/c++ compile if you used ccache to compile your software. See ccache.samba.org.
ccache and its predecessor, compilercache, save me large quantities of time compiling things for work and play.