Slashdot Mirror
Linux 2.2.0 pre4
Prothonotar
writes "Kernel 2.2.0 pre4 is out." www.kernel.org being
overloaded use a mirror. Mirrors are expressions of the form
ftp.country.kernel.org or www.country.kernel.org,
such as ftp.us.kernel.org.
← Back to Stories (view on slashdot.org)
oh well.
Just curious, anyone got a list of what's new and improved? Guessing fat32 fix.
-matt
hehe I JUST compiled/booted to pre3....
btw.. WHERE THE FUCK ARE ALL THE BOOTUP MESSAGES....
I kinda like seeing these things.. lets me know my system is WORKING..
for example..
it does not display my adaptec 2940 init stuff....
but it IS loading support for it..
I have never patched a kernel to another version before, I'm asumming its the same thing as patching a kernel any other way but I downloaded the patch and ungziped it but it dosen't look like a patch. It won't untar wtf do I do to it?
I got into ftp.kernel.org!!! Woohoo!!
What I want to know is what exactly was it that was left out of 2.2.0 that Linus was talking about when he was talking about all the features that didn't make it?
Inquiring minds and all that.
Go to http://mirror.sit.wisc.edu/pub/linux/kernel if you are in the US.
Check out the patch to README. ;-)
Unfortunately, not all US mirrors are equal. Keep
trying until you get through to a US mirror that
does have it.
Well...for me, since Pre3 ( same deal with Pre4 ) make xconfig has worked...kinda. I cannot see all of the options on screen. The buttons ( with the config sections ) just keep trailing off the bottom of the screen. Is this the same for everyone else right now, and does anyone know a fix?
Does any one else notice that when you run ifconfig your ppp connection shows a hell of alot of errors and droped packets. This is a big fuckin error people! I know its not just my system because some one posted the same complaint about it on freshmeat.
or you need a newer ifconfig
I'm pretty certain that the pre patches are in sucession. Maybe you are thinking of the ac patches which do need to be reversed.
-matt
] Since edge.linuxhq.com is lagging
Well, initially, I didn't plan on making changelogs for the prepatches. I'd need to rewrite the CGI scripts I use. I'm thinking about it now, though.
-Myrdraal
I fear restarting my system with pre4...I have a cyrix chip.
Nothing at ftp.ir.kernel.org either.
- Saddam
No problems here with m2-300.
-ultimus-
The firewalling has switched from ipfwadm to ipchains. You'll have to figure it out from there since that's all I know....
Good thing you didn't waste any time RTFMing but instead rushed right over to /. to complain.
There's this thing called multitasking... :)
Even though I've never used the ntfs support in 2.1.x, I've been keeping that option enabled in my config. When I compiled pre4, it complained about it, causing the compile to error out:
:)
inode.c: In function `ntfs_decompress_run':
inode.c:421: `ntfs_s16' undeclared (first use in this function)
inode.c:421: (Each undeclared identifier is reported only once
inode.c:421: for each function it appears in.)
inode.c:423: `ntfs_s32' undeclared (first use in this function)
I then removed it from my config and all compiled well. Just thought I'd give Linus and Alan something to do!
Primer on #linuxos in efnet
Its available in AU at : ftp://ftp.linux.org.au/pub/kernel/v2.1
If you're looking for the kernel on ftp.us.kernel.org...it's in the 2.1 directory and it's fast as heck for me at the moment :)
BTW, the ftp.jp.kernel.org isn't located in
Japan at all. Last time I checked it was in
Canada (so said traceroute)
Isn't there anybody in Japan willing to take
it's place?
# I'd love to but at 128kbps bandwidth.......:)
Who do a newbie submit bugs to?
I don't feel like writing email to like Alan or Linus...
Best regards,
Steen Suder
CAB at Earthling.net
Hi, this is the AC whose 2.2.0pre3 was trashing his HD. Thumbs up to 2.2.0-pre4!
NOTE: this is a vanilla setup; IDE hdd with two partitions (linux and linux swap), networking but no masquerading/gatewaying, etc. I just wanted to get it to boot. I'm sure there'll be more work ahead... :-)
You need the following tarballs:
Install them in this order: libc, binutils, ld.so, and then it doesn't matter.
Exercise care in upgrading libc. If you do something naive like 'rm libc.so.5; ln -s /lib/libc.so.5.4.46 /lib/libc.so.5' your system will likely be hosed, because /bin/ln requires libc to load, but it doesn't exist anymore! Instead, do 'ln -s -f /lib/libc.so.5.4.46 /lib/libc.so.5'.
You don't need to worry about UNIX98 pty's yet, at least under a virgin Slackware 3.4 and later. It doesn't use /dev/ttyp* by default. If you've got a Slackware 3.4 distro that has been upgraded from earlier distros, proceed cautiously, or again, you may be locked out of your system.
Have fun!
Oh dear, I know I had this address some time
ago (a web page with several lists?)
From what I read on ipchains, (http://www.rustcorp.com/linux/ipchains/HOWTO-4.ht ml) it sounds like you may want to add a couple more things just for completeness sake. There are 3 main chains: input, output, and forward. I'm not too worried about security for the moment because I have a dynamic IP, so I just used the following commands:
ipchains -F input
ipchains -F output
ipchains -F forward
ipchains -P input ACCEPT
ipchains -P output ACCEPT
ipchains -P forward MASQ
The first 3 flush all the rules for all the chains, the second 3 set policy for each chain. Since I basically want to accept and to send all packets on all network interfaces, I set the policies to ACCEPT. It's the forwarding that I'd like to control better. What I found through experimenting is that doing the "ipchains -P forward DENY" and "ipchains -A -j MASQ -s 192.168.0.0/24 -d 0.0.0.0/0" didn't work for some reason... I couldn't get to the web (except for nslookups which bypass masquerading) from behind the masquerade box. I'm guessing that the "DENY" policy is killing the return packets from the internet-side servers.
Maybe someone with a little more knowledge could suggest a slightly more secure setup.
Don't disable SMP support in the make menu or you'll be sorry!
It could be an irq-conflict. /etc/rc.d/rc.modules)
Try booting using the distribution bootdisks.
Then mount your linux partition and stop it from loading the printerport.
(with slackware: comment out the line loading the lp module in
If you didn't compile lp as a module: steal the kernel from the bootdisk, reboot and recompile again. Try compiling both sound and lp as modules, make sure they don't load at startup.
could be: both sound and lp is using irq7 (common)
I think Alan has a fix for it, though.
Anyone been able to compile or have a working patch for the sun4 architecture ie Sparc 2 Sparc 5?MY compiles have been failing since .132 of the same error with cycles_t and noone seems to be working on the sparc port of the kernel. If anyone has any advice or has a patch please email me at orac@nospam.darkcrystal.org (remove nospam)
Thanks,
MaShaun Jones
It's just credit card info and stuff like that. Us developers need to eat too you fuggin ingrates!
Mail it to me
Linux panoramix 2.2.0-pre4-ac1 #1 SMP Sun Jan 3 16:54:29 CET 1999 i686 unknown
5:52pm up 9 min, 9 users, load average: 1.21, 0.98, 0.51
No, you are exactly correct. Anonymous access should be disabled on ftp.kernel.org. The primary site should be used to only distribute the source to authorized mirrors on down the line. It preserves the tree heirarchy and everything runs more smoothly because people learn how to use the mirror system. We have a DNS setup that works very well where we have one primary machine that no one is able to query directly but all zone updates are performed on and then 3 secondaries do zone transfers off that and then the people use the 3 secondaries instead. It's called distributing the load. So if anyone at Transmeta is listening.. close public anonymous access to non-mirror sites!
If it is not on one server, try connecting again. They're just using round robin DNS. One of the other US mirror sites might have it already.
Use tcpdump to see what is going out
linux-kernel mailing-list.
'cause merely patching the kernel is taxing enough to most newbies (the target for the 2.2.0prerelease series), you want them to back out too? it'll never happen... :-)
I was just wondering if anyone knows if any distributions have boot disks with any of the 2.2pre kernels on them. Just thought it might be a nice way to quickly test them.
Well, I had some problems with pppd myself.. My data transfer rates seem to be peeking at about 1k/s and not getting any higher. I have updated everything already.. I'm not too sure why its doing this.. 2.0.x kernels work just fine.
you can grab patches from 2.1.120 on up and several full kernel trees at
ftp://ns3.sinister.com/pub/linux
On an Alpha (LX164) - it won't compile - get this message:
util.c:16: asm/ipc.h: No such file or directory
make[2]: *** [util.o] Error 1
make[2]: Leaving directory `/usr/src/linux-2.1.132/ipc'
make[1]: *** [first_rule] Error 2
make[1]: Leaving directory `/usr/src/linux-2.1.132/ipc'
make: *** [_dir_ipc] Error 2
No kidding, we need like massive frame buffer improvements. I run a Voodoo Banshee and I have to do silly little back flips to make X Windows works. The vesa fbcon makes my card perform all shitty. PLEASE SOMEONE MAKE XFREE86 SIGN A NDA SO I CAN GET A DESENT WORKING XSERVER! shit there plaining on doing it anyway, i wish thay would hurry up.
Im SOOO glad linus finally got to getting all thouse misc patches he had sitting around merged in.. one of them happened to be a fix for buggy isa/dma setups (like in my motherboard) that when I tryed to use ANY dma's on a isa card I would get a hardware lockup..
I FINALLY HAVE SOUND FROM MY SB AWE 64 GOLD AGAIN!!!!
rock on to whoever figured out that fix btw!!!!!
Check ftp.kernel.org if you can get through - I haven't seen it on any mirrors yet.
it's crazy!
Um, dude? Who said anything about downloading "patches" (whatever those are) from kernel.org? I just download the entire source tree every time - why not? That's what a cable modem is for, fast downloading! Sure beats having to figure out cryptic UNIX shit like "patches".
I have an old Cyrix P150+ from a while ago. It's been working fine with my 2.0.34 kernel. I keep hearing strange things about Cyrix chips with the new 2.2Pre series kernel, however. What exactly is broken with older non MMX Cyrix chips? What should I do to avoid problems?
Of course it works with Cyrix! Linux is the greatest OS ever written.
More FUD from Microsoft.
It really needs to be in development soon. Everyone seems to be coming out with usb stuff now.
My major problem is getting the X to run properly.
I am currently sending out this SOS through LYNX.
The X gives me error as below
_X11TransSocketOpen: socket () failed for local
_X11TransSocketOpenCOTSClient: Unable to open socket for local
_X11TransOpen: transport open failed for local/simon:0
I am running KDE 1.0 as my WM. SuSE 5.2 distribution. Would someone tell me how to get it running? It seems that I have this problem with 2.1.132 version as well. Thank you very much for your attention.
My major problem is getting the X to run properly.
I am currently sending out this SOS through LYNX.
The X gives me error as below
_X11TransSocketOpen: socket () failed for local
_X11TransSocketOpenCOTSClient: Unable to open socket for local
_X11TransOpen: transport open failed for local/simon:0
I am running KDE 1.0 as my WM. SuSE 5.2 distribution. Would someone tell me how to get it running? It seems that I have this problem with 2.1.132 version as well. Thank you very much for your attention.
Sounds like a different error I have.
I managed to get a X (mouse pointer), and the
wall paper never came out after that.
When I switch to my Terminal mode, i see all
those error I mentioned above.
I just installed it on my home workstation, which is a shitty old p200 [overdrive... which means it performs like a p150, maybe].
...Might I say, it's operating a *shitload* faster. Very amazing. X is about 100 times more responsive.
The version that I was previously running was 2.0.35.
I presume you tried that with a clean untar.
Otherwise, a lot of stuff will be precompiled
There is a small issue with Cyrix chips regarding the timestamp register. Alan Cox has it patched in his kernels, but that patch hasn't been integrated by Linus yet. It's a very small fix, and it'll be in the final 2.2.0 release.
Art Cancro
UNCENSORED! BBS
While I am not having this problem, I appear to be having a problem with sound. I can play .wav and .mp3 with no problem, but whenever I try to play any real audio I hear about 1 sec, then get error #1 (Unknown error). When I close the player netscape gives me a dialog "subprocess diagnostics" with errors like "****audio: write error: 287 bytes errno: 11". Real Audio did work for me with 2.0.36. I am running an AWE64.
Compile it straight in or make it a module and load unix.o.
People tried to help you, and you thank them by telling them to fuck off? What's the matter with you? Are the other kids being mean to you in school?
All the devel kernels seemed to do that while ftping a file. Got really annoying seeing a pile of checksum spew across your irc console.
hmmmm. i'm away from home and, thus, away from the information stored in my real brain, but i'll try my best with the stuff i keep in my head.
all the virge cards are, IIRC, VESA1.2 . fbcon needs VESA2.0 implemented to work, such as is found in the matrox millenium, the newer ATI cards, etc.
HTH.
bdb151 at psu dot edu
I'm afraid that "NDA" and Open Source / Free software are incompatible.
:-)
Open source by definition gives away all the things that an NDA forbids you from telling others.
If Open Source is a matter of principle, then you can't sign an NDA for your Open Source products.
If it's not, write&release a binary-only product or use MacOS/BeOS/LooseOS/WhateverOtherOS (except the various *BSD's).
Happy Hacking, or as it might be, Happy Gaming, and best wishes for the new year
--Tim
If all has not helped so far, there are two options:
:-)
* Upgrade net-tools to the latest version (or well, at least a 1.4 version instead of 1.3). The layout of some info has changed and the older net-tools report tons of bogus errors/collisions.
* There are also many legitimate problems. From my lukring on the linux kernel mailinglist I gathered that these problems can crop up if VJ compression is turned on in your PPP connection: Turn it off. (The error btw is probably caused by buggy compression at your ISP, not nessecarily by Linux.)
I hope it fixes your problems!
Happy Hacking & Compiling
--Tim
For upgrading libc (libc5.4.46), simply /lib/libc.so.5, instead you
follow the release file. You _don't need_
to delete the
have to run
/sbin/ldconfig
to make sure that the new ld.so cache is
refreshed.
i've been running p4 for well over a day -- anyone know what's holding up p5?
(this is a joke, if you couldn't figure it out on your own)
Where can i find "sh-utils-1.16.tar.gz"? All I found was "sh-utils-1.12+security.tar.gz"
Please reply to me at jbgoode1@NOSPAM.usa.net (remove "NOSPAM." of course). Thanks!!
What? Xfree86 signing a NDA?? They therefore would have to stop releasing the source!! Don't be crazy, it's GNU!
Dont the virge cards have a 1.2 and not a 2.0 vesa. If you go to the s3 home page, there is a TSR you can download to give you vesa 2.0.
If you load the TSR and then use loadlin. You can get the framebuffer working.
6x86L works fine with me
As far as I can tell its right before the kernel prints "Sound initialization started" durring the bootup sequence (for me, right after it says "lp0: using parport0 (polling)."). Once it gets to this position it wont continute. Only a cold boot (reset button) lets me get by this part. I've had this problem since I've started using devel kernel (starting with 2.1.130). Anyone know whats going on?
Anyway, this goes back to 2.1.9x, when something in
Though, if anyone out there is running pppd 2.3.5 and wants to try the new kernel, upgrade first... (actually, RTFM and do ALL the required upgrades)
---
This looks like Alan Cox writing (the message to which I'm replying), but it's not identified as such - better restore your Slashdot cookie, Alan..
Looks like the FAT32 fix is in there, along with what looks like a major NTFS change, some changes for parallel port zip drives and zip plus. Some swap code changes, and a new ksymoops seem to be the highlights from my brief look at it...
Posted by BEEPO!!:
Like a flicking candle in a breeze, BEEPO's joke makes a dance like gesture at slapdash, before being darkened by the moderators moistened thumb and forefinger:
Posted by Pepi:
Report the problem to the linux-kernel@vger.rutgers.edu. It's important that
we get a chance to fix these things before 2.2 gets out the door. Either that, or mail the maintainer. Does it oops for you? Enabled debugging in scsi driver/cdrom.c?
Contact me privately, if need any help.
Posted by Moritz Moeller - Herrmann:
I just noticed that xfstt hangs with linux 2.2.0pre3
Is it just me? Or anyone else, too?
I would assume the FAT32 was fixed.. I'm compiling now. LinuxHQ, (since edge.linuxhq.com is lagging) has a nice patch browser so you can at least see where things were changed.
---
Don Rude - AKA - RudeDude
RudeDude
Perl/Linux/PHP hacker
Or you can get linux-nfs.0.4.22.i386.rpm and use the kernel nfs daemon. I think linuxhq had the best tutorial on how to do this.~
^~~^~^^~~^~^~^~^^~^^~^~^~~^^^~^^~~^~~~^~~^
ABORTED effort:
Close all that you have.
Or you can get linux-nfs.0.4.22.i386.rpm and use the kernel nfs daemon. I think linuxhq had the best tutorial on how to do this.
And don't forget to get a samba with smbmount compiled in it. I had to make my own.
^~~^~^^~~^~^~^~^^~^^~^~^~~^^^~^^~~^~~~^~~^~
ABORTED effort:
Close all that you have.
...he is, only it is sick to think about it.
ABORTED effort:
^~~^~^^~~^~^~^~^^~^^~^~^~~^^^~^^~~^~~~^~~^~
Close all that you have.
Serious recommendation guys:
Get your huge tarball now and get the patches one at a time, even if you don't recompile in between. The netload of a few hundred thousand people trying to get a 12-meg tarball all at the same time is going to be like nothing the internet has ever seen before. Get the major download over with now, get the patches as they come out, and then when the final patch-2.2.0.gz comes, apply it and recompile. This is going to be huger than the IE 4.0 release, mark my words. Let's show we know how to be better prepared than MS was.
On a side note, has anyone else noticed incredibly huge numbers of TCP checksum errors when trying to FTP something? I'm talking to Alan about this, but I'm sure we could use a few more case studies if anyone else has noticed the same problem.
linux/arch/i386/kernel/smp.c line 1055:
if(!cpu_hz) cpu_hz = 150000000;
Well, I have this controller, but I have to admit I haven't used my CD in a while. ;-) I'll test it out for you.
sigs are a waste of space
Well I have hardly any. :-) I've had no problems with FAT32 nor UP/SMP configs... mind you pre3 had a blank /proc/cpuinfo on SMP.
/dev/sndstat. modprobing with the kernel GUS Max drivers works though.
:-)
:-)
The only gripe I have is that kernel support for Esoniq 1370 is busted. I can use the commercial OSS drivers just fine except that my card sounds like someone's drowning for the first few seconds, then resumes the watery behaviour on heavy disk activity (no IRQs are shared). 2.1.130+ did this too. Using the kernel 1370 drivers loads, but gives a blank
Any ideas there?
I figure when 2.2.0 is officially released I'll try the switch to glibc too... Perhaps I am a bit masochistic.
oh yeah -- remove the spambusters from the email if you could help me out in any way.
Pardon my ignorance, but....
I didn't think Cyrix' supported SMP. Is smp.c always compiled in?
Compile
read slashdot
Compile
read slashdot
Compile
read slashdot
damm
Linux is only free if your time has no value. Windows is only free if you threaten to use Linux.
Well, we'll see how much got fixed. I've read through the patch, but not being a coder menas I'll wait till one of the patch summaries goes up to see what I've missed. A lot of it looks like the Cyrix and TCS stuff, along with just simple doc fixes.
does pre4 not work with cyrix 6x86? pre3 worked fine with it, but not pre4. i did everything the same, but, i got a divide by zero and synching error on boot. im going to try again, but i dont think it will work. any other cyrix users out there having trouble? email me!
--
ok, after putting my cyrix as 586 instead of a 486/Cx86, it worked. everything seems to be running smoothly now. if any of you fellow cyrix users out there have any questions, mail me.
--
Anyone else have problems with the kernel dying when using SCSI cdroms and an 7880? Ive been running 2.1.x since .99 and this is the first time ive had a problem..
But pre 1-4 have been crashing hard on me..
Oops... 'linux/include/asm-alpha/ipc.h' was not
provided with this patch, even though 'linux/ipc/util.c' now needs it.
/* MAGIC THEATRE
ENTRANCE NOT FOR EVERYBODY
MADMEN ONLY */
Just wondering. I'd test it my self, except my 5.1 gig drive died Friday night. Life kind of sucks right now w/o it & linux.
As of this message, the only place I was able to find it was via http directly to www.kernel.org. Go to www.linuxhq.com and hit the http link. The mirrors (ftp.us.kernel.org and ftp.tux.org are the only ones I tried) don't have it yet and the main ftp site is too overcrowded. Http doesn't deny anonymous users ;-)
--
Aaron Gaudio
"The fool finds ignorance all around him.
"Every man is a mob, a chain gang of idiots." - Jonathan Nolan, Memento Mori
Again, I have a Cyrix P150 (first generation) and have compiled and booted on 2.2.0 pre4 fine (I applied the ac1 patch to get ntfs to compile). I have disabled SMP (and also changed HZ to 1024 in param.h).
--
Aaron Gaudio
"The fool finds ignorance all around him.
"Every man is a mob, a chain gang of idiots." - Jonathan Nolan, Memento Mori
Have you upgraded everything it said in the Changes file?
Also, are you upgrading to glibc from a libc5 system? That's a tricky endevour as I tried that once and got a lot of core dumps from glibc programs as my reward.
--
Aaron Gaudio
"The fool finds ignorance all around him.
"Every man is a mob, a chain gang of idiots." - Jonathan Nolan, Memento Mori
I finally get the new kernel up and running, and they go and release a new one...at least it sounds like linus got the user base he was requesting...
I got Pre1 when it first came out...
:-)
I haven't had any real trouble with it, Is it worth me getting Pre4, patching and compiling it?
or will Pre5 be out tomorrow?
Compiled every version since pre1, ie pre2 pre3 and now almost pre4
I downloaded pre-2 and patched to pre-3. Patches are tiny downloads; people should use them!
Try finger @linux.kernel.org...
And I only compile -pre3 like 20 mins ago... haven't even rebooted yet! d'oh!
The README changes still leave things like
gzip -cd linux-2.1.XX.tar.gz | tar xfv -
in there though.
Jonathan
> gcc -D__KERNEL__ -I/usr/src/2.2.0pre4/include -Wall
> -Wstrict-prototypes -O2 -fomit-frame-pointer
> -pipe -fno-strength-reduce -m486 -malign-loops=2
> -malign-jumps=2 -malign-functions=2 -DCPU=686
> -DNTFS_IN_LINUX_KERNEL -DNTFS_VERSION=\"990102\"
> -c -o inode.o inode.c
> inode.c: In function `ntfs_decompress_run':
> inode.c:421: `ntfs_s16' undeclared (first use this function)
> inode.c:421: (Each undeclared identifier is reported only once
> inode.c:421: for each function it appears in.)
> inode.c:423: `ntfs_s32' undeclared (first use this function)
> make[3]: *** [inode.o] Error 1
> make[3]: Leaving directory `/usr/src/2.2.0pre4/fs/ntfs'
> make[2]: *** [first_rule] Error 2
> make[2]: Leaving directory `/usr/src/2.2.0pre4/fs/ntfs'
> make[1]: *** [_subdir_ntfs] Error 2
> make[1]: Leaving directory `/usr/src/2.2.0pre4/fs'
> make: *** [_dir_fs] Error 2
Jonathan
My system is sending out packets all the time, just a little bit like when you're pinging someone (but without the reply's) and every once in a while a large peak (3-4k/s).
I can't figure out what it's doing and I can't find it in any of the logfiles.
Does anybody know what this is?
And it doesn't send packets when I stick a firewall between me and the outside world, so it looks like it's reacting to some incoming stuff.
FWIW, My machine's an AMD K6-2 300 with a NE-2000 clone card. It has done this in 2.2pre as well as 2.1.*, but not in my regular 2.0.35 kernel. And I noticed this because I have the wmnet dockapplet running, which shows incoming and outgoing packets in a graph.
Since noone has actually posted it here, and several people have asked for it... Here is the announcement off of the linux-kernel mailing list.
If you are interested in the mailing list check out http://www.tux.org/lkml/
Ok, you know the drill by now. This fixes:
- yes, people told me about the new and improved ksymoops. Much better,
no need for C++, and this one actually seems to compile and work
reliably.
- ntfs fixes
- the vfat thing _really_ works now
- NFS fix for deleting files while writebacks active.
- ppa/imm driver updated
- minor mm balancing patches
- Alan took the gauntlet and cleaned up some CONFIG_PROC_FS stuff.
More on Monday,
Linus
Would you do it for some scoobie crack?
pre-4 isn't on the US mirror yet and ftp.kernel.org is full...
Wonder how long it takes to replicate (if it's out)...
Justin
Mu. P.S. The address you see is real. =)
I got into ftp.kernel.org!!! Woohoo!!
Then you're part of the fucking problem, asshole.
What packages do I have to upgrade on RedHat 5.0 system inorder to get the 2.2pre* kernels to work.
anyone else having problems with DHCP and RedHat 5.2?
Vivez sans temps mort!
This went across linux-kernel a while back. Can't wait to see it actually implemented. When it was discussed, plans were made to include it in 2.3 kernels. Sounds like fun to me!
How can you release the source for a card without giving away the information about the specs? It is somewhat trivial to figure out how a card works by reading the source.
Oh, good. I was looking into this problem myself, :-)
but got tired and went home for a long weekend instead. (I'm building it on my desktop workstation at work, which is faster than my home machine.) I got as far as determining that the b_size (blocksize) parameter was set to 0, causing the error. The piece of code looked entirely new; I was tempted to comment it out.
Does anyone know offhand what the fix was? Was it just an uninitialized data structure?
Deven
"Simple things should be simple, and complex things should be possible." - Alan Kay
Has anyone been succesful with ip masquerading using any of the 2.2.0 kernels?
I've pre1,pre2, and pre3, and non have worked for me. I'm going from the instructions in Ip-masquerading-howto, which worked perfect in 2.0.35.
Is there something special which has to be done in these kernels?
I like cheese, cheese is good, cheeeeeeese is gooood.
i'm at ftp://ftp.us.kernel.org/pub/linux/kernel/v2.1/
it says latest is pre1 and i don't see pre4 anywhere.
I like cheese, cheese is good, cheeeeeeese is gooood.
Use these 3 lines in a startup script .. changing the appropriate things
/proc/sys/net/ipv4/ip_forward
...
.. this is an example thats not working
..
#ipchains -P forward DENY
#ipchains -A forward -j MASQ -s 192.168.0.0/24 -d 0.0.0.0/0
#echo 1 >
without the #'s of course
If anyone here has managed to get ipmasqadm working I need some help. I cant get it to forward any ports and I did it exactly as the man file reads
ipmasqadm portfw -a -P tcp -L 192.168.0.1 6112 -R 192.168.0.2 6112
talking windows users is where I draw the line
Tis better to be silent and thought a fool, than to open your mouth and remove all doubt --Abraham Lincoln
Cyrix 6x86MX 200 at 2x75 booted fine with pre4
..
I just swapped out an ATI card for a S3 Virge and my frame buffer device just stopped working. Anyone have any idea on that?
talking windows users is where I draw the line
Tis better to be silent and thought a fool, than to open your mouth and remove all doubt --Abraham Lincoln
ok, I found a little tidbit in the ipchains howto that helped me enough to figure it out .. im gonna show you my working network file and you can see how it works between .36 and 2.2pre
/proc/sys/net/ipv4/ip_forward
..
###### 2.0.X ######################
#ipfwadm -F -p deny
#ipfwadm -F -a m -S 192.168.0.0/24 -D 0.0.0.0/0
#ipautofw -A -r tcp 113 113 -h 192.168.0.2
#ipautofw -A -r tcp 2000 2010 -h 192.168.0.2
#ipautofw -A -r tcp 1034 1035 -h 192.168.0.2
#ipautofw -A -r udp 1034 1035 -h 192.168.0.2
#ipautofw -A -r tcp 6112 6112 -h 192.168.0.2
#ipautofw -A -r udp 6112 6112 -h 192.168.0.2
#ipautofw -A -r udp 7070 7070 -h 192.168.0.2
#ipautofw -A -r udp 7071 7071 -h 192.168.0.2
#ipautofw -A -r tcp 26029 26029 -h 192.168.0.2
#ipautofw -A -r udp 26029 26029 -h 192.168.0.2
###### 2.2.X ######################
ipchains -P forward DENY
ipchains -A forward -j MASQ -s 192.168.0.0/24 -d 0.0.0.0/0
echo 1 >
ipmasqadm portfw -f
ipmasqadm autofw -A -r tcp 113 113 -h 192.168.0.2
ipmasqadm autofw -A -r tcp 2000 2010 -h 192.168.0.2
ipmasqadm autofw -A -r tcp 1034 1035 -h 192.168.0.2
ipmasqadm autofw -A -r udp 1034 1035 -h 192.168.0.2
ipmasqadm autofw -A -r tcp 6112 6112 -h 192.168.0.2
ipmasqadm autofw -A -r udp 6112 6112 -h 192.168.0.2
ipmasqadm autofw -A -r udp 7070 7070 -h 192.168.0.2
ipmasqadm autofw -A -r udp 7071 7071 -h 192.168.0.2
ipmasqadm autofw -A -r tcp 26029 26029 -h 192.168.0.2
ipmasqadm autofw -A -r udp 26029 26029 -h 192.168.0.2
talking windows users is where I draw the line
Tis better to be silent and thought a fool, than to open your mouth and remove all doubt --Abraham Lincoln
Whenever I try to place the line
/etc/lilo.conf
..
video=vesa
inside of lilo.conf and run lilo I get
Syntax error near line 8 in file
is there anywhere special I need to put it or anything or does it not even belong in lilo.conf
I really need this to get ywrap working for scrollback
talking windows users is where I draw the line
Tis better to be silent and thought a fool, than to open your mouth and remove all doubt --Abraham Lincoln
Peace to my homies If you don't like it you can suck it.
www.espn.com"Are you satisfied with fucking?" - Dave Matthews from "Halloween"
buy a monitor which can do 3000x3000 (suggested by Alan Cox) ;-)
I was just curious how I would go about submitting a document I wrote on how to get the kernel drivers working with an ESS1869 soundcard to be included with the kernel...??