2.4.9 Kernel Released

Justin writes: "Linus is off to Finland for a week or so and released 2.4.9. " Here is the Changelog for those of you interested. Yeah, it's probably gonna be a little crowded for a bit. Please post mirrors in the comments.

Comments

drivers/net/rrunner.c looks to be borked.

[AgTiger]

Submitted for amusement, a segment of make modules:

make -C net modules
make[2]: Entering directory `/usr/src/linux-2.4.9/drivers/net'
gcc -D__KERNEL__ -I/usr/src/linux-2.4.9/include -Wall -Wstrict-prototypes -Wno-trigraphs -O2 -fomit-frame-pointer -fno-strict-aliasing -fno-common -pipe -mpreferred-stack-boundary=2 -march=i686 -malign-functions=4 -DMODULE -DMODVERSIONS -include /usr/src/linux-2.4.9/include/linux/modversions.h -c -o rrunner.o rrunner.c
rrunner.c:1241: macro `min' used with only 2 args
rrunner.c:1252: macro `min' used with only 2 args
rrunner.c: In function `rr_dump':
rrunner.c:1241: parse error before `__x'
rrunner.c:1241: `__x' undeclared (first use in this function)
rrunner.c:1241: (Each undeclared identifier is reported only once
rrunner.c:1241: for each function it appears in.)
rrunner.c:1241: `__y' undeclared (first use in this function)
rrunner.c:1252: parse error before `__x'
rrunner.c:1221: warning: `len' might be used uninitialized in this function
make[2]: *** [rrunner.o] Error 1
make[2]: Leaving directory `/usr/src/linux-2.4.9/drivers/net'
make[1]: *** [_modsubdir_net] Error 2
make[1]: Leaving directory `/usr/src/linux-2.4.9/drivers'
make: *** [_mod_drivers] Error 2

The problem with frequent kernel releases

[BierGuzzl]

With longer gaps between kernel releases, people can write apps that interact with the kernel and actually have a reasonable expectation that they won't spend the majority of their time trying to update their code to work with the most recent version of the kernel. People who just want bug fix patches should be able to get just that, leaving the added features,etc which will invariably introduce new bugs to be introduced on the next release cycle. The debian distribution system would be perfectly suited for this concept, and would pretty much do away with the need for a "development" branch of the kernel.

Is kerneli still being maintained?

[typedef]

Does anyone know what the deal is with the kerneli patch? Short of an unofficial hacked patch to work with newer kernels, kerneli hasn't been updated since 2.4.3, and there seems to be some serious issues currently with file corruption when using kerneli with 2.4 series kernels. Alternative projects like cryptoapi and loop-aes have sprung up from currently foobar'd kerneli, and while they work great, I can't help but wonder what's going on with the kerneli project that it hasn't been updated in so long.

Re:NTFS filesystem

[Haggis Muncher]

You can fix this by editing "/usr/src/linux/fs/ntfs/unistr.h" and adding the following at line 30:

#include <linux/kernel.h>

and then recompiling. I've not bothered to submit an official patch... there's probably dozens already.

Bug in NTFS compilation

[the way]

I just compiled 2.4.9 with read-only support for NTFS. It turns out that there is a small bug that stops it from compiling. To fix the bug, edit fs/ntfs/unistr.c, and add somewhere near the top (line 24 or 25 is fine):

#include <linux/kernel.h>

I can't wait until XFS is standard...

[wowbagger]

I can't wait until SGI gets XFS merged into the main tree. I'm running XFS on all my systems, and so I have to wait until SGI gets the changes merged back into their port.

XFS (especially when combined with LVM) is great. No fscks, big files, ACLs, and you can grow a mounted file system (great with LVM and hot-swap drives).