ftape and kernel 2.4 problem
Hi!
As the linux-ftape mailing list is gone I'm asking you guys.
Can someone tell me how to adapt the ftape driver that I can use it under kernel 2.4.x (x >= 5)? I'm not that into kernel hacking that I know what changed from 2.2.x to 2.4.x. Below is the output of make.
BTW why wasn't the newer ftape driver ported to 2.4 but the stone age ftape driver is still in 2.4?
PS: Please CC me because I'm not on linux-kernel.
friedl:/usr/src/ftape-4.04a # make
for i in ftape ; do make -C $i all ; done
make[1]: Entering directory `/mount/2/src/ftape-4.04a/ftape'
Quoted 1 line
../include/linux/modftversions.h
for i in lowlevel internal parport zftape compressor; \
do \
make -C $i NODEP=true versions; \
done
make[2]: Entering directory `/mount/2/src/ftape-4.04a/ftape/lowlevel'
rm -f ../../include/linux/modules/ftape_syms.ver.tmp; gcc -D__KERNEL__ -I. -I../../include -I/usr/src/linux/include -E -D__GENKSYMS__ ftape_syms.c | /sbin/genksyms ../../include/linux/modules 2> /dev/null; rm -f ../../include/linux/modules/ftape_syms.ver.tmp make[2]: Leaving directory `/mount/2/src/ftape-4.04a/ftape/lowlevel'
make[2]: Entering directory `/mount/2/src/ftape-4.04a/ftape/internal'
make[2]: Nothing to be done for `versions'.
make[2]: Leaving directory `/mount/2/src/ftape-4.04a/ftape/internal'
make[2]: Entering directory `/mount/2/src/ftape-4.04a/ftape/parport'
make[2]: Nothing to be done for `versions'.
make[2]: Leaving directory `/mount/2/src/ftape-4.04a/ftape/parport'
make[2]: Entering directory `/mount/2/src/ftape-4.04a/ftape/zftape'
rm -f ../../include/linux/modules/zftape_syms.ver.tmp; gcc -D__KERNEL__ -I. -I../../include -I/usr/src/linux/include -E -D__GENKSYMS__ zftape_syms.c | /sbin/genksyms ../../include/linux/modules 2> /dev/null; rm -f ../../include/linux/modules/zftape_syms.ver.tmp updating ../../include/linux/modftversions.h
make[2]: Leaving directory `/mount/2/src/ftape-4.04a/ftape/zftape'
make[2]: Entering directory `/mount/2/src/ftape-4.04a/ftape/compressor'
make[2]: Nothing to be done for `versions'.
make[2]: Leaving directory `/mount/2/src/ftape-4.04a/ftape/compressor'
set -e; for i in lowlevel internal parport zftape compressor; do make -C $i modules; done
make[2]: Entering directory `/mount/2/src/ftape-4.04a/ftape/lowlevel'
make[2]: Leaving directory `/mount/2/src/ftape-4.04a/ftape/lowlevel'
make[2]: Entering directory `/mount/2/src/ftape-4.04a/ftape/lowlevel'
gcc -Wall -Wstrict-prototypes -O2 -fomit-frame-pointer -fno-strength-reduce -D__KERNEL__ -I. -I../../include -I/usr/src/linux/include -DMODULE -DMODVERSIONS -include ../../include/linux/modftversions.h -DEXPORT_SYMTAB -c ftape_syms.c In file included from /usr/src/linux/include/linux/fs.h:12,
from /usr/src/linux/include/linux/capability.h:17,
from /usr/src/linux/include/linux/binfmts.h:5,
from /usr/src/linux/include/linux/sched.h:9,
from ../../include/linux/ftape.h:35,
from ftape_syms.c:32:
/usr/src/linux/include/linux/wait.h: In function `__add_wait_queue_tail':
/usr/src/linux/include/linux/wait.h:220: warning: implicit declaration of function `list_add_tail'
In file included from ftape-tracing.h:35,
from ftape_syms.c:33:../lowlevel/ftape-init.h: In function `ft_sigtest':../lowlevel/ftape-init.h:70: structure has no member named `signal' ../lowlevel/ftape-init.h:71: warning: control reaches end of non-void function make[2]: *** [ftape_syms.o] Error 1
make[2]: Leaving directory `/mount/2/src/ftape-4.04a/ftape/lowlevel' make[1]: *** [modules] Error 2
make[1]: Leaving directory `/mount/2/src/ftape-4.04a/ftape'
make: *** [all] Error 2
friedl:/usr/src/ftape-4.04a #
Signature · 2 lines
MfG / Regards Friedrich Lobenstock
Here's a patch I wrote to allow ftape to compile against 2.4.something. It still works with 2.4.5. I'm not sure if it works entirely (it seems to), but it compiles and seems to work. Enjoy!
Quoted 12 lines · Friedrich Lobenstock
Hi!
As the linux-ftape mailing list is gone I'm asking you guys.
Can someone tell me how to adapt the ftape driver that I can use it under kernel 2.4.x (x >= 5)? I'm not that into kernel hacking that I know what changed from 2.2.x to 2.4.x. Below is the output of make.
BTW why wasn't the newer ftape driver ported to 2.4 but the stone age ftape driver is still in 2.4?
PS: Please CC me because I'm not on linux-kernel.
Signature · 3 lines
-Steven In a time of universal deceit, telling the truth is a revolutionary act. -- George Orwell
ftape/lowlevel/ftape-init.h +1 -1
@@ -67,7 +67,7 @@ }extern inline int ft_sigtest (unsigned long mask) { -return (current->signal.sig[0 ] & mask); +return (current->sigpending & mask); }extern inline int ft_killed (void ) {
include/linux/ftape.h +1 -1
@@ -28,7 +28,7 @@* for the QIC-40/80/3010/3020 floppy-tape driver for Linux. * */ - +#define __initlocaldata __initdata#define FTAPE_VERSION "ftape v4.04a 07/25/2000"#ifdef __KERNEL__
Quoted 3 lines · Steven Walter
Here's a patch I wrote to allow ftape to compile against 2.4.something. It still works with 2.4.5. I'm not sure if it works entirely (it seems to), but it compiles and seems to work. Enjoy!
I really do enjoy it despite compiling gives me a lot of warnings :)
Alan, how could we get the current in kernel ftape driver updated? What needs to be done?
PS: Please CC me because I'm not on linux-kernel.
Signature · 2 lines
MfG / Regards Friedrich Lobenstock
Quoted 2 lines
Alan, how could we get the current in kernel ftape driver updated? What needs to be done?
Someone needs to do it. To be honest the last ftape contribution I can find is back in 2.3.30 or so and was some fixes to make it compile which were done I think by someone without the hardware.
If you are actually using ftape you are I suspect the most qualified volunteer for doing something about it I've seen for two years 8)
Quoted 3 lines · Alan Cox
If you are actually using ftape you are I suspect the most qualified volunteer for doing something about it I've seen for two years 8)
Hi Alan!
Sure, but what I wanted to know are there some documents that describe the changes between 2.2 and 2.4? And how would you accept patches?
For sure I could only make it work - maybe with help by Steven Walter - and then provide a patch for the whole thing. I know you and Linus only want small patches. I think this will only relate to the ftape drive, so I ask would the aformentioned big patch do? Or how'd we split it up?
Maybe you could take a quick look at the latest state of the driver (dont forget to count the small patch of Steven Walter) and tell me what you think about the whole thing.
http://www.instmath.rwth-aachen.de/~heine/ftape/getting.html
Signature · 2 lines
MfG / Regards Friedrich Lobenstock