Discussion:
Kernel 2.5.33 successfully compiled
Oliver Pitzeier
2002-09-03 10:59:46 UTC
Permalink
Hi volks!

After beeing on holiday and having a lot of work the last few weeks I had time
to try compile 2.5.33 on my AS1000 (Noritake system).

It compiled successfull. I have not yet tried to reboot the machine because it
is used currently...

I will try later and tell you if it works or not...

Thanks a lot to all those alpha developers who made this possible. I guess
Ivan was one of the most involved people. :o)

Best regards!
--
Oliver Pitzeier
UNIX Administrator
-
Linux 2.4.19 i686 Load: 1.04 0.52 0.20
Oliver Pitzeier
2002-09-03 11:57:48 UTC
Permalink
On Tuesday 03 September 2002 12:59, Oliver Pitzeier wrote:
[ ... ]
Post by Oliver Pitzeier
It compiled successfull. I have not yet tried to reboot the machine because
it is used currently...
OK. I was in an error! I didn't saw the error because of a "make -j2"...

I'll try to fix it now...
--
Oliver Pitzeier
UNIX Administrator
-
Linux 2.4.19 i686 Load: 0.08 0.02 0.03
Leslie Donaldson
2002-09-03 15:08:46 UTC
Permalink
Hello,

I have had 3 errors so far.

1. add the following line to
/usr/src/linux-2.5.X/linux-2.5.33/sound/core/device.c

#include <asm-generic/errno-base.h>

2. The following file need copied (a quick hack)

cd include/asm
cp ../asm-i386/kmap_types.h .


3. The following error I haven't tried to fix yet.

process.c: In function `alpha_clone':
process.c:268: too few arguments to function `do_fork'
process.c: In function `alpha_vfork':
process.c:277: too few arguments to function `do_fork'
make[1]: *** [process.o] Error 1
make[1]: Leaving directory
`/usr/src/linux-2.5.X/linux-2.5.33/arch/alpha/kernel'
make: *** [arch/alpha/kernel] Error 2

Just my 2 cents worth.

Leslie D.
Post by Oliver Pitzeier
[ ... ]
Post by Oliver Pitzeier
It compiled successfull. I have not yet tried to reboot the machine because
it is used currently...
OK. I was in an error! I didn't saw the error because of a "make -j2"...
I'll try to fix it now...
--
/----------------------------\ Current Contractor: None
| Leslie F. Donaldson | Current Customer : None
| Computer Contractor | Skills: Unix/OS9/VMS/Linux/SUN-OS/C/C++/assembly
| Have Computer will travel. | WWW : http://www.cs.rose-hulman.edu/~donaldlf
\----------------------------/ Email: mail://***@cs.rose-hulman.edu
Goth Code V1.1: GoCS$$ TYg(T6,T9) B11Bk!^1 C6b-- P0(1,7) M+ a24 n--- b++:+
H6'11" g m---- w+ r+++ D--~!% h+ s10 k+++ R-- Ssw LusCA++
Padraig Brady
2002-09-03 15:26:03 UTC
Permalink
Just a me too:
compile breaks in OSS & inet6
Oliver Pitzeier
2002-09-03 15:53:09 UTC
Permalink
On Tuesday 03 September 2002 17:08, Leslie Donaldson wrote:
[ ... ]
Post by Leslie Donaldson
I have had 3 errors so far.
1. add the following line to
/usr/src/linux-2.5.X/linux-2.5.33/sound/core/device.c
I don't have sound on this machine, so I don't get any errors in this module.

[ ... ]
Post by Leslie Donaldson
2. The following file need copied (a quick hack)
cd include/asm
cp ../asm-i386/kmap_types.h .
I used the same quick-hack, but I was wondering why kmap_types.h is missing...
Any ideas? Anybody?
Post by Leslie Donaldson
3. The following error I haven't tried to fix yet.
process.c:268: too few arguments to function `do_fork'
process.c:277: too few arguments to function `do_fork'
make[1]: *** [process.o] Error 1
make[1]: Leaving directory
`/usr/src/linux-2.5.X/linux-2.5.33/arch/alpha/kernel'
make: *** [arch/alpha/kernel] Error 2
The same problems happens here. I'm currently searching for the changelog for
do_fork, so I may find out what arguments I should add to make this working
again. At the i386-tree there is sometimes "NULL" as the last argument added,
but I don't believe that it will be good to simply add "NULL" in every
function call. :o)

Also the quick hack with kmap_types.h doesn't seem very nice to me, but I just
tried to compile it fine for the first...

Any suggestions from anyone are very welcome! :o)
--
Oliver Pitzeier
UNIX Administrator
-
Linux 2.4.19 i686 Load: 0.50 0.39 0.37
Thunder from the hill
2002-09-03 16:14:01 UTC
Permalink
Hi,

I must admit that Alpha is a few more out of sync than it deserves.
Post by Oliver Pitzeier
Post by Leslie Donaldson
2. The following file need copied (a quick hack)
cd include/asm
cp ../asm-i386/kmap_types.h .
That's really a bad hack, if i386 kmap types match up with your
expectations, you must be a lucky guy. i386 and alpha are still 100%
different machines.
Post by Oliver Pitzeier
I used the same quick-hack, but I was wondering why kmap_types.h is missing...
Any ideas? Anybody?
It wasn't introduced, I guess.
Post by Oliver Pitzeier
Post by Leslie Donaldson
process.c:268: too few arguments to function `do_fork'
process.c:277: too few arguments to function `do_fork'
make[1]: *** [process.o] Error 1
make[1]: Leaving directory
Yes, the syntax changed.
Post by Oliver Pitzeier
The same problems happens here. I'm currently searching for the changelog for
do_fork, so I may find out what arguments I should add to make this working
again. At the i386-tree there is sometimes "NULL" as the last argument added,
but I don't believe that it will be good to simply add "NULL" in every
function call. :o)
The NULL only needs to be non-NULL if we clone with CLONE_CLEARTID.

Thunder
--
--./../...-/. -.--/---/..-/.-./..././.-../..-. .---/..-/.../- .-
--/../-./..-/-/./--..-- ../.----./.-../.-.. --./../...-/. -.--/---/..-
.- -/---/--/---/.-./.-./---/.--/.-.-.-
--./.-/-.../.-./.././.-../.-.-.-
Oliver Pitzeier
2002-09-03 16:18:16 UTC
Permalink
On Tuesday 03 September 2002 18:14, Thunder from the hill wrote:
[ ... ]
Post by Thunder from the hill
Post by Oliver Pitzeier
I used the same quick-hack, but I was wondering why kmap_types.h is
missing... Any ideas? Anybody?
It wasn't introduced, I guess.
No. It seems simply missing. In 2.5.32 it is there. On the right place. :o))))
Post by Thunder from the hill
Post by Oliver Pitzeier
Post by Leslie Donaldson
process.c:268: too few arguments to function `do_fork'
process.c:277: too few arguments to function `do_fork'
make[1]: *** [process.o] Error 1
make[1]: Leaving directory
Yes, the syntax changed.
I saw. But how should it be called now from within process.c?
Post by Thunder from the hill
Post by Oliver Pitzeier
The same problems happens here. I'm currently searching for the changelog
for do_fork, so I may find out what arguments I should add to make this
working again. At the i386-tree there is sometimes "NULL" as the last
argument added, but I don't believe that it will be good to simply add
"NULL" in every function call. :o)
The NULL only needs to be non-NULL if we clone with CLONE_CLEARTID.
I c...

That what I expected.
--
Oliver Pitzeier
UNIX Administrator
-
Linux 2.4.19 i686 Load: 0.59 0.37 0.32
Thunder from the hill
2002-09-03 16:34:17 UTC
Permalink
Hi,
Post by Oliver Pitzeier
No. It seems simply missing. In 2.5.32 it is there. On the right place. :o))))
Then copy it from 2.5.32 rather than from i386.
Post by Oliver Pitzeier
Post by Thunder from the hill
Post by Leslie Donaldson
process.c:268: too few arguments to function `do_fork'
process.c:277: too few arguments to function `do_fork'
make[1]: *** [process.o] Error 1
make[1]: Leaving directory
Yes, the syntax changed.
I saw. But how should it be called now from within process.c?
What about += NULL? I'm not sure, but it may work.

The only thing that makes me sad is the other two zeroes. They simply seem
down and out.

Thunder
--
--./../...-/. -.--/---/..-/.-./..././.-../..-. .---/..-/.../- .-
--/../-./..-/-/./--..-- ../.----./.-../.-.. --./../...-/. -.--/---/..-
.- -/---/--/---/.-./.-./---/.--/.-.-.-
--./.-/-.../.-./.././.-../.-.-.-
Oliver Pitzeier
2002-09-03 16:27:16 UTC
Permalink
[ ... ]
Post by Oliver Pitzeier
No. It seems simply missing. In 2.5.32 it is there. On the right place. :o))))
Post by Thunder from the hill
Post by Leslie Donaldson
process.c:268: too few arguments to function `do_fork'
process.c:277: too few arguments to function `do_fork'
make[1]: *** [process.o] Error 1
make[1]: Leaving directory
Yes, the syntax changed.
OK, after calling do_fork with "NULL" as extra argument. It compiles.

(No, I do not want to make the kernel run fine [currently], but I want to
trace down the count of bugs to a countable value. :o) )

[ ... ]

OK. The next problem seems harded to me (Maybe Ivan can help a bit!?):

gcc -Wp,-MD,./.irq.o.d -D__KERNEL__ -I/usr/src/linux-2.5.33/include -Wall
-Wstrict-prototypes -Wno-trigraphs -O2 -fomit-frame-pointer
-fno-strict-aliasing -fno-common -pipe -mno-fp-regs -ffixed-8 -msmall-data
-mcpu=ev5 -Wa,-mev6 -nostdinc -iwithprefix include -DKBUILD_BASENAME=irq
-c -o irq.o irq.c
In file included from irq.c:15:
/usr/src/linux-2.5.33/include/linux/ptrace.h:28: warning: `struct task_struct'
declared inside parameter list
/usr/src/linux-2.5.33/include/linux/ptrace.h:28: warning: its scope is only
this definition or declaration, which is probably not what you want
/usr/src/linux-2.5.33/include/linux/ptrace.h:29: warning: `struct task_struct'
declared inside parameter list
/usr/src/linux-2.5.33/include/linux/ptrace.h:30: warning: `struct task_struct'
declared inside parameter list
/usr/src/linux-2.5.33/include/linux/ptrace.h:31: warning: `struct task_struct'
declared inside parameter list
/usr/src/linux-2.5.33/include/linux/ptrace.h:32: warning: `struct task_struct'
declared inside parameter list
/usr/src/linux-2.5.33/include/linux/ptrace.h:33: warning: `struct task_struct'
declared inside parameter list
/usr/src/linux-2.5.33/include/linux/ptrace.h:35: warning: `struct task_struct'
declared inside parameter list
/usr/src/linux-2.5.33/include/linux/ptrace.h:36: warning: `struct task_struct'
declared inside parameter list
/usr/src/linux-2.5.33/include/linux/ptrace.h:39: warning: `struct task_struct'
declared inside parameter list
/usr/src/linux-2.5.33/include/linux/ptrace.h: In function `ptrace_link':
/usr/src/linux-2.5.33/include/linux/ptrace.h:41: dereferencing pointer to
incomplete type
/usr/src/linux-2.5.33/include/linux/ptrace.h:42: warning: passing arg 1 of
`__ptrace_link' from incompatible pointer type
/usr/src/linux-2.5.33/include/linux/ptrace.h:42: warning: passing arg 2 of
`__ptrace_link' from incompatible pointer type
/usr/src/linux-2.5.33/include/linux/ptrace.h: At top level:
/usr/src/linux-2.5.33/include/linux/ptrace.h:44: warning: `struct task_struct'
declared inside parameter list
/usr/src/linux-2.5.33/include/linux/ptrace.h: In function `ptrace_unlink':
/usr/src/linux-2.5.33/include/linux/ptrace.h:46: dereferencing pointer to
incomplete type
/usr/src/linux-2.5.33/include/linux/ptrace.h:47: warning: passing arg 1 of
`__ptrace_unlink' from incompatible pointer type
make[1]: *** [irq.o] Error 1
make[1]: Leaving directory `/usr/src/linux-2.5.33/arch/alpha/kernel'
make: *** [arch/alpha/kernel] Error 2
--
Oliver Pitzeier
UNIX Administrator
-
Linux 2.4.19 i686 Load: 0.32 0.25 0.27
Thunder from the hill
2002-09-03 16:39:22 UTC
Permalink
Hi,
Post by Oliver Pitzeier
/usr/src/linux-2.5.33/include/linux/ptrace.h:28: warning: `struct task_struct'
declared inside parameter list
Looks like we need to include <linux/sched.h> prior to that.

Thunder
--
--./../...-/. -.--/---/..-/.-./..././.-../..-. .---/..-/.../- .-
--/../-./..-/-/./--..-- ../.----./.-../.-.. --./../...-/. -.--/---/..-
.- -/---/--/---/.-./.-./---/.--/.-.-.-
--./.-/-.../.-./.././.-../.-.-.-
Remco Post
2002-09-03 22:22:43 UTC
Permalink
I now do have a successful compile, will try to boot it in a sec.... This is from the linuxppc-2.5 tree, will do the same patches on Linus' tree to see if that compiles as well. Ok, it sort of boots as well:

Kernel command line: console=ttyS0,9600 console=tty0 root=/dev/sda2
time_init: decrementer frequency = 16.666782 MHz
Calibrating delay loop... 266.24 BogoMIPS
Memory: 62280k available (1568k kernel code, 804k data, 100k init, 0k highmem)
Security Scaffold v1.0.0 initialized
Dentry-cache hash table entries: 8192 (order: 4, 65536 bytes)
Inode-cache hash table entries: 4096 (order: 3, 32768 bytes)
Mount-cache hash table entries: 512 (order: 0, 4096 bytes)
POSIX conformance testing by UNIFIX
Linux NET4.0 for Linux 2.4
Based upon Swansea University Computer Society NET3.039
Initializing RT netlink socket
PCI: Probing PCI hardware
Oops: kernel access of bad area, sig: 11
NIP: C000EFA0 LR: C00DF404 SP: C0363B50 REGS: c0363aa0 TRAP: 0301 Not taintedMSR: 00001032 EE: 0 PR: 0 FP: 0 ME: 1 IR/DR: 11
DAR: 00000000, DSISR: 42000000
TASK = c0360040[1] 'swapper' Last syscall: 120
last math 00000000 last altivec 00000000
GPR00: C75C0080 C0363B50 C0360040 C03F94C0 00000000 0000000E 00000001 C0363B78
GPR08: 00000000 00000000 C03F94CC C75C0000 64676500
Call backtrace:
C00DF404 C00E026C C00E036C C00E057C C020F130 C020C5F4 C020C638
C00038E0 C0008A3C
Kernel panic: Attempted to kill init!

Remember this is a prep board....

On Tue, 3 Sep 2002 18:27:16 +0200
Post by Oliver Pitzeier
[ ... ]
Post by Oliver Pitzeier
No. It seems simply missing. In 2.5.32 it is there. On the right place. :o))))
Post by Thunder from the hill
Post by Leslie Donaldson
process.c:268: too few arguments to function `do_fork'
process.c:277: too few arguments to function `do_fork'
make[1]: *** [process.o] Error 1
make[1]: Leaving directory
Yes, the syntax changed.
OK, after calling do_fork with "NULL" as extra argument. It compiles.
(No, I do not want to make the kernel run fine [currently], but I want to
trace down the count of bugs to a countable value. :o) )
[ ... ]
Attached patch worked for _compiling_ on my powermac, it might help for you as well....
Post by Oliver Pitzeier
gcc -Wp,-MD,./.irq.o.d -D__KERNEL__ -I/usr/src/linux-2.5.33/include -Wall
-Wstrict-prototypes -Wno-trigraphs -O2 -fomit-frame-pointer
-fno-strict-aliasing -fno-common -pipe -mno-fp-regs -ffixed-8 -msmall-data
-mcpu=ev5 -Wa,-mev6 -nostdinc -iwithprefix include -DKBUILD_BASENAME=irq
-c -o irq.o irq.c
/usr/src/linux-2.5.33/include/linux/ptrace.h:28: warning: `struct task_struct'
declared inside parameter list
/usr/src/linux-2.5.33/include/linux/ptrace.h:28: warning: its scope is only
this definition or declaration, which is probably not what you want
/usr/src/linux-2.5.33/include/linux/ptrace.h:29: warning: `struct task_struct'
declared inside parameter list
/usr/src/linux-2.5.33/include/linux/ptrace.h:30: warning: `struct task_struct'
declared inside parameter list
/usr/src/linux-2.5.33/include/linux/ptrace.h:31: warning: `struct task_struct'
declared inside parameter list
/usr/src/linux-2.5.33/include/linux/ptrace.h:32: warning: `struct task_struct'
declared inside parameter list
/usr/src/linux-2.5.33/include/linux/ptrace.h:33: warning: `struct task_struct'
declared inside parameter list
/usr/src/linux-2.5.33/include/linux/ptrace.h:35: warning: `struct task_struct'
declared inside parameter list
/usr/src/linux-2.5.33/include/linux/ptrace.h:36: warning: `struct task_struct'
declared inside parameter list
/usr/src/linux-2.5.33/include/linux/ptrace.h:39: warning: `struct task_struct'
declared inside parameter list
/usr/src/linux-2.5.33/include/linux/ptrace.h:41: dereferencing pointer to
incomplete type
/usr/src/linux-2.5.33/include/linux/ptrace.h:42: warning: passing arg 1 of
`__ptrace_link' from incompatible pointer type
/usr/src/linux-2.5.33/include/linux/ptrace.h:42: warning: passing arg 2 of
`__ptrace_link' from incompatible pointer type
/usr/src/linux-2.5.33/include/linux/ptrace.h:44: warning: `struct task_struct'
declared inside parameter list
/usr/src/linux-2.5.33/include/linux/ptrace.h:46: dereferencing pointer to
incomplete type
/usr/src/linux-2.5.33/include/linux/ptrace.h:47: warning: passing arg 1 of
`__ptrace_unlink' from incompatible pointer type
make[1]: *** [irq.o] Error 1
make[1]: Leaving directory `/usr/src/linux-2.5.33/arch/alpha/kernel'
make: *** [arch/alpha/kernel] Error 2
--
Oliver Pitzeier
UNIX Administrator
-
Linux 2.4.19 i686 Load: 0.32 0.25 0.27
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
Thunder from the hill
2002-09-03 22:37:48 UTC
Permalink
Hi,
Post by Remco Post
Attached patch worked for _compiling_ on my powermac, it might help for you as well....
I'd rather not include sched.h into ptrace.h, for that way lies madness.
You get all the crappy headers when you only include one of them. I'm not
saying the change itself is wrong. It's indeed effective. But it's that
cleanup vs. messup thing.

Thunder
--
--./../...-/. -.--/---/..-/.-./..././.-../..-. .---/..-/.../- .-
--/../-./..-/-/./--..-- ../.----./.-../.-.. --./../...-/. -.--/---/..-
.- -/---/--/---/.-./.-./---/.--/.-.-.-
--./.-/-.../.-./.././.-../.-.-.-
Loading...