Kernel compilation problem

Kernel compilation problem

[Ben Hague]

Hi,
  I'm trying to compile the latest CVS kernel for a Philips Nino 300, and I 
keep getting the error;
make[3]: Entering directory `/home/ben/pdalinux/linux/drivers/char'
mipsel-linux-gcc -I /home/ben/pdalinux/linux/include/asm/gcc -D__KERNEL__ 
-I/home/ben/pdalinux/linux/include -Wall -Wstrict-prototypes -Wno-trigraphs 
-O2 -fomit-frame-pointer -fno-strict-aliasing -fno-common -G 0 -mno-abicalls 
-fno-pic -mcpu=r3000 -mips1 -pipe    -c -o serial_tx3912.o serial_tx3912.c
In file included from serial_tx3912.h:14,
                 from serial_tx3912.c:32:
/home/ben/pdalinux/linux/include/linux/serialP.h:50: field `icount' has 
incomplete type
In file included from serial_tx3912.h:15,
                 from serial_tx3912.c:32:
/home/ben/pdalinux/linux/include/linux/generic_serial.h:29: warning: `struct 
serial_struct' declared inside parameter list
/home/ben/pdalinux/linux/include/linux/generic_serial.h:29: warning: its 
scope is only this definition or declaration,
/home/ben/pdalinux/linux/include/linux/generic_serial.h:29: warning: which is 
probably not what you want.
/home/ben/pdalinux/linux/include/linux/generic_serial.h:100: warning: `struct 
serial_struct' declared inside parameter list
/home/ben/pdalinux/linux/include/linux/generic_serial.h:101: warning: `struct 
serial_struct' declared inside parameter list
In file included from serial_tx3912.c:32:
serial_tx3912.h:88: field `icount' has incomplete type
serial_tx3912.c: In function `transmit_char_pio':
serial_tx3912.c:189: `SERIAL_XMIT_SIZE' undeclared (first use in this 
function)
serial_tx3912.c:189: (Each undeclared identifier is reported only once
serial_tx3912.c:189: for each function it appears in.)
serial_tx3912.c: In function `rs_open':
serial_tx3912.c:605: `ASYNC_SPLIT_TERMIOS' undeclared (first use in this 
function)
serial_tx3912.c: In function `rs_ioctl':
serial_tx3912.c:675: sizeof applied to an incomplete type
serial_tx3912.c:676: warning: passing arg 2 of `gs_getserial' from 
incompatible pointer type
serial_tx3912.c:680: sizeof applied to an incomplete type
serial_tx3912.c:681: warning: passing arg 2 of `gs_setserial' from 
incompatible pointer type
make[3]: *** [serial_tx3912.o] Error 1
make[3]: Leaving directory `/home/ben/pdalinux/linux/drivers/char'
make[2]: *** [first_rule] Error 2
make[2]: Leaving directory `/home/ben/pdalinux/linux/drivers/char'
make[1]: *** [_subdir_char] Error 2
make[1]: Leaving directory `/home/ben/pdalinux/linux/drivers'
make: *** [_dir_drivers] Error 2

Am I doing something wrong, or is there a bug in the CVS sources?
Thanks
Ben Hague

Re: Kernel compilation problem

[Ralf Baechle]

On Sat, Aug 25, 2001 at 12:18:59PM +0000, Ben Hague wrote:

>   I'm trying to compile the latest CVS kernel for a Philips Nino 300, and I 
> keep getting the error;

Known problem.  I asked the Nino maintainer to fix it a few days ago and
he promised to do this weekend.

  Ralf

Kernel compilation problem

[Zgembo Adislic]

Hello,
 
 I using linux kernel 2.6.8.1 for my embedded ARM target and I am able to compile it without any problems using gcc-3.4-0 cross-compiler. For some reason I had to modify one of my device drivers to work with files and I had to include "stdio.h" library for that reason. But now when I execute make command I got an error that says that
 
 "stdio.h: No such file or directory"
And all functions that I used form that library are not recognized. 
 
 I really see no reason why this should not work?! All suggestions are welcome.
 
---------------------------------
8:00? 8:25? 8:40?  Find a flick in no time
 with theYahoo! Search movie showtime shortcut.

RE: Kernel compilation problem

[Woodruff, Richard]

>  "stdio.h: No such file or directory"
> And all functions that I used form that library are not recognized.

They don't exist in that form in the kernel context.  Only
user/application context is this available.

>  I really see no reason why this should not work?! All suggestions are
> welcome.

Your expectation is not correct.  Things in the kernel are built and
have different attributes as compared for things meant for user space.

Subsets of the common C Library functions are available in the kernel.
They do not always carry all the features of the
glibc/uclibc/newlibc/... counterparts.  The kernel environment is much
more restrictive.

Regards,
Richard W.

RE: kernel compilation problem

[Pacheco Jason NPRI]

> The spurious 8259A interrupt error is an incredibly common and 
> non-dangerous error message.  Most likely it is caused by the 
> uniprocessor APIC (Advanced Programmable Interupt Conrtoller) linux 
> code.  

I blame the sunspots...
-
To unsubscribe from this list: send the line "unsubscribe linux-newbie" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.linux-learn.org/faqs

RE: kernel compilation problem

[Ray Olszewski]

At 08:46 AM 12/23/2003 -0500, Pacheco Jason NPRI wrote:
> >       Hi,
> > I was compiling the 2.4.18 kernel as a practice run and when I booted
> > thru it I recieved a message during boottime:
> > Kernel panic: VFS unable to mount root fs on 03:05.
> > any clues why?

Where is the host's actual root filesystem? 03:05 says the kernel is 
looking for it on /dev/hda5 ... is that the right place or not? (It's not 
typical for a Debian install, but not impossible either.)

If it is correct -- what filesystem type is it? (ext2, most likely). Make 
sure your kernel has support for that filesystem type compiled in.

If it is incorrect -- what partition should it be? Are you compiling the 
kernel on the same system you are trying to boot it on? What happens if you 
specify the correct partition at the lilo (or grub or whatever you use for 
a bootloader) prompt? Are you booting from hda?


>Did you copy over your /usr/src/linux/System.map file into /boot? Also,
>did you build an initial ramdisk via mkinitrd? A lot of times that last
>step is overlooked but is necessary because it loads the initial drivers
>needed to mount the hard disk partitions.

No it's not. It is only necessary if the kernel does not have ide support 
compiled in (not usual) or dos not have support for your filesystem type 
compiled in (again unusual, if the filesysystem is ext2). Typical, 
"vanilla" systems do not need the initrd step to boot/init.

> > I run the Debian woody release.

[rest omitted]



-
To unsubscribe from this list: send the line "unsubscribe linux-newbie" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.linux-learn.org/faqs

Re: kernel compilation problem

[joy]

Sorry to find a trivial answer to the Q. The settings were alright .
thanks a lot , Jason and Ray, All I needed to do was to to copy the
System.map as this Kernel does not use the other two System.map(s)  in
/boot (they are for kernels compiled with diff options).
But still ,I would like to get rid of the irritating
              spurious 8259A interrupt : IRQ7
message I get on Boottime.
I run an athlon 2400+ , with a biostar nforce2 mobo and the normal
peripherals, nothing out of the ordinary.


-
To unsubscribe from this list: send the line "unsubscribe linux-newbie" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.linux-learn.org/faqs

Re: kernel compilation problem

[Beolach]

joy wrote:
> Sorry to find a trivial answer to the Q. The settings were alright .
> thanks a lot , Jason and Ray, All I needed to do was to to copy the
> System.map as this Kernel does not use the other two System.map(s)  in
> /boot (they are for kernels compiled with diff options).
> But still ,I would like to get rid of the irritating
>              spurious 8259A interrupt : IRQ7
> message I get on Boottime.
> I run an athlon 2400+ , with a biostar nforce2 mobo and the normal
> peripherals, nothing out of the ordinary.
> 

The spurious 8259A interrupt error is an incredibly common and 
non-dangerous error message.  Most likely it is caused by the 
uniprocessor APIC (Advanced Programmable Interupt Conrtoller) linux 
code.  It's really not dangerous or bad or anything.  But if you really 
do want to get rid of it, try recompiling the kernel without 
uniprocessor APIC support ("Local APIC support on uniprocessors" in make 
menuconfig).  This should get rid of the message.

A couple good references for this issue are: 
<http://www.linuxchix.org/pipermail/techtalk/2002-August/012697.html>

and this thread on the LKML (Linux Kernel Mailing List):
<http://www.cs.helsinki.fi/linux/linux-kernel/2001-47/0697.html>

HTH,
Conway S. Smith

-
To unsubscribe from this list: send the line "unsubscribe linux-newbie" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.linux-learn.org/faqs

RE: kernel compilation problem

[Pacheco Jason NPRI]

>       Hi,
> I was compiling the 2.4.18 kernel as a practice run and when I booted 
> thru it I recieved a message during boottime:
> Kernel panic: VFS unable to mount root fs on 03:05.
> any clues why?

Did you copy over your /usr/src/linux/System.map file into /boot? Also,
did you build an initial ramdisk via mkinitrd? A lot of times that last
step is overlooked but is necessary because it loads the initial drivers
needed to mount the hard disk partitions.

> I run the Debian woody release.
> also when I boot thru my working kernel, I get a message 
> during boottime:
> spurious 8259A interrupt : IRQ7
> any ideas?

I just recently built my first Debian system and I get the same error,
except I think the number is different. The system was a junk testbed, so
I figured it was my piece of garbage mobo, which his a Soyo K7ADA+. Anyway,
the error doesn't seem to cause a problem other than having that annoying
message popup here and there, but I would like to figure out how to solve
it as well.
-
To unsubscribe from this list: send the line "unsubscribe linux-newbie" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.linux-learn.org/faqs

kernel compilation problem

[joy]

      Hi,
I was compiling the 2.4.18 kernel as a practice run and when I booted 
thru it I recieved a message during boottime:
Kernel panic: VFS unable to mount root fs on 03:05.
any clues why?
I run the Debian woody release.
also when I boot thru my working kernel, I get a message during boottime:
spurious 8259A interrupt : IRQ7
any ideas?
thanx in advance,
Joy.M.M

-
To unsubscribe from this list: send the line "unsubscribe linux-newbie" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.linux-learn.org/faqs

kernel compilation problem

[Charles Lepple]

While configuring the kernel (2.2.5 from ftp.kernel.org), I get the
following message:

 Sound card support (CONFIG_SOUND) [N/y/m/?]
>scripts/Configure: drivers/sgi/char/Config.in: No such file or
directory
 *
 * Kernel hacking

Then, while executing 'make dep', I see a bunch of messages about
'SOCK_DGRAM' being redefined.

And when I try 'make zImage', it bombs compiling init/main.c -- there's
a "parse error" in linux/sched.h, and lots of stuff breaks. I have the
compile log if anyone can decipher it, but is this just a simple case of
me overlooking a patch? Or where can I find a preconfigured kernel
source tree?

Thanks in advance,

--
Charles Lepple
System Administrator, Virginia Tech EE Workstation Labs
clepple@ee.vt.edu || http://www.foo.tho.org/charles/

Re: kernel compilation problem

[Ralf Baechle]

On Thu, Apr 15, 1999 at 04:20:04AM +0000, Charles Lepple wrote:

> Then, while executing 'make dep', I see a bunch of messages about
> 'SOCK_DGRAM' being redefined.
> 
> And when I try 'make zImage', it bombs compiling init/main.c -- there's
> a "parse error" in linux/sched.h, and lots of stuff breaks. I have the
> compile log if anyone can decipher it, but is this just a simple case of
> me overlooking a patch? Or where can I find a preconfigured kernel
> source tree?

Get the kernel from {ftp,www}.linux.sgi.com.

  Ralf

Re: kernel compilation problem

[Matthias Kleinschmidt]

[-- Attachment #1: Type: text/plain, Size: 763 bytes --]

Hi,

On Thu, Apr 15, 1999 at 04:20:04AM +0000, Charles Lepple wrote:
> While configuring the kernel (2.2.5 from ftp.kernel.org), I get the
> following message:

You have to get the kernel from cvs:

export CVSROOT=":pserver:cvs@linus.linux.sgi.com:/cvs"
cvs login (cvs as password)
cvs -z3 checkout linux

The config is a little tricky. Look at
ftp.linux.sgi.com/pub/linux/mips/test/vmlinux-indy-990212.conf
as example

> And when I try 'make zImage', it bombs compiling init/main.c

"make zImage" doesn't work. Try "make vmlinux".

Matthias

-- 
Matthias Kleinschmidt
Cedar Dell 568B, Box 5398
UMass Dartmouth
285 Old Westport Rd.
North Dartmouth, MA 02747
Tel.:  +1 508 910 6851
email: matthias@fmc-container.mach.uni-karlsruhe.de

[-- Attachment #2: Type: application/pgp-signature, Size: 290 bytes --]