by Kamalesh Babulal

[permalink] [raw]

Subject: Re: 2.6.24-rc8-mm1 Build Failure on scsi driver

Hi Andrew,

The kernel build fails with following error

drivers/scsi/aha152x.o: In function `aha152x_host_reset_host':
/home/kamalesh/scrap/linux-2.6.24-rc8/drivers/scsi/aha152x.c:1324: multiple definition of `aha152x_host_reset_host'
drivers/scsi/pcmcia/built-in.o:/home/kamalesh/scrap/linux-2.6.24-rc8/drivers/scsi/aha152x.c:1324: first defined here
drivers/scsi/aha152x.o: In function `aha152x_release':
/home/kamalesh/scrap/linux-2.6.24-rc8/drivers/scsi/aha152x.c:908: multiple definition of `aha152x_release'
drivers/scsi/pcmcia/built-in.o:/home/kamalesh/scrap/linux-2.6.24-rc8/drivers/scsi/aha152x.c:908: first defined here
ld: Warning: size of symbol `aha152x_release' changed from 68 in drivers/scsi/pcmcia/built-in.o to 100 in drivers/scsi/aha152x.o
drivers/scsi/aha152x.o: In function `aha152x_probe_one':
/home/kamalesh/scrap/linux-2.6.24-rc8/drivers/scsi/aha152x.c:772: multiple definition of `aha152x_probe_one'
drivers/scsi/pcmcia/built-in.o:/home/kamalesh/scrap/linux-2.6.24-rc8/drivers/scsi/aha152x.c:772: first defined here
drivers/scsi/fdomain.o:(.data+0x0): multiple definition of `fdomain_driver_template'
drivers/scsi/pcmcia/built-in.o:(.data+0x5a0): first defined here
drivers/scsi/fdomain.o: In function `fdomain_16x0_bus_reset':
/home/kamalesh/scrap/linux-2.6.24-rc8/drivers/scsi/fdomain.c:1568: multiple definition of `fdomain_16x0_bus_reset'
drivers/scsi/pcmcia/built-in.o:/home/kamalesh/scrap/linux-2.6.24-rc8/drivers/scsi/fdomain.c:1568: first defined here
drivers/scsi/fdomain.o: In function `__fdomain_16x0_detect':
/home/kamalesh/scrap/linux-2.6.24-rc8/drivers/scsi/fdomain.c:894: multiple definition of `__fdomain_16x0_detect'
drivers/scsi/pcmcia/built-in.o:/home/kamalesh/scrap/linux-2.6.24-rc8/drivers/scsi/fdomain.c:894: first defined here
ld: Warning: size of symbol `__fdomain_16x0_detect' changed from 1206 in drivers/scsi/pcmcia/built-in.o to 1700 in drivers/scsi/fdomain.o
drivers/scsi/fdomain.o: In function `fdomain_setup':
/home/kamalesh/scrap/linux-2.6.24-rc8/drivers/scsi/fdomain.c:554: multiple definition of `fdomain_setup'
drivers/scsi/pcmcia/built-in.o:/home/kamalesh/scrap/linux-2.6.24-rc8/drivers/scsi/fdomain.c:554: first defined here
make[2]: *** [drivers/scsi/built-in.o] Error 1
make[1]: *** [drivers/scsi] Error 2
make: *** [drivers] Error 2

--
Thanks & Regards,
Kamalesh Babulal,
Linux Technology Center,
IBM, ISTL.

2008-01-17 16:49:25

2008-01-17 18:11:31

by Mariusz Kozlowski

[permalink] [raw]

Subject: Re: [-mm Patch] uml: fix a building error

Hello,

> This patch fixes this building error:
> ...
> drivers/char/mem.c: In function ‘read_mem’:
> drivers/char/mem.c:136: error: implicit declaration of function ‘unxlate_dev_mem_ptr’
> ...

I see this on sparc64 as well:

CC drivers/char/mem.o
drivers/char/mem.c: In function 'read_mem':
drivers/char/mem.c:136: error: implicit declaration of function 'unxlate_dev_mem_ptr'
make[2]: *** [drivers/char/mem.o] Error 1
make[1]: *** [drivers/char] Error 2
make: *** [drivers] Error 2

Does sparc64 need similar fix?

Regards,

Mariusz

2008-01-17 18:18:34

On Thu, 17 Jan 2008 21:45:39 +0530 Kamalesh Babulal <[email protected]> wrote:

> Hi Andrew,
>
> The kernel build fails with following error
>
> drivers/scsi/aha152x.o: In function `aha152x_host_reset_host':
> /home/kamalesh/scrap/linux-2.6.24-rc8/drivers/scsi/aha152x.c:1324: multiple definition of `aha152x_host_reset_host'
> drivers/scsi/pcmcia/built-in.o:/home/kamalesh/scrap/linux-2.6.24-rc8/drivers/scsi/aha152x.c:1324: first defined here
> drivers/scsi/aha152x.o: In function `aha152x_release':
> /home/kamalesh/scrap/linux-2.6.24-rc8/drivers/scsi/aha152x.c:908: multiple definition of `aha152x_release'
> drivers/scsi/pcmcia/built-in.o:/home/kamalesh/scrap/linux-2.6.24-rc8/drivers/scsi/aha152x.c:908: first defined here
> ld: Warning: size of symbol `aha152x_release' changed from 68 in drivers/scsi/pcmcia/built-in.o to 100 in drivers/scsi/aha152x.o
> drivers/scsi/aha152x.o: In function `aha152x_probe_one':

Neat. Seems that the scsi build system is linking together two copies of
drivers/scsi/aha152x.o. One via drivers/scsi/aha152x.o directly and the
other via drivers/scsi/pcmcia/built-in.o.

Please send the .config.

I'm looking suspiciously at this, from git-scsi-misc:

commit 8ae732a91df051aba6820068a47b631a06599d84
Author: Tejun Heo <[email protected]>
Date: Fri Dec 7 22:36:23 2007 +0900

[SCSI] make pcmcia directory use obj-y|m instead of subdir-y|m

subdir-y|m isn't supposed to contain modules or built-in components.
Change subdir-$(CONFIG_PCMCIA) to obj-$(CONFIG_PCMCIA).

Signed-off-by: Tejun Heo <[email protected]>
Acked-by: Sam Ravnborg <[email protected]>
Signed-off-by: James Bottomley <[email protected]>

diff --git a/drivers/scsi/Makefile b/drivers/scsi/Makefile
index b5441f5..93e1428 100644
--- a/drivers/scsi/Makefile
+++ b/drivers/scsi/Makefile
@@ -17,7 +17,7 @@
CFLAGS_aha152x.o = -DAHA152X_STAT -DAUTOCONF
CFLAGS_gdth.o = # -DDEBUG_GDTH=2 -D__SERIAL__ -D__COM2__ -DGDTH_STATISTICS

-subdir-$(CONFIG_PCMCIA) += pcmcia
+obj-$(CONFIG_PCMCIA) += pcmcia/

obj-$(CONFIG_SCSI) += scsi_mod.o
obj-$(CONFIG_SCSI_TGT) += scsi_tgt.o

2008-01-17 19:14:59

by Randy Dunlap

[permalink] [raw]

Subject: Re: 2.6.24-rc8-mm1 Kernel oops will running kernbench

* Kamalesh Babulal <[email protected]> [2008-01-18 16:14:00]:

> Paul Mackerras wrote:
> > Kamalesh Babulal writes:
> >
> >>>>> NIP: 0000000000004570 LR: 000000000fc42dc0 CTR: 0000000000000000
> >>>>> REGS: c00000077b6bf8c0 TRAP: 0300 Not tainted (2.6.24-rc8-mm1-autotest)
> >>>>> MSR: 8000000000001000 <ME> CR: 28022422 XER: 00000000
> >>>>> DAR: c00000077b6bfce0, DSISR: 000000000a000000
> >
> > Actually, how much RAM does this machine have? If it has less than
> > 32GB, then the problem is that the kernel stack pointer is bogus.
> > (How it got to be bogus is the interesting question, of course. :)
> >
> > Paul.
> >
> Hi Paul,
>
> The machine has around 30GB of RAM, do you want me to try, by taking
> the git-powerpc.patch out of the series and try reproducing the oops.
>

Kamalesh, I thought I saw Paul's request for trying without
git-powerpc.patch (it's in a separate email).

--
Warm Regards,
Balbir Singh
Linux Technology Center
IBM, ISTL

2008-01-18 13:32:46

by Rafael J. Wysocki

[permalink] [raw]

2008-01-21 14:59:51

by James Bottomley

[permalink] [raw]

Subject: Re: [PATCH] SCSI: fix isa/pcmcia compile problem

On Mon, 2008-01-21 at 09:56 +0000, Christoph Hellwig wrote:
> On Fri, Jan 18, 2008 at 04:20:40PM +0900, Tejun Heo wrote:
> > aha152x.c and fdomain are built twice - once for the isa driver and
> > once for the PCMCIA one. Through #ifdefs, the compiled codes are
> > slightly different; thus, global symbols need to be given different
> > names depending on which flavor is being built. This patch adds
> > GLOBAL() macro to aha152x.h and fdomain.h which change the symbol
> > depending on PCMCIA.
> >
> > This bug has always existed but has been masked by the fact the
> > drivers/scsi/pcmcia used subdir-(y|m) instead of obj-(y|m) which made
> > drivers/scsi/pcmcia/built_in.o not linked into the kernel and thus
> > avoided the duplicate symbols during compilation.
>
> The right fix would be to compile it only once and attach it to both
> busses. It would be nice if someone could look into that instead of
> hacking around the issue.

I agree in principle, but without the hardware such a change would be
untested ... which is what makes me worry about it.

James

2008-01-21 18:31:31

by Valdis Klētnieks

[permalink] [raw]

Subject: Re: 2.6.24-rc8-mm1 - SELinux issues

On Thu, 17 Jan 2008 02:35:14 PST, Andrew Morton said:

> ftp://ftp.kernel.org/pub/linux/kernel/people/akpm/patches/2.6/2.6.24-rc8/2.6.24-rc8-mm1/
>
> - selinux is busted on one of my two selinux-enabled test machines.

This problem is fixed in Paul Moore's latest spin of the networking patches - I
was able to quilt up a fixed-up -rc8-mm1 with an updated git-lblnet.patch built
from his latest patch-bomb (labelled as 'RFC PATCH v12' posted to the selinux
and linux-security-module lists on Thursday, so the next pull of that git tree
into -mm should be OK...

Attachments:

(No filename) (226.00 B)

2008-01-21 18:53:24

On Tue, 22 Jan 2008 15:13:58 -0800
Dave Hansen <[email protected]> wrote:

> @@ -566,10 +567,26 @@ static void mark_files_ro(struct super_b
> {
> struct file *f;
>
> +retry:
> file_list_lock();
> list_for_each_entry(f, &sb->s_files, f_u.fu_list) {
> - if (S_ISREG(f->f_path.dentry->d_inode->i_mode) && file_count(f))
> - f->f_mode &= ~FMODE_WRITE;
> + struct vfsmount mnt;
> + if (!S_ISREG(f->f_path.dentry->d_inode->i_mode))
> + continue;
> + if (!file_count(f))
> + continue;
> + if (!(f->f_mode & FMODE_WRITE))
> + continue;
> + f->f_mode &= ~FMODE_WRITE;
> + f->f_mnt_write_state |= FILE_MNT_WRITE_RELEASED;
> + mnt = f->f_path.mnt;
> + file_list_unlock();
> + /*
> + * This can sleep, so we can't hold
> + * the file_list_lock() spinlock.
> + */
> + mnt_drop_write(mnt);
> + goto retry;
> }
> file_list_unlock();
> }
> _

this doesn't even compile. How much confidence am I supposed
to have that once I've fixed it, it will actually work?