Skip to contentSkip to conversation

Kernel compile error with 2.5.66

DISCUSSION John Goerzen 3 participants 4 messages Mar 29, 2003 – Mar 31, 2003

lore
Share & tools
lore
#0 lore raw

This is on PowerPC:

gcc -Wp,-MD,mm/.fremap.o.d -D__KERNEL__ -Iinclude -Wall -Wstrict-prototypes -Wno-trigraphs -O2 -fno-strict-aliasing -fno-common -Iarch/ppc -msoft-float -pipe -ffixed-r2 -Wno-uninitialized -mmultiple -mstring -fomit-frame-pointer -nostdinc -iwithprefix include -DKBUILD_BASENAME=fremap -DKBUILD_MODNAME=fremap -c -o mm/.tmp_fremap.o mm/fremap.c
In file included from mm/fremap.c:14:
include/linux/swapops.h: In function `pte_to_swp_entry':
include/linux/swapops.h:54: warning: implicit declaration of function pte_file' mm/fremap.c: In function sys_remap_file_pages':
mm/fremap.c:139: `PTE_FILE_MAX_BITS' undeclared (first use in this function) mm/fremap.c:139: (Each undeclared identifier is reported only once mm/fremap.c:139: for each function it appears in.)
make[2]: *** [mm/fremap.o] Error 1
make[1]: *** [mm] Error 2
make[1]: Leaving directory `/home/jgoerzen/kernel/linux-2.5.66'
make: *** [stamp-build] Error 2

#1 lore raw
Quoted 6 lines · John Goerzen

This is on PowerPC:
<snip>
In file included from mm/fremap.c:14:
include/linux/swapops.h: In function `pte_to_swp_entry':
include/linux/swapops.h:54: warning: implicit declaration of function pte_file' mm/fremap.c:139: PTE_FILE_MAX_BITS' undeclared (first use in this function)

Yeap, I'm getting this too. There seems to be a whole bunch of changes in include/asm-ppc64/pgtable.h that aren't in include/asm-ppc/pgtable.h

This is at least my guess.
By doing the dumb cut&paste business I've gotten further along, but haven't done enough to actually compile fully yet. I'm half nervous about doing a lot of this as I have no idea about this level on ppc (maybe benh, the maintainer of linux for power mac could help out?).

If I get anything going, I'll let you know.

on an architectural question, there seems to be a lot of duplication between ppc and ppc64, should they be more closely melded or is there a really-good-reason(tm) behind the seperation?

------------------------------
Stewart Smith
[email protected]
Ph: +61 4 3884 4332
ICQ: 6734154

#2 lore raw
Quoted 3 lines · Stewart Smith

Yeap, I'm getting this too. There seems to be a whole bunch of changes in include/asm-ppc64/pgtable.h that aren't in include/asm-ppc/pgtable.h

I should say that after my post, I tried out the linuxppc-2.5 2.5.66 tree at http://www.penguinppc.org/dev/kernel.shtml, and it did not have this problem.

However, I could not get it to compile for numerous other reasons:

* If PnP BIOS is enabled, it fails to compile the x86 assembler.

* All Modules fail to load due to some sort of missing symbol (sorry, I didn't take notes) Q_MODULE or something maybe.

* A non-module-enabled kernel fails to build, complaining of a missing isa_virt_to_bus symbol. And that happened even after I enabled the unnecessary ISA support.

* hdparm causes a kernel oops and apparent lockup (again, I didn't take notes, I was just trying to make the darn thing boot)

In short, 2.5 on PowerPC does not seem to be in a very good state right now.

-- John