Hi,
Any ideas how I could configure the kernel to get past this
problem? Currently the config file says this about MCE:
CONFIG_GART_IOMMU=y
CONFIG_SWIOTLB=y
CONFIG_X86_MCE=y
# CONFIG_X86_MCE_INTEL is not set
Can I safely set CONFIG_X86_MCE to no or not set? Or is this something
else completely?
Thanks,
Mark
lightning linux-2.6.14-rc2-rt7 # make
CHK include/linux/version.h
CHK include/linux/compile.h
CHK usr/initramfs_list
CC arch/x86_64/kernel/mce.o
In file included from arch/x86_64/kernel/mce.c:17:
include/linux/fs.h: In function `lock_super':
include/linux/fs.h:847: warning: implicit declaration of function `down'
include/linux/fs.h: In function `unlock_super':
include/linux/fs.h:853: warning: implicit declaration of function `up'
arch/x86_64/kernel/mce.c: In function `mce_read':
arch/x86_64/kernel/mce.c:392: warning: type defaults to `int' in
declaration of `DECLARE_MUTEX'
arch/x86_64/kernel/mce.c:392: warning: parameter names (without types)
in function declaration
arch/x86_64/kernel/mce.c:401: error: `mce_read_sem' undeclared (first
use in this function)
arch/x86_64/kernel/mce.c:401: error: (Each undeclared identifier is
reported only once
arch/x86_64/kernel/mce.c:401: error: for each function it appears in.)
make[1]: *** [arch/x86_64/kernel/mce.o] Error 1
make: *** [arch/x86_64/kernel] Error 2
lightning linux-2.6.14-rc2-rt7 #
On Thu, 2005-09-29 at 19:07 -0700, Mark Knecht wrote:
> Hi,
> Any ideas how I could configure the kernel to get past this
> problem? Currently the config file says this about MCE:
>
> CONFIG_GART_IOMMU=y
> CONFIG_SWIOTLB=y
> CONFIG_X86_MCE=y
> # CONFIG_X86_MCE_INTEL is not set
>
> Can I safely set CONFIG_X86_MCE to no or not set? Or is this something
> else completely?
I think it's something else completely .. You would be better off
turning on complete preemption .
Daniel
On 9/29/05, Daniel Walker <[email protected]> wrote:
> On Thu, 2005-09-29 at 19:07 -0700, Mark Knecht wrote:
> > Hi,
> > Any ideas how I could configure the kernel to get past this
> > problem? Currently the config file says this about MCE:
> >
> > CONFIG_GART_IOMMU=y
> > CONFIG_SWIOTLB=y
> > CONFIG_X86_MCE=y
> > # CONFIG_X86_MCE_INTEL is not set
> >
> > Can I safely set CONFIG_X86_MCE to no or not set? Or is this something
> > else completely?
>
> I think it's something else completely .. You would be better off
> turning on complete preemption .
>
> Daniel
Thanks.
make allnoconfig builds
make defconfig fails as per my earlier message
make defconfig and then turning on complete preemption is doing this:
CC mm/fadvise.o
CC mm/page_alloc.o
CC mm/page-writeback.o
CC mm/pdflush.o
CC mm/readahead.o
CC mm/slab.o
mm/slab.c:2404: error: conflicting types for 'kmem_cache_alloc_node'
include/linux/slab.h:122: error: previous declaration of
'kmem_cache_alloc_node' was here
mm/slab.c:2404: error: conflicting types for 'kmem_cache_alloc_node'
include/linux/slab.h:122: error: previous declaration of
'kmem_cache_alloc_node' was here
make[1]: *** [mm/slab.o] Error 1
make: *** [mm] Error 2
lightning linux-2.6.14-rc2-rt7 #
I will continue on. Thanks very much for your help.
Thanks,
Mark
On 9/29/05, Mark Knecht <[email protected]> wrote:
> On 9/29/05, Daniel Walker <[email protected]> wrote:
> > I think it's something else completely .. You would be better off
> > turning on complete preemption .
> >
> > Daniel
>
> Thanks.
>
> make allnoconfig builds
> make defconfig fails as per my earlier message
>
> make defconfig and then turning on complete preemption is doing this:
>
> CC mm/fadvise.o
> CC mm/page_alloc.o
> CC mm/page-writeback.o
> CC mm/pdflush.o
> CC mm/readahead.o
> CC mm/slab.o
> mm/slab.c:2404: error: conflicting types for 'kmem_cache_alloc_node'
> include/linux/slab.h:122: error: previous declaration of
> 'kmem_cache_alloc_node' was here
> mm/slab.c:2404: error: conflicting types for 'kmem_cache_alloc_node'
> include/linux/slab.h:122: error: previous declaration of
> 'kmem_cache_alloc_node' was here
> make[1]: *** [mm/slab.o] Error 1
> make: *** [mm] Error 2
> lightning linux-2.6.14-rc2-rt7 #
>
> I will continue on. Thanks very much for your help.
>
> Thanks,
> Mark
>
Great! It still failed but built after I turned NUMA off.
Thanks VERY much!!
- Mark