vapier | 1 Mar 2008 12:29
Favicon

mainline kernel building

Summary:

   Building blackfin-2.6
### BF527-EZKIT_defconfig
###### BF522: OK
###### BF523: OK
###### BF524: OK
###### BF525: OK
###### BF526: OK
###### BF527: OK
### BF533-EZKIT_defconfig
###### BF531: OK
###### BF532: OK
###### BF533: OK
### BF533-STAMP_defconfig
###### BF531: OK
###### BF532: OK
###### BF533: OK
### BF537-STAMP_defconfig
###### BF534: OK
###### BF536: OK
###### BF537: OK
### BF548-EZKIT_defconfig
###### BF542: FAIL
###### BF544: FAIL
###### BF547: FAIL
###### BF549: FAIL
###### BF548: FAIL
### BF561-EZKIT_defconfig
###### BF561: OK
(Continue reading)

Mike Frysinger | 1 Mar 2008 21:08
Picon

regression: ntfs build failure in linux-2.6.git on no-mmu: no is_vmalloc_addr()

On Saturday 01 March 2008 06:29:59 vapier@... wrote:
> --- blackfin-2.6/BF548-EZKIT.BF542
> --- blackfin-2.6/BF548-EZKIT.BF544
> --- blackfin-2.6/BF548-EZKIT.BF547
> --- blackfin-2.6/BF548-EZKIT.BF548
> --- blackfin-2.6/BF548-EZKIT.BF549
> --- linux-2.6/BF548-EZKIT.BF542
> --- linux-2.6/BF548-EZKIT.BF544
> --- linux-2.6/BF548-EZKIT.BF547
> --- linux-2.6/BF548-EZKIT.BF548
> --- linux-2.6/BF548-EZKIT.BF549
> --- next-20080229/BF548-EZKIT.BF542
> --- next-20080229/BF548-EZKIT.BF544
> --- next-20080229/BF548-EZKIT.BF547
> --- next-20080229/BF548-EZKIT.BF548
> --- next-20080229/BF548-EZKIT.BF549
> <stdin>:1377:2: warning: #warning syscall timerfd_create not implemented
> <stdin>:1389:2: warning: #warning syscall timerfd_settime not implemented
> <stdin>:1393:2: warning: #warning syscall timerfd_gettime not implemented
> In file included from fs/ntfs/attrib.c:32:
> fs/ntfs/malloc.h: In function 'ntfs_free':
> fs/ntfs/malloc.h:88: error: implicit declaration of function
> 'is_vmalloc_addr' make[2]: *** [fs/ntfs/attrib.o] Error 1
> make[1]: *** [fs/ntfs] Error 2
> make: *** [fs] Error 2

commit 9e2779fa281cfda13ac060753d674bbcaa23367e changed fs/ntfs/malloc.h from
using VMALLOC_* defines to is_vmalloc_addr() (which makes sense), but Blackfin
was defining VMALLOC_* (to basically the range of all memory addresses like
most no-mmu ports)
(Continue reading)

Vitja Makarov | 2 Mar 2008 14:14
Picon
Gravatar

toolchain

Hi!

What toolchain do you use to build trunk?
I found a bug in trunk one.

http://blackfin.uclinux.org/gf/project/toolchain/tracker/?action=TrackerItemEdit&tracker_item_id=3948

vitja.
vapier | 3 Mar 2008 12:28
Favicon

mainline kernel building

Summary:

   Building blackfin-2.6
### BF527-EZKIT_defconfig
###### BF522: OK
###### BF523: OK
###### BF524: OK
###### BF525: OK
###### BF526: OK
###### BF527: OK
### BF533-EZKIT_defconfig
###### BF531: OK
###### BF532: OK
###### BF533: OK
### BF533-STAMP_defconfig
###### BF531: OK
###### BF532: OK
###### BF533: OK
### BF537-STAMP_defconfig
###### BF534: OK
###### BF536: OK
###### BF537: OK
### BF548-EZKIT_defconfig
###### BF542: FAIL
###### BF544: FAIL
###### BF547: FAIL
###### BF549: FAIL
###### BF548: FAIL
### BF561-EZKIT_defconfig
###### BF561: OK
(Continue reading)

Mike Frysinger | 3 Mar 2008 18:55
Picon

Re: toolchain

On Sun, Mar 2, 2008 at 8:14 AM, Vitja Makarov
<vitja.makarov@...> wrote:
>  What toolchain do you use to build trunk?

this is in the FAQ:
http://docs.blackfin.uclinux.org/doku.php?id=faq

usually the mailing list is for development while the forums is for questions

>  I found a bug in trunk one.

it's probably always been there.  i doubt it's specific to trunk.
-mike
Robert Fries | 3 Mar 2008 22:13

Re: bf561 / SMP-like patch

The attachment is a patch to add irq affinity to SMP.  This patch 
should be applied on top of the patch I submitted on 21Jan08 
(patch.Smp.2007R1.1-RC3.git).    That earlier patch was a refactoring 
of the SMP patch for 2007R1.1.

This patch uses the PROG{0,1,2}_INTB interrupts to control which PF 
pin interrupts go to coreB while the PROG{0,1,2}_INTA interrupts 
control those that go to coreA.  This creates a conflict with the PM 
facility which uses the PROG{0,1,2}_INTB interrupts to wake up the 
processor.  This patch uses a config variable BFIN_IRQ_AFFINITY to 
control whether the irq affinity code is active so that it can be 
disabled when the PM facility is needed.

This patch also corrects what I believe to be a typo in the routine 
platform_secondary_init in the file arch/blackfin/mach-bf561/smp.c.  
A comment at the top of that routine indicates that it clones 
registers from coreA to coreB.  However it just copies the registers 
SICB_IAR{0-9} onto themselves.  This patch changes that code to copy 
from the registers SICA_IAR{0-9} to the corresponding register in the 
set SICB_IAR{0-9}. 

-bob
_______________________________________________
Uclinux-dist-devel mailing list
Uclinux-dist-devel@...
http://blackfin.uclinux.org/mailman/listinfo/uclinux-dist-devel
Philippe Gerum | 3 Mar 2008 22:30
Favicon

Re: Re: bf561 / SMP-like patch

Robert Fries wrote:
> The attachment is a patch to add irq affinity to SMP.  This patch 
> should be applied on top of the patch I submitted on 21Jan08 
> (patch.Smp.2007R1.1-RC3.git).    That earlier patch was a refactoring 
> of the SMP patch for 2007R1.1.
> 
> This patch uses the PROG{0,1,2}_INTB interrupts to control which PF 
> pin interrupts go to coreB while the PROG{0,1,2}_INTA interrupts 
> control those that go to coreA.  This creates a conflict with the PM 
> facility which uses the PROG{0,1,2}_INTB interrupts to wake up the 
> processor.  This patch uses a config variable BFIN_IRQ_AFFINITY to 
> control whether the irq affinity code is active so that it can be 
> disabled when the PM facility is needed.
> 
> This patch also corrects what I believe to be a typo in the routine 
> platform_secondary_init in the file arch/blackfin/mach-bf561/smp.c.

Oops, indeed.

> A comment at the top of that routine indicates that it clones 
> registers from coreA to coreB.  However it just copies the registers 
> SICB_IAR{0-9} onto themselves.  This patch changes that code to copy 
> from the registers SICA_IAR{0-9} to the corresponding register in the 
> set SICB_IAR{0-9}. 
> 
> -bob
> 
> 
> ------------------------------------------------------------------------
> 
(Continue reading)

Bryan Wu | 3 Mar 2008 23:16

Re: Re: bf561 / SMP-like patch

On Mon, Mar 3, 2008 at 1:13 PM, Robert Fries <robert_fries@...> wrote:
> The attachment is a patch to add irq affinity to SMP.  This patch
>  should be applied on top of the patch I submitted on 21Jan08
>  (patch.Smp.2007R1.1-RC3.git).    That earlier patch was a refactoring
>  of the SMP patch for 2007R1.1.
>

Is there any plan to support SMP patch for 2008R1 or SVN trunk?

Thanks
-Bryan
vapier | 4 Mar 2008 12:28
Favicon

mainline kernel building

Summary:

   Building blackfin-2.6
### BF527-EZKIT_defconfig
###### BF522: OK
###### BF523: OK
###### BF524: OK
###### BF525: OK
###### BF526: OK
###### BF527: OK
### BF533-EZKIT_defconfig
###### BF531: OK
###### BF532: OK
###### BF533: OK
### BF533-STAMP_defconfig
###### BF531: OK
###### BF532: OK
###### BF533: OK
### BF537-STAMP_defconfig
###### BF534: OK
###### BF536: OK
###### BF537: OK
### BF548-EZKIT_defconfig
###### BF542: FAIL
###### BF544: FAIL
###### BF547: FAIL
###### BF549: FAIL
###### BF548: FAIL
### BF561-EZKIT_defconfig
###### BF561: OK
(Continue reading)

vapier | 5 Mar 2008 12:28
Favicon

mainline kernel building

Summary:

   Building blackfin-2.6
### BF527-EZKIT_defconfig
###### BF522: OK
###### BF523: OK
###### BF524: OK
###### BF525: OK
###### BF526: OK
###### BF527: OK
### BF533-EZKIT_defconfig
###### BF531: OK
###### BF532: OK
###### BF533: OK
### BF533-STAMP_defconfig
###### BF531: OK
###### BF532: OK
###### BF533: OK
### BF537-STAMP_defconfig
###### BF534: OK
###### BF536: OK
###### BF537: OK
### BF548-EZKIT_defconfig
###### BF542: FAIL
###### BF544: FAIL
###### BF547: FAIL
###### BF549: FAIL
###### BF548: FAIL
### BF561-EZKIT_defconfig
###### BF561: OK
(Continue reading)


Gmane