* Re: about ap4evb HDMI video
2010-12-20 8:18 about ap4evb HDMI video Kuninori Morimoto
@ 2010-12-20 23:00 ` Guennadi Liakhovetski
2010-12-21 0:36 ` Kuninori Morimoto
` (3 subsequent siblings)
4 siblings, 0 replies; 6+ messages in thread
From: Guennadi Liakhovetski @ 2010-12-20 23:00 UTC (permalink / raw)
To: linux-sh
On Mon, 20 Dec 2010, Kuninori Morimoto wrote:
>
> Dear Guennadi, Magnus, Paul
>
> Now I'm using paul/rmobile-latest tree,
> and HDMI video doens't work on mach-ap4evb board.
>
> It print below when boot.
> Why is this happen ?
Strange, it works for me with the same board and branch as of commit
09dd7ded60019d6a4fd2ae20a08c4ad2bc3ed3e9 and without any warnings. I'll
forward you off-list my .config.
Thanks
Guennadi
>
> ---------------
> ...
> msgmni has been set to 500
> io scheduler noop registered (default)
> sh-mobile-hdmi sh-mobile-hdmi: Detected HDMI controller 0x0:0x0
> sh-mobile-hdmi sh-mobile-hdmi: clocks managed by runtime pm
> sh_mobile_lcdc_fb sh_mobile_lcdc_fb.1: clocks managed by runtime pm
> ------------[ cut here ]------------
> WARNING: at /opt/usr/src/WORK/morimoto/gitlinux/linux-2.6/mm/page_alloc.c:1985 )
> Modules linked in:
> [<c002a220>] (unwind_backtrace+0x0/0xe4) from [<c0033b80>] (warn_slowpath_commo)
> [<c0033b80>] (warn_slowpath_common+0x4c/0x64) from [<c0033bb0>] (warn_slowpath_)
> [<c0033bb0>] (warn_slowpath_null+0x18/0x1c) from [<c0063808>] (__alloc_pages_no)
> [<c0063808>] (__alloc_pages_nodemask+0x144/0x528) from [<c002ac90>] (__dma_allo)
> [<c002ac90>] (__dma_alloc+0x90/0x2a4) from [<c002af1c>] (dma_alloc_coherent+0x5)
> [<c002af1c>] (dma_alloc_coherent+0x54/0x60) from [<c026616c>] (sh_mobile_lcdc_p)
> [<c026616c>] (sh_mobile_lcdc_probe+0x5ac/0x820) from [<c017e8fc>] (platform_drv)
> [<c017e8fc>] (platform_drv_probe+0x14/0x18) from [<c017dab8>] (driver_probe_dev)
> [<c017dab8>] (driver_probe_device+0xc8/0x184) from [<c017dbd4>] (__driver_attac)
> [<c017dbd4>] (__driver_attach+0x60/0x84) from [<c017d324>] (bus_for_each_dev+0x)
> [<c017d324>] (bus_for_each_dev+0x48/0x74) from [<c017cc90>] (bus_add_driver+0x9)
> [<c017cc90>] (bus_add_driver+0x9c/0x20c) from [<c017dec4>] (driver_register+0xa)
> [<c017dec4>] (driver_register+0xa8/0x138) from [<c00253ac>] (do_one_initcall+0x)
> [<c00253ac>] (do_one_initcall+0xcc/0x1a0) from [<c00083c4>] (kernel_init+0x94/0)
> [<c00083c4>] (kernel_init+0x94/0x14c) from [<c0026830>] (kernel_thread_exit+0x0)
> ---[ end trace dec6997083161631 ]---
> sh_mobile_lcdc_fb sh_mobile_lcdc_fb.1: unable to allocate buffer
> sh_mobile_lcdc_fb: probe of sh_mobile_lcdc_fb.1 failed with error -12
> sh_mobile_lcdc_fb sh_mobile_lcdc_fb.0: clocks managed by runtime pm
> Console: switching to colour frame buffer device 100x30
> graphics fb0: registered sh_mobile_lcdc_fb/mainlcd as 800x480 16bpp.
> ...
> ---------------
>
> Best regards
> --
> Kuninori Morimoto
>
>
---
Guennadi Liakhovetski, Ph.D.
Freelance Open-Source Software Developer
http://www.open-technology.de/
^ permalink raw reply [flat|nested] 6+ messages in thread* Re: about ap4evb HDMI video
2010-12-20 8:18 about ap4evb HDMI video Kuninori Morimoto
2010-12-20 23:00 ` Guennadi Liakhovetski
@ 2010-12-21 0:36 ` Kuninori Morimoto
2010-12-21 0:36 ` Paul Mundt
` (2 subsequent siblings)
4 siblings, 0 replies; 6+ messages in thread
From: Kuninori Morimoto @ 2010-12-21 0:36 UTC (permalink / raw)
To: linux-sh
Dear Guennadi
> > It print below when boot.
> > Why is this happen ?
>
> Strange, it works for me with the same board and branch as of commit
> 09dd7ded60019d6a4fd2ae20a08c4ad2bc3ed3e9 and without any warnings. I'll
> forward you off-list my .config.
Thanks
your .config works well
I check my .config
Thanks again
Best regards
--
Kuninori Morimoto
^ permalink raw reply [flat|nested] 6+ messages in thread* Re: about ap4evb HDMI video
2010-12-20 8:18 about ap4evb HDMI video Kuninori Morimoto
2010-12-20 23:00 ` Guennadi Liakhovetski
2010-12-21 0:36 ` Kuninori Morimoto
@ 2010-12-21 0:36 ` Paul Mundt
2010-12-21 1:32 ` Kuninori Morimoto
2010-12-22 4:42 ` Paul Mundt
4 siblings, 0 replies; 6+ messages in thread
From: Paul Mundt @ 2010-12-21 0:36 UTC (permalink / raw)
To: linux-sh
On Tue, Dec 21, 2010 at 09:36:39AM +0900, Kuninori Morimoto wrote:
> > > It print below when boot.
> > > Why is this happen ?
> >
> > Strange, it works for me with the same board and branch as of commit
> > 09dd7ded60019d6a4fd2ae20a08c4ad2bc3ed3e9 and without any warnings. I'll
> > forward you off-list my .config.
>
> Thanks
> your .config works well
> I check my .config
> Thanks again
>
Please let us know what the differences are. If you have a config option
enabled that causes an -ENOMEM then we need to know what that is, and
what is causing it.
^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: about ap4evb HDMI video
2010-12-20 8:18 about ap4evb HDMI video Kuninori Morimoto
` (2 preceding siblings ...)
2010-12-21 0:36 ` Paul Mundt
@ 2010-12-21 1:32 ` Kuninori Morimoto
2010-12-22 4:42 ` Paul Mundt
4 siblings, 0 replies; 6+ messages in thread
From: Kuninori Morimoto @ 2010-12-21 1:32 UTC (permalink / raw)
To: linux-sh
Dear Paul, Guennadi, Magnus
> Please let us know what the differences are. If you have a config option
> enabled that causes an -ENOMEM then we need to know what that is, and
> what is causing it.
I checked diff my.config guennadi.config
The critical config was
CONFIG_FORCE_MAX_ZONEORDER
guennadi's config was 12
my config was 11
default value seems 11 ?
Best regards
--
Kuninori Morimoto
^ permalink raw reply [flat|nested] 6+ messages in thread* Re: about ap4evb HDMI video
2010-12-20 8:18 about ap4evb HDMI video Kuninori Morimoto
` (3 preceding siblings ...)
2010-12-21 1:32 ` Kuninori Morimoto
@ 2010-12-22 4:42 ` Paul Mundt
4 siblings, 0 replies; 6+ messages in thread
From: Paul Mundt @ 2010-12-22 4:42 UTC (permalink / raw)
To: linux-sh
On Tue, Dec 21, 2010 at 10:32:13AM +0900, Kuninori Morimoto wrote:
> > Please let us know what the differences are. If you have a config option
> > enabled that causes an -ENOMEM then we need to know what that is, and
> > what is causing it.
>
> I checked diff my.config guennadi.config
> The critical config was
>
> CONFIG_FORCE_MAX_ZONEORDER
>
> guennadi's config was 12
> my config was 11
>
> default value seems 11 ?
>
For boards that need to do large contiguous allocations it of course
makes sense to have this number bumped up. Most boards however do not
require this, so we always stick with a conservative default.
What happened here is that you tried to alloc a large contiguous block
that exceeded the max order and got an -ENOMEM, so we don't have any
option but to bump the max order for this case.
^ permalink raw reply [flat|nested] 6+ messages in thread