[U-Boot] u-boot spl with fat32 on arm

[U-Boot] u-boot spl with fat32 on arm

[Michal Simek]

Hi guys,

I do play a little bit with u-boot spl and I am curious
about fat32 implementation. There is MAX_CLUSTSIZE which is 65k
and there are 3 buffers in fat.c code
do_fat_read_at_block
get_contents_vfatname_block
get_dentfromdir_block

All these 3 are placed to bss section.
I have just 192k of memory for SPL that's why I tried 2 things
1. Place these 3 fat arrays to different memory which is working fine
2. Decrease MAX_CLUSTSIZE to any low value (I do use 8k).

Unfortunately if I decrease MAX_CLUSTSIZE and add them to bss section
SPL doesn't work. If I add them to a different section in front of
or behind bss everything is just fine.

That's why I have a question where can be the problem in this behaviour
that keeping 8k MAX_CLUSTSIZE in bss section doesn't work
and keeping 8k MAX_CLUSTSIZE out of bss section works.

bss relocation is not done in SPL.
Maybe someone could know where the problem can be before I look
at ASM code.

Thanks for your hints,
Michal

-- 
Michal Simek, Ing. (M.Eng), OpenPGP -> KeyID: FE3D1F91
w: www.monstr.eu p: +42-0-721842854
Maintainer of Linux kernel - Microblaze cpu - http://www.monstr.eu/fdt/
Maintainer of Linux kernel - Xilinx Zynq ARM architecture
Microblaze U-BOOT custodian and responsible for u-boot arm zynq platform


-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 263 bytes
Desc: OpenPGP digital signature
URL: <http://lists.denx.de/pipermail/u-boot/attachments/20140128/fbb1f6f9/attachment.pgp>

[U-Boot] u-boot spl with fat32 on arm

[Tom Rini]

On Tue, Jan 28, 2014 at 03:23:23PM +0100, Michal Simek wrote:

> Hi guys,
> 
> I do play a little bit with u-boot spl and I am curious
> about fat32 implementation. There is MAX_CLUSTSIZE which is 65k
> and there are 3 buffers in fat.c code
> do_fat_read_at_block
> get_contents_vfatname_block
> get_dentfromdir_block
> 
> All these 3 are placed to bss section.
> I have just 192k of memory for SPL that's why I tried 2 things
> 1. Place these 3 fat arrays to different memory which is working fine
> 2. Decrease MAX_CLUSTSIZE to any low value (I do use 8k).
> 
> Unfortunately if I decrease MAX_CLUSTSIZE and add them to bss section
> SPL doesn't work. If I add them to a different section in front of
> or behind bss everything is just fine.
> 
> That's why I have a question where can be the problem in this behaviour
> that keeping 8k MAX_CLUSTSIZE in bss section doesn't work
> and keeping 8k MAX_CLUSTSIZE out of bss section works.
> 
> bss relocation is not done in SPL.

To be clear, are you saying you cannot do this in your case?  We support
CONFIG_SPL_BSS_START_ADDR / CONFIG_SPL_BSS_MAX_SIZE which is how on TI
platforms for example we put BSS into DDR, as we've configured that
before needing BSS.

-- 
Tom
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 836 bytes
Desc: Digital signature
URL: <http://lists.denx.de/pipermail/u-boot/attachments/20140129/171c59a6/attachment.pgp>

[U-Boot] u-boot spl with fat32 on arm

[Michal Simek]

On 01/29/2014 10:57 PM, Tom Rini wrote:
> On Tue, Jan 28, 2014 at 03:23:23PM +0100, Michal Simek wrote:
> 
>> Hi guys,
>>
>> I do play a little bit with u-boot spl and I am curious
>> about fat32 implementation. There is MAX_CLUSTSIZE which is 65k
>> and there are 3 buffers in fat.c code
>> do_fat_read_at_block
>> get_contents_vfatname_block
>> get_dentfromdir_block
>>
>> All these 3 are placed to bss section.
>> I have just 192k of memory for SPL that's why I tried 2 things
>> 1. Place these 3 fat arrays to different memory which is working fine
>> 2. Decrease MAX_CLUSTSIZE to any low value (I do use 8k).
>>
>> Unfortunately if I decrease MAX_CLUSTSIZE and add them to bss section
>> SPL doesn't work. If I add them to a different section in front of
>> or behind bss everything is just fine.
>>
>> That's why I have a question where can be the problem in this behaviour
>> that keeping 8k MAX_CLUSTSIZE in bss section doesn't work
>> and keeping 8k MAX_CLUSTSIZE out of bss section works.
>>
>> bss relocation is not done in SPL.
> 
> To be clear, are you saying you cannot do this in your case?  We support
> CONFIG_SPL_BSS_START_ADDR / CONFIG_SPL_BSS_MAX_SIZE which is how on TI
> platforms for example we put BSS into DDR, as we've configured that
> before needing BSS.

I am not saying that I can't place BSS to the different location out of
my 192k OCM. Moving whole BSS to DDR is possible but I would rather
have it in OCM.
I was talking to Albert on IRC about this and I will debug it
to see what's wrong.

Thanks,
Michal


-- 
Michal Simek, Ing. (M.Eng), OpenPGP -> KeyID: FE3D1F91
w: www.monstr.eu p: +42-0-721842854
Maintainer of Linux kernel - Microblaze cpu - http://www.monstr.eu/fdt/
Maintainer of Linux kernel - Xilinx Zynq ARM architecture
Microblaze U-BOOT custodian and responsible for u-boot arm zynq platform


-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 263 bytes
Desc: OpenPGP digital signature
URL: <http://lists.denx.de/pipermail/u-boot/attachments/20140130/5570466a/attachment.pgp>