FreeScale MC68360 support
PATCH 1/1 FreeScale MC68360 support
Good day!
This patch adds support for MC68360 based board for u-boot-1.1.6. Mail me if you have any questions :)
Best regards,
Matvejchikov Ilya.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.denx.de/pipermail/u-boot/attachments/20070408/3f74b315/attachment.htm
-------------- next part --------------
A non-text attachment was scrubbed...
Name: u-boot-1.1.6-mc68360-i0.patch.gz
Type: application/x-gzip
Size: 28186 bytes
Desc: not available
Url : http://lists.denx.de/pipermail/u-boot/attachments/20070408/3f74b315/attachment.bin
PATCH 1/1 FreeScale MC68360 support
Dear Ilya,
Quoted 3 lines · in message <[email protected]> you…
This patch adds support for MC68360 based board for u-boot-1.1.6. Mail me if you have any questions :)
Please correct me if I'm wrong, but MC68360 is not the name of a board, but of the Freescale (resp. ex Motorola) QICC CPU. There are many different boards using this processor, so you will have to chose a more descriptive board name.
Also, U-Boot 1.1.6 is outdated. Please submit your patch against a recent version of U-Boot (i. e. top of tree in git repository, or at least release 1.2.0).
Then, there are some coding style issues with your patch (indentation not by TAB, indentation not by multiple of 8 columns, trailing white space, C++ comments, too long lines, etc.) as well as other formal issues (missing Copyright entries, missing license headers, missing
Signed-off-by: line).
It seems you use a private flash driver for a device that looks as if it was CFI conformant. Please explain why you cannot use the CFI driver instead.
I also wonder if you really need a new ethernet driver
(cpu/mc68360/enet.c) and a new serial driver (cpu/mc68360/serial.c) - IIRC the CPM on the 360 is mostly compatible (with very few minor deviations that can be handled easily) to the CPM on the PowerQICC I (= MPC8xx) processors. I think most of this could could (and should) be shared? The same holds for some other files like for example include/asm-m68k/arch-mc68360/commproc.h which is highly redundand with the MPC8xx commproc.h
Some of your files contain version ID stuff like here:
include/asm-m68k/arch-mc68360/mc68360_enet.h:
*********************************** * $Id: m68360_enet.h,v 1.1.1.1 2001/05/18 17:10:11 hamilton Exp $ ***********************************
Please remove this.
I see some "#if 0" blocks in your code; please remove these.
I see:
#define CONFIG_BOOTFILE vmlinuz
This looks broken to me. U-Boot does not boot a vmlinuz file, but U-Boot images created by mkimage.
#define CFG_BAUDRATE_TABLE { 19200 }
This is *very* restrictive. I recommend to support all commonly used baudrates instead (i. e. at least the range from 9600 through 115200).
#define CFG_HZ (unsigned long int)32768000
This is broken. CFG_HZ is required to be 1000 (i. e. millisecond ticks).
#define CFG_SDRAM_SIZE 0x00800000
Please don't do that. U-Boot style is to allow auto-adjustment to the actual RAM size.
#define CFG_MALLOC_LEN (5*1024*1024)
You have 8 MB of RAM and reserve 5 MB for malloc()? This seems broken to me.
You comment out relevant parts of lib_m68k/board.c - you must not do that!
lib_m68k/time.c - your "implementation" of udelay() is broken. Please replace by a working version.
Please clean up these issues and resubmit.
Thanks.
Best regards,
Wolfgang Denk
Signature · 4 lines
DENX Software Engineering GmbH, HRB 165235 Munich, CEO: Wolfgang Denk Office: Kirchenstr. 5, D-82194 Groebenzell, Germany Phone: (+49)-8142-66989-10 Fax: (+49)-8142-66989-80 Email: wd at denx.de Include the success of others in your dreams for your own success.
PATCH 1/1 FreeScale MC68360 support
2007/4/9, Wolfgang Denk <[email protected]>:
Quoted 6 lines
Dear Ilya,
in message <[email protected]>
Quoted 1 line · you
Quoted 8 lines
Quoted 1 line
This patch adds support for MC68360 based board for u-boot-1.1.6. Mail
me if
Quoted 1 line
you have any questions :)
Please correct me if I'm wrong, but MC68360 is not the name of a board, but of the Freescale (resp. ex Motorola) QICC CPU. There are many different boards using this processor, so you will have to chose a more descriptive board name.
Yes, you are right. MC68360 is the name of the CPU. But the board I'm working
at right now is my own design. That's why I can't do it.
Also, U-Boot 1.1.6 is outdated. Please submit your patch against a
Quoted 2 lines
recent version of U-Boot (i. e. top of tree in git repository, or at least release 1.2.0).
Ooops, I was not aware of it. As far as I know, there was no official information about it.
Then, there are some coding style issues with your patch (indentation
Quoted 4 lines
not by TAB, indentation not by multiple of 8 columns, trailing white space, C++ comments, too long lines, etc.) as well as other formal issues (missing Copyright entries, missing license headers, missing
Signed-off-by: line).
Ok. Could you tell me where I can get exact information about it?
It seems you use a private flash driver for a device that looks as if
Quoted 2 lines
it was CFI conformant. Please explain why you cannot use the CFI driver instead.
This is because I have old non-CFI flash chips :(
I also wonder if you really need a new ethernet driver
Quoted 7 lines
(
cpu/mc68360/enet.c) and a new serial driver (cpu/mc68360/serial.c) - IIRC the CPM on the 360 is mostly compatible (with very few minor deviations that can be handled easily) to the CPM on the PowerQICC I (= MPC8xx) processors. I think most of this could could (and should) be shared? The same holds for some other files like for exampleinclude/asm-m68k/arch-mc68360/commproc.hwhich is highly redundand with the MPC8xxcommproc.h
I don't know this processors family well enough, but I see what I can do with it....
Some of your files contain version ID stuff like here:
Quoted 8 lines
include/asm-m68k/arch-mc68360/mc68360_enet.h:*********************************** * $Id: m68360_enet.h,v 1.1.1.1 2001/05/18 17:10:11 hamilton Exp $ ***********************************Please remove this.
ok
I see some "#if 0" blocks in your code; please remove these.
ok
I see:
Quoted 5 lines
#define CONFIG_BOOTFILE vmlinuzThis looks broken to me. U-Boot does not boot a vmlinuz file, but U-Boot images created by mkimage.
Yes, this name is incorrect... Let it be uImage.
#define CFG_BAUDRATE_TABLE { 19200 }
Quoted 4 lines
This is *very* restrictive. I recommend to support all commonly used baudrates instead (i. e. at least the range from 9600 through 115200).
ok
#define CFG_HZ (unsigned long int)32768000
Quoted 3 lines
This is broken.
CFG_HZis required to be 1000 (i. e. millisecond ticks).
okkk..
#define CFG_SDRAM_SIZE 0x00800000
Quoted 9 lines
Please don't do that. U-Boot style is to allow auto-adjustment to the actual RAM size.
#define
CFG_MALLOC_LEN(5*1024*1024)You have 8 MB of RAM and reserve 5 MB for
malloc()? This seems broken to me.
Why not? I really have it working :)
You comment out relevant parts of lib_m68k/board.c - you must not do
Quoted 1 line
that!
Could you tell me the reason why I should not do it?
lib_m68k/time.c - your "implementation" of udelay() is broken. Please
Quoted 1 line
replace by a working version.
I'll try.
Please clean up these issues and resubmit.
Quoted 2 lines
I'm grateful to you for your help and for what you are doing :)
I will take into consideration your advice and recommendations. And I'll try to correct my patch.
Best regards,
Matvejchikov Ilya.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.denx.de/pipermail/u-boot/attachments/20070411/2ac7b8c7/attachment.htm
PATCH 1/1 FreeScale MC68360 support
Quoted 7 lines · In message <[email protected]> you…
Quoted 3 lines
many different boards using this processor, so you will have to chose a more descriptive board name.
Yes, you are right. MC68360 is the name of the CPU. But the board I'm working
at right now is my own design. That's why I can't do it.
Can't do it? You cannot give your own board a name? Why not?!?
Quoted 4 lines
Quoted 1 line
least release 1.2.0).
Ooops, I was not aware of it. As far as I know, there was no official information about it.
Yes, there was.
Quoted 7 lines
Then, there are some coding style issues with your patch (indentation
Quoted 4 lines
not by TAB, indentation not by multiple of 8 columns, trailing white space, C++ comments, too long lines, etc.) as well as other formal issues (missing Copyright entries, missing license headers, missing
Signed-off-by: line).Ok. Could you tell me where I can get exact information about it?
Start at http://www.denx.de/wiki/UBoot, and make sure to read at lkeast the README.
Quoted 11 lines
#define
CFG_SDRAM_SIZE0x00800000Quoted 8 lines
Please don't do that. U-Boot style is to allow auto-adjustment to the actual RAM size.
#define
CFG_MALLOC_LEN(5*1024*1024)You have 8 MB of RAM and reserve 5 MB for
malloc()? This seems broken to me.Why not? I really have it working :)
You cannot load a Linux kernel and a ramdisk on such a system, while there are still ~5 MB of malloc space unused.
Quoted 4 lines
You comment out relevant parts of
lib_m68k/board.c- you must not doQuoted 1 line
that!
Could you tell me the reason why I should not do it?
Because lib_m68k/board.c is common code which you share with all other m68k boards. You must not mess with this.
Quoted 2 lines
I will take into consideration your advice and recommendations. And I'll try to correct my patch.
Thanks.
Quoted 4 lines
------=_Part_4889_10802702.1176321492201
Content-Type: text/html; charset=ISO-8859-1
Content-Transfer-Encoding: 7bit
Content-Disposition: inline
But please NEVER post HTML here again!
Best regards,
Wolfgang Denk
Signature · 5 lines
DENX Software Engineering GmbH, HRB 165235 Munich, CEO: Wolfgang Denk Office: Kirchenstr. 5, D-82194 Groebenzell, Germany Phone: (+49)-8142-66989-10 Fax: (+49)-8142-66989-80 Email: wd at denx.de The idea of male and female are universal constants. -- Kirk, "Metamorphosis", stardate 3219.8
PATCH 1/1 FreeScale MC68360 support
In reply to Wolfgang DenkWas the MC68360 patch ever released?
Signature · 2 lines
View this message in context: http://www.nabble.com/-PATCH--FreeScale-MC68360-support-tf3543268.html#a10288952 Sent from the Uboot - Users mailing list archive at Nabble.com.