[U-Boot] v4 patch:SMDK6400: Fix some build bug
Change from V1: patch 1: patch 2: Delete some compile information from commit. patch 3: Add LED modify information and Delete some compile information from commit. patch 4: Add new patch for SDRAM init.
Change from v2: patch2: Modify Makefile for arch/arm/cpu/arm1176/s3c64xx/cpu_init.s in build error. It will generat redefine information for "mem_ctrl_asm_init" before the Modify. patch4: Modify the Sergei Shtylyov comments and change the SDRAM size variable for PHYS_SDRAM_SIZE_1.
Change from v3: Add new patch 3: I discuss with Amicalement, I find the mutiple-link issue: The first, the cpu_init.o have already been link for cmd_link_o_target atfer compile. But, The link script re-link the point file. So the link machine will generate multiple definition error information.
The second, Since the first 4kB of nand boot featue code move to nand_spl, So It is not necessary to force the cpu_init.o in non-nand boot.
Delete the cpu_init.o from u-boot-nand.lds is safe.
patch 4: patch 4 and patch 5 is split from v2 patch4. patch 6: Modify according Minkyu Kang comment
Fix CONFIG_SYS_INIT_SP_ADDR undefined issue.
Signed-off-by: seedshope bocui107@gmail.com --- include/configs/smdk6400.h | 5 +++++ 1 files changed, 5 insertions(+), 0 deletions(-)
diff --git a/include/configs/smdk6400.h b/include/configs/smdk6400.h index 671f2c7..c9acf58 100644 --- a/include/configs/smdk6400.h +++ b/include/configs/smdk6400.h @@ -44,6 +44,11 @@ #define CONFIG_PERIPORT_BASE 0x70000000 #define CONFIG_PERIPORT_SIZE 0x13
+#define CONFIG_SYS_IRAM_BASE 0x0c000000 /* Internal SRAM base address */ +#define CONFIG_SYS_IRAM_SIZE 0x2000 /* 8 KB of internal SRAM memory */ +#define CONFIG_SYS_IRAM_END (CONFIG_SYS_IRAM_BASE + CONFIG_SYS_IRAM_SIZE) +#define CONFIG_SYS_INIT_SP_ADDR (CONFIG_SYS_IRAM_END - GENERATED_GBL_DATA_SIZE) + #define CONFIG_SYS_SDRAM_BASE 0x50000000
/* input clock of PLL: SMDK6400 has 12MHz input clock */
Dear seedshope,
In message 1295624053-8060-2-git-send-email-bocui107@gmail.com you wrote:
Fix CONFIG_SYS_INIT_SP_ADDR undefined issue.
Signed-off-by: seedshope bocui107@gmail.com
I'm sorry, but please consider all your patches rejected unless you provide a Signed-off-by: using your real name (sorry, no pseudonyms or anonymous contributions.)
Best regards,
Wolfgang Denk
On 01/22/2011 04:42 AM, Wolfgang Denk wrote:
Dear seedshope,
In message1295624053-8060-2-git-send-email-bocui107@gmail.com you wrote:
Fix CONFIG_SYS_INIT_SP_ADDR undefined issue.
Signed-off-by: seedshopebocui107@gmail.com
I'm sorry, but please consider all your patches rejected unless you provide a Signed-off-by: using your real name (sorry, no pseudonyms or anonymous contributions.)
ok, This is the first to join the free software for me. So I use my pseudonyms(My English name), I will use the spelling of my name.
Thanks, seedshope
Best regards,
Wolfgang Denk
Modify Makefile for cpu_init.c and Start.s use some label,this defined u-boot.lds of arch/arm/cpu/arm1176. But SMDK6400 use the link script board/samsung/smdk6400/u-boot-nand.lds. So add some label form u-boot.lds to u-boot-nand.lds
Signed-off-by: seedshope bocui107@gmail.com --- board/samsung/smdk6400/u-boot-nand.lds | 27 ++++++++++++++++++++++++--- 1 files changed, 24 insertions(+), 3 deletions(-)
diff --git a/board/samsung/smdk6400/u-boot-nand.lds b/board/samsung/smdk6400/u-boot-nand.lds index 29a4f61..6771981 100644 --- a/board/samsung/smdk6400/u-boot-nand.lds +++ b/board/samsung/smdk6400/u-boot-nand.lds @@ -56,7 +56,28 @@ SECTIONS .mmudata : { *(.mmudata) }
. = ALIGN(4); - __bss_start = .; - .bss : { *(.bss) . = ALIGN(4); } - _end = .; + + .rel.dyn : { + __rel_dyn_start = .; + *(.rel*) + __rel_dyn_end = .; + } + + .dynsym : { + __dynsym_start = .; + *(.dynsym) + } + + .bss __rel_dyn_start (OVERLAY) : { + __bss_start = .; + *(.bss) + . = ALIGN(4); + _end = .; + } + + /DISCARD/ : { *(.dynstr*) } + /DISCARD/ : { *(.dynamic*) } + /DISCARD/ : { *(.plt*) } + /DISCARD/ : { *(.interp*) } + /DISCARD/ : { *(.gnu*) } }
The first, the cpu_init.o have already been link for cmd_link_o_target atfer compile, But, The link script re-link the point file. So the link machine will generate multiple definition error information.
The second, Since the first 4kB of nand boot featue code move to nand_spl, So It is not necessary to force the cpu_init.o in non-nand boot.
Signed-off-by: seedshope bocui107@gmail.com --- board/samsung/smdk6400/u-boot-nand.lds | 1 - 1 files changed, 0 insertions(+), 1 deletions(-)
diff --git a/board/samsung/smdk6400/u-boot-nand.lds b/board/samsung/smdk6400/u-boot-nand.lds index 6771981..6bf4971 100644 --- a/board/samsung/smdk6400/u-boot-nand.lds +++ b/board/samsung/smdk6400/u-boot-nand.lds @@ -35,7 +35,6 @@ SECTIONS .text : { arch/arm/cpu/arm1176/start.o (.text) - arch/arm/cpu/arm1176/s3c64xx/cpu_init.o (.text) *(.text) }
In the nand_spl feature of SMDK6400. Add some relocation symbols to nand_spl/board/samsung/smdk6400/u-boot.lds to fix the compile error.
Signed-off-by: seedshope bocui107@gmail.com --- nand_spl/board/samsung/smdk6400/u-boot.lds | 17 ++++++++++++++++- 1 files changed, 16 insertions(+), 1 deletions(-)
diff --git a/nand_spl/board/samsung/smdk6400/u-boot.lds b/nand_spl/board/samsung/smdk6400/u-boot.lds index 3ac6aa1..30b1573 100644 --- a/nand_spl/board/samsung/smdk6400/u-boot.lds +++ b/nand_spl/board/samsung/smdk6400/u-boot.lds @@ -55,7 +55,22 @@ SECTIONS __u_boot_cmd_end = .;
. = ALIGN(4); + + .rel.dyn : { + __rel_dyn_start = .; + *(.rel*) + __rel_dyn_end = .; + } + + .dynsym : { + __dynsym_start = .; + *(.dynsym) + } + + .bss __rel_dyn_start (OVERLAY) : { __bss_start = .; - .bss : { *(.bss) . = ALIGN(4); } + *(.bss) + . = ALIGN(4); _end = .; + } }
Since nand boot have some limit for the first 4KB, We only disable the LED function to reduce the code space. At the same time, Fix the compile error for LED function undefined in the compile time of nand_spl.
Signed-off-by: seedshope bocui107@gmail.com --- arch/arm/cpu/arm1176/start.S | 2 ++ 1 files changed, 2 insertions(+), 0 deletions(-)
diff --git a/arch/arm/cpu/arm1176/start.S b/arch/arm/cpu/arm1176/start.S index 237dcfe..ae3706a 100644 --- a/arch/arm/cpu/arm1176/start.S +++ b/arch/arm/cpu/arm1176/start.S @@ -354,9 +354,11 @@ clbss_l:str r2, [r0] /* clear loop... */ cmp r0, r1 bne clbss_l
+#ifndef CONFIG_NAND_SPL bl coloured_LED_init bl red_LED_on #endif +#endif
/* * We are done. Do not return, instead branch to second part of board
Since SDRAM init function have already change, So the SDRAM initial function must be change.
Signed-off-by: seedshope bocui107@gmail.com --- board/samsung/smdk6400/smdk6400.c | 10 +++++++++- 1 files changed, 9 insertions(+), 1 deletions(-)
diff --git a/board/samsung/smdk6400/smdk6400.c b/board/samsung/smdk6400/smdk6400.c index 35aa40b..1d03b7a 100644 --- a/board/samsung/smdk6400/smdk6400.c +++ b/board/samsung/smdk6400/smdk6400.c @@ -78,10 +78,18 @@ int board_init(void) return 0; }
-int dram_init(void) +void dram_init_banksize(void) { + DECLARE_GLOBAL_DATA_PTR; + gd->bd->bi_dram[0].start = PHYS_SDRAM_1; gd->bd->bi_dram[0].size = PHYS_SDRAM_1_SIZE; +} + +int dram_init(void) +{ + gd->ram_size = get_ram_size((long *)CONFIG_SYS_SDRAM_BASE, + PHYS_SDRAM_1_SIZE);
return 0; }
Hello.
seedshope wrote:
Since SDRAM init function have already change, So the SDRAM initial function must be change.
This description sounds somewhat tautological...
Signed-off-by: seedshope bocui107@gmail.com
Your real name is required in the signoff.
board/samsung/smdk6400/smdk6400.c | 10 +++++++++- 1 files changed, 9 insertions(+), 1 deletions(-)
diff --git a/board/samsung/smdk6400/smdk6400.c b/board/samsung/smdk6400/smdk6400.c index 35aa40b..1d03b7a 100644 --- a/board/samsung/smdk6400/smdk6400.c +++ b/board/samsung/smdk6400/smdk6400.c @@ -78,10 +78,18 @@ int board_init(void) return 0; }
-int dram_init(void) +void dram_init_banksize(void) {
- DECLARE_GLOBAL_DATA_PTR;
- gd->bd->bi_dram[0].start = PHYS_SDRAM_1; gd->bd->bi_dram[0].size = PHYS_SDRAM_1_SIZE;
+}
+int dram_init(void) +{
- gd->ram_size = get_ram_size((long *)CONFIG_SYS_SDRAM_BASE,
PHYS_SDRAM_1_SIZE);
Could you move this last line more to the right?
WBR, Sergei
On 01/22/2011 01:52 AM, Sergei Shtylyov wrote:
Hello.
seedshope wrote:
Since SDRAM init function have already change, So the SDRAM initial function must be change.
This description sounds somewhat tautological...
If I describe as following: Since SDRAM init function have already change, Modify SDRAM inital function to adapt to it.
How about it?
Signed-off-by: seedshope bocui107@gmail.com
Your real name is required in the signoff.
I use the name for my pen name. It is not problem.
board/samsung/smdk6400/smdk6400.c | 10 +++++++++- 1 files changed, 9 insertions(+), 1 deletions(-)
diff --git a/board/samsung/smdk6400/smdk6400.c b/board/samsung/smdk6400/smdk6400.c index 35aa40b..1d03b7a 100644 --- a/board/samsung/smdk6400/smdk6400.c +++ b/board/samsung/smdk6400/smdk6400.c @@ -78,10 +78,18 @@ int board_init(void) return 0; }
-int dram_init(void) +void dram_init_banksize(void) {
- DECLARE_GLOBAL_DATA_PTR;
- gd->bd->bi_dram[0].start = PHYS_SDRAM_1; gd->bd->bi_dram[0].size = PHYS_SDRAM_1_SIZE;
+}
+int dram_init(void) +{
- gd->ram_size = get_ram_size((long *)CONFIG_SYS_SDRAM_BASE,
PHYS_SDRAM_1_SIZE);Could you move this last line more to the right?
Ya, the orig is ok, But I re-do the patch, It is miss. sorry.
Thanks seedshope
WBR, Sergei
On 01/22/2011 02:05 AM, seedshope wrote:
On 01/22/2011 01:52 AM, Sergei Shtylyov wrote:
Hello.
seedshope wrote:
Since SDRAM init function have already change, So the SDRAM initial function must be change.
This description sounds somewhat tautological...
If I describe as following: Since SDRAM init function have already change, Modify SDRAM inital function to adapt to it.
How about it?
Signed-off-by: seedshope bocui107@gmail.com
Your real name is required in the signoff.
I use the name for my pen name. It is not problem.
board/samsung/smdk6400/smdk6400.c | 10 +++++++++- 1 files changed, 9 insertions(+), 1 deletions(-)
diff --git a/board/samsung/smdk6400/smdk6400.c b/board/samsung/smdk6400/smdk6400.c index 35aa40b..1d03b7a 100644 --- a/board/samsung/smdk6400/smdk6400.c +++ b/board/samsung/smdk6400/smdk6400.c @@ -78,10 +78,18 @@ int board_init(void) return 0; }
-int dram_init(void) +void dram_init_banksize(void) {
- DECLARE_GLOBAL_DATA_PTR;
- gd->bd->bi_dram[0].start = PHYS_SDRAM_1; gd->bd->bi_dram[0].size = PHYS_SDRAM_1_SIZE;
+}
+int dram_init(void) +{
- gd->ram_size = get_ram_size((long *)CONFIG_SYS_SDRAM_BASE,
PHYS_SDRAM_1_SIZE);Could you move this last line more to the right?
Ya, the orig is ok, But I re-do the patch, It is miss. sorry.
Hi Sergei,
I feel this may be you e-mail issue. I open my patch, It is display as following: + gd->ram_size = get_ram_size((long *)CONFIG_SYS_SDRAM_BASE, + PHYS_SDRAM_1_SIZE);
Thanks, seedshope
Thanks seedshope
WBR, Sergei
Le 21/01/2011 19:15, seedshope a écrit :
On 01/22/2011 02:05 AM, seedshope wrote:
On 01/22/2011 01:52 AM, Sergei Shtylyov wrote:
Hello.
seedshope wrote:
Since SDRAM init function have already change, So the SDRAM initial function must be change.
This description sounds somewhat tautological...If I describe as following: Since SDRAM init function have already change, Modify SDRAM inital function to adapt to it.
How about it?
Still unclear, due to the fact you're using the same three terms ("init/initial, RAM, function") for two apparently different things.
Signed-off-by: seedshopebocui107@gmail.com
Your real name is required in the signoff.I use the name for my pen name. It is not problem.
I think Sergei means pen names should not be used. I won't personally pass judgment, but so far I've always seen contributors using their actual names.
I feel this may be you e-mail issue. I open my patch, It is display as following:
gd->ram_size = get_ram_size((long *)CONFIG_SYS_SDRAM_BASE,PHYS_SDRAM_1_SIZE);
Your patch, pulled from patchwork and viewed in vi, has three tabs on that second line, which does not align properly. You should check your code editor settings re: tabs.
Amicalement,
On 01/22/2011 02:29 AM, Albert ARIBAUD wrote:
Le 21/01/2011 19:15, seedshope a écrit :
On 01/22/2011 02:05 AM, seedshope wrote:
On 01/22/2011 01:52 AM, Sergei Shtylyov wrote:
Hello.
seedshope wrote:
Since SDRAM init function have already change, So the SDRAM initial function must be change.
This description sounds somewhat tautological...If I describe as following: Since SDRAM init function have already change, Modify SDRAM inital function to adapt to it.
How about it?
Still unclear, due to the fact you're using the same three terms ("init/initial, RAM, function") for two apparently different things.
Ya, Maybe, But I don't know to describe it.
The patch is only to modify the dram_init() and dram_init_banksize(), Could you help me to describe?
Thank you very much！ seedshope
Signed-off-by: seedshopebocui107@gmail.com
Your real name is required in the signoff.I use the name for my pen name. It is not problem.
I think Sergei means pen names should not be used. I won't personally pass judgment, but so far I've always seen contributors using their actual names.
ok
I feel this may be you e-mail issue. I open my patch, It is display as following:
gd->ram_size = get_ram_size((long *)CONFIG_SYS_SDRAM_BASE,PHYS_SDRAM_1_SIZE);Your patch, pulled from patchwork and viewed in vi, has three tabs on that second line, which does not align properly. You should check your code editor settings re: tabs.
My patch is ok, I just two tabs in my e-mail, But I sent the mail, It is change.
Thanks hongbo
Amicalement,
Le 21/01/2011 19:43, seedshope a écrit :
On 01/22/2011 02:29 AM, Albert ARIBAUD wrote:
Le 21/01/2011 19:15, seedshope a écrit :
On 01/22/2011 02:05 AM, seedshope wrote:
On 01/22/2011 01:52 AM, Sergei Shtylyov wrote:
Hello.
seedshope wrote:
Since SDRAM init function have already change, So the SDRAM initial function must be change.
This description sounds somewhat tautological...If I describe as following: Since SDRAM init function have already change, Modify SDRAM inital function to adapt to it.
How about it?
Still unclear, due to the fact you're using the same three terms ("init/initial, RAM, function") for two apparently different things.
Ya, Maybe, But I don't know to describe it.
The patch is only to modify the dram_init() and dram_init_banksize(), Could you help me to describe?
Thank you very much！ seedshope
The reason for the change to dram_init is not actually about DRAM. If you look up similar patches, you'll find out it is about not being able to access gd->bd because bd does not exist, and this is so since the ELF relocation was introduced. So some good descriptions could be "do not use gd->bd any more" or "fix dram_init for relocation support", for instance.
Signed-off-by: seedshopebocui107@gmail.com
Your real name is required in the signoff.I use the name for my pen name. It is not problem.
I think Sergei means pen names should not be used. I won't personally pass judgment, but so far I've always seen contributors using their actual names.
ok
I feel this may be you e-mail issue. I open my patch, It is display as following:
gd->ram_size = get_ram_size((long *)CONFIG_SYS_SDRAM_BASE,PHYS_SDRAM_1_SIZE);Your patch, pulled from patchwork and viewed in vi, has three tabs on that second line, which does not align properly. You should check your code editor settings re: tabs.
My patch is ok, I just two tabs in my e-mail, But I sent the mail, It is change.
Do you send the patch through git format-patch and git send-email? Many e-mail softwares have weird issues when posting git patches, which is why git has its own tools for sending patches via e-mail.
Thanks hongbo
Amicalement,
On 01/22/2011 03:11 AM, Albert ARIBAUD wrote:
Le 21/01/2011 19:43, seedshope a écrit :
On 01/22/2011 02:29 AM, Albert ARIBAUD wrote:
Le 21/01/2011 19:15, seedshope a écrit :
On 01/22/2011 02:05 AM, seedshope wrote:
On 01/22/2011 01:52 AM, Sergei Shtylyov wrote:
Hello.
seedshope wrote:
> Since SDRAM init function have already change, So the SDRAM > initial function must be change. This description sounds somewhat tautological...
If I describe as following: Since SDRAM init function have already change, Modify SDRAM inital function to adapt to it.
How about it?
Still unclear, due to the fact you're using the same three terms ("init/initial, RAM, function") for two apparently different things.
Ya, Maybe, But I don't know to describe it.
The patch is only to modify the dram_init() and dram_init_banksize(), Could you help me to describe?
Thank you very much！ seedshope
The reason for the change to dram_init is not actually about DRAM. If you look up similar patches, you'll find out it is about not being able to access gd->bd because bd does not exist, and this is so since the ELF relocation was introduced. So some good descriptions could be "do not use gd->bd any more" or "fix dram_init for relocation support", for instance.
ok,
> Signed-off-by: seedshopebocui107@gmail.com Your real name is required in the signoff.
I use the name for my pen name. It is not problem.
I think Sergei means pen names should not be used. I won't personally pass judgment, but so far I've always seen contributors using their actual names.
ok
I feel this may be you e-mail issue. I open my patch, It is display as following:
gd->ram_size = get_ram_size((long *)CONFIG_SYS_SDRAM_BASE,PHYS_SDRAM_1_SIZE);Your patch, pulled from patchwork and viewed in vi, has three tabs on that second line, which does not align properly. You should check your code editor settings re: tabs.
My patch is ok, I just two tabs in my e-mail, But I sent the mail, It is change.
Do you send the patch through git format-patch and git send-email?
Yes, I use the git format-patch and git send-email
Many e-mail softwares have weird issues when posting git patches, which is why git has its own tools for sending patches via e-mail.
ok
Thanks seedshope
Thanks hongbo
Amicalement,
Hi seedshope,
Le 22/01/2011 02:56, seedshope a écrit :
My patch is ok, I just two tabs in my e-mail, But I sent the mail, It is change.
Do you send the patch through git format-patch and git send-email?
Yes, I use the git format-patch and git send-email
Many e-mail softwares have weird issues when posting git patches, which is why git has its own tools for sending patches via e-mail.
ok
Since you're using git format-patch and git send-email, then your original change is not correctly aligned. I suggest you check your code editor's settings on indentation and use of tabulations, notably the "tab size": tabs should align on 8-space multiples; also check that your editor uses a fixed font -- you never know.
Thanks hongbo
Amicalement,
On 01/22/2011 03:31 PM, Albert ARIBAUD wrote:
Hi seedshope,
Le 22/01/2011 02:56, seedshope a écrit :
My patch is ok, I just two tabs in my e-mail, But I sent the mail, It is change.
Do you send the patch through git format-patch and git send-email?
Yes, I use the git format-patch and git send-email
Many e-mail softwares have weird issues when posting git patches, which is why git has its own tools for sending patches via e-mail.
ok
Since you're using git format-patch and git send-email, then your original change is not correctly aligned. I suggest you check your code editor's settings on indentation and use of tabulations, notably the "tab size": tabs should align on 8-space multiples; also check that your editor uses a fixed font -- you never know.
Hi Amicalement
I check my patch 6 on the http://news.gmane.org/gmane.comp.boot-loaders.u-boot, It look fine. I have a bit despondent. Why do you think it has a format problem.
Thanks seedshope
Amicalement,
Hi seedshope,
Le 22/01/2011 20:23, seedshope a écrit :
Hi Amicalement
That's Albert, actually. :)
I check my patch 6 on the http://news.gmane.org/gmane.comp.boot-loaders.u-boot, It look fine. I have a bit despondent. Why do you think it has a format problem.
V5 of your patch has one more tab on as V4 had on the line we're discussion. It is a bit better ; Sergei will tell if that's enough for him.
Thanks seedshope
Amicalement,
On 01/23/2011 04:28 AM, Albert ARIBAUD wrote:
Hi seedshope,
Le 22/01/2011 20:23, seedshope a écrit :
Hi Amicalement
That's Albert, actually. :)
I check my patch 6 on the http://news.gmane.org/gmane.comp.boot-loaders.u-boot, It look fine. I have a bit despondent. Why do you think it has a format problem.
V5 of your patch has one more tab on as V4 had on the line we're discussion. It is a bit better ; Sergei will tell if that's enough for him.
yes, I just found the error in web site. I miss something in my thunderbird. such as tab convert space, So the format is change. Here, I beg you to forgot my miss.
BR seedshope
Thanks seedshope
Amicalement,
Dear seedshope,
In message 4D3B40AC.8090105@gmail.com you wrote:
yes, I just found the error in web site. I miss something in my thunderbird. such as tab convert space, So the format is change. Here, I beg you to forgot my miss.
It is usually helpful to search for and read the available documentation.
See file Documentation/email-clients.txt in your Linux source tree, or: http://git.kernel.org/?p=linux/kernel/git/torvalds/linux-2.6.git;a=blob;f=Do...
Best regards,
Wolfgang Denk
On 01/22/2011 01:52 AM, Sergei Shtylyov wrote:
Hello.
seedshope wrote:
Since SDRAM init function have already change, So the SDRAM initial function must be change.
This description sounds somewhat tautological...
Signed-off-by: seedshope bocui107@gmail.com
Your real name is required in the signoff.
board/samsung/smdk6400/smdk6400.c | 10 +++++++++- 1 files changed, 9 insertions(+), 1 deletions(-)
diff --git a/board/samsung/smdk6400/smdk6400.c b/board/samsung/smdk6400/smdk6400.c index 35aa40b..1d03b7a 100644 --- a/board/samsung/smdk6400/smdk6400.c +++ b/board/samsung/smdk6400/smdk6400.c @@ -78,10 +78,18 @@ int board_init(void) return 0; }
-int dram_init(void) +void dram_init_banksize(void) {
- DECLARE_GLOBAL_DATA_PTR;
gd->bd->bi_dram[0].start = PHYS_SDRAM_1; gd->bd->bi_dram[0].size = PHYS_SDRAM_1_SIZE; +}
+int dram_init(void) +{
- gd->ram_size = get_ram_size((long *)CONFIG_SYS_SDRAM_BASE,
- PHYS_SDRAM_1_SIZE);
You can look at the web site(http://news.gmane.org/gmane.comp.boot-loaders.u-boot) for the patch, It is inconsistent with your description.
Thanks seedshope
Could you move this last line more to the right?
WBR, Sergei
Dear seedshope,
On 22 January 2011 00:34, seedshope bocui107@gmail.com wrote:
Since SDRAM init function have already change, So the SDRAM initial function must be change.
Signed-off-by: seedshope bocui107@gmail.com
board/samsung/smdk6400/smdk6400.c | 10 +++++++++- 1 files changed, 9 insertions(+), 1 deletions(-)
diff --git a/board/samsung/smdk6400/smdk6400.c b/board/samsung/smdk6400/smdk6400.c index 35aa40b..1d03b7a 100644 --- a/board/samsung/smdk6400/smdk6400.c +++ b/board/samsung/smdk6400/smdk6400.c @@ -78,10 +78,18 @@ int board_init(void) return 0; }
-int dram_init(void) +void dram_init_banksize(void) {
- DECLARE_GLOBAL_DATA_PTR;
Please remove it.
gd->bd->bi_dram[0].start = PHYS_SDRAM_1; gd->bd->bi_dram[0].size = PHYS_SDRAM_1_SIZE;
Thanks Minkyu Kang
Hi seedshope,
seedshope <bocui107 <at> gmail.com> writes:
-int dram_init(void) +void dram_init_banksize(void) {
- DECLARE_GLOBAL_DATA_PTR;
This declaration should be done on file scope, not in a function.
- gd->bd->bi_dram[0].start = PHYS_SDRAM_1; gd->bd->bi_dram[0].size = PHYS_SDRAM_1_SIZE;
+}
Best regards, Thomas
participants (6)
-
Albert ARIBAUD -
Minkyu Kang -
seedshope -
Sergei Shtylyov -
Thomas Langer -
Wolfgang Denk