[U-Boot] [PATCH 0/3] arm: socfpga: clean up socfpga_common.h
This series cleans up the include/configs/socfpga_common.h file a bit.
It removes some defines that are used nowhere and cleans up some leftovers after various subsystems have been converted to use DM.
Simon Goldschmidt (3): arm: socfpga: move legacy i2c defines to vining arm: socfpga: clean up socfpga_common.h arm: socfpga: remove CONFIG_SYS_BOOTMAPSZ
include/configs/socfpga_common.h | 41 --------------------------- include/configs/socfpga_vining_fpga.h | 24 ++++++++++++++++ 2 files changed, 24 insertions(+), 41 deletions(-)
All socfpga boards exept for vining_fpga use DM_I2C. Move the non-DM i2c defines from socfpga_common.h to socfpga_vining_fpga.h to prevent i2c getting enabled for those boards when DM_I2C is deselected.
Signed-off-by: Simon Goldschmidt simon.k.r.goldschmidt@gmail.com ---
include/configs/socfpga_common.h | 26 -------------------------- include/configs/socfpga_vining_fpga.h | 24 ++++++++++++++++++++++++ 2 files changed, 24 insertions(+), 26 deletions(-)
diff --git a/include/configs/socfpga_common.h b/include/configs/socfpga_common.h index bd8f5c8c41..087ed6e96d 100644 --- a/include/configs/socfpga_common.h +++ b/include/configs/socfpga_common.h @@ -137,32 +137,6 @@ #define CONFIG_SYS_NAND_DATA_BASE SOCFPGA_NANDDATA_ADDRESS #endif
-/* - * I2C support - */ -#ifndef CONFIG_DM_I2C -#define CONFIG_SYS_I2C -#define CONFIG_SYS_I2C_BASE SOCFPGA_I2C0_ADDRESS -#define CONFIG_SYS_I2C_BASE1 SOCFPGA_I2C1_ADDRESS -#define CONFIG_SYS_I2C_BASE2 SOCFPGA_I2C2_ADDRESS -#define CONFIG_SYS_I2C_BASE3 SOCFPGA_I2C3_ADDRESS -/* Using standard mode which the speed up to 100Kb/s */ -#define CONFIG_SYS_I2C_SPEED 100000 -#define CONFIG_SYS_I2C_SPEED1 100000 -#define CONFIG_SYS_I2C_SPEED2 100000 -#define CONFIG_SYS_I2C_SPEED3 100000 -/* Address of device when used as slave */ -#define CONFIG_SYS_I2C_SLAVE 0x02 -#define CONFIG_SYS_I2C_SLAVE1 0x02 -#define CONFIG_SYS_I2C_SLAVE2 0x02 -#define CONFIG_SYS_I2C_SLAVE3 0x02 -#ifndef __ASSEMBLY__ -/* Clock supplied to I2C controller in unit of MHz */ -unsigned int cm_get_l4_sp_clk_hz(void); -#define IC_CLK (cm_get_l4_sp_clk_hz() / 1000000) -#endif -#endif /* CONFIG_DM_I2C */ - /* * QSPI support */ diff --git a/include/configs/socfpga_vining_fpga.h b/include/configs/socfpga_vining_fpga.h index 5517ed722d..2c6fee6ece 100644 --- a/include/configs/socfpga_vining_fpga.h +++ b/include/configs/socfpga_vining_fpga.h @@ -16,6 +16,30 @@ #define CONFIG_LOADADDR 0x01000000 #define CONFIG_SYS_LOAD_ADDR CONFIG_LOADADDR
+/* I2C support */ +#ifndef CONFIG_DM_I2C +#define CONFIG_SYS_I2C +#define CONFIG_SYS_I2C_BASE SOCFPGA_I2C0_ADDRESS +#define CONFIG_SYS_I2C_BASE1 SOCFPGA_I2C1_ADDRESS +#define CONFIG_SYS_I2C_BASE2 SOCFPGA_I2C2_ADDRESS +#define CONFIG_SYS_I2C_BASE3 SOCFPGA_I2C3_ADDRESS +/* Using standard mode which the speed up to 100Kb/s */ +#define CONFIG_SYS_I2C_SPEED 100000 +#define CONFIG_SYS_I2C_SPEED1 100000 +#define CONFIG_SYS_I2C_SPEED2 100000 +#define CONFIG_SYS_I2C_SPEED3 100000 +/* Address of device when used as slave */ +#define CONFIG_SYS_I2C_SLAVE 0x02 +#define CONFIG_SYS_I2C_SLAVE1 0x02 +#define CONFIG_SYS_I2C_SLAVE2 0x02 +#define CONFIG_SYS_I2C_SLAVE3 0x02 +#ifndef __ASSEMBLY__ +/* Clock supplied to I2C controller in unit of MHz */ +unsigned int cm_get_l4_sp_clk_hz(void); +#define IC_CLK (cm_get_l4_sp_clk_hz() / 1000000) +#endif +#endif /* CONFIG_DM_I2C */ + /* I2C EEPROM */ #ifdef CONFIG_CMD_EEPROM #define CONFIG_SYS_I2C_EEPROM_ADDR 0x50
On 1/9/19 8:56 PM, Simon Goldschmidt wrote:
All socfpga boards exept for vining_fpga use DM_I2C. Move the non-DM i2c defines from socfpga_common.h to socfpga_vining_fpga.h to prevent i2c getting enabled for those boards when DM_I2C is deselected.
Can you just flip the board to DM I2C ? I can test the patch .
Signed-off-by: Simon Goldschmidt simon.k.r.goldschmidt@gmail.com
include/configs/socfpga_common.h | 26 -------------------------- include/configs/socfpga_vining_fpga.h | 24 ++++++++++++++++++++++++ 2 files changed, 24 insertions(+), 26 deletions(-)
diff --git a/include/configs/socfpga_common.h b/include/configs/socfpga_common.h index bd8f5c8c41..087ed6e96d 100644 --- a/include/configs/socfpga_common.h +++ b/include/configs/socfpga_common.h @@ -137,32 +137,6 @@ #define CONFIG_SYS_NAND_DATA_BASE SOCFPGA_NANDDATA_ADDRESS #endif
-/*
- I2C support
- */
-#ifndef CONFIG_DM_I2C -#define CONFIG_SYS_I2C -#define CONFIG_SYS_I2C_BASE SOCFPGA_I2C0_ADDRESS -#define CONFIG_SYS_I2C_BASE1 SOCFPGA_I2C1_ADDRESS -#define CONFIG_SYS_I2C_BASE2 SOCFPGA_I2C2_ADDRESS -#define CONFIG_SYS_I2C_BASE3 SOCFPGA_I2C3_ADDRESS -/* Using standard mode which the speed up to 100Kb/s */ -#define CONFIG_SYS_I2C_SPEED 100000 -#define CONFIG_SYS_I2C_SPEED1 100000 -#define CONFIG_SYS_I2C_SPEED2 100000 -#define CONFIG_SYS_I2C_SPEED3 100000 -/* Address of device when used as slave */ -#define CONFIG_SYS_I2C_SLAVE 0x02 -#define CONFIG_SYS_I2C_SLAVE1 0x02 -#define CONFIG_SYS_I2C_SLAVE2 0x02 -#define CONFIG_SYS_I2C_SLAVE3 0x02 -#ifndef __ASSEMBLY__ -/* Clock supplied to I2C controller in unit of MHz */ -unsigned int cm_get_l4_sp_clk_hz(void); -#define IC_CLK (cm_get_l4_sp_clk_hz() / 1000000) -#endif -#endif /* CONFIG_DM_I2C */
/*
- QSPI support
*/ diff --git a/include/configs/socfpga_vining_fpga.h b/include/configs/socfpga_vining_fpga.h index 5517ed722d..2c6fee6ece 100644 --- a/include/configs/socfpga_vining_fpga.h +++ b/include/configs/socfpga_vining_fpga.h @@ -16,6 +16,30 @@ #define CONFIG_LOADADDR 0x01000000 #define CONFIG_SYS_LOAD_ADDR CONFIG_LOADADDR
+/* I2C support */ +#ifndef CONFIG_DM_I2C +#define CONFIG_SYS_I2C +#define CONFIG_SYS_I2C_BASE SOCFPGA_I2C0_ADDRESS +#define CONFIG_SYS_I2C_BASE1 SOCFPGA_I2C1_ADDRESS +#define CONFIG_SYS_I2C_BASE2 SOCFPGA_I2C2_ADDRESS +#define CONFIG_SYS_I2C_BASE3 SOCFPGA_I2C3_ADDRESS +/* Using standard mode which the speed up to 100Kb/s */ +#define CONFIG_SYS_I2C_SPEED 100000 +#define CONFIG_SYS_I2C_SPEED1 100000 +#define CONFIG_SYS_I2C_SPEED2 100000 +#define CONFIG_SYS_I2C_SPEED3 100000 +/* Address of device when used as slave */ +#define CONFIG_SYS_I2C_SLAVE 0x02 +#define CONFIG_SYS_I2C_SLAVE1 0x02 +#define CONFIG_SYS_I2C_SLAVE2 0x02 +#define CONFIG_SYS_I2C_SLAVE3 0x02 +#ifndef __ASSEMBLY__ +/* Clock supplied to I2C controller in unit of MHz */ +unsigned int cm_get_l4_sp_clk_hz(void); +#define IC_CLK (cm_get_l4_sp_clk_hz() / 1000000) +#endif +#endif /* CONFIG_DM_I2C */
/* I2C EEPROM */ #ifdef CONFIG_CMD_EEPROM #define CONFIG_SYS_I2C_EEPROM_ADDR 0x50
Am Mi., 9. Jan. 2019, 22:27 hat Marek Vasut marex@denx.de geschrieben:
On 1/9/19 8:56 PM, Simon Goldschmidt wrote:
All socfpga boards exept for vining_fpga use DM_I2C. Move the non-DM i2c defines from socfpga_common.h to socfpga_vining_fpga.h to prevent i2c getting enabled for those boards when DM_I2C is deselected.
Can you just flip the board to DM I2C ? I can test the patch .
I tried that first and it wouldn't compile. But if you can test it (I can't), I can give it a 2nd look.
Regards, Simon
Signed-off-by: Simon Goldschmidt simon.k.r.goldschmidt@gmail.com
include/configs/socfpga_common.h | 26 -------------------------- include/configs/socfpga_vining_fpga.h | 24 ++++++++++++++++++++++++ 2 files changed, 24 insertions(+), 26 deletions(-)
diff --git a/include/configs/socfpga_common.h
b/include/configs/socfpga_common.h
index bd8f5c8c41..087ed6e96d 100644 --- a/include/configs/socfpga_common.h +++ b/include/configs/socfpga_common.h @@ -137,32 +137,6 @@ #define CONFIG_SYS_NAND_DATA_BASE SOCFPGA_NANDDATA_ADDRESS #endif
-/*
- I2C support
- */
-#ifndef CONFIG_DM_I2C -#define CONFIG_SYS_I2C -#define CONFIG_SYS_I2C_BASE SOCFPGA_I2C0_ADDRESS -#define CONFIG_SYS_I2C_BASE1 SOCFPGA_I2C1_ADDRESS -#define CONFIG_SYS_I2C_BASE2 SOCFPGA_I2C2_ADDRESS -#define CONFIG_SYS_I2C_BASE3 SOCFPGA_I2C3_ADDRESS -/* Using standard mode which the speed up to 100Kb/s */ -#define CONFIG_SYS_I2C_SPEED 100000 -#define CONFIG_SYS_I2C_SPEED1 100000 -#define CONFIG_SYS_I2C_SPEED2 100000 -#define CONFIG_SYS_I2C_SPEED3 100000 -/* Address of device when used as slave */ -#define CONFIG_SYS_I2C_SLAVE 0x02 -#define CONFIG_SYS_I2C_SLAVE1 0x02 -#define CONFIG_SYS_I2C_SLAVE2 0x02 -#define CONFIG_SYS_I2C_SLAVE3 0x02 -#ifndef __ASSEMBLY__ -/* Clock supplied to I2C controller in unit of MHz */ -unsigned int cm_get_l4_sp_clk_hz(void); -#define IC_CLK (cm_get_l4_sp_clk_hz() /
-#endif -#endif /* CONFIG_DM_I2C */
/*
- QSPI support
*/ diff --git a/include/configs/socfpga_vining_fpga.h
b/include/configs/socfpga_vining_fpga.h
index 5517ed722d..2c6fee6ece 100644 --- a/include/configs/socfpga_vining_fpga.h +++ b/include/configs/socfpga_vining_fpga.h @@ -16,6 +16,30 @@ #define CONFIG_LOADADDR 0x01000000 #define CONFIG_SYS_LOAD_ADDR CONFIG_LOADADDR
+/* I2C support */ +#ifndef CONFIG_DM_I2C +#define CONFIG_SYS_I2C +#define CONFIG_SYS_I2C_BASE SOCFPGA_I2C0_ADDRESS +#define CONFIG_SYS_I2C_BASE1 SOCFPGA_I2C1_ADDRESS +#define CONFIG_SYS_I2C_BASE2 SOCFPGA_I2C2_ADDRESS +#define CONFIG_SYS_I2C_BASE3 SOCFPGA_I2C3_ADDRESS +/* Using standard mode which the speed up to 100Kb/s */ +#define CONFIG_SYS_I2C_SPEED 100000 +#define CONFIG_SYS_I2C_SPEED1 100000 +#define CONFIG_SYS_I2C_SPEED2 100000 +#define CONFIG_SYS_I2C_SPEED3 100000 +/* Address of device when used as slave */ +#define CONFIG_SYS_I2C_SLAVE 0x02 +#define CONFIG_SYS_I2C_SLAVE1 0x02 +#define CONFIG_SYS_I2C_SLAVE2 0x02 +#define CONFIG_SYS_I2C_SLAVE3 0x02 +#ifndef __ASSEMBLY__ +/* Clock supplied to I2C controller in unit of MHz */ +unsigned int cm_get_l4_sp_clk_hz(void); +#define IC_CLK (cm_get_l4_sp_clk_hz() /
+#endif +#endif /* CONFIG_DM_I2C */
/* I2C EEPROM */ #ifdef CONFIG_CMD_EEPROM #define CONFIG_SYS_I2C_EEPROM_ADDR 0x50
-- Best regards, Marek Vasut
On 1/9/19 10:31 PM, Simon Goldschmidt wrote:
Am Mi., 9. Jan. 2019, 22:27 hat Marek Vasut <marex@denx.de mailto:marex@denx.de> geschrieben:
On 1/9/19 8:56 PM, Simon Goldschmidt wrote: > All socfpga boards exept for vining_fpga use DM_I2C. Move the non-DM > i2c defines from socfpga_common.h to socfpga_vining_fpga.h to prevent > i2c getting enabled for those boards when DM_I2C is deselected. Can you just flip the board to DM I2C ? I can test the patch .I tried that first and it wouldn't compile. But if you can test it (I can't), I can give it a 2nd look.
Yes please, let's get rid of this ad-hoc non-DM stuff.
Remove outdated defines (not used any more, outdated due to DM conversion) from socfpga_common.h.
Signed-off-by: Simon Goldschmidt simon.k.r.goldschmidt@gmail.com ---
include/configs/socfpga_common.h | 13 ------------- 1 file changed, 13 deletions(-)
diff --git a/include/configs/socfpga_common.h b/include/configs/socfpga_common.h index 087ed6e96d..e9b368d93a 100644 --- a/include/configs/socfpga_common.h +++ b/include/configs/socfpga_common.h @@ -58,10 +58,6 @@ #define CONFIG_SYS_BARGSIZE CONFIG_SYS_CBSIZE /* Boot argument buffer size */
-#ifndef CONFIG_SYS_HOSTNAME -#define CONFIG_SYS_HOSTNAME CONFIG_SYS_BOARD -#endif - /* * Cache */ @@ -150,15 +146,6 @@ unsigned int cm_get_qspi_controller_clk_hz(void); #define CONFIG_CQSPI_REF_CLK cm_get_qspi_controller_clk_hz() #endif
-/* - * Designware SPI support - */ - -/* - * Serial Driver - */ -#define CONFIG_SYS_NS16550_SERIAL - /* * USB */
On 1/9/19 8:56 PM, Simon Goldschmidt wrote:
Remove outdated defines (not used any more, outdated due to DM conversion) from socfpga_common.h.
Signed-off-by: Simon Goldschmidt simon.k.r.goldschmidt@gmail.com
Acked-by: Marek Vasut marex@denx.de
include/configs/socfpga_common.h | 13 ------------- 1 file changed, 13 deletions(-)
diff --git a/include/configs/socfpga_common.h b/include/configs/socfpga_common.h index 087ed6e96d..e9b368d93a 100644 --- a/include/configs/socfpga_common.h +++ b/include/configs/socfpga_common.h @@ -58,10 +58,6 @@ #define CONFIG_SYS_BARGSIZE CONFIG_SYS_CBSIZE /* Boot argument buffer size */
-#ifndef CONFIG_SYS_HOSTNAME -#define CONFIG_SYS_HOSTNAME CONFIG_SYS_BOARD -#endif
/*
- Cache
*/ @@ -150,15 +146,6 @@ unsigned int cm_get_qspi_controller_clk_hz(void); #define CONFIG_CQSPI_REF_CLK cm_get_qspi_controller_clk_hz() #endif
-/*
- Designware SPI support
- */
-/*
- Serial Driver
- */
-#define CONFIG_SYS_NS16550_SERIAL
/*
- USB
*/
On 09.01.19 20:56, Simon Goldschmidt wrote:
Remove outdated defines (not used any more, outdated due to DM conversion) from socfpga_common.h.
Signed-off-by: Simon Goldschmidt simon.k.r.goldschmidt@gmail.com
Reviewed-by: Stefan Roese sr@denx.de
Thanks, Stefan
socfpga_common.h defines CONFIG_SYS_BOOTMAPSZ to 64 MiB.
Since having this define overrides the 'bootm_size' env variable for the whole socfpga platform, let's remove this define from socfpga_common.h and instead rely on the 'bootm_size' env variable (which is initialized to 160 MiB in the same file's default env). This gives users the chance to override it in their own environment.
Signed-off-by: Simon Goldschmidt simon.k.r.goldschmidt@gmail.com ---
include/configs/socfpga_common.h | 2 -- 1 file changed, 2 deletions(-)
diff --git a/include/configs/socfpga_common.h b/include/configs/socfpga_common.h index e9b368d93a..04e0f06230 100644 --- a/include/configs/socfpga_common.h +++ b/include/configs/socfpga_common.h @@ -10,8 +10,6 @@ */ #define CONFIG_CLOCKS
-#define CONFIG_SYS_BOOTMAPSZ (64 * 1024 * 1024) - #define CONFIG_TIMESTAMP /* Print image info with timestamp */
/* add target to build it automatically upon "make" */
On 1/9/19 8:56 PM, Simon Goldschmidt wrote:
socfpga_common.h defines CONFIG_SYS_BOOTMAPSZ to 64 MiB.
Since having this define overrides the 'bootm_size' env variable for the whole socfpga platform, let's remove this define from socfpga_common.h and instead rely on the 'bootm_size' env variable (which is initialized to 160 MiB in the same file's default env). This gives users the chance to override it in their own environment.
Signed-off-by: Simon Goldschmidt simon.k.r.goldschmidt@gmail.com
include/configs/socfpga_common.h | 2 -- 1 file changed, 2 deletions(-)
diff --git a/include/configs/socfpga_common.h b/include/configs/socfpga_common.h index e9b368d93a..04e0f06230 100644 --- a/include/configs/socfpga_common.h +++ b/include/configs/socfpga_common.h @@ -10,8 +10,6 @@ */ #define CONFIG_CLOCKS
-#define CONFIG_SYS_BOOTMAPSZ (64 * 1024 * 1024)
#define CONFIG_TIMESTAMP /* Print image info with timestamp */
/* add target to build it automatically upon "make" */
Can you at least "imply" it to 64 MiB, so we don't change the behavior ?
Am Mi., 9. Jan. 2019, 22:27 hat Marek Vasut marex@denx.de geschrieben:
On 1/9/19 8:56 PM, Simon Goldschmidt wrote:
socfpga_common.h defines CONFIG_SYS_BOOTMAPSZ to 64 MiB.
Since having this define overrides the 'bootm_size' env variable for the whole socfpga platform, let's remove this define from
socfpga_common.h
and instead rely on the 'bootm_size' env variable (which is initialized to 160 MiB in the same file's default env). This gives users the chance to override it in their own environment.
Signed-off-by: Simon Goldschmidt simon.k.r.goldschmidt@gmail.com
include/configs/socfpga_common.h | 2 -- 1 file changed, 2 deletions(-)
diff --git a/include/configs/socfpga_common.h
b/include/configs/socfpga_common.h
index e9b368d93a..04e0f06230 100644 --- a/include/configs/socfpga_common.h +++ b/include/configs/socfpga_common.h @@ -10,8 +10,6 @@ */ #define CONFIG_CLOCKS
-#define CONFIG_SYS_BOOTMAPSZ (64 * 1024 * 1024)
#define CONFIG_TIMESTAMP /* Print image info with timestamp
*/
/* add target to build it automatically upon "make" */
Can you at least "imply" it to 64 MiB, so we don't change the behavior ?
You mean change the "boot_size" value in default environment to 64 MiB? Sure.
Regards, Simon
On 1/9/19 10:30 PM, Simon Goldschmidt wrote:
Am Mi., 9. Jan. 2019, 22:27 hat Marek Vasut <marex@denx.de mailto:marex@denx.de> geschrieben:
On 1/9/19 8:56 PM, Simon Goldschmidt wrote: > socfpga_common.h defines CONFIG_SYS_BOOTMAPSZ to 64 MiB. > > Since having this define overrides the 'bootm_size' env variable for > the whole socfpga platform, let's remove this define from socfpga_common.h > and instead rely on the 'bootm_size' env variable (which is initialized > to 160 MiB in the same file's default env). This gives users the > chance to override it in their own environment. > > Signed-off-by: Simon Goldschmidt <simon.k.r.goldschmidt@gmail.com <mailto:simon.k.r.goldschmidt@gmail.com>> > --- > > include/configs/socfpga_common.h | 2 -- > 1 file changed, 2 deletions(-) > > diff --git a/include/configs/socfpga_common.h b/include/configs/socfpga_common.h > index e9b368d93a..04e0f06230 100644 > --- a/include/configs/socfpga_common.h > +++ b/include/configs/socfpga_common.h > @@ -10,8 +10,6 @@ > */ > #define CONFIG_CLOCKS > > -#define CONFIG_SYS_BOOTMAPSZ (64 * 1024 * 1024) > - > #define CONFIG_TIMESTAMP /* Print image info with timestamp */ > > /* add target to build it automatically upon "make" */ > Can you at least "imply" it to 64 MiB, so we don't change the behavior ?You mean change the "boot_size" value in default environment to 64 MiB? Sure.
Sure, just so we won't change the behavior, some people like to run gigantic kernels.
Am Mi., 9. Jan. 2019, 22:31 hat Marek Vasut marex@denx.de geschrieben:
On 1/9/19 10:30 PM, Simon Goldschmidt wrote:
Am Mi., 9. Jan. 2019, 22:27 hat Marek Vasut <marex@denx.de mailto:marex@denx.de> geschrieben:
On 1/9/19 8:56 PM, Simon Goldschmidt wrote: > socfpga_common.h defines CONFIG_SYS_BOOTMAPSZ to 64 MiB. > > Since having this define overrides the 'bootm_size' env variablefor
> the whole socfpga platform, let's remove this define from socfpga_common.h > and instead rely on the 'bootm_size' env variable (which is initialized > to 160 MiB in the same file's default env). This gives users the > chance to override it in their own environment. > > Signed-off-by: Simon Goldschmidt <simon.k.r.goldschmidt@gmail.com <mailto:simon.k.r.goldschmidt@gmail.com>> > --- > > include/configs/socfpga_common.h | 2 -- > 1 file changed, 2 deletions(-) > > diff --git a/include/configs/socfpga_common.h b/include/configs/socfpga_common.h > index e9b368d93a..04e0f06230 100644 > --- a/include/configs/socfpga_common.h > +++ b/include/configs/socfpga_common.h > @@ -10,8 +10,6 @@ > */ > #define CONFIG_CLOCKS > > -#define CONFIG_SYS_BOOTMAPSZ (64 * 1024 * 1024) > - > #define CONFIG_TIMESTAMP /* Print image info with timestamp */ > > /* add target to build it automatically upon "make" */ > Can you at least "imply" it to 64 MiB, so we don't change thebehavior ?
You mean change the "boot_size" value in default environment to 64 MiB? Sure.
Sure, just so we won't change the behavior, some people like to run gigantic kernels.
Well, 28th this patch, the size should grow from 64 MiB (from the define) to 160 MiB (bootm_size value). I figured that would be ok for everyone... But if you want me to shrink that back to 64 MiB, I'm ok with that either...
Regards, Simon
On 1/9/19 10:34 PM, Simon Goldschmidt wrote:
Am Mi., 9. Jan. 2019, 22:31 hat Marek Vasut <marex@denx.de mailto:marex@denx.de> geschrieben:
On 1/9/19 10:30 PM, Simon Goldschmidt wrote: > > > Am Mi., 9. Jan. 2019, 22:27 hat Marek Vasut <marex@denx.de <mailto:marex@denx.de> > <mailto:marex@denx.de <mailto:marex@denx.de>>> geschrieben: > > On 1/9/19 8:56 PM, Simon Goldschmidt wrote: > > socfpga_common.h defines CONFIG_SYS_BOOTMAPSZ to 64 MiB. > > > > Since having this define overrides the 'bootm_size' env variable for > > the whole socfpga platform, let's remove this define from > socfpga_common.h > > and instead rely on the 'bootm_size' env variable (which is > initialized > > to 160 MiB in the same file's default env). This gives users the > > chance to override it in their own environment. > > > > Signed-off-by: Simon Goldschmidt <simon.k.r.goldschmidt@gmail.com <mailto:simon.k.r.goldschmidt@gmail.com> > <mailto:simon.k.r.goldschmidt@gmail.com <mailto:simon.k.r.goldschmidt@gmail.com>>> > > --- > > > > include/configs/socfpga_common.h | 2 -- > > 1 file changed, 2 deletions(-) > > > > diff --git a/include/configs/socfpga_common.h > b/include/configs/socfpga_common.h > > index e9b368d93a..04e0f06230 100644 > > --- a/include/configs/socfpga_common.h > > +++ b/include/configs/socfpga_common.h > > @@ -10,8 +10,6 @@ > > */ > > #define CONFIG_CLOCKS > > > > -#define CONFIG_SYS_BOOTMAPSZ (64 * 1024 * 1024) > > - > > #define CONFIG_TIMESTAMP /* Print image info with > timestamp */ > > > > /* add target to build it automatically upon "make" */ > > > Can you at least "imply" it to 64 MiB, so we don't change the behavior ? > > > You mean change the "boot_size" value in default environment to 64 MiB? > Sure. Sure, just so we won't change the behavior, some people like to run gigantic kernels.Well, 28th this patch, the size should grow from 64 MiB (from the define) to 160 MiB (bootm_size value). I figured that would be ok for everyone... But if you want me to shrink that back to 64 MiB, I'm ok with that either...
160 is fine too.
Am Mi., 9. Jan. 2019, 22:39 hat Marek Vasut marex@denx.de geschrieben:
On 1/9/19 10:34 PM, Simon Goldschmidt wrote:
Am Mi., 9. Jan. 2019, 22:31 hat Marek Vasut <marex@denx.de mailto:marex@denx.de> geschrieben:
On 1/9/19 10:30 PM, Simon Goldschmidt wrote: > > > Am Mi., 9. Jan. 2019, 22:27 hat Marek Vasut <marex@denx.de <mailto:marex@denx.de> > <mailto:marex@denx.de <mailto:marex@denx.de>>> geschrieben: > > On 1/9/19 8:56 PM, Simon Goldschmidt wrote: > > socfpga_common.h defines CONFIG_SYS_BOOTMAPSZ to 64 MiB. > > > > Since having this define overrides the 'bootm_size' env variable for > > the whole socfpga platform, let's remove this define from > socfpga_common.h > > and instead rely on the 'bootm_size' env variable (which is > initialized > > to 160 MiB in the same file's default env). This gives usersthe
> > chance to override it in their own environment. > > > > Signed-off-by: Simon Goldschmidt <simon.k.r.goldschmidt@gmail.com <mailto:simon.k.r.goldschmidt@gmail.com> > <mailto:simon.k.r.goldschmidt@gmail.com <mailto:simon.k.r.goldschmidt@gmail.com>>> > > --- > > > > include/configs/socfpga_common.h | 2 -- > > 1 file changed, 2 deletions(-) > > > > diff --git a/include/configs/socfpga_common.h > b/include/configs/socfpga_common.h > > index e9b368d93a..04e0f06230 100644 > > --- a/include/configs/socfpga_common.h > > +++ b/include/configs/socfpga_common.h > > @@ -10,8 +10,6 @@ > > */ > > #define CONFIG_CLOCKS > > > > -#define CONFIG_SYS_BOOTMAPSZ (64 * 1024 * 1024) > > - > > #define CONFIG_TIMESTAMP /* Print image infowith
> timestamp */ > > > > /* add target to build it automatically upon "make" */ > > > Can you at least "imply" it to 64 MiB, so we don't change the behavior ? > > > You mean change the "boot_size" value in default environment to 64 MiB? > Sure. Sure, just so we won't change the behavior, some people like to run gigantic kernels.Well, 28th this patch, the size should grow from 64 MiB (from the define) to 160 MiB (bootm_size value). I figured that would be ok for everyone... But if you want me to shrink that back to 64 MiB, I'm ok with that either...
160 is fine too.
Ok, so no V2 for this one. I'll check i2c DM tomorrow.
Regards, Simon
On 1/9/19 10:42 PM, Simon Goldschmidt wrote:
Am Mi., 9. Jan. 2019, 22:39 hat Marek Vasut <marex@denx.de mailto:marex@denx.de> geschrieben:
On 1/9/19 10:34 PM, Simon Goldschmidt wrote: > > > Am Mi., 9. Jan. 2019, 22:31 hat Marek Vasut <marex@denx.de <mailto:marex@denx.de> > <mailto:marex@denx.de <mailto:marex@denx.de>>> geschrieben: > > On 1/9/19 10:30 PM, Simon Goldschmidt wrote: > > > > > > Am Mi., 9. Jan. 2019, 22:27 hat Marek Vasut <marex@denx.de <mailto:marex@denx.de> > <mailto:marex@denx.de <mailto:marex@denx.de>> > > <mailto:marex@denx.de <mailto:marex@denx.de> <mailto:marex@denx.de <mailto:marex@denx.de>>>> geschrieben: > > > > On 1/9/19 8:56 PM, Simon Goldschmidt wrote: > > > socfpga_common.h defines CONFIG_SYS_BOOTMAPSZ to 64 MiB. > > > > > > Since having this define overrides the 'bootm_size' env > variable for > > > the whole socfpga platform, let's remove this define from > > socfpga_common.h > > > and instead rely on the 'bootm_size' env variable (which is > > initialized > > > to 160 MiB in the same file's default env). This gives users the > > > chance to override it in their own environment. > > > > > > Signed-off-by: Simon Goldschmidt > <simon.k.r.goldschmidt@gmail.com <mailto:simon.k.r.goldschmidt@gmail.com> > <mailto:simon.k.r.goldschmidt@gmail.com <mailto:simon.k.r.goldschmidt@gmail.com>> > > <mailto:simon.k.r.goldschmidt@gmail.com <mailto:simon.k.r.goldschmidt@gmail.com> > <mailto:simon.k.r.goldschmidt@gmail.com <mailto:simon.k.r.goldschmidt@gmail.com>>>> > > > --- > > > > > > include/configs/socfpga_common.h | 2 -- > > > 1 file changed, 2 deletions(-) > > > > > > diff --git a/include/configs/socfpga_common.h > > b/include/configs/socfpga_common.h > > > index e9b368d93a..04e0f06230 100644 > > > --- a/include/configs/socfpga_common.h > > > +++ b/include/configs/socfpga_common.h > > > @@ -10,8 +10,6 @@ > > > */ > > > #define CONFIG_CLOCKS > > > > > > -#define CONFIG_SYS_BOOTMAPSZ (64 * 1024 * 1024) > > > - > > > #define CONFIG_TIMESTAMP /* Print image info with > > timestamp */ > > > > > > /* add target to build it automatically upon "make" */ > > > > > Can you at least "imply" it to 64 MiB, so we don't change the > behavior ? > > > > > > You mean change the "boot_size" value in default environment to 64 > MiB? > > Sure. > > Sure, just so we won't change the behavior, some people like to run > gigantic kernels. > > > Well, 28th this patch, the size should grow from 64 MiB (from the > define) to 160 MiB (bootm_size value). I figured that would be ok for > everyone... But if you want me to shrink that back to 64 MiB, I'm ok > with that either... 160 is fine too.Ok, so no V2 for this one. I'll check i2c DM tomorrow.
Thanks, appreciated.
On Wed, Jan 9, 2019 at 10:44 PM Marek Vasut marex@denx.de wrote:
On 1/9/19 10:42 PM, Simon Goldschmidt wrote:
Am Mi., 9. Jan. 2019, 22:39 hat Marek Vasut <marex@denx.de mailto:marex@denx.de> geschrieben:
On 1/9/19 10:34 PM, Simon Goldschmidt wrote: > > > Am Mi., 9. Jan. 2019, 22:31 hat Marek Vasut <marex@denx.de <mailto:marex@denx.de> > <mailto:marex@denx.de <mailto:marex@denx.de>>> geschrieben: > > On 1/9/19 10:30 PM, Simon Goldschmidt wrote: > > > > > > Am Mi., 9. Jan. 2019, 22:27 hat Marek Vasut <marex@denx.de <mailto:marex@denx.de> > <mailto:marex@denx.de <mailto:marex@denx.de>> > > <mailto:marex@denx.de <mailto:marex@denx.de> <mailto:marex@denx.de <mailto:marex@denx.de>>>> geschrieben: > > > > On 1/9/19 8:56 PM, Simon Goldschmidt wrote: > > > socfpga_common.h defines CONFIG_SYS_BOOTMAPSZ to 64 MiB. > > > > > > Since having this define overrides the 'bootm_size' env > variable for > > > the whole socfpga platform, let's remove this define from > > socfpga_common.h > > > and instead rely on the 'bootm_size' env variable (which is > > initialized > > > to 160 MiB in the same file's default env). This gives users the > > > chance to override it in their own environment. > > > > > > Signed-off-by: Simon Goldschmidt > <simon.k.r.goldschmidt@gmail.com <mailto:simon.k.r.goldschmidt@gmail.com> > <mailto:simon.k.r.goldschmidt@gmail.com <mailto:simon.k.r.goldschmidt@gmail.com>> > > <mailto:simon.k.r.goldschmidt@gmail.com <mailto:simon.k.r.goldschmidt@gmail.com> > <mailto:simon.k.r.goldschmidt@gmail.com <mailto:simon.k.r.goldschmidt@gmail.com>>>> > > > --- > > > > > > include/configs/socfpga_common.h | 2 -- > > > 1 file changed, 2 deletions(-) > > > > > > diff --git a/include/configs/socfpga_common.h > > b/include/configs/socfpga_common.h > > > index e9b368d93a..04e0f06230 100644 > > > --- a/include/configs/socfpga_common.h > > > +++ b/include/configs/socfpga_common.h > > > @@ -10,8 +10,6 @@ > > > */ > > > #define CONFIG_CLOCKS > > > > > > -#define CONFIG_SYS_BOOTMAPSZ (64 * 1024 * 1024) > > > - > > > #define CONFIG_TIMESTAMP /* Print image info with > > timestamp */ > > > > > > /* add target to build it automatically upon "make" */ > > > > > Can you at least "imply" it to 64 MiB, so we don't change the > behavior ? > > > > > > You mean change the "boot_size" value in default environment to 64 > MiB? > > Sure. > > Sure, just so we won't change the behavior, some people like to run > gigantic kernels. > > > Well, 28th this patch, the size should grow from 64 MiB (from the > define) to 160 MiB (bootm_size value). I figured that would be ok for > everyone... But if you want me to shrink that back to 64 MiB, I'm ok > with that either... 160 is fine too.Ok, so no V2 for this one. I'll check i2c DM tomorrow.
Thanks, appreciated.
I had a quick look, but I2C eemprom access has change too much from legacy to DM for me to just dry-code that.
Hoever, I just saw the PHY1 extension board for my SoCrates has an I2C eeprom, too. So I'll first try to get that running before just converting the Vining. But this will probably need more time.
Regards, Simon
On 1/10/19 9:19 AM, Simon Goldschmidt wrote:
On Wed, Jan 9, 2019 at 10:44 PM Marek Vasut marex@denx.de wrote:
On 1/9/19 10:42 PM, Simon Goldschmidt wrote:
Am Mi., 9. Jan. 2019, 22:39 hat Marek Vasut <marex@denx.de mailto:marex@denx.de> geschrieben:
On 1/9/19 10:34 PM, Simon Goldschmidt wrote: > > > Am Mi., 9. Jan. 2019, 22:31 hat Marek Vasut <marex@denx.de <mailto:marex@denx.de> > <mailto:marex@denx.de <mailto:marex@denx.de>>> geschrieben: > > On 1/9/19 10:30 PM, Simon Goldschmidt wrote: > > > > > > Am Mi., 9. Jan. 2019, 22:27 hat Marek Vasut <marex@denx.de <mailto:marex@denx.de> > <mailto:marex@denx.de <mailto:marex@denx.de>> > > <mailto:marex@denx.de <mailto:marex@denx.de> <mailto:marex@denx.de <mailto:marex@denx.de>>>> geschrieben: > > > > On 1/9/19 8:56 PM, Simon Goldschmidt wrote: > > > socfpga_common.h defines CONFIG_SYS_BOOTMAPSZ to 64 MiB. > > > > > > Since having this define overrides the 'bootm_size' env > variable for > > > the whole socfpga platform, let's remove this define from > > socfpga_common.h > > > and instead rely on the 'bootm_size' env variable (which is > > initialized > > > to 160 MiB in the same file's default env). This gives users the > > > chance to override it in their own environment. > > > > > > Signed-off-by: Simon Goldschmidt > <simon.k.r.goldschmidt@gmail.com <mailto:simon.k.r.goldschmidt@gmail.com> > <mailto:simon.k.r.goldschmidt@gmail.com <mailto:simon.k.r.goldschmidt@gmail.com>> > > <mailto:simon.k.r.goldschmidt@gmail.com <mailto:simon.k.r.goldschmidt@gmail.com> > <mailto:simon.k.r.goldschmidt@gmail.com <mailto:simon.k.r.goldschmidt@gmail.com>>>> > > > --- > > > > > > include/configs/socfpga_common.h | 2 -- > > > 1 file changed, 2 deletions(-) > > > > > > diff --git a/include/configs/socfpga_common.h > > b/include/configs/socfpga_common.h > > > index e9b368d93a..04e0f06230 100644 > > > --- a/include/configs/socfpga_common.h > > > +++ b/include/configs/socfpga_common.h > > > @@ -10,8 +10,6 @@ > > > */ > > > #define CONFIG_CLOCKS > > > > > > -#define CONFIG_SYS_BOOTMAPSZ (64 * 1024 * 1024) > > > - > > > #define CONFIG_TIMESTAMP /* Print image info with > > timestamp */ > > > > > > /* add target to build it automatically upon "make" */ > > > > > Can you at least "imply" it to 64 MiB, so we don't change the > behavior ? > > > > > > You mean change the "boot_size" value in default environment to 64 > MiB? > > Sure. > > Sure, just so we won't change the behavior, some people like to run > gigantic kernels. > > > Well, 28th this patch, the size should grow from 64 MiB (from the > define) to 160 MiB (bootm_size value). I figured that would be ok for > everyone... But if you want me to shrink that back to 64 MiB, I'm ok > with that either... 160 is fine too.Ok, so no V2 for this one. I'll check i2c DM tomorrow.
Thanks, appreciated.
I had a quick look, but I2C eemprom access has change too much from legacy to DM for me to just dry-code that.
Hoever, I just saw the PHY1 extension board for my SoCrates has an I2C eeprom, too. So I'll first try to get that running before just converting the Vining. But this will probably need more time.
The MW opens next week, so that's fine .
On 10.01.2019 09:38, Marek Vasut wrote:
On 1/10/19 9:19 AM, Simon Goldschmidt wrote:
On Wed, Jan 9, 2019 at 10:44 PM Marek Vasut marex@denx.de wrote:
On 1/9/19 10:42 PM, Simon Goldschmidt wrote:
Am Mi., 9. Jan. 2019, 22:39 hat Marek Vasut <marex@denx.de mailto:marex@denx.de> geschrieben:
On 1/9/19 10:34 PM, Simon Goldschmidt wrote: > > > Am Mi., 9. Jan. 2019, 22:31 hat Marek Vasut <marex@denx.de <mailto:marex@denx.de> > <mailto:marex@denx.de <mailto:marex@denx.de>>> geschrieben: > > On 1/9/19 10:30 PM, Simon Goldschmidt wrote: > > > > > > Am Mi., 9. Jan. 2019, 22:27 hat Marek Vasut <marex@denx.de <mailto:marex@denx.de> > <mailto:marex@denx.de <mailto:marex@denx.de>> > > <mailto:marex@denx.de <mailto:marex@denx.de> <mailto:marex@denx.de <mailto:marex@denx.de>>>> geschrieben: > > > > On 1/9/19 8:56 PM, Simon Goldschmidt wrote: > > > socfpga_common.h defines CONFIG_SYS_BOOTMAPSZ to 64 MiB. > > > > > > Since having this define overrides the 'bootm_size' env > variable for > > > the whole socfpga platform, let's remove this define from > > socfpga_common.h > > > and instead rely on the 'bootm_size' env variable (which is > > initialized > > > to 160 MiB in the same file's default env). This gives users the > > > chance to override it in their own environment. > > > > > > Signed-off-by: Simon Goldschmidt > <simon.k.r.goldschmidt@gmail.com <mailto:simon.k.r.goldschmidt@gmail.com> > <mailto:simon.k.r.goldschmidt@gmail.com <mailto:simon.k.r.goldschmidt@gmail.com>> > > <mailto:simon.k.r.goldschmidt@gmail.com <mailto:simon.k.r.goldschmidt@gmail.com> > <mailto:simon.k.r.goldschmidt@gmail.com <mailto:simon.k.r.goldschmidt@gmail.com>>>> > > > --- > > > > > > include/configs/socfpga_common.h | 2 -- > > > 1 file changed, 2 deletions(-) > > > > > > diff --git a/include/configs/socfpga_common.h > > b/include/configs/socfpga_common.h > > > index e9b368d93a..04e0f06230 100644 > > > --- a/include/configs/socfpga_common.h > > > +++ b/include/configs/socfpga_common.h > > > @@ -10,8 +10,6 @@ > > > */ > > > #define CONFIG_CLOCKS > > > > > > -#define CONFIG_SYS_BOOTMAPSZ (64 * 1024 * 1024) > > > - > > > #define CONFIG_TIMESTAMP /* Print image info with > > timestamp */ > > > > > > /* add target to build it automatically upon "make" */ > > > > > Can you at least "imply" it to 64 MiB, so we don't change the > behavior ? > > > > > > You mean change the "boot_size" value in default environment to 64 > MiB? > > Sure. > > Sure, just so we won't change the behavior, some people like to run > gigantic kernels. > > > Well, 28th this patch, the size should grow from 64 MiB (from the > define) to 160 MiB (bootm_size value). I figured that would be ok for > everyone... But if you want me to shrink that back to 64 MiB, I'm ok > with that either... 160 is fine too.Ok, so no V2 for this one. I'll check i2c DM tomorrow.
Thanks, appreciated.
I had a quick look, but I2C eemprom access has change too much from legacy to DM for me to just dry-code that.
Hoever, I just saw the PHY1 extension board for my SoCrates has an I2C eeprom, too. So I'll first try to get that running before just converting the Vining. But this will probably need more time.
The MW opens next week, so that's fine .
I failed to find the time to do this. It did not "just work" and I could not get it to run in the short time I had for this. Could we still merge 2/3 and 3/3 for this release? Or even leave the vining I2C cleanup to that board's maintainer (don't know who that is actually) and apply 1/3 too?
Regards, Simon
On 3/1/19 8:38 PM, Simon Goldschmidt wrote:
On 10.01.2019 09:38, Marek Vasut wrote:
On 1/10/19 9:19 AM, Simon Goldschmidt wrote:
On Wed, Jan 9, 2019 at 10:44 PM Marek Vasut marex@denx.de wrote:
On 1/9/19 10:42 PM, Simon Goldschmidt wrote:
Am Mi., 9. Jan. 2019, 22:39 hat Marek Vasut <marex@denx.de mailto:marex@denx.de> geschrieben:
On 1/9/19 10:34 PM, Simon Goldschmidt wrote: > > > Am Mi., 9. Jan. 2019, 22:31 hat Marek Vasut <marex@denx.de mailto:marex@denx.de > <mailto:marex@denx.de mailto:marex@denx.de>> geschrieben: > > On 1/9/19 10:30 PM, Simon Goldschmidt wrote: > > > > > > Am Mi., 9. Jan. 2019, 22:27 hat Marek Vasut <marex@denx.de mailto:marex@denx.de > <mailto:marex@denx.de mailto:marex@denx.de> > > <mailto:marex@denx.de mailto:marex@denx.de <mailto:marex@denx.de mailto:marex@denx.de>>> geschrieben: > > > > On 1/9/19 8:56 PM, Simon Goldschmidt wrote: > > > socfpga_common.h defines CONFIG_SYS_BOOTMAPSZ to 64 MiB. > > > > > > Since having this define overrides the 'bootm_size' env > variable for > > > the whole socfpga platform, let's remove this define from > > socfpga_common.h > > > and instead rely on the 'bootm_size' env variable (which is > > initialized > > > to 160 MiB in the same file's default env). This gives users the > > > chance to override it in their own environment. > > > > > > Signed-off-by: Simon Goldschmidt > <simon.k.r.goldschmidt@gmail.com mailto:simon.k.r.goldschmidt@gmail.com > <mailto:simon.k.r.goldschmidt@gmail.com mailto:simon.k.r.goldschmidt@gmail.com> > > <mailto:simon.k.r.goldschmidt@gmail.com mailto:simon.k.r.goldschmidt@gmail.com > <mailto:simon.k.r.goldschmidt@gmail.com mailto:simon.k.r.goldschmidt@gmail.com>>> > > > --- > > > > > > include/configs/socfpga_common.h | 2 -- > > > 1 file changed, 2 deletions(-) > > > > > > diff --git a/include/configs/socfpga_common.h > > b/include/configs/socfpga_common.h > > > index e9b368d93a..04e0f06230 100644 > > > --- a/include/configs/socfpga_common.h > > > +++ b/include/configs/socfpga_common.h > > > @@ -10,8 +10,6 @@ > > > */ > > > #define CONFIG_CLOCKS > > > > > > -#define CONFIG_SYS_BOOTMAPSZ (64 * 1024
> > > - > > > #define CONFIG_TIMESTAMP /* Print image info with > > timestamp */ > > > > > > /* add target to build it automatically upon "make" */ > > > > > Can you at least "imply" it to 64 MiB, so we don't change the > behavior ? > > > > > > You mean change the "boot_size" value in default environment to 64 > MiB? > > Sure. > > Sure, just so we won't change the behavior, some people like to run > gigantic kernels. > > > Well, 28th this patch, the size should grow from 64 MiB (from the > define) to 160 MiB (bootm_size value). I figured that would be ok for > everyone... But if you want me to shrink that back to 64 MiB, I'm ok > with that either...
160 is fine too.
Ok, so no V2 for this one. I'll check i2c DM tomorrow.
Thanks, appreciated.
I had a quick look, but I2C eemprom access has change too much from legacy to DM for me to just dry-code that.
Hoever, I just saw the PHY1 extension board for my SoCrates has an I2C eeprom, too. So I'll first try to get that running before just converting the Vining. But this will probably need more time.
The MW opens next week, so that's fine .
I failed to find the time to do this. It did not "just work" and I could not get it to run in the short time I had for this. Could we still merge 2/3 and 3/3 for this release? Or even leave the vining I2C cleanup to that board's maintainer (don't know who that is actually) and apply 1/3 too?
Can you rebase/repost those patches , so I can take a look ? Unless they are fixes, those will be for next.
On 09.01.19 20:56, Simon Goldschmidt wrote:
socfpga_common.h defines CONFIG_SYS_BOOTMAPSZ to 64 MiB.
Since having this define overrides the 'bootm_size' env variable for the whole socfpga platform, let's remove this define from socfpga_common.h and instead rely on the 'bootm_size' env variable (which is initialized to 160 MiB in the same file's default env). This gives users the chance to override it in their own environment.
Signed-off-by: Simon Goldschmidt simon.k.r.goldschmidt@gmail.com
Reviewed-by: Stefan Roese sr@denx.de
Thanks, Stefan
participants (3)
-
Marek Vasut -
Simon Goldschmidt -
Stefan Roese