Kconfig: Introduce CONFIG_SYS_HAS_SRAM
Subsystems (5)
PATCH v4 1/17 Kconfig: Introduce CONFIG_SYS_HAS_SRAM
In order to be able to replace "#ifdef CONFIG_SYS_SRAM_BASE" sequences with the IS_ENABLED() equivalent, introduce a new boolean Kconfig option that signals whether the platform has SRAM support.
---
Kconfig | 11 +++++++++++ 1 file changed, 11 insertions(+)
Kconfig +11 -0
@@ -350,6 +350,17 @@ config PLATFORM_ELFENTRY default "__start" if MIPS default "_start" +config SYS_HAS_SRAM + bool + default y if TARGET_PIC32MZDASK + default y if TARGET_DEVKIT8000 + default y if TARGET_TRICORDER + default n + help + Enable this to allow support for the on board SRAM. + SRAM base address is controlled by CONFIG_SYS_SRAM_BASE. + SRAM size is controlled by CONFIG_SYS_SRAM_SIZE. + endmenu # General setup menu "Boot images"
Signature · 1 line
2.17.1
PATCH v4 2/17 Kconfig: Convert CONFIG_SYS_SRAM_BASE to Kconfig
This converts ad-hoc CONFIG_SYS_SRAM_BASE to Kconfig.
---
Kconfig | 5 +++++ include/configs/pic32mzdask.h | 1 - scripts/config_whitelist.txt | 1 - 3 files changed, 5 insertions(+), 2 deletions(-)
Kconfig +5 -0
@@ -361,6 +361,11 @@ config SYS_HAS_SRAM SRAM base address is controlled by CONFIG_SYS_SRAM_BASE. SRAM size is controlled by CONFIG_SYS_SRAM_SIZE. +config SYS_SRAM_BASE + hex + default 0x80000000 if TARGET_PIC32MZDASK + default 0x0 + endmenu # General setup menu "Boot images"
include/configs/pic32mzdask.h +0 -1
@@ -19,7 +19,6 @@/*---------------------------------------------------------------------- * Memory Layout */ -#define CONFIG_SYS_SRAM_BASE 0x80000000#define CONFIG_SYS_SRAM_SIZE 0x00080000/* 512K */ /* Initial RAM for temporary stack, global data */
scripts/config_whitelist.txt +0 -1
@@ -3788,7 +3788,6 @@ CONFIG_SYS_SPL_LEN CONFIG_SYS_SPL_MALLOC_SIZE CONFIG_SYS_SPL_MALLOC_START CONFIG_SYS_SPR -CONFIG_SYS_SRAM_BASE CONFIG_SYS_SRAM_SIZE CONFIG_SYS_SRAM_START CONFIG_SYS_SRIO
Signature · 1 line
2.17.1
PATCH v4 2/17 Kconfig: Convert CONFIG_SYS_SRAM_BASE to Kconfig
Quoted 4 lines · Ovidiu Panait
This converts ad-hoc
CONFIG_SYS_SRAM_BASEtoKconfig.Reviewed-by Simon Glass <[email protected]>Signed-off-by Ovidiu Panait <[email protected]>
Applied to u-boot/master, thanks!
Signature · 8 lines
Tom -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 659 bytes Desc: not available URL: <https://lists.denx.de/pipermail/u-boot/attachments/20200806/cdf64d8b/attachment.sig>
PATCH v4 3/17 Kconfig: Convert CONFIG_SYS_SRAM_SIZE to Kconfig
In reply to Ovidiu PanaitThis converts ad-hoc CONFIG_SYS_SRAM_SIZE to Kconfig.
---
Kconfig | 7 +++++++ include/configs/devkit8000.h | 1 - include/configs/pic32mzdask.h | 2 -- include/configs/tricorder.h | 1 - scripts/config_whitelist.txt | 1 - 5 files changed, 7 insertions(+), 5 deletions(-)
Kconfig +7 -0
@@ -366,6 +366,13 @@ config SYS_SRAM_BASE default 0x80000000 if TARGET_PIC32MZDASK default 0x0 +config SYS_SRAM_SIZE + hex + default 0x00080000 if TARGET_PIC32MZDASK + default 0x10000 if TARGET_DEVKIT8000 + default 0x10000 if TARGET_TRICORDER + default 0x0 + endmenu # General setup menu "Boot images"
include/configs/devkit8000.h +0 -1
@@ -140,7 +140,6 @@/* SRAM config */ #define CONFIG_SYS_SRAM_START 0x40200000 -#define CONFIG_SYS_SRAM_SIZE 0x10000/* Defines for SPL */
include/configs/pic32mzdask.h +0 -2
@@ -19,8 +19,6 @@/*---------------------------------------------------------------------- * Memory Layout */ -#define CONFIG_SYS_SRAM_SIZE 0x00080000/* 512K */ -/* Initial RAM for temporary stack, global data */ #define CONFIG_SYS_INIT_RAM_SIZE 0x10000#define CONFIG_SYS_INIT_RAM_ADDR \
include/configs/tricorder.h +0 -1
@@ -202,7 +202,6 @@/* SRAM config */ #define CONFIG_SYS_SRAM_START 0x40200000 -#define CONFIG_SYS_SRAM_SIZE 0x10000/* Defines for SPL */
scripts/config_whitelist.txt +0 -1
@@ -3788,7 +3788,6 @@ CONFIG_SYS_SPL_LEN CONFIG_SYS_SPL_MALLOC_SIZE CONFIG_SYS_SPL_MALLOC_START CONFIG_SYS_SPR -CONFIG_SYS_SRAM_SIZE CONFIG_SYS_SRAM_START CONFIG_SYS_SRIO CONFIG_SYS_SRIO1_MEM_BASE
Signature · 1 line
2.17.1
PATCH v4 3/17 Kconfig: Convert CONFIG_SYS_SRAM_SIZE to Kconfig
Quoted 4 lines · Ovidiu Panait
This converts ad-hoc
CONFIG_SYS_SRAM_SIZEtoKconfig.Reviewed-by Simon Glass <[email protected]>Signed-off-by Ovidiu Panait <[email protected]>
Applied to u-boot/master, thanks!
Signature · 8 lines
Tom -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 659 bytes Desc: not available URL: <https://lists.denx.de/pipermail/u-boot/attachments/20200806/12d98f2e/attachment.sig>
PATCH v4 4/17 Kconfig: Remove CONFIG_SYS_SRAM_START
In reply to Ovidiu PanaitRemove ad-hoc CONFIG_SYS_SRAM_START and use CONFIG_SYS_SRAM_BASE instead.
---
Kconfig | 2 ++ include/configs/devkit8000.h | 3 --- include/configs/tricorder.h | 3 --- scripts/config_whitelist.txt | 1 - 4 files changed, 2 insertions(+), 7 deletions(-)
Kconfig +2 -0
@@ -364,6 +364,8 @@ config SYS_HAS_SRAM config SYS_SRAM_BASE hex default 0x80000000 if TARGET_PIC32MZDASK + default 0x40200000 if TARGET_DEVKIT8000 + default 0x40200000 if TARGET_TRICORDER default 0x0 config SYS_SRAM_SIZE
include/configs/devkit8000.h +0 -3
@@ -138,9 +138,6 @@/* Boot Argument Buffer Size */ -/* SRAM config */ -#define CONFIG_SYS_SRAM_START 0x40200000 -/* Defines for SPL */ /* NAND boot config */
include/configs/tricorder.h +0 -3
@@ -200,9 +200,6 @@ CONFIG_SYS_INIT_RAM_SIZE - \ GENERATED_GBL_DATA_SIZE) -/* SRAM config */ -#define CONFIG_SYS_SRAM_START 0x40200000 -/* Defines for SPL */ #define CONFIG_SPL_NAND_BASE
scripts/config_whitelist.txt +0 -1
@@ -3788,7 +3788,6 @@ CONFIG_SYS_SPL_LEN CONFIG_SYS_SPL_MALLOC_SIZE CONFIG_SYS_SPL_MALLOC_START CONFIG_SYS_SPR -CONFIG_SYS_SRAM_START CONFIG_SYS_SRIO CONFIG_SYS_SRIO1_MEM_BASE CONFIG_SYS_SRIO1_MEM_BUS
Signature · 1 line
2.17.1
PATCH v4 4/17 Kconfig: Remove CONFIG_SYS_SRAM_START
Quoted 4 lines · Ovidiu Panait
Remove ad-hoc
CONFIG_SYS_SRAM_STARTand useCONFIG_SYS_SRAM_BASEinstead.Reviewed-by Simon Glass <[email protected]>Signed-off-by Ovidiu Panait <[email protected]>
Applied to u-boot/master, thanks!
Signature · 8 lines
Tom -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 659 bytes Desc: not available URL: <https://lists.denx.de/pipermail/u-boot/attachments/20200806/9048b2c2/attachment.sig>
PATCH v4 5/17 cmd: bdinfo: Move sram info prints to generic code
In reply to Ovidiu Panaitbi_sramstart and bi_sramsize are generic members of the bd_info structure, so move the m68k/powerpc-specific prints to generic code. Also, print them only if SRAM support is enabled via CONFIG_SYS_HAS_SRAM.
---
arch/m68k/lib/bdinfo.c | 4 ---- arch/powerpc/lib/bdinfo.c | 4 ---- cmd/bdinfo.c | 4 ++++ 3 files changed, 4 insertions(+), 8 deletions(-)
arch/m68k/lib/bdinfo.c +0 -4
@@ -15,10 +15,6 @@ void arch_print_bdinfo(void) {struct bd_info *bd = gd->bd; -#if defined(CONFIG_SYS_INIT_RAM_ADDR ) -bdinfo_print_num ("sramstart" , (ulong )bd->bi_sramstart); -bdinfo_print_num ("sramsize" , (ulong )bd->bi_sramsize); -#endif bdinfo_print_mhz ("busfreq" , bd->bi_busfreq);#if defined(CONFIG_SYS_MBAR )bdinfo_print_num ("mbar" , bd->bi_mbar_base);
arch/powerpc/lib/bdinfo.c +0 -4
@@ -20,10 +20,6 @@ void arch_print_bdinfo(void) {struct bd_info *bd = gd->bd; -#if defined(CONFIG_SYS_INIT_RAM_ADDR ) -bdinfo_print_num ("sramstart" , (ulong )bd->bi_sramstart); -bdinfo_print_num ("sramsize" , (ulong )bd->bi_sramsize); -#endif bdinfo_print_mhz ("busfreq" , bd->bi_busfreq);#if defined(CONFIG_MPC8xx) || defined(CONFIG_E500 )bdinfo_print_num ("immr_base" , bd->bi_immr_base);
cmd/bdinfo.c +4 -0
@@ -75,6 +75,10 @@ int do_bdinfo(struct cmd_tbl *cmdtp, int flag, int argc, char *const argv[])print_bi_dram (bd);bdinfo_print_num ("memstart" , (ulong )bd->bi_memstart);print_lnum ("memsize" , (u64 )bd->bi_memsize); +if (IS_ENABLED (CONFIG_SYS_HAS_SRAM )) { +bdinfo_print_num ("sramstart" , (ulong )bd->bi_sramstart); +bdinfo_print_num ("sramsize" , (ulong )bd->bi_sramsize); + }bdinfo_print_num ("flashstart" , (ulong )bd->bi_flashstart);bdinfo_print_num ("flashsize" , (ulong )bd->bi_flashsize);bdinfo_print_num ("flashoffset" , (ulong )bd->bi_flashoffset);
Signature · 1 line
2.17.1
PATCH v4 5/17 cmd: bdinfo: Move sram info prints to generic code
Quoted 6 lines · Ovidiu Panait
bi_sramstartandbi_sramsizeare generic members of thebd_infostructure, so move the m68k/powerpc-specific prints to generic code. Also, print them only if SRAM support is enabled viaCONFIG_SYS_HAS_SRAM.Reviewed-by Simon Glass <[email protected]>Signed-off-by Ovidiu Panait <[email protected]>
Applied to u-boot/master, thanks!
Signature · 8 lines
Tom -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 659 bytes Desc: not available URL: <https://lists.denx.de/pipermail/u-boot/attachments/20200806/d2815f47/attachment.sig>
PATCH v4 6/17 board_f: Introduce setup_bdinfo initcall
In reply to Ovidiu PanaitIntroduce setup_bdinfo initcall as a generic routine to populate bdinfo fields.
---
common/board_f.c | 6 ++++++ include/init.h | 10 ++++++++++ 2 files changed, 16 insertions(+)
common/board_f.c +6 -0
@@ -598,6 +598,11 @@ static int display_new_sp(void)return 0 ; } +int setup_bdinfo (void ) +{ +return 0 ; +} +#if defined(CONFIG_M68K ) || defined(CONFIG_MIPS ) || defined(CONFIG_PPC ) || \ defined(CONFIG_SH )static int setup_board_part1 (void )
@@ -975,6 +980,7 @@ static const init_fnc_t init_sequence_f[] = { reserve_stacks, dram_init_banksize, show_dram_config, + setup_bdinfo, #if defined(CONFIG_M68K) || defined(CONFIG_MIPS) || defined(CONFIG_PPC) || \ defined(CONFIG_SH) setup_board_part1,
include/init.h +10 -0
@@ -141,6 +141,16 @@ int arch_reserve_stacks(void);*/ int arch_reserve_mmu (void ); +/** +* setup_bdinfo() - Generic boardinfo setup +* +* Routine for populating various generic boardinfo fields of +* gd->bd. It is called during the generic board init sequence. +* +* Return: 0 if OK +*/ +int setup_bdinfo (void ); +/** * init_cache_f_r() - Turn on the cache in preparation for relocation *
Signature · 1 line
2.17.1
PATCH v4 6/17 board_f: Introduce setup_bdinfo initcall
Quoted 5 lines · Ovidiu Panait
Introduce
setup_bdinfoinitcall as a generic routine to populate bdinfo fields.Reviewed-by Simon Glass <[email protected]>Signed-off-by Ovidiu Panait <[email protected]>
Applied to u-boot/master, thanks!
Signature · 8 lines
Tom -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 659 bytes Desc: not available URL: <https://lists.denx.de/pipermail/u-boot/attachments/20200806/f9e58c6e/attachment.sig>
PATCH v4 7/17 board_f: Introduce arch_setup_bdinfo initcall
In reply to Ovidiu PanaitCertain architectures (ppc, mips, sh, m68k) use setup board_part1 and setup_board_part2 calls during pre-relocation init to populate gd->bd boardinfo fields. This makes the generic init sequence cluttered with arch-specific ifdefs.
In order to clean these arch-specific sequences from generic init, introduce arch_setup_bdinfo weak initcall so that everyone can define their own bdinfo setup routines.
---
common/board_f.c | 7 ++++++- include/init.h | 12 ++++++++++++ 2 files changed, 18 insertions(+), 1 deletion(-)
common/board_f.c +6 -1
@@ -598,11 +598,16 @@ static int display_new_sp(void)return 0 ; } -int setup_bdinfo (void ) +__weak int arch_setup_bdinfo (void ) {return 0 ; } +int setup_bdinfo (void ) +{ +return arch_setup_bdinfo (); +} +#if defined(CONFIG_M68K ) || defined(CONFIG_MIPS ) || defined(CONFIG_PPC ) || \ defined(CONFIG_SH )static int setup_board_part1 (void )
include/init.h +12 -0
@@ -141,6 +141,18 @@ int arch_reserve_stacks(void);*/ int arch_reserve_mmu (void ); +/** +* arch_setup_bdinfo() - Architecture dependent boardinfo setup +* +* Architecture-specific routine for populating various boardinfo fields of +* gd->bd. It is called during the generic board init sequence. +* +* If an implementation is not provided, it will just be a nop stub. +* +* Return: 0 if OK +*/ +int arch_setup_bdinfo (void ); +/** * setup_bdinfo() - Generic boardinfo setup *
Signature · 1 line
2.17.1
PATCH v4 7/17 board_f: Introduce arch_setup_bdinfo initcall
Quoted 11 lines · Ovidiu Panait
Certain architectures (ppc, mips, sh, m68k) use setup
board_part1andsetup_board_part2calls during pre-relocation init to populategd->bdboardinfo fields. This makes the generic init sequence cluttered with arch-specific ifdefs.In order to clean these arch-specific sequences from generic init, introduce
arch_setup_bdinfoweak initcall so that everyone can define their own bdinfo setup routines.Reviewed-by Simon Glass <[email protected]>Signed-off-by Ovidiu Panait <[email protected]>
Applied to u-boot/master, thanks!
Signature · 8 lines
Tom -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 659 bytes Desc: not available URL: <https://lists.denx.de/pipermail/u-boot/attachments/20200806/765a6e1f/attachment.sig>
PATCH v4 8/17 board_f: Factor out bdinfo bi_mem{start, size} to setup_bdinfo
In reply to Ovidiu PanaitMove all assignments to gd->bd->bi_mem{start,size} to generic code in setup_bdinfo.
Xtensa architecture is special in this regard as it defines its own handling of gd->bd->bi_mem{start,size} fields. In order to avoid defining a weak SDRAM function, let arch_setup_bdinfo overwrite the generic flags.
For ARC architecture, remove ARCH_EARLY_INIT_R from Kconfig since it is not needed anymore.
Also, use gd->ram_base to populate bi_memstart to avoid an ifdef.
---
arch/Kconfig | 1 - arch/arc/lib/cpu.c | 7 ------- arch/xtensa/lib/Makefile | 2 +- arch/xtensa/lib/bdinfo.c | 22 ++++++++++++++++++++++ board/cadence/xtfpga/xtfpga.c | 3 --- common/board_f.c | 11 +++++------ 6 files changed, 28 insertions(+), 18 deletions(-) create mode 100644 arch/xtensa/lib/bdinfo.c
arch/Kconfig +0 -1
@@ -10,7 +10,6 @@ choice config ARC bool "ARC architecture" - select ARCH_EARLY_INIT_R select ARC_TIMER select CLK select HAVE_PRIVATE_LIBGCC
arch/arc/lib/cpu.c +0 -7
@@ -25,13 +25,6 @@ int arch_cpu_init(void)return 0 ; } -int arch_early_init_r (void ) -{ - gd->bd->bi_memstart =CONFIG_SYS_SDRAM_BASE ; - gd->bd->bi_memsize =CONFIG_SYS_SDRAM_SIZE ; -return 0 ; -} -/* This is a dummy function on arc */ int dram_init (void ) {
arch/xtensa/lib/Makefile +1 -1
@@ -5,4 +5,4 @@ obj-$(CONFIG_CMD_BOOTM) += bootm.o -obj-y += cache.o misc.o relocate.o time.o +obj-y += cache.o misc.o relocate.o time.o bdinfo.o
arch/xtensa/lib/bdinfo.c new file +22 -0
@@ -0,0 +1,22 @@ +// SPDX-License-Identifier: GPL-2.0+ +/* +* XTENSA-specific information for the 'bd' command +* +* (C) Copyright 2003 +* Wolfgang Denk, DENX Software Engineering, wd at denx.de. +*/ + +#include <common.h> +#include <init.h> + +DECLARE_GLOBAL_DATA_PTR ; + +int arch_setup_bdinfo (void ) +{ +struct bd_info *bd = gd->bd; + + bd->bi_memstart =PHYSADDR (CONFIG_SYS_SDRAM_BASE ); + bd->bi_memsize =CONFIG_SYS_SDRAM_SIZE ; + +return 0 ; +}
board/cadence/xtfpga/xtfpga.c +0 -3
@@ -51,9 +51,6 @@ int checkboard(void)int dram_init_banksize (void ) { - gd->bd->bi_memstart =PHYSADDR (CONFIG_SYS_SDRAM_BASE ); - gd->bd->bi_memsize =CONFIG_SYS_SDRAM_SIZE ; -return 0 ; }
common/board_f.c +5 -6
@@ -605,6 +605,11 @@ __weak int arch_setup_bdinfo(void)int setup_bdinfo (void ) { +struct bd_info *bd = gd->bd; + + bd->bi_memstart = gd->ram_base;/* start of memory */ + bd->bi_memsize = gd->ram_size;/* size in bytes */ +return arch_setup_bdinfo (); }
@@ -614,12 +619,6 @@ static int setup_board_part1(void) {struct bd_info *bd = gd->bd; -/* -* Save local variables to board info struct -*/ - bd->bi_memstart =CONFIG_SYS_SDRAM_BASE ;/* start of memory */ - bd->bi_memsize = gd->ram_size;/* size in bytes */ -#ifdef CONFIG_SYS_SRAM_BASE bd->bi_sramstart =CONFIG_SYS_SRAM_BASE ;/* start of SRAM */ bd->bi_sramsize =CONFIG_SYS_SRAM_SIZE ;/* size of SRAM */
Signature · 1 line
2.17.1
PATCH v4 8/17 board_f: Factor out bdinfo bi_mem{start, size} to setup_bdinfo
Hi Ovidiu,
Quoted 44 lines
Move all assignments to gd->bd->bi_mem{start,size} to generic code in
setup_bdinfo.Xtensa architecture is special in this regard as it defines its own handling of gd->bd->bi_mem{start,size} fields. In order to avoid defining a weak SDRAM function, let
arch_setup_bdinfooverwrite the generic flags.For ARC architecture, remove
ARCH_EARLY_INIT_RfromKconfigsince it is not needed anymore.Also, use
gd->ram_baseto populatebi_memstartto avoid an ifdef.Signed-off-by Ovidiu Panait <[email protected]>---
arch/Kconfig | 1 - arch/arc/lib/cpu.c | 7 ------- arch/xtensa/lib/Makefile | 2 +- arch/xtensa/lib/bdinfo.c | 22 ++++++++++++++++++++++ board/cadence/xtfpga/xtfpga.c | 3 --- common/board_f.c | 11 +++++------ 6 files changed, 28 insertions(+), 18 deletions(-) create mode 100644 arch/xtensa/lib/bdinfo.carch/Kconfig +0
-1@@ -10,7 +10,6 @@ choice config ARC bool "ARC architecture" - select ARCH_EARLY_INIT_R select ARC_TIMER select CLK select HAVE_PRIVATE_LIBGCCarch/arc/lib/cpu.c +0
-0@@ -25,13 +25,6 @@ int arch_cpu_init(void)return 0 ; }
arch/arc/lib/cpu.c int arch_cpu_init(void) · line 30
Quoted 10 lines
arch/arc/lib/cpu.c +0
-7-int arch_early_init_r (void ) -{ - gd->bd->bi_memstart =CONFIG_SYS_SDRAM_BASE ; - gd->bd->bi_memsize =CONFIG_SYS_SDRAM_SIZE ; -return 0 ; -} -/* This is a dummy function on arc */ int dram_init (void ) {
For ARC part...
PATCH v4 8/17 board_f: Factor out bdinfo bi_mem{start, size} to setup_bdinfo
In reply to Ovidiu PanaitQuoted 16 lines · Ovidiu Panait
Move all assignments to gd->bd->bi_mem{start,size} to generic code in
setup_bdinfo.Xtensa architecture is special in this regard as it defines its own handling of gd->bd->bi_mem{start,size} fields. In order to avoid defining a weak SDRAM function, let
arch_setup_bdinfooverwrite the generic flags.For ARC architecture, remove
ARCH_EARLY_INIT_RfromKconfigsince it is not needed anymore.Also, use
gd->ram_baseto populatebi_memstartto avoid an ifdef.Signed-off-by Ovidiu Panait <[email protected]>---
Quoted 8 lines
arch/Kconfig | 1 - arch/arc/lib/cpu.c | 7 ------- arch/xtensa/lib/Makefile | 2 +- arch/xtensa/lib/bdinfo.c | 22 ++++++++++++++++++++++ board/cadence/xtfpga/xtfpga.c | 3 --- common/board_f.c | 11 +++++------ 6 files changed, 28 insertions(+), 18 deletions(-) create mode 100644 arch/xtensa/lib/bdinfo.c
Quoted 115 lines
arch/Kconfig +0
-1@@ -10,7 +10,6 @@ choice config ARC bool "ARC architecture" - select ARCH_EARLY_INIT_R select ARC_TIMER select CLK select HAVE_PRIVATE_LIBGCCarch/arc/lib/cpu.c +0
-7@@ -25,13 +25,6 @@ int arch_cpu_init(void)return 0 ; } -int arch_early_init_r (void ) -{ - gd->bd->bi_memstart =CONFIG_SYS_SDRAM_BASE ; - gd->bd->bi_memsize =CONFIG_SYS_SDRAM_SIZE ; -return 0 ; -} -/* This is a dummy function on arc */ int dram_init (void ) {arch/xtensa/lib/Makefile +1
-1@@ -5,4 +5,4 @@ obj-$(CONFIG_CMD_BOOTM) += bootm.o -obj-y += cache.o misc.o relocate.o time.o +obj-y += cache.o misc.o relocate.o time.o bdinfo.oarch/xtensa/lib/bdinfo.c new file +22
-0@@ -0,0 +1,22 @@ +// SPDX-License-Identifier: GPL-2.0+ +/* +* XTENSA-specific information for the 'bd' command +* +* (C) Copyright 2003 +* Wolfgang Denk, DENX Software Engineering, wd at denx.de. +*/ + +#include <common.h> +#include <init.h> + +DECLARE_GLOBAL_DATA_PTR ; + +int arch_setup_bdinfo (void ) +{ +struct bd_info *bd = gd->bd; + + bd->bi_memstart =PHYSADDR (CONFIG_SYS_SDRAM_BASE ); + bd->bi_memsize =CONFIG_SYS_SDRAM_SIZE ; + +return 0 ; +}board/cadence/xtfpga/xtfpga.c +0
-3@@ -51,9 +51,6 @@ int checkboard(void)int dram_init_banksize (void ) { - gd->bd->bi_memstart =PHYSADDR (CONFIG_SYS_SDRAM_BASE ); - gd->bd->bi_memsize =CONFIG_SYS_SDRAM_SIZE ; -return 0 ; }common/board_f.c +5
-7@@ -605,6 +605,11 @@ __weak int arch_setup_bdinfo(void)int setup_bdinfo (void ) { +struct bd_info *bd = gd->bd; + + bd->bi_memstart = gd->ram_base;/* start of memory */ + bd->bi_memsize = gd->ram_size;/* size in bytes */ +return arch_setup_bdinfo (); }@@ -614,12 +619,6 @@ static int setup_board_part1(void) {struct bd_info *bd = gd->bd; -/* -* Save local variables to board info struct -*/ - bd->bi_memstart =CONFIG_SYS_SDRAM_BASE ;/* start of memory */ - bd->bi_memsize = gd->ram_size;/* size in bytes */ -#ifdef CONFIG_SYS_SRAM_BASE bd->bi_sramstart =CONFIG_SYS_SRAM_BASE ;/* start of SRAM */ bd->bi_sramsize =CONFIG_SYS_SRAM_SIZE ;/* size of SRAM */ --2.17.1
PATCH v4 8/17 board_f: Factor out bdinfo bi_mem{start, size} to setup_bdinfo
In reply to Ovidiu PanaitQuoted 7 lines · Ovidiu Panait
Move all assignments to gd->bd->bi_mem{start,size} to generic code in
setup_bdinfo.Xtensa architecture is special in this regard as it defines its own handling of gd->bd->bi_mem{start,size} fields. In order to avoid defining a weak SDRAM function, let
arch_setup_bdinfooverwrite the generic flags.
Quoted 8 lines
For ARC architecture, remove
ARCH_EARLY_INIT_RfromKconfigsince it is not needed anymore.Also, use
gd->ram_baseto populatebi_memstartto avoid an ifdef.Signed-off-by Ovidiu Panait <[email protected]>Acked-by Alexey Brodkin <[email protected]>Reviewed-by Simon Glass <[email protected]>
Applied to u-boot/master, thanks!
Signature · 8 lines
Tom -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 659 bytes Desc: not available URL: <https://lists.denx.de/pipermail/u-boot/attachments/20200806/2494081c/attachment.sig>
PATCH v4 9/17 board_f: Move sram bdinfo assignments to generic code
In reply to Ovidiu PanaitMove sram related bdinfo from arch-specific setup_board_part1 to generic code in setup_bdinfo. Also use "if (IS_ENABLED(CONFIG_SYS_HAS_SRAM))" instead of "#ifdef CONFIG_SYS_SRAM_BASE".
---
common/board_f.c | 10 +++++----- 1 file changed, 5 insertions(+), 5 deletions(-)
common/board_f.c +5 -5
@@ -610,6 +610,11 @@ int setup_bdinfo(void) bd->bi_memstart = gd->ram_base;/* start of memory */ bd->bi_memsize = gd->ram_size;/* size in bytes */ +if (IS_ENABLED (CONFIG_SYS_HAS_SRAM )) { + bd->bi_sramstart =CONFIG_SYS_SRAM_BASE ;/* start of SRAM */ + bd->bi_sramsize =CONFIG_SYS_SRAM_SIZE ;/* size of SRAM */ + } +return arch_setup_bdinfo (); }
@@ -619,11 +624,6 @@ static int setup_board_part1(void) {struct bd_info *bd = gd->bd; -#ifdef CONFIG_SYS_SRAM_BASE - bd->bi_sramstart =CONFIG_SYS_SRAM_BASE ;/* start of SRAM */ - bd->bi_sramsize =CONFIG_SYS_SRAM_SIZE ;/* size of SRAM */ -#endif -#if defined(CONFIG_E500 ) || defined(CONFIG_MPC86xx) bd->bi_immr_base =CONFIG_SYS_IMMR ;/* base of IMMR register */ #endif
Signature · 1 line
2.17.1
PATCH v4 9/17 board_f: Move sram bdinfo assignments to generic code
Quoted 6 lines · Ovidiu Panait
Move sram related bdinfo from arch-specific
setup_board_part1to generic code insetup_bdinfo. Also use "if (IS_ENABLED(CONFIG_SYS_HAS_SRAM))" instead of "#ifdefCONFIG_SYS_SRAM_BASE".Reviewed-by Simon Glass <[email protected]>Signed-off-by Ovidiu Panait <[email protected]>
Applied to u-boot/master, thanks!
Signature · 8 lines
Tom -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 659 bytes Desc: not available URL: <https://lists.denx.de/pipermail/u-boot/attachments/20200806/3d713778/attachment.sig>
PATCH v4 10/17 board_f: m68k: Factor out m68k-specific bdinfo setup
In reply to Ovidiu PanaitFactor out m68k-specific bdinfo setup to arch_setup_bdinfo in arch/m68k/lib/bdinfo.c. Also, use if(IS_ENABLED()) instead of #ifdef where possible.
---
arch/m68k/lib/bdinfo.c | 21 +++++++++++++++++++++ common/board_f.c | 19 ++++--------------- 2 files changed, 25 insertions(+), 15 deletions(-)
arch/m68k/lib/bdinfo.c +21 -0
@@ -11,6 +11,27 @@DECLARE_GLOBAL_DATA_PTR ; +int arch_setup_bdinfo (void ) +{ +struct bd_info *bd = gd->bd; + + bd->bi_mbar_base =CONFIG_SYS_MBAR ;/* base of internal registers */ + + bd->bi_intfreq = gd->cpu_clk;/* Internal Freq, in Hz */ + bd->bi_busfreq = gd->bus_clk;/* Bus Freq, in Hz */ + +if (IS_ENABLED (CONFIG_PCI )) + bd->bi_pcifreq = gd->pci_clk; + +#if defined(CONFIG_EXTRA_CLOCK ) + bd->bi_inpfreq = gd->arch.inp_clk;/* input Freq in Hz */ + bd->bi_vcofreq = gd->arch.vco_clk;/* vco Freq in Hz */ + bd->bi_flbfreq = gd->arch.flb_clk;/* flexbus Freq in Hz */ +#endif + +return 0 ; +} +void arch_print_bdinfo (void ) {struct bd_info *bd = gd->bd;
common/board_f.c +4 -15
@@ -618,7 +618,7 @@ int setup_bdinfo(void)return arch_setup_bdinfo (); } -#if defined(CONFIG_M68K ) || defined(CONFIG_MIPS ) || defined(CONFIG_PPC ) || \ +#if defined(CONFIG_MIPS ) || defined(CONFIG_PPC ) || \ defined(CONFIG_SH )static int setup_board_part1 (void ) {
@@ -627,9 +627,6 @@ static int setup_board_part1(void)#if defined(CONFIG_E500 ) || defined(CONFIG_MPC86xx) bd->bi_immr_base =CONFIG_SYS_IMMR ;/* base of IMMR register */ #endif -#if defined(CONFIG_M68K ) - bd->bi_mbar_base =CONFIG_SYS_MBAR ;/* base of internal registers */ -#endif #if defined(CONFIG_MPC83xx) bd->bi_immrbar =CONFIG_SYS_IMMR ;#endif
@@ -638,7 +635,7 @@ static int setup_board_part1(void) }#endif -#if defined(CONFIG_PPC ) || defined(CONFIG_M68K ) +#if defined(CONFIG_PPC )static int setup_board_part2 (void ) {struct bd_info *bd = gd->bd;
@@ -651,14 +648,6 @@ static int setup_board_part2(void) bd->bi_sccfreq = gd->arch.scc_clk; bd->bi_vco = gd->arch.vco_out;#endif /* CONFIG_CPM2 */ -#if defined(CONFIG_M68K ) && defined(CONFIG_PCI ) - bd->bi_pcifreq = gd->pci_clk; -#endif -#if defined(CONFIG_EXTRA_CLOCK ) - bd->bi_inpfreq = gd->arch.inp_clk;/* input Freq in Hz */ - bd->bi_vcofreq = gd->arch.vco_clk;/* vco Freq in Hz */ - bd->bi_flbfreq = gd->arch.flb_clk;/* flexbus Freq in Hz */ -#endif return 0 ; }
@@ -985,11 +974,11 @@ static const init_fnc_t init_sequence_f[] = { dram_init_banksize, show_dram_config, setup_bdinfo, -#if defined(CONFIG_M68K) || defined(CONFIG_MIPS) || defined(CONFIG_PPC) || \ +#if defined(CONFIG_MIPS) || defined(CONFIG_PPC) || \ defined(CONFIG_SH) setup_board_part1, #endif -#if defined(CONFIG_PPC) || defined(CONFIG_M68K) +#if defined(CONFIG_PPC) INIT_FUNC_WATCHDOG_RESET setup_board_part2, #endif
Signature · 1 line
2.17.1
PATCH v4 10/17 board_f: m68k: Factor out m68k-specific bdinfo setup
Quoted 6 lines · Ovidiu Panait
Factor out m68k-specific bdinfo setup to
arch_setup_bdinfoinarch/m68k/lib/bdinfo.c. Also, use if(IS_ENABLED()) instead of #ifdef where possible.Reviewed-by Simon Glass <[email protected]>Signed-off-by Ovidiu Panait <[email protected]>
Applied to u-boot/master, thanks!
Signature · 8 lines
Tom -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 659 bytes Desc: not available URL: <https://lists.denx.de/pipermail/u-boot/attachments/20200806/868ffd92/attachment.sig>
PATCH v4 11/17 board_f: ppc: Factor out ppc-specific bdinfo setup
In reply to Ovidiu PanaitFactor out ppc-specific bdinfo setup from generic init sequence to arch_setup_bdinfo in arch/powerpc/lib/bdinfo.c.
---
arch/powerpc/lib/bdinfo.c | 25 +++++++++++++++++++++++++ common/board_f.c | 36 ++---------------------------------- 2 files changed, 27 insertions(+), 34 deletions(-)
arch/powerpc/lib/bdinfo.c +25 -0
@@ -11,6 +11,31 @@DECLARE_GLOBAL_DATA_PTR ; +int arch_setup_bdinfo (void ) +{ +struct bd_info *bd = gd->bd; + +#if defined(CONFIG_E500 ) || defined(CONFIG_MPC86xx) + bd->bi_immr_base =CONFIG_SYS_IMMR ;/* base of IMMR register */ +#endif + +#if defined(CONFIG_MPC83xx) + bd->bi_immrbar =CONFIG_SYS_IMMR ; +#endif + + bd->bi_intfreq = gd->cpu_clk;/* Internal Freq, in Hz */ + bd->bi_busfreq = gd->bus_clk;/* Bus Freq, in Hz */ + +#if defined(CONFIG_CPM2 ) + bd->bi_cpmfreq = gd->arch.cpm_clk; + bd->bi_brgfreq = gd->arch.brg_clk; + bd->bi_sccfreq = gd->arch.scc_clk; + bd->bi_vco = gd->arch.vco_out; +#endif /* CONFIG_CPM2 */ + +return 0 ; +} +void __weak board_detail (void ) {/* Please define board_detail() for your PPC platform */
common/board_f.c +2 -34
@@ -618,37 +618,9 @@ int setup_bdinfo(void)return arch_setup_bdinfo (); } -#if defined(CONFIG_MIPS ) || defined(CONFIG_PPC ) || \ - defined(CONFIG_SH ) +#if defined(CONFIG_MIPS ) || defined(CONFIG_SH )static int setup_board_part1 (void ) { -struct bd_info *bd = gd->bd; - -#if defined(CONFIG_E500 ) || defined(CONFIG_MPC86xx) - bd->bi_immr_base =CONFIG_SYS_IMMR ;/* base of IMMR register */ -#endif -#if defined(CONFIG_MPC83xx) - bd->bi_immrbar =CONFIG_SYS_IMMR ; -#endif - -return 0 ; -} -#endif - -#if defined(CONFIG_PPC ) -static int setup_board_part2 (void ) -{ -struct bd_info *bd = gd->bd; - - bd->bi_intfreq = gd->cpu_clk;/* Internal Freq, in Hz */ - bd->bi_busfreq = gd->bus_clk;/* Bus Freq, in Hz */ -#if defined(CONFIG_CPM2 ) - bd->bi_cpmfreq = gd->arch.cpm_clk; - bd->bi_brgfreq = gd->arch.brg_clk; - bd->bi_sccfreq = gd->arch.scc_clk; - bd->bi_vco = gd->arch.vco_out; -#endif /* CONFIG_CPM2 */ -return 0 ; }#endif
@@ -974,14 +946,10 @@ static const init_fnc_t init_sequence_f[] = { dram_init_banksize, show_dram_config, setup_bdinfo, -#if defined(CONFIG_MIPS) || defined(CONFIG_PPC) || \ - defined(CONFIG_SH) +#if defined(CONFIG_MIPS) || defined(CONFIG_SH) setup_board_part1, #endif -#if defined(CONFIG_PPC) INIT_FUNC_WATCHDOG_RESET - setup_board_part2, -#endif display_new_sp, #ifdef CONFIG_OF_BOARD_FIXUP fix_fdt,
Signature · 1 line
2.17.1
PATCH v4 11/17 board_f: ppc: Factor out ppc-specific bdinfo setup
Quoted 5 lines · Ovidiu Panait
Factor out ppc-specific bdinfo setup from generic init sequence to
arch_setup_bdinfoinarch/powerpc/lib/bdinfo.c.Reviewed-by Simon Glass <[email protected]>Signed-off-by Ovidiu Panait <[email protected]>
Applied to u-boot/master, thanks!
Signature · 8 lines
Tom -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 659 bytes Desc: not available URL: <https://lists.denx.de/pipermail/u-boot/attachments/20200806/142b1298/attachment.sig>
PATCH v4 12/17 board_f: Remove setup_board_part1
In reply to Ovidiu PanaitNow that all arch specific code was converted to setup_bdinfo, we can remove setup_board_part1.
---
common/board_f.c | 12 +----------- 1 file changed, 1 insertion(+), 11 deletions(-)
common/board_f.c +1 -11
@@ -618,13 +618,6 @@ int setup_bdinfo(void)return arch_setup_bdinfo (); } -#if defined(CONFIG_MIPS ) || defined(CONFIG_SH ) -static int setup_board_part1 (void ) -{ -return 0 ; -} -#endif -#ifdef CONFIG_POST static int init_post (void ) {
@@ -945,11 +938,8 @@ static const init_fnc_t init_sequence_f[] = { reserve_stacks, dram_init_banksize, show_dram_config, - setup_bdinfo, -#if defined(CONFIG_MIPS) || defined(CONFIG_SH) - setup_board_part1, -#endif INIT_FUNC_WATCHDOG_RESET + setup_bdinfo, display_new_sp,#ifdef CONFIG_OF_BOARD_FIXUP fix_fdt,
Signature · 1 line
2.17.1
PATCH v4 12/17 board_f: Remove setup_board_part1
Quoted 5 lines · Ovidiu Panait
Now that all arch specific code was converted to
setup_bdinfo, we can removesetup_board_part1.Reviewed-by Simon Glass <[email protected]>Signed-off-by Ovidiu Panait <[email protected]>
Applied to u-boot/master, thanks!
Signature · 8 lines
Tom -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 659 bytes Desc: not available URL: <https://lists.denx.de/pipermail/u-boot/attachments/20200806/cd7a3f0c/attachment.sig>
PATCH v4 13/17 dm: blk: Use IS_ENABLED() instead of #ifdefs in blk_post_probe
In reply to Ovidiu PanaitUse IS_ENABLED() instead of #ifdef in blk_post_probe function.
No functional change intended.
---
drivers/block/blk-uclass.c | 9 +++++---- 1 file changed, 5 insertions(+), 4 deletions(-)
drivers/block/blk-uclass.c +5 -4
@@ -644,11 +644,12 @@ int blk_unbind_all(int if_type)static int blk_post_probe (struct udevice *dev) { -#if defined(CONFIG_PARTITIONS ) && defined(CONFIG_HAVE_BLOCK_DEVICE ) -struct blk_desc *desc =dev_get_uclass_platdata (dev); +if (IS_ENABLED (CONFIG_PARTITIONS ) && +IS_ENABLED (CONFIG_HAVE_BLOCK_DEV )) { +struct blk_desc *desc =dev_get_uclass_platdata (dev); -part_init (desc); -#endif +part_init (desc); + }return 0 ; }
Signature · 1 line
2.17.1
PATCH v4 13/17 dm: blk: Use IS_ENABLED() instead of #ifdefs in blk_post_probe
Quoted 6 lines · Ovidiu Panait
Use
IS_ENABLED()instead of #ifdef inblk_post_probefunction.No functional change intended.
Reviewed-by Simon Glass <[email protected]>Signed-off-by Ovidiu Panait <[email protected]>
Applied to u-boot/master, thanks!
Signature · 8 lines
Tom -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 659 bytes Desc: not available URL: <https://lists.denx.de/pipermail/u-boot/attachments/20200806/5a375df9/attachment.sig>
PATCH v4 14/17 drivers: serial: Make serial_initialize return int
In reply to Ovidiu Panaitserial_initialize is called only during the common init sequence, after relocation (in common/board_r.c). Because it has a void return value, it has to wrapped in initr_serial. In order to be able to get rid of this indirection, make serial_initialize return int.
Remove extern from prototype in order to silence the following checkpatch warning:
check: extern prototypes should be avoided in .h files
---
drivers/serial/serial-uclass.c | 4 ++-- drivers/serial/serial.c | 4 +++- include/serial.h | 2 +- 3 files changed, 6 insertions(+), 4 deletions(-)
drivers/serial/serial-uclass.c +2 -2
@@ -170,9 +170,9 @@ int serial_init(void) }/* Called after relocation */ -void serial_initialize (void ) +int serial_initialize (void ) { -serial_init (); +return serial_init (); }static void _serial_putc (struct udevice *dev,char ch)
drivers/serial/serial.c +3 -1
@@ -170,7 +170,7 @@ void serial_register(struct serial_device *dev)* serial port to the serial core. That serial port is then used as a * default output. */ -void serial_initialize (void ) +int serial_initialize (void ) {atmel_serial_initialize ();mcf_serial_initialize ();
@@ -183,6 +183,8 @@ void serial_initialize(void)mtk_serial_initialize ();serial_assign (default_serial_console ()->name); + +return 0 ; }static int serial_stub_start (struct stdio_dev *sdev)
include/serial.h +1 -1
@@ -42,10 +42,10 @@ extern struct serial_device eserial5_device;extern struct serial_device eserial6_device;extern void serial_register (struct serial_device *); -extern void serial_initialize (void );extern void serial_stdio_init (void );extern int serial_assign (const char *name);extern void serial_reinit_all (void ); +int serial_initialize (void );/* For usbtty */ #ifdef CONFIG_USB_TTY
Signature · 1 line
2.17.1
PATCH v4 14/17 drivers: serial: Make serial_initialize return int
Quoted 11 lines · Ovidiu Panait
serial_initializeis called only during the common init sequence, after relocation (incommon/board_r.c). Because it has a void return value, it has to wrapped ininitr_serial. In order to be able to get rid of this indirection, makeserial_initializereturn int.Remove extern from prototype in order to silence the following checkpatch warning:
check: extern prototypes should be avoided in .h filesReviewed-by Simon Glass <[email protected]>Signed-off-by Ovidiu Panait <[email protected]>
Applied to u-boot/master, thanks!
Signature · 8 lines
Tom -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 659 bytes Desc: not available URL: <https://lists.denx.de/pipermail/u-boot/attachments/20200806/a6eb0547/attachment.sig>
PATCH v4 15/17 common/board_r: Remove initr_serial wrapper
In reply to Ovidiu PanaitRemove the initr_serial->serial_initialize indirection and call serial_initialize directly.
---
common/board_r.c | 8 +------- 1 file changed, 1 insertion(+), 7 deletions(-)
common/board_r.c +1 -7
@@ -187,12 +187,6 @@ static int initr_reloc_global_data(void)return 0 ; } -static int initr_serial (void ) -{ -serial_initialize (); -return 0 ; -} -#if defined(CONFIG_PPC ) || defined(CONFIG_M68K ) || defined(CONFIG_MIPS )static int initr_trap (void ) {
@@ -705,7 +699,7 @@ static init_fnc_t init_sequence_r[] = {#endif initr_dm_devices, stdio_init_tables, - initr_serial, + serial_initialize, initr_announce, #if CONFIG_IS_ENABLED(WDT) initr_watchdog,
Signature · 1 line
2.17.1
PATCH v4 15/17 common/board_r: Remove initr_serial wrapper
Quoted 5 lines · Ovidiu Panait
Remove the
initr_serial->serial_initializeindirection and callserial_initializedirectly.Reviewed-by Simon Glass <[email protected]>Signed-off-by Ovidiu Panait <[email protected]>
Applied to u-boot/master, thanks!
Signature · 8 lines
Tom -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 659 bytes Desc: not available URL: <https://lists.denx.de/pipermail/u-boot/attachments/20200806/18ea8c4b/attachment.sig>
PATCH v4 15/17 common/board_r: Remove initr_serial wrapper
In reply to Ovidiu PanaitQuoted 29 lines · Ovidiu Panait
Remove the
initr_serial->serial_initializeindirection and callserial_initializedirectly.Reviewed-by Simon Glass <[email protected]>Signed-off-by Ovidiu Panait <[email protected]>---
common/board_r.c | 8 +------- 1 file changed, 1 insertion(+), 7 deletions(-)common/board_r.c +0
-6@@ -187,12 +187,6 @@ static int initr_reloc_global_data(void)return 0 ; } -static int initr_serial (void ) -{ -serial_initialize (); -return 0 ; -} -#if defined(CONFIG_PPC ) || defined(CONFIG_M68K ) || defined(CONFIG_MIPS )static int initr_trap (void ) {
common/board_r.c static init_fnc_t init_sequence_r[] = { · line 705
Quoted 8 lines
common/board_r.c +1
-1@@ -705,7 +699,7 @@ static init_fnc_t init_sequence_r[] = {#endif initr_dm_devices, stdio_init_tables, - initr_serial, + serial_initialize, initr_announce, #if CONFIG_IS_ENABLED(WDT) initr_watchdog,
This commit breaks my patch series [1]. I suspect it is the previous patch which does the breaking. Here are two versions of my series, with the only difference being that one is applied before this patch and the other after [2, 3]. As a few additional notes, this patch does *not* break booting without my series applied. In addition, setting CONFIG_LOGLEVEL to 5 or higher causes the board to start booting again.
I am pretty confused by all this, as I never directly touch any serial code in my series. Further, I have no idea why it would start working again just by changing the loglevel.
--Sean
[1] https://patchwork.ozlabs.org/project/uboot/list/?series=185489
[2] https://github.com/Forty-Bot/u-boot/tree/maix_gpio_good
[2] https://github.com/Forty-Bot/u-boot/tree/maix_gpio_bad
PATCH v4 15/17 common/board_r: Remove initr_serial wrapper
Quoted 50 lines · Sean Anderson
Quoted 29 lines · Ovidiu Panait
Remove the
initr_serial->serial_initializeindirection and callserial_initializedirectly.Reviewed-by Simon Glass <[email protected]>Signed-off-by Ovidiu Panait <[email protected]>---
common/board_r.c | 8 +------- 1 file changed, 1 insertion(+), 7 deletions(-)common/board_r.c +0
-6@@ -187,12 +187,6 @@ static int initr_reloc_global_data(void)return 0 ; } -static int initr_serial (void ) -{ -serial_initialize (); -return 0 ; -} -#if defined(CONFIG_PPC ) || defined(CONFIG_M68K ) || defined(CONFIG_MIPS )static int initr_trap (void ) {common/board_r.c static init_fnc_t init_sequence_r[] = { · line 705
Quoted 8 lines
common/board_r.c +1
-1@@ -705,7 +699,7 @@ static init_fnc_t init_sequence_r[] = {#endif initr_dm_devices, stdio_init_tables, - initr_serial, + serial_initialize, initr_announce, #if CONFIG_IS_ENABLED(WDT) initr_watchdog,This commit breaks my patch series [1]. I suspect it is the previous patch which does the breaking. Here are two versions of my series, with the only difference being that one is applied before this patch and the other after [2, 3]. As a few additional notes, this patch does *not* break booting without my series applied. In addition, setting
CONFIG_LOGLEVELto 5 or higher causes the board to start booting again.I am pretty confused by all this, as I never directly touch any serial code in my series. Further, I have no idea why it would start working again just by changing the loglevel.
Quoted 6 lines
--Sean
[1] https://patchwork.ozlabs.org/project/uboot/list/?series=185489
[2] https://github.com/Forty-Bot/u-boot/tree/maix_gpio_good
[2] https://github.com/Forty-Bot/u-boot/tree/maix_gpio_bad
+Heinrich can you verify this behavior?
--Sean
PATCH v4 16/17 blkcache: Extend blkcache_init to cover CONFIG_NEEDS_MANUAL_RELOC
In reply to Ovidiu PanaitExtend manual relocation of block_cache list pointers to all platforms that enable CONFIG_NEEDS_MANUAL_RELOC. Remove m68k-specific checks and provide a single implementation that adds gd->reloc_off to the pre-relocation pointers.
---
common/board_r.c | 2 +- drivers/block/blkcache.c | 13 +++++++------ 2 files changed, 8 insertions(+), 7 deletions(-)
common/board_r.c +1 -1
@@ -836,7 +836,7 @@ static init_fnc_t init_sequence_r[] = {#if defined(CONFIG_PRAM ) initr_mem,#endif -#if defined(CONFIG_M68K ) && defined(CONFIG_BLOCK_CACHE ) +#if defined(CONFIG_NEEDS_MANUAL_RELOC ) && defined(CONFIG_BLOCK_CACHE ) blkcache_init, #endif run_main_loop,
drivers/block/blkcache.c +7 -6
@@ -12,6 +12,8 @@#include <linux/ctype.h> #include <linux/list.h> +DECLARE_GLOBAL_DATA_PTR ; +struct block_cache_node {struct list_head lh;int iftype;
@@ -22,21 +24,20 @@ struct block_cache_node {char *cache; }; -#ifndef CONFIG_M68K static LIST_HEAD (block_cache); -#else -static struct list_head block_cache; -#endif static struct block_cache_stats _stats = { .max_blocks_per_entry =8 , .max_entries =32 }; -#ifdef CONFIG_M68K +#ifdef CONFIG_NEEDS_MANUAL_RELOC int blkcache_init (void ) { -INIT_LIST_HEAD (&block_cache); +struct list_head *head = &block_cache; + + head->next = (uintptr_t )head->next + gd->reloc_off; + head->prev = (uintptr_t )head->prev + gd->reloc_off;return 0 ; }
Signature · 1 line
2.17.1
PATCH v4 16/17 blkcache: Extend blkcache_init to cover CONFIG_NEEDS_MANUAL_RELOC
Quoted 10 lines · Ovidiu Panait
Extend manual relocation of
block_cachelist pointers to all platforms that enableCONFIG_NEEDS_MANUAL_RELOC. Remove m68k-specific checks and provide a single implementation that addsgd->reloc_offto the pre-relocation pointers.Tested-by Angelo Dureghello <[email protected]>Reviewed-by Simon Glass <[email protected]>Reviewed-by Eric Nelson <[email protected]>Signed-off-by Ovidiu Panait <[email protected]>
Applied to u-boot/master, thanks!
Signature · 8 lines
Tom -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 659 bytes Desc: not available URL: <https://lists.denx.de/pipermail/u-boot/attachments/20200806/7f754ff5/attachment.sig>
PATCH v4 17/17 common/board_r: Move blkcache_init call earlier in the boot sequence
In reply to Ovidiu Panaitblkcache_init manually relocates blkcache list pointers when CONFIG_NEEDS_MANUAL_RELOC is enabled. However, it is called very late in the boot sequence, which could be a problem if previous boot calls execute blkcache operations with the non-relocated pointers. For example, mmc is initialized earlier and might call blkcache_invalidate (in mmc_select_hwpart()) when trying to load the environment from mmc via env_load().
To fix this issue, move blkcache_init boot call earlier, before mmc gets initialized.
---
common/board_r.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-)
common/board_r.c +3 -3
@@ -705,6 +705,9 @@ static init_fnc_t init_sequence_r[] = { initr_watchdog, #endif INIT_FUNC_WATCHDOG_RESET +#if defined(CONFIG_NEEDS_MANUAL_RELOC ) && defined(CONFIG_BLOCK_CACHE ) + blkcache_init, +#endif #ifdef CONFIG_NEEDS_MANUAL_RELOC initr_manual_reloc_cmdtable,#endif
@@ -835,9 +838,6 @@ static init_fnc_t init_sequence_r[] = {#endif #if defined(CONFIG_PRAM ) initr_mem, -#endif -#if defined(CONFIG_NEEDS_MANUAL_RELOC ) && defined(CONFIG_BLOCK_CACHE ) - blkcache_init, #endif run_main_loop, };
Signature · 1 line
2.17.1
PATCH v4 17/17 common/board_r: Move blkcache_init call earlier in the boot sequence
Quoted 8 lines · Ovidiu Panait
blkcache_initmanually relocates blkcache list pointers whenCONFIG_NEEDS_MANUAL_RELOCis enabled. However, it is called very late in the boot sequence, which could be a problem if previous boot calls execute blkcache operations with the non-relocated pointers. For example, mmc is initialized earlier and might callblkcache_invalidate(inmmc_select_hwpart()) when trying to load the environment from mmc viaenv_load().
Quoted 7 lines
To fix this issue, move
blkcache_initboot call earlier, before mmc gets initialized.Tested-by Angelo Dureghello <[email protected]>Reviewed-by Simon Glass <[email protected]>Signed-off-by Ovidiu Panait <[email protected]>
Applied to u-boot/master, thanks!
Signature · 8 lines
Tom -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 659 bytes Desc: not available URL: <https://lists.denx.de/pipermail/u-boot/attachments/20200806/a6c4d42e/attachment.sig>
PATCH v4 1/17 Kconfig: Introduce CONFIG_SYS_HAS_SRAM
In reply to Ovidiu PanaitQuoted 6 lines · Ovidiu Panait
In order to be able to replace "#ifdef
CONFIG_SYS_SRAM_BASE" sequences with theIS_ENABLED()equivalent, introduce a new booleanKconfigoption that signals whether the platform has SRAM support.Reviewed-by Simon Glass <[email protected]>Signed-off-by Ovidiu Panait <[email protected]>
Applied to u-boot/master, thanks!
Signature · 8 lines
Tom -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 659 bytes Desc: not available URL: <https://lists.denx.de/pipermail/u-boot/attachments/20200806/4d21fa5f/attachment.sig>