Skip to contentSkip to conversation

Tidy up the 'bd' command.

PATCH New Simon Glass 8 participants 93 messages May 4, 2020 – May 10, 2020 4 messages not in the archive

Subsystems (7)
Arm (32-bit) x86 PowerPC RISC-V +3 more subsystems
lore
Share & tools
lore
#0 lore raw

cover Tidy up the 'bd' command.

The code for the 'bd' command never got the 'generic board' treatment many years ago when global_data and bd_info were converted. As a result it still has a lot of arch-specific duplication of generic code.

This series aims to make as much code in this file generic as possible, so that it is easy to add new info on all architectures.

For the three architectures that actually need additional code (ARM, PPC and m68k) this is moved into arch-specific files.

With this series, bdinfo.c drops from nearly 500 lines to just over 100.

It also makes x86 report the frame buffer address properly (the original goal of my effort).

Simon Glass (36):
bdinfo: nds32: Use generic bd_info
bdinfo: riscv: Use generic bd_info
bdinfo: m68k: Drop bd_info->bi_ipbfreq
bdinfo: xtensa: Create a generic do_bdinfo for xtensa
bdinfo: mips: Use the generic bd command
bdinfo: nios2: Use the generic bd command
bdinfo: microblaze: Use the generic bd command
bdinfo: sh: Use the generic bd command
bdinfo: x86: Use the generic bd command
bdinfo: sandbox: Use the generic bd command
bdinfo: nds32: Use the generic bd command
bdinfo: riscv: Use the generic bd command
bdinfo: arm: Use the generic bd command
bdinfo: Drop the option to not use the generic 'bd' command
bdinfo: Drop unused __maybe_unused
bdinfo: microblaze: sh: nios2: Drop arch-specific flash info
bdinfo: Drop unnecessary inline on functions
bdinfo: Drop print_std_bdinfo()
bdinfo: ppc: Drop arch-specific print_baudrate()
bdinfo: sh: arc: Drop arch-specific print_bi_mem()
bdinfo: Drop print_bi_boot_params()
bdinfo: Drop print_bi_flash()
bdinfo: Drop print_cpu_word_size()
bdinfo: net: ppc: Drop bi_enet1addr and other similar info
bdinfo: net: ppc: Drop prints for CONFIG_HAS_ETHn
bdinfo: net: Drop legacy ethernet bdinfo
bdinfo: net: Inline print_eths()
bdinfo: net: Inline print_eth_ip_addr()
bdinfo: Export some basic printing functions
bdinfo: arm: Move ARM-specific info into its own file
bdinfo: ppc: Move PPC-specific info into its own file
bdinfo: m68k: Move m68k-specific info into its own file
bdinfo: m68k: ppc: Move arch-specific code from bdinfo
bdinfo: Update the file comments
bdinfo: dm: Update fb_base when using driver model
bdinfo: x86: vesa: Update fb_base to the correct value

 arch/arm/lib/Makefile           |   1 +
 arch/arm/lib/bdinfo.c           |  51 ++++
 arch/m68k/lib/Makefile          |   1 +
 arch/m68k/lib/bdinfo.c          |  34 +++
 arch/nds32/include/asm/u-boot.h |  20 +-
 arch/powerpc/lib/Makefile       |   2 +
 arch/powerpc/lib/bdinfo.c       |  46 ++++
 arch/riscv/include/asm/u-boot.h |  19 +-
 arch/x86/lib/fsp/fsp_graphics.c |   5 +-
 cmd/bdinfo.c                    | 451 +++-----------------------------
 common/board_r.c                |  16 +-
 drivers/video/video-uclass.c    |   1 +
 include/asm-generic/u-boot.h    |  18 --
 include/init.h                  |   9 +
 14 files changed, 185 insertions(+), 489 deletions(-)
 create mode 100644 arch/arm/lib/bdinfo.c
 create mode 100644 arch/m68k/lib/bdinfo.c
 create mode 100644 arch/powerpc/lib/bdinfo.c
Signature · 1 line
2.26.2.526.g744177e7f7-goog
#1 lore raw

PATCH 1/36 bdinfo: nds32: Use generic bd_info

At present nds32 still uses its own private bd_info struct. Move it over to use the generic one like other archs.

---

 arch/nds32/include/asm/u-boot.h | 20 ++------------------
 1 file changed, 2 insertions(+), 18 deletions(-)
arch/nds32/include/asm/u-boot.h +2 -18
index 8c949e7fb7..7b6e905f2a 100644
@@ -19,26 +19,10 @@
 #ifndef _U_BOOT_H_
 #define _U_BOOT_H_	1
 
+/* Use the generic board which requires a unified bd_info */
+#include <asm-generic/u-boot.h>
 #include <asm/u-boot-nds32.h>
 
-
-typedef struct bd_info {
-	unsigned long	bi_arch_number;	/* unique id for this board */
-	unsigned long	bi_boot_params;	/* where this board expects params */
-	unsigned long	bi_memstart;	/* start of DRAM memory */
-	unsigned long	bi_memsize;	/* size	 of DRAM memory in bytes */
-	unsigned long	bi_flashstart;	/* start of FLASH memory */
-	unsigned long	bi_flashsize;	/* size	 of FLASH memory */
-	unsigned long	bi_flashoffset; /* reserved area for startup monitor */
-	unsigned char	bi_enetaddr[6];
-
-	struct				/* RAM configuration */
-	{
-		unsigned long start;
-		unsigned long size;
-	} bi_dram[CONFIG_NR_DRAM_BANKS];
-} bd_t;
-
 /* For image.h:image_check_target_arch() */
 #define IH_ARCH_DEFAULT IH_ARCH_NDS32
 
Signature · 1 line
2.26.2.526.g744177e7f7-goog

A message that is not in the archive (sent off-list or not yet received): [email protected]

#3 lore raw

PATCH 1/36 bdinfo: nds32: Use generic bd_info

Quoted 11 lines

From: Simon Glass [mailto:sjg at chromium.org]
Sent: Tuesday, May 05, 2020 7:17 AM
To: U-Boot Mailing List
Cc: Tom Rini; Simon Glass; Joe Hershberger; Macpaul Lin; Rick Jian-Zhi Chen(???)
Subject: [PATCH 01/36] bdinfo: nds32: Use generic bd_info

At present nds32 still uses its own private bd_info struct. Move it over to use the generic one like other archs.

---

Quoted 36 lines
 arch/nds32/include/asm/u-boot.h | 20 ++------------------
 1 file changed, 2 insertions(+), 18 deletions(-)
arch/nds32/include/asm/u-boot.h +2 -19
@@ -19,26 +19,10 @@
 #ifndef _U_BOOT_H_
 #define _U_BOOT_H_     1
 
+/* Use the generic board which requires a unified bd_info */ #include
+<asm-generic/u-boot.h>
 #include <asm/u-boot-nds32.h>
 
-
-typedef struct bd_info {
-       unsigned long   bi_arch_number; /* unique id for this board */
-       unsigned long   bi_boot_params; /* where this board expects params */
-       unsigned long   bi_memstart;    /* start of DRAM memory */
-       unsigned long   bi_memsize;     /* size  of DRAM memory in bytes */
-       unsigned long   bi_flashstart;  /* start of FLASH memory */
-       unsigned long   bi_flashsize;   /* size  of FLASH memory */
-       unsigned long   bi_flashoffset; /* reserved area for startup monitor */
-       unsigned char   bi_enetaddr[6];
-
-       struct                          /* RAM configuration */
-       {
-               unsigned long start;
-               unsigned long size;
-       } bi_dram[CONFIG_NR_DRAM_BANKS];
-} bd_t;
-
 /* For image.h:image_check_target_arch() */  #define IH_ARCH_DEFAULT IH_ARCH_NDS32
 
--

2.26.2.526.g744177e7f7-goog

#5 lore raw

PATCH 2/36 bdinfo: riscv: Use generic bd_info

In reply to Simon Glass

At present riscv still uses its own private bd_info struct. Move it over to use the generic one like other archs.

---

 arch/riscv/include/asm/u-boot.h | 19 ++-----------------
 1 file changed, 2 insertions(+), 17 deletions(-)
arch/riscv/include/asm/u-boot.h +2 -17
index 5ba8e77812..d5e1d5f323 100644
@@ -18,25 +18,10 @@
 #ifndef _U_BOOT_H_
 #define _U_BOOT_H_	1
 
+/* Use the generic board which requires a unified bd_info */
+#include <asm-generic/u-boot.h>
 #include <asm/u-boot-riscv.h>
 
-
-typedef struct bd_info {
-	unsigned long	bi_boot_params;	/* where this board expects params */
-	unsigned long	bi_memstart;	/* start of DRAM memory */
-	unsigned long	bi_memsize;	/* size	 of DRAM memory in bytes */
-	unsigned long	bi_flashstart;	/* start of FLASH memory */
-	unsigned long	bi_flashsize;	/* size	 of FLASH memory */
-	unsigned long	bi_flashoffset; /* reserved area for startup monitor */
-	unsigned char	bi_enetaddr[6];
-
-	struct				/* RAM configuration */
-	{
-		unsigned long start;
-		unsigned long size;
-	} bi_dram[CONFIG_NR_DRAM_BANKS];
-} bd_t;
-
 /* For image.h:image_check_target_arch() */
 #define IH_ARCH_DEFAULT IH_ARCH_RISCV
 
Signature · 1 line
2.26.2.526.g744177e7f7-goog

A message that is not in the archive (sent off-list or not yet received): [email protected]

#7 lore raw

PATCH 2/36 bdinfo: riscv: Use generic bd_info

Quoted 11 lines

From: Simon Glass [mailto:sjg at chromium.org]
Sent: Tuesday, May 05, 2020 7:17 AM
To: U-Boot Mailing List
Cc: Tom Rini; Simon Glass; Joe Hershberger; Rick Jian-Zhi Chen(???)
Subject: [PATCH 02/36] bdinfo: riscv: Use generic bd_info

At present riscv still uses its own private bd_info struct. Move it over to use the generic one like other archs.

---

Quoted 34 lines
 arch/riscv/include/asm/u-boot.h | 19 ++-----------------
 1 file changed, 2 insertions(+), 17 deletions(-)
arch/riscv/include/asm/u-boot.h +2 -18
@@ -18,25 +18,10 @@
 #ifndef _U_BOOT_H_
 #define _U_BOOT_H_     1
 
+/* Use the generic board which requires a unified bd_info */ #include
+<asm-generic/u-boot.h>
 #include <asm/u-boot-riscv.h>
 
-
-typedef struct bd_info {
-       unsigned long   bi_boot_params; /* where this board expects params */
-       unsigned long   bi_memstart;    /* start of DRAM memory */
-       unsigned long   bi_memsize;     /* size  of DRAM memory in bytes */
-       unsigned long   bi_flashstart;  /* start of FLASH memory */
-       unsigned long   bi_flashsize;   /* size  of FLASH memory */
-       unsigned long   bi_flashoffset; /* reserved area for startup monitor */
-       unsigned char   bi_enetaddr[6];
-
-       struct                          /* RAM configuration */
-       {
-               unsigned long start;
-               unsigned long size;
-       } bi_dram[CONFIG_NR_DRAM_BANKS];
-} bd_t;
-
 /* For image.h:image_check_target_arch() */  #define IH_ARCH_DEFAULT IH_ARCH_RISCV
 
--
#9 lore raw

PATCH 3/36 bdinfo: m68k: Drop bd_info->bi_ipbfreq

In reply to Simon Glass

This field is not used anymore. Drop it.

---

 include/asm-generic/u-boot.h | 1 -
 1 file changed, 1 deletion(-)
include/asm-generic/u-boot.h +0 -1
index cc94d39069..be0014c3b3 100644
@@ -60,7 +60,6 @@ typedef struct bd_info {
 	unsigned long	bi_vco;		/* VCO Out from PLL, in MHz */
 #endif
 #if defined(CONFIG_M68K)
-	unsigned long	bi_ipbfreq;	/* IPB Bus Freq, in MHz */
 	unsigned long	bi_pcifreq;	/* PCI Bus Freq, in MHz */
 #endif
 #if defined(CONFIG_EXTRA_CLOCK)
Signature · 1 line
2.26.2.526.g744177e7f7-goog
#11 lore raw

PATCH 4/36 bdinfo: xtensa: Create a generic do_bdinfo for xtensa

In reply to Simon Glass

This arch uses only the generic function. It would be nice if all the archs did the same. As a first step, create a new generic function for the 'bd' command and make xtensa use it.

---

 cmd/bdinfo.c | 14 ++++++++++----
 1 file changed, 10 insertions(+), 4 deletions(-)
cmd/bdinfo.c +10 -4
index 9edcf8d74e..9247180a29 100644
@@ -474,14 +474,20 @@ int do_bdinfo(cmd_tbl_t *cmdtp, int flag, int argc, char * const argv[])
 
 #elif defined(CONFIG_XTENSA)
 
-int do_bdinfo(cmd_tbl_t *cmdtp, int flag, int argc, char * const argv[])
+#define USE_GENERIC
+
+#else
+ #error "a case for this architecture does not exist!"
+#endif
+
+/* Temporary check for archs that use generic bdinfo. Eventually all will */
+#ifdef USE_GENERIC
+int do_bdinfo(cmd_tbl_t *cmdtp, int flag, int argc, char *const argv[])
 {
 	print_std_bdinfo(gd->bd);
+
 	return 0;
 }
-
-#else
- #error "a case for this architecture does not exist!"
 #endif
 
 /* -------------------------------------------------------------------- */
Signature · 1 line
2.26.2.526.g744177e7f7-goog
#12 lore raw

PATCH 4/36 bdinfo: xtensa: Create a generic do_bdinfo for xtensa

Quoted 11 lines · Simon Glass

This arch uses only the generic function. It would be nice if all the archs did the same. As a first step, create a new generic function for the 'bd' command and make xtensa use it.

---

 cmd/bdinfo.c | 14 ++++++++++----
 1 file changed, 10 insertions(+), 4 deletions(-)
#13 lore raw

PATCH 5/36 bdinfo: mips: Use the generic bd command

In reply to Simon Glass

MIPS currently has a few extra things which are generally useful. Add them to the generic function and move MIPS over to use it.

---

 cmd/bdinfo.c | 13 ++++---------
 1 file changed, 4 insertions(+), 9 deletions(-)
cmd/bdinfo.c +4 -9
index 9247180a29..6ccbd2f50f 100644
@@ -299,15 +299,7 @@ int do_bdinfo(cmd_tbl_t *cmdtp, int flag, int argc, char * const argv[])
 
 #elif defined(CONFIG_MIPS)
 
-int do_bdinfo(cmd_tbl_t *cmdtp, int flag, int argc, char * const argv[])
-{
-	print_std_bdinfo(gd->bd);
-	print_num("relocaddr", gd->relocaddr);
-	print_num("reloc off", gd->reloc_off);
-	print_cpu_word_size();
-
-	return 0;
-}
+#define USE_GENERIC
 
 #elif defined(CONFIG_ARM)
 
@@ -485,6 +477,9 @@ int do_bdinfo(cmd_tbl_t *cmdtp, int flag, int argc, char * const argv[])
 int do_bdinfo(cmd_tbl_t *cmdtp, int flag, int argc, char *const argv[])
 {
 	print_std_bdinfo(gd->bd);
+	print_num("relocaddr", gd->relocaddr);
+	print_num("reloc off", gd->reloc_off);
+	print_cpu_word_size();
 
 	return 0;
 }
Signature · 1 line
2.26.2.526.g744177e7f7-goog
#14 lore raw

PATCH 5/36 bdinfo: mips: Use the generic bd command

Quoted 10 lines · Simon Glass

MIPS currently has a few extra things which are generally useful. Add them to the generic function and move MIPS over to use it.

---

 cmd/bdinfo.c | 13 ++++---------
 1 file changed, 4 insertions(+), 9 deletions(-)
#15 lore raw

PATCH 5/36 bdinfo: mips: Use the generic bd command

In reply to Simon Glass

Am 05.05.20 um 01:17 schrieb Simon Glass:

Quoted 8 lines

MIPS currently has a few extra things which are generally useful. Add them to the generic function and move MIPS over to use it.

---

 cmd/bdinfo.c | 13 ++++---------
 1 file changed, 4 insertions(+), 9 deletions(-)
Quoted 33 lines
cmd/bdinfo.c +4 -9
index 9247180a29..6ccbd2f50f 100644
@@ -299,15 +299,7 @@ int do_bdinfo(cmd_tbl_t *cmdtp, int flag, int argc, char * const argv[])
 
 #elif defined(CONFIG_MIPS)
 
-int do_bdinfo(cmd_tbl_t *cmdtp, int flag, int argc, char * const argv[])
-{
-	print_std_bdinfo(gd->bd);
-	print_num("relocaddr", gd->relocaddr);
-	print_num("reloc off", gd->reloc_off);
-	print_cpu_word_size();
-
-	return 0;
-}
+#define USE_GENERIC
 
 #elif defined(CONFIG_ARM)
 
@@ -485,6 +477,9 @@ int do_bdinfo(cmd_tbl_t *cmdtp, int flag, int argc, char * const argv[])
 int do_bdinfo(cmd_tbl_t *cmdtp, int flag, int argc, char *const argv[])
 {
 	print_std_bdinfo(gd->bd);
+	print_num("relocaddr", gd->relocaddr);
+	print_num("reloc off", gd->reloc_off);
+	print_cpu_word_size();
 
 	return 0;
 }
Signature · 1 line
- Daniel
#16 lore raw

PATCH 6/36 bdinfo: nios2: Use the generic bd command

In reply to Simon Glass

Nios2 currently has some code to output SRAM information which is behind an #ifdef. No nios2 boards define this option, so the code can be removed.

Move Nios2 over to use the generic function.

---

 cmd/bdinfo.c | 19 +------------------
 1 file changed, 1 insertion(+), 18 deletions(-)
cmd/bdinfo.c +1 -18
index 6ccbd2f50f..405a915861 100644
@@ -220,24 +220,7 @@ int do_bdinfo(cmd_tbl_t *cmdtp, int flag, int argc, char * const argv[])
 
 #elif defined(CONFIG_NIOS2)
 
-int do_bdinfo(cmd_tbl_t *cmdtp, int flag, int argc, char * const argv[])
-{
-	bd_t *bd = gd->bd;
-
-	print_bi_dram(bd);
-	print_bi_flash(bd);
-
-#if defined(CONFIG_SYS_SRAM_BASE)
-	print_num ("sram start",	(ulong)bd->bi_sramstart);
-	print_num ("sram size",		(ulong)bd->bi_sramsize);
-#endif
-
-	print_eth_ip_addr();
-	print_baudrate();
-	print_cpu_word_size();
-
-	return 0;
-}
+#define USE_GENERIC
 
 #elif defined(CONFIG_MICROBLAZE)
 
Signature · 1 line
2.26.2.526.g744177e7f7-goog
#17 lore raw

PATCH 6/36 bdinfo: nios2: Use the generic bd command

Quoted 4 lines · Simon Glass

Nios2 currently has some code to output SRAM information which is behind an #ifdef. No nios2 boards define this option, so the code can be removed.

Quoted 8 lines

Move Nios2 over to use the generic function.

---

 cmd/bdinfo.c | 19 +------------------
 1 file changed, 1 insertion(+), 18 deletions(-)
#18 lore raw

PATCH 7/36 bdinfo: microblaze: Use the generic bd command

In reply to Simon Glass

Microblaze prints out ethernet and FDT information. This is useful to most archs, so move it into the generic code and move microblaze over to use it.

---

 cmd/bdinfo.c | 30 +++++++-----------------------
 1 file changed, 7 insertions(+), 23 deletions(-)
cmd/bdinfo.c +7 -23
index 405a915861..dc5a09f8ce 100644
@@ -224,29 +224,7 @@ int do_bdinfo(cmd_tbl_t *cmdtp, int flag, int argc, char * const argv[])
 
 #elif defined(CONFIG_MICROBLAZE)
 
-int do_bdinfo(cmd_tbl_t *cmdtp, int flag, int argc, char * const argv[])
-{
-	bd_t *bd = gd->bd;
-
-	print_bi_dram(bd);
-	print_bi_flash(bd);
-#if defined(CONFIG_SYS_SRAM_BASE)
-	print_num("sram start     ",	(ulong)bd->bi_sramstart);
-	print_num("sram size      ",	(ulong)bd->bi_sramsize);
-#endif
-#if defined(CONFIG_CMD_NET) && !defined(CONFIG_DM_ETH)
-	print_eths();
-#endif
-	print_baudrate();
-	print_num("relocaddr", gd->relocaddr);
-	print_num("reloc off", gd->reloc_off);
-	print_num("fdt_blob", (ulong)gd->fdt_blob);
-	print_num("new_fdt", (ulong)gd->new_fdt);
-	print_num("fdt_size", (ulong)gd->fdt_size);
-	print_cpu_word_size();
-
-	return 0;
-}
+#define USE_GENERIC
 
 #elif defined(CONFIG_M68K)
 
@@ -463,6 +441,12 @@ int do_bdinfo(cmd_tbl_t *cmdtp, int flag, int argc, char *const argv[])
 	print_num("relocaddr", gd->relocaddr);
 	print_num("reloc off", gd->reloc_off);
 	print_cpu_word_size();
+#if defined(CONFIG_CMD_NET) && !defined(CONFIG_DM_ETH)
+	print_eths();
+#endif
+	print_num("fdt_blob", (ulong)gd->fdt_blob);
+	print_num("new_fdt", (ulong)gd->new_fdt);
+	print_num("fdt_size", (ulong)gd->fdt_size);
 
 	return 0;
 }
Signature · 1 line
2.26.2.526.g744177e7f7-goog
#19 lore raw

PATCH 7/36 bdinfo: microblaze: Use the generic bd command

Quoted 11 lines · Simon Glass

Microblaze prints out ethernet and FDT information. This is useful to most archs, so move it into the generic code and move microblaze over to use it.

---

 cmd/bdinfo.c | 30 +++++++-----------------------
 1 file changed, 7 insertions(+), 23 deletions(-)
#20 lore raw

PATCH 7/36 bdinfo: microblaze: Use the generic bd command

In reply to Simon Glass

Am 05.05.20 um 01:17 schrieb Simon Glass:

Quoted 46 lines

Microblaze prints out ethernet and FDT information. This is useful to most archs, so move it into the generic code and move microblaze over to use it.

---

 cmd/bdinfo.c | 30 +++++++-----------------------
 1 file changed, 7 insertions(+), 23 deletions(-)
cmd/bdinfo.c +1 -23
index 405a915861..dc5a09f8ce 100644
@@ -224,29 +224,7 @@ int do_bdinfo(cmd_tbl_t *cmdtp, int flag, int argc, char * const argv[])
 
 #elif defined(CONFIG_MICROBLAZE)
 
-int do_bdinfo(cmd_tbl_t *cmdtp, int flag, int argc, char * const argv[])
-{
-	bd_t *bd = gd->bd;
-
-	print_bi_dram(bd);
-	print_bi_flash(bd);
-#if defined(CONFIG_SYS_SRAM_BASE)
-	print_num("sram start     ",	(ulong)bd->bi_sramstart);
-	print_num("sram size      ",	(ulong)bd->bi_sramsize);
-#endif
-#if defined(CONFIG_CMD_NET) && !defined(CONFIG_DM_ETH)
-	print_eths();
-#endif
-	print_baudrate();
-	print_num("relocaddr", gd->relocaddr);
-	print_num("reloc off", gd->reloc_off);
-	print_num("fdt_blob", (ulong)gd->fdt_blob);
-	print_num("new_fdt", (ulong)gd->new_fdt);
-	print_num("fdt_size", (ulong)gd->fdt_size);
-	print_cpu_word_size();
-
-	return 0;
-}
+#define USE_GENERIC
 
 #elif defined(CONFIG_M68K)
 

cmd/bdinfo.c int do_bdinfo(cmd_tbl_t *cmdtp, int flag, int argc, char *const argv[]) · line 449

Quoted 9 lines
cmd/bdinfo.c +6 -0
@@ -463,6 +441,12 @@ int do_bdinfo(cmd_tbl_t *cmdtp, int flag, int argc, char *const argv[])
 	print_num("relocaddr", gd->relocaddr);
 	print_num("reloc off", gd->reloc_off);
 	print_cpu_word_size();
+#if defined(CONFIG_CMD_NET) && !defined(CONFIG_DM_ETH)
+	print_eths();
+#endif
+	print_num("fdt_blob", (ulong)gd->fdt_blob);
+	print_num("new_fdt", (ulong)gd->new_fdt);
+	print_num("fdt_size", (ulong)gd->fdt_size);

shouldn't that be guarded with CONFIG_IS_ENABLED(OF_LIBFDT) ?

Quoted 4 lines
cmd/bdinfo.c +0 -0
 	return 0;
 }
 
Signature · 1 line
- Daniel
#21 lore raw

PATCH 7/36 bdinfo: microblaze: Use the generic bd command

Hi Daniel,

Quoted 49 lines · Daniel Schwierzeck

Am 05.05.20 um 01:17 schrieb Simon Glass:

Quoted 45 lines

Microblaze prints out ethernet and FDT information. This is useful to most archs, so move it into the generic code and move microblaze over to use it.

---

 cmd/bdinfo.c | 30 +++++++-----------------------
 1 file changed, 7 insertions(+), 23 deletions(-)
cmd/bdinfo.c +1 -23
index 405a915861..dc5a09f8ce 100644
@@ -224,29 +224,7 @@ int do_bdinfo(cmd_tbl_t *cmdtp, int flag, int argc, char * const argv[])
 
 #elif defined(CONFIG_MICROBLAZE)
 
-int do_bdinfo(cmd_tbl_t *cmdtp, int flag, int argc, char * const argv[])
-{
-     bd_t *bd = gd->bd;
-
-     print_bi_dram(bd);
-     print_bi_flash(bd);
-#if defined(CONFIG_SYS_SRAM_BASE)
-     print_num("sram start     ",    (ulong)bd->bi_sramstart);
-     print_num("sram size      ",    (ulong)bd->bi_sramsize);
-#endif
-#if defined(CONFIG_CMD_NET) && !defined(CONFIG_DM_ETH)
-     print_eths();
-#endif
-     print_baudrate();
-     print_num("relocaddr", gd->relocaddr);
-     print_num("reloc off", gd->reloc_off);
-     print_num("fdt_blob", (ulong)gd->fdt_blob);
-     print_num("new_fdt", (ulong)gd->new_fdt);
-     print_num("fdt_size", (ulong)gd->fdt_size);
-     print_cpu_word_size();
-
-     return 0;
-}
+#define USE_GENERIC
 
 #elif defined(CONFIG_M68K)
 
Quoted 12 lines

cmd/bdinfo.c int do_bdinfo(cmd_tbl_t *cmdtp, int flag, int argc, char *const argv[]) · line 449

Quoted 10 lines
cmd/bdinfo.c +6 -0
@@ -463,6 +441,12 @@ int do_bdinfo(cmd_tbl_t *cmdtp, int flag, int argc, char *const argv[])
      print_num("relocaddr", gd->relocaddr);
      print_num("reloc off", gd->reloc_off);
      print_cpu_word_size();
+#if defined(CONFIG_CMD_NET) && !defined(CONFIG_DM_ETH)
+     print_eths();
+#endif
+     print_num("fdt_blob", (ulong)gd->fdt_blob);
+     print_num("new_fdt", (ulong)gd->new_fdt);
+     print_num("fdt_size", (ulong)gd->fdt_size);

shouldn't that be guarded with CONFIG_IS_ENABLED(OF_LIBFDT) ?

No it doesn't need to be. The commands are not available in SPL, and U-Boot proper is supposed to use device tree for all boards.

Regards,
Simon

#22 lore raw

PATCH 8/36 bdinfo: sh: Use the generic bd command

In reply to Simon Glass

This arch has no code that is not already in the generic function. Drop the arch-specific function and change sh over to use the generic one.

---

 cmd/bdinfo.c | 13 +------------
 1 file changed, 1 insertion(+), 12 deletions(-)
cmd/bdinfo.c +1 -12
index dc5a09f8ce..0b9ac21848 100644
@@ -325,18 +325,7 @@ static int do_bdinfo(cmd_tbl_t *cmdtp, int flag, int argc,
 
 #elif defined(CONFIG_SH)
 
-int do_bdinfo(cmd_tbl_t *cmdtp, int flag, int argc, char * const argv[])
-{
-	bd_t *bd = gd->bd;
-
-	print_bi_mem(bd);
-	print_bi_flash(bd);
-	print_eth_ip_addr();
-	print_baudrate();
-	print_cpu_word_size();
-
-	return 0;
-}
+#define USE_GENERIC
 
 #elif defined(CONFIG_X86)
 
Signature · 1 line
2.26.2.526.g744177e7f7-goog
#23 lore raw

PATCH 8/36 bdinfo: sh: Use the generic bd command

Quoted 10 lines · Simon Glass

This arch has no code that is not already in the generic function. Drop the arch-specific function and change sh over to use the generic one.

---

 cmd/bdinfo.c | 13 +------------
 1 file changed, 1 insertion(+), 12 deletions(-)
#24 lore raw

PATCH 9/36 bdinfo: x86: Use the generic bd command

In reply to Simon Glass

This arch shows 'ethspeed' info but only the freescale drivers use it, so it can be dropped.

It also calls print_bi_dram() which is safe to call from any arch since it has an #ifdef inside it. Add this to the generic do_bdinfo() and move x86 over to use it. Put it first since pytests rely on seeing it before memstart in find_ram_base().

---

 cmd/bdinfo.c | 21 ++-------------------
 1 file changed, 2 insertions(+), 19 deletions(-)
cmd/bdinfo.c +2 -19
index 0b9ac21848..0b903d234e 100644
@@ -329,25 +329,7 @@ static int do_bdinfo(cmd_tbl_t *cmdtp, int flag, int argc,
 
 #elif defined(CONFIG_X86)
 
-int do_bdinfo(cmd_tbl_t *cmdtp, int flag, int argc, char * const argv[])
-{
-	bd_t *bd = gd->bd;
-
-	print_bi_boot_params(bd);
-
-	print_bi_dram(bd);
-
-	print_num("relocaddr", gd->relocaddr);
-	print_num("reloc off", gd->reloc_off);
-#if defined(CONFIG_CMD_NET)
-	print_eth_ip_addr();
-	print_mhz("ethspeed",	    bd->bi_ethspeed);
-#endif
-	print_baudrate();
-	print_cpu_word_size();
-
-	return 0;
-}
+#define USE_GENERIC
 
 #elif defined(CONFIG_SANDBOX)
 
@@ -426,6 +408,7 @@ int do_bdinfo(cmd_tbl_t *cmdtp, int flag, int argc, char * const argv[])
 #ifdef USE_GENERIC
 int do_bdinfo(cmd_tbl_t *cmdtp, int flag, int argc, char *const argv[])
 {
+	print_bi_dram(gd->bd);
 	print_std_bdinfo(gd->bd);
 	print_num("relocaddr", gd->relocaddr);
 	print_num("reloc off", gd->reloc_off);
Signature · 1 line
2.26.2.526.g744177e7f7-goog
#25 lore raw

PATCH 9/36 bdinfo: x86: Use the generic bd command

Quoted 9 lines · Simon Glass

This arch shows 'ethspeed' info but only the freescale drivers use it, so it can be dropped.

It also calls print_bi_dram() which is safe to call from any arch since it has an #ifdef inside it. Add this to the generic do_bdinfo() and move x86 over to use it. Put it first since pytests rely on seeing it before memstart in find_ram_base().

Quoted 6 lines

---

 cmd/bdinfo.c | 21 ++-------------------
 1 file changed, 2 insertions(+), 19 deletions(-)
#26 lore raw

PATCH 10/36 bdinfo: sandbox: Use the generic bd command

In reply to Simon Glass

Sandbox has a printout of 'FB base' but this code is not used since sandbox uses driver model for everything.

Move sandbox over to use the generic do_bdinfo().

---

 cmd/bdinfo.c | 16 +---------------
 1 file changed, 1 insertion(+), 15 deletions(-)
cmd/bdinfo.c +1 -15
index 0b903d234e..dce24b43fa 100644
@@ -333,21 +333,7 @@ static int do_bdinfo(cmd_tbl_t *cmdtp, int flag, int argc,
 
 #elif defined(CONFIG_SANDBOX)
 
-int do_bdinfo(cmd_tbl_t *cmdtp, int flag, int argc, char * const argv[])
-{
-	bd_t *bd = gd->bd;
-
-	print_bi_boot_params(bd);
-	print_bi_dram(bd);
-	print_eth_ip_addr();
-
-#if defined(CONFIG_LCD) || defined(CONFIG_VIDEO)
-	print_num("FB base  ", gd->fb_base);
-#endif
-	print_cpu_word_size();
-
-	return 0;
-}
+#define USE_GENERIC
 
 #elif defined(CONFIG_NDS32)
 
Signature · 1 line
2.26.2.526.g744177e7f7-goog
#27 lore raw

PATCH 10/36 bdinfo: sandbox: Use the generic bd command

Quoted 4 lines · Simon Glass

Sandbox has a printout of 'FB base' but this code is not used since sandbox uses driver model for everything.

Quoted 8 lines

Move sandbox over to use the generic do_bdinfo().

---

 cmd/bdinfo.c | 16 +---------------
 1 file changed, 1 insertion(+), 15 deletions(-)
#28 lore raw

PATCH 11/36 bdinfo: nds32: Use the generic bd command

In reply to Simon Glass

This arch has none of its own info to show. Move it over to use the generic do_bdinfo().

---

 cmd/bdinfo.c | 14 +-------------
 1 file changed, 1 insertion(+), 13 deletions(-)
cmd/bdinfo.c +1 -13
index dce24b43fa..6a2bbfb348 100644
@@ -337,19 +337,7 @@ static int do_bdinfo(cmd_tbl_t *cmdtp, int flag, int argc,
 
 #elif defined(CONFIG_NDS32)
 
-int do_bdinfo(cmd_tbl_t *cmdtp, int flag, int argc, char * const argv[])
-{
-	bd_t *bd = gd->bd;
-
-	print_num("arch_number",	bd->bi_arch_number);
-	print_bi_boot_params(bd);
-	print_bi_dram(bd);
-	print_eth_ip_addr();
-	print_baudrate();
-	print_cpu_word_size();
-
-	return 0;
-}
+#define USE_GENERIC
 
 #elif defined(CONFIG_RISCV)
 
Signature · 1 line
2.26.2.526.g744177e7f7-goog

A message that is not in the archive (sent off-list or not yet received): [email protected]

#30 lore raw

PATCH 11/36 bdinfo: nds32: Use the generic bd command

Quoted 11 lines

From: U-Boot [mailto:u-boot-bounces at lists.denx.de] On Behalf Of Simon Glass
Sent: Tuesday, May 05, 2020 7:17 AM
To: U-Boot Mailing List
Cc: Tom Rini; Simon Glass; Bin Meng; Heiko Schocher; Macpaul Lin
Subject: [PATCH 11/36] bdinfo: nds32: Use the generic bd command

This arch has none of its own info to show. Move it over to use the generic do_bdinfo().

---

Quoted 28 lines
 cmd/bdinfo.c | 14 +-------------
 1 file changed, 1 insertion(+), 13 deletions(-)
cmd/bdinfo.c +1 -13
@@ -337,19 +337,7 @@ static int do_bdinfo(cmd_tbl_t *cmdtp, int flag, int argc,
 
 #elif defined(CONFIG_NDS32)
 
-int do_bdinfo(cmd_tbl_t *cmdtp, int flag, int argc, char * const argv[]) -{
-       bd_t *bd = gd->bd;
-
-       print_num("arch_number",        bd->bi_arch_number);
-       print_bi_boot_params(bd);
-       print_bi_dram(bd);
-       print_eth_ip_addr();
-       print_baudrate();
-       print_cpu_word_size();
-
-       return 0;
-}
+#define USE_GENERIC
 
 #elif defined(CONFIG_RISCV)
 
--

2.26.2.526.g744177e7f7-goog

#32 lore raw

PATCH 12/36 bdinfo: riscv: Use the generic bd command

In reply to Simon Glass

This arch has none of its own info to show. Move it over to use the generic do_bdinfo().

---

 cmd/bdinfo.c | 15 +--------------
 1 file changed, 1 insertion(+), 14 deletions(-)
cmd/bdinfo.c +1 -14
index 6a2bbfb348..0e96dd619c 100644
@@ -341,20 +341,7 @@ static int do_bdinfo(cmd_tbl_t *cmdtp, int flag, int argc,
 
 #elif defined(CONFIG_RISCV)
 
-int do_bdinfo(cmd_tbl_t *cmdtp, int flag, int argc, char * const argv[])
-{
-	bd_t *bd = gd->bd;
-
-	print_bi_boot_params(bd);
-	print_bi_dram(bd);
-	print_num("relocaddr", gd->relocaddr);
-	print_num("reloc off", gd->reloc_off);
-	print_eth_ip_addr();
-	print_baudrate();
-	print_cpu_word_size();
-
-	return 0;
-}
+#define USE_GENERIC
 
 #elif defined(CONFIG_ARC)
 
Signature · 1 line
2.26.2.526.g744177e7f7-goog

A message that is not in the archive (sent off-list or not yet received): [email protected]

#34 lore raw

PATCH 12/36 bdinfo: riscv: Use the generic bd command

Quoted 11 lines

From: U-Boot [mailto:u-boot-bounces at lists.denx.de] On Behalf Of Simon Glass
Sent: Tuesday, May 05, 2020 7:17 AM
To: U-Boot Mailing List
Cc: Tom Rini; Simon Glass; Bin Meng; Heiko Schocher
Subject: [PATCH 12/36] bdinfo: riscv: Use the generic bd command

This arch has none of its own info to show. Move it over to use the generic do_bdinfo().

---

Quoted 29 lines
 cmd/bdinfo.c | 15 +--------------
 1 file changed, 1 insertion(+), 14 deletions(-)
cmd/bdinfo.c +1 -14
@@ -341,20 +341,7 @@ static int do_bdinfo(cmd_tbl_t *cmdtp, int flag, int argc,
 
 #elif defined(CONFIG_RISCV)
 
-int do_bdinfo(cmd_tbl_t *cmdtp, int flag, int argc, char * const argv[]) -{
-       bd_t *bd = gd->bd;
-
-       print_bi_boot_params(bd);
-       print_bi_dram(bd);
-       print_num("relocaddr", gd->relocaddr);
-       print_num("reloc off", gd->reloc_off);
-       print_eth_ip_addr();
-       print_baudrate();
-       print_cpu_word_size();
-
-       return 0;
-}
+#define USE_GENERIC
 
 #elif defined(CONFIG_ARC)
 
--

2.26.2.526.g744177e7f7-goog

#36 lore raw

PATCH 13/36 bdinfo: arm: Use the generic bd command

In reply to Simon Glass

Unfortunately ARM has a lot of special stuff. Move it into the generic function for now, so we can have it all in one place.

Add in the frame-buffer printout, since it supports driver model too.

---

 cmd/bdinfo.c | 250 +++++++++++++++++++++------------------------------
 1 file changed, 104 insertions(+), 146 deletions(-)
cmd/bdinfo.c +104 -146
index 0e96dd619c..f07f92463c 100644
@@ -170,53 +170,8 @@ static inline void __maybe_unused print_std_bdinfo(const bd_t *bd)
 }
 
 #if defined(CONFIG_PPC)
-void __weak board_detail(void)
-{
-	/* Please define board_detail() for your platform */
-}
-
-int do_bdinfo(cmd_tbl_t *cmdtp, int flag, int argc, char * const argv[])
-{
-	bd_t *bd = gd->bd;
-
-#ifdef DEBUG
-	print_num("bd address",		(ulong)bd);
-#endif
-	print_bi_mem(bd);
-	print_bi_flash(bd);
-	print_num("sramstart",		bd->bi_sramstart);
-	print_num("sramsize",		bd->bi_sramsize);
-#if	defined(CONFIG_MPC8xx) || defined(CONFIG_E500)
-	print_num("immr_base",		bd->bi_immr_base);
-#endif
-	print_num("bootflags",		bd->bi_bootflags);
-#if defined(CONFIG_CPM2)
-	print_mhz("vco",		bd->bi_vco);
-	print_mhz("sccfreq",		bd->bi_sccfreq);
-	print_mhz("brgfreq",		bd->bi_brgfreq);
-#endif
-	print_mhz("intfreq",		bd->bi_intfreq);
-#if defined(CONFIG_CPM2)
-	print_mhz("cpmfreq",		bd->bi_cpmfreq);
-#endif
-	print_mhz("busfreq",		bd->bi_busfreq);
-
-#ifdef CONFIG_ENABLE_36BIT_PHYS
-#ifdef CONFIG_PHYS_64BIT
-	puts("addressing  = 36-bit\n");
-#else
-	puts("addressing  = 32-bit\n");
-#endif
-#endif
-
-	print_eth_ip_addr();
-	print_baudrate();
-	print_num("relocaddr", gd->relocaddr);
-	board_detail();
-	print_cpu_word_size();
 
-	return 0;
-}
+#define USE_GENERIC
 
 #elif defined(CONFIG_NIOS2)
 
@@ -228,35 +183,7 @@ int do_bdinfo(cmd_tbl_t *cmdtp, int flag, int argc, char * const argv[])
 
 #elif defined(CONFIG_M68K)
 
-int do_bdinfo(cmd_tbl_t *cmdtp, int flag, int argc, char * const argv[])
-{
-	bd_t *bd = gd->bd;
-
-	print_bi_mem(bd);
-	print_bi_flash(bd);
-#if defined(CONFIG_SYS_INIT_RAM_ADDR)
-	print_num("sramstart",		(ulong)bd->bi_sramstart);
-	print_num("sramsize",		(ulong)bd->bi_sramsize);
-#endif
-#if defined(CONFIG_SYS_MBAR)
-	print_num("mbar",		bd->bi_mbar_base);
-#endif
-	print_mhz("cpufreq",		bd->bi_intfreq);
-	print_mhz("busfreq",		bd->bi_busfreq);
-#ifdef CONFIG_PCI
-	print_mhz("pcifreq",		bd->bi_pcifreq);
-#endif
-#ifdef CONFIG_EXTRA_CLOCK
-	print_mhz("flbfreq",		bd->bi_flbfreq);
-	print_mhz("inpfreq",		bd->bi_inpfreq);
-	print_mhz("vcofreq",		bd->bi_vcofreq);
-#endif
-	print_eth_ip_addr();
-	print_baudrate();
-	print_cpu_word_size();
-
-	return 0;
-}
+#define USE_GENERIC
 
 #elif defined(CONFIG_MIPS)
 
@@ -264,64 +191,7 @@ int do_bdinfo(cmd_tbl_t *cmdtp, int flag, int argc, char * const argv[])
 
 #elif defined(CONFIG_ARM)
 
-static int do_bdinfo(cmd_tbl_t *cmdtp, int flag, int argc,
-			char * const argv[])
-{
-	bd_t *bd = gd->bd;
-
-	print_num("arch_number",	bd->bi_arch_number);
-	print_bi_boot_params(bd);
-	print_bi_dram(bd);
-
-#ifdef CONFIG_SYS_MEM_RESERVE_SECURE
-	if (gd->arch.secure_ram & MEM_RESERVE_SECURE_SECURED) {
-		print_num("Secure ram",
-			  gd->arch.secure_ram & MEM_RESERVE_SECURE_ADDR_MASK);
-	}
-#endif
-#ifdef CONFIG_RESV_RAM
-	if (gd->arch.resv_ram)
-		print_num("Reserved ram", gd->arch.resv_ram);
-#endif
-#if defined(CONFIG_CMD_NET) && !defined(CONFIG_DM_ETH)
-	print_eths();
-#endif
-	print_baudrate();
-#if !(CONFIG_IS_ENABLED(SYS_ICACHE_OFF) && CONFIG_IS_ENABLED(SYS_DCACHE_OFF))
-	print_num("TLB addr", gd->arch.tlb_addr);
-#endif
-	print_num("relocaddr", gd->relocaddr);
-	print_num("reloc off", gd->reloc_off);
-	print_num("irq_sp", gd->irq_sp);	/* irq stack pointer */
-	print_num("sp start ", gd->start_addr_sp);
-#if defined(CONFIG_LCD) || defined(CONFIG_VIDEO) || defined(CONFIG_DM_VIDEO)
-	print_num("FB base  ", gd->fb_base);
-#endif
-	/*
-	 * TODO: Currently only support for davinci SOC's is added.
-	 * Remove this check once all the board implement this.
-	 */
-#ifdef CONFIG_CLOCKS
-	printf("ARM frequency = %ld MHz\n", gd->bd->bi_arm_freq);
-	printf("DSP frequency = %ld MHz\n", gd->bd->bi_dsp_freq);
-	printf("DDR frequency = %ld MHz\n", gd->bd->bi_ddr_freq);
-#endif
-#ifdef CONFIG_BOARD_TYPES
-	printf("Board Type  = %ld\n", gd->board_type);
-#endif
-#if CONFIG_VAL(SYS_MALLOC_F_LEN)
-	printf("Early malloc usage: %lx / %x\n", gd->malloc_ptr,
-	       CONFIG_VAL(SYS_MALLOC_F_LEN));
-#endif
-#if CONFIG_IS_ENABLED(MULTI_DTB_FIT)
-	print_num("multi_dtb_fit", (ulong)gd->multi_dtb_fit);
-#endif
-	if (gd->fdt_blob)
-		print_num("fdt_blob", (ulong)gd->fdt_blob);
-	print_cpu_word_size();
-
-	return 0;
-}
+#define USE_GENERIC
 
 #elif defined(CONFIG_SH)
 
@@ -345,17 +215,7 @@ static int do_bdinfo(cmd_tbl_t *cmdtp, int flag, int argc,
 
 #elif defined(CONFIG_ARC)
 
-int do_bdinfo(cmd_tbl_t *cmdtp, int flag, int argc, char * const argv[])
-{
-	bd_t *bd = gd->bd;
-
-	print_bi_mem(bd);
-	print_eth_ip_addr();
-	print_baudrate();
-	print_cpu_word_size();
-
-	return 0;
-}
+#define USE_GENERIC
 
 #elif defined(CONFIG_XTENSA)
 
@@ -367,10 +227,22 @@ int do_bdinfo(cmd_tbl_t *cmdtp, int flag, int argc, char * const argv[])
 
 /* Temporary check for archs that use generic bdinfo. Eventually all will */
 #ifdef USE_GENERIC
+void __weak board_detail(void)
+{
+	/* Please define board_detail() for your PPC platform */
+}
+
 int do_bdinfo(cmd_tbl_t *cmdtp, int flag, int argc, char *const argv[])
 {
-	print_bi_dram(gd->bd);
-	print_std_bdinfo(gd->bd);
+	bd_t *bd = gd->bd;
+
+#ifdef DEBUG
+	print_num("bd address", (ulong)bd);
+#endif
+	if (IS_ENABLED(CONFIG_ARM))
+		print_num("arch_number", bd->bi_arch_number);
+	print_bi_dram(bd);
+	print_std_bdinfo(bd);
 	print_num("relocaddr", gd->relocaddr);
 	print_num("reloc off", gd->reloc_off);
 	print_cpu_word_size();
@@ -380,6 +252,92 @@ int do_bdinfo(cmd_tbl_t *cmdtp, int flag, int argc, char *const argv[])
 	print_num("fdt_blob", (ulong)gd->fdt_blob);
 	print_num("new_fdt", (ulong)gd->new_fdt);
 	print_num("fdt_size", (ulong)gd->fdt_size);
+#if defined(CONFIG_LCD) || defined(CONFIG_VIDEO) || defined(CONFIG_DM_VIDEO)
+	print_num("FB base  ", gd->fb_base);
+#endif
+
+	/* This section is used only by ARM */
+#ifdef CONFIG_ARM
+#ifdef CONFIG_SYS_MEM_RESERVE_SECURE
+	if (gd->arch.secure_ram & MEM_RESERVE_SECURE_SECURED) {
+		print_num("Secure ram",
+			  gd->arch.secure_ram & MEM_RESERVE_SECURE_ADDR_MASK);
+	}
+#endif
+#ifdef CONFIG_RESV_RAM
+	if (gd->arch.resv_ram)
+		print_num("Reserved ram", gd->arch.resv_ram);
+#endif
+#if !(CONFIG_IS_ENABLED(SYS_ICACHE_OFF) && CONFIG_IS_ENABLED(SYS_DCACHE_OFF))
+	print_num("TLB addr", gd->arch.tlb_addr);
+#endif
+	print_num("irq_sp", gd->irq_sp);	/* irq stack pointer */
+	print_num("sp start ", gd->start_addr_sp);
+	/*
+	 * TODO: Currently only support for davinci SOC's is added.
+	 * Remove this check once all the board implement this.
+	 */
+#ifdef CONFIG_CLOCKS
+	printf("ARM frequency = %ld MHz\n", gd->bd->bi_arm_freq);
+	printf("DSP frequency = %ld MHz\n", gd->bd->bi_dsp_freq);
+	printf("DDR frequency = %ld MHz\n", gd->bd->bi_ddr_freq);
+#endif
+#ifdef CONFIG_BOARD_TYPES
+	printf("Board Type  = %ld\n", gd->board_type);
+#endif
+#if CONFIG_VAL(SYS_MALLOC_F_LEN)
+	printf("Early malloc usage: %lx / %x\n", gd->malloc_ptr,
+	       CONFIG_VAL(SYS_MALLOC_F_LEN));
+#endif
+#if CONFIG_IS_ENABLED(MULTI_DTB_FIT)
+	print_num("multi_dtb_fit", (ulong)gd->multi_dtb_fit);
+#endif
+#endif /* CONFIG_ARM */
+
+	/* This section is used only by ppc */
+#if defined(CONFIG_MPC8xx) || defined(CONFIG_E500)
+	print_num("immr_base", bd->bi_immr_base);
+#endif
+	if (IS_ENABLED(CONFIG_PPC)) {
+		print_num("bootflags", bd->bi_bootflags);
+		print_mhz("intfreq", bd->bi_intfreq);
+#ifdef CONFIG_ENABLE_36BIT_PHYS
+		if (IS_ENABLED(CONFIG_PHYS_64BIT))
+			puts("addressing  = 36-bit\n");
+		else
+			puts("addressing  = 32-bit\n");
+#endif
+		board_detail();
+	}
+#if defined(CONFIG_CPM2)
+	print_mhz("cpmfreq", bd->bi_cpmfreq);
+	print_mhz("vco", bd->bi_vco);
+	print_mhz("sccfreq", bd->bi_sccfreq);
+	print_mhz("brgfreq", bd->bi_brgfreq);
+#endif
+
+	/* This is used by m68k and ppc */
+#if defined(CONFIG_SYS_INIT_RAM_ADDR)
+	print_num("sramstart", (ulong)bd->bi_sramstart);
+	print_num("sramsize", (ulong)bd->bi_sramsize);
+#endif
+	if (IS_ENABLED(CONFIG_PPC) || IS_ENABLED(CONFIG_M68K))
+		print_mhz("busfreq", bd->bi_busfreq);
+
+	/* The rest are used only by m68k */
+#ifdef CONFIG_M68K
+#if defined(CONFIG_SYS_MBAR)
+	print_num("mbar", bd->bi_mbar_base);
+#endif
+	print_mhz("cpufreq", bd->bi_intfreq);
+	if (IS_ENABLED(CONFIG_PCI))
+		print_mhz("pcifreq", bd->bi_pcifreq);
+#ifdef CONFIG_EXTRA_CLOCK
+	print_mhz("flbfreq", bd->bi_flbfreq);
+	print_mhz("inpfreq", bd->bi_inpfreq);
+	print_mhz("vcofreq", bd->bi_vcofreq);
+#endif
+#endif
 
 	return 0;
 }
Signature · 1 line
2.26.2.526.g744177e7f7-goog
#37 lore raw

PATCH 13/36 bdinfo: arm: Use the generic bd command

Hi Simon,

Quoted 17 lines · Simon Glass

Unfortunately ARM has a lot of special stuff. Move it into the generic function for now, so we can have it all in one place.

Add in the frame-buffer printout, since it supports driver model too.

---

 cmd/bdinfo.c | 250 +++++++++++++++++++++------------------------------
 1 file changed, 104 insertions(+), 146 deletions(-)
cmd/bdinfo.c +0 -0
index 0e96dd619c..f07f92463c 100644

cmd/bdinfo.c static inline void __maybe_unused print_std_bdinfo(const bd_t *bd) · line 172

Quoted 3 lines
cmd/bdinfo.c +0 -0
@@ -170,53 +170,8 @@ static inline void __maybe_unused print_std_bdinfo(const bd_t *bd)
 }
 
 #if defined(CONFIG_PPC)

This commit has both PPC and ARM changes. Could them be split into 2 patches?

cmd/bdinfo.c static inline void __maybe_unused print_std_bdinfo(const bd_t *bd) · line 177

Quoted 5 lines
cmd/bdinfo.c +0 -5
-void __weak board_detail(void)
-{
-       /* Please define board_detail() for your platform */
-}
-

Regards,
Bin

#38 lore raw

PATCH 14/36 bdinfo: Drop the option to not use the generic 'bd' command

In reply to Simon Glass

Now that all architectures are using the generic do_bdinfo(), drop the option to not use it. When new architectures are added, they will get at least some useful information from the generic implementation.

---

 cmd/bdinfo.c | 59 ----------------------------------------------------
 1 file changed, 59 deletions(-)
cmd/bdinfo.c +0 -59
index f07f92463c..13b4f2ebba 100644
@@ -169,64 +169,6 @@ static inline void __maybe_unused print_std_bdinfo(const bd_t *bd)
 	print_baudrate();
 }
 
-#if defined(CONFIG_PPC)
-
-#define USE_GENERIC
-
-#elif defined(CONFIG_NIOS2)
-
-#define USE_GENERIC
-
-#elif defined(CONFIG_MICROBLAZE)
-
-#define USE_GENERIC
-
-#elif defined(CONFIG_M68K)
-
-#define USE_GENERIC
-
-#elif defined(CONFIG_MIPS)
-
-#define USE_GENERIC
-
-#elif defined(CONFIG_ARM)
-
-#define USE_GENERIC
-
-#elif defined(CONFIG_SH)
-
-#define USE_GENERIC
-
-#elif defined(CONFIG_X86)
-
-#define USE_GENERIC
-
-#elif defined(CONFIG_SANDBOX)
-
-#define USE_GENERIC
-
-#elif defined(CONFIG_NDS32)
-
-#define USE_GENERIC
-
-#elif defined(CONFIG_RISCV)
-
-#define USE_GENERIC
-
-#elif defined(CONFIG_ARC)
-
-#define USE_GENERIC
-
-#elif defined(CONFIG_XTENSA)
-
-#define USE_GENERIC
-
-#else
- #error "a case for this architecture does not exist!"
-#endif
-
-/* Temporary check for archs that use generic bdinfo. Eventually all will */
-#ifdef USE_GENERIC
 void __weak board_detail(void)
 {
 	/* Please define board_detail() for your PPC platform */
@@ -341,7 +283,6 @@ int do_bdinfo(cmd_tbl_t *cmdtp, int flag, int argc, char *const argv[])
 
 	return 0;
 }
-#endif
 
 /* -------------------------------------------------------------------- */
 
Signature · 1 line
2.26.2.526.g744177e7f7-goog
#39 lore raw

PATCH 14/36 bdinfo: Drop the option to not use the generic 'bd' command

Quoted 11 lines · Simon Glass

Now that all architectures are using the generic do_bdinfo(), drop the option to not use it. When new architectures are added, they will get at least some useful information from the generic implementation.

---

 cmd/bdinfo.c | 59 ----------------------------------------------------
 1 file changed, 59 deletions(-)
#40 lore raw

PATCH 15/36 bdinfo: Drop unused __maybe_unused

In reply to Simon Glass

Some of these are not needed now. Drop them to avoid cluttering the code.

---

 cmd/bdinfo.c | 6 ++----
 1 file changed, 2 insertions(+), 4 deletions(-)
cmd/bdinfo.c +2 -4
index 13b4f2ebba..204d4c5db1 100644
@@ -15,12 +15,11 @@
 
 DECLARE_GLOBAL_DATA_PTR;
 
-__maybe_unused void print_cpu_word_size(void)
+void print_cpu_word_size(void)
 {
 	printf("%-12s= %u-bit\n", "Build", (uint)sizeof(void *) * 8);
 }
 
-__maybe_unused
 static void print_num(const char *name, ulong value)
 {
 	printf("%-12s= 0x%0*lx\n", name, 2 * (int)sizeof(value), value);
@@ -67,7 +66,6 @@ static void print_lnum(const char *name, unsigned long long value)
 	printf("%-12s= 0x%.8llX\n", name, value);
 }
 
-__maybe_unused
 static void print_mhz(const char *name, unsigned long hz)
 {
 	char buf[32];
@@ -160,7 +158,7 @@ static inline void print_baudrate(void)
 #endif
 }
 
-static inline void __maybe_unused print_std_bdinfo(const bd_t *bd)
+static inline void print_std_bdinfo(const bd_t *bd)
 {
 	print_bi_boot_params(bd);
 	print_bi_mem(bd);
Signature · 1 line
2.26.2.526.g744177e7f7-goog
#42 lore raw

PATCH 16/36 bdinfo: microblaze: sh: nios2: Drop arch-specific flash info

In reply to Simon Glass

The differences these architectures have are minor and not worth the extra code. Use the generic version always.

Tidy up the code style by removing unnecessary tabs.

---

 cmd/bdinfo.c | 17 +++--------------
 1 file changed, 3 insertions(+), 14 deletions(-)
cmd/bdinfo.c +3 -14
index 204d4c5db1..3d16f4cc1a 100644
@@ -110,20 +110,9 @@ static inline void print_bi_dram(const bd_t *bd)
 
 static inline void print_bi_flash(const bd_t *bd)
 {
-#if defined(CONFIG_MICROBLAZE) || defined(CONFIG_SH)
-	print_num("flash start    ",	(ulong)bd->bi_flashstart);
-	print_num("flash size     ",	(ulong)bd->bi_flashsize);
-	print_num("flash offset   ",	(ulong)bd->bi_flashoffset);
-
-#elif defined(CONFIG_NIOS2)
-	print_num("flash start",	(ulong)bd->bi_flashstart);
-	print_num("flash size",		(ulong)bd->bi_flashsize);
-	print_num("flash offset",	(ulong)bd->bi_flashoffset);
-#else
-	print_num("flashstart",		(ulong)bd->bi_flashstart);
-	print_num("flashsize",		(ulong)bd->bi_flashsize);
-	print_num("flashoffset",	(ulong)bd->bi_flashoffset);
-#endif
+	print_num("flashstart", (ulong)bd->bi_flashstart);
+	print_num("flashsize", (ulong)bd->bi_flashsize);
+	print_num("flashoffset", (ulong)bd->bi_flashoffset);
 }
 
 static inline void print_eth_ip_addr(void)
Signature · 1 line
2.26.2.526.g744177e7f7-goog
#43 lore raw

PATCH 16/36 bdinfo: microblaze: sh: nios2: Drop arch-specific flash info

Quoted 4 lines · Simon Glass

The differences these architectures have are minor and not worth the extra code. Use the generic version always.

Quoted 8 lines

Tidy up the code style by removing unnecessary tabs.

---

 cmd/bdinfo.c | 17 +++--------------
 1 file changed, 3 insertions(+), 14 deletions(-)
#44 lore raw

PATCH 17/36 bdinfo: Drop unnecessary inline on functions

In reply to Simon Glass

This serves no purpose since the compiler will inline the functions automatically. Drop use of inline in this file.

---

 cmd/bdinfo.c | 15 +++++++--------
 1 file changed, 7 insertions(+), 8 deletions(-)
cmd/bdinfo.c +7 -8
index 3d16f4cc1a..6e5b3d997d 100644
@@ -60,7 +60,6 @@ static void print_eths(void)
 }
 #endif
 
-__maybe_unused
 static void print_lnum(const char *name, unsigned long long value)
 {
 	printf("%-12s= 0x%.8llX\n", name, value);
@@ -74,12 +73,12 @@ static void print_mhz(const char *name, unsigned long hz)
 }
 
 
-static inline void print_bi_boot_params(const bd_t *bd)
+static void print_bi_boot_params(const bd_t *bd)
 {
 	print_num("boot_params",	(ulong)bd->bi_boot_params);
 }
 
-static inline void print_bi_mem(const bd_t *bd)
+static void print_bi_mem(const bd_t *bd)
 {
 #if defined(CONFIG_SH)
 	print_num("mem start      ",	(ulong)bd->bi_memstart);
@@ -93,7 +92,7 @@ static inline void print_bi_mem(const bd_t *bd)
 #endif
 }
 
-static inline void print_bi_dram(const bd_t *bd)
+static void print_bi_dram(const bd_t *bd)
 {
 #ifdef CONFIG_NR_DRAM_BANKS
 	int i;
@@ -108,14 +107,14 @@ static inline void print_bi_dram(const bd_t *bd)
 #endif
 }
 
-static inline void print_bi_flash(const bd_t *bd)
+static void print_bi_flash(const bd_t *bd)
 {
 	print_num("flashstart", (ulong)bd->bi_flashstart);
 	print_num("flashsize", (ulong)bd->bi_flashsize);
 	print_num("flashoffset", (ulong)bd->bi_flashoffset);
 }
 
-static inline void print_eth_ip_addr(void)
+static void print_eth_ip_addr(void)
 {
 #if defined(CONFIG_CMD_NET)
 	print_eth(0);
@@ -138,7 +137,7 @@ static inline void print_eth_ip_addr(void)
 #endif
 }
 
-static inline void print_baudrate(void)
+static void print_baudrate(void)
 {
 #if defined(CONFIG_PPC)
 	printf("baudrate    = %6u bps\n", gd->baudrate);
@@ -147,7 +146,7 @@ static inline void print_baudrate(void)
 #endif
 }
 
-static inline void print_std_bdinfo(const bd_t *bd)
+static void print_std_bdinfo(const bd_t *bd)
 {
 	print_bi_boot_params(bd);
 	print_bi_mem(bd);
Signature · 1 line
2.26.2.526.g744177e7f7-goog
#45 lore raw

PATCH 17/36 bdinfo: Drop unnecessary inline on functions

Hi Simon,

Quoted 15 lines · Simon Glass

This serves no purpose since the compiler will inline the functions automatically. Drop use of inline in this file.

---

 cmd/bdinfo.c | 15 +++++++--------
 1 file changed, 7 insertions(+), 8 deletions(-)
cmd/bdinfo.c +0 -0
index 3d16f4cc1a..6e5b3d997d 100644

cmd/bdinfo.c static void print_eths(void) · line 63

Quoted 4 lines
cmd/bdinfo.c +0 -1
@@ -60,7 +60,6 @@ static void print_eths(void)
 }
 #endif
 
-__maybe_unused

This should be squashed into patch [15/36] bdinfo: Drop unused __maybe_unused

Quoted 54 lines
cmd/bdinfo.c +6 -6
 static void print_lnum(const char *name, unsigned long long value)
 {
        printf("%-12s= 0x%.8llX\n", name, value);
@@ -74,12 +73,12 @@ static void print_mhz(const char *name, unsigned long hz)
 }
 
 
-static inline void print_bi_boot_params(const bd_t *bd)
+static void print_bi_boot_params(const bd_t *bd)
 {
        print_num("boot_params",        (ulong)bd->bi_boot_params);
 }
 
-static inline void print_bi_mem(const bd_t *bd)
+static void print_bi_mem(const bd_t *bd)
 {
 #if defined(CONFIG_SH)
        print_num("mem start      ",    (ulong)bd->bi_memstart);
@@ -93,7 +92,7 @@ static inline void print_bi_mem(const bd_t *bd)
 #endif
 }
 
-static inline void print_bi_dram(const bd_t *bd)
+static void print_bi_dram(const bd_t *bd)
 {
 #ifdef CONFIG_NR_DRAM_BANKS
        int i;
@@ -108,14 +107,14 @@ static inline void print_bi_dram(const bd_t *bd)
 #endif
 }
 
-static inline void print_bi_flash(const bd_t *bd)
+static void print_bi_flash(const bd_t *bd)
 {
        print_num("flashstart", (ulong)bd->bi_flashstart);
        print_num("flashsize", (ulong)bd->bi_flashsize);
        print_num("flashoffset", (ulong)bd->bi_flashoffset);
 }
 
-static inline void print_eth_ip_addr(void)
+static void print_eth_ip_addr(void)
 {
 #if defined(CONFIG_CMD_NET)
        print_eth(0);
@@ -138,7 +137,7 @@ static inline void print_eth_ip_addr(void)
 #endif
 }
 
-static inline void print_baudrate(void)
+static void print_baudrate(void)
 {
 #if defined(CONFIG_PPC)
        printf("baudrate    = %6u bps\n", gd->baudrate);

cmd/bdinfo.c static inline void print_baudrate(void)

Quoted 9 lines
cmd/bdinfo.c +1 -2
@@ -147,7 +146,7 @@ static inline void print_baudrate(void)
 #endif
 }
 
-static inline void print_std_bdinfo(const bd_t *bd)
+static void print_std_bdinfo(const bd_t *bd)
 {
        print_bi_boot_params(bd);
        print_bi_mem(bd);
--

Other than that,

Regards,
Bin

#46 lore raw

PATCH 18/36 bdinfo: Drop print_std_bdinfo()

In reply to Simon Glass

Everything in this function is standard now so it serves no purpose. Move it into the generic do_bdinfo() function.

---

 cmd/bdinfo.c | 14 +++++---------
 1 file changed, 5 insertions(+), 9 deletions(-)
cmd/bdinfo.c +5 -9
index 6e5b3d997d..4e08d9e40a 100644
@@ -146,15 +146,6 @@ static void print_baudrate(void)
 #endif
 }
 
-static void print_std_bdinfo(const bd_t *bd)
-{
-	print_bi_boot_params(bd);
-	print_bi_mem(bd);
-	print_bi_flash(bd);
-	print_eth_ip_addr();
-	print_baudrate();
-}
-
 void __weak board_detail(void)
 {
 	/* Please define board_detail() for your PPC platform */
@@ -169,7 +160,12 @@ int do_bdinfo(cmd_tbl_t *cmdtp, int flag, int argc, char *const argv[])
 #endif
 	if (IS_ENABLED(CONFIG_ARM))
 		print_num("arch_number", bd->bi_arch_number);
+	print_bi_boot_params(bd);
 	print_bi_dram(bd);
+	print_bi_mem(bd);
+	print_bi_flash(bd);
+	print_eth_ip_addr();
+	print_baudrate();
 	print_std_bdinfo(bd);
 	print_num("relocaddr", gd->relocaddr);
 	print_num("reloc off", gd->reloc_off);
Signature · 1 line
2.26.2.526.g744177e7f7-goog
#47 lore raw

PATCH 18/36 bdinfo: Drop print_std_bdinfo()

Quoted 10 lines · Simon Glass

Everything in this function is standard now so it serves no purpose. Move it into the generic do_bdinfo() function.

---

 cmd/bdinfo.c | 14 +++++---------
 1 file changed, 5 insertions(+), 9 deletions(-)
#48 lore raw

PATCH 19/36 bdinfo: ppc: Drop arch-specific print_baudrate()

In reply to Simon Glass

This function outputs the same basic info. Since the baud rate is commonly 115200 these is often no difference. Drop the arch-specific code and inline it to avoid a one-line function.

---

 cmd/bdinfo.c | 12 +-----------
 1 file changed, 1 insertion(+), 11 deletions(-)
cmd/bdinfo.c +1 -11
index 4e08d9e40a..62eea010c5 100644
@@ -137,15 +137,6 @@ static void print_eth_ip_addr(void)
 #endif
 }
 
-static void print_baudrate(void)
-{
-#if defined(CONFIG_PPC)
-	printf("baudrate    = %6u bps\n", gd->baudrate);
-#else
-	printf("baudrate    = %u bps\n", gd->baudrate);
-#endif
-}
-
 void __weak board_detail(void)
 {
 	/* Please define board_detail() for your PPC platform */
@@ -165,8 +156,7 @@ int do_bdinfo(cmd_tbl_t *cmdtp, int flag, int argc, char *const argv[])
 	print_bi_mem(bd);
 	print_bi_flash(bd);
 	print_eth_ip_addr();
-	print_baudrate();
-	print_std_bdinfo(bd);
+	printf("baudrate    = %u bps\n", gd->baudrate);
 	print_num("relocaddr", gd->relocaddr);
 	print_num("reloc off", gd->reloc_off);
 	print_cpu_word_size();
Signature · 1 line
2.26.2.526.g744177e7f7-goog
#49 lore raw

PATCH 19/36 bdinfo: ppc: Drop arch-specific print_baudrate()

Quoted 11 lines · Simon Glass

This function outputs the same basic info. Since the baud rate is commonly 115200 these is often no difference. Drop the arch-specific code and inline it to avoid a one-line function.

---

 cmd/bdinfo.c | 12 +-----------
 1 file changed, 1 insertion(+), 11 deletions(-)
#50 lore raw

PATCH 19/36 bdinfo: ppc: Drop arch-specific print_baudrate()

In reply to Simon Glass
Quoted 5 lines · Simon Glass

This function outputs the same basic info. Since the baud rate is commonly 115200 these is often no difference. Drop the arch-specific code and inline it to avoid a one-line function.

Thanks,
Stefan

Quoted 27 lines

---

  cmd/bdinfo.c | 12 +-----------
  1 file changed, 1 insertion(+), 11 deletions(-)
cmd/bdinfo.c +0 -9
index 4e08d9e40a..62eea010c5 100644
@@ -137,15 +137,6 @@ static void print_eth_ip_addr(void)
  #endif
  }
 
-static void print_baudrate(void)
-{
-#if defined(CONFIG_PPC)
-	printf("baudrate    = %6u bps\n", gd->baudrate);
-#else
-	printf("baudrate    = %u bps\n", gd->baudrate);
-#endif
-}
-
  void __weak board_detail(void)
  {
  	/* Please define board_detail() for your PPC platform */

cmd/bdinfo.c int do_bdinfo(cmd_tbl_t *cmdtp, int flag, int argc, char *const argv[]) · line 162

Quoted 9 lines
cmd/bdinfo.c +1 -2
@@ -165,8 +156,7 @@ int do_bdinfo(cmd_tbl_t *cmdtp, int flag, int argc, char *const argv[])
  	print_bi_mem(bd);
  	print_bi_flash(bd);
  	print_eth_ip_addr();
-	print_baudrate();
-	print_std_bdinfo(bd);
+	printf("baudrate    = %u bps\n", gd->baudrate);
  	print_num("relocaddr", gd->relocaddr);
  	print_num("reloc off", gd->reloc_off);
  	print_cpu_word_size();

Viele Gr??e,
Stefan

Signature · 3 lines
DENX Software Engineering GmbH,      Managing Director: Wolfgang Denk
HRB 165235 Munich, Office: Kirchenstr.5, D-82194 Groebenzell, Germany
Phone: (+49)-8142-66989-51 Fax: (+49)-8142-66989-80 Email: sr at denx.de
#51 lore raw

PATCH 20/36 bdinfo: sh: arc: Drop arch-specific print_bi_mem()

In reply to Simon Glass

It isn't worth having arch-specific code for such minor output differences. In fact it is better if all archs are consistent.

Drop the arch-specific code in print_bi_mem() and inline it to avoid a two-line function.

---

 cmd/bdinfo.c | 17 ++---------------
 1 file changed, 2 insertions(+), 15 deletions(-)
cmd/bdinfo.c +2 -15
index 62eea010c5..5ffd86e508 100644
@@ -78,20 +78,6 @@ static void print_bi_boot_params(const bd_t *bd)
 	print_num("boot_params",	(ulong)bd->bi_boot_params);
 }
 
-static void print_bi_mem(const bd_t *bd)
-{
-#if defined(CONFIG_SH)
-	print_num("mem start      ",	(ulong)bd->bi_memstart);
-	print_lnum("mem size       ",	(u64)bd->bi_memsize);
-#elif defined(CONFIG_ARC)
-	print_num("mem start",		(ulong)bd->bi_memstart);
-	print_lnum("mem size",		(u64)bd->bi_memsize);
-#else
-	print_num("memstart",		(ulong)bd->bi_memstart);
-	print_lnum("memsize",		(u64)bd->bi_memsize);
-#endif
-}
-
 static void print_bi_dram(const bd_t *bd)
 {
 #ifdef CONFIG_NR_DRAM_BANKS
@@ -153,7 +139,8 @@ int do_bdinfo(cmd_tbl_t *cmdtp, int flag, int argc, char *const argv[])
 		print_num("arch_number", bd->bi_arch_number);
 	print_bi_boot_params(bd);
 	print_bi_dram(bd);
-	print_bi_mem(bd);
+	print_num("memstart", (ulong)bd->bi_memstart);
+	print_lnum("memsize", (u64)bd->bi_memsize);
 	print_bi_flash(bd);
 	print_eth_ip_addr();
 	printf("baudrate    = %u bps\n", gd->baudrate);
Signature · 1 line
2.26.2.526.g744177e7f7-goog
#52 lore raw

PATCH 20/36 bdinfo: sh: arc: Drop arch-specific print_bi_mem()

Quoted 4 lines · Simon Glass

It isn't worth having arch-specific code for such minor output differences. In fact it is better if all archs are consistent.

Quoted 9 lines

Drop the arch-specific code in print_bi_mem() and inline it to avoid a two-line function.

---

 cmd/bdinfo.c | 17 ++---------------
 1 file changed, 2 insertions(+), 15 deletions(-)
#53 lore raw

PATCH 20/36 bdinfo: sh: arc: Drop arch-specific print_bi_mem()

Hi Simon,

Quoted 15 lines

Subject: Re: [PATCH 20/36] bdinfo: sh: arc: Drop arch-specific print_bi_mem()

Quoted 12 lines · Simon Glass

It isn't worth having arch-specific code for such minor output differences. In fact it is better if all archs are consistent.

Drop the arch-specific code in print_bi_mem() and inline it to avoid a two-line function.

---

 cmd/bdinfo.c | 17 ++---------------
 1 file changed, 2 insertions(+), 15 deletions(-)

I guess this entire series is available in the following tree, right?
https://gitlab.denx.de/u-boot/custodians/u-boot-dm/-/tree/bd-working

In that case I'm very happy with that change as well as removal of the duplicated spaghetti-code - thanks for taking care of that!

#54 lore raw

PATCH 21/36 bdinfo: Drop print_bi_boot_params()

In reply to Simon Glass

This function only has one line of code in it so inline it.

---

 cmd/bdinfo.c | 8 +-------
 1 file changed, 1 insertion(+), 7 deletions(-)
cmd/bdinfo.c +1 -7
index 5ffd86e508..b65eec332d 100644
@@ -72,12 +72,6 @@ static void print_mhz(const char *name, unsigned long hz)
 	printf("%-12s= %6s MHz\n", name, strmhz(buf, hz));
 }
 
-
-static void print_bi_boot_params(const bd_t *bd)
-{
-	print_num("boot_params",	(ulong)bd->bi_boot_params);
-}
-
 static void print_bi_dram(const bd_t *bd)
 {
 #ifdef CONFIG_NR_DRAM_BANKS
@@ -137,7 +131,7 @@ int do_bdinfo(cmd_tbl_t *cmdtp, int flag, int argc, char *const argv[])
 #endif
 	if (IS_ENABLED(CONFIG_ARM))
 		print_num("arch_number", bd->bi_arch_number);
-	print_bi_boot_params(bd);
+	print_num("boot_params", (ulong)bd->bi_boot_params);
 	print_bi_dram(bd);
 	print_num("memstart", (ulong)bd->bi_memstart);
 	print_lnum("memsize", (u64)bd->bi_memsize);
Signature · 1 line
2.26.2.526.g744177e7f7-goog
#56 lore raw

PATCH 22/36 bdinfo: Drop print_bi_flash()

In reply to Simon Glass

This function only has three lines of code in it so inline it.

---

 cmd/bdinfo.c | 11 +++--------
 1 file changed, 3 insertions(+), 8 deletions(-)
cmd/bdinfo.c +3 -8
index b65eec332d..f22dc7a129 100644
@@ -87,13 +87,6 @@ static void print_bi_dram(const bd_t *bd)
 #endif
 }
 
-static void print_bi_flash(const bd_t *bd)
-{
-	print_num("flashstart", (ulong)bd->bi_flashstart);
-	print_num("flashsize", (ulong)bd->bi_flashsize);
-	print_num("flashoffset", (ulong)bd->bi_flashoffset);
-}
-
 static void print_eth_ip_addr(void)
 {
 #if defined(CONFIG_CMD_NET)
@@ -135,7 +128,9 @@ int do_bdinfo(cmd_tbl_t *cmdtp, int flag, int argc, char *const argv[])
 	print_bi_dram(bd);
 	print_num("memstart", (ulong)bd->bi_memstart);
 	print_lnum("memsize", (u64)bd->bi_memsize);
-	print_bi_flash(bd);
+	print_num("flashstart", (ulong)bd->bi_flashstart);
+	print_num("flashsize", (ulong)bd->bi_flashsize);
+	print_num("flashoffset", (ulong)bd->bi_flashoffset);
 	print_eth_ip_addr();
 	printf("baudrate    = %u bps\n", gd->baudrate);
 	print_num("relocaddr", gd->relocaddr);
Signature · 1 line
2.26.2.526.g744177e7f7-goog
#57 lore raw

PATCH 22/36 bdinfo: Drop print_bi_flash()

Hi Simon,

Quoted 2 lines · Simon Glass

This function only has three lines of code in it so inline it.

What's our guideline for inlining functions? To me if there is only one line, inline the function looks good to me. But 3 lines? I am not sure.

Quoted 7 lines

---

 cmd/bdinfo.c | 11 +++--------
 1 file changed, 3 insertions(+), 8 deletions(-)

Regards,
Bin

#58 lore raw

PATCH 22/36 bdinfo: Drop print_bi_flash()

Quoted 16 lines · Bin Meng

Hi Simon,

Quoted 2 lines · Simon Glass

This function only has three lines of code in it so inline it.

What's our guideline for inlining functions? To me if there is only one line, inline the function looks good to me. But 3 lines? I am not sure.

Quoted 6 lines

---

 cmd/bdinfo.c | 11 +++--------
 1 file changed, 3 insertions(+), 8 deletions(-)

I'm ambivalent on if this should be moved in to the caller or not (I first read inlining as marking with inline). It kinda matters how the overall function looks when you're done.

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/20200506/974204a4/attachment.sig>
#59 lore raw

PATCH 22/36 bdinfo: Drop print_bi_flash()

In reply to Bin Meng

Hi Bin,

Quoted 10 lines · Bin Meng

Hi Simon,

Quoted 2 lines · Simon Glass

This function only has three lines of code in it so inline it.

What's our guideline for inlining functions? To me if there is only one line, inline the function looks good to me. But 3 lines? I am not sure.

Me neither. But in this case I want to get all the code into once place and the end result is a function with about 30 lines, which seems OK.

Quoted 9 lines
Quoted 8 lines

---

 cmd/bdinfo.c | 11 +++--------
 1 file changed, 3 insertions(+), 8 deletions(-)

Regards,
SImon

#60 lore raw

PATCH 23/36 bdinfo: Drop print_cpu_word_size()

In reply to Simon Glass

This function only has one line of code in it so inline it.

---

 cmd/bdinfo.c | 7 +------
 1 file changed, 1 insertion(+), 6 deletions(-)
cmd/bdinfo.c +1 -6
index f22dc7a129..d0afef5b5e 100644
@@ -15,11 +15,6 @@
 
 DECLARE_GLOBAL_DATA_PTR;
 
-void print_cpu_word_size(void)
-{
-	printf("%-12s= %u-bit\n", "Build", (uint)sizeof(void *) * 8);
-}
-
 static void print_num(const char *name, ulong value)
 {
 	printf("%-12s= 0x%0*lx\n", name, 2 * (int)sizeof(value), value);
@@ -135,7 +130,7 @@ int do_bdinfo(cmd_tbl_t *cmdtp, int flag, int argc, char *const argv[])
 	printf("baudrate    = %u bps\n", gd->baudrate);
 	print_num("relocaddr", gd->relocaddr);
 	print_num("reloc off", gd->reloc_off);
-	print_cpu_word_size();
+	printf("%-12s= %u-bit\n", "Build", (uint)sizeof(void *) * 8);
 #if defined(CONFIG_CMD_NET) && !defined(CONFIG_DM_ETH)
 	print_eths();
 #endif
Signature · 1 line
2.26.2.526.g744177e7f7-goog
#62 lore raw

PATCH 24/36 bdinfo: net: ppc: Drop bi_enet1addr and other similar info

In reply to Simon Glass

These values were 'old' in 2013 so it should be safe to remove them. They are never set in U-Boot anyway, so the values will always be zero.

---

 common/board_r.c             | 16 +---------------
 include/asm-generic/u-boot.h | 17 -----------------
 2 files changed, 1 insertion(+), 32 deletions(-)
common/board_r.c +1 -15
index d9015cd057..4876afba6b 100644
@@ -525,21 +525,7 @@ static int initr_ethaddr(void)
 
 	/* kept around for legacy kernels only ... ignore the next section */
 	eth_env_get_enetaddr("ethaddr", bd->bi_enetaddr);
-#ifdef CONFIG_HAS_ETH1
-	eth_env_get_enetaddr("eth1addr", bd->bi_enet1addr);
-#endif
-#ifdef CONFIG_HAS_ETH2
-	eth_env_get_enetaddr("eth2addr", bd->bi_enet2addr);
-#endif
-#ifdef CONFIG_HAS_ETH3
-	eth_env_get_enetaddr("eth3addr", bd->bi_enet3addr);
-#endif
-#ifdef CONFIG_HAS_ETH4
-	eth_env_get_enetaddr("eth4addr", bd->bi_enet4addr);
-#endif
-#ifdef CONFIG_HAS_ETH5
-	eth_env_get_enetaddr("eth5addr", bd->bi_enet5addr);
-#endif
+
 	return 0;
 }
 #endif /* CONFIG_CMD_NET */
include/asm-generic/u-boot.h +0 -17
index be0014c3b3..0ec4522653 100644
@@ -67,23 +67,6 @@ typedef struct bd_info {
 	unsigned long bi_vcofreq;	/* vco Freq in MHz */
 	unsigned long bi_flbfreq;	/* Flexbus Freq in MHz */
 #endif
-
-#ifdef CONFIG_HAS_ETH1
-	unsigned char   bi_enet1addr[6];	/* OLD: see README.enetaddr */
-#endif
-#ifdef CONFIG_HAS_ETH2
-	unsigned char	bi_enet2addr[6];	/* OLD: see README.enetaddr */
-#endif
-#ifdef CONFIG_HAS_ETH3
-	unsigned char   bi_enet3addr[6];	/* OLD: see README.enetaddr */
-#endif
-#ifdef CONFIG_HAS_ETH4
-	unsigned char   bi_enet4addr[6];	/* OLD: see README.enetaddr */
-#endif
-#ifdef CONFIG_HAS_ETH5
-	unsigned char   bi_enet5addr[6];	/* OLD: see README.enetaddr */
-#endif
-
 	ulong	        bi_arch_number;	/* unique id for this board */
 	ulong	        bi_boot_params;	/* where this board expects params */
 #ifdef CONFIG_NR_DRAM_BANKS
Signature · 1 line
2.26.2.526.g744177e7f7-goog
#63 lore raw

PATCH 24/36 bdinfo: net: ppc: Drop bi_enet1addr and other similar info

Quoted 11 lines · Simon Glass

These values were 'old' in 2013 so it should be safe to remove them. They are never set in U-Boot anyway, so the values will always be zero.

---

 common/board_r.c             | 16 +---------------
 include/asm-generic/u-boot.h | 17 -----------------
 2 files changed, 1 insertion(+), 32 deletions(-)
#64 lore raw

PATCH 24/36 bdinfo: net: ppc: Drop bi_enet1addr and other similar info

In reply to Simon Glass
Quoted 4 lines · Simon Glass

These values were 'old' in 2013 so it should be safe to remove them. They are never set in U-Boot anyway, so the values will always be zero.

Thanks,
Stefan

Quoted 52 lines

---

  common/board_r.c             | 16 +---------------
  include/asm-generic/u-boot.h | 17 -----------------
  2 files changed, 1 insertion(+), 32 deletions(-)
common/board_r.c +1 -15
index d9015cd057..4876afba6b 100644
@@ -525,21 +525,7 @@ static int initr_ethaddr(void)
 
  	/* kept around for legacy kernels only ... ignore the next section */
  	eth_env_get_enetaddr("ethaddr", bd->bi_enetaddr);
-#ifdef CONFIG_HAS_ETH1
-	eth_env_get_enetaddr("eth1addr", bd->bi_enet1addr);
-#endif
-#ifdef CONFIG_HAS_ETH2
-	eth_env_get_enetaddr("eth2addr", bd->bi_enet2addr);
-#endif
-#ifdef CONFIG_HAS_ETH3
-	eth_env_get_enetaddr("eth3addr", bd->bi_enet3addr);
-#endif
-#ifdef CONFIG_HAS_ETH4
-	eth_env_get_enetaddr("eth4addr", bd->bi_enet4addr);
-#endif
-#ifdef CONFIG_HAS_ETH5
-	eth_env_get_enetaddr("eth5addr", bd->bi_enet5addr);
-#endif
+
  	return 0;
  }
  #endif /* CONFIG_CMD_NET */
include/asm-generic/u-boot.h +0 -10
index be0014c3b3..0ec4522653 100644
@@ -67,23 +67,6 @@ typedef struct bd_info {
  	unsigned long bi_vcofreq;	/* vco Freq in MHz */
  	unsigned long bi_flbfreq;	/* Flexbus Freq in MHz */
  #endif
-
-#ifdef CONFIG_HAS_ETH1
-	unsigned char   bi_enet1addr[6];	/* OLD: see README.enetaddr */
-#endif
-#ifdef CONFIG_HAS_ETH2
-	unsigned char	bi_enet2addr[6];	/* OLD: see README.enetaddr */
-#endif
-#ifdef CONFIG_HAS_ETH3
-	unsigned char   bi_enet3addr[6];	/* OLD: see README.enetaddr */
-#endif

include/asm-generic/u-boot.h typedef struct bd_info { · line 72

Quoted 10 lines
include/asm-generic/u-boot.h +0 -7
-#ifdef CONFIG_HAS_ETH4
-	unsigned char   bi_enet4addr[6];	/* OLD: see README.enetaddr */
-#endif
-#ifdef CONFIG_HAS_ETH5
-	unsigned char   bi_enet5addr[6];	/* OLD: see README.enetaddr */
-#endif
-
  	ulong	        bi_arch_number;	/* unique id for this board */
  	ulong	        bi_boot_params;	/* where this board expects params */
  #ifdef CONFIG_NR_DRAM_BANKS

Viele Gr??e,
Stefan

Signature · 3 lines
DENX Software Engineering GmbH,      Managing Director: Wolfgang Denk
HRB 165235 Munich, Office: Kirchenstr.5, D-82194 Groebenzell, Germany
Phone: (+49)-8142-66989-51 Fax: (+49)-8142-66989-80 Email: sr at denx.de
#65 lore raw

PATCH 25/36 bdinfo: net: ppc: Drop prints for CONFIG_HAS_ETHn

In reply to Simon Glass

These config options have not been migrated to Kconfig. This should be handled using driver model, iterating over the available Ethernet devices. For now, remove the code.

---

 cmd/bdinfo.c | 15 ---------------
 1 file changed, 15 deletions(-)
cmd/bdinfo.c +0 -15
index d0afef5b5e..6fa8b32389 100644
@@ -86,21 +86,6 @@ static void print_eth_ip_addr(void)
 {
 #if defined(CONFIG_CMD_NET)
 	print_eth(0);
-#if defined(CONFIG_HAS_ETH1)
-	print_eth(1);
-#endif
-#if defined(CONFIG_HAS_ETH2)
-	print_eth(2);
-#endif
-#if defined(CONFIG_HAS_ETH3)
-	print_eth(3);
-#endif
-#if defined(CONFIG_HAS_ETH4)
-	print_eth(4);
-#endif
-#if defined(CONFIG_HAS_ETH5)
-	print_eth(5);
-#endif
 	printf("IP addr     = %s\n", env_get("ipaddr"));
 #endif
 }
Signature · 1 line
2.26.2.526.g744177e7f7-goog
#66 lore raw

PATCH 25/36 bdinfo: net: ppc: Drop prints for CONFIG_HAS_ETHn

Quoted 11 lines · Simon Glass

These config options have not been migrated to Kconfig. This should be handled using driver model, iterating over the available Ethernet devices. For now, remove the code.

---

 cmd/bdinfo.c | 15 ---------------
 1 file changed, 15 deletions(-)
#67 lore raw

PATCH 25/36 bdinfo: net: ppc: Drop prints for CONFIG_HAS_ETHn

In reply to Simon Glass
Quoted 5 lines · Simon Glass

These config options have not been migrated to Kconfig. This should be handled using driver model, iterating over the available Ethernet devices. For now, remove the code.

Thanks,
Stefan

Quoted 22 lines

---

  cmd/bdinfo.c | 15 ---------------
  1 file changed, 15 deletions(-)
cmd/bdinfo.c +0 -8
index d0afef5b5e..6fa8b32389 100644
@@ -86,21 +86,6 @@ static void print_eth_ip_addr(void)
  {
  #if defined(CONFIG_CMD_NET)
  	print_eth(0);
-#if defined(CONFIG_HAS_ETH1)
-	print_eth(1);
-#endif
-#if defined(CONFIG_HAS_ETH2)
-	print_eth(2);
-#endif
-#if defined(CONFIG_HAS_ETH3)
-	print_eth(3);

cmd/bdinfo.c static void print_eth_ip_addr(void) · line 91

Quoted 10 lines
cmd/bdinfo.c +0 -7
-#endif
-#if defined(CONFIG_HAS_ETH4)
-	print_eth(4);
-#endif
-#if defined(CONFIG_HAS_ETH5)
-	print_eth(5);
-#endif
  	printf("IP addr     = %s\n", env_get("ipaddr"));
  #endif
  }

Viele Gr??e,
Stefan

Signature · 3 lines
DENX Software Engineering GmbH,      Managing Director: Wolfgang Denk
HRB 165235 Munich, Office: Kirchenstr.5, D-82194 Groebenzell, Germany
Phone: (+49)-8142-66989-51 Fax: (+49)-8142-66989-80 Email: sr at denx.de
#68 lore raw

PATCH 26/36 bdinfo: net: Drop legacy ethernet bdinfo

In reply to Simon Glass

This code pre-dates driver model and the migration date is nearly upon us. Pare the print_eths() function down and enable it for driver model, since it works correctly.

The IP address is already printed in print_eth_ip_addr() so we can remove that.

---

 cmd/bdinfo.c | 21 ++-------------------
 1 file changed, 2 insertions(+), 19 deletions(-)
cmd/bdinfo.c +2 -19
index 6fa8b32389..ba69575fc2 100644
@@ -34,26 +34,10 @@ static void print_eth(int idx)
 	printf("%-12s= %s\n", name, val);
 }
 
-#ifndef CONFIG_DM_ETH
-__maybe_unused
 static void print_eths(void)
 {
-	struct eth_device *dev;
-	int i = 0;
-
-	do {
-		dev = eth_get_dev_by_index(i);
-		if (dev) {
-			printf("eth%dname    = %s\n", i, dev->name);
-			print_eth(i);
-			i++;
-		}
-	} while (dev);
-
 	printf("current eth = %s\n", eth_get_name());
-	printf("ip_addr     = %s\n", env_get("ipaddr"));
 }
-#endif
 
 static void print_lnum(const char *name, unsigned long long value)
 {
@@ -116,9 +100,8 @@ int do_bdinfo(cmd_tbl_t *cmdtp, int flag, int argc, char *const argv[])
 	print_num("relocaddr", gd->relocaddr);
 	print_num("reloc off", gd->reloc_off);
 	printf("%-12s= %u-bit\n", "Build", (uint)sizeof(void *) * 8);
-#if defined(CONFIG_CMD_NET) && !defined(CONFIG_DM_ETH)
-	print_eths();
-#endif
+	if (IS_ENABLED(CONFIG_CMD_NET))
+		print_eths();
 	print_num("fdt_blob", (ulong)gd->fdt_blob);
 	print_num("new_fdt", (ulong)gd->new_fdt);
 	print_num("fdt_size", (ulong)gd->fdt_size);
Signature · 1 line
2.26.2.526.g744177e7f7-goog
#69 lore raw

PATCH 26/36 bdinfo: net: Drop legacy ethernet bdinfo

Hi Simon,

Quoted 29 lines · Simon Glass

This code pre-dates driver model and the migration date is nearly upon us. Pare the print_eths() function down and enable it for driver model, since it works correctly.

The IP address is already printed in print_eth_ip_addr() so we can remove that.

---

 cmd/bdinfo.c | 21 ++-------------------
 1 file changed, 2 insertions(+), 19 deletions(-)
cmd/bdinfo.c +0 -5
index 6fa8b32389..ba69575fc2 100644
@@ -34,26 +34,10 @@ static void print_eth(int idx)
        printf("%-12s= %s\n", name, val);
 }
 
-#ifndef CONFIG_DM_ETH
-__maybe_unused
 static void print_eths(void)
 {
-       struct eth_device *dev;
-       int i = 0;
-

cmd/bdinfo.c static void print_eth(int idx) · line 39

Quoted 10 lines
cmd/bdinfo.c +0 -9
-       do {
-               dev = eth_get_dev_by_index(i);
-               if (dev) {
-                       printf("eth%dname    = %s\n", i, dev->name);
-                       print_eth(i);
-                       i++;
-               }
-       } while (dev);
-
        printf("current eth = %s\n", eth_get_name());

Now this is an one-line function. Should we inline it?

Quoted 9 lines
cmd/bdinfo.c +0 -2
-       printf("ip_addr     = %s\n", env_get("ipaddr"));
 }
-#endif
 
 static void print_lnum(const char *name, unsigned long long value)
 {
@@ -116,9 +100,8 @@ int do_bdinfo(cmd_tbl_t *cmdtp, int flag, int argc, char *const argv[])
        print_num("relocaddr", gd->relocaddr);
        print_num("reloc off", gd->reloc_off);

cmd/bdinfo.c int do_bdinfo(cmd_tbl_t *cmdtp, int flag, int argc, char *const argv[])

Quoted 10 lines
cmd/bdinfo.c +2 -4
        printf("%-12s= %u-bit\n", "Build", (uint)sizeof(void *) * 8);
-#if defined(CONFIG_CMD_NET) && !defined(CONFIG_DM_ETH)
-       print_eths();
-#endif
+       if (IS_ENABLED(CONFIG_CMD_NET))
+               print_eths();
        print_num("fdt_blob", (ulong)gd->fdt_blob);
        print_num("new_fdt", (ulong)gd->new_fdt);
        print_num("fdt_size", (ulong)gd->fdt_size);
--

Regards,
Bin

#70 lore raw

PATCH 26/36 bdinfo: net: Drop legacy ethernet bdinfo

Hi Bin,

Quoted 37 lines · Bin Meng

Hi Simon,

Quoted 33 lines · Simon Glass

This code pre-dates driver model and the migration date is nearly upon us. Pare the print_eths() function down and enable it for driver model, since it works correctly.

The IP address is already printed in print_eth_ip_addr() so we can remove that.

---

 cmd/bdinfo.c | 21 ++-------------------
 1 file changed, 2 insertions(+), 19 deletions(-)
cmd/bdinfo.c +0 -9
index 6fa8b32389..ba69575fc2 100644
@@ -34,26 +34,10 @@ static void print_eth(int idx)
        printf("%-12s= %s\n", name, val);
 }
 
-#ifndef CONFIG_DM_ETH
-__maybe_unused
 static void print_eths(void)
 {
-       struct eth_device *dev;
-       int i = 0;
-
-       do {
-               dev = eth_get_dev_by_index(i);
-               if (dev) {
-                       printf("eth%dname    = %s\n", i, dev->name);
Quoted 8 lines

cmd/bdinfo.c static void print_eth(int idx) · line 39

Quoted 6 lines
cmd/bdinfo.c +0 -5
-                       print_eth(i);
-                       i++;
-               }
-       } while (dev);
-
        printf("current eth = %s\n", eth_get_name());

Now this is an one-line function. Should we inline it?

Yes it goes away in a later commit. Trying to avoid too much confusing for reviewers.

Regards.

Simon

#71 lore raw

PATCH 27/36 bdinfo: net: Inline print_eths()

In reply to Simon Glass

This function has only one line of code so inline it.

---

 cmd/bdinfo.c | 7 +------
 1 file changed, 1 insertion(+), 6 deletions(-)
cmd/bdinfo.c +1 -6
index ba69575fc2..e6237daf93 100644
@@ -34,11 +34,6 @@ static void print_eth(int idx)
 	printf("%-12s= %s\n", name, val);
 }
 
-static void print_eths(void)
-{
-	printf("current eth = %s\n", eth_get_name());
-}
-
 static void print_lnum(const char *name, unsigned long long value)
 {
 	printf("%-12s= 0x%.8llX\n", name, value);
@@ -101,7 +96,7 @@ int do_bdinfo(cmd_tbl_t *cmdtp, int flag, int argc, char *const argv[])
 	print_num("reloc off", gd->reloc_off);
 	printf("%-12s= %u-bit\n", "Build", (uint)sizeof(void *) * 8);
 	if (IS_ENABLED(CONFIG_CMD_NET))
-		print_eths();
+		printf("current eth = %s\n", eth_get_name());
 	print_num("fdt_blob", (ulong)gd->fdt_blob);
 	print_num("new_fdt", (ulong)gd->new_fdt);
 	print_num("fdt_size", (ulong)gd->fdt_size);
Signature · 1 line
2.26.2.526.g744177e7f7-goog
#72 lore raw

PATCH 27/36 bdinfo: net: Inline print_eths()

Hi Simon,

Quoted 26 lines · Simon Glass

This function has only one line of code so inline it.

---

 cmd/bdinfo.c | 7 +------
 1 file changed, 1 insertion(+), 6 deletions(-)
cmd/bdinfo.c +0 -5
index ba69575fc2..e6237daf93 100644
@@ -34,11 +34,6 @@ static void print_eth(int idx)
        printf("%-12s= %s\n", name, val);
 }
 
-static void print_eths(void)
-{
-       printf("current eth = %s\n", eth_get_name());
-}
-
 static void print_lnum(const char *name, unsigned long long value)
 {
        printf("%-12s= 0x%.8llX\n", name, value);

cmd/bdinfo.c int do_bdinfo(cmd_tbl_t *cmdtp, int flag, int argc, char *const argv[])

Quoted 9 lines
cmd/bdinfo.c +1 -2
@@ -101,7 +96,7 @@ int do_bdinfo(cmd_tbl_t *cmdtp, int flag, int argc, char *const argv[])
        print_num("reloc off", gd->reloc_off);
        printf("%-12s= %u-bit\n", "Build", (uint)sizeof(void *) * 8);
        if (IS_ENABLED(CONFIG_CMD_NET))
-               print_eths();
+               printf("current eth = %s\n", eth_get_name());
        print_num("fdt_blob", (ulong)gd->fdt_blob);
        print_num("new_fdt", (ulong)gd->new_fdt);
        print_num("fdt_size", (ulong)gd->fdt_size);
--

I think this patch should be squashed into [26/36] bdinfo: net: Drop legacy ethernet bdinfo

Regards,
Bin

#73 lore raw

PATCH 28/36 bdinfo: net: Inline print_eth_ip_addr()

In reply to Simon Glass

This function only has two lines of code now, so inline it.

---

 cmd/bdinfo.c | 15 ++++-----------
 1 file changed, 4 insertions(+), 11 deletions(-)
cmd/bdinfo.c +4 -11
index e6237daf93..e30d530411 100644
@@ -20,7 +20,6 @@ static void print_num(const char *name, ulong value)
 	printf("%-12s= 0x%0*lx\n", name, 2 * (int)sizeof(value), value);
 }
 
-__maybe_unused
 static void print_eth(int idx)
 {
 	char name[10], *val;
@@ -61,14 +60,6 @@ static void print_bi_dram(const bd_t *bd)
 #endif
 }
 
-static void print_eth_ip_addr(void)
-{
-#if defined(CONFIG_CMD_NET)
-	print_eth(0);
-	printf("IP addr     = %s\n", env_get("ipaddr"));
-#endif
-}
-
 void __weak board_detail(void)
 {
 	/* Please define board_detail() for your PPC platform */
@@ -90,13 +81,15 @@ int do_bdinfo(cmd_tbl_t *cmdtp, int flag, int argc, char *const argv[])
 	print_num("flashstart", (ulong)bd->bi_flashstart);
 	print_num("flashsize", (ulong)bd->bi_flashsize);
 	print_num("flashoffset", (ulong)bd->bi_flashoffset);
-	print_eth_ip_addr();
 	printf("baudrate    = %u bps\n", gd->baudrate);
 	print_num("relocaddr", gd->relocaddr);
 	print_num("reloc off", gd->reloc_off);
 	printf("%-12s= %u-bit\n", "Build", (uint)sizeof(void *) * 8);
-	if (IS_ENABLED(CONFIG_CMD_NET))
+	if (IS_ENABLED(CONFIG_CMD_NET)) {
 		printf("current eth = %s\n", eth_get_name());
+		print_eth(0);
+		printf("IP addr     = %s\n", env_get("ipaddr"));
+	}
 	print_num("fdt_blob", (ulong)gd->fdt_blob);
 	print_num("new_fdt", (ulong)gd->new_fdt);
 	print_num("fdt_size", (ulong)gd->fdt_size);
Signature · 1 line
2.26.2.526.g744177e7f7-goog
#75 lore raw

PATCH 29/36 bdinfo: Export some basic printing functions

In reply to Simon Glass

At present the functions to print a number and a frequency are static. We want to move some of the code in here to an arch-specific file. For consistency that code should use these same functions. So export them with an appropriate name.

---

 cmd/bdinfo.c   | 78 +++++++++++++++++++++++++-------------------------
 include/init.h |  6 ++++
 2 files changed, 45 insertions(+), 39 deletions(-)
cmd/bdinfo.c +39 -39
index e30d530411..c6096a5fd3 100644
@@ -15,7 +15,7 @@
 
 DECLARE_GLOBAL_DATA_PTR;
 
-static void print_num(const char *name, ulong value)
+void bdinfo_print_num(const char *name, ulong value)
 {
 	printf("%-12s= 0x%0*lx\n", name, 2 * (int)sizeof(value), value);
 }
@@ -38,7 +38,7 @@ static void print_lnum(const char *name, unsigned long long value)
 	printf("%-12s= 0x%.8llX\n", name, value);
 }
 
-static void print_mhz(const char *name, unsigned long hz)
+void bdinfo_print_mhz(const char *name, unsigned long hz)
 {
 	char buf[32];
 
@@ -52,9 +52,9 @@ static void print_bi_dram(const bd_t *bd)
 
 	for (i = 0; i < CONFIG_NR_DRAM_BANKS; ++i) {
 		if (bd->bi_dram[i].size) {
-			print_num("DRAM bank",	i);
-			print_num("-> start",	bd->bi_dram[i].start);
-			print_num("-> size",	bd->bi_dram[i].size);
+			bdinfo_print_num("DRAM bank",	i);
+			bdinfo_print_num("-> start",	bd->bi_dram[i].start);
+			bdinfo_print_num("-> size",	bd->bi_dram[i].size);
 		}
 	}
 #endif
@@ -70,50 +70,50 @@ int do_bdinfo(cmd_tbl_t *cmdtp, int flag, int argc, char *const argv[])
 	bd_t *bd = gd->bd;
 
 #ifdef DEBUG
-	print_num("bd address", (ulong)bd);
+	bdinfo_print_num("bd address", (ulong)bd);
 #endif
 	if (IS_ENABLED(CONFIG_ARM))
-		print_num("arch_number", bd->bi_arch_number);
-	print_num("boot_params", (ulong)bd->bi_boot_params);
+		bdinfo_print_num("arch_number", bd->bi_arch_number);
+	bdinfo_print_num("boot_params", (ulong)bd->bi_boot_params);
 	print_bi_dram(bd);
-	print_num("memstart", (ulong)bd->bi_memstart);
+	bdinfo_print_num("memstart", (ulong)bd->bi_memstart);
 	print_lnum("memsize", (u64)bd->bi_memsize);
-	print_num("flashstart", (ulong)bd->bi_flashstart);
-	print_num("flashsize", (ulong)bd->bi_flashsize);
-	print_num("flashoffset", (ulong)bd->bi_flashoffset);
+	bdinfo_print_num("flashstart", (ulong)bd->bi_flashstart);
+	bdinfo_print_num("flashsize", (ulong)bd->bi_flashsize);
+	bdinfo_print_num("flashoffset", (ulong)bd->bi_flashoffset);
 	printf("baudrate    = %u bps\n", gd->baudrate);
-	print_num("relocaddr", gd->relocaddr);
-	print_num("reloc off", gd->reloc_off);
+	bdinfo_print_num("relocaddr", gd->relocaddr);
+	bdinfo_print_num("reloc off", gd->reloc_off);
 	printf("%-12s= %u-bit\n", "Build", (uint)sizeof(void *) * 8);
 	if (IS_ENABLED(CONFIG_CMD_NET)) {
 		printf("current eth = %s\n", eth_get_name());
 		print_eth(0);
 		printf("IP addr     = %s\n", env_get("ipaddr"));
 	}
-	print_num("fdt_blob", (ulong)gd->fdt_blob);
-	print_num("new_fdt", (ulong)gd->new_fdt);
-	print_num("fdt_size", (ulong)gd->fdt_size);
+	bdinfo_print_num("fdt_blob", (ulong)gd->fdt_blob);
+	bdinfo_print_num("new_fdt", (ulong)gd->new_fdt);
+	bdinfo_print_num("fdt_size", (ulong)gd->fdt_size);
 #if defined(CONFIG_LCD) || defined(CONFIG_VIDEO) || defined(CONFIG_DM_VIDEO)
-	print_num("FB base  ", gd->fb_base);
+	bdinfo_print_num("FB base  ", gd->fb_base);
 #endif
 
 	/* This section is used only by ARM */
 #ifdef CONFIG_ARM
 #ifdef CONFIG_SYS_MEM_RESERVE_SECURE
 	if (gd->arch.secure_ram & MEM_RESERVE_SECURE_SECURED) {
-		print_num("Secure ram",
+		bdinfo_print_num("Secure ram",
 			  gd->arch.secure_ram & MEM_RESERVE_SECURE_ADDR_MASK);
 	}
 #endif
 #ifdef CONFIG_RESV_RAM
 	if (gd->arch.resv_ram)
-		print_num("Reserved ram", gd->arch.resv_ram);
+		bdinfo_print_num("Reserved ram", gd->arch.resv_ram);
 #endif
 #if !(CONFIG_IS_ENABLED(SYS_ICACHE_OFF) && CONFIG_IS_ENABLED(SYS_DCACHE_OFF))
-	print_num("TLB addr", gd->arch.tlb_addr);
+	bdinfo_print_num("TLB addr", gd->arch.tlb_addr);
 #endif
-	print_num("irq_sp", gd->irq_sp);	/* irq stack pointer */
-	print_num("sp start ", gd->start_addr_sp);
+	bdinfo_print_num("irq_sp", gd->irq_sp);	/* irq stack pointer */
+	bdinfo_print_num("sp start ", gd->start_addr_sp);
 	/*
 	 * TODO: Currently only support for davinci SOC's is added.
 	 * Remove this check once all the board implement this.
@@ -131,17 +131,17 @@ int do_bdinfo(cmd_tbl_t *cmdtp, int flag, int argc, char *const argv[])
 	       CONFIG_VAL(SYS_MALLOC_F_LEN));
 #endif
 #if CONFIG_IS_ENABLED(MULTI_DTB_FIT)
-	print_num("multi_dtb_fit", (ulong)gd->multi_dtb_fit);
+	bdinfo_print_num("multi_dtb_fit", (ulong)gd->multi_dtb_fit);
 #endif
 #endif /* CONFIG_ARM */
 
 	/* This section is used only by ppc */
 #if defined(CONFIG_MPC8xx) || defined(CONFIG_E500)
-	print_num("immr_base", bd->bi_immr_base);
+	bdinfo_print_num("immr_base", bd->bi_immr_base);
 #endif
 	if (IS_ENABLED(CONFIG_PPC)) {
-		print_num("bootflags", bd->bi_bootflags);
-		print_mhz("intfreq", bd->bi_intfreq);
+		bdinfo_print_num("bootflags", bd->bi_bootflags);
+		bdinfo_print_mhz("intfreq", bd->bi_intfreq);
 #ifdef CONFIG_ENABLE_36BIT_PHYS
 		if (IS_ENABLED(CONFIG_PHYS_64BIT))
 			puts("addressing  = 36-bit\n");
@@ -151,32 +151,32 @@ int do_bdinfo(cmd_tbl_t *cmdtp, int flag, int argc, char *const argv[])
 		board_detail();
 	}
 #if defined(CONFIG_CPM2)
-	print_mhz("cpmfreq", bd->bi_cpmfreq);
-	print_mhz("vco", bd->bi_vco);
-	print_mhz("sccfreq", bd->bi_sccfreq);
-	print_mhz("brgfreq", bd->bi_brgfreq);
+	bdinfo_print_mhz("cpmfreq", bd->bi_cpmfreq);
+	bdinfo_print_mhz("vco", bd->bi_vco);
+	bdinfo_print_mhz("sccfreq", bd->bi_sccfreq);
+	bdinfo_print_mhz("brgfreq", bd->bi_brgfreq);
 #endif
 
 	/* This is used by m68k and ppc */
 #if defined(CONFIG_SYS_INIT_RAM_ADDR)
-	print_num("sramstart", (ulong)bd->bi_sramstart);
-	print_num("sramsize", (ulong)bd->bi_sramsize);
+	bdinfo_print_num("sramstart", (ulong)bd->bi_sramstart);
+	bdinfo_print_num("sramsize", (ulong)bd->bi_sramsize);
 #endif
 	if (IS_ENABLED(CONFIG_PPC) || IS_ENABLED(CONFIG_M68K))
-		print_mhz("busfreq", bd->bi_busfreq);
+		bdinfo_print_mhz("busfreq", bd->bi_busfreq);
 
 	/* The rest are used only by m68k */
 #ifdef CONFIG_M68K
 #if defined(CONFIG_SYS_MBAR)
 	print_num("mbar", bd->bi_mbar_base);
 #endif
-	print_mhz("cpufreq", bd->bi_intfreq);
+	bdinfo_print_mhz("cpufreq", bd->bi_intfreq);
 	if (IS_ENABLED(CONFIG_PCI))
-		print_mhz("pcifreq", bd->bi_pcifreq);
+		bdinfo_print_mhz("pcifreq", bd->bi_pcifreq);
 #ifdef CONFIG_EXTRA_CLOCK
-	print_mhz("flbfreq", bd->bi_flbfreq);
-	print_mhz("inpfreq", bd->bi_inpfreq);
-	print_mhz("vcofreq", bd->bi_vcofreq);
+	bdinfo_print_mhz("flbfreq", bd->bi_flbfreq);
+	bdinfo_print_mhz("inpfreq", bd->bi_inpfreq);
+	bdinfo_print_mhz("vcofreq", bd->bi_vcofreq);
 #endif
 #endif
 
include/init.h +6 -0
index b5a167b6ed..af4deed566 100644
@@ -261,6 +261,12 @@ void relocate_code(ulong start_addr_sp, struct global_data *new_gd,
 	__attribute__ ((noreturn));
 #endif
 
+/* Print a numeric value (for use in arch_print_bdinfo()) */
+void bdinfo_print_num(const char *name, ulong value);
+
+/* Print a clock speed in MHz */
+void bdinfo_print_mhz(const char *name, unsigned long hz);
+
 #endif	/* __ASSEMBLY__ */
 /* Put only stuff here that the assembler can digest */
 
Signature · 1 line
2.26.2.526.g744177e7f7-goog
#76 lore raw

PATCH 29/36 bdinfo: Export some basic printing functions

Quoted 6 lines · Simon Glass

At present the functions to print a number and a frequency are static. We want to move some of the code in here to an arch-specific file. For consistency that code should use these same functions. So export them with an appropriate name.

Quoted 7 lines

---

 cmd/bdinfo.c   | 78 +++++++++++++++++++++++++-------------------------
 include/init.h |  6 ++++
 2 files changed, 45 insertions(+), 39 deletions(-)
#77 lore raw

PATCH 30/36 bdinfo: arm: Move ARM-specific info into its own file

In reply to Simon Glass

We don't really want to have ARM-specific code in a generic file. Create a new arch-specific function to hold it, and move it into that.

Make the function weak so that any arch can implement it.

---

 arch/arm/lib/Makefile |  1 +
 arch/arm/lib/bdinfo.c | 51 +++++++++++++++++++++++++++++++++++++++++++
 cmd/bdinfo.c          | 44 +++++--------------------------------
 include/init.h        |  3 +++
 4 files changed, 61 insertions(+), 38 deletions(-)
 create mode 100644 arch/arm/lib/bdinfo.c
arch/arm/lib/Makefile +1 -0
index b839aa7a50..27b12e7f2b 100644
@@ -43,6 +43,7 @@ obj-$(CONFIG_$(SPL_TPL_)USE_ARCH_MEMSET) += memset.o
 obj-$(CONFIG_$(SPL_TPL_)USE_ARCH_MEMCPY) += memcpy.o
 obj-$(CONFIG_SEMIHOSTING) += semihosting.o
 
+obj-y	+= bdinfo.o
 obj-y	+= sections.o
 obj-y	+= stack.o
 ifdef CONFIG_CPU_V7M
arch/arm/lib/bdinfo.c new file +51 -0

51 lines not on this page · raw message

cmd/bdinfo.c +6 -38

62 lines not on this page · raw message

include/init.h +3 -0
index af4deed566..e727031514 100644
@@ -267,6 +267,9 @@ void bdinfo_print_num(const char *name, ulong value);
 /* Print a clock speed in MHz */
 void bdinfo_print_mhz(const char *name, unsigned long hz);
 
+/* Show arch-specific information for the 'bd' command */
+void arch_print_bdinfo(void);
+
 #endif	/* __ASSEMBLY__ */
 /* Put only stuff here that the assembler can digest */
 
Signature · 1 line
2.26.2.526.g744177e7f7-goog
#78 lore raw

PATCH 30/36 bdinfo: arm: Move ARM-specific info into its own file

Quoted 7 lines · Simon Glass

We don't really want to have ARM-specific code in a generic file. Create a new arch-specific function to hold it, and move it into that.

Make the function weak so that any arch can implement it.

Quoted 9 lines

---

 arch/arm/lib/Makefile |  1 +
 arch/arm/lib/bdinfo.c | 51 +++++++++++++++++++++++++++++++++++++++++++
 cmd/bdinfo.c          | 44 +++++--------------------------------
 include/init.h        |  3 +++
 4 files changed, 61 insertions(+), 38 deletions(-)
 create mode 100644 arch/arm/lib/bdinfo.c
#79 lore raw

PATCH 30/36 bdinfo: arm: Move ARM-specific info into its own file

In reply to Simon Glass

Am 05.05.20 um 01:17 schrieb Simon Glass:

Quoted 77 lines

We don't really want to have ARM-specific code in a generic file. Create a new arch-specific function to hold it, and move it into that.

Make the function weak so that any arch can implement it.

---

 arch/arm/lib/Makefile |  1 +
 arch/arm/lib/bdinfo.c | 51 +++++++++++++++++++++++++++++++++++++++++++
 cmd/bdinfo.c          | 44 +++++--------------------------------
 include/init.h        |  3 +++
 4 files changed, 61 insertions(+), 38 deletions(-)
 create mode 100644 arch/arm/lib/bdinfo.c
arch/arm/lib/Makefile +1 -0
index b839aa7a50..27b12e7f2b 100644
@@ -43,6 +43,7 @@ obj-$(CONFIG_$(SPL_TPL_)USE_ARCH_MEMSET) += memset.o
 obj-$(CONFIG_$(SPL_TPL_)USE_ARCH_MEMCPY) += memcpy.o
 obj-$(CONFIG_SEMIHOSTING) += semihosting.o
 
+obj-y	+= bdinfo.o
 obj-y	+= sections.o
 obj-y	+= stack.o
 ifdef CONFIG_CPU_V7M
arch/arm/lib/bdinfo.c new file +43 -0
new file mode 100644
index 0000000000..ce8edd0313
@@ -0,0 +1,51 @@
+// SPDX-License-Identifier: GPL-2.0+
+/*
+ * ARM-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;
+
+void arch_print_bdinfo(void)
+{
+#ifdef CONFIG_SYS_MEM_RESERVE_SECURE
+	if (gd->arch.secure_ram & MEM_RESERVE_SECURE_SECURED) {
+		bdinfo_print_num("Secure ram",
+				 gd->arch.secure_ram &
+				 MEM_RESERVE_SECURE_ADDR_MASK);
+	}
+#endif
+#ifdef CONFIG_RESV_RAM
+	if (gd->arch.resv_ram)
+		bdinfo_print_num("Reserved ram", gd->arch.resv_ram);
+#endif
+#if !(CONFIG_IS_ENABLED(SYS_ICACHE_OFF) && CONFIG_IS_ENABLED(SYS_DCACHE_OFF))
+	bdinfo_print_num("TLB addr", gd->arch.tlb_addr);
+#endif
+	bdinfo_print_num("irq_sp", gd->irq_sp);	/* irq stack pointer */
+	bdinfo_print_num("sp start ", gd->start_addr_sp);
+	/*
+	 * TODO: Currently only support for davinci SOC's is added.
+	 * Remove this check once all the board implement this.
+	 */
+#ifdef CONFIG_CLOCKS
+	printf("ARM frequency = %ld MHz\n", gd->bd->bi_arm_freq);
+	printf("DSP frequency = %ld MHz\n", gd->bd->bi_dsp_freq);
+	printf("DDR frequency = %ld MHz\n", gd->bd->bi_ddr_freq);
+#endif
+#ifdef CONFIG_BOARD_TYPES
+	printf("Board Type  = %ld\n", gd->board_type);
+#endif
 

arch/arm/lib/bdinfo.c line 51

Quoted 10 lines
arch/arm/lib/bdinfo.c new file +8 -0
 
 
+#if CONFIG_VAL(SYS_MALLOC_F_LEN)
+	printf("Early malloc usage: %lx / %x\n", gd->malloc_ptr,
+	       CONFIG_VAL(SYS_MALLOC_F_LEN));
+#endif
+#if CONFIG_IS_ENABLED(MULTI_DTB_FIT)
+	bdinfo_print_num("multi_dtb_fit", (ulong)gd->multi_dtb_fit);
+#endif
+}

that part is not really ARM specific. Maybe it's useful in the generic function?

Signature · 1 line
- Daniel
#80 lore raw

PATCH 31/36 bdinfo: ppc: Move PPC-specific info into its own file

In reply to Simon Glass

We don't really want to have PPC-specific code in a generic file. Create a new arch-specific function to hold it, and move it into that.

Make the function weak so that any arch can implement it.

---

 arch/powerpc/lib/Makefile |  2 ++
 arch/powerpc/lib/bdinfo.c | 41 +++++++++++++++++++++++++++++++++++++++
 cmd/bdinfo.c              | 27 --------------------------
 3 files changed, 43 insertions(+), 27 deletions(-)
 create mode 100644 arch/powerpc/lib/bdinfo.c
arch/powerpc/lib/Makefile +2 -0
index 01c9dd51be..f61809ab05 100644
@@ -15,6 +15,8 @@ MINIMAL=y
 endif
 endif
 
+obj-y	+= bdinfo.o
+
 ifdef MINIMAL
 obj-y += cache.o time.o
 ifndef CONFIG_TIMER
arch/powerpc/lib/bdinfo.c new file +41 -0

41 lines not on this page · raw message

cmd/bdinfo.c +0 -27
index 55a9244aef..570022052c 100644
@@ -60,11 +60,6 @@ static void print_bi_dram(const bd_t *bd)
 #endif
 }
 
-void __weak board_detail(void)
-{
-	/* Please define board_detail() for your PPC platform */
-}
-
 __weak void arch_print_bdinfo(void)
 {
 }
@@ -103,28 +98,6 @@ int do_bdinfo(cmd_tbl_t *cmdtp, int flag, int argc, char *const argv[])
 
 	arch_print_bdinfo();
 
-	/* This section is used only by ppc */
-#if defined(CONFIG_MPC8xx) || defined(CONFIG_E500)
-	bdinfo_print_num("immr_base", bd->bi_immr_base);
-#endif
-	if (IS_ENABLED(CONFIG_PPC)) {
-		bdinfo_print_num("bootflags", bd->bi_bootflags);
-		bdinfo_print_mhz("intfreq", bd->bi_intfreq);
-#ifdef CONFIG_ENABLE_36BIT_PHYS
-		if (IS_ENABLED(CONFIG_PHYS_64BIT))
-			puts("addressing  = 36-bit\n");
-		else
-			puts("addressing  = 32-bit\n");
-#endif
-		board_detail();
-	}
-#if defined(CONFIG_CPM2)
-	bdinfo_print_mhz("cpmfreq", bd->bi_cpmfreq);
-	bdinfo_print_mhz("vco", bd->bi_vco);
-	bdinfo_print_mhz("sccfreq", bd->bi_sccfreq);
-	bdinfo_print_mhz("brgfreq", bd->bi_brgfreq);
-#endif
-
 	/* This is used by m68k and ppc */
 #if defined(CONFIG_SYS_INIT_RAM_ADDR)
 	bdinfo_print_num("sramstart", (ulong)bd->bi_sramstart);
Signature · 1 line
2.26.2.526.g744177e7f7-goog
#81 lore raw

PATCH 31/36 bdinfo: ppc: Move PPC-specific info into its own file

Quoted 6 lines · Simon Glass

We don't really want to have PPC-specific code in a generic file. Create a new arch-specific function to hold it, and move it into that.

Make the function weak so that any arch can implement it.

Quoted 9 lines

---

 arch/powerpc/lib/Makefile |  2 ++
 arch/powerpc/lib/bdinfo.c | 41 +++++++++++++++++++++++++++++++++++++++
 cmd/bdinfo.c              | 27 --------------------------
 3 files changed, 43 insertions(+), 27 deletions(-)
 create mode 100644 arch/powerpc/lib/bdinfo.c
#82 lore raw

PATCH 31/36 bdinfo: ppc: Move PPC-specific info into its own file

In reply to Simon Glass
Quoted 6 lines · Simon Glass

We don't really want to have PPC-specific code in a generic file. Create a new arch-specific function to hold it, and move it into that.

Make the function weak so that any arch can implement it.

Thanks,
Stefan

Quoted 104 lines

---

  arch/powerpc/lib/Makefile |  2 ++
  arch/powerpc/lib/bdinfo.c | 41 +++++++++++++++++++++++++++++++++++++++
  cmd/bdinfo.c              | 27 --------------------------
  3 files changed, 43 insertions(+), 27 deletions(-)

create mode 100644 arch/powerpc/lib/bdinfo.c

arch/powerpc/lib/Makefile +2 -0
index 01c9dd51be..f61809ab05 100644
@@ -15,6 +15,8 @@ MINIMAL=y
  endif
  endif
 
+obj-y	+= bdinfo.o
+
  ifdef MINIMAL
  obj-y += cache.o time.o
  ifndef CONFIG_TIMER
arch/powerpc/lib/bdinfo.c new file +41 -0
new file mode 100644
index 0000000000..da09bb276f
@@ -0,0 +1,41 @@
+// SPDX-License-Identifier: GPL-2.0+
+/*
+ * PPC-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;
+
+void __weak board_detail(void)
+{
+	/* Please define board_detail() for your PPC platform */
+}
+
+void arch_print_bdinfo(void)
+{
+	bd_t *bd = gd->bd;
+
+#if defined(CONFIG_MPC8xx) || defined(CONFIG_E500)
+	bdinfo_print_num("immr_base", bd->bi_immr_base);
+#endif
+	bdinfo_print_num("bootflags", bd->bi_bootflags);
+	bdinfo_print_mhz("intfreq", bd->bi_intfreq);
+#ifdef CONFIG_ENABLE_36BIT_PHYS
+	if (IS_ENABLED(CONFIG_PHYS_64BIT))
+		puts("addressing  = 36-bit\n");
+	else
+		puts("addressing  = 32-bit\n");
+#endif
+	board_detail();
+#if defined(CONFIG_CPM2)
+	bdinfo_print_mhz("cpmfreq", bd->bi_cpmfreq);
+	bdinfo_print_mhz("vco", bd->bi_vco);
+	bdinfo_print_mhz("sccfreq", bd->bi_sccfreq);
+	bdinfo_print_mhz("brgfreq", bd->bi_brgfreq);
+#endif
+}
cmd/bdinfo.c +0 -20
index 55a9244aef..570022052c 100644
@@ -60,11 +60,6 @@ static void print_bi_dram(const bd_t *bd)
  #endif
  }
 
-void __weak board_detail(void)
-{
-	/* Please define board_detail() for your PPC platform */
-}
-
  __weak void arch_print_bdinfo(void)
  {
  }
@@ -103,28 +98,6 @@ int do_bdinfo(cmd_tbl_t *cmdtp, int flag, int argc, char *const argv[])
 
  	arch_print_bdinfo();
 
-	/* This section is used only by ppc */
-#if defined(CONFIG_MPC8xx) || defined(CONFIG_E500)
-	bdinfo_print_num("immr_base", bd->bi_immr_base);
-#endif
-	if (IS_ENABLED(CONFIG_PPC)) {
-		bdinfo_print_num("bootflags", bd->bi_bootflags);
-		bdinfo_print_mhz("intfreq", bd->bi_intfreq);
-#ifdef CONFIG_ENABLE_36BIT_PHYS
-		if (IS_ENABLED(CONFIG_PHYS_64BIT))
-			puts("addressing  = 36-bit\n");
-		else
-			puts("addressing  = 32-bit\n");
-#endif
-		board_detail();
-	}

cmd/bdinfo.c int do_bdinfo(cmd_tbl_t *cmdtp, int flag, int argc, char *const argv[]) · line 103

Quoted 10 lines
cmd/bdinfo.c +0 -7
-#if defined(CONFIG_CPM2)
-	bdinfo_print_mhz("cpmfreq", bd->bi_cpmfreq);
-	bdinfo_print_mhz("vco", bd->bi_vco);
-	bdinfo_print_mhz("sccfreq", bd->bi_sccfreq);
-	bdinfo_print_mhz("brgfreq", bd->bi_brgfreq);
-#endif
-
  	/* This is used by m68k and ppc */
  #if defined(CONFIG_SYS_INIT_RAM_ADDR)
  	bdinfo_print_num("sramstart", (ulong)bd->bi_sramstart);

Viele Gr??e,
Stefan

Signature · 3 lines
DENX Software Engineering GmbH,      Managing Director: Wolfgang Denk
HRB 165235 Munich, Office: Kirchenstr.5, D-82194 Groebenzell, Germany
Phone: (+49)-8142-66989-51 Fax: (+49)-8142-66989-80 Email: sr at denx.de
#83 lore raw

PATCH 32/36 bdinfo: m68k: Move m68k-specific info into its own file

In reply to Simon Glass

We don't really want to have m68k-specific code in a generic file. Create a new arch-specific function to hold it, and move it into that.

Make the function weak so that any arch can implement it.

---

 arch/m68k/lib/Makefile |  1 +
 arch/m68k/lib/bdinfo.c | 29 +++++++++++++++++++++++++++++
 cmd/bdinfo.c           | 15 ---------------
 3 files changed, 30 insertions(+), 15 deletions(-)
 create mode 100644 arch/m68k/lib/bdinfo.c
arch/m68k/lib/Makefile +1 -0
index a040f40eb8..b66d66afd2 100644
@@ -7,6 +7,7 @@
 ## if the user asked for it
 lib-$(CONFIG_USE_PRIVATE_LIBGCC) += lshrdi3.o muldi3.o ashldi3.o
 
+obj-y	+= bdinfo.o
 obj-$(CONFIG_CMD_BOOTM) += bootm.o
 obj-y	+= cache.o
 obj-y	+= interrupts.o
arch/m68k/lib/bdinfo.c new file +29 -0
new file mode 100644
index 0000000000..09a1abfc21
@@ -0,0 +1,29 @@
+// SPDX-License-Identifier: GPL-2.0+
+/*
+ * PPC-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;
+
+void arch_print_bdinfo(void)
+{
+	bd_t *bd = gd->bd;
+
+#if defined(CONFIG_SYS_MBAR)
+	bdinfo_print_num("mbar", bd->bi_mbar_base);
+#endif
+	bdinfo_print_mhz("cpufreq", bd->bi_intfreq);
+	if (IS_ENABLED(CONFIG_PCI))
+		bdinfo_print_mhz("pcifreq", bd->bi_pcifreq);
+#ifdef CONFIG_EXTRA_CLOCK
+	bdinfo_print_mhz("flbfreq", bd->bi_flbfreq);
+	bdinfo_print_mhz("inpfreq", bd->bi_inpfreq);
+	bdinfo_print_mhz("vcofreq", bd->bi_vcofreq);
+#endif
+}
cmd/bdinfo.c +0 -15
index 570022052c..f0afdb153f 100644
@@ -106,21 +106,6 @@ int do_bdinfo(cmd_tbl_t *cmdtp, int flag, int argc, char *const argv[])
 	if (IS_ENABLED(CONFIG_PPC) || IS_ENABLED(CONFIG_M68K))
 		bdinfo_print_mhz("busfreq", bd->bi_busfreq);
 
-	/* The rest are used only by m68k */
-#ifdef CONFIG_M68K
-#if defined(CONFIG_SYS_MBAR)
-	bdinfo_print_num("mbar", bd->bi_mbar_base);
-#endif
-	bdinfo_print_mhz("cpufreq", bd->bi_intfreq);
-	if (IS_ENABLED(CONFIG_PCI))
-		bdinfo_print_mhz("pcifreq", bd->bi_pcifreq);
-#ifdef CONFIG_EXTRA_CLOCK
-	bdinfo_print_mhz("flbfreq", bd->bi_flbfreq);
-	bdinfo_print_mhz("inpfreq", bd->bi_inpfreq);
-	bdinfo_print_mhz("vcofreq", bd->bi_vcofreq);
-#endif
-#endif
-
 	return 0;
 }
 
Signature · 1 line
2.26.2.526.g744177e7f7-goog
#84 lore raw

PATCH 32/36 bdinfo: m68k: Move m68k-specific info into its own file

Quoted 6 lines · Simon Glass

We don't really want to have m68k-specific code in a generic file. Create a new arch-specific function to hold it, and move it into that.

Make the function weak so that any arch can implement it.

Quoted 9 lines

---

 arch/m68k/lib/Makefile |  1 +
 arch/m68k/lib/bdinfo.c | 29 +++++++++++++++++++++++++++++
 cmd/bdinfo.c           | 15 ---------------
 3 files changed, 30 insertions(+), 15 deletions(-)
 create mode 100644 arch/m68k/lib/bdinfo.c
#85 lore raw

PATCH 32/36 bdinfo: m68k: Move m68k-specific info into its own file

In reply to Simon Glass
Quoted 6 lines · Simon Glass

We don't really want to have m68k-specific code in a generic file. Create a new arch-specific function to hold it, and move it into that.

Make the function weak so that any arch can implement it.

Quoted 8 lines

---

 arch/m68k/lib/Makefile |  1 +
 arch/m68k/lib/bdinfo.c | 29 +++++++++++++++++++++++++++++
 cmd/bdinfo.c           | 15 ---------------
 3 files changed, 30 insertions(+), 15 deletions(-)
 create mode 100644 arch/m68k/lib/bdinfo.c

Tested on stmark2

stmark2 $ bdi
boot_params = 0x47d96770

DRAM bank   = 0x00000000
-> start    = 0x40000000
-> size     = 0x08000000
memstart    = 0x40000000
memsize     = 0x08000000
flashstart  = 0x00000000
flashsize   = 0x00000000
flashoffset = 0x00000000
baudrate    = 115200 bps
relocaddr   = 0x47dd5000
reloc off   = 0xfffd4c00
Build       = 32-bit
fdt_blob    = 0x47d90670
new_fdt     = 0x47d90670
fdt_size    = 0x00002860
sramstart   = 0x00000000
sramsize    = 0x00000000
busfreq     =    120 MHz
mbar        = 0xfc000000
cpufreq     =    240 MHz
flbfreq     =     60 MHz
inpfreq     =     30 MHz
#86 lore raw

PATCH 33/36 bdinfo: m68k: ppc: Move arch-specific code from bdinfo

In reply to Simon Glass

We don't have an easy way to share these three lines of code with two architectures. We also want to make it clear that this code is actually arch-specific.

So just duplicate it in each arch-specific file.

---

 arch/m68k/lib/bdinfo.c    | 5 +++++
 arch/powerpc/lib/bdinfo.c | 5 +++++
 cmd/bdinfo.c              | 8 --------
 3 files changed, 10 insertions(+), 8 deletions(-)
arch/m68k/lib/bdinfo.c +5 -0
index 09a1abfc21..971c47c306 100644
@@ -15,6 +15,11 @@ void arch_print_bdinfo(void)
 {
 	bd_t *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);
 #endif
arch/powerpc/lib/bdinfo.c +5 -0
index da09bb276f..d8c64155f0 100644
@@ -20,6 +20,11 @@ void arch_print_bdinfo(void)
 {
 	bd_t *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);
 #endif
cmd/bdinfo.c +0 -8
index f0afdb153f..aea7cf904a 100644
@@ -98,14 +98,6 @@ int do_bdinfo(cmd_tbl_t *cmdtp, int flag, int argc, char *const argv[])
 
 	arch_print_bdinfo();
 
-	/* This is used by m68k and ppc */
-#if defined(CONFIG_SYS_INIT_RAM_ADDR)
-	bdinfo_print_num("sramstart", (ulong)bd->bi_sramstart);
-	bdinfo_print_num("sramsize", (ulong)bd->bi_sramsize);
-#endif
-	if (IS_ENABLED(CONFIG_PPC) || IS_ENABLED(CONFIG_M68K))
-		bdinfo_print_mhz("busfreq", bd->bi_busfreq);
-
 	return 0;
 }
 
Signature · 1 line
2.26.2.526.g744177e7f7-goog
#87 lore raw

PATCH 33/36 bdinfo: m68k: ppc: Move arch-specific code from bdinfo

Quoted 7 lines · Simon Glass

We don't have an easy way to share these three lines of code with two architectures. We also want to make it clear that this code is actually arch-specific.

So just duplicate it in each arch-specific file.

Quoted 8 lines

---

 arch/m68k/lib/bdinfo.c    | 5 +++++
 arch/powerpc/lib/bdinfo.c | 5 +++++
 cmd/bdinfo.c              | 8 --------
 3 files changed, 10 insertions(+), 8 deletions(-)
#88 lore raw

PATCH 33/36 bdinfo: m68k: ppc: Move arch-specific code from bdinfo

In reply to Simon Glass
Quoted 7 lines · Simon Glass

We don't have an easy way to share these three lines of code with two architectures. We also want to make it clear that this code is actually arch-specific.

So just duplicate it in each arch-specific file.

Thanks,
Stefan

Quoted 49 lines

---

  arch/m68k/lib/bdinfo.c    | 5 +++++
  arch/powerpc/lib/bdinfo.c | 5 +++++
  cmd/bdinfo.c              | 8 --------
  3 files changed, 10 insertions(+), 8 deletions(-)
arch/m68k/lib/bdinfo.c +5 -0
index 09a1abfc21..971c47c306 100644
@@ -15,6 +15,11 @@ void arch_print_bdinfo(void)
  {
  	bd_t *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);
  #endif
arch/powerpc/lib/bdinfo.c +5 -0
index da09bb276f..d8c64155f0 100644
@@ -20,6 +20,11 @@ void arch_print_bdinfo(void)
  {
  	bd_t *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);
  #endif
cmd/bdinfo.c +0 -1
index f0afdb153f..aea7cf904a 100644
@@ -98,14 +98,6 @@ int do_bdinfo(cmd_tbl_t *cmdtp, int flag, int argc, char *const argv[])
 
  	arch_print_bdinfo();
 
-	/* This is used by m68k and ppc */

cmd/bdinfo.c int do_bdinfo(cmd_tbl_t *cmdtp, int flag, int argc, char *const argv[]) · line 103

Quoted 10 lines
cmd/bdinfo.c +0 -7
-#if defined(CONFIG_SYS_INIT_RAM_ADDR)
-	bdinfo_print_num("sramstart", (ulong)bd->bi_sramstart);
-	bdinfo_print_num("sramsize", (ulong)bd->bi_sramsize);
-#endif
-	if (IS_ENABLED(CONFIG_PPC) || IS_ENABLED(CONFIG_M68K))
-		bdinfo_print_mhz("busfreq", bd->bi_busfreq);
-
  	return 0;
  }
 

Viele Gr??e,
Stefan

Signature · 3 lines
DENX Software Engineering GmbH,      Managing Director: Wolfgang Denk
HRB 165235 Munich, Office: Kirchenstr.5, D-82194 Groebenzell, Germany
Phone: (+49)-8142-66989-51 Fax: (+49)-8142-66989-80 Email: sr at denx.de
#89 lore raw

PATCH 34/36 bdinfo: Update the file comments

In reply to Simon Glass

Update the comment at the top of the file to indicate what this file does. Also drop the line at the bottom and an unnecessary header include.

---

 cmd/bdinfo.c | 8 ++------
 1 file changed, 2 insertions(+), 6 deletions(-)
cmd/bdinfo.c +2 -6
index aea7cf904a..8c498d4290 100644
@@ -1,17 +1,15 @@
 // SPDX-License-Identifier: GPL-2.0+
 /*
+ * Implements the 'bd' command to show board information
+ *
  * (C) Copyright 2003
  * Wolfgang Denk, DENX Software Engineering, wd at denx.de.
  */
 
-/*
- * Boot support
- */
 #include <common.h>
 #include <command.h>
 #include <env.h>
 #include <vsprintf.h>
-#include <linux/compiler.h>
 
 DECLARE_GLOBAL_DATA_PTR;
 
@@ -101,8 +99,6 @@ int do_bdinfo(cmd_tbl_t *cmdtp, int flag, int argc, char *const argv[])
 	return 0;
 }
 
-/* -------------------------------------------------------------------- */
-
 U_BOOT_CMD(
 	bdinfo,	1,	1,	do_bdinfo,
 	"print Board Info structure",
Signature · 1 line
2.26.2.526.g744177e7f7-goog
#90 lore raw

PATCH 34/36 bdinfo: Update the file comments

Quoted 10 lines · Simon Glass

Update the comment at the top of the file to indicate what this file does. Also drop the line at the bottom and an unnecessary header include.

---

 cmd/bdinfo.c | 8 ++------
 1 file changed, 2 insertions(+), 6 deletions(-)
#91 lore raw

PATCH 35/36 bdinfo: dm: Update fb_base when using driver model

In reply to Simon Glass

Update this value with the address of a video device so that it shows with the 'bd' command.

It would be better to obtain the address from the uclass by looking in struct video_uc_platdata for each device. We can move over to that once DM_VIDEO migration is complete.

---

 drivers/video/video-uclass.c | 1 +
 1 file changed, 1 insertion(+)
drivers/video/video-uclass.c +1 -0
index 3d658e61d7..e7229cefaf 100644
@@ -82,6 +82,7 @@ int video_reserve(ulong *addrp)
 		      __func__, size, *addrp, dev->name);
 	}
 	gd->video_bottom = *addrp;
+	gd->fb_base = *addrp;
 	debug("Video frame buffers from %lx to %lx\n", gd->video_bottom,
 	      gd->video_top);
 
Signature · 1 line
2.26.2.526.g744177e7f7-goog
#92 lore raw

PATCH 35/36 bdinfo: dm: Update fb_base when using driver model

Quoted 8 lines · Simon Glass

Update this value with the address of a video device so that it shows with the 'bd' command.

It would be better to obtain the address from the uclass by looking in struct video_uc_platdata for each device. We can move over to that once DM_VIDEO migration is complete.

Quoted 6 lines

---

 drivers/video/video-uclass.c | 1 +
 1 file changed, 1 insertion(+)
#93 lore raw

PATCH 36/36 bdinfo: x86: vesa: Update fb_base to the correct value

In reply to Simon Glass

Set this value in global_data so that it is reported correctly on x86 boards.

In fact, space is allocated for the frame buffer even though it is not used. The FSP picks the address itself, which is why the current value is wrong.

---

 arch/x86/lib/fsp/fsp_graphics.c | 5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)
arch/x86/lib/fsp/fsp_graphics.c +3 -2
index 98b762209f..46fb907dc3 100644
@@ -96,6 +96,7 @@ static int fsp_video_probe(struct udevice *dev)
 	 * For IGD, it seems to be always on BAR2.
 	 */
 	vesa->phys_base_ptr = dm_pci_read_bar32(dev, 2);
+	gd->fb_base = vesa->phys_base_ptr;
 
 	ret = vbe_setup_video_priv(vesa, uc_priv, plat);
 	if (ret)
@@ -104,8 +105,8 @@ static int fsp_video_probe(struct udevice *dev)
 	mtrr_add_request(MTRR_TYPE_WRCOMB, vesa->phys_base_ptr, 256 << 20);
 	mtrr_commit(true);
 
-	printf("%dx%dx%d\n", uc_priv->xsize, uc_priv->ysize,
-	       vesa->bits_per_pixel);
+	printf("%dx%dx%d @ %x\n", uc_priv->xsize, uc_priv->ysize,
+	       vesa->bits_per_pixel, vesa->phys_base_ptr);
 
 	return 0;
 
Signature · 1 line
2.26.2.526.g744177e7f7-goog
#94 lore raw

PATCH 36/36 bdinfo: x86: vesa: Update fb_base to the correct value

Hi Simon,

Quoted 7 lines · Simon Glass

Set this value in global_data so that it is reported correctly on x86 boards.

In fact, space is allocated for the frame buffer even though it is not used. The FSP picks the address itself, which is why the current value is wrong.

I don't get it about the "current value is wrong"?

Quoted 20 lines

---

 arch/x86/lib/fsp/fsp_graphics.c | 5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)
arch/x86/lib/fsp/fsp_graphics.c +1 -0
index 98b762209f..46fb907dc3 100644
@@ -96,6 +96,7 @@ static int fsp_video_probe(struct udevice *dev)
         * For IGD, it seems to be always on BAR2.
         */
        vesa->phys_base_ptr = dm_pci_read_bar32(dev, 2);
+       gd->fb_base = vesa->phys_base_ptr;
 
        ret = vbe_setup_video_priv(vesa, uc_priv, plat);
        if (ret)

arch/x86/lib/fsp/fsp_graphics.c static int fsp_video_probe(struct udevice *dev) · line 112

Quoted 10 lines
arch/x86/lib/fsp/fsp_graphics.c +2 -2
@@ -104,8 +105,8 @@ static int fsp_video_probe(struct udevice *dev)
        mtrr_add_request(MTRR_TYPE_WRCOMB, vesa->phys_base_ptr, 256 << 20);
        mtrr_commit(true);
 
-       printf("%dx%dx%d\n", uc_priv->xsize, uc_priv->ysize,
-              vesa->bits_per_pixel);
+       printf("%dx%dx%d @ %x\n", uc_priv->xsize, uc_priv->ysize,
+              vesa->bits_per_pixel, vesa->phys_base_ptr);
 
        return 0;
 

Regards,
Bin

#95 lore raw

cover Tidy up the 'bd' command.

In reply to Simon Glass

Hi Simon,

Quoted 4 lines

-----Original Message-----
From: Simon Glass <[email protected]>
Sent: Tuesday, May 5, 2020 2:17 AM
To: U-Boot Mailing List <[email protected]>

[snip]

Quoted 17 lines

Subject: [PATCH 00/36] Tidy up the 'bd' command.

The code for the 'bd' command never got the 'generic board' treatment many years ago when global_data and bd_info were converted. As a result it still has a lot of arch-specific duplication of generic code.

This series aims to make as much code in this file generic as possible, so that it is easy to add new info on all architectures.

For the three architectures that actually need additional code (ARM, PPC and m68k) this is moved into arch-specific files.

With this series, bdinfo.c drops from nearly 500 lines to just over 100.

It also makes x86 report the frame buffer address properly (the original goal of my effort).

Quoted 14 lines

Simon Glass (36):
bdinfo: nds32: Use generic bd_info
bdinfo: riscv: Use generic bd_info
bdinfo: m68k: Drop bd_info->bi_ipbfreq
bdinfo: xtensa: Create a generic do_bdinfo for xtensa
bdinfo: mips: Use the generic bd command
bdinfo: nios2: Use the generic bd command
bdinfo: microblaze: Use the generic bd command
bdinfo: sh: Use the generic bd command
bdinfo: x86: Use the generic bd command
bdinfo: sandbox: Use the generic bd command
bdinfo: nds32: Use the generic bd command
bdinfo: riscv: Use the generic bd command
bdinfo: arm: Use the generic bd command

Looks like this one contains much more than just ARM changes. Instead it looks like a massive migration to a generic bdinfo. So I would propose either split it into per-arch changes (though given this series size it might not be a good idea) or alternatively maybe just ask for affected architecture maintainers for an ack?

Also IMHO it would be very convenient to have a link to a git tree with all these changes - it will significantly simplify review process.

But regardless all my comments above I'd like to note that this series is very welcome as a long needed clean-up of that per-arch nonsense. Thanks for doing this significant and useful work!

-Alexey

#96 lore raw

cover Tidy up the 'bd' command.

Hi Alexey,

Quoted 42 lines · Alexey Brodkin

Hi Simon,

Quoted 4 lines

-----Original Message-----
From: Simon Glass <[email protected]>
Sent: Tuesday, May 5, 2020 2:17 AM
To: U-Boot Mailing List <[email protected]>

[snip]

Quoted 17 lines

Subject: [PATCH 00/36] Tidy up the 'bd' command.

The code for the 'bd' command never got the 'generic board' treatment many years ago when global_data and bd_info were converted. As a result it still has a lot of arch-specific duplication of generic code.

This series aims to make as much code in this file generic as possible, so that it is easy to add new info on all architectures.

For the three architectures that actually need additional code (ARM, PPC and m68k) this is moved into arch-specific files.

With this series, bdinfo.c drops from nearly 500 lines to just over 100.

It also makes x86 report the frame buffer address properly (the original goal of my effort).

Quoted 14 lines

Simon Glass (36):
bdinfo: nds32: Use generic bd_info
bdinfo: riscv: Use generic bd_info
bdinfo: m68k: Drop bd_info->bi_ipbfreq
bdinfo: xtensa: Create a generic do_bdinfo for xtensa
bdinfo: mips: Use the generic bd command
bdinfo: nios2: Use the generic bd command
bdinfo: microblaze: Use the generic bd command
bdinfo: sh: Use the generic bd command
bdinfo: x86: Use the generic bd command
bdinfo: sandbox: Use the generic bd command
bdinfo: nds32: Use the generic bd command
bdinfo: riscv: Use the generic bd command
bdinfo: arm: Use the generic bd command

Quoted 6 lines

Looks like this one contains much more than just ARM changes. Instead it looks like a massive migration to a generic bdinfo. So I would propose either split it into per-arch changes (though given this series size it might not be a good idea) or alternatively maybe just ask for affected architecture maintainers for an ack?

This is not a very large migration, just a small change to one command, mostly. We already have generic bdinfo (same 'struct bd') for all but two archs.

The series is split by arch - see the tags for each commit. That is why some maintainers get some patches and not others, or at least that is my hope, since patman runs get_maintainers separately on each commit.

Quoted 3 lines

Also IMHO it would be very convenient to have a link to a git tree with all these changes - it will significantly simplify review process.

See u-boot-dm/bd-working for the tree. I always post a series there before sending, but don't always mention the name, sorry.

Quoted 3 lines

But regardless all my comments above I'd like to note that this series is very welcome as a long needed clean-up of that per-arch nonsense. Thanks for doing this significant and useful work!

:-)

Regards,
Simon