sunxi: minor cleanups and refactoring
Subsystems (3)
cover sunxi: minor cleanups and refactoring
Just a collection of patches I came up with while trying to refactor and reorganise the Allwinner SPL code.
Patch 1/4 aims to replace Tom's patch [1], which made CONFIG_MMC_SUNXI_SLOT a proper Kconfig variable. A closer inspection reveals that we don't really need that at all.
Patch 2 and 3 are just random cleanups.
Patch 4 is a low hanging fruit in the effort to separate ARM specific code from strictly Allwinner platform routines. It also removes some #ifdef's.
Please have a look!
Cheers,
Andre
[1] https://lore.kernel.org/u-boot/[email protected]/
Andre Przywara (4):
sunxi: remove unused CONFIG_MMC_SUNXI_SLOT
sunxi: remove bogus mmc_pinmux_setup() prototype
sunxi: board: annotate #endif lines
sunxi: move arch timer setup out of board/ directory
arch/arm/mach-sunxi/Makefile | 4 +++ arch/arm/mach-sunxi/arch_timer.c | 39 +++++++++++++++++++++ arch/arm/mach-sunxi/board.c | 22 +++++++----- arch/arm/mach-sunxi/clock_sun6i.c | 4 +-- board/sunxi/board.c | 58 +++++++------------------------ include/configs/sunxi-common.h | 3 -- scripts/config_whitelist.txt | 1 - 7 files changed, 72 insertions(+), 59 deletions(-) create mode 100644 arch/arm/mach-sunxi/arch_timer.c
Signature · 1 line
2.35.5
PATCH 1/4 sunxi: remove unused CONFIG_MMC_SUNXI_SLOT
There is a CONFIG_MMC_SUNXI_SLOT definition in our sunxi_common.h config header, which was used to note the first MMC controller to initialise. The definition in that header was always set to 0, with no easy way of overriding this, and certainly none of the existing boards made any use of that (non-)feature.
Remove that definition and replace it with a constant 0 in the only user, in board.c. It turns out that this is safe, as this is only used in the SPL, and the BROM also unconditionally initialises MMC0. This also removes the last legacy config symbol with SUN*I in it from the whitelist.
--- board/sunxi/board.c | 9 +++++++-- include/configs/sunxi-common.h | 3 --- scripts/config_whitelist.txt | 1 - 3 files changed, 7 insertions(+), 6 deletions(-)
board/sunxi/board.c +7 -2
@@ -525,9 +525,14 @@ static void mmc_pinmux_setup(int sdc)int board_mmc_init (struct bd_info *bis) { +/* +* The BROM always accesses MMC port 0 (typically an SD card), and +* most boards seem to have such a slot. The others haven't reported +* any problem with unconditionally enabling this in the SPL. +*/ if (!IS_ENABLED (CONFIG_UART0_PORT_F )) { -mmc_pinmux_setup (CONFIG_MMC_SUNXI_SLOT ); -if (!sunxi_mmc_init (CONFIG_MMC_SUNXI_SLOT )) +mmc_pinmux_setup (0 ); +if (!sunxi_mmc_init (0 ))return -1 ; }
include/configs/sunxi-common.h +0 -3
@@ -75,9 +75,6 @@#define CONFIG_SYS_NAND_MAX_ECCPOS 1664#endif -/* mmc config */ -#define CONFIG_MMC_SUNXI_SLOT 0 -/* * Miscellaneous configurable options */
scripts/config_whitelist.txt +0 -1
@@ -153,7 +153,6 @@ CONFIG_MISC_COMMON CONFIG_MIU_2BIT_21_7_INTERLEAVED CONFIG_MIU_2BIT_INTERLEAVED CONFIG_MMC_DEFAULT_DEV -CONFIG_MMC_SUNXI_SLOT CONFIG_MONITOR_IS_IN_RAM CONFIG_MPC85XX_FEC CONFIG_MPC85XX_FEC_NAME
Signature · 1 line
2.35.5
Re: [PATCH 1/4] sunxi: remove unused CONFIG_MMC_SUNXI_SLOT
Quoted 11 lines · Andre Przywara
There is a
CONFIG_MMC_SUNXI_SLOTdefinition in oursunxi_common.hconfig header, which was used to note the first MMC controller to initialise. The definition in that header was always set to 0, with no easy way of overriding this, and certainly none of the existing boards made any use of that (non-)feature.
Remove that definition and replace it with a constant 0 in the only user, inboard.c. It turns out that this is safe, as this is only used in the SPL, and the BROM also unconditionally initialises MMC0. This also removes the last legacy config symbol with SUN*I in it from the whitelist.
Quoted 6 lines
Signed-off-by Andre Przywara <[email protected]>--- board/sunxi/board.c | 9 +++++++-- include/configs/sunxi-common.h | 3 --- scripts/config_whitelist.txt | 1 - 3 files changed, 7 insertions(+), 6 deletions(-)
PATCH 2/4 sunxi: remove bogus mmc_pinmux_setup() prototype
In reply to Andre PrzywaraSince all callers of mmc_pinmux_setup() are located after the definition of that function, there is no need for a forward declaration (anymore?).
Remove the prototype along with its #ifdef guards.
--- board/sunxi/board.c | 4 ---- 1 file changed, 4 deletions(-)
board/sunxi/board.c +0 -4
@@ -184,10 +184,6 @@ enum env_location env_get_location(enum env_operation op, int prio)return ENVL_UNKNOWN ; } -#ifdef CONFIG_DM_MMC -static void mmc_pinmux_setup (int sdc); -#endif -/* add board specific code here */ int board_init (void ) {
Signature · 1 line
2.35.5
Re: [PATCH 2/4] sunxi: remove bogus mmc_pinmux_setup() prototype
Quoted 9 lines · Andre Przywara
Since all callers of
mmc_pinmux_setup()are located after the definition of that function, there is no need for a forward declaration (anymore?).Remove the prototype along with its #ifdef guards.
Signed-off-by Andre Przywara <[email protected]>--- board/sunxi/board.c | 4 ---- 1 file changed, 4 deletions(-)
PATCH 3/4 sunxi: board: annotate #endif lines
In reply to Andre PrzywaraThe legacy Allwinner code is cluttered with #ifdef's, some of them even nested, which makes the code hard to read and error prone. Eventually we will get rid of most of them, but for now let's at least annotate the #endif lines with the corresponding symbol the bracket started with.
--- arch/arm/mach-sunxi/board.c | 8 ++++---- arch/arm/mach-sunxi/clock_sun6i.c | 4 ++-- board/sunxi/board.c | 11 +++++------ 3 files changed, 11 insertions(+), 12 deletions(-)
arch/arm/mach-sunxi/board.c +4 -4
@@ -73,7 +73,7 @@ phys_size_t board_get_usable_ram_top(phys_size_t total_size) return gd->ram_top; } -#endif +#endif /* CONFIG_ARM64 */ #ifdef CONFIG_SPL_BUILD static int gpio_init (void )
@@ -196,7 +196,7 @@ static int spl_board_load_image(struct spl_image_info *spl_image, return 0; }SPL_LOAD_IMAGE_METHOD ("FEL" ,0 ,BOOT_DEVICE_BOARD , spl_board_load_image); -#endif +#endif /* CONFIG_SPL_BUILD */ #define SUNXI_INVALID_BOOT_SOURCE -1
@@ -457,7 +457,7 @@ void board_init_f(ulong dummy)#endif sunxi_board_init(); } -#endif +#endif /* CONFIG_SPL_BUILD */ #if !CONFIG_IS_ENABLED (SYSRESET )void reset_cpu (void )
@@ -490,7 +490,7 @@ void reset_cpu(void)while (1 ) { } #endif } -#endif +#endif /* CONFIG_SYSRESET */ #if !CONFIG_IS_ENABLED (SYS_DCACHE_OFF ) && defined(CONFIG_CPU_V7A )void enable_caches (void )
arch/arm/mach-sunxi/clock_sun6i.c +2 -2
@@ -63,7 +63,7 @@ void clock_init_safe(void)setbits_le32 (&ccm->sata_clk_cfg,CCM_SATA_CTRL_ENABLE ); #endif } -#endif +#endif /* CONFIG_SPL_BUILD */ void clock_init_sec (void ) {
@@ -172,7 +172,7 @@ void clock_set_pll1(unsigned int clk) &ccm->cpu_axi_cfg); } } -#endif +#endif /* CONFIG_SPL_BUILD */ void clock_set_pll3 (unsigned int clk) {
board/sunxi/board.c +5 -6
@@ -351,7 +351,7 @@ void board_nand_init(void)sunxi_nand_init (); #endif } -#endif +#endif /* CONFIG_NAND_SUNXI */ #ifdef CONFIG_MMC static void mmc_pinmux_setup (int sdc)
@@ -554,7 +554,7 @@ int mmc_get_env_dev(void) } }#endif -#endif +#endif /* CONFIG_MMC */ #ifdef CONFIG_SPL_BUILD
@@ -670,7 +670,7 @@ void sunxi_board_init(void) else printf("Failed to set core voltage! Can't set CPU frequency\n"); } -#endif +#endif /* CONFIG_SPL_BUILD */ #ifdef CONFIG_USB_GADGET int g_dnl_board_usb_cable_connected (void )
@@ -699,7 +699,7 @@ int g_dnl_board_usb_cable_connected(void) return sun4i_usb_phy_vbus_detect(&phy); } -#endif +#endif /* CONFIG_USB_GADGET */ #ifdef CONFIG_SERIAL_TAG void get_board_serial (struct tag_serialnr *serialnr)
@@ -928,7 +928,6 @@ int ft_board_setup(void *blob, struct bd_info *bd) }#ifdef CONFIG_SPL_LOAD_FIT -static void set_spl_dt_name (const char *name) {struct boot_file_head *spl =get_spl_header (SPL_ENV_HEADER_VERSION );
@@ -996,4 +995,4 @@ int board_fit_config_name_match(const char *name) return ret; } -#endif +#endif /* CONFIG_SPL_LOAD_FIT */
Signature · 1 line
2.35.5
Re: [PATCH 3/4] sunxi: board: annotate #endif lines
Quoted 6 lines · Andre Przywara
The legacy Allwinner code is cluttered with #ifdef's, some of them even nested, which makes the code hard to read and error prone. Eventually we will get rid of most of them, but for now let's at least annotate the #endif lines with the corresponding symbol the bracket started with.
Quoted 6 lines
Signed-off-by Andre Przywara <[email protected]>--- arch/arm/mach-sunxi/board.c | 8 ++++---- arch/arm/mach-sunxi/clock_sun6i.c | 4 ++-- board/sunxi/board.c | 11 +++++------ 3 files changed, 11 insertions(+), 12 deletions(-)
PATCH 4/4 sunxi: move arch timer setup out of board/ directory
In reply to Andre PrzywaraAt the moment we have an #ifdef-protected routine in board/sunxi/board.c, which sets up the arch timer CNTFRQ register, if that hasn't been done already.
This only applies to ARMv7 SoCs running in secure state, and so is much better located in the arch/arm/mach-sunxi directory.
Move that routine into a separate function and file in said directory, which also allows to trigger the compilation for ARMv7 in the Makefile. Also move the call to it into the arch/arm version of board.c.
--- arch/arm/mach-sunxi/Makefile | 4 ++++ arch/arm/mach-sunxi/arch_timer.c | 39 ++++++++++++++++++++++++++++++++ arch/arm/mach-sunxi/board.c | 14 ++++++++---- board/sunxi/board.c | 34 +--------------------------- 4 files changed, 54 insertions(+), 37 deletions(-) create mode 100644 arch/arm/mach-sunxi/arch_timer.c
arch/arm/mach-sunxi/Makefile +4 -0
@@ -26,8 +26,12 @@ obj-$(CONFIG_MACH_SUN8I) += clock_sun6i.o endif obj-$(CONFIG_MACH_SUN9I) += clock_sun9i.o gtbus_sun9i.o obj-$(CONFIG_SUN50I_GEN_H6) += clock_sun50i_h6.o + ifndef CONFIG_ARM64 obj-y += timer.o +ifndef CONFIG_MACH_SUNIV +obj-y += arch_timer.o +endif endif ifdef CONFIG_SPL_BUILD
arch/arm/mach-sunxi/arch_timer.c new file +39 -0
@@ -0,0 +1,39 @@ +// SPDX-License-Identifier: GPL-2.0+ +/* +* ARM Generic Timer CNTFRQ setup +* +* (C) Copyright 2013 Oliver Schinagl <[email protected]> +*/ + +#include <stdint.h> +#include <log.h> +#include <asm/armv7.h> + +void setup_arch_timer (void ) +{ +uint32_t id_pfr1, freq; + + asmvolatile ("mrc p15, 0, %0, c0, c1, 1" :"=r" (id_pfr1)); +debug ("id_pfr1: 0x%08x\n" , id_pfr1); +/* Generic Timer Extension available? */ +if (((id_pfr1 >>CPUID_ARM_GENTIMER_SHIFT ) &0xf ) ==0 ) +return ; + +/* +* CNTFRQ is a secure register, so we will crash if we try to +* write this from the non-secure world (read is OK, though). +* In case some bootcode has already set the correct value, +* we avoid the risk of writing to it. +*/ + asmvolatile ("mrc p15, 0, %0, c14, c0, 0" :"=r" (freq)); +if (freq ==CONFIG_COUNTER_FREQUENCY ) +return ; + +debug ("arch timer frequency is %d Hz, should be %d, fixing ...\n" , + freq,CONFIG_COUNTER_FREQUENCY ); +if (IS_ENABLED (CONFIG_NON_SECURE )) +printf ("arch timer frequency is wrong, but cannot adjust it\n" ); +else + asmvolatile ("mcr p15, 0, %0, c14, c0, 0" + : :"r" (CONFIG_COUNTER_FREQUENCY )); +}
arch/arm/mach-sunxi/board.c +10 -4
@@ -347,10 +347,6 @@ u32 spl_boot_device(void)return sunxi_get_boot_device (); } -__weak void sunxi_sram_init (void ) -{ -} -/* * When booting from an eMMC boot partition, the SPL puts the same boot * source code into SRAM A1 as when loading the SPL from the normal
@@ -434,10 +430,20 @@ u32 spl_mmc_boot_mode(struct mmc *mmc, const u32 boot_device)return result; } +__weak void sunxi_sram_init (void ) +{ +} + +__weak void setup_arch_timer (void ) +{ +} +void board_init_f (ulong dummy) {sunxi_sram_init (); +setup_arch_timer (); +#if definedCONFIG_MACH_SUN6I || definedCONFIG_MACH_SUN8I_H3 /* Enable non-secure access to some peripherals */ tzpc_init ();
board/sunxi/board.c +1 -33
@@ -34,9 +34,6 @@#include <asm/global_data.h> #include <linux/delay.h> #include <u-boot/crc.h> -#ifndef CONFIG_ARM64 -#include <asm/armv7.h> -#endif #include <asm/gpio.h> #include <asm/io.h> #include <u-boot/crc.h>
@@ -187,39 +184,10 @@ enum env_location env_get_location(enum env_operation op, int prio)/* add board specific code here */ int board_init (void ) { -__maybe_unused int id_pfr1, ret, satapwr_pin, macpwr_pin; +int ret, satapwr_pin, macpwr_pin; gd->bd->bi_boot_params = (PHYS_SDRAM_0 +0x100 ); -#if !defined(CONFIG_ARM64 ) && !defined(CONFIG_MACH_SUNIV ) - asmvolatile ("mrc p15, 0, %0, c0, c1, 1" :"=r" (id_pfr1)); -debug ("id_pfr1: 0x%08x\n" , id_pfr1); -/* Generic Timer Extension available? */ -if ((id_pfr1 >>CPUID_ARM_GENTIMER_SHIFT ) &0xf ) { -uint32_t freq; - -debug ("Setting CNTFRQ\n" ); - -/* -* CNTFRQ is a secure register, so we will crash if we try to -* write this from the non-secure world (read is OK, though). -* In case some bootcode has already set the correct value, -* we avoid the risk of writing to it. -*/ - asmvolatile ("mrc p15, 0, %0, c14, c0, 0" :"=r" (freq)); -if (freq !=CONFIG_COUNTER_FREQUENCY ) { -debug ("arch timer frequency is %d Hz, should be %d, fixing ...\n" , - freq,CONFIG_COUNTER_FREQUENCY ); -#ifdef CONFIG_NON_SECURE -printf ("arch timer frequency is wrong, but cannot adjust it\n" ); -#else - asmvolatile ("mcr p15, 0, %0, c14, c0, 0" - : :"r" (CONFIG_COUNTER_FREQUENCY )); -#endif - } - } -#endif /* !CONFIG_ARM64 && !CONFIG_MACH_SUNIV */ - ret =axp_gpio_init ();if (ret)return ret;
Signature · 1 line
2.35.5
Re: [PATCH 4/4] sunxi: move arch timer setup out of board/ directory
Hi Andre,
Quoted 24 lines · Andre Przywara
At the moment we have an #ifdef-protected routine in
board/sunxi/board.c, which sets up the arch timer CNTFRQ register, if that hasn't been done already.
This only applies to ARMv7 SoCs running in secure state, and so is much better located in thearch/arm/mach-sunxidirectory.Move that routine into a separate function and file in said directory, which also allows to trigger the compilation for ARMv7 in the
Makefile. Also move the call to it into thearch/armversion ofboard.c.Signed-off-by Andre Przywara <[email protected]>--- arch/arm/mach-sunxi/Makefile | 4 ++++ arch/arm/mach-sunxi/arch_timer.c | 39 ++++++++++++++++++++++++++++++++ arch/arm/mach-sunxi/board.c | 14 ++++++++---- board/sunxi/board.c | 34 +--------------------------- 4 files changed, 54 insertions(+), 37 deletions(-) create mode 100644 arch/arm/mach-sunxi/arch_timer.carch/arm/mach-sunxi/Makefile +0
-0
arch/arm/mach-sunxi/Makefile obj-$(CONFIG_MACH_SUN8I) += clock_sun6i.o · line 32
Quoted 7 lines
arch/arm/mach-sunxi/Makefile +2
-0@@ -26,8 +26,12 @@ obj-$(CONFIG_MACH_SUN8I) += clock_sun6i.o endif obj-$(CONFIG_MACH_SUN9I) += clock_sun9i.o gtbus_sun9i.o obj-$(CONFIG_SUN50I_GEN_H6) += clock_sun50i_h6.o + ifndef CONFIG_ARM64 obj-y += timer.o +ifndef CONFIG_MACH_SUNIV
Probably Allwinner will not release any new ARMv5 SoC, but would it be more future-proof (and a bit more meaningful) to condition this on CONFIG_CPU_V7A?
Alternatively, would it make sense to put this code inside arch/arm/cpu/armv7/arch_timer.c? I don't know why we wouldn't want to enable CONFIG_SYS_ARCH_TIMER on sunxi SoCs where it is available. And making this generic could eventually remove the need for the version in arch/arm/cpu/armv7/nonsec_virt.S. But this patch is still an improvement as-is.
Quoted 11 lines
arch/arm/mach-sunxi/Makefile +2
-0+obj-y += arch_timer.o +endif endif ifdef CONFIG_SPL_BUILDarch/arm/mach-sunxi/arch_timer.c new file +0
-0
arch/arm/mach-sunxi/arch_timer.c line 5
Quoted 5 lines
arch/arm/mach-sunxi/arch_timer.c new file +5
-0@@ -0,0 +1,39 @@ +// SPDX-License-Identifier: GPL-2.0+ +/* +* ARM Generic Timer CNTFRQ setup +* +* (C) Copyright 2013 Oliver Schinagl <[email protected]>
git blame points to cba69eeeaa67d3fb93ec6f3abab1f653abf895a9 and d96ebc468d0dff6eb6f069bba03b3f0e33aa22de as the source for this code, so this authorship is incorrect.
arch/arm/mach-sunxi/arch_timer.c line 8
Quoted 3 lines
arch/arm/mach-sunxi/arch_timer.c new file +3
-0+*/ + +#include <stdint.h>
This does not do what you expect, or really anything at all. The comment there suggests using linux/types.h.
Regards,
Samuel
Quoted 122 lines
arch/arm/mach-sunxi/arch_timer.c new file +31
-0+#include <log.h> +#include <asm/armv7.h> + +void setup_arch_timer (void ) +{ +uint32_t id_pfr1, freq; + + asmvolatile ("mrc p15, 0, %0, c0, c1, 1" :"=r" (id_pfr1)); +debug ("id_pfr1: 0x%08x\n" , id_pfr1); +/* Generic Timer Extension available? */ +if (((id_pfr1 >>CPUID_ARM_GENTIMER_SHIFT ) &0xf ) ==0 ) +return ; + +/* +* CNTFRQ is a secure register, so we will crash if we try to +* write this from the non-secure world (read is OK, though). +* In case some bootcode has already set the correct value, +* we avoid the risk of writing to it. +*/ + asmvolatile ("mrc p15, 0, %0, c14, c0, 0" :"=r" (freq)); +if (freq ==CONFIG_COUNTER_FREQUENCY ) +return ; + +debug ("arch timer frequency is %d Hz, should be %d, fixing ...\n" , + freq,CONFIG_COUNTER_FREQUENCY ); +if (IS_ENABLED (CONFIG_NON_SECURE )) +printf ("arch timer frequency is wrong, but cannot adjust it\n" ); +else + asmvolatile ("mcr p15, 0, %0, c14, c0, 0" + : :"r" (CONFIG_COUNTER_FREQUENCY )); +}arch/arm/mach-sunxi/board.c +10
-4@@ -347,10 +347,6 @@ u32 spl_boot_device(void)return sunxi_get_boot_device (); } -__weak void sunxi_sram_init (void ) -{ -} -/* * When booting from an eMMC boot partition, the SPL puts the same boot * source code into SRAM A1 as when loading the SPL from the normal @@ -434,10 +430,20 @@ u32 spl_mmc_boot_mode(struct mmc *mmc, const u32 boot_device)return result; } +__weak void sunxi_sram_init (void ) +{ +} + +__weak void setup_arch_timer (void ) +{ +} +void board_init_f (ulong dummy) {sunxi_sram_init (); +setup_arch_timer (); +#if definedCONFIG_MACH_SUN6I || definedCONFIG_MACH_SUN8I_H3 /* Enable non-secure access to some peripherals */ tzpc_init ();board/sunxi/board.c +1
-33@@ -34,9 +34,6 @@#include <asm/global_data.h> #include <linux/delay.h> #include <u-boot/crc.h> -#ifndef CONFIG_ARM64 -#include <asm/armv7.h> -#endif #include <asm/gpio.h> #include <asm/io.h> #include <u-boot/crc.h> @@ -187,39 +184,10 @@ enum env_location env_get_location(enum env_operation op, int prio)/* add board specific code here */ int board_init (void ) { -__maybe_unused int id_pfr1, ret, satapwr_pin, macpwr_pin; +int ret, satapwr_pin, macpwr_pin; gd->bd->bi_boot_params = (PHYS_SDRAM_0 +0x100 ); -#if !defined(CONFIG_ARM64 ) && !defined(CONFIG_MACH_SUNIV ) - asmvolatile ("mrc p15, 0, %0, c0, c1, 1" :"=r" (id_pfr1)); -debug ("id_pfr1: 0x%08x\n" , id_pfr1); -/* Generic Timer Extension available? */ -if ((id_pfr1 >>CPUID_ARM_GENTIMER_SHIFT ) &0xf ) { -uint32_t freq; - -debug ("Setting CNTFRQ\n" ); - -/* -* CNTFRQ is a secure register, so we will crash if we try to -* write this from the non-secure world (read is OK, though). -* In case some bootcode has already set the correct value, -* we avoid the risk of writing to it. -*/ - asmvolatile ("mrc p15, 0, %0, c14, c0, 0" :"=r" (freq)); -if (freq !=CONFIG_COUNTER_FREQUENCY ) { -debug ("arch timer frequency is %d Hz, should be %d, fixing ...\n" , - freq,CONFIG_COUNTER_FREQUENCY ); -#ifdef CONFIG_NON_SECURE -printf ("arch timer frequency is wrong, but cannot adjust it\n" ); -#else - asmvolatile ("mcr p15, 0, %0, c14, c0, 0" - : :"r" (CONFIG_COUNTER_FREQUENCY )); -#endif - } - } -#endif /* !CONFIG_ARM64 && !CONFIG_MACH_SUNIV */ - ret =axp_gpio_init ();if (ret)return ret;
Re: [PATCH 4/4] sunxi: move arch timer setup out of board/ directory
On Wed, 14 Dec 2022 00:14:49 -0600
Samuel Holland <[email protected]> wrote:
Hi Samuel,
thanks for having a look!
Quoted 20 lines · Andre Przywara
Quoted 19 lines
At the moment we have an #ifdef-protected routine in
board/sunxi/board.c, which sets up the arch timer CNTFRQ register, if that hasn't been done already.
This only applies to ARMv7 SoCs running in secure state, and so is much better located in thearch/arm/mach-sunxidirectory.Move that routine into a separate function and file in said directory, which also allows to trigger the compilation for ARMv7 in the
Makefile. Also move the call to it into thearch/armversion ofboard.c.Signed-off-by Andre Przywara <[email protected]>--- arch/arm/mach-sunxi/Makefile | 4 ++++ arch/arm/mach-sunxi/arch_timer.c | 39 ++++++++++++++++++++++++++++++++ arch/arm/mach-sunxi/board.c | 14 ++++++++---- board/sunxi/board.c | 34 +--------------------------- 4 files changed, 54 insertions(+), 37 deletions(-) create mode 100644 arch/arm/mach-sunxi/arch_timer.c
Quoted 16 lines
arch/arm/mach-sunxi/Makefile obj-$(CONFIG_MACH_SUN8I) += clock_sun6i.o · line 32
Quoted 12 lines
arch/arm/mach-sunxi/Makefile +2
-0@@ -26,8 +26,12 @@ obj-$(CONFIG_MACH_SUN8I) += clock_sun6i.o endif obj-$(CONFIG_MACH_SUN9I) += clock_sun9i.o gtbus_sun9i.o obj-$(CONFIG_SUN50I_GEN_H6) += clock_sun50i_h6.o + ifndef CONFIG_ARM64 obj-y += timer.o +ifndef CONFIG_MACH_SUNIVProbably Allwinner will not release any new ARMv5 SoC, but would it be more future-proof (and a bit more meaningful) to condition this on
CONFIG_CPU_V7A?
Ah, of course, I forgot about that symbol, and just saw !CONFIG_ARM64.
Quoted 6 lines
Alternatively, would it make sense to put this code inside
arch/arm/cpu/armv7/arch_timer.c? I don't know why we wouldn't want to enableCONFIG_SYS_ARCH_TIMERon sunxi SoCs where it is available. And making this generic could eventually remove the need for the version inarch/arm/cpu/armv7/nonsec_virt.S. But this patch is still an improvement as-is.
Yeah, that indeed sounds even better, I will give it a try. I guess sunxi will stay the only user for now, but we can extend this later on.
Thanks for the suggestion!
Cheers,
Andre
Quoted 156 lines
Tested-by Samuel Holland <[email protected]>Quoted 11 lines
arch/arm/mach-sunxi/Makefile +2
-0+obj-y += arch_timer.o +endif endif ifdef CONFIG_SPL_BUILDarch/arm/mach-sunxi/arch_timer.c new file +0
-0arch/arm/mach-sunxi/arch_timer.c line 5
Quoted 5 lines
arch/arm/mach-sunxi/arch_timer.c new file +5
-0@@ -0,0 +1,39 @@ +// SPDX-License-Identifier: GPL-2.0+ +/* +* ARM Generic Timer CNTFRQ setup +* +* (C) Copyright 2013 Oliver Schinagl <[email protected]>
git blamepoints tocba69eeeaa67d3fb93ec6f3abab1f653abf895a9andd96ebc468d0dff6eb6f069bba03b3f0e33aa22deas the source for this code, so this authorship is incorrect.arch/arm/mach-sunxi/arch_timer.c line 8
Quoted 3 lines
arch/arm/mach-sunxi/arch_timer.c new file +3
-0+*/ + +#include <stdint.h> This does not do what you expect, or really anything at all. The comment there suggests using
linux/types.h.Regards,
SamuelQuoted 122 lines
arch/arm/mach-sunxi/arch_timer.c new file +31
-0+#include <log.h> +#include <asm/armv7.h> + +void setup_arch_timer (void ) +{ +uint32_t id_pfr1, freq; + + asmvolatile ("mrc p15, 0, %0, c0, c1, 1" :"=r" (id_pfr1)); +debug ("id_pfr1: 0x%08x\n" , id_pfr1); +/* Generic Timer Extension available? */ +if (((id_pfr1 >>CPUID_ARM_GENTIMER_SHIFT ) &0xf ) ==0 ) +return ; + +/* +* CNTFRQ is a secure register, so we will crash if we try to +* write this from the non-secure world (read is OK, though). +* In case some bootcode has already set the correct value, +* we avoid the risk of writing to it. +*/ + asmvolatile ("mrc p15, 0, %0, c14, c0, 0" :"=r" (freq)); +if (freq ==CONFIG_COUNTER_FREQUENCY ) +return ; + +debug ("arch timer frequency is %d Hz, should be %d, fixing ...\n" , + freq,CONFIG_COUNTER_FREQUENCY ); +if (IS_ENABLED (CONFIG_NON_SECURE )) +printf ("arch timer frequency is wrong, but cannot adjust it\n" ); +else + asmvolatile ("mcr p15, 0, %0, c14, c0, 0" + : :"r" (CONFIG_COUNTER_FREQUENCY )); +}arch/arm/mach-sunxi/board.c +10
-4@@ -347,10 +347,6 @@ u32 spl_boot_device(void)return sunxi_get_boot_device (); } -__weak void sunxi_sram_init (void ) -{ -} -/* * When booting from an eMMC boot partition, the SPL puts the same boot * source code into SRAM A1 as when loading the SPL from the normal @@ -434,10 +430,20 @@ u32 spl_mmc_boot_mode(struct mmc *mmc, const u32 boot_device)return result; } +__weak void sunxi_sram_init (void ) +{ +} + +__weak void setup_arch_timer (void ) +{ +} +void board_init_f (ulong dummy) {sunxi_sram_init (); +setup_arch_timer (); +#if definedCONFIG_MACH_SUN6I || definedCONFIG_MACH_SUN8I_H3 /* Enable non-secure access to some peripherals */ tzpc_init ();board/sunxi/board.c +1
-33@@ -34,9 +34,6 @@#include <asm/global_data.h> #include <linux/delay.h> #include <u-boot/crc.h> -#ifndef CONFIG_ARM64 -#include <asm/armv7.h> -#endif #include <asm/gpio.h> #include <asm/io.h> #include <u-boot/crc.h> @@ -187,39 +184,10 @@ enum env_location env_get_location(enum env_operation op, int prio)/* add board specific code here */ int board_init (void ) { -__maybe_unused int id_pfr1, ret, satapwr_pin, macpwr_pin; +int ret, satapwr_pin, macpwr_pin; gd->bd->bi_boot_params = (PHYS_SDRAM_0 +0x100 ); -#if !defined(CONFIG_ARM64 ) && !defined(CONFIG_MACH_SUNIV ) - asmvolatile ("mrc p15, 0, %0, c0, c1, 1" :"=r" (id_pfr1)); -debug ("id_pfr1: 0x%08x\n" , id_pfr1); -/* Generic Timer Extension available? */ -if ((id_pfr1 >>CPUID_ARM_GENTIMER_SHIFT ) &0xf ) { -uint32_t freq; - -debug ("Setting CNTFRQ\n" ); - -/* -* CNTFRQ is a secure register, so we will crash if we try to -* write this from the non-secure world (read is OK, though). -* In case some bootcode has already set the correct value, -* we avoid the risk of writing to it. -*/ - asmvolatile ("mrc p15, 0, %0, c14, c0, 0" :"=r" (freq)); -if (freq !=CONFIG_COUNTER_FREQUENCY ) { -debug ("arch timer frequency is %d Hz, should be %d, fixing ...\n" , - freq,CONFIG_COUNTER_FREQUENCY ); -#ifdef CONFIG_NON_SECURE -printf ("arch timer frequency is wrong, but cannot adjust it\n" ); -#else - asmvolatile ("mcr p15, 0, %0, c14, c0, 0" - : :"r" (CONFIG_COUNTER_FREQUENCY )); -#endif - } - } -#endif /* !CONFIG_ARM64 && !CONFIG_MACH_SUNIV */ - ret =axp_gpio_init ();if (ret)return ret;