* [U-Boot] [PATCH v5 00/18] rockchip: back-to-bootrom: replace assembly-implementation with C-code
@ 2017-10-10 14:21 Philipp Tomsich
2017-10-10 14:21 ` [U-Boot] [PATCH v5 01/18] arm: boot0 hook: move boot0 hook before '_start' Philipp Tomsich
` (18 more replies)
0 siblings, 19 replies; 46+ messages in thread
From: Philipp Tomsich @ 2017-10-10 14:21 UTC (permalink / raw)
To: u-boot
Recent discussions confirmed (what the code always assumed): the
Rockchip BROM always enters U-Boot with the stack-pointer valid
(i.e. the U-Boot startup code is running off the BROM stack).
We can thus replace the back-to-bootrom code (i.e. both the
save_boot_params and back_to_bootrom implementations) using C-code
based on setjmp/longjmp. The new implementation is already structured
to allow an easy drop-in of Andy's changes to enter download-mode when
returning to the BROM.
This turned out to require a some tweaking to system.h (making sure
that the prototype for save_boot_params_ret is visible for A64)and
start.S (so binutils knows that this is a possible function entry and
it can correctly insert A32-to-Thumb transitions) and taking the axe
to setjmp.h (which created quite a few issues with it not expecting
A32/T32/Thumb call-sites and some fragility from GCC being smart about
the clobber-list of the inline assembly... which led to r9 not being
saved or restored).
For v4+: To fix issues with the RK3188 support, this also updates the
boot0 hook changes (i.e. allowing the boot0-hook to insert code/data
before the ARM vector table) that Kever had submitted this spring and
implements a similar "early back-to-bootrom" as suggested by Pawel
(for the RK3066) as a generic mechanism.
Changes in v5:
- set return value to 0 before returning to the BROM
Changes in v4:
- after merging the 'back-to-bootrom' series with the 'boot0-hook'
series, this drops the TPL stub and builds only a single SPL image
that uses the 'early back-to-bootrom' logic originally implemented
by Pawel for the RK3066.
- changes the SPL_STACK_BASE to +0x800 (from +0x804), as the boot0
hook already reserves the space for the SPL magic (previously
inserted by mkimage)
- no longer updates rk3188-board-tpl.c (as we have just removed it
in an earlier commit)
Changes in v3:
- tracked the root-cause why no interwork branch was emitted and fixed
it using a '.type'-directive in start.S to mark save_boot_params_ret
as a (possible) function-entry.
- converted setjmp/longjmp from inline-assembly to separate .S files
to improve predicatability if emitted code
Changes in v2:
- [added in v2] chain back_to_bootrom calls for SPL, first returning
to the TPL (using the same mechanism) and the to the BROM from the
TPL
- also covers the RK3188 (which I had originally missed)
Kever Yang (3):
rockchip: boot0: align to 0x20 for armv7 '_start'
rockchip: mkimage: use spl_boot0 for all Rockchip SoCs
rockchip: rk3288: use aligned address for SPL_TEXT_BASE
Philipp Tomsich (15):
arm: boot0 hook: move boot0 hook before '_start'
rockchip: enable boot0-hook for all Rockchip SoCs
rockchip: rk3036: use aligned address for SPL_TEXT_BASE
socfpga: boot0 hook: adjust to unified boot0 semantics
bcm235xx: boot0 hook: adjust to unified boot0 semantics
bcm281xx: boot0 hook: adjust to unified boot0 semantics
rockchip: boot0 hook: support early return for RK3188/RK3066-style
BROM
arm: make save_boot_params_ret prototype visible for AArch64
arm: mark save_boot_params_ret as a function
arm: provide a PCS-compliant setjmp implementation
rockchip: back-to-bootrom: replace assembly-implementation with C-code
rockchip: rk3188: use boot0 hook to load up SPL in 2 steps
rockchip: back-to-bootrom: allow passing a cmd to the bootrom
rockchip: rk3188: move CONFIG_SPL_* entries from rk3188_common.h to
Kconfig
rockchip: mkimage: remove unused code-paths (spl_boot0 is now implied)
arch/arm/Kconfig | 1 +
arch/arm/cpu/armv7/start.S | 1 +
arch/arm/include/asm/arch-bcm235xx/boot0.h | 2 +
arch/arm/include/asm/arch-bcm281xx/boot0.h | 2 +
arch/arm/include/asm/arch-rockchip/boot0.h | 38 ++++++++---
arch/arm/include/asm/arch-rockchip/bootrom.h | 30 ++++++---
arch/arm/include/asm/setjmp.h | 94 ++++------------------------
arch/arm/include/asm/system.h | 62 +++++++++---------
arch/arm/lib/Makefile | 6 ++
arch/arm/lib/setjmp.S | 37 +++++++++++
arch/arm/lib/setjmp_aarch64.S | 42 +++++++++++++
arch/arm/lib/vectors.S | 54 ++++++++++------
arch/arm/mach-rockchip/Kconfig | 39 ++++++++++--
arch/arm/mach-rockchip/Makefile | 5 +-
arch/arm/mach-rockchip/bootrom.c | 54 +++++++++++++++-
arch/arm/mach-rockchip/rk3036-board-spl.c | 2 +-
arch/arm/mach-rockchip/rk3188-board-spl.c | 14 +----
arch/arm/mach-rockchip/rk3188-board-tpl.c | 86 -------------------------
arch/arm/mach-rockchip/rk322x-board-spl.c | 2 +-
arch/arm/mach-rockchip/rk3288-board-spl.c | 4 +-
arch/arm/mach-rockchip/rk3288-board-tpl.c | 2 +-
arch/arm/mach-rockchip/rk3368-board-tpl.c | 2 +-
arch/arm/mach-rockchip/rk3399-board-spl.c | 2 +-
arch/arm/mach-rockchip/save_boot_param.S | 69 --------------------
arch/arm/mach-socfpga/include/mach/boot0.h | 3 +
doc/README.rockchip | 10 +--
include/configs/rk3036_common.h | 2 +-
include/configs/rk3188_common.h | 18 +-----
include/configs/rk3288_common.h | 2 +-
tools/rkcommon.c | 45 +++++--------
30 files changed, 348 insertions(+), 382 deletions(-)
create mode 100644 arch/arm/lib/setjmp.S
create mode 100644 arch/arm/lib/setjmp_aarch64.S
delete mode 100644 arch/arm/mach-rockchip/rk3188-board-tpl.c
delete mode 100644 arch/arm/mach-rockchip/save_boot_param.S
--
2.1.4
^ permalink raw reply [flat|nested] 46+ messages in thread
* [U-Boot] [PATCH v5 01/18] arm: boot0 hook: move boot0 hook before '_start'
2017-10-10 14:21 [U-Boot] [PATCH v5 00/18] rockchip: back-to-bootrom: replace assembly-implementation with C-code Philipp Tomsich
@ 2017-10-10 14:21 ` Philipp Tomsich
2017-11-07 14:18 ` [U-Boot] [U-Boot, v5, " Philipp Tomsich
2017-10-10 14:21 ` [U-Boot] [PATCH v5 02/18] rockchip: boot0: align to 0x20 for armv7 '_start' Philipp Tomsich
` (17 subsequent siblings)
18 siblings, 1 reply; 46+ messages in thread
From: Philipp Tomsich @ 2017-10-10 14:21 UTC (permalink / raw)
To: u-boot
The boot0 hook on ARM does not insert its payload before the vector
table. This is both a mismatch with thec comment above it and
contradict usage of the boot0 hook on ARM64.
To fix this (and unify the semantics for ARM and ARM64), we change the
boot0-hook semantics on ARM to match those on ARM64:
(1) if a boot0-hook is present it is inserted at the start of
the image
(2) if a boot0-hook is present, emitting the ARM vector table
(and the _start) symbol are suppressed in vectors.S and
the boot0-hook has full control over where and when it
wants to emit these
Signed-off-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com>
---
Changes in v5: None
Changes in v4: None
Changes in v3: None
Changes in v2: None
arch/arm/include/asm/arch-rockchip/boot0.h | 1 -
arch/arm/lib/vectors.S | 54 +++++++++++++++++++-----------
2 files changed, 35 insertions(+), 20 deletions(-)
diff --git a/arch/arm/include/asm/arch-rockchip/boot0.h b/arch/arm/include/asm/arch-rockchip/boot0.h
index 72d264b..455d842 100644
--- a/arch/arm/include/asm/arch-rockchip/boot0.h
+++ b/arch/arm/include/asm/arch-rockchip/boot0.h
@@ -1,4 +1,3 @@
-
/*
* Copyright 2017 Theobroma Systems Design und Consulting GmbH
*
diff --git a/arch/arm/lib/vectors.S b/arch/arm/lib/vectors.S
index 1019091..9cb0d2e 100644
--- a/arch/arm/lib/vectors.S
+++ b/arch/arm/lib/vectors.S
@@ -16,6 +16,22 @@
#include <config.h>
/*
+ * A macro to allow insertion of an ARM exception vector either
+ * for the non-boot0 case or by a boot0-header.
+ */
+ .macro ARM_VECTORS
+ b reset
+ ldr pc, _undefined_instruction
+ ldr pc, _software_interrupt
+ ldr pc, _prefetch_abort
+ ldr pc, _data_abort
+ ldr pc, _not_used
+ ldr pc, _irq
+ ldr pc, _fiq
+ .endm
+
+
+/*
*************************************************************************
*
* Symbol _start is referenced elsewhere, so make it global
@@ -35,6 +51,23 @@
.section ".vectors", "ax"
+#if defined(CONFIG_ENABLE_ARM_SOC_BOOT0_HOOK)
+/*
+ * Various SoCs need something special and SoC-specific up front in
+ * order to boot, allow them to set that in their boot0.h file and then
+ * use it here.
+ *
+ * To allow a boot0 hook to insert a 'special' sequence after the vector
+ * table (e.g. for the socfpga), the presence of a boot0 hook supresses
+ * the below vector table and assumes that the vector table is filled in
+ * by the boot0 hook. The requirements for a boot0 hook thus are:
+ * (1) defines '_start:' as appropriate
+ * (2) inserts the vector table using ARM_VECTORS as appropriate
+ */
+#include <asm/arch/boot0.h>
+
+#else
+
/*
*************************************************************************
*
@@ -46,28 +79,11 @@
*/
_start:
-
#ifdef CONFIG_SYS_DV_NOR_BOOT_CFG
.word CONFIG_SYS_DV_NOR_BOOT_CFG
#endif
-
- b reset
- ldr pc, _undefined_instruction
- ldr pc, _software_interrupt
- ldr pc, _prefetch_abort
- ldr pc, _data_abort
- ldr pc, _not_used
- ldr pc, _irq
- ldr pc, _fiq
-
-#ifdef CONFIG_ENABLE_ARM_SOC_BOOT0_HOOK
-/*
- * Various SoCs need something special and SoC-specific up front in
- * order to boot, allow them to set that in their boot0.h file and then
- * use it here.
- */
-#include <asm/arch/boot0.h>
-#endif
+ ARM_VECTORS
+#endif /* !defined(CONFIG_ENABLE_ARM_SOC_BOOT0_HOOK) */
/*
*************************************************************************
--
2.1.4
^ permalink raw reply related [flat|nested] 46+ messages in thread
* [U-Boot] [PATCH v5 02/18] rockchip: boot0: align to 0x20 for armv7 '_start'
2017-10-10 14:21 [U-Boot] [PATCH v5 00/18] rockchip: back-to-bootrom: replace assembly-implementation with C-code Philipp Tomsich
2017-10-10 14:21 ` [U-Boot] [PATCH v5 01/18] arm: boot0 hook: move boot0 hook before '_start' Philipp Tomsich
@ 2017-10-10 14:21 ` Philipp Tomsich
2017-11-07 14:18 ` [U-Boot] [U-Boot, v5, " Philipp Tomsich
2017-11-09 12:59 ` [U-Boot] [PATCH v5 " Andy Yan
2017-10-10 14:21 ` [U-Boot] [PATCH v5 03/18] rockchip: enable boot0-hook for all Rockchip SoCs Philipp Tomsich
` (16 subsequent siblings)
18 siblings, 2 replies; 46+ messages in thread
From: Philipp Tomsich @ 2017-10-10 14:21 UTC (permalink / raw)
To: u-boot
From: Kever Yang <kever.yang@rock-chips.com>
The '_start' is using as vector table base address, and will write
to VBAR register, so it needs to be aligned to 0x20 for armv7.
Signed-off-by: Kever Yang <kever.yang@rock-chips.com>
[Updated to current code base:]
Signed-off-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com>
---
Changes in v5: None
Changes in v4: None
Changes in v3: None
Changes in v2: None
arch/arm/include/asm/arch-rockchip/boot0.h | 14 ++++++++++++--
1 file changed, 12 insertions(+), 2 deletions(-)
diff --git a/arch/arm/include/asm/arch-rockchip/boot0.h b/arch/arm/include/asm/arch-rockchip/boot0.h
index 455d842..f7c6146 100644
--- a/arch/arm/include/asm/arch-rockchip/boot0.h
+++ b/arch/arm/include/asm/arch-rockchip/boot0.h
@@ -6,12 +6,13 @@
/*
* Execution starts on the instruction following this 4-byte header
- * (containing the magic 'RK33').
+ * (containing the magic 'RK30', 'RK31', 'RK32' or 'RK33'). This
+ * magic constant will be written into the final image by the rkimage
+ * tool, but we need to reserve space for it here.
*
* To make life easier for everyone, we build the SPL binary with
* space for this 4-byte header already included in the binary.
*/
-
#ifdef CONFIG_SPL_BUILD
/*
* We need to add 4 bytes of space for the 'RK33' at the
@@ -26,6 +27,15 @@
b reset /* may be overwritten --- should be 'nop' or a 'b reset' */
#endif
b reset
+#if !defined(CONFIG_ARM64)
+ /*
+ * For armv7, the addr '_start' will used as vector start address
+ * and write to VBAR register, which needs to aligned to 0x20.
+ */
+ .align(5)
+_start:
+ ARM_VECTORS
+#endif
#if defined(CONFIG_ROCKCHIP_RK3399) && defined(CONFIG_SPL_BUILD)
.space CONFIG_ROCKCHIP_SPL_RESERVE_IRAM /* space for the ATF data */
--
2.1.4
^ permalink raw reply related [flat|nested] 46+ messages in thread
* [U-Boot] [PATCH v5 03/18] rockchip: enable boot0-hook for all Rockchip SoCs
2017-10-10 14:21 [U-Boot] [PATCH v5 00/18] rockchip: back-to-bootrom: replace assembly-implementation with C-code Philipp Tomsich
2017-10-10 14:21 ` [U-Boot] [PATCH v5 01/18] arm: boot0 hook: move boot0 hook before '_start' Philipp Tomsich
2017-10-10 14:21 ` [U-Boot] [PATCH v5 02/18] rockchip: boot0: align to 0x20 for armv7 '_start' Philipp Tomsich
@ 2017-10-10 14:21 ` Philipp Tomsich
2017-11-07 14:18 ` [U-Boot] [U-Boot, v5, " Philipp Tomsich
2017-10-10 14:21 ` [U-Boot] [PATCH v5 04/18] rockchip: mkimage: use spl_boot0 " Philipp Tomsich
` (15 subsequent siblings)
18 siblings, 1 reply; 46+ messages in thread
From: Philipp Tomsich @ 2017-10-10 14:21 UTC (permalink / raw)
To: u-boot
Rockchip SoCs bootrom design is like this:
- First 2KB or 4KB internal memory is for bootrom stack and heap;
- Then the first 4-byte suppose to be a TAG like 'RK33';
- The the following memory address end with '0004' is the first
instruction load and running by bootrom;
Let's use the boot0 hook to reserve the first 4-byte tag for all
the Rockchip SoCs.
Signed-off-by: Philipp Tomsich <philipp.tomsich@theobroma-systems>
[Commit message:]
Signed-off-by: Kever Yang <kever.yang@rock-chips.com>
Signed-off-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com>
---
Changes in v5: None
Changes in v4: None
Changes in v3: None
Changes in v2: None
arch/arm/Kconfig | 1 +
arch/arm/mach-rockchip/Kconfig | 2 --
2 files changed, 1 insertion(+), 2 deletions(-)
diff --git a/arch/arm/Kconfig b/arch/arm/Kconfig
index 64e0ee4..585e5b0 100644
--- a/arch/arm/Kconfig
+++ b/arch/arm/Kconfig
@@ -1117,6 +1117,7 @@ config ARCH_ROCKCHIP
select DM_USB if USB
select DM_PWM
select DM_REGULATOR
+ select ENABLE_ARM_SOC_BOOT0_HOOK
imply CMD_FASTBOOT
imply FASTBOOT
imply FAT_WRITE
diff --git a/arch/arm/mach-rockchip/Kconfig b/arch/arm/mach-rockchip/Kconfig
index d9b25d5..31e9864 100644
--- a/arch/arm/mach-rockchip/Kconfig
+++ b/arch/arm/mach-rockchip/Kconfig
@@ -74,7 +74,6 @@ config ROCKCHIP_RK3368
imply SPL_SEPARATE_BSS
imply SPL_SERIAL_SUPPORT
imply TPL_SERIAL_SUPPORT
- select ENABLE_ARM_SOC_BOOT0_HOOK
select DEBUG_UART_BOARD_INIT
select SYS_NS16550
help
@@ -112,7 +111,6 @@ config ROCKCHIP_RK3399
select SPL_SEPARATE_BSS
select SPL_SERIAL_SUPPORT
select SPL_DRIVERS_MISC_SUPPORT
- select ENABLE_ARM_SOC_BOOT0_HOOK
select DEBUG_UART_BOARD_INIT
help
The Rockchip RK3399 is a ARM-based SoC with a dual-core Cortex-A72
--
2.1.4
^ permalink raw reply related [flat|nested] 46+ messages in thread
* [U-Boot] [PATCH v5 04/18] rockchip: mkimage: use spl_boot0 for all Rockchip SoCs
2017-10-10 14:21 [U-Boot] [PATCH v5 00/18] rockchip: back-to-bootrom: replace assembly-implementation with C-code Philipp Tomsich
` (2 preceding siblings ...)
2017-10-10 14:21 ` [U-Boot] [PATCH v5 03/18] rockchip: enable boot0-hook for all Rockchip SoCs Philipp Tomsich
@ 2017-10-10 14:21 ` Philipp Tomsich
2017-11-07 14:18 ` [U-Boot] [U-Boot, v5, " Philipp Tomsich
2017-10-10 14:21 ` [U-Boot] [PATCH v5 05/18] rockchip: rk3288: use aligned address for SPL_TEXT_BASE Philipp Tomsich
` (14 subsequent siblings)
18 siblings, 1 reply; 46+ messages in thread
From: Philipp Tomsich @ 2017-10-10 14:21 UTC (permalink / raw)
To: u-boot
From: Kever Yang <kever.yang@rock-chips.com>
Enable the spl_boot0 in SPL and use the pre-padding TAG memory,
the mkimage do not need to pad it but only need to replace the value
with correct TAG value.
Signed-off-by: Kever Yang <kever.yang@rock-chips.com>
[Updated:]
Signed-off-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com>
---
Changes in v5: None
Changes in v4: None
Changes in v3: None
Changes in v2: None
tools/rkcommon.c | 14 +++++++-------
1 file changed, 7 insertions(+), 7 deletions(-)
diff --git a/tools/rkcommon.c b/tools/rkcommon.c
index 1a24e16..c83260a 100644
--- a/tools/rkcommon.c
+++ b/tools/rkcommon.c
@@ -72,15 +72,15 @@ struct spl_info {
};
static struct spl_info spl_infos[] = {
- { "rk3036", "RK30", 0x1000, false, false },
- { "rk3128", "RK31", 0x1800, false, false },
- { "rk3188", "RK31", 0x8000 - 0x800, true, false },
- { "rk322x", "RK32", 0x8000 - 0x1000, false, false },
- { "rk3288", "RK32", 0x8000, false, false },
- { "rk3328", "RK32", 0x8000 - 0x1000, false, false },
+ { "rk3036", "RK30", 0x1000, false, true },
+ { "rk3128", "RK31", 0x1800, false, true },
+ { "rk3188", "RK31", 0x8000 - 0x800, true, true },
+ { "rk322x", "RK32", 0x8000 - 0x1000, false, true },
+ { "rk3288", "RK32", 0x8000, false, true },
+ { "rk3328", "RK32", 0x8000 - 0x1000, false, true },
{ "rk3368", "RK33", 0x8000 - 0x1000, false, true },
{ "rk3399", "RK33", 0x30000 - 0x2000, false, true },
- { "rv1108", "RK11", 0x1800, false, false},
+ { "rv1108", "RK11", 0x1800, false, true },
};
static unsigned char rc4_key[16] = {
--
2.1.4
^ permalink raw reply related [flat|nested] 46+ messages in thread
* [U-Boot] [PATCH v5 05/18] rockchip: rk3288: use aligned address for SPL_TEXT_BASE
2017-10-10 14:21 [U-Boot] [PATCH v5 00/18] rockchip: back-to-bootrom: replace assembly-implementation with C-code Philipp Tomsich
` (3 preceding siblings ...)
2017-10-10 14:21 ` [U-Boot] [PATCH v5 04/18] rockchip: mkimage: use spl_boot0 " Philipp Tomsich
@ 2017-10-10 14:21 ` Philipp Tomsich
2017-11-07 14:18 ` [U-Boot] [U-Boot, v5, " Philipp Tomsich
2017-10-10 14:21 ` [U-Boot] [PATCH v5 06/18] rockchip: rk3036: " Philipp Tomsich
` (13 subsequent siblings)
18 siblings, 1 reply; 46+ messages in thread
From: Philipp Tomsich @ 2017-10-10 14:21 UTC (permalink / raw)
To: u-boot
From: Kever Yang <kever.yang@rock-chips.com>
After we use boot0 hook, we can use offset '000' instead of '004' as
SPL_TEXT_BASE.
Signed-off-by: Kever Yang <kever.yang@rock-chips.com>
[Updated tag in commit summary:]
Signed-off-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com>
---
Changes in v5: None
Changes in v4: None
Changes in v3: None
Changes in v2: None
include/configs/rk3288_common.h | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/include/configs/rk3288_common.h b/include/configs/rk3288_common.h
index 34f2558..9201cb6 100644
--- a/include/configs/rk3288_common.h
+++ b/include/configs/rk3288_common.h
@@ -35,7 +35,7 @@
#if defined(CONFIG_SPL_BUILD) && defined(CONFIG_TPL_BOOTROM_SUPPORT)
# define CONFIG_SPL_TEXT_BASE 0x0
#else
-# define CONFIG_SPL_TEXT_BASE 0xff704004
+# define CONFIG_SPL_TEXT_BASE 0xff704000
#endif
/* MMC/SD IP block */
--
2.1.4
^ permalink raw reply related [flat|nested] 46+ messages in thread
* [U-Boot] [PATCH v5 06/18] rockchip: rk3036: use aligned address for SPL_TEXT_BASE
2017-10-10 14:21 [U-Boot] [PATCH v5 00/18] rockchip: back-to-bootrom: replace assembly-implementation with C-code Philipp Tomsich
` (4 preceding siblings ...)
2017-10-10 14:21 ` [U-Boot] [PATCH v5 05/18] rockchip: rk3288: use aligned address for SPL_TEXT_BASE Philipp Tomsich
@ 2017-10-10 14:21 ` Philipp Tomsich
2017-11-07 14:18 ` [U-Boot] [U-Boot, v5, " Philipp Tomsich
2017-10-10 14:21 ` [U-Boot] [PATCH v5 07/18] socfpga: boot0 hook: adjust to unified boot0 semantics Philipp Tomsich
` (12 subsequent siblings)
18 siblings, 1 reply; 46+ messages in thread
From: Philipp Tomsich @ 2017-10-10 14:21 UTC (permalink / raw)
To: u-boot
With the boot0-hook inserting the additional padding to receive our
SPL magic, the SPL_TEXT_BASE can be aligned again.
Signed-off-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com>
---
Changes in v5: None
Changes in v4: None
Changes in v3: None
Changes in v2: None
include/configs/rk3036_common.h | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/include/configs/rk3036_common.h b/include/configs/rk3036_common.h
index 9ac0df5..a387099 100644
--- a/include/configs/rk3036_common.h
+++ b/include/configs/rk3036_common.h
@@ -26,7 +26,7 @@
#define CONFIG_SYS_INIT_SP_ADDR 0x60100000
#define CONFIG_SYS_LOAD_ADDR 0x60800800
#define CONFIG_SPL_STACK 0x10081fff
-#define CONFIG_SPL_TEXT_BASE 0x10081004
+#define CONFIG_SPL_TEXT_BASE 0x10081000
#define CONFIG_ROCKCHIP_MAX_INIT_SIZE (4 << 10)
#define CONFIG_ROCKCHIP_CHIP_TAG "RK30"
--
2.1.4
^ permalink raw reply related [flat|nested] 46+ messages in thread
* [U-Boot] [PATCH v5 07/18] socfpga: boot0 hook: adjust to unified boot0 semantics
2017-10-10 14:21 [U-Boot] [PATCH v5 00/18] rockchip: back-to-bootrom: replace assembly-implementation with C-code Philipp Tomsich
` (5 preceding siblings ...)
2017-10-10 14:21 ` [U-Boot] [PATCH v5 06/18] rockchip: rk3036: " Philipp Tomsich
@ 2017-10-10 14:21 ` Philipp Tomsich
2017-11-06 17:26 ` Dr. Philipp Tomsich
2017-11-07 14:18 ` [U-Boot] [U-Boot, v5, " Philipp Tomsich
2017-10-10 14:21 ` [U-Boot] [PATCH v5 08/18] bcm235xx: " Philipp Tomsich
` (11 subsequent siblings)
18 siblings, 2 replies; 46+ messages in thread
From: Philipp Tomsich @ 2017-10-10 14:21 UTC (permalink / raw)
To: u-boot
With the updated boot0 semantics (i.e. giving the boot0-hook control
over when and where the vector table is emitted), the boot0-hook for
the socfpga needs to be adjusted.
Signed-off-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com>
---
Changes in v5: None
Changes in v4: None
Changes in v3: None
Changes in v2: None
arch/arm/mach-socfpga/include/mach/boot0.h | 3 +++
1 file changed, 3 insertions(+)
diff --git a/arch/arm/mach-socfpga/include/mach/boot0.h b/arch/arm/mach-socfpga/include/mach/boot0.h
index 22d9e7f..d6b9435 100644
--- a/arch/arm/mach-socfpga/include/mach/boot0.h
+++ b/arch/arm/mach-socfpga/include/mach/boot0.h
@@ -7,6 +7,9 @@
#ifndef __BOOT0_H
#define __BOOT0_H
+_start:
+ ARM_VECTORS
+
#ifdef CONFIG_SPL_BUILD
.balignl 64,0xf33db33f;
--
2.1.4
^ permalink raw reply related [flat|nested] 46+ messages in thread
* [U-Boot] [PATCH v5 08/18] bcm235xx: boot0 hook: adjust to unified boot0 semantics
2017-10-10 14:21 [U-Boot] [PATCH v5 00/18] rockchip: back-to-bootrom: replace assembly-implementation with C-code Philipp Tomsich
` (6 preceding siblings ...)
2017-10-10 14:21 ` [U-Boot] [PATCH v5 07/18] socfpga: boot0 hook: adjust to unified boot0 semantics Philipp Tomsich
@ 2017-10-10 14:21 ` Philipp Tomsich
2017-11-07 14:18 ` [U-Boot] [U-Boot, v5, " Philipp Tomsich
2017-10-10 14:21 ` [U-Boot] [PATCH v5 09/18] bcm281xx: " Philipp Tomsich
` (10 subsequent siblings)
18 siblings, 1 reply; 46+ messages in thread
From: Philipp Tomsich @ 2017-10-10 14:21 UTC (permalink / raw)
To: u-boot
This updates the BCM235xx boot0-hook to the updated boot0 semantics
by emitting _start and the vector table before the boot0 hook (as
was the case before).
Signed-off-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com>
Version-changes: 5
- ran 'whitespace-cleanup'
---
Changes in v5: None
Changes in v4: None
Changes in v3: None
Changes in v2: None
arch/arm/include/asm/arch-bcm235xx/boot0.h | 2 ++
1 file changed, 2 insertions(+)
diff --git a/arch/arm/include/asm/arch-bcm235xx/boot0.h b/arch/arm/include/asm/arch-bcm235xx/boot0.h
index a747bd3..28f05ba 100644
--- a/arch/arm/include/asm/arch-bcm235xx/boot0.h
+++ b/arch/arm/include/asm/arch-bcm235xx/boot0.h
@@ -5,5 +5,7 @@
*/
/* BOOT0 header information */
+_start:
+ ARM_VECTORS
.word 0xbabeface
.word _end - _start
--
2.1.4
^ permalink raw reply related [flat|nested] 46+ messages in thread
* [U-Boot] [PATCH v5 09/18] bcm281xx: boot0 hook: adjust to unified boot0 semantics
2017-10-10 14:21 [U-Boot] [PATCH v5 00/18] rockchip: back-to-bootrom: replace assembly-implementation with C-code Philipp Tomsich
` (7 preceding siblings ...)
2017-10-10 14:21 ` [U-Boot] [PATCH v5 08/18] bcm235xx: " Philipp Tomsich
@ 2017-10-10 14:21 ` Philipp Tomsich
2017-11-07 14:18 ` [U-Boot] [U-Boot, v5, " Philipp Tomsich
2017-10-10 14:21 ` [U-Boot] [PATCH v5 10/18] rockchip: boot0 hook: support early return for RK3188/RK3066-style BROM Philipp Tomsich
` (9 subsequent siblings)
18 siblings, 1 reply; 46+ messages in thread
From: Philipp Tomsich @ 2017-10-10 14:21 UTC (permalink / raw)
To: u-boot
This updates the BCM281xx boot0-hook to the updated boot0 semantics
by emitting _start and the vector table before the boot0 hook (as
was the case before).
Signed-off-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com>
---
Changes in v5: None
Changes in v4: None
Changes in v3: None
Changes in v2: None
arch/arm/include/asm/arch-bcm281xx/boot0.h | 2 ++
1 file changed, 2 insertions(+)
diff --git a/arch/arm/include/asm/arch-bcm281xx/boot0.h b/arch/arm/include/asm/arch-bcm281xx/boot0.h
index a747bd3..28f05ba 100644
--- a/arch/arm/include/asm/arch-bcm281xx/boot0.h
+++ b/arch/arm/include/asm/arch-bcm281xx/boot0.h
@@ -5,5 +5,7 @@
*/
/* BOOT0 header information */
+_start:
+ ARM_VECTORS
.word 0xbabeface
.word _end - _start
--
2.1.4
^ permalink raw reply related [flat|nested] 46+ messages in thread
* [U-Boot] [PATCH v5 10/18] rockchip: boot0 hook: support early return for RK3188/RK3066-style BROM
2017-10-10 14:21 [U-Boot] [PATCH v5 00/18] rockchip: back-to-bootrom: replace assembly-implementation with C-code Philipp Tomsich
` (8 preceding siblings ...)
2017-10-10 14:21 ` [U-Boot] [PATCH v5 09/18] bcm281xx: " Philipp Tomsich
@ 2017-10-10 14:21 ` Philipp Tomsich
2017-11-07 14:18 ` [U-Boot] [U-Boot, v5, " Philipp Tomsich
2017-10-10 14:21 ` [U-Boot] [PATCH v5 11/18] arm: make save_boot_params_ret prototype visible for AArch64 Philipp Tomsich
` (8 subsequent siblings)
18 siblings, 1 reply; 46+ messages in thread
From: Philipp Tomsich @ 2017-10-10 14:21 UTC (permalink / raw)
To: u-boot
Some Rockchip BROM versions (e.g. the RK3188 and RK3066) first read 1KB data
from NAND into SRAM and executes it. Then, following a return to bootrom, the
BROM loads additional code to SRAM (not overwriting the first block read) and
reenters at the same address as the first time.
To support booting either a TPL (on the RK3066) or SPL (on the RK3188) using
this model of having to count entries, this commit adds code to the boot0
hook to track the number of entries and handle them accordingly.
Signed-off-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com>
Signed-off-by: Paweł Jarosz <paweljarosz3691@gmail.com>
Tested-by: Andy Yan <andy.yan@rock-chips.com>
---
Changes in v5:
- set return value to 0 before returning to the BROM
Changes in v4: None
Changes in v3: None
Changes in v2: None
arch/arm/include/asm/arch-rockchip/boot0.h | 25 +++++++++++++++++++------
arch/arm/mach-rockchip/Kconfig | 28 ++++++++++++++++++++++++++++
2 files changed, 47 insertions(+), 6 deletions(-)
diff --git a/arch/arm/include/asm/arch-rockchip/boot0.h b/arch/arm/include/asm/arch-rockchip/boot0.h
index f7c6146..af3a733 100644
--- a/arch/arm/include/asm/arch-rockchip/boot0.h
+++ b/arch/arm/include/asm/arch-rockchip/boot0.h
@@ -19,12 +19,25 @@
* beginning of the executable. However, as we want to keep
* this generic and make it applicable to builds that are like
* the RK3368 (TPL needs this, SPL doesn't) or the RK3399 (no
- * TPL, but extra space needed in the SPL), we simply repeat
- * the 'b reset' with the expectation that the first one will
- * be overwritten, if this is the first stage contained in the
- * final image created with mkimage)...
+ * TPL, but extra space needed in the SPL), we simply insert
+ * a branch-to-next-instruction-word with the expectation that
+ * the first one may be overwritten, if this is the first stage
+ * contained in the final image created with mkimage)...
*/
- b reset /* may be overwritten --- should be 'nop' or a 'b reset' */
+ b 1f /* if overwritten, entry-address is at the next word */
+1:
+#endif
+#if CONFIG_IS_ENABLED(ROCKCHIP_EARLYRETURN_TO_BROM)
+ adr r3, entry_counter
+ ldr r0, [r3]
+ cmp r0, #1 /* check if entry_counter == 1 */
+ beq reset /* regular bootup */
+ add r0, #1
+ str r0, [r3] /* increment the entry_counter in memory */
+ mov r0, #0 /* return 0 to the BROM to signal 'OK' */
+ bx lr /* return control to the BROM */
+entry_counter:
+ .word 0
#endif
b reset
#if !defined(CONFIG_ARM64)
@@ -32,7 +45,7 @@
* For armv7, the addr '_start' will used as vector start address
* and write to VBAR register, which needs to aligned to 0x20.
*/
- .align(5)
+ .align(5), 0x0
_start:
ARM_VECTORS
#endif
diff --git a/arch/arm/mach-rockchip/Kconfig b/arch/arm/mach-rockchip/Kconfig
index 31e9864..d59a1d5 100644
--- a/arch/arm/mach-rockchip/Kconfig
+++ b/arch/arm/mach-rockchip/Kconfig
@@ -158,6 +158,34 @@ config ROCKCHIP_SPL_RESERVE_IRAM
config ROCKCHIP_BROM_HELPER
bool
+config SPL_ROCKCHIP_EARLYRETURN_TO_BROM
+ bool "SPL requires early-return (for RK3188-style BROM) to BROM"
+ depends on SPL && ENABLE_ARM_SOC_BOOT0_HOOK
+ help
+ Some Rockchip BROM variants (e.g. on the RK3188) load the
+ first stage in segments and enter multiple times. E.g. on
+ the RK3188, the first 1KB of the first stage are loaded
+ first and entered; after returning to the BROM, the
+ remainder of the first stage is loaded, but the BROM
+ re-enters at the same address/to the same code as previously.
+
+ This enables support code in the BOOT0 hook for the SPL stage
+ to allow multiple entries.
+
+config TPL_ROCKCHIP_EARLYRETURN_TO_BROM
+ bool "TPL requires early-return (for RK3188-style BROM) to BROM"
+ depends on TPL && ENABLE_ARM_SOC_BOOT0_HOOK
+ help
+ Some Rockchip BROM variants (e.g. on the RK3188) load the
+ first stage in segments and enter multiple times. E.g. on
+ the RK3188, the first 1KB of the first stage are loaded
+ first and entered; after returning to the BROM, the
+ remainder of the first stage is loaded, but the BROM
+ re-enters at the same address/to the same code as previously.
+
+ This enables support code in the BOOT0 hook for the TPL stage
+ to allow multiple entries.
+
config SPL_MMC_SUPPORT
default y if !SPL_ROCKCHIP_BACK_TO_BROM
--
2.1.4
^ permalink raw reply related [flat|nested] 46+ messages in thread
* [U-Boot] [PATCH v5 11/18] arm: make save_boot_params_ret prototype visible for AArch64
2017-10-10 14:21 [U-Boot] [PATCH v5 00/18] rockchip: back-to-bootrom: replace assembly-implementation with C-code Philipp Tomsich
` (9 preceding siblings ...)
2017-10-10 14:21 ` [U-Boot] [PATCH v5 10/18] rockchip: boot0 hook: support early return for RK3188/RK3066-style BROM Philipp Tomsich
@ 2017-10-10 14:21 ` Philipp Tomsich
2017-11-07 14:18 ` [U-Boot] [U-Boot, v5, " Philipp Tomsich
2017-10-10 14:21 ` [U-Boot] [PATCH v5 12/18] arm: mark save_boot_params_ret as a function Philipp Tomsich
` (7 subsequent siblings)
18 siblings, 1 reply; 46+ messages in thread
From: Philipp Tomsich @ 2017-10-10 14:21 UTC (permalink / raw)
To: u-boot
The save_boot_params_ret() prototype (for those of us, that have a
valid SP on entry and can implement save_boot_params() in C), was
previously only defined for !defined(CONFIG_ARM64).
This moves the declaration to a common block to ensure the prototype
is available to everyone that might need it.
Signed-off-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com>
Tested-by: Andy Yan <andy.yan@rock-chips.com>
---
Changes in v5: None
Changes in v4: None
Changes in v3: None
Changes in v2: None
arch/arm/include/asm/system.h | 62 +++++++++++++++++++++----------------------
1 file changed, 31 insertions(+), 31 deletions(-)
diff --git a/arch/arm/include/asm/system.h b/arch/arm/include/asm/system.h
index 79bd19a..e2af296 100644
--- a/arch/arm/include/asm/system.h
+++ b/arch/arm/include/asm/system.h
@@ -332,37 +332,6 @@ void psci_arch_init(void);
#ifndef __ASSEMBLY__
-/**
- * save_boot_params() - Save boot parameters before starting reset sequence
- *
- * If you provide this function it will be called immediately U-Boot starts,
- * both for SPL and U-Boot proper.
- *
- * All registers are unchanged from U-Boot entry. No registers need be
- * preserved.
- *
- * This is not a normal C function. There is no stack. Return by branching to
- * save_boot_params_ret.
- *
- * void save_boot_params(u32 r0, u32 r1, u32 r2, u32 r3);
- */
-
-/**
- * save_boot_params_ret() - Return from save_boot_params()
- *
- * If you provide save_boot_params(), then you should jump back to this
- * function when done. Try to preserve all registers.
- *
- * If your implementation of save_boot_params() is in C then it is acceptable
- * to simply call save_boot_params_ret() at the end of your function. Since
- * there is no link register set up, you cannot just exit the function. U-Boot
- * will return to the (initialised) value of lr, and likely crash/hang.
- *
- * If your implementation of save_boot_params() is in assembler then you
- * should use 'b' or 'bx' to return to save_boot_params_ret.
- */
-void save_boot_params_ret(void);
-
#ifdef CONFIG_ARMV7_LPAE
void switch_to_hypervisor_ret(void);
#endif
@@ -556,6 +525,37 @@ void mmu_page_table_flush(unsigned long start, unsigned long stop);
#ifndef __ASSEMBLY__
/**
+ * save_boot_params() - Save boot parameters before starting reset sequence
+ *
+ * If you provide this function it will be called immediately U-Boot starts,
+ * both for SPL and U-Boot proper.
+ *
+ * All registers are unchanged from U-Boot entry. No registers need be
+ * preserved.
+ *
+ * This is not a normal C function. There is no stack. Return by branching to
+ * save_boot_params_ret.
+ *
+ * void save_boot_params(u32 r0, u32 r1, u32 r2, u32 r3);
+ */
+
+/**
+ * save_boot_params_ret() - Return from save_boot_params()
+ *
+ * If you provide save_boot_params(), then you should jump back to this
+ * function when done. Try to preserve all registers.
+ *
+ * If your implementation of save_boot_params() is in C then it is acceptable
+ * to simply call save_boot_params_ret() at the end of your function. Since
+ * there is no link register set up, you cannot just exit the function. U-Boot
+ * will return to the (initialised) value of lr, and likely crash/hang.
+ *
+ * If your implementation of save_boot_params() is in assembler then you
+ * should use 'b' or 'bx' to return to save_boot_params_ret.
+ */
+void save_boot_params_ret(void);
+
+/**
* Change the cache settings for a region.
*
* \param start start address of memory region to change
--
2.1.4
^ permalink raw reply related [flat|nested] 46+ messages in thread
* [U-Boot] [PATCH v5 12/18] arm: mark save_boot_params_ret as a function
2017-10-10 14:21 [U-Boot] [PATCH v5 00/18] rockchip: back-to-bootrom: replace assembly-implementation with C-code Philipp Tomsich
` (10 preceding siblings ...)
2017-10-10 14:21 ` [U-Boot] [PATCH v5 11/18] arm: make save_boot_params_ret prototype visible for AArch64 Philipp Tomsich
@ 2017-10-10 14:21 ` Philipp Tomsich
2017-11-07 14:18 ` [U-Boot] [U-Boot, v5, " Philipp Tomsich
2017-10-10 14:21 ` [U-Boot] [PATCH v5 13/18] arm: provide a PCS-compliant setjmp implementation Philipp Tomsich
` (6 subsequent siblings)
18 siblings, 1 reply; 46+ messages in thread
From: Philipp Tomsich @ 2017-10-10 14:21 UTC (permalink / raw)
To: u-boot
As no '.type' was set for save_boot_params_ret in start.S, binutils
did not track whether it was emitted as A32 or T32. By properly
marking save_boot_params_ret as a potential function entry, we can
make sure that the compiler will insert the appropriate instructions
for branching to save_boot_params_ret both for call-sites emitted as
A32 and T32.
Reported-by: Andy Yan <andy.yan@rock-chips.com>
Signed-off-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com>
Tested-by: Andy Yan <andy.yan@rock-chips.com>
---
Changes in v5: None
Changes in v4: None
Changes in v3:
- tracked the root-cause why no interwork branch was emitted and fixed
it using a '.type'-directive in start.S to mark save_boot_params_ret
as a (possible) function-entry.
Changes in v2: None
arch/arm/cpu/armv7/start.S | 1 +
1 file changed, 1 insertion(+)
diff --git a/arch/arm/cpu/armv7/start.S b/arch/arm/cpu/armv7/start.S
index 7b84a7a..95a0b52 100644
--- a/arch/arm/cpu/armv7/start.S
+++ b/arch/arm/cpu/armv7/start.S
@@ -31,6 +31,7 @@
.globl reset
.globl save_boot_params_ret
+ .type save_boot_params_ret,%function
#ifdef CONFIG_ARMV7_LPAE
.global switch_to_hypervisor_ret
#endif
--
2.1.4
^ permalink raw reply related [flat|nested] 46+ messages in thread
* [U-Boot] [PATCH v5 13/18] arm: provide a PCS-compliant setjmp implementation
2017-10-10 14:21 [U-Boot] [PATCH v5 00/18] rockchip: back-to-bootrom: replace assembly-implementation with C-code Philipp Tomsich
` (11 preceding siblings ...)
2017-10-10 14:21 ` [U-Boot] [PATCH v5 12/18] arm: mark save_boot_params_ret as a function Philipp Tomsich
@ 2017-10-10 14:21 ` Philipp Tomsich
2017-11-07 14:18 ` [U-Boot] [U-Boot, v5, " Philipp Tomsich
2017-10-10 14:21 ` [U-Boot] [PATCH v5 14/18] rockchip: back-to-bootrom: replace assembly-implementation with C-code Philipp Tomsich
` (5 subsequent siblings)
18 siblings, 1 reply; 46+ messages in thread
From: Philipp Tomsich @ 2017-10-10 14:21 UTC (permalink / raw)
To: u-boot
The previous setjmp-implementation (as a static inline function that
contained an 'asm volatile' sequence) was extremely fragile: (some
versions of) GCC optimised the set of registers. One critical example
was the removal of 'r9' from the clobber list, if -ffixed-reg9 was
supplied.
To increase robustness and ensure PCS-compliant behaviour, the setjmp
and longjmp implementation are now in assembly and closely match what
one would expect to find in a libc implementation.
Signed-off-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com>
Tested-by: Andy Yan <andy.yan@rock-chips.com>
---
Changes in v5: None
Changes in v4: None
Changes in v3:
- converted setjmp/longjmp from inline-assembly to separate .S files
to improve predicatability if emitted code
Changes in v2: None
arch/arm/include/asm/setjmp.h | 94 ++++++-------------------------------------
arch/arm/lib/Makefile | 6 +++
arch/arm/lib/setjmp.S | 37 +++++++++++++++++
arch/arm/lib/setjmp_aarch64.S | 42 +++++++++++++++++++
4 files changed, 98 insertions(+), 81 deletions(-)
create mode 100644 arch/arm/lib/setjmp.S
create mode 100644 arch/arm/lib/setjmp_aarch64.S
diff --git a/arch/arm/include/asm/setjmp.h b/arch/arm/include/asm/setjmp.h
index c3399a7..517beeb 100644
--- a/arch/arm/include/asm/setjmp.h
+++ b/arch/arm/include/asm/setjmp.h
@@ -1,6 +1,6 @@
/*
- * (C) Copyright 2016
- * Alexander Graf <agraf@suse.de>
+ * (C) Copyright 2017 Theobroma Systems Design und Consulting GmbH
+ * (C) Copyright 2016 Alexander Graf <agraf@suse.de>
*
* SPDX-License-Identifier: GPL-2.0+
*/
@@ -8,89 +8,21 @@
#ifndef _SETJMP_H_
#define _SETJMP_H_ 1
+/*
+ * This really should be opaque, but the EFI implementation wrongly
+ * assumes that a 'struct jmp_buf_data' is defined.
+ */
struct jmp_buf_data {
- ulong target;
- ulong regs[5];
- int ret;
-};
-
-typedef struct jmp_buf_data jmp_buf[1];
-
-static inline int setjmp(jmp_buf jmp)
-{
- jmp->ret = 0;
-
-#ifdef CONFIG_ARM64
- asm volatile(
- "adr x1, jmp_target\n"
- "str x1, %0\n"
- "stp x26, x27, %1\n"
- "stp x28, x29, %2\n"
- "mov x1, sp\n"
- "str x1, %3\n"
- "jmp_target: "
- : "=m" (jmp->target), "=m" (jmp->regs[0]),
- "=m" (jmp->regs[2]), "=m" (jmp->regs[4])
- :
- : "x0", "x1", "x2", "x3", "x4", "x5", "x6", "x7",
- "x8", "x9", "x10", "x11", "x12", "x13", "x14", "x15",
- "x16", "x17", "x18", "x19", "x20", "x21", "x22",
- "x23", "x24", "x25", /* x26, x27, x28, x29, sp */
- "x30", "cc", "memory");
-#else
- asm volatile(
-#if CONFIG_IS_ENABLED(SYS_THUMB_BUILD)
- ".align 2\n"
- "adr r0, jmp_target\n"
- "add r0, r0, $1\n"
+#if defined(__aarch64__)
+ u64 regs[13];
#else
- "adr r0, jmp_target\n"
-#endif
- "mov r1, %0\n"
- "mov r2, sp\n"
- "stm r1!, {r0, r2, r4, r5, r6, r7}\n"
- ".align 2\n"
- "jmp_target: \n"
- :
- : "l" (&jmp->target)
- : "r0", "r1", "r2", "r3", /* "r4", "r5", "r6", "r7", */
- "r8", "r9", "r10", "r11", /* sp, */ "ip", "lr",
- "cc", "memory");
-#endif
-
- return jmp->ret;
-}
-
-static inline __noreturn void longjmp(jmp_buf jmp, int ret)
-{
- jmp->ret = ret;
-
-#ifdef CONFIG_ARM64
- asm volatile(
- "ldr x0, %0\n"
- "ldr x1, %3\n"
- "mov sp, x1\n"
- "ldp x26, x27, %1\n"
- "ldp x28, x25, %2\n"
- "mov x29, x25\n"
- "br x0\n"
- :
- : "m" (jmp->target), "m" (jmp->regs[0]), "m" (jmp->regs[2]),
- "m" (jmp->regs[4])
- : "x0", "x1", "x25", "x26", "x27", "x28");
-#else
- asm volatile(
- "mov r1, %0\n"
- "ldm r1!, {r0, r2, r4, r5, r6, r7}\n"
- "mov sp, r2\n"
- "bx r0\n"
- :
- : "l" (&jmp->target)
- : "r1");
+ u32 regs[10]; /* r4-r9, sl, fp, sp, lr */
#endif
+};
- while (1) { }
-}
+typedef struct jmp_buf_data jmp_buf[1];
+int setjmp(jmp_buf jmp);
+void longjmp(jmp_buf jmp, int ret);
#endif /* _SETJMP_H_ */
diff --git a/arch/arm/lib/Makefile b/arch/arm/lib/Makefile
index 6e1c436..abffa10 100644
--- a/arch/arm/lib/Makefile
+++ b/arch/arm/lib/Makefile
@@ -17,6 +17,12 @@ else
obj-y += vectors.o crt0.o
endif
+ifdef CONFIG_ARM64
+obj-y += setjmp_aarch64.o
+else
+obj-y += setjmp.o
+endif
+
ifndef CONFIG_SPL_BUILD
ifdef CONFIG_ARM64
obj-y += relocate_64.o
diff --git a/arch/arm/lib/setjmp.S b/arch/arm/lib/setjmp.S
new file mode 100644
index 0000000..6746e5e
--- /dev/null
+++ b/arch/arm/lib/setjmp.S
@@ -0,0 +1,37 @@
+/*
+ * (C) 2017 Theobroma Systems Design und Consulting GmbH
+ *
+ * SPDX-License-Identifier: GPL-2.0+
+ */
+
+#include <config.h>
+#include <asm/assembler.h>
+#include <linux/linkage.h>
+
+.pushsection .text.setjmp, "ax"
+ENTRY(setjmp)
+ /*
+ * A subroutine must preserve the contents of the registers
+ * r4-r8, r10, r11 (v1-v5, v7 and v8) and SP (and r9 in PCS
+ * variants that designate r9 as v6).
+ */
+ mov ip, sp
+ stm a1, {v1-v8, ip, lr}
+ mov a1, #0
+ bx lr
+ENDPROC(setjmp)
+.popsection
+
+.pushsection .text.longjmp, "ax"
+ENTRY(longjmp)
+ ldm a1, {v1-v8, ip, lr}
+ mov sp, ip
+ mov a1, a2
+ /* If we were passed a return value of zero, return one instead */
+ cmp a1, #0
+ bne 1f
+ mov a1, #1
+1:
+ bx lr
+ENDPROC(longjmp)
+.popsection
diff --git a/arch/arm/lib/setjmp_aarch64.S b/arch/arm/lib/setjmp_aarch64.S
new file mode 100644
index 0000000..b68edb8
--- /dev/null
+++ b/arch/arm/lib/setjmp_aarch64.S
@@ -0,0 +1,42 @@
+/*
+ * (C) 2017 Theobroma Systems Design und Consulting GmbH
+ *
+ * SPDX-License-Identifier: GPL-2.0+
+ */
+
+#include <config.h>
+#include <asm/macro.h>
+#include <linux/linkage.h>
+
+.pushsection .text.setjmp, "ax"
+ENTRY(setjmp)
+ /* Preserve all callee-saved registers and the SP */
+ stp x19, x20, [x0,#0]
+ stp x21, x22, [x0,#16]
+ stp x23, x24, [x0,#32]
+ stp x25, x26, [x0,#48]
+ stp x27, x28, [x0,#64]
+ stp x29, x30, [x0,#80]
+ mov x2, sp
+ str x2, [x0, #96]
+ mov x0, #0
+ ret
+ENDPROC(setjmp)
+.popsection
+
+.pushsection .text.longjmp, "ax"
+ENTRY(longjmp)
+ ldp x19, x20, [x0,#0]
+ ldp x21, x22, [x0,#16]
+ ldp x23, x24, [x0,#32]
+ ldp x25, x26, [x0,#48]
+ ldp x27, x28, [x0,#64]
+ ldp x29, x30, [x0,#80]
+ ldr x2, [x0,#96]
+ mov sp, x2
+ /* Move the return value in place, but return 1 if passed 0. */
+ adds x0, xzr, x1
+ csinc x0, x0, xzr, ne
+ ret
+ENDPROC(longjmp)
+.popsection
--
2.1.4
^ permalink raw reply related [flat|nested] 46+ messages in thread
* [U-Boot] [PATCH v5 14/18] rockchip: back-to-bootrom: replace assembly-implementation with C-code
2017-10-10 14:21 [U-Boot] [PATCH v5 00/18] rockchip: back-to-bootrom: replace assembly-implementation with C-code Philipp Tomsich
` (12 preceding siblings ...)
2017-10-10 14:21 ` [U-Boot] [PATCH v5 13/18] arm: provide a PCS-compliant setjmp implementation Philipp Tomsich
@ 2017-10-10 14:21 ` Philipp Tomsich
2017-11-07 14:18 ` [U-Boot] [U-Boot, v5, " Philipp Tomsich
2017-10-10 14:21 ` [U-Boot] [PATCH v5 15/18] rockchip: rk3188: use boot0 hook to load up SPL in 2 steps Philipp Tomsich
` (4 subsequent siblings)
18 siblings, 1 reply; 46+ messages in thread
From: Philipp Tomsich @ 2017-10-10 14:21 UTC (permalink / raw)
To: u-boot
The back-to-bootrom implementation for Rockchip has always relied on
the stack-pointer being valid on entry, so there was little reason to
have this as an assembly implementation.
This provides a new C-only implementation of save_boot_params and
back_to_bootrom (relying on setjmp/longjmp) and removes the older
assembly-only implementation.
Signed-off-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com>
Tested-by: Andy Yan <andy.yan@rock-chips.com>
---
Changes in v5: None
Changes in v4: None
Changes in v3: None
Changes in v2: None
arch/arm/include/asm/arch-rockchip/bootrom.h | 27 ++++++++---
arch/arm/mach-rockchip/Makefile | 4 +-
arch/arm/mach-rockchip/bootrom.c | 52 ++++++++++++++++++++-
arch/arm/mach-rockchip/save_boot_param.S | 69 ----------------------------
4 files changed, 73 insertions(+), 79 deletions(-)
delete mode 100644 arch/arm/mach-rockchip/save_boot_param.S
diff --git a/arch/arm/include/asm/arch-rockchip/bootrom.h b/arch/arm/include/asm/arch-rockchip/bootrom.h
index 169cc5e..2f61a33 100644
--- a/arch/arm/include/asm/arch-rockchip/bootrom.h
+++ b/arch/arm/include/asm/arch-rockchip/bootrom.h
@@ -1,5 +1,6 @@
/*
* (C) Copyright 2017 Heiko Stuebner <heiko@sntech.de>
+ * (C) Copyright 2017 Theobroma Systems Design und Consulting GmbH
*
* SPDX-License-Identifier: GPL-2.0
*/
@@ -14,15 +15,27 @@
extern u32 SAVE_SP_ADDR;
/**
- * Hand control back to the bootrom to load another
- * boot stage.
+ * back_to_bootrom() - return to bootrom (for TPL/SPL), passing a
+ * result code
+ *
+ * Transfer control back to the Rockchip BROM, restoring necessary
+ * register context and passing a command/result code to the BROM
+ * to instruct its next actions (e.g. continue boot sequence, enter
+ * download mode, ...).
+ *
+ * This function does not return.
*/
-void back_to_bootrom(void);
+enum rockchip_bootrom_cmd {
+ /*
+ * These can not start at 0, as 0 has a special meaning
+ * for setjmp().
+ */
-/**
- * Assembler component for the above (do not call this directly)
- */
-void _back_to_bootrom_s(void);
+ BROM_BOOT_NEXTSTAGE = 1, /* continue boot-sequence */
+ BROM_BOOT_ENTER_DNL, /* have BROM enter download-mode */
+};
+
+void back_to_bootrom(void);
/**
* Boot-device identifiers as used by the BROM
diff --git a/arch/arm/mach-rockchip/Makefile b/arch/arm/mach-rockchip/Makefile
index daafc8d..b875dfc 100644
--- a/arch/arm/mach-rockchip/Makefile
+++ b/arch/arm/mach-rockchip/Makefile
@@ -8,8 +8,8 @@
# this may have entered from ATF with the stack-pointer pointing to
# inaccessible/protected memory (and the bootrom-helper assumes that
# the stack-pointer is valid before switching to the U-Boot stack).
-obj-spl-$(CONFIG_ROCKCHIP_BROM_HELPER) += bootrom.o save_boot_param.o
-obj-tpl-$(CONFIG_ROCKCHIP_BROM_HELPER) += bootrom.o save_boot_param.o
+obj-spl-$(CONFIG_ROCKCHIP_BROM_HELPER) += bootrom.o
+obj-tpl-$(CONFIG_ROCKCHIP_BROM_HELPER) += bootrom.o
obj-tpl-$(CONFIG_ROCKCHIP_RK3188) += rk3188-board-tpl.o
obj-tpl-$(CONFIG_ROCKCHIP_RK3288) += rk3288-board-tpl.o
diff --git a/arch/arm/mach-rockchip/bootrom.c b/arch/arm/mach-rockchip/bootrom.c
index 8380e4e..7b9b307 100644
--- a/arch/arm/mach-rockchip/bootrom.c
+++ b/arch/arm/mach-rockchip/bootrom.c
@@ -6,11 +6,61 @@
#include <common.h>
#include <asm/arch/bootrom.h>
+#include <asm/setjmp.h>
+#include <asm/system.h>
+
+/*
+ * Force the jmp_buf to the data-section, as .bss will not be valid
+ * when save_boot_params is invoked.
+ */
+static jmp_buf brom_ctx __section(".data");
void back_to_bootrom(void)
{
#if CONFIG_IS_ENABLED(LIBCOMMON_SUPPORT)
puts("Returning to boot ROM...\n");
#endif
- _back_to_bootrom_s();
+ longjmp(brom_ctx, BROM_BOOT_NEXTSTAGE);
+}
+
+/*
+ * All Rockchip BROM implementations enter with a valid stack-pointer,
+ * so this can safely be implemented in C (providing a single
+ * implementation both for ARMv7 and AArch64).
+ */
+int save_boot_params(void)
+{
+ int ret = setjmp(brom_ctx);
+
+ switch (ret) {
+ case 0:
+ /*
+ * This is the initial pass through this function
+ * (i.e. saving the context), setjmp just setup up the
+ * brom_ctx: transfer back into the startup-code at
+ * 'save_boot_params_ret' and let the compiler know
+ * that this will not return.
+ */
+ save_boot_params_ret();
+ while (true)
+ /* does not return */;
+ break;
+
+ case BROM_BOOT_NEXTSTAGE:
+ /*
+ * To instruct the BROM to boot the next stage, we
+ * need to return 0 to it: i.e. we need to rewrite
+ * the return code once more.
+ */
+ ret = 0;
+ break;
+
+ default:
+#if CONFIG_IS_ENABLED(LIBCOMMON_SUPPORT)
+ puts("FATAL: unexpected command to back_to_bootrom()\n");
+#endif
+ hang();
+ };
+
+ return ret;
}
diff --git a/arch/arm/mach-rockchip/save_boot_param.S b/arch/arm/mach-rockchip/save_boot_param.S
deleted file mode 100644
index 50fce20..0000000
--- a/arch/arm/mach-rockchip/save_boot_param.S
+++ /dev/null
@@ -1,69 +0,0 @@
-/*
- * (C) Copyright 2016 Rockchip Electronics Co., Ltd
- * (C) Copyright 2017 Theobroma Systems Design und Consulting GmbH
- *
- * SPDX-License-Identifier: GPL-2.0+
- */
-
-#include <linux/linkage.h>
-
-#if defined(CONFIG_ARM64)
-.globl SAVE_SP_ADDR
-SAVE_SP_ADDR:
- .quad 0
-
-ENTRY(save_boot_params)
- sub sp, sp, #0x60
- stp x29, x30, [sp, #0x50]
- stp x27, x28, [sp, #0x40]
- stp x25, x26, [sp, #0x30]
- stp x23, x24, [sp, #0x20]
- stp x21, x22, [sp, #0x10]
- stp x19, x20, [sp, #0]
- ldr x8, =SAVE_SP_ADDR
- mov x9, sp
- str x9, [x8]
- b save_boot_params_ret /* back to my caller */
-ENDPROC(save_boot_params)
-
-.globl _back_to_bootrom_s
-ENTRY(_back_to_bootrom_s)
- ldr x0, =SAVE_SP_ADDR
- ldr x0, [x0]
- mov sp, x0
- ldp x29, x30, [sp, #0x50]
- ldp x27, x28, [sp, #0x40]
- ldp x25, x26, [sp, #0x30]
- ldp x23, x24, [sp, #0x20]
- ldp x21, x22, [sp, #0x10]
- ldp x19, x20, [sp]
- add sp, sp, #0x60
- mov x0, xzr
- ret
-ENDPROC(_back_to_bootrom_s)
-#else
-.globl SAVE_SP_ADDR
-SAVE_SP_ADDR:
- .word 0
-
-/*
- * void save_boot_params
- *
- * Save sp, lr, r1~r12
- */
-ENTRY(save_boot_params)
- push {r1-r12, lr}
- ldr r0, =SAVE_SP_ADDR
- str sp, [r0]
- b save_boot_params_ret @ back to my caller
-ENDPROC(save_boot_params)
-
-
-.globl _back_to_bootrom_s
-ENTRY(_back_to_bootrom_s)
- ldr r0, =SAVE_SP_ADDR
- ldr sp, [r0]
- mov r0, #0
- pop {r1-r12, pc}
-ENDPROC(_back_to_bootrom_s)
-#endif
--
2.1.4
^ permalink raw reply related [flat|nested] 46+ messages in thread
* [U-Boot] [PATCH v5 15/18] rockchip: rk3188: use boot0 hook to load up SPL in 2 steps
2017-10-10 14:21 [U-Boot] [PATCH v5 00/18] rockchip: back-to-bootrom: replace assembly-implementation with C-code Philipp Tomsich
` (13 preceding siblings ...)
2017-10-10 14:21 ` [U-Boot] [PATCH v5 14/18] rockchip: back-to-bootrom: replace assembly-implementation with C-code Philipp Tomsich
@ 2017-10-10 14:21 ` Philipp Tomsich
2017-11-07 14:18 ` [U-Boot] [U-Boot, v5, " Philipp Tomsich
2017-10-10 14:21 ` [U-Boot] [PATCH v5 16/18] rockchip: back-to-bootrom: allow passing a cmd to the bootrom Philipp Tomsich
` (3 subsequent siblings)
18 siblings, 1 reply; 46+ messages in thread
From: Philipp Tomsich @ 2017-10-10 14:21 UTC (permalink / raw)
To: u-boot
For the RK3188, the BROM will attempt to load up the first stage
image (SPL for the RK3188) in two steps: first 1KB to offset 0x800
in the SRAM and then the remainder to offset 0xc00 in the SRAM.
It always enters at 0x804, though.
With this changeset, the RK3188 boot removes the TPL (stub) stage and
builds a single SPL binary that utilizes the early back-to-bootrom via
the boot0-hook.
Consequently, the passing of the saved boot params via pmu->os_reg[2]
is also removed.
Signed-off-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com>
---
Changes in v5: None
Changes in v4:
- after merging the 'back-to-bootrom' series with the 'boot0-hook'
series, this drops the TPL stub and builds only a single SPL image
that uses the 'early back-to-bootrom' logic originally implemented
by Pawel for the RK3066.
- changes the SPL_STACK_BASE to +0x800 (from +0x804), as the boot0
hook already reserves the space for the SPL magic (previously
inserted by mkimage)
Changes in v3: None
Changes in v2:
- [added in v2] chain back_to_bootrom calls for SPL, first returning
to the TPL (using the same mechanism) and the to the BROM from the
TPL
arch/arm/mach-rockchip/Kconfig | 2 +-
arch/arm/mach-rockchip/Makefile | 1 -
arch/arm/mach-rockchip/rk3188-board-spl.c | 10 ----
arch/arm/mach-rockchip/rk3188-board-tpl.c | 86 -------------------------------
doc/README.rockchip | 10 ++--
include/configs/rk3188_common.h | 12 ++---
6 files changed, 9 insertions(+), 112 deletions(-)
delete mode 100644 arch/arm/mach-rockchip/rk3188-board-tpl.c
diff --git a/arch/arm/mach-rockchip/Kconfig b/arch/arm/mach-rockchip/Kconfig
index d59a1d5..09f2c45 100644
--- a/arch/arm/mach-rockchip/Kconfig
+++ b/arch/arm/mach-rockchip/Kconfig
@@ -18,7 +18,7 @@ config ROCKCHIP_RK3188
select SUPPORT_SPL
select SUPPORT_TPL
select SPL
- select TPL
+ select SPL_ROCKCHIP_EARLYRETURN_TO_BROM
select BOARD_LATE_INIT
select ROCKCHIP_BROM_HELPER
help
diff --git a/arch/arm/mach-rockchip/Makefile b/arch/arm/mach-rockchip/Makefile
index b875dfc..c15e9bf 100644
--- a/arch/arm/mach-rockchip/Makefile
+++ b/arch/arm/mach-rockchip/Makefile
@@ -11,7 +11,6 @@
obj-spl-$(CONFIG_ROCKCHIP_BROM_HELPER) += bootrom.o
obj-tpl-$(CONFIG_ROCKCHIP_BROM_HELPER) += bootrom.o
-obj-tpl-$(CONFIG_ROCKCHIP_RK3188) += rk3188-board-tpl.o
obj-tpl-$(CONFIG_ROCKCHIP_RK3288) += rk3288-board-tpl.o
obj-tpl-$(CONFIG_ROCKCHIP_RK3368) += rk3368-board-tpl.o
diff --git a/arch/arm/mach-rockchip/rk3188-board-spl.c b/arch/arm/mach-rockchip/rk3188-board-spl.c
index 406207e..05d4ae6 100644
--- a/arch/arm/mach-rockchip/rk3188-board-spl.c
+++ b/arch/arm/mach-rockchip/rk3188-board-spl.c
@@ -101,7 +101,6 @@ static int setup_arm_clock(void)
void board_init_f(ulong dummy)
{
struct udevice *pinctrl, *dev;
- struct rk3188_pmu *pmu;
int ret;
/* Example code showing how to enable the debug UART on RK3188 */
@@ -145,15 +144,6 @@ void board_init_f(ulong dummy)
return;
}
- /*
- * Recover the bootrom's stackpointer.
- * For whatever reason needs to run after rockchip_get_clk.
- */
- pmu = syscon_get_first_range(ROCKCHIP_SYSCON_PMU);
- if (IS_ERR(pmu))
- pr_err("pmu syscon returned %ld\n", PTR_ERR(pmu));
- SAVE_SP_ADDR = readl(&pmu->sys_reg[2]);
-
ret = uclass_get_device(UCLASS_PINCTRL, 0, &pinctrl);
if (ret) {
debug("Pinctrl init failed: %d\n", ret);
diff --git a/arch/arm/mach-rockchip/rk3188-board-tpl.c b/arch/arm/mach-rockchip/rk3188-board-tpl.c
deleted file mode 100644
index b458ef6..0000000
--- a/arch/arm/mach-rockchip/rk3188-board-tpl.c
+++ /dev/null
@@ -1,86 +0,0 @@
-/*
- * (C) Copyright 2015 Google, Inc
- *
- * SPDX-License-Identifier: GPL-2.0+
- */
-
-#include <common.h>
-#include <debug_uart.h>
-#include <spl.h>
-#include <asm/io.h>
-#include <asm/arch/bootrom.h>
-#include <asm/arch/pmu_rk3188.h>
-
-DECLARE_GLOBAL_DATA_PTR;
-
-/* track how often we were entered */
-static int rk3188_num_entries __attribute__ ((section(".data")));
-
-#define PMU_BASE 0x20004000
-#define SPL_ENTRY 0x10080C00
-
-static void jump_to_spl(void)
-{
- typedef void __noreturn (*image_entry_noargs_t)(void);
-
- struct rk3188_pmu * const pmu = (void *)PMU_BASE;
- image_entry_noargs_t tpl_entry =
- (image_entry_noargs_t)(unsigned long)SPL_ENTRY;
-
- /* Store the SAVE_SP_ADDR in a location shared with SPL. */
- writel(SAVE_SP_ADDR, &pmu->sys_reg[2]);
- tpl_entry();
-}
-
-void board_init_f(ulong dummy)
-{
- /* Example code showing how to enable the debug UART on RK3188 */
-#ifdef EARLY_UART
-#include <asm/arch/grf_rk3188.h>
- /* Enable early UART on the RK3188 */
-#define GRF_BASE 0x20008000
- struct rk3188_grf * const grf = (void *)GRF_BASE;
-
- rk_clrsetreg(&grf->gpio1b_iomux,
- GPIO1B1_MASK << GPIO1B1_SHIFT |
- GPIO1B0_MASK << GPIO1B0_SHIFT,
- GPIO1B1_UART2_SOUT << GPIO1B1_SHIFT |
- GPIO1B0_UART2_SIN << GPIO1B0_SHIFT);
- /*
- * Debug UART can be used from here if required:
- *
- * debug_uart_init();
- * printch('a');
- * printhex8(0x1234);
- * printascii("string");
- */
- debug_uart_init();
-
- printch('t');
- printch('p');
- printch('l');
- printch('-');
- printch(rk3188_num_entries + 1 + '0');
- printch('\n');
-#endif
-
- rk3188_num_entries++;
-
- if (rk3188_num_entries == 1) {
- /*
- * The original loader did some very basic integrity
- * checking at this point, but the remaining few bytes
- * could be used for any improvement making sense
- * really early on.
- */
-
- back_to_bootrom();
- } else {
- /*
- * TPL part of the loader should now wait for us
- *@offset 0xC00 in the sram. Should never return
- * from there.
- */
- jump_to_spl();
- }
-}
diff --git a/doc/README.rockchip b/doc/README.rockchip
index 4b7be0b..57f551b 100644
--- a/doc/README.rockchip
+++ b/doc/README.rockchip
@@ -176,17 +176,17 @@ described above, but the image creation needs a bit more care.
The bootrom of rk3188 expects to find a small 1kb loader which returns
control to the bootrom, after which it will load the real loader, which
-can then be up to 29kb in size and does the regular ddr init.
+can then be up to 29kb in size and does the regular ddr init. This is
+handled by a single image (built as the SPL stage) that tests whether
+it is handled for the first or second time via code executed from the
+boot0-hook.
Additionally the rk3188 requires everything the bootrom loads to be
rc4-encrypted. Except for the very first stage the bootrom always reads
and decodes 2kb pages, so files should be sized accordingly.
# copy tpl, pad to 1020 bytes and append spl
-cat tpl/u-boot-tpl.bin > tplspl.bin
-truncate -s 1020 tplspl.bin
-cat spl/u-boot-spl.bin >> tplspl.bin
-tools/mkimage -n rk3188 -T rksd -d tplspl.bin out
+tools/mkimage -n rk3188 -T rksd -d spl/u-boot-spl.bin out
# truncate, encode and append u-boot.bin
truncate -s %2048 u-boot.bin
diff --git a/include/configs/rk3188_common.h b/include/configs/rk3188_common.h
index 5e46234..9824a10 100644
--- a/include/configs/rk3188_common.h
+++ b/include/configs/rk3188_common.h
@@ -37,14 +37,9 @@
#define CONFIG_ROCKCHIP_MAX_INIT_SIZE (0x8000 - 0x800)
#define CONFIG_ROCKCHIP_CHIP_TAG "RK31"
-#ifdef CONFIG_TPL_BUILD
-#define CONFIG_SPL_TEXT_BASE 0x10080804
-/* tpl size 1kb - 4byte RK31 header */
-#define CONFIG_SPL_MAX_SIZE (0x400 - 0x4)
-#elif defined(CONFIG_SPL_BUILD)
-/* spl size 32kb sram - 2kb bootrom - 1kb spl */
-#define CONFIG_SPL_MAX_SIZE (0x8000 - 0xC00)
-#define CONFIG_SPL_TEXT_BASE 0x10080C00
+#define CONFIG_SPL_TEXT_BASE 0x10080800
+/* spl size 32kb sram - 2kb bootrom */
+#define CONFIG_SPL_MAX_SIZE (0x8000 - 0x800)
#define CONFIG_SPL_FRAMEWORK 1
#define CONFIG_SPL_CLK 1
#define CONFIG_SPL_PINCTRL 1
@@ -53,7 +48,6 @@
#define CONFIG_SPL_RAM 1
#define CONFIG_SPL_DRIVERS_MISC_SUPPORT 1
#define CONFIG_ROCKCHIP_SERIAL 1
-#endif
#define CONFIG_SPL_STACK 0x10087fff
--
2.1.4
^ permalink raw reply related [flat|nested] 46+ messages in thread
* [U-Boot] [PATCH v5 16/18] rockchip: back-to-bootrom: allow passing a cmd to the bootrom
2017-10-10 14:21 [U-Boot] [PATCH v5 00/18] rockchip: back-to-bootrom: replace assembly-implementation with C-code Philipp Tomsich
` (14 preceding siblings ...)
2017-10-10 14:21 ` [U-Boot] [PATCH v5 15/18] rockchip: rk3188: use boot0 hook to load up SPL in 2 steps Philipp Tomsich
@ 2017-10-10 14:21 ` Philipp Tomsich
2017-11-07 14:18 ` [U-Boot] [U-Boot, v5, " Philipp Tomsich
2017-10-10 14:21 ` [U-Boot] [PATCH v5 17/18] rockchip: rk3188: move CONFIG_SPL_* entries from rk3188_common.h to Kconfig Philipp Tomsich
` (2 subsequent siblings)
18 siblings, 1 reply; 46+ messages in thread
From: Philipp Tomsich @ 2017-10-10 14:21 UTC (permalink / raw)
To: u-boot
The BROM supports forcing it to enter download-mode, if an appropriate
result/cmd-word is returned to it. There already is a series to
support this in review, so this prepares the (newly C-version) of the
back-to-bootrom code to accept a cmd to passed on to the BROM.
All the existing call-sites are adjusted to match the changed function
signature.
Signed-off-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com>
Tested-by: Andy Yan <andy.yan@rock-chips.com>
---
Changes in v5: None
Changes in v4:
- no longer updates rk3188-board-tpl.c (as we have just removed it
in an earlier commit)
Changes in v3: None
Changes in v2:
- also covers the RK3188 (which I had originally missed)
arch/arm/include/asm/arch-rockchip/bootrom.h | 5 ++++-
arch/arm/mach-rockchip/bootrom.c | 4 ++--
arch/arm/mach-rockchip/rk3036-board-spl.c | 2 +-
arch/arm/mach-rockchip/rk3188-board-spl.c | 4 ++--
arch/arm/mach-rockchip/rk322x-board-spl.c | 2 +-
arch/arm/mach-rockchip/rk3288-board-spl.c | 4 ++--
arch/arm/mach-rockchip/rk3288-board-tpl.c | 2 +-
arch/arm/mach-rockchip/rk3368-board-tpl.c | 2 +-
arch/arm/mach-rockchip/rk3399-board-spl.c | 2 +-
9 files changed, 15 insertions(+), 12 deletions(-)
diff --git a/arch/arm/include/asm/arch-rockchip/bootrom.h b/arch/arm/include/asm/arch-rockchip/bootrom.h
index 2f61a33..103b799 100644
--- a/arch/arm/include/asm/arch-rockchip/bootrom.h
+++ b/arch/arm/include/asm/arch-rockchip/bootrom.h
@@ -24,6 +24,9 @@ extern u32 SAVE_SP_ADDR;
* download mode, ...).
*
* This function does not return.
+ *
+ * @brom_cmd: indicates how the bootrom should continue the boot
+ * sequence (e.g. load the next stage)
*/
enum rockchip_bootrom_cmd {
/*
@@ -35,7 +38,7 @@ enum rockchip_bootrom_cmd {
BROM_BOOT_ENTER_DNL, /* have BROM enter download-mode */
};
-void back_to_bootrom(void);
+void back_to_bootrom(enum rockchip_bootrom_cmd brom_cmd);
/**
* Boot-device identifiers as used by the BROM
diff --git a/arch/arm/mach-rockchip/bootrom.c b/arch/arm/mach-rockchip/bootrom.c
index 7b9b307..e369fdc 100644
--- a/arch/arm/mach-rockchip/bootrom.c
+++ b/arch/arm/mach-rockchip/bootrom.c
@@ -15,12 +15,12 @@
*/
static jmp_buf brom_ctx __section(".data");
-void back_to_bootrom(void)
+void back_to_bootrom(enum rockchip_bootrom_cmd brom_cmd)
{
#if CONFIG_IS_ENABLED(LIBCOMMON_SUPPORT)
puts("Returning to boot ROM...\n");
#endif
- longjmp(brom_ctx, BROM_BOOT_NEXTSTAGE);
+ longjmp(brom_ctx, brom_cmd);
}
/*
diff --git a/arch/arm/mach-rockchip/rk3036-board-spl.c b/arch/arm/mach-rockchip/rk3036-board-spl.c
index 9458201..550e3a1 100644
--- a/arch/arm/mach-rockchip/rk3036-board-spl.c
+++ b/arch/arm/mach-rockchip/rk3036-board-spl.c
@@ -40,7 +40,7 @@ void board_init_f(ulong dummy)
sdram_init();
/* return to maskrom */
- back_to_bootrom();
+ back_to_bootrom(BROM_BOOT_NEXTSTAGE);
}
/* Place Holders */
diff --git a/arch/arm/mach-rockchip/rk3188-board-spl.c b/arch/arm/mach-rockchip/rk3188-board-spl.c
index 05d4ae6..8e3b8ae 100644
--- a/arch/arm/mach-rockchip/rk3188-board-spl.c
+++ b/arch/arm/mach-rockchip/rk3188-board-spl.c
@@ -158,7 +158,7 @@ void board_init_f(ulong dummy)
setup_arm_clock();
#if CONFIG_IS_ENABLED(ROCKCHIP_BACK_TO_BROM) && !defined(CONFIG_SPL_BOARD_INIT)
- back_to_bootrom();
+ back_to_bootrom(BROM_BOOT_NEXTSTAGE);
#endif
}
@@ -219,7 +219,7 @@ void spl_board_init(void)
preloader_console_init();
#if CONFIG_IS_ENABLED(ROCKCHIP_BACK_TO_BROM)
- back_to_bootrom();
+ back_to_bootrom(BROM_BOOT_NEXTSTAGE);
#endif
return;
diff --git a/arch/arm/mach-rockchip/rk322x-board-spl.c b/arch/arm/mach-rockchip/rk322x-board-spl.c
index 4ddb8ba..35f4f97 100644
--- a/arch/arm/mach-rockchip/rk322x-board-spl.c
+++ b/arch/arm/mach-rockchip/rk322x-board-spl.c
@@ -76,6 +76,6 @@ void board_init_f(ulong dummy)
/* Disable the ddr secure region setting to make it non-secure */
rk_clrreg(SGRF_DDR_CON0, 0x4000);
#if defined(CONFIG_ROCKCHIP_SPL_BACK_TO_BROM) && !defined(CONFIG_SPL_BOARD_INIT)
- back_to_bootrom();
+ back_to_bootrom(BROM_BOOT_NEXTSTAGE);
#endif
}
diff --git a/arch/arm/mach-rockchip/rk3288-board-spl.c b/arch/arm/mach-rockchip/rk3288-board-spl.c
index 7b7fd5a..f64a548 100644
--- a/arch/arm/mach-rockchip/rk3288-board-spl.c
+++ b/arch/arm/mach-rockchip/rk3288-board-spl.c
@@ -216,7 +216,7 @@ void board_init_f(ulong dummy)
#endif
#if CONFIG_IS_ENABLED(ROCKCHIP_BACK_TO_BROM) && !defined(CONFIG_SPL_BOARD_INIT)
- back_to_bootrom();
+ back_to_bootrom(BROM_BOOT_NEXTSTAGE);
#endif
}
@@ -283,7 +283,7 @@ void spl_board_init(void)
preloader_console_init();
#if CONFIG_IS_ENABLED(ROCKCHIP_BACK_TO_BROM)
- back_to_bootrom();
+ back_to_bootrom(BROM_BOOT_NEXTSTAGE);
#endif
return;
err:
diff --git a/arch/arm/mach-rockchip/rk3288-board-tpl.c b/arch/arm/mach-rockchip/rk3288-board-tpl.c
index 3d08b5b..150beea 100644
--- a/arch/arm/mach-rockchip/rk3288-board-tpl.c
+++ b/arch/arm/mach-rockchip/rk3288-board-tpl.c
@@ -69,7 +69,7 @@ void board_init_f(ulong dummy)
void board_return_to_bootrom(void)
{
- back_to_bootrom();
+ back_to_bootrom(BROM_BOOT_NEXTSTAGE);
}
u32 spl_boot_device(void)
diff --git a/arch/arm/mach-rockchip/rk3368-board-tpl.c b/arch/arm/mach-rockchip/rk3368-board-tpl.c
index b3e6ffa..60d5aea 100644
--- a/arch/arm/mach-rockchip/rk3368-board-tpl.c
+++ b/arch/arm/mach-rockchip/rk3368-board-tpl.c
@@ -148,7 +148,7 @@ void board_init_f(ulong dummy)
void board_return_to_bootrom(void)
{
- back_to_bootrom();
+ back_to_bootrom(BROM_BOOT_NEXTSTAGE);
}
u32 spl_boot_device(void)
diff --git a/arch/arm/mach-rockchip/rk3399-board-spl.c b/arch/arm/mach-rockchip/rk3399-board-spl.c
index 9c20f56..b96903e 100644
--- a/arch/arm/mach-rockchip/rk3399-board-spl.c
+++ b/arch/arm/mach-rockchip/rk3399-board-spl.c
@@ -23,7 +23,7 @@ DECLARE_GLOBAL_DATA_PTR;
void board_return_to_bootrom(void)
{
- back_to_bootrom();
+ back_to_bootrom(BROM_BOOT_NEXTSTAGE);
}
static const char * const boot_devices[BROM_LAST_BOOTSOURCE + 1] = {
--
2.1.4
^ permalink raw reply related [flat|nested] 46+ messages in thread
* [U-Boot] [PATCH v5 17/18] rockchip: rk3188: move CONFIG_SPL_* entries from rk3188_common.h to Kconfig
2017-10-10 14:21 [U-Boot] [PATCH v5 00/18] rockchip: back-to-bootrom: replace assembly-implementation with C-code Philipp Tomsich
` (15 preceding siblings ...)
2017-10-10 14:21 ` [U-Boot] [PATCH v5 16/18] rockchip: back-to-bootrom: allow passing a cmd to the bootrom Philipp Tomsich
@ 2017-10-10 14:21 ` Philipp Tomsich
2017-11-07 14:18 ` [U-Boot] [U-Boot, v5, " Philipp Tomsich
2017-10-10 14:21 ` [U-Boot] [PATCH v5 18/18] rockchip: mkimage: remove unused code-paths (spl_boot0 is now implied) Philipp Tomsich
2017-10-11 13:07 ` [U-Boot] [PATCH v5 00/18] rockchip: back-to-bootrom: replace assembly-implementation with C-code Heiko Stuebner
18 siblings, 1 reply; 46+ messages in thread
From: Philipp Tomsich @ 2017-10-10 14:21 UTC (permalink / raw)
To: u-boot
There still are a few CONFIG_SPL_* options selected using defines from
rk3188_common.h instead of via Kconfig. This migrates those over to
Kconfig.
Signed-off-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com>
---
Changes in v5: None
Changes in v4: None
Changes in v3: None
Changes in v2: None
arch/arm/mach-rockchip/Kconfig | 7 ++++++-
include/configs/rk3188_common.h | 6 ------
2 files changed, 6 insertions(+), 7 deletions(-)
diff --git a/arch/arm/mach-rockchip/Kconfig b/arch/arm/mach-rockchip/Kconfig
index 09f2c45..36df484 100644
--- a/arch/arm/mach-rockchip/Kconfig
+++ b/arch/arm/mach-rockchip/Kconfig
@@ -16,8 +16,13 @@ config ROCKCHIP_RK3188
select CPU_V7
select SPL_BOARD_INIT if SPL
select SUPPORT_SPL
- select SUPPORT_TPL
select SPL
+ select SPL_CLK
+ select SPL_PINCTRL
+ select SPL_REGMAP
+ select SPL_SYSCON
+ select SPL_RAM
+ select SPL_DRIVERS_MISC_SUPPORT
select SPL_ROCKCHIP_EARLYRETURN_TO_BROM
select BOARD_LATE_INIT
select ROCKCHIP_BROM_HELPER
diff --git a/include/configs/rk3188_common.h b/include/configs/rk3188_common.h
index 9824a10..0382450 100644
--- a/include/configs/rk3188_common.h
+++ b/include/configs/rk3188_common.h
@@ -41,12 +41,6 @@
/* spl size 32kb sram - 2kb bootrom */
#define CONFIG_SPL_MAX_SIZE (0x8000 - 0x800)
#define CONFIG_SPL_FRAMEWORK 1
-#define CONFIG_SPL_CLK 1
-#define CONFIG_SPL_PINCTRL 1
-#define CONFIG_SPL_REGMAP 1
-#define CONFIG_SPL_SYSCON 1
-#define CONFIG_SPL_RAM 1
-#define CONFIG_SPL_DRIVERS_MISC_SUPPORT 1
#define CONFIG_ROCKCHIP_SERIAL 1
#define CONFIG_SPL_STACK 0x10087fff
--
2.1.4
^ permalink raw reply related [flat|nested] 46+ messages in thread
* [U-Boot] [PATCH v5 18/18] rockchip: mkimage: remove unused code-paths (spl_boot0 is now implied)
2017-10-10 14:21 [U-Boot] [PATCH v5 00/18] rockchip: back-to-bootrom: replace assembly-implementation with C-code Philipp Tomsich
` (16 preceding siblings ...)
2017-10-10 14:21 ` [U-Boot] [PATCH v5 17/18] rockchip: rk3188: move CONFIG_SPL_* entries from rk3188_common.h to Kconfig Philipp Tomsich
@ 2017-10-10 14:21 ` Philipp Tomsich
2017-11-07 14:18 ` [U-Boot] [U-Boot, v5, " Philipp Tomsich
2017-10-11 13:07 ` [U-Boot] [PATCH v5 00/18] rockchip: back-to-bootrom: replace assembly-implementation with C-code Heiko Stuebner
18 siblings, 1 reply; 46+ messages in thread
From: Philipp Tomsich @ 2017-10-10 14:21 UTC (permalink / raw)
To: u-boot
With all targets converted to generate prepadded images, this removes
the spl_boot0 field from our config structure and removes the unused
code-path (for images that are not prepadded): i.e. spl_boot0 is now
implied as 'true' and the code is specialised by removing the other
case.
Signed-off-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com>
---
Changes in v5: None
Changes in v4: None
Changes in v3: None
Changes in v2: None
tools/rkcommon.c | 45 ++++++++++++++-------------------------------
1 file changed, 14 insertions(+), 31 deletions(-)
diff --git a/tools/rkcommon.c b/tools/rkcommon.c
index c83260a..aed2b95 100644
--- a/tools/rkcommon.c
+++ b/tools/rkcommon.c
@@ -58,9 +58,6 @@ struct header1_info {
* @spl_hdr: Boot ROM requires a 4-bytes spl header
* @spl_size: Spl size(include extra 4-bytes spl header)
* @spl_rc4: RC4 encode the SPL binary (same key as header)
- * @spl_boot0: A new-style (ARM_SOC_BOOT0_HOOK) image that should
- * have the boot magic (e.g. 'RK33') written to its first
- * word.
*/
struct spl_info {
@@ -68,19 +65,18 @@ struct spl_info {
const char *spl_hdr;
const uint32_t spl_size;
const bool spl_rc4;
- const bool spl_boot0;
};
static struct spl_info spl_infos[] = {
- { "rk3036", "RK30", 0x1000, false, true },
- { "rk3128", "RK31", 0x1800, false, true },
- { "rk3188", "RK31", 0x8000 - 0x800, true, true },
- { "rk322x", "RK32", 0x8000 - 0x1000, false, true },
- { "rk3288", "RK32", 0x8000, false, true },
- { "rk3328", "RK32", 0x8000 - 0x1000, false, true },
- { "rk3368", "RK33", 0x8000 - 0x1000, false, true },
- { "rk3399", "RK33", 0x30000 - 0x2000, false, true },
- { "rv1108", "RK11", 0x1800, false, true },
+ { "rk3036", "RK30", 0x1000, false },
+ { "rk3128", "RK31", 0x1800, false },
+ { "rk3188", "RK31", 0x8000 - 0x800, true },
+ { "rk322x", "RK32", 0x8000 - 0x1000, false },
+ { "rk3288", "RK32", 0x8000, false },
+ { "rk3328", "RK32", 0x8000 - 0x1000, false },
+ { "rk3368", "RK33", 0x8000 - 0x1000, false },
+ { "rk3399", "RK33", 0x30000 - 0x2000, false },
+ { "rv1108", "RK11", 0x1800, false },
};
static unsigned char rc4_key[16] = {
@@ -158,16 +154,6 @@ bool rkcommon_need_rc4_spl(struct image_tool_params *params)
return info->spl_rc4;
}
-bool rkcommon_spl_is_boot0(struct image_tool_params *params)
-{
- struct spl_info *info = rkcommon_get_spl_info(params->imagename);
-
- /*
- * info would not be NULL, because of we checked params before.
- */
- return info->spl_boot0;
-}
-
static void rkcommon_set_header0(void *buf, uint file_size,
struct image_tool_params *params)
{
@@ -366,15 +352,12 @@ int rkcommon_vrec_header(struct image_tool_params *params,
* have the first 4 bytes reserved for the spl_name). Reserving
* these 4 bytes is done using the BOOT0_HOOK infrastructure.
*
- * Depending on this, the header is either 0x800 (if this is a
- * 'boot0'-style payload, which has reserved 4 bytes at the
- * beginning for the 'spl_name' and expects us to overwrite
- * its first 4 bytes) or 0x804 bytes in length.
+ * The header is always at 0x800 (as we now use a payload
+ * prepadded using the boot0 hook for all targets): the first
+ * 4 bytes of these images can safely be overwritten using the
+ * boot magic.
*/
- if (rkcommon_spl_is_boot0(params))
- tparams->header_size = RK_SPL_HDR_START;
- else
- tparams->header_size = RK_SPL_HDR_START + 4;
+ tparams->header_size = RK_SPL_HDR_START;
/* Allocate, clear and install the header */
tparams->hdr = malloc(tparams->header_size);
--
2.1.4
^ permalink raw reply related [flat|nested] 46+ messages in thread
* [U-Boot] [PATCH v5 00/18] rockchip: back-to-bootrom: replace assembly-implementation with C-code
2017-10-10 14:21 [U-Boot] [PATCH v5 00/18] rockchip: back-to-bootrom: replace assembly-implementation with C-code Philipp Tomsich
` (17 preceding siblings ...)
2017-10-10 14:21 ` [U-Boot] [PATCH v5 18/18] rockchip: mkimage: remove unused code-paths (spl_boot0 is now implied) Philipp Tomsich
@ 2017-10-11 13:07 ` Heiko Stuebner
18 siblings, 0 replies; 46+ messages in thread
From: Heiko Stuebner @ 2017-10-11 13:07 UTC (permalink / raw)
To: u-boot
Am Dienstag, 10. Oktober 2017, 16:21:00 CEST schrieb Philipp Tomsich:
>
> Recent discussions confirmed (what the code always assumed): the
> Rockchip BROM always enters U-Boot with the stack-pointer valid
> (i.e. the U-Boot startup code is running off the BROM stack).
>
> We can thus replace the back-to-bootrom code (i.e. both the
> save_boot_params and back_to_bootrom implementations) using C-code
> based on setjmp/longjmp. The new implementation is already structured
> to allow an easy drop-in of Andy's changes to enter download-mode when
> returning to the BROM.
>
> This turned out to require a some tweaking to system.h (making sure
> that the prototype for save_boot_params_ret is visible for A64)and
> start.S (so binutils knows that this is a possible function entry and
> it can correctly insert A32-to-Thumb transitions) and taking the axe
> to setjmp.h (which created quite a few issues with it not expecting
> A32/T32/Thumb call-sites and some fragility from GCC being smart about
> the clobber-list of the inline assembly... which led to r9 not being
> saved or restored).
>
> For v4+: To fix issues with the RK3188 support, this also updates the
> boot0 hook changes (i.e. allowing the boot0-hook to insert code/data
> before the ARM vector table) that Kever had submitted this spring and
> implements a similar "early back-to-bootrom" as suggested by Pawel
> (for the RK3066) as a generic mechanism.
This series on rk3188-radxarock
Tested-by: Heiko Stuebner <heiko@sntech.de>
It can enter SPL and from there also enter into proper uboot.
At first I was about to complain that it broke mid-series until I
realized that I'll of course need to adapt my image build script
to not try to use a TPL binary :-D
Heiko
^ permalink raw reply [flat|nested] 46+ messages in thread
* [U-Boot] [PATCH v5 07/18] socfpga: boot0 hook: adjust to unified boot0 semantics
2017-10-10 14:21 ` [U-Boot] [PATCH v5 07/18] socfpga: boot0 hook: adjust to unified boot0 semantics Philipp Tomsich
@ 2017-11-06 17:26 ` Dr. Philipp Tomsich
2017-11-06 17:54 ` Marek Vasut
2017-11-07 14:18 ` [U-Boot] [U-Boot, v5, " Philipp Tomsich
1 sibling, 1 reply; 46+ messages in thread
From: Dr. Philipp Tomsich @ 2017-11-06 17:26 UTC (permalink / raw)
To: u-boot
Marek & Tien Fong,
could either of you have a look at the below change (in the context of
https://patchwork.ozlabs.org/patch/823872/, which is also contained in
the same series) and give me an ok on this, so I can merge this series
early in the upcoming merge window?
Thanks,
Philipp.
> On 10 Oct 2017, at 16:21, Philipp Tomsich <philipp.tomsich@theobroma-systems.com> wrote:
>
> With the updated boot0 semantics (i.e. giving the boot0-hook control
> over when and where the vector table is emitted), the boot0-hook for
> the socfpga needs to be adjusted.
>
> Signed-off-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com>
> ---
>
> Changes in v5: None
> Changes in v4: None
> Changes in v3: None
> Changes in v2: None
>
> arch/arm/mach-socfpga/include/mach/boot0.h | 3 +++
> 1 file changed, 3 insertions(+)
>
> diff --git a/arch/arm/mach-socfpga/include/mach/boot0.h b/arch/arm/mach-socfpga/include/mach/boot0.h
> index 22d9e7f..d6b9435 100644
> --- a/arch/arm/mach-socfpga/include/mach/boot0.h
> +++ b/arch/arm/mach-socfpga/include/mach/boot0.h
> @@ -7,6 +7,9 @@
> #ifndef __BOOT0_H
> #define __BOOT0_H
>
> +_start:
> + ARM_VECTORS
> +
> #ifdef CONFIG_SPL_BUILD
> .balignl 64,0xf33db33f;
>
> --
> 2.1.4
>
^ permalink raw reply [flat|nested] 46+ messages in thread
* [U-Boot] [PATCH v5 07/18] socfpga: boot0 hook: adjust to unified boot0 semantics
2017-11-06 17:26 ` Dr. Philipp Tomsich
@ 2017-11-06 17:54 ` Marek Vasut
2017-11-06 17:55 ` Dr. Philipp Tomsich
0 siblings, 1 reply; 46+ messages in thread
From: Marek Vasut @ 2017-11-06 17:54 UTC (permalink / raw)
To: u-boot
On 11/06/2017 06:26 PM, Dr. Philipp Tomsich wrote:
> Marek & Tien Fong,
>
> could either of you have a look at the below change (in the context of
> https://patchwork.ozlabs.org/patch/823872/, which is also contained in
> the same series) and give me an ok on this, so I can merge this series
> early in the upcoming merge window?
Please stop top-posting.
Does it still generate that structure at offset 0x40 in u-boot*sfp on
SoCFPGA ? If so, fine by me.
> Thanks,
> Philipp.
>
>> On 10 Oct 2017, at 16:21, Philipp Tomsich <philipp.tomsich@theobroma-systems.com> wrote:
>>
>> With the updated boot0 semantics (i.e. giving the boot0-hook control
>> over when and where the vector table is emitted), the boot0-hook for
>> the socfpga needs to be adjusted.
>>
>> Signed-off-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com>
>> ---
>>
>> Changes in v5: None
>> Changes in v4: None
>> Changes in v3: None
>> Changes in v2: None
>>
>> arch/arm/mach-socfpga/include/mach/boot0.h | 3 +++
>> 1 file changed, 3 insertions(+)
>>
>> diff --git a/arch/arm/mach-socfpga/include/mach/boot0.h b/arch/arm/mach-socfpga/include/mach/boot0.h
>> index 22d9e7f..d6b9435 100644
>> --- a/arch/arm/mach-socfpga/include/mach/boot0.h
>> +++ b/arch/arm/mach-socfpga/include/mach/boot0.h
>> @@ -7,6 +7,9 @@
>> #ifndef __BOOT0_H
>> #define __BOOT0_H
>>
>> +_start:
>> + ARM_VECTORS
>> +
>> #ifdef CONFIG_SPL_BUILD
>> .balignl 64,0xf33db33f;
>>
>> --
>> 2.1.4
>>
>
> _______________________________________________
> U-Boot mailing list
> U-Boot at lists.denx.de
> https://lists.denx.de/listinfo/u-boot
>
--
Best regards,
Marek Vasut
^ permalink raw reply [flat|nested] 46+ messages in thread
* [U-Boot] [PATCH v5 07/18] socfpga: boot0 hook: adjust to unified boot0 semantics
2017-11-06 17:54 ` Marek Vasut
@ 2017-11-06 17:55 ` Dr. Philipp Tomsich
2017-11-06 17:57 ` Marek Vasut
0 siblings, 1 reply; 46+ messages in thread
From: Dr. Philipp Tomsich @ 2017-11-06 17:55 UTC (permalink / raw)
To: u-boot
> On 6 Nov 2017, at 18:54, Marek Vasut <marek.vasut@gmail.com> wrote:
>
> On 11/06/2017 06:26 PM, Dr. Philipp Tomsich wrote:
>> Marek & Tien Fong,
>>
>> could either of you have a look at the below change (in the context of
>> https://patchwork.ozlabs.org/patch/823872/, which is also contained in
>> the same series) and give me an ok on this, so I can merge this series
>> early in the upcoming merge window?
>
> Please stop top-posting.
>
> Does it still generate that structure at offset 0x40 in u-boot*sfp on
> SoCFPGA ? If so, fine by me.
Yes it does.
And the binary still looks the same around the magic offset.
Regards,
Philipp.
>> Thanks,
>> Philipp.
>>
>>> On 10 Oct 2017, at 16:21, Philipp Tomsich <philipp.tomsich@theobroma-systems.com> wrote:
>>>
>>> With the updated boot0 semantics (i.e. giving the boot0-hook control
>>> over when and where the vector table is emitted), the boot0-hook for
>>> the socfpga needs to be adjusted.
>>>
>>> Signed-off-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com>
>>> ---
>>>
>>> Changes in v5: None
>>> Changes in v4: None
>>> Changes in v3: None
>>> Changes in v2: None
>>>
>>> arch/arm/mach-socfpga/include/mach/boot0.h | 3 +++
>>> 1 file changed, 3 insertions(+)
>>>
>>> diff --git a/arch/arm/mach-socfpga/include/mach/boot0.h b/arch/arm/mach-socfpga/include/mach/boot0.h
>>> index 22d9e7f..d6b9435 100644
>>> --- a/arch/arm/mach-socfpga/include/mach/boot0.h
>>> +++ b/arch/arm/mach-socfpga/include/mach/boot0.h
>>> @@ -7,6 +7,9 @@
>>> #ifndef __BOOT0_H
>>> #define __BOOT0_H
>>>
>>> +_start:
>>> + ARM_VECTORS
>>> +
>>> #ifdef CONFIG_SPL_BUILD
>>> .balignl 64,0xf33db33f;
>>>
>>> --
>>> 2.1.4
>>>
>>
>> _______________________________________________
>> U-Boot mailing list
>> U-Boot at lists.denx.de
>> https://lists.denx.de/listinfo/u-boot
>>
>
>
> --
> Best regards,
> Marek Vasut
^ permalink raw reply [flat|nested] 46+ messages in thread
* [U-Boot] [PATCH v5 07/18] socfpga: boot0 hook: adjust to unified boot0 semantics
2017-11-06 17:55 ` Dr. Philipp Tomsich
@ 2017-11-06 17:57 ` Marek Vasut
2017-11-07 4:46 ` Chee, Tien Fong
0 siblings, 1 reply; 46+ messages in thread
From: Marek Vasut @ 2017-11-06 17:57 UTC (permalink / raw)
To: u-boot
On 11/06/2017 06:55 PM, Dr. Philipp Tomsich wrote:
>
>> On 6 Nov 2017, at 18:54, Marek Vasut <marek.vasut@gmail.com> wrote:
>>
>> On 11/06/2017 06:26 PM, Dr. Philipp Tomsich wrote:
>>> Marek & Tien Fong,
>>>
>>> could either of you have a look at the below change (in the context of
>>> https://patchwork.ozlabs.org/patch/823872/, which is also contained in
>>> the same series) and give me an ok on this, so I can merge this series
>>> early in the upcoming merge window?
>>
>> Please stop top-posting.
>>
>> Does it still generate that structure at offset 0x40 in u-boot*sfp on
>> SoCFPGA ? If so, fine by me.
>
> Yes it does.
> And the binary still looks the same around the magic offset.
OK, fine by me then.
--
Best regards,
Marek Vasut
^ permalink raw reply [flat|nested] 46+ messages in thread
* [U-Boot] [PATCH v5 07/18] socfpga: boot0 hook: adjust to unified boot0 semantics
2017-11-06 17:57 ` Marek Vasut
@ 2017-11-07 4:46 ` Chee, Tien Fong
0 siblings, 0 replies; 46+ messages in thread
From: Chee, Tien Fong @ 2017-11-07 4:46 UTC (permalink / raw)
To: u-boot
On Isn, 2017-11-06 at 18:57 +0100, Marek Vasut wrote:
> On 11/06/2017 06:55 PM, Dr. Philipp Tomsich wrote:
> >
> >
> > >
> > > On 6 Nov 2017, at 18:54, Marek Vasut <marek.vasut@gmail.com>
> > > wrote:
> > >
> > > On 11/06/2017 06:26 PM, Dr. Philipp Tomsich wrote:
> > > >
> > > > Marek & Tien Fong,
> > > >
> > > > could either of you have a look at the below change (in the
> > > > context of
> > > > https://patchwork.ozlabs.org/patch/823872/, which is also
> > > > contained inÂ
> > > > the same series) and give me an ok on this, so I can merge this
> > > > series
> > > > early in the upcoming merge window?
> > > Please stop top-posting.
> > >
> > > Does it still generate that structure at offset 0x40 in u-
> > > boot*sfp on
> > > SoCFPGA ? If so, fine by me.
> > Yes it does.
> > And the binary still looks the same around the magic offset.
> OK, fine by me then.
>
Fine by me too.
^ permalink raw reply [flat|nested] 46+ messages in thread
* [U-Boot] [U-Boot, v5, 01/18] arm: boot0 hook: move boot0 hook before '_start'
2017-10-10 14:21 ` [U-Boot] [PATCH v5 01/18] arm: boot0 hook: move boot0 hook before '_start' Philipp Tomsich
@ 2017-11-07 14:18 ` Philipp Tomsich
0 siblings, 0 replies; 46+ messages in thread
From: Philipp Tomsich @ 2017-11-07 14:18 UTC (permalink / raw)
To: u-boot
> The boot0 hook on ARM does not insert its payload before the vector
> table. This is both a mismatch with thec comment above it and
> contradict usage of the boot0 hook on ARM64.
>
> To fix this (and unify the semantics for ARM and ARM64), we change the
> boot0-hook semantics on ARM to match those on ARM64:
> (1) if a boot0-hook is present it is inserted at the start of
> the image
> (2) if a boot0-hook is present, emitting the ARM vector table
> (and the _start) symbol are suppressed in vectors.S and
> the boot0-hook has full control over where and when it
> wants to emit these
>
> Signed-off-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com>
> ---
>
> Changes in v5: None
> Changes in v4: None
> Changes in v3: None
> Changes in v2: None
>
> arch/arm/include/asm/arch-rockchip/boot0.h | 1 -
> arch/arm/lib/vectors.S | 54 +++++++++++++++++++-----------
> 2 files changed, 35 insertions(+), 20 deletions(-)
>
Applied to u-boot-rockchip/next, thanks!
^ permalink raw reply [flat|nested] 46+ messages in thread
* [U-Boot] [U-Boot, v5, 02/18] rockchip: boot0: align to 0x20 for armv7 '_start'
2017-10-10 14:21 ` [U-Boot] [PATCH v5 02/18] rockchip: boot0: align to 0x20 for armv7 '_start' Philipp Tomsich
@ 2017-11-07 14:18 ` Philipp Tomsich
2017-11-09 12:59 ` [U-Boot] [PATCH v5 " Andy Yan
1 sibling, 0 replies; 46+ messages in thread
From: Philipp Tomsich @ 2017-11-07 14:18 UTC (permalink / raw)
To: u-boot
> From: Kever Yang <kever.yang@rock-chips.com>
>
> The '_start' is using as vector table base address, and will write
> to VBAR register, so it needs to be aligned to 0x20 for armv7.
>
> Signed-off-by: Kever Yang <kever.yang@rock-chips.com>
> [Updated to current code base:]
> Signed-off-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com>
> ---
>
> Changes in v5: None
> Changes in v4: None
> Changes in v3: None
> Changes in v2: None
>
> arch/arm/include/asm/arch-rockchip/boot0.h | 14 ++++++++++++--
> 1 file changed, 12 insertions(+), 2 deletions(-)
>
Applied to u-boot-rockchip/next, thanks!
^ permalink raw reply [flat|nested] 46+ messages in thread
* [U-Boot] [U-Boot, v5, 03/18] rockchip: enable boot0-hook for all Rockchip SoCs
2017-10-10 14:21 ` [U-Boot] [PATCH v5 03/18] rockchip: enable boot0-hook for all Rockchip SoCs Philipp Tomsich
@ 2017-11-07 14:18 ` Philipp Tomsich
0 siblings, 0 replies; 46+ messages in thread
From: Philipp Tomsich @ 2017-11-07 14:18 UTC (permalink / raw)
To: u-boot
> Rockchip SoCs bootrom design is like this:
> - First 2KB or 4KB internal memory is for bootrom stack and heap;
> - Then the first 4-byte suppose to be a TAG like 'RK33';
> - The the following memory address end with '0004' is the first
> instruction load and running by bootrom;
>
> Let's use the boot0 hook to reserve the first 4-byte tag for all
> the Rockchip SoCs.
>
> Signed-off-by: Philipp Tomsich <philipp.tomsich@theobroma-systems>
> [Commit message:]
> Signed-off-by: Kever Yang <kever.yang@rock-chips.com>
>
> Signed-off-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com>
> ---
>
> Changes in v5: None
> Changes in v4: None
> Changes in v3: None
> Changes in v2: None
>
> arch/arm/Kconfig | 1 +
> arch/arm/mach-rockchip/Kconfig | 2 --
> 2 files changed, 1 insertion(+), 2 deletions(-)
>
Applied to u-boot-rockchip/next, thanks!
^ permalink raw reply [flat|nested] 46+ messages in thread
* [U-Boot] [U-Boot, v5, 05/18] rockchip: rk3288: use aligned address for SPL_TEXT_BASE
2017-10-10 14:21 ` [U-Boot] [PATCH v5 05/18] rockchip: rk3288: use aligned address for SPL_TEXT_BASE Philipp Tomsich
@ 2017-11-07 14:18 ` Philipp Tomsich
0 siblings, 0 replies; 46+ messages in thread
From: Philipp Tomsich @ 2017-11-07 14:18 UTC (permalink / raw)
To: u-boot
> From: Kever Yang <kever.yang@rock-chips.com>
>
> After we use boot0 hook, we can use offset '000' instead of '004' as
> SPL_TEXT_BASE.
>
> Signed-off-by: Kever Yang <kever.yang@rock-chips.com>
> [Updated tag in commit summary:]
> Signed-off-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com>
> ---
>
> Changes in v5: None
> Changes in v4: None
> Changes in v3: None
> Changes in v2: None
>
> include/configs/rk3288_common.h | 2 +-
> 1 file changed, 1 insertion(+), 1 deletion(-)
>
Applied to u-boot-rockchip/next, thanks!
^ permalink raw reply [flat|nested] 46+ messages in thread
* [U-Boot] [U-Boot, v5, 04/18] rockchip: mkimage: use spl_boot0 for all Rockchip SoCs
2017-10-10 14:21 ` [U-Boot] [PATCH v5 04/18] rockchip: mkimage: use spl_boot0 " Philipp Tomsich
@ 2017-11-07 14:18 ` Philipp Tomsich
0 siblings, 0 replies; 46+ messages in thread
From: Philipp Tomsich @ 2017-11-07 14:18 UTC (permalink / raw)
To: u-boot
> From: Kever Yang <kever.yang@rock-chips.com>
>
> Enable the spl_boot0 in SPL and use the pre-padding TAG memory,
> the mkimage do not need to pad it but only need to replace the value
> with correct TAG value.
>
> Signed-off-by: Kever Yang <kever.yang@rock-chips.com>
> [Updated:]
> Signed-off-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com>
> ---
>
> Changes in v5: None
> Changes in v4: None
> Changes in v3: None
> Changes in v2: None
>
> tools/rkcommon.c | 14 +++++++-------
> 1 file changed, 7 insertions(+), 7 deletions(-)
>
Applied to u-boot-rockchip/next, thanks!
^ permalink raw reply [flat|nested] 46+ messages in thread
* [U-Boot] [U-Boot, v5, 06/18] rockchip: rk3036: use aligned address for SPL_TEXT_BASE
2017-10-10 14:21 ` [U-Boot] [PATCH v5 06/18] rockchip: rk3036: " Philipp Tomsich
@ 2017-11-07 14:18 ` Philipp Tomsich
0 siblings, 0 replies; 46+ messages in thread
From: Philipp Tomsich @ 2017-11-07 14:18 UTC (permalink / raw)
To: u-boot
> With the boot0-hook inserting the additional padding to receive our
> SPL magic, the SPL_TEXT_BASE can be aligned again.
>
> Signed-off-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com>
> ---
>
> Changes in v5: None
> Changes in v4: None
> Changes in v3: None
> Changes in v2: None
>
> include/configs/rk3036_common.h | 2 +-
> 1 file changed, 1 insertion(+), 1 deletion(-)
>
Applied to u-boot-rockchip/next, thanks!
^ permalink raw reply [flat|nested] 46+ messages in thread
* [U-Boot] [U-Boot, v5, 07/18] socfpga: boot0 hook: adjust to unified boot0 semantics
2017-10-10 14:21 ` [U-Boot] [PATCH v5 07/18] socfpga: boot0 hook: adjust to unified boot0 semantics Philipp Tomsich
2017-11-06 17:26 ` Dr. Philipp Tomsich
@ 2017-11-07 14:18 ` Philipp Tomsich
1 sibling, 0 replies; 46+ messages in thread
From: Philipp Tomsich @ 2017-11-07 14:18 UTC (permalink / raw)
To: u-boot
> With the updated boot0 semantics (i.e. giving the boot0-hook control
> over when and where the vector table is emitted), the boot0-hook for
> the socfpga needs to be adjusted.
>
> Signed-off-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com>
> ---
>
> Changes in v5: None
> Changes in v4: None
> Changes in v3: None
> Changes in v2: None
>
> arch/arm/mach-socfpga/include/mach/boot0.h | 3 +++
> 1 file changed, 3 insertions(+)
>
Applied to u-boot-rockchip/next, thanks!
^ permalink raw reply [flat|nested] 46+ messages in thread
* [U-Boot] [U-Boot, v5, 08/18] bcm235xx: boot0 hook: adjust to unified boot0 semantics
2017-10-10 14:21 ` [U-Boot] [PATCH v5 08/18] bcm235xx: " Philipp Tomsich
@ 2017-11-07 14:18 ` Philipp Tomsich
0 siblings, 0 replies; 46+ messages in thread
From: Philipp Tomsich @ 2017-11-07 14:18 UTC (permalink / raw)
To: u-boot
> This updates the BCM235xx boot0-hook to the updated boot0 semantics
> by emitting _start and the vector table before the boot0 hook (as
> was the case before).
>
> Signed-off-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com>
>
> Version-changes: 5
> - ran 'whitespace-cleanup'
> ---
>
> Changes in v5: None
> Changes in v4: None
> Changes in v3: None
> Changes in v2: None
>
> arch/arm/include/asm/arch-bcm235xx/boot0.h | 2 ++
> 1 file changed, 2 insertions(+)
>
Applied to u-boot-rockchip/next, thanks!
^ permalink raw reply [flat|nested] 46+ messages in thread
* [U-Boot] [U-Boot, v5, 10/18] rockchip: boot0 hook: support early return for RK3188/RK3066-style BROM
2017-10-10 14:21 ` [U-Boot] [PATCH v5 10/18] rockchip: boot0 hook: support early return for RK3188/RK3066-style BROM Philipp Tomsich
@ 2017-11-07 14:18 ` Philipp Tomsich
0 siblings, 0 replies; 46+ messages in thread
From: Philipp Tomsich @ 2017-11-07 14:18 UTC (permalink / raw)
To: u-boot
> Some Rockchip BROM versions (e.g. the RK3188 and RK3066) first read 1KB data
> from NAND into SRAM and executes it. Then, following a return to bootrom, the
> BROM loads additional code to SRAM (not overwriting the first block read) and
> reenters at the same address as the first time.
>
> To support booting either a TPL (on the RK3066) or SPL (on the RK3188) using
> this model of having to count entries, this commit adds code to the boot0
> hook to track the number of entries and handle them accordingly.
>
> Signed-off-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com>
> Signed-off-by: Paweł Jarosz <paweljarosz3691@gmail.com>
> Tested-by: Andy Yan <andy.yan@rock-chips.com>
> ---
>
> Changes in v5:
> - set return value to 0 before returning to the BROM
>
> Changes in v4: None
> Changes in v3: None
> Changes in v2: None
>
> arch/arm/include/asm/arch-rockchip/boot0.h | 25 +++++++++++++++++++------
> arch/arm/mach-rockchip/Kconfig | 28 ++++++++++++++++++++++++++++
> 2 files changed, 47 insertions(+), 6 deletions(-)
>
Applied to u-boot-rockchip/next, thanks!
^ permalink raw reply [flat|nested] 46+ messages in thread
* [U-Boot] [U-Boot, v5, 09/18] bcm281xx: boot0 hook: adjust to unified boot0 semantics
2017-10-10 14:21 ` [U-Boot] [PATCH v5 09/18] bcm281xx: " Philipp Tomsich
@ 2017-11-07 14:18 ` Philipp Tomsich
0 siblings, 0 replies; 46+ messages in thread
From: Philipp Tomsich @ 2017-11-07 14:18 UTC (permalink / raw)
To: u-boot
> This updates the BCM281xx boot0-hook to the updated boot0 semantics
> by emitting _start and the vector table before the boot0 hook (as
> was the case before).
>
> Signed-off-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com>
> ---
>
> Changes in v5: None
> Changes in v4: None
> Changes in v3: None
> Changes in v2: None
>
> arch/arm/include/asm/arch-bcm281xx/boot0.h | 2 ++
> 1 file changed, 2 insertions(+)
>
Applied to u-boot-rockchip/next, thanks!
^ permalink raw reply [flat|nested] 46+ messages in thread
* [U-Boot] [U-Boot, v5, 12/18] arm: mark save_boot_params_ret as a function
2017-10-10 14:21 ` [U-Boot] [PATCH v5 12/18] arm: mark save_boot_params_ret as a function Philipp Tomsich
@ 2017-11-07 14:18 ` Philipp Tomsich
0 siblings, 0 replies; 46+ messages in thread
From: Philipp Tomsich @ 2017-11-07 14:18 UTC (permalink / raw)
To: u-boot
> As no '.type' was set for save_boot_params_ret in start.S, binutils
> did not track whether it was emitted as A32 or T32. By properly
> marking save_boot_params_ret as a potential function entry, we can
> make sure that the compiler will insert the appropriate instructions
> for branching to save_boot_params_ret both for call-sites emitted as
> A32 and T32.
>
> Reported-by: Andy Yan <andy.yan@rock-chips.com>
> Signed-off-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com>
> Tested-by: Andy Yan <andy.yan@rock-chips.com>
> ---
>
> Changes in v5: None
> Changes in v4: None
> Changes in v3:
> - tracked the root-cause why no interwork branch was emitted and fixed
> it using a '.type'-directive in start.S to mark save_boot_params_ret
> as a (possible) function-entry.
>
> Changes in v2: None
>
> arch/arm/cpu/armv7/start.S | 1 +
> 1 file changed, 1 insertion(+)
>
Applied to u-boot-rockchip/next, thanks!
^ permalink raw reply [flat|nested] 46+ messages in thread
* [U-Boot] [U-Boot, v5, 11/18] arm: make save_boot_params_ret prototype visible for AArch64
2017-10-10 14:21 ` [U-Boot] [PATCH v5 11/18] arm: make save_boot_params_ret prototype visible for AArch64 Philipp Tomsich
@ 2017-11-07 14:18 ` Philipp Tomsich
0 siblings, 0 replies; 46+ messages in thread
From: Philipp Tomsich @ 2017-11-07 14:18 UTC (permalink / raw)
To: u-boot
> The save_boot_params_ret() prototype (for those of us, that have a
> valid SP on entry and can implement save_boot_params() in C), was
> previously only defined for !defined(CONFIG_ARM64).
>
> This moves the declaration to a common block to ensure the prototype
> is available to everyone that might need it.
>
> Signed-off-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com>
> Tested-by: Andy Yan <andy.yan@rock-chips.com>
> ---
>
> Changes in v5: None
> Changes in v4: None
> Changes in v3: None
> Changes in v2: None
>
> arch/arm/include/asm/system.h | 62 +++++++++++++++++++++----------------------
> 1 file changed, 31 insertions(+), 31 deletions(-)
>
Applied to u-boot-rockchip/next, thanks!
^ permalink raw reply [flat|nested] 46+ messages in thread
* [U-Boot] [U-Boot, v5, 13/18] arm: provide a PCS-compliant setjmp implementation
2017-10-10 14:21 ` [U-Boot] [PATCH v5 13/18] arm: provide a PCS-compliant setjmp implementation Philipp Tomsich
@ 2017-11-07 14:18 ` Philipp Tomsich
0 siblings, 0 replies; 46+ messages in thread
From: Philipp Tomsich @ 2017-11-07 14:18 UTC (permalink / raw)
To: u-boot
> The previous setjmp-implementation (as a static inline function that
> contained an 'asm volatile' sequence) was extremely fragile: (some
> versions of) GCC optimised the set of registers. One critical example
> was the removal of 'r9' from the clobber list, if -ffixed-reg9 was
> supplied.
>
> To increase robustness and ensure PCS-compliant behaviour, the setjmp
> and longjmp implementation are now in assembly and closely match what
> one would expect to find in a libc implementation.
>
> Signed-off-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com>
> Tested-by: Andy Yan <andy.yan@rock-chips.com>
> ---
>
> Changes in v5: None
> Changes in v4: None
> Changes in v3:
> - converted setjmp/longjmp from inline-assembly to separate .S files
> to improve predicatability if emitted code
>
> Changes in v2: None
>
> arch/arm/include/asm/setjmp.h | 94 ++++++-------------------------------------
> arch/arm/lib/Makefile | 6 +++
> arch/arm/lib/setjmp.S | 37 +++++++++++++++++
> arch/arm/lib/setjmp_aarch64.S | 42 +++++++++++++++++++
> 4 files changed, 98 insertions(+), 81 deletions(-)
> create mode 100644 arch/arm/lib/setjmp.S
> create mode 100644 arch/arm/lib/setjmp_aarch64.S
>
Applied to u-boot-rockchip/next, thanks!
^ permalink raw reply [flat|nested] 46+ messages in thread
* [U-Boot] [U-Boot, v5, 15/18] rockchip: rk3188: use boot0 hook to load up SPL in 2 steps
2017-10-10 14:21 ` [U-Boot] [PATCH v5 15/18] rockchip: rk3188: use boot0 hook to load up SPL in 2 steps Philipp Tomsich
@ 2017-11-07 14:18 ` Philipp Tomsich
0 siblings, 0 replies; 46+ messages in thread
From: Philipp Tomsich @ 2017-11-07 14:18 UTC (permalink / raw)
To: u-boot
> For the RK3188, the BROM will attempt to load up the first stage
> image (SPL for the RK3188) in two steps: first 1KB to offset 0x800
> in the SRAM and then the remainder to offset 0xc00 in the SRAM.
> It always enters at 0x804, though.
>
> With this changeset, the RK3188 boot removes the TPL (stub) stage and
> builds a single SPL binary that utilizes the early back-to-bootrom via
> the boot0-hook.
>
> Consequently, the passing of the saved boot params via pmu->os_reg[2]
> is also removed.
>
> Signed-off-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com>
> ---
>
> Changes in v5: None
> Changes in v4:
> - after merging the 'back-to-bootrom' series with the 'boot0-hook'
> series, this drops the TPL stub and builds only a single SPL image
> that uses the 'early back-to-bootrom' logic originally implemented
> by Pawel for the RK3066.
> - changes the SPL_STACK_BASE to +0x800 (from +0x804), as the boot0
> hook already reserves the space for the SPL magic (previously
> inserted by mkimage)
>
> Changes in v3: None
> Changes in v2:
> - [added in v2] chain back_to_bootrom calls for SPL, first returning
> to the TPL (using the same mechanism) and the to the BROM from the
> TPL
>
> arch/arm/mach-rockchip/Kconfig | 2 +-
> arch/arm/mach-rockchip/Makefile | 1 -
> arch/arm/mach-rockchip/rk3188-board-spl.c | 10 ----
> arch/arm/mach-rockchip/rk3188-board-tpl.c | 86 -------------------------------
> doc/README.rockchip | 10 ++--
> include/configs/rk3188_common.h | 12 ++---
> 6 files changed, 9 insertions(+), 112 deletions(-)
> delete mode 100644 arch/arm/mach-rockchip/rk3188-board-tpl.c
>
Applied to u-boot-rockchip/next, thanks!
^ permalink raw reply [flat|nested] 46+ messages in thread
* [U-Boot] [U-Boot, v5, 14/18] rockchip: back-to-bootrom: replace assembly-implementation with C-code
2017-10-10 14:21 ` [U-Boot] [PATCH v5 14/18] rockchip: back-to-bootrom: replace assembly-implementation with C-code Philipp Tomsich
@ 2017-11-07 14:18 ` Philipp Tomsich
0 siblings, 0 replies; 46+ messages in thread
From: Philipp Tomsich @ 2017-11-07 14:18 UTC (permalink / raw)
To: u-boot
> The back-to-bootrom implementation for Rockchip has always relied on
> the stack-pointer being valid on entry, so there was little reason to
> have this as an assembly implementation.
>
> This provides a new C-only implementation of save_boot_params and
> back_to_bootrom (relying on setjmp/longjmp) and removes the older
> assembly-only implementation.
>
> Signed-off-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com>
> Tested-by: Andy Yan <andy.yan@rock-chips.com>
> ---
>
> Changes in v5: None
> Changes in v4: None
> Changes in v3: None
> Changes in v2: None
>
> arch/arm/include/asm/arch-rockchip/bootrom.h | 27 ++++++++---
> arch/arm/mach-rockchip/Makefile | 4 +-
> arch/arm/mach-rockchip/bootrom.c | 52 ++++++++++++++++++++-
> arch/arm/mach-rockchip/save_boot_param.S | 69 ----------------------------
> 4 files changed, 73 insertions(+), 79 deletions(-)
> delete mode 100644 arch/arm/mach-rockchip/save_boot_param.S
>
Applied to u-boot-rockchip/next, thanks!
^ permalink raw reply [flat|nested] 46+ messages in thread
* [U-Boot] [U-Boot, v5, 18/18] rockchip: mkimage: remove unused code-paths (spl_boot0 is now implied)
2017-10-10 14:21 ` [U-Boot] [PATCH v5 18/18] rockchip: mkimage: remove unused code-paths (spl_boot0 is now implied) Philipp Tomsich
@ 2017-11-07 14:18 ` Philipp Tomsich
0 siblings, 0 replies; 46+ messages in thread
From: Philipp Tomsich @ 2017-11-07 14:18 UTC (permalink / raw)
To: u-boot
> With all targets converted to generate prepadded images, this removes
> the spl_boot0 field from our config structure and removes the unused
> code-path (for images that are not prepadded): i.e. spl_boot0 is now
> implied as 'true' and the code is specialised by removing the other
> case.
>
> Signed-off-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com>
> ---
>
> Changes in v5: None
> Changes in v4: None
> Changes in v3: None
> Changes in v2: None
>
> tools/rkcommon.c | 45 ++++++++++++++-------------------------------
> 1 file changed, 14 insertions(+), 31 deletions(-)
>
Applied to u-boot-rockchip/next, thanks!
^ permalink raw reply [flat|nested] 46+ messages in thread
* [U-Boot] [U-Boot, v5, 16/18] rockchip: back-to-bootrom: allow passing a cmd to the bootrom
2017-10-10 14:21 ` [U-Boot] [PATCH v5 16/18] rockchip: back-to-bootrom: allow passing a cmd to the bootrom Philipp Tomsich
@ 2017-11-07 14:18 ` Philipp Tomsich
0 siblings, 0 replies; 46+ messages in thread
From: Philipp Tomsich @ 2017-11-07 14:18 UTC (permalink / raw)
To: u-boot
> The BROM supports forcing it to enter download-mode, if an appropriate
> result/cmd-word is returned to it. There already is a series to
> support this in review, so this prepares the (newly C-version) of the
> back-to-bootrom code to accept a cmd to passed on to the BROM.
>
> All the existing call-sites are adjusted to match the changed function
> signature.
>
> Signed-off-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com>
> Tested-by: Andy Yan <andy.yan@rock-chips.com>
> ---
>
> Changes in v5: None
> Changes in v4:
> - no longer updates rk3188-board-tpl.c (as we have just removed it
> in an earlier commit)
>
> Changes in v3: None
> Changes in v2:
> - also covers the RK3188 (which I had originally missed)
>
> arch/arm/include/asm/arch-rockchip/bootrom.h | 5 ++++-
> arch/arm/mach-rockchip/bootrom.c | 4 ++--
> arch/arm/mach-rockchip/rk3036-board-spl.c | 2 +-
> arch/arm/mach-rockchip/rk3188-board-spl.c | 4 ++--
> arch/arm/mach-rockchip/rk322x-board-spl.c | 2 +-
> arch/arm/mach-rockchip/rk3288-board-spl.c | 4 ++--
> arch/arm/mach-rockchip/rk3288-board-tpl.c | 2 +-
> arch/arm/mach-rockchip/rk3368-board-tpl.c | 2 +-
> arch/arm/mach-rockchip/rk3399-board-spl.c | 2 +-
> 9 files changed, 15 insertions(+), 12 deletions(-)
>
Applied to u-boot-rockchip/next, thanks!
^ permalink raw reply [flat|nested] 46+ messages in thread
* [U-Boot] [U-Boot, v5, 17/18] rockchip: rk3188: move CONFIG_SPL_* entries from rk3188_common.h to Kconfig
2017-10-10 14:21 ` [U-Boot] [PATCH v5 17/18] rockchip: rk3188: move CONFIG_SPL_* entries from rk3188_common.h to Kconfig Philipp Tomsich
@ 2017-11-07 14:18 ` Philipp Tomsich
0 siblings, 0 replies; 46+ messages in thread
From: Philipp Tomsich @ 2017-11-07 14:18 UTC (permalink / raw)
To: u-boot
> There still are a few CONFIG_SPL_* options selected using defines from
> rk3188_common.h instead of via Kconfig. This migrates those over to
> Kconfig.
>
> Signed-off-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com>
> ---
>
> Changes in v5: None
> Changes in v4: None
> Changes in v3: None
> Changes in v2: None
>
> arch/arm/mach-rockchip/Kconfig | 7 ++++++-
> include/configs/rk3188_common.h | 6 ------
> 2 files changed, 6 insertions(+), 7 deletions(-)
>
Applied to u-boot-rockchip/next, thanks!
^ permalink raw reply [flat|nested] 46+ messages in thread
* [U-Boot] [PATCH v5 02/18] rockchip: boot0: align to 0x20 for armv7 '_start'
2017-10-10 14:21 ` [U-Boot] [PATCH v5 02/18] rockchip: boot0: align to 0x20 for armv7 '_start' Philipp Tomsich
2017-11-07 14:18 ` [U-Boot] [U-Boot, v5, " Philipp Tomsich
@ 2017-11-09 12:59 ` Andy Yan
2017-11-09 13:03 ` Dr. Philipp Tomsich
1 sibling, 1 reply; 46+ messages in thread
From: Andy Yan @ 2017-11-09 12:59 UTC (permalink / raw)
To: u-boot
Hi Phipipp, Kever:
On 2017年10月10日 22:21, Philipp Tomsich wrote:
> From: Kever Yang <kever.yang@rock-chips.com>
>
> The '_start' is using as vector table base address, and will write
> to VBAR register, so it needs to be aligned to 0x20 for armv7.
>
> Signed-off-by: Kever Yang <kever.yang@rock-chips.com>
> [Updated to current code base:]
> Signed-off-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com>
>
> ---
>
> Changes in v5: None
> Changes in v4: None
> Changes in v3: None
> Changes in v2: None
>
> arch/arm/include/asm/arch-rockchip/boot0.h | 14 ++++++++++++--
> 1 file changed, 12 insertions(+), 2 deletions(-)
>
> diff --git a/arch/arm/include/asm/arch-rockchip/boot0.h b/arch/arm/include/asm/arch-rockchip/boot0.h
> index 455d842..f7c6146 100644
> --- a/arch/arm/include/asm/arch-rockchip/boot0.h
> +++ b/arch/arm/include/asm/arch-rockchip/boot0.h
> @@ -6,12 +6,13 @@
>
> /*
> * Execution starts on the instruction following this 4-byte header
> - * (containing the magic 'RK33').
> + * (containing the magic 'RK30', 'RK31', 'RK32' or 'RK33'). This
> + * magic constant will be written into the final image by the rkimage
> + * tool, but we need to reserve space for it here.
> *
> * To make life easier for everyone, we build the SPL binary with
> * space for this 4-byte header already included in the binary.
> */
> -
> #ifdef CONFIG_SPL_BUILD
> /*
> * We need to add 4 bytes of space for the 'RK33' at the
> @@ -26,6 +27,15 @@
> b reset /* may be overwritten --- should be 'nop' or a 'b reset' */
> #endif
> b reset
Do we really need the "b reset" here? the macro ARM_VECTORS already
has a b reset.
Besides Joseph found that the irq function will not work with this "b
reset"
> +#if !defined(CONFIG_ARM64)
> + /*
> + * For armv7, the addr '_start' will used as vector start address
> + * and write to VBAR register, which needs to aligned to 0x20.
> + */
> + .align(5)
> +_start:
> + ARM_VECTORS
> +#endif
>
> #if defined(CONFIG_ROCKCHIP_RK3399) && defined(CONFIG_SPL_BUILD)
> .space CONFIG_ROCKCHIP_SPL_RESERVE_IRAM /* space for the ATF data */
^ permalink raw reply [flat|nested] 46+ messages in thread
* [U-Boot] [PATCH v5 02/18] rockchip: boot0: align to 0x20 for armv7 '_start'
2017-11-09 12:59 ` [U-Boot] [PATCH v5 " Andy Yan
@ 2017-11-09 13:03 ` Dr. Philipp Tomsich
2017-11-10 3:43 ` Kever Yang
0 siblings, 1 reply; 46+ messages in thread
From: Dr. Philipp Tomsich @ 2017-11-09 13:03 UTC (permalink / raw)
To: u-boot
Andy,
On 9 Nov 2017, at 13:59, Andy Yan <andy.yan@rock-chips.com> wrote:
>
> Hi Phipipp, Kever:
>
>
> On 2017年10月10日 22:21, Philipp Tomsich wrote:
>> From: Kever Yang <kever.yang@rock-chips.com>
>>
>> The '_start' is using as vector table base address, and will write
>> to VBAR register, so it needs to be aligned to 0x20 for armv7.
>>
>> Signed-off-by: Kever Yang <kever.yang@rock-chips.com>
>> [Updated to current code base:]
>> Signed-off-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com>
>>
>> ---
>>
>> Changes in v5: None
>> Changes in v4: None
>> Changes in v3: None
>> Changes in v2: None
>>
>> arch/arm/include/asm/arch-rockchip/boot0.h | 14 ++++++++++++--
>> 1 file changed, 12 insertions(+), 2 deletions(-)
>>
>> diff --git a/arch/arm/include/asm/arch-rockchip/boot0.h b/arch/arm/include/asm/arch-rockchip/boot0.h
>> index 455d842..f7c6146 100644
>> --- a/arch/arm/include/asm/arch-rockchip/boot0.h
>> +++ b/arch/arm/include/asm/arch-rockchip/boot0.h
>> @@ -6,12 +6,13 @@
>> /*
>> * Execution starts on the instruction following this 4-byte header
>> - * (containing the magic 'RK33').
>> + * (containing the magic 'RK30', 'RK31', 'RK32' or 'RK33'). This
>> + * magic constant will be written into the final image by the rkimage
>> + * tool, but we need to reserve space for it here.
>> *
>> * To make life easier for everyone, we build the SPL binary with
>> * space for this 4-byte header already included in the binary.
>> */
>> -
>> #ifdef CONFIG_SPL_BUILD
>> /*
>> * We need to add 4 bytes of space for the 'RK33' at the
>> @@ -26,6 +27,15 @@
>> b reset /* may be overwritten --- should be 'nop' or a 'b reset' */
>> #endif
>> b reset
>
> Do we really need the "b reset" here? the macro ARM_VECTORS already has a b reset.
> Besides Joseph found that the irq function will not work with this "b reset”
The quoted code was from an older version of the series, but the ‘b reset’ is still there
in the newer version.
What IRQ function does not work?
Shouldn't the IRQ be jumping into the ARM_VECTORS and never see the ‘b reset’?
>> +#if !defined(CONFIG_ARM64)
>> + /*
>> + * For armv7, the addr '_start' will used as vector start address
>> + * and write to VBAR register, which needs to aligned to 0x20.
>> + */
>> + .align(5)
>> +_start:
>> + ARM_VECTORS
>> +#endif
>> #if defined(CONFIG_ROCKCHIP_RK3399) && defined(CONFIG_SPL_BUILD)
>> .space CONFIG_ROCKCHIP_SPL_RESERVE_IRAM /* space for the ATF data */
>
>
^ permalink raw reply [flat|nested] 46+ messages in thread
* [U-Boot] [PATCH v5 02/18] rockchip: boot0: align to 0x20 for armv7 '_start'
2017-11-09 13:03 ` Dr. Philipp Tomsich
@ 2017-11-10 3:43 ` Kever Yang
0 siblings, 0 replies; 46+ messages in thread
From: Kever Yang @ 2017-11-10 3:43 UTC (permalink / raw)
To: u-boot
Andy,
On 11/09/2017 05:03 AM, Dr. Philipp Tomsich wrote:
> Andy,
>
> On 9 Nov 2017, at 13:59, Andy Yan <andy.yan@rock-chips.com> wrote:
>> Hi Phipipp, Kever:
>>
>>
>> On 2017年10月10日 22:21, Philipp Tomsich wrote:
>>> From: Kever Yang <kever.yang@rock-chips.com>
>>>
>>> The '_start' is using as vector table base address, and will write
>>> to VBAR register, so it needs to be aligned to 0x20 for armv7.
>>>
>>> Signed-off-by: Kever Yang <kever.yang@rock-chips.com>
>>> [Updated to current code base:]
>>> Signed-off-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com>
>>>
>>> ---
>>>
>>> Changes in v5: None
>>> Changes in v4: None
>>> Changes in v3: None
>>> Changes in v2: None
>>>
>>> arch/arm/include/asm/arch-rockchip/boot0.h | 14 ++++++++++++--
>>> 1 file changed, 12 insertions(+), 2 deletions(-)
>>>
>>> diff --git a/arch/arm/include/asm/arch-rockchip/boot0.h b/arch/arm/include/asm/arch-rockchip/boot0.h
>>> index 455d842..f7c6146 100644
>>> --- a/arch/arm/include/asm/arch-rockchip/boot0.h
>>> +++ b/arch/arm/include/asm/arch-rockchip/boot0.h
>>> @@ -6,12 +6,13 @@
>>> /*
>>> * Execution starts on the instruction following this 4-byte header
>>> - * (containing the magic 'RK33').
>>> + * (containing the magic 'RK30', 'RK31', 'RK32' or 'RK33'). This
>>> + * magic constant will be written into the final image by the rkimage
>>> + * tool, but we need to reserve space for it here.
>>> *
>>> * To make life easier for everyone, we build the SPL binary with
>>> * space for this 4-byte header already included in the binary.
>>> */
>>> -
>>> #ifdef CONFIG_SPL_BUILD
>>> /*
>>> * We need to add 4 bytes of space for the 'RK33' at the
>>> @@ -26,6 +27,15 @@
>>> b reset /* may be overwritten --- should be 'nop' or a 'b reset' */
>>> #endif
>>> b reset
>> Do we really need the "b reset" here? the macro ARM_VECTORS already has a b reset.
>> Besides Joseph found that the irq function will not work with this "b reset”
> The quoted code was from an older version of the series, but the ‘b reset’ is still there
> in the newer version.
>
> What IRQ function does not work?
> Shouldn't the IRQ be jumping into the ARM_VECTORS and never see the ‘b reset’?
We need the 'b reset' here, this is not present at the same time with
ARM_VECTORS,
I think what we need to handle is that we don't need boot0_hook in U-Boot,
let me send a patch to fix this.
Thanks,
- Kever
>
>>> +#if !defined(CONFIG_ARM64)
>>> + /*
>>> + * For armv7, the addr '_start' will used as vector start address
>>> + * and write to VBAR register, which needs to aligned to 0x20.
>>> + */
>>> + .align(5)
>>> +_start:
>>> + ARM_VECTORS
>>> +#endif
>>> #if defined(CONFIG_ROCKCHIP_RK3399) && defined(CONFIG_SPL_BUILD)
>>> .space CONFIG_ROCKCHIP_SPL_RESERVE_IRAM /* space for the ATF data */
>>
>
^ permalink raw reply [flat|nested] 46+ messages in thread
end of thread, other threads:[~2017-11-10 3:43 UTC | newest]
Thread overview: 46+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2017-10-10 14:21 [U-Boot] [PATCH v5 00/18] rockchip: back-to-bootrom: replace assembly-implementation with C-code Philipp Tomsich
2017-10-10 14:21 ` [U-Boot] [PATCH v5 01/18] arm: boot0 hook: move boot0 hook before '_start' Philipp Tomsich
2017-11-07 14:18 ` [U-Boot] [U-Boot, v5, " Philipp Tomsich
2017-10-10 14:21 ` [U-Boot] [PATCH v5 02/18] rockchip: boot0: align to 0x20 for armv7 '_start' Philipp Tomsich
2017-11-07 14:18 ` [U-Boot] [U-Boot, v5, " Philipp Tomsich
2017-11-09 12:59 ` [U-Boot] [PATCH v5 " Andy Yan
2017-11-09 13:03 ` Dr. Philipp Tomsich
2017-11-10 3:43 ` Kever Yang
2017-10-10 14:21 ` [U-Boot] [PATCH v5 03/18] rockchip: enable boot0-hook for all Rockchip SoCs Philipp Tomsich
2017-11-07 14:18 ` [U-Boot] [U-Boot, v5, " Philipp Tomsich
2017-10-10 14:21 ` [U-Boot] [PATCH v5 04/18] rockchip: mkimage: use spl_boot0 " Philipp Tomsich
2017-11-07 14:18 ` [U-Boot] [U-Boot, v5, " Philipp Tomsich
2017-10-10 14:21 ` [U-Boot] [PATCH v5 05/18] rockchip: rk3288: use aligned address for SPL_TEXT_BASE Philipp Tomsich
2017-11-07 14:18 ` [U-Boot] [U-Boot, v5, " Philipp Tomsich
2017-10-10 14:21 ` [U-Boot] [PATCH v5 06/18] rockchip: rk3036: " Philipp Tomsich
2017-11-07 14:18 ` [U-Boot] [U-Boot, v5, " Philipp Tomsich
2017-10-10 14:21 ` [U-Boot] [PATCH v5 07/18] socfpga: boot0 hook: adjust to unified boot0 semantics Philipp Tomsich
2017-11-06 17:26 ` Dr. Philipp Tomsich
2017-11-06 17:54 ` Marek Vasut
2017-11-06 17:55 ` Dr. Philipp Tomsich
2017-11-06 17:57 ` Marek Vasut
2017-11-07 4:46 ` Chee, Tien Fong
2017-11-07 14:18 ` [U-Boot] [U-Boot, v5, " Philipp Tomsich
2017-10-10 14:21 ` [U-Boot] [PATCH v5 08/18] bcm235xx: " Philipp Tomsich
2017-11-07 14:18 ` [U-Boot] [U-Boot, v5, " Philipp Tomsich
2017-10-10 14:21 ` [U-Boot] [PATCH v5 09/18] bcm281xx: " Philipp Tomsich
2017-11-07 14:18 ` [U-Boot] [U-Boot, v5, " Philipp Tomsich
2017-10-10 14:21 ` [U-Boot] [PATCH v5 10/18] rockchip: boot0 hook: support early return for RK3188/RK3066-style BROM Philipp Tomsich
2017-11-07 14:18 ` [U-Boot] [U-Boot, v5, " Philipp Tomsich
2017-10-10 14:21 ` [U-Boot] [PATCH v5 11/18] arm: make save_boot_params_ret prototype visible for AArch64 Philipp Tomsich
2017-11-07 14:18 ` [U-Boot] [U-Boot, v5, " Philipp Tomsich
2017-10-10 14:21 ` [U-Boot] [PATCH v5 12/18] arm: mark save_boot_params_ret as a function Philipp Tomsich
2017-11-07 14:18 ` [U-Boot] [U-Boot, v5, " Philipp Tomsich
2017-10-10 14:21 ` [U-Boot] [PATCH v5 13/18] arm: provide a PCS-compliant setjmp implementation Philipp Tomsich
2017-11-07 14:18 ` [U-Boot] [U-Boot, v5, " Philipp Tomsich
2017-10-10 14:21 ` [U-Boot] [PATCH v5 14/18] rockchip: back-to-bootrom: replace assembly-implementation with C-code Philipp Tomsich
2017-11-07 14:18 ` [U-Boot] [U-Boot, v5, " Philipp Tomsich
2017-10-10 14:21 ` [U-Boot] [PATCH v5 15/18] rockchip: rk3188: use boot0 hook to load up SPL in 2 steps Philipp Tomsich
2017-11-07 14:18 ` [U-Boot] [U-Boot, v5, " Philipp Tomsich
2017-10-10 14:21 ` [U-Boot] [PATCH v5 16/18] rockchip: back-to-bootrom: allow passing a cmd to the bootrom Philipp Tomsich
2017-11-07 14:18 ` [U-Boot] [U-Boot, v5, " Philipp Tomsich
2017-10-10 14:21 ` [U-Boot] [PATCH v5 17/18] rockchip: rk3188: move CONFIG_SPL_* entries from rk3188_common.h to Kconfig Philipp Tomsich
2017-11-07 14:18 ` [U-Boot] [U-Boot, v5, " Philipp Tomsich
2017-10-10 14:21 ` [U-Boot] [PATCH v5 18/18] rockchip: mkimage: remove unused code-paths (spl_boot0 is now implied) Philipp Tomsich
2017-11-07 14:18 ` [U-Boot] [U-Boot, v5, " Philipp Tomsich
2017-10-11 13:07 ` [U-Boot] [PATCH v5 00/18] rockchip: back-to-bootrom: replace assembly-implementation with C-code Heiko Stuebner
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.