Mirrors/u-boot
2
0
Fork 0
mirror of https://github.com/AsahiLinux/u-boot synced 2024-12-02 01:19:49 +00:00
Code Issues Projects Releases Packages Wiki Activity
u-boot/arch/arc/lib/start.S

116 lines
2.7 KiB
ArmAsm
Raw Normal View History

SPDX: Convert all of our single license tags to Linux Kernel style When U-Boot started using SPDX tags we were among the early adopters and there weren't a lot of other examples to borrow from. So we picked the area of the file that usually had a full license text and replaced it with an appropriate SPDX-License-Identifier: entry. Since then, the Linux Kernel has adopted SPDX tags and they place it as the very first line in a file (except where shebangs are used, then it's second line) and with slightly different comment styles than us. In part due to community overlap, in part due to better tag visibility and in part for other minor reasons, switch over to that style. This commit changes all instances where we have a single declared license in the tag as both the before and after are identical in tag contents. There's also a few places where I found we did not have a tag and have introduced one. Signed-off-by: Tom Rini <trini@konsulko.com>
2018-05-06 21:58:06 +00:00
/* SPDX-License-Identifier: GPL-2.0+ */
arc: add cpu files Signed-off-by: Alexey Brodkin <abrodkin@synopsys.com> Cc: Vineet Gupta <vgupta@synopsys.com> Cc: Francois Bedard <fbedard@synopsys.com> Cc: Wolfgang Denk <wd@denx.de> Cc: Heiko Schocher <hs@denx.de>
2014-02-04 08:56:14 +00:00
/*
* Copyright (C) 2013-2014 Synopsys, Inc. All rights reserved.
*/
#include <asm-offsets.h>
#include <config.h>
arc: merge common start-up code between ARC and ARCv2 Even though ARCompact and ARCv2 are not binary compatible most of assembly instructions are used in both. With this change we'll get rid of duplicate code. Still IVTs are implemented differently so we're keeping them in separate files. Signed-off-by: Alexey Brodkin <abrodkin@synopsys.com>
2015-02-19 15:40:58 +00:00
#include <linux/linkage.h>
arc: add cpu files Signed-off-by: Alexey Brodkin <abrodkin@synopsys.com> Cc: Vineet Gupta <vgupta@synopsys.com> Cc: Francois Bedard <fbedard@synopsys.com> Cc: Wolfgang Denk <wd@denx.de> Cc: Heiko Schocher <hs@denx.de>
2014-02-04 08:56:14 +00:00
#include <asm/arcregs.h>
arc: merge common start-up code between ARC and ARCv2 Even though ARCompact and ARCv2 are not binary compatible most of assembly instructions are used in both. With this change we'll get rid of duplicate code. Still IVTs are implemented differently so we're keeping them in separate files. Signed-off-by: Alexey Brodkin <abrodkin@synopsys.com>
2015-02-19 15:40:58 +00:00
ENTRY(_start)
arc: introduce separate section for interrupt vector table Even though existing implementation works fine in preparation to submission of ARCv2 architecture we need this change. In case of ARCv2 interrupt vector table consists of just addresses of corresponding handlers. And if those addresses will be in .text section then assembler will encode them as everything in .text section as middle-endian and then on real execution CPU will read swapped addresses and will jump into the wild. Once introduced new section is situated so .text section remains the first which allows us to use common linker option for linking everything to a specified CONFIG_SYS_TEXT_BASE. Signed-off-by: Alexey Brodkin <abrodkin@synopsys.com> Signed-off-by: Igor Guryanov <guryanov@synopsys.com>
2014-12-24 14:17:11 +00:00
/* Setup interrupt vector base that matches "__text_start" */
sr __ivt_start, [ARC_AUX_INTR_VEC_BASE]
arc: significant cache rework [1] Align cache management functions to those in Linux kernel. I.e.: a) Use the same functions for all cache ops (D$ Inv/Flush) b) Split cache ops in 3 sub-functions: "before", "lineloop" and "after". That way we may re-use "before" and "after" functions for region and full cache ops. [2] Implement full-functional L2 (SLC) management. Before SLC was simply disabled early on boot. It's also possible to enable or disable L2 cache from config utility. [3] Disable/enable corresponding caches early on boot. So if U-Boot is configured to use caches they will be used at all times (this is useful in partucular for speed-up of relocation). Signed-off-by: Alexey Brodkin <abrodkin@synopsys.com>
2015-05-18 13:56:26 +00:00
; Disable/enable I-cache according to configuration
lr r5, [ARC_BCR_IC_BUILD]
breq r5, 0, 1f ; I$ doesn't exist
lr r5, [ARC_AUX_IC_CTRL]
#ifndef CONFIG_SYS_ICACHE_OFF
bclr r5, r5, 0 ; 0 - Enable, 1 is Disable
#else
bset r5, r5, 0 ; I$ exists, but is not used
#endif
sr r5, [ARC_AUX_IC_CTRL]
ARC: Invalidate instruction and data caches early on boot This is useful to make sure no stale data exists in caches after bootloaders. The worst thing could be some lines of cache were locked in a bootloader for example during DDR recalibration and never unlocked. This may lead to really unpredictable issues later down the line. Signed-off-by: Eugeniy Paltsev <Eugeniy.Paltsev@synopsys.com> Signed-off-by: Alexey Brodkin <abrodkin@synopsys.com>
2018-01-16 18:52:25 +00:00
mov r5, 1
sr r5, [ARC_AUX_IC_IVIC]
; As per ARC HS databook (see chapter 5.3.3.2)
; it is required to add 3 NOPs after each write to IC_IVIC.
nop
nop
nop
arc: significant cache rework [1] Align cache management functions to those in Linux kernel. I.e.: a) Use the same functions for all cache ops (D$ Inv/Flush) b) Split cache ops in 3 sub-functions: "before", "lineloop" and "after". That way we may re-use "before" and "after" functions for region and full cache ops. [2] Implement full-functional L2 (SLC) management. Before SLC was simply disabled early on boot. It's also possible to enable or disable L2 cache from config utility. [3] Disable/enable corresponding caches early on boot. So if U-Boot is configured to use caches they will be used at all times (this is useful in partucular for speed-up of relocation). Signed-off-by: Alexey Brodkin <abrodkin@synopsys.com>
2015-05-18 13:56:26 +00:00
1:
; Disable/enable D-cache according to configuration
lr r5, [ARC_BCR_DC_BUILD]
breq r5, 0, 1f ; D$ doesn't exist
lr r5, [ARC_AUX_DC_CTRL]
bclr r5, r5, 6 ; Invalidate (discard w/o wback)
#ifndef CONFIG_SYS_DCACHE_OFF
bclr r5, r5, 0 ; Enable (+Inv)
#else
bset r5, r5, 0 ; Disable (+Inv)
#endif
sr r5, [ARC_AUX_DC_CTRL]
arc: introduce separate section for interrupt vector table Even though existing implementation works fine in preparation to submission of ARCv2 architecture we need this change. In case of ARCv2 interrupt vector table consists of just addresses of corresponding handlers. And if those addresses will be in .text section then assembler will encode them as everything in .text section as middle-endian and then on real execution CPU will read swapped addresses and will jump into the wild. Once introduced new section is situated so .text section remains the first which allows us to use common linker option for linking everything to a specified CONFIG_SYS_TEXT_BASE. Signed-off-by: Alexey Brodkin <abrodkin@synopsys.com> Signed-off-by: Igor Guryanov <guryanov@synopsys.com>
2014-12-24 14:17:11 +00:00
ARC: Invalidate instruction and data caches early on boot This is useful to make sure no stale data exists in caches after bootloaders. The worst thing could be some lines of cache were locked in a bootloader for example during DDR recalibration and never unlocked. This may lead to really unpredictable issues later down the line. Signed-off-by: Eugeniy Paltsev <Eugeniy.Paltsev@synopsys.com> Signed-off-by: Alexey Brodkin <abrodkin@synopsys.com>
2018-01-16 18:52:25 +00:00
mov r5, 1
sr r5, [ARC_AUX_DC_IVDC]
arc: significant cache rework [1] Align cache management functions to those in Linux kernel. I.e.: a) Use the same functions for all cache ops (D$ Inv/Flush) b) Split cache ops in 3 sub-functions: "before", "lineloop" and "after". That way we may re-use "before" and "after" functions for region and full cache ops. [2] Implement full-functional L2 (SLC) management. Before SLC was simply disabled early on boot. It's also possible to enable or disable L2 cache from config utility. [3] Disable/enable corresponding caches early on boot. So if U-Boot is configured to use caches they will be used at all times (this is useful in partucular for speed-up of relocation). Signed-off-by: Alexey Brodkin <abrodkin@synopsys.com>
2015-05-18 13:56:26 +00:00
1:
arc: add support for SLC (System Level Cache, AKA L2-cache) ARCv2 cores may have built-in SLC (System Level Cache, AKA L2-cache). This change adds functions required for controlling SLC: * slc_enable/disable * slc_flush/invalidate For now we just disable SLC to escape DMA coherency issues until either: * SLC flush/invalidate is supported in DMA APIin U-Boot * hardware DMA coherency is implemented (that might be board specific so probably we'll need to have a separate Kconfig option for controlling SLC explicitly) Signed-off-by: Alexey Brodkin <abrodkin@synopsys.com>
2015-03-30 10:36:04 +00:00
#ifdef CONFIG_ISA_ARCV2
arc: significant cache rework [1] Align cache management functions to those in Linux kernel. I.e.: a) Use the same functions for all cache ops (D$ Inv/Flush) b) Split cache ops in 3 sub-functions: "before", "lineloop" and "after". That way we may re-use "before" and "after" functions for region and full cache ops. [2] Implement full-functional L2 (SLC) management. Before SLC was simply disabled early on boot. It's also possible to enable or disable L2 cache from config utility. [3] Disable/enable corresponding caches early on boot. So if U-Boot is configured to use caches they will be used at all times (this is useful in partucular for speed-up of relocation). Signed-off-by: Alexey Brodkin <abrodkin@synopsys.com>
2015-05-18 13:56:26 +00:00
; Disable System-Level Cache (SLC)
lr r5, [ARC_BCR_SLC]
breq r5, 0, 1f ; SLC doesn't exist
lr r5, [ARC_AUX_SLC_CTRL]
bclr r5, r5, 6 ; Invalidate (discard w/o wback)
bclr r5, r5, 0 ; Enable (+Inv)
sr r5, [ARC_AUX_SLC_CTRL]
1:
arc: add support for SLC (System Level Cache, AKA L2-cache) ARCv2 cores may have built-in SLC (System Level Cache, AKA L2-cache). This change adds functions required for controlling SLC: * slc_enable/disable * slc_flush/invalidate For now we just disable SLC to escape DMA coherency issues until either: * SLC flush/invalidate is supported in DMA APIin U-Boot * hardware DMA coherency is implemented (that might be board specific so probably we'll need to have a separate Kconfig option for controlling SLC explicitly) Signed-off-by: Alexey Brodkin <abrodkin@synopsys.com>
2015-03-30 10:36:04 +00:00
#endif
arc: significant cache rework [1] Align cache management functions to those in Linux kernel. I.e.: a) Use the same functions for all cache ops (D$ Inv/Flush) b) Split cache ops in 3 sub-functions: "before", "lineloop" and "after". That way we may re-use "before" and "after" functions for region and full cache ops. [2] Implement full-functional L2 (SLC) management. Before SLC was simply disabled early on boot. It's also possible to enable or disable L2 cache from config utility. [3] Disable/enable corresponding caches early on boot. So if U-Boot is configured to use caches they will be used at all times (this is useful in partucular for speed-up of relocation). Signed-off-by: Alexey Brodkin <abrodkin@synopsys.com>
2015-05-18 13:56:26 +00:00
Fix board init code to respect the C runtime environment board_init_f_mem() alters the C runtime environment's stack it is actually already using. This is not a valid behaviour within a C runtime environment. Split board_init_f_mem into C functions which do not alter their own stack and always behave properly with respect to their C runtime environment. Signed-off-by: Albert ARIBAUD <albert.u.boot@aribaud.net> Acked-by: Thomas Chou <thomas@wytron.com.tw>
2015-11-25 16:56:32 +00:00
/* Establish C runtime stack and frame */
arc: significant cache rework [1] Align cache management functions to those in Linux kernel. I.e.: a) Use the same functions for all cache ops (D$ Inv/Flush) b) Split cache ops in 3 sub-functions: "before", "lineloop" and "after". That way we may re-use "before" and "after" functions for region and full cache ops. [2] Implement full-functional L2 (SLC) management. Before SLC was simply disabled early on boot. It's also possible to enable or disable L2 cache from config utility. [3] Disable/enable corresponding caches early on boot. So if U-Boot is configured to use caches they will be used at all times (this is useful in partucular for speed-up of relocation). Signed-off-by: Alexey Brodkin <abrodkin@synopsys.com>
2015-05-18 13:56:26 +00:00
mov %sp, CONFIG_SYS_INIT_SP_ADDR
mov %fp, %sp
arc: introduce separate section for interrupt vector table Even though existing implementation works fine in preparation to submission of ARCv2 architecture we need this change. In case of ARCv2 interrupt vector table consists of just addresses of corresponding handlers. And if those addresses will be in .text section then assembler will encode them as everything in .text section as middle-endian and then on real execution CPU will read swapped addresses and will jump into the wild. Once introduced new section is situated so .text section remains the first which allows us to use common linker option for linking everything to a specified CONFIG_SYS_TEXT_BASE. Signed-off-by: Alexey Brodkin <abrodkin@synopsys.com> Signed-off-by: Igor Guryanov <guryanov@synopsys.com>
2014-12-24 14:17:11 +00:00
Fix board init code to respect the C runtime environment board_init_f_mem() alters the C runtime environment's stack it is actually already using. This is not a valid behaviour within a C runtime environment. Split board_init_f_mem into C functions which do not alter their own stack and always behave properly with respect to their C runtime environment. Signed-off-by: Albert ARIBAUD <albert.u.boot@aribaud.net> Acked-by: Thomas Chou <thomas@wytron.com.tw>
2015-11-25 16:56:32 +00:00
/* Allocate reserved area from current top of stack */
arc: get rid of CONFIG_SYS_GENERIC_GLOBAL_DATA As discussed on mailing list we're drifting away from CONFIG_SYS_GENERIC_GLOBAL_DATA in favour to use of board_init_f_mem() for global data. So do this for ARC architecture. Signed-off-by: Alexey Brodkin <abrodkin@synopsys.com>
2015-02-25 15:10:18 +00:00
mov %r0, %sp
Fix board init code to respect the C runtime environment board_init_f_mem() alters the C runtime environment's stack it is actually already using. This is not a valid behaviour within a C runtime environment. Split board_init_f_mem into C functions which do not alter their own stack and always behave properly with respect to their C runtime environment. Signed-off-by: Albert ARIBAUD <albert.u.boot@aribaud.net> Acked-by: Thomas Chou <thomas@wytron.com.tw>
2015-11-25 16:56:32 +00:00
bl board_init_f_alloc_reserve
/* Set stack below reserved area, adjust frame pointer accordingly */
arc: get rid of CONFIG_SYS_GENERIC_GLOBAL_DATA As discussed on mailing list we're drifting away from CONFIG_SYS_GENERIC_GLOBAL_DATA in favour to use of board_init_f_mem() for global data. So do this for ARC architecture. Signed-off-by: Alexey Brodkin <abrodkin@synopsys.com>
2015-02-25 15:10:18 +00:00
mov %sp, %r0
mov %fp, %sp
Fix board init code to respect the C runtime environment board_init_f_mem() alters the C runtime environment's stack it is actually already using. This is not a valid behaviour within a C runtime environment. Split board_init_f_mem into C functions which do not alter their own stack and always behave properly with respect to their C runtime environment. Signed-off-by: Albert ARIBAUD <albert.u.boot@aribaud.net> Acked-by: Thomas Chou <thomas@wytron.com.tw>
2015-11-25 16:56:32 +00:00
/* Initialize reserved area - note: r0 already contains address */
bl board_init_f_init_reserve
ARC: init debug uart in early common arc code The debug UART is intended for use very early in U-Boot to debug problems before serial drivers are up. Call debug_uart_init right before board_init_f. Signed-off-by: Eugeniy Paltsev <Eugeniy.Paltsev@synopsys.com> Signed-off-by: Alexey Brodkin <abrodkin@synopsys.com>
2018-05-03 12:01:58 +00:00
#ifdef CONFIG_DEBUG_UART
/* Earliest point to set up early debug uart */
bl debug_uart_init
#endif
arc: introduce separate section for interrupt vector table Even though existing implementation works fine in preparation to submission of ARCv2 architecture we need this change. In case of ARCv2 interrupt vector table consists of just addresses of corresponding handlers. And if those addresses will be in .text section then assembler will encode them as everything in .text section as middle-endian and then on real execution CPU will read swapped addresses and will jump into the wild. Once introduced new section is situated so .text section remains the first which allows us to use common linker option for linking everything to a specified CONFIG_SYS_TEXT_BASE. Signed-off-by: Alexey Brodkin <abrodkin@synopsys.com> Signed-off-by: Igor Guryanov <guryanov@synopsys.com>
2014-12-24 14:17:11 +00:00
/* Zero the one and only argument of "board_init_f" */
mov_s %r0, 0
arc: Introduce a possibility to not relocate U-boot Disabling relocation might be useful on ARC for 2 reasons: a) For advanced debugging with Synopsys proprietary MetaWare debugger which is capable of accessing much more specific hardware resources compared to gdb. For example it may show contents of L1 and L2 caches, internal states of some hardware blocks etc. But on the downside MetaWare debugger still cannot work with PIE. Even though that limitation could be work-arounded with change of ELF's header and stripping down all debug info but with it we won't have debug info for source-level debugging which is quite inconvenient. b) Some platforms which might benefit from usage of U-Boot basically don't have enough RAM to accommodate relocation of U-Boot so we keep code in flash and use as much of RAM as possible for more interesting things. Signed-off-by: Alexey Brodkin <abrodkin@synopsys.com> Cc: Simon Glass <sjg@chromium.org> Cc: Bin Meng <bmeng.cn@gmail.com> Cc: Heiko Schocher <hs@denx.de> Cc: York Sun <york.sun@nxp.com> Cc: Stefan Roese <sr@denx.de>
2015-12-16 16:24:10 +00:00
bl board_init_f
/* We only get here if relocation is disabled by GD_FLG_SKIP_RELOC */
/* Make sure we don't lose GD overwritten by zero new GD */
mov %r0, %r25
mov %r1, 0
bl board_init_r
arc: merge common start-up code between ARC and ARCv2 Even though ARCompact and ARCv2 are not binary compatible most of assembly instructions are used in both. With this change we'll get rid of duplicate code. Still IVTs are implemented differently so we're keeping them in separate files. Signed-off-by: Alexey Brodkin <abrodkin@synopsys.com>
2015-02-19 15:40:58 +00:00
ENDPROC(_start)
arc: introduce separate section for interrupt vector table Even though existing implementation works fine in preparation to submission of ARCv2 architecture we need this change. In case of ARCv2 interrupt vector table consists of just addresses of corresponding handlers. And if those addresses will be in .text section then assembler will encode them as everything in .text section as middle-endian and then on real execution CPU will read swapped addresses and will jump into the wild. Once introduced new section is situated so .text section remains the first which allows us to use common linker option for linking everything to a specified CONFIG_SYS_TEXT_BASE. Signed-off-by: Alexey Brodkin <abrodkin@synopsys.com> Signed-off-by: Igor Guryanov <guryanov@synopsys.com>
2014-12-24 14:17:11 +00:00
arc: add cpu files Signed-off-by: Alexey Brodkin <abrodkin@synopsys.com> Cc: Vineet Gupta <vgupta@synopsys.com> Cc: Francois Bedard <fbedard@synopsys.com> Cc: Wolfgang Denk <wd@denx.de> Cc: Heiko Schocher <hs@denx.de>
2014-02-04 08:56:14 +00:00
/*
arc: clean-up init procedure Intention behind this work was elimination of as much assembly-written code as it is possible. In case of ARC we already have relocation fix-up implemented in C so why don't we use C for U-Boot copying, .bss zeroing etc. It turned out x86 uses pretty similar approach so we re-used parts of code in "board_f.c" initially implemented for x86. Now assembly usage during init is limited to stack- and frame-pointer setup before and after relocation. Signed-off-by: Alexey Brodkin <abrodkin@synopsys.com> Cc: Simon Glass <sjg@chromium.org>
2015-02-24 16:40:36 +00:00
* void board_init_f_r_trampoline(stack-pointer address)
arc: add cpu files Signed-off-by: Alexey Brodkin <abrodkin@synopsys.com> Cc: Vineet Gupta <vgupta@synopsys.com> Cc: Francois Bedard <fbedard@synopsys.com> Cc: Wolfgang Denk <wd@denx.de> Cc: Heiko Schocher <hs@denx.de>
2014-02-04 08:56:14 +00:00
*
* This "function" does not return, instead it continues in RAM
* after relocating the monitor code.
*
arc: clean-up init procedure Intention behind this work was elimination of as much assembly-written code as it is possible. In case of ARC we already have relocation fix-up implemented in C so why don't we use C for U-Boot copying, .bss zeroing etc. It turned out x86 uses pretty similar approach so we re-used parts of code in "board_f.c" initially implemented for x86. Now assembly usage during init is limited to stack- and frame-pointer setup before and after relocation. Signed-off-by: Alexey Brodkin <abrodkin@synopsys.com> Cc: Simon Glass <sjg@chromium.org>
2015-02-24 16:40:36 +00:00
* r0 = new stack-pointer
arc: add cpu files Signed-off-by: Alexey Brodkin <abrodkin@synopsys.com> Cc: Vineet Gupta <vgupta@synopsys.com> Cc: Francois Bedard <fbedard@synopsys.com> Cc: Wolfgang Denk <wd@denx.de> Cc: Heiko Schocher <hs@denx.de>
2014-02-04 08:56:14 +00:00
*/
arc: clean-up init procedure Intention behind this work was elimination of as much assembly-written code as it is possible. In case of ARC we already have relocation fix-up implemented in C so why don't we use C for U-Boot copying, .bss zeroing etc. It turned out x86 uses pretty similar approach so we re-used parts of code in "board_f.c" initially implemented for x86. Now assembly usage during init is limited to stack- and frame-pointer setup before and after relocation. Signed-off-by: Alexey Brodkin <abrodkin@synopsys.com> Cc: Simon Glass <sjg@chromium.org>
2015-02-24 16:40:36 +00:00
ENTRY(board_init_f_r_trampoline)
/* Set up the stack- and frame-pointers */
mov %sp, %r0
arc: add cpu files Signed-off-by: Alexey Brodkin <abrodkin@synopsys.com> Cc: Vineet Gupta <vgupta@synopsys.com> Cc: Francois Bedard <fbedard@synopsys.com> Cc: Wolfgang Denk <wd@denx.de> Cc: Heiko Schocher <hs@denx.de>
2014-02-04 08:56:14 +00:00
mov %fp, %sp
arc: clean-up init procedure Intention behind this work was elimination of as much assembly-written code as it is possible. In case of ARC we already have relocation fix-up implemented in C so why don't we use C for U-Boot copying, .bss zeroing etc. It turned out x86 uses pretty similar approach so we re-used parts of code in "board_f.c" initially implemented for x86. Now assembly usage during init is limited to stack- and frame-pointer setup before and after relocation. Signed-off-by: Alexey Brodkin <abrodkin@synopsys.com> Cc: Simon Glass <sjg@chromium.org>
2015-02-24 16:40:36 +00:00
/* Update position of intterupt vector table */
lr %r0, [ARC_AUX_INTR_VEC_BASE]
ld %r1, [%r25, GD_RELOC_OFF]
add %r0, %r0, %r1
sr %r0, [ARC_AUX_INTR_VEC_BASE]
arc: add cpu files Signed-off-by: Alexey Brodkin <abrodkin@synopsys.com> Cc: Vineet Gupta <vgupta@synopsys.com> Cc: Francois Bedard <fbedard@synopsys.com> Cc: Wolfgang Denk <wd@denx.de> Cc: Heiko Schocher <hs@denx.de>
2014-02-04 08:56:14 +00:00
arc: clean-up init procedure Intention behind this work was elimination of as much assembly-written code as it is possible. In case of ARC we already have relocation fix-up implemented in C so why don't we use C for U-Boot copying, .bss zeroing etc. It turned out x86 uses pretty similar approach so we re-used parts of code in "board_f.c" initially implemented for x86. Now assembly usage during init is limited to stack- and frame-pointer setup before and after relocation. Signed-off-by: Alexey Brodkin <abrodkin@synopsys.com> Cc: Simon Glass <sjg@chromium.org>
2015-02-24 16:40:36 +00:00
/* Re-enter U-Boot by calling board_init_f_r */
j board_init_f_r
ENDPROC(board_init_f_r_trampoline)
Reference in a new issue Copy permalink