The omap3 L2 cache enable/disable function to omap3 dependent code
Kim
PATCH 1/1 The omap3 L2 cache enable/disable function to omap3 dependent code
---
The L2 cache enable/disable function in the cpu/arm_cortexa8/cpu.c moved to cpu/arm_cortexa8/omap3/cache.c.
This patches fixes the First issue in the following
http://lists.denx.de/pipermail/u-boot/2009-May/053433.html
The Second issue is fixed by
http://lists.denx.de/pipermail/u-boot/2009-May/053490.html
cpu/arm_cortexa8/cpu.c | 68 +--------------------- cpu/arm_cortexa8/omap3/Makefile | 2 +- cpu/arm_cortexa8/omap3/board.c | 5 +- cpu/arm_cortexa8/omap3/cache.c | 99 ++++++++++++++++++++++++++++++++ include/asm-arm/arch-omap3/sys_proto.h | 1 - include/asm-arm/cache.h | 37 ++++++++++++ 6 files changed, 143 insertions(+), 69 deletions(-) create mode 100644 cpu/arm_cortexa8/omap3/cache.c create mode 100644 include/asm-arm/cache.h
cpu/arm_cortexa8/cpu.c +3 -65
@@ -35,15 +35,12 @@#include <command.h> #include <asm/arch/sys_proto.h> #include <asm/system.h> +#include <asm/cache.h> #ifdef CONFIG_USE_IRQ DECLARE_GLOBAL_DATA_PTR ;#endif -#ifndef CONFIG_L2_OFF -void l2cache_disable (void ); -#endif -static void cache_flush (void );int cpu_init (void )
@@ -80,7 +77,7 @@ int cleanup_before_linux(void)#ifndef CONFIG_L2_OFF /* turn off L2 cache */ -l2cache_disable (); +l2_cache_disable ();/* invalidate L2 cache also */ v7_flush_dcache_all (get_device_type ());#endif
@@ -89,71 +86,12 @@ int cleanup_before_linux(void) asm("mcr p15, 0, %0, c7, c10, 4" : :"r" (i));#ifndef CONFIG_L2_OFF -l2cache_enable (); +l2_cache_enable ();#endif return 0 ; } -void l2cache_enable () -{ -unsigned long i; -volatile unsigned int j; - -/* ES2 onwards we can disable/enable L2 ourselves */ -if (get_cpu_rev () >=CPU_3XX_ES20 ) { - __asm__ __volatile__("mrc p15, 0, %0, c1, c0, 1" :"=r" (i)); - __asm__ __volatile__("orr %0, %0, #0x2" :"=r" (i)); - __asm__ __volatile__("mcr p15, 0, %0, c1, c0, 1" :"=r" (i)); - }else { -/* Save r0, r12 and restore them after usage */ - __asm__ __volatile__("mov %0, r12" :"=r" (j)); - __asm__ __volatile__("mov %0, r0" :"=r" (i)); - -/* -* GP Device ROM code API usage here -* r12 = AUXCR Write function and r0 value -*/ - __asm__ __volatile__("mov r12, #0x3" ); - __asm__ __volatile__("mrc p15, 0, r0, c1, c0, 1" ); - __asm__ __volatile__("orr r0, r0, #0x2" ); -/* SMI instruction to call ROM Code API */ - __asm__ __volatile__(".word 0xE1600070" ); - __asm__ __volatile__("mov r0, %0" :"=r" (i)); - __asm__ __volatile__("mov r12, %0" :"=r" (j)); - } - -} - -void l2cache_disable () -{ -unsigned long i; -volatile unsigned int j; - -/* ES2 onwards we can disable/enable L2 ourselves */ -if (get_cpu_rev () >=CPU_3XX_ES20 ) { - __asm__ __volatile__("mrc p15, 0, %0, c1, c0, 1" :"=r" (i)); - __asm__ __volatile__("bic %0, %0, #0x2" :"=r" (i)); - __asm__ __volatile__("mcr p15, 0, %0, c1, c0, 1" :"=r" (i)); - }else { -/* Save r0, r12 and restore them after usage */ - __asm__ __volatile__("mov %0, r12" :"=r" (j)); - __asm__ __volatile__("mov %0, r0" :"=r" (i)); - -/* -* GP Device ROM code API usage here -* r12 = AUXCR Write function and r0 value -*/ - __asm__ __volatile__("mov r12, #0x3" ); - __asm__ __volatile__("mrc p15, 0, r0, c1, c0, 1" ); - __asm__ __volatile__("bic r0, r0, #0x2" ); -/* SMI instruction to call ROM Code API */ - __asm__ __volatile__(".word 0xE1600070" ); - __asm__ __volatile__("mov r0, %0" :"=r" (i)); - __asm__ __volatile__("mov r12, %0" :"=r" (j)); - } -} -static void cache_flush (void ) { asm ("mcr p15, 0, %0, c7, c5, 0" : :"r" (0 ));
cpu/arm_cortexa8/omap3/Makefile +1 -1
@@ -26,7 +26,7 @@ include $(TOPDIR)/config.mk LIB = $(obj)lib$(SOC).a SOBJS := lowlevel_init.o -COBJS := sys_info.o board.o clock.o interrupts.o mem.o syslib.o +COBJS := sys_info.o board.o clock.o interrupts.o mem.o syslib.o cache.o SRCS := $(SOBJS:.o=.S) $(COBJS:.o=.c) OBJS := $(addprefix $(obj),$(COBJS) $(SOBJS))
cpu/arm_cortexa8/omap3/board.c +3 -2
@@ -36,6 +36,7 @@#include <asm/io.h> #include <asm/arch/sys_proto.h> #include <asm/arch/mem.h> +#include <asm/cache.h> extern omap3_sysinfo sysinfo;
@@ -206,9 +207,9 @@ void s_init(void)#endif #ifdef CONFIG_L2_OFF -l2cache_disable (); +l2_cache_disable ();#else -l2cache_enable (); +l2_cache_enable ();#endif /* * Writing to AuxCR in U-boot using SMI for GP DEV
cpu/arm_cortexa8/omap3/cache.c new file +99 -0
99 lines not on this page · raw message
(quoted diff) +0 -0
b/include/asm-arm/arch-omap3/sys_proto.h
index 7361d08..9314c7c 100644
include/asm-arm/arch-omap3/sys_proto.h +0 -1
@@ -50,7 +50,6 @@ u32 is_running_in_sdram(void);u32 is_running_in_sram (void );u32 is_running_in_flash (void );u32 get_device_type (void ); -void l2cache_enable (void );void secureworld_exit (void );void setup_auxcr (void );void try_unlock_memory (void );
include/asm-arm/cache.h new file +37 -0
@@ -0,0 +1,37 @@ +/* +* (C) Copyright 2008 Texas Insturments +* +* (C) Copyright 2002 +* Sysgo Real-Time Solutions, GmbH <www.elinos.com> +* Marius Groeger <[email protected]> +* +* (C) Copyright 2002 +* Gary Jennejohn, DENX Software Engineering, <[email protected]> +* +* See file CREDITS for list of people who contributed to this +* project. +* +* This program is free software; you can redistribute it and/or +* modify it under the terms of the GNU General Public License as +* published by the Free Software Foundation; either version 2 of +* the License, or (at your option) any later version. +* +* This program is distributed in the hope that it will be useful, +* but WITHOUT ANY WARRANTY; without even the implied warranty of +* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the +* GNU General Public License for more details. +* +* You should have received a copy of the GNU General Public License +* along with this program; if not, write to the Free Software +* Foundation, Inc., 59 Temple Place, Suite 330, Boston, +* MA 02111-1307 USA +*/ + +#ifndef _L2_CACHE_H_ +#define _L2_CACHE_H_ + +void l2_cache_disable (); +void l2_cache_enable (); + +#endif +
--
1.5.6.3
PATCH 1/1 The omap3 L2 cache enable/disable function to omap3 dependent code
Quoted 8 lines · Heung Jun
CC Dirk Behme <[email protected]>Signed-off-by HeungJun, Kim <[email protected]>---
The L2 cache enable/disable function in the
cpu/arm_cortexa8/cpu.cmoved tocpu/arm_cortexa8/omap3/cache.c.
Quoted 8 lines
This patches fixes the First issue in the following
http://lists.denx.de/pipermail/u-boot/2009-May/053433.html
The Second issue is fixed by
Hi Riverful,
1. the patch title is missing the verb 'move'
2. your signoff is in the correct location, however the commit message is not. It belongs above the line with the '---'.
3. while making backreferences by pointing to messages in the list archive is cool, they don't stand the test of time - the URL may one day cease to exist. So it is better to either reference a commitish in the repository itself, or simply restate/copy the original data.
Thanks,
Kim
Kim
PATCH 1/1 The omap3 L2 cache enable/disable function to omap3 dependent code
Dear Kim Philips,
Thanks to review & help.
I agree with 1, 2 ideas.
Quoted 4 lines
3. while making backreferences by pointing to messages in the list archive is cool, they don't stand the test of time - the URL may one day cease to exist. So it is better to either reference a commitish in the repository itself, or simply restate/copy the original data.
But, I don't understand idea 3 well.
Let me know more detailed? :-)
I don't understand the word "commitish" & the meaning of "original data". And, I think that you tell me to simplify the comments.
Is it right or something like?
BTW, Thanks one more :)
Best Regards,
riverful
2009/6/2 Kim Phillips <[email protected]>:
Quoted 34 lines
Quoted 8 lines · Heung Jun
CC Dirk Behme <[email protected]>Signed-off-by HeungJun, Kim <[email protected]>---
The L2 cache enable/disable function in the
cpu/arm_cortexa8/cpu.cmoved tocpu/arm_cortexa8/omap3/cache.c.Quoted 8 lines
This patches fixes the First issue in the following
http://lists.denx.de/pipermail/u-boot/2009-May/053433.html
The Second issue is fixed by
Hi Riverful,
1. the patch title is missing the verb 'move'
2. your signoff is in the correct location, however the commit message is not. ?It belongs above the line with the '---'.
3. while making backreferences by pointing to messages in the list archive is cool, they don't stand the test of time - the URL may one day cease to exist. ?So it is better to either reference a commitish in the repository itself, or simply restate/copy the original data.Thanks,
Kim
PATCH 1/1 The omap3 L2 cache enable/disable function to omap3 dependent code
Quoted 8 lines · Heung Jun
Dear Kim Philips,
Thanks to review & help.
Quoted 4 lines
3. while making backreferences by pointing to messages in the list archive is cool, they don't stand the test of time - the URL may one day cease to exist. So it is better to either reference a commitish in the repository itself, or simply restate/copy the original data.
Quoted 5 lines
But, I don't understand idea 3 well.
Let me know more detailed? :-)
I don't understand the word "commitish" & the meaning of "original data". And, I think that you tell me to simplify the comments.
Is it right or something like?
basically, I'm saying that backreferences should only be git commit ID's ('commitish' in git terminology). See e.g., "Commit 574b319512 introduced" and "A later fix in commit 8ec6e332ea" below:
commit af75a45d23b72a59ac5cc0427696c7f634fdc94b
Author: Wolfgang Denk <[email protected]> Date: Fri May 15 09:27:58 2009 +0200
IDE: bail out of dev_print() for unknown device types
Commit 574b319512 introduced a subtle bug by mixing a list of tests for "dev_desc->type" and "dev_desc->if_type" into one switch(), which then mostly did not work because "dev_desc->type" cannot take any "IF_*" type values. A later fix in commit 8ec6e332ea changed the switch() into testing "dev_desc->if_type", but at this point the initial test for unknown device types was completely lost, which resulted in output like that for IDE ports without device attached:
Device 1: Model: Firm: Ser#:
Type: # 1F #
Capacity: not available
This patch re-introduces the missing test for unknown device types.
...and if your commit message wants to reference something that is _not_ already in the repository, e.g., in your case I think you have this compiler error:
cpu.c:99: warning: function declaration isn't a prototype cpu.c: In function 'l2cache_enable':
cpu.c:104: warning: implicit declaration of function 'get_cpu_rev' cpu.c:104: error: 'CPU_3XX_ES20' undeclared (first use in this function) cpu.c:104: error: (Each undeclared identifier is reported only once cpu.c:104: error: for each function it appears in.)cpu.c: At top level:
cpu.c:129: warning: function declaration isn't a prototype cpu.c: In function 'l2cache_disable':
cpu.c:134: error: 'CPU_3XX_ES20' undeclared (first use in this function) make[1]: *** [cpu.o] Error 1
...so please copy it into the commit message, so that it lives as long as the life of the project (as opposed to the life of the maillist archive server URL).
Kim
PATCH 1/1 The omap3 L2 cache enable/disable function to omap3 dependent code
In reply to Kim PhillipsQuoted 22 lines · Kim Phillips
Quoted 8 lines · Heung Jun
CC Dirk Behme <[email protected]>Signed-off-by HeungJun, Kim <[email protected]>---
The L2 cache enable/disable function in the
cpu/arm_cortexa8/cpu.cmoved tocpu/arm_cortexa8/omap3/cache.c.Quoted 8 lines
This patches fixes the First issue in the following
http://lists.denx.de/pipermail/u-boot/2009-May/053433.html
The Second issue is fixed by
Hi Riverful,
Quoted 7 lines
1. the patch title is missing the verb 'move'
2. your signoff is in the correct location, however the commit message is not. It belongs above the line with the '---'.
3. while making backreferences by pointing to messages in the list archive is cool, they don't stand the test of time - the URL may one day cease to exist. So it is better to either reference a commitish in the repository itself, or simply restate/copy the original data.
As long as references to the mailing list archive are in patch comment (below --- as they are here) and don't go into git I think it's fine to use them.
Best regards
Dirk
PATCH 1/1 The omap3 L2 cache enable/disable function to omap3 dependent code
Quoted 10 lines · On Mon, 01 Jun 2009 18:46:21 +0200 Dirk Behme <[email protected]> wrote:
Quoted 6 lines
2. your signoff is in the correct location, however the commit message is not. It belongs above the line with the '---'.
3. while making backreferences by pointing to messages in the list archive is cool, they don't stand the test of time - the URL may one day cease to exist. So it is better to either reference a commitish in the repository itself, or simply restate/copy the original data.As long as references to the mailing list archive are in patch comment (below --- as they are here) and don't go into git I think it's fine to use them.
agreed (and I just did that myself in another thread) - it's just that, in this case, the referenced subject matter needs to go in the git commit message (see 2 above).
Kim