davidlt/kernel-ark
1
0
Fork 0
Code Issues Pull Requests Releases Wiki Activity
cdfc83075f
kernel-ark/arch/mips/mm/c-r3k.c

347 lines
7.9 KiB
C
Raw Normal View History

Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 22:20:36 +00:00
/*
* r2300.c: R2000 and R3000 specific mmu/cache code.
*
update David Miller's old email address Signed-off-by: Justin P. Mattock <justinmattock@gmail.com> Acked-by: David S. Miller <davem@davemloft.net> Signed-off-by: Jiri Kosina <jkosina@suse.cz>
2011-04-04 21:15:29 +00:00
* Copyright (C) 1996 David S. Miller (davem@davemloft.net)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 22:20:36 +00:00
*
* with a lot of changes to make this thing work for R3000s
* Tx39XX R4k style caches added. HK
* Copyright (C) 1998, 1999, 2000 Harald Koerfgen
* Copyright (C) 1998 Gleb Raiko & Vladimir Roganov
[MIPS] c-r3k: Implement flush_cache_range() Contrary to the belief of some, the R3000 and related processors did have caches, both a data and an instruction cache. Here is an implementation of r3k_flush_cache_page(), which is the processor-specific back-end for flush_cache_range(), done according to the spec in Documentation/cachetlb.txt. While at it, remove an unused local function: get_phys_page(), do some trivial formatting fixes and modernise debugging facilities. Signed-off-by: Maciej W. Rozycki <macro@linux-mips.org> Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
2007-10-17 10:51:39 +00:00
* Copyright (C) 2001, 2004, 2007 Maciej W. Rozycki
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 22:20:36 +00:00
*/
#include <linux/kernel.h>
#include <linux/sched.h>
MIPS: Build fix - include <linux/smp.h> into all smp_processor_id() users. Some of the were relying into smp.h being dragged in by another header which of course is fragile. <asm/cpu-info.h> uses smp_processor_id() only in macros and including smp.h there leads to an include loop, so don't change cpu-info.h. Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
2009-06-19 13:05:26 +00:00
#include <linux/smp.h>
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 22:20:36 +00:00
#include <linux/mm.h>
#include <asm/page.h>
#include <asm/pgtable.h>
#include <asm/mmu_context.h>
#include <asm/isadep.h>
#include <asm/io.h>
#include <asm/bootinfo.h>
#include <asm/cpu.h>
static unsigned long icache_size, dcache_size; /* Size in bytes */
static unsigned long icache_lsize, dcache_lsize; /* Size in bytes */
MIPS: Delete __cpuinit/__CPUINIT usage from MIPS code commit 3747069b25e419f6b51395f48127e9812abc3596 upstream. The __cpuinit type of throwaway sections might have made sense some time ago when RAM was more constrained, but now the savings do not offset the cost and complications. For example, the fix in commit 5e427ec2d0 ("x86: Fix bit corruption at CPU resume time") is a good example of the nasty type of bugs that can be created with improper use of the various __init prefixes. After a discussion on LKML[1] it was decided that cpuinit should go the way of devinit and be phased out. Once all the users are gone, we can then finally remove the macros themselves from linux/init.h. Note that some harmless section mismatch warnings may result, since notify_cpu_starting() and cpu_up() are arch independent (kernel/cpu.c) and are flagged as __cpuinit -- so if we remove the __cpuinit from the arch specific callers, we will also get section mismatch warnings. As an intermediate step, we intend to turn the linux/init.h cpuinit related content into no-ops as early as possible, since that will get rid of these warnings. In any case, they are temporary and harmless. Here, we remove all the MIPS __cpuinit from C code and __CPUINIT from asm files. MIPS is interesting in this respect, because there are also uasm users hiding behind their own renamed versions of the __cpuinit macros. [1] https://lkml.org/lkml/2013/5/20/589 [ralf@linux-mips.org: Folded in Paul's followup fix.] Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com> Cc: linux-mips@linux-mips.org Patchwork: https://patchwork.linux-mips.org/patch/5494/ Patchwork: https://patchwork.linux-mips.org/patch/5495/ Patchwork: https://patchwork.linux-mips.org/patch/5509/ Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
2013-06-18 13:38:59 +00:00
unsigned long r3k_cache_size(unsigned long ca_flags)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 22:20:36 +00:00
{
unsigned long flags, status, dummy, size;
volatile unsigned long *p;
p = (volatile unsigned long *) KSEG0;
flags = read_c0_status();
/* isolate cache space */
write_c0_status((ca_flags|flags)&~ST0_IEC);
*p = 0xa5a55a5a;
dummy = *p;
status = read_c0_status();
if (dummy != 0xa5a55a5a || (status & ST0_CM)) {
size = 0;
} else {
for (size = 128; size <= 0x40000; size <<= 1)
*(p + size) = 0;
*p = -1;
for (size = 128;
(size <= 0x40000) && (*(p + size) == 0);
size <<= 1)
;
if (size > 0x40000)
size = 0;
}
write_c0_status(flags);
return size * sizeof(*p);
}
MIPS: Delete __cpuinit/__CPUINIT usage from MIPS code commit 3747069b25e419f6b51395f48127e9812abc3596 upstream. The __cpuinit type of throwaway sections might have made sense some time ago when RAM was more constrained, but now the savings do not offset the cost and complications. For example, the fix in commit 5e427ec2d0 ("x86: Fix bit corruption at CPU resume time") is a good example of the nasty type of bugs that can be created with improper use of the various __init prefixes. After a discussion on LKML[1] it was decided that cpuinit should go the way of devinit and be phased out. Once all the users are gone, we can then finally remove the macros themselves from linux/init.h. Note that some harmless section mismatch warnings may result, since notify_cpu_starting() and cpu_up() are arch independent (kernel/cpu.c) and are flagged as __cpuinit -- so if we remove the __cpuinit from the arch specific callers, we will also get section mismatch warnings. As an intermediate step, we intend to turn the linux/init.h cpuinit related content into no-ops as early as possible, since that will get rid of these warnings. In any case, they are temporary and harmless. Here, we remove all the MIPS __cpuinit from C code and __CPUINIT from asm files. MIPS is interesting in this respect, because there are also uasm users hiding behind their own renamed versions of the __cpuinit macros. [1] https://lkml.org/lkml/2013/5/20/589 [ralf@linux-mips.org: Folded in Paul's followup fix.] Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com> Cc: linux-mips@linux-mips.org Patchwork: https://patchwork.linux-mips.org/patch/5494/ Patchwork: https://patchwork.linux-mips.org/patch/5495/ Patchwork: https://patchwork.linux-mips.org/patch/5509/ Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
2013-06-18 13:38:59 +00:00
unsigned long r3k_cache_lsize(unsigned long ca_flags)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 22:20:36 +00:00
{
unsigned long flags, status, lsize, i;
volatile unsigned long *p;
p = (volatile unsigned long *) KSEG0;
flags = read_c0_status();
/* isolate cache space */
write_c0_status((ca_flags|flags)&~ST0_IEC);
for (i = 0; i < 128; i++)
*(p + i) = 0;
*(volatile unsigned char *)p = 0;
for (lsize = 1; lsize < 128; lsize <<= 1) {
*(p + lsize);
status = read_c0_status();
if (!(status & ST0_CM))
break;
}
for (i = 0; i < 128; i += lsize)
*(volatile unsigned char *)(p + i) = 0;
write_c0_status(flags);
return lsize * sizeof(*p);
}
MIPS: Delete __cpuinit/__CPUINIT usage from MIPS code commit 3747069b25e419f6b51395f48127e9812abc3596 upstream. The __cpuinit type of throwaway sections might have made sense some time ago when RAM was more constrained, but now the savings do not offset the cost and complications. For example, the fix in commit 5e427ec2d0 ("x86: Fix bit corruption at CPU resume time") is a good example of the nasty type of bugs that can be created with improper use of the various __init prefixes. After a discussion on LKML[1] it was decided that cpuinit should go the way of devinit and be phased out. Once all the users are gone, we can then finally remove the macros themselves from linux/init.h. Note that some harmless section mismatch warnings may result, since notify_cpu_starting() and cpu_up() are arch independent (kernel/cpu.c) and are flagged as __cpuinit -- so if we remove the __cpuinit from the arch specific callers, we will also get section mismatch warnings. As an intermediate step, we intend to turn the linux/init.h cpuinit related content into no-ops as early as possible, since that will get rid of these warnings. In any case, they are temporary and harmless. Here, we remove all the MIPS __cpuinit from C code and __CPUINIT from asm files. MIPS is interesting in this respect, because there are also uasm users hiding behind their own renamed versions of the __cpuinit macros. [1] https://lkml.org/lkml/2013/5/20/589 [ralf@linux-mips.org: Folded in Paul's followup fix.] Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com> Cc: linux-mips@linux-mips.org Patchwork: https://patchwork.linux-mips.org/patch/5494/ Patchwork: https://patchwork.linux-mips.org/patch/5495/ Patchwork: https://patchwork.linux-mips.org/patch/5509/ Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
2013-06-18 13:38:59 +00:00
static void r3k_probe_cache(void)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 22:20:36 +00:00
{
dcache_size = r3k_cache_size(ST0_ISC);
if (dcache_size)
dcache_lsize = r3k_cache_lsize(ST0_ISC);
icache_size = r3k_cache_size(ST0_ISC|ST0_SWC);
if (icache_size)
icache_lsize = r3k_cache_lsize(ST0_ISC|ST0_SWC);
}
static void r3k_flush_icache_range(unsigned long start, unsigned long end)
{
unsigned long size, i, flags;
volatile unsigned char *p;
size = end - start;
if (size > icache_size || KSEGX(start) != KSEG0) {
start = KSEG0;
size = icache_size;
}
p = (char *)start;
flags = read_c0_status();
/* isolate cache space */
write_c0_status((ST0_ISC|ST0_SWC|flags)&~ST0_IEC);
for (i = 0; i < size; i += 0x080) {
MIPS: Whitespace cleanup. Having received another series of whitespace patches I decided to do this once and for all rather than dealing with this kind of patches trickling in forever. Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
2013-01-22 11:59:30 +00:00
asm( "sb\t$0, 0x000(%0)\n\t"
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 22:20:36 +00:00
"sb\t$0, 0x004(%0)\n\t"
"sb\t$0, 0x008(%0)\n\t"
"sb\t$0, 0x00c(%0)\n\t"
"sb\t$0, 0x010(%0)\n\t"
"sb\t$0, 0x014(%0)\n\t"
"sb\t$0, 0x018(%0)\n\t"
"sb\t$0, 0x01c(%0)\n\t"
[MIPS] War on whitespace: cleanup initial spaces followed by tabs. Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
2006-03-11 08:18:41 +00:00
"sb\t$0, 0x020(%0)\n\t"
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 22:20:36 +00:00
"sb\t$0, 0x024(%0)\n\t"
"sb\t$0, 0x028(%0)\n\t"
"sb\t$0, 0x02c(%0)\n\t"
"sb\t$0, 0x030(%0)\n\t"
"sb\t$0, 0x034(%0)\n\t"
"sb\t$0, 0x038(%0)\n\t"
"sb\t$0, 0x03c(%0)\n\t"
"sb\t$0, 0x040(%0)\n\t"
"sb\t$0, 0x044(%0)\n\t"
"sb\t$0, 0x048(%0)\n\t"
"sb\t$0, 0x04c(%0)\n\t"
"sb\t$0, 0x050(%0)\n\t"
"sb\t$0, 0x054(%0)\n\t"
"sb\t$0, 0x058(%0)\n\t"
"sb\t$0, 0x05c(%0)\n\t"
[MIPS] War on whitespace: cleanup initial spaces followed by tabs. Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
2006-03-11 08:18:41 +00:00
"sb\t$0, 0x060(%0)\n\t"
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 22:20:36 +00:00
"sb\t$0, 0x064(%0)\n\t"
"sb\t$0, 0x068(%0)\n\t"
"sb\t$0, 0x06c(%0)\n\t"
"sb\t$0, 0x070(%0)\n\t"
"sb\t$0, 0x074(%0)\n\t"
"sb\t$0, 0x078(%0)\n\t"
"sb\t$0, 0x07c(%0)\n\t"
: : "r" (p) );
p += 0x080;
}
write_c0_status(flags);
}
static void r3k_flush_dcache_range(unsigned long start, unsigned long end)
{
unsigned long size, i, flags;
volatile unsigned char *p;
size = end - start;
if (size > dcache_size || KSEGX(start) != KSEG0) {
start = KSEG0;
size = dcache_size;
}
p = (char *)start;
flags = read_c0_status();
/* isolate cache space */
write_c0_status((ST0_ISC|flags)&~ST0_IEC);
for (i = 0; i < size; i += 0x080) {
MIPS: Whitespace cleanup. Having received another series of whitespace patches I decided to do this once and for all rather than dealing with this kind of patches trickling in forever. Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
2013-01-22 11:59:30 +00:00
asm( "sb\t$0, 0x000(%0)\n\t"
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 22:20:36 +00:00
"sb\t$0, 0x004(%0)\n\t"
"sb\t$0, 0x008(%0)\n\t"
"sb\t$0, 0x00c(%0)\n\t"
[MIPS] War on whitespace: cleanup initial spaces followed by tabs. Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
2006-03-11 08:18:41 +00:00
"sb\t$0, 0x010(%0)\n\t"
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 22:20:36 +00:00
"sb\t$0, 0x014(%0)\n\t"
"sb\t$0, 0x018(%0)\n\t"
"sb\t$0, 0x01c(%0)\n\t"
[MIPS] War on whitespace: cleanup initial spaces followed by tabs. Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
2006-03-11 08:18:41 +00:00
"sb\t$0, 0x020(%0)\n\t"
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 22:20:36 +00:00
"sb\t$0, 0x024(%0)\n\t"
"sb\t$0, 0x028(%0)\n\t"
"sb\t$0, 0x02c(%0)\n\t"
[MIPS] War on whitespace: cleanup initial spaces followed by tabs. Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
2006-03-11 08:18:41 +00:00
"sb\t$0, 0x030(%0)\n\t"
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 22:20:36 +00:00
"sb\t$0, 0x034(%0)\n\t"
"sb\t$0, 0x038(%0)\n\t"
"sb\t$0, 0x03c(%0)\n\t"
[MIPS] War on whitespace: cleanup initial spaces followed by tabs. Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
2006-03-11 08:18:41 +00:00
"sb\t$0, 0x040(%0)\n\t"
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 22:20:36 +00:00
"sb\t$0, 0x044(%0)\n\t"
"sb\t$0, 0x048(%0)\n\t"
"sb\t$0, 0x04c(%0)\n\t"
[MIPS] War on whitespace: cleanup initial spaces followed by tabs. Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
2006-03-11 08:18:41 +00:00
"sb\t$0, 0x050(%0)\n\t"
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 22:20:36 +00:00
"sb\t$0, 0x054(%0)\n\t"
"sb\t$0, 0x058(%0)\n\t"
"sb\t$0, 0x05c(%0)\n\t"
[MIPS] War on whitespace: cleanup initial spaces followed by tabs. Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
2006-03-11 08:18:41 +00:00
"sb\t$0, 0x060(%0)\n\t"
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 22:20:36 +00:00
"sb\t$0, 0x064(%0)\n\t"
"sb\t$0, 0x068(%0)\n\t"
"sb\t$0, 0x06c(%0)\n\t"
[MIPS] War on whitespace: cleanup initial spaces followed by tabs. Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
2006-03-11 08:18:41 +00:00
"sb\t$0, 0x070(%0)\n\t"
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 22:20:36 +00:00
"sb\t$0, 0x074(%0)\n\t"
"sb\t$0, 0x078(%0)\n\t"
"sb\t$0, 0x07c(%0)\n\t"
: : "r" (p) );
p += 0x080;
}
write_c0_status(flags);
}
static inline void r3k_flush_cache_all(void)
{
}
static inline void r3k___flush_cache_all(void)
{
r3k_flush_dcache_range(KSEG0, KSEG0 + dcache_size);
r3k_flush_icache_range(KSEG0, KSEG0 + icache_size);
}
static void r3k_flush_cache_mm(struct mm_struct *mm)
{
}
static void r3k_flush_cache_range(struct vm_area_struct *vma,
[MIPS] c-r3k: Implement flush_cache_range() Contrary to the belief of some, the R3000 and related processors did have caches, both a data and an instruction cache. Here is an implementation of r3k_flush_cache_page(), which is the processor-specific back-end for flush_cache_range(), done according to the spec in Documentation/cachetlb.txt. While at it, remove an unused local function: get_phys_page(), do some trivial formatting fixes and modernise debugging facilities. Signed-off-by: Maciej W. Rozycki <macro@linux-mips.org> Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
2007-10-17 10:51:39 +00:00
unsigned long start, unsigned long end)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 22:20:36 +00:00
{
}
[MIPS] c-r3k: Implement flush_cache_range() Contrary to the belief of some, the R3000 and related processors did have caches, both a data and an instruction cache. Here is an implementation of r3k_flush_cache_page(), which is the processor-specific back-end for flush_cache_range(), done according to the spec in Documentation/cachetlb.txt. While at it, remove an unused local function: get_phys_page(), do some trivial formatting fixes and modernise debugging facilities. Signed-off-by: Maciej W. Rozycki <macro@linux-mips.org> Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
2007-10-17 10:51:39 +00:00
static void r3k_flush_cache_page(struct vm_area_struct *vma,
unsigned long addr, unsigned long pfn)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 22:20:36 +00:00
{
[MIPS] c-r3k: Implement flush_cache_range() Contrary to the belief of some, the R3000 and related processors did have caches, both a data and an instruction cache. Here is an implementation of r3k_flush_cache_page(), which is the processor-specific back-end for flush_cache_range(), done according to the spec in Documentation/cachetlb.txt. While at it, remove an unused local function: get_phys_page(), do some trivial formatting fixes and modernise debugging facilities. Signed-off-by: Maciej W. Rozycki <macro@linux-mips.org> Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
2007-10-17 10:51:39 +00:00
unsigned long kaddr = KSEG0ADDR(pfn << PAGE_SHIFT);
int exec = vma->vm_flags & VM_EXEC;
struct mm_struct *mm = vma->vm_mm;
pgd_t *pgdp;
pud_t *pudp;
pmd_t *pmdp;
pte_t *ptep;
pr_debug("cpage[%08lx,%08lx]\n",
cpu_context(smp_processor_id(), mm), addr);
/* No ASID => no such page in the cache. */
if (cpu_context(smp_processor_id(), mm) == 0)
return;
pgdp = pgd_offset(mm, addr);
pudp = pud_offset(pgdp, addr);
pmdp = pmd_offset(pudp, addr);
ptep = pte_offset(pmdp, addr);
/* Invalid => no such page in the cache. */
if (!(pte_val(*ptep) & _PAGE_PRESENT))
return;
r3k_flush_dcache_range(kaddr, kaddr + PAGE_SIZE);
if (exec)
r3k_flush_icache_range(kaddr, kaddr + PAGE_SIZE);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 22:20:36 +00:00
}
[MIPS] R3000: local_flush_data_cache_page take a pointer argument. Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
2007-03-24 13:20:15 +00:00
static void local_r3k_flush_data_cache_page(void *addr)
[MIPS] Handle IDE PIO cache aliases on SMP. Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
2006-04-05 19:42:04 +00:00
{
}
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 22:20:36 +00:00
static void r3k_flush_data_cache_page(unsigned long addr)
{
}
static void r3k_flush_cache_sigtramp(unsigned long addr)
{
unsigned long flags;
[MIPS] c-r3k: Implement flush_cache_range() Contrary to the belief of some, the R3000 and related processors did have caches, both a data and an instruction cache. Here is an implementation of r3k_flush_cache_page(), which is the processor-specific back-end for flush_cache_range(), done according to the spec in Documentation/cachetlb.txt. While at it, remove an unused local function: get_phys_page(), do some trivial formatting fixes and modernise debugging facilities. Signed-off-by: Maciej W. Rozycki <macro@linux-mips.org> Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
2007-10-17 10:51:39 +00:00
pr_debug("csigtramp[%08lx]\n", addr);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 22:20:36 +00:00
flags = read_c0_status();
write_c0_status(flags&~ST0_IEC);
/* Fill the TLB to avoid an exception with caches isolated. */
MIPS: Whitespace cleanup. Having received another series of whitespace patches I decided to do this once and for all rather than dealing with this kind of patches trickling in forever. Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
2013-01-22 11:59:30 +00:00
asm( "lw\t$0, 0x000(%0)\n\t"
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 22:20:36 +00:00
"lw\t$0, 0x004(%0)\n\t"
: : "r" (addr) );
write_c0_status((ST0_ISC|ST0_SWC|flags)&~ST0_IEC);
MIPS: Whitespace cleanup. Having received another series of whitespace patches I decided to do this once and for all rather than dealing with this kind of patches trickling in forever. Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
2013-01-22 11:59:30 +00:00
asm( "sb\t$0, 0x000(%0)\n\t"
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 22:20:36 +00:00
"sb\t$0, 0x004(%0)\n\t"
: : "r" (addr) );
write_c0_status(flags);
}
MIPS: cache: Provide cache flush operations for XFS Until now flush_kernel_vmap_range() and invalidate_kernel_vmap_range() did not exist on MIPS resulting in heavy cache corruption on XFS filesystems. Left for the post-3.0 time: optimization and make this work with highmem, too. Since the combination of highmem + cache aliases atm doesn't work this isn't a regression. Signed-off-by: Ralf Baechle <ralf@linux-mips.org> Patchwork: https://patchwork.linux-mips.org/patch/2505/
2011-06-17 15:20:28 +00:00
static void r3k_flush_kernel_vmap_range(unsigned long vaddr, int size)
{
BUG();
}
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 22:20:36 +00:00
static void r3k_dma_cache_wback_inv(unsigned long start, unsigned long size)
{
/* Catch bad driver code */
BUG_ON(size == 0);
iob();
r3k_flush_dcache_range(start, start + size);
}
MIPS: Delete __cpuinit/__CPUINIT usage from MIPS code commit 3747069b25e419f6b51395f48127e9812abc3596 upstream. The __cpuinit type of throwaway sections might have made sense some time ago when RAM was more constrained, but now the savings do not offset the cost and complications. For example, the fix in commit 5e427ec2d0 ("x86: Fix bit corruption at CPU resume time") is a good example of the nasty type of bugs that can be created with improper use of the various __init prefixes. After a discussion on LKML[1] it was decided that cpuinit should go the way of devinit and be phased out. Once all the users are gone, we can then finally remove the macros themselves from linux/init.h. Note that some harmless section mismatch warnings may result, since notify_cpu_starting() and cpu_up() are arch independent (kernel/cpu.c) and are flagged as __cpuinit -- so if we remove the __cpuinit from the arch specific callers, we will also get section mismatch warnings. As an intermediate step, we intend to turn the linux/init.h cpuinit related content into no-ops as early as possible, since that will get rid of these warnings. In any case, they are temporary and harmless. Here, we remove all the MIPS __cpuinit from C code and __CPUINIT from asm files. MIPS is interesting in this respect, because there are also uasm users hiding behind their own renamed versions of the __cpuinit macros. [1] https://lkml.org/lkml/2013/5/20/589 [ralf@linux-mips.org: Folded in Paul's followup fix.] Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com> Cc: linux-mips@linux-mips.org Patchwork: https://patchwork.linux-mips.org/patch/5494/ Patchwork: https://patchwork.linux-mips.org/patch/5495/ Patchwork: https://patchwork.linux-mips.org/patch/5509/ Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
2013-06-18 13:38:59 +00:00
void r3k_cache_init(void)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 22:20:36 +00:00
{
extern void build_clear_page(void);
extern void build_copy_page(void);
r3k_probe_cache();
flush_cache_all = r3k_flush_cache_all;
__flush_cache_all = r3k___flush_cache_all;
flush_cache_mm = r3k_flush_cache_mm;
flush_cache_range = r3k_flush_cache_range;
flush_cache_page = r3k_flush_cache_page;
flush_icache_range = r3k_flush_icache_range;
[MIPS] Fix WARNING: at kernel/smp.c:290 trap_init issues flush_icache_range(), which uses ipi functions to get icache flushing done on all cpus. But this is done before interrupts are enabled and caused WARN_ON messages. This changeset introduces a new local_flush_icache_range() and uses it before interrupts (and additional CPUs) are enabled to avoid this problem. Signed-off-by: Thomas Bogendoerfer <tsbogend@alpha.franken.de> Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
2008-08-04 18:53:57 +00:00
local_flush_icache_range = r3k_flush_icache_range;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 22:20:36 +00:00
MIPS: cache: Provide cache flush operations for XFS Until now flush_kernel_vmap_range() and invalidate_kernel_vmap_range() did not exist on MIPS resulting in heavy cache corruption on XFS filesystems. Left for the post-3.0 time: optimization and make this work with highmem, too. Since the combination of highmem + cache aliases atm doesn't work this isn't a regression. Signed-off-by: Ralf Baechle <ralf@linux-mips.org> Patchwork: https://patchwork.linux-mips.org/patch/2505/
2011-06-17 15:20:28 +00:00
__flush_kernel_vmap_range = r3k_flush_kernel_vmap_range;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 22:20:36 +00:00
flush_cache_sigtramp = r3k_flush_cache_sigtramp;
[MIPS] Handle IDE PIO cache aliases on SMP. Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
2006-04-05 19:42:04 +00:00
local_flush_data_cache_page = local_r3k_flush_data_cache_page;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 22:20:36 +00:00
flush_data_cache_page = r3k_flush_data_cache_page;
_dma_cache_wback_inv = r3k_dma_cache_wback_inv;
_dma_cache_wback = r3k_dma_cache_wback_inv;
_dma_cache_inv = r3k_dma_cache_wback_inv;
printk("Primary instruction cache %ldkB, linesize %ld bytes.\n",
icache_size >> 10, icache_lsize);
printk("Primary data cache %ldkB, linesize %ld bytes.\n",
dcache_size >> 10, dcache_lsize);
build_clear_page();
build_copy_page();
}
Reference in New Issue Copy Permalink